Oncogene (1997) 14, 223 ± 231
 1997 Stockton Press All rights reserved 0950 ± 9232/97 $12.00
The Nck SH2/SH3 adaptor protein is present in the nucleus and associates
with the nuclear protein SAM68
Deirdre C Lawe1, Christine Hahn3 and Albert J Wong1,2
Departments of 1Microbiology and Immunology, 2Pharmacology, The Kimmel Cancer Institute, and 3Je€erson Medical College,
Philadelphia, Pennsylvania 19107
SH2/SH3 adaptor proteins are essential components of
the signal transduction pathways initiated by tyrosine
kinases. Nck is a ubiquitously expressed adaptor protein
whose function has been enigmatic. We performed
confocal microscopy to localize Nck in NIH3T3 and
A431 cells. Surprisingly, Nck was identi®ed in the
nucleus as well as the cytoplasm with no visible change
in localization due to PDGF or EGF stimulation.
Western blot analysis of nuclear and cytosolic fractions
con®rmed that there was no translocation in response to
growth factor and that tyrosine phosphorylation was
speci®c to only cytosolic Nck. Far Western blot analysis
with either Nck, the SH2 domain, or the SH3 domains
revealed di€erential binding in nuclear and cytosolic
lysates, indicating speci®c binding partners for each
subcellular location. The major target of c-Src during
mitosis is SAM68, a RNA-binding protein ordinarily
localized to the nucleus. SAM68 was identi®ed as a
nuclear speci®c binding partner of Nck in both nonmitotic and mitotic cells. Several tyrosine kinases can be
found in the nucleus but their signal transduction remains
unde®ned. The discovery of an adaptor protein in the
nucleus suggests there are signal transduction mechanisms within the nucleus that recapitulate those found in
the cytoplasm.
Keywords: Nck; SAM68; SH3 domains; signal transduction; tyrosine kinases
Introduction
Eukaryotic signal transduction pathways initiated by
receptor protein tyrosine kinase (RPTK) activation
utilize both phosphorylation and protein-protein
interactions to disseminate their signals throughout
the cell. Two motifs that frequently mediate these latter
interactions are SH2 and SH3 domains (van der Geer
et al., 1994; Pawson, 1995; Cohen et al., 1995). These
domains have now been identi®ed in many types of
proteins including enzymes and transcription factors
(Pawson, 1995; Cohen et al., 1995). One intriguing
family of SH2/SH3 containing proteins are the adaptor
proteins. These proteins consist primarily of SH2 and
SH3 domains and have no intrinsic enzymatic activity
(Pawson, 1995; Cohen et al., 1995), indicating that the
primary role of adaptor proteins is to couple an
upstream signaling molecule with a downstream
e€ector. The prototype of the adaptor proteins is
Correspondence: A Wong
Received 12 April 1996; revised 12 September 1996; accepted 12
September 1996
Grb2. The SH2 domain of Grb2 binds to an
autophosphorylation site on a RPTK, and the SH3
domains bind to the RAS guanine nucleotide exchange
factor SOS thus linking RPTKs to the RAS signaling
pathway (Olivier et al., 1993; Simon et al., 1993;
Lowenstein et al., 1992).
The SH2 domains of adaptor proteins can also bind
to several types of cytoplasmic proteins. For example,
the SH2 domain of Grb2 can bind to the tyrosine
phosphatase SHPTP2/syp (Li et al., 1994) and focal
adhesion kinase (FAK) (Schlaepfer et al., 1994). SH3
domains recognize proline rich sequences in a constitutive manner and a wide variety of proteins can be targets
of SH3 domains (Pawson, 1995). The SH3 domains of
Grb2 can bind to the GTPase dynamin (Gout et al.,
1993), the recently described Gab1 docking protein
(Holgado-Madruga et al., 1996), and the p85 subunit of
PI-3-kinase (Wang et al., 1995). SH3 domains can also
in¯uence signal transduction by localizing proteins to
speci®c subcellular locations. Microinjection experiments have shown that the SH3 domains of Grb2
localize it to membrane ru‚es (Bar-Sagi et al., 1993).
Thus, adaptor proteins may be utilized for diverse types
of interactions throughout the cell.
Nck is a 47 kDa protein consisting of three SH3
domains followed by one SH2 domain (Lehmann et al.,
1990). It is expressed in all tissues examined thus far
and is phosphorylated on serine, threonine and
tyrosine after growth factor addition (Li et al., 1992;
Meisenhelder and Hunter, 1992; Park and Rhee, 1992;
Chou et al., 1992). It has no known enzymatic activity
and so is thought to function as an adaptor protein. In
support of this role, Nck has been shown to associate
via its SH2 domain with activated RPTKs, FAK
(Nishimura et al., 1993; Schlaepfer et al., 1994) and the
docking protein IRS-1 (Lee et al., 1993). The SH3
domains of Nck have been shown to associate with
SOS (Hu et al., 1995), a 65 kDa serine/threonine kinase
termed NAK (Chou and Hanafusa, 1995), and the
Wiskott-Aldrich syndrome protein, WASP (RiveroLezcano et al., 1995). The Drosophila homologue of
Nck has recently been identi®ed as dreadlocks (dock)
(Garrity et al., 1996). Mutations in this gene disrupt
photoreceptor cell (R cell) axon guidance and
targeting, although the signal transduction machinery
involved in axon guidance is unknown. Overexpression
of Nck can lead to tranformation of ®broblasts (Li et
al., 1992; Chou et al., 1992); however, dominant
negative Nck mutants have no e€ect on ERK1
activation by oncogenic Abl or in response to EGF
(Tanaka et al., 1995). Consequently, the relationship
between these protein interactions and cell proliferation
has yet to be elucidated, and so the function of Nck
remains incompletely de®ned.
The Nck adaptor protein is present in the nucleus
D Lawe et al
224
In the present study, we demonstrate that Nck is a
nuclear as well as a cytosolic protein. Immunolocalization and biochemical studies indicate that the
subcellular location of Nck is independent of growth
factor stimulation, and that tyrosine phosphorylation is
speci®c to the cytosolic Nck. We show binding partners
speci®c to the nuclear and cytosolic fractions, and that
one of the nuclear targets of Nck is the SAM68
protein. The discovery of Nck in the nucleus may
further illuminate its role in cellular signaling.
Results
Immunolocalization of Nck
To further elucidate the function of Nck in intracellular signaling, we determined its subcellular localization in quiescent NIH3T3 and A431 cells by confocal
microscopy using an anti-Nck speci®c monoclonal
antibody. Interestingly, we found Nck localized to
both the nuclear and the cytoplasmic region of these
cells as seen in Figure 1. The staining had a speckled
appearance in both the cytoplasm and nucleus, with
occasional intense punctate staining that was more
prominent in the nucleus. The nuclear distribution of
Nck excludes the nucleoli but also illuminates the
periphery of the nucleus, suggesting localization to the
nuclear envelope as well. Using confocal microscopy, it
is possible to construct a vertical cross section of the
cell known as a Z scan. As shown in Figure 1, Z scans
con®rmed the distinct nuclear localization of Nck in
both cell types. The same pattern of distribution was
observed using methanol to ®x and permeabilize the
cells (data not shown). No staining was seen after
treatment with secondary antibody alone (Figure 1) or
after competition with the recombinant GST-K-Nck
fusion protein, indicating the speci®city of the
immunostaining.
E€ect of PDGF stimulation on the localization of Nck in
NIH3T3 cells
Nck has previously been shown to associate with the
EGF and PDGF receptors upon stimulation with
ligand (Li et al., 1992; Meisenhelder and Hunter,
1992; Park and Rhee, 1992), and the autophosphorylation site to which Nck binds has been identi®ed on the
PDGF receptor (Nishimura et al., 1993). However,
following RPTK activation certain signal transduction
PDGF(min.):
0
proteins translocate to the nucleus such as the STAT
proteins or MAP kinases (Davis, 1993; Karin and
Hunter, 1995). We wished to determine the e€ect
growth factor stimulation would have on the subcellular localization of Nck. In order to test this,
quiescent NIH3T3 cells were stimulated with PDGF
for either 10 min or 60 min, and the subcellular
localization of Nck was determined as above.
Compared to quiescent cells (Figure 2), the pattern of
Nck staining in the stimulated cells revealed no
obvious alteration in the distribution of Nck. The
same was true for serum stimulation of these cells, as
well as EGF stimulation of serum starved A431 cells
(data not shown). These results demonstrate that there
is no extensive redistribution of Nck following growth
factor or serum stimulation. The failure to detect any
A431
NIH-3T3
Anti-Nck
Z Scan
No 10Ab
Figure 1 Confocal microscopy reveals the presence of Nck in
both the nucleus and cytoplasm. Quiescent A431 and NIH3T3
were ®xed with 2% formaldehyde and incubated with anti-Nck
mAb, followed by FITC conjugated anti-mouse antibody. The
cells were visualized by confocal laser scanning microscopy. Note
the presence of Nck in both the cytoplasm and nucleus. A vertical
cross section, or Z scan, of the confocal image of each cell type is
also shown demonstrating bright staining of the nucleus. The
control for each cell type shows no labeling when only the antimouse secondary antibody was used for staining. Bar=10 mm
10
60
Figure 2 Subcellular localization of Nck is independent of growth factor stimulation. Quiescent NIH3T3 cells were stimulated with
50 ng/ml of PDGF for the indicated times, then incubated with anti-Nck mAb and visualized as described in Figure 1. Bar=10 mm
The Nck adaptor protein is present in the nucleus
D Lawe et al
membrane association of Nck after PDGF stimulation
may indicate that the fraction of total cellular Nck that
associates with receptors after ligand stimulation is
possibly too small to detect using confocal microscopy.
Biochemical characterization of the subcellular location
of Nck
The nuclear and cytoplasmic localization of Nck was
con®rmed by Western blot analysis of subcellular
fractions. Quiescent or PDGF stimulated NIH3T3
cells were fractionated into their cytosolic and nuclear
components and the blots incubated with anti-Nck
antibody. In quiescent cells, Nck was present in both
the nuclear and cytosolic fractions (Figure 3a, lanes 1
and 4). PDGF stimulation had no detectable e€ect on
Nck protein levels in either the nuclear (Figure 3a,
lanes 2 and 3) or cytosolic fractions (Figure 3a, lanes 5
and 6), demonstrating that there is no substantial
translocation of Nck to either subcellular location due
to PDGF stimulation. Similarly, treatment of NIH3T3
cells with serum or A431 cells with EGF had no e€ect
on the levels of Nck found in either the nuclear or
cytosolic fractions of these cells (data not shown).
A recent report by Rivero-Lezcano et al. described
Nck as present in both the cytoplasmic and membrane
fractions of HL60 cells, but absent from the nuclear
fraction (Rivero-Lezcano et al., 1995). To help clarify
this discrepancy, both HL60 cells and another
a
Nucleus
PDGF (min.):
Cytosol
Total
0 10 60
0 10 60
0 10 60
Blot: Anti-SHPTP2
— 69
Anti-Nck
— 46
Anti-SAM68
— 69
1
b
2
3
4
5
BaF3
N
C
6
7
8
9
HL60
T
N
C
T
Blot: Anti-SHPTP2
— 69
Anti-Nck
— 46
Anti-SAM68
— 69
Figure 3 Analysis of subcellular fractions shows that the
subcellular localization of Nck is independent of growth factor
stimulation. (a) 50 mg of total cell lysate, cytosolic and nuclear
fractions from NIH3T3 cells that were either quiescent (0) or
stimulated for the indicated times with PDGF were separated by
SDS ± PAGE, transferred to nitrocellulose and the blots incubated
with anti-Nck mAb. (b) 50 mg of total cell lysate (T), cytosolic (C)
and nuclear (N) fractions from BaF3 and HL60 cells were also
immunoblotted with anti-Nck antibody. To demonstrate proper
fractionation, the same ®lters were reblotted with anti-SHPTP2/
syp mAb as a cytosolic marker. Duplicate ®lters were blotted with
anti-SAM68 as a nuclear marker
hematopoietic cell line, BaF3, were fractionated and
analysed as above. Nck was detected in the nuclear and
cytosolic fractions of both cell lines. While the 47 kDa
form of Nck was present in the nuclear fraction of
HL60 cells, the predominant band detected by the antiNck antibody was a 56 kDa band (Figure 3b) which
may be preferentially seen in this compartment due to
the high salt conditions used to extract the nuclei. This
band may be due to a cellular modi®cation of Nck or
represent an alternatively spliced form. The presence of
an alternative form might explain the relative lack of
the 47 kDa version of Nck in the nucleus.
To control for appropriate fractionation, the same
®lters were reprobed with an antibody to SHPTP2/syp,
a protein-tyrosine-phosphatase known to be cytosolic
(Rivard et al., 1995; Peraldi et al., 1994), and duplicate
®lters were reprobed with an antibody against SAM68,
a nuclear protein (Wong et al., 1992; Taylor et al.,
1995). As expected, SHPTP2/syp was detected only in
the cytosolic fraction and SAM68 only in the nuclear
fraction, indicating that these subcellular fractions were
not measurably contaminated with each other (Figure
3).
Tyrosine phosphorylation of Nck does not alter its
subcellular localization
Previous work has shown that the activation of several
RPTKs by their cognate ligands results in the tyrosine
phosphorylation of Nck (Li et al., 1992; Meisenhelder
and Hunter, 1992; Park and Rhee, 1992; Ryan and
Gillespie, 1994; Guo et al., 1995). Since the level of
Nck present in the nuclear and cytosolic fractions did
not change signi®cantly following growth factor
stimulation, we wished to determine if the tyrosine
phosphorylated subpopulation of Nck altered its
location upon growth factor addition. Whole cell
lysates, nuclear and cytosolic fractions from NIH3T3
cells that were either quiescent or stimulated with
PDGF were used in immunoprecipitations with antiNck antibody. Western blots of these samples were
then incubated with an anti-phosphotyrosine antibody.
As previously described (Li et al., 1992; Park and
Rhee, 1992), the phosphotyrosine content of Nck in
total cell lysates increased after 10 min of PDGF
stimulation and was maintained for up to 60 min (data
not shown). This same pattern of tyrosine phosphorylation was reiterated in the cytosolic fraction (Figure
4), while the nuclear counterpart of Nck showed no
tyrosine phosphorylation in either quiescent or PDGF
stimulated fractions. The same ®lter was then reprobed
with anti-Nck antibody to show equal levels of Nck
precipitated from each lysate (Figure 4). These results
demonstrate di€erential tyrosine phosphorylation of
Nck based on its subcellular location. This fact, taken
together with the subcellular location of Nck being
independent of growth factor stimulation, indicates
that the nuclear subpopulation of Nck is distinct from
the cytoplasmic.
Far Western analysis identi®es Nck binding partners
speci®c to each subcellular location
Having identi®ed two distinct populations of Nck
based on subcellular location, we reasoned that there
should be binding partners of Nck speci®c to each site.
225
The Nck adaptor protein is present in the nucleus
D Lawe et al
Whole cell lysates, as well as nuclear and cytosolic
fractions that were either quiescent or stimulated with
PDGF were separated by SDS ± PAGE and transferred
to nitrocellulose. GST-K-Nck fusion protein was
phosphorylated in vitro with [g-32P]ATP, cleaved with
Factor Xa, and the labeled Nck protein that was
liberated was used to probe this ®lter. [g32P]Nck was
Cytosol
PDGF (min.):
0
10
found to bind a series of proteins in total cell lysate and
the most prominent migrated as 25, 47, 68, 69, 105 and
180 kDa bands (Figure 5a). Certain proteins appeared
to be recognized in a growth factor dependent manner,
such as the 25 kDa protein (Figure 5a), although the
binding of this band was variable. These same proteins
were identi®ed in GST-K-Nck precipitates of the same
lysates (data not shown). Analysis of cell fractions
revealed that some of these binding partners had a
speci®c subcellular localization. The 47, 68 and
105 kDa bands were nuclear proteins, while the 25
and 180 kDa bands were found in the cytoplasm. The
69 kDa protein was present in both fractions.
Since both SH2 and SH3 domains have been shown
to mediate protein-protein interactions, we wished to
delineate which domain(s) of Nck were required to
bind these proteins. A fusion protein containing the
three SH3 domains of Nck (GKN3S) was used in far
Western analysis as above and bound to the 25 kDa
protein found in the cytoplasmic fraction (data not
shown). The most prominent proteins bound by the
SH3 domains in the nuclear fraction were the 68, 69
and 105 kDa proteins (Figure 5b). No binding was
seen when labeled GST fusion protein was used to
probe identical ®lters (data not shown). Binding
partners speci®c to the SH2 domain were identi®ed
using GST-K-SH2 for far Western analysis in place of
a g-32P-labeled probe in order to avoid high back-
Nucleus
60
0
10
60
Blot: Anti-PTyr
— 46
Anti-Nck
— 46
IP: Anti-Nck
Figure 4 Di€erential tyrosine phosphorylation of Nck based on
its subcellular location. Cytosolic and nuclear fractions were
prepared from quiescent NIH3T3 cells or cells stimulated for the
indicated times with PDGF. 150 mg of each were used in
immunoprecipitations with anti-Nck mAb and Western blots
containing the precipitates were incubated with anti-phosphotyrosine antibody. The same ®lter was incubated with anti-Nck mAb
to show that equal amounts of Nck were precipitated from each
lysate
a
Total
Nucleus
0 10 60
0 10 60
▲
▲
PDGF (min.): 0 10 60
220 —
Cytosol
▲ ▲
97 —
69 —
46 —
30 —
▲
22 —
14 —
Probe:
32P-Nck
b
Nucleus
PDGF (min.): 0 10 60
220 —
Nucleus
0 10 60
220 —
▲▲ ▲
97 —
69 —
46 —
Nucleus
0 10 60
220 —
▲ ▲
226
97 —
69 —
46 —
97 —
69 —
46 —
30 —
22 —
30 —
22 —
30 —
22 —
14 —
14 —
14 —
Probe:
GKN3S
GST-K-SH2
GST-K
Figure 5 Nck has distinct nuclear and cytosolic binding partners. (a) 25 mg of total cell lysate, 20 mg of cytosolic and 5 mg of
nuclear lysate from NIH3T3 cells that were either quiescent or stimulated with PDGF were separated by SDS ± PAGE, transferred
to nitrocellulose and incubated with g-32P-labeled Nck protein. Prominent cytoplasmic binding partners are indicated by arrow
heads to the left, while those on the right indicate nuclear binding partners. (b) Identical nuclear lysates were incubated with GSTK-SH2 (or GST-K as a control) followed by antibody to GST and detected with 125I secondary antibody to identify SH2 domain
binding partners. SH3 domain binding partners were identi®ed using a g-32P-labeled GKN3S probe. Numbers to the left indicate the
size of molecular weight markers in kDa
The Nck adaptor protein is present in the nucleus
D Lawe et al
To con®rm that Nck and SAM68 could interact
directly, SAM68 was immunoprecipitated from whole
cell lysates and the resulting blot probed with in vitro
labeled Nck. [g-32P]Nck did bind directly to SAM68
(data not shown) and the association between Nck and
SAM68 seen in these experiments was not dependent
on growth factor addition, suggesting it was mediated
ground. The Nck SH2 domain identi®ed the 47, 69 and
105 kDa proteins as binding partners in the nuclear
fraction (Figure 5b), as well as the 180 kDa band in
the cytoplasmic fraction (data not shown). Again, no
binding was seen with the GST fusion protein itself
(Figure 5b). Thus, Nck recognized proteins that were
speci®c to the nucleus and the cytoplasm further
supporting the notion that the nuclear subpopulation
of Nck serves a di€erent function from its cytosolic
counterpart.
a
SAM68 associates with Nck
PDGF (min.):
69 —
46 —
Blot:
0
10
60
69 —
Blot:
Anti-SAM68
b
Lysate
M
IP: Anti-Nck
A
M
A
Blot: Anti-Sam68
— 69
— 46
Anti-Nck
Figure 6 Association of SAM68 with Nck in NIH3T3 nuclear
lysates. (a) Nuclear fractions were prepared from quiescent
NIH3T3 cells or cells stimulated for the indicated times with
PDGF. 300 mg of each lysate was subjected to immunoprecipitation with anti-Nck mAb. Western blots containing the precipitates
were incubated with anti-SAM68 antibody. (b) 500 mg of nuclear
lysate from mitotic (M) or asynchronous (A) NIH3T3 cells was
immunoprecipitated with anti-Nck antibody. 50 mg of each lysate
along with the immunoprecipitates were then resolved by SDS ±
PAGE, transferred to nitrocellulose, the upper half of which was
incubated with anti-SAM68 antibody and the lower half with
anti-Nck antibody
GST-K-SH3C
GST-K-SH3B
GST-K-SH3A
b
GST
GKN3S
GST-K-Nck
PPT:
97 —
GST-K-SH2
SAM68 is an RNA binding protein recently identi®ed as
the predominant substrate and binding partner of c-Src
in mitotic cells (Taylor et al., 1995; Taylor and
Shalloway, 1994; Fumagalli et al., 1994). When the cSrc tyrosine kinase is activated during mitosis it
associates with the predominantly nuclear SAM68 via
both its SH2 and SH3 domains, presumably through
nuclear envelope breakdown (Wong et al., 1992; Taylor
et al., 1995; Taylor and Shalloway, 1994; Fumagalli et
al., 1994). It has been suggested that SAM68 may act as a
sca€olding protein for c-Src during mitosis (Taylor et al.,
1995; Richard et al., 1995), as several other SH2/SH3
proteins have been shown to associate with SAM68 such
as the adaptor protein Grb2, the p85 subunit of PI-3kinase and PLCg-1. Since SAM68 is principally localized
to the nucleus (Wong et al., 1992) we asked if SAM68
could associate with Nck in non-mitotic cells. Analysis of
anti-Nck immunoprecipitates using nuclear lysates from
either quiescent cells or cells stimulated with PDGF
revealed the co-precipitation of SAM68 (Figure 6a). This
interaction was independent of growth factor addition.
The Nck-SAM68 interaction was also examined in cells
undergoing mitosis. Nck was immunoprecipitated from
nuclear lysates that were either asynchronous or blocked
in mitosis by nocodazole. There was *30% less SAM68
associated with Nck from mitotic vs asynchronous
lysates, even though *25% more Nck was found in
immunoprecipitations from mitotic cells (Figure 6b).
There was no change in SAM68 levels during mitosis.
One possible explanation is that SAM68 is being
sequestered by activated c-Src during mitosis resulting
in a loss of association with Nck.
a
IP: Anti-Nck
69 —
IP:
Anti-SAM68
Anti-SAM68
Figure 7 The SH3 domains of Nck mediate SAM68 binding. (a) 150 mg of total cell lysate from quiescent NIH3T3 cells were
subjected to precipitations with either GST-K-Nck, GKN3S, GST-K-SH2, or GST alone. Blots containing the precipitates were
then incubated with anti-SAM68 antibody. (b) 400 mg of lysate from NIH3T3 cells was immunoprecipitated with anti-SAM68
antibody. The precipitates were divided in half, separated by SDS ± PAGE, and then transferred to nitrocellulose. One half of each
precipitate was probed as a far Western with either GST-K-SH3A, GST-K-SH3B or GST-K-SH3C. The other half was incubated
with anti-SAM68 antibody to con®rm the identity of the protein bound by Nck (data not shown)
227
The Nck adaptor protein is present in the nucleus
D Lawe et al
228
by the SH3 domains of Nck. The nature of this
interaction was investigated by performing precipitations on cell lysates with GST-K-Nck, GKN3S, GSTK-SH2 or GST fusion proteins. SAM68 bound only to
GST-K-Nck and GKN3S (Figure 7a), con®rming that
the Nck-SAM68 interaction was mediated by the SH3
domains of Nck. To delineate which of the three SH3
domains encoded by Nck was responsible for this
interaction, each SH3 domain was expressed individually as a GST fusion protein and used in far Western
analysis of SAM68 immunoprecipitated from whole
cell lysate. The fusion protein encoding the SH3
domain most amino-terminal in Nck, GST-K-SH3A,
bound directly to SAM68 while the other two did not
(Figure 7b). These results indicate that Nck is
constitutively bound to SAM68 and that the ®rst
SH3 domain is essential for the Nck-SAM68 interaction.
Discussion
We have used subcellular localization as a means to
further elucidate the function of Nck in signaling. We
have demonstrated that Nck is both a nuclear and a
cytoplasmic protein and that this subcellular location
is independent of growth factor stimulation by both
confocal microscopy and Western blot analysis of
subcellular fractions. The amount of Nck in the
nucleus could be substantial, but as with the
quantitation of MAP kinase (Chen et al., 1992) a
more precise determination is dicult due to the
signi®cant amount of protein leakage from nuclei
during cell fractionation (Paine et al., 1983).
Since Nck does not contain a known nuclear
localization signal (NLS), its presence in the nucleus
may be through its association with other proteins.
Two Nck binding partners have recently been identi®ed
as WASP (Rivero-Lezcano et al., 1995) and c-Cbl
(Rivero-Lezcano et al., 1994), each containing an NLS.
A certain proportion of the WASP is localized to the
nucleus, and the same has been found for the viral
form of c-Cbl (Rivero-Lezcano et al., 1995; Blake et
al., 1993). Nck has also been shown to bind c-Abl in
vitro, a predominantly nuclear tyrosine kinase containing an NLS (Ren et al., 1994; Wang, 1993). SH3
interactions have been shown to be constitutive, as
opposed to SH2 interactions which are induced by a
tyrosine phosphorylation event, and this agrees with
our ®nding that the nuclear localization of Nck is
independent of growth factor stimulation.
We have identi®ed one of these nuclear speci®c
binding partners of Nck as SAM68, an RNA binding
protein. SAM68 is the predominant substrate and
binding partner of c-Src in c-Src transformed fibroblasts
(Taylor et al., 1995). We have demonstrated that Nck is
constitutively associated with SAM68 via its most
amino-terminal SH3 domain. PLCg-1 has been shown
to bind SAM68 in a similar fashion (Richard et al.,
1995). Although we have found that Nck protein levels
increase slightly during mitosis, the quantity of SAM68
associated with Nck is decreased. This reduction is most
likely explained by the association of c-Src with SAM68
during mitosis. A functional role for SAM68 in the cell
has not yet been clearly de®ned, but it has been shown
that binding of the SH3 domain of c-Src to SAM68 can
interfere with the binding of SAM68 to RNA (Taylor et
al., 1995). The consequences of the Nck-SAM68
interaction in non-mitotic as well as mitotic cells awaits
further clari®cation.
The discovery of Nck in both the nucleus and
cytoplasm suggests several possibilities related to cell
signaling. Sequestration to a particular subcellular
location may regulate a proteins' interactions with its
substrates and this partitioning may control subsequent cellular events, as is the case for c-Abl. c-Abl is
predominantly located in the nucleus and has been
shown to be a negative regulator of cell growth when
overexpressed (Sawyers et al., 1994). Mutation of its
SH3 domain activates its transforming abilities and
results in a cytoplasmic localization (Van Etten et al.,
1995). v-Abl and Bcr-Abl, the oncogenic versions of cAbl, are also found in the cytoplasm (Wang, 1993).
Whether the nuclear localization of Nck can enhance
or suppress cell growth is presently not clear, but we
have noted that Nck is predominantly found in the
nucleus of primary glial tumor cells by immunohistochemistry and in cultured glial tumor cells by
confocal microscopy (DCL, CH, and AJW, unpublished observations).
A second possibility is that Nck is involved in
communicating signals from the cytoplasm to the
nucleus, or vice versa. However, our experiments have
suggested that the cytosolic and nuclear pools of Nck do
not undergo any signi®cant exchange. Thus, another
possibility is that Nck functions as an adaptor protein in
the nucleus. Several non-receptor protein tyrosine
kinases are localized to and are active in the nucleus.
Much as they are important in linking tyrosine kinases to
e€ector molecules in the cytoplasm, adaptor proteins
may be essential in linking nuclear tyrosine kinases and
their substrates to speci®c nuclear e€ectors. In support of
this notion, we have identi®ed speci®c SH2 and SH3
domain binding partners of Nck in the nuclear fraction
of cells. Many of the putative e€ectors of the SH3
domains of Nck that have recently been identi®ed are
found in the nucleus (Rivero-Lezcano et al., 1995; Blake
et al., 1993; Wang, 1993). Speci®c binding partners have
also been identi®ed for the SH2 and SH3 domains of the
adaptor protein c-Crk in cytosolic and nuclear lysates
(Feller et al., 1995). Using nucelar extracts from v-Crk
transformed 3Y1 cells, several tyrosine phosphorylated
nuclear speci®c binding partners for the SH2 domain of
Nck were also identi®ed (Feller et al., 1995). These
®ndings, together with our studies, lead us to speculate
that the role of the nuclear subpopulation of Nck is
primarily to serve as an adaptor protein.
Interestingly, it has been found that a phosphoinositide metabolism pathway exists within the nuclear
envelope similar to that which takes place in the
cytoplasmic membrane (Divecha et al., 1991, 1993).
Furthermore, PI-3-kinase has been localized to the
nucleus (Peles et al., 1992; Neri et al., 1994). The Vav
proto-oncogene, which can interact with Grb2 and
other cytoplasmic proteins, binds to the nuclear
antigen Ku-70 and has been localized to the nucleus
(Romero et al., 1996). These ®ndings, along with
previous studies showing tyrosine kinases in the
nucleus and the results presented here, add to a
growing body of data suggesting that there are
intricate signal transduction mechanisms within the
nucleus that recapitulate those found in the cytoplasm.
The Nck adaptor protein is present in the nucleus
D Lawe et al
Materials and methods
Generation of recombinant proteins
Full length Nck (nucleotides 101 ± 1235) (Lehmann et al.,
1990) was ampli®ed from human glial tumor RNA by
reverse transcriptase-polymerase chain reaction using a 5'
oligonucleotide that contained the initiation methionine
and a BamHI site (TGGGATCCCGATGGCAGAAGAAGTGGTG) and a 3' primer that contained an
EcoRI site (GGAATTCCATGATAAATGCTTGACAAGA). The sequence of the cDNA was veri®ed and cloned
in frame into the BamHI-EcoRI sites of pGEX-5x-3
(Pharmacia, Piscataway NJ). An oligonucleotide encoding
the cAMP-dependent protein kinase phosphorylation site
(Ron and Dressler, 1992) was inserted into the BamHI site
in-frame with the GST. This construct was named GSTK-Nck. A fusion construct encoding only the three SH3
domains of Nck (nucleotides 101 ± 942) was generated
exactly as described above with the exception of a
di€erent 3' oligonucleotide (GGAATTCCGATTGCCAGCAAACTT-TCCA) and named GKN3S. GST-K-SH2, the
fusion protein consisting of GST and the SH2 domain of
Nck was generated by digesting GST-K-Nck with AATIEcoRI and gel purifying the 317 bp SH2 domain. This
was subcloned into pGEX-5X-3 containing the same PKA
site as above, except the 5' end of the PKA oligonucleotide was modi®ed to a BglII site such that the BamHI site
at the 3' end of the oligonucleotide was unique. The
resulting vector named GST-K was digested with BamHI,
blunt ended, digested with EcoRI, into which the SH2
domain was then subcloned. The three individual SH3
domains of Nck were each PCR ampli®ed from GST-KNck and cloned into the BamHI-EcoRI sites of GST-K.
The resulting constructs were named GST-K-SH3A
(nucleotides 101 ± 368), GST-K-SH3B (nucleotides 387 ±
623) and GST-K-SH3C (nucleotides 606 ± 942). These
fusion constructs were used to transform competent
DH5a cells (Gibco/BRL, Bethesda, MD), and the
resulting GST fusion proteins were expressed by induction with 1 mM isopropyl-b-D-thiogalactoside for 2 h at
378C. The expressed GST fusion proteins were then
isolated from bacterial lysates by anity chromatography
using a glutathione-sepharose column, followed by elution
with glutathione, as described (Smith and Johnson, 1988).
Cell culture
NIH3T3 murine ®broblasts were maintained in Dulbecco's
modi®ed Eagle's medium (DMEM) supplemented with
100 U/ml penicillin, 100 mg/ml streptomycin, 100 mg/ml
kanamycin and 10% calf serum (Gibco/BRL). Con¯uent
cells were made quiescent by culturing in low serum (0.5%) for
18 to 22 h prior to stimulation with 50 ng/ml of PDGF-BB
(recombinant human, Gibco/BRL) or 10% calf serum. A431
human squamous carcinoma cells were maintained in DMEM
supplemented with 100 U/ml penicillin, 100 mg/ml streptomycin, 100 mg/ml kanamycin and 10% fetal bovine serum
(Gibco/BRL). Cells con¯uent for at least 48 h were serum
starved in DMEM for 24 h prior to stimulation with 100 ng/
ml EGF (recombinant human, Gibco/BRL). Mitotic cells
were isolated as previously described (Taylor et al., 1994) with
the exception that whole cells were not lysed, but were
fractionated. Human promyelomonocytic HL60 cells were
maintained in RPMI 1640 medium supplemented with 10%
fetal bovine serum (Gibco/BRL). Murine hematopoietic cells
BaF3 were maintained in RPMI 1640 medium supplemented
with 10% fetal bovine serum and 2 U/ml of IL-3 (Genzyme).
Immuno¯uorescence
Cells were grown to 40 ± 50% con¯uence on coverslips,
made quiescent and were either unstimulated or stimu-
lated as above. Coverslips were washed in cold phosphate
bu€ered saline (PBS) and ®xed in 2% formaldehyde/PBS
for 10 min at 48C, followed by permeabilization with
0.2% Triton X-100/PBS for 10 min at 48C. Alternatively,
cells were ®xed for 5 min in 7108C methanol and air
dried. Cells were then washed twice with PBS and blocked
with 5% nonfat dry milk, 0.9% goat serum/PBS for
30 min at 48C, followed by a 5 min incubation with 1%
BSA/PBS. Cells were incubated with 2 mg/ml of anti-Nck
monoclonal antibody (mAb) (Transduction Laboratories,
Lexington, KY) in 1% BSA/PBS for 30 min at 48C. Cells
were then washed three times for 5 min each in PBS
followed by incubation with FITC conjugated goat antimouse antibody (Boehringer Mannheim, Indianapolis,
IN), diluted 1 : 25 in 5% nonfat dry milk/PBS for
30 min at 48C. Cells were post ®xed in 2% formaldehyde/PBS. Coverslips were mounted onto slides with
SLOW-FADE (Molecular Probes, Eugene, OR) and
sealed with nail polish. Microscopy was performed at
the Kimmel Cancer Institute Laser Confocal Microscope
Facility with a Zeiss Axiovert 100 confocal microscope
with a Bio Rad MRC600 Krypton-Argon mixed gas
multi-line scanning laser rated at 15 milliwatts. The cells
were viewed using optical sections of 0.25 mm thickness.
The excitation and emission wavelengths were 488 nm and
520 nm for FITC.
For blocking experiments, the anti-Nck antibody was
incubated overnight with 40 mg of GST-K-Nck, and the same
amount of fusion protein was added to the blocking solution.
Subcellular fractionation
Fractions were prepared as previously described (McCaffrey et al., 1992) with several modi®cations. Cells were
washed twice with ice cold PBS and nuclear extracts
prepared by scraping the cells into 0.8 ml of Dignum bu€er
A (10 mM Tris pH 7.4, 10 m M NaCl, 3 mM MgCl2, 0.1 mM
EGTA, 0.5 mM DTT, 1 mM PMSF, 2 mg/ml leupeptin,
2 mg/ml aprotinin, and 1 mM sodium orthovanadate). After
10 min on ice, 50 ml of 10% NP-40 was added and the cells
were immediately vortexed. Nuclei were pelted by
centrifugation at 500 g at 48C for 30 s. Nuclei were
washed twice in 0.8 ml of bu€er A without NP-40, and
the nuclei were lysed in 50 ml of Dignum bu€er C (20 mM
HEPES, pH 7.9, 420 mM NaCl, 1.5 mM MgCl2, 0.1 mM
EDTA, 25% glycerol, 0.5 mM DTT, 0.5 mM PMSF, 2 mg/
ml leupeptin, 2 mg/ml aprotinin, 1 mM sodium orthovanadate) by 15 min of constant shaking at 48C. After lysis,
nuclear preparations were centrifuged at 12 000 g for
10 min at 48C and protein concentrations of the resulting
supernatants were determined by the Bio Rad DC protein
assay (Bio Rad, Hercules, CA) and stored at 7808C.
The cytosolic fraction was prepared by scraping cells into
0.8 ml of lysis bu€er containing 7.5 mM Tris, pH 7.6, 1 mM
MgCl2, 0.5 mM DTT, 0.1 mM EDTA, 2 mM PMSF, 20 mg/ml
leupeptin, 20 mg/ml aprotinin and 1 mM sodium orthovanadate. Lysates were subjected to one freeze/thaw cycle and
centrifuged at 12 000 g for 10 min at 48C.
Whole cell lysates were prepared by lysing cells in 1 ml of
PBS/TDS bu€er (10 mM Na2HPO4, 150 mM NaCl, 1%
Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 0.2%
sodium azide, 0.004% sodium ¯uoride, 1 mg/ml aprotinin,
1 mg/ml leupeptin, 1 mM sodium orthovanadate, pH 7.25)
and cellular debris removed by centrifugation at 12 000 g at
48C for 10 min.
Immunoblot and far Western analysis
Lysates were resolved by SDS ± PAGE (4 ± 20% Trisglycine gels) and transferred to nitrocellulose membranes
(Schleicher and Schuell, Keene, NH). For Western blot
analysis, membranes were blocked in TTBS (100 mM Tris,
229
The Nck adaptor protein is present in the nucleus
D Lawe et al
230
pH 7.5, 150 mM NaCl, 0.1% Tween 20) containing 1%
BSA for monoclonal antibodies or 5% nonfat dry milk for
polyclonal antibodies at r.t. for 1 h. Membranes were
incubated with primary antibody in TTBS containing
either 1% BSA or 5% nonfat dry milk at r.t. for 1 h,
using a concentration of 1 mg/ml for the anti-Nck mAb,
1 mg/ml for the recombinant anti-phosphotyrosine antibody conjugated to biotin, RC20B (Transduction Laboratories), 1 mg/ml for a second anti-Nck mAb (UBI, Lake
Placid, NY), 1 mg/ml for the anit-SHPTP2/SYP mAb
(Transduction Laboratories) and 1 mg/ml for the antiGAP-associated p62 polyclonal antibody (Santa Cruz
Biotechnology, Santa Cruz, CA) used to detect SAM68.
The cDNA reported for GAP-associated p62 (Wong et al.,
1992) is actually that of SAM68 (Lock et al., 1996) and
peptides used to make SAM68 antibody are derived from
this sequence. After washing with TTBS, blots were
incubated with 125I-sheep anti-mouse IgG antibody or 125Igoat anti-rabbit antibody (Dupont/NEN, Wilmington, DE)
at 56105 c.p.m./ml in the same bu€ers. The RC20B
antibody was visualized with 125I-streptavidin (Amersham,
Arlington Heights, IL) at 56105 c.p.m./ml.
For far Western analysis, membranes were blocked in 5%
nonfat dry milk in TTBS with 5 mM DTT for 1 h at 48C.
Membranes were incubated with 2.5 ± 56105 c.p.m./ml of the
relevant g-32P-labeled fusion protein in 1% BSA/TTBS with
5 mM DTT overnight at 48C and then washed three times for
10 min in cold TTBS. The GST fusion proteins were labeled
as described by Blanar and Rutter (1992) with several
modi®cations. 2 mg of fusion protein was bound to
glutathione sepharose and washed once with HMK bu€er
(20 mM Tris, pH 7.5, 1 mM DTT, 100 mM NaCl, and 12 mM
MgCl2), and then resuspended in 40 ml of HMK. 50 U of the
catalytic subunit of cAMP-dependent protein kinase (Sigma,
St Louis, MO) and 600 mCi of [g-32P]ATP (Dupont/NEN)
were added and the reaction proceeded for 30 min at 378C
until stopped by washing once with 100 ml of MDF bu€er
(50 mM Tris, pH 7.5, 100 mM NaCl, 0.5% Nonidet P-40,
1 mM DTT, and 5 mM sodium ¯uoride). Cleavage of the
labeled fusion protein from the GST was done by washing
the sepharose three times with Factor Xa cleavage bu€er
(50 mM Tris, pH 7.5, 150 mM NaCl, and 1 mM CaCl2),
resuspending in 70 ml of Factor Xa cleavage bu€er along with
Factor Xa (10% w/w fusion protein) (New England Biolabs,
Beverly, MA), and rocking overnight at r.t. The cleaved
protein was then puri®ed on a Centrisep column (Princeton
Separations, Adelphia, NJ). GST-K-Nck, GKN3S, and GST-
K were all labeled in this manner. To detect SH2 domain
binding partners, ®lters were blocked and probed as above
except 5 mg/ml of GST-K-SH2 (or 5 mg/ml of GST-K as a
control) was used in place of a g-32P-labeled probe. Far
Western analysis with the individual SH3 domain fusion
proteins was also performed this way. The binding of the
fusion proteins was detected by immunoblotting with 1 mg/ml
of anti-GST mAb (Santa Cruz Biotechnology) in 1% BSA/
TTBS for 1 h at 48C and visualized with 125I-sheep antimouse IgG antibody as described above.
Immunoprecipitations
Nuclear and cytosolic fractions were prepared as above
and adjusted to 150 mM NaCl, 1% Triton X-100, 0.5%
sodium deoxycholate, and 0.1% SDS to match the total
cell lysates. Lysates were incubated for 3 ± 12 h at 48C with
3 ± 4 mg of either anti-Nck mAb (UBI) or anti-p62
polyclonal antibody that had been prebound to 20 ml of
protein G/A agarose (Oncogene Science, Uniondale, NY)
and 5 ml of Staph A (10% w/v, Gibco/BRL). The
complexes were then centrifuged for 4 min at 3000 g, the
supernatants were removed, and the pellets were washed
four times with 1 ml of PBS/TDS. The pellets were then
resuspended in 16 SDS ± PAGE sample bu€er, boiled and
analysed by SDS ± PAGE and Westerm blotting as
described above.
For GST protein precipitations, 2 mg of GST fusion
protein was bound to glutathione sepharose beads that had
been preblocked in 5% nonfat dry milk in PBS/TDS. 150 mg
of total cell lysate was preincubated with GST bound to
beads for 1 h at 48C, pelted by centrifuation, and the
supernatant then incubated with either GST or GST fusion
protein bound to beads for 2 ± 3 h at 48C. Following
centrifugation, the pellets were washed four times in 1 ml
of PBS/TDS, and analysed by SDS ± PAGE and Western
blotting as above.
Acknowledgements
We thank Perry Hall, Marc Antonyak, David Emlet, and
David Moscatello for reading the manuscript, and Janice
Disposito for assistance with confocal microscopy. This
work was supported by T32-CA09678 (to DCL) and CA53149, NS-31102, and the Ronald McDonald Children's
Charities Foundation (to AJW).
References
Bar-Sagi D, Rotin D, Batzer A, Mandiyan V and Schlessinger J. (1993). Cell, 74, 83 ± 91.
Blake TJ, Heath KG and Langdon WY. (1993). EMBO J.,
12, 2017 ± 2026.
Blanar MA and Rutter WJ. (1992). Science, 256, 1014 ± 1018.
Chen R-H, Sarnecki C and Blenis J. (1992). Mol. Cell. Biol.,
12, 915 ± 927.
Chou MM and Hanafusa H. (1995). J. Biol. Chem., 270,
7359 ± 7364.
Chou MM, Fajardo JE and Hanafusa H. (1992). Mol. Cell.
Biol., 12, 5834 ± 5842.
Cohen GB, Ren R and Baltimore D. (1995). Cell, 80, 237 ±
248.
Davis R. (1993). J. Biol. Chem., 268, 14553 ± 14556.
Divecha N, Ban®c H and Irvine RF. (1991). EMBO J., 10,
3207 ± 3214.
Divecha N, Ban®c H and Irvine RF. (1993). Cell, 74, 405 ±
407.
Feller SM, Knudsen B and Hanafusa H. (1995). Oncogene,
10, 1465 ± 1473.
Fumagalli S, Totty NF, Hsuan JJ and Courtneldge SA.
(1994). Nature, 368, 871 ± 874.
Garrity PA, Rao Y, Salecker I, McGlade J, Pawson T and
Zipursky SL. (1996). Cell, 85, 639 ± 650.
Gout I, Dhand R, Hiles ID, Fry MJ, Panayotou G, Das P,
Truong O, Totty NF, Hsuan J, Booker GW, Campbell ID
and Water®eld MD. (1993). Cell, 75, 25 ± 36.
Guo D, Jia Q, Song H-Y, Warren RS and Donner B. (1995).
J. Biol. Chem., 270, 6729 ± 6733.
Holgado-Madruga M, Emlet DR, Moscatello DK, Godwin
AK and Wong AJ. (1996). Nature, 379, 560 ± 564.
Hu G, Milfay D and Williams LT. (1995). Mol. Cell. Biol.,
15, 1169 ± 1174.
Karin M and Hunter T. (1995). Curr. Biol., 5, 747 ± 757.
Lee C-H, Li W, Nishimura R, Zhou M, Batzer AG, Myers Jr
MG, White MF, Schlessinger J and Skolnik EY. (1993).
Proc. Natl. Acad. Sci. USA, 90, 11713 ± 11717.
Lehmann JM, Riethmuller G and Johnson JP. (1990).
Nucleic Acids Res., 18, 1048.
The Nck adaptor protein is present in the nucleus
D Lawe et al
Li W, Hu P, Skolnik EY, Ullrich A and Schlessinger J.
(1992). Mol. Cell. Biol., 12, 5824 ± 5833.
Li W, Nishimura R, Kashishian A, Batzer AG, Kim WJH,
Cooper JA and Schlessinger J. (1994). Mol. Cell. Biol., 14,
509 ± 517.
Lock P, Fumagalli S, Polakis P, McCormick F and
Courtneidge SA. (1996). Cell, 84, 23 ± 24.
Lowenstein EJ, Daly RJ, Batzer AG, Li W, Margolis B,
Lammers R, Ullrich A, Skolnik EY, Bar-Sagi D and
Schlessinger J. (1992). Cell, 70, 431 ± 442.
McCa€ey PG, Jain J, Jamieson C, Sen R and Rao A. (1992).
J. Biol. Chem., 267, 1864 ± 1871.
Meisenhelder J and Hunter T. (1992). Mol. Cell. Biol., 12,
5843 ± 5856.
Neri LM, Milani D, Bertolaso L, Stroscio M, Bertagnolo V
and Capitani S. (1994). Cell. Mol. Biol., 40, 619 ± 626.
Nishimura R, Li W, Kashishian A, Mondino A, Zhou M,
Cooper J and Schlessinger J. (1993). Mol. Cell. Biol., 13,
6889 ± 6896.
Olivier JP, Raabe T, Henkemeyer M, Dickson B, Mbamalu
G, Margolis B, Schelssinger J, Hafen E and Pawson T.
(1993). Cell, 73, 179 ± 191.
Paine PL, Austerberry CF, Desjarlais LJ and Horowitz SB.
(1983). J. Cell. Biol., 97, 1240 ± 1242.
Park D and Rhee SG. (1992). Mol. Cell. Biol., 12, 5816 ±
5823.
Pawson T. (1995). Nature, 373, 573 ± 580.
Peles E, Lamprecht R, Ben-Levy R, Tzahar E and Yarden Y.
(1992). J. Biol. Chem., 267, 12266 ± 12274.
Peraldi P, Zhao Z, Filloux C, Fischer EH and Van
Obberghen E. (1994). Proc. Natl. Acad. Sci. USA, 91,
5002 ± 5006.
Ren R, Ye Z-S and Baltimore D. (1994). Genes Dev., 8, 783 ±
795.
Richard S, Yu D, Blumer KJ, Hausladen D, Olszowy MW,
Connelly PA and Shaw A. (1995). Mol. Cell. Biol., 15,
186 ± 197.
Rivard N, McKenzie FR, Brondello JM and Pouyssegur J.
(1995). J. Biol. Chem., 270, 11017 ± 11024.
Rivera-Lezcano OM, Sameshima JH, Marcilla A and
Robbins KC. (1994). J. Biol. Chem., 269, 17363 ± 17366.
Rivero-Lezcano OM, Marcilla A, Sameshima JH and
Robbins KC. (1995). Mol. Cell. Biol., 15, 5725 ± 5731.
Romero F, Dargemont C, Pozo F, Reeves WH, Camonis J,
Gisselbrecht S and Fischer S. (1996). Mol. Cell. Biol., 16,
37 ± 44.
Ron D and Dressler H. (1992). BioTechniques, 13, 866 ± 869.
Ryan PJ and Gillespie LL. (1994). Dev. Biol., 166, 101 ± 111.
Sawyers CL, McLaughlin J, Goga A, Havlik M and Witte O.
(1994). Cell, 77, 121 ± 131.
Schlaepfer DD, Hanks SK, Hunter T and van der Geer P.
(1994). Nature, 372, 786 ± 791.
Simon MA, Dodson GS and Rubin GM. (1993). Cell, 73,
169 ± 177.
Smith DB and KS Johnson. (1988). Gene, 67, 31 ± 40.
Tanaka M, Gupta R and Mayer BJ. (1995). Mol. Cell. Biol.,
15, 6829 ± 6837.
Taylor S and Shalloway D. (1994). Nature, 368, 867 ± 871.
Taylor SJ, Ana® M, Pawson T and Shalloway D. (1995). J.
Biol. Chem., 270, 10120 ± 10124.
van der Geer P, Hunter T and Lindberg RA. (1994). Annu.
Rev. Cell. Biol., 10, 251 ± 337.
Van Etten RA, Debnath J, Zhou H and Casasnovas JM.
(1995). Oncogene, 10, 1977 ± 1988.
Wang JYJ. (1993). Curr. Opin. Genet. Dev., 3, 35 ± 43.
Wang J, Auger KR, Jarvis L, Shi Y and Roberts TM. (1995).
J. Biol. Chem., 270, 12774 ± 12780.
Wong G, Muller O, Clark R, Conroy L, Moran MF, Polakis
P and McCormick F. (1992). Cell, 69, 551 ± 558.
231