Adhesion (Cd11b/CD18) in Regulating Neutrophil Mac-1/Complement Receptor Type 3 Function of the Lectin Domain Of

Home , Urokinase receptor

Function of the Lectin Domain of Mac-1/Complement Receptor Type 3 (CD11b/CD18) in Regulating Neutrophil Adhesion This information is current as of September 25, 2021. Yu Xia, Gita Borland, Jibiao Huang, Ikuko F. Mizukami, Howard R. Petty, Robert F. Todd III and Gordon D. Ross J Immunol 2002; 169:6417-6426; ; doi: 10.4049/jimmunol.169.11.6417 http://www.jimmunol.org/content/169/11/6417 Downloaded from

References This article cites 60 articles, 38 of which you can access for free at:

http://www.jimmunol.org/content/169/11/6417.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• 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 by guest on September 25, 2021

*average

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

Function of the Lectin Domain of Mac-1/Complement Receptor Type 3 (CD11b/CD18) in Regulating Neutrophil Adhesion1

Yu Xia,2* Gita Borland,* Jibiao Huang,† Ikuko F. Mizukami,‡ Howard R. Petty,† Robert F. Todd III,‡ and Gordon D. Ross3*

A lectin function within CD11b mediates both cytotoxic priming of Mac-1/complement receptor type 3 (CR3) by ␤-glucan and the formation of transmembrane signaling complexes with GPI-anchored glycoproteins such as CD16b (Fc␥RIIIb). A requirement for GPI-anchored urokinase plasminogen activator receptor (uPAR; CD87) in neutrophil adhesion and diapedesis has been demonstrated with uPAR-knockout mice. In this study, neutrophil activation conditions generating high-affinity (H-AFN) or low-affinity ␤ (L-AFN) 2 integrin adhesion were explored. A role for the Mac-1/CR3 lectin domain and uPAR in mediating H-AFN or L-AFN adhesion was suggested by the inhibition of Mac-1/CR3-dependent adhesion to ICAM-1 or fibrinogen by ␤-glucan or anti-uPAR. The formation of uPAR complexes with Mac-1/CR3 activated for L-AFN adhesion was demonstrated by fluorescence resonance Downloaded from energy transfer. Conversely, Jurkat cell LFA-1 H-AFN-adhesion to ICAM-1 was not associated with uPAR/LFA-1 complexes, any requirement for GPI-anchored glycoproteins, or inhibition by ␤-glucan. A single CD11b lectin site for ␤-glucan and uPAR was suggested because the binding of either ␤-glucan or uPAR to Mac-1/CR3 selectively masked two CD11b epitopes adjacent to the transmembrane domain. Moreover, treatment with phosphatidylinositol-specific phospholipase C that removed GPI-anchored proteins increased CD11b-specific binding of 125I-labeled ␤-glucan by 3-fold and this was reversed with soluble recombinant

uPAR. Conversely, neutrophil activation for generation of Mac-1/CR3/uPAR complexes inhibited CD11b-dependent binding of http://www.jimmunol.org/ 125I-labeled ␤-glucan by 75%. These data indicate that the same lectin domain within CD11b regulates both the cytotoxic and adhesion functions of Mac-1/CR3. The Journal of Immunology, 2002, 169: 6417–6426.

␣ ␤ he leukocyte M 2 integrin known also as Mac-1, com- can to a distinct lectin domain contained within the C-terminal plement receptor type 3 (CR3),4 and CD11b/CD18 func- region of CD11b (7–9). Ligation of fungal ␤(1, 3)-glucans to the T tions both as an adhesion molecule facilitating diapedesis lectin domain of CD11b results in priming of the receptor, such and as a C3R enabling phagocytosis or degranulation in response that yeast cells bound to the I-domain via iC3b trigger “outside-in” to factor I-cleaved C3b fragment of C3 (iC3b)-opsonized micro- signaling for phagocytosis or degranulation (10, 11). In addition, organisms (1–5). Important protein ligands such as ICAM-1, iC3b, the binding of soluble ␤-glucan or yeast cell walls to the lectin by guest on September 25, 2021 and fibrinogen bind to overlapping sites contained within an “in- domain can generate the H-AFN MIDAS conformation within the serted” I-domain at the N terminus of the CD11b subunit that is I-domain (11, 12). induced to express a high-affinity (H-AFN) metal ion-dependent Several lines of evidence indicate that adhesion via Mac-1/CR3 adhesion site (MIDAS) following cell activation. Notably, adhe- binding to endothelial cell ICAM-1 requires the formation of sion may also occur through the cytoskeleton-regulated clustering membrane complexes between Mac-1/CR3 and GPI-anchored of integrins that retain a low-affinity (L-AFN) binding site state urokinase plasminogen activator receptor (uPAR). mAbs to differ- (6). Phagocytosis of iC3b-opsonized fungi that are captured first by ent epitopes of uPAR can either inhibit or induce Mac-1-dependent the I-domain requires secondary ligation of fungal cell wall ␤-glu- adhesion, and removal of GPI-anchored proteins with phosphatidylinositol-specific phospholipase C (PiPLC) (13) or inhibition of uPAR synthesis with an antisense oligonucleotide (14, 15) pre- *Chemoattractant Group of the James Graham Brown Cancer Center, Departments of vents Mac-1-dependent adhesion until the cells are reconstituted Pathology, and of Microbiology and Immunology, University of Louisville, Louis- ville, KY 40202; †Department of Biological Sciences, Wayne State University, De- with soluble recombinant uPAR (sr-uPAR) (13). Moreover, neu- troit, MI 48202; and ‡Division of Hematology/Oncology, Department of Internal trophils from uPAR-deficient mice exhibit defective diapedesis Medicine, Comprehensive Cancer Center, University of Michigan, Ann Arbor, MI 48109 into certain inflammatory sites (13, 16). A lectin-like interaction Received for publication May 9, 2002. Accepted for publication September 26, 2002. appears to be involved in uPAR-dependent adhesion because the surface complexes between uPAR and Mac-1 are disrupted by sug- The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance ars such as N-acetyl-D-glucosamine (NADG) (17). These data sug- with 18 U.S.C. Section 1734 solely to indicate this fact. gest that uPAR may bind to the same lectin domain within the C 1 This work was supported by National Institutes of Health Grants CA86412 (to terminus of CD11b that is used for cytotoxic degranulation in re- G.D.R.), CA42246 (to R.F.T.), and AI27409 (to H.R.P.). sponse to iC3b-opsonized yeast. Moreover, because similar sugar- 2 Current address: Celera, 180 Kimball Way, South San Francisco, CA 84080 inhibitable complexes have been observed between uPAR and 3 Address correspondence and reprint requests to Dr. Gordon D. Ross, James Graham ␤ ␤ CR4 (CD11c/CD18) (18), as well as between uPAR and 1 or 3 Brown Cancer Center, University of Louisville, 529 South Jackson Street, Room 429, Louisville, KY 40202. E-mail address: [email protected] integrins (19), it appears possible that lectin-dependent complexes formed with uPAR may be important for adhesion with a broad 4 Abbreviations used in this paper: CR3, complement receptor type 3; iC3b, factor I-cleaved C3b fragment of C3; H-AFN, high affinity; L-AFN, low affinity; MIDAS, range of integrins. Although sugar-inhibitable complexes between metal ion-dependent adhesion site; uPAR, urokinase plasminogen activator receptor; LFA-1 (CD11a/CD18) and Fc␥RIIIB have been demonstrated us- PiPLC, phosphatidylinositol-specific phospholipase C; sr-uPAR, soluble recombinant uPAR; NADG,N-acetyl-D-glucosamine; RET, resonance energy transfer; TRITC, ing resonance energy transfer (RET) techniques (20), the forma- tetramethylrhodamine isothiocyanate; 125I-␤-glucan, 125I-labeled ␤-glucan. tion of LFA-1 complexes with uPAR has not been investigated.

Nevertheless, other investigators have demonstrated the recovery All other chemicals and reagents, except where specified, were purchased of uPAR within anti-CD11a immunoprecipitates from monocytes, from Sigma-Aldrich (St. Louis, MO). as well as sparse LFA-1/uPAR cocapping (21). Neutrophils and T cells A lectin site has been identified in both human and murine CR3 that binds soluble or particulate ␤-glucan. However, studies of Peripheral blood neutrophils were isolated under LPS-free conditions using ␤ two-step Ficoll/Hypaque density gradient centrifugation (34). Peripheral other 2 integrins have failed to demonstrate a similar lectin ac- blood T lymphocytes were isolated using RosetteSep T Cell Enrichment tivity (8, 10, 22). The exact location of the lectin site within the C Cocktail according to the manufacturer’s protocol (StemCell Technologies, terminus of CD11b has not been determined, but its blockade by Vancouver, British Columbia, Canada). Isolated T cell preparations were ϩ ␤-glucan oligosaccharides containing as few as seven glucose sub- Ն96% CD3 , but only weak staining for uPAR was detectable by indirect units (23) suggests that it represents a relatively small portion of immunofluorescence. The T cell line Jurkat E6-1 was obtained from the American Type Culture Collection and maintained in RPMI 1640 medium the C-terminal domain. Complicating mapping of the lectin site is supplemented with 10% FBS. The majority of Jurkat cells bore readily the molecular flexibility of integrins that allows a ligand bound to detectable LFA-1 and uPAR but not Mac-1/CR3. one end of the molecule to generate a conformational change at the opposite end of the molecule. Thus, the binding of mAbs to the Assay of neutrophils and Jurkat cells for uPAR N-terminal I-domain results in a masking of the C-terminal lectin membrane complexes with Mac-1/CR3 or LFA-1 by immunofluorescence RET site, and conversely, small (10 kDa) ␤-glucans that bind to the C-terminal region can generate the H-AFN MIDAS within the I- Neutrophils were suspended in 100 ␮l HBSS/5 mM Ca2ϩ/1 ␮M calcium domain. An important finding was that small ␤-glucans bound with ionophore A23187, seeded on cover glasses, and incubated at 37°C for 15 min. After washing with ice-cold HBSS, the neutrophils were reacted se- sufficiently H-AFN (50 nM) to block the uptake of C-terminal Ј Ј Downloaded from quentially with Mo1/44 F(ab )2 anti-CD11b-FITC and 3B10 F(ab )2 anti- domain-specific mAbs without blocking the uptake of mAbs to uPAR-TRITC in 1% BSA/PBS at 4°C for 20 min. Next, the stained neu- N-terminal epitopes (8). Moreover, recombinant C-terminal frag- trophils were washed with ice-cold PBS/1 mM EGTA and rinsed with ϩ ments of CD11b expressed on insect cells bound ␤-glucan with an ice-cold PBS/EGTA containing 10 mM Mg2 and then sealed on the cover affinity similar to native CD11b/CD18 on neutrophils (9). glasses in the latter solution. As a control, cells were washed with and sealed in PBS/EGTA medium without Mg2ϩ. Duplicate slides prepared in The current investigation sought to determine whether uPAR this way were analyzed at timed intervals in parallel by immunofluores- and small ␤-glucans competed for binding to the same lectin site cence microscopy with a temperature controlled stage regulated at either http://www.jimmunol.org/ within CD11b, and whether uPAR binding to this lectin site played 4°Cor37°C. Jurkat T cells cultured for 2 days in 10% FCS/RPMI 1640 a role in regulating the adhesion function of CR3. Although the medium without phenol red were stained and analyzed in a similar way as neutrophils using sequential treatments with TS1/22 anti-LFA-1 IgG plus primary focus was on neutrophil Mac-1/CR3 and its interaction Ј rabbit F(ab )2 antimouse IgG-Fc-specific Ab-FITC followed by 3B10 with uPAR mediating adhesion to ICAM-1, the expression of Ј F(ab )2 anti-uPAR-TRITC. Calcium ionophore that was used to up-regu- LFA-1 on neutrophils complicated the interpretation of data and late neutrophil surface CR3 to levels comparable to uPAR was unnecessary necessitated the inclusion of experiments with T cells that express with Jurkat cells that maximally expressed LFA-1 and uPAR without stim- only LFA-1 and not CR3. ulation. The instrumentation and methods used to measure FITC fluorescence and the TRITC fluorescence resulting from RET of excited FITC to TRITC molecules clustered to within Յ7 nm of each other have been by guest on September 25, 2021 Materials and Methods previously described (35, 36). Abs and other reagents Activation of neutrophils with PMA and analysis of CD11b All mAbs were used as IgG purified from ascite fluid or culture medium by epitopes anion exchange chromatography (22, 24). The hybridomas secreting anti- Neutrophils were incubated with or without PiPLC (0.25 U, 1 ϫ 107 cells/ CD11b mAbs MN-41 and OKM1 were obtained from Drs. A. Eddy and A. ml) in RPMI 1640 medium/0.2% BSA at 37°C for 1 h. Typically, this Michael, University of Minnesota (Minneapolis, MN) and American Type treatment caused a 70–80% reduction in staining for GPI-anchored CD16b Culture Collection (Manassas, VA), respectively. Other anti-CD11b mAbs, and uPAR using 3G8 and 3B10 mAbs, respectively. After washing, cells CBRM1/5, CBRM1/10, CBRM1/21, and CBRM1/23 (25, 26) were pro- were incubated first with or without PMA (20 ng/ml) at 37°C for 20 min, vided by Dr. T. Springer (Center for Blood Research and Harvard Medical and then with various FITC or Oregon Green 488-labeled mAbs at 4°C School, Boston, MA), and mAb24 (27) was a gift from Dr. N. Hogg (Im- (37°C for mAb 24; Ref. 37) for 30 min. After washing, cells were analyzed perial Cancer Research Fund, London, U.K.). The generation of anti- by flow cytometry using a Coulter Profile II (Beckman Coulter, Miami Ј CD11b Mo1/44 mAb and preparation of F(ab )2 coupled to FITC were Lakes, FL). For all flow cytometry assays, 10 ␮g/ml propidium iodide was previously described (20, 28). The epitopes and function blocking effects of added to the stained cells just before analysis to allow exclusion of dead these mAbs have recently been reviewed (5). Anti-uPAR mAb 3B10 and cells that stain nonspecifically with any labeled mAb. These data consis- Ј the generation of its F(ab )2 coupled to tetramethylrhodamine isothiocya- tently demonstrated Ն96% viability of neutrophils following treatment nate (TRITC) were previously described (17, 29, 30). Anti-CD11a mAb with PMA and/or PiPLC. TS1/22 was provided by Dr. Springer and G43-25B was purchased from BD PharMingen (San Diego, CA). The hybridoma-secreting anti-human Assay of neutrophils for CD11b-specific binding of 125I-labeled MHC class I mAb, DX17 (31), was a gift from Dr. L. L. Lanier (University ␤-glucan (125I-␤-glucan) of California, San Francisco, CA) and the hybridoma-secreting 3G8 anti- CD16 (Fc␥RIII) was obtained from Dr. J. Unkeless (Mt. Sinai School of Neutrophils were tested for uptake of 125I-␤-glucan as previously described Medicine, New York, NY). Anti-CD3-PE, anti-CD55-FITC, and anti- (8, 22) using 125I-␤-glucan of ϳ10 kDa. Briefly, neutrophils were incu- CD59-FITC were purchased from BD PharMingen. For flow cytometry, bated on ice with 2–4 ␮g/ml of 125I-␤-glucan for 15 min. Triplicate ali- mAbs were coupled to FITC or the fluorescein derivative Oregon Green quots of the cell suspension (1 ϫ 106 cells) were layered onto mineral oil 488 dye according to the manufacturer’s instructions (Molecular Probes, and the cells with bound radioactivity were separated from fluid phase Eugene, OR). Soluble rICAM-1-Fc (32) was generously provided by Dr. unbound 125I-␤-glucan by centrifugation at 14,000 ϫ g for 1 min in 500 ␮l D. Staunton (ICOS, Seattle, WA). A portion of this rICAM-1 was labeled of conical centrifuge tubes. After freezing these tubes at Ϫ140°C, the tips with either Oregon Green 488 or FITC. sr-uPAR was generated as de- of the tubes containing the cells with bound 125I-␤-glucan were cut off and scribed (33). PiPLC was a gift from Dr. M. Lowe, (Columbia University, analyzed with a gamma scintillation counter. The proportion of cell-asso- New York, NY). Various preparations of a soluble zymosan-derived poly- ciated radioactivity that was bound specifically to Mac-1/CR3 was calcu- saccharide fraction made up primarily of (1,3) ␤-D-glucan (␤-glucan; vary- lated as the net bound cpm measured with neutrophils that had been treated ing in size from ϳ2–20 kDa) were isolated and characterized for approx- with 10 ␮g/ml of anti-CD11b mAb as compared with untreated neutro- imate size by Superdex 75 molecular sieve column chromatography vs phils. In some experiments, neutrophils were treated with PiPLC as de- dextran-FITC molecular mass standards (8, 22). For use in radioactive scribed above to remove a proportion of all GPI-anchored proteins and then binding assays, an ϳ10 kDa ␤-glucan was first coupled to tyramine by membrane uPAR was selectively reconstituted by addition of sr-uPAR as reductive amination, and then radiolabeled with Na125I using Iodogen (9). previously described (33). Briefly, after cells had been treated with PiPLC, The Journal of Immunology 6419

the cells were divided into two portions that were incubated either with or identify the requirements of uPAR and/or lectin site interactions. without 10 ␮M sr-uPAR for 20 min at 25°C followed by two washes of the In this study, H-AFN adhesion was defined as the ability to bind cells to remove any remaining unbound sr-uPAR. These assays required fluid-phase rICAM-1-FITC that was inhibitable with mAbs to neutrophil viability to exceed 95% to prevent nonspecific uptake of 125I- ␤-glucan, and therefore any neutrophil isolates exhibiting Ͻ95% viability CD11a and/or CD11b (6, 39). L-AFN adhesion was defined as were not used. stimulated adhesion to a surface that did not involve generation of H-AFN binding sites and was inhibitable by mAbs to CD11a Assay for staining with soluble rICAM-1-FITC and/or CD11b. With neutrophils, T lymphocytes, and Jurkat T For analysis of the binding of soluble rICAM-1 by flow cytometry, cells cells, staining with rICAM-1-FITC was detectable following stim- ␮ were stained by incubation in 1.0 g/ml rICAM-1-FITC for 30 min at ulation with 10 mM Mg2ϩ/2 mM EGTA, but not with 10 mM 37°C. After staining, the cells were washed two times and suspended in 2ϩ 2ϩ ice-cold PBS/0.2% BSA containing 10 ␮g/ml propidium iodide, and ana- Mg /1 mM Ca (Fig. 1). PMA (20 ng/ml) was effective only lyzed for staining by flow cytometry. For analysis of the specificity of with neutrophils, and corresponded to expression of the H-AFN rICAM-1-FITC staining, neutrophils, Jurkat cells, or T cells were incu- binding site of Mac-1/CR3 as shown by staining with CBRM1/5- ␤ bated first with or without mAbs or soluble -glucan at 4°C for 20 min, FITC, a specific marker of the Mac-1/CR3 H-AFN MIDAS. Tests activated either with 20 ng/ml PMA or 10 mM Mg2ϩ in the presence of either 2 mM EGTA or 1 mM Ca2ϩ at 37°C for 10 min, and finally stained for the specificity of rICAM-1-FITC staining stimulated by 10 mM 2ϩ at 37°C for 30 min with 1.0 ␮g/ml rICAM-1-FITC. Mg /EGTA (Fig. 2) showed that this condition was selective for stimulation of H-AFN LFA-1 because staining was inhibited com- Assay for neutrophil and T cell adhesion to immobilized pletely by a mAb to CD11a, whereas a mAb to CD11b I-domain rICAM-1 had no effect on staining. It had previously been reported that ϩ Costar EIA/RIA eight-well strips (Corning, Corning, NY) were coated with stimulation with 10 mM Mg2 /EGTA also failed to generate the Downloaded from ␮ ␮ 2ϩ 2ϩ 100 l rICAM-1 (2 g/ml in PBS with 1 mM each of Ca and Mg ), CD11b I-domain CBRM1/5 neoepitope (11). Such LFA-1 H-AFN incubated overnight at 4°C, and unbound binding sites were blocked with ␤ 2.5% BSA/PBS/Ca2ϩ/Mg2ϩ at room temperature for 1 h. Where indicated, binding of fluid ICAM-1 did not require uPAR or a -glucan- T cells or neutrophils were treated with PiPLC to remove GPI-anchored reactive lectin site, as there was no significant inhibition by either proteins as described for neutrophils above. Neutrophils or T cells were anti-uPAR or soluble ␤-glucan. As expected, studies with PMA- labeled with 51Cr by incubating 1 ϫ 107 cells in 0.5 ml of calcium/mag- stimulated neutrophils showed that inhibition of staining with nesium-free HBSS/1% BSA medium with 250 ␮Ci of Na51Crfor1hat 51 ICAM-1-FITC required a combination of anti-CD11a and CD11b http://www.jimmunol.org/ 37°C, followed by three washes with this medium to remove unbound Cr. Ն For neutrophil adhesion assay, 51Cr-labeled cells (8 ϫ 104) were added into (that produced 75% inhibition) indicating that PMA stimulated wells of eight-well strips coated with rICAM-1 and the strips were incu- both H-AFN LFA-1 and Mac-1/CR3 (not shown). bated at 37°C for 3.5–5 min in a water bath. For T lymphocyte or Jurkat cell adhesion assays, 51Cr-labeled cells (1 ϫ 105) were added into the wells GPI-anchored glycoproteins contribute to formation of the coated with rICAM-1 and the strips were incubated first at 4°C for 1 h and H-AFN Mac-1/CR3 binding site then incubated at 37°C for 10 min. Labeled cell suspensions exhibited Ն95% viability or were not used in this assay. Cells were activated for PMA is an activator of neutrophil adhesion and the Mac-1/CR3 adhesion by mixture with PMA (10 ng/ml) or 10 mM Mg2ϩ in the presence H-AFN MIDAS conformation. PMA-stimulated neutrophils ␮ or absence of mAbs to CD11a, CD11b, uPAR, or MHC class I (10 g/ml), bound soluble rICAM-1-FITC, and this was reduced 35% ( p Ͻ

␤ ␮ ␣ ␮ by guest on September 25, 2021 -glucan (5 g/ml), or -mannan (25 g/ml) just before addition to the ϳ rICAM-1-coated wells. With T lymphocytes and Jurkat cells, the 10 mM 0.05) by prior treatment with PiPLC that removed 70% of cell Mg2ϩ activation medium included 1 mM EGTA to chelate Ca2ϩ (38). surface GPI-anchored proteins such as uPAR (Fig. 3) or CD16b After incubation, the wells were washed four times with PBS/Ca2ϩ/Mg2ϩ, (not shown). By contrast, when neutrophils were activated instead separated into single wells, and the radioactivity of each well was deter- with 10 mM Mg2ϩ/EGTA in a manner that activated only LFA-1 mined with a gamma scintillation counter. for H-AFN adhesion, PiPLC treatment had no effect on neutrophil Assay for LPS-stimulated neutrophil adhesion to immobilized binding of soluble rICAM-1-FITC (Fig. 3). These data suggest that fibrinogen GPI-anchored proteins function only in generation of H-AFN Costar eight-well strips were coated with fibrinogen in the same way as with rICAM-1. To minimize neutrophil stimulation before the assay, neutrophils were washed and suspended in medium containing 10 ␮g/ml polymyxin B and the time for labeling with 51Cr was reduced from1hto20 min. Following labeling, the neutrophils were washed three times and suspended in medium lacking polymyxin B. Viability was Ն95%. The cells were dispensed into 12 ϫ 75-mm plastic tubes, suspended in calcium/ magnesium-free HBSS/BSA, and incubated with or without 5 ␮g/ml of mAbs to CD11a (TS1/22), CD11b (MN-41), uPAR, or 10 ␮g/ml ␤-glucan (ϳ3 kDa) for 20 min on ice, and then triplicate 50-␮l samples from each tube were added to the fibrinogen-coated wells containing 50 ␮l of HBSS/ BSA with 2 mM each of Ca2ϩ and Mg2ϩ and 2 ␮g/ml of LPS. The cells were rapidly sedimented onto the well surfaces by 2-min centrifugation of the plates at 100 ϫ g, and then the plates were placed at 37°C for 40 min to allow LPS-stimulated adhesion. After transferring the plates to an ice bath, unbound neutrophils were removed by four washes of each well with ice-cold HBSS/BSA, the individual wells were separated from the plate, and each well was analyzed for 51Cr with a gamma scintillation counter.

Controls included wells lacking LPS or containing HBSS/EDTA to prevent 2ϩ all integrin-mediated adhesion. FIGURE 1. Effect of Ca on the generation of integrin H-AFN binding site(s) as indicated by staining with soluble rICAM-1-FITC. Stimulation of Results integrin H-AFN binding site(s) by 10 mM Mg2ϩ required chelation of ϩ Cell stimulation conditions required for generation of H- or calcium, whereas stimulation by PMA occurred in the presence of Ca2 . L-AFN integrin binding sites The results shown represent the mean Ϯ SD of more than or equal to three assays. The single asterisks indicate that the difference in the staining ob- ␤ 2ϩ Because 2 integrins mediate adhesion through either H- or L- tained following stimulation with PMA and/or 10 mM Mg /EGTA was AFN binding sites, it was important to deﬁne cell stimulation con- signiﬁcantly greater (p Ͻ 0.01) than the staining obtained in the presence ditions with both Mac-1/CR3 and LFA-1 that could be used to of 10 mM Mg2ϩ/1 mM Ca2ϩ. 6420 LECTIN DOMAIN REGULATION OF Mac-1/CR3 ADHESION Downloaded from

FIGURE 3. Role of GPI-anchored glycoproteins in the generation of ␤ 2-integrin H-AFN ICAM-1 binding site(s) as indicated by the reduced ability of PiPLC-treated cells to bind soluble rICAM-1-FITC. Cells with or

without treatment with PiPLC to remove GPI-anchored proteins were stim- http://www.jimmunol.org/ ulated with PMA or 10 mM Mg2ϩ/EGTA and examined for staining by soluble rICAM-1-FITC. Staining was also conducted with anti- uPAR-FITC to determine the proportion of GPI-anchored uPAR removed by this PiPLC treatment. The results shown represent the mean Ϯ SD of more than or equal to three assays, and the asterisk indicates that the difference between bracketed bars was signiﬁcant (p Ͻ 0.05)

ϩ ϩ by incubation in 10 mM Mg2 plus1mMCa2 to prevent the by guest on September 25, 2021 FIGURE 2. Mg2ϩ/EGTA-stimulation of the H-AFN binding of soluble formation of H-AFN LFA-1. Under these conditions, adhesion to rICAM-1-FITC occurs only with LFA-1 and not with Mac-1/CR3, and shows no requirement for a ␤-glucan-reactive lectin site or uPAR. Anti- ICAM-1 was mutually dependent upon L-AFN LFA-1 and Mac- CD11a (TS1/22) and anti-CD11b (MN-41) were used to block LFA-1- and 1/CR3, because mAbs to CD11a and CD11b each produced less Mac-1/CR3-dependent binding of rICAM-1-FITC at a concentration of 2.5 inhibition (22 and 26%) than did a mixture of both mAbs (75%, ␮g/ml. Soluble ␤-glucan was tested over a range of concentrations and the Fig. 4). As little as 5 ␮g/ml soluble ␤-glucan also produced 26% data shown were obtained with the highest concentration examined (25 inhibition of adhesion, and ␤-glucan slightly augmented the inhi- ␮g/ml). The results shown represent the mean Ϯ SD of more than or equal bition activity produced by mAbs to either CD11a or CD11b. Its to three assays. The asterisks indicate that the staining obtained with cells 25% inhibition of adhesion was similar to the 23% produced by a 2ϩ Ͻ stimulated with 10 mM Mg /2 mM EGTA was significantly greater (p mAb to uPAR. However, adding ␤-glucan to the mixture of anti- 2ϩ 2ϩ 0.01) than that obtained in the presence of 10 mM Mg /1 mM Ca , and CD11a and anti-CD11b did not produce greater inhibition than that that the inhibition of staining obtained with anti-CD11a (that was nearly observed with the mAb mixture without ␤-glucan (data not 100% in all cases) was significant (p Ͻ 0.001). The effect on staining produced by the other mAbs and ␤-glucan was not significant (p Ͼ 0.05). shown). Moreover, there was no significant inhibition of adhesion produced by either a mAb to class I or 25 ␮g/ml soluble yeast ␣-mannan (Fig. 4). Even though these data suggest that a lectin site Mac-1/CR3 and not LFA-1. Tests of PiPLC-treated uPARϪ T lym- functions in promoting adhesion by L-AFN receptors, they do not ϩ ␤ phocytes (40) and uPAR Jurkat cells after stimulation with allow distinction of which of these 2 integrins use lectin site Mg2ϩ/EGTA similarly failed to show any role for GPI-anchored interactions under these conditions. glycoproteins in generation of H-AFN LFA-1. Although T lymphocytes did not express uPAR, tests for two other GPI-anchored A role for uPAR and lectin interactions in Mac-1/CR3- T cell proteins, CD55 and CD59, showed a Ͼ90% reduction in T dependent neutrophil adhesion but not LFA-1-dependent T cell cell staining for these proteins following PiPLC treatment (not adhesion shown). Because the dual binding of LFA-1 and Mac-1/CR3 made it dif- ficult to define the individual requirements for either integrin, con- Lectin domain regulation of neutrophil adhesion to ICAM-1 ditions were used that were selective for each integrin. LPS is Previous reports showed that surface complexes between uPAR known to stimulate the H-AFN MIDAS of neutrophil CD11b, and and Mac-1/CR3 were dependent upon lectin-carbohydrate interac- neutrophils treated with 1 ␮g/ml LPS were readily stained with tions because they were disrupted by NADG (18). If such com- either soluble rICAM-FITC or CBRM1/5-FITC (not shown). Be- plexes were required for Mac-1/CR3-dependent adhesion to cause adhesion to ICAM-1 could occur through either LFA-1 or ICAM-1, then disruption of these complexes should inhibit adhe- Mac-1/CR3 (Fig. 4), fibrinogen was used to coat surfaces instead sion. Neutrophil adhesion to immobilized rICAM-1 was induced of ICAM-1, because fibrinogen binds avidly to Mac-1/CR3 (41, The Journal of Immunology 6421 Downloaded from

FIGURE 5. Function of uPAR and the ␤-glucan-reactive lectin site of CR3 in neutrophil LPS-stimulated adhesion to immobilized fibrinogen. Neutrophils labeled with 51Cr under LPS-free conditions were first incubated with 1 ␮g/ml mAb or ␤-glucan (ϳ3 kDa; 5 ␮g/ml) and then added FIGURE 4. Specificity of neutrophil adhesion to immobilized rICAM-1 2ϩ 2ϩ 51 to fibrinogen-coated wells in the presence of 1 ␮g/ml LPS. Neutrophil stimulated with 10 mM Mg /1 mM Ca . Stimulated and Cr-labeled http://www.jimmunol.org/ adhesion was analyzed after a 40-min incubation of the wells at 37°C. The neutrophils were incubated with or without mAbs, ␤-glucan, or ␣-mannan, results shown represent the mean Ϯ SD of four or more assays conducted and then the cells were added into wells coated with rICAM-1 for analysis in triplicate. Asterisks indicate a significant difference (p Ͻ 0.05) compared of adhesion. mAbs to CD11b used were MN-41 or CBRM1/21. Each gave with the LPS-stimulated neutrophils not treated with mAb or ␤-glucan. similar results and only the results obtained with MN-41 are shown. mAbs Anti-CD11b, anti-uPAR, and ␤-glucan each produced approximately to CD11a used were TS1/22 or G43-25B. TS1/22 produced more inhibition equivalent inhibition of adhesion, and there was no significant additional than did G43-25B, and only the data obtained with TS1/22 are shown. The decrease in adhesion when anti-CD11b was combined with anti-uPAR or results shown represent the mean Ϯ SD of three or more assays conducted ␤-glucan, or when a mixture of anti-uPAR and ␤-glucan was examined. in triplicate. Anti-CD11a, anti-CD11b, anti-uPAR, and soluble ␤-glucan each produced significant inhibition of adhesion relative to the control (asterisk over individual bars indicates p Ͻ 0.05). In contrast, adhesion was by guest on September 25, 2021 not inhibited significantly by either anti-class I or soluble ␣-mannan (n.s. To examine a possible function of uPAR in LFA-1-dependent over these bars indicates not significant). Mixtures of anti-CD11a or anti- adhesion to ICAM-1 without assay interference from Mac-1/CR3 CD11b plus another mAb or soluble ␤-glucan each produced significantly as had occurred with neutrophils (Fig. 4), Jurkat cells and isolated more inhibition than did the single mAb alone. The brackets over pairs of peripheral blood T lymphocytes were tested instead of neutrophils bars indicate statistical analysis of tests in which an individual mAb is because they predominantly express LFA-1 and Յ5% of periph- ␤ compared with a mixture of that mAb with another mAb or soluble -glu- eral blood T cells express Mac-1/CR3 (43, 44). Both types of cells can, with a single asterisk indicating p Ͻ 0.05 and two asterisks, p Ͻ 0.01. were analyzed for a possible role of uPAR and lectin site interactions in Mg2ϩ/EGTA-induced H-AFN adhesion to ICAM-1 (Fig. 42) but poorly, if at all, to LFA-1 (Fig. 5). Demonstration of LPS- 6). Unlike LFA-1-dependent L-AFN adhesion with neutrophils induced neutrophil Mac-1/CR3-dependent adhesion to fibrinogen (Fig. 4), T cell adhesion cannot be induced in the presence of ϩ required alteration of some of the assay conditions. Stimulation of physiologic concentrations of Ca2 , and adhesion induced by ϩ ϩ adhesion by LPS was relatively slow compared with PMA or Mg2 requires chelation of Ca2 with EGTA (38). Such H-AFN Mg2ϩ, requiring 40 min rather than 5 min, and its demonstration LFA-1-dependent T cell adhesion was inhibited almost completely was enhanced by accelerating neutrophil sedimentation by a brief by a single mAb to either CD11a or ICAM-1, while there was no 2-min centrifugation step before incubation at 37°C. Under these effect on adhesion by soluble ␤-glucan or mAbs to CD11b or conditions, adhesion of LPS-stimulated neutrophils to fibrinogen- uPAR. In addition, treatment of T cells or Jurkat cells with PiPLC coated surfaces was blocked 75% by anti-CD11b whereas anti- that removed 60–70% of detectable uPAR from Jurkat cells and CD11a failed to produce significant inhibition. With such adhesion Ͼ90% of detectable GPI-anchored CD55 and CD59 from T cells, that was primarily H-AFN Mac-1/CR3-dependent, anti-uPAR and had no detectable effect on their ability to adhere to ICAM-1- ϩ soluble ␤-glucan each produced much more inhibition of adhesion coated surfaces following stimulation with 10 mM Mg2 /EGTA (48 and 65%) than they did with immobilized ICAM-1 (23 and (data not shown). 25%) where adhesion was mediated by both L-AFN Mac-1/CR3 and LFA-1 (Fig. 4). However, because anti-uPAR and ␤-glucan Activation of Mac-1/CR3 for adhesion is associated with the each presumably functioned through a similar mechanism (i.e., formation of membrane clusters between Mac-1/CR3 and uPAR disruption of uPAR/CR3 membrane complexes) a mixture of anti- Previous research on resting neutrophils had demonstrated the uPAR and ␤-glucan did not produce significantly more inhibition presence of lectin-dependent membrane clusters of Mac-1/CR3 of adhesion than did either agent individually, and in addition, with Fc␥RIIIB and reversible lectin-dependent complexes between there was no augmentation of the inhibition of anti-CD11b pro- Mac-1/CR3 and uPAR that were induced by cell stimulation. How- duced by addition of either ␤-glucan or anti-uPAR to anti-CD11b ever, such uPAR/CR3 complexes were dissociated on the leading (Fig. 5). edge of polarized cells, making it unclear how such complexes 6422 LECTIN DOMAIN REGULATION OF Mac-1/CR3 ADHESION Downloaded from http://www.jimmunol.org/

FIGURE 6. Role of LFA-1 but not uPAR or a ␤-glucan-reactive lectin site in T-lymphocyte or Jurkat cell adhesion to immobilized ICAM-1. T lymphocytes or Jurkat cells stimulated with 10 mM Mg2ϩ/EGTA were incubated with or without mAb or ␤-glucan, and then the cells were added to wells coated with rICAM-1 and analyzed for adhesion. The results shown represent the mean Ϯ SD of more than or equal to assays conducted in triplicate. by guest on September 25, 2021 might function in adhesion and diapedesis of leukocytes (17, 45). The current study examined neutrophils for the formation of membrane clusters between Mac-1/CR3 and uPAR using RET of an excitation signal from FITC-labeled Mac-1/CR3 to TRITC-labeled uPAR (Fig. 7). With this system, excitation of TRITC by FITC requires that the excited FITC be within 7 nm of the TRITC (46). Neutrophils were stimulated by incubation in 10 mM Mg2ϩ/EGTA that generates adhesion via H-AFN LFA-1 and L-AFN Mac-1/ CR3. When the labeled cells in 10 mM Mg2ϩ/EGTA were maintained at 4°C, minimal RET was observed, whereas when the cells FIGURE 7. Modulating effect of 10 mM Mg2ϩ/EGTA on the asso- were warmed to 37°C for 20 min a strongly positive RET signal ciation of Mac-1/CR3 with uPAR on neutrophils but not on Jurkat cells. was observed indicating the formation of greatly increased num- Cells were stained with FITC-conjugated anti-CD11b (donor) or anti- bers of Mac-1/CR3 clusters with uPAR as compared with the CD11a (donor) and/or TRITC-conjugated anti-uPAR (acceptor), as described in Materials and Methods. Incubation of neutrophils in 10 sparse clusters on unstimulated cells (Fig. 7, top panel). A similar 2ϩ 2ϩ mM Mg /EGTA for 20 min at 37°C produced a significant increase in 37°C incubation in the absence of 10 mM Mg /EGTA stimula- RET (indicated by arrow) between Mac-1/CR3 and uPAR as shown in tion did not induce such an increase in uPAR clusters with Mac- the top panel (RET counts ϭ 13,010 Ϯ 895 at 4°C vs 16,023 Ϯ 962 at 1/CR3 (Fig. 7, middle panel). Finally, virtually no RET was ob- 37°C; p Ͻ 0.001). By contrast, in the absence of Mg2ϩ during a similar served between Mg/EGTA-induced H-AFN LFA-1 and uPAR incubation of neutrophils at 37°C, there was no change in the RET from ϩ when uPAR Jurkat T cells were examined (Fig. 7, bottom panel). Mac-1/CR3 to uPAR as shown in the middle panel (RET counts ϭ 13,531 Ϯ 886 at 4°C vs 12, 789 Ϯ 873 at 37°C; p Ͼ 0.05). When Jurkat GPI-anchored proteins mask specific C-terminal epitopes of cells were similarly tested for a Mg2ϩ/EGTA-induced RET from LFA-1 CD11b to uPAR, none was observed as shown in the bottom panel. Dotted lines in the histograms represent cells stained with FITC-conjugated anti- The lectin domain of Mac-1/CR3 was previously localized to the CD11b or anti-CD11a alone. These experiments were conducted five C-terminal region of CD11b in experiments that showed that li- times examining a total of 181 neutrophils for RET in the presence of ␤ ϩ ϩ gation of a 10 kDa -glucan to Mac-1/CR3 resulted in the selective Mg2 and three times without Mg2 examining a total of 136 neutro- masking of epitopes in the C-terminal region (8). A subsequent phils. Jurkat cells were examined four times and a total of 152 report mapped these C-terminal epitopes (47), allowing a compar- individual Jurkat cells were evaluated for RET. Representative micro- ison of epitope locations vs the amount that specific epitopes were scopic images are shown to the right of each fluorescence histogram. The Journal of Immunology 6423 masked by ␤-glucan attachment (Fig. 8). The CBRM1/23 mAb defines an epitope (Fig. 8, epitope subregion no. 4) that is both closest to the transmembrane domain of CD11b (47) and inhibited most by ␤-glucan binding (81%) to Mac-1/CR3 (8). By contrast, OKM1 defines an epitope subregion (no. 1) that is furthest from the transmembrane domain (47) and inhibited least (32%) by ␤-glucan binding to Mac-1/CR3 (8). The CBRM1/10 epitope (no. 3) is located adjacent and N-terminal to the epitope defined by CBRM1/23, and was masked by 74% following ␤-glucan ligation to Mac-1/CR3. These data suggest that the lectin domain responsible both for binding soluble ␤-glucan, and potentially also for generating lectin-dependent complexes between Mac-1/CR3 and GPI-anchored molecules, may be located near the transmembrane domain and epitope no. 4. Based on these data, experiments were conducted to determine whether the formation of Mac-1/CR3 complexes with uPAR blocked any of these C-terminal epitopes in a similar manner to FIGURE 9. Neutrophil activation for adhesion that stimulates uPAR soluble ␤-glucan (Fig. 9). Activation of neutrophils with PMA was binding to Mac-1/CR3 results in masking of the same C-terminal domain used to generate the CBRM1/5 neoepitope that is a reporter for the epitopes of CD11b that were previously shown to be masked by the bind- Downloaded from ␤ H-AFN MIDAS (48). PMA activation produced a 140% increase ing of -glucan to Mac-1/CR3. Neutrophils were activated by treatment with PMA, with or without the prior removal of GPI-anchored proteins in staining with CBRM1/5, and this increase was limited to 80% with PiPLC, and then monitored for staining with various FITC or Oregon by the removal of 70–80% of GPI-anchored proteins via PiPLC Green 488-labeled mAbs. The percent change in neutrophil staining vs before treatment with PMA (Fig. 9). Neutrophil treatment with untreated cells incubated in buffer medium alone is shown on the y-axis. PiPLC alone had no effect on CBRM1/5 staining. Treatment with Staining for the GPI-anchored proteins CD16b, uPAR, CD55, and CD59

PiPLC, with or without subsequent activation by PMA had no indicated that PiPLC removed only 70–90% of GPI-anchored proteins, and http://www.jimmunol.org/ significant effect on detection of the mAb24 H-AFN reporter neo- that PMA stimulation (without PiPLC treatment) caused the shedding of epitope that is now recognized to be located within the I-domain of 60–70% of membrane CD16b but no detectable shedding of uPAR, CD55, CD18 (49, 50) rather than a region common to CD11a or CD11b or CD59 (not shown). The results represent the mean Ϯ SD of more than Ͻ (37). Of particular note, treatment of neutrophils with PMA caused or equal to three experiments. Significant differences (p 0.05) in brack- a 25% reduction in staining for the CBRM1/23 epitope ( p Ͻ 0.05), eted comparisons are indicated by a single asterisk. despite the overall 90–100% increase in total Mac-1/CR3 surface expression as shown by staining with either OKM1 or I-domain- specific MN-41. This reduction in CBRM1/23 epitope staining de- epitope as compared with PMA-treated cells bearing native spite a large overall increase in Mac-1/CR3 expression suggested amounts of GPI-anchored proteins. However, even though there by guest on September 25, 2021 that the CBRM1/23 epitope might be masked by uPAR at its at- was no decrease in CBMR1/23 staining following PiPLC and tachment site to Mac-1/CR3. Removal of GPI-anchored proteins PMA treatments, there was still partial masking of the CBRM1/23 with PiPLC unmasked 40% additional CBRM1/23 epitopes on un- epitope because there was no net increase in CBRM1/23 staining activated neutrophils ( p Ͻ 0.05), and treatment with PMA after despite the overall 90–100% increase in total Mac-1/CR3 surface PiPLC resulted in significantly less masking of the CBRM1/23 expression. A similar but reduced unmasking effect by PiPLC was also noted with the CBRM1/10 epitope that is adjacent to the CBRM1/23 epitope. As with the CBRM1/23 epitope, the large increase in total Mac-1/CR3 surface expression induced by PMA was not detectable by a comparable increase in staining with CBRM1/10 indicating some masking of this epitope on newly expressed Mac-1/CR3. However, as with CBRM1/23, treatment of neutrophils with PiPLC to remove GPI-anchored proteins caused a significant increase in the amount of CBRM1/10 staining after PMA stimulation. By contrast, exposure of the OKM1 epitope that is separated by three epitope subdomains from the transmembrane region of CD11b, was not significantly affected by PiPLC treatment. PiPLC treatment had also no effect on expression of the I-domain epitope recognized by MN-41 (Fig. 9) whose expression is also unaffected by the presence of ␤-glucan bound to CR3 (8). FIGURE 8. Schematic diagram of the C-terminal region of CD11b These data suggest that GPI-anchored proteins such as CD16b and showing the location of epitopes recognized by specific mAbs (OKM1, uPAR bind to CD11b, masking the same C-terminal epitopes of CBRM1/16, etc.) as reported by Lu et al. (47). The numbers are the posi- CD11b that are masked by the binding of ␤-glucan to CR3. tions of amino acid residues within CD11b, and the bracketed subdomains numbered 1–4 contain the epitopes recognized by these mAbs. On the GPI-anchored proteins compete with soluble ␤-glucan for bottom are shown the percentage that each of the epitope subdomains are binding to Mac-1/CR3 blocked by the prior ligation of a small soluble ␤-glucan polysaccharide to Mac-1/CR3 (8). Each percentage value represents the mean from six tests If the removal of GPI-anchored proteins enhanced the detection of in which neutrophils were incubated first with ␤-glucan and then stained by the CBRM1/23 epitope because it was located adjacent to the lec- an individual mAb using indirect immunofluorescence. The percentages tin domain, then the removal of GPI-anchored proteins should en- shown for each epitope were derived from averaging the ␤-glucan masking hance exposure of the lectin site and promote CD11b-specific activity for each of the mAbs that bound to a specific epitope subdomain. binding of 125I-␤-glucan. Both untreated and PiPLC-treated cells 6424 LECTIN DOMAIN REGULATION OF Mac-1/CR3 ADHESION

totoxic degranulation in response to iC3b-opsonized target cells. Not only does ␤-glucan dissociate membrane complexes between uPAR and CR3 (51, 52), but also the current studies demonstrated that ␤-glucan also inhibits CR3-dependent adhesion to ICAM-1 or fibrinogen. Moreover, uPAR competes with ␤-glucan for binding to CR3. Finally, the generation of adhesion through formation of membrane complexes between uPAR and CR3 was shown to mask selectively the same C-terminal epitopes of CD11b that had been reported previously to be masked by the binding of soluble ␤-glucan to CR3. An unexpected finding was that uPAR complexes with Mac-1/ CR3 were involved in mediating adhesion via L-AFN as well as H-AFN binding sites. The current investigation used fluorescence microscopy measurements of RET to demonstrate the formation of FIGURE 10. ␤-Glucan binding to Mac-1/CR3 is enhanced following uPAR complexes with Mac-1/CR3 under conditions of neutrophil removal of GPI-anchored proteins from neutrophils and this enhancement stimulation with 10 mM Mg2ϩ/EGTA that did not generate either is reversed by sr-uPAR. Neutrophils were incubated with or without the CBRM1/5 reporter neoepitope or CD11b-dependent binding of PiPLC, and then a portion of the PiPLC-treated neutrophils were uPAR- fluid-phase ICAM-1. Previous research using RET analysis had reconstituted by incubation with 10 ␮g/ml sr-uPAR. Neutrophils were then Downloaded from tested for CD11b-specific binding of 125I-␤-glucan by determining the net shown similar uPAR complexes with Mac-1/CR3 when the H- binding of 125I-␤-glucan cpm following blockade with a mAb to CD11b. AFN MIDAS was generated through neutrophil stimulation with 1 ϩ The results represent the mean Ϯ SD for three assays. The increase in mM Mn2 /EGTA (17). With T cells, avid adhesion via L-AFN uptake of 125I-␤-glucan after PiPLC treatment was significant (p Ͻ 0.03), LFA-1 has been proposed to require cytoskeleton-dependent mem- 125 as was the decrease in I-␤-glucan uptake following sr-uPAR reconsti- brane clustering of LFA-1 (6). From the current data it is hypoth- tution of the PiPLC-treated cells (p Ͻ 0.05).

esized that L-AFN Mac-1/CR3 adhesion may involve a similar http://www.jimmunol.org/ clustering of L-AFN Mac-1/CR3 that is regulated by uPAR. Initial attempts to map the lectin site of CR3 by use of blocking exhibited CD11b-specific binding of 125I-␤-glucan that could be mAbs to specific epitopes of CD11b had been hampered by mAb- blocked by mAbs to CD11b (Fig. 10). Moreover, PiPLC-treated induced conformational changes that produced an allosteric block- neutrophils exhibited a 3.5-fold higher CD11b-specific binding of ade of ␤-glucan binding to CR3 such that mAbs specific for nearly 125I-␤-glucan than did untreated cells, despite the finding that such any epitope location on CD11b inhibited ␤-glucan binding to CR3 treatment with PiPLC did not increase the overall expression of (8). However, because of the relatively high binding affinity of Mac-1/CR3 (Fig. 9). Because PiPLC removes a proportion of all small ␤-glucans to CR3 (50 nM), it was possible to show that the various types of GPI-anchored proteins and not just uPAR, the attachment of a small (10 kDa) ␤-glucan to CR3 selectively in- by guest on September 25, 2021 specific activity of uPAR was explored by reconstituting a portion hibited the binding of mAbs directed to epitopes of CD11b that of the PiPLC-treated neutrophils with 10 ␮g/ml sr-uPAR in a man- were located in the large region of CD11b located C-terminal to ner that had been previously shown to restore membrane-bound both the I-domain and divalent-cation-binding region. The epitope uPAR (33). Reconstitution of uPAR to the PiPLC-treated neutro- that was most effectively masked by ␤-glucan attachment to CR3 125 ␤ phils reduced the level of CD11b-specific binding of I- -glucan was defined by mAb CBRM1/23 (8), and a subsequent report to a similar level as untreated neutrophils (Fig. 10). showed that this epitope was at the C-terminal end of CD11b, Although lectin-dependent complexes between GPI-anchored adjacent to the transmembrane domain (47). The current investi- Fc␥RIIIB and Mac-1/CR3 exist on unstimulated neutrophils (20), gation provided evidence for the masking of two C-terminal the current investigation suggested that additional lectin site-de- epitopes following activation of CR3 for adhesion and formation pendent complexes between Mac-1/CR3 and uPAR might be gen- of lectin-dependent membrane complexes between CR3 and erated with cellular activation for adhesion. It was thus hypothe- uPAR. As with blocking by ␤-glucan, the epitope most effectively sized that activation of neutrophils for adhesion would cause the masked by complex formation with uPAR was CBRM1/23, fol- lectin site to be masked by uPAR. Tests for the uptake of 125I-␤- lowed by the adjacent CBRM1/10 epitope. By contrast, CR3 ac- glucan (ϳ10 kDa) by neutrophils that had been incubated for 30 min at 37°C in RPMI 1640 medium containing 10 mM Mg2ϩ/1 tivation had no effect on detection of the OKM1 epitope that is at mM Ca2ϩ (that generates only L-AFN Mac-1/CR3 adhesion) the N-terminal end of the C-terminal region of CD11b. Taken to- showed that the Mac-1/CR3-specific uptake of 125I-␤-glucan was gether these data suggest that the lectin site may be near to the reduced from 4100 Ϯ 422 cpm to 1024 Ϯ 85 cpm (a reduction of CBRM1/23 epitope at the membrane proximal end of CD11b. 75%; p Ͻ 0.01) as compared with cells incubated in normal RPMI A direct competition between GPI-anchored proteins and 1640 medium (not shown). These data suggest that uPAR com- soluble ␤-glucan for binding to Mac-1/CR3 was demonstrated. petes with ␤-glucan for binding to the same lectin site of Mac-1/ Treatment of unstimulated neutrophils with PiPLC enhanced 125 CR3 and, along with RET (Fig. 7), serve as an additional demon- CD11b-dependent binding of I-␤-glucan, indicating that the lec- stration that a complex formation between uPAR and Mac-1/CR3 tin-dependent binding of CD16b that has been demonstrated with is associated with L-AFN as well as H-AFN adhesion. resting neutrophils probably also occurs via the same ␤-glucan- binding lectin site. Although the enhanced binding of 125I-␤-glu- Discussion can was reversed by addition of soluble uPAR, it probably would This investigation confirmed the hypothesis that uPAR, and prob- have been possible to accomplish the same blockade of the lectin ably also other GPI-anchored glycoproteins, form membrane com- site by reconstitution with soluble CD16b. In this regard, others plexes with CR3 via the same lectin site that had previously been have reported that soluble CD16b is taken up by neutrophils via a shown to be responsible for priming CR3 for phagocytosis or cy- lectin-dependent/sugar inhibitable binding to Mac-1/CR3 (53). The Journal of Immunology 6425

Because sugar-inhibitable complexes between LFA-1 and complex develops a binding site for vitronectin (60). A peptide Fc␥RIIIB had been shown (20), as well as the coimmunoprecipi- corresponding to this interaction site designated M25 and identi- tation of uPAR with LFA-1 (21), tests were also conducted to ﬁed initially by phage display was shown to inhibit both the uroki- determine whether uPAR or some other GPI-anchored receptor nase-induced adhesion to vitronectin as well as Mac-1/CR3/uPAR might also regulate LFA-1 adhesion to ICAM-1. Normal blood T adhesion to ﬁbrin (58). These data indicate that this interaction site cells do not express uPAR, but uPAR is a marker of activated T at residues 424–440 is required for development of H-AFN cells (40), and T cells migrating into tumors have been shown to MIDAS-dependent adhesion in addition to the lectin site located express uPAR (54). On neutrophils that express both LFA-1 and near the CBRM1/23 epitope at residues 943-1047. It is unknown CR3, adhesion to ICAM-1 was inhibited equally by mAbs to whether there is a different type of interaction at this secondary site CD11a and CD11b, and a mixture of both mAbs had an additive when urokinase binds to uPAR that results in disruption of the effect indicating that these integrins functioned together in medi- H-AFN MIDAS conformation or alternatively whether there is a ating adhesion to ICAM-1. Nevertheless, anti-uPAR alone inhib- third site of Mac-1/CR3/uPAR interaction. ited adhesion to the same extent as anti-CD11b or soluble ␤-glucan Although these studies failed to show a function of uPAR in alone, suggesting that uPAR functioned only with CR3. Moreover, LFA-1-dependent adhesion, other studies have shown an involve- ␤ ␤ ␤ there was only a slight additive effect on anti-CD11b inhibition of ment of uPAR in the adhesion function of 1, 3, and 5 integrins ␤ ␤ ␤ adhesion by mixing anti-CD11b with -glucan or anti-uPAR, in- (19, 61). Moreover, the complexes between uPAR and 1 or 3 dicating that each have the same CR3 target and cannot block integrins were shown to be disrupted by NADG in the same way ␤ the residual adhesion mediated by LFA-1. T lymphocytes and the as this sugar disrupts uPAR complexes with 2 integrins, suggest-

Jurkat T cell line differed from neutrophils in that PiPLC treatment ing that a lectin domain may be a common feature of integrins Downloaded from had no significant effect on adhesion to ICAM-1. Moreover, with allowing them to interact with uPAR (19). Jurkat cells that expressed uPAR on the majority of cells, there was also no effect on LFA-1 adhesion by anti-uPAR or ␤-glucan. These Acknowledgments experiments suggest that LFA-1 differs from CR3 in use of uPAR We thank Dr. Timothy A. Springer (Center for Blood Research) and Dr. for generating a H-AFN adhesion site. However, other conditions Nancy Hogg (Imperial Cancer Research Fund) for the generous donation of may be required to demonstrate a function for T cell uPAR be- several well-characterized mAbs specific for CD11a and CD11b. We also http://www.jimmunol.org/ cause experiments with uPAR-deficient mice have demonstrated a thank Dr. Martin Low (Columbia University) for providing the quantities reduced T cell migration into inflammatory sites resembling nor- of purified PiPLC required for these studies. Other colleagues who gener- mal mice treated with anti-CD11a or anti-ICAM-1 (13). ously provided mAbs or hybridomas include Dr. Lewis Lanier (University This investigation highlights the role of the lectin domain of of California, San Francisco) and Dr. Jay Unkeless (Mt. Sinai School of Medicine). We especially thank Dr. Donald Stanton (ICOS Pharmaceuti- CD11b in promoting the H-AFN MIDAS conformation in the dis- cals) for the generous gift of rICAM-1. tal I-domain. Previous studies had shown induction of the CBRM1/5 H-AFN MIDAS reporter neoepitope on neutrophils in- ␤ References cubated with soluble -glucan (11). In addition, the CD11b lectin 1. Springer, T. A. 1994. Traffic signals for lymphocyte recirculation and leukocyte by guest on September 25, 2021 domain binding to Candida albicans yeast cell walls had also been emigration: the multistep paradigm. Cell 76:301. shown to induce the H-AFN MIDAS conformation in the distal 2. Hogg, N., and C. Berlin. 1995. Structure and function of adhesion receptors in leukocyte trafficking. Immunol. Today 16:327. I-domain. It is of interest that when the C-terminal region of 3. Sugimori, T., D. L. Griffith, and M. A. Arnaout. 1997. Emerging paradigms of CD11b was replaced with the C-terminal region of CD11a, the integrin ligand binding and activation. Kidney Int. 51:1454. rMac-1/CR3 expressed a H-AFN MIDAS conformation and bound 4. Van Kooyk, Y., M. Lub, and C. G. Figdor. 1998. Adhesion and signaling mediated by the cytoplasmic tails of leucocyte integrins. Cell Adhes. Commun. to C. albicans in a sugar/lectin-independent manner (12). The lack 6:247. of need for a lectin domain in LFA-1 for adhesion corresponds to 5. Ross, G. D. 2000. Regulation of the adhesion versus cytotoxic functions of the ␣ ␤ the current data showing an apparent lack of use of GPI-anchored Mac-1/CR3/ M 2-integrin glycoprotein. Crit. Rev. Immunol. 20:197. 6. Stewart, M. P., C. Cabanãs, and N. Hogg. 1996. T cell adhesion to intercellular proteins in T cell adhesion. adhesion molecule-1 (ICAM-1) is controlled by cell spreading and the activation In what appears to be a contradiction to the anti-adhesive ac- of integrin LFA-1. J. Immunol. 156:1810. ␤ 7. Cain, J. A., S. L. Newman, and G. D. Ross. 1987. Role of complement receptor tivity of -glucan shown in this study, previous studies had shown type three and serum opsonins in the neutrophil response to yeast. Complement that soluble ␤-glucan could induce the CBRM1/5 epitope that is a 4:75. 8. Thornton, B. P., V. Vetvicka, M. Pitman, R. C. Goldman, and G. D. Ross. 1996. reporter for the high-affinity I-domain MIDAS. Nevertheless, this ␤ ␤ Analysis of the sugar specificity and molecular location of the -glucan-binding same report also noted that soluble -glucan failed to induce neu- lectin site of complement receptor type 3 (CD11b/CD18). J. Immunol. 156:1235. trophil spreading or adhesion (11). The current investigation sug- 9. Xia, Y., and G. D. Ross. 1999. Generation of recombinant fragments of CD11b gests that adhesion requires more than a H-AFN binding site, and expressing the functional ␤-glucan-binding lectin site of CR3 (CD11b/CD18). J. Immunol. 162:7285. additionally requires the formation of CR3 complexes with uPAR 10. Ross, G. D., J. A. Cain, B. L. Myones, S. L. Newman, and P. J. Lachmann. 1987. that may induce clustering of CR3 (55). Thus, although ␤-glucan Specificity of membrane complement receptor type three (CR3) for ␤-glucans. Complement 4:61. binding to CR3 induces its H-AFN binding site, it prevents adhe- 11. Vetvicka, V., B. P. Thornton, and G. D. Ross. 1996. Soluble ␤-glucan polysac- sion by blocking CR3 complex formation with uPAR. Such a charide binding to the lectin site of neutrophil or NK cell complement receptor membrane clustering of ␤ integrins has been proposed to be re- type 3 (CD11b/CD18) generates a primed state of the receptor capable of medi- 2 ating cytotoxicity of iC3b-opsonized target cells. J. Clin. Invest. 98:50. quired for adhesion via L-AFN or H-AFN binding sites (56, 57). 12. Forsyth, C. B., E. F. Plow, and L. Zhang. 1998. Interaction of the fungal pathogen Although the highly glycosylated uPAR appears to initiate its Candida albicans with integrin CD11b/CD18: recognition by the I domain is interaction with CR3 through the lectin domain, there are likely to modulated by the lectin-like domain and the CD18 subunit. J. Immunol. 161: 6198. be two or three additional sites of interaction between uPAR and 13. May, A. E., S. M. Kanse, L. R. Lund, R. H. Gisler, B. A. Imhof, and Mac-1/CR3. An additional interaction has been shown to occur at K. T. Preissner. 1998. Urokinase receptor (CD87) regulates leukocyte recruitment ␤ via 2 integrins in vivo. J. Exp. Med. 188:1029. a discrete site within the divalent cation-binding region that is 14. Gyetko, M. R., R. F. Todd III, C. C. Wilkinson, and R. G. Sitrin. 1994. The C-terminal to the I-domain at residues 424–440 (58). Following urokinase receptor is required for human monocyte chemotaxis in vitro. J. Clin. uPAR complex formation with Mac-1/CR3, occupation of uPAR Invest. 93:1380. 15. Sitrin, R. G., R. F. Todd III, H. R. Petty, T. G. Brock, S. B. Shollenberger, with its ligand urokinase causes Mac-1/CR3 to reverse its adhesion E. Albrecht, and M. R. Gyetko. 1996. The urokinase receptor (CD87) facilitates to ICAM-1 or fibrin (59), and secondarily, the uPAR portion of the CD11b/CD18-mediated adhesion of human monocytes. J. Clin. Invest. 97:1942. 6426 LECTIN DOMAIN REGULATION OF Mac-1/CR3 ADHESION

16. Gyetko, M. R., S. Sud, T. Kendall, J. A. Fuller, M. W. Newstead, and 38. McDowall, A., B. Leitinger, P. Stanley, P. A. Bates, A. M. Randi, and N. Hogg. T. J. Standiford. 2000. Urokinase receptor-deficient mice have impaired neutro- 1998. The I domain of integrin leukocyte function-associated antigen-1 is in- phil recruitment in response to pulmonary Pseudomonas aeruginosa infection. volved in a conformational change leading to high affinity binding to ligand J. Immunol. 165:1513. intercellular adhesion molecule 1 (ICAM-1). J. Biol. Chem. 273:27396. 17. Xue, W., A. L. Kindzelskii, R. F. Todd III, and H. R. Petty. 1994. Physical 39. Pyszniak, A. M., C. A. Welder, and F. Takei. 1994. Cell surface distribution of association of complement receptor type 3 and urokinase-type plasminogen ac- high-avidity LFA-1 detected by soluble ICAM-1-coated microspheres. J. Immu- tivator receptor in neutrophil membranes. J. Immunol. 152:4630. nol. 152:5241. 18. Petty, H. R., A. L. Kindzelskii, Y. Adachi, and R. F. Todd III. 1997. Ectodomain 40. Nykjaer, A., B. Moller, R. F. Todd III, T. Christensen, P. A. Andreasen, interactions of leukocyte integrins and pro-inflammatory GPI-linked membrane J. Gliemann, and C. M. Petersen. 1994. Urokinase receptor: an activation antigen proteins. J. Pharmacol. Biomed. Anal. 15:1405. in human T lymphocytes. J. Immunol. 152:505. 19. Xue, W., I. Mizukami, R. F. Todd III, and H. R. Petty. 1997. Urokinase-type 41. Altieri, D. C., R. Bader, P. M. Mannucci, and T. S. Edgington. 1988. Oligospeci- ␤ ␤ plasminogen activator receptors associate with 1 and 3 integrins of fibrosar- ficity of the cellular adhesion receptor MAC-1 encompasses an inducible recog- coma cells: dependence on extracellular matrix components. Cancer Res. 57: nition specificity for fibrinogen. J. Cell Biol. 107:1893. 1682. 42. Weber, C., and T. A. Springer. 1997. Neutrophil accumulation on activated, 20. Zhou, M., R. F. Todd III, J. G. J. Van de Winkel, and H. R. Petty. 1993. Co- surface-adherent platelets in flow is mediated by interaction of Mac-1 with fi- ␣ ␤ capping of the leukoadhesin molecules complement receptor type 3 and lympho- brinogen bound to IIb 3 and stimulated by platelet-activating factor. J. Clin. cyte function-associated antigen-1 with Fc␥ receptor III on human neutrophils: Invest. 100:2085. possible role of lectin-like interactions. J. Immunol. 150:3030. 43. Miller, L. J., R. Schwarting, and T. A. Springer. 1986. Regulated expression of 21. Bohuslav, J., V. Horejsı´, C. Hansmann, J. Sto¨ckl, U. H. Weidle, O. Majdic, the Mac-1, LFA-1, p150,95 glycoprotein family during leukocyte differentiation. I. Bartke, W. Knapp, and H. Stockinger. 1995. Urokinase plasminogen activator J. Immunol. 137:2891. ␤ receptor, 2-integrins, and Src-kinases within a single receptor complex of human 44. Muto, S., V. Vetvicka, and G. D. Ross. 1993. CR3 (CD11b/CD18) expressed by monocytes. J. Exp. Med. 181:1381. cytotoxic T cells and NK cells is upregulated in a manner similar to neutrophils 22. Xia, Y., V. Vetvicka, J. Yan, M. Hanikyrova, T. N. Mayadas, and G. D. Ross. following stimulation with various activating agents. J. Clin. Immunol. 13:175. 1999. The ␤-glucan-binding lectin site of mouse CR3 (CD11b/CD18) and its 45. Kindzelskii, A. L., Z. O. Laska, R. F. Todd III, and H. R. Petty. 1996. Urokinase- function in generating a primed state of the receptor that mediates cytotoxic type plasminogen activator receptor reversibly dissociates from complement re- ␣ ␤ Downloaded from activation in response to iC3b-opsonized target cells. J. Immunol. 162:2281. ceptor type 3 ( M 2, CD11b/CD18) during neutrophil polarization. J. Immunol. 23. Janusz, M. J., K. F. Austen, and J. K. Czop. 1989. Isolation of a yeast heptaglu- 156:297. coside that inhibits monocyte phagocytosis of zymosan particles. J. Immunol. 46. Zhou, M.-J., H. Poo, R. F. Todd III, and H. R. Petty. 1992. Surface-bound im- 142:959. mune complexes trigger transmembrane proximity between complement receptor 24. Myones, B. L., J. G. Dalzell, N. Hogg, and G. D. Ross. 1988. Neutrophil and type 3 and the neutrophil’s cortical microfilaments. J. Immunol. 148:3550. monocyte cell surface p150,95 has iC3b-receptor (CR4) activity resembling CR3. 47. Lu, C. F., C. Oxvig, and T. A. Springer. 1998. The structure of the ␤-propeller J. Clin. Invest. 82:640. domain and C-terminal region of the integrin ␣ M subunit-dependence on ␤ 25. Diamond, M. S., and T. A. Springer. 1993. A subpopulation of Mac-1 (CD11b/ subunit association and prediction of domains. J. Biol. Chem. 273:15138.

CD18) molecules mediates neutrophil adhesion to ICAM-1 and fibrinogen. 48. Oxvig, C., C. Lu, and T. A. Springer. 1999. Conformational changes in tertiary http://www.jimmunol.org/ J. Cell Biol. 120:545. structure near the ligand binding site of an integrin I domain. Proc. Natl. Acad. 26. Diamond, M. S., J. Garcia-Aguilar, J. K. Bickford, A. L. Corbi, and Sci. USA 96:2215. T. A. Springer. 1993. The I domain is a major recognition site on the leukocyte 49. Lu, C., M. Shimaoka, Q. Zang, J. Takagi, and T. A. Springer. 2001. Locking in ␣ ␤ integrin Mac-1 (CD11b/CD18) for four distinct adhesion ligands. J. Cell Biol. alternate conformations of the integrin L 2 I domain with disulfide bonds re- 120:1031. veals functional relationships among integrin domains. Proc. Natl. Acad. Sci. 27. Dransfield, I., and N. Hogg. 1989. Regulated expression of Mg2ϩ binding epitope USA 98:2393. on leukocyte integrin ␣ subunits. EMBO J. 8:3759. 50. Kamata, T., K. K. Tieu, T. Tarui, W. Puzon-McLaughlin, N. Hogg, and ␤ 28. Todd, R. F., III, and M. A. Arnaout. 1986. Monoclonal antibodies that identify Y. Takada. 2002. The role of the CPNKEKEC sequence in the 2 subunit I ␣ ␤ Mo1 and LFA-1, two human leukocyte membrane glycoproteins: a review. In domain in regulation of integrin L 2 (LFA-1). J. Immunol. 168:2296. Leukocyte Typing II. Volume 3: Human Myeloid and Hematopoietic Cells. 51. Kindzelskii, A. L., W. Xue, R. F. Todd III, L. A. Boxer, and H. R. Petty. 1994. E. L. Reinherz, B. F. Haynes, L. M. Nadler, and I. D. Bernstein, eds. Springer- Aberrant capping of membrane proteins on neutrophils from patients with leu-

Verlag, New York, p. 95. kocyte adhesion deficiency. Blood 83:1650. by guest on September 25, 2021 ␤ 29. Mizukami, I. F., S. D. Vinjamuri, R. D. Trochelman, and R. F. Todd III. 1990. 52. Todd, R. F., III, and H. R. Petty. 1997. 2(CD11/CD18) integrins can serve as A structural characterization of the Mo3 activation antigen expressed on the signaling partners for other leukocyte receptors. J. Lab. Clin. Med. 129:492. plasma membrane of human mononuclear phagocytes. J. Immunol. 144:1841. 53. Galon, J., J. F. Gauchat, N. Mazie`res, R. Spagnoli, W. Storkus, M. Lo¨tze, 30. Mizukami, I. F., B. A. Garni-Wagner, L. M. DeAngelo, M. Liebert, A. Flint, J. Y. Bonnefoy, W. H. Fridman, and C. Saute`s. 1996. Soluble Fc␥ receptor type D. A. Lawrence, R. L. Cohen, and R. F. Todd III. 1994. Immunologic detection III (Fc␥RIII:CD16) triggers cell activation through interaction with complement of the cellular receptor for urokinase plasminogen activator. Clin. Immunol. Im- receptors. J. Immunol. 157:1184. munopathol. 71:96. 54. Bianchi, E., E. Ferrero, F. Fazioli, F. Mangili, J. Wang, J. R. Bender, F. Blasi, and 31. Litwin, V., J. Gumperz, P. Parham, J. H. Phillips, and L. L. Lanier. 1994. NKB1: R. Pardi. 1996. Integrin-dependent induction of functional urokinase receptors in A natural killer cell receptor involved in the recognition of polymorphic HLA-B primary T lymphocytes. J. Clin. Invest. 98:1133. molecules. J. Exp. Med. 180:537. 55. Sitrin, R. G., P. M. Pan, H. A. Harper, R. F. Todd III, D. M. Harsh, and 32. Martin, S., J. M. Casasnovas, D. E. Staunton, and T. A. Springer. 1993. Efficient R. A. Blackwood. 2000. Clustering of urokinase receptors (uPAR; CD87) induces neutralization and disruption of rhinovirus by chimeric ICAM-1/immunoglobulin proinflammatory signaling in human polymorphonuclear neutrophils. J. Immunol. molecules. J. Virol. 67:3561. 165:3341. 33. Mizukami, I. F., and R. F. Todd III. 1998. A soluble form of the urokinase 56. Stewart, M., and N. Hogg. 1996. Regulation of leukocyte integrin function: af- plasminogen activator receptor (suPAR) can bind to hematopoietic cells. J. Leu- finity vs. avidity. J. Cell. Biochem. 61:554. kocyte Biol. 64:203. 57. Hogg, N., and B. Leitinger. 2001. Shape and shift changes related to the function 34. Ross, G. D., and J. D. Lambris. 1982. Identification of a C3bi-specific membrane of leukocyte integrins LFA-1 and Mac-1. J. Leukocyte Biol. 69:893. complement receptor that is expressed on lymphocytes, monocytes, neutrophils, 58. Simon, D. I., Y. Wei, L. Zhang, N. K. Rao, H. Xu, Z. Chen, Q. Liu, S. Rosenberg, and erythrocytes. J. Exp. Med. 155:96. and H. A. Chapman. 2000. Identification of a urokinase receptor-integrin inter- 35. Zhou, M.-J., R. F. Todd III, and H. R. Petty. 1991. Detection of transmembrane action site: promiscuous regulator of integrin function. J. Biol. Chem. 275:10228. linkages between immunoglobulin or complement receptors and the neutrophil’s 59. Simon, D. I., N. K. Rao, H. Xu, Y. Wei, O. Majdic, E. Ronne, L. Kobzik, and cortical microfilaments by resonance energy transfer microscopy. J. Mol. Biol. H. A. Chapman. 1996. Mac-1 (CD11b/CD18) and the urokinase receptor (CD87) 218:263. form a functional unit on monocytic cells. Blood 88:3185. 36. Sehgal, G., K. Zhang, R. F. Todd III, L. A. Boxer, and H. R. Petty. 1993. Lectin- 60. Wei, Y., D. A. Waltz, N. Rao, R. J. Drummond, S. Rosenberg, and like inhibition of immune complex receptor-mediated stimulation of neutrophils: H. A. Chapman. 1994. Identification of the urokinase receptor as an adhesion effects on cytosolic calcium release and superoxide production. J. Immunol. 150: receptor for vitronectin. J. Biol. Chem. 269:32380. 4571. 61. Yebra, M., G. C. N. Parry, S. Stromblad, N. Mackman, S. Rosenberg, 37. Dransfield, I., C. Cabanãs, J. Barrett, and N. Hogg. 1992. Interaction of leukocyte B. M. Mueller, and D. A. Cheresh. 1996. Requirement of receptor-bound uroki- ␣ ␤ integrins with ligand is necessary but not sufficient for function. J. Cell Biol. nase-type plasminogen activator for integrin v 5-directed cell migration. J. Biol. 116:1527. Chem. 271:29393.