| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
subunit and the tyrosine phosphatase SHP-2
INSERM U151, IFR 31, CHU Rangueil, 31403 Toulouse, France;
* Département d'Anatomie et de Biologie Cellulaire, Faculté de Médecine, Université de Sherbrooke, Canada; and
Howard Hughes Medical Institute, The University of Michigan, Medical School, Ann Arbor, Michigan 48109-0606, USA
1Correspondence: INSERM U151, Institut Louis Bugnard, 1 avenue J. Poulhès, CHU Rangueil, 31403 Toulouse, Cédex 4, France. E-mail: louis.buscail{at}rangueil.inserm.fr
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-guanidinoglutaric acid. We identified the NOS
activated by the peptide as the neuronal isoform: the expression of the
C415A neuronal NOS mutant inhibited both CCK-induced stimulation of NOS
activity and cell proliferation. These two effects were also inhibited
after expression of the C459S tyrosine phosphatase SHP-2 mutant and the
ßARK1 (495–689) sequestrant peptide, indicating the requirement of
activated SHP-2 and G-ß
subunit. Kinetic analysis (Western blot
after coimmunoprecipitation and specific SHP-2 activity) revealed that
in response to CCK-treatment, SHP-2 associated to G-ß1 subunit,
became activated, and then dephosphorylated the neuronal NOS through a
direct association. These data demonstrate that the neuronal NOS is
implicated in proliferative effect evoked by CCK. A novel growth
signaling pathway is described, involving the activation of neuronal
NOS by dephosphorylation of tyrosyl residues.—Cordelier, P.,
Estève, J.-P., Rivard, N., Marletta, M., Vaysse, N., Susini, C.,
Buscail, L. The activation of neuronal no synthase is mediated by
G-protein ß subunit and the tyrosine phosphatase SHP-2.
Key Words: cholecystokinin • nitric oxide • cell growth • Gß complex
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. CCK has also been found to be expressed in brain and
in nerve endings in many parts of the body, and thus acts as a
neuropeptide in both the central and peripheral nervous systems. In the
central nervous system, the CCK-8 form that possesses the complete
biological activity of CCK is mainly represented and is involved in
behavioral processes such as anxiety or satiety.
CCK is known to stimulate pancreatic growth in vitro and
in vivo in the mouse, rat, or hamster (2)
. In
rat exocrine pancreas, this effect has been shown to be mediated by the
CCKA receptor subtype (3
, 4)
. In
this latter cellular model, CCKA receptors
interact with G-proteins (guanine nucleotide binding proteins) and
thereby mediate the activation of different intracellular pathways,
leading to 1) the stimulation of intracellular calcium
mobilization (via phospholipase C/inositol phosphate systems),
2) an increase of cyclic AMP, cyclic GMP (cGMP), arachidonic
acid, and phosphatidic acid levels, and 3) the stimulation
of tyrosine kinase or tyrosine phosphatase activities (1
, 5)
. However, the mechanism(s) involved in the proliferative
effect induced by CCK through CCKA receptors is
not completely elucidated.
We recently demonstrated that CHO-K1 cells expressed endogenous
CCKA receptors (6)
. In this model we
also observed that CCK-8 stimulated cell proliferation. This effect
occurred through an activation of a soluble guanylate cyclase via a
pertussis toxin-sensitive Gi/G0 protein, followed by activation of the
protein kinase G and the MAP kinase system (6)
. It
represented a new example of a positive effect of CCK on cell
proliferation that is mediated by CCKA receptors.
It was an additional observation of the positive effect of cGMP on cell
proliferation, the nucleotide being more frequently responsible for an
inhibition of cell proliferation or migration.
The potent activator of soluble guanylate cyclase is nitric oxide (NO).
NO is a ubiquitous endogenous messenger molecule that participates in a
variety of physiological and pathophysiological processes. NO is
produced by NO synthases (NOS), which catalyze the oxidation of
L-arginine to form L-citrulline and NO (7
8
9)
. NO
synthases constitute at least three distinct isoforms, which are the
products of individual genes. All catalyze the formation of NO. Two of
these isoforms, neuronal (nNOS) and endothelial (eNOS) NOS, are
constitutively expressed (including resting cells) and regulated by
Ca2+/calmodulin. The third one is inducible
(iNOS) in a number of cell types, including macrophages, after exposure
to bacterial lipopolysaccharide in conjunction with cytokines that
stimulate iNOS protein formation (9)
. NO is known to act
as a messenger in neuronal transmission, control of vascular tone, and
immune response-induced cytostasis. It is also implicated in many other
processes such as cerebral blood flow and ischemia, platelets and
neutrophil aggregation, coronary perfusion, lung ventilation,
glomerular perfusion, endocrine and exocrine secretion of pancreas, gut
peristaltism and secretion, cell proliferation, and migration
(10)
.
To further characterize the growth-promoting effect evoked by CCK-8, we have investigated the possible involvement of NO, the production of which could activate the soluble guanylate cylase/cGMP pathways and subsequently cell proliferation. Other aims of the study were to identify the NOS isoform that could be activated by CCK treatment and the mechanism of activation of this NOS.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-33P]ATP (3000 Ci/mmol) was purchased from
Isotopchim (Ganagobie, France). AG50W-X8 resin (sodium form) and
Polyprep columns were from Bio-Rad (Ivry sur Seine, France).
[14C] L-arginine, enhanced chemiluminescence
(ECL) immunodetection system, and Hybond ECL nitrocellulose membrane
were from Amersham Corp (Les Ulis, France). CHAPS was from Serva
(Heidelberg, Germany). Leupeptin, tetrahydro-L-bioptherin
(BH4), ß-NADPH, FMN, FAD,
NG-nitro-L-arginine (L-NAME),
calmodulin, pertussis toxin, soybean trypsin inhibitor, poly(Glu, Tyr),
cholesterol hemisuccinate, dithiothreitol (DTT), L-citrulline,
L-arginine, Sepharose-protein A beads, and geneticin (G418) were from
Sigma (Saint Quentin Fallavier, France). -Guanidinoglutaric acid
(GGA) and bovine ß12 recombinant protein were from Calbiochem
(Meudon, France). Fugene-6 was from Boehringer Mannheim (Meylan,
France).
Antibodies
Monoclonal anti-human neuronal NOS, monoclonal
antiphosphotyrosine-PY20 antibodies, and monoclonal anti-human SHP-2
antibody were from Transduction Laboratories (Montluçon, France).
Rabbit polyclonal anti-human neuronal NOS antibody was from Biomol
(Plymouth, Pa.). Anti-ß1 subunit antibody was from Gramsch
Laboratories (Schwabhôusen, Germany). Polyclonal anti-SHP-2
antibody was kindly provided by Dr. F. McKenzie (CNRS-UMR134, Nice,
France).
Cell culture and transfection
CHO cells (CHO K1 strain) were cultured in -modified Eagle's
medium (MEM) containing 10% fetal calf serum (FCS), fungizone,
streptomycin, and penicillin. The bovine carboxyl-terminal
ß-adrenergic receptor kinase 1 (ßARK1) (495–689) ß-globin
fragment cDNA, subcloned in pRK5, was kindly provided by Drs. M. Lohse
and R. Lefkovitz (HHMI, Duke University Medical Center, Durham, N.C.).
Cells were concomitantly transfected with pRK5-ßARK1 and pSV2neo
(Clontech, Palo Alto, Calif.) vectors. Stable transfectants were
selected in MEM containing geneticin at 600 µg/ml. After
selection, cells were cultured in MEM containing 10% FCS and
geneticin (200 µg/ml). pSV2neo vector alone stable transfectants were
used as control clones.
The rat heme-deficient neuronal nitric oxide synthase mutant C415A nNOS
was subcloned in pCMV5 vector, kindly provided by P. Rouet (INSERM
U317, Toulouse, France). The C459S SHP-2 mutant (11)
was
subcloned in the pcDNA3 vector, a gift from C. Nahmias (ICGM Cochin,
Paris). For transient transfection, CHO cells were grown in 60 mm
diameter dishes for 18 h in MEM containing 10% FCS. After
MEM was removed, cells were transfected with 2 µg of pCMV5/C415A
nNOS or 1 µg of pcDNA3/C459S SHP-2 vector and 1 µg of ßgal vector
containing ß-galactosidase cDNA (kindly provided by H. Paris, INSERM
U388, Toulouse, France), used as transfection internal control, and 6
µl Fugene-6 in MEM without FCS for 30 h. ß-galactosidase
activity was assayed using
o-nitrophenolß-D-galactopyranoside as substrate. Cells
transfected with pCMV5 or pcDNA3 vectors alone were used as control. In
these conditions of transient transfection, percentage of cell
transfected at 30 h was 48 ± 4%
(mean±SE), as measured after transient
expression of green fluorescent protein (driven by the CMV promoter)
and analysis under microscope (image-analysis system Viso-Lab 2000,
Biocom, Paris, France).
NO synthase activity
CHO cells were plated in 60 mm diameter dishes at 50 x
103 cells/ml (5 ml of MEM containing 10% FCS
per dish) until subconfluence and, after an 18 h period of serum
deprivation, were treated with CCK-8 with or without other agents
tested. Cells were then homogenized using a Dounce homogenizer (60
strokes at 4°C) in 50 mM Tris buffer (pH 7.5) containing 0.1 mM EDTA,
0.1 mM EGTA, 10% glycerol, 20 µM leupeptin, 1 mM PMSF, 1 mM soybean
trypsin inhibitor, 0.1% CHAPS. Protein was assayed using the Bio-Rad
Reagent and 25 µg of cell proteins were then incubated for 15 min at
37°C in 50 mM Tris buffer (pH 7.4) containing 50 µM
[14C]L-arg (150,000 cpm, specific activity 58.7
Ci/mmol), 10 mM ßNADPH, 1 mM DTT, 4 µM FMN, 4 µM FAD, 10 µM
BH4, 2 µg of calmodulin, and 1 mM
CaCl2 in a final incubation volume of 200 µl.
The reaction was terminated by addition of 1 ml of quench buffer
containing 20 mM HEPES, 2 mM EDTA, 0.2 mM EGTA, 1 mM L-citrulline. The
samples were applied to 2 ml columns of Dowex AG50W-X8
(Na+ form) preequilibrated with quench buffer.
Columns were then eluted with 2 ml of quench buffer and radioactivity
in the eluates was measured by liquid scintillation.
Cell growth assay
CHO cells were cultured in MEM containing 10% FCS and plated
in 35 mm dishes at 50 x 103 cells/ml (2 ml
per dish). After an overnight attachment phase, the medium was changed
to MEM containing either 10% FCS or 0% FCS, with or without other
different agents tested. Cell growth was measured after 24 h by
cell counting with a Coulter counter model ZM, as described previously
(12)
. Cell growth assays using treatment with exogenous
L-arginine were performed in L-arginine-free MEM.
SHP-2 activity assay
CHO cells were plated in 100 mm diameter dishes at 100 x
103 cells/ml for 16 h (10 ml of MEM
containing 10% FCS per dish) until subconfluence; after an 18 h
period of serum deprivation, they were treated with CCK-8 or other
agents. CHO and CHO/ßARK1 cells were washed twice with
phosphate-buffered saline (PBS) and solubilized for 10 min on ice with
50 mM Tris buffer (pH 7.6) containing 140 mM NaCl, 5 mM
MgCl2, 5 mM sodium orthovanadate, and 0.05%
soybean trypsin (buffer A) in the presence of 1.5% CHAPS and 0.5 mg/ml
cholesterol hemisuccinate. The mixture was gently agitated for 30 min
at 4°C and centrifuged at 13,000 x g for 20 min.
Soluble proteins (100 µg) were incubated for 2 h at 4°C with
anti-SHP-2 or preimmune serum prebound to Sepharose-protein A beads
prewashed in buffer A in the presence of 0.1% CHAPS, 60 µg/ml
cholesterol hemisuccinate, and 0.3% bovine serum albumin.
Immunoprecipitated proteins were then washed once with 40 mM Tris
buffer (pH 7) containing 0.5 mM EDTA, 0.1% Tween 20, and 0.1% sodium
dodecyl sulfate (SDS), resuspended in 380 µl of the same buffer, and
duplicated for tyrosine phosphatase assay. The substrate poly-(Glu,
Tyr) was phosphorylated with [-33P]ATP as
described previously (13)
. Dithiothreitol (5 mM) was
added; the reaction was initiated by the addition of 30,000 cpm of
33P-labeled poly(Glu, Tyr) and allowed to proceed
for 10 min at 30°C. PTPase activity was assayed as described
previously (12)
and expressed in picomoles of inorganic
phosphate released per minute at 30°C from radiolabeled substrate.
Immunoprecipitation and immunoblotting
CHO cells were plated in 100 mm diameter dishes at 100 x
103 cells/ml (10 ml per dish) until
subconfluence, and after an 18 h period of serum deprivation were
treated with CCK-8, with or without other agents tested. CHO cells were
washed twice in PBS and solubilized for 10 min on ice with 50 mM Tris
buffer (pH 7.6) containing 140 mM NaCl, 1 mM EDTA, 0.05% soybean
trypsin inhibitor, and 0.1 mM phenylmethylsulfonyl fluoride (buffer A)
in the presence of 1.5% CHAPS and 0.5 mM sodium orthovanadate. The
mixture was gently agitated for 30 min at 4°C and centrifuged at
13,000 x g for 20 min. Soluble proteins (500 µg to 1
mg) were incubated for 3 h at 4°C with antiphosphotyrosine or
anti-nNOS, or anti-SHP-2 antibodies or preimmune serum prebound to
Sepharose-protein A beads prewashed in buffer A. The beads were then
washed twice with buffer A and resuspended in 50 µl of Laemmli buffer
3% SDS.
For immunoblotting, 50 µl of immunoprecipitated proteins (see above) or 50 µg of solubilized proteins were resolved through 7.5% or 12% SDS-polyacrylamide gels, transferred to a nitrocellulose membrane, and immunoblotted with anti-human nNOS, anti-PY20, anti-ß1 subunit, or anti-SHP-2 antibodies. Immunoreactive proteins were visualized by the ECL immunodetection system and quantified by image analysis using a Biocom apparatus.
For visualization of monomers and dimers of nNOS, low-temperature
SDS-polyacrylamide gels were performed. Briefly, cells treated or not
with CCK-8 and transfected or not with the mutant nNOS cDNA were
solubilized for 30 min at 4°C in a Tris 50 mM buffer (pH 7.6)
containing 140 mM NaCl, 5 mM MgCl2, 0.5 mM
ortho-vanadate, 0.05% soybean trypsin inhibitor, and 0.1 mM
phenylmethylsulfonyl fluoride in the presence of 1% CHAPS and 10%
glycerol. After addition of Laemmli buffer 3% SDS without
ß-mercaptoethanol or heating, 100 µg of solubilized proteins was
resolved through discontinuous 5% to 15% SDS-polyacrylamide gradient
gels performed at a constant current of 30 mA. Gels and buffers were
equilibrated at 4°C prior to electrophoresis, which was performed at
temperature below 15°C, as described previously (14)
.
Immunoblotting was performed as described above using anti-human nNOS
antibodies.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. NO is the activator of
soluble guanylate cyclase. To further characterize this effect, CHO-K1
cells were treated or not with 0.1 µM CCK-8 for different times. As
shown in Fig. 1
A, NOS activity was stimulated in a time-dependent manner,
with a maximal effect after 30 s of CCK treatment, followed by a
decrease until 5 min. As shown in Fig. 1B, an
18 h
pretreatment of cells with 100 ng/ml of pertussis toxin reduced by
79 ± 4% the CCK-induced stimulation of NOS activity.
|
To investigate the role of NO production on cell proliferation caused
by CCK, CHO cells cultured in serum-free medium were treated with the
NO donor sodium nitroprusside (1 nM to 1 µM). This treatment resulted
in a dose-dependent increase of cell proliferation when compared to
control cells cultured in serum-free medium, maximal effect being
observed at 0.1 µM sodium nitroprusside (SNP; +38±9%,
ED50: 1.22±0.4 nM) (data not shown from three
experiments performed in triplicate). Treatment of cells with the
substrate of NOS L-arginine (100 µM), resulted also in an increase of
cell proliferation when compared to control cells (Fig. 2
). We also observed that cell proliferation induced by 0.1 µM SNP was
inhibited by 57 ± 1% when cells were concomitantly treated with
the soluble guanylate cyclase inhibitor LY 83583 (0.1 µM) (Fig. 2)
.
|
To further characterize the possible involvement of NO production in
the proliferative effect evoked by CCK, CHO cells were treated for
24 h with 0.1 µM CCK-8 in the presence or not of the inhibitors
of NOS L-NAME and GGA. As shown in Fig. 2
, CCK-induced CHO cell
proliferation was inhibited by incubating cells with the two
inhibitors. These inhibitors had no effect on the basal cell
proliferation (data not shown). No additive effect could be observed on
cell proliferation when cells were concomitantly treated with CCK-8
(0.1 µM) and L-arginine (100 µM), indicating that the same pathway
was involved. Taken together, these results indicate that CCK
stimulated CHO cell proliferation through the activation of a NOS and
by a mechanism involving a pertussis-toxin-sensitive Gi/G0 protein.
The constitutive nNOS is involved in the growth signaling
pathway of CCK
In CHO cells, the conditions and the kinetics of
activation of NOS activity were consistent with the involvement of a
constitutive NOS isoform. As described previously (15)
,
immunoblotting revealed that CHO-K1 cells expressed endogenous
constitutive nNOS (Fig. 3
A). However, we also detected the endothelial NOS isoform by
Western blotting, this isoform being expressed less than nNOS (data not
shown). In addition, we found that the NOS inhibitor GGA abolished the
CCK-induced stimulation of both cell proliferation and NOS activity in
CHO cells. This inhibitor, also known to be an anticonvulsant compound,
is highly active in neuronal tissues that predominantly express nNOS.
To obtain direct evidence that nNOS was activated by CCK in CHO cells
and that this isoform was implicated in the positive effect of CCK on
CHO cell proliferation, we transiently expressed the rat C415A nNOS
mutant (Cys415 mutated to Ala). This mutant does
not bind heme and tetrahydrobiopterin, essential cofactors required for
NO synthesis (16
, 17)
. After transient expression of the
C415A nNOS mutant, immunoblotting demonstrated expression of two forms
of nNOS (Fig. 3A
) in cells transfected with the pCMV5/C415A
nNOS vector when compared to cells transfected with the pCMV5 control
vector alone. The 155 kDa form corresponds to the recombinant rat
isoform and the 160 kDa form to the endogenous hamster nNOS form, also
detected in the hamster cerebellum soluble extracts used as positive
control. As shown in Fig. 3B
, CCK-8 was unable to stimulate
NOS activity in CHO cells transiently expressing the nNOS mutant when
compared to cells transiently transfected with the pCMV5 vector alone.
In addition, CCK-8 could not induce cell proliferation in cells
expressing the nNOS mutant (Fig. 3B
, insert). To investigate
the effect of C415A nNOS mutant, solubilized cells transfected with
pCMV5/C415A nNOS or pCMV5 vectors were subjected to low-temperature
SDS-polyacrylamide gel electrophoresis (PAGE) in order to evaluate the
amount of dimeric nNOS (320 kDa) relative to monomeric nNOS (155–160
kDa). As shown in Fig. 3C
(lanes 1 and 3), the amount of
nNOS dimers was not increased after expression of the nNOS mutant when
compared to the control cells. In addition, under CCK treatment, the
amount of endogenous nNOS dimers was increased (lane 4) in control CHO
cells. This increase was abolished after nNOS mutant expression (lane
2). All these results suggest that CCK can induce endogenous nNOS
dimerization. This effect is prevented after transfection of inactive
nNOS without apparent heterodimerization between endogenous nNOS and
heterologous mutant nNOS.
|
SHP-2 mediates the stimulation the nNOS activity and cell
proliferation induced by CCK
The activity of the constitutive NOS isoforms such as
nNOS is critically controlled by an elevated level of intracellular
calcium, which promotes binding of calmodulin to the enzymes. In CHO-K1
cells, we previously demonstrated that CCK can induce mobilization of
intracellular calcium (12)
. However, this effect appeared
to be pertussis toxin independent, and calcium mobilization was not
implicated in the proliferative effect of CCK in these cells
(6)
. Evidence exists that endothelial isoform of NOS could
be activated by pathways other than the increase of intracellular
calcium concentrations, such as by phosphorylation or dephosphorylation
of proteins on tyrosine or serine-threonine residues (18)
.
Previous reports demonstrated that CCK activated tyrosine kinase and
tyrosine phosphatase activities, and these effects could be involved in
the trophic effect of CCK in vivo (5)
.
We first tested the possible implication of a tyrosine phosphatase
activity in the CCK-induced cell proliferation. CHO cells were
concomitantly treated or not for 24 h with CCK-8 and the tyrosine
phosphatase inhibitor pervanadate. At concentrations of 0.1, 0.3, and
0.5 µM, pervanadate inhibited CCK-induced cell proliferation
(-72±8.5% at 0.3 µM) (Fig. 4
). The basal cell proliferation of CHO-K1 was not affected at these
three concentrations, but was inhibited at 1 µM of pervanadate (data
not shown). In parallel, we tested the effect of okadaic acid, the
inhibitor of the serine threonine phosphatases, which had no effect on
CCK-induced cell proliferation at a concentration of 0.1 µM (data not
shown from three experiments performed in triplicate). All these
results suggested that CCK-induced cell proliferation depends on the
activation of a tyrosine phosphatase.
|
The phosphotyrosine phosphatase SHP-2 is a positive mediator of the
mitogenic signaling induced by several growth factors
(19)
. To determine whether CCK activated the tyrosine
phosphatase SHP-2 in CHO cells, cells were incubated in the presence of
CCK-8 for various times (after which they were solubilized) and SHP-2
activity was measured in SHP-2 immunoprecipitates. As shown in
Fig. 5
, treatment by CCK evoked a rapid and time-dependent stimulation of
SHP-2 activity. The stimulation of SHP-2 activity was maximal after 1
min of CCK treatment and declined for up to 15 min.
|
To obtain direct evidence that tyrosine phosphatase SHP-2 was
implicated in the positive effect of CCK on CHO cell proliferation, we
transiently expressed the dominant negative mutant SHP-2 (C459S mutant:
Cys459 mutated to serine) (11)
.
After transient expression of the mutant, immunoblotting with
anti-SHP-2 antibodies demonstrated that expression of the protein in
cells (treated or untreated with CCK-8) transfected with pcDNA3/C459S
SHP-2 vector was increased when compared to cells transfected with
pCDNA3 vector alone (Fig. 6
A). This increase may correspond to the recombinant mutant
transiently expressed in CHO cells. After transient expression of the
SHP-2 mutant, CCK-8 was unable to stimulate cell proliferation when
compared to cells transiently transfected with the pcDNA3 control
vector alone (Fig. 6B
). As shown in Fig. 6B
(insert), the inhibition of CCK-8-induced cell proliferation depended
on the concentration of the pcDNA3/C459S SHP-2 plasmid transfected,
maximal effect being observed with 1.5–2 µg.
|
To determine whether the NOS activity stimulated by CCK depends on a
tyrosine phosphatase activation, CHO cells were treated or untreated
with CCK-8 for 30 s after or without a 15 min preincubation with
pervanadate at a concentration of 0.3 µM. As shown in Fig. 7
A, pretreatment of cells with pervanadate had no effect on
basal NOS activity, but inhibited by 90 ± 1% the CCK-induced
stimulation of NOS activity, suggesting that activation of NOS depends
on the activation of a tyrosine phosphatase.
|
To ask whether CCK regulates the level of tyrosine phosphorylation of
nNOS, CCK-treated or untreated CHO cells were subjected to
immunoprecipitation with antiphosphotyrosine antibodies. Immunoblot
with anti-nNOS antibodies revealed that nNOS of untreated cells was
tyrosine phosphorylated and rapidly dephosphorylated after CCK
treatment, maximal dephosphorylation being observed at 15 and 30 s, which slowly decreased until 5 min. Pretreatment of cells with 0.3
µM pervanadate for 15 min inhibited nNOS tyrosine dephosphorylation
induced by a 30 s CCK treatment (Fig. 7B, D
). In
addition, converse experiments revealed that a similar profile and
kinetic of nNOS dephosphorylation under CCK treatment were observed
after immunoprecipitation of CHO cells with anti-nNOS antibodies,
followed by immunoblot with antiphosphotyrosine antibodies (Fig. 7C
).
To obtain evidence that CCK may activate a NOS activity in CHO cells by
stimulating SHP-2, we transiently expressed the inactive SHP-2 C459S
mutant (immunoblotting of transfected cells with anti-SHP-2 antibodies
is shown in Fig. 8
A). After transient expression of this mutant, CCK-8 was
unable to stimulate NOS activity when compared to cells transiently
transfected with the pcDNA3 control vector alone (Fig. 8B
).
As shown in Fig. 8C
, dephosphorylation of nNOS induced by
CCK-8 treatment of CHO cells (lanes 1 and 2) was abolished after
transfection of inactive SHP-2 mutant, the basal phosphorylation of
nNOS remaining unchanged (lanes 3 and 4). These results indicate that
activation of SHP-2 is implicated in the activation of NOS activity and
nNOS dephosphorylation induced by CCK in CHO cells.
|
The Gß subunit complex is involved in the coupling of CCK
receptor to nNOS
Whereas it was thought that G was the sole or major
activator of effectors, it is now recognized that Gß is also an
important independent activator of many different effectors of
G-protein actions (including Gi proteins) (20)
. To
investigate the implication of Gß in the CCK-induced activation of
NOS, CHO cells were first treated or not with exogenous Gß12
subunit at final concentrations of 50, 100, and 150 nM under the same
conditions used for CCK-8 treatment. As shown in Fig. 9
A, treatment of solubilized CHO cells with the exogenous
Gß12 subunit evoked a stimulation of NO synthase activity. The
maximal effect was observed at a concentration of 100 nM, the
stimulatory effect being similar to that observed after CCK-8 treatment
of wild-type CHO cells or CHO cells stably expressing the
neomycin-resistant gene alone.
|
The G-protein receptor kinases are known to phosphorylate multiple
receptor serine and threonine residues. Among this kinases, the
ß-adrenergic receptor kinases 1 and 2 translocate to a variety of
G-protein-coupled receptors by forming a complex with the Gß
released upon activation of heterotrimeric G-proteins. A 125 amino acid
domain at the carboxyl terminus of these kinases provides the binding
site for Gß. Expression of a minigene corresponding to the
carboxyl terminus amino acid sequence of ßARK1 has been shown to act
as a sequestrant sequence for Gß subunits when expressed in COS
cells (21)
. To demonstrate the implication of Gß in
the CCK-induced stimulation of both CHO cell proliferation and NOS
activity, the cDNA encoding the carboxyl-terminal 195 amino acids of
the bovine ßARK1-(495–689) polypeptide were stably transfected in
CHO cells. Cellular expression of the minigene product could be
demonstrated by immunoblotting of solubilized cells, using a specific
antibody raised against the last 221 amino acids of the carboxyl
terminus domain of the rat ßARK2 (kindly provided by C. Stone, Duke
University Medical Center), which recognizes the isoforms ßARK-1 and
ßARK-2, as well as the minigene ßARK1 (495–689) (20)
.
An immunoreactive protein of 28 kDa was detected in CHO cells
expressing ßARK1 (495–689), but not in control CHO cells expressing
a neoresistance gene (data not shown). In addition, no effect was
observed on the apparent expression of CCK receptor (assayed by CCK
ligand binding, as described in 6
) in CHO cells stably
expressing the ßARK1-(495–689) polypeptide when compared to
wild-type CHO cells or CHO cells stably expressing the neomycin
resistance gene (data not shown).
As shown in Fig. 9A, B
, in cells expressing the sequestrant
polypeptide, CCK-induced stimulation of NOS activity and cell
proliferation were both abolished. In this clone, the basal NOS
remained unchanged and cell proliferation could be induced by serum.
Similar results were observed in another clone of CHO cells expressing
ßARK1-(495–689) carboxyl-terminal polypeptide (data not shown).
Moreover, treatment of CHO cells stably expressing the
ßARK1-(495–689) polypeptide with CCK-8 (0.1 µM) did not induce
stimulation of SHP-2 activity whenever the time of treatment (data not
shown from two experiments in duplicate). These results suggest that in
CHO cells, CCK stimulated cell proliferation, NOS activity, and SHP-2
activity via a Gß-dependent pathway.
SHP-2 associates with G-ß subunits and nNOS in response to CCK
treatment
To further characterize the implication of G-ß, SHP-2, and
nNOS in the growth signal evoked by CCK, CHO cells were either treated
not with CCK for various times prior to solubilization and
immunoprecipitation with anti-SHP-2 antibodies. The amounts of G-ß1
subunits (always dynamically associated with the subunit) and nNOS
were then analyzed by immunoblotting. The blots were reprobed with
anti-SHP-2 antibodies to ensure that comparable amounts of SHP-2
molecules were immunoprecipitated at each time point of CCK treatment.
As observed in Fig. 10
A, the ß1 subunit was immunoprecipitated with SHP-2
antibodies in resting cells. The treatment with CCK resulted in a rapid
increase of the amount of ß1 subunit immunoprecipitated, which was
maximal after 15 and 30 s of treatment (315±45% of control) and
decreased until 5 min. In parallel, we observed that nNOS was also
immunoprecipitated with SHP-2 antibodies in resting cells. In response
to CCK treatment, the amount of immunoprecipitated nNOS was increased
by 240%, with a maximum observed after 1 min of CCK treatment and
followed by a decrease after 2 and 5 min. The kinetics of association
of nNOS with SHP-2 was shifted in time when compared to the association
of G-ß1 with SHP-2.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The effect of LY
83583 suggests that NO acts through the activation of guanylate cyclase
and the production of cGMP.
Besides VIP, neurotensin, angiotensin, or muscarinic receptors
(14
, 22)
, CCKA receptor represents a
novel example of G-protein-coupled receptor that can be positively
coupled to a NOS activity. However, to our knowledge this is the first
report demonstrating that this positive coupling is implicated in cell
growth stimulation. NO has been shown to inhibit cell proliferation and
cell migration and to induce apoptosis. However, it has also been shown
to promote proliferation of various cell types in vitro and
to facilitate tumor growth and neovascularization in vivo
(23
, 24)
. Effects of NO on cell proliferation are known to
depend on concentration of the molecule—high concentrations often
exerting negative effects and low concentrations evoking a positive
effect (25)
. The results we obtained with low
concentrations of sodium nitroprusside on CHO cells and a low amount of
intracellular NO produced under CCK treatment correlated well with
these observations.
In the present work, the rapid activation of NOS activity suggested the
involvement of a constitutive NOS. Moreover, CCK treatment promoted the
formation of nNOS dimers, which appear to be the active form of nNOS.
The ability of the heme/BH4-deficient nNOS mutant
to inhibit CCK-induced stimulation of NOS activity and cell
proliferation as well as CCK-induced nNOS dimers formation suggests
that nNOS may be required to mediate the effect of CCK. As previously
demonstrated, dimerization of nNOS requires heme and
BH4 (14
, 26)
. Thus dimerization of
the nNOS mutant could not be observed after its transfection in CHO
cells. However, heterodimerization between mutant and wild-type nNOS
did not occur. We hypothesized that the inactive nNOS could act as a
competitive inhibitor of endogenous nNOS regarding the transduction
pathway molecules activated by CCK such as tyrosine phosphatase SHP-2.
Neuronal NO synthase displays a widespread expression in central and
peripheral nervous systems such as the gut and urogenital tract
(27)
. This isoform is considered to be responsible for the
largest proportion of tissue NO synthase activity. The enzyme has thus
been implicated in the regulation of neuronal cell biology and in
neuroendocrine biology as a major nonadrenergic noncholinergic
neurotransmitter in enteric nerves in the contraction of skeletal
muscle and relaxation of smooth muscle, among other roles
(28)
. Many authors speculate that nNOS may be implicated
in nervous system morphogenesis, neuronal differentiation, and
synaptogenesis. Targeted disruption of the nNOS in mice by homologous
recombination did not reveal major abnormalities of central nervous
development (29)
. Some authors have approached the role of
nNOS in cell growth or differentiation by investigating the expression
of the isoform on neuronal or phaeochromocytoma cells. Neuronal NOS
has been induced by growth factors or expressed in surviving neurons
(30)
. In addition, induction of different isoform of NOS
in phaeochromocytoma cells has been responsible for growth arrest
(31)
. During Drosophila development, high
expression of NOS and the resulting NO production were responsible for
cell growth arrest (32)
. To our knowledge, however, direct
evidence of the implication of nNOS in the positive regulation of cell
proliferation of eukaryotic cells has rarely been observed.
The physiological relevance of the coupling of CCK receptor with nNOS
remains to be established. The mouse exocrine pancreas represents a
tissue model that expresses only the CCKA
receptor subtype, as we found in our CHO-K1 strain. CCK is known to
stimulate exocrine pancreatic secretion and growth in mice. NO is known
to positively modulate basal and stimulated exocrine pancreatic
secretion and blood flow in rodents. However, controversies still exist
in the possible direct implication of the NO/cGMP system on amylase
secretion by acinar cells (33)
. Interestingly, we found
that CCK-8 evoked a time-dependent stimulation of NOS activity on mouse
acinar cells and that nNOS was expressed in pancreatic acinar cells
(unpublished results). Activation of nNOS could thus be implicated in
the mitogenic effect of CCK on exocrine pancreas. The involvement of
the coupling between CCKA receptors and nNOS in
the other biological effects of CCK that implicate NO, such as
relaxation of sphincter of Oddi and lower esophagus, colonic, or
gastroduodenal motility, and gastroprotection, remain to be
investigated.
Activation of nNOS requires activation of the tyrosine phosphatase
SHP-2 and G-ß subunit
The SH2-containing phosphotyrosine phosphatases such as SHP-1 and
SHP-2 regulate various protein tyrosine kinase signaling pathways,
interacting with receptor autophosphorylation or with signaling
proteins. Whereas SHP-1 appears to be a negative regulator of growth
factors receptor signaling, SHP-2 is a positive mediator of the
mitogenic signaling induced by several growth factors (19
, 34)
. Many of these growth factors act through receptors that
possess an intrinsic tyrosine kinase activity such as insulin, PDGF,
EGF receptors, but also act though G-protein-coupled receptors such as
thrombin or proteinase-activated receptor 2 receptors (11
, 35)
.
We observed that CHO cell proliferation induced by CCK was dependent on
a tyrosine phosphatase and required the stimulation of SHP-2 activity.
The kinetic of activation of SHP-2 correlated well with that of
stimulation of NOS activity, dephosphorylation of nNOS on tyrosine
residues, and association of SHP-2 with nNOS evoked by CCK treatment.
Expression of inactive SHP-2 inhibited CCK-induced cell proliferation,
nNOS dephosphorylation on tyrosine residues, and NOS activity. As
previously observed (11
, 36)
, this catalytically inactive
protein can bind to substrates without dephosphorylating them, thereby
competitively interfering with the access of these substrates to the
endogenous wild-type SHP-2. In the present work, nNOS could thus be a
substrate of SHP-2. Taken together, these results suggest that CCK
treatment led to the activation of SHP-2 that can dephosphorylate and
then activate the nNOS.
The mechanism of association of SHP-2 with nNOS and the tyrosyl residue
involved on nNOS remain to be identified. Nevertheless, this is the
first evidence that nNOS activity can be regulated by tyrosine
dephosphorylation. Using kinase or phosphatase inhibitors, the eNOS
isoform has been found to be regulated by phosphorylation on serine
threonine or tyrosine residues, but from indirect observations
(9
, 37)
. Tyrosine phosphorylation also seems to regulate
association of eNOS with caveolin-1 within endothelial caveola.
Tyrosine phosphorylation thus may regulate activity and subcellular
trafficking of eNOS (18)
. It has recently been
suggested that phosphorylation/dephosphorylation may alter
conformation and/or protein coupling of constitutive NOS, facilitating
their interactions with specific phospholipids or proteins that enhance
or maintain a Ca2+-independent activation
(38
, 39)
.
Expanding examples of cellular responses regulated by Gß have
already been described. Cell growth and differentiation could implicate
or Gß protein subunits. In the present work, we observed that
the recombinant Gß12 subunit could induce a stimulation of NOS
activity. In addition, in the expression of the carboxyl terminus of
ß-adrenergic receptor kinase-1, a Gß subunit-sequestering agent
inhibited the CCK-induced stimulation of SHP-2 activity, NOS activity,
and cell proliferation. These results also suggested that CCK
stimulated SHP-2 activity in CHO cells via a Gß-dependent pathway.
The kinetic of G-ß1 and SHP-2 associations correlated well with those
of CCK-induced stimulation of SHP-2 activity, CCK-induced stimulation
of NOS activity, and CCK-induced dephosphorylation of nNOS on tyrosine
residues. All these observations suggest that upon CCK stimulation, the
occupation of CCKA receptors induced the
recruitment of G-ß subunit to SHP-2, which was activated and then
recruited nNOS to dephosphorylate and activate it.
In conclusion, the new concept of nNOS acting as an important molecule in the signal cascade of growth factors may have some practical implication taking into account the wide distribution of this isoform both in neuronal and nonneuronal tissues. Moreover, activation of nNOS by dephosphorylation could be involved in other physiological roles of nNOS in epithelial, neuronal, or vascular systems.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. G. Lee and D. L. Mykles Proteomics and signal transduction in the crustacean molting gland Integr. Comp. Biol., December 1, 2006; 46(6): 965 - 977. [Abstract] [Full Text] [PDF]
|
|
|
