The JI PBL Intereron Source
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Moffatt, O. D.
Right arrow Articles by Gregory, C. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Moffatt, O. D.
Right arrow Articles by Gregory, C. D.
The Journal of Immunology, 1999, 162: 6800-6810.
Copyright © 1999 by The American Association of Immunologists

Macrophage Recognition of ICAM-3 on Apoptotic Leukocytes1

Odette D. Moffatt*, Andrew Devitt*, Elaine D. Bell{dagger}, David L. Simmons{dagger},{ddagger} and Christopher D. Gregory2,*

* Institute of Cell Signaling and School of Biomedical Sciences, University of Nottingham Medical School, Queen’s Medical Centre, Nottingham, United Kingdom; {dagger} Institute of Molecular Medicine, John Radcliffe Hospital, Headington, Oxford, United Kingdom; and {ddagger} SmithKline Beecham Pharmaceuticals, Harlow, Essex, United Kingdom


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Cells undergoing apoptosis are cleared rapidly by phagocytes, thus preventing tissue damage caused by loss of plasma membrane integrity. In this study, we show that the surface of leukocytes is altered during apoptosis such that the first Ig-like domain of ICAM-3 (CD50) can participate in the recognition and phagocytosis of the apoptotic cells by macrophages. Macrophage recognition of apoptotic cell-associated ICAM-3 was demonstrated both on leukocytes and, following transfection of exogenous ICAM-3, on nonleukocytes. The change in ICAM-3 was a consistent consequence of apoptosis triggered by various stimuli, suggesting that it occurs as part of a final common pathway of apoptosis. Alteration of ICAM-3 on apoptotic cells permitting recognition by macrophages resulted in a switch in ICAM-3-binding preference from the prototypic ICAM-3 counterreceptor, LFA-1, to an alternative macrophage receptor. Using mAbs to block macrophage/apoptotic cell interactions, we were unable to obtain evidence that either the alternative ICAM-3 counterreceptor {alpha}dß2 or the apoptotic cell receptor {alpha}vß3 was involved in the recognition of ICAM-3. By contrast, mAb blockade of macrophage CD14 inhibited ICAM-3-dependent recognition of apoptotic cells. These results show that ICAM-3 can function as a phagocytic marker of apoptotic leukocytes on which it acquires altered macrophage receptor-binding activity.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Cell death by apoptosis plays a significant role in the control of cell population sizes (1). Breakdown in the regulation of this process contributes to a variety of diseases, including autoimmune and neurodegenerative disorders, immunodeficiency, and cancer (2). Apoptosis is a cell-autonomous process with two main consequences: 1) the organized degradation of the cell, a well-studied series of events involving the caspase family of proteases (3), and one or more DNases (4), and 2) rapid recognition and engulfment of the cell and its membrane-bound fragments (apoptotic bodies) by phagocytes (5). The mechanisms underlying the latter are, as yet, ill understood, but the capacity to be phagocytosed rapidly is a critically important facet of apoptosis (5, 6, 7). Thus, clearance of cells before loss of plasma membrane integrity, as occurs in apoptosis, militates against inflammation and neighboring histotoxicity. Such catastrophic consequences of cell death are the hallmarks of necrosis (1), an accidental form of cell death with no physiological relevance. Although engulfment of apoptotic cells can be mediated by nonprofessional phagocytes, such as neighboring epithelial cells and fibroblasts, professional phagocytes, the macrophages, play a major role in clearance in areas of high frequency apoptosis (1, 5).

In vitro studies of the mechanisms of recognition and phagocytosis of apoptotic cells have to date implicated, on the macrophage surface, five glycoprotein receptors, the vitronectin receptor {alpha}vß3 (8), CD36 (9), the ATP-binding cassette transporter ABC-1 (10), class A scavenger receptor (11), and CD14 (12) in apoptotic cell clearance. The oxidized low density lipoprotein receptor (CD68) may also play a role (13). On the apoptotic cell, however, only one cell surface change, the exposure of phosphatidylserine, has been firmly linked with the clearance mechanism (14), although there is some evidence that carbohydrate changes on the surface of apoptotic cells may also be important (15, 16, 17). In this work, we define a novel change in the cell surface of apoptotic leukocytes involving ICAM-3 (CD50) that allows apoptotic-leukocyte recognition by phagocytes.

ICAM-3 is a highly glycosylated Ig-superfamily member that is constitutively expressed on leukocytes (18, 19, 20, 21, 22). It has five extracellular Ig-like domains and binds to the leukointegrin LFA-1 (CD11a ({alpha}L)/CD18 (ß2)) through its most membrane-distal Ig-like domain (D1)3 (23, 24, 25). It also binds to the LFA-1-related leukointegrin, {alpha}dß2 (26). Functional studies to date define ICAM-3 as an adhesion molecule with signal-transducing functions involved in immune regulation. Thus, while the functional consequences of binding of ICAM-3 to {alpha}dß2 have yet to be realized, LFA-1/ICAM-3 interactions between APC and resting T cells provide costimulatory signals that are important for initiating immune responses (20, 27, 28, 29, 30). In addition, homotypic LFA-1/ICAM-3 interactions can exert inhibitory effects on proliferating T cells (31), and ligation of ICAM-3 by mAbs has been found to increase the rate of apoptosis in human thymocytes (31, 32). Roles for ICAM-3 in signal transduction are implied from its ability to induce intracellular Ca2+ mobilization, to undergo phosphorylation on tyrosine, and to associate with the tyrosine phosphatase, CD45 (33, 34, 35). Soluble forms of ICAM-3 have been found to be elevated in the circulation of patients with autoimmune disorders (36, 37), but the significance of soluble ICAM-3 is as yet unknown.

In this work, we present evidence for a novel function of ICAM-3 in the recognition of apoptotic leukocytes by macrophages. The results show that apoptotic leukocyte-associated ICAM-3 interacts with macrophages via D1 in a leukointegrin-independent manner.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Monoclonal Abs

The anti-ICAM-3 and anti-LFA-1 mAbs used in these studies are summarized in Table IGo. Anti-ICAM-3 mAbs CAL 3.10, CAL 3.38, CAL 3.41, and ICAM-3.3 (CH3.3) were produced in the laboratory of D.L.S. (38). Gifts of purified CAL 3.10 and 182B were made by Dr. Ian Collins, R&D Systems (Abingdon, U.K.), and by Dr. Joel Hayflick (ICOS, Bothel, WA), respectively. BU68 was kindly provided by Debbie Hardie (Department of Immunology, University of Birmingham, U.K.), and 3A9 was provided by Innogenetics (Gent, Belgium). ICAM-3.3 was purchased from R&D Systems (Minneapolis, MN). Other anti-ICAM-3 mAbs (KS128, BY44, 186-269, B-N2, B-P12, B-R1, AZN-ICAM-3.1) and anti-LFA-1 mAbs were obtained from the Sixth International Workshop on Leukocyte Differentiation Antigens. The ß2-specific mAb KIM 127 and the {alpha}v-specific mAb 13C2 were kindly provided by Dr. Martin Robinson (Celltech, Slough, U.K.) and by Professor Michael Horton (University College, London, U.K.), respectively. The {alpha}d-specific mAbs 212D, 217K, and 217L were generously supplied by Dr. Mike Gallatin and Pat Hoffman (ICOS). Anti-CD14 mAbs 61D3 and 63D3 were obtained as described (39).


View this table:
[in this window]
[in a new window]
  		Table I. Summary of specificities and functions of anti-ICAM-3 and anti-LFA-1 mAbs used

 
Cells

The group I Burkitt lymphoma (BL) cell line MUTU (40), the Jurkat T cell line, and the erythroleukemia line K562 were cultured in suspension in RPMI 1640 medium (Life Technologies, Grand Island, NY) containing 10% Serum Supreme (BioWhittaker, Walkersville, MD). K562 cells stably expressing LFA-1 (KL/4 cells) were kindly provided by Dr. Martin Robinson (Celltech, Slough, U.K.). COS-1 cells were grown on 15-cm dishes (Falcon) in DMEM (Life Technologies) containing 10% FCS and 2 mM glutamine. The human embryonic kidney-derived lines HEK 293 and HEK 293T were kindly provided by Dr. Barbara Spruce (University of Dundee, Dundee, U.K.). Transient transfection of 293T cells was conducted by the calcium phosphate method using human ICAM-3 (18) in pCDM8. Immunofluorescence staining or apoptosis induction was conducted 24 h after transfection. Macrophages were obtained by culture of human monocytes obtained by Percoll fractionation of defibrinated venous blood, as described (12). Adherent cells were maintained for 7 days on glass multiwell slides (Hendley, Essex, U.K.) in IMDM (Life Technologies) containing 10% heat-inactivated autologous serum. Neutrophil granulocytes were obtained by Percoll fractionation of EDTA-treated venous blood following erythrocyte sedimentation, as described (41).

Induction and measurement of apoptosis

Apoptosis was induced in BL cells by culture with ionomycin (1 µg/ml; Calbiochem, La Jolla, CA) or staurosporine (1 µM; Sigma, St. Louis, MO) for 18 h at 37°C or by cold-shock treatment: maintenance of cells at 1°C for 4 h and subsequent return to 37°C for an additional 4 h (42). In some experiments, ionomycin-induced apoptosis was followed over a period of 2–18 h. Staurosporine treatment was also used to induce apoptosis in Jurkat T cells and HEK 293 (T) cells. Transiently transfected HEK/ICAM-3 cells displayed identical responses to staurosporine as parental HEK cells. Default apoptosis in neutrophils was obtained by culture of freshly isolated cells in RPMI 1640 containing 10% Serum Supreme for 24 h at 37°C. Apoptosis was assessed either by fluorescence microscopy of acridine orange-stained samples (43) or by flow-cytometric light scatter analysis of unstained samples (44) using the Coulter (Palo Alto, CA) XL flow cytometer.

Macrophage/apoptotic cell interaction assay

Apoptotic cell populations were coincubated for 1 h at 37°C with 7-day monocyte-derived macrophages, as we have previously described (12, 39). For studies of viable cell binding, untreated BL cell populations were coincubated with macrophages. Where indicated, Abs were included throughout the coincubation period. All Abs were used at concentrations that were saturating according to cell surface immunofluorescence staining: ascitic fluids were used at 1:100–1:200; purified mAbs were used at 10–40 µg/ml. In some experiments, mAb treatment of either apoptotic cells or macrophages was conducted before the interaction assay. Before use, mAb-treated cells were washed twice in RPMI. In experiments using KIM 127, mAb was included in the interaction assays at 10 µg/ml. All interaction assays were stopped by dipping slides in ice-cold PBS; three changes were used to wash away unbound cells. Slides were fixed in methanol and stained in Jenner-Giemsa (39). Coded slides were scored by light microscopy by two independent observers: the number of macrophages interacting with apoptotic cells was recorded using established criteria (39) for 200 macrophages counted per well, with at least duplicate wells being scored for each experimental sample. Data are generally expressed as percentage of macrophages interacting with apoptotic cells. In certain experiments, viable cells were observed to become interactive with macrophages in these assays, in which case the morphology of the macrophage-bound cells was noted and assays were additionally scored as percentage of macrophages interacting with viable cells. In experiments in which K562 and KL/4 cells were used as surrogate macrophages, cells adherent to plastic coverslips in serum-free RPMI were treated as described for macrophages. Since recognition of apoptotic cells by human macrophages can vary widely with macrophage donor, the majority of the results are presented as representative experiments of at least three using different donors. Unless otherwise stated, results are expressed as mean ± SD of replicate samples with statistical comparisons of untreated (apoptotic cells alone) and mAb treated, samples being conducted using Student’s t test.

Immunofluorescence staining and flow cytometry

Indirect immunofluorescence staining was conducted as previously described (39) using saturating concentrations of primary Abs and goat anti-mouse Ig FITC (1:50; Sigma) for visualization. Macrophages were cultured for 7 days in 25-cm2 flasks (Life Technologies) and stained in suspension after removal using 5 mM EDTA and gentle scraping. Immunofluorescence profiles were analyzed using the Coulter XL flow cytometer. Light scatter properties were used to gate viable and apoptotic cells, as described (44).

Soluble rICAM-3 proteins

The production of ICAM-3(D1-D2)-Fc (extracellular D1 and D2 fused to the Fc region of human IgG1) and the derivation of single point mutants of this protein have been described (18, 23, 25). D1 mutants were: E2K (glutamic acid residue at position 2 substituted by lysine residue), E2A, E8K, E8A, D27A, E32K, K33D, K33A, E43A, L66K, S68K, Q75A, and Q75H. D2 mutants were: H155D, H155A, G156K, and P158K. Wild-type (wt) or mutant chimeric plasmids were transiently expressed in COS-1 cells cultured in DMEM containing 0.1% IgG-free FCS after transfection using DEAE-dextran. Recombinant proteins were collected from the supernatants over 7 days and purified on protein A columns (Pharmacia, Piscataway, NJ). Reactivity of ICAM-3 mAbs with recombinant proteins was assessed by ELISA. Ninety-six-well immunoplates (Nunc) were coated with sheep anti-human IgG (The Binding Site, Birmingham, U.K.; 5 µg/ml in 0.05 M carbonate buffer, pH 9.6) and loaded with purified ICAM-3(D1-D2)-Fc protein (50 µg/ml) for 1 h at 37°C. Plates were incubated with saturating concentrations of ICAM-3 mAbs for 1 h at 37°C, followed by sheep anti-mouse HRP conjugate (1:5000, provided by Chandra Raykundalia, University of Birmingham, Birmingham, U.K.) for a further hour. Extensive washing between all steps was conducted with PBS containing 0.05% Tween-20 (PBS-T), and all reagents were diluted in PBS-T. o-phenylenediamine substrate (Sigma) was added in the presence of H2O2 and, after 30 min at 37°C, the reaction was stopped and absorbances at 492 nm were read using an Anthos 2001 plate reader.

Western blotting

ICAM-3 was detected in Western blots of HEK 293T cell lysates. ICAM-3-transfected or mock-transfected cells were harvested 40 h posttransfection. Where indicated, staurosporine (1 µM) was included for the last 16 h. Sonicated lysates were solubilized in Laemmli buffer and subjected to SDS-PAGE (25 µg total protein per lane). Western blots were probed with a mixture of ICAM-3 mAbs, CAL 3.38 (D1 specific) and 182B (D4–5 specific, kindly provided by Dr. J. Hayflick), and visualized using ECL (Amersham, Arlington Heights, IL).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Interaction of apoptotic B lymphocytes with macrophages is inhibited by ICAM-3 mAbs 3A9 and BU68

We have established previously that, following induction of apoptosis, B lymphocytes obtained from BL cell lines of group I (biopsy-like) phenotype (40) bind to, and are phagocytosed by, human monocyte-derived macrophages in vitro (39). Viable B cells from such lines do not interact with macrophages in this way ((39) and see later: Fig. 8Go). Following preliminary investigations (45), we studied a panel of 13 ICAM-3 mAbs (Table IGo) in assays of macrophage recognition of apoptotic B cells and identified two, 3A9 and BU68, that markedly inhibited macrophage/apoptotic B cell interactions (Fig. 1Go). The degree of inhibition was found to be comparable with that of mAb 61D3, which, as we have previously shown, inhibits macrophage/apoptotic B cell interactions through binding to macrophage CD14 (12). The mAbs 3A9 and BU68 were found both to inhibit the number of macrophages interacting with apoptotic B cells and, of the B cell-bound macrophages, the number of apoptotic cells interacting per macrophage (Fig. 1Go and data not shown).



View larger version (56K):
[in this window]
[in a new window]
  		FIGURE 8. Preferential binding of viable, but not apoptotic, BL cells to LFA-1-active macrophages. A, Macrophages (±KIM 127) were assessed for their ability to interact with apoptotic cells (AC, speckled bars) and viable cells (black bars) after coincubation with BL cell populations, which were either predominantly apoptotic (ionomycin treated, 68% apoptotic, shown to the left of the vertical line) or predominantly viable (untreated, 25% apoptotic, shown to the right of the vertical line). Means ± SD. Experiment representative of three similar. Student’s t test: a, interacting VC (AC alone), untreated vs KIM 127 treated, p < 0.05; b, interacting VC (KIM 127 treated), AC alone vs 7E4 treated, p < 0.05; c, interacting VC (VC alone), untreated vs KIM 127 treated, p < 0.05; d, interacting VC (KIM 127 treated), VC alone vs 7E4 treated, p < 0.01; e, interacting VC (KIM 127 treated), VC alone vs CAL 3.38 treated, p < 0.05; f, interacting AC (KIM 127 treated), VC alone vs BU68 treated, p < 0.05; g, interacting VC (KIM 127 treated), VC alone vs BU68 treated, p < 0.05. B, Morphological detail of BL cells (using Jenner Giemsa stain) interacting with untreated 7-day monocyte-derived macrophage (left) or KIM 127-activated macrophage (right). Note viable morphology of clustered BL cells at right. Arrows: apoptotic cells.

 


View larger version (58K):
[in this window]
[in a new window]
  		FIGURE 1. Inhibition of interaction of apoptotic B cells with macrophages by anti-ICAM-3 mAbs BU68 and 3A9. Apoptotic BL cells were coincubated with 7-day monocyte-derived macrophages for 1 h in the presence of the indicated mAbs. Effects of anti-ICAM-3 mAbs (hatched bars) in comparison with the CD14 mAbs 61D3 (blocker) and 63D3 (nonblocker). Epitopes of anti-ICAM-3 mAbs: D1 = domain 1, D2 = domain 2, D1-D2 = interdomain 1 and 2. AC, ionomycin-treated apoptotic cells. Data are means ± SD. Student’s t test: ***, p < 0.001; **, p < 0.01; *, p < 0.05. Experiments shown in A and B are separate experiments, each representative of three identical.

 
We also determined whether the nature of the stimulus for apoptosis affected the involvement of ICAM-3 in macrophage recognition of apoptotic B cells. Three triggers of apoptosis were compared: 1) the Ca2+ ionophore ionomycin, 2) the protein kinase C-inhibitor staurosporine, and 3) low temperature pretreatment (cold shock (42)). As shown in Fig. 2Go, while the capacity of B cell populations to interact with macrophages varied with the stimulus for apoptosis, involvement of ICAM-3 in the macrophage/apoptotic cell interactions was remarkably consistent, with BU68 (but not CAL 3.38) causing similar relative levels of inhibition regardless of the apoptosis-inducing stimulus.



View larger version (41K):
[in this window]
[in a new window]
  		FIGURE 2. Occurrence of ICAM-3-dependent macrophage/apoptotic cell interactions with different apoptosis-inducing stimuli. BL cells were induced to undergo apoptosis by treatment with ionomycin, staurosporine, or by brief incubation at 1°C (cold shock), and subsequently exposed to macrophages in the presence or absence of BU68 or CAL 3.38. Data are means ± SD. Experiment representative of three identical. Student’s t test: ***, p < 0.001; **, p < 0.01; *, p < 0.05.

 
These results confirm and extend our initial observations with mAb 3A9 (45) and demonstrate that apoptotic B lymphocytes, apparently independently of the apoptosis-inducing stimulus, interact with macrophages through a mechanism involving specific site(s) on ICAM-3 that can be blocked by 3A9 and BU68, but not by 11 other ICAM-3 mAbs.

BU68 and 3A9 map to similar epitopes in D1 of ICAM-3

To obtain more information about the region of ICAM-3 involved in apoptotic-leukocyte binding to macrophages, wt and mutant chimeric ICAM-3-Fc fusion proteins were used to map the epitopes recognized by BU68 and 3A9, together with those of a panel of ICAM-3 mAbs (CAL 3.10, CAL 3.38, CAL 3.41, 186-269, B-P12, and AZN-ICAM3.1) that failed to inhibit in assays of macrophage recognition of apoptotic cells (see Fig. 1Go). As shown in Fig. 3Go, all mAbs bound effectively to the wt ICAM-3(D1-D2)-Fc recombinant protein and to D2 mutants, confirming their D1 specificity (Table IGo). The patterns of reactivity of BU68 and 3A9 with various mutant ICAM-3(D1-D2)-Fc recombinant proteins were remarkably similar, being clearly distinct from those of CAL 3.10, CAL 3.38, CAL 3.41, 186–269, B-P12, and AZN-ICAM3.1 (Fig. 3Go). While the K33D mutation resulted in the loss of binding to ICAM-3 of all eight mAbs tested (without affecting D2 mAb binding; data not shown), E32K, K33A, S68K, and Q75H mutations particularly affected binding of BU68 and 3A9. Thus, only BU68 and 3A9 were affected by all four of these mutations (Fig. 3Go). In addition, the Q75A mutation markedly inhibited binding of BU68, but not 3A9. According to recent models of ICAM-3 structure (23, 24, 25), these four residues lie either on, or close to, the CFG face of ICAM-3 D1, which has been implicated as forming part of the binding site for LFA-1.



View larger version (71K):
[in this window]
[in a new window]
  		FIGURE 3. Reactivity of ICAM-3 mAbs against recombinant chimeric ICAM-3(D1-D2)-Fc mutants. The indicated mAbs were tested in ELISA against the mutants shown on the y-axis. Data (absorbances) are expressed in relation to reactivity with the wt recombinant protein (first black bar), which was normalized at 100% for each experiment. Results are means ± SEM of three separate experiments. CD14-Fc (second black bar) is included as a negative control recombinant chimeric protein. Clear and hatched bars delineate D1 mutants, grey bars D2 mutants. Of the D1 mutants, the hatched bars highlight a pattern of four mutations that inhibited the binding of BU68 and 3A9 only.

 
These results suggest strongly that the epitopes recognized by BU68 and 3A9 are closely similar to each other, and also overlap, at least partially, with the LFA-1 binding site in D1 of ICAM-3.

Interaction of apoptotic B cells with macrophages involves B cell-associated, but not macrophage-associated, ICAM-3

ICAM-3 is expressed at the surface of the macrophages and B cells used in the recognition assays, and both the blocking mAbs BU68 and 3A9 as well as nonblocking mAbs bound effectively to each cell type (data not shown). We therefore tested the polarity of the ICAM-3 involvement in macrophage/apoptotic B cell interactions. As shown in Fig. 4Go, treatment of apoptotic B cells with either BU68 or 3A9 before coincubation with macrophages inhibited macrophage/apoptotic B cell interactions, whereas no inhibition was observed when macrophages were similarly pretreated. Pretreatment of either macrophages or B cells with ICAM-3 mAb CAL 3.38 failed to affect interactions between the two cell types (Fig. 4Go).



View larger version (50K):
[in this window]
[in a new window]
  		FIGURE 4. Polarized role of ICAM-3 in macrophage/apoptotic cell interactions. Interaction of apoptotic BL cells (AC) with 7-day monocyte-derived macrophages (M{phi}). Assays were conducted following pretreatment of either BL cells (gray bars) or macrophages (hatched bars) with ICAM-3 mAbs BU68, 3A9, or CAL 3.38. Means ± SD, experiment representative of three identical. Student’s t test: **, p < 0.01.

 
These results show that apoptotic B cell-associated ICAM-3 is functional in the recognition of these cells by macrophages, whereas macrophage ICAM-3 (like viable B cell ICAM-3) is not. This suggests that apoptotic B cells display ICAM-3 in a form that is qualitatively different from that of viable B cells. Taken together with the results of Fig. 2Go, our observations indicate that this qualitative change may occur downstream of the activation of biochemical events that are common to multiple proapoptosis signals. Since the epitope(s) recognized by BU68 and 3A9 is present on ICAM-3 of viable (B cells and macrophages), as well as apoptotic, cells, it is likely that this region of the molecule is not recognized directly by the putative macrophage receptor, or at least does not form the major binding site on ICAM-3 for this receptor.

Macrophages recognize ICAM-3 on apoptotic leukocytes and on apoptotic nonleukocytes expressing exogenous ICAM-3

To determine whether macrophage recognition of apoptotic leukocytes other than B cells can be mediated by ICAM-3, we analyzed ICAM-3 dependence of the interaction of apoptotic T lymphocytes (staurosporine-treated Jurkat cells) and of apoptotic neutrophils (aged in vitro) with macrophages. As shown in Fig. 5Go, interaction of apoptotic T cells with macrophages was markedly inhibited by BU68, but not CAL 3.38. Similar results were obtained with apoptotic neutrophils whose capacity to interact with macrophages was inhibited substantially and to similar degrees by BU68 and by the vitronectin receptor mAb 13C2 (Fig. 5Go). Confirmation that the involvement of ICAM-3 in apoptotic cell clearance is restricted to constitutively ICAM-3-expressing cells (naturally leukocytes) was provided by the nonleukocyte, ICAM-3-negative kidney-derived line HEK 293, whose interaction with macrophages was inhibited by the CD14 mAb 61D3, but was not affected by BU68 (Figs. 5Go and 6Go). However, apoptotic HEK 293 cells expressing high levels of exogenous ICAM-3 (Fig. 6GoA) acquired the ability to interact with macrophages via the ICAM-3-dependent pathway. As shown in Fig. 6GoB, ICAM-3-expressing HEK cells were more readily recognized by macrophages than their ICAM-3-negative counterparts, such recognition being substantially inhibited by BU68, but not CAL 3.38. The apoptotic morphology of staurosporine-induced HEK/ICAM-3 cells that interact with macrophages is illustrated in Fig. 6GoC. Staurosporine-treated HEK cells showed identical features; viable HEK or HEK/ICAM-3 cells failed to interact with macrophages (data not shown). Intriguingly, the enhanced interaction of apoptotic ICAM-3-expressing HEK cells was also markedly inhibited by 61D3, but not by 63D3.



View larger version (54K):
[in this window]
[in a new window]
  		FIGURE 5. ICAM-3-dependent interactions between macrophages and apoptotic leukocytes, but not nonleukocytes. Apoptotic T cells (Jurkat, staurosporine induced), neutrophils (peripheral blood derived, aged in culture), and embryonic kidney-derived HEK 293 cells (staurosporine induced) were exposed to macrophages in the presence or absence of the anti-ICAM-3 mAbs BU68 and CAL 3.38, the anti-{alpha}v mAb 13C2, and the anti-CD14 mAbs 61D3 and 63D3. Assays containing mAbs are presented in relation to the results of assays without mAbs, the levels of macrophages interacting with apoptotic cells in the absence of mAbs (i.e., control levels) being normalized to 100%. Data shown are means ± SEM of three separate experiments for each class of apoptotic cell shown. Student’s t test: ***, p < 0.001; *, p < 0.05.

 


View larger version (25K):
[in this window]
[in a new window]
  		FIGURE 6. ICAM-3-dependent recognition of apoptotic nonleukocytes following expression of exogenous ICAM-3. A, Immunofluorescence histograms of HEK 293T cells stained with CAL 3.38 24 h after mock transfection (HEK) or after transfection with pCDM8 containing ICAM-3 cDNA (HEK/ICAM-3). Sixty percent ICAM-3+ cells were obtained in the transfected population. Filled histograms: background staining of FITC conjugate. B, Interaction of apoptotic HEK and HEK/ICAM-3 cells (identical to those analyzed in A) with 7-day monocyte-derived macrophages. Apoptosis was induced by treatment with staurosporine (1 µM) for 18 h (HEK 70%, HEK/ICAM-3 72% in the experiment shown), and apoptotic cells (AC) were either coincubated with macrophages alone or in the presence of the indicated mAbs. Means ± SD, experiment representative of three identical. Student’s t test: ***, p < 0.001; *, p < 0.05. C, Macrophage interacting with apoptotic (staurosporine-treated) HEK cells. Arrows indicate apoptotic cells and bodies. Giemsa stain.

 
These results indicate that ICAM-3 contributes to a general mechanism of macrophage recognition of apoptotic leukocytes that normally express this glycoprotein constitutively. The observation that ICAM-3, when expressed artificially, can function in a similar manner on apoptotic nonleukocytes implies that, if additional apoptotic cell-associated molecules are required to act in concert with ICAM-3 to achieve macrophage recognition, these are not leukocyte lineage restricted.

ICAM-3-mediated interaction of apoptotic leukocytes and macrophages is independent of macrophage LFA-1, {alpha}dß2, and {alpha}vß3, but involves CD14

We next sought to determine whether either of the known ICAM-3 receptors, LFA-1 or {alpha}dß2, could function as a receptor on macro-phages for apoptotic cell-associated ICAM-3. First, a panel of six mAbs (three CD11a and three CD18), which are effective in inhibiting binding of LFA-1 to ICAM-1, ICAM-2, and ICAM-3 (Table IGo), was tested for their capacity to inhibit the interaction of apoptotic B cells with macrophages. None were effective (Fig. 7GoA). Furthermore, K562 cells expressing exogenous LFA-1 failed to support binding of apoptotic cells above that observed with parental, LFA-1-negative K562 cells (data not shown). We also tested the effects of three {alpha}d-specific mAbs. All failed to affect apoptotic leukocyte recognition by macrophages (Fig. 7GoA). Similar results were obtained when ICAM-3-expressing HEK 293 cells were used as the apoptotic cell source, with neither the CD18-blocking mAb 7E4 nor the {alpha}d mAb 217L inhibiting interaction of these cells with macrophages (Fig. 7GoB). Lack of macrophage ß2 integrin involvement in interaction with apoptotic leukocyte-associated ICAM-3 was further supported by experiments in which activated macrophage LFA-1 was investigated. Activation of macrophage LFA-1 with KIM 127, a mAb that promotes binding of LFA-1 to ICAM-3 and ICAM-1 (46), failed to affect apoptotic cell interactions with macrophages (Fig. 8GoA). Macrophage binding of viable B cells, however, was markedly enhanced following KIM 127 treatment (Fig. 8Go, A and B). Significantly, viable cell binding involved LFA-1/ICAM-3 interactions, as confirmed by inhibition with mAb 7E4, and with the ICAM-3 mAbs CAL 3.38 and BU68 (Fig. 7GoB). The near complete inhibition of macrophage/viable cell interactions by the CD18 mAb 7E4 compared with the partial inhibition by anti-ICAM-3 mAbs suggests that viable B cell binding to macrophages stimulated by KIM 127 is entirely ß2-integrin (presumably LFA-1) dependent and is likely to involve ICAM-1 as well as ICAM-3, since the BL cells are known to express low levels of ICAM-1 (40). Viable cell binding to macrophages, in contrast to apoptotic cell binding, did not lead to phagocytosis ((39) and data not shown). Taken together, these results demonstrate not only that apoptotic cell-associated ICAM-3 interacts with macrophages independently of LFA-1, and probably ß2 integrins in general, including {alpha}dß2, but also that, during apoptosis, ICAM-3 loses its ability to interact with LFA-1.



View larger version (38K):
[in this window]
[in a new window]
  		FIGURE 7. Macrophage/apoptotic cell interactions are independent of LFA-1 and {alpha}dß2. A, Failure of CD11a mAbs (gray bars), CD18 mAbs (narrow-hatched bars), and {alpha}d mAbs (wide-hatched bars) to inhibit interaction of macrophages with apoptotic BL cells. Separate experiments shown for CD11a/CD18 and {alpha}d mAbs, each representative of three identical. B, Failure of CD18 and {alpha}d mAbs to inhibit interaction of macrophages with apoptotic HEK or HEK/ICAM-3 cells. Representative experiment of three identical. Means ± SD. Student’s t test: ****, p < 0.0001; ***, p < 0.001.

 
Of other possible receptors on macrophages for apoptotic cell-associated ICAM-3, we next considered two further candidates, {alpha}vß3 and CD14, each having proven ability in phagocytic clearance of apoptotic cells (8, 12, 39). Using HEK/ICAM-3 transfectants as the source of apoptotic cells to highlight the apoptotic-ICAM-3 pathway, we found that 13C2, an {alpha}v-specific mAb known to inhibit uptake of apoptotic leukocytes by macrophages (see Ref. 8 and Fig. 5Go), failed to affect the interaction of apoptotic HEK cells with macrophages, regardless of whether ICAM-3 was present (Fig. 9Go). By contrast, 61D3, a CD14 mAb known to inhibit macrophage/apoptotic leukocyte interactions (12, 39), markedly reduced macrophage/apoptotic HEK cell interactions. Thus, as shown in Fig. 9Go, 61D3, but not its nonblocking counterpart 63D3, markedly inhibited not only the low levels of macrophage interactions observed with apoptotic HEK cells, but also the enhanced levels observed with apoptotic HEK/ICAM-3 cells. Indeed, 61D3-mediated blockade of macrophage recognition of apoptotic HEK/ICAM-3 cells was equivalent to that observed with BU68. Furthermore, when BU68 and 61D3 were used in combination, no additional inhibitory effects over and above the individual mAbs were obtained (Fig. 9Go). These results suggest strongly that, while the {alpha}vß3 macrophage vitronectin receptor is not involved in ICAM-3-dependent clearance of apoptotic cells, macrophage CD14 is.



View larger version (43K):
[in this window]
[in a new window]
  		FIGURE 9. ICAM-3-mediated macrophage/apoptotic cell interactions are independent of {alpha}vß3, but involve CD14. Failure of anti-{alpha}v mAb (13C2) and potency of CD14 mAb 61D3 in inhibiting interaction of macrophages with apoptotic HEK or HEK/ICAM-3 cells. Effects of combining 61D3 and the inhibitory ICAM-3 mAb, BU68, are also shown. Means ± SD. Student’s t test: ****, p < 0.0001; ***, p < 0.001. HEK/ICAM-3 transfectants were 48% ICAM-3 positive by immunofluorescence in the experiment shown (representative of three identical); staurosporine treatment resulted in 73% apoptotic HEK cells and 63% apoptotic HEK/ICAM-3 cells.

 
Quantitative, but not qualitative, changes in ICAM-3 from apoptotic cells are detected by Western blotting

In an effort to obtain additional information of possible relevance to the mechanism underlying the observed altered receptor-binding function of ICAM-3 on apoptotic cells, ICAM-3 from viable and apoptotic cells was compared by immunoblotting. As shown in Fig. 10Go, lysates of viable and of apoptotic ICAM-3-expressing HEK cells contained two molecular species that were detected by the ICAM-3 mAbs: a major band at 96 kDa, and a minor band at 124 kDa. No qualitative differences between viable or apoptotic cell-derived ICAM-3 were detectable. However, a clear quantitative difference was noted, with apoptotic cell lysates containing lower levels of ICAM-3 than their viable counterparts. Lower levels of ICAM-3 on the surface of apoptotic cells as compared with viable cells were also found in immunofluorescence/flow cytometry assays (our unpublished observations).



View larger version (64K):
[in this window]
[in a new window]
  		FIGURE 10. Western blot comparing ICAM-3 from viable and apoptotic cells. Lysates of mock- or ICAM-3-transfected HEK cells were prepared either before (-) or after (+) induction of apoptosis with staurosporine. ICAM-3 was detected as two bands (arrows, right). Molecular weight markers are shown at left (arrowheads).

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Little is yet known of molecules on the surface of apoptotic cells that support the essential recognition of such cells by phagocytes. While evidence has been provided that changes in surface charge and carbohydrate composition at the plasma membrane of apoptotic cells are important components for phagocytic recognition (15, 16, 17, 47), the molecules involved have not been identified. Indeed, to date, the only apoptotic cell-associated molecule to be shown definitively to play a role in the recognition process is the anionic phospholipid, phosphatidylserine, which becomes externalized during apoptosis before loss of plasma membrane integrity (14, 48). The results presented in this work show that apoptotic cell-associated ICAM-3 is involved in the recognition of apoptotic leukocytes by macrophages. ICAM-3 is therefore only the second molecule to be identified as an apoptotic cell surface entity that is involved in the clearance process. Furthermore, support of phagocytic clearance of apoptotic cells is a novel ICAM function. Our overall conclusions are that: 1) ICAM-3 supports a common mechanism of phagocytic elimination of ICAM-3-bearing cells, that is leukocytes, undergoing apoptosis; (2) the ICAM-3-mediated marking of apoptotic leukocytes for recognition by phagocytes may form part of a final common pathway for apoptosis; 3) the key region of ICAM-3 for recognition by phagocytes is located in the most distal Ig-like domain, D1; 4) although this region lies close to, or overlaps with, the region of ICAM-3 that interacts with macrophage LFA-1 to bind viable leukocytes, apoptotic cell recognition is LFA-1 independent, and is probably also independent of the related leukointegrin and ICAM-3 receptor, {alpha}dß2; and 5) macrophage CD14 plays a role in ICAM-3-dependent apoptotic cell clearance.

A qualitative change in ICAM-3 during apoptosis?

Our results are consistent with the idea that ICAM-3 exists in viable and apoptotic forms, with phagocytic interactions between apoptotic leukocytes and macrophages being restricted to the apoptotic form of the molecule. Thus, viable ICAM-3 was not routinely involved in interactions between macrophages and B cells since, in the standard macrophage interaction assays used (which measure a combination of macrophage adhesion to B cells and macrophage phagocytosis of B cells (12, 39)), interactions with apoptotic cells predominated (see Figs. 1Go and 8Go) (12, 39). On occasions when significant levels of viable cell interactions were observed, either by overloading the macrophage assays with viable cells or by including the LFA-1-activating mAb KIM 127 (Fig. 8Go), such interactions were limited to adhesion and did not lead to phagocytosis (see Fig. 8GoB and data not shown). As discussed below, adhesion of viable cells via ICAM-3 was LFA-1 dependent, whereas adhesion/phagocytosis of apoptotic cells was not. In addition, while the macrophage populations used in all interaction assays expressed ICAM-3 (in viable form), macrophage ICAM-3 did not participate in apoptotic cell recognition (Fig. 4Go). We therefore propose that ICAM-3 undergoes a qualitative change as a result of activation of the apoptosis program, which alters its receptor-binding properties and allows it to interact, directly or indirectly, with a macrophage-borne molecule or molecular complex. The nature of the macrophage receptor for apoptotic ICAM-3 is considered below, and the likelihood that it is not LFA-1, the prototypic counterreceptor, or {alpha}dß2, the alternative counterreceptor, for viable ICAM-3, lends further support to the notion that different forms of ICAM-3 exist on viable and apoptotic leukocytes.

The basis for the qualitative difference between apoptotic and viable forms of ICAM-3 remains to be determined. Differential glycosylation is an attractive possibility, particularly since ICAM-3 is known to be highly and variably glycosylated, being the most heavily glycosylated member of the ICAMs. Differentially glycosylated forms of neural Ig-superfamily members have altered binding properties (49, 50). Furthermore, evidence has been presented that ICAM-1 undergoes a change from LFA-1 binding to Mac-1 binding that may result from differential glycosylation (51). Notably, N-linked glycosylation sites exist in ICAM-3 at Asn62 and Asn72 (23, 24), in the vicinity of the BU68 and 3A9 binding sites. Although we do not address possible mechanisms underlying the putative qualitative change in ICAM-3 on apoptotic cells in detail in this study, we found by SDS-PAGE and immunoblotting identical species of ICAM-3 in both viable cell and apoptotic cell lysates. The observed bands of 96 and 124 kDa fell within the known m.w. range from ~60 kDa for the nonglycosylated backbone up to ~160 kDa for the fully glycosylated protein, depending on the cell type (18, 22). On the basis of gross m.w., therefore, we were unable to obtain any indication that a different glycoform(s) of ICAM-3 was associated with apoptotic cells (Fig. 10Go). Formally, the proposed qualitative change in ICAM-3 during apoptosis remains speculative: it is conceivable that the apoptotic form of ICAM-3 is, in itself, indistinguishable from the viable form, but on apoptotic cells is a participant in a molecular recognition complex, other components of which are either absent from viable cells or are present but incapable of associating with ICAM-3 (see below and Fig. 11Go). In this respect, it is notable that lower levels of ICAM-3 are present on apoptotic cells than on their viable counterparts (Fig. 10Go and our unpublished observations). In future studies, it will be important to define the molecular basis for the observed qualitative and quantitative changes in ICAM-3 that occur during apoptosis, particularly in relation to the apoptosis machinery.



View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 11. Schematic representation of some models for interaction of macrophage receptor with apoptotic cell-associated ICAM-3. Epitopes for BU68 and 3A9 are present on D1 of scheme 1, viable, and schemes 2–4, apoptotic, cell-associated ICAM-3. ICAM-3 may undergo a qualitative change during apoptosis (scheme 2), which allows direct interaction with macrophage receptors (M{phi}R), but BU68 and 3A9 inhibit this interaction because their epitopes impinge upon, or lie close to, the M{phi}R binding site on D1. In scheme 3, apoptotic cell-associated ICAM-3 interacts with M{phi}R through an intermediate bridging molecule (B) that binds ICAM-3 at or near the BU68/3A9 epitopes, while in scheme 4, the M{phi}R is required to interact with a complex ligand composed in part by ICAM-3 and in part by an additional protein (C) that again binds ICAM-3 at or near the BU68/3A9 epitopes. B and C either associate only with qualitatively changed, apoptotic ICAM-3 (as depicted), or are only available (or able) to interact with ICAM-3 at apoptotic cell surfaces. B and C may be integral or peripheral components of the plasma membrane of apoptotic leukocytes. CD14 may function as a M{phi}R in these models.

 
The macrophage receptor for apoptotic cell-associated ICAM-3

Using a series of mutant rICAM-3 proteins, the inhibitory mAbs BU68 and 3A9 were shown to recognize similar epitopes in D1 of ICAM-3 (Fig. 3Go). Circumstantial evidence from recent molecular models of ICAM-3 structure (23, 24, 25) would suggest that these epitopes are likely to overlap with the LFA-1 binding site(s) of D1. Further evidence that the epitopes recognized by BU68 and 3A9 overlap with the LFA-1 binding site of ICAM-3 D1 is provided by additional observations (Table IGo) that both Abs are partial blockers of binding of cell-associated LFA-1 to rICAM-3. In addition, BU68 was found to be as potent as CAL 3.38 (an LFA-1/rICAM-3 blocker; Table IGo) in inhibition of macrophage/viable cell-associated ICAM-3 interactions (Fig. 8GoA). Nevertheless, additional evidence supports our conclusion that LFA-1 is an unlikely macrophage receptor for apoptotic cell-associated ICAM-3. First, CD11a and CD18 mAbs that inhibit the interaction of LFA-1 with ICAM-1, ICAM-2, and ICAM-3 failed to prevent ICAM-3-dependent interaction of apoptotic cells with macrophages (Fig. 8GoA). Similarly, several ICAM-3 D1-specific mAbs that are proven and potent blockers of LFA-1/ICAM-3 binding (Table IGo) failed to inhibit macrophage/apoptotic leukocyte interactions (Figs. 1Go, 5Go, and 6Go). LFA-1 expressed on the surface of transfectant K562 cells, like the endogenous LFA-1 of macrophages, also failed to support binding of apoptotic leukocytes. On macrophages activated with KIM 127, a mAb that promotes ICAM-1 and ICAM-3 binding to LFA-1 (46), binding of viable cells was LFA-1 dependent, whereas apoptotic cell binding was LFA-1 independent (Fig. 8Go).

Our attempts to inhibit macrophage/apoptotic cell interactions with mAbs specific for the other known counterreceptor for viable ICAM-3, the leukointegrin {alpha}dß2 (Fig. 7Go), suggest that, like its relative LFA-1, it too fails to bind apoptotic ICAM-3. Only a limited number of {alpha}d-specific mAbs are yet available, and it remains a formal possibility that epitopes other than those bound by the available Abs are important in interacting with apoptotic cell-associated ICAM-3. However, our results with CD18 mAbs (Fig. 7Go) and earlier work indicating that the leuko (ß2)-integrin family as a whole does not play a major role in recognition and clearance of apoptotic leukocytes by human monocyte-derived macrophages (8, 47), support the view that {alpha}dß2 is unlikely to be involved in apoptotic ICAM-3/macrophage interactions.

In an effort to resolve the question of the identity of the macrophage moiety that interacts with apoptotic cell-associated ICAM-3, we investigated two macrophage receptors, the vitronectin receptor {alpha}vß3 and CD14, which have been firmly linked with apoptotic cell clearance through the inhibitory action of certain vitronectin receptor- and CD14-specific mAbs (8, 12). The {alpha}v-specific mAb 13C2, which is known to inhibit clearance of apoptotic leukocytes by human monocyte-derived macrophages (8) (Fig. 5Go), failed to affect the ICAM-3-dependent interaction of apoptotic HEK/ICAM-3 cells with macrophages. By contrast, the CD14 mAb 61D3, a known inhibitor of macrophage clearance of apoptotic leukocytes, substantially inhibited macrophage interactions with HEK/ICAM-3 cells (Fig. 9Go). Involvement of CD14 was epitope specific since 63D3, a CD14 mAb that has been shown to bind CD14 at a location distinct from that of 61D3 (12), failed to affect ICAM-3-dependent apoptotic cell clearance. Furthermore, since the effects of simultaneously blocking apoptotic ICAM-3 (with BU68, on the side of the HEK/ICAM-3 cell) and CD14 (with 61D3, on the side of the macrophage) were not additive (indeed, mixing these Abs was no more effective than using them individually), we conclude that ICAM-3 and CD14 belong to a common pathway of interaction between apoptotic cell and macrophage. Our results do not preclude, however, the involvement of macrophage receptors in addition to CD14 in ICAM-3-dependent apoptotic cell clearance. Indeed, the inability of 61D3 to reduce the enhanced (i.e., ICAM-3-dependent) recognition of apoptotic HEK/ICAM-3 cells to the level of the apoptotic parental HEK cells (Fig. 9Go) suggests a role for macrophage receptors in addition to CD14. The capacity of CD14 and apoptotic cell-associated ICAM-3 to interact directly will require further investigation. Since CD14 is known to display lectin-like activity (52, 53), it is tempting to speculate that CD14 and apoptotic ICAM-3 could bind directly via lectin-sugar interactions. Whatever the basis for the relationship between CD14 and ICAM-3, the role of CD14 in apoptotic cell clearance is not limited to one involving ICAM-3 since CD14 can also function in the clearance of apoptotic cells that are ICAM-3 negative (Figs. 5Go and 9Go). By implication, therefore, alternative CD14 ligands may also be available on apoptotic cells. This notion would accord with the known capacity of CD14 to interact with a broad array of lipid, carbohydrate, and protein ligands (53, 54, 55, 56, 57, 58, 59, 60).

Apoptotic HEK cells overexpressing ICAM-3 permitted demonstration of the ICAM-3-dependent pathway of apoptotic cell clearance in the absence of any Ab effects. However, such overactivity of this pathway in no way addresses the question of the relative physiological activity of the pathway compared with other pathways involving either alternative surface components of apoptotic cells (e.g., exposed phosphatidylserine (14)) or well-characterized macrophage receptors other than CD14 that have been imlicated in apoptotic cell removal, including {alpha}vß3 (8), CD36 (9), the ATP-binding cassette transporter ABC-1 (10), and class A scavenger receptor (11). Although it is not yet clear why multiple pathways are required for apoptotic cell clearance, the apparent multiplicity in the receptor-ligand interactions that contribute to apoptotic cell removal most likely illustrates that 1) apoptotic cell removal in vivo is of the utmost importance to homeostasis and consequently is served by redundant molecular mechanisms, and 2) front-line and backup clearance mechanisms are involved in the removal of apoptotic cells at sequential stages of apoptosis (7). The relative importance of individual clearance pathways awaits detailed clarification of the various phagocyte-receptor/apoptotic cell-ligand interactions that support the clearance process. The indication, as discussed above, that CD14 and ICAM-3 may comprise an apoptotic cell clearance pathway that overlaps with others involving alternative phagocyte receptors and additional apoptotic cell ligands suggests that, given appropriate expression and context, these molecules would be predicted to play significant roles in apoptotic cell clearance.

Nature of the apoptotic ICAM-3/macrophage receptor interaction

The observation that the epitopes of ICAM-3 D1 recognized by the inhibitory mAbs BU68 and 3A9 are present on viable, as well as apoptotic, cell-associated ICAM-3 implies that these epitopes do not interact directly with the macrophage receptor in question. With reference to Fig. 11Go, we suggest either 1) qualitatively changed apoptotic ICAM-3 interacts with the putative macrophage receptor via a region in D1 that lies sufficiently close to the binding sites of BU68 and 3A9 so as to allow stearic hindrance of macrophage-receptor binding by these Abs, or 2) qualitatively changed or unchanged ICAM-3 participates in a molecular complex in which an additional component contributes significantly to macrophage-receptor binding, possibly by forming a molecular bridge between ICAM-3 and the macrophage receptor. In the latter situation, BU68 and 3A9 epitopes would be envisaged to play a role in binding of the putative bridging molecule to ICAM-3. Work by Savill and colleagues on the clearance of apoptotic granulocytes by macrophages provides a precedent for the molecular bridge model in that thrombospondin, an adhesive glycoprotein secreted by many cell types, is proposed to form a bridge between the macrophage vitronectin receptor/CD36 complex and an undefined component of the apoptotic granulocyte (9). It is conceivable that other peripheral or integral membrane proteins or lipid membrane components interact with ICAM-3 on apoptotic cells, providing the apoptotic leukocyte-specific ligand for a macrophage receptor that does not directly recognize ICAM-3 (Fig. 11Go). How CD14 fits into this picture remains to be determined. One obvious possibility is that CD14 is the putative macrophage receptor illustrated in Fig. 11Go, or forms part of a receptor complex. Alternatively, since CD14 can be cleaved from the cell surface following ligand binding to act on other, ill-defined cell surface receptors (61), it is conceivable that apoptotic ICAM-3 interacts with the putative macrophage receptor as a complex with CD14. Extensive future studies will be required to answer these critical questions.

In conclusion, the results presented in this work 1) implicate ICAM-3 in the phagocytic clearance of apoptotic leukocytes, and 2) demonstrate that a change in counterreceptor-binding specificity of ICAM-3 is a common feature of leukocyte apoptosis. The alteration in counterreceptor preference may provide an important mechanism for functional isolation of ICAM-3-bearing cells. For example, the inability of APC, upon engagement of their apoptotic program, to mediate ICAM-3/LFA-1 interactions may be an important aspect of normal immunological physiology. Furthermore, defective clearance of apoptotic cells is likely to contribute to inflammatory and autoimmune disorders (see 7). Since high levels of circulating ICAM-3 are a feature of such disorders (36, 37), it is tempting to speculate that ICAM-3 could be linked causally to disease pathology, perhaps by competition of circulating ICAM-3 for the macrophage receptors involved in apoptotic cell clearance. Our results will help to further understanding of the contribution of ICAM-3 to normal, as well as pathological, immune mechanisms.


   Acknowledgments
 
We thank Mike Gallatin, Pat Hoffman, and Joel Hayflick (ICOS), Ian Collins (R&D Systems), Debbie Hardie (University of Birmingham), Mike Horton (University College London), and Innogenetics for gifts of Abs; Martyn Robinson (Celltech) for KIM 127 and K562 transfectants; Barbara Spruce (University of Dundee) for HEK cells; and Chandra Raykundalia for technical assistance.


   Footnotes
 
1 This work was supported by grants from the Medical Research Council (U.K.) and the University of Nottingham Research Opportunity Fund. Back

2 Address correspondence and reprint requests to Dr. Christopher D. Gregory, Institute of Cell Signaling and School of Biomedical Sciences, University of Nottingham Medical School, D Floor, Queen’s Medical Centre, Nottingham, NG7 2UH, U.K. E-mail address: Back

3 Abbreviations used in this paper: D, domain; BL, Burkitt lymphoma; HEK, human embryonic kidney; wt, wild-type. Back

Received for publication October 13, 1998. Accepted for publication March 19, 1999.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Wyllie, A. H., J. F. R. Kerr, A. R. Currie. 1980. Cell death: the significance of apoptosis. Int. Rev. Cytol. 68:251.[Medline]
  2. Thompson, C. B.. 1995. Apoptosis in the pathogenesis and treatment of disease. Science 267:1456.[Abstract/Free Full Text]
  3. Nicholson, D. W., N. A. Thornberry. 1997. Caspases: killer proteases. Trends Biochem. Sci. 22:299.[Medline]
  4. Enari, M., H. Sakahira, H. Yokoyama, K. Okawa, A. Iwamatsu, S. Nagata. 1998. A caspase-activated DNase that degrades DNA during apoptosis, and its inhibitor ICAD. Nature 391:43.[Medline]
  5. Savill, J., V. Fadok, P. Henson, C. Haslett. 1993. Phagocyte recognition of cells undergoing apoptosis. Immunol. Today 14:131.[Medline]
  6. Fadok, V. A., D. L. Bratton, S. C. Frasch, M. L. Warner, P. M. Henson. 1998. The role of phosphatidylserine in recognition of apoptotic cells by phagocytes. Cell Death Differ. 5:551.[Medline]
  7. Ren, Y., J. Savill. 1998. Apoptosis: the importance of being eaten. Cell Death Differ. 5:563.[Medline]
  8. Savill, J., I. Dransfield, N. Hogg, C. Haslett. 1990. Vitronectin receptor-mediated phagocytosis of cells undergoing apoptosis. Nature 343:170.[Medline]
  9. Savill, J., N. Hogg, Y. Ren, C. Haslett. 1992. Thrombospondin cooperates with CD36 and the vitronectin receptor in macrophage recognition of neutrophils undergoing apoptosis. J. Clin. Invest. 90:1513.
  10. Luciani, M.-F., G. Chimini. 1996. The ATP binding cassette transporter ABC1, is required for the engulfment of corpses generated by apoptotic cell death. EMBO J. 15:226.[Medline]
  11. Platt, N., H. Suzuki, Y. Kurihara, T. Kodama, S. Gordon. 1996. Role for the class A macrophage scavenger receptor in the phagocytosis of apoptotic thymocytes in vitro. Proc. Natl. Acad. Sci. USA 93:12456.[Abstract/Free Full Text]
  12. Devitt, A., O. D. Moffatt, C. Raykundalia, J. D. Capra, D. L. Simmons, C. D. Gregory. 1998. Human CD14 mediates recognition and phagocytosis of apoptotic cells. Nature 392:505.[Medline]
  13. Sambrano, G. R., D. Steinberg. 1995. Recognition of oxidatively damaged and apoptotic cells by an oxidized low-density-lipoprotein receptor on mouse peritoneal macrophages role of membrane phosphatidylserine. Proc. Natl. Acad. Sci. USA 92:1396.[Abstract/Free Full Text]
  14. Fadok, V. A., D. R. Voelker, P. A. Campbell, J. J. Cohen, D. L. Bratton, P. M. Henson. 1992. Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. J. Immunol. 148:2207.[Abstract]
  15. Duvall, E., A. H. Wyllie, R. G. Morris. 1985. Macrophage recognition of cells undergoing programmed cell death (apoptosis). Immunology 56:351.[Medline]
  16. Dini, L., F. Autuori, A. Lentini, S. Oliverio, M. Piacentini. 1992. The clearance of apoptotic cells in the liver is mediated by the asialoglycoprotein receptor. FEBS Lett. 296:174.[Medline]
  17. Hall, S. E., J. S. Savill, P. M. Henson, C. Haslett. 1994. Apoptotic neutrophils are phagocytosed by fibroblasts with participation of the fibroblast vitronectin receptor and involvement of a mannose/fucose-specific lectin. J. Immunol. 153:3218.[Abstract]
  18. Fawcett, J., C. L. L. Holness, L. A. Needham, H. Turley, K. C. Gatter, D. Y. Mason, D. L. Simmons. 1992. Molecular cloning of ICAM-3, a third ligand for LFA-1, constitutively expressed on resting leukocytes. Nature 360:481.[Medline]
  19. Vazeux, R., P. A. Hoffman, J. K. Tomita, E. S. Dickinson, R. L. Jasman, T. Stjohn, W. M. Gallatin. 1992. Cloning and characterization of a new intercellular adhesion molecule ICAM-R. Nature 360:485.[Medline]
  20. De Fougerolles, A. R., T. A. Springer. 1992. Intercellular adhesion molecule 3, a third adhesion counter-receptor for lymphocyte function-associated molecule 1 on resting lymphocytes. J. Exp. Med. 175:185.[Abstract/Free Full Text]
  21. De Fougerolles, A. R., L. B. Klickstein, T. A. Springer. 1993. Cloning and expression of intercellular adhesion molecule 3 reveals strong homology to other immunoglobulin family counter-receptors for lymphocyte function-associated antigen 1. J. Exp. Med. 177:1187.[Abstract/Free Full Text]
  22. De Fougerolles, A. R., M. S. Diamond, T. A. Springer. 1995. Heterogenous glycosylation of ICAM-3 and lack of interaction with mac-1 and p150,95. Eur. J. Immunol. 25:1008.[Medline]
  23. Holness, C. L., P. A. Bates, A. J. Littler, C. D. Buckley, A. McDowall, D. Bossy, N. Hogg, D. L. Simmons. 1995. Analysis of the binding site on intercellular adhesion molecule-3 for the leukocyte integrin lymphocyte function-associated antigen-1. J. Biol. Chem. 270:877.[Abstract/Free Full Text]
  24. Klickstein, L. B., M. R. York, A. R. Defougerolles, T. A. Springer. 1996. Localization of the binding site on intercellular adhesion molecule-3 (ICAM-3) for lymphocyte function-associated antigen-1 (LFA-1). J. Biol. Chem. 271:23920.[Abstract/Free Full Text]
  25. Bell, E. D., A. P. May, D. L. Simmons. 1998. The leukocyte function-associated antigen-1 (LFA-1)-binding site on ICAM-3 comprises residues on both faces of the first immunoglobulin domain. J. Immunol. 161:1363.[Abstract/Free Full Text]
  26. Van der Vieren, M., H. Le Trong, C. L. Wood, P. F. Moore, T. St. John, D. E. Staunton, W. M. Gallatin. 1995. A novel leukointegrin, {alpha}dß2, binds preferentially to ICAM-3. Immunity 3:683.[Medline]
  27. Campanero, M. R., M. A. Delpozo, A. G. Arroyo, P. Sanchez-Mateos, T. Hernandez-Caselles, A. Craig, R. Pulido, F. Sanchez-Madrid. 1993. ICAM-3 interacts with LFA-1 and regulates the LFA-1/ICAM-1 cell adhesion pathway. J. Cell Biol. 123:1007.[Abstract/Free Full Text]
  28. Hernandez-Caselles, T., G. Rubio, M. R. Campanero, M. A. Delpozo, M. Muro, F. Sanchez-Madrid, P. Aparicio. 1993. ICAM-3, the third LFA-1 counter-receptor, is a costimulatory molecule for both resting and activated T lymphocytes. Eur. J. Immunol. 23:2799.[Medline]
  29. Griffiths, C. E. M., D. Railan, W. M. Gallatin, K. D. Cooper. 1995. The ICAM-3/LFA-1 interaction is critical for epidermal Langerhans cell alloantigen presentation to CD4+ T cells. Br. J. Dermatol. 133:823.[Medline]
  30. Starling, G. C., A. D. McLellan, W. Egner, R. V. Sorg, J. Fawcett, D. L. Simmons, D. N. J. Hart. 1995. Intercellular adhesion molecule 3 is the predominant costimulatory ligand for leukocyte function antigen 1 on human blood dendritic cells. Eur. J. Immunol. 25:2528.[Medline]
  31. Green, J. M., C. B. Thompson. 1996. Homotypic interactions mediated through LFA-1/ICAM-3 decrease the proliferative response of activated T cells. Cell. Immunol. 171:126.[Medline]
  32. Martinez-Caceres, E., G. Ruggiero, H. Spits, M. Juan, J. Barcelo, J. Vives, J. Martorell, O. Vinas. 1996. Stimulation through CD50 (ICAM-3) induces both activation and programmed cell death of human thymocytes. Tissue Antigens 48:626.[Medline]
  33. Arroyo, A. G., M. R. Campanero, P. Sanchez-Mateos, J. M. Zapata, M. A. Ursa, M. A. Delpozo, F. Sanchez-Madrid. 1994. Induction of tyrosine phosphorylation during ICAM-3 and LFA-1-mediated intercellular adhesion, and its regulation by the CD45 tyrosine phosphatase. J. Cell Biol. 126:1277.[Abstract/Free Full Text]
  34. Juan, M., O. Vinas, M. R. Pinootin, L. Places, E. Martinez-Caceres, J. J. Barcelo, A. Miralles, R. Vilella, M. A. Delafuente, J. Vives, J. Yague, A. Gaya. 1994. CD50 (intercellular adhesion molecule 3) stimulation induces calcium mobilization and tyrosine phosphorylation through p59fyn and p56lck in Jurkat T-cell line. J. Exp. Med. 179:1747.[Abstract/Free Full Text]
  35. Skubitz, K. M., K. Ahmed, K. D. Campbell, A. P. N. Skubitz. 1995. CD50 (ICAM-3) is phosphorylated on tyrosine and is associated with tyrosine kinase activity in human neutrophils. J. Immunol. 154:2888.[Abstract]
  36. Martin, S., P. Rieckmann, I. Melchers, R. Wagner, J. Bertrams, A. E. Voskuyl, B. O. Roep, J. Zielasek, E. Heidenthal, I. Weichselbraun, W. M. Gallatin, H. Kolb. 1995. Circulating forms of ICAM-3 (cICAM-3): elevated levels in autoimmune diseases and lack of association with cICAM-1. J. Immunol. 154:1951.[Abstract]
  37. Littler, A. J., C. D. Buckley, P. Wordsworth, I. Collins, J. Martinson, D. L. Simmons. 1997. Distinct profile of six soluble adhesion molecules (ICAM-1, ICAM-3, VCAM-1, E-selectin, L-selectin and P-selectin) in rheumatoid arthritis. Br. J. Rheumatol. 36:164.[Abstract/Free Full Text]
  38. Bossy, D., C. D. Buckley, C. L. Holness, A. J. Littler, N. Murray, I. Collins, D. L. Simmons. 1995. Epitope mapping and functional properties of anti-intercellular adhesion molecule-3 (CD50) monoclonal antibodies. Eur. J. Immunol. 25:459.[Medline]
  39. Flora, P. K., C. D. Gregory. 1994. Recognition of apoptotic cells by human macrophages: inhibition by a monocyte/macrophage-specific monoclonal antibody. Eur. J. Immunol. 24:2625.[Medline]
  40. Gregory, C. D., M. Rowe, A. B. Rickinson. 1990. Different Epstein-Barr virus-B cell interactions in phenotypically distinct clones of a Burkitt’s lymphoma cell line. J. Gen. Virol. 71:1481.[Abstract/Free Full Text]
  41. Jepsen, L. V., T. A. Skottun. 1982. A rapid one-step method for the isolation of human granulocytes from whole blood. Scand. J. Clin. Lab. Invest. 42:235.[Medline]
  42. Gregory, C. D., A. E. Milner. 1994. Regulation of cell survival in Burkitt lymphoma: implications from studies of apoptosis following cold-shock treatment. Int. J. Cancer 57:419.[Medline]
  43. Gregory, C. D., C. Dive, S. Henderson, C. A. Smith, G. T. Williams, J. Gordon, A. B. Rickinson. 1991. Activation of Epstein-Barr virus latent genes protects human B cells from death by apoptosis. Nature 349:612.[Medline]
  44. Dive, C., C. D. Gregory, D. J. Phipps, D. L. Evans, A. E. Milner, A. H. Wyllie. 1992. Analysis and discrimination of necrosis and apoptosis (programmed cell death) by multiparameter flow cytometry. Biochim. Biophys. Acta 1133:275