MINIREVIEW ERK and cell death: ERK location and T cell selection Emma Teixeiro and Mark A. Daniels Department of Molecular Microbiology and Immunology, Center for Cellular and Molecular Immunology, School of Medicine, University of Missouri, Columbia, MO, USA Introduction The development of a healthy immune system depends on the generation of a diverse pool of T cells capable of providing protection against a broad range of pathogens, while avoiding an autoimmune attack on healthy tissue. T cells recognize specific peptide anti- gens presented in the context of self-major histocom- patibility complex (MHC) molecules through clonally distributed T cell antigen receptors (TCRs) expressed early during T cell ontogeny. The TCR is formed by the random association of variable and constant genetic elements. Although this process leads to the production of a diverse pool of pathogen-specific T cells, it also leads to the generation of T cells that are either useless, due to an inability to recognize MHC, or extremely dangerous, due to the potential for an overt reaction against self. Therefore, it is essential that T cell development includes a selection process by which only the useful cells are instructed to mature and the nonfunctional and potentially harmful T cells are eliminated before they can fully develop. The shaping of the T cell repertoire begins when immature T cell precursors, called thymocytes, are selected by the ability of their TCR to recognize self- peptides presented by MHC (pMHC) on the various populations of antigen-presenting cells present in the thymus [1]. A ‘Goldilocks’ affinity model of selection has been proposed to describe this process. T cells that are unable to recognize self-peptide MHC undergo Keywords ERK; MAPK; T cell selection; TCR signaling; thymocyte Correspondence M. A. Daniels, Department of Molecular Microbiology and Immunology, Center for Cellular and Molecular Immunology, M616 Medical Sciences Bldg, One Hospital Drive, Columbia, MO 65212, USA Fax: +573 882 4287 Tel: +573 884 1659 E-mail: danielsma@missouri.edu (Received 19 June 2009, revised 14 August 2009, accepted 26 August 2009) doi:10.1111/j.1742-4658.2009.07368.x The selection of functional T cells is mediated by interactions between the T cell antigen receptor and self-peptide major histocompatibility complex expressed on thymic epithelium. These interactions either lead to survival and development or death. The T cell antigen receptor is an unusual recep- tor able to signal multiple cell fates. The precise mechanism by which this is achieved has been an area of intense research effort over the years. One model proposes that the differential activation of mitogen-activated protein kinase pathways contributes to this decision. Here, the role of extracellular signal-regulated kinase in promoting or preventing apoptosis during thymic selection is discussed. Abbreviations BIM, Bcl2 like 11; DAG, diacylglycerol; Egr, early growth response protein; ERK, extracellular signal-regulated kinase; ITAM, immunoreceptor tyrosine-based activation motif; JNK, c-JunNH 2 -terminal kinase; LAT, linker for activation of T cells; MAPK, mitogen-activated protein kinase; MHC, major histocompatibility complex; pERK, phospho-extracellular signal-regulated kinase; pJNK, phospho-c-JunNH 2 -terminal kinase; PLC, phospholipase C; SLP-76, SH2 domain containing leukocyte protein of 76 kDa; SAP-1, SRF accessory protein 1 (ELK4); SOS, son of sevenless; TCR, T cell antigen receptor. 30 FEBS Journal 277 (2010) 30–38 ª 2009 The Authors Journal compilation ª 2009 FEBS apoptosis (death by neglect). Weak or intermediate TCR ⁄ pMHC interactions induce positive selection and lead to the survival, development and maturation of a self-restricted, yet self-tolerant, T cell repertoire (reviewed in [2]). TCRs that bind self-peptide MHC with high affinity induce either anergy [3], receptor editing [4], deviation into regulatory T cell lineage [5] or clonal deletion (apoptosis), which collectively are considered to be negative selection [6]. Clonal deletion is thought to be the dominant form of negative selec- tion. Although this model of selection is generally accepted, the question remains as to how the TCR can translate subtle changes in ligand binding parameters to signal such distinct cell fates as survival ⁄ differentia- tion and death. The goal then is to establish the point where the TCR signals diverge, leading to the ultimate fate of either life or death for the developing thymocyte. A differential signaling model, where different mitogen-activated protein kinase (MAPK) signals lead to either positive or negative selection, has begun to emerge (reviewed in [2,7,8]). MAPK signaling is important for determining cell fate decisions in a diverse number of organisms and cell types (reviewed in [7,8]). In thymocytes, c-JunNH 2 -terminal kinase (JNK) [9], p38 [10] and extracellular signal-regulated kinase 5 (ERK5) [11] are MAPKs essential for negative selection, but do not influence positive selec- tion. The small GTPase Ras initiates the MAPK cas- cade that leads to Raf1–MEK1 ⁄ 2–ERK1 ⁄ 2 activation. The phosphorylation of ERK1 ⁄ 2 is important for positive selection and dispensable for negative selection [12–15]. Interestingly, positive and negative selecting ligands activate all four of these pathways. The conun- drum is how does a T cell integrate these signals to distinguish positive from negative selection? One possi- ble explanation may be the location of the active form of these signaling molecules within a cell. It was recently shown, that in thymocytes, positive and nega- tive selecting ligands induce the localization of the components of the Ras ⁄ ERK cascade and active ERK into distinct subcellular compartments (Figs 1 and 2) [16]. Several groups have demonstrated that the biolog- ical outcome of Ras ⁄ MAPK activation is determined by its subcellular localization [8,17]. The role of ERK in promoting or preventing clonal deletion (apoptosis) during the thymic selection decision-making process is the subject of this review. Is LAT the fork in the TCR signaling road? A long-time goal of immunologists is to establish where signals emanating from the TCR diverge and lead to such distinct cell fates as survival and death. Much is known about the events that occur immedi- ately upon TCR engagement of pMHC. Lck is acti- vated by CD45 and recruited to the TCR ⁄ CD3 complex by the coreceptors CD4 or CD8. Lck then Fig. 2. Localization of pattern Ras ⁄ MAPK signaling intermediates during negative selection. The figure depicts the location of mole- cules described in the text in the case of TCR engagement by a negative selecting ligand. Note the separate location of pJNK and pERK, and Ras-GRP1 ⁄ Grb2 ⁄ SOS ⁄ Ras ⁄ Raf at the plasma membrane. Fig. 1. Localization of pattern Ras ⁄ MAPK signaling intermediates during positive selection. The figure depicts the location of mole- cules described in the text in the case of TCR engagement by a positive selecting ligand. Note the similar location of pJNK and pERK, and Ras-GRP1 ⁄ Ras ⁄ Raf at the Golgi. E. Teixeiro and M. A. Daniels ERK location and T cell selection FEBS Journal 277 (2010) 30–38 ª 2009 The Authors Journal compilation ª 2009 FEBS 31 phosphorylates the immunoreceptor tyrosine-based activation motifs (ITAMs) of the CD3 subunits and TCRf. This allows for the recruitment of the kinase Zap-70 to the TCR and induces its activation (reviewed in [18]). The TCRf chain contains six ITAMs. Initially, studies comparing variants of agonist peptides suggested that qualitative differences in TCRf phosphorylation of individual ITAMs contributed to the selection decision. In these studies, weak ligands, capable of inducing positive selection, generated the p21 form of phopho-f. High-affinity ligands, capable of inducing negative selection, generated p23-f [19,20]. More recently, the mutation of selected ITAMs and studies on f-chain phosphorylation have provided more support for a quantitative than a qualitative TCRf phosphorylation model [21–23]. One study has suggested that phosphorylation of a defined number of CD3 ITAMs is required for each developmental step in selection [23]. Another study showed that a gradual decrease in TCR affinity correlates with a gradual decrease in total TCRf phosphorylation. Interestingly, in spite of the small change in total TCRf phosphory- lation between ligands that lie on either side of the boundary of selection, the recruitment of active Zap- 70 to the membrane is markedly enhanced for negative selecting ligands [16]. The TCR f chain does not have the capacity to recruit a wide variety of signaling molecules [24]. How- ever, one of the downstream targets of active Zap-70, the linker for activation of T cells (LAT), contains nine phosphorylatable tyrosines that are able to specifi- cally recruit several signaling molecules essential for thymocyte selection (reviewed in [25]). This suggests that LAT could be a more suitable candidate as a branch point in TCR signal transduction. Mutational analyses specifically determined that the four distal LAT tyrosines are critically important for T cell devel- opment [26]. Phosphorylation of these tyrosines facili- tates the recruitment of phospholipase C (PLC)c1, Gads, SLP-76 and Grb-2. PLCc1 is essential for the mobilization of calcium. The role of calcium, upstream of calcineurin, has long been appreciated as being important for thymocyte selection [27]. Activation of PLCc1 also induces the generation of diacylglycerol (DAG), which is essential for the activation of protein kinase C-h and the guanine exchange factor Ras- GRP1. Protein kinase C-h and Ras-GRP1 are impor- tant for the activation of Ras ⁄ Raf1 ⁄ ERK and have also been linked to positive selection in thymocytes [28,29]. Ras-GRP1, on the other hand, appears to be much less important for negative selection [30]. Gads mediates the association of SLP-76, another Zap-70 substrate, to LAT. Mice deficient in Gads or SLP-76 are largely defective for positive and negative selection [31,32]. Members of the Tec family of kinases associate with LAT ⁄ Gads ⁄ SLP-76 and contribute to the stability of the complex and enhance the activation of PLCc1. Tec kinase deficiency alters both positive and negative selection [33,34]. Grb2 is the only adaptor molecule recruited to LAT that is uniquely required for negative selection. Its ability to associate with the guanine exchange factor son of sevenless (SOS) in T cells makes it an important activator of the Ras ⁄ Raf ⁄ ERK pathway. Grb2 haploid insufficient mice demonstrate decreased induction of active JNK and p38 with a concomitant reduction in negative selection. Interestingly, these mice do not have a defect in the generation of phospho-ERK (pERK) [35]. Together with the role of Ras-GRP1 in ERK activation and positive selection, these data sug- gest that negative selecting ligands would exclusively activate Grb2 ⁄ SOS and positive selectors would acti- vate Ras-GRP1. In support of this, positive selectors do not recruit Grb2 ⁄ SOS to LAT. They only activate Ras-GRP1 and induce its recruitment to the Golgi [16]. However, the fact that negative selecting ligands activate and recruit both Ras-GRP1 and Grb2 ⁄ SOS to the plasma membrane argues against this. Mathemati- cal modeling of LAT phosphorylation [36] and studies on the phosphorylation kinetics of individual LAT ty- rosines [37], indicate that the lack of Grb2–LAT inter- action during positive selection may be due to partial phosphorylation of LAT. In line with this, positive and negative selecting ligands show quantitative differ- ences in total phosphorylation of LAT [16], although the phosphorylation state of individual tyrosines in thymocytes has not been assessed. Taken together these data favor a model where the signal emanating from the TCR begins to diverge at LAT. The location of ERK1 ⁄ 2 and selection outcome The differential recruitment of signaling intermediates to LAT appears to have a direct consequence on the regulation of downstream signaling pathways that are important for determining the selection decision. One of these is the Ras ⁄ ERK pathway. The role of the Ras ⁄ ERK cascade in thymic selection has been well studied (reviewed in [7]). Phosphorylation of ERK1 ⁄ 2 is essential for positive selection, but their role in nega- tive selection is dispensable [12–15]. The activation of ERK is linked to LAT through PLCc1 ⁄ DAG ⁄ Ras- GRP1 ⁄ Ras and Grb2⁄ SOS ⁄ Ras (reviewed in [8]). When thymocytes are stimulated by negative selecting ligands, they induce a rapid, robust, yet transient ERK location and T cell selection E. Teixeiro and M. A. Daniels 32 FEBS Journal 277 (2010) 30–38 ª 2009 The Authors Journal compilation ª 2009 FEBS induction of pERK that is localized to the plasma membrane (Fig. 2). On the other hand, positive select- ing ligands induce a slow and sustained activation of ERK originating from the Golgi and leading to pERK being distributed throughout the cell (Fig. 1) [16,38,39]. The differences in the kinetics of pERK induction may be explained by both the location and the identity of the upstream activators of Ras. Ras activation by only Ras-GRP1 follows a graded response that correlates with the stimulus, whereas SOS, on the other hand, contains a positive feedback loop that dramatically increases the rate of Ras activa- tion [40]. Furthermore, there is a marked increase in negative regulation of Ras at the plasma membrane versus the Golgi [8]. Therefore, the differential recruit- ment of Grb2 ⁄ SOS and Ras compartmentalization describe a potential mechanism for the differences in pERK kinetics observed between positive and negative selecting ligands [16,38,39]. The method that thymocytes utilize for the compart- mentalization of the components of the Ras ⁄ ERK pathway is not completely understood. The regulation of activation and membrane recruitment of Ras-GRP1 is dependent on calcium and DAG. One could imagine a scenario where the slow generation of calcium (and DAG) induced by positive selectors activates Ras- GRP1, which then binds to the DAG-rich Golgi [8,16]. Conversely, the robust calcium flux induced by nega- tive selectors could correlate with the generation of large quantities of DAG at the plasma membrane and lead to Ras-GRP1 recruitment to that location. Recent work has demonstrated that in T cells, PLCc1 activa- tion leads to activation of Ras-GRP1 and its recruit- ment to the Golgi and lymphocyte function-associated antigen (LFA-1)-mediated activation of phopholipase D2 leads to activation of Ras-GRP1 on both the Golgi and the plasma membrane [41]. In addition, DAG kin- ases have also been shown to play an important role in the regulation of Ras-GRP1 localization [42,43]. The location of Ras-GRP1 and SOS in turn lead to the recruitment, activation and compartmentalization of Ras to the different membranes within a cell. Downstream of Ras, various MAPK scaffolds that are restricted to different membrane compartments are probably responsible for the localization of pERK [8]. One such scaffold, kinase suppressor of Ras (KSR)1, has been shown to be important for the membrane localization of ERK and to somehow play a role in T cell development [44,45]. How these processes combine to determine the localization pattern of Ras ⁄ ERK in thymocytes remains to be seen. The classic paradigm of ERK activation would predict that once phosphory- lated, ERK dissociates from the scaffold where it can either act on cytosolic targets or move to the nucleus to activate transcription factors (reviewed in [7,8]). In light of this, the localization of pERK at the plasma membrane by negative selecting ligands is most curious (Fig. 2). Whether it is somehow sequestered at the plasma membrane or subject to rapid dephosphoryla- tion is a question that remains to be answered. When all of this is considered together, differences in the location of pERK may provide the developing thymo- cyte with the ability to distinguish positive and nega- tive selecting ligands. Although the precise mechanism of how differential MAPK compartmentalization con- tributes to this decision is open to debate, possible implications of pERK localization are discussed below. Several transcription factors that are downstream targets of ERK1 ⁄ 2 play an important role in mediating positive selection. SRF accessory protein 1 (ELK4) (SAP-1) is a ternary complex factor subfamily member of Ets transcription factors that is activated by pERK1 ⁄ 2. Deficiency of SAP-1 leads to a block in positive selection [46]. Activation of SAP-1 leads to the expression of early growth response protein (Egr)1. Overexpression of Egr1 results in positive selection on a nonselecting background [47], and Egr1-deficient mice are impaired for positive selection [48]. ERK1 ⁄ 2 activation also leads to a reduction in DNA binding by the basic helix–loop–helix protein E2A through the increased expression of the inhibitor of basic helix– loop–helix protein Id3 [49]. E2A-deficient mice have enhanced positive selection [50]; Id3-deficient mice demonstrate a profound block in positive selection [51]. Therefore, the sequestration of active ERK at the plasma membrane by negative selectors may effectively block the activation of these nucleus-resident transcrip- tion factors tipping the balance in favor of negative selection. Active ERK1 ⁄ 2 may also contribute to thymic selec- tion by regulating the balance of proapoptotic and prosurvival proteins in the cytosol (reviewed in [52]). Apoptosis induced by negative selection does not involve classical death receptor pathways, rather it depends (at least in part) on the nuclear orphan steroid receptor Nur77 (discussed later) and the proapoptotic BH3-only Bcl-2 family member Bcl2 like 11 (BIM). The prosurvival molecules Bcl-2 and Bcl-x L are able to bind and sequester the proapoptotic molecules Bax and Bak to prevent them from inducing apoptosis. Once active, BIM is able to bind Bcl-2 or Bcl-x L , lead- ing to the release and activation of Bax ⁄ Bak and apop- tosis [53]. Thymocytes from male mice lacking Bim are severely impaired for negative selection of the auto- reactive male antigen specific HY-TCR [54]. In addi- tion, the defect in negative selection in the nonobese E. Teixeiro and M. A. Daniels ERK location and T cell selection FEBS Journal 277 (2010) 30–38 ª 2009 The Authors Journal compilation ª 2009 FEBS 33 diabetic mouse strain has been linked to defective induction of BIM, among other proapototic molecules, enhancing its importance for negative selection [55,56]. Post-translational modifications can affect both the level of expression and the proapoptotic activity of BIM. For example, ERK-mediated phosphorylation of BIM can target it for ubiquitination and degradation (reviewed in [52,57]) or inhibit its proapoptotic activity by reducing its binding to the prosurvival molecules Mcl-1 and Bcl-x L [58,59]. Interestingly, JNK phospho- rylates BIM on the same residue as ERK. However, JNK also recruits the prolyl-isomerase Pin1 and induces a conformational change in BIM that enhances its proapoptotic potency in neuronal cells [60]. JNK has also been implicated in the upregulation of BIM expression [52] and JNK-mediated phosphorylation of BIM facilitates its release from sequestration by dynein motor complex [61]. Whether these findings hold true in developing thymocytes is not known. During nega- tive selection, whether BIM is regulated through tran- scription, post-translational modification or both remains to be determined. In summary, although BIM is recognized to be critically important for negative selection, the mechanism of its regulation is still unclear. Negative selecting ligands induce a rapid and robust induction of phopho-ERK1 ⁄ 2, whereas positive select- ing ligands induce slow and sustained activation of pERK1 ⁄ 2 [38,39]. These data suggest the possibility of a kinetic discrimination model for thymic selection. However, this cannot explain how strong induction of pERK1 ⁄ 2 by negative selectors does not rescue the thymocyte from apoptosis, especially in the light of the roles of pERK1 ⁄ 2 just described. Consider then, active JNK is distributed throughout the cell and has the same kinetics regardless of ligand strength [16,39]. Fur- thermore, positive selecting ligands induce pERK1 ⁄ 2 throughout the cell similar to phospho-JNK (pJNK). By contrast, negative selecting ligands lead to the acti- vation and retention of pERK at the plasma mem- brane. The net result is that negative selectors induce segregation of pERK1 ⁄ 2 and pJNK [16]. This suggests that the localization of pERK1 ⁄ 2 determines the selec- tion outcome. Along this line, targeting of the Raf ⁄ MEK ⁄ ERK MAPK module to either the cytoplasm or the plasma membrane in a neuronal cell line leads to switch-like differences in biological outcome [62]. Additionally, studies from several groups have demon- strated that the subcellular localization pattern of Ras ⁄ MAPK determines the signaling output in a vari- ety of cell types (reviewed in [8]). Given the competing roles of ERK1 ⁄ 2 and JNK in determining selection, these data suggest that retention of pERK1 ⁄ 2 at the plasma membrane mimics the effect of an ERK knock- out and gives pJNK an unopposed opportunity or at least a head start in activating the proapoptotic effec- tor molecules necessary for negative selection. Alterna- tively, a model where a unique signal is provided by membrane-bound pERK cannot be ruled out. Regard- less, it is attractive to hypothesize that the differential compartmentalization of Ras ⁄ ERK pathways provides the thymocyte with the ability to distinguish between positive and negative selecting ligands [16]. Future studies are needed to establish whether the localization pattern is sufficient to determine selection outcome. ERK5, Nur77 and negative selection The orphan steroid receptor Nur77 is part of a small family of transcription factors (which also includes Nurr1 and Nor1) that is thought to play an important role in mediating TCR-induced apoptosis in immature thymocytes. It acts in a pathway that is not redundant, but rather parallel to BIM (reviewed in [52]). The importance of Nur77 as a proapoptotic molecule in T cells was first described in hybridomas [63,64]. Using various models of negative selection, dominant nega- tive Nur77 resulted in a decrease in negative selection, whereas constitutively active mutants led to an increase in negative selection [65,66]. However, Nur77-deficient mice do not have a defect in negative selection. This apparent discrepancy can be explained by considering the redundant role of the related family member Nor1 in mediating clonal deletion and the fact that the dom- inant negative form of Nur77 is able to inhibit the function of the other family members and block dele- tion of autoreactive thymocytes [67]. Upon TCR stimulation, Nur77 transcription is upregulated through the ERK5 ⁄ MEF2 pathway [11,68,69]. The activation of Nur77 occurs by a cal- cium-dependent pathway that ultimately leads to its phosphorylation by ERK5 [70,71]. On the other hand, Akt-mediated phosphorylation of Nur77 inhib- its its DNA-binding activity [72], and there is specula- tion that ERK2 phophorylation can also inhibit Nur77 function by phosphorylation on a site distinct from the ERK5 target site [73]. Given these and other roles of Akt and ERK2 [15], these data suggest an additional mechanism by which thymocytes could distinguish positive from negative selecting ligands. The mechanism by which Nur77 mediates the induc- tion of apoptosis is less clear. Transcriptional activity correlates with apoptosis in Nur77 transgenic thymo- cytes and Nur77-deficient mice [70,74]. In fact, Nur77 induces the proapoptotic gene Nur77 downstream gene 1 ( NDG1), FasL and TNF-related apoptosis- ERK location and T cell selection E. Teixeiro and M. A. Daniels 34 FEBS Journal 277 (2010) 30–38 ª 2009 The Authors Journal compilation ª 2009 FEBS inducing ligand (TRAIL) [75], but the physiological relevance of some of these molecules in clonal dele- tion of thymocytes has not been tested. Other studies have shown that Nur77 can translocate from the nucleus to the mitochondria, where it binds to Bcl-2, converting it from a prosurvival factor into a proapo- totic molecule [76–78]. A conflicting study reported that efficient export of Nur77 was only observed in mature T cells and immature thymocytes did not translocate Nur77 to the cytosol [79]. The observed differences, which could be due to the type of cell examined, experimental technique, model system or maturation state of thymocytes tested, need to be resolved to make an accurate conclusion. Further- more, although these two models of Nur77-induced apoptosis are not necessarily mutually exclusive, it is difficult to reconcile the mitochondrial data with the correlation between transcriptional activity and apop- tosis [52]. Interestingly, the activation of ERK5, dis- pensable for positive selection, has a kinetic level of induction that is similar between positive and negative selecting ligands [11]. This resembles what has been reported for JNK and p38 [39] and again suggests that sequestration of pERK at the plasma membrane by negative selecting ligands may be necessary for the induction of signals necessary for negative selection. Conclusions Understanding the mechanisms that determine central tolerance is essential to the regulation of autoimmuni- ty, infectious disease and cancer. The mechanism by which T cells translate the parameters of ligand engagement into positive or negative selection has been elusive. The default for a preselection double-positive thymocyte is death. The time involved to complete the selection process and the requirement for intact thymic architecture have made studying the process of nega- tive selection extremely difficult. In spite of this, MAP- Ks, ERK1 ⁄ 2, ERK5, JNK and p38, are known to be involved in the induction of positive versus negative selection in the thymus. ERK5, JNK and p38 are required for negative selection and dispensable for positive selection. On the other hand, ERK1 ⁄ 2 is only involved in positive selection. At first glance, this appears to describe the mechanism of selection. Yet, the fact remains that these molecules are activated by both positive and negative selecting ligands. Differen- tial subcellular localization of these, and possibly other signaling intermediates, provides the developing thymocyte with the tools to overcome this apparent problem. Interestingly, JNK activity and subcellular location are the same regardless of the ligand strength, whereas ERK activity and location change depending of the nature of the selecting ligand. In addition, the kinetics of the other MAPK important for negative selection are the same independent of the ligand. Given their opposing function and downstream targets, this appears to give ERK a prominent role in determining the outcome of thymic selection. Further studies are needed to demonstrate how ERK and JNK compete and if location is sufficient to determine selection out- come. Acknowledgements We would like to thank Dr Bumsuk Hahm and Dr Mark McIntosh for critical reading of the manuscript and Dr Ed Palmer for his support. Work from the laboratory of MAD and ET are supported by the University of Missouri Mission Enhancement Fund. References 1 Anderson G, Lane PJ & Jenkinson EJ (2007) Generat- ing intrathymic microenvironments to establish T-cell tolerance. Nat Rev Immunol 7, 954–963. 2 Starr TK, Jameson SC & Hogquist KA (2003) Positive and negative selection of T cells. Annu Rev Immunol 21, 139–176. 3 Hammerling GJ, Schonrich G, Momburg F, Auphan N, Malissen M, Malissen B, Schmitt-Verhulst AM & Arnold B (1991) Non-deletional mechanisms of periph- eral and central tolerance: studies with transgenic mice with tissue-specific expression of a foreign MHC class I antigen. Immunol Rev 122, 47–67. 4 McGargill MA, Derbinski JM & Hogquist KA (2000) Receptor editing in developing T cells. Nat Immunol 1, 336–341. 5 Fontenot JD & Rudensky AY (2005) A well adapted regulatory contrivance: regulatory T cell development and the forkhead family transcription factor Foxp3. Nat Immunol 6, 331–337. 6 Baldwin TA, Hogquist KA & Jameson SC (2004) The fourth way? Harnessing aggressive tendencies in the thymus. J Immunol 173, 6515–6520. 7 Alberola-Ila J & Herna ´ ndez-Hoyos G (2003) The Ras ⁄ MAPK cascade and the control of positive selection. Immunol Rev 191, 79–96. 8 Mor A & Philips MR (2006) Compartmentalized Ras ⁄ MAPK signaling. Annu Rev Immunol 24, 771– 800. 9 Rinco ´ n M, Whitmarsh A, Yang DD, Weiss L, De ´ rijard B, Jayaraj P, Davis RJ & Flavell RA (1998) The JNK pathway regulates the in vivo deletion of immature CD4(+)CD8(+) thymocytes. J Exp Med 188, 1817– 1830. E. Teixeiro and M. A. Daniels ERK location and T cell selection FEBS Journal 277 (2010) 30–38 ª 2009 The Authors Journal compilation ª 2009 FEBS 35 10 Sugawara T, Moriguchi T, Nishida E & Takahama Y (1998) Differential roles of ERK and p38 MAP kinase pathways in positive and negative selection of T lym- phocytes. Immunity 9, 565–574. 11 Sohn SJ, Lewis G & Winoto A (2008) Non-redundant function of the MEK5-ERK5 pathway in thymocyte apoptosis. EMBO J 27, 1896–1906. 12 Alberola-Ila J, Hogquist KA, Swan KA, Bevan MJ & Perlmutter RM (1996) Positive and negative selection invoke distinct signaling pathways. J Exp Med 184, 9–18. 13 Alberola-Ila J, Forbush KA, Seger R, Krebs EG & Perlmutter RM (1995) Selective requirement for MAP kinase activation in thymocyte differentiation. Nature 373, 620–623. 14 Page ` s G, Gue ´ rin S, Grall D, Bonino F, Smith A, Anjuere F, Auberger P & Pouysse ´ gur J (1999) Defective thymocyte maturation in p44 MAP kinase (Erk 1) knockout mice. Science 286, 1374–1377. 15 Fischer AM, Katayama CD, Page ` s G, Pouysse ´ gur J & Hedrick SM (2005) The role of erk1 and erk2 in mul- tiple stages of T cell development. Immunity 23, 431–443. 16 Daniels MA, Teixeiro E, Gill J, Hausmann B, Roubaty D, Holmberg K, Werlen G, Holla ¨ nder GA, Gascoigne NR & Palmer E (2006) Thymic selection threshold defined by compartmentalization of Ras ⁄ MAPK signal- ling. Nature 444, 724–729. 17 Hancock JF (2003) Ras proteins: different signals from different locations. Nat Rev Mol Cell Biol 4, 373–384. 18 Germain RN & Stefanova I (1999) The dynamics of T cell receptor signaling: complex orchestration and the key roles of tempo and cooperation. Annu Rev Immunol 17, 467–522. 19 Sloan-Lancaster J, Shaw AS, Rothbard JB & Allen PM (1994) Partial T cell signaling: altered phospho-zeta and lack of zap70 recruitment in APL-induced T cell anergy. Cell 79, 913–922. 20 Madrenas J, Wange RL, Wang JL, Isakov N, Samelson LE & Germain RN (1995) Zeta phosphorylation with- out ZAP-70 activation induced by TCR antagonists or partial agonists. Science 267, 515–518. 21 Smyth LA, Williams O, Huby RD, Norton T, Acuto O, Ley SC & Kioussis D (1998) Altered peptide ligands induce quantitatively but not qualitatively different intracellular signals in primary thymocytes. Proc Natl Acad Sci USA 95, 8193–8198. 22 Love PE & Shores EW (2000) ITAM multiplicity and thymocyte selection: how low can you go? Immunity 12, 591–597. 23 Holst J, Wang H, Eder KD, Workman CJ, Boyd KL, Baquet Z, Singh H, Forbes K, Chruscinski A, Smeyne R et al. (2008) Scalable signaling mediated by T cell antigen receptor-CD3 ITAMs ensures effective negative selection and prevents autoimmunity. Nat Immunol 9, 658–666. 24 van Oers NS, Love PE, Shores EW & Weiss A (1998) Regulation of TCR signal transduction in murine thymocytes by multiple TCR zeta-chain signaling motifs. J Immunol 160, 163–170. 25 Sommers CL, Samelson LE & Love PE (2004) LAT: a T lymphocyte adapter protein that couples the antigen receptor to downstream signaling pathways. BioEssays 26, 61–67. 26 Sommers CL, Menon RK, Grinberg A, Zhang W, Samelson LE & Love PE (2001) Knock-in mutation of the distal four tyrosines of linker for activation of T cells blocks murine T cell development. J Exp Med 194, 135–142. 27 Wang CR, Hashimoto K, Kubo S, Yokochi T, Kubo M, Suzuki M, Suzuki K, Tada T & Nakayama T (1995) T cell receptor-mediated signaling events in CD4+CD8+ thymocytes undergoing thymic selection: requirement of calcineurin activation for thymic positive selection but not negative selection. J Exp Med 181, 927–941. 28 Morley SC, Weber KS, Kao H & Allen PM (2008) Protein kinase C-theta is required for efficient positive selection. J Immunol 181, 4696–4708. 29 Dower NA, Stang SL, Bottorff DA, Ebinu JO, Dickie P, Ostergaard HL & Stone JC (2000) RasGRP is essential for mouse thymocyte differentiation and TCR signaling. Nat Immunol 1 , 317–321. 30 Priatel JJ, Teh SJ, Dower NA, Stone JC & Teh HS (2002) RasGRP1 transduces low-grade TCR signals which are critical for T cell development, homeostasis, and differentiation. Immunity 17, 617–627. 31 Yoder J, Pham C, Iizuka YM, Kanagawa O, Liu SK, McGlade J & Cheng AM (2001) Requirement for the SLP-76 adaptor GADS in T cell development. Science 291, 1987–1991. 32 Maltzman JS, Kovoor L, Clements JL & Koretzky GA (2005) Conditional deletion reveals a cell-autonomous requirement of SLP-76 for thymocyte selection. J Exp Med 202, 893–900. 33 Houtman JC, Barda-Saad M & Samelson LE (2005) Examining multiprotein signaling complexes from all angles. FEBS J 272, 5426–5435. 34 Prince AL, Yin CC, Enos ME, Felices M & Berg LJ (2009) The Tec kinases Itk and Rlk regulate conven- tional versus innate T-cell development. Immunol Rev 228, 115–131. 35 Gong Q, Cheng AM, Akk AM, Alberola-Ila J, Gong G, Pawson T & Chan AC (2001) Disruption of T cell signaling networks and development by Grb2 haploid insufficiency. Nat Immunol 2, 29–36. 36 Prasad A, Zikherman J, Das J, Roose JP, Weiss A & Chakraborty AK (2009) Origin of the sharp boundary ERK location and T cell selection E. Teixeiro and M. A. Daniels 36 FEBS Journal 277 (2010) 30–38 ª 2009 The Authors Journal compilation ª 2009 FEBS that discriminates positive and negative selection of thymocytes. Proc Natl Acad Sci USA 106, 528–533. 37 Houtman JC, Houghtling RA, Barda-Saad M, Toda Y & Samelson LE (2005) Early phosphorylation kinetics of proteins involved in proximal TCR-mediated signal- ing pathways. J Immunol 175, 2449–2458. 38 Mariathasan S, Zakarian A, Bouchard D, Michie AM, Zu´ n ˜ iga-Pflu ¨ cker JC & Ohashi PS (2001) Duration and strength of extracellular signal-regulated kinase signals are altered during positive versus negative thymocyte selection. J Immunol 167, 4966–4973. 39 Werlen G, Hausmann B & Palmer E (2000) A motif in the alphabeta T-cell receptor controls positive selection by modulating ERK activity. Nature 406, 422–426. 40 Das J, Ho M, Zikherman J, Govern C, Yang M, Weiss A, Chakraborty AK & Roose JP (2009) Digital signaling and hysteresis characterize ras activation in lymphoid cells. Cell 136, 337–351. 41 Mor A, Campi G, Du G, Zheng Y, Foster DA, Dustin ML & Philips MR (2007) The lymphocyte function- associated antigen-1 receptor costimulates plasma mem- brane Ras via phospholipase D2. Nat Cell Biol 9, 713–719. 42 Carrasco S & Merida I (2004) Diacylglycerol-dependent binding recruits PKCtheta and RasGRP1 C1 domains to specific subcellular localizations in living T lympho- cytes. Mol Biol Cell 15, 2932–2942. 43 Zha Y, Marks R, Ho AW, Peterson AC, Janardhan S, Brown I, Praveen K, Stang S, Stone JC & Gajewski TF (2006) T cell anergy is reversed by active Ras and is reg- ulated by diacylglycerol kinase-alpha. Nat Immunol 7, 1166–1173. 44 Giurisato E, Lin J, Harding A, Cerutti E, Cella M, Lewis RE, Colonna M & Shaw AS (2009) The mitogen- activated protein kinase scaffold KSR1 is required for recruitment of extracellular signal-regulated kinase to the immunological synapse. Mol Cell Biol 29, 1554– 1564. 45 Laurent MN, Ramirez DM & Alberola-Ila J (2004) Kinase suppressor of Ras couples Ras to the ERK cas- cade during T cell development. J Immunol 173, 986– 992. 46 Costello PS, Nicolas RH, Watanabe Y, Rosewell I & Treisman R (2004) Ternary complex factor SAP-1 is required for Erk-mediated thymocyte positive selection. Nat Immunol 5, 289–298. 47 Miyazaki T & Lemonnier FA (1998) Modulation of thymic selection by expression of an immediate-early gene, early growth response 1 (Egr-1). J Exp Med 188, 715–723. 48 Bettini M, Xi H, Milbrandt J & Kersh GJ (2002) Thy- mocyte development in early growth response gene 1-deficient mice. J Immunol 169, 1713–1720. 49 Bain G, Cravatt CB, Loomans C, Alberola-Ila J, Hedrick SM & Murre C (2001) Regulation of the helix–loop–helix proteins, E2A and Id3, by the Ras-ERK MAPK cascade. Nat Immunol 2, 165–171. 50 Bain G, Quong MW, Soloff RS, Hedrick SM & Murre C (1999) Thymocyte maturation is regulated by the activity of the helix–loop–helix protein, E47. J Exp Med 190, 1605–1616. 51 Rivera RR, Johns CP, Quan J, Johnson RS & Murre C (2000) Thymocyte selection is regulated by the helix– loop–helix inhibitor protein, Id3. Immunity 12, 17–26. 52 Strasser A, Puthalakath H, O’Reilly L & Bouillet P (2008) What do we know about the mechanisms of elimination of autoreactive T and B cells and what challenges remain. Immunol Cell Biol 86, 57–66. 53 Cheng EH, Wei MC, Weiler S, Flavell RA, Mak TW, Lindsten T & Korsmeyer SJ (2001) BCL-2, BCL-X(L) sequester BH3 domain-only molecules preventing BAX- and BAK-mediated mitochondrial apoptosis. Mol Cell 8, 705–711. 54 Bouillet P, Purton JF, Godfrey DI, Zhang LC, Coultas L, Puthalakath H, Pellegrini M, Cory S, Adams JM & Strasser A (2002) BH3-only Bcl-2 family member Bim is required for apoptosis of autoreactive thymocytes. Nature 415, 922–926. 55 Liston A, Lesage S, Gray DH, O’Reilly L, Strasser A, Fahrer AM, Boyd RL, Wilson J, Baxter AG, Gallo EM et al. (2004) Generalized resistance to thymic deletion in the NOD mouse; a polygenic trait characterized by defective induction of Bim. Immunity 21, 817–830. 56 Zucchelli S, Holler P, Yamagata T, Roy M, Benoist C & Mathis D (2005) Defective central tolerance induction in NOD mice: genomics and genetics. Immunity 22, 385–396. 57 Ley R, Ewings KE, Hadfield K & Cook SJ (2005) Reg- ulatory phosphorylation of Bim: sorting out the ERK from the JNK. Cell Death Differ 12, 1008–1014. 58 Bunin A, Khwaja FW & Kersh GJ (2005) Regulation of Bim by TCR signals in CD4 ⁄ CD8 double-positive thymocytes. J Immunol 175, 1532–1539. 59 Ewings KE, Hadfield-Moorhouse K, Wiggins CM, Wickenden JA, Balmanno K, Gilley R, Degenhardt K, White E & Cook SJ (2007) ERK1 ⁄ 2-dependent phos- phorylation of BimEL promotes its rapid dissociation from Mcl-1 and Bcl-xL. EMBO J 26, 2856–2867. 60 Becker EB & Bonni A (2006) Pin1 mediates neural- specific activation of the mitochondrial apoptotic machinery. Neuron 49, 655–662. 61 Puthalakath H, Huang DC, O’Reilly LA, King SM & Strasser A (1999) The proapoptotic activity of the Bcl-2 family member Bim is regulated by interaction with the dynein motor complex. Mol Cell 3, 287–296. 62 Harding A, Tian T, Westbury E, Frische E & Hancock JF (2005) Subcellular localization determines MAP kinase signal output. Curr Biol 15, 869–873. 63 Liu ZG, Smith SW, McLaughlin KA, Schwartz LM & Osborne BA (1994) Apoptotic signals delivered through E. Teixeiro and M. A. Daniels ERK location and T cell selection FEBS Journal 277 (2010) 30–38 ª 2009 The Authors Journal compilation ª 2009 FEBS 37 the T-cell receptor of a T-cell hybrid require the imme- diate-early gene nur77. Nature 367, 281–284. 64 Woronicz JD, Calnan B, Ngo V & Winoto A (1994) Requirement for the orphan steroid receptor Nur77 in apoptosis of T-cell hybridomas. Nature 367, 277–281. 65 Calnan BJ, Szychowski S, Chan FK, Cado D & Winoto A (1995) A role for the orphan steroid receptor Nur77 in apoptosis accompanying antigen-induced negative selection. Immunity 3, 273–282. 66 Zhou T, Cheng J, Yang P, Wang Z, Liu C, Su X, Bluethmann H & Mountz JD (1996) Inhibition of Nur77 ⁄ Nurr1 leads to inefficient clonal deletion of self-reactive T cells. J Exp Med 183, 1879–1892. 67 Cheng LE, Chan FK, Cado D & Winoto A (1997) Functional redundancy of the Nur77 and Nor-1 orphan steroid receptors in T-cell apoptosis. EMBO J 16, 1865– 1875. 68 Kasler HG, Victoria J, Duramad O & Winoto A (2000) ERK5 is a novel type of mitogen-activated protein kinase containing a transcriptional activation domain. Mol Cell Biol 20, 8382–8389. 69 Sohn SJ, Li D, Lee LK & Winoto A (2005) Transcrip- tional regulation of tissue-specific genes by the ERK5 mitogen-activated protein kinase. Mol Cell Biol 25, 8553–8566. 70 Woronicz JD, Lina A, Calnan BJ, Szychowski S, Cheng L & Winoto A (1995) Regulation of the Nur77 orphan steroid receptor in activation-induced apoptosis. Mol Cell Biol 15, 6364–6376. 71 Fujii Y, Matsuda S, Takayama G & Koyasu S (2008) ERK5 is involved in TCR-induced apoptosis through the modification of Nur77. Genes Cells 13, 411–419. 72 Masuyama N, Oishi K, Mori Y, Ueno T, Takahama Y & Gotoh Y (2001) Akt inhibits the orphan nuclear receptor Nur77 and T-cell apoptosis. J Biol Chem 276, 32799–32805. 73 Katagiri Y, Takeda K, Yu ZX, Ferrans VJ, Ozato K & Guroff G (2000) Modulation of retinoid signalling through NGF-induced nuclear export of NGFI-B. Nat Cell Biol 2, 435–440. 74 Kuang AA, Cado D & Winoto A (1999) Nur77 tran- scription activity correlates with its apoptotic function in vivo. Eur J Immunol 29, 3722–3728. 75 Rajpal A, Cho YA, Yelent B, Koza-Taylor PH, Li D, Chen E, Whang M, Kang C, Turi TG & Winoto A (2003) Transcriptional activation of known and novel apoptotic pathways by Nur77 orphan steroid receptor. EMBO J 22, 6526–6536. 76 Lin B, Kolluri SK, Lin F, Liu W, Han YH, Cao X, Dawson MI, Reed JC & Zhang XK (2004) Conversion of Bcl-2 from protector to killer by interaction with nuclear orphan receptor Nur77 ⁄ TR3. Cell 116, 527–540. 77 Li H, Kolluri SK, Gu J, Dawson MI, Cao X, Hobbs PD, Lin B, Chen G, Lu J, Lin F et al. (2000) Cyto- chrome c release and apoptosis induced by mitochon- drial targeting of nuclear orphan receptor TR3. Science 289, 1159–1164. 78 Thompson J & Winoto A (2008) During negative selection, Nur77 family proteins translocate to mito- chondria where they associate with Bcl-2 and expose its proapoptotic BH3 domain. J Exp Med 205, 1029– 1036. 79 Cunningham NR, Artim SC, Fornadel CM, Sellars MC, Edmonson SG, Scott G, Albino F, Mathur A & Punt JA (2006) Immature CD4+CD8+ thymocytes and mature T cells regulate Nur77 distinctly in response to TCR stimulation. J Immunol 177 , 6660–6666. ERK location and T cell selection E. Teixeiro and M. A. Daniels 38 FEBS Journal 277 (2010) 30–38 ª 2009 The Authors Journal compilation ª 2009 FEBS . at the plasma membrane and lead to Ras-GRP1 recruitment to that location. Recent work has demonstrated that in T cells, PLCc1 activa- tion leads to activation of Ras-GRP1 and its recruit- ment. hypothesize that the differential compartmentalization of Ras ⁄ ERK pathways provides the thymocyte with the ability to distinguish between positive and negative selecting ligands [16]. Future studies. regardless of the ligand strength, whereas ERK activity and location change depending of the nature of the selecting ligand. In addition, the kinetics of the other MAPK important for negative selection