Trogocytosis Is a Gateway to Characterize Functional Diversity in Melanoma-Specific CD8+ Clones

This information is current as Ronny Uzana, Galit Eisenberg, Yael Sagi, Shoshana of October 2, 2021. Frankenburg, Sharon Merims, Ninette Amariglio, Eitan Yefenof, Tamar Peretz, Arthur Machlenkin and Michal Lotem J Immunol 2012; 188:632-640; Prepublished online 7 December 2011; doi: 10.4049/jimmunol.1101429 Downloaded from http://www.jimmunol.org/content/188/2/632

Supplementary http://www.jimmunol.org/content/suppl/2011/12/07/jimmunol.110142 http://www.jimmunol.org/ Material 9.DC1 References This article cites 50 articles, 25 of which you can access for free at: http://www.jimmunol.org/content/188/2/632.full#ref-list-1

Why The JI? Submit online.

by guest on October 2, 2021 • 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

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 © 2012 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Trogocytosis Is a Gateway to Characterize Functional Diversity in Melanoma-Specific CD8+ T Cell Clones

Ronny Uzana,* Galit Eisenberg,* Yael Sagi,† Shoshana Frankenburg,* Sharon Merims,* Ninette Amariglio,‡ Eitan Yefenof,x Tamar Peretz,* Arthur Machlenkin,*,1 and Michal Lotem*,1

Trogocytosis, the transfer of membrane patches from target to immune effector cells, is a signature of tumor–T cell interaction. In this study, we used the trogocytosis phenomenon to study functional diversity within tumor-specific T cell clones with identical

TCR specificity. MART-126–35–specific CD8 T cell clones, which differed in their trogocytosis capacity (low [2D11], intermediate [2G1], high [2E2]), were generated from melanoma patients. Functional evaluation of the clones showed that the percentage of trogocytosis-capable T cells closely paralleled each clone’s IFN-g and TNF-a production, lysosome degranulation, and lysis of

peptide-pulsed targets and unmodified melanoma. The highly cytotoxic 2E2 clone displayed the highest TCR peptide binding Downloaded from affinity, whereas the low-activity 2D11 clone showed TCR binding to peptide-MHC in a CD8-dependent manner. TCR analysis revealed Vb16 for clones 2E2 and 2G1 and Vb14 for 2D11. When peptide-affinity differences were bypassed by nonspecific TCR stimulation, clones 2E2 and 2D11 still manifested distinctive signaling patterns. The high-activity 2E2 clone displayed prolonged phosphorylation of ribosomal protein S6, an integrator of MAPK and AKT activation, whereas the low-activity 2D11 clone generated shorter and weaker phosphorylation. Screening the two clones with identical TCR Vb by immunoreceptor array

showed higher phosphorylation of NK, T, and Ag (NTB-A), a SLAM family homophilic receptor, in clone 2E2 compared http://www.jimmunol.org/ with 2G1. Specific blocking of NTB-A on APCs markedly reduced cytokine production by CD8 , pointing to a possible contribution of NTB-A costimulation to T cell functional diversity. This finding identifies NTB-A as a potential target for improving anti-cancer immunotherapy. The Journal of Immunology, 2012, 188: 632–640.

ytotoxic CD8+ T cells specific for a single antigenic (pMHC) not only determines the activation capacity of the for- epitope can originate from one or several progenitors, mer, but also that of its clonal progeny (6). Short and loose C distinguishable by their TCR sequences (1). The structure interactions yield subclones of low potency whereas long and tight of the TCR is the major determinant of T cell functional robust- contact results in highly reactive T cells, thus leading to clonal ness, specifying its binding affinity to a cognate peptide presented diversity (2, 7). Although the mechanism underlying the func- by guest on October 2, 2021 by an appropriate MHC class I molecule (2). In vivo, many other tional imprint of T cells is not yet fully elucidated, it is clear that factors affect the actual cytotoxic capacity of T cells, including in addition to TCR affinity, other interactions take place at the their state of differentiation, shifting from naive cells to terminally initial immunological synapse, which shape T cell functional ca- differentiated effectors, length of activation (3), cytokine milieu pacity. Immunological synapse refers to the focal contact between (4), and costimulation (5). Hence, T cell competence reflects the an immune effector cell and APCs. It is a contact-secretion pro- overall outcome of all these variables. cess that leads to effector cell activation by reciprocal molecular Recently, it has been acknowledged that the initial encoun- clustering of ligands and receptors on the cytoplasmic membranes ter between a naive CD8+ T cell with its cognate peptide-MHC of the cellular partners involved (8, 9). The linkage between receptors and ligands leads to the unidirectional transfer of mem- brane fragments from APCs to effector cells, a process termed *Sharett Institute of Oncology, Hadassah Medical Organization, Jerusalem 91120, Israel; †Division of Oncology, School of Medicine, Stanford University, Stanford, CA “trogocytosis” (10–12). It is readily visible by labeling Ag-loaded 94305; ‡Cancer Research Center, Chaim Sheba Medical Center, Ramat Gan 52621, x APCs with a membrane dye that can be detected on cognate Israel; and Lautenberg Center for General and Tumor Immunology, Hebrew T cells by immunofluorescence (13). Thus, trogocytosis provides University of Jerusalem, Jerusalem 91120, Israel a useful signature for T cells that have interacted with APCs. 1A.M. and M.L. share senior authorship of this work. We have previously shown that trogocytosis also occurs between Received for publication May 16, 2011. Accepted for publication November 2, 2011. tumor-reactive CD8+ T cells and melanoma target cells (14), and This work was supported by the Israel Cancer Association, the Chief Scientist of that it is preferentially associated with a highly cytotoxic T cell the Israel Ministry of Health, and the Dr. Miriam and Sheldon G. Adelson Medical Research Foundation. population. Notably, TCR specificity was not the only determinant Address correspondence and reprint requests to Dr. Arthur Machlenkin and Dr. of trogocytosis, as a proportion of Ag-specific T cells with the Michal Lotem, Sharett Institute of Oncology, Hadassah Medical Organization, P.O. same TCR specificity did not capture target membranes. In the Box 12000, Jerusalem 91120, Israel. E-mail addresses: [email protected] present study, we aim to elucidate the reasons for the functional (A.M.) and [email protected] (M.L.) diversity within CD8+ T cells with the same TCR specificity. The online version of this article contains supplemental material. Abbreviations used in this article: LCK, -specific protein tyrosine kinase; NTB-A, NK, T, and B cell Ag; PLC, phospholipase C; pMHC, peptide-MHC; rhIL-2, Materials and Methods recombinant human IL-2; SLP-76, Src homology 2 domain-containing leukocyte Tumor cell lines and lymphocyte cultures protein of 76 kDa; TIL, tumor-infiltrating lymphocyte. Melanoma cell line M171 (HLA-A22/MART-1+) was established in the Copyright Ó 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00 Sharett Institute of Oncology, Hadassah Medical Organization (Jerusalem, www.jimmunol.org/cgi/doi/10.4049/jimmunol.1101429 The Journal of Immunology 633

Israel) (15). 624mel (HLA-A2+/MART-1+) was a gift of M. Parkhurst cells were then harvested, resuspended in FACS buffer, and surface stained (Surgery Branch, National Institutes of Health, Bethesda, MD). T2 is a with fluorophore-conjugated anti-CD8 mAb for 30 min at 4˚C. Following TAP-2–deficient lymphoblastoid line of HLA-A2 genotype. All cell lines permeabilization with Cytofix/Cytoperm buffer, intracellular proteins were were cultured in RPMI 1640 supplemented with 10% heat-inactivated stained with one of the following mAb: anti–IFN-g-PE or -FITC, anti– FCS, 2 mmol/l L-glutamine, and combined antibiotics. Tumor-infiltrat- TNF-a-PE, anti–perforin-PE, anti-granzyme B-PE (all from eBioscience), ing lymphocyte (TIL) microcultures were initiated and expanded from and anti–TCRz-chain-PE (kind gift from M. Baniyash). Appropriate iso- tumor specimens taken from resected metastases of melanoma patients, as type controls were used. The data were analyzed using FCS Express described (16). Human lymphocytes were cultured in complete medium software (De Novo Software, Los Angeles, CA). consisting of RPMI 1640 supplemented with 10% heat-inactivated hu- man AB serum, 6000 IU/ml recombinant human IL-2 (rhIL-2; Chiron, Cytotoxicity assay Amsterdam, The Netherlands), 2 mmol/l L-glutamine, 1 mmol/l sodium In vitro cytotoxicity assays were performed as described (17). Briefly, pyruvate, 1% nonessential amino acids, 25 mmol/l HEPES (pH 7.4), 50 lymphocytes were washed, resuspended in complete medium, and mixed mmol/l 2-ME, and combined antibiotics (all from Invitrogen Life Tech- 35 nologies). The study was approved by the Institutional Review Board, and at different ratios with 5000 S-L-methionine–labeled melanoma or g all patients gave their informed consent prior to initiation of melanoma peptide-pulsed T2 cells. In some experiments, IFN- –treated (100 U/ml and lymphocyte cell cultures. for 48 h; ProSpec) melanoma cells were used as targets. CTL assays were done in U-shaped microtiter wells at 37˚C, 5% CO2 for 5 h. Percentage of Cloning of peptide-specific TIL specific lysis was calculated as follows: % lysis = [(cpm in experimental well 2 cpm spontaneous release)/(cpm maximal release 2 cpm sponta- On day 14 of TIL initiation, the lymphocytes were washed with PBS, neous release)] 3 100. resuspended in PBS supplemented with 0.5% BSA, and stained with FITC- b conjugated HLA-A*0201/MART-126–35 dextramer (Immudex, Copenha- TCR -chain analysis gen, Denmark) for 30 min at 4˚C. The lymphocytes were then incubated TCR b-chain spectratyping was performed as described previously (18) with allophycocyanin-conjugated mouse anti-human CD8 (eBioscience) Downloaded from + with minor modifications. Briefly, RNA was isolated from logarithmically for an additional 30 min at 4˚C and washed. CD8 lymphocytes, positively + + growing clones using TRI Reagent (Sigma-Aldrich, St. Louis, MO). Total stained by the dextramer (CD8 dextramer cells), were sorted by a BD RNA (4 mg) was reverse transcribed into cDNA with 200 U SuperScript II FACSAria and directly cloned at one or two cells per well in 96-well plates 2 RNase H reverse transcriptase (Life Technologies). cDNA was used as in the presence of 30 ng/ml Ortho anti-CD3 (eBioscience), 6000 IU/ml 4 a template for multiplex PCR amplification of the rearranged TCR b-chain rhIL-2, and 4 Gy-irradiated 5 3 10 allogeneic PBMCs as feeder cells. CDR3 region. Each multiplex reaction contained an antisense primer spe- Five days later, 6000 IU/ml rhIL-2 was added and renewed every 2 d cific for the TCR b-chain C region 59-TTC TGA TGG CTC AAA CAC-39 thereafter. On day 14, the clones were assayed for IFN-g secretion in a

and six or seven TCR b-chain variable gene-specific sense primers, labeled http://www.jimmunol.org/ peptide-specific manner following their coincubation with MART-1 – 26–35 with FAM, hexachlorofluorescein, or NED (Applied Biosystems, Foster pulsed T2 cells (commercially synthesized and purified [.95%] by City, CA). All 24 functional Vb families were analyzed (Supplemental reverse-phase HPLC by Biomer Technology) using commercially available Table 1). Each reaction contained cDNA template, 500 mM dNTPs, 2 mM ELISA reagents (R&D Systems, Minneapolis, MN). The MART-1 – 26–35 MgCl , and 1 U AmpliTaq Gold DNA polymerase (PerkinElmer) in reactive clones were further expanded in a second-round exposure to 30 ng/ 2 AmpliTaq Gold buffer in a final volume of 20 ml. After completion, an ml Ortho anti-CD3 and 6000 IU/ml rhIL-2 in the presence of 50-fold excess aliquot of the PCR product was diluted 1:50 and analyzed via capillary irradiated feeder cells. electrophoresis using an Applied Biosystems 3100 genetic analyzer. The Trogocytosis assay and flow cytometry output of the DNA analyzer is a distribution of fluorescence intensity versus time, which was converted to a distribution of fluorescence intensity T2 cells were pulsed with A2-restricted peptides MART-126–35 or HIV-1 versus length by comparison with the fluorescence intensity trace of a by guest on October 2, 2021 Gag p2419–27 for negative control, 1 mmol/l in Opti-MEM (Life Tech- reference sample containing known size standards. nologies) for 1 h at 37˚C, and washed twice in PBS. Melanoma cells and peptide-pulsed T2 targets were either 1) labeled with PKH26 (Sigma- TCR signaling profile by flow cytometry Aldrich) according to the manufacturer’s protocol or 2) surface bio- Signaling profile analysis of the T cell clones was characterized by detection tinylated with D-biotinoyl-ε-aminocaproic acid-N-hydroxysuccinimide es- of phosphorylated proteins involved in TCR signal propagation. T cells ter (biotin-7-NHS; Roche, Mannheim, Germany) according to the man- were aliquoted evenly in test tubes at 1 3 105 cells per tube in RPMI 1640 ufacturer’s instructions with minor modifications. Briefly, cells were and 10% human AB serum and kept on ice. Soluble mouse anti-human suspended in buffer (20 mM HEPES, 150 mM NaCl, 0.1 mM CaCl , and 1 2 CD3 (clone OKT3; eBioscience) or T2 cells pulsed with 5 mMMART- mM MgCl in PBS) containing 1 mg/ml biotin-7-NHS (Roche) and in- 2 1 or HIV-1 Gag p24 were added to the clones while on ice. The cubated for 10 min at room temperature. Cold RPMI 1640 medium con- 26–35 19–27 cell/stimulus mix was transferred to a 37˚C bath at given time points to taining 10% AB serum was added (1:1 vol) for 10 min at 4˚C to stop yield incubation periods of 1, 4, 15, and 45 min. The reaction was termi- labeling, and the cells were washed three times with PBS containing 1% nated and signaling fixed by adding 1 ml fixation-permeabilization reagent AB serum. Following resuspension in complete medium, labeled tar- (eBioscience). Signaling response was determined on the same day by gets were cocultured with MART-1 –reactive T cells (1–5 3 105 26–35 flow cytometry using fluorochrome-Abs (Becton Dickinson) to detect lymphocytes/well in 200 ml final volume) in U-shaped 96-well plates at phosphorylated lymphocyte-specific protein tyrosine kinase (LCK; Y505) a 1:1 E:T ratio. Following incubation for 1–5 h at 37˚C, cells were washed clone 4/LCK-Y505, CD3z (Y142) clone K25-407.69, ZAP70 (Y319) clone twice in PBS containing 0.5 mM EDTA to ensure conjugate dissociation P-ZAP70, Src homology 2 domain-containing leukocyte protein of 76 kDa and then resuspended in FACS buffer consisting of PBS supplemented with (SLP-76; Y128) clone J141-668.36.58, ERK (T202/Y204) clone 20A, and 0.5% BSA and 0.1% NaN . The cells were stained with allophycocyanin- 3 phospholipase C (PLC)g (Y759) clone K86-689.37. An Ab against phos- conjugated anti-CD8 Abs and PE-conjugated streptavidin (both from phorylated ribosomal protein S6 (S235/S236) clone D57.2.2E was pur- eBioscience) for 30 min at 4˚C and subjected to flow cytometry on a BD chased from Cell Signaling Technologies. The fixed and permeate T cells LSRII (BD Biosciences). The capacity of each clone to perform trogo- were washed with FACS buffer (eBioscience) and incubated with the stain cytosis was defined as percentage of streptavidin+CD8+ cells, gated on mix for 30 min at room temperature. Abs from Becton Dickinson were lymphocytes, following doublets exclusion (13). The gating strategy is used to detect expression of CD3, CD20, and CD19 to gate on the T cell shown in Supplemental Fig. 1. Immunophenotyping of clones was per- clones and exclude T2 cells from the signal analysis. The BD LSRII FACS formed using the following Abs: anti–CD107A-FITC or -allophycocyanin cytometer and Cytobank Web site (http://www.cytobank.org) were used for (eBioscience), anti–CD137-allophycocyanin (BD Biosciences), and anti– data acquisition and analysis. CD8b-chain-PE (Beckman Coulter). For MART-126–35–specific TCR stain- ing in a CD8-dependent and -independent manner, lymphocytes were Profiling immunoreceptor phosphorylation in T cells stained with FITC-conjugated dextramers (Immudex) and PE-conjugated iTAg tetramers (Beckman Coulter), respectively, according to the manu- T cells were activated by plate-bound anti-CD3 mAb (eBioscience) for 30 facturers’ instructions. For intracellular cytokine/marker staining, a Cyto- min at 37˚C. The cells were harvested and solubilized in lysis buffer, and fix/Cytoperm Plus kit with BD GolgiPlug (BD Biosciences) was used sample protein concentrations were determined using Bradford’s reagent according to the manufacturer’s recommendations. Briefly, 1–5 3 105 (Bio-Rad Laboratories). Then, 400 mg cell lysate of each clone was applied lymphocytes were cocultured with peptide-pulsed T2 or melanoma cells on a human phospho-immunoreceptor array (Proteome Profiler array; for 6 h at a 1:1 ratio, as described above. Brefeldin A (1 mg/ml, GolgiPlug) R&D Systems) that assays a panel of phosphotyrosine molecules associ- was added for the last 4 h to enable intracellular protein accumulation. The ated with immune response. Cell lysates, conditioning of the membranes, 634 A STUDY OF T CELL FUNCTIONAL DIVERSITY BY TROGOCYTOSIS hybridization, and development were performed according to the manu- facturer’s protocol. The array signals were visualized using luminescent image analyzer LAS-3000 (Fujifilm) by exposure to ECL Western blotting substrate (Pierce), and then quantified with TINA 2.0 software using MiniBIS Pro (DNR Bio-Imaging Systems).

NK, T, and B cell Ag blocking on target cells To mask NK, T, and B cell Ag (NTB-A) on target cells, 5 mg/ml monoclonal anti–NTB-A Ab (clone NT-7) or mouse IgG1 k isotype control Abs (both from BioLegend, San Diego, CA) were added to 106/ml peptide-pulsed T2 or 624mel cells, suspended in complete medium. Following incubation for 30 min at 4˚C, cells were washed twice in cold PBS and cocultured with effectors at 1:1 ratio for 6 h at 37˚C.

Results Generation of MART-126–35–specific TIL-derived clones with variable trogocytosis capacity To elucidate key elements that contribute to functional diversity among CD8+ T cells of the same TCR specificity, we generated a panel of MART-126–35–specific T cell cultures by single cell Downloaded from sorting (Fig. 1A). MART-126–35 specificity of the clones was confirmed by dextramer staining, demonstrating that 5 of 10 growing clones were peptide specific (Fig. 1B). The dextramer- positive clones were examined for their ability to capture target membranes. The percentage of trogocytosis+ T cells following coculture with MART-126–35–pulsed T2 cells ranged from 11% for clone 2D11 to 44% for clone 2E2. Additional clones (2E6, 2G1, http://www.jimmunol.org/ and 1C8) exhibited intermediate trogocytosis, ranging from 23 to 28% (Fig. 1C). To validate the observed interclonal difference in trogocytosis against T2 cells, we used as targets unmodified 624mel melanoma cells, which naturally present MART-126–35 on HLA-A2. The most powerful clone, 2E2, preserved its superiority + (50% trogocytosis ), and the other clones were scaled in a similar FIGURE 1. FACS-based single cell sorting and generation of MART- + order to that detected with peptide-pulsed T2 targets. Only 8% of 126–35–specific CD8 T cell clones with different trogocytosis capacity. A, + trogocytosis cells were found in the 2D11 clone, whereas 13– TIL from an HLA-A2+ patient was stained with anti-CD8 mAb and by guest on October 2, 2021 30% were evident for 2E6, 2G1, and 1C8 cells (Fig. 1D). In all MART-126–35/HLA-A2 dextramers. Following single cell sorting of dex- cases, coculture of the clones with control peptide-pulsed T2 cells tramer+CD8+ T cells, the clones were generated and expanded with anti- or with irrelevant melanoma resulted in ,1.2% nonspecific tro- CD3 mAb, rhIL-2, and feeder cells. B, MART-126–35/HLA-A2 specificity gocytosis. Throughout the study, the clones underwent multiple of CD8 T cells. The clones were double stained with anti-CD8 mAb and rounds of expansion while consistently preserving their pattern of MART-126–35/HLA-A2 dextramers and monitored by a BD LSRII: MART- 1 dextramer staining of CD8-gated T cells (open curve); background trogocytosis. Therefore, each pattern represents a coherent char- 26–35 staining with irrelevant (gp100154–162/HLA-A2) dextramers (filled curve). acteristic of a given T cell population rather than an inconsistent C and D, T cell clones were incubated with surface-biotinylated, specific, trait affected by in vitro conditions. Hence, we generated a panel or irrelevant peptide-pulsed T2 cells (C) or with 624mel (A2+/MART-1+) + of CD8 T cell clones of the same TCR specificity but with dis- or M-171(A22/MART-1+) melanomas (D). Numbers indicate the per- tinct capacities to perform trogocytosis: from low (2D11), through centage of trogocytosis-performing (streptavidin+CD8+) T cells. One rep- intermediate (2G1), to high (2E2). resentative experiment of five performed is shown.

The extent of trogocytosis performed by T cell clones grades them along a functional scale these results reflect a dampened machinery of cytokine production Recently, we showed that trogocytosis reflects the strength of in 2D11 as compared with 2E2 cells, we measured IFN-g and dynamic interaction between a lymphocyte and a tumor cell and TNF-a production following stimulation with PMA/ionomycin. might be used as a functional signature of tumor-specific T cells Our results revealed comparable potential for cytokine produc- (14). To gain a deeper insight into the link between trogocytosis tion by 2D11, suggesting intact ability to produce IFN-g and TNF-a intensity and tumor-specific responsiveness of distinct T cell by this clone (Fig. 2A,2B, histograms). clones, we compared 2D11, 2G1, and 2E2 clones in several func- Next, we directly correlated trogocytosis capacity of the clones tional assays: cytokine secretion, lysosomal degranulation, and with their cytolytic activity as measured by cell surface expression cytotoxicity. of CD107A (LAMP-1), a surrogate marker of lysosomal degran- Ag-specific reactivity was initially evaluated by intracellular ulation (19). In this set of experiments, peptide-pulsed targets and staining of IFN-g and TNF-a following stimulation with 624mel melanoma cells were labeled with the lypophilic dye PKH26 and cells. The percentage of IFN-g–producing cells ranged from 20% coincubated with T cells. Trogocytosis was measured by detection for 2D11 to 54% for 2E2, with an intermediate value of 34% for of PKH26 on CD8 lymphocytes and plotted against CD107A (Fig. 2G1 (Fig. 2A). The pattern for TNF-a production was similar, 2C). In line with the results obtained with streptavidin (Fig. 1C, with a low cytokine production by 2D11 cells (13%) and a 2- and 1D), PKH26 acquisition was low in 2D11, intermediate in 2G1, 3.5-fold increase in the percentage of cytokine-producing cells for and high in 2E2. Plotting trogocytosis (PKH26) against lyso- 2G1 and 2E2 clones, respectively (Fig. 2B). To determine whether somal degranulation (CD107A) exposed the highest percentage of The Journal of Immunology 635

FIGURE 2. Intracellular cytokine expression and lysosomal degranulation by MART-126–35–specific CD8+ Tcellclones.A and B, Dot plots show the lymphocytes that were stimulated with melanoma in the presence of brefeldin A, followed by double staining for CD8 and IFN-g (A) or TNF-a (B), as de- scribed in Materials and Methods. Numbers indicate the percentage of cytokine-producing CD8+ T cells. Histograms indicate intracellular cytokine production that was detected in 2E2 (solid curve), 2D11 (broken curve), and 2G1 (dotted curve) clones following acti- vation with PMA and ionomycin; a filled histogram shows background staining with the mouse isotype

control Ab. One representative experiment of five per- Downloaded from formed is shown. C, Lysosomal degranulation was measured by concomitant CD8 and CD107A stain- ing and PKH26 acquisition following stimulation with PKH26-labeled 624mel or MART-1–pulsed T2 cells. Irrelevant peptide-pulsed T2 cells and M171 melanoma were used as nonspecific targets. Numbers in upper

right and lower right quadrants indicate the percentage http://www.jimmunol.org/ of PKH26+CD107A+ and PKH26+CD107A2 cells, re- spectively. CD8+ populations are shown. D, Expression of CD137 against CD107A on CD8+ T cells. Numbers in upper right and lower right quadrants indicate the percentages of CD8+CD137+ cells that are CD107A+ and CD107A2, respectively. One representative ex- periment of four performed is shown. by guest on October 2, 2021

PKH26+CD107A+ cells in 2E2 and lowest in 2D11, further sup- killing melanoma cells (Fig. 3B). Reduced susceptibility of tumor porting a tight association between trogocytosis and cytolytic cells to T cell attack may be due to low MHC expression or im- activity (Fig. 2C). To correlate trogocytosis with activation- paired Ag processing (21). To compare CTL clones in a scenario induced phenotypic alterations, 2E2, 2G1, and 2D11 clones were where tumor cells are less likely to escape T cell killing due to low compared with respect to CD137 expression following incuba- expression of pMHC, 624mel cells were treated with IFN-g prior tion with specific melanoma. CD137 has been described as a to their usage as targets in cytotoxicity assays. Exposure to IFN-g marker for activation of effector immune cells (20). In line with increased expression of MHC class I by 5-fold (data not shown). the data of Wolfl et al. (20), the vast majority of Ag-specific Whereas IFN-g treatment notably increased the sensitivity of T cells displayed elevated levels of CD137 after incubation with 624mel cells to 2E2 cytotoxicity, it barely affected killing by 2D11 624mel cells, with a peak at 24 h (data not shown). Notably, 2D11 (Fig. 3C). Therefore, low pMHC expression is unlikely to account and 2G1, the clones with lower effector activity, exhibited a higher for the impaired functionality of 2D11 cells, and IFN-g pretreat- percentage of CD137+ cells, compared with the 2E2 clone (Fig. ment does not restore its poor functionality. Taken together, these 2D, upper and lower right quadrants). However, the percentage of data indicate that the level of trogocytosis reliably reflects anti- CD137+CD107A+ cells was the highest for 2E2 (Fig. 2D). Thus, tumor reactivity and allows functional scaling of Ag-specific double staining for CD107A and CD137, which highlights Ag- T cells. reactive cells within the overall activated population, best corre- lates with trogocytosis levels of each clone. TCR analysis of T cell clones Finally, we compared cytotoxic activity of 2E2, 2G1 and 2D11 We next addressed the possible role of TCR binding affinity to against peptide-pulsed targets and MART-1+HLA-A2+ melanoma pMHC in the functional diversity of T cell clones. TCR interaction cells. Clone 2E2 displayed the highest lytic activity against with pMHC relies on intrinsic affinity and costabilization of the peptide-pulsed T2 cells; 2G1 clone was 10–15% less effective complex via the CD8 molecule and the a3 domain of the MHC than 2E2, and 2D11 cells exhibited only marginal CTL activity class I H chain (22). The clones were evaluated for pMHC binding (Fig. 3A). Similarly, highest anti-melanoma cytotoxicity was dis- in a CD8-dependent and -independent manner. To this end, we played by clone 2E2, whereas 2D11 was entirely ineffective at used multimers consisting of either intact or a3-mutated HLA-A2, 636 A STUDY OF T CELL FUNCTIONAL DIVERSITY BY TROGOCYTOSIS

FIGURE 4. TCR staining of MART-126–35–specific T cell clones. 2E2 (solid curve), 2G1 (dotted curve), and 2D11 (broken curve) clones were

stained with wild-type (left panel)ora3-mutated pMHC/MART-126–35 multimers (right panel). The filled histogram shows background staining

with irrelevant (gp100154–162/HLA-A2) multimers. One representative ex- periment of three performed is shown.

in 2E2 than in 2D11, as evident in nontreated cells and in cells stimulated with control peptide-pulsed T2 cells. Additionally, a distinct timeline pattern of p-S6 in 2E2 and 2D11 was observed upon activation with peptide-pulsed T2 cells, with enhanced Downloaded from phosphorylation emerging between 15 and 45 min later (Fig. 5A). Peptide stimulation did not yield detectable levels of other TCR- related protein phosphorylation (data not shown). To exclude variability in phosphorylation patterns as a result of different TCR/pMHC affinities, anti-CD3 Ab, a potent activator of T cells, was used. p-S6 showed the most distinctive patterns differentiating FIGURE 3. Cytotoxic activity of tumor Ag-specific T cell clones. CTL http://www.jimmunol.org/ 35 2E2 from 2D11. 2E2 showed an intense and protracted phos- assays were done with S-labeled–specific (MART-126–35) or irrelevant 2 phorylation of S6 commencing after 1 min, peaking at 15 min, and (HIV-derived) peptide-pulsed T2 cells (A) or with A2+ 624mel and A2 M171 melanoma cells (B) as targets, as described (50). C, Cytotoxicity of continuing at 45 min. The functionally inferior 2D11 displayed 2E2 (left panel) and 2D11 (right panel) is shown against untreated 624mel stronger initial phosphorylation of S6, followed by a weaker and (striped bars), M171 (open bars), or 624mel cells pretreated with IFN-g shorter course that ended after 45 min (Fig. 5B). Whereas minor (filled bars). Data are means 6 SE (n = 3/group) percentage of target lysis differences in phosphorylation of SLP-76 and CD3z were ob- at E:T ratios of 1:1 to 0.25:1 (A), 8:1 (B), and 30:1 to 3.25:1 (C). Statistical served between the clones, no differences in phosphorylation significance of the cytotoxic activity of T cells against 624mel cells pre- levels between the two clones were detected for LCK, PLCg, and treated with IFN-g versus untreated 624mel was determined by an un-

ERK1/2 (Fig. 5B and data not shown). These results indicate that by guest on October 2, 2021 , paired Student t test: *p 0.008, **, insignificant. One representative even when T cell clones are triggered by anti-CD3 in a pMHC- experiment of four performed is shown. independent manner, they show distinct signaling patterns, sug- gesting that factors other than TCR affinity affect the function with the latter being deficient in CD8-mediated binding (23). All level of each clone. three clones exhibited equal levels of TCR staining with wild-type Screening array of phosphorylated immunoreceptors. T cell acti- multimers. However, only 2E2 and 2G1 retained the same level of vation initiates cascades of tyrosine phosphorylation of ITAM and staining when the a3-mutated multimer was used, whereas TCR ITIM (reviewed in Ref. 24). To compare ITAM and ITIM patterns staining of 2D11 was markedly decreased, confirming lower TCR in 2D11, 2G1, and 2E2, cells were activated by anti-CD3 mAb and affinity (Fig. 4). To further characterize the TCRs, RNA was pu- screened by a phospho-immunoreceptor array. TCR triggering rified from each clone and subjected to TCR b-chain spectratyp- resulted in strong CD3z phosphorylation in all tested clones (Fig. ing. This analysis revealed Vb region 16 restriction for clones 2E2 6A). Phosphorylation of NTB-A was high in 2E2, lower in 2D11, and 2G1, and Vb14 for clone 2D11 (Table I). Taken together, the and almost undetectable in 2G1 cells (Fig. 6A). Quantitative image data suggest that the lower function of 2D11 is at least partly analysis revealed a 2- and 4-fold increase of phosphorylated NTB-A related to its low-affinity TCR. However, the different activity in 2E2 compared with 2D11 and 2G1 cells, respectively (Fig. levels of 2E2 and 2G1 are likely due to TCR affinity-independent 6B). Total NTB-A expression in unstimulated and activated T cells factors. was similar for all clones (Fig. 6C and data not shown). NTB-A is Profiling immunoreceptor phosphorylation of the T cell clones a surface receptor engaged in homophilic interactions (25). To ascribe a role for NTB-A–mediated signaling in CD8+ T cell ac- To delineate the phosphorylation pattern of proteins associated with tivation, 2E2 cells were stimulated with MART-1 –pulsed T2 signal transduction in each T cell clone, we used two comple- 26–35 cells whose NTB-A was masked by specific Ab (26). T2 cells mentary approaches: 1) flow cytometry-based longitudinal analy- sis of phosphorylated proteins downstream of the TCR; and 2) screening array of phosphorylated immunoreceptors. Table I. Characterization of MART-126–35–specific T cell clones of Flow cytometry-based analysis. To evaluate whether the functional different avidity/affinity differences between the clones are reflected in distinctive signaling patterns, phosphorylation of CD3z, SLP-76, ribosomal protein S6, TCR Trogocytosis Avidity CD8-Independent LCK, PLCg, and ERK1/2 proteins was recorded in the high- and Vb Intensity (IFN-g Curve) Tetramer Staining low-reactive 2E2 and 2D11 clones, respectively. Differences in 2E2 16 High High Yes phosphorylation patterns were mainly observed for the ribosomal 2G1 16 Medium-high High Yes protein S6. Baseline phosphorylation of S6 (p-S6) was stronger 2D11 14 Low Low No The Journal of Immunology 637

FIGURE 5. Diverse phosphorylation of proteins as- sociated with TCR signal transduction. A, Phosphoflow analysis of phosphorylated ribosomal protein S6 at different time points following 2E2 and 2D11 clone stimulation with specific (MART-126–35) or irrelevant (HIV) peptide-pulsed T2 cells. The gate was set on the CD8+CD192 population. B, Longitudinal array of phosphorylated proteins S6, SLP-76, and CD3z in anti- Downloaded from CD3 mAb-stimulated T cells. 2E2 and 2D11 clones were stimulated by soluble anti-CD3, as described in Materials and Methods, followed by phosphoflow anal- ysis at different time points. One representative ex- periment of three performed is shown. nt, not treated. http://www.jimmunol.org/ by guest on October 2, 2021

treated with isotype-matched Ab and NTB-A–negative 624mel based on cytokine secretion, phenotypic activation-induced alter- melanoma were used as controls. As shown in Fig. 7, pretreatment ations, and cytotoxic activity. We conclude that the level of tro- of T2 cells with anti–NTB-A resulted in a decreased percentage of gocytosis reliably reflects anti-tumor reactivity and allows INF-g– and TNF-a–producing cells, decreasing from 81 to 50% functional scaling of Ag-specific T cells. This is exemplified in the and from 34 to 13%, respectively. In an attempt to increase the present study by the 2E2, 2G1, and 2D11 clones of high, inter- inhibitory effect of the blocking Ab, in some experiments anti– mediate, and low activity, respectively. Importantly, the diverse NTB-A mAb was concomitantly added to target and effector cells. functional traits of these clones were stable throughout repeated However, this did not lead to additional decrease in T cell activity cycles of in vitro expansion. This observation reinforces the prog- (data not shown). Pretreatment of T2 cells with an isotype control eny functional consistency principle formulated by Beuneu et al. mAb and 624mel treated with anti–NTB-A Ab did not affect cy- (6), which states that the activation level induced by peptide- tokine production by T cells (Fig. 7). These results suggest that loaded APCs is consistently displayed by the entire progeny. specific blocking of NTB-A homophilic interactions markedly The three T cell clones made two pairs: 2E2 and 2D11, which reduces the responsiveness of CD8 T cells. differed markedly in their function and bore different TCR Vb; and 2E2 and 2G1, which expressed the same TCR Vb but differed Discussion in function (Table I). In line with the fact that TCR affinity is In this study, we used the phenomenon of trogocytosis as a sig- the primary factor dictating T cell activation threshold (29), 2E2 nature of function for clonal T cells. Trogocytosis was used to and 2D11 express TCRs of strikingly different affinities, as dem- distinguish strong effector T cells from counterparts with reduced onstrated by CD8-dependent and -independent tetramer staining activity and to define molecules that partake in the functional (Fig. 4) and dose-dependent IFN-g secretion (not shown). In con- diversification of cells with identical TCR specificity. trast, the second pair of clones, 2E2 and 2G1, varied in their A potent tool developed to advance this study was a panel of degree of anti-tumor activity, despite having the same TCR Vb MART-126–35–specific T cell clones endowed with distinct ca- region and similar TCR affinities. Several factors could contribute pacities to perform trogocytosis. The presence of membrane frag- to the functional disparity between 2E2 and 2G1, including 1) ments on T cells that have recently interacted with target cells TCR ab expression levels, 2) z-chain downmodulation in the low- primarily reflects TCR specificity, avidity, and cytotoxic activity function clone (30), and 3) changes in the ratio between CD8 (14, 27, 28). In the present study, a correlation between the extent homodimers (aa) and heterodimers (ab) (31, 32). The expression of trogocytosis and the functional status of T cells was established levels of TCR ab, z, CD8a, and CD8b were indistinguishable 638 A STUDY OF T CELL FUNCTIONAL DIVERSITY BY TROGOCYTOSIS

FIGURE 6. NTB-A (SLAMF6) expression on MART-

126–35–specific T cell clones. A, Phospho-immunore- ceptor array of multiple tyrosine phosphorylated immu- noreceptors in lysates prepared from designated T cell clones following activation with anti-CD3 mAb. B, Quantitative evaluation of phosphorylated NTB-A in the clones analyzing the array image file using image analysis software, as described in Materials and Methods. C, Expression of surface NTB-A on 2E2 (solid curve), 2G1 (broken curve), and 2D11 (dotted curve) cells following double staining with anti-CD8 and anti– NTB-A mAbs and flow cytometry on a BD LSRII. Filled histogram shows background staining with isotype con- trol Ab. PC, positive control. Downloaded from

between the two clones and therefore could not account for their NTB-A results in its phosphorylation and association with the small http://www.jimmunol.org/ functional diversity (data not shown). Experiments searching for adaptor molecules SAP and EAT2, as well as activation of Fyn and other factors that could affect T cell function are currently un- downstream transcription through NF-kB (43). NTB-A engage- derway. ment in activated T cells leads to T cell proliferation, Tc1 cytokine Trogocytosis takes place at the immunological synapse, the focal secretion, and enhanced lytic function (25). Our data show that point for signal initiation and integration that defines the level of NTB-A blockade on APCs inhibits TNF-a and IFN-g production, T cell activation (33). The extent of trogocytosis could re- emphasizing the need for NTB-A costimulation (Fig. 7). Sur- flect changes in signaling pathways not necessarily related to prisingly, NTB-A blockade had a stronger impact on TNF-a than pMHC/TCR affinity. In addressing this issue, we characterized by guest on October 2, 2021 the signaling pattern of clones 2E2 and 2D11 in a TCR/pMHC- dependent and –independent manner, using peptide-pulsed APCs and anti-CD3 mAbs as stimulators. Of the signals detected, phos- phorylation of S6 ribosomal protein showed the most distinct difference between the clones upon peptide and anti-CD3 stimu- lation. 2E2, the most potent clone, endured longer phosphoryla- tion, whereas 2D11 displayed only a brief signal. S6 is phosphor- ylated downstream of the AKT/mTOR pathway and integrates multiple signals, including those of the MAPK pathway (13, 14). Sustained activation of T cells, regardless of the initiating trigger, is associated with S6 kinase activity (34–36). The higher phos- phorylation state of S6 in clone 2E2 is in accord with these data. Thus, S6 phosphorylation is a good indicator of the overall downstream signal propagation in activated T cells. The data suggest that different signaling properties of the low and high effector clones, regardless of TCR affinity, account for functional diversity. TCR/CD3 signaling is augmented and sustained by a series of ligand–receptor interactions of costimulatory molecules on the surface of T cells and APCs (37). Screening by the phospho- FIGURE 7. Interfering with the NTB-A/NTB-A homophilic interaction immunoreceptor array technology revealed augmented phosphor- decreases cytokine production by CD8+ T cells. 2E2 T cells were stimu- + lated with MART-126–35–pulsed T2 cells (NTB-A ) or 624mel melanoma ylation patterns of NTB-A in clone 2E2 as compared with 2G1 2 and 2D11. Surprisingly, in the two clones with identical TCR (NTB-A ), pretreated with either anti–NTB-A or with matched isotype affinities, 2E2 and 2G1, the differences in NTB-A phosphorylation control mAb. Following 6 h stimulation in the presence of brefeldin A, the cells were double stained for CD8 and intracellular IFN-g (A) or TNF-a were particularly prominent. NTB-A (SLAMF6) is a surface re- (B) and subjected to flow cytometry on a BD LSRII. Columns show mean ceptor of the SLAM family, which belongs to the CD2 super- of three replicates for IFN-g+CD8+ (A) or TNF-a+CD8+ (B) cells; bars family, and is expressed on NK, T, and B lymphocytes (38–41). indicate SE. Statistical significance of the effect of anti–NTB-A versus This receptor acts through homophilic interactions with other NTB-A isotype control on T cell function was determined by an unpaired Student t molecules, providing a costimulatory signal for lymphocyte activa- test. *p , 0.001, **p , 0.0001. One representative experiment of three tion, with a focus on innate-like effectors (41, 42). Cross-linking of performed is shown. The Journal of Immunology 639 on IFN-g production. Although the reason for this finding is un- 12. Daubeuf, S., M. A. Lindorfer, R. P. Taylor, E. Joly, and D. Hudrisier. 2010. The direction of plasma membrane exchange between lymphocytes and accessory clear, the impact of NTB-A on production of different cytokines cells by trogocytosis is influenced by the nature of the accessory cell. J. was shown to be variable (44, 45). Immunol. 184: 1897–1908. To the best of our knowledge, this report is the first evidence for 13. Daubeuf, S., A. L. Puaux, E. Joly, and D. Hudrisier. 2006. A simple trogocytosis- based method to detect, quantify, characterize and purify antigen-specific live the importance of NTB-A as a costimulator in tumor Ag-specific lymphocytes by flow cytometry, via their capture of membrane fragments from CD8 T cells. Further studies are being conducted to explore the antigen-presenting cells. Nat. Protoc. 1: 2536–2542. relevance of this molecule for anti-tumor response through direct 14. Machlenkin, A., R. Uzana, S. Frankenburg, G. Eisenberg, L. Eisenbach, J. Pitcovski, R. Gorodetsky, A. Nissan, T. Peretz, and M. Lotem. 2008. Capture engagement of NTB-A on tumor-specific T cells. Several co- of tumor cell membranes by trogocytosis facilitates detection and isolation of stimulatory molecules, including CD28, ICOS, OX40, and 4-1BB, tumor-specific functional CTLs. Cancer Res. 68: 2006–2013. are recognized. Although CD28 is considered to be the most potent 15. Lotem, M., A. Machlenkin, T. Hamburger, A. Nissan, L. Kadouri, S. Frankenburg, Z. Gimmon, O. Elias, I. B. David, A. Kuznetz, et al. 2009. Autologous melanoma costimulatory pathway (46–48), its targeting with an agonistic Ab vaccine induces antitumor and self-reactive immune responses that affect patient in a clinical trial tragically ended up with a fatal cytokine storm survival and depend on MHC class II expression on vaccine cells. Clin. Cancer (49). The addition of NTB-A to the short list of costimulators Res. 15: 4968–4977. 16. Dudley, M. E., J. R. Wunderlich, T. E. Shelton, J. Even, and S. A. Rosenberg. contributes a new candidate to be evaluated for anti-cancer im- 2003. Generation of tumor-infiltrating lymphocyte cultures for use in adoptive munotherapy. transfer therapy for melanoma patients. J. Immunother. 26: 332–342. 17. Machlenkin, A., A. Paz, E. Bar Haim, O. Goldberger, E. Finkel, B. Tirosh, Understanding the different mechanisms that account for T cell I. Volovitz, E. Vadai, G. Lugassy, S. Cytron, et al. 2005. Human CTL epitopes diversification has a practical side: one of the main obstacles in prostatic acid phosphatase-3 and six-transmembrane epithelial antigen of prostate- cancer immunotherapy is the low affinity and impaired function of 3 as candidates for prostate cancer immunotherapy. Cancer Res. 65: 6435–6442. 18. Akatsuka, Y., E. G. Martin, A. Madonik, A. A. Barsoukov, and J. A. Hansen. 1999. anti-tumor T cells. Even when a substantial number of T cells can Rapid screening of T-cell receptor (TCR) variable gene usage by multiplex PCR: Downloaded from be generated and transferred to the patient, cancer regression has application for assessment of clonal composition. Tissue Antigens 53: 122–134. not been guaranteed. Although the lack of specificity could be 19. Rubio, V., T. B. Stuge, N. Singh, M. R. Betts, J. S. Weber, M. Roederer, and P. P. Lee. 2003. Ex vivo identification, isolation and analysis of tumor-cytolytic resolved with the advent of TCR engineering, the low reactivity and T cells. Nat. Med. 9: 1377–1382. short survival of T cells still restricts the clinical benefit of cancer 20. Wolfl, M., J. Kuball, W. Y. Ho, H. Nguyen, T. J. Manley, M. Bleakley, and immunotherapy. Trogocytosis provides new opportunities to reveal P. D. Greenberg. 2007. Activation-induced expression of CD137 permits detection, isolation, and expansion of the full repertoire of CD8+ T cells responding to an- and evaluate molecules that could improve this situation. tigen without requiring knowledge of epitope specificities. Blood 110: 201–210. 21. Qin, Z., C. Harders, X. Cao, C. Huber, T. Blankenstein, and B. Seliger. 2002. http://www.jimmunol.org/ Increased tumorigenicity, but unchanged immunogenicity, of transporter for Acknowledgments antigen presentation 1-deficient tumors. Cancer Res. 62: 2856–2860. We thank Prof. Ronald Levy and Dr. Jonathan Irish from the Division of 22. Garcia, K. C., C. A. Scott, A. Brunmark, F. R. Carbone, P. A. Peterson, Oncology, School of Medicine, Stanford University, for discussions on I. A. Wilson, and L. Teyton. 1996. CD8 enhances formation of stable T-cell receptor/MHC class I molecule complexes. Nature 384: 577–581. phospho-signaling analysis. We appreciate the help of Dr. Eyal Zcharia from 23. Bodinier, M., M. A. Peyrat, C. Tournay, F. Davodeau, F. Romagne, the Sharett Institute of Oncology, Hadassah Medical Organization, for the M. Bonneville, and F. Lang. 2000. Efficient detection and immunomagnetic molecular analysis of TCR. We thank Inna Ben David, Anna Kuznetz, and sorting of specific T cells using multimers of MHC class I and peptide with Yael Gelfand for skilled technical work. reduced CD8 binding. Nat. Med. 6: 707–710. 24. Billadeau, D. D., and P. J. Leibson. 2002. ITAMs versus ITIMs: striking a bal- ance during cell regulation. J. Clin. Invest. 109: 161–168. Disclosures 25. Flaig, R. M., S. Stark, and C. Watzl. 2004. Cutting edge: NTB-A activates NK by guest on October 2, 2021 cells via homophilic interaction. J. Immunol. 172: 6524–6527. The authors have no financial conflicts of interest. 26. Ward, J., M. Bonaparte, J. Sacks, J. Guterman, M. Fogli, D. Mavilio, and E. Barker. 2007. HIV modulates the expression of ligands important in triggering natural killer cell cytotoxic responses on infected primary T-cell blasts. Blood References 110: 1207–1214. 27. Riond, J., J. Elhmouzi, D. Hudrisier, and J. E. Gairin. 2007. Capture of mem- 1. Johnson, L. A., B. Heemskerk, D. J. Powell, Jr., C. J. Cohen, R. A. Morgan, brane components via trogocytosis occurs in vivo during both dendritic cells and M. E. Dudley, P. F. Robbins, and S. A. Rosenberg. 2006. Gene transfer of tumor- + target cells encounter by CD8 T cells. Scand. J. Immunol. 66: 441–450. reactive TCR confers both high avidity and tumor reactivity to nonreactive pe- 28. Puaux, A. L., J. Campanaud, A. Salles, X. Pre´ville, B. Timmerman, E. Joly, and ripheral blood mononuclear cells and tumor-infiltrating lymphocytes. J. Immu- D. Hudrisier. 2006. A very rapid and simple assay based on trogocytosis to detect nol. 177: 6548–6559. and measure specific T and B cell reactivity by flow cytometry. Eur. J. Immunol. 2. Zeh, H. J., III, D. Perry-Lalley, M. E. Dudley, S. A. Rosenberg, and J. C. Yang. 36: 779–788. 1999. High avidity CTLs for two self-antigens demonstrate superior in vitro and 29. Kedzierska, K., N. L. La Gruta, M. P. Davenport, S. J. Turner, and P. C. Doherty. in vivo antitumor efficacy. J. Immunol. 162: 989–994. 2005. Contribution of T cell receptor affinity to overall avidity for virus-specific 3. Wherry, E. J., S. J. Ha, S. M. Kaech, W. N. Haining, S. Sarkar, V. Kalia, CD8+ T cell responses. Proc. Natl. Acad. Sci. USA 102: 11432–11437. S. Subramaniam, J. N. Blattman, D. L. Barber, and R. Ahmed. 2007. Molecular + 30. Roberts, J. L., J. P. Lauritsen, M. Cooney, R. E. Parrott, E. O. Sajaroff, signature of CD8 T cell exhaustion during chronic viral infection. Immunity 27: C. M. Win, M. D. Keller, J. H. Carpenter, J. Carabana, M. S. Krangel, et al. 2007. 670–684. T-B+NK+ severe combined immunodeficiency caused by complete deficiency of 4. Nembrini, C., B. Abel, M. Kopf, and B. J. Marsland. 2006. Strong TCR sig- the CD3z subunit of the T-cell antigen receptor complex. Blood 109: 3198–3206. naling, TLR ligands, and cytokine redundancies ensure robust development of 31. Pang, D. J., A. C. Hayday, and M. J. Bijlmakers. 2007. CD8 raft localization is type 1 effector T cells. J. Immunol. 176: 7180–7188. induced by its assembly into CD8ab heterodimers, not CD8aa homodimers. J. + 5. Dolfi, D. V., and P. D. Katsikis. 2007. CD28 and CD27 costimulation of CD8 Biol. Chem. 282: 13884–13894. T cells: a story of survival. Adv. Exp. Med. Biol. 590: 149–170. 32. Laugel, B., H. A. van den Berg, E. Gostick, D. K. Cole, L. Wooldridge, 6. Beuneu, H., F. Lemaıˆtre, J. Deguine, H. D. Moreau, I. Bouvier, Z. Garcia, J. Boulter, A. Milicic, D. A. Price, and A. K. Sewell. 2007. Different T cell M. L. Albert, and P. Bousso. 2010. Visualizing the functional diversification of receptor affinity thresholds and CD8 coreceptor dependence govern cytotoxic + CD8 T cell responses in lymph nodes. Immunity 33: 412–423. T lymphocyte activation and tetramer binding properties. J. Biol. Chem. 282: 7. Alexander-Miller, M. A., G. R. Leggatt, and J. A. Berzofsky. 1996. Selective 23799–23810. expansion of high- or low-avidity cytotoxic T lymphocytes and efficacy for 33. Hudrisier, D., A. Aucher, A. L. Puaux, C. Bordier, and E. Joly. 2007. Capture of adoptive immunotherapy. Proc. Natl. Acad. Sci. USA 93: 4102–4107. target cell membrane components via trogocytosis is triggered by a selected set 8. Monks, C. R., B. A. Freiberg, H. Kupfer, N. Sciaky, and A. Kupfer. 1998. Three- of surface molecules on T or B cells. J. Immunol. 178: 3637–3647. dimensional segregation of supramolecular activation clusters in T cells. Nature 34. Arechiga, A. F., B. D. Bell, S. Leverrier, B. M. Weist, M. Porter, Z. Wu, 395: 82–86. Y. Kanno, S. J. Ramos, S. T. Ong, R. Siegel, and C. M. Walsh. 2007. A Fas- 9. Bromley, S. K., W. R. Burack, K. G. Johnson, K. Somersalo, T. N. Sims, associated death domain protein/caspase-8-signaling axis promotes S-phase C. Sumen, M. M. Davis, A. S. Shaw, P. M. Allen, and M. L. Dustin. 2001. The entry and maintains S6 kinase activity in T cells responding to IL-2. J. Immu- immunological synapse. Annu. Rev. Immunol. 19: 375–396. nol. 179: 5291–5300. 10. Joly, E., and D. Hudrisier. 2003. What is trogocytosis and what is its purpose? 35. Lockyer, H. M., E. Tran, and B. H. Nelson. 2007. STAT5 is essential for Akt/ Nat. Immunol. 4: 815. p70S6 kinase activity during IL-2-induced lymphocyte proliferation. J. Immunol. 11. Aucher, A., E. Magdeleine, E. Joly, and D. Hudrisier. 2008. Capture of plasma 179: 5301–5308. membrane fragments from target cells by trogocytosis requires signaling in 36. Xie, C., R. Patel, T. Wu, J. Zhu, T. Henry, M. Bhaskarabhatla, R. Samudrala, T cells but not in B cells. Blood 111: 5621–5628. K. Tus, Y. Gong, H. Zhou, et al. 2007. PI3K/AKT/mTOR hypersignaling in 640 A STUDY OF T CELL FUNCTIONAL DIVERSITY BY TROGOCYTOSIS

autoimmune lymphoproliferative disease engendered by the epistatic interplay of the SLAM family receptor Ly108 controls T cell and neutrophil functions. J. Sle1b and FAS lpr. Int. Immunol. 19: 509–522. Immunol. 174: 5931–5935. 37. Jenkins, M. R., A. Tsun, J. C. Stinchcombe, and G. M. Griffiths. 2009. The 45. Chatterjee, M., K. Kis-Toth, T. H. Thai, C. Terhorst, and G. C. Tsokos. 2011. strength of T cell receptor signal controls the polarization of cytotoxic machinery SLAMF6-driven co-stimulation of human peripheral T cells is defective in SLE to the immunological synapse. Immunity 31: 621–631. T cells. Autoimmunity 44: 211–218. 38. Veillette, A. 2006. Immune regulation by SLAM family receptors and SAP- 46. Greenwald, R. J., G. J. Freeman, and A. H. Sharpe. 2005. The B7 family related adaptors. Nat. Rev. Immunol. 6: 56–66. revisited. Annu. Rev. Immunol. 23: 515–548. 39. Veillette, A., and S. Latour. 2003. The SLAM family of immune-cell receptors. 47.Laux,I.,A.Khoshnan,C.Tindell,D.Bae,X.Zhu,C.H.June,R.B.Effros,and Curr. Opin. Immunol. 15: 277–285. A. Nel. 2000. Response differences between human CD4+ and CD8+ T-cells during 40. Green, J. M., V. Karpitskiy, S. L. Kimzey, and A. S. Shaw. 2000. Coordinate CD28 costimulation: implications for immune cell-based therapies and studies related regulation of T cell activation by CD2 and CD28. J. Immunol. 164: 3591–3595. to the expansion of double-positive T-cells during aging. Clin. Immunol. 96: 187–197. 41. Valdez, P. A., H. Wang, D. Seshasayee, M. van Lookeren Campagne, A. Gurney, 48. Zhang, H., K. M. Snyder, M. M. Suhoski, M. V. Maus, V. Kapoor, C. H. June, and W. P. Lee, and I. S. Grewal. 2004. NTB-A, a new activating receptor in T cells C. L. Mackall. 2007. 4-1BB is superior to CD28 costimulation for generating CD8+ that regulates autoimmune disease. J. Biol. Chem. 279: 18662–18669. cytotoxic lymphocytes for adoptive immunotherapy. J. Immunol. 179: 4910–4918. 42. Veillette, A., Z. Dong, and S. Latour. 2007. Consequence of the SLAM-SAP sig- 49. Suntharalingam, G., M. R. Perry, S. Ward, S. J. Brett, A. Castello-Cortes, naling pathway in innate-like and conventional lymphocytes. Immunity 27: 698–710. M. D. Brunner, and N. Panoskaltsis. 2006. Cytokine storm in a phase 1 trial of the 43. Chan, B., A. Lanyi, H. K. Song, J. Griesbach, M. Simarro-Grande, F. Poy, anti-CD28 TGN1412. N. Engl. J. Med. 355: 1018–1028. D. Howie, J. Sumegi, C. Terhorst, and M. J. Eck. 2003. SAP couples Fyn to 50. Machlenkin, A., O. Goldberger, B. Tirosh, A. Paz, I. Volovitz, E. Bar-Haim, SLAM immune receptors. Nat. Cell Biol. 5: 155–160. S. H. Lee, E. Vadai, E. Tzehoval, and L. Eisenbach. 2005. Combined den- 44. Howie, D., F. S. Laroux, M. Morra, A. R. Satoskar, L. E. Rosas, W. A. Faubion, dritic cell cryotherapy of tumor induces systemic antimetastatic immunity. Clin. A. Julien, S. Rietdijk, A. J. Coyle, C. Fraser, and C. Terhorst. 2005. Cutting edge: Cancer Res. 11: 4955–4961. Downloaded from http://www.jimmunol.org/ by guest on October 2, 2021