Ligand Binding but Undetected Functional Response of FcR after Their Capture by T Cells via Trogocytosis

This information is current as Denis Hudrisier, Béatrice Clemenceau, Stéphanie Balor, of September 28, 2021. Sandrine Daubeuf, Eddy Magdeleine, Marc Daëron, Pierre Bruhns and Henri Vié J Immunol 2009; 183:6102-6113; Prepublished online 19 October 2009;

doi: 10.4049/jimmunol.0900821 Downloaded from http://www.jimmunol.org/content/183/10/6102

References This article cites 53 articles, 32 of which you can access for free at: http://www.jimmunol.org/content/183/10/6102.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists by guest on September 28, 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 © 2009 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Ligand Binding but Undetected Functional Response of FcR after Their Capture by T Cells via Trogocytosis1

Denis Hudrisier,2*† Be´atrice Clemenceau,‡ Ste´phanie Balor,† Sandrine Daubeuf,*† Eddy Magdeleine,*† Marc Dae¨ron,§¶ Pierre Bruhns,§¶ and Henri Vie´‡

Intercellular transfer of surface proteins by trogocytosis is common and could affect responses. Yet, the role of trogocytosis in T cell function is still elusive, and it is unknown whether a molecule, once captured by T cells, harbors the same biological properties as in donor APC. In this study, we showed that Fc␥R as well as the associated FcR␥ subunit could be detected at high levels on murine and human T cells after their intercellular transfer from Fc␥R-expressing APC. Capture of Fc␥R occurred during coculture of T cells with Fc␥R-expressing APC upon Ab- or Ag-mediated T cell stimulation. Once captured by T cells, Fc␥R were expressed in a conformation compatible with physiological function and conferred upon T cells the ability to bind immune complexes and to provision B cells with this source of Ag. However, we were unable to detect downstream signal or Downloaded from signaling-dependent function following the stimulation of Fc␥R captured by T cells, and biochemical studies suggested the im- proper integration of Fc␥R in the recipient T . Thus, our study demonstrates that T cells capture Fc␥R that can efficiently exert ligand-binding activity, which, per se, could have functional consequences in T cell- cooperation. The Journal of Immunology, 2009, 183: 6102–6113.

ecognition by T cells of Ag at the surface of APC is a The transfer of receptors by trogocytosis has been reported (8, 9), http://www.jimmunol.org/ critical event for subsequent T cell activation and acqui- but it is unclear whether they could transduce biochemical signals R sition of effector functions. A well-recognized event as- after transfer. This question is critical because it could greatly af- sociated with Ag recognition is the cointernalization of the TCR fect the functional outcome of receptor capture through trogocy- together with its Ag, which is formed by the peptide-MHC com- tosis by cell subsets that intrinsically do not express these given plex present at the membrane of APC (1–3). Membrane-bound receptors. peptide-MHC complexes are captured by T cells via membrane The uptake by T cells of FcR released from macrophages was fragments containing both and a large panel of membrane- among the first descriptions of intercellular transfer of surface mol- bound proteins in a process termed trogocytosis (4, 5). Conse- ecules (10). However, no mechanism was reported for this transfer, quently, T cells harbor on their surface not only peptide-MHC and the type of Fc␥R taken up by T cells as well as their potential by guest on September 28, 2021 complexes, but also a series of molecules that they do not synthe- function after uptake have yet to be defined. Fc␥R play important size themselves. Many different roles have been hypothesized for roles in immune responses by coupling innate and adaptive im- trogocytosis in T cell biology, but their physiological relevance munity through their ability to bind Ab-Ag complexes and Ab- and underlying mechanisms have not been demonstrated (1, 5–7). opsonized target cells, and thus, their role in and Some proposed functions of trogocytosis solely require that mol- Ab-dependent cell cytotoxicity (ADCC).3 The function of Fc␥R ecules captured by T cells interact with their ligand on the same or depends on their type (activatory or inhibitory, FcR␥ subunit de- different cells without the need for intracellular signaling events. pendent or independent, high or low affinity) and on the cells ex- For instance, capture by T cells of the peptide-MHC complex and pressing the Fc␥R (11, 12). of costimulatory molecules, followed by interaction with the TCR In this study, we thus examined the mechanisms of Fc␥R cap- and CD28 on other T cells, results in T cell-T cell interactions ture by T cells through trogocytosis and documented the capture of leading to activation, tolerance, or exhaustion (1). In contrast, it is various types of Fc␥RbyCD4ϩ and CD8ϩ murine and human T expected that a proper integration of the captured molecule would cells. Furthermore, we explored whether, once captured by T cells, be required for downstream intracellular signals to be triggered (5). Fc␥R could act as a receptor both in terms of ligand-binding and functional (signaling-based) properties.

*Centre National de la Recherche Scientifique, Institut de Pharmacologie et de Bi- ologie Structurale, Toulouse, France; †Universite´de Toulouse, UPS, Institut de Phar- Materials and Methods macologie et de Biologie Structurale, Toulouse, France; ‡INSERM, Unite´ 601, Cell lines and mice Nantes, France; §Institut Pasteur, De´partement d’Immunologie, Unite´d’Allergologie Mole´culaire et Cellulaire, Paris, France; and ¶INSERM, Unite´760, Paris, France The plasmacytoma cell line P815 or HEK 293 cells and their stable trans- ␥ Received for publication March 13, 2009. Accepted for publication September fectants expressing Fc R were used as target cells. Dendritic cells (DC) 7, 2009. generated from B6 bone marrow (BM-DC) (13) or peritoneal B6 macro- phages were used as APC. The OT-I CD8ϩ T cells specific for OVA The costs of publication of this article were defrayed in part by the payment of page peptide 257–264 (SIINFEKL) presented by the H-2Kb MHC class I mol- charges. This article must therefore be hereby marked advertisement in accordance ϩ with 18 U.S.C. Section 1734 solely to indicate this fact. ecules and the OT-II CD4 T cells specific for the OVA peptide 322–332 1 This work was supported by the Region Midi-Pyre´ne´es and the Agence Nationale pour la Recherche. 3 ␤ ␤ Abbreviations used in this paper: ADCC, Ab-dependent cell cytotoxicity; 2m, 2- 2 Address correspondence and reprint requests to Dr. Denis Hudrisier, Institut de microglobulin; BM-DC, dendritic cells generated from B6 bone marrow; DC, den- Pharmacologie et de Biologie Structurale, Centre National de la Recherche Scienti- dritic cell; HEL, hen egg lysozyme; IC, immune complex; mFc␥, murine Fc␥. fique Unite´ Mixte de Recherche 5089, 205 route de Narbonne, 31077 Toulouse. E-mail address: [email protected] Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00 www.jimmunol.org/cgi/doi/10.4049/jimmunol.0900821 The Journal of Immunology 6103 Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 1. Murine CD8ϩ and CD4ϩ T cells capture Fc␥R from P815 cells during redirected trogocytosis. A, P815 cells previously stained with the fluorescent lipophilic probe DiO were incubated with CD8ϩ OT-I cells coated with the indicated mAb. At the end of the 1-h coculture, cells were labeled with anti-CD8 mAb, and the capture of DiO (left panels)orFc␥R(right panels) was analyzed by flow cytometry. FI is the fold induction of capture calculated as the ratio of the median fluorescence intensity of Fc␥R(right panels) or DiO (left panels) staining on gated CD8ϩ T cells measured in the presence or absence of the mAb directed against the indicated molecules. Note that quadrants were placed such that median fluorescence intensities can be read on T cells (upper left quadrant) and P815 cells (lower right quadrant), regardless of the experimental conditions. Numbers represent the percentage of cells in each quadrant. Note that we used a narrow forward/side light scatter window to gate out cell conjugates. B,AsinA, except that the intensity of the capture of the lipophilic probe (membrane capture) and of Fc␥R is plotted for every mAb used in trogocytosis experiments. Anti-CD18 does not trigger trogocytosis, and therefore, constitutes a negative control (21). C,AsinA, except that CD4ϩ OT-II T cells were used instead of OT-I T cells, and that only the capture of Fc␥R is shown. The experiment shown is representative of three independent experiments. presented by the MHC class II I-Ab were obtained from the corresponding or BCR ␬-chain (187.1), and mAb against human CD4 (RPA-T4), CD8 TCR-transgenic mice (Charles River Laboratories) (14, 15). Total spleno- (RPA-T8), Fc␥RIIIA (3G8), and streptavidin were from BD Pharmingen. cytes from TCR-transgenic animals were cultured in the presence of 0.1 Unlabeled mAb to CD2 (RM2-2), CD3␧ (2C11), CD4 (RM4-4 or RM4-5), ␮ ␤ b ␤ M appropriate antigenic peptide and used in trogocytosis or functional CD8 (H35), CD27 (LG3A10), CD71 (C2F2), H-2K (Y3), or 2-micro- ␤ assays 4–6 days after stimulation. B cells specific for HEL were from globulin ( 2m; S-19) were from BD Biosciences. Abs to GFP were from BCR-transgenic MD4 mice (H-2d background) (16). The 1H11-34 cell line Abcam, and anti-rabbit IgG coupled to Alexa647 were from Invitrogen/ is a T cell hybridoma specific for HEL107–116 presented by I-Ed (17). Ac- Molecular Probes. Anti-human CD3 (OKT-3) and MHC class I (W6/32) tivated human T cells transduced with the retroviral vector encoding hu- were obtained from culture supernatant of the corresponding hybridoma. man Fc␥RIIIA-FcR␥ chimera were generated, as described previously was from Roche. Unlabeled rabbit polyclonal Ab to the FcR␥ (18). subunit was from Upstate Biotechnology/Euromedex. Construction of a vector encoding FcR␥GFP was done, as described previously (19). Vectors Reagents, Abs, and molecular biology encoding murine flagged Fc␥RIIB or Fc␥RIIIA and the FcR␥ subunit have Ͼ been described previously (20). Retroviral vector encoding the human Peptides were synthesized in our laboratory and HPLC purified ( 98%), Fc␥RIIIA-FcR␥ chimera was previously reported (18). and their identity was confirmed by mass spectrometry. The fluorescent DiO, Indo-1, hen egg lysozyme (HEL), anti-Flag, and anti-OVA Abs Trogocytosis were from Sigma-Aldrich. HEL-specific mAb F9 and F10 and their Fab were a gift of J.-C. Gue´ry (INSERM U563, Toulouse, France). Fluores- Trogocytosis was performed, as previously described (21). Target cells cently labeled mAb against mouse CD8␣ (53.6.7.2), CD4 (GK1.5, RM4-4 labeled or not with the lipophilic probe DiO (22) were incubated in U- or RM4-5), Fc␥RIIB/Fc␥RIIIA (2.4G2), TNF (TNF-␣), CD69 (H1.2F3), bottom 96-well plates (0.5 ϫ 106 cells/well in 100 ␮l final volume) with T 6104 FUNCTIONALITY OF Fc␥R CAPTURED BY T CELLS Downloaded from http://www.jimmunol.org/

FIGURE 2. Both Fc␥RIIB and Fc␥RIIIA trigger trogocytosis and are captured by OT-I and OT-II T cells. HEK cells transfected with a vector encoding murine flagged Fc␥RIIB or Fc␥RIIIA plus the FcR␥ subunit or left untransfected (mock) were incubated with OT-I T cells coated or not with the anti-H-2Kb mAb. A, Cells were analyzed by flow cytometry using anti-CD8 mAb and a biotinylated mAb against Flag, followed by fluorescent streptavidin. Histograms show Fc␥R expression on gated OT-I T cells exposed to the indicated target cells in the absence (gray histograms) or presence (white histograms) of the anti-H-2Kb mAb. B,AsinA, except that flow cytometry was performed using anti-CD8 and anti-Fc␥RIIB/RIIIA mAbs. C, HEK cells transfected with a vector encoding murine flagged Fc␥RIIB or Fc␥RIIIA plus the FcR␥ subunit were incubated with OT-II T cells coated or not with the indicated mAb. Cells were analyzed by flow cytometry using anti-CD4 mAb and anti-Fc␥RIIB/RIIIA mAbs. Numbers represent the median fluorescence intensity of anti- by guest on September 28, 2021 Fc␥RIIB/IIIA staining on gated CD4ϩ T cells. Note that HEK cells that were mock transfected do not allow triggering of trogocytosis by OT-I or OT-II T cells (data not shown). cells (0.2 ϫ 106 cells/well in 100 ␮l final volume) for1hat37°C. For Indo-1 (2 mM, 1 ϫ 106 cells/ml) at 37°C for 1 h. After washing (three redirected trogocytosis experiments, T cells were previously incubated times) and staining with anti-CD4 or anti-CD8 mAb, T cells were resus- with 5 ␮g/ml different unlabeled mAbs for 30 min at 4°C. For Ag-depen- pended at 1 ϫ 106 cells/ml and the calcium-dependent fluorescence of dent trogocytosis experiments, the corresponding peptide Ag (final, satu- Indo-1 in T cells was assessed by flow cytometry on a LSRII cytofluorom- rating concentration ϭ 1 ␮M) was incubated with target cells or target cells eter (BD Biosciences). The total recording time was 10 min. At different b ␤ were transfected with a construct encoding the OVA-H-2K - 2m complex time points, IC or PMA/ionomycin was added. (23). T cells were then analyzed by flow cytometry on a FACSCalibur or Intracellular cytokine staining was performed, as described previously LSRII (BD Biosciences) using anti-CD8 or anti-CD4 mAb. When trans- (21). OT-I T cells (0.5 ϫ 106 cells in 200 ␮l) separated from target cells fected HEK cells were used as target cells, T cells coated or not with a mAb at the end of trogocytosis experiments were incubated with IC or the OVA (5 ϫ 105 cells) were incubated with adherent HEK-transfected cells (2.5 ϫ peptide SIINFEKL for2hat37°C in U-bottom 96-well plates. Brefeldin 106 cells/well of a 6-well plate) during the time of trogocytosis experiment, A was then added, and incubation was extended for additional 2 h. Cells and T cells were harvested before flow cytometry analysis. Staining with were stained with anti-CD8 mAb, fixed with 2% p-formaldehyde, perme- anti-Fc␥RIIB/IIIA or anti-Flag mAb was used to detect trogocytosis. In abilized with a permeabilization buffer (PBS, 0.5% saponin, 1% BSA), and assays in which target cells were labeled with DiO, trogocytosis was mea- labeled with anti-TNF-␣ mAb. Cells were then analyzed by flow sured directly by the detection of DiO on the effectors. In some cases, cells cytometry. were treated for 1 min at room temperature with citrate/phosphate buffer (pH 3.0), as previously described (24, 25), before staining for flow cytom- ADCC assays etry analysis. C2F2 anti-CD71 mAb were added at the end of the coculture between P815 and T cells, where indicated. Raji cells were labeled with 100 ␮Ci (3.7 MBq) of 51Cr (PerkinElmer) for 1 h at 37°C and washed three times with culture medium. Cells were Binding of immune complexes incubated with the humanized anti-CD20 mAb rituximab (5 ␮g/ml) or with medium as control, and 5000 cells were placed in a 96-well U-bottom Cells were incubated for1hat4°Cwith immune complexes (IC) previ- plate. Effector T cells (previously incubated with the indicated Fc␥R-ex- ously formed by incubating for 30 min at 37°C an equimolar mixture of pressing target during trogocytosis experiments) were added at the indi- HEL-specific F9 and F10 mAbs (murine IgG1, k) with 10-fold lower molar cated E:T ratio to Raji cells (final volume: 200 ␮l) for a 5-h incubation equivalent of biotinylated HEL. Binding of IC was analyzed by flow cy- period at 37°C. A total of 100 ␮l of the supernatant was then harvested and tometry (FACSCalibur; BD Biosciences) after staining with anti-CD8 or counted in a gamma counter (Packard Instrument). anti-CD4 mAb and fluorescent streptavidin. Measurement of intracellular Ca2ϩ flux T cell-B cell cooperation assays -To assess intracellular calcium (Ca2؉ mobilization), T cells were separated OT-I T cells (having captured or not Fc␥RIIB or Fc␥RIIIA through pre from target cells at the end of trogocytosis experiments and incubated with vious incubation with HEK-murine Fc␥ (mFc␥)RIIB or HEK-mFc␥RIIIA The Journal of Immunology 6105

FIGURE 3. The FcR␥ subunit is captured and displayed by CD4ϩ and CD8ϩ T cells in a correct orientation and topology. A, HEK cells trans- fected with a vector encoding the FcR␥GFP subunit, flagged Fc␥RIIB plus FcR␥GFP, or flagged Fc␥RIIIA plus FcR␥GFP or left untransfected (mock) were incubated with OT-I T cells coated or not with the anti-H-2Kb mAb. Cells were analyzed by flow cy- tometry using anti-CD8 mAb and de- tection of the GFP fluorescence. Histo- grams show GFP fluorescence on gated CD8ϩ OT-I T cells exposed to target cells in the absence (gray histograms) or presence (white histograms) of the anti-H-2Kb mAb. The experiment shown is representative of three inde- pendent experiments. B, OT-II T cells Downloaded from coated with the indicated mAb were in- cubated with P815 cells for 1 h. At the end of the incubation period, cells were labeled with anti-CD4 mAb, and then fixed and permeabilized (eight dot plots presented at the bottom of the figure; total staining) or fixed only (eight dot http://www.jimmunol.org/ plots at the top of the figure; surface staining) before incubation in the ab- sence or presence of a primary rabbit antiserum directed against the FcR␥ subunit, followed by staining with flu- orescent anti-rabbit Abs. Similar results were obtained in a second independent experiment. Numbers represent the per-

centage of cells in each quadrant. by guest on September 28, 2021

b ␤ ϩ cells coexpressing or not the construct encoding H-2K -OVA- 2m, as de- CD4 T cells (Fig. 1C), and a similar correlation between the scribed above) were loaded and then washed with IC formed with anti- capture of a lipophilic dye and that of Fc␥R was observed (data not OVA Ab and the OVA-HEL Ag, prepared as described previously (26). shown). Captured Fc␥R were still detectable on T cells following Then, 105 MD4 B cells, expressing a HEL-specific IgM BCR, were coin- ϫ 5 5 an overnight culture in the absence of P815 cells (data not shown). cubated at 37°C for 24 h with 2 10 of these OT-I T cells and 10 ϩ 1H11-34 cells. The levels of IL-2 in the harvested culture supernatants Capture of Fc␥R by T cells was also observed with other Fc␥R were measured using an ELISA kit (BD Biosciences). In some assays, tumor cell lines such as the B lymphomas A20 or LB27.4 (data not CD69 up-regulation or BCR down-modulation on MD4 B cells was ana- shown). Furthermore, using HEK cells transfected with cDNA en- ␥ lyzed 2 h after their incubation with OT-I T cells having acquired Fc Ror ␥ not and loaded or not with HEL-OVA IC. coding murine flagged Fc RIIB or with cDNA encoding murine Fc␥RIIIA plus the FcR␥ subunit (also called Fc␧RI␥ or ␥) neces- Results sary for Fc␥RIII expression (20), we found that the expression of ϩ ϩ ␥ ␥ CD4 and CD8 T cells acquire murine Fc␥R upon coculture Fc RIIB or Fc RIIIA allowed trogocytosis triggered by stimula- ␥ ␥ with Fc␥R-expressing cells tory mAb to occur. Indeed, Fc RIIB or Fc RIIIA was captured by OT-I (Fig. 2, A and B) or OT-II T cells (Fig. 2C), as detected by We recently showed that trogocytosis can be triggered in T cells in using either anti-Flag or anti-Fc␥RIIB/IIIA mAb. The fact that we a so-called redirected manner (21). This is observed when T cells could detect Fc␥R using anti-Flag mAb demonstrates that the ex- ␥ are cocultured with Fc R-expressing target cells in the presence of pression of Fc␥R by T cells resulted from capture from transfected Abs directed against certain molecules of the surface of the T HEK cells and not from rapid de novo synthesis or mobilization of ␤ (such as CD3, CD2, or MHC class I/ 2m complex for intracellular stores, as described previously (27). Taken together, instance) (21). For the current study, we determined whether the our results indicate that both Fc␥RIIB and Fc␥RIIIA can be cap- ␥ Fc R itself was included in the set of molecules captured by T tured by T cells during redirected trogocytosis. cells during this process. We found that OT-I CD8ϩ T cells cap- tured the fluorescent lipophilic dye DiO as well as Fc␥R when ␥ exposed to P815 cells in the presence of various mAb triggering T cells capture the signal-transducing FcR subunit along with ␥ trogocytosis, but not in control conditions with no mAb (Fig. 1, A Fc R during trogocytosis and B), with isotype controls or mAb not triggering trogocytosis Because the FcR␥ subunit is necessary for signal transduction after (data not shown) (21). For any given stimulatory mAb, the inten- Fc␥R stimulation, we analyzed whether ⌻ cells captured the FcR␥ sity of Fc␥R capture was proportional to that of fluorescent li- subunit together with Fc␥R during trogocytosis. For these exper- pophilic dye (Fig. 1B). Similar results were obtained with OT-II iments, we generated HEK cells expressing mFc␥RIIB or 6106 FUNCTIONALITY OF Fc␥R CAPTURED BY T CELLS mFc␥RIIIA and a fusion protein in which FcR␥ is fused to GFP. Using these target cells in redirected trogocytosis experiments, we found that the FcR␥GFP subunit was captured efficiently by OT-I T cells when the coculture was performed in the presence, but not in the absence of stimulatory mAb (Fig. 3A). ⌻ cells cocultured with HEK expressing both Fc␥RIIB and the FcR␥GFP subunit in the presence of a stimulatory mAb captured the FcR␥GFP subunit (Fig. 3A) together with Fc␥RIIB (data not shown), but no capture oc- curred when we used ⌯⌭⌲ cells expressing the FcR␥GFP subunit in the absence of Fc␥R (Fig. 3A). In all cases, when the FcR␥ subunit was captured by T cells, its detection using anti-GFP Ab required permeabilization of the cells (data not shown). Capture of the FcR␥ subunit was confirmed using OT-II T cells that were cocultured with P815 cells in the presence or absence of stimula- tory mAb before analysis by flow cytometry. As shown in Fig. 3B, the FcR␥ subunit was strongly detected in OT-II cells exposed to P815 cells in the presence of a mAb triggering trogocytosis, and provided that staining with anti-FcR␥ mAb was performed in per- meabilized cells. In the absence of permeabilization, the FcR␥ sub- Downloaded from unit was only barely detectable (Fig. 3B). Taken together, our re- sults indicate that the FcR␥ subunit is captured by CD4ϩ T cells together with Fc␥R and appears to be mostly expressed in a correct orientation after capture. Capture of Fc␥R is triggered by Ag recognition both on http://www.jimmunol.org/ immortalized cell lines and primary APC In the previous experiments, Fc␥R capture was triggered by stim- ulatory mAb. To determine whether Fc␥R were captured upon physiological stimulation of T cells with Ag, we first used immor- talized cell lines expressing the Ag. As shown in Fig. 4A, OT-I T cells captured Fc␥RIIIA when cultured with HEK cells cotrans- b ␤ ␥ fected with a vector encoding the covalent H-2K -SIINFEKL- 2m FIGURE 4. Capture of Fc R by OT-I T cells from primary DC or mac- complex (23), but not in the absence of the latter construct. As rophages occurs during mAb- or Ag-mediated trogocytosis. A, HEK cells ϩ expected, no capture was detected when HEK cells did not express (left panel) or a stable transfectant expressing Fc␥RIIIA FcR␥ (right by guest on September 28, 2021 Fc␥RIIIA (Fig. 4A). We next examined whether Fc␥R capture by panel) were used as target cells and were incubated with OT-I T cells in the T cells would occur in the presence of primary APC. For these absence (gray histograms) or presence (dotted line) of the anti-CD2 mAb. For Ag-mediated trogocytosis, HEK cells and HEK-Fc␥RIIIAϩFcR␥ were experiments, peritoneal macrophages or BM-DC generated from transiently transfected with a vector encoding a covalent H-2Kb-OVA B6 mice were used as APC and cocultured with OT-I T cells in the complex 48 h before incubation with OT-I T cells (solid line). Transfection presence of OVA. As shown in Fig. 4B, OT-I T cells readily cap- efficiency was Ͼ80% (data not shown). At the end of the incubation period, tured Fc␥R following stimulation with BM-DC or peritoneal mac- cells were labeled with anti-CD8 and anti-Fc␥RIIB/RIIIA mAb. Histo- rophages in the presence of the OVA peptide. The efficiency of grams show Fc␥RIIB/IIIA staining on gated CD8ϩ T cells. B,AsinA, Fc␥R capture triggered by the Ag was comparable to that observed except that BM-DC (left panels) or peritoneal macrophages (M␾; right in redirected trogocytosis experiments with anti-CD2 mAb (Fig. panels) from B6 mice were pulsed (OVA) or not (medium) with the 4B). Similar results were obtained with OT-II T cells (data not SIINFEKL peptide and then incubated with OT-I T cells. OT-I T cells were shown). Thus, Ag recognition by T cells on artificial as well as coated (anti-CD2) or not with the anti-CD2 mAb in control experiments. At primary APC-expressing Fc␥R results in the capture of Fc␥RbyT the end of the 1-h incubation period, cells were analyzed by flow cytometry using anti-CD8 and anti-Fc␥RIIB/RIIIA mAbs. Numbers in bold indicate cells. ϩ the median fluorescence intensity of anti-Fc␥R staining on CD8 T cells. T cells bind IC after Fc␥R acquisition The other numbers represent the percentage of cells in each quadrant. Sim- ilar results were obtained in a second independent experiment. An important question regarding receptors captured by trogocyto- sis is whether they remain functional. To determine whether Fc␥R retained their ability to bind IC once captured by OT-I CTL, T cells nor P815 cells were stained with IC formed with FabЈ frag- cells were first subjected to trogocytosis in the presence of Fc␥R- ments of F9 and F10 instead of whole mAb molecule, confirming expressing APC and then incubated with IC formed by an equimo- that binding of IC by captured as well as native Fc␥R occurred lar mixture of the F9 and F10 anti-HEL mAbs complexed with through interactions with the Fc fragment of mAb. Similarly, biotinylated HEL. As shown in the top panels of Fig. 5A, OT-I OT-II T cells could bind IC after capture of Fc␥R from P815 or cells that had captured Fc␥R (i.e., cocultured with P815 in the LB27 cells (data not shown). Thus, Fc␥R captured by T cells con- presence of mAb) strongly bound IC, but their Fc␥R-negative ferred onto these cells the ability to efficiently bind IC. counterparts (i.e., cocultured with P815 in the absence of mAb) did not. No staining was detected when OT-I cells bearing Fc␥R were Failure to detect biochemical or functional signal from the ␥ incubated with HEL alone or with a mixture of biotinylated F9 and stimulation of captured Fc R F10 mAb alone (Fig. 5B). In contrast, P815 cells (lower right To determine whether Fc␥R could transduce intracellular signals quadrants) bound IC both in the presence and absence of mAb. after their capture by trogocytosis, we stimulated Fc␥R on T cells Furthermore, as shown in the lower panels of Fig. 5A, neither OT-I having acquired Fc␥R (FcRacqϩ) and, as a control, on T cells The Journal of Immunology 6107

FIGURE 5. OT-I bind IC after Fc␥R acquisition. A, OT-I T cells (ap- pearing in the upper quadrants) were incubated with P815 (appearing in the lower quadrants) in the presence of the indicated mAb, as described in Fig. 1. At the end of the incubation period, cells were incubated at 4°C for 1 h with IC formed between bio- tinylated HEL and either whole anti- HEL F9 and F10 IgG (top panels)or Ј F(ab )2 of the same mAb (bottom panels). Cells were then analyzed us- ing anti-CD8 mAb and fluorescent streptavidin. Numbers represent the percentage of cells in each quadrant. The experiment shown is representa-

tive of three independent experi- Downloaded from ments. B,AsinA, except that OT-I T cells having captured Fc␥RIIB/IIIA during coculture with P815 cells in the presence or absence of the anti- CD2 mAb were then incubated at 4°C for 1 h with biotinylated HEL (left two panels) or biotinylated anti-HEL http://www.jimmunol.org/ F9 and F10 IgG (right two panels).

cocultured with P815 cell in the absence of mAb (FcRacqϪ), and we failed in transducing OT-I and OT-II T cells with a sufficient we measured various biochemical and functional responses. Upon efficiency using the viral vector encoding Fc␥RIIIA-FcR␥ chimera stimulation with IC, we were unable to detect the production of (data not shown). Therefore, we transposed our findings to a hu- cytokines such as TNF-␣ by FcRacqϩ or FcRacqϪ OT-I T cells man system in which PBMC transduced with Fc␥RIIIA-FcR␥ by guest on September 28, 2021 (Fig. 6A), whereas TNF-␣ was produced in response to antigenic could be readily generated (18) (see also Fig. 7A). Similar to our stimulation (Fig. 6B). Similarly, we could not detect Ca2ϩ flux observations made in the murine model (21), we found that addi- upon IC stimulation of OT-I FcRacqϩ T cells above levels ob- tion of anti-CD3 OKT3 mAb or W6/32 anti-MHC class I mAb in served for control FcRacqϪ OT-I T cells. In contrast, both types of cocultures of HEK cells expressing Fc␥RIIIA-FcR␥ chimera and T cells strongly mobilized their Ca2ϩ stores upon treatment with human CD4ϩ and CD8ϩ T cells led to the capture of the PMA and ionomycin (Fig. 6, C and D). Similar results were ob- Fc␥RIIIA-FcR␥ chimera by T cells (data not shown and Fig. 7B). tained with OT-II T cells (data not shown). As a control, we found Interestingly, T cells having captured Fc␥RIIIA-FcR␥ chimera that P815 target cells displayed Ca2ϩ mobilization upon stimula- (Fc␥RIIIA-FcR␥acqϩ T cells) did not lyse Raji cells coated with tion with IC (data not shown). No ADCC activity against Raji cells rituximab better than control T cells incubated with HEK cells in coated with the anti-CD20 Ab rituximab was detected with OT-I or the absence of any mAb (Fc␥RIIIA-FcR␥acqϪ T cells) (Fig. 7C)or OT-II T cells having acquired Fc␥RIIB/IIIA by mAb-triggered not incubated with HEK cells (data not shown). In contrast, T cells trogocytosis (Fig. 6E). Similarly, we could not detect ADCC ac- transduced with the retroviral vector expressing Fc␥RIIIA-FcR␥ tivity (but efficient cytotoxicity against OVA-pulsed target cells) efficiently lysed Raji cells coated with rituximab, showing that T with OT-I T cells having acquired Fc␥RIIB/IIIA during trogocy- cells do possess a functional machinery able to translate Fc␥R tosis-triggered Ag recognition on HEK-Fc␥RIIIAϩFcR␥ or stimulation into a functional response. Taken together, although T BM-DC (Fig. 6F), suggesting that trogocytosis triggered by Ab or cells can efficiently capture Fc␥R, we were unable to detect func- Ag is comparable. Altogether, we were unable to detect any Ca2ϩ tional signal upon stimulation, whereas we could do so when Fc␥R mobilization, TNF-␣ production, or cytotoxicity upon stimulation were endogenously expressed by T cells. of T cells having acquired Fc␥R, suggesting that Fc␥R captured by T cells may not be functional. To determine whether our inability ␥ to detect function of captured Fc␥R could be due to their improper Fc R captured by T cells exhibit different features as compared ␥ connection to the T cell signal transduction machinery or to the with Fc R endogenously expressed by target cells lack of some of the biochemical mediators necessary to transduce The discrepancy between the strong ligand-binding ability and signals downstream of Fc␥R that are present in APC, we used a the undetected function of Fc␥R captured by T cells could be previously generated retroviral vector (18) encoding a human due to the fact that these Fc␥R may not be properly integrated Fc␥RIIIA-FcR␥ chimeric protein to assess the function of Fc␥R in the T cell membrane, and consequently, could not properly acquired vs endogenously expressed. We generated HEK cells ex- connect to the signaling machinery. Therefore, we examined pressing high levels of Fc␥RIIIA-FcR␥ chimera and found that whether evidence of improper integration of Fc␥R could be these target cells could be used in redirected trogocytosis experi- found. First, as shown in Fig. 8, we found that upon noncyto- ments with OT-I or OT-II T cells (data not shown). Unfortunately, lytic acid treatment, a large proportion of Fc␥R acquired by T 6108 FUNCTIONALITY OF Fc␥R CAPTURED BY T CELLS Downloaded from http://www.jimmunol.org/

FIGURE 6. Failure to detect biochemical or functional signals emanating from Fc␥R stimulation on T cells after trogocytosis. A, OT-I T cells were incubated with HEK-expressing Fc␥RIIIAϩFcR␥ in the presence (bottom panel,Fc␥Racqϩ) or absence (top panel,Fc␥Racq-) of the anti-H-Kb mAb triggering trogocytosis, separated from adherent HEK cells, and then stimulated with IC before analysis of intracellular TNF-␣ production by intracellular staining using anti-CD8 and anti-TNF-␣ mAb. Numbers represent the percentage of cells in each quadrant. B,AsinA, except that OT-I T cells were by guest on September 28, 2021 incubated for4hat37°C with (bottom panel) or without (top panel) OVA peptide, followed by measurement of TNF-␣ production. C,AsinB, except that after separation of OT-I T cells from HEK cells, OT-I T cells were labeled with Indo-1 and recorded for Ca2ϩ mobilization over time. After baseline recording for 1 min, IC were added (first arrow from the left), and then after 4 min PMA/ionomycin was added (second row). D,AsinC, except that HEK-expressing Fc␥RII were used as targets during incubation with OT-I T cells. E, OT-I (f) or OT-II (u) T cells before (Ϫ) or after (ϩ) incubation with HEK-Fc␥RIIIAϩFcR␥ in the absence or presence of the anti-H-2Kb mAb were introduced in a 5-h chromium release assay using Raji cells coated with rituximab as target cells (E:T ratio was 60:1). As a positive control for the lysis of Raji cellsϩ rituximab, activated human PBMC (Ⅺ) were used as effector cells (E:T ratio was 60:1). Note that specific lysis of Raji cells not coated with rituximab was below 5%. F, OT-I T cells were exposed to HEK-Fc␥RIIIAϩFcR␥ that were transfected (ϩOVA Ag) or not (ϪOVA Ag) with the construct encoding covalent H-2Kb-OVA complex or to BM-DC pulsed (ϩOVA Ag) or not (ϪOVA Ag) with the OVA peptide, as indicated by the plus and minus signs below the graph. Then OT-I T cells were introduced in a 5-h chromium release assay using Raji cells coated with rituximab as target cells (f, E:T ratio was 60:1) or to EL-4 cells that were pulsed with the OVA (Ⅺ, E:T ratio was 60:1). Lysis of Raji cells alone or EL-4 cells alone by OT-I T cells was below 10% in all cases (data not shown).

cells was removed from the T cell surface (76%, loss as calcu- Fc␥RIIB/IIIA-expressing OT-I T cells present IC to B cells, lated using median fluorescence intensities), whereas this treat- allowing T cell help ment had little or no effect on Fc␥R expressed by P815 cells Having shown that FcR capture allows T cells to bind IC in the (11% loss). Addition of polybrene, known to favor fusion, dur- absence of detected signaling-dependent function, we wondered ing or after acid treatment, did not lead to increased acid resis- whether immune functions involving solely IC binding could be tance of Fc␥R captured by T cells (data not shown). Second, we bestowed on T cells by Fc␥RIIB/IIIA acquisition. For that, we found that Fc␥R captured by T cells were not down-modulated ␥ upon treatment with the C2F2 Ab against CD71, whereas this determined whether Fc RIIB/IIIA capture by T cells allowed them treatment induced a massive down-modulation of Fc␥R endo- to present IC to B cells in order for B cells to benefit from T cell genously expressed by P815 cells (97% loss, as calculated using help. To address this question, we used the experimental system median fluorescence intensities), further strengthening the hy- presented in Fig. 9A, in which OT-I T cells having captured pothesis of a distinct molecular layout of captured vs endoge- Fc␥RIIB or not in a previous culture with HEK-Fc␥RIIB were nous Fc␥R in the membrane (Fig. 8). By using these two dif- incubated with OVA-HEL IC. Then after extensive washing, these ferent approaches, we could thus show that the majority of cells were incubated with MD4 B cells and the activation of the B Fc␥R molecules captured by T cells had features suggesting cells by CD69 up-regulation (presented in Fig. 9B). Note that, in their improper insertion in the T cell membrane. this case, we used an OVA-HEL Ag complexed with anti-OVA Ab The Journal of Immunology 6109 Downloaded from http://www.jimmunol.org/ FIGURE 7. Human CD4ϩ and CD8ϩ T cells perform detectable ADCC after endogenous expression, but not capture of human Fc␥RIIIA-␥. Activated PBMC were transduced or not with a retroviral vector encoding the chimeric Fc␥RIIIA-FcR␥ molecule. Nontransduced PBMC were incubated with HEK cells expressing Fc␥RIIIA-FcR␥ in the presence or absence of the OKT-3 anti-CD3 mAb found to trigger the capture of Fc␥RIIIA by T cells. Then T cells were separated from adherent HEK cells and used in a 5-h chromium release assay against Raji cells, coated or not with the anti-CD20 mAb rituximab. A, Flow cytometry analysis showing Fc␥RIIIA expression of transduced (right panel) or nontransduced PBMC (left panel). Numbers represent the percentage of cells in each quadrant. B, Analysis by flow cytometry of the capture of Fc␥RIIIA-FcR␥ by nontransduced PBMC (as shown in A) incubated with HEK expressing Fc␥RIIIA-FcR␥ in the presence (open histogram) or absence (closed histograms) of the OKT3 anti-CD3 mAb. Top panel, Represents Fc␥R staining on gated CD8ϩ T cells; bottom panel, on gated CD4ϩ T cells. C, Percent specific lysis of Raji cells coated with rituximab in the presence of PBMC exposed to HEK cells expressing Fc␥RIIIA-FcR␥ in the presence of OKT3 (Fc␥RIIIAacqϩ, gray symbols) or not (Fc␥RIIIAacqϪ, white symbols) or PBMC endogenously expressing Fc␥RIIIA-FcR␥ (Fc␥RIIIAϩ, black symbols) after retroviral transduction. Note that killing of Raji cells not coated with by guest on September 28, 2021 rituximab was comparable to killing obtained with Fc␥RIIIacqϩ and Fc␥RIIIacqϪ (data not shown). to allow recognition of HEL by the MD4 BCR. To determine not acquired Fc␥RIIB/IIIA (i.e., exposed in a previous culture to whether IC presented by OT-I T cells were captured and processed HEK-Fc␥RIIB or HEK-Fc␥RIIIA with no mAb added or no Ag by the B cells, we then added the 1H11-34 T cell hybridoma spe- expressed). No IL-2 production above background was detected cific for the HEL Ag presented by the MHC II I-Ed molecule and when cocultures were performed between OT-I T cells (that had measured IL-2 production by ELISA (presented in Fig. 9, C and acquired Fc␥RIIB/IIIA) and either MD4 B cells, or 1H11-34 T D). Importantly, the source of IL-2 in this system can only be cells, even in the presence of HEL-OVA IC (Fig. 9D). As ex- identified as the 1H11-34 T cell hybridoma and not the OT-I T pected, cocultures between MD4 B cells and 1H11-34 led to pro- cells because OT-I T cells are of the H-2b haplotype, whereas MD4 duction of IL-2 when HEL-OVA or HEL-OVA IC were added B cells (and the 1H11-34 T cell hybridoma) are of the H-2d hap- (Fig. 9D). Thus, based on these in vitro experiments, Fc␥R cap- lotype. As shown in Fig. 9B, CD69 was up-regulated at the surface tured by T cells may play a functional role in vivo by provisioning of MD4 B cells exposed to OT-I T cells having acquired FcR when B cells with immobilized Ag and permitting T cell help. the latter cells were coated with IC, but not when left uncoated. No detectable CD69 up-regulation was noticed in the presence of OT-I T cells that had not captured FcR. As a control, CD69 was mark- Discussion edly up-regulated on B cells exposed to HEL-OVA and slightly Although the transfer of proteins between immune cells has been up-regulated when exposed to IC. Each time we could detect CD69 anecdotically reported for years, the widespread nature of this phe- up-regulation, we noticed a correlated BCR ␬-chain internalization nomenon and its potential biological significance have only been (data not shown). Thus, T cells having captured FcR can present described recently. Indeed, protein transfer, and in particular tro- IC to and activate B cells. As shown in Fig. 9C, OT-I cells having gocytosis, has been proposed to play a role in fundamental pro- acquired Fc␥RIIB from HEK-Fc␥RIIB present OVA-HEL IC to cesses such as the initiation and regulation of immune responses HEL-specific B cells, leading to IL-2 production by the 1H11-34 (5, 6). Thus, it is critical to identify molecules that can be captured T cell hybridoma. Comparable results were obtained when OT-I T by immune cells, understand the molecular basis of the transfer, cells that had captured Fc␥RIIIA in a previous culture with HEK- and, most importantly, define the function of captured molecules Fc␥RIIIAϩ FcR␥ and that were transfected to express the H-2Kb- on recipient cells. In this study, we showed that T cells efficiently ␤ ␥ OVA- 2m complex (as in Fig. 4A) were used. In contrast, no IL-2 acquired Fc R from APC during mAb- or Ag-mediated stimula- production above background (measured in the absence of IC stim- tion and expressed it in a correct orientation and topology on their ulation; data not shown) was observed with OT-I T cells that had surface. After capture of Fc␥R, T cells could bind Fc␥R ligands 6110 FUNCTIONALITY OF Fc␥R CAPTURED BY T CELLS

in the TCR complex (37, 43, 44). Besides endogenous expression, it was suspected years ago that T cells could uptake Fc␥R released from macrophages (10). However, it was not known whether a proteolytic cleavage or other mechanisms more closely related to trogocytosis were involved in this protein transfer. In this study, we demonstrated that the expression of Fc␥R by T cells shortly after coculture with Fc␥R-expressing target cells reflected capture by trogocytosis. T cells could capture both Fc␥RIIB/Fc␥RIIIA from immortalized cell lines, but also from primary, bona fide APC upon stimulation with the Ag or with defined stimulatory mAb. Furthermore, Fc␥R capture was observed for both CD4ϩ and CD8ϩ T cells and for murine as well as human T cells, sug- gesting that this process commonly occurs in all T cells. Thus, Fc␥R extends the list of molecules captured by T cells upon pro- ductive recognition of APC. We found that Fc␥R and the FcR␥ subunit were expressed in a correct conformation and topology after their capture by trogocy- tosis. This conclusion is based on the facts that extracellular

epitopes were detected with conformational mAb, intracellular Downloaded from epitopes were detected mainly after cell permeabilization, and ac- quired Fc␥R could efficiently bind IC. Our data are reminiscent of those obtained by Davis and colleagues (24, 25) regarding the capture of Cw6-GFP molecules by NK cells, and support the no- tion that both the conformation and orientation of molecules cap-

tured by trogocytosis are correct. In the case of Fc␥R, this result http://www.jimmunol.org/ FIGURE 8. Fc␥R captured by T cells display different stability and in- bears the additional significance that the ability of captured Fc␥R ternalization as compared with Fc␥R expressed by target cells. OT-I T cells to bind IC could endow T cells with novel and unexpected func- were incubated with P815 cells in the absence (left panels) or presence tional properties. (right panels) of the anti-CD2 mAb. At the end of the incubation period, Multiple hypotheses have been proposed regarding the role cells were left untreated (top panels), incubated 1 min in a noncytolytic (pH played by molecules captured by T cells. Usually, these roles were 3.0) buffer, and then extensively washed (middle panels) or incubated 1 h inferred from those played by the same molecules when expressed at 37°C with anti-CD71 mAb. Cells were then analyzed by flow cytometry on the donor cell, assuming that the same roles would be played on using anti-CD8 and anti-Fc␥RIIB/RIIIA mAb. Quadrants were placed such recipient T cells. However, little data support this notion in many upper quadrants lower that T cells appear in the and target cells in the by guest on September 28, 2021 quadrants. Numbers represent the percentage of cells in each quadrant. cases. The most convincing function demonstrated for trogocytosis Percentages of loss of Fc␥R expression upon the various treatments are concerns the impact of captured molecules (such as peptide-MHC indicated in the text. Note that we do not want to mean that anti-CD71- complexes) on immune regulation through T-T cell interactions (1, mediated Fc␥RIIB/IIIA internalization has any physiological relevance, 6). Noticeably, these novel functions depend on the T cell ability but is simply yet another way to show that Fc␥R do not behave in recipient to present the ligands they acquired to receptors present on their T cells as in donor P815 cells. own membrane or on surrounding cells. Similar to these reports, we found that Fc␥R was perfectly able to bind physiological li- gands such as IC after capture by T cells. As suggested by our very efficiently, but we were unable to detect biochemical or func- results presented in Fig. 9, this could affect the immune response tional response resulting from this binding. Furthermore, we found by providing T cells with the ability to bind and present IC to evidence for a different behavior of Fc␥R captured by T cells as neighboring cells expressing Fc␥R or BCR. By engaging Fc␥RIIB, compared with receptors endogenously expressed by donor cells the inhibitory Fc␥R expressed by B cells, soluble Ag-Ab com- that was most likely secondary to the incomplete integration of plexes inhibit B cell activation (45). In contrast, as suggested by captured Fc␥R into the T cell membrane. Based on our results, we our results, Ag-Ab complexes bound to FcR captured by T cells propose that the capture of Fc␥R by T cells could provide them could activate rather than inhibit B cells most likely because Fc novel opportunities to interact with ligands such as IC or IgG- fragments engage FcR on T cells and would not be available for opsonized cells. This may possibly occur without triggering sig- Fc␥RIIB recognition on B cells. Furthermore, acquired Fc␥R naling-dependent responses in these recipient T cells because we could function similarly to another member of the Fc␥R family were unable to detect such signals. whose function does not depend on intracellular signaling, i.e., the Fc␥R are not generally considered as typical molecules ex- glycophosphatidylinositol-anchored Fc␥RIIIB, resulting, for ex- pressed by conventional T cells (28). However, it is now clear that ample, in recruitment of cells to IC (12, 46). Nevertheless, addi- subpopulations of conventional T cells both in the mouse (10, 29– tional studies are clearly required to understand whether any roles 35) and in humans (36–39) do endogenously express Fc␥Ror could be played by T cells having acquired Fc␥R by trogocytosis subunits of the Fc␥R complex. Apart from ␥␦T cells and NK T in vivo. cells, this is for instance the case of double-negative T cells (32), Although the capture of several receptors by T and NK cells via some conventional CD4ϩ or CD8ϩ ␣␤T cells exhibiting an acti- trogocytosis has already been documented, their functionality on vated phenotype (18, 36, 39), and some tumor-infiltrating lympho- recipient cells was always examined in terms of ligand-binding, cytes (40). Expression of Fc␥R by these cells was shown to confer but not signaling properties (6, 8, 47, 48). We thus examined onto them novel biological functions, such as the ability to perform whether Fc␥R captured by T cells could induce intracellular sig- ADCC (18, 35, 36, 39), to produce lymphokines (35, 41, 42) and, nals and/or signaling-dependent functions. Studies of Fc␥R func- in the case of the FcR␥ subunit, to functionally replace the ␨-chain tion in T cell subpopulations endogenously expressing Fc␥R have The Journal of Immunology 6111 Downloaded from http://www.jimmunol.org/

FIGURE 9. After Fc␥R acquisition, OT-I present IC to B cells to allow B cell-T cell cooperation. A, Cartoon showing the cell cultures and molecular signals allowing us to determine whether OT-I T cells having acquired Fc␥R could present IC to MD4 B cells and allow them to participate in a T cell-B cell cooperation. B, MD4 B cells were incubated with OT-I T cells, which, in a first coculture, had acquired Fc␥RIIB (OT-I-Fc␥RIIBacqϩ)ornot by guest on September 28, 2021 (OT-I-Fc␥RIIBacqϪ) from HEK-Fc␥RIIB. OT-I cells were pulsed (open histograms) or not with HEL-OVA IC (gray histograms) before extensive washing and addition to MD4 B cells. After 2 h, cells were stained with mAbs directed against CD69 and B220 before analysis by flow cytometry. Shown are overlaid histograms of CD69 expression on gated B220ϩ cells in conditions in which IC were present or not on OT-I T cells. Controls show CD69 expression on MD4 B cells alone exposed or not to IC or, on the left panel, to HEL-OVA. C, OT-I T cells were incubated for1hat37°C with HEK-Fc␥RIIB or Fc␥RIIIAϩ FcR␥ coexpressing or not the OVA Ag through transfection by the construct encoding the covalent H-2Kb-OVA complex (to trigger Ag-mediated trogocytosis), or in the presence of the anti-CD2 mAb (to trigger mAb-mediated trogocytosis), or with no particular stimulus (none). As in B, except that the HEL-specific I-Ed-restricted 1H11-34 T hybridoma was added to the culture and that IL-2 production was measured by ELISA in duplicate wells in the experimental and control conditions indicated below the bars. f and Ⅺ, Represent two independent experiments. D,AsinC, except that this panel shows the controls performed as indicated below the graph for the experiment presented in f in C. established that the cellular machinery necessary for Fc␥R signal- Fc␥R captured by T cells or at levels too low to be detected. In ing is usually present in T cells (18, 36, 39, 41). However, we were agreement with our results, it was recently found that bystander B unable to detect any Ca2ϩ flux or cytokine production upon IC cells rapidly acquire Ag receptors from activated B cells by tro- stimulation of Fc␥R captured by T cells. Furthermore, T cells hav- gocytosis, and that they display Ca2ϩ flux upon stimulation of the ing acquired Fc␥R did not induce detectable ADCC even though endogenously expressed BCR, but not the acquired BCR (9). As a they expressed high levels of Fc␥R after capture and could effi- counterexample, the observations that CTL (2) or DC (49) having ciently bind IC. In contrast, T cells endogenously expressing Fc␥R acquired peptide-MHC complexes are lysed by CTL specific for after transduction are quite efficient at performing ADCC, which is the acquired Ag could suggest that peptide-MHC complexes are a very sensitive assay. These results show that, at least in this case, well integrated in the recipient T cell membrane and functional as the fact that we could not detect ADCC with captured Fc␥Risnot they sensitize T cells to fratricide killing. However, it is unknown due to an improper association with the FcR␥ subunit. Note that whether peptide-MHC integration in the recipient T cell membrane we previously reported that ADCC could easily be detected in the is mandatory for the execution of perforin-mediated killing. Fur- case of some memory ␣␤T cells naturally expressing Fc␥RIIIA in thermore, the observation that acquired molecules such as peptide- humans, at levels lower than those observed in this study after MHC complexes can participate in signaling in the recipient T capture (39) (our unpublished data), indicating that detectable cells (3) is compatible with our data, because in this case, the function could be expected from the high levels of Fc␥R captured. acquired molecules interact with a functional, endogenously ex- Taken together, although we were not able to detect whether very pressed receptor to maintain signal propagation in recipient T cells. early activation events could be triggered by IC stimulation of T The fact that acquired FcR may not transmit intracellular signal cells having acquired Fc␥RII/III, one possible interpretation of could be seen as a means to prevent T cells from responding to these results is that biochemical signals might not be transduced by stimulation by promiscuous IC. 6112 FUNCTIONALITY OF Fc␥R CAPTURED BY T CELLS

FIGURE 10. Cartoon depicting possible scenarios for efficient ligand binding, but undetected functional response of Fc␥R after their capture by T cells via trogocytosis. A, Fc␥R molecules captured by T cells are present in a correct orientation and topology within a membrane fragment ad- sorbed on the T cell membrane. B,Fc␥R molecules cap- tured by T cells are present in a correct orientation and topology within a membrane fragment fused to the T cell membrane, but disconnected from the Fc␥R-signaling ma- chinery expressed in T cells. A similar scenario could re- sult from the improper association between Fc␥RIIIA and the FcR␥ subunit in the case of the natural receptor. Note that these two models are not mutually exclusive.

Several models could explain how captured Fc␥R may not trig- thermore, although these studies involved other hematopoietic ger signaling or signaling-dependent function in T cells. It is pos- cells, trogocytosis triggered by therapeutic mAb such as rituximab sible that either receptors are not properly inserted within the has been reported both in vitro (52) and in vivo (53), suggesting plasma membrane of the recipient cell (Fig. 10A) or they are well that this process has biological relevance. integrated in the plasma membrane, but not properly connected to Our results have important implications for the consequences of Downloaded from the signaling machinery (Fig. 10B). Although we do not exclude trogocytosis in T cell biology. Indeed, not only did we show that this last possibility, we found indications that Fc␥R captured by T T cells could capture Fc␥R that subsequently bound IC with great cells exhibited features best compatible with the first hypothesis. efficiency, but our data further advance the two major questions First, a large proportion of Fc␥R captured by T cells was shed concerning the mechanisms of trogocytosis, e.g., the molecular upon noncytolytic acid treatment (whereas this treatment had little basis of protein transfer and the extent of the function of captured

effect on the expression of endogenous Fc␥R by donor cells). molecules. http://www.jimmunol.org/ However, whether the acid-resistant molecules are integrated in the recipient cell membrane or are not integrated, but still resistant Acknowledgments to release, remains unclear. Second, we showed that Fc␥R cap- We thank Etienne Joly, Anne Aucher, Ste´phanie Cabantous, and members tured by T cells were not down-modulated upon treatment with an of the I3 Club at the Institut de Pharmacologie et de Biologie Structurale for Ab against CD71, whereas this treatment induced a massive down- stimulating discussions, as well as Jean-Charles Guery for the gift of anti- modulation of Fc␥R endogenously expressed by P815 cells (pos- HEL Abs and Ted Hansen for the vector encoding H-2Kb-OVA complex. sibly reflecting comodulation of these two molecules in recycling We thank He´le`ne Bour-Jordan for scientific editing of our paper. This work compartments upon anti-CD71 mAb stimulation), further strength- benefited from the assistance of the electron microscopy facility of the IFR ening the hypothesis of a distinct molecular layout of captured vs 109 (TRI platform). by guest on September 28, 2021 endogenous Fc␥R in the membrane. Finally, reminiscent of the structures observed by Davis and colleagues (25) on NK cells, our Disclosures preliminary electron microscopy analysis of FcR␥GFP expressed The authors have no financial conflict of interest. on T cells after capture supports the notion that Fc␥R are present in membranous structures loosely attached to the plasma mem- References brane of T cells (data not shown). Finally, Patel et al. (50) observed 1. Wetzel, S., and D. Parker. 2006. MHC transfer from APC to T cells following antigen recognition. Crit. Rev. Immunol. 26: 1–21. that cells synthesizing, but not those having acquired a given mol- 2. Huang, J. F., Y. Yang, H. Sepulveda, W. Shi, I. Hwang, P. A. Peterson, ecule, were sensitive to complement lysis triggered by mAb M. R. Jackson, J. Sprent, and Z. Cai. 1999. TCR-mediated internalization of against this molecule. Considering that the completion of fusion peptide-MHC complexes acquired by T cells. Science 286: 952–954. 3. Wetzel, S., T. McKeithan, and D. Parker. 2005. Peptide-specific intercellular events is usually so fast that the different steps are very difficult to transfer of MHC class II to CD4 T cells directly from the immunological synapse observe (51), we believe that membrane fragments containing upon cellular dissociation. J. Immunol. 174: 80–89. Fc␥R could readily fuse with the T cell membrane in a proper way 4. Hudrisier, D., J. Riond, H. Mazarguil, J. E. Gairin, and E. Joly. 2001. Cutting edge: CTLs rapidly capture membrane fragments from target cells in a TCR within the time frame of our coculture assays (up to 5 h). Note that signaling-dependent manner. J. Immunol. 166: 3645–3649. the inhibition of functional properties of T cells having captured 5. Joly, E., and D. Hudrisier. 2003. What is trogocytosis and what is its purpose? . Nat. Immunol. 4: 815–815. these molecules (6) could simply be due to masking of critical 6. Davis, D. M. 2007. Intercellular transfer of cell-surface proteins is common and ligands or receptors by vesicles acquired from other cells rather can affect many stages of an immune response. Nat. Rev. Immunol. 7: 238–243. than by signaling events transduced from captured receptors. 7. Hudrisier, D., and P. Bongrand. 2002. Intercellular transfer of antigen-presenting cell determinants onto T cells: molecular mechanisms and biological significance. In conclusion, we propose that molecules captured by trogocy- FASEB J. 16: 477–486. tosis may perform functions that do not require their integration in 8. Patel, D. M., P. Y. Arnold, G. A. White, J. P. Nardella, and M. D. Mannie. 1999. the plasma membrane, such as interactions with ligands on the Class II MHC/peptide complexes are released from APC and are acquired by T cell responders during specific antigen recognition. J. Immunol. 163: 5201–5210. same or different cells. Whether this property could be generalized 9. Quah, B. J., V. P. Barlow, V. McPhun, K. I. Matthaei, M. D. Hulett, and to other receptors and immune cells has yet to be determined be- C. R. Parish. 2008. Bystander B cells rapidly acquire antigen receptors from cause no other receptors were reported to date for their ability to activated B cells by membrane transfer. Proc. Natl. Acad. Sci. USA 105: 4259–4264. transmit intracellular signals once captured by T cells. 10. Lee, S. T., and F. Paraskevas. 1978. Macrophage–T cell interactions. I. The Most of the experiments performed in this study used mAb to uptake by T cells of Fc receptors released from macrophages. Cell. Immunol. 40: 141–153. trigger trogocytosis. This choice was made because mAb often 11. Bruhns, P., B. Iannascoli, P. England, D. A. Mancardi, N. Fernandez, S. Jorieux, trigger trogocytosis more efficiently than Ag. Importantly, we and M. Daeron. 2009. Specificity and affinity of human Fc␥ receptors and their found no qualitative differences in trogocytosis triggered by Ab or polymorphic variants for human IgG subclasses. Blood 113: 3716–3725. ␥ 12. Ravetch, J., and J. Kinet. 1991. Fc receptors. Annu. Rev. Immunol. 9: 457–492. Ag stimulation with regard to the capture of Fc R by T cells, 13. Lutz, M. B., N. Kukutsch, A. L. Ogilvie, S. Rossner, F. Koch, N. Romani, and suggesting that both processes exhibit comparable features. Fur- G. Schuler. 1999. An advanced culture method for generating large quantities of The Journal of Immunology 6113

highly pure dendritic cells from mouse bone marrow. J. Immunol. Methods 223: 34. Mizoguchi, H., J. J. O’Shea, D. L. Longo, C. M. Loeffler, D. W. McVicar, and 77–92. A. C. Ochoa. 1992. Alterations in signal transduction molecules in T lymphocytes 14. Hogquist, K. A., S. C. Jameson, W. R. Heath, J. L. Howard, M. J. Bevan, and from tumor-bearing mice. Science 258: 1795–1798. F. R. Carbone. 1994. T-cell receptor antagonist peptides induce positive selec- 35. Dhanji, S., K. Tse, and H. S. Teh. 2005. The low affinity Fc receptor for IgG tion. Cell 76: 17–27. functions as an effective cytolytic receptor for self-specific CD8 T cells. J. Im- 15. Barnden, M., J. Allison, W. Heath, and F. Carbone. 1998. Defective TCR ex- munol. 174: 1253–1258. pression in transgenic mice constructed using cDNA-based ␣- and ␤-chain genes 36. Bjorkstrom, N. K., V. D. Gonzalez, K. J. Malmberg, K. Falconer, A. Alaeus, under the control of heterologous regulatory elements. Immunol. Cell Biol. 76: G. Nowak, C. Jorns, B. G. Ericzon, O. Weiland, J. K. Sandberg, and 34–40. H. G. Ljunggren. 2008. Elevated numbers of Fc␥RIIIAϩ (CD16ϩ) effector CD8 16. Goodnow, C. C., J. Crosbie, S. Adelstein, T. B. Lavoie, S. J. Smith-Gill, T cells with NK cell-like function in chronic hepatitis C virus infection. J. Im- R. A. Brink, H. Pritchard-Briscoe, J. S. Wotherspoon, R. H. Loblay, K. Raphael, munol. 181: 4219–4228. et al. 1988. Altered immunoglobulin expression and functional silencing of self- 37. Krishnan, S., V. G. Warke, M. P. Nambiar, G. C. Tsokos, and D. L. Farber. 2003. reactive B lymphocytes in transgenic mice. Nature 334: 676–682. The FcR␥ subunit and Syk kinase replace the CD3 ␨-chain and ZAP-70 kinase in 17. Adorini, L., A. Sette, S. Buus, H. Grey, M. Darsley, P. Lehmann, G. Doria, the TCR signaling complex of human effector CD4 T cells. J. Immunol. 170: Z. Nagy, and E. Appella. 1988. Interaction of an immunodominant epitope with 4189–4195. Ia molecules in T cell activation. Proc. Natl. Acad. Sci. USA 85: 5181–5185. 38. Zupo, S., L. Azzoni, R. Massara, A. D’Amato, B. Perussia, and M. Ferrarini. 18. Clemenceau, B., N. Congy-Jolivet, G. Gallot, R. Vivien, J. Gaschet, G. Thibault, 1993. Coexpression of Fc␥ receptor IIIA and interleukin-2 receptor ␤ chain by a and H. Vie. 2006. Antibody-dependent cellular cytotoxicity (ADCC) is mediated subset of human CD3ϩ/CD8ϩ/CD11bϩ lymphocytes. J. Clin. Immunol. 13: by genetically modified antigen-specific human T lymphocytes. Blood 107: 228–236. 4669–4677. 39. Clemenceau, B., R. Vivien, M. Berthome, N. Robillard, R. Garand, G. Gallot, 19. Mukasa, R., Y. Terada, M. Shiroishi, H. Fujiwara, K. Hayata, K. Morishita, S. Vollant, and H. Vie. 2008. Effector memory ␣␤ T lymphocytes can express C. Ra, and T. Takashi. 2005. Rapid receptor-proximal signaling assays for FcR␥- Fc␥RIIIa and mediate antibody-dependent cellular cytotoxicity. J. Immunol. 180: containing receptors. J. Immunol. Methods 303: 105–121. 5327–5334. 20. Mancardi, D. A., B. Iannascoli, S. Hoos, P. England, M. Daeron, and P. Bruhns. 2008. Fc␥RIV is a mouse IgE receptor that resembles macrophage Fc␧RI in 40. Mizoguchi, A., E. Mizoguchi, R. N. Smith, F. I. Preffer, and A. K. Bhan. 1997.

␣ Downloaded from humans and promotes IgE-induced lung inflammation. J. Clin. Invest. 118: Suppressive role of B cells in chronic colitis of T cell receptor mutant mice. 3738–3750. J. Exp. Med. 186: 1749–1756. 21. Hudrisier, D., A. Aucher, A. L. Puaux, C. Bordier, and E. Joly. 2007. Capture of 41. Neville, M. E., and H. W. Lischner. 1981. Activation of Fc receptor-bearing target cell membrane components via trogocytosis is triggered by a selected set lymphocytes by immune complexes. II. Killer lymphocytes mediate Fc ligand- of surface molecules on T or B cells. J. Immunol. 178: 3637–3647. induced lymphokine production. J. Exp. Med. 154: 1868–1880. 22. Daubeuf, S., A. Puaux, E. Joly, and D. Hudrisier. 2007. A simple trogocytosis- 42. Neville, M. E., and H. W. Lischner. 1982. Activation of Fc receptor-bearing based method to detect, quantify, characterize and purify antigen-specific live lymphocytes by immune complexes. I. Stimulation of lymphokine production by lymphocytes by flow cytometry, via their capture of membrane fragments from nonadherent human peripheral blood lymphocytes. J. Immunol. 128: 1063–1069.

antigen-presenting cells. Nat. Protoc. 1: 2536–2542. 43. Liu, C. P., W. J. Lin, M. Huang, J. W. Kappler, and P. Marrack. 1997. Devel- http://www.jimmunol.org/ 23. Yu, Y. Y., N. Netuschil, L. Lybarger, J. M. Connolly, and T. H. Hansen. 2002. opment and function of T cells in T cell antigen receptor/CD3␨ knockout mice Cutting edge: single-chain trimers of MHC class I molecules form stable struc- reconstituted with Fc␧RI␥. Proc. Natl. Acad. Sci. USA 94: 616–621. tures that potently stimulate antigen-specific T cells and B cells. J. Immunol. 168: 44. Shores, E., V. Flamand, T. Tran, A. Grinberg, J. P. Kinet, and P. E. Love. 1997. 3145–3149. Fc␧RI␥ can support T cell development and function in mice lacking endogenous 24. Vanherberghen, B., K. Andersson, L. M. Carlin, E. N. Nolte-’t Hoen, TCR ␨-chain. J. Immunol. 159: 222–230. G. S. Williams, P. Hoglund, and D. M. Davis. 2004. Human and murine inhib- 45. Amigorena, S., C. Bonnerot, J. Drake, D. Choquet, W. Hunziker, J. Guillet, itory receptors transfer from natural killer cells to target cells. P. Webster, C. Sautes, I. Mellman, and W. Fridman. 1993. Cytoplasmic domain Proc. Natl. Acad. Sci. USA 101: 16873–16878. of IgG Fc receptors in B lymphocytes. Science 256: 1808–1812. 25. Williams, G. S., L. M. Collinson, J. Brzostek, P. Eissmann, C. R. Almeida, 46. Coxon, A., X. Cullere, S. Knight, S. Sethi, M. W. Wakelin, G. Stavrakis, F. E. McCann, D. Burshtyn, and D. M. Davis. 2007. Membranous structures F. W. Luscinskas, and T. N. Mayadas. 2001. Fc␥RIII mediates neutrophil re- transfer cell surface proteins across NK cell immune synapses. Traffic 8: cruitment to immune complexes: a mechanism for neutrophil accumulation in 1190–1204. immune-mediated inflammation. Immunity 14: 693–704. by guest on September 28, 2021 26. Garside, P., E. Ingulli, R. R. Merica, J. G. Johnson, R. J. Noelle, and 47. Bourbie´-Vaudaine, S., N. Blanchard, C. Hivroz, and P.-H. Rome´o. 2006. Den- M. K. Jenkins. 1998. Visualization of specific B and T interactions dritic cells can turn CD4ϩ T lymphocytes into vascular endothelial growth factor- in the lymph node. Science 281: 96–99. carrying cells by intercellular neuropilin-1 transfer. J. Immunol. 177: 1460–1469. 27. Tosi, M., and H. Zakem. 1992. Surface expression of Fc␥ receptor III (CD16) on 48. Tabiasco, J., A. Vercellone, F. Meggetto, D. Hudrisier, P. Brousset, and chemoattractant-stimulated neutrophils is determined by both surface shedding J. J. Fournie. 2003. Acquisition of viral receptor by NK cells through immuno- and translocation from intracellular storage compartments. J. Clin. Invest. 90: logical synapse. J. Immunol. 170: 5993–5998. 462–470. 28. Ravetch, J. V., and S. Bolland. 2001. IgG Fc receptors. Annu. Rev. Immunol. 19: 49. Russo, V., D. Zhou, C. Sartirana, P. Rovere, A. Villa, S. Rossinin, C. Traversari, 275–290. and C. Bordignon. 2000. Acquisition of intact allogeneic human leukocyte anti- 29. Basten, A., J. F. Miller, N. L. Warner, R. Abraham, E. Chia, and J. Gamble. 1975. gen molecules by human dendritic cells. Blood 95: 3473–3477. A subpopulation of T cells bearing Fc receptors. J. Immunol. 115: 1159–1165. 50. Patel, D., R. Dudek, and M. Mannie. 2001. Intercellular exchange of class II 30. Krammer, P. H., L. Hudson, and J. Sprent. 1975. Fc-receptors, Ia-antigens, and MHC complexes: ultrastructural localization and functional presentation of ad- immunoglobulin on normal and activated mouse T lymphocytes. J. Exp. Med. sorbed I-A/peptide complexes. Cell. Immunol. 214: 21–34. 142: 1403–1415. 51. Chernomordik, L., and M. Kozlov. 2003. Protein-lipid interplay in fusion and 31. Stout, R. D., and L. A. Herzenberg. 1975. The Fc receptor on thymus-derived fission of biological membranes. Annu. Rev. Biochem. 72: 175–207. lymphocytes. I. Detection of a subpopulation of murine T lymphocytes bearing 52. Beum, P., D. Mack, A. Pawluckowycz, M. Lindorfer, and R. Taylor. 2008. Bind- the Fc receptor. J. Exp. Med. 142: 611–621. ing of rituximab, trastuzumab, cetuximab or mAb T101 to cancer cells promotes 32. Thomson, C. W., W. A. Teft, W. Chen, B. P. Lee, J. Madrenas, and L. Zhang. trogocytosis mediated by THP-1 cells and monocytes. J. Immunol. 181: 2006. FcR␥ presence in TCR complex of double-negative T cells is critical for 8120–8132. their regulatory function. J. Immunol. 177: 2250–2257. 53. Li, Y., M. Williams, J. Cousar, A. Pawluczkowycz, M. Lindorfer, and R. Taylor. 33. Titus, J. A., F. D. Finkelman, D. A. Stephany, J. F. Jones, and D. M. Segal. 1984. 2007. Rituximab-CD20 complexes are shaved from Z138 mantle cell lymphoma Quantitative analysis of Fc␥ receptors on murine spleen cell populations by using cells in intravenous and subcutaneous SCID mouse models. J. Immunol. 179: dual parameter flow cytometry. J. Immunol. 133: 556–561. 4263–4271.