δγ + T Cells Involvement in Viral Immune Control of Chronic Human Herpesvirus 8 Infection

This information is current as Serge Barcy, Stephen C. De Rosa, Jeffrey Vieira, Kurt Diem, of October 1, 2021. Minako Ikoma, Corey Casper and Lawrence Corey J Immunol 2008; 180:3417-3425; ; doi: 10.4049/jimmunol.180.5.3417 http://www.jimmunol.org/content/180/5/3417 Downloaded from

References This article cites 53 articles, 22 of which you can access for free at: http://www.jimmunol.org/content/180/5/3417.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on October 1, 2021

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

؉ T Cells Involvement in Viral Immune Control of Chronic␦␥ Human Herpesvirus 8 Infection1

Serge Barcy,2* Stephen C. De Rosa,* Jeffrey Vieira,* Kurt Diem,* Minako Ikoma,* Corey Casper,† and Lawrence Corey*†‡

Little is known about what effector populations are associated with the control of human herpesvirus 8 (HHV-8) infection in vivo. We compared T lymphocyte subsets among HIV؊HHV-8؉ and HIV؊HHV-8؊ infected human individuals. ␣␤؉ T cells from HHV-8-infected individuals displayed a significantly higher percentage of differentiated effector cells among both CD4؉ and CD8؉ T cell subsets. HHV-8 infection was associated with significant expansion of ␥␦؉ V␦1 T cells expressing a differentiated effector cell phenotype in peripheral blood. In vitro stimulation of PBMC from HHV-8-infected individuals with either infectious viral particles or different HHV-8 viral proteins resulted in ␥␦ V␦1 T cell activation. In addition, ␥␦ V␦1 T cells displayed a strong reactivity against HHV-8-infected cell lines and prevented the release of infectious viral particles following the induction of lyric Downloaded from replication. These data indicate that ␥␦ T cells play a role in both innate and adaptive T cell responses against HHV-8 in immunocompetent individuals. The Journal of Immunology, 2008, 180: 3417–3425.

uman herpesvirus 8 (HHV-8),3 also known as Kaposi’s cells have been described. One, expressing the TCR variable sarcoma-associated herpesvirus (KSHV) is the etiolog- region V␦2, represents the majority of peripheral blood ␥␦ lym- ical agent of several infectious diseases including phocytes. ␥␦ T cells of this subset play a role in the defense

H http://www.jimmunol.org/ Kaposi’s sarcoma, primary effusion lymphoma (PEL), and multi- against intracellular pathogens and hematological malignancies centric Castleman’s disease (1). Like other herpesviruses, HHV-8 (14, 15). By contrast, the second subset of V␦1 T cells is res- is able to establish a predominantly latent, life-long infection in its ident mainly in the oral and intestinal epithelia, where these host. The increased incidence of these diseases in immunocom- cells might provide a first line of defense against viral infections promised individuals suggests that host immune control may be or malignancies (16). ␥␦ T cells have been implicated in anti- essential in preventing HHV-8-associated diseases (2). Several viral immune responses on the basis of their selective expansion studies have demonstrated the anti-HHV-8 specificity of TCR in the peripheral blood of patients infected with HIV, CMV, ␣␤ϩCD8ϩ CTL (3–10). HHV-8 specific CTL responses to latent EBV, and HSV (17–21).

and lytic viral proteins have been detected during primary HHV-8 In the present study, we examined the relative frequency of by guest on October 1, 2021 infection (11). The contribution of these T cells in the resolution of differentiated effector T cells in peripheral blood samples from HHV-8 infection has yet to be documented, because several im- HHV-8-infected and uninfected individuals. We report that mune evasion mechanisms targeting infected cell recognition by HHV-8-infected immunocompetent individuals have a signifi- CTL have been described including blocking Ag presentation, in- cant expansion of T lymphocytes expressing the ␥␦ V␦1 TCR hibiting costimulatory molecule surface expression, and deregu- and that these cells exhibit anti-HHV-8 specificity and antiviral lating T cell activation signaling (12). activity. The large majority of T cells present in the peripheral blood of healthy individuals express ␣␤ TCR, with T cells expressing Materials and Methods the complementary ␥␦ TCR typically accounting for Ͻ5% of Study population ␥␦ the circulating T cells (13). Two main subsets of human T Participants for this study were recruited from the Seattle, WA area for participation in studies of the natural shedding of HHV-8 infection (22). All HHV-8 positive subjects we studied were men who have with men, *Department of Laboratory Medicine and †Department of Medicine, University of were HIV-1-negative as shown by ELISA, had Abs to HHV-8 in a com- Washington and ‡Clinical Research Division, Cancer Research Cen- bined whole ELISA plus a confirmatory immunofluorescence assay ter, Seattle WA 98109 (IFA), and were also observed to shed HHV-8 DNA in saliva on Ն2 days Received for publication June 18, 2007. Accepted for publication December 29, 2007. of observation. HHV-8 seronegative controls were enrolled from cohorts followed with known low rates of HHV-8 and who demonstrated persistent The costs of publication of this article were defrayed in part by the payment of page seronegativity to HHV-8 over time. Both HHV-8 seropositive and sero- charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. negative populations were age matched. The study protocol was approved by the University of Washington Institutional Review Board, Seattle, WA. 1 This work was supported by National Institutes of Health Grants AI30731, R37 AI42528, RO1 DE016809, RO1 DE14149, and K23 AI054162-04. This research was Flow cytometry also supported by the University of Washington Center for AIDS Research, a National Institutes of Health-funded program (P30 AI 27757). PBMC were isolated from 50 ml of heparinized blood by Ficoll-Hypaque 2 Address correspondence and reprint requests to Dr. Serge Barcy, Department of centrifugation and cryopreserved. Cells were stained with fluorescently la- Laboratory Medicine, University of Washington, 1959 Northeast Pacific Street, P.O. beled Abs as described previously. The staining combinations used were: Box 358070, Seattle, WA 98109. E-mail address: [email protected] CD3-Qdot 605, CD4-Alexa Fluor 405 (Caltag Laboratories/Invitrogen Life 3 Technologies), CD8-allophycocyanin-Alexa Fluor 750 (Caltag Laborato- Abbreviations used in this paper: HHV-8, human herpesvirus 8; HVS, Herpesvirus ␦ ␦ saimiri; KSHV, Kaposi’s sarcoma associated herpesvirus; , open reading frame; ries/Invitrogen Life Technologies), CD27-Qdot 655, V 1-FITC ( TCS1 PEL, primary effusion lymphoma; rh, recombinant human. clone; Pierce/Endogen), V␦2-PE (BD/Pharmingen), CD45R0-Texas Red-PE (Beckman/Coulter), CD11a-allophycocyanin (BD/Pharmingen), Copyright © 2008 by The American Association of Immunologists, Inc. 0022-1767/08/$2.00 and CD57-Alexa Fluors680. All Abs (mAb) were purchased labeled except www.jimmunol.org 3418 ␥␦ T CELL STIMULATION FOLLOWING HHV-8 INFECTION

CD3, CD27, and CD57, which were purchased purified from BD Pharmingen culture at a final concentration of 1 ␮g/ml. In some experiments cells from and conjugated in the laboratory using standard protocols (www.drmr.com/ different PEL lines were mixed with ␥␦ T cells (2 ϫ 106 cells/ml) at abcon). Qdot and Alexa dyes were purchased from Invitrogen Life Technol- various dilutions. When indicated, purified OKT3 or isotype control were ogies. Dead cells were excluded following staining with propidium iodide added at a final concentration of 5 ␮g/ml. For purified viral protein stim- (Sigma-Aldrich). Samples were analyzed on a LSR II flow cytometer (BD ulation, HVS-immortalized ␥␦ T cells (5 ϫ 105 cells/ml) were mixed with Biosciences). Data analysis, including postacquisition compensation, was per- irradiated (5,000 rad) heterologous PBMC (105 cells/ml) in the presence of formed using FlowJo software (Tree Star). the purified HHV-8 recombinant proteins gB, K8.1, ORF65, and ORF73 (0.5 ␮g/ml). After 48 h, culture supernatants from triplicate wells were Statistical analysis pooled and tested for the presence of cytokines. Measurements of hu- ␥ ␣ JMP software produced by the SAS Institute was used for all statistical man IFN- and TNF- were analyzed using a sandwich ELISA. Sam- ples were tested in duplicate. The coefficient of variation was always analyses. Significant values for comparisons between groups were deter- Ͻ mined by the nonparametric Wilcoxon’s rank sum analysis. Data are shown 10%. Matched pair mAbs were purchased from Endogen. The lowest detection limit for the IFN-␥ assay is 0.3 pg/ml and 1 pg/ml for the as box plots in which the ends of the box are the 25th and 75th percentiles, ␣ and the line across the middle indicates the median. The lines above and TNF- assay. ϫ below the box extend to the outermost data that falls within 1.5 inter- ␥␦ V␦1 T cell repertoire diversity quartile range. Highly purified ␥␦ V␦1 T cells from different cell lines were isolated using Cell lines flow cytometry. Following an automated process, the positive cells were distributed in 24 wells at 10 cells/well into a 96-well PCR plate preloaded All cell lines were maintained in complete medium consisting of DMEM ␮ ϫ (Invitrogen Life Technologies) supplemented with antibiotics (100 IU/ml with 5 l of lysis buffer (1 recombinant Thermus thermophilus buffer penicillin and 0.1 mg/ml streptomycin) and 10% heat-inactivated bovine (Applied Biosystems), 0.005% Triton X-100, 0.5% 2-ME, and 3.3% pro- teinase K). Cell lysis was achieved by incubating the plate at 55°C for 15 serum (Gemini Bio-Products). The cell lines BCBL-1, JSC-1, Raji, Vero, o

min followed by a 5-min incubation at 95 C. Samples were then subjected Downloaded from and Ramos were obtained from the American Type Culture Collection. The ␥␦ ␦ lymphoblastoid cell line TM was obtained from Dr. S. R. Ridell (Fred to a nested PCR using V 1-specific primers as previously described Hutchinson Cancer Research Center, Seattle, WA). The Bjab cell line was (28). Following the first PCR (recombinant Thermus thermophilus DNA obtained from M. Lagunoff (University of Washington, Seattle, WA). C1R, polymerase; profile: 60°C for 30 min, 95°C for 5 min, and then at 95°C for 20 s/52°C for 20 s/60°C for 1 min for 45 cycles), 10 ␮l of the reaction mix a human lymphoblastoid cell line, was provided by V. Groh (Fred Hutchin- ␮ son Cancer Research Center). The rKSHV.219 infected JSC-1 cell line was transferred into 90 l of fresh PCR buffer before running the second PCR (AmpliTaq DNA polymerase; profile: 95°C for 2 min and then at (JSC/rKSHV.219) has been described elsewhere (23). o Bjab infection by rKSHV.219 was achieved using cell to cell virus 95°C for 30 s/54° for 30 s/72 C for 30 s for 35 cycles and then at 72°C for http://www.jimmunol.org/ spread (24). Briefly, rKSHV.219 lytic replication was induced in latently 7 min). Positive amplifications were visualized on a 2% agarose gel and infected Vero cells (105 cells/well in a 6-well plate) using sodium butyrate cloned into the PCR2.1 TOPO vector (Invitrogen Life Technologies). Pos- (0.1 ␮M, Sigma) and a recombinant baculovirus expressing KSHV ORF50 itive colonies were expanded overnight in 2 ml of Luria-Bertani medium. as described elsewhere (23). After overnight incubation, culture medium Plasmid DNA extraction was performed using a Qiagen kit according to was removed and cells were washed with PBS. Induced Vero cells were the manufacturer’s instructions. M13 forward and reverse primers were 6 used for the sequencing reactions. Nucleotide sequences were assigned to cocultured with Bjab (10 cells/well) for 4 days, after which puromycin ␦ selection (0.5 mg/ml; Calbiochem) began. After 2 wk KSHV-infected Bjab TCR- gene segments based on identities of V and J germline sequences were tested for GFP expression and Vero cells’ residual presence by flow published in GenBank using Blastn. cytometry using anti-CD19 Ab (BD Pharmingen). Expression and purification of HHV-8 viral proteins ␥␦ T cell lines The HHV-8-GST fusion proteins gB, ORF73, ORF65, and K8.1A were a by guest on October 1, 2021 PBMC (2 ϫ 106 cells/ml, 5 ml/well in 6-well plate) were stimulated with gift from Dr. B. Chandran (University of Kansas Medical Center, Kansas infectious viral particles (5 ϫ 104 infectious U/ml) in the presence of City, KS) and were expressed from recombinant baculoviruses (29). Pro- rhIL-2 (20 U/ml; Hemagen Diagnostics) and rhIL-7 (100 U/ml; R&D Sys- tein expression was achieved by infecting SF9 cells with recombinant bac- tems). After 2 wk, ␥␦ T cells were purified using magnetic cell separation ulovirus. Supernatants were harvested after 5 days. Recombinant proteins according to manufacturer’s protocol (Miltenyi Biotec). The derived cell were isolated by ammonium sulfate precipitation followed by dialysis lines were further expanded under polyclonal expansion using PHA as against PBS. The proteins were further purified using a metal affinity resin previously described (25). Briefly, purified ␥␦ T cells were expanded using (BD Biosciences) according to manufacturer protocol. Proteins concentra- PHA stimulation (0.8 ␮g/ml; Murex Diagnostics) in the presence of irra- tion was assessed by a Bradford protein assay (Pierce). ϫ 6 diated heterologous feeders (2 10 cell/ml, 5,000 rad). Virus titration ␥␦ T cells immortalization with Herpesvirus saimiri (HVS) Cells from the JSC/rKSHV.219 cell line were incubated with pooled ␦ HVS (strain C4488) was kindly provided by R. Desrosiers (New England culture supernatants from HVS-immortalized V 1 T cells previously stim- Regional Primate Research Center, MA). Owl monkey kidney (OMK) cells ulated for 48 h with JSC/rKSHV.219 as described above. Cells were cul- ␥ were used for propagation of HVS. ␥␦ T cell immortalization by HVS tured for 24 h in the presence of a blocking Ab specific for the IFN- ␣ ␮ infection was performed as previously described (26). Briefly, ␥␦ T cells receptor -chain (10 g/ml; BD Pharmingen) or an isotype control before ␮ were simultaneously stimulated with PHA (0.8 ␮g/ml; Murex Diagnostics) the induction of lytic replication with sodium butyrate (0.1 M). Cell-free and irradiated heterologous PBMC (5,000 rad) with recombinant human supernatants were recovered 48 h later. Virus titrations were performed as (rh)IL-2 (20 U/ml) and rhIL-7 (100 U/ml) addition on the following day. described elsewhere (23). Briefly, virus yield was determined using a suscep- Sample wells were infected over a 7-day period with 10% (v/v) prepared tible cell line by evaluating the number of GFP-expressing cells. Because this HVS virus supernatant. Immortalization was assessed based on the ␥␦ T is not a plaque assay, viral yields were reported as infectious units per milliliter cell line’s ability to proliferate extensively in vitro independently of any instead of PFU per milliliter. Infectious units of rKSHV.219 were determined external stimulation. T cell clones established from an immortalized ␥␦ T by titering virus present in cell-free supernatants on 293 cells. The number of cell line were generated as previously described (27). ␥␦ T cells from an GFP-positive cells was determined 2 days postinfection by visually counting uninfected individual were isolated by FACS using a V␦1-specific Ab cells using an inverted Nikon fluorescence microscope. No GFP-positive cells (Pierce/Endogen). Cells were then immortalized by HVS as described were present on the same day as inoculation. Treatment of the 293 cell line ␥ above. with recombinant human IFN- did not affect its susceptibility to HHV-8 infection (data not shown). In vitro stimulation and cytokine detection ϫ 6 Results PBMC (2 10 cells/ml) were stimulated in the presence of the purified ␣␤ϩ ␥␦ϩ HHV-8 recombinant proteins gB, K8.1, open reading frame (ORF)65, and Phenotypic changes among and T lymphocyte subsets ORF73 (0.5 ␮g/ml) or infectious viral particles (105 infectious U/ml) in in HHV-8-infected individuals and uninfected controls complete medium supplemented with rhIL-2 and rhIL-7. ␥␦ T cells (5 ϫ 105 cells/ml) were cocultured in a 96-well round-bottom plate with differ- We assessed T cell phenotypes by using memory and differentia- ent cell lines (5 ϫ 104 cells/ml). When specified, commercially available tion markers to determine whether there were any significant viral lysates from HHV-8 or CMV (Applied Biosystems) were added to the differences between HHV-8-infected and seronegative men. As The Journal of Immunology 3419

Table I. Phenotypic characteristic of ␣␤ and ␥␦ T lymphocytes in HHV-8 seropositive versus HHV-8 seronegative individualsa

HHV-8ϩ (n ϭ 7) HHV-8Ϫ (n ϭ 5) p Valueb

Age 45c 42 Gender Male Male HIV status Negative Negative HHV-8 immunofluorescence assay Positive Negative HHV-8 PCR (saliva) Positive NAd Percentage CD4 among CD3 cells 57.9 63.8 0.1 Percentage CD8 among CD3 cells 37.7 31.3 0.1 Percentage V␦1 among CD3 cells 4.2 (1.8–6.5)e 0.5 (0.3–1.2) 0.01 Percentage V␦2 among CD3 cells 1.3 4.1 0.3 V␦1/V␦2 ratio 3.4 (0.8–4.8) 0.30 (0.08–0.6) 0.01

a All patients were clinically normal. b Determined by nonparametric Wilcoxon test. c Data are median. d NA, not available. e Interquantile range.

shown in Table I, no significant difference in the relative per- cells in both CD4ϩ (median 4.8/infected vs 0.9/uninfected) and Downloaded from centage of CD4ϩ or CD8ϩ T cell subpopulations was noted CD8ϩ (median 43/infected vs 13.7/uninfected) peripheral T cell between HHV-8-infected individuals and controls. However, populations from HHV-8-infected individuals. We have previ- we noted a significant increase in ␥␦ lymphocytes expressing ously shown that specific surface markers can be used to de- the V␦1 receptor chain and a relative decrease in ␥␦ lympho- termine an effector T cell population not only among ␣␤ but cytes expressing V␦2 receptor chain in PBMC from HHV-8- also for ␥␦ T cells (30, 33). Consequently, we also used CD57 infected individuals as compared with seronegative controls. as a marker to determine the relative percentage of effector http://www.jimmunol.org/ Although the decrease in the percentage of V␦2 T cells was not memory cells in both V␦1 and V␦2 ␥␦ T cell subpopulations. As statistically significant, the V␦1 T cell subset expansion and the seen in Fig. 1, among the HHV-8-infected individuals only the overall reduction of the V␦2 T cell subset led to a significant V␦1 T cell subset had a significant increase in the relative per- alteration in the peripheral V␦1/V␦2 T cell ratio in HHV-8- centage of CD57ϩ effector cells similar to what we observed for infected individuals (Table I). the ␣␤ϩ T cells (median 64.8/infected vs 24.3/uninfected). In Previous studies have shown that CD57 surface expression contrast, the frequency of effector cells among the V␦2 T cell on both CD4ϩ and CD8ϩ T lymphocyte subsets is characteristic subpopulation remained unchanged between HHV-8-infected of terminally differentiated cells in individuals with persistent individuals and seronegative controls. For three infected indi- viral infection (30–32). Using this specific marker, we deter- viduals, PBMC from sequential blood draws were available. by guest on October 1, 2021 mined the relative percentage of effector cells among ␣␤ϩ T The observed V␦1 T cell frequency always remained above the lymphocytes. As shown in Fig. 1, A and B, we observed a sig- V␦2 T cell frequency in all samples for a period of up to 28 mo nificant increase in the relative percentage of CD57ϩ effector (Fig. 1C).

FIGURE 1. ␣␤ϩ and ␥␦ϩ T cell phenotypes in HHV-8-infected indi- viduals and uninfected controls. A, PBMC (107 cells) were stained using a panel of Abs as described in Mate- rial and Methods. Cells were scatter- gated on lymphocytes and then gated on CD3 to identify T cells. T cells not expressing V␦1orV␦2 were then gated on CD4ϩ or CD8ϩ. Relative percentages of cells expressing CD57ϩ within each T cell subset for representative HHV-8 seropositive and seronegative individuals are shown. B, Statistical analyses of the relative percentage of CD57ϩ cells in each T cell subset for both popula- tions. C, Relative percentages of cells expressing V␦1 and V␦2 in sequential blood samples from a representative HHV-8-infected subject. 3420 ␥␦ T CELL STIMULATION FOLLOWING HHV-8 INFECTION

In vitro stimulation of PBMC from infected individuals with HHV-8 specifically induces ␥␦ T cell expansion Table II. ␥␦ T cell expansion following stimulation with HHV-8 viral particles and purified viral recombinant proteinsa We next investigated whether ␥␦ V␦1 T cells were able to respond to HHV-8 stimulation in vitro. PBMC from HHV-8-infected pa- ϩ ϩ ϩ ϩ Subjects CD4 CD8 V␦1 V␦2 tients or uninfected controls were stimulated in the presence of IL-2 and IL-7 with or without infectious viral particles. Infectious HHV-8 seropositive Serpos I viral particles were obtained from HHV-8-infected Vero cells in Vector 46.3 36.0 2.4 3.3 which lytic viral replication was artificially triggered as previously gB 36.1 47.1 2.3 4.3 described (23). As shown in Table II, stimulation with HHV-8 K8.1 34.6 47.9 2.0 2.2 infectious viral particles demonstrated no significant expansion ORF65 38.3 46.8 2.5 2.3 ϩ ϩ ϩ ϩ ϩ ϩ ORF73 38.2 43.2 2.2 2.6 among either ␣␤ CD4 CD8 or ␥␦ V␦1 V␦2 T cell subsets Infectious particles 46.3 37.6 10.2 2.4 in PBMC from uninfected individuals. In contrast, stimulation of Seropos II PBMC from all HHV-8-infected individuals resulted in a 3- to Vector 49.9 34.1 2.0 0.8 6-fold expansion of the ␥␦ V␦1 T cell population. gB 51.4 32.4 4.1 1.6 K8.1 50.3 30.6 5.9 1.4 We next determined whether the observed expansion of the ␥␦ ORF65 53.3 31.4 3.3 1.9 V␦1 T cell population could also be triggered in response to viral ORF73 43.1 40.8 3.2 1.6 protein stimulation. To do so, PBMC from seropositive (n ϭ 3) Infectious particles 50.6 30.3 5.9 2.4 and seronegative (n ϭ 3) individuals were cultured in the presence Seropos III Vector 54.0 34.2 1.2 1.0 of the purified HHV-8 GST-fusion-proteins gB, ORF65, ORF73, gB 55.2 33.1 2.5 1.1 Downloaded from and K8.1. As seen in Table II, PBMC stimulation with several K8.1 50.1 37.4 2.5 1.0 purified viral proteins (gB, K8.1, and ORF65) resulted in a 2- to ORF65 45.1 44.6 3.3 1.1 ␥␦ ␦ ORF73 47.5 41.1 3.0 2.2 3-fold expansion of the V 1 T cell population in two of three Infectious particles 58.2 30.9 7.9 0.8 HHV-8-infected individuals. In contrast, following similar stimu- Seropos IV lation no ␥␦ V␦1 expansion was observed in PBMC from sero- Medium 36.6 49.0 4.4 0.9 negative individuals. No consistent expansion in the ␣␤ϩ CD4ϩ or Infectious particles 14.9 68.0 17.9 6.3 ϩ ϩ http://www.jimmunol.org/ CD8 T cell populations were seen in the HHV-8 persons. Seropos V We did, however, note some expansion in the V␦2 T cell pop- Medium 39.0 53.1 3.4 0.5 Infectious particles 46.0 36.0 17.0 1.3 ulation in three of seven HHV-8 seropositive individuals (Seropos II, IV, and VI in Table II), an effect not observed ex vivo in PBMC Seropos VI Medium 49.9 31.4 7.3 0.8 from HHV-8 seropositive individuals. It is possible that this could Infectious particles 10.3 35.5 42.2 4.0 reflect a bystander activation effect from a previous exposure to Seropos VII other related herpesviruses, especially EBV, which are known to Medium 40.2 55.7 0.6 0.4 stimulate this particular ␥␦ T cell subset (34, 35). The serological Infectious particles 39.0 53.1 3.4 0.5 status of the individuals enrolled in our study for these was HHV-8 seronegative by guest on October 1, 2021 not available. Seroneg I Vector 33.5 24.4 3.0 33.6 gB 32.8 26.5 3.1 27.5 ␥␦ ␦ K8.1 28.2 28.9 4.1 29.0 V 1 T cells specifically recognize HHV-8-infected cell lines ORF65 26.1 30.3 3.3 27.0 Following in vitro stimulation with infectious viral particles ␥␦ T ORF73 31.2 27.4 3.1 27.5 Infectious particles 27.2 33.3 3.4 27.0 cells were purified and the derived cell lines were further expanded Seroneg II with PHA. Flow cytometry analysis revealed that the resulting cell Vector 66.3 17.1 1.7 0.9 lines were exclusively expressing the V␦1 TCR chain (data not gB 54.8 25.5 1.9 0.7 shown). We then mixed the expanded ␥␦ V␦1 T cells with two K8.1 55.3 26.8 2.3 1.4 ϩ ORF65 69.6 19.6 1.8 0.6 HHV-8 infected cell lines: 1) JSC-1, a PEL cell line coinfected ORF73 76.3 15.2 1.4 0.7 with EBV and HHV-8; and 2) BCBL-1, a PEL cell line infected Infectious particles 69.4 20.2 0.7 0.9 with just HHV-8. We used the Burkitt’s lymphoma Raji (HHV- Seroneg III Ϫ ϩ 8 EBV ) as the negative control in these experiments because Vector 72.6 10.2 1.3 7.5 there is no HHV-8 uninfected PEL line. Following coculture, large gB 66.2 14.6 1.7 9.7 ␥ ␥␦ ␦ K8.1 71.7 11.7 1.6 7.0 amounts of IFN- were detected for two V 1 cell lines when ORF65 72.9 10.8 1.7 6.7 stimulated with either BCBL-1 or JSC-1 (Fig. 2A). In addition, ORF73 74.7 10.3 1.4 5.8 significant levels of TNF-␣ secretion were also detected with one Infectious particles 74.4 9.7 1.0 6.0 of the ␥␦ V␦1 cell lines in response to both HHV-8-infected PEL Seroneg IV ␥ Medium 44.0 32.0 8.0 2.3 lines. In all experiments only minimal amounts of IFN- and Infectious particles 49.0 30.0 7.6 1.6 TNF-␣ were measured following stimulation with Raji. Similar ␦ Seroneg V stimulation of two V 1 T cell lines established from seronegative Medium 46.9 19.7 2.4 11.7 individuals did not result in any significant IFN-␥ secretion (Fig. Infectious particles 42.2 19.0 2.5 12.6 ␥␦ ␦ 2B). These data suggested that the V 1 T cell reactivity we a PBMC (2 ϫ 106 cells/ml) from HHV-8 seropositive (Seropos I–VII) and observed was directed against the HHV-8-infected tumor cell lines seronegative (Seroneg I–V) subjects were stimulated with the purified recombi- independently of EBV presence. nant viral proteins gB, K8.1, ORF65, and ORF73 (0.5 ␮g/ml) or infectious viral particles (105 infectious units/ml) in the presence of rhIL-2 and rhIL-7. PBMC Because the reactivity of the V␦1 cell lines against HHV-8- cultures in the presence of GST (vector) or cytokines alone (medium) were used infected PEL lines did not preclude a possible specific recognition as controls. After 2 wk, cells were recovered and analyzed by flow cytometry using a panel of Abs as described in Material and Methods. Cells were scatter- of nonviral Ags, we established a new HHV-8-infected cell line gated on lymphocytes. Relative percentages of positive cells for each marker 2 wk from a Burkitt’s lymphoma line called Bjab using a previously poststimulation are shown. described HHV-8 recombinant virus, rKSHV.219 (23). This The Journal of Immunology 3421

FIGURE 3. ␥␦ T cells stimulation with HHV-8-infected cell lines. A, ␥␦ Downloaded from T cell lines (5 ϫ 105 cells/ml) established from a HHV-8 seropositive and a seronegative individual were stimulated with cells from different HHV- 8-infected cell lines or uninfected controls at a responder/stimulator ratio of 10:1.Viral lysates from HHV-8 or CMV were added to a coculture of ␥␦ ␥␦ FIGURE 2. T cell stimulation with HHV-8-infected PEL lines. A, uninfected Bjab and ␥␦ T cells at a final concentration of 1 ␮g/ml. B, ␥␦ ϫ 6 T cell lines (2 10 cells/ml) established from two HHV-8 seropositive T cells (2 ϫ 106 cells/ml) from an HHV-8-infected individual were stim- individuals were stimulated with cells from different HHV-8-infected PEL ulated with JSC-1 cells (responder/stimulator ration of 20:1) in the pres- http://www.jimmunol.org/ lines (JSC-1 and BCBL-1) or uninfected control (Raji) at various dilutions. ence of an OKT3 Ab (5 ␮g/ml) or an isotype control. IFN-␥ secretion was ␥␦ ϫ 6 B, T cell lines (2 10 cells/ml) established from two seronegative measured after 48 h by ELISA in pooled supernatants from triplicate cul- individuals were cocultured with the JSC-1, BCBL-1, or Raji cell lines at tures. The coefficient of variation for cytokine titration was Ͻ10%. Data a responder/stimulator ratio of 25:1. Cytokines secretion was measured at are representative of three independent experiments. 48 h in pooled supernatants from triplicate cultures. Cytokines titers were determined by ELISA. The coefficient of variation for cytokine titration was Ͻ10%. Data are representative of three independent experiments. HHV-8-infected or uninfected individuals. We determined the CDR3 sequences expressed by ␥␦ V␦1 T cells following in vitro by guest on October 1, 2021 recombinant carried the puromycin gene resistance that allowed a stimulation with or without HHV-8 as described in Table II. To quick selection of a long-term infected cell population. It also pro- do so, we developed a new method to determine the sequence of vided an easy identification of latent or lytic replication in living the third hypervariable CDR (CDR3) of the V␦1 gene segment. cells thanks to the selective expression of the green (GFP) or red Our method combines the use of flow cytometry and highly fluorescent protein (RFP), respectively. Bjab cells were cocultured sensitive PCR. It allows us to gather CDR3 sequences from an with rKSHV.219-infected Vero cells in which lytic viral replica- extremely low number of T cells. The CDR3 repertoire ex- tion was artificially induced. Successfully infected cells expressed pressed by each V␦1 T cell population was analyzed based on GFP and could be expanded under drug selection. As shown in how many different sequences are obtained (diversity) and how Fig. 3A, substantial amounts of IFN-␥ were detected when V␦1T many times the same sequence is detected (complexity/oligo- cells were stimulated with HHV-8-infected Bjab/rKSHV.219 or clonality). As shown in Fig. 4, the ␥␦ TCR repertoire in the PEL lines. A similar level of IFN-␥ secretion was measured when unstimulated PBMC from an HHV-8-seronegative individual a commercially available viral lysate from HHV-8 but not from contained a restricted number of V␦1 clonotypes (2/10 distinct CMV was used to stimulate V␦1 T cells in the presence of unin- V␦1 sequences). Following HHV-8 stimulation, the V␦1 T cell fected Bjab cells. Uninfected cell lines including Bjab did not in- repertoire became more diverse (7/9 distinct V␦1 sequences) duce any significant IFN-␥ secretion. Stimulation of a V␦1 T cell and no clonal amplification was detected. In contrast, the ␥␦ line established from an HHV-8-seronegative individual did not V␦1 repertoire in unstimulated PBMC from an HHV-8-infected result in any significant IFN-␥ secretion. These data indicated that patient exhibited a large junctional diversity (12/19 distinct V␦1 the specificity of V␦1 T cell lines from seropositive subjects is sequences). The V␦1 transcripts were oligoclonal and a domi- directed against HHV-8 viral Ags. To assess whether the activation nant clonotype was detected. In vitro stimulation with HHV-8 of V␦1 T cells by HHV-8-infected cells was mediated through resulted in a restricted V␦1 repertoire (5/14 distinct V␦1 se- engagement of the ␥␦ TCR, cocultures of T cells with the JSC-1 quences) characterized by the clonal expansion of the same PEL line were performed in the presence of OKT3. As shown in dominant transcript detected in unstimulated PBMC. This par- Fig. 3B, the addition of OKT3 resulted in a strong inhibition of ticular clonotype was detected in 6 of 19 independent PCR in IFN-␥ secretion. These data indicate that the HHV-8-infected PEL unstimulated PBMC and in 9 of 14 independent PCR in PBMC line triggered the production of IFN-␥ by V␦1 T cells through a stimulated with HHV-8. The sequence of the dominant V␦1 mechanism involving TCR engagement. transcript was identical with the TCR sequence of a HHV-8- specific ␥␦ T cell clone isolated from the same individual. ␥␦ ␦ Analysis of the V 1 repertoire These data provided direct evidence supporting enrichment for To further examine whether V␦1 T cell expansion was Ag driven, HHV-8-reactive ␥␦ V␦1 T cells within PBMC from infected we compared the ␥␦ V␦1 T cell repertoire diversity in PBMC from patients. 3422 ␥␦ T CELL STIMULATION FOLLOWING HHV-8 INFECTION

FIGURE 4. Junctional diversity of the V␦1 T cell subpopulation. PBMC from a seropositive and a seronega- tive individual were stimulated with or without HHV-8 in the presence of rhIL-2 and rhIL-7. After 2 wk, ␥␦ V␦1 T cells were isolated and V␦1 TCR transcripts from each individual were cloned and sequenced as de- scribed in Materials and Methods. These sequences are available from GenBank under accession numbers EF656621–EF656644. Downloaded from http://www.jimmunol.org/

Characterization of HHV-8 specific response by ␥␦ effector T cells Identification of the Ags recognized by ␥␦ V␦1 T cells is critical for further understanding of their potential antiviral functions. This approach, however, requires a large number of effector cells and maintaining long-term T cell culture without losing their original by guest on October 1, 2021 functions. Therefore, we used HVS, an oncogenic tumor virus of New World monkeys, to immortalize different ␥␦ V␦1 T cell lines. Multiple lines of evidence have put forth the model that TCR ␥␦ cell Ag reactivity is dependent on conformationally intact Ags, similar to Ig Ag recognition (13, 36, 37). As such, we further investigated whether different HHV-8 viral proteins known to be relevant serological Ags were able to induce stimulation of HVS- transformed ␥␦ V␦1 T cells. Immortalized V␦1 T cells were cul- tured in the presence of the purified HHV-8-GST fusion proteins gB, ORF65, ORF73, and K8.1. Because ␥␦ T cell activation has been shown to depend on costimulatory molecules and to require TCR cross-linking, PBMC were also added to the culture (38, 39). As shown in Fig. 5D, viral protein stimulation of HVS-trans- formed ␥␦ V␦1 T cells from an HHV-8 seronegative individual induced a minimal amount of IFN-␥ secretion. In contrast, the ␥␦ V␦1 line established from three different HHV-8 seropositive in- dividuals produced significant amounts of IFN-␥ following stim- ulation with the different viral proteins (Fig. 5, A–C). FIGURE 5. ␥␦ T cells stimulation with HHV-8-purified viral proteins. As shown in Fig. 5 there was a broad range of IFN-␥ values Three HVS-immortalized ␥␦ V␦1 T cell lines derived from separate observed, likely reflecting the heterogeneity of the different im- HHV-8ϩ patients (A–C) and an immortalized cell line derived from an mortalized ␥␦ V␦1 T cell lines established from these HHV-8- HHV-8 seronegative person (D) were stimulated with different purified infected individuals. In Fig. 3A, different ␥␦ V␦1 T cell lines pro- recombinant viral proteins (gB, K8.1, ORF 65, and ORF73) or control duced a broad range of IFN-␥ and TNF-␣ secretions in response to empty GST vector (vector). The specificity of a T cell clone established from the ␥␦ V␦1 line shown in A was also determined by stimulation in the the same stimulation with HHV-8-infected cell lines. Similarly, ␥␦ ␥␦ ␦ presence of different purified viral proteins (E). HVS-immortalized T each immortalized V 1 T cell line is heterogeneous. The high- ϫ 5 ␥ ␥␦ ␦ cells (5 10 cells/ml) were cultured with irradiated (5,000 rad) heterol- est value for IFN- secretion is observed for the V 1 T cell line ogous PBMC (105 cells/ml) in the presence of different purified viral pro- shown in Fig. 5A. Analysis of the T cell repertoire diversity ex- teins (0.5 mg/ml) for 48 h. IFN-␥ secretion was measured by ELISA in pressed by the PBMC used to establish this particular cell line is pooled supernatants from triplicate cultures. The coefficient of variation for shown in Fig. 4. The V␦1 transcripts were oligoclonal and a dom- cytokine titration was Ͻ10%. The data are representative of five indepen- inant clonotype was detected. The sequence of the dominant V␦1 dent experiments with similar results. The Journal of Immunology 3423

Previous studies have shown that IFN-␥ treatment of HHV-8- infected cell lines reduced the amount of infectious virus released following the induction of lytic replication with 12-O-tetradeca- noylphorbol-13-acetate (40). Therefore, we investigated whether IFN-␥ secreted by different V␦1 T cell lines in the presence of JSC/rKSHV.219 could inhibit HHV-8 propagation in vitro. To avoid V␦1 T cell activation due to exposure to sodium butyrate, cells from the JSC/rKSHV.219 cell line were incubated with cul- ture supernatants from different V␦1 T cell lines previously stim- ulated with JSC/rKSHV.219. Cells were cultured in the presence of a blocking Ab specific for the IFN-␥ receptor ␣-chain or isotype control before the induction of lytic replication with sodium bu- tyrate. As shown in Fig. 6B, viral particle production by JSC/ rKSHV.219 in the presence of supernatants from stimulated ␥␦ V␦1 T cells was dramatically reduced. In contrast, the addition of a blocking Ab specific for the IFN-␥ receptor restored the produc- tion of infectious viral particles. FIGURE 6. ␥␦ T cells antiviral activity. A, HVS-immortalized ␥␦ T These data demonstrate that IFN-␥ production by ␥␦ V␦1T (1.5 ϫ 106/ml) cells from a HHV-8-infected individual were cocultured in cells established from HHV-8-infected patients has an antiviral a 96-well plate with cells from JSC-1 or JSC/rKSHV.219 at various dilu- Downloaded from tions. After 48 h, pooled supernatants from triplicate cultures were recov- effect by preventing infectious viral particle release. ered. The presence of IFN-␥ was determined by ELISA. The coefficient of variation for cytokine titration was Ͻ10%. The data are representative of Discussion three independent experiments with similar results. B, Cells from the JSC/ Our analysis indicates that HHV-8 infection, while asymptomatic rKSHV.219 cell line were incubated with pooled culture supernatants from clinically in the HIVϪ person, is associated with persistent phe- three different HVS-immortalized V␦1 T cell lines previously stimulated notypic changes in different peripheral blood T cell subpopula- for 48 h with JSC/rKSHV.219. Cells were cultured for 24 h in the presence tions. In contrast with the relative percentage of ␣␤ϩCD3ϩ T cells http://www.jimmunol.org/ ␥ ␮ of a blocking Ab specific for the IFN- receptor (10 g/ml) or an isotype that remained stable in both CD4ϩ and CD8ϩ lymphocyte popu- control before induction of lytic replication with sodium butyrate (1 ␮g/ lations, we observed an 8-fold expansion among ␥␦ CD3ϩ T cells ml). Cell-free supernatants were recovered 48 h later. Infectious viral par- ␦ ␥␦ ␦ ticles titration was assessed by counting GFP-positive 293 cells 48 h expressing the V 1 receptor chain. A specific expansion of V 1 postinfection using an inverted fluorescent microscope. Data represent the T cells was also observed in vitro following PBMC stimulation ϩ ϩ mean value of three separate counts. with HHV-8. The absence of consistent expansion of ␣␤ CD4 or CD8ϩ T cell subpopulations is in agreement with a previous report showing low lymphoproliferative response to HHV-8 de- tectable in only 42% of a similar cohort of HIVϪHHV-8ϩ men by guest on October 1, 2021 transcript was identical with the TCR sequence of a HHV-8-spe- who have sex with men (6). cific ␥␦ T cell clone isolated from the same individual. The V␦1 We also demonstrated a 2.5- to 3-fold increase in the relative junctional sequence of the TCR expressed by this clone is shown frequency of cells expressing an effector phenotype among the in Fig. 4. As shown in Fig. 5E, a significant amount of IFN-␥ was ␣␤ϩCD4ϩ, CD8ϩ, and ␥␦ V␦1 T cell subpopulations, suggesting detected following stimulation of this dominant T cell clone with immune activation. Significant increases in relative frequency of the ORF65 purified fusion protein. Hence, the presence of a dom- CD57ϩ T cells have been reported in situations of persistent viral inant clonotype is responsible for the high value for IFN-␥ secre- infection and chronic inflammation (41, 42). Recently, CD57 ex- tion observed for this particular immortalized V␦1 cell line. pression on CD4 and CD8 lymphocytes was also associated with replicative senescence and terminal differentiation (31, 32). It is ␥␦ T cells inhibit HHV-8 propagation in vitro commonly accepted that this is the result of chronic activation due We next examined the ability of ␥␦ T cells to inhibit HHV-8 prop- to persistent Ag exposure. agation in vitro. Analysis of in vitro HHV-8 interaction with hosts The observation of CD57 expression on a large fraction of cells and quantification of infection have been hampered by the ␣␤ϩCD4ϩ, CD8ϩ, and ␥␦ V␦1 T cells in asymptomatically in- absence of a lytic replication cycle and a reliable plaque assay. fected patients with HHV-8 is of interest. Mucosal shedding of Therefore, we have developed a cell line made of JSC-1 cells that virus has been difficult to discern in many people and our data coexpress the HHV-8 recombinant virus rKSHV.219. We have suggest that HHV-8 persists in a lytic state more frequently than previously shown that the fluorescent proteins expressed by currently appreciated. Moreover, the presence of an expanded pool rKSHV.219 can be used as indicators of viral entry and infection. of lymphocytes expressing CD57 in PBMC from HHV-8-infected As reported before, upon treatment with sodium butyrate the JSC/ individuals could lead to a lesser capacity to proliferate in vitro rKSHV.219 cell line was able to support viral replication with the following viral stimulation. Therefore, the extent and potential of release of infectious viral particles (23). The virus yield can easily the cellular immune response to HHV-8 in previous studies may be determined using a susceptible cell line by evaluating the num- have been underestimated. ber of GFP-expressing cells. The most novel observation made during our study was the se- We initially evaluated whether the JSC/rKSHV.219 line was lective long-lasting expansion of ␥␦ T cells specifically from the able to trigger ␥␦ V␦1 activation as well as the parental JSC-1 line. V␦1 subtype in peripheral blood samples from HHV-8-infected As shown in Fig. 6A, the amounts of IFN-␥ detected following individuals. To date, V␦1 T cell expansion has been described stimulation by JSC-1 or JSC/rKSHV.219 were similar. following viral infection in two instances: transplant patients un- Having established that both lines are as potent activator for ␥␦ dergoing active CMV infection or HIV infected patients (17, 18). V␦1 T cells, we then assessed whether immortalized V␦1 T cells A similar V␦1 T cell expansion has also been observed in patients affected JSC/rKSHV.219 viral yield in vitro. with lyme arthritis in response to Borrelia burgdorferi stimulation 3424 ␥␦ T CELL STIMULATION FOLLOWING HHV-8 INFECTION

(43). To our knowledge, HHV-8 is the only virus other than HIV Disclosures for which a ␥␦ V␦1 T cell expansion is observed and the first to be The authors have no financial conflict of interest. described in immunocompetent individuals. We observed a long- term persistence of this expansion suggesting sustained ␥␦ T cell References activation. 1. Schulz, T. F. 1998. Kaposi’s sarcoma-associated herpesvirus (human herpesvi- One common peculiarity of the viral infections leading to V␦1 rus-8). J. Gen. Virol. 79: 1573–1591. 2. Pellet, C., D. Kerob, A. Dupuy, M. V. Carmagnat, S. Mourah, M. P. Podgorniak, T cell expansion is the involvement of mucosae (44). Our previous C. Toledano, P. Morel, O. Verola, C. Dosquet, et al. 2006. Kaposi’s sarcoma- studies have shown that saliva is the major mucosal site of HHV-8 associated herpesvirus viremia is associated with the progression of classic and replication in humans, and our patients were selected as having endemic Kaposi’s sarcoma. J. Invest. Dermatol. 126: 621–627. ␦ 3. Wilkinson, J., A. Cope, J. Gill, D. Bourboulia, P. Hayes, N. Imami, T. Kubo, salivary HHV-8 infection. V 1 cells are mainly located in intes- A. Marcelin, V. Calvez, R. Weiss, et al. 2002. Identification of Kaposi’s sarcoma- tinal epithelia where they represent 70–90% of the ␥␦ T cells (45). associated herpesvirus (KSHV)-specific cytotoxic T-lymphocyte epitopes and Little is known about the replication of HHV-8 in the gastrointes- evaluation of reconstitution of KSHV-specific responses in human immunodefi- ciency virus type 1-Infected patients receiving highly active antiretroviral ther- tinal tract and whether this is a source of V␦1 (46). Our data also apy. J. Virol. 76: 2634–2640. indicate that there appear to be an in vivo Ag-driven expansion of 4. Wang, Q. J., X. L. Huang, G. Rappocciolo, F. J. Jenkins, W. H. Hildebrand, ␦ Z. Fan, E. K. Thomas, and C. R. Rinaldo, Jr. 2002. Identification of an HLA V 1 T cells during the course of HHV-8 infection. Consistent with A*0201-restricted CD8ϩ T-cell epitope for the glycoprotein B homolog of hu- this notion, we have demonstrated that V␦1 T cell lines were spe- man herpesvirus 8. Blood 99: 3360–3366. 5. Wang, Q. J., F. J. Jenkins, L. P. Jacobson, Y. X. Meng, P. E. Pellett, cifically activated in the presence of HHV-8-infected cell lines. ϩ ␥␦ ␦ L. A. Kingsley, K. G. Kousoulas, A. Baghian, and C. R. Rinaldo, Jr. 2000. CD8 Furthermore, our analysis of the V 1 T cell repertoire from cytotoxic T lymphocyte responses to lytic proteins of human herpes virus 8 in an infected individual provides the first evidence that the expanded human immunodeficiency virus type 1-infected and -uninfected individuals. J. In- peripheral ␥␦ V␦1 T cell population responsive to HHV-8 is re- fect. Dis. 182: 928–932. Downloaded from 6. Strickler, H. D., J. J. Goedert, F. R. Bethke, C. M. Trubey, T. R. O’Brien, flective of a selective, clonally restricted response to viral Ags. Our J. Palefsky, J. E. Whitman, D. Ablashi, S. Zeichner, and G. M. Shearer. 1999. observations support the model that the peripheral V␦1 T cell rep- Human herpesvirus 8 cellular immune responses in homosexual men. J. Infect. Dis. 180: 1682–1685. ertoire is shaped by positive selection in response to an antigenic 7. Stebbing, J., D. Bourboulia, M. Johnson, S. Henderson, I. Williams, N. Wilder, stimulation (47, 48). M. Tyrer, M. Youle, N. Imami, T. Kobu, et al. 2003. Kaposi’s sarcoma-associated ␥␦ T cell specificity against HHV-8 was further demonstrated herpesvirus cytotoxic T lymphocytes recognize and target Darwinian positively

selected autologous K1 epitopes. J. Virol. 77: 4306–4314. http://www.jimmunol.org/ by their specific recognition of purified viral proteins. Only few 8. Ribechini, E., C. Fortini, M. Marastoni, S. Traniello, S. Spisani, P. Monini, and examples of Ag-specific ␥␦ T cells have been reported. The pre- R. Gavioli. 2006. Identification of CD8ϩ T cell epitopes within lytic of viously described Ag specificities identified included nonclassical human herpes virus 8. J. Immunol. 176: 923–930. 9. Micheletti, F., P. Monini, C. Fortini, P. Rimessi, M. Bazzaro, M. Andreoni, class I MHC molecules, MHC-like molecules (CD1c, MICA, M. Giuliani, S. Traniello, B. Ensoli, and R. Gavioli. 2002. Identification of cy- MICB), heat shock proteins, monoalkyl phosphate (49), and bac- totoxic T lymphocyte epitopes of human herpesvirus 8. Immunology 106: ␥␦ ␦ 395–403. terial superantigens (38). A specific reactivity of V 1 T cells to 10. Woodberry, T., T. J. Suscovich, L. M. Henry, J. N. Martin, S. Dollard, P. G. primary leukemia blasts has also been reported (50). Specific rec- O’Connor, J. K. Davis, D. Osmond, T. H. Lee, D. H. Kedes, et al. 2005. Impact ognition of viral proteins by ␥␦ T cells is infrequent; it is of interest of Kaposi sarcoma-associated herpesvirus (KSHV) burden and HIV coinfection on the detection of T cell responses to KSHV ORF73 and ORF65 proteins. that the other described virus was another herpesvirus (HSV-1 gly- J. Infect. Dis. 192: 622–629. by guest on October 1, 2021 coprotein gI) (51). 11. Wang, Q. J., F. J. Jenkins, L. P. Jacobson, L. A. Kingsley, R. D. Day, Furthermore, we demonstrated that V␦1 T cells are able to pre- Z. W. Zhang, Y. X. Meng, P. E. Pellett, K. G. Kousoulas, A. Baghian, and C. R. Rinaldo, Jr. 2001. Primary human herpesvirus 8 infection generates a vent in vitro the release of infectious particles from HHV-8-in- broadly specific CD8ϩ T-cell response to viral lytic cycle proteins. Blood 97: fected cells, suggesting a protective antiviral role. Previous studies 2366–2373. ␥ 12. Rezaee, S. A., C. Cunningham, A. J. Davison, and D. J. Blackbourn. 2006. Ka- have shown that IFN- treatment of HHV-8-infected cell lines posi’s sarcoma-associated herpesvirus immune modulation: an overview. J. Gen. reduced the amount of infectious virus that resulted when HHV-8 Virol. 87: 1781–1804. was induced into the lytic cascade using 12-O-tetradecanoylphor- 13. Kronenberg, M. 1994. Antigens recognized by ␥␦T cells. Curr. Opin. Immunol. 6: 64–71. bol-13-acetate (40, 52, 53). This is in agreement with our data 14. Fisch, P., E. Meuer, D. Pende, S. Rothenfusser, O. Viale, S. Kock, S. Ferrone, demonstrating a decrease in infectious viral particle release due to D. Fradelizi, G. Klein, L. Moretta, et al. 1997. Control of B cell lymphoma ␥ IFN-␥ production by HHV-8 stimulated V␦1 T cells. recognition via natural killer inhibitory receptors implies a role for human V 9/ V␦2 T cells in tumor immunity. Eur. J. Immunol. 27: 3368–3379. In summary, our studies indicate a specific expansion of ␥␦ V␦1 15. Jouen-Beades, F., E. Paris, C. Dieulois, J. F. Lemeland, V. Barre-Dezelus, T cells in the peripheral blood of asymptomatic HHV-8-infected S. Marret, G. Humbert, J. Leroy, and F. Tron. 1997. In vivo and in vitro activation ␦ and expansion of ␥␦ T cells during Listeria monocytogenes infection in humans. individuals. Our results support the role of V 1 T cells specific for Infect. Immun. 65: 4267–4272. viral proteins in control of HHV-8 chronic infection. This situation 16. Falini, B., L. Flenghi, S. Pileri, P. Pelicci, M. Fagioli, M. F. Martelli, L. Moretta, ␥␦ and E. Ciccone. 1989. Distribution of T cells bearing different forms of the T cell represents a unique opportunity to better understand the T cell ␥␦ ␥␦ receptor in normal and pathological human tissues. J. Immunol. 143: physiology. Studies to determine whether alteration in T cell 2480–2488. formation is a factor in HHV-8-induced Kaposi’s sarcoma lesions 17. Rossol, R., J. M. Dobmeyer, T. S. Dobmeyer, S. A. Klein, S. Rossol, D. Wesch, ␦ ϩ ␥␦ in immunocompromised hosts are warranted. D. Hoelzer, D. Kabelitz, and E. B. Helm. 1998. Increase in V 1 T cells in the peripheral blood and bone marrow as a selective feature of HIV-1 but not other virus infections. Br. J. Haematol. 100: 728–734. 18. Dechanet, J., P. Merville, F. Berge, G. Bone-Mane, J. L. Taupin, P. Michel, Acknowledgments P. Joly, M. Bonneville, L. Potaux, and J. F. Moreau. 1999. Major expansion of ␥␦ T lymphocytes following infection in kidney allograft recip- We thank Jie Wang, Steve Kuntz, and Anna Wald (University of Wash- ients. J. Infect. Dis. 179: 1–8. ington, Seattle, WA) for specimen collection, Alan Fox for technical as- 19. Oyoshi, M. K., H. Nagata, N. Kimura, Y. Zhang, A. Demachi, T. Hara, sistance, Bala Chandran (Rosalind Franklin University of Medicine and H. Kanegane, Y. Matsuo, T. Yamaguchi, T. Morio, et al. 2003. Preferential ex- ␥ ␥ ␦ ␦ ␥␦ Science, Chicago, IL) for providing the different recombinant baculovirus, pansion of V 9-J P/V 2-J 3 T cells in nasal T-cell lymphoma and chronic active Epstein-Barr virus infection. Am. J. Pathol. 162: 1629–1638. Ronald Desrosiers (New England Regional Primate Research Center, 20. Verjans, G. M., R. W. Roest, A. van der Kooi, G. van Dijk, Southborough, MA) for kindly providing HVS, Michael Lagunoff (Uni- W. I. van der Meijden, and A. M. Osterhaus. 2004. Isopentenyl pyrophosphate- versity of Washington, Seattle, WA) for providing the Bjab cell line, reactive V␥9V␦ 2 T helper 1-like cells are the major ␥␦ T cell subset recovered Veronika Groh (Fred Hutchinson Cancer Research Center, Seattle WA) for from lesions of patients with genital herpes. J. Infect. Dis. 190: 489–493. 21. Bluestone, J. A., R. Khattri, R. Sciammas, and A. I. Sperling. 1995. TCR ␥␦ providing the C1R cell line, and Nancy LaCroix (Fred Hutchinson Cancer cells: a specialized T-cell subset in the immune system. Annu. Rev. Cell. Dev. Research Center, Seattle, WA) for help in preparing this manuscript. Biol. 11: 307–353. The Journal of Immunology 3425

22. Casper, C., E. Krantz, H. Taylor, J. Dalessio, D. Carrell, A. Wald, L. Corey, and 38. Kabelitz, D., A. Glatzel, and D. Wesch. 2000. recognition by human ␥␦ R. Ashley. 2002. Assessment of a combined testing strategy for detection of T lymphocytes. Int. Arch. Allergy Immunol. 122: 1–7. to human herpesvirus 8 (HHV-8) in persons with Kaposi’s sarcoma, 39. Sireci, G., E. Champagne, J. J. Fournie, F. Dieli, and A. Salerno. 1997. Patterns persons with asymptomatic HHV-8 infection, and persons at low risk for HHV-8 of phosphoantigen stimulation of human V␥9/V␦2 T cell clones include Th0 infection. J. Clin. Microbiol. 40: 3822–3825. cytokines. Hum. Immunol. 58: 70–82. 23. Vieira, J., and P. M. O’Hearn. 2004. Use of the red fluorescent protein as a marker 40. Pozharskaya, V. P., L. L. Weakland, and M. K. Offermann. 2004. Inhibition of of Kaposi’s sarcoma-associated herpesvirus lytic gene expression. 325: infectious human herpesvirus 8 production by ␥ interferon and ␣ interferon in 225–240. BCBL-1 cells. J. Gen. Virol. 85: 2779–2787. 24. Sakurada, S., H. Katano, T. Sata, H. Ohkuni, T. Watanabe, and S. Mori. 2001. 41. Ibegbu, C. C., Y. X. Xu, W. Harris, D. Maggio, J. D. Miller, and A. P. Kourtis. Effective human herpesvirus 8 infection of human umbilical vein endothelial cells 2005. Expression of killer cell lectin-like receptor G1 on antigen-specific human by cell-mediated transmission. J. Virol. 75: 7717–7722. CD8ϩ T lymphocytes during active, latent, and resolved infection and its relation 25. Barcy, S., M. L. Huang, L. Corey, and D. M. Koelle. 2005. Longitudinal analysis ϩ with CD57. J. Immunol. 174: 6088–6094. of virus-specific CD4 cell clonotypes in infected tissues and 42. Lynne, J. E., I. Schmid, J. L. Matud, K. Hirji, S. Buessow, D. M. Shlian, and blood. J. Infect. Dis. 191: 2012–2021. J. V. Giorgi. 1998. Major expansions of select CD8ϩ subsets in acute Epstein- 26. Fickenscher, H., C. Bokel, A. Knappe, B. Biesinger, E. Meinl, B. Fleischer, Barr virus infection: comparison with chronic human immunodeficiency virus B. Fleckenstein, and B. M. Broker. 1997. Functional phenotype of transformed disease. J. Infect. Dis. 177: 1083–1087. ␣␤ ␥␦ human and T cells determined by different subgroup C strains of herpes- 43. Rutkowski, S., D. H. Busch, and H. I. Huppertz. 1997. Lymphocyte proliferation virus Saimiri. J. Virol. 71: 2252–2263. assay in response to Borrelia burgdorferi in patients with Lyme arthritis: analysis 27. Pon, R. A., and M. S. Freedman. 2003. Study of Herpesvirus saimiri immortal- of lymphocyte subsets. Rheumatol. Int. 17: 151–158. ␥␦ ization of T cells derived from peripheral blood and CSF of multiple sclerosis 44. Dechanet, J., P. Merville, V. Pitard, X. Lafarge, and J. F. Moreau. 1999. Human patients. J. Neuroimmunol. 139: 119–132. ␥␦ T cells and viruses. Microbes Infect. 1: 213–217. 28. Chowers, Y., W. Holtmeier, J. Harwood, E. Morzycka-Wroblewska, and 45. Deusch, K., F. Luling, K. Reich, M. Classen, H. Wagner, and K. Pfeffer. 1991. M. F. Kagnoff. 1994. The V ␦ 1 T cell receptor repertoire in human small intestine A major fraction of human intraepithelial lymphocytes simultaneously expresses and colon. J. Exp. Med. 180: 183–190. the ␥␦ T cell receptor, the CD8 accessory molecule and preferentially uses the V 29. Zhu, L., R. Wang, A. Sweat, E. Goldstein, R. Horvat, and B. Chandran. 1999. ␦ 1 gene segment. Eur. J. Immunol. 21: 1053–1059. Comparison of human sera reactivities in immunoblots with recombinant human 46. Peacock, J. W., and K. L. Bost. 2000. Infection of intestinal epithelial cells and Downloaded from herpesvirus (HHV)-8 proteins associated with the latent (ORF73) and lytic (ORFs development of systemic disease following gastric instillation of murine ␥her- 65, K8.1A, and K8.1B) replicative cycles and in immunofluorescence assays with pesvirus-68. J. Gen. Virol. 81: 421–429. HHV-8-infected BCBL-1 cells. Virology 256: 381–392. 47. Jouen-Beades, F., F. Halary, L. Drouot, M. A. Peyrat, E. Paris, P. Joly, D. Gilbert, 30. De Rosa, S. C., D. K. Mitra, N. Watanabe, L. A. Herzenberg, and M. Roederer. ␥ ␦ 2001. V␦1 and V␦2 ␥␦ T cells express distinct surface markers and might be M. Bonneville, and F. Tron. 1999. Expansion of circulating V 9/V 1 T cells developmentally distinct lineages. J. Leukocyte Biol. 70: 518–526. in a patient with a syndrome of recurrent : evidence for an unusual antigen- 31. Palmer, B. E., N. Blyveis, A. P. Fontenot, and C. C. Wilson. 2005. Functional and driven process leading to selection of recurrent motifs within TCR junctional phenotypic characterization of CD57ϩCD4ϩ T cells and their association with loops of diverse lengths. Eur. J. Immunol. 29: 3338–3349. 48. Beldjord, K., C. Beldjord, E. Macintyre, P. Even, and F. Sigaux. 1993. Peripheral HIV-1-induced T cell dysfunction. J. Immunol. 175: 8415–8423. http://www.jimmunol.org/ ␦ ϩ ␦ 32. Brenchley, J. M., N. J. Karandikar, M. R. Betts, D. R. Ambrozak, B. J. Hill, selection of V 1 cells with restricted T cell receptor gene junctional reper- L. E. Crotty, J. P. Casazza, J. Kuruppu, S. A. Migueles, M. Connors, et al. 2003. toire in the peripheral blood of healthy donors. J. Exp. Med. 178: 121–127. Expression of CD57 defines replicative senescence and antigen-induced apoptotic 49. Tanaka, Y., S. Sano, E. Nieves, G. De Libero, D. Rosa, R. L. Modlin, death of CD8ϩ T cells. Blood 101: 2711–2720. M. B. Brenner, B. R. Bloom, and C. T. Morita. 1994. Nonpeptide ligands for ␥␦ 33. De Rosa, S. C., J. P. Andrus, S. P. Perfetto, J. J. Mantovani, L. A. Herzenberg, human T cells. Proc. Natl. Acad. Sci. USA 91: 8175–8179. and M. Roederer. 2004. Ontogeny of ␥␦T cells in humans. J. Immunol. 172: 50. Meeh, P. F., M. King, R. L. O’Brien, S. Muga, P. Buckhalts, R. Neuberg, and 1637–1645. L. S. Lamb, Jr. 2006. Characterization of the ␥␦ T cell response to acute leuke- 34. De Paoli, P., D. Gennari, P. Martelli, V. Cavarzerani, R. Comoretto, and mia. Cancer Immunol. Immunother. 55: 1072–1080. G. Santini. 1990. ␥␦T cell receptor-bearing lymphocytes during Epstein-Barr 51. Sciammas, R., R. M. Johnson, A. I. Sperling, W. Brady, P. S. Linsley, virus infection. J. Infect. Dis. 161: 1013–1016. P. G. Spear, F. W. Fitch, and J. A. Bluestone. 1994. Unique antigen recognition 35. Maccario, R., M. G. Revello, P. Comoli, D. Montagna, F. Locatelli, and by a herpesvirus-specific TCR-␥␦cell. J. Immunol. 152: 5392–5397.

G. Gerna. 1993. HLA-unrestricted killing of HSV-1-infected mononuclear cells: 52. Krug, L. T., V. P. Pozharskaya, Y. Yu, N. Inoue, and M. K. Offermann. 2004. by guest on October 1, 2021 involvement of either ␥␦ϩ or ␣␤ϩ human cytotoxic T lymphocytes. J. Immunol. Inhibition of infection and replication of human herpesvirus 8 in microvascular 150: 1437–1445. endothelial cells by ␣ interferon and phosphonoformic acid. J. Virol. 78: 36. Rock, E. P., P. R. Sibbald, M. M. Davis, and Y. H. Chien. 1994. CDR3 length in 8359–8371. antigen-specific immune receptors. J. Exp. Med. 179: 323–328. 53. Milligan, S., M. Robinson, E. O’Donnell, and D. J. Blackbourn. 2004. Inflam- 37. Chien, Y. H., R. Jores, and M. P. Crowley. 1996. Recognition by ␥␦ T cells. matory cytokines inhibit Kaposi’s sarcoma-associated herpesvirus lytic gene tran- Annu. Rev. Immunol. 14: 511–532. scription in in vitro-infected endothelial cells. J. Virol. 78: 2591–2596.