The T lymphocyte glycoprotein CD2 binds the cell surface ligand LFA-3

CD2 (known also as T11 (ref. 1), LFA-2 (ref. 2) and the erythrocyte rosette receptor (ref. 3] is a functionally important T lymphocyte surface glycoprotein of relative molecular mass 50,000 to 58,000 (Mr 50-58 K) which appears early in thymocyte ontogeny and is present on all mature T cells. Monoclonal antibodies to CD2 inhibit cytotoxic T-lymphocyte (CTL)-mediated killing by binding to the T lymphocyte and blocking adhesion to the target cell. Such antibodies also inhibit T helper cell responses including antigen-stimulated proliferation, interleukin-2 (IL-2) secretion, and IL-2 receptor expression. Certain combinations of monoclonal antibodies to CD2 epitopes trigger proliferation of peripheral blood T lymphocytes, cytotoxic effector function and expression of IL-2 receptors by thymocytes, resulting in thymocyte proliferation in the presence of exogenous IL-2 (ref. 11). These findings suggest that CD2 can function in signalling as well as being an adhesion molecule. To understand the role of CD2 in T-cell adhesion and activation, it is essential to define its natural ligand. Our previous observation that purified CD2 inhibits rosetting of T lymphocytes with sheep erythrocytes and can be absorbed by sheep erythrocytes suggested it also might bind with detectable affinity to human cells. We now report that CD2 binds to a cell-surface antigen known as lymphocyte function-associated antigen-3 (LFA-3) with high affinity, and can mediate adhesion of lymphoid cells via interaction with LFA-3.     

An LFA-3 cDNA encodes a phospholipid-linked membrane protein homologous to its receptor CD2

Recently the human T cell erythrocyte receptor CD2 has been shown to bind human erythrocytes through LFA-3, a heavily glycosylated surface protein of broad tissue distribution. CD2-LFA-3 interactions are important for cytolytic conjugate formation, for thymocyte adhesion, and for T cell activation. A complementary DNA clone encoding LFA-3 was isolated using a complementary DNA clone encoding LFA-3 was isolated using a novel transient expression system of mouse cells. The cDNA encodes a phospholipid-linked membrane protein whose extracellular domain shares significant homology with CD2. As CD2 is homologous with the neural cell adhesion molecule NCAM in immunoglobulin-like domains, cellular adhesion molecules in both neural and lymphoid tissues could have a common ancestor.     

Regulatory properties of LFA-1 alpha and beta chains in human T-lymphocyte activation

Lymphocyte function-associated antigen-1 (LFA-1) is a heterodimer composed of an alpha and beta chain that is expressed on the surface of most leukocytes and is an essential molecule for adhesion reactions between cells participating in the immune response. A putative ligand for LFA-1 is the intercellular adhesion molecule ICAM-1 (refs 3-5). Leukocyte adhesion abnormality is found in patients with LFA-1 deficiency. It is not clear whether binding of ligand to the LFA-1 molecule merely spatially orientates cells towards each other or can also induce signals that regulate cell activation and differentiation. We have recently developed a T-cell proliferation assay which uses immobilized anti-CD3 monoclonal antibodies as stimulant and is independent of LFA-1-mediated cellular adhesion. As there is no interference by anti-LFA-1 monoclonal antibodies with the adhesion-dependent activation steps, this T-cell activation system allows us to investigate whether transmembrane signals are induced by binding of ligand to LFA-1 on T cells. Our data indicate that binding of ligand to LFA-1 results in the transduction of regulatory signal across the plasma membrane, rather like other molecules (CD2, CD4, CD8) (refs 8-11) with signal-modifying properties involved in the adhesion of T cells to target/stimulator cells. Indeed, adhesion molecules might generally be important in signal transduction, even in cells not belonging to the immune system.     

CD2-mediated adhesion facilitates T lymphocyte antigen recognition function

The CD2 T lymphocyte-surface glycoprotein serves to mediate adhesion between T lymphocytes and their cognate cellular partners which express the specific ligand LFA-3. In addition, CD2 by itself or in conjunction with T-cell receptor stimulation, transduces signals resulting in T-lymphocyte activation. One or both of these functions seems to be physiologically important, given that certain anti-CD2 monoclonal antibodies block T-cell activation and that antigen-responsive memory T cells express a high level of CD2 relative to virgin T cells, which are largely antigen-unresponsive. Nevertheless, the contribution of the individual CD2 functions in T-cell responses has not been independently examined. To this end, human CD2 complementary DNAs encoding an intact LFA-3-binding adhesion domain, but lacking a functional cytoplasmic signal transduction element (CD2trans-), were introduced into an ovalbumin-specific, I-Ad restricted murine T-cell hybridoma. The antigen-specific response of T hybridoma cells expressing human CD2trans- protein was enhanced up to 400% when the human LFA-3 ligand was introduced into the I-Ad expressing murine antigen-presenting cells. In contrast, no augmentation was observed if human LFA-3 was absent or expressed on a third-party cell lacking the I-Ad restriction element. These results directly demonstrate the functional significance of adhesion events mediated between CD2 on the antigen-responsive T lymphocyte and LFA-3 on the presenting cell in optimizing antigen-specific T-cell activation.     

Enhancement of LFA-1-mediated cell adhesion by triggering through CD2 or CD3 on T lymphocytes

The lymphocyte function-associated molecule LFA-1 (CD11a/CD18) plays a key part in lymphocyte adhesion. Lymphocytes do not adhere spontaneously; activation of protein kinase C (PKC) by phorbol esters, however, gives rise to strong LFA-1-dependent adhesion, indicating that activation of LFA-1 is required to induce cell adhesion. We have now investigated whether the functionally important CD2 and CD3 surface structures on T lymphocytes are involved in the activation of LFA-1. The stimulation of these molecules, which causes activation of PKC, strongly promoted LFA-1-dependent adhesion. Furthermore, we demonstrate by using cells from an LFA-1-deficient patient that this enhanced lymphocyte adhesion is caused by activation of the LFA-1 molecule and not by activation of its ligands. LFA-1 was persistently activated by triggering through CD2 but only transiently by triggering through CD3. We postulate that CD2 and CD3 can differentially regulate the affinity of LFA-1 for its ligands by modulating its molecular conformation through PKC-dependent mechanisms.     

Alternative pathway activation of T cells by binding of CD2 to its cell-surface ligand

Activation of resting T lymphocytes is initiated by the interaction of cell-surface receptors with their corresponding ligands. In addition to activation through the CD3 (T3)-Ti antigen-receptor complex, recent experiments have demonstrated induction of T-cell proliferation through the CD2 (T11) molecule, traditionally known as the erythrocyte(E)-receptor, through which T cells can bind red blood cells (RBC). This 'alternative pathway' of T-cell activation was observed in vitro in response to combinations of anti-CD2 monoclonal antibodies (mAbs) that bind to distinct epitopes of CD2, such as mAbs against T11(2) plus T11(3). The physiological importance of this activation pathway can be assessed only by studying the effects of a naturally occurring ligand of CD2 on T-cell activation. We have recently described such a ligand, a glycoprotein of apparent relative molecular mass 42,000 (Mr 42K) that is expressed on all blood cells and some other tissues. Here we demonstrate that binding of this cell surface molecule, termed T11 target structure or T11TS, to CD2 (T11) induces reactivity in resting T cells to a mitogenic stimulus given by a mAb to the T11(3) determinant or by submitogenic concentrations of anti-T11(2+3) mAbs. Thus, one of the signals required for T-cell activation through the alternative pathway is provided by the interaction of CD2 with a naturally occurring complementary cell-surface molecule.     

Two antigen-independent adhesion pathways used by human cytotoxic T-cell clones

Cell-cell adhesion is essential for many immunological functions, including interaction of cytotoxic T lymphocytes (CTLs) with their targets. We have explored CTL-target interactions using well-characterized cloned human CTLs. Conjugate formation between these CTLs and many antigen-negative targets is almost as efficient as with specific target cells, but does not lead to target-cell lysis. Thus, on specific target cells, adhesion by antigen-independent pathways may occur concurrently with or precede antigen recognition. The molecules LFA-1, CD2 (T11, LFA-2) and LFA-3 have been shown to be involved in human CTL conjugation with and lysis of specific target cells. Here we describe monoclonal antibody inhibition studies using individual monoclonal antibodies and mixes which demonstrate (1) that LFA-1, CD2 and LFA-3 are involved in antigen-independent conjugate formation; and (2) suggest that CD2 and LFA-3 are involved in one pathway and LFA-1 in another. We confirmed the existence of distinct pathways by the demonstration that LFA-1-dependent adhesion requires divalent cations and is temperature-sensitive whereas CD2- and LFA-3-dependent adhesion does not require divalent cations and is temperature-insensitive. Together with previous data, our studies suggest that CD2 on the effector interacts with LFA-3 as its ligand on targets.     

Structure of domain 1 of rat T lymphocyte CD2 antigen.

The CD2 antigen is largely restricted to cells of the T-lymphocyte lineage and has been established as an important adhesion molecule in interactions between human T lymphocytes and accessory cells. In the adhesion reaction, CD2 on T cells binds to LFA-3 on other cells, with binding through domain 1 of CD2. CD2 can also be a target for the delivery of mitogenic signals to T lymphocytes cultured with combinations of anti-CD2 antibodies. Two predictions that are contradictory have been made for the structure of CD2 domain 1. One suggests an immunoglobulin (Ig) fold, on the basis of sequence patterns conserved in the Ig-superfamily (IgSF), whilst the other proposes a pattern of alternating alpha-helices and beta-strands, on the basis of secondary structure predictions. Thus CD2 domain 1 is an important test case for the validity of IgSF assignments based on sequence patterns. We report here the expression of domain 1 of rat CD2 in an Escherichia coli expression system and have determined a low-resolution solution structure by NMR spectroscopy.     

Functional cloning of ICAM-2, a cell adhesion ligand for LFA-1 homologous to ICAM-1

The leukocyte adhesion molecule LFA-1 mediates a wide range of lymphocyte, monocyte, natural killer cell, and granulocyte interactions with other cells in immunity and inflammation. LFA-1 (CD11a/CD18) is a receptor for intercellular adhesion molecule 1 (ICAM-1, CD54), a surface molecule which is constitutively expressed on some tissues and induced on other in inflammation. Induction of ICAM-1 on epithelial cells, endothelial cells and fibroblasts mediates LFA-1-dependent adhesion of lymphocytes. Several lines of evidence have suggested the existence of a second LFA-1 ligand: homotypic adhesion of one cell line was inhibited by a monoclonal antibody to LFA-1, but not by one to ICAM-1; there exists an LFA-1-dependent, ICAM-1-independent pathway of adhesion to endothelial cells; and also, there are some types of target cells in which LFA-1-dependent T-lymphocyte adhesion and lysis are independent of ICAM-1. We have cloned this second ligand, designated ICAM-2, using a novel method for identifying ligands of adhesion molecules. ICAM-2 is an integral membrane protein with two immunoglobulin-like domains, whereas ICAM-1 has five. Remarkably, ICAM-2 is much more closely related to the two most N-terminal domains of ICAM-1 (34% identity) than either ICAM-1 or ICAM-2 is to other members of the immunoglobulin superfamily, demonstrating the existence of a subfamily of immunoglobulin-like ligands that bind the same integrin receptor.     

Evidence for existence of a close association between CD14 and CXCR4 on monocytic cell line U937

The surface expression of HIV-1 coreceptors (CXCR4 and CCR5) on monocytes can be regulated by the ligand of CD14,and the susceptibility of the cells to HIV-1 is then changed.Our previous study found that monoclonal antibody against CD14 could dramatically inhibit CXCR4-mediated chemotaxis and cell-cell fusion.Based on these studies,we explored potential relationship between CD14 and CXCR4 on monocytic cell line U937.Flow cytometry analysis showed that anti-CXCR4 monoclonal antibody (mAb) 12G5 strongly inhibited binding of the FITC-conjugated anti-CD14 monoclonal antibodies (TUK4 and UCHM1) to U937,while another CX- CR4-specific mAb B-R24 did not show any effect on this binding.On the other hand,two anti-CD14 monoclonal antibodies (TUK4 and UCH-M1) obviously inhibited the binding of the PE-conjugated anti-CXCR4 mAb 12G5 to U937 but did not inhibit the binding of mAb 12G5 to CXCR4-transfected 3T3 cells (3T3.T4.CXCR4),which indicates that the blocking of mAb 12G5 binding to CXCR4 by CD14- specific mAbs is not involved in the possibility that CD14-specific mAbs directly bind to CXCR4.These results suggested existence of a close association between CD14 and CXCR4 on monocytic cell line U937.     

Evidence for existence of a close association between CD14 and CXCR4 on monocytic cell line U937

The surface expression of HIV-1 coreceptors (CXCR4 and CCR5) on monocytes can be regulated by the ligand of CD14, and the susceptibility of the cells to HIV-1 is then changed. Our previous study found that monoclonal antibody against CD14 could dramatically inhibit CXCR4-mediated chemotaxis and cell-cell fusion. Based on these studies, we explored potential relationship between CD14 and CXCR4 on monocytic cell line U937. Flow cytometry analysis showed that anti-CXCR4 monoclonal antibody (mAb) 12G5 strongly inhibited binding of the FITC-conjugated anti-CD14 monoclonal antibodies (TUK4 and UCHM1) to U937, while another CXCR4-specific mAb B-R24 did not show any effect on this binding. On the other hand, two anti-CD14 monoclonal antibodies (TUK4 and UCH-M1) obviously inhibited the binding of the PE-conjugated anti-CXCR4 mAb 12G5 to U937 but did not inhibit the binding of mAb 12G5 to CXCR4-transfected 3T3 cells (3T3.T4.CXCR4), which indicates that the blocking of mAb 12G5 binding to CXCR4 by CD14-specific mAbs is not involved in the possibility that CD14-specific mAbs directly bind to CXCR4. These results suggested existence of a close association between CD14 and CXCR4 on monocytic cell line U937.     

Activated CD8 binding to class I protein mediated by the T-cell receptor results in signalling

The CD8 glycoprotein of T cells bind nonpolymorphic regions of class I major histocompatibility complex proteins on target cells and these interactions promote antigen recognition and signalling by the T-cell receptor. Studies using artificial membranes indicated that effective CD8/class I interaction is critical for response by alloantigen-specific cytotoxic T lymphocytes when class I protein is the only ligand on the antigen-bearing surface. But significant CD8-mediated binding of cytotoxic T lymphocytes to non-antigenic class I protein could not be detected in the absence of the alloantigen. These apparently contradictory findings indicate that CD8 binding to class I protein might be activated through the T-cell receptor and the results reported here demonstrate that this is the case. Treatment of cytotoxic T lymphocytes with soluble anti-T-cell receptor antibody activates adhesion of the cytotoxic T lymphocytes to class I, but not class II proteins. The specificity of this binding implies that it is mediated by CD8 and blocking by anti-CD8 antibodies confirmed this. Furthermore, binding of CD8 to class I protein resulted in generation of an additional signal(s) necessary to initiate response at low T-cell receptor occupancy levels.     

ICAM, an adhesion ligand of LFA-1, is homologous to the neural cell adhesion molecule NCAM

Antigen-specific cell contacts in the immune system are strengthened by antigen-nonspecific interactions, mediated in part by lymphocyte-function associated (LFA) antigens. The LFA-1 antigen is widely expressed on cells of haematopoietic origin and is a major receptor of T cells, B cells and granulocytes. LFA-1 mediates the leukocyte adhesion reactions underlying cytolytic conjugate formation, helper T-cell interactions, and antibody-dependent killing by natural killer cells and granulocytes. Recently, ICAM-1 (intercellular adhesion molecule-1) has been defined as a ligand for LFA-1. Monoclonal antibodies to ICAM-1 block T lymphocyte adhesion to fibroblasts and endothelial cells and disrupt the interaction between cytotoxic T cells and target cells. In addition, purified ICAM-1 reconstituted into artificial membranes binds LFA-1+ cells. ICAM-1 is found on leukocytes, fibroblasts, epithelial cells and endothelial cells and its expression is regulated by inflammatory cytokines. LFA-1 has been placed in the integrin family of cell surface receptors by virtue of the high sequence similarity between the LFA-1 and integrin beta chains. The adhesion ligands of the integrin family are glycoproteins bearing the Arg-Gly-Asp (RGD) sequence motif, for example, fibronectin, fibrinogen, vitronectin and von Willebrand factor. Here we show that a complementary DNA clone ICAM-1 contains no RGD motifs, but instead is homologous to the neural cell adhesion molecule NCAM.     

The B-cell antigen CD22 mediates monocyte and erythrocyte adhesion.

Interaction with antigen-presenting accessory cells is thought to be an important step in B-cell activation, and the B-cell receptor CD22, which is coordinately expressed with surface immunoglobulin, has been proposed to participate in the antigen response. Here we show that CD22 has a structure closely related to myelin-associated glycoprotein (MAG, a neuronal adhesion protein), and mediates monocyte and erythrocyte adhesion. Like CD2, the T-cell erythrocyte receptor, CD22 may facilitate antigen recognition by promoting antigen-nonspecific contacts with accessory cells.     

Identification of a CD4 binding site on the beta 2 domain of HLA-DR molecules.

The CD4 and CD8 molecules are transmembrane glycoproteins expressed by functionally distinct subsets of mature T cells. CD4+ and CD8+ T cells recognize antigens on major histocompatibility complex (MHC) class II-bearing and class I-bearing target cells respectively. The ability of monoclonal antibodies against CD4 and CD8 to block antigen recognition by T cells, as well as cell-cell adhesion assays, indicate that CD4 and CD8 bind to nonpolymorphic determinants of class II or class I MHC. Here we demonstrate that soluble recombinant HLA-DR4 molecules from insect cells and HLA-DR-derived peptides bind to immobilized recombinant soluble CD4. CD4 binds recombinant soluble DR4 heterodimers, as well as the soluble DR4-beta chain alone. Furthermore, two out of twelve DR4-beta peptides could interact specifically with CD4. These findings show that CD4 interacts with a region of MHC class II molecules analogous to a previously identified loop in class I MHC proteins that binds CD8 (refs 8, 9).     

Activation-independent binding of human memory T cells to adhesion molecule ELAM-1.

The induction of an ensemble of adhesion molecules on endothelial cells by inflammatory cytokines is likely to be crucial to the differential migration of T-lymphocyte subsets into inflammatory sites. Two molecular pathways involving the VLA-4 and LFA-1 integrins are known to mediate T-cell adhesion to activated endothelium. Here we show that a third pathway involving the rapidly inducible endothelial cell-surface adhesion molecule ELAM-1 contributes to the binding of resting CD4+ T cells to IL-1-induced human endothelial cells. All three pathways contribute to the greater adhesion to endothelium of memory T cells than naive T cells. There are two unique features of T-cell adhesion to purified ELAM-1: first, ELAM-1 exclusively mediates adhesion of memory T cells; second, memory T-cell binding to ELAM-1 is independent of acute activation events that regulate integrin-mediated adhesion. Thus, ELAM-1 may be of primary importance in the initial attachment of memory T cells to inflamed endothelium in vivo and to the preferential migration of memory T cells into tissue and inflammatory sites.     

Suppressor effects and cyclic AMP accumulation by the CD29 molecule of CD4+ lymphocytes

Integrins are a superfamily of related molecules whose function, where known, is to mediate adhesion. The so-called very-late-activation antigen (VLA) family includes at least five distinct heterodimers, each composed of a unique alpha-subunit non-covalently associated with a common beta-subunit. Several members of the family have been shown to bind extracellular matrix proteins, but the function of VLA-4 is so far unknown. VLA-4 is the only member of the family detected on thymocytes and resting T cells. We show here that an antibody which recognizes the beta-subunit of VLA-4 (CD29) on T cells can inhibit CD4+ cell proliferation triggered by CD2 or CD3, and that binding of this antibody to activated T cells leads to an increase in cyclic AMP levels which is comparable to that elicited by forskolin. These negative signalling effects are unique to this antibody: other CD29 antibodies do not affect the growth of activated CD4 cells but enhance the proliferation of whole T cell populations and abrogate the suppressive effects of mitomycin-treated CD8 cells on CD4-cell growth. Taken together, our results indicate that VLA-4 functions in cell-cell interactions and that it is the target for the suppressive effects of CD8 cells on CD4 cells.     

ICAM-1 a ligand for LFA-1-dependent adhesion of B, T and myeloid cells

Cell-cell adhesion is essential for many immunological functions. The LFA-1 molecule, a member of a superfamily of adhesion molecules, participates in adhesion which is critical to the function of each of the three major subsets of leukocytes: lymphocytes, monocytes and granulocytes. Putative LFA-1 ligands have been identified functionally in different laboratories using three different monoclonal antibodies that inhibit LFA-1-mediated leukocyte adhesion in particular model systems; however, there may be more than one LFA-1 ligand. We have directly compared the three relevant monoclonal antibodies, and show that each binds to the same molecule, intercellular-adhesion molecule-1 (ICAM-1). Most important, B, T and myeloid cells adhere specifically to purified ICAM-1-coated surfaces; such adhesion has distinctive requirements for Mg2+ and Ca2+. This constitutes biochemical evidence that ICAM-1 functions as a ligand for LFA-1-dependent adhesion by a variety of leukocytes.     

Thy-1-mediated T-cell activation requires co-expression of CD3/Ti complex

In addition to monoclonal antibodies against the CD3 (T3)-T-cell antigen receptor (CD3/Ti) complex, several other monoclonals directed towards distinct cell surface structures on human (CD2 (T11) and Tp44) and murine (Thy-1, TAP, and Ly-6) T lymphocytes are capable of activating T cells. It has been proposed that such structures may function as alternative pathways of stimulation. To examine directly whether any relationship exists between Thy-1-dependent activation phenomena and T-cell activation mediated through the CD3/Ti complex, we have transfected several CD3/Ti- variants of the human T-cell line Jurkat with the murine Thy-1.2 gene. Our data indicate that in CD3/Ti-, Thy-1.2+ transfectants, monoclonal antibodies against Thy-1.2 can induce a rise in cytoplasmic free calcium ([Ca2+]i), but fail to stimulate interleukin-2 (IL-2) production. The only defect in these variant cell lines responsible for the inability to produce IL-2 in response to Thy-1 stimulation was in the expression of the CD3/Ti complex, because replacement of defective Ti alpha- or beta-chain genes reconstributed both surface expression of CD3/Ti and responsiveness to Thy-1 in the IL-2 production assay.     

Cytotoxic T-lymphocyte activation involves a cascade of signalling and adhesion events.

In addition to the antigen-specific T-cell receptor (TCR), T cells bear an array of 'accessory' molecules that can contribute to stable adhesion to the antigen-bearing cell and provide costimulatory signals. For several of these, T-cell adhesion to the ligand can be activated by TCR-dependent signalling (a signal from the TCR primes the coreceptor to bind to its ligand). It is unclear whether the individual coreceptors share common mechanisms of priming and cosignalling, and perhaps act in a redundant manner, or whether they act in a distinct way and contribute uniquely to the activation process. We report here the use of isolated alloantigen, class I proteins and fibronectin ligands to show that coreceptors on cytotoxic T lymphocytes are activated sequentially and deliver distinct biochemical signals on binding to their ligands. TCR engagement activates CD8 by a protein tyrosine kinase-dependent pathway, and CD8 then acts as a signal for initiation of polyphosphoinositide hydrolysis on binding to class I. In contrast, activated adhesion to fibronectin does not initiate polyphosphoinositide hydrolysis, but amplifies hydrolysis once it has been initiated. Thus, cytotoxic T-lymphocyte activation involves a TCR-initiated cascade of adhesion and signalling events leading to response.