Cytolytic Mechanisms and Expression of Activation-Regulating Receptors on Effector-Type CD8+CD45RA+CD27− Human T Cells This information is current as of September 29, 2021. Paul A. Baars, Laura M. Ribeiro do Couto, Jeanette H. W. Leusen, Berend Hooibrink, Taco W. Kuijpers, Susanne M. A. Lens and René A. W. van Lier J Immunol 2000; 165:1910-1917; ; doi: 10.4049/jimmunol.165.4.1910 Downloaded from http://www.jimmunol.org/content/165/4/1910

References This article cites 45 articles, 19 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/165/4/1910.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 29, 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 © 2000 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Cytolytic Mechanisms and Expression of Activation-Regulating Receptors on Effector-Type CD8؉CD45RA؉CD27؊ Human T Cells

Paul A. Baars,1* Laura M. Ribeiro do Couto,*1,2,3 Jeanette H. W. Leusen,† Berend Hooibrink,* Taco W. Kuijpers,* Susanne M. A. Lens,4* and Rene´A. W. van Lier5*

Circulating CD8؉ T cells with a CD45RA؉CD27؊ phenotype resemble cytolytic effector cells because they express various cytolytic mediators and are able to execute cytotoxicity without prior stimulation in vitro. We here demonstrate that .CD8؉CD45RA؉CD27؊ T cells can use both granule exocytosis and Fas/Fas pathways to induce in target cells The availability of these cytolytic mechanisms in circulating T cells suggests that the activity of these cells must be carefully

controlled to prevent unwanted tissue damage. For this reason, we analyzed the expression of surface receptors that either enhance Downloaded from or inhibit function. Compared with memory-type cells, effector cells were found to express normal levels of CD3⑀ and TCR␨ and relatively high levels of CD8. CTLA-4 was absent from freshly isolated effector cells, whereas a limited number of unstimu- lated memory cells expressed this molecule. In line with recent findings on CD8؉CD28؊ T cells, CD45RA؉CD27؊ T cells were unique in the abundant expression of NK cell-inhibitory receptors, both of Ig superfamily and C-type lectin classes. Binding of NK cell-inhibitory receptors to classical and nonclassical MHC class I molecules may inhibit the activation of the cytolytic machinery ؉ ؉ ؊ induced by either Ag -specific or nonspecific signals in CD8 CD45RA CD27 T cells. The Journal of Immunology, 2000, http://www.jimmunol.org/ 165: 1910–1917.

ffective immune responses against viruses depend for a assumption, secretion profile, activation requirements, large part on the induction of CTL that lyse virally in- and the expression of adhesion and homing receptors on fected cells. Studies initially performed in mice have CD45RAϪCD45R0ϩ T cells resemble those of Ag-experienced E Ϫ ϩ shown that after infection, viral peptide-specific T cells undergo T cells (8, 9). Next to the CD45RA CD45R0 subset, a second massive expansion and rapidly acquire effector function (1–3). Af- subpopulation with features of in vivo priming is found in hu- ter elimination of the pathogen, most of the CD8ϩ Ag-specific man peripheral blood. This population is characterized by the die as a result of apoptosis. However, a portion of absence of the costimulatory molecules CD28 (10) and CD27 by guest on September 29, 2021 specific T cells persists (1) and constitutes a pool of memory cells, (7), the presence of both CD45RA and CD57 Ags (11), and which on challenge with the same or a similar pathogen is able to abundant expression of CD11a (12). Because CD8ϩ T cells mount a more forceful immune response (4). Recent studies on within this subset express cytolytic mediators such as perforin, EBV infection have shown that similar mechanisms are opera- granzyme A and B, and (FasL) 6 mRNA and are able tional in the human system (5). to execute cytotoxicity without prior in vitro stimulation, it is In humans, cell surface marker analysis has been thoroughly suggested that they represent effector cells in vivo (13). Inter- ϩ used to discern functional distinct subsets of CD8 T cells. estingly, although these effector type cells have poor prolifer- Proliferative responses to viral Ags are predominantly confined ative potential in vitro, it has been documented that these cells Ϫ ϩ to the CD45RA CD45R0 subset, which suggests that this increase with age (14). The recent observation (15) that these population contains memory-type cells (6, 7). In line with this cells lack the CCR7 indicates that they do not recirculate through the secondary lymphoid organs but *Department of Immunobiology and Laboratory for Experimental and Clinical Im- rather migrate to sites of inflammation. Finally, compatible with munology, Academic Medical Center, CLB, University of Amsterdam, Amsterdam, selection by specific Ag in vivo is the observation that their † The Netherlands; and Department of Cell Biology, University Medical Center Utre- ␤ cht, Utrecht, The Netherlands TCR V repertoire is strongly skewed when compared with ϩ Received for publication January 18, 2000. Accepted for publication June 6, 2000. either naive or memory CD8 T cells (16). Indeed, recent stud- ies using HLA/viral-peptide tetrameric complexes have demon- The costs of publication of this article were defrayed in part by the payment of page ϩ Ϫ charges. This article must therefore be hereby marked advertisement in accordance strated that specific CD8 T cells with CD28 and with 18 U.S.C. Section 1734 solely to indicate this fact. CD45RAϩCD27Ϫ phenotypes can be found in EBV, hepatitis C 1 P.A.B. and L.M.R. contributed equally to this study. virus, and CMV carriers (17, 18). 2 L.R. was supported by Fundac¸a˜o para a Cieˆncia e a Tecnologia, Portugal (Grant Studies with in vitro expanded T cell clones and experiments in PRAXIS XXI/BD/9156/96). mutant mice have provided evidence that CTL can exert their ef- 3 Current address: Department of Molecular Immunology, National Institute of Public fector functions by at least two independent pathways: granule Health and Environmental Protection, P.O. Box 1, 3720 BA Bilthoven, The Netherlands. exocytosis and the Fas/FasL pathway (19). The granule exocytosis 4 Current address: Institut Biochimi Universite´de Lausanne, Chemin de Boveresses 155, 1066 Epalinges, Switzerland. 5 Address correspondence and reprint requests to Dr. Rene´A. W. van Lier, Depart- ment of Immunobiology, Plesmanlaan 125, 1066 CX Amsterdam, The Netherlands. 6 Abbreviations used in this paper: FasL, Fas ligand; RT, room temperature; CMA, E-mail address: [email protected] concanamycin A; NKRs, NK cell-inhibitory receptors.

Copyright © 2000 by The American Association of Immunologists 0022-1767/00/$02.00 The Journal of Immunology 1911 pathway involves secretion of granules containing cytotoxic effec- cells with anti-human granzyme A Ab (45 min at RT). Cells were then tor molecules onto the surface of target cells. Perforin plays a washed three times and incubated with goat anti-mouse IgG FITC (30 min critical role in this pathway, because it can polymerize to form at RT). Moviol containing 2.5% diazabicyclo octane (Fluka, Buchs, Swit- zerland) was used to mount on the cells. For immediate analysis, we made channel-like structures in target cell membrane, through which use of a Leitz DMIRB fluorescence microscope (Leica, Voorburg, The granzymes can enter and subsequently activate the death machin- Netherlands) interfaced with a Leica TCS4D confocal laser microscope ery (20). Alternatively, CTLs can use the Fas pathway to kill their (Leica, Heidelberg, Germany). Images were imported in Adobe Photoshop target cells. This cytolytic pathway is mainly based on cell-cell 4.0 (Adobe Systems, San Jose, CA). interaction between Fas, expressed on the target cell, and FasL, Induction and measurement of granzyme release expressed on the CTL (21). Engagement of the results ϩ in the aggregation of its intracellular death domains, followed by CD8 subpopulations were isolated as described above. Cells were cul- tured in triplicate in 96-well plates (0.3 ϫ 106 cells/well) in either the the activation of several with the ultimate death of the presence or the absence of 1 ng/ml PMA and 1 ␮M ionomycin (both from target cell (22). We here show that both granule exocytosis and Sigma, St. Louis, MO). Supernatants were harvested after 1 h culture, and FasL pathways are readily operational in circulating effector-type the levels of soluble granzyme A and B were measured by ELISA. Stim- CD8ϩ T cells. Furthermore, in search for potential mechanisms ulated cells were used for intracellular evaluation of these molecules ac- that could control the cytotoxic machinery of these cells in vivo, cording to the protocol described below. ⑀ ␨ Granzyme A and B concentrations in culture supernatants were mea- we found normal expression of CD3 and CD3 chains and ab- sured by specific solid phase sandwich ELISA as previously described (27) sence of CTLA-4 expression. However, in agreement with recent with some modifications. The granzyme A ELISA was performed using the finding on CD28Ϫ T cells (23, 24), a strong increase in the ex- GA29 mAb (coating Ab; CLB) and the biotinylated GA28 mAb (detection pression of various classes of killer-inhibitory receptors was ob- Ab; CLB). The limits of detection of the granzyme A and B ELISA sys- Downloaded from tems were 12 and 3 pg/ml, respectively. served when cells maturate into CD8ϩCD45RAϩCD27Ϫ effector T cells (25). Cytotoxicity assay To evaluate the contribution of the Fas/FasL pathway by effector CD8ϩ T Materials and Methods cells, a redirected cytotoxic assay was performed as previously described Antibodies (28) with some modifications. Briefly, to specifically block cytotoxicity ϩ

mediated by perforin and granzymes, isolated CD8 T cells were pre- http://www.jimmunol.org/ For the analysis of expression of cell surface and intracellular molecules, treated (or not) for 4 h with 400 nM concanamycin A (CMA) (Sigma), the following mouse anti-human mAbs were used: unlabeled Fas2 and washed once, and cultured in the presence of CMA. Next, cells were cocul- CD3, CD27 FITC, and CD3 FITC and biotinylated (GB12) tured in 24-well plates (106 cells/well) with FsA target cells (0.2 ϫ 106 (CLB, Amsterdam, The Netherlands); CD45RA PECy5 and TCR␨ FITC cells/well) (E:T 5:1). The redirected cytotoxic assay conditions were imi- (Serotec Kidlington, Oxford, U.K.); CD94, NKG2a, CD158a, CD158b, tated by incubating the cells for 18 h with anti-CD3 (clone CLB-T3/4.1) to and CD45RA, CTLA-4 (CD152), all PE labeled (Coulter-Immunotech, ϩ stimulate effector CD8 T cells and bind to the FcR expressed on Miami, FL); CD8 peridinin chlorophyll , CD45R0 APC, CD16 PE, Ramos.FSA cells. To evaluate the existence of Fas/FasL-dependent cyto- NKB1-FITC (Becton Dickinson, San Jose, CA); unlabeled FasL (Nok-2) toxicity, blocking Abs (anti-FasL (NOK-2) or anti-Fas (Fas2)) were added (PharMingen, San Diego, CA); unlabeled granzyme A, perforin FITC, to the cultures. Apoptosis of Ramos.FSA cells was evaluated using FITC- (Ho¨lzel Diagnostika, Cologne, Germany); and annexin V FITC (Nexins labeled annexin V as described previously (26). Ramos.FSA cells were ϩ

Research, Hoeven, The Netherlands). As a conjugate for unlabeled or bi- by guest on September 29, 2021 discriminated from CD8 T cells by gating on forward and side scatter otinylated mAbs, goat F(abЈ) anti-mouse IgG2a PE (Southern Biotech- 2 parameters (data not shown). nology Associates, Birmingham, AL), goat anti-mouse FITC (CLB), and streptavidin red 670 (Life Technologies, Gaithersburg, MD) were used, Flow cytometry respectively. Isotype-matched mAb served as controls. Expression of cell surface molecules. Staining for NK cell-inhibitory re- Cells ceptors (NKRs) was performed by incubating freshly isolated PBMC with PBMC were isolated from buffy coats of healthy blood donors by density saturating amounts of directly labeled CD8, CD45RA, CD27, and an anti- centrifugation with Ficoll-Isopaque (Pharmacia Biotech, Uppsala, Swe- MHC class I killer-inhibitory receptor mAb in PBS containing 0.5% BSA den). Subsequently, CD8ϩ T cells (Ͼ97% TCR␣␤ϩCD8ϩ cells as assessed (Bayer, Kankakee, IL) (30 min at 4°C). Expression of the different markers was measured on a FACScalibur (Becton Dickinson) and analyzed with the by flow cytometry; data not shown) were prepared by incubating PBMC ϩ with anti-CD8 microbeads (Miltenyi Biotec, Bergisch Gladbach, Germa- Cell Quest program (Becton Dickinson). CD8 T cells were gated, and ny), followed by positive selection with the VarioMACS (Miltenyi Biotec) NKRs were analyzed on the different CD8 subsets. according to the manufacturer’s recommendations. For subset purification Expression of cytoplasmic molecules/epitopes. Intracellular content of ϩ (7), CD8ϩ T cells were stained with CD45RA and CD27 and sorted on a FasL was measured in freshly isolated CD8 T cells before and after in FACStar (Becton Dickinson) into CD45RAϩCD27ϩ (naive fraction), vitro stimulation. FasL on stimulated cells was measured in the presence of CD45RAϩCD27Ϫ (effector fraction), and CD45RAϪCD27ϩ (memory an inhibitor of protein secretion, resulting in the cytoplasmic accumulation fraction) populations (Ͼ95% purity as assessed by flow cytometry; data not of the synthesized FasL. After cell fixation and permeabilization, intracel- shown). For NK isolation, PBMC were stained with CD3 and CD16 mAb lular staining was performed according to a protocol originally described ϩ and sorted on a FACStar into CD3ϪCD16ϩ cells (Ͼ96% purity). by Jung et al. (29) with some modifications. Briefly, isolated CD8 T cells The Fas-sensitive subclone Ramos.FsA of the Burkitt lymphoma Ramos were stimulated (106 cells/ml) for 4 h with PMA (1 ng/ml) and ionomycin (26) has been previously described. (1 ␮M) in the presence of the protein secretion inhibitor monensin (1 ␮M) (all from Sigma). Next, the cells were washed twice in cold PBS-0.5% Cell culture BSA and stained with CD45RA and CD27 (30 min at 4°C, washed twice with cold PBS, and fixed with PBS containing 4% paraformaldehyde (5 All culture experiments were performed in IMDM (Life Technologies, min at 4°C). Fixation was followed by permeabilization with PBS con- Gaithersburg, MD) enriched with 10% heat-inactivated FCS (Euro Bio- ␮ taining 0.1% saponin (Calbiochem, La Jolla, CA) and 0.5% BSA. Non- chem, Bierges, Belgium), gentamicin (43 g/ml), and 2-ME (0.0035%) specific binding was blocked by incubating the cells in the same buffer (culture medium). supplemented with 10% human pooled serum (CLB) (20 min at 4°C). For Confocal laser microscopy all subsequent incubation and washing steps, PBS, 0.1% saponin, 0.5% BSA was used. Cells were then washed once and stained with 5 ␮g/ml Purified CD45RAϩCD27ϩ, CD45RAϩCD27Ϫ, and CD3ϪCD16ϩ popula- anti-FasL (30 min at 4°C). After another washing step, cells were stained tions were isolated as described above. After adhesion to poly-L-lysine- with PE-labeled goat anti-mouse IgG2a isotype-specific mAb (20 min at coated object glasses (30 min at room temperature (RT), cell populations 4°C). Expression of the different markers was measured on a FACScalibur were fixated in PBS containing 3% paraformaldehyde (60 min at RT). and analyzed with the Cell Quest program. CD8ϩ T cells were gated, and

Next, cells were incubated in PBS containing 50 mM NH4Cl (5 min at RT) FasL was analyzed on the different CD8 subsets. and subsequently permeabilized in PBS/0.1% saponin/0.5% BSA (30 min Expression of CTLA-4 and TCR␨ was performed using a permeabili- at RT). After washing, intracellular staining was performed by incubating zation and staining protocol identical with that described above. 1912 CYTOLYTIC PATHWAYS AND INHIBITORY RECEPTORS OF EFFECTOR CD8ϩ T CELLS

FIGURE 1. Confocal laser microscopy anal- ysis of granzyme A expression in purified CD8ϩ subsets and NK cells. Expression is shown in the total CD8ϩ population (a) and is compared with the sorted naive CD45RAϩCD27ϩ (b), effector CD45RAϩCD27Ϫ(c), and NK cell (CD3ϪCD16ϩ) population (d). Downloaded from http://www.jimmunol.org/

pacities of the different subsets, nearly all effector CD8ϩ T cells Results by guest on September 29, 2021 Utilization of cytotoxic pathways by circulating effector cells contained granzyme A in a granular fashion (Fig. 1c), whereas in contrast, naive (CD45RAϩCD27ϩ) CD8ϩ T cells did not express We previously found by flow cytometry that CD8ϩCD45RAϩCD27Ϫ granzyme A (Fig. 1b). Moreover, the expression of this molecule effector-type T cells abundantly express components of the exo- in CD8ϩCD45RAϩCD27Ϫ T cells parallels in both qualitative and cytotic cytolysis pathway, i.e., granzyme A, B and perforin (Ref. 7 quantitative terms the expression observed in freshly isolated NK and data not shown). To test whether in circulating CTL these cells (Fig. 1d). cytotoxic enzymes are contained within granules and are thereby Next we investigated the capacity of CD8ϩ T cells to release ready for receptor-induced exocytosis, the intracellular distribution granzyme A and B on stimulation with PMA and ionomycin. Con- of granzyme A in CD8ϩ T cells (Fig. 1a) was studied by confocal siderable concentrations of both granzyme A and granzyme B scanning laser microscopy. In accordance with the cytotoxic ca- could be measured in supernatants from cultures of effector CD8ϩ

FIGURE 2. Release of granzyme A and B after 1 h stimulation with PMA (1 ng/ml) and ionomycin (iono) (1 ␮M). CD8ϩ total (Ⅺ), CD45RAϩCD27ϩ (u), CD45RAϪCD27ϩ (1 ), CD45RAϩCD27Ϫ (f) were sorted and stimulated for 1 h. Granzymes A and B were measured in the supernatants by ELISA. The Journal of Immunology 1913

FIGURE 3. Extracellular (top) and intracellu- lar (bottom) staining of FasL on CD8ϩ subsets. Unstimulated cells (left) and 4-h PMA/ionomycin/ monensin-stimulated cells (right) were stained ex- tracellularly for CD45RA and CD27. After fixation and permeabilization, cells were stained intracellu- larly for FasL. In the histograms, the solid line shows FasL expression, whereas the dotted line shows the isotype-matched control mAb. FasL- positive cells were colored black and plotted back into the CD45RA and CD27 dot plot, whereas FasL-negative cells were colored gray. Numbers in the dot plot give the percentage of FasL-positive within the four main subsets and correspond clock- Downloaded from wise with the CD45RAϩCD27ϩ, CD45RAϪ CD27ϩ, CD45RAϪCD27Ϫ, and CD45RAϩ CD27Ϫ subset. http://www.jimmunol.org/

T cells (Fig. 2). Moderate levels of granzyme A could also be detected supernatants corresponded with a decrease in the intracellular levels in supernatants of memory (CD45RAϪCD27ϩ) cells, in accordance on stimulated cells as evaluated by flow cytometry (data not shown). with low levels of expression of granzyme A in this subset (data not These results indicate that the stored granzymes (A and B) are rapidly shown). In contrast, no significant release of granzyme B was ob- released from effector cells after stimulation. by guest on September 29, 2021 served when naive or memory cells were stimulated in our experi- mental conditions. Furthermore, the presence of granzymes in the

FIGURE 4. FasL-dependent killing of CD8ϩ effector cells. Isolated CD8ϩ cells were not pretreated (A) or pretreated (B) with CMA to abolish perforin-granzyme-containing granules. A redirected killer assay was per- FIGURE 5. Intracellular staining of TCR␨ in CD8 subsets. Purified formed with the Fas-sensitive Ramos clone FsA with addition of anti-CD3. CD8ϩ cells were stained extracellularly for CD45RA and CD27. After E:T ratio, 5:1. FACS staining with annexin V-FITC in combination with fixation and permeabilization, cells were stained intracellularly with TCR␨. propidium iodide assayed the percentage of apoptotic cells. After 18 h of TCR␨ brightly expressing cells were colored black and plotted back into culture, ϳ60% of the Ramos.FSA cells specifically bound annexin the CD45RA and CD27 dot plot, whereas TCR␨ dull cells were colored V-FITC. Anti-FasL (Nok-2) and anti-Fas (Fas-2) mAbs were added to gray. The dotted line shows the negative control. Numbers in the dot plot block FasL-dependent killing, whereas ␥1 and ␥2a irrelevant mAbs were give the percentage of TCR␨ positive within the four main subsets and added as controls. The condition with anti-CD3 alone was set to 100%, and correspond clockwise with the CD45RAϩCD27ϩ, CD45RAϪCD27ϩ, the percentage of apoptotic cells is given as the percentage of this control. CD45RAϪCD27Ϫ, and CD45RAϩCD27Ϫ subset. 1914 CYTOLYTIC PATHWAYS AND INHIBITORY RECEPTORS OF EFFECTOR CD8ϩ T CELLS

Table I. CTLA-4 expression in CD8 T cell subsetsa

“Naive” “Memory” “Effector” CD45RAϩ CD45RAϪ CD45RAϪ CD45RAϩ Stimulus CD8 Total CD27ϩ CD27ϩ CD27Ϫ CD27Ϫ

Medium 4b 11114 2 CD2 24 19 48 42 11 CD2 ϩ CD28 34 10 50 53 32

a Purified CD8 cells were stimulated as indicated for 16 h in the presence of monensin. Extracellular staining was performed for CD45RA and CD27. After fixation with PFA, cells were intracellularly stained for CTLA-4. All isotype-matched controls were Ͻ1%. One representative experiment is shown of three. b Percentage of CTLA-4ϩ cells within given subset.

FasL mRNA is readily detectable in CD45RAϩCD27Ϫ cells but of Fas on the Ramos.FSA cells, because addition of either Fas- hardly in one of the other circulating CD8ϩ T cell subsets (7). We blocking mAb (Fas2) or FasL mAbs reduced apoptosis by 74 and evaluated the extracellular and intracellular expression of FasL 46%, respectively (Fig. 4). Because combined usage of CMA and protein in either unstimulated or stimulated CD8ϩ T cell subsets. Fas-blocking mAb reduced target cell lysis by ϳ90%, we believe Unstimulated cells did not show extracellular expression of FasL, that the granule exocytosis and FasL pathway account for the ma- whereas intracellular expression of FasL was abundantly found in jority of cytolytic activity of these circulating effectors. Downloaded from CD45RAϩCD27Ϫ cells (Fig. 3, left). To assess the expression of FasL after stimulation, purified CD8ϩ T cells were cultured for 4 h Expression of activation-regulating molecules on effector-type in the presence or absence of PMA and ionomycin. Even in the T cells presence of the metalloproteinase inhibitor KB8301 only a mini- The abundance of cytolytic mediators in circulating effector CTLs mal amount of extracellular FasL could be detected (data not suggests that the activity of these cells must be carefully controlled

shown). Stimulation did, however, result in an increase in the in- to prevent unwanted tissue damage. In a number of in vivo con- http://www.jimmunol.org/ tracellular expression of FasL in the memory compartment (Fig. 3, ditions, such as autoimmunity, tumor growth, and HIV infection right). (31–34), low responsiveness of T cells may be induced by down- To investigate whether both granule exocytosis and the Fas/ regulation of the expression of TCR␨, which is an essential sig- FasL pathways contribute to the cytotoxic potential of circulating naling component in the CD3/TCR complex. To investigate effector cells, a redirected killer assay was performed. For this, whether low expression of TCR␨ would be involved in the control effector CD8ϩ cells were tested for their ability to induce apoptosis of the activation of cytotoxic effector cells, we analyzed the ex- in a Fas-sensitive subclone of the Burkitt lymphoma cell line pression of TCR␨ on the different CD8ϩ T cell subsets. As shown Ramos, i.e., Ramos.FSA (26). Cytotoxicity was initiated by adding in Fig. 5, memory and effector CTLs expressed comparable high CD3 mAb to stimulate CD8ϩ cells and bind to the FcR of the levels of TCR␨ whereas the naive subset expressed TCR␨ less by guest on September 29, 2021 Ramos.FSA cells. Evaluation of apoptotic cell death on the target intensively. Moreover, because effector CTLs also express rela- cells was performed by annexin V staining. To estimate the con- tively high levels of CD3⑀ and the coreceptor CD8 (data not tribution of the exocytosis pathway to the killing by shown), we conclude that it is unlikely that circulating effectors are CD8ϩCD45RAϩCD27Ϫ T cells, CMA, a selective inhibitor of kept at a low level of activation by suboptimal signaling via perforin-based killing, was used (30). Apoptosis of Ramos.FSA TCR/CD3. cells by freshly isolated effector CTLs could be blocked by ϳ55% CTLA-4 acts as a negative regulator of T cell activation and is when cells were incubated with CMA showing contribution of the vital for the control of homeostasis (35). To evaluate granule exocytosis pathway (data not shown). The remaining cy- whether this molecule could be involved in the regulation of totoxic response appeared to be largely dependent on the triggering activation of circulating effector CTLs, CTLA-4 expression was

FIGURE 6. Extracellular staining of CD94 and CD158a in CD8ϩ subsets. Purified CD8ϩ cells were stained extracellularly for CD45RA, CD27, together with CD94 or CD158a. CD94- or CD158a-positive cells were colored black and plotted back into the CD45RA and CD27 dot plot whereas negative cells were colored gray. The dotted line shows the negative control. Numbers in the dot plot give the percentage of CD94- or CD158a-positive cells within the four main subsets and correspond clockwise with the CD45RAϩCD27ϩ, CD45RAϪ CD27ϩ, CD45RAϪCD27Ϫ, and CD45RAϩ CD27Ϫ subset. The Journal of Immunology 1915

Table II. Expression of NKRs on CD8 T cell subsets in comparison with NK cells

CD45RAϩ CD45RAϪ CD45RAϪ CD45RAϩ NK Cell CD3Ϫ CD8 Total CD27ϩ CD27ϩ CD27Ϫ CD27Ϫ CD56ϩ

CD94 43 Ϯ 11a 3 Ϯ 126Ϯ 14 64 Ϯ 24 90 Ϯ 493Ϯ 4 (31–59)b (2–5) (9–44) (34–93) (86–98) (90–98) NKG2a 6 Ϯ 3 0.6 Ϯ 0.3 7 Ϯ 212Ϯ 11 10 Ϯ 747Ϯ 7 (4–12) (0–0.9) (4–9) (4–36) (2–20) (42–55) CD158a 11 Ϯ 14 0.4 Ϯ 0.4 0.3 Ϯ 0.2 3 Ϯ 632Ϯ 37 32 Ϯ 21 (0.6–39) (0.1–1.2) (0.1–0.5) (0.2–15) (6–92) (10–51) CD158b 9 Ϯ 6 1.4 Ϯ 0.8 2.0 Ϯ 0.8 12 Ϯ 10 22 Ϯ 11 45 Ϯ 12 (3–20) (0.4–3) (0.7–3) (2–30) (9–37) (33–56) NKB-1 1.9 Ϯ 0.8 0.3 Ϯ 0.3 0.1 Ϯ 0.1 3 Ϯ 54Ϯ 214Ϯ 1 (1.1–3.2) (0–0.7) (0–0.4) (0–11) (2–7) (13–15)

a Percentage of positive cells within given subset with SEM (n ϭ 6). b Numbers in parentheses, range.

analyzed. CD8ϩ T cells were stained for CD45RA, CD27, and to reliably demonstrate FasL on the surface of effector cells. How-

CTLA-4 before and after stimulation with CD2 and CD28 mAbs. ever, the contribution of FasL in the induction of apoptosis in a Downloaded from On unstimulated cells, this T cell regulator was almost absent in Fas-sensitive Burkitt lymphoma line demonstrates that, in accor- effector CTLs whereas ϳ10% of memory CTLs expressed this dance with previous findings in mice (38, 39), in vivo matured molecule (Table I). On activation with anti-CD2 mAbs, both ef- human CTLs can make use of the FasL effector pathway fector CTLs and memory-type cells had comparable but moderate The strong expression of these cytolytic molecules and the func- expression, i.e., 32 and 50% positive cells, respectively (Table I). tional availability of these compounds upon activation suggest that

The absence on freshly isolated cells and the low expression on in the activity of these circulating effectors must be carefully con- http://www.jimmunol.org/ vitro activated T cells render it unlikely that CTLA-4 acts as a trolled to prevent unwanted damage to healthy cells, e.g., after the major negative regulator of effector CTL activation. recognition of cross-reactive peptides. NKRs, which have been Ϫ MHC class I inhibitory receptors have been found on CD28 T orginally described on NK cells, can after binding to MHC class I cells, and studies in which T cell clones were activated by (super) molecules transduce inhibitory signals for cellular cytotoxicity Ags have shown that inhibitory receptors can modulate effector (40). Concordant with previous findings on CD28Ϫ T cells, we CTL function (36). We analyzed the expression of NKRs during found that CD8ϩCD45RAϩCD27Ϫ T cells express a variety of proposed stages of CD8ϩ T cell differentiation (25). Naive ϩ ϩ ϩ NKR, both of the C-type lectin and Ig superfamily classes (23, 24). (CD45RA CD27 ) CD8 T cells virtually lacked any of the NKRs ϩ Ϫ ϩ Ϫ ϩ In healthy individuals the majority of CD8 CD27 T cells express by guest on September 29, 2021 analyzed. In contrast, circulating effector (CD45RA CD27 ) CD8 CD45RA but lack CD28 showing that under physiological condi- T cells showed a strong expression of all NKRs analyzed which was Ϫ Ϫ tions CD27 and CD28 subsets represent largely overlapping with respect to CD94 and CD158a,b comparable with the expression ϩ populations of CD8 T cells (7). However, when patients with levels on NK cells (Fig. 6, Table II). Interestingly, within the Ϫ Ϫ acute viral infections are being analyzed a very considerable por- CD45RA subset, CD27 cells had a relatively high NKR ex- Ϫ Ϫ tion of CD45RA CD28 T cells does express CD27 (25, 47). We pression. The present data corroborate that the differentiation of ϩ have postulated that by combined analysis of CD27, CD28, and human CD8 T cells toward effector CTL is accompanied by an CD45RA expression, distinct stages of differentiated CD8 T cells increase in the expression of NKRs. can be identified (25). We here show that the differentiation toward ϩ ϩ Ϫ Discussion CD8 CD45RA CD27 “effector-type” T cell is not only accom- panied by an enhancement of intracellular cytolytic mediators but Class I-restricted effector CTL specifically recognize cells that ex- also coincides with an strong increase in the expression of various press peptides derived from intracellular replicating pathogens and NKRs. A complicating finding in appreciating the functional con- may subsequently induce death of these cells. The data presented here show that CTL that have developed in vivo and that can be sequences of the expression of these molecules is that the mAb that identified by the CD8ϩCD45RAϩCD27Ϫ surface phenotype have are being used to detect NKR of the Ig family will also bind to two separate death-inducing mechanisms at their disposal, the splice variants of these molecules that can act as activating recep- granule exocytosis and the FasL pathway. The coexpression of tors, i.e., killer cell activating receptor (41). Differing with Mingari these systems suggests that, dependent on properties of the in- et al. (23), we find highest expression of these regulating receptors on CD45RAϩCD27Ϫ and therefore CD28Ϫ T cells (7) and not on fected cell, distinct elimination strategies can be used by these Ϫ Ϫ effectors in vivo. In addition, IFN-␥ and TNF-␣ that are highly CD45RA CD27 T cells. Although it has recently been reported produced by these cells (7) likely aid in coping adequately with that IL15 can induce CD94/NKG-2A expression on mitogenically infections, e.g., by limiting virus replication (37). activated T cells (42), it is less clear which specific signals do Although circulating effector T cells are well equipped with induce the expression of Ig superfamily type NKRs on T cells. T these cell death-inducing pathways, they do not, without stimula- cell cloning experiments have indicated that the expression of tion, release granzymes or perforin, nor do they express FasL on NKRs is a stable phenotypic trait. In some donors, very high per- the plasma membrane. Short term stimulation in vitro, however, centages of both C-type lectin and Ig superfamily NKRs were Ϫ induces a very rapid release of the constituents of the exocytosis found specifically in the CD27 T cell fraction. This finding could pathway in culture supernatants. On the other hand, although suggest that concomitant with their differentiation toward differ- CD8ϩCD45RAϩCD27Ϫ T cells contain FasL mRNA (7) and ex- entiated effector cells, CD8ϩCD45RAϩCD27 receive specific sig- press limited but discernible amounts of FasL intracellularly, we nals that induce up-regulation of these cell activation-regulating were unable, even in the presence of specific proteinase inhibitors, receptors. 1916 CYTOLYTIC PATHWAYS AND INHIBITORY RECEPTORS OF EFFECTOR CD8ϩ T CELLS

ϩ Ϫ Multiple studies have documented that CD8 CD28 , normal elderly humans: the T cell equivalent to “benign monoclonal gammapa- CD8ϩCD45RAϩCD27Ϫ,orNKRϩ subsets have a restricted and thy.” J. Exp. Med. 179:609. 11. Wang, E. C. Y., J. Taylor-Wiedeman, P. Perera, J. Fisher, and L. K. Borysiewicz. selected usage of the TCR repertoire suggestive of selection by Ag 1993. Subsets of CD8ϩ, CD57ϩ cells in normal, healthy individuals: correlation (10, 16). Indeed, more recently, it was shown that virus-specific with human cytomegalovirus (HCMV) carrier status, phenotypic and functional and tumor Ag-specific CD8ϩCD45RAϩCD27Ϫ T cells can be analyses. Clin. Exp. Immunol. 94:297. 12. Okumura, M., Y. Fujii, Y. Takeuchi, K. Inada, K. Nakahara, and H. Matsuda. found in hepatitis C virus- and CMV-infected persons and mela- 1993. Age-related accumulation of LFA-1high cells in a CD8ϩCD45RAhigh T cell noma patients, respectively (18, 43). Several reports have documented population. Eur. J. Immunol. 23:1057. Ϫ that NKRs can inhibit both superantigen- and Ag-induced activation 13. Azuma, M., J. H. Phillips, and L. L. Lanier. 1993. CD28 T lymphocytes: an- tigenic and functional properties. J. Immunol. 150:1147. of T cells (36, 44, 45). Interestingly, Lee et al. (43) reported that 14. Fagnoni, F. F., R. Vescovini, M. Mazzola, G. Bologna, E. Nigro, G. Lavagetto, ϩ ϩ Ϫ despite the presence of Ag-specific CD8 CD45RA CD27 in the C. Franceshi, M. Passeri, and P. Sansoni. 1996. Expansion of CD8ϩCD28Ϫ T circulation of a melanoma patient, no lysis of melanoma cell line cells in healthy ageing people, including centenarians. Immunology 88:501. 15. Sallusto, F., D. Lenig, R. Forster, M. Lipp, and A. Lanzavecchia. 1999. Two in vitro was observed. Because our data suggest that these cells subsets of memory T lymphocytes with distinct homing potentials and effector will also express high levels of NKRs, inhibition of TCR-induced functions. Nature 401:708. killing can be envisaged. For this reason, it is important to more 16. Hamann, D., S. Kostense, K. C. Wolthers, S. A. Otto, P. A. Baars, F. Miedema, and R. A. van Lier. 1999. Evidence that human CD8ϩCD45RAϩCD27Ϫ cells are precisely define the signals that induce expression of NKRs on induced by antigen and evolve through extensive rounds of division. Int. Immu- effector T cells, because these receptors will seriously interfere nol. 11:1027. with vaccination strategies for, e.g., cancer. 17. He, X. S., B. Rehermann, F. X. Lopez-Labrador, J. Boisvert, R. Cheung, J. Mumm, H. Wedemeyer, M. Berenguer, T. L. Wright, M. M. Davis, and The question remains on the physiological function of circulat- H. B. Greenberg. 1999. Quantitative analysis of hepatitis C virus-specific CD8ϩ ing effector-type cells. The functional relevance of these cells ap- T cells in peripheral blood and liver using peptide-MHC tetramers. Proc. Natl. Downloaded from pears to be supported by the fact that circulating effector-type cells Acad. Sci. USA 96:5692. 18. Kern, F., E. Khatamzas, I. Surel, C. Frommel, P. Reinke, S. L. Waldrop, increase with age (46). Although the presence of differentiated and L. J. Picker, and H. D. Volk. 1999. Distribution of human CMV-specific memory functionally competent CTL could provide a first line of defense T cells among the CD8pos subsets defined by CD57, CD27, and CD45 isoforms. against invading pathogens to which a protective response has al- Eur. J. Immunol. 29:2908. 19. Shresta, S., C. T. Pham, D. A. Thomas, T. A. Graubert, and T. J. Ley. 1998. How ready been mounted once, the expression of NKR will preclude do cytotoxic lymphocytes kill their targets? Curr. Opin. Immunol. 10:581.

rapid activation by these cells by any dose of peptide. Yet, either 20. Henkart, P.A. 1985. Mechanism of lymphocyte-mediated cytotoxicity. Annu. http://www.jimmunol.org/ with decreasing amounts of MHC molecules (which may limit the Rev. Immunol 3:31:31. 21. Berke, G. 1995. The CTL’s kiss of death. Cell 81:9. inhibitory signal) or with increasing specific peptide concentration 22. Cohen, G. M. 1997. Caspases: the executioners of apoptosis. Biochem. J. 326:1. (which will augment the stimulatory signal), these cells may be- 23. Mingari, M. C., F. Schiavetti, M. Ponte, C. Vitale, E. Maggi, S. Romagnani, ϩ come activated. In fact, the integration between these positive and J. Demarest, G. Pantaleo, A. S. Fauci, and L. Moretta. 1996. Human CD8 T lymphocyte subsets that express HLA class I-specific inhibitory receptors repre- negative signals allows the effector CTL to detect the specific load- sent oligoclonally or monoclonally expanded cell populations. Proc. Natl. Acad. ing of MHC molecules with foreign peptides irrespective of the Sci. USA 93:12433. number of MHC molecules expressed on a target cell. 24. Speiser, D. E., D. Valmori, D. Rimoldi, M. J. Pittet, D. Lienard, V. Cerundolo, H. R. MacDonald, J. C. Cerottini, and P. Romero. 1999. CD28-negative cytolytic effector T cells frequently express NK receptors and are present at variable pro- portions in circulating lymphocytes from healthy donors and melanoma patients. by guest on September 29, 2021 Acknowledgments Eur. J. Immunol. 29:1990. 25. Hamann, D., M. T. Roos, and R. A. van Lier. 1999. Faces and phases of human We thank Jo¨rg Hamann, Do¨rte Hamann, and Claire Boog for critical read- CD8 T-cell development. Immunol. Today 20:177. ing of and comments on the manuscript. We also thank Angela Wolbink 26. Lens, S. M., B. F. den Drijver, A. J. Potgens, K. Tesselaar, M. H. van Oers, and and Erik Hack for performing granzyme A and B ELISAs. R. A. van Lier. 1998. Dissection of pathways leading to antigen receptor-induced and Fas/CD95-induced apoptosis in human B cells. J. Immunol. 160:6083. 27. Spaeny-Dekking, E. H., W. L. Hanna, A. M. Wolbink, P. C. Wever, A. J. Kummer, A. J. Swaak, J. M. Middeldorp, H. G. Huisman, C. J. Froelich, and References C. E. Hack. 1998. Extracellular granzymes A and B in humans: detection of 1. Zimmermann, C., K. Brduscha-Riem, C. Blaser, R. M. Zinkernagel, and native species during CTL responses in vitro and in vivo. J. Immunol. 160:3610. H. Pircher. 1996. Visualisation, characterisation, and turnover of CD8ϩ memory 28. Jong, R., M. Brouwer, Rebel, V. I., Van Seventer,G. A. F. Miedema, and T cells in virus-infected hosts. J. Exp. Med. 183:1367. R. A. W. Lier. 1990. Generation of alloreactive cytolytic T lymphocytes by im- 2. Murali-Krishna, K., J. D. Altman, M. Suresh, D. J. Sourdive, A. J. Zajac, mobilized anti-CD3 monoclonal antibodies: requirements for human cytolytic J. D. Miller, J. Slansky, and R. Ahmed. 1998. Counting antigen-specific CD8 T T-lymphocyte differentiation. Immunology 70:357. cells: a reevaluation of bystander activation during viral infection. Immunity 29. Jung, T., U. Schauer, C. Heusser, C. Neuman, and C. Rieger. 1993. Detection of 8:177. intracellular by flow cytometry. J. Immunol. Methods 159:197. 3. Jacob, J., and D. Baltimore. 1999. Modelling T-cell memory by genetic marking 30. Kataoka, T., N. Shinobara, H. Takayama, K. Takaku, S. Kondo, S., S.Yonehara, of memory T cells in vivo. Nature 399:593. and K. Nagai. 1996. Concanamycin A, a powerful tool for characterization and 4. Doherty, P. C. 1998. The new numerology of immunity mediated by virus-spe- estimation of contribution of perforin- and Fas-based lytic pathways in cell-me- cific CD8ϩ T cells. Curr. Opin. Microbiol. 1:419. diated cytotoxicity. J. Immunol. 156: 3678. 5. Callan, M. F. C., L. Tan, N. Annels, G. S. Ogg, J. Wilson, C. A. O’Callaghan, 31. Liossis, S. N., X. Z. Ding, G. J. Dennis, and G. C. Tsokos. 1998. Altered pattern A. J. McMichael, and A. B. Rickinson. 1998. Direct visualization of antigen of TCR/CD3-mediated protein-tyrosyl phosphorylation in T cells from patients specific CD8ϩ T cells during the primary immune response to Epstein Barr virus with systemic erythematosus: deficient expression of the T cell receptor ␨ in vivo. J. Exp. Med. 187:1395. chain. J. Clin. Invest. 101:1448. 6. Merkenschlager, M., L. Terry, R. Edwards, and P. C. Beverley. 1988. Limiting 32. Mulder, W. M., E. Bloemena, M. J. Stukart, J. A. Kummer, J. Wagstaff, and dilution analysis of proliferative responses in human lymphocyte populations R. J. Scheper. 1997. T cell receptor-␨ and granzyme B expression in mononuclear defined by the UCHL1: implications for differential CD45 cell infiltrates in normal colon mucosa and colon carcinoma. Gut 40:113. expression in T cell memory formation. Eur. J. Immunol. 18:1653. 33. Maurice, M. M., A. C. Lankester, A. C. Bezemer, M. F. Geertsma, P. P. Tak, 7. Hamann, D., P. A. Baars, M. H. G. Rep, B. Hooibrink, S. R. Kerkhof-Garde, F. C. Breedveld, R. A. van Lier, and C. L. Verweij. 1997. Defective TCR-me- M. R. Klein, and R. A. W. Van Lier. 1997. Phenotypic and functional separation diated signaling in synovial T cells in rheumatoid arthritis. J. Immunol. 159:2973. of memory and effector human CD8pos T cells. J. Exp. Med. 186:1407. 34. Stefanova, I., M. W. Saville, C. Peters, F. R. Cleghorn, D. Schwartz, 8. Sanders, M. E., M. W. Makgoba, S. O. Sharrow, D. Stephany, T. A. Springer, D. J. Venzon, K. J. Weinhold, N. Jack, C. Bartholomew, W. A. Blattner, et al. H. A. Young, and S. Shaw. 1988. Human memory T lymphocytes express in- 1996. HIV infection–induced posttranslational modification of T cell signaling creased levels of three cell adhesion molecules (LFA-3, CD2, and LFA-1) and molecules associated with disease progression. J Clin. Invest. 98:1290. three other molecules (UCHL-1, CDw29, and Pgp-1) and have enhanced IFN-␥ 35. Walunas, T. L., D. J. Lenschow, C. Y. Bakker, P. S. Linsley, G. J. Freeman, production. J. Immunol. 140:1401. J. M. Green, C. B. Thompson, and J. A. Bluestone. 1994. CTLA-4 can function 9. De Jong, R., M. Brouwer, F. Miedema, and R. A. W. Van Lier. 1991. Human as a negative regulator of T cell activation. Immunity 1:405. ϩ ϩ ϩ CD8 T lymphocytes can be divided into CD45RA and CD45R0 cells with 36. Phillips, J. H., J. E. Gumperz, P. Parham, and L. L. Lanier. 1995. Superantigen- different requirements for activation and differentiation. J. Immunol. 146:2088. dependent, cell-mediated cytotoxicity inhibited by MHC class I receptors on T 10. Posnett, D. N., R. Sinha, S. Kabak, and C. Russo. 1994. Clonal populations in lymphocytes. Science 268:403. The Journal of Immunology 1917

37. Feduchi, E., M. A. Alonso, and L. Carrasco. 1989. Human ␥ interferon and tumor Characterization of circulating T cells specific for tumor-associated antigens in necrosis factor exert a synergistic blockade on the replication of herpes simplex melanoma patients. Nat. Med. 5:677. virus. J. Virol. 63:1354. 44. Bakker, A. B., J. H. Phillips, C. G. Figdor, and L. L. Lanier. 1998. Killer cell 38. Lowin, B., M. Hahne, C. Mattmann, and J. Tschopp. 1994. Cytolytic T-cell inhibitory receptors for MHC class I molecules regulate lysis of melanoma cells cytotoxicity is mediated through perforin and Fas lytic pathways. Nature 370: mediated by NK cells, ␥␦ T cells, and antigen-specific CTL. J. Immunol. 160: 650. 5239. 39. Kojima, H., N. Shinohara, S. Hanaoka, Y. Someya-Shirota, Y. Takagaki, 45. Ikeda, H., B. Lethe, F. Lehmann, N. van Baren, J. F. Baurain, C. de Smet, H. Ohno, T. Saito, T. Katayama, H. Yagita, and K. Okumura. 1994. Two distinct H. Chambost, M. Vitale, A. Moretta, T. Boon, et al. 1997. Characterization of an pathways of specific killing revealed by perforin mutant cytotoxic T lympho- antigen that is recognized on a melanoma showing partial HLA loss by CTL cytes. Immunity 1:357. expressing an NK inhibitory receptor. Immunity 6:199. 40. Lanier, L. L. 1998. NK cell receptors. Annu. Rev. Immunol 16:359. 41. Moretta, A., Sivori, S., Ponte, M., Minganri, M. C., and Loretta, L. 1998. Stim- 46. Rufer, N., T. H. Brummendorf, S. Kolvraa, C. Bischoff, K. Christensen, ulatory receptors in NK and T cells. Curr. Top. Microbiol. Immunol. 230:15. L. Wadsworth, M. Schulzer, and P. M. Lansdorp. 1999. Telomere fluorescence 42. Mingari, M. C., C. Vitale, C. Cantoni, R. Bellomo, M. Ponte, F. Schiavetti, measurements in granulocytes and T lymphocyte subsets point to a high turnover S. Bertone, A. Moretta, and L. Moretta. 1997. Interleukin-15-induced maturation of hematopoietic stem cells and memory T cells in early childhood. J. Exp. Med. of human natural killer cells from early thymic precursors: selective expression 190:157. of CD94/NKG2-A as the only HLA class I-specific inhibitory receptor. Eur. 47. Roos, T. L., R. A. W. van Lier, D. Hamann, G. J. Knol, I. Verhoofstad, D. van J. Immunol. 27:1374. Baarle, F. Miedema, and P. T. A. Schellekens. 2000. Changes in the compostion 43. Lee, P. P., C. Yee, P. A. Savage, L. Fong, D. Brockstedt, J. S. Weber, D. Johnson, of circulating CD8ϩ T-cell subsets during acute Epstein-Barr virus and human S. Swetter, J. Thompson, P. D. Greenberg, M. Roederer, and M. M. Davis. 1999. immunodeficiency virus infection in man. J. Infect. Dis. In press. Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021