Proc. Natl. Acad. Sci. USA Vol. 88, pp. 2623-2627, April 1991 Immunology
Adhesion versus coreceptor function of CD4 and CD8: Role of the cytoplasmic tail in coreceptor activity (T-cell
activation/p56kk/T-ceil receptor/accessory molecules)
M. CARRIE MICELI, PAUL VON HOEGEN, AND JANE R. PARNES* Division of Immunology and Rheumatology, Department of Medicine, Stanford University Medical Center, Stanford, CA 94305-5111
Communicated by Leonard A. Herzenberg, December 28, 1990
ABSTRACT CD4 and CD8 play an important role in T-cell recognition and activation; however, their mechanisms of action are not well understood. We compare the effects of expressing CD4 and CD8a either individually or together in a class U-restricted T-cell hybridoma. We also compare the effects of expressing truncated forms of CD4 or CD8a that do not have a cytoplasmic tail and thus do not associate with the T-cell-specifIc tyrosine kinase p56Ick, which has been implicated in T-cell activation. We demonstrate that, although CD4 and CD8a can specifically enhance interleukin 2 secretion, maximal potentiation occurs with expression of CD4, which, unlike CD8, can bind to the same major histocompatibility complex protein as the T-cell receptor. Our data further indicate that the cytoplasmic tail and/or the associated p56&k are primarily significant for interleukin 2 secretion by the hybridomas we have examined when CD4 or CD8 can bind to the same major histocompatibility complex ligand as the T-cell receptor.
CD4 with anti-CD4 monoclonal antibody (mAb) increases the in vitro kinase activity of the CD4 associated p56 ck (25), and that crosslinking of the TCR and CD4 enhances the degree of TCR-mediated tyrosine phosphorylation (26) and level of lymphokine response (27, 28). To further define the mechanism(s) by which CD4 and CD8 enhance T-cell responses to antigen, we have compared the ability of these proteins to stimulate responses when they can or cannot bind to the same MHC molecule as the TCR. BI-141 is a CD4-/CD8- T-cell hybridoma specific for beef insulin associated with a hybrid class II molecule (29). Since class I and class II molecules are present on the APC and since a baseline response to beef insulin is seen in the absence of either CD4 or CD8, this system allows for the direct comparison of the consequences of CD4 binding to the same MHC molecule as the TCR and the binding of CD8a to class I MHC distinct from the TCR ligand.
MATERIALS AND METHODS
CD4 and CD8 are cell surface glycoproteins expressed on subsets of thymocytes and mature T cells. Transfection of genes encoding CD4 or CD8 into T-cell hybridomas has demonstrated that these proteins augment interleukin 2 (IL-2) secretion in response to specific antigen (1-6) by binding to their respective class 11 (7) and class 1 (8, 9) major histocompatibility complex (MHC) ligands. Most thymocytes express CD4 and CD8, though as they mature, they selectively retain expression of either CD4 or CD8. CD4 is generally expressed on mature T cells that recognize class II-restricted antigen, whereas CD8 is usually expressed on mature cells that recognize class I-restricted antigen. This correlation between T-cell specificity and CD4 or CD8 expression has led to the suggestion that CD4 and CD8 are involved in T-cell recognition and thymic selection. The introduction of CD4 or CD8 into T-cell hybridomas in which they cannot bind to the same MHC ligand as the T-cell receptor (TCR) has allowed for independent assessment of CD4- or CD8-ligand and TCR-ligand interactions (3, 4). These experiments demonstrated that CD4 and CD8 can increase T-cell responses by binding independently of the TCR, presumably by acting as adhesion molecules to strengthen the overall avidity of the T cell for its antigenpresenting cell (APC). However, a growing body of evidence suggests that there is a physical association between the TCR and CD4 (10-12) and that CD4 and CD8 may primarily function complexed with the TCR (13-18). They have therefore been referred to as coreceptors (19). Additionally, it has been suggested that CD4 and CD8 may play an active role in T-cell signaling. Support for this idea comes from the findings that the T-cell-specific tyrosine kinase p561ck is bound to the cytoplasmic tail of CD4 and CD8 (20-24), that stimulation of
Transfection and Expression of CD4 or CD8a. cDNAs encoding CD4, CD8a, or CD8a' were subcloned into the expression vector pH3APr-2-neo (30). The L3T4cyt- construct encoding the truncated CD4 molecule was genetically engineered by cutting the cDNA encoding CD4 at the naturally occurring Nae I site, inserting stop codons, and subcloning into the pHBAPr-2-neo (30). Thirty micrograms of DNA was transfected into 1 x 107 BI-141 T-hybridoma cells by electroporation with a single discharge of 3.5 kV/cm. Transfectants were selected by growth in RPMI 1640 medium (GIBCO) supplemented with 10o heat-inactivated fetal calf serum (Gemini Biological Products, Calabasas, CA), 2 mM L-glutamine (Irvine Scientific), and 600 ug of G418 per ml (GIBCO). Functional Analysis of BI-141 and IL-2 Assay. Irradiated (4500 rads; 1 rad = 0.01 Gy) AabAI'k-expressing L cells were plated in 96-well plates at a concentration of 5 x 104 cells per well and pulsed with various concentrations of either beef or pork insulin (Sigma). The following day 1 x 105 responders per well were added and 24-hr supernatants from triplicate cultures were collected and assayed for growth by the IL-2-dependent cell line HT-2 (31) (1 x 104 cells per well). HT-2 cells were pulsed with 1 ,Ci of [3H]thymidine per well (1 Ci = 37 GBq; Amersham) 16-18 hr after culture with supernatants. Cultures were harvested on glass filters after 24 hr and assayed in a scintillation counter. Data were reported as cpm of [3H]thymidine incorporated. Antibodies used for blocking studies were partially purified from tissue culture supernatant by passing over protein A- or protein G-conjugated Sepharose beads (Pharmacia) as per the manufacturer's recommendations. mAb GK1.5 (anti-CD4)
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Abbreviations: MHC, major histocompatibility complex; IL-2, interleukin 2; TCR, T-cell receptor; mAb, monoclonal antibody; APC, antigen-presenting cell. *To whom reprint requests should be addressed. 2623
Immunology: Miceli et al.
2624
Proc. Natl. Acad. Sci. USA 88 (1991)
(32) and mAb 2.43 (anti-CD8, Lyt-2.2 allele) (33) were added to T-cell cultures at a concentration of 4 ,ug/ml and were present throughout the assay.
RESULTS Relative Effects of Expression of CD4 or CD8a in a Class 11-Restricted Hybridoma. BI-141 is a CD4-, CD8- T helper cell hybridoma specific for beef insulin associated with the hybrid class II IA molecule, AabAf8k (29). We have previously shown that expression of mouse CD4 introduced into BI-141 by gene transfer results in a marked increase in reactivity to beef insulin along with a new specificity for pork insulin (6, 34). To compare the relative effects upon IL-2 production of CD4 binding to the same (class II) MHC protein as the TCR and CD8 binding to a distinct (class I) MHC protein, we transfected an expression vector containing the cDNA for either CD4 or CD8a and the neomycin resistance gene into BI-141. Transfectants were selected for growth in the antibiotic G418 and subjected to immunofluorescence staining and analysis on the fluorescence-activated cell sorter (FACS) for cell surface expression of CD4, CD8, and the TCR Vf38 chain used for the insulin-specific response. Maximal levels of IL-2 production in response to antigen occurred when physiological or greater levels of CD4 or CD8 were present on the cell surface. Thus, to ensure that we were comparing the relative levels of maximal augmentation due to CD4 and CD8a, transfectants were sorted on a FACS for expression of high levels of CD4 or CD8 and similar levels of V,B8 (data not shown). Whereas CD8a can potentiate the response to beef insulin above that of the parental BI-141 hybridoma, it does so much less efficiently than CD4 (Fig. 1 Upper). All of the transfectants demonstrate similar plateau levels of IL-2 production. The difference between CD4 and
0
1010 A
0
x
E 0
*
B141 mCD&xLC5
0
mCD8cLD5
A
/
mCD4.C5 0 mCD4.5b v mCD4+mCD8acC6
0
100 Beef insulin, ug/ml
101 0
x
E 0. 0
1
i100
10
Pork insulin,
Ag/ml
FIG. 1. CD8a augments IL-2 release by BI-141, though not as well as CD4: Relative effects of CD4 or CD8 upon stimulation with either beef insulin or pork insulin presented by irradiated (4500 rads) AabAfBk-expressing L cells (5 x 104 cells per well). IL-2 production was assessed as described in the text and is reported as cpm of [3H]thymidine incorporation of HT-2 cells.
CD8a transfectants is demonstrated even more dramatically in the response to pork insulin (Fig. 1 Lower). Although significant IL-2 production by CD4 transfectants is observed in response to pork insulin, no IL-2 response is detected by either the parental BI-141 hybridoma or CD8a transfectants at the concentrations of pork insulin tested in Fig. 1. The effects of either CD4 or CD8a are specific since they can be blocked by mAbs directed against CD4 or CD8, respectively (data not shown and Fig. 2). Additionally, the effects of CD8a can be blocked by preincubating the APCs with anti-class I mAbs (data not shown). Coexpression of CD4 and CD8a in the Same Class 11Restricted Hybridoma. BI-141 cells cotransfected with CD4 and CD8a were sorted on a FACS until they expressed comparable levels of CD4 and CD8a when normalized to levels physiologically seen on mouse lymph node cells (data not shown). These CD4',CD8' transfectants were as efficient at augmenting IL-2 production as those transfectants expressing CD4 alone (Fig. 2). Blocking experiments on cells expressing CD4 and CD8 further demonstrate a predominant role for CD4 in this class II-restricted response (Fig. 2). Whereas mAb GK1.5, directed against CD4, significantly blocks the response at all antigen concentrations, mAb 2.43, directed against CD8, either inhibits minimally only at very low antigen concentrations or fails to inhibit at all. In each case the concentrations of mAb GK1.5 or 2.43 used were shown to specifically block transfectants bearing only CD4 or only CD8, respectively. Role of the CD8a Cytoplasmic Tail in Augmenting IL-2 Production in a Class lI-Restricted Hybridoma. Mouse CD8a naturally occurs in two forms as a result of alternative mRNA splicing (35). CD8a has a cytoplasmic tail of 28 amino acids by which it associates with the intracellular tyrosine kinase p561ck (22, 24). In contrast, CD8a' has only 3 amino acids in the cytoplasm and does not associate with p56lck (22). We have previously been involved in the characterization of a T-cell hybridoma, DC27.10, in which IL-2 production in response to the class I alloantigen Kb is entirely dependent on the expression of CD8a (2, 22). CD4, which cannot bind to the same MHC protein as the TCR, had no effect upon antigen responsiveness in this hybridoma (22). Presumably, the TCR expressed by DC27.10 is of sufficiently low affinity that it is entirely dependent on "coreceptor functions" of CD8, and any effects due to independent binding of CD4 to class II molecules present on the APC are insufficient to elicit a measurable response. In the DC27.10 hybridoma we determined that CD8a augments IL-2 production to a greater degree than does CD8a' (22). This demonstrated a direct functional role for the cytoplasmic tail of CD8a and/or the associated p56Ick when CD8a and the TCR could potentially bind to the same MHC molecule. To assess the role of the CD8a cytoplasmic tail and the associated p561ck when the TCR and CD8a bind to distinct MHC ligands, we transfected BI-141 with an expression vector containing the cDNA encoding CD8a or CD8a'. Appropriate expression of either CD8a or CD8a' was confirmed in the transfectants by immunoprecipitation with anti-CD8 mAbs, and comparable cell surface levels were demonstrated by FACS analysis (data not shown). As shown in Fig. 3, hybridomas transfected with either CD8a or CD8a' are equally efficient at augmenting IL-2 production in response to beef insulin. The potentiating effects seen with CD8a and CD8a' are specific, since they are blocked by mAb 2.43 directed against CD8 (Fig. 3). These data further indicate that the blocking seen with anti-CD8a mAb does not involve transmission of a negative signal through the cytoplasmic tail but rather results from steric interference with a receptor-ligand (CD8/class I) interaction. Role of the CD4 Cytoplasmic Tail in Augmenting IL-2 Production in a Class 11-Restricted Hybridoma. To assess the role of the cytoplasmic tail of CD4 in coreceptor function
Immunology: Miceli et al. Experiment 1
Proc. Natl. Acad. Sci. USA 88 (1991) o B-1 41
O mCD4.5b 0 mCD4+mCD8awD55 solid symbols: mAb GK1.5 open symbols: no mAb
2625
A&mCD8a B3 open symbols: no mAb
solid symbols: mAb 2.43
(antl-CD4)
100
0
x
(antl-CD8)
100 .1-0
--O-
4
0
10
.*
0
E
1
0
0
A
1
0
0
0
ExperimeXnt 2
100
10 Beef insulin, /Ag/ml
100
10
ABIM141
omCD4.C5
_orA *mCD8mD55 0MCw -
.
V mCD4+mCD8u.C6 OmCD8ftB3 open symbols: no mAb open symbols: no mAb solid symbols: mAb GKI .5 (antl-CD4)
10
0 .10
0
x
E 0
(antl-CD8)
50
50
co0
solid symbols: mAb 2.43
.,
1
3
100
10
Beef insulin, jug/ml FIG. 2. Anti-CD4 blocks the response of CD4+CD8a double transfectants to agreaterdegree than anti-CD8. Responder cells were stimulated to produce IL-2 in response to antigen in the presence of mAb (4 ,ug/ml) directed against either CD4 (GK1.5, open symbols) or CD8a (2.43, closed symbols). IL-2 production is reported
as cpm
of [3H]thymidine incorporation of HT-2 cells.
(i.e., under circumstances where CD4 and the TCR can bind to the same MHC molecule), the cDNA encoding CD4 was truncated by cutting at the naturally occurring Nae I site, and stop codons were inserted. The resulting construct encodes
the extracellular and transmembrane domains of CD4, but only the first three amino acids of the cytoplasmic tail. When transfected into BI-141 cells, the truncated molecule was expressed on the cell surface as demonstrated by immunofluorescence staining (data not shown). Northern blot analysis confirmed that CD4cytF transfectants expressed an mRNA species '100 base pairs shorter than that observed with transfectants expressing the full-length CD4 protein (data not shown). CD4cyt- transfectants were sorted on the FACS to obtain cells expressing CD4 levels comparable to the full-length CD4 transfectants and were then tested for their ability to produce IL-2 in response to beef insulin. As shown in Fig. 4, CD4cyt- transfectants could not be distinguished from CD8a transfectants in their ability to augment IL-2 secretion by the BI-141 hybridoma. Although CD4cyttransfectants produce much lower levels of IL-2 than the full-length CD4 transfectants at any given antigen concentration (Fig. 4), they do release more IL-2 than the untransfected parental line. The effects of CD4cyt- expression can be blocked with anti-CD4 mAbs, demonstrating the specificity of the augmentation (data not shown).
DISCUSSION To better evaluate the mechanisms by which CD4 and CD8 enhance T-cell activation, we have directly compared the consequences of CD4 or CD8 binding to the same or different
MHC molecules as the TCR in a single syngeneic, class II-restricted antigen-driven system. We have measured the relative potentiating effects on antigen-induced IL-2 production of expressing full-length or truncated forms of CD4 and CD8 in the class II-restricted BI-141 T-cell hybridoma. We have shown that CD8a can augment IL-2 production by BI-141. The effect of CD8 was seen in multiple transfectants assayed several times each and is specific, since it can be blocked by antibodies directed against CD8 or class I MHC. These results are in agreement with those of others demonstrating that the expression of CD4 or CD8 in T cells under circumstances where they could not bind to the same MHC ligand as the TCR results in the augmentation of antigen-induced responses (3, 4, 36). Since in BI-141, CD8a and CD8a' augment reactivity to the same extent, it is unlikely that this augmentation is the direct result of signal transduction through the CD8a cytoplasmic tail or the associated p561ck. Instead, increased reactivity most likely results from the increased avidity of the T cell for the APC due to CD8-class I interactions. This mechanism may be relevant in low-affinity T cells where an additional interaction between CD4 or CD8 and MHC alleles not directly involved in TCR antigen recognition may be significant. Indeed, support for such an effect was described by Goldstein and Mescher (36) in a system analyzing alloreactive T-cell responses to cellsized artificial membranes bearing H-2 class I. Recent data by O'Rourke et al. (37) demonstrate that CD8 binding to class I MHC is increased upon activation of the TCR with an anti-TCR antibody. If such a mechanism is involved in CD8 potentiation of antigen-induced IL-2 production in the BI-141 hybridoma, the finding that CD8a and
2626
Proc. Natl. Acad. Sci. USA 88 (1991)
Immunology: Miceli et al. Experiment 1
BAA-141
Omr CD8aB3
* mCD4+mCD&a.D55
Vm(CD8a.A6
OmCD8a.B5 A mCD8a.C3
no mAb
0
co0
10+
x
A: A
E
A
0
0Il
1
20
Experiment 2
_0 * *A
1000
100
BI-141
A v mCD4.5b
0 mCD8ot.C3
o mCD8a.B3 V mCD8a A6
0 mCD8ce.D6
no mAb
100
mCD8u .C5 0 mCD&u.A1 * mCD8d.C6 blocking with mAb 2.43 (antl-CD8) A
v
100o
0
co0 x
10.1
10
E
c00 A~~~~~~
1
20
100
A0 1
20 1000 Beef insulin, ,ug/ml
V 100
1 000
FIG. 3. CD8a' augments IL-2 release by BI-141 to the same extent as CD8a. The contribution of the CD8a cytoplasmic tail was assessed by comparing the abilities of CD8a and CD8a' transfectants to produce IL-2 in response to antigen. IL-2 production is reported as cpm of [3H]thymidine incorporation of HT-2 cells. Blocking experiments were performed as described in the legend to Fig. 2.
CD8a' augment IL-2 production to the same degree may imply that this mechanism is not dependent on the cytoplasmic tail of CD8a. Alternatively, this mechanism may not be operative in stimulating IL-2 production by the BI-141 hybridoma or may play less of a role when the TCR is stimulated by binding to antigen as opposed to specific antibody. Although CD4 and CD8a can augment antigen-induced IL-2 production in BI-141, CD4 augments the response to a much greater degree than CD8a. We attribute the greater ability of CD4 to potentiate this class II-restricted response to its ability to bind to the same MHC molecule as the TCR. The expression of CD4 and CD8 in BI-141 provides additional evidence of a predominant role for CD4 in this class IIrestricted response, since CD4+CD8 double transfectants are blocked more easily and to a greater degree with anti-CD4 than with anti-CD8 antibody. These data are in agreement with blocking studies on anomolous hybridomas that express CD4 and CD8 (17, 18). Our data demonstrate a major functional role for the cytoplasmic tail of CD4 in class II-restricted coreceptor function in a syngeneic system using intact APCs. Sleckman et al. (38) have previously demonstrated a potentiating effect due to the cytoplasmic tail of human CD4 under conditions of limiting antigen concentration in planar membranes using a xenogeneic cytoplasmic-tailless human CD4 construct in a mouse anti-human T-cell hybridoma. Previous data from the class I-specific DC27.10 hybridoma (22) and current evidence from the BI-141 hybridoma suggest that there is an element of coreceptor function that is dependent on the cytoplasmic tail of CD4 and CD8 and/or the associated p561ck. This may be the result of increased signaling due to activated p56 ck bound to the cytoplasmic tail of
CD4 or CD8a. Alternatively, or additionally, the cytoplasmic tail may be directly involved in TCR-CD8 or TCR-CD4 complex formation. We and others (22, 39) have recently published data using the class I allo-specific DC27.10 hybridoma suggesting that there may be some element of coreceptor function that is independent of the cytoplasmic tail (i.e., removal of the cytoplasmic tail of CD8 diminishes but does not completely abrogate CD8-dependent coreceptor activity). We saw no such effect with CD4 coreceptor function in the BI-141 system, since CD8a and CD4cyt- appear to potentiate IL-2 production to the same degree. These results may reflect a fundamental difference between CD4 and CD8 in their association with the TCR. On the other hand, the BI-141 system may not be sensitive enough in this response range to detect a difference between the effects of CD4cyt- and CD8a. Tyrosine phosphorylation of a number of as yet uncharacterized proteins and of the r chain are known to occur concomitantly with T-cell activation (40, 41) and are among the earliest events in T-cell activation (26, 40, 41). It has recently been demonstrated that treatment of CD4+ T cells with anti-CD4 mAbs results in increased in vitro p561ck activity accompanied by increased tyrosine phosphorylation of the chain of the TCR complex (25) and that when CD4 and the TCR are crosslinked using heteroconjugate antibodies, TCR-induced tyrosine phosphorylation is augmented (26). Perhaps increased responsiveness when CD4 or CD8 binds to the same ligand as the TCR results from their bringing activated p561ck bound to their cytoplasmic tails within proximity of a TCR-associated p56lck substrate(s). Alternatively, when CD4 and CD8 bind to the same molecule as the TCR, they may bring the associated p56Ick within proximity of a
Proc. Natl. Acad. Sci. USA 88 (1991)
Immunology: Miceli et al. A
0
T
.
* mCD4Cyt-2B4 * mCD4cyl-25814 * mCD4Cyt-.2l
mCD8ue.C3
B1-141
o V mCDEa.B3
* mCD4.C5
0
100
0
10
x
EQ. C.)
1
100 Beef insulin, pg/ml B
0
B-141
V mCD8LSB3.a
mCD4.5b
0
mCD4cyt-.25B1 * mCD4cyl-.284
*
mCD6czB3.b
v
co
1000
mCD4cyt-.25B4
100
0 T-
IK/
x
E 0
/
0~~~
10j
T20
-100
Beef insulin,
1000
pg/ml
FIG. 4. A genetically engineered cytoplasmic-tailless CD4 only as CD8a or CD8a'. The contribution of the CD4 cytoplasmic tail was assessed by companng the abilities of transfectants expressing a genetically engineered cytoplasmic tailless CD4 molecule (CD4cyt-) and transfectants expressing full-length CD4 to release IL-2 in response to antigen. A and B represent two independent experiments. augments IL-2 release by BI-141 as well
TCR-associated tyrosine phosphaitase responsible for actiVating p56lck (possibly CD45). The dependence of CD4 coreceptor function on the cytoplasmic tail may partially result from a role of the CD4 cytoplasmic tail in forming a complex with the TCR to create a receptor either more capable of binding antigen and MHC or more capable of signaling. The BI-141 transfectants described here should provide a model system for further dissection of the role of the cytoplasmic tail in coreceptor function. We are grateful to Dr. R. Germain for AabApk transfected L cells and Dr. A. Reske-Kunz for the BI-141 T-cell hybridoma. This work
supported by National Institutes of Health Grants GM34991, CA46507, and AI19512 (to J.R.P.). M.C.M. was supported by National Institutes of Health Training Grant A107290 and a postdoctoral fellowship from the Arthritis Foundation. P.v.H. was supported by a postdoctoral fellowship from the Deutsche Forschungsgemeinschaft. J.R.P. is an Established Investigator of the American Heart Association. was
1. Dembic, Z., Hass, W., Zamoyska, R., Parnes, J. R., Steinmetz, M. & von Boehmer, H. (1987) Nature (London) 326, 510-511.
2627
2. Gabert, J., Langlet, C., Zamoyska, R., Parnes, J. R., Schmitt-Verhulst, A. M. & Malissen, B. (1987) Cell 50, 545-554. 3. Gay, D., Maddon, P., Sekaly, R., Talle, M. A., Godfrey, M., Long, E., Goldstein, G., Chess, L., Axel, R., Kappler, J. & Marrack, P. (1987) Nature (London) 328, 626-629. 4. Ratnofsky, S. E., Peterson, A., Greenstein, J. L. & Burakoff, S. J. (1987) J. Exp. Med. 166, 1747-1757. 5. Sleckman, B. P., Peterson, A., Jones, W. K., Foran, J. A., Greenstein, J., Seed, B. & Burakoff, S. J. (1987) Nature (London) 328, 351-353. 6. Ballhausen, W. G., Reske-Kunz, A. B., Tourvieille, B., Ohashi, P. 5., Parnes, J. R. & Mak, T. M. (1988) J. Exp. Med. 167, 1493-1498. 7. Doyle, C. & Strominger, J. L. (1987) Nature (London) 330, 256-257. 8. Norment, A., Salter, R. D., Parham, P., Englehard, V. H. & Littman, D. (1988) Nature (London) 336, 79-81. 9. Rosenstein, Y., Ratnofsky, S., Burakoff, S. J. & Hermann, S. H. (1989) J. Exp. Med. 169, 149-160. 10. Gallagher, P. F., Fazekas de St. Groth, B. & Miller, J. F. A. P. (1989) Proc. Nati. Acad. Sci. USA 86, 10044-10048. 11. Anderson, P., Blue, M. L. & Schlossman, S. F. (1988) J. Immunol. 140, 1732-1737. 12. Rojo, J. M., Saizawa, K. & Janeway, C. A., Jr. (1989) Proc. Nati. Acad. Sci. USA 86, 3311-3315. 13. Potter, T. A., Rajan, T. V., Dick, R. F. & Bluestone, J. A. (1989) Nature (London) 337, 73-75. 14. Connolly, J. M., Hansen, T. H., Ingold, A. L. & Potter, T. A. (1990) Proc. Nati. Acad. Sci. USA 87, 2137-2141. 15. Salter, R. D., Norment, A. M., Chen, A. P., Clayberger, C., Krensky, A. M., Littman, D. & Parham, P. (1989) Nature (London) 338, 345-347. 16. Salter, R. D., Benjamin, R. J., Wesley, P. K., Buxton, S. E., Garrett, T. P. J., Clayberger, C., Krensky, A. M., Norment, A. M., Littman, D. R. & Parham, P. (1990) Nature (London) 345, 41-46. 17. Fazekas de St. Groth, B., Gallagher, P. F. & Miller, J. F. A. P. (1986) Proc. Natl. Acad. Sci. USA 83, 2594-2598. 18. Jones, B., Khavari, P. A., Conrad, P. J. & Janeway, C. A., Jr. (1987) J. Immunol. 139, 380-384. 19. Janeway, C. A., Jr., Rojo, J., Saizawa, K., Dianzani, U., Portoles, P., Tite, J., Haque, S. & Jones, B. (1989) Immunol. Rev. 109, 77-92. 20. Rudd, C. E., Trevillyan, J. M., Dasgupta, J. D., Wong, L. L. & Schlossman, S. L. (1988) Proc. Nati. Acad. Sci. USA 85, 5190-5194. 21. Veillette, A., Bookman, M. B., Horak, E. M. & Bolen, J. B. (1988) Cell 55, 301-308. 22. Zamoyska, R., Derham, P., Gorman, S. D., von Hoegen, P., Bolen, J. B., Veillette, A. & Parnes, J. R. (1989) Nature (London) 342, 278-281. 23. Shaw, A. S., Amrein, K. E., Hammond, C., Stern, D. F., Sefton, B. F. & Rose, J. F. (1989) Cell 59, 627-636. 24. Turner, J. M., Brodsky, M. H., Irving, B. A., Levin, S. D., Perlmutter, R. M. & Littman, D. R. (1990) Cell 60, 755-765. 25. Veillette, A., Bookman, M. A., Horak, E. M., Samelson, L. E. & Bolen, J. B. (1989) Nature (London) 338, 257-259. 26. June, C. H., Fletcher, M. C., Ledbetter, J. A. & Samelson, L. E. (1990) J. Immunol. 144, 1591-1599. 27. Anderson, P., Blue, M. L., Morimoto, C. & Schlossman, S. F. (1987) J. Immunol. 139, 678-682. 28. Ledbetter, J. A., June, C. H., Rabinovich, P. S., Grossman, A., Tsu, T. T. & Imboden, J. B. (1988) Eur. J. Immunol. 17, 529-534. 29. Reske-Kunz, A. B. & Rude, E. (1985) Eur. J. Immunol. 15, 1048-1054. 30. Gunning, P., Leavitt, J., Muscat, G., Ng, Y. S. & Kedes, L. (1987) Proc. Natl. Acad. Sci. USA 83, 4831-4837. 31. Mossman, T. R. (1983) J. Immunol. Methods 65, 55-58. 32. Dialynas, D. P., Quan, Z. S., Wall, K. A., Pierres, A., Quintans, J., Loken, M. R., Pierres, M. & Fitch, F. W. (1984) J. Immunol. 131, 2445-2451. 33. Sarmiento, M., Glasebrook, A. L. & Fitch, F. W. (1980) J. Immunol. 125, 2665-2672. 34. Parnes, J. R., von Hoegen, P. & Miceli, M. C. (1990) Cold Spring Harbor Symp. Quant. Biol. 54, 649-655. 35. Liaw, C. W., Zamoyska, R. & Parnes, J. R. (1986) J. Immunol. 137, 1037-1043. 36. Goldstein, S. A. N. & Mescher, M. F. (1988) J. Immunol. 140, 37073711. 37. O'Rourke, A. M., Rogers, J. & Mescher, M. (1990) Nature (London) 346, 187-189. 38. Sleckman, B. P., Peterson, A., Foran, A., Gorga, J. C., Kara, C. J., Strominger, J. L., Burakoff, S. J. & Greenstein, J. L. (1988) J. Immunol. 141, 49-54. 39. Letourneur, F., Gabert, J., Cosson, P., Blanc, D., Davoust, J. & Malissen, B. (1990) Proc. Natl. Acad. Sci. USA 87, 2339-2343. 40. Samelson, L. E., Patel, M. D., Weissman, A. M., Harford, J. H. & Klausner, R. D. (1986) Cell 46, 1083-1090. 41. Hsi, E. D., Siegel, J. M., Minami, Y., Luong, E. T., Klausner, R. D. & Samelson, L. E. (1989) J. Biol. Chem. 264, 10836-10842.
Adhesion versus coreceptor function of CD4 and CD8: Role of the cytoplasmic tail in coreceptor activity (T-cell
activation/p56kk/T-ceil receptor/accessory molecules)
M. CARRIE MICELI, PAUL VON HOEGEN, AND JANE R. PARNES* Division of Immunology and Rheumatology, Department of Medicine, Stanford University Medical Center, Stanford, CA 94305-5111
Communicated by Leonard A. Herzenberg, December 28, 1990
ABSTRACT CD4 and CD8 play an important role in T-cell recognition and activation; however, their mechanisms of action are not well understood. We compare the effects of expressing CD4 and CD8a either individually or together in a class U-restricted T-cell hybridoma. We also compare the effects of expressing truncated forms of CD4 or CD8a that do not have a cytoplasmic tail and thus do not associate with the T-cell-specifIc tyrosine kinase p56Ick, which has been implicated in T-cell activation. We demonstrate that, although CD4 and CD8a can specifically enhance interleukin 2 secretion, maximal potentiation occurs with expression of CD4, which, unlike CD8, can bind to the same major histocompatibility complex protein as the T-cell receptor. Our data further indicate that the cytoplasmic tail and/or the associated p56&k are primarily significant for interleukin 2 secretion by the hybridomas we have examined when CD4 or CD8 can bind to the same major histocompatibility complex ligand as the T-cell receptor.
CD4 with anti-CD4 monoclonal antibody (mAb) increases the in vitro kinase activity of the CD4 associated p56 ck (25), and that crosslinking of the TCR and CD4 enhances the degree of TCR-mediated tyrosine phosphorylation (26) and level of lymphokine response (27, 28). To further define the mechanism(s) by which CD4 and CD8 enhance T-cell responses to antigen, we have compared the ability of these proteins to stimulate responses when they can or cannot bind to the same MHC molecule as the TCR. BI-141 is a CD4-/CD8- T-cell hybridoma specific for beef insulin associated with a hybrid class II molecule (29). Since class I and class II molecules are present on the APC and since a baseline response to beef insulin is seen in the absence of either CD4 or CD8, this system allows for the direct comparison of the consequences of CD4 binding to the same MHC molecule as the TCR and the binding of CD8a to class I MHC distinct from the TCR ligand.
MATERIALS AND METHODS
CD4 and CD8 are cell surface glycoproteins expressed on subsets of thymocytes and mature T cells. Transfection of genes encoding CD4 or CD8 into T-cell hybridomas has demonstrated that these proteins augment interleukin 2 (IL-2) secretion in response to specific antigen (1-6) by binding to their respective class 11 (7) and class 1 (8, 9) major histocompatibility complex (MHC) ligands. Most thymocytes express CD4 and CD8, though as they mature, they selectively retain expression of either CD4 or CD8. CD4 is generally expressed on mature T cells that recognize class II-restricted antigen, whereas CD8 is usually expressed on mature cells that recognize class I-restricted antigen. This correlation between T-cell specificity and CD4 or CD8 expression has led to the suggestion that CD4 and CD8 are involved in T-cell recognition and thymic selection. The introduction of CD4 or CD8 into T-cell hybridomas in which they cannot bind to the same MHC ligand as the T-cell receptor (TCR) has allowed for independent assessment of CD4- or CD8-ligand and TCR-ligand interactions (3, 4). These experiments demonstrated that CD4 and CD8 can increase T-cell responses by binding independently of the TCR, presumably by acting as adhesion molecules to strengthen the overall avidity of the T cell for its antigenpresenting cell (APC). However, a growing body of evidence suggests that there is a physical association between the TCR and CD4 (10-12) and that CD4 and CD8 may primarily function complexed with the TCR (13-18). They have therefore been referred to as coreceptors (19). Additionally, it has been suggested that CD4 and CD8 may play an active role in T-cell signaling. Support for this idea comes from the findings that the T-cell-specific tyrosine kinase p561ck is bound to the cytoplasmic tail of CD4 and CD8 (20-24), that stimulation of
Transfection and Expression of CD4 or CD8a. cDNAs encoding CD4, CD8a, or CD8a' were subcloned into the expression vector pH3APr-2-neo (30). The L3T4cyt- construct encoding the truncated CD4 molecule was genetically engineered by cutting the cDNA encoding CD4 at the naturally occurring Nae I site, inserting stop codons, and subcloning into the pHBAPr-2-neo (30). Thirty micrograms of DNA was transfected into 1 x 107 BI-141 T-hybridoma cells by electroporation with a single discharge of 3.5 kV/cm. Transfectants were selected by growth in RPMI 1640 medium (GIBCO) supplemented with 10o heat-inactivated fetal calf serum (Gemini Biological Products, Calabasas, CA), 2 mM L-glutamine (Irvine Scientific), and 600 ug of G418 per ml (GIBCO). Functional Analysis of BI-141 and IL-2 Assay. Irradiated (4500 rads; 1 rad = 0.01 Gy) AabAI'k-expressing L cells were plated in 96-well plates at a concentration of 5 x 104 cells per well and pulsed with various concentrations of either beef or pork insulin (Sigma). The following day 1 x 105 responders per well were added and 24-hr supernatants from triplicate cultures were collected and assayed for growth by the IL-2-dependent cell line HT-2 (31) (1 x 104 cells per well). HT-2 cells were pulsed with 1 ,Ci of [3H]thymidine per well (1 Ci = 37 GBq; Amersham) 16-18 hr after culture with supernatants. Cultures were harvested on glass filters after 24 hr and assayed in a scintillation counter. Data were reported as cpm of [3H]thymidine incorporated. Antibodies used for blocking studies were partially purified from tissue culture supernatant by passing over protein A- or protein G-conjugated Sepharose beads (Pharmacia) as per the manufacturer's recommendations. mAb GK1.5 (anti-CD4)
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Abbreviations: MHC, major histocompatibility complex; IL-2, interleukin 2; TCR, T-cell receptor; mAb, monoclonal antibody; APC, antigen-presenting cell. *To whom reprint requests should be addressed. 2623
Immunology: Miceli et al.
2624
Proc. Natl. Acad. Sci. USA 88 (1991)
(32) and mAb 2.43 (anti-CD8, Lyt-2.2 allele) (33) were added to T-cell cultures at a concentration of 4 ,ug/ml and were present throughout the assay.
RESULTS Relative Effects of Expression of CD4 or CD8a in a Class 11-Restricted Hybridoma. BI-141 is a CD4-, CD8- T helper cell hybridoma specific for beef insulin associated with the hybrid class II IA molecule, AabAf8k (29). We have previously shown that expression of mouse CD4 introduced into BI-141 by gene transfer results in a marked increase in reactivity to beef insulin along with a new specificity for pork insulin (6, 34). To compare the relative effects upon IL-2 production of CD4 binding to the same (class II) MHC protein as the TCR and CD8 binding to a distinct (class I) MHC protein, we transfected an expression vector containing the cDNA for either CD4 or CD8a and the neomycin resistance gene into BI-141. Transfectants were selected for growth in the antibiotic G418 and subjected to immunofluorescence staining and analysis on the fluorescence-activated cell sorter (FACS) for cell surface expression of CD4, CD8, and the TCR Vf38 chain used for the insulin-specific response. Maximal levels of IL-2 production in response to antigen occurred when physiological or greater levels of CD4 or CD8 were present on the cell surface. Thus, to ensure that we were comparing the relative levels of maximal augmentation due to CD4 and CD8a, transfectants were sorted on a FACS for expression of high levels of CD4 or CD8 and similar levels of V,B8 (data not shown). Whereas CD8a can potentiate the response to beef insulin above that of the parental BI-141 hybridoma, it does so much less efficiently than CD4 (Fig. 1 Upper). All of the transfectants demonstrate similar plateau levels of IL-2 production. The difference between CD4 and
0
1010 A
0
x
E 0
*
B141 mCD&xLC5
0
mCD8cLD5
A
/
mCD4.C5 0 mCD4.5b v mCD4+mCD8acC6
0
100 Beef insulin, ug/ml
101 0
x
E 0. 0
1
i100
10
Pork insulin,
Ag/ml
FIG. 1. CD8a augments IL-2 release by BI-141, though not as well as CD4: Relative effects of CD4 or CD8 upon stimulation with either beef insulin or pork insulin presented by irradiated (4500 rads) AabAfBk-expressing L cells (5 x 104 cells per well). IL-2 production was assessed as described in the text and is reported as cpm of [3H]thymidine incorporation of HT-2 cells.
CD8a transfectants is demonstrated even more dramatically in the response to pork insulin (Fig. 1 Lower). Although significant IL-2 production by CD4 transfectants is observed in response to pork insulin, no IL-2 response is detected by either the parental BI-141 hybridoma or CD8a transfectants at the concentrations of pork insulin tested in Fig. 1. The effects of either CD4 or CD8a are specific since they can be blocked by mAbs directed against CD4 or CD8, respectively (data not shown and Fig. 2). Additionally, the effects of CD8a can be blocked by preincubating the APCs with anti-class I mAbs (data not shown). Coexpression of CD4 and CD8a in the Same Class 11Restricted Hybridoma. BI-141 cells cotransfected with CD4 and CD8a were sorted on a FACS until they expressed comparable levels of CD4 and CD8a when normalized to levels physiologically seen on mouse lymph node cells (data not shown). These CD4',CD8' transfectants were as efficient at augmenting IL-2 production as those transfectants expressing CD4 alone (Fig. 2). Blocking experiments on cells expressing CD4 and CD8 further demonstrate a predominant role for CD4 in this class II-restricted response (Fig. 2). Whereas mAb GK1.5, directed against CD4, significantly blocks the response at all antigen concentrations, mAb 2.43, directed against CD8, either inhibits minimally only at very low antigen concentrations or fails to inhibit at all. In each case the concentrations of mAb GK1.5 or 2.43 used were shown to specifically block transfectants bearing only CD4 or only CD8, respectively. Role of the CD8a Cytoplasmic Tail in Augmenting IL-2 Production in a Class lI-Restricted Hybridoma. Mouse CD8a naturally occurs in two forms as a result of alternative mRNA splicing (35). CD8a has a cytoplasmic tail of 28 amino acids by which it associates with the intracellular tyrosine kinase p561ck (22, 24). In contrast, CD8a' has only 3 amino acids in the cytoplasm and does not associate with p56lck (22). We have previously been involved in the characterization of a T-cell hybridoma, DC27.10, in which IL-2 production in response to the class I alloantigen Kb is entirely dependent on the expression of CD8a (2, 22). CD4, which cannot bind to the same MHC protein as the TCR, had no effect upon antigen responsiveness in this hybridoma (22). Presumably, the TCR expressed by DC27.10 is of sufficiently low affinity that it is entirely dependent on "coreceptor functions" of CD8, and any effects due to independent binding of CD4 to class II molecules present on the APC are insufficient to elicit a measurable response. In the DC27.10 hybridoma we determined that CD8a augments IL-2 production to a greater degree than does CD8a' (22). This demonstrated a direct functional role for the cytoplasmic tail of CD8a and/or the associated p56Ick when CD8a and the TCR could potentially bind to the same MHC molecule. To assess the role of the CD8a cytoplasmic tail and the associated p561ck when the TCR and CD8a bind to distinct MHC ligands, we transfected BI-141 with an expression vector containing the cDNA encoding CD8a or CD8a'. Appropriate expression of either CD8a or CD8a' was confirmed in the transfectants by immunoprecipitation with anti-CD8 mAbs, and comparable cell surface levels were demonstrated by FACS analysis (data not shown). As shown in Fig. 3, hybridomas transfected with either CD8a or CD8a' are equally efficient at augmenting IL-2 production in response to beef insulin. The potentiating effects seen with CD8a and CD8a' are specific, since they are blocked by mAb 2.43 directed against CD8 (Fig. 3). These data further indicate that the blocking seen with anti-CD8a mAb does not involve transmission of a negative signal through the cytoplasmic tail but rather results from steric interference with a receptor-ligand (CD8/class I) interaction. Role of the CD4 Cytoplasmic Tail in Augmenting IL-2 Production in a Class 11-Restricted Hybridoma. To assess the role of the cytoplasmic tail of CD4 in coreceptor function
Immunology: Miceli et al. Experiment 1
Proc. Natl. Acad. Sci. USA 88 (1991) o B-1 41
O mCD4.5b 0 mCD4+mCD8awD55 solid symbols: mAb GK1.5 open symbols: no mAb
2625
A&mCD8a B3 open symbols: no mAb
solid symbols: mAb 2.43
(antl-CD4)
100
0
x
(antl-CD8)
100 .1-0
--O-
4
0
10
.*
0
E
1
0
0
A
1
0
0
0
ExperimeXnt 2
100
10 Beef insulin, /Ag/ml
100
10
ABIM141
omCD4.C5
_orA *mCD8mD55 0MCw -
.
V mCD4+mCD8u.C6 OmCD8ftB3 open symbols: no mAb open symbols: no mAb solid symbols: mAb GKI .5 (antl-CD4)
10
0 .10
0
x
E 0
(antl-CD8)
50
50
co0
solid symbols: mAb 2.43
.,
1
3
100
10
Beef insulin, jug/ml FIG. 2. Anti-CD4 blocks the response of CD4+CD8a double transfectants to agreaterdegree than anti-CD8. Responder cells were stimulated to produce IL-2 in response to antigen in the presence of mAb (4 ,ug/ml) directed against either CD4 (GK1.5, open symbols) or CD8a (2.43, closed symbols). IL-2 production is reported
as cpm
of [3H]thymidine incorporation of HT-2 cells.
(i.e., under circumstances where CD4 and the TCR can bind to the same MHC molecule), the cDNA encoding CD4 was truncated by cutting at the naturally occurring Nae I site, and stop codons were inserted. The resulting construct encodes
the extracellular and transmembrane domains of CD4, but only the first three amino acids of the cytoplasmic tail. When transfected into BI-141 cells, the truncated molecule was expressed on the cell surface as demonstrated by immunofluorescence staining (data not shown). Northern blot analysis confirmed that CD4cytF transfectants expressed an mRNA species '100 base pairs shorter than that observed with transfectants expressing the full-length CD4 protein (data not shown). CD4cyt- transfectants were sorted on the FACS to obtain cells expressing CD4 levels comparable to the full-length CD4 transfectants and were then tested for their ability to produce IL-2 in response to beef insulin. As shown in Fig. 4, CD4cyt- transfectants could not be distinguished from CD8a transfectants in their ability to augment IL-2 secretion by the BI-141 hybridoma. Although CD4cyttransfectants produce much lower levels of IL-2 than the full-length CD4 transfectants at any given antigen concentration (Fig. 4), they do release more IL-2 than the untransfected parental line. The effects of CD4cyt- expression can be blocked with anti-CD4 mAbs, demonstrating the specificity of the augmentation (data not shown).
DISCUSSION To better evaluate the mechanisms by which CD4 and CD8 enhance T-cell activation, we have directly compared the consequences of CD4 or CD8 binding to the same or different
MHC molecules as the TCR in a single syngeneic, class II-restricted antigen-driven system. We have measured the relative potentiating effects on antigen-induced IL-2 production of expressing full-length or truncated forms of CD4 and CD8 in the class II-restricted BI-141 T-cell hybridoma. We have shown that CD8a can augment IL-2 production by BI-141. The effect of CD8 was seen in multiple transfectants assayed several times each and is specific, since it can be blocked by antibodies directed against CD8 or class I MHC. These results are in agreement with those of others demonstrating that the expression of CD4 or CD8 in T cells under circumstances where they could not bind to the same MHC ligand as the TCR results in the augmentation of antigen-induced responses (3, 4, 36). Since in BI-141, CD8a and CD8a' augment reactivity to the same extent, it is unlikely that this augmentation is the direct result of signal transduction through the CD8a cytoplasmic tail or the associated p561ck. Instead, increased reactivity most likely results from the increased avidity of the T cell for the APC due to CD8-class I interactions. This mechanism may be relevant in low-affinity T cells where an additional interaction between CD4 or CD8 and MHC alleles not directly involved in TCR antigen recognition may be significant. Indeed, support for such an effect was described by Goldstein and Mescher (36) in a system analyzing alloreactive T-cell responses to cellsized artificial membranes bearing H-2 class I. Recent data by O'Rourke et al. (37) demonstrate that CD8 binding to class I MHC is increased upon activation of the TCR with an anti-TCR antibody. If such a mechanism is involved in CD8 potentiation of antigen-induced IL-2 production in the BI-141 hybridoma, the finding that CD8a and
2626
Proc. Natl. Acad. Sci. USA 88 (1991)
Immunology: Miceli et al. Experiment 1
BAA-141
Omr CD8aB3
* mCD4+mCD&a.D55
Vm(CD8a.A6
OmCD8a.B5 A mCD8a.C3
no mAb
0
co0
10+
x
A: A
E
A
0
0Il
1
20
Experiment 2
_0 * *A
1000
100
BI-141
A v mCD4.5b
0 mCD8ot.C3
o mCD8a.B3 V mCD8a A6
0 mCD8ce.D6
no mAb
100
mCD8u .C5 0 mCD&u.A1 * mCD8d.C6 blocking with mAb 2.43 (antl-CD8) A
v
100o
0
co0 x
10.1
10
E
c00 A~~~~~~
1
20
100
A0 1
20 1000 Beef insulin, ,ug/ml
V 100
1 000
FIG. 3. CD8a' augments IL-2 release by BI-141 to the same extent as CD8a. The contribution of the CD8a cytoplasmic tail was assessed by comparing the abilities of CD8a and CD8a' transfectants to produce IL-2 in response to antigen. IL-2 production is reported as cpm of [3H]thymidine incorporation of HT-2 cells. Blocking experiments were performed as described in the legend to Fig. 2.
CD8a' augment IL-2 production to the same degree may imply that this mechanism is not dependent on the cytoplasmic tail of CD8a. Alternatively, this mechanism may not be operative in stimulating IL-2 production by the BI-141 hybridoma or may play less of a role when the TCR is stimulated by binding to antigen as opposed to specific antibody. Although CD4 and CD8a can augment antigen-induced IL-2 production in BI-141, CD4 augments the response to a much greater degree than CD8a. We attribute the greater ability of CD4 to potentiate this class II-restricted response to its ability to bind to the same MHC molecule as the TCR. The expression of CD4 and CD8 in BI-141 provides additional evidence of a predominant role for CD4 in this class IIrestricted response, since CD4+CD8 double transfectants are blocked more easily and to a greater degree with anti-CD4 than with anti-CD8 antibody. These data are in agreement with blocking studies on anomolous hybridomas that express CD4 and CD8 (17, 18). Our data demonstrate a major functional role for the cytoplasmic tail of CD4 in class II-restricted coreceptor function in a syngeneic system using intact APCs. Sleckman et al. (38) have previously demonstrated a potentiating effect due to the cytoplasmic tail of human CD4 under conditions of limiting antigen concentration in planar membranes using a xenogeneic cytoplasmic-tailless human CD4 construct in a mouse anti-human T-cell hybridoma. Previous data from the class I-specific DC27.10 hybridoma (22) and current evidence from the BI-141 hybridoma suggest that there is an element of coreceptor function that is dependent on the cytoplasmic tail of CD4 and CD8 and/or the associated p561ck. This may be the result of increased signaling due to activated p56 ck bound to the cytoplasmic tail of
CD4 or CD8a. Alternatively, or additionally, the cytoplasmic tail may be directly involved in TCR-CD8 or TCR-CD4 complex formation. We and others (22, 39) have recently published data using the class I allo-specific DC27.10 hybridoma suggesting that there may be some element of coreceptor function that is independent of the cytoplasmic tail (i.e., removal of the cytoplasmic tail of CD8 diminishes but does not completely abrogate CD8-dependent coreceptor activity). We saw no such effect with CD4 coreceptor function in the BI-141 system, since CD8a and CD4cyt- appear to potentiate IL-2 production to the same degree. These results may reflect a fundamental difference between CD4 and CD8 in their association with the TCR. On the other hand, the BI-141 system may not be sensitive enough in this response range to detect a difference between the effects of CD4cyt- and CD8a. Tyrosine phosphorylation of a number of as yet uncharacterized proteins and of the r chain are known to occur concomitantly with T-cell activation (40, 41) and are among the earliest events in T-cell activation (26, 40, 41). It has recently been demonstrated that treatment of CD4+ T cells with anti-CD4 mAbs results in increased in vitro p561ck activity accompanied by increased tyrosine phosphorylation of the chain of the TCR complex (25) and that when CD4 and the TCR are crosslinked using heteroconjugate antibodies, TCR-induced tyrosine phosphorylation is augmented (26). Perhaps increased responsiveness when CD4 or CD8 binds to the same ligand as the TCR results from their bringing activated p561ck bound to their cytoplasmic tails within proximity of a TCR-associated p56lck substrate(s). Alternatively, when CD4 and CD8 bind to the same molecule as the TCR, they may bring the associated p56Ick within proximity of a
Proc. Natl. Acad. Sci. USA 88 (1991)
Immunology: Miceli et al. A
0
T
.
* mCD4Cyt-2B4 * mCD4cyl-25814 * mCD4Cyt-.2l
mCD8ue.C3
B1-141
o V mCDEa.B3
* mCD4.C5
0
100
0
10
x
EQ. C.)
1
100 Beef insulin, pg/ml B
0
B-141
V mCD8LSB3.a
mCD4.5b
0
mCD4cyt-.25B1 * mCD4cyl-.284
*
mCD6czB3.b
v
co
1000
mCD4cyt-.25B4
100
0 T-
IK/
x
E 0
/
0~~~
10j
T20
-100
Beef insulin,
1000
pg/ml
FIG. 4. A genetically engineered cytoplasmic-tailless CD4 only as CD8a or CD8a'. The contribution of the CD4 cytoplasmic tail was assessed by companng the abilities of transfectants expressing a genetically engineered cytoplasmic tailless CD4 molecule (CD4cyt-) and transfectants expressing full-length CD4 to release IL-2 in response to antigen. A and B represent two independent experiments. augments IL-2 release by BI-141 as well
TCR-associated tyrosine phosphaitase responsible for actiVating p56lck (possibly CD45). The dependence of CD4 coreceptor function on the cytoplasmic tail may partially result from a role of the CD4 cytoplasmic tail in forming a complex with the TCR to create a receptor either more capable of binding antigen and MHC or more capable of signaling. The BI-141 transfectants described here should provide a model system for further dissection of the role of the cytoplasmic tail in coreceptor function. We are grateful to Dr. R. Germain for AabApk transfected L cells and Dr. A. Reske-Kunz for the BI-141 T-cell hybridoma. This work
supported by National Institutes of Health Grants GM34991, CA46507, and AI19512 (to J.R.P.). M.C.M. was supported by National Institutes of Health Training Grant A107290 and a postdoctoral fellowship from the Arthritis Foundation. P.v.H. was supported by a postdoctoral fellowship from the Deutsche Forschungsgemeinschaft. J.R.P. is an Established Investigator of the American Heart Association. was
1. Dembic, Z., Hass, W., Zamoyska, R., Parnes, J. R., Steinmetz, M. & von Boehmer, H. (1987) Nature (London) 326, 510-511.
2627
2. Gabert, J., Langlet, C., Zamoyska, R., Parnes, J. R., Schmitt-Verhulst, A. M. & Malissen, B. (1987) Cell 50, 545-554. 3. Gay, D., Maddon, P., Sekaly, R., Talle, M. A., Godfrey, M., Long, E., Goldstein, G., Chess, L., Axel, R., Kappler, J. & Marrack, P. (1987) Nature (London) 328, 626-629. 4. Ratnofsky, S. E., Peterson, A., Greenstein, J. L. & Burakoff, S. J. (1987) J. Exp. Med. 166, 1747-1757. 5. Sleckman, B. P., Peterson, A., Jones, W. K., Foran, J. A., Greenstein, J., Seed, B. & Burakoff, S. J. (1987) Nature (London) 328, 351-353. 6. Ballhausen, W. G., Reske-Kunz, A. B., Tourvieille, B., Ohashi, P. 5., Parnes, J. R. & Mak, T. M. (1988) J. Exp. Med. 167, 1493-1498. 7. Doyle, C. & Strominger, J. L. (1987) Nature (London) 330, 256-257. 8. Norment, A., Salter, R. D., Parham, P., Englehard, V. H. & Littman, D. (1988) Nature (London) 336, 79-81. 9. Rosenstein, Y., Ratnofsky, S., Burakoff, S. J. & Hermann, S. H. (1989) J. Exp. Med. 169, 149-160. 10. Gallagher, P. F., Fazekas de St. Groth, B. & Miller, J. F. A. P. (1989) Proc. Nati. Acad. Sci. USA 86, 10044-10048. 11. Anderson, P., Blue, M. L. & Schlossman, S. F. (1988) J. Immunol. 140, 1732-1737. 12. Rojo, J. M., Saizawa, K. & Janeway, C. A., Jr. (1989) Proc. Nati. Acad. Sci. USA 86, 3311-3315. 13. Potter, T. A., Rajan, T. V., Dick, R. F. & Bluestone, J. A. (1989) Nature (London) 337, 73-75. 14. Connolly, J. M., Hansen, T. H., Ingold, A. L. & Potter, T. A. (1990) Proc. Nati. Acad. Sci. USA 87, 2137-2141. 15. Salter, R. D., Norment, A. M., Chen, A. P., Clayberger, C., Krensky, A. M., Littman, D. & Parham, P. (1989) Nature (London) 338, 345-347. 16. Salter, R. D., Benjamin, R. J., Wesley, P. K., Buxton, S. E., Garrett, T. P. J., Clayberger, C., Krensky, A. M., Norment, A. M., Littman, D. R. & Parham, P. (1990) Nature (London) 345, 41-46. 17. Fazekas de St. Groth, B., Gallagher, P. F. & Miller, J. F. A. P. (1986) Proc. Natl. Acad. Sci. USA 83, 2594-2598. 18. Jones, B., Khavari, P. A., Conrad, P. J. & Janeway, C. A., Jr. (1987) J. Immunol. 139, 380-384. 19. Janeway, C. A., Jr., Rojo, J., Saizawa, K., Dianzani, U., Portoles, P., Tite, J., Haque, S. & Jones, B. (1989) Immunol. Rev. 109, 77-92. 20. Rudd, C. E., Trevillyan, J. M., Dasgupta, J. D., Wong, L. L. & Schlossman, S. L. (1988) Proc. Nati. Acad. Sci. USA 85, 5190-5194. 21. Veillette, A., Bookman, M. B., Horak, E. M. & Bolen, J. B. (1988) Cell 55, 301-308. 22. Zamoyska, R., Derham, P., Gorman, S. D., von Hoegen, P., Bolen, J. B., Veillette, A. & Parnes, J. R. (1989) Nature (London) 342, 278-281. 23. Shaw, A. S., Amrein, K. E., Hammond, C., Stern, D. F., Sefton, B. F. & Rose, J. F. (1989) Cell 59, 627-636. 24. Turner, J. M., Brodsky, M. H., Irving, B. A., Levin, S. D., Perlmutter, R. M. & Littman, D. R. (1990) Cell 60, 755-765. 25. Veillette, A., Bookman, M. A., Horak, E. M., Samelson, L. E. & Bolen, J. B. (1989) Nature (London) 338, 257-259. 26. June, C. H., Fletcher, M. C., Ledbetter, J. A. & Samelson, L. E. (1990) J. Immunol. 144, 1591-1599. 27. Anderson, P., Blue, M. L., Morimoto, C. & Schlossman, S. F. (1987) J. Immunol. 139, 678-682. 28. Ledbetter, J. A., June, C. H., Rabinovich, P. S., Grossman, A., Tsu, T. T. & Imboden, J. B. (1988) Eur. J. Immunol. 17, 529-534. 29. Reske-Kunz, A. B. & Rude, E. (1985) Eur. J. Immunol. 15, 1048-1054. 30. Gunning, P., Leavitt, J., Muscat, G., Ng, Y. S. & Kedes, L. (1987) Proc. Natl. Acad. Sci. USA 83, 4831-4837. 31. Mossman, T. R. (1983) J. Immunol. Methods 65, 55-58. 32. Dialynas, D. P., Quan, Z. S., Wall, K. A., Pierres, A., Quintans, J., Loken, M. R., Pierres, M. & Fitch, F. W. (1984) J. Immunol. 131, 2445-2451. 33. Sarmiento, M., Glasebrook, A. L. & Fitch, F. W. (1980) J. Immunol. 125, 2665-2672. 34. Parnes, J. R., von Hoegen, P. & Miceli, M. C. (1990) Cold Spring Harbor Symp. Quant. Biol. 54, 649-655. 35. Liaw, C. W., Zamoyska, R. & Parnes, J. R. (1986) J. Immunol. 137, 1037-1043. 36. Goldstein, S. A. N. & Mescher, M. F. (1988) J. Immunol. 140, 37073711. 37. O'Rourke, A. M., Rogers, J. & Mescher, M. (1990) Nature (London) 346, 187-189. 38. Sleckman, B. P., Peterson, A., Foran, A., Gorga, J. C., Kara, C. J., Strominger, J. L., Burakoff, S. J. & Greenstein, J. L. (1988) J. Immunol. 141, 49-54. 39. Letourneur, F., Gabert, J., Cosson, P., Blanc, D., Davoust, J. & Malissen, B. (1990) Proc. Natl. Acad. Sci. USA 87, 2339-2343. 40. Samelson, L. E., Patel, M. D., Weissman, A. M., Harford, J. H. & Klausner, R. D. (1986) Cell 46, 1083-1090. 41. Hsi, E. D., Siegel, J. M., Minami, Y., Luong, E. T., Klausner, R. D. & Samelson, L. E. (1989) J. Biol. Chem. 264, 10836-10842.
