Microsoft Word - Paper 4 Deswal Author Copy.doc Nepal  Journal  of  Biotechnology.    Jan  2012,  Vol.  2,  No.  1:  62  –  71                                                                                                                    Biotechnology  Society  of  Nepal  (BSN),  All  rights  reserved   62 REVIEW ARTICLE Models  of  T  cell  antigen  receptor  activation:  the  puzzle  still  remained   Sumit Deswala,b aDepartment of Molecular Immunology, Max Planck Institute of Immunobiology and Epigenetics and Faculty of Biology III, University of Freiburg, Stübeweg 51, 79108 Freiburg, Germany bSpemann Graduate School of Biology and Medicine (SGBM), University of Freiburg, Germany Corresponding address: Max Planck Institute of Immunobiology and Epigenetics, Stübeweg 51, 79108 Freiburg, Germany. E-mail: deswal@immunbio.mpg.de, Tel.: +49-761-5108-314, FAX: +49-761-5108-423 Abstract T cell antigen receptor (TCR) is a protein-complex expressed on all T cells of the immune system and is responsible for the activation of T cells when infectious agent is presented by an antigen presenting cell (APC) in the form of peptides bound to the major histocompatibility complex (pMHC). Despite numerous studies it is not clear what biochemical changes upon binding of antigen ligand to the extracellular domains of TCR leads to activation of intracellular signaling (a process known as TCR triggering). This review summarizes possible biochemical mechanisms for TCR triggering and discusses their comparative limitations and advantages in explaining various experimental observations. Keywords: T cell antigen receptor, activation, model Introduction T cell antigen receptor (TCR or TCR-CD3) is a protein complex expressed exclusively in T cells and is composed of the variable TCR α and β chains and the constant CD3γ, CD3δ, CD3ε and CD3ζ chains (Figure 1). CD3γ and CD3δ chain are glycoproteins each of which form a heterodimer with non-glycosylated CD3ε chain. Along with the CD3ζ-ζ homodimer, these chains associate with the TCRαβ heterodimer to generate the full TCR- CD3 complex [1]. The TCR-CD3 is responsible for activation of a T cell upon antigen encounter which then help in the activation of B cells by releasing helper cytokines (helper T cells) or kill the target cell directly by the secretion of cytotoxic effector molecules such as granzymes, perforin and granulysin. Antigenic ligand binding to the TCR leads to the phosphorylation of immunoreceptor Nepal  Journal  of  Biotechnology.    Jan  2012,  Vol.  2,  No.  1:  62  –  71                                                                                                                    Biotechnology  Society  of  Nepal  (BSN),  All  rights  reserved   63 tyrosine-based activation motifs (ITAMs) of the CD3 complex by the Src family protein tyrosine kinase Lck, resulting in the membrane targeting and activation of another kinase, ZAP-70. Once activated, ZAP-70 phosphorylates several substrates, including the transmembrane adaptor protein, linker for activation of T cells (LAT). Phospho-LAT serves as an important point of divergence for signals initiated from the TCR by recruiting several effector molecules to the plasma membrane thus initiating multiple pathways essential for full T cell activation [2]. ! " # $% &'& # ++ + -- ITAM -- V C V C -- Fig.1: The TCR-CD3 complex. V and C represent variable and constant domains respectively for the TCR α and β. The hypervariable region that binds to the pMHC complex is shown in red color. Although the downstream signaling mechanisms of T cell activation have been explored to quite detailed extent, the basic mechanism of how ligand binding to extracellular domains of TCRαβ leads to downstream signaling remained unclear. Several models of TCR triggering have been proposed based on the experimental findings, but none of the models so far could explain every aspect of TCR triggering. Broadly, these models can be grouped in three major processes which involve aggregation, conformational change or segregation (Figure 2). This review summarizes the proposed models of TCR triggering and discusses their comparative abilities for explaining various experimental observations (Table 1). 1. Antigen induced clustering Several forms of clustering have been proposed as the triggering mechanism for the TCR. These include homodimer, heterodimer and the pseudodimer models as discussed below. Nepal  Journal  of  Biotechnology.    Jan  2012,  Vol.  2,  No.  1:  62  –  71                                                                                                                    Biotechnology  Society  of  Nepal  (BSN),  All  rights  reserved   64 b c d e f Lck TCR pMHC Lck LckLck binding TCR Lck CD45 pMHC Lck Lck Lck LckLck binding lipid raft Lck TCR pMHC Co-receptor Lck TCR pMHC Lck TCR pMHC LckLck CD45 TCR pMHC binding ! Lck Lck TCR pMHC Lck Pullbinding TCR pMHC binding a g h Clustering models Conformational change models Segregation models Fig.2: Various models of TCR triggering. a) Homodimer model, b) Heterodimer model, c) Pseudo-dimer model, d) Membrane binding model, e) Kinetic deformation model, f) Permissive geometry model, g) Kinetic segregation, h) Lipid raft mediated segregation model. Full description of each model is given in the main text. Nepal  Journal  of  Biotechnology.    Jan  2012,  Vol.  2,  No.  1:  62  –  71                                                                                                                    Biotechnology  Society  of  Nepal  (BSN),  All  rights  reserved   65 Table 1: Abilities of different models to explain various experimental observations. Observation     Triggering  model       Homodimer   Heterodimer   Pseudo- dimer   Membrane binding   Kinetic deformation   Permissive geometry   Kinetic segregation   Lipid raft   Pre-­‐clustered   TCR   No   Yes   Yes   Yes   Yes   Yes   Yes   Yes   Coreceptor   independent   triggering   Yes   No   No   Yes   Yes   Yes   Yes   Yes   Triggering  by   soluble   monomer  pMHC   No   Yes   No   Yes   No   No   No   Yes   Triggering  by   anti-­‐CD3   antibodies   Yes   No   No   Yes   No   Yes   No   No   Activation  by   pervanadate   No   No   No   No   No   No   Yes   Yes   Inhibition  by   truncated  CD45   No   No   No   No   No   No   Yes   No     Inhibition  by   elongated  pMHC   No   No   No   No   No   No   Yes   No     Triggering  in   absence  of  self   pMHC   Yes   Yes   No   Yes   Yes   Yes   Yes   Yes   a. Homodimer model Clustering of TCR–CD3 complexes following TCR engagement could lead to enhanced phosphorylation, for example by increasing the proximity of associated Lck molecules, resulting in the activation of the second receptor in the cluster by trans- autophosphorylation (or this enhanced proximity of kinases might lead to their trans- phosphorylation which enhance their activity to phosphorylate the receptor). This model is supported by the observations that artificial ligands such as pMHC tetramer and the anti- TCRαβ and anti-CD3 antibodies that crosslink the TCR lead to its activation whereas Fab fragment of these antibodies or the monomeric pMHC in solution fail to activate the TCR [3]. Also the observations that MHC class I [4, 5] and MHC class II [6] exist in pre-formed clusters, support the homodimer model. However, since several studies have now shown the existence of pre-clustered TCRs on the T cell surface [7, 8], it is difficult to imagine that clustering alone could trigger the Nepal  Journal  of  Biotechnology.    Jan  2012,  Vol.  2,  No.  1:  62  –  71                                                                                                                    Biotechnology  Society  of  Nepal  (BSN),  All  rights  reserved   66 TCR. In fact in a study by Schamel and colleagues [9], both clustering and conformational change were shown to be required for TCR triggering (permissive geometry model, as discussed later). b. Heterodimer model In the heterodimer model [10], coreceptor CD8 or CD4 binding to the same pMHC complex as the TCR brings the coreceptor- associated Lck kinase into proximity with TCR–CD3 ITAMs and their phosphorylation. Further it has been proposed that to fully activate the TCR, a pMHC ligand need to interact with a TCR–CD8 pair with a threshold time to induce stable zippering between the membrane-proximal domain of CD8 and the connecting peptide motif in the TCRα [11]. This allows stable association of Lck with the CD3 complex and results in complete phosphorylation of the CD3 ITAMs. A low-affinity pMHC ligand that interacts with a TCR–CD8 pair with less than threshold time induces incomplete zippering and therefore allows only transient Lck association and partial CD3 phosphorylation. However, the observations that T cells can develop and function normally in the mice lacking both CD4 and CD8 suggest that coreceptors are not absolutely required for TCR triggering and T cell activation [12, 13]. Also the fact that anti- CD3 antibodies and their F(ab’)2 fragments that do not engage the co-receptors can still activate the TCR [14, 15], indicates that co- receptors are not required for the TCR triggering itself, though they might be important for full activation of a T cell. c. Pseudo-dimer model Only a few MHCs carry high affinity agonist peptides for the TCRs, whereas a great majority of MHCs carry endogenous self- peptides. On the basis of the crystallographic studies which showed that the CD4 tail associating with Lck was far from their own TCR/CD3 complex, the pseudodimer model postulates that two TCRs are brought together by binding low-affinity self or high-affinity agonist pMHC ligands and that the CD4 associated with a TCR engaging the agonist pMHC complex can assists in phosphorylation of neighboring TCR (engaged to low affinity self peptide) by the associated kinase Lck [16]. Thus according to this model, a ‘pseudodimer’ consists of one foreign and one self-antigen engaged receptor linked via CD4 molecule is the primary unit of TCR signal initiation module. 2. Conformational change model Several studies have proposed conformational Nepal  Journal  of  Biotechnology.    Jan  2012,  Vol.  2,  No.  1:  62  –  71                                                                                                                    Biotechnology  Society  of  Nepal  (BSN),  All  rights  reserved   67 change as a requirement for TCR triggering [17, 18]. However, how the ligand binding to TCRαβ ectodomains leads to a conformational change in the cytoplasmic tails of CD3 subunits is not clear. The situation is complicated by the fact that the crystal structure of the fully assembled TCR-CD3 complex is not known, although individual subunits have been crystallized partially [19- 22]. Some of the important models for the mechanism of conformational change in TCR- CD3 are discussed below. a. Membrane binding model According to this model, cytoplasmic tails of CD3ε [23, 24] and CD3ζ [25] associate with the lipids present in the plasma membrane of a T cell. Thereby, the ITAMs are buried in the plasma membrane, making them inaccessible for phosphorylation by the kinases. Ligand binding leads to release of these CD3 chains from the plasma membrane, making them accessible to kinase and hence the phosphorylation. b. Kinetic deformation model Several line of evidence have proposed that the mechanical effects (such as pulling or shearing) of pMHC binding to the TCR leads to a piston-like displacement of the TCR–CD3 complex [26, 27]. This induces a change in the conformation of the CD3 cytoplasmic domains, allowing ITAM phosphorylation. c. Permissive geometry model Based on the experimental evidence for the existence of TCR-CD3s in a pre-clustered form [7] and the fact that pMHC monomers in solution fail to trigger TCR[28, 29] while pMHC dimer or higher oligomer can induce TCR triggering, the permissive geometry model was proposed[30]. According to this model, before ligand binding, the mono and multivalent TCR-CD3s exist in an autoinhibited state where ITAMs are inaccessible for phosphorylation by the kinase. Simultaneous dimeric (or higher order) ligand binding to the two TCR-CD3 complexes leads to a scissor-like movement in the two TCR- CD3 complexes leading to exposure of their cytoplasmic tails, thus making the ITAMs accessible for phosphorylation. 3. Segregation Model In a resting T cell, phosphorylation of the TCR-CD3 is kept in check by the active phosphotases. This is supported by the fact that pervanadate, a phosphatase inhibitor, Nepal  Journal  of  Biotechnology.    Jan  2012,  Vol.  2,  No.  1:  62  –  71                                                                                                                    Biotechnology  Society  of  Nepal  (BSN),  All  rights  reserved   68 treatment leads to strong phosphorylation of the TCR-CD3 ITAMs and also initiates the downstream signaling [31]. Thus any process that favors the kinase-phosphtase balance towards kinase, can lead to TCR-CD3 phosphorylation and the signaling. Redistribution of the TCR-CD3, kinases and phosphatases in which phosphatase is segregated from the TCR can interfere with the dephosphorylation. There are two mechanisms proposed for such ligand dependent re-distribution, which are discussed below. a. Kinetic segregation Kinetic segregation model was proposed based on the topological view of the cell surface molecules at the T cell–APC interface [32, 33]. The tight intercellular contact causes the segregation of the molecules by sizes of their ectodomain resulting in physical separation of TCR from the large inhibitory tyrosine phosphatase CD45, leading to stable phosphorylation of TCR–CD3 ITAMs by Lck. Though sounding good in the context of a T- cell – APC contact, the model fails to explain the activation of T cells by soluble anti- TCRαβ and anti-CD3 antibodies and the pMHC tetramer [14, 34]. b. Lipid raft mediated segregation model Lipid rafts are the detergent (such as Triton X- 100, Brij-series, NP-40 or CHAPS)-resistant membrane microdomains which are enriched in glycosphingolipids, cholesterol and lipid- modified proteins such as the GPI-anchored proteins. Also these microdomains are enriched in double-acylated (myristoylated, palmitoylated) Src-family kinases and the important signaling molecules such as the coreceptors CD4 and CD8 and the adaptor protein LAT. Lipid raft mediated segregation model postulates that pMHC engagement results in partitioning of the TCR–CD3 complex into lipid rafts enriched in Lck and deficient in CD45 [35, 36]. Partial support for this model comes from the observations that palmitoylation-deficient (and raft excluded) mutants of Lck and LAT are functionally defective. In addition, artificial targeting of other cytoplasmic molecules, such as SHP-1 [37], CD45 [38] or PLCγ [39], to membrane rafts has marked functional effects on TCR- induced signaling. A connecting peptide in the TCRα chain and the CD3δ chain were identified as the critical sites essential for effective raft association. Mutations in these components interfere with TCR signaling [40]. The mechanism of directing association of the engaged TCR complex with lipid rafts is not clear but might be based on co-engagement of the raft resident CD4/CD8 coreceptors. Nepal  Journal  of  Biotechnology.    Jan  2012,  Vol.  2,  No.  1:  62  –  71                                                                                                                    Biotechnology  Society  of  Nepal  (BSN),  All  rights  reserved   69 Conclusion Analysis of various proposed mechanisms reveals that probably a combination of these mechanisms is responsible for TCR triggering. Existence of pre-clustered TCR molecules rules out the homodimerization as a requirement for TCR triggering. It is possible that CD3 ITAMs are inaccessible for phosphorylation by the kinase in the resting state and some sort of conformational change is necessary for exposing the ITAMs, but additional studies are required to find out what mechanism is responsible for such conformational change. Segregation models argue the balance of kinase mediated phosphorylation and phosphatase mediated dephosphorylation shifts in favour of kinase upon ligand binding. Though such segregation might contribute for TCR triggering in the context of a T cell-antigen presenting cell interaction, it seems not the sole mechanism of TCR triggering as the soluble anti-TCR or anti-CD3 antibodies which do not induce segregation can still cause TCR triggering. Further studies on the structure of fully assembled TCR-CD3 complex will be required for a better understanding of the TCR triggering. References 1. Call  ME,  Pyrdol  J,  Wiedmann  M,  Wucherpfennig   KW:  The  organizing  principle   in  the  formation   of  the  T  cell  receptor-­‐CD3  complex.  Cell  2002,   111:967-­‐979.     2. Smith-­‐Garvin  JE,  Koretzky  GA,  Jordan  MS:  T  cell   activation.   Annu   Rev   Immunol   2009,   27:591-­‐ 619.     3. Cochran   JR,   Aivazian   D,   Cameron   TO,   Stern   LJ:   Receptor   clustering   and   transmembrane   signaling   in   T   cells.   Trends   Biochem   Sci   2001,   26(5):304-­‐310.     4. Matko  J,  Bushkin  Y,  Wei  T,  Edidin  M:  Clustering   of   class   I   HLA   molecules   on   the   surfaces   of   activated   and   transformed   human   cells.   J   Immunol  1994,  152(7):3353-­‐3360.     5. Fooksman   DR,   Gronvall   GK,   Tang   Q,   Edidin   M:   Clustering  class  I  MHC  modulates  sensitivity  of   T  cell  recognition.  Journal  of  immunology  2006,   176(11):6673-­‐6680.     6. Unternaehrer   JJ,   Chow   A,   Pypaert   M,   Inaba   K,   Mellman   I:   The   tetraspanin   CD9   mediates   lateral  association  of  MHC  class  II  molecules  on   the   dendritic   cell   surface.   Proceedings   of   the   National   Academy   of   Sciences   of   the   United   States  of  America  2007,  104(1):234-­‐239.     7. Schamel   WW,   Arechaga   I,   Risueno   RM,   van   Santen   HM,   Cabezas   P,   Risco   C,   Valpuesta   JM,   Alarcon   B:   Coexistence   of   multivalent   and   monovalent  TCRs  explains  high  sensitivity  and   wide   range   of   response.   J   Exp   Med   2005,   202:493-­‐503.     8. Lillemeier   BF,   Mortelmaier   MA,   Forstner   MB,   Huppa  JB,  Groves  JT,  Davis  MM:  TCR  and  Lat  are   expressed  on  separate  protein  islands  on  T  cell   membranes  and  concatenate  during  activation.   Nat  Immunol  2010,  11(1):90-­‐96.     9. Minguet   S,   Swamy   M,   Alarcon   B,   Luescher   IF,   Schamel   WW:   Full   activation   of   the   T   cell   receptor   requires   both   clustering   and   Nepal  Journal  of  Biotechnology.    Jan  2012,  Vol.  2,  No.  1:  62  –  71                                                                                                                    Biotechnology  Society  of  Nepal  (BSN),  All  rights  reserved   70 conformational   changes   at   CD3.   Immunity   2007,  26(1):43-­‐54.     10. Trautmann   A,   Randriamampita   C:   Initiation   of   TCR  signalling  revisited.  Trends  Immunol  2003,   24(8):425-­‐428.     11. Palmer   E,   Naeher   D:   Affinity   threshold   for   thymic   selection   through   a   T-­‐cell   receptor-­‐co-­‐ receptor   zipper.   Nat   Rev   Immunol   2009,   9(3):207-­‐213.     12. Locksley   RM,   Reiner   SL,   Hatam   F,   Littman   DR,   Killeen  N:  Helper  T  cells  without  CD4:  control  of   leishmaniasis   in   CD4-­‐deficient   mice.   Science   1993,  261(5127):1448-­‐1451.     13. Schilham   MW,   Fung-­‐Leung   WP,   Rahemtulla   A,   Kuendig   T,   Zhang   L,   Potter   J,   Miller   RG,   Hengartner  H,  Mak  TW:  Alloreactive  cytotoxic  T   cells  can  develop  and  function  in  mice  lacking   both   CD4   and   CD8.   European   journal   of   immunology  1993,  23(6):1299-­‐1304.     14. Chang   TW,   Kung   PC,   Gingras   SP,   Goldstein   G:   Does  OKT3  monoclonal  antibody  react  with  an   antigen-­‐recognition   structure   on   human   T   cells?  Proc  Natl  Acad  Sci  U  S  A  1981,  78(3):1805-­‐ 1808.     15. Kaye  J,  Janeway  CA,  Jr.:  The  Fab  fragment  of  a   directly   activating   monoclonal   antibody   that   precipitates   a   disulfide-­‐linked   heterodimer   from  a  helper  T  cell  clone  blocks  activation  by   either  allogeneic  Ia  or  antigen  and  self-­‐Ia.  J  Exp   Med  1984,  159(5):1397-­‐1412.     16. Krogsgaard  M,  Li  QJ,  Sumen  C,  Huppa  JB,  Huse   M,   Davis   MM:   Agonist/endogenous   peptide-­‐ MHC   heterodimers   drive   T   cell   activation   and   sensitivity.  Nature  2005,  434(7030):238-­‐243.   17. Levin   SE,   Weiss   A:   Twisting   tails   exposed:   the   evidence  for  TCR  conformational  change.  J  Exp   Med  2005,  201(4):489-­‐492.     18. Gil   D,   Schamel   WW,   Montoya   M,   Sanchez-­‐ Madrid  F,  Alarcon  B:  Recruitment  of  Nck  by  CD3   epsilon   reveals   a   ligand-­‐induced   conformational   change   essential   for   T   cell   receptor  signaling  and  synapse  formation.  Cell   2002,  109(7):901-­‐912.     19. Sun   ZY,   Kim   ST,   Kim   IC,   Fahmy   A,   Reinherz   EL,   Wagner   G:   Solution   structure   of   the   CD3epsilondelta   ectodomain   and   comparison   with  CD3epsilongamma  as  a  basis  for  modeling   T   cell   receptor   topology   and   signaling.   Proc   Natl  Acad  Sci  U  S  A  2004,  101(48):16867-­‐16872.     20. Arnett   KL,   Harrison   SC,   Wiley   DC:   Crystal   structure  of  a  human  CD3-­‐epsilon/delta  dimer   in  complex  with  a  UCHT1  single-­‐chain  antibody   fragment.   Proc   Natl   Acad   Sci   U   S   A   2004,   101(46):16268-­‐16273.     21. Sun   ZYS,   Seok   Kim   K,   Wagner   G,   Reinherz   EL:   Mechanisms   contributing   to   T   cell   receptor   signaling   and   assembly   revealed   by   the   solution  structure  of  an  ectodomain  fragment   of  the  CD3eg  heterodimer.  Cell  2001,  105:913-­‐ 923.     22. Kjer-­‐Nielsen  L,  Dunstone  MA,  Kostenko  L,  Ely  LK,   Beddoe   T,   Mifsud   NA,   Purcell   AW,   Brooks   AG,   McCluskey  J,  Rossjohn  J:  Crystal  structure  of  the   human   T   cell   receptor   CD3   epsilon   gamma   heterodimer   complexed   to   the   therapeutic   mAb   OKT3.   Proc   Natl   Acad   Sci   U   S   A   2004,   101(20):7675-­‐7680.     23. Xu  C,  Gagnon  E,  Call  ME,  Schnell  JR,  Schwieters   CD,   Carman   CV,   Chou   JJ,   Wucherpfennig   KW:   Regulation   of   T   cell   receptor   activation   by   dynamic  membrane  binding  of  the  CD3epsilon   cytoplasmic   tyrosine-­‐based   motif.   Cell   2008,   135(4):702-­‐713.     24. Deford-­‐Watts   LM,   Tassin   TC,   Becker   AM,   Medeiros   JJ,   Albanesi   JP,   Love   PE,   Wulfing   C,   van  Oers  NS:  The  cytoplasmic  tail  of  the  T  cell   receptor   CD3   epsilon   subunit   contains   a   phospholipid-­‐binding   motif   that   regulates   T   cell   functions.   J   Immunol   2009,   183(2):1055-­‐ 1064.     Nepal  Journal  of  Biotechnology.    Jan  2012,  Vol.  2,  No.  1:  62  –  71                                                                                                                    Biotechnology  Society  of  Nepal  (BSN),  All  rights  reserved   71 25. Aivazian   D,   Stern   LJ:   Phosphorylation   of   T   cell   receptor  zeta  is  regulated  by  a  lipid  dependent   folding   transition.   Nat   Struct   Biol   2000,   7(11):1023-­‐1026.     26. Ma  Z,  Sharp  KA,  Janmey  PA,  Finkel  TH:  Surface-­‐ anchored   monomeric   agonist   pMHCs   alone   trigger   TCR   with   high   sensitivity.   PLoS   Biol   2008,  6(2):e43.   27. Ma   Z,   Janmey   PA,   Finkel   TH:   The   receptor   deformation   model   of   TCR   triggering.   Faseb   J   2008,  22(4):1002-­‐1008.     28. Stone  JD,  Stern  LJ:  CD8  T  Cells,  Like  CD4  T  Cells,   Are   Triggered   by   Multivalent   Engagement   of   TCRs   by   MHC-­‐Peptide   Ligands   but   Not   by   Monovalent   Engagement.   J   Immunol   2006,   176(3):1498-­‐1505.     29. Cochran   JR,   Cameron   TO,   Stern   LJ:   The   relationship  of  MHC-­‐peptide  binding  and  T  cell   activation   probed   using   chemically   defined   MHC   class   II   oligomers.   Immunity   2000,   12(3):241-­‐250.     30. Minguet   S,   Schamel   WWA:   A   permissive   geometry  model  for  TCR-­‐CD3  activation.  Trends   Biochem  Sci  2008,  33(2):51-­‐57.     31. Secrist   JP,   Burns   LA,   Karnitz   L,   Koretzky   GA,   Abraham  RT:  Stimulatory  effects  of  the  protein   tyrosine   phosphatase   inhibitor,   pervanadate,   on   T-­‐cell   activation   events.   J   Biol   Chem   1993,   268(8):5886-­‐5893.     32. Choudhuri  K,  Wiseman  D,  Brown  MH,  Gould  K,   van  der  Merwe  PA:  T-­‐cell  receptor  triggering  is   critically   dependent   on   the   dimensions   of   its   peptide-­‐MHC   ligand.   Nature   2005,   436(7050):578-­‐582.     33. Davis   SJ,   van   der   Merwe   PA:   The   kinetic-­‐ segregation  model:  TCR  triggering  and  beyond.   Nat  Immunol  2006,  7(8):803-­‐809.     34. Boniface  JJ,  Rabinowitz  JD,  Wulfing  C,  Hampl  J,   Reich   Z,   Altman   JD,   Kantor   RM,   Beeson   C,   McConnell   HM,   Davis   MM:   Initiation   of   signal   transduction   through   the   T   cell   receptor   requires   the   multivalent   engagement   of   peptide/MHC   ligands   [corrected].   Immunity   1998,  9(4):459-­‐466.     35. Horejsi   V:   Lipid   rafts   and   their   roles   in   T-­‐cell   activation.  Microbes  Infect  2005,  7(2):310-­‐316.     36. Horejsi   V:   The   roles   of   membrane   microdomains   (rafts)   in   T   cell   activation.   Immunol  Rev  2003,  191:148-­‐164.     37. Su  MW,  Yu  CL,  Burakoff  SJ,  Jin  YJ:  Targeting  Src   homology   2   domain-­‐containing   tyrosine   phosphatase   (SHP-­‐1)   into   lipid   rafts   inhibits   CD3-­‐induced   T   cell   activation.   Journal   of   immunology  2001,  166(6):3975-­‐3982.     38. He  X,  Woodford-­‐Thomas  TA,  Johnson  KG,  Shah   DD,   Thomas   ML:   Targeting   of   CD45   protein   tyrosine   phosphatase   activity   to   lipid   microdomains  on  the  T  cell  surface  inhibits  TCR   signaling.   European   journal   of   immunology   2002,  32(9):2578-­‐2587.     39. Veri   MC,   DeBell   KE,   Seminario   MC,   DiBaldassarre   A,   Reischl   I,   Rawat   R,   Graham   L,   Noviello   C,   Rellahan   BL,   Miscia   S   et   al:   Membrane   raft-­‐dependent   regulation   of   phospholipase   Cgamma-­‐1   activation   in   T   lymphocytes.   Molecular   and   cellular   biology   2001,  21(20):6939-­‐6950.     40. Werlen   G,   Hausmann   B,   Palmer   E:   A   motif   in   the  alphabeta  T-­‐cell  receptor  controls  positive   selection   by   modulating   ERK   activity.   Nature   2000,  406:422-­‐426.