(Received for publication, October 2, 1996, and in revised form, January 13, 1997)
From the Department of Immunology, The Weizmann Institute of Science, Rehovot 76100, Israel
The protein tyrosine kinase Syk is activated upon
engagement of immune recognition receptors. We have focused on the
identification of signaling elements immediately downstream to Syk in
the pathway leading to T cell activation. To circumvent T cell receptor
(TCR)·CD3 activation of Src family kinases, we constructed a
signaling molecule with an extracellular single chain Fv of an anti-TNP
antibody, attached via a transmembrane region to Syk (scFv-Syk). In a
murine T cell hybridoma, direct aggregation of chimeric Syk with
antigen culminates in interleukin-2 production and target cell lysis. Initially, it causes an increase in the association between scFv-Syk and the cytosolic protein Cbl and subsequently promotes tyrosine phosphorylation of Cbl. Interestingly, although both Cbl and
phospholipase C- (PLC-
) are phosphorylated in this hybridoma upon
TCR·CD3 cross-linking, these two events are uncoupled in
scFv-Syk-transfected cells, in which we were unable to detect
antigen-driven PLC-
phosphorylation. These results support a model
in which Syk can initiate and directly activate the T cell's signaling
machinery and position Cbl as a primary tyrosine kinase substrate in
this pathway. Furthermore, for efficient PLC-
phosphorylation to
occur in these cells, the combined actions of different tyrosine kinase families may be required.
A variety of biochemical changes occur in T lymphocytes after antigen stimulation including increases in protein phosphorylation, lipid turnover, and intracellular Ca2+ levels, activation of specific enzymes, and induction of gene expression (1, 2). The earliest of these T cell receptor (TCR)1-triggered events is activation of protein tyrosine kinases (3, 4), although none of the subunits of the TCR·CD3 complex possess an intrinsic tyrosine kinase activity. The ability of the ligated receptors to induce increased tyrosine phosphorylation lies in their ability to interact with cytoplasmic non-receptor protein tyrosine kinases (5), including members of the Src, Csk, and Syk kinase families (1, 2).
Syk, a 72-kDa protein tyrosine kinase, is abundant in several
hematopoietic lineages, such as B cells, myeloid cells, and thymocytes
(6). Syk has been shown to associate with the cytoplasmic domains of
many immune recognition receptors including the Ig and Ig
chains
of the B cell antigen receptor (7), the Fc
receptor of mast cells
(8), and the Fc
receptors of monocytes and macrophages (9-12). Upon
engagement of these receptors, Syk becomes phosphorylated and is
thereby activated to phosphorylate itself and additional cellular
proteins. Participation of Syk in T cell activation was first suggested
by the finding that this kinase, when fused to the transmembrane and
extracellular domains of CD7 and CD16, respectively, could induce
complete T cell activation (13). More recently, Syk was found to be
constitutively associated with the TCR·CD3 complex in a basal state,
then rapidly autophosphorylated and enzymatically activated up to
20-fold after T cell stimulation (14). Importantly, the activation of
Syk does not depend on the presence of Lck in T cells, although both
Lck and Fyn can act as downstream amplifiers of Syk's initial signal
(14).
A model of sequential activation of Src and Syk family kinases has been
proposed (15, 16) whereby receptor clustering stimulates one or more of
the membrane-associated Src kinases, resulting in phosphorylation of
immunoreceptor tyrosine-based activation motif (ITAM) tyrosines within
receptor chains. The tandem SH2 domains of Syk (and Zap-70 in T cells)
can then bind to these newly created docking sites, leading to the
phosphorylation of Syk and an increase in its intrinsic kinase activity
(17). The molecular events leading to downstream processes such as
phosphatidylinositol 4,5-bisphosphate breakdown, PI3-kinase activation,
and stimulation of the Ras pathway is likely attributed to
phosphorylation of target proteins by both Src and Syk families of
kinases. Since Src kinases are localized within particulate fractions,
whereas Syk kinase is present mostly in the cytosolic fraction of cell lysates, it has been proposed that each type of kinase is positioned within the cell to interact with and phosphorylate a distinct subset of
proteins (18). Recently a number of reports have emerged implicating
the proteins HS1, cortactin, -tubulin, and phospholipase C-
(PLC-
) as specific substrates for Syk (18-21).
In this study, we focused on the identification of signaling elements
immediately downstream to Syk in the pathway leading to complete T cell
activation. Most studies on T cell signaling are carried out on cells
stimulated with a cross-linking antibody to the TCR·CD3 complex. In
this manner, both Src and Syk kinase families are rapidly activated,
and intricate signaling complexes are formed (22), making it difficult
to assign a particular T cell substrate to one kinase. To circumvent
this problem, we have constructed a signaling molecule with an
extracellular single chain Fv (scFv) motif attached via a transmembrane
region to Syk. This design, similar to one devised by Kolanus et
al. (13), allows for simple and direct clustering of Syk molecules
with antigen, thus bypassing the initial activation of Src family
kinases and their phosphorylation of receptor ITAM sequences. We show here that direct aggregation of Syk molecules in a T cell hybridoma transmits a signal culminating in the effective production of IL-2 and
target cell lysis. Initially, Syk aggregation causes an increase in the
association between Syk and Cbl and promotes the tyrosine
phosphorylation of Cbl, a cytosolic multidomain protein. Interestingly,
in this system we were unable to detect significant phosphorylation of
PLC-, a proposed substrate of T cell tyrosine kinases.
Materials
The following antibodies were used in these experiments: GK20.5
(anti-Sp6 monoclonal antibody), anti-Cbl and anti-Syk (Santa Cruz
Biotechnology, Santa Cruz, CA), anti-PLC-1 (for Western blotting from Transduction Laboratories, Lexington, KY and for immunoprecipitation an anti-peptide polyclonal antibody kindly provided
by Yosef Yarden at the Weizmann Institute (23)), and anti-phosphotyrosine 4G10 (Upstate Biotechnology, Inc., Lake Placid, NY). Protein A- and protein G-Sepharose were from Pharmacia (Uppsala, Sweden). Oligonucleotide synthesis and DNA sequencing were performed by
the Biological Services at the Weizmann Institute of Science. Trinitrophenyl-fowl
-globulin (TNP-F
G) or TNP-A.20 were made as
described previously (24, 25).
Construction of Chimeric Syk Molecules
Construction of the anti-TNP single chain Fv genes from the Sp6 anti-TNP monoclonal antibody has been described (25). For isolation of the Syk tyrosine kinase cDNA, reverse transcription-polymerase chain reaction was carried out on RNA from both Jurkat human leukemia and anti-CD3-stimulated human T cells. The polymerase chain reaction was performed in two stages, and DNA pieces were later joined at a native HindIII site. Primers were designed based on the porcine sequence with comparison to the predicted human amino acid sequence (13). Similarly, the CD8 transmembrane and hinge regions were amplified by reverse transcription-polymerase chain reaction from Jurkat RNA. The construct (scFv-Syk) was sequenced in its entirety before insertion into the pRSV expression vector. Oligonucleotides for construction of chimeric genes were synthesized as follows.
SykAmino-terminal fragment 5 primer:
5
-CGTCTAGAACCATGGCAGACAGTGCCAACCACTTGCCCTTCTTCT-3
;
amino-terminal fragment 3
primer: 5
-TTCTTCCCCTCGGGGATGGAAAGCTTCCC3
; carboxyl-terminal fragment 5
primer:
5
-AAGGACAAAACTGGGAAGCTTTCCATCCC-3
; carboxyl-terminal fragment 3
primer: 5
-CGCTCGAGTTTAATTAACCACATCGTAGTAGTA-3
.
5 primer:
5
-CCGGTCACCGTCTCTTCAGCGCTGAGCAACTCCATCATGTACTTCAG-3
; 3
primer:
5
-CGTCTAGAGTGGTTGCAGTAAAGGGTGATAAC-3
.
Expression of Chimeric Syk Molecules
Twenty µg of pRSV-scFv-Syk DNA were transfected into 27J
murine hybridoma cells by electroporation as described elsewhere (25).
Transfectants were selected in G418 at 2 mg/ml. Expression of chimeric
protein on the surface of transfected cells was evaluated by
immunofluorescence staining using the GK20.5 anti-Sp6 idiotype and
fluorescein isothiocyanate-labeled anti-mouse Fab antibody. For
Western blotting, cells were washed in phosphate-buffered saline and
solubilized in lysis buffer (50 mM Hepes, pH 7.5, 150 mM NaCl, 1% Triton X-100, 2 mM EDTA, 2 mM EGTA, 50 mM NaF, 2 mM Na3VO4, 1 mM phenylmethylsulfonyl
fluoride, 0.4% aprotonin (24.4 trypsin inhibitory units/ml), 5 µg/ml
leupeptin, 10 µg/ml soybean trypsin inhibitor) for 15 min on ice and
then centrifuged at 17,000 × g for 15 min at 4 °C
for collection of supernatants. Protein was quantitated using the
Bio-Rad protein assay dye reagent (Bio-Rad, Hercules, CA), and 50-100
µg of total lysate protein, in partially reducing (lysates plus
reducing sample buffer were incubated for 20 min at room temperature)
or nonreducing conditions, were separated on 10% SDS-polyacrylamide
gels. Proteins were then transferred to nitrocellulose membranes, and
reacted with the GK20.5 antibody. Recognition of the primary antibody
was visualized with the ECL Western blotting detection system
(Amersham, Buckinghamshire, UK).
Cellular Activation, Detection of Phosphotyrosine Proteins, and Immunoprecipitation
Immobilized TNP11-FG was used for antigen
stimulation of transfectants. Six-well plates were coated with 50 µg/ml TNP-F
G in phosphate buffered saline overnight and then
washed in DMEM, no fetal calf serum. Cells were washed in DMEM without
fetal calf serum and resuspended at 3 × 106/ml in the
same medium. 1.5 ml of cell suspension were placed on each well for the
indicated time periods at 37 °C, after which cells were collected,
quickly spun, and solubilized as above in lysis buffer without NaF but
containing 10 mM sodium pyrophosphate and 80 mM
-glycerophosphate disodium salt. The TCR-expressing hybridoma, MD45,
was stimulated by incubating 1 × 107 cells with 0.5 ml of supernatant from the anti-CD3 hybridoma 2C11. After indicated
time periods, cells were quickly spun and solubilized as above. Upon
separation by SDS-PAGE under reducing conditions and transfer of
proteins to nitrocellulose, immunoblotting was performed with the
monoclonal anti-phosphotyrosine antibody 4G10 followed by detection
with peroxidase-labeled goat anti-mouse antibody and ECL.
For immunoprecipitation experiments, 10 µg of anti-Cbl or 7 µl of
PLC-1 antiserum were added to equivalent amounts of
lysate protein for 2 h at 4 °C, after which 30-40 µl of
protein A- or protein G-Sepharose were added for another 2 h with
constant agitation. Anti-Sp6 antibody GK20.5, first bound to protein
G-Sepharose, was used for immunoprecipitation of chimeric Syk through
the scFv portion. Immunoprecipitates were washed three times in wash
buffer (10 mM Tris, pH 7.5, 140 mM NaCl, 0.1%
SDS, 1% Nonidet P-40, 10 mM EDTA, 2 mM EGTA, 1 mM phenylmethylsulfonyl fluoride, 0.4% aprotonin (24.4 trypsin inhibitory units/ml), 5 µg/ml leupeptin, 10 µg/ml soybean
trypsin inhibitor), resuspended in 1 × reducing SDS sample buffer, and separated by SDS-PAGE. Immunoblots were developed by ECL
detection after incubation with primary antibodies. Stripping was
performed according to the instructions provided by the supplier.
Kinase assays were performed on anti-Sp6 immunoprecipitates, which were
washed three times in a buffer containing 50 mM Hepes, pH
7, 150 mM NaCl, 0.1% Triton, 10% glycerol, 1 mM Na3VO4, and 5 mM
NaF. Sepharose-bound proteins were then incubated in the same buffer
containing 10 mM magnesium acetate, 10 mM
MnCl2, and 20 µCi of [-32P]ATP (3000 Ci/mmol, DuPont NEN) for 10 min at room temperature and washed three
more times in wash buffer before resuspension in reducing sample
buffer. After separation by SDS-PAGE, proteins were transferred to
polyvinylidene difluoride (MSI, Westboro, MA), and the membrane was
exposed to x-ray film both before and after treatment with 1 N KOH for 1 h at 55 °C.
Functional Assays
To measure specific IL-2 production, 105
transfectants expressing the scFv-Syk chimera were incubated with
3 × 105 TNP-modified A.20 or L1210 B lymphoma cells
in DMEM containing 10% fetal calf serum for 18-24 h. Alternatively,
transfectants were reacted with plastic-immobilized TNP-FG, at a
molar ratio of TNP11-F
G. Supernatants were collected,
and IL-2 production was assayed as described previously (26), using an
IL-2-dependent CTL line and methyltetrazolium acid
staining. To evaluate the ability of transfectants to specifically lyse
TNP-labeled target cells, the 51Cr-release assay was
performed over 4-8 h (26).
To achieve an
effective and direct means of clustering Syk molecules, we constructed
a chimeric molecule in which the scFv of an anti-TNP antibody, Sp6, is
attached via the hinge and transmembrane domains of CD8 to the
cytoplasmic tyrosine kinase. The scFv moiety serves as an efficient
method for aggregating Syk molecules using polyvalent TNP antigen and
an experimental handle to detect and isolate chimeric Syk molecules.
The CD8
hinge region has been shown to be important in the
expression and extension of Ig-like domains (27). Once constructed as
described under "Experimental Procedures," the chimeric signaling
molecule was cloned into an expression vector that contains the Rous
sarcoma virus-long terminal repeat promoter and the neomycin resistance
gene (28). DNA was transfected into a mutant of the murine T cell
hybridoma MD45, named 27J, which lacks the T cell receptor as a result
of defective production of its
chain (29). Stable transfectants
were selected in medium containing G418 and assayed for chimeric
protein expression. When total cell lysates were separated by SDS-PAGE
and then immunoblotted with an anti-idiotype antibody (GK20.5),
chimeric Syk molecules are detected at the expected size of 112 kDa
(Fig. 1A). The larger protein band at
approximately 200 kDa seen in Fig. 1A is the dimeric form of
the chimera (since recognition of the Sp6 antigen by GK20.5 is
destroyed under completely reducing conditions, boiling of lysates was
avoided).
To assure that the Syk molecules in this configuration retained protein
tyrosine kinase activity, chimeras were immunoprecipitated from
scFv-Syk-transfected cells (clone S8-1.9) with anti-idiotype antibodies, and washed immunoprecipitates were subjected to an in
vitro kinase assay. As seen in Fig. 1B, there was
effective incorporation of [-32P]ATP into several
molecules, the major one corresponding in size to that of the scFv-Syk
chimera itself. The gel in Fig. 1B was treated with 1 N KOH before exposure to film, yet the same pattern of
phosphorylated protein bands was seen before treatment, confirming that
the incorporation of [
-32P]ATP was for the most part
on tyrosine residues. The same antibody did not precipitate any kinase
activity from nontransfected 27J cells.
Fig. 1C shows fluorescence-activated cell sorter analysis of three representative transfectants and nontransfected 27J cells after staining with antibody to the extracellular scFv, and then with a secondary fluorescein isothiocyanate-conjugated antibody. These results confirm the surface expression of the chimeric scFv-Syk protein.
Cross-linking of Chimeric scFv-Syk Initiates Production of IL-2 and Target Cell CytolysisTo assess the ability of scFv-Syk to
initiate IL-2 production in the T cell hybridoma, transfectants were
incubated with antigen either as TNP-modified carrier protein
(TNP-FG) or TNP-modified mouse B lymphoma cells (TNP-A.20 or
TNP-L1210). Supernatants were then tested for their ability to support
the growth of an IL-2-dependent CTL-L line. The results in
Fig. 2A show the significant secretion of
IL-2 from chimeric Syk-transfected cells upon stimulation with either
type of antigen. The homologous kinase Zap-70 in chimeric form,
however, was not able to trigger IL-2 production upon antigen triggering, despite being present on the cell surface (not shown). The
scFv-Syk chimeras can also mediate specific, non-major
histocompatibility complex-restricted cytolytic activity toward
TNP-modified A.20 target cells as shown in Fig. 2B.
Triggering of Chimeric Syk Molecules Results in Cbl Phosphorylation
To reveal the major proteins that undergo
tyrosine phosphorylation due to antigen-induced clustering of Syk
molecules, nontransfected 27J and scFv-Syk bearing cells (clone S8-1.9)
were stimulated with TNP-FG, lysed, and analyzed by
anti-phosphotyrosine immunoblotting (Fig.
3A). In transfected cells only, a protein
with a molecular mass of ~112 kDa is detected at all time points
tested, corresponding in size to that of the chimeric scFv-Syk
molecule. Immunoprecipitation experiments with anti-idiotype antibodies
(GK20.5), followed by anti-phosphotyrosine immunoblotting, confirmed
that scFv-Syk chimeras are indeed constitutively phosphorylated (Fig.
3C). In many experiments we have noted a slight decline in
chimera phosphorylation upon longer stimulation periods (Fig.
3C, 30-min stimulation). However, we were unable to observe
any reproducible change in Syk's kinase activity upon antigen
stimulation when tested by in vitro phosphorylation of
immunoprecipitated molecules (data not shown).
After 10-min stimulation with antigen, the first
tyrosine-phosphorylated protein detected in S8-1.9 transfectants has a
molecular mass of ~120 kDa (Fig. 3A). As expected,
nontransfected 27J cells show no phosphotyrosine response upon
incubation with TNP-FG. Since the TCR tyrosine kinase substrate Cbl
is of the same size, we tested whether Cbl is identical to the protein
phosphorylated upon stimulation of chimeric Syk. This 120-kDa
cytoplasmic protein contains potential nuclear localization and zinc
finger domains, proline-rich, and leucine zipper regions (30). Cbl is
the cellular homolog of v-cbl, first discovered as the
transforming gene of the Cas NS-1 murine retrovirus which causes pre-B
lymphomas and myelogenous leukemias (31).
When the blot in Fig. 3A was stripped and reprobed with
anti-Cbl antibodies, Cbl protein exactly aligned with the upper
tyrosine-phosphorylated band (Fig. 3B). To confirm this
finding, immunoprecipitation experiments were performed with anti-Cbl
antibodies as shown in Fig. 4. Antigen stimulation of
scFv-Syk molecules for 10 min results in the specific tyrosine
phosphorylation of Cbl, which peaks at 20 min (Fig. 4, left
panel). Importantly, the amounts of Cbl immunoprecipitated from
the various lysates were approximately equal (Fig. 4, middle panel). The lower band in the anti-phosphotyrosine immunoblot corresponds to the scFv-Syk chimera (see below). Interestingly, we also
see a transient increase in phosphorylation of this Cbl-bound chimera,
which was undetectable in anti-idiotype immunoprecipitates (see Fig.
3C).
An Increase in Syk-Cbl Association Precedes the Syk-mediated Phosphorylation of Cbl
The finding that aggregation of chimeric
Syk triggers phosphorylation of Cbl suggested a possible association
between these two molecules. Indeed, we do detect a constitutive
association between chimeric Syk and Cbl, as seen by immunoblotting Cbl
immunoprecipitates with anti-Syk antibodies (Fig. 4, right
panel, 0 min). This basal association is most likely
not mediated by tyrosine phosphorylation since addition of up to 50 mM of p-nitrophenyl phosphate to
immunoprecipitation reactions does not disrupt it (Fig.
5A). This compound has been shown to inhibit
binding of anti-phosphotyrosine antibodies (32) and to disrupt protein
interactions that depend on the presence of phosphotyrosine (33, 34).
Interestingly, when chimeric Syk molecules were immunoprecipitated with the GK20.5 anti-idiotype antibody, we detected a significant increase in their association with Cbl upon antigen stimulation (Fig. 5B). As a result of this switch in experimental protocol, the coprecipitation of Cbl through anti-scFv idiotypic antibodies is often much lower than in the reciprocal configuration, and the detectable increase in Syk-Cbl association is more pronounced (compare Fig. 4, right panel, and Fig. 5B, left panel). This is not surprising considering that scFv-Syk chimeras are greatly overexpressed in S8-1.9 cells, in excess of possible Cbl molecules to bind, whereas a large proportion of endogenous Cbl protein does bind the Syk chimera in the basal state. Fig. 5B shows a peak in Syk-Cbl association after 10-min antigen stimulation, just preceding that seen for Cbl phosphorylation (Fig. 4). This heightened association between Syk and Cbl molecules persisted for at least 40 min. Although the amount of basal and induced Syk-Cbl association can vary between experiments, densitometer scanning of autoradiograms from several experiments confirmed a 2-7-fold increase upon stimulation of chimeric Syk molecules with immobilized antigen.
Activation of Chimeric Syk Molecules Does Not Cause Significant Phosphorylation of PLC-Many studies on T cell activation have
shown the importance of PLC- as a substrate for T cell tyrosine
kinases (35-38). In fact, a number of reports have suggested this
enzyme is a direct substrate for the Syk kinase (13, 21, 39-41). To
assess the state of tyrosine phosphorylation of PLC-
in our system,
cells containing chimeric Syk molecules were stimulated with antigen as
above, and cell lysates were probed with anti-PLC-
1
antibodies. When PLC-
1 immunoprecipitates were
immunoblotted with anti-phosphotyrosine antibodies, we could not detect
significant PLC-
1 phosphorylation upon cross-linking of
chimeric Syk, despite notable levels of PLC-
in these cells (Fig.
6, panels A and B). This is in
marked contrast to the pronounced phosphorylation of Cbl in stimulated cells of the same experiment (Fig. 6, panels C and
D). Interestingly, despite the lack of PLC-
1
phosphorylation in antigen-stimulated S8-1.9 cells, there is a
constitutive association between PLC-
1 and the scFv-Syk
chimera (Fig. 7). This is likely due to an interaction between the SH2 domains of PLC-
1 and tyrosine
phosphorylated scFv-Syk (see Fig. 3C), as has been shown to
occur in antigen receptor-stimulated B cells (41).
To verify that PLC-1 could serve as a substrate for
activated tyrosine kinases in these cells, we tested whether it
underwent phosphorylation upon stimulation of the endogenous TCR
complex in the TCR-expressing hybridoma, MD45. These cells indeed
display a pronounced increase in both PLC-
1 and Cbl
phosphorylation in response to anti-CD3 triggering (Fig.
8). Thus, although the activation of
PLC-
1 may be an important step in the pathway emanating
from the TCR·CD3 complex, it appears to be much less crucial in the signal propagated through chimeric Syk molecules.
Given the importance of Syk kinase activity in lymphocyte activation (2, 16), identification of its early targets of phosphorylation and association in T cells is an immediate goal for understanding propagation of the signal leading to T cell activation. In the direct triggering of the chimeric scFv-Syk molecules by an antigen stimulus, we found that the first detectable tyrosine kinase event was phosphorylation of the proto-oncogene c-cbl (Figs. 3 and 4). Cbl is strongly phosphorylated on tyrosine residues upon stimulation of a variety of growth factor and antigen receptors (42-47). As a multidomain protein, Cbl is capable of interacting in vivo with a variety of molecules involved in cellular signaling, including protein tyrosine (43, 48-52) and lipid (33, 53) kinases, and adaptor proteins (44, 51, 53-57). Although its biological role is at present unclear, its participation in cellular transformation (58-60) suggests an important function for Cbl in cellular proliferation.
Cbl is also one of the earliest and most prominent protein tyrosine kinase substrates in this T cell hybridoma when stimulated through the TCR·CD3 receptor complex (Fig. 8, C and D). It has recently been suggested that by interacting with adaptor proteins, such as Grb2, Crk, and Crk-L, Cbl may affect the function of nucleotide exchange factors including Sos, Vav, and C3G (57, 61). Since direct triggering of Syk allows for Cbl phosphorylation, we speculate that this kinase may be an important initiator of the pathway that regulates guanine nucleotide exchange on small G proteins in T cells.
We observed a basal level of association between scFv-Syk chimeras and Cbl (Figs. 4 and 5), which is increased in a time-dependent fashion upon antigen stimulation (Fig. 5). Thus our experiments demonstrate that direct aggregation of Syk kinase molecules in T cells augments their association with Cbl, while the kinetics of this event just precedes that of aggregation-induced Cbl phosphorylation. A similar activation-dependent complex was shown to form between Syk and Cbl upon stimulation of the B cell antigen receptor (51). Furthermore, we could only detect antigen-induced phosphorylation of scFv-Syk chimeras which coimmunoprecipitate with Cbl (Fig. 4), and not in the larger cellular pool of chimeric molecules (Fig. 3). These data suggest a direct phosphorylation of Cbl by the chimeric Syk kinase, although the indirect involvement of additional kinases cannot be ruled out (51). The Src family member Fyn is one candidate kinase, as several studies have shown it to exist in preformed complexes with Cbl in unstimulated T cells (45, 62). We have also noted a constitutive association between Fyn and Cbl in both scFv-Syk-transfected and TCR-expressing MD45 cells (not shown). However, although anti-CD3 stimulation results in an increase in Fyn-associated phospho-Cbl in MD45 cells, the same is not true for antigen stimulation of scFv-Syk receptors (not shown). Another Src family member, Lck, although expressed in the hybridoma (not shown), is likely not involved in the phosphorylation of Cbl, since this event can occur efficiently in the mutant Jurkat cell line, JCaM1.6, which lacks Lck expression (63).
The nature of the Syk-Cbl association in scFv-Syk-transfected cells is
not clear. Band and colleagues (64) just recently described a novel
phosphotyrosine-binding domain in the NH2-terminal region
of Cbl that can directly interact with Zap-70 in stimulated T cells.
During the course of our studies we were unaware of the identity of
this region and therefore did not directly assess its involvement.
However, in the resting state at least, it appears that the association
between Cbl and scFv-Syk is not mediated by phosphotyrosine, because we
were unable to disrupt it with a high concentration of the
phosphotyrosine analog, p-nitrophenyl phosphate (Fig.
5A). Despite the detected increase in association upon
scFv-Syk aggregation, a nondirect interaction cannot be discounted, since we also see the presence of additional signaling proteins interacting with either chimeric Syk (Fyn, PLC-1) or Cbl
(Fyn, Grb2, PLC-
1) prior to stimulation (Fig. 7 and not
shown). Syk contains two SH2 domains which, upon cell activation, could
bind phosphotyrosines within Cbl, resulting in the increased
association detected. There are in fact sequences within Cbl which
resemble the general motif most recognized by the COOH-terminal Syk SH2 domain (65). Previous studies by Marcilla et al. (43) showed a small amount of Cbl to be associated with Syk immunoprecipitates in
the monocytic cell line HL-60. However, despite effective
phosphorylation of Cbl upon Fc
R receptor engagement, complex
formation was not affected, further suggesting a nondirect interaction
between the two molecules.
PLC- is a prominent and early substrate of T cell tyrosine kinases
upon activation of the TCR·CD3 complex (35-38). This is also true in
MD45 cells when activated in the same fashion (Fig. 8, A and
B), confirming that the PLC-
pathway is intact in the original hybridoma. However, when stimulated via aggregation of Syk, we
were unable to detect significant phosphorylation of
PLC-
1 in the scFv-Syk transfected line S8-1.9 (Fig. 6,
A and B), despite the fact that
PLC-
1 and scFv-Syk are constitutively associated (Fig.
7). Thus we see an uncoupling of these two tyrosine phosphorylation events (Cbl and PLC-
) in antigen stimulation of S8-1.9 cells, which
still effectively leads to their activation. We propose that the
transmission of signals to the nucleus for IL-2 gene transcription is,
at least in part, qualitatively different in cells stimulated through
the TCR·CD3 complex versus direct aggregation of Syk
kinase.
The lack of significant PLC-1 phosphorylation in our
system was unexpected in light of recent results supporting a model in
which this enzyme is a direct substrate of Syk kinase (13, 21, 39-41).
PLC-
1 may indeed be phosphorylated at very low levels, yet our repeated efforts to detect this event, using a panel of both
monoclonal and polyclonal antibodies, make this an unlikely possibility. Antibody aggregation of similar chimeric proteins (based
on CD16 extracellular, CD7 transmembrane, and Syk intracellular regions) in Jurkat leukemia cells was in fact shown to result in
PLC-
1 phosphorylation (13). Differences in the cell type studied may be an important differentiating factor since we have noticed that, in MD45 cells (although there is clear phosphorylation of
this protein upon anti-CD3 stimulation (Fig. 8)), the ratio of
phosphorylated to total PLC-
1 immunoprecipitated is much
less than that in Jurkat cells (not shown), in which most of the
studies on PLC-
phosphorylation have been performed.
Earlier reports have suggested a role for Src family kinases in
coupling the TCR to the activation of PLC-1 (66, 67). In
fact, five different Src family kinases were shown to efficiently phosphorylate purified PLC-
1 and PLC-
2
in vitro (68). Furthermore, a recent study of chicken
B cells deficient in Bruton's tyrosine kinase revealed that
PLC-
2 phosphorylation and activity are regulated by this
kinase as well (69). Homologous Bruton's tyrosine kinase family
members, Emt and Itk, are present in T cells (70, 71) where they may
play a role, along with Src and Syk family kinases, in the activation
of PLC-
. Interestingly, in a heterologous system in which B cell
receptor-induced signaling was reconstituted in a nonlymphoid cell line
(72), Syk activity was necessary but not sufficient to couple the B
cell receptor to activation of PLC-
1. Collectively,
these data suggest that for efficient PLC-
1 phosphorylation and/or activation, the combined actions of different tyrosine kinase families is required, a situation which may be unattainable by direct triggering of chimeric Syk molecules.
While several cytoplasmic protein tyrosine kinases have been implicated in the development and activation of T lymphocytes (1, 2), and a number of models proposed to describe the sequence of events following engagement of the TCR, discrepancies still remain. One open question concerns the requirement for Src kinases in the activation of Syk (14, 73). The construction of a chimeric protein based on this key intracellular tyrosine kinase, linked to the extracellular scFv, has allowed for direct triggering of the T cell-signaling machinery. In this T cell hybridoma, clustering of Syk is sufficient to achieve complete T cell activation, leading to antigen-specific IL-2 production and target cell lysis (Fig. 2). Similar conclusions were drawn from studies in Jurkat T cells transduced with CD16-CD7-Syk chimeric molecules (13), as discussed above. In these studies, aggregation of Syk chimeras alone, but coaggregation of chimeras bearing Fyn and Zap-70 kinases, allowed for initiation of cytolytic effector function. In the MD45 T cell hybridoma we have also observed the inability of Zap-70 to initiate T cell functions, despite efficient surface expression and activity of scFv-Zap-70 chimeras (data not shown). Likewise, the function of a Zap-70 transgene in COS cells (73, 74) is dependent on Src kinases Lyn or Fyn.
The situation with Syk is more complex. In B cell receptor signaling a requirement for Lyn to induce tyrosine phosphorylation and activation of Syk was demonstrated (75). In transfected COS cells, a similar dependence on Src kinases has been reported by some investigators (73, 75) but not others (14). In fact, Syk was shown to positively regulate Lck phosphorylation and activity in transfected COS cells (76), while in T cells Syk is constitutively bound to the TCR·CD3 complex and rapidly activated upon stimulation, even in the absence of Lck (14). Furthermore, a recent report showed the intrinsic kinase activity of Syk to be significantly superior to that of Zap-70 (77), possibly explaining its functional autonomy. The results we present herein with chimeric Syk molecules support a model in which Syk, as an integral part of the TCR·CD3 complex, possesses initiating capabilities once aggregated. This aggregation can result from engagement of the TCR·CD3 complex and may be Src family kinase-independent.