CD11b/CD18-Dependent Interactions of Neutrophils with Intestinal Epithelium Are Mediated by Fucosylated

This information is current as Ke Zen, Yuan Liu, Dana Cairo and Charles A. Parkos of September 29, 2021. J Immunol 2002; 169:5270-5278; ; doi: 10.4049/jimmunol.169.9.5270 http://www.jimmunol.org/content/169/9/5270 Downloaded from

References This article cites 54 articles, 28 of which you can access for free at: http://www.jimmunol.org/content/169/9/5270.full#ref-list-1

Why The JI? Submit online. http://www.jimmunol.org/

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on September 29, 2021

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2002 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

CD11b/CD18-Dependent Interactions of Neutrophils with Intestinal Epithelium Are Mediated by Fucosylated Proteoglycans1

Ke Zen,2 Yuan Liu, Dana Cairo, and Charles A. Parkos

CD11b/CD18-mediated adhesive interactions play a key role in regulating polymorphonuclear leukocytes (PMN)) migration across intestinal epithelium. However, the identity of epithelial ligands for migrating PMN remains obscure. In this study we investigated the role of carbohydrates in mediating adhesive interactions between T84 intestinal epithelial cells and CD11b/CD18 purified from PMN. Fucoidin, heparin/heparin sulfate, N-acetyl-D-glucosamine, mannose-6-phosphate, and laminarin were found to inhibit adhesion of T84 cells to CD11b/CD18. The most potent inhibitory effects were observed with fucoidin (50% inhibition -at 1Ð5 ؋ 10؊8 M). Binding assays demonstrated that fucoidin directly bound to CD11b/CD18 in a divalent cation- and sulfation Downloaded from dependent fashion that was blocked by anti-CD11b mAbs. Experiments employing CD11b/CD18 as a probe to blot T84 cell fucosylated proteins purified via fucose-specific lectin column revealed several candidate CD11b/CD18 binding proteins with molecular masses of 95, 50, 30, 25, and 20 kDa. Fucosidase treatment of T84 cells resulted in significantly reduced cell adhesion to CD11b/CD18, while no inhibition was observed after neuraminidase treatment. Finally, significant inhibition of T84 cell ad- hesion to CD11b/CD18 was observed after blocking cell synthesis with p-nitrophenyl-␤-D-xylopyranoside. These

findings implicate epithelial cell surface proteoglycans decorated with sulfated fucose moieties as ligands for CD11b/CD18 during http://www.jimmunol.org/ PMN migration across mucosal surfaces. The Journal of Immunology, 2002, 169: 5270Ð5278.

pithelial dysfunction and patient symptoms in inflamma- CD18 (6). While the precise localization of the lectin-like domain tory intestinal diseases such as ulcerative colitis and on CD11b is not known, results from a previous study suggested Crohn’s disease correlate with migration of polymorpho- localization between the I domain and C-terminal regions of E 3 nuclear leukocytes (PMN) across epithelial surfaces (1, 2). In CD11b (7). studies modeling the process of PMN migration across mucosal The unique structure of CD11b/CD18 enables it to bind to a surfaces, it has been shown that PMN transepithelial migration is wide variety of proteins and carbohydrates. Previously reported

␤ by guest on September 29, 2021 dependent on the leukocyte 2 integrin CD11b/CD18 (Mac-1, ligands for CD11b/CD18 include proteins such as CR3), but not CD11a/CD18 (3). It is generally believed that ICAM-1 (6, 8) and soluble factors such as iC3b (9, 10), fibrinogen CD11b/CD18 mediates initial adhesion of PMN to epithelial (11), LPS (12, 13), elastase (14), oligodeoxynucleotide (15), zy- monolayers by interactions with a counter-receptor(s) on the epi- mosan (16), ␤-glucan (17), heparin (heparin sulfate) (18, 19), de- thelial cell surface. The nature of these counter-receptors or adhe- natured proteins (20), and a hookworm-derived neutrophil adhe- sive ligands that play an important role in regulating PMN trans- sion inhibitor (21). While ICAM-1 is the only cell membrane ␤ epithelial migration, however, have not been defined. Like other 2 protein ligand that has been reported for CD11b/CD18 and has integrins, CD11b/CD18 contains both an I or an A domain and a been shown to play an important role in PMN transendothelial lectin-like domain that mediate ligand specificity. An ϳ200-aa re- migration, it does not appear to play a role in PMN migration gion comprising the I domain has been extensively characterized, across intestinal epithelium. In particular, ICAM-1 expression is revealing homology to plasma proteins such as von Willebrand induced on the apical membrane of intestinal epithelial cells only factor, complement factor II, and extracellular matrix proteins, in- under certain inflammatory conditions (3, 22). Under these condi- cluding collagen and cartilage matrix protein (4, 5). It is also re- tions apically expressed ICAM-1 is not accessible as a ligand for garded as a major recognition site for several ligands of CD11b/ migrating PMN, because transepithelial migration is dependent on interactions of PMN with the basolateral membrane of intestinal

Division of Gastrointestinal Pathology, Department of Pathology and Laboratory epithelial cells. These observations argue against a role for Medicine, Emory University, Atlanta, GA 30322 ICAM-1 as an adhesive ligand for PMN during transepithelial mi- Received for publication June 19, 2002. Accepted for publication August 30, 2002. gration in the intestine. Previous studies have also suggested that The costs of publication of this article were defrayed in part by the payment of page heparin binds to CD11b/CD18 (18, 19). It was proposed that ep- charges. This article must therefore be hereby marked advertisement in accordance ithelial cell surface proteoglycans decorated with heparin sulfate with 18 U.S.C. Section 1734 solely to indicate this fact. moieties (HSPGs) might serve as adhesive ligands for CD11b/ 1 This work was supported by National Institutes of Health Grant HL54229 (to CD18. However, no such HSPGs have yet been identified. Fur- C.A.P.). thermore, syndecan-1, a proteoglycan that is ex- 2 Address correspondence and reprint requests to Dr. Ke Zen, Department of Pathol- ogy and Laboratory Medicine, Emory University, Whitehead Biomedical Building, pressed primarily on epithelial cells, does not support PMN Room 115, 615 Michael Street, Atlanta, GA 30322. E-mail address: [email protected] adhesion (18). 3 Abbreviations used in this paper: PMN, polymorphonuclear leukocytes; BCECF- Lectin-like properties of CD11b/CD18 have been borne out by AM, 2Ј,7Ј-bis-(2-carboxyethyl)-5-(and -6)-carboxyfluorescein, acetoxymethyl ester; several reports demonstrating carbohydrate binding to CD11b/ biotin-X-NHS, biotin-⑀-aminocaproic acid N-hydroxysuccinimide ester; BSA-F, BSA-fucoidin conjugate; HSPG, heparin sulfate proteoglycan; ␣-xy, p-nitrophenyl- CD18 (7, 16, 18). However, the role of epithelial cell surface car- ␣-D-xylopyranoside; ␤-xy, p-nitrophenyl-␤-D-xylopyranoside. bohydrates in CD11b/CD18-mediated adhesive interactions has

Copyright © 2002 by The American Association of Immunologists, Inc. 0022-1767/02/$02.00 The Journal of Immunology 5271 not been characterized. Previously, it was shown that certain car- characteristic of CD11b and CD18, respectively (not shown). The typical bohydrates, such as mannose-6-phosphate, glucose-6-phosphate, yield of CD11b/CD18 from 1010 PMN ranged from 1 to 2 mg. heparin, and fucoidin, are effective inhibitors of PMN migration across T84 cell monolayers (23). These results strongly suggested T84 cell adhesion to purified CD11b/CD18 that carbohydrates play a role in regulating PMN transepithelial T84 cell adhesion to immobilized, purified CD11b/CD18 was performed as migration. However, the mechanism by which these carbohydrates previously described (25). Briefly, purified CD11b/CD18 (ϳ100 ␮g/ml in regulate PMN transepithelial migration and demonstration of their 150 mM NaCl, 2 mM MgCl2, 2 mM CaCl2, 100 mM Tris, and 1% N-octyl- ␤-D-glucopyranoside, pH 7.4) was diluted 20-fold with HBSS and imme- existence on the epithelial cell surface have not been reported. diately added to 96-well, flat-bottom, microtiter plates (ICN Biomedical, In the present study we tested a wide range of mono- and poly- Aurora, OH; 50 ␮l/well). Microtiter plates were kept at 4¡C overnight to saccharides for inhibition of epithelial T84 cell adhesion to puri- allow for protein binding. Nonspecific protein binding was blocked by fied CD11b/CD18. We demonstrate that several carbohydrates, in- adding 1% BSA in HBSS solution for1hatroom temperature. Trypsin/ EDTA-elicited T84 cells were then washed twice with HBSS and incu- cluding fucoidin, heparin/heparin sulfate, N-acetyl-D-glucosamine, bated with 5 ␮g/ml BCECF-AM (Molecular Probe) in HBSS for 15 min at mannose-6-phosphate, and laminarin, can significantly inhibit the 37¡C. After washing by centrifugation, fluorescence-labeled T84 cells were adhesion of T84 cells to purified CD11b/CD18. We have deter- added to CD11b/CD18-coated plates (ϳ2.5ϫ105 cells/well in a total vol- mined that fucoidin is the most potent inhibitor of cell adhesion ume of 150 ␮l), followed by stationary incubation at 37¡Cfor1htoallow and does so by binding to CD11b/CD18 in a divalent cation-de- for cell adhesion. To quantify adhesion, plates were gently washed three times, and the fluorescence of each well was determined using a fluores- pendent fashion. We also determined the role of cell surface fucose cence microtiter plate reader at excitation/emission wavelengths of 485/ moieties and fucosylated proteoglycans in regulating T84 cell ad- 535 nm (Millipore, Milford, MA). Cell adherence was calculated as the herence to CD11b/CD18. We show that fucosylated proteins de- percentage of total applied cells: 100 ϫ (postwash fluorescence/pre-wash Downloaded from rived from solubilized intestinal epithelial cells via the fucose- fluorescence). In some experiments, T84 cells were pretreated with test specific lectin column contain several candidate CD11b/CD18 compounds/mAbs as indicated before addition to CD11b/CD18-coated wells. binding proteins. Based on these results we conclude that epithelial T84 cell surface proteoglycans decorated with sulfated fucose moi- Biochemical modification of fucoidin eties serve as adhesive ligands for CD11b/CD18 during the trans- Desulfation of fucoidin was achieved by solvolysis in dimethylsulfoxide as migration response. http://www.jimmunol.org/ previously described (27). Briefly, the sodium salt of fucoidin (250 mg) was converted into free acid using Dowex 50 W-X8 (Hϩ) (Sigma), neu- Materials and Methods tralized with pyridine, and lyophilized. The pyridinium salt was then dis-

Cell lines and reagents solved in 10 ml of 10% aqueous Me2SO. The solution was heated at 80¡C for 90 min and cooled in an ice bath. The reaction mixture was diluted to T84 intestinal epithelial cells were grown in a 1/1 mixture of DMEM and 20 ml with water, and the pH was adjusted to 9.0 using 0.1 M NaOH. The Ham’s F-12 medium supplemented with 15 mM HEPES buffer (pH 7.5), ␮ ␮ ␮ mixture was then dialyzed and lyophilized. 14 mM NaHCO3,40 g/ml penicillin, 8 g/ml ampicillin, 90 g/ml strep- Fucoidin (normal and desulfated) was biotinylated using biotin-X-NHS tomycin, and 5% newborn calf serum (Life Technologies, Gaithersburg, Ј Ј after cyanogen bromide activation and aminoalkylation with diaminohex- MD). The fluorescent probe 2 ,7 -bis-(2-carboxyethyl)-5-(and -6)-carboxy- ane. This method has been shown to effectively label polysaccharides while fluorescein acetoxymethyl ester (BCECF-AM) was purchased from Mo- ␣ preserving the polysaccharide structure (28). Briefly, 500 mg of polysac- by guest on September 29, 2021 lecular Probes (Eugene, OR). The glycosidase -L-fucosidase from bovine charide was mixed with 875 mg of cyanogen bromide in 28 ml of 0.5 M kidney (EC 3.2.1.51) and neuraminidase (sialidase) from arthrobacter ure- Na CO solution (pH 11) with vigorous stirring for 30 min. Activated afaciens (EC 3.2.1.18), which cleaves both ␣2,3- and ␣2,6-linked sialic 2 3 ⑀ carbohydrates were de-salted over a Sephadex G-50 column and mixed acid (24), were obtained from Roche (Indianapolis, IN). Biotin- -aminoca- with 750 mg of diaminohexane in 15 ml of 0.4 M NaHCO solution (pH proic acid N-hydroxysuccinimide ester (biotin-X-NHS), heparin (from por- 3 ϭ 9.5). The mixture was gently stirred at 4¡C for 14 h and dialyzed against cine intestinal mucosa; Mr 13,000Ð15,000), and heparan sulfate (from ϭ PBS. After lyophilization, 200 mg of aminoalkyl derivative dissolved in bovine intestinal mucosa; Mr 7, 500) were purchased from Calbiochem 100 ml of 0.1 M NaHCO (pH 8.0) was reacted with 27 mg of biotin-X- (La Jolla, CA). Other mono- and polysaccharides, including fucoidin, dex- 3 ϭ ϭ NHS (Calbiochem). After6hat25¡C the solution was extensively dia- tran (Mr 428,000), dextran sulfate (Mr 500,000), mannose-6-phos- lyzed. The biotinylated aminoalkyl derivatives were recovered by phate, glucose-6-phosphate, chondroitin sulfate C (from shark cartilage), lyophilization. galactose, L-fucose, and laminarin were purchased from Sigma (St. Louis, MO). FITC-labeled lectin from tetragonolobus purpureas, tetragonolobus purpureas immobilized on 4% beaded agarose, and an anion exchanger Fucoidin-CD11b/CD18 binding assay ϩ (Dowex H 50 WX8-100) were also obtained from Sigma. A panel of The binding of fucoidin to CD11b/CD18 was studied using two reciprocal functionally characterized mAbs that had been previously shown to bind to methods. First, to assay binding of biotinylated fucoidin to immobilized the defined regions of the CD11b extracellular domain (18, 25) was used CD11b/CD18, purified CD11b/CD18 was diluted (ϳ20-fold) with HBSS as purified IgG. This panel of mAbs included CBRM1/23 (C domain bind- and bound to 96-well, flat-bottom, microtiter plates as described above. ing, IgG2a), CBRM1/29 (I domain binding, IgG1), CBRM1/31 (I domain After blocking with 2% BSA in HBSS, biotinylated fucoidin (10 ␮g/ml) binding, IgG1), LM2/1 (I domain binding, purified IgG1), and OKM1 (C was added (37¡C, 30 min). Plates were washed three times with HBSS domain binding, IgG2b). A rabbit polyclonal Ab against human CD11b containing 0.5% BSA and then incubated with HRP-conjugated streptavi- R7928A was raised by immunizing rabbit with a peptide (DMMSEGGP- din in HBSS containing 2% BSA for 30 min at room temperature. Poly- PGAEPQ) corresponding to the C terminus of CD11b as previously de- saccharide/CD11b complexes were detected colorimetrically using ABTS. scribed (26). Hybridoma cells secreting a functionally inhibitory mAb Controls included BSA-coated wells or wells without fucoidin (streptavi- against CD18 (TS1/18, IgG1) were purchased from American Type Cul- din alone). As a reciprocal method, CD11/CD18 was assayed for binding ture Collection (Manassas, VA). A protease inhibitor mixture consisting of to immobilized fucoidin. Here, 100 ␮l of fucoidin (5 mg/ml) in HBSS was soybean trypsin inhibitor, benzamidine, leupeptin, pepstatin A, bestatin, added to microtiter wells, followed by incubation for2hat37¡C. After and aprotinin was obtained from Calbiochem. GelCode Blue stain Reagent washing off unbound fucoidin, fucoidin-coated wells were blocked with was obtained from Pierce (no. 24590; Rockford, IL). Nitrocellulose mem- 2% BSA in HBSS for1hatroom temperature. Purified CD11b/CD18 (5 brane and pre-stained m.w. markers were purchased from Bio-Rad (Rich- ␮g/ml) in HBSS containing 0.1% Triton X-100 was added to the fucoidin- mond, CA). All other reagents, unless stated, were obtained from Sigma. coated wells. After 1-h incubation at 37¡C, the plates were washed three Purification of CD11b/CD18 times with HBSS containing 0.1% Triton X-100, followed by incubation with anti-CD18 mAb TS1/18 (1/500 dilution) or polyclonal Ab R7928A Functionally active CD11b/CD18 was purified to homogeneity from large (1/200 dilution) in HBSS containing 2% BSA. After washing away un- quantities of human PMN (ϳ1010 cells) by immunoaffinity chromatogra- bound Ab, wells were incubated with HRP-conjugated secondary Ab, fol- phy using LM2/1-Sepharose, as previously described by Diamond et al. lowed by addition of substrate (ABTS) for color development and OD (8). SDS-PAGE of the purified integrin, followed by Coomassie blue stain- measurement. Wells coated with BSA (no fucoidin) or chondroitin sulfate ϳ ing, revealed two prominent protein bands with Mr of 150 and 95 kDa, C served as controls. 5272 T84 CELL ADHESION TO CD11b/CD18

Neoproteoglycan preparation and CD11b/CD18 binding Blockage of T84 cell proteoglycan synthesis experiments Biosynthesis of proteoglycans was blocked in T84 cells as previously de- A neoproteoglycan derivative of fucoidin was prepared according to the scribed (31). T84 cells in 12-well tissue culture plates (ϳ75% confluence) method described by Matsumoto et al. (29). Briefly, 30 mg of fucoidin was were grown in cell culture medium (1:1 mixture of DMEM and Ham’s dissolved in 4 ml of distilled water and mixed with 30 mg of N-ethoxy- F-12 medium in 5% dialyzed FCS) containing 2 mM p-nitrophenyl-␤-D- ␤ carbonyl-2-ethoxy-1,2-dihydroquinoline in 6 ml of ethanol. After preincu- xylopyranoside ( -xy) overnight (37¡C, 5% CO2). As a control, T84 mono- bation at room temperature for 2 h, 60 mg of BSA was added, and the layers were incubated in cell culture medium containing 2 mM p-nitrophe- mixture was incubated at 4¡C for 2 days with gentle shaking. The mixture nyl-␣-D-xylopyranoside (␣-xy), which does not inhibit biosynthesis of cell was then dialyzed against PBS, followed by lyophilization and storage at proteoglycans. T84 cell monolayers were washed twice with cell culture 4¡C for further use. The formation of neoproteoglycan was confirmed by medium before harvesting, labeling with BCECF-AM, and used in cell SDS-PAGE, demonstrating a shift in the m.w. of BSA under nonreducing adhesion assays. In some experiments cells were also treated with a com- conditions. To test the binding of fucoidin neoproteoglycan to CD11b/ bination of fucosidase and 2 mM ␤-xy. The cells that were grown in the CD18, fucoidin neoproteoglycan (fucoidin-BSA conjugate) and BSA (con- presence of 2 mM ␤-xy were harvested with trypsin/EDTA and further trol) were subjected to SDS-PAGE analysis on a 4Ð15% gradient gel (5 incubated with fucosidase (1.0 U/ml) in PBS containing 2.5 mM EDTA, 1 ␮g/lane). Proteins were directly visualized with silver staining or were mM benzamidine, and 10 ␮g/ml each of PMSF and pepstatin A for 60 min electrophoretically transferred onto nitrocellulose filter (Bio-Rad, Rich- at 37¡C. After washing with cold HBSS, cells were labeled with mond, CA), followed by probing with CD11b/CD18. To probe with BCECF-AM in HBSS and assayed for adhesion to immobilized CD11b/ CD11b/CD18, nitrocellulose filters were first incubated with blocking so- CD18. Trypan blue exclusion was assessed to ensure cell viability. lution (HBSS containing 0.2% Triton X-100 and 1% BSA) for1hatroom temperature. The filters were then incubated with purified CD11b/CD18 Statistics ␮ (final concentration, 5Ð10 g/ml) in blocking solution for2hat37¡C. Data are presented as the mean Ϯ SD and were compared by Student’s After washing filters with HBSS containing 0.1% Triton X-100 (three t test. Downloaded from times for 5Ð10 min each time), nitrocellulose filters were incubated se- quentially with anti-CD18 mAb TS1/18 (1/1000 dilution) and HRP-con- jugated secondary Ab, followed by detection with ECL. All reactions were Results performed in the presence of 2 mM Mg2ϩ and Ca2ϩ. Certain carbohydrates inhibit T84 cell adhesion to immobilized CD11b/CD18 Probing T84 cell fucose residue-containing proteins with We tested a wide variety of carbohydrates for inhibition of epi-

CD11b/CD18 http://www.jimmunol.org/ thelial cell adhesion to purified CD11b/CD18. We previously dem- Isolation of fucosylated proteins from detergent lysate of T84 cells was onstrated that purified CD11b/CD18 immobilized in microtiter performed using a fucose-specific lectin column. Briefly, T84 cells (ϳ4 ϫ 109) were harvested with trypsin/EDTA and lysed with 30Ð35 ml of lysis wells supported binding of 30Ð55% of the total applied T84 cell load after1hofincubation at 37¡C, while Ͻ5% of applied cells buffer containing 100 mM Tris, 150 mM NaCl, 2.5 mM CaCl2, 2.5 mM MgCl2, 1% Triton X-100, and a mixture of proteinase inhibitors, pH 7.3. adhered to wells coated with BSA alone (25). Complete inhibition After a 1-h high speed (45,000 ϫ g) centrifugation, the cell lysate super- of T84 cell adhesion to CD11b/CD18 by mAb CBRM1/29 dem- natant was pumped through a fucose-specific lectin column (tetragonolo- onstrates the specificity of this interaction. Table I represents a bus purpureas) at 8Ð10 ml/h. After extensive washing with HBSS contain- ing 0.1% Triton X-100, the column was then eluted with a solution of 200 summary of the effects of a panel of carbohydrate moieties on T84 ␣ mM Tris, 50 mg/ml -L-fucose, and 0.8% SDS, pH 9.5. The protein col- cell adhesion to CD11b/CD18. As shown, among all carbohydrates by guest on September 29, 2021 lection was immediately neutralized with 1.5 M Tris buffer (pH 7.2), dia- tested, fucoidin potently inhibited adhesion in a concentration-de- lyzed extensively against HBSS using 10-kDa cut-off dialysis bag, and pendent fashion. At a low concentration of 0.1 mg/ml, fucoidin concentrated before further analysis. Purified proteins were analyzed by Ϯ SDS-PAGE (10 ␮g/lane), followed by either direct GelCode Blue staining significantly inhibited cell-integrin adhesion (79.3 6.8% inhibi- or CD11b/CD18 probing after transfer onto nitrocellulose filters as de- tion; p Ͻ 0.001). Adhesion of T84 cells to CD11b/CD18 was also scribed above. significantly reduced by heparin (46.5 Ϯ 8.1% inhibition; p Ͻ 0.03), heparin sulfate (43.2 Ϯ 9.4% inhibition; p Ͻ 0.03), Cell surface deglycosylation experiments N-acetyl-D-glucosamine (40.8 Ϯ 10.2% inhibition; p Ͻ 0.03), Removal of T84 cell surface fucose residues was performed as described mannose-6-phosphate (38.5 Ϯ 6.4% inhibition; p Ͻ 0.03), and by Brennan et al. (24) with slight modifications. Briefly, trypsin/EDTA- elicited T84 cells (5 ϫ 105 cell/ml) were incubated at 37¡C for different lengths of time with fucosidase (0.1, 0.5, and 1.0 U/ml) in PBS containing ␮ 2.5 mM EDTA, 1 mM benzamidine, and 10 g/ml each of PMSF and Table I. Summary of carbohydrate effects on T84 cell-CD11b/CD18 pepstatin A, pH 7.0. Cells were then washed with cold HBSS before adhesiona BCECF-AM labeling with BCECF-AM and cell adhesion assays. To re- move T84 cell surface sialic acid residues, trypsin/EDTA-elicited cells were incubated with neuraminidase (0.1, 0.5, and 1.0 U/ml) in PBS (pH % Inhibition Ϯ SD 7.0) containing 2.5 mM EDTA, 1 mM benzamidine, and 10 ␮g/ml each of PMSF and pepstatin A at 37¡C for different lengths of time. Cells were then Carbohydrate 0.1 mg/ml 1.0 mg/ml 10 mg/ml washed with cold HBSS before BCECF-AM labeling and use in adhesion L-fucose 4.1 Ϯ 2.1 3.9 Ϯ 1.7 4.5 Ϯ 1.7 assays. Controls included cells incubated in the same buffers and condi- Galactose 3.5 Ϯ 1.9 Ϫ4.6 Ϯ 3.6 Ϫ6.4 Ϯ 4.2 tions, but without enzymes. Trypan blue exclusion was assessed to verify Glucose-6-phosphate 1.9 Ϯ 1.2 4.5 Ϯ 2.1 6.9 Ϯ 3.6 cell viability after enzyme treatment. Dextran 3.9 Ϯ 2.8 5.4 Ϯ 3.8 8.6 Ϯ 5.2 Dextran sulfate 2.8 Ϯ 2.1 6.7 Ϯ 4.3 7.3 Ϯ 6.1 Cell surface staining with fucose-specific lectin Chondroitin C 1.8 Ϯ 1.1 4.2 Ϯ 2.2 5.8 Ϯ 2.6 b Cell surface staining with fucose-specific lectin was performed as previ- Heparin 5.8 Ϯ 3.6 10.6 Ϯ 3.9 46.5 Ϯ 8.1 b ously described (30). Trypsin/EDTA elicited T84 cells were washed three Heparin sulfate 4.9 Ϯ 2.1 9.7 Ϯ 2.6 43.2 Ϯ 9.4 b times with cold HBSS and then blocked with HBSS containing 5% FCS for N-acetyl-D-glucosamine 5.5 Ϯ 1.8 11.3 Ϯ 3.5 40.8 Ϯ 10.2 b 1 h at room temperature. Cells (5 ϫ 105/ml) were then incubated with Lamanire 7.6 Ϯ 3.1 13.2 Ϯ 4.7 39.7 Ϯ 11.3 b FITC-conjugated fucose-specific lectin from tetragonolobus purpureas (fi- Mannose-6-phosphate 9.1 Ϯ 3.6 14.6 Ϯ 6.2 38.5 Ϯ 6.4 c c c nal concentration, 20 ␮g/ml) in HBSS containing 5% FCS for 30 min at Fucoidin 79.3 Ϯ 6.8 94.6 Ϯ 8.4 97.1 Ϯ 7.1 c room temperature. Cells were washed three times with HBSS, shortly fixed CBRM 1/29 (5 ␮g/ml) 98.9 Ϯ 4.1 with 1% paraformaldehyde (1Ð5 min), washed twice with HBSS, mounted a Data are presented as the percent inhibition by each carbohydrate Ϯ SD de- on slides, and observed under a fluorescence microscope. In this experi- scribed in Materials and Methods. Results are from three separate experiments. ment T84 cells were also treated with 1.0 U/ml fucosidase at 37¡Cfor1h b p Ͻ 0.03. before labeling with FITC-lectin. c p Ͻ 0.001. The Journal of Immunology 5273 laminarin (39.7 Ϯ 11.3%; p Ͻ 0.03), but only at high concentra- tion (10 mg/ml). However, no significant inhibition was found in the presence of glucose-6-phosphate, dextran, dextran sulfate, chondroitin sulfate C, L-fucose, or galactose. Unlike fucoidin, which is a polymer of sulfated L-fucose, the monosaccharide L- fucose had no effect on T84 cell adhesion to CD11b/CD18. As shown in Fig. 1, 50% inhibition of T84 cell binding to CD11b/ CD18 was observed at fucoidin concentrations of 1Ð5 ϫ 10Ϫ8 M. To determine whether the observed inhibitory effects on adhe- sion were the result of fucoidin binding to CD11b/CD18, trypsin/ EDTA-elicited T84 cells or microtiter wells containing immobi- lized CD11b/CD18 were preincubated with fucoidin as detailed in Materials and Methods. After washing away unbound fucoidin FIGURE 2. Fucoidin-mediated inhibition of T84 cell adhesion to CD11b/CD18 occurs through direct interactions with CD11b/CD18. Be- with HBSS, cell adhesion assays were performed in the absence of fore cell adhesion assays, T84 cells (FϩT84 cells) or CD11b/CD18-coated fucoidin. As shown in Fig. 2, preincubation of T84 cells with fu- microtiter wells (FϩCD11b/CD18) were preincubated with 10 mg/ml of coidin did not affect cell adhesion to CD11b/CD18 whereas pre- fucoidin for 30 min at 37¡C, followed by washing. Cell adhesion assays incubation of immobilized CD11b/CD18 with fucoidin strongly were then performed in the absence of fucoidin as detailed in Materials and reduced the T84 cell adhesion. These results suggest that fucoidin Methods. Control wells (no treatment) represent T84 cell binding to inhibits T84 cell adhesion to CD11b/CD18 through direct interac- CD11b/CD18 in the absence of fucoidin pretreatment, whereas blank rep- Downloaded from tions with CD11b/CD18. resents binding of T84 cells to wells coated with BSA only. Values are the mean Ϯ SD of three separate experiments, each performed in triplicate. Fucoidin binds directly to CD11b/CD18 The cell adhesion results suggested the possibility of binding in- been mapped to various domains on the extracellular portion of teractions between fucoidin and CD11b/CD18, and additional ex- CD11b (18, 25) and in the absence of divalent cations. As shown periments were performed to examine this possibility. The binding http://www.jimmunol.org/ in Fig. 4, the binding of biotinylated fucoidin to CD11b/CD18 was of fucoidin to CD11b/CD18 was investigated using two reciprocal largely blocked by anti-CD11b Abs with epitopes on the mem- binding assays: biotinylated fucoidin binding to immobilized brane-proximal, C-terminal region of CD11b (CBRM1/23 and CD11b/CD18 and CD11b/CD18 binding to immobilized fucoidin. OKM1) as well as anti-CD11b Abs with epitopes residing within For the first assay fucoidin was biotinylated (28) and added to the I domain and amino-terminal region (CBRM1/29, CBRM1/31, CD11b/CD18-coated microtiter wells. After incubation at 37¡C for 1 h, bound fucoidin was detected with HRP-streptavidin and color development. We observed that biotinylated fucoidin bound to CD11b/CD18, but not to BSA (Fig. 3A). In the reciprocal assay microtiter wells were first coated with unlabeled fucoidin, fol- by guest on September 29, 2021 lowed by addition of purified CD11b/CD18 in buffer containing Mg2ϩ and detergent. CD11b/CD18 binding was then detected with anti-CD18 Ab (TS1/18), HRP-conjugated goat anti-mouse second- ary Ab, and color development. As shown in Fig. 3B, CD11b/ CD18 bound to fucoidin-coated wells; in contrast, there was little binding of CD11b/CD18 to either BSA-coated wells or chon- droitin sulfate C-coated wells. To further characterize the binding properties of fucoidin to CD11b/CD18, binding assays were performed using of a panel of well-characterized anti-CD11b/CD18 mAbs that had previously

FIGURE 3. Binding of fucoidin to CD11b/CD18. A, Binding of biotin- ylated fucoidin (B-fucoidin) to immobilized CD11b. As detailed in Mate- rials and Methods, microtiter wells coated with CD11b/CD18 were blocked with BSA, followed by addition of 100 ␮g/ml B-fucoidin for 1 h at 37¡C. B-fucoidin binding was assessed after addition of HRP-streptavi- din, followed by color development and OD measurement (405 nm). B, Binding of CD11b/CD18 to immobilized fucoidin and chondroitin sulfate C (control). Polysaccharides (5 mg/ml in HBSS) were added to wells (50 ␮l/well, 37¡C, 2 h), followed by blocking with BSA and addition of CD11b/CD18 (2.5 ␮g/ml, 37¡C, 1 h). CD11b/CD18 binding to fucoidin FIGURE 1. Concentration-dependent inhibition of T84 cell adhesion to was assessed after addition of mAb TS1/18, followed by incubation with CD11b/CD18 by fucoidin. As detailed in Materials and Methods, BCECF- HRP-conjugated secondary mAb, color development, and OD measure- labeled T84 cells were incubated with CD11b/CD18-coated wells for 1 h ment (405 nm). Binding assays were performed in the presence of 2 mM at 37¡C in the presence of polysaccharides. T84 cell adhesion is expressed Ca2ϩ and Mg2ϩ. BSA-coated wells served as controls in both experiments. as a percentage of the total applied cells. Values are the mean Ϯ SD of Values are the mean Ϯ SD of three separate experiments, each performed three separate experiments, each performed in triplicate. in triplicate. 5274 T84 CELL ADHESION TO CD11b/CD18

Sulfation of fucoidin determines binding to CD11b/CD18 Sulfation of polysaccharides has been shown to be important for a number of glycosylated proteins (proteoglycans) interacting with integrins (32Ð34). Thus, experiments were performed to examine whether sulfation of fucoidin plays a role in binding to CD11b. For these experiments fucoidin was chemically desulfated, followed by biotinylation before use in binding assays with CD11b/CD18. The desulfation method we used in the experiments did not destroy the polymer structure of fucoidin and resulted in 75%Ð90% desulfa- tion of fucoidin polysaccharide. As shown in Fig. 6A, the binding FIGURE 4. Effects of anti-CD11b/CD18 mAbs and EDTA on the bind- of desulfated fucoidin to immobilized CD11b/CD18 was dimin- ing of fucoidin to CD11b/CD18. Binding assays of biotinylated fucoidin ished by 65% compared with that observed with untreated fucoi- (B-fucoidin, 100 ␮g/ml) to immobilized CD11b/CD18 were performed as din. In Fig. 6B, desulfated fucoidin was also tested for inhibition of detailed in Fig. 3, with the inclusion of CD11b/CD18 mAbs (40 ␮g/ml) or 5 mM EDTA. Values are the mean Ϯ SD of three separate experiments, T84 cell adhesion to purified CD11b/CD18. As shown, binding of each performed in triplicate. T84 cells to CD11b/CD18 in the presence of 10 mg/ml of desul- fated fucoidin was only mildly inhibited (ϳ25% decrease), whereas binding in the presence of untreated fucoidin was reduced by 95%. Downloaded from and LM2/1). Interestingly, mAb TS1/18, which binds to CD18, did not inhibit fucoidin-CD11b/CD18 binding. These results suggest Fucosidase treatment of T84 cells diminishes adhesion to that the fucoidin binding site is on CD11b, but is not localized to CD11b/CD18 a specific domain, as defined by previously characterized mAbs. To determine whether epithelial cell surface fucosylation plays a We also observed that the binding of fucoidin to CD11b/CD18 was role in adhesion to CD11b/CD18, T84 cells were treated with fu- abolished by 5 mM EDTA, indicating that such binding is depen- cosidase and assayed for adhesion. For these experiments trypsin/ dent on divalent cation (Fig. 4). EDTA-elicited T84 cells were labeled with BCECF, followed by http://www.jimmunol.org/ incubation with different amounts of fucosidase at 37¡C. As a con- Fucosylated neoproteoglycans bind to CD11b/CD18 trol, parallel experiments were performed in which cells were in- A modified Western blotting protocol was employed using purified cubated with neuraminidase to remove cell surface sialic acid res- CD11b/CD18 as a probe to evaluate binding to synthetic fucoidin- idues. The enzyme-treated cells were still 95% viable based on containing neoproteoglycans. For these experiments a BSA-fucoi- trypan blue exclusion 2 h after enzymatic treatments. Fucosidase din conjugate (BSA-F) was prepared as previously described (29) treatment significantly reduced T84 cell adhesion to CD11b/CD18 with minor modifications. SDS-PAGE of BSA-F demonstrated a

large shift in the molecular mass of the complex to a higher ap- by guest on September 29, 2021 parent molecular mass on a 4Ð15% gradient gel (Fig. 5A, lane 1) compared with normal BSA alone (Fig. 5A, lane 2). Fig. 5B rep- resents a modified Western blot in which samples of BSA-F (lane 1) and BSA (lane 2) were subjected to SDS-PAGE and transferred to nitrocellulose filters, followed by incubation with CD11b/CD18 in the presence of 2 mM Mg2ϩ and detergent at room temperature. Bound CD11b/CD18 was detected by probing washed blots with anti-CD18 mAb (TS1/18) and HRP-conjugated secondary Ab, fol- lowed by ECL. As shown in Fig. 5B, a strong CD11b/CD18 bind- ing signal was detected in BSA-F (lane 1), but was completely absent in the BSA control (lane 2).

FIGURE 6. Fucoidin binding to CD11b/CD18 is dependent on sulfation of polysaccharide. A, Binding of fucoidin or desulfated-fucoidin (De-fu- coidin) to immobilized CD11b/CD18. Fucoidin was biotinylated and des- ulfated as detailed in Materials and Methods. Biotinylated polysaccharide (100 ␮g/ml) was incubated with CD11b/CD18-coated wells or blanks (no FIGURE 5. Fucoidin neoproteoglycan binds to CD11b/CD18. As de- CD11b/CD18) for1hat37¡C, followed by addition of HRP-streptavidin tailed in Materials and Methods, a neoproteoglycan derivative of fucoidin and color development. B, Effect of fucoidin or De-fucoidin on T84 cell (F-BSA) was prepared for this experiment. Purified F-BSA (5 ␮g/lane; adhesion to CD11b/CD18. Suspensions of BCECF-labeled T84 cells (50 lane 1) and BSA (lane 2) were separated by nonreducing SDS-PAGE (4Ð ␮l, ϳ106 cells/ml) were incubated with CD11b/CD18-coated wells or 15% gel). The proteins were either visualized by silver staining (A)or blanks (no CD11b/CD18) for1hat37¡C in the presence or the absence of transferred to nitrocellulose filters and probed with purified CD11b/CD18 10 mg/ml polysaccharides as indicated. The control represents binding of (B), followed by incubation with mAb TS1/18 and HRP-conjugated sec- T84 cells to CD11b/CD18 in the absence of polysaccharide. Values are the ondary mAb, and visualized by ECL. All buffers contained 2 mM Mg2ϩ. mean Ϯ SD of three separate experiments, each performed in triplicate. The Journal of Immunology 5275

FIGURE 7. Effect of fucosidase on epithelial cell adherence to CD11b/CD18. As detailed in Materials and Methods, trypsin/EDTA-elicited T84 cells

were pretreated with fucosidase or neuraminidase (0, 0.1, 0.5, and 1.0 U/ml, respectively) for 30 min at 37¡C, washed twice with HBSS and used in adhesion Downloaded from assays (A). B, Cells were pretreated with fucosidase (0.5 U/ml) or neuraminidase (0.5 U/ml) at 37¡C for different lengths of time as indicated, followed by washing with HBSS, and were used in CD11b/CD18 adhesion assays. Values are the mean Ϯ SD of three separate experiments, each performed in triplicate. in a concentration- and time-dependent fashion (Fig. 7). In con- no decrease in labeling with FITC-wheat germ agglutinin after trast, no significant inhibition was observed after pretreatment of fucosidase treatment, nor was tetragonolobus purpureas labeling http://www.jimmunol.org/ cells with neuraminidase. These results suggest that T84 cells con- altered by neuraminidase treatment (data not shown). tain cell surface fucose-like structures that partially mediate bind- ing to CD11b/CD18. Inhibition of proteoglycan synthesis reduces T84 cell adhesion Labeling experiments were also performed to examine the effect to CD11b/CD18 of fucosidase treatment on binding of fucose-specific lectin (tet- Since the above results and those reported previously (35, 36) con- ragonolobus purpureas) to T84 cells. Suspensions of T84 cells firm the existence of fucose residues on the epithelial cell surface, were treated with fucosidase (1.0 U/ml, 60 min), followed by fix- we reasoned that inhibition of the synthesis of cell surface proteins ation and staining with FITC-labeled lectin (Fig. 8). Untreated T84 decorated with such residues might influence adhesive interactions by guest on September 29, 2021 cells were strongly labeled with the lectin (Fig. 8B). In contrast, with CD11b/CD18. Candidate cell surface molecules containing fucosidase treatment demonstrated a dramatic reduction in labeling fucose moieties include glycolipids and proteoglycans that are dec- of T84 cells with FITC-lectin (Fig. 8D). As in controls, there was orated with sulfated fucose polymers. To inhibit proteoglycan syn- thesis, T84 cell monolayers were cultured overnight in the pres- ence of 2 mM ␤-xy, followed by labeling and use in cell adhesion assays. As a control, parallel incubations were performed in the presence of 2 mM ␣-xy, which does not inhibit proteoglycan syn- thesis. In both conditions cell viability was Ͼ95% after treatment as assessed by trypan blue exclusion. As shown in Fig. 9, treatment of T84 cells with ␣-xy had no significant effect on cell adhesion to CD11b/CD18 compared with untreated cells (52.5 Ϯ 6.7 vs 51.7 Ϯ 5.9% of total applied cells adhering to CD11b/CD18 for ␣-xy vs

FIGURE 8. Fluorescence staining of T84 cell surface for fucosylated structures. As detailed in Materials and Methods, trypsin/EDTA-elicited T84 cells were incubated with (C and D) or without (A and B) fucosidase FIGURE 9. Inhibition of T84 cell proteoglycan synthesis reduces cell (1.0 U/ml). Cells were then washed three times with cold blocking buffer adhesion to CD11b/CD18. As detailed in Materials and Methods, T84 cells (HBSS with 1% BSA) and incubated with FITC-labeled lectin from tet- were cultured in the presence of 2 mM ␤-xyl or 2 mM ␣-xyl in DMEM ragonolobus purpureas (25 ␮g/ml) on ice for 30 min. After a short fixation overnight, followed by harvest and BCECF labeling for cell adhesion as- with paraformaldehyde, cells were then mounted on glass slides and visu- says. In a subset of experiments T84 cells pretreated with ␤-xyl were fur- alized under a fluorescence microscope. A and C, Phase contrast images of ther incubated with fucosidase (1.0 U/ml) for 30 min at 37¡C before ad- the fluorescence photomicrographs of cells in B and D, respectively. hesion assays. Values are the mean Ϯ SD of three separate experiments, .p Ͻ 0.01 ,ء .Bars ϭ 50 ␮m. each performed in triplicate 5276 T84 CELL ADHESION TO CD11b/CD18

tion (23). Since these experiments consisted of cell-cell adhesion and transmigration assays, the molecular mechanisms of the car- bohydrate-mediated inhibition observed were not clear. The present study was undertaken to better define the mechanism of carbohydrate-mediated inhibition of transepithelial migration and the nature of epithelial ligands for migrating PMN. Here, using assays of T84 cell adhesion to highly purified CD11b/CD18, we screened a wide range of mono- and polysaccharides and identified several carbohydrates that significantly inhibit such adhesive in- teractions. We found that the sulfated, L-fucose-rich polysaccha- ride fucoidin is the most potent inhibitor (Ͼ90% inhibition) and Ϫ FIGURE 10. Identification of fucosylated CD11b/CD18 binding pro- produces 50% inhibition at a concentration of 1Ð5.0 ϫ 10 8 M. teins in T84 cells. Fucosylated proteins were isolated from T84 cell lysate Using a number of binding assays, we determined that fucoidin via a fucose-specific lectin column as detailed in Materials and Methods. binds directly and specifically to CD11b/CD18. ␮ Fucosylated proteins were analyzed by 12% SDS-PAGE (10 g/lane) and The CD11b/CD18 binding properties of fucoidin are similar to stained with GelCode Blue (lane 1). In parallel, proteins were transferred those of L-selectin, which has also been shown to bind fucoidin to nitrocellulose filters and probed with CD11b/CD18 (lane 3), followed by (40, 41). In particular, fucoidin binds to the lectin-like domain of incubation with mAb TS1/18 and HRP-conjugated secondary mAb, and were visualized by ECL. All buffers in this experiment contained 2 mM L-selectin with properties characteristic of C-type lectin-carbohy- Downloaded from Mg2ϩ. Lane 2, Control nitrocellulose filter performed as described for lane drate binding interactions. CD11b also has a lectin-like domain 3, but without incubation with CD11b/CD18. that is most likely present between the I domain and C-terminal regions of the extracellular domain (7, 16). Our results indicate that the binding of fucoidin to CD11b/CD18 is divalent cation dependent, which is a major binding property of C-type lectins. no drug treatment, respectively). In contrast, adhesion of ␤-xy- The similarity of carbohydrate binding patterns between CD11b/

treated T84 cells to purified CD11b/CD18 was significantly re- http://www.jimmunol.org/ CD18 and selectin have also been demonstrated in other cases. For duced compared with that of ␣-xy-treated cells (33.4 Ϯ 4.8 vs 52.5 Ϯ 6.7% of applied cells adhering to CD11b/CD18 for ␤-xy vs example, heparin and complement factor H have been shown as ␣-xy, respectively). Interestingly, incubation of ␤-xy-treated T84 ligands for both integrins (13, 18, 19, 42, 43). We have attempted cells with fucosidase (1.0 U/ml, 60 min, 37¡C) did not result in to map the fucoidin binding site on CD11b/CD18 using anti- further significant inhibition of adhesion to CD11b/CD18. CD11b mAbs with defined epitopes. Since fucoidin binding to CD11b/CD18 was strongly inhibited by I domain and C-terminal Identification of fucose-containing proteins from T84 cells that mAbs (Fig. 4), it is likely that fucoidin binds to a range of sites on bind to CD11b/CD18 CD11b/CD18. Interestingly, in assays of PMN transepithelial mi-

Experiments were performed to identify fucosylated protein(s) in gration and T84 cell adhesion to purified CD11b/CD18, it is clear by guest on September 29, 2021 T84 cells that bind to CD11b/CD18. As detailed in Materials and that the I domain is the most critical functional binding domain Methods, fucose-containing proteins were first purified from T84 (25). Furthermore, mAb TS1/18, which binds to CD18, strongly cells using a fucose-specific lectin column (tetragonolobus purpu- inhibits T84 cell binding to CD11b/CD18, but has no effect on reas immobilized on 4% beaded agarose). The mixture of fucosy- fucoidin binding to CD11b/CD18, suggesting that the fucoidin lated proteins was dialyzed, concentrated, and separated by SDS- binding site is localized on CD11b. Despite these differences, fu- PAGE (12% gel concentration). Proteins were also transferred to coidin is a potent inhibitor of PMN transepithelial migration (23). nitrocellulose filters for our modified Western blot analysis (prob- The binding of fucoidin to CD11b/CD18 was determined to be ing filters with purified CD11b/CD18 protein in the presence of strongly dependent on the degree of sulfation and polymer struc- Mg2ϩ and detergent). In our experiment, ϳ800 ␮g of total fuco- ture. As shown in Table I, the monosaccharide L-fucose does not sylated proteins were isolated from ϳ4 ϫ 109 T84 cells. As shown affect cell-CD11b/CD18 adhesion, nor does D-fucose even at con- centrations as high as 20 mM (data not shown). The functional in Fig. 10, CD11b/CD18 recognized several major bands at Mr of 95, 50, 30, 25, and 20 kDa (lane 3, arrowheads). Those protein activity of fucoidin in our CD11b/CD18 binding assays was sig- bands were absent in the control experiment in which no CD11b/ nificantly reduced by desulfation, suggesting a critical role of sul- CD18, but only anti-CD18 mAb and HRP-conjugated secondary fation in fucoidin-CD11b/CD18 interactions. One explanation for mAb, was used (Fig. 10, lane 2). The same amount of all fucosy- this observation would be that basic amino acid residues in CD11b lated proteins from T84 cells was displayed by staining with play a role in binding to sulfated fucans, as reported for other GelCode Blue (Fig. 10, lane 1). adhesion proteins (44). However, other sulfated polysaccharides, such as chondroitin sulfate C and dextran sulfate, with comparable Discussion charge densities and m.w. did not inhibit T84 cell binding to Transepithelial migration of PMN is a pathological hallmark of CD11b/CD18. These observations suggest that sulfation of fucoi- disease states such as ulcerative colitis, Crohn’s disease, and in- din may play an important role in modifying or maintaining the ␤ fectious enterocolitis. It has been well established that the 2 in- structure necessary for binding to CD11b/CD18 instead of simple tegrin CD11b/CD18 plays a key role in mediating the initial stage charge contributions. This hypothesis is further supported by our of transepithelial migration consisting of PMN adhesion to the finding that neuraminidase treatment of T84 cells, which results in basolateral epithelial surface (25, 37Ð39). In the present study we removal of negatively charged sialic acid residues, did not reduce have further characterized the interaction of carbohydrates with adhesion to CD11b/CD18. Interestingly, CD18 contains a basic CD11b/CD18 and demonstrated a role of fucosylated proteogly- amino acid domain that has been implicated in carbohydrate bind- cans in CD11b/CD18-mediated epithelial cell adhesion. ing interactions (45). While we only tested one CD18-reactive In a previous study we reported that several sulfated polysac- mAb TS1/18, we failed to detect inhibition with this function charides, including fucoidin, inhibited PMN transepithelial migra- blocking mAb. The Journal of Immunology 5277

As a ligand for L-selectin (40, 41) or the scavenger receptor (46, integrin Mac-1 (CD11b/CD18) for four distinct adhesion ligands. J. Cell Biol. 47), fucoidin has been implicated in a wide range of physiological 120:1031. 7. Xia, Y., V. Vetvicka, J. Yan, M. Hanikyrova, T. Mayadas, and G. D. Ross. 1999. events, including lymphocyte tethering and rolling (48) and phago- The ␤-glucan-binding lectin site of mouse CR3 (CD11b/CD18) and its function cytosis (47, 49). A role for fucoidin in cell adhesion has also been in generating a primed state of the receptor that mediates cytotoxic activation in reported. In particular, bovine sperm binding to oviductal epithe- response to iC3b-opsonized target cells. J. Immunol. 162:2281. 8. Diamond, M. S., D. E. Staunton, A. R. de Fougerolles, S. A. Stacker, lium has been shown to be dependent on fucose residues (50) and J. Garcia-Aguilar, M. L. Hibbs, and T. A. Springer. 1990. ICAM-1 (CD54): a can be abolished by treatment of the epithelium with fucosidase counter-receptor for Mac-1 (CD11b/CD18). J. Cell Biol. 111:3129. (36). Fucoidin and other sulfated polysaccharides have also been 9. Bilsland, C. A., M. S. Diamond, and T. A. Springer. 1994. The leukocyte integrin p150,95 (CD11c/CD18) as a receptor for iC3b: activation by a heterologous ␤ shown to be involved in sperm-egg interactions in birds (30). In subunit and localization of a ligand recognition site to the I domain. J. Immunol. vitro studies of human cells have shown that sulfated polysaccha- 152:4582. rides, including fucoidin, can inhibit lymphocyte-to-epithelial 10. Beller, D. I., T. A. Springer, and R. D. Schreiber. 1982. Anti-Mac-1 selectively inhibits the mouse and human type three complement receptor. J. Exp. Med. transmission of HIV-1 (51, 52). Furthermore, others have reported 156:1000. that adhesion of both T and B lymphocytes to cultured fibroblasts 11. Altieri, D. C., F. R. Agbanyo, J. Plescia, M. H. Ginsberg, T. S. Edgington, and was strongly inhibited by fucoidin, whereas fucose and mannan E. F. Plow. 1990. A unique recognition site mediates the interaction of fibrinogen with the leukocyte integrin Mac-1 (CD11b/CD18). J. Biol. Chem. 265:12119. had no effect (53). These results are in agreement with our obser- 12. White, S. R., D. R. Dorscheid, K. F. Rabe, K. R. Wojcik, and K. J. Hamann. 1999. vations and support the functional relevance of fucoidin in epithe- Role of very late adhesion integrins in mediating repair of human airway epi- lial cell interactions with CD11b/CD18. It is possible that such thelial cell monolayers after mechanical injury. Am. J. Respir. Cell Mol. Biol. 20:787. interactions might be exploited therapeutically. For example in an- 13. DiScipio, R. G., P. J. Daffern, I. U. Schraufstatter, and P. Sriramarao. 1998. imal models, cellular infiltrates associated with meningitis can be Human polymorphonuclear leukocytes adhere to complement factor H through an Downloaded from ␣ ␤ reduced by infusion of fucoidin (22). Whether fucoidin could be interaction that involves M 2 (CD11b/CD18). J. Immunol. 160:4057. 14. Cai, T. Q., and S. D. Wright. 1996. Human leukocyte elastase is an endogenous used to inhibit neutrophil transepithelial migration in ulcerative ␣ ␤ ligand for the integrin CR3 (CD11b/CD18, Mac-1, M 2) and modulates poly- colitis or Crohn’s disease remains to be determined. morphonuclear leukocyte adhesion. J. Exp. Med. 184:1213. While the nature of fucoidin-like epithelial counter-receptors for 15. Benimetskaya, L., J. D. Loike, Z. Khaled, G. Loike, S. C. Silverstein, L. Cao, J. el Khoury, T. Q. Cai, and C. A. Stein. 1997. Mac-1 (CD11b/CD18) is an CD11b/CD18 remains undefined, candidate structures have been oligodeoxynucleotide-binding protein. Nat. Med. 3:414. reported on epithelial cell surface (35, 54). In particular, there are 16. Ross, G. D., J. A. Cain, and P. J. Lachmann. 1985. Membrane complement http://www.jimmunol.org/ a number of structures on the epithelial cell surface that contain receptor type three (CR3) has lectin-like properties analogous to bovine conglu- tinin as functions as a receptor for zymosan and rabbit erythrocytes as well as a fucose residues, including glycolipids, membrane receptor for iC3b. J. Immunol. 134:3307. with terminal fucosylation, and cell surface proteoglycans deco- 17. Yan, J., V. Vetvicka, Y. Xia, A. Coxon, M. C. Carroll, T. N. Mayadas, and ␤ rated with fucose or sulfated fucose sugar chains. Our data suggest G. D. Ross. 1999. -Glucan, a “specific” biologic response modifier that uses antibodies to target tumors for cytotoxic recognition by leukocyte complement that intestinal epithelial cell surface proteoglycans decorated with receptor type 3 (CD11b/CD18). J. Immunol. 163:3045. sulfated fucose polymers play an important role in CD11b/CD18- 18. Diamond, M. S., R. Alon, C. A. Parkos, M. T. Quinn, and T. A. Springer. 1995. mediated adhesive interactions. This is consistent with our finding Heparin is an adhesive ligand for the leukocyte integrin Mac-1 (CD11b/CD1). J. Cell Biol. 130:1473. that blockage of biosynthesis of epithelial cell proteoglycans by 19. Peter, K., M. Schwarz, C. Conradt, T. Nordt, M. Moser, W. Kubler, and C. Bode. ␤-xy reduced T84 cell adhesion to CD11b/CD18 by the same 1999. Heparin inhibits ligand binding to the leukocyte integrin Mac-1 (CD11b/ by guest on September 29, 2021 ␣ CD18). Circulation 100:1533. amount observed after treatment with -fucosidase (Fig. 9). While ␣ ␤ 20. Davis, G. E. 1992. Affinity of integrins for damaged extracellular matrix: v 3 the exact identities of such proteoglycans are not known, we have binds to denatured collagen type I through RGD sites. Biochem. Biophys. Res. identified several candidate fucosylated proteins that can bind to Commun. 182:1025. 21. Rieu, P., T. Ueda, I. Haruta, C. P. Sharma, and M. A. Arnaout. 1994. The A- CD11b/CD18. Due to the material (whole cell lysates from epi- ␤ domain of 2 integrin CR3 (CD11b/CD18) is a receptor for the hookworm- thelial cells) we used in purification, it is possible that some of derived neutrophil adhesion inhibitor NIF. J. Cell Biol. 127:2081. these fucosylated proteins recognized by CD11b/CD18 are not ex- 22. Chow, J., R. B. Hartley, C. Jagger, and S. A. Dilly. 1992. ICAM-1 expression in pressed on the cell surface. Further characterization, including use renal disease. J. Clin. Pathol. 45:880. 23. Colgan, S. P., C. A. Parkos, D. McGuirk, H. R. Brady, A. A. Papayianni, of purified epithelial cell plasma membranes or selective cell sur- G. Frendl, and J. L. Madara. 1995. Receptors involved in carbohydrate binding face labeling with tagged lectin as starting material, will facilitate modulate intestinal epithelial-neutrophil interactions. J. Biol. Chem. 270:10531. identification of the subset of epithelial cell surface proteoglycans 24. Brennan, J., F. Takei, S. Wong, and D. L. Mager. 1995. Carbohydrate recognition by a natural killer cell receptor, Ly-49C. J. Biol. Chem. 270:9691. that serves as adhesive ligand for CD11b/CD18. 25. Balsam, L. B., T. W. Liang, and C. A. Parkos. 1998. Functional mapping of CD11b/CD18 epitopes important in neutrophil-epithelial interactions: a central role of the I domain. J. Immunol. 160:5058. Acknowledgments 26. Liu, Y., D. Merlin, S. L. Burst, M. Pochet, J. L. Madara, and C. A. Parkos. 2001. We thank Susan Voss for her help with the T84 and H29 cell culture. The role of CD47 in neutrophil transmigration. Increased rate of migration cor- relates with increased cell surface expression of CD47. J. Biol. Chem. 276:40156. 27. Nagasawa, K., Y. Inoue, and T. Kamata. 1977. Solvolytic desulfation of glyco- References saminoglycuronan sulfates with dimethyl sulfoxide containing water or methanol. 1. Kumar, N. B., T. T. Nostrant, and H. D. Appelman. 1982. The histopathologic Carbohydrate Res. 58:47. spectrum of acute self-limited colitis (acute infectious-type colitis). Am. J. Surg. 28. Debbage, P. L., W. Lange, T. Hellmann, and H. J. Gabius. 1988. Detection of Pathol. 6:523. receptors for sulfated polysaccharides in human placenta by biotinylated probes. 2. Hawker, P. C., J. S. McKay, and L. A. Turnberg. 1980. Electrolyte transport J. Histochem. Cytochem. 36:1097. across colonic mucosa from patients with inflammatory bowel disease. Gastro- 29. Matsumoto, H., A. Natsume, H. Ueda, T. Saitoh, and H. Ogawa. 2001. Screening enterology. 79:508. of a unique lectin from 16 cultivable mushrooms with hybrid and 3. Parkos, C. A., S. P. Colgan, M. S. Diamond, A. Nusrat, T. W. Liang, neoproteoglycan probes and purification of a novel N-acetylglucosamine-specific T. A. Springer, and J. L. Madara. 1996. Expression and polarization of intercel- lectin from Oudemansiella platyphylla fruiting body. Biochim. Biophys. Acta lular adhesion molecule-1 on human intestinal epithelia: consequences for 1526:(37. CD11b/CD18-mediated interactions with neutrophils. Mol. Med. 2:489. 30. Robertson, L., G. J. Wishart, and A. J. Horrocks. 2000. Identification of peri- 4. Huang, C., and T. A. Springer. 1995. A binding interface on the I domain of vitelline N-linked glycans as mediators of sperm-egg interaction in chickens. lymphocyte function-associated antigen-1 (LFA-1) required for specific interac- J. Reprod. Fertil. 120:397. tion with intercellular adhesion molecule 1 (ICAM-1). J. Biol. Chem. 270:19008. 31. Ortega-Barria, E., and J. C. Boothroyd. 1999. A toxoplasma lectin-like activity 5. Michishita, M., V. Videm, and M. A. Arnaout. 1993. A novel divalent cation- specific for sulfated polysaccharides is involved in host cell infection. J. Biol. ␤ binding site in the A domain of the 2 integrin CR3 (CD11b/CD18) is essential Chem. 274:1267. for ligand binding. Cell 72:857. 32. Haroun-Bouhedja, F., M. Ellouali, C. Sinquin, and C. Boisson-Vidal. 2000. Re- 6. Diamond, M. S., J. Garcia-Aguilar, J. K. Bickford, A. L. Corbi, and lationship between sulfate groups and biological activities of fucans. Thrombosis T. A. Springer. 1993. The I domain is a major recognition site on the leukocyte Res. 100:453. 5278 T84 CELL ADHESION TO CD11b/CD18

33. Vischer, P., and E. Buddecke. 1991. Different action of heparin and fucoidan on 44. DeAngelis, P. L., and C. G. Glabe. 1988. Role of basic amino acids in the in- arterial smooth muscle cell proliferation and thrombospondin and fibronectin teraction of bindin with sulfated fucans. Biochemistry 27:8189. metabolism. Eur. J. Cell Biol. 56:407. 45. Barbe, L. L., B. M. Boval, M. P. Wautier, and J. L. Wautier. 2001. An in vitro 34. DeAngelis, P. L., and C. G. Glabe. 1987. Polysaccharide structural features that system for testing leucocyte and leukaemic cell line adhesion to synthetic fibres. are critical for the binding of sulfated fucans to bindin, the adhesive protein from Br. J. Haematol. 115:664. sea urchin sperm. J. Biol. Chem. 262:13946. 46. Santiago-Garcia, J., J. Mas-Oliva, T. L. Innerarity, and R. E. Pitas. 2001. Secreted 35. Dorscheid, D. R., A. E. Conforti, K. J. Hamann, K. F. Rabe, and S. R. White. forms of the amyloid-␤ precursor protein are ligands for the class A scavenger 1999. Characterization of cell surface lectin-binding patterns of human airway receptor. J. Biol. Chem. 276:30655. epithelium. Histochem. J. 31:145. ␣ 36. Suarez, S. S., I. Revah, M. Lo, and S. Kolle. 1998. Bull sperm binding to ovi- 47. Kawamura, M., J. W. Heinecke, and A. Chait. 2000. Increased uptake of -hy- ductal epithelium is mediated by a Ca2ϩ-dependent lectin on sperm that recog- droxy aldehyde-modified low density by macrophage scavenger re- nizes Lewis-a trisaccharide. Biol. Reprod. 59:39. ceptors. J. Lipid Res. 41:1054. 37. Parkos, C. A. 1997. Molecular events in neutrophil transepithelial migration. 48. Tangemann, K., A. Bistrup, S. Hemmerich, and S. D. Rosen. 1999. Sulfation of BioEssays 19:865. a high endothelial venule-expressed ligand for L-selectin: effects on tethering and 38. Parkos, C. A. 1997. Cell adhesion and migration. I. Neutrophil adhesive inter- rolling of lymphocytes. J. Exp. Med. 190:935. actions with intestinal epithelium. Am. J. Physiol. 273:G763. 49. Kritharides, L., J. Upston, W. Jessup, and R. T. Dean. 1998. Accumulation and 39. Parkos, C. A., C. Delp, M. A. Arnaout, and J. L. Madara. 1991. Neutrophil metabolism of low density lipoprotein-derived cholesteryl linoleate hydroperox- migration across a cultured intestinal epithelium: dependence on a CD11b/CD18- ide and hydroxide by macrophages. J. Lipid Res. 39:2394. mediated event and enhanced efficiency in physiological direction. J. Clin. Invest. 50. Lefebvre, R., M. C. Lo, and S. S. Suarez. 1997. Bovine sperm binding to ovi- 88:1605. ductal epithelium involves fucose recognition. Biol. Reprod. 56:1198. 40. Shimaoka, M., M. Ikeda, T. Iida, N. Taenaka, I. Yoshiya, and T. Honda. 1996. 51. Stoolman, L. M., and S. D. Rosen. 1983. Possible role for cell-surface carbohy- Fucoidin, a potent inhibitor of leukocyte rolling, prevents neutrophil influx into drate-binding molecules in lymphocyte recirculation. J. Cell Biol. 96:722. phorbol-ester-induced inflammatory sites in rabbit lungs. Am. J. Respir. Crit. Care Med. 153:307. 52. Pearce-Pratt, R., and D. M. Phillips. 1996. Sulfated polysaccharides inhibit lym- 41. Granert, C., J. Raud, X. Xie, L. Lindquist, and L. Lindbom. 1994. Inhibition of phocyte-to-epithelial transmission of human immunodeficiency virus-1. Biol. Re- leukocyte rolling with polysaccharide fucoidin prevents pleocytosis in experi- prod. 54:173. Downloaded from mental meningitis in the rabbit. J. Clin. Invest. 93:929. 53. Abraham, D., G. Bou-Gharios, H. Muir, and I. Olsen. 1989. Adhesion of lym- 42. Yanaka, K., N. Kato, and T. Nose. 2000. Heparin also interacts with selectins and phoid cells to fibroblasts in tissue culture. Cell. Immunol. 122:33. modulates the cell adhesion process. Circulation 102:E169. 54. Dorscheid, D. R., K. R. Wojcik, K. Yule, and S. R. White. 2001. Role of cell 43. Malhotra, R., M. Ward, R. B. Sim, and M. I. Bird. 1999. Identification of human surface glycosylation in mediating repair of human airway epithelial cell mono- complement factor H as a ligand for L-selectin. Biochem. J. 341:61. layers. Infect. Immun. 281:L982. http://www.jimmunol.org/ by guest on September 29, 2021