(Received for publication, October 24, 1996, and in revised form, January 27, 1997)
From the Third Department of Internal Medicine, Insulin receptor substrate (IRS)-2 is
structurally and functionally similar to IRS-1. Indeed, stimulation
with insulin or insulin-like growth factor I led to the rapid tyrosine
phosphorylation of both IRS-1 and IRS-2, which in turn activated
phosphatidylinositol (PI) 3-kinase in L6 cells and rat skeletal muscle.
However, IRS-2 was rapidly dephosphorylated (3-10 min after the
addition of insulin/insulin-like growth factor I), whereas IRS-1
phosphorylation continued for at least 60 min. The time courses of the
PI 3-kinase activity associated with IRS-1 and IRS-2 paralleled the
tyrosine phosphorylation of these proteins. Preincubation with sodium
orthovanadate, an inhibitor of protein tyrosine phosphatase, blocked
the rapid dephosphorylation of IRS-2, suggesting the involvement of
tyrosine phosphatase. The activation of PI 3-kinase apparently plays an
important role in the rapid dephosphorylation of IRS-2, as IRS-2
dephosphorylation was inhibited markedly by suppressing PI 3-kinase
activity with wortmannin or overexpression of the dominant negative p85
subunit of PI 3-kinase, which cannot bind the p110 catalytic subunit. In addition, platelet-derived growth factor stimulation prior to
insulin stimulation decreased IRS-associated PI 3-kinase and significantly inhibited the dephosphorylation of IRS-2. Taken together,
these observations suggest that IRS-2 plays a unique role in mediating
the signals from the insulin receptor to downstream molecules and
that this effect is more transient than that of IRS-1. Tyrosine
phosphatase and IRS-associated PI 3-kinase activity thus contribute to
the rapid dephosphorylation of IRS-2.
Insulin binding to its receptor induces activation of the receptor
tyrosine kinase followed by the phosphorylation of several cytosolic
substrates. A major substrate of the insulin receptor is a 165-185-kDa
protein termed insulin receptor substrate (IRS)1-1 (1).
After insulin stimulation, IRS-1 is rapidly
phosphorylated on multiple tyrosine residues and binds to several
Src-homology 2 domains containing proteins (SH2 proteins) (2), which
include regulatory subunits for phosphatidylinositol (PI) 3-kinase
(p85 (3), p55 However, studies using IRS-1-deficient mice, derived from targeted gene
disruption, have demonstrated that a 180-190-kDa protein functions as
an alternative substrate for the insulin receptor (10, 11). Cloning of
this protein revealed a structure similar to that of IRS-1, which led
this protein to be designated IRS-2 (12). IRS-2 has multiple conserved
tyrosine phosphorylation sites that can bind to various SH2 proteins
(12), indicating that IRS-2 is functionally similar to IRS-1. Herein,
we investigated the functional differences between IRS-1 and IRS-2. We
found that IRS-2 is dephosphorylated much more rapidly and activates PI
3-kinase more transiently than IRS-1 in skeletal muscle cells.
Furthermore, important roles of its associated PI 3-kinase in the rapid
dephosphorylation of IRS-2 are discussed.
Polyclonal anti-IRS-1 and -IRS-2 antibodies were
prepared by immunizing rabbits with synthetic peptides derived from
amino acids based on the unique COOH termini of mouse, rat, and human IRS-1 (RRSSEDLSNYASINFQKQPEDRQ, corresponding to residues 1211-1233 (mouse), 1213-1235 (rat), and 1220-1242 (human)) and mouse IRS-2 (TYASIDFLSHHLKEATVVKE, residues 1302-1321). Antibody against the p85
subunit of PI 3-kinase was raised as described (5). The antibodies were
affinity purified and concentrated as described (14). The
anti-phosphotyrosine ( L6 myoblasts were grown in Dulbecco's
modified Eagle's medium containing 10% fetal calf serum at 37 °C
in 5% (v/v) CO2 in air. The cells were seeded onto
10-cm-diameter plastic culture dishes at a density of 3,000 cells/cm2. L6 cells were rendered quiescent in Dulbecco's
modified Eagle's medium containing 2% serum for 10 days to promote
fusion into myotubes. Myotube formation was determined as the
percentage of nuclei present in multinucleated myotubes. In this
experiment, 80-90% of myoblasts were fused into myotubes.
A
full-length p85 The recombinant
adenoviruses Adex1CALacZ (18) and Adex1CA Male Sprague Dawley rats (7 weeks, 244-272 g) were
fasted for 12 h before the experiments and anesthetized with
sodium pentobarbital. The portal vein was exposed, and 4 ml of normal
saline with or without 10 Confluent monolayers
of L6 cells were incubated for 3 h in serum-free Dulbecco's
modified Eagle's medium prior to insulin stimulation. In some
experiments, the cells were first incubated with 2 mM
Na3VO4 for 15 min or 10 nM
wortmannin (Sigma) for 10 min. The cells were stimulated with the
indicated concentrations of insulin for the indicated periods at
37 °C. In some experiments, the cells were stimulated with
10 The L6 cells (6.0 × 105 cells) in 10-cm dishes were solubilized in 6 ml of
ice-cold lysis buffer containing 20 mM Tris, pH 7.5, 137 mM NaCl, 1 mM CaCl2, 1% Nonidet
P-40, 10 mg/ml aprotinin, 1 mM phenylmethylsulfonyl
fluoride, and 100 µM Na3VO4.
Lysates were extracted by centrifugation at 15,000 × g
for 10 min, and 1-ml quantities of the lysates (~0.3 mg of total
protein) were incubated with anti-IRS-1 or anti-IRS-2 antibody (10 µg
each) for 1 h. Immunocomplexes were precipitated with 20 µl of
protein A-Sepharose for 1 h. PI 3-kinase was assayed in the
immunoprecipitates as described (4).
We initially examined the
expression and insulin-induced tyrosine phosphorylation of IRS-1 and
IRS-2 in L6 cells. Cell lysates were immunoprecipitated with anti-IRS-1
or anti-IRS-2 antibody and immunoblotted with anti-IRS-1 or anti-IRS-2
antibody. As shown in Fig. 1A, anti-IRS-1 and
anti-IRS-2 antibodies specifically recognize IRS-1 and IRS-2,
respectively (lanes 1-4). After the incubation with
10
L6 cells were then stimulated with the indicated concentrations of
insulin for 3 min (Fig. 1B). The tyrosine phosphorylations of IRS-1 and IRS-2 were increased by insulin in a
concentration-dependent manner without altering the amounts of
either protein. IRS-1- and IRS-2-associated PI 3-kinase activities were
also increased in a similar insulin concentration-dependent
manner (Fig. 1C).
Fig. 2A shows the
time courses of IRS-1 and IRS-2 tyrosine phosphorylation. IRS-1 and
IRS-2 were both tyrosine-phosphorylated within 1 min after the addition
of insulin/IGF-I, with peaks at 3 min. However, IRS-2 was rapidly
dephosphorylated, and phosphorylation was essentially undetectable
beyond 10 min, whereas IRS-1 phosphorylation persisted for more than 60 min. The amounts of IRS-1 and IRS-2 proteins did not change until at
least 60 min. To quantitate the sensitivity of
Fig. 2B illustrates the time courses of IRS-1 and IRS-2
tyrosine phosphorylation in rat skeletal muscle. Although IRS-1 and IRS-2 were both tyrosine-phosphorylated within 3 min after the injection of insulin into the portal vein, tyrosine phosphorylation of
IRS-2 was attenuated much more rapidly than that of IRS-1. Thus, IRS-2
is more transiently phosphorylated by insulin/IGF-1 stimulation than
IRS-1 in L6 cells and rat skeletal muscle.
Next, we compared insulin-induced PI 3-kinase association with IRS-1
and IRS-2. Relative PI 3-kinase activities were expressed as the
percentage of the maximal increase from the basal level. The 100%
values of PI 3-kinase associated with IRS-1 and IRS-2 (derived from
1.0 × 105 cells) were 2199 and 302 cpm, respectively.
Thus, the relative amount of PI 3-kinase associated with IRS-2 was
13.6% that of IRS-1. Fig. 2C (left panel) shows
that whereas IRS-1-associated PI 3-kinase activity was attenuated to
48% of the maximal increase from the basal level at 60 min after the
addition of insulin, that of IRS-2 decreased to 12% of the maximum at
60 min. Similar results were obtained with IGF-I stimulation (Fig.
2C, right panel). This rapid decrease in
IRS-2-associated PI 3-kinase activity is considered to be reflected by
the rapid dephosphorylation of IRS-2 shown in Fig. 2A. Thus,
it is likely that IRS-2 mediates signals from the insulin/IGF-I
receptor to PI 3-kinase more transiently than does IRS-1 in skeletal
muscle cells.
To examine whether the rapid dephosphorylation of IRS-2
occurs via protein tyrosine phosphatase, we incubated L6 cells with Na3VO4, an inhibitor of tyrosine phosphatase,
prior to insulin stimulation. After incubation with 2 mM
Na3VO4 for 15 min, insulin-induced phosphorylation of IRS-2 continued for at least 60 min after the addition of insulin (Fig. 3). This result suggests the
presence of tyrosine phosphatase(s), which is involved in the rapid
dephosphorylation of IRS-2.
Furthermore, we attempted to identify
the molecular event that constitutes the signal leading to the
dephosphorylation of IRS-2. We examined the possible involvement of PI
3-kinase in IRS-2 dephosphorylation utilizing two independent
procedures. First, we incubated L6 cells with 10 nM
wortmannin, a potent inhibitor of PI 3-kinase, for 10 min prior to
insulin stimulation. The wortmannin treatment markedly inhibited the
rapid dephosphorylation of IRS-2, whereas no significant effect was
observed for IRS-1 (Fig. 3).
Second, the
We observed that overexpression of The time courses of tyrosine phosphorylation of IRS-1 and IRS-2 in L6
cells overexpressing In
addition, we performed an experiment to determine whether it is PI
3-kinase activation alone or the binding of active PI 3-kinase to IRS-2
which induces the rapid dephosphorylation of IRS-2. L6 cells were
stimulated with 50 ng/ml PDGF for 30 min and then with
10
The binding of insulin to its receptor activates receptor tyrosine
kinase. The activated insulin receptor phosphorylates several cytosolic
substrates, which contribute to the divergent biological effects of
insulin. These include a major insulin receptor substrate, IRS-1, and
the recently cloned IRS-2. These two proteins had been considered to be
structurally and functionally similar (12, 22). To date, little
information has been available on the different characteristics of
IRS-1 and IRS-2. It was reported that IRS-2 binds to the insulin
receptor through its newly identified domain, which is absent in IRS-1
(23, 24). A recent report showed that after treatment with tumor
necrosis factor- Next, we attempted to clarify the molecular mechanisms leading to rapid
dephosphorylation of IRS-2. When the cells were incubated with
Na3VO4 to block tyrosine phosphatase, IRS-2
tyrosine phosphorylation was prolonged. The rapid dephosphorylation of
IRS-2 occurs via protein tyrosine phosphatase, as expected. However,
when the cells were incubated with 10 nM wortmannin to
inhibit PI 3-kinase, the dephosphorylation was also markedly inhibited.
In addition, when the Although the mechanism by which IRS-associated PI 3-kinase triggers the
rapid dephosphorylation of IRS-2 remains unknown, the evidence supports
two possibilities. First, IRS-1-associated PI 3-kinase leads to
phosphorylation of serine residues of IRS-1. PI 3-kinase possesses not
only lipid kinase, but also serine kinase activity (26). In
vitro kinase assays of anti-p85 immunoprecipitates have shown that
IRS-1 is serine- phosphorylated by IRS-1-associated PI 3-kinase (27,
28). We speculate that IRS-2-associated PI 3-kinase phosphorylates
serine residues of IRS-2 as well. Meanwhile, serine/threonine
phosphorylation of IRSs affects their tyrosine phosphorylation (29,
30). When the serine/threonine phosphorylation of IRS-1 is augmented by
okadaic acid (29) or calyculin A (30), which are serine/threonine
phosphatase inhibitors, insulin-induced tyrosine phosphorylation of
IRS-1 is impaired. It is possible that serine/threonine phosphorylation
of IRSs modulates not only their tyrosine phosphorylation but also
dephosphorylation by tyrosine phosphatase. Different serine/threonine
phosphorylation sites in IRS-1 and IRS-2 might be responsible for the
different degree of accessibility to tyrosine phosphatase. We speculate
that IRS-2-associated PI 3-kinase, which phosphorylates serine residues
of IRS-2, modulates rapid tyrosine dephosphorylation of IRS-2.
Second, a role for PI 3-kinase in protein sorting has been suggested by
the finding that a yeast PI 3-kinase homolog, VSP34, is involved in
sorting proteins to vacuoles (31). PI 3-kinase binding to the PDGF
receptor is reportedly essential for the postendocytic sorting of this
receptor (32). We speculated that PI 3-kinase binding to IRS-1 and
IRS-2 is necessary for their intracellular trafficking and that they
are sorted and dephosphorylated differently.
In conclusion, IRS-2 is dephosphorylated more rapidly than IRS-1, and
IRS-2 dephosphorylation is regulated to some extent by IRS-associated
PI 3-kinase activity. Based on the results of this study, we have
devised a model for the roles of IRS-1 and IRS-2. We propose that IRS-1
transmits continuous signals from the insulin receptor, whereas IRS-2
mediates transient signals regulated by its associated PI 3-kinase
activity. The continuous IRS-1 pathway and the duration-regulated IRS-2
pathway might operate in combination to fine tune the acute and chronic
actions of insulin.
We are grateful to Dr. Izumu Saito (Institute
of Medical Science, University of Tokyo) for the generous gift of the
recombinant Adex1CALacZ and the cassette cosmid for constructing
recombinant adenovirus.
Laboratory of Molecular and Cellular Morphology,
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENT
REFERENCES
(4, 5), and p55
(5)), Grb2 (6), SHPTP2 (7), and Nck
(8). Consequently, IRS-1 mediates activation of PI 3-kinase,
p21ras and mitogen-activated protein kinase (9),
resulting in the promotion of glucose uptake, glycogen synthesis,
mitogenesis, or gene expression. Therefore, IRS-1 was originally
thought to play important roles in insulin signaling.
Antibodies
PY) monoclonal antibody (4G10) was purchased
from Upstate Biotechnology Inc. Anti-hemagglutinin epitope antibody
(12CA5) was from Boehringer Mannheim. Anti-IRS-1, -IRS-2, -p85, and -PY
antibodies immunoprecipitated 70-90% of each protein, as assessed by
immunoblotting the lysate before and after immunoprecipitation. The
efficiencies of immunoprecipitation with anti-IRS-1 and anti-IRS-2
antibodies were comparable (80-90%).
p85)
cDNA was isolated by screening a cDNA
library from MIN6, a mouse insulinoma cell line (15), on the basis of
the reported sequence (16). Comparison with the reported sequence
revealed codon CCC (Pro46) to have been replaced with CAG
(Gln46) in our p85
cDNA. To construct a
p85
mutant that cannot bind to the p110 catalytic subunit, we synthesized
two oligonucleotide primers as follows: sense, TTCGACTCTATACAGAACAC
corresponding to nucleotides 994-1013 of p85
; antisense,
GCCGGCATAGTCTGGAACATCGTATGGATATTGGATTTCCTGGGAAGTACG containing an
NaeI site (GCCGGC), the antisense sequence of the influenza
virus hemagglutinin epitope tag (YPYDVPDYA) and nucleotides 1970-1990
of p85
. Polymerase chain reaction amplification was carried out
using the two primers and p85
cDNA as a template. A fragment
containing nucleotides 994-1990 of p85
, as well as nucleotide
sequences for the hemagglutinin tag and NaeI site at its
3
-end, was obtained. The PstI-DraI fragment
(nucleotides 1033-2092) of p85
was replaced with the
PstI-NaeI fragment of the polymerase chain
reaction product, such that amino acids 479-512 of p85
,
i.e. those responsible for binding with p110 (17), were
replaced with a hemagglutinin tag sequence.
p85, which encode the
Escherichia coli lacZ (
-galactosidase) gene and
p85,
respectively, were constructed as described (19, 20). For protein
production, L6 cells were infected with these adenoviruses at a
multiplicity of infection of 10~30 and incubated for 48 h.
lacZ gene expression was confirmed by histochemical staining
with 5-bromo-4-chloro-3-indolyl
-D-galactopyranoside (0.5 mg/ml). When adenovirus Adex1CALacZ was applied, lacZ
expression gene was observed in more than 90% of L6 cells (data not
shown). In addition, L6 cells infected with Adex1CALacZ exhibited no
significant difference in the expression of p85 protein or in PI
3-kinase activity compared with noninfected cells (data not shown).
Therefore, in this study, L6 cells infected with Adex1CALacZ were used
as a control. The expression of
p85 protein was confirmed by
immunoblotting the anti-p85 antibody immunoprecipitate with anti-p85
and anti-hemagglutinin antibodies.
5 M insulin was
injected. After the indicated periods, the soleus muscle was removed
and homogenized immediately with a Polytron aggregate generator
(Kinematica) operated at maximum speed for 30 s in ice-cold
solubilization buffer (10 ml/g tissue), composed of 50 mM
Hepes at pH 7.5, 137 mM NaCl, 1 mM
MgCl2, 1 mM CaCl2, 10 mM sodium pyrophosphate, 10 mM sodium fluoride,
2 mM EDTA, 2 mM sodium orthovanadate
(Na3VO4), 1% Nonidet P-40, 1 mM
phenylmethylsulfonyl fluoride, and 10 mg/ml aprotinin. The homogenates
were centrifuged at 15,000 × g at 4 °C for 30 min
to remove insoluble material, and the supernatant was
immunoprecipitated (as described below).
7 M insulin-like growth factor I (IGF-I).
In other experiments, the cells were stimulated with 50 ng/ml
platelet-derived growth factor (PDGF) (Upstate Biotechnology Inc.) for
30 min before insulin stimulation. After stimulation, the cells were
lysed at 4 °C with ice-cold phosphate-buffered saline (2 ml/10-cm
dish of L6 cells) containing 10% Triton X-100, 1 mM
phenylmethylsulfonyl fluoride, and 100 µM
Na3VO4. Insoluble material was removed by
centrifugation at 15,000 × g for 10 min at 4 °C.
One-ml quantities of the cell lysates (~1 mg of total protein) were
incubated with anti-IRS-1, anti-IRS-2, or
PY antibody (10 µg each)
for 1 h and precipitated by incubation with 20 µl of protein
A-Sepharose (Pharmacia Biotech Inc.) for 1 h. The immunocomplexes
were washed five times with phosphate-buffered saline containing 10%
Triton X-100 and boiled in 30 µl of Laemmli sample buffer containing
10 mM dithiothreitol. The proteins were resolved on 7.5%
SDS-polyacrylamide gel electrophoresis and transferred to
nitrocellulose membranes (Schleicher & Schuell). After blocking with
TBS-T (10 mM Tris, pH 7.6, 150 mM NaCl, 0.05% Tween 20) containing 3% bovine serum albumin for 1 h, the
membranes were incubated with the appropriate antibodies. The proteins
were visualized by blotting with 125I-protein A or by
enhanced chemiluminescence (ECL) using horseradish peroxidase-labeled
anti-rabbit or mouse IgG (Amersham Corp.). In some experiments using
125I-protein A, the signals were analyzed
stoichiometrically using Bioimage analyzer BAS2000 (Fuji).
L6 Cells Express IRS-1 and IRS-2, Which Are Tyrosine-phosphorylated
in Response to Insulin Stimulation
7 M insulin for 3 min, cell lysates were
immunoprecipitated with anti-IRS-1 or anti-IRS-2 antibody.
Immunoblotting with
PY antibody revealed tyrosine phosphorylation of
IRS-1 and IRS-2 (lanes 5 and 6). The migration of
IRS-2 being slightly behind of that of IRS-1 on SDS-polyacrylamide gel
electrophoresis is consistent with the apparent molecular mass of IRS-2
being about 10 kDa higher than that of IRS-1, as reported (12). These
results demonstrated that L6 cells express IRS-1 and IRS-2, both of
which are tyrosine-phosphorylated in response to insulin stimulation.
In this experiment, the IRS-1 immunoprecipitate derived from 5 µl of
cell lysate and the IRS-2 immunoprecipitate derived from 15 µl of
cell lysate yielded equivalent PY signals in the
PY immunoblotting
(data not shown). As the immunoprecipitation efficiencies of anti-IRS-1
and anti-IRS-2 antibodies were comparable, the relative amount of the
PY form of IRS-2 was roughly 30% of that of IRS-1.
Fig. 1.
IRS-1 and IRS-2 expression, insulin-induced
tyrosine phosphorylation, and PI 3-kinase activation in L6 cells.
Panel A, IRS-1 and IRS-2 expression, insulin-induced
tyrosine phosphorylation. L6 cells were incubated with
107 M insulin for 3 min. Cell lysates were
immunoprecipitated with anti-IRS-1 (
IRS-1) or anti-IRS-2 (
IRS-2)
antibody. The immunoprecipitates were resolved by SDS-polyacrylamide
gel electrophoresis, transferred to nitrocellulose membranes, and
immunoblotted with the indicated antibodies. Panel B,
insulin concentration dependence of insulin-induced IRS-1 and IRS-2
tyrosine phosphorylation. L6 cells were incubated with the indicated
concentrations of insulin for 3 min. Cell lysates were
immunoprecipitated with
IRS-1 or
IRS-2 antibody and immunoblotted with
PY antibody (upper panel) and with
IRS-1 or
IRS-2 antibody (lower panel). Panel C, insulin
concentration dependence of IRS-1- and IRS-2-associated PI 3-kinase
activity. PI 3-kinase activities in the immunoprecipitates derived with
IRS-1 or
IRS-2 antibody were assayed as described (4). Fold
increases in basal PI 3-kinase activity are presented. Bars
indicate S.E. from three independent experiments.
[View Larger Version of this Image (33K GIF file)]
PY blotting, we did a
PY blot of 0.5 volume (1/2 ×, 1/4 ×, 1/8 ×) of the samples of IRS-1
and IRS-2 immnunoprecipitated at 3 min after the addition of insulin,
which were those most heavily tyrosine-phosphorylated. At 1/2 ×, 1/4
×, and 1/8 ×, PY signals of IRS-1 and IRS-2 immunoprecipitates were
detectable, but no signal was seen at 1/16 × (data not shown).
These results indicate that the amount of IRS-2 phosphorylation
decreased to less than one-eighth of the maximal level beyond 10 min.
Fig. 2.
Time courses of insulin/IGF-I-induced IRS-1
and IRS-2 tyrosine phosphorylation and associated PI 3-kinase activity.
Panel A, time courses of insulin/IGF-I-induced IRS-1 and
IRS-2 tyrosine phosphorylation in L6 cells. L6 cells were stimulated
with 107 M insulin or IGF-I for the indicated
periods. Cell lysates were immunoprecipitated with
IRS-1 or
IRS-2
antibody and subsequently immunoblotted with
PY antibody
(upper panel) and with
IRS-1 or
IRS-2 antibody
(lower panel). Panel B, time courses of the insulin-induced tyrosine phosphorylation of IRS-1 and IRS-2 in rat
skeletal muscle. Rats were fasted for 12 h before experiments, and
4 ml of normal saline with or without 10
5 M
insulin was injected through the portal vein. After the indicated periods, the soleus muscle was removed and homogenized as described under "Experimental Procedures." Cell lysates were
immunoprecipitated with
IRS-1 or
IRS-2 antibody and immunoblotted
with
PY antibody (upper panel) and with
IRS-1 or
IRS-2 antibody (lower panel). The data are representative
of three independent experiments. Panel C, time courses of
IRS-1- and IRS-2-associated PI 3-kinase activity in L6 cells. After
stimulation with 10
7 M insulin or IGF-I, PI
3-kinase activities in the immunoprecipitates derived with
IRS-1 or
IRS-2 antibody were assayed as described. The relative PI 3-kinase
activity, as a percent of the maximal increase from the basal level, is
indicated. Bars indicate S.E. from three experiments.
[View Larger Version of this Image (38K GIF file)]
Fig. 3.
Effects of Na3VO4 and
wortmannin on the phosphorylation of IRS-1 and IRS-2. L6 cells
were incubated with 2 mM Na3VO4 for
15 min or 10 nM wortmannin (WT) for 10 min, then
stimulated with 107 M insulin for the
indicated periods. Cell lysates were immunoprecipitated with
IRS-1
or
IRS-2 antibody and immunoblotted with
PY antibody.
[View Larger Version of this Image (26K GIF file)]
p85 Mutant Inhibit the Faster
Dephosphorylation of IRS-2
p85 mutant of PI 3-kinase, which cannot bind to the p110
catalytic subunit and contains a hemagglutinin epitope tag, was
overexpressed using an adenovirus expression system. The expression of
p85 protein was confirmed by immunoprecipitation with anti-p85
antibody and immunoblotting with anti-hemagglutinin antibody (Fig.
4A). Fig. 4A (left
panel) shows the anti-p85 immunoblotting of anti-p85
immunoprecipitates from the control cells (overexpressing lacZ, lane 1) and the cells overexpressing
p85
(lane 2).
p85 overexpression reached a level
approximately 8-fold that of the wild type p85 protein, which was
assessed by blotting with 125I-protein A and quantitated
with BAS2000. The expression of
p85 protein was also confirmed by
immunoprecipitation with anti-p85 antibody followed by immunoblotting
with anti-hemagglutinin antibody (Fig. 4A, right
panel).
p85 protein was detected by anti-hemagglutinin blotting
in L6 cells overexpressing
p85 (lane 4), but not in control cells (lane 3). The PI 3-kinase activities of
PY
antibody immunoprecipitates from control cells and from those
expressing
p85 were then compared. The PI 3-kinase activity was
suppressed by
p85 overexpression to 60 and 15% of that of control
cells in the absence and presence of insulin, respectively (Fig.
4B).
Fig. 4.
Overexpression of p85 and the effect of
p85 on phosphorylation of IRS-1 and IRS-2. Panel A, the
expression of
p85 protein was confirmed by immunoprecipitation with
anti-p85 (
p85) antibody followed by immunoblotting with anti-p85 and
anti-hemagglutinin (
HA) antibody. Lanes 1 and
3, lacZ-overexpressing L6 cells as a control;
lanes 2 and 4,
p85-overexpressing cells.
Panel B, PI 3-kinase activities in immunoprecipitates
derived with
PY antibody from control cells and cells overexpressing
p85 with or without 10
7 M insulin
stimulation for 3 min. Panel C, L6 cells overexpressing
p85 were stimulated with 10
7 M insulin for
the indicated periods. Cell lysates were immunoprecipitated with
IRS-1 or
IRS-2 antibody and immunoblotted with
PY
antibody.
[View Larger Version of this Image (26K GIF file)]
p85 affected neither the binding
of Grb2 to IRS-1 and IRS-2 nor mitogen-activated protein kinase
activation in response to insulin stimulation in L6 cells (data not
shown). These observations were consistent with those made in 3T3-L1
adipocytes (13).
p85 are shown in Fig. 4C. With
p85 overexpression, IRS-2 dephosphorylation was markedly inhibited, whereas no change was observed in IRS-1. These results suggest that PI
3-kinase activation is necessary for the rapid dephosphorylation of
IRS-2.
7 M insulin for the indicated periods. Fig.
5A shows that the PI 3-kinase activity
associated with phosphotyrosine was increased 10-24-fold by PDGF
stimulation prior to insulin stimulation compared with insulin
stimulation alone, whereas insulin-induced IRS-associated PI 3-kinase
activations were remarkably decreased. This is considered to be
attributable to most of the PI 3-kinase having associated with
autophosphorylated PDGF receptor due to the preceding treatment with
PDGF, resulting in a decrease in the PI 3-kinase associated with IRSs.
Under these conditions, IRS-2 dephosphorylation was significantly
inhibited (Fig. 5B, lower right). These
observations support the notion that, rather than total PI 3-kinase
activity, it is IRS-associated PI 3-kinase activity that is important
for the rapid dephosphorylation of IRS-2.
Fig. 5.
Effect of PDGF stimulation on associated PI
3-kinase activity and tyrosine phosphorylation of IRS-1 and IRS-2.
L6 cells were stimulated with 50 ng/ml PDGF for 30 min, then stimulated with 107 M insulin for the indicated periods.
Panel A, PI 3-kinase activities in the immunoprecipitates
derived with
PY,
IRS-1, and
IRS-2 antibodies were assayed as
described. The relative PI 3-kinase activity as a percent of the
maximal increase from the basal level or fold increase is indicated.
Bars indicate S.E. from three experiments. Panel
B, cell lysates were immunoprecipitated with
IRS-1 or
IRS-2 antibody and immunoblotted with
PY antibody.
[View Larger Version of this Image (28K GIF file)]
, IRS-1 inhibits insulin receptor tyrosine kinase
activity, whereas IRS-2 does not (25). In this study, we demonstrated
that IRS-2 is dephosphorylated more rapidly and activates PI 3-kinase
more transiently than IRS-1 in skeletal muscle cells. This is the first
report detailing the different features of IRS-1 and IRS-2 in tyrosine
phosphorylation and signal transduction to downstream molecules.
p85 mutant was overexpressed, IRS-2
dephosphorylation was markedly inhibited, suggesting that PI 3-kinase
activation is necessary for rapid dephosphorylation of IRS-2. It is
unlikely that
p85 binds to phosphotyrosine residues of IRS-1 and
IRS-2 in a nonspecific manner because the overexpression of
p85
affected neither the binding of Grb2 to IRS-1 and IRS-2 nor
mitogen-activated protein kinase activation. When IRS-1- and
IRS-2-associated PI 3-kinase was reduced, despite PI 3-kinase activity
with phosphotyrosine being increased, by sequential addition of PDGF
and insulin, IRS-2 dephosphorylation was reduced. We have come to the
conclusion that IRS-associated PI 3-kinase activity rather than total
PI 3-kinase activity is important for rapid dephosphorylation of IRS-2.
*
This work was supported by a grant-in-aid for scientific
research from the Japanese Ministry of Education and by a grant from TMFC.The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
To whom correspondence should be addressed. Tel.:
81-3-3815-5411 (ext. 3133); Fax: 81-3-5803-1874; E-mail:
asano-tky{at}umin.u-tokyo.ac.jp.
1
The abbreviations used are; IRS, insulin
receptor substrate; PI, phosphatidylinositol; PY,
anti-phosphotyrosine;
p85, dominant negative p85 mutant;
Na3VO4, sodium orthovanadate; IGF-I,
insulin-like growth factor I; PDGF, platelet-derived growth
factor.
©1997 by The American Society for Biochemistry and Molecular Biology, Inc.