Naive T Cells Are Resistant to Anergy Induction by Anti-CD3 Fabienne Andris, Sébastien Denanglaire, Fabrizio de Mattia, Jacques Urbain and Oberdan Leo This information is current as of May 10, 2019. J Immunol 2004; 173:3201-3208; ; doi: 10.4049/jimmunol.173.5.3201 http://www.jimmunol.org/content/173/5/3201 Downloaded from References This article cites 48 articles, 25 of which you can access for free at: http://www.jimmunol.org/content/173/5/3201.full#ref-list-1

Why The JI? Submit online. http://www.jimmunol.org/ • Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on May 10, 2019 Subscription Information about subscribing to The Journal of 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 © 2004 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Naive T Cells Are Resistant to Anergy Induction by Anti-CD3 Antibodies1

Fabienne Andris,2 Se´bastien Denanglaire, Fabrizio de Mattia,3 Jacques Urbain, and Oberdan Leo

Anti-CD3 mAbs are potent immunosuppressive agents used in clinical transplantation. It has been generally assumed that one of the anti-CD3 mAb-mediated tolerance mechanisms is through the induction of naive unresponsiveness, often referred to as anergy. We demonstrate in this study that naive T cells stimulated by anti-CD3 mAbs both in vivo and in vitro do not respond to the staphylococcal enterotoxin B nor to soluble forms of anti-CD3 mAbs and APC, but express increased reactivity to plastic-coated forms of the same anti-CD3 mAbs and to their nominal Ag/class II MHC, a finding that is difficult to rationalize with the concept of anergy. Phenotypic and detailed kinetic studies further suggest that a strong signal 1 delivered by anti-CD3 mAbs in the absence of costimulatory molecules does not lead to anergy, but rather induces naive T cells to change their mitogen responsiveness and acquire features of memory T cells. In marked contrast, Ag-experienced T cells are sensitive to anergy Downloaded from induction under the same experimental settings. Collectively, these studies demonstrate that exposure of naive T cells in vivo and in vitro to a strong TCR stimulus does not induce Ag unresponsiveness, indicating that sensitivity to negative signaling through .TCR/CD3 triggering is developmentally regulated in CD4؉ T cells. The Journal of Immunology, 2004, 173: 3201–3208

ptimal T cell stimulation requires Ag recognition by the anergy was induced in vitro following triggering of naive T cells

TCR (signal 1) and signals provided by the APC (col- with anti-CD3 mAbs (13–15). http://www.jimmunol.org/ O lectively referred to as signal 2 or costimulatory signals) Anti-CD3 mAb is a potent immunosuppressive drug altering T (1, 2). An important concept that emerged from studies performed cell functions in vivo that is widely used for the prevention of mostly on cloned murine and human CD4ϩ cells is that provision allograft rejection or the treatment of autoimmune diseases (16). of signal 1 in the absence of costimulation does not represent a The precise mechanism of T cell unresponsiveness induced by neutral event, but rather is perceived by T cells as a negative stim- anti-CD3 mAbs remains poorly defined. Among others, it has been ulus, leading to a state of cell-autonomous unresponsiveness re- proposed that anti-CD3 mAbs induced T cell anergy in vivo by ferred to as anergy (3, 4). As a consequence, negative signaling by delivering a partial T cell activation signal (16–18). Ag recognition in the absence of adequate accessory cell functions Because the ability to induce T cell anergy in naive is thought to represent a powerful mechanism for inducing periph- is of both theoretical and clinical importance, we wished to inves- by guest on May 10, 2019 eral cell tolerance (5, 6). Central to this proposed mechanism of tigate the consequences of anti-CD3-mediated signaling on resting is the concept that naive CD4ϩ T cells (thymic naive T cells, both in vivo and in vitro. We demonstrate in this emigrant cells that have not encountered Ag) are sensitive to an- work that naive T cells are not sensitive to anergy induction by ergy induction. Although work performed on clonal cell popula- anti-CD3 mAbs. Anti-CD3-treated cells do not respond to conven- tions has clearly demonstrated the signaling dichotomy of the TCR tional in vitro polyclonal mitogens such as soluble anti-CD3 mAbs (activation vs anergy induction) (7–9), the extent to which TCR or bacterial , but respond to nominal Ag/MHC com- signals induce T cell anergy in naive populations has been a matter plexes. Thus, lack of response to noncognate classical T cell mi- of debate. Naive T cells were found to be resistant to anergy in- togens does not result from anergy induction, but rather reveals a duction following exposure to peptide-MHC complexes in the ab- change in mitogen responsiveness that is characteristic of memory- sence of costimulation (10–12). These reports contrast with stud- type cells. Collectively, these observations demonstrate that in ies performed in both human and murine models in which T cell contrast to Ag-experienced T cells, Ab-mediated TCR ligation in the absence of costimulation does not induce anergy in naive CD4ϩ T cells. Laboratoire de Physiologie Animale, Institut de Biologie et de Me´decine Mole´cu- laires, Universite´Libre de Bruxelles, Gosselies, Belgium Received for publication July 10, 2003. Accepted for publication June 30, 2004. Materials and Methods The costs of publication of this article were defrayed in part by the payment of page Medium and reagents charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. The medium in all experiments was RPMI 1640 supplemented with 5% 1 This work was supported by the Belgian Program in Interuniversity Poles of At- FCS, penicillin, streptomycin, glutamine, nonessential amino acids, and ϫ Ϫ5 traction initiated by the Belgian State, Prime Minister’s office, Science Policy Pro- 5 10 M 2-ME. The chicken OVA peptide (OVA223–239) was synthe- gramming, and by a Research Concerted Action of Communaute´ Franc¸aise de sized by Neosystem (Strasbourg, France). Superantigen staphylococcal en- Belgique. terotoxin B (SEB)4 was purchased from Toxin Technology (Sarasota, FL). 2 Address correspondence and reprint requests to Dr. Fabienne Andris, Laboratoire de Ionomycin and PMA were obtained from Sigma-Aldrich (St. Louis, MO). Physiologie Animale, Institut de Biologie et de Me´decine Mole´culaires, Universite´ The hamster mAb 37.51 to murine CD28 (19) was kindly provided by J. Libre de Bruxelles, Rue Pr Jeener et Brachet 12, 6041 Gosselies, Belgium. E-mail Allison (University of California, Berkeley, CA). address: [email protected] 3 Current address: 162, Department of Cell Physiology, Nijmegen Center for Molec- ular Life Sciences, University Medical Center, St. Radboud, P.O. Box 9101, 6500 HB 4 Abbreviations used in this paper: SEB, superantigen staphylococcal enterotoxin B; Nijmegen, The Netherlands. SSC, side light scatter.

Copyright © 2004 by The American Association of Immunologists, Inc. 0022-1767/04/$02.00 3202 ANTI-CD3 mAbs DO NOT INDUCE ANERGY IN NAIVE T CELLS

Cell lines and mice report of Hashemi et al. (28). Polystyrene 6.0-␮m beads (Polysciences, Warrington, PA) were incubated overnight with 10 ␮g/ml anti-CD3 mAbs Six- to 8-wk-old BALB/c mice (Harlan, Horst, The Netherlands) and or PBS (4 ϫ 107 beads/ml) on a rotator at 4°C. Beads were then stored at DO11.10 mice (20) were maintained in a -free environment. The 4°C in the suspension with the Ab or PBS for extended periods of time, DO11.10 T cell line was derived following three rounds of biweekly stim- extensively washed to eliminate unbound Ab, and resuspended at 2 ϫ 107 ulations of T cells from the DO11.10 transgenic mice by the OVA223–239 beads/ml before being used in the experimental test. peptide and APC, as described (21). Aliquots of Fluo-3-loaded T cells were warmed to 37°C 5 min before 5 Cell purifications and in vitro treatments each experiment. Typically, 400 ␮l of cells (8 ϫ 10 cells) were mixed with 40 ␮l of beads (to yield to 1:1 T cell-bead ratio), centrifuged at 1000 CD4ϩ T cells were purified from naive animals by magnetic depletion of rpm for 30 s, then gently resuspended and incubated for 30 s at 37°C before B cells, , dendritic cells, NK cells, , and CD8ϩ T being analyzed by flow cytometry. We used a forward light scatter/side cells. Spleen cells (2 ϫ 107/ml) were incubated (30 min at 4°C) in culture light scatter (SSC) gating procedure to separate the three populations of medium containing 1 ␮g/ml of the following biotinylated mAbs: anti- free cells, free beads, and cell-bead conjugates (as shown in Fig. 8). The CD19, anti-pan NK, anti-GR1 (all from BD Biosciences, Mountain View, free beads were smaller than T cells and displayed a strong SSC signal, due CA), anti-CD11c (clone N418), anti-I-Ed (clone 14.4.4), and anti-CD8 to their opacity. When anti-CD3-coated beads were added to T cells, a third (clone H35). The cells were then centrifuged and resuspended in 70 ␮lof population appeared in the sample mixture that had the size of the T lym- avidin-conjugated magnetic beads (Miltenyi Biotec, Bergisch-Gladbach, phocytes and the SSC characteristic of the beads. Our condition of mixing Germany)/108 spleen cells, for 30 min at 4°C before centrifugation and cells and beads typically yielded a maximum of ϳ20% of the total number separation over an autoMACS column (Miltenyi Biotec) using the De- of cells in the form of cell-bead conjugates. We chose to keep this number pleteS program, according to the manufacturer’s recommendations. The low to minimize the occurrence of the more complex situation of multiple percentage of CD4ϩ T cells obtained ranged between 95 and 98%, as cell-bead aggregates that are difficult to interpret and also clog the cytometer. estimated by flow cytometry in all experiments. Fluo-3 emission was measured on a linear scale in the free-cell and Feeder cells were obtained following complement-mediated T cell de- cell-bead doublet gates. The fluorescence 1 signal for Fluo-3 was calibrated Ϫ 2ϩ pletion using an anti-Thy-1 mAb (clone HO13.4). Thy-1 cells (containing by saturating Ca with 150 ␮g/ml ionomycin to obtain the maximum Downloaded from 2ϩ B cells, dendritic cells, and macrophages) were irradiated at 1200 rad (from signal (Fmax) and then adding Mn (2 mM MnCl2) to obtain the mini- 2ϩ 2ϩ a cobalt source) before use. mum signal (Fmin). The intracellular Ca concentration ([Ca ]i) can be Purified CD4ϩ cells (1.5 ϫ 106/ml) were incubated for 2 days in com- calculated from the Fluo-3 fluorescence intensity using the formula derived 2ϩ ϭ ϭ plete medium in 24-well plates coated with anti-CD3 mAbs (clone 145- by Tsien et al. (29): [Ca ]i Kd (F –Fmin)/(Fmax –F), where Kd 400 2C11, 10 ␮g/ml) (22) and then rested for 2 days in medium alone. Viable nM for the Fluo-3 intracellular dye. cells were recovered by centrifugation on Lympholyte M solution (Cedar- lane Laboratories, Hornby, Canada) and restimulated (0.5–2 ϫ 105/ml in 200 ␮l of 96-well plates) in the presence of serial dilutions of plate-bound Results http://www.jimmunol.org/ anti-CD3 mAbs or in the presence of freshly isolated syngenic feeder cells Anti-CD3 mAbs do not induce naive T cell anergy in vivo and graded doses of anti-CD3 mAbs, SEB, or OVA223–239. Culture super- natants were tested 24 h later for IL-2 content (using an IL-2-sensitive To evaluate the effect of in vivo anti-CD3 mAb therapy on T cell ␥ VDA/2 cell line) and after 48 h for IFN- and IL-4 content by ELISA, as immunocompetence, equal numbers of purified CD4ϩ T cells iso- described (23, 24). Proliferation was assessed 48–72 h after the initiation of the culture by [3H]thymidine incorporation. As cells were cultured in the lated from control (-matched, Ab-treated) and anti-CD3- absence of exogenous growth factors, a progressive decline in cell viability treated mice were stimulated by graded doses of soluble anti-CD3 was observed in control cultures lacking anti-CD3 stimuli. However, cul- mAbs in the presence of syngeneic, Fc␥R-bearing APC or by ture conditions did not modify the ability of control populations to respond graded doses of plastic-absorbed anti-CD3 mAbs in the absence of to TCR-mediated stimulations, as cells cultured in the absence of agonists by guest on May 10, 2019 and freshly purified cells responded in similar fashion to receptor triggering accessory cells. As previously demonstrated by us and others (17, (15, 25) (data not shown). Therefore, in most experiments, freshly purified 25), anti-CD3 mAb administration in mice led to T cell unrespon- CD4ϩ T cells were used as unstimulated controls. siveness to a subsequent in vitro stimulation by soluble anti-CD3 ϩ In vivo treatments and accessory cells (Fig. 1A). The same CD4 T cell population retained, however, the ability to respond to plastic-coated anti- Naive animals were injected i.v. with 10 ␮g of anti-CD3 mAbs ascite or CD3 mAbs (Fig. 1B), indicating that these cells were fully capable control hamster mAbs (clone PARSI-19, generated in our laboratory) as- cite. Both ascite solutions were endotoxin free, as judged by the Limulus of transmitting activation signals, leading to cell proliferation amebocyte lysate assay. CD4ϩ T cells were purified and restimulated in through the TCR/CD3 receptor. Both cell populations responded to vitro 72 h later, as described in the previous section. pharmacological agents bypassing early receptor signaling (a com- Flow cytometry bination of ionomycin and PMA), further illustrating their func- tional proliferative competence (Fig. 1C). To evaluate the ability Specific cell surface staining was performed using standard procedures and of anti-CD3 mAbs to induce unresponsiveness to nominal Ag, analyzed with a FACSort cytometer (BD Biosciences). Cells were labeled 7 OVA-specific, TCR transgenic DO11.10 mice were inoculated with CFSE, as described (26). Briefly, T cells were resuspended at 10 /ml ϩ in serum-free medium in the presence of 10 mM CFSE and incubated for with control and anti-CD3 mAbs, as above. Purified CD4 T cells 10 min at 37°C. The reaction was stopped by adding 5 vol of ice-cold from both experimental groups were subjected to in vitro restimu- medium, and, after two washes, CFSE-labeled cells were used for in vitro lation in the presence of anti-CD3 mAbs or Ag/MHC complexes. cultures and FACS analysis. As previously shown, CD4ϩ T cells isolated from anti- Calcium mobilization assays CD3-treated mice responded poorly to soluble anti-CD3 mAbs and T cells were loaded with the Fluo-3 calcium-sensitive dye, as described APC (Fig. 1D). However, anti-CD3-treated T cells proliferated in (27). Briefly, cells were washed three times in calcium/magnesium-free response to the cognate OVA peptide Ag (Fig. 1E), indicating that HBSS (Invitrogen Life Technologies, Carlsbad, CA) and incubated at anti-CD3 therapy in vivo did not down-regulate T cell responsive- 107/ml with F-127 pluronic acid (100 ␮g/ml) and 5 ␮M Fluo-3 for 30 min ness to Ag. at 37°C. Cells were then washed twice in complete medium and stored Repeated anti-CD3 mAb injections in mice have been shown to (2 ϫ 106/ml) at 4°C in the dark before use. For stimulations with soluble anti-CD3 mAbs, Fluo-3-loaded T cells favor the development of Th2-like cells, a cell population display- (400 ␮l) were incubated with anti-CD3 mAbs (clone 7D6, 0.5 ␮g/ml) for ing reduced sensitivity to classical anergy induction (30). How- 2 min at 37°C, cross-linked by secondary rabbit anti-mouse Abs, and an- ever, in vitro restimulation of anti-CD3-treated CD4ϩ T cells by alyzed for Fluo-3 emission at 525 nm with the CellQuest software (BD plastic-coated anti-CD3 mAbs or OVA peptide/MHC complexes Biosciences), as previously described (25). ␥ For stimulations with plastic-coated anti-CD3 mAbs, we devised a led to sustained or increased production of both IFN- and IL-4 method for simultaneous measurement of cytoplasmic calcium for free (Fig. 1, F and G), suggesting that the anti-CD3 mAb in vivo treat- cells and cells conjugated with anti-CD3-coated plastic beads based on a ment did not induce a shift toward the Th2 phenotype. The Journal of Immunology 3203

FIGURE 2. Selective depletion of Ag-experienced T cells following anti- CD3 injection. DO11.10 transgenic mice were inoculated with OVA pep- tide and adjuvant. Five days later, one group of mice received control isotype (A), while the other was injected i.v. with anti-CD3 mAbs (B). Splenocytes were recovered on day 4 and triple stained with anti-CD4 FITC, KJ1.26 PE, and anti-CD62L-biotin/avidin-CyChrome Abs. Panels show CD62L vs KJ1.26ϩ expression on CD4ϩ-gated splenocytes. Percent-

age of gated cells in each quadrant is indicated. Downloaded from

injected in these mice, we reproducibly observed a loss of the CD62Llow transgenic (KJ1.26ϩ) CD4ϩ T cells, suggesting that anti-CD3 mAb treatment selectively depleted activated/memory T

(Fig. 2, compare lower right quadrants). Typically, anti-CD3 treat- http://www.jimmunol.org/ ment led to the loss of 50–90% of memory OVA-specific T cells, indicating that Ag-experienced T cells are extremely sensitive (compared with naive transgene-bearing T lymphocytes) to acti- vation-induced cell death in vivo. This conclusion was further sup- ported by transfer experiments in which BALB/c mice were re- constituted with naive or in vitro activated DO11.10 T cells before anti-CD3 treatment. Anti-CD3 mAb injection induced clonal ex- pansion of transferred naive transgenic cells, while the same treat-

ment caused massive depletion of the Ag-experienced cells (data by guest on May 10, 2019 not shown). The responsiveness status of the remaining Ag-activated anti- CD3-treated splenocytes was assessed following Ag restimulation FIGURE 1. Responsiveness of in vivo anti-CD3 mAb-treated CD4ϩ T ϩ in vitro. In the experiment shown in Fig. 3, naive and OVA-im- lymphocytes. AÐE, Proliferative response of CD4 T cells purified from munized transgenic mice were treated with anti-CD3 mAbs or iso- BALB/c (AÐC) or DO11.10 (D and E) mice inoculated 3 days before with type control mAbs, and CD4ϩ splenic T cells were tested for OVA control or anti-CD3 mAbs and restimulated in vitro (105 cells/well) with peptide responsiveness. As shown above, naive T cells treated in graded doses of: A and D, soluble anti-CD3 and T-depleted syngenic APC (2 ϫ 105 cells/well); B, plastic-coated anti-CD3 mAbs; C, calcium iono-

phore and PMA (10 ng/ml); and E, OVA223–239 and APC. Proliferation was evaluated after a 48-h culture, and results are expressed as cpm of [3H]thy- midine incorporation. F and G, secretion induced following re- stimulation of CD4ϩ T cells from control and anti-CD3-treated DO11.10 ␮ ␮ mice with coated anti-CD3 (10 g/ml) or OVA223–239 (1 g/ml) and APC. IFN-␥ (F) and IL-4 (G) culture supernatant titers are estimated by ELISA. Results are representative of 3 (F and G) to at least 10 (AÐC) individual experiments.

Anti-CD3 mAbs induce Ag-experienced T cell depletion and anergy in vivo To evaluate the effects of the anti-CD3 injection on Ag-experi- enced T cell responsiveness, we inoculated DO11.10 transgenic mice with the cognate OVA peptide Ag in adjuvant before treat- FIGURE 3. Anergy induction in Ag-experienced T cells following anti- ment with control or anti-CD3 mAbs. As shown in Fig. 2A, im- CD3 inoculation. DO11.10 mice were injected with saline (A)orOVA munization of transgenic mice with the OVA peptide Ag in adju- peptide (B). Five days later, one-half of the mice in each group were in- vant induced Ag-specific effector/memory T cell differentiation in oculated i.v. with anti-CD3 mAb or the control isotype mAb. Splenocytes vivo, as judged by reduced expression of the CD62L marker on a were tested on day 4 after Ab injection for their responsiveness to the OVA large fraction of the transgene-bearing T cells specifically (Fig. 2A, peptide Ag in vitro. Results are expressed as mean Ϯ SD of [3H]thymidine compare left and right quadrants). When anti-CD3 mAbs were incorporation (two mice/group). 3204 ANTI-CD3 mAbs DO NOT INDUCE ANERGY IN NAIVE T CELLS

vivo with anti-CD3 mAbs displayed a vigorous response to OVA peptide in vitro (Fig. 3A). In contrast, anti-CD3 mAb treatment led to a reduced response to the same Ag in immunized animals (Fig. 3B), suggesting that sensitivity to anergy induction in this exper- imental model could be a feature of Ag-experienced CD4ϩ T cells.

Anti-CD3 mAbs do not induce naive T cell anergy in vitro To further determine whether TCR triggering in the absence of costimulation was able to induce T cell anergy in naive T cells, purified CD4ϩ T cells from naive BALB/c or DO11.10 mice were cultured on anti-CD3 mAb-coated dishes, according to a protocol originally developed by Jenkins et al. (31). Naive CD4ϩ T cells were exposed to anti-CD3 mAbs for 2 days, allowed to rest for an additional period of 2 days in fresh medium, and assayed for im- munocompetence during secondary restimulations with a panel of mitogenic stimuli. Control cells were either cultured for 4 days in the presence of medium alone or freshly purified from naive syn- genic mice, as detailed in Materials and Methods. This anergy protocol was first validated using a DO11.10 T cell line derived in vitro following repetitive rounds of antigenic stimulation. Prein- Downloaded from cubation of the DO11.10 line with plate-bound anti-CD3 mAbs led to a global hyporesponsiveness characterized by a defective pro- liferation to both anti-CD3 mAb triggering (in soluble and coated forms; Fig. 4, A and B) and antigenic stimulation (Fig. 4C). In marked contrast, the same protocol failed to induce classical T cell

anergy in naive T cells of identical genetical background. In keep- http://www.jimmunol.org/ ing with the in vivo model previously described, in vitro anti-CD3- treated CD4ϩ T cells displayed a reduced response to soluble anti- CD3 mAbs (Fig. 5, A and B) and to the V␤8-specific SEB superantigen (Fig. 5E). These cells, however, proliferated vigor- ously to plastic-bound anti-CD3 mAbs (note the shift response of pretreated vs control cells toward lower activation threshold in Fig. 5, C and D) and in response to the cognate OVA/MHC complexes (Fig. 5F). Moreover, restimulation of preactivated cells in the pres- ence of OVA peptide or insolubilized anti-CD3 led to increased by guest on May 10, 2019

FIGURE 5. Responsiveness of plastic-coated anti-CD3 mAb-treated CD4ϩ T lymphocytes. AÐF, Proliferative response of CD4ϩ T cells (105/ well) pretreated for 2 days with plastic-coated anti-CD3 mAbs (10 ␮g/ml) and control cells to graded doses of: A and B, soluble anti-CD3 and T- depleted syngenic APC (2 ϫ 105 cells/well); C and D, plastic-coated anti- ϩ CD3 mAbs; E, SEB and APC; and F, OVA223–239 and APC. CD4 T cells were purified from BALB/c (A, C, and E) or DO11.10 (B, D, and F) mice. Proliferation was evaluated after a 48-h culture, and results are expressed as cpm of [3H]thymidine incorporation. GÐI, Cytokine secretion induced following restimulation of control and anti-CD3-treated CD4ϩ T cells from ␮ ␮ DO11.10 mice with coated anti-CD3 (10 g/ml) or OVA223–239 (1 g/ml) and APC. IL-2 (G), IFN-␥ (H), and IL-4 (I) culture supernatant titers are estimated, as described in Materials and Methods and Fig. 1. Results are representative of 3 (E and GÐI) to at least 10 (AÐD and F) individual experiments.

secretion of both Th1 (IL-2, IFN-␥) and Th2 (IL-4) (Fig. 5, GÐI). FIGURE 4. Anergy induction in a DO11.10 T cell line following anti- CD3-treatment. DO11.10-derived T cells were cultured for 2 days with Anti-CD3-treated CD4ϩ T cells display phenotypic and plastic-coated anti-CD3 mAbs (1 ␮g/ml) or medium. After a 2-day rest in functional characteristics of memory T cells fresh medium, cell lines (105/well) were restimulated with graded doses of the indicated stimuli (AÐD), and proliferation was assessed by [3H]thymi- The distinct ability of anti-CD3-treated cells to respond to different dine incorporation at the end of a 48-h culture. This experiment has been forms of anti-CD3 mAbs was reminiscent of early work performed reproduced three times with similar results. on memory T cells. Memory T cells differ from naive T cells by the The Journal of Immunology 3205 expression of a different set of surface markers (32, 33), a rapid production of polarized cytokines, and a rapid initiation of expan- sion after antigenic restimulation (34). Of relevance to this study, mouse memory CD4ϩ T cells have also been shown to be hypo- responsive to several noncognate TCR stimuli, including bacterial superantigens (35) and APC-directed anti-CD3 mAbs (36), a pat- tern of responsiveness very similar to the anti-CD3 mAb-treated cells described in the present study. Based on these considerations, we hypothesized that anti-CD3 treatment led to the differentiation of naive T cells into memory-like cells. The phenotypical analysis shown in Fig. 6 is in agreement with this hypothesis. Indeed, when compared with naive control lymphocytes, anti-CD3-treated CD4ϩ T cells expressed equivalent levels of the CD69 early ac- tivation markers and reduced levels of CD45RB and CD62L, a phenotype often associated with Ag-experienced cells. As memory T cells respond rapidly to Ag stimulation, a kinetic analysis of the proliferative response to distinct stimuli was performed. As a pos- itive control, a population of T cells previously exposed to a com- bination of anti-CD3 and anti-CD28 mAbs was included in this analysis. The response of naive T cells to all stimuli tested peaked Downloaded from at ϳdays 2–4 (Fig. 7, AÐC). In agreement with published evi- dence, anti-CD3/CD28 mAb-treated cells displayed a typical ef- fector cell profile, characterized by a fast response (peak responses between days 1 and 2, depending on the stimulus) and a higher efficacy (note the increased thymidine uptake when compared with

naive T cells). Anti-CD3 mAb-treated cells presented a unique http://www.jimmunol.org/ pattern of responsiveness. These cells proliferated with a typical secondary-type response (faster kinetics and increased thymidine uptake) to both the OVA cognate Ag and the plate-bound 323–339 FIGURE 7. Kinetic responses of control and anti-CD3 mAb-treated anti-CD3 mAbs (Fig. 7, C and B, respectively). Surprisingly, the ϩ CD4 T cells. AÐC, Control, plastic-coated anti-CD3- or plastic-coated response of anti-CD3-treated cells to soluble anti-CD3 peaked anti-CD3 ϩ anti-CD28-treated naive T cells were restimulated with: A, early, but faded rapidly when compared with the response of both soluble anti-CD3 mAbs (30 ng/ml) and syngenic APC; B, plastic-coated control and anti-CD3/CD28-stimulated cells (Fig. 7A). This early ␮ ϩ anti-CD3 mAbs (3 g/ml); or C, OVA223–239 (300 ng/ml) APC. Prolif- and transient [3H]thymidine incorporation suggested that anti- eration was measured at the indicated times by adding [3H]thymidine dur- CD3-treated cells underwent an abortive cell proliferation. This ingthelast6hoftheculture. D and E, Control and anti-CD3-treated cells by guest on May 10, 2019 observation was supported by the FACS analysis of cell divisions were loaded with CFSE and stimulated (2 ϫ 105 cells, in 96-well plate) for with the CFSE fluorescent dye, showing that anti-CD3-treated 4 days in the presence of APC and soluble anti-CD3 mAbs (30 ng/ml) CD4ϩ T cells did not undergo extensive cellular division in re- (solid lines) or medium (dotted lines) and analyzed for their fluorescence sponse to stimulation with APC-bound anti-CD3 mAbs, when content. Similar results were obtained in three independent experiments. compared with naive T cells (Fig. 7D).

Anti-CD3 treatment affects early TCR-signaling capacities in their TCR activation requirement. To further support this con- The previous observations indicated that rather than inducing an- tention at the level of early TCR signaling, control and anti-CD3- ergy, exposure of naive T cells to anti-CD3 mAbs led to a change treated cells were loaded with the Fluo-3 calcium detector dye, and their response to distinct forms of anti-CD3 mAbs was analyzed by flow cytometry (Fig. 8 and Table I). In particular, control and anti-CD3-treated cells were incubated with a plastic-coated form of anti-CD3 mAbs suitable for flow cytometry analysis (anti-CD3- coupled beads; see Materials and Methods for details). Preliminary studies showed that anti-CD3-treated cells proliferate strongly in response to anti-CD3 mAbs coated on the plastic beads (data not shown). A significant proportion of the CD4ϩ T cells associated in doublets with the anti-CD3-coated beads as shown in the forward light scatter/SSC dot plot in Fig. 8, A and B. There was no signif- icant difference in the proportion of cell-bead doublets obtained in the two T cell populations (ranging between 15 and 24% in three different experiments), and no doublet formation was observed with uncoated beads (data not shown). The mean fluorescence emitted by the Fluo-3 indicator dye was then compared in the single cell and cell-bead doublet gates of both T cell populations FIGURE 6. Surface phenotype of naive and anti-CD3-treated CD4ϩ T (Fig. 8, C and D). Calcium mobilization only occurred in T cells cells. Control (upper panels) and in vitro anti-CD3-treated cells (lower bound to Ab-coated beads. In agreement with the previous obser- panels) were stained with Abs to CD45RB, CD62L, CD69 (solid lines), or vations, anti-CD3-treated cells displayed an enhanced calcium re- control isotypes (dotted lines). sponse to anti-CD3-coated beads. As a control, calcium influx 3206 ANTI-CD3 mAbs DO NOT INDUCE ANERGY IN NAIVE T CELLS

Discussion The ability of the TCR to deliver negative signals is thought to represent an important mechanism for self/nonself discrimination in the periphery (2). Although it has been clearly established that signals delivered by the APC are required for optimal stimulation of naive T cells, the extent to which these cells are sensitive to anergy induction by TCR engagement in the absence of costimu- lation is still a matter of debate. A series of in vitro observations indicate that engagement of the TCR of a naive T cell by an Ag/ MHC complex in the absence of costimulation represents a neutral event (11, 12). In particular, naive T cells are resistant to anergy induction by Ag pulsed on 1-ethyl-3-(3-dimethylaminopropyl)- carbodiimide-fixed APC, a classical protocol leading to unrespon- siveness of T cell clones (37). Similarly, exposure of naive T cells FIGURE 8. Calcium response of control and anti-CD3-treated naive T to ionomycin and nonmitogenic anti-CD3 mAbs failed to induce T cells to plastic bead-insolubilized anti-CD3 mAbs. A and B, Illustration of cell unresponsiveness (38, 39), supporting the notion that these flow cytometric gating for cell population analysis. Lower box gates, Sin- gle cell gates; upper box gates, cell/anti-CD3-coated bead doublet gates. C cells are insensitive to multiple forms of anergy induction. These and D, CD4ϩ T cells were incubated for 1 min with beads coated with observations are in marked contrast with studies performed with ⑀ anti-CD3 mAbs, as described in Materials and Methods. Dot plots repre- mitogenic anti-CD3 mAbs, demonstrating the sensitivity of naive Downloaded from sent the fluorescence emission of the Fluo-3-loaded T cells in the single T cells to this form of anergy induction (14, 15). These observa- cell and cell-bead doublet gates. Results are representative of three differ- tions could be rationalized by assuming that naive T cells display ent experiments. a higher threshold for anergy induction, requiring a strong TCR signal in the absence of costimulation (such as provided by mito- genic anti-CD3 mAbs) to enter the anergic state.

upon addition of ionomycin was equivalent in both cell popula- These considerations led us to re-evaluate the functional conse- http://www.jimmunol.org/ tions (data not shown). We next compared the ability of naive and quences of TCR complex ligation by anti-CD3⑀ mAbs in naive anti-CD3-treated cells to increase intracellular calcium levels in CD4ϩ T cells both in vivo and in vitro. Using this approach, we response to distinct forms of anti-CD3 mAbs. Cells were stimu- demonstrate in this study that naive T cells are not sensitive to lated by anti-CD3-coated beads or by soluble anti-CD3 mAbs anergy induction by mitogenic anti-CD3⑀ mAbs. The major ob- cross-linked by a secondary anti-Ig reagent. As shown in Table I, servation from our study is that although anti-CD3-treated cells naive and anti-CD3-treated cells displayed distinct activation re- failed to respond to soluble anti-CD3⑀ mAbs in the presence of quirement. Naive T cells preferentially responded to soluble Ab costimulatory-bearing APCs, they retained the ability to proliferate forms, while anti-CD3-treated cells displayed enhanced responses and produce cytokines in response to nominal Ag (Figs. 1, 5, and to coated forms of the Ab. Thus, and in agreement with the in vitro 7). These data call for caution in extrapolation from the common by guest on May 10, 2019 cell-based studies (cell proliferation and cytokine production), anti- perception that failure to respond to soluble anti-CD3 mAbs re- CD3-treated cells are not unresponsive to TCR complex triggering, flects T cell unresponsiveness. Several reports have indeed dem- but rather display an altered activation profile when compared with onstrated that while naive T cells are optimally stimulated by sol- naive cells. uble anti-CD3 mAbs, memory cells preferentially respond to insoluble forms of the same Ab (36, 40). Similarly, other widely used T cell mitogens (including bacterial superantigens or lectins) Table I. Calcium influx in control and anti-CD3-treated cells upon have been shown to preferentially stimulate naive T cells, further restimulation with soluble or coated forms of anti-CD3 mAbs illustrating the differences in the integration of TCR signals be- tween T cell subsets (35, 41). These observations suggest that re- Stimulations sponsiveness to anti-CD3 mAbs reflects cell differentiation rather Anti-CD3- Anti-CD3-coated than immunocompetence, and suggest that conventional mitogens a b Experiment Pretreatment None anti-Ig beads (including lectins, bacterial superantigens, and anti-TCR complex 1 None 41c 1056 487 Abs) are not suited for the evaluation of T cell competence in Plastic-coated 154 754 1576 anergy-related studies. These observations suggest that some of the anti-CD3 conflicting results previously reported on naive T cell anergy (see 2 None 20 443 516 Refs. 14 and 15) may derive from inadequate methodology. Plastic-coated 95 289 1023 In contrast to studies performed with nominal Ag presented by anti-CD3 costimulatory-deficient APCs (37), stimulation of naive T cells by anti-CD3 mAbs cannot be considered as a neutral event. In vivo or 3 None 60 ndd 547 Plastic-coated 74 nd 1169 in vitro anti-CD3 triggering induced T cell blastogenesis (as shown anti-CD3 by cell size increase) and proliferation (as judged by thymidine

a Fluo-3-loaded CD4ϩ T cells were incubated with anti-CD3 mAbs for 2 min at uptake and CFSE-labeling studies; data not shown). Moreover, 37°C, cross-linked by secondary rabbit anti-Ig Abs. Calcium concentration was cal- both in vivo and in vitro anti-CD3-treated cells expressed reduced culated from the mean fluorescence at the peak level. b ϩ levels of CD45RB and CD62L molecules (Fig. 6, and data not Fluo-3-loaded CD4 T cells were incubated for 1 min with anti-CD3-coated ϩ beads, and calcium concentration was calculated from the mean fluorescence of the shown), a phenotype characteristic of Ag-experienced cells. CD4 cell-bead doublets. experienced T cells have been shown to represent a heterogeneous c Results are expressed as mean calcium concentration (see Materials and Methods). pool of lymphocytes composed of effector (recently activated) and d nd, Not done. long-lived memory T cells (42). Ahmadzadeh et al. (40) recently The Journal of Immunology 3207 developed an activation profile assay, using anti-CD3 and anti- unresponsiveness apparently develops following a productive re- genic stimuli, that clearly identified two subsets of Ag-experienced sponse (proliferation and cytokine secretion) and is readily revers- CD4ϩ T cells. Compared with naive T cells, both effector ible upon removal of the antigenic stimulus. The observation that (CD62Llow) and resting memory (CD62Lhigh) T cells responded the biochemical block in signal transduction appears to be different vigorously to their nominal Ag/MHC, but differed in their respon- between these two forms of unresponsiveness (see Table I in Ref. siveness to anti-CD3-mediated triggering. Effector cells prolifer- 4) concurs with the concept that these two phenomenon are bio- ated vigorously to both plastic-coated and APC-bearing anti-CD3 logically distinct. This conclusion may have important functional mAbs, while resting memory T cells exhibited hyporesponsiveness consequences in clinical studies. Indeed, anergizing protocols to anti-CD3 mAbs insolubilized on FcR-bearing APC. According based on acute exposure of naive T cells to a costimulatory-defi- to this view, anti-CD3/CD28-treated cells presented a typical ef- cient TCR stimulus (such as anti-CD3 mAbs or Ag-loaded, co- fector cell responsiveness (a higher efficacy with a faster kinetic to stimulatory-deficient APCs) may prove inefficient in inducing long all TCR stimuli tested; Fig. 7), while T cells treated with plate-bound lasting unresponsiveness in vivo. Rather, induction of adaptive tol- anti-CD3 mAbs alone displayed activation requirement often associ- erance may require previous activation of T cell in vivo (possibly ated with a resting memory T cell phenotype (Figs. 5 and 7). in a costimulatory-dependent fashion), followed by continuous ex- Although the mechanism responsible for the selective unrespon- posure to Ag. However, the functional status of these adapted cells siveness of anti-CD3-treated cells to soluble forms of the same Ab in vivo is still unclear, as despite defective proliferative responses, was beyond the scope of this study, we demonstrate in this study these cells acquire and retain cytolytic potential (see Bandeira et al. that this change in activation requirement is probably due to a (46) and Stamou et al. (47)), calling for caution in translating these modification in the signaling machinery, as anti-CD3-treated cells protocols in a clinical setting. were preferentially stimulated by coated forms of the Ab at an In conclusion, the present study extends previous observations Downloaded from early signaling step, as emphasized by studies of intracellular cal- and strongly suggests that naive T cells (with some possible ex- cium influx described in Table I. These studies are in keeping with ceptions; see Ref. 48) are not sensitive to T cell clonal anergy, as observations indicating that memory and naive cells differ in their defined above. Thus, the ultimate impact of the cell-autonomous, TCR-associated signaling responses (43). Nonresponsiveness of antiproliferative states (clonal anergy and/or adaptive tolerance) on memory CD4ϩ T cells to soluble anti-CD3⑀ mAb has been pre- peripheral tolerance is still unclear, and needs to be integrated with

viously linked to inhibitory signals delivered by CD4 upon MHC the emerging role of regulatory cells in containing the immune http://www.jimmunol.org/ class II interaction (36, 44, 45). This hypothesis is hard to reconcile response to self constituents. with the observations reported in this study, as: 1) anti-CD3- treated cells displayed a reduced response to soluble anti-CD3 Acknowledgments mAbs even when presented by APCs from MHC class II knockout We thank P. Veirman for animal care, and M. Moser and F. Bureau for mice (our unpublished observations), and 2) cells unresponsive to critical review of the manuscript. We dedicate this work to the memory of soluble anti-CD3 mAbs or SEB responded to Ag presented in the our colleague Laurent Conde Da Silva Fraga. context of MHC class II molecules (Figs. 1, 5, and 7). The biochemical mechanism responsible for the selective unre- References by guest on May 10, 2019 sponsiveness of anti-CD3-treated T cells to bacterial superantigens 1. Babbitt, B. P., P. M. Allen, G. Matsueda, E. Haber, and E. R. Unanue. 1985. has not been identified to date. In particular, based on the notion Binding of immunogenic peptides to Ia histocompatibility molecules. Nature that superantigens and peptide agonists interact with distinct re- 317:359. 2. Weaver, C. T., and E. R. Unanue. 1990. The costimulatory function of - gions of both the TCR and the MHC molecules, it is tempting to presenting cells. Immunol. Today 11:49. speculate that the composition of the macromolecular complex re- 3. Schwartz, R. H. 1990. A cell culture model for T clonal anergy. cruited by these ligands at the immune synapse may be slightly Science 248:1349. 4. Schwartz, R. H. 2003. T cell anergy. Annu. Rev. Immunol. 21:305. different, a hypothesis that is presently under investigation. 5. Schwartz, R. H. 1996. Models of T cell anergy: is there a common molecular Finally, cell-mixing studies failed to provide any support for a mechanism? J. Exp. Med. 184:1. 6. Lechler, R., J. G. Chai, F. Marelli-Berg, and G. Lombardi. 2001. The contribu- regulatory influence of anti-CD3-treated cells upon naive T cells, tions of T-cell anergy to peripheral T-cell tolerance. Immunology 103:262. indicating that the altered responsiveness to anti-CD3 mAbs re- 7. Jenkins, M. K., and R. H. Schwartz. 1987. by chemically flected a cell-autonomous event (data not shown). modified splenocytes induces antigen-specific T cell unresponsiveness in vitro and in vivo. J. Exp. Med. 165:302. Collectively, our observations indicate that anti-CD3 stimula- 8. Gimmi, C. D., G. J. Freeman, J. G. Gribben, G. Gray, and L. M. Nadler. 1993. tion in the absence of costimulation does not induce Ag unrespon- Human T-cell clonal anergy is induced by antigen presentation in the absence of siveness in naive T cells, but favors the differentiation of naive T costimulation. Proc. Natl. Acad. Sci. USA 90:6586. 9. Sloan-Lancaster, J., B. D. Evavold, and P. M. Allen. 1993. Induction of T-cell cells toward a memory-like phenotype. anergy by altered T-cell-receptor ligand on live antigen-presenting cells. Nature The conclusion that naive T cells are resistant to several aner- 363:156. gizing protocols in vitro is at variance with numerous observations 10. Sagerstrom, C. G., E. M. Kerr, J. P. Allison, and M. M. Davis. 1993. Activation and differentiation requirements of primary T cells in vitro. Proc. Natl. Acad. Sci. suggesting that T cells can be rendered unresponsive in vivo. A USA 90:8987. recent proposal by Schwartz (4) may help resolve this apparent 11. Perez, V. L., L. Van Parijs, A. Biuckians, X. X. Zheng, T. B. Strom, and paradox. Accordingly, T cell unresponsiveness can be classified A. K. Abbas. 1997. Induction of peripheral T cell tolerance in vivo requires CTLA-4 engagement. 6:411. into two broad categories. T cell clonal anergy can be defined as a 12. Ragazzo, J. L., M. E. Ozaki, L. Karlsson, P. A. Peterson, and S. R. Webb. 2001. growth arrest state that develops following acute exposure of T Costimulation via lymphocyte function-associated antigen 1 in the absence of CD28 ligation promotes anergy of naive CD4ϩ T cells. Proc. Natl. Acad. Sci. cells to an inadequate stimulus (such as a TCR ligand in the ab- USA 98:241. sence of costimulatory signals). In most situations, this anergic 13. Wolf, H., Y. Muller, S. Salmen, W. Wilmanns, and G. Jung. 1994. Induction of state is long lasting and persists in the absence of further TCR anergy in resting human T lymphocytes by immobilized anti-CD3 antibodies. Eur. J. Immunol. 24:1410. engagement. Adaptive tolerance has been proposed to unfold in 14. Chai, J. G., and R. I. Lechler. 1997. Immobilized anti-CD3 mAb induces anergy vivo following chronic exposure of naive T cells to a persisting Ag in murine naive and memory CD4ϩ T cells in vitro. Int. Immunol. 9:935. in the absence of inflammation. In this setting, naive T cells re- 15. Frauwirth, K. A., M. L. Alegre, and C. B. Thompson. 2000. Induction of T cell anergy in the absence of CTLA-4/B7 interaction. J. Immunol. 164:2987. spond vigorously to the Ag, but gradually down-regulate their pro- 16. Chatenoud, L. 2003. CD3-specific -induced active tolerance: from bench liferative response and cytokine production capacity. This form of to bedside. Nat. Rev. Immunol. 3:123. 3208 ANTI-CD3 mAbs DO NOT INDUCE ANERGY IN NAIVE T CELLS

17. Hirsch, R., M. Eckhaus, H. Auchincloss, Jr., D. H. Sachs, and J. A. Bluestone. 32. Lee, W. T., X. M. Yin, and E. S. Vitetta. 1990. Functional and ontogenetic 1988. Effects of in vivo administration of anti-T3 on T cell analysis of murine CD45Rhi and CD45Rlo CD4ϩ T cells. J. Immunol. 144:3288. function in mice. I. Immunosuppression of transplantation responses. J. Immunol. 33. Bradley, L. M., G. G. Atkins, and S. L. Swain. 1992. Long-term CD4ϩ memory 140:3766. T cells from the spleen lack MEL-14, the lymph node homing receptor. J. Im- 18. Chau, L. A., J. Y. Tso, J. Melrose, and J. Madrenas. 2001. HuM291(Nuvion), a munol. 148:324. humanized Fc receptor-nonbinding antibody against CD3, anergizes peripheral 34. Rogers, P. R., C. Dubey, and S. L. Swain. 2000. Qualitative changes accompany blood T cells as partial agonist of the T cell receptor. Transplantation 71:941. memory T cell generation: faster, more effective responses at lower doses of 19. Harding, F. A., J. G. McArthur, J. A. Gross, D. H. Raulet, and J. P. Allison. 1992. antigen. J. Immunol. 164:2338. CD28-mediated signalling co-stimulates murine T cells and prevents induction of 35. Lee, W. T., and E. S. Vitetta. 1992. Memory T cells are anergic to the superan- anergy in T-cell clones. Nature 356:607. tigen staphylococcal enterotoxin B. J. Exp. Med. 176:575. 20. Murphy, K. M., A. B. Heimberger, and D. Y. Loh. 1990. Induction by antigen of 36. Farber, D. L., M. Luqman, O. Acuto, and K. Bottomly. 1995. Control of memory ϩ ϩ lo intrathymic apoptosis of CD4 CD8 TCR in vivo. Science CD4 T cell activation: MHC class II molecules on APCs and CD4 ligation inhibit 250:1720. memory but not naive CD4 T cells. Immunity 2:249. 21. De Mattia, F., S. Chomez, F. Van Laethem, V. Moulin, J. Urbain, M. Moser, 37. Hayashi, R. J., D. Y. Loh, O. Kanagawa, and F. Wang. 1998. Differences between O. Leo, and F. Andris. 1999. Antigen-experienced T cells undergo a transient responses of naive and activated T cells to anergy induction. J. Immunol. 160:33. phase of unresponsiveness following optimal stimulation. J. Immunol. 163:5929. 38. Davis, L. S., and P. E. Lipsky. 1993. Tolerance induction of human CD4ϩ T 22. Leo, O., M. Foo, D. H. Sachs, L. E. Samelson, and J. A. Bluestone. 1987. Iden- cells: markedly enhanced sensitivity of memory versus naive T cells to peripheral tification of a monoclonal antibody specific for a murine T3 polypeptide. Proc. anergy. Cell. Immunol. 146:351. Natl. Acad. Sci. USA 84:1374. 39. Smith, J. A., J. Y. Tso, M. R. Clark, M. S. Cole, and J. A. Bluestone. 1997. 23. De Smedt, T., M. Van Mechelen, G. De Becker, J. Urbain, O. Leo, and M. Moser. Nonmitogenic anti-CD3 monoclonal antibodies deliver a partial T cell receptor 1997. Effect of interleukin-10 on maturation and function. Eur. signal and induce clonal anergy. J. Exp. Med. 185:1413. J. Immunol. 27:1229. 40. Ahmadzadeh, M., S. F. Hussain, and D. L. Farber. 2001. Heterogeneity of the 24. Maldonado-Lopez, R., T. De Smedt, P. Michel, J. Godfroid, B. Pajak, ϩ memory CD4 T cell response: persisting effectors and resting memory T cells. C. Heirman, K. Thielemans, O. Leo, J. Urbain, and M. Moser. 1999. CD8␣ and Ϫ J. Immunol. 166:926. CD8␣ subclasses of dendritic cells direct the development of distinct T helper hi - cells in vivo. J. Exp. Med. 189:587. 41. Lerner, A., T. Yamada, and R. A. Miller. 1989. Pgp-1 T lymphocytes accumu late with age in mice and respond poorly to concanavalin A. Eur. J. Immunol. 25. Andris, F., M. Van Mechelen, F. De Mattia, E. Baus, J. Urbain, and O. Leo. 1996. Downloaded from 19:977. Induction of T cell unresponsiveness by anti-CD3 antibodies occurs indepen- dently of co-stimulatory functions. Eur. J. Immunol. 26:1187. 42. Farber, D. L., and M. Ahmadzadeh. 2002. Dissecting the complexity of the mem- 26. Weston, S. A., and C. R. Parish. 1990. New fluorescent dyes for lymphocyte ory T cell response. Immunol. Res. 25:247. 43. Hall, S. R., B. M. Heffernan, N. T. Thompson, and W. C. Rowan. 1999. migration studies: analysis by flow cytometry and fluorescence microscopy. ϩ ϩ ϩ ϩ J. Immunol. Methods 133:87. CD4 CD45RA and CD4 CD45RO T cells differ in their TCR-associated 27. Vandenberghe, P. A., and J. L. Ceuppens. 1990. Flow cytometric measurement of signaling responses. Eur. J. Immunol. 29:2098. cytoplasmic free calcium in human peripheral blood T lymphocytes with fluo-3, 44. Metz, D. P., D. L. Farber, R. Konig, and K. Bottomly. 1997. Regulation of a new fluorescent calcium indicator. J. Immunol. Methods 127:197. memory CD4 T cell adhesion by CD4-MHC class II interaction. J. Immunol. 28. Hashemi, B. B., J. P. Slattery, D. Holowka, and B. Baird. 1996. Sustained T cell 159:2567. http://www.jimmunol.org/ receptor-mediated Ca2ϩ responses rely on dynamic engagement of receptors. 45. Metz, D. P., and K. Bottomly. 1999. Function and regulation of memory CD4 T J. Immunol. 156:3660. cells. Immunol. Res. 19:127. 29. Tsien, R. Y., T. Pozzan, and T. J. Rink. 1982. Calcium homeostasis in intact 46. Bandeira, A., J. Mengel, O. Burlen-Defranoux, and A. Coutinho. 1991. Prolif- lymphocytes: cytoplasmic free calcium monitored with a new, intracellularly erative T cell anergy to MIs-1a does not correlate with in vivo tolerance. Int. trapped fluorescent indicator. J. Cell Biol. 94:325. Immunol. 3:923. 30. Wissing, M., M. L. Alegre, D. Abramowicz, F. Willems, V. Donckier, 47. Stamou, P., J. de Jersey, D. Carmignac, C. Mamalaki, D. Kioussis, and M. L. Vanderhaeghen, O. Leo, and M. Goldman. 1995. Emergence of Th2-like B. Stockinger. 2003. Chronic exposure to low levels of antigen in the periphery cells after anti-CD3 monoclonal antibody administration in mice. Transplant. causes reversible functional impairment correlating with changes in CD5 levels Proc. 27:378. in monoclonal CD8 T cells. J. Immunol. 171:1278. 31. Jenkins, M. K., C. A. Chen, G. Jung, D. L. Mueller, and R. H. Schwartz. 1990. 48. Thomas, S., R. Kumar, A. Preda-Pais, S. Casares, and T.-D. Brumeanu. 2003. A Inhibition of antigen-specific proliferation of type 1 murine T cell clones after model for antigen-specific T-cell anergy: displacement of CD4–p56lck signalo- by guest on May 10, 2019 stimulation with immobilized anti-CD3 monoclonal antibody. J. Immunol. some from the lipid rafts by a soluble, dimeric peptide-MHC class II chimera. 144:16. J. Immunol. 170:5981.