Eur. J. Immunol. 1991. 21: 1277-1282

Nitin K. Damle, Peter S. Lmsley and Jeffrey A. Ledbetter Oncogen Division, Bristol-Myers Squibb Pharmaceutical Research Institute, Seattle

Cognate Th-B cell interaction involves CD28 and B7

1277

Direct helper T cell-induced B cell differentiation involves interaction between T cell antigen C D B and B cell activation antigen B7 Cognate interactions between major histocompatibility complex class I1 antigen (Ag)-reactive CD4+ T helper (Th) and Ag-presenting B cells induce first the activation of B cells and their subsequent differentiation into Ig-secreting cells (IgSC). The Th cell-associated homodimeric glycoprotein CD28 has been implicated as an important regulator of T h activation. Recently, B cell-associated early activation Ag B7 has been identified as a ligand for the CD28 molecule. I n this study, we have examined using monoclonal antibodies (mAb) the roles of CD28 and B7 molecules during the Th-B cell cognate interactions leading to the differentiation of B7+ B cells. Anti-CD28 mAb 9.3 specifically inhibited proliferative responses of CD4+ T cells to both allogeneic B cells and soluble Ag-presenting autologous non-Tcells. In addition, anti-CD28 mAb 9.3 inhibited Th-induced differentiation of alloantigen-presenting B cells into ISC. Similar inhibition of both Ag-induced Th activation and B cell differentiation into ISC was observed using mAb BB1 which recognizes a B cell-associated molecule B7. In contrast, non-cognateTh-independentexogenous interleukin 6-induced differentiation of B7+ B cells into ISC was not inhibited by mAb to either molecule. These results clearly demonstrate the involvement of CD28 on Th and its ligand B7 on B cells during cognate Th-Binteractions leading to the differentiation of B cells. Furthermore, these results also suggest the development of new mAb-based therapeutic approaches for exaggerated B cell activation associated with certain autoimmune diseases such as systemic lupus erythematosus.

1 Introduction

include CD2 (LFA-2): CD58 (LFA-3), CD4: MHC class 11, CDllalCD18 (LFA-1): CD54 (ICAM-1) [8, 91.

Optimal activation of B lymphocytes and their subsequent differentiation into Ig-secreting cells (IgSC) is critically dependent on the helper effects of CD4+ Th cells and is mediated via both direct (cognate) Th-B cell intercellular contact-mediated interactions and the elaboration of Agnonspecific cytokines (non-cognate activation; [1-41). Although Th-derived cytokines can stimulate B cells [4], their synthesis and directional exocytosis is initiated and sustained via cognate interactions between Ag-primed T h and Ag-presenting B cells [4, 51. B cells make use of their sIg receptors to intracellularly concentrate and process native Ag and this processed nominal Ag in association with MHC class I1 molecules can then be presented to T h cells [6,7]. While the Ag-specificity of Th-B interactions is determined by the interaction between a nominal Agpresented in conjunction with MHC class I1 molecules on the surface of B cells and the TcR/CD3 complex on the surface of Th cells leading to the activation of Th and B cells, the successful outcome of Th-B interactions requires participation of additional transmembrane receptor-ligand (conter-receptor) pairs of co-stimulatory accessory/adhesion molecules on the surface of T h and B cells which

One such co-stimulatory/accessory molecule is CD28, a homodimeric glycoprotein belonging to the Ig superfamily and expressed on almost all CD4+ T h cells [lo]. Current evidence suggests that this molecule functions in an alternative Tcell activation pathway which is distinct from that initiated via the TcR/CD3 complex [ ll] . mAb to the CD28 molecule can augment T cell responses initiated by various polyclonal stimuli [12-171, in part as a consequence of not only increased production of cytokines and their mRNA stability but also increased responsiveness to these cytokines [18-201. In contrast, anti-CD28 mAb can also exert inhibitory effects on the activation of Th cells by alloantigens or soluble Ag indicating that CD28-mediated regulatory effects on the activation of Th cells are dictated by the nature of the stimulus [17,21,22].We have recently shown that the CD28 molecule onTh cells can be utilized to mediate adhesion with activated B cells by interacting with another member of Ig superfamily, B7, which is an early B cell activation Ag [23-261. In the present study, using alloantigen-driven maturation of B cells as a model system, we demonstrate for the first time the involvement of CD28 molecule on the surface of CD4+ T h cells and that of B7 on B cells during the COgnateTh-Binteraction leading to B cell differentiation into IgSC.

[I 91881 Correspondence: Nitin K. Damle, Oncogen Division, BristolMyers Squibb Pharmaceutical Research Institute, 3005 First Avenue, Seattle, WA 98121, USA Abbreviations:

CM: Complete culture medium IgSC: Ig-

secreting cell(s) 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991

2 Materials and methods 2.1 Culture medium Complete culture medium (CM) consisted of RPh4I 1640 (Irvine Scientific, Santa Ana, CA) supplemented with 0014-2980/91/0505-1277$3.50+ .25IO

1278

Eur. J. Immunol. 1991. 22: 1277-1282

N . K. Damle, I? S. Linsley and J. A. Ledbetter

100 U/ml of penicillin G, 100 pg/ml of streptomycin, 2 mM L-glutamine, 5 X M 2-ME, and 10% FBS (Irvine Scientific). 2.2 Cells and mAb

EBV-transformed B cell lines CESS (HLA-A1,A3; B5,B17; DR7), JIJOYE, and SKW6.4 (HLA-A2a; B27,B51; DR7), were obtained from American Type Culture Collection (ATCC, Rockville, MD). EBV-transformed B cell lines ARENT (HLA-A2; B38,B39; DRw6) and MSAB (HLA-A1,A2; B57; DR7) were generously provided by Dr. Edgar G. Engleman, Stanford University School of Medicine, Stanford, CA. Hybridomas OKT4 (IgG anti-CD4), OKT8 (IgG anti-CD8) and HNKl (IgM anti-CD57) were obtained from ATCC and ascitic fluids from these hybridomas were generated in pristane-primed BALB/c mice. Production and characterization of antiCD28 mAb9.3 (IgGza) has been described [13-151. mAb 4H9 (IgGz, anti-CD7; [20]) was generously provided by Dr. Edgar E. Engleman and mAb anti-B7 (BB1; IgM; [26]) by Dr. Edward A. Clark, University of Washington, Seattle, WA . PBMC from healthy donors were separated into T and non-Tcells using sheep erythrocyte rosetting technique, and Tcells were separated by panning into CD4+ subset and further into CD4+CD45RKCD45RO+memory subpopulation as described [27]. 2.3 Proliferative responses of T cells

To examine the effect of Th ceUs on Ig production, 2 x 104-2.5 x 104cells from HLA-DR7+ EBV-transformed B cell lines, IgM-producing SKW or IgG-producing CESS, were cultured with varying numbers of DR7-primed CD4+CD45ROf Th cells for 96 h after which cell-free SN from these cultures were collected and assayed for the quantitation of IgM (SKWcultures) or IgG (CESS cultures) using solid-phase ELISA. Exogenous IL6 (1-100 U/ml) induced Ig production by these B cells was also examined.

2.5 Ig quantitation IgG or IgM in culture SN were measured using solid-phase ELISA as described [29]. Briefly, 96-well flat-bottom microtiter ELISA plates (Corning, Corning, NY) were coated with 200 pl/well of sodium carbonate buffer (pH 9.6) containing 10 pg/ml of affinity-purified goat antihuman IgG or IgM Ab (Tago, Burlingame, CA), incubated overnight at 4 "C, and then washed with PBS and wells were further blocked with 2% BSA in PBS (BSA-PBS). Samples to be assayed were added at appropriate dilution to these wells and incubated with 200 pl/well of 1 : lo00 dilution of horse radish peroxidase (HRP)-conjugated F(ab'):! fraction of affinity-purified goat anti-human IgG or IgM Ab (Tago). The plates were then washed, and 100pl/well of ophenylenediamine (Sigma, St. Louis, MO) solution (0.6 mg/ml in citrate-phosphate buffer with pH 5.5 and 0.045% hydrogen peroxide). Color development was stopped with 2 N sulfuric acid. Absorbance at 490 nm was measured with an automated ELISA plate reader. Test and control samples were run in triplicate and the values of absorbance were compared to those obtained with known IgG or IgM standards run simultaneously with the SN samples to generate the standard curve using which the concentrations of Ig in culture SN were quantitated. Data are expressed as ng/ml of Ig f SEM of either triplicate of cultures.

Fifty thousand CD4+CD45RO+T cells were cultured with 1 x lo4 irradiated (8000 rad from a 13'Cs source) EBVtransformed allogeneic B cells in 0.2 ml of CM in roundbottom microtiter wells in a humidified 5% COz and 95% air atmosphere. CD4+CD45RO+Tcells also were independently stimulated with 100 pg/ml of soluble purified protein derivative of tuberculin (PPD, Connought Laboratories, Willowdale, Ontario, Canada) in the presence of 1 x lo4 irradiated (3000 rad) autologous non-T cells. Triplicate 3 Results cultures were pulsed with 1 @/well = 37 kBq/well of ["H]dThd (6.7 Ci/mmol, NEN, Boston, MA) 16 h before Cognate interaction between CD4+ Th and Ag-presenting harvesting of cells for measurement of radiolabel incorpo- B cells result in the activation and differentiation of both ration into newly synthesized DNA. The results are cell types consequently leading to the development of Ig-secreting cells [4]. Allogeneic MLR offers an ideal expressed as cpm f SEM. system to analyze cognate Th-B cell interaction because alloantigen-specific CD4+ Th induce both the activation and differentiation of alloantigen-bearing B cells into 2.4 T h cell-induced Ig production by B cells Ig-secreting cells [30-341. The involvement of CD28 on Th DR7-primed CD4+Th cells were derived from the allogene- cells and its ligand B7 during the activation of Th and B cells ic MLC consisting of responder CD4+CD45RO+ T cells in the allogeneic MLR was first examined using murine (HLA-A26,A29; B7,B55; DR9,DRlO) and irradiated mAb directed at these molecules. Fig. 1shows the results of MSAB (DR7+) B cells as stimulator cells as described one such experiment in which the proliferative responses of before [28].The isolation of resting CD4+CD45RO+Tcells CD4+CD45RO+ T cells to various allogeneic B cells were and that of DR7-primed CD4+CD45RO+ T lymphoblasts examined in the presence of murine mAb to CD7, CD28, using discontinuous Percoll density gradient centrifugation CD57 or B7. EBV-transformed B cell lines were used as was also previously described [28]. These DR7-primed Th stimulator cells in these experiments because these B cells cells were continuously propagated in the presence of exhibit various features of activated B cells such as the irradiated MSAB B cells and 50 U/ml of IL2. Prior to their expression of high levels of MHC class I1 and B7 molecules functional analysis, viable DR7-primed Th cells were [24, 261. isolated by Ficoll-Hypaque gradient centrifugation and maintained overnight in CM without DR7+ feeder cells or The presence of anti-CD28 mAb (9.3; IgG2,) but not that of IL 2. isotype-matched anti-CD7 mAb (4H9; IgG2,) consistently

Cognate Th-B cell interaction involves CD28 and B7

Eur. J. Immunol. 1991. 21: 1277-1282

1279

Zunn,

No mAb AntLCD7

zoo00

AntLCD28 AntLCD57

lunn,

Anti-BB1

Figure 1. Effect of anti-CD28and anti-B7 mAb on the day-7 proliferative responses of T cells. Fifty thousand freshly isolated resting CD4+CD45RO+T cells were stimulated with irradiated allogeneic cells (1 x 104 EBV-transformed B cells or 2.5 X 104 non-T cells) in the presence of 10 clg/mlof mAb to CD7, CD28, CD57 or B7 molecule. CD4+CD45RO+Tcells were also stimulated with PPD (100 pglml) and 1 x 104 irrdiated autologous non-Tcells in the presence of the above mAb. Proliferative responses were examined o n day 7 of culture.

loo00

sood

0

Arent

Ccss

Jijoyc

Allo. NoD-T

Auto. NOD-" + PPD

Stimulus

inhibited the MLR proliferative response of CD4+ Tcells to allogeneic B cells. Similarly, the addition of anti-B7 mAb (BB1; IgM) but not that of isotype-matched anti-CD57 HNK1; IgM) to the allogeneic MLR resulted in the inhibition of T cell proliferation. The inhibitory effects of anti-CD28 mAb9.3 on the MLR responses of T cells are consistent with our previous observations [17,21]. Similar to the allogeneic MLR, proliferative response of CD4+ T cells to soluble Ag PPD presented by autologous non-Tcells was also inhibited by anti-CD28 and anti-B7 mAb. Although both anti-CD28 mAb9.3 (IgG2,) and anti-B7 mAb BB1 (IgM) inhibited the allogeneic MLR and the soluble Ag-induced proliferative responses, anti-CD28mediated inhibition was always stronger than that by anti-B7 for all the responder-stimulator combinations examined, perhaps due t o the fact that anti-B7, unlike anti-CD28, is an IgM mAb and may have a lower affinity for B7 than that of anti-CD28 for CD28. These observations are also consistent with the weaker ability of anti-B7 mAb to block the CD28-mediated adhesion to B7+ B cells ~31. Allogeneic MHC class 11-activated CD4+ Th cells induce first the activation and subsequently the differentiation of appropriate MHC class 11-bearing B cells into IgSC [4, 30-341. To examine the roles of CD28 and B7 during such cognateTh-B interactions, we used two EBV-transformed B cell lines, IgG-secreting DR7+ CESS and IgM-secreting DR7+ SKW. When appropriately stimulated both these B cell lines significantly increase their production of the respective Ig isotype. We first examined the effects of DR7-specific CD4+CD45ROf Th line on the Ig production by both CESS and SKW B cells.Thus, B cells were cultured for 4 days with various concentrations of DR7-primed CD4+ T h cells in the absence of any exogenous growtwdifferentiation factor@)after which Ig secreted in the cell-free SN was quantitated by using a solid-phase ELISA. Freshly isolated CD4+CD45RO+ T cells (autologous to the DR7primed CD4+ T h ) were used as a control for DR7-primed CD4+Th cells. IL 6 was used as a positive control to monitor the non-cognate Ig production by these B cell lines. Fig. 2

o

h

-.--.-T

-

~

*

-. '

--...-I

10' lo2 lo3 lo4 l o 5 lo6 CD4+CD45RO+ T cells / 25,000 SKW B cells

*

..---I

10' lo2 l o 3 lo4 lo5 lo6 CD4+CD45RO+ T cells I 25,000 CESS B cells Figure2 Th cell-induced differentiation of B cells into Igsc. DR7-primed CD4+CD45RO+Th cells were added in increasing numbers to 2.5 x 104 CESS or SKW B cells and cultured for 4 days after which IgM (a) or IgG (b) in culture SN were quantitated. Ig production induced by frehsly isolated resting CD4+CD45RO+T cells was also simultaneously examined.

N. K. Damle, €? S. Linsley and J. A. Ledbetter

1280

shows the Ig production by either B cell line as a function of the concentration of DR7-primed Th with optimal Ig producion induced at either 1: 1 or 1:2 T h :B ratios. At Th : B ratios higher than 1 : 1inhibition of Ig production was observed. Hence, all further experiments were carried out at Th:B ratio of 1 :2. Freshly isolated resting CD4+CD45RO+ T cells, autologous to the above DR7primed CD4+ Th, were also simultaneously examined for their ability to induce Ig synthesis by CESS or SKW B cells. As shown in Fig. 2, these unprimed resting CD4+ Tcells slightly induced IgM production by SKW B cells but had no effect on the IgG production by CESS B cells in 4-day cultures. This slight helper effect observed with unprimed CD4+CD45ROf T cells may be due to recruitment and partial activation of DR7-responsive T cells from the CD4+CD45RO+ population during the Ig induction cultures. The reasons for the inability of the same CD4+CD45ROf T cells to exert similar helper effects on

.-

lo00

-

800

-

v1

Y v)

5 h

600-

v)

5

I

400

3000

-

2500

-

2000

-

E . oc

5

.-

1

-

Anti-CD7

\

AntLCD28

+ Anti-CDS7 + Anti-BBl

5

'h

v1

-

Anti-CD7

.*

Anti-BBI

.

P 1500

T

- + Antl-CD57

...-

The roles of CD28 and B7 during cognateTh : B-induced Ig production was further examined. Both CESS and SKW B cells constitutively express B7 Ag on their surface and thus, represent a source of uniformly activated B cell populations for their use in Th-B cognate interactions or in cytokinedriven non-cognate maturation. Thus, either DR7+ B cells were cultured for 4 days with DR7-specific CD4+ Th line at Th :B ratio of 1: 2 and to those cultures mAb to CD28 or B7 were added at different concentrations. Ig production at the end of these 3-day cultures was examined. Fig. 3 shows that both anti-CD28 and anti-B7 but not their isotype-matched mAb controls (anti-CD7 and anti-CD57, respectively) inhibited in a dose-dependent manner Th-induced Ig production by B cells. Once again, anti-CD28 mAb-mediated inhibition of Ig production was stronger than that by anti-B7 mAb. In contrast, Ig production by either B cells induced with exogenous IL 6 (noncognate differentiation) was not affected by any of the above mAb (data not shown). These results strongly suggest that the interaction between CD28 and B7 during cognate Th-B collaboration is pivotal to the differentiation of activated B cells in ISC.

The present study adds CD28 and its ligand B7 [23] to the list of co-stimulatory transmembrane receptor-ligand pairs influencing T h :B interactions. Involvement of both CD28 and B7 during Th:B collaboration was evident from inhibition by anti-CD28 and anti-B7 of not only Th cell activation but alsoTh-induced differentiation of B cells into IgSC. At which step during T h : B collaboration CD28 : B7 interaction manifests its effects is not known at present. However, taking into account the receptor-ligand relationship between CD28 and B7 [23], the observed inhibitory effects of anti-CD28 and anti-B7 mAb might have been due to the inhibition of CD28 :B7 interaction underlying these responses.

Anti-CD28

' I .-.-."I 1000 0.0001 0.001 0.01

DR7+ CESS B cells remain unclear and may be due to the differences in the states of maturation of CESS and SKW B cells in addition to the less optimal activation of CD4+CD45RO+ T cells. The production of Ig by CESS (IgG) or SKW (IgM) B cells induced by DR7-primed CD4+ Th was specific for HLA-DR7 because similarly activated DRw6-primed CD4+ T h (stimulated with DRw6+ ARENT B cells and autologous to the DR7-primed Th) were unable to induce Ig production by either CESS or SKW B cells (data not shown).

4 Discussion

v)

v1

Eur. J. Immunol. 1991. 22: 1277-1282

.-'-0.1

. 1

-

-4

..'"I

10

100

mAb pglml

Figure 3. Effect of anti-CD28 and anti-B7 mAb on the Th cellinduced production of Ig by B cells.'Ik.enty thousand DR7+ B cells (CESS or SKW) were cultured for 4 days with 1 x 104 DR7-primed CD4+CD45RO+Th cells in the presence of increasing concentrations of mAb against CD7, CD28, CD57 or B7. IgM (a) and IgG (b) contents of cell-free SN were then quantitated.The SEM of these cultures was < 10% of the mean for each data point.

During cognate Th :B interaction although both T h and B cells cross-stimulate each other, their functional differentiation is critically dependent on the provision by T h cells of growth- and differentiation-inducing cytokines such as IL 2, IL 4 and IL6 [l-41. Interaciton between CD28 and B7 may significantly influence the production of cytokines which may profoundly influence B cell differentiation [18, 191. Indeed, recent studies from our laboratory indicate that binding of B7 to CD28 enhances accumulation of IL2 mRNA and subsequent Th cell proliferation [35]. Studies by Po0 et al. [5] on cloned T h :B interaction have indicated that interaction of TcR with nominal Ag-MHCclass I1 on B cells results in focused release of Th cell-derived cytokines in the area of Th and B cell contact (vectorially oriented exocytosis).This may ensure the activation of only B cells

Eur. J. Immunol. 1991. 21: 1277-1282

presenting Ag to T h cells, and also avoids activation of bystander B cells. Although the role of CD28 during the above TcR-directed vectorially oriented exocytosis of Th cell-derived cytokines has not been explored, ligation of CD28 by B7 during Th : B collaboration may facilitate sustained synthesis and delivery of these cytokines for their utilization during the differentiation of B cells into IgSC. Of interest in this context was lack of inhibition by anti-CD28 and anti-B7 mAb of Th cell-independent differentiation of CESS or SKW B cells induced with exogenous IL4 or IL 6, further suggesting that CD28 :B7 interaction controls either production of these cytokines or their focused delivery on B cells, or both of these events. Although the role of CD28 : B7 interaction during cognate Th : B collaboration was emphasized in the present study, involvement of CD28 and B7 may not be restricted toTh : B interactions, and may apply more generally to other Ag-presenting cells such as monocyteMQ, dendritic cells, epidermal Langerhans cells and perhaps other APC. AntiCD28 mAb significantly inhibits the responses of CD4+ T cells to not only allogeneic stimuli but also to soluble Ag presented by autologous non-T cells which include both monocytes and B cells [17, 21, 221. Resting B cells do not express B7 on their surface [24,36], but activation of B cells by cross-linking sIg or MHC class I1 with Ab or mitogens up-regulates the expression of B7 [24]. Studies by Freedman et al. [25] have shown that monocytes express low levels of mRNA for B7, although the surface expression of B7 on monocytes was not examined in their study. It remains to be seen whether monocytes upon activation up-regulate surface expression of B7. In any case, it is conceivable that ligation of nominal Ag : MHC class I1 on APC by TcR on Th leads to elevated expression of B7 by these APC [36] which then facilitates via its interaction with CD28, the production of various cytokines by Th which would in turn drive both growth and differentiation of both Th and B cells. Although CD28 on Th plays a pivotal role during both Th activation and Th-induced B cell differentiation, it should be pointed out that in the absence of participation of TcR/CD3 complex on Th cells, CD28:B7 interaction is unable to exert its various functional effects. This was clearly evident when HLA-DRw6-primed CD4+CD28+T h were unable to direct maturation of CESS or SKW B cells (both B7+HLA-DR7+; N. K. Damle, unpublished observation). These observations are consistent with earlier reports on the MHC class I1 specificity of the allogeneic Th :B interaction inducing B cell activation [32-341. Thus, TcR on Th cells directs not only the antigen specificity of Th:B interaction but also the initiation of Th activation resulting in the transcription of various cytokine genes as well as directed exocytosis of Th cell-synthesized cytokines toward apposing interacting B cells. Ligation of CD28 by B7 may sustain the production of these cytokines by increasing via as yet unknown mechanism(s) the stability of cytokine mRNA as shown in the studies of Lindsten et al. [18, 191. In addition, although not addressed in this study, engagement of TcR/CD3 on Th cells may influence avidity of interaction between CD28 and B7, similar to that between CDlldCD18 (LFA-1) and CD54 (ICAM-1) on apposing cells [37]. In summary, this study shows that CD28 and its ligand B7 play a pivotal role during both the activation of CD4+ T h

Cognate Th-B cell interaction involves CD28 and B7

1281

cell and Th-induced differentiation of B cells. Furthermore, inhibitory effects of anti-CD28 and anti-B7 mAb on the cognate Th : B interaction demonstrated in this study may also provide the basis for devising new strategies targeting CD28 and B7 molecules for possible clinically advantageous therapeutic applications in the treatment of various autoimmune disorders such as insulin-dependent diabetes mellitus, myasthenia gravis, rheumatoid arthritis and SLE. We thank Drs. Edgar G . Engleman and Edward A . Clark for their generous supply of mAb, and Drs. Ingegerd Hellstrom and Karl-Eric Hellstrom for their continued support and encouragement.

Received December 28, 1990.

5 References 1 Noel, R. J. and Snow, E.C., Immunol. Today 1990. 11: 361. 2 Kupfer, A . , Swain, S. L., Janeway, C. A . , Jr. and Singer, S. J.. Proc. Natl. Acad. Sci. USA 1986. 83: 6080. 3 Brian, A. A . , Proc. Natl. Acad. Sci. USA 1988. 85: 654. 4 Moller, G. (Ed.), Immunol. Rev. 1987. 99: 1. 5 Poo, W.-J., Conrad, L. and Janeway, C. A , , Jr., Nature 1988. 332: 378. 6 Rock, K. L., Benacerraf, B. and Abbas, A . K . , J. Exp. Med. 1984. 160: 1102. 7 Lanzavecchia, A., Nature 1985. 314: 537. 8 Springer,T. A., Nature 1990. 346: 425. 9 Moller, G. (Ed.), Immunol. Rev. 1990. 114: 1. 10 June, C., Ledbetter, J. A., Linsley, €! S. and Thompson. C. B.. Immunol. Today 1990. 11: 211. 11 June, C. H . , Ledbetter, J. A . , , Gillespie, M. M . , Lindsten, T. and Thompson, C. B., Mol. Cell Biol. 1987. 7: 4472. 12 Gmunder, H. and Lesslauer,W. A., Eur. J. Biochem. 1984.142: 153. 13 Ledbetter, J. A . , Martin, F’. J., Spooner, C. E.,Wofsky, D.,Tsu, T.T., Beatty, I? G. and Gladstone, F’., J. Imrnunol. 1985. 13s: 2331. 14 Hara,T. Fu, S. M. and Hansen, J. A . , J. Exp. Med. 1985. 161: 1513. 15 Martin, I? J., Ledbetter, J. A . , Morishita, Y.,June, C. H . , Beatty, P. G. and Hansen, J. A . , J. Immunol. 1986. 136: 3282. 16 Weiss, A., Manger, B. and Imboden, J., J. Immunol. 1986.137: 819. 17 Damle, N. K., Doyle, L.V., Grosmaire, L. S. and Ledbetter, J.A., J. Immunol. 1988. 140: 1753. 18 Thomson, C. B., Lindsten,T., Ledbetter, J. A., Kunkel, S. L.. Young, H. A., Emerson, S. G., Leiden, J. M. and June, C. H., Proc. Natl. Acad. Sci. USA 1989. 86: 1333. 19 Lindsten, T., June, C. H., Ledbetter, J. A . , Stella, G. and Thompson, C. B., Science 1989. 244: 339. 20 Damle, N. K. and Doyle, L. V., J. Immunol. 1989. 143: 1761. 21 Damle, N. K., Hansen, J. A., Good, R. A . and Gupta, S., Proc. Natl. Acad. Sci. USA 1981. 78: 5096. 22 Lesslauer,W., Koning, F., Ottenhoff,T., Giphart, M., Goulmy. E. and Van Rood, J. J., Eur. J. Immunol. 1986. 16: 1289. 23 Linsley, I? S., Clark, E. A. and Ledbetter, J. A . , Proc. Natl. Acad. Sci. USA 1990. 87: 5031. 24 Freeman, A. S., Freeman, G . , Horowitz, J. C., Daley, J. and Nadler, L. M., J. Immunol. 1987. 139: 3260. 25 Freeman, G. J., Freedman, A . S., Segil, J. M., Lee, G . . Whitman, J. F. and Nadler, L. M., J. Immunol. 1989. 143: 2714. 26 Tokochi,T., Holly, R. D. and Clark, E.A., J. Immunol. 1981. 128: 823.

1282

N. K. Damle, P. S. Linsley and J. A. Ledbetter

27 Damle, N. K., Childs, A. L. and Doyle, L.V., J. Imrnunol. 1987. 139: 1501. 28 Damle, N. K., Mohagheghpour, N. and Engleman, E. G., J. Irnrnunol. 1984. 133: 1235. 29 Volkman, D. H., Lane, H. C. and Fauci, A. S., Proc. Natl. Acad. Sci. USA 1981. 78: 2528. 30 Chiorazzi, N., Fu, S. M. and Kunkel, H. G., lrnrnunof. Rev. 1979. 45: 219. 31 Kotzin, B. L., Benike, C. J. and Engleman, E. G., J. Irnrnunol. 1981. 127: 931. 32 Friedman, S. M.,Thompson, G. S. and Pricipato, M. A., J. Irnrnunol. 1982. 129: 2541.

Eur. J. Immunol. 1991. 21: 1277-1282 33 Goldberg, D., Green, A., Gottlieb, A. B., Crow, M. K., Lewison, A. and Friedman, S. M., J. Irnrnunol. 1985. I35: 1012. 34 Crow, M. K., Jover, J. A. and Friedman, S. M., J. Exp. Med. 1986. 164: 1760. 35 Linsley, P. S., Brady,W., Grosmaire, L., Aruffo, A., Damle, N. K. and Ledbetter, J. A., J. Exp. Med. 1991. 173: 721. 36 Koulova, L., Clark, E. A., Shu, G. and Dupont, B., J. Exp. Med. 1991. 173: 759. 37 Dustin, M. L. and Springer,T. A., Nature 1989. 341: 619.