FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1996364990> ?p ?o ?g. }

• W1996364990 endingPage "2454" @default.
• W1996364990 startingPage "2448" @default.
• W1996364990 abstract "Collectins are C-type animal lectins with both collagenous and carbohydrate recognition domains and are involved in the first line host defense against pathogens. We report here a novel Ca2+-dependent and GlcNAc-binding lectin consisting of subunits of 35 kDa (P35) with a collagen-like sequence. When P35 is isolated from human serum, it forms a homopolymer by means of intermolecular disulfide bonding, as is the case with collectins. P35 cDNA was cloned from a human liver cDNA library, and the deduced amino acid sequence of 313 residues revealed that the mature form of P35 consists mainly of collagen- and fibrinogen-like domains. The latter contained two potential Ca2+-binding sites that may be involved in carbohydrate binding. The overall sequence of P35 was highly homologous to porcine ficolins α and β. Northern blots of various human tissues showed that the major product of the 1.3-kilobase-long P35 transcript is expressed in liver. P35 enhanced phagocytosis of Salmonella typhimurium by neutrophils, suggesting an opsonic effect via the collagen region. P35 was found to bind to GlcNAc-conjugated bovine serum albumin, a neoglycoprotein, as well as to neoglycolipids containing complex-type oligosaccharides derived from glycoproteins, suggesting that P35 recognizes GlcNAc residues such as those found in microbial glycoconjugates and complex-type oligosaccharides. Therefore, P35 represents a new type of GlcNAc-binding lectin with structural and functional similarities to collectins involved in innate immunity. Collectins are C-type animal lectins with both collagenous and carbohydrate recognition domains and are involved in the first line host defense against pathogens. We report here a novel Ca2+-dependent and GlcNAc-binding lectin consisting of subunits of 35 kDa (P35) with a collagen-like sequence. When P35 is isolated from human serum, it forms a homopolymer by means of intermolecular disulfide bonding, as is the case with collectins. P35 cDNA was cloned from a human liver cDNA library, and the deduced amino acid sequence of 313 residues revealed that the mature form of P35 consists mainly of collagen- and fibrinogen-like domains. The latter contained two potential Ca2+-binding sites that may be involved in carbohydrate binding. The overall sequence of P35 was highly homologous to porcine ficolins α and β. Northern blots of various human tissues showed that the major product of the 1.3-kilobase-long P35 transcript is expressed in liver. P35 enhanced phagocytosis of Salmonella typhimurium by neutrophils, suggesting an opsonic effect via the collagen region. P35 was found to bind to GlcNAc-conjugated bovine serum albumin, a neoglycoprotein, as well as to neoglycolipids containing complex-type oligosaccharides derived from glycoproteins, suggesting that P35 recognizes GlcNAc residues such as those found in microbial glycoconjugates and complex-type oligosaccharides. Therefore, P35 represents a new type of GlcNAc-binding lectin with structural and functional similarities to collectins involved in innate immunity. Animal lectins are mainly involved in endocytosis, cell-cell interactions, and host defense. Collectins (1.Malhotra R. Haurum J. Thiel S. Sim R.B. Eur. J. Immunol. 1992; 22: 1437-1445Crossref PubMed Scopus (121) Google Scholar, 2.Sastry K. Ezekowitz R.A. Curr. Opin. Immunol. 1993; 5: 59-66Crossref PubMed Scopus (185) Google Scholar, 3.Holmskov U. Malhotra R. Sim R.B. Jensenius J.C. Immunol. Today. 1994; 15: 67-74Abstract Full Text PDF PubMed Scopus (421) Google Scholar, 4.Hoppe H.J. Reid K.B.M. Protein Sci. 1994; 3: 1143-1158Crossref PubMed Scopus (192) Google Scholar) are a group of oligomeric C-type (Ca2+-dependent) (5.Drickamer K. J. Biol. Chem. 1988; 263: 9557-9560Abstract Full Text PDF PubMed Google Scholar) animal lectins and include three types of serum lectins (mannose-binding protein (MBP)( 1The abbreviations used are: MBPmannose-binding proteinCRDcarbohydrate recognition domainBSAbovine serum albuminGHBSSHanks' balanced salt solution containing 0.1% gelatinPMNpolymorphonuclear neutrophilskbkilobase(s).) (6.Thiel S. Immunopharmacology. 1992; 24: 91-99Crossref PubMed Scopus (43) Google Scholar, 7.Turner M. Biochem. Soc. Trans. 1994; 22: 88-94Crossref PubMed Scopus (30) Google Scholar), conglutinin, and CL-43(8.Jensenius J.C. Laursen S.B. Zheng Y. Holmskov U. Biochem. Soc. Trans. 1994; 22: 95-100Crossref PubMed Scopus (10) Google Scholar)) and two pulmonary surfactant proteins (SP-A and SP-D(9.Kuroki Y. Voelker D.R. J. Biol. Chem. 1994; 269: 25943-25946Abstract Full Text PDF PubMed Google Scholar)). Their structures and functions are related to C1q, a subcomponent of the first complement component (C1). mannose-binding protein carbohydrate recognition domain bovine serum albumin Hanks' balanced salt solution containing 0.1% gelatin polymorphonuclear neutrophils kilobase(s). Collectins possess a structural unit made up of three polypeptide chains, each containing at least three domains. The first domain consists of an NH2-terminal sequence containing cysteine residues that make it possible for structural units to form oligomers via disulfide bonds. The second domain is a collagenous region consisting of Gly-X-Y triplet repeats (where X and Y represent any amino acid). The third domain contains a COOH-terminal sequence consisting of a carbohydrate recognition domain (CRD) (10.Weis W.I. Kahn R. Fourme R. Drickamer K. Hendrickson W.A. Science. 1991; 254: 1608-1615Crossref PubMed Scopus (486) Google Scholar) that is also involved in Ca2+ binding. It has 14 invariant and 18 highly conserved residues spread over ∼120 amino acids and is found in many animal lectins(11.Weis W.I. Drickamer K. Hendrickson W.A. Nature. 1992; 360: 127-134Crossref PubMed Scopus (856) Google Scholar, 12.Drickamer K. Nature. 1992; 360: 183-186Crossref PubMed Scopus (453) Google Scholar). Collectins play a role in the first line of host defense against microorganisms possessing certain oligosaccharides on their surface. Opsonization appears to be their common function, as is the case with MBP (13.Kuhlman M. Joiner K. Ezekowitz R.A.B. J. Exp. Med. 1989; 169: 1733-1745Crossref PubMed Scopus (393) Google Scholar, 14.Hartshorn K.L. Sastry K. White M.R. Anders E.M. Super M. Ezekowitz R.A.B. Tauber A.I. J. Clin. Invest. 1993; 91: 1414-1420Crossref PubMed Scopus (183) Google Scholar) and conglutinin(15.Hartshorn K.L. Sastry K. Brown D. White M.R. Okarma T.B. Lee Y.-M. Tauber A.I. J. Immunol. 1993; 151: 6265-6273PubMed Google Scholar), and may occur through the interaction of collagenous regions with the C1q receptor on phagocytes (16.Malhotra R. Thiel S. Reid K.B.M. Sim R.B. J. Exp. Med. 1990; 172: 955-959Crossref PubMed Scopus (224) Google Scholar, 17.Malhotra R. Willis A.C. Jensenius J.C. Jackson J. Sim R.B. Immunology. 1993; 78: 341-348PubMed Google Scholar). MBP and conglutinin have been shown to bind to gp120 and gp160 of the envelope glycoproteins of human immunodeficiency virus type 1, respectively(18.Larkin M. Childs R.A. Matthews T.J. Thiel S. Mizuochi T. Lawson A.M. Savill J.S. Haslett C. Diaz R. Feizi T. AIDS (Phila.). 1989; 3: 793-798Crossref PubMed Scopus (89) Google Scholar, 19.Andersen O. Sorensen A.-M. Svehay S.-E. Fenouillet E. Scand. J. Immunol. 1991; 33: 81-88Crossref PubMed Scopus (28) Google Scholar), thereby inhibiting virus infection of cells (20.Ezekowitz R.A.B. Kuhlman M. Groopman L.E. Byrn R.A. J. Exp. Med. 1989; 169: 185-196Crossref PubMed Scopus (302) Google Scholar) or the interaction between gp160 and CD4(19.Andersen O. Sorensen A.-M. Svehay S.-E. Fenouillet E. Scand. J. Immunol. 1991; 33: 81-88Crossref PubMed Scopus (28) Google Scholar). In addition, MBP has the unique property of being able to activate the complement system(21.Ikeda K. Sannoh T. Kawasaki N. Kawasaki T. Yamashina I. J. Biol. Chem. 1987; 262: 7451-7454Abstract Full Text PDF PubMed Google Scholar, 22.Kawasaki N. Kawasaki T. Yamashina I. J. Biochem. (Tokyo). 1989; 106: 483-489Crossref PubMed Scopus (151) Google Scholar, 23.Schweinle J.E. Ezekowitz R.A.B. Tenner A.J. Kuhlman M. Joiner K.A. J. Clin. Invest. 1989; 84: 1821-1829Crossref PubMed Scopus (116) Google Scholar, 24.Haurum J.S. Thiel S. Jones I.M. Fischer P.B. Laursen S.B. Jensenius J.C. AIDS (Phila.). 1993; 7: 1307-1313Crossref PubMed Scopus (130) Google Scholar), probably by utilizing a C1s-like serine protease termed MBP-associated serine protease(25.Matsushita M. Fujita T. J. Exp. Med. 1992; 176: 1497-1502Crossref PubMed Scopus (569) Google Scholar, 26.Sato T. Endo Y. Matsushita M. Fujita T. Int. Immunol. 1994; 6: 665-669Crossref PubMed Scopus (154) Google Scholar), which is capable of cleaving C4 and C2, leading to the killing of bacteria or viruses by direct lysis or opsonization. We describe here a novel type of human serum Ca2+-dependent and GlcNAc-binding lectin with opsonic activity. This lectin has a collagenous domain similar to that of collectins. Unlike the CRD in collectins, however, it possesses a fibrinogen-like domain that is probably involved in carbohydrate binding. Mannan from Saccharomyces cerevisiae, GlcNAc-agarose, ovalbumin, ribonuclease B, human IgG, fetuin, asialofetuin, human α1-acid glycoprotein, and bovine serum albumin (BSA) conjugated with lactose (lactose-BSA) and cellobiose (cellobiose-BSA) were purchased from Sigma. Human fibrinogen was from KabiVitrum (Stockholm, Sweden), and neoglycoproteins of Man-BSA, Glc-BSA, GlcNAc-BSA, and Gal-BSA were from E. Y. Laboratories (San Mateo, CA). In each neoglycoprotein, ∼19-42 mol of monosaccharide were conjugated to 1 mol of BSA. Mannose and GlcNAc were from Wako Pure Chemical Industries (Osaka, Japan). CNBr-activated Sepharose 4B was from Pharmacia Fine Chemicals (Uppsala, Sweden). Coupling of mannan or asialofetuin to CNBr-activated Sepharose was performed according to the manufacturer's instructions. Na125I and [α-32P]dCTP were purchased from Amersham International (Buckinghamshire, United Kingdom). P35 was labeled using Na125I and IODO-GEN from Pierce, and its specific activity was 0.24 MBq/μg. Human serum obtained by recalcification of outdated plasma was treated with polyethylene glycol 4000 at a final concentration of 7%. The precipitate was dissolved in 50 mM Tris, 1 M NaCl, 50 mM CaCl2, pH 7.8 (starting buffer), and applied to a mannan-Sepharose column. After washing the column, MBP was eluted with starting buffer containing 300 mM mannose. P35 was then eluted with starting buffer containing 150 mM GlcNAc. Since the eluate was contaminated with IgG and IgM, the eluted fractions were dialyzed against 50 mM Tris, 50 mM NaCl, pH 8.0, and then chromatographed on a Mono Q column that had been equilibrated with the above buffer. Elution was carried out by applying a linear NaCl gradient to 0.5 M. P35 was eluted at a NaCl concentration between 0.15 and 0.2 M. The preparation of P35 was analyzed by SDS-polyacrylamide gel electrophoresis using the Laemmli system(27.Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207856) Google Scholar). Isolated P35 (10 μg) was injected with complete Freund's adjuvant into BALB/c mice. After 3 weeks, the mice were boosted with 10 μg of P35, and 3 days later, spleens were removed and used for fusion with P3U1 cells. Hybridoma cells producing anti-P35 antibodies were obtained by screening culture supernatants for antibodies reactive against P35 using an enzyme-linked immunosorbent assay. One of the positive monoclonal antibodies (2F5) was purified from ascites fluid of mice bearing the hybridoma by precipitation with ammonium sulfate at 50% saturation followed by chromatography on DEAE-Toyopearl. 2F5 was of the IgG1 isotype. Based on analysis of the NH2-terminal amino acid sequence of the purified P35 protein (see Fig. 2), two oligonucleotides were synthesized as probes for the cloning of P35 cDNA: 5′-GA(A/G)GTNAA(A/G)ATGGTNGG-3′ corresponding to amino acids 9-14 and 5′-ATGGTIGGICTIGAIGGITCIGAIAAICTIAC-3′ corresponding to amino acids 12-22 of mature P35. By screening for a clone that was positive for both oligonucleotides, a P35 cDNA clone lacking a portion of the 5′-end was isolated from a human liver λZAP cDNA library. By further screening using the cDNA clone as a probe, 11 positive clones were obtained. These positive clones were subcloned into pBluescript II SK− by in vivo excision according to the instructions provided by Stratagene (La Jolla, CA) and sequenced by the dideoxy method with a Sequenase DNA sequencing kit (U. S. Biochemical Corp.). By overlapping the sequences of these clones, a stretch of P35 cDNA was constructed. A membrane filter containing 2 μg of poly(A)+ mRNA from various human tissues (CLONTECH, Palo Alto, CA) was hybridized with 32P-labeled cDNA (EcoRI-XhoI fragment of 971 base pairs corresponding to nucleotides 110-1080) at 42°C overnight in 50% formamide, 5 × Denhardt's solution, 5 × SSPE (1 × SSPE is 9 mM sodium phosphate, 150 mM NaCl, 1 mM EDTA, pH 7.4) 0.5% SDS, and 200 μg/ml salmon sperm DNA. After a final washing at 60°C for 30 min in 0.1 × SSC (15 mM sodium citrate, 150 mM NaCl, pH 7.0) containing 0.1% SDS, the filter was exposed to an autoradiogram imaging system, and the image was then read using a bioimaging analyzing system (BAS 1000, Fujifilm, Tokyo, Japan). The NH2-terminal amino acid sequence of P35 was determined with a gas-phase protein sequencer (PSQ-1, Shimadzu Corp., Kyoto, Japan). Microtiter plates (MaxiSorp, Nunc, Roskilde, Denmark) were coated with 100 μl of mannan (0.5 mg/ml 50 mM carbonate buffer, pH 9.6) by incubation at 4°C overnight. Wells were washed with Veronal buffer containing 0.1% Tween 20 and then blocked with Blockace (Dainippon Pharmaceutical Co., Ltd., Osaka, Japan) at 37°C for 2 h. After washing, wells were treated with 30 μl of P35 diluted in Veronal buffer containing either 2 mM CaCl2 or 10 mM EDTA at 37°C for 2 h. Wells were then washed with Veronal buffer/Tween 20 containing 2 mM CaCl2 or 10 mM EDTA and then with Veronal buffer/Tween 20 containing 2 mM CaCl2, followed by incubation with 30 μl of anti-P35 monoclonal antibody (2F5) at 37°C for 2 h. After washing, horseradish peroxidase-conjugated goat anti-mouse IgG (Caltag Lab, San Francisco, CA) was added to the wells, and incubation was carried out at 37°C for 2 h. The color was developed by addition of azinodi-(3-ethylbenzthiazoline)sulfonic acid (ABTS, Zymed Laboratories, Inc., South San Francisco, CA) in citrate buffer, pH 4.4, containing 0.025% hydrogen peroxide. The absorbance at 405 nm was determined using a Bio-Rad photometer (Model 450 Microplate Reader). Oligosaccharides were quantitatively released by hydrazinolysis, as described previously in detail(28.Mizuochi T. Hounsell E.F. Methods in Molecular Biology: Glycoprotein Analysis in Biomedicine. 14. Humana Press Inc., Totowa, NJ1993: 55-68Google Scholar, 29.Matsui T. Titani K. Mizuochi T. J. Biol. Chem. 1992; 267: 8723-8731Abstract Full Text PDF PubMed Google Scholar), from the following six glycoproteins: ovalbumin with high mannose-type and hybrid-type oligosaccharides(30.Tai T. Yamashita K. Ito S. Kobata A. J. Biol. Chem. 1977; 252: 6687-6694Abstract Full Text PDF PubMed Google Scholar); bovine pancreatic RNase B with only high mannose-type oligosaccharides(31.Liang C.-J. Yamashita K. Kobata A. J. Biochem. (Tokyo). 1980; 88: 51-58PubMed Google Scholar); human IgG with di-, mono-, and non-galactosylated biantennary complex-type oligosaccharides(32.Mizuochi T. Taniguchi T. Shimizu A. Kobata A. J. Immunol. 1982; 129: 2016-2020PubMed Google Scholar); human fibrinogen with digalactosylated biantennary complex-type oligosaccharides as the major oligosaccharides(33.Mizuochi T. Taniguchi T. Asami Y. Takamatsu J. Okude M. Iwanaga S. Kobata A. J. Biochem. (Tokyo). 1982; 92: 283-293Crossref PubMed Scopus (37) Google Scholar); fetuin with triantennary complex-type oligosaccharides as the major oligosaccharides(34.Takasaki S. Kobata A. Biochemistry. 1986; 25: 5709-5715Crossref PubMed Scopus (123) Google Scholar); and human α1-acid glycoprotein with tetraantennary complex-type oligosaccharides as the major oligosaccharides(35.Yoshima H. Matsumoto A. Mizuochi T. Kawasaki T. Kobata A. J. Biol. Chem. 1981; 256: 8476-8484Abstract Full Text PDF PubMed Google Scholar). Oligosaccharide mixtures containing complex-type oligosaccharides were converted into neutral oligosaccharides by sialidase treatment. The neutral oligosaccharide mixtures obtained from the six glycoproteins were conjugated to dipalmitoylphosphatidylethanolamine as described previously(36.Mizuochi T. Loveless R.W. Lawson A.M. Chai W. Lachmann P.J. Childs R.A. Thiel S. Feizi T. J. Biol. Chem. 1989; 264: 13834-13839Abstract Full Text PDF PubMed Google Scholar, 37.Mizuochi T. Trends Glycosci. Glycotechnol. 1991; 3: 435-437Crossref Scopus (8) Google Scholar). The reaction mixtures containing 2 μg of oligosaccharides were subjected to TLC using chloroform/methanol/water (105:100:28, by volume). Since all the neoglycolipids thus prepared contained a common lipid, they were separated according to their oligosaccharide structures using this TLC system. After developing to ∼7 cm from the origin, the TLC plates were air-dried. They were then overlaid with 125I-labeled P35, and binding to the neoglycolipids was assessed by autoradiography and orcinol staining as reported previously (36.Mizuochi T. Loveless R.W. Lawson A.M. Chai W. Lachmann P.J. Childs R.A. Thiel S. Feizi T. J. Biol. Chem. 1989; 264: 13834-13839Abstract Full Text PDF PubMed Google Scholar, 38.Loveless R.W. Feizi T. Childs R.A. Mizuochi T. Stoll M.S. Oldroyd R.G. Lachmann P.J. Biochem. J. 1989; 258: 109-113Crossref PubMed Scopus (48) Google Scholar). The six glycoproteins used were those whose oligosaccharides had been used for preparation of the neoglycolipids for the P35 overlay assay, and the six neoglycoproteins used were those in which BSA had been conjugated with mannose, glucose, GlcNAc, galactose, lactose, and cellobiose. These were dot-blotted on an Immobilon-P membrane (Millipore Co., Bedford, MA), which was then blocked and incubated with 125I-labeled P35. After the membranes were washed, binding of P35 to the glycoprotein and neoglycoprotein dots was assessed by autoradiography as reported previously(36.Mizuochi T. Loveless R.W. Lawson A.M. Chai W. Lachmann P.J. Childs R.A. Thiel S. Feizi T. J. Biol. Chem. 1989; 264: 13834-13839Abstract Full Text PDF PubMed Google Scholar, 38.Loveless R.W. Feizi T. Childs R.A. Mizuochi T. Stoll M.S. Oldroyd R.G. Lachmann P.J. Biochem. J. 1989; 258: 109-113Crossref PubMed Scopus (48) Google Scholar). Two strains of Salmonella typhimurium, TV119 and LT2, were kindly provided by Dr. K. Yasunaka (Fukuoka University School of Medicine). TV119 belongs to an Ra chemotype strain bearing rough core polysaccharides, including GlcNAc, at the nonreducing termini, whereas LT2 has additional O-polysaccharides. Bacteria were cultivated overnight at 37°C in LB broth and then diluted 1:50 in fresh broth. After incubation at 37°C for 3 h, bacteria at the midlag phase were washed with Hanks' balanced salt solution containing 0.1% gelatin (GHBSS) and resuspended in GHBSS containing 10 mM CaCl2 at A660 = 0.5 (1 × 108 colony-forming units/ml). PMN were isolated from peripheral blood of healthy volunteers. Heparinized whole blood was centrifuged at 250 × g for 10 min. After removal of the supernatant, blood cells were diluted 2-fold with phosphate-buffered saline. PMN were isolated by centrifugation on Polymorphoprep (Nycomed Pharma A/S, Oslo, Norway) and resuspended at 1 × 107/ml in GHBSS containing 10 mM CaCl2. Binding of P35 to S. typhimurium was analyzed by flow cytometry. S. typhimurium cells (TV119 or LT2; 1 × 107 colony-forming units) in 100 μl of gelatin/Veronal buffer containing 2 mM CaCl2 were treated with 200 μl of gelatin/Veronal buffer containing 20 μg/ml P35 at 4°C for 30 min. In inhibition experiments, P35 was preincubated for 30 min with 100 mM GlcNAc or mannan (50 mg/ml) before it was added to the bacteria. Bacteria were then incubated with 20 μl of anti-P35 monoclonal antibody 2F5 (100 μg/ml) at 4°C for 30 min and stained with 20 μl of 100 μg/ml fluorescein isothiocyanate-conjugated rabbit anti-mouse immunoglobulins (Dako Japan Co., Ltd., Kyoto, Japan) at 4°C for 30 min. The bacteria were washed twice between each reaction with gelatin/Veronal buffer containing 2 mM CaCl2. Reactivities were evaluated on a FACScan flow cytometer (Becton Dickinson, Mountain View, CA) and compared with a control consisting of bacteria treated with 2F5 and fluorescein isothiocyanate-conjugated rabbit anti-mouse immunoglobulins. S. typhimurium cells (TV119 or LT2; 2 × 107 colony-forming units) suspended in 200 μl of GHBSS were incubated in the presence or absence of P35 at 4°C for 30 min. Bacteria were then washed and resuspended in 200 μl of GHBSS, followed by addition of 200 μl of PMN (1 × 107/ml). The mixture containing bacterial cells and PMN was incubated at 37°C for 20 min. After the cells were washed with buffer, they were stained with 5-Giemsa, and phagocytosis of bacteria by PMN was observed by light microscopy. The number of bacteria ingested per 100 PMN was defined as the phagocytic index. Significance was tested using Student's t test. In the purification scheme of MBP, the 7% polyethylene glycol precipitate of human serum was first subjected to affinity chromatography on a yeast mannan-Sepharose column, and MBP was eluted with mannose in the presence of calcium ions as described previously(25.Matsushita M. Fujita T. J. Exp. Med. 1992; 176: 1497-1502Crossref PubMed Scopus (569) Google Scholar). In the present study, we then carried out elution with GlcNAc to examine whether lectins such as conglutinin-like protein were retained on mannan-Sepharose after MBP had been eluted. As a result, a protein with a molecular mass of ∼35 kDa under reducing conditions was found in the eluate from mannan-Sepharose. We tentatively named this lectin P35. Further purification of P35 was achieved by ion-exchange chromatography using a Mono Q column since the P35 preparation from mannan-Sepharose was contaminated with IgG and IgM. As shown in Fig. 1, P35 showed a single band that was slightly larger than the MBP subunit under reducing conditions. Under nonreducing conditions, it appeared as a 320-kDa moiety. These results suggest that P35 is composed of subunits linked to form a homopolymer via disulfide bonds, as is the case with MBP. The NH2-terminal 43-amino acid sequence of P35 determined using a protein sequencer shows glycine residues at intervals of two amino acids, suggesting a collagenous type of sequence (Fig. 2). Based on the NH2-terminal amino acid sequence of the P35 subunit, two oligonucleotides were synthesized and used to screen for cDNA clones of P35 in a human liver λZAP cDNA library. We eventually obtained 11 positive clones that we sequenced. cDNA sequence analysis showed that P35 contains a 939-base pair open reading frame encoding 313 amino acids preceded by a leader peptide of 25 amino acids (Fig. 3). The predicted molecular size of mature P35 was calculated to be 31.4 kDa, which nearly matched that of the P35 subunit. There are two potential N-glycosylation sites at residues 240 and 300 and two potential calcium-binding sites in the COOH-terminal half. P35 consists of three major domains: an NH2-terminal noncollagenous region of 1-13 amino acids; a collagen-like sequence starting at amino acid 14 with two Gly-X-Y triplets followed by a gap of six amino acid residues and 15 Gly-X-Y repeats; and a COOH terminus of 218 amino acids, which in P35 has fibrinogen and fibrinogen-like characteristics. By conducting a homology search, we found that P35 shows 77.4 and 75.4% homology to ficolins α and β, respectively, at the amino acid level (Fig. 4). Ficolin was first identified in porcine uterus membranes and characterized as a transforming growth factor-β-binding protein by Ichijo et al.(39.Ichijo H. Ronnstrand L. Miyagawa K. Ohashi H. Heldin C.-H. Miyazono K. J. Biol. Chem. 1991; 266: 22459-22464Abstract Full Text PDF PubMed Google Scholar), although its physiological functions remain unknown. These investigators isolated two closely related cDNA clones encoding two proteins designated ficolins α and β, which have 83% homology and which consist of collagen- and fibrinogen-like domains(40.Ichijo H. Hellman U. Wernstedt C. Gonez L.J. Claesson-Welsh L. Heldin C.-H. Miyazono K. J. Biol. Chem. 1993; 268: 14505-14513Abstract Full Text PDF PubMed Google Scholar), as seen in P35. Northern blot analysis of mRNA from various human tissues showed that the major transcript of P35 is expressed in liver and is 1.3 kb in length (Fig. 5). Liver also contains three additional transcripts of 3.0, 3.2, and 4.0 kb. Lung and placenta expressed cross-hybridized messages of 1.4 kb in length, which is slightly longer than the major transcript in liver. To examine whether P35 is Ca2+-dependent, P35 was incubated with mannan coated on enzyme-linked immunosorbent assay plates in the presence of calcium ions or EDTA, and P35 binding to mannan was then determined using monoclonal anti-P35 antibody. The results show that P35 bound to mannan in the presence of calcium ions (A405 = 0.22), whereas EDTA prevented this binding (A405 = 0.06). Furthermore, purified P35 bound to mannan-Sepharose in the presence of Ca2+ and was later eluted with EDTA (data not shown). Therefore, it is likely that Ca2+ is required for P35 binding to mannan. P35 specificity for various glycoproteins and neoglycoproteins was evaluated by a binding assay. As shown in Fig. 6, P35 bound to GlcNAc-BSA, one of the neoglycoproteins, but did not bind to the others (Man-BSA, Glc-BSA, Gal-BSA, lactose-BSA, and cellobiose-BSA). It also did not bind to any of the glycoproteins tested, i.e. ovalbumin (OVA) with high mannose-type and hybrid-type oligosaccharides(30.Tai T. Yamashita K. Ito S. Kobata A. J. Biol. Chem. 1977; 252: 6687-6694Abstract Full Text PDF PubMed Google Scholar); bovine pancreatic RNase B (RN) with only high mannose-type oligosaccharides (31.Liang C.-J. Yamashita K. Kobata A. J. Biochem. (Tokyo). 1980; 88: 51-58PubMed Google Scholar); human IgG with di-, mono-, and non-galactosylated biantennary complex-type oligosaccharides(32.Mizuochi T. Taniguchi T. Shimizu A. Kobata A. J. Immunol. 1982; 129: 2016-2020PubMed Google Scholar); human fibrinogen (Fib) with digalactosylated biantennary complex-type oligosaccharides as the major oligosaccharides(33.Mizuochi T. Taniguchi T. Asami Y. Takamatsu J. Okude M. Iwanaga S. Kobata A. J. Biochem. (Tokyo). 1982; 92: 283-293Crossref PubMed Scopus (37) Google Scholar); fetuin (Fet) with triantennary complex-type oligosaccharides as the major oligosaccharides(34.Takasaki S. Kobata A. Biochemistry. 1986; 25: 5709-5715Crossref PubMed Scopus (123) Google Scholar); and human α1-acid glycoprotein (AGP) with tetraantennary complex-type oligosaccharides as the major oligosaccharides(35.Yoshima H. Matsumoto A. Mizuochi T. Kawasaki T. Kobata A. J. Biol. Chem. 1981; 256: 8476-8484Abstract Full Text PDF PubMed Google Scholar). From these results, it appears likely that P35 recognizes GlcNAc. The reason why P35 bound to GlcNAc-BSA but not to human IgG with non-galactosylated (GlcNAc-exposed) oligosaccharides may have been due to steric hindrance by the polypeptide moiety. Another possibility may involve a lack of residue clusters because GlcNAc-BSA, to which P35 bound, contains clustered GlcNAc residues (29 mol of GlcNAc/mol of BSA), whereas the six glycoproteins tested, to which P35 did not bind, contain only one to five N-linked oligosaccharide chains with GlcNAc residues. We next performed a similar binding assay using neoglycolipids containing neutral oligosaccharides derived from the above six glycoproteins. Neoglycolipids were prepared by conjugating neutral oligosaccharides derived from the glycoproteins to dipalmitoylphosphatidylethanolamine. These conjugates were chromatographed on a TLC plate, which was then overlaid with 125I-P35. As shown in Fig. 7, radiolabeled P35 bound to neoglycolipids containing fully galactosylated bi-, tri-, and tetraantennary complex-type oligosaccharides derived from fibrinogen, fetuin, and α1-acid glycoprotein (Fig. 7, lanes d-f, respectively). In addition, P35 bound to neoglycolipids containing di-, mono-, and non-galactosylated biantennary complex-type oligosaccharides of IgG (lane c). However, P35 did not bind to those containing high mannose-type or hybrid-type oligosaccharides derived from ovalbumin and RNase B (lanes a and b, respectively). The observation that P35 bound to IgG oligosaccharides lacking terminal galactose residues (bands indicated by arrows in lane c) and to complex-type oligosaccharides with terminal galactose residues, but not to high mannose-type or hybrid-type oligosaccharides with GlcNAc residues in the trimannosyl core structures, suggests that P35 recognizes the GlcNAc residue that is next to galactose and that is linked to the trimannosyl core in complex-type oligosaccharides. The finding that the binding of P35 to glycoproteins and neoglycolipids was different, although both contain the same oligosaccharide structures, may be ascribed to a difference in clustering of the carbohydrate structures. On natural glycoprotein molecules, unlike neoglycoproteins, oligosaccharide chains are dispersed. However, the oligosaccharides of neoglycolipids are probably clustered on TLC plates coated with silica gel because of the hydrophobic nature of their lipid moiety. Therefore, it was suggested that P35 may prefer clustered GlcNAc residues and clustered complex-type oligosaccharide chains with GlcNAc residues linked to the trimannosyl core. The possibility still remains that oligosaccharides are hidden by the polypeptide moiety or by sialic acid residues linked to galactose residues of oligosaccharide chains in glycoproteins and that they are exposed in neoglycolipids in which the hindering moieties have been removed. To confirm the above results, purified P35 was applied to both GlcNAc-agarose and asialofetuin-Sepharose columns, which were then washed and eluted with GlcNAc. As shown in Fig. 8, P35 bound to both columns and was eluted with GlcNAc. Therefore, P35 is clearly a GlcNAc-binding lectin. The Ra chemotype strain of S. typhimurium possesses a rough core polysaccharide and lacks O-polysaccharides. GlcNAc is present at the nonreducing termini of the Ra polysaccharide. We therefore examined whether P35 is able to bind to TV119, an Ra chemotype strain. As shown in Fig. 9, binding of P35 to TV119 was demonstrated by flow cytometric analysis. P35 binding was abolished by GlcNAc or mannan, indicating that the binding is carbohydrate-specific. On the other hand, P35 was unable to bind to LT2, a smooth-type strain of S. typhimurium that possesses additional O-polysaccharides. The ability of P35 to bind to the Ra chemotype strain of S. typhimurium prompted us to examine whether P35 serves as an opsonin for phagocytosis of these bacteria. S. typhimurium cells (TV119 or LT2) were preincubated with various amounts of P35. PMN were then mixed with the bacteria, and incubation was continued for an additional 20 min. Phagocytosis of S. typhimurium by PMN was determined by microscopy. As shown in Fig. 10, P35 enhanced uptake of TV119 by PMN in a dose-dependent manner, whereas it showed no opsonic activity toward LT2. We isolated and characterized a human serum lectin tentatively named P35 that is capable of binding to yeast mannan-Sepharose and elutes with GlcNAc. Binding studies with neoglycoproteins and neoglycolipids indicate that P35 does not bind to mannose-BSA or to high mannose-type or hybrid-type oligosaccharides, although it binds to GlcNAc in GlcNAc-BSA and complex-type oligosaccharide chains with GlcNAc residues linked to the trimannosyl core. In experiments with mannan-Sepharose, P35 may not have bound to the mannose residue but to GlcNAc-containing oligosaccharides present as contaminants in the yeast mannan used since yeast mannan itself does not contain this type of oligosaccharide(41.Nakajima T. Ballou C.E. J. Biol. Chem. 1974; 249: 7679-7684Abstract Full Text PDF PubMed Google Scholar). The binding specificity of P35 for GlcNAc was also confirmed by affinity chromatography using GlcNAc-agarose and asialofetuin-Sepharose. Purified P35 was able to bind to both columns and eluted with GlcNAc (Fig. 8). P35, MBP, and conglutinin can be classified as serum lectins specific for GlcNAc. Based on binding studies with neoglycolipids, it has been shown that MBP(42.Childs R.A. Drickamer K. Kawasaki T. Thiel S. Mizuochi T. Feizi T. Biochem. J. 1989; 262: 131-138Crossref PubMed Scopus (93) Google Scholar, 43.Childs R.A. Feizi T. Yuen C.-T. Drickamer K. Quesenberry M.S. J. Biol. Chem. 1990; 265: 20770-20777Abstract Full Text PDF PubMed Google Scholar), conglutinin(36.Mizuochi T. Loveless R.W. Lawson A.M. Chai W. Lachmann P.J. Childs R.A. Thiel S. Feizi T. J. Biol. Chem. 1989; 264: 13834-13839Abstract Full Text PDF PubMed Google Scholar), and P35 recognize a GlcNAc next to a terminal galactose in oligosaccharides derived from IgG when this terminal galactose is missing. Unlike P35, MBP and conglutinin are unable to bind to GlcNAc when an adjacent terminal galactose is present. MBP and conglutinin bind to mannose in oligosaccharides derived from glycoproteins such as RNase B and ovalbumin. In contrast, P35 is unable to bind to high mannose-type or hybrid-type oligosaccharides or to mannose-BSA. The lack of binding to hybrid-type oligosaccharides bearing a GlcNAcβ1-Man group may be due to steric hindrance by the outer chain on the Manα1-6Manβ1 side. It can be concluded that P35 shows binding specificity toward GlcNAc and that, unlike MBP and conglutinin, it does not recognize mannose. The precise mechanism by which P35 is able to bind to neoglycolipids is unknown, as is the reason why it does not bind to glycoproteins even though both contain the same oligosaccharide structure. As described under “Results,” however, P35 may prefer clustered GlcNAc residues and clustered complex-type oligosaccharide chains with GlcNAc residues linked to the trimannosyl core. This preference is in accord with the binding of P35 to microorganisms that have a large number of nonreducing terminal GlcNAc residues exposed on their surface, as is the case with bacteria of S. typhimurium strain TV119. On the other hand, P35 was unable to bind to LT2, a smooth-type strain of S. typhimurium that possesses additional O-polysaccharides. These extra O-polysaccharides might render GlcNAc inaccessible to P35 due to steric hindrance. Therefore, we could not exclude the possibility that the GlcNAc residues of the glycoprotein oligosaccharide chains may be hidden by the polypeptide portion or by sialic acid residues that are linked to galactose residues on the these oligosaccharides, resulting in the prevention of a P35 interaction. By searching for sequence homology in a protein data base, we found that the NH2-terminal six-amino acid sequence of P35 completely coincides with that of a 36-kDa protein that was identified as one of three soluble class I human leukocyte antigen proteins in human plasma by immunoprecipitation and immunoblotting using anti-human leukocyte antigen monoclonal antibodies(44.Haga J.A. She J.-X. Kao K.-J. J. Biol. Chem. 1991; 266: 3695-3701Abstract Full Text PDF PubMed Google Scholar). Of the three proteins, the NH2-terminal amino acid sequences of two species were identical to that of cellular human leukocyte antigen, whereas no homology was found for the other 36-kDa protein. It is possible that the 36-kDa protein is identical to P35 and was obtained because of its ability to bind to IgG through complex-type oligosaccharides. cDNA analysis revealed that P35 and porcine ficolin are highly homologous and share collagen- and fibrinogen-like domains. Two types of ficolin with sequence homology, ficolins α and β, have been identified from a porcine uterus cDNA library. The distribution of ficolin in various human tissues has been studied using porcine ficolin probes and mRNAs from these tissues; and ficolin α was shown to be present in lung and placenta, and ficolin β in skeletal muscle (40.Ichijo H. Hellman U. Wernstedt C. Gonez L.J. Claesson-Welsh L. Heldin C.-H. Miyazono K. J. Biol. Chem. 1993; 268: 14505-14513Abstract Full Text PDF PubMed Google Scholar). With Northern blot analysis, we showed that the major P35 transcript of 1.3 kb is expressed in liver. This transcript seems to correspond to P35. Three larger signals that were also detected in liver probably correspond to products of alternative splicing because they were found in the cDNA sequences, and Northern blotting with the intron sequences as a probe showed that the probes were reacted with these three bands, but not with the 1.3-kb band. ( 2Y. Endo, S. Taira, Y. Sato, M. Matsushita, and T. Fujita, manuscript in preparation.) Lung and placenta expressed cross-hybridized messages 1.4 kb in length, which is slightly longer than that in liver, suggesting that these two tissues produce P35-related proteins that may differ in amino acid sequence from the one expressed in liver. It is possible that P35 and P35-related protein might be human analogues of the two types of porcine ficolin. P35 and collectins share structural and functional similarities. They are both Ca2+-dependent lectins as well as hybrid proteins that have collagenous domains with either a fibrinogen-like domain (in P35) or a CRD (in collectins). Amino acid sequences in the CRD responsible for calcium and carbohydrate binding have been thoroughly investigated(11.Weis W.I. Drickamer K. Hendrickson W.A. Nature. 1992; 360: 127-134Crossref PubMed Scopus (856) Google Scholar, 12.Drickamer K. Nature. 1992; 360: 183-186Crossref PubMed Scopus (453) Google Scholar). Although there is no sequence homology between the CRD and the fibrinogen-like domain, P35 has potential calcium-binding sites within the latter. It is therefore reasonable to expect that carbohydrate-binding sites also reside in this domain. It has been reported that fibrinogen-like domains occur in certain proteins such as tenascin and cytotactin (45.Pearson C.A. Pearson D. Shibahara S. Hofsteenge J. Chiquet-Ehrismann R. EMBO J. 1988; 7: 2977-2982Crossref PubMed Scopus (336) Google Scholar, 46.Jones F.S. Hoffman S. Cunningham B.A. Edelman G.M. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 1905-1909Crossref PubMed Scopus (152) Google Scholar) and the scabrous protein of Drosophila melanogaster(47.Baker N.E. Mlodzik M. Rubin G.M. Science. 1990; 250: 1370-1377Crossref PubMed Scopus (209) Google Scholar); however, their functions in these proteins are not yet fully understood. This is the first report demonstrating that one possible function of the fibrinogen-like domain might be to bind to carbohydrates in a calcium-dependent manner. P35 has been demonstrated to bind to TV119, an Ra chemotype strain of S. typhimurium bearing rough core polysaccharides containing a terminal GlcNAc. Binding of P35 to TV119 was carbohydrate-specific and mediated enhancement of phagocytosis by PMN. These activities were not directed to LT2, a smooth-type strain of S. typhimurium. O-Polysaccharides in LT2 might render P35 inaccessible to GlcNAc due to steric hindrance. Collagen domains in collectins are crucial for exhibiting opsonic activities. Upon binding to their ligands, collectins interact with the C1q receptor on phagocytes, resulting in enhancement of phagocytosis. Since P35 is also a lectin containing a collagen-like domain structure, it is possible that in P35, this domain binds to the C1q receptor on PMN. In conclusion, P35 is a human serum Ca2+-dependent lectin specific for GlcNAc. It exhibits a high structural similarity to porcine ficolin and possesses both collagen- and fibrinogen-like domains. P35 also resembles collectins in that the collagen-like domains in both are presumably involved in opsonic activities toward microorganisms. Instead of the CRD found in collectins, however, P35 has a fibrinogen-like domain that could be responsible for carbohydrate binding. P35 is therefore a novel type of serum Ca2+-dependent and GlcNAc-binding lectin with hybrid domains that are probably involved in innate immunity. We thank Dr. William Campbell for English editing and Mihoko Mogi for technical assistance." @default.
• W1996364990 created "2016-06-24" @default.
• W1996364990 creator A5008048202 @default.
• W1996364990 creator A5009306852 @default.
• W1996364990 creator A5021633231 @default.
• W1996364990 creator A5030342357 @default.
• W1996364990 creator A5058817968 @default.
• W1996364990 creator A5065631088 @default.
• W1996364990 creator A5072839852 @default.
• W1996364990 creator A5082731774 @default.
• W1996364990 date "1996-02-01" @default.
• W1996364990 modified "2023-10-14" @default.
• W1996364990 title "A Novel Human Serum Lectin with Collagen- and Fibrinogen-like Domains That Functions as an Opsonin" @default.
• W1996364990 cites W136318563 @default.
• W1996364990 cites W1497265739 @default.
• W1996364990 cites W1521344931 @default.
• W1996364990 cites W1522572101 @default.
• W1996364990 cites W1524114939 @default.
• W1996364990 cites W1549239087 @default.
• W1996364990 cites W1571263856 @default.
• W1996364990 cites W1573617710 @default.
• W1996364990 cites W1590966264 @default.
• W1996364990 cites W1604340849 @default.
• W1996364990 cites W1625540616 @default.
• W1996364990 cites W1794811372 @default.
• W1996364990 cites W1814747440 @default.
• W1996364990 cites W1852243429 @default.
• W1996364990 cites W1889222490 @default.
• W1996364990 cites W194897549 @default.
• W1996364990 cites W1952749171 @default.
• W1996364990 cites W1964871627 @default.
• W1996364990 cites W1984469505 @default.
• W1996364990 cites W1991051424 @default.
• W1996364990 cites W2006679866 @default.
• W1996364990 cites W2013946123 @default.
• W1996364990 cites W2017520388 @default.
• W1996364990 cites W2022798056 @default.
• W1996364990 cites W2033450022 @default.
• W1996364990 cites W2038113812 @default.
• W1996364990 cites W2047774312 @default.
• W1996364990 cites W2054292263 @default.
• W1996364990 cites W2069007382 @default.
• W1996364990 cites W2070894409 @default.
• W1996364990 cites W2071993331 @default.
• W1996364990 cites W2072343511 @default.
• W1996364990 cites W2074497479 @default.
• W1996364990 cites W2100837269 @default.
• W1996364990 cites W2119799948 @default.
• W1996364990 cites W2121783932 @default.
• W1996364990 cites W2136124161 @default.
• W1996364990 cites W2136212260 @default.
• W1996364990 cites W2310509661 @default.
• W1996364990 cites W2332948747 @default.
• W1996364990 cites W2376576252 @default.
• W1996364990 cites W3102457749 @default.
• W1996364990 doi "https://doi.org/10.1074/jbc.271.5.2448" @default.
• W1996364990 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/8576206" @default.
• W1996364990 hasPublicationYear "1996" @default.
• W1996364990 type Work @default.
• W1996364990 sameAs 1996364990 @default.
• W1996364990 citedByCount "294" @default.
• W1996364990 countsByYear W19963649902012 @default.
• W1996364990 countsByYear W19963649902013 @default.
• W1996364990 countsByYear W19963649902014 @default.
• W1996364990 countsByYear W19963649902015 @default.
• W1996364990 countsByYear W19963649902016 @default.
• W1996364990 countsByYear W19963649902017 @default.
• W1996364990 countsByYear W19963649902018 @default.
• W1996364990 countsByYear W19963649902019 @default.
• W1996364990 countsByYear W19963649902020 @default.
• W1996364990 countsByYear W19963649902021 @default.
• W1996364990 countsByYear W19963649902022 @default.
• W1996364990 countsByYear W19963649902023 @default.
• W1996364990 crossrefType "journal-article" @default.
• W1996364990 hasAuthorship W1996364990A5008048202 @default.
• W1996364990 hasAuthorship W1996364990A5009306852 @default.
• W1996364990 hasAuthorship W1996364990A5021633231 @default.
• W1996364990 hasAuthorship W1996364990A5030342357 @default.
• W1996364990 hasAuthorship W1996364990A5058817968 @default.
• W1996364990 hasAuthorship W1996364990A5065631088 @default.
• W1996364990 hasAuthorship W1996364990A5072839852 @default.
• W1996364990 hasAuthorship W1996364990A5082731774 @default.
• W1996364990 hasBestOaLocation W19963649901 @default.
• W1996364990 hasConcept C124594742 @default.
• W1996364990 hasConcept C160448771 @default.
• W1996364990 hasConcept C185592680 @default.
• W1996364990 hasConcept C189413060 @default.
• W1996364990 hasConcept C203014093 @default.
• W1996364990 hasConcept C2779036427 @default.
• W1996364990 hasConcept C2780758401 @default.
• W1996364990 hasConcept C55493867 @default.
• W1996364990 hasConcept C86803240 @default.
• W1996364990 hasConcept C89372155 @default.
• W1996364990 hasConcept C95444343 @default.
• W1996364990 hasConceptScore W1996364990C124594742 @default.
• W1996364990 hasConceptScore W1996364990C160448771 @default.
• W1996364990 hasConceptScore W1996364990C185592680 @default.
• W1996364990 hasConceptScore W1996364990C189413060 @default.

• next