Phenotype, Function, and Expression Profiles of Programmed Death-1 hi CD8 T Cells in Healthy Human Adults

This information is current as Jaikumar Duraiswamy, Chris C. Ibegbu, David Masopust, of September 25, 2021. Joseph D. Miller, Koichi Araki, Gregory H. Doho, Pramila Tata, Satish Gupta, Michael J. Zilliox, Helder I. Nakaya, Bali Pulendran, W. Nicholas Haining, Gordon J. Freeman and Rafi Ahmed

J Immunol 2011; 186:4200-4212; Prepublished online 7 Downloaded from March 2011; doi: 10.4049/jimmunol.1001783 http://www.jimmunol.org/content/186/7/4200 http://www.jimmunol.org/ References This article cites 63 articles, 33 of which you can access for free at: http://www.jimmunol.org/content/186/7/4200.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 25, 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 © 2011 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Phenotype, Function, and Profiles of Programmed Death-1hi CD8 T Cells in Healthy Human Adults

Jaikumar Duraiswamy,*,†,1 Chris C. Ibegbu,*,† David Masopust,*,†,2 Joseph D. Miller,*,†,3 Koichi Araki,*,† Gregory H. Doho,‡ Pramila Tata,x Satish Gupta,x Michael J. Zilliox,*,† Helder I. Nakaya,*,{ Bali Pulendran,*,{ W. Nicholas Haining,‖ Gordon J. Freeman,#,** and Rafi Ahmed*,†

T cell dysfunction is an important feature of many chronic viral infections. In particular, it was shown that programmed death-1 (PD-1) regulates T cell dysfunction during chronic lymphocytic choriomeningitis virus infection in mice, and PD-1hi cells exhibit an intense exhausted gene signature. These findings were extended to human chronic infections such as HIV, hepatitis C virus, and hi

hepatitis B virus. However, it is not known if PD-1 cells of healthy humans have the traits of exhausted cells. In this study, we Downloaded from provide a comprehensive description of phenotype, function, and gene expression profiles of PD-1hi versus PD-1lo CD8 T cells in the peripheral blood of healthy human adults as follows: 1) the percentage of naive and memory CD8 T cells varied widely in the peripheral blood cells of healthy humans, and PD-1 was expressed by the memory CD8 T cells; 2) PD-1hi CD8 T cells in healthy humans did not significantly correlate with the PD-1hi exhausted gene signature of HIV-specific human CD8 T cells or chronic lymphocytic choriomeningitis virus-specific CD8 T cells from mice; 3) PD-1 expression did not directly affect the ability of CD8

T cells to secrete cytokines in healthy adults; 4) PD-1 was expressed by the effector memory compared with terminally differ- http://www.jimmunol.org/ entiated effector CD8 T cells; and 5) finally, an interesting inverse relationship between CD45RA and PD-1 expression was observed. In conclusion, our study shows that most PD-1hi CD8 T cells in healthy adult humans are effector memory cells rather than exhausted cells. The Journal of Immunology, 2011, 186: 4200–4212.

D8 T cells are a critical component of the for enhancing natural immunological control over chronic viral in- and are responsible for killing virus-infected cells and fections. C control of persistent and reactivating viruses. However, Programmed death-1 (PD-1) is a member of the CD28 family of persistent antigenic stimulation leads to CD8 T cell exhaustion, immune modulators (2–4). It is upregulated on CD8 and CD4 T cells characterized by the induction of a hypoproliferative state and loss upon activation. PD-1 binds to its ligands PD- (B7-H1) or PD-L2 by guest on September 25, 2021 of the ability to produce antiviral cytokines (1). Reversing CD8 (B7-DC). Ligation of PD-1 results in dephosphorylation of signal- T cell exhaustion could provide a promising therapeutic approach ing molecules downstream of the TCR, thus dampening T cell sensitivity to antigenic stimulation. Significant evidence suggests that this pathway inhibits T cell responses upon persistent antigenic † *Emory Vaccine Center, Emory University, Atlanta, GA 30322; Department of Mi- stimulation (5–7). It has been shown that immunoreceptor tyrosine- crobiology and Immunology, Emory University, Atlanta, GA 30322; ‡Emory Biomarker Service Center, Emory University, Atlanta, GA 30322; xStrand Life Sciences, Bangalore based switch motif of PD-1 cytoplasmic tail recruits tyrosine { ‖ 560024, India; Department of Pathology, Emory University, Atlanta, GA 30322; De- phosphatases, Src homology region 2 domain-containing phos- partment of Pediatric Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA 02115; #Department of Medical Oncology, Dana-Farber Cancer Institute, phatase-1, and -2, which in turn interferes with proximal TCR sig- Harvard Medical School, Boston, MA 02115; and **Department of Medicine, Harvard naling pathways to affect T cell functions (8). PD-1–deficient mice, Medical School, Boston, MA 02115 when crossed to the NOD background, experience more rapid and 1Current address: Ovarian Cancer Research Center and Department of Obstetrics and severe diabetes (9–12). PD-L12/2 mice die of excessive T cell im- Gynecology, University of Pennsylvania School of Medicine, Philadelphia, PA. munopathology following chronic lymphocytic choriomeningitis 2 Current address: Department of Microbiology and Center for Immunology, Univer- virus (LCMV) infection (13). We have shown that PD-1 regulates sity of Minnesota, Minneapolis, MN. T cell dysfunction during chronic LCMV infection in mice and 3Current address: Centers for Disease Control and Prevention, Atlanta, GA. demonstrated that in vivo blockade of PD-1–PD-L1 interactions: 1) Received for publication June 17, 2010. Accepted for publication January 29, 2011. restores effector functions of exhausted CD8 T cells; and 2) leads to This work was supported by National Institutes of Health Grant P01 AI080192-01 (to substantial reduction in virus replication (13). These findings were R.A.) and Grand Challenges in Global Health Initiative Grant 05GCGH0 (to R.A.). extended to HIV, hepatitis B virus (HBV), and hepatitis C virus The sequences presented in this article have been submitted to the Gene Expression Omnibus under accession number GSE26495. (HCV) infections in humans and SIV infection in macaques, in- Address correspondence and reprint requests to Dr. Rafi Ahmed, Emory Vaccine dicating that PD-1 overexpression on T cells plays an important role Center and Department of Microbiology and Immunology, Emory University School in these infections (14–20). These data suggest that abrogation of of Medicine, 1510 Clifton Road, G211, Rollins Research Center, Atlanta, GA 30322. the PD-1 inhibitory pathway may contribute to successful treatment E-mail address: [email protected] of life-threatening chronic infections in humans. Abbreviations used in this article: CMV, cytomegalovirus; DEG, differentially ex- pressed gene; GSEA, gene set enrichment analysis; HBV, hepatitis B virus; HCV, hepa- In addition to chronic viral infections, PD-1 inhibitory pathway titis C virus; KIR, killer cell Ig-like receptor; KLR, killer cell lectin-like receptor; plays an important role in tumors. Expression of PD-L1 is reported LCMV, lymphocytic choriomeningitis virus; PD-1, programmed death-1; TCM,central on a variety of human tumors. Expression of PD-L1 by tumor cells memory T cell; T , effector memory T cell; T , effector memory RA T cell. EM EMRA correlates with a very poor prognosis, suggesting that cancer cells Copyright Ó 2011 by The American Association of Immunologists, Inc. 0022-1767/11/$16.00 purloin this inhibitory pathway to evade the host immune response www.jimmunol.org/cgi/doi/10.4049/jimmunol.1001783 The Journal of Immunology 4201

(21, 22). Recent studies also show that the tumor microenviron- For this reason, we examined the levels of PD-1 expression on ment may play a role in the induction and maintenance of PD-1 the resting CD8 T cells from healthy individuals. We found that expression on tumor-reactive T cells and that expression of PD-1 PD-1 is expressed by ∼60% of memory CD8 T cells in the pe- on tumor-infiltrating lymphocytes impairs the antitumor immune ripheral blood cells of healthy humans. First, we analyzed the responses in humans (23). gene expression profile of PD-1hi CD8 T cells in the peripheral Although PD-1 expression on virus-specific CD8 T cells of blood cells of healthy human adults in comparison with PD-1lo chronically infected patients is well described, little is known about and naive CD8 T cells. We found that the gene expression profiles its pattern expression on CD8 T cells of healthy human adults. of PD-1hi and PD-1lo CD8 T cells were closely related. Then Although PD-1 contributes to functional defects of memory CD8 we compared gene expression profiles of the PD-1hi CD8 T cell T cells, deficiencies in the PD-1 are associated with autoimmune subset of healthy humans with PD-1hi exhausted signatures of diseases (5, 10, 24). PD-1 is transiently expressed on activated HIV-specific human CD8 T cells or LCMV-specific CD8 T cells T cells, and frequent TCR stimulation is required to maintain PD- from chronically (LCMV clone-13 strain) infected mice. We 1 expression (13, 25, 26). However, the proportion of activated found that signatures of characteristic of exhausted CD8 versus exhausted cells expressing PD-1 is not known. Hence it is T cells were not enriched in the PD-1hi CD8 T cell subset in also essential to delineate when PD-1 expression is the signal of healthy humans. Next, we compared the gene expression profile of physiologic regulatory mechanisms of activated cells, a marker of PD-1hi CD8 T cells with their phenotypic and functional charac- exhausted cells, or a mediator of functional exhaustion. teristics. Phenotypic analysis also revealed that the majority of Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 1. PD-1 is expressed by activated/memory CD8 T lymphocytes in the blood of healthy human adults. A, Representative data from five subjects (i) and summary (ii) showing the ratio of naive (CD45RA+/CCR7+) versus memory subsets among CD8+/CD3+ T cells (n = 27). B, Percentage of PD-1 expression among naive versus memory subsets of CD8+/CD3+ T cells. C, Percentage of PD-1 expression among CD11ahi CD8+/CD3+ T cells. D, Per- centage of PD-1 expression among CCR7lo CD8+/CD3+ T cells. 4202 CHARACTERISTICS OF PD-1hi CD8 T CELLS IN HEALTHY HUMANS

PD-1–expressing cells do not show exhausted characteristics. Both streptomycin, and L-glutamine as described (27). All subjects were re- the gene array and phenotypic analysis demonstrated that the cruited at the Emory University Vaccine Center and gave informed con- sent for the study. majority of PD-1 is expressed by the effector memory (TEM) compared with terminally differentiated effector memory RA Isolation, amplification, and labeling of mRNA for gene array

T cells (TEMRA). Interestingly, we observed a relationship between hi lo + + CD45RA and PD-1 expression that CD45RA and PD-1 expression PD-1 , PD-1 , and naive human CD8 CD3 T cells were FACS sorted, and total RNA was isolated from cells in TRIzol (Invitrogen) according to levels were inversely related. CD8 T cells with the lowest CD45RA the manufacturer’s protocol. cDNA was synthesized using the SuperScript expression contained the highest proportion of PD-1hi cells and Choice cDNA synthesis (Invitrogen) and an oligo(dT) primer con- CD8 T cells with the highest CD45RA expression contained the taining a T7 promoter. The T7 MEGAscript kit (Ambion) was used to lowest proportion of PD-1hi cells. Our study suggests that expres- amplify cRNA from the cDNA. The cRNA was reverse transcribed with biotinylated nucleotides using the Enzo BioArray High Yield RNA tran- sion of PD-1 may be the signal of regulatory mechanisms of both script labeling kit (Enzo Life Sciences) in the second round of cRNA activated and exhausted cells in healthy human adults, and PD-1 synthesis, fragmented, and hybridized on Affymetrix U133 expression should not be regarded as a definitive marker for ex- Plus 2.0 arrays (Affymetrix) at the Vanderbilt University Microarray Shared hausted cells. Resource, according to the manufacturer’s protocols. Microarray analysis Materials and Methods Blood samples Gene expression data were summarized with RMA and merged according to each cell type (PD-1hi, PD-1lo, and naive). Each cell-type group was then Peripheral blood samples were obtained from healthy human adults (25–40 compared by t test (p , 0.05). The genes with low fold change values (1.6 y of age). PBMC were isolated from the blood using vacutainer cell cutoff) and normalized intensity values ,100 in both compared conditions Downloaded from preparation tubes (BD Biosciences, San Diego, CA). RBCs were lysed by were further removed to reduce false positivity. Each group’s signature incubation with ACK lysis buffer (Life Technologies) for 2 min at room genes were then cross examined to extract the unique gene sets among PD- temperature. After extensive washing, the PBMC were resuspended in 1hi, PD-1lo, and naive CD8 T cell subsets. These gene sets were examined RPMI 1640 medium containing 10% FCS supplemented with penicillin, through downstream analysis methods. Using the Database for Annotation, http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 2. Genome-wide microarray analysis of PD-1hi and PD-1lo CD8 T cells of healthy adults. A, Representative data showing sorting strategy of PD-1hi, PD-1lo, and naive subpopulations among CCR7loCD8+/CD3+ T cells. Expression level of mRNA was measured by microarray analysis of sorted PD-1hi, PD-1lo, and naive cells. B, Total number of transcripts differentially expressed (up- or downregulated; 1.6-fold cutoff) between the indicated comparison groups is shown. C, Relative intensities of the DEGs in any comparison with a 1.6-fold cutoff in naive, PD-1hi, and PD-1lo populations are plotted as heat maps to depict the relationship among these three populations. Expression level of each gene is represented by the number of standard deviations above (red) or below (green) the average value for that gene across all samples. D, Heat map showing relative intensities of the DEGs between PD-1hi and PD-1lo populations with a 1.6-fold cutoff. E, GSEA showing enriched profile of KIR gene set in PD-1lo cells. The Journal of Immunology 4203

Visualization, and Integrated Discovery, we categorized the biological meabilization buffer, and stained for cytokines using anti–IFN-g, TNF-a, processes in which the genes were involved, their molecular function, or IL-2, or MIP-1b. After washing, the cells were acquired on an FACSCa- related pathways. libur (BD Biosciences) or LSR II flow cytometer (BD Biosciences) using We have done two types of analyses with the data set as described (28– FACSDiva software (BD Immunocytometry Systems). Flow cytometry 34). 1) Comparative marker analysis: to test differentially expressed genes analysis was performed using FlowJo software (Tree Star, Ashland, OR). (DEGs) among PD-1hi and PD-1lo cells. For direct comparison of gene expression profiles of PD-1hi and PD-1lo cells, PD-1lo cell groups were Statistical analysis hi t used as controls for baseline transformation of PD-1 cell groups. The Comparison between groups (p value) was calculated using the Student t p test with value of 0.05 and fold change value of 1.6 with no correction test and Wilcoxon t test. All statistical analysis was performed using the was used. 2) Gene set enrichment analysis (GSEA): it is a method for GraphPad Prism program (GraphPad). determining whether a -ordered list of genes for a particular com- parison of interest is enriched in genes derived from an independently Microarray data generated gene set (e.g., PD-1hi and PD-1lo cells). GSEA provides an enrichment score that measures the degree of enrichment of a given gene Data have been deposited in the Gene Expression Omnibus under accession set at the top (highly correlated) or bottom (anticorrelated) of the second, number GSE26495 and can be viewed at: http://www.ncbi.nlm.nih.gov/geo/ rank-ordered data set (28–33). A nominal p value is used to assess the query/acc.cgi?acc=GSE26495. significance of the enrichment score. GSEA was performed with the following three different sets of genes: 1) Results exhausted gene sets, to examine if exhausted genes were enriched. These gene sets were those upregulated in LCMV exhausted (versus naive, ef- Healthy human adults show wide variation in the percentages fector, memory) and HIV progressors (versus elites) (32); 2) KIR gene set of naive and memory CD8 T cells (Table II); and 3) naive, TEM, and TEMRA subsets (34–37).

We examined the percentage of naive and memory CD8 T cells in Downloaded from Immunophenotyping the peripheral blood cells of healthy adults (n = 27). As shown in Fig. 1A, the ratio of naive versus memory CD8 T cells varied wide- For phenotypic analysis, 1 3 106 PBMC were stained with Abs for surface markers. Anti-human PD-1 (EH12, mouse IgG1) and anti-CD160 Abs ly among healthy individuals (Fig. 1A). The percentage of naive were kindly provided by G.J. Freeman, Dana-Farber Cancer Institute cells ranged from 10–60% (mean 35%) and memory cells from (Boston, MA). The following directly conjugated Abs were used: anti- 40–90% (mean 60%) of the total CD8 T cells (Fig. 1A). We then CD3, -CD8, -CD11a, -CD27, -CD28, -CD38, -CD45RA, -CD57, -CD95, examined PD-1 expression among naive and memory CD8 T cells. -CD127, -CD195, –HLA-DR, –Ki-67, –Bcl-2, -perforin, and –CTLA-4 http://www.jimmunol.org/ (BD Biosciences); anti-CCR7, -KIR, –LAG-3, and –Tim-3 (R&D Sys- Naive cells did not express PD-1. In contrast, 60% of memory cells tems, Minneapolis, MN); anti-granzyme B (Caltag Laboratories); and anti- expressed PD-1 (Fig. 1B). In addition, the majority of PD-1 was 2B4 (eBioscience). Anti–killer cell lectin-like receptor (KLR) G-1 Ab was expressed by CCR7lo/CD11ahi CD8 T cells (Fig. 1C,1D). These kindly provided by Dr. H. Pircher, University of Freiburg (Freiburg, Ger- data demonstrate that the majority of PD-1 was expressed by the many). Anti-a4b7 Ab was kindly provided by Dr. E. Butcher, Stanford memory CD8 T cells in the peripheral blood cells of healthy humans. University (Menlo Park, CA). For staining intracellular , cells were permeabilized using FACS permeabilization solution (BD Pharmingen) and Genome-wide microarray analysis of PD-1–expressing CD8 incubated with anti–Ki-67, –Bcl-2, -granzyme B, -perforin, or –CTLA-4 Ab T cells from the blood of healthy human adults according to the manufacturer’s protocol (BD Biosciences). To study the characteristics of PD-1–expressing CD8 T cells in Intracellular cytokine analysis by guest on September 25, 2021 healthy human adults, we performed gene array analysis. An ex- For intracellular cytokine staining, fresh PBMC were stimulated in vitro for ample of postsort analysis of CCR7loPD-1hi and CCR7loPD-1lo 6 h with plate-bound anti-CD3/CD28 or PMA/ionomycin in 96-well round- CD8 T cells is shown (Fig. 2A). We used highly purified PD-1hi, bottom plates in the presence of brefeldin A (1 ml/ml). The cells were lo washed once with FACS buffer and stained with relevant T cell markers for PD-1 , and naive CD8 T cell populations from six healthy adult 30 min at room temperature. Then the cells were fixed and permeabilized individuals for gene expression studies using Affymetrix Human with the Cytofix/Cytoperm kit (BD Biosciences), washed twice with per- genome U133 Plus 2.0 arrays (Affymetrix) containing ∼54,000

Table I. Genes upregulated on CCR7loPD-1hi versus CCR7loPD-1lo CD8 T cells of healthy human adults

Affymetrix Gene Name Symbol Biological Process/Function/Pathway Fold Change Identification No. Costimulation CD28 molecule CD28 T cell costimulation 2.91 206545_at CD27 molecule CD27 T cell costimulation 2.84 206150_at Cytotoxic T lymphocyte- associated 4 CTLA-4 Inhibitory receptor 2.54 236341_at Signaling and transcription factors Regulator of G-protein signaling 1 RGS1 Signal transduction 2.55 205645_at Signal-regulatory protein g SIRPG Signal transduction 2.47 220485_s_at Multiple EGF-like domains 6 MEGF6 Cell growth, proliferation and differentiation 2.44 226869_at PAS domain containing serine/threonine kinase PASK Intracellular signaling 2.11 213534_s_at Zinc finger protein 512B ZNF512B Transcriptional regulation 1.77 55872_at Homing and cell adhesion Chemokine (C-X-C motif) receptor 6 CXCR6 Homing 2.42 206974_at Chemokine (C-X-C motif) receptor 4 CXCR4 Homing 1.74 211919_at Effector function Granzyme K (granzyme 3; tryptase II) GZMK Granulocyte-mediated immunity; apoptosis 2.16 206666_at Cell proliferation Integral 2A ITM2A Cell differentiation 1.79 202746_at Miscellaneous Sphingosine 1-phosphate phosphatase 2 SGPP2 Phosphohydrolase activity 1.86 244780_at Tetratricopeptide repeat domain 9 TTC9 Tumor cell metastasis 1.80 213172_at Membrane-bound O-acyltransferase MBOAT1 Phospholipid remodeling 1.76 227379_at domain containing 1 Table II. Genes downregulated on CCR7 loPD-1hi versus CCR7 loPD-1lo CD8 T cells of healthy human adults 4204

Biological Affymetrix Gene Name Symbol Process/Function/Pathway Fold Change Identification No. TCR signaling and T cell activation TYRO protein tyrosine kinase binding protein TYROBP (DAP12) T cell activation/ZAP70 23.85 204122_at binding Src homology 2 domain containing 1B SH2D1B Signal transduction 24.55 1553177_at TCRa locus/TCRd locus TRA/TRD T cell activation 24.50 217143_s_at v-yes-1 Yamaguchi sarcoma viral-related oncogene homolog LYN T cell activation 22.57 202626_s_at Linker for activation of T cells family, member 2 LAT2 Ras-MAPK pathway Ca+ 22.00 221581_s_at signaling RAP2A, member of RAS oncogene family RAP2A Signal transduction 21.91 221830_at Vav 3 guanine nucleotide exchange factor VAV3 Actin cytoskeleton 21.78 218807_at rearrangement pathways Killer-cell Ig-like receptors Killer cell Ig-like receptor, three domains, long cytoplasmic tail, 2 KIR3DL2 Cytotoxicity 24.50 207314_x_at Killer cell lectin-like receptor subfamily C, member 3 KLRC3 Cytotoxicity 23.53 207723_s_at Killer cell Ig-like receptor, three domains, long/short cytoplasmic tail, 1 KIR3DL1/KIR3DS1 Cytotoxicity 23.40 211389_x_at Killer cell Ig-like receptor, two domains, long cytoplasmic tail, 2/3 KIR2DL2/3 Cytotoxicity 23.28 211532_x_at and short cytoplasmic tail 1/2/4/5 KIR2DS1/2/4/5 Killer cell lectin-like receptor subfamily C, member 1/2 KLRC1/2 Cytotoxicity 22.76 206785_s_at Killer cell lectin-like receptor subfamily F, member 1 KLRF1 Cytotoxicity 22.51 220646_s_at Killer cell lectin-like receptor subfamily C, member 4 KLRC4 Cytotoxicity 22.31 210690_at PD-1 OF CHARACTERISTICS Killer cell Ig-like receptor, two domains, short cytoplasmic tail, 3 KIR2DS3 Cytotoxicity 22.07 208122_x_at Killer cell lectin-like receptor subfamily D, member 1 KLRD1 Cytotoxicity 22.04 207795_s_at Killer cell Ig-like receptor, two domains, long cytoplasmic tail, 1 KIR2DL1 Cytotoxicity 21.88 210890_x_at Killer cell Ig-like receptor, two domains, long cytoplasmic tail, 4 KIR2DL4 Cytotoxicity 21.80 208426_x_at Natural cytotoxicity triggering receptor 1 NCR1 Cytotoxicity 21.71 207860_at Killer cell Ig-like receptor, two domains, long cytoplasmic tail, 5A KIR2DL5A Cytotoxicity 21.68 211410_x_at Cell adhesion Neural cell adhesion molecule 1 NCAM1 Cell adhesion 25.88 212843_at Integrin aM (complement component 3 receptor 3 subunit) ITGAM (CD11b) Cell adhesion 22.24 205786_s_at Palladin, cytoskeletal-associated protein PALLD Cell adhesion 22.13 200907_s_at Integrin aX (complement component 3 receptor 4 subunit) ITGAX (CD11c) Cell adhesion 22.06 210184_at Transcription factor IKAROS family zinc finger 2 IKZF2 Transcription factor 22.60 231929_at hi Kruppel-like factor 11 KLF11 Apoptosis 1.61 218486_at HUMANS HEALTHY IN CELLS T CD8 Zinc finger protein 683 ZNF683 Transcriptional regulation 22.32 230756_at Transcription factor CP2-like 1 TFCP2L1 Transcriptional regulator 21.93 227642_at Methyltransferase like 7A METTL7A Transmethylation 22.66 207761_s_at Effector function Granulysin GNLY T cell /NK cell cytotoxicity 23.19 205495_s_at Granzyme B (granzyme 2, cytotoxic T lymphocyte-associated GZMB Cytolytic activity 21.90 210164_at serine esterase 1) Protein metabolism and transport Myosin VI MYO6 Intracellular trafficking 22.04 203215_s_at Golgi integral membrane protein 4 GOLIM4 Protein trafficking 21.76 238002_at Chemokines and cytokines IL-8R, b IL8RB (CXCR2) Cytokine/chemokine- 22.30 207008_at mediated immunity IL-7 IL7 Cytokine 21.84 206693_at Chemokine-like receptor 1 CMKLR1 21.69 207652_s_at

(Table continues)

Downloaded from from Downloaded http://www.jimmunol.org/ by guest on September 25, 2021 25, September on guest by h ora fImmunology of Journal The Table II. (Continued)

Biological Affymetrix Gene Name Symbol Process/Function/Pathway Fold Change Identification No. Cell cycle and differentiation C-terminal binding protein 2 CTBP2 Cell localization 22.74 201218_at BCL2-related ovarian killer BOK Cell cycle 21.84 223349_s_at Ras association (RalGDS/AF-6) domain family member 4 RASSF4 Cell cycle and apoptosis 21.83 226436_at Platelet-derived growth factor D PDGFD Cell growth/differentiation 21.65 219304_s_at G protein-coupled receptor 56 GPR56 Cell–cell interaction 21.60 212070_at Miscellaneous Hepatoma-derived growth factor, related protein 3 HDGFRP3 DNA synthesis/cell 22.91 209524_at proliferation Adrenergic, b-1-, receptor ADRB1 Angiogenesis 22.70 229309_at Protease, serine, 23 PRSS23 Protein metabolism 22.24 229441_at Golgi membrane protein 1 GOLM1 Tumor biomarker 22.10 217771_at Galactosidase, b 1-like 2 GLB1L2 22.09 213713_s_at Transmembrane 6 superfamily member 1 TM6SF1 Transmembrane protein 22.09 219892_at Rho-related BTB domain containing 3 RHOBTB3 22.04 225202_at Synaptogyrin 1 SYNGR1 Gene for schizophrenia 21.99 210613_s_at Oxysterol binding protein-like 5 OSBPL5 Intracellular lipid receptor 21.91 223464_at Tetraspanin 2 TSPAN2 Transmembrane protein 21.90 227236_at Transmembrane and coiled-coil domain family 3 TMCC3 Transmembrane protein 21.88 226489_at Leukocyte Ig-like receptor, subfamily B (with TM and ITIM domains), member 1 LILRB1 IgR 21.88 229937_x_at Dedicator of cytokinesis 5 DOCK5 Autosomal recessive gene 21.85 230263_s_at Arrestin, b 1 ARRB1 21.81 222912_at Phosphodiesterase 4A, cAMP-specific (phosphodiesterase E2 dunce homolog, PDE4A 21.80 204735_at Drosophila) PGD2 synthase (brain) PTGDS 21.79 212187_x_at Fc fragment of IgG, low-affinity IIIb, receptor (CD16b) FCGR3B CD16; FcgR 21.65 204007_at Carboxylesterase 1 (monocyte/macrophage serine esterase 1) CES1 21.65 209616_s_at Basonuclin 2 BNC2 Keratinocyte transcription factor 21.63 238478_at Synovial sarcoma, X breakpoint 2 interacting protein SSX2IP 21.62 203017_s_at Matrix-remodeling associated 7 MXRA7 21.62 235836_at

4205

Downloaded from from Downloaded http://www.jimmunol.org/ by guest on September 25, 2021 25, September on guest by 4206 CHARACTERISTICS OF PD-1hi CD8 T CELLS IN HEALTHY HUMANS human transcripts. PD-1hi and PD-1lo subsets exhibited differen- tial expression (upregulated or downregulated) of 184 genes (Fig. 2B). Among them, 54 genes were expressed at relatively higher levels in PD-1hi cells and 130 genes in PD-1lo cells (presented in Tables I, II). Relative to naive cells, PD-1hi cells differentially expressed 2645 genes, and PD-1lo cells differentially expressed 2964 genes (Fig. 2B). Fig. 2C shows heat map analysis of 3328 DEGs in any comparison among naive, PD-1hi, and PD-1lo sub- sets. These data demonstrate that gene expression profiles of PD- 1hi and PD-1lo subsets were closely related. However, these two subsets were different from naive CD8 T cells in terms of their global gene expression profiles (Fig. 2C). Direct comparison of FIGURE 3. PD-1hi CD8 T cells of healthy adults do not show enrich- hi lo PD-1 and PD-1 subsets showed 184 differentially expressed ment of global exhausted PD-1hi gene signature from LCMV- or HIV- (upregulated or downregulated) genes (Fig. 2D). specific CD8 T cells as previously published (29, 32). GSEA showing A list of genes that are up- and downregulated on PD-1hi cells is enrichment profile of exhausted (LCMV and HIV) signature genes in PD- presented in Tables I and II. Compared to PD-1lo cells, PD-1hi 1hi and PD-1lo data sets of healthy adults (p . 0.05, NS). cells upregulated expression of genes encoding costimulatory (CD28 and CD27) and coinhibitory (CTLA-4) receptors, effector for each subset to run GSEA to test how these unique gene sets are molecule (granzyme K), cell signaling and proliferation molecules enriched in our data set comparing PD-1hi versus PD-1lo data sets. Downloaded from (PASK and E2F3), and homing receptors (CXCR6, CXCR4, and As expected, the naive gene set did not enrich in either PD-1hi or CCR5). In addition, PD-1hi cells also show downregulation of PD-1lo data sets (Fig. 4). In contrast, we found significant enrich- hi some positive TCR signaling molecules such as TYROBP, LAT2, ment of the TEM gene set in PD-1 cells (p = 0.035) and the TEMRA LYN, and VAV3. gene set in PD-1lo cells (p = 0.005) (Fig. 4). One of the striking observations in this study is that all of the

killer cell Ig-like receptors (KIRs) and KLRs were completely http://www.jimmunol.org/ downregulated on PD-1hi cells (Table II). We tested enrichment of genes encoding all of the KIRs on PD-1lo cells using GSEA as des- cribed in the Materials and Methods (28–33). For this, we compiled a set of KIR genes from the Database for Annotation, Visualiza- tion, and Integrated Discovery. Then we determined if this curated KIR gene set was enriched in genes that were increased in ex- pression in PD-1hi versus PD-1lo CD8 T cells of healthy subjects. As shown in Fig. 2E, we confirmed highly significant (p , 0.001) enrichment of genes encoding KIRs on PD-1lo cells (Fig. 2E). by guest on September 25, 2021 We next tested whether PD-1hi cells showed global similarity to exhausted cells by comparing gene expression profiles from PD-1hi CD8 T cells to previously published microarray data from exhaus- ted, virus-specific CD8 T cells in humans and LCMV mouse models (29, 32) (Fig. 3). We identified sets of genes upregulated in HIV- specific CD8 T cells from chronic progressors compared with HIV- specific CD8 T cells from controllers (32) and queried our PD-1hi versus PD-1lo data set for global enrichment of the exhausted pro- gressor signature (28–32). If PD-1hi cells of healthy subjects rep- resented a more exhausted population, one would predict that the exhausted gene set would be more enriched in the PD-1hi than PD- 1lo subset. However, GSEA showed no significant enrichment of exhausted gene signature in the PD-1hi subset compared with PD-1lo subset (Fig. 3). We repeated this analysis with a set of genes up- regulated in exhausted LCMV-specific CD8 T cells in mice and found no enrichment of exhausted signature genes in our PD-1hi subset from healthy individuals (Fig. 3). These findings suggest that PD-1hi CD8 T cells from healthy individuals do not share a global gene expression pattern with exhausted CD8 T cells in chronic viral infection. hi We then examined the relationship between T cell differentiation FIGURE 4. TEM and TEMRA gene signatures are enriched in PD-1 and lo state and PD-1 expression. For this, we identified sets of genes that PD-1 cells, respectively. Heat map shows unique gene sets of naive, TEM, were characteristic of human memory CD8 T cell subsets—naive, and TEMRA gene sets (34) (left panel). Three modules of unique gene T , and T —from the published data (34). For each class of sets—naive, TEM, and TEMRA—comprising 681, 53, and 226 genes, re- EM EMRA spectively, are shown (middle panel). GSEA was applied to PD-1 data set samples, we identified unique sets of upregulated genes; for ex- hi lo , , using unique sets to compare PD-1 and PD-1 cells (right panel). p ample, naive unique genes are those specifically upregulated (p 0.05, significant. Left panel reprinted from Willinger, T., T. Freeman, H. 0.05) in the naive subset compared with TEM and TEMRA. Similarly, Hasegawa, A. J. McMichael, and M. F. Callan. 2005. Molecular signatures we obtained specifically expressed unique genes by TEM and TEMRA distinguish human central memory from effector memory CD8 T cell subsets. We found that 681 of naive, 53 of TEM, and 226 of TEMRA subsets. J. Immunol. 175: 5895–5903. Copyright 2005. The American were unique genes (Fig. 4). Then we created three custom modules Association of Immunologists, Inc. The Journal of Immunology 4207

Phenotypic analysis of PD-1–expressing CD8 T cells from the characteristics of CCR7loPD-1hi or CCR7loPD-1lo in the context of blood of healthy adults naive (CD45RA+CCR7+) cells (Fig. 5B). Recent evidence from chronic LCMV infection in mice and HIV, We analyzed coexpression of PD-1 and several T cell differ- HBV, or HCV infections in humans indicates that upregulation entiation markers (26, 35–40). CD27 and CD28 are costimulatory of PD-1 correlates with impaired CD8 T cell phenotypic and receptors involved, respectively, in the generation of Ag-primed functional properties (14–19). However, in our study, microarray cells and the regulation of T cell activation (39, 40). In this study, hi analysis of PD-1hi CD8 T cells from healthy adults demonstrated we observed that PD-1 CD8 T cells in healthy adults expressed lo that these cells did not exhibit exhausted characteristics of PD-1hi both CD28 and CD27 at higher levels compared with PD-1 CD8 CD8 T cells of LCMV or HIV. Hence, we carried out phenotypic T cells. analysis to correlate the results of microarray analysis. The gating It has been well demonstrated that CD8 T cells downregulate strategy of PD-1hi, PD-1lo, and naive CD8 T cells from a repre- cytokine receptor CD127 (IL-7Ra) expression as they differentiate sentative subject is shown in Fig. 5A. We compared the phenotypic following Ag encounter and selectively re-express on a subset Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 5. Phenotypic analysis of PD-1hi, PD-1lo, and naive CD8 T cells of healthy adults. A, An example of gating strategy of PD-1hi and PD-1lo CD8+/ CD3+ T cells that are CCR7 negative is shown. Naive cells (CD45RA+/CCR7+) that do not express PD-1 are shown. B, Flow cytometry analysis showing percent expression and mean fluorescence intensity (MFI) of indicated markers on PD-1hi and PD-1lo CCR7loCD8+/CD3+ T cells in comparison with naive CD8+/CD3+ T cells of the corresponding donors are shown (n = 5–15). The percentage of CD8 T cells expressing the relevant receptor for each individual (open circle) and the group mean (horizontal line) are shown. 4208 CHARACTERISTICS OF PD-1hi CD8 T CELLS IN HEALTHY HUMANS destined to form precursors of the memory pool (41–43). We expression of antiapoptotic protein Bcl-2 was expressed at lower observed substantial expression of CD127 on memory cells, levels on PD-1hi cells compared with PD-1lo cells, suggesting that though not as high as by naive cells. Among the memory cells, PD-1hi cells are prone to apoptosis. Naive cells showed expression PD-1hi cells showed higher expression levels of CD127 compared of higher levels of Bcl-2. Opposing expression of PD-1 and CD57 with PD-1lo cells. It was reported that memory cells retain ex- during CD8 T cell maturation was described, with high-level PD-1 pression of activation marker CCR5, although not as high as ac- expression having an impact upon ex vivo sensitivity to apoptosis tivated cells (38). Our study also showed that CD8 T cells of (47). However, both PD-1hi and PD-1lo CD8 T cells of healthy healthy humans expressed CCR5; however, there was no differ- humans express similar levels of CD57. In addition, PD-1 ex- ence in expression levels among PD-1hi and PD-1lo CD8 T cells. pression was linked to a proapoptotic phenotype characterized by Cytolytic potential of the effector T cells is characterized by the high expression of CD95/Fas (48). Consistent with this, our study expression of granzyme B and perforin. Our results showed that showed positive correlation between PD-1 and CD95 expression both PD-1hi and PD-1lo CD8 T cells of healthy humans express among CD8 T cells of healthy humans. similar levels of granzyme B. However, perforin expression was In mice, naive CD8 T cells express moderate levels of a4b7 lower among PD-1hi CD8 T cells compared with PD-1lo CD8 integrin, which is required for homing to Peyer’s patches (49), T cells of healthy humans. It has been shown that resting cells whereas recently activated effector CD8 T cells express high levels do not express Ki-67, whereas cycling or very recently divided of a4b7 and migrate to effector sites including small intestinal T cells upregulated Ki-67 expression (44). Consistent with this, epithelium (50). We found that a4b7 expression delineated two PD-1hi or PD-1lo CD8 T cells of healthy subjects do not express populations among CCR7lo CD8 T cells in healthy adult human

Ki-67. In addition, the absence of Ki-67 expression clearly cor- PBMC: a subset that lacked a4b7 expression and another that Downloaded from related with the lack of CD38 and HLA-DR expression on PD-1hi expressed similar levels as naive CD8 T cells. PD-1hi CD8 T cells or PD-1lo CD8 T cells in our study. were enriched for a4b7+ cells, suggesting that PD-1hi cells may KLRG-1 is an NK cell receptor expressed by T cells that exhibit have different trafficking patterns than PD-1lo CD8 T cells. terminal effector properties (45, 46). Our results showed that both It has been shown that naive cells express CD45RA, its ex- PD-1hi and PD-1lo cells express similar levels of KLRG-1. The pression is downregulated during effector differentiation, and it is

re-expressed on resting memory cells in the absence of further http://www.jimmunol.org/ restimulation (38, 51). Our results showed that PD-1hi CD8 T cells

p=0.05

% 2B4

% CTLA-4 by guest on September 25, 2021

% LAG-3

% Tim-3

p=0.06

% KIR

% CD160

pp=0.01=0.01 p=0.01

% KIR3DL1

% KIR3DL3

FIGURE 6. Coexpression of PD-1 and other inhibitory receptors on FIGURE 7. No direct association between PD-1 expression and cyto- CD8 T cells of healthy adults. Flow cytometry analysis showing expression kine production by CD8 T cells of healthy adults. PBLs from healthy of inhibitory receptors on PD-1hi, PD-1lo, and naive CD8+/CD3+ T cells of individuals were stimulated with anti-CD3/CD28 for 6 h at 37˚C, and healthy human adults (n = 5). CTLA-4 staining was done intracellularly intracellular staining of IFN-g, TNF-a, IL-2, and MIP-1b was done. following anti-CD3/CD28 stimulation. The percentage of CD8 T cells Representative data from one subject (of 16 healthy adults tested) showing expressing the relevant receptor for each individual (open circle) and the percentage of IFN-g, TNF-a, IL-2, and MIP-1b secretion by naive, PD-1hi, group mean (horizontal line) are shown. and PD-1lo CD8+/CD3+ T cell subsets. The Journal of Immunology 4209 downregulate CD45RA compared with PD-1lo CD8 T cells of Expression of PD-1 does not affect the ability of CD8 T cells in healthy humans. healthy adults to secrete cytokines High-level expression of PD-1 was associated with a diminished Coexpression of PD-1 and other inhibitory receptors on the ability of LCMV-specific CD8 T cells to secrete cytokines in CD8 T cells of healthy human adults chronically infected mice (13). However, human HIV-specific CD8 Coexpression of multiple inhibitory receptors during chronic T cells showed that PD-1 expression was not directly associated LCMV infection and their association with greater T cell ex- with impaired cytokine secretion (15). In this study, we tested if haustion was reported (52). Comparing global gene expression stimulation of PBMC with anti-CD3/CD28 shows any difference profiles of exhausted CD8 T cells to functional LCMV-specific in the secretion (or lack thereof) of IFN-g, TNF-a, IL-2, or MIP- effector and memory CD8 T cells revealed upregulation of a 1b by PD-1hi or PD-1lo CD8 T cells. We stimulated total CD8 number of inhibitory receptor genes in addition to PD-1 (29). T cells with anti-CD3/CD28 for 6 h at 37˚C and then assessed In addition, another study demonstrated that HIV-specific CD4 cytokine secretion by intracellular cytokine staining. We did not T cells coexpress PD-1 and another inhibitory molecule, CTLA-4 observe any difference in production of IFN-g, TNF-a, IL-2, or (53). Hence, we tested if CD8 T cell responses in humans are MIP-1b between PD-1hi and PD-1lo CD8 T cells (Fig. 7). Our regulated by coexpression of other inhibitory receptors, including results indicate that PD-1 expression in healthy adults did not CTLA-4 (CD152), 2B4 (CD244) (54), LAG-3 (55), Tim-3 (56), affect the ability of CD8 T cells to secrete cytokines. CD160 (57), and KIR (CD158) (58–62) by flow cytometry. PBMC from healthy human adults were stained for surface and intracel- PD-1 is expressed by the majority of TEM CD8 cells in the lular expression of these receptors. Our results showed no signif- blood of healthy humans Downloaded from icant difference in expression levels of these receptors between Gene expression analysis showed that the majority of PD-1 was hi lo PD-1 or PD-1 CD8 T cells (Fig. 6). However, upon anti-CD3/ expressed by TEM compared with TEMRA (Fig. 4). To confirm this, CD28 stimulation, we found PD-1hi cells expressed significantly we assigned PD-1–expressing CD3+/CD8+ T cells into four sub- higher levels of expression of CTLA-4 compared with PD-1lo CD8 sets based on expression of CCR7, lymphoid tissue homing re- T cells (Fig. 6). ceptor, and CD45RA, a transmembrane tyrosine phosphatase: http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 8. Expression of PD-1 among CD8 T cell subsets of healthy adults. Representative data from one subject (A) and summary (B) showing percentage of PD-1 expression among naive, TCM, TEM, and TEMRA CD8+/CD3+ subsets (n = 27). Histograms demonstrate CCR72/CD45RA2 CD8 T cells express high level of PD-1 compared to CCR72/CD45RA+ cells. C, Representative datashowingdecreaseinPD-1expression (histograms) levels as the CD45RA level goes up in CCR7-CD8+CD3+ subsets. D, Summary of MFI and percentage PD-1 ex- pression among CCR7-CD8+CD3+ T cells. 4210 CHARACTERISTICS OF PD-1hi CD8 T CELLS IN HEALTHY HUMANS

+ + + 1) naive (CCR7 CD45RA ); 2) central memory (TCM: CCR7 with acute (influenza and vaccinia) infections (p , 0.05) (Fig. 9B). 2 2 2 CD45RA ); 3) effector memory (TEM: CCR7 CD45RA ); and These results also highlight correlation of PD-1 expression with 2 + 4) terminal effectors (TEMRA: CCR7 CD45RA ) (35–37). As viral Ag persistence. shown in Fig. 8A and 8B, naive cells did not express PD-1. In contrast, 60% of TEM expressed PD-1, and approximately one Discussion third of TEMRA express PD-1. We also observed about one fourth The T cell response is regulated by balance between costimulatory of TCM expressed PD-1. and coinhibitory signals. T cell exhaustion was initially charac- To further examine PD-1 levels among TEM and TEMRA CD8 terized in LCMV model in mice, and these findings were extended T cells, we further subdivided CCR7lo CD8 T cells into four to human chronic infections. All of these data demonstrate that PD- populations based on the levels of CD45RA expression. We ob- 1 is highly expressed on chronic virus-specific CD8 T cells, and its served an interesting relationship between CD45RA and PD-1 ex- upregulation is correlated with high viral load and associated with pression in that their expression levels are inversely related (Fig. a high level of viremia. In contrast to continuous productive 8C–E). Specifically, CD8 T cells with the lowest CD45RA ex- replication of HIV, HBV, or HCV, common human herpesviruses pression contained the highest proportion of PD-1+ cells, and CD8 such as EBV and CMV are characterized by a stable cell–virus T cells with the highest CD45RA expression contained the lowest relationship and CD8 T cell-mediated control. PD-1 is not only proportion of PD-1+ cells. Overall, these data demonstrate that the expressed on exhausted T cells but also transiently expressed on majority of PD-1 is expressed by the TEM CD8 T cells in the activated T cells. The mechanism of PD-1 regulation in activated peripheral blood cells of healthy humans. and exhausted cells is still poorly defined. In this study, we per-

formed an in-depth analysis of PD-1hi and PD-1lo CD8 T cells to Downloaded from PD-1 is expressed at low to moderate levels on understand whether PD-1hi cells in healthy adult humans represent cytomegalovirus- and EBV-specific CD8 T cells exhausted cells. Numerous studies report high-level expression of PD-1 on CD8 We found that the ratio of naive and memory CD8 T cells vary T cells specific for HIV, HBV, and HCV infections (14–20, 32). In widely among healthy adults. Also, PD-1 expression level in this study, we analyzed expression of PD-1 on EBV-, cytomegalo- memory cells vary among these subjects from 40 to 80% (Fig. 1B).

virus (CMV)-, influenza-, and vaccinia virus-specific CD8 T cells in This variation in PD-1 expression could be because of the flux http://www.jimmunol.org/ the PBMC of healthy seropositive humans. We observed that these occurring between maturation steps depending on the time of viral cells express low levels of PD-1 compared to the previously pub- reactivation or in response to changes in viral load. When the Ag lished reports on HIV-, HBV-, or HCV-specific CD8 T cells (14–19). is low, cells are PD-1lo. During recrudescence, some of these PD- We also observed varying levels of PD-1 expression on memory 1lo cells are recruited back to effector phenotype, resulting in CD8 cells of different specificities. EBV-specific CD8 T cells upregulation of PD-1 expression. Thus, in response to fluctuations expressed higher levels of PD-1 (Fig. 9) than CD8 T cells specific to in the Ag load, change in the distribution of PD-1hi CD8 T cells CMV. In contrast, influenza virus-specific memory CD8 T cells across different subsets can occur. expressed low levels of PD-1, and vaccinia virus-specific CD8 We compared gene expression profiles of PD-1hi CD8 T cell T cells rarely expressed PD-1. Hence, memory CD8 T cells specific subset of healthy humans with PD-1hi exhausted signatures of by guest on September 25, 2021 for chronic infections (EBV and CMV) expressed PD-1 compared LCMV- or HIV-specific CD8 T cells. We found that genes char- acteristic of functionally exhausted CD8 T cells were not enriched in PD-1hi CD8 T cells in healthy humans. This suggests that the PD-1hi CD8 T cell subset in the peripheral blood of healthy human subjects does not share global similarity in gene expression to exhausted T cells. However, the PD-1hi compartment consists of a heterogeneous set of Ag-experienced cells, and we therefore cannot exclude the possibility that a minority of this population may share features of exhaustion. The phenotypic analysis of PD-1hi CD8 T cells of healthy hu- man adults does not exhibit exhausted characteristics. For exam- ple, PD-1hi CD8 T cell responses to chronic viral infections, such as HIV, in humans report progressive downregulation of CD127 and CD28 (14–19). This phenotype is likely a direct consequence of chronic stimulation in a situation of high Ag load (63, 64). In contrast, healthy adults carry asymptomatic EBV and CMV that are characterized by low level of stimulation, Ag load, and re- currence, which do not downregulate CD127 and CD28 but main- tain PD-1 at low to intermediate levels (Fig. 9). In addition, our data show that expression of PD-1 had no direct effect upon the ability of human CD8 T cells to produce cytokines. FIGURE 9. PD-1 is expressed at low to intermediate levels by CD8 Gene expression profiles using gene sets representative of naive, T cells specific for common chronic viral infections (EBV and CMV). A, TEM, and TEMRA show that the majority of PD-1 is expressed by Representative PD-1 staining on EBV, CMV, influenza, and vaccinia virus- T compared with T CD8 T cells. Recent studies empha- specific CD8+/CD3+ T cells. Peptide epitopes are abbreviated to the first EM EMRA sized the heterogeneity of effector and memory cell populations three amino acids. Percentage and MFI of PD-1 expression among tetra- + with the description of multiple cellular subsets based on pheno- mer cells are indicated. B, Summary of percentage and MFI of PD-1 + expression on CMV- and EBV-specific CD8 T cells in comparison with type and function. Ag-experienced human CD8 Tcellswere influenza and vaccinia virus-specific CD8 T cells from healthy subjects is delineated into naive, TCM,TEM, and TEMRA subsets based on the shown. Individual results from each subject (individual symbols) and patterns of CD45RA and CCR7 expression (35–37). Previous group mean (horizontal line) are shown. studies report that TEMRA phenotype is associated with terminal The Journal of Immunology 4211 effector cells with little proliferative capacity, sensitive to apo- hepatitis B virus (HBV)-specific T-cell dysfunction in chronic HBV infection. J. Virol. 81: 4215–4225. ptosis, and associated with senescence. In contrast, this study 19. Radziewicz, H., C. C. Ibegbu, M. L. Fernandez, K. A. Workowski, K. Obideen, shows that PD-1 is predominantly expressed by TEM rather than M. Wehbi, H. L. Hanson, J. P. Steinberg, D. Masopust, E. J. Wherry, et al. 2007. TEMRA. In fact, we recently showed that yellow fever virus-spe- Liver-infiltrating lymphocytes in chronic human hepatitis C virus infection display an exhausted phenotype with high levels of PD-1 and low levels of cific memory CD8 T cells had a TEMRA phenotype (38). Hence, it CD127 expression. J. Virol. 81: 2545–2553. is also possible that the low-level expression of PD-1 by TEMRA 20. Velu, V., K. Titanji, B. Zhu, S. Husain, A. Pladevega, L. Lai, T. H. Vanderford, could be associated with the presence of memory CD8 T cell L. Chennareddi, G. Silvestri, G. J. Freeman, et al. 2009. Enhancing SIV-specific immunity in vivo by PD-1 blockade. Nature 458: 206–210. populations that do not express PD-1. 21. Dong, H., S. E. Strome, D. R. Salomao, H. Tamura, F. Hirano, D. B. Flies, In conclusion, our study shows that most of the PD-1–expressing P. C. Roche, J. Lu, G. Zhu, K. Tamada, et al. 2002. Tumor-associated B7-H1 cells in healthy human adults do not represent characteristics of promotes T-cell apoptosis: a potential mechanism of immune evasion. Nat. Med. 8: 793–800. exhausted cells. 22. Thompson, R. H., H. Dong, and E. D. Kwon. 2007. Implications of B7-H1 expression in clear cell carcinoma of the kidney for prognostication and ther- apy. Clin. Cancer Res. 13: 709s–715s. Acknowledgments 23. Ahmadzadeh, M., L. A. Johnson, B. Heemskerk, J. R. Wunderlich, M. E. Dudley, We thank R. Karaffa, S. Durham, and S. Mertens for assistance with FACS D. E. White, and S. A. Rosenberg. 2009. Tumor antigen-specific CD8 T cells and members of the Ahmed laboratory for helpful discussions. infiltrating the tumor express high levels of PD-1 and are functionally impaired. Blood 114: 1537–1544. 24. Francisco, L. M., P. T. Sage, and A. H. Sharpe. 2010. The PD-1 pathway in Disclosures tolerance and autoimmunity. Immunol. Rev. 236: 219–242. G.J.F. draws royalties from patents regarding PD-1. The authors have no 25. Okazaki, T., and T. Honjo. 2006. Rejuvenating exhausted T cells during chronic

viral infection. Cell 124: 459–461. Downloaded from other financial conflicts of interest. 26. Miller, J. D., R. G. van der Most, R. S. Akondy, J. T. Glidewell, S. Albott, D. Masopust, K. Murali-Krishna, P. L. Mahar, S. Edupuganti, S. Lalor, et al. 2008. Human effector and memory CD8+ T cell responses to smallpox and References yellow fever vaccines. Immunity 28: 710–722. 1. Wherry, E. J., and R. Ahmed. 2004. Memory CD8 T-cell differentiation during 27. Brown, J. A., D. M. Dorfman, F. R. Ma, E. L. Sullivan, O. Munoz, C. R. Wood, viral infection. J. Virol. 78: 5535–5545. E. A. Greenfield, and G. J. Freeman. 2003. Blockade of programmed death-1 2. Freeman, G. J., and A. H. Sharpe. 2003. The PD-1: PD-1 ligand pathway. In The ligands on dendritic cells enhances T cell activation and cytokine production. J. Immunol. 170: 1257–1266.

B7-CD28 Family Molecules. L. Chen, ed. R.G. Landes, Austin, TX, p. 59–66. http://www.jimmunol.org/ 3. Freeman, G. J., E. J. Wherry, R. Ahmed, and A. H. Sharpe. 2006. Reinvigorating 28. Subramanian, A., P. Tamayo, V. K. Mootha, S. Mukherjee, B. L. Ebert, exhausted HIV-specific T cells via PD-1-PD-1 ligand blockade. J. Exp. Med. M. A. Gillette, A. Paulovich, S. L. Pomeroy, T. R. Golub, E. S. Lander, and 203: 2223–2227. J. P. Mesirov. 2005. Gene set enrichment analysis: a knowledge-based approach 4. Sharpe, A. H., E. J. Wherry, R. Ahmed, and G. J. Freeman. 2007. The function of for interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. USA programmed cell death 1 and its ligands in regulating autoimmunity and in- 102: 15545–15550. fection. Nat. Immunol. 8: 239–245. 29. Wherry, E. J., S. J. Ha, S. M. Kaech, W. N. Haining, S. Sarkar, V. Kalia, 5. Okazaki, T., and T. Honjo. 2006. The PD-1-PD-L pathway in immunological S. Subramaniam, J. N. Blattman, D. L. Barber, and R. Ahmed. 2007. Molecular tolerance. Trends Immunol. 27: 195–201. signature of CD8+ T cell exhaustion during chronic viral infection. Immunity 27: 6. Sharpe, A. H., and G. J. Freeman. 2002. The B7-CD28 superfamily. Nat. Rev. 670–684. Immunol. 2: 116–126. 30. Haining, W. N., B. L. Ebert, A. Subrmanian, E. J. Wherry, Q. Eichbaum, 7. Chen, L. 2004. Co-inhibitory molecules of the B7-CD28 family in the control of J. W. Evans, R. Mak, S. Rivoli, J. Pretz, J. Angelosanto, et al. 2008. Identification T-cell immunity. Nat. Rev. Immunol. 4: 336–347. of an evolutionarily conserved transcriptional signature of CD8 memory dif- by guest on September 25, 2021 8. Chemnitz, J. M., R. V. Parry, K. E. Nichols, C. H. June, and J. L. Riley. 2004. ferentiation that is shared by T and B cells. J. Immunol. 181: 1859–1868. 31. Sarkar, S., V. Kalia, W. N. Haining, B. T. Konieczny, S. Subramaniam, and SHP-1 and SHP-2 associate with immunoreceptor tyrosine-based switch motif of R. Ahmed. 2008. Functional and genomic profiling of effector CD8 T cell programmed death 1 upon primary human T cell stimulation, but only receptor subsets with distinct memory fates. J. Exp. Med. 205: 625–640. ligation prevents T cell activation. J. Immunol. 173: 945–954. 32. Quigley, M., F. Pereyra, B. Nilsson, F. Porichis, C. Fonseca, Q. Eichbaum, 9. Nishimura, H., M. Nose, H. Hiai, N. Minato, and T. Honjo. 1999. Development B. Julg, J. L. Jesneck, K. Brosnahan, S. Imam, et al. 2010. Transcriptional of lupus-like autoimmune diseases by disruption of the PD-1 gene encoding an analysis of HIV-specific CD8+ T cells shows that PD-1 inhibits T cell function ITIM motif-carrying immunoreceptor. Immunity 11: 141–151. by upregulating BATF. Nat. Med. 16: 1147–1151. 10. Keir, M. E., S. C. Liang, I. Guleria, Y. E. Latchman, A. Qipo, L. A. Albacker, 33. Haining, W. N., and E. J. Wherry. 2010. Integrating genomic signatures for M. Koulmanda, G. J. Freeman, M. H. Sayegh, and A. H. Sharpe. 2006. Tissue immunologic discovery. Immunity 32: 152–161. expression of PD-L1 mediates peripheral T cell tolerance. J. Exp. Med. 203: 34. Willinger, T., T. Freeman, H. Hasegawa, A. J. McMichael, and M. F. Callan. 883–895. 2005. Molecular signatures distinguish human central memory from effector 11. Nishimura, H., T. Okazaki, Y. Tanaka, K. Nakatani, M. Hara, A. Matsumori, memory CD8 T cell subsets. J. Immunol. 175: 5895–5903. S. Sasayama, A. Mizoguchi, H. Hiai, N. Minato, and T. Honjo. 2001. Autoimmune 35. Sallusto, F., D. Lenig, R. Fo¨rster, M. Lipp, and A. Lanzavecchia. 1999. Two dilated cardiomyopathy in PD-1 receptor-deficient mice. Science 291: 319–322. subsets of memory T lymphocytes with distinct homing potentials and effector 12. Ansari, M. J., A. D. Salama, T. Chitnis, R. N. Smith, H. Yagita, H. Akiba, functions. Nature 401: 708–712. T. Yamazaki, M. Azuma, H. S. Iwai, S. J. Khoury, et al. 2003. The programmed 36. Appay, V., R. A. van Lier, F. Sallusto, and M. Roederer. 2008. Phenotype and death-1 (PD-1) pathway regulates autoimmune diabetes in nonobese diabetic function of human T lymphocyte subsets: consensus and issues. Cytometry A 73: (NOD) mice. J. Exp. Med. 198: 63–69. 975–983. 13. Barber, D. L., E. J. Wherry, D. Masopust, B. Zhu, J. P. Allison, A. H. Sharpe, 37. Zhang, J.-Y., Z. Zhang, X. Wang, J.-L. Fu, J. Yao, Y. Jiao, L. Chen, H. Zhang, G. J. Freeman, and R. Ahmed. 2006. Restoring function in exhausted CD8 J. Wei, L. Jin, et al. 2007. PD-1 up-regulation is correlated with HIV-specific T cells during chronic viral infection. Nature 439: 682–687. memory CD8+ T-cell exhaustion in typical progressors but not in long-term 14. Day, C. L., D. E. Kaufmann, P. Kiepiela, J. A. Brown, E. S. Moodley, S. Reddy, nonprogressors. Blood 109: 4671–4678. E. W. Mackey, J. D. Miller, A. J. Leslie, C. DePierres, et al. 2006. PD-1 ex- 38. Akondy, R. S., N. D. Monson, J. D. Miller, S. Edupuganti, D. Teuwen, H. Wu, pression on HIV-specific T cells is associated with T-cell exhaustion and disease F. Quyyumi, S. Garg, J. D. Altman, C. Del Rio, et al. 2009. The yellow fever progression. Nature 443: 350–354. virus vaccine induces a broad and polyfunctional human memory CD8+ T cell 15. Petrovas, C., J. P. Casazza, J. M. Brenchley, D. A. Price, E. Gostick, response. J. Immunol. 183: 7919–7930. W. C. Adams, M. L. Precopio, T. Schacker, M. Roederer, D. C. Douek, and 39. Hamann, D., P. A. Baars, M. H. Rep, B. Hooibrink, S. R. Kerkhof-Garde, R. A. Koup. 2006. PD-1 is a regulator of virus-specific CD8+ T cell survival in M. R. Klein, and R. A. van Lier. 1997. Phenotypic and functional separation of HIV infection. J. Exp. Med. 203: 2281–2292. memory and effector human CD8+ T cells. J. Exp. Med. 186: 1407–1418. 16. Trautmann, L., L. Janbazian, N. Chomont, E. A. Said, S. Gimmig, B. Bessette, 40. Romero, P., A. Zippelius, I. Kurth, M. J. Pittet, C. Touvrey, E. M. Iancu, M. R. Boulassel, E. Delwart, H. Sepulveda, R. S. Balderas, et al. 2006. Up- P. Corthesy, E. Devevre, D. E. Speiser, and N. Rufer. 2007. Four functionally regulation of PD-1 expression on HIV-specific CD8+ T cells leads to reversible distinct populations of human effector-memory CD8+ T lymphocytes. J. immune dysfunction. Nat. Med. 12: 1198–1202. Immunol. 178: 4112–4119. 17. Boettler, T., E. Panther, B. Bengsch, N. Nazarova, H. C. Spangenberg, 41. Kaech, S. M., J. T. Tan, E. J. Wherry, B. T. Konieczny, C. D. Surh, and H. E. Blum, and R. Thimme. 2006. Expression of the interleukin-7 receptor R. Ahmed. 2003. Selective expression of the interleukin 7 receptor identifies alpha chain (CD127) on virus-specific CD8+ T cells identifies functionally and effector CD8 T cells that give rise to long-lived memory cells. Nat. Immunol. 4: phenotypically defined memory T cells during acute resolving hepatitis B virus 1191–1198. infection. J. Virol. 80: 3532–3540. 42. Schluns, K. S., W. C. Kieper, S. C. Jameson, and L. Lefranc¸ois. 2000. 18. Boni, C., P. Fisicaro, C. Valdatta, B. Amadei, P. Di Vincenzo, T. Giuberti, Interleukin-7 mediates the homeostasis of naı¨ve and memory CD8 T cells D. Laccabue, A. Zerbini, A. Cavalli, G. Missale, et al. 2007. Characterization of in vivo. Nat. Immunol. 1: 426–432. 4212 CHARACTERISTICS OF PD-1hi CD8 T CELLS IN HEALTHY HUMANS

43. Sauce, D., M. Larsen, R. J. Abbott, A. D. Hislop, A. M. Leese, N. Khan, 53. Kaufmann, D.E., D.G.Kavanagh,F.Pereyra, J.J.Zaunders, E.W.Mackey,T.Miura, L. Papagno, G. J. Freeman, and A. B. Rickinson. 2009. Upregulation of in- S. Palmer, M. Brockman, A. Rathod, A. Piechocka-Trocha, et al. 2007. Upregulation terleukin 7 receptor alpha and programmed death 1 marks an epitope-specific of CTLA-4 by HIV-specific CD4+ T cells correlates with disease progression and CD8+ T-cell response that disappears following primary Epstein-Barr virus in- defines a reversible immune dysfunction. Nat. Immunol. 8: 1246–1254. fection. J. Virol. 83: 9068–9078. 54. McNerney, M. E., K. M. Lee, and V. Kumar. 2005. 2B4 (CD244) is a non-MHC 44. Gerdes, J., H. Lemke, H. Baisch, H. H. Wacker, U. Schwab, and H. Stein. 1984. binding receptor with multiple functions on natural killer cells and CD8+ T cells. Cell cycle analysis of a cell proliferation-associated human nuclear antigen Mol. Immunol. 42: 489–494. defined by the monoclonal antibody Ki-67. J. Immunol. 133: 1710–1715. 55. Triebel, F. 2003. LAG-3: a regulator of T-cell and DC responses and its use in 45. Voehringer, D., M. Koschella, and H. Pircher. 2002. Lack of proliferative ca- therapeutic vaccination. Trends Immunol. 24: 619–622. pacity of human effector and memory T cells expressing killer cell lectinlike 56. Golden-Mason, L., B. E. Palmer, N. Kassam, L. Townshend-Bulson, receptor G1 (KLRG1). Blood 100: 3698–3702. S. Livingston, B. J. McMahon, N. Castelblanco, V. Kuchroo, D. R. Gretch, and 46. Ibegbu, C. C., Y. X. Xu, W. Harris, D. Maggio, J. D. Miller, and A. P. Kourtis. H. R. Rosen. 2009. Negative immune regulator Tim-3 is overexpressed on T cells 2005. Expression of killer cell lectin-like receptor G1 on antigen-specific human in hepatitis C virus infection and its blockade rescues dysfunctional CD4+ CD8+ T lymphocytes during active, latent, and resolved infection and its relation and CD8+ T cells. J. Virol. 83: 9122–9130. with CD57. J. Immunol. 174: 6088–6094. 57. Cai, G., A. Anumanthan, J. A. Brown, E. A. Greenfield, B. Zhu, and 47. Petrovas, C., B. Chaon, D. R. Ambrozak, D. A. Price, J. J. Melenhorst, B. J. Hill, G. J. Freeman. 2008. CD160 inhibits activation of human CD4+ T cells through C. Geldmacher, J. P. Casazza, P. K. Chattopadhyay, M. Roederer, et al. 2009. interaction with herpesvirus entry mediator. Nat. Immunol. 9: 176–185. Differential association of programmed death-1 and CD57 with ex vivo survival 58. Ravetch, J. V., and L. L. Lanier. 2000. Immune inhibitory receptors. Science 290: of CD8+ T cells in HIV infection. J. Immunol. 183: 1120–1132. 84–89. 48. Mueller, Y. M., S. C. De Rosa, J. A. Hutton, J. Witek, M. Roederer, J. D. Altman, 59. Arlettaz, L., S. Degermann, C. De Rham, E. Roosnek, and B. Huard. 2004. and P. D. Katsikis. 2001. Increased CD95/Fas-induced apoptosis of HIV-specific Expression of inhibitory KIR is confined to CD8+ effector T cells and limits their CD8(+) T cells. Immunity 15: 871–882. proliferative capacity. Eur. J. Immunol. 34: 3413–3422. 49. Lefranc¸ois, L., C. M. Parker, S. Olson, W. Muller, N. Wagner, M. P. Scho¨n, and 60. Williams, A. P., A. R. Bateman, and S. I. Khakoo. 2005. Hanging in the balance. L. Puddington. 1999. The role of beta7 integrins in CD8 T cell trafficking during KIR and their role in disease. Mol. Interv. 5: 226–240. an antiviral immune response. J. Exp. Med. 189: 1631–1638. 61. Carrington, M., and M. P. Martin. 2006. The impact of variation at the KIR gene 50. Masopust, D., D. Choo, V. Vezys, E. J. Wherry, J. Duraiswamy, R. Akondy, cluster on human disease. Curr. Top. Microbiol. Immunol. 298: 225–257. Downloaded from J. Wang, K. A. Casey, D. L. Barber, K. S. Kawamura, et al. 2010. Dynamic T cell 62. Lazuardi, L., D. Herndler-Brandstetter, S. Brunner, G. T. Laschober, migration program provides resident memory within intestinal epithelium. J. G. Lepperdinger, and B. Grubeck-Loebenstein. 2009. Microarray analysis Exp. Med. 207: 553–564. reveals similarity between CD8+CD28- T cells from young and elderly persons, 51. Carrasco, J., D. Godelaine, A. Van Pel, T. Boon, and P. van der Bruggen. 2006. but not of CD8+CD28+ T cells. Biogerontology 10: 191–202. CD45RA on human CD8 T cells is sensitive to the time elapsed since the last 63. Shin, H., S. D. Blackburn, J. N. Blattman, and E. J. Wherry. 2007. Viral antigen antigenic stimulation. Blood 108: 2897–2905. and extensive division maintain virus-specific CD8 T cells during chronic in- 52. Blackburn, S. D., H. Shin, W. N. Haining, T. Zou, C. J. Workman, A. Polley, fection. J. Exp. Med. 204: 941–949.

M. R. Betts, G. J. Freeman, D. A. Vignali, and E. J. Wherry. 2009. Coregulation 64. Mueller, S. N., and R. Ahmed. 2009. High antigen levels are the cause of T cell http://www.jimmunol.org/ of CD8+ T cell exhaustion by multiple inhibitory receptors during chronic viral exhaustion during chronic viral infection. Proc. Natl. Acad. Sci. USA 106: 8623– infection. Nat. Immunol. 10: 29–37. 8628. by guest on September 25, 2021