Thymocyte Egress in Regulating Mst1 a Cell-Intrinsic Role

Home , MST1

A Cell-Intrinsic Role for Mst1 in Regulating Thymocyte Egress Yongli Dong, Xingrong Du, Jian Ye, Min Han, Tian Xu, Yuan Zhuang and Wufan Tao This information is current as of September 27, 2021. J Immunol 2009; 183:3865-3872; Prepublished online 19 August 2009; doi: 10.4049/jimmunol.0900678 http://www.jimmunol.org/content/183/6/3865 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2009/08/19/jimmunol.090067 Material 8.DC1

References This article cites 43 articles, 17 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/183/6/3865.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 27, 2021

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2009 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

A Cell-Intrinsic Role for Mst1 in Regulating Thymocyte Egress1

Yongli Dong,* Xingrong Du,* Jian Ye,* Min Han,*† Tian Xu,*‡ Yuan Zhuang,2*§ and Wufan Tao2*

The MST1 kinase was recently identified as playing an essential role in the promotion of lymphocyte polarization and adhesion stimulated by chemokines and TCR signaling. However, the physiological relevance of the Mst1 pathway in thymocyte development is not completely understood. In this study, we analyzed the effect of Mst1 disruption on thymocyte ,development and migration. Mst1-deficient (Mst1؊/؊) mice displayed an accumulation of mature thymocytes in the thymus a dramatic reduction of lymphocytes in blood and peripheral lymphoid tissues, and a decrease of homing ability to peripheral -lymph nodes. Mst1؊/؊ thymocytes were impaired in chemotactic response to chemokines, such as CCL19, but not to sphin gosine-1-phosphate. Further analyses of Mst1؊/؊ mice revealed a severe impairment in the egress of mature T cells from the thymus. T lineage-specific knockout of the Mst1 gene demonstrates a cell-intrinsic role for Mst1 in regulating T cell development. Our study indicates that Mst1 is crucial in controlling lymphocyte chemotaxis and thymocyte emigration. The Downloaded from Journal of Immunology, 2009, 183: 3865–3872.

hymocyte maturation follows a distinct pattern of cell mi- Therefore, S1P1-deﬁcient mice are lymphopenic in the periphery gration in the thymus and ends with the egress of fully (6, 7). matured T cells. Chemokines and their receptors play cru- Chemokine receptors activate multiple signaling pathways, in-

T http://www.jimmunol.org/ cial roles in regulating thymocyte migration and egress (1–3). For cluding the Ras/Rho family of small GTPases such as Rac and example, thymocyte migration from the cortex to the medulla is Rap1 (8, 9). Lymphocytes lacking DOCK2, a Rac guanine ex- strictly controlled by chemokine receptor CCR7, whereas subse- change factor, show partial defects in the polymerization of F-actin quent egress from the medulla to the circulation is tightly regulated induced by the S1P1 ligand S1P, as well as impaired emigration by a chemoattractant lipid receptor, the sphingosine-1-phosphate from lymph nodes (10). The actin-nucleating and polymerization 3 Ϫ/Ϫ (S1P) receptor 1 (S1P1). Disruption of the CCR7 gene (CCR7 protein mDia1, acting as a downstream Rho GTPase effector, is mice) or of the locus encoding the CCR7-speciﬁc chemokines also needed for thymic egress (11). The Rap1 effector molecule, CCL19/21 ( pltϪ/Ϫ mice) prevented the migration of thymocytes ␣ RAPL, is required for lymphocyte adhesion through LFA-1 and 4 from the cortex to the medulla (4). Consequently, thymocytes were

integrins and for cell polarization triggered by chemokines (12, by guest on September 27, 2021 Ϫ/Ϫ found trapped in the thymus of newborn CCR7 mice (5) and 13). RAPL-deficient mice display defects in thymic emigration and inappropriately emigrated via the corticomedullary junction in lymphocyte homing to peripheral lymph nodes. RAPL forms a Ϫ/Ϫ Ϫ/Ϫ adult CCR7 and plt mice (4). S1P1-deficient T cells de- complex with MST1 and activates MST1 kinase activity. Knock- velop normally but fail to leave the thymus, resulting in an accu- down of Mst1 abolished the RAPL-mediated polarizing morphol- mulation of mature thymocytes in the medullar area of the thymus. ogy and integrin-dependent lymphocyte adhesion (14). Mst1 is a member of the Hpo pathway identified in both inver- *Institute of Developmental Biology and Molecular Medicine, School of Life Sci- tebrates and vertebrates (15, 16). Hpo,aDrosophila homologue of ence, Fudan University, Shanghai, China; †Howard Hughes Medical Institute, De- partment of Molecular, Cell, and Developmental Biology, University of Colorado, the mammalian Mst1 gene, controls organ size by regulating cell Boulder, CO 80309; ‡Howard Hughes Medical Institute, Department of Genetics, growth, survival, and proliferation (17–20). Hpo also regulates til- Yale University School of Medicine, New Haven, CT 06536; and §Department of ing and maintenance, two complementary aspects of dendrite de- Immunology, Duke University Medical Center, Durham, NC 27701 velopment in the nervous system (21). Cst-1,anMst1 homologue Received for publication March 4, 2009. Accepted for publication July 14, 2009. in Caenorhabditis elegans, is responsible for maintenance of the The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance normal life span and prevention of tissue aging (22). Mst1 was with 18 U.S.C. Section 1734 solely to indicate this fact. demonstrated in a variety of mammalian cultured cells to promote 1 This study was supported by Chinese Key Projects for Basic Research (973) apoptosis through caspase-mediated proteolytic activation and Grant 2006CB806700, Hi-Tech Research and Development Project (863) Grant phosphorylation (23–25) or via the Ki-Ras/Nore1 proapoptotic 2007AA022101, National Natural Science Foundation of China Grant 30630043, Key Projects Grant for Basic Research 08JC1400800 from the Science and Technology pathway (26). Recently, a study of Mst1 mutant mice derived from Committee of Shanghai Municipality, Shanghai Pujiang Program Grant 05PJ14024, a gene trap embryonic stem (ES) cell line revealed a reduction of and the 211 and 985 projects of the Chinese Ministry of Education. lymphocytes in the mutant and confirmed the role for Mst1 in the 2 Address correspondence and reprint requests to Dr. Wufan Tao, Institute of Devel- polarization and adhesion of lymphocytes upon stimulation by opmental Biology and Molecular Medicine, Fudan University, Shanghai 200433, the People’s Republic of China. E-mail address: [email protected] or Dr. Yuan TCR and CCL21 (27). Zhuang, Department of Immunology, Duke University Medical Center, Durham, NC In this article we report the generation and analysis of an Mst1 27701. E-mail address: [email protected] knockout allele that is incapable of producing any functional 3 Abbreviations used in this paper: S1P, sphingosine-1-phosphate; 7AAD, 7-amino MST1 proteins. In addition to a severe reduction of lymphocytes in actinomycin D; CMFDA, 5-chloromethylfluorescein diacetate; CMTMR, 5-(and-6)- (((4-chloromethyl)benzoyl)amino)tetramethylrhodamine; DP, double positive; ES, blood and peripheral lymphoid tissues, our Mst1-deficient mice embryonic stem; HET, heterozygous; HO, homozygous; PGK, phosphoglycerate ki- displayed an accumulation of mature thymocytes in the thymus. nase; S1P , S1P receptor 1; SP, single positive; WT, wild type. 1 We show that the Mst1-deficient thymocytes are impaired in che- Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00 motactic responses to chemokines such as CCL19, but not to S1P. www.jimmunol.org/cgi/doi/10.4049/jimmunol.0900678 3866 Mst1 REGULATES THYMOCYTE EMIGRATION

These mice also displayed severe defects in the egress of mature T mocytes were calculated as following: FITCϩCD4ϩ T cells in the spleen are divided by the sum of FITCϩCD4ϩ mature T cells in the thymus and cells from thymus. A cell-intrinsic role for Mst1 in regulating T ϩ ϩ cell development is further demonstrated by T lineage-specific de- FITC CD4 T cells in the spleen. Mst1 letion of the gene. Our studies reveal a novel and essential Adoptive transfer role for Mst1 in thymocyte emigration. Lymphocytes from WT and Mst1Ϫ/Ϫ spleen and lymph nodes were labeled Materials and Methods with 5 ␮M 5-(and-6)-(((4-chloromethyl)benzoyl)amino)tetramethylrhodamine (CMTMR) and 1 ␮M 5-chloromethylfluorescein diacetate Generation of Mst1-deficient mice (CMFDA) (Molecular Probes), respectively, as described (32). The labeled ϩ The genomic clones containing exon 2 to exon 9 of Mst1 were isolated cells were mixed with an equal number of T cells or CD4 T cells and injected into the tail vein of 129/Sv mice. The labeled CD4ϩ and CD8ϩ T from a 129/Sv mouse genomic phage library (Stratagene). The Mst1 gene ϩ was modified by adding loxP sequences to XhoI and NcoI sites flanking cells (or CD4 T cells only) from the spleen, lymph nodes (axillary and exon 4. A neomycin (Neo) resistance cassette flanked by Flp recognition brachial), mesenteric lymph node, and peripheral blood and lymph were target (FRT) sites and a diphtheria toxin A (DTA) expression cassette were analyzed separately by flow cytometry 1 and 24 h after injection. For the used as positive and negative selection markers, respectively. Gene target- survival test, similar analyses were performed on day 5 after injection. ing was conducted in the W4 ES cells line (ES-W4129S6; Taconic Trans- genic). Chimeric mice were bred with phosphoglycerate kinase (PGK)-Flp In vitro and organ chemotaxis assay (stock no. 003946; The Jackson Laboratory) transgenic mice with 129/Sv Thymocyte chemotaxis assays were performed as described previously (7, genetic background to remove the PGK-Neo positive selection marker 6 ϩ ϩ Ϫ ϩ 33). Briefly, 1 ϫ 10 thymocytes were loaded into the upper chamber and and generate Mst1 /fl mice. Mst1 / mice were derived from Mst1 /fl allowed to transmigrate into the lower chamber containing either 1 ␮g/ml mice by removing exon 4 through crossing with PGK-Cre (28) trans- ␮ ␮ CCL19 (Peprotech), 1 g/ml CCL21 (Peprotech), 2 g/ml CCL25 (R&D Downloaded from genic mice. Progenies were genotyped by PCR containing a common Systems), 300 ng/ml CXCL12 (Peprotech), or S1P at the indicated con- oligo (Mst1-F2, 5Ј-GCTGATCCATGTCTCTACTCC-3Ј), a wild-type- centrations for 3 h. The input cells and the cells that migrated to the lower specific oligo (Mst1-R2, 5Ј-GCTGCCATAATACTTGACTACG-3Ј), chamber were stained with anti-CD4, -CD8, and -CD62L Abs for flow and a mutant-specific oligo (Mst1-R3, 5Ј-CCAGGCATGGTAGG- cytometry analysis. GAGAATG-3Ј). All mice mentioned above were maintained on a Organ chemotaxis assay was performed as described (34) with minor 129/Sv genetic background. ϩ Ϫ modifications; 100 ␮l of RPMI 1640 and 450 ␮l of RPMI 1640 with 500 To specifically mutate Mst1 gene in T cells, Mst1 / :LCK-Cre mice ϩ/Ϫ ng/ml CCL19 were added to the upper and lower chambers, respectively.

were generated by crossing Mst1 mice with the LCK-Cre transgenic http://www.jimmunol.org/ ϩ Ϫ Thymocytes were allowed to emigrate out of thymus and transmigrate into mice (29). Mst1fl/fl mice were then crossed with Mst1 / :LCK-Cre mice on the lower chamber for 3 h before subjection to Ab staining with CD4-FITC 129/Sv and C57BL/6J mixed background. Progenies were genotyped by and CD8-PE and flow cytometry analysis. PCR using two sets of oligos specific for Mst1 (Mst1-F2 and Mst1-R2) and Ј Ј Cre (Neo-Cre F1,5-GGAAAATGCTTCTGTCCGTTTG-3 ; and Cre-R2, 5Ј-CGCATAACCAGTGAAACAGCATTGC-3Ј) genes, respectively. Six- Transendothelial migration assay to 9-wk-old littermates were used for all experiments described in this Assay conditions were adapted from Shulman et al. (35). bEnd.3 cells report. All experiments were performed in accordance with protocols ap- (7.5 ϫ 104)in200␮l of RPMI 1640 were seeded on the upper surface of proved by the Animal Care and Use Committee of the Institute of Devel- a Transwell insert coated by 0.1% gelatin and cultured at 37°C in 5% CO2. opmental Biology and Molecular Medicine at Fudan University, Shanghai, Three days later the bEnd.3 cells were stimulated with 20 ng/ml TNF-␣ for People’s Republic of China. 24 h. After washing the endothelial cell layer with PBS, thymocytes (1 ϫ 6 ␮ by guest on September 27, 2021 RT-PCR 10 )in100 l of RPMI 1640 with 0.5% fatty acid-free BSA were added to the upper chamber, and 600 ␮l of medium with 100 nM S1P was added

Total mRNAs were isolated from livers of adult mice using TRIzol (Invitro- to the lower chamber. After incubation for4hat37°C in 5% CO2, trans- gen). cDNAs were reverse transcribed using an TaKaRa RNA PCR (avian migrated thymocytes were collected to quantify the numbers by flow myeloblastosis virus) kit, version 3.0, and amplified with the combination of cytometry. specific primers for exon 1 and exon 3 or exon 4 and exon 9. The primers are as follows: exon 1, 5Ј-AGCTGAGGAACCCACCGC-3Ј; exon 3, 5Ј-GG Annexin V assay CACTTGCTTGATTGCAACAATC-3Ј; exon 4, 5Ј-CAGCTACTTTAAGA ␮ ACACAGACCTG-3Ј; and exon 9, 5Ј-CTTCCTCTTCCTCGTCCTCCG-3Ј. Freshly isolated splenocytes and lymphocytes were resuspended in 100 l of binding buffer (10 mM HEPES (pH7.4), 0.14 M NaCl, and 2.5 mM

Western blotting CaCl2) supplied with anti-CD4 (or anti-CD8), anti-CD44 Abs, annexin V-FITC (BD Pharmingen), and 7AAD. After incubation for 15 min at Proteins from mouse kidneys were extracted by radioimmune precipitation room temperature in the dark, cells were supplied with 300 ␮l of binding ϫ assay (RIPA) buffer containing 1 mM PMSF and 1 proteinase inhibitor buffer followed by FACS analysis within 1 h. The annexin V-positive and (Roche). The proteins were separated by electrophoresis on 10% SDS- 7AAD-negative cells were defined as apoptotic. polyacrylamide gels followed by Western blotting with anti-MST1 (Cell Signaling) as described by Callus et al. (30). The same amounts of proteins were used for the control probed with anti-actin (catalog no. sc1615; Santa Results Cruz Biotechnology). Generation of Mst1-deficient mice Flow cytometric analysis The Mst1 gene was mutated in ES cells via gene targeting (Fig. Single cell suspensions were prepared from thymus, bone marrow, spleen, 1A). Homologous recombinants were identified from G418-resis- lymph nodes (left axillary), and Peyer’s patches in PBS containing 5% tant colonies by PCR and then confirmed by Southern blotting

FBS. Erythrocytes from blood were removed in 155 mM NH4Cl lysis (Fig. 1B) using the probe shown in Fig. 1A. Targeted ES cell buffer. Cells were stained with Abs (eBioscience or Caltag Laboratories) clones were injected into C57BL/6 blastocysts to establish germ- and 7-aminoactinomycin D (7AAD) (Sigma-Aldrich), and washed once line transmission of the target allele. The PGK-Neo marker and with 1ϫ PBS containing 5% FBS before being analyzed on FACSCalibur (BD Biosciences). In all cases, statistical analyses between wild type (WT) exon 4 of the Mst1 coding region were sequentially removed by and heterozygous (HET) or WT and homozygous (HO) mice were per- crossing to the PGK-Flp and PGK-Cre transgenic lines, respec- formed using Student’s t test. tively. Deletion of exon 4 resulted in a frame shift and a premature Analysis of recent thymic emigration stop codon in exon 5. WT Mst1 mRNA transcripts were not de- tected in the HO mutant mice by RT-PCR (Fig. 1C). Western The intrathymic labeling was described previously (31). Each thymus lobe blotting showed that the MST1 protein level was reduced in HET of an anesthetized mouse was directly injected with 10 ␮lof600␮g/ml FITC (F7250; Sigma-Aldrich) solution in PBS. Recent thymic emigration and became undetectable in HO mice (Fig. 1D). These results con- was assessed 24 h after dye injection. Spleens were harvested and analyzed firmed the disruption of the Mst1 gene by the targeted allele, which for FITCϩCD4ϩ T cells by flow cytometry. The recently emigrated thy- is referred as deficient or negative (Ϫ) hereafter. Animals from the The Journal of Immunology 3867 Downloaded from http://www.jimmunol.org/

FIGURE 1. The generation of Mst1 knockout mice. A, Schematic illustrations of the knockout strategy for Mst1 gene. A1, Schematic structure of the targeting vector, pMst1-KO3. DTA and Neo represent the diphtheria toxin A and the neomycin expression cassettes, respectively, with the PGK promoter. Arrows in the boxes indicate the transcriptional direction of the Neo and DTA genes. E4 and E5 are exons 4 and 5, and H is the HindIII restriction enzyme. A2, Restriction map of the mouse Mst1 locus. E3 represents exon 3, N and X are the NcoI and XhoI restriction enzymes, respectively. Note that the NcoI and XhoI sites shown are not unique sites in the locus and that other sites than those shown also exist in the locus. A3, Structure of the modified Mst1 locus after homologous recombination. A4, Structure of the mutated locus after removing the Neo cassette by Flp and a fragment containing exon 4 by CRE. FRT and loxP sequences are represented by black and white arrowheads, respectively. Arrows a and b in A3 are PCR primers for detecting the modified Mst1 allele (Mst1-MF2, 5Ј-GTCTGTTACCTCCTGTCCTCGTGTTC-3Ј; and Mst1-R4, 5Ј-CATTTGTCACGTCCTGCACGAC-3Ј). The probe for Southern blotting is shown as a bold line in A2. The expected sizes of the HindIII fragments hybridized with the probe are indicated for the WT in A2 by guest on September 27, 2021 and for modified Mst1 alleles in A3. B, Southern blot analysis of genomic DNA from targeted ES clones. The 3.3- and 5.1-kb HindIII fragments corresponding to the WT and modified alleles are indicated. One representative ES clone containing the modified allele (M) and a WT clone (ϩ/ϩ) are shown. C, RT-PCR analysis of Mst1 expression in livers using primers specific to Mst1 cDNA (see Materials and Methods). D, Western blot analysis of MST1 protein expression in kidneys. Actin was used as loading controls. ϩ/ϩ, WT; ϩ/Ϫ, HET; Ϫ/Ϫ, HO; Neg, negative control; KD, Kilodalton.

Mst1ϩ/Ϫ intercross exhibited Mendelian frequencies of inheri- (CD44low) and activated (CD44high) T cells in blood and second- tance. Mst1Ϫ/Ϫ mice were viable and phenotypically indistinguish- ary lymphoid organs (Fig. 3A). Although activated T cells isolated able from WT and HET littermates in a 6-mo window of analysis. from lymph nodes showed a marginal enhancement of annexin V staining, no significant alteration in ongoing apoptosis was ob- A severe reduction of peripheral lymphocytes in Mst1-deficient served among naive T cells freshly isolated from spleen and lymph mice nodes (Fig. 3, B and C). We also assessed the effect of Mst1 de- To investigate the physiological roles of Mst1 in lymphocyte de- ficiency on the survival of peripheral T cells in vivo in an adoptive velopment, we first examined the distribution of lymphocytes in transfer assay. Differentially labeled HO and WT T cells were blood and peripheral lymphoid organs in Mst1Ϫ/Ϫ mice. The av- cotransferred into 129/Sv hosts. The ratio of HO to WT at day 5 erage numbers of CD4ϩ T cells, CD8ϩ T cells, and B cells in (0.94 Ϯ 0.05) after transfer was essentially unchanged from that blood were decreased by 85.9 Ϯ 1.3%, 88.1 Ϯ 5%, and 76 Ϯ 1%, before injection (0.95 Ϯ 0.06) (supplemental data Fig. S1).4 These Ϫ Ϫ respectively, compared with the WT controls (Fig. 2C). The av- results indicate that the reduction of peripheral T cells in Mst1 / erage numbers of CD4ϩ and CD8ϩ T cells in the spleen were mice, particularly naive T cells, is unlikely due to a general en- reduced to 45 Ϯ 1.8% and 26 Ϯ 0.4%, respectively, of that in WT hancement of apoptosis. controls. Lymphocyte paucity was also observed in axillary lymph Mst1-deficient T cells are impaired for homing to peripheral nodes and Peyer’s patches (Fig. 2D). lymphoid organs Mst1-deficient naive T cells are phenotypically viable Lymphocyte adhesion and polarization play a critical role in lym- To understand the cause of lymphocyte paucity, we used an an- phocyte homing to peripheral lymphoid organs (37). Given the nexin V assay to assess the ongoing apoptosis of peripheral T cells reported roles of Mst1 in lymphocyte adhesion and polarization freshly isolated from the Mst1-deficient mice. Because the in vivo (14), we next investigated whether the homing of peripheral Ϫ/Ϫ apoptotic rates of naive T cells and activated T cells are different Mst1 T cells into secondary lymphoid organs was affected. (36), we analyzed the apoptosis of naive and activated T cells separately. Mst1 disruption altered the proportion of naive 4 The online version of this article contains supplemental material. 3868 Mst1 REGULATES THYMOCYTE EMIGRATION Downloaded from FIGURE 2. Lymphopenia in Mst1-deficient mice. A and B, Reduced proportions of Mst1Ϫ/Ϫ CD4ϩ, CD8ϩ T cells and B220ϩ B cells in blood. Representative profiles of CD4ϩ, CD8ϩ cells (A) and B220ϩ cells (B) in peripheral blood from mice with the indicated genotypes (ϩ/ϩ, WT; ϩ/Ϫ, HET; Ϫ/Ϫ, HO). C, The numbers of total lymphocyte (left) and CD4ϩ, CD8ϩ and B220ϩ subset (right) cells in peripheral blood. D, The numbers of CD4ϩ and CD8ϩ T cells in spleen (SPL), left axillary lymph nodes (LN), and Peyer’s patches (PP). The means and SEM of cell numbers are shown. In C and D, p Ͻ 0.01; n ϭ 5, except ,ءء p Ͻ 0.05; and ,ء .statistical analyses between WT and HET mice and WT and HO mice were performed using Student’s t test lymph nodes (n ϭ 10). http://www.jimmunol.org/

Equal numbers of Mst1Ϫ/Ϫ and WT T cells labeled with the flu- cells in the spleen was increased to ϳ1.2:1. However, the ratio of orescent dyes CMFDA and CMTMR, respectively, were mixed Mst1Ϫ/Ϫ to WT CD4ϩ T cells in blood as well as in lymph was and transferred into WT recipient mice by tail vein injection. The reduced to ϳ0.5:1 (Fig. 4C), consistent with the idea that Mst1Ϫ/Ϫ homing ability of T cells to lymph nodes was then examined 1 h T cells are defective in emigrating into the circulation after enter- posttransfer. The ratio of labeled Mst1Ϫ/Ϫ T cells to labeled WT ing secondary lymphoid organs or nonlymphoid tissues. Taken cells in the lymph nodes had decreased to one-half of the starting together, Mst1 is crucial for peripheral lymphocyte trafficking. value (Fig. 4B), indicating a homing defect for Mst1Ϫ/Ϫ T cells. by guest on September 27, 2021 The ratio for the two donor types in blood was found to be ϳ1.5:1, An accumulation of mature SP thymocytes in Mst1-deficient which is consistent with a delayed entry of Mst1Ϫ/Ϫ T cells into mice lymphoid organs (Fig. 4B). These results indicate that Mst1-defi- We investigated whether the reduction of peripheral T cells was cient T cells are impaired for homing to peripheral lymphoid or- due to any defects in thymocyte development. Mst1-deficient mice gans. We also performed similar analyses 24 h after transfer. In- showed normal numbers of double-positive (DP) thymocytes but a terestingly, it was found that the ratio of Mst1Ϫ/Ϫ to WT CD4ϩ T significant increase in single-positive (SP) thymocytes (Fig. 5, A

FIGURE 3. Status of apoptosis among naive and activated T cells in spleen and LNs. A, Ratios of CD44low (CD44lo) naive and CD44high (CD44hi) activated T cells freshly isolated from blood (BL), spleen (SPL), mesenteric lymph nodes (MN), and axillary lymph nodes (LN). B and C, Percentages of apoptotic cells (annexin Vϩ and 7AADϪ) among CD4ϩCD44low and CD4ϩCD44high (B) and CD8ϩCD44low and CD8ϩCD44high (C) T cells freshly isolated from spleen and lymph nodes. The means and SEM of apoptotic cells (%) are shown. Statistical analyses between WT and HET mice and WT and HO .p Ͻ 0.05; n ϭ 5 ,ء .mice were performed using Student’s t test The Journal of Immunology 3869

FIGURE 4. Differential homing rates of T cells from WT and Mst1Ϫ/Ϫ mice. A, Representative proﬁles of CMTMR- or CMFDA-labeled T cells Downloaded from before injection and in the blood and the axillary/brachial lymph nodes from recipients 1 h postinjection. B, Ratios of labeled Mst1Ϫ/Ϫ (HO) and WT T cells before injection and in lymphoid tissues as indicated 1 h postinjection. C, Ratios of labeled CD4ϩ T cells in the tissues 24 h postinjection. BI, Before injection; BL, blood; SPL, spleen; LN, axillary/brachial lymph node; MN, mesenteric lymph node; LY, lymph. The means and SEM of http://www.jimmunol.org/ ratios are shown; n ϭ 3. and B). The accumulation of SP thymocytes did not lead to any obvious perturbation of thymic cortex and medulla structures (supplemental Fig. S2), suggesting a proper occupancy of the cortex area with SP thymocytes. CD4 and CD8 SP cells in the thymus can be divided into two subgroups: CD62Llow and CD69high immature

SP cells and CD62Lhigh and CD69low mature SP cells (38). Mst1- by guest on September 27, 2021 deficient SP T cell populations contained a higher proportion of mature cells with CD62Lhigh and CD69low phenotypes (Fig. 5C). This result suggests that Mst1 disruption leads to an accumulation FIGURE 5. Accumulation of mature CD4 and CD8 SP T cells in Ϫ/Ϫ of mature SP cells in the thymus. Mst1 mice. A, Representative CD4 and CD8 staining of total live thymocytes. Percentages of DP (ϩ/ϩ) and SP cells (Ϫ/Ϫ) are indicated in the Mst1 is required for thymocyte egress dot plots. B, Total numbers of thymocytes and DP T cells (left panel), and CD4 and CD8 SP T cells (right panel) in thymi from mice with the indi- The accumulation of mature SP cells in the thymus and the reduc- cated genotypes. C, Expression of the maturation markers CD62L and Ϫ Ϫ tion of peripheral lymphocytes in Mst1 / mice suggests a defect CD69 on Mst1Ϫ/Ϫ (open) and WT (gray) CD4 SP thymocytes. The same Ϫ Ϫ in thymic egress. To examine whether Mst1 / thymocytes are numbers of total thymocytes were analyzed for the WT and Mst1Ϫ/Ϫ sam- impaired in exiting the thymus, we analyzed the in vivo ability of ples. D, The recent thymus emigrants in Mst1Ϫ/Ϫ and WT mice. FITC- ϩ Ϫ Ϫ mature thymocytes to emigrate from the thymus by tracing the labeled CD4 T cells in the spleens of the Mst1 / and WT mice were recent thymus emigrants after intrathymic injection of the fluores- analyzed by FACS 24 h after intrathymic injection of FITC. E, Total numbers of thymocytes and DP T cells in Mst1fl/Ϫ:LCK-Cre mice. F, Accu- cent dye FITC. As shown in Fig. 5D, the recent emigration from fl/Ϫ Ϫ Ϫ mulation of mature CD4 and CD8 SP T cells in the thymi of Mst1 : the Mst1 / thymus into the spleen was significantly less than that LCK-Cre mice. The means and SEM of cell numbers are shown. in WT control mice 24 h after FITC thymic injection. This result Statistical analyses between WT and HET mice and WT and HO mice strongly suggests a defect in the output of mature T cells from the were performed using Student’s t test; n ϭ 5 for B and n ϭ 3 for D, E, Ϫ/Ϫ .p Ͻ 0.01 ,ءء ;p Ͻ 0.05 ,ء .thymus to the periphery in adult Mst1 mice. and F A T cell intrinsic role for Mst1 The above studies cannot determine whether T cell develop- Taken together, we conclude that disruption of the Mst1 gene in mice mental defects reflect a T cell-intrinsic role of Mst1. To address results in a cell-intrinsic defect of thymic emigration. this question, we generated Mst1 T cell-specific knockout mice (Mst1fl/Ϫ:LCK-Cre) by crossing Mst1fl/fl with Mst1ϩ/Ϫ:LCK- Mst1-deficient thymocytes show normal chemotactic response to Cre mice (supplemental Fig. S3A). Similarly as the Mst1Ϫ/Ϫ S1P mice, Mst1fl/Ϫ:LCK-Cre mice displayed a severe reduction of Egress of mature SP cells has been shown to depend on the che- peripheral T cells (supplemental data S3C) and an accumulation motactic response of the receptor S1P1 to the chemoattractant lipid of SP T cells in the thymus in addition to an overall increase of S1P (6, 7). We thus examined whether Mst1 plays a role in S1P1- the number of total thymocytes (Fig. 5E). Further analyses also mediated chemotaxis of thymocytes. As expected, immature T showed that only mature thymocytes accumulated in the thymi cells (CD4 SP, CD62Llow and CD8 SP, CD62Llow) from neither of Mst1fl/Ϫ:LCK-Cre mice, not immature SP thymocytes (Fig. 5F). WT nor Mst1Ϫ/Ϫ mice responded to S1P (Fig. 6, A and B, left 3870 Mst1 REGULATES THYMOCYTE EMIGRATION

FIGURE 6. Thymocyte chemotactic response to S1P. A and B, Chemo- taxis of Mst1Ϫ/Ϫ and WT CD4 (left) and CD8 SP thymocytes (right)toS1P at the indicated concentrations in Transwell assays. C and D, The chemotactic response of different thymocyte populations to 100 nM S1P in transendothelial migration assays. CD62hi, CD62Lhigh; CD62Llo, CD63Llow. The average percentage of input cells of the indicated population that migrated to S1P is shown; n ϭ 3. Downloaded from panels). Mature CD4 SP (CD62Lhigh) T cells from both WT and thymocytes are not affected by Mst1 knockout (supplemental Fig. Mst1Ϫ/Ϫ thymi displayed a similar chemotactic response to S1P 5, A and B). Collectively, these results indicate that Mst1 may play (Fig. 6A, right panel). Furthermore, Mst1Ϫ/Ϫ mature CD8 SP a generic role downstream of multiple chemokine receptors during (CD62Lhigh) T cells showed a hyper-response to S1P (Fig. 6B, thymic migration and egress. Ϫ/Ϫ right panel). To further evaluate the ability of Mst1 thymocytes http://www.jimmunol.org/ to undergo transendothelial migration in response to S1P, we per- Discussion formed an in vitro transendothelial migration assay. The result Genetic ablation of the Mst1 gene resulted in lymphopenia. We showed that Mst1 deﬁciency did not inﬂuence the migration of provide compelling evidence that the lymphopenia phenotype was mature T cells across endothelial cells (Fig. 6, C and D). These primarily caused by an impaired egress of mature T cells from the Ϫ Ϫ data indicate that Mst1 / thymocytes are capable of undergoing thymus. We further show that Mst1Ϫ/Ϫ thymocytes displayed im- migration in responding to the S1P signal. paired chemotactic response to several chemokines such as

Ϫ/Ϫ CCL19, CCL21, CXCL12, and CCL25. However, S1P-mediated Mst1 thymocytes are defective in chemotaxis to chemokine chemotactic migration of mature thymocytes remained intact, in- CCL19, CCL21, CXCL12, and CCL25 dicating that Mst1 is unlikely to be involved in supporting S1P- by guest on September 27, 2021 CCL19 and CCL21 are two chemokines involved in T cell migra- induced thymic egress. These results support the notion that thy- tion in thymus and thymocyte egress in neonatal mice (4, 5, 39). mic emigration is regulated by both S1P-dependent and Knockdown Mst1 in T cells abolished CCL21-induced polarity -independent mechanisms (41) and that Mst1 plays a particularly and adhesion to ICAM (14). To determine whether Mst1 is re- important role in S1P-independent chemotactic responses. quired for CCR7-mediated thymocyte migration, we examined the CCL19 has been demonstrated to be highly enriched on med- chemotactic response of Mst1Ϫ/Ϫ thymocytes to CCL19 and ullar endothelial venules and is speculated to mediate egress by CCL21 in chemotaxis assays. Our results showed that both guiding the positioning of mature thymocytes close to blood ves- Mst1Ϫ/Ϫ CD4 and CD8 SP T cells exhibited a lower response to sels (5). Recently, CCL19/CCL21 has been shown to directly ac- CCL21 than WT controls (Fig. 7, A and B). Mst1Ϫ/Ϫ CD4 SP T tivate Mst1 through engagement of CCR7 in T cells. Mst1 regu- cells, but not CD8 SP cells, also showed significantly reduced lates T lymphocytes homing through induction of both a polarized response to CCL19 (Fig. 7, C and D, left panels). Further analysis morphology and integrin clustering and adhesion triggered by the revealed that immature CD8 SP T cells significantly reduced the ligands of CCR7, CCL19/CCL21 (14, 27). We propose that a sim- response to CCR19, whereas mature CD8 SP T cells did not (Fig. ilar mechanism could be involved in regulating thymocyte egress. 7D). To further examine whether Mst1 is directly involved in This idea is supported by the observation that integrin LFA-1 ex- CCR7-mediated egress, we performed in vitro chemotaxis exper- pression is down-regulated in CD8 SP and mature CD4 SP T cell iments using whole thymi. The study revealed that the number of in Mst1Ϫ/Ϫ mice (supplemental data Fig. S6, A and B). mobilized thymocytes from a Mst1Ϫ/Ϫ thymus in response to Although our study indicated a potential role for Mst1 in the CCL19 were significantly lower than that from a WT thymus (Fig. CCR7-mediated pathway in thymocyte migration, the phenotypes 7E). These studies suggest that the CCR7-Mst1 pathway may be of Mst1 knockout mice do not overlap with those from CCR7 involved in thymocyte migration and egress. knockout or plt mice. Although both CCR7 knockout and plt mice Next, we investigated whether Mst1 plays a generic role down- showed a dramatically reduced medulla area due to the retention of stream of chemokine receptors during thymocyte development. SP cells in the cortex (4), the cortex-medulla structure appears The chemokines CXCL12 and CCL25, which use CXCR4 and normal in Mst1Ϫ/Ϫ mice (supplemental Fig. S2). There could be CCR9, respectively, as receptors, are known to be expressed in the multiple explanations for this difference. One possibility is that the thymus (40). CXCL12 is expressed in the medulla, the cortico- absence of Mst1 may only reduce the efficiency of, rather than medullary junction, and few cortical cells, and CCL25 is expressed completely block, migration from cortex to medulla. A slower but in the cortex and the medulla (40). In the Transwell assay, constant rate of migration from the cortex to the medulla could Mst1Ϫ/Ϫ SP thymocytes displayed a significantly reduced response lead to an accumulation of SP cells in the medulla if the subse- to CXCL12 and CCL25 (supplemental Fig. S4, A and B). Pheno- quent egress is blocked. Alternatively, the loss of Mst1 function typing assay showed that the expressions of CXCR4 and CCR7 on could also be partially compensated by the activity of the Mst1 The Journal of Immunology 3871 Downloaded from http://www.jimmunol.org/

FIGURE 7. Impaired chemotactic response of Mst1Ϫ/Ϫ thymocytes to CCL21 and CCL19. A and B, Chemotactic response of SP thymocytes toward a fixed concentration of CCL21 (1 ␮g/ml) for a duration of 3 h. C and D, Migration rate of SP thymocytes toward a fixed concentration of CCL19 (1 ␮g/ml) hi high lo low for a duration of 3 h. The means and SEM of input cells migrated are shown in A–D. CD62 , CD62L ; CD62L , CD63L . E, Emigration of thymocytes by guest on September 27, 2021 toward CCL19 from thymic lobes in Transwell chemotactic chambers. CD4 and CD8 profiles of cells that emigrated to CCL19 from an Mst1Ϫ/Ϫ thymus after3h(left and middle panels) and the numbers of emigrated CD4ϩ and CD8ϩ cells (right panel) are shown. Statistical analyses were performed using .p Ͻ 0.01. Ctrl, Control ,ءء ;p Ͻ 0.05 ,ء .Student’s t test. n ϭ 3 for A–D and n ϭ 4 for E

homologue Mst2 (42); cortex to medulla migration may appear Several possibilities may account for the different results ob- normal even though Mst1 is involved in regulating this process. tained in these two studies. First, these two alleles were generated Finally, Mst1-mediated thymic egress may be induced by several by completely different methods and were structurally different. alternative chemotactic cues. Given that Mst1Ϫ/Ϫ thymocytes are The gene trap allele carries a trapping vector (pGT0lxr; www. defective in chemotactic responses to several chemokines in addi- sanger.ac.uk/PostGenomics/genetrap/) between the first and the tion to CCL19, it remains to be determined which chemokine and second exon without physical elimination of any part of the Mst1 chemokine receptor pair is directly involved in the activation of coding sequence. Although Western blot analysis clearly demon- MST1 during thymic egress. strated the loss of MST1 proteins in the spleen and thymus, one During the preparation of this manuscript, an Mst1 mutant cannot rule out the possibility of leaky expression that may be mouse strain derived from a gene trap ES clone (AJ0315) was undetectable by the method used. In contrast, our targeted allele reported to display a reduced number of peripheral T cells (27). was designed to remove the entire coding region of exon 4. This The reduction of peripheral lymphocytes is now recapitulated in deletion will also result in a frame shift truncation if splicing oc- the study of our Mst1 knockout mice. However, several pheno- curs between exon 3 and exon 5. Therefore, the WT MST1 protein types are clearly different between these two mutant alleles. First, cannot be produced from our Mst1 knockout allele. Second, the lymphocyte reduction in the peripheral lymphoid organs was more discrepancy between the two studies could be partially due to the severe in our Mst1 knockout mice than in the gene trap mutant difference in genetic background between the C57BL/6 and 129Sv mouse. Second, a significant increase in the number of SP thymo- mouse strains. Our Mst1-deficient mice are pure 129Sv, whereas cytes was only observed in our Mst1 knockout mice. A small but the gene trap mutant mice were progeny backcrossed with insignificant increase in SP thymocytes was mentioned in the study C57BL/6 for six generations. However, this background difference of the gene trap mice without additional information. Third, the study alone is not sufficient to explain the discrepancy between the two of Mst1 gene trap mice revealed a significant enhancement of apo- mutant strains, because we also observed the developmental defect ptosis among freshly isolated peripheral CD3ϩ T cells (27). In con- among mice bred to 129Sv and C57BL/6 mixed background in the trast, analysis of our Mst1 knockout allele only showed a marginal case of T lineage-specific knockout of the Mst1 gene. enhancement of apoptosis among the activated (CD44high) CD4ϩ T Although Mst1 is broadly expressed, Mst1 disruption only cells freshly isolated from lymph nodes. results in a relatively restricted defect in lymphocyte migration 3872 Mst1 REGULATES THYMOCYTE EMIGRATION and egress. Thus, the MST1 kinase and the factors in its sig- 19. Udan, R. S., M. Kango-Singh, R. Nolo, C. Tao, and G. Halder. 2003. Hippo naling pathway represent attractive targets for the development promotes proliferation arrest and apoptosis in the Salvador/Warts pathway. Nat. Cell Biol. 5: 914–920. of therapeutic drugs for immune suppression. The genetic sys- 20. Ryoo, H. D., and H. Steller. 2003. Hippo and its mission for growth control. Nat. tem established in this study provides a critical tool for further Cell Biol. 5: 853–855. delineation of the Mst1 signaling pathway relevant to thymo- 21. Emoto, K., J. Z. Parrish, L. Y. Jan, and Y. N. Jan. 2006. The tumour suppressor Hippo acts with the NDR kinases in dendritic tiling and maintenance. Nature 443: cyte and lymphocyte migration and the development of thera- 210–213. peutic strategies. 22. Lehtinen, M. K., Z. Yuan, P. R. Boag, Y. Yang, J. Villen, E. B. Becker, During the revision of this manuscript, Katagiri, et al. reported S. DiBacco, N. de la Iglesia, S. Gygi, T. K. Blackwell, and A. Bonni. 2006. A conserved MST-FOXO signaling pathway mediates oxidative-stress responses similar findings (43). and extends life span. Cell 125: 987–1001. 23. Cheung, W. L., K. Ajiro, K. Samejima, M. Kloc, P. Cheung, C. A. Mizzen, Acknowledgments A. Beeser, L. D. Etkin, J. Chernoff, W. C. Earnshaw, and C. D. Allis. 2003. Apoptotic phosphorylation of histone H2B is mediated by mammalian sterile We thank Shiyu Zhang, Boying Tan, Yanfeng Tan, Yanling Yang, and twenty kinase. Cell 113: 507–517. Fang Wang for technical assistance, Dr. Xiaohui Wu and Zengli Guo for 24. Glantschnig, H., G. A. Rodan, and A. A. Reszka. 2002. Mapping of MST1 kinase scientific discussion, and Xiaofeng Tao for the editing of English. sites of phosphorylation: activation and autophosphorylation. J. Biol. Chem. 277: 42987–42996. 25. Graves, J. D., K. E. Draves, Y. Gotoh, E. G. Krebs, and E. A. Clark. 2001. Both Disclosures phosphorylation and caspase-mediated cleavage contribute to regulation of the The authors have no financial conflict of interest. Ste20-like protein kinase Mst1 during CD95/Fas-induced apoptosis. J. Biol. Chem. 276: 14909–14915. References 26. Khokhlatchev, A., S. Rabizadeh, R. Xavier, M. Nedwidek, T. Chen, X. F. Zhang, B. Seed, and J. Avruch. 2002. Identification of a novel Ras-regulated proapop- Downloaded from 1. Takahama, Y. 2006. Journey through the thymus: stromal guides for T-cell de- totic pathway. Curr. Biol. 12: 253–265. velopment and selection. Nat. Rev. Immunol. 6: 127–135. 27. Zhou, D., B. D. Medoff, L. Chen, L. Li, X. F. Zhang, M. Praskova, M. Liu, 2. Forster, R., A. C. Davalos-Misslitz, and A. Rot. 2008. CCR7 and its ligands: A. Landry, R. S. Blumberg, V. A. Boussiotis, et al. 2008. The Nore1B/Mst1 balancing immunity and tolerance. Nat. Rev. Immunol. 8: 362–371. complex restrains antigen receptor-induced proliferation of naive T cells. Proc. 3. Misslitz, A., G. Bernhardt, and R. Forster. 2006. Trafficking on serpentines: mo- Natl. Acad. Sci. USA 105: 20321–20326. lecular insight on how maturating T cells find their winding paths in the thymus. 28. Sun, L., X. Wu, M. Han, T. Xu, and Y. Zhuang. 2008. A mitotic recombination Immunol. Rev. 209: 115–128. system for mouse chromosome 17. Proc. Natl. Acad. Sci. USA 105: 4237–4241. 4. Ueno, T., F. Saito, D. H. Gray, S. Kuse, K. Hieshima, H. Nakano, T. Kakiuchi, M. Lipp, R. L. Boyd, and Y. Takahama. 2004. CCR7 signals are essential for 29. Pan, L., J. Hanrahan, J. Li, L. P. Hale, and Y. Zhuang. 2002. An analysis of T cell http://www.jimmunol.org/ cortex-medulla migration of developing thymocytes. J. Exp. Med. 200: 493–505. intrinsic roles of E2A by conditional gene disruption in the thymus. J. Immunol. 5. Ueno, T., K. Hara, M. S. Willis, M. A. Malin, U. E. Hopken, D. H. Gray, 168: 3923–3932. K. Matsushima, M. Lipp, T. A. Springer, R. L. Boyd, et al. 2002. Role for CCR7 30. Callus, B. A., A. M. Verhagen, and D. L. Vaux. 2006. Association of mammalian ligands in the emigration of newly generated T lymphocytes from the neonatal sterile twenty kinases, Mst1 and Mst2, with hSalvador via C-terminal coiled-coil thymus. Immunity 16: 205–218. domains, leads to its stabilization and phosphorylation. FEBS J. 273: 4264–4276. 6. Allende, M. L., J. L. Dreier, S. Mandala, and R. L. Proia. 2004. Expression of the 31. Scollay, R. G., E. C. Butcher, and I. L. Weissman. 1980. Thymus cell migration: sphingosine 1-phosphate receptor, S1P1, on T-cells controls thymic emigration. quantitative aspects of cellular traffic from the thymus to the periphery in mice. J. Biol. Chem. 279: 15396–15401. Eur. J. Immunol. 10: 210–218. 7. Matloubian, M., C. G. Lo, G. Cinamon, M. J. Lesneski, Y. Xu, V. Brinkmann, 32. Berlin-Rufenach, C., F. Otto, M. Mathies, J. Westermann, M. J. Owen, M. L. Allende, R. L. Proia, and J. G. Cyster. 2004. Lymphocyte egress from A. Hamann, and N. Hogg. 1999. Lymphocyte migration in lymphocyte function- thymus and peripheral lymphoid organs is dependent on S1P receptor 1. Nature associated antigen (LFA)-1-deficient mice. J. Exp. Med. 189: 1467–1478. 427: 355–360. 33. Ngo, V. N., H. L. Tang, and J. G. Cyster. 1998. Epstein-Barr virus-induced by guest on September 27, 2021 8. Spiegel, S., and S. Milstien. 2003. Sphingosine-1-phosphate: an enigmatic sig- molecule 1 ligand chemokine is expressed by dendritic cells in lymphoid tissues nalling lipid. Nat. Rev. Mol. Cell Biol. 4: 397–407. and strongly attracts naive T cells and activated B cells. J. Exp. Med. 188: 9. Durand, C. A., J. Westendorf, K. W. Tse, and M. R. Gold. 2006. The Rap 181–191. GTPases mediate CXCL13- and sphingosine1-phosphate-induced chemotaxis, 34. Kim, C. H., L. M. Pelus, J. R. White, and H. E. Broxmeyer. 1998. Differential adhesion, and Pyk2 tyrosine phosphorylation in B lymphocytes. Eur. J. Immunol. chemotactic behavior of developing T cells in response to thymic chemokines. 36: 2235–2249. Blood 91: 4434–4443. 10. Nombela-Arrieta, C., T. R. Mempel, S. F. Soriano, I. Mazo, M. P. Wymann, 35. Shulman, Z., R. Pasvolsky, E. Woolf, V. Grabovsky, S. W. Feigelson, N. Erez, E. Hirsch, A. C. Martinez, Y. Fukui, U. H. von Andrian, and J. V. Stein. 2007. Y. Fukui, and R. Alon. 2006. DOCK2 regulates chemokine-triggered lateral lym- A central role for DOCK2 during interstitial lymphocyte motility and sphin- phocyte motility but not transendothelial migration. Blood 108: 2150–2158. gosine-1-phosphate-mediated egress. J. Exp. Med. 204: 497–510. 36. Swain, S. L., M. Croft, C. Dubey, L. Haynes, P. Rogers, X. Zhang, and 11. Sakata, D., H. Taniguchi, S. Yasuda, A. Adachi-Morishima, Y. Hamazaki, L. M. Bradley. 1996. From naive to memory T cells. Immunol. Rev. 150: R. Nakayama, T. Miki, N. Minato, and S. Narumiya. 2007. Impaired T lympho- 143–167. cyte trafficking in mice deficient in an actin-nucleating protein, mDia1. J. Exp. 37. von Andrian, U. H., and T. R. Mempel. 2003. Homing and cellular traffic in Med. 204: 2031–2038. lymph nodes. Nat. Rev. Immunol. 3: 867–878. 12. Katagiri, K., A. Maeda, M. Shimonaka, and T. Kinashi. 2003. RAPL, a Rap1- 38. Weinreich, M. A., and K. A. Hogquist. 2008. Thymic emigration: when and how binding molecule that mediates Rap1-induced adhesion through spatial regulation T cells leave home. J. Immunol. 181: 2265–2270. of LFA-1. Nat. Immunol. 4: 741–748. 13. Katagiri, K., N. Ohnishi, K. Kabashima, T. Iyoda, N. Takeda, Y. Shinkai, 39. Kurobe, H., C. Liu, T. Ueno, F. Saito, I. Ohigashi, N. Seach, R. Arakaki, K. Inaba, and T. Kinashi. 2004. Crucial functions of the Rap1 effector molecule Y. Hayashi, T. Kitagawa, M. Lipp, et al. 2006. CCR7-dependent cortex-to-me- RAPL in lymphocyte and dendritic cell trafficking. Nat. Immunol. 5: 1045–1051. dulla migration of positively selected thymocytes is essential for establishing 14. Katagiri, K., M. Imamura, and T. Kinashi. 2006. Spatiotemporal regulation of the central tolerance. Immunity 24: 165–177. kinase Mst1 by binding protein RAPL is critical for lymphocyte polarity and 40. Misslitz, A., O. Pabst, G. Hintzen, L. Ohl, E. Kremmer, H. T. Petrie, and adhesion. Nat. Immunol. 7: 919–928. R. Forster. 2004. Thymic T cell development and progenitor localization depend 15. Zeng, Q., and W. Hong. 2008. The emerging role of the hippo pathway in cell on CCR7. J. Exp. Med. 200: 481–491. contact inhibition, organ size control, and cancer development in mammals. Can- 41. Vianello, F., P. Kraft, Y. T. Mok, W. K. Hart, N. White, and M. C. Poznansky. cer Cell 13: 188–192. 2005. A CXCR4-dependent chemorepellent signal contributes to the emigration 16. Pan, D. 2007. Hippo signaling in organ size control. Genes Dev. 21: 886–897. of mature single-positive CD4 cells from the fetal thymus. J. Immunol. 175: 17. Harvey, K. F., C. M. Pfleger, and I. K. Hariharan. 2003. The Drosophila Mst 5115–5125. ortholog, hippo, restricts growth and cell proliferation and promotes apoptosis. 42. Ling, P., T. J. Lu, C. J. Yuan, and M. D. Lai. 2008. Biosignaling of mammalian Cell 114: 457–467. Ste20-related kinases. Cell. Signal. 20: 1237–1247. 18. Wu, S., J. Huang, J. Dong, and D. Pan. 2003. hippo encodes a Ste-20 family 43. Katagiri, K., T. Katakai, Y. Ebisuno, Y. Ueda, T. Okada, and T. Kinashi. 2009. protein kinase that restricts cell proliferation and promotes apoptosis in conjunc- Mst1 controls lymphocyte trafficking and interstitial motility within lymph nodes. tion with salvador and warts. Cell 114: 445–456. EMBO J. 28: 1319–1331.