CD155 Is Involved in Negative Selection and Is Required to Retain Terminally Maturing CD8 T Cells in Thymus

CD155 Is Involved in Negative Selection and Is Required To Retain Terminally Maturing CD8 T Cells in Thymus

This information is current as Quan Qiu, Inga Ravens, Sebastian Seth, Anchana of October 1, 2021. Rathinasamy, Michael K. Maier, Ana Davalos-Misslitz, Reinhold Forster and Günter Bernhardt J Immunol 2010; 184:1681-1689; Prepublished online 4 January 2010; doi: 10.4049/jimmunol.0900062 http://www.jimmunol.org/content/184/4/1681 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2010/01/04/jimmunol.090006

Material 2.DC1 http://www.jimmunol.org/ References This article cites 56 articles, 25 of which you can access for free at: http://www.jimmunol.org/content/184/4/1681.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on October 1, 2021

• No Triage! Every submission reviewed by practicing scientists

• 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 © 2010 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

CD155 Is Involved in Negative Selection and Is Required To Retain Terminally Maturing CD8 T Cells in Thymus

Quan Qiu, Inga Ravens, Sebastian Seth, Anchana Rathinasamy, Michael K. Maier, Ana Davalos-Misslitz, Reinhold Forster, and Gu¨nter Bernhardt

During their final maturation in the medulla, semimature single-positive (SP) thymocytes downregulate activation markers and subsequently exit into the periphery. Although semimature CD4+ SP cells are sensitive to negative selection, the timing of when negative selection occurs in the CD8 lineage remains elusive. We show that the abundance of terminally matured CD8+ SP cells in adult thymus is modulated by the genetic background. Moreover, in BALB/c mice, the frequency of terminally matured CD8+ SP cells, but not that of CD4+ SP cells present in thymus, varies depending on age. In mice lacking expression of the adhesion receptor CD155, a selective deficiency of mature CD8+ SP thymocytes was observed, emerging first in adolescent animals at the age when these cells start to accumulate in wild-type thymus. Evidence is provided that the mature cells emigrate prematurely when CD155 is absent, cutting Downloaded from short their retention time in the medulla. Moreover, in nonmanipulated wild-type mice, semimature CD8+ SP thymocytes are subjected to negative selection, as reflected by the diverging TCR repertoires present on semimature and mature CD8+ T cells. In CD155-deficient animals, a shift was found in the TCR repertoire displayed by the pool of CD8+ SP cells, demonstrating that CD155 is involved in negative selection. The Journal of Immunology, 2010, 184: 1681–1689.

mong hematopoietic cells, T cells are unique because they CD28, CD40, CD43, and CD152), as well as the chemokine receptor accomplish their development in thymus and not in bone CCR7 are capable of participating and cooperate in negative se- http://www.jimmunol.org/ A marrow. To achieve this, thymic precursors enter the organ lection (8–11). In contrast, little is known about the involvement of at the cortico-medullary junction and start a differentiation program adhesion receptors in this process, although evidence exists that (1). This is connected to an intrathymic migration path guiding the functional interference with integrins/ligands or cadherins causes developing cells through the cortex and subsequently into the me- a perturbed genesis of thymocytes at several stages of their de- dulla (2). The early cortical steps encompass the double-negative velopment (12–14). During their final maturation, SP cells down- stages 1–4 prior to the onset of CD8 expression, giving rise to im- regulate activation markers, such as CD69 and CD24, and mature single-positive (SP) cells (1). Shortly thereafter, CD4 ex- upregulate CD62L (15, 16), rendering cells competent for emigra- pression commences, yielding double-positive (DP) thymocytes tion. Egress of matured naive SP cells crucially depends on the by guest on October 1, 2021 that are subjected to positive selection (3). Cells successfully expression of the sphingosine 1-phosphate (S1P) receptor S1P1. + + passing positive selection develop into semimature CD4 or CD8 Cells lacking S1P1 or treated with FTY720, a drug interfering with SP cells, a step that is believed to be associated with their transit into S1P receptor-mediated activity, are blocked in emigration (17–19). the medulla. Although negative selection eliminating self-reactive The adhesion receptor CD155 was originally identified as the SP cells by apoptosis can take place to a certain extent in the cortex cellular receptor for poliovirus, representing the major factor de- (4), it is generally assumed that these processes are mediated by termining the host and tissue tropism of the virus (20). CD155 be- medullary epithelial cells (mTECs) as well as dendritic cells (DCs) longs to the nectins, a family of Ig-like transmembrane proteins amply present in the medulla (5, 6). Particularly DCs, but also (21). All known physiological and pathogenic functions of CD155 mTECs, express costimulatory molecules as well as a broad array of are mediated via its outermost V-like domain harboring the docking self-Ags required to eliminate self-reactive T cells (5, 7). According epitopes for its known ligands: nectin-3, vitronectin, CD226, CD96, to a redundancy model, many costimulatory molecules (e.g., CD5, and poliovirus (22–29). CD155 is involved in the establishment of adherens junctions between endothelial and epithelial cells (21) and was shown to influence cell motility, migration, and proliferation Institute of Immunology, Hannover Medical School, Hannover, Germany (30, 31). In addition, CD155 is overexpressed in several tumor types, Received for publication January 9, 2009. Accepted for publication December 2, an observation that explains the susceptibility of tumors to NK- 2009. driven eradication along the CD226/CD155 axis (32–34). However, This work was supported by Grants BE1886/2-1/-2 from the Deutsche Forschungs- the latter interaction is also relevant for NK-mediated elimination of gemeinschaft. Q.Q. was supported by the Wilhelm Hirte Foundation. S.S. and A.R. were awarded a Georg Christoph Lichtenberg scholarship provided by the Ph.D. DCs; thus, it may contribute to shape the DC pool that is able to Program in Infection Biology sponsored by the Ministry of Science and Culture of induce T cell responses (35, 36). Moreover, we demonstrated re- Lower Saxony. cently that CD155 is implicated in the generation of a humoral Address correspondence and reprint requests to Dr. Gu¨nter Bernhardt, Institute of immune response upon oral ingestion of Ags (37). This may be Immunology, Hannover Medical School, Carl-Neuberg Str. 1, 30625 Hannover, Ger- many. E-mail address: [email protected] explained by the finding that CD155-deficient mice harbor reduced The online version of this article contains supplemental material. numbers of follicular helper T cells in their Peyer’s patches (38). Abbreviations used in this paper: DC, dendritic cell; DP, double-positive; ISP, im- Inthisstudy,weshowthatthelackofCD155provokesashiftinthe + mature single-positive; KO, knockout; mLN, mesenteric lymph node; mTEC, med- TCRrepertoire displayed byCD8 SPthymocytes.Wealso observed ullary epithelial cell; pLN, peripheral lymph node; RTE, recent thymic emigrant; a significant reduction in the number of fully matured CD8+ SP cells S1P, sphingosine 1-phosphate; SP, single-positive; WT, wild-type. in adult thymi, whereas the frequency of semimature CD8+ SP cells + Copyright Ó 2010 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00 is largely unaffected. Interestingly, mature CD8 SP cells seem to www.jimmunol.org/cgi/doi/10.4049/jimmunol.0900062 1682 FUNCTION OF CD155 IN DEVELOPMENT OF CD8 THYMOCYTES accumulate in an age-dependent fashion in wild-type (WT) BALB/c differentiated cells by applying a lineage mixture of rat anti-mouse Abs thymus, offering an explanation for why the paucity of CD8+ SP recognizing Ter119, CD4 (RMCD4-2), CD8b (RMCD8-2), CD11b (M1/70, cells triggered by CD155 deficiency starts to emerge in adolescent eBioscience), CD19 (1D3), DX5 (Biolegend), and GR1, followed by in- cubation with sheep anti-rat IgG coupled to Dynabeads (Invitrogen). mice. Because we failed to observe defects in maturation and the Magnetic separation was performed according to the manufacturer’s in- degree of emigration or apoptosis regarding CD155-deficient CD8+ structions. Lineage negative cells were concentrated in PBS to allow in- SP thymocytes, our findings suggest that newly matured CD8+ trahepatic injection of 2 3 105 cells in a total of 25 ml. Mice were analyzed thymocytes cannot be retained in the medulla when CD155 ex- 8–10 wk following the bone marrow transfer. All animals exerted normal behavior, yet they were distinguished by a substantially reduced weight pression is missing. This demonstrates that the adhesive mecha- compared with age-matched controls. nisms keeping SP cells inside the medulla differ between CD4+ and CD8+ SP cells, semimature and mature CD8+ SP cells, and mature Real-time PCR analysis + CD8 SP cells produced in young versus adult mice. Cells were prepared from thymi pooled from five WT or CD155-deficient mice. CD8 SP cells were enriched by negative selection, removing CD4+ Materials and Methods cells via anti CD4-biotin/M-280 Streptavidin Dynabeads (Invitrogen). Then staining with anti CD8a-Cy5/CD24-PECy7 was done, followed by Mice flow cytometric sorting into CD8+CD24hi and CD8+CD24lo fractions. 3 5 WT BALB/c (H2d) and C57BL/6 (H2b) mice were purchased from Charles Successful separation was controlled by reanalysis. Up to 5 10 cells River Laboratories (Sulzfeld, Germany) or bred in Hannover Medical were used for RNA isolation (Absolutely RNA Microprep kit, Stratagene, School’s animal facility. DBA/1 (H2q), C3H (H2k), and SJL (H2s) mice La Jolla, CA); subsequently, synthesis of the first strand of cDNA (Su- were purchased from Janvier (Le Genest St. Isle, France). CD155 knockout perscript II reverse transcriptase; Invitrogen) was done using random (KO) mice were described before (37). Mice were crossed onto the BALB/c hexamer primers. The expression of GAPDH and S1P1 was analyzed using background for at least six generations. All experiments including animals a Lightcycler 2.0 (Roche) and the Fast Start DNA Master plus SYBR Downloaded from were conducted according to the regulations of the local government and Green Kit (Roche) or the SYBR Premix Ex Taq Kit (Takara Shuzo, Otsu, institutional guidelines. Japan). The primers applied and the relative comparative quantification of expression levels was done based on GAPDH expression, as described Flow cytometry earlier, except that a cloned PCR product representing the GAPDH am- plicon was used for standardization (39). Single-cell suspensions of thymi were obtained by mincing the organs through a cell strainer dish. Cells were washed in FACS buffer (PBSd/2%

FBS), counted, and stained in 96-well plates, according to standard proce- Results http://www.jimmunol.org/ dures, in 50 ml reaction volume. After three final washes, DAPI was added to The number of CD8+ SP cells is severely reduced in thymus of exclude dead cells from analysis. The Annexin-V-Fluos kit (Roche, Basel, CD155-deficient mice Switzerland) was used to analyze apoptosis. The following Abs were used: Vb 2, 3, 4, 5, 6, 7, 8, 11, and 12 TCR (BD In our earlier analyses of CD155-deficient mice, we noticed regular Biosciences, San Jose, CA); CD24-FITC, CD69-PerCp-Cy5.5, CD45.2- numbers of peripheral T cell subsets in all secondary lymphoid PerCp-Cy5.5, Ly51, and SA-PerCP (all from BD Biosciences); CD25-Alexa organs investigated (37). Moreover, we also failed to observe any Fluor 488, CD44-Alexa Fluor 405, CD44-PE, CD44-APC, CD11b-PE, DX5- bio, abTCR, and SA-Pacific Orange (all from Invitrogen, Karlsruhe, Ger- differences in size or obvious anatomical and morphological aber- many); CD117-PE; (eBioscience, Frankfurt, Germany); EpCAM-PE (eBio- rations in the lymphoid organs (37). When investigating CD155- science); and CD24-PE-Cy7 (Biolegend, Uithoorn, The Netherlands). The deficient thymi of adult BALB/c mice (8–10 wk) by immune his- by guest on October 1, 2021 following homemade Abs were used: CD3-bio (17A2), CD4-bio (RMCD4- tology, regularly sized and distributed cortico-medullary areas were 2), CD8b-bio (RMCD8-2), CD19 (1D3), TER119, GR1, gdTCR-bio (GL3), CD62L (MEL-14), B220 (TIB146), CD155 (4G3), and CD226 (3B3); they detected (not shown). In addition, determination of overall thymic were either used directly labeled as indicated or detected with fluorochrome- cell counts failed to reveal significant differences compared with labeled standard secondary Abs. Flow cytometry was performed on an LSRII WT mice. However, when analyzing the cellular composition of flow cytometer (BD Biosciences); the data were evaluated using WinList 5.0. thymocytes by flow cytometry, we found a significant reduction in + Ab treatment of mice the frequency of CD8 SP cells in mutant thymi (Fig. 1A;WT: 5.08% 6 1.06%; KO: 3.23% 60.60%), which was paralleled by WT BALB/c mice were injected i.v. with 400 mg anti-CD155 mAb 4G3 (37) a slight increase in the percentage of DP cells (data not shown). In or an isotype control mAb every 5 d. The animals were sacrificed 28 d fol- + lowing the first injection. contrast, the frequency of CD4 SP cells remained unchanged (WT: 10.28% 6 1.17%; KO: 9.39% 6 1.50%). A considerable proportion FTY720 application of the flow cytometric determinations of CD8+ SP frequencies were Mice were administered sterile water or FTY720 (40 mg/mouse; Calbio- also done more stringently, including markers revealing other cell chem, San Diego, CA) in drinking water or by oral gavage every 3 d. Ten types that are present in the thymus, and therefore, may contribute to to 12 d later, mice were sacrificed and analyzed. Drug effectiveness was the observed imbalance. However, when gating was done to exclude confirmed by analyzing lymphopenia induction in the blood of treated hi + animals (data not shown). memory cells (CD44 ), regulatory T cells (CD25 ), TCRgd T cells (GL3+), and NK T cells (CD3+DX5+), the same degree of dis- Intrathymic FITC injection crepancy between WT and KO thymocyte numbers was found, + Animals were anesthetized, and the chest was disinfected with 70% ethanol. confirming that regular abTCR CD8 T cells are affected by CD155 An ∼5-mm longitudinal incision was made into the skin directly above the deficiency (data not shown). sternum, then the area was cleaned with 70% ethanol by wiping carefully We recently established a panel of anti-CD155 mAbs; two of with a lint-free cloth, followed by scanning the sternal/rib area for proper these (3F1 and 4G3) block the interaction of CD155 with its ligands injection (the thymus is located underneath the second rib, close to the sternum). A 27G needle was used to inject 10 ml 1 mg/ml FITC solution CD96 and CD226 [(37, 40) and data not shown]. Therefore, mAb (Sigma-Aldrich, St. Louis, MO) per lobe. The injection deepness was ∼5 4G3 was injected repeatedly into WT BALB/c mice for 4 wk, and mm. The skin incision was closed with surgical clips. The spleen and the thymi and the peripheral lymph nodes (PLNs) were analyzed. lymph nodes were harvested for analysis ∼40 h postinjection. mAb 4G3 was able to provoke a loss of CD8+ SP cells to an extent Bone marrow chimeras seen in CD155-deficient mice (Fig. 1B). The frequencies of peripheral T cells were not influenced by mAb treatment, demon- To enable the analysis of bone marrow chimeras at an age when thymus strating that T cells were not depleted. This suggested that the development culminates, newborn mice no older than 1 d were subjected to + two irradiations of 2.5 Gy each, separated by 4 h, followed by bone marrow observed deficiency in CD8 SP cell numbers can be triggered transfer. Bone marrow was harvested from donor mice and depleted for instantaneously by interrupting CD155 adhesion. The Journal of Immunology 1683

CD155 deficiency affects primarily mature CD8+ SP thymocytes We next sought to characterize the CD8+ SP thymocytes in more detail with regard to their maturation stage. It is known that SP cells downregulate CD69 and CD24 during their final maturation. CD24 staining revealed a characteristic distribution pattern among WT CD8+ SP cells. The narrow majority of cells displayed intermediate levels of CD24, representing semimature CD8+ SP cells. A considerable proportion of cells was CD24lo (mature CD8+ SP cells), whereas a minor fraction was CD24hi (Fig. 1C). The latter population is composed of immature SP cells (TCRab negative, data not shown), which develop into DP cells. When comparing this pattern with that obtained from CD155-deficient cells, it was evident that the lack of CD155 almost exclusively affects the most mature CD24lo cells because, to a large extent, these cells were missing in the KO thymi (Fig. 1C). The CD24-staining profiles of the CD4+ SP WTand mutant cells did not diverge significantly from each other, in accordance with the finding that the cell frequencies were identical. As mentioned above, injection of anti-CD155 mAb +

4G3 caused a deficiency of thymic CD8 SP cells. Analysis of the Downloaded from CD24 expression profile of mAb-treated CD8+ SP cells found that they displayed the same phenotype as seen in the classical KO cells, with the mature CD24lo cells almost completely missing in the pool of CD8+ SP thymocytes (data not shown). + Mature CD4 SP thymocytes upregulate their S1P1 expression, rendering them competent for emigration (19). Therefore, we in- http://www.jimmunol.org/ vestigated the expression of this receptor by real-time PCR in semimature and mature CD8 SP cells separately (Fig. 1C). In both, + WT and CD155 mutant CD8 SP S1P1 expression was virtually absent in semimature cells. A pronounced upregulation of S1P1 expression was observed in the mature subpopulation to an extent that did not differ between WT and CD155-deficient cells. This indicates that the lack of mature CD8+ SP cells in CD155-deficient thymi is not caused by an aberrantly regulated S1P1 expression. by guest on October 1, 2021 Abundance of CD8+ SP cells is strain dependent In general, C57BL/6 thymi harbor a lower percentage of CD8+ SP cells compared with the BALB/c background, indicating that the abundance of SP cells varies among different mouse strains. When analyzing the cellular composition of the CD8+ SP cell pool, it was evident that only a small proportion of mature CD8+CD24lo cells is present in adult thymus of C57BL/6 mice (Fig. 2A). We selected additional inbred mouse strains of differing MHC class I haplotypes (see Materials and Methods) and investigated the composition of their CD8+ SP cell populations in more detail. To this end, the expression analysis of the CD155 counter-receptor CD226 was in- FIGURE 1. Lack of CD155 causes a deficiency in mature CD8+ thy- cluded. CD226 is expressed in semimature and mature, but not mocytes. A, Flow cytometric analysis of thymocytes and peripheral T cells + 2/2 immature, CD8 SP cells, thus allowing a better discrimination of of WT and CD155 origin showing the separation according to CD4 and + CD8 expression. Dead cells (DAPI2) and doublets were excluded. For immature and semimature CD8 SP cells (36). In addition, CD226 + peripheral cells (pLNs), a pregate was set to include DAPI2CD62L+CD3+ expression declines in mature CD8 SP cells of BALB/c origin. By cells only. The lower panels depict the percentages of CD4+ SP and CD8+ and large, the investigated strains fall into two categories: DBA/1, SP cells. Each circle represents one animal. Closed circles represent WT SJL, and BALB/c mice possess considerable amounts of mature mice; open circles denote CD155-deficient mice. B, Distribution of SP cells CD8+ SP cells in their thymi as illustrated by their low ratio of and peripheral T cells in mice treated with anti-CD155 mAb. Closed circles semimature to mature CD8+ SP cells, whereas C3H and C57BL/6 and open circles represent untreated and mAb-treated mice, respectively. mice largely lack that cell type (Fig. 2B,2C). In addition, a char- ppp , + + p 0.001. C, Left panels: CD24 expression of CD8 SP and CD4 SP acteristic CD226 expression pattern could be found in each strain, cells from WT and CD155-deficient thymus as given in A. Shown are with SJL CD8+ SP cells expressing this CD155 counter-receptor at representative graphs as overlays of WT and KO cells. Right panel: Ex- a low level and DBA/1 mice exhibiting the biggest difference in pression of S1P1 was determined in FACS-sorted populations as indicated. + Expression is given in relation to GAPDH. Results from two independent CD226 expression between semimature and mature CD8 SP cells. experiments are shown; runs were done in duplicate. Error bars depict SD. This is an interesting observation considering that BALB/c thymocytes express slightly more CD155 than C57BL/6 thymocytes (data not shown). We analyzed heterozygous BALB/c animals to test whether the level of CD155 expression is critical for modulating the abundance of mature CD8+ SP cells. CD155 is expressed in 1684 FUNCTION OF CD155 IN DEVELOPMENT OF CD8 THYMOCYTES Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 2. Abundance of mature CD8+ thymocytes is strain dependent. A, Flow cytometric analyses of CD8+ SP cells are shown as in Fig. 1C for the strains as indicated. B, Refinement of the analysis seen in A including CD226. A and B show representative graphs and scatter plots, whereas the results obtained by gating according to B are summarized in C. The ratios of semimature/mature CD8+ SP cells are shown. Each dot represents one thymus. Het, heterozygous CD155+/2 BALB/c animals; ISP, immature SP. a gene dosage-dependent fashion, resulting in half-maximum mutant CD8+ SP cells remained remarkably constant over the entire CD155 levels on CD155+/2 cells (data not shown). However, study (i.e., not reproducing WT characteristics) (Fig. 3B). Until ∼3– CD155+/2CD8+ SP cells resembled their WT counterparts, whereas 4 wk of age, only a small percentage of the WT CD8+ SP cells are of the CD1552/2CD8+ SP cells of BALB/c origin mimicked the the mature (CD24lo) phenotype (Fig. 3C). Shortly thereafter, these CD57BL/6 WT phenotype with regard to the ratio of semimature to cells seem to accumulate in thymus, reaching a plateau ∼6–10 wk mature CD8+ SP cells (Fig. 2C). after birth, before their fraction starts to gradually decline again. In All data presented in the following sections refer to BALB/c mice. contrast, the separate analysis of CD24lo versus CD24int/hiCD8+ mutant SP cells revealed that, by and large, the percentage of mature + Paucity of mature CD8 SP cells is age dependent CD8+ SP cells remained at a low level, because the age-dependent Thymus represents a highly dynamic organ that grows tremendously accumulation of these cells does not take place when CD155 is in size after birth. At ∼3 wk of age, thymi have grown to their missing (Fig. 3C). maximal size, which remains constant during the following weeks 2/2 + until atrophy begins (41). When analyzing the emergence of SP cells Paucity of mature CD155 CD8 SP cells is triggered by in WT BALB/c thymus, we found that the frequency of CD4+ SP premature exit cells reached a plateau at ∼3 wk of age, concurring with thymic The selective absence of mature CD8+ SP cells in mutant thymi of growth dynamics (Fig. 3A). In contrast, expansion of the CD8+ SP BALB/c mice may be caused by a defective maturation, a pre- cell pool was delayed (peaking at ∼6–8 wk; Fig. 3B), confirming mature exit, or an increased apoptotic rate of these particular cells. earlier reports (41). Considering this, it was of interest to monitor the Determination of the latter did not reveal a significant difference progression of the observed CD8+ SP phenotype, caused by CD155 between WT and mutant thymocytes (Fig. 4), even though apo- deficiency, from suckling to advanced adulthood that is distin- ptotic rates seemed to be slightly elevated in the KO mice. guished by thymus involution. The frequencies of CD155-deficient To address thymic maturation and the exit of mutant CD8 T cells, CD4+ SP cells did not differ significantly from those of WT cells at animals were treated for 10–12 d with FTY720, an immunomodulatory any of the ages investigated (Fig. 3A). In contrast, the percentage of drug. FTY720 selectively interferes with S1P receptor activity, blocks The Journal of Immunology 1685

FIGURE 4. Apoptosis is not increased in CD155-deficient CD8+ SP thymocytes. Apoptotic CD8+ cells were identified by flow cytometry as Annexin-V+PI2 cells. Shown are representative scatter plots. Percentages 6 SD result from the analysis of four WT and four CD155-deficient mice.

pronounced compared with WT thymocytes (Fig. 6A), yet we failed to note signiﬁcant differences in the rates of RTEs in all secondary lymphoid organs analyzed. Although the numbers of RTEs are identical, the maturation stage may differ between WT and CD155- deﬁcient RTEs. To test whether the latter are more or less mature compared with their WT counterparts, we analyzed the expression

of CD44, CD62L, and CD24 on both types of RTEs. We observed Downloaded from that the levels of CD62L did not differ between RTEs and FITC2 peripheral CD8+ T cells, whereas RTEs had not yet completely downregulated CD44 and, in particular, CD24 to the levels found on the peripheral cells (Fig. 6B). This is in accordance with data published by other investigators (43, 44). More importantly, RTEs of 2/2

CD155 and WT origin displayed an identical proﬁle with regard http://www.jimmunol.org/ to the expression of CD44, CD62L, and CD24. We conclude that the rate of production of mature SP thymocytes, as well as their rate of emigration, is not affected by CD155 deﬁciency. by guest on October 1, 2021

FIGURE 3. The CD155 phenotype develops only with time in BALB/c mice. The frequencies of SP cells were determined by flow cytometry from thymi of various ages. A, CD4+ SP thymocytes. B, CD8+ SP thymocytes. C, The CD8+ SP pool was separated into CD24lo and CD24int/hi fractions. The number of thymi analyzed (n) is shown below each bar. pp , 0.05; pppp , 0.001. the exit of mature T cells from thymus and pLNs, and induces phenotypic maturation of semimature SP thymocytes (18, 19, 42). Upon FTY720 application, the CD8+ SP cell frequency became in- distinguishable between WT and mutant thymi, whereas in both cases, the overall number of SP cells increased considerably (Fig. 5A). This is most likely owed to the exit block triggered by the presence of FTY720. Subsequently, the profiles of expression of CD24, CD44, CD69, and CD62L were determined by flow cytometry (Fig. 5B and data not shown). The vast majority of SP cells in FTY720-treated thymi represent mature CD69loCD24loCD62hi cells. Most strikingly, WT and mutant cells gave rise to identical expression patterns of these maturation markers. In conjunction with the identical frequencies of CD8+ SP cells mentioned above, this indicates that CD155-deficient semimature SP thymocytes are able to mature and accumulate to an extent comparable to WT thymocytes. This suggests that the rates of SP cell FIGURE 5. FTY720 treatment abolished the differences in CD8+ SP production are similar, if not identical, in CD155-deficient thymi. composition caused by the lack of CD155. A, WT and CD155-deficient We then analyzed the frequency of recent thymic emigrants mice were fed FTY720 or PBSd as described in Materials and Methods. (RTEs) by injecting the fluorescent dye FITC into WTand KO thymi. The thymi were analyzed for their content of SP cells 10–12 d later. Each ThedyeunspecificallylabelsthymocytesthatcanbetrackedasRTEs circle represents one thymus. pppp , 0.001. B, SP cells were further 40 h later when analyzing secondary lymphoid organs. For unknown characterized with regard to their expression of CD24, CD69, and CD62L. reasons,thefrequencyoflabeledmutantthymocyteswasslightlyless Representative graphs are shown. 1686 FUNCTION OF CD155 IN DEVELOPMENT OF CD8 THYMOCYTES

cells). Therefore, it was of interest to investigate whether CD155 deficiency correlates with a shift in the TCR repertoire present in the pool of mutant SP cells. To this end, the expression rate of several TCRVb chains (Vb2, 3, 4, 5, 6, 7, 8, 11, and 12) was determined by flow cytometry in 8–10-wk-old mice (Fig. 7A). The panel of chains analyzed was also guided by the earlier finding that a profound al- teration in the frequency of diverse Vb subsets especially prone to Downloaded from http://www.jimmunol.org/

FIGURE 6. RTEs do not differ in frequency and phenotype between WT and CD155-deficient cells. A, FITC dye was injected intrathymically as described in Materials and Methods. The thymi, pLNs, and mLNs were analyzed 42 h later for CD8+ RTEs identified as FITC+ cells among un- labeled T cells. Each circle represents one animal. Closed circles: WT mice; open circles: CD155-deficient animals. B, Representative graphs show the expression pattern of CD62L, CD24, and CD44 in WT and CD155 deficient-CD8+ RTEs. The corresponding expression profile of by guest on October 1, 2021 peripheral naive CD8+ T cells that are FITC2 is also shown.

Taking into account that mutant CD8+ SP thymocytes did not differ from their WT counterparts, with regard to the critical pa- rameters of emigration, maturation, and sensitivity to apoptosis, the most likely explanation for the observed paucity of mature CD8+ SP cells in mutant thymus is a defect in the capacity to retain these cells in the medulla during their final maturation. This is in line with the finding that not only SP thymocytes themselves express CD155 and its ligands CD226 and CD96 [(37, 40) and data not shown]. In addition, mTECs and thymic DCs are positive for CD155 (data not shown), thus providing a dense CD155-driven adhesive network in the medulla. Shift in the TCR Vb repertoire used by CD8+ SP cells in the absence of CD155 The finding that CD4+ SP cells and semimature CD8+ SP cells are not affected in mutant thymi suggests that positive selection pro- ceeded normally in CD155-deficient cells. In general, positive selection determining lineage fate is assumed to precede negative selection. Passing through their final differentiation steps, thymocytes transit from the cortex to the medulla with semimature CD4+ FIGURE 7. Negative selection of CD8+ SP cells in WT and CD155- SP cells commencing negative selection (2, 6). However, doubts deficient mice. A, A series of Vb chain-specific mAbs was exploited to determine the repertoire of TCRs expressed by WT and CD155-mutant DP were raised regarding the general validity of this selection schedule + + (45). In addition, virtually nothing is known about at which stage(s) cells, CD4 SP cells, and CD8 SP cells. Each circle represents one thymus. B, Separately analyzed Vb repertoire in thymocytes of WT animals negative selection occurs in the CD8 lineage. Provided that CD155 according to their CD24 expression. This was compared with the corre- is involved in establishing or maintaining contacts with medullary sponding usage found in the CD8+CD62L+ fraction of mLN cells. A total + cells, this may influence the dwell time of CD8 SP cells as well as of eight mice were analyzed in two separate experiments. Statistical indirectly influence TCR-driven selection by decreasing the average analysis refers to the comparison of the CD24low fraction with the CD24int TCR-mediated signal intensity (possibly also in the case of CD4+ SP or mLN fraction. pp , 0.05; ppp , 0.01; pppp , 0.001. The Journal of Immunology 1687 recognizing endogenous retroviral Ags is indicative of defective is evident that the original differences in Vb3, 5, and 12 frequency negative selection in the CD4+ cell pool. In adult BALB/c animals, found were largely caused by the biased contributions of the semi- the strain-specific retroviral Ags cause massive deletion of CD4+ mature and mature cells to the pool of mutant CD8+ SP cells T cells expressing Vb3, 5, 11, and 12 (8, 9). When comparing the (Supplemental Fig. 2). However, this does not apply for Vb2, and TCR Vb repertoire between WT and mutant thymocytes, we ob- particularly not for Vb4. These discrepancies in the repertoire se- served moderate, but highly statistically significant, differences in lection between WTand KO CD8+ thymocytes must have come into the frequencies of Vb2, 3, 4, 5, and 12 present on CD8+ SP cells, existence before cells completed their transition to the mature stage. whereas only a marginal difference in Vb8 usage was found for This assumption is supported by another observation. As shown CD4+ SP cells. Likewise, the subchain composition in the DP above, following treatment with FTY720, the composition of the fraction was virtually identical, suggesting that, up to this stage, T CD8 SP compartment no longer differed between WT and CD155- cell development is regular in CD155-deficient mice. These findings deficient thymi (Fig. 5). When analyzing the frequencies of Vb2-, correlate well with the above-described selective deficiency in 3-, 4-, 5-, and 12-expressing T cells, we observed that FTY720 CD8+ SP cells, whereas CD4+ SP cells are largely unaffected; this application equalized the different frequencies in Vb3, 5, and 12 supports the idea that adhesion receptor-mediated dwell time and but not in Vb2 and 4 (Supplemental Fig. 3). Remarkably, how- negative selection represent coupled phenomena. ever, Vb4 was slightly increased by FTY720 treatment in KO Surprisingly, a different result was obtained when analyzing animals, whereas it was reduced in untreated mice compared with peripheral naive T cells (Supplemental Fig. 1). Some of the ob- the corresponding WT controls. Such a phenomenon may relate to served shifts displayed by the mutant thymocytes appear attenuated the finding that FTY720 can interfere with negative selection (46). in the periphery. CD8+ T cells isolated from mesenteric lymph Considering this, the shortened residency of mature CD8+ SP cells nodes (mLNs) resemble their thymic counterparts with respect to caused by CD155 deficiency and the concomitant bias in the rep- Downloaded from Vb4 but not Vb2, 3, and 12 (no difference). Interestingly, the ertoire selection may not always share a common origin (i.e., neg- increased presence of Vb5 on CD8+ SP thymocytes is opposed by ative selection and length of stay do not trace back to the same a reduced frequency of this chain among peripheral CD8+ T cells. adhesive events); thus, they may be mediated by different cell types contacted by the thymocytes. However, a defect in adhesion may Semimature and mature WT CD8+ SP cells and peripheral + account for both phenomena, and both seem to overlap chronolog-

b http://www.jimmunol.org/ CD8 T cells display divergent TCR V patterns, indicating an ically; the results presented above admit the interpretation that ongoing selection negative selection continues to shape the TCR repertoire to a certain In theory, the biases in the Vb repertoire selection observed in the extent beyond the semimature stage. Thus, a prolonged residency of CD155-deficient CD8+ SP pool may represent a numerical artifact CD8+ SP cells in medulla beginning with adolescence may influence caused by the highly divergent contributions of the semimature the TCR repertoire of naive CD8+ cells by increasing the probability CD24intCD8+ and mature CD24loCD8+ SP cell numbers in the KO of productive encounters between selectors and CD8+ thymocytes. scenario compared with the WT one. Such an assumption antici- pates that the TCR Vb distributions among semimature and mature CD8+ SP cells differ from each other, consequently Discussion by guest on October 1, 2021 postulating that negative selection operates predominantly in the When analyzing thymi of CD155-deficient mice, we found a sig- semimature or mature stage. Following the concept that semi- nificant reduction in the frequency of CD8+ SP but not CD4+ SP mature CD24intCD8+ SP cells are sensitive to negative selection, cells. Inclusion of maturation markers, such as CD24, CD69, or as suggested for the CD4 lineage (6), it would be more appropriate CD62L, in the flow cytometric analyses allowed assignment of the to compare the CD24int fractions when determining whether observed phenotype to the most mature CD8 cell population negative selection is impaired in thymus of CD155-deficient mice. present in thymus: CD24loCD69loCD62Lhi cells ready to exit into Likewise, the repertoire of the peripheral CD8+ T cells should be the periphery. A further refinement of the investigations of this cell compared with that of the mature CD24loCD8+ thymocytes. type including thymi of various ages revealed that the frequency Therefore, we analyzed the Vb chain repertoire of semimature profile of mature WT CD8+ SP cells is subject to significant CD24intCD8+ and mature CD24loCD8+ SP cells in WT animals variations, whereas that of semimature CD8+ SP cells remained separately. Unfortunately, a corresponding investigation in remarkably constant. Beginning at ∼3–4 wk after birth, mature CD155-deficient thymi was affected by the paucity of mature CD8+ SP cells accumulate in thymus, reaching a plateau at 6–10 cells, rendering such an analysis statistically insignificant for low- wk. In contrast, the frequencies of mature CD8+ SP cells remained frequency Vb chains in the mature compartment. The results invariantly low in CD155-deficient mice of all ages analyzed. obtained from WT cells sustain the supposition of ongoing se- Regarding these dynamics, the BALB/c CD1552/2 thymocytes lection in the semimature CD24intCD8+ SP stage, because statis- displayed characteristics resembling mature WT C57BL/6 CD8+ tically significant differences in the usage of Vb chains 3, 4, 5, 6, SP cells that failed to accumulate during adolescence (Fig. 2 and 7, and 11 were found compared with mature CD8+ SP cells (Fig. data not shown). Interestingly, the genetic background seems to 7B). In support of this, the frequency patterns of CD24loCD8+ SP exert a profound influence on the capability of mature CD8+ SP cells and peripheral naive CD8+ T cells resemble each other more cells to accumulate in thymus, but conclusive statements await closely than do those of CD24loCD8+ and CD24intCD8+ SP thy- more detailed analyses for various strains as done in this study for mocytes. This is in accordance with the above-mentioned model. BALB/c animals. A strain-dependent modulation of SP frequen- Nevertheless, the minor divergences in Vb usage found between cies and population dynamics (potentially also affecting CD4+ SP the mature CD24loCD8+ thymocytes and peripheral CD8+ T cells cells) represents a thus-far neglected parameter in the discussion (Vb2, 6, 7, 8, 11, and 12) indicate an ongoing shaping of the TCR of experimental results addressing, for example, the dwell time of repertoire finally engaged by peripheral naive CD8+ T cells. SP cells in thymus (see Ref. 47 for a concise overview). Therefore, However, it is uncertain whether such fine-tuning occurs in the considering this aspect may help to reconcile conflicting inter- pool of mature thymocytes and/or in the periphery. pretations of studies that used different mouse model strains. When comparing the WT CD24int fraction with the CD155-de- CD155 represents an adhesion receptor. Together with the results ficient CD8+ SP population that is dominated by the CD24int pool, it obtained from other experiments presented in this study (FITC 1688 FUNCTION OF CD155 IN DEVELOPMENT OF CD8 THYMOCYTES injection, FTY720 treatment, and determination of apoptosis rate), ception of Vb12 being slightly underrepresented on CD155- we hypothesize that matured CD8+ SP cells cannot be retained in deficient cells [Supplemental Fig. 5]). Moreover, the juvenile Vb the medulla of BALB/c thymi under CD155 deficiency, thereby repertoire is significantly different from the corresponding com- curtailing their dwell time in this compartment. In general, the position in adulthood. Thus, the output of RTEs differing in their length of time that CD8+ SP cells spend in the medulla may TCR signature from previous waves may slowly, but constantly, correlate positively with the efficacy/duration and the frequency of supersede the repertoire currently in use in the periphery (52). adhesion events that occur between themselves and other de- It is remarkable that CD155 deficiency almost exclusively affects veloping thymocytes, stroma cells, DCs, or extracellular matrix. adult CD8+ SP cells. Our observations illustrate that the de- Currently, we can provide only provisional evidence that the lack veloping thymocytes rely on different adhesion systems that de- of mature CD8+ SP cells is caused primarily by a thymocyte-born pend on the T cell subset as well as their developmental stage defect or by CD155 deficiency on DCs. The analyses of bone (semimature versus mature). Moreover, starting at ∼3–4 wk, marrow chimeric animals were obscured by the severe side effects a change in the adhesive settings causes an accumulation of ma- of irradiation; it caused a substantial loss of mature CD8+ SP cells ture CD8+ SP cells in thymus now critically engaging CD155. in the control group, which consisted of WT animals that had This event may correlate with birth. It is estimated that one round received WT bone marrow (Supplemental Fig. 4). In this study, we of thymocyte development requires ∼3 wk after the precursors observed a much lower frequency of CD8+ SP cells and a ratio of start their differentiation program (2). Filling of thymus with semimature/mature CD8+ SP cells of 5.66 6 0.43, which re- precursors occurs in waves (53), and it seems that the first wave of sembled that found in untreated BALB/c KO or BL6 mice (Fig. prothymocytes settling the organ perinatally follows a slightly 2C). Yet we noticed that the residual retention of the mature cells different path of maturation. This fits into a time frame that the in irradiated mice was due to the presence of CD155 on T cells majority of freshly produced mature T cells 3–4 wk later consists Downloaded from and/or DCs, because KO animals reconstituted with WT bone of progeny derived from the perinatal immigration wave. In- marrow displayed a similar ratio of semimature/mature CD8+ SP terestingly, at the same time, animals experience some profound cells (4.69 6 1.30). In contrast, in KO to WT reconstituted ani- immunological changes related to weaning. A reorientation of the mals, an even more pronounced deficiency of mature CD8+ SP immune system is required to accommodate this change in the cells (ratio, 8.34 6 0.70) was found, which approached the ratio individual’s exposure to exogenous burden regarding harmless seen in the KO/KO control group (11.19 6 1.54). Ags, such as food, as well as commensal microflora and patho- http://www.jimmunol.org/ By separately analyzing the TCR repertoire present on semi- gens. The end of weaning and the concomitant discontinuation of mature, mature CD8+ SP cells and naive peripheral CD8+ T cells, the protective maternal Ig shield coincide with structural alter- it was possible to document an ongoing selection process in oth- ations/maturation in GALT and remote lymphoid organs, a phe- erwise nonmanipulated mice. Even if we studied only a selected nomenon that was found to correlate with commensal microflora set of Vb chains, the comparison of their patterns would suggest now present in the gut (54–56). As outlined above, newly released that the main step in repertoire shaping is completed in the stage RTEs refresh the precedent peripheral pool of T cells, possibly of semimature CD24int CD8+ SP cells. This is illustrated by the facilitating to adopt to the requirement of changing immunologi- finding that mature CD8+ SP thymocytes and their peripheral cal challenges met during subsequent stages in life (52). There- by guest on October 1, 2021 counterparts diverged only modestly from each other with regard fore, it is tempting to speculate that the generation of CD8+ T cells to their TCR repertoire. We provided evidence that CD155 par- starting to settle the periphery at ∼3–4 wk must fulfill different ticipates in delaying the exit of mature CD8+ SP thymocytes and criteria that manifest as an extended residency in the medulla. influences TCR repertoire selection. As outlined above, these two Taken together, the discrepancies in repertoire selection as well phenotypes may be evoked by the same type of molecular adhe- as age-dependent dwell time of CD8+ SP thymocytes, but not sion events but in the context of different cellular interaction CD4+ SP thymocytes, related to CD155 function helped disclose partners. Whatever applies, the premature release of cells distin- hidden aspects of the life cycle of terminally maturing CD8+ SP guished by uncompleted thymic selection may cause problems in cells. To our knowledge, the provoked abnormalities due to that potentially autoreactive cells escape into the periphery. CD155 deficiency are not sufficient to elicit macroscopic pheno- However, such cells may be kept in check by regulatory T cells as types, such as overt signs of autoimmunity, even in old mice (data well as by an operative peripheral negative selection in a fashion not shown). However, final conclusions regarding these aspects resembling thymic self-education including Aire-expressing cells await more detailed analyses of CD155-deficient mice. (48). Particularly, the inverted frequencies of the Vb5 chain ex- + pressed by mature thymic versus peripheral CD8 T cells would Acknowledgments support such an assumption. Thus, there is evidence that the TCR We thank O. Pabst for critically reading the manuscript. repertoire is continuously surveyed and shaped at many stages inside thymus as well as in the periphery. The Ags required for presentation are made available locally or by transporting them Disclosures into thymus as well as by expressing them in the periphery on DCs The authors have no financial conflicts of interest. or in a manner comparable to mTECs (6, 48–51). However, a drift in the TCR repertoire in the pool of peripheral naive CD8+ T cells References may also be evoked by a biased self-renewal or a preferential 1. Hayday, A. C., and D. J. Pennington. 2007. Key factors in the organized chaos of supply of survival signals to cells displaying distinct TCRs. The early T cell development. Nat. Immunol. 8: 137–144. interpretation of the results is complicated further by the concept 2. Mißlitz, A., G. Bernhardt, and R. Fo¨rster. 2006. Trafficking on serpentines: molecular insight on how maturating T cells find their winding paths in the that thymocyte development may follow slightly different guide- thymus. Immunol. Rev. 209: 115–128. lines in adults compared with juvenile (,3–4-wk-old) mice; this 3. Singer, A., S. Adoro, and J. H. Park. 2008. Lineage fate and intense debate: may also affect the selective forces driving the adjustment of the myths, models and mechanisms of CD4- versus CD8-lineage choice. Nat. Rev. Immunol. 8: 788–801. TCR repertoire in use. Such an idea is supported by the fact that 4. McCaughtry, T. M., T. A. Baldwin, M. S. Wilken, and K. A. Hogquist. 2008. we failed to find significant differences when comparing WT and Clonal deletion of thymocytes can occur in the cortex with no involvement of the KO CD8+ SP cells isolated from 3–4-wk-old mice (with the ex- medulla. J. Exp. Med. 205: 2575–2584. The Journal of Immunology 1689

5. Hogquist, K. A., T. A. Baldwin, and S. C. Jameson. 2005. Central tolerance: 31. Sakamoto, Y., H. Ogita, H. Komura, and Y. Takai. 2008. Involvement of nectin in learning self-control in the thymus. Nat. Rev. Immunol. 5: 772–782. inactivation of integrin alpha(v)beta(3) after the establishment of cell-cell ad- 6. Sprent, J., and H. Kishimoto. 2002. The thymus and negative selection. Immunol. hesion. J. Biol. Chem. 283: 496–505. Rev. 185: 126–135. 32. Castriconi, R., A. Dondero, M. V. Corrias, E. Lanino, D. Pende, L. Moretta, 7. Anderson, M. S., E. S. Venanzi, Z. Chen, S. P. Berzins, C. Benoist, and C. Bottino, and A. Moretta. 2004. Natural killer cell-mediated killing of freshly D. Mathis. 2005. The cellular mechanism of Aire control of T cell tolerance. isolated neuroblastoma cells: critical role of DNAX accessory molecule-1-poliovirus Immunity 23: 227–239. receptor interaction. Cancer Res. 64: 9180–9184. 8. Davalos-Misslitz, A. C., T. Worbs, S. Willenzon, G. Bernhardt, and R. Fo¨rster. 33. Gilfillan, S., C. J. Chan, M. Cella, N. M. Haynes, A. S. Rapaport, K. S. Boles, 2007. Impaired responsiveness to T-cell receptor stimulation and defective D. M. Andrews, M. J. Smyth, and M. Colonna. 2008. DNAM-1 promotes acti- negative selection of thymocytes in CCR7-deficient mice. Blood 110: 4351– vation of cytotoxic lymphocytes by nonprofessional antigen-presenting cells and 4359. tumors. J. Exp. Med. 205: 2965–2973. 9. Foy, T. M., D. M. Page, T. J. Waldschmidt, A. Schoneveld, J. D. Laman, 34. Iguchi-Manaka, A., H. Kai, Y. Yamashita, K. Shibata, S. Tahara-Hanaoka, S. R. Masters, L. Tygrett, J. A. Ledbetter, A. Aruffo, E. Claassen, et al. 1995. An S. Honda, T. Yasui, H. Kikutani, K. Shibuya, and A. Shibuya. 2008. Accelerated essential role for gp39, the ligand for CD40, in thymic selection. J. Exp. Med. tumor growth in mice deficient in DNAM-1 receptor. J. Exp. Med. 205: 2959– 182: 1377–1388. 2964. 10. Kishimoto, H., and J. Sprent. 1999. Several different cell surface molecules 35. Pende, D., R. Castriconi, P. Romagnani, G. M. Spaggiari, S. Marcenaro, control negative selection of medullary thymocytes. J. Exp. Med. 190: 65–73. A. Dondero, E. Lazzeri, L. Lasagni, S. Martini, P. Rivera, et al. 2006. Expression 11. Takahashi, S., H. Kataoka, S. Hara, T. Yokosuka, K. Takase, S. Yamasaki, of the DNAM-1 ligands, Nectin-2 (CD112) and poliovirus receptor (CD155), on W. Kobayashi, Y. Saito, and T. Saito. 2005. In vivo overexpression of CTLA-4 dendritic cells: relevance for natural killer-dendritic cell interaction. Blood 107: suppresses lymphoproliferative diseases and thymic negative selection. Eur. J. 2030–2036. Immunol. 35: 399–407. 36. Seth, S., A. M. Georgoudaki, B. J. Chambers, Q. Qiu, E. Kremmer, M. K. Maier, 12. Schmeissner, P. J., H. Xie, L. B. Smilenov, F. Shu, and E. E. Marcantonio. 2001. N. Czeloth, I. Ravens, R. Foerster, and G. Bernhardt. 2009. Heterogeneous ex- Integrin functions play a key role in the differentiation of thymocytes in vivo. J. pression of the adhesion receptor CD226 on murine NK and T cells and its Immunol. 167: 3715–3724. function in NK-mediated killing of immature dendritic cells. J. Leukoc. Biol. 86: 13. Kutlesa, S., J. T. Wessels, A. Speiser, I. Steiert, C. A. Mu¨ller, and G. Klein. 2002. 91–101. E-cadherin-mediated interactions of thymic epithelial cells with CD103+ thy-

37. Maier, M. K., S. Seth, N. Czeloth, Q. Qiu, I. Ravens, E. Kremmer, M. Ebel, Downloaded from mocytes lead to enhanced thymocyte cell proliferation. J. Cell Sci. 115: 4505– W. Mu¨ller, O. Pabst, R. Fo¨rster, and G. Bernhardt. 2007. The adhesion receptor 4515. CD155 determines the magnitude of humoral immune responses against orally 14. Revilla, C., A. L. Gonza´lez, C. Conde, M. Lo´pez-Hoyos, and J. Merino. 1997. ingested antigens. Eur. J. Immunol. 37: 2214–2225. Treatment with anti-LFA-1 alpha monoclonal antibody selectively interferes 2 38. Seth, S., I. Ravens, E. Kremmer, M. K. Maier, U. Hadis, S. Hardtke, R. Fo¨rster, with the maturation of CD4 8+ thymocytes. Immunology 90: 550–556. and G. Bernhardt. 2009. Abundance of follicular helper T cells in Peyer’s 15. Ge, Q., and W. F. Chen. 1999. Phenotypic identiﬁcation of the subgroups of patches is modulated by CD155. Eur. J. Immunol. 39: 3160–3170. murine T-cell receptor alphabeta+ CD4+ CD82 thymocytes and its implication 39. Czeloth, N., G. Bernhardt, F. Hofmann, H. Genth, and R. Fo¨rster. 2005. in the late stage of thymocyte development. Immunology 97: 665–671. Sphingosine-1-phosphate mediates migration of mature dendritic cells. J. Im-

16. Kishimoto, H., and J. Sprent. 1997. Negative selection in the thymus includes http://www.jimmunol.org/ munol. 175: 2960–2967. semimature T cells. J. Exp. Med. 185: 263–271. 40. Seth, S., M. K. Maier, Q. Qiu, I. Ravens, E. Kremmer, R. Fo¨rster, and 17. Weinreich, M. A., and K. A. Hogquist. 2008. Thymic emigration: when and how G. Bernhardt. 2007. The murine pan T cell marker CD96 is an adhesion receptor T cells leave home. J. Immunol. 181: 2265–2270. for CD155 and nectin-1. Biochem. Biophys. Res. Commun. 364: 959–965. 18. Rosen, H., C. Alfonso, C. D. Surh, and M. G. McHeyzer-Williams. 2003. Rapid 41. Penit, C., and F. Vasseur. 1989. Cell proliferation and differentiation in the fetal induction of medullary thymocyte phenotypic maturation and egress inhibition and early postnatal mouse thymus. J. Immunol. 142: 3369–3377. by nanomolar sphingosine 1-phosphate receptor agonist. Proc. Natl. Acad. Sci. 42. Yagi, H., R. Kamba, K. Chiba, H. Soga, K. Yaguchi, M. Nakamura, and T. Itoh. USA 100: 10907–10912. 2000. Immunosuppressant FTY720 inhibits thymocyte emigration. Eur. J. Im- 19. Matloubian, M., C. G. Lo, G. Cinamon, M. J. Lesneski, Y. Xu, V. Brinkmann, munol. M. L. Allende, R. L. Proia, and J. G. Cyster. 2004. Lymphocyte egress from 30: 1435–1444. 43. Boursalian, T. E., J. Golob, D. M. Soper, C. J. Cooper, and P. J. Fink. 2004. thymus and peripheral lymphoid organs is dependent on S1P receptor 1. Nature 427: 355–360. Continued maturation of thymic emigrants in the periphery. Nat. Immunol. 5: 418–425. 20. Mendelsohn, C. L., E. Wimmer, and V. R. Racaniello. 1989. Cellular receptor for by guest on October 1, 2021 poliovirus: molecular cloning, nucleotide sequence, and expression of a new 44. Gabor, M. J., D. I. Godfrey, and R. Scollay. 1997. Recent thymic emigrants are member of the immunoglobulin superfamily. Cell 56: 855–865. distinct from most medullary thymocytes. Eur. J. Immunol. 27: 2010–2015. 21. Takai, Y., K. Irie, K. Shimizu, T. Sakisaka, and W. Ikeda. 2003. Nectins and 45. von Boehmer, H., and P. Kisielow. 2006. Negative selection of the T-cell rep- nectin-like molecules: roles in cell adhesion, migration, and polarization. Cancer ertoire: where and when does it occur? Immunol. Rev. 209: 284–289. Sci. 94: 655–667. 46. Shimizu, C., X. Li, M. Kimura, K. Hashimoto, K. Sugaya, M. Kubo, S. Suzuki, 22. Bernhardt, G., J. J. Harber, A. Zibert, M. deCrombrugghe, and E. Wimmer. 1994. and T. Nakayama. 1998. A novel immunosuppressant, FTY720, increases the The poliovirus receptor: identification of domains and amino acid residues efficiency of a superantigen-induced peripheral T-cell deletion whilst inhibiting critical for virus binding. Virology 203: 344–356. negative selection in the thymus. Immunology 94: 503–512. 23. Bottino, C., R. Castriconi, D. Pende, P. Rivera, M. Nanni, B. Carnemolla, 47. McCaughtry, T. M., M. S. Wilken, and K. A. Hogquist. 2007. Thymic emigration C. Cantoni, J. Grassi, S. Marcenaro, N. Reymond, et al. 2003. Identification of revisited. J. Exp. Med. 204: 2513–2520. PVR (CD155) and Nectin-2 (CD112) as cell surface ligands for the human 48. Gardner, J. M., J. J. Devoss, R. S. Friedman, D. J. Wong, Y. X. Tan, X. Zhou, DNAM-1 (CD226) activating molecule. J. Exp. Med. 198: 557–567. K. P. Johannes, M. A. Su, H. Y. Chang, M. F. Krummel, and M. S. Anderson. 24. Fuchs, A., M. Cella, E. Giurisato, A. S. Shaw, and M. Colonna. 2004. Cutting 2008. Deletional tolerance mediated by extrathymic Aire-expressing cells. Sci- edge: CD96 (tactile) promotes NK cell-target cell adhesion by interacting with ence 321: 843–847. the poliovirus receptor (CD155). J. Immunol. 172: 3994–3998. 49. Steinman, R. M., D. Hawiger, and M. C. Nussenzweig. 2003. Tolerogenic 25. Ikeda, W., S. Kakunaga, S. Itoh, T. Shingai, K. Takekuni, K. Satoh, Y. Inoue, dendritic cells. Annu. Rev. Immunol. 21: 685–711. A. Hamaguchi, K. Morimoto, M. Takeuchi, et al. 2003. Tage4/Nectin-like 50. Probst, H. C., K. McCoy, T. Okazaki, T. Honjo, and M. van den Broek. 2005. molecule-5 heterophilically trans-interacts with cell adhesion molecule Nectin-3 Resting dendritic cells induce peripheral CD8+ T cell tolerance through PD-1 and enhances cell migration. J. Biol. Chem. 278: 28167–28172. and CTLA-4. Nat. Immunol. 6: 280–286. 26. Lange, R., X. Peng, E. Wimmer, M. Lipp, and G. Bernhardt. 2001. The polio- 51. Bonasio, R., M. L. Scimone, P. Schaerli, N. Grabie, A. H. Lichtman, and virus receptor CD155 mediates cell-to-matrix contacts by specifically binding to U. H. von Andrian. 2006. Clonal deletion of thymocytes by circulating dendritic vitronectin. Virology 285: 218–227. cells homing to the thymus. Nat. Immunol. 7: 1092–1100. 27. Meyer, D., S. Seth, J. Albrecht, M. K. Maier, L. du Pasquier, I. Ravens, 52. Tanchot, C., and B. Rocha. 1997. Peripheral selection of T cell repertoires: the L. Dreyer, R. Burger, M. Gramatzki, R. Schwinzer, et al. 2009. CD96 interaction role of continuous thymus output. J. Exp. Med. 186: 1099–1106. with CD155 via its first Ig-like domain is modulated by alternative splicing or 53. Foss, D. L., E. Donskoy, and I. Goldschneider. 2001. The importation of he- mutations in distal Ig-like domains. J. Biol. Chem. 284: 2235–2244. matogenous precursors by the thymus is a gated phenomenon in normal adult 28. Mueller, S., and E. Wimmer. 2003. Recruitment of nectin-3 to cell-cell junctions mice. J. Exp. Med. 193: 365–374. through trans-heterophilic interaction with CD155, a vitronectin and poliovirus 54. Fo¨rster, R., O. Pabst, and G. Bernhardt. 2008. Homeostatic chemokines in de- receptor that localizes to alpha(v)beta3 integrin-containing membrane micro- velopment, plasticity, and functional organization of the intestinal immune domains. J. Biol. Chem. 278: 31251–31260. system. Semin. Immunol. 20: 171–180. 29. Ravens, I., S. Seth, R. Fo¨rster, and G. Bernhardt. 2003. Characterization and 55. Hooper, L. V. 2004. Bacterial contributions to mammalian gut development. identification of Tage4 as the murine orthologue of human poliovirus receptor/ Trends Microbiol. 12: 129–134. CD155. Biochem. Biophys. Res. Commun. 312: 1364–1371. 56. Mazmanian, S. K., C. H. Liu, A. O. Tzianabos, and D. L. Kasper. 2005. An 30. Ogita, H., and Y. Takai. 2006. Nectins and nectin-like molecules: roles in cell immunomodulatory molecule of symbiotic bacteria directs maturation of the adhesion, polarization, movement, and proliferation. IUBMB Life 58: 334–343. host immune system. Cell 122: 107–118.