Subsets Antibodies Differentially Regulate Individual Th Fcr-Nonbinding Anti-CD3 Monoclonal Partial TCR Signals Delivered By

Partial TCR Signals Delivered by FcR-Nonbinding Anti-CD3 Monoclonal Antibodies Differentially Regulate Individual Th Subsets This information is current as of September 24, 2021. Judith A. Smith, Qizhi Tang and Jeffrey A. Bluestone J Immunol 1998; 160:4841-4849; ; http://www.jimmunol.org/content/160/10/4841 Downloaded from

References This article cites 38 articles, 19 of which you can access for free at: http://www.jimmunol.org/content/160/10/4841.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 September 24, 2021

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 1998 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Partial TCR Signals Delivered by FcR-Nonbinding Anti-CD3 Monoclonal Antibodies Differentially Regulate Individual Th Subsets1

Judith A. Smith, Qizhi Tang, and Jeffrey A. Bluestone2

Anti-CD3 mAbs with low FcR affinity prolong graft survival in the absence of the cytokine-mediated toxicity observed with conventional anti-CD3 treatment. Previous studies have shown that FcR-nonbinding anti-CD3 mAbs suppress immune responses, at least in part, by delivering a partial signal resulting in Th1 unresponsiveness. In this study, the biochemical and functional consequences of FcR-nonbinding anti-CD3 treatment for various activated T cell populations were examined. In contrast to Th1 cells, FcR-nonbinding anti-CD3-treated Th2 cells secreted IL-4 and proliferated. Furthermore, Th2 cells cultured with the mAb were not rendered unresponsive. Mixed “Th0” populations responded to FcR-nonbinding anti-CD3 by producing IL-4, and Downloaded from showed a selective decrease in IL-2 production following preculture with the mAb. The stimulation of IL-4-producing cells did not reflect a more complete TCR signal, since similar defects in ␨, ZAP-70, and MAP kinase phosphorylation were observed in Th1 and Th2 cells. Despite the proximal signaling defects, FcR-nonbinding anti-CD3 induced nuclear translocation of NF-ATc. Thus, Abs that deliver partial TCR signals may promote development of a Th2 phenotype during the course of an immune response via selective effects on different Th subsets. The Journal of Immunology, 1998, 160: 4841–4849. http://www.jimmunol.org/ nti-CD3 mAbs, such as OKT3, have been shown to be activation and a rise in intracellular calcium, the other pathway effective in the treatment of organ graft rejection (1). involving ras activation (7). Previous studies have shown that A However, the morbidity associated with the initial FcR- FcR-nonbinding anti-CD3 delivered a partial signal characterized dependent cytokine release triggered by OKT3 has prompted the proximally by deficient TCR ␨ and ZAP-70 phosphorylation, but exploration of alternative therapies with FcR-nonbinding anti-CD3 full ZAP-70 association with the TCR. The partial signal resulted mAbs (2–4). In the murine model, a chimeric Ab was constructed in functional inactivation of Th1 clones manifested by the inability 3 containing the CD3⑀-specific 145-2C11 (2C11) variable portion to proliferate or secrete IL-2 (8). These findings were strikingly and a murine IgG3 Fc portion (an isotype with low FcR affinity).

similar to studies with altered peptide ligands (APLs), wherein the by guest on September 24, 2021 This mAb failed to induce cytokine production or proliferation of engagement of the TCR by APL led to partial activation of TCR naive T cells in vitro. In vivo, anti-CD3 IgG3 was shown to pro- signal transduction pathways and T cell unresponsiveness (9–11). long graft survival as well as its FcR-binding 2C11 counterpart, In contrast to the FcR-nonbinding anti-CD3-treated Th1 cells, na- but in the apparent absence of toxic side effects. Previous studies ive T cells exposed to the mAb remained responsive (8). These examining the mechanism of immunosuppression by anti-CD3 results suggested that the prolongation of graft survival by FcR- IgG3 mAbs in vivo indicated that FcR-nonbinding anti-CD3 Abs nonbinding anti-CD3 mAbs may be a consequence of selective down-modulated TCR expression, but had no long-term effect on inactivation of Ag-stimulated T cells. In support of this hypothesis, naive T cell functional responses (5). a recent study has demonstrated that low dose anti-CD3 or non- One of the earliest events following ligation of the TCR is the mitogenic anti-CD3 F(abЈ) fragments induce tolerance in overtly tyrosine phosphorylation of multiple components of the TCR com- 2 plex (␨ and CD3⑀, ␥, and ␦) (6). This phosphorylation, believed to diabetic NOD mice, but do not prevent diabetes if administered be mediated by the src family kinases lck and fyn, permits the before disease onset (12). Thus, the mechanism by which these association of other proteins with the TCR complex, including the mAbs suppress immune responses may depend upon the selective ZAP-70 kinase. Two major signaling cascades are set in motion by effects of FcR-nonbinding anti-CD3 on activated T cells. these proximal events, one involving phospholipase C␥1 (PLC␥-1) During the course of an immune response, naive T cells differentiate into Th phenotypes defined by their pattern of cytokine secretion and immunomodulatory properties; Th1 cells secrete Ben May Institute for Cancer Research, Committee on Immunology, Department of TNF-␣, IL-2, and IFN-␥, which enhance inflammatory cell-medi- Pathology, University of Chicago, IL 60637 ated responses, whereas Th2 cells secrete IL-4, IL-5, IL-10, and Received for publication September 11, 1997. Accepted for publication January IL-13, cytokines that suppress inflammatory responses while po- 22, 1998. tentiating humoral immunity (13). Multiple studies have suggested The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance that the induction and maintenance of tolerance in both transplant with 18 U.S.C. Section 1734 solely to indicate this fact. and autoimmune diseases is a direct consequence of enhanced Th2 1 This work was supported by Grants P01AI29531-07 and CA-1459, and by the activity at the expense of the Th1 subset (14, 15). For example, Howard Hughes Medical Institute. treatments that prolong graft survival, such as CTLA4-Ig and anti- 2 Address correspondence and reprint requests to Dr. Jeffrey Bluestone, Committee on CD4 mAbs, correlate with increased IL-4 and IL-10 production in Immunology, 5841 S. Maryland Avenue, MC 1089, Chicago, IL 60637. E-mail address: jbluest@flowcity.bsd.uchicago.edu accepted grafts (16, 17). Thus, any treatment that might “tip the 3 Abbreviations used in this paper: 2c11, 145-2c11; PLC␥-1, phospholipase C␥1; balance” toward a Th2 phenotype would have important therapeu- APL, altered peptide ligand; KO, knockout. tic implications.

Ј Evidence from one in vivo study involving anti-CD3 F(ab )2 tained by restimulation every 7 to 14 days with 0.2 mg/ml OVA peptide, fragments suggests that FcR-nonbinding anti-CD3 mAbs do not 12.5 U/ml rIL-2, and irradiated (3000 rad) H-2d splenic APC. suppress all activated Th subsets. In a murine collagen arthritis Mixed T cell lines were generated as follows. In a 24-well dish, 1 to 1.5 ϫ 105 DO 11.10 lymph node cells per well were activated with 0.3 to model, the resolution of disease by FcR-nonbinding anti-CD3 1 ␮g/ml of OVA peptide in the presence of 6 ϫ 106 irradiated (2000 rad) Ј ␥ d F(ab )2 fragments correlated with suppressed IL-2 and IFN- pro- H-2 splenic APC and 12.5 U/ml of IL-2 for 8 to 12 days before challenge duction, and preserved IL-4 production (18). Thus, selective reg- with anti-CD3 IgG3. For the IL-4 KO and IFN-␥ KO lines, lymph node ulation of Th subsets may contribute to the in vivo efficacy of cells were CD8 depleted with the 3.155 mAb and complement, then 5 ϫ 105 cells per well were stimulated with 0.03 to 0.1 ␮g/ml anti-CD3 (145- FcR-nonbinding anti-CD3 mAbs. 2C11) and 4.5 to 5 ϫ 106 anti-Thy-1 T-depleted irradiated H-2d spleno- To gain a better understanding of the consequences of FcR- cytes for 7 to 12 days. In subsequent rounds of cytokine KO T cell stim- nonbinding anti-CD3 treatment for Th responses, an in vitro anal- ulation, 1 ϫ 105 cells were plated per well. Similar results were obtained ysis of the mAb’s effect on different populations of activated from first round cultures with non-CD8-depleted lymph node cells. All T CD4ϩ T cells was undertaken. In contrast to what had been ob- cell lines were restimulated every 7 to 14 days. served in Th1 clones, Th2 clones and polyclonal IL-4-secreting T Proliferation and responsiveness assays cell populations proliferated, and were not rendered unresponsive The following experiments were performed in 5% FCS supplemented by the FcR-nonbinding anti-CD3 mAbs. Moreover, polyclonal ac- DMEM (Life Technologies, Grand Island, NY). In a 96-well flat-bottom tivated populations exposed to FcR-nonbinding anti-CD3 main- dish, 1 ϫ 105 T cells per well were cultured with media alone, 1 ␮g/ml tained their ability to produce IL-4, but secreted much less IL-2 in plate-immobilized anti-CD3 (145-2C11), a single dose of soluble anti-CD3 IgG3 (1 ␮g/ml), or serial log dilutions of soluble anti-CD3 IgG3. For a secondary response. Examination of the proximal signals in- 4 cytokine-blocking assays, 5 ϫ 10 T cells from the T cell clone 4.5 (or Downloaded from duced by FcR-nonbinding anti-CD3 mAb in Th1 and Th2 cells 24.5) were stimulated in the presence of 1 ␮g/ml of anti-CD3 (or anti-CD3 revealed qualitatively similar deficiencies in ␨, ZAP-70, and MAP IgG3) with or without APC (2.5 ϫ 105 Thy-1-depleted splenocytes irra- kinase phosphorylation. The reduced proximal signals were suffi- diated at 2000 rad), 25 ␮g/ml of anti-IL-4 mAb, 10 ␮g/ml each of anti-IL-2 cient to drive NF-ATc translocation in both Th subsets. Together, plus anti-IL-2R mAbs, or 25 ␮g/ml of rat control Ig mAb. After 40 h, cultures were pulsed with [3H]TdR for a further 8 h, harvested on a 96-well these results suggest that FcR-nonbinding anti-CD3 delivers a par- Filermate 196 plate harvester (Packard Instrument, Meriden, CT), and then tial signal that has different functional consequences for Th1 or counted on a Packard TopCount microplate scintillation counter. Results Th2 populations. The promotion of Th2 cytokine secretion and are represented as the mean of triplicate determinations with a SEM http://www.jimmunol.org/ proliferation, and the concomitant suppression of Th1 responses of Յ20%. ϫ 6 are likely to account for the ability of FcR-nonbinding anti-CD3 to For responsiveness assays, in a 24-well dish, 1 10 T cells per well were cultured with or without 1 ␮g/ml soluble anti-CD3 IgG3 for 24 h, skew in vivo immune responses toward a Th2 phenotype. washed extensively (3 ϫ 10 ml), and then rested for 72 h. After this rest period, in a 96-well flat-bottom dish, 4 ϫ 104 T cells per well were stim- Materials and Methods ulated with 1 mg/ml of OVA, 1 ␮g/ml of OVA peptide, or 1 ␮g/ml of 145-2C11 plus 5 ϫ 105 T-depleted irradiated H-2d APC. Cultures were Animals pulsed at 48 h for a further 12 to 16 h before harvesting and counting. To Six- to eight-week-old BALB/c, B10.A, and DBA/2 mice were purchased test IL-2 and IL-4 cytokine production during proliferation assays or re- from Frederick Cancer Research Laboratories (Frederick, MD). The OVA stimulation assays, supernatants were taken at 24 h or 40 to 48 h and by guest on September 24, 2021 ␥ peptide (323–339)-specific TCR-transgenic DO 11.10 mice were gifts of analyzed by ELISA (Endogen, Cambridge, MA). Anti-IFN- reagents for ␥ Drs. D. Loh and K. Murphy (Washington University, St. Louis, MO) and the IFN- ELISA were provided by Dr. Robert Schrieber (Washington the IL-4 knockout (KO) mice were a gift of Dr. S. Reiner (University of University, St. Louis, MO). Chicago, Chicago, IL). These animals were maintained in a specific patho- Biochemistry gen-free facility at the University of Chicago. Analysis of anti-CD3 IgG3 mAb-induced TCR phosphorylation has been Abs and reagents previously described in detail (8). Briefly, 4 to 7 ϫ 106 T cell clones were precoated in PBS with 5 ␮g/ml of anti-CD3 IgG3 for 10 min on ice. The The following Abs were used in this study: 145-2C11 (anti-CD3); AT83A cells were then stimulated with an equal volume of goat anti-mouse IgG3 (anti-Thy-1); 3.155 (anti-CD8) mAb; anti-CD3 IgG3 (5, 8); 11B11 mAb (1:100 final dilution) for 2.5 min in a 37°C water bath and then lysed with (anti-IL-4) (from Dr. E. Vitteta, University of Texas, Dallas, TX); S4B6 2ϫ Lysis buffer (0.5% Triton X-100 final). After immunoprecipitation with (anti-IL-2), 7D4 (anti-IL-2R), PC615.3 (anti-IL-2R), SFR8-B6 (anti- anti-␨ (H146)-coated protein A-agarose beads (Pharmacia-UpJohn, Upp- human HLA Bw6 rat control Ig) (anti-IL-2/IL-2R and rat control Ig are sala, Sweden), the samples were eluted with reducing sample buffer: 10% protein G-purified mAbs provided by Dr. R. L. Hendricks, University of of the sample was reserved for an anti-␨ blot (387), and the rest of the Illinois, Chicago, IL); goat anti-mouse IgG3 antisera (Sigma, St. Louis, ␨ sample was resolved on 12% SDS-PAGE, transferred to PVDF membrane MO); H146 (anti- ) mAb supernatant (provided by Dr. F. Fitch, University (Millipore, Bedford, MA), blocked with 10% BSA (Sigma), and probed of Chicago, Chicago, IL); 4G10 (anti-phosphotyrosine) (UBI, Lake Placid, ␨ with anti-phosphotyrosine. In some experiments, these blots were stripped NY); 387 (anti- ) antiserum (Dr. L. Samelson, National Institutes of and reprobed with anti-ZAP-70. For analysis of MAP kinase activation, T Health, Bethesda, MD); 1598-8 (anti-ZAP-70) antiserum (Dr. A. Weiss, cells were stimulated as above, and then 1 ϫ 106 cell equivalents of whole University of California, San Francisco, CA); anti-active MAP kinase (Pro- cell lysate was resolved on 10% SDS-PAGE. The blots were probed with mega, Madison, WI), anti-ERK1 and ERK2 (Zymed, San Francisco, CA); anti-active MAP kinase, stripped, and then reprobed with anti-MAP kinase. anti-NF-ATc1 (Affinity BioReagents, Golden, CO); and goat anti-mouse After incubation with the appropriate horseradish peroxidase-coupled sec- FITC (Kirkegaard & Perry Laboratories, Gaithersburg, MD). Goat serum ondary Abs, the blots were developed by enhanced chemiluminescence was purchased from Vector Laboratories (Burlingame, CA). The OVA and (Amersham, Arlington Heights, IL). Densitometry measurements of the pigeon cytochrome c Ag were purchased from Sigma and the rIL-2 was a MAP kinase bands were performed using an AMBIS Image Acquisition gift of Cetus (San Francisco, CA). OVA peptide was provided by Dr. Fitch. and Analysis instrument (San Diego, CA). Low toxicity rabbit complement was purchased from Pel-Freez (Brown Deer, WI). Confocal microscopy

6 T cell clones and lines In serum-free media, 2 to 5 ϫ 10 T cells were stimulated in 100 to 200 ␮l with5to10␮g/ml anti-CD3 IgG3 for 20 min at 37°C. The cells were then The pigeon cytochrome c-specific Th1 clone, AE.7, was provided by Dr. added to an equal volume of 2 to 4% paraformaldehyde for 10 min at room M. Jenkins (University of Minnesota, Minneapolis, MN). The OVA-spe- temperature, washed (1% BSA containing PBS), and further permeabilized cific Th1 clone pGL10 and Th2 clone pL104 were provided by Dr. F. Fitch. with Ϫ70°C methanol for 2 min on ice. Cells were washed and rehydrated pGL10, pL104, and AE.7 T cell clones were maintained as previously with wash buffer for 15 min at room temperature. Nonspecific staining was described except that the APC feeders for pL104 were irradiated at 3000 blocked with 5% normal goat serum for 20 min at room temperature, and rad (19, 20). Th0 clones (24.5 and 4.5) derived from the DO 11.10 TCR then cells were incubated with a 1:300 to 1:1000 dilution of anti-NF-ATc transgenic were also obtained from Dr. Fitch. These clones were main- overnight at 4°C. The cells were washed, incubated for 45 min at 37°C with The Journal of Immunology 4843

of the autocrine growth factor IL-4, although the amount produced was consistently less than that observed in response to immobilized anti-CD3 mAbs (Fig. 1B). Activated T cells designated as “Th0” make both IL-2 and IL-4 before commitment to a Th lineage. The responses of two OVA peptide-speciﬁc Th0 clones (4.5 and 24.5) to the anti-CD3 IgG3 mAb were examined. Both Th0 clones proliferated to soluble and immobilized anti-CD3 mAbs (Fig. 2A). As evidence of their Th0 phenotype, the T cell clones produced IL-2 and IL-4 upon culture with immobilized anti-CD3 mAb; however, in response to the anti- CD3 IgG3 mAb, the Th0 clones secreted only IL-4 (Fig. 2B). It is possible that the IL-2 was undetectable due to consumption by the proliferating cells. If true, the Th0 clones would have preferentially consumed IL-2 vs IL-4 when cultured in the absence of anti- CD3 cross-linking. However, anti-IL-4 mAb, but not anti-IL-2/ IL-2R mAbs, blocked anti-CD3 IgG3-induced proliferation in Th0 clones (Fig. 3).

In an in vivo setting, the presence of costimulation-bearing APC Downloaded from might alter the cytokine profile induced by anti-CD3 IgG3 in Th0 cells. To address this possibility, splenic accessory cells were in- FIGURE 1. Anti-CD3 IgG3 induces IL-4 production and proliferation cluded in the in vitro stimulations (Fig. 3). Despite the presence of in Th2 clones. A, pGL10 (Th1) or pL104 (Th2) T cell clones were cultured APC, Th0 clones produced IL-4 (data not shown) and depended in the presence of media, soluble anti-CD3 IgG3 (open squares) or plastic upon IL-4 for proliferation in response to the anti-CD3 IgG3 mAb. immobilized anti-CD3 (filled squares) for 40 h and then pulsed for 8 h with http://www.jimmunol.org/ [3H]TdR. B, Supernatants (40 h) were examined for the presence of IL-4 by In contrast, both anti-IL-4 and anti-IL-2/IL-2R mAbs partially ELISA. Results are representative of three independent experiments. blocked FcR-binding anti-CD3 mAb-induced proliferation. Simi- lar results were obtained with the Th0 clone 24.5 (data not shown). Thus, IL-4 appears to be the preferred growth factor produced in goat anti-mouse FITC (1:50 final), and then incubated for 15 min with response to the anti-CD3 IgG3 mAb. wash buffer at 37°C. After one more wash, the cells were resuspended in One caveat in using T cell clones to predict the behavior of Fluoromount-G (Southern Biotech, Birmingham, AL), and mounted on slides for analysis on a Zeiss 410 confocal microscope. activated T cells is that clones have been restimulated many times in vitro and thus selected for long-term survival in tissue culture.

Results During the course of passage, clonal responses could potentially by guest on September 24, 2021 FcR-nonbinding anti-CD3 mAbs induce proliferation and IL-4 deviate from what might be observed with “normal” activated T production in Th0 and Th2 cells cells. Thus, bulk T cells from the DO 11.10 TCR transgenic were The ability to trigger or suppress different activated Th populations activated with Ag and APC in vitro one to three times, then chal- may contribute to the in vivo efﬁcacy of FcR-nonbinding anti- lenged with the anti-CD3 IgG3 mAb. At the time of analysis, these CD3. Therefore, the effect of the anti-CD3 IgG3 mAb on Th1 and polyclonal activated T cells were capable of producing IL-2, IL-4, ␥ Th2 responses was compared (Fig. 1A). As previously shown, Th1 and IFN- . Previously activated DO 11.10 T cells proliferated in T cell clones did not proliferate in response to the soluble bivalent response to the soluble anti-CD3 IgG3, in contrast to the lack of anti-CD3 mAb. However, multivalent cross-linking provided by a response seen in naive T cells (data not shown, and Ref. 8) (Fig. secondary anti-IgG Ab (8), or immobilization of the anti-CD3 4). The T cells stimulated with soluble anti-CD3 IgG3 produced mAb on a plastic surface resulted in proliferation. By comparison, IL-4, and not IL-2, even though (as seen in response to immobi- the Th2 clone, pL104, incorporated [3H]TdR in the absence of lized anti-CD3) the T cells were capable of producing both exogenous mAb cross-linking. In the presence of splenic APC, cytokines. anti-CD3 IgG3 also promoted clonal expansion of the Th2 cells To better deﬁne which cells were proliferating in response to the (data not shown). An examination of the Th2 culture supernatants anti-CD3 IgG3 mAb within the polyclonal activated T cell popu- revealed that the soluble anti-CD3 IgG3 mAb induced production lation, IL-4 KO and IFN-␥ KO mice were used to generate Th1

FIGURE 2. Th0 clones proliferate and produce IL-4 in response to the anti-CD3 IgG3 mAb. A, The Th0 clone 4.5 was cultured in the presence of media, soluble anti- CD3 IgG3, or immobilized anti-CD3 for 40 h, and then pulsed for 8 h. B, A second Th0 clone, 24.5, was stimulated as indicated, and then 40-h supernatants were tested by ELISA for IL-2 (open bars) and IL-4 (hatched bars) production. Results are representative of three separate experiments. In individual Th0 experiments, similar results were obtained with clone 4.5 and 24.5. *, Below limit of detection (0.2 ng/ml). 4844 PARTIAL TCR SIGNALS DIFFERENTIALLY REGULATE Th SUBSETS

FIGURE 3. Anti-IL-4 mAb, but not anti- IL-2/IL-2R mAbs, block anti-CD3 IgG3-induced proliferation in a Th0 T cell clone. 4.5 T cells (Th0) were stimulated with 1 ␮g/ml of soluble anti-CD3 IgG3 or anti-CD3 in the presence or absence of APC. Anti-IL-4 mAb, anti-IL-2/IL-2R, or rat control Ig were added as indicated. Cultures were pulsed with [3H]TdR at 40 h. Results are representative of four independent experiments. Similar results were obtained with the Th0 clone 24.5 (data not shown).

and Th2 populations, respectively. After one round of in vitro ac- T cells to respond in subsequent stimulations. As seen in Figure Downloaded from tivation with mitogenic anti-CD3 (2C11) and APC, T cells from 6B, Th0 clones precultured with the soluble anti-CD3 IgG3 were the IL-4 KO mice produced IFN-␥ whereas T cells from the IFN-␥ hyporesponsive in a secondary antigenic stimulation (20% of con- KO mice produced IL-4 (data not shown). When challenged with trol proliferation). The anti-CD3 IgG3-treated Th0 clones pro- the anti-CD3 IgG3, the activated IFN-␥ KO T cells proliferated to duced readily detectable IL-4 (40% of control), similar to what has both soluble and immobilized anti-CD3. In contrast, the activated been observed in other anergy systems (21).

IL-4 KO cells proliferated to immobilized, but not soluble anti- Finally, the consequences of anti-CD3 IgG3 pretreatment of http://www.jimmunol.org/ CD3 IgG3 (Fig. 5). Thus, anti-CD3 IgG3 induced proliferation bulk Ag-activated T cells were examined (Fig. 7). Culture with the only in the Th2-like, IL-4-secreting populations. anti-CD3 IgG3 mAb had a minimal effect on T cell proliferation in FcR-nonbinding anti-CD3 induces unresponsiveness in Th1 and the secondary antigenic challenge in the majority of experiments. Th0 cells, but not Th2 T cells In some experiments, treatment with anti-CD3 IgG3 following two to three rounds of in vitro stimulation resulted in a significant The preceding experiments suggested that enhanced outgrowth of decrease (up to 67%) in proliferative response to Ag. This varia- IL-4-producing cells following FcR-nonbinding anti-CD3 treat- tion in the effect of anti-CD3 IgG3 on proliferative responsiveness ment may contribute to the Th cytokine deviation observed in sev- may reflect the relative mixture of Th0, Th1, and Th2 cells that eral in vivo models. These alterations in Th phenotype could also by guest on September 24, 2021 developed during the repeated antigenic stimulations. However, in reflect the selective induction of Th1 unresponsiveness. To exam- all the experiments, anti-CD3 IgG3 treatment induced a profound ine this latter possibility, the effect of anti-CD3 pretreatment on deviation in the cytokine profile evident upon restimulation with Th1 vs Th2 clonal responsiveness was determined. T cells were Ag: T cells exposed to anti-CD3 IgG3 produced equal or slightly cultured for 24 h with anti-CD3 IgG3, washed extensively, rested for 3 days, and then restimulated with optimal doses of Ag and greater IL-4, and significantly less IL-2 compared with controls. APC (Fig. 6). This 3-day rest period was sufficient for TCR reex- The observed alteration in the cytokine profile reflected poten- pression (5, 8). Preculturing the Th1 clone, pGL10, with anti-CD3 tial contributions from selective outgrowth of IL-4-producing cells IgG3 resulted in proliferative hyporesponsiveness that correlated as well as Th1 unresponsiveness. However, it was not clear from with reduced IL-2 production (Ref. 8, and data not shown). The previous studies whether in vitro-activated nonclonal Th1 cells addition of costimulation-bearing splenic APC did not affect the could be rendered anergic. The effects of anti-CD3 IgG3 treatment ability of anti-CD3 IgG3 to induce unresponsiveness in Th1 clones on the responsiveness of activated T cells derived from the IL-4 (8). In contrast to the Th1 clone, preculture of the Th2 clone pL104 KO and IFN-␥ KO mice were compared. Consistent with the ob- with anti-CD3 IgG3 did not affect the ability of the T cells to servations in T cell clones, anti-CD3 IgG3 induced hyporespon- respond to Ag, or produce IL-4 in the restimulation assay (Fig. siveness in IL-4 KO, but not in IFN-␥ KO T cells (Fig. 8). Taken 6A). Next, Th0 clones were examined to determine the effect of together, the results from clonal and polyclonal populations sug- anti-CD3 IgG3 treatment on the ability of dual cytokine-producing gest that selective Th hyporesponsiveness may contribute to

FIGURE 4. Polyclonal activated T cell populations produce IL-4 and proliferate in response to anti-CD3 IgG3. DO 11.10 lymph node cells were activated with OVA peptide, irradiated splenic APC, and IL-2 one to three times in vitro. The T cells were then cultured with media, anti-CD3 IgG3, or immobilized anti- CD3 for 40 h, and pulsed with [3H]TdR for 8 h. Proliferation results are representative of four independent experiments. Supernatants were harvested at 40 h and analyzed by ELISA for IL-2 and IL-4 production. Similar results were obtained with supernatants harvested at 24 h (data not shown). The Journal of Immunology 4845

tated with anti-␨ and the resulting blot probed with anti- phosphotyrosine Abs. Portions of the anti-␨ immunoprecipitations were probed with anti-␨ Abs to conﬁrm that an equivalent amount of TCR complex was present in the different samples. In both Th subsets, similar qualitative differences were observed between cross-linked and non-cross-linked anti-CD3 signaling. Non-cross- linked anti-CD3 IgG3 mAb induced less of the hyperphosphorylated p23 ␨ vs p21 ␨, and less ZAP-70 phosphorylation (Fig. 9). In Th2 cells, phosphorylated CD3⑀ and p18 ␨ were diminished as FIGURE 5. Soluble anti-CD3 IgG3 induces proliferation in in vitro- well. Examination of aliquots (10% of volume) by Western blot- activated IFN-␥ KO T cells, but not IL-4 KO T cells. CD8-depleted lymph ting with an anti-␨ antiserum demonstrated comparable amounts of ␥ node cells from IL-4 KO or IFN- KO mice were activated in vitro one or ␨ in each preparation (lower panel). Probing the TCR blots with two times with anti-CD3 (145-2C11), IL-2, and T-depleted irradiated anti-ZAP-70 revealed that even in the apparent absence of ZAP-70 splenic APC. The activated T cells were then cultured with soluble or immobilized anti-CD3 for 48 h. Results are representative of three separate phosphorylation in Th2 clones, ZAP-70 was physically associated experiments. with the TCR complex (data not shown). Similar results have been obtained with Th0 clones (data not shown). Thus, at the level of the TCR, anti-CD3 does not appear to induce Th2 proliferation by the cytokine proﬁle changes induced by anti-CD3 IgG3 mAb delivering a more complete TCR signal. Downloaded from treatment.

Th1 and Th2 clones show similar proximal signaling defects in Anti-CD3 IgG3 induces defective downstream signals in both response to anti-CD3 IgG3 Th1 and Th2 clones Experiments with T cell clones and polyclonal activated popula- It was possible that even though the proximal anti-CD3 IgG3 sig-

tions indicated that anti-CD3 IgG3 could induce proliferation only http://www.jimmunol.org/ nals were defective, the minimal phosphorylation observed was in cells capable of producing IL-4. Thus, either the anti-CD3 IgG3 sufficient to induce more complete downstream events in Th2 delivers biochemically distinct TCR signals to Th1 and Th2 cells cells. TCR-induced ras activity has been shown to be essential for or anti-CD3 IgG3 delivers a similar TCR signal with different out- comes. It had been demonstrated previously that triggering of the T cell activation. Ras triggers the activation of a series of serine/ TCR on Th1 clones by non-cross-linked anti-CD3 IgG3 resulted in threonine kinases leading to MAP kinase phosphorylation, activa- partial phosphorylation of ␨ and inefficient phosphorylation of tion, and translocation into the nucleus. This signaling cascade TCR-associated ZAP-70. This proximal signal resulted in down- culminates in the activation of a composite transcription factor, stream decreases in PLC␥-1 activation. For Th1 cells, this pertur- AP-1, which binds multiple cytokine promoters (22). Therefore, bation of tyrosine phosphorylation correlated with a tolerogenic MAP kinase phosphorylation was evaluated as an indicator of ras by guest on September 24, 2021 signal (8). pathway induction in anti-CD3 IgG3-triggered T cell responses To address whether anti-CD3 IgG3 delivers a more “complete” (Fig. 10). In the presence of an anti-IgG cross-linker, anti-CD3 signal to responder Th2 cell types, proximal signaling events in IgG3 induced significant MAP kinase phosphorylation. By com- pGL10 (Th1) or pL104 (Th2) cells were compared. After stimu- parison, the non-cross-linked anti-CD3 IgG3 resulted in much lation with anti-CD3 IgG3 in the presence or absence of an anti- weaker MAP kinase phosphorylation (fourfold less for ERK2 and IgG cross-linking reagent, the TCR complex was immunoprecipi- sevenfold less for ERK1). A functional assay for MAPK activation

FIGURE 6. Anti-CD3 IgG3 renders Th1 and Th0 clones, but not Th2 clones, unresponsive. A, pGL10 (Th1) or pL104 (Th2) clones were cultured with media alone or soluble anti-CD3 IgG3 for 24 h, washed three times, and then rested for 3 days. At this point, the T cell clones were restimulated with 1 ␮g/ml of OVA Ag and T-depleted irradiated splenic APC for 48 h, and then pulsed for a further 12 to 16 h. Supernatants were harvested at 48 h and examined for the presence of IL-4 by ELISA. Results are representative of two independent experiments. B, Th0 clones were cultured with or without anti-CD3 IgG3, and restimulated as in A. Three experiments were performed using both Th0 clones 24.5 and 4.5 (similar proliferation results were obtained with each). Over multiple experiments, IL-4 production by anti-CD3 IgG3-pretreated T cells during the secondary stimulation ranged from 40 to 240% of media pretreated controls. 4846 PARTIAL TCR SIGNALS DIFFERENTIALLY REGULATE Th SUBSETS

FIGURE 7. Anti-CD3 IgG3 treatment of polyclonal populations results in decreased IL-2 production. Bulk T cells, activated as in Figure 4, were cultured with anti-CD3 IgG3 for 24 h, washed, rested, and restimulated with OVA peptide (or OVA Ag) and T-depleted irradiated splenic APC for 48 h. At 48 h, supernatants were collected for analysis by ELISA and cultures were pulsed with [3H]TdR. Cytokine results are representative of seven independent experiments. In the restimulation, proliferation results varied from no effect (as shown) up to a 67% decrease after pretreatment with anti-CD3 IgG3. was consistent with the MAPK phosphorylation 644/42 MAP Ki- treatment with anti-CD3 IgG3 induced a shift in NF-ATc local- nase Assay Kit, New England Biolabs, data not shown). This re- ization from the cytoplasm (evident as a thin ring) to more diffuse duced phosphorylation in the absence of anti-CD3 cross-linking central areas containing bright spots. Thus, in spite of the proximal was not merely due to delayed kinetics (data not shown). Signif- deficits in TCR signaling manifested in reduced PLC␥-1 and MAP icantly, ras pathway signaling was compromised to the same ex- kinase activation, anti-CD3 IgG3 delivered a sufficient signal to Downloaded from tent in both Th1 and Th2 clones following anti-CD3 IgG3 induce NF-ATc translocation into the nucleus (8). stimulation. Upon ligation of the TCR, a second lipid-mediated pathway is activated as well. A key player in this pathway, PLC␥-1, cleaves phosphoinositol bisphosphate to yield diacylglycerol and IP3.IP3 triggers a Ca2ϩ flux, which ultimately leads to nuclear transloca- http://www.jimmunol.org/ tion of NF-AT (the cytoplasmic portion of the nuclear factor of activated T cells) (7). Both NF-ATp and NF-ATc family members have been shown to translocate upon activation (23). NF-AT is a critical transcription factor for several cytokine genes, including IL-2 and IL-4 (24). In previous studies, an anti-CD3 IgG3-induced calcium flux was not detectable by FACS in Th1 cells (8). How- ever, there were several indirect indications suggesting that anti- CD3 IgG3 might induce a subtle calcium signal. The ability of by guest on September 24, 2021 cyclosporin A to block anti-CD3 IgG3-induced Th1 anergy im- plied that anti-CD3 IgG3 delivered a calcium signal that might be required for the tolerogenic activity of the mAb (8). Furthermore, anti-CD3 IgG3 delivered a sufficient Ca2ϩ signal to synergize with PMA in causing IL-2 production and proliferation in both naive T cells and Th1 clones (data not shown). Finally, in an APL model, where IP3 generation had been historically undetectable, extremely sensitive video imaging revealed transient low amplitude calcium fluxes (25). Based on these results, it was likely that anti-CD3 IgG3 delivers a weak calcium signal that might affect NF-AT translocation into the nucleus. Therefore, the translocation of NF- ATc was examined in Th1 and Th2 clones stimulated with anti- CD3 IgG3. The stimulated cells were fixed, stained with anti-NF- ATc, and analyzed by confocal microscopy. As seen in Figure 11,

FIGURE 9. Defective ␨ and ZAP-70 phosphorylation induced by anti- CD3 IgG3 in Th1 and Th2 cells. Th1 (pGL10) or Th2 (pL104) clones were stimulated with anti-CD3 IgG3 in the presence or absence of anti-IgG3 cross-linker for 2.5 min at 37°C. The cells were then lysed and the TCR complex immunoprecipitated with anti-␨ mAb. Aliquots (10% of vol) were FIGURE 8. Soluble anti-CD3 IgG3 induces hyporesponsiveness in ac- removed, and then the remaining samples were resolved on 12% SDS- tivated IL-4 KO, but not IFN-␥ KO T cells. Cytokine KO T cells were PAGE, transferred to PVDF membrane, and subjected to Western blotting activated as in Figure 5, cultured with anti-CD3 IgG3 or media for 24 h, with anti-phosphotyrosine (upper panel). The aliquots of the above sam- and then rested for 72 h. In the restimulation, the T cells were stimulated ples were loaded separately and subjected to Western blotting with anti-␨ with anti-CD3 (145-2C11) and T-depleted irradiated splenic APC. Results antiserum (lower panel). Results are representative of three independent are representative of two independent experiments. experiments. The Journal of Immunology 4847

Th0 clones, and Th1 lines, were rendered hyporesponsive following treatment with anti-CD3 IgG3. For Th0 clones, the reduced responsiveness most likely resulted from the combined lack of IL-2 production and the blockade of IL-4 responsiveness previously reported in other anergy models (21, 26). The contrasting effects of anti-CD3 IgG3 on Th0 and Th2 responsiveness suggests that the induction of unresponsiveness does not strictly correlate with proliferation during the primary culture. The biochemical signals triggered by anti-CD3 IgG3 mAbs in Th1 and Th2 cells were qualitatively similar. In both T cell subsets, stimulation with the non-cross-linked anti-CD3 IgG3 resulted in a reduced ratio of hyperphosphorylated p23 ␨ compared with p21 ␨ and minimal ZAP-70 phosphorylation. These proximal deficits were exaggerated in Th2 clones, possibly due to the decreased FIGURE 10. Anti-CD3 IgG3 stimulation of Th1 and Th2 clones results overall level of tyrosine phosphorylation seen when T cells were in suboptimal MAP kinase phosphorylation. Th1 (pGL10) and Th2 stimulated with either cross-linked or non-cross-linked anti-CD3. (pL104) clones were stimulated with cross-linked or non-cross-linked anti- The quantitative differences may reﬂect clonal variation, since CD3 IgG3 for 2.5 min at 37°C. Whole cell lysates were resolved by 10%

SDS-PAGE, transferred to PVDF, and then Western blotted with anti- such differences have been observed among Th1 clones (J. Smith Downloaded from active MAP kinase Ab (top panel). The blots were subsequently stripped and J. Bluestone, unpublished observations). Similar proximal sig- and reprobed with anti-MAP kinase (bottom panel). Results are represen- naling defects have been demonstrated in the APL system and tative of two separate experiments. under conditions of CD4 coreceptor blockade (10, 11, 27). Previ- ous reports in these two model systems have stressed the correla- tion between these specific signaling deficits and the induction of Discussion unresponsiveness in T cell clones. However, the results presented http://www.jimmunol.org/ In this study we have shown that an anti-CD3 mAb with low FcR here using a variety of T cell clones and short-term T cell lines affinity, anti-CD3 IgG3, delivers a characteristic partial signal with provide evidence that an altered ratio of p23 to p21 ␨, and defective different functional consequences depending upon the Th pheno- ZAP-70 phosphorylation, do not always lead to the induction of type of the population. Anti-CD3 IgG3 treatment of mixed acti- unresponsiveness. Rather, the consequences of the proximal sig- vated populations resulted in a relative decrease in the ability of naling defects induced by anti-CD3 IgG3 varied, depending upon these populations to produce IL-2, without diminishing IL-4 pro- the Th phenotype. duction, recapitulating the findings from in vivo studies of anti- It should be noted that the downstream biochemical conse- Ј CD3 F(ab )2 treatment (18). The ability of anti-CD3 IgG3 to quences of this characteristic proximal pattern of signaling are clonally expand Th2 (IL-4-secreting) cells while suppressing the largely unknown. In this study, two major TCR signaling cascades, by guest on September 24, 2021 responsiveness of IL-2-secreting cells provides a mechanism for involving the PLC␥-1 and ras pathways, were evaluated by ex- the Ab-induced Th cytokine deviation evident in vitro and in vivo. amining events proximal to the nucleus. Ras activates a series of Specifically, anti-CD3 IgG3 induced proliferation in populations serine/threonine kinases, ultimately resulting in phosphorylation of activated T cells capable of producing the IL-4 growth factor. (and thus activation) of the MAP kinases ERK1 and ERK2 (22). In Unlike Th2 cells, IL-2-secreting populations, such as Th1 clones, Th1 and Th2 cells, non-cross-linked anti-CD3 IgG3 induced weak

FIGURE 11. Anti-CD3 IgG3 induces NF-ATc translocation in Th1 and Th2 cells. Th1 (AE.7) or Th2 (pL104) cells were stimulated with media (top row) or anti-CD3 IgG3 (bottom row) for 20 min at 37°C. Af- ter paraformaldehyde and methanol fixa- tion, the cells were stained with anti-NF- ATc and goat anti-mouse FITC, and then mounted on slides for analysis by confocal microscopy. Results are representative or two separate Th1 experiments and three Th2 experiments. 4848 PARTIAL TCR SIGNALS DIFFERENTIALLY REGULATE Th SUBSETS phosphorylation of the MAP kinases compared with the cross- PMA may contribute by activating ras (thus enhancing AP-1 ac- linked mAb, indicative of suboptimal ras signaling. Quantitatively tivity) or PKC (NF-␬B) (22). The IL-2 promoter requires the co- similar defects were observed in Th1 and Th2 cells. Since the ERK operative binding of multiple different transcription factors, in- kinases regulate the fos component of the AP-1 transcription fac- cluding NF-AT, AP-1, and NF-␬B. In fact, the NF-AT binding tor, these results suggest that soluble anti-CD3 IgG3 may induce sites within the IL-2 promoter represent composite NF-AT/AP-1 less AP-1 than a cross-linked anti-CD3 stimulus (22). sites, where AP-1 is required for activity (24). In contrast, it has Previous studies have shown that the anti-CD3 IgG3 mAb in- been suggested that NF-AT, in the presence of other easily induc- duced little PLC␥-1 phosphorylation, and calcium flux was not ible factors (such as c-maf ), may be sufficient to drive minimal detectable by FACS analysis (8). These results had suggested that IL-4 transcription. Unlike the IL-2 promoter, the IL-4 promoter the calcium signal delivered by anti-CD3 IgG3 must be very low. contains NF-AT binding sites that do not require AP-1 (24). De- In a system using APLs, demonstration of a low amplitude calcium spite the presence of these sites, it has been demonstrated that signal required sensitive video imaging techniques (25). The data NF-AT and AP-1 greatly synergize in enhancing IL-4 transcription presented here are the first indication that a partial TCR signal, (33–35). This difference between NF-AT activity in the presence characterized by an altered ratio of phospho-␨ and defective or absence of AP-1 suggests a basis for the lower levels of IL-4 ZAP-70 phosphorylation (and a low level calcium signal), is suf- observed in the absence of anti-CD3 cross-linking. Taken together, ficient to induce translocation of NF-ATc into the nucleus. NF-AT these results suggest that the decreased level of MAP kinase ac- translocation occurred in both Th1 and Th2 cells stimulated by tivity (and thus AP-1) induced by non cross-linked anti-CD3 could anti-CD3 IgG3. These results imply either that ZAP-70 phosphor- be more deleterious for IL-2 than for IL-4 production. Further- Downloaded from ylation is dispensable for this event, or that low levels are suffi- more, the NF-AT that translocates in response to non-cross-linked cient. In a recent study examining B cell signaling, Dolmetsch et anti-CD3 may be sufficient for IL-4 production. al. showed that low levels of calcium resulted in NF-AT translo- The data presented in this study have implications for how other cation (consistent with our findings), yet higher levels of calcium TCR signaling-related therapies (such as APLs or nondepleting were required for JNK and NF-␬B activation (28). Together with anti-CD4) may exert their protective effects in vivo, as well as for the MAP kinase data, these results suggest that stimulation by general mechanisms of tolerance induction. Consistent with the in non-cross-linked anti-CD3 Abs may result in a qualitatively and vivo findings with FcR-nonbinding anti-CD3 mAbs, effective anti- http://www.jimmunol.org/ quantitatively different array of activated transcription factors than CD4 therapy in transplantation and autoimmune diseases strongly those induced by a cross-linked anti-CD3 stimulus. correlates with Th deviation from a Th1 to a Th2 phenotype (17, A major question raised by the apparent similarity in anti-CD3 36). It may be more than a coincidence that the proximal signals IgG3-mediated signal transduction in Th1 and Th2 cells is why the delivered by anti-CD3 IgG3 and under conditions of coreceptor mAb selectively induced proliferation and unresponsiveness in blockade resemble each other (31). In a recent study, T cell lines specific subsets. The selective stimulation of proliferation by anti- and clones derived from mice injected with altered proteolipopro- CD3 IgG3 could reflect either quantitatively or qualitatively dif- tein peptides displayed a Th0/Th2 phenotype, and adoptively pre-

ferent requirements for driving IL-2 vs IL-4 transcription. For in- vented experimental allergic encephalomyelitis (37). The results by guest on September 24, 2021 stance, it is possible that all the correct signals are being sent by from the present study have shown that anti-CD3 IgG3 induces anti-CD3 IgG3 at a reduced level, but the cytokine promoters have NF-AT translocation, but not efficient MAP kinase phosphoryla- quantitatively different hierarchical thresholds for triggering. In the tion. Interestingly, in B cells, a toleragenic signal has been shown absence of cross-linking, anti-CD3 IgG3 induced 10-fold less IL-4 to consist of NF-AT and ERK activation, but not JNK kinase ac- in Th2 clones. Anti-CD3 IgG3 stimulation of the Th1 clone, tivation (38). The selective activity of specific transcription factors pGL10, resulted in two logs less IFN-␥ production compared with (such as NF-AT) in the absence of others may translate into a immobilized anti-CD3 stimulation (data not shown). The subopti- toleragenic signal in multiple cell types. Thus, different models of mal levels of cytokine transcription factors induced by anti-CD3 altered Ag receptor signaling may reflect the use of common bio- IgG3 may fall below the threshold for effective IL-2 production. chemical pathways that lead to tolerance as manifested by lym- Alternatively, differential association of transcription factors with phocyte inactivation or cytokine deviation. the IL-2 and IL-4 promoters may account for the disparate sensi- The results presented here also have specific implications for the tivity (29). This quantitative hypothesis is consistent with studies use of FcR-nonbinding Abs in the clinical setting. During an in examining the effect of Ag dose on Th development. Several vivo immune response, such as graft rejection, T cells differentiate groups have reported that extremely low levels of nominal Ag into both Th1 and Th2 phenotypes. Besides the direct pro-Th2 preferentially induced a Th2 subset phenotype, and that higher effect of FcR-nonbinding anti-CD3 mAbs on activated cells, the levels of Ag were required for Th1 differentiation (30, 31). How- development of a Th2 response could be magnified through re- ever, low doses of Ag deliver proximal signals that are qualita- cruitment of uncommitted cells. Cytokines, such as IL-4, promote tively different from the pattern of signaling triggered by APL (and selective Th development (13). Thus, anti-CD3 IgG3-induced thus anti-CD3 IgG3) (10, 27). Also, there is a complete lack of modification of the cytokine milieu could alter Th differentiation of dose response to the anti-CD3 IgG3, wherein high amounts of naive T cells responding to Ag. The ability of anti-CD3 IgG3 to mAb failed to induce proliferation in Th1 cells. suppress Th1 responses while promoting Th2 responses in vitro Alternatively, selective cytokine induction by anti-CD3 IgG3 suggests a mechanism that may explain the efficacy of these mAbs could reflect qualitative differences in the transcription factors re- in prolonging graft survival in the absence of global anergy induc- quired for cytokine promoter activity. For instance, IL-4 transcription. Both the low toxicity of FcR-nonbinding anti-CD3 mAbs and tion could be less dependent on triggering of all of the TCR-related their potential for Th2 cytokine deviation suggest these Abs may signaling cascades. On a gross level, Th2 clones have been re- be effective in suppressing Th1-mediated autoimmune diseases. ported to produce IL-4 in response to calcium ionophores alone, whereas Th1 cells require another signal (e.g., PMA) to produce IL-2 (32). Similarly, although anti-CD3 IgG3 induced Th2 prolif- Acknowledgments eration, the mAb only elicited IL-2 production and proliferation in We thank Julie Auger for her substantial assistance with the confocal mi- naive cells or Th1 clones in the presence of PMA (data not shown). croscopy experiments. The Journal of Immunology 4849

References 20. Quill, H., and R. H. Schwartz. 1987. Stimulation of normal inducer T cell clones with antigen presented by purified Ia molecules in planar lipid membranes: spe- 1. Ortho Multicenter Transplant Study Group. 1985. A randomized clinical trial of cific induction of a long-lived state of proliferative nonresponsiveness. J. Immu- OKT3 monoclonal antibody for acute rejection of cadaveric renal transplants. nol. 138:3704. N. Engl. J. Med 313:337. 21. Gajewski, T. F., D. W. Lancki, R. Stack, and F. W. Fitch. 1994. Anergy of TH0 2. Kan, E. A. R., S. D. Wright, K. Welte, and C. Y. Wang. 1986. Fc receptors on helper T lymphocytes induces downregulation of TH1 characteristics and a tran- monocytes cause OKT3-treated lymphocytes to internalize T3 and to secrete sition to a TH2-like phenotype. J. Exp. Med. 179:481. IL-2. Cell. Immunol. 98:181. 22. Cantrell, D. 1996. T cell antigen receptor signal transduction pathways. Annu. 3. Ferran, C., K. Sheehan, M. Dy, R. Schreiber, S. Merite, P. Landais, L. Noel, Rev. Immunol. 14:259. G. Grau, J. A. Bluestone, J. F. Bach, and L. Chatenoud. 1990. Cytokine related 23. Timmerman, L. A., N. A. Clipstone, S. N. Ho, J. P. Northrop, and G. R. Crabtree. syndrome following injection of anti-CD3 monoclonal antibody: further evidence 1996. Rapid shuttling of NF-AT in discrimination of Ca2ϩ signals and immuno- for transient in vivo T cell activation. Eur. J. Immunol. 20:509. suppression. Nature 383:837. 4. Charpentier, B., C. Hiesse, O. Lantz, C. Ferran, S. Stephens, D. O’Shaugnessy, 24. Rao, A. 1994. NF-ATp: a transcription factor required for the coordinate induc- M. Bodmer, G. Benoit, J. F. Bach, and L. Chatenoud. 1992. Evidence that anti- tion of several cytokine genes. Immunol. Today 15:274. human tumor necrosis factor monoclonal antibody prevents OKT3-induced acute 25. Sloan-Lancaster, J., T. H. Steinberg, and P. M. Allen. 1996. Selective activation syndrome. Transplantation 54:997. of the calcium signaling pathway by altered peptide ligands. J. Exp. Med. 184: 5. Alegre, M. L., J. Y. Tso, H. A. Sattar, J. Smith, F. Desalle, M. Cole, and 1525. J. A. Bluestone. 1995. An anti-murine CD3 monoclonal antibody with a low 26. Mueller, D. L., L. Chiodetti, P. A. Bacon, and R. H. Schwartz. 1991. Clonal ␥ affinity for Fc receptors suppresses transplantation responses while minimizing anergy blocks the response to IL-4, as well as the production of IL-2, in dual- acute toxicity and immunogenicity. J. Immunol. 155:1544. producing T helper cell clones. J. Immunol. 147:4118. 6. Qian, D., I. Griswold-Prenner, M. R. Rosner, and F. W. Fitch. 1993. Multiple 27. Madrenas, J., L. A. Chau, J. Smith, J. A. Bluestone, and R. N. Germain. 1997. components of the T cell antigen receptor complex become tyrosine-phosphory- The efficiency of CD4 recruitment to ligand-engaged TCR controls the agonist/ lated upon activation. J. Biol. Chem. 268:4488. partial agonist properties of peptide-MHC molecule ligands. J. Exp. Med. 185: 7. Weiss, A., and D. R. Littman. 1994. Signal transduction by lymphocyte antigen 219. receptors. Cell 76:263. 28. Dolmetsch, R. E., R. S. Lewis, C. C. Goodnow, and J. I. Healy. 1997. Differential Downloaded from 8. Smith, J. A., J. Y. Tso, M. R. Clark, M. S. Cole, and J. A. Bluestone. 1997. activation of transcription factors induced by Ca2ϩ response amplitude and du- Nonmitogenic anti-CD3 monoclonal antibodies deliver a partial T cell receptor ration. Nature 386:855. signal and induce clonal anergy. J. Exp. Med. 185:1413. 29. Tara, D., D. L. Weiss, and M. A. Brown. 1995. Characterization of the consti- 9. Sloan-Lancaster, J., B. D. Evavold, and P. M. Allen. 1993. Induction of T-cell tutive and inducible components of a T cell IL-4 activation-responsive element. anergy by altered T-cell-receptor ligand on live antigen-presenting cells. Nature J. Immunol. 154:4592. 363:156. 30. Constant, S., C. Pfeiffer, A. Woodard, T. Pasqualini, and K. Bottomly. 1995. 10. Sloan-Lancaster, J., A. S. Shaw, J. B. Rothbard, and P. M. Allen. 1994. Partial T Extent of T cell receptor ligation can determine the functional differentiation of cell signaling: altered phospho-zeta and lack of ZAP70 recruitment in APL- ϩ naive CD4 T cells. J. Exp. Med. 182:1591. http://www.jimmunol.org/ induced T cell anergy. Cell 79:913. 31. Hosken, N. A., K. Shibuya, A. W. Heath, K. M. Murphy, and A. O’Garra. 1995. 11. Madrenas, J., R. L. Wange, J. L. Wang, N. Isakov, L. E. Samelson, and The effect of antigen dose on CD4ϩ T helper cell phenotype development in a T R. N. Germain. 1995. Zeta phosphorylation without ZAP-70 activation induced cell receptor-alpha beta-transgenic model. J. Exp. Med. 182:1579. by TCR antagonists or partial agonists. Science 267:515. 12. Chatenoud, L., J. Primo, and J.-F. Bach. 1997. CD3 antibody-induced dominant 32. Tamura, T., T. Yanagida, and H. Nariuchi. 1993. Difference in signal transduc- self tolerance in overtly diabetic NOD mice. J. Immunol. 158:2947. tion pathway for IL-2 and IL-4 production in T helper 1 and T helper 2 cell clones 13. Abbas, A. K., K. M. Murphy, and A. Sher. 1996. Functional diversity of helper in response to anti-CD3. J. Immunol. 151:6051. T lymphocytes. Nature 383:787. 33. Rooney, J. W., M. R. Hodge, P. G. McCaffrey, A. Rao, and L. H. Glimcher. 1994. 14. Strom, T. B., P. Roy-Chaudhury, R. Manfro, X. X. Zheng, P. W. Nickerson, A common factor regulates both Th1- and Th2-specific cytokine gene expression. K. Wood, and A. Bushell. 1996. The Th1/Th2 paradigm and the allograft re- EMBO J. 13:625. sponse. Curr. Opin. Immunol. 8:688. 34. Rooney, J. W., T. Hoey, and L. H. Glimcher. 1995. Coordinate and cooperative roles for NF-AT and AP-1 in the regulation of the murine IL-4 gene. Immunity

15. Nicholson, L. B., and V. K. Kuchroo. 1996. Manipulation of the Th1/Th2 balance by guest on September 24, 2021 in autoimmune disease. Curr. Opin. Immunol. 8:837. 2:473. 16. Sayegh, M. H., E. Akalin, W. W. Hancock, M. E. Russell, C. B. Carpenter, 35. Ho, I.-C., M. R. Hodge, J. W. Rooney, and L. H. Glimcher. 1996. The proto- P. S. Linsley, and L. A. Turka. 1995. CD28-B7 blockade after alloantigenic oncogene c-maf is responsible for tissue-speciﬁc expression of interleukin-4. Cell challenge in vivo inhibits Th1 cytokines but spares Th2. J. Exp. Med. 181:1869. 85:973. 17. Mottram, P. L., W. R. Han, L. J. Purcell, I. F. McKenzie, and W. W. Hancock. 36. Chu, C. Q., and M. Londei. 1996. Induction of Th2 cytokines and control of 1995. Increased expression of IL-4 and IL-10 and decreased expression of IL-2 collagen-induced arthritis by nondepleting anti-CD4 Abs. J. Immunol. 157:2685. and interferon-gamma in long-surviving mouse heart allografts after brief CD4- 37. Nicholson, L. B., A. Muraza, B. P. Haﬂer, A. Sette, and V. K. Kuchroo. 1997. A monoclonal antibody therapy. Transplantation 59:559. T cell receptor antagonist peptide induces T cells that mediate bystander sup- 18. Hughes, C., J. A. Wolos, E. H. Giannini, and R. Hirsch. 1994. Induction of T pression and prevent autoimmune encephalomyelitis induced with multiple my- helper cell hyporesponsiveness in an experimental model of autoimmunity by elin antigens. Proc. Natl. Acad. Sci. USA 94:9279. using nonmitogenic anti-CD3 monoclonal antibody. J. Immunol. 153:3319. 38. Healy, J. I., R. E. Dolmetsch, L. A. Timmerman, J. G. Cyster, M. L. Thomas, 19. Stack, R. M., D. J. Lenschow, G. S. Gray, J. A. Bluestone, and F. W. Fitch. 1994. G. R. Crabtree, R. S. Lewis, and C. C. Goodnow. 1997. Different nuclear signals IL-4 treatment of small splenic B cells induces costimulatory molecules B7-1 and are activated by the B cell receptor during positive versus negative signaling. B7-2. J. Immunol. 152:5723. Immunity 6:419.