Oncogene (2004) 23, 3659–3669
& 2004 Nature Publishing Group All rights reserved 0950-9232/04 $25.00
www.nature.com/onc
A negative role of SHP-2 tyrosine phosphatase in growth factor-dependent
hematopoietic cell survival
Jing Chen1, Wen-Mei Yu1, Kevin D Bunting1,2 and Cheng-Kui Qu*,1,2
1
Department of Hematopoiesis, Jerome H Holland Laboratory for the Biomedical Sciences, American Red Cross, 15601 Crabbs
Branch Way, Rockville, MD 20855, USA; 2Department of Anatomy and Cell Biology, George Washington University Medical
Center, 2300 I Street, NW, Washington, DC 20037, USA
SHP-2 tyrosine phosphatase is highly expressed in
hematopoietic cells; however, the function of SHP-2 in
hematopoietic cell processes is not fully understood.
Recent identification of SHP-2 mutations in childhood
leukemia further emphasizes the importance of SHP-2
regulation in hematopoietic cells. We previously reported
that SHP-2 played a positive role in IL-3-induced
activation of Jak2 kinase in a catalytic-dependent
manner. Interestingly, enforced expression of wild-type
(WT) SHP-2 in Ba/F3 cells enhanced growth factor
deprivation-induced apoptosis. Biochemical analyses revealed that although IL-3 activation of Jak2 kinase was
increased, tyrosyl phosphorylation of its downstream
substrate STAT5 was disproportionately decreased by
the overexpression of SHP-2. Following IL-3 deprivation,
the tyrosyl phosphorylation of STAT5 that is required for
its antiapoptotic activity was rapidly diminished in SHP-2
overexpressing cells. As a result, reduction of the putative
downstream targets of STAT5–Bcl-XL and pim-1 was
accelerated by overexpression of SHP-2. Further investigation showed that SHP-2 associated with STAT5, and
that it was indeed able to dephosphorylate STAT5.
Finally, overexpression of SHP-2 in primary bone marrow
hematopoietic progenitor cells compromised their differentiative and proliferative potential, and enhanced growth
factor withdrawal-induced cell death. And, the effect of
SHP-2 overexpression on growth factor-dependent survival was diminished in STAT5-deficient hematopoietic
cells. Taken together, these results suggest that SHP-2
tyrosine phosphatase negatively regulates hematopoietic
cell survival by dephosphorylation of STAT5.
Oncogene (2004) 23, 3659–3669. doi:10.1038/sj.onc.1207471
Published online 19 April 2004
Keywords: SHP-2; tyrosine phosphatase; hematopoietic
cell survival; STAT5; IL-3
Introduction
Growth factor-induced intracellular signaling pathways
play an important role in maintaining homeostasis of
*Correspondence: Dr C-K Qu; E-mail: [email protected]
Received 22 July 2003; revised 25 October 2003; accepted 20 December
2003; Published online 19 April 2004
the hematopoietic system (Dragovich et al., 1998;
Wickremasinghe and Hoffbrand, 1999). Committed
hematopoietic progenitor cells require growth/survival
factors such as interleukin (IL)-3 for both growth and
survival; depletion of these trophic factors leads to
programmed cell death (apoptosis). Since extracellular
factors function through intracellular signal transduction, growth/survival factor-dependent signaling cascades such as the PI3K/Akt, Erk, and signal transducer
and activator of transcription (STAT) pathways play
important roles in determining hematopoietic cell
survival or death. Accumulating evidence has demonstrated that dys-regulation of intracellular signal transduction induced by extracellular survival factors results
in abnormal hematopoietic homeostasis, which may
eventually lead to a variety of blood disorders including
leukemia.
Recent studies have shown that growth/survival
factor induced signaling events are functionally linked
to the Bcl-2 family proteins that directly govern cell
survival or death through apoptosis. For example,
growth factor activated Akt kinase phosphorylates
Bad (Datta et al., 1997; del Peso et al., 1997), a
proapoptotic member of the Bcl-2 family. The phosphorylated form of Bad associates with 14-3-3 adaptor
protein (Zha et al., 1996) which in turn sequesters Bad
away from Bcl-XL (an antiapoptotic protein), thus
freeing Bcl-XL to exert its antiapoptotic effect. Another
connection between growth factor signaling cascades
and the apoptotic machinery is that growth factors
promote hematopoietic cell survival by upregulation of
Bcl-XL and pim-1 through activation of STAT5
(Dumon et al., 1999; Nosaka et al., 1999; Silva et al.,
1999; Socolovsky et al., 1999).
STAT5 is one of the six mammalian members of the
STAT family that have been characterized. Both
STAT5a and STAT5b isoforms are tyrosine phosphorylated and activated by a wide range of cytokines and
hormones such as IL-3, GM-CSF, IL-2, erythropoietin,
thrombopoietin, prolactin, and growth hormone (Rane
and Reddy, 2002). Activated STAT5 forms homo- or
heterodimers, enters the nucleus, and binds to the
specific DNA sequences in the promoter regions of
various responsive genes, resulting in gene activation or
repression. STAT5 has been shown to promote the IL-3
and IL-2 dependent survival of hematopoietic and
SHP-2 phosphatase and hematopoietic cell survival
J Chen et al
3660
lymphoid cells, respectively (Zamorano et al., 1998;
Dumon et al., 1999). Recent studies have revealed that
STAT5 promotes hematopoietic cell survival by regulating the expression of a number of antiapoptotic proteins
such as Bcl-XL (Dumon et al., 1999; Socolovsky et al.,
1999; Shinjyo et al., 2001 ) and pim-1 (Nosaka et al.,
1999), and that tyrosyl phosphorylation of STAT5 is
critical for this biological activity. STAT5 is phosphorylated and activated by Jak family kinases, particularly
Jak2; however, it is not fully understood how this
transcription factor is inactivated. Previous reports have
suggested that STAT5 might be dephosphorylated by
tyrosine phosphatases PTP1B and SHP-2 in the
prolactin, IL-2, and erythropoietin signaling pathways
(Aoki and Matsuda, 2000; Yu et al., 2000; Chughtai
et al., 2002; Chen et al., 2003), but the biological
significance of their functional interactions remains
uncharacterized.
SHP-2, an SH2 domain-containing protein tyrosine
phosphatase, is ubiquitously expressed in a variety of
tissues and cell types, and has been implicated in
diverse signaling pathways, including those initiated
by growth factors, cytokines, insulin, and interferons (Tonks and Neel, 2001; Qu, 2002). A large body
of evidence has demonstrated that, in most circumstances, SHP-2 plays a positive role in transducing
the signal elicited from receptor tyrosine kinases such
as EGF and FGF receptors (Tang et al., 1995;
Bennett et al., 1996; Hadari et al., 1998; Shi et al.,
1998). This phosphatase is also highly expressed in
hematopoietic cells and has been indicated to be
involved in signal transduction of a number of
hematopoietic growth factors, including IL-3 (Welham
et al., 1994; Bone et al., 1997; Gu et al., 1998; Yu et al.,
2003). SHP-2 plays an essential role in the onset of
hematopoietic development. We previously mutated the
SHP-2 gene in mouse embryonic stem (ES) cells.
Analyses of in vitro (Qu et al., 1997) and in vivo (Qu
et al., 1998) hematopoietic cell differentiation from
SHP-2/ ES cells demonstrated that SHP-2 was
required for primitive hematopoiesis. These results
stand in contrast to the negative role of the closely
related SHP-1 phosphatase in hematopoietic cell signaling. Indeed, our subsequent genetic analyses of hematopoietic development in the SHP-2/SHP-1 double
mutant embryos confirmed opposing roles for SHP-2
and SHP-1 in regulating hematopoietic-lineage determination (Qu et al., 2001).
Most recently, several gain-of-function mutations of
SHP-2 have been identified in human juvenile myelomonocytic leukemia, myelodysplastic syndromes, and
acute myeloid leukemia (Tartaglia et al., 2003). These
findings highlight the importance of studies on the
function of SHP-2 in hematopoietic cells. Unfortunately, the detailed biochemical activities of SHP-2
phosphatase in the regulation of hematopoietic
cell processes are not fully appreciated. We previously reported that SHP-2 played a positive role in
IL-3-induced activation of Jak2 and Erk kinases in a
catalytic-dependent manner (Yu et al., 2003). In the
course of that work, we noticed that overexpression of
Oncogene
WT SHP-2 in hematopoietic cells compromised their
hematopoietic activities and enhanced their responses to
growth factor deprivation-induced apoptosis. We have
therefore further characterized the biological and
biochemical activities of SHP-2 in the regulation of
hematopoietic cell survival.
Results
Overexpression of SHP-2 increases the susceptibility of
hematopoietic cells to growth factor withdrawal-induced
death
In the course of our previous studies focusing on the role
of tyrosine phosphatase SHP-2 in IL-3-induced signaling and cellular responses (Yu et al., 2003), we observed
that Ba/F3 cells overexpressing catalytically inactive
SHP-2 (SHP-2 C/S) were more susceptible to cell death.
Following IL-3 withdrawal, cell survival rates of SHP-2
C/S clones were decreased compared to that of GFP
control cells (Figure 1a). Consistent with this observation, activation of caspase 3, an executioner protease in
cellular apoptosis was increased in these cells
(Figure 1b). Since IL-3 is required for Ba/F3 cell
survival and growth, the increased apoptosis appears
to be attributed to the decrease in IL-3 signaling in these
cells. We previously showed that IL-3-induced Jak2
kinase activation was significantly decreased in SHP-2
C/S cells (Yu et al., 2003). In agreement with that
observation, activation (tyrosyl phosphorylation) of its
downstream substrate STAT5 was attenuated (Figure 2).
Intriguingly, overexpression of WT SHP-2 in Ba/F3 cells
also resulted in accelerated cell death following IL-3
starvation, and this effect was even greater than that
seen in the C/S expressing cells (Figure 1a and c). The
percentage of hypodiploid cells with sub-G1 DNA
content, known to be apoptotic, was dramatically
increased in WT SHP-2 overexpressing cells following
starvation (Figure 1c). These results indicate that in
addition to the regulatory role in hematopoietic cell
proliferation and differentiation, SHP-2 phosphatase might also be involved in the regulation of
hematopoietic cell survival/death.
Overexpression of WT SHP-2 phosphatase decreases
tyrosyl phosphorylation of STAT5 while activation of
Jak2 kinase is increased
To elucidate the underlying mechanisms by which SHP2 modulates hematopoietic cell survival, we focused on
IL-3 addition and deprivation-induced biochemical
processes in WT SHP-2 overexpressing cells. As shown
in Figure 3a, in response to IL-3 stimulation, activation
of Jak2 kinase, defined by its tyrosyl phosphorylation,
was significantly increased by overexpression of WT
SHP-2. This result is consistent with our previous
finding that SHP-2 plays a positive role in IL-3-induced
Jak2 activation in a catalytic-dependent manner (Yu
et al., 2003). Unexpectedly, activation of STAT5, a
downstream target of Jak2 kinase, was found to be
SHP-2 phosphatase and hematopoietic cell survival
J Chen et al
3661
Figure 1 Overexpression of WT and catalytically inactive SHP-2 increases IL-3 deprivation-induced apoptosis in Ba/F3 cells. Ba/F3
cells were transduced with the empty vector (GFP) or WT SHP-2 (WT) retroviruses. Transduced cells were sorted for GFP expression.
WT SHP-2 transduced stable cell lines were selected by limiting dilution and anti-SHP-2 immunoblotting. Control (GFP positive
whole-cell population sorted from the empty vector transduced cells), two independent WT SHP-2, and twao independent SHP-2 C/S
clones previously generated (Yu et al., 2003) were used for the experiments. (a) Exponentially growing cells were starved of IL-3 for the
indicated periods of time. Cell survival rates were determined by the MTS assay (Yu et al., 2003). Expression of SHP-2 or SHP-2 C/S in
individual clones is shown on the lower panel. (b) Caspase-3 activities of the cells were determined using the Caspase-3 Assay Kit
according to the manufacturer’s instructions 48 h following IL-3 starvation. (c) Exponentially growing cells were starved for 48 h. Cells
were then harvested and fixed with 70% ethanol. Fixed cells were treated with RNase A (20 mg/ml) at 371C for 30 min, washed with
PBS, and then stained with propidium iodide (50 mg/ml in PBS). Cellular DNA content was monitored using BD-LSR flowcytometry
(BD Biosciences). The percentage of cells with sub-G1 DNA content (M1) was determined using the CELLQuestt software (BD
Biosciences) (Yuan et al., 2003). Two to four independent experiments were performed and similar results were obtained in each.
Results shown in (a) and (b) are the mean7s.d. of triplicates from one experiment
decreased rather than proportionally increased
(Figure 3a). Multiple experiments were conducted
with several independent clones, producing similar
results. In contrast, IL-3 activation of Akt and Erk
kinases was slightly enhanced in WT SHP-2 overexpressing cells (Figure 3b). Since Akt and Erk kinases
promote cell survival (Datta et al., 1997; del Peso et al.,
1997; Songyang et al., 1997), this data indicates that
the increased apoptosis of the cells overexpressing
WT SHP-2 is not related to the PI3 kinase or Erk
pathways.
SHP-2 phosphatase associates with and dephosphorylates
STAT5
The observation that tyrosyl phosphorylation of STAT5
is decreased while Jak2 activation is increased in WT
SHP-2 overexpressing cells raises the possibility that
SHP-2 might directly inactivate STAT5. To test this
possibility, we examined tyrosyl phosphorylation of
STAT5 following IL-3 starvation. As illustrated in
Figure 4a, after IL-3 withdrawal from the cell culture
medium, the decay of STAT5 phosphorylation in Ba/F3
cells overexpressing WT SHP-2 was significantly
Oncogene
SHP-2 phosphatase and hematopoietic cell survival
J Chen et al
3662
GFP
C/S6
0 5 10 15 30
0 5 10 15 30
C/S7
0
5 10 15 30
a
IL-3 (min)
GFP
0
WT10
5 15 30
0
5 15 30
WT12
0
5
15
30
IL-3 (min)
Phospho-STAT5b
Phospho-Jak2
IP: Anti-STAT5b IB: Anti-PY
IP: Anti-Jak2 IB: Anti-PY
STAT5b
Jak2
IP: Anti-STAT5b IB: Anti-STAT5b
Figure 2 IL-3-induced tyrosyl phosphorylation of STAT5 is
decreased by interference of the catalytic activity of endogenous
SHP-2. SHP-2 C/S overexpressing Ba/F3 cells and the GFP control
cells were starved for 5 h and then stimulated with IL-3 (2 ng/ml)
for the indicated periods of time. Whole-cell lysates were prepared
and immunoprecipitated (IP) with anti-STAT5b Ab followed by
anti-PY immunoblotting (IB). The blot was stripped and reprobed
with anti-STAT5b Ab to examine protein loading. Representative
results from two independent experiments are shown
IP: Anti-Jak2 IB: Anti-Jak2
GFP
WT10
0 5 15 30
0 5 15 30
WT12
0 5 15 30
IL-3 (min)
Phospho-STAT5b
IP: Anti-STAT5b IB: Anti-PY
accelerated relative to GFP control cells. Tyrosyl
phosphorylation of STAT5 was barely detected 1 h after
starvation vs 5 h in control cells. This phenotypic change
appears to be specific to STAT5 as Jak2 kinase activity
diminished comparably in both cell types. Since STAT5
regulates cellular survival by upregulation of antiapoptotic protein Bcl-XL (Dumon et al., 1999; Nosaka et al.,
1999; Silva et al., 1999; Socolovsky et al., 1999) and pim1 (Nosaka et al., 1999), the expression of Bcl-XL and
pim-1 was examined next. Consistent with the STAT5
phosphorylation response, following IL-3 deprivation,
reduction of Bcl-XL and pim-1 expression in WT SHP-2
overexpressing cells was accelerated (Figure 4b).
Together these results imply that SHP-2 modulates
IL-3-induced hematopoietic cell survival by dephosphorylating STAT5.
To better define the molecular mechanism by which
SHP-2 phosphatase regulates STAT5 activity, we
examined their physical and functional interactions.
Consistent with previous reports (Yu et al., 2000;
Chughtai et al., 2002; Chen et al., 2003), anti-STAT5
immunoprecipitation followed by anti-SHP-2 immunoblotting showed that SHP-2 constitutively associated
with STAT5 independent of IL-3 stimulation in SHP-2
overexpressing cells (data not shown). Further in vitro
GST pull down assay demonstrated that only the
phosphatase fragment could pull down STAT5
(data not shown), suggesting that the SHP-2/STAT5
interaction is direct and that this association is mediated
by the C-terminal part of SHP-2 but not the SH2
domains.
Since SHP-2 associates with STAT5, their functional
interplay was next determined. Anti-STAT5 immunoprecipitates prepared from IL-3 stimulated parental Ba/
F3 cells were incubated in tyrosine phosphatase assay
buffer for various periods of time and the phosphorylation status of STAT5 in the immunoprecipitates was
examined. As expected, tyrosyl phosphorylation of
STAT5 decreased along with the incubation time, and
the dephosphorylation could be blocked by a tyrosine
phosphatase inhibitor, orthovanadate (Figure 4c). Since
SHP-2 physically interacts with STAT5, this result
indicates that the SHP-2 existing in the anti-STAT5
Oncogene
STAT5b
IP: Anti-STAT5b IB: Anti-STAT5b
b
GFP
WT10
WT12
0 5 15 30
0 5 15 30
0 5 15 30
IL-3 (min)
Phospho-Akt
IB: Anti-phospho-Akt
Akt
IB: Anti-Akt
GFP
WT10
0 5 15 30 0 5 15 30
IL-3 (min)
Phospho-p44
Phospho-p42
IB: Anti-phospho-Erk
p44
p42
IB: Anti-Erk
Figure 3 Overexpression of WT SHP-2 increases IL-3-induced
activation of Jak2 kinase but decreases tyrosyl phosphorylation of
STAT5. GFP control and two independent WT SHP-2 overexpressing Ba/F3 clones were starved and stimulated with IL-3 as
described above. (a) Whole-cell lysates were immunoprecipitated
with anti-Jak2 and anti-STAT5b Abs followed by anti-PY
immunoblottings. (b) Cell lysates were also examined for Erk and
Akt activation using specific anti-phospho-Erk and anti-phosphoAkt immunoblottings. Blots were stripped and reprobed with antiJak2, anti-STAT5, anti-Erk, or anti-Akt Abs to examine protein
loading. Representative results from two independent experiments
are shown
immunocomplex dephosphorylates STAT5. To further
examine the phosphatase activity of SHP-2 on STAT5,
we incubated anti-STAT5 immunoprecipitates with
purified GST-SHP-2 fusion protein and analyzed tyrosyl
phosphorylation of STAT5. GST-SHP-2 but not
SHP-2 phosphatase and hematopoietic cell survival
J Chen et al
3663
5
0
1
2
5
0
1
2
5
Starvation (hr)
Starved
WT
10
WT
12
2
Control
GFP
1
WT12
WT
12
0
WT10
GFP
GFP
b
WT
10
a
Phospho-STAT5a
Bcl-XL
IB: Anti-Bcl-XL
IP: Anti-STAT5a IB: Anti-pY
pim-1
STAT5a
IB: Anti-pim-1
Tubulin
IP: Anti-STAT5a IB: Anti-STAT5a
IB: Anti-Tubulin
GFP
0
1
2
WT10
5
0
1
2
WT12
5
0
1
2
5
Starvation (hr)
Phospho-Jak2
IP: Anti-Jak2 IB: Anti-pY
Jak2
IP: Anti-Jak2 IB: Anti-Jak2
c
d
0
-
-
-
+
30 60 120 120
Na3VO4
Incubation (min)
Buffer
SHP-2 SH2
C
0.1 1
0
SHP2
10 0.1 1
10
(µg)
Phospho-STAT5b
Phospho-STAT5b
IP: Anti-STAT5b IB: Anti-PY
IP: anti-STAT5b IB: anti-PY
STAT5b
STAT5b
IP: Anti-STAT5b IB: Anti-STAT5b
IP: anti-STAT5b IB: anti-STAT5b
Figure 4 Decay of STAT5 phosphorylation following IL-3 deprivation is dramatically enhanced by overexpression of SHP-2.
Exponentially growing Ba/F3 clones, GFP control and two independent WT SHP-2 overexpressing, were starved for 5 h and then
stimulated with IL-3 for 30 min. The cells were washed and seeded into IL-3-free medium for the indicated periods of time before being
harvested for the following analyses. (a) Tyrosyl phosphorylation of STAT5a and Jak2 kinase was examined as described in Figure 3.
(b) Cell lysates prepared from the cells cultured in IL-3-free medium for 24 h were examined for Bcl-XL and pim-1 expression. (c) AntiSTAT5b immunoprecipitates prepared from IL-3 stimulated GFP control cells as described above were incubated in phosphatase assay
buffer with or without addition of sodium orthovanadate (1 mM) for the indicated periods of time. Tyrosine phosphorylation of
STAT5 was then analysed by anti-PY immunoblotting. (d) Anti-STAT5b immunoprecipitates were incubated with GST-SHP-2 (full
length) or GST-SHP-2 SH2 domains (amino acids 1–243 including two SH2 domains) GST fusion proteins at various concentrations in
phosphatase assay buffer for 30 min. Tyrosyl phosphorylation of STAT5 was then examined. ‘C’ indicates the nonincubated control.
Blots were stripped and reprobed with anti-Jak2, anti-STAT5a, anti-tubulin, and anti-STAT5b Abs, respectively, to examine protein
loading. Representative results from two independent experiments are shown
GST-SHP-2 SH2 domains dephosphorylated STAT5
and this function was clearly dose-dependent. In all,
10 mg GST-SHP-2 completely dephosphorylated STAT5
(Figure 4d). Taken together, these results further
confirm that SHP-2 interacts with STAT5 both physically and functionally.
Stat5 plays an important role in hematopoietic cell
survival
The above results showed that STAT5 activity was
reduced in SHP-2 overexpressing cells, to determine
whether the decreased STAT5 activity was connected to
Oncogene
SHP-2 phosphatase and hematopoietic cell survival
J Chen et al
3664
the enhanced hematopoietic cell apoptosis, we next
examined the role of STAT5 in hematopoietic cell
survival. Bone marrow-derived macrophages (BMDM)
were generated from WT and STAT5-deficient mice as
we previously reported (Yu et al., 2002). They were
maintained in IL-3-containing medium for 2 days and
cell survival rates following IL-3 deprivation were then
assessed. As shown in Figure 5a, STAT5/ cells showed
increased cell death compared to WT control cells, and
expression level of Bcl-XL was barely detectable in
STAT5/ cells following 24 h of starvation, consistent
with the essential role for STAT5 in maintaining
erythroid progenitor and mast cell survival and the
observation that Bcl-XL was not detected in growth
factor-starved STAT5/ mast cells (Socolovsky et al.,
1999; Shelburne et al., 2003). To further confirm the role
of STAT5 in hematopoietic cell survival, we transduced
Ba/F3 cells with a dominant-negative STAT5 mutant
(STAT5aD757) (the C-terminal transactivation domain
is deleted) (Wang et al., 2000; Shelburne et al., 2003).
Transduced cells were sorted and assayed for survival
after IL-3 withdrawal. In agreement with the results
obtained from STAT5-deficient BMDMs, interference
a
of endogenous STAT5 function by a dominant-negative
mutant STAT5 also enhanced growth factor withdrawal-induced Ba/F3 cell death (Figure 5b). Reduction
of pim-1 expression induced by starvation was enhanced
in the dominant-negative STAT5 overexpressing cells.
These results together confirm an important role for the
STAT5/Bcl-XL, pim-1 pathway in hematopoietic cell
survival. In addition, we examined colony-forming
potential of STAT5-deficient bone marrow progenitor
cells in IL-3-containing methylcellulose and liquid
medium. The results also support that STAT5 activity
is required for optimal response to IL-3, an effect that
was most evident at low doses of the cytokine (data not
shown).
Enforced expression of shp-2 compromises hematopoietic
potential of primary bone marrow progenitor cells
Hematopoietic stem/progenitor cell fate is finely coordinated by cell survival signals and proliferative and
differentiative signals. Since SHP-2 phosphatase appears
to modulate hematopoietic cell survival by dephosphorylating STAT5, we next examined the effect of
b
STAT5+/+
100
100
GFP
STAT5a∆757
Cell Survival (%)
Cell Survival (%)
STAT5-/75
50
25
0
75
50
25
0
0
24
0
48
24
48
Starvation time (hr)
Starvation time (hr)
STAT5+/+ STAT5-/-
GFP STAT5a∆757
C
S
C
S
C
S
C
S
STAT5 &
STAT5a∆757
Bcl-XL
IB: Anti-Bcl-XL
96
IB: Anti-STAT5
pim-1
STAT5
IB: Anti-pim-1
IB: Anti-STAT5
Tubulin
Tubulin
IB: Anti-Tubulin
IB: Anti-Tubulin
Figure 5 Hematopoietic cell sensitivity to growth factor withdrawal-induced apoptosis is increased by loss of STAT5 function. (a)
BMDM were generated from WT and STAT5-deficient mice as we previously reported (Yu et al., 2002). They were maintained in IL-3containing medium (20 ng/ml) for 2 days and cell survival rates were then assessed 24 and 48 h following IL-3 withdrawal. Expression
of Bcl-XL was examined 24 h after starvation. (b) Ba/F3 cells were transduced with dominant-negative STAT5aD757. Transduced cells
were sorted and subjected to the cell survival assay at the indicated time points after IL-3 deprivation. The expression level of pim-1
was examined 24 h after IL-3 starvation. ‘C’ and ‘S’ indicate control and starved cells, respectively. The blots were stripped and
reprobed with anti-STAT5 Ab from BD Transduction Laboratories (that detects both STAT5a and STAT5b) and anti-tubulin Ab.
Results shown are a representative of two independent experiments
Oncogene
SHP-2 phosphatase and hematopoietic cell survival
J Chen et al
3665
overexpression of SHP-2 on hematopoietic progenitor
cell function. Primary bone marrow cells were transduced with WT SHP-2 by retroviral-mediated gene
transfer. Transduced cells were sorted based on GFP
expression and then assayed for colony formation in
methylcellulose medium containing a full combination
of hematopoietic growth factors (see Materials and
methods) or IL-3 alone. As shown in Figure 6a,
overexpression of SHP-2 attenuates colony-forming
capacities of the progenitor cells. In agreement with
this result, the expansion rate of progenitor cells
overexpressing SHP-2 in IL-3-containing liquid culture
was also reduced (Figure 6b).
To test whether the above phenotypic changes of the
SHP-2 transduced cells resulted from enhanced apoptosis, we transduced WT bone marrow cells with SHP-2
and the vector. Cells (unsorted) were maintained in
growth factor containing medium for 48 h. Percentages
of gene transduced cells (GFP positive) in these mixed
cell populations were then determined by FACS before
and 48 h after growth factor starvation. As shown in
Figure 7a, after 2 days of starvation, the percentage of
SHP-2 transduced cells dropped much faster than that
of the GFP vector transduced cells, suggesting that WT
SHP-2 overexpressing cells died much faster than
nontransduced cells.
Furthermore, we transduced WT and STAT5-deficient bone marrow progenitor cells with SHP-2
phosphatase and the GFP vector control. Transduced cells were sorted and maintained in IL-3-containing medium for 2 days and then assayed for survival
following IL-3 withdrawal. Consistent with the
results shown in Figure 6, overexpression of SHP-2
phosphatase in WT primary hematopoietic cells
decreased cell survival, suggesting that excessive
SHP-2 enzyme sensitized hematopoietic cells to growth
factor
deprivation-induced
death.
Interestingly,
the effect of SHP-2 overexpression on cell survival
in STAT5-deficient hematopoietic cells was diminished (Figure 7b), indicating an important role for
STAT5 in the SHP-2 regulation of hematopoietic cell
survival.
Discussion
In this report we have provided evidence that SHP-2
tyrosine phosphatase negatively regulates hematopoietic
cell survival. Enforced expression of SHP-2 in bone
marrow progenitor cells compromised their hematopoietic potential and enhanced their susceptibility to
apoptosis following growth factor deprivation. Biochemical analyses suggest that SHP-2 modulates hematopoietic cell survival by interaction with STAT5. We
have shown that SHP-2 associates with and dephosphorylates STAT5. Although IL-3 activation of Jak2
kinase was increased by overexpression of WT SHP-2,
tyrosyl phosphorylation of the Jak2 downstream substrate, STAT5, was disproportionately decreased. Following IL-3 starvation, tyrosyl phosphorylation of
Figure 6 Overexpression of SHP-2 in hematopoietic progenitor
cells reduces IL-3-induced differentiation and proliferation. Bone
marrow cells harvested from femurs of WT mice were transduced
with WT SHP-2 or GFP control retroviruses using the co-culture
system described in Materials and methods. (a) Control and WT
SHP-2 transduced cells were sorted and plated (2 104/ml) in
IMDM methylcellulose medium supplemented with a full combination of hematopoietic growth factors (Full) (see Materials and
methods) or IL-3 (20 ng/ml) only. Hematopoietic colonies were
scored after incubation for 7 days. Expression of SHP-2 in FACSsorted cells transduced with control and WT SHP-2 is shown on
the lower panel. (b) Control or SHP-2 transduced cells were sorted
and seeded into IL-3-containing (20 ng/ml) RPMI 1640 medium
with 10% FBS. Viable (trypan blue negative) cells were counted on
a hematocytometer 72 h later. Two independent experiments were
performed and similar results were obtained. Results shown are the
mean7s.d. of triplicates from one experiment
STAT5 required for its antiapoptotic activity was
rapidly diminished in SHP-2 overexpressing cells. These
results suggest that, in addition to its role in hematopoietic cell proliferation and differentiation, SHP-2 is
also involved in regulation of hematopoietic cell
survival, thereby providing new insights into the
biological function of SHP-2 phosphatase in hematopoietic cell processes.
Oncogene
SHP-2 phosphatase and hematopoietic cell survival
J Chen et al
3666
Figure 7 Negative role of SHP-2 in hematopoietic cell survival is mediated by STAT5. (a) WT bone marrow cells were transduced
with SHP-2 phosphatase and the GFP vector control through retroviral-mediated gene transfer. Following the retroviral infection,
cells (unsorted) were maintained and expanded in IL-3, IL-6, and SCF- containing medium for 48 h. Percentages of gene transduced
cells (GFP positive) in the mixed cell populations were then examined before and 48 h after growth factor withdrawal. (b) WT and
STAT5-deficient bone marrow cells were transduced with SHP-2 phosphatase and the GFP vector control as above. Transduced cells
were sorted and maintained in IL-3-containing medium (20 ng/ml) for 2 days. Cell survival rates were then determined at the indicated
time points after the medium was changed to IL-3-free medium. Expression of SHP-2 and STAT5 in the sorted WT and STAT5/ cells
transduced with SHP-2 and control vector was examined (the right panel)
Complex roles of SHP-2 phosphatase in hematopoietic
cell regulation
SHP-2 appears to play compound roles in hematopoietic
cell processes. Although SHP-2 targeted deletion mutation severely suppresses hematopoietic development (Qu
et al., 1997, 1998), this might be mainly due to loss of its
promoting role in early differentiation process of ES
cells (Qu and Feng, 1998; Chan et al., 2003). Therefore,
it has been unclear how SHP-2 phosphatase is exactly
involved in hematopoietic stem/progenitor cell processes. Since SHP-2 knockout embryos do not survive
past midgestation (Qu et al., 1997; Saxton et al., 1997),
the number of SHP-2 mutant hematopoietic cells
available is extremely limited. Dominant-negative inhibition and overexpression approaches are very useful
for addressing certain aspects of SHP-2 function in the
hematopoietic compartment. We have been investigatOncogene
ing the role of SHP-2 phosphatase in IL-3-mediated
hematopoietic cell responses. By using a dominantnegative approach, we previously reported that SHP-2
functioned in both enzymatic activity-dependent and independent manners in IL-3-induced signaling and
cellular responses (Yu et al., 2003). Complete removal of
SHP-2 protein results in loss of all its functions, so IL-3
signal transduction in SHP-2-deficient cells was essentially blocked. However, in catalytically inactive SHP-2 overexpressing Ba/F3 cells, only the
IL-3-induced Jak/STAT and Erk pathways were
attenuated. Due to the important role in Jak2 and Erk
activation, SHP-2 catalytic activity is required for
optimal hematopoietic cell responses to IL-3-induced
proliferation and differentiation (Wheadon et al., 2003;
Yu et al., 2003).
By using an overexpression approach, we now
provide evidence that SHP-2 also modulates growth
SHP-2 phosphatase and hematopoietic cell survival
J Chen et al
3667
factor-dependent hematopoietic cell survival. Biological
and biochemical data in this report showed that in spite
of the positive role of SHP-2 in IL-3 activation of Jak2
and Erk kinases, enforced expression of WT SHP-2
increased Ba/F3 cell susceptibility to IL-3 depletioninduced apoptosis. Moreover, overexpression of SHP-2
in primary hematopoietic progenitor cells in fact
compromised their hematopoietic potential both in vitro
(Figure 6 and 7) and in vivo (Chen and Qu, unpublished
data).
Recently, several gain-of-function mutations of SHP2 have been identified in human juvenile myelomonocytic leukemia, myelodysplastic syndromes, and acute
myeloid leukemia (Tartaglia et al., 2003). These findings
would indicate that excessive SHP-2 phosphatase might
contribute to hematopoietic cell transformation. However, our studies in this report clearly suggest that this is
not likely to be the case, thus raising an interesting
question on how the gain-of-function mutations of
SHP-2 contribute to hematopoietic cell transformation.
It is likely that in addition to the increased catalytic
activities, the gain-of-function SHP-2 mutants also have
altered protein–protein interacting properties and that
these changes together contribute to leukemogenesis.
Certainly, further studies are needed to determine how
these mutants disturb hematopoietic cell signaling and
cellular processes.
SHP-2 appears to modulate hematopoietic cell survival by
dephosphorylation of STAT5
One interesting observation in this report is that while
IL-3-induced activation of Jak2 kinase was increased by
overexpression of SHP-2, tyrosyl phosphorylation of the
Jak2 substrate–STAT5 was decreased (Figure 3a), and
that following IL-3 withdrawal, the decay of STAT5
phosphorylation in SHP-2 overexpressing cells was
dramatically accelerated (Figure 4a). These results,
together with subsequent physical association analysis
and the in vitro dephosphorylation assay (Figure 4c and
d), strongly support the functional interplay between
SHP-2 and STAT5. It is important to point out that
although SHP-2 inactivates STAT5, the net tyrosine
phosphorylation status of STAT5 in the catalytically
inactive SHP-2 (SHP-2 C/S) cells was not increased, but
instead was decreased (Figure 2), this is because
activation of upstream Jak2 kinase in SHP-2 C/S
overexpressing cells was significantly decreased by the
interference of the catalytic activity of endogenous
SHP-2 (Yu et al., 2003).
The interaction between SHP-2 and STAT5, in
particular, the physiological significance of their functional interplay has not been well characterized. The
association between SHP-2 and STAT5 was first
reported in the IL-2-induced signal transduction (Yu
et al., 2000). Following that, SHP-2 and STAT5 were
found to be in the same immunocomplex in mouse
mammary HC11 and human breast cancer T47D cells
and to translocate into the nucleus upon stimulation by
prolactin (Chughtai et al., 2002). More recently, SHP-2
was shown to dephosphorylate STAT5a in mouse
fibroblast cells (Chen et al., 2003). However, the
physiological significance of their interaction, particularly, in hematopoietic cell processes, has not been
defined. Our studies have now demonstrated that SHP-2
associates with STAT5 and that this interaction
negatively regulates IL-3-induced cell survival and
hematopoietic activity.
It appears that enhanced dephosphorylation of
STAT5 by excessive SHP-2 accounts for increased cell
death. This notion is supported by the following
evidence. First, STAT5 plays an important role in
hematopoietic cell survival. Following growth factor
withdrawal, STAT5-deficient primary hematopoietic
cells and dominant-negative STAT5 overexpressing
Ba/F3 cells are more susceptible to undergo apoptosis
than WT or the vector control transduced counterparts
(Figure 5). Moreover, STAT5-deficient bone marrow
hematopoietic cells showed decreased hematopoietic
potential in sub-optimal IL-3-containing medium (data
now shown). Second, starvation-induced reduction of
the expression levels of antiapoptotic proteins Bcl-XL
and pim-1, two STAT5 downstream targets, was
enhanced by overexpression of SHP-2 phosphatase.
Third, overexpression of SHP-2 decreases cell survival
only in WT but not in STAT5-deficient hematopoietic
cells. No significant difference in cell survival was
observed between SHP-2 overexpressing and the vector
transduced STAT5/ cells (Figure 7b), suggesting that
the negative regulation of growth factor-dependent
hematopoietic cell survival by SHP-2 is mediated by
STAT5. However, since STAT5 plays an important role
in hematopoietic cell survival (Figure 5), it remains
possible that diminished SHP-2 effect in STAT5/
hematopoietic cells might be attributed to lack of
STAT5 activation of downstream antiapoptotic molecules. Certainly, further experiments are required to
address this concern.
Materials and methods
Mice, cell lines, and reagents
STAT5a/b þ / mice and dominant-negative STAT5aD757 were
generously provided by Dr James Ihle (Memphis, TN, USA).
The mouse colony was maintained at the American Red Cross
Vivarium. STAT5ab/ mice were produced and genotyped as
previously reported (Teglund et al., 1998; Bunting et al., 2002).
Ba/F3, a murine pro-B lymphoma cell line was routinely
maintained in RPMI-1640 medium with 10% fetal bovine
serum (FBS) and 10% conditioned medium produced by
murine IL-3 cDNA transfected XB-30 hematopoietic cells.
Anti-SHP-2, anti-STAT5a, anti-STAT5b, anti-Bcl-XL, antiErk, and anti-phospho-Erk antibodies (Abs) were purchased
from Santa Cruz Biotechnology (Santa Cruz, CA, USA). AntiSTAT5 Ab that detects both STAT5a and STAT5b was
supplied by BD Transduction Laboratories (San Diego, CA,
USA). Anti-phospho-tyrosine Ab (PY) (4G10) and Jak2
antiserum were obtained from Upstate Biotechnology Inc.
(Lake Placid, NY, USA). Anti-phospho-Akt and anti-Akt Abs
were purchased from Cell Signaling Technology (Beverly, MA,
USA). The Cellular Caspase-3 Activity Assay Kit was supplied
by Calbiochem (La Jolla, CA, USA).
Oncogene
SHP-2 phosphatase and hematopoietic cell survival
J Chen et al
3668
Immunoprecipitation and immunoblotting
Cells were lysed in RIPA buffer (50 mM Tris-HCl pH 7.4, 1%
NP-40, 0.25% Na-deoxycholate, 150 mM NaCl, 1 mM EDTA,
1 mM NaF, 2 mM Na3VO4, 10 mg/ml leupeptin, 10 mg/ml
aprotinin, and 1 mM PMSF). Whole-cell lysates (500 mg) were
immunoprecipitated with 1–2 mg purified Abs or 2 ml antiserum
Abs. Immunoprecipitates were washed three times with
HNTG buffer (20 mM HEPES pH 7.5, 150 mM NaCl, 1%
Glycerol, 0.1% Triton X-100, and 1 mM Na3VO4) and resolved
by SDS–PAGE followed by immunoblotting with the indicated Abs.
Dephosphorylation assay
GFP empty vector transduced Ba/F3 cells were starved for 5 h
and then stimulated with IL-3 for 15 min. Whole-cell lysates
were prepared and immunoprecipitated with anti-STAT5b Ab.
The immunoprecipitates were washed twice with phosphatase
assay buffer (150 mM NaCl, 50 mM Tris-HCl, pH 7.4, 5 mM
dithiothreitol) and then incubated in the same buffer with or
without purified GST fusion proteins at 371C for various
periods of time. Reactions were terminated by adding SDS–
PAGE loading buffer. Samples were resolved by SDS–PAGE
followed by anti-PY immunoblottings.
Generation of SHP-2 retroviral producer cell line and
transduction of primary bone marrow hematopoietic stem/
progenitor cells
WT SHP-2 cDNA was cloned into the MSCV-IR-GFP
retroviral vector (Persons et al., 1999) containing an internal
ribosomal entry sequence (IRES) driving expression of a
downstream GFP gene to facilitate tracking of transduced
cells. Ecotropic GP þ E86-based retroviral producer cell lines
were generated by transduction with retroviral supernatant
produced by 293T cells that were transiently cotransfected
with pQEPAM3 (Minus E) packaging plasmid, pSraG
(VSV-G) envelope plasmid, and the recombinant retroviral
plasmid. To transduce primary hematopoietic stem/progenitor
cells with WT SHP-2, nucleated bone marrow cells harvested
from femurs of 4-week-old mice were prestimulated in RPMI1640 medium containing 10% FBS, SCF (50 ng/ml), IL-3
(20 ng/ml), and IL-6 (50 ng/ml) for 2 days, and then cocultured with irradiated (1500 rad) retroviral producer cells in
the presence of polybrene (6 mg/ml) for 48 h. Transduced cells
were sorted by FACS based on GFP expression
Generation of BMDM
BMDM were prepared from 4-week-old STAT5/ mice and
WT littermates as we previously described (Yu et al., 2002).
Briefly, bone marrow cells harvested from femurs were
incubated in Dulbecco modified Eagle medium (DMEM)
supplemented with 15% FBS and 20% L1 cell conditional
medium (L-CM) as a source of M-CSF. On the second day,
nonadherent cells were collected and seeded into new tissue
culture plates at the concentration of 2 105 cells/ml. After 4–5
days of culture in L-CM and IL-3-containing medium,
nonadherent cells were then collected for cell survival assay
after growth factor deprivation.
Hematopoietic progenitor assay
Bone marrow cells (2 104 cells/ml) were assayed for colonyforming units (CFUs) in 0.9%. methylcellulose IMDM
medium containing 30% FBS, glutamine (104 M), b-mecaptoethanol (3.3 105 M), and a combination of hematopoietic
growth factors (5% pokeweed mitogen-stimulated spleen cell
conditioned medium, 20 ng/ml IL-3, 50 ng/ml SCF, 2 U/ml
EPO, and 0.1 mM hemin) (Qu et al., 1998, 2001; Yu et al.,
2002). After 7 days of culture at 371C in a 5% CO2 incubator,
hematopoietic cell colonies were counted under an inverted
microscope.
Acknowledgements
We thank Ms. Teresa Hawley for cell sorting and Drs James
Ihle, Robert Hawley, Achsah Keegan, and Christine Couldrey
for the reagents, helpful discussions, and critical reading of the
manuscript. This work was supported by The US National
Institutes of Health Grant R01 HL68212 (to CKQ).
References
Aoki N and Matsuda T. (2000). J. Biol. Chem., 275, 39718–
39726.
Bennett AM, Hausdorff SF, O’Reilly AM, Freeman RM and
Neel BG. (1996). Mol. Cell. Biol., 16, 1189–1202.
Bone H, Dechert U, Jirik F, Schrader JW and Welham MJ.
(1997). J. Biol. Chem., 272, 14470–14476.
Bunting KD, Bradley HL, Hawley TS, Moriggl R, Sorrentino
BP and Ihle JN. (2002). Blood, 99, 479–487.
Chan RJ, Johnson SA, Li Y, Yoder MC and Feng GS. (2003).
Blood, 102, 2074–2080.
Chen Y, Wen R, Yang S, Schuman J, Zhang EE, Yi T,
Feng GS and Wang D. (2003). J. Biol. Chem., 278, 165
20–16527.
Chughtai N, Schimchowitsch S, Lebrun JJ and Ali S. (2002).
J. Biol. Chem., 277, 31107–31114.
Datta SR, Dudek H, Tao X, Masters S, Fu H, Gotoh Y and
Greenberg ME. (1997). Cell, 91, 231–241.
del Peso L, Gonzalez-Garcia M, Page C, Herrera R and Nunez
G. (1997). Science, 278, 687–689.
Dragovich T, Rudin CM and Thompson CB. (1998).
Oncogene, 17, 3207–3213.
Oncogene
Dumon S, Santos SC, Debierre-Grockiego F, GouilleuxGruart V, Cocault L, Boucheron C, Mollat P, Gisselbrecht
S and Gouilleux F. (1999). Oncogene, 18, 4191–4199.
Gu H, Pratt JC, Burakoff SJ and Neel BG. (1998). Mol. Cell,
2, 729–740.
Hadari YR, Kouhara H, Lax I and Schlessinger J. (1998). Mol.
Cell. Biol., 18, 3966–3973.
Nosaka T, Kawashima T, Misawa K, Ikuta K, Mui AL and
Kitamura T. (1999). EMBO J., 18, 4754–4765.
Persons DA, Allay JA, Allay ER, Ashmun RA, Orlic D, Jane
SM, Cunningham JM and Nienhuis AW. (1999). Blood, 93,
488–499.
Qu CK. (2002). Biochim. Biophys. Acta., 1592, 297–301.
Qu CK and Feng GS. (1998). Oncogene, 17, 433–439.
Qu CK, Nguyen S, Chen J and Feng GS. (2001). Blood, 97,
911–914.
Qu CK, Shi ZQ, Shen R, Tsai FY, Orkin SH and Feng GS.
(1997). Mol. Cell. Biol., 17, 5499–5507.
Qu CK, Yu WM, Azzarelli B, Cooper S, Broxmeyer HE and
Feng GS. (1998). Mol. Cell. Biol., 18, 6075–6082.
Rane SG and Reddy EP. (2002). Oncogene, 21, 3334–3358.
SHP-2 phosphatase and hematopoietic cell survival
J Chen et al
3669
Saxton TM, Henkemeyer M, Gasca S, Shen R, Rossi DJ,
Shalaby F, Feng GS and Pawson T. (1997). EMBO J., 16,
2352–2364.
Shelburne CP, McCoy ME, Piekorz R, Sexl V, Roh KH,
Jacobs-Helber SM, Gillespie SR, Bailey DP, Mirmonsef P,
Mann MN, Kashyap M, Wright HV, Chong HJ, Bouton
LA, Barnstein B, Ramirez CD, Bunting KD, Sawyer S,
Lantz CS and Ryan JJ. (2003). Blood, 102, 1290–1297.
Shi ZQ, Lu W and Feng GS. (1998). J. Biol. Chem., 273,
4904–4908.
Shinjyo T, Kuribara R, Inukai T, Hosoi H, Kinoshita T,
Miyajima A, Houghton PJ, Look AT, Ozawa K and Inaba
T. (2001). Mol. Cell. Biol., 21, 854–864.
Silva M, Benito A, Sanz C, Prosper F, Ekhterae D, Nunez G
and Fernandez-Luna JL. (1999). J. Biol. Chem., 274,
22165–22169.
Socolovsky M, Fallon AE, Wang S, Brugnara C and Lodish
HF. (1999). Cell, 98, 181–191.
Songyang Z, Baltimore D, Cantley LC, Kaplan DR and
Franke TF. (1997). Proc. Natl. Acad. Sci. USA, 94,
11345–11350.
Tang TL, Freeman Jr RM, O’Reilly AM, Neel BG and Sokol
SY. (1995). Cell, 80, 473–483.
Tartaglia M, Niemeyer CM, Fragale A, Song X, Buechner J,
Jung A, Hahlen K, Hasle H, Licht JD and Gelb BD. (2003).
Nat. Genet., 34, 148–150.
Teglund S, McKay C, Schuetz E, van Deursen JM,
Stravopodis D, Wang D, Brown M, Bodner S, Grosveld G
and Ihle JN. (1998). Cell, 93, 841–850.
Tonks NK and Neel BG. (2001). Curr. Opin. Cell Biol., 13,
182–195.
Wang D, Moriggl R, Stravopodis D, Carpino N, Marine JC,
Teglund S, Feng J and Ihle JN. (2000). EMBO J., 19,
392–399.
Welham MJ, Dechert U, Leslie KB, Jirik F and Schrader JW.
(1994). J. Biol. Chem., 269, 23764–23768.
Wheadon H, Edmead C and Welham MJ. (2003). Biochem. J.,
376, 147–157.
Wickremasinghe RG and Hoffbrand AV. (1999). Blood, 93,
3587–3600.
Yu CL, Jin YJ and Burakoff SJ. (2000). J. Biol. Chem., 275,
599–604.
Yu WM, Hawley TS, Hawley RG and Qu CK. (2002). Blood,
99, 2351–2359.
Yu WM, Hawley TS, Hawley RG and Qu CK. (2003).
Oncogene, 22, 5995–6004.
Yuan L, Yu WM, Yuan Z, Haudenschild CC and Qu CK.
(2003). J. Biol. Chem., 278, 15208–15216.
Zamorano J, Wang HY, Wang R, Shi Y, Longmore GD and
Keegan AD. (1998). J. Immunol., 160, 3502–3512.
Zha J, Harada H, Yang E, Jockel J and Korsmeyer SJ. (1996).
Cell, 87, 619–628.
Oncogene