Elevated Numbers of FcγRIIIA+ (CD16+) Effector CD8 T Cells with NK Cell-Like Function in Chronic Hepatitis C Virus Infection This information is current as of September 27, 2021. Niklas K. Björkström, Veronica D. Gonzalez, Karl-Johan Malmberg, Karolin Falconer, Annette Alaeus, Greg Nowak, Carl Jorns, Bo-Göran Ericzon, Ola Weiland, Johan K. Sandberg and Hans-Gustaf Ljunggren

J Immunol 2008; 181:4219-4228; ; Downloaded from doi: 10.4049/jimmunol.181.6.4219 http://www.jimmunol.org/content/181/6/4219 http://www.jimmunol.org/ References This article cites 42 articles, 23 of which you can access for free at: http://www.jimmunol.org/content/181/6/4219.full#ref-list-1

Why The JI? Submit online.

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

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

• Fast Publication! 4 weeks from acceptance to publication

*average

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

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2008 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Elevated Numbers of Fc␥RIIIA؉ (CD16؉) Effector CD8 T Cells with NK Cell-Like Function in Chronic Hepatitis C Virus Infection1

Niklas K. Bjo¨rkstro¨m,2* Veronica D. Gonzalez,* Karl-Johan Malmberg,* Karolin Falconer,† Annette Alaeus,† Greg Nowak,‡ Carl Jorns,‡ Bo-Go¨ran Ericzon,‡ Ola Weiland,† Johan K. Sandberg,* and Hans-Gustaf Ljunggren2*

CTL are crucial in the defense against viral infections. In the course of investigating peripheral blood and intrahepatic CD8 T cells in patients with chronic hepatitis C virus (HCV) infection, we observed a significant population of CD8 T cells expressing the Fc␥RIIIA (CD16) . This observation led us to characterize these cells with respect to their phenotype and function in a cohort of patients with chronic HCV infection as well as in healthy blood donors. On average, 10% of peripheral blood CD8 T Downloaded from cells from HCV-infected patients expressed CD16 compared with only a few percent in healthy donors. CD16؉ CD8 T cells displayed a late-stage effector phenotype with high levels of perforin. These cells exhibited a restricted TCR profile suggesting underlying clonal expansion. Stimulation of CD16 on CD8 T cells evoked a vigorous response similar to that of CD16 stimulation in NK cells. Our data suggest that CD8 T cells, during chronic HCV infection in humans, continue to differentiate beyond defined stages of terminal effector cells, acquiring CD16 and NK cell-like functional properties. The Journal of Immunology, 2008, 181: 4219–4228. http://www.jimmunol.org/

ore than 120 million people are estimated to be in- on many NK cells as well as on several other immune cells in- fected with hepatitis C virus (HCV)3 worldwide (1). cluding and (8). CD16 is best characterized M Acute HCV infection can, in some cases, spontane- for its role in Ab-dependent cellular cytotoxicity (ADCC) (8, 9). ously be cleared by a robust multi-specific CD8 response Stimulation of CD16 on NK cells also results in production of (2–4). However, chronic disease frequently develops following several cytokines including IFN-␥ and TNF-␣ (10). HCV infection with high morbidity and mortality, strongly as- The observation, per se, of CD16 expression on a small number sociated with development of liver cirrhosis and hepatocellular of T cells in healthy peripheral blood donors is not new (11). by guest on September 27, 2021 carcinoma (5). Failure to clear the infection is at least in part Already in 1985, Lanier et al. (12) provided the first functional associated with impaired CD8 T cell function, including defec- analysis of such cells in humans. At this time, however, no dis- tive maturation and/or cellular exhaustion, as well as viral es- tinction was made between TCR␣␤ and TCR␥␦ expression, and cape (6, 7). the phenotypic and functional characterization was limited to the In a characterization of CD8 T cells from patients with chronic methods available at that time. Since then, CD16 expression on HCV infection, we observed a significant proportion (often up to TCR␥␦ T cells has been analyzed quite extensively (11, 13–15). 10%) of TCR␣␤ CD8 T cells expressing Fc␥RIII〈 (CD16). CD16 Few studies in the literature have, however, characterized human is a low-affinity FcR for IgG. This receptor is commonly expressed CD16ϩ TCR␣␤ CD8 T cells in healthy individuals and in chronic infection with respect to phenotype and functionality. Our results reveal that a subset of phenotypically terminally dif- *Center for Infectious Medicine and †Department of Medicine, Unit of Infectious ferentiated effector CD16ϩ CD8 T cells are present in elevated Diseases; and ‡Department of Clinical Science, Intervention and Technology, Divi- sion of Transplantation Surgery, Karolinska Institutet, Karolinska University Hospi- numbers in patients with chronic untreated HCV infection and at ϩ tal, Stockholm, Sweden low but clearly detectable numbers in healthy controls. CD16 Received for publication March 26, 2008. Accepted for publication July 9, 2008. CD8 T cells were found not only in peripheral blood but also in the ϩ The costs of publication of this article were defrayed in part by the payment of page liver. The high numbers of CD16 TCR␣␤ CD8 T cells observed charges. This article must therefore be hereby marked advertisement in accordance here in individuals chronically infected with HCV prompted us to with 18 U.S.C. Section 1734 solely to indicate this fact. characterize this subset of cells in detail with respect to matura- 1 This work was supported by grants from the Swedish Foundation for Strategic tional stage, tissue distribution, TCR specificity, and functionality. Research, the Swedish Research Council, the Swedish Cancer Society, the Swedish International Development Cooperation Agency, the Royal Swedish Academy of Sci- ences, the Cancer Society of Stockholm, the Karolinska Institutet, and the Karolinska University Hospital. Materials and Methods 2 Address correspondence and reprint requests to Niklas K. Bjo¨rkstro¨m or Dr. Patient material Hans-Gustaf Ljunggren, Department of Medicine, Center for Infectious Medicine, F59, Karolinska Institutet, Karolinska University Hospital Huddinge, S-141 86 Peripheral blood was obtained from 53 patients chronically infected with Stockholm, Sweden. E-mail addresses: [email protected] or Hans- HCV (genotype 1, n ϭ 16; genotype 2, n ϭ 18; genotype 3, n ϭ 19). All [email protected] patients were treatment naive for their HCV infection and HIV-1 negative. 3 Abbreviations used in this paper: HCV, hepatitis C virus; ADCC, Ab-dependent Median age for the HCV cohort was 54 years. Buffy coats were collected cellular cytotoxicity; IHL, intrahepatic ; KIR, killer cell Ig-like receptor; from 34 healthy blood donors. Median age for the healthy blood donors NKR, NK cell receptor. was 45 years. Liver tissue and whole blood were also acquired from five HCV-infected patients and one patient with HCV-related hepatocellular Copyright © 2008 by The American Association of Immunologists, Inc. 0022-1767/08/$2.00 carcinoma undergoing liver transplantation. This study was approved by www.jimmunol.org 4220 CD16ϩ CD8 T CELLS IN CHRONIC HCV INFECTION

the regional ethics committee (approval nos. 03-541, 2006/971-31/1, and (GolgiSTOP; BD Biosciences), 1/150 dilution, and brefeldin A (Golgi- 2006/229-31/3). Written informed consent was obtained from all patients. PLUG; BD Biosciences), 1/250 dilution, were included after1hofcocul- ture. Following incubation, the cells were washed and surface Ag staining Preparation of PBMC and intrahepatic (IHL) was performed. Following this step, cells were washed twice, fixed, and For isolation of PBMC, whole blood from patients and buffy coats from permeabilized using a Cytofix/Cytoperm (BD Biosciences). Finally, the healthy donors were separated by Lymphoprep gradient centrifugation cells were washed twice in the permeabilization buffer and stained for (Axis-Shield). PBMC were analyzed immediately after isolation or frozen intracellular cytokines for 20 min on ice and then washed twice. Data were in 10% DMSO (Sigma-Aldrich) and 90% FCS (Life Technologies) and immediately acquired after completion of the staining procedure. stored in liquid nitrogen for later analysis. Liver tissues were mechanically disrupted into 1-mm3 fragments, incubated in RPMI 1640 supplemented Long-term culture assay with 0.05% collagenase type I (Worthington Biomedical) and 0.002% PBMC were isolated as previously described, washed twice, and enriched DNase I (Roche) with gentle agitation for 30 min at 37°C, then filtered for CD8 T cells by negative selection using a CD8 T cell isolation kit 2 through a 100-␮m nylon mesh, washed twice, and separated by Lym- from Miltenyi Biotec. Next, the enriched CD8 T cells were stained with phoprep gradient centrifugation for isolation of IHL. IHL were stained and anti-CD3, anti-CD4, anti-CD8, anti-CD45RA, anti-CD27, and anti-CD16 analyzed immediately after isolation. and then sorted using a FACSAria instrument (BD Biosciences). In brief, 1 ϫ 105 of each purified CD8 T cell subset, naive T cells (CD16ϪCD27ϩ mAbs and tetramers ϩ Ϫ ϩ Ϫ CD45RA ), central memory T cells (CD16 CD27 CD45RA ), and ef- Ϫ Ϫ ϩ Anti-CD3 Pacific Blue and anti-CD3 Cascade Yellow were obtained from fector memory T cells (CD16 CD27 CD45RA ) were cultured in RPMI DakoCytomation. Anti-CD56 PE-Cy7, anti-CD16 Alexa Fluor 647, anti- 1640 medium supplemented with 10% FCS and 250 IU/ml rIL-2 (Chiron) CD16 Pacific Blue, anti-CD14 allophycocyanin-Cy7, anti-CD8 PerCP, anti- for 3 or 7 days in 96-well round-bottom plates at 37°C and 5% CO2.Atthe CD8 PE-Cy7, anti-CCR7 PE-Cy7, anti-CD28 allophycocyanin, anti-CD28 end of the culture, the cells were stained for CD3, CD8, CD16, and the

FITC, anti-CD27 FITC, anti-CD27 allophycocyanin-Cy7, anti-TCR␣␤ nucleic acid stain 7-aminoactinomycin D, counted, and analyzed for CD16 Downloaded from FITC, anti-TCR␥␦ FITC, anti-perforin FITC, anti-CD3 PE, anti-CD3 al- expression on the CyAn ADP instrument. lophycocyanin, anti-CD3 PE-Cy7, anti-granzyme B Alexa Fluor 647, anti- NKG2D PE, anti-KIR3DL1 FITC, anti-CD38 allophycocyanin, and anti- Results CD107a FITC were purchased from BD Biosciences. Anti-CD56 PE-Cy5, Elevated numbers of CD16ϩ CD8 T cells in patients with anti-KIR2DL2/3 PE, anti-NKG2A PE, anti-2B4 PE, and anti-CD57 FITC were obtained from Beckman Coulter. Anti-KIR2DL1 allophycocyanin chronic HCV infection and anti-NKG2C PE were purchased from R&D Systems. Anti-CD7 PE

In the course of analysis of CD8 T cells in patients with chronic http://www.jimmunol.org/ was obtained from Coulter Immunotech. Anti-CD45RA Pacific Blue was from Invitrogen. Anti-CD27 allophycocyanin-Cy7 was from Biolegend. HCV infection, we observed that a large proportion of the CD8 T IgG1, IgG2b, and IgM isotypes were from BD Biosciences. Intracellular cell pool expressed CD16 (Fig. 1A). This finding prompted us to ϩ cytokines were visualized with anti-IFN-␥ PE (BD Biosciences) and anti- study the relative frequencies and absolute numbers of CD16 TNF-␣ Alexa Fluor 647 from eBioscience. The TCR V␤ kit FITC/PE was CD8 T cells in a larger cohort encompassing 53 patients chroni- from Beckman Coulter. Polystyrene beads (BD Biosciences) were used for cally infected with HCV and 34 healthy controls. These studies compensation purposes. ϩ Subjects were evaluated for HLA-A2 expression by staining with anti- revealed a marked increase in frequency of CD16 CD8 T cells in HLA-A2 (BD Biosciences). CD8 T cells specific for viruses were identified HCV-infected individuals compared with healthy controls (mean, and enumerated using allophycocyanin-conjugated HLA-A2 tetrameric 10.1 Ϯ 9.5% CD16ϩ CD8 T cells in HCV-infected individuals; ϩ complexes refolded with the CMV pp65 epitope NLVPMVATV, HCV mean, 3.8 Ϯ 4.6% CD16 CD8 T cells in healthy controls; p Ͻ by guest on September 27, 2021 NS3 1073 epitope CIMGVCWTV, and the HCV NS3 1406 epitope KLVALGINAV (Coulter Immunotech). 0.0001) (Fig. 1B). Since the percentage out of total CD8 T cells is a relative mea- surement, we also determined absolute numbers of CD16ϩ CD8 T For cell surface staining, purified PBMC or IHL were incubated in PBS cells in peripheral blood in patients with chronic HCV infection with 5% FCS for 20 min on ice in the dark with saturating amounts of the and healthy controls (Fig. 1C). A 5-fold increase in absolute num- desired mAb combination, washed twice, and fixed in PBS containing 1% bers of CD16ϩ CD8 T cells were observed in patients with chronic formaldehyde. Isotype control mAb were used at corresponding concen- HCV infection compared with healthy controls (mean, 29.9 Ϯ 38.8 trations. When intracellular staining was conducted, cells were permeabil- ␮ ϩ ized with Cytofix/Cytoperm (BD Biosciences), washed in permeabilization cells/ l CD16 CD8 T cells in HCV-infected individuals; mean, ϩ buffer, and then stained with intracellular Abs for 20 min on ice and finally 5.8 Ϯ 3.8 cells/␮l CD16 CD8 T cells in healthy controls; p Ͻ washed twice. Data were acquired on a CyAn ADP nine-color flow cy- 0.01). When comparing expression levels of CD16 in HCV-in- tometer (Beckman Coulter) and analyzed with FlowJo software version 8 fected patients and healthy donors, there was a trend toward higher (Tree Star) as previously described (16). levels of CD16 present on CD8 T cells in the patient group (Fig. Absolute T cell counts 1D) (HCV: n ϭ 25, mean, 73.3 Ϯ 36.0; healthy: n ϭ 9, mean, Ϯ ϭ MultiTEST CD3 FITC, CD8 PE, CD45 PerCP, and CD4 allophycocyanin 51.6 18.2; p 0.06). Notably, compared with the high expres- reagent (BD Biosciences) and heparinized whole blood were mixed in a sion levels of CD16 on NK cells, expression levels of CD16 were TrueCount tube (BD Biosciences) and incubated for 20 min at room lower on CD8 T cells from HCV-infected patients as well as temperature in the dark. Next, FACS lysing solution (BD Biosciences) healthy controls (Fig. 1E). was added and the samples were incubated for another 30 min at room Taken together, these results demonstrate the presence of a ma- temperature in the dark. The samples were immediately acquired on a ϩ FACSCalibur four-color flow cytometer (BD Biosciences) and analyzed jor CD16 CD8 T cell population in HCV-infected individuals, with CellQuest software. representing ϳ10% of the total CD8 T cell pool. This observation led us to characterize CD16ϩ CD8 T cells in more detail with Functional CD16 assay regard to phenotype and function. PBMC were maintained in complete medium (RPMI 1640 supplemented with 100 ␮g/ml L-glutamine, 10% heat-inactivated FCS, 100 U/ml peni- CD16ϩ CD8 T cells are terminally differentiated effector ␮ cillin G, and 100 g/ml streptomycin). For Ab coating, P815 cells (Amer- T cells ican Type Culture Collection) propagated in complete medium were incu- bated at 1 ϫ 106 cells/ml for 30 min at room temperature with 10 ␮g/ml To determine the TCR constitution of CD16ϩ CD8 T cells, PBMC anti-CD16 mAb or the corresponding isotype control and then washed from eight blood donors were stained with Abs against TCR␣␤ twice. PBMC were then mixed with P815 cells at a ratio of 10:1 in round- ␥␦ ϩ bottom 96-well plates in a final volume of 200 ␮l. Anti-CD107a FITC and and TCR . The CD16 CD8 T cells identified exclusively ex- a corresponding IgG1 isotype control were added at the beginning of the pressed the TCR␣␤ (Fig. 2A). Skewed maturation of Ag-specific

assay and then cells were incubated at 37°C and 5% CO2 for 6 h. Monensin CD8 T cells is a hallmark of persistent viral infections (17). To The Journal of Immunology 4221 Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 1. CD16ϩ CD8 T cells are expanded in patients with chronic HCV infection. A, Flow cytometric identification of CD16ϩ CD8 T cells excluding doublet events (FSC Lin vs FSC Area) and CD14ϩ cells in a HCV-infected patient. B, Frequency (percent) of CD16ϩ CD8 T cells among total CD8 T cells in healthy and HCV-infected individuals. C, Absolute numbers (cells/␮l) of CD16ϩ CD8 T cells in healthy and HCV-infected patients. D, Expression of CD16 on CD8 T cells from one representative HCV-infected patient and one representative healthy control. E, Expression of CD16 and CD3 on CD56ϩCD3-NK cells (blue contour plot) and CD8 T cells (red contour plot) from one HCV-infected patient. An unpaired t test was used for statistical .p Ͻ 0.0001. FSC, Forward scatter ,ءءء p Ͻ 0.01 and ,ءء ;analysis in B and the Mann-Whitney U test in C assess the maturational status of the CD16ϩ CD8 T cell subset, we cells from HCV-infected patients displayed a profile similar to that stained this population for CCR7, CD45RA, CD27, CD28, and of CD16ϩ CD8 T cells obtained from healthy blood donors (data CD7. CD16ϩ CD8 T cells displayed a uniform differentiation pro- not shown). file when compared with CD16Ϫ CD8 T cells (Fig. 2B). Eighty- To further elucidate the phenotypic profile of CD16ϩ CD8 T four percent of the CD16ϩ CD8 T cells were CCR7ϪCD45RAϩ cells, different subsets of CD16Ϫ CD8 T cells were sorted on the compared with 27% among the CD16Ϫ CD8 T cells ( p Ͻ 0.001). basis of differentiation status and put in cultures supplemented Only 24% of the CD16ϩ CD8 T cells expressed the costimulatory with a low dose of IL-2 to support cellular survival. Terminally receptors CD27 and CD28, whereas 85% of the CD16Ϫ CD8 T differentiated (CD45RAϩCD27Ϫ) CD16Ϫ CD8 T cells showed ev- cells expressed CD27 or CD28 ( p Ͻ 0.001). Furthermore, CD16ϩ idence of CD16 surface expression after 3 days in culture and CD8 T cells expressed low levels of CD7 on their surface (Fig. 2, levels of CD16 expression were further elevated at day 7 (Fig. 2F). B and C). In contrast, neither naive (CD45RAϩCD27ϩ) nor central memory Effector T cells characteristically express intracellular effector (CD45RAϪCD27ϩ) CD16Ϫ CD8 T cells up-regulated CD16 upon molecules, including perforin and granzymes. A majority of the culture. CD16ϩ CD8 T cells expressed perforin and virtually all CD16ϩ We also assessed whether the increased frequency of CD16ϩ CD8 T cells contained granzyme B, as detected by intracellular CD8 T cells found in patients with HCV infection correlated with staining (Fig. 2, D and E). Significantly higher frequencies of per- markers of T cell activation. CD8 T cells from healthy and HCV- forin and granzyme Bϩ cells were found among the CD16ϩ CD8 infected patients were investigated for CD38 expression. CD8 T T cells compared with CD16Ϫ CD8 T cells (perforin expression: cells from HCV-infected patients expressed significantly higher 68.6 Ϯ 19.6% vs 23.7 Ϯ 16.9%, p Ͻ 0.0001; granzyme B expres- frequencies of CD38 compared with healthy (mean, 5.74 Ϯ 4.7% sion: 92.7 Ϯ 5.5% vs 33.2 Ϯ 14.9%, p Ͻ 0.005). CD16ϩ CD8 T vs 2.5 Ϯ 1.63%, p Ͻ 0.05). CD38 expression correlated positively 4222 CD16ϩ CD8 T CELLS IN CHRONIC HCV INFECTION Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 2. CD16ϩ CD8 T cells are late-stage effector T cells. A, Expression of TCR␣␤ on CD16Ϫ and CD16ϩ CD8 T cells of one representative donor. B, Expression of CD45RA, CCR7, CD27, CD28, and CD7 on CD16Ϫ (black density plot background) and CD16ϩ (red contour plot overlay) CD8 T cells. C, Compilation of expression patterns of CD45RA and CCR7, CD27 and CD28, and CD27 and CD7 on CD16Ϫ and CD16ϩ CD8 T cells from six donors p Ͻ 0.001). D, Expression of perforin and granzyme B on ,ءءء p Ͻ 0.05 and ,ء ,two-way ANOVA with Bonferroni post tests for statistical evaluation) CD16Ϫ (black density plot background) and CD16ϩ (red contour plot overlay) CD8 T cells. E, Compilation of expression patterns of perforin and granzyme p Ͻ 0.001. F, Expression of CD16 on sorted ,ءءء p Ͻ 0.01 and ,ءء ;B. Paired t test was used when n Ն15 and Wilcoxon matched pairs test when n Ͻ15 naive (CD45RAϩCD27ϩ), central memory (CD45RAϪCD27ϩ), and terminally differentiated (CD45RAϩCD27Ϫ) CD16-negative CD8 T cells after 3 or 7 days in culture. One representative experiment of three independent experiments is shown. G, Compilation of expression of CD38 compared with CD16 on CD8 T cells from 18 healthy donors. H, Compilation of expression of CD38 compared with CD16 on CD8 T cells from 10 HCV-infected patients. Linear regression was used for statistical evaluation (G and H). The Journal of Immunology 4223

FIGURE 3. Frequency and pheno- type of intrahepatic CD16ϩ CD8 T cells in patients with chronic HCV in- fection. A, Frequency of CD16ϩ CD8 T cells among total CD8 T cells in PBL and IHL. B, Expression pattern of CCR7, CD27, CD28, and perforin on CD16ϩ CD8 T cells from PBL and IHL determined by eight-color flow cyto- metric analysis and Boolean gating summarized for six patients with chronic HCV infection. Mann-Whitney U test used for statistical comparison. .p Ͻ 0.01 ,ءء ;n.s., Not significant with CD16 expression in both healthy and HCV-infected individ- from peripheral blood in the same patients exhibited a terminally uals (Fig. 2, G and H). differentiated phenotype (CCR7ϪCD27ϪCD28Ϫperforinϩ; Fig. In summary, CD16ϩ CD8 T cells exhibited a homogeneous dif- 3B). These data suggest that a similar high number of CD16ϩ CD8 ferentiation profile with the majority of the cells being T cells found in peripheral blood of HCV-infected individuals is Downloaded from CCR7ϪCD45RAϩCD27ϪCD28ϪCD7low/neg. Furthermore, CD16ϩ found also at the site of active HCV replication. However, intra- CD8 T cells contained granules with effector molecules. Collec- hepatic CD16ϩ CD8 T cells exhibit a unique phenotypical profile tively, these data strongly suggest that CD16ϩ CD8 T cells found expressing both costimulatory and effector molecules as well as a in peripheral blood are terminally differentiated effector CD8 T receptor associated with entry into secondary lymphoid tissues. cells. http://www.jimmunol.org/ Distinct phenotype of intrahepatic CD16ϩ CD8 T cells TCR V␤ repertoire analysis of CD16ϩ CD8 T cells discloses HCV exhibits strong tropism for hepatocytes. Therefore, we in- a restricted repertoire vestigated whether CD16ϩ CD8 T cells also resided in the liver of A key step in the eradication of viral infections is clonal expansion patients with chronic HCV infection. For this purpose, liver sam- of Ag-specific CD8 T cells. To compare the relative levels of ples and peripheral blood were obtained from HCV-infected pa- clonality among CD16ϩ and CD16Ϫ CD8 T cell subsets, we as- tients undergoing liver transplantation due to HCV-related liver sessed the TCR V␤ distribution in these populations. The TCR disease. Similar frequencies of CD16ϩ CD8 T cells were found in composition of different CD8 T cell subsets was determined using the liver and peripheral blood in these patients (6.8 Ϯ 3.3% vs an array of 24 Abs specific for different TCR V␤. The repertoire 7.5 Ϯ 5.4%; Fig. 3A). In contrast to peripheral blood, the major of CD16ϩ CD8 T cells was, in the majority of donors studied, by guest on September 27, 2021 phenotype of intrahepatic CD16ϩ CD8 T cells was dominated by a single (or a few) TCR V␤ (Fig. 4A). This was CCR7ϩCD27ϩCD28ϩperforinϩ while the CD16ϩ CD8 T cells in contrast to the broad TCR V␤ repertoires of naive

B A 50 Example 1 CD16-CD45RA+CCR7+ *** * ** CD16-CD45RA+/-CCR7- 60 CD16+ 25

40 0 50 Example 2 20

25 0 CD16- CD16- CD16+

within the most represented Vbeta CD45RA+ CD45RA+/-

0 cells T Frequency (%) of positive CD8 CCR7+ CCR7- 50 Example 3

25 Frequency (%) of positive cells within CD8+ T cells T Frequency (%) of positive cells within CD8+ 0

Vbeta 1 2 3 4 5.1 5.2 5.3 7.1 7.2 8 9 11 12 13.1 13.2 13.6 14 16 17 18 20 21.3 22 23

FIGURE 4. CD16ϩ CD8 T cells exhibit a highly restricted TCR repertoire. A, TCR V␤ chain repertoire analysis of CD16ϩ, CD16ϪCD45RAϩCCR7ϩ and CD16ϪCD45RA/ϪCCR7Ϫ CD8 T cells from three representative donors. B, Frequency of the most prominent V␤ within CD16ϩ, ;p Ͻ 0.05 ,ء ;CD16ϪCD45RAϩCCR7ϩ, and CD16ϪCD45RAϩ/ϪCCR7Ϫ CD8 T cells, n ϭ 11. Wilcoxon matched pairs test used for statistical comparison .p Ͻ 0.001 ,ءءء p Ͻ 0.01; and ,ءء 4224 CD16ϩ CD8 T CELLS IN CHRONIC HCV INFECTION

FIGURE 5. CD16 stimulation trig- gers a multifunctional response profile in CD16ϩ CD8 T cells. A, CD107a,

IFN-␥, and TNF-␣ production from Downloaded from one representative donor and a Bool- ean platform used to determine multi- functional responses. B, Frequency of CD16ϩ CD8 T cells and NK cells re- sponding with CD107a, IFN-␥, and TNF-␣. C, Responses are grouped ac- cording to level of multifunctionality. http://www.jimmunol.org/ D, Complete multifunctional response profile of responding cells. B–D, n ϭ 7. Wilcoxon matched pairs test used for statistical comparison (B and D); .p Ͻ 0.05 ,ء by guest on September 27, 2021

(CD16ϪCD45RAϩCCR7ϩ) and effector memory/terminally dif- Taken together, the data presented suggest that CD16ϩ CD8 T ferentiated (CD16ϪCD45RAϩ/ϪCCR7Ϫ) CD8 T cells. The dom- cells have a restricted TCR repertoire compared with CD16Ϫ ef- inant TCR V␤ in the CD16ϩ CD8 T cell subset differed between fector memory and terminally differentiated CD8 T cells. This in- donors (dominant V␤13.1, in 3 of 11 donors; dominant V␤3, in 3 dicates that a substantial degree of clonal expansion have taken of 11 donors; dominant V␤2, -5.1, -7.2, -13.2, and -17, in 1 of 11 place within the CD16ϩ CD8 T cell subset. donors). The frequency of the dominant V␤ ranged from 8 to 59%, Ϯ ϩ CD16 stimulation triggers a multifunctional response with a mean of 32 19% in the CD16 CD8 T cells compared ϩ with 9.3 Ϯ 4.9% for CD16Ϫ effector memory/terminally differen- in CD16 CD8 T cells tiated and 5.0 Ϯ 1.2% for CD16Ϫ naive CD8 T cells (CD16ϩ CD8 Following the assessment of frequencies and phenotypic charac- T cells vs CD16Ϫ naive CD8 T cells, p ϭ 0.001; CD16ϩ CD8 T teristics of CD16ϩ CD8 T cells, we set out to investigate the re- cells vs CD16Ϫ effector CD8 T cells, p ϭ 0.002; Fig. 4B). sponse evoked upon specific triggering of CD16 on such cells. For HCV-specific CD8 T cells are notoriously hard to identify in this purpose, CD16ϩ CD8 T cells from healthy donors were used. peripheral blood of infected individuals using specific tetramers P815 cells coated with anti-CD16 mAb were used and a multi- (18). In the present study, we failed to detect HCV-specific CD8 T functional readout system simultaneously monitoring degranula- cells using HCV-specific HLA-A2 tetramers in HLA-A2ϩ tion (CD107a up-regulation), IFN-␥, and TNF-␣ expression was patients. designed (Fig. 5A, top row). The responses were compared with The Journal of Immunology 4225 Downloaded from

FIGURE 6. Expression of NK receptors on CD16ϩ CD8 T cells from patients with HCV infection. A, Expression of 2B4 and NG2D (black lines) compared with isotype control (gray solid line), (B) KIR2DL1, KIR2DL2/3, KIR3DL1, (C) CD56 and CD57, and (D) NKG2A and NKG2C, all on CD16ϩ CD8 T cells from one representative HCV-infected patient. http://www.jimmunol.org/ responses by donor-matched NK cells stimulated in a similar fash- CD16Ϫ: 24.7 Ϯ 7.6; n ϭ 18, p Ͻ 0.001). On the contrary, CD16ϩ ion. CD16ϩ CD8 T cells readily responded with CD8 T cells expressed lower levels of NKG2D compared with (CD107aϩ cells, 34.9 Ϯ 19.4%) and cytokine production (IFN-␥ϩ CD16Ϫ CD8 T cells (CD16ϩ: mean,78.1 Ϯ 14.8; CD16Ϫ: 87.7 Ϯ cells, 21.0 Ϯ 14.5%; TNF-␣ϩ cells, 19.7 Ϯ 13.4%) after6hof 10.7; n ϭ 18, p Ͻ 0.01). Subsets of CD16ϩ CD8 T cells were stimulation. No statistically significant differences with respect to positive for inhibitory killer cell Ig-like receptors (KIRs) (Fig. 6B). frequencies of responding cells were observed when compared A Boolean platform analysis was performed simultaneously as- with NK cells (Fig. 5B). There was, however, a tendency toward sessing KIR2DL1, KIR2DL2/3, and KIR3DL1 expression on by guest on September 27, 2021 less degranulation in CD16ϩ CD8 T cells as compared with NK CD16ϩ CD8 T cells in 25 HCV-infected patients. With the ex- cells. ception of KIR2DL2/3ϩKIR2DL1ϪKIR3DL1Ϫ cells (mean, Robust multifunctional T cell responses have recently been as- 27.8 Ϯ 25.6%), none of the other six subsets represented Ͼ5% of sociated with slower disease progression in chronic infections in the CD16ϩ CD8 T cells. On average, 60% of the CD16ϩ CD8 T humans and with protection in vaccination models (19, 20). We cells were negative for KIRs. therefore asked whether CD16ϩ CD8 T cells have the capacity to In line with a terminally differentiated phenotype, roughly 80% respond with two or more functions simultaneously. Indeed, of the CD16ϩ CD8 T cells expressed the senescent marker CD57 ϳ33% of the CD16ϩ CD8 T cells responded with concurrent (Fig. 6C) (21). CD57-positive cells were significantly more fre- CD107a up-regulation, IFN-␥, and TNF-␣ production or any two quent within CD16ϩ CD8 T cells compared with CD16Ϫ CD8 T combinations of these three functions (Fig. 5C). No significant cells (CD16ϩ: mean, 76.8 Ϯ 13.5%; CD16Ϫ: mean, 24.9 Ϯ ϩ differences in multifunctionality existed between CD16 CD8 T 13.2%; n ϭ 25, p Ͻ 0.001). A characteristic of activated T cells is cells and NK cells upon CD16 triggering. Production of IFN-␥ and surface up-regulation of CD56 along with the acquisition of NK TNF-␣ without simultaneous degranulation was more common cell-like properties (22). We therefore evaluated CD56 expression ϩ among the CD16 CD8 T cells than in NK cells (Fig. 5D). on the CD16ϩ CD8 T cells from HCV-infected patients and These results demonstrate that stimulation through CD16, in the healthy controls (Fig. 6C). Roughly 40% of the CD16ϩ CD8 T absence of TCR triggering, is sufficient for induction of cytokine cells from HCV-infected patients expressed CD56. CD56 expres- production and degranulation with similar levels of multifunction- sion was more frequent on CD16ϩ CD8 T cells from HCV-in- ality as compared with NK cells. This also illustrates that the levels fected patients compared with healthy individuals (HCV: n ϭ 35, of CD16 expressed on CD8 T cells are sufficient for eliciting bi- mean 41.0 Ϯ 20.3%; healthy n ϭ 27, mean 30.8 Ϯ 19.8%; p Ͻ ological responses. 0.05). Furthermore, CD56 expression was much lower on CD16Ϫ ϭ Ϯ ϩ CD8 T cells in both cohorts (HCV: n 35, mean 12.6 8.4%, NK receptor expression profile of CD16 CD8 T cells healthy: n ϭ 27, mean 7.4 Ϯ 7.5%). Finally, minor populations Since CD16ϩ CD8 T cells responded to CD16 triggering indepen- within the CD16ϩ CD8 T cell subset in HCV-infected patients dently of the TCR with a profile similar to that of NK cells, we expressed inhibitory CD94/NKG2A (mean, 9.6 Ϯ 10.9%) and ac- finally set out to determine the expression profile of major acti- tivating CD94/NKG2C (mean, 11.8 Ϯ 12.7%) receptors (Fig. 6D). vating and inhibitory NK cell receptors on the CD16ϩ CD8 T cells These data suggest that CD16ϩ CD8 T cells in HCV-infected in HCV-infected patients. All CD16ϩ CD8 T cells showed evi- individuals express a wide array of different NK cell-associated dence of 2B4 and NKG2D surface expression (Fig. 6A). Interest- activating and inhibitory receptors. The expression of these recep- ingly, 2B4 was more highly expressed on CD16ϩ CD8 T cells tors is in most, but not all cases, more prominent on CD16ϩ CD8 compared with CD16Ϫ CD8 T cells (CD16ϩ: mean, 40.2 Ϯ 8.3; T cells compared with CD16Ϫ CD8 T cells. Future studies will 4226 CD16ϩ CD8 T CELLS IN CHRONIC HCV INFECTION have to address the importance of the clustering of NK cell-asso- infection show evidence for clonal expansion as well as TCR- ciated receptors on CD16ϩ CD8 T cells in patients with HCV independent effector functions. In line with the up-regulation of infection. multiple activating NKR on intraepithelial CTL in patients with celiac disease, we were able to detect expression and accumulation Discussion of activating NKR on CD16ϩ CD8 T cells from HCV-infected In the course of studying CD8 T cells in patients with chronic patients. Interestingly, we could also find expression of some in- HCV infection, we discovered a sizable fraction of Fc␥RIII〈 hibitory NKR but at low frequencies. When intrahepatic IL-15 (CD16)-expressing CD8 T cells. This unique population of CD8 T levels were investigated, we could not detect a difference between cells was significantly expanded both in relative frequencies (per- HCV-infected and healthy individuals (N.K.B., unpublished ob- centages) and absolute numbers in peripheral blood of HCV-in- servations). It is thus less likely that dysregulated IL-15 expression fected patients as compared with healthy subjects. CD16ϩ CD8 T is responsible for the expansion of the CD16ϩ CD8 T cell subset. cells exhibited a terminally differentiated phenotype and had a CD8 T lymphocytes have a central role in the resolution of viral highly restricted TCR V␤ repertoire. Moreover, CD16 functioned infections. Naive CD8 T cells encounter Ag in secondary lym- independently of the TCR and yielded a prominent multifunctional phoid organs and activation by Ag induces proliferation and dif- response consisting of both effector cytokine production and de- ferentiation into diverse effector and memory stages. Complex ex- granulation. Taken together, the findings indicate that an effector pression patterns of cell surface maturation markers have been CD8 T cell population with NK cell-like properties can persist and used to describe and identify different maturational stages within expand in chronic HCV infection, despite that this infection is the CD8 T cell pool (17, 37). Naive and central memory CD8 T known to be associated with T cell exhaustion (23). cells expresses CCR7 and CD62L, both necessary for homing to Downloaded from The function of CD16 has been best studied on NK cells. These lymph nodes, whereas effector memory and terminally differ- cells are regulated by the integrated input from a wide array of entiated effector CD8 T cells lack expression of these markers activating and inhibitory NK cell receptors (NKR) (24, 25). The (37). A second model for T cell differentiation employs the combined input from at least two activating receptors is necessary presence and subsequent stepwise down-regulation of the co- for granular polarization and release with subsequent cytotoxicity stimulatory molecules CD27, CD28, and CD7 along with ϩ ϩ high ϩ in resting NK cells. CD16 is the only activating receptor on resting CD45RA to define naive (CD27 CD28 CD7 CD45RA ), http://www.jimmunol.org/ NK cells that can autonomously trigger these events (25). In hu- memory (CD27ϩ/ϪCD28ϩ/ϪCD7high/lowCD45RAϪ), and effec- man NK cells, CD16 associates with FcR␥ or CD3␨ homodimers tor (CD27ϪCD28ϪCD7low/negCD45RAϩ) CD8 T cells (38, 39). for signal transduction (26, 27). CD3␨ is the major subunit utilized In the present study, we used five surface markers to define in by the TCR complex for signal transduction. FcR␥-containing detail the maturational status of CD16ϩ CD8 T cells. Interest- TCR complexes have been reported in rare subsets of T cells such ingly, CD16ϩ CD8 T cells exhibited a distinct and homoge- as NKT cells, activated ␥␦ T cells, and double-negative T cells, but neous profile. The combination of the two models presented not in CD8ϩTCR␣␤ϩ T cells (28). Therefore, TCR-independent above strongly suggests that the CD16ϩ CD8 T cells are ter- CD16 signaling in CD16ϩ CD8 T cells is likely to occur through minally differentiated. This observation is further strengthened CD3␨. by the significant intracellular accumulation of cytolytic effec- by guest on September 27, 2021 Many NKR are not restricted in their expression to NK cells but tor molecules that was found in the CD16ϩ CD8 T cell subset. can also be found on CD8 T cells (22, 29–31). In humans, this HCV exhibits strong tropism for hepatocytes. Consequently, includes KIR CD94/NKG2A, 2B4, and NKG2D and, more re- HCV-related morbidity is usually manifested as a progressive liver cently, in the context of HCV infection CD161 (32). Some NKR inflammation with fibrosis development finally leading to end- are uniformly induced on CD8 T cells responding to Ag, whereas stage liver disease. In an attempt to address the clinical signifi- others are restricted to subsets of differentiated CD8 T cells. NKR cance of elevated numbers of CD16ϩ CD8 T cells in HCV infec- on CD8 T cells are in most situations dependent on TCR signaling tion, HCV viral load and serum alanine aminotransferase levels for their function and works as utensils modulating T cell re- were compared and correlated to frequencies of CD16ϩ CD8 T sponses (22, 29, 32–34). In contrast, CD16 seems to be functional cells in the HCV-infected patients. Neither HCV viral load nor on its own in CD8 T cells evoking prominent cytokine production serum alanine aminotransferase correlated with higher frequencies and high levels of degranulation. This indicates that CD16 on T of CD16ϩ CD8 T cells found in HCV-infected individuals (data cells works in a similar fashion as CD16 expressed by NK cells, not shown). However, we were able to detect similar levels of i.e., without the necessity for combined input from multiple acti- CD16ϩ CD8 T cells in the liver as compared with peripheral vating receptors. Furthermore, it is interesting to note that the mag- blood. It has previously been shown that intrahepatic and periph- nitude of responses evoked by CD16ϩ CD8 T cells was very sim- eral blood HCV-specific CD8 T cells differ in phenotype, with a ilar to those of NK cells. predominance of effector memory (CCR7ϪCD45RAϪ) CD8 T It has been suggested that the stepwise transformation of CTL cells in the liver and a more diverse profile found in peripheral into more “innate-like” cells with NK cell properties is one un- blood (18). Furthermore, defective maturation of HCV-specific derlying mechanism for immunopathology in chronic inflamma- CD8 T cells may be associated with imperfect down-regulation of tory diseases (35). Some evidence for this comes from the repro- CD27 and CD28 (40). Interestingly, the ambiguous profile of in- gramming occurring within intraepithelial CTL in patients trahepatic CD16ϩ CD8 T cells with both immature markers (CD27 suffering from celiac disease. Multiple activating NKR, including and CD28) and effector molecules is more similar to that of HCV- NKG2C, NKG2D, NKp44, and NKp46, appear to be strongly up- specific CD8 T cells. We have not had the possibility to function- regulated on oligoclonally expanded IFN-␥-producing effector T ally characterize intrahepatic CD16ϩ CD8 T cells due to scarcity cells (35, 36). Dysregulated IL-15 expression was suggested to of primary human liver material. Possibly, costimulation received govern this transformation into NK-like lymphokine-activated through CD27 or CD28 would further elevate the already strong killer cells (36). A hallmark of chronic progressing HCV infection functional response evoked upon specific CD16 triggering. Be- is hepatic inflammation yielding fibrosis development and even- cause CD16ϩ CD8 T cells found in peripheral blood to a large tually liver cirrhosis. Similar to reprogrammed CD8 T cells in extent lack costimulatory molecules, we have not been able to celiac disease, CD16ϩ CD8 T cells expanded in chronic HCV address this hypothesis using peripheral blood-derived cells. The Journal of Immunology 4227

The phenomenon of T cell exhaustion has been extensively 5. Tong, M. J., N. S. el-Farra, A. R. Reikes, and R. L. Co. 1995. Clinical outcomes studied in two major human chronic viral infections, HIV-1 and after transfusion-associated hepatitis C. N. Engl. J. Med. 332: 1463–1466. 6. Wherry, E. J., and R. Ahmed. 2004. Memory CD8 T-cell differentiation during HCV. One characteristic affecting CD8 T cells is the hierarchical viral infection. J. Virol. 78: 5535–5545. loss of T cell functions. The ability to produce IL-2 and convey 7. Bowen, D. G., and C. M. Walker. 2005. Mutational escape from CD8ϩ T cell immunity: HCV evolution, from chimpanzees to man. J. Exp. Med. 201: cytotoxicity are lost early in the process followed by deficient 1709–1714. TNF-␣ expression, whereas IFN-␥ production is the function re- 8. Ravetch, J. V., and S. Bolland. 2001. IgG Fc receptors. Annu. Rev. Immunol. 19: maining intact for the longest time (23). On the other hand, pa- 275–290. 9. Lanier, L. L., A. M. Le, J. H. Phillips, N. L. Warner, and G. F. Babcock. 1983. tients who control (HIV-1) or clear (HCV) their chronic infection Subpopulations of human natural killer cells defined by expression of the Leu-7 exhibit highly multifunctional responses much similar to those of (HNK-1) and Leu-11 (NK-15) . J. Immunol. 131: 1789–1796. T cells that respond and clear acute infections or suppress patho- 10. Bryceson, Y. T., M. E. March, D. F. Barber, H. G. Ljunggren, and E. O. Long. 2005. Cytolytic granule polarization and degranulation controlled by different gens into latency (19). In this context, we evaluated the functional receptors in resting NK cells. J. Exp. Med. 202: 1001–1012. profile obtained after triggering CD16 on CD8 T cells. Interest- 11. Sandor, M., and R. G. Lynch. 1993. Lymphocyte Fc receptors: the special case ingly, high frequencies of CD16ϩ CD8 T cells responded with of T cells. Immunol. Today 14: 227–231. 12. Lanier, L. L., T. J. Kipps, and J. H. Phillips. 1985. Functional properties of a degranulation and/or TNF-␣ production. Furthermore, ϳ35% of unique subset of cytotoxic CD3ϩ T lymphocytes that express Fc receptors for the CD16ϩ CD8 T cells also possessed the ability to respond with IgG (CD16/Leu-11 ). J. Exp. Med. 162: 2089–2106. 13. Groh, V., S. Porcelli, M. Fabbi, L. L. Lanier, L. J. Picker, T. Anderson, two or more functions simultaneously. R. A. Warnke, A. K. Bhan, J. L. Strominger, and M. B. Brenner. 1989. Human In a model with self-Ag-specific murine T cells, Dhanji et al. lymphocytes bearing T cell receptor ␥/␦ are phenotypically diverse and evenly (41) reported the specific expression of CD16 on a subset of distributed throughout the lymphoid system. J. Exp. Med. 169: 1277–1294. ϩ 14. Lafont, V., J. Liautard, J. P. Liautard, and J. Favero. 2001. Production of TNF-␣ ␣␤ Downloaded from murine TCR CD8 T cells. Similar to our observation that by human V␥9V␦2 T cells via engagement of Fc␥RIIIA, the low affinity type 3 ϩ the CD16 CD8 T cells expanded in chronic HCV infection receptor for the Fc portion of IgG, expressed upon TCR activation by nonpeptidic exhibit a late-stage phenotype, CD16 expression in this murine antigen. J. Immunol. 166: 7190–7199. high 15. Angelini, D. F., G. Borsellino, M. Poupot, A. Diamantini, R. Poupot, G. Bernardi, model was confined to the CD44 CD8 T cells of memory F. Poccia, J. J. Fournie´, and L. Battistini. 2004. Fc␥RIII discriminates between 2 phenotype and was not found on naive CD8 T cells. Further- subsets of V␥9V␦2 effector cells with different responses and activation path- more, CD16 expression in this model was dependent on either ways. Blood 104: 1801–1807. 16. Gonzalez, V. D., N. K. Bjo¨rkstro¨m, K. J. Malmberg, M. Moll, C. Kuylenstierna, infection with Listeria monocytogenes or bystander activation J. Michae¨lsson, H. G. Ljunggren, and J. K. Sandberg. 2008. Application of nine- http://www.jimmunol.org/ by IL-2 or IL-15 but independent of TCR triggering. CD16 color flow cytometry for detailed studies of the phenotypic complexity and func- ␥ ϩ tional heterogeneity of human lymphocyte subsets. J. Immunol. Methods 330: signaled through FcR and the murine CD16 CD8 T cells 64–74. responded with ADCC as well as effector cytokine production 17. Kaech, S. M., and E. J. Wherry. 2007. Heterogeneity and cell-fate decisions in ϩ (41), a functional response much similar to that here described effector and memory CD8 T cell differentiation during viral infection. Immunity ϩ 27: 393–405. for CD16 CD8 human T cells. 18. Racanelli, V., M. A. Frassanito, P. Leone, C. Brunetti, S. Ruggieri, and Related to the present studies, Clemenceau et al. (42) have re- F. Dammacco. 2007. Bone marrow of persistently hepatitis C virus-infected in- ϩ cently considered the possibility of arming large populations of T dividuals accumulates memory CD8 T cells specific for current and historical viral antigens: a study in patients with benign hematological disorders. J. Immu- cells with CD16. They have demonstrated that human T cells can nol. 179: 5387–5398. be modified by CD16 gene transfer to gain the ability to perform 19. Betts, M. R., M. C. Nason, S. M. West, S. C. De Rosa, S. A. Migueles, by guest on September 27, 2021 ADCC. After transduction with a CD16/␥ gene, monoclonal and J. Abraham, M. M. Lederman, J. M. Benito, P. A. Goepfert, M. Connors, et al. ϩ ϩ 2006. HIV nonprogressors preferentially maintain highly functional HIV-specific polyclonal CD4 and CD8 T cells displayed stable expression of CD8ϩ T cells. Blood. 107: 4781–4789. the CD16 receptor at the cell surface. When tested against tumor 20. Darrah, P. A., D. T. Patel, P. M. De Luca, R. W. Lindsay, D. F. Davey, targets coated with the humanized anti-CD20 mAb rituximab, the B. J. Flynn, S. T. Hoff, P. Andersen, S. G. Reed, S. L. Morris, et al. 2007. Multifunctional TH1 cells define a correlate of vaccine-mediated protection newly expressed CD16 receptor enabled the T cells to kill the against Leishmania major. Nat. Med. 13: 843–850. tumors through ADCC (42). 21. Brenchley, J. M., N. J. Karandikar, M. R. Betts, D. R. Ambrozak, B. J. Hill, L. E. Crotty, J. P. Casazza, J. Kuruppu, S. A. Migueles, M. Connors, et al. 2003. The expression of CD16 on T cells has been studied most ex- Expression of CD57 defines replicative senescence and antigen-induced apoptotic tensively within the context of TCR␥␦ T cells in humans (11, death of CD8ϩ T cells. Blood 101: 2711–2720. 22. McMahon, C. W., and D. H. Raulet. 2001. Expression and function of NK cell 13–15). Few studies, if any, have attempted a more thorough anal- ϩ ϩ receptors in CD8 T cells. Curr. Opin. Immunol. 13: 465–470. ysis of CD16 TCR␣␤ CD8 T cells in humans. The findings pre- 23. Shin, H., and E. J. Wherry. 2007. CD8 T cell dysfunction during chronic viral sented here are to our knowledge the first thorough study evalu- infection. Curr. Opin. Immunol. 19: 408–415. ating the phenotype and function of CD16ϩ human TCR␣␤ CD8 24. Lanier, L. L. 2005. NK cell recognition. Annu. Rev. Immunol. 23: 225–274. 25. Bryceson, Y. T., M. E. March, H. G. Ljunggren, and E. O. Long. 2006. Activa- T cells. We propose that settings of chronic immune activation, as tion, coactivation, and costimulation of resting human natural killer cells. Immu- seen in chronic human viral infections, might lead to the differen- nol. Rev. 214: 73–91. 26. Lanier, L. L., G. Yu, and J. H. Phillips. 1989. Co-association of CD3␨ with a tiation of terminal effector CD8 T cells beyond defined stages, receptor (CD16) for IgG Fc on human natural killer cells. Nature 342: 803–805. acquiring NK cell-like properties. 27. Wirthmueller, U., T. Kurosaki, M. S. Murakami, and J. V. Ravetch. 1992. Signal transduction by Fc␥RIII (CD16) is mediated through the ␥ chain. J. Exp. Med. 175: 1381–1390. Disclosures 28. Thomson, C. W., W. A. Teft, W. Chen, B. P. Lee, J. Madrenas, and L. Zhang. The authors have no financial conflict of interest. 2006. FcR␥ presence in TCR complex of double-negative T cells is critical for their regulatory function. J. Immunol. 177: 2250–2257. 29. Ugolini, S., and E. Vivier. 2000. Regulation of T cell function by NK cell re- References ceptors for classical MHC class I molecules. Curr. Opin. Immunol. 12: 295–300. 1. Shepard, C. W., L. Finelli, and M. J. Alter. 2005. Global epidemiology of hep- 30. Kambayashi, T., E. Assarsson, J. Michae¨lsson, P. Berglund, A. D. Diehl, atitis C virus infection. Lancet Infect. Dis. 5: 558–567. B. J. Chambers, and H. G. Ljunggren. 2000. Emergence of CD8ϩ T cells ex- 2. Thimme, R., D. Oldach, K. M. Chang, C. Steiger, S. C. Ray, and F. V. Chisari. pressing NK cell receptors in influenza A virus-infected mice. J. Immunol. 165: 2001. Determinants of viral clearance and persistence during acute hepatitis C 4964–4969. virus infection. J. Exp. Med. 194: 1395–1406. 31. Assarsson, E., T. Kambayashi, J. K. Sandberg, S. Hong, M. Taniguchi, 3. Cooper, S., A. L. Erickson, E. J. Adams, J. Kansopon, A. J. Weiner, D. Y. Chien, L. Van Kaer, H. G. Ljunggren, and B. J. Chambers. 2000. CD8ϩ T cells rapidly M. Houghton, P. Parham, and C. M. Walker. 1999. Analysis of a successful acquire NK1.1 and NK cell-associated molecules upon stimulation in vitro and in immune response against hepatitis C virus. Immunity 10: 439–449. vivo. J. Immunol. 165: 3673–3679. 4. Lechner, F., D. K. Wong, P. R. Dunbar, R. Chapman, R. T. Chung, 32. Northfield, J. W., V. Kasprowicz, M. Lucas, N. Kersting, B. Bengsh, A. Kim, P. Dohrenwend, G. Robbins, R. Phillips, P. Klenerman, and B. D. Walker. 2000. R. E. Phillips, B. D. Walker, R. Thimme, G. Lauer, and P. Klenerman. 2008. Analysis of successful immune responses in persons infected with hepatitis C CD161 expression on hepatitis C virus-specific CD8ϩ T cells suggests a distinct virus. J. Exp. Med. 191: 1499–1512. pathway of T cell differentiation. Hepatology 47: 396–406. 4228 CD16ϩ CD8 T CELLS IN CHRONIC HCV INFECTION

33. Laouar, A., M. Manocha, M. Wan, H. Yagita, R. A. van Lier, and N. Manjunath. 38. Aandahl, E. M., J. K. Sandberg, K. P. Beckerman, K. Taske´n, W. J. Moretto, and 2007. Cutting edge: distinct NK receptor profiles are imprinted on CD8 T cells in D. F. Nixon. 2003. CD7 is a differentiation marker that identifies multiple CD8 the mucosa and periphery during the same antigen challenge: role of tissue- T cell effector subsets. J. Immunol. 170: 2349–2355. specific factors. J. Immunol. 178: 652–656. 39. Aandahl, E. M., M. F. Quigley, W. J. Moretto, M. Moll, V. D. Gonzalez, A. So¨nnerborg, S. Lindba¨ck, F. M. Hecht, S. G. Deeks, M. G. Rosenberg, et al. 34. Ugolini, S., C. Arpin, N. Anfossi, T. Walzer, A. Cambiaggi, R. Fo¨rster, M. Lipp, 2004. Expansion of CD7low and CD7negative CD8 T-cell effector subsets in HIV-1 R. E. Toes, C. J. Melief, J. Marvel, and E. Vivier. 2001. Involvement of inhibitory ϩ infection: correlation with antigenic load and reversion by antiretroviral treat- NKRs in the survival of a subset of memory-phenotype CD8 T cells. Nat. ment. Blood 104: 3672–3678. Immunol. 2: 430–435. 40. Appay, V., P. R. Dunbar, M. Callan, P. Klenerman, G. M. Gillespie, L. Papagno, 35. Meresse, B., S. A. Curran, C. Ciszewski, G. Orbelyan, M. Setty, G. Bhagat, G. S. Ogg, A. King, F. Lechner, C. A. Spina, et al. 2002. Memory CD8ϩ T cells L. Lee, M. Tretiakova, C. Semrad, E. Kistner, et al. 2006. Reprogramming of vary in differentiation phenotype in different persistent virus infections. Nat. Med. CTLs into natural killer-like cells in celiac disease. J. Exp. Med. 203: 1343–1355. 8: 379–385. 41. Dhanji, S., K. Tse, and H. S. Teh. 2005. The low affinity for IgG 36. Meresse, B., Z. Chen, C. Ciszewski, M. Tretiakova, G. Bhagat, T. N. Krausz, functions as an effective cytolytic receptor for self-specific CD8 T cells. J. Im- D. H. Raulet, L. L. Lanier, V. Groh, T. Spies, et al. 2004. Coordinated induction munol. 174: 1253–1258. by IL15 of a TCR-independent NKG2D signaling pathway converts CTL into 42. Cle´menceau, B., N. Congy-Jolivet, G. Gallot, R. Vivien, J. Gaschet, G. Thibault, lymphokine-activated killer cells in celiac disease. Immunity 21: 357–366. and H. Vie´. 2006. Antibody-dependent cellular cytotoxicity (ADCC) is mediated ϩ 37. Harty, J. T., and V. P. Badovinac. 2008. Shaping and reshaping CD8 T-cell by genetically modified antigen-specific human T lymphocytes. Blood 107: memory. Nat. Rev. Immunol. 8: 107–119. 4669–4677. Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021