Activation of Tyrosine Kinase p6OfYn following T Cell Antigen Receptor Cross-linking*

Activation of T cells by specific antigens in the context of major histocompatibility complex encoded proteins is mediated by the T cell antigen receptor (TcR), consisting of a variable (Ti) and an invariant (CD3) subunits. Tyrosine phosphorylation is considered to be one of the earliest steps in TcR-mediated signal transduction. There are indications that the p60fyn protein tyrosine kinase is involved in signaling via TcR. However, enzymatic activation of the Src-related tyrosine kinases upon TcR triggering has not been shown yet, therefore the identity of TcR-activated tyrosine kinase(s) remains unclear. We demonstrate that cross-linking of CD3 activates p60fyn and induces tyrosine phosphorylation of cellular proteins in human T cells (resting peripheral T cells, a helper T cell clone, a helper T cell clone immortalized with Herpesvirus saimiri, and a leukemic T cell line). Activation of p60fyn was fast, and its maximum (2-4-fold activation as compared with the basal activity) was followed by a decline. The amount of p60fyn in the cells remained unchanged. None of the other T cell Src-related tyrosine kinases was activated after cross-linking of CD3. Activation of p60fyn was induced by anti-CD3, but not by anti-CD4, anti-CD2, or anti-CD28. The activation was correlated with an increase of the phosphotyrosine content of p60fyn. These studies provide direct proof for the functional association between p60fyn and the TcR.

rosine phosphorylation is thought to be one of the earliest steps in TcR-mediated signal transduction (5-7). However, the identity of a tyrosine kinase activated by TcR triggering remains unclear. There are several indications that p6OfYn may be involved in signal transduction by the TcR. First, p6OfYn can be co-precipitated with CD3, although with an apparently low stoichiometry (8,9). Also, co-distribution of p6OfYn with the TcR has been shown following cross-linking of the TcR (10). Overexpression of p6OrYn in thymocytes renders these cells hyperresponsive to TcR stimuli, whereas overexpression of an enzymatically defective mutant of p6OrYn suppresses TcR-mediated signal transduction (11). Finally, the phosphate content of p6Oryn increases in permeabilized thymocytes after cross-linking of CD3 (12). However, in none of these studies has the activity of p60fyn nor any other tyrosine kinases been demonstrated to be activated following TcR engagement.
Here we report that cross-linking of CD3 increases ~60""protein kinase activity in human T cells.

MATERIALS AND METHODS
Cells-Peripheral T cells were separated from donor blood according to Ref. 13, except glass bead columns with anti-human IgG were used to remove B cells from non-adherent peripheral blood mononuclear cells (PBMC). Resulting preparations were >90% CD3+.
EBR-65/57 is a human CD4'CD8-, interleukin-2-dependent T cell clone obtained from a normal donor by limiting dilution. The clone was restimulated with phytohemagglutinin in the presence of yirradiated PBMC, and used for experiments 7-10 days later. The human CD4+CD8-T cell clone TAR-61/48 HVS, immortalized with Herpesvirus saimiri (14), acquired the ability to grow in culture without restimulation retaining interleukin-2 dependence and the ability to respond to TcR-mediated stimuli (data not shown). CEM-6 is a CD3+ subline of the leukemic T cell line CEM. Cells were grown in RPMI 1640 supplemented with 10% fetal bovine serum (FBS) and 4 mM L-glutamine, 20 mM HEPES, and penicillin/streptomycin (complete medium). In the case of clones, 1% human AB serum was added to complete medium.

RESULTS AND DISCUSSION
The results shown in Fig. 1-4 demonstrate that crosslinking of CD3 on the surface of a variety of different T cells leads to an increase in ~60'"" tyrosine protein kinase activity in immune complex assay, as well as to the stimulation of total tyrosine phosphorylation in whole cell lysates. Activation of p60f."" protein kinase activity was observed as early as 30 s following TcR cross-linking, and its maximum a t 1-3 min was followed by a decline in detectable activity by 10 min. The activation of ~60'"" is likely the result of a change in specific activity, since the amount of ~60'"" in the cells and immune complexes remained unchanged.
All four types of human T cells used in these experiments (peripheral blood T cells (Fig. l ) , a T cell clone (Fig. 2), a T cell clone immortalized by Herpesvirus saimiri (Fig. 3), and a leukemic T cell line (Fig. 4)) demonstrated qualitatively similar patterns of cellular tyrosine phosphorylation and p60'v" activation kinetics, although their extents were not the same for different cells. The average activation indexes (maximum/ control ratios) were about 2 for resting T cells, the TAR-61/ 48 HVS clone, and the CEM-6 line (for both autophosphorylation of ~6 0 '~" and phosphorylation of enolase). The EBR-65/57 clone demonstrated activation index 3 for autophosphorylation, and 4 for enolase. This activation was reproducihle, demonstrating the standard deviations between experi-   ments less than 30% for each t-ype of cell. The fact, that the activation of p60"" protein kinase activity followed crosslinking of CD3 in four distinct kinds of T cells suggests that this represents a general feature of TcR ligation in 'r cells.
Among the similarities in tyrosine phosphorylation ohserved hetween the different T cells is the apparent coprecipitation of two cellular proteins (~11.5-120 and p8.5) with p60"". The level of tyrosine phosphorylation of both these proteins was suhstantially increased in Fyn immunoprecipitates after cross-linking of CD.7 in all cell t-ypes. Interestingly, a few differences between the T cells used in our experiments were observed. First, the amount of p60"" expressed in CEV-  lunes3-7, anti-p60fy" serum at 0,0.5,1,3, and 10 min, respectively;  lanes 8 and 9, anti-p56Ick serum, control ( 8 ) and 3 min after crosslinking (9). Positions of tyrosine kinases are indicated. Cells were stimulated by OKT3 cross-linking. b, tyrosine phosphorylation of p6OfY" following cross-linking of CD3 (OKT3) or CD4 (Max. 16H5). Immunoprecipitation of p6Ory" followed by APT immunoblotting. Lanes 1-7, same as in panel a; lane 8, untreated control; lane 9 , 3 min after CD4 cross-linking. 6 cells was substantially lower than that expressed in other T cells studied rendering p6OfYn in CEM-6 cells undetectable by immunoblotting. Second, the protein band corresponding to molecular mass 145 kDa was highly phosphorylated on tyrosine following TcR stimulation in T cell clones and peripheral T cells, but was not detected in CEM-6 cells. Third, the ratio between autophosphorylation and phosphorylation of enolase was different in different cells. The average ratios were about 2, 1, 0.5, and 0.3 for resting T cells, EBR-65/57, TAR-61/ 48HVS, and CEM-6 respectively. Activation did not change this ratio dramatically, indicating that both autophosphorylation and phosphorylation of enolase increased proportionally. This finding, although not yet understood, may reflect fundamental differences in physiology of normal and continuously proliferating cells.
Interestingly, moderate changes in p6OrY" activity correlated with dramatic increase in total tyrosine phosphorylation. This can be explained by several reasons. First, not only p6OfY", but also other tyrosine kinases, such as the TcR {-chain bound tyrosine kinase in Jurkat cells (19), may become activated or recruited to the activated TcR complex. Second, binding of substrates to activated tyrosine kinases can contribute to the increase in tyrosine phosphorylation (15,(20)(21)(22)(23). Third, the total tyrosine phosphorylation might better correlate with the TcR-bound p6OfYn activity. However, this activity could not be assessed, since ~6 0 '~" did not detectably associate with CD3 under conditions of these experiments (data not shown).
Only ~6 0 '~" immunoprecipitates demonstrated enhanced kinase activity following CD3 cross-linking. No detectable activation of ~5 6 "~ or ~62''~'" was found (data not shown). These data are supported by the finding that the phosphate content of p6Ory", but not ~5 6 ' '~ or ~6 2 "~'~ is increased following CD3 cross-linking ( Fig. 5a and data not shown). At least a portion of the increased phosphate incorporation into p6OrYn is due to phosphotyrosine since the reactivity of the fyn protein with APT antibodies increased after CD3 cross-linking (Fig. 5b).
As indicated above, no detectable increase of phosphate incorporation in p561ck was observed (Fig. 5a). There was also no observed increase in the tyrosine kinase activity of p56ICk or in its reactivity with APT antibody following CD3-crosslinking (data not shown). However, a significant activation of p56Iek followed cross-linking of CD4 (data not shown). These findings indicate that CD4-mediated signal transduction is independent of TcR-signaling.
To investigate the potential specificity of the observed ~6 0 '~" activation with regard to cell surface protein engagement, we examined ~60"'" protein kinase activity following surface cross-linking of a variety of T cell surface molecules.
These studies indicate that cross-linking of CD3 with different anti-CD3 antibodies resulted in activation of p6OrY", although cross-linking of CEM-6 CD4, CD2, or CD28 did not lead to observable differences in p60ryn activity (data not shown). Similar results were obtained using other types of T cells studied (data not shown). We cannot rule out that T cell stimulation by engagement of CD28 (being dependent on the functional state of T cells; Ref. 24) or by ligation of CD2 under different circumstances or in other T cells may alter p6OfY" activity.
Our results demonstrate that engagement of TcR results in the rapid activation of p6OrY" in human T cells. The stimulation of p6OfYn tyrosine protein kinase activity appears to be specific with regard to the T cell surface molecules crosslinked as well as specific for this particular Src-family member. The observed increase in the phosphotyrosine content of p6OfYn after engagement of TcR is consistent with the concept that activation accompanies enzyme autophosphorylation. These results identify the fyn gene product as at least one of the T cell tyrosine protein kinases that participate in TcR signal transduction pathways.