Báo cáo khoa học: Phosphatidylinositol 3,4,5-trisphosphate modulation in SHIP2-deficient mouse embryonic fibroblasts pot

Phosphatidylinositol 3,4,5-trisphosphate modulation
in SHIP2-deficient mouse embryonic fibroblasts
Daniel Blero
1
, Jing Zhang
1
, Xavier Pesesse
1
, Bernard Payrastre
2
, Jacques E. Dumont
1
,
Ste
´
phane Schurmans
3
and Christophe Erneux
1
1 Interdisciplinary Research Institute (IRIBHM), Universite
´
Libre de Bruxelles, Belgium
2 INSERM U563, Departement d’Oncogenese et Signalization dans les Cellules Hematopoietiques, Ho
ˆ
pital Purpan, Toulouse Cedex, France
3 IRIBHM, IBMM, Gosselies Belgique
The SHIPs (SH2 domain containing inositol 5-phos-
phatases) are members of the inositol 5-phosphatase
family. Two isoenzymes, named SHIP1 and SHIP2
have been identified and characterized [1–4]. The cellu-
lar and tissue distribution of SHIP2 is very wide [5],

particularly in cells that do not express SHIP1 (e.g. in
heart, muscle or adipocytes). Tyrosine phosphorylation
of SHIP2 occurs in response to treatment of cells with
various stimuli, e.g. epidermal growth factor (EGF),
platelet-derived growth factor (PDGF), insulin or
macrophage colony-stimulating factor (M-CSF) but
the biological significance of this phosphorylation is
unknown [6–8]. In 3T3-L1 preadipocytes, SHIP2
translocation to the plasma membrane occurs in
response to insulin or PDGF. In this model, SHIP2
translocation does not seem to require its tyrosine
Keywords
inositol 5-phosphatase; mouse embryonic
fibroblasts; phosphatidylinositol
3,4,5-trisphosphate; SH2 domain; signal
transduction.
Correspondence
C. Erneux, Institute of Interdisciplinary
Research (IRIBHM), Campus Erasme
Building C, 808 Route de Lennik, 1070
Brussels, Belgium
Fax: +32 2 555 4655
Tel: +32 2 555 4162
E-mail:
Note
D. Blero and J. Zhang contributed equally to
this work.
(Received 27 July 2004, revised 6 February
2005, accepted 21 March 2005)
doi:10.1111/j.1742-4658.2005.04672.x

SHIP2, the ubiquitous SH2 domain containing inositol 5-phosphatase,
includes a series of protein interacting domains and has the ability to
dephosphorylate phosphatidylinositol 3,4,5-trisphosphate [PtdIns(3,4,5)P
3
]
in vitro. The present study, which was undertaken to evaluate the impact of
SHIP2 on PtdIns(3,4,5)P
3
levels, was performed in a mouse embryonic
fibroblast (MEF) model using SHIP2 deficient (– ⁄ –) MEF cells derived
from knockout mice. PtdIns(3,4,5)P
3
was upregulated in serum stimulated
– ⁄ – MEF cells as compared to + ⁄ + MEF cells. Although the absence of
SHIP2 had no effect on basal PtdIns(3,4,5)P
3
levels, we show here that
this lipid was significantly upregulated in SHIP2 – ⁄ – cells but only after
short-term (i.e. 5–10 min) incubation with serum. The difference in
PtdIns(3,4,5)P
3
levels in heterozygous fibroblast cells was intermediate
between the + ⁄ + and the – ⁄ – cells. In our model, insulin-like growth
factor-1 stimulation did not show this upregulation. Serum stimulated
phosphoinositide 3-kinase (PI 3-kinase) activity appeared to be comparable
between + ⁄ + and – ⁄ – cells. Moreover, protein kinase B, but not mitogen
activated protein kinase activity, was also potentiated in SHIP2 deficient
cells stimulated by serum. The upregulation of protein kinase B activity
in serum stimulated cells was totally reversed in the presence of the
PI 3-kinase inhibitor LY-294002, in both + ⁄ + and – ⁄ – cells. Altogether,

these data establish a link between SHIP2 and the acute control of
PtdIns(3,4,5)P
3
levels in intact cells.
Abbreviations
CHO-IR, chinese hamster ovary cells overexpressing the insulin receptor; EGF, epidermal growth factor; FBS, foetal bovine serum;
FGF, fibroblast growth factor; HGF, hepatocyte growth factor; IGF, insulin-like growth factor; MAP, mitogen activated protein; M-CSF,
macrophage colony-stimulating factor; MEF, mouse embryonic fibroblast; PDGF, platelet-derived growth factor; PI 3-kinase, phosphoinositide
3-kinase; PKB, protein kinase B; PtdIns(3,4)P
2
, phosphatidylinositol 3,4-bisphosphate; PtdIns(3,4,5)P
3
, phosphatidylinositol
3,4,5-trisphosphate; PtdIns4P, phosphatidylinositol 4-phosphate; PtdIns(4,5)P
2
, phosphatidylinositol 4,5-bisphosphate; PTEN, phosphate and
tension homolog deleted on chromosome 10; SHIP, SH2 domain containing inositol phosphatase.
2512 FEBS Journal 272 (2005) 2512–2522 ª 2005 FEBS
phosphorylation [9]. As phosphatidylinositol 3,4,5-tris-
phosphate [PtdIns(3,4,5)P
3
] is a major intracellular sig-
nal generated by insulin and insulin-like growth factor
(IGF)-1, it has been suggested that SHIP2 is a physio-
logically negative regulator of their signalling. Indeed,
overexpression of SHIP2 inhibited insulin-induced glu-
cose uptake and glycogen synthesis in 3T3-L1 adipo-
cytes and L6 myotubes [10,11]. SHIP2 also appears to
inhibit the insulin-induced phosphorylation of Akt2,
but not Akt1, in 3T3-L1 adipocytes. Upon insulin sti-

mulation, SHIP2 is translocated to the plasma mem-
brane, where it inhibits the insulin-specific subcellular
redistribution of Akt2 [12]. The expression of SHIP2
was enhanced in an animal model of type 2 diabetes
which was accompanied by an attenuation of insulin
signalling [13]. However, a role for SHIP2 has also
been suggested in other pathways: in rat vascular
smooth muscle cells, SHIP2 downregulates PDGF and
IGF-1 mediated signalling downstream of PI 3-kinase
[14]. In glioblastoma cells, SHIP2 inhibits protein kin-
ase B (PKB) and provokes a potent cell cycle arrest in
G
1
[15]. SHIP2 could play an essential role in cell
adhesion and spreading as shown in HeLa cells [16]. A
regulatory role for SHIP2 in M-CSF-induced signalling
has been recently suggested [8]. SHIP2 also functions
in the maintenance and dynamic remodelling of actin
structures as well as in endocytosis and downregula-
tion of the EGF receptor [17].
In vivo, homozygous disruption of SHIP2 by remo-
ving exons 19–29 causes severe hypoglycemia and
death within a few hours after birth. Heterozygous dis-
ruption of this gene leads to hypersensitivity to insulin
demonstrated by the increased glycogen synthesis in
skeletal muscles in response to insulin. Injection of
d-glucose resulted in a more rapid glucose clearance
in SHIP2+ ⁄ – than in SHIP2+ ⁄ + mice. Moreover,
the incidences of spontaneous or irradiated-induced
tumours were not affected in SHIP2+ ⁄ – mice [18].

Removal of exons 1–18 of SHIP2 resulted in a differ-
ent phenotype: the mice were viable and had no
increased insulin sensitivity but they were smaller in
body weight and length, and were highly resistant to
weight gain when placed on a high-fat diet [19]. The
reason for this discrepancy between the two pheno-
types is currently not understood but several explana-
tions have been proposed [19].
We and others previously reported that SHIP2
displays inositol 5-phosphatase activity when
PtdIns(3,4,5)P
3
and phosphatidylinositol 4,5-bisphos-
phate [PtdIns(4,5)P
2
] were used as substrate in vitro.
Inositol tetrakisphosphate was also a substrate of the
enzyme expressed in bacteria [15,20,21]. Moreover, both
in COS-7 cells and in chinese hamster ovary cells over-
expressing the insulin receptor (CHO-IR) cells transfected
with SHIP2, the levels of PtdIns(3,4,5)P
3
were decreased
in both EGF and insulin stimulated cells [22,23]. This
was also observed in rat vascular smooth muscle cells,
where PtdIns(3,4,5)P
3
levels were decreased in SHIP2
transfected cells stimulated by PDGF or IGF-1 [14].
Both PKB and mitogen activated protein (MAP) kinase

activities were also decreased in SHIP2 transfected cells
suggesting that SHIP2 is a down-regulator of both arms
of receptor tyrosine kinase activation [10,15,22,23]. The
present study was therefore undertaken to establish the
extent of PtdIns(3,4,5)P
3
regulation in SHIP2 – ⁄ – cells
derived from MEF cells. Although the absence of
SHIP2 had no effect on basal PtdIns(3,4,5)P
3
levels, we
show here that this lipid was significantly upregulated
in SHIP2 – ⁄ – MEF cells but only after short-term
(i.e. 5–10 min) incubation with serum. In our model,
IGF-1 stimulation did not show this upregulation
and PtdIns(4,5)P
2
levels were comparable between
SHIP2+ ⁄ + and – ⁄ – MEF cells.
Results
Status of SHIP2, PTEN, insulin and IGF-1 receptor
expression in SHIP2 +/+ and –/– MEF cells
SHIP2 – ⁄ – mice were obtained as reported previously
[18]. As our SHIP2 – ⁄ – mice died very shortly after
birth, we chose to work with MEF cells as a model
to measure the 3-phosphorylated phosphoinositides.
MEF cells were prepared from embryos of hetero-
zygous crosses and genotyped by PCR analysis. Two
series of MEF cells (1 and 2) were prepared from
two independent crosses to validate the measurements

of phosphoinositides (see below). Western blot analy-
sis of SHIP2 was performed to confirm the absence
of expression of SHIP2 in – ⁄ – MEF cells (Figs 1A
and B). The expression of SHIP2 in + ⁄ – MEF cells
was decreased as compared to wild type (+ ⁄ +)
MEF cells (Fig. 1A) as reported previously [18].
Although we detected the presence of the IGF-1
receptor in MEF cells by western blotting, the b sub-
unit of the insulin receptor was not be seen by this
method suggesting that it is either not expressed or
was below the detection level of the antibodies used
in our immunodetection method (Fig. 1B). The
expression of the PtdIns(3,4,5)P
3
3-phosphatase, phos-
phatase and tension homolog deleted on chromosome
10 (PTEN) [24,25] was not significantly modified
between SHIP2+ ⁄ + and – ⁄ – MEF cells (Fig. 1B). No
changes in expression of the regulatory subunits of PI
3-kinase p85 were seen between the two types of cells
(data not shown).
D. Blero et al. Phosphatidylinositol 3,4,5-trisphosphate levels in mouse embryonic fibroblasts
FEBS Journal 272 (2005) 2512–2522 ª 2005 FEBS 2513
No change in PtdIns(4,5)P
2
levels in SHIP2
+/+ and –/– MEF cells
As in vitro, PtdIns(4,5)P
2
is also a substrate of SHIP2

[15], we compared the levels of [
3
H]PtdIns(4,5)P
2
and
[
3
H] phosphatidylinositol 4-phosphate (PtdIns4P) after
labelling the cells with [
3
H]inositol in the presence of
10% FBS for 72 h: the amount of [
3
H]PtdIns(4,5)P
2
and [
3
H]PtdIns4P did not change significantly between
+ ⁄ + and – ⁄ – MEF cells (Fig. 2A). Similar results
were obtained when we labelled the cells with
[
32
P]orthophosphate for more than 4 h. In our assay
SHIP2
MEF SHIP2
+/+
1
MEF SHIP2
+/+
2

MEF SHIP2
+/
-
MEF SHIP2
-
/
-
2
MEF SHIP2
-
/
-
1
CHO
-
IR
150
100
75
250
kDa
AB
SHIP2
InsR
IGF-IR
PTEN
MEF SHIP2
+
/+
2

MEF SHIP2
-
/
-
1
MEF
SHIP2
+/+
1
MEF SHIP2
-
/
-
2
CHO
-
IR
Fig. 1. Western blot analysis of MEF
SHIP2 + ⁄ +, + ⁄ –and–⁄ – cells. Twenty
micrograms of proteins from a lysate made
of MEF cells or CHO-IR were applied to
SDS gels. Immunodetection was performed
with antibodies against SHIP2, PTEN, IGF-1
and the insulin receptor (IGF-1R and InsR).
MEF cells 1 and 2 were from two
independent preparations of cells.
PtdIns(3,4,5)P
3
levelsLabelling with [
3

H] inositol
0
0,2
0,4
0,6
0,8
1
1,2
010203040
TIME (minute)
%(of PtdIns(4,5)P2)
SHIP2+/+
SHIP2-/-
0
1
2
3
4
5
6
AB
CD
PtdInsP PtdIns(4,5)P2
% (of PtdIns)
SHIP2+/+
SHIP2-/-
PtdIns(3,4,5)P
3
levels
0

0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
0 min 10 min
%(of PtdIns(4,5)P2)
SHIP2+/+
SHIP2-/-
PtdIns(3,4,5)P
3
levels
0
0,2
0,4
0,6
0,8
1
1,2
1,4
0 min 10 min
%(of PtdIns(4,5)P2)
SHIP2+/+
SHIP2+/-
SHIP2-/-
Fig. 2. PtdIns(3,4,5)P

3
levels in serum stimulated SHIP2 + ⁄ +, + ⁄ –and–⁄ – MEFs. (A) MEF cells were labelled with 50 lCi [
3
H]inositol for
72 h in the presence of FBS. [
3
H]phosphoinositides were isolated as described. The data were normalized with respect to the total radioac-
tivity present in the phosphatidylinositol fraction. The data are means of triplicates ± SD. (B) + ⁄ + and – ⁄ – MEF cells were labelled with
[
32
P]orthophosphate for 4 h and stimulated with 10% serum for various periods of time. [
32
P]PtdIns(3,4,5)P
3
was isolated as described. The
data are expressed as a percentage of total [
32
P]PtdIns(4,5)P
2
measured in the same HPLC profile and are means of triplicates ± SD.
(C) MEF cells were labelled with [
32
P] for 4 h and stimulated by 10% FBS for 10 min. The data are expressed as a percentage of total [
32
P]
PtdIns(4,5)P
2
measured in the same HPLC profile. The data are means of three independent experiments using the two series of MEF cells
(1 and 2) ± SD. (D) + ⁄ +,+ ⁄ – and – ⁄ – MEF cells were labelled with [
32

P] and stimulated for 10 min by 10% FBS. The data are means of
duplicates ± SD. The data are representative of two different experiments.
Phosphatidylinositol 3,4,5-trisphosphate levels in mouse embryonic fibroblasts D. Blero et al.
2514 FEBS Journal 272 (2005) 2512–2522 ª 2005 FEBS
of [
32
P]-labelled lipids, we scraped off together the
region of the TLC containing both [
32
P]PtdIns(3,4,5)P
3
and [
32
P]PtdIns(4,5)P
2
, the levels of [
32
P]-labelled
3-phosphoinositides were normalized with respect to
[
32
P]PtdIns(4,5)P
2
.
SHIP2 modulated PtdIns(3,4,5)P
3
levels after
short-term serum stimulation
The levels of PtdIns(3,4,5)P
3

were compared between
the two types of cells that had been stimulated by
serum. PtdIns(3,4,5)P
3
was increased following stimu-
lation of both types of cells by 10% FBS. In res-
ponse to the addition of serum, the production of
PtdIns(3,4,5)P
3
was upregulated in SHIP2 – ⁄ – cells as
compared to + ⁄ + cells. A maximal effect was seen
between 5 and 10 min after stimulation by 10% serum
(Figs 2B and 3A). This effect was observed in the two
independently prepared MEF cells (Fig. 2C). In these
experiments, the levels of phosphatidylinositol 3,4-bis-
phosphate; [PtdIns(3,4)P
2
] were not significantly differ-
ent between the two types of cells (see below). This
probably reflects the complex pathway of PtdIns(3,4)P
2
production ⁄ degradation in response to FBS via PI
3-kinase, PTEN, SHIP2 and several other phosphati-
dylinositol 5-phosphatases [25]. The difference in
PtdIns(3,4,5)P
3
levels between serum stimulated + ⁄ +
and – ⁄ – cells was also observed in heterozygous MEF
cells but the effect was intermediate as compared to
the + ⁄ + and – ⁄ – MEF cells. Basal levels of

PtdIns(3,4,5)P
3
were not different between + ⁄ + and
– ⁄ – cells (Fig. 2D).
In contrast to serum, IGF-1 stimulation resulted in
maximal production of PtdIns(3,4,5)P
3
after 2 min and
no significant differences in PtdIns(3,4,5)P
3
levels were
seen between SHIP2+ ⁄ +and – ⁄ – cells (Fig. 3B).
When the MEF cells were stimulated with insulin
(1–100 nm), we did not see any significant increase in
PtdIns(3,4,5)P
3
levels in contrast to CHO-IR cells sti-
mulated with insulin that were used as positive control
(data not shown).
SHIP2 did not modulate PtdIns(3,4,5)P
3
levels
after long-term serum stimulation
Previous studies in MEF cells deficient in PTEN have
shown that PtdIns(3,4,5)P
3
was upregulated about
twofold in – ⁄ – cells after 4 h of incubation in the
presence of 5% FBS. The data suggested that
PtdIns(3,4,5)P

3
is a physiological substrate of PTEN
[26]. We therefore measured PtdIns(3,4,5)P
3
levels in
SHIP2 deficient cells after long-term stimulation by
FBS (under the same conditions used in the study of
PTEN deficient MEF cells). PtdIns(3,4,5)P
3
levels in
SHIP2 – ⁄ – cells were not significantly different as com-
pared to + ⁄ + cells after 30 min or 4 h of stimulation
by 5% FBS (Figs 2B and 4, respectively). Therefore,
no significant differences in PtdIns(3,4,5)P
3
levels
between SHIP2 + ⁄ + and – ⁄ – were observed under
the conditions where PtdIns(3,4,5)P
3
levels were upreg-
ulated in PTEN deficient cells.
PI 3-kinase activity in SHIP2 +/+ and –/– MEF
cells
PI 3-kinase activity in SHIP2 + ⁄ + and – ⁄ – cells was
determined in both the presence and absence of the PI 3-
kinase inhibitor LY-294002. Basal activity was stimula-
ted in the presence of 10% FBS or IGF-1 at 10 nm. This
activity was reversed when lysates were prepared in the
presence of LY-294002 (Fig. 5). No differences in the
0

0,2
0,4
0,6
0,8
1
1,2
0510
TIME (minute)
0510
TIME (minute)
%(of PtdIns(4,5)P
2
)
%(of PtdIns(4,5)P
2
)
SHIP2+/+
SHIP2-/-
A
PtdIns(3,4,5)P
3
levels
PtdIns(3,4,5)P
3
levels
+serum
0
0,1
0,2
0,3

0,4
0,5
0,6
SHIP2+/+
SHIP2-/-
B
+ IGF-1
Fig. 3. PtdIns(3,4,5)P
3
levels in serum and IGF-1 stimulated MEF
cells. Time course of [
32
P] PtdIns(3,4,5)P
3
production in (A) serum
and (B) IGF-1 stimulated MEF cells. + ⁄ + and – ⁄ – MEF cells were
labelled with [
32
P]orthophosphate for 4 h and stimulated with 10%
serum or IGF-1 at 10 n
M for the indicated times. [
32
P] PtdIns(3,4,5)P
3
was quantified as before.
D. Blero et al. Phosphatidylinositol 3,4,5-trisphosphate levels in mouse embryonic fibroblasts
FEBS Journal 272 (2005) 2512–2522 ª 2005 FEBS 2515
level of activation was determined in the SHIP2 + ⁄ +
and – ⁄ – MEF cells at the time of maximal production of
PtdIns(3,4,5)P

3
, i.e. 5 min for FBS and 2 min for IGF-1.
Therefore, changes in PI 3-kinase activity are probably
not responsible for the upregulation of PtdIns(3,4,5)P
3
levels in SHIP2 – ⁄ – MEF cells.
Effect of serum on PKB and MAP kinase activities
In overexpression studies, SHIP2 causes PKB inactiva-
tion and MAP kinase inhibition [10,15,22,23]. Activa-
ted PKB was detected using phosphospecific antibodies
against T308 and S473. PKB activity was upregulated
in serum stimulated SHIP2 –⁄ – cells as compared to
+ ⁄ + cells (Fig. 6A). A similar result was obtained by
using an enzymatic assay for PKB after immunopre-
cipitation of PKB. The net increase in PKB activity in
serum stimulated cells was approximately two times
higher in SHIP2 – ⁄ – cells as compared to + ⁄ + cells
(Fig. 6B). We also showed that the upregulation of
PKB phosphorylation was totally reversed when cells
were preincubated in the presence of the PI 3-kinase
inhibitor LY-294002 (Fig. 7). In contrast, MAK kinase
activities (p-Erk1 ⁄ 2) following serum stimulation were
not different between the two types of cells (Fig. 6A).
Stimulation of MEF with various agonists
MEF cells were also stimulated for 5 min by various
agonists: EGF, hepatocyte growth factor (HGF), b
fibroblast growth factor (FGF), IGF-1 and PDGF
(Fig. 8). HGF, b FGF, IGF-1 (1 nm), did not increase
PtdIns(3,4,5)P
3

and PtdIns(3,4)P
2
production. EGF sti-
mulated both PtdIns(3,4,5)P
3
and PtdIns(3,4)P
2
levels
but the differences in the two lipid levels were not signifi-
cant between the two types of cells (SHIP2 deficient or
not). IGF-1 at 10 nm stimulated PtdIns(3,4,5)P
3
produc-
tion but PtdIns(3,4,5)P
3
levels were not significantly
upregulated in the – ⁄ – cells compared to the + ⁄ + cells
SHIP2 MEF +/+
-/-
+/+
-/-
Origin
PI3P
Control Control with LY No lysates
A
SHIP2 MEF
+/+
-/-
+/+
-/-

Origin
PI3P
FBS10%
FBS10% +LY
B
+/+
-/-
+/+
-/-
Origin
PI3P
SHIP2 MEF
IGF-1 10 nM IGF-1 10 nM +LY
C
-
LY + LY
Fig. 5. PI 3-kinase activity in serum or IGF-1
stimulated SHIP2 + ⁄ +and–⁄ – MEF cells
PI 3-kinase activty was measured in (A) con-
trol, and in stimulated cells (B) 10% serum
for 5 min and (C) 10 n
M IGF-1 for 2 min.
The assay was performed as described.
When used, LY-294002 (25 l
M) was added
to the preincubation for 30 min.
PtdIns(3,4,5)P
3
levels
0

0.05
0.1
0.15
0.2
0.25
MEF SHIP2+/+ MEF SHIP2-/-
% (of PtdIns(4,5)P2)
+ serum
Fig. 4. PtdIns(3,4,5)P
3
levels in serum stimulated SHIP2 + ⁄ + and
– ⁄ – MEF cells after long-term stimulation. + ⁄ + and – ⁄ – MEF cells
were labelled with [
32
P] and stimulated with 5% serum for 4 h.
[
32
P]PtdIns(3,4,5)P
3
was quantified as before.
Phosphatidylinositol 3,4,5-trisphosphate levels in mouse embryonic fibroblasts D. Blero et al.
2516 FEBS Journal 272 (2005) 2512–2522 ª 2005 FEBS
as shown before in the kinetics of the lipid production
(Fig. 3B). In PDGF stimulated cells, PtdIns(3,4,5)P
3
levels increased but no significant differences were seen
between +⁄ + and – ⁄ – cells although PtdIns(3,4)P
2
was
lower in SHIP2 deficient cells as compared to control

cells (Fig. 8). In order to verify that the agonists used
stimulated MEF cells efficiently, we determined MAP
kinase activity (p-Erk1 ⁄ 2) in the two types of cells: the
agonists tested also stimulated phospho MAP kinase
activity in both SHIP2 + ⁄ + and – ⁄ – cells (Fig. 9).
Discussion
SHIP2 is a typical signalling enzyme potentially
involved in the biochemical cascade of many growth
factors and insulin [2,4,7,10–15,23]. Its sequence shows
the presence of an SH2 domain, proline rich sequences,
a NPXY site that can be phosphorylated on tyrosine
and a catalytic domain which is typical for a member
of the inositol 5-phosphatase family. SHIP2 appears to
be able to dephosphorylate at the 5-position of the
inositol ring of PtdIns(4,5)P
2
, PtdIns(3,4,5)P
3
and ino-
sitol tetrakisphoshate in vitro [7,15,21].
SHIP2
p-ser
473
PKB
A
B
p-thr
308
PKB
total PKB

p-Erk1/2
Erk2
1 2 3 4 5 6 7 8
Western blotting
PKB assay
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
0min 10min FBS
CPM
SHIP2+/+
SHIP2-/-
Fig. 6. PKB and MAP kinase activities in SHIP2 + ⁄ + and – ⁄ – MEF
cells. (A) SHIP2 + ⁄ + (lane 1,2 and 5,6) and – ⁄ – (lane 3,4 and 7,8)
MEF cells were stimulated by 10% FBS for 10 min (lane 1–4
unstimulated, 5–8 stimulated by FBS). Protein (20 lg) wase immuno-
blotted with the indicated antibodies. Phosphorylation was assayed
by using phospho-specific antibodies against p-Thr308 or p-Ser473
for PKB and p-Erk1 ⁄ 2 for MAK kinase activities. Western blot
against total PKB or Erk2 is shown below. Results are representa-
tive of three experiments. (B) PKB activities determined after
immunoprecipitation of endogenous PKB as described. Protein cell
lysate (100 lg) was used in the assay. Results are means of dupli-

cates ± SD.
SHIP2 MEF
+/+
-/-
+/+
-/-
+/+
-/-
+/+
-/-
LY 294002
FBS 10%
FBS 10% + LY294002
Control
+/+
-/-
+/+
-/-
+/+
-/-
+/+
-/-
P-Ser
473
PKB
Total PKB
Fig. 7. Phospho PKB activity in SHIP2 + ⁄ +and–⁄ – cells MEF cells.
MEF cells were stimulated in the 10% serum for 5 min in the pres-
ence and absence of LY-294002 (25 l
M). Protein (20 lg) was

immunoblotted. Phosphorylation of PKB was assayed in the pres-
ence of antibodies against p-Ser473 and total PKB is shown below.
PtdIns(3,4)P
2
levels in MEF cells (5 min stimulation)
0
0,5
1
1,5
2
2,5
3
3,5
NS FBS 10% EGF
50n
g
/ml
HGF
15n
g
/ml
β-FGF
100n
g
/ml
IGF-1
1nM
IGF-1
10nM
PDGF

30n
g
/ml
%(of tP dI sn4(,5P)2)
SHIP2+/+
SHIP2-/-
0
0,2
0,4
0,6
0,8
1
1,2
S
N
BF
S
%01

G
E
F
5
0
g
n
m
/
l
GH

F
1

n
5
g
m
/
l
β-
G
FF01

0
n
g
/
m
l
I
G
-F
1
n
1
M
GI
F
-
1

1
n0
M
P
D
FG
3
n0
g
/
lm
SHIP2+/+
SHIP2-/-
PtdIns(3,4,5)P
3
levels ( 5 min stimulation)
%(of tP dI sn4(5,P)2)
Fig. 8. PtdIns(3,4,5)P
3
and PtdIns(3,4)P
2
levels in SHIP2 + ⁄ +and
– ⁄ – MEF cells. + ⁄ + and – ⁄ – MEF cells were stimulated by 10%
FBS, EGF 50 ngÆmL
)1
, HGF 15 ngÆmL
)1
, b FGF 100 ngÆmL
)1
, IGF-1

1 and 10 n
M or PDGF 30 ngÆmL
)1
for 5 min. [
32
P]PtdIns(3,4,5)P
3
(upper panel) and [
32
P]PtdIns(3,4)P
2
(lower panel) are expressed as
a percentage of total [
32
P]PtdIns(4,5)P
2
and are means of dupli-
cates ± SD. NS ¼ non stimulated cells.
D. Blero et al. Phosphatidylinositol 3,4,5-trisphosphate levels in mouse embryonic fibroblasts
FEBS Journal 272 (2005) 2512–2522 ª 2005 FEBS 2517
The ability of SHIP2 to be a physiological regulator
of PtdIns(3,4,5)P
3
is often assumed based on a similar
function attributed to SHIP1 [27,28]. Two SHIP2
knockout mice have now been reported. The first
knockout mice showed an increased sensitivity to insu-
lin although it was later recognized that another gene
Phox2a was also deleted in the targeting construct
together with the 19–29 exons of SHIP2 [18,29]. It is

not known whether the phenotype was influenced by
this second gene deletion. The second knockout mice,
in which the first 18 exons were deleted, showed a dif-
ferent phenotype with normal insulin and glucose tol-
erances; the mice were however, resistant to dietary
obesity [19]. Interestingly, an increased activation of
PKB phosphorylation was observed in skeletal muscle
and liver of these SHIP2 null mice on stimulation with
insulin [19]. The data obtained in analysing the pheno-
types of both knockout mice are consistent with
SHIP2 being directly responsible for the dephosphory-
lation of PtdIns(3,4,5)P
3
. This is not to say that SHIP2
is only acting as a phosphoinositide 5-phosphatase
(EC 3.1.3.36). For example, the SHIP2 C-terminal
region is quite specific as compared to that of SHIP1
and could interact with specific protein partners such
as filamin or c-Cbl associated protein thereby provi-
ding multiple molecular interactions and possible bio-
chemical regulation mechanisms of its activity and
localization [30,31]. The presence of SHIP2 in mem-
brane ruffles has been reported and this could account
for the regulation of actin rearrangement by regulating
local levels of SHIP2 lipid substrate and ⁄ or interacting
with cytoskeleton regulatory proteins [30]. The trans-
location of SHIP2 to plasma membranes upon insulin
stimulation and the requirement for the negative regu-
lation of insulin signalling could account for SHIP2
specificity [32]. The role of tyrosine phosphorylation of

SHIP2 is unclear in preadipocytes [9] and multiple
phosphorylation sites have been identified, e.g. in
Jurkat cells [33]. Thus, the biochemistry of SHIP2 in
the insulin (and probably other signalling cascades) is
not fully understood.
We were interested to measure PtdIns(3,4,5)P
3
lev-
els in SHIP2 depleted cells. As SHIP2 did not appear
to be specific for a given signalling pathway in cellu-
lar models, we compared the effects of a series of
agonists including EGF, PDGF and IGF-1 that
had been previously used in SHIP2 overexpression
studies. We also compared our data with PTEN
which is often presented as the principal regulator of
PtdIns(3,4,5)P
3
levels [25]. This is the first report
showing a direct comparison of PtdIns(3,4,5)P
3
pro-
duction in SHIP2 + ⁄ + and – ⁄ – cells. PtdIns(3,4,5)P
3
levels [but not PtdIns(4,5)P
2
levels] were potentiated
in serum stimulated SHIP2 – ⁄ – MEF cells as com-
pared to + ⁄ + cells; this effect was not observed in
unstimulated cells, or after long-term stimulation (i.e.
1 or 4 h) of the cells. PKB activity (but not MAP

kinase activity) was potentiated in serum stimulated
SHIP2 – ⁄ – cells with this effect being completely
reversed in the presence of LY-294002. Serum stimu-
lated PI 3-kinase activity appeared to be comparable
between SHIP2 + ⁄ + and – ⁄ – cells and in both
cases, the activity was decreased in the presence of
LY-294002. Therefore, the results obtained with
PtdIns(3,4,5)P
3
in our study could not be explained
by an upregulation of PI 3-kinase in serum stimulated
SHIP2 – ⁄ – cells. We concluded that the increase in
PtdIns(3,4,5)P
3
levels and PKB activity measured in
our study is a consequence of an effect on
PtdIns(3,4,5)P
3
dephosphorylation. Stambolic et al.
[26] reported that PtdIns(3,4,5)P
3
levels were also
potentiated (about twofold) in PTEN – ⁄ – cells that
had been incubated and labelled with 5% FBS for
4 h; however, no kinetics were provided for compari-
son with our data. In contrast, in our study no
change in PtdIns(3,4,5)P
3
levels were seen in SHIP2
depleted cells as compared to + ⁄ + cells either at the

basal level or after long-term stimulation. We found
that SHIP2 acted after short-term (5–10 min) serum
stimulation as a modulator of PtdIns(3,4,5)P
3
levels.
Previous data obtained with SHIP1 also indicated
that SHIP1 does not regulate basal PtdIns(3,4,5)P
3
EGF 50ng/ml
HGF 15ng/ml
FGF 100ng/ml
IGF
-
1
1
n
M
IGF 1 10nM
PDGF 30ng/ml
FBS 10%
control
SHIP2 +/+
SHIP2 -/-
EGF 50ng/ml
HGF 15ng/ml
FGF 100ng/ml
IGF
-
1
1 nM

IGF 1 10nM
PDGF 30ng/ml
FBS 10%
control
Erk2Erk2Erk2
p-Erk1/2
Fig. 9. Phospho MAP kinase activity in SHI-
P2 + ⁄ + and – ⁄ – MEF cells. + ⁄ +and–⁄ – MEF
cells were stimulated as in Fig. 8. Protein
(100 lg) was immunoblotted and probed
with p-Erk1 ⁄ 2 for MAP kinase activity. Total
MAP kinase (Erk2) is shown below.
Phosphatidylinositol 3,4,5-trisphosphate levels in mouse embryonic fibroblasts D. Blero et al.
2518 FEBS Journal 272 (2005) 2512–2522 ª 2005 FEBS
levels but that it may control the duration and mag-
nitude of stimulated increases in this lipid [25,34].
We did not detect any upregulation of
PtdIns(3,4,5)P
3
by stimulation of the cells with HGF,
b-FGF, IGF-1 or EGF. We excluded any difference in
regulation of PI 3-kinase activity between serum and
IGF-1 stimulated cells as no differences in PI 3-kinase
activity could be detected between SHIP2 + ⁄ + and
– ⁄ – MEF cells (this effect was reversed in the presence
of LY-294002). The reason for not observing an upreg-
ulation of PtdIns(3,4,5)P
3
with every agonist is not
yet understood, however, maximal production of

PtdIns(3,4,5)P
3
was observed after 5 min stimulation
by serum and after only 2–3 min by IGF-1; the ampli-
tude of the PtdIns(3,4,5)P
3
production was also not
comparable being higher for serum as compared to
IGF-1 (as shown in Fig. 3). Finally, we cannot ignore
the fact that our method does not allow the measure-
ment of minor or local changes in PtdIns(3,4,5)P
3
levels that are observed only in the presence of serum.
These observations could contribute to the specificity
we have observed with serum. However, in addition to
serum, we found that short-term H
2
O
2
treatment of
the SHIP2 – ⁄ – MEF cells also upregulates the phos-
phorylation of PKB (J. Zhang, unpublished data). This
observation suggests the following model: it is possible
that serum is producing some reactive oxygen species
which could be responsible for the inactivation of
PTEN. The production of reactive oxygen species in
response to growth factors or insulin and inactivation
of PTEN has been recently reported by others [35,36].
In this model, we assume that SHIP2 is less active than
PTEN in terms of enzymatic activity and that SHIP2

will only be able to control the PtdIns(3,4,5)P
3
levels
once PTEN is inactivated by oxidation. This mechan-
ism may be dependent on both the type of agonist and
the cell type. In another study, others have recently
reported enhanced PKB activation in response to
M-CSF in fetal liver-derived macrophages prepared
from SHIP2 knockout mice [8].
Our data also suggest that one or several compo-
nents of the serum allows SHIP2 to be effectively
recruited near the sites of PtdIns(3,4,5)P
3
production.
The localization of SHIP2 at the membrane is import-
ant for its lipid phosphatase activity as shown in 3T3-
L1 adipocytes where insulin provokes a redistribution
of SHIP2 from the cytosol to the plasma membrane
fraction following a mechanism which is in part
dependent on PI 3-kinase activity [14]. Moreover, as
discussed above, PTEN is also competing for SHIP2 in
the regulation of PtdIns(3,4,5)P
3
and PtdIns(3,4)P
2
lev-
els and this may affect the kinetics of the phosphoino-
sitides in stimulated cells. The influence of PTEN
activity in this complex pathway is not known but we
clearly established in our study that PTEN expression

is unchanged between SHIP2 wild type and SHIP2
deficient cells.
In contrast to data obtained in cells transfected with
SHIP2, no difference in MAP kinase activity was
observed between serum stimulated SHIP2 + ⁄ + and
– ⁄ – cells; this suggests that the SHIP2 pathway does not
interfere with MAP kinase activity in serum stimulated
cells.
In conclusion, our data are consistent with SHIP2
affecting the transient control of PtdIns(3,4,5)P
3
levels
which influences PKB activity, at least in cells stimula-
ted by serum. The specificity of PtdIns(3,4,5)P
3
hydro-
lysis by SHIP2 with regard to serum stimulation and
the short-term kinetics of the SHIP2 action add an
interesting twist to the sophistication of these signal-
ling systems. This concept, is quite reminiscent of some
of the characteristics of Ca
2+
signalling, e.g. the differ-
ent aspects of neuronal differentiation encoded by the
frequency of Ca
2+
transients [37].
Experimental procedures
Materials
SHIP1 and SHIP2 antibodies have been described previously

[5]. PTEN antibody was from A.G. Scientific, Inc. (San
Diego, CA, USA) Anti-PI 3-Kinase p85 and antiphosphotyr-
osine were from Upstate (AH Veemendaal, the Netherlands).
Anti-Insulin receptor (InsR) and anti IGF-1 antibodies
were kindly provided by K. Siddle (Department of Clinical
Biochemistry, Cambridge University, UK). HGF, IGF-1 and
PDGF were provided by Upstate. LY-294002, PI and phos-
phatidylserine were from Sigma (Bonnem, Belgium). Easi-
tides [c-
32
P] ATP (3000 CiÆmmol
)1
) was from NEM.
[
32
P]Orthophosphate (10 mCiÆmL
)1
) was from Amersham
(Rosendaal, the Netherlands).
PtdIns(3,4,5)P
3
measurements
Cells (1.5 · 10
6
) were cultured in 10% serum overnight.
Cells were washed twice in medium without serum and
twice in medium without either phosphate or serum. They
were labelled for at least 4 h in medium with [
32
P]ortho-

phosphate (250 lCiÆmL
)1
) but without serum. Cells were
stimulated with various agents as indicated in the text.
The reaction was terminated by 5 mL cold NaCl ⁄ Pi. Cells
were lysed in 3.75 mL 2.4 N HCl. Lipids were extracted
in 3 mL methanol and 4.5 mL CHCl
3
. After TLC and
deacylation of the phosphoinositides, separation was per-
formed by HPLC on Whatman SAX columns (Leuven,
Belgium) [23]. Radioactivity was estimated with an
online detector from Raytest (Straubenhardt, Germany).
D. Blero et al. Phosphatidylinositol 3,4,5-trisphosphate levels in mouse embryonic fibroblasts
FEBS Journal 272 (2005) 2512–2522 ª 2005 FEBS 2519
PtdIns(4,5)P
2
was also determined by labelling the cells
with [
3
H] inositol as reported [38]. The various 3-phos-
phorylated phosphoinositides standards were prepared in
insulin stimulated CHO-IR cells or in platelets as reported
previously [22,39].
Preparation of MEF cells
Primary MEF cells were isolated from 14-day postcoitus
C57BL6 mouse embryos [18]. Embryos were surgically
removed and separated from maternal tissues following pro-
tocol approved by the Ethics Committee for animals. The
head of each embryo was removed and kept for genotyping.

After visceral organ removal, the rest of the body was minced
finely by repetitive syringe aspiration, then washed twice with
1 · NaCl ⁄ Pi and incubated in 500 lL trypsin⁄ EDTA (2.5%
trypsin, 1 mm EDTA) at 37 °C for 60 min. The embryo
fragments were resuspended by adding 1.5 mL of complete
medium (DMEM, 2% streptomycin ⁄ ampicillin, 50 lm
b-mercaptoethanol) with a 2-mL glass pipette. The cells were
dissociated with a 10-mL pipette by adding another 7.5 mL
complete medium. The supernatant was transferred to a T75
culture flask after 2 min resting. MEF were obtained after
incubation of the cells at 37 °C for 2–3 days.
Genotyping of MEF cells
Genotyping of SHIP2 MEF was performed by PCR using
specific primers to amplify the neo gene and a specific exon
deleted in the recombinant allele [18]. The same forward
primer was used for each of the + ⁄ + and – ⁄ – alleles:
5¢-GGGTCTTTGGAGCTGTGGACT-3¢. While specific
reverse primers were used for the + ⁄ + allele:
5¢-CCCAAGTGTCTCCCATCATCC-3¢ and for the – ⁄ –
allele: 5¢-TAAGGGTTCCGGATCTGCC-3¢. The PCR
reaction was performed under the following conditions:
denaturation at 95 °C for 3 min, followed by 40 cycles at
95 °C for 30 s, 60 °C for 30 s, 72 °C for 30 s and elonga-
tion of 72 °C for 7 min.
Cell lysates, PKB and MAP kinase assay
MEF cells were lysed in 50 mm Tris ⁄ HCl pH 7.4, 1% NP-
40, 0.5% cholate, 0.1% Triton X-100, 1 mm EDTA, 1 mm
EGTA, 50 mm NaF, 20 mm b glycerophosphate, 15 mm
sodium pyrophosphate, 2 mm orthovanadate, 10 nm oka-
daic acid, protease inhibitors (Roche, Vilvoorde, Belgium),

0.1 m NaCl. Activated PKB was detected using phospho-
specific antibodies against T308 or S473. Activated MAP
kinase was detected using phospho-p44 ⁄ 42 MAP (Erk1 ⁄ 2)
kinase antibody (Cell Signalling, Leusden, the Netherlands).
Antibodies against total PKB and MAP kinase (Erk2) were
from Cell Signalling. For PKB assay, immunoprecipitation
was performed with anti Akt1 ⁄ PKB (Upstate) following the
protocol provided. PKB activities were determined in the
presence of 30 lm Crosstide substrate peptide and
[c-
32
P]ATP. Phosphorylated substrate was measured on P81
phosphocellulose papers.
Measurement of PI 3-kinase activity
Cell lysates of serum or IGF-1 stimulated MEF cells
(2 · 10
6
cells per condition) were immunoprecipitated with
2 lL antiphosphotyrosine antibodies overnight at 4 °C. The
immunoprecipitates were washed twice in 50 mm Tris ⁄ HCl
pH 7.4, 200 mm NaCl, 0.1% Brij, protease inhibitors
(Roche) and once in the kinase reaction buffer (see below).
The pellet ( 30 lL) was incubated at 37 °C for 30 min in
the presence of 80 lL2· kinase buffer containing 100 mm
Tris ⁄ HCl pH 7.4, 200 mm NaCl, 10 mm MgCl
2
,1mm
EDTA, 200 lm ATP together with [c-
32
P]ATP (10 lCi per

condition) and 50 lL sonicated vesicles of PI and phos-
phatidylserine. When the effect of the PI 3-kinase inhibitor
LY-294002 was tested, it was added to the kinase buffer at
25 lm. The lipids were extracted following a Bligh and
Dyer modified procedure and resuspended in 30 lLof
CHCl
3
⁄ CH
3
OH (1 : 1). Separation of the reaction product
was performed by TLC on a silica plate in acetone ⁄
CH
3
OH ⁄ acetic acid ⁄ H
2
O ⁄ CHCl
3
(30 : 26 : 24 : 14 : 80,
v ⁄ v ⁄ v ⁄ v ⁄ v). The corresponding spots were analysed by
autoradiography.
Acknowledgements
We would like to thank Mrs Colette Moreau, Dr Len
Stephens, Louis Hue, Mark Rider, Franc¸ ois Willer-
main, Fabrice Vandeput, Vale
´
rie Dewaste and Natha-
lie Paternotte for many helpful discussions. This work
was supported by grants of the Fonds de la Recherche
Scientifique Me
´

dicale, Action de Recherche Concerte
´
e
of the Communaute
´
Franc¸ aise de Belgique and
INSERM-Communaute
´
Franc¸ aise de Belgique exchange
contract. This work was executed in the framework of
research network IAPV-O5 (Belgium Science Policy).
Daniel Blero and Xavier Pesesse are Charge
´
de
Recherche FNRS.
References
1 Damen JE, Liu L, Rosten P, Humphries RK, Jefferson
AB, Majerus PW & Krystal G (1996) The 145-kDa pro-
tein induced to associate with Shc by multiple cytokines
is an inositol tetraphosphate and phosphatidylinositol
3,4,5- triphosphate 5-phosphatase. Proc Natl Acad Sci
USA 93, 1689–1693.
2 Ishihara H, Sasaoka T, Hori H, Wada T, Hirai H, Har-
uta T, Langlois WJ & Kobayashi M (1999) Molecular
Phosphatidylinositol 3,4,5-trisphosphate levels in mouse embryonic fibroblasts D. Blero et al.
2520 FEBS Journal 272 (2005) 2512–2522 ª 2005 FEBS
cloning of rat SH2-containing inositol phosphatase 2
(SHIP2) and its role in the regulation of insulin signal-
ing. Biochem Biophys Res Commun 260, 265–272.
3 Lioubin MN, Algate PA, Tsai S, Carlberg K, Aebersold

A & Rohrschneider LR (1996) p150Ship, a signal trans-
duction molecule with inositol polyphosphate-5-phos-
phatase activity. Genes Dev 10, 1084–1095.
4 Pesesse X, Deleu S, De Smedt F, Drayer L & Erneux C
(1997) Identification of a second SH2-domain-contain-
ing protein closely related to the phosphatidylinositol
polyphosphate 5-phosphatase SHIP. Biochem Biophys
Res Commun 239, 697–700.
5 Muraille E, Pesesse X, Kuntz C & Erneux C (1999) Dis-
tribution of the src-homology-2-domain-containing ino-
sitol 5- phosphatase SHIP-2 in both non-haemopoietic
and haemopoietic cells and possible involvement of
SHIP-2 in negative signalling of B-cells. Biochem J 342,
697–705.
6 Habib T, Hejna JA, Moses RE & Decker SJ (1998)
Growth factors and insulin stimulate tyrosine phosphory-
lation of the 51C ⁄ SHIP2 protein. J Biol Chem 273,
18605–18609.
7 Wisniewski D, Strife A, Swendeman S, Erdjument-Bro-
mage H, Geromanos S, Kavanaugh WM, Tempst P &
Clarkson B (1999) A novel SH2-containing phosphati-
dylinositol 3,4,5-trisphosphate 5-phosphatase (SHIP2) is
constitutively tyrosine phosphorylated and associated
with src homologous and collagen gene (SHC) in
chronic myelogenous leukemia progenitor cells. Blood
93, 2707–2720.
8 Wang YJ, Keogh RJ, Hunter MG, Mitchell CA, Frey
RS, Javaid K, Malik AB, Schurmans S, Tridandapani S
& Marsh CB (2004) SHIP2 is recruited to the cell mem-
brane upon macrophage colony-stimulating factor

(M-CSF) stimulation and regulates M-CSF-induced
signaling. J Immunol 173, 6820–6830.
9 Gagnon A, ArtemenkoY, Crapper T & Sorisky A (2003)
Regulation of endogenous SH2 domain-containing
inositol 5-phosphatase (SHIP2) in 3T3-L1 and human
preadipocytes. J Cellular Physiol 197, 243–250.
10 Sasaoka T, Hori H, Wada T, Ishiki M, Haruta T, Ishi-
hara H & Kobayashi M (2001) SH2-containing inositol
phosphatase 2 negatively regulates insulin- induced gly-
cogen synthesis in L6 myotubes. Diabetologia 44, 1258–
1267.
11 Wada T, Sasaoka T, Funaki M, Hori H, Murakami S,
Ishiki M, Haruta T, Asano T, Ogawa W, Ishihara H &
Kobayashi M (2001) Overexpression of SH2-containing
inositol phosphatase 2 results in negative regulation of
insulin-induced metabolic actions in 3T3-L1 adipocytes
via its 5¢-phosphatase catalytic activity. Mol Cell Biol
21, 1633–1646.
12 Sasaoka T, Wada T, Fukui K, Murakami S, Ishihara
H, Suzuki R, Tobe K, Kadowaki T & Kobayashi M
(2004) SH2-containing inositol phosphatase 2 predomi-
nantly regulates Akt2, and not Akt1, phosphorylation
at the plasma membrane in response to insulin in
3T3-L1 adipocytes. J Biol Chem 279, 14835–14843.
13 Hori H, Sasaoka T, Ishihara H, Wada T, Murakami S,
Ishiki M & Kobayashi M (2002) Association of SH2-
containing inositol phosphatase 2 with the insulin resis-
tance of diabetic db ⁄ db mice. Diabetes 51, 2387–2394.
14 Sasaoka T, Kikuchi K, Wada T, Sato A, Hori H,
Murakami S, Fukui K, Ishihara H, Aota R, Kimura I

& Kobayashi M (2003) Dual role of Src homology
domain 2-containing inositol phosphatase 2 in the regu-
lation of platelet-derived growth factor and insulin-like
growth factor I signaling in rat vascular smooth muscle
cells. Endocrinology 144, 4204–4214.
15 Taylor V, Wong M, Brandts C, Reilly L, Dean NM,
Cowsert LM, Moodie S & Stokoe D (2000) 5¢ phos-
pholipid phosphatase SHIP-2 causes protein kinase B
inactivation and cell cycle arrest in glioblastoma cells.
Mol Cell Biol 20, 6860–6871.
16 Prasad N, Topping RS & Decker SJ (2001) SH2-con-
taining inositol 5¢-phosphatase SHIP2 associates with
the p130 (Cas) adapter protein and regulates cellular
adhesion and spreading. Mol Cell Biol 21, 1416–1428.
17 Prasad N & Decker SJ (2005) SH2-containing 5¢ inositol
phosphatase SHIP2 regulates cytoskeleton organization
and ligand-dependent downregulation of the epidermal
growth factor receptor. J Biol Chem 280, 13129–13136.
18 Clement S, Krause U, Desmedt F, Tanti JF, Behrends
J, Pesesse X, Sasaki T, Penninger J, Doherty M, Mala-
isse W, Dumont JE, Marchand-Brustel Y, Erneux C,
Hue L & Schurmans S (2001) The lipid phosphatase
SHIP2 controls insulin sensitivity. Nature 409, 92–97.
19 Sleeman MW, Wortley KEV, Lai KM, Gowen LC,
Kintner J, Kline WO, Garcia K, Stitt TN, Yancopoulos
GD, Wiegand SJ & Glass DJ (2005) Absence of the
lipid phosphatase SHIP2 confers resistance to diary
obesity. Nat Med 11, 199–205.
20 Giuriato S, Blero D, Robaye B, Bruyns C, Payrastre B
& Erneux C (2002) SHIP2 overexpression strongly

reduces the proliferation rate of K562 erythroleukemia
cell line. Biochem Biophys Res Commun 296, 106–110.
21 Pesesse X, Moreau C, Drayer AL, Woscholski R, Par-
ker P & Erneux C (1998) The SH2 domain containing
inositol 5-phosphatase SHIP2 displays phosphatidylino-
sitol 3,4,5-trisphosphate and inositol 1,3,4,5- tetraki-
sphosphate 5-phosphatase activity. FEBS Lett 437,
301–303.
22 Blero D, De Smedt F, Pesesse X, Paternotte N, Moreau
C, Payrastre B & Erneux C (2001) The SH2 domain
containing inositol 5-phosphatase SHIP2 controls phos-
phatidylinositol 3,4,5-trisphosphate levels in CHO-IR
cells stimulated by insulin. Biochem Biophys Res Com-
mun 282, 839–843.
23 Pesesse X, Dewaste V, De Smedt F, Laffargue M, Giur-
iato S, Moreau C, Payrastre B & Erneux C (2001) The
D. Blero et al. Phosphatidylinositol 3,4,5-trisphosphate levels in mouse embryonic fibroblasts
FEBS Journal 272 (2005) 2512–2522 ª 2005 FEBS 2521
Src homology 2 domain containing inositol 5-phospha-
tase SHIP2 is recruited to the epidermal growth factor
(EGF) receptor and dephosphorylates phosphatidylino-
sitol 3,4,5-trisphosphate in EGF- stimulated COS-7
cells. J Biol Chem 276, 28348–28355.
24 McConnachie G, Pass I, Walker SM & Downes CP
(2003) Interfacial kinetic analysis of the tumour suppres-
sor phosphatase, PTEN: evidence for activation by
anionic phospholipids. Biochem J 371, 947–955.
25 Leslie NR & Downes CP (2002) PTEN: The down side
of PI 3-kinase signalling. Cell Signal 14, 285–295.
26 Stambolic V, Suzuki A, de la Pompa JL, Brothers GM,

Mirtsos C, Sasaki T, Ruland J, Penninger JM, Siderov-
ski DP & Mak TW (1998) Negative regulation of
PKB ⁄ Akt-dependent cell survival by the tumor suppres-
sor PTEN. Cell 95, 29–39.
27 Backers K, Blero D, Paternotte N, Zhang J & Erneux C
(2003) The termination of PI3K signalling by SHIP1
and SHIP2 inositol 5-phosphatases. Adv Enzyme Regul
43, 15–28.
28 Scharenberg AM, El HillalO, Fruman DA, Beitz LO,
Li ZM, Lin SQ, Gout I, Cantley LC, Rawlings DJ &
Kinet JP (1998) Phosphatidylinositol-3,4,5-trisphosphate
(PtdIns-3,4,5-P 3) Tec kinase-dependent calcium signal-
ing pathway: a target for SHIP-mediated inhibitory
signals. EMBO J 17, 1961–1972.
29 Clement S, Krause U, Desmedt F, Tanti JF, Behrends
J, Pesesse X, Sasaki T, Penninger J, Doherty M, Mala-
isse W, Dumont JE, Marchand-Brustel Y, Erneux C,
Hue L & Schurmans S (2005) Corrigendum: the lipid
phosphatase SHIP2 controls insulin sensitivity. Nature
431, 878.
30 Dyson JM, O’Malley CJ, Becanovic J, Munday AD,
Berndt MC, Coghill ID, Nandurkar HH, Ooms LM &
Mitchell CA (2001) The SH2-containing inositol poly-
phosphate 5-phosphatase, SHIP-2, binds filamin and
regulates submembraneous actin. J Cell Biol 155, 1065–
1079.
31 Vandenbroere I, Paternotte N, Dumont JE, Erneux C &
Pirson I (2003) The c-Cbl-associated protein and c-Cbl
are two new partners of the SH2-containing inositol
polyphosphate 5-phosphatase SHIP2. Biochem Biophys

Res Commun 300, 494–500.
32 Ishihara H, Sasaoka T, Ishiki M, Wada T, Hori H, Ka-
gawa S & Kobayashi M (2002) Membrane localization
of Src homology 2-containing inositol 5¢-phosphatase 2
via Shc association is required for the negative regula-
tion of insulin signaling in Rat1 fibroblasts overexpress-
ing insulin receptors. Mol Endocrinol 16, 2371–2381.
33 Salomon AR, Ficarro SB, Brill LM, Brinker A, Phung
QT, Ericson C, Sauer K, Brock A, Horn DM, Schultz
PG & Peters EC (2003) Profiling of tyrosine phosphory-
lation pathways in human cells using mass spectrome-
try. Proc Natl Acad Sci USA 100, 443–448.
34 Liu QR, Sasaki T, Kozieradzki I, Wakeham A, Itie A,
Dumont DJ & Penninger JM (1999) SHIP is a negative
regulator of growth factor receptor-mediated PKB ⁄ Akt
activation and myeloid cell survival. Genes Dev 13,
786–791.
35 Kwon J, Lee SR, Yang KS & Ahn Y (2004) Kim,
Stadtman ER & Rhee SG Reversible oxidation and
inactivation of the tumor suppressor PTEN in cells sti-
mulated with peptide growth factors. Proc Natl Acad
Sci USA 101, 16419–16424.
36 Seo JH, Ahn Y, Lee SR, Yeo CY & Hur KC (2005)
The major target of the endogenously generated reactive
oxygen species in response to insulin stimulation is
phosphatase and tensin homolog and not PI-3 kinase in
the PI-3 kinase pathway. Mol Biol Cell 16, 348–357.
37 Gu XN & Spitzer NC (1995) Distinct Aspects of Neuro-
nal Differentiation Encoded by Frequency of Sponta-
neous Ca

2+
Transients. Nature 375, 784–787.
38 Serunian LA, Auger K & Cantley LC (1991) Identifica-
tion and quantification of polyphosphoinositides pro-
duced in response to platelet-derived growth-factor
stimulation. Methods Enzymol 198, 78–87.
39 Giuriato S, Pesesse X, Bodin S, Sasaki T, Viala C, Mar-
ion E, Penninger J, Schurmans S, Erneux C & Payrastre
B (2003) SH2-containing inositol 5-phosphatases 1 and
2 in blood platelets. their interactions and roles in the
control of phosphatidylinositol 3,4,5-trisphosphate
levels. Biochem J 376, 199–207.
Phosphatidylinositol 3,4,5-trisphosphate levels in mouse embryonic fibroblasts D. Blero et al.
2522 FEBS Journal 272 (2005) 2512–2522 ª 2005 FEBS