![]()
|
|
||||||||
(Received for publication, June 15, 1995; and in revised form, September 8, 1995) From the
Hepatocyte growth factor/scatter factor (HGF/SF) is a
multifunctional cytokine that induces mitogenesis, motility, invasion,
and morphogenesis of several epithelial and endothelial cell lines in
culture. The receptor for HGF/SF has been identified as the Met
tyrosine kinase. To investigate the signaling pathways that are
involved in these events, we have generated chimeric receptors
containing the extracellular domain of the colony-stimulating factor-1
(CSF-1) receptor fused to the transmembrane and intracellular domains
of the Met receptor (MET). Madin-Darby canine kidney (MDCK) epithelial
cells expressing the CSF-MET chimera dissociate and scatter in response
to CSF-1. However, cells expressing a mutant CSF-MET receptor
containing a phenylalanine substitution for tyrosine 1356 were unable
to scatter or form branching tubules following stimulation with CSF-1.
Tyrosine 1356 is essential for the recruitment of multiple substrates
including the p85 subunit of PI3-kinase, phospholipase C
Cell motility is a fundamental process required during normal
embryonic development, wound healing, inflammatory responses, and tumor
progression toward metastasis(1) . Hepatocyte growth
factor/scatter factor (HGF/SF) ( Phosphorylated tyrosine residues
within receptor tyrosine kinases provide binding sites for molecules
containing SH2 and PTB domains that act to transduce extracellular
signals to the interior of the cell (24) . Following
stimulation of cells with HGF/SF, several proteins are phosphorylated,
activated, and/or associated with the Met receptor. These include p120
GTPase-activating protein, mitogen activated protein kinase, Src,
phospholipase C To investigate the signaling
pathways that are involved in these events, we have generated chimeric
receptors containing the extracellular domain of the CSF-1 receptor
fused to the transmembrane and intracellular domains of the Met
receptor. Madin-Darby canine kidney (MDCK) epithelial cells dissociate,
scatter, and form branching tubules in response to HGF/SF. MDCK cells
expressing the CSF-MET chimera dissociate and scatter in response to
CSF-1(39) ; however, MDCK cells expressing a mutant CSF-MET
receptor containing a phenylalanine substitution for tyrosine 1356
(Y1356F) are unable to scatter or form branching tubules following
stimulation with CSF-1(29) . This has suggested that one or
more of the substrates that bind to Tyr
Figure 1:
E-cadherin, desmoplakins I/II, and ZO-1
remain insoluble following CSF-1 stimulation of MDCK cells expressing
the Y1356F CSF-MET mutant receptor. Colonies of MDCK cells expressing
the Y1356F CSF-MET mutant receptor were serum-starved for 24 h and
further incubated for 24 h in the same medium (a-d), in
medium containing HGF/SF (5 units/ml) (e-h), or in
medium containing CSF-1 (50 ng/ml) (i-l). Cells were
extracted in situ with CSK buffer and fixed with 1.75%
formaldehyde in PBS as described under ``Experimental
Procedures.'' Cells were then processed for indirect
immunofluorescence with antibodies against E-cadherin (b, f, and j), desmoplakins I/II (c, g,
and k), and ZO-1 (d, h, and l) as
primary antibodies, followed by rhodamine-coupled anti-mouse antibodies
(for E-cadherin and ZO-1) and fluorescein isothiocyanate-coupled
anti-rabbit antibodies (for desmoplakins I/II) as secondary
antibodies.
Figure 3:
The redistribution of E-cadherin and
desmoplakins I/II, cell spreading, and scatter in HGF/SF-stimulated
MDCK cells are inhibited by wortmannin. A, colonies of
serum-starved MDCK cells were incubated for 3.5 h in medium containing
0.1% Me
MDCK cells expressing
a CSF-MET chimera dissociate and scatter in a similar manner in
response to CSF-1 or HGF/SF(39) . However, MDCK cells
expressing a mutant CSF-MET receptor containing a phenylalanine
substitution for tyrosine 1356 (Y1356F) fail to scatter following
stimulation with CSF-1(29) . We show here that junctional
complexes remained intact and in an insoluble form at cell-cell
interfaces in CSF-1-stimulated MDCK cells expressing the Y1356F CSF-MET
mutant (Fig. 1, i-l).
Figure 2:
PI3-kinase is activated following HGF/SF
stimulation of MDCK cells and is inhibited by wortmannin in vitro and in vivo. A, MDCK cells starved for 48 h were
stimulated with HGF/SF (10 units/ml) for different time intervals, and
cell lysates were then immunoprecipitated with the anti-phosphotyrosine
antibody PY20. Immune complexes were adsorbed using protein A-Sepharose
and subjected to a PI3-kinase assay. The products of the reaction were
analyzed by thin layer chromatography, visualized by autoradiography,
and quantified by a PhosphorImager. B,
phosphotyrosine-containing proteins from lysates of MDCK cells were
immunoprecipitated with the PY20 antibody and assayed for PI3-kinase
activity in vitro in the presence of either 0.1%
Me
To examine the role of PI3-kinase in vivo, serum-starved MDCK cells were pretreated for 30 min
at 37 °C with either wortmannin or its solvent (Me To establish
which processes were inhibited by wortmannin, we examined the effect of
wortmannin on the redistribution of junctional complex proteins. The
spreading and changes in morphology observed in MDCK cells, following
HGF/SF stimulation for 3 h, were accompanied by a reduction in the
amount of insoluble E-cadherin and desmoplakins I/II at cell-cell
interfaces (Fig. 3A, f and g). At
this time however, the concentration of insoluble ZO-1 at cell-cell
interfaces (Fig. 3A, h) was comparable to that
observed in control unstimulated MDCK cells (Fig. 3A, d). A redistribution of ZO-1 was only apparent when cells
began to scatter (data not shown). However, in the presence of 500
nM wortmannin, all three proteins from junctional complexes
were retained in an insoluble compartment at the plasma membrane (Fig. 3A, j-l). Thus, the ability of
wortmannin to inhibit MDCK cell spreading and scatter correlated with
the retention of insoluble junctional complexes and tight cell-cell
interactions, suggesting that PI3-kinase activity is required for the
redistribution of junctional complex proteins and cell dissociation
induced by HGF/SF in MDCK cells. Chemotaxis of some cell types
transduced by the PDGF-
Figure 4:
pp70
To establish whether pp70
Figure 5:
Inhibition of pp70
HGF/SF is a multifunctional cytokine that stimulates
dissociation, scatter, and morphogenesis of epithelial
cells(2, 4, 5) . The response of MDCK cells
to HGF/SF can be visualized first as cell spreading (after 2-3
h), when cells are still associated and present within the colony (Fig. 3A), followed by cell dissociation (after
4-6 h) and cell scatter (from 6 h) (Fig. 1). Breakdown of
cell contacts is a prerequisite for cell dissociation, and we have
shown that dissociation of MDCK cells in response to HGF/SF is
concomitant with the loss of stable insoluble junctional complexes at
sites of cell contact. Interestingly, following HGF/SF stimulation of
MDCK cells, a redistribution of E-cadherin and desmoplakins I/II, which
are components of adherens junctions and desmosomes, is observed prior
to the redistribution of ZO-1, a component of tight junctions ( Fig. 1and 3A). This may reflect the ability of HGF/SF
to stimulate the phosphorylation of MDCK cells expressing a CSF-MET mutant
receptor (Y1356F) fail to stimulate scatter (29) or
redistribute junctional complexes (Fig. 1). Tyrosine 1356 in the
carboxyl terminus of the Met receptor is essential for association with
PI3-kinase, phospholipase C The ability of wortmannin to
inhibit HGF/SF-induced spreading and scatter of MDCK cells correlated
directly with the extent of PI3-kinase inhibition in vivo,
thus supporting a crucial role for PI3-kinase activity in the
dissociation and scatter of MDCK cells. The concentrations of
wortmannin required for the inhibition of cell spreading,
redistribution and solubilization of junctional complex proteins, and
cell scatter were higher than that reported for the inhibition of
diverse biological responses involving membrane ruffling (70) ,
histamine secretion(55) , respiratory
burst(53, 71) , or glucose transport (72) (between 50-100 nM). Although high
concentrations of wortmannin (in the µM range in
vitro) have been reported to inhibit myosin light chain
kinase(55, 73) , an enzyme thought to be involved in
cell motility, treatment of MDCK cells with an inhibitor of myosin
light chain kinase (ML-9) had no effect on HGF/SF-induced MDCK cell
scatter (57) . Moreover, our data demonstrated that the canine
PI3-kinase was sensitive to wortmannin in vitro, with an
IC Wortmannin is also an unstable compound
when maintained at 37 °C (54) . Thus, the ability of MDCK
cells treated with wortmannin to begin to spread at 7 h post HGF/SF
stimulation (Fig. 3B) may reflect the instability of
wortmannin. Consistent with this possibility, MDCK cells stimulated
with HGF/SF for 7 h remained as tight cell colonies when the medium
containing wortmannin and HGF/SF was replaced at 2-h intervals (data
not shown). Although MDCK cells showed some signs of toxicity under
these conditions, this suggests that PI3-kinase is essential for cell
spreading. We therefore conclude that PI3-kinase is required for MDCK
cell dissociation and thus, scatter, following stimulation with HGF/SF,
but due to the instability of wortmannin, we cannot rule out that other
factors are involved in these events. The pp70 The requirement for a functional
PI3-kinase has been implicated in actin reorganization at the plasma
membrane (membrane ruffling) stimulated by PDGF (70, 76) and insulin(76) . Moreover,
PI3-kinase is involved in the activation of the small GTP-binding
protein Rac(77) , which is required for membrane ruffling in
response to growth factors (76, 78) . Interestingly,
microinjection of MDCK cells with a dominant negative mutant of Rac
(N17Rac1) inhibits cell spreading and actin reorganization induced by
HGF/SF(79) , thus supporting a crucial role for Rac in these
events. In addition, Ras is also essential for the dissociation and
scatter of MDCK cells(79, 80) . Expression of a
dominant negative mutant Ras protein (N17Ras) (80) or the
injection of a neutralizing antibody for Ras (Y13-259) (79) block HGF/SF-induced cell dissociation and scatter,
whereas microinjection of an activated Ras (V12H-Ras) in MDCK cells
promotes cell spreading in the absence of HGF/SF(79) .
Furthermore, GTP-bound Ras interacts with PI3-kinase and may contribute
to its activation(81, 82) . Thus, we propose that
activation of PI3-kinase in MDCK cells following stimulation of the Met
receptor by HGF/SF promotes cell dissociation, which is independent
from the activation of pp70
Volume 270,
Number 46,
Issue of November 17, 1995 pp. 27780-27787
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
, and
Grb2. In this study, we have investigated the role of PI3-kinase and a
downstream target of PI3-kinase, pp70
, in the induction
of MDCK cell scatter in response to HGF/SF. Our results demonstrate
that following stimulation with HGF/SF, activation of PI3-kinase but
not pp70
is essential for MDCK cell scatter.
)is a multifunctional
cytokine with activities on a wide variety of normal and neoplastic
cells. HGF/SF is a mitogen, dissociation, and motility factor for many
epithelial cells (2, 3, 4) and stimulates
tubulogenesis of tubular epithelial cells (5) as well as the
invasion of carcinoma cells(6) . In vivo, HGF/SF is a
potent angiogenic factor (7, 8) and is involved in
organ regeneration (9) and
tumorigenesis(10, 11) . A high affinity receptor for
HGF/SF has been identified as the product of the met proto-oncogene(12, 13) , which encodes a receptor
tyrosine kinase originally isolated as the tpr-met oncogene(14, 15) . The Met receptor is
predominantly expressed in epithelial cells in culture (15, 16, 17) and in epithelium in vivo (18, 19). (
)The mature form of the Met receptor is a
heterodimeric protein of 190 kDa, which consists of a 45-kDa
extracellular
-subunit linked by disulfide bonds to a 145-kDa
-subunit that spans the membrane and contains the catalytic kinase
domain(16, 20, 21, 22) . Binding of
HGF/SF induces activation of the kinase and auto/transphosphorylation
of the receptor (13) on specific tyrosine residues in the
chain(12, 23) ., phosphatidylinositol 3-kinase (PI3-kinase),
Grb2(25, 26, 27, 28, 29, 30) ,
Ras(31) , focal adhesion kinase(32) ,
-catenin,
plakoglobin(33) , and the Shc adaptor protein(34) . The
Met receptor tyrosine kinase is highly phosphorylated on two tyrosine
residues (1234 and 1235) within the kinase domain (35, 36) that are essential for the catalytic activity
of the receptor(29, 35) . In addition, tyrosine 1356
within the carboxyl terminus of the
-subunit of the Met receptor
is phosphorylated (29) and is essential for the recruitment of
multiple substrates including the p85 subunit of PI3-kinase,
phospholipase C, and
Grb2(29, 30, 37, 38) . Although a
role for many of these signal transduction pathways has been
established in cell mitogenesis, their role in dissociation and scatter
of epithelial cells is unknown.
are required for
the induction of MDCK cell scatter following stimulation with HGF/SF.
Our results demonstrate that following stimulation with HGF/SF,
activation of PI3-kinase but not phospholipase C
or
pp70
, a downstream target of PI3-kinase, is essential for
MDCK cell scatter.
Cell Lines
MDCK epithelial cells were maintained
in Dulbecco's modified Eagle's medium supplemented with 10%
fetal bovine serum and 50 µg/ml gentamycin (all from Life
Technologies, Inc.). The MDCK stable cell line expressing the CSF-MET
mutant receptor Y1356F has been described previously(29) .Immunofluorescence Labeling
Cells (1
10
cells/well) were seeded on glass coverslips (Bellco
Glass Inc.) in 24-well culture dishes (Nunc). 24 or 48 h later, cells
were serum-starved in 0.2% fetal bovine serum for 24 h and stimulated
with HGF/SF (5 or 10 units/ml) or CSF (50 ng/ml) at 37 °C for the
chosen time intervals. When appropriate, cells were preincubated with
wortmannin/Me
SO (0.1%) (Sigma), rapamycin/ETOH (0.1%)
(Calbiochem), staurosporine/Me
SO (0.1%) (Sigma),
U-73122/Me
SO (0.1%), or U-73343/Me
SO (0.1%)
(Biomol, Plymouth Meeting, PA) for 30 min at 37 °C at
concentrations indicated in the text. Following stimulation with
HGF/SF, cells were washed in PBS and treated with CSK buffer (10 mM PIPES, pH 7.0, 300 mM sucrose, 50 mM NaCl, 3
mM MgCl
, 0.5% Triton X-100, 1 mM phenylmethylsulfonyl fluoride, 0.1 mg/ml DNase, and 0.1 mg/ml
RNase A) for 20 min at 4 °C, followed by fixation with 1.75%
formaldehyde (in PBS) for 5-10 min at room temperature as
described(40, 41) . For phase contrast pictures, cells
were fixed for 30 min in 0.2% glutaraldehyde (JBS-Chem, Montreal,
Canada) at 4 °C. Primary antibodies used for staining proteins from
junctional complexes were: a 1/50 dilution of a mouse monoclonal
antibody, 3G8, directed against E-cadherin; a 1/200 dilution of a
rabbit polyclonal antibody, DPI/II, directed against desmoplakins I and
II (both kindly provided by Dr. M. Pasdar, University of Alberta,
Edmonton, Canada); and a 1/200 dilution of a mouse monoclonal antibody,
R26-4C, recognizing ZO-1 (Developmental Studies Hybridoma Bank,
University of Iowa). These primary antibody reactions were revealed
using the appropriate fluorescein isothiocyanate-labeled anti-rabbit
antibodies (1/50 dilution) (Jackson ImmunoResearch Laboratories Inc.)
or rhodamine-conjugated anti-mouse antibodies (1/200 dilution)
(Boehringer Mannheim). All incubations with antibodies were for 30 min
at room temperature, and cells were washed with PBS after each
incubation. The cells were mounted onto slides in Immuno-Fluore medium
(ICN), and observations were made on a Nikon Labophot-2 epifluorescence
microscope. Images were photographed using Kodak TMY-400 films.Phosphatidylinositol 3-Kinase Assay
MDCK cells (4
10
) were seeded in 100-mm dishes and then
serum-starved for 48 h in 0.2% fetal bovine serum. Monolayers were
generally 70-80% confluent and contained mid-size colonies. When
appropriate, cells were preincubated with wortmannin/Me
SO
(0.1%) for 30 min at 37 °C prior to stimulation with HGF/SF (10
units/ml) on ice for 10 min followed by incubation at 37 °C for the
chosen time intervals. Following stimulation, cells were lysed in a
buffer containing 50 mM HEPES, pH 7.5, 1% Triton X-100, 10%
glycerol, 150 mM NaCl, 1.5 mM MgCl
, 1
mM EDTA, 10 mM
Na
P
O
, 100 mM NaF, 1
mM phenylmethylsulfonyl fluoride, 1 mM
Na
VO 10 µg/ml aprotinin, and 10 µg/ml
leupeptin on ice for 30 min. The PI3-kinase assay was performed
essentially as described with minor modifications (37) .
Immunoprecipitation was performed on clarified cell lysates (15,000 rpm
for 15 min at 4 °C) using the monoclonal antibody PY20 against
phosphotyrosine-containing proteins (Signal Transduction Laboratories
Inc.). Equal amounts of proteins were incubated for 1.5 h with PY20
alone followed by a 1.5-h incubation in the presence of protein
A-Sepharose (20% suspension) (Pharmacia) at 4 °C with mixing. The
Sepharose-immune complexes were then subjected to several washes with
ice-cold solutions: three times with lysis buffer; once with PBS; once
with 0.1 M Tris-HCl, pH 7.5, 0.5 M LiCl; once with
distilled water; once with 20 mM Tris-HCl, pH 7.5, 100
mM NaCl, 1 mM EDTA; and once with kinase buffer (20
mM Tris-HCl, pH 7.5, 100 mM NaCl, and 0.5 mM EGTA). The beads were resuspended in 50 µl of kinase buffer
containing 0.2 mg/ml of phosphatidylinositol (sonicated 15 min at 4
°C; Sigma) preincubated for 10 min at room temperature, at which
time, for in vitro assays, wortmannin or an equivalent amount
of Me
SO (0.1%) were added. Then 20 µCi of
[-
P]ATP and 20 mM MgCl
were added for 10 min at room temperature. Reactions were
terminated by adding 150 µl of chloroform/methanol/11.6 M HCl (50:100:1), and phosphatidylinositol was extracted with 100
µl of chloroform. The organic phase was washed with methanol/1 M HCl (1:1) and lyophilized. Phosphatidylinositol, resuspended
in 15 µl of chloroform, was spotted on a silica gel 60 thin layer
chromatography plate (Merck) and resolved in chloroform/methanol/28%
ammonium hydroxide/water (86:76:10:14) for 45 min. Phosphorylated
products were visualized by autoradiography and quantified by the Fujix
Bio-Imaging Analyzer Bas 1000.S6 Kinase Immunoprecipitation, Western Blotting, and
in Vitro Kinase Assay
MDCK cells (1.2
10
) were plated in 60-mm dishes and serum-starved as
described above. When appropriate, cells were preincubated with
rapamycin/ETOH (0.1%) or with wortmannin/Me
SO (0.1%) for 30
min at 37 °C prior to stimulation. For immunoprecipitations, cells
were rapidly washed with ice-cold PBS and lysed in 0.4 ml of lysis
buffer containing 50 mM Tris-HCl, pH 7.4, 1% Triton X-100, 100
mM NaCl, 5 mM EDTA, 50 mM NaF, 40
mM
-glycerophosphate, 1 mM
Na
VO 1 mM phenylmethylsulfonyl
fluoride, 10 µg/ml aprotinin, and 10 µg/ml leupeptin on ice for
30 min. As described above and in (42) , equal amounts of
protein from clarified cell lysates were immunoprecipitated with a
rabbit polyclonal antibody recognizing the carboxyl terminus of
p70
(Santa Cruz Biotechnology, Inc.) for 2 h. Protein
A-Sepharose was then added for 2 h. Immune complexes were washed four
times with lysis buffer, eluted by boiling in Laemmli sample buffer,
subjected to SDS-polyacrylamide (8%) gel electrophoresis, and
transferred to nitrocellulose (Schleicher & Schuell). For Western
blotting, membranes were blocked for 1 h at room temperature with 3%
bovine serum albumin in TBST buffer (10 mM Tris-HCl, pH 7.5,
2.5 mM EDTA, 150 mM NaCl, and 0.1% Tween 20), washed
several times in TBST buffer, and probed with anti-p70
antibodies (1:200 dilution) for 1 h at room temperature. The
primary antibody reaction was detected with horseradish
peroxidase-coupled protein A (1:5000 dilution) by the enhanced
chemiluminescence method (Amersham Corp.) according to the
manufacturer's recommendations. For the in vitro kinase
assay, p70
was immunoprecipitated as above, and the
immune complexes were washed three times with lysis buffer and once
with kinase buffer (20 mM HEPES, pH 7.4, 10 mM MgCl
, 1 mM dithiothreitol, 10 mM
-glycerophosphate, 1 mM phenylmethylsulfonyl fluoride, 10
µg/ml aprotinin, and 10 µg/ml leupeptin). The beads were
resuspended in 25 µl of kinase buffer containing 10 µCi of
[-
P]ATP and 0.2 mM of the S6
peptide RRRLSSLRA (Santa Cruz Biotechnology, Inc.) and incubated at 30
°C for 20 min. The reactions were terminated by application to 1
cm
of phosphocellulose p81 paper (Whatman). The filters
were then extensively washed (in 500 ml for 15 min: four times in 1%
phosphoric acid, twice with distilled water, and once in ETOH) and
counted in a liquid scintillation counter. Blank reactions where the S6
peptide was omitted were also performed to control for background
levels of radioactivity.
A CSF-MET Mutant Chimera, Y1356F, Fails to Promote Cell
Dissociation and Redistribution of Junctional Protein Complexes in
Response to CSF-1
MDCK cells form discrete colonies when
sparsely seeded on plastic or on glass coverslips. Following
stimulation of MDCK cell colonies maintained in 0.2% serum with HGF/SF,
we have observed that MDCK cells first spread (2-3 h), dissociate
(4-6 h), and then adopt a fibroblastoid cell morphology as they
begin to scatter (from 6 h) ( Fig. 1and Fig. 3A). In MDCK epithelial cells, the lateral plasma
membranes are connected by the formation of junctional complexes,
including tight junctions, zonula adherens, and
desmosomes(43) . Adherens type junctions and desmosomes play
important roles in maintaining cell-cell
adhesion(44, 45) . Therefore, scattering of MDCK cells
requires the breakdown of these intercellular adhesions. In polarized
epithelial cells, proteins of tight junctions and zonula adherens are
coupled to the actin filaments, whereas desmosomal proteins are
associated with intermediate filaments. In this context, these proteins
are insoluble, whereas in the absence of cell-cell contact, they are
redistributed into the soluble compartment of the
cytoplasm(40, 46, 47, 48, 49, 50) .
To examine the dissociation of MDCK cells in response to HGF/SF, we
have analyzed the localization of junctional complex proteins by
indirect immunofluorescence microscopy. We have demonstrated that in
colonies of MDCK cells where the membranes were solubilized with CSK
buffer, E-cadherin (from zonula adherens), desmoplakins I/II (from
desmosomes), and ZO-1 (from tight junctions) were insoluble and
distributed uniformly along regions of cell-cell contact (Fig. 1, a-d). However, scatter of MDCK cells in
response to HGF/SF resulted in the redistribution of each of these
proteins, and as a consequence, they were no longer organized in
insoluble complexes associated with the cytoskeleton at the plasma
membrane (Fig. 1, e-h).
SO (a-d), preincubated in medium
containing 0.1% Me
SO for 30 min followed by stimulation
with HGF/SF (10 units/ml) for 3 h (e-h), or preincubated
in medium containing 500 nM wortmannin (in 0.1%
Me
SO) for 30 min followed by stimulation with HGF/SF for 3
h (i-l). Cells were extracted in situ with CSK
buffer, fixed with 1.75% formaldehyde in PBS, and processed for
indirect immunofluorescence with antibodies against E-cadherin (b, f, and j), desmoplakins I/II (c, g, and k), and ZO-1 (d, h, and l). B, colonies of serum-starved MDCK
cells were incubated for 7.5 h in medium containing 0.1%
Me
SO (a), preincubated in medium containing 0.1%
Me
SO for 30 min followed by stimulation with HGF/SF (10
units/ml) for 7 h (b), preincubated in medium containing 1
µM wortmannin (in 0.1% Me
SO) for 30 min
followed by stimulation with HGF/SF for 7 h (c), preincubated
in medium containing 5 nM staurosporine (in 0.1%
Me
SO) for 30 min followed by stimulation with HGF/SF for 7
h (d), preincubated in medium containing 2 µM U-73122 (e), or 2 µM U-73343 (f)
(in 0.1% Me
SO) for 30 min followed by stimulation with
HGF/SF for 7 h. Cells were fixed in 0.2% glutaraldehyde in
PBS.
Wortmannin Inhibits PI3-Kinase Activity and Scatter of
HGF/SF-stimulated MDCK Cells
Chemotaxis transduced by the
PDGF-
receptor requires the binding sites for
PI3-kinase(51, 52) , suggesting that this substrate
may play an important role in this event. PI3-kinase is activated
following treatment of A-549 (lung carcinoma) cells with
HGF/SF(25) . Following HGF/SF stimulation of MDCK cells,
PI3-kinase activity was increased 4-fold within 1 min, and this level
of activation was maintained over 10 min (Fig. 2A). To
investigate the involvement of PI3-kinase in the scatter of MDCK cells
in response to HGF/SF, we have used the potent inhibitor wortmannin,
which is a fungal metabolite that directly binds to and inhibits the
p110 subunit of PI3-kinase(53, 54, 55) . When
assayed in vitro, wortmannin inhibited the canine MDCK cell
PI3-kinase immunoprecipitated using antibodies to phosphotyrosine with
an IC (50% inhibitory concentration) below 10 nM (Fig. 2B).
SO or various concentrations of wortmannin for 10 min
prior to the kinase reaction. C, serum-starved MDCK cells were
preincubated with either 0.1% Me
SO or wortmannin (10 nM to 10 µM) for 30 min at 37 °C, followed by
stimulation with HGF/SF (10 units/ml) for 1 min at 37 °C. Cells
were lysed, and phosphotyrosine-containing proteins were
immunoprecipitated with the PY20 antibody. Immune complexes were then
processed as described above. The position of phosphatidylinositol
3-phosphate (PI(3)P) is indicated.
SO
0.1%) and stimulated with HGF/SF, and PI3-kinase was immunoprecipitated
with anti-phosphotyrosine antibodies and assayed in vitro.
Wortmannin, at concentrations of 100 nM and 1 µM,
reduced the level of PI3-kinase activity in HGF/SF-stimulated MDCK
cells by 35 and 65% respectively (Fig. 2C). To evaluate
the effect of wortmannin on HGF/SF-induced cell scatter, colonies of
serum-starved MDCK cells were pretreated for 30 min at 37 °C with
wortmannin (10 nM-1 µM) and then stimulated with
HGF/SF. MDCK cells, following stimulation with HGF/SF, changed shape
and flattened by 3 h (Fig. 3A, e), whereas
MDCK cells in the presence of 500 nM wortmannin remained as
tight colonies (Fig. 3A, i). Moreover, in the
presence of 1 µM wortmannin, MDCK cells stimulated with
HGF/SF for 7 h had flattened (Fig. 3B, c) but
were inhibited in their ability to scatter when compared with control
HGF/SF-stimulated MDCK cells (Fig. 3B, b). The
ability of wortmannin to inhibit both cell spreading and scatter was
concentration-dependent and correlated with its ability to inhibit
PI3-kinase activity in vivo (data not shown).
and epidermal growth factor receptors
requires phospholipase C(51, 56) ; however, the
inhibition of protein kinase C, a downstream target of phospholipase
C
, did not inhibit MDCK cell scatter in response to
HGF/SF(57) . To investigate the involvement of phospholipase
C
in HGF/SF-induced MDCK cell scatter, we used an inhibitor of
phospholipase C (U-73122; IC
of 1-2 µM) (56, 58, 59) and the protein kinase C
inhibitor staurosporine (IC
of 0.7
nM)(60) . Colonies of serum-starved MDCK cells were
pretreated with Me
SO (0.1%), staurosporine (1
nM-20 nM), U-73122 (0.25-2 µM), or
an inactive analogue of U-73122 (U-73343; 0.25-2 µM)
for 30 min at 37 °C. MDCK cells in the presence of staurosporine (Fig. 3B, d, and data not shown), the
phospholipase C inhibitor U-73122 (Fig. 3B, e,
and data not shown), the inactive analogue U-73343 (Fig. 3B, f, and data not shown), or
Me
SO (0.1%) (Fig. 3B, b) scattered
following stimulation with HGF/SF for 7 h. Therefore, in contrast to
PI3-kinase, phospholipase C and protein kinase C are not essential
for MDCK cell spreading and scatter.
pp70
To further characterize the involvement of
PI3-kinase-regulated pathways in HGF/SF-induced cell scatter, we have
investigated the role of pp70Is Activated Following
HGF/SF Stimulation of MDCK Cells But Is Not Required for
Scatter
, a potential downstream
target of PI3-kinase(61, 62) . Activation of
pp70
is associated with its phosphorylation on Ser/Thr
residues (63, 64) and its decreased electrophoretic
mobility. Phosphorylation of pp70
was increased following
stimulation of MDCK cells with HGF/SF (Fig. 4).
Immunoprecipitation of pp70
, followed by immunoblotting
with a specific p70
antibody, revealed a shift in
mobility of pp70
that was maximal at 20 min following
stimulation with HGF/SF and decreased by 2 h (Fig. 4A, top). In agreement with this result, an increase in
pp70
in vitro kinase activity, when assayed
against an S6 peptide, was observed following stimulation of MDCK cells
by HGF/SF for 20 min (Fig. 4A, bottom). To
establish if pp70
activity was inhibited by wortmannin in
MDCK cells, pp70
activation was assayed in serum-starved
cells pretreated with wortmannin (10 nM-10 µM) or
Me
SO (0.1%) for 30 min and then stimulated with HGF/SF for
20 min. In the presence of 100 nM wortmannin, HGF/SF-induced
phosphorylation of pp70 was inhibited, and pp70
activity was reduced to 30% of control (HGF-stimulated MDCK
cells), whereas maximal inhibition, where pp70
activity
was reduced to 10% of control, was observed in the presence of 1
µM wortmannin (Fig. 4B, bottom).
is activated following
HGF/SF stimulation of MDCK cells and is inhibited by wortmannin or
rapamycin. A, serum-starved MDCK cells were stimulated with
HGF/SF (10 units/ml) for different time intervals, and proteins from
cell lysates were immunoprecipitated with antibodies against
p70
. Immune complexes were adsorbed using protein
A-Sepharose and subjected to SDS-polyacrylamide (8%) gel
electrophoresis, transferred onto nitrocellulose and immunoblotted with
an anti-p70
antibody (top), or subjected to an in vitro kinase assay using S6 peptide as a substrate as
described under ``Experimental Procedures'' (bottom). The results are reported as counts incorporated into
S6 peptide. B, serum-starved MDCK cells were preincubated with
either 0.1% Me
SO or various concentrations of wortmannin
for 30 min at 37 °C, followed by further incubation in
Me
SO or by stimulation with HGF/SF (10 units/ml) for 20 min
at 37 °C. Cells were lysed, pp70 was
immunoprecipitated, and immune complexes were then processed as
described above. C, serum-starved MDCK cells were preincubated
with either 0.1% ETOH or various concentrations of rapamycin for 30 min
at 37 °C, followed by further incubation in ETOH or by stimulation
with HGF/SF (10 units/ml) for 20 min at 37 °C. Cells were lysed and
pp70
-immune complexes were processed as described
above.
is essential for MDCK cell
dissociation and scatter induced by HGF/SF, serum-starved MDCK cells
were pretreated with the macrolide antibiotic rapamycin, which inhibits
the PI3-kinase-related protein RAFT1 (65) and as a consequence
its downstream target pp70
(66, 67, 68) . HGF/SF-induced pp70
phosphorylation (Fig. 4C, top) and in vitro kinase activity (Fig. 4C, bottom) were inhibited equally by rapamycin (1-50 ng/ml)
or wortmannin (1-10 µM) (Fig. 4B).
However, preincubation of MDCK cells with 20 ng/ml of rapamycin for 30
min followed by stimulation with HGF/SF did not inhibit scattering of
MDCK cells (Fig. 5i). Moreover, rapamycin pretreatment
did not inhibit the redistribution or solubilization of E-cadherin,
desmoplakins I/II, and ZO-1 following stimulation of MDCK cells with
HGF/SF (Fig. 5, j-l). Therefore, although
pp70
is activated following stimulation with HGF/SF and
is a possible downstream target of PI3-kinase in MDCK cells, it is not
required for scatter of MDCK cells in response to HGF/SF.
activation in HGF/SF-stimulated MDCK cells by rapamycin does not block
cell spreading and dissociation. Colonies of serum-starved MDCK cells
were incubated for 24 h in medium containing 0.1% ETOH (a-d), preincubated in medium containing 0.1% ETOH for
30 min and then stimulated by HGF/SF (10 units/ml) for 24 h (e-h), or preincubated in medium containing 20 ng/ml of
rapamycin for 30 min and then stimulated with HGF/SF for 24 h (i-l). Cells were extracted in situ with CSK
buffer and fixed with 1.75% formaldehyde in PBS and then processed for
indirect immunofluorescence with antibodies against E-cadherin (b, f, and j), desmoplakins I/II (c, g, and k), and ZO-1 (d, h, and l).
-catenin and
plakoglobin(33) , which could contribute to the destabilization
of adherens junctions and desmosomes but not tight junctions, in which
these proteins are absent., and
Grb2(29, 30, 37, 38) , suggesting
that at least one of these signaling pathways is required for cell
dissociation and scatter. However, a mutant CSF-MET receptor containing
a substitution of a histidine residue for the asparagine downstream
from tyrosine 1356 (N1358H), which failed to bind only the Grb2 adaptor
protein, stimulated MDCK cell scatter in response to ligand, (
)demonstrating that association of Grb2 with the Met
receptor is not essential. Because binding sites for PI3-kinase (51, 52) or phospholipase C (51) in the
PDGF receptor-
are important for PDGF-BB-induced chemotaxis and
because phospholipase C is involved in epidermal growth
factor-mediated chemotaxis(56) , we have investigated the role
of these signaling pathways in HGF/SF-induced cell dissociation and
scatter of MDCK epithelial cells. We have shown that PI3-kinase plays
an essential role in HGF/SF-mediated scatter of MDCK cells. In the
presence of wortmannin, a potent inhibitor for PI3-kinase, MDCK cells
stimulated with HGF/SF for 3 h remained as tight colonies and retained
insoluble E-cadherin, desmoplakins I/II, and ZO-1 at cell-cell
interfaces. Conversely, control MDCK cell colonies contained cells with
a flattened appearance, which showed a redistribution of junctional
complex proteins involving E-cadherin and desmoplakins I/II, whereas,
as discussed above, ZO-1 from tight junctions was maintained while
cells were in contact (Fig. 3A)(69) . Moreover,
concurrent with the ability of wortmannin to inhibit the spreading of
MDCK cells 3 h after HGF/SF stimulation, wortmannin also inhibited the
scatter of MDCK cells at 7 h (Fig. 3B). In contrast, an
inhibitor of phospholipase C (U-73122) or protein kinase C
(staurosporine) at concentrations equal or higher than their respective
IC
did not inhibit scatter of MDCK cells in response to
HGF/SF (Fig. 3B).
below 10 nM, which is comparable to other
studies(53, 54, 55, 72) . In
addition, wortmannin is a lipophilic compound and is expected to
inactivate PI3-kinase localized at the plasma membrane. However, the
Met receptor is localized to the basolateral surface of MDCK
cells(74) . Thus the application of wortmannin to the apical
surface of a colony of tightly associated MDCK cells may be unable to
efficiently inactivate the Met-stimulated PI3-kinase localized to the
basolateral compartment.
has
been described as a downstream target of PI3-kinase in various cell
types(61, 62) . We show that pp70
is
activated following stimulation of MDCK cells with HGF/SF and that this
activity is also inhibited by wortmannin, suggesting that pp70
is a downstream target of PI3-kinase in MDCK cells (Fig. 4). pp70
is required for the progression
through G1 in response to serum and growth factors in a variety of
cells (66, 67, 68, 75) . However,
although pretreatment of MDCK cells with rapamycin inhibited the
activation of HGF/SF-induced pp70
, this had no effect on
MDCK cell spreading, redistribution of junctional complex proteins, or
cell scatter (Fig. 5). These results suggest that although
PI3-kinase is required for cell dissociation and scatter following
stimulation of MDCK cells with HGF/SF, this is independent from the
activation of pp70
.
, but may involve the small
GTP-binding proteins Rac and Ras. The relationship between PI3-kinase
and these proteins in MDCK cells is currently under investigation.
)
)
)
We thank members of the Park laboratory for critical
reading of the manuscript and Dr. Allison Haggarty for the use of the
fluorescent microscope.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
![]()
CiteULike
Complore
Connotea
Del.icio.us
Digg
Reddit
Technorati What's this?
This article has been cited by other articles:
![]() |
K. Hellmuth, S. Grosskopf, C. T. Lum, M. Wurtele, N. Roder, J. P. von Kries, M. Rosario, J. Rademann, and W. Birchmeier Specific inhibitors of the protein tyrosine phosphatase Shp2 identified by high-throughput docking PNAS, May 20, 2008; 105(20): 7275 - 7280. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. de Graauw, I. Tijdens, M. B. Smeets, P. J. Hensbergen, A. M. Deelder, and B. van de Water Annexin A2 Phosphorylation Mediates Cell Scattering and Branching Morphogenesis via Cofilin Activation Mol. Cell. Biol., February 1, 2008; 28(3): 1029 - 1040. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Zeng, M.-Z. Zhang, A. B. Singh, R. Zent, and R. C. Harris ErbB4 Isoforms Selectively Regulate Growth Factor induced Madin-Darby Canine Kidney Cell Tubulogenesis Mol. Biol. Cell, November 1, 2007; 18(11): 4446 - 4456. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Kimura, T. Sakisaka, T. Baba, T. Yamada, and Y. Takai Involvement of the Ras-Ras-activated Rab5 Guanine Nucleotide Exchange Factor RIN2-Rab5 Pathway in the Hepatocyte Growth Factor-induced Endocytosis of E-cadherin J. Biol. Chem., April 14, 2006; 281(15): 10598 - 10609. [Abstract] [Full Text] [PDF] |
||||
|
|