| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 279, Issue 26, 27466-27471, June 25, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 Integrin Antibodies That Stabilize the Low Affinity State by Preventing the Swing-out of the Hybrid Domain*
From the
The CBR Institute for Biomedical Research and Department of Pathology and the Department of Cell Biology, Harvard Medical School, Boston, Massachusetts 02115 and the ¶Institute for Protein Research, Laboratory of Protein Synthesis and Expression, Osaka University, 3-2 Yamadaoka, Suita, Osaka 565-0871, Japan
Received for publication, April 20, 2004 , and in revised form, April 27, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
1 antibody, SG/19, that had been reported to inhibit the function of the 1 integrin on the cell surface. SG/19 recognized the wild type 1 subunit that exists in a conformational equilibrium between the high and low affinity states but bound poorly to a mutant 1 integrin that had been locked in a high affinity state. Epitope mapping of SG/19 revealed that Thr82 in the 1 subunit, located at the outer face of the boundary between the I-like and hybrid domains, was the key binding determinant for this antibody. Direct visualization of the 
51 headpiece fragment in complex with SG/19 Fab with electron microscopy confirmed the location of the binding surface and showed that the ligand binding site is not occluded by the bound Fab. Surface plasmon resonance showed that 51 integrin bound by SG/19 maintained a low affinity toward its physiological ligand fibronectin (Fn) whereas binding by function-blocking anti-5 antibodies resulted in a complete loss of fibronectin binding. Thus a class of the anti- antibodies represented by SG/19 attenuate the ligand binding function by restricting the conformational shift to the high affinity state involving the swing-out of the hybrid domain without directly interfering with ligand docking. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Monoclonal antibodies have been instrumental to studies on the structure and function of integrins (9). Integrin mAbs can be divided into three classes: (i) "inhibitory mAbs" that perturb biological function i.e. ligand binding, (ii) "stimulatory mAbs" that augment ligand binding, and (iii) "neutral mAbs" that do not have any impact on activity. Although some inhibitory mAbs directly recognize the ligand binding site on the integrin molecule and act as ligand-mimetic competitive inhibitors, another class of inhibitory mAbs bind outside the ligand binding site and seem to inhibit integrin function allosterically (10, 11). It is assumed that this class of mAbs exert their effect by stabilizing the unoccupied state of the receptor or by preventing a conformational change necessary for ligand occupancy, in a manner opposite from that of stimulating mAbs (12). However, the mechanism by which these mAbs affect the conformational change in integrins has not been determined nor has the effect of these mAbs on the ligand binding kinetics of integrins been investigated. Here we present direct evidence that a class of allosteric anti-1 mAbs, represented by SG/19, binds distal to the ligand binding site across the interface between the I-like and hybrid domains of the 1 subunit and prevents interconversion of the receptor conformation by blocking the hybrid domain swing-out. The restriction imposed by mAb binding maintains the integrin in a low, but not zero, affinity state.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
1 mAbs SG/19 (IgG1) and SG/7 (IgA) were kindly provided by Dr. K. Miyake (13). Murine anti-human 1 mAbs TS2/16 (14), K20 (15), rat anti-human 5 mAbs 16 (16), 11 (17), and rat anti-human 1 mAb 13 (16) were obtained from the Fifth International Leukocyte Workshop (18). Murine anti-1 mAbs JB1 (19), 12G10 (20), HUTS-4 (21), and 5 mAb P1D6 (22) were purchased from Chemicon International Inc. (Temecula, CA). The Fab fragment of SG/19 IgG was prepared as described previously (23).
Transfection and mAb Binding—Plasmids coding for full-length wild type or G429N mutant human 1 were subcloned into pEF1/V5-HisA and introduced into the CHO-K1 cells using calcium phosphate precipitation. After selecting the transfected cells in a medium containing 1 mg/ml Geneticin G418 for 1 week, the surviving cells were sorted by fluorescence-activated cell sorting to obtain cell lines expressing the desired level of human 1 integrins. For epitope mapping, CHO-K1 cells transfected with a human 1 chain carrying single point mutation were cultured in a medium containing 1 mg/ml Geneticin G418 for 2 weeks, and the surviving cell population was tested for reactivity against various mAbs without further cloning. Bindings of mAbs were evaluated by fluorescence-activated cell sorting analysis using either anti-mouse or anti-rat IgG conjugated with fluorescein isothiocyanate. Fibronectin binding was evaluated by incubating human 1-expressing cells with 50 µg/ml bovine plasma fibronectin labeled with fluorescein isothiocyanate in the presence or absence of activating mAb TS2/16 (10 µg/ml) at room temperature for 30 min followed by flow cytometric analysis.
Negative Stain EM of 51 Head Fragment—A soluble 51 head-piece fragment containing 5 residues 1–623 and 1 residues 1–445 was produced and purified as described previously (24). The construct contained a disulfide-bonded C-terminal "clasp," which was removed specifically by tobacco etch virus protease treatment prior to the analysis. Unclasped 51 fragment was incubated with a saturating concentration of SG/19 Fab fragment or Fn7–10 fragment and purified on a Superdex 200 HR column equilibrated with 50 mM Tris HCl, pH 7.5, 150 mM NaCl, containing 1 mM MnCl2. The peak fraction was adsorbed to glow-discharged carbon-coated copper grids, stained with uranyl formate, and inspected with an FEI Tecnai 12 electron microscope operated at 120 kV. Image processing was performed using the SPIDER image processing package (25) as described previously (24).
Surface Plasmon Resonance (SPR) Binding Assay—Interaction between the recombinant full-length 51 and recombinant fibronectin fragment was assessed by SPR analysis with a BIAcore 3000 (Biacore AB) as described previously (26). Briefly, 1.7 µg of 51 was incubated with 4.2 µg of purified IgG for >30 min in Tris-buffered saline containing 1 mM MnCl2, diluted to give a final concentration of 20–40 nM, and flowed at 20 µl/min at 25 °C over a streptavidin-coated sensor chip immobilized with biotinylated Fn7–10 at a density of 
500 resonance units. All measurements were base line-corrected by subtracting the sensorgram obtained with a control streptavidin surface, and kinetic parameters were determined by fitting the data to a 1:1 Langmuir binding model using BIAevaluation software version 3.0.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
1 Subunit—SG/19 was originally identified as a mouse mAb that blocked binding of the Nalm-6 human pre-B cell line to the murine stromal cell line BMS2 (13). It strongly inhibits 1 integrin-mediated cell adhesion at 10 µg/ml (13). Whereas most function-blocking mAb to 1 bind to the I-like domain (9, 27), SG/19 maps to a more N-terminal region in the 1 subunit (28). Within this N-terminal region, only 4 residues in the PSI domain and 7 residues in the hybrid domain differ between human and mouse 1. Although these domains are not known to be implicated in the ligand binding function, inspection of a homology model of 51 reveals that 2 of these 11 species-specific residues (i.e. Thr82 and Lys87) are positioned very close to the ligand binding I-like domain (Fig. 1) and thus are likely candidates for the SG/19 epitope. These residues were individually mutated to the corresponding murine residue, and the reactivity of SG/19 was evaluated. As shown in Table I, the binding of SG/19 to the 1 chain was critically dependent on Thr82, whereas the mutation at Lys87 did not affect the binding. Conversely, the binding of another function-blocking mAb, SG/7, which had been reported to recognize a Ca2+-dependent epitope (29) was totally abolished by the mutation of Lys87 and only partially by the mutation of Thr82. Mutation of species-specific residues in the I-like domain, i.e. N207D, K208R, V211F, K218Q, M287V, and A342G, did not affect the binding of either mAb.
|
|
1 mAb 13 has been characterized extensively and reported to block human 1 integrin function by an allosteric mechanism (11). The epitope for mAb 13 had been mapped to residues 207–218 in the 1 I-like domain (27). Our results revealed that mAb 13 recognizes a combinatorial epitope made by Asn207, Lys208, and Val211 with Asn207 contributing the most (Table I). Next we explored whether SG/19 and mAb 13 compete with each other for binding to 1. Cy3-labeled SG/19 was competed out by mAb 13 for binding to wild type human 1 (Table II). Reciprocally, the binding of mAb 13 was competed by mAb SG/19. Reciprocal competition was also observed between SG/19 and SG/7 mAbs (data not shown) and between mAb SG/7 and mAb 13 (13). In contrast, SG/19 did not compete with two other mAbs, TS2/16 and 4B4, mapped to residues 207–218 (data not shown) indicating that the binding site for mAb 13 is slightly different from that for TS2/16 and 4B4. In the three-dimensional model, Asn207, Thr82, and Lys87 are all located in close proximity (Fig. 1B). Because the antigen-binding footprints for the three inhibitory mAbs 13, SG/19, and SG/7 are overlapping, they are likely to have similar impacts on the conformation of the junction of the 1 I-like and hybrid domains to which they bind.
|
1 Integrin Locked in a High Affinity State—Stimulatory mAbs, upon binding to integrins, shift the conformational equilibrium toward the active conformer. Therefore, stimulatory mAbs generally show increased binding toward activated receptors (9, 23, 30). Allosteric inhibitory mAbs that work in the opposite way are expected to show reduced binding to the active conformer. We have successfully engineered a mutant 3 integrin constitutively in a high affinity state by introducing an N-glycosylation site, i.e. a glycan wedge, at the inner side of the bend between the I-like and hybrid domains (31). A previously designed 1 glycan wedge mutant, 1-P333N, has a high affinity for ligands (31), although it is less well expressed than wild type and only partially expresses activation epitopes. Therefore, another mutant, 1-G429N, was designed to introduce an N-glycosylation site at Asn429 to wedge open the I-like hybrid domain interface (Fig. 1C). This mutant shows maximal ligand binding activity in the absence of any activating agents (Fig. 2A). This activated 1 mutant assumes the high affinity conformation in the absence of a bound ligand, as shown by the markedly increased binding of the stimulatory mAbs 12G10 and HUTS-4 compared with binding to the wild type receptor (Fig. 2B). Conversely, binding of the inhibitory mAbs 13 and SG/19 was decreased more than 4-fold to the 1-G429N mutant compared with wild type (Fig. 2B). SG/7 also reacted much less well (70% reduction from wild type) to the wedge mutant (data not shown). A reduction of mAb 13 binding and augmentation of mAb HUTS-4 by an activating mutation in the 1 I-like domain 7-helix has been reported (32). These effects of the L358A mutation are much less pronounced (50% reduction and 50% increase in binding of mAbs 13 and HUTS-4, respectively) than the G429N mutation. As Gly429 is located far from the SG/19 epitope at Thr82 and the mAb 13 epitope at Asn207 (Fig. 1), it is unlikely that mutation of Gly429 directly affects binding to these epitopes. Taken together, the data suggest strongly that the inhibitory mAbs 13 and SG/19 prefer the resting (i.e. low affinity) conformer and stabilize this conformation upon binding thereby allosterically modulating the ligand binding activity of 1 integrin.
|
51 Integrin Head-piece—We next sought to clarify how SG/19 could stabilize a particular conformation of an integrin. In a recent study (24), we utilized negative stain EM and image analysis to visualize a headpiece fragment of 51 integrin in the absence and presence of a bound ligand. In the absence of a bound ligand, the 51 headpiece assumes the "closed" conformation with the 1 hybrid domain close to the subunit (Fig. 3A and Ref. 24). Binding to a fragment of the ligand fibronectin changes the conformation dramatically, with an outward swing of the hybrid domain away from the subunit to assume the "open" conformation (Fig. 3B and Ref. 24). The complex of the 51 headpiece and SG/19 Fab reveals a clear density for the Fab bound to the side of the 1 subunit (Fig. 3C). The Fab fragment can be easily distinguished by its typical oval shape with a "dimple" corresponding to the thin central part between the VL/VH and CL/CH1 lobes. The very uniform orientation of SG/19 Fab relative to the rest of the particle makes it possible to define the binding surface. The antigen binding site of the SG/19 Fab docks to the boundary between the I-like and hybrid domains of the 1 subunit at the "side" away from the 5 subunit. This is in perfect agreement with the result of the epitope mapping (see Fig. 1). Importantly, the ligand binding interface located on the top of the receptor is completely exposed and not occluded by the bound Fab (Fig. 3, compare B with C). The images show that the binding site of SG/19 centers on the junction between the I-like and hybrid domains, very close to the location of the key antigenic residue Thr82, which is at the domain boundary (see Fig. 1). The binding site of SG/19 clearly includes significant portions of both the I-like and hybrid domains (Fig. 3C). This binding site is significantly rear-ranged when the hybrid domain swings out 70° upon binding fibronectin (Fig. 3B) so that it is unlikely that SG/19 could bind to it. These findings suggest that SG/19 prevents the outward swing of the hybrid domain, trapping the integrin in a low affinity conformation.
|
51 Bound by Inhibitory Anti-1 mAbs Has Low but Detectable Affinity for Ligands—The ligand binding characteristics of 51 integrin bound by inhibitory mAbs was measured using SPR to determine whether these mAbs abolished ligand binding, or maintained 51 in a low affinity state. To ensure saturation, purified recombinant 51 integrin was preincubated with purified mAbs in large excess. Incubation with 1.8 µM SG/19 Fab for 10 min resulted in saturation binding of the 51 headpiece with no dissociation detected during 20 min of gel filtration (Fig. 3D). Saturation was also observed with 100 nM SG/19 Fab (not shown). Reaction mixtures containing 140 nM 51 headpiece and 560 nM mAb (i.e. 1120 nM of Fab equivalents) were diluted up to 7-fold (thus maintaining the concentration of uncomplexed Fab equivalents at >140 nM) and directly infused onto sensor chips coated with fibronectin fragment in a buffer containing 1 mM Mn2+. 51 complexed with the non-blocking, non-activating 5 mAb 11 showed an elevated signal compared with the uncomplexed 51 because of the increased mass of the analyte (Fig. 4, compare tracing b with c). Although mAb 11 binding slightly changed the binding kinetics by decreasing the dissociation rate, this seems to be a general effect of mAb binding, because another non-blocking, non-activating 1 mAb JB1 (Fig. 4a) and an anti-coiled coil tag mAb 2H11(not shown) had a similar effect. The inhibitory 5 mAbs 16 (Fig. 4g) and P1D6 (Fig. 4f) completely abolished the binding of 51 headpiece to fibronectin. In contrast, 51 complexed with mAb 13 (Fig. 4d) or SG/19 (Fig. 4e) showed decreased but significant binding to the ligand. Increasing the mAb concentration 2-fold did not alter the binding (not shown). Together with the saturation binding experiments described above, the known KD of 0.34 nM of mAb 13 for 51 (11), and the high concentrations of mAb used in SPR, these results demonstrate that the incomplete inhibition is not because of incomplete saturation of the binding sites by mAbs. Therefore, mAbs SG/19 and 13 decrease the affinity of 51 for fibronectin. The kinetic constants (Table III) show that SG/19 binding affected both association and dissociation rate constants resulting in an overall 30-fold drop in the KD value.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
1 inhibitory mAb SG/19 exerts its effect allosterically. The inhibitory mAbs 13 and SG/7 show similar binding and inhibition characteristics and have an overlapping binding footprint, suggesting that the allosteric mode of inhibition is shared among these three mAbs. In fact, mAb 13 has been reported to be an allosteric inhibitor for ligand binding by 51 (11). This notion was based on the fact that mAb 13 binding to 1 was attenuated by ligand, and the concentration of ligand required for half-maximal inhibition was independent of the mAb concentration. Not only occupation by ligand but also mutations that lock the 1 integrin in a high affinity state perturb mAb 13 binding (Fig. 2B and Ref. 32), supporting the indirect mode of inhibition by this class of mAbs. EM images of the integrin-SG/19 Fab complex clearly show that the mAb docks at the outer hinge connecting the I-like domain and the hybrid domain. This would prevent the outward swing of the hybrid domain and fix the closed conformation of the headpiece that represents the low affinity state (5). Recently, we (33) and others (32) have reported that shape shifting in the 7 helix region of the I-like domain upregulates ligand affinity. In the context of the whole molecule, this shape shifting causes an outward swing of the hybrid domain (3, 5, 26). Thus inhibitory mAbs represented by mAb 13 and SG/19 exert their effect by disabling the conformational transition of I-like domain to the high affinity state. This is exactly the reverse of the mechanism by which the glycan-wedge mutation activates integrins (31). It is not clear whether all anti-1 inhibitory mAbs work with a similar indirect mechanism. The inhibitory mAb 4B4 maps to Lys218 of the 1 I-like domain and does not compete with mAb 13, SG/19, or SG/7.2 The Lys218 epitope of 4B4 is more proximal to the ligand binding site (Fig. 1). However, because 4B4 binding is also diminished in the 1 wedge mutant (data not shown), it is likely that inhibition by 4B4 is, at least in part, allosteric in nature.
Is the allosteric mode specific to 1 mAb or shared with inhibitory mAb to other integrin subunits? Inhibitory mAbs to human 2 are divided into two classes based on the epitope location, one has been mapped to the top of the I-like domain primarily recognizing Glu175 or Arg122 and the other recognizes a combinatorial epitope involving the side chains of Lys133, His322, and Tyr339 (34). Homology modeling of the 2 I-like domain reveals that the latter residues are situated on the side/bottom face of the I-like domain in a location similar to the epitope residues for mAb 13, SG/19, or SG/7, suggesting the latter class of mAbs to inhibit 2 integrin function in a way similar to the mAbs described in this paper. All of the anti-3 inhibitory mAbs in which the epitopes have been mapped in detail to date recognize the top part of the I-like domain (35), and it is not known whether these mAbs inhibit 3 integrin function directly or indirectly. Inspection of the 3 integrin structure reveals that human/mouse species-specific residues are excluded from the outer hinge region between the I-like and hybrid domains, suggesting that mouse anti-human allosteric inhibitory mAbs against 3 similar to SG/19 could not be obtained because of the sequence conservation.
Mould et al. (10) have reported that anti-5 mAbs also primarily work as allosteric inhibitors. This notion was again based on the fact that the binding of inhibitory anti-5 mAbs were non-competitively attenuated by the bound ligand. However, our SPR experiments clearly showed that 51 bound by anti-5 mAb 16 or P1D6 had no detectable affinity for ligand, in contrast to the decreased but significant ligand affinity of the SG/19-bound form (Fig. 4). This strongly suggests that the mechanisms for the inhibition are quite different between these two classes of mAbs. EM imaging of the complex of Fn7–10 with the 51 headpiece fragment revealed that the 10th module docked onto the / interface with the 7–9th modules located close to the 5 subunit, although a direct interaction between the 9th module and the 5 -propeller domain was not evident (Fig. 3B and Ref. 24). It is therefore possible that binding of anti-5 mAbs to the top of the 5 -propeller sterically precludes the approach of the Fn7–9 portion, even if the binding site is distinct from the ligand binding pocket.
The present study underscores the strength of EM imaging in analyzing the conformation of protein complexes. The determination of epitope residues by mutagenesis, combined with three-dimensional structure information, generally leads to a reasonable assumption about the binding mode of an antibody. Direct visualization of the antigen-antibody complex, however, reveals not only the location of the actual interface but also elucidates the impact of antibody binding on the conformation of the target protein. In contrast to x-ray crystal structures, which usually show one of several possible low energy conformations, molecular EM samples the entire set of conformations of single particles and can thus produce structures for all of the major conformations present in the protein population. Another advantage is that EM imaging can be accomplished with a relatively small amount of dilute sample.
In conclusion, we have shown that one class of allosteric anti-1 mAbs blocks ligand binding by preventing the "swingout" of the hybrid domain from the I-like domain, thus stabilizing the closed conformation of the integrin headpiece. There may exist another class of allosterically inhibitory mAbs that stabilize the "bent" conformation of integrins by stabilizing the head-tail or -tail--tail interactions. The use of mAbs as a probe for integrin structure and function has yielded a profound understanding of the mechanisms underlying ligand recognition and affinity regulation of integrins. The unique mode of functional perturbation by anti-1 integrin mAbs described here may have important implications in developing anti-integrin pharmaceuticals. For example, an antibody that stabilizes the low affinity conformation of an integrin without completely eliminating its function may prove desirable to prevent the oversuppression of integrin activity. Stabilizing the natural low affinity conformation may also be advantageous, because an antagonist-induced conformational change has been implicated in the development of autoantibodies in patients receiving anti-integrin drugs (36).
|
FOOTNOTES |
|---|
|| To whom correspondence should be addressed. Tel.: 81-6-6879-8607; Fax: 81-6-6879-8609; E-mail: takagi{at}protein.osaka-u.ac.jp.
1 The abbreviations used are: mAb, monoclonal antibody; EM, electron microscopic; Fn, fibronectin; SPR, surface plasmon resonance; CHO, Chinese hamster ovary.
2 J. Takagi, unpublished result.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  X.-Y. Tang, Y.-F. Li, and S.-M. Tan Intercellular Adhesion Molecule-3 Binding of Integrin {alpha}L{beta}2 Requires Both Extension and Opening of the Integrin Headpiece J. Immunol., April 1, 2008; 180(7): 4793 - 4804. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Tchesnokova, P. Aprikian, O. Yakovenko, C. LaRock, B. Kidd, V. Vogel, W. Thomas, and E. Sokurenko Integrin-like Allosteric Properties of the Catch Bond-forming FimH Adhesin of Escherichia coli J. Biol. Chem., March 21, 2008; 283(12): 7823 - 7833. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Zhu, B. Boylan, B.-H. Luo, P. J. Newman, and T. A. Springer Tests of the Extension and Deadbolt Models of Integrin Activation J. Biol. Chem., April 20, 2007; 282(16): 11914 - 11920. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Puklin-Faucher, M. Gao, K. Schulten, and V. Vogel How the headpiece hinge angle is opened: new insights into the dynamics of integrin activation J. Cell Biol., October 23, 2006; 175(2): 349 - 360. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Zhang, W. D. Marcus, N. H. Goyal, P. Selvaraj, T. A. Springer, and C. Zhu Two-dimensional Kinetics Regulation of {alpha}L{beta}2-ICAM-1 Interaction by Conformational Changes of the {alpha}L-Inserted Domain J. Biol. Chem., December 23, 2005; 280(51): 42207 - 42218. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R.-H. Tang, E. Tng, S. K. A. Law, and S.-M. Tan Epitope Mapping of Monoclonal Antibody to Integrin {alpha}L {beta}2 Hybrid Domain Suggests Different Requirements of Affinity States for Intercellular Adhesion Molecules (ICAM)-1 and ICAM-3 Binding J. Biol. Chem., August 12, 2005; 280(32): 29208 - 29216. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B.-H. Luo, C. V. Carman, J. Takagi, and T. A. Springer Disrupting integrin transmembrane domain heterodimerization increases ligand binding affinity, not valency or clustering PNAS, March 8, 2005; 102(10): 3679 - 3684. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Chen, J. Takagi, C. Xie, T. Xiao, B.-H. Luo, and T. A. Springer The Relative Influence of Metal Ion Binding Sites in the I-like Domain and the Interface with the Hybrid Domain on Rolling and Firm Adhesion by Integrin {alpha}4{beta}7 J. Biol. Chem., December 31, 2004; 279(53): 55556 - 55561. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Tng, S.-M. Tan, S. Ranganathan, M. Cheng, and S. K. A. Law The Integrin {alpha}L{beta}2 Hybrid Domain Serves as a Link for the Propagation of Activation Signal from Its Stalk Regions to the I-like Domain J. Biol. Chem., December 24, 2004; 279(52): 54334 - 54339. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||
