QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Vol. 17, Issue 10, 4270-4281, October 2006

This Article Abstract Full Text (PDF) Supplemental Material All Versions of this Article: E05-12-1098v1 17/10/4270    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cambi, A. Articles by Figdor, C. G. Search for Related Content PubMed PubMed Citation Articles by Cambi, A. Articles by Figdor, C. G.

Organization of the Integrin LFA-1 in Nanoclusters Regulates Its Activity

Alessandra Cambi^*, Ben Joosten^*, Marjolein Koopman, Frank de Lange^*,, Inge Beeren^*, Ruurd Torensma^*, Jack A. Fransen, Maria Garcia-Parajó^,, Frank N. van Leeuwen^*, and Carl G. Figdor^*

*Department of Tumor Immunology and Department of Cell Biology and Microscopic Imaging Centre, Nijmegen Centre for Molecular Life Sciences, Radboud University Nijmegen Medical Centre, 6500 HB Nijmegen, The Netherlands; and Applied Optics Group and Department of Applied Physics and MESA+ Research Institute, University of Twente, 7522 NM Enschede, The Netherlands

Submitted December 2, 2005; Revised June 29, 2006; Accepted July 6, 2006
Monitoring Editor: Martin A. Schwartz

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The 2-integrin LFA-1 facilitates extravasation of monocytes (MOs) into the underlying tissues, where MOs can differentiate into dendritic cells (DCs). Although DCs express LFA-1, unlike MOs, they cannot bind to ICAM-1. We hypothesized that an altered integrin organization on the DC plasma membrane might cause this effect and investigated the relationship between membrane organization and function of LFA-1 on MOs and DCs. High-resolution mapping of LFA-1 surface distribution revealed that on MOs LFA-1 function is associated with a distribution in well-defined nanoclusters (100–150-nm diameter). Interestingly, a fraction of these nanoclusters contains primed LFA-1 molecules expressing the specific activation-dependent L16-epitope. Live imaging of MO–T-cell conjugates showed that only these primed nanoclusters are dynamically recruited to the cellular interface forming micrometer-sized assemblies engaged in ligand binding and linked to talin. We conclude that besides affinity regulation, LFA-1 function is controlled by at least three different avidity patterns: random distributed inactive molecules, well-defined ligand-independent proactive nanoclusters, and ligand-triggered micrometer-sized macroclusters.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Integrins are transmembrane / heterodimers that regulate cell–cell and cell–extracellular matrix interactions. Lymphocyte function-associated antigen-1 (LFA-1; L2; CD11a/CD18) is a leukocyte specific integrin that mediates migration across the endothelium and within tissues and formation of immunological synapse (Dustin and Springer, 1989; van Kooyk et al., 1989; Lub et al., 1995; Grakoui et al., 1999; Carman et al., 2003). LFA-1 binds to its major counterreceptor ICAM-1 (Marlin and Springer, 1987) and with lower affinity also to ICAM-2 (Staunton et al., 1989) and -3 (de Fougerolles et al., 1991; de Fougerolles and Springer, 1992).

Two not-mutually-exclusive mechanisms were proposed explaining how integrins such as LFA-1 become activated. First, conformational changes lead to an increased affinity for the ligands (Shimaoka et al., 2003). Evidence comes from mutagenesis studies (Lu et al., 2001a; Lu et al., 2001b) and from the identification of "activation reporter" epitopes (Dransfield and Hogg, 1989; Beals et al., 2001). Indeed, NMR and negative stain electron microscopy (EM) have revealed conformational rearrangement and movement, such as separation of the cytoplasmic tails and extension of the extracellular domains, that lead to a general mechanism of integrin activation (Takagi et al., 2002; Vinogradova et al., 2002). Early studies also showed that a Ca²⁺-dependent epitope recognized by the monoclonal antibody (mAb) NKI-L16 (Keizer et al., 1988; van Kooyk et al., 1991)—further referred to as L16 epitope—is an activation-reporter epitope resulting from a Ca²⁺-bound extended conformation of the L subunit (Xie et al., 2004).

A second mechanism that regulates integrin activation is a dynamic reorganization of LFA-1 receptors into multimolecular assemblies at the cell surface that locally increase the binding valency (avidity; van Kooyk et al., 1999; Katagiri et al., 2003). So far, experimental evidence for the formation of such LFA-1 clusters on activated cells is limited and primarily is based on images of micrometer-sized patches of molecules on the plasma membrane of polarized cells (van Kooyk et al., 1994). Whether the formation of integrin clusters precedes or follows ligand binding remains controversial (Bazzoni and Hemler, 1998; Carman and Springer, 2003). It was shown that on T-lymphocytes recruitment of LFA-1 within specialized membrane lipid microdomains enhances LFA-1 clustering (Krauss and Altevogt, 1999; Marwali et al., 2003), which might have direct implications for LFA-1 as a signaling molecule (Leitinger and Hogg, 2002).

Human monocytes (MOs) use LFA-1 to mediate antibody (Ab)-dependent cytotoxicity, to adhere to endothelial cells and extravasate into the underlying tissue (Martz, 1987; Springer, 1990). Recently, endothelial cells were shown to form upright microvilli-like projections that are enriched in ICAM-1 molecules (Carman et al., 2003). Transmigrating MOs present regions of increased density of LFA-1 molecules that form linear clusters and colocalize with the ICAM-1–enriched projections (Carman et al., 2003). As precursors of antigen-presenting immature dendritic cells (DCs; Banchereau and Steinman, 1998), MOs can differentiate in DCs upon reverse transmigration over endothelial monolayers (Randolph et al., 1998) or when cultured in vitro in presence of GM-CSF and IL-4 (moDCs; Romani et al., 1996). Although moDCs express LFA-1, unlike MOs, adhesion to ICAM-2 and -3 is completely mediated by the C-type lectin DC-SIGN (CD209; Geijtenbeek et al., 2000a, 2000b).

These observations suggest that the plasma membrane organization of adhesion receptors on MOs and DCs must change dynamically, along with the specific alterations occurring during the development from precursors toward DCs. Our previous studies showed that distribution of DC-SIGN at the cell surface dynamically changed during DC development along with its binding properties (Cambi et al., 2004). This prompted us to investigate and compare the relationship between cell membrane organization and adhesiveness of LFA-1 on MOs and DCs. Here, we demonstrate that LFA-1–mediated binding to ICAM-1 is completely lost during development of moDCs. This coincides with gradual exclusion of LFA-1 from lipid microdomains. We used high-resolution transmission EM (TEM) on whole-mount samples of MOs and DCs to map the cell surface distribution of LFA-1 at submicrometer level. Our findings demonstrate the existence of three levels of avidity for LFA-1: randomly distributed inactive molecules, well-defined ligand-independent nanoclusters, and ligand-triggered micrometer-sized macroclusters.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Monoclonal Antibodies and Chemicals
The mouse mAbs against L chain: NKI-L15, NKI-L16, and TS2/4 (kindly provided by E. Martz). Activating KIM185, kindly provided by M. Robinson, Celltech (Slough, United Kingdom), and the blocking NKI-L19 recognize the 2 chain. REK-1 is a mAb anti-ICAM-1. CD46 and CD55 were detected by E4.3 (PharMingen) and 143–30 (CLB, Amsterdam, The Netherlands), respectively. Alexa-conjugated secondary Abs, Alexa-488–conjugated cholera toxin B subunit, and the Alexa-647 labeling kit were from Molecular Probes (Eugene, OR) and Invitrogen; methyl--cyclodextrin (MCD) from Sigma (St. Louis, MO); rhodamine-conjugated goat anti-mouse IgG H&L Fab fragment from Abcam (Cambridge, United Kingdom); goat anti-CTxB antibody from Calbiochem (La Jolla, CA); goat anti-mouse-conjugated 10-nm gold from Aurion Biosystems (Vienna, Austria). The rat anti-cytohesin-1 Ab 7H2 was a gift of W. Kolanus (University of Bonn, Germany). The rat anti-RAPL Ab E11.2 was a gift of K. Katagiri (Kyoto University, Japan).

Cells
MOs were obtained from buffy coats of healthy individuals and were purified using Ficoll density centrifugation. Immature DCs were obtained as already reported elsewhere (Romani et al., 1994; Geijtenbeek et al., 2000c). The Jurkat T-cell line was kept in culture in Iscove's medium supplemented with 5% fetal calf serum.

Fluorescent Bead Adhesion Assay
Carboxylate-modified streptavidin-coated TransFluorSpheres (488/645 nm, 1 µm ; Molecular Probes) were coated with ICAM-1-Fc, and the bead adhesion assay was performed as described (Geijtenbeek et al., 1999). Briefly, the fluorescent beads were coated with biotinylated goat anti-human Fc antbodies and subsequently with ICAM-1-Fc chimeras. This guarantees the outwards orientation of the ICAM-1 molecules with respect to the bead surface. After each coating step, several thorough washing steps ensure that the excess of unbound molecules is washed away. For the incubation, a ratio of 20 beads per cell was used. The blocking or stimulating Abs were preincubated with the cells before adding the ligand-coated beads. When the lipid raft–disrupting agent MCD was used, the cells were resuspended in serum-free medium containing 20 mM MCD and preincubated for 30 min at 37°C. Adhesion was determined as the percentage of cells that bound fluorescent beads by flow cytometry on an FAC-Scalibur (Becton Dickinson, Oxnard, CA).

Flow Cytometry
For flow cytometry analysis, cells were incubated (30 min, 4°C) in PBS, 0.5% BSA, and 0.01% sodium azide, with different mAbs (5 µg/ml), followed by incubation with FITC-labeled goat anti-mouse IgG antibody (GAM-FITC; Zymed, South San Francisco, CA) for 30 min at 4°C. The relative fluorescence intensity was measured on a FACSCalibur. Isotype-specific controls were included.

Confocal Microscopy
For studies of colocalization with lipid rafts, cells were stained with mAbs and CTxB as already described (Cambi et al., 2004). Isotype-specific controls were always included. Analysis was done with a Bio-Rad MRC1024 Confocal Laser Scanning Microscope (Richmond, CA). Signals were collected sequentially to avoid bleed-through. To prevent loss of NKI-L16 binding in the double- or triple-labeling experiments, cells were not fixed before adding the primary Abs. They were allowed to adhere on FN-coated coverslips and stained with anti-LFA-1 (NKI-L16 or NKI-L15, or TS2/4) and anti-ICAM-1 (10 µg/ml) at 4°C, after prolonged incubation in CLSM buffer to minimize a specific binding. After removing unbound Abs by extensive washing in ice-cold PBS, cells were quickly fixed in 1% PFA, and after a blocking step in CLSM buffer, secondary staining for LFA-1 and ICAM-1 was performed using isotype-specific Alexa-conjugated goat anti-mouse antibodies. Manders coefficient M1 was calculated as already reported (Costes et al., 2004) and reflects the amount of LFA-1 colocalizing with GM1 separately for each cell.

Electron Microscopy Labeling Procedure
For TEM, MOs and DCs were allowed to spread on glass coverslips covered by a thin layer of poly-L-lysine (PLL)-coated Formvar for 1 h at 37°C, washed to remove unbound cells, and immediately fixed with 1% PFA for 15 min. Subsequent Ab and gold labeling was performed as already published (Cambi et al., 2004). When labeling was performed using NKI-L16, no prefixation with 1% PFA was performed to prevent loss of NKI-L16 binding. Therefore, after adhesion on the substrate, cells were washed in ice-cold buffer and incubated on ice with cold NKI-L16. Isotype-specific controls were always included. It should be noted that the effect of temperature changes on NKI-L16 labeling is negligible (Supplementary Figure 2.

Analysis of Gold Particle Distribution Pattern
After gold labeling and fixation, the specimens were dehydrated and transferred from the glass onto copper grids as already published (Cambi et al., 2004). The specimens were observed in a JEOL 1010 transmission electron microscope (Welwyn Garden City, United Kingdom), operating at 60–80 kV. Because MOs and DCs widely spread, the membrane available for gold particle analysis represented up to 60–70% of the whole labeled plasma membrane. For each cell several areas were analyzed at random. The digital images of electron micrographs were processed by custom-written software based on Labview (National Instruments, Austin, TX) as already described (Cambi et al., 2004).

Live Imaging
MOs adhered onto a FN-coated glass Petri dish (30 min, 37°C). After extensive washing in cold PBS, MOs were labeled with NKI-L16 (15 min, 4°C) and subsequently with the rhodamine-conjugated goat anti-mouse IgG H&L Fab fragment (15 min at 4°C). After thorough washing, the Alexa-647–conjugated TS2/4 (or NKI-L15) was added for 15 min at RT. After washing in Imaging medium (RPMI 1640, without phenol red), labeled MOs were put on the microscope setup at 37°C and 5% CO₂ for 10 min before imaging started. Subsequently, unlabeled Jurkat T-cells were added, and the interactions were analyzed with a Zeiss LSM 510 microscope, using a PlanApochromatic 63 x 1.4 oil immersion DIC lens (Carl Zeiss GmbH, Jena, Germany). While recording, the pinhole was completely open. Cells were imaged using Zeiss LSM Image Browser version 3.2 (Carl Zeiss) and processed with Image J version 1.32j software (National Institutes of Health, http://rsb.info.nih.gov/ij).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
LFA-1–mediated Binding Is Lost During Development of moDCs
We previously showed that moDCs express high amounts of DC-SIGN and moderate amounts of LFA-1. Both receptors are able to bind to ICAM-2 and -3, additionally LFA-1 binds to ICAM-1. While on MOs LFA-1 mediates adhesion to all ICAMs, on DCs adhesion to ICAM-2 and -3 is completely mediated by DC-SIGN (Geijtenbeek et al., 2000a, 2000b; Cambi et al., 2004). In an attempt to understand this change in LFA-1–binding capacity from MOs to DCs, we measured LFA-1–mediated binding to ICAM-1 Fc-coated beads during DC development (Geijtenbeek et al., 1999). MOs and DCs profoundly differ in their capacity to bind ICAM-1–coated fluorescent beads (Figure 1A). More than 60% of MOs spontaneously bound ICAM-1, and this interaction was predominantly 2-integrin mediated, as shown by the effective block in the presence of anti-2 blocking mAb NKI-L19. The use of anti-L blocking mAb (NKI-L15) indicated that on MOs, binding to ICAM-1 was predominantly mediated by LFA-1 (40%), whereas the other two 2-integrins MAC-1 and p150.95 mediated the residual binding.

View larger version (28K): [in this window] [in a new window]   Figure 1. Binding to ICAM-1 and L16 epitope expression decrease during development of moDCs. (A) Adhesion to ICAM-1 during development of moDCs (see drawing) was determined using 1-µm ligand-coated fluorescent beads, prepared as described in Materials and Methods. NKI-L15 and NKI-L19 mAbs were used to block LFA-1 and all 2 integrins, respectively. To enhance binding, the anti-2 activating mAb KIM185 was used. Neither blocking nor activation was observed in presence of isotype controls (unpublished data). One representative experiments out of three is shown. (B) The expression levels of LFA-1 on MOs and DCs were assessed by FACS analysis. , the isotype control; , the specific staining with anti-LFA-1 mAb. Mean fluorescence intensity is indicated. One representative donor is shown.

To exclude that the decreased binding to ICAM-1 on DCs was due to decreased expression levels, we detected LFA-1 expression by flow cytometry. Figure 1B shows that LFA-1 expression levels, detected with the mAb NKI-L15, remained unaltered during DC development.

The transition of LFA-1 from an inactive into an active state is known to depend on extracellular Ca²⁺ ions, and Ca²⁺ occupancy is reported by the chain–specific mAb NKI-L16, which detects the L16 epitope (Keizer et al., 1988; van Kooyk et al., 1991, 1994). Recent studies more specifically showed that NKI-L16 recognizes a Ca²⁺-bound extended conformation of the subunit, typical of primed LFA-1 (Xie et al., 2004). Therefore, we analyzed the L16 epitope expression on MOs and DCs. As shown in Figure 1B, the L16 epitope was highly expressed on MOs but was barely detectable on DCs.

It should be noted that the high expression level of the other 2 integrins on both MOs and DCs prevented us from analyzing other known activation reporter epitopes that involve the 2 subunit to monitor LFA-1 activation state.

Together, these observations indicate that major changes in LFA-1 adhesiveness occur during DC development.

Involvement of Lipid Rafts on LFA-1 Activity on MOs and DCs
The recruitment of active LFA-1 into specific cholesterol- and glycosphingolipids-enriched microdomains, known as lipid rafts, was proposed as additional mechanism to regulate integrin activity (Krauss and Altevogt, 1999; Leitinger and Hogg, 2002; Marwali et al., 2003). Therefore, we investigated whether lipid rafts played a role in LFA-1 adhesiveness on MOs and DCs.

On MOs, the effect of MCD, a lipid raft–disrupting agent that extracts membrane cholesterol, was tested on the ICAM-1 bead adhesion assay (Figure 2A). MCD treatment inhibits binding for 40%, indicating that lipid raft disruption partially affected LFA-1–binding capacity. Furthermore, we examined the codistribution of LFA-1 and the lipid raft marker GM1, a glycosphingolipid, on MOs and DCs, by Ab patching and confocal microscopy (Figure 2B). When cocapping was induced on MOs, LFA-1 completely colocalized with GM1 to the same extent as for the lipid raft–associated GPI-anchored protein CD55. By contrast, on DCs, LFA-1 is completely excluded from the lipid raft fraction.

View larger version (36K): [in this window] [in a new window]   Figure 2. LFA-1 resides in lipid rafts on MOs, not on DCs. (A) LFA-1–mediated adhesion to ICAM-1–coated beads was measured on MOs after cholesterol depletion by preincubation with 20 mM MCD for 30 min at 37°C. Data shown are means ± SD of one representative experiment (out of three) performed in triplicate. (B) Confocal microscopy analysis of copatching of LFA-1 (NKI-L15 labeled) and GM1 on MOs and DCs. Receptor copatching and staining were performed as described in Material and Methods. CD55 and CD46 are positive and negative lipid rafts marker, respectively. Results are representatives of multiple cells in three independent experiments. Bar, 5 µm. (C) To quantify the degree of colocalization between either NKI-L15 (total LFA-1) or NKI-L16 (primed LFA-1) with GM1 on MOs, the Manders coefficient (M1) was calculated. M1 can vary between 0 and 1 (1 = colocalization). Receptor copatching and staining were performed as described in Material and Methods, and cells were analyzed by confocal microscopy (n = 22).

On MOs, we quantified the colocalization of either NKI-L15 (total LFA-1) or NKI-L16 (primed LFA-1) with GM1 by calculating the Manders coefficient (Costes et al., 2004). No significant differences in colocalization with lipid rafts were observed between primed and total LFA-1 (Figure 2C), suggesting that on MO LFA-1 resides in lipid rafts even if the molecule is not in the Ca²⁺-bound extended primed conformation. Finally, the colocalization coefficient for LFA-1 and GM1 on DCs was found to be 0.14 ± 0.24 (n = 25), where 1 indicate full colocalization, thus indicating a significant exclusion of LFA-1 from lipid raft domains on DCs.

LFA-1 Is Organized in Nanoclusters on MOs
The organization of the cell membrane has been described as a mosaic of numerous lipid and protein microdomains ranging from 20 to 500 nm (Mayor and Rao, 2004). The fact that LFA-1 differentially associates with lipid rafts on MOs and DCs (Figure 2) prompted us to investigate the cell membrane organization of LFA-1 at high resolution.

We exploited TEM to study whole-mount samples of MOs and DCs, after specific labeling of LFA-1 with gold particles. Recently, we and others demonstrated the potential of this method to map and quantitate the distribution patterns of cell membrane receptors (Vereb et al., 2000; Panyi et al., 2003; Cambi et al., 2004).

MOs and DCs were allowed to adhere and stretch before labeling with mAb against LFA-1 and gold particles. It should be noted that the stretched cells are so thin (down to 200 nm at cell periphery) that sectioning is not required, thus making the whole dorsal membrane available for gold labeling and subsequent TEM analysis. To exclude any effect of the substrate on LFA-1 distribution, TEM analysis was also performed on cells that were gold-labeled in suspension and then mounted onto PLL. No differences were seen between cells stretched on the support or cells adhering to PLL (our unpublished data). We found that LFA-1 distribution changed dramatically during DC development (Figure 3). Although on MOs LFA-1 is organized in well-defined nanoclusters, on DCs the gold particles are evenly distributed over the cell surface.

View larger version (121K): [in this window] [in a new window]   Figure 3. LFA-1 is clustered on MOs and random on DCs. MOs and DCs were specifically labeled with 10-nm gold and treated for TEM (see Materials and Methods). Results represent multiple cells in several independent experiments. The top pictures represent whole cells imaged by TEM. Middle and bottom pictures are higher magnifications where 10-nm gold particles are visible. Bar, 200 nm unless otherwise indicated.

View larger version (28K): [in this window] [in a new window]   Figure 4. Quantitative analysis of the distribution of gold particles labeling LFA-1. The digital images were processed by a custom-written software based on Labview. Gold labels were counted, and coordinates were assigned to each feature. Interparticle distances were calculated using a nearest neighbor (nn) distance algorithm. (A) nn distance values were calculated for each image, and the data of several independent experiments were pooled. Subsequently, the nn distances were divided into three classes: 0–50, 50–100, and >100 nm, and the percentage of nn distance values falling into each class was plotted. (B) The partitioning of gold labels in clusters of various size (i.e., number of particles/cluster) was also quantified. Clusters were defined when gold particles were <50 nm apart from a neighboring particle. The percentage of gold particles involved in the formation of a certain cluster size was calculated. The insets are two representative processed digital images, where each type of cluster is shown in a different color. One representative experiment out of three is shown.

LFA-1 nanoclusters appear either as round or slightly elongated features, with an indicative average diameter of 150 nm, and are randomly localized on the cell. Similarly to NKI-L15, NKI-L16 staining also showed clustered distribution of gold-labeled LFA-1 on MO cell membrane (Figure 5A). However, when LFA-1 was labeled by NKI-L16, the number of gold particles per µm² was approximately fourfold lower when compared with NKI-L15 labeled cells (Figure 5B). Labeling with TS2/4 mAb, which recognizes another (inert) LFA-1 epitope, gave results similar to NKI-L15 (our unpublished data).

View larger version (38K): [in this window] [in a new window]   Figure 5. Resting and primed ligand-independent LFA-1 nanoclusters are expressed on the MO plasma membrane. (A) MOs were labeled as described in Figure 4, and the distribution patterns obtained with different anti-L mAbs were compared. (B) Quantitative analysis of the distribution of gold particle labeling LFA-1 was performed as described in Figure 4B, and gold labels detected per µm2 were counted. (C) Also, the number of clusters per µm2 was calculated. (D) The percentage of nn distance values falling into each class was plotted. Data are the mean of three independent experiments ±SD. (E) The partitioning of gold labels in clusters of various size was also quantified and compared between the mAbs. Clusters were defined as described in Figure 4B. One representative experiment out of three is shown. Scale bar, 200 nm.

Therefore, on MO LFA-1 nanoclusters are formed in a ligand-independent manner. Although most LFA-1 nanoclusters are in a resting state, 25% are in a primed state (L16 epitope expressed). Notably, L16 epitope expression is confined to a subset of nanoclusters rather than equally distributed over all clusters, suggesting the existence of clusters representing distinct states of integrin activation. MO differentiation into DC leads to complete dispersion of the nanoclusters and subsequent random distribution of single inactive LFA-1 molecules.

Primed LFA-1 Nanoclusters Readily Bind to ICAM-1
To further investigate the difference between primed and resting LFA-1 nanoclusters, we performed double labeling of LFA-1 using NKI-L16 and TS2/4 (nonoverlapping epitopes). As shown in Figure 6A, all LFA-1 nanoclusters are detectable by TS2/4, but only a subpopulation of these also shows expression of the L16 epitope (indicated in yellow). Interestingly, when MOs were seeded at high density, primed L16 epitope expressing LFA-1 was specifically enriched at the contact zone between adjacent cells, indicating that binding of ligand leads to preferential recruitment of primed LFA-1 clusters (Figure 6B).

View larger version (53K): [in this window] [in a new window]   Figure 6. The primed LFA-1 molecules colocalize with ICAM-1 and talin at the cell–cell contacts. (A) MOs were allowed to adhere onto PLL-coated glass coverslips for 30 min at 37°C. After extensive washing with cold PBS, cells were incubated with TS2/4 (red) and NKI-L16 (green) on ice for 30 min to label total LFA-1 and primed LFA-1, respectively. Unbound Abs were removed, and cells were fixed with 2% PFA. Isotype specific Alexa-conjugated goat anti-mouse Abs were added. The cells were analyzed by confocal microscopy, and colocalization is indicated in yellow. (B) MOs were seeded at high density on PLL-coated glass coverslips and labeled as described for A. Total LFA-1 is shown in green and primed fraction in red. Colocalization is indicated in yellow in the merged picture. Subsequently, triple labeling for total LFA-1 (green), primed LFA-1 (blue), and various interactors (red), such as the ligand ICAM-1 (C), talin (D), cytohesin-1 (E), and RapL (F), was performed on high-density seeded MOs. Colocalization is shown in the merged pictures. Bar, 10 µm, unless otherwise indicated.

Although several proteins have been identified that bind to the cytoplasmic tails of integrins (Liu et al., 2000), accumulating evidence suggests that talin plays a central role in activating LFA-1 (Kupfer and Singer, 1989; Monks et al., 1998; Sampath et al., 1998). In verifying which cytoplasmic regulators of LFA-1 were present at the contact sites, we observed that LFA-1 macroclusters engaged in binding with ICAM-1 completely colocalize with talin, which is clearly enriched at the contact sites (Figure 6D). Although we did not find a similar enrichment of other known regulators of LFA-1 such as cytohesin-1 (Geiger et al., 2000) and RapL (Katagiri et al., 2003; Figure 6, E and F), we cannot exclude that these proteins may have either a transient or a weaker role in regulating LFA-1 adhesion on MOs.

Next, we investigated whether the interaction of LFA-1 with talin is essential in the formation of primed nanoclusters even in absence of ligand binding. Ligand-independent Ab-induced capping of primed LFA-1 did not result in clear colocalization of the cytoplasmic proteins talin, cytohesin, and RapL, suggesting that the expression of the L16 epitope is not the result of pre-existing ligand-independent association of the cytoplasmic tails of LFA-1 with any of its known regulators (Supplementary Figure 1).

From these results we conclude that on the cell membrane of MO LFA-1 molecules are organized in nanoclusters formed without prior engagement of ligand and independent from known intracellular regulators such as talin.

Primed LFA-1 Nanoclusters Are Dynamically Recruited at MO–T-cell Interface and Form Macroclusters
In an attempt to understand if these LFA-1 macroclusters localized at the contact site between adjacent cells were derived from the dynamic recruitment of pre-existing nanoclusters of primed LFA-1, we analyzed contact formation between MOs and T-cells in real time. Freshly isolated MOs were allowed to adhere onto glass coverslips and prelabeled with fluorescent mAb TS2/4 and with NKI-L16, which was detected by a fluorescent secondary Fab fragment. After washing unbound Abs, T-cells were added, and the formation of MO–T-cell conjugates was followed in time. It is important to note that resting T-cells express ICAM-1 but do not express the L16 epitope and do not spontaneously bind to ICAM-1, indicating that on these T-cells LFA-1 is resting (our unpublished data). In Figure 7A, a clear enrichment of primed LFA-1 molecules can be observed at the contact area between MOs and T-cells from 270 to 810 s after contact initiation. Importantly, the presence of neither TS2/4 nor NKI-L16 significantly increases binding of MOs (nor of DCs) to ICAM-1–coated beads (our unpublished data), suggesting that the cell–cell interaction cannot be attributed to the presence of these Abs. When the interaction between MOs and T-cells occurred in presence of the blocking mAb NKI-L15, hardly any conjugate was observed, and no recruitment of primed LFA-1 could be detected (Figure 7B).

View larger version (89K): [in this window] [in a new window]   Figure 7. The primed LFA-1 nanoclusters are rapidly recruited at the MO–T-cell contact. (A) MOs were allowed to adhere at low density onto a glass coverslip for 30 min at 37°C. After washing with PBS, MOs were labeled with anti-aL mAbs NKI-L16 and TS2/4, as described in Materials and Methods. After removal of unbound Abs, Jurkat T-cells were added. The formation of spontaneous conjugates between MOs and T-cells was followed at 37°C with a Zeiss LSM 510 microscope. (B) In presence of the blocking mAb NKI-L15, the number of conjugates was lower, and no enrichment of NKI-L16–positive LFA-1 molecules at the cell–cell contact was observed.

View larger version (27K): [in this window] [in a new window]   Figure 8. Quantification of the recruitment of the primed LFA-1 fraction at the MO–T-cell contact site. As shown in the cartoon, the recruitment index (RI) of each Ab was defined as the fluorescence intensity at the contact site divided by the fluorescence intensity of the whole cell. Data in A and B are referred to the images shown in Figure 7, A and B, respectively, and are representative of multiple cells in three independent experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The leukocyte specific 2 integrin LFA-1 regulates cell adhesion, migration, and immunological synapse formation. Here we demonstrate for the first time, exploiting high-resolution TEM and spatial-point pattern analysis, that LFA-1–mediated cell adhesion is directly correlated with submicrometer-sized avidity changes of LFA-1 when MOs differentiate into DCs.

On DC LFA-1 Is Inactive
MOs can readily differentiate into antigen-presenting DC upon reverse transmigration of endothelial monolayers (Randolph et al., 1998) or when cultured in vitro in presence of GM-CSF and IL-4 (Romani et al., 1996). Therefore, by using MOs and moDCs instead of LFA-1 transfectants we followed changes in the behavior of endogenous adhesion receptors in a semiphysiological system. Moreover, moDC are currently manipulated ex vivo for antitumor and antiviral therapy (Figdor et al., 2004). Detailed characterization of the migratory and adhesive properties of moDCs is therefore very important and can have direct consequences for the currently executed clinical trials.

Although MOs and DCs express equivalent levels of LFA-1, there is a marked difference in LFA-1 activation state on the two cell types (Figure 1A). High-resolution mapping of LFA-1 cell surface distribution patterns by TEM and quantitative spatial point pattern analysis now revealed that on DCs the resting LFA-1 molecules are randomly distributed on the cell membrane as single features (Figure 3). Similar observations were made on K562 cells transfected with LFA-1 (K-LFA-1), which do not bind ICAM-1 and exhibit a random cell surface distribution of LFA-1 (our unpublished data). These data are in agreement with recent observation that the absence of FRET-detectable microclustering corresponds with lack of binding activity on K-LFA-1 cells (Kim et al., 2004).

Crystallography, NMR, and negative stain EM showed different conformational states of integrin fragments, revealing structural rearrangements of both and subunits upon activation (Beglova et al., 2002; Takagi et al., 2002; Xiao et al., 2004). In the low-affinity conformation, the headpiece of the integrin faces down toward the cell membrane; in the high-affinity conformation, the headpiece extends upward. This molecular switch occurs at the "genu" of the integrin subunits (Takagi and Springer, 2002). In particular, the subunit leg has been shown to extend at the Ca²⁺-dependent epitope near the thigh–genu interface (Xie et al., 2004). This epitope appears only in the extended conformation and in the case of L is specifically recognized by the conformation-sensitive mAb NKI-L16 (Keizer et al., 1988; Xie et al., 2004). We have shown that LFA-1–binding capacity is directly correlated to the expression of this epitope (Figure 1B). Because in the L inactive conformation the L16 epitope is shielded from Ab recognition, the complete lack of L16 epitope on DCs suggests that all LFA-1 molecules are in the "off," i.e., bent conformation. A recent publication of van Gisbergen et al. (2005) suggests that on DCs inactive LFA-1 may be triggered upon interaction of DCs with T-cells (van Gisbergen et al., 2005). Further studies are needed to unravel the exact signaling pathways that may trigger resting LFA-1 molecules on DCs.

LFA-1 Nanoclusters Are Either Resting or Primed
On MOs, LFA-1 is organized in primed nanoclusters (L16 positive) and resting nanoclusters (L16 negative). Because no apparent differences were found in gold particles proximity or cluster size when comparing resting and primed nanoclusters (Figure 5), we hypothesize that clustering and priming represent two distinct events and LFA-1 molecules must all adopt either the bent inactive or the extended primed conformation within a nanocluster. The underlying mechanism remains unknown, but it is tempting to speculate that a cooperative effect between the packed LFA-1 molecules may exist. Recently, Nishiuchi et al. (2005) described the role of the tetraspanin CD151 in modulating the integrin 31 adhesiveness by stabilizing its active conformation through physical interaction. Other tetraspanins were shown to be involved in the regulation of LFA-1 adhesiveness (Shibagaki et al., 1999; Van Compernolle et al., 2001). We are currently investigating the role of these tetraspanins in the formation of LFA-1 nanoclusters.

The arrangement of LFA-1 in nanoclusters may also guarantee an efficient binding and rebinding of LFA-1 to the ICAM-1 homodimers expressed on the endothelium. Biochemical studies showed that a divalent LFA-1 is the minimum configuration for forming efficient bonds with ICAM-1 (Sarantos et al., 2005). LFA-1 nanoclusters may further increase the likelihood of LFA-1 binding to ICAM-1. Shamri et al. (2005) recently showed that on rolling lymphocytes LFA-1 activation occurs locally and abruptly and that priming of LFA-1 molecules (i.e., L16 epitope induction) is a prerequisite for ICAM-1–induced activation of the subunit I-domain.

LFA-1 Nanoclusters Are Formed Independently of Ligand Binding
It has long been debated whether clustering of integrins triggers or rather results from ligand binding. Here, we present "snapshots" of the MO cell membrane, in the absence of ligand binding, where LFA-1 shows a well-defined clustered distribution pattern. Because integrin activity on circulating leukocytes is up-regulated within seconds (Shamri et al., 2005), organization of LFA-1 in small predefined nanoclusters would favor instant interaction with the endothelial surface. Carman and colleagues showed that on MO linear clusters of LFA-1 colocalize with ICAM-1–containing microvilli on the endothelium before MO extravasation (Carman and Springer, 2004). Our observations extend their findings and suggest that on MO cell surface, preformed LFA-1 nanoclusters may facilitate or accelerate the interaction of MOs with these endothelial ICAM-1–enriched projections.

Although it was suggested that localized clustering of LFA-1 at the cell surface might simply result from previous encounters of LFA-1 with ICAM ligands exposed by adjacent cells (Kim et al., 2004), our current findings suggest a more complex mechanism. Kim et al. (2004) showed that only binding of multimeric ligand (i.e., ICAM-1Fc/Anti-IgA complexes) to LFA-1–induced FRET-detectable microclustering on K-LFA-1 cells. We observed that when MOs are allowed to adhere at low density to prevent cell contact with ICAM-1 molecules from adjacent cells, nanoclusters are still present. Moreover, inhibition of LFA-1/ICAMs interactions in cis does not alter LFA-1 nanocluster organization (Supplementary Figure 3). Furthermore, DCs, which do not have LFA-1 nanoclusters, cannot be induced to form LFA-1 aggregates, even when cultured at high cell density (our unpublished data). However, we do not exclude that upon ligand binding a further increase in receptor packing might occur within the LFA-1 nanoclusters that would induce FRET-detectable changes in intradomain receptor proximity. Further high-resolution analysis of the integrin molecules within these assemblies is needed to provide detailed insight in the molecular and spatial organization of LFA-1 nanoclusters.

LFA-1 Macroclusters Derive from the Recruitment of Pre-existing Nanoclusters
We investigated the interaction between MOs double-labeled for LFA-1 (NKI-L16 and TS2/4) and unlabeled ICAM-1–bearing T-cells by live imaging and quantified LFA-1 accumulation at the cell–cell contact (Figures 7 and 8). The RI of primed LFA-1 molecules at the MO–T-cell contact site is significantly higher than the RI of the total LFA-1 (Figure 8). This indicates that the preformed L16-positive nanoclusters act as "primed" scaffolds of LFA-1 molecules that readily bind the ligand. These scaffolds are recruited at the contact site where larger micrometer-sized macroclusters are subsequently formed (Figure 9).

View larger version (19K): [in this window] [in a new window]   Figure 9. Major changes in LFA-1 avidity and adhesiveness occur during DC development. This cartoon summarizes the major phenotypical changes observed on DCs compared with their precursor MO. Although on immature DCs, LFA-1 is unable to bind to its ligands, is randomly distributed at the cell surface and completely excluded from the lipid rafts, on MO LFA-1 is organized in well- defined submicrometer sized nanoclusters. LFA-1 nanoclusters reside in lipid rafts but differ in terms of activation state: a fraction is primed and express the L16 epitope, the rest is in a bent resting conformation. By interaction with ICAM-1–bearing cells, the primed nanoclusters are recruited at the contact site, generating micrometer-sized macroclusters. There, LFA-1 binds to the counterreceptor ICAM-1 and establishes additional interactions with the cytoplasmic protein talin. In cell types other than MO, other regulators, such as RapL, might be then recruited from the cytoplasm and further strengthen binding and modulate signaling.

Our observation that nanocluster formation does not require interaction with talin, whereas ligand-engaged macroclusters fully colocalize with talin, is in agreement with findings of Shamri et al. (2005). They observed that LFA-1 tethering to low-density ICAM-1 under shear flow does not require intact talin, whereas at increasing ICAM-1 density, the contribution of talin also increases (Shamri et al., 2005). In addition, Xie et al. (2004) have mapped the L16 epitope by using a chimeric construct where extracellular and chains were fused to the same transmembrane anchoring domain of platelet-derived growth factor receptor. This construct allows the chimeric protein to be displayed on the cell surface but lacks the integrin cytoplasmic tail motifs that modulate the interaction with talin. Therefore, the formation of ligand-independent primed nanoclusters does not require interaction between LFA-1 and talin, whereas strengthening of cell–cell contacts is specifically supported by the presence of talin.

LFA-1 Nanoclusters and Lipid Rafts
The organization of membrane receptors in microdomains is often associated with their colocalization with lipid rafts. On T-cells, only active LFA-1 was found to reside in lipid rafts (Leitinger and Hogg, 2002). Interestingly, our studies on MOs show that both primed and resting LFA-1 nanoclusters colocalize with lipid rafts, indicating that neither clustering nor association with lipid rafts per se are sufficient for priming of LFA-1. However, on DCs random inactive LFA-1 is completely excluded by the lipid raft environment (Figure 2). The mechanism by which integrins are recruited into lipid rafts remains unclear. Because no clear lipid modification (such as myristoylation or palmitoylation) has been reported for integrin molecules, the differentiated association of LFA-1 with lipid rafts is most likely regulated by complex formation between the integrin and other as yet undefined membrane proteins.

The organization of LFA-1 in nanoclusters residing in lipid rafts may also explain why HIV-1 particles bearing host-derived ICAM-1 are efficiently spread by MOs but not by moDCs (Bounou et al., 2004). Recently, we made similar observations for DC-SIGN, which binds and internalizes HIV-1 particles only when it is organized in submicrometer-sized microdomains, not when randomly distributed (Cambi et al., 2004). Therefore, LFA-1 nanoclusters on MOs may be exploited by host-derived ICAM-1-bearing HIV-1 through a similar mechanism using LFA-1 nanoclusters as docking platforms.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
We demonstrated that phenotypical changes occurring during development of precursor cells such as MOs into DCs involve also rearrangement of receptor organization at the cell surface. By using high resolution microscopy, we visualized distinct stages of integrin avidity: random inactive molecules, ligand-independent nanoclusters, and ligand-triggered macroclusters. Our findings reveal a novel mechanism to regulate integrin function. Well-defined pre-existing nanoclusters of LFA-1 equip the cell with a machinery to respond instantly when action is required, such as binding to endothelium or interacting with other leukocytes. The absence of these nanoclusters on human moDCs explains why these cells lack the capacity to bind ICAM-1 despite significant expression of LFA-1. The present challenge is to better understand the mechanism that regulates the submicrometer-sized avidity changes of LFA-1 during DC development.

	ACKNOWLEDGMENTS

 
We thank Huib Croes for his help with the EM studies. This work was supported by Stichting voor Fundamenteel Onderzoek der Materie (FOM) to M.K. and the Dutch Cancer Society (KWF) to F.v.L. C.G.F. is supported by Grant NWO 901-10-092 from the Netherlands Organisation for Scientific Research. The authors do not have any conflicting financial interests.

	Footnotes

 
The online version of this article contains supplemental material at MBC Online (http://www.molbiolcell.org).

This article was published online ahead of print in MBC in Press (http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E05-12-1098) on July 19, 2006.

Present address: Laboratory of NanoBioengineering, Parc Científic de Barcelona (PCB), Josep Samitier 1-5, 08028 Barcelona, Spain, and ICREA-Institució Catalana de Recerca i Estudis Avançats, 08015 Barcelona, Spain.

Address correspondence to: Carl G. Figdor (c.figdor{at}ncmls.ru.nl)

Abbreviations used: LFA-1, lymphocyte function–associated antigen-1; MO, monocytes; DC, dendritic cell; moDC, monocyte-derived dendritic cell; TEM, transmission electron microscopy.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Banchereau, J. and Steinman, R. M. (1998). Dendritic cells and the control of immunity. Nature 392, 245–252.[CrossRef][Medline]

Bazzoni, G. and Hemler, M. E. (1998). Are changes in integrin affinity and conformation overemphasized? Trends Biochem. Sci 23, 30–34.[CrossRef][Medline]

Beals, C. R., Edwards, A. C., Gottschalk, R. J., Kuijpers, T. W., Staunton, D. E. (2001). CD18 activation epitopes induced by leukocyte activation. J. Immunol 167, 6113–6122.[Abstract/Free Full Text]

Beglova, N., Blacklow, S. C., Takagi, J., Springer, T. A. (2002). Cysteine-rich module structure reveals a fulcrum for integrin rearrangement upon activation. Nat. Struct. Biol 9, 282–287.[CrossRef][Medline]

Bounou, S., Giguere, J. F., Cantin, R., Gilbert, C., Imbeault, M., Martin, G., Tremblay, M. J. (2004). The importance of virus-associated host ICAM-1 in human immunodeficiency virus type 1 dissemination depends on the cellular context. FASEB J 18, 1294–1296.[Abstract/Free Full Text]

Cambi, A., et al. (2004). Microdomains of the C-type lectin DC-SIGN are portals for virus entry into dendritic cells. J. Cell Biol 164, 145–155.[Abstract/Free Full Text]

Carman, C. V., Jun, C. D., Salas, A., Springer, T. A. (2003). Endothelial cells proactively form microvilli-like membrane projections upon intercellular adhesion molecule 1 engagement of leukocyte LFA-1. J. Immunol 171, 6135–6144.[Abstract/Free Full Text]

Carman, C. V. and Springer, T. A. (2003). Integrin avidity regulation: are changes in affinity and conformation underemphasized? Curr. Opin. Cell Biol 15, 547–556.[CrossRef][Medline]

Carman, C. V. and Springer, T. A. (2004). A transmigratory cup in leukocyte diapedesis both through individual vascular endothelial cells and between them. J. Cell Biol 167, 377–388.[Abstract/Free Full Text]

Costes, S. V., Daelemans, D., Cho, E. H., Dobbin, Z., Pavlakis, G., Lockett, S. (2004). Automatic and quantitative measurement of protein-protein colocalization in live cells. Biophys. J 86, 3993–4003.[Medline]

de Fougerolles, A. R. and Springer, T. A. (1992). Intercellular adhesion molecule 3, a third adhesion counter-receptor for lymphocyte function-associated molecule 1 on resting lymphocytes. J. Exp. Med 175, 185–190.[Abstract/Free Full Text]

de Fougerolles, A. R., Stacker, S. A., Schwarting, R., Springer, T. A. (1991). Characterization of ICAM-2 and evidence for a third counter-receptor for LFA-1. J. Exp. Med 174, 253–267.[Abstract/Free Full Text]

Dransfield, I. and Hogg, N. (1989). Regulated expression of Mg2+ binding epitope on leukocyte integrin alpha subunits. EMBO J 8, 3759–3765.[Medline]

Dustin, M. L. and Springer, T. A. (1989). T-cell receptor cross-linking transiently stimulates adhesiveness through LFA-1. Nature 341, 619–624.[CrossRef][Medline]

Figdor, C. G., de Vries, I. J., Lesterhuis, W. J., Melief, C. J. (2004). Dendritic cell immunotherapy: mapping the way. Nat. Med 10, 475–480.[CrossRef][Medline]

Geiger, C., et al. (2000). Cytohesin-1 regulates beta-2 integrin-mediated adhesion through both ARF-GEF function and interaction with LFA-1. EMBO J 19, 2525–2536.[CrossRef][Medline]

Geijtenbeek, T. B., Krooshoop, D. J., Bleijs, D. A., van Vliet, S. J., van Duijnhoven, G. C., Grabovsky, V., Alon, R., Figdor, C. G., van Kooyk, Y. (2000a). DC-SIGN-ICAM-2 interaction mediates dendritic cell trafficking. Nat. Immunol 1, 353–357.[CrossRef][Medline]

Geijtenbeek, T. B., et al. (2000b). DC-SIGN, a dendritic cell-specific HIV-1-binding protein that enhances trans-infection of T cells. Cell 100, 587–597.[CrossRef][Medline]

Geijtenbeek, T. B., Torensma, R., van Vliet, S. J., van Duijnhoven, G. C., Adema, G. J., van Kooyk, Y., Figdor, C. G. (2000c). Identification of DC-SIGN, a novel dendritic cell-specific ICAM-3 receptor that supports primary immune responses. Cell 100, 575–585.[CrossRef][Medline]

Geijtenbeek, T. B., van Kooyk, Y., van Vliet, S. J., Renes, M. H., Raymakers, R. A., Figdor, C. G. (1999). High frequency of adhesion defects in B-lineage acute lymphoblastic leukemia. Blood 94, 754–764.[Abstract/Free Full Text]

Grakoui, A., Bromley, S. K., Sumen, C., Davis, M. M., Shaw, A. S., Allen, P. M., Dustin, M. L. (1999). The immunological synapse: a molecular machine controlling T cell activation. Science 285, 221–227.[Abstract/Free Full Text]

Katagiri, K., Maeda, A., Shimonaka, M., Kinashi, T. (2003). RAPL, a Rap1-binding molecule that mediates Rap1-induced adhesion through spatial regulation of LFA-1. Nat. Immunol 4, 741–748.[CrossRef][Medline]

Keizer, G. D., Visser, W., Vliem, M., Figdor, C. G. (1988). A monoclonal antibody (NKI-L16) directed against a unique epitope on the alpha-chain of human leukocyte function-associated antigen 1 induces homotypic cell-cell interactions. J. Immunol 140, 1393–1400.[Abstract]

Kim, M., Carman, C. V., Yang, W., Salas, A., Springer, T. A. (2004). The primacy of affinity over clustering in regulation of adhesiveness of the integrin áLâ2. J. Cell Biol 167, 1241–1253.[Abstract/Free Full Text]

Krauss, K. and Altevogt, P. (1999). Integrin leukocyte function-associated antigen-1-mediated cell binding can be activated by clustering of membrane rafts. J. Biol. Chem 274, 36921–36927.[Abstract/Free Full Text]

Kupfer, A. and Singer, S. J. (1989). The specific interaction of helper T cells and antigen-presenting B cells. IV. Membrane and cytoskeletal reorganizations in the bound T cell as a function of antigen dose. J. Exp. Med 170, 1697–1713.[Abstract/Free Full Text]

Leitinger, B. and Hogg, N. (2002). The involvement of lipid rafts in the regulation of integrin function. J. Cell Sci 115, 963–972.[Abstract/Free Full Text]

Liu, S., Calderwood, D. A., Ginsberg, M. H. (2000). Integrin cytoplasmic domain-binding proteins. J. Cell Sci 113, (Pt 20)3563–3571.[Abstract]

Lu, C., Shimaoka, M., Ferzly, M., Oxvig, C., Takagi, J., Springer, T. A. (2001a). An isolated, surface-expressed I domain of the integrin alphaLbeta2 is sufficient for strong adhesive function when locked in the open conformation with a disulfide bond. Proc. Natl. Acad. Sci. USA 98, 2387–2392.[Abstract/Free Full Text]

Lu, C., Shimaoka, M., Zang, Q., Takagi, J., Springer, T. A. (2001b). Locking in alternate conformations of the integrin alphaLbeta2 I domain with disulfide bonds reveals functional relationships among integrin domains. Proc. Natl. Acad. Sci. USA 98, 2393–2398.[Abstract/Free Full Text]

Lub, M., van Kooyk, Y., Figdor, C. G. (1995). Ins and outs of LFA-1. Immunol. Today 16, 479–483.[CrossRef][Medline]

Marlin, S. D. and Springer, T. A. (1987). Purified intercellular adhesion molecule-1 (ICAM-1) is a ligand for lymphocyte function-associated antigen 1 (LFA-1). Cell 51, 813–819.[CrossRef][Medline]

Martz, E. (1987). LFA-1 and other accessory molecules functioning in adhesions of T and B lymphocytes. Hum. Immunol 18, 3–37.[Medline]

Marwali, M. R., Rey-Ladino, J., Dreolini, L., Shaw, D., Takei, F. (2003). Membrane cholesterol regulates LFA-1 function and lipid raft heterogeneity. Blood 102, 215–222.[Abstract/Free Full Text]

Mayor, S. and Rao, M. (2004). Rafts: scale-dependent, active lipid organization at the cell surface. Traffic 5, 231–240.[CrossRef][Medline]

Monks, C. R., Freiberg, B. A., Kupfer, H., Sciaky, N., Kupfer, A. (1998). Three-dimensional segregation of supramolecular activation clusters in T cells. Nature 395, 82–86.[CrossRef][Medline]

Nishiuchi, R., Sanzen, N., Nada, S., Sumida, Y., Wada, Y., Okada, M., Takagi, J., Hasegawa, H., Sekiguchi, K. (2005). Potentiation of the ligand-binding activity of integrin alpha3beta1 via association with tetraspanin CD151. Proc. Natl. Acad. Sci. USA 102, 1939–1944.[Abstract/Free Full Text]

Panyi, G., et al. (2003). Colocalization and nonrandom distribution of Kv1.3 potassium channels and CD3 molecules in the plasma membrane of human T lymphocytes. Proc. Natl. Acad. Sci. USA 100, 2592–2597.[Abstract/Free Full Text]

Randolph, G. J., Beaulieu, S., Lebecque, S., Steinman, R. M., Muller, W. A. (1998). Differentiation of monocytes into dendritic cells in a model of transendothelial trafficking. Science 282, 480–483.[Abstract/Free Full Text]

Romani, N., Gruner, S., Brang, D., Kampgen, E., Lenz, A., Trockenbacher, B., Konwalinka, G., Fritsch, P. O., Steinman, R. M., Schuler, G. (1994). Proliferating dendritic cell progenitors in human blood. J. Exp. Med 180, 83–93.[Abstract/Free Full Text]

Romani, N., Reider, D., Heuer, M., Ebner, S., Kampgen, E., Eibl, B., Niederwieser, D., Schuler, G. (1996). Generation of mature dendritic cells from human blood. An improved method with special regard to clinical applicability. J. Immunol. Methods 196, 137–151.[CrossRef][Medline]

Sampath, R., Gallagher, P. J., Pavalko, F. M. (1998). Cytoskeletal interactions with the leukocyte integrin beta2 cytoplasmic tail. Activation-dependent regulation of associations with talin and alpha-actinin. J. Biol. Chem 273, 33588–33594.[Abstract/Free Full Text]

Sarantos, M. R., Raychaudhuri, S., Lum, A. F., Staunton, D. E., Simon, S. I. (2005). Leukocyte function-associated antigen 1-mediated adhesion stability is dynamically regulated through affinity and valency during bond formation with intercellular adhesion molecule-1. J. Biol. Chem 280, 28290–28298.[Abstract/Free Full Text]

Shamri, R., Grabovsky, V., Gauguet, J. M., Feigelson, S., Manevich, E., Kolanus, W., Robinson, M. K., Staunton, D. E., von Andrian, U. H., Alon, R. (2005). Lymphocyte arrest requires instantaneous induction of an extended LFA-1 conformation mediated by endothelium-bound chemokines. Nat. Immunol 6, 497–506.[CrossRef][Medline]

Shibagaki, N., Hanada, K., Yamashita, H., Shimada, S., Hamada, H. (1999). Overexpression of CD82 on human T cells enhances LFA-1/ICAM-1-mediated cell-cell adhesion: functional association between CD82 and LFA-1 in T cell activation. Eur. J. Immunol 29, 4081–4091.[CrossRef]