By CD2 and CD28 Coordinate Regulation of T Cell Activation

Coordinate Regulation of T Cell Activation by CD2 and CD28 Jonathan M. Green, Vladimir Karpitskiy, Stephanie L. Kimzey and Andrey S. Shaw This information is current as of October 2, 2021. J Immunol 2000; 164:3591-3595; ; doi: 10.4049/jimmunol.164.7.3591 http://www.jimmunol.org/content/164/7/3591 Downloaded from References This article cites 27 articles, 15 of which you can access for free at: http://www.jimmunol.org/content/164/7/3591.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 October 2, 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 © 2000 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Coordinate Regulation of T Cell Activation by CD2 and CD281

Jonathan M. Green,2*† Vladimir Karpitskiy,† Stephanie L. Kimzey,* and Andrey S. Shaw†

T cell activation requires co-engagement of the TCR with accessory and costimulatory molecules. However, the exact mechanism of costimulatory function is unknown. Mice lacking CD2 or CD28 show only mild deficits, demonstrating that neither protein is essential for T cell activation. In this paper we have generated mice lacking both CD2 and CD28. T cells from the double-deficient mice have a profound defect in activation by soluble anti-CD3 Ab and Ag, yet remain responsive to immobilized anti-CD3. This suggests that CD2 and CD28 may function together to facilitate interactions of the T cell and APC, allowing for efficient signal transduction through the TCR. The Journal of Immunology, 2000, 164: 3591–3595.

cell activation is initiated by recognition of Ag by the binds ligands expressed mainly on APC. When bound to their li- TCR. Although necessary, engagement of the TCR is not gands, CD2 and CD28 can help to span a distance of ϳ15 nm, the sufﬁcient for full activation; rather participation of acces- same distance spanned by the TCR bound to its ligand, MHC/

T Downloaded from sory or co-stimulator molecules is also required (1, 2). This ob- peptide (13). Because co-engagement of CD2 with the TCR can servation forms the basis for a two-signal model of lymphocyte strongly potentiate T cell activation, CD2 has also been considered activation (3). In this model, the ﬁrst signal is provided by the a costimulatory molecule. Although T cells from mice lacking ex- TCR/CD3 complex, while the second is provided by a costimula- pression of CD2 were initially reported to have normal activation tory protein on the surface of the T cell. Engagement of the TCR parameters, a recent study suggests that CD2 may have a small role in the absence of costimulation results in an abortive T cell re- in enhancing T cell activation (14, 15).

sponse and a subsequent state of Ag-specific unresponsiveness The fact that T cells from mice lacking CD2 or CD28 are only http://www.jimmunol.org/ termed anergy (2). Although intensively studied over the last de- mildly affected led us to explore whether CD2 and CD28 might cade, the exact mechanism by which costimulatory molecules reg- share some overlapping functions. We therefore generated mice ulate T cell activation is unclear. Most current models favor the deficient in both CD2 and CD28 expression. Our data demonstrate idea that the accessory or costimulatory molecules provide an es- that CD2/CD28 double-deficient T cells have a profound defect in sential second biochemical signal that is required for T cell activation and proliferation. These data suggest that CD2 and activation. CD28 have redundant functions in T cell activation and together Although many molecules have been implicated as costimula- regulate some of the initial steps in T cell activation. tory receptors, CD28 appears to be the most important. Engage- ment of CD28 either by its ligand on the APC, B7-1 (CD80) or Materials and Methods by guest on October 2, 2021 B7-2 (CD86), or by Ab can strongly enhance TCR signaling re- Mice sponses (4). Importantly, blocking CD28 engagement inhibits T CD28-deficient mice backbred five generations to C57BL/6 have previ- cell activation and may result in T cell anergy (5–8). Surprisingly, ously been described (10). CD2-deficient mice were generously provided studies of mice that lack expression of CD28 demonstrate that from N. Killeen (University of California, San Francisco) via J. Bromberg significant residual T cell function exists in the absence of CD28 (University of Michigan, Ann Arbor). The mice were crossed to generate mice homozygous for the null alleles at CD2 and CD28. Littermates were (9, 10). Although T cells from CD28-deficient mice have de- bred to either CD2- or CD28-deficient mice to generate control mice. Ge- creased overall responses, they are still capable of mounting sig- notypes were determined by PCR and confirmed by southern blotting. nificant responses to mitogenic stimuli, and are capable of strong CD2/CD28-deficient mice described above were then crossed to CD28- responses against transplanted tissues and some viral infections deficient D011.10 mice (provided by S. Reiner and C. Thompson, Univer- (10, 11). Thus, CD28 is not absolutely required for T cell function. sity of Chicago). Pups deficient at both CD2 and CD28 were then screened for the presence of the TCR transgene by flow cytometry using the clono- One potential explanation for these results is that CD28 may typic mAb KJ1-26. MHC haplotype was also determined by flow cytom- share functions with other T cell membrane proteins. One candi- etry of peripheral blood. Mice were 6–12 wk of age for all experiments, date is the CD2 molecule. Like CD28, CD2 is a member of the Ig and were age and sex matched within each experiment. All mice are housed supergene family (12). It is expressed primarily on T cells and in specific pathogen free environments maintained by the Department of Comparative Medicine at the Washington University School of Medicine. Antibodies Departments of *Medicine and †Pathology, Washington University School of Med- Anti-CD3 (145-2c11, hamster IgG) was provided by J. Bluestone, (Uni- icine, St. Louis, MO 63110 versity of Chicago). The clonotypic mAb KJ1-26 was generously provided Received for publication November 8, 1999. Accepted for publication January by K. Murphy (Washington University, St. Louis, MO). All other Abs were 27, 2000. purchased from PharMingen (San Diego, CA). Flow cytometric analysis The costs of publication of this article were defrayed in part by the payment of page was performed on a FACSCalibur flow cytometer using CellQuest soft- charges. This article must therefore be hereby marked advertisement in accordance ware (Becton Dickinson Corporation, Mountain View, CA). with 18 U.S.C. Section 1734 solely to indicate this fact. Proliferation assays 1 This work was supported in part by National Institutes of Health Grants HL03408 (J.M.G.) and AI34094 (A.S.S.) and by a grant from the American Lung Association Lymph node cells or splenocytes were isolated, and single cell suspensions of Eastern Missouri (J.M.G.). were prepared following standard protocols. Cultures were stimulated with 2 Address correspondence and reprint requests to Dr. Jonathan M. Green, Washington the indicated doses of anti-CD3 mAb and pulsed for the final8hofa48-h 3 University School of Medicine, Box 8052, 660 South Euclid Avenue, St. Louis, MO culture with 1.0 ␮Ci/well tritiated thymidine ([ H]TdR; ICN, Costa Mesa, 63110. E-mail address: [email protected] CA). Each condition was plated in quadruplicate and the mean Ϯ SD

FIGURE 1. T cell development in CD2/CD28-deficient mice. Lymph node cells or thymocytes were isolated from 6-wk-old wild-type, CD2-, CD28-, or CD2/CD28-deficient mice and analyzed for surface expression of CD4 and CD8 by two-color flow cytometry. The percentage of cells in each quadrant is indicated. Downloaded from

shown. Exogenous IL-2 (100 U/ml) was added at the start of the culture as CD2/CD28 double-deﬁcient T cells fail to proliferate in indicated in the experiments. Puriﬁed T cells were prepared from lymph response to soluble anti-CD3 but respond to immobilized anti- node cells by negative selection with magnetic bead conjugated anti-B220 and anti-MHC class II mAb according to the manufacturers instructions CD3

(Miltenyi Biotec, Auburn, CA). Cells prepared in this manner were Ն99% http://www.jimmunol.org/ Thy1.2 positive by FACS analysis and unresponsive to Con A stimulation To determine whether T cells lacking both CD2 and CD28 were (data not shown). Purified lymph node T cells were stimulated with plate able to respond to TCR engagement, we stimulated bulk lymph immobilized anti-CD3 following standard methods. For stimulation of node cells with soluble anti-CD3 and measured proliferation by cells in the D011.10 background, 1.5 ϫ 105 splenocytes were stimulated thymidine incorporation (Fig. 2A). Wild-type mice or mice defi- with the OVA323–339 peptide in round-bottom microtiter plates. Prolifera- tion was determined by [3H]TdR incorporation. Thymidine incorporation cient in only CD2 or CD28 mounted a robust proliferative re- was measured at 48, 72, and 96 h. All experiments were repeated a min- sponse to anti-CD3 stimulation. In contrast, T cells from mice imum of three times, and representative data are presented. deficient in both CD2 and CD28 were virtually unresponsive to Conjugate assay CD3 engagement by soluble Ab. However, T cells from all genotypes proliferated strongly to stimulation with PMA and ionomy- by guest on October 2, 2021 The murine B cell line A20 that expresses I-Ad was pulsed with the indicated dose of Ag for 18 h and then labeled with Cell Tracker Green (0.5 cin (Fig. 2B). Similar results were obtained in experiments per- ␮M; Molecular Probes, Eugene, OR) per the manufacturer’s protocol. formed with splenocytes or lymph node cells and when examined Lymph node cells were isolated and labeled with Cell Tracker Orange (0.5 at 24, 48, and 72 h. No reproducible differences were observed in ␮ M; Molecular Probes). The ability to form T cell/APC cell conjugates the response of mice homozygous or heterozygous for the wild- was assessed as previously described (16). Briefly, T cells and APC were resuspended in cold DMEM following labeling at 5 ϫ 106 cells per ml. type alleles (data not shown). Twenty-five microliters of each were then placed into a 0.2 ml tube and the Surprisingly, when purified T cells from the CD2/CD28 double- cells briefly pelleted at 4°C and then incubated at 37°C for 6 min. The cells deficient mice were activated using immobilized anti-CD3, near were then resuspended in 2 ml of PBS containing 1% BSA and gently mixed. The presence of T cell:APC conjugates was assessed using two- normal proliferative responses were measured (Fig. 2C). This color flow cytometric analysis on a FACSCalibur flow cytometer. Data are demonstrates that signal transduction pathways initiated through expressed as the percentage of conjugates relative to wild type, where the the TCR are functional, and suggests that the defect may involve wild type is set at 100%. The mean of two independent experiments is interactions between the T cell and the APC. In support of this presented. idea, the CD2/CD28-deficient mice failed to respond to Con A Results treatment, which is also dependent on interactions with accessory T cell development in CD2/CD28-deficient mice cells (Fig. 2B). We also examined the expression of the activation markers To assess the role of CD2 and CD28 in T cell development, we CD69 and CD25 following stimulation of bulk lymph node cells bred mice singly deficient in either CD2 or CD28 to obtain mice with soluble anti-CD3. Expression of both CD69 and CD25 were deficient in both genes. The genotype of each mouse was deter- decreased in the double knockout mice (Table I). Consistent with mined by PCR analysis using primers specific for the wild-type and knockout alleles, and confirmed by Southern blotting. In ad- the failure to induce CD25 expression, addition of exogenous IL-2 dition, surface expression of CD2 and CD28 was assessed by flow did not restore the proliferative response of the CD2/CD28-double cytometry for each experiment. Analysis of peripheral T cell sub- deficient T cells (Fig. 2D). sets demonstrated no significant alterations in cell number or dis- CD28 has been demonstrated to influence cell survival follow- tribution (Fig. 1). Thymocyte cellularity and subsets were also ing stimulation (17). To determine of the proliferative defect was similar to wild-type mice. The percentage of T cells and B cells in solely due to increased cell death, we assessed cell viability fol- the lymph nodes as assessed by Thy1.2 and B220 staining were lowing stimulation with anti-CD3 (Table I). Viability was Ͼ90% similar in all genotypes examined (data not shown). Thus, T cell for all genotypes at the initiation of the culture (data not shown). development was not grossly altered in mice lacking both CD2 At 48 h following activation, there was a decrease in cell survival and CD28. for both the CD28 and CD2/CD28-deficent cells. However, the The Journal of Immunology 3593 Downloaded from

FIGURE 2. CD2/CD28-double deficient T cells respond to immobilized but not to soluble anti-CD3. A, Bulk lymph node cells from wild-type, CD2-, CD28-, or CD2/CD28-deficient mice were isolated and stimulated with graded doses of soluble anti-CD3. B, Lymph node cells from each genotype were stimulated with PMA (5 ng/ml) and ionomycin (0.1 ␮g/ml) or Con A (5 ␮g/ml) and proliferation determined at 48 h. C, Purified resting lymph node T cells were isolated from wild-type or CD2/CD28-deficient mice and stimulated with graded doses of plate immobilized anti-CD3. Proliferation was determined by thymidine incorporation for the final8hofa48-h culture. D, Bulk lymph node cells were stimulated with soluble anti-CD3 (10 ␮g/ml) alone http://www.jimmunol.org/ or with addition of exogenous IL-2 (100 U/ml). Proliferation was determined at 48 h. Data from one representative experiment are shown. magnitude of the difference is insufficient to account for the dif- but unlike the double knockout T cells, this impairment was over- ferences in proliferation suggesting that induction of cell death come at higher doses of Ag and was less significant at later time alone cannot explain the lack of proliferative response. points. CD2-deficient mice proliferate at levels similar to wild-type mice (Ref. 15 and data not shown). No significant differences were Ag-specific proliferation of CD2/CD28-deficient T cells is observed in stimulation of lymph node cells or splenocytes (data impaired in a dose- and time-dependent manner not shown). Thus, T cells lacking both CD2 and CD28 exhibit a by guest on October 2, 2021 To examine the proliferative response following stimulation in a profound impairment to Ag activation. more physiologically relevant context, we bred double-deficient Decreased T cell/APC conjugate formation in the absence of mice to the TCR transgenic mouse expressing the DO11.10 TCR. CD2 and CD28 The D011.10 mouse expresses a transgenic TCR specific for the d OVA323–339 peptide in the context of I-A (18). Expression of the As both CD2 and CD28 bind to ligands expressed on APC, both TCR transgene as assessed by flow cytometry with a clonotype- proteins may be involved in facilitating interactions between T specific Ab revealed similar levels of TCR expression in all genotypes cells and APC. We tested this hypothesis by measuring the ability (data not shown). Stimulation of CD2/CD28 double-deficient of double-deficient cells to form T cell:APC conjugates. DO11.10 splenocytes with OVA323–339 peptide demonstrated a dose- and T cells were co-incubated with Ag-pulsed APC and assessed for time-dependent impairment in the proliferative response (Fig. 3). conjugate formation by flow cytometry. As shown in Fig. 4, T cells At low Ag doses, no significant proliferative response was gener- lacking both CD2 and CD28 had a marked reduction in the number ated in the double-deficient T cells. Proliferation was detected only of stable conjugates formed as compared with either wild-type T at the highest peptide dose used (3 ␮M). In addition, the impaired cells or T cells from mice deficient in only CD2 or CD28. At the response of T cells from the double knockout mice was most sig- highest dose of Ag examined, no difference was observed between nificant at later time points. Mice lacking CD28 alone also dem- genotypes, consistent with the observation that some proliferative onstrated a time- and dose-dependent response to peptide/MHC, response can be elicited at these Ag doses. This suggests that both molecules work together to facilitate T cell interactions with APC.

Table I. Viability and activation marker expressiona Discussion To determine whether there is redundancy in the function of the CD2 CD28 % CD69ϩ % IL-2Rϩ % Viable accessory molecules CD2 and CD28, we generated mice deficient ϩ/Ϫϩ/Ϫ 34 34 51 in both proteins. In contrast to T cells lacking only CD2 or CD28, Ϫ/Ϫϩ/Ϫ 29 29 55 we found that T cells deficient in both proteins were severely im- ϩ/ϪϪ/Ϫ 38 24 44 paired in their ability to be activated and proliferate in response to Ϫ/ϪϪ/Ϫ 13 11 31 stimulation by either soluble anti-CD3 Ab or peptide Ag. T cells a Bulk lymph node cells were stimulated with soluble anti-CD3 (1.0 mg/ml), and from CD2 and CD28 double-deficient mice had a profound defect expression of CD69 and CD25 was determined after 18 and 48 h, respectively, by in response to soluble anti-CD3 stimulation. This was not due flow cytometry. The percentage of cells positive for each marker is shown. Viability was determined at 48 h by propidium iodide exclusion, and the percentage of live cells solely to impaired IL-2 secretion, as addition of exogenous IL-2 is shown. had little effect. Consistent with this finding, we found that double 3594 CD2 AND CD28 DETERMINE ACTIVATION THRESHOLD

FIGURE 3. Time- and dose-dependent impairment of Ag-specific proliferation in the absence of CD2 and CD28. Splenocytes from wild-type, CD28- deficient, and CD2/CD28-deficient D011.10 mice were isolated and stimulated with OVA323–339 peptide. Proliferation of replicate plates was measured at 48, 72, and 96 h by tritiated thymidine incorporation. knockout mice failed to efficiently up-regulate the early activation activation of intracellular signaling cascades, as potential mecha- markers, CD69 and CD25. This suggests that the signaling defect nisms for either CD2 or CD28 function. Our data suggest that involves the earliest steps in T cell activation. ligation of either CD2 or CD28 is required for the initial activation Surprisingly, we found that purified T cells from the double- of T cells. However, their mechanism of action may be either deficient mice could be activated with immobilized anti-CD3. similar or distinct, but in the absence of both, the T cell is severely Differences in the mechanism by which soluble vs immobilized

impaired in its ability to respond to Ag. Downloaded from anti-CD3 activates T cells may therefore provide insights into how Most current models of T cell activation subscribe to the two- CD2 and CD28 function. Stimulation of T cells using soluble signal hypothesis which proposes that two signals are required for anti-CD3 relies on the presence of Fc-bearing cells to cross-link T cell activation (3). The first is transduced by the TCR, and the the TCR. In contrast, immobilized anti-CD3 results in activation in second by a costimulatory molecule. Although many molecules the absence of any accessory cells as the solid phase substrate on have been implicated as costimulatory molecules, the best studied which the Ab is immobilized provides for high degrees of cross- is CD28. CD28 is thought to transduce a required and specific http://www.jimmunol.org/ linking. That immobilized anti-CD3 activates double-deficient T biochemical signal that integrates with the signal initiated by the cells demonstrates that the intrinsic signaling capabilities of the TCR to effect T cell activation. As T cells lacking CD28 can still TCR are intact. The discrepancy in the ability of soluble vs im- be activated, either the signal transduced by CD28 can be delivered mobilized anti-CD3 to activate double-deficient T cells suggests by another molecule on the surface of the T cell- or TCR-mediated that CD2 and CD28 may function primarily by stabilizing the in- signals alone may be sufficient. teraction of the T cell with the APC. In the absence of CD2 and Evidence is accumulating to support the latter possibility. These CD28, interactions with Fc-bearing B cells and macrophages data suggest that CD28 may function mainly to enhance and am- might be impaired, resulting in ineffective TCR cross-linking. plify signals transduced by the TCR. Because CD28 is an adhesion

Breeding the double knockout mouse to a TCR transgenic molecule, it may function in concert with other adhesion molecules by guest on October 2, 2021 mouse enabled us to examine T cell activation by peptide Ag. such as CD2 to stabilize the formation of a close membrane con- Similar to soluble anti-CD3 stimulation, we found that Ag-induced tact between the T cell and the APC. CD28 engagement also pro- T cell proliferation was significantly impaired. Given that the TCR motes a cytoskeletal mechanism that recruits lipid rafts to the T is competent to signal in these cells, the most compelling expla- cell/APC contact (19, 20). These adhesive and cytoskeletal func- nation for the defect is the inability of T cells to form stable con- tions of CD28 result ultimately in promoting and stabilizing T cell tacts with the APC. In confirmation of this, direct assay of the interactions with APCs. This interaction is critical as TCR recog- ability of T cells to form conjugates with APC demonstrated a nition of Ag is dependent upon formation of a stable and organized marked reduction in T cell:APC pairs, particularly at submaximal contact structure known as the immunological synapse (13, 21). Ag stimulation. In support of this model, CD28 has been shown to be particu- While both CD2 and CD28 contribute to the formation of the T larly important for Ags with a short half-life. CD28-deficient mice cell:APC contact, this does not exclude other mechanisms, such as cannot respond to Ags that are rapidly cleared from the mouse, but can mount normal responses to Ags that are present in the mouse for several days (22, 23). These data suggest that one mechanism by which CD28 may regulate T cell proliferation is by enhancing TCR sensitivity and length of engagement rather than transducing a unique signal. Like CD28, CD2 is thought to play an important role in facilitating interactions between the T cell and the APC. CD2 binds specific ligands expressed on a wide range of APC. In humans, the principle ligand for CD2 is CD58, also known as LFA-3 (24). In rodents, the related molecule CD48 is the most active natural ligand for CD2 (12). Although the affinity of CD2 for its ligands in vitro is relatively low (10Ϫ5), recent evidence suggest that this affinity is physiologically relevant (25). FIGURE 4. T cell:APC conjugate formation. Lymph node T cells and As CD2 ligation can enhance signaling mediated by the TCR, it Ag-pulsed A20 cells were labeled with Cell Tracker Orange or Green dyes, respectively, and assayed for their ability to form stable conjugates. Data has been considered by some to be a costimulatory molecule (26, are expressed as the percentage of conjugates relative to wild type, where 27). CD2 also shares some structural features in common with wild type is set at 100%. The mean of two independent experiments is CD28. Both are members of the Ig superfamily, and each contains shown. Ig repeats in the extracellular domain. Bound to their ligands, they The Journal of Immunology 3595 span the same distance, 15 nm, similar to the distance spanned by 7. Linsley, P. S., P. M. Wallace, J. Johnson, M. G. Gibson, J. L. Greene, the TCR bound to peptide/MHC. Lastly, both contain relatively J. A. Ledbetter, C. Singh, and M. A. Tepper. 1992. Immunosuppression in vivo by the soluble form of the CTLA-4 T cell activation molecule. Science 257:792. long cytoplasmic domains that contain potential sites of binding of 8. Turka, L. A., P. S. Linsley, H. Lin, W. Brady, J. M. Leiden, R.-Q. Wei, Src homology 3 (SH3) domain containing proteins. Both mole- M. L. Gibson, X.-G. Zheng, S. Myrdal, D. Gordon, et al. 1992. T-cell activation cules are likely to be involved in facilitating cell-cell contact be- by the CD28 ligand B7 is required for cardiac allograft rejection in vivo. Proc. Natl. Acad. Sci. USA 89:11102. tween the T cell and the APC. 9. Green, J. M., P. J. Noel, A. I. Sperling, T. L. Walunas, G. S. Gray, As mice lacking CD2 are relatively normal, it seemed likely that J. A. Bluestone, and C. B. Thompson. 1994. Absence of B7-dependent responses the role of CD2 in T cell biology was not essential and could be in CD28-deficient mice. Immunity 1:501. 10. Shahinian, A., K. Pfeffer, K. P. Lee, T. M. Ku¨ndig, K. Kishihara, A. Wakeham, substituted by other molecules on the surface of the T cell (15). In K. Kawai, P. S. Ohashi, C. B. Thompson, and T. W. Mak. 1993. Differential T this paper we present data supporting a model where CD2 and cell costimulatory requirements in CD28-deficient mice. Science 261:609. CD28 share some functions. We have shown previously that CD2 11. Lin, H., J. C. Rathmell, G. S. Gray, C. B. Thompson, J. M. Leiden, and M. L. Alegre. 1998. Cytotoxic T lymphocyte antigen 4 (CTLA4) blockade ac- clustering plays an essential role in helping to form a specialized celerates the acute rejection of cardiac allografts in CD28-deficient mice: CTLA4 cell contact called the immunological synapse (21). Preliminary can function independently of CD28. J. Exp. Med. 188:199. experiments suggest that CD28 can also induce protein clustering 12. Davis, S. J., and P. A. van der Merwe. 1996. The structure and ligand interactions of CD2: implications for T-cell function. Immunol. Today 17:177. and segregation and thus may function in a manner similar to CD2 13. Shaw, A. S., and M. L. Dustin. 1997. Making the T cell receptor go the distance: (S. Bromley and M. Dustin, personal communication). Further- a topological view of T cell activation. Immunity 6:361. more, it is also possible that the cytoplasmic tails of CD2 and 14. Teh, S. J., N. Killeen, A. Tarakhovsky, D. R. Littman, and H. S. Teh. 1997. CD2 regulates the positive selection and function of antigen-specific CD4ϪCD8ϩ T CD28 interact with the same or similar intracellular proteins. Al- cells. Blood 89:1308. though the precise mechanism by which CD2 and CD28 coordi- 15. Killeen, N., S. G. Stuart, and D. R. Littman. 1992. Development and function of nately regulate the process of T cell activation remains to be de- T cells in mice with a disrupted CD2 gene. EMBO J. 11:4329. Downloaded from fined, the data presented in this manuscript demonstrate that each 16. Luce, G. G., S. O. Sharrow, S. Shaw, and P. M. Gallop. 1985. Enumeration of cytotoxic cell-target conjugates by flow cytometry using internal fluorescent may function to facilitate the interaction with APCs. The binding stains. BioTechniques 3:270. of CD2 and CD28 with their ligands would allow for efficient TCR 17. Noel, P. J., L. H. Boise, J. M. Green, and C. B. Thompson. 1996. CD28 co- engagement, especially in the setting of limiting Ag. Thus, the stimulation prevents cell death during primary T cell activation. J. Immunol. 157:636. engagement of the accessory molecules CD2 and CD28 may ef- 18. Murphy, K. M., A. B. Heimberger, and D. Y. Loh. 1990. Induction by antigen of fectively set the threshold for T cell activation by Ag. intrathymic apoptosis of CD4ϩCD8ϩTCRlo thymocytes in vivo. Science 250: http://www.jimmunol.org/ 1720. 19. Viola, A., S. Schroeder, Y. Sakakibara, and A. Lanzavecchia. 1999. T lympho- Acknowledgments cyte costimulation mediated by reorganization of membrane microdomains. Sci- ence 283:680. We thank M. Dustin, A. Chan and R. Arch for helpful discussion and 20. Wulfing, C., and M. M. Davis. 1998. A receptor/cytoskeletal movement triggered review of the manuscript. We also thank N. Killeen and J. Bromberg for by costimulation during T cell activation. Science 282:2266. providing CD2-deficient mice and S. Reiner and C. Thompson for 21. Dustin, M. L., M. W. Olszowy, A. D. Holdorf, J. Li, S. Bromley, N. Desai, CD28-deficient mice. P. Widder, F. Rosenberger, P. A. van der Merwe, P. M. Allen, and A. S. Shaw. 1998. A novel adaptor protein orchestrates receptor patterning and cytoskeletal polarity in T-cell contacts. Cell 94:667.

References 22. Ku¨ndig, T. M., A. Shahinian, K. Kawai, H.-W. Mittrucker, E. Sebzda, by guest on October 2, 2021 M. F. Bachmann, T. W. Mak, and P. S. Ohashi. 1996. Duration of TCR stimu- 1. Quill, H., and R. H. Schwartz. 1987. Stimulation of normal inducer T cell clones lation determines costimulatory requirement of T cells. Immunity 5:41. with antigen presented by purified Ia molecules in planar lipid membranes: spe- 23. Bachmann, M. F., K. McKall-Faienza, R. Schmits, D. Bouchard, J. Beach, cific induction of a long-lived state of proliferative nonresponsiveness. J. Immu- D. E. Speiser, T. W. Mak, and P. S. Ohashi. 1997. Distinct roles for LFA-1 and nol. 138:3704. CD28 during the activation of naive T cells: adhesion versus costimulation. Im- 2. Jenkins, M. K., and R. H. Schwartz. 1987. Antigen presentation by chemically munity 7:549. modified splenocytes induces antigen-specific T cell unresponsiveness in vitro and in vivo. J. Exp. Med. 165:302. 24. Moingeon, P., H. C. Chang, P. H. Sayre, L. K. Clayton, A. Alcover, P. Gardner, 3. Bretscher, P., and M. Cohn. 1970. A theory of self-nonself discrimination. Sci- and E. L. Reinherz. 1989. The structural biology of CD2. Immunol. Rev. 111:111. ence 169:1042. 25. Dustin, M. L., L. M. Ferguson, P. Y. Chan, T. A. Springer, and D. E. Golan. 1996. 4. June, C. H., J. A. Bluestone, L. M. Nadler, and C. B. Thompson. 1994. The B7 Visualization of CD2 interaction with LFA-3 and determination of the two-di- and CD28 receptor families. Immunol. Today 15:321. mensional dissociation constant for adhesion receptors in a contact area. J. Cell 5. Harding, F. A., J. G. McArthur, J. A. Gross, D. H. Raulet, and J. P. Allison. 1992. Biol. 132:465. CD28-mediated signalling co-stimulates murine T cells and prevents induction of 26. Bierer, B. E., A. Peterson, J. C. Gorga, S. H. Hermann, and S. J. Burakoff. 1988. anergy in T-cell clones. Nature 356:607. Synergistic T cell activation via the physiological ligands for CD2 and the T cell 6. Lenschow, D. J., Y. Zeng, J. R. Thistlethwaite, A. Montag, W. Brady, receptor. J. Exp. Med. 168:1145. M. G. Gibson, P. S. Linsley, and J. A. Bluestone. 1992. Long-term survival of 27. Bierer, B. E., and S. J. Burakoff. 1989. T-lymphocyte activation: the biology and xenogeneic pancreatic islet grafts induced by CTLA4Ig. Science 257:789. function of CD2 and CD4. Immunol. Rev. 111:267.