T Cells This Information Is Current As of September 25, 2021

Human Cytomegalovirus Infection Induces a Rapid and Sustained Change in the Expression of NK Cell Receptors on CD8+ T Cells This information is current as of September 25, 2021. Amber van Stijn, Ajda T. Rowshani, Sila L. Yong, Frank Baas, Eddy Roosnek, Ineke J. M. ten Berge and René A. W. van Lier J Immunol 2008; 180:4550-4560; ; doi: 10.4049/jimmunol.180.7.4550 Downloaded from http://www.jimmunol.org/content/180/7/4550

References This article cites 36 articles, 21 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/180/7/4550.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 25, 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 © 2008 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Human Cytomegalovirus Infection Induces a Rapid and Sustained Change in the Expression of NK Cell Receptors on CD8؉ T Cells1

Amber van Stijn,2,3*† Ajda T. Rowshani,2† Sila L. Yong,2* Frank Baas,2‡ Eddy Roosnek,2§ Ineke J. M. ten Berge,2† and Rene´A. W. van Lier2*

The CD8؉ T cell compartment of human CMV-seropositive individuals characteristically contains a high proportion of cells that express NK cell receptors (NKRs) which may contribute to the surveillance of virus-infected cells. To test whether this enhanced expression is a direct and immediate result of CMV infection, we used DNA microarrays to analyze putative changes in the RNA expression level of NKRs in CMV-specific CD8؉ T cells of renal transplant recipients experiencing primary CMV infection. Already in the acute phase 39 of infection 29 NKRs were induced, of which 19 remained high 1 year after cessation of viral replication. Activating and inhibitory NKRs Downloaded from were induced to a similar extent. Detailed longitudinal flow cytometric analyses confirmed NKR changes at the protein level. Strikingly, a strong induction of CD94 on CD3؉ T cells was observed with surface expression of activating CD94dim NKG2C dimers appearing before inhibitory CD94bright NKG2A ones. After the acute phase of infection, the balance between inhibitory and activating receptors did not change. Thus, CMV infection induces a rapid and lasting change in the expression of NKRs on human CD8؉ T cells. The Journal of Immunology, 2008, 180: 4550–4560. http://www.jimmunol.org/ ytomegalovirus is a persistent ␤ herpes virus that in the contained within this distinctive subset (3). Frequently, CMV-spe- Western world infects 60–90% of the population. In im- cific CD8ϩ T cells in the latency stage lack the lymph node-hom- C munocompetent individuals, primary CMV infection is ing receptor CCR7 and the costimulatory receptors CD27 and infrequently diagnosed because of the lack of prototypical symp- CD28, but express CD57 and CD45RA. Longitudinal studies in toms. Immunocompromised patients, however, may experience se- kidney transplant recipients have indicated that the acquisition of vere complaints due to pulmonary and gastrointestinal pathology this phenotype by Ag-specific CD8ϩ T cells is a stepwise process (1). In the suppression of initial CMV replication and the mainte- that continues for an extended period after primary CMV infection nance of viral latency, both NK cells and T cells have been im- (8). Because these cells typically lack expression of the classical plicated (2). CMV appears to be unique among the persisting vi- costimulatory receptors, it has been suggested that other surface by guest on September 25, 2021 ruses in that it stably augments the number of circulating resting molecules might contribute to their activation upon encountering cytolytic T cells (3). Within the CD4ϩ pool, these cells are char- virus-infected cells (9–11). acterized by the absence of CD28 and the presence of cytolytic NK cell receptors (NKRs)4 have first been described as surface granules containing perforin and granzyme B (4). For a fraction of receptors on NK cells that bind to specific HLA class I molecules these cells, specificity for CMV peptides presented by HLA class (12–15). These receptors may belong to either the Ig- or C-type II molecules can indeed be demonstrated (5). Also, the CD8ϩ T lectin superfamilies and transmit, upon binding their ligands, in- cell compartment of CMV carriers contains an expanded fraction hibitory or activating signals to the cell’s interior. The quality of of perforin- and granzyme-expressing cells (6). These cells have the signal depends on the composition of the intracellular parts of been characterized by various phenotypic markers (for a review, the individual receptors or their associated molecules. Addition- see Ref. 7) and CMV peptide-specific CD8ϩ T cells are often ally, NK cells express natural cytotoxic receptors (NCR), like NKp44 and NKp46, for which no ligands have been conclusively defined but that appear to function as strong activators of NK cell *Department of Experimental Immunology and †Department of Internal Medicine, Division of Nephrology and ‡Department of Neurogenetics, Academic Medical Cen- function. Apart from NK cells, expression of NKRs is also found ter, Amsterdam, The Netherlands; and §Division of Hematology, Geneva University on the resting cytotoxic effector cells of CMV carriers (16). In Hospital, Geneva, Switzerland agreement with this, transsectional cohort studies have shown that Received for publication August 21, 2007. Accepted for publication January 30, 2008. CMV infection leaves an imprint in the NKR repertoire on T cells The costs of publication of this article were defrayed in part by the payment of page (17). Whether NKRs on T cells regulate recognition and subse- charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. quent elimination of CMV-infected cells is not clear. Still, certain 1 This work was supported by Grant C03.2034 from the Dutch Kidney Foundation (to NKRs are able to down-modulate cytotoxicity induced via TCR/ A.v.S.). R.A.W.v.L. was supported by VICI Grant 918.46.606 from The Netherlands CD3 or to inhibit cytolysis after cross-linking other NKRs (18– Organization of Scientific Research. 21). Triggering of the C-type lectin receptor NKG2D on Ag-spe- 2 A.v.S. designed research, analyzed data, and wrote the article. A.T.R. designed cific CD8ϩ cells does not trigger cytokine production, calcium research. S.L.Y. performed research and analyzed data. F.B. contributed vital new reagents and analytical tools. E.R. contributed vital new reagents and analytical tools. I.J.t.B. designed research and revised the manuscript critically for important intellectual content. R.A.W.v.L. designed research and revised the manuscript critically for 4 Abbreviations used in this paper: NKR, NK cell receptor; NCR, natural cytotoxic important intellectual content and wrote the article. receptor; p.i., postinfection; KIR, killer Ig-like receptor; aNKR, activating NKR; 3 Address correspondence and reprint requests to Dr. Amber van Stijn, Department of iNKR, inhibitory/inhibiting NKR. Experimental Immunology, Academic Medical Center, Meibergdreef 9, 1105AZ Am- sterdam, The Netherlands. E-mail address: [email protected] Copyright © 2008 by The American Association of Immunologists, Inc. 0022-1767/08/$2.00 www.jimmunol.org The Journal of Immunology 4551

Table I. Patient characteristics

Age Days to First No. of Days Patient (Years) Positive PCR with CMVϩ PCR HLA Typing of Recipient (R) and Donor (D) Assays

1 69 31 219 R: A2/3 B7/39(16) DR15(2)/12(5) Microarray D: A2/23(9) B62(15)/49(21) DR11(5)/13(6) 2 35 45 99 R: A2/3 B7/35 DR1/15(2) Microarray D: A2/11 B7/8 DR4/15(2) 3 27 62 144 R: A1/3 B7/35 DR15(2)/4 Microarray D: A1/3 B7/35 DR4 4 35 47 68 R: A11/3 B51(5)/39(16) DR1/15(2) FACS D: A11/28 B51(5)/27 DR1/9 5 73 25 122 R: A1/2 B7/60(40) DR13(6) FACS D: A1 B63(15)/8 DR17(3)/13(6) 6 42 46 118 R: A26(10)/32(19) B44(12)/8 DR3/12(5) FACS D: A1/31(19) B39(16)/8 DR17(3)/8 7 35 25 79 R: A3/11 B7/51(5) DR15(2)/14(6) FACS D: A11/2 B51(5) DR12(5)/14(6) 8 47 28 83 R:A1/3 B7/8 DR15(2)/3 FACS D: A3/11 B7/45(12) DR15(2)/4 9 30 29 80 R: A1/2 B8/8 DR17(3) FACS D: A2 B8/41 DR3/7 Downloaded from 10 33 NA 0 R: A3/24(9) B51(5)/55(22) DR4/13(6) FACS D: A2/24(9) B51(5)/55(22) DR13(6)/14(6) 11 49 NA 0 R: A2/3 B60(40)/13 DR13(6)/7 FACS D: A1/3 B8/13 DR3/7 12 32 96 14 R: A28 B60(40)/44(12) DR1/13(6) FACS D: A2/28 B44(12)/57(22) DR1/7

13 26 33 37 R: A3/31(19) B51(5) DR3/14(6) FACS http://www.jimmunol.org/ D: A2/3 B27/61(40) DR16(2)/17(3) 14 32 61 45 R: A11/31(19) B51(5)/60(40) DR1/4 FACS D: A2/3 B13/51(5) DR1/7 15 32 49 74 R: A3/11 B7/48 DR15(2) 15(2) FACS D: A3/11 B7/18 DR15(2)/3

NA, Not applicable. mobilization, nor cytotoxicity as it does in NK cells, but, instead, experienced a primary CMV infection. Antiviral therapy consisted of gan- it augments cytotoxic and proliferative responses of T cells (9). ciclovir. Cells from three patients were used in microarray analyses. Lon- by guest on September 25, 2021 Other NKRs may lower activation thresholds (22) and thereby gitudinal samples from 10others were used for detailed flow cytometry. Two CMV-seronegative renal transplant patients who did not develop pri- function as costimulatory receptors for T cells. Ibegbu et al. (23) mary CMV infection were used to verify whether the phenotypical changes described the presence of the inhibitory C-type lectin KLRG1 in were indeed inflicted by CMV infection rather than by the transplantation combination with CD57 on terminally differentiated memory itself. Patient characteristics are listed in Table I. All patients gave written CD8ϩ cells. These cells secrete cytokines but have a poor repli- informed consent and the study was approved by the local medical ethical committee. cative proliferation, suggesting a modifying role of KLRG1 in these CD8ϩ cells. Tetrameric complexes Elegant studies in mice have revealed that NKRs are able to Allophycocyanin-conjugated HLA-A2 tetramer loaded with the CMV modify the functional competence of virus-specific T cells in vivo. pp65-derived NLVPTMVATV peptide and allophycocyanin-conjugated CD94-NKG2A, an inhibitory receptor of the C-type lectin class, HLA-B7 tetramer loaded with the CMV pp65-derived TPRVTGGGAM was shown to induce anergy in a polyomavirus-specific CD8ϩ T peptide were obtained from Sanquin. cell population, contributing to the development of tumors (24). Cells Thus, NKRs on T cells may be indispensable for an adequate balance of activation and control of T cells during viral infection. PBMC were isolated from heparinized blood using standard density gra- dient centrifugation and subsequently cryopreserved in liquid nitrogen until The development of CMV-specific T cell responses can longi- the day of analysis. tudinally be followed in recipients of kidney allografts that en- To isolate naive CD8ϩ cells, cells were isolated from buffy coats counter CMV for the first time as a result of virus reactivation from from healthy donors by a two-step procedure. After Ficoll, CD8ϩ T cells were isolated by CD8ϩ microbeads (Miltenyi Biotec) and stored the transplanted organ (25). We used DNA microarrays combined ϩ with flow cytometry to obtain a comprehensive view on the reg- overnight at 4°C in 10% (v/v) serum-containing medium. CD8 T cells ϩ were then labeled with CD27-FITC (7C9, homemade), CD45RA-RD1 ulation of NKR expression on CMV-induced CD8 T cells. Spe- (Coulter), and CD8-allophycocyanin (BD Pharmingen) and FACS cifically, we asked whether 1) NKR expression is a direct and sorted using a FACS Aria (BD Biosciences) in naive CD8ϩ T cells immediate consequence of CMV infection and 2) whether the bal- (CD8ϩCD45RAhighCD27high). To isolate CMV-specific CD8ϩ effector ance between activating NKRs (aNKRs) and inhibitory NKRs cells at the peak of the CMV response, PBMC were stained with HLA- DR-FITC (BD Biosciences, CD38-PE (BD Biosciences), and CD8-allo- (iNKRs) changes between the acute and latent stage. phycocyanin (BD Pharmingen). HLA-DRhighCD38highCD8ϩ were sorted using a FACS Aria (BD Biosciences). Materials and Methods To obtain CMV-specific CD8ϩ cells in the latency phase, PBMC ob- Subjects tained from 40 to 60 wk after transplantation (long term, 1 year postinfection (p.i.)) were stained with allophycocyanin-conjugated tetramers and Renal transplant recipients were treated with basiliximab prophylactically subsequently allophycocyanin microbeads (Miltenyi Biotec) were used to and with basic immunosuppressive drug therapy consisting of pred- isolate the cells. Upon reanalysis, the purified populations contained be- nisolone, mycophenolate mofetil, and cyclosporin. Thirteen of 15 patients tween 95 and 97% tetramer-binding cells. 4552 NKRs ON CD8ϩ T CELLS DURING PRIMARY CMV

RNA isolation, amplification, labeling, and hybridization RNA was isolated using the nucleospin RNA isolation kit (Machery-Na- gel) according to the manufacturer’s instructions. mRNA was amplified using the MessageAmp II Kit (Ambion). Labeling, hybridization, and data extraction were performed at ServiceXS (Leiden, The Netherlands). Briefly, 100 ng of total RNA was mixed with 1 ␮l of T7 oligo(dT) primer in a total volume of 12 ␮l. Primer and template were denatured by incubating at 70°C for 10 min and annealed by putting the reaction tubes on ice. The first-strand reaction was performed by adding 8 ␮l of Reverse Tran- scription Master Mix (containing 10ϫ First Strand buffer, RNase inhibitor, dNTP mix, and reverse transcriptase) and incubating at 42°C for 2 h. Sec- ond- strand cDNA synthesis was done by adding 63 ␮l of Nuclease-Free Water, 10 ␮lof10ϫ second-strand buffer, 4 ␮l of dNTP mix, 2 ␮lofDNA polymerase, and 1 ␮l of RNase H and incubating at 16°C for 2 h. cDNA purification was done according to the manufacturer’s protocol (Ambion). In vitro transcription was initiated by addition of 2 ␮l of aaUTP Solution (50 mM), 12 ␮l of ATP, CTP, and GTP mix (25 mM), 3 ␮lofUTP Solution (50 mM), 4 ␮lofT710ϫ reaction buffer, and 4 ␮l of T7 enzyme mix and incubated at 37°C for 9 h. Machery-Nagel RNA Clean up mini- spin columns were used for purification of the cRNA. Dye coupling reaction was performed using 5 ␮g of amino allyl aRNA in 3.33 ␮l, 5 ␮l of DMSO, and ␮ 1.66 l of NaCO3 buffer added to the monoreactive dye, prepared according Downloaded from to the manufacturer’s protocol (Amersham Biosciences). After an incubation at room temperature for 60 min, 4.5 ␮l of 4 M hydroxylamine was added and incubated at room temperature for 15 min. Dye-labeled aRNA was purified with Machery-Nagel RNA Clean up mini-spin columns and the samples were checked on concentration and dye incorporation on the Nanodrop ND-1000. Hybridization was performed with 600 ng of each labeled target along with fragmentation and hybridization buffer at 60°C for 17 h onto Human Whole Genome (WHG) 44K Oligo Microarrays from Agilent Technologies per the http://www.jimmunol.org/ manufacturer’s protocol. Microarray imaging and data analysis The microarray slides were washed following the user manual instructions and scanned using the Agilent dual-laser DNA microarray scanner. Default settings of Agilent Feature Extraction preprocessing protocols were used to obtain normalized expression values from the raw scans. Exact protocol and parameter settings are described in the Agilent Feature Extraction Software User Manual 7.5 (http://chem.agilent.com/scripts/LiteraturePDF.

asp?iWHIDϭ37629). The default Agilent normalization procedure, called by guest on September 25, 2021 Linear & Lowess, was applied. Rosetta Resolver (Rosetta Biosoftware, Seattle, WA) was used for analysis of the data. Flow cytometric analyses

PBMC were washed in PBS containing 0.01% (w/v) NaN3 and 0.5% (w/v) BSA (PBA). PBMC (500,000) were incubated with appropriate concentrations of tetrameric complexes in a small volume for 30 min at 4°C, protected from light. Then fluorescent-labeled conjugated mAbs (concentrations according to the manufacturer’s instructions) were added and incubated for 30 min at 4°C, protected from light. For analysis of expression of surface markers, the following mAbs were used in different combina- tions: CD56-allophycocyanin, NKG2D-allophycocyanin (BD Pharmin- gen), KIR2DS1/L1-PE (CD158a and h, clone EB6), KIR2DL2/L3/S2-PE (CD158b and j, clone GL183), NKp44-PE, NKp46-PE, NKG2A-PE (all from Beckman Coulter), NKG2C-PE (R&D Systems), CD27-FITC (homemade clone 3A12 for FACS analysis and homemade clone 7C9 for FACS sorting), CD27-PE (clone L128), CD3-PerCP-Cy5.1, CD94-PE, CD94-allophycocyanin, NKB1 (KIR3DL1)-FITC, CD38-PE, CD45RA-FITC, anti- HLA-DR-FITC (all BD Biosciences). Cells were washed in PBA and analyzed using a FACSCalibur flow cytometer (BD Biosciences) or FACS Canto flow cytometer (BD Biosciences) and CellQuest software (BD Bio- sciences) or FlowJo software (Tree Star). Quantitative PCR for KIR2DL2/DL3/DS2 From patients 4 and 8 (see Table I), CD8ϩCD158ϩ cells were sorted at two time points: at the peak of the viral load (T1) and 2 mo later (T2: patient FIGURE 1. Confirmation of the microarray results by FACS analyses of 4) or 6 mo later (T2: patient 8) when the virus was not detectable. RNA was selected CD8ϩ T cell proteins that are known to be regulated during a primary isolated and quantitative PCR was performed on cDNA with the Light- CMV infection. Graphs show the RNA fold changes found by microarray Cycler System (Roche Diagnostics) in microcapillary tubes with a Quan- analysis (as compared with naive CD8ϩ cells) for the genes indicated. The tiTect SYBR green PCR kit solution (Qiagen). After a 15-min denaturation protein confirmation of the selected genes is performed by flow cytometry. step at 95°C, 50 PCR cycles of 15 s at 94°C, 30 s at 58°C, 45 s at 72°C, ϩ Density plots are gated on CD8 T cells. Depicted on the x-axis is the pp65 and5sat79°C were performed. To confirm the purity and specificity of the reaction, a melting curve analysis was performed at the end of the PCR CMV-specific tetramer staining, on the y-axis is the protein as indicated. peak, Peak of CMV infection, 1 year; ϳ1 year after CMV infection. Quadrants by slowly increasing (0.1°C/s) the temperature of the reaction from 65 to ϩ 95°C. A standard curve was established with a dilution of purified PCR depicted as percentages of total CD8 T cells. Shown is one representative product from 10 to 107 molecules. GAPDH was used as an internal patient (patient 15). The Journal of Immunology 4553

Table II. Changes in NKR during primary CMVa

Alternative Name Naiveb Peakc 1 Yeard mAbse

aNKR MHC class I speciﬁc KIRs KIR2DS2 CD158j 1 3,6 1,4 GL183f KIR2DS4 CD158l 1,2 6,4 1,9 KIR3DL5A CD158f 1,1 2,3 1,1 Leukocyte Ig-like LILRA1 LIR6/CD85l 0,9 10,6 13,4 LILRA3 ILT6/LIR4/CD85e 0,9 2 2,9 LILRA4 ILT7 0,9 1,3 1,3 LILRA5 CD85f 0,9 3,7 2,3 Non-MHC class I speciﬁc NCRs NCR1 NKp46 0,9 47,4 39,1 NKp46 NCR2 NKp44 1,1 0,9 0,9 NKp44 NCR3 NKp30 1,1 1,4 0,9 C-type lectin receptors KLRC2 NKG2C 1 9,8 7,8 NKG2C KLRK1 NKG2D 1 3,6 2,3 NKG2D Downloaded from KLRC3 NKG2E 1 2,8 1,2 KLRC4 NKG2F 1 2,2 1,9 KLRF1 0,9 5 12,8 CLEC2B AICL 1 3,8 2,2 Other CD16 FcyIIIa 0,9 8,5 14

CD244 2b4 1,1 13,5 10,6 http://www.jimmunol.org/ NKG7 1 8,5 14,9 iNKR MHC class I speciﬁc KIRs KIR2DL1 CD158a 0,8 6 1,6 EB6g KIR2DL2 CD158b1 1 5 1,5 GL183f KIR2DL4 CD158d 1 2,9 1 KIR3DL1 CD158e1/NKB1 1 2,9 1,1 NKB1 KIR3DL2 CD158k 1,1 7,1 1,4

KIR3DL3 CD158z 1 1,1 1,3 by guest on September 25, 2021 Leukocyte Ig-like receptors LILR1 0,9 2,3 2,4 LILRB1 ILT2/LIR1/CD85j 1 28,3 61,4 LILRB2 ILT4/LIR2/CD85d 1,6 13,9 17,9 LILRB3 ILT5/LIR3/CD85a 1 2,1 2,2 LILRB4 ILT3/LIR5/CD85k 1 0,9 1,1 LILRB5 LIR8/CB85c 0,9 1,4 1 Non-MHC class I speciﬁc NCRs None C-type lectin receptors KLRA1 1 2,2 2,4 KLRB1 CD161 1 9,3 17,1 KLRC1 NKG2A/CD159A 1,1 98NKG2A KLRG1 MAFAL 0,9 11,5 14,5 Other LAIR1 1 1,1 2,6 LAIR2 1,2 11,8 34,4 Either aNKR or iNKR Leukocyte Ig-like receptors LILRP2 ILT10/CD85m 1 1,5 2,1 C-type lectin receptors KLRD1 CD94 1 36,5 38,9 CD94

a Bold numbers represent fold changes of significantly changed genes according to the criteria described in Materials and Methods. b Naive: CD8ϩCD45ϩCD27ϩ. c Peak: CD8ϩCD38ϩHLA-DRϩ. d One year: CMV-specific tetramer pp65ϩ cells. e mAbs used in FACS analyses. f Clone GL183 detects KIR2DS2/2DL2/2DL3. g Clone EB6 detects KIR2DS1/L1. reference. The killer Ig-like receptor (KIR) primers used were either the same AA-3Ј; reverse, 5Ј-GCCCTGCAGAGAACCTACA-3Ј; KIR2DL3, 5Ј-CC as those published by Uhrberg et al. (26) or one nucleotide modifications ACTGAACCAAGCTCCG-3Ј; reverse, 5Ј-GCAGGAGACAACTTTGGA thereof: KIR2DS2, 5Ј-CTGCACAGAGAGGGGAAGTA-3Ј; reverse, 5Ј- TCA-3Ј; and GAPDH, 5Ј-GGACCTGACCTGCCGTCTAG-3Ј; reverse, ACGCTCTCTCCTGCCAA-3Ј; KIR2DL2, 5Ј-CATGATGGGGTCTCCA 5Ј-CCACCACCCTGTTGCTGTAG-3Ј. 4554 NKRs ON CD8ϩ T CELLS DURING PRIMARY CMV Downloaded from

FIGURE 3. Acquisition and maintenance of iNKRs and aNKRs during primary CMV infection detected by microarray. A, Cumulative fold change was calculated by summing all of the fold changes of genes that change Ն2 at the peak and/or 1 year p.i. observed from aNKRs (n ϭ 15) or iNKRs ;p Ͻ 0.001 ,ء .(n ϭ 13); all NKRs (n ϭ 31, including unknown NKRs)

Wilcoxon-signed rank test. Multiple observations and genes that did not http://www.jimmunol.org/ change Ն (n ϭ 8) were ignored. B, Mean fold changes and SD were calculated. #, Mann-Whitney U test. FIGURE 2. Two expression profiles of significantly regulated NKRs upon primary CMV infection. Trend plots are made with the Rosetta Re- solver software; bold lines connect significantly changed NKRs. Back- lated from the same patients between 40 and 60 wk after the pri- ground lines represent all gene expression data of the microarray. A, Tran- mary response. At this time point, no CMV could be detected in sient up-regulation during acute phase of infection. B, Up-regulation during the blood by sensitive PCR. Because we used the two-color mi- the acute phase and ϳ40 wk p.i. croarray Agilent WHG chips to perform RNA expression profiling, the expression level of each gene in a specific population is given by guest on September 25, 2021 as a ratio relative to the expression level of naive CD8ϩ cells Statistics obtained from healthy donors (subset CD8ϩ cells/naive CD8ϩ Trend plotting was performed in a Rosetta Resolver with a fold change cells). cutoff of Ն2 and a one-way ANOVA test with a p value cutoff of Յ0.05 Microarray analysis revealed a number of changes in gene ex- with the Benjamini and Hochberg false-discovery rate correction to select pression levels of several molecules (Fig. 1). The changes in sequences with differences across the different cell populations. The paired, mRNA expression levels of Ki-67, CD28, CCR7, CD27, CD127, nonparametric Wilcoxon-signed rank test was used for analysis of genes perforin, granzyme B, and IFN-␥ were confirmed at the protein that differ between different cell populations. The nonparametric Mann- Whitney U test was used for analysis of differences between aNKRs and level by FACS analyses; Ki-67, a proliferation marker, was only iNKRs. Linear regression analysis was performed to show the relationship elevated during the peak of viral replication. In contrast, CD28, between NKG2A and CD94bright expression. Significance was indicated by CD27, CD127, and CCR7 expression was low on virus-specific T p Ͻ 0.05. cells. Finally, the cytolytic proteins granzyme B and perforin and the antiviral cytokine IFN-␥ were high in virus-specific cells. Results On the Agilent’s WHG array, 39 NKR-related sequences were CMV infection induces a rapid and persistent induction of NKR present. Changes in their relative expression levels are represented expression in Table II and Fig. 2. Three distinct expression profiles could be To analyze regulation of NKR gene expression by CMV in hu- distinguished. First, the expression of only eight NKRs such as mans, we took advantage of the fact that dynamic change in virus- LILRA4 or KIR3DL3 did not show significant changes (Table II). specific T cell populations can carefully be tracked in CMV-sero- Second, a limited number of receptors, including KIR2DS2, negative renal transplant recipients that experience a primary KIR2DS4, and KIR3DL2, were up-regulated during the peak of CMV infection due to the transplantation of a CMV-carrying do- the CMV infection but returned to baseline levels in the latency nor kidney. As described earlier (8), circulating CMV-tetramer- stage (Fig. 2A). However, the majority of NKR genes were in- binding CD8ϩ cells emerge ϳ2–4 wk after the first positive CMV duced early in the response and remained high in the follow-up. PCR, but their number is too low to obtain enough cells for mi- Two NKRs (LAIR1 and LILRP2) were only induced in the latent croarray analyses. To be able to include enough cells eligible for state (Table II). analysis, we isolated all virus-activated CD8ϩ T cells by sorting Remarkably, KIRs, that bind epitopes on HLA-C and HLA-B CD8ϩHLA-DRϩCD38ϩ T cells. We previously showed that these class I alleles were almost exclusively present in the second pro- activated lymphocytes appear at a high frequency early in the anti- file, e.g., induced during the acute phase of the viral infection, but CMV response and that at this point all CMV-tetramerϩ cells are absent in CMV pp65-specific tetramer-positive CD8ϩ cells in the contained within this population (8). To obtain virus-specific cells renal transplant patients 1 year p.i. The other classes of receptors in the latency stage, CMV-specific pp65 tetramerϩ cells were iso- did not show predominance for a particular expression profile. The Journal of Immunology 4555 Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 4. Induction of NKRs on CD3ϩ cells during primary CMV infection detected by FACS analysis. Dot plots are gated on CD3ϩ T cells in which the CD27 marker is plotted against the staining for EB6 (KIR2DS1/KIR2DL1), GL183 (KIR2DL2, KIR2DL3, KIR2DS2), NKB1 (KIR3DL1), NKG2D, NKp44, NKp46, or CD94. Quadrant percentages depicted as percentages from CD3ϩ cells. Shown is one representative patient (patient 4). PCR viral load is given as copies per ml. Time is deﬁned as weeks from the peak of the PCR viral load. All stainings are CD27-FITC, NKR-PE, CD3-PerCP, and CD56-allophycocyanin, except in the NKB1 staining, which is FITC labeled and CD27-PE is used. 4556 NKRs ON CD8ϩ T CELLS DURING PRIMARY CMV

FIGURE 5. Induction of NKRs on CD3ϩ T cells during primary CMV infection detected by FACS analysis. A–F, Percentages are calculated as means and SEM from T cells in six patients with a primary

CMV infection. Time is deﬁned as weeks from the Downloaded from peak of the PCR viral load determined for each patient separately. G and H, Quantitative PCR for KIR2DL2/DL3/DS2, which can be detected by ﬂow cytometric staining with GL138 mAb. G, Patient 4, T1; peak CMV, T2; week 4 mo after peak. H, Pa- tient 8, T1; peak CMV, T2; 6 mo after peak. http://www.jimmunol.org/ by guest on September 25, 2021

The balance between inhibitory and activating receptors does expected to radically change the results since tetramerϩ T cells also not change in the course of infection have a CD38ϩHLA-DRϩ phenotype in the acute phase. To visualize the CMV-induced changes on NKR gene expression in CD8ϩ T cells, we calculated a cumulative fold change and the mean Individual NKRs show different kinetics in cell surface fold change of all receptors. When all NKR genes were grouped to- expression gether, a strong increase of NKR expression was observed both in the acute and latency stages (Fig. 3A). Of the 39 NKR genes on the chip, To obtain a more thorough insight into the detailed kinetics of 19 are activating, 18 are inhibiting, while 2 have yet an unknown NKR expression, four-color flow cytometric analyses were per- function. CD94 forms heterodimers with either inhibiting NKG2A or formed. Due to the limited number of fluorescence channels, we activating NKG2C. Separating the NKRs in aNKRs and iNKRs re- chose to analyze the NKR phenotype in combination with CD3, vealed a similar expression pattern for both types of receptors (Fig. CD56, and CD27 markers. CD27 was included to monitor the 3B). Since a small change level in RNA can have a major impact at maturation of virus-specific cells by the gradual loss of CD27. The the protein level, we also counted the number of aNKRs and iNKRs combination of CD3 and CD56 proved to be a good combination ϩ that changed during infection. This showed that there was no selective to distinguish between virus-specific T cells, including CD4 and ϩ induction of either aNKRs or iNKRs at a particular stage; at the peak CD8 T cells, and NK cells. of CMV, 16 aNKRs, 12 iNKRs and CD94 were elevated, whereas 11 Representative FACS plots of one patient are shown in Fig. 4 aNKRs, 8 iNKRs, and 2 “unknown function” NKRs (LILRP2 and and the data of six patients analyzed are represented in Fig. 5. A CD94) were up-regulated after ϳ1 year of infection. Thus, the array clear induction of NKR expression was observed when the viral data did not provide any indication that the balance between activating load reached its maximum. Although expression on CD3ϩ cells and inhibitory receptor genes differs between virus-specific cells in the was low (0.4 Ϯ 0.07%) and remained low after transplantation (up acute vs the latent stage. It should be mentioned that in the acute phase to 1.9 Ϯ 0.4%), a consistent and significant increase of KIR2DS1/ CMV-specific CD8ϩ T cells containing activated T cells were stud- KIR2DL1 (detected by the EB6 mAb, further referred to as EB6ϩ; ied, while in the latent phase pp65-tetramerϩ CMV-specific cells were Fig. 5A) was observed when CMV DNA became detectable. Al- studied. This could potentially influence the results, although it is not though with clearly different kinetics, consistent increases in cell The Journal of Immunology 4557 Downloaded from

FIGURE 6. CD94dim, CD94bright, NKG2A, and NKG2C on T cells during primary CMV. Total CD94 (A), CD94dim (B), and CD94bright http://www.jimmunol.org/ (C) expression on CD3ϩ T cells. D, Linear regression analysis was performed between CD94bright and NKG2A expression on CD3ϩ cells (R2 ϭ 0.98). NKG2A (E) and NKG2C (F) expression on CD8ϩ cells. Shown is one representative patient of three patients. NKG2A (G) and NKG2C (H) on CMV-speciﬁc CD8ϩ pp65- tetramerϩ cells. Shown is one representative patient (patient 4). Dot plots are gated on vital cells. Numbers indicate the percentages within by guest on September 25, 2021 the corresponding quadrants from the gated population. Time is deﬁned as weeks from the peak of the PCR viral load determined for each patient separately. PreTx, Pretransplantation; n.a., not applicable.

surface expression were also found for KIR2DS2/2DL2/2DL3 (de- and NKp46 (Fig. 5, E and F), and CD94 (Fig. 6A). In two CMV- tected by the GL183 mAb, further referred to GL183ϩ; Fig. 5B), seronegative individuals that received a CMV-negative donor kid- KIR3DL1 (NKB1; Fig. 5C), NKG2D (Fig. 5D), the NCRs NKp44 ney, no changes in NKR expression were observed (data not 4558 NKRs ON CD8ϩ T CELLS DURING PRIMARY CMV shown), showing that the alterations in NKR expression were in- proach, we found that mRNA expression levels of different mem- deed a consequence of the CMV infection and not related to the bers of all NKR families, including KIRs, leukocyte Ig-like recep- transplantation procedure. In the follow-up periods of ϳ1 year, we tors, NCRs, and C-type lectin receptors in virus-induced CD8ϩ T did not observe a specific association of NKR with CD27 expres- cells, increased during the infection. Many of these genes re- sion, indicating that induction of these molecules is independent mained increased up to 50 wk after the peak of the initial viral ϩ from other differentiation steps of human CMV-specific CD8 T burst when the virus load had long dropped below detection by cells (Fig. 4). sensitive PCR. We investigated whether the presence of KIR ligands corre- Changes in KIR2DL1/S1 (EB6), KIR2DL2/S2 (GL183), and sponded to the elevation of specific KIRs during the primary CMV KIR3DL1 (NKB1) mRNA were observed in the acutely acti- infection. No evidence for this assumption was found, because vated CD8ϩHLA-DRϩCD38ϩ cells, but expression levels in patients 5, 8, and 9 nor their transplanted kidneys expressed HLA- the pp65 tetramer CD8ϩ cells obtained during latency were Bw4, the ligand for NKB1, but still NKB1 was strongly induced in comparable to those of naive CD8ϩ T cells. This observation is these individuals (data not shown). in agreement with previous observations (16, 28) where expres- ϩ The increase in GL183 was very prominent, as at the end of sion of these KIRs was found on the CD8ϩCD45RAϩCD27Ϫ ϩ the follow-up up to 13% of the circulating CD3 pool expressed effector-type cells typical for CMV- infected individuals this marker. Because the Ab GL183 recognizes both activating present during CMV infection but absent on the pp65 tetramer- (KIR2DS2) and inhibitory members (KIR2DL2 and DL3) of the binding cells. The reason for this difference is unclear. KIR family, we analyzed, by quantitative PCR, whether CMV in- CD8ϩCD45RAϩCD27Ϫ only expand to high numbers after duced the selective outgrowth of a monoclonal population express- Downloaded from ϩ ϩ CMV infection (6) and it therefore seems likely that many ing only one KIR species. CD8 GL183 cells from two patients CMV-reactive cells are contained within this subset. Elegant (patients 4 and 8) at two time points were obtained and in these studies by Picker et al. (29) have shown that the T cell response cells all three transcripts could be detected with values that are in to CMV is very robust and is directed against many peptides of the normal range of KIR-purified cells (Fig. 5G). This shows that the virus. Studies until now have mainly focused on pp65-spe- the strongly expanded GL183ϩ T cell population does not repre- ϩ cific T cells and therefore it is possible that CMV-specific sent the outgrowth of a monoclonal KIR population. CD8ϩ T cells, reactive toward epitopes generated from other http://www.jimmunol.org/ CD94 dimerizes with different kinetics with NKG2C and NKG2A viral proteins, will have different phenotypes with regard to, for ϩ example, KIR. The increase in CD94 expression was quite prominent on CD3 ϩ NKR expression on CD8 T cells is not associated with other T cells (Figs. 4 and Fig. 6A), and we therefore analyzed in more latent viruses such as EBV and HIV (28), suggesting that expres- detail whether the molecule would operate in transmitting ac- sion of these molecules has a particular function in the anti-CMV tivating and/or inhibiting signals. CD94 can only exert its sig- response. Since CMV-specific T cells generally reach later differ- naling function when heterodimerized with either NKG2A (in- entiation stages than T cells specific for other persisting viruses hibiting dimer) or NKG2C (activating dimer). The intensity of

(3), one might assume that the expression of NKRs is an integral by guest on September 25, 2021 CD94 expression is, as reported by Arlettaz et al. (27), high in part of the late differentiation phenotype. However, our array data conjunction with NKG2A but low when bound to NKG2C. Dur- appear to argue against this notion. Up-regulation of a vast number ing the first weeks of the CMV response, predominantly of NKRs already takes place during the early phase of the response CD94dim cells appeared in the circulation (Fig. 6B) and some- when cells have an activated early differentiation phenotype (16). what later CD94bright cells became detectable (Fig. 6C). As expected (27), we found a correlation between the CD94bright and Rather, our data suggest that the expression of NKRs as a consequence of CMV infection is an early and adaptive response of the NKG2A expression in our samples (Fig. 6D). Because the ma- ϩ jority of CD94-positive T cells are CTLs, we continued the CD8 effector cell pool to the virus. Whether specific viral prod- analysis on CD8ϩ T cells. Already before CMV infection, ucts and/or the virus-induced modulation of HLA expression have NKG2C is present on a small subset of CD8ϩ cells, whereas created an evolutionary necessity for this broad expression of the NKG2A is almost absent on CD8ϩ cells. Confirming the dif- NKR repertoire will have to be resolved. ferential rise in the dimerization pattern of CD94 with NKG2C The early up-regulation of the NCR NKp44 was found by flow ϩ cytometry but was not revealed by the microarrays. This is perhaps and NKG2A, we found that a NKG2C increase on CD8 cells ϩ could be detected shortly after the start of the infection (2 wk not surprising since the up-regulation of NKp44 in CD3 cells is before CMV peak; Fig. 6F), whereas the increase of NKG2A transient and low, reaching a maximum at the peak of the viral load started later in the response (Fig. 6E). To analyze the expression of only mean 1.7%. Two weeks after the peak of the viral load, the of these dimers on established virus-specific cells, pp65 tetram- frequency of NKp44-expressing cells is already decreasing in most ers were used to analyze cells of a HLA-A2ϩ individual. The patients. It is possible that the patients selected for the RNA iso- inhibitory CD94:NKG2A dimer was found on the majority of lation had too low levels to be detectable by microarray analysis. the CMV-specific tetramerϩ cells irrespective of the stage of In any case, it suggests that NKp44, as in NK cells (30, 31), is infection (Fig. 6G). In marked contrast, the activating NKG2C preferentially expressed on the mitogenically activated cells that dimer was induced on the CD3ϩ cells but conspicuously absent can only be found early in infection (8). ϩ from the tetramer-binding cells (Fig. 6H). The absence of CD94 was found to be expressed on both CD3 cells (pro- ϩ CD94/NKG2C on cells with proven virus-specificity is remi- tein) and CD8 CMV-specific cells (RNA). CD94 forms het- niscent of earlier findings showing that tetramer-binding cells erodimers with NKG2x proteins (where x can be A, B, C, E, or also lack EB6 and GL183 (CD158a,h and CD158b, j expres- H). The NKG2A/B/CD94 is an inhibitory receptor, whereas the sion) (16). NKG2C/CD94 and NKG2E/H/CD94 are activating receptors (32). Arlettaz et al. (27) reported that CD94bright expression is Discussion associated with NKG2A and CD94dim is associated with In this study, we show the impact of a primary CMV response on NKG2C, although also other NKG2 proteins may dimerize with NKR virus-induced CD8ϩ T lymphocytes. Using a microarray ap- CD94. Our observation that CD94dim was primarily enhanced The Journal of Immunology 4559 during the acute phase of the infection, whereas CD94bright ac- Frequencies of circulating cytolytic, CD45RAϩCD27Ϫ, CD8ϩ T lymphocytes depend on infection with CMV. J. Immunol. 170: 4342–4348. counted for the latent phase was confirmed by the quick rise in ϩ 7. van Lier, R. A., I. ten Berge, and L. E. Gamadia. 2003. Human CD8 T-cell NKG2C followed by robust expansion of NKG2A which con- differentiation in response to viruses. Nat. Rev. Immunol. 3: 931–939. tinued to expand for at least 40 wk after the peak of the viral 8. Gamadia, L. E., E. B. Remmerswaal, J. F. Weel, F. Bemelman, R. A. van Lier, and I. ten Berge. 2003. Primary immune responses to human CMV: a critical role load. A strict correlation between microarray data and surface ϩ for IFN-␥-producing CD4 T cells in protection against CMV disease. Blood expression of NKG2C was not apparent in our analyses. 101: 2686–2692. Indeed, it has been reported (27, 33) that if NKG2A is ex- 9. Groh, V., R. Rhinehart, J. Randolph-Habecker, M. S. Topp, S. R. Riddell, and ␣␤ pressed, no surface NKG2C expression could be observed even T. Spies. 2001. Costimulation of CD8 T cells by NKG2D via engagement by MIC induced on virus-infected cells. Nat. Immunol. 2: 255–260. with abundant transcription of NKG2C. This discrepancy might be 10. Mandelboim, O., D. M. Davis, H. T. Reyburn, M. Vales-Gomez, E. G. Sheu, due to different affinities of CD94 for binding partners and a pref- L. Pazmany, and J. L. Strominger. 1996. Enhancement of class II-restricted T cell erence to dimerize with NKG2A. Also in our analyses, whereas responses by costimulatory NK receptors for class I MHC proteins. Science 274: ϩ 2097–2100. NKG2A was expressed at the cell surface of tetramer cells, no 11. Namekawa, T., M. R. Snyder, J. H. Yen, B. E. Goehring, P. J. Leibson, NKG2C expression could be detected in apparent contrast with the C. M. Weyand, and J. J. Goronzy. 2000. Killer cell activating receptors function as costimulatory molecules on CD4ϩCD28null T cells clonally expanded in rheu- high NKG2C mRNA content detected in the microarray analysis. matoid arthritis. J. Immunol. 165: 1138–1145. The coordinated changes of aNKRs and iNKRs on recently 12. Moretta, A., C. Bottino, D. Pende, G. Tripodi, G. Tambussi, O. Viale, A. Orengo, primed and late differentiated CD8ϩ T cells at different time M. Barbaresi, A. Merli, E. Ciccone, et al. 1990. Identification of four subsets of human CD3ϪCD16ϩ natural killer (NK) cells by the expression of clonally dis- points during the viral infection suggest that NKRs contribute tributed functional surface molecules: correlation between subset assignment of to the fine tuning of antiviral effector functions, keeping the NK clones and ability to mediate specific alloantigen recognition. J. Exp. Med. delicate balance between efficient destruction of virus-infected 172: 1589–1598. Downloaded from 13. Moretta, A., M. Vitale, C. Bottino, A. M. Orengo, L. Morelli, R. Augugliaro, cells and loss of organ function due to immunopathology. This M. Barbaresi, E. Ciccone, and L. Moretta. 1993. P58 molecules as putative re- idea is in line with several observations on the function of KIRs ceptors for major histocompatibility complex (MHC) class I molecules in human and leukocyte Ig-like receptors showing that they can, for ex- natural killer (NK) cells: anti-p58 antibodies reconstitute lysis of MHC class I-protected cells in NK clones displaying different specificities. J. Exp. Med. 178: ample, down-regulate killing efficiency (34), inhibit cytokine 597–604. production (35), and weaken proliferative potential (28) of 14. Vitale, M., S. Sivori, D. Pende, L. Moretta, and A. Moretta. 1995. Coexpression ϩ of two functionally independent p58 inhibitory receptors in human natural killer CD8 T cell clones. Henel et al. (36) recently demonstrated that http://www.jimmunol.org/ ϩ cell clones results in the inability to kill all normal allogeneic target cells. Proc. with low Ag concentrations, KIR expression on memory CD4 Natl. Acad. Sci. USA 92: 3536–3540. T cells uncoupled granule release from gene activation, reason- 15. Bottino, C., M. Vitale, L. Olcese, S. Sivori, L. Morelli, R. Augugliaro, ing that a general inflammation is inhibited to circumvent E. Ciccone, L. Moretta, and A. Moretta. 1994. The human natural killer cell receptor for major histocompatibility complex class I molecules: surface modu- pathological events. In view of this, it can be reasoned that in lation of p58 molecules and their linkage to CD3 ␨ chain, Fc␧RI ␥ chain, and the immune reactions against persistent viruses, like CMV, a neg- p56lck kinase. Eur. J. Immunol. 24: 2527–2534. 16. Gamadia, L. E., R. J. Rentenaar, P. A. Baars, E. B. Remmerswaal, S. Surachno, ative regulation of general inflammation is essential to prevent J. F. Weel, M. Toebes, T. N. Schumacher, I. ten Berge, and R. A. van Lier. 2001. damage by continued activation of T lymphocytes. Differentiation of cytomegalovirus-specific CD8ϩ T cells in healthy and immu- Cohort studies have indicated that the NKR profile on NK and nosuppressed virus carriers. Blood 98: 754–761.

17. Guma, M., A. Angulo, C. Vilches, N. Gomez-Lozano, N. Malats, and by guest on September 25, 2021 T cells is imprinted by CMV (17). We here have demonstrated a M. Lopez-Botet. 2004. Imprint of human cytomegalovirus infection on the NK strong induction of NKRs on T cells as a direct result of CMV cell receptor repertoire. Blood 104: 3664–3671. infection. These NKRs might provide essential control mecha- 18. Ferrini, S., A. Cambiaggi, R. Meazza, S. Sforzini, S. Marciano, M. C. Mingari, and L. Moretta. 1994. T cell clones expressing the natural killer cell-related p58 nisms for the activation of T cells by regulating costimulatory receptor molecule display heterogeneity in phenotypic properties and p58 func- signals, dampening T cell activation when viral loads diminishes, tion. Eur. J. Immunol. 24: 2294–2298. and controlling harmful activation to the host during latent viral 19. Mingari, M. C., C. Vitale, A. Cambiaggi, F. Schiavetti, G. Melioli, S. Ferrini, and A. Poggi. 1995. Cytolytic T lymphocytes displaying natural killer (NK)-like ac- infections. tivity: expression of NK-related functional receptors for HLA class I molecules (p58 and CD94) and inhibitory effect on the TCR-mediated target cell lysis or lymphokine production. Int. Immunol. 7: 697–703. Acknowledgments 20. Mingari, M. C., F. Schiavetti, M. Ponte, C. Vitale, E. Maggi, S. Romagnani, We thank Dr. Perry Moerland for assistance with the bioinformatics and J. Demarest, G. Pantaleo, A. S. Fauci, and L. Moretta. 1996. Human CD8ϩ T statistics of the microarray analysis. lymphocyte subsets that express HLA class I-specific inhibitory receptors represent oligoclonally or monoclonally expanded cell populations. Proc. Natl. Acad. Sci. USA 93: 12433–12438. Disclosures 21. Mingari, M. C., M. Ponte, C. Cantoni, C. Vitale, F. Schiavetti, S. Bertone, The authors have no financial conflict of interest. R. Bellomo, A. T. Cappai, and R. Biassoni. 1997. HLA-class I-specific inhibitory receptors in human cytolytic T lymphocytes: molecular characterization, distri- bution in lymphoid tissues and co-expression by individual T cells. Int. Immunol. References 9: 485–491. 1. Rowshani, A. T., F. J. Bemelman, E. M. van Leeuwen, R. A. van Lier, and 22. Jamieson, A. M., A. Diefenbach, C. W. McMahon, N. Xiong, J. R. Carlyle, and I. ten Berge. 2005. Clinical and immunologic aspects of cytomegalovirus infec- D. H. Raulet. 2002. The role of the NKG2D immunoreceptor in immune cell tion in solid organ transplant recipients. Transplantation 79: 381–386. activation and natural killing. Immunity 17: 19–29. 2. Polic, B., H. Hengel, A. Krmpotic, J. Trgovcich, I. Pavic, P. Luccaronin, 23. Ibegbu, C. C., Y. X. Xu, W. Harris, D. Maggio, J. D. Miller, and A. P. Kourtis. 2005. Expression of killer cell lectin-like receptor G1 on antigen-specific human S. Jonjic, and U. H. Koszinowski. 1998. Hierarchical and redundant lymphocyte ϩ subset control precludes cytomegalovirus replication during latent infection. CD8 T lymphocytes during active, latent, and resolved infection and its relation J. Exp. Med. 188: 1047–1054. with CD57. J. Immunol. 174: 6088–6094. 24. Moser, J. M., J. Gibbs, P. E. Jensen, and A. E. Lukacher. 2002. CD94-NKG2A 3. Appay, V., P. R. Dunbar, M. Callan, P. Klenerman, G. M. Gillespie, L. Papagno, ϩ G. S. Ogg, A. King, F. Lechner, C. A. Spina, et al. 2002. Memory CD8ϩ T cells receptors regulate antiviral CD8 T cell responses. Nat. Immunol. 3: 189–195. vary in differentiation phenotype in different persistent virus infections. Nat. Med. 25. Snydman, D. R., B. G. Werner, B. Heinze-Lacey, V. P. Berardi, N. L. Tilney, 8: 379–385. R. L. Kirkman, E. L. Milford, S. I. Cho, H. L. Bush, Jr., A. S. Levey, et al. 1987. 4. Appay, V., J. J. Zaunders, L. Papagno, J. Sutton, A. Jaramillo, A. Waters, Use of cytomegalovirus immune globulin to prevent cytomegalovirus disease in P. Easterbrook, P. Grey, D. Smith, A. J. McMichael, et al. 2002. Characterization renal-transplant recipients. N. Engl. J. Med. 317: 1049–1054. of CD4ϩ CTLs ex vivo. J. Immunol. 168: 5954–5958. 26. Uhrberg, M., N. M. Valiante, B. P. Shum, H. G. Shilling, K. Lienert-Weidenbach, 5. van Leeuwen, E. M., E. B. Remmerswaal, M. T. Vossen, A. T. Rowshani, B. Corliss, D. Tyan, L. L. Lanier, and P. Parham. 1997. Human diversity in killer P. M. Wertheim-van Dillen, R. A. van Lier, and I. ten Berge. 2004. Emergence cell inhibitory receptor genes. Immunity 7: 753–763. of a CD4ϩCD28Ϫgranzyme Bϩ, cytomegalovirus-specific T cell subset after 27. Arlettaz, L., J. Villard, C. de Rham, S. Degermann, B. Chapuis, B. Huard, and recovery of primary cytomegalovirus infection. J. Immunol. 173: 1834–1841. E. Roosnek. 2004. Activating CD94:NKG2C and inhibitory CD94:NKG2A re- 6. Kuijpers, T. W., M. T. Vossen, M. R. Gent, J. C. Davin, M. T. Roos, ceptors are expressed by distinct subsets of committed CD8ϩ TCR ␣␤ lympho- P. M. Wertheim-van Dillen, J. F. Weel, P. A. Baars, and R. A. van Lier. 2003. cytes. Eur. J. Immunol. 34: 3456–3464. 4560 NKRs ON CD8ϩ T CELLS DURING PRIMARY CMV

28. Anfossi, N., J. M. Doisne, M. A. Peyrat, S. Ugolini, O. Bonnaud, D. Bossy, ligands display overlapping affinities and thermodynamics. J. Immunol. 174: V. Pitard, P. Merville, J. F. Moreau, J. F. Delfraissy, et al. 2004. Coordinated 2878–2884. expression of Ig-like inhibitory MHC class I receptors and acquisition of cyto- 33. Noppen, C., C. Schaefer, P. Zajac, A. Schutz, T. Kocher, J. Kloth, M. Heberer, toxic function in human CD8ϩ T cells. J. Immunol. 173: 7223–7229. M. Colonna, G. De Libero, and G. C. Spagnoli. 1998. C-type lectin-like receptors 29. Waldrop, S. L., C. J. Pitcher, D. M. Peterson, V. C. Maino, and L. J. Picker. 1997. ϩ in peptide-specific HLA class I-restricted cytotoxic T lymphocytes: differential Determination of antigen-specific memory/effector CD4 T cell frequencies by flow cytometry: evidence for a novel, antigen-specific homeostatic mechanism in expression and modulation of effector functions in clones sharing identical TCR HIV-associated immunodeficiency. J. Clin. Invest 99: 1739–1750. structure and epitope specificity. Eur. J. Immunol. 28: 1134–1142. 30. Cantoni, C., C. Bottino, M. Vitale, A. Pessino, R. Augugliaro, A. Malaspina, 34. Saverino, D., M. Fabbi, F. Ghiotto, A. Merlo, S. Bruno, D. Zarcone, C. Tenca, S. Parolini, L. Moretta, A. Moretta, and R. Biassoni. 1999. NKp44, a triggering M. Tiso, G. Santoro, G. Anastasi, et al. 2000. The CD85/LIR-1/ILT2 inhibitory receptor involved in tumor cell lysis by activated human natural killer cells, is a receptor is expressed by all human T lymphocytes and down-regulates their func- novel member of the immunoglobulin superfamily. J. Exp. Med. 189: 787–796. tions. J. Immunol. 165: 3742–3775. 31. Vitale, M., C. Bottino, S. Sivori, L. Sanseverino, R. Castriconi, E. Marcenaro, 35. D’Andrea, A., C. Chang, J. H. Phillips, and L. L. Lanier. 1996. Regulation of T R. Augugliaro, L. Moretta, and A. Moretta. 1998. NKp44, a novel triggering cell lymphokine production by killer cell inhibitory receptor recognition of self surface molecule specifically expressed by activated natural killer cells, is in- J. Exp. Med. volved in non-major histocompatibility complex-restricted tumor cell lysis. HLA class I alleles. 184: 789–794. J. Exp. Med. 187: 2065–2072. 36. Henel, G., K. Singh, D. Cui, S. Pryshchep, W. W. Lee, C. M. Weyand, and 32. Kaiser, B. K., F. Barahmand-Pour, W. Paulsene, S. Medley, D. E. Geraghty, and J. J. Goronzy. 2006. Uncoupling of T-cell effector functions by inhibitory killer R. K. Strong. 2005. Interactions between NKG2x immunoreceptors and HLA-E immunoglobulin-like receptors. Blood 107: 4449–4457. Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021