The leukocyte-restricted integrin αLβ2 is required in immune processes such as leukocyte adhesion, migration, and immune synapse formation. Activation of αLβ2 by conformational changes promotes αLβ2 binding to its ligands, ICAMs. It was reported that different affinity states of αLβ2 are required for binding ICAM-1 and ICAM-3. Recently, the bent, extended with a closed headpiece, and extended with open headpiece conformations of αLβ2, was reported. To address the overall conformational requirements of αLβ2 that allow selective binding of these ICAMs, we examined the adhesion properties of these αLβ2 conformers. αLβ2 with different conformations were generated by mutations, and verified by using a panel of reporter mAbs that detect αLβ2 extension, hybrid domain movement, or I-like domain activation. We report a marked difference between extended αLβ2 with closed and open headpieces in their adhesive properties to ICAM-1 and ICAM-3. Our data show that the extension of αLβ2 alone is sufficient to mediate ICAM-1 adhesion. By contrast, an extended αLβ2 with an open headpiece is required for ICAM-3 adhesion.

Integrins are type I membrane proteins having an α and a β subunit that are noncovalently associated (1). Twenty-four human integrin heterodimers have been identified. Coupled with a growing list of cytosolic signaling cascades emanating from each of these heterodimers, integrins serve pivotal roles in a wide spectrum of biological processes, including cell motility, growth, and differentiation (1). The importance of integrins on platelets and leukocytes is well exemplified in Glanzmann’s thrombasthenia, leukocyte adhesion deficiency (LAD)3 I, and LAD III (2, 3). Glanzmann’s thrombasthenia is a bleeding disorder due to defective expression or function of integrin αIIbβ3. Leukocytes of LAD I showed defective adhesive properties to substratum as a result of mutations in the integrin β2 subunit that forms the subfamily of β2 integrins that comprise αLβ2, αMβ2, αXβ2, and αDβ2. Many of these mutations affect the ligand-binding affinity of these integrins. In LAD III, the expression, structure, and intrinsic adhesive property of the platelet and leukocyte integrins remain intact, but they show defects in G protein-coupled receptor-triggered activation (3). This is attributed to the dysfunctional regulation of Rap-1, a Ras-related GTPase involved in integrin activation (4, 5).

Integrin affinity modulation is driven by receptor conformational changes promoted by extracellular and/or cytosolic factors (6). The emerging paradigm of integrin activation describes the transition of an integrin from a bent conformation with an obtuse angle to one that is highly extended as a hallmark of integrin activation that is necessary for effective ligand binding (7, 8). However, there are also reports that support ligand binding by integrins in a bent conformation (9, 10, 11, 12). Three predominant integrin conformers that may depict the conformation of integrin during different stages of its activation are observed from electron microscopy studies of αLβ2 and αXβ2 (13). These are the bent conformer, the extended conformer with a closed headpiece, and the extended conformer with an open headpiece. The swing-out of the hybrid domain is a distinguishing feature between the closed and open headpieces apart from other structural changes in the I-like domain (also referred to as the βI or βA domain) (Fig. 1) (14). These conformers may bear significance in physiology because they can present different ligand-binding properties, which are necessary for the fine regulation of cell adhesiveness in different microenvironments.

FIGURE 1.

Illustration of αLβ2 in three conformations. The model of αLβ2 was generated as described under Materials and Methods. The three predominant populations of αLβ2 reported are the bent conformer, an extended conformer with closed headpiece, and an extended conformer with open headpiece (13 ).

FIGURE 1.

Illustration of αLβ2 in three conformations. The model of αLβ2 was generated as described under Materials and Methods. The three predominant populations of αLβ2 reported are the bent conformer, an extended conformer with closed headpiece, and an extended conformer with open headpiece (13 ).

Close modal

Integrin αLβ2 (LFA-1; CD11aCD18) serves major roles in leukocyte biology, which include leukocyte adhesion, migration, and immune synapse formation (15). The ligands of αLβ2 are the ICAMs (16, 17, 18, 19, 20) and the junctional adhesion molecule (21). ICAM-1 is expressed on many cell types, including activated endothelial cells and leukocytes, whereas ICAM-3 is predominantly expressed at high levels on leukocytes and epidermal Langerhans cells (18, 22, 23). ICAM-1 is a key ligand of αLβ2 for leukocyte rolling adhesion in shear flow and for the formation of the immune synapse (24, 25). ICAM-3 was reported to play a role in early Ag-independent interactions between T cells and APCs, and cross-linking of ICAM-3 on T cells regulates αLβ2-ICAM-1 interaction (26, 27). Interestingly, it was also reported that different ICAMs trigger differential αLβ2 signaling events that regulate T cell differentiation (28).

In line with the binding affinity of recombinant αL I domain to ICAMs that is in the order of ICAM-1 > ICAM-2 > ICAM-3 (29), the adhesion of αLβ2-bearing cells to ICAM-3 as compared with ICAM-1 required higher level of αLβ2 activation as defined by the number of exogenous activating agents required (30). We showed previously that disrupting salt-bridge formation of the αLβ2 cytoplasmic tails by mutation or the overexpression of talin head domain promotes effective ICAM-1, but not ICAM-3 adhesion (31). Furthermore, we reported an LAD 1 mutation in the β2 I-like domain that promoted constitutive cell adhesion to ICAM-3 (32). These are highly suggestive, but incomplete data that support the requirement of different αLβ2 conformations for binding to ICAM-1 and ICAM-3. The αLβ2 in the bent, extended with closed headpiece, and extended with open headpiece conformations, compose a model of αLβ2 activation that could provide selective binding to the ICAMs. Thus, we set out to systematically analyze the adhesive properties of these αLβ2 conformers to the ICAMs in this study. These αLβ2 conformers were generated by mutations, and verified by conformation-sensitive reporter mAbs. We showed that αLβ2 extension on its own is sufficient for ICAM-1, but not ICAM-3 adhesion. The latter requires both αLβ2 extension and the opening of the headpiece.

These mAbs were provided by others, as follows: MHM24 hybridoma (αL-specific, function-blocking mAb) (33) and MHM23 (β2-specific, heterodimer reporter mAb) (34, 35) were obtained from A. McMichael (John Radcliffe Hospital, Oxford, U.K.); m24 (β2 specific; reports I-like domain activation) (36, 37) was from N. Hogg (Leukocyte Adhesion Laboratory, London Research Institute, Lincoln’s Inn Fields Laboratories, London, U.K.). MEM148 (β2 specific; reports hybrid domain movement) (30) was purchased from AbD Serotec. The hybridomas of KIM185 (β2 specific; activates β2 integrins) (38), KIM127 (β2 specific; reports an extended β2 integrin) (39, 40), and IB4 (β2 specific that recognizes only the heterodimer form of β2 integrins) (41) hybridoma were obtained from American Type Culture Collection. The preparation of recombinant human ICAM-1-Fc and ICAM-3-Fc was described previously (42). All general chemicals and reagents were purchased from Sigma-Aldrich or Merck, unless otherwise stated.

The expression plasmid pcDNA3 (Invitrogen Life Technologies) containing the integrin αL and β2 cDNAs was described previously (42). The amino acids of the integrins are numbered with reference to Barclay et al. (43). The following αLβ2 mutants were reported: mutant β2D709R with salt-bridge disruption (31); activating LAD 1 mutation in the β2 I-like domain β2N329S (32); and substitutions of Lys287 and Lys294 to cysteines in the αL I domain generating a high-affinity αLcch (h for high affinity) (37). The cysteine-lock low-affinity I domain mutant αLccl (l for low affinity) was generated by replacing Leu289 and Lys294 of the αL I domain with cysteines (37). The β2 glycan mutant was generated by introducing two amino acid substitutions Q295T and P296L in the β2 sequence (44). All point mutations aforementioned were generated using the Quikchange site-directed mutagenesis kit (Stratagene) with relevant primer pairs.

The fluorescence resonance energy transfer (FRET) pair expression vectors pEYFP-N1 and pECFP-N1 (BD Clontech) were used to generate αL and β2 subunits with cyan fluorescent protein (CFP) and yellow fluorescent protein (YFP) fused to the cytoplasmic tails, respectively. To inhibit their inherent tendency to form hetero- or homodimers, monomeric CFP (mCFP) and monomeric YFP (mYFP) were generated by L221K substitution found at the dimer interface (45). mCFP and mYFP were subcloned into integrin expression vectors to generate αL mCFP, β2 mYFP, β2D709R mYFP, β2glycan mYFP, and β2N329S mYFP. The αL mCFP has 5-aa linker and the β2 mCFP contructs have 6-aa linker inserted between the integrin and fluorophore sequences, as described (46). All constructs were verified by sequencing (Research Biolabs).

The 293T cells (American Type Culture Collection) were maintained in DMEM containing 10% (v/v) heat-inactivated FBS (HI-FBS), 100 IU/ml penicillin, and 100 μg/ml streptomycin (HyClone). Cells were transfected with integrin expression plasmids using the transfection reagent Polyfect (Qiagen). K562 cells (American Type Culture Collection) were maintained in RPMI 1640 supplemented with 10% (v/v) HI-FBS with penicillin and streptomycin. K562 cells were electroporated with relevant αL and β2 expression plasmids using a pipette-type microporator MP-100 (NanoEnTek), according to manufacturer’s instructions.

To detect integrin expression on transfectants, cells were incubated with primary mAb IB4 (β2 specific, heterodimer dependent) (20 μg/ml) on ice for 30 min. Cells were washed in medium and incubated in medium containing FITC-conjugated sheep anti-mouse IgG (Sigma-Aldrich) (dilution 1/400) on ice for 30 min. Stained cells were fixed in PBS containing 1% (v/v) formaldehyde, followed by analyses on a FACSCalibur using the software CellQuest (BD Biosciences). To examine I-like domain activation, transfectants bearing wild-type αLβ2 and mutants were stained with mAb m24 with or without Mg/EGTA at 37°C for 30 min, washed, and stained with secondary FITC-conjugated Ab, as described (42).

Cell surface proteins were labeled with biotin, as described (32). The 293T transfectants were washed twice in PBS and incubated in PBS containing 0.5 mg/ml sulfo-NHS-biotin (Pierce) for 30 min at room temperature. Cells were washed in PBS containing 10 mM Tris-HCl (pH 8.0) and 0.1% (w/v) BSA to stop the labeling reaction. Cells were incubated in medium containing 5% (v/v) HI-FBS and 10 mM HEPES with the relevant mAb (2 μg each) at 37°C for 30 min. Unbound mAb was removed by washing the cells twice in medium. Cells were lysed in lysis buffer (10 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1% (v/v) Nonidet P-40) with protease inhibitor mixture (Roche) at 4°C for 30 min. Immunoprecipitation was performed as described (32) using the rabbit anti-mouse IgG (Sigma-Aldrich) coupled to protein A-Sepharose beads (Amersham). Proteins immunoprecipitated were resolved on 7.5% SDS-PAGE under reducing conditions, followed by electroblotting onto Immobilon P membrane (Millipore). Biotinylated protein bands were detected with streptavidin-HRP, followed by ECL detection using the ECL-plus kit (Amersham). For detection of additional N-linked glycosylation on αLβ2 glycan, cell surface integrins were labeled with biotin and immunoprecipitated with mAb MHM23 from lysates of transfectants. Immunoprecipitated proteins were treated with peptide N-glycosidase F (PNGase F) (New England Biolabs), according to manufacturer’s instructions.

Adhesion assays on immobilized ICAMs were performed as described (42). Polysorb microtiter well (Nunc) was coated with 0.5 μg of goat anti-human IgG Fc specific (Sigma-Aldrich) in 50 mM bicarbonate buffer (pH 9.2) overnight at 4°C, followed by blocking the nonspecific binding sites with 0.5% (w/v) BSA in PBS. Each well was coated with 50 ng of ICAM-Fc in PBS at room temperature for 2 h. For the titration experiment using different amount of ICAM-1, each well was coated with 10, 2, or 0.4 ng of ICAM-1. Wells were washed twice with wash buffer (RPMI 1640 containing 5% HI-FBS and 10 mM HEPES (pH 7.4)) before used. Transfectants labeled with 2′7′-bis-(2-carboxyethyl)-5-(and-6) carboxyfluorescein (Molecular Probes) were seeded at ∼2 × 104 cells/well in medium, followed by incubation at 37°C for 30 min in a humidified 5% CO2 incubator. Unbound cells were removed by washing the wells twice in wash buffer. The percentage of adherent cells was determined by measuring the fluorescence signal of adherent cells in a fluorescent plate reader (FL600) (Bio-Tek Instruments). The activating conditions were Mg/EGTA (5 mM MgCl2 and 1.5 mM EGTA) or mAb KIM185 (10 μg/ml), or both. The mAb MHM24 (10 μg/ml) was used as blocking mAb to verify αLβ2-mediated adhesion specificity.

FRET analyses to detect integrin cytoplasmic tail separation were performed by the method of acceptor photobleaching (47) on a Zeiss LSM510 confocal microscope (Carl Zeiss). K562 cells were used for FRET analyses, as described (46). K562 transfected with relevant αL and β2 FRET pairs was cytospun onto poly(l-lysine)-coated glass slides. The laser and emission filter parameters used for photobleaching and mCFP/mYFP detection pre- and postbleaching are: mCFP, excitation wavelength 458 nm and emission filter BP 470–500 nm; mYFP, excitation wavelength 514 nm and emission filter LP 530 nm. The oil immersion ×63 objective lens was used. Photobleaching of mYFP of an entire cell was achieved by scanning the cell 20 times using the 514 argon laser line that was set at the maximum intensity. For acquisition and data analyses, the cell membrane was selected as region of interest. mCFP signals within the region of interest pre- and post-mYFP bleaching were acquired. FRET efficiency (EF) was calculated using the equation EF = (I6 − I5) × 100/I6, where In is the mCFP intensity at the nth time point (47). Bleaching was performed between the fifth and sixth time points. Similar analyses of unbleach cells using the equation CF = (I6 − I5) × 100/I6 were made. The mean noise computed as NF = (I5 − I4) × 100/I5 in which the mCFP signals at the fourth and fifth time points before the bleaching process were close to zero in all cases. All image acquisitions, analyses, and intensity measurements were performed using the software LSM 510 version 2.

The partial bent model of αLβ2 was generated by Modeler8v1 using the structural coordinates of bent αVβ3 (1JV2) (48). The Plexin-Semaphorin-Integrin, hybrid domain, integrin epidermal growth factor (I-EGF)1–3 were not resolved in the αVβ3 structure. The structure of these domains of β2 was obtained from the crystal coordinates 2P26 and 2P28 (49). The coordinates 1LFA (50) were used for the αLβ2 I domain. The model only serves to provide an illustration of possible global αLβ2 conformations. The detail positions of the domains, especially the I domain on the β propeller, and the I-EGFs with reference to the β TD require additional structural information from an intact αLβ2 or I domain-containing integrin. All figures depicting integrin structures were generated by PyMOL (68).

The main objective of this study is to examine the ligand-binding properties of αLβ2 in three conformations, as follows: the bent αLβ2, extended αLβ2 with a closed headpiece, and extended αLβ2 with an open headpiece. To this end, we have generated a panel of αLβ2 mutants described in this study and in subsequent sections. First, we constructed an extended αLβ2 with a closed headpiece by replacing Asp709 of the β2 cytoplasmic tail with an Arg, which disrupts the salt-bridge linking the αL and β2 cytoplasmic tails. The conformation of this mutant αLβ2D709R was verified by assessing its reactivity to two conformational sensitive reporter mAbs MEM148 and KIM127. MEM148 recognizes a masked epitope in the hybrid domain that becomes exposed when the hybrid domain is displaced (30). Displacement of the hybrid domain generates an integrin with an open headpiece (14). KIM127 recognizes an epitope located in the I-EGF2 and it reports αLβ2 extension (40). The 293T transfectants expressing wild-type αLβ2 and αLβ2D709R were surface labeled with biotin, and immunoprecipitations were performed with the reporter mAbs (Fig. 2 A). The mAb MHM23 that reacts with β2 integrin heterodimers was included as a control to assess the level of αLβ2 heterodimer formation and expression. In wild-type αLβ2, MEM148 and KIM127 coprecipitated αL with β2 only when Mg/EGTA, which is an αLβ2-activating agent, was included. Similarly, in αLβ2D709R, MEM148 failed to coprecipitate αL with β2D709R unless activated with Mg/EGTA. By contrast, KIM127 coprecipitated αL with β2D709R even without Mg/EGTA supplement. As reported previously, unassociated β2 subunits were detected with MEM148 and KIM127 (32). These data demonstrate that the salt-bridge-disrupted mutant αLβ2D709R adopts an extended conformation with a closed headpiece.

FIGURE 2.

A high-affinity I domain on an extended αLβ2 promotes ICAM-3 adhesion. A, Immunoprecipitation of cell surface biotin-labeled wild-type αLβ2 and αLβ2D709R with reporter mAbs MEM148 and KIM127 under different conditions. mAb MHM23 (β2 integrins heterodimer specific) was included as a control. ME: Mg/EGTA. B, Immunoprecipitation of αLcchβ2 and αLcchβ2D709R with the aforementioned mAbs. C, Surface expressions of αLcchβ2 and αLcchβ2D709R on transfectants were assessed by mAb IB4 staining (shaded histogram), followed by flow cytometry analyses. Open histogram represents staining with an irrelevant mAb. D, Adhesion of transfectants bearing different αLβ2 mutants to ICAM-1 and ICAM-3 under different activating conditions. K185: KIM185. Adhesion specificity was demonstrated using the function-blocking mAb MHM24. Two independent experiments were performed, and data from a representative experiment are shown. Data points represent means ± SD of triplicates. E, Adhesion of transfectants expressing a high-affinity I domain or not in an extended αLβ2 on different concentrations of ICAM-1. Representative data of two independent experiments are shown. Data points represent means ± SD of triplicates. Cell surface expressions of all mutants were comparable to that of wild-type αLβ2 as reported by mAb IB4 staining (shaded histogram), followed by flow cytometry analyses. Open histogram represents staining with an irrelevant mAb.

FIGURE 2.

A high-affinity I domain on an extended αLβ2 promotes ICAM-3 adhesion. A, Immunoprecipitation of cell surface biotin-labeled wild-type αLβ2 and αLβ2D709R with reporter mAbs MEM148 and KIM127 under different conditions. mAb MHM23 (β2 integrins heterodimer specific) was included as a control. ME: Mg/EGTA. B, Immunoprecipitation of αLcchβ2 and αLcchβ2D709R with the aforementioned mAbs. C, Surface expressions of αLcchβ2 and αLcchβ2D709R on transfectants were assessed by mAb IB4 staining (shaded histogram), followed by flow cytometry analyses. Open histogram represents staining with an irrelevant mAb. D, Adhesion of transfectants bearing different αLβ2 mutants to ICAM-1 and ICAM-3 under different activating conditions. K185: KIM185. Adhesion specificity was demonstrated using the function-blocking mAb MHM24. Two independent experiments were performed, and data from a representative experiment are shown. Data points represent means ± SD of triplicates. E, Adhesion of transfectants expressing a high-affinity I domain or not in an extended αLβ2 on different concentrations of ICAM-1. Representative data of two independent experiments are shown. Data points represent means ± SD of triplicates. Cell surface expressions of all mutants were comparable to that of wild-type αLβ2 as reported by mAb IB4 staining (shaded histogram), followed by flow cytometry analyses. Open histogram represents staining with an irrelevant mAb.

Close modal

We have shown previously that the mutant αLβ2D709R is constitutively activated, and it binds ICAM-1, but not ICAM-3 (31). Together with our present data, it may be conjectured that an extended αLβ2 with a closed headpiece supports ICAM-1, but not ICAM-3 binding. What additional conformational change(s) is required for an extended αLβ2 to bind ICAM-3? To address this, two approaches were taken. The first approach is to assess the conformation of the αL I domain that is required for an extended αLβ2 to bind ICAM-3. The second approach seeks to determine the conformation of the αLβ2 headpiece that is required for ICAM-3 binding, and it will be described in the subsequent sections.

The αL I domain can be engineered into closed or open conformation having different affinities to the ICAMs (37, 51). ICAM-3 binding of an isolated recombinant αL I domain with an engineered disulfide bond (Lys287 and Lys294 mutated into Cys) has also been reported (29), and the conformation of this I domain was assigned as the high-affinity conformation. Previously, we reported that the mutant αLcchβ2 (cch: Cys-Cys lock; high-affinity conformation) had a propensity to bind ICAM-1, but not ICAM-3. Interestingly, it adopts a global conformation indistinguishable from wild-type αLβ2. It is bent with a closed headpiece as determined by KIM127 and MEM148 (32). It may be rationalized that the lack of ICAM-3 binding of mutant αLcchβ2 is attributed to the fact that it is not extended.

In this study, we compared αLcchβ2 with αLcchβ2D709R. Both mutants present a high-affinity I domain, but the former adopts a bent conformation, whereas the latter adopts an extended conformation. This was confirmed by immunoprecipitation analyses using MEM148 and KIM127 (Fig. 2,B). The expression levels of these mutants were also comparable as determined by mAb IB4 staining, followed by flow cytometry analyses (Fig. 2,C). Previously, we reported that wild-type αLβ2-mediated cell adhesion to the ICAMs can be induced by exogenous activating agents such as Mg/EGTA or the β2 integrin-activating mAb KIM185 (30). Adhesion to ICAM-1 requires only one activating agent, either Mg/EGTA or KIM185, but adhesion to ICAM-3 requires both agents (30). In this study, cells expressing αLcchβ2 or αLcchβ2D709R showed similar constitutive ICAM-1 adhesion profiles because both promoted cell adhesion to ICAM-1 without the requirement of Mg/EGTA activation (Fig. 2 D). Addition of Mg/EGTA only marginally increased the levels of adhesion in both samples. Adhesion specificity mediated by these αLβ2 mutants was verified when adhesion was inhibited in the presence of the αLβ2-specific function-blocking mAb MHM24. A different profile was observed when ICAM-3 adhesion assays were performed. In the absence of Mg/EGTA or KIM185 supplement, the cells expressing αLcchβ2 failed to adhere to ICAM-3, but the cells expressing αLcchβ2D709R showed constitutive adhesion to ICAM-3. Adding the activating agents promoted αLcchβ2 cells to adhere to ICAM-3. Thus, we may conclude that the extension of αLβ2 is necessary for ICAM-3 adhesion, but an extended αLβ2 conformation alone is insufficient for ICAM-3 adhesion. This conclusion is drawn from these collective observations. A bent αLβ2 with a high-affinity I domain (αLcchβ2) reported in this study, and an extended αLβ2 without a high-affinity I domain (αLβ2D709R) reported previously (31) failed to promote constitutive ICAM-3 adhesion. However, an extended αLβ2 with a high-affinity I domain (αLcchβ2D709R) binds ICAM-3 constitutively.

It is also interesting to note that the bent αLβ2 with a high-affinity I domain (αLcchβ2) could promote cell adhesion to ICAM-1 to a level comparable to that of an extended αLβ2 with a high-affinity I domain (αLcchβ2D709R) (Fig. 2,D). An extended αLβ2 is important for rolling adhesion of cells on ICAM-1 (52), and the arrest of lymphocytes on endothelium-presenting chemokines involves αLβ2 extension (53). Thus, we check whether there can be a difference in the adhesive properties of αLcchβ2, αLβ2D709R, and αLcchβ2D709R at lower coating concentrations of ICAM-1. Each well was coated with 10, 2, or 0.4 ng of ICAM-1 instead of the 50 ng used previously (Fig. 2 E). Of note, cells expressing αLcchβ2 or αLβ2D709R adhered less well than cells expressing αLcchβ2D709R when 10 ng/well ICAM-1 was used. This difference was accentuated at 2 ng/well ICAM-1. No significant difference between samples was detected when 0.4 ng/well ICAM-1 was used. The expression levels of these mutants were similar to that of wild-type αLβ2 as determined by flow cytometry. These data further differentiate the αLβ2 that is extended with a high-affinity I domain from αLβ2 that is only extended or αLβ2 that has a high-affinity I domain, but bent.

We have shown by the first approach that a high-affinity I domain is required for an extended αLβ2 to bind ICAM-3. This directly assessed the conformation of the ligand-binding domain presented on an extended αLβ2 that is required for ICAM-3 binding. However, it did not provide information relating to the importance of allostery found in other regions of αLβ2. Hence, the second approach was pursued to examine the conformational changes in the αLβ2 headpiece that are required for effective ICAM-3 binding.

In the αLβ2 headpiece, three key domains are considered for ligand binding. The αL I domain directly participates in ligand binding. The β2 I-like domain binds to an intrinsic ligand in the αL I domain, and it regulates the activity of the αL I domain via allostery. Finally, the activity of the β2 I-like domain is linked to the movement of the hybrid domain (54, 55). The primary difference between αLβ2 with closed and open headpieces is the position of the hybrid domains (13). A displaced or swing-out of the hybrid domain away from the integrin α subunit presents an integrin with an open headpiece; conversely, the lack of hybrid domain movement defines an integrin with a closed headpiece (13, 56).

We sought to generate αLβ2 with an open headpiece. It was reported that the introduction of an N-linked glycan near the interface of the I-like domain and the hybrid domain generates integrins αIIbβ3, αVβ3, and α5β1 with open headpieces, and the ligand-binding activities of these integrins were up-regulated (44). Using the same approach, we generated the mutant β2 glycan, which has an N-linked glycosylation site introduced at Asn293 by replacing two neighboring residues Gln295 and Pro296 to Thr and Leu, respectively (Fig. 3,A). The position of Asn293 in the β2 I-like domain is shown (Fig. 3,B). The additional glycosylation in β2 glycan was confirmed by immunoprecipitating αLβ2 glycan, followed by PNGase F treatment (Fig. 3 C). The β2 glycan band migrated slower than wild-type β2 before treatment with PNGase F, but migrated similarly after treatment. Thus, an additional glycosylation was introduced in the β2 glycan.

FIGURE 3.

Introduction of a glycan in the β2 I-like domain promotes constitutive αLβ2-mediated adhesion to ICAM-3. A, Amino acid sequences showing the position where an additional N-linked glycosylation site at Asn293 was introduced in β2 by the substitutions Q295T and P296L. The sequence of β3, which N-linked glycan was introduced at Asn303 (44 ), as reported, was included for comparison. B, Model of β2 I-like domain and hybrid domain to illustrate the additional N-linked glycosylation site at Asn293, and Asn329 in which mutation to Ser, identified from LAD-1, promotes an extended αLβ2 with open headpiece (32 ). The structural coordinates of bent αVβ3 1JV2 (48 ) were used as the template. C, The additional glycan introduced was verified by immunoprecipitation of cell surface-labeled integrins, followed by PNGase F treatment. The β2 glycan protein band migrated slower than that of the wild-type β2 before treatment, but migrated similarly after treatment. D, Immunoprecipitation of cell surface-biotinylated αLβ2 mutants by KIM127 and MEM148 under different conditions. MHM23 was included as a control. E, Adhesion of transfectants bearing wild-type αLβ2 and αLβ2 glycan to ICAMs. ME: Mg/EGTA; K185: KIM185. Adhesion specificity was demonstrated using the function-blocking mAb MHM24. Two independent experiments were performed, and data from a representative experiment are shown. Data points represent means ± SD of triplicates. Expressions of wild-type αLβ2 and αLβ2 glycan were comparable, as determined by mAb IB4 staining, followed by flow cytometry. Open histogram (irrelevant mAb); shaded histogram (IB4).

FIGURE 3.

Introduction of a glycan in the β2 I-like domain promotes constitutive αLβ2-mediated adhesion to ICAM-3. A, Amino acid sequences showing the position where an additional N-linked glycosylation site at Asn293 was introduced in β2 by the substitutions Q295T and P296L. The sequence of β3, which N-linked glycan was introduced at Asn303 (44 ), as reported, was included for comparison. B, Model of β2 I-like domain and hybrid domain to illustrate the additional N-linked glycosylation site at Asn293, and Asn329 in which mutation to Ser, identified from LAD-1, promotes an extended αLβ2 with open headpiece (32 ). The structural coordinates of bent αVβ3 1JV2 (48 ) were used as the template. C, The additional glycan introduced was verified by immunoprecipitation of cell surface-labeled integrins, followed by PNGase F treatment. The β2 glycan protein band migrated slower than that of the wild-type β2 before treatment, but migrated similarly after treatment. D, Immunoprecipitation of cell surface-biotinylated αLβ2 mutants by KIM127 and MEM148 under different conditions. MHM23 was included as a control. E, Adhesion of transfectants bearing wild-type αLβ2 and αLβ2 glycan to ICAMs. ME: Mg/EGTA; K185: KIM185. Adhesion specificity was demonstrated using the function-blocking mAb MHM24. Two independent experiments were performed, and data from a representative experiment are shown. Data points represent means ± SD of triplicates. Expressions of wild-type αLβ2 and αLβ2 glycan were comparable, as determined by mAb IB4 staining, followed by flow cytometry. Open histogram (irrelevant mAb); shaded histogram (IB4).

Close modal

Next, we examined the conformation and the ligand-binding properties of αLβ2 glycan. To assess the conformation of αLβ2 glycan, immunoprecipitations of αLβ2 glycan with reporter mAbs MEM148 and KIM127 were performed (Fig. 3,D). We included a mutant αLβ2N329S in this experiment. This mutant showed positive reactivity with MEM148 and KIM127, which suggests an open headpiece and extended conformation, and it promotes constitutive adhesion to ICAM-3 (32). The position of Asn329 in the β2 I-like domain is shown (Fig. 3 B). Interestingly, in the absence of Mg/EGTA, very low level of αLβ2 glycan was precipitated by MEM148, whereas high level of αLβ2N329S was detected. Addition of Mg/EGTA markedly increased the level of αLβ2 glycan precipitated by MEM148. These data suggest that αLβ2 glycan is extended, but it appears to have a closed headpiece.

The ligand-binding properties of αLβ2 glycan were examined (Fig. 3 E). Cells expressing wild-type αLβ2 did not showed significant adhesion to ICAM-1 and ICAM-3 without Mg/EGTA or KIM185 treatment. In the presence of Mg/EGTA, cells expressing wild-type αLβ2 adhered to ICAM-1, but not ICAM-3. Adhesion to ICAM-3 was detected when Mg/EGTA and KIM185 were included. These are consistent with our previous observations (30). By contrast, cells expressing αLβ2 glycan adhered constitutively to ICAM-1 and ICAM-3 without the requirement of an activating agent. The expression level of αLβ2 glycan was comparable to that of wild-type αLβ2 as determined by mAb IB4 staining and flow cytometry.

We were intrigued by these findings. The mutant αLβ2 glycan clearly exhibits a capacity for ICAM-3 binding without requiring exogenous activation by Mg/EGTA or KIM185. This suggests a high-affinity receptor (30). On the contrary, the immunoprecipitation data point to an extended integrin with a closed headpiece. As discussed previously, αLβ2 extension alone is insufficient for effective ICAM-3 adhesion because a high-affinity I domain is also required. What accounts for this apparent disparity in observations? It is possible that the introduced glycan moiety in the interface between the β2 I-like domain and the hybrid domain could only effect subtle movement of the hybrid domain, which is insufficient to unmask fully the epitope of mAb MEM148 in αLβ2 glycan. However, the displacement of the hybrid domain by the glycan still triggers the activation of the I-like domain, which converts the I domain to a high-affinity conformation. Hence, we tested the hypothesis by examining the activity of the I-like domain in αLβ2 glycan by using the mAb m24 that reports an activated I-like domain (36, 37). All other αLβ2 mutants examined to date were also included for comparison (Fig. 4).

FIGURE 4.

The activation of the I-like domain in the αLβ2 mutants. Transfectants bearing wild-type αLβ2 and mutants were stained with reporter mAb m24 under different conditions, followed by flow cytometry analyses. The mAb IB4 was included to determine the total expression of the integrins. The expression index of m24 (calculated using the equation, expression index = % cells gated-positive × geo-mean fluorescence) was expressed as percentage of total integrin expression index based on IB4 signal. ME: Mg/EGTA. Two independent experiments were performed, and data from a representative experiment are shown.

FIGURE 4.

The activation of the I-like domain in the αLβ2 mutants. Transfectants bearing wild-type αLβ2 and mutants were stained with reporter mAb m24 under different conditions, followed by flow cytometry analyses. The mAb IB4 was included to determine the total expression of the integrins. The expression index of m24 (calculated using the equation, expression index = % cells gated-positive × geo-mean fluorescence) was expressed as percentage of total integrin expression index based on IB4 signal. ME: Mg/EGTA. Two independent experiments were performed, and data from a representative experiment are shown.

Close modal

Transfectants bearing wild-type αLβ2 or mutants were incubated in medium containing m24 in the absence or presence of Mg/EGTA at 37°C, followed by flow cytometry analyses (Fig. 4). Wild-type αLβ2 showed low level of m24 staining unless activated with Mg/EGTA. The same profile was observed for αLcchβ2. The mutants αLβ2D709R and αLcchβ2D709R showed moderate levels of m24 staining, but these could be augmented in the presence of Mg/EGTA. Noteworthily, αLβ2 glycan and αLβ3N329S showed high level of m24 staining even without activating treatment. These data showed that the I-like domains of the mutants αLβ2 glycan and αLβ3N329S were highly activated. It is evident that the lesser degree of hybrid domain movement in αLβ2 glycan as compared with that of αLβ3N329S is nonetheless capable of inducing a high level of I-like domain activation. Coupled with the findings that both mutants exhibit constitutive ICAM-3-binding properties, we may infer the requirement of an open αLβ2 headpiece for ICAM-3 binding.

We have shown that the extension and opening of the αLβ2 headpiece promote ICAM-3 adhesion. This also suggests that an open headpiece αLβ2 conformation induces a high-affinity I domain that binds ICAM-3. Thus, we conjectured that locking the I domains of mutants αLβ2 glycan and αLβ2N329S in low-affinity conformations will attenuate their constitutive ICAM-3 adhesive properties. The low-affinity mutant αLccl (ccl: Cys-Cys lock; low-affinity conformation) was made by substituting Leu289 and Lys294 of the αL I domain with cysteines (37). We validated the blunting effect of this mutation on αLβ2 ligand binding by transfecting cells with αLcclβ2, and examined the adhesion profiles of these cells (Fig. 5). The cells expressing αLcclβ2 showed minimal adhesion to ICAM-1 and ICAM-3 even under conditions that were activating, although expression level of αLcclβ2 was comparable to wild-type αLβ2 as determined by IB4 staining, followed by flow cytometry analyses. This was consistent with the low adhesive property of αLcclβ2 reported (37).

FIGURE 5.

The adhesive properties of αLβ2 with an engineered low-affinity I domain to ICAM-1 and ICAM-3 were assessed. Expression of wild-type αLβ2 and αLcclβ2 on transfectants was comparable, as determined by IB4 staining (shaded histogram), followed by flow cytometry. Irrelevant mAb (open histogram). ICAM adhesion assays were performed, as described in previous figure legends. Two independent experiments were performed, and data from a representative experiment are shown. Data points represent means ± SD of triplicates.

FIGURE 5.

The adhesive properties of αLβ2 with an engineered low-affinity I domain to ICAM-1 and ICAM-3 were assessed. Expression of wild-type αLβ2 and αLcclβ2 on transfectants was comparable, as determined by IB4 staining (shaded histogram), followed by flow cytometry. Irrelevant mAb (open histogram). ICAM adhesion assays were performed, as described in previous figure legends. Two independent experiments were performed, and data from a representative experiment are shown. Data points represent means ± SD of triplicates.

Close modal

Next, we assessed the adhesive properties of αLcclβ2glycan and αLcclβ2N329S to the ICAMs. Cells were transfected with αLβ2 glycan, αLcclβ2glycan, αLβ2N329S, or αLcclβ2N329S, and the expression levels of these mutants were comparable as determined by flow cytometry (Fig. 6,A). Cells expressing αLβ2 glycan adhered constitutively to ICAM-1 and ICAM-3 (Fig. 6,B). By contrast, cells expressing αLcclβ2glycan adhered constitutively to ICAM-1, but the level of adhesion was reduced as compared with αLβ2 glycan. The αLcclβ2glycan transfectants also showed marked diminution of ICAM-3 adhesion. Cells expressing αLβ2N329S adhered constitutively to ICAM-1 and ICAM-3. Cells expressing αLcclβ2N329S showed no significant changes in ICAM-1 adhesion, but ICAM-3 adhesion was abrogated. Both αLcclβ2glycan and αLcclβ2N329S retained activated I-like domains as determined by flow cytometry with m24 staining (Fig. 6,C). The observed effects were not due to conformational changes of the hybrid domain because locking the I domain in a low-affinity conformation did not alter the reactivity of αLβ2glycan and αLβ2N329S with MEM148 (Fig. 6 D). Furthermore, locking the I domain in a low-affinity conformation did not alter the extended conformation of these mutants as revealed by KIM127 immunoprecipitation.

FIGURE 6.

Locking the I domain in a low-affinity conformation abrogates ICAM-3 adhesion mediated by αLβ2 glycan and αLβ2N329S. A, Flow cytometry analyses of integrin expression on transfectants using mAb IB4 (shaded histogram). Irrelevant mAb (open histogram). B, ICAM adhesion assays of transfectants bearing different αLβ2 mutants. ME: Mg/EGTA; K185: KIM185. Adhesion specificity was demonstrated using the function-blocking mAb MHM24. Two independent experiments were performed, and data from a representative experiment are shown. Data points represent means ± SD of triplicates. C, Detection of I-like domain activation in αLβ2 mutants with m24 under different conditions. Representative data from two independent experiments are presented. D, Immunoprecipitation of cell surface biotin-labeled αLβ2 mutants with KIM127 and MEM148 under different conditions.

FIGURE 6.

Locking the I domain in a low-affinity conformation abrogates ICAM-3 adhesion mediated by αLβ2 glycan and αLβ2N329S. A, Flow cytometry analyses of integrin expression on transfectants using mAb IB4 (shaded histogram). Irrelevant mAb (open histogram). B, ICAM adhesion assays of transfectants bearing different αLβ2 mutants. ME: Mg/EGTA; K185: KIM185. Adhesion specificity was demonstrated using the function-blocking mAb MHM24. Two independent experiments were performed, and data from a representative experiment are shown. Data points represent means ± SD of triplicates. C, Detection of I-like domain activation in αLβ2 mutants with m24 under different conditions. Representative data from two independent experiments are presented. D, Immunoprecipitation of cell surface biotin-labeled αLβ2 mutants with KIM127 and MEM148 under different conditions.

Close modal

In activated integrins, the cytoplasmic tails are separated (46). We extended the investigation by examining whether the cytoplasmic tails of mutants αLβ2 glycan and αLβ3N329S are separated using FRET by the acceptor-photobleach method (47). The αL and β2 cytoplasmic tails of wild-type and mutant integrins were fused to mCFP and mYFP, respectively, as described (46). The separation of the αL and β2 cytoplasmic tails will lead to a reduction in FRET signal. K562 cells were transfected with αLβ2D709R, αLβ2glycan, or αLβ2N329S, and subjected to FRET analyses (Fig. 7). The salt-bridge-disrupted mutant αLβ2D709R was included as a control for the separation of cytoplasmic tails. All αLβ2 mutants showed reduction in FRET efficiency when compared with the basal level of wild-type αLβ2 (Fig. 7,A). Representative fluorescence intensity plots of mCFP and mYFP of cells expressing wild-type and mutant αLβ2 before and after mYFP photobleaching are also shown (Fig. 7 B). These data suggest that the C-terminal halves of these αLβ2 mutants underwent conformational changes. Importantly, the activating mutations in the ectodomain of αLβ2 as in αLβ2 glycan and αLβ2N329S could trigger not only extension of these integrins, but also the separation of their cytoplasmic tails. The relay of structural changes from the αLβ2 headpiece, as a result of these mutations, to its cytoplasmic tails corroborates well with the report on integrin bidirectional signaling by its cytoplasmic tails (46). Ligand binding induces integrin signaling, and it was shown that binding of soluble ICAM-1 to αLβ2 in the presence of Mn2+ could induce both extension of αLβ2 and the separation of its cytoplasmic tails (46).

FIGURE 7.

Analyzing the separation of cytoplasmic tails in αLβ2 mutants. Separation of integrin cytoplasmic tails was assessed by acceptor-photobleaching FRET analyses, as described under Materials and Methods. Three independent experiments were performed. A, Data from a representative experiment with calculated percentage of FRET efficiency are shown. B, Fluorescence intensity measurements, using the software LSM 510 version 2, of representative cells pre- and postacceptor photobleach are also shown. Bleach occurred between the 5th and 6th time points.

FIGURE 7.

Analyzing the separation of cytoplasmic tails in αLβ2 mutants. Separation of integrin cytoplasmic tails was assessed by acceptor-photobleaching FRET analyses, as described under Materials and Methods. Three independent experiments were performed. A, Data from a representative experiment with calculated percentage of FRET efficiency are shown. B, Fluorescence intensity measurements, using the software LSM 510 version 2, of representative cells pre- and postacceptor photobleach are also shown. Bleach occurred between the 5th and 6th time points.

Close modal

Once activated by extra- or intracellular signals, integrins undergo marked conformational changes that are necessary for the up-regulation of its ligand-binding affinity (7, 8, 12). Electron microscopy images of αLβ2 and αXβ2 revealed three species of conformers that may represent snapshots of integrin conformational transition (13). These are the bent conformer, extended conformer with a closed headpiece, and extended conformer with an open headpiece. A hallmark of integrin activation is the transition from a bent conformation to an extended conformation. The unbending of an integrin projects the ligand-binding headpiece away from the membrane with its genu serving as the pivot. The αLβ2 on leukocytes unbends when stimulated by endothelium-bound chemokines, which is most likely attributed to the separation of its cytoplasmic tails by the cytosolic protein talin (31, 46, 53). The two species of extended integrin aforementioned differ in the conformation of their headpieces. The closed and open headpieces differ in the angle between the hybrid domain and the I-like domain (14, 57). Collective reports that examined integrins without an I domain suggest affinity difference between an extended integrin with a closed headpiece and one with an open headpiece (14, 44, 58, 59, 60).

Integrin αLβ2 binds to the ICAMs. It is well documented that ICAM-1 and ICAM-3 have different tissue distributions and they have important roles at different stages of leukocyte homeostasis and development, as described above (18, 22, 23, 24, 25, 26, 27, 28). Cell adhesion to ICAM-1 and ICAM-3 requires distinct αLβ2 activation states (30), and the binding affinity of isolated αL I domain to the ICAMs is ICAM-1 > ICAM-2 > ICAM-3 (29). Together, these observations suggest the requirement of selective ICAM-1 and ICAM-3 cell adhesion mediated by different conformational states of αLβ2 in immune processes. In this study, we show that selective binding to these ICAMs can be regulated by distinct changes in the overall conformation of αLβ2. The results obtained from this study are summarized (Fig. 8 A).

FIGURE 8.

Summary and illustration of αLβ2 conformational states. A, Summarized results obtained from this study. Note: the αLcchβ2 and αLβ2D709R exhibit poor adhesion to low coating concentrations of ICAM-1 when compared with αLcchβ2D709R. B, Schematic with cartoons illustrating possible integrin αLβ2 conformational transitions. The adhesive properties to ICAMs are indicated for each possible conformer. ICAM-1∗: the bent conformer I may bind ICAM-1 when it undergoes certain degree of unbending. Cartoons, shaded: αL; unshaded: β2.

FIGURE 8.

Summary and illustration of αLβ2 conformational states. A, Summarized results obtained from this study. Note: the αLcchβ2 and αLβ2D709R exhibit poor adhesion to low coating concentrations of ICAM-1 when compared with αLcchβ2D709R. B, Schematic with cartoons illustrating possible integrin αLβ2 conformational transitions. The adhesive properties to ICAMs are indicated for each possible conformer. ICAM-1∗: the bent conformer I may bind ICAM-1 when it undergoes certain degree of unbending. Cartoons, shaded: αL; unshaded: β2.

Close modal

A bent αLβ2 with a high-affinity I domain (αLcchβ2) could effect ICAM-1, but not ICAM-3 adhesion in the absence of exogenous activating agents, which is consistent with our previous study (32). It is possible that the engineered high-affinity I domain triggers certain degree of αLβ2 extension, but not to the point of a highly extended molecule that could react with the reporter mAb KIM127 (40). Indeed, different degrees of unbending could be extrapolated from the electron microscopy images of bent αXβ2 (13). FRET-based study of αLβ2 and α4β1 provides additional evidence of varied integrin conformers that are not fully extended (11, 61). Furthermore, an engineered bent αLβ2 based on the x-ray coordinates of a bent β2 fragment that comprised the Plexin-Semaphorin-Integrin, hybrid domain, I-EGF-1, and I-EGF-2 showed constitutive adhesion to ICAM-1 as compared with wild-type αLβ2 (49). This difference possibly lies in the angle of the bent, with that of wild-type αLβ2 being more acute than that of the mutant. It is tempting to speculate that if the nonactivated wild-type αLβ2 adopts an obtuse conformation as in the αVβ3 structure, the angle of the β2 genu in the mutant would be wider than the wild-type αLβ2, which could favor certain degree of ICAM-1 adhesion as reported (48, 49), and illustrated in the schematic (Fig. 8 B).

An interesting finding is that the levels of ICAM-1 adhesion mediated by αLβ2 with a high-affinity I domain, but with an overall bent conformation (αLcchβ2) and an extended αLβ2 with wild-type I domain (αLβ2D709R), were comparable. It may seem that there is no significant difference between the two αLβ2 conformers in ICAM-1 binding, but this should be interpreted carefully. In this study, we have used the conventional flat-bottom plate immobilized ligand adhesion assay to discern the conformational status of αLβ2 required for selective ICAM-1 and ICAM-3 binding. This system may not be highly sensitive to report subtle, but important differences between engagement of ICAM-1 by bent and extended αLβ2 because of its long contact period of cells to ligands and its stringent washing steps for cell detachment. Ligand density, contact time, and detachment force are parameters that will determine whether these differences are detected. Indeed, it was shown that K562 cells transfected with wild-type αLβ2 mediated rolling of cells on ICAM-1 in medium containing Ca2+ and Mg2+ in shear flow, although these cells failed to adhere to ICAM-1 in conventional plate adhesion assay (24). In another study, the induction of αLβ2 extension on K562 transfectants by the α/β I-like allosteric antagonist XVA143 increased the number of rolling cells on ICAM-1 when compared with untreated cells in shear flow, suggesting the importance of αLβ2 extension in rolling adhesion (52). It was also shown that the up-regulation of I domain affinity and receptor extension are important events that promote the transition of αLβ2-expressing K562 cells from rolling to firm adhesion on ICAM-1 (62). These data also point to the finer details in the process of leukocyte rolling adhesion on endothelium involving the integrins (24, 52, 53, 62).

In this study, an extended αLβ2 is not effective in mediating ICAM-3 adhesion when the I domain is not of a high-affinity conformation. Salt-bridge disruption induced an extended αLβ2 that allowed ICAM-1 adhesion, but it did not promote ICAM-3 adhesion unless the I domain was made high affinity by an introduced cysteine lock. This was recapitulated in αLβ2 with I-like domain modifications glycan and N329S, which promoted constitutive adhesion to ICAM-3. Adhesion to ICAM-3 was abrogated when the I domain of these mutants was engineered in a low-affinity conformation. Taking into consideration the concept of I domain regulation by the I-like domain that is coupled to hybrid domain movement (8), we hypothesize that shape changes in the αLβ2 headpiece could fine regulate the adhesion of cells to ICAM-1 and ICAM-3. Importantly, this may allow discrimination between different ligands by αLβ2 based on its conformation under different cellular conditions. Indeed, two additional pieces of evidence in this study lend support to the importance of headpiece in defining adhesion to ICAM-1 only or to ICAM-1 and ICAM-3. First, the I-like domain was activated as reported by m24 in the αLβ2N329S and αLβ2 glycan mutants. The level of m24 staining was significantly higher than that of the salt-bridge-disrupted mutant αLβ2D709R that favored ICAM-1 adhesion, but failed to mediate ICAM-3 adhesion in the absence of exogenous activating agents. Second, hybrid domain movement was detected in αLβ2 glycan and αLβ2N329S, although the former showed low reactivity with MEM148. Collectively, these suggest the additional requirement of an open headpiece for an extended αLβ2 to mediate effective adhesion to ICAM-3.

In the previous section, we conjectured that the hybrid domain is displaced to a lesser degree in αLβ2 glycan as compared with αLβ2N329S. This may explain the low, but detectable reactivity of αLβ2 glycan to MEM148. This line of reasoning would also suggest that the overall shape of the headpiece of αLβ2 glycan differs from that of αLβ2N329S. Despite the unavailability of structural data at present, which are interesting follow-on studies to pursue, there is indirect evidence that infers shape differences in the headpieces of these mutants. It is evident that when the I domain was locked in a low-affinity conformation, αLβ2 glycan- or αLβ2N329S-mediated adhesion to ICAM-3 was similarly abrogated. Noteworthily, the low-affinity I domain reduced the level of constitutive ICAM-1 adhesion mediated by αLβ2 glycan, but had minimal effect on that mediated by αLβ2N329S. Similar effect was detected when the m24 staining of these mutants with lock low-affinity I domains was compared. Although we could not exclude the possibility that the introduced glycan in the I-like domain directly triggers shape changes in the I-like domain other than the displacement of the hybrid domain, the adhesion profiles of αLβ2 glycan and αLβ2N329S together with m24 stainings support the requirement of further conformational changes in the headpiece of extended αLβ2 to facilitate ICAM-3 adhesion.

We also propose that the conformations of αLβ2 glycan and αLβ2N329S are different, and may be illustrated by cartoons representing conformer III and IV, respectively (Fig. 8 B). The difference in the angle of movement of the hybrid domain between these conformers would entail different degrees of splaying at their C-terminal halves. The separation of the C-terminal halves was further verified by FRET study, which showed reduction in FRET signal from basal level in these mutants when compared with wild-type αLβ2. From the electron microscopy images of recombinant αLβ2 and αXβ2, each having a C-terminal acidic-basic coiled-coil clasp, conformer III may be similar to the unclasped αXβ2 in the presence of XVA143 that binds to the metal ion-dependent adhesion site of the β2 I-like domain and triggers αLβ2 extension (13, 62). Conformer IV may be similar to that of an unclasped extended open αLβ2 (13). The transition from conformer III to IV may be induced by ligand binding.

Next, the distinction between conformer II and III lies in the headpieces, which could account for the difference in their ICAM-3 adhesion properties. Conformer II, as induced by salt-bridge disruption (αLβ2D709R), mediated ICAM-3 adhesion only when a high-affinity I domain was introduced. Conversely, constitutive adhesion to ICAM-3 of conformer III (αLβ2 glycan) was abrogated by the introduction of a low-affinity I domain. In both cases, constitutive ICAM-1 adhesion was detected regardless of I domain mutations. The m24 reactivity of these mutants also points to a difference in headpiece conformation, although cytoplasmic tail separation was detected in these mutants, as determined by FRET. The αLβ2 with certain degree of unbending as depicted by conformer I could allow ICAM-1 adhesion that is dependent on the ligand density, the duration of the contact time, and the force of detachment aforementioned. The equilibrium is shifted toward conformer II when, for example, the integrin is activated by talin, which triggers its full extension (31, 46, 63). Because the αLβ2 is extended, it allows certain degree of hybrid domain movement, generating two populations of αLβ2 that are illustrated by conformer II and III. When these conformers interact with ligand, further displacement of the hybrid domain ensues, which shifts the equilibrium toward conformer IV. It is also tempting to speculate that these transitions hint at an asymmetric nature of integrin bidirectional signaling.

Finally, cell type-specific differences in the manifestation of integrin adhesiveness have been reported. In a detailed study addressing the role of integrin cytoplasmic tail contribution toward inside-out activation, it was demonstrated that integrin α5β1 in peripheral blood T cells, for example, bound to its ligand fibronectin with a lower affinity when compared with α5β1 in Chinese hamster ovary cells (64). The authors went further to show that the platelet integrin αIIbβ3 with its cytoplasmic tails replaced by the corresponding tails of α5β1 was activated when compared with wild-type αIIbβ3 expressed in Chinese hamster ovary cells. The study made by Sigal et al. (24) also showed stronger adhesion of the T cell line Jurkat and peripheral blood T cells than transfected K562-expressing αLβ2 to ICAM-1. Thus, cytosolic factors that are cell-type specific can influence the adhesive properties of integrins. As examples, the inhibitory Rho family member RhoH that is expressed only in hematopoietic cells is found to maintain the nonadhesive state of αLβ2 (65, 66). RAPL (Nore1B), a molecule that binds the small GTPase Rap1, is preferentially expressed in lymphocytes, and it modulates the adhesive property and membrane distribution of αLβ2 (67). In this regard, it will be appropriate to extend our present findings that made use of primarily 293T cells to, for example, T cells in future work. This would also provide an interesting approach to investigate the intrinsic signaling capabilities of αLβ2 with different conformations for outside-in signaling.

We thank Ardcharaporn Vararattanavech for assistance in performing FRET, and L.S. Kong for assistance in αLβ2 modeling. We thank S.K.A. Law for providing the ICAMs and mAbs, and for helpful comments on the study.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This study was supported by the Singapore A*STAR BMRC Grant 06/1/22/19/445. X.-Y.T. and Y.-F.L. performed the experiments; X.-Y.T. and S.-M.T. analyzed the results and made the figures; and S.-M.T. designed the study and wrote the paper.

3

Abbreviations used in this paper: LAD, leukocyte adhesion deficiency; FRET, fluorescence resonance energy transfer; HI-FBS, heat-inactivated FBS; I-EGF, integrin epidermal growth factor; CFP, cyan fluorescent protein; YFP, yellow fluorescent protein; mCFP, monomeric CFP; mYFP, monomeric YFP; PNGase F, peptide N-glycosidase F.

1
Hynes, R. O..
2002
. Integrins: bidirectional, allosteric signaling machines.
Cell
110
:
673
-687.
2
Hogg, N., P. A. Bates.
2000
. Genetic analysis of integrin function in man: LAD-1 and other syndromes.
Matrix Biol.
19
:
211
-222.
3
Alon, R., A. Etzioni.
2003
. LAD-III, a novel group of leukocyte integrin activation deficiencies.
Trends Immunol.
24
:
561
-566.
4
Kinashi, T., M. Aker, M. Sokolovsky-Eisenberg, V. Grabovsky, C. Tanaka, R. Shamri, S. Feigelson, A. Etzioni, R. Alon.
2004
. LAD-III, a leukocyte adhesion deficiency syndrome associated with defective Rap1 activation and impaired stabilization of integrin bonds.
Blood
103
:
1033
-1036.
5
Pasvolsky, R., S. W. Feigelson, S. S. Kilic, A. J. Simon, G. Tal-Lapidot, V. Grabovsky, J. R. Crittenden, N. Amariglio, M. Safran, A. M. Graybiel, et al
2007
. A LAD-III syndrome is associated with defective expression of the Rap-1 activator CalDAG-GEFI in lymphocytes, neutrophils, and platelets.
J. Exp. Med.
204
:
1571
-1582.
6
Carman, C. V., T. A. Springer.
2003
. Integrin avidity regulation: are changes in affinity and conformation underemphasized?.
Curr. Opin. Cell Biol.
15
:
547
-556.
7
Arnaout, M. A., B. Mahalingam, J. P. Xiong.
2005
. Integrin structure, allostery, and bidirectional signaling.
Annu. Rev. Cell. Dev. Biol.
21
:
381
-410.
8
Luo, B. H., C. V. Carman, T. A. Springer.
2007
. Structural basis of integrin regulation and signaling.
Annu. Rev. Immunol.
25
:
619
-647.
9
Adair, B. D., J. P. Xiong, C. Maddock, S. L. Goodman, M. A. Arnaout, M. Yeager.
2005
. Three-dimensional EM structure of the ectodomain of integrin αVβ3 in a complex with fibronectin.
J. Cell Biol.
168
:
1109
-1118.
10
Chigaev, A., A. M. Blenc, J. V. Braaten, N. Kumaraswamy, C. L. Kepley, R. P. Andrews, J. M. Oliver, B. S. Edwards, E. R. Prossnitz, R. S. Larson, L. A. Sklar.
2001
. Real time analysis of the affinity regulation of α4 integrin: the physiologically activated receptor is intermediate in affinity between resting and Mn2+ or antibody activation.
J. Biol. Chem.
276
:
48670
-48678.
11
Chigaev, A., A. Waller, G. J. Zwartz, T. Buranda, L. A. Sklar.
2007
. Regulation of cell adhesion by affinity and conformational unbending of α4β1 integrin.
J. Immunol.
178
:
6828
-6839.
12
Arnaout, M. A., S. L. Goodman, J. P. Xiong.
2007
. Structure and mechanics of integrin-based cell adhesion.
Curr. Opin. Cell Biol.
19
:
495
-507.
13
Nishida, N., C. Xie, M. Shimaoka, Y. Cheng, T. Walz, T. A. Springer.
2006
. Activation of leukocyte β2 integrins by conversion from bent to extended conformations.
Immunity
25
:
583
-594.
14
Xiao, T., J. Takagi, B. S. Coller, J. H. Wang, T. A. Springer.
2004
. Structural basis for allostery in integrins and binding to fibrinogen-mimetic therapeutics.
Nature
432
:
59
-67.
15
Kinashi, T..
2007
. Integrin regulation of lymphocyte trafficking: lessons from structural and signaling studies.
Adv. Immunol.
93
:
185
-227.
16
Staunton, D. E., M. L. Dustin, H. P. Erickson, T. A. Springer.
1990
. The arrangement of the immunoglobulin-like domains of ICAM-1 and the binding sites for LFA-1 and rhinovirus.
Cell
61
:
243
-254.
17
De Fougerolles, A. R., S. A. Stacker, R. Schwarting, T. A. Springer.
1991
. Characterization of ICAM-2 and evidence for a third counter-receptor for LFA-1.
J. Exp. Med.
174
:
253
-267.
18
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
-190.
19
Bailly, P., E. Tontti, P. Hermand, J. P. Cartron, C. G. Gahmberg.
1995
. The red cell LW blood group protein is an intercellular adhesion molecule which binds to CD11/CD18 leukocyte integrins.
Eur. J. Immunol.
25
:
3316
-3320.
20
Mizuno, T., Y. Yoshihara, J. Inazawa, H. Kagamiyama, K. Mori.
1997
. cDNA cloning and chromosomal localization of the human telencephalin and its distinctive interaction with lymphocyte function-associated antigen-1.
J. Biol. Chem.
272
:
1156
-1163.
21
Ostermann, G., K. S. Weber, A. Zernecke, A. Schroder, C. Weber.
2002
. JAM-1 is a ligand of the β2 integrin LFA-1 involved in transendothelial migration of leukocytes.
Nat. Immunol.
3
:
151
-158.
22
Dustin, M. L., R. Rothlein, A. K. Bhan, C. A. Dinarello, T. A. Springer.
1986
. Induction by IL 1 and interferon-γ: tissue distribution, biochemistry, and function of a natural adherence molecule (ICAM-1).
J. Immunol.
137
:
245
-254.
23
Acevedo, A., M. A. del Pozo, A. G. Arroyo, P. Sanchez-Mateos, R. Gonzalez-Amaro, F. Sanchez-Madrid.
1993
. Distribution of ICAM-3-bearing cells in normal human tissues: expression of a novel counter-receptor for LFA-1 in epidermal Langerhans cells.
Am. J. Pathol.
143
:
774
-783.
24
Sigal, A., D. A. Bleijs, V. Grabovsky, S. J. van Vliet, O. Dwir, C. G. Figdor, Y. van Kooyk, R. Alon.
2000
. The LFA-1 integrin supports rolling adhesions on ICAM-1 under physiological shear flow in a permissive cellular environment.
J. Immunol.
165
:
442
-452.
25
Grakoui, A., S. K. Bromley, C. Sumen, M. M. Davis, A. S. Shaw, P. M. Allen, M. L. Dustin.
1999
. The immunological synapse: a molecular machine controlling T cell activation.
Science
285
:
221
-227.
26
Montoya, M. C., D. Sancho, G. Bonello, Y. Collette, C. Langlet, H. T. He, P. Aparicio, A. Alcover, D. Olive, F. Sanchez-Madrid.
2002
. Role of ICAM-3 in the initial interaction of T lymphocytes and APCs.
Nat. Immunol.
3
:
159
-168.
27
Campanero, M. R., M. A. del Pozo, 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
-1016.
28
Perez, O. D., D. Mitchell, G. P. Nolan.
2007
. Differential role of ICAM ligands in determination of human memory T cell differentiation.
BMC Immunol.
8
:
2
29
Shimaoka, M., C. Lu, R. T. Palframan, U. H. von Andrian, A. McCormack, J. Takagi, T. A. Springer.
2001
. Reversibly locking a protein fold in an active conformation with a disulfide bond: integrin αL I domains with high affinity and antagonist activity in vivo.
Proc. Natl. Acad. Sci. USA
98
:
6009
-6014.
30
Tang, R. H., E. Tng, S. K. Law, S. M. Tan.
2005
. Epitope mapping of monoclonal antibody to integrin αLβ2 hybrid domain suggests different requirement of affinity states for intercellular adhesion molecules (ICAM)-1 and ICAM-3 binding.
J. Biol. Chem.
280
:
29208
-29216.
31
Li, Y. F., R. H. Tang, K. J. Puan, S. K. Law, S. M. Tan.
2007
. The cytosolic protein talin induces an intermediate affinity integrin αLβ2.
J. Biol. Chem.
282
:
24310
-24319.
32
Cheng, M., S. Y. Foo, M. L. Shi, R. H. Tang, L. S. Kong, S. K. Law, S. M. Tan.
2007
. Mutation of a conserved asparagine in the I-like domain promotes constitutively active integrins αLβ2 and αIIbβ3.
J. Biol. Chem.
282
:
18225
-18232.
33
Hildreth, J. E., F. M. Gotch, P. D. Hildreth, A. J. McMichael.
1983
. A human lymphocyte-associated antigen involved in cell-mediated lympholysis.
Eur. J. Immunol.
13
:
202
-208.
34
Hildreth, J. E., J. T. August.
1985
. The human lymphocyte function-associated (HLFA) antigen and a related macrophage differentiation antigen (HMac-1): functional effects of subunit-specific monoclonal antibodies.
J. Immunol.
134
:
3272
-3280.
35
Poloni, F., P. Puddu, F. Moretti, M. Flego, G. Romagnoli, M. Tombesi, I. Capone, A. Chersi, F. Felici, M. Cianfriglia.
2001
. Identification of a LFA-1 region involved in the HIV-1-induced syncytia formation through phage-display technology.
Eur. J. Immunol.
31
:
57
-63.
36
Dransfield, I., C. Cabanas, A. Craig, N. Hogg.
1992
. Divalent cation regulation of the function of the leukocyte integrin LFA-1.
J. Cell Biol.
116
:
219
-226.
37
Lu, C., M. Shimaoka, Q. Zang, J. Takagi, T. A. Springer.
2001
. Locking in alternate conformations of the integrin αLβ2 I domain with disulfide bonds reveals functional relationships among integrin domains.
Proc. Natl. Acad. Sci. USA
98
:
2393
-2398.
38
Robinson, M. K., D. Andrew, H. Rosen, D. Brown, S. Ortlepp, P. Stephens, E. C. Butcher.
1992
. Antibody against the Leu-CAM β chain (CD18) promotes both LFA-1- and CR3-dependent adhesion events.
J. Immunol.
148
:
1080
-1085.
39
Stephens, P., J. T. Romer, M. Spitali, A. Shock, S. Ortlepp, C. G. Figdor, M. K. Robinson.
1995
. KIM127, an antibody that promotes adhesion, maps to a region of CD18 that includes cysteine-rich repeats.
Cell Adhes. Commun.
3
:
375
-384.
40
Beglova, N., S. C. Blacklow, J. Takagi, T. A. Springer.
2002
. Cysteine-rich module structure reveals a fulcrum for integrin rearrangement upon activation.
Nat. Struct. Biol.
9
:
282
-287.
41
Wright, S. D., P. E. Rao, W. C. Van Voorhis, L. S. Craigmyle, K. Iida, M. A. Talle, E. F. Westberg, G. Goldstein, S. C. Silverstein.
1983
. Identification of the C3bi receptor of human monocytes and macrophages by using monoclonal antibodies.
Proc. Natl. Acad. Sci. USA
80
:
5699
-5703.
42
Tan, S. M., R. H. Hyland, A. Al-Shamkhani, W. A. Douglass, J. M. Shaw, S. K. Law.
2000
. Effect of integrin β2 subunit truncations on LFA-1 (CD11a/CD18) and Mac-1 (CD11b/CD18) assembly, surface expression, and function.
J. Immunol.
165
:
2574
-2581.
43
Barclay, A. N., M. H. Brown, S. K. Law, A. J. McKnight, M. G. Tomlinson, P. A. van der Merwe.
1997
.
The Leucocyte Antigen Facts Book
Academic Press, London.
44
Luo, B. H., T. A. Springer, J. Takagi.
2003
. Stabilizing the open conformation of the integrin headpiece with a glycan wedge increases affinity for ligand.
Proc. Natl. Acad. Sci. USA
100
:
2403
-2408.
45
Zacharias, D. A., J. D. Violin, A. C. Newton, R. Y. Tsien.
2002
. Partitioning of lipid-modified monomeric GFPs into membrane microdomains of live cells.
Science
296
:
913
-916.
46
Kim, M., C. V. Carman, T. A. Springer.
2003
. Bidirectional transmembrane signaling by cytoplasmic domain separation in integrins.
Science
301
:
1720
-1725.
47
Karpova, T. S., C. T. Baumann, L. He, X. Wu, A. Grammer, P. Lipsky, G. L. Hager, J. G. McNally.
2003
. Fluorescence resonance energy transfer from cyan to yellow fluorescent protein detected by acceptor photobleaching using confocal microscopy and a single laser.
J. Microsc.
209
:
56
-70.
48
Xiong, J. P., T. Stehle, B. Diefenbach, R. Zhang, R. Dunker, D. L. Scott, A. Joachimiak, S. L. Goodman, M. A. Arnaout.
2001
. Crystal structure of the extracellular segment of integrin αVβ3.
Science
294
:
339
-345.
49
Shi, M., S. Y. Foo, S. M. Tan, E. P. Mitchell, S. K. Law, J. Lescar.
2007
. A structural hypothesis for the transition between bent and extended conformations of the leukocyte β2 integrins.
J. Biol. Chem.
282
:
30198
-30206.
50
Qu, A., D. J. Leahy.
1995
. Crystal structure of the I-domain from the CD11a/CD18 (LFA-1, αLβ2) integrin.
Proc. Natl. Acad. Sci. USA
92
:
10277
-10281.
51
Lu, C., M. Shimaoka, M. Ferzly, C. Oxvig, J. Takagi, T. A. Springer.
2001
. An isolated, surface-expressed I domain of the integrin αLβ2 is sufficient for strong adhesive function when locked in the open conformation with a disulfide bond.
Proc. Natl. Acad. Sci. USA
98
:
2387
-2392.
52
Salas, A., M. Shimaoka, A. N. Kogan, C. Harwood, U. H. von Andrian, T. A. Springer.
2004
. Rolling adhesion through an extended conformation of integrin αLβ2 and relation to α I and β I-like domain interaction.
Immunity
20
:
393
-406.
53
Shamri, R., V. Grabovsky, J. M. Gauguet, S. Feigelson, E. Manevich, W. Kolanus, M. K. Robinson, D. E. Staunton, U. H. von Andrian, R. Alon.
2005
. Lymphocyte arrest requires instantaneous induction of an extended LFA-1 conformation mediated by endothelium-bound chemokines.
Nat. Immunol.
6
:
497
-506.
54
Takagi, J., T. A. Springer.
2002
. Integrin activation and structural rearrangement.
Immunol. Rev.
186
:
141
-163.
55
Yang, W., M. Shimaoka, A. Salas, J. Takagi, T. A. Springer.
2004
. Intersubunit signal transmission in integrins by a receptor-like interaction with a pull spring.
Proc. Natl. Acad. Sci. USA
101
:
2906
-2911.
56
Mould, A. P., S. J. Barton, J. A. Askari, P. A. McEwan, P. A. Buckley, S. E. Craig, M. J. Humphries.
2003
. Conformational changes in the integrin β A domain provide a mechanism for signal transduction via hybrid domain movement.
J. Biol. Chem.
278
:
17028
-17035.
57
Takagi, J., K. Strokovich, T. A. Springer, T. Walz.
2003
. Structure of integrin α5β1 in complex with fibronectin.
EMBO J.
22
:
4607
-4615.
58
Takagi, J., B. M. Petre, T. Walz, T. A. Springer.
2002
. Global conformational rearrangements in integrin extracellular domains in outside-in and inside-out signaling.
Cell
110
:
599
-511.
59
Mould, A. P., E. J. Symonds, P. A. Buckley, J. G. Grossmann, P. A. McEwan, S. J. Barton, J. A. Askari, S. E. Craig, J. Bella, M. J. Humphries.
2003
. Structure of an integrin-ligand complex deduced from solution x-ray scattering and site-directed mutagenesis.
J. Biol. Chem.
278
:
39993
-39999.
60
Luo, B. H., K. Strokovich, T. Walz, T. A. Springer, J. Takagi.
2004
. Allosteric β1 integrin antibodies that stabilize the low affinity state by preventing the swing-out of the hybrid domain.
J. Biol. Chem.
279
:
27466
-27471.
61
Larson, R. S., T. Davis, C. Bologa, G. Semenuk, S. Vijayan, Y. Li, T. Oprea, A. Chigaev, T. Buranda, C. R. Wagner, L. A. Sklar.
2005
. Dissociation of I domain and global conformational changes in LFA-1: refinement of small molecule-I domain structure-activity relationships.
Biochemistry
44
:
4322
-4331.
62
Salas, A., M. Shimaoka, U. Phan, M. Kim, T. A. Springer.
2006
. Transition from rolling to firm adhesion can be mimicked by extension of integrin αLβ2 in an intermediate affinity state.
J. Biol. Chem.
281
:
10876
-10882.
63
Ginsberg, M. H., A. Partridge, S. J. Shattil.
2005
. Integrin regulation.
Curr. Opin. Cell Biol.
17
:
509
-516.
64
O’Toole, T. E., Y. Katagiri, R. J. Faull, K. Peter, R. Tamura, V. Quaranta, J. C. Loftus, S. J. Shattil, M. H. Ginsberg.
1994
. Integrin cytoplasmic domains mediate inside-out signal transduction.
J. Cell Biol.
124
:
1047
-1059.
65
Li, X., X. Bu, B. Lu, H. Avraham, R. A. Flavell, B. Lim.
2002
. The hematopoiesis-specific GTP-binding protein RhoH is GTPase deficient and modulates activities of other Rho GTPases by an inhibitory function.
Mol. Cell. Biol.
22
:
1158
-1171.
66
Cherry, L. K., X. Li, P. Schwab, B. Lim, L. B. Klickstein.
2004
. RhoH is required to maintain the integrin LFA-1 in a nonadhesive state on lymphocytes.
Nat. Immunol.
5
:
961
-967.
67
Katagiri, K., A. Maeda, M. Shimonaka, T. Kinashi.
2003
. RAPL, a Rap1-binding molecule that mediates Rap1-induced adhesion through spatial regulation of LFA-1.
Nat. Immunol.
4
:
741
-748.
68
Delano, W. L. 2002. http://pymol.sourceforgenet/.