Shigeyuki Fukui

Oligosaccharide microarrays for high-throughput detection and specificity assignments of carbohydrate-protein interactions

We describe microarrays of oligosaccharides as neoglycolipids and their robust display on nitrocellulose. The arrays are sourced from glycoproteins, glycolipids, proteoglycans, polysaccharides, whole organs, or from chemically synthesized oligosaccharides. We show that carbohydrate-recognizing proteins single out their ligands not only in arrays of homogeneous oligosaccharides but also in arrays of heterogeneous oligosaccharides. Initial applications have revealed new findings, including: (i) among O-glycans in brain, a relative abundance of the Lewisx sequence based on N-acetyllactosamine recognized by anti-L5, and a paucity of the Lewisx sequence based on poly-N-acetyllactosamine recognized by anti-SSEA-1; (ii) insights into chondroitin sulfate oligosaccharides recognized by an antiserum and an antibody (CS-56) to chondroitin sulfates; and (iii) binding of the cytokine interferon-γ (IFN-γ) and the chemokine RANTES to sulfated sequences such as HNK-1, sulfo-Lewisx, and sulfo-Lewisa, in addition to glycosaminoglycans. The approach opens the way for discovering new carbohydrate-recognizing proteins in the proteome and for mapping the repertoire of carbohydrate recognition structures in the glycome.

Structure and Expression of the mRNA Encoding a Novel Fibroblast Growth Factor, FGF-18

We isolated the cDNA encoding a novel member (207 amino acids) of the fibroblast growth factor (FGF) family from rat embryos. Because this protein is the 18th documented member of the FGF family, we tentatively termed it FGF-18. We have also determined mouse and human FGF-18 with high amino acid identity (99.5 and 99.0%) to rat FGF-18, respectively. Among FGF family members, FGF-18 is most similar (52.7% amino acid identity) to FGF-8 and FGF-17. FGF-18 has a typical signal sequence at its amino terminus. Recombinant rat FGF-18, which was efficiently secreted by High Five insect cells infected with recombinant baculovirus containing the cDNA, induced neurite outgrowth in PC12 cells. The expression of FGF-18 mRNA was examined in adult rat tissues and embryos by Northern blotting analysis and in situ hybridization. FGF-18mRNA of ∼2.7 kilobases was preferentially detected in the lung among adult rat tissues examined. In rat embryos, FGF-18mRNA was detected in several discrete regions at embryonic days 14.5 and 19.5 but not at E10.5. The temporal and spatial patterns ofFGF-18 mRNA expression in embryos are quite different from those of FGF-8 and FGF-17 mRNAs reported. The present results indicate that FGF-18 is a unique secreted signaling molecule in the adult lung and developing tissues.

A monoclonal antibody directed to Tn antigen

A murine monoclonal antibody, MLS 128, that was assigned to an anti-Tn antibody has been established by immunizing mice with human colonic cancer cells (LS 180). MLS 128 bound to mucin glycopeptides from LS 180 cells and their asialo forms to the same extent as well as to ovine submaxillary mucin (OSM) and asialo OSM. Special non-sialylated GalNAc residue(s) attached to a certain peptide region in the antigens seems to be involved in the binding since N-acetylgalactosaminidase treatment of the antigen abolished the binding and pronase digestion diminished the binding markedly.

Mucin-carbohydrate directed monoclonal antibody

To raise monoclonal antibodies recognizing cancer‐associated alterations of the carbohydrate structure of glycoproteins, Balb/c mice were immunized with human colonic cancer cells (LS 180 from ATCC). One of the generated hybridomas produced a monoclonal antibody that bound to the carbohydrate moiety of mucin‐type glycoproteins from LS 180. The antibody did not bind to glycoproteins from another colonic cancer cell line, SW 1116, or to glycolipids from any of the colonic cancer cell lines. The antibody bound to ovine and bovine submaxillary mucins (OSM and BSM). NeuAcα2→6Ga1NAc seemed to be involved in the epitope.

Two Related but Distinct Chondroitin Sulfate Mimetope Octasaccharide Sequences Recognized by Monoclonal Antibody WF6

Chondroitin sulfate (CS) proteoglycans are major components of cartilage and other connective tissues. The monoclonal antibody WF6, developed against embryonic shark cartilage CS, recognizes an epitope in CS chains, which is expressed in ovarian cancer and variably in joint diseases. To elucidate the structure of the epitope, we isolated oligosaccharide fractions from a partial chondroitinase ABC digest of shark cartilage CS-C and established their chain length, disaccharide composition, sulfate content, and sulfation pattern. These structurally defined oligosaccharide fractions were characterized for binding to WF6 by enzyme-linked immunosorbent assay using an oligosaccharide microarray prepared with CS oligosaccharides derivatized with a fluorescent aminolipid. The lowest molecular weight fraction recognized by WF6 contained octasaccharides, which were split into five subfractions. The most reactive subfraction contained several distinct octasaccharide sequences. Two octasaccharides, ΔD-C-C-C and ΔC-C-A-D (where A represents GlcUAβ1-3GalNAc(4-O-sulfate), C is GlcUAβ1-3Gal-NAc(6-O-sulfate), D is GlcUA(2-O-sulfate)β1-3GalNAc(6-O-sulfate), ΔCis Δ4,5HexUAα1-3GalNAc(6-O-sulfate), and ΔDis Δ4,5HexUA(2-O-sulfate)α1-3GalNAc(6-O-sulfate)), were recognized by WF6, but other related octasaccharides, ΔC-A-D-C and ΔC-C-C-C, were not. The structure and sequences of both the binding and nonbinding octasaccharides were compared by computer modeling, which revealed a remarkable similarity between the shape and distribution of the electrostatic potential in the two different octasaccharide sequences that bound to WF6 and that differed from the nonbinding octasaccharides. The strong similarity in structure predicted for the two binding CS octasaccharides (ΔD-C-C-C and ΔC-C-A-D) provided a possible explanation for their similar affinity for WF6, although they differed in sequence and thus form two specific mimetopes for the antibody.

Characterization of mucin-type oligosaccharides with the sialyl-Lea structure from human colorectal adenocarcinoma cells

Oligosaccharides with the sialyl-Le(a) structure have been isolated on an affinity column of a monoclonal antibody, MSW 113, from mucin-type glycoproteins derived from the surfaces of SW 1116 and LS 180 cells, and their secretions. The oligosaccharides were polydisperse with respect to molecular size, the oligosaccharides derived from glycoproteins in culture media being larger than those in cell lysates, as assessed by gel filtration. Some of the oligosaccharides were susceptible to degradation by endo-beta-galactosidase (E. freundii), as judged from the change in the gel filtration pattern. These results indicate that oligosaccharides with the sialyl-Le(a) structure derived from mucin-type glycoproteins produced by human colonic cancer cells are extremely large in size and complex in structure, and that some of them contain the poly-N-acetyllactosamine structure.

Expression of the Tn antigen on T-lymphoid cell line Jurkat

Expression of the Tn antigen on a T-lymphoid cell line, Jurkat, was investigated using an anti-Tn monoclonal antibody, MLS 128. Immunoprecipitation or immunoaffinity chromatography of a lysate of Jurkat cells led to the isolation of a 120 kDa glycoprotein carrying the Tn antigen. This glycoprotein and leukosialin (CD43) were indistinguishable on SDS-PAGE and as to immunoreactivity with MLS 128. Leukosialin from an erythroid cell line, K562, exhibited no reactivity with MLS 128 despite that this leukosialin has several GalNAc alpha-Ser(Thr) structures. Pulse-chase experiments with the Jurkat leukosialin showed that newly synthesized leukosialin acquired the antigenecity after a lag of about 30 min, whereas incorporation of GalNAc into the leukosialin occurred earlier. These results indicate that the Tn antigen is expressed on leukosialin and that its epitopic structure is more complex than GalNAc alpha-Ser(Thr).

Detection of oligosaccharide ligands for Hepatocyte growth factor/Scatter factor (HGF/SF), Keratinocyte growth factor (KGF/FGF-7), RANTES and Heparin cofactor II by neoglycolipid microarrays of glycosaminoglycan-derived oligosaccharide fragments

Neoglycolipid technology is eminently adaptable for microarray design for high-throughput detection and specificity assignments of carbohydrate-protein interactions. Dermatan sulfate (DS) is known to play an important role because of its ability to bind growth factors as well as chemokines and to modulate their biological activities during inflammation and response to injury. We prepared various iduronic acid-rich fragments from DS by complete digestion with chondroitinase ACI, and investigated whether the DS-binding proteins, such as HGF/SF, RANTES, KGF/FGF-7 and HCII, can detect their oligosaccharide ligands in a neoglycolipid microarray. First, a comparison of the intensity of binding signals obtained from chondroitin oligosaccharides with those of heparin oligosaccharides showed that our microarray system is feasible not only to single-out the oligosaccharide ligands, but also to detect the difference between an intrinsic interaction unrelated only to electrostatic interaction and non-specific electrostatic interaction. Second, HGF/SF, KGF/FGF-7 and HCII showed preferential binding to iduronic acid-rich fragments of DS oligosaccharides that are greater than 8-mers in lengths. In contrast, RANTES binding seemed to depend only on the negative charges; their binding intensity towards the DS oligosaccharides was somewhat stronger than the binding of HGF/SF, KGF/FGF-7 and HCII. Third, the use of polyvinylpyrrolidone-40 (PVP-40), ovalbumin (OV) and Tween 20 in place of BSA as a blotting agent was useful in these glycosaminoglycan dependent reactions to minimize background due to non-specific interactions.

Highly sulfated hexasaccharide sequences isolated from chondroitin sulfate of shark fin cartilage: Insights into the sugar sequences with bioactivities

Chondroitin sulfate (CS) chains regulate the development of the central nervous system in vertebrates and are linear polysaccharides consisting of variously sulfated repeating disaccharides, [-4GlcUAβ1-3GalNAcβ1-](n), where GlcUA and GalNAc represent D-glucuronic acid and N-acetyl-D-galactosamine, respectively. CS chains containing D-disaccharide units [GlcUA(2-O-sulfate)-GalNAc(6-O-sulfate)] are involved in the development of cerebellar Purkinje cells and neurite outgrowth-promoting activity through interaction with a neurotrophic factor, pleiotrophin, resulting in the regulation of signaling. In this study, to obtain further structural information on the CS chains containing d-disaccharide units involved in brain development, oligosaccharides containing D-units were isolated from a shark fin cartilage. Seven novel hexasaccharide sequences, ΔO-D-D, ΔA-D-D, ΔC-D-D, ΔE-A-D, ΔD-D-C, ΔE-D-D and ΔA-B-D, in addition to three previously reported sequences, ΔC-A-D, ΔC-D-C and ΔA-D-A, were isolated from a CS preparation of shark fin cartilage after exhaustive digestion with chondroitinase AC-I, which cannot act on the galactosaminidic linkages bound to D-units. The symbol Δ stands for a 4,5-unsaturated bond of uronic acids, whereas A, B, C, D, E and O represent [GlcUA-GalNAc(4-O-sulfate)], [GlcUA(2-O-sulfate)-GalNAc(4-O-sulfate)], [GlcUA-GalNAc(6-O-sulfate)], [GlcUA(2-O-sulfate)-GalNAc(6-O-sulfate)], [GlcUA-GalNAc(4-O-, 6-O-sulfate)] and [GlcUA-GalNAc], respectively. In binding studies using an anti-CS monoclonal antibody, MO-225, the epitopes of which are involved in cerebellar development in mammals, novel epitope structures, ΔA-D-A, ΔA-D-D and ΔA-B-D, were revealed. Hexasaccharides containing two consecutive D-units or a B-unit will be useful for the structural and functional analyses of CS chains particularly in the neuroglycobiological fields.

Production of Monoclonal Antibodies Directed against Carbohydrate Moieties of Cell Surface Glycoproteins

Through the use of a technique for raising monoclonal antibodies, coupled with a solid‐phase radioimmunoassay utilizing immobilized glyeopeptides prepared from the surface membranes of the colorectal cancer cells (LS 180) used for the immunization, carbohydrate‐directed monoclonal antibodies were obtained. One of the monoclonal antibodies, MLS 102, reacted immunohistochemically intensely with the colorectal cancer cell surface and the mutinous glycoproteins secreted by the cancer cells, but only weakly with normal colon tissue. The antigenic determinant recognized by MLS 102 was the carbohydrate moiety of glycoproteins with terminal sialic acid. The antigens defined by other monoclonal antibodies, MLS 103 and 104, were immunohistochemically detected in both normal colonic epithelial and cancer cells. These antibodies seemed to recognize the carbohydrate moieties of both glycoproteins and glycolipids. The method described in this report can be generally applied to raise cell surface carbohydrate‐directed antibodies.