ShblinksSLP-76andVavwiththeCD3complexinJurkatT cells
Cecilia K. Lindholm
1
, Maria L. Henriksson
2
, Bengt Hallberg
2
and Michael Welsh
1
1
Department of Medical Cell Biology, Biomedicum, Uppsala University, Sweden;
2
Department of Medical
Biosciences/Pathology, Umea
˚
University, Sweden
This study addresses the interactions between the adaptor
protein Shband components involved inT cell signalling,
including SLP-76, Gads, Vavand ZAP70. We show that
both SLP-76and ZAP70 co-immunoprecipitate withShb in
Jurkat Tcellsand that ShbandVav co-immunoprecipitate
when cotransfected in COS cells. We also demonstrate,
utilizing fusion protein constructs, that SLP-76, Gads and
Vav associate independently of each other to different
domains or regions, of Shb. Overexpression of an SH2
domain-defective Shb causes diminished phosphorylation of
SLP-76 andVavand consequently decreased activation of
c-Jun kinase upon T cell receptor (TCR) stimulation. Shb
was also found to localize to glycolipid-enriched membrane
microdomains (GEMs), also called lipid rafts, after TCR
stimulation. Our results indicate that upon TCR stimula-
tion, Shb is targeted to these lipid rafts where Shb aids in
recruiting the SLP-76–Gads–Vav complex to theT cell
receptor f-chain and ZAP70.
Keywords:Shb;Vav;SLP-76;Tcellreceptor;lipidrafts.
The early events inT cell signalling involve the activation of
Lck, which subsequently phosphorylates the TCR f-chain
and the e-chain of theCD3complex on ITAM motifs
(immunoreceptor tyrosine-based activation motifs). This
recruits the kinases ZAP70 or Syk that consequently induces
tyrosine phosphorylation of several intracellular substrates,
for example, phospholipase C-c1(PLC-c1), LAT [1], Vav
and SLP-76 [2]. Functional interactions between these
effector molecules are believed to be necessary for T cell
maturation, proliferation and differentiation.
One aspect of T cell receptor (TCR) activation relates to
specific areas (microdomains) of theT cell plasma mem-
brane, rich in glycosphingolipids, cholesterol and GPI-
anchored proteins, but poor in phospholipids (reviewed in
[3]). These glycolipid-enriched membrane microdomains
(GEMs), or lipid rafts, are rich in PTKs (protein tyrosine
kinases), monomeric and trimeric G proteins and several
other signaling proteins. In resting Tcells LAT, Lck, Fyn,
Syk, Cbl and Ras are located in rafts, whereas some other
proteins are recruited to the rafts upon T cell receptor
stimulation. The latter group includes SLP-76, Gads, Vav,
ZAP70, TCR f-chain, PLC-c1, Shc and PKC [4,5].
The adapter proteins SLP-76and Gads andthe guanine
nucleotide exchange factor Vav are critical for appropriate
T cell activation. SLP-76andVav are both phosphorylated
upon CD3 stimulation inJurkatT cells. SLP-76 is
phosphorylated by ZAP70 [6] at multiple residues
(Tyr113, Tyr128 and Tyr145) of which Tyr113 and
Tyr128 are responsible for allowing the binding of the
Vav SH2 domain [7]. SLP-76 has also been shown to
associate with Lck, LAT, Grb2 and SLAP-130 inJurkat T
cells. Vav is reported to be phosphorylated by Lck on
Tyr174 and this process activates the exchange activity of
Vav [8,9]. However, other studies have suggested tyrosine-
174 on Vav as a putative candidate for phosphorylation by
Syk and Zap70 [10]. Vav is a guanine nucleotide exchange
factor for the GTPases Rac1 and Cdc42, which cause,
among other things, activation of the c-Jun N-terminal
kinases (JNK). Gads (Grb2-related adaptor downstream of
Shc) is not phosphorylated in response to TCR stimulation,
but it has been shown to associate with several proteins
upon T cell activation, including Shc, SLP-76and LAT
[11,12].
The Shb adapter protein was cloned from a b-cell library
in 1994 [13], but it has since been found to be ubiquitously
expressed [14]. Shb is involved in FGFR-1 and PDGFR
signaling, and also associates with several signaling proteins,
including CrkII, Eps8, Grap and Src [14,15]. Shb is also of
importance for the TCR-dependent immune response in
Jurkat cells. TCR-mediated activation of NFAT was totally
abolished inJurkatcells expressing a mutant form of Shb
with a defective SH2 domain, and endogenous IL-2
production was also decreased in these cells [16]. Overex-
pression of wild-type Shb, however, had no effects on the
CD3 mediated responses inJurkatcells [14,16]. In Jurkat
cells, Shb has previously been found to exhibit domain-
dependent interactions with Grb2, LAT, PLC-c1andthe
f-chain of theT cell receptor [14–16]. The SH2 domain of
Shb bound the f-chain, the PTB domain bound LAT and
the proline-rich regions associate with PLC-c1 and Grb2.
There are also four putative tyrosine phosphorylation sites
in Shb that show extensive homology with similar sequences
in other adapter proteins [17], and conform with the
consensus sequence Y-X-D/E/T/Q-P-F/Y/W-D/E.
Correspondence to C. Lindholm, Department of Medical Cell Biology,
Box 571, Biomedicum, 75123, Uppsala, Sweden.
Fax: + 46 18 556401, Tel.: + 46 18 4714033,
E-mail: Cecilia.Lindholm@medcellbiol.uu.se
Abbreviations: TCR, T cell receptor; SH2, Src homology 2; PTB,
phosphotyrosine binding; ITAM, immunoreceptor tyrosine-based
activation motif; PTK, protein tyrosine kinase; ECL, enhanced
chemiluminescence system; JNK, c-Jun N-terminal kinase; GST,
glutathione-S-transferase; PDGF, platelet-derived growth factor;
FGF, fibroblast growth factor; P-Y, phosphotyrosine; MAPK,
mitogen-activated protein kinase; GEMs, glycolipid-enriched
membrane microdomains; LAT, Linker for Activation of T cells.
(Received 13 February 2002, revised 16 May 2002,
accepted 21 May 2002)
Eur. J. Biochem. 269, 3279–3288 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03008.x
In this study, we describe interactions between Shb and
SLP-76, Gads, Vavand ZAP70 and how a functional Shb
SH2 domain is important for phosphorylation of SLP76
and Vav, possibly by ZAP70. TheShb SH2 domain is also
of importance for the activation of the JNK, downstream of
Vav intheT cell signalling cascade. We have also observed
the recruitment of Shb to lipid rafts upon TCR stimulation
in JurkatT cells.
MATERIALS AND METHODS
Antibodies
Anti-phosphotyrosine Ig (4G10) and anti-Vav Ig were
purchased from Upstate Biotechnology (Lake Placid, NY,
USA). CD3 antibody (CD3 pure) was from Becton-
Dickinson (San Jose, CA). Anti-JNK Ig and anti-(phos-
pho-JNK) Ig were from New England BioLabs (Beverly,
MA). Anti-HA antibody was from Santa Cruz. Anti-
ZAP70Ig,anti-SLP-76Igandanti-Rac1Igwerefrom
Transduction Laboratories (Lexington, KY). Rabbit poly-
clonal anti-Gads Ig was a gift from Jane McGlade, Hospital
for Sick Children Research Institute, Toronto, Canada.
Affinity purified anti-Shb Ig has been described previously
[18].
DNA constructs
The Shb–SH2–GST fusion protein plasmid has been
described previously [13]. The Shb–PTB-Pro–GST plasmid
was described previously under the name p55 ShbDSH2 [14]
and produces a fusion protein corresponding to the PTB
domain and two proline-rich sequences. The p55 Shb–pET
plasmid has been described previously [14] and produces the
full-length p55 Shbwith an additional His-tag for purifica-
tion on Ni-beads according to the manufacturer (Novagen).
The Shb–PTB–GST plasmid was constructed using the
primers 5¢-GGGATCCTTCCAGGACCCCTAC-3¢ and
5¢-AGAATTCAGGGCTCCCATGTTT-3¢ corresponding
to theShb cDNA nucleotides 840–1740. The amplified
fragment was digested using EcoRI, and ligated into EcoRI
digested pGEX-2T vector. The SLP-76-Pro–GST plasmid
was constructed using the primers 5¢-CGAGGGATCCCT
GCAGAACTCCATCCTGCCTG-3¢ and 5¢-CATTTAAT
GAATTCTCTTCCTCCGC-3¢ corresponding to SLP-76
nucleotides 466–1245. The amplified fragment was cut using
BamH1 and EcoR1 and ligated into pGEX-2TK. The
SLP-76-SH2-GST plasmid was constructed using the
primers 5¢-GGAAGGATCCAATTCATTAAATGAAGA
GTG-3¢ and 5¢-GGCTATAACGAATTCTGGGTACCC
TGCAGCATG-3¢. This fragment was also cut with BamH1
and EcoR1 and ligated into pGEX-2TK. The PAK-CD-
GST (PAK Crib domain) plasmid has previously been
described previously [19].
The Shb wild-type plasmid contains theShb cDNA
inserted in pcDNA1 vector, and has been described
previously. TheShb R522K plasmid has also been described
previously [14]. Briefly, the arginine at position 522 was
converted to a lysine, in full-length Shb cDNA, and inserted
in pcDNA1 expression vector. TheShb DPTB-Tyr plasmid
was constructed by the deletion of the nucleotides 880–1603
in theShb cDNA (containing the four tyrosine phosphory-
lation sites andthe PTB domain), using the restriction
enzymes FspIandBstEII. The hemeagglutinin-tagged
SLP-76 vectors were constructed as follows. The SLP-76
fragment (1–534), the SLP-76-Tyr fragment (1–155) and the
SLP-76-SH2 fragment (415–534) were cut out from the
above mentioned pGEX-2TK vectors using BamH1 and
EcoR1 and ligated into pSG5A vector, containing an HA
tag. TheVav expression plasmid was a kind gift from A.
Weiss, San Francisco, CA, USA.
Transient transfections
COS cells were maintained in Dulbecco’s modified
Eagle’s medium (DMEM) supplemented with 10% FcII
and antibiotics. Thecells were washed three times in
serum-free DMEM and transfected with 2 lgofeach
plasmid or emply vector, as indicated inthe figures,
using Lipofectamine as recommended by the manufac-
turer. JurkatTcells were washed twice in NaCl/P
i
and
transfected by electroporation (380 V, 950 lFina1-cm
cuvette) with 15 lg of each plasmid as indicated in
figure 3.
Binding experiments and immunoprecipitations
Jurkat cells or Jurkatcells overexpressing Shb R522K [16],
were maintained in RPMI medium supplemented with 10%
FcII (Hyclone) and antibiotics. COS cells were maintained
in DMEM medium supplemented with 10% FcII. Jurkat
cells were collected by centrifugation and suspended in
RPMI 1640 medium lacking serum before stimulation with
the CD3 antibody at 37 °C for 2 min. COS cells were either
unstimulated or stimulated with pervanadate for 20 min at
37 °C. Jurkatcells were pelleted by centrifugation and COS
cells were scraped together using a rubber policeman. The
cells were lysed in Triton lysis buffer (0.15
M
NaCl, 0.05
M
Tris pH 7.5, 0.5% Triton X-100, 1 m
M
NaF, 0.1 m
M
orthovanadate, 100 UÆmL
)1
trasylol, 2 m
M
phenylmethane-
sulfonyl fluoride) for 10 min. Nuclei were pelleted by
centrifugation and cell extracts were incubated with the
immobilized fusion proteins on glutatione Sepharose beads
(Amersham-Pharmacia Biotech, Uppsala, Sweden) or
Ni-beads (Novagen) for 30 min on ice and then washed
three times with NaCl/P
i
/1% Triton. Alternatively, the cell
extracts were preincubated with protein A Sepharose for
15 min (Jurkat cell extracts only) and then immunoprecipi-
tated for 1 h with antibodies against Shb, SLP-76, Vav,
ZAP70, HA or preimmune sera (IgG). Immunoprecipitates
were pelleted using either protein A or protein G Sepharose
(50 lL) and then washed three times with NaCl/P
i
/1%
Triton. The fusion protein complexes, cell extracts or the
immuno-complexes were then resolved on SDS/PAGE and
subjected to Western blotting onto Immobilon filters
(Millipore) in 20% methanol, 190 m
M
glycine, 23 m
M
Tris
and 0.02% SDS. The blots were subsequently incubated
with blocking solution (5% BSA in NaCl/P
i
/0.5% Tween)
and primary antibody as indicated. Immuno-reactivity
was detected using horseradish peroxidase-conjugated
secondary antibodies and ECL (Amersham–Pharmacia
Biotech, Uppsala, Sweden) according to the manufacturer’s
instructions. For detection of Rac1 activity, the amount of
Rac1 bound to the PAK-CD fusion protein was determined
using Rac1 antisera, and normalized to total Rac1, in cell
lysate.
3280 C. K. Lindholm et al. (Eur. J. Biochem. 269) Ó FEBS 2002
Peptide synthesis and peptide inhibition experiments
Peptides corresponding to tyrosine-phosphorylation sites in
SLP-76 andVav were synthesized. The peptides used in this
paper include: SLP-Y113: WSSFEEDDYESPND, SLP-
Y128: QDGEDDGDYESPNE, SLP-Y145: APVEDDA
DYEPPPS, Vav-Y174: NEEAEGDEIYEDLMRL and
Vav-Y692: ILANRSDGTYLVRQRV, where Y indicates
phosphotyrosine. For peptide inhibition experiments, cell
lysates prepared as above were incubated with Shb–PTB-
Pro–GST, Shb–PTB–GST or p55 Shb–pET fusion protein in
the presence of 100 l
M
phosphorylated peptide or phospho-
tyrosine (10 m
M
)at4 °C for 30 min. After three washes with
NaCl/P
i
/1% Triton, the samples were separated by SDS/
PAGE and transferred onto Immobilon filter as described
above. Anti-SLP76 Ig and anti-Vav Ig were used to detect the
amount of these proteins bound to theShb fusion protein.
Isolation of GEM fractions
Jurkat cells (5 · 10
7
) where either left unstimulated or
stimulated withCD3 antibody at 37 °C for 3 min, followed
by lysis at 4 °C for 30 min in 1 mL of Mes-buffered saline
(25 m
M
Mes, pH 6.5, 150 m
M
NaCl, 5 m
M
EDTA, 1%
Triton X-100, 0.05 m
M
orthovanadate, 100 U mL
)1
trasy-
lol, 10 m
M
NaF, 1 m
M
Pefablock). The lysates where then
mixed with 1 mL 80% sucrose in Mes-buffered saline and
transferred to ultracentrifuge tubes. The samples were
overlaid with 2 mL of 30% sucrose in Mes-buffered saline,
followed by 5% sucrose in Mes-buffered saline. The Triton-
insoluble fractions were separated from the cell lysates by
ultracentrifugation for 22 h at 250 000 g in a Beckman
SW50.1 rotor at 4 °C. Fractions (400 lL) were removed
sequentially starting from the top of the gradient. The
proteins in each fraction were precipitated using 20%
trichloroacetic acid. The precipitates were washed with
acetone, dried and resuspended in 40 lLofSDS-sample
buffer. The fractions (fraction 1–6 contain the lipid rafts)
were then resolved on SDS/PAGE and subjected to Western
blotting as described above. The blots were subsequently
incubated with blocking solution (5% BSA in NaCl/P
i
/
0.5% Tween) and primary antibody as indicated. Immuno-
reactivity was detected as described above.
RESULTS
Expression of Shbwith an R522K mutation inthe SH2
domain affects the co-immunoprecipitation of several
tyrosine phosphorylated proteins
To further investigate the protein-interactions of the Shb
adapter protein, we attempted to identify additional binding
partners for ShbinJurkatT cells. Shb immunoprecipitation
in JurkatTcells revealed several coimmunoprecipitated
proteins (Fig. 1). We detected phosphoproteins correspond-
ing to the previously identified proteins PLC-c1 at 160 kDa,
Linker for Activation of Tcells (LAT) at 35 kDa and the
TCR f-chain at 22 kDa, but also proteins of 100, 75 and
70 kDa. The association between Shband PLC-c1was
previously shown to be independent of TCR activation,
whereas the interactions between LAT or the f-chain and
Shb were found to be dependent on TCR activation [14,16].
We have also described the R522K mutation inthe SH2
domain of Shb [14,16], and how expression of this Shb
mutant decreases the tyrosine phosphorylation of PLC-c1,
LAT andthe TCR f-chain. Jurkat R522K-2 cell lysates
displayed reduced tyrosine phosphorylation of proteins
migrating as 160, 100, 75 and 35 kDa (Fig. 1) upon CD3
stimulation. Shb immunoprecipitation of the R522K cell
lysates revealed decreased phosphotyrosine content of the
bands corresponding to PLC-c1 (160 kDa), LAT (36 kDa)
and the f-chain (22 kDa). In addition, theShb immuno-
precipitations exhibited reduced tyrosine phosphorylation
of the 75 and 100 kDa proteins, intheJurkat R522K-2 cells,
unlike the 70-kDa product, which was not, or only slightly,
affected. Probing the blot with an antibody reactive with
SLP-76 revealed a weak band at 75 kDa that was detected
in the CD3-stimulated control cells. The R522K-2 cells
exhibited the presence of SLP-76 regardless of whether the
cells were stimulated withCD3 or not. The reduced
phosphorylation of the 160-, 100-, 75-, 35- and 22-kDa
proteins inJurkat cell lysates has previously been verified in
another clone overexpressing R522K Shband after tran-
sient transfection of the R522K Shb cDNA [14,16].
Fig. 1. Shb coimmunoprecipitates with several tyrosine phosphorylated
proteins. Jurkat-neo or Jurkat R522K-2 cells (10
7
) were left unstimu-
lated (–) or were stimulated with anti-CD3 Ig for 2 min in 37 °C(+).
The cells were then lysed in triton lysis buffer and cell extracts were
precleared with protein A Sepharose and then subjected to a-Shb
immunoprecipitation. The precipitates and cell extracts were resolved
on SDS/PAGE and blotted for SLP-76, Shband phosphotyrosine
using the 4G10 antibody. The positions of SLP-76, Shb, LAT and
molecular mass markers are indicated.
Ó FEBS 2002 Association of Shb, SLP-76andVavinJurkatcells (Eur. J. Biochem. 269) 3281
Shb associates with SLP-76, ZAP70 and Vav
in CD3 stimulated JurkatT cells
To further elucidate the identity of the 75-kDa protein seen
after a-Shb immunoprecipitation, another blot was probed
with anti-(SLP-76) Ig. SLP-76 was present in a-Shb
immunoprecipitates after CD3 stimulation, but not after
immunoprecipitation with preimmune serum (Fig. 2A).
Likewise, the 70-kDa protein that coimmunoprecipitated
with Shb was identified as ZAP70 (Fig. 2A).
To assess the mode of interactions between Shband SLP-
76 or ZAP70, GST fusion proteins comprising p55
ShbDSH2 (Shb–PTB-Pro–GST) [14], Shb–PTB–GST,
Shb–SH2–GST [13] or GST (as control) were incubated
with cell extracts from CD3 stimulated or unstimulated
Jurkat cells (Fig. 2B). We also tested if phosphotyrosine
could inhibit the binding between fusion protein and cell
protein. ZAP70 was found to associate with both the Shb–
PTB-Pro-GST fusion protein andthe Shb–SH2–GST
fusion protein, and these associations were slightly increased
after CD3 stimulation. The interaction between the SH2
domain and ZAP70 was completely blocked with phospho-
tyrosine, whereas binding of ZAP70 to the Shb–PTB-Pro–
GST fusion protein was only partially inhibited under these
conditions. The blot was also stained with amidoblack to
verify that equal amounts of Shb–PTB-Pro–GST fusion
protein were used.
SLP-76 was found to associate withthe Shb–PTB-Pro–
GST fusion protein, containing both the PTB domain and
two of the N-terminal proline-rich domains. This interac-
tion was only slightly inhibited by phosphotyrosine. To test
if theShb PTB domain could be involved in this interaction
we searched the protein-sequence of SLP-76 for D-D-X-Y,
which is the consensus sequence for theShb PTB domain
Fig. 2. Association of ShbwithSLP-76and ZAP70 inCD3 stimulated Tcellsand pervanadate stimulated COS cells. Jurkatcells (10
7
)were
unstimulated (–) or stimulated by CD3 cross-linking for 2 min in 37 °C (+) before lysis in triton lysis buffer. (A) Cell extracts were subjected to
immunoprecipitation using normal rabbit serum (IgG) or anti-Shb antibody (Shb). (B) Cell extracts were incubated for 30 min on ice with the
indicated immobilized fusion proteins inthe presence (+) or absence (–) of 10 m
M
phosphotyrosine (P-Y). (C) Cell extracts were incubated for
30 min on ice with immobilized p55 Shb–pET fusion protein inthe absence or presence of tyrosine-phosphorylated peptides or phosphotyrosine
(P-Y). The peptides correspond to tyrosines 113, 128 and 145 inSLP-76and tyrosines 174 and 692 in Vav. SLP-3Y contains all three SLP-peptides.
(D) Cell extracts were incubated for 30 min on ice with immobilized SLP-76-Pro-GST or SLP-76-SH2-GST fusion proteins. (E) COS cells,
transfected withShband different HA-tagged SLP-76 constructs or vector, as indicated, were stimulated, or not, with pervanadate, for 15 min in
37 °C, before lysis. The lysates were subjected to immunoprecipitations using anti-HA Ig. (F) Jurkat cell extracts were incubated with immobilized
p55 Shb–pET fusion protein inthe absence (–) or presence (+) of phosphotyrosine (P-Y). The samples in (A–F) were resolved on SDS/PAGE. The
blots were probed withthe antibodies indicated andthe blot in (B) was stained using amido-black. The positions of SLP-76, ZAP70, Gads and Shb
are indicated with arrows.
3282 C. K. Lindholm et al. (Eur. J. Biochem. 269) Ó FEBS 2002
binding site [14]. Three such sequences were found, all
previously reported to be involved in cell signalling [7,20].
Three synthetic phosphopeptides were made corresponding
to these three sites (SLP-Y113, SLP-Y128 and SLP-Y145).
However, no inhibition of SLP-76 binding to theShb fusion
protein could be seen using these peptides (Fig. 2C). In
addition, theShb PTB domain fusion protein does not
allow binding of SLP-76 (results not shown). The two LAT-
peptides previously reported to inhibit the LAT–Shb
association [16] were also tested, and had no inhibitory
effect on SLP-76, ZAP70 or Vav association to Shb (results
not shown).
To characterize the Shb–SLP-76 interaction, we utilized
fusion proteins corresponding to the SH2 domain of
SLP76 andthe proline-rich regions of SLP-76. We found
that Shb associates withthe SH2 domain of SLP-76, but
not to the proline-rich regions of SLP-76 (Fig. 2D), and
that this association was increased by CD3-stimulation. To
test this further, we overexpressed HA-tagged SLP-76 or
different domains of SLP-76, together with wild type Shb
in COS cells, and performed HA-immunoprecipitations.
Figure 2E shows that Shb is associated with wild-type
SLP-76 andthe SH2 domain of SLP-76 upon pervanadate
stimulation (to increase the degree of protein tyrosine
phosphorylation) in COS cells. We then hypothesized that
the interaction between ShbandSLP-76 might be of a
trimeric nature, where the SH2 domain of SLP-76 asso-
ciates with a phosphotyrosine motif in Shb, andthe proline
rich sequences of Shb interact with some other adaptor,
which also has the ability to bind SLP-76 (Fig. 8). After
investigating possible SLP-76-interacting adaptor proteins,
we considered the Grb2-related adaptor Gads a candidate.
Gads was found to associate withthe p55 Shb fusion
protein, in a phosphotyrosine-independent manner
(Fig. 2F), which was very similar to our findings concern-
ing the association of SLP-76 to the Shb–PTB-Pro–GST
fusion protein. The same blot was also probed with anti-
(SLP-76) Ig, andthe pattern of association was found to be
quite similar to that of Gads. The significance of the
SLP-76–Shb interaction was further verified by transient
transfections of Jurkatcellswith different Shb mutants, to
study the pattern of tyrosine phosphorylation. Cells
cotransfected withSLP-76and wild-type Shb, exhibited a
normal patern of tyrosine phosphorylation in response to
CD3 stimulation (Fig. 3). However, when SLP-76 was
cotransfected withShb R522K (with a nonfunctional SH2
domain) or Shb DPTB-Tyr (with the four tyrosine
phosphorylation sites andthe PTB domain deleted), the
cells exhibited a decreased response to CD3 stimulation
(Fig. 3), which was verified using densitometric scannings.
Particularly, decreased tyrosine phosphorylation of pro-
teins corresponding to 160 (PLC-c1), 75 and 36 kDa
(LAT) was observed and described in Fig. 3, with the
relative phosphorylation of these proteins compared to the
total amount of SLP-76intheCD3 treated lanes. The
75-kDa protein comigrated exactly with SLP-76. These
results indicate that both the SH2 domain andthe PTB-
Tyr region of Shb, are vital for the phosphorylation of
several proteins upon TCR engagement, including SLP-76,
PLC-c1 and LAT.
The guanine nucleotide exchange factor Vav was
also found to associate with Shb, utilizing fusion pro-
teins (Fig. 4A). The PTB domain of Shb mediated this
interaction andthe binding could be blocked out with
phosphotyrosine (Fig. 4A,B). To identify a putative binding
site for ShbintheVav protein, we searched the Vav
sequence for the PTB domain consensus binding-site D-D-
X-Y. Two such motifs were found inVavand synthetic
peptides corresponding to them were made. These peptides
were used in binding experiments with a GST fusion protein
containing only theShb PTB domain. The PTB domain of
Shb was found to associate withVavand this binding is
blocked by both phosphotyrosine andthe more specific
synthetic peptide corresponding to tyrosine 174 in Vav
(Vav-Y174) (Fig. 4B), which is more known as the
autoregulatory site of Vav [21]. We also tried to block this
binding withtheSLP-76 peptides, as Y113 and Y128 in
SLP76 have been reported to be responsible for Vav binding
to SLP76 [20], but observed no effect.
Attempts to coimmunoprecipitate VavwithShbin Jurkat
cells were unsuccessful, due to an IgG-related band of the
same size as Vav, which made detection of Vav impossible.
An alternative approach was to express VavandShb in
COS cells, and then perform a-Shb immunoprecipitations.
Figure 4C shows how Vav coimmunoprecipitates with Shb
upon pervanadate treatment of COS cells transfected with
Shb and Vav.
Our results suggest that although VavandSLP-76 are
known to interact with each other, they both bind directly
and independently to Shb.
Phosphorylation of SLP-76andVav is dependent
on Shbwith a functional SH2 domain
To further study the effects of an R522K mutation in the
SH2 domain of Shb on SLP-76andVav phosphorylation,
we performed immunoprecipitations of SLP-76andVav in
Fig. 3. Shb is vital for the phosphorylation of several proteins in Jurkat
T cells. Jurkat cells, transfected withSLP-76and different Shb con-
structs, as indicated, were stimulated (+) or not (–) withCD3 antibody
for2minin37°C, and lysed. The cell extracts were resolved on SDS/
PAGE andthe blots were probed with anti-(SLP-76) Ig and 4G10 anti-
phosphotyrosine Ig. Relative phosphorylation, as assessed by densio-
metric scanning of the proteins indicated, is given inthe table below the
blot. The values are given relative the total amount of SLP-76.
Ó FEBS 2002 Association of Shb, SLP-76andVavinJurkatcells (Eur. J. Biochem. 269) 3283
Jurkat-R522K-2 and Jurkat-neo cells (Fig. 5A,B). Western
blot analyses of these immunoprecipitates revealed in the
Shb SH2 defective clone, Jurkat-R522K-2, a failure of CD3
stimulation to increase the phosphorylation of either
SLP-76 or Vav. The basal phosphorylation of Vav was
also lower inthe clone expressing the R522K mutant. The
same experiment was performed on ZAP70 in these cells,
with a minor difference in stimulation between mutant and
control (Fig. 5C). The degree of phosphorylation, relative
unstimulated Jurkat cells, are given below each blot. These
findings are consistent withthe view that ZAP70 operates
upstream of Shb, whereas SLP-76andVav are effectors
downstream of Shbinthe signalling pathway following
TCR engagement.
Effects of expression of R522K ShbinJurkatT cells
on JNK and Rac1 activation
The guanine nucleotide exchange factor Vav is known to
activate Rac1, RhoA and Cdc42. Both Rac1 and Cdc42 are
among other things known to cause activation of JNK. As
Vav phosphorylation is decreased inthecells expressing Shb
with a defective SH2 domain (Jurkat R522K-2), we decided
to also examine Rac1 and JNK activation upon CD3
stimulation in these cells. As displayed in Fig. 6A, Rac1
activity (GTP-Rac1), as assessed by association with the
PAK-CD fusion protein, is decreased intheJurkat R522K-
2 cells upon CD3 stimulation, compared to normal Jurkat
cells. Consequently, JNK activation is also abolished upon
CD3 stimulation intheJurkat R522K-2 cells compared to
the Jurkat-neo cells (Fig. 6B).
Shb is recruited to GEMs after TCR ligation
It has recently been shown that several proteins involved in
TCR signal transduction, including SLP-76, Vav, ZAP70
and Gads, localize to GEMs upon TCR stimulation [22–26].
To assess if this was also the case for Shb, JurkatTcells were
left unstimulated or stimulated with anti-CD3 Ig and then
lysed in a Triton X-100-based buffer. Lysates were subjected
to sucrose density gradient ultracentrifugation to separate
the detergent resistant GEMs from the Triton-soluble
Fig. 5. Phosphorylation of SLP-76andVav is dependent on the Shb
SH2 domain. Jurkat-neo or Jurkat R522K-2 cells (10
7
) were left
unstimulated (–) or were stimulated with anti-CD3 Ig for 2 min in
37 °C (+). Thecells were then lysed in triton lysis buffer and cell
extracts were precleared with protein A sepharose and then subjected
to immunoprecipitation with (A) anti-SLP-76 Ig (B) anti-Vav Ig or (C)
anti-ZAP70 Ig. The precipitated proteins were resolved on SDS/PAGE
and blotted for phosphotyrosine (4G10), SLP-76, Vavand ZAP70.
The positions of SLP-76, Vavand ZAP-70 are indicated. The relative
phosphorylation, after normalization for the total amount of protein,
compared to unstimulated Jurkat cells, is also indicated under each
blot.
Fig. 4. Association of ShbandVavinCD3 stimulated Tcells and
pervanadate stimulated COS cells. (A,B) Jurkatcells (10
7
)were
unstimulated (–) or stimulated by CD3 cross-linking for 2 min in 37 °C
(+) before lysis in triton lysis buffer. (A) Cell extracts were incubated
for 30 min on ice withthe indicated immobilized fusion proteins in the
presence (+) or absence (–) of 10 m
M
phosphotyrosine (P-Y). (B)
Jurkat cell extracts were incubated for 30 min on ice with immobilized
Shb–PTB–GST fusion protein inthe absence or presence of the indi-
cated tyrosine-phosphorylated peptides or phosphotyrosine (P-Y). (C)
COS cells, transfected withShband Vav, as indicated, were stimulated,
or not, with pervanadate, for 15 min in 37 °C, before lysis. The lysates
were subjected to immunoprecipitations using Shb antibody. The
samplesin(A–C)werewashedandresolvedonSDS/PAGEand
immunoblotted withthe antibodies indicated. The positions of Vav
and Shb are indicated with arrows.
3284 C. K. Lindholm et al. (Eur. J. Biochem. 269) Ó FEBS 2002
fractions. Both Shb isoforms (55 and 66 kDa) were found to
localize to the GEM fraction (mainly fractions 3 and 4) after
CD3 stimulation (Fig. 7A). The GEM fraction was defined
by the presence of LAT, which is constitutively located to
lipid rafts. LAT might therefore be responsible for the
recruitment of Shb to the lipid rafts upon TCR engagment
(Fig. 7B,C). SLP-76, Vavand Gads were also seen to
localize to fraction 3, 4 and 5 upon CD3 stimulation
(Fig. 7D–F), which is in agreement with previous studies
[23–26].
DISCUSSION
We have previously established a role for ShbinT cell
signalling by describing the domain-specific binding of Shb
to the TCR f-chain,LAT,PLC-c1 and Grb2 [14] [16]. We
have also demonstrated that a mutation inthe SH2 domain
of Shb affects TCR signalling through MAPK and Ca
2+
causing abolished activation of the NFAT element in the
IL-2 promoter [16]. In this report, we show interactions
between Shband SLP-76, Gads, Vavand ZAP70 and that
Shb is recruited to GEMs upon TCR ligation. We also
examine how some of these interactions are affected by a
mutation inthe SH2 domain of Shb, and consequently how
the activation of JNK is affected.
To further study the effects of the mutation inthe Shb
SH2 domain, we examined the proteins found to coprecipi-
tate withShbinJurkat cells. Except for the previously
studied LAT, PLC-c1andtheTCRf-chain, we also noted
the presence of other tyrosine phosphorylated proteins
possibly corresponding to SLP-76, Vavand ZAP70. In the
cells expressing a nonfunctional SH2 domain, the phos-
phorylation of SLP-76andVav is decreased. This indicates
that Shb forms a signalling complexwith LAT, SLP-76,
Vav and PLC-c1, and that this complex is linked to the
TCR via the f-chain, through association of theShb SH2
domain. Phosphorylation of ZAP70 is not affected, or
affected very little, by the R522K mutation intheShb SH2
domain. This is consistent with our model (Fig. 8), where
ZAP70 is bound to the ITAMs of the f-chain, to which Shb
with its associated proteins also are attached, and thus
brought in proximity to the tyrosine kinase ZAP70.
Introduction of theShb SH2 point mutation decreases the
ability of the Shb–LAT–SLP76–Vav complex to interact
with ZAP70. In addition, expression of Shbwith a deletion
of the PTB domain andthe tyrosine phosphorylation sites,
also reduces the phosphorylation of phospho-proteins
corresponding to PLC-c1, SLP-76and LAT, implicating
the involvement of these regions of Shbin TCR signaling.
We have previously investigated the effects of overex-
pression of wild-type ShbinJurkat cells, and seen no
differences inthe CD3-mediated tyrosine phosphorylation,
when Shb is overexpressed compared to mock transfected
Jurkat cells [14]. This is further confirmed in Fig. 3, where
cotransfection of ShbandSLP-76 produces Jurkatcells with
a normal CD3-induced tyrosine phosphorylation pattern.
To elucidate how SLP-76, Vavand ZAP70 associate with
Shb, we utilized fusion proteins comprising different parts of
Shb. An association of the adapter protein SLP-76 to the
p55 Shb–pET andtheShb PTB-Proline-rich fusion protein
could be seen both in unstimulated cellsand after CD3
crosslinking. As the phosphopeptides did not efficiently
displace this association and as SLP-76 did not bind directly
to theShb PTB-domain fusion protein only, we considered
it plausible that theShb proline-rich motifs associated with
SLP-76 via some adapter protein, as SLP-76 lacks an SH3
domain. However, this interaction does not explain the
CD3-dependency of the SLP-76–Shb interaction; we there-
fore performed binding experiments using fusion proteins
corresponding to different domains of SLP-76. We also
cotransfected Shb together with different SLP-76 mutants in
COS cells. Both these approaches showed that the SLP-76
SH2 domain associated withShb after CD3 stimulation in
Jurkat cells, or pervanadate treatment in COS cells. We
have previously reported tyrosine phosphorylation of Shb
after CD3 stimulation [16], and therefore suggest that
phosphorylated tyrosines inShb are binding sites for the
SLP-76 SH2 domain. Gads is a potential adapter that links
the Shb proline-rich motifs withSLP-76 as the latter
associates with Gads via an SH3-domain/proline–rich motif
interaction. The strong association of Gads to the Shb
fusion protein, andthe previously reported association
between Shband other proteins of the Gads family, Grb2
and Grap, all suggest that an association between Shb and
Gads exists. Accordingly, Gads bridges SLP-76and Shb,
Fig. 6. Activation of Rac1 and JNK is dependent on a functional Shb
SH2 domain. Jurkat-neo or Jurkat R522K-2 cells (10
7
)were
unstimulated (–) or stimulated by CD3 crosslinking for 2 min (+)
before lysis in triton lysis buffer. (A) Jurkat cell extracts were incubated
for 30 min on ice with immobilized PAK–CD–GST fusion protein.
Fusion-protein complexes and cell extracts were resolved on SDS/
PAGE, andthe blot was probed with Rac1 antisera. (B) Cell extracts
were resolved on SDS/PAGE andthe blot was probed with anti-
(phospho-JNK) Ig, and after stripping, reprobed with anti-JNK Ig.
The positions of Rac1, phospho-JNK and total JNK are indicated in
the figures. The relative amount of active Rac1, after normalization for
the total amount of Rac1, is also indicated under the blot in (A).
Ó FEBS 2002 Association of Shb, SLP-76andVavinJurkatcells (Eur. J. Biochem. 269) 3285
and theSLP-76 SH2 domain binds to tyrosine phosphor-
ylated Shb.
The guanine nucleotide exchange factor Vav associates
directly withthe PTB domain of Shb through tyrosine-174.
Tyrosine 174 has previously been shown to be of import-
ance for Vav GEF function [9] and is a possible site for
negative regulation of Vav. The binding of Vav to the Shb
PTB domain might stabilize its association to SLP-76, and
thus aids inthe activation process of Vav.
Our data also show abolished CD3-mediated Rac1
activation and JNK phosphorylation, as a consequence of
the mutation intheShb SH2 domain (R522K). This is
probably due to the abolished activation of Vavin these
cells, as several reports have shown that activated Vav is
required for Rac1-mediated JNK activation in lymphoid
cells [27,28].
Recent attention has focused on the concentration of
effector molecules into subdomains, so-called GEMs or
lipid rafts, upon receptor stimulation. These membrane
subdomains are characterized by their detergent insolubility.
Some proteins involved inT cell signalling are always
present in GEMs, for example LAT, which is anchored to
the membrane via its palmitoylated tail. We presently show
that Shb is recruited to the GEMs upon TCR stimulation,
where SLP-76, Vav, ZAP70, Gads and Grb2 have all
previously been shown to target after CD3 ligation
[23,24,26,29].
In conclusion, our results suggest that Shb is important
for the early events of T cell signalling. Shb is recruited to
GEMs, possibly by LAT, after antigen stimulation, and can
also associate with several important signalling proteins
such as PLC-c1, Vav, Gads, Grb2, GRAP and SLP-76.
This generates a signalling complex that is brought in
proximity withthe ZAP70 tyrosine kinase, the main
phosphate-donor and thus activator, of these signaling
proteins. These interactions are then of importance for the
activation of both the MAP kinase andthe c-Jun kinase
pathways.
Fig. 7. Recruitment of Shb to GEMs after TCR ligation. JurkatTcells were not treated (NT) or stimulated withCD3 antibody (CD3) for 3 min,
followed by lysis in Mes lysis buffer. Lysates were subjected to sucrose gradient ultracentrifugation. Sequential fractions were removed from the top
of the gradient and indicated as fraction number. The fractions were precipitated using TCA, and after washing with acetone, dissolved in 40 lLof
SDS sample buffer. The samples were resolved on SDS/PAGE, followed by immunoblot analysis using anti-Shb Ig (A), anti-phosphotyrosine Ig
(4G10) (B), anti-LAT Ig (C), anti-Gads Ig (D), anti-Vav Ig (E) and anti-(SLP-76) Ig (F).
3286 C. K. Lindholm et al. (Eur. J. Biochem. 269) Ó FEBS 2002
ACKNOWLEDGEMENTS
We gratefully acknowledge the skillful technical assistance of Ing-Britt
Hallgren and Ing-Marie Mo
¨
rsare and peptide synthesis by Dr A
˚
ke
Engstro
¨
m. We thank Dr Jane McGlade for providing the anti-Gads Ig
and Dr Arthur Weiss for providing theVav expression plasmid. The
work has been supported by grants from the Juvenile Diabetes
Foundation International, Swedish Medical Research Council (31X-
10822), the Swedish Diabetes association, The Novo-Nordisk Foun-
dation andthe Family Ernfors Fund.
REFERENCES
1. Zhang, W., Sloan-Lancaster, J., Kitchen, J., Trible, R.P. &
Samelson, L.E. (1998) LAT: the ZAP-70 tyrosine kinase substrate
that linksT cell receptor to cellular activation. Cell 92, 83–92.
2. Wu, J., Motto, D.G., Koretzky, G.A. & Weiss, A. (1996) Vav and
SLP-76 interact and functionally cooperate in IL-2 gene activa-
tion. Immunity 4, 593–602.
3. Simons, K. & Ikonen, E. (1997) Functional rafts in cell mem-
branes. Nature 387, 569–572.
4. Simons, K. & Toomre, D. (2000) Lipid rafts and signal trans-
duction. Nat.Rev.MolCellBiol.1, 31–39.
5. Cherukuri, A., Dykstra, M. & Pierce, S.K. (2001) Floating the raft
hypothesis: lipid rafts play a role in immune cell activation.
Immunity 14, 657–660.
6. Raab, M., da Silva, A.J., Findell, P.R. & Rudd, C.E. (1997)
Regulation of Vav-SLP-76 binding by ZAP-70 and its relevance to
TCR zeta/CD3 induction of interleukin-2. Immunity 6, 155–164.
7. Fang, N. & Koretzky, G.A. (1999) SLP-76andVav function in
separate, but overlapping pathways to augment interleukin-2
promoter activity. J. Biol Chem. 274, 16206–16212.
8. Han, J., Das, B., Wei, W., Van Aelst, L., Mosteller, R.D.,
Khosravi-Far, R., Der Westwick, J.K., C.J. & Broek, D. (1997)
Lck regulates Vav activation of members of the Rho family of
GTPases. MolCellBiol.17, 1346–1353.
9. Han, J., Luby-Phelps, K., Das, B., Shu, X., Xia, Y., Mosteller,
R.D.,Krishna,U.M.,Falck,J.R.,White,M.A.&Broek,D.
(1998) Role of substrates and products of PI 3-kinase in regulating
activation of Rac-related guanosine triphosphatases by Vav.
Science 279, 558–560.
10. Deckert, M., Tartare-Deckert, S., Couture, C., Mustelin, T. &
Altman, A. (1996) Functional and physical interactions of Syk
family kinases withtheVav proto-oncogene product. Immunity 5,
591–604.
11. Liu, S.K. & McGlade, C.J. (1998) Gads is a novel SH2 and SH3
domain-containing adaptor protein that binds to tyrosine-
phosphorylated Shc. Oncogene 17, 3073–3082.
12. Liu,S.K.,Fang,N.,Koretzky,G.A.&McGlade,C.J.(1999)The
hematopoietic-specific adaptor protein gads functions in T-cell
signaling via interactions withtheSLP-76and LAT adaptors.
Curr. Biol. 9, 67–75.
13. Welsh,M.,Mares,J.,Karlsson,T.,Lavergne,C.,Breant,B.&
Claesson-Welsh, L. (1994) Shb is a ubiquitously expressed Src
homology 2 protein. Oncogene 9, 19–27.
14. Welsh, M., Songyang, Z., Frantz, J.D., Trub, T., Reedquist, K.A.,
Karlsson, T., Miyazaki, M., Cantley, L.C., Band, H. & Shoelson,
S.E. (1998) Stimulation through theT cell receptor leads to
interactions between SHBand several signaling proteins. Onco-
gene 16, 891–901.
15. Karlsson, T. & Welsh, M. (1997) Modulation of Src homology 3
proteins by the proline-rich adaptor protein Shb. Exp. Cell Res.
231, 269–275.
16. Lindholm, C.K., Gylfe, E., Zhang, W., Samelson, L.E. & Welsh,
M. (1999) Requirement of the Src homology 2 domain protein
Shb for T cell receptor-dependent activation of the interleukin-2
gene nuclear factor for activation of Tcells element inJurkat T
cells. J. Biol. Chem. 274, 28050–28057.
17. Lindholm, C.K., Frantz, J.D., Shoelson, S.E. & Welsh, M. (2000)
Shf, a Shb-like adapter protein, is involved in PDGF-alpha-
receptor regulation of apoptosis. Biochem. Biophys. Res. Commun.
278, 537–543.
18. Karlsson, T. & Welsh, M. (1996) Apoptosis of NIH3T3 cells
overexpressing the Src homology 2 domain protein Shb. Oncogene
13, 955–961.
19. Sander, E.E., van Delft, S., ten Klooster, J.P., Reid, T., van
der Kammen, R.A., Michiels, F. & Collard, J.G. (1998)
Matrix-dependent Tiam1/Rac signaling in epithelial cells
promotes either cell-cell adhesion or cell migration and is
regulated by phosphatidylinositol 3-kinase. J. Cell Biol. 143, 1385–
1398.
20. Fang, N., Motto, D.G., Ross, S.E. & Koretzky, G.A. (1996)
Tyrosines 113, 128, and 145 of SLP-76 are required for optimal
augmentation of NFAT promoter activity. J. Immunol. 157, 3769–
3773.
21. Kuhne, M.R., Ku, G. & Weiss, A. (2000) A guanine nucleotide
exchange factor-independent function of Vav1 in transcriptional
activation. J. Biol. Chem. 275, 2185–2190.
22. Zhang, W., Trible, R.P. & Samelson, L.E. (1998) LAT palmi-
toylation: its essential role in membrane microdomain targeting
and tyrosine phosphorylation during T cell activation. Immunity 9,
239–246.
23. Boerth,N.J.,Sadler,J.J.,Bauer,D.E.,Clements,J.L.,Gheith,
S.M. & Koretzky, G.A. (2000) Recruitment of SLP-76 to the
membrane and glycolipid-enriched membrane microdomains
replaces the requirement for linker for activation of TcellsinT cell
receptor signaling. J. Exp. Med. 192, 1047–1058.
24. Ishiai, M., Kurosaki, M., Inabe, K., Chan, A.C., Sugamura, K. &
Kurosaki, T. (2000) Involvement of LAT, Gads, and Grb2 in
compartmentation of SLP-76 to the plasma membrane. J. Exp.
Med. 192, 847–856.
25. Salojin, K.V., Zhang, J., Meagher, C. & Delovitch, T.L. (2000)
ZAP-70 is essential for theT cell antigen receptor-induced plasma
Fig. 8. Model of T cell receptor interactions with Shb. TCR engagement
results in recruitment of ZAP70 tyrosine kinase to the phosphorylated
ITAMs of the TCR f-chain. Shb is also recruited to the f-chain, via its
SH2 domain. Shb is also associated with LAT, PLC-c1, Grb2, SLP-76,
Gads and Vav, which are brought into the vicinity of ZAP70 and are
thus phosphorylated. This signaling complex can then recruit a number
of effectors, leading to activation of JNK, ERK1 and ERK2.
Ó FEBS 2002 Association of Shb, SLP-76andVavinJurkatcells (Eur. J. Biochem. 269) 3287
membrane targeting of SOS andVavinT cells. J. Biol. Chem. 275,
5966–5975.
26. Montixi, C., Langlet, C., Bernard, A.M., Thimonier, J., Dubois,
C., Wurbel, M.A., Chauvin, J.P., Pierres, M. & He, H.T. (1998)
Engagement of T cell receptor triggers its recruitment to low-
density detergent-insoluble membrane domains. EMBO J. 17,
5334–5348.
27. Teramoto, H., Salem, P., Robbins, K.C., Bustelo, X.R. &
Gutkind, J.S. (1997) Tyrosine phosphorylation of thevav proto-
oncogene product links FcepsilonRI to the Rac1-JNK pathway.
J. Biol. Chem. 272, 10751–10755.
28. Song, J.S., Haleem-Smith, H., Arudchandran, R., Gomez, J.,
Scott, P.M., Mill, J.F., Tan, T.H. & Rivera, J. (1999) Tyrosine
phosphorylation of Vav stimulates IL-6 production in mast
cells by a Rac/c-Jun N-terminal kinase-dependent pathway.
J. Immunol. 163, 802–810.
29. Arudchandran, R., Brown, M.J., Peirce, M.J., Song, J.S., Zhang,
J., Siraganian, R.P., Blank, U. & Rivera, J. (2000) The Src
homology 2 domain of Vav is required for its compartmentation
to the plasma membrane and activation of c-Jun NH(2)-terminal
kinase 1. J. Exp Med. 191, 47–60.
3288 C. K. Lindholm et al. (Eur. J. Biochem. 269) Ó FEBS 2002
. co-immunoprecipitation of several
tyrosine phosphorylated proteins
To further investigate the protein-interactions of the Shb
adapter protein, we attempted to identify. protein, containing both the PTB domain and
two of the N-terminal proline-rich domains. This interac-
tion was only slightly inhibited by phosphotyrosine.