B and T Lymphocyte Attenuator Is Highly Expressed on CMV-Specific T Cells during Infection and Regulates Their Function

This information is current as Nacer-Eddine Serriari, Françoise Gondois-Rey, Yves of September 28, 2021. Guillaume, Ester B. M. Remmerswaal, Sonia Pastor, Nassima Messal, Alemseged Truneh, Ivan Hirsch, René A. W. van Lier and Daniel Olive J Immunol 2010; 185:3140-3148; Prepublished online 6

August 2010; Downloaded from doi: 10.4049/jimmunol.0902487 http://www.jimmunol.org/content/185/6/3140

References This article cites 43 articles, 21 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/185/6/3140.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 28, 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 © 2010 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

B and T Lymphocyte Attenuator Is Highly Expressed on CMV-Specific T Cells during Infection and Regulates Their Function

Nacer-Eddine Serriari,* Franc¸oise Gondois-Rey,* Yves Guillaume,* Ester B. M. Remmerswaal,† Sonia Pastor,* Nassima Messal,* Alemseged Truneh,‡ Ivan Hirsch,* Rene´ A. W. van Lier,† and Daniel Olive*

B and T lymphocyte attenuator (BTLA), like its relative programmed cell death-1 (PD-1), is a that negatively regulates murine activation. However, its expression and function on human T cells is currently unknown. We report in this study on the expression of BTLA in human T cell subsets as well as its regulation on virus-specific T cells during primary human CMV infection. BTLA is expressed on human CD4+ T cells during different stages of differentiation, whereas on CD8+ T cells, it is found on naive Downloaded from T cells and is progressively downregulated in memory and differentiated effector-type cells. During primary CMV infection, BTLA was highly induced on CMV-specific CD8+ T cells immediately following their differentiation from naive cells. After control of CMV infection, BTLA expression went down on memory CD8+ cells. Engagement of BTLA by mAbs blocked CD3/CD28- mediated T cell proliferation and Th1 and Th2 cytokine secretion. Finally, in vitro blockade of the BTLA pathway augmented, as efficient as anti–PD-1 mAbs, allogeneic as well as CMV-specific CD8+ T cell proliferation. Thus, our results suggest that, like PD-1, BTLA provides a potential target for enhancing the functional capacity of CTLs in viral infections. The Journal of http://www.jimmunol.org/ Immunology, 2010, 185: 3140–3148.

mmune responses are controlled by cosignaling molecules and produce effector cytokines (4–10), raising the possibility that that regulate T lymphocyte activation either positively or neg- this pathway has been exploited by a variety of viruses to establish I atively. The B7/CD28 family of negative receptors include chronic infection. CTLA4, programmed cell death 1 (PD-1) (also called CD279), BTLA (CD272) is the most recently described member of the and a recently identified molecule known as B and T lymphocyte CD28 family. BTLA was first identified as a transcript highly attenuator (BTLA) (1, 2). In mice, PD-1 is found on activated, but specific to Th1 cells (11), but it was later shown to also be ex- by guest on September 28, 2021 not resting, T and B cells (3). PD-1 has two ligands called PD- pressed by thymocytes (12). BTLA interacts with herpes virus- (also called B7-H1 and CD274) and PD-L2 (also called B7-DC entry mediator (HVEM), a member of the TNFR family (13). The and CD273), which have distinct patterns of expression. Recently, interaction of BTLA, an Ig superfamily member, and HVEM, it has been reported that PD-1 is highly expressed on exhausted a costimulatory receptor homologous to TNFR, is quite unique murine T cells during chronic lymphocytic choriomeningitis virus in that it defines a cross talk between these two families of struc- (LCMV) infection and that it is upregulated on human HIV and turally distinct receptors (14). BTLA is expressed on a broad array HCV-specific CD8+ T cells. Blocking of PD-1/PD-1 inter- of murine immune cells, including B cells, CD4+ T cells, CD8+ actions reinvigorates exhausted T cells, allowing them to expand T cells, NK cells, macrophages, and dendritic cells (DCs). Like PD-1, BTLA contains a membrane proximal ITIM and mem- brane distal immunoreceptor -based switch motif (ITSM). *Institut National de la Sante´ et de la Recherche Me´dicale Unite´ Mixte de Recherche 891, Institut Paoli Calmettes, Universite´ de la Me´diterrane´e, Infrastructures en Biologie Disruption of either the ITIM or ITSM abrogated the ability of BTLA Sante et Agronomie Cancer Immunomonitoring Platform, Institut Fe´de´ratif de Re- to recruit either Src homology region 2 domain-containing phospha- † cherche 137, Marseille, France; Department of Experimental Immunology, Academic tase (SHP) 1 or SHP2 (15), suggesting that BTLA recruits SHP1 and Medical Center, Amsterdam, The Netherlands; and ‡Synaptex, Sudbury, MA 01776 SHP2 in a manner distinct from PD-1 and that both tyrosine motifs Received for publication July 30, 2009. Accepted for publication July 1, 2010. are required to block T cell activation. The human BTLA cytoplasmic This work was supported by Institut National de la Sante´ et de la Recherche Me´d- icale, Institut Paoli Calmettes, Groupement des Entreprises Francaises dans la Lutte region also contains two Grb-2 binding sites, YDN and YSN (16). Contre le Cancer, Infrastructures en Biologie Sante et Agronomie, and Association PD-1 and BTLA are both implicated in the negative regulation of pour la Recherche sur le Cancer. immune responses. PD-1, moreover, may not only to be involved in Address correspondence and reprint requests to Dr. Daniel Olive, Institut National de control, but also in immune exhaustion in la Sante´ et de la Recherche Me´dicale Unite´ Mixte de Recherche 891, Centre de Recherche en Cance´rologie de Marseille, 232 Boulevard de Sainte-Marguerite, Mar- chronic infections and cancer. The role of BTLA appears to be lim- seille, 13009, France. E-mail address: [email protected] ited to lymphocyte activation as shown in BTLA-deficient mice, Abbreviations used in this paper: aAPC, artificial APC; BTLA, B and T lymphocyte which have normal lymphoid organ development. However, these attenuator; CM, central memory; CsA, cyclosporin A; DC, ; EM, effector mice display hyperproliferative T cell responses to TCR-mediated memory; HVEM, herpes virus-entry mediator; iDC, immature dendritic cell; IE, im- mediate-early; ITSM, immunoreceptor tyrosine-based switch motif; LCMV, lympho- activation (11, 17). Furthermore, loss of BTLA function leads to cytic choriomeningitis virus; mDC, mature dendritic cell; mfi, median of fluorescence; increased susceptibility to experimental autoimmune encephalo- MHC I, MHC class I; PD-1, programmed cell death-1; SHP, Src homology region 2 myelitis and MHC-mismatched allograft rejection, supporting domain-containing phosphatase; TEMRA, effector memory CD45RA+ T cell. a role for BTLA in modulating T cell activation and effector Copyright Ó 2010 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00 responses (11, 18). In contrast to PD-1, BTLA is not upregulated www.jimmunol.org/cgi/doi/10.4049/jimmunol.0902487 The Journal of Immunology 3141 in the LCMV murine model of chronic infection, and conse- Alexa Fluor 488-conjugated anti-BTLA mAbs and a corresponding line- quently, it is not supposed to play a prominent role in the function age-specific mAb: anti-CD3–ECD (Beckman Coulter, Paris, France), anti- of antiviral-specific CD8 T cells, at least during chronic infection CD27–allophycocyanin-Alexa 750 (Invitrogen), anti-CD4–Pacific Blue, anti- CD8–Alexa 700, or anti-CD45RA–PE-Cy7 (all from Becton Dickinson). (19). Yet, its role in the specific human antiviral T cell responses Dead cells were eliminated using aqua LIVE/DEAD Fixable Dead Cell Stain has not been demonstrated. Kit (Invitrogen), and nonstudied cell populations were excluded from the Hence, we have performed a comprehensive study of human analysis by the use of anti-CD14 and anti-CD19–PC5 mAbs (Beckman BTLA expression on the subsets of CD4+ and CD8+ T cells and Coulter). Cells were washed twice in cold PBS with 2% FCS and 0.02% sodium azide, fixed in 4% paraformaldehyde, and data were acquired on an compared its expression to that of its relative PD-1. In addition, FACS LSR II SORP 4 lasers flow cytometer (BD Immunocytometry Systems, we have analyzed BTLA expression in the course of CMV-specific Le Pont De Claix, France) and analyzed with FlowJo (Tree Star, Ashland, CD8+ T cell differentiation. Finally, we have compared the effect OR). PBMCs from transplant recipients were washed in PBS containing of PD-1 and BTLA on T cell activation of T cell-specific alloge- 0.01% (w/v) NaN3 and 0.5% (w/v) BSA. One million PBMCs were incu- neic and CMV responses in vitro. bated with an appropriate concentration of tetrameric complexes in a small volume for 30 min at 4˚C, protected from . Fluorescence-labeled mAbs were then added and incubated for 30 min at 4˚C, protected from light, Materials and Methods at concentrations according to the manufacturer’s instructions. For surface Generation of anti-human, PD-1, and BTLA mAbs marker expression analysis, the following Abs were used: PD-1–PE, CD27- allophycocyanin–Alexa Fluor 750 (eBioscience, San Diego, CA), CD28- All mAbs were produced similarly. Female BALB/c mice were immunized Alexa Fluor 700 (Biolegend, San Diego, CA), CD8-PE–Alexa Fluor 610 by i.p. injection with 10 mg human Ig fusion protein with 250 ml Freund (Invitrogen), HLA-DR–PerCP-Cy5.5, and CD45RA–PE-Cy7 (BD Bioscien- adjuvant. Immunization was repeated three times at 2-wk intervals; the ces, San Jose, CA). Cells were washed and measured on an FACSCanto flow fourth immunization was made by i.v. injecting with 10 mg Ig fusion protein. cytometer (BD Biosciences) and analyzed with FlowJo software (Tree Star). Downloaded from Three days later, cells were fused with X63Ag8 myeloma cells with PEG 1500 (Roche, Basel, Switzerland) and cloned with HAT selection Artificial APCs and T cell assays (Sigma-Aldrich, St. Louis, MO) and hybridoma cloning factor (Origen, + + Igen, Rockville, MD). The hybridoma supernatants were screened by cell Human CD4 and CD8 T cells were purified by negative selection from surface staining of human PD-1 and BTLA-transfected COS cell lines, re- PBMCs using magnetic beads (Miltenyi Biotec, Paris, France) and were routinely .97% CD3+, .98% CD4+ for CD4+ T cell isolation, and .95% spectively, and for lack of reactivity with untransfected COS cells. Anti– + + PD-1 (PD1 3.1, IgG2b; PD1 6.4, IgG1) and anti-BTLA (BTLA 7.1, IgG2b; CD8 for CD8 T cell isolation as determined by flow cytometry. T cells http://www.jimmunol.org/ BTLA 8.2, IgG1) were selected as reagents for FACS analysis and func- were stimulated with artificial APCs (aAPCs) at a ratio of 3:1 (cells/beads) tional studies. In addition, these reagents have already been described else- comprised of magnetic beads (Dynabeads M-450 Epoxy, Dynal Biotech, where (20, 21). Invitrogen) coated with the following Abs: anti-CD3 (OKT3), anti-human CD28 (CD28.2), anti-human PD-1 (PD1 6.4), anti-human BTLA (BTLA Ethics statement 7.1), and anti-MHC class I (MHC I) (YJ4). As previously described (23), these aAPCs were coated with suboptimal anti-CD3 mAb (5%), subopti- All subjects gave written informed consent, and the medical ethics com- mal levels of anti-CD28 mAb (10%), and anti-MHC I mAb (CD3/28/MHC mittee of the Academic Medical Center, Amsterdam, The Netherlands, I), anti–PD-1 mAb (CD3/28/PD-1 plus MHC I), anti-BTLA mAb (CD3/ approved the study. 28/BTLA plus MHC I), or anti–PD-1 plus anti-BTLA (CD3/28/PD-1 plus BTLA), constituting the remaining 85% of protein added to the bead. Subjects T cells (1.5 3 105 cells/well) were stimulated in round-bottom 96-well by guest on September 28, 2021 plates. Supernatants were harvested at 48 h to analyze cytokine secretion Healthy volunteers were obtained from Marseille’s Blood Bank (Eta- with a human Th1/Th2 cytokine Kit (BD Cytometric Bead Array, BD blissement Francais du Sang, Marseille, France). Two CMV-seronegative Biosciences). Proliferation was measured at day 5 by tritiated thymidine renal transplant recipients of a CMV-seropositive were studied incorporation for the last 18 h. longitudinally. Heparinized PBMCs were isolated at regular intervals using standard density gradient centrifugation techniques and subsequently cry- Preparation of monocyte-derived DCs opreserved. Both recipients were treated with basic immunosuppressive therapy (cyclosporine A, prednisolone, and mycophenolate mofetil). Pa- PBMCs were obtained from healthy donors and isolated by Lymphoprep tient 1 also received CD25 mAb induction therapy. The HLA typing for (Abcys) density gradient centrifugation. CD14+ monocytes were then im- patient 1 was A1/11 B8/45 DR3/12 and patient 2 was A2/24 B51/39 DR1/ munomagnetically purified with CD14 mAb-conjugated microbeads (Mil- 11. Quantitative PCR for CMV was performed in EDTA whole-blood tenyi Biotec). Purity of the CD14+ cells by flow cytometry analysis samples, as described (22). To determine CMV serostatus, anti-CMV IgG was .98%. For generation of monocyte-derived DCs, CD14+ monocytes was measured in serum using the AxSYM microparticle enzyme immuno- were cultured for 5 d in six-well plates at 2 3 106 cells/well (Falcon; BD assay (Abbott Laboratories, Abbott Park, IL) according to the manufacturer’s Biosciences) in RPMI 1640 medium containing 10% FCS (Life Technol- instructions. Measurements were calibrated relative to a standard serum. ogies, Paisley, U.K.) and supplemented on days 0, 3, and 5 with 100 ng/ml GM-CSF and 20 ng/ml IL-4 (both cytokines from Abcys). Immature DCs Tetrameric complexes (iDCs) were harvested at day 5, and their maturation was accomplished The following HLA–peptide tetrameric complexes were used: HLA-A1 by coculturing them for 2 d with 50 Gy-irradiated CD40L-transfected cells 3 5 tetramer loaded with the CMV pp65-derived YSEHPTFTSQY peptide (2 10 /well). (CMV A1 pp65 YSE allophycocyanin), HLA-A2 tetramer loaded with the CMV pp65-derived NLVPMVATV peptide (CMV A2 pp65 NLV allo- Allogeneic stimulation of T cells with monocyte-derived DCs phycocyanin), the CMV immediate-early (IE)-derived VLEETSVML pep- T cells were separated from PBMCs by means of a Pan T isolation II tide (CMV A2 IE VLE allophycocyanin), HLA-B8 tetramer loaded with (Miltenyi Biotec) according to the manufacturer’s protocol. T cells were the CMV IE-derived QIKVRVDMV peptide (CMV B8 IE QIK allophy- then labeled with 0.5 mM CFSE (Invitrogen) for 10 min at 37˚C. The cocyanin), or the CMV IE-derived ELRRKMMYM peptide (CMV B8 IE 5 labeled T cells (2 3 10 cells/well) were washed and cultured in 96-well ELR allophycocyanin) (Sanquin, Amsterdam, The Netherlands). All used flat-bottom plates in RPMI 1640 medium containing 10% FCS in the tetramers were allophycocyanin conjugated. presence of blocking Fab to PD-1 (PD-1.3.1), BTLA (BTLA 8.2), or 4 Immunofluorescence staining and flow cytometry matched isotype Fab, with immature or mature allogeneic DCs (2 3 10 cells/well) in triplicate. After 5 d of culture, cells were stained with CD25- PBMCs were obtained from healthy donors and isolated by fraction- allophycocyanin, CD8-PC7, and CD4-PE. Data were acquired in a CAN- ation over Lymphoprep gradients (Abcys, Paris, France). mAbs to PD-1 TO-A (BD Immunocytometry Systems) and analyzed with FlowJo soft- (PD-1.3.1) and BTLA (BTLA7.1) were conjugated with Alexa Fluor 488 ware (Tree Star). CD25 expression was analyzed, and proliferation was or Alexa 647 using a commercially available kit (Invitrogen, Paisley, measured by quantification of CFSE dilution in the CD8+ T cell popula- U.K.). Anti–PD-1–allophycocyanin and CD272/BTLA-PE (Becton Dick- tion. To determine the production of cytokines, cell-free supernatants were inson, Le Pont De Claix, France) were used for comparison with the collected after 48 h or 96 h stimulation and assayed for IFN-g and TNF-a homemade conjugated Abs. Briefly, PBMCs were incubated with both by ELISA using OptEIA kits (BD Pharmingen, San Diego, CA) according the optimized dilution of the Alexa Fluor 647-conjugated anti–PD-1 and to the manufacturer’s instructions. 3142 BTLA AND T CELL-SPECIFIC hCMV RESPONSES

Induction of specific anti-pp65 CD8+ T cells In vitro, PD-1 is upregulated on activated CD4+ and CD8+ The iDCs and mature DCs (mDCs) were pulsed for 2 h at 37˚C in RPMI 1% T lymphocytes, and maximal expression is observed after 48 h (3, FCS with a 10 mg/ml CMV pp65 NLVPMVATV peptide. After two washes, 26, 27). To compare the expression profile of BTLA upon T cell peptide-pulsed DCs (2 3 104/well) were cultured with autologous CFSE- activation, purified CD4 and CD8 T cells were stimulated with 5 labeled T cells (2 3 10 /well) in the presence of anti–PD-1, anti-BTLA– plate-bound anti-CD3 (OKT3) and soluble anti-CD28 (CD28.2) blocking mAbs, or isotype control. On day 7, T cells were harvested and and stained every day with anti-BTLA mAbs. A progressive decrease restimulated with 10 mg/ml CMV pp65 NLVPMVATV peptide for 6 h. Cells + were stained with CMV-pp65-allophycocyanin and CD8-PC5. After gating in the intensity of BTLA expression was demonstrated on CD8 but + of CMVpp65+CD8+ T cells, proliferation was measured by CFSE dilution not CD4 T cells through 48 h (data not shown). and expressed as a division index calculated with the proliferation tool of FlowJo (Tree Star). For intracellular cytokine staining, cells were permeabi- BTLA and PD-1 expression during primary CMV infection lized using Cytofix/Cytoperm kit (Becton Dickinson) and stained with IFN-g– + PE (Becton Dickinson) mAb. We next followed the expression of BTLA on CD8 CMV-specific T cells during posttransplantation primary infection in two patients Statistical analysis using three distinct tetramers identifying responses against peptides All data were analyzed using GraphPad Prism version 5.00 for MacIntosh derived from pp65 and IE-1 of CMV. These three distinct tetramers (GraphPad, San Diego, CA). The Wilcoxon matched pairs test was used to together with CD8, CD45RA, and CD27 staining permitted us to examine the variations of PD-1 and BTLA expression on T lymphocyte analyze the differentiation of CMV-specific T cell populations dur- subsets of differentiation. Comparisons were made between either the pre- vious subset of differentiation according to a linear differentiation pathway ing the course of primary CMV infection. Viremia, monitored by (in Fig. 1), between CMV positive T CD8 lymphocytes and total T CD8 as levels of CMV DNA copies in blood, allowed us to identify acute indicated (in Fig. 2), or between different conditions of stimulation (in and controlled phase of CMV infection (Fig. 2). We have provided Downloaded from Figs. 3, 4). Differences were considered as statistically significant when data corresponding to the analyzed patients (Fig. 2B). In addition, , p 0.05. because of the low number of tetramer-positive cells, especially during the early stages of infection, the data obtained from both Results patients tested with three tetramers at several time points were Distinct expression of BTLA and PD-1 on different grouped according to the sampling performed during the acute or differentiation subsets of T lymphocytes the controlled phase of infection (Fig. 2C) and compared with the http://www.jimmunol.org/ T cell populations can be divided into several subsets based on the total CD8+ T cell population of patients at an early time postinfection. expression of CD45RA and CD27 (24, 25). The putative differ- The analysis was performed as depicted in Fig. 2A. As shown in entiation pathway proceeds from the naive (CD45RA+ CD27+) via Fig. 2B, viremia was detected during 9 wk in patient 1 and 2, the central memory (CM; CD45RA2CD27+) to the late effector although in this latter patient, it disappeared between wk 9 and 12 memory (EM) stage (CD45RA2CD272). A distinct subset of cells to reappear at wks 12–16. During the follow-up period postinfec- expressing a CD45RAhighCD272 phenotype is found predomi- tion, in both patients 1 and 2, the CMV-tetramer+ CD8 cells pro- nantly in the CD8 T cell population and is called the effector gressively moved from a naive and CM phenotype to become EM memory CD45RA+ (TEMRA) subset. To analyze the expression and TEMRA (Fig. 2B,2C, left panels). BTLA and PD-1 expres- of BTLA and PD-1 on CD8+ and CD4+ T cell subpopulations, sion were upregulated at early times postinfection and decreased by guest on September 28, 2021 seven-color FACS analyses were performed on six healthy PBMC at wk 25–30 (Fig. 2B). Interestingly, in patient 2, the decreased donors. The gating strategy allowed us to study BTLA and PD-1 viremia during wk 9–12 was associated with the concurrent de- expression and coexpression in three CD4+ T cell and four CD8+ crease expression of both BTLA and PD-1 as well as the activa- T cell subpopulations (Fig. 1A). The expression of PD-1 and tion marker HLA-DR on tetramer-positive cells. The expression BTLA within CD8+ and CD4+ T cell subpopulations showed dif- levels of BTLA and PD-1 during the acute phase of infection were ferent patterns (Fig. 1B–D). In CD4+ T cells, no difference in ex- significantly increased as compared with the total CD8 population pression of BTLA among differentiation subsets was seen, whereas in (p = 0.0007 and p = 0.05, respectively). In the controlled phase of CD8+ T cells, BTLA expression in the naive subset decreased sig- infection, BTLA decreased to reach lower expression levels sig- nificantly in CM and EM subsets (p =0.03)andisevenlowerin nificantly different from the acute phase (p = 0.03), whereas TEMRA subsets (p = 0.03). PD-1 expression was almost undetect- PD-1 remained at a high level not significantly different from able on naive CD4 and CD8 T cells and appeared in CM and the acute phase (Fig. 2C, middle panel). At the early time points, EM cells (p = 0.03). Expression levels dropped upon differentiation tetramer-positive cells are readily activated, as shown by the of CD8 to the TEMRA stage. Similar profiles were obtained with higher expression levels of the activation marker HLA-DR as different mAb clones and fluorochrome conjugates (Fig. 1C,1D). compared with the total CD8 population (p = 0.002) that decreases Thus, BTLA expression on CD8+ and CD4+ T cells seems to during the controlled phase of infection (p = 0.03) (Fig. 2B, left follow different rules. Whereas BTLA expression was maintained at bottom panel,2C). almost similar level along the CD4+ T cell differentiation pathway, it Because the analyzed patients were under immunosuppressive appeared to be regulated on CD8 T cells. In contrast, PD-1 regula- treatment, cyclosporin A could contribute to BTLA or PD-1 reg- tion appeared to be very similar along both CD4+ and CD8+ T cell ulation. Therefore, we analyzed the effect of cyclosporin A on differentiation pathways: it increased after the priming of naive cells the expression of PD-1 and BTLA in stimulated T cells of healthy and reached a maximum of expression at the EM stages. However, donors (Fig. 3). Consistent with previously reported results (28), PD-1 is both higher and more frequently expressed on CD8+ than cyclosporin A inhibited 45% of the expression of BTLA and 30% CD4+ T cells (Fig. 1C–E). Only a limited number of CD4+ and of the expression of PD-1 on T cells. Thus, the elevated levels of CD8+ cells coexpressed both markers, and CD8 PD-1+ and BTLA+ BTLA on the patient cells in the acute phase of infection (Fig. 2C) subpopulations appeared to be specifically excluded (Fig. 1E). High might be underestimated to a larger extent than that of PD-1. levels of expression of both markers on separate cell populations might be related to the different regulation of their expression along PD-1 and BTLA are potent inhibitors of CD3/CD28 costimulation the differentiation pathways of CD4+ and CD8+ cells. Previous studies have described inhibition of T cell responses using Unlike BTLA, which is constitutively expressed on CD4+ and anti-human PD-1 and anti-human BTLA mAbs (26, 29). However, CD8+ T cells, the expression of PD-1 is low on resting T cells. the relative efficiency of both systems has not been investigated. The Journal of Immunology 3143 Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 1. BTLA is downregulated in the course of CD8 but not CD4 T cell differentiation. CD8+ and CD4+ T cells from PBMCs of healthy volunteers (n = 6) were analyzed by flow cytometry for expression of PD1 and BTLA. A, Gating strategy for identification of T cell subsets. PBMCs were gated according to their size then into singlets and postexclusion of dead cells (Live/dead+); live cells were gated further by exclusion of CD14+CD19+ cells into CD3+ T cells then into CD3+CD4+ and CD3+CD8+ populations. CD4+ and CD8+ T cells were analyzed for CD45RA and CD27 expression resulting in the following subsets of T lymphocytes: naive CD45RA+CD27+, CM CD45RA2CD27+, EM CD45RA2CD272, and TEMRA CD45RA+CD27. B, Repre- sentative histograms of BTLA and PD-1 expression in CD8+ and CD4+ T cell differentiation subsets. BTLA and PD-1 expression on T CD8+ and T CD4+ subsets of differentiation determined by means of homemade mAbs (C) and commercially available reagents (Becton Dickinson) (D). The median, minimum, and maximum values of mfi are shown. Each population is compared with the previous subset of differentiation in the sequence naive, CM, 3144 BTLA AND T CELL-SPECIFIC hCMV RESPONSES

We have made anti-BTLA and PD-1 mAbs that are described in were selected for their ability to inhibit the binding of PD-1 to its Fig. 4A. Both BTLA 7.1 and BTLA 8.2 mAbs efficiently inhibited ligands PDL-1 and PDL-2 and BTLA to HVEM (Fig. 4A), and their the binding of HVEM-Ig to BTLA expressing COS 7 cells. effect was controlled by an isotypic mAb control. Proliferation of PD-1 3.1 but not PD-1 6.4 inhibited the binding of both PD-L1 CMVpp65+-specific CD8+ T cells was analyzed after 7 d and cal- and PD-L2 to PD-1–expressing COS 7 cells. To investigate culated as a division index resulting from the CFSE dilution. The whether PD-1 and BTLA exert similar functions on T cells, we percentage of CMVpp65+ T cells and the percentage of IFN-g used aAPCs (23). Purified CD4+ or CD8+ T cells were stimulated producing cells were determined. Both mAbs in a soluble form with beads coated with anti-CD3 and anti-CD28 mAbs together enhanced T cell proliferation to a similar extent independently of with anti-human PD-1, anti-human BTLA, or both. PD-1 and the maturation of DCs used for stimulation, suggesting a poten- BTLA significantly inhibited to a similar level (3–7 times) both tial role of these molecules in controlling CTL expansion (Fig. CD4+ and CD8+ T cell CD3/CD28-induced proliferation (Fig. 4D, top panel). The percentage of CMVpp65+ T cells among the 4B). The inhibitory effect of PD-1 or BTLA on CD3/CD28- CD8 population showed a slight decrease that suggests a similar induced proliferation of T cells was highly significant, and no of proliferating cells. We did not detect a significant in- additive or synergistic effects were detected between the two mol- crease in the production of IFN-g (Fig. 4D, lower panel) and ecules. Both PD-1 and BTLA inhibited also CD3/CD28-mediated TNF-a (data not shown) by intracellular staining. These later upregulation of IFN-g and TNF-a in CD4+ T cells and CD8+ data are compatible with recent demonstration that PD-1 mAbs T cells (Fig. 4B). IL-4 and IL-5 were also strongly inhibited by do not affect IFN-g production in short-term in vitro stimulation PD-1 and BTLA engagement on both CD4+ and CD8+ T cells (30). The trend toward enhancement was generally greater post-

(data not shown). BTLA had a stronger effect on T cell proliferation exposure of CD8+ T cells to a weaker stimulating iDC population Downloaded from (p = 0.0001) and inhibited more efficiently cytokine secretion (p # than to strongly stimulating mDCs, as shown by allogeneic stimu- 0.004) than PD-1 (Fig. 4B, middle and bottom panels). Using lation results (Fig. 4C). The enhancing effect of BTLA and PD-1 on suboptimal T cell activation, PD-1 and BTLA mAbs strongly in- CMV-specific T cells was similar in both weak and strong DC hibited T cell proliferation and Th1/Th2 cytokine secretion, and stimulation conditions. This could be related to the differentiated again, no functional cooperation was detected between the two phenotype of CMV-specific T cells.

molecules (data not shown). http://www.jimmunol.org/

Blockade of PD-1 or BTLA pathway enhances both allogeneic + Discussion and CMV-specific CD8 T cell proliferation We describe in this study for the first time the dynamic expression We investigated the functions of PD-1 and BTLA in stimulated of the negative regulator of T cell activation BTLA on human CD8+ populations of T lymphocytes in which their ligands PD-L1, T cells in the course of primary CMV infection and have de- PD-L2, and HVEM have been expressed using T cell allogeneic termined its function in the control of memory T cell expansion. responses against DCs presenting weak (iDCs) or strong (mDCs) The differentiation of human naive T cells into effector and stimulatory capacity. Purified CFSE-labeled T cells were cultured memory cells is accompanied by changes in the expression of for 5 d with allogeneic iDCs or mDCs in the presence of blocking surface molecules and effector molecules. This topic has been re- by guest on September 28, 2021 Fab specific to PD-1 and BTLA or isotype Fab control. Prolifera- cently extensively debated (31). Several markers include receptors tion was quantified as the percentage of proliferating CD8+ T cells involved in T cell activation, costimulation, chemokine receptors, calculated by CFSE dilution (Fig. 4C, top panel). Activation was adhesion molecules, and, more recently, expression of the cosig- quantified by CD25 expression on total T cells (Fig. 4C, middle naling molecule PD-1 (5, 6, 32–35). Interestingly, PD-1 expression panel), and IFN-g release in supernatants was quantified by depends on two settings in vivo. PD-1 is upregulated early postacti- ELISA (Fig. 4C, bottom panel). In both iDC and mDC allo- vation, but it is downregulated at late stages of differentiation. How- geneic stimulations, the blockade of BTLA or PD-1 resulted in ever, it is also linked to T cell activation because PD-1 overexpression a significant increase in proliferation, CD25 expression, and IFN-g occurs along with the upregulation of activation markers. release. Differences between the effects of the blockade of BTLA To understand how BTLA could affect the function of CD4+ and and PD-1 were observed, depending on the strength of the stim- CD8+ T cells, we first examined the expression of BTLA on ulation. In a weak allostimulation condition (iDCs), the effects of different T cell subsets and compared it to PD-1, which has been the blockade of BTLA on proliferation (p = 0.03), CD25 expres- extensively studied during viral infections in humans. As already sion (p = 0.0025), and IFN-g release (p = 0.015) were signif- shown, PD-1 is barely detected on naive T cells and upregulated icantly lower than the effects of the blockade of PD-1. In a stronger on effector, EM, and CM CD8+ T cells, as well as memory-type allostimulation condition (mDCs), the effects of the blockade of CD4+ T cells from healthy donors. In contrast, BTLA expression BTLA were similar to PD-1, excepted for IFN-g release, in which is stable on different CD4+ T cell subsets, but it is downregulated a lower effect of the blockade of BTLA was observed (p = 0.013). on effector and EM CD8+ T cells as compared with naive cells. The expression of BTLA and PD-1 on CMV-specific T cells These observations were reproduced in vitro because T cell acti- prompted us to analyze their function in the activation of CMV- vation induced progressive PD-1 upregulation on CD4+ and CD8+ specific CD8+ T cells. To address this issue, we compared the effect T cells. In contrast, BTLA was downregulated upon activation of of PD-1 and BTLA pathways on CMVpp65+-specific CD8+ T cell CD8+ but not CD4+ T cells. It would suggest a differential regu- proliferation and function (Fig. 4D). To this end, we blocked lation of BTLA expression in CD4+ and CD8+ cells. The basis for PD-1 or BTLA by soluble forms of Ab in autologous cocultures this differential decreased expression is unclear. The only avail- of T cells with iDCs or mDCs pulsed with pp65 CMV peptide. able data on of the regulation of BTLA expression show that it is Anti–PD-1 (PD1.3.1) or anti-BTLA (BTLA 8.2)-blocking Abs induced by phorbol esters and calcium ionophores, but it is

EM, and TEMRA, and the significant p values are indicated as pp , 0.05 (Wilcoxon test). E, BTLA and PD1 coexpression on CD4 and CD8 T cells. PBMCs from healthy donors were stained and analyzed using the gating strategy described in A. An example of BTLA and PD-1 coexpression on CD4+ CD3+and CD8+CD3+ T cells is shown. mfi, median of fluorescence. The Journal of Immunology 3145 Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 2. BTLA and PD-1 expression during CMV infection. A, Gates used for the analysis of patient CD8+CMV-tetramer+ T cells. We show the gating of CD8+ T cells on patient 1, wk 3, and CMV-tetramer+CD8+ T cells on patient 1, wk 145, after gating on viable cells (not shown). Gates for subsets of differentiation according to CD45RA and CD27 expression are set on CD8+ T cells (patient 1, wk 3). Gates for BTLA, PD-1, and HLA-DR expression are set on negatively expressing populations within CD8+ T cells (patient 1, wk 3), respectively, as TEMRA for BTLA and naive for PD-1 and HLA-DR. B, Follow-up of markers expressed on CD8+CMV-tetramer+ T cells and CMV viremia in the two transplanted patients over time. Left panel shows raw data of CD45RA/CD27 coexpression, BTLA/PD-1 coexpression, and HLA-DR expression for each blood sampling in a representative CD8+CMV-tetramer+ T cell population (IE-elr for patient 1 and pp65 for patient 2). The number of CD8+CMV-tetramer+ T cells analyzed ranged from 50–20,000 for patient 1 and 50– 100,000 for patient 2 from early to late time points. Right panel shows median, maximal, and minimum mfi values of BTLA, PD-1, and HLA-DR for the three CMV tetramers plotted over time for each patient. Viremia is represented as a gray zone and scored positive when .1 DNA copy/ml that defines the acute phase of infection. Viremia ,1 DNA copy/ml defines the controlled phase of infection and is not shown on graphs. C, Global analysis of CD8+CMV- tetramer+ T cell differentiation phenotype and BTLA, PD-1, and HLA-DR expression in acute and controlled phases of infection in the two transplanted patients. Acute and controlled phases of infection are defined in C. Median, maximum, and minimum values measured for the two patient cells tested with 3146 BTLA AND T CELL-SPECIFIC hCMV RESPONSES

Whereas both PD-1 and BTLA are upregulated on CMV- specific cells during acute infection, PD-1 alone remains signifi- cantly overexpressed when viremia is controlled. This finding would indicate differences in the regulation of the expression of PD-1 and BTLA molecules. What are the likely roles of the expres- sion of negative regulators on Ag-specific cells? The upregulation of PD-1 was proposed to be a likely mechanism for the control of proliferation and apoptosis of activated cells (6). Based upon the recent publications in mice using BTLA2/2 cells, two roles have been proposed. First, BTLA was initially identified as having in- hibitory actions on T cell proliferation and cytokine production FIGURE 3. Inhibition of BTLA and PD-1 expression on activated T in vitro (11, 12). Inhibitory actions were observed in vivo in mod- CD4+ cells by CsA. Purified CD4+ T lymphocytes were stimulated by els of experimental allergic encephalomyelitis and allergic airway means of beads coated with CD3 and CD28 mAbs in the presence or inflammation (11, 38). However, a prosurvival role of BTLA was not of CsA (100 nM). Expression of BTLA and PD-1 was determined + + also proposed (39). Hurchla et al. (39) tested BTLA function in 24 h poststimulation on CD4 CD25 T cells and represented as a ratio a mice in vivo model of chronic allostimulation using BTLA2/2 of mfi to isotypic staining control. Median, maximum, and minimum val- cells. They proposed that, under chronic allostimulation, BTLA ues are shown. The percentage of reduction of expression in the presence of CsA is shown. The p values were calculated between both stimulation is acting in a prosurvival manner and is necessary for the main- conditions using the Wilcoxon test. pp , 0.05. CsA, cyclosporin A. tenance, rather than the initiation, of the effector T cell response. Downloaded from In contrast, Deppong et al. (40), also using BTLA-deficient mice, had increased survival in comparison with wild-type mice in the downregulated by cyclosporin A (28). This report indicates that asthma inhaled-allergen model. In humans, our in vitro results multiple events, some being dependent on calcineurin, are in- suggest that BTLA exerts mainly inhibitory function on CMV- volved. However, these data do not explain the downregulation specific memory T cells and controls the expansion of the + + of BTLA expression and the differences between CD8 and CD4 CMV-specific T cells. Additional roles, such as a role in survival, http://www.jimmunol.org/ cells. will have to be analyzed. Although the downregulation of BTLA upon interaction with its Recently, Blackburn et al. (4–8) investigated the different in- ligand HVEM has not been reported, a similar mechanism is hibitory receptors that might be involved in the coregulation of reported for various receptors, including internalization of HVEM CD8+ T cell exhaustion in the murine LCMV model, including the itself (36). Moreover, this does not explain the differences between already identified receptor PD-1. These results suggest that in CD8+ and CD4+ cells. CMV encodes the UL144 protein that was addition to PD-1, LAG-3 and 2B4 were upregulated on exhausted shown to interact with BTLA and to decrease T cell activation CD8+ effectors (19). In contrast, BTLA expression was not mod- (37). Our observation demonstrating that BTLA was upregulated ified in naive, memory, and exhausted CD8+ T cells. Additional during the acute phase of CMV infection would make BTLA information regarding the expression of the inhibitory receptors by guest on September 28, 2021 a likely candidate to be targeted because of its inhibitory role in including BTLA during the acute phase of the infection was not T cell activation during CMV (re)activation. Whether the decreased included. Our data indicate that during reactivation of human expression of BTLA on CD8+ T cells is due the internalization, CMV, BTLA is highly regulated on the CD8+ subsets. In contrast, masking of the binding site, or transcriptional regulation will deserve when the patients had controlled CMV viremia, the expression of further investigation. In addition, because the data were obtained in BTLA diminished, suggesting that BTLA is more involved in the transplanted patients, the immunosuppressive treatment may lower control of acute than controlled infection. Further studies will be BTLA and PD-1 overexpression, as cyclosporin A was shown to necessary to assess a putative role of BTLA in T cell inhibitio- regulate the promoters of both molecules and underestimate their n/exhaustion during HIV, HCV, and other viral infections. increase. Our data suggest that PD-1 and BTLA cross-linking elicits similar The expression and the role of negative regulators of immune inhibitory functions on T cell function regarding CD4 and CD8 pro- functions are instrumental for the understanding of the basis of the liferation and cytokine production. This demonstrates that upon ap- T cells responses during the course of viral infections. This study propriate ligation, they both are strong inhibitors of T cell responses. demonstrates that the negative regulator BTLA is highly expressed Like PD-1, BTLA contains in its cytoplasmic domain two motifs: an during the acute phase of infection on CMV-specific T cells when ITIM and an ITSM, both requiring the tyrosine phosphatases SHP1 the population is constituted by naive, CM, or TEMRA T cells and and SHP2 (11, 12, 15). At present, the targets of SHP1 and SHP2 that expression of BTLA decreases when viremia is controlled, and recruited to BTLA are unknown. It is possible that they also have CMV population is constituted in a large majority by CM, EM, or a role in dephosphorylation of signaling intermediates downstream TEMRA T cells. In this regard, BTLA expression follows closely of Ag receptors in lymphocytes (11, 15, 41) or in specifically target- the expression of the activation marker HLA-DR, which is highly ing the PI3K-PKB pathway, as proposed for PD-1. This would be in upregulated on CMV-specific T cells during the acute phase of in- linewith the fact that they were neither additive nor synergistic in their fection and returns to lower levels when viremia is controlled. function using CD3+CD28 costimulation. In addition, PD-1 and PD-1 has also been described as regulated by activation (35). BTLA mAb increased both allogeneic and CMV-specific T cell pro- However, a striking difference in expression is found when com- liferation. The likely mechanisms rely on the inhibition of the inter- paring PD-1 and BTLA expression. action between BTLA and HVEM because the mAbs were selected

the different tetramers are plotted. Left panel shows the repartition of CD8+CMV-tetramer+ T cells into the four subsets of differentiation based on CD45RA and CD27 expression. Middle panel shows BTLA and PD-1 mfi in CMV pp65 tetramer+ T cells and in total CD8+ T cells of both patients at an early time point (wk 3 and wk 5, respectively). Right panel shows HLA-DR expression. All p values are calculated with the Wilcoxon paired test. pp , 0.05; pp,0.001 , p , 0.01; pppp , 0.001. The Journal of Immunology 3147 Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 4. Inhibition of T cell activation by BTLA and PD-1 mAb. A, Characterization of PD-1 and BTLA mAbs. COS 7 cells transfected to express BTLA (left panel) or PD-1 (middle and right panels) were preincubated with the indicated concentrations of anti-BTLA (BTLA7.1 and BTLA8.2) or anti- PD-1 (PD1.3.1 and PD1.6.4), respectively, followed by the addition of HVEM-Ig (left panel), PD-L1–Ig (middle panel), or PD-L2–Ig (right panel). B, Inhibition of T cell activation by BTLA and PD-1 mAbs. Purified CD4 T and CD8 T lymphocytes were stimulated by beads coated with CD3 and CD28 mAbs and with different combinations of mAb directed against BTLA and PD-1 (n = 6). Proliferation was measured by [3H]thymidine incorporation. IFN-g and TNF-a release in supernatants was quantified by ELISA. The p values indicated with asterisks are calculated to compare differences between each Ab mix and the isotype control, except for combination of the BTLA and PD-1, which is compared with PD-1 alone; BTLA and PD1 are also compared with each other. C, Blockade of BTLA or PD-1 pathways in a T allogeneic stimulation. iDCs or mDCs were used to stimulate allogeneic purified CFSE-labeled T lymphocytes in the presence of blocking Fab of BTLA 8.2, PD-1 3.1, or isotype. Proliferation was quantified by CFSE dilution and represented as the percentage of proliferating cells among the CD8+ T cells. CD25 expression is represented as a ratio of mfi to isotypic staining in total T cell population. IFN-g release in supernatants was quantified by ELISA. D, Inhibition of CMV-specific activation by BTLA and PD-1 mAbs. iDCs or mDCs were stimulated by pp65 CMV peptide and cocultivated with autologous purified CFSE-labeled T lymphocytes in the presence of BTLA or PD-1 mAbs or matched isotype. Proliferation of CMVpp65+ CD8+ T lymphocytes was analyzed after 7 d and represented by the division index calculated with the proliferation tool of FlowJo (Tree Star) as a result of CFSE dilution. Median, maximum, and minimum values of six separated experiments are shown. The percentage of CMVpp65+ CD8+ T lymphocytes producing IFN-g was determined after an intracellular staining (n = 4). The p values were calculated using the Wilcoxon paired test to compare differences between BTLA or PD1 mABs to the isotype control, and between BTLA and PD1. pp , 0.05; pp0.001 , p , 0.01; pppp , 0.001. p . 0.05 is not shown. for their efficient inhibition of the interaction on BTLA with HVEM. murine CMV model. In this model, some CD8 T cell populations This hypothesis would be in line with the role of HVEM in the in- may be sustained or increase over time in tissues. This accumu- hibition of T cell functions via HVEM (13, 42). However, agonist lation of virus-specific CD8+ T cells is called memory inflation. effects of BTLA mAbs cannot be ruled out. In this model, Snyder et al. (43) suggest that large pools of Memory maintenance against CMV has been recently revisited. differentiated cells found in the blood and tissues are turning over Snyder et al. (43) have studied the host–virus balance in the continuously with rapid resupply from self-renewing memory 3148 BTLA AND T CELL-SPECIFIC hCMV RESPONSES pools as well as naive precursors. This question is important to lymphoma and B-cell small lymphocytic lymphoma/chronic lymphocytic leuke- mia. Hum. Pathol. 39: 1050–1058. keep large functional pools of effectors. Altogether, these data add 21. Derre´, L., J. P. Rivals, C. Jandus, S. Pastor, D. Rimoldi, P. Romero, O. Michielin, new information on the pathways of viral-specific T cell responses D. Olive, and D. E. Speiser. 2010. BTLA mediates inhibition of human tumor- regulation and on the role of negative regulators of T cell activa- specific CD8+ T cells that can be partially reversed by vaccination. J. Clin. Invest. 120: 157–167. tion during human CMV reactivation. 22. Boom, R., C. Sol, J. Weel, Y. Gerrits, M. de Boer, and P. Wertheim-van Dillen. 1999. A highly sensitive assay for detection and quantitation of human cytomeg- alovirus DNA in serum and plasma by PCR and electrochemiluminescence. Acknowledgments J. Clin. Microbiol. 37: 1489–1497. We thank M. Ghiotto, S. Amir, and F. Mallet for excellent technical assis- 23. Riley, J. L., K. Schlienger, P. J. Blair, B. Carreno, N. Craighead, D. Kim, tance. R. G. Carroll, and C. H. June. 2000. Modulation of susceptibility to HIV-1 infection by the cytotoxic T lymphocyte antigen 4 costimulatory mole- cule. J. Exp. Med. 191: 1987–1997. Disclosures 24. Appay, V., P. R. Dunbar, M. Callan, P. Klenerman, G. M. Gillespie, L. Papagno, G. S. Ogg, A. King, F. Lechner, C. A. Spina, et al. 2002. Memory CD8+ T cells The authors have no financial conflicts of interest. vary in differentiation phenotype in different persistent virus infections. Nat. Med. 8: 379–385. 25. van Lier, R. A., I. J. ten Berge, and L. E. Gamadia. 2003. Human CD8(+) T-cell References differentiation in response to viruses. Nat. Rev. Immunol. 3: 931–939. 1. Sharpe, A. H., and G. J. Freeman. 2002. The B7-CD28 superfamily. Nat. Rev. 26. Bennett, F., D. Luxenberg, V. Ling, I. M. Wang, K. Marquette, D. Lowe, Immunol. 2: 116–126. N. Khan, G. Veldman, K. A. Jacobs, V. E. Valge-Archer, et al. 2003. Program 2. Greenwald, R. J., G. J. Freeman, and A. H. Sharpe. 2005. The B7 family death-1 engagement upon TCR activation has distinct effects on costimulation revisited. Annu. Rev. Immunol. 23: 515–548. and cytokine-driven proliferation: attenuation of ICOS, IL-4, and IL-21, but not CD28, IL-7, and IL-15 responses. J. Immunol. 170: 711–718.

3. Agata, Y., A. Kawasaki, H. Nishimura, Y. Ishida, T. Tsubata, H. Yagita, and Downloaded from T. Honjo. 1996. Expression of the PD-1 antigen on the surface of stimulated 27. Carter, L., L. A. Fouser, J. Jussif, L. Fitz, B. Deng, C. R. Wood, M. Collins, mouse T and B lymphocytes. Int. Immunol. 8: 765–772. T. Honjo, G. J. Freeman, and B. M. Carreno. 2002. PD-1:PD-L inhibitory path- 4. Barber, D. L., E. J. Wherry, D. Masopust, B. Zhu, J. P. Allison, A. H. Sharpe, way affects both CD4(+) and CD8(+) T cells and is overcome by IL-2. Eur. J. G. J. Freeman, and R. Ahmed. 2006. Restoring function in exhausted CD8 Immunol. 32: 634–643. T cells during chronic viral infection. Nature 439: 682–687. 28. Zeng, C., Y. Zhen, S. A. Shu, T. Wu, H. Yi, Z. X. Lian, and Y. Zhao. 2007. The 5. Day, C. L., D. E. Kaufmann, P. Kiepiela, J. A. Brown, E. S. Moodley, S. Reddy, influence of cyclosporin A on lymphocyte attenuator expression. J. Autoimmun. E. W. Mackey, J. D. Miller, A. J. Leslie, C. DePierres, et al. 2006. 28: 234–244. PD-1 expression on HIV-specific T cells is associated with T-cell exhaustion and 29. Chemnitz, J. M., R. V. Parry, K. E. Nichols, C. H. June, and J. L. Riley. 2004. disease progression. Nature 443: 350–354. SHP-1 and SHP-2 associate with immunoreceptor tyrosine-based switch motif of http://www.jimmunol.org/ 6. Petrovas, C., J. P. Casazza, J. M. Brenchley, D. A. Price, E. Gostick, programmed death 1 upon primary human T cell stimulation, but only receptor W. C. Adams, M. L. Precopio, T. Schacker, M. Roederer, D. C. Douek, and ligation prevents T cell activation. J. Immunol. 173: 945–954. R. A. Koup. 2006. PD-1 is a regulator of virus-specific CD8+ T cell survival in 30. Fourcade, J., P. Kudela, Z. Sun, H. Shen, S. R. Land, D. Lenzner, P. Guillaume, HIV infection. J. Exp. Med. 203: 2281–2292. I. F. Luescher, C. Sander, S. Ferrone, et al. 2009. PD-1 is a regulator of NY-ESO-1- 7. Golden-Mason, L., B. Palmer, J. Klarquist, J. A. Mengshol, N. Castelblanco, and specific CD8+ T cell expansion in melanoma patients. J. Immunol. 182: 5240–5249. H. R. Rosen. 2007. Upregulation of PD-1 expression on circulating and intra- 31. Appay, V., R. A. van Lier, F. Sallusto, and M. Roederer. 2008. Phenotype and hepatic hepatitis C virus-specific CD8+ T cells associated with reversible im- function of human T lymphocyte subsets: consensus and issues. Cytometry A 73: mune dysfunction. J. Virol. 81: 9249–9258. 975–983. 8. D’Souza, M., A. P. Fontenot, D. G. Mack, C. Lozupone, S. Dillon, A. Meditz, 32. Freeman, G. J., E. J. Wherry, R. Ahmed, and A. H. Sharpe. 2006. Reinvigorating C. C. Wilson, E. Connick, and B. E. Palmer. 2007. Programmed death 1 ex- exhausted HIV-specific T cells via PD-1-PD-1 ligand blockade. J. Exp. Med. pression on HIV-specific CD4+ T cells is driven by viral replication and asso- 203: 2223–2227. ciated with T cell dysfunction. J. Immunol. 179: 1979–1987. 33. Trautmann, L., L. Janbazian, N. Chomont, E. A. Said, S. Gimmig, B. Bessette, by guest on September 28, 2021 9. Boni, C., P. Fisicaro, C. Valdatta, B. Amadei, P. Di Vincenzo, T. Giuberti, M. R. Boulassel, E. Delwart, H. Sepulveda, R. S. Balderas, et al. 2006. Upre- D. Laccabue, A. Zerbini, A. Cavalli, G. Missale, et al. 2007. Characterization of gulation of PD-1 expression on HIV-specific CD8+ T cells leads to reversible virus (HBV)-specific T-cell dysfunction in chronic HBV infection. immune dysfunction. Nat. Med. 12: 1198–1202. J. Virol. 81: 4215–4225. 34. Urbani, S., B. Amadei, D. Tola, M. Massari, S. Schivazappa, G. Missale, and 10. Peng, G., S. Li, W. Wu, X. Tan, Y. Chen, and Z. Chen. 2008. PD-1 upregulation C. Ferrari. 2006. PD-1 expression in acute hepatitis C virus (HCV) infection is is associated with HBV-specific T cell dysfunction in chronic hepatitis B associated with HCV-specific CD8 exhaustion. J. Virol. 80: 11398–11403. patients. Mol. Immunol. 45: 963–970. 35. Sauce, D., J. R. Almeida, M. Larsen, L. Haro, B. Autran, G. J. Freeman, and 11. Watanabe, N., M. Gavrieli, J. R. Sedy, J. Yang, F. Fallarino, S. K. Loftin, V. Appay. 2007. PD-1 expression on human CD8 T cells depends on both state of M. A. Hurchla, N. Zimmerman, J. Sim, X. Zang, et al. 2003. BTLA is differentiation and activation status. AIDS 21: 2005–2013. a lymphocyte inhibitory receptor with similarities to CTLA-4 and PD-1. Nat. 36. Morel, Y., J. M. Schiano de Colella, J. Harrop, K. C. Deen, S. D. Holmes, Immunol. 4: 670–679. T. A. Wattam, S. S. Khandekar, A. Truneh, R. W. Sweet, J. A. Gastaut, et al. 12. Han, P., O. D. Goularte, K. Rufner, B. Wilkinson, and J. Kaye. 2004. An in- 2000. Reciprocal expression of the TNF family receptor herpes virus entry hibitory Ig superfamily protein expressed by lymphocytes and APCs is also an mediator and its ligand LIGHT on activated T cells: LIGHT down-regulates its early marker of thymocyte positive selection. J. Immunol. 172: 5931–5939. own receptor. J. Immunol. 165: 4397–4404. 13. Gonzalez, L. C., K. M. Loyet, J. Calemine-Fenaux, V. Chauhan, B. Wranik, 37. Cheung, T. C., I. R. Humphreys, K. G. Potter, P. S. Norris, H. M. Shumway, W. Ouyang, and D. L. Eaton. 2005. A coreceptor interaction between the B. R. Tran, G. Patterson, R. Jean-Jacques, M. Yoon, P. G. Spear, et al. 2005. CD28 and TNF receptor family members B and T lymphocyte attenuator and Evolutionarily divergent herpesviruses modulate T cell activation by targeting herpesvirus entry mediator. Proc. Natl. Acad. Sci. USA 102: 1116–1121. the herpesvirus entry mediator cosignaling pathway. Proc. Natl. Acad. Sci. USA 14. Murphy, K. M., C. A. Nelson, and J. R. Sedy´. 2006. Balancing co-stimulation 102: 13218–13223. and inhibition with BTLA and HVEM. Nat. Rev. Immunol. 6: 671–681. 38. Deppong, C., T. I. Juehne, M. Hurchla, L. D. Friend, D. D. Shah, C. M. Rose, 15. Gavrieli, M., N. Watanabe, S. K. Loftin, T. L. Murphy, and K. M. Murphy. 2003. T. L. Bricker, L. P. Shornick, E. C. Crouch, T. L. Murphy, et al. 2006. Cutting edge: Characterization of phosphotyrosine binding motifs in the cytoplasmic domain of B B and T lymphocyte attenuator and programmed death receptor-1 inhibitory recep- and T lymphocyte attenuator required for association with protein tyrosine phos- tors are required for termination of acute allergic airway inflammation. J. Immunol. phatases SHP-1 and SHP-2. Biochem. Biophys. Res. Commun. 312: 1236–1243. 176: 3909–3913. 16. Chemnitz, J. M., A. R. Lanfranco, I. Braunstein, and J. L. Riley. 2006. B and 39. Hurchla, M. A., J. R. Sedy, and K. M. Murphy. 2007. Unexpected role of B and T lymphocyte attenuator-mediated signal transduction provides a potent inhibi- T lymphocyte attenuator in sustaining cell survival during chronic allostimula- tory signal to primary human CD4 T cells that can be initiated by multiple tion. J. Immunol. 178: 6073–6082. phosphotyrosine motifs. J. Immunol. 176: 6603–6614. 40. Deppong, C., J. M. Degnan, T. L. Murphy, K. M. Murphy, and J. M. Green. 2008. 17. Krieg, C., O. Boyman, Y. X. Fu, and J. Kaye. 2007. B and T lymphocyte at- B and T lymphocyte attenuator regulates T cell survival in the lung. J. Immunol. tenuator regulates CD8+ T cell-intrinsic homeostasis and memory cell genera- 181: 2973–2979. tion. Nat. Immunol. 8: 162–171. 41. Okazaki, T., A. Maeda, H. Nishimura, T. Kurosaki, and T. Honjo. 2001. 18. Tao, R., L. Wang, R. Han, T. Wang, Q. Ye, T. Honjo, T. L. Murphy, PD-1 immunoreceptor inhibits receptor-mediated signaling by recruiting K. M. Murphy, and W. W. Hancock. 2005. Differential effects of B and src homology 2-domain-containing tyrosine phosphatase 2 to phosphotyrosine. T lymphocyte attenuator and programmed death-1 on acceptance of partially Proc. Natl. Acad. Sci. USA 98: 13866–13871. versus fully MHC-mismatched cardiac allografts. J. Immunol. 175: 5774–5782. 42. Sedy, J. R., M. Gavrieli, K. G. Potter, M. A. Hurchla, R. C. Lindsley, K. Hildner, 19. Blackburn, S. D., H. Shin, W. N. Haining, T. Zou, C. J. Workman, A. Polley, S. Scheu, K. Pfeffer, C. F. Ware, T. L. Murphy, and K. M. Murphy. 2005. B and M. R. Betts, G. J. Freeman, D. A. Vignali, and E. J. Wherry. 2009. Coregulation T lymphocyte attenuator regulates T cell activation through interaction with of CD8+ T cell exhaustion by multiple inhibitory receptors during chronic viral herpesvirus entry mediator. Nat. Immunol. 6: 90–98. infection. Nat. Immunol. 10: 29–37. 43. Snyder, C. M., K. S. Cho, E. L. Bonnett, S. van Dommelen, G. R. Shellam, and 20. Xerri, L., B. Chetaille, N. Seriari, C. Attias, Y. Guillaume, C. Arnoulet, and A. B. Hill. 2008. Memory inflation during chronic viral infection is maintained by D. Olive. 2008. Programmed death 1 is a marker of angioimmunoblastic T-cell continuous production of short-lived, functional T cells. Immunity 29: 650–659.