cytoekeletal polarization luciana

seminars in I M M U N OL OG Y, Vol 12, 2000: pp. 5]21 doi: 10.1006rsmim.1999.0203, available online at http:rrwww.ideali...

4 downloads 77 Views 790KB Size
seminars in I M M U N OL OG Y, Vol 12, 2000: pp. 5]21 doi: 10.1006rsmim.1999.0203, available online at http:rrwww.idealibrary.com on

Cytoskeletal polarization and redistribution of cellsurface molecules during T cell antigen recognition P. Anton van der Merwe 1U , Simon J. Davis 2, Andrey S. Shaw 3 and Michael L. Dustin 3

lymphocytes by, for example, chemotactic factors or ligation of antigen receptors results in increased adhesiveness,3 cytoskeletal polarization and the development of morphological asymmetry.4 It has long been recognized that cytoskeletal changes contribute to lymphocyte effector functions such as polarized cytokine secretion and target cell killing.5,6 This review focuses on the role of cytoskeletally-mediated processes in T cell antigen recognition. Asymmetry of activated lymphocytes also extends to the distribution of molecules on their surface. In migrating lymphocytes, chemokine receptors are concentrated at the leading edge, whereas molecules such as CD43, CD44 and CD50 are found at the trailing edge.4 Dramatic changes in cell-surface molecule distribution are also observed when the T lymphocytes recognise antigens on the antigen presenting cells ŽAPCs..5,7 ] 10 In this review we first describe the redistribution that occurs during T cell antigen recognition, placing it in the context of other cellular events such as changes in cell motility and morphology, increases in intracellular calcium, and polarization of the cytoskeleton. We then discuss in detail the molecular mechanisms underlying redistribution. Finally, we speculate on the functional significance of redistribution, and attempt to integrate the experimental data into a general model of T cell antigen recognition.

T cell antigen recognition is accompanied by cytoskeletal polarization towards the APC and large-scale redistribution of cell surface molecules into ‘supramolecular activation clusters’ (SMACs), forming an organized contact interface termed the ‘immunological synapse’ (IS). Molecules are arranged in the IS in a micrometer scale bull’s eye pattern with a central accumulation of TCR r peptide]MHC (the cSMAC) surrounded by a peripheral ring of adhesion molecules (the pSMAC). We propose that segregation of cell surface molecules on a much smaller scale initiates TCR triggering, which drives the formation of the IS by active transport processes. IS formation may function as a checkpoint for full T cell activation, integrating information on the presence and quality of TCR ligands and the nature and activation state of the APC. Key words: T cell receptor r T cell antigen recognition r SMAC r immunological synapse r segregation Q2000 Academic Press

Introduction LYMPHOCYTES ARE PROPELLED through the blood as relatively undifferentiated spheres uniformly covered with short microvilli, but without obvious asymmetry or polarity. These cells also relatively non-adhesive, partly because integrins such as LFA-1 are immobilized by attachment to the actin cytoskeleton and maintained in a low avidity state.1,2 Activation of

Observations Snapshots of cytoskeletal polarization and cell-surface molecule redistribution

From the 1Sir William Dunn School of Pathology, University of Oxford, Oxford OX1 3RE, UK, 2Nuffield Department of Clinical Medicine, John Radcliffe Hospital, University of Oxford, Headington OX3 9DU, UK and 3Centre for Immunology and the Department of Pathology, Washington University School of Medicine, U 660 South Euclid Avenue, St. Louis, MO 63110, USA. Corresponding author. Q2000 Academic Press 1044-5323r 00 r 010005q 17 $35.00r 0

Studies performed over a decade ago showed that, following recognition of the antigen on target cells or APCs, T cells become polarized with respect to the target cell or APC Žreviewed in ref 5.. The microtubule organising centre ŽMTOC., which is usually 5

P. Anton van der Merwe et al

6

Redistribution of T cell surface molecules

found in the trailing edge of a migrating T cell, repositions itself between the T cell nucleus and the target cell.11 This is accompanied by, and is probably responsible for, the movement of the Golgi apparatus to the same side of the cell.11 There is also an increase in actin polymerization adjacent to the contact interface, and the actin-binding protein talin accumulates in this area.12 The T cell receptor and its CD3 signalling module ŽTCRrCD3., the TCR coreceptor CD4, and the adhesion receptorrligand pair LFA-1rICAM-1 all accumulate at the contact interface between CD4 T cells and APCs presenting specific antigens.5 Much more recently, Wulfing and Davis13,14 have reported evidence for a more generalized, cytoskeletally-driven movement of cell surface molecules towards the T cellrAPC contact interface. Recently the redistribution of molecules within the contact interface has been visualized both in fixed T cellrAPC conjugates7,8 as well as in live T cells interacting with planar, glass-supported lipid bilayers into which fluorescently-labelled, lipid-anchored molecules have been incorporated.10,15 These bilayers are capable of stimulating T cell proliferation, suggesting that they can function as surrogate APCs. While less physiological than APCs, the planar bilayer system has provided higher resolution images and the first quantitative and dynamic data on this redistribution. Both approaches reveal that cell-surface molecules segregate within T cellrAPC contacts into distinct regions or clusters, referred to by Monks et al 8 as supramolecular activation clusters ŽSMACs., thereby forming an organized interface which we refer to as an immunological synapse10 ŽFigure 1.. The phrase immunological synapse ŽIS. was coined several years ago 6 to draw attention to some of the similarities between these contact areas and neuronal synapses.16 We have extended the term to describe the charac-

teristic molecular patterns formed in the process of physiological T cell activation.10 IS formation passes through several stages ŽFigure 2., culminating in the ‘mature’ IS ŽFigure 1.. In the mature IS the TCRrCD3 and several molecules involved in TCR triggering congregate at the centre of the interface.7,8,10 These include peptide]MHC, the accessory molecule CD28 and its ligand CD80, and cytoplasmic signalling molecules Že.g. lck, fyn, PKC-u .. This central area has been termed the central cluster or cSMAC.8 A second group of molecules, including LFA-1, ICAM-1 and talin, form a ring surrounding the cSMAC, which has been termed the ‘outer adhesion ring’ or peripheral SMAC ŽpSMAC..7,8,10 In the planar bilayer system10 the accessory molecule CD2 and its ligand CD48 appear to be confined to an ‘inner adhesion ring’. This surrounds the central zone but lies within the outer LFA-1 adhesion ring. Finally, CD43, a large, highly-abundant mucin-like molecule, is mostly excluded from the contact interface.9 Dynamics of polarization and redistribution Studies in live cells have provided insights into the timing and the sequence of events described above. T cell clones and hybridomas are highly mobile in cell culture, migrating at speeds of 4]10 m mrmin.4,17 When they encounter surrogate APCs Žusually cell lines transfected with suitable MHC molecules. that are not presenting specific antigens, they show no changes in motility or morphology.17,18 In contrast, when they encounter APCs presenting specific antigens they arrest, undergo shape changes, and expand their contact with the APC.17,18 T cells also migrate rapidly on artificial substrates coated with ICAM-1, but stop migrating when they encounter regions

Figure 1. Structure of the mature immunological synapse. Ža. Schematic view of the mature IS. The position of the IS is shown schematically above. The structure of the IS is shown below as viewed perpendicular to the contact interface. Not that some of the molecules are expressed on T cells and others on APCs. CD43 is excluded from the entire interface. ŽB. Fluorescent images revealing the position of the CD28rCD80 Žleft. and CD2rCD48 Žright. complexes in the IS. The left image shows the positions of peptide]MHC Žgreen., CD80 Žred. and ICAM-1 Žblue. at the interface between 3A9 T cells and planar bilayers. The right image shows the positions of peptide]MHC Žgreen., CD48 Žred. and ICAM-1 Žblue. at interface between 2B4 T cells and planar bilayers. The CD28rCD80 complex colocalizes with the TCRrpeptide]MHC whereas the CD2rCD48 complex segregates to an inner ring between TCRrpeptide]MHC and LFA-1rICAM-1. Data from Grakoui et al.10 Figure 2. Steps in the formation of the IS. ŽA. Fluorescent images illustrating the dynamics of IS formation. T cells are brought into contact with planar lipid bilayers at t s 0. The positions of engaged peptide]MHC Žgreen. and engaged ICAM-1 Žred. at the indicated times after initial contact are shown superimposed on an image of the contact area Žwhite.. Reproduced with permission from Grakoui et al.10 ŽB. The three stages of IS formation in planar lipid bilayer system. Reproduced with permission from Grakoui et al.10

7

P. Anton van der Merwe et al

coated with ICAM-1 and specific peptide]MHC complex.19 Together these data suggest that an encounter with specific antigens delivers a stop signal to migrating T cells, which serves to keep them associated with the APC that is presenting specific antigen.19 Naive T cells are less mobile but also show rapid changes in morphology and motility upon contact with B cells presenting specific antigens.20,21 These changes in motility and morphology are followed in a variable time Žseconds to minutes. later by increases, sometimes oscillating, in intracellular calcium concentration.17,18,22 A consistent feature in all the above studies is that changes in T cell motility, morphology and intracellular calcium concentration are only seen if the APC or planar bilayers present specific antigens, indicating that they require TCR triggering. In contrast, when dendritic cells are used as APCs, morphological changes, conjugate formation, and increases in calcium are seen in the absence of antigens,21 even in the absence of MHC class II.23 This suggests that, unlike surrogate APCs or B cells, dendritic cells have the ability to engage and activate naıve ¨ T cells prior to any TCR engagement, perhaps through the expression of high levels of the CD28 ligands CD80 and CD86 or chemokines.24,25 Technical innovations such as the use of green fluorescent protein chimeras13 and the planar bilayer system15 have made it possible to visualise the redistribution of cell surface molecules in live T cells. A recent study 10 using primary T cells in the planar lipid bilayer system has revealed that the formation of the IS takes several minutes following initial contact and TCR triggering, and can be divided into at least three stages ŽFigure 2.. Following contact with planar bilayers presenting specific antigens, the T cells arrest and within 30 s a central zone containing the LFA-1 ligand ICAM-1 is formed. Analysis by interference reflectance microscopy indicated that an area of close contact surrounds the central zone. An unexpected finding was that clusters of engaged peptide]MHC first appeared in this outer close contact ring. Since no other accessory molecules were present in this system we ŽA.S. and M.L.D.. proposed that the cells were using the LFA-1rICAM-1 interaction as an ‘anchor’ to force the membranes into close proximity. This raises the interesting possibility that active processes contribute to the membrane approximation that is necessary for the TCR to engage peptide]MHC complexes Žsee below..

Over the next 1.5]30 min the clusters of peptide]MHC in the outer ring moved towards the centre of the interface, forming a compact central cluster that is surrounded by a ring of ICAM-1 engaged to LFA-1. This organized interface, which is termed the ‘mature’ IS,10 is very similar to the pattern of SMACs described by Monks et al.8 The latter were visualized at the interface of T cells and specific antigens presenting APCs that had been fixed after 30 min of coculture. This transport of peptide]MHC is inhibited by cytochalasin D, suggesting that IS formation is dependent on the actin cytoskeleton.10 Using peptide variants it was shown that the final density of peptide]MHC complexes in the central cluster correlated best with the half-life of the TCRrpeptide]MHC interaction.10 A final stage of IS formation was analyzed by fluorescent photobleaching recovery ŽFPR. experiments. This revealed that, after formation, the central cluster of the engaged peptide]MHC complexes does not exchange with free peptide]MHC outside the bleached area. The reasons for this are currently unclear, and it is not known whether the TCR is also stably associated with this central cluster. One possible explanation for this stability is that peptide]MHC complexes, possibly with bound TCR, oligomerise directly with each other, forming a two-dimensional lattice.26 A second possibility is that a complex network of associations between TCRrCD3, the co-receptors CD4 or CD8, accessory molecules, and their various extracellular and intracellular ligands, is stabilized through the cooperative effects of multiple weak interactions. How can this stability be reconciled with other data27,28 suggesting that TCRr peptide]MHC complexes are continuously forming and dissociating during T cell antigen recognition? One proposed explanation10 is that there is a high turnover of TCRrpeptide]MHC complexes during the formation of the mature IS in the steps preceding stabilization of the central cluster. A key finding to emerge from these dynamic studies is that TCR triggering is a very early event, and that changes in morphology, motility, calcium concentration and, in particular, IS formation all follow, and are dependent on initial TCR triggering. This raises three important questions that will be addressed in the remainder of this review. How does the initial TCR triggering occur? How does this triggering drive formation of the IS? Given that it is not required for initial TCR triggering, what is the purpose of IS formation? 8

Redistribution of T cell surface molecules

Mechanisms

ment region’., precise alignment of membranes enables the CD2rCD58 interaction, which has a low solution affinity Že.g. K d ; 2 m M., to achieve a high two-dimensional affinity Ž2D K d ; 1 moleculer m m2 ..15,32 The presence of large molecules at the cell]cell interface would be expected to act as a constraint on TCR engagement of peptide]MHC by preventing sufficiently close membrane approximation and membrane alignment.37 Indeed, T cell antigen recognition is enhanced when CD43 is absent from T cells.38 Notably, however, many T cell ‘accessory’ molecules which contribute to T cell antigen recognition are also small ŽFigure 3.. These observation led to the suggestion30,39,40 that, in order for the TCR to engage peptide]MHC, molecules at the interface between T cells and APCs would have to segregate according to the size ŽFigure 4., with small accessory molecules contributing to the formation of ‘closecontact zones’, within which the inter-membrane separation distance Ž; 15 nm. matched the dimensions of the TCRrpeptide]MHC complex, and from which large molecules such as CD43, CD45 and integrins are excluded. Structural41 ] 44 and mutagenesis36 studies of CD2 and its ligands CD48 and CD58 have shown that these molecules bind in a head-to-head orientation, forming a complex that spans the same dimensions Ž; 14 nm. as the TCRrpeptide]MHC complex. On this basis it was proposed that one function of CD2 is to facilitate the formation of close-contact zones, thereby enhancing TCR engagement of peptide]MHC.36,39,40 The importance of close membrane approximation is strongly supported by the recent observation that, whereas formation of the wild-type Ž; 14 nm. CD2rCD48 complex strongly enhances TCR engagement of peptide]MHC, an elongated Ž; 21 nm. CD2rCD48 complex is strongly inhibitory.37

Overview The processes contributing to the redistribution and segregation of cell surface molecules fall into two groups. Firstly there are passive mechanisms which are dependent on properties intrinsic to the molecules and their interactions. These include interactions with counter-receptors at the contact interface, physical dimensions, and partition into lipid rafts. Although not active in the sense of consuming cellular energy, these mechanisms do produce physical forces. Secondly, redistribution can be driven by active cellular processes, such as polarized secretion or remodelling of the cortical actin cytoskeleton. Biophysical considerations Molecular dimensions, membrane alignment and two-dimensional affinity The molecular composition of the T cell surface has been intensively studied and is now reasonably well characterized.29 One striking observation30,31 is that these molecules vary enormously in size ŽFigure 3.. The extracellular portions of the TCR and its ligand peptide]MHC are small Ž; 7 nm. compared with several highly-abundant cell surface molecules such as CD43 Ž; 45 nm. and CD45 Ž; 28]50 nm., as well as important adhesion molecules such as integrins Ž; 21 nm.. It is conceivable that undulations in the plasma membrane could accommodate the size differences of interacting molecules. However, a second striking observation is that interactions between membrane-tethered adhesion molecules generate precise alignment of apposing membranes.15,32 Consequently, molecular interactions spanning different dimensions tend to segregate at the interface into distinct, homogenous zones.33 While the existence and significance of membrane alignment had been proposed earlier,34 ] 36 it was demonstrated for the first time by a comparison of the solution affinity and the membrane Žor two-dimensional. affinity of the CD2rCD58 interaction, using soluble and membrane-tethered forms of these molecules, respectively.15,32 The membrane affinity constant is two-dimensional because the concentration of membrane-tethered molecules is measured as surface density, a two-dimensional parameter Žmoleculesrmm2 .. This comparison revealed that, by confining both interacting binding sites to a small volume Ž; 5 nm high. between apposing membranes Žthe ‘confine-

Receptor lateral mobility When two cells make contact, complementary surface receptors will usually need to move laterally within the plane of the membrane in order to encounter each other. This movement will depend on their lateral mobility,45 a term which refers collectively to two distinct properties; the proportion of the molecule that is mobile Žthe mobile fraction., and the diffusion rates of the mobile molecules within the plane of the membrane Žthe lateral diffusion coefficient.. There is currently only limited information on the lateral mobility of T cell surface molecules. Approxi9

P. Anton van der Merwe et al

mately 75% of the CD2 molecules expressed in Jurkat cells were shown to be mobile,46 similar to observations with other molecules.47 This agreed with the

observation that a similar proportion of the CD2 accumulates in contact areas with CD58-bearing planar bilayers.15 In contrast, only 10]20% of the TCR

10

Redistribution of T cell surface molecules

on mature T cells is mobile.10 The reasons for this low mobility are not clear but may include cytoskeletal associations48,49 or diffusion barriers.50 The latter may be in the cytoplasm Že.g. cytoskeletal ‘fences’. or in the membrane itself Že.g. other membrane proteins or lipid rafts.. This low mobility needs to be reconciled with experiments suggesting that up to 80% of cell surface TCRrCD3 engages peptide]MHC during antigen recognition.27 One possible explanation is that the immobile TCRrCD3 48,49 is actively transported to the contact interface over a long period. Alternatively, the T cell may form multiple contacts Žsimultaneous or sequential. with several APCs.

Two distinct transport processes have been implicated in the redistribution of cell-surface molecules; polarized secretion of cytoplasmic vesicles containing the molecules, and redistribution of molecules already at the surface. Polarized secretion Both CD8 T cells 5,53 and CD4 T cells 5,54,55 are capable of polarized secretion, in which cytoplasmic vesicles fuse to the plasma membrane at the contact interface with target cells or APCs. This allows the contents of these vesicles Že.g. perforin, granzymes, and cytokines. as well as the vesicle membranes to be directed to the interface.5,53 ] 55 Recent data suggest that the T cell surface molecules FasL56 and CTLA-4 52 reside predominantly in cytoplasmic vesicular stores, and that these molecules are targeted to the site of TCR ligation by polarized secretion. There is some evidence that lipid rafts and their associated molecules Žsee below. are concentrated in intracellular, presumably vesicular, stores in naıve ¨ T cells.57,58 Polarized secretion is known to target rafts to the apical membrane of epithelial cells.59 This raises the question as to whether lipid rafts and associated molecules are transported to the IS by polarized secretion.

Active processes There is compelling evidence that active processes contribute to the redistribution of cell surface molecules during T cell antigen recognition. Firstly, much of the observed redistribution requires and follows TCR triggering.5,8 ] 10,13,14,51 Secondly, cytoskeletal inhibitors inhibit redistribution.10,13,14 Thirdly, a number of studies Žsee, e.g. refs 5,52. have shown that cross-linking of the TCRrCD3 complex with antibodies can trigger redistribution of TCRrCD3 Ž‘capping’. as well as other molecules Ž‘co-capping’. to one region Žthe ‘cap’. of the T cell surface. And finally, a cytoplasmic protein has been identified that interacts with the cytoplasmic domain of CD2 ŽCD2AP. and influences its distribution within a T cellrmembrane interface.33

Movement of molecules on the cell surface Several lines of evidence implicate the cortical actin cytoskeleton ŽCAC. in the redistribution of surface molecules to the contact interface. Firstly, treatment of T cells with cytoskeletal inhibitors abrogates the

Figure 3. Dimensions of molecules involved in T cell antigen recognition. A schematic view of the dimensions of cell-surface molecules involved in T cell antigen recognition. Sizes are estimated from structural studies of these or related molecules29 with the exception of CD148. The size of CD148 Ž; 55 nm. is estimated by adding 10 fibronectin type III domains strung end-to-end Ž; 40 nm. to an ; 80 amino acid mucin-like region Ž; 15 nm.. Figure 4. The kinetic]segregation model of TCR triggering and immunological synapse formation. The redistribution of cell surface molecules Ždepicted as in Figure 3. in T cell activation. The green and red boxes on the left are enlarged views of the boxes on the right. Ži. Before cell]cell contact the TCRrCD3 complex is subject to constitutive tyrosine phosphorylation and dephosphorylation. Dephosphorylation dominates, leading to a low steady-state level of tyrosine phosphorylation. Žii. Initial T cell contact with an APC leads the formation of multiple close-contact zones within the contact area as a result of small-scale segregation of molecules. This is a passive process driven by size and ligand binding. Within these close-contact zones tyrosine phosphorylation is favoured because tyrosine phosphatases Že.g. CD45. are excluded, and tyrosine kinases are concentrated. Žiii. Engagement by TCR of specific-peptide]MHC occurs within these close-contact zones and results in the trapping of the TCRrCD3 complex within a zone for a period, the length of which is determined primarily by the stability of the TCRrpeptide]MHC complex. While in the zone, TCRrCD3 is tyrosinephosphorylated, initiating the multi-step signalling process required for triggering. Živ. A sufficiently stable TCRrpeptide]MHC interaction Žcomplex II. allows these signalling steps to be completed, leading to triggering. Unstable complexes ŽI and III. lead to no or partial triggering because, on leaving the close-contact zone, TCRrCD3, and any recruited signalling molecules, are re-exposed to high levels of tyrosine-phosphatase activity. Žv. TCR triggering initiates active large-scale segregation of molecules, culminating in the formation of the mature IS. Žvi. Steps Žii. to Žv. are enhanced by ‘co-stimulatory signals’ such as chemokines or the expression of CD28 ligands on APCs. Žvii. Sustained signalling through the IS results in full T cell activation.

11

P. Anton van der Merwe et al

redistribution of ICAM-1 and other molecules to the interface13,14 and inhibits transport of peptide]MHC within the interface.10 Secondly, a large number of cell surface molecules which migrate towards and away from the interface have been shown to associate with the cytoskeleton or with linker molecules which themselves associate with the cytoskeleton Žsee below.. Finally, ligation of the TCR leads to localized talin accumulation5 and actin polymerization.12,60,61 How might the CAC organise the redistribution of cell surface molecules upon T cell activation? The CAC moves in a centripetal direction within the cellular processes that extend from cells, as a result of a combination of actin filament remodelling and myosin motor proteins.62 Following recognition, T cells extend analogous processes around the APC, forming a cup-shape interface. Centripetal movement of the CAC within this ‘cup’ might move molecules coupled to the CAC, such as TCRrCD3,48,49 radially towards the centre of the contact area. How are some molecules Že.g. CD2 and LFA-1. prevented from moving into the centre and others Že.g. CD43. actively excluded? Possible explanations include the absence of links to the CAC, physical exclusion on the basis of size, and attachment to other cytoskeletal structures.

are discussed in detail in the accompanying reviews by Janes et al and Zhang and Samelson, respectively.

Linking TCR triggering to cytoskeletal polarization and receptor redistribution Role of TCR triggering Anti-TCR antibodies can induce both re-orientation of the MTOC,61,70 polarized exocytosis52 and localized actin polymerization,61,70 suggesting that TCR triggering may be sufficient for these processes. However, an important caveat is that these experiments were performed in the presence of serum Ža source of extracellular matrix proteins. or intact target cells, leaving open the possibility that integrin engagement is an additional requirement.71 As far as is known, TCR triggering is necessary for the re-orientation of the MTOC and Golgi apparatus towards the T cellrAPC contact interface.5,19,70 In contrast, localized actin polymerization does not require TCR triggering.70 An interaction between LFA-1 on T cells and ICAM-1 on APCs is sufficient to induce cortical actin polymerization and talin accumulation at the contact site.70 Anti-TCR antibodies can also induce redistribution of cell surface molecules, suggesting that TCR triggering is also sufficient for this to occur.5,52 It is unclear whether TCR triggering is absolutely necessary for active redistribution of cell surface molecules. The active redistribution described to date at T cell contact interfaces has always been preceded by TCR ligation.5,7,8,10 However, the various antigen-independent responses seen when T cells contact dendritic cells raise the possibility that TCR triggering may not be essential for redistribution at T cellrdendritic cell interfaces.21,23,24

Lipid rafts The plasma membrane of many cells,63 including T cells,64,65 contains microdomains which are enriched in glycosphingolipids and cholesterol, termed lipid rafts. Some T cell surface molecules important in TCR triggering Že.g. lck, fyn, LAT, CD48 and CD4. are preferentially associated with these structures64 ] 66 . Others, most notably CD45 64,67 appear to be depleted from lipid rafts. This being so, the redistribution of lipid rafts could contribute to the segregation of cell surface molecules at the T cellrAPC interface. Recent data suggest that triggering of the TCR can induce active redistribution of lipid rafts towards the site of TCR triggering.57 If cell-surface molecules that associate with lipid rafts, such as the GPI-anchored molecule CD48, bind to a ligand on APCs, this could provide a passive mechanism for the accumulation of rafts at the interface.66 The importance of lipid rafts in TCR triggering is highlighted by the finding that chemical disruption of rafts abrogates TCR triggering.64,68 Furthermore, mutations which disrupt the association with rafts of the essential signalling molecules LAT 69 or lck 51 also disrupt TCR triggering. The role of lipid rafts and LAT in TCR signalling

Role of accessory molecules: a mechanism of co-stimulation? Recent studies have implicated accessory molecules in the active redistribution of T cell surface molecules. Wulfing and Davis showed that bulk redistribution of T cell surface molecules to the T cellrAPC interface is inhibited when CD28rCD80 or LFA-1rICAM-1 interactions are blocked.14 Viola et al found that the redistribution of lipid rafts to the area in contact with anti-CD3 antibody-coated beads is dramatically enhanced when the beads are also coated with anti-CD28 antibodies.57 These findings suggest that accessory molecules may enhance T cell activation by contributing to this redistribution process. The ability of the CD28rligand interaction to activate Rho family 12

Redistribution of T cell surface molecules

GTPases Žsee below. independently of TCR triggering support this notion.72 As noted above, the LFA1rICAM-1 interaction can trigger changes in the CAC at the contact site.70 The finding that cytoskeletal polarization and large-scale receptor redistribution depend on TCR triggering, and are enhanced by engagement of accessory molecules such as CD28, suggests that IS formation is the point at which the T cell integrates information about the presence of the antigen on the APC and its activation state Že.g. expression of chemokines or CD28 ligands.. Or, to rephrase in the terminology of the two-signal hypothesis of T cell activation, IS formation may integrate signal 1 Žstimulus from antigen. with signal 2 Žco-stimulus from the APC.. This is supported by the finding that dendritic cells, the most effective APCs, are uniquely capable of adhering to and inducing antigen-independent responses in T cells.21,23 ] 25

Thus, the picture emerging is that TCR triggering activates Rho family GTPases through activation of proteins such as vav, and that Rho GTPases regulate various effector proteins, including protein kinases and WASp, which regulate in turn actin binding proteins. Direct links between vav1 and the actin73 and microtubule80,81 cytoskeleton have also been identified. Since the cytoskeleton and associated motor proteins control the active transport processes implicated in receptor redistribution, the vavrRho pathway is likely to couple TCR triggering to cytoskeletal polarization and the redistribution of cellsurface molecules. Linking the cytoskeleton to T cell surface molecules For TCR-induced changes in the CAC to contribute the redistribution of cell-surface molecules, these molecules need to associate with the CAC. Linkages have been reported for LFA-1, which associates via talin,82 CD2, which associates with CD2AP 33 and tubulin,83 the TCR, which associates via the CD3z chain,48,49 and CD45, which associates with the linker protein fodrin.84 Furthermore, members of the ezrin-radixin-moesin ŽERM. family of proteins couple CD43, CD44, and ICAM-1, -2, and -3 to the CAC.85 With the exception of CD2AP Žsee below., the role of these linker molecules in surface molecule redistribution is not known. However, their diversity provides a possible explanation for the distinct localization of cell-surface molecules within the IS.

Signalling pathways Recently a tentative outline has emerged of the signal transduction pathways that couple TCR triggering to cytoskeletal polarization and receptor redistribution.72,73 Attention has focused on the Rho family of small GTP binding proteins Žespecially RhoA, Cdc42Hs and Rac1. since they have been shown to regulate cytoskeletal and vesicle transport processes in other cells.72,74 Using a T cell hybridoma, Stower et al 60 have shown that a dominant-negative mutant of Cdc42Hs inhibits actin and microtubule polarization towards the T cellrAPC interface. How does TCR triggering activate rho GTPases? The most likely candidate is the GTPrGDP exchange factor vav1, which activates Rac1 and has been reported to be essential for TCR-induced actin polymerization and capping.75 ] 77 Vav1 is activated by tyrosine phosphorylation andror by binding to polyphosphoinositide products generated following activation of phosphatidylinositol ŽPI. 3-kinase.73 The mechanisms by which the rho GTPases regulate the cytoskeleton is not well-understood but is likely to involve effector proteins which regulate actin-binding proteins, such as the protein kinases ROCK and LIM-kinase.72,78 An effector protein already implicated in coupling Rho GTPases to the cytoskeleton in T cells is WASp, so named because it is defective in patients with Wiscott-Aldrich immunodeficiency syndrome.79 WASp interacts with Cdc42Hs and is required for capping induced by anti-CD3 antibodies, suggesting that it couples Cdc42Hs activation to the actin cytoskeleton.

CD2AP CD2AP was first identified as a molecule that binds to the cytoplasmic domain of CD2 and influences its distribution at the contact interface between a T cell and planar lipid bilayers.33 The finding that CD2rCD48 complexes occupy a special position in the IS10 is further evidence that CD2 is being actively moved at the interface, and it seems likely that this movement is also mediated by CD2AP. A recent study on CD2AP-deficient mice suggests that this molecule has a wider role in lymphoid and non-lymphoid tissues.86 These mice die from complications of excessive leakage of serum proteins into the urine Žnephrotic syndrome..86 This leakage appears to result from defects in the epithelial cell Žpodocyte. processes that contribute to the glomerular filtration barrier. Interestingly, CD2AP-deficient T cells show a substantial reduction in response to anti-CD3 antibodies,86 whereas CD2-deficient mice 13

P. Anton van der Merwe et al

triggering can be divided into aggregation,88,89 conformational change,90,91 or segregation39,40,92,93 models. Segregation models have in common the proposal that peptide]MHC binding leads to triggering by holding the TCRrCD3 complex in a membrane environment in which tyrosine kinases have been segregated from tyrosine-phosphatases. Kinetic proof-reading,94 kinetic discrimination95 and serial triggering 27 models do not fall neatly into the above classification because they do not address the central question of how binding transduces a signal. Instead they try to account for particular features of TCR triggering, such as the ability of the TCR to discriminate between very similar ligands94,95 and the ability of a few peptide]MHC complexes to engage many TCRs.27 Consequently they can and have been incorporated into aggregation, conformational change, or segregation models.

are normal in this respect, suggesting that CD2AP plays a broader role in T cell function than CD2. What might be the purpose of locating CD2rCD48 complexes in an inner ring surrounding the TCRrpeptide]MHC-containing central zone? One intriguing possibility, consistent with a generalized role for CD2AP in forming filtration junctions, is that this ring of CD2rCD48 functions as a molecular filter within the IS keeping larger molecules out of the central zone in the face of centripetal actinomyosinbased transport processes.10,14

Functions What is the purpose of cytoskeletal polarization and redistribution of cell surface molecules? One probable role for cytoskeletal polarization is to facilitate the targeting by T cells of cytokines and the contents of lytic granules to the site of contact with APCs or target cells, respectively.5,55 As argued above, cytoskeletal polarization is also likely to contribute to the large-scale redistribution of cell surface molecules leading to the formation of the IS. Here we speculate on the role of IS formation in T cell activation, placing it in the context of a recently-proposed model of TCR triggering.39,40 It is important to re-emphasise here that the formation of the IS follows and depends on TCR triggering, indicating that it cannot be the mechanism of initial TCR triggering. Nevertheless IS formation does appear to be required for full T cell activation. What is the evidence for this? Firstly, in the few studies performed to date, IS formation correlated perfectly with full T cell activation.8,10 Secondly, disruption of the actin cytoskeleton20,87 or the regulation thereof,75 ] 77 manipulations which would be expected to disrupt IS formation, leads to disruption of full T cell activation Žproliferation andror cytokine production. and negative selection. While not conclusive, these findings are consistent with an essential role for the IS in full T cell activation.

Problems with aggregation and conformational change models Aggregation models postulate that peptide]MHC engagement leads to clustering or aggregation of TCRrCD3 complexes. At first glance this is an attractive notion since dimerizationraggregation is a wellestablished mechanism of signal transduction,96 and it is well known that aggregation of TCRrCD3, with either antibodies or multivalent peptide]MHC,97 will trigger TCRs. However, there are reasons for doubting that this is the physiological mechanism for TCR triggering.92 In systems in which dimerizationraggregation lead to signalling, the mechanism of multimerization is usually clear;96 a multivalent ligand Žeither a monomer with two binding sites or a multimer. binds two or more receptors, thereby bringing them into close proximity. In the case of the TCR the ligand, peptide]MHC, is monovalent, and present at an exceptionally low density, with as few as 10]100 molecules of a specific peptide]MHC on an entire cell. Thus, only a very small number of TCRs will be bound at any one time. Furthermore, each interaction is likely to be very short-lived, with a half-time of a few seconds.89 This suggests that simultaneous engagement of adjacent TCRs Ži.e. aggregation. will be exceedingly rare. Furthermore, if TCR triggering depended on aggregation it should be very sensitive to the surface density of specific peptide]MHC, whereas the converse is true Žsee, e.g. refs 10,98.. A refinement of the aggregation model, which attempts to address these problems, proposes that the binding of TCR to peptide]MHC generates a new binding site that enables TCRrpeptide]MHC complexes to self-

Small-scale segregation provides a novel mechanism for signal transduction across membranes TCR triggering requires binding of the TCR to specific peptide presented on MHC class I or II molecules. One of the key, unresolved questions in T cell biology is how this binding generates a signal. In other words how does the T cell ‘know’ that a particular TCR has bound peptide]MHC? Based on the way they address this question, current models of TCR 14

Redistribution of T cell surface molecules

associate.26 However, given the huge diversity in TCR and peptide]MHC structures, and the resulting diversity in the structure of TCRrpeptide]MHC complexes,99 it is difficult to envisage how such binding sites could be conserved in all TCRrpeptide]MHC complexes. A further difficulty is understanding how direct self-association occurs on the cell surface when each TCR is surrounded by several other molecules Že.g. CD3 subunits and a co-receptor., all of which are heavily glycosylated. A variant of the aggregation model is that triggering is a consequence of heterodimerization of the TCR with its co-receptor ŽCD4 or CD8.. While this may be sufficient for TCR triggering in certain circumstances,100 it cannot be the universal mechanism of triggering, since TCR triggering in vitro and in vivo 101,102 can occur in the complete absence of CD4 or CD8. Conformational change models propose that peptide]MHC binding leads to a conformational change in the TCR, which is somehow transduced into the cell interior. This model requires that every TCR is able to undergo the same binding-induced conformational change in the face of huge diversity in the structure of the antigen binding sites and peptide]MHC, which would seem highly implausible. Furthermore, structural data available to date show that there are no conformational changes in the TCR induced by binding peptide]MHC except for adjustments within the antigen-binding site.91,99 Taking the latter data into account, Wiley et al 91 recently proposed a variant of the conformational change model based on studies of signalling through the erythropoetin receptor.103 In this model the TCR exists as a preformed dimer. When both TCRs in the dimer bind to peptide]MHC the two TCRs, reposition with respect to each other, and this repositioning is detected in some way. Unfortunately this model has the same drawback as the aggregation models when applied to TCR triggering Žsee above.. A further difficulty is envisaging how the independent binding of two specific peptide]MHC complexes could change the relative positions of the two TCRs?

the same mechanism.40 These models emerged from several key observations described below.39,40 Firstly, the TCR and T cell accessory molecules are much smaller than other abundant cell surface molecules, including CD43, CD45, and integrins30,31 ŽFigure 3.. Taken together with the observation that membrane alignment is important for optimal binding of membrane-tethered molecules,15,32 this suggested that these molecules must segregate according to size at the contact interface.36,39,40 Secondly, tyrosine phosphatase inhibitors can activate T cells,104,105 producing tyrosine phosphorylation patterns very similar to those seen during TCR triggering.106,107 Furthermore, an inhibitor of src kinases ŽPP1. rapidly and dramatically decreases the level of basal tyrosine phosphorylation in T cells.57 These findings show that the tyrosine kinases implicated in TCR triggering are constitutively active and under tonic inhibitory control by tyrosine phosphatases, and imply that altering the local balance between these two competing, constitutive processes would be sufficient to trigger intracellular signals. Thirdly, one of the most active and abundant tyrosine phosphatases is the large cell surface molecule CD45.108 Furthermore, CD45 may associate directly or indirectly Žvia CD45-associated protein. with the TCR and its signalling apparatus.109 ] 112 This suggested that, in addition to its generally accepted role maintaining lck in a primed state,93 CD45 may also tonically inhibit TCR triggering by dephosphorylating key tyrosine kinase substrates such as CD3z ,109 ZAP-70 106 and lck itself.113 There is clear evidence that membrane-associated tyrosine phosphatases other than CD45 also reverse constitutively active tyrosine kinases.107 One possible candidate is CD148, which has recently been shown to inhibit TCR triggering.114 Interestingly, like CD45, CD148 is predicted to have a very large extracellular domain ŽFigure 3.. Finally, the TCR is able to discriminate between, and produce very different responses to, very similar ligands.115 This is difficult to explain purely on the basis of differences in affinity; all else being equal, a twofold change in affinity will produce a less than twofold change in the total number of engaged TCRs. In contrast, small kinetic differences can readily be amplified to produce very different biochemical responses.94,95 These observations were synthesized into a new model of TCR triggering,39,40 which we now refer to as the kinetic]segregation model to emphasise its two key elements, molecular segregation and binding ki-

The kinetic-segregation model of TCR triggering The arguments outlined above against conformational change or dimerizationraggregation models led to the proposal that an entirely novel mechanism, based on molecular segregation in contact areas, was responsible for TCR triggering.39 A related ‘topological model’, based on the implications of new data on two-dimensional affinity,32 subsequently converged on 15

P. Anton van der Merwe et al

netics. The model postulates the following sequence of events in TCR triggering ŽFigure 4.. When T cells encounter target cells or APCs, molecules at the contact interface will segregate according to size. Small molecules such as CD2, CD4, CD8 and CD28 will contribute to the formation of multiple, small close-contact zones, from which large molecules such as CD45, CD43, CD148 and LFA-1 are excluded. As a result of this segregation, close-contact zones will be a ‘pro-signalling’ environment, relatively depleted of tyrosine phosphatases Žsuch as CD45 and perhaps CD148. and enriched in tyrosine kinases such as lck. The close membrane approximation within the close-contact zone will facilitate TCR encounters with specific peptide]MHC. The binding of TCR to specific peptide]MHC will result in the TCR and its associated CD3 complex being held within a close contact zone, where they show a net increase in phosphorylation as a result of exposure to tyrosine kinases such as lck in the absence of tyrosine phosphatases such as CD45. It is important to emphasise here that ligation of a single TCRrCD3 can generate a signal Žno dimerization or aggregation is required.. Furthermore, tyrosine phosphorylation of TCRrCD3 is not strictly dependent on its localization to a ‘prosignalling’ close contact zone, given that it occurs constitutively Žsee above.. It is the downstream signalling events that are crucially dependent on localization to close-contact zones because this prolongs the half-life of tyrosine phosphorylated species by segregating the TCRrCD3 and associated molecules from tyrosine phosphatases. This tyrosine phosphorylation initiates a cascade of events including recruitment and activation of the tyrosine kinase ZAP-70 and phosphorylation and activation of ZAP-70 substrates.116 This sequence will be interrupted when the TCRrCD3 complex and associated molecules leaves such a close-contact zone. Because the TCRrCD3 is held within the contact zone through binding to peptide]MHC, tyrosine phosphorylation and TCR triggering will depend crucially on the half-life of the TCRrpeptide]MHC interaction117 and a decrease in the half-life will lead to interruption of the cascade of signalling events.118,119 To provide kinetic discrimination the model requires that initial close-contact zones be relatively small or short-lived so that inappropriately phosphorylated TCRrCD3 complexes do not have to diffuse far prior to their dephosphorylation by CD45 or other excluded phosphatases. This may be a key difference between the respective signalling roles of

the initial close-contact zones and the IS, which forms over a much larger area. It also seems likely that there will be many such close-contact zones forming within the contact area between T cells and APCs. Unfortunately, given the limited resolution of the light microscope, the existence of such small closecontact zones remains speculative. Recent studies relevant to the kinetic]segregation model Since the kinetic]segregation model was first proposed a number of findings have provided support for key elements of the model. First, mice with reduced levels of CD45 expression show evidence of enhanced signalling through the TCR,120 consistent with an inhibitory effect of CD45. Second, CD45 has been shown to inhibit the function of constitutivelyactive lck by dephosphorylating its substrates.121 Third, studies on planar bilayer systems have shown that engaged CD2 and LFA-1, which differ considerably in size ŽFigure 3., segregate in the contact interface independent of cytoskeletal interactions.33 This strongly supports the proposal that size differences drive the segregation of molecules within the contact interface. Fourth, it has been demonstrated that increasing the dimensions of an accessory molecule strongly inhibits TCR engagement.37 Fifth, a form of CD45 with a truncated extracellular domain has been shown to inhibit TCR triggering by antigens ŽC. Irles, A. Symons, F. Michel, P.A. van der Merwe, and O. Acuto, unpublished observations.. Sixth, T cells are able to discriminate between peptide]MHC ligands with similar affinities but with differences in half-life for TCR binding.89,117 Finally, as with early TCR signalling events,117 the number of peptide]MHC complexes that accumulated in the central region of the IS correlated strongly with the half-life of the TCR binding.10 This provided the first direct evidence that IS formation is dictated by kinetics of the TCRrpeptide]MHC interaction,10 most likely through the dependence of large-scale segregation on early TCR signalling events that are triggered, we suggest, by small-scale segregation. TCR triggering is inhibited by cell-surface receptors such as CTLA-4 and Killer cell Immunoglobulin Receptors ŽKIRs., which associate with the cytoplasmic tyrosine phosphatases SHP-1 and SHP-2.122,123 Interestingly, the complexes formed by CTLA-4 and KIRs and their ligands ŽCD80rCD86 and MHC class I, respectively. are predicted to have dimensions compatible with their co-localization with the TCR in contact areas ŽFigure 3.. Furthermore, CTLA-4 inhibits the earliest tyrosine phosphorylation events fol16

Redistribution of T cell surface molecules

lowing TCR triggering.124,125 This suggests that inhibitory receptors may function by altering the balance between tyrosine phosphorylation and dephosphorylation in the immediate vicinity of the TCRrCD3. As noted earlier, CTLA-4 is found predominantly in vesicular stores and is directed to the surface at the site of TCR triggering by polarized secretion.52 This suggests that CTLA-4 may be targeted to the centre of the IS as a means of modulating or terminating TCR signalling. The kinetic]segregation model is compatible with evidence that lipid rafts play a crucial role in TCR triggering Žsee above.. Lipid rafts may accumulate in or adjacent to the close-contact zones through either active 57 or passive mechanisms,66 helping to contribute to the ‘pro-signalling’ environment by recruiting tyrosine kinases and other pro-signalling molecules and excluding CD45 Žsee accompanying review by Janes et al.67 Based on these recent data, new models of TCR triggering have been proposed which postulate that TCR binding to peptide]MHC brings the TCR and lipid rafts together.92,93 These ‘lipid raft models’ are similar in principle to the kinetic]segregation model in that they also propose that triggering is a consequence of the redistribution of the TCR into an environment enriched in tyrosine kinases and depleted of tyrosine phosphatases. These models may, however, need to be adjusted since recent evidence suggests that TCRs may be constitutively associated with rafts Žsee Janes et al .. The main drawback of lipid raft models is that they do not suggest a clear mechanism, independent of aggregation or conformational change, by which engagement of the TCR with peptide]MHC either enhances its association with lipid rafts or promotes their aggregation.

viewed in ref 126; see also refs 127,128.. These same peripheral T cells require a much higher affinity TCR ligand in order to become fully activated. How is the TCR able to respond to ligands with such a wide range of affinities and yet retain the ability to discriminate between ligand which differ only slightly in their binding properties? The kinetic]segregation model would allow very weak interactions to be detected because TCRrCD3 complexes need only be transiently detained in close-contact zones to become phosphorylated. Thus, tyrosine phosphorylation could occur even with very weak TCR binding to peptide]MHC. Perhaps this is sufficient to generate the signals required for positive selection in the thymus or survival in the periphery. The finding that ZAP-70 is constitutively associated with partially-phosphorylated TCR-z in peripheral T cells, but that this is reduced in T cells from mice lacking self-peptide]MHC and absent in T cells grown in culture, is consistent with continuous weak signalling through the TCR.126 Because it would be inappropriate for peripheral T cells to be fully-activated by self-peptide]MHC, a mechanism must exist that enables a T cell to clearly distinguish between the signals resulting from weak versus strong self-peptide]MHC engagement. We suggest that the IS provides such a mechanism because it appears only to form when TCR binds peptide]MHC above a threshold affinity or half-life.10 Nevertheless the mechanisms that leads to IS formation are exquisitely sensitive since these structures form at very low surface densities of peptide]MHC.10 Taken together, these findings suggest that IS formation may function as a checkpoint for full T cell activation, facilitating discrimination between low affinity Žpresumably self-. peptide]MHC engagement and high affinity Žpresumably foreign-. peptide]MHC engagement. Why might IS formation be required for full T cell activation? There are at least two possibilities. IS formation may lead to the generation of a unique signal. A candidate for such a signal is activated protein]kinase C-u , which selectively accumulates in the central portion of the IS or cSMAC 7 Žsee accompanying review by Kativar and Mochly-Rosen.. A second possibility is that IS formation may be necessary for the prolonged Ž) 1 h. triggering necessary for full T cell activation.129 If the IS is required for full T cell activation, it follows that any factor that contributes to IS formation will contribute to full T cell activation. Like full T cell activation, IS formation requires that the

The immunological synapse as an activation checkpoint and signal integrator In addition to being structurally plausible, any model of TCR triggering has to explain a central paradox. The TCR detects peptide]MHC ligands with a wide range of affinities and surface densities but can nevertheless discriminate between peptide]MHC complexes with very subtle differences in binding properties.115,117,126 The ability of a given TCR to respond to a wide range of affinities is clearly illustrated by the ability of thymocytes to undergo positive and negative selection using the same TCR. More recently it has been shown that mature peripheral T cells also recognise and respond to self-peptide]MHC Žre17

P. Anton van der Merwe et al

3. Dustin ML, Springer TA Ž1989. T-cell receptor cross-linking transiently stimulates adhesiveness through LFA-1. Nature 341:619]624 4. Sanchez-Madrid F, del Pozo MA Ž1999. Leukocyte polarization in cell migration and immune interactions. EMBO J 18:501]511 5. Kupfer A, Singer SJ Ž1989. Cell biology of cytotoxic and helper T cell functions: immunofluorescence microscopic studies of single cells and cell couples. Annu Rev Immunol 7:309]337 6. Paul WE, Seder RA Ž1994. Lymphocyte responses and cytokines. Cell 76:241]251 7. Monks CR, Kupfer H, Tamir I, Barlow A, Kupfer A Ž1997. Selective modulation of protein kinase C-theta during T-cell activation. Nature 385:83]86 8. Monks CR, Freiberg BA, Kupfer H, Sciaky N, Kupfer A Ž1998. Three-dimensional segregation of supramolecular activation clusters in T cells. Nature 395:82]86 9. Sperling AI et al Ž1998. Cutting edge: TCR signaling induces selective exclusion of CD43 from the T cell]antigen-presenting cell contact site. J Immunol 161:6459]6462 10. Grakoui A et al Ž1999. The immunological synapse: a molecular machine controlling T cell activation. Science 285: 221]227 11. Geiger B, Rosen D, Berke G Ž1982. Spatial relationships of microtubule-organizing centers and the contact area of cytotoxic T lymphocytes and target cells. J Cell Biol 95:137]143 12. Ryser JE, Rungger-Brandle E, Chaponnier C, Gabbiani G, Vassalli P Ž1982. The area of attachment of cytotoxic T lymphocytes to their target cells shows high motility and polarization of actin, but not myosin. J Immunol 128: 1159]1162 13. Wulfing C, Sjaastad MD, Davis MM Ž1998. Visualizing the dynamics of T cell activation: intracellular adhesion molecule 1 migrates rapidly to the T cellrB cell interface and acts to sustain calcium levels. Proc Natl Acad Sci USA 95:6302]6307 14. Wulfing C, Davis MM Ž1998. A receptorrcytoskeletal movement triggered by costimulation during T cell activation. Science 282:2266]2269 15. Dustin ML, Ferguson LM, Chan P, Springer TA, Golan DE Ž1996. Visualization of the CD2 interaction with LFA-3 and determination of the two-dimensional dissociation constant for adhesion receptors in a contact area. J Cell Biol 132: 465]474 16. Norcross MA Ž1984. A synaptic basis for T-lymphocyte activation. Ann Immunol ŽParis. 135D:113]134 17. Negulescu PA, Krasieva TB, Khan A, Kerschbaum HH, Cahalan MD Ž1996. Polarity of T cell shape, motility, and sensitivity to antigen. Immunity 4:421]430 18. Donnadieu E, Bismuth G, Trautmann A Ž1994. Antigen recognition by helper T cells elicits a sequence of distinct changes of their shape and intracellular calcium. Curr Biol 4:584]595 19. Dustin ML, Bromley SK, Kan Z, Peterson DA, Unanue ER Ž1997. Antigen receptor engagement delivers a stop signal to migrating T lymphocytes. Proc Natl Acad Sci USA 94: 3909]3913 20. Delon J, Bercovici N, Liblau R, Trautmann A Ž1998. Imaging antigen recognition by naive CD4q T cells: compulsory cytoskeletal alterations for the triggering of an intracellular calcium response. Eur J Immunol 28:716]729 21. Delon J, Bercovici N, Raposo G, Liblau R, Trautmann A Ž1998. Antigen-dependent and -independent Ca2q responses triggered in T cells by dendritic cells compared with B cells. J Exp Med 188:1473]1484 22. Wulfing C et al Ž1997. Kinetics and extent of T cell activation as measured with the calcium signal. J Exp Med 185:1815]1825 23. Montes M, McIlroy D, Hosmalin A, Trautmann A Ž1999.

TCRrpeptide]MHC interaction affinity or half-life is above a threshold value.10 There is also intriguing evidence that ligation of accessory molecules such as CD28 and LFA-1 enhances some of the transport processes that lead to IS formation.14,57 Interactions with these molecules are regulated by soluble mediators Že.g. chemokines will activate LFA-1. and will depend on the activation or maturation state of the APC.24,25 Finally, there is evidence that dendritic cells are distinct from other APCs in their ability adhere to and evoke changes in naıve ¨ T cells in the absence of and prior to antigen recognition.21,23,24 These data suggest that the formation of the IS serves to integrate information relating to the presence and quality of the antigen Že.g. surface density and binding properties. and information on the nature and activation state of the APC.24,130 Based on these considerations we suggest the following integrated model of full T cell activation ŽFigure 4.. Small scale-segregation, driven primarily by ligand binding and steric factors, leads to initial TCR triggering, by the mechanism outlined in the kinetic]segregation model. This is sufficient to stimulate certain T cell functions, the nature of which Žpositive selection or survival. depends on the developmental state of the T cell. Full activation, however, requires the formation of the IS, which involves active transport processes. IS formation requires a threshold level of triggering through the TCR and is influenced by the nature and activation state of the APC.

Acknowledgements P.A.V. is supported by the Medical Research Council and S.J.D. by the Wellcome Trust. M.L.D. is supported by the Arthritis Foundation, The Whitaker Foundation and The National Institutes of Health. A.S. is supported by the National Institutes of Health. We thank Don Mason, Neil Barclay, Marion Brown, Matt Thomas, and Talitha Bakker for critical comments on the manuscript. We thank S. Bromley and C. Sumen for the use of images in Figure 2.

References 1. Kucik DF, Dustin ML, Miller JM, Brown EJ Ž1996. Adhesion activating phorbol ester increases the mobility of leukocyte integrin LFA-1 in cultured lymphocytes. J Clin Invest 97:2139]2144 2. Dustin ML Ž1998. Making a little affinity go a long way: a topological view of LFA-1 regulation. Cell Adhes Commun 6:255]262

18

Redistribution of T cell surface molecules

24. 25. 26. 27. 28. 29. 30. 31. 32.

33. 34. 35. 36.

37. 38. 39. 40. 41. 42. 43. 44. 45. 46.

47.

Calcium responses elicited in human T cells and dendritic cells by cell]cell interaction and soluble ligands. Int Immunol 11:561]568 Steinman RM Ž1991. The dendritic cell system and its role in immunogenicity. Annu Rev Immunol 9:271]296 Banchereau J, Steinman RM Ž1998. Dendritic cells and the control of immunity. Nature 392:245]252 Reich Z et al Ž1997. Ligand-specific oligomerization of T-cell receptor molecules. Nature 387:617]620 Valitutti S, Muller S, Cella M, Padovan E, Lanzavecchia A ¨ Ž1995. Serial triggering of many T-cell receptors by a few peptide]MHC complexes. Nature 375:148]151 Beeson C et al Ž1996. Early biochemical signals arise from low affinity TCR]ligand reactions at the cell]cell interface. J Exp Med 184:777]782 Barclay AN et al Ž1997. The leucocyte antigen factsbook. Academic Press, London Springer TA Ž1990. Adhesion receptors of the immune system. Nature 346:425]434 Barclay AN et al Ž1993. The leucocyte antigen factsbook. Academic Press, London Dustin ML et al Ž1997. Low affinity interaction of human or rat cell adhesion molecule CD2 with ligands aligns adhering membranes to achieve high physiological affinity. J Biol Chem 272:30889]30898 Dustin ML et al Ž1998. A novel adaptor protein orchestrates receptor patterning and cytoskeletal polarity in T cell contacts. Cell 94:667]677 Singer SJ Ž1992. Intercellular communication and cell]cell adhesion. Science 255:1671]1677 Karjalainen K Ž1994. High sensitivity, low affinity}paradox of T-cell receptor recognition. Curr Opin Immunol 6:9]12 van der Merwe PA, McNamee PN, Davies EA, Barclay AN, Davis SJ Ž1995. Topology of the CD2]CD48 cell-adhesion molecule complex: implications for antigen recognition by T cells. Curr Biol 5:74]84 Wild MK et al Ž1999. Dependence of T cell antigen recognition on the dimensions of an accessory receptor]ligand complex. J Exp Med 190:31]41 Manjunath N, Correa M, Ardman M, Ardman B Ž1995. Negative regulation of T-cell adhesion and activation by CD43. Nature 377:535]538 Davis SJ, van der Merwe PA Ž1996. Structure and ligand interactions of CD2: implications for T-cell function. Immunol Today 17:177]187 Shaw AS, Dustin ML Ž1997. Making the T cell receptor go the distance: a topological view of T cell activation. Immunity 6:361]369 Jones EY, Davis SJ, Williams AF, Harlos K, Stuart DA Ž1992. Crystal structure of a soluble form of the cell adhesion ˚ Nature 360:232]239 molecule CD2 at 2.8 A. Bodian DL et al Ž1994. Crystal structure of the extracellular ˚ region of the human cell adhesion molecule CD2 at 2.5 A resolution. Structure 2:755]766 Ikemizu S et al Ž1999. Crystal structure of the CD2-binding domain of CD58 Žlymphocyte function-associated antigen 3. at 1.8-A resolution. Proc Natl Acad Sci USA 96:4289]4294 Wang JH et al Ž1999. Structure of a heterophilic adhesion complex between the human CD2 and CD58 ŽLFA-3. counterreceptors. Cell 97:791]803 Jacobson K, Ishihara A, Inman R Ž1987. Lateral diffusion of proteins in membranes. Annu Rev Physiol 49:163]175 Liu SJ, Hahn WC, Bierer BE, Golan DE Ž1995. Intracellular mediators regulate CD2 lateral diffusion and cytoplasmic calcium ion mobilization upon CD2-mediated T cell activation. Biophys J 68:459]470 Saxton MJ, Jacobson K Ž1997. Single-particle tracking: applications to membrane dynamics. Annu Rev Biophys Biomol Struct 26:373]399

48. Caplan S, Zeliger S, Wang L, Baniyash M Ž1995. Cellsurface-expressed T-cell antigen]receptor zeta chain is associated with the cytoskeleton. Proc Natl Acad Sci USA 92:4768]4772 49. Rozdzial MM, Malissen B, Finkel TH Ž1995. Tyrosine-phosphorylated T cell receptor zeta chain associates with the actin cytoskeleton upon activation of mature T lymphocytes. Immunity 3:623]633 50. Jacobson K, Sheets ED, Simson R Ž1995. Revisiting the fluid mosaic model of membranes. Science 268:1441]1442 51. Kabouridis PS, Magee AI, Ley SC Ž1997. S-acylation of LCK protein tyrosine kinase is essential for its signalling function in T lymphocytes. EMBO J 16:4983]4998 52. Linsley PS et al Ž1996. Intracellular trafficking of CTLA-4 and focal localization towards sites of TCR engagement. Immunity 4:535]543 53. Griffiths GM Ž1995. The cell biology of CTL killing. Curr Opin Immunol 7:343]348 54. Kupfer A, Mosmann TR, Kupfer H Ž1991. Polarized expression of cytokines in cell conjugates of helper T cells and splenic B cells. Proc Natl Acad Sci USA 88:775]779 55. Kupfer H, Monks CR, Kupfer A Ž1994. Small splenic B cells that bind to antigen specific T helper ŽTh. cells and face the site of cytokine production in the Th cells selectively proliferate: immunofluorescence microscopic studies of Th-B antigen-presenting cell interactions. J Exp M ed 179:1507]1515 56. Bossi G, Griffiths GM Ž1999. Degranulation plays an essential part in regulating cell surface expression of Fas ligand in T cells and natural killer cells. Nat Med 5:90]96 57. Viola A, Schroeder S, Sakakibara Y, Lanzavecchia A Ž1999. T lymphocyte costimulation mediated by reorganization of membrane microdomains. Science 283:680]682 58. Bachmann MF et al Ž1999. Developmental regulation of Lck targeting to the CD8 coreceptor controls signaling in naive and memory T cells. J Exp Med 189:1521]1530 59. Brown DA, Rose JK Ž1992. Sorting of GPI-anchored proteins to glycolipid enriched membrane subdomains during transport to the apical cell surface. Cell 68:533]544 60. Stowers L, Yelon D, Berg LJ, Chant J Ž1995. Regulation of the polarization of T cells toward antigen-presenting cells by Ras-related GTPase CDC42. Proc Natl Acad Sci USA 92: 5027]5031 61. Lowin-Kropf B, Shapiro VS, Weiss A Ž1998. Cytoskeletal polarization of T cells is regulated by an immunoreceptor tyrosine-based activation motif-dependent mechanism. J Cell Biol 140:861]871 62. Lin CH, Thompson CA, Forscher P Ž1994. Cytoskeletal reorganization underlying growth cone motility. Curr Opin Neurobiol 4:640]647 63. Simons K, Ikonen E Ž1997. Functional rafts in cell membranes. Nature 387:569]572 64. Xavier R, Brennan T, Li Q, McCormack C, Seed B Ž1998. Membrane compartmentation is required for efficient T cell activation. Immunity 8:723]732 65. Montixi C et al Ž1998. Engagement of T cell receptor triggers its recruitment to low density detergent-insoluble membrane domains. EMBO J 17:5334]5348 66. Moran M, Miceli MC Ž1998. Engagement of GPI-linked CD48 contributes to TCR signals and cytoskeletal reorganization: a role for lipid rafts in T cell activation. Immunity 9:787]796 67. Rodgers W, Rose JK Ž1996. Exclusion of CD45 inhibits activity of p56lck associated with glycolipid-enriched membrane domains. J Cell Biol 135:1515]1523 68. Stulnig TM et al Ž1998. Polyunsaturated fatty acids inhibit T cell signal transduction by modification of detergent-insoluble membrane domains. J Cell Biol 143:637]644 69. Zhang W, Trible RP, Samelson LE Ž1998. LAT palmitoyla-

19

P. Anton van der Merwe et al

70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81.

82.

83. 84.

85. 86. 87.

88. 89. 90. 91.

92. 93.

tion: its essential role in membrane microdomain targeting and tyrosine phosphorylation during T cell activation. Immunity 9:239]246 Sedwick CE et al Ž1999. TCR, LFA-1, and CD28 play unique and complementary roles in signaling T cell cytoskeletal reorganization. J Immunol 162:1367]1375 Giancotti FG, Ruoslahti E Ž1999. Integrin signaling. Science 285:1028]1033 Reif K, Cantrell DA Ž1998. Networking Rho family GTPases in lymphocytes. Immunity 8:395]401 Fischer KD, Tedford K, Penninger JM Ž1998. Vav links antigen]receptor signaling to the actin cytoskeleton. Semin Immunol 10:317]327 Hall A Ž1998. Rho GTPases and the actin cytoskeleton. Science 279:509]514 Fischer KD et al Ž1998. Vav is a regulator of cytoskeletal reorganization mediated by the T-cell receptor. Curr Biol 8:554]562 Holsinger LJ et al Ž1998. Defects in actin-cap formation in Vav-deficient mice implicate an actin requirement for lymphocyte signal transduction. Curr Biol 8:563]572 Kong YY et al Ž1998. Vav regulates peptide-specific apoptosis in thymocytes. J Exp Med 188:2099]2111 Maekawa M et al Ž1999. Signaling from rho to the actin cytoskeleton through protein kinases ROCK and LIM-kinase. Science Snapper SB, Rosen FS Ž1999. The Wiskott-Aldrich Syndrome Protein ŽWASP.: roles in signaling and cytoskeletal organization. Annu Rev Immunol 17:905]929 Ley SC et al Ž1994. Tyrosine phosphorylation of alpha tubulin in human T lymphocytes. Eur J Immunol 24:99]106 Fernandez JA et al Ž1999. Phosphorylation- and activationindependent association of the tyrosine kinase Syk and the tyrosine kinase substrates Cbl and Vav with tubulin in B-cells. J Biol Chem 274:1401]1406 Burn P, Kupfer A, Singer SJ Ž1988. Dynamic membraneeytoskeletal interactions: specific association of integrin and talin arises in vivo after phorbol ester treatment of peripheral blood lymphocytes. Proc Natl Acad Sci USA 85:497]501 Offringa R, Bierer BE Ž1993. Association of CD2 with tubulin. Evidence for a role of the cytoskeleton in T cell activation. J Biol Chem 268:4979]4988 Lokeshwar VB, Bourguignon LY Ž1992. Tyrosine phosphatase activity of lymphoma CD45 ŽGP180. is regulated by a direct interaction with the cytoskeleton. J Biol Chem 267:21551]21557 Bretscher A Ž1999. Regulation of cortical structure by the ezrin]radixin]moesin protein family. Curr Opin Cell Biol 11:109]116 Shih NY et al Ž1999. Congenital nephrotic syndrome in mice lacking CD2AP. Science 286:312]314. Valitutti S, Dessing M, Aktories K, Gallati H, Lanzavecchia A Ž1995. Sustained signalling leading to T cell activation results from prolonged T cell receptor occupancy. Role of the T cell actin cytoskeleton. J Exp Med 181:577]584 Germain RN Ž1997. The importance of receptor clustering. Curr Biol 7:R640]R644 Davis MM et al Ž1998. Ligand recognition by alpha beta T cell receptors. Annu Rev Immunol 16:523]544 Janeway CA Ž1995. Ligands for the T-cell receptor: hard times for avidity models. Immunol Today 16:223]225 Ding YH, Baker BM, Garboczi DN, Biddison WE, Wiley DC Ž1999. Four A6-TCRrpeptiderHLA-A2 structures that generate very different T cell signals are nearly identical. Immunity 11:45]56 Lanzavecchia A, Lezzi G, Viola A Ž1999. From TCR engagement to T cell activation: a kinetic view of T cell behavior. Cell 96:1]4 Thomas ML Ž1999. The regulation of antigen]receptor sig-

94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104.

105.

106. 107. 108.

109.

110.

111. 112.

113.

114.

20

naling by protein tyrosine phosphatases: a hole in the story. Curr Opin Immunol 11:270]276 McKeithan TW Ž1995. Kinetic proofreading in T-cell receptor signal transduction. Proc Natl Acad Sci USA 92: 5042]5046 Rabinowitz JD, Beeson C, Lyons DS, Davis MM, McConnell HM Ž1996. Kinetic discrimination in T-cell activation. Proc Natl Acad Sci USA 93:1401]1405 Klemm JD, Schreiber SL, Crabtree GR Ž1998. Dimerization as a regulatory mechanism in signal transduction. Annu Rev Immunol 16:569]592 Boniface JJ et al Ž1998. Initiation of signal transduction through the T cell receptor requires the multivalent engagement of peptiderMHC ligands. Immunity 9:459]466 Viola A, Lanzavecchia A Ž1996. T cell activation determined by T cell receptor number and tunable thresholds. Science 273:104]106 Garcia KC, Teyton L, Wilson IA Ž1999. Structural basis of T cell recognition. Annu Rev Immunol 17:369]397 Delon J et al Ž1998. CD8 expression allows T cell signaling by monomeric peptide]MHC complexes. Immunity 9: 467]473 Locksley RM, Reiner SL, Hatam F, Littman DR, Killeen N Ž1993. Helper T cells without CD4: control of Leishmaniasis in CD4-deficient mice. Science 261:1448]1451 Schilham MW et al Ž1993. Alloreactive cytotoxic T cells can develop and function in mice lacking both CD4 and CD8. Eur J Immunol 23:1299]1304 Remy I, Wilson IA, Michnick SW Ž1999. Erythropoietin receptor activation by a ligand-induced conformation change. Science 283:990]993 O’Shea JJ, McVicar DW, Bailey TL, Burns C, Smyth MJ Ž1992. Activation of human peripheral blood T lymphocytes by pharmacological induction of protein-tyrosine phosphorylation. Proc Natl Acad Sci USA 89:10306]10310 Secrist JP, Burns LA, Karnitz L, Koretzky GA, Abraham RT Ž1993. Stimulatory effects of the protein tyrosine phosphatase inhibitor, pervanadate, on T-cell activation events. J Biol Chem 268:5886]5893 Mustelin T et al Ž1995. Regulation of the p70zap tyrosine protein kinase in T cells by the CD45 phosphotyrosine phosphatase. Eur J Immunol 25:942]946 Jin YJ, Friedman J, Burakoff SJ Ž1998. Regulation of tyrosine phosphorylation in isolated T cell membrane by inhibition of protein tyrosine phosphatases. J Immunol 161:1743]1750 Trowbridge IS, Thomas ML Ž1994. CD45}an emerging role as a protein-tyrosine-phosphatase required for lymphocyte-activation and development. Annu Rev Immunol 12: 85]116 Furukawa T, Itoh M, Krueger NX, Streuli M, Saito H Ž1994. Specific interaction of the CD45 protein-tyrosine phosphatase with tyrosine-phosphorylated CD3 zeta chain. Proc Natl Acad Sci USA 91:10928]10932 Leitenberg D, Boutin Y, Lu DD, Bottomly K Ž1999. Biochemical association of CD45 with the T cell receptor complex: regulation by CD45 isoform and during T cell activation. Immunity 10:701]711 Motoya S et al Ž1999. Interaction between CD45-AP and protein-tyrosine kinases involved in T cell receptor signaling. J Biol Chem 274:1407]1414 Veillette A, Soussou D, Latour S, Davidson D, Gervais FG Ž1999. Interactions of CD45-associated protein with the antigen receptor signaling machinery in T-lymphocytes. J Biol Chem 274:14392]14399 D’Oro U, Sakaguchi K, Appella E, Ashwell JD Ž1996. Mutational analysis of Lck in CD45-negative T cells: dominant role of tyrosine 394 phosphorylation in kinase activity. Mol Cell Biol 16:4996]5003 Tangye SG et al Ž1998. Negative regulation of human T cell

Redistribution of T cell surface molecules

115. 116. 117.

118.

119. 120. 121.

activation by the receptor-type protein tyrosine phosphatase CD148. J Immunol 161:3803]3807 Kersh GJ, Allen PM Ž1996. Essential flexibility in T-cell recognition of antigen. Nature 380:495]498 Weiss A, Littman DR Ž1994. Signal transduction by lymphocyte antigen receptors. Cell 76:263]274 Kersh GJ, Kersh EN, Fremont DH, Allen PM Ž1998. Highand low-potency ligands with similar affinities for the TCR: the importance of kinetics in TCR signaling. Immunity 9:817]826 Sloane-Lancaster J, Shaw AS, Rothbard JB, Allen PA Ž1994. Partial T cell signalling: altered phospho-z and lack of Zap70 recruitment in APL-induced T cell anergy. Cell 79:913]922 Madrenas J et al Ž1995. z phosphorylation without ZAP-70 activation induced by TCR antagonists or partial agonists. Science 267:515]518 Wallace VA et al Ž1997. Alterations in the level of CD45 surface expression affect the outcome of thymic selection. J Immunol 158:3205]3214 Gervais FG, Veillette A Ž1997. Reconstitution of interactions between protein-tyrosine phosphatase CD45 and tyrosineprotein kinase p56Žlck. in nonlymphoid cells. J Biol Chem 272:12754]12761

122. Thompson CB, Allison JP Ž1997. The emerging role of CTLA-4 as an immune attenuator. Immunity 7:445]450 123. Long EO Ž1999. Regulation of immune responses through inhibitory receptors. Annu Rev Immunol 17:875]904 124. Marengere ´ LEM et al Ž1996. Regulation of T cell receptor signalling by tyrosine phosphatase SYP association with CTLA-4. Science 272:1170]1173 125. Lee KM et al Ž1998. Molecular basis of T cell inactivation by CTLA-4. Science 282:2263]2266 126. Germain RN, Stefanova I Ž1999. The dynamics of T cell receptor signaling: complex orchestration and the key roles of tempo and cooperation. Annu Rev Immunol 17:467]522 127. Goldrath AW, Bevan MJ Ž1999. Low-affinity ligands for the TCR drive proliferation of mature CD8q T cells in lymphopenic hosts. Immunity 11:183]190 128. Ernst B, Lee D, Chang JM, Sprent J, Surh CD Ž1999. The peptide ligands mediating positive selection in the thymus control T cell survival and homeostatic proliferation in the periphery. Immunity 11:173]181 129. Crabtree GR Ž1989. Contingent genetic regulatory events in T lymphocyte activation. Science 243:355]361 130. Janeway CA Jr Ž1989. Approaching the asymptote? Evolution and revolution in immunology. Cold Spring Harb Symp Quant Biol 54:1]13

21