View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector Immunity Article

Agonist-Selected Development Requires Strong T Cell Receptor Signaling and Store-Operated Calcium Entry

Masatsugu Oh-hora,1,2,3,4,* Noriko Komatsu,1,3 Mojgan Pishyareh,2 Stefan Feske,5 Shohei Hori,6 Masaru Taniguchi,7 Anjana Rao,2,9,* and Hiroshi Takayanagi1,3,8,10 1Department of Cell Signaling, Graduate School of Medical and Dental Sciences, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8549, Japan 2Department of Pathology, Immune Disease Institute and Program in Cellular and Molecular Medicine, Children’s Hospital, Harvard Medical School, 200 Longwood Avenue, Boston, MA 02115, USA 3Global Center of Excellence (GCOE) Program, International Research Center for Molecular Science in Tooth and Bone Diseases, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8549, Japan 4Japan Science and Technology Agency (JST), Precursory Research for Embryonic Science and Technology (PRESTO) program, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8549, Japan 5Department of Pathology, New York University, School of Medicine, New York, NY 10016, USA 6Research Unit for Immune 7Laboratory for Immune Regulation RIKEN Research Center for Allergy and Immunology, 1-7-22 Suehiro-cho, Tsurumi-ku, Yokohama, Kanagawa 230-0045, Japan 8Japan Science and Technology Agency (JST), Explorative Research for Advanced Technology (ERATO) program, Takayanagi Osteonetwork Project, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8549, Japan 9Present address: Division of Signaling and Expression, La Jolla Institute for Allergy & Immunology, La Jolla, CA 92037, USA 10Present address: Department of Immunology, Graduate School of Medicine and Faculty of Medicine, The University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo 113-0033, Japan *Correspondence: [email protected] (M.O.), [email protected] (A.R.) http://dx.doi.org/10.1016/j.immuni.2013.02.008

SUMMARY pMHC) interactions. In conventional TCRab+ T cell development, weak or absent TCR-pMHC interactions cannot support thymo- T cell receptor (TCR) signaling driven by interaction cyte survival, leading to death by neglect. Moderate-affinity of the TCR with specific complexes of self- TCR-pMHC interactions lead to the development of functional and the major histocompatibility complex deter- T cells through positive selection. High-affinity TCR-pMHC inter- mines T cell fate in thymic development. However, actions normally trigger apoptosis of self-reactive thymocytes the signaling pathway through which TCR signal through negative selection. However, some self-reactive thymo- strength regulates distinct T cell lineages remains cytes mature into unconventional T-lineage cells through an alternative selection process defined as agonist selection (Bald- unknown. Here we have used mice lacking the endo- 2+ win et al., 2004; Stritesky et al., 2012). Agonist-selected uncon- plasmic reticulum Ca sensors stromal interaction ventional T cell subsets are thought to have a regulatory role in molecule 1 (STIM1) and STIM2 to show that STIM- and are classified into three main cell types: 2+ induced store-operated Ca entry is not essential forkhead box P3 (Foxp3)+ regulatory T (Treg) cells, invariant + for thymic development of conventional TCRab natural killer T (iNKT) cells, and TCRab+ CD8aa+ intestinal intra- T cells but is specifically required for the develop- epithelial lymphocytes (IELs) (Hsieh et al., 2012; Kronenberg and ment of agonist-selected T cells (regulatory T cells, Gapin, 2002; Lambolez et al., 2007). It has been proposed that invariant natural killer T cells, and TCRab+ CD8aa+ these T cells require relatively strong and sustained TCR signals intestinal intraepithelial lymphocytes). The severe for their development (Baldwin et al., 2004). Although this affinity impairment of agonist-selected T cell development model is acknowledged, there remains a longstanding question is mainly due to a defect in interleukin-2 (IL-2) or concerning how the TCR signal strength and duration regulates the development of these distinct T cell subsets. IL-15 signaling. Thus, STIM1 and STIM2-mediated 2+ Engagement of TCR-pMHC activates several tyrosine store-operated Ca influx, leading to efficient kinases and ultimately phospholipase C (PLC)-g1. Activated activation of NFAT (nuclear factor of activated PLC-g1 hydrolyzes phosphatidylinositol 4,5-bisphosphate into

T cells), is critical for the postselection maturation diacylglycerol (DAG) and inositol-1,4,5-trisphosphate (IP3), of agonist-selected T cells. which induces the release of Ca2+ from the endoplasmic retic- ulum (ER). In turn, ER store depletion triggers store-operated INTRODUCTION Ca2+ entry, the predominant mechanism for sustained increase 2+ 2+ of intracellular free Ca ([Ca ]i) downstream of the TCR. Thymic selection depends on the affinity of T cell receptor- Store-operated Ca2+ entry leads to activation of the phospha- peptide major histocompatibility complex glycoprotein (TCR- tase calcineurin, which in turn activates the transcription factor

Immunity 38, 881–895, May 23, 2013 ª2013 Elsevier Inc. 881 Immunity T Cell Development and Calcium Signaling

A C 220 Ca2+

WT 1KO 4 DKO 4 4 10 10 10 10.7 84.5 10.3 83.6 13.3 80.3 3 3 μ 1200 3 10 TG (1 M) 10 10 1200

2 2 10 2 10 (nM) WT 10 (nM) i i 800 ]

800 ] 2+ 1 2+ 1 1 10

CD4 10 10 400 400 [Ca [Ca 0 3.7 10 0 4.9 10 0 3.5 10 0 1 2 3 4 DKO * 0 1 2 3 4 10 10 10 10 10 10 0 10 1 102 103 104 10 10 10 10 10 CD8 0 0 0 200 400 600 WT DKO B 0Ca2 2+ WT 1KO DKO WT 1KO DKO 800 Bio- S. Avidin WT 1200 STIM1 STIM2 600 αCD3 (nM) i ]

(nM) 800 i ] 2+ 400 2+

[Ca 200 400 DKO [Ca *** Actin Actin 0 0 0 200 400 600 WT DKO Time (s) D Male HY TCR tg E PBS SEB 30 Lymph nodes 25 4 Thymus 4 10 10 20 20

3 3 15 10 10 Control 58.8 10 34.6 10 10 2 1.25 10 2 13.4 Control 5

0 Gated on 10 1 10 1 0 10 0 10 1 102 103 104 10 0 10 1 102 103 104

50 + + 25 β 10 0 10 0 CD4 TCRV 8 10 0 10 1 102 103 104 10 0 10 1 102 103 104 4 4 40 10 10 20

30 15 3 3 10 10 DKO 20 71.7 59.4 DKO 10 2 2 10 2.86 10 15.2 10 5

0 0 1 1 10 10 10 0 10 1 102 103 104 10 0 10 1 102 103 104 CD24 10 0 10 0 CD4 10 0 10 1 102 103 104 10 0 10 1 102 103 104 CD8 Vβ8Vβ6 80 NS 80 NS Thymus LNs 3 20 NS ** NS 60 60 15 ** 2 40 40 10 T cells (%) immature cells (%) immature cells (%) T cells (%) 20 20

1 high high high high 5 CD8

CD8 0 0 CD24 00 CD24 PBS SEB PBS SEB PBS SEB PBS SEB DKO DKO Control Control Control DKO Control DKO F Female HY TCR tg

4 Thymus 4Lymph nodes 10 10 600 Thymus Lymph nodes 3 3 30 80 10 10

400 Control 10 2 11.5 10 2 37.1 60 200 20 10 1 10 1 *** 40 0 0 0 **

10 10 T cells (%) NS 10 0 10 1 102 103 104 T cells (%) 10 0 10 1 102 103 104 10 0 10 1 102 103 104 4 4 **

10 10 high

high 10 20 300 3 3 10 10 CD8 CD8

DKO 200 0 0 10 2 8.0 10 2 17.6

100 10 1 10 1 DKO DKO DKO DKO Control Control Control Control

0 10 0 10 0 Cell number 0 1 2 3 4 CD4 10 10 10 10 10 10 0 10 1 102 103 104 10 0 10 1 102 103 104 6w 8w 6w 8w CD8 T3.70

(legend on next page) 882 Immunity 38, 881–895, May 23, 2013 ª2013 Elsevier Inc. Immunity T Cell Development and Calcium Signaling

NFAT (Feske, 2007; Hogan et al., 2010). The induction of store- tion of IL-15 partially rescued differentiation into TCRab+ operated Ca2+ entry is controlled by two major molecules, the CD8aa+ IELs. These results suggest that postselection matura- ER Ca2+ sensor stromal interaction molecule 1 (STIM1) (Liou tion of all agonist-selected precursors requires continuous high et al., 2005; Roos et al., 2005) and calcium release-activated activity of NFAT by TCR-mediated store-operated Ca2+ entry, calcium (CRAC) channels ORAI1 (Feske et al., 2006; Vig et al., to induce efficient expression of downstream of NFAT 2006; Zhang et al., 2006). STIM1 is an acknowledged positive and NFAT-target transcription factors. regulator of store-operated CRAC channels. Loss of STIM1 abrogates TCR-induced store-operated Ca2+ entry and NFAT RESULTS activation, resulting in impaired proliferation and cytokine production by peripheral human and mouse T cells (McCarl Store-Operated Ca2+ Entry Is Dispensable for the et al., 2010; Oh-Hora et al., 2008; Picard et al., 2009). The related Development of Conventional TCRab+ T Cells protein STIM2 regulates sustenance of calcium entry and NFAT In both humans and mice, deficiency of STIM1 alone does not activation in mouse CD4+ T cells (Oh-Hora et al., 2008), but also influence the number of conventional TCRab+ T cells in the 2+ regulates basal concentration of [Ca ]i in HeLa cells (Brandman periphery (Beyersdorf et al., 2009; Picard et al., 2009) (data not et al., 2007). shown), suggesting that STIM2 may compensate for Ca2+ influx In thymocytes, TCR signal strength well correlates with magni- in thymocytes. To examine the role of STIM-dependent store- tude and duration of Ca2+ influx. An in vitro study demonstrated operated Ca2+ entry in thymocyte development, we first gener- that a strong TCR signal elicited by promoting negative ated chimeric mice by transplanting fetal liver cells derived 2+ fl/fl / fl/fl fl/fl selection sustained a high concentration of [Ca ]i with large from Stim1 Cmv-Cre (Stim1 ) mice or Stim1 Stim2 Ca2+ influx, whereas a weak TCR signal by peptides promoting Cmv-Cre (Stim1/Stim2/) mice into CD45.1+ Rag2-deficient positive selection induces a small Ca2+ influx and increased mice. Surprisingly, the resulting chimeric mice displayed normal 2+ + [Ca ]i concentration gradually (Daniels et al., 2006; Nakayama development of conventional TCRab T cells (Figure 1A), in et al., 1992). In contrast, an ex vivo study showed that thymo- contrast to the reported phenotype of mice with a disruption cytes undergoing positive selection showed a substantial of Cnb1, encoding a regulatory subunit of the Ca2+-dependent 2+ 2+ increase of [Ca ]i through sustained Ca oscillations (Bhakta phosphatase calcineurin (Neilson et al., 2004). We con- et al., 2005). Because store-operated Ca2+ entry provides both firmed that CD4+CD8+ double-positive (DP) thymocytes from large and sustained Ca2+ influx with T cells, store-operated Stim1/Stim2/ mice did not express STIM1 and STIM2 Ca2+ entry has long been thought to be a critical Ca2+ entry and showed no store-operated Ca2+ entry after stimula- pathway in T cell development. However, there is no direct tion with or anti-CD3 (Figures 1B and 1C). To assess evidence for this assumption. the role of STIM proteins in thymocyte development in more To elucidate the role of Ca2+ influx during T cell ontogeny, we detail, we established Stim1fl/flStim2fl/fl Vav-iCre mice. Again, analyzed mice in which STIM1 and its homolog STIM2 were store-operated Ca2+ entry was abrogated in stimulated deleted in T cells or hematopoietic cells. We found that STIM- CD4CD8 double-negative (DN) and DP thymocytes from dependent store-operated Ca2+ entry is not essential for the Stim1fl/flStim2fl/fl Vav-iCre mice (see Figures S1A and S1B avail- development or positive selection of conventional TCRab+ able online). Despite this defect, thymic cell numbers and T cell but specifically regulates the development of agonist- expression of the cell surface markers CD4 and CD8 were selected T cells. The ablation of STIM1 and STIM2 significantly comparable to those in control mice (Figures S1C and S1D). compromised the cytokine-driven expansion and functional Thus, although store-operated Ca2+ entry exists in a functional maturation of agonist-selected precursors. Absence of store- form in DP thymocytes, it is not absolutely essential for thymic operated Ca2+ entry resulted in substantially decreased expres- development of conventional TCRab+ T cells. 2+ sion of NFAT target genes, which led to impaired upregulation of Sustained oscillations of [Ca ]i are essential for positive Il2rb in iNKT cells and TCRab+ CD8aa+ IELs. The administration selection of conventional TCRab+ T cells in the thymus of agonist complexes of IL-2 and anti-IL-2 rescued Treg cell (Bhakta et al., 2005). To examine the function of STIM proteins proliferation, but not suppressive function, whereas administra- in positive and negative selection of thymocytes in vivo, we

Figure 1. STIM-Dependent Store-Operated Ca2+ Entry Is Dispensable for Thymic Selection of T Cells (A) Conventional T cell development in mice transplanted with fetal liver cells isolated from Stim1/ (1KO), Stim1/Stim2/ (DKO), and wild-type (WT, Stim1+/+Stim2+/+) littermate control embryos at embryonic day 14.5. (B) Protein expression of STIM1 (left) and STIM2 (right) in DP thymocytes from (A). (C) Ca2+ influx in response to 1 mM thapsigargin (TG) (top, n = 5) or anti-CD3 crosslinking (bottom, n = 3) in littermate wild-type control (black) and Stim1/Stim2/ (gray) DP thymocytes from (A). (D) Flow cytometric analysis of negative selection in male Stim1+/+Stim2+/+ Lck-Cre Rag2/ HY TCR-transgenic (tg) mice (control) and Stim1fl/flStim2fl/fl Lck-Cre Rag2/ HY TCR-tg mice (DKO) (top) and frequency of CD8hi T cells in the thymus or lymph nodes (bottom). Three of six weeks old mice were analyzed in each case. (E) SEB-induced negative selection in Stim1fl/flStim2fl/fl (control) and Stim1fl/flStim2fl/fl Vav-iCre (DKO) mice. Frequencies of SEB-sensitive CD24hiCD4+TCRVb8+ and SEB-insensitive CD24hiCD4+TCRVb6+ immature thymocytes were analyzed 2 days after the administration of 200 mg SEB. Four mice were analyzed in each case. (F) Flow cytometric analysis of positive selection in female Stim1+/+Stim2+/+ Lck-Cre Rag2/ HY TCR-tg mice (control) and Stim1fl/flStim2fl/fl Lck-Cre Rag2/ HY TCR-tg mice (DKO) (left and center) and frequency of CD8hi T cells in the thymus or lymph nodes (right). Three mice were analyzed in each case. Data are representative of two (A, B, C, and E) and more than three (D and F) independent experiments. Error bars (C, D, E, and F) denote mean ± SEM. *p < 0.001; **p < 0.03; ***p < 0.05; NS, not significant. See also Figure S1.

Immunity 38, 881–895, May 23, 2013 ª2013 Elsevier Inc. 883 Immunity T Cell Development and Calcium Signaling

A Stim1fl/flStim2+/+ Stim1fl/flStim2fl/fl Vav-iCre Vav Vav 4 4 -iCre 4 -iCre 10 10 10 12 NS

3 3 3 10 9.36 10 8.3 10 1.43 8 + hi 10 2 10 2 10 2 Treg cells Foxp3 CD25 Treg cells + + 4 in splenic CD4 Tcells (%) 10 1 10 1 10 1 β CD25 (CD4 TCR ) * 10 0 10 0 10 0 0 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 Foxp3 4 4 4 10 10 10 40

3 3 3 10 10 10 30 iNKT cells 2 33.3 2 1.52 2 10 10 10 0.33 20 iNKT cells in CD19- liver mononuclear cells (%) - 10 1 10 1 10 1 (CD19 ) 10 *

CD1d-tet * 10 0 10 0 10 0 0 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 TCRβ 4 4 4 10 10 10 25

3 3 3 20 10 10 10 NS

β TCRαβ+ 15 10 2 10 2 10 2 CD8αα+ IELs CD8αα+ IELs 10 in intestinal TCRαβ+ T cells (%) 10 1 10 1 10 1 CD8 5 12.8 14.7 0.17 * 10 0 10 0 10 0 0 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 α CD8 1KO DKO Control B C Stim1fl/flStim2fl/fl Stim1fl/flStim2fl/fl Gated on whole Wild-type Vav-iCre Lck Lck thymocytes + + 4 -Cre 4 -Cre 4 4 (CD45.1 ) 4 (CD45.2 ) 10 10 10 10 10

3 3 3 3 3 10 10 10 41.4 10 10 11.9 2.16 10 2 10 2 10 2 10 2 15 10 2 2.81 Treg cells 52.8 Treg cells CD25 CD25 + + + + 10 1 10 1 (CD4 TCRβ ) 10 1 10 1 10 1 (CD4 TCRβ ) 15.4 0.93 10 0 10 0 10 0 10 0 10 0 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 DKO(CD45.2) Foxp3 Wild-type Foxp3

4 4 4 4 10 10 (CD45.1) 10 10

3 3 3 3 10 10 10 2.45 10 0.067

10 2 15.9 10 2 0.21 iNKT cells 10 2 10 2 iNKT cells - - 10 1 10 1 (CD19 ) 10 1 10 1 (CD19 ) CD1d-tet CD1d-tet

10 0 10 0 10 0 10 0 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 TCRβ TCRβ

4 4 4 4 10 10 10 10

3 3 3 3 10 10 10 10 β TCRαβ+ β TCRαβ+ 2 2 2 2 10 10 + 10 10 + CD8αα IELs 6.31 0.25 CD8αα CD8 10 1 10 1 CD8 10 1 10 1 IELs 46.7 0.24 10 0 10 0 10 0 10 0 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 CD8α CD8α

D Treg cells E iNKT cells F IELs 15 35 15 50 20 15 NS 30 ** ** *** 40 * 25 15 10 10 10 20 30 IELs (%)

cells (%) 10 ** + cells (%) cells (%) cells (%) + + + 15 20 + 5 5 αα 5 10

iNKT cells (%) 5 BrdU BrdU BrdU

Foxp3 10 5 CD8 0 0 0 0 0 0

DKO DKO DKO DKO DKO DKO Control Control Control Control Control Control

(legend on next page) 884 Immunity 38, 881–895, May 23, 2013 ª2013 Elsevier Inc. Immunity T Cell Development and Calcium Signaling

crossed Stim1fl/fl Stim2fl/fl Lck-Cre mice to HY TCR-transgenic to observe a major reduction in the numbers of Foxp3+ Treg cells (TCR-tg) Rag2/ mice (Kisielow et al., 1988). In male STIM-defi- and TCRab+ CD8aa+ IELs (Figure 2A), as we previously showed cient HY TCR-tg mice, there was a moderate increase in the for Foxp3+ Treg cells (Oh-Hora et al., 2008). Consistent with this number of CD8+ T cells in the thymus compared to littermate observation, a recent report also demonstrated that STIM1 defi- controls (Figure 1D), indicating that loss of store-operated Ca2+ ciency alone causes a severe reduction of iNKT cells in the entry might affect the efficiency of negative selection. To further periphery in humans (Fuchs et al., 2012). Thus these three assess the contribution of store-operated Ca2+ entry to positive agonist-selected subsets of T cells show different hierarchies and negative selection in vivo, we injected staphylococcal of dependence on store-operated Ca2+ entry. Moreover, the enterotoxin B (SEB) intraperitoneally into littermate control and requirement for store-operated Ca2+ entry is cell-intrinsic: in Stim1fl/flStim2fl/fl Vav-iCre mice (Kishimoto et al., 1998). In competitive bone marrow chimeras generated by transplanta- contrast to littermate control mice, Stim1fl/flStim2fl/fl Vav-iCre tion of a 1:1 mixture of Stim1fl/flStim2fl/fl Vav-iCre (CD45.2) mice displayed a higher frequency of TCRVb8+CD4+CD24hi bone marrow and wild-type (CD45.1) bone marrow into irradi- immature thymocytes (58.8% versus 71%; Figure 1E) but no ated Rag1/ mice, STIM-deficient bone marrow gave rise to difference of CD24lo mature CD4+ T cells in the periphery (Fig- fewer numbers of all three agonist-selected subsets of T-lineage ure S1E), suggesting delayed positive selection of immature cells (Figure 2C). thymocytes after positive selection. After SEB injection, the cor- responding numbers of CD24hi immature thymocytes were The Reduced Number of Peripheral Agonist-Selected 34.6% in littermate controls versus 59.4% in Stim1fl/flStim2fl/fl T Cells Is Not Due to Impaired Cell Survival Vav-iCre mice (Figure 1E), suggesting that loss of store-operated To examine whether STIM proteins regulate the survival of Ca2+ entry causes a mild impairment of negative selection. Posi- agonist-selected T cells in the periphery, we carried out tive selection was essentially normal in female STIM-deficient in vivo 5-bromo-20-deoxyuridine (BrdU) labeling experiments in HY TCR-tg mice (Figure 1F), as judged by unaltered TCR reper- Stim1fl/flStim2fl/fl Mx1-Cre mice in which the Stim1 and Stim2 toires and normal numbers of TCRint CD69int thymocytes under- genes were acutely deleted by administration of polyinosinic- going selection (Starr et al., 2003)(Figures S1F–S1H). However, polycytidylic acid (poly IC). Mice were given BrdU in their the frequencies of CD8hi T cells in the thymus and lymph nodes drinking water for 10 days, injected with poly IC over the course of female STIM-deficient HY TCR-tg mice were slightly lower of the next 4 days, and analyzed 1 month later (Figure S2A). We compared to those in control mice. Overall, these results demon- confirmed that the administration of poly IC efficiently deleted strate that store-operated Ca2+ entry contributes to negative both the Stim1 gene and the Stim2 gene in T cells (Figure S2B). selection but is not absolutely essential for positive selection Although both the splenic Foxp3+ Treg population (Figure 2D; during the development of conventional TCRab+ T cells. Figure S2C) and the fraction of iNKT cells in liver mononuclear cells (Figure 2E; Figure S2D) showed a moderate decrease in Absence of Store-Operated Ca2+ Entry Results in Stim1fl/flStim2fl/fl Mx1-Cre mice a month after poly IC treatment, Impaired Development of Agonist-Selected T Cells perhaps due to decreased homing or new generation of these The observed partial impairment of negative selection, with no cells, the frequency of BrdU-labeled cells was not altered. In effect on positive selection, suggested that store-operated TCRab+ CD8aa+ IELs from Stim1fl/flStim2fl/fl Mx1-Cre mice, the Ca2+ entry might preferentially regulate the differentiation of percentages of BrdU-retaining cells were also increased self-reactive thymocytes. We therefore asked whether Stim compared to those from littermate controls (Figure 2F; Fig- deficiency influenced the development of agonist-selected ure S2E). Thus, these results suggest that store-operated Ca2+ T cells. In contrast to conventional TCRab+ T cells, the frequen- entry regulates agonist-selected T cell development in the cies of Foxp3+ Treg cells, iNKT cells, and TCRab+ CD8aa+ IELs thymus rather than the expansion and survival of these were significantly reduced in Stim1fl/flStim2fl/fl Vav-iCre and agonist-selected T cells in the periphery. Stim1fl/flStim2fl/fl Lck-Cre mice (Figures 2A and 2B). Notably, the three different lineages of self-reactive T cells were differen- Impaired Post-Selection Maturation of Agonist- tially affected in Stim1fl/flStim2+/+ Vav-iCre and Stim1fl/flStim2fl/fl Selected T Cells in Mice Lacking STIM1 and STIM2 Vav-iCre mice (Figure 2A). Loss of STIM1 alone caused We next examined the effect of store-operated Ca2+ entry on the a substantial reduction in the numbers of iNKT cells, to <5% of thymic development of these agonist-selected T cells. iNKT cells control, whereas loss of both STIM1 and STIM2 was required develop from DP thymocytes and are selected by lipid antigens

Figure 2. STIM-Deficient Mice Show a Significant Decrease of Regulatory T Cells, iNKT Cells, and TCRab+ CD8aa+ IELs in the Periphery (A) Flow cytometric analysis of Foxp3+ regulatory T cells (top), CD1d-tetramer+ iNKT cells (middle), and TCRab+ CD8aa+ IELs (bottom) in Stim1+/+Stim2+/+ Vav-iCre (left), Stim1fl/flStim2+/+ Vav-iCre (center), and Stim1fl/flStim2fl/fl Vav-iCre (right) mice. Right panels show average frequencies of each population. Control is shown by the dark gray bar); 1KO is shown by the open bar; DKO is shown by the light gray bar. n = 4 or 5. *p < 0.01. (B) Flow cytometric analysis of spleen, liver mononuclear cells, and intraepithelial cells as in Stim1+/+Stim2+/+ Lck-Cre (left) and Stim1fl/flStim2fl/fl Lck-Cre (right) mice. (C) Flow cytometric analysis of Foxp3+ regulatory T cells (top), CD1d-tetramer+ iNKT cells (middle), and TCRab+ CD8aa+ IELs (bottom) in competitive mixed bone-marrow chimeras. (D–F) Frequencies of Foxp3+ regulatory T cells in splenic CD4+TCRb+ cells (D), CD1d-tetramer+ iNKT cells in CD19 liver mononuclear cells (E) and CD8aa+ IELs in intestinal TCRab+ T cells (F) (left) from the BrdU-treated, poly IC-injected mice and the frequencies of BrdU-retaining cells in Foxp3+ cells, iNKT cells, and TCRab+ CD8aa+ IELs (right). n = 4–6. *p < 0.005; **p < 0.05; ***p = 0.067; NS, not significant. Data are representative of two (C–H) and more than three (A and B) independent experiments. Error bars (A, D, E, and F) denote mean ± SEM. See also Figure S2.

Immunity 38, 881–895, May 23, 2013 ª2013 Elsevier Inc. 885 Immunity T Cell Development and Calcium Signaling

presented by MHC class I-like molecule CD1d. After the expression of CD122 and PLZF in iNKT cell precursors agonist selection, iNKT cell precursors express promyelocytic (Seiler et al., 2012). These data suggest that the NFAT-Egr2 leukemia zinc finger (PLZF) encoded by Zbtb16 (Kovalovsky signaling pathway regulates TCRab+ CD8aa+ IEL development, et al., 2008; Savage et al., 2008), a master regulator of iNKT as well as iNKT cell development. cell development, and then expand and mature in response to Finally, we interrogated the early stages of Foxp3+ Treg cell IL-15 signaling in association with upregulation of CD122, development in detail. We found that the frequencies and cell the common b chain of the IL-2 and IL-15 receptors (Godfrey numbers of the mature Foxp3+CD25+ Treg cells (population 3) and Berzins, 2007). Because the population of iNKT cells in were significantly decreased in Stim1fl/flStim2fl/fl Vav-iCre the thymus is very small, we investigated their development in neonates on postnatal day 2, whereas those of the immature whole thymocytes as well as in a pre-enriched population Foxp3-CD25+ cells (population 1) and Foxp3+CD25- cells (popu- defined by staining with CD1d tetramer loaded with a-galacto- lation 2) were only modestly decreased (Figure 5A). Interestingly, sylceramide (aGalCer). In Stim1fl/flStim2fl/fl Vav-iCre mice, the amount of Foxp3 expression in Stim1fl/flStim2fl/fl Vav-iCre we detected slightly reduced frequencies and numbers of mice is lower than that of control mice (Figure 5B), suggesting CD1d tetramer loaded with aGalCer-binding CD24+ iNKT cell that maximum expression of Foxp3 requires store-operated precursors (Figures 3A and 3B, Stage 0), the earliest subset of Ca2+ entry. Foxp3-CD25hi cells are thought to be a major popu- iNKT cell precursors after agonist selection (Godfrey and Ber- lation containing Treg cell precursors (Lio and Hsieh, 2008). zins, 2007). By contrast, more differentiated aGalCer-binding Indeed, STIM-deficient Foxp3-CD25+ cells differentiated into CD24 iNKT cell precursors (Stages 1, 2, 3) were all abolished Foxp3+CD25+ Treg cells in vitro (Figure 5C), suggesting that in the absence of store-operated Ca2+ entry (Figures 3A Treg cell precursors can be generated in the absence of and 3B), suggesting that differentiation of iNKT cell precursors store-operated Ca2+ entry. Thus, these findings indicate that after agonist selection is severely impaired by the loss of agonist selection of Treg cells is intact in the absence of store- store-operated Ca2+ entry. Indeed, the expression of CD122 operated Ca2+ entry. The decreased frequency of Treg cells in expression was substantially impaired in residual STIM-deficient Stim1fl/flStim2fl/fl Vav-iCre mice reflected decreased proliferation iNKT cell precursors (Figure 3C). Furthermore, expression of rather than impaired expression of the prosurvival molecule, PLZF was also markedly reduced by the loss of store-operated Bcl-2 (Figures 5D and S3). CD25 upregulation was substantially Ca2+ entry (Figure 3D). These results indicate that the post- decreased in STIM-deficient Foxp3+CD25+ Treg cells, whereas selection expansion and functional maturation of iNKT cells CD122 expression was reduced mildly (Figure 5E). also requires store-operated Ca2+ entry and potentially IL-15 Collectively, these data demonstrate that store-operated Ca2+ signals as well. entry via the STIM-ORAI pathway is critical for the differentiation It is generally thought that TCRab+ CD8aa+ IELs arise in the of all three agonist-selected T cells after agonist selection medi- thymus (Leishman et al., 2002); however, the molecular mecha- ated by the TCR. Notably, developing agonist-selected T cells all nism underlying their development is poorly understood. showed impaired upregulation of cytokine receptor subunits, Although the thymic precursors of TCRab+ CD8aa+ IELs after suggesting that the impaired maturation of these T cell subsets agonist selection were defined as DN TCRb+ CD5+ cells (Gang- might be due to weakened IL-2 or IL-15 signaling. adharan et al., 2006), this population contains NKT cells and mucosal-associated invariant T (MAIT) cells expressing invariant Cytokine Administration Facilitates Differentiation of Va19 TCR coupled preferentially with TCRVb6 and TCRVb8 Post-Selection Precursors into Mature Cells, but Does (Treiner et al., 2003). Therefore, we refer here to DN TCRb+ Not Restore Their Effector Functions CD5+ cells as ‘‘rough pre-IELs,’’ and define ‘‘pre-IELs’’ as DN Both IL-2 and IL-15 signaling facilitate the differentiation of CD90.2+ CD5+ CD69+ CD122+ TCRb+ NK1.1 TCRgd agonist-selected T cell precursors into mature cells (Gangad- TCRVb6 and TCRVb8 cells. The number of pre-IELs was haran et al., 2006; Lio and Hsieh, 2008). We asked whether we substantially decreased but not abolished in Stim1fl/flStim2fl/fl could restore functionally mature Treg cells in STIM-deficient Vav-iCre mice (Figure 4A), suggesting that agonist selection of mice by injection with immune complexes of either IL-2 and IELs occurs to some degree in the absence of store-operated anti-IL-2 (Boyman et al., 2006) or IL-15 and IL-15Ra-Fc (Elpek Ca2+ entry. Proliferation of DN TCRb+ CD5+ ‘‘rough pre-IEL’’ cells et al., 2010) to provide sustained IL-2 and IL-15 signaling, in Stim1fl/flStim2fl/fl Vav-iCre mice, which contain true pre-IELs, respectively. In Stim1fl/flStim2fl/fl Vav-iCre mice treated with IL- was substantially diminished relative to littermate controls due 2-anti-IL-2 immune complexes, the number of CD25+Foxp3+ to impaired upregulation of CD122 (Figures 4B and 4C), another Treg cells was restored to that observed in untreated littermate marker of pre-IELs (Lambolez et al., 2007), suggesting that store- control mice (4%–5% of CD4+ T cells) (Figure 6A; Figure S4A). operated Ca2+ entry might influence IEL development by modu- Moreover, proliferation of Foxp3+ Treg cells was also remarkably lating cytokine receptor signaling. We further analyzed the gene enhanced (Figure 6B; Figure S4B). The Foxp3+ Treg cells expression pattern of DN TCRb+ CD5+ ‘‘rough pre-IEL’’ cells. In recovered from Stim1fl/flStim2fl/fl Vav-iCre mice treated with addition to CD122, we found that expression of the early growth IL-2-anti-IL-2 immune complexes were Helios-positive, indi- factor 2 (Egr2) and Egr3 genes, well-known NFAT target genes, cating that they are derived from thymic Treg cells rather was reduced at the mRNA level by the loss of store-operated than peripheral naive CD4+ T cells (Thornton et al., 2010)(Fig- Ca2+ entry (Table S1). Indeed, not only rough pre-IELs but also ure S4C). Foxp3 expression was slightly augmented by treat- iNKT cell precursors in Stim1fl/flStim2fl/fl Vav-iCre mice showed ment of Stim1fl/flStim2fl/fl Vav-iCre mice with IL-2-anti-IL-2 substantial reduction of Egr2 protein expression (Figure 4D). immune complexes (Figure 6C; Figure S4D) but these STIM- This result is consistent with a recent report that Egr2 regulates deficient Foxp3+ Treg cells nevertheless failed to inhibit the

886 Immunity 38, 881–895, May 23, 2013 ª2013 Elsevier Inc. D LFepeso ntyi NTclsa tg – ncnrl(lc ie n K ga ie ie ahdln ersnsioyecnrl p<00;N,not SEM. NS, ± 0.01; mean < denote *p B) control. and isotype represents (A line bars Dashed Error mice. experiments. line) independent (gray (A) DKO mice. four and lines) and line) (gray (B–D) (black DKO control two and in of lines) 1–3 representative (black stage are control at Data cells from significant. iNKT precursors thymic cell in iNKT expression in PLZF components (D) receptor IL-15 of Expression (C) B lwctmti nlss(et n rqece rgt fpeerce hmciK rcrosby precursors iNKT thymic pre-enriched of (right) frequencies and 5. = (left) (Stim1 n analysis thymus. cytometric the in Flow precursors (B) cell iNKT of (right) number B A Signaling Calcium and Development Cell T Immunity A lwctmti nlssof analysis Ca cytometric Flow Store-Operated (A) of Loss 3. Figure C Stim1

10 CD1d-tet10 10 10 10 10 10 10 10 10 Gated onCD19 +/+ 0 1 2 3 4 0 1 2 3 4 10 10 fl/fl

Relative cell #100 20 40 60 80 0 0.0045 0 Stim2 0 Stim2 10 0.15 0 10 10 CD24

10 CD1d-tet10 10 10 10 10 10 10 10 10 1 1 CD122 +/+ 0 1 2 3 4 0 1 2 3 4 fl/fl 10 10 10 1 0 0 or Vav 0.019 0.37 26.1 10 10 0.0093 2 2 Stim1 10 10 CD24 10 ir)adDO(Stim1 DKO and -iCre) 2 1 1 10 10 3 3 +/+ 10 10 10 3 2 2 0.16 0.14 Stim2 10 10 a - 4 4 glcoycrmd-oddC1 ermrpstv NTcl rcrosi h hmsioae rmcnrl(Stim1 control from isolated thymus the in precursors cell iNKT tetramer-positive CD1d -galactosylceramide-loaded DKO Control 10 10 10 4 3 3 100 +/+ 20 40 60 80 0 10 0 Vav 10 10 4 4 DKO Control IL-15R ir)adDO( DKO and -iCre) 10 2+ 1 fl/fl nr eut nIpie NTCl eeomn u oRdcdEpeso fC12adPLZF and CD122 of Expression Reduced to Due Development Cell iNKT Impaired in Results Entry Stim2 10 2 iNKT precursors (%) 0.05 0.15 0.25 fl/fl 10 10 10 10 10 0.1 0.2 0.3 10 α 0 1 2 3 4 3 10 Vav 0 0 Control Stim1 CD1d-tet ir)mc lf) rqec fiK elpeusr nCD19 in precursors cell iNKT of frequency (left), mice -iCre) 10 tg0Stage1-3 Stage0 0 CD24 4 10 100 20 40 60 80 1 0 10 fl/fl 0 DKO Stim2 10 + 2 NS 10 CD132 + 1 0 fl/fl 10 3 10 Vav

2 Control CD1d-tet 10 ir)mice. -iCre) 4 J CD24 10 3 α DKO 18 KO * 10 Immunity 4 - + Tcraj-18 D

4 38

Relative cell #100 Absolute cell number (x10 ) / 20 40 60 80 10 20 30 40 50 60 0 10 8–9,My2,2013 23, May 881–895, , 0 0 a ie(J mice CD1d-tet glcoycrmd-oddC1 ermri control in tetramer CD1d -galactosylceramide-loaded

tg0Stage1-3 Stage0 Control Pre-enriched iNKT precursors (%) CD24 10 0.05 0.15 0.25 1 PLZF 0.1 0.2 a 0 8K)wr sda eaiecnrl 5. = n control. negative a as used were KO) 18 10 + DKO 2 Control NS + 10 Stage0 3 hmcts(etr n bouecell absolute and (center) thymocytes

DKO CD1d-tet Control 10 NS 4 CD24 ª Control DKO 03Esve Inc. Elsevier 2013 DKO * - + 20 30 10 0 Control Stage1-3 fl/fl

Stim2 DKO * fl/fl 887 or Immunity T Cell Development and Calcium Signaling

CD4,CD8, A NK1.1,TCRγδ (-) IEL precursors

4 4 4 4 10 10 10 10

3 3 3 3 10 10 10 10 56.3 2.53 37.9 10 2 10 2 10 2 10 2 Control

10 1 10 1 10 1 10 1

10 0 2.5 10 0 10 0 10 0 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 4 4 4 4 10 10 10 10 8 β

3 3 3 3 10 10 10 10

6,V 63.8 0.91 44.4 10 2 10 2 10 2 10 2 β DKO

10 1 10 1 10 1 10 1

10 0 2.16 10 0 10 0 10 0 CD5 TCRV CD122 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 CD90.2 TCRβ CD69 )

4 B ) 2 30

-2 100 100 100

80 80 80

1.5 60 60 60 20 40 40 40

1 20 20 20 Relative cell # ** 0 0 0 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 * 10 α 0.5 CD122 IL-15R CD132 CD

IEL precursors (%)(x10 Pre-IEL iNKT(Stage1-3) Absolute cell number (x10 0 0 100 100 100

80 80 80

DKO DKO 60 60 60 Control Control 40 40 40

20 20 20

Relative cell # 0 0 0 Relative cell # 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 Ki67 Egr2

Figure 4. STIM-Deficient Mice Have a Smaller Population of Thymic TCRab+ CD8aa+ IEL Precursors (A) Flow cytometric analysis of TCRab+ CD8aa+ IEL precursors (pre-IELs) among CD4 CD8 thymocytes from control mice and VavDKO mice and frequency in total thymocytes (left) and absolute cell number (right) of pre-IELs. Data are representative of three independent experiments. Error bars denote mean ± SEM n = 5. *p < 0.01, **p < 0.05. (B) Expression of IL-15 receptor components in DN CD5+TCRb+ thymocytes from control (black lines) and DKO (gray lines) mice. Data are representative of three independent experiments. (C) Proliferation of DN CD5+TCRb+ thymocytes from control (black lines) and DKO (gray lines) mice was assessed by Ki-67 staining. Data are representative of three independent experiments. (D) Egr2 expression in DN CD5+TCRb+ thymocytes and iNKT cell precursors in control (black line) and DKO (gray line) mice. Dashed line represents isotype control. Data are representative of two (D) and more than three (A to C) independent experiments. See also Table S1. proliferation of responder T cells in vitro (Figure 6D). The is crucial for IL-2-driven differentiation or proliferation and impaired suppressive activity cells was not due to down- the functional maturation of Treg cells by modulating NFAT regulation of Foxp3 expression or decreased survival (Figures activity. 6E and 6F), but rather to lower expression of inhibitory molecules In contrast to IL-2-anti-IL-2 immune complexes, IL-15- such as CTLA-4, LAG-3 and TIGIT (Figure 6G). Consistent immune complexes did not efficiently rescue Treg cell numbers with our previous data (Wu et al., 2006), these results collectively (Figure 6H). Treatment with IL-15-immune complexes did, indicate that STIM-ORAI-mediated store-operated Ca2+ entry however, induce the expansion of thymic DN TCRb+ CD5+

888 Immunity 38, 881–895, May 23, 2013 ª2013 Elsevier Inc. Immunity T Cell Development and Calcium Signaling

+ A Gated on CD4

4 4 4 10 10 10 3 2.63 1.07 0 0 3 3 3 10 10 10 2.5 P1 P3 * 2 2 2 2 10 10 10 Control

10 1 1 1 CD25 10 10 1.5 P2 0 0 96 0.34 0 100 0 10 10 10 1 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 4 4 10 10 * Foxp3 2.22 0.055 0 0 0.5 3 3 10 10 * 0 10 2 10 2 DKO Cells in CD4SP thymocytes (%) DKO DKO DKO 10 1 10 1 Control Control Control P1 P2 P3 10 0 97.6 0.15 10 0 100 0 ) CD25 Isotype control 4 10 0 10 1 102 103 104 10 0 10 1 102 103 104 5 Foxp3 Isotype control 4 NS B C 3 4 Control4 DKO 100 10 10 34.6 19.9 80 3 3 2 10 10 60 NS 10 2 10 2 1 40

1 1 20 CD25 10 10 **

Absolute cell number (x10 0

0 Relative cell # 10 0 10 0 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 DKO DKO DKO Foxp3 Foxp3 Control Control Control Gated on CD4+ P1 P2 P3

D E

100 100 100 100 100

80 P1 80 P3 80 80 80

60 60 60 60 60

40 40 40 40 40

20 20 20 20 20

0 0 Relative cell # 0 0 0 Relative cell # 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 KC76i D25221DC CD132

Figure 5. Absence of Store-Operated Ca2+ Entry Blocks Treg Cell Development (A) Staining of Foxp3 versus CD25 on CD4-single positive (CD4SP) thymocytes in neonates at day 2 after birth (left) and frequency (upper right) and absolute cell number (lower right) of each population. Error bars denote mean ± SEM n = 6. *p < 0.002, **p < 0.01; NS, not significant. (B) Foxp3 expression in Foxp3+CD25+ cells (population 3) from control (black lines) and DKO (gray lines) mice. (C) In vitro differentiation of Treg cell precursors into Foxp3+CD25hi Treg cells. (D) Proliferation of Foxp3+CD25+ cells from control (black lines) and DKO (gray lines) mice was assessed by Ki-67 staining. (E) Expression of IL-2 receptor components in Foxp3+CD25+ from control (black lines) and DKO (gray lines) mice. Data are representative of two (A, C, and E) and more than three (B and D) independent experiments. See also Figure S3.

‘‘rough pre-IEL’’ cells in vivo; however, the numbers of mature absence of STIM may be due to their inability to properly localize TCRab+ CD8aa+ IELs, the numbers of thymic iNKT precursor to IL-15-rich tissues such as cryptopatches. cells and peripheral mature iNKT cells, all remained suppressed (Figures S4E and S4F). Notably, in vitro culture of DN TCRb+ NFAT Activity Is Required for Efficient Development of CD5+ ‘‘rough pre-IEL’’ cells in IL-15 resulted in partial recovery Agonist-Selected T Cells of TCRab+ CD8aa+ IELs (Figure S4G), demonstrating that The above results strongly suggest that the maturation of STIM-deficient pre-IELs possess the ability to differentiate into agonist-selected T cells requires sustained large Ca2+ influx. TCRab+ CD8aa+ IELs in the presence of IL-15. These data raise Because the natural ligands of Treg cells, IELs, and conventional the possibility that the impaired development of IELs in the thymocytes undergoing positive selection remain unknown,

Immunity 38, 881–895, May 23, 2013 ª2013 Elsevier Inc. 889 Immunity T Cell Development and Calcium Signaling

A Thymus B Control DKO Control DKO 20 100 100 4 4 10 10 * 80 80

3 3 10 10 60 60

40 40 2 4.14 2 0.24 10 10 IL-2 15 + Isotype 20 20 10 1 10 1

0.75 Relative cell # 0 0 2.8 10 0 10 1 102 103 104 10 0 10 1 102 103 104 0 0 10 10 Cells (%)

10 0 10 1 102 103 104 10 0 10 1 102 103 104 hi 10 Ki67 4 4 10 10 C IL-2+Isotype IL-2+anti-IL-2 3 3 10 10 CD25 100 100 +

IL-2 80 80 2 10.6 2 4.75 10 10 + anti-IL-2 5 60 60 Foxp3 10 1 10 1 3.31 1.2 40 40

0 0 CD25 10 10 20 20 10 0 10 1 102 103 104 10 0 10 1 102 103 104 0

Foxp3 Relative cell # 0 0 IL-2 IL-2 10 0 10 1 102 103 104 10 0 10 1 102 103 104 Gated on CD4+ + Isotype + anti-IL-2 Foxp3

D Control DKO EF

300 100

3 150 ) 7 4 200 4 2 100 80 6 5 100 50 60 5 Cell number 1 0 0 0 10 0 10 1 102 103 104 10 0 10 1 102 103 104 40 4 CFSE

Relative cell # 3 20 60 2 0 50 0 1 2 3 4 10 10 10 10 10 1 40 Foxp3 Absolute cell number (x10 0 30 DKO DKO 20 Control Control 10 Treg Tresp 0

Ratio in responder T cells (%) 01234 5 Numbers of division

G H Control DKO 4 4 10 10 100 100

80 80 3 3 10 10

60 60 10 2 3.44 10 2 0.41 40 40 PBS

10 1 10 1 20 20 3.06 1.02 Relative cell # 0 0 10 0 10 0 0 1 2 3 4 0 1 2 3 4 10 10 10 10 10 10 10 10 10 10 10 0 10 1 102 103 104 10 0 10 1 102 103 104 4 4 CTLA-4 LAG-3 10 10

3 3 100 100 10 10 1.53 80 80 10 2 5.04 10 2 IL-15 + IL-15Rα-Fc 60 60 10 1 10 1 1.99 40 40 7.47

0 0 CD25 10 10 20 20 10 0 10 1 102 103 104 10 0 10 1 102 103 104 Relative cell # 0 0 0 1 2 3 4 0 1 2 3 4 Foxp3 10 10 10 10 10 10 10 10 10 10 + TIGIT GITR Gated on CD4 Control DKO

Figure 6. Store-Operated Ca2+ Entry Controls Proliferation and Functional Maturation of Treg Cell Precursors (A) Flow cytometric analysis of thymic CD4+ Foxp3+CD25+ Treg cells in control (Stim1fl/flStim2fl/fl; dark gray bar) and DKO (Stim1fl/flStim2fl/fl Vav-iCre; light gray bar) mice treated with IL-2-immune complexes (left) and frequency of Foxp3+CD25+ Treg cells in CD4SP thymocytes (right). Error bars denote mean ± SEM n = 3. *p < 0.05. (legend continued on next page) 890 Immunity 38, 881–895, May 23, 2013 ª2013 Elsevier Inc. Immunity T Cell Development and Calcium Signaling

we stimulated whole thymocytes with anti-CD3 antibody Previous studies have suggested that sustained TCR signals and measured Ca2+ influx by flow cytometry. Wild-type imma- allow agonist-selected T cell precursors to mature (Baldwin ture Foxp3CD25+ cells, as well as CD4loCD8lo DP cells con- et al., 2004); however, little is known about the molecular path- taining iNKT precursors and pre-IELs, showed larger and more ways responsible for their development. Self-reactive thymo- sustained Ca2+ influx in response to the stimulation with anti- cytes show high and rapid Ca2+ influx through the interaction CD3 antibody, whereas STIM-deficient immature Foxp3CD25+ of self-peptides with TCRs. Given that STIM-ORAI signaling cells showed little Ca2+ influx (Figure 7A; Figure S5). Lastly, induces rapid large Ca2+ influxes but also maintains high 2+ 2+ we examined the effect of the Ca -responsive transcription concentrations of [Ca ]i, which supports long-term conse- factor NFAT on agonist-selected T cell development. NFAT quences of Ca2+ signaling such as gene expression, store-oper- consists of a family of four transcription factors (NFAT1-4, also- ated Ca2+ entry seems to fulfill both modes of Ca2+ influx during known as NFATc1-c4), each of which is preferentially expressed thymic development of agonist-selected T cells. Indeed, our in different subsets of T cells (Hogan et al., 2003). Because results have shown that lack of store-operated Ca2+ entry specific deletion of NFAT4 in T cells causes a partial block in influences postselection maturation of agonist-selected T cells positive selection of conventional TCRab+ T cells (Cante´ -Barrett rather than agonist selection itself. We speculate that differenti- et al., 2007), we examined mice in which NFAT1 and NFAT2 ation of these cells after agonist selection in response to were deleted in T cells to avoid the effect of NFAT4 deficiency prolonged Ca2+ influx via store-operated Ca2+ entry involves on T cell development. In Nfat1/Nfat2fl/fl Lck-Cre mice, the efficient expression of NFAT target genes, e.g., Cd25 and development of Treg cells and iNKT cells, but not conventional Egr2, or sustained cooperation of NFAT with other transcription TCRab+ T cells, was partially blocked in the thymus, whereas factors (Chen et al., 1998; Wu et al., 2006). Consistent with the postthymic development of all agonist-selected T cells was this idea, hyperactive TCR signaling enables higher Egr2 ex- substantially impaired (Figures 7B and 7C). Given the predomi- pression, which allows thymocytes to differentiate into iNKT nant expression of NFAT4 in DP thymocytes, NFAT4 may also cells (Seiler et al., 2012). contribute to the development of agonist-selected T cells. Even a complete lack of store-operated Ca2+ entry caused Together with the data from STIM-deficient mice, these data only a mild impairment of negative selection in our STIM- indicate that STIM-mediated store-operated Ca2+ entry regu- deficient mice. A more severe impairment is observed in lates the postselection maturation of agonist-selected T cells LAT(Y136F) mutant mice and T cell-specific Plcg1/ mice in through activation of all expressed in T cells. which Ca2+ influx downstream of TCR activation is abolished (Fu et al., 2010; Sommers et al., 2005). The expression of Bim, DISCUSSION a proapoptotic protein that plays a critical role in negative selection (Bouillet et al., 2002), can be induced by Ca2+-respon- We demonstrate here that agonist-selected T cells require store- sive protein kinase C as well as by treatment with phorbol myris- operated Ca2+ entry for their development. Loss of store-oper- tate acetate (PMA), indicating that pathways other than Ca2+ ated Ca2+ entry results in severe impairment of postselection signaling are involved in negative selection (Cante´ -Barrett proliferation and functional maturation of agonist-selected et al., 2006). Thus, the stronger impairment of negative selection T cells, by modulating the expression of cytokine receptors observed in LAT(Y136F) mutant and Plcg1/ mice compared and effector molecules. In contrast, development of conven- to STIM-deficient mice reflects the fact that both the LAT and tional TCRab+ T cells either does not require store-operated PLC-g1 mutations interfere with additional, non-Ca2+-depen- Ca2+ entry or can be compensated for by non-store-operated dent (i.e., DAG-dependent) signaling pathways downstream of pathways of Ca2+ influx. These data suggest that sustained PLC-g1 activation. store-operated Ca2+ entry, full NFAT activation, and efficient In contrast to negative selection, several studies have reported expression of NFAT-driven genes are important factors in that the store-operated Ca2+ entry-calcineurin-NFAT pathway is agonist-selected T cell development. Importantly, our data in essential for positive selection of conventional TCRab+ T cells. the STIM-deficient mouse model are consistent with phenotypes Calcineurin-deficient mice show a complete block of positive observed in human patients with nonfunctional mutations or selection in vivo (Neilson et al., 2004), and genetic disruption of point mutations in the STIM1 gene, who show specific reduction NFAT4 activity partially impairs positive selection due to of peripheral agonist-selected T cells including Foxp3+ Treg cells increased cell death (Cante´ -Barrett et al., 2007). Another study and iNKT cells (Fuchs et al., 2012; Picard et al., 2009). has shown that positively-selecting thymocytes display

(B) Enhanced proliferation of Foxp3+CD25+ Treg cells following injection of IL-2-anti-IL-2 immune complexes (black lines) but not immune complexes of IL-2 and isotype control monoclonal antibody (gray lines), in the same experiment as that shown in (A). (C) Amounts of Foxp3 expression in control (black lines) and DKO (gray lines) thymocytes following injection of IL-2-anti-IL-2 immune complexes. (D) Suppressive function of the recovered Treg cells from control (Stim1fl/flStim2fl/flFoxp3hCD2 or Stim1+/+Stim2+/+ Foxp3hCD2; closed circle) and DKO (Stim1fl/flStim2fl/fl CD4-Cre Foxp3hCD2; open circle) mice. Carboxyfluorescein succinimidyl ester (CFSE)-labeled responder T cells were cocultured with Treg cells treated with IL-2-anti-IL-2 immune complexes. Error bars denote mean ± SEM n = 3. (E) Amounts of Foxp3 expression on the recovered control (black lines) and DKO (gray lines) Treg cells. (F) Cell numbers of each population after 3 days of coculture. Control is shown by the open circle; DKO is shown by the closed circle. (G) Flow cytometric analysis of the expression of inhibitory molecules on thymic Foxp3+CD25+ Treg cells in control (black lines) and DKO (gray lines), in the same experiment as that shown in (D). (H) Flow cytometric analysis of thymic Foxp3+CD25+ Treg cells in control (Stim1fl/flStim2fl/fl) and DKO (Stim1fl/flStim2fl/fl Vav-iCre) mice treated with IL-15-immune complexes. Data are representative of two (B to H) and more than three (A) independent experiments. See also Figure S4.

Immunity 38, 881–895, May 23, 2013 ª2013 Elsevier Inc. 891 Immunity T Cell Development and Calcium Signaling

A 5 CD3 3 * CD4+CD25hiFoxp3- ** 6 * 4 CD4loCD8lo * 2 3 4

2 1 DP 2 Peak of ratio Mean of ratio

1 (Fluo-4/Fura Red) (Fluo-4/Fura Red)

Ratio (Fluo-4/Fura Red) 0 0 lo hi lo hi 0 DP DP lo CD8 lo CD8 0 300 600 900 1200 + CD25- + CD25- CD4 CD4 CD4 CD4 Time (s) Foxp3 Foxp3

B Treg cells iNKT cells TCRαβ+ CD8αα+ IELs

Nfat1-/-Nfat2fl/fl Nfat1-/-Nfat2fl/fl Nfat1-/-Nfat2fl/fl Lck Lck Lck Lck Lck Lck 4 -Cre 4 -Cre 4 -Cre 4 -Cre 4 -Cre 4 -Cre 10 10 10 10 10 10 2.52 1.33 6.22 93.5 33.5 60.8 3 3 3 3 3 3 10 10 10 10 10 10 44.1 36.6 10 2 10 2 10 2 10 2 10 2 10 2 Thymus

1 1 1 1 1 1

CD25 10 10 CD44 10 10 10 10 CD122

10 0 10 0 10 0 0.22 10 0 5.69 10 0 10 0 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 Foxp3 NK1.1 CD69 Gated on CD4+ Gated on CD19-CD1d-tet+ Gated on DN TCRβ+CD5+ iNKT(-),MAIT(-)

4 4 4 4 4 4 10 10 10 10 10 10 10.5 5.36 3 3 3 3 3 3 10 10 10 12.4 10 4.33 10 10 β 10 2 10 2 10 2 10 2 10 2 10 2 Periphery 43.7 21.2 CD8 CD25 10 1 10 1 10 1 10 1 10 1 10 1 CD1d-tet

10 0 10 0 10 0 10 0 10 0 10 0 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 10 0 10 1 102 103 104 Foxp3 TCRβ CD8α Gated on CD4+TCRβ+ Gated on CD19- Gated on TCRβ+

Thymus Periphery Thymus Periphery Thymus Periphery

4 12 100 20 0.1 50 80 0.08 40 3 15 NS 8 * * 60 0.06 30 Treg cells (%) IELs (%) Treg cells (%) + hi 2 10 * * hi 40 0.04 20 4 * αα 1 5

CD25 20 0.02 10 iNKT cells (%) + CD25 CD8 + IEL precursors (%) in iNKT cells (%) 0 0 Stage 3 precursors 0 0 0 0 Foxp3 DKO Foxp3 DKO DKO DKO DKO DKO Control Control Control Control Control Control

C Nfat1-/-Nfat2fl/fl Lck-Cre Lck-Cre 4 4 10 10 7.41 83.9 6.7 84.9 3 3 10 10

10 2 10 2 CD4 10 1 10 1

10 0 5.95 2.76 10 0 5.01 3.35 10 0 10 1 102 103 104 10 0 10 1 102 103 104 CD8

(legend on next page) 892 Immunity 38, 881–895, May 23, 2013 ª2013 Elsevier Inc. Immunity T Cell Development and Calcium Signaling

characteristic sustained Ca2+ oscillations, which allows thymo- control. In this scenario, store-operated Ca2+ entry might be cytes to form stable interactions with stromal cells (Bhakta essential to reach a higher threshold required for agonist- et al., 2005). In contrast, although STIM-regulated store-oper- selected T cell development. ated Ca2+ entry is a major Ca2+ influx pathway in T cells, our In addition to TCR signaling, cytokine signaling is another key results demonstrate that the loss of store-operated Ca2+ entry process that regulates development of agonist-selected T cells. does not affect positive selection. IL-2 and IL-15 are critical for terminal differentiation of all Why is positive selection normal in the absence of store-oper- agonist-selected T cells (Godfrey and Berzins, 2007; Lambolez ated Ca2+ entry? Potentially, DP thymocytes could use one or et al., 2007; Lio and Hsieh, 2008). Our data show that IL-2 or more non-store-operated Ca2+ channels during positive selec- IL-15 signaling alone restores the expansion of Treg cell precur- tion to compensate for the absence of store-operated Ca2+ sors but not the suppressive function of mature Treg cells, sug- entry. Indeed, CRAC channel inhibitors did not inhibit the gesting that acquisition of mature effector function further Ca2+ oscillations elicited by TCR stimulation in thymocytes requires both TCR signaling and IL-2 or IL-15 signaling. Indeed, undergoing positive selection (Bhakta et al., 2005), suggesting both Treg cells (Oh-Hora et al., 2008) and iNKT cells (Zullo et al., that Ca2+ influx might be independent of the STIM-ORAI 2007) need NFAT activity downstream of the TCR for their pathway. DP thymocytes express many diverse Ca2+-perme- effector functions. able channels (data not shown). Among these, transient In addition to IL-2 and IL-15, transforming growth factor (TGF)- receptor potential cation channel M7 (TRPM7) is highly ex- b is also essential for agonist-selected T cell development. Mice pressed in thymocytes and mediates the permeation of both in which TGF-b signaling is disrupted show a significant Mg2+ and Ca2+. Mice with a targeted deletion of the Trpm7 decrease of agonist-selected T cell populations (Doisne et al., gene in T cells show a partial developmental block in the DN 2009; Konkel et al., 2011; Li et al., 2006; Liu et al., 2008; Marie to DP thymocyte transition without altering Mg2+ homeostasis et al., 2006). However, in contrast to STIM-deficient mice, these because of impaired differentiation of thymic medullary epithe- mice show expansion of Treg cells in the thymus of adult mice lial cells (Jin et al., 2008), suggesting that immature thymocytes and upregulation of CD122 in iNKT cell precursors and pre- utilize TRPM7 to take up Ca2+ from the extracellular space. The IELs. Moreover, lack of TGF-b signaling exacerbates negative 2+ + L-type voltage-dependent Ca channel Cav1.4 may also selection of conventional TCRab T cells (Ouyang et al., 2010). compensate for the loss of store-operated Ca2+ entry because Hence, it is unlikely that defects in TGF-b signaling underlie

Cav1.4-deficient mice show a subtle but clear reduction of the impaired development of agonist-selected T cells in STIM- mature SP thymocytes (Omilusik et al., 2011). Finally, other deficient mice. pathways may contribute to Ca2+ entry during positive selec- In contrast to Treg cells and iNKT cells, the mechanism of + + tion: for instance, a very small number of IP3 receptors in TCRab CD8aa IEL development is still enigmatic. Although the plasma membrane (estimated at 1 or 2 receptors) Ca2+ signaling seems to be central to the development of accounts for as much as 50% of total Ca2+ influx in B cells TCRab+ CD8aa+ IELs, our studies did not fully elucidate the (Dellis et al., 2006). role of NFAT in pre-IELs. In the absence of store-operated A ‘‘gauntlet’’ model proposes that thymocytes mature by Ca2+ entry, the expression of Egr2, an NFAT target gene, was receiving continuous weak stimulation from multiple contacts substantially downregulated in pre-IELs, and this was accompa- with peptide-expressing stromal cells, which allows them to nied by decreased expression of CD122 and PLZF. Because acquire positively selecting signals without making stable PLZF-deficient mice show normal frequency of TCRab+ contact with thymic stromal cells (Ebert et al., 2008). Consistent CD8aa+ IELs in the periphery (Kreslavsky et al., 2009), CD122 with this model, peripheral primary T cells serially encounter is a preferential target of NFAT-Egr2 pathway. However, antigen-presenting dendritic cells over hours and accumulate CD122 expression seems to be only part of the program that incremental activation signals until T cells reach a threshold regulates the terminal differentiation of TCRab+ CD8aa+ IELs that triggers stable interactions with dendritic cells (Mempel in vivo. Exogenous IL-15 administration rescued TCRab+ et al., 2004). Because positive selection occurs over a period CD8aa+ IELs differentiation of STIM-deficient pre-IELs in vitro of 3–5 days as thymocytes migrate from the outer cortex to the but not in vivo, suggesting that NFAT might regulate recruitment medulla (Huesmann et al., 1991; Shortman et al., 1991), multiple of pre-IELs to IL-15-rich tissues in the small intestine. Because contacts with thymic stromal cells, which generate small Ca2+ CCL25 is rich in small intestine, its receptor CCR9 seems to be signals induced by STIM-independent Ca2+ influx, might allow essential for the recruitment of pre-IELs to the gut. Although DP thymocytes to differentiate into CD4 or CD8 single-positive the expression of CCR9 is regulated by the cooperation of thymocytes after accumulating signals reach a threshold NFAT1 with retinoic acid receptor-retinoid X receptor complex required to positive selection. Indeed, we observed delayed in T cells (Ohoka et al., 2011), genetic disruption of CCR9 in maturation of conventional STIM-deficient T cells compared to mice results in an increased frequency of TCRab+ CD8aa+

Figure 7. Efficient Agonist-T Cell Development Requires Full Activation of NFATs (A) Ca2+ influx in response to anti-CD3 crosslinking (n = 3) in wild-type CD4+CD25hiFoxp3- (Treg precursors, solid line), CD4loCD8lo (population including iNKT and pre-IELs) and DP thymocytes (conventional T cells). Error bars denote mean ± SEM n = 3. *p < 0.05; **p = 0.07. (B and C) Flow cytometric analysis of agonist-selected T-lineage cells (B) and conventional T cell development (C) in NFAT1 and NFAT2-doubly deficient mice. Frequencies represent Foxp3+CD25hi Treg cells in CD4SP thymocytes or in splenic CD4+TCRb+ cells, iNKT cells in CD19 liver mononuclear cells, IEL precursors in total thymocytes, or CD8aa+ IELs in intestinal TCRab+ T cells. Data are representative of two (B and C) and three (A) independent experiments. Error bars denote mean ± SEM n = 4. *p < 0.05; NS, not significant. See also Figure S5.

Immunity 38, 881–895, May 23, 2013 ª2013 Elsevier Inc. 893 Immunity T Cell Development and Calcium Signaling

IELs (Uehara et al., 2002; Wurbel et al., 2001). Thus it is likely Brandman, O., Liou, J., Park, W.S., and Meyer, T. (2007). STIM2 is a feedback 2+ that NFAT regulates other chemokine receptors or adhesion regulator that stabilizes basal cytosolic and Ca molecules. Alternatively, NFAT might induce or cooperate with levels. Cell 131, 1327–1339. an uncharacterized master regulator of TCRab+ CD8aa+ IELs. Cante´ -Barrett, K., Gallo, E.M., Winslow, M.M., and Crabtree, G.R. (2006). 2+ Further studies are needed to elucidate the molecular identity Thymocyte negative selection is mediated by protein kinase C- and Ca - dependent transcriptional induction of bim. J. Immunol. 176, 2299–2306. and mechanism of these cells. In conclusion, our findings suggest a common TCR signaling pathway of agonist-selected Cante´ -Barrett, K., Winslow, M.M., and Crabtree, G.R. (2007). Selective role of T cells in which Foxp3+ Treg cells, iNKT cells, and TCRab+ NFATc3 in positive selection of thymocytes. J. Immunol. 179, 103–110. CD8aa+ IELs require strong and sustained Ca2+ influx by store- Chen, L., Glover, J.N., Hogan, P.G., Rao, A., and Harrison, S.C. (1998). operated Ca2+ entry for their postselection maturation. Structure of the DNA-binding domains from NFAT, Fos and Jun bound specif- ically to DNA. Nature 392, 42–48. Daniels, M.A., Teixeiro, E., Gill, J., Hausmann, B., Roubaty, D., Holmberg, K., EXPERIMENTAL PROCEDURES Werlen, G., Holla¨ nder, G.A., Gascoigne, N.R., and Palmer, E. (2006). Thymic selection threshold defined by compartmentalization of Ras/MAPK signalling. Mice Nature 444, 724–729. All mice were maintained in specific pathogen-free barrier facilities at Harvard Medical School or Tokyo Medical and Dental University and were used in Dellis, O., Dedos, S.G., Tovey, S.C., Taufiq-Ur-Rahman, Dubel, S.J., and 2+ accordance with protocols approved by the Center for Animal Resources Taylor, C.W. (2006). Ca entry through plasma membrane IP3 receptors. and Comparative Medicine of Harvard Medical School or the Institutional Science 313, 229–233. Animal Care and Use Committee of Tokyo Medical and Dental University Doisne, J.M., Bartholin, L., Yan, K.P., Garcia, C.N., Duarte, N., Le Luduec, J.B., and conformed to relevant guidelines and laws. Vincent, D., Cyprian, F., Horvat, B., Martel, S., et al. (2009). iNKT cell develop- ment is orchestrated by different branches of TGF-b signaling. J. Exp. Med. SUPPLEMENTAL INFORMATION 206, 1365–1378. Ebert, P.J., Ehrlich, L.I., and Davis, M.M. (2008). Low ligand requirement for Supplemental Information includes five figures, one table, and Supplemental deletion and lack of synapses in positive selection enforce the gauntlet of Experimental Procedures and can be found with this article online at http:// thymic T cell maturation. Immunity 29, 734–745. dx.doi.org/10.1016/j.immuni.2013.02.008. Elpek, K.G., Rubinstein, M.P., Bellemare-Pelletier, A., Goldrath, A.W., and ACKNOWLEDGMENTS Turley, S.J. (2010). Mature natural killer cells with phenotypic and functional alterations accumulate upon sustained stimulation with IL-15/IL-15Ra We thank T. Kitamura (The University of Tokyo) for providing Plat-E packaging complexes. Proc. Natl. Acad. Sci. USA 107, 21647–21652. cell lines, as well as K. Okamoto, M. Guerrini, A. Terashima, Y. Muratani, and T. Feske, S. (2007). Calcium signalling in lymphocyte activation and disease. Kato for discussion and assistance. This work was supported in part by Grant- Nat. Rev. Immunol. 7, 690–702. in-Aid for Scientific Research on Priority Areas from the Japan Society for the Feske, S., Gwack, Y., Prakriya, M., Srikanth, S., Puppel, S.H., Tanasa, B., Promotion of Science (JSPS), PRESTO from the Japan Science and Tech- Hogan, P.G., Lewis, R.S., Daly, M., and Rao, A. (2006). A mutation in Orai1 nology Agency (JST) (to M.O.), Grants-in-Aid for GCOE Program from the causes immune deficiency by abrogating CRAC channel function. Nature Ministry of Education, Culture, Sports, Science and Technology of Japan (to 441, 179–185. M.O. and H.T.), ERATO, Takayanagi Osteonetwork Project from the JST (to H.T.) and NIH grants to S.F. (AI 066128) and A.R. (AI40127 and AI84167). Fu, G., Chen, Y., Yu, M., Podd, A., Schuman, J., He, Y., Di, L., Yassai, M., M.O. was also supported by grants from Takeda Life Science Foundation Haribhai, D., North, P.E., et al. (2010). Phospholipase Cg1 is essential for and the Mochida Memorial Foundation for Medical and Pharmaceutical T cell development, activation, and tolerance. J. Exp. Med. 207, 309–318. Research. N.K. is supported by JSPS Research Fellowships for Young Scien- Fuchs, S., Rensing-Ehl, A., Speckmann, C., Bengsch, B., Schmitt-Graeff, A., tists. S.F. and A.R. are scientific founders of CalciMedica, a company that Bondzio, I., Maul-Pavicic, A., Bass, T., Vraetz, T., Strahm, B., et al. (2012). seeks to identify novel treatments for immune-related diseases. Antiviral and regulatory T cell immunity in a patient with stromal interaction molecule 1 deficiency. J. Immunol. 188, 1523–1533. Received: June 21, 2012 Gangadharan, D., Lambolez, F., Attinger, A., Wang-Zhu, Y., Sullivan, B.A., and Accepted: February 11, 2013 Cheroutre, H. (2006). Identification of pre- and postselection TCRab+ intraepi- Published: March 14, 2013 thelial lymphocyte precursors in the thymus. Immunity 25, 631–641.

REFERENCES Godfrey, D.I., and Berzins, S.P. (2007). Control points in NKT-cell develop- ment. Nat. Rev. Immunol. 7, 505–518. Baldwin, T.A., Hogquist, K.A., and Jameson, S.C. (2004). The fourth way? Hogan, P.G., Chen, L., Nardone, J., and Rao, A. (2003). Transcriptional Harnessing aggressive tendencies in the thymus. J. Immunol. 173, 6515–6520. regulation by calcium, calcineurin, and NFAT. Genes Dev. 17, 2205–2232. Beyersdorf, N., Braun, A., Vo¨ gtle, T., Varga-Szabo, D., Galdos, R.R., Kissler, Hogan, P.G., Lewis, R.S., and Rao, A. (2010). Molecular basis of calcium S., Kerkau, T., and Nieswandt, B. (2009). STIM1-independent T cell develop- signaling in lymphocytes: STIM and ORAI. Annu. Rev. Immunol. 28, 491–533. ment and effector function in vivo. J. Immunol. 182, 3390–3397. Hsieh, C.S., Lee, H.M., and Lio, C.W. (2012). Selection of regulatory T cells in Bhakta, N.R., Oh, D.Y., and Lewis, R.S. (2005). Calcium oscillations regulate the thymus. Nat. Rev. Immunol. 12, 157–167. thymocyte motility during positive selection in the three-dimensional thymic environment. Nat. Immunol. 6, 143–151. Huesmann, M., Scott, B., Kisielow, P., and von Boehmer, H. (1991). Kinetics and efficacy of positive selection in the thymus of normal and T cell receptor Bouillet, P., Purton, J.F., Godfrey, D.I., Zhang, L.C., Coultas, L., Puthalakath, transgenic mice. Cell 66, 533–540. H., Pellegrini, M., Cory, S., Adams, J.M., and Strasser, A. (2002). BH3-only Bcl-2 family member Bim is required for apoptosis of autoreactive thymocytes. Jin, J., Desai, B.N., Navarro, B., Donovan, A., Andrews, N.C., and Clapham, Nature 415, 922–926. D.E. (2008). Deletion of Trpm7 disrupts embryonic development and thymo- poiesis without altering Mg2+ homeostasis. Science 322, 756–760. Boyman, O., Kovar, M., Rubinstein, M.P., Surh, C.D., and Sprent, J. (2006). Selective stimulation of T cell subsets with antibody-cytokine immune Kishimoto, H., Surh, C.D., and Sprent, J. (1998). A role for Fas in negative complexes. Science 311, 1924–1927. selection of thymocytes in vivo. J. Exp. Med. 187, 1427–1438.

894 Immunity 38, 881–895, May 23, 2013 ª2013 Elsevier Inc. Immunity T Cell Development and Calcium Signaling

Kisielow, P., Blu¨ thmann, H., Staerz, U.D., Steinmetz, M., and von Boehmer, H. channel is a critical regulator of T cell receptor signaling and naive T cell (1988). Tolerance in T-cell-receptor transgenic mice involves deletion of non- homeostasis. Immunity 35, 349–360. + + mature CD4 8 thymocytes. Nature 333, 742–746. Ouyang, W., Beckett, O., Ma, Q., and Li, M.O. (2010). Transforming growth Konkel, J.E., Maruyama, T., Carpenter, A.C., Xiong, Y., Zamarron, B.F., Hall, factor-b signaling curbs thymic negative selection promoting regulatory B.E., Kulkarni, A.B., Zhang, P., Bosselut, R., and Chen, W. (2011). Control of T cell development. Immunity 32, 642–653. + the development of CD8aa intestinal intraepithelial lymphocytes by TGF-b. Picard, C., McCarl, C.A., Papolos, A., Khalil, S., Lu¨ thy, K., Hivroz, C., LeDeist, Nat. Immunol. 12, 312–319. F., Rieux-Laucat, F., Rechavi, G., Rao, A., et al. (2009). STIM1 mutation asso- Kovalovsky, D., Uche, O.U., Eladad, S., Hobbs, R.M., Yi, W., Alonzo, E., Chua, ciated with a syndrome of immunodeficiency and autoimmunity. N. Engl. J. K., Eidson, M., Kim, H.J., Im, J.S., et al. (2008). The BTB-zinc finger transcrip- Med. 360, 1971–1980. tional regulator PLZF controls the development of invariant natural killer T cell Roos, J., DiGregorio, P.J., Yeromin, A.V., Ohlsen, K., Lioudyno, M., Zhang, S., effector functions. Nat. Immunol. 9, 1055–1064. Safrina, O., Kozak, J.A., Wagner, S.L., Cahalan, M.D., et al. (2005). STIM1, an 2+ Kreslavsky, T., Savage, A.K., Hobbs, R., Gounari, F., Bronson, R., Pereira, P., essential and conserved component of store-operated Ca channel function. Pandolfi, P.P., Bendelac, A., and von Boehmer, H. (2009). TCR-inducible PLZF J. Cell Biol. 169, 435–445. transcription factor required for innate phenotype of a subset of gd T cells with Savage, A.K., Constantinides, M.G., Han, J., Picard, D., Martin, E., Li, B., restricted TCR diversity. Proc. Natl. Acad. Sci. USA 106, 12453–12458. Lantz, O., and Bendelac, A. (2008). The transcription factor PLZF directs the Kronenberg, M., and Gapin, L. (2002). The unconventional lifestyle of NKT effector program of the NKT cell lineage. Immunity 29, 391–403. cells. Nat. Rev. Immunol. 2, 557–568. Seiler, M.P., Mathew, R., Liszewski, M.K., Spooner, C., Barr, K., Meng, F., Lambolez, F., Kronenberg, M., and Cheroutre, H. (2007). Thymic differentiation Singh, H., and Bendelac, A. (2012). Elevated and sustained expression of of TCRab+ CD8aa+ IELs. Immunol. Rev. 215, 178–188. the transcription factors Egr1 and Egr2 controls NKT lineage differentiation in response to TCR signaling. Nat. Immunol. 13, 264–271. Leishman, A.J., Gapin, L., Capone, M., Palmer, E., MacDonald, H.R., Kronenberg, M., and Cheroutre, H. (2002). Precursors of functional MHC class Shortman, K., Vremec, D., and Egerton, M. (1991). The kinetics of T cell antigen + + I- or class II-restricted CD8aa+ T cells are positively selected in the thymus by receptor expression by subgroups of CD4 8 thymocytes: delineation of + + 2+ agonist self-peptides. Immunity 16, 355–364. CD4 8 3 thymocytes as post-selection intermediates leading to mature T cells. J. Exp. Med. 173, 323–332. Li, M.O., Sanjabi, S., and Flavell, R.A. (2006). Transforming growth factor- b controls development, homeostasis, and tolerance of T cells by regulatory Sommers, C.L., Lee, J., Steiner, K.L., Gurson, J.M., Depersis, C.L., El-Khoury, T cell-dependent and -independent mechanisms. Immunity 25, 455–471. D., Fuller, C.L., Shores, E.W., Love, P.E., and Samelson, L.E. (2005). Mutation of the phospholipase C-g1-binding site of LAT affects both positive and nega- Lio, C.W., and Hsieh, C.S. (2008). A two-step process for thymic regulatory tive thymocyte selection. J. Exp. Med. 201, 1125–1134. T cell development. Immunity 28, 100–111. Starr, T.K., Jameson, S.C., and Hogquist, K.A. (2003). Positive and negative Liou, J., Kim, M.L., Heo, W.D., Jones, J.T., Myers, J.W., Ferrell, J.E., Jr., and selection of T cells. Annu. Rev. Immunol. 21, 139–176. Meyer, T. (2005). STIM is a Ca2+ sensor essential for Ca2+-store-depletion- Stritesky, G.L., Jameson, S.C., and Hogquist, K.A. (2012). Selection of self- triggered Ca2+ influx. Curr. Biol. 15, 1235–1241. reactive T cells in the thymus. Annu. Rev. Immunol. 30, 95–114. Liu, Y., Zhang, P., Li, J., Kulkarni, A.B., Perruche, S., and Chen, W. (2008). A Thornton, A.M., Korty, P.E., Tran, D.Q., Wohlfert, E.A., Murray, P.E., Belkaid, critical function for TGF-b signaling in the development of natural Y., and Shevach, E.M. (2010). Expression of Helios, an Ikaros transcription CD4+CD25+Foxp3+ regulatory T cells. Nat. Immunol. 9, 632–640. factor family member, differentiates thymic-derived from peripherally induced Marie, J.C., Liggitt, D., and Rudensky, A.Y. (2006). Cellular mechanisms of fatal Foxp3+ T regulatory cells. J. Immunol. 184, 3433–3441. early-onset autoimmunity in mice with the T cell-specific targeting of trans- Treiner, E., Duban, L., Bahram, S., Radosavljevic, M., Wanner, V., Tilloy, F., forming growth factor-b receptor. Immunity 25, 441–454. Affaticati, P., Gilfillan, S., and Lantz, O. (2003). Selection of evolutionarily McCarl, C.A., Khalil, S., Ma, J., Oh-hora, M., Yamashita, M., Roether, J., conserved mucosal-associated invariant T cells by MR1. Nature 422, 164–169. Kawasaki, T., Jairaman, A., Sasaki, Y., Prakriya, M., and Feske, S. (2010). Uehara, S., Grinberg, A., Farber, J.M., and Love, P.E. (2002). A role for CCR9 in Store-operated Ca2+ entry through ORAI1 is critical for T cell-mediated auto- T lymphocyte development and migration. J. Immunol. 168, 2811–2819. immunity and allograft rejection. J. Immunol. 185, 5845–5858. Vig, M., Peinelt, C., Beck, A., Koomoa, D.L., Rabah, D., Koblan-Huberson, M., Mempel, T.R., Henrickson, S.E., and Von Andrian, U.H. (2004). T-cell priming Kraft, S., Turner, H., Fleig, A., Penner, R., and Kinet, J.P. (2006). CRACM1 is a by dendritic cells in lymph nodes occurs in three distinct phases. Nature plasma membrane protein essential for store-operated Ca2+ entry. Science 427, 154–159. 312, 1220–1223. Nakayama, T., Ueda, Y., Yamada, H., Shores, E.W., Singer, A., and June, C.H. Wu, Y., Borde, M., Heissmeyer, V., Feuerer, M., Lapan, A.D., Stroud, J.C., (1992). In vivo calcium elevations in thymocytes with T cell receptors that are Bates, D.L., Guo, L., Han, A., Ziegler, S.F., et al. (2006). FOXP3 controls regu- specific for self ligands. Science 257, 96–99. latory T cell function through cooperation with NFAT. Cell 126, 375–387. Neilson, J.R., Winslow, M.M., Hur, E.M., and Crabtree, G.R. (2004). Calcineurin Wurbel, M.A., Malissen, M., Guy-Grand, D., Meffre, E., Nussenzweig, M.C., B1 is essential for positive but not negative selection during thymocyte devel- Richelme, M., Carrier, A., and Malissen, B. (2001). Mice lacking the CCR9 opment. Immunity 20, 255–266. CC-chemokine receptor show a mild impairment of early T- and B-cell devel- Oh-Hora, M., Yamashita, M., Hogan, P.G., Sharma, S., Lamperti, E., Chung, opment and a reduction in T-cell receptor gd+ gut intraepithelial lymphocytes. W., Prakriya, M., Feske, S., and Rao, A. (2008). Dual functions for the endo- Blood 98, 2626–2632. plasmic reticulum calcium sensors STIM1 and STIM2 in T cell activation and Zhang, S.L., Yeromin, A.V., Zhang, X.H., Yu, Y., Safrina, O., Penna, A., Roos, tolerance. Nat. Immunol. 9, 432–443. J., Stauderman, K.A., and Cahalan, M.D. (2006). Genome-wide RNAi screen of Ohoka, Y., Yokota, A., Takeuchi, H., Maeda, N., and Iwata, M. (2011). Retinoic Ca2+ influx identifies genes that regulate Ca2+ release-activated Ca2+ channel acid-induced CCR9 expression requires transient TCR stimulation and coop- activity. Proc. Natl. Acad. Sci. USA 103, 9357–9362. erativity between NFATc2 and the retinoic acid receptor/retinoid X receptor Zullo, A.J., Benlagha, K., Bendelac, A., and Taparowsky, E.J. (2007). complex. J. Immunol. 186, 733–744. Sensitivity of NK1.1-negative NKT cells to transgenic BATF defines a role for Omilusik, K., Priatel, J.J., Chen, X., Wang, Y.T., Xu, H., Choi, K.B., Gopaul, R., activator protein-1 in the expansion and maturation of immature NKT cells in

McIntyre-Smith, A., Teh, H.S., Tan, R., et al. (2011). The Cav1.4 calcium the thymus. J. Immunol. 178, 58–66.

Immunity 38, 881–895, May 23, 2013 ª2013 Elsevier Inc. 895