FTY720 and two novel butterfly derivatives exert a general anti-inflammatory potential by reducing immune cell adhesion to endothelial cells through activation of S1P3 and phosphoinositide 3-kinase

Sphingosine-1-phosphate (S1P) is a key lipid regulator of a variety of cellular responses including cell proliferation and survival, cell migration, and inflammatory reactions. Here, we investigated the effect of S1P receptor activation on immune cell adhesion to endothelial cells under inflammatory conditions. We show that S1P reduces both tumor necrosis factor (TNF)-α- and lipopolysaccharide (LPS)-stimulated adhesion of Jurkat and U937 cells to an endothelial monolayer. The reducing effect of S1P was reversed by the S1P1+3 antagonist VPC23019 but not by the S1P1 antagonist W146. Additionally, knockdown of S1P3, but not S1P1, by short hairpin RNA (shRNA) abolished the reducing effect of S1P, suggesting the involvement of S1P3. A suppression of immune cell adhesion was also seen with the immunomodulatory drug FTY720 and two novel butterfly derivatives ST-968 and ST-1071. On the molecular level, S1P and all FTY720 derivatives reduced the mRNA expression of LPS- and TNF-α-induced adhesion molecules including ICAM-1, VCAM-1, E-selectin, and CD44 which was reversed by the PI3K inhibitor LY294002, but not by the MEK inhibitor U0126.In summary, our data demonstrate a novel molecular mechanism by which S1P, FTY720, and two novel butterfly derivatives acted anti-inflammatory that is by suppressing gene transcription of various endothelial adhesion molecules and thereby preventing adhesion of immune cells to endothelial cells and subsequent extravasation.

The physiological roles of ICAM-1 and ICAM-2 in neutrophil migration into tissues

PURPOSE OF REVIEW Neutrophil extravasation from the blood into tissues is initiated by tethering and rolling of neutrophils on endothelial cells, followed by neutrophil integrin activation and shear resistant arrest, crawling, diapedesis and breaching the endothelial basement membrane harbouring pericytes. Endothelial intercellular cell adhesion molecule (ICAM)-1 and ICAM-2, in conjunction with ICAM-1 on pericytes, critically contribute to each step. In addition, epithelial ICAM-1 is involved in neutrophil migration to peri-epithelial sites. The most recent findings on the role of ICAM-1 and ICAM-2 for neutrophil migration into tissues will be reviewed here. RECENT FINDINGS Signalling via endothelial ICAM-1 and ICAM-2 contributes to stiffness of the endothelial cells at sites of chronic inflammation and junctional maturation, respectively. Endothelial ICAM-2 contributes to neutrophil crawling and initiation of paracellular diapedesis, which then proceeds independent of ICAM-2. Substantial transcellular neutrophil diapedesis across the blood-brain barrier is strictly dependent on endothelial ICAM-1 and ICAM-2. Endothelial ICAM-1 or ICAM-2 is involved in neutrophil-mediated plasma leakage. ICAM-1 on pericytes assists the final step of neutrophil extravasation. Epithelial ICAM-1 rather indirectly promotes neutrophil migration to peri-epithelial sites. SUMMARY ICAM-1 and ICAM-2 are involved in each step of neutrophil extravasation, and have redundant but also distinct functions. Analysis of the role of endothelial ICAM-1 requires simultaneous consideration of ICAM-2.

STUDIES ON THE CLASS I GLYCOPROTEINS OF THE MURINE MAJOR HISTOCOMPATIBILITY COMPLEX (TRANSPLANTATION ANTIGEN, LYMPHOCYTE, CELL SURFACE MOLECULE)

A series of studies were undertaken to analyze and compare various aspects of murine class I glycoproteins. An initial area of investigation characterized the Qa-1 alloantigens using two-dimensional gel electrophoresis. Analysis of the products of the Qa-1('b), Qa-1('c) and Qa-1('d) alleles indicated that these were distinct molecules as determined by their lack of comigration upon comparative two-dimensional gel analysis. The importance of asparagine-linked glycosylation in the cell surface expression of class I molecules was also examined. These studies employed tunicamycin, an inhibitor of N-linked glycosylation. Tunicamycin treatment of activated T lymphocytes diminished the surface expression of Qa-1 to undetectable levels; the levels of other class I molecules exhibited little or no decrease. These results indicated that N-linked glycosylation has a differential importance in the cell surface expression of various class I molecules. The molecular weight diversity of class I molecules was also investigated. Molecular weight determination of both the fully glycosylated and unglycosylated forms of H-2 and Qa/Tla region encoded molecules established that there is a significant variation in the sizes of these forms of various class I molecules. The most significant difference ((TURN)9,000 daltons) exists between the unglycosylated forms of H-2K('b) and Qa-2, suggesting that the structural organization of these two molecules may be very different. A comparative two-dimensional gel analysis of various class I glycoproteins isolated from resting and activated T and B lymphocytes indicated that class I molecules expressed on activated T cells exhibited an isoelectrophoretic pattern that was distinct from the isoelectrophoretic pattern of class I molecules expessed on the other cell populations. This difference was attributed to a lower sialic acid content of the molecules expressed on activated T cells. Analysis of cell homogenates determined that activated T cells contained a higher level of endogenous neuraminidase activity than was detected in the other populations, suggesting that this may be the basis of the lower sialic acid content. The relationship of the Qa-4 and Qa-2 alloantigens was also examined. It was established that upon mitogen activation, the expression of Qa-4 was greatly decreased, whereas Qa-2 expression was not decreased. However, an anti-Qa-2 monoclonal antibody blocked the binding of an anti-Qa-4 monoclonal antibody to resting cells. These studies established that Qa-4 is a determinant restricted to resting cells, which is closely associated on the surface with the Qa-2 molecule. ^

T cell adhesion regulation from clustering GM1 lipid rafts

Lipid rafts are small laterally mobile cell membrane structures that are highly enriched in lymphocyte signaling molecules. Lipid rafts can form from the assembly of specialized lipids and proteins through hydrophobic associations from saturated acyl chains. GM1 gangliosides are a common lipid raft component and have been shown to be essential in many T cell functions. Current lipid raft theory hypothesizes that certain aspects of T cell signaling can be initiated from the coalescence of these signaling-enriched lipid rafts to sites of receptor engagement. We have described how the specific aggregation of GM1 lipid rafts can cause a reorganization of cell surface molecular associations which include dynamic associations of β1 integrins with GM1 lipid rafts. These associations had pronounced effects on T cell adhesive and migratory states. We show that GM1 lipid raft aggregation can dramatically inhibit T cell migration and chemotaxis on the extracellular matrix constituent fibronectin. This inhibition of migration function was shown to be dependent on the src kinase Lck and PKC-regulated F-actin polymerization to extending pseudopods. Furthermore, GM1 lipid raft clustering could activate T cell adhesion-strengthening mechanisms. These include an increase in cellular rigidity, the creation of polymerized cortical F-actin structures, the induction of high affinity integrin states, an increase in surface area and symmetry of the contact plane, and resistance to shear flow detachment while adherent to fibronectin. This indicates that GM1 lipid raft aggregation defines a novel stimulus to regulate lymphocyte motility and cellular adhesion which could have important implications in T cell homing mechanisms. ^

Intercellular adhesion and membrane polarity in rat hepatocytes: Localization of cell -CAM 105 and other domain-specific membrane proteins {\it in situ\/} and {\it in vitro\/} by electron microscopy

Cell-CAM 105 has been identified as a cell adhesion molecule (CAM) based on the ability of monospecific and monovalent anti-cell-CAM 105 antibodies to inhibit the reaggregation of rat hepatocytes. Although one would expect to find CAMs concentrated in the lateral membrane domain where adhesive interactions predominate, immunofluorescence analysis of rat liver frozen sections revealed that cell-CAM 105 was present exclusively in the bile canalicular (BC) domain of the hepatocyte. To more precisely define the in situ localization of cell-CAM 105, immunoperoxidase and electron microscopy were used to analyze intact and mechanically dissociated fixed liver tissue. Results indicate that although cell-CAM 105 is apparently restricted to the BC domain in situ, it can be detected in the pericanalicular region of the lateral membranes when accessibility to lateral membranes is provided by mechanical dissociation. In contrast, when hepatocytes were labeled following incubation in vitro under conditions used during adhesion assays, cell-CAM 105 had redistributed to all areas of the plasma membrane. Immunofluorescence analysis of primary hepatocyte cultures revealed that cell-CAM 105 and two other BC proteins were localized in discrete domains reminscent of BC while cell-CAM 105 was also present in regions of intercellular contact. These results indicate that the distribution of cell-CAM 105 under the experimental conditions used for cell adhesion assays differs from that in situ and raises the possibility that its adhesive function may be modulated by its cell surface distribution. The implications of these and other findings are discussed with regard to a model for BC formation.^ Analysis of molecular events involved in BC formation would be accelerated if an in vitro model system were available. Although BC formation in culture has previously been observed, repolarization of cell-CAM 105 and two other domain-specific membrane proteins was incomplete. Since DMSO had been used by Isom et al. to maintain liver-specific gene expression in vitro, the effect of this differentiation system on the polarity of these membrane proteins was examined. Based on findings presented here, DMSO apparently prolongs the expression and facilitates polarization of hepatocyte membrane proteins in vitro. ^

The structure of the two amino-terminal domains of human ICAM-1 suggests how it functions as a rhinovirus receptor and as an LFA-1 integrin ligand

The normal function of human intercellular adhesion molecule-1 (ICAM-1) is to provide adhesion between endothelial cells and leukocytes after injury or stress. ICAM-1 binds to leukocyte function-associated antigen (LFA-1) or macrophage-1 antigen (Mac-1). However, ICAM-1 is also used as a receptor by the major group of human rhinoviruses and is a catalyst for the subsequent viral uncoating during cell entry. The three-dimensional atomic structure of the two amino-terminal domains (D1 and D2) of ICAM-1 has been determined to 2.2-Å resolution and fitted into a cryoelectron microscopy reconstruction of a rhinovirus–ICAM-1 complex. Rhinovirus attachment is confined to the BC, CD, DE, and FG loops of the amino-terminal Ig-like domain (D1) at the end distal to the cellular membrane. The loops are considerably different in structure to those of human ICAM-2 or murine ICAM-1, which do not bind rhinoviruses. There are extensive charge interactions between ICAM-1 and human rhinoviruses, which are mostly conserved in both major and minor receptor groups of rhinoviruses. The interaction of ICAMs with LFA-1 is known to be mediated by a divalent cation bound to the insertion (I)-domain on the α chain of LFA-1 and the carboxyl group of a conserved glutamic acid residue on ICAMs. Domain D1 has been docked with the known structure of the I-domain. The resultant model is consistent with mutational data and provides a structural framework for the adhesion between these molecules.

The CXC chemokine stromal cell-derived factor 1 is not responsible for CD8+ T cell suppression of syncytia-inducing strains of HIV-1

Primary CD8+ T cells from HIV+ asymptomatics can suppress virus production from CD4+ T cells acutely infected with either non-syncytia-inducing (NSI) or syncytia-inducing (SI) HIV-1 isolates. NSI strains of HIV-1 predominantly use the CCR5 chemokine receptor as a fusion cofactor, whereas fusion of T cell line-adapted SI isolates is mediated by another chemokine receptor, CXCR4. The CCR5 ligands RANTES (regulated on activation, normal T cell expressed and secreted), macrophage inflammatory protein 1α (MIP-1α), and MIP-1β are HIV-1 suppressive factors secreted by CD8+ cells that inhibit NSI viruses. Recently, the CXC chemokine stromal cell-derived factor 1 (SDF-1) was identified as a ligand for CXCR4 and shown to inhibit SI strains. We speculated that SDF-1 might be an effector molecule for CD8+ suppression of SI isolates and assessed several SDF-1 preparations for inhibition of HIV-1LAI-mediated cell–cell fusion, and examined levels of SDF-1 transcripts in CD8+ T cells. SDF-1 fusion inhibitory activity correlated with the N terminus, and the α and β forms of SDF-1 exhibited equivalent fusion blocking activity. SDF-1 preparations having the N terminus described by Bleul et al. (Bleul, C.C., Fuhlbrigge, R.C., Casasnovas, J.M., Aiuti, A. & Springer, T.A. (1996) J. Exp. Med. 184, 1101–1109) readily blocked HIV-1LAI-mediated fusion, whereas forms containing two or three additional N-terminal amino acids lacked this activity despite their ability to bind and/or signal through CXCR4. Though SDF-1 is constitutively expressed in most tissues, CD8 T cells contained extremely low levels of SDF-1 mRNA transcripts (<1 transcript/5,000 cells), and these levels did not correlate with virus suppressive activity. We conclude that suppression of SI strains of HIV-1 by CD8+ T cells is unlikely to involve SDF-1.

The human natural killer cell immune synapse

Inhibitory killer Ig-like receptors (KIR) at the surface of natural killer (NK) cells induced clustering of HLA-C at the contacting surface of target cells. In this manner, inhibitory immune synapses were formed as human NK cells surveyed target cells. At target/NK cell synapses, HLA-C/KIR distributed into rings around central patches of intercellular adhesion molecule-1/lymphocyte function-associated antigen-1, the opposite orientation to mature murine T cell-activating synapses. This organization of protein was stable for at least 20 min. Cells could support multiple synapses simultaneously, and clusters of HLA-C moved as NK cells crawled over target cells. Clustering required a divalent metal cation, explaining how metal chelators inhibit KIR function. Surprisingly, however, formation of inhibitory synapses was unaffected by ATP depletion and the cytoskeletal inhibitors, colchicine and cytochalsins B and D. Clearly, supramolecular organization within plasma membranes is critical for NK cell immunosurveillance.

Specific Activation of Leukocyte β2 Integrins Lymphocyte Function–associated Antigen-1 and Mac-1 by Chemokines Mediated by Distinct Pathways via the α Subunit Cytoplasmic Domains

We show that CC chemokines induced a sustained increase in monocyte adhesion to intercellular adhesion molecule-1 that was mediated by Mac-1 (αMβ2) but not lymphocyte function–associated antigen-1 (LFA-1; αLβ2). In contrast, staining for an activation epitope revealed a rapid and transient up-regulation of LFA-1 activity by monocyte chemotactic protein-1 (MCP-1) in monocytes and Jurkat CCR2 chemokine receptor transfectants or by stromal-derived factor-1α in Jurkat cells. Differential kinetics for activation of Mac-1 (sustained) and LFA-1 (transient) avidity in response to stromal-derived factor-1α were confirmed by expression of αM or αL in αL-deficient Jurkat cells. Moreover, expression of chimeras containing αL and αM cytoplasmic domain exchanges indicated that α cytoplasmic tails conferred the specific mode of regulation. Coexpressing αM or chimeras in mutant Jurkat cells with a “gain of function” phenotype that results in constitutively active LFA-1 demonstrated that Mac-1 was not constitutively active, whereas constitutive activity was mediated via the αL cytoplasmic tail, implying the presence of distinct signaling pathways for LFA-1 and Mac-1. Transendothelial chemotaxis of monocytes in response to MCP-1 was dependent on LFA-1; however, Mac-1 was involved at MCP-1 concentrations stimulating its avidity, showing differential contributions of β2 integrins. Our data suggest that a specific regulation of β2 integrin avidity by chemokines may be important in leukocyte extravasation and may be triggered by distinct activation pathways transduced via the α subunit cytoplasmic domains.

The PH Domain and the Polybasic c Domain of Cytohesin-1 Cooperate specifically in Plasma Membrane Association and Cellular Function

Recruitment of intracellular proteins to the plasma membrane is a commonly found requirement for the initiation of signal transduction events. The recently discovered pleckstrin homology (PH) domain, a structurally conserved element found in ∼100 signaling proteins, has been implicated in this function, because some PH domains have been described to be involved in plasma membrane association. Furthermore, several PH domains bind to the phosphoinositides phosphatidylinositol-(4,5)-bisphosphate and phosphatidylinositol-(3,4,5)-trisphosphate in vitro, however, mostly with low affinity. It is unclear how such weak interactions can be responsible for observed membrane binding in vivo as well as the resulting biological phenomena. Here, we investigate the structural and functional requirements for membrane association of cytohesin-1, a recently discovered regulatory protein of T cell adhesion. We demonstrate that both the PH domain and the adjacent carboxyl-terminal polybasic sequence of cytohesin-1 (c domain) are necessary for plasma membrane association and biological function, namely interference with Jurkat cell adhesion to intercellular adhesion molecule 1. Biosensor measurements revealed that phosphatidylinositol-(3,4,5)-trisphosphate binds to the PH domain and c domain together with high affinity (100 nM), whereas the isolated PH domain has a substantially lower affinity (2–3 μM). The cooperativity of both elements appears specific, because a chimeric protein, consisting of the c domain of cytohesin-1 and the PH domain of the β-adrenergic receptor kinase does not associate with membranes, nor does it inhibit adhesion. Moreover, replacement of the c domain of cytohesin-1 with a palmitoylation–isoprenylation motif partially restored the biological function, but the specific targeting to the plasma membrane was not retained. Thus we conclude that two elements of cytohesin-1, the PH domain and the c domain, are required and sufficient for membrane association. This appears to be a common mechanism for plasma membrane targeting of PH domains, because we observed a similar functional cooperativity of the PH domain of Bruton’s tyrosine kinase with the adjacent Bruton’s tyrosine kinase motif, a novel zinc-containing fold.

Use of a β1 Integrin-deficient Human T Cell to Identify β1 Integrin Cytoplasmic Domain Sequences Critical for Integrin Function

T cell activation rapidly and transiently regulates the functional activity of integrin receptors. Stimulation of CD3/T cell receptor, CD2 or CD28, as well as activation with phorbol esters, can induce within minutes an increase in β1 integrin-mediated adhesion of T cells to fibronectin. In this study, we have produced and utilized a mutant of the Jurkat T cell line, designated A1, that lacks protein and mRNA expression of the β1 integrin subunit but retains normal levels of CD2, CD3, and CD28 on the cell surface. Activation-dependent adhesion of A1 cells to fibronectin could be restored upon transfection of a wild-type human β1 integrin cDNA. Adhesion induced by phorbol 12-myristate 13-acetate-, CD3-, CD2-, and CD28 stimulation did not occur if the carboxy-terminal five amino acids of the β1 tail were truncated or if either of two well-conserved NPXY motifs were deleted. Scanning alanine substitutions of the carboxy-terminal five amino acids demonstrated a critical role for the tyrosine residue at position 795. The carboxy-terminal truncation and the NPXY deletions also reduced adhesion induced by direct stimulation of the β1 integrin with the activating β1 integrin-specific mAb TS2/16, although the effects were not as dramatic as observed with the other integrin-activating signals. These results demonstrate a vital role for the amino-terminal NPXY motif and the carboxy-terminal end of the β1 integrin cytoplasmic domain in activation-dependent regulation of integrin-mediated adhesion in T cells. Furthermore, the A1 cell line represents a valuable new cellular reagent for the analysis of β1 integrin structure and function in human T cells.

NMR and mutagenesis evidence for an I domain allosteric site that regulates lymphocyte function-associated antigen 1 ligand binding

The leukocyte integrin, lymphocyte function-associated antigen 1 (LFA-1) (CD11a/CD18), mediates cell adhesion and signaling in inflammatory and immune responses. To support these functions, LFA-1 must convert from a resting to an activated state that avidly binds its ligands such as intercellular adhesion molecule 1 (ICAM-1). Biochemical and x-ray studies of the Mac-1 (CD11b/CD18) I domain suggest that integrin activation could involve a conformational change of the C-terminal α-helix. We report the use of NMR spectroscopy to identify CD11a I domain residues whose resonances are affected by binding to ICAM-1. We observed two distinct sites in the CD11a I domain that were affected. As expected from previous mutagenesis studies, a cluster of residues localized around the metal ion-dependent adhesion site (MIDAS) was severely perturbed on ICAM-1 binding. A second cluster of residues distal to the MIDAS that included the C-terminal α-helix of the CD11a I domain was also affected. Substitution of residues in the core of this second I domain site resulted in constitutively active LFA-1 binding to ICAM-1. Binding data indicates that none of the 20 substitution mutants we tested at this second site form an essential ICAM-1 binding interface. We also demonstrate that residues in the I domain linker sequences can regulate LFA-1 binding. These results indicate that LFA-1 binding to ICAM-1 is regulated by an I domain allosteric site (IDAS) and that this site is structurally linked to the MIDAS.

Na,K-ATPase β-Subunit Is Required for Epithelial Polarization, Suppression of Invasion, and Cell Motility

The cell adhesion molecule E-cadherin has been implicated in maintaining the polarized phenotype of epithelial cells and suppression of invasiveness and motility of carcinoma cells. Na,K-ATPase, consisting of an α- and β-subunit, maintains the sodium gradient across the plasma membrane. A functional relationship between E-cadherin and Na,K-ATPase has not previously been described. We present evidence that the Na,K-ATPase plays a crucial role in E-cadherin–mediated development of epithelial polarity, and suppression of invasiveness and motility of carcinoma cells. Moloney sarcoma virus-transformed Madin-Darby canine kidney cells (MSV-MDCK) have highly reduced levels of E-cadherin and β1-subunit of Na,K-ATPase. Forced expression of E-cadherin in MSV-MDCK cells did not reestablish epithelial polarity or inhibit the invasiveness and motility of these cells. In contrast, expression of E-cadherin and Na,K-ATPase β1-subunit induced epithelial polarization, including the formation of tight junctions and desmosomes, abolished invasiveness, and reduced cell motility in MSV-MDCK cells. Our results suggest that E-cadherin–mediated cell-cell adhesion requires the Na,K-ATPase β-subunit's function to induce epithelial polarization and suppress invasiveness and motility of carcinoma cells. Involvement of the β1-subunit of Na,K-ATPase in the polarized phenotype of epithelial cells reveals a novel link between the structural organization and vectorial ion transport function of epithelial cells.

Characterization of Syntenin, a Syndecan-binding PDZ Protein, as a Component of Cell Adhesion Sites and Microfilaments

Syntenin is a PDZ protein that binds the cytoplasmic C-terminal FYA motif of the syndecans. Syntenin is widely expressed. In cell fractionation experiments, syntenin partitions between the cytosol and microsomes. Immunofluorescence microscopy localizes endogenous and epitope-tagged syntenin to cell adhesion sites, microfilaments, and the nucleus. Syntenin is composed of at least three domains. Both PDZ domains of syntenin are necessary to target reporter tags to the plasma membrane. The addition of a segment of 10 amino acids from the N-terminal domain of syntenin to these PDZ domains increases the localization of the tags to stress fibers and induces the formation of long, branching plasma membrane extensions. The addition of the complete N-terminal region, in contrast, reduces the localization of the tags to plasma membrane/adhesion sites and stress fibers, and reduces the morphotypical effects. Recombinant domains of syntenin with the highest plasma membrane localization display the lowest nuclear localization. Syndecan-1, E-cadherin, β-catenin, and α-catenin colocalize with syntenin at cell-cell contacts in epithelial cells, and coimmunoprecipitate with syntenin from extracts of these cells. These results suggest a role for syntenin in the composition of adherens junctions and the regulation of plasma membrane dynamics, and imply a potential role for syntenin in nuclear processes.

Dimerization and the effectiveness of ICAM-1 in mediating LFA-1-dependent adhesion

Dimeric intercellular adhesion molecule-1 (ICAM-1) binds more efficiently to lymphocyte function-associated antigen-1 (LFA-1) than monomeric ICAM-1. However, it is unknown whether dimerization enhances binding simply by providing two ligand-binding sites and thereby increasing avidity, or whether it serves to generate a single “fully competent” LFA-1-binding surface. Domain 1 of ICAM-1 contains both the binding site for LFA-1, centered on residue E34, and a homodimerization interface. Whether the LFA-1-binding site extends across the homodimerization interface has not been tested. To address this question, we constructed four different heterodimeric soluble forms of ICAM-1 joined at the C terminus via an α-helical coiled coil (ACID-BASE). These heterodimeric ICAM-1 constructs include, (i) E34/E34 (two intact LFA-1-binding sites), (ii) E34/K34 (one disrupted LFA-1-binding site), (iii) E34/ΔD1–2 (one deleted LFA-1-binding site), and (iv) K34/K34 (two disrupted LFA-1-binding sites). Cells bearing activated LFA-1 bound similarly to surfaces coated with either E34/K34 or E34/ΔD1–2 and with an ≈2-fold reduction in efficiency compared with E34/E34, suggesting that D1 dimerization, which is precluded in E34/ΔD1-D2, is not necessary for optimal LFA-1 binding. Furthermore, BIAcore (BIAcore, Piscataway, NJ) affinity measurements revealed that soluble open LFA-1 I domain bound to immobilized soluble ICAM-1, E34/E34, E34/K34, and E34/ΔD1-D2 with nearly identical affinities. These studies demonstrate that a single ICAM-1 monomer, not dimeric ICAM-1, represents the complete, “fully competent” LFA-1-binding surface.