| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
*
Research Center, Notre-Dame Hospital, University of Montreal, Montreal, Quebec, Canada;
Department of Surgery, the Second Affiliated Hospital of the Zhejiang Medical School, Zhejiang University, People’s Republic of China;
Nephrology Service, Notre-Dame Hospital, University of Montreal, Montreal, Quebec, Canada; and
Department of Surgery, McGill University, Montreal, Quebec, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
25% of the known RTKs; they are now named EphAs
(EphA1 to EphA8) and EphBs (EphB1 to EphB6) according to their sequence
homology (1). At the protein level, EphB6 shares
47–60% homology with other members of EphB subgroups (2, 3), whereas homology between human and mouse EphB6 is as high as
91.1% (2, 3). Such a high degree of homology suggests
important conserved functions of this molecule. Eph ligands are mainly expressed on cell surfaces and are now called ephrins. EphrinAs (ephrinA1 to ephrinA5) are ligands of EphAs and are GPI-anchored proteins. EphrinBs (ephrinB1 to ephrinB3) are ligands of EphBs and are transmembrane proteins. EphAs bind ephrinAs and EphBs bind ephrinBs with loose specificity. Generally, they do not bind ligands of the other group. The transmembrane ephrinBs can also function as reciprocal receptors for EphB molecules and transduce signals into cells (4).
Because both Eph receptor kinases and their ligands are all cell surface molecules, they can only interact with each other if expressed on adjacent cells. Not surprisingly, the clearly demonstrated function of these receptors and ligands is to control accurate spatial patterning and cell positioning. Most of these findings are derived from studies on the CNS where most Eph kinases have high expression levels (5, 6). Recently, it has been found that ephrinB2 and its ligand EphB4 are involved in angiogenesis, and such a function is consistent with the known roles of Eph kinases in controlling spatial structure formation (7).
A few Eph RTK members, such as EphB6 and EphB4, have high levels of expression in hemopoietic cells and in the thymus (2, 3). However, despite their expression in these tissues and their presumed importance as both cell surface receptors and tyrosine kinases, we have no knowledge as to the function of any member of this important RTK family in the immune system. In this paper, we report a novel finding on functional changes of Jurkat T cells after EphB6 and CD3 co-cross-linking. Signal transduction after EphB6 cross-linking was also examined. This work represents the first endeavor in exploring the roles of Eph kinases in the immune system.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cells were cultured in RPMI 1640 supplemented with 10% FCS and antibiotics (8). Tonsillar T cells, B cells, and PBMC were prepared as described before (8, 9). Thymidine uptake was measured for cell proliferation as described before (8). In some experiments, the culture wells were coated with mAbs. For coating, mAbs of indicated amount in PBS were added to wells (50 µl/well), and the plates were incubated overnight at 4°C. The wells were washed three times with PBS before use.
Northern blot analysis
Total cellular RNA, extracted from different cell lines, tissues, PBMC, or tonsillar B lymphocytes, was analyzed by Northern blotting as described in a previous publication (9). In some experiments, mRNA isolated with FastTrack 2.0 mRNA isolation kits (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions was used for Northern blotting. A 1.25-kb SmaI/EcoRI fragment of human EphB6 cDNA (from positions 2741–3989) was used as a probe.
Preparation of recombinant proteins corresponding to the extracellular domain of human EphB6
A 0.5-kb PCR fragment coding for part of the extracellular domain of human EphB6 (from amino acid positions 31 to 202) was cloned into a prokaryotic expression vector, pGEX-2TK. The resulting construct, pGST-2TK-human protein tyrosine kinase (HPTK), encoded a recombinant fusion protein between GST and the Eph extracellular domain. The recombinant protein, designated as GST-HPTK, was expressed in Escherichia coli BL21 and was purified with glutathione-Sepharose beads.
Preparation of mAb against the extracellular domain of EphB6
Basic procedures for mAb preparation have been described by Coligan et al. (10). Briefly, a BALB/c mouse was immunized with the recombinant protein GST-HPTK. Spleen cells of the mouse were fused to SP2/0-Ag14 myeloma cells. Culture supernatants of hybridomas were screened by flow cytometry according to their binding to Jurkat cells. Limiting dilutions were performed to obtain hybridoma clones.
Flow cytometry
To detect EphB6 expression, Jurkat cells were stained with mAb 4F12 followed by PE-conjugated goat anti-mouse IgG F(ab')2 (Jackson Laboratories, West Grove, PA). Detailed procedures of staining were described in a previous publication (11). To detect apoptotic cells, Jurkat cells were double-stained with annexin V-FITC (BD PharMingen, San Diego, CA) and propidium iodide (PI). The PI-negative population was gated to measure their expression of annexin V. TUNEL was used to monitor Jurkat cell apoptosis with kits from Roche Diagnostics (Laval, Quebec, Canada) according to the manufacturer’s instructions.
Cross-linking of EphB6 and CD3
Jurkat cells were incubated in serum-free RPMI 1640 medium for 45 min on ice in the presence of mAb 4F12 (anti-EphB6), mAb OKT3 (anti-CD3), or both (all at 2 µg per 106 cells). The cells were washed and resuspended in 37°C serum-free medium in the presence of rabbit anti-mouse IgG (10 µg/ml, final concentration). Cross-linking was terminated by adding 10 ml cold PBS to the cells, which were spun down and lysed in TNE buffer, as detailed in a previous publication (12).
In vitro lymphokine production
To measure lymphokine production, supernatants from Jurkat cell
culture were collected 72 h after anti-CD3 and anti-EphB6
cross-linking. IL-4, IFN-
, and GM-CSF in the supernatants were
quantified by ELISA (Quantikine; R&D Systems, Minneapolis, MN)
according to the manufacturer’s instructions.
Dot blotting, immunoblotting, and immunoprecipitation (IP)
For dot blotting, different undenatured recombinant GST fusion
proteins were spotted on nitrocellulose membranes. The membranes were
then hybridized with mAb 4F12 in the absence or presence of competitor
recombinant protein GST-HPTK. Signals were revealed by ECL. For
immunoblotting, Jurkat cell lysates or immune complexes precipitated by
various Abs were resolved in SDS-PAGE and blotted onto
polyvinylidene difluoride membranes. The membranes were hybridized
first with anti-EphB6 mAb 4F12, anti-phosphotyrosine mAb RC20
(Transduction Laboratories, Lexington, KY), polyclonal goat Ab against
EphB6 (Santa Cruz Biotechnology, Santa Cruz, CA), mAb against
protein-tyrosine phosphatase 1C/Src homology 2 domain-containing
tyrosine phosphatase-1 (SHP-1) (Transduction Laboratories), or
polyclonal rabbit Ab against Cbl (Santa Cruz Biotechnology). Signals
were detected by ECL. For IP, Jurkat cell lysates were reacted with
rabbit Abs against CrkII, CrkL, GRAB2, p85
phosphatidylinositol
3-kinase (PI-3K), Vav, or Cbl (Santa Cruz Biotechnology). Immune
complexes were brought down by protein A+G-Sepharose beads. Detailed
methods of immunoblotting and IP have been described in a previous
publication (12).
Detection of soluble Fas ligand (FasL)
Soluble FasL in culture supernatants was detected using a commercial ELISA from Oncogen Research Products (Boston, MA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
EphB6 messages are reported to have high expression levels in the
adult human brain and pancreas (3) and in the adult mouse
brain and thymus (2). Here, its expression was
investigated in selected human tissues, tumors, and immune cells.
According to total RNA blotting (Fig. 1
A), EphB6 had no detectable
expression in the heart and muscle, and low expression in the colon,
colon tumor, lung, and lung tumor. Its expression in the spleen and B
cells was also quite low, whereas Jurkat cells had high expression.
Among the cell lines tested (Fig. 1B), Jurkat cells had the
highest expression, followed by a chronic myelogenous leukemia line,
K562, and Burkitt B cell lymphoma, Namalwa cells. EphB6 expression was
at a low level in histiocytic lymphoma U937 cells and promyelocytic
leukemia HL-60 cells, but was not detectable in cervical cancer HeLa
cells and Burkitt lymphomas Jijoye and Daudi cells.
|
C). According to Northern blotting using total
RNA of T cells (Fig. 1C, left), EphB6 expression
in T cells was reasonably high. After their activation by PHA, the
expression was decreased, and treating the T cells with an
immunosuppressant rapamycin during the PHA activation prevented to a
certain degree such decrease. The biological significance of such
modulation is yet to be understood. Poly-A-enriched RNA from
rosette-enriched tonsillar T cells (90% CD3 positive), and resting or
activated tonsillar B cells (85% CD19 positive) was also tested in
Northern blotting (Fig. 1C, right). In this
experiment, EphB6 mRNA could be easily detected, and the T cells had
higher expression than B cells. However, EphB6 expression had no
apparent modulation in B cells activated by Staphylococcus
aureus Cowan I bacteria plus IL-2 for 16 or 40 h.
Thus, EphB6 mRNA was expressed in normal T and B cells and was
down-regulated after activation in T but not B cells. Preparation of recombinant EphB6 and mAb against EphB6
To verify EphB6 expression at the protein level as well as for
further experimentation on EphB6 function, we produced a mAb against
the extracellular domain of EphB6. A GST-HPTK fusion protein (GST fused
with a part of the human EphB6 extracellular domain) was expressed in
E. coli and purified with glutathione-Sepharose beads (Fig. 2A). The highly purified
fusion protein was used to immunize a BALB/c mouse, and hybridomas were
produced. The hybridomas were screened with flow cytometry for mAbs
that bound on surfaces of Jurkat cells. A positive clone 4F12 (IgG1)
was identified, and almost all Jurkat cells, which had high EphB6 mRNA
expression, were highly positive for 4F12 staining (Fig. 2B). K562 cells, which had modest EphB6 mRNA expression,
were modestly 4F12 positive, whereas Daudi cells, which had no
detectable EphB6 mRNA, were 4F12 negative. Thus, the 4F12 staining
correlated to the EphB6 mRNA expression. The expression pattern of
EphB6 in these malignant cells according to 4F12 staining is consistent
with that reported by Shimoyama et al. (13) using their
anti-human EphB6 mAb T49-25. Moreover, EphB6 could be detected on
3–5% peripheral T cells according to 4F12 staining (data not shown),
and this percentage is also in agreement with that in the publication
of Shimoyama using mAb T49-25 (13).
|
C). 4F12 bound
strongly to undenatured immunogen GST-HPTK, but not to GST, GST fused
with a different region of the human EphB6 extracellular domain
(GST-control-1), or GST fused with the mouse EphB6 intracellular domain
(GST-control-2) (Fig. 2D). 4F12 binding to solid-phase
GST-HPTK could be blocked by soluble GST-HPTK (Fig. 2E).
These results illustrate the specificity of mAb 4F12.
The specificity of mAb 4F12 was further confirmed by cell adhesion
assay. Jurkat cells grow normally in suspension and form clumps, as
shown in the left panel of Fig. 3A. When the cells were
cultured in wells coated with 4F12, the mAb anchored the cells to the
bottom of the wells and they could no longer form clumps (Fig. 3A, second column). This effect was dose
dependent in that when mAb concentrations used for coating were reduced
from 1 to 0.2 µg/well/50 µl, small clumps started to appear.
Isotypic control mAb (Fig. 3A, third
column) or 4F12 in solution (Fig. 3A,
fourth column) had no effect on clump formation.
When recombinant EphB6 extracellular domain GST-HPTK was added to the
coated wells, it reversed the anchoring effect of 4F12, as shown in
Fig. 3B. This reversing effect appeared in wells coated with
0.5 µg/50 µl/well 4F12, and was more obvious in wells coated with
less 4F12 (0.2 µg/50 µl/well) (Fig. 3B, third
column). For all these cultures, Jurkat cell proliferation was not
affected (data not shown), indicating that 4F12 or recombinant proteins
had not toxic effects on the cells.
|
Co-cross-linking of EphB6 and CD3 results in activation-induced death of Jurkat cells
So far, there are no reports pertaining to the function of any Eph receptor kinase in immune cells, nor there are known functions of EphB6 in any type of tissues or cells. With mAb available against the EphB6 extracellular domain, we searched for EphB6 function in the immune system, using Jurkat cells as a model.
First, we examined lymphokine production by Jurkat cells after
cross-linking their surface CD3, EphB6, or both. Jurkat cells were
cultured in wells coated with a suboptimal concentration of OKT3
(anti-CD3, 0.1 µg/50 µl/well), an adequate concentration of
4F12 (anti-EphB6, 0.25 µg/50 µl/well), or both for 3 days.
IL-4, IFN-, and GM-CSF in the supernatants were assayed with ELISA.
As illustrated in Fig. 4A,
IL-4 produced by Jurkat cells cross-linked with anti-CD3,
anti-EphB6, or both showed no changes. The anti-CD3 alone
moderately enhanced IFN- production, but anti-EphB6 alone had no
effect. However, co-cross-linking of these two molecules led to reduced
IFN- production. In contrast, such co-cross-linking stimulated
GM-CSF production, whereas cross-linking with anti-CD3 or
anti-EphB6 alone exerted no effect. The increase of GM-CSF
indicates that co-cross-linking was not a null or nonspecific toxic
event, but activated a certain cellular program in Jurkat cells.
|
B, Jurkat cells cultured in wells
coated with the anti-CD3 or the anti-EphB6 had proliferation
rates comparable to those cultured in uncoated wells. However, cells
cultured in wells coated with both the anti-CD3 and anti-EphB6
had significantly reduced proliferation. It is to be emphasized that
the anti-CD3 alone had no effect because it was used at a
suboptimal concentration. When it was used at a high concentration, it
alone caused apoptosis and consequently reduced thymidine uptake of the
Jurkat cells (data not shown).
We wondered whether inhibition of proliferation reflected
activation-induced cell death. To evaluate this possibility, Jurkat
cells were cultured in wells coated with a suboptimal concentration of
anti-CD3, an adequate concentration of anti-EphB6, or both for
24 h and stained with anti-annexin V and PI. PI-negative cells
were gated for their annexin V expression. As shown in Fig. 4C, cross-linking of Jurkat cells with the anti-CD3 or
the anti-EphB6 alone did not cause apoptosis, but co-cross-linking
of both molecules did in that annexin V-positive cells increased from
<2.4 to 31.4%.
In lymphocytes, apoptosis is often a Fas-mediated process. We thus
assessed soluble FasL produced by these Jurkat cells. Supernatants from
uncoated wells, wells coated with a suboptimal concentration of
anti-CD3 (0.1 µg/50 µl/well for coating) alone, or an adequate
concentration of anti-EphB6 (0.25 µg/50 µl/well for coating)
alone had very low levels of soluble FasL (<25 ng/ml) (Fig. 4D). When Jurkat cells were cultured in wells coated with
the anti-CD3 and anti-EphB6 for 16 h, soluble FasL in the
culture supernatant rose to >200 pg/ml, which was at a comparable
level by treating Jurkat cells with optimal solid-phase anti-CD3 (1
µg/50 µl/well for coating). To verify that the apoptosis induced by
CD3 and EphB6 co-cross-linking was indeed mediated by Fas and FasL
interaction, we added soluble recombinant Fas-Fc protein to the above
described culture system. As shown in Fig. 4E, anti-CD3
and EphB6 co-cross-linking resulted in apoptosis of 34.7% Jurkat cells
after 24 h according to the TUNEL assay. The presence of Fas-Fc
(50 µg/ml) significantly suppressed apoptosis with only 17%
TUNEL-positive cells. A control recombinant protein with an identical
Fc tail had no effect on the apoptosis (Fig. 4E,
bottom panel, 33.6%).
The results of this section showed that after co-cross-linking of CD3 and EphB6 on Jurkat cells, a certain cellular program was activated, which leads to altered cytokine production and Fas-mediated apoptosis.
EphB6 could transduce signals into Jurkat cells
It has been reported that both human and mouse EphB6 have
mutations in the conserved kinase domain, and as a consequence they
lack intrinsic tyrosine kinase activity according to tests done using
transfected recombinant EphB6 (2, 3). Thus, it has been
speculated that EphB6 is a silent receptor with no signaling
capability. This speculation was obviously in conflict with our
observation that co-cross-linking of CD3 and EphB6 led to functional
changes of Jurkat cells. Certain signals must have been transmitted
from natural surface EphB6 into the cells. To examine this possibility,
tyrosine phosphorylation of total cellular proteins was assayed in
Jurkat cells cross-linked with mAb 4F12, anti-CD3 mAb (OKT3), or
both. The cell lysates were examined by immunoblotting using
anti-phosphotyrosine mAb RC20. After 4F12 cross-linking, there was
an transient increase in tyrosine phosphorylation of several proteins
of 50, 90, and 200 kDa in size. The phosphorylation declined after 5
min, and at 30 min the intensity of all the phosphoproteins was below
that at time 0. CD3 cross-linking induced stronger protein tyrosine
phosphorylation in many bands at 1 and 5 min. The overall pattern was
similar to that resulted from EphB6 cross-linking, but several bands of
35, 50, and 90 kDa in size were more prominently phosphorylated
between 1 and 5 min. Co-cross-linking of EphB6 with CD3 resulted in
pattern similar to that of anti-CD3 alone, with the exception that
the 90-kDa band was not strongly phosphorylated and the 200-kDa band
had a faster phosphorylation and dephosphorylation kinetics
(Fig. 5).
|
Interaction between EphB6 and Cbl
EphB6 had no intrinsic kinase activity, but it could transduce
signals and lead to functional changes in Jurkat cells. A logical
explanation for this is that EphB6 might associate with molecules
involved in signal transduction in T cells. We examined a list of
signaling molecules that might form complexes with EphB6. Included in
the list are adaptor proteins GRB2 and p85 of PI-3K that are known
to associate with other Eph kinases (14, 15). Several
other adaptor proteins, such as CrkL, CrkII, and Vav, were also
included. Also included was Cbl, which has adaptor function but also
has ubiquitin ligase activity (16). Jurkat lysates were
immunoprecipitated with rabbit Abs against the above-mentioned
molecules, and immune complexes were analyzed with immunoblotting using
a rabbit anti-EphB6 Ab. As shown in Fig. 6A, EphB6 was prominently
present in Cbl precipitates. EphB6 was also coprecipitated with GRB2,
CrkL, and CrkII, but to lesser degrees. No detectable EphB6 was found
in immunoprecipitates of p85 PI-3K or Vav. It is to be noted that
p85 PI-3K and Vav could be detected by immunoblotting in the p85
PI-3K and Vav immunoprecipitates (data not shown), indicating that the
anti-p85 PI-3K and anti-Vav Abs had adequate affinity
for IP.
|
B,
Cbl had constitutive phosphorylation, and 3 min after EphB6
cross-linking there was a decline in Cbl tyrosine phosphorylation. By
10 min, most of Cbl was dephosphorylated. In contrast, the level of Cbl
protein remained the same. This result indicates that Cbl is actively
involved in EphB6 signaling.
Cbl is a substrate of SHP-1 (18), and SHP-1 activity is
regulated by allosteric mechanisms involving interaction of its Src
homology 2 domain with other proteins (19, 20). We
wondered whether dephosphorylation of Cbl after EphB6 cross-linking was
related to changes in the interaction between Cbl and SHP-1. Jurkat
lysates were immunoprecipitated with anti-Cbl, and SHP-1 in immune
complexes was revealed by immunoblotting. As seen in Fig. 6C, SHP-1 could be detected directly in the Jurkat lysate
and anti-Cbl IP (first and second
lanes, respectively). The SHP-1 level started to diminish 5 min
after EphB6 cross-linking, and continued to do so at 15 min after EphB6
cross-linking. This decline was correlated with reduced tyrosine
phosphorylation of Cbl as shown in Fig. 6B. Cross-linking of
CD3 resulted in a faster disappearance of SHP-1 in the Cbl IP, and
co-cross-linking of EphB6 with anti-CD3 had a similar effect as
anti-CD3 cross-linking alone. Cbl protein levels showed no apparent
changes after different ways of cross-linking as expected
(lower panel of Fig. 6C). If SHP-1 is the
enzyme responsible for Cbl dephosphorylation in our model as reported
in other studies (18), it dissociates from its substrates
once its job is done.
We were not able to detect bands corresponding to phospho-SHP-1 in Cbl
precipitates (Fig. 6B), whereas SHP-1 protein could be
detected in the anti-Cbl precipitate (Fig. 6C). Because
the anti-phospho-protein mAb and anti-SHP-1 mAb are not
necessarily comparable in their sensitivity, these results are not
self-contradictory. The Cbl-associated SHP-1 proteins could be
phosphorylated but were below the detection level by the
anti-phospho-Ab RC-20. With that said, we cannot rule out the
possibility that most SHP-1 proteins associated with Cbl were not
phosphorylated, because SHP-1 is mainly regulated by changing of its
tertiary structure depending its association with other proteins.
The results of this section suggest that signal transduction from EphB6 is through proteins it associates with, Cbl being one of them.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Eph family kinases are in the center of attention in neurobiology due to their newly found functions in the guidance of neuron growth and spatial pattern formation, although the function of EphB6 is still unknown. A few of the Eph kinases, including EphB6, are expressed in lymphoid tissues. For example, EphA3 is expressed in pre-B cell lines (21); EphA2 (22), EphB4 (23), and EphB6 (2, 3) are expressed in the thymus. The inherent importance of these RTKs, due to their being receptors as well as tyrosine kinases, and due to their expression in lymphoid tissues, is self-evident.
The observed biological function of EphB6 in Jurkat T cells is very interesting. This is the first report that an Eph receptor is involved in inducing apoptosis. We suspected that the apoptosis involved the FasL/Fas pathway. However, the surface FasL expression on Jurkat cells was not detectable after stimulation by anti-EphB6 alone, anti-EphB6 in combination with anti-CD3, or even with an optimal concentration of anti-CD3 (data not shown). It has been reported that FasL can be rapidly shed from the cells surface, and soluble FasL can also induce apoptosis (24). Indeed, we could detect a rapid and significant increase of soluble FasL at 16 h in the supernatant of Jurkat cells stimulated by anti-EphB6 and anti-CD3, and this is correlated to the occurrence of apoptosis of the Jurkat cells as detected by annexin V expression at 24 h. The increase of the soluble FasL in the supernatant was not due to release of cellular FasL from dead cells, because at 16 h, few cells were PI positive (data not shown). The apoptosis could be blocked partially by soluble Fas-Fc, and this serves as an additional evidence that the apoptosis is Fas mediated. The partial block is probably due to the fact that the soluble FasL shed from the Jurkat cells functions as an autocrine or paracrine with high local concentrations, and that the Fas-Fc used was not sufficient to completely neutralize the soluble FasL. With that said, we cannot exclude the possibility that pathways other than the Fas-mediated one are so involved.
It is noteworthy that with strong cross-linking of TCR, or with strong activation by PMA plus ionomycin, Jurkat cells undergo apoptosis without the help of EphB6. In our model, anti-CD3 was used at a suboptimal concentration that was not sufficient to induce apoptosis alone without the anti-EphB6. Thus, we speculate that one biological function of EphB6 might be to enhance the signal strength of TCR cross-linking. If this is so, then an outcome of EphB6 ligation in normal T cells might be activation or apoptosis, depending on the maturation of T cells and the nature of TCR cross-linking. Indeed, we have found that costimulation of normal T cells with anti-CD3 and anti-EphB6 led to drastic changes in their proliferation and cytokine production (data not shown).
Munthe et al. (25) recently reported that ephrinB2 is a candidate ligand for EphB6 according to cell surface binding assays. In a separate study, we performed in situ hybridization on the expression patterns of mouse ephrinB2 and EphB6 and found that their expression was colocalized in white pulp of the spleen and cortex of the thymus (data not shown), unlike reciprocal expression patterns between other Eph kinases and their ligands in the CNS or vascular system. The colocalization suggests two things. First, EphB6 has relevant functions in lymphocytes, because its putative ligand has chances of interacting with it. Second, a third party might be required to trigger a biological function in immune cells, or else EphB6 will be perennially activated. Consistent with this prediction, we found that cross-linking of CD3 in addition to EphB6 in Jurkat cells was required to trigger GM-CSF production, inhibit proliferation, and induce apoptosis.
Initial reports on sequences and kinase activities of mouse and human EphB6 were perplexing. The two share a very high degree of homology (91% at the peptide level), suggesting that EphB6 is a well-conserved gene across species and must have important functions. Yet several mutations are found in the kinase domain of both human and mouse EphB6 and, consequently, they have no detectable kinase activities according to assays using recombinant proteins (2, 3). We have confirmed that a recombinant human intracellular domain of EphB6 had no kinase activity (data not shown). One might speculate that this RTK is a "dumb" receptor and serves to "damp" other Eph family members that share the same ligands. However, our study demonstrates that EphB6 is a functional receptor and is fully capable of inducing functions and transducing signals in Jurkat cells. Cross-linking of EphB6 results in transient tyrosine phosphorylation of several cellular proteins. Because EphB6 has no intrinsic kinase activity, the kinases responsible for signaling must be the one(s) associated directly or indirectly with EphB6. We found a trace amount of EphB6 in Fyn precipitates (data not shown), and conceivably some tyrosine kinase activities after EphB6 cross-linking might be derived from the associated Fyn. However, this does not exclude the possibilities that additional signaling molecules are also involved.
Indeed, we have demonstrated that EphB6 could be detected in Cbl precipitates. Conversely, Cbl was found in EphB6 precipitates (data not shown). Therefore, we have convincingly established that Cbl is associated with EphB6. Cbl proto-oncoproteins are primarily expressed in hemopoietic cells (26, 27) and is involved in the signaling of receptor protein tyrosine kinases as well as in the signaling of cell surface receptors that are associated with cytoplasmic protein tyrosine kinases (17). Therefore, the association of Cbl with EphB6 is in keeping with the general property of Cbl and is a novel signaling mechanism for Eph family kinases.
Cbl contains several protein-protein interaction domains such as a Src
homology 2 domain, a RING finger domain, a large proline-rich
Src homology 3 binding domain, and a leucine zipper. These are the
structural basis for its interaction with various signaling proteins,
including GRB2, the Crk adaptor family, Vav, p85 of PI-3K
(17), and EphB6, which also has a leucine zipper in its
carboxyl-terminal. In our study, EphB6 could be detected in GRB2, CrkL,
and CrkII precipitates, and abundant Cbl was coprecipitated with EphB6.
These findings raise a possibility that EphB6 forms multiunit complexes
with GRB2, CrkI, and/or CrkII using Cbl as an intermediate. This
possibility is under further investigation.
Although Cbl has been reported to associate with many other cell surface receptors, it seems that most phosphorylated Cbl molecules in Jurkat cells are involved in the EphB6 signaling pathway, because cross-linking of EphB6 leads to rapid tyrosine dephosphorylation of most Cbl molecules. If the phosphorylated Cbl are at the same time associated with other receptors, then we will arrive at a logical speculation that EphB6 can influence the signaling and functioning of those receptors.
Cross-linking of EphB6 also led to dissociation of SHP-1 from Cbl, which is a known substrate of SHP-1 (18). This finding suggests that EphB6, Cbl, and SHP-1 might form a trimolecule complex constitutively, and SHP-1 might dephosphorylate Cbl during the course of EphB6 activation and then dissociate itself from Cbl once the dephosphorylation is completed. Such a possibility is supported by our finding that all the three molecules could be detected in the membrane fraction of Jurkat cells (data not shown). We are currently investigating whether the EphB6, Cbl, and SHP-1 trimolecule complex does exist in Jurkat cells.
Cbl is also known as ubiquitin ligase (16), and its role as a negative regulator in many signaling pathways (28) might related to its function in channeling the kinase receptors it associates with to the proteasome degradation machinery. The association between EphB6 with Cbl certainly raises an interesting question whether EphB6 degradation is via the ubiquitin-proteasome pathway. When K562 cells were cultured in the presence of a proteasome inhibitor, dipeptide boronic acid (29), their surface expression of EphB6 as measured by 4F12 binding was significantly increased (data not shown), suggesting that EphB6 is degraded via the proteasome.
At present, it is not yet understood how the observed signaling events are related to the functional changes of Jurkat cells after EphB6 cross-linking. Because no prior knowledge is available, our study marks the beginning rather than the end of an exploration into the physiological roles and signaling of Eph kinases in the immune system.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Jiangping Wu, Laboratory of Transplantation Immunology, Pavilion DeSève, Research Center, Room Y-5616, Notre-Dame Hospital, Centre Hospitalier Universitaire de Montreal, 1560 Sherbrooke Street East, Montreal, Quebec H2L 4M1, Canada. E-mail address: jianping.wu{at}umontreal.ca
3 Abbreviations used in this paper: RTK, receptor tyrosine kinase; HPTK, human protein tyrosine kinase; IP, immunoprecipitation; SHP-1, Src homology 2 domain-containing tyrosine phosphatase-1; PI, propidium iodide; PI-3K, phosphatidylinositol 3-kinase; FasL, Fas ligand.
Received for publication November 20, 2000. Accepted for publication June 1, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B in streptococcal cell wall-induced polyarthritis. Proc. Natl. Acad. Sci. USA 95:15671.This article has been cited by other articles:
|
|
 |
|
  C. Bassi, S. Mello, R. Cardoso, P. Godoy, A. Fachin, C. Junta, P Sandrin-Garcia, C. Carlotti, R. Falcao, E. Donadi, et al. Transcriptional changes in U343 MG-a glioblastoma cell line exposed to ionizing radiation Human and Experimental Toxicology, December 1, 2008; 27(12): 919 - 929. [Abstract] [PDF]
|
|
|
|
 |
|
 S. H. Wimmer-Kleikamp, E. Nievergall, K. Gegenbauer, S. Adikari, M. Mansour, T. Yeadon, A. W. Boyd, N. R. Patani, and M. Lackmann Elevated protein tyrosine phosphatase activity provokes Eph/ephrin-facilitated adhesion of pre-B leukemia cells Blood, August 1, 2008; 112(3): 721 - 732. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Pfaff, U. Fiedler, and H. G. Augustin Emerging roles of the Angiopoietin-Tie and the ephrin-Eph systems as regulators of cell trafficking J. Leukoc. Biol., October 1, 2006; 80(4): 719 - 726. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Mao, X. Qiao, H. Luo, and J. Wu Transgenic Drak2 Overexpression in Mice Leads to Increased T Cell Apoptosis and Compromised Memory T Cell Development J. Biol. Chem., May 5, 2006; 281(18): 12587 - 12595. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Yu, J. Mao, Y. Wu, H. Luo, and J. Wu Ephrin-B1 Is Critical in T-cell Development J. Biol. Chem., April 14, 2006; 281(15): 10222 - 10229. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Matsuoka, H. Obama, M. L. Kelly, T. Matsui, and M. Nakamoto Biphasic Functions of the Kinase-defective Ephb6 Receptor in Cell Adhesion and Migration J. Biol. Chem., August 12, 2005; 280(32): 29355 - 29363. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Vearing, F.-T. Lee, S. Wimmer-Kleikamp, V. Spirkoska, C. To, C. Stylianou, M. Spanevello, M. Brechbiel, A. W. Boyd, A. M. Scott, et al. Concurrent Binding of Anti-EphA3 Antibody and Ephrin-A5 Amplifies EphA3 Signaling and Downstream Responses: Potential as EphA3-Specific Tumor-Targeting Reagents Cancer Res., August 1, 2005; 65(15): 6745 - 6754. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Miao, K. Strebhardt, E. B. Pasquale, T.-L. Shen, J.-L. Guan, and B. Wang Inhibition of Integrin-mediated Cell Adhesion but Not Directional Cell Migration Requires Catalytic Activity of EphB3 Receptor Tyrosine Kinase: ROLE OF RHO FAMILY SMALL GTPases J. Biol. Chem., January 14, 2005; 280(2): 923 - 932. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Yu, H. Luo, Y. Wu, and J. Wu EphrinB1 Is Essential in T-cell-T-cell Co-operation during T-cell Activation J. Biol. Chem., December 31, 2004; 279(53): 55531 - 55539. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Hafner, G. Schmitz, S. Meyer, F. Bataille, P. Hau, T. Langmann, W. Dietmaier, M. Landthaler, and T. Vogt Differential Gene Expression of Eph Receptors and Ephrins in Benign Human Tissues and Cancers Clin. Chem., March 1, 2004; 50(3): 490 - 499. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. G. Wohlfahrt, C. Karagiannidis, S. Kunzmann, M. M. Epstein, W. Kempf, K. Blaser, and C. B. Schmidt-Weber Ephrin-A1 Suppresses Th2 Cell Activation and Provides a Regulatory Link to Lung Epithelial Cells J. Immunol., January 15, 2004; 172(2): 843 - 850. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Yu, H. Luo, Y. Wu, and J. Wu Mouse EphrinB3 Augments T-cell Signaling and Responses to T-cell Receptor Ligation J. Biol. Chem., November 21, 2003; 278(47): 47209 - 47216. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. K. Murai and E. B. Pasquale `Eph'ective signaling: forward, reverse and crosstalk J. Cell Sci., July 15, 2003; 116(14): 2823 - 2832. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Yu, H. Luo, Y. Wu, and J. Wu Ephrin B2 Induces T Cell Costimulation J. Immunol., July 1, 2003; 171(1): 106 - 114. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Sharfe, A. Freywald, A. Toro, and C. M. Roifman Ephrin-A1 Induces c-Cbl Phosphorylation and EphA Receptor Down-Regulation in T Cells J. Immunol., June 15, 2003; 170(12): 6024 - 6032. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Freywald, N. Sharfe, C. Rashotte, T. Grunberger, and C. M. Roifman The EphB6 Receptor Inhibits JNK Activation in T Lymphocytes and Modulates T Cell Receptor-mediated Responses J. Biol. Chem., March 14, 2003; 278(12): 10150 - 10156. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
