(Received for publication, November 4, 1996, and in revised form, December 17, 1996)
From the Terry Fox Laboratory, British Columbia Cancer Agency, and The Biomedical Research Centre, University of British Columbia, Vancouver, British Columbia V5Z 1L3, Canada
In this study we have investigated the role that the Src homology 2 domain (SH2) of the 145-kDa 5-phosphatase, SH2-containing inositol phosphatase (SHIP), plays in three of the properties that have been associated with this protein following cytokine stimulation: its association with Shc, its tyrosine phosphorylation, and its inhibition of hemopoietic cell growth. In vitro studies using this SH2 domain revealed that it was capable of binding directly to the Tyr(P)317 motif of Shc with a KD of approximately 290 nM, in keeping with other specific SH2/Tyr(P) interactions. In vivo analysis revealed the SH2 and NPXpY motifs of SHIP acted together, with the Tyr(P)317 and phosphotyrosine binding (PTB) domains of Shc, respectively, to ensure a high affinity SHIP·Shc complex. Expression of cDNAs encoding hemagglutinin-tagged wild type and SH2-inactivated forms of SHIP in the murine hemopoietic cell line DA-ER revealed that wild type SHIP becomes both tyrosine-phosphorylated and associated with Shc following interleukin-3 stimulation, as expected, but the SH2-inactivated SHIPs do neither. Moreover, while the growth rates of parental DA-ER cells and cells expressing these various SHIP constructs are identical, the wild type SHIP-expressing cells die, via programmed cell death, far more rapidly than parental cells. Cells expressing SH2-inactivated SHIPs, on the other hand, show either a reduced or no effect on apoptosis. These results suggest that the SH2 domain of SHIP is required not only for the tyrosine phosphorylation of SHIP and Shc association following cytokine stimulation but also for its induction of apoptosis.
Several years ago we (1, 2) and others (3, 4) identified a 145-kDa
protein (p145) within hemopoietic cell lines that became
tyrosine-phosphorylated and associated with Shc following stimulation
with various cytokines. Interestingly, we found that the association
between p145 and Shc could be inhibited with phosphopeptides corresponding to the pY317VNV sequence within Shc (2).
Based on these results we proposed that p145 contained a Src homology 2 domain (SH2)1, and that it competed with
Grb2 for binding to the Tyr(P)317 sequence within Shc (2).
More recently we cloned the cDNA for this 145-kDa protein (5) and
found that the predicted amino acid sequence did indeed contain an SH2
domain at its amino terminus as well as two phosphotyrosine binding
(PTB) consensus sequences (i.e. NPXY sequences)
(6), several proline rich SH3 binding regions, and two motifs highly
conserved among inositol polyphosphate 5-phosphatases. Based on these
properties we called this protein SHIP for 2-containing
nositol hosphatase (5). Unlike most inositol
polyphosphate 5-phosphatases that hydrolyze phosphatidylinositol 4,5-P2-bisphosphate (PI-4,5-P2) and/or inositol
1,4,5-trisphosphate (7), SHIP selectively hydrolyzes the 5-phosphate
from inositol 1,3,4,5-tetraphosphate, and phosphatidylinositol
3,4,5-trisphosphate (PI-3,4,5-P3) (5), two inositol
polyphosphates recently implicated in growth factor-mediated signaling
(8-11).
Concurrent with our cloning of SHIP, Lioubin et al. (12) independently obtained the cDNA for this protein using a yeast two-hybrid system based on its affinity for the PTB domain of Shc. They further went on to show that ectopic expression of SHIP in FDC-P1 cells expressing the macrophage colony-stimulating factor receptor, c-Fms, led to a reduction in the size of the macrophage colony-stimulating factor receptor or interleukin-3 (IL-3)-stimulated colonies in soft agar (12). Subsequently, Kavanaugh et al. monitored the purification of this 145-kDa protein by its ability to bind to the Shc PTB in vitro. Their strategies indicated that Shc was capable of binding via its PTB domain to SHIP, at least in vitro, and this was potentially at odds with our finding that phosphopeptides corresponding to the pY317VNV motif within Shc inhibit SHIP binding to Shc (2). Because of the potential importance of the interaction of SHIP with Shc in mediating cytokine-induced signals we were interested in establishing whether the SH2 domain of SHIP played a role in this interaction as well as in its reported inhibitory effect on cell growth.
The production and purification of COS
cell-derived murine IL-3 and granulocyte macrophage colony-stimulating
factor were as described previously (14). Glutathione
S-transferase (GST) fusion proteins consisting of the 27-kDa
amino terminus of GST linked to the SH2 domain of murine SHIP (amino
acids 7-133) (5), bovine PLC-1 (amino-terminal) (residues
547-659), and murine Grb2 (residues 60-158) were expressed in
Escherichia coli in pGEX-2T plasmids (Pharmacia Biotech
Inc., Baie d'Urfe, Quebec, Canada) and purified from the sonicated
bacteria using glutathione-agarose (Pharmacia) as described previously
(1). Rabbit antiserum to SHIP was generated by immunizing animals with
the GST-SHIP SH2 fusion protein described above. The phosphopeptides
used for the inhibition studies consisted of the 12-mer RRASVSpYVNVQN
corresponding to the sequence flanking Tyr317 within Shc,
the 11-mer EMINPNpYIGMG corresponding to the sequence flanking the NPNY
within SHIP (both synthesized by the Sequencing Center at the
University of Victoria, British Columbia, Canada), and the 11-mer
STDpYSSGGSQG corresponding to an intracellular region of the
erythropoietin receptor (15) (generously provided by Dr. Taolin Yi,
Cleveland Clinic Foundation, Cleveland, OH). The anti-phosphotyrosine
(anti-Tyr(P)) monoclonal antibody (mAb) was purchased from Upstate
Biotechnology (Lake Placid, NY). Both affinity-purified rabbit
polyclonal antibodies to Shc (for immunoprecipitations) and mAb to Shc
(for Western blotting) were obtained from Transduction Laboratories
(Lexington, KY). The anti-hemagglutinin (HA) mAb was from Babco
(Richmond, CA). Horseradish peroxidase-conjugated second antibodies
were purchased from Jackson ImmunoResearch Laboratories, Inc. (West
Grove, PA). Protein grade Nonidet P-40 was from Calbiochem. The
enhanced chemiluminescence Western blotting reagents were obtained from
Pierce.
HA-tagged forms of wild type (WT)-SHIP, SH2-SHIP
(lacking the second half of the SH2 domain), and R34G-SHIP (in which
the critical arginine in the FLVR sequence of the SH2 domain has been replaced with a glycine (9)) were generated by first fusing the HA-tag
in-frame at the 5
end to the entire murine WT-SHIP cDNA in a
murine stem cell virus vector containing the puromycin resistance gene,
pac (16). The
SH2-SHIP was then generated from this
construct by deleting the FspI, AccI fragment
corresponding to amino acids 44-149 within the SH2 domain. The R34G
point mutant was generated from the HA-tagged WT-SHIP using the
QuikChangTM site-directed mutagenesis kit (Stratagene). All
three plasmids were calcium phosphate transfected into the producer
cell line BOSC 23, and 48-h retroviral supernatants were used to infect DA-ER cells. DA-ER cells were selected for 2 weeks in puromycin, and
clones were analyzed for SHIP expression by Western analysis with
anti-HA mAbs.
Murine
B6SUtA1 cells, maintained in RPMI 1640 medium with 10%
fetal calf serum and 5 ng/ml granulocyte macrophage colony-stimulating factor or murine DA-ER cells (DA-3 cells expressing cell surface erythropoietin receptors (1)) infected with HA-tagged WT-, R34G-, or
SH2-SHIP, maintained in RPMI 1640 medium with 10% fetal calf serum
and 5 ng/ml IL-3, were growth factor-deprived for 4-6 h at 37 °C in
RPMI 1640 medium containing 0.1% bovine serum albumin and then
stimulated at 37 °C for 5 min with murine IL-3 (400 ng/ml). The
cells were then washed once with phosphate-buffered saline, solubilized
at 2 × 107 cells/ml with 0.5% Nonidet P-40 in
4 °C phosphorylation solubilization buffer (50 mM HEPES,
pH 7.4, 100 mM NaF, 10 mM NaPPi, 2 mM Na3VO4, 4 mM EDTA, 2 mM phenylmethylsulfonyl fluoride, 10 µg/ml leupeptin, and
2 µg/ml aprotinin), and subjected to immunoprecipitation and Western
blotting as described previously (2).
Cell lysates from B6SUtA1 cells, treated with or without IL-3 for 5 min at 37 °C, were either incubated immediately for 1 h at 4 °C with glutathione-agarose beads bearing a GST-fusion protein containing the SH2 domain of SHIP or supplemented with SDS to a final concentration of 1%, boiled for 5 min, and diluted 20-fold with phosphorylation solubilization buffer containing 0.5% Nonidet P-40 before incubation with the beads. The beads were then washed three times with the same buffer and boiled in SDS-sample buffer, and the eluted proteins were subjected to Western blot analysis with the anti-Tyr(P) mAb, 4G10.
Phosphopeptide Inhibition AssaysB6SUtA1 cells were incubated with IL-3 for 5 min at 37 °C and the cells rapidly lysed with 0.5% Nonidet P-40 in phosphorylation solubilization buffer in the presence or absence of 50 µM of either the phosphorylated 12-mer corresponding to the sequence flanking Tyr317 within Shc, a phosphorylated 11-mer corresponding to the NPXpY sequence recognized by the PTB domain of Shc (17), or a control tyrosine-phosphorylated peptide corresponding to an intracellular region of the erythropoietin receptor. The cell lysates were incubated with anti-SHIP antibodies, and the immune complexes were collected on protein A-Sepharose beads. The bound proteins were then eluted off the beads by heating at 100 °C for 3 min in SDS-sample buffer and subjected to Western analysis with the anti-Tyr(P) mAb, 4G10.
SH2 Binding Studies Using Surface Plasmon ResonanceGST
fusion proteins containing the SH2 domains of Grb2, SHIP,
and the amino-terminal SH2 of bovine PLC-1 were eluted from
glutathione-agarose beads at 23 °C with 75 mM HEPES, 150 mM NaCl, 5 mM dithiothreitol, 0.1% Triton
X-100, and 20 mM reduced glutathione and dialyzed for 2 days in phosphate-buffered saline, and protein concentrations were
determined using the bicinchoninic acid protein assay reagent (Pierce).
The binding affinities of the GST-SH2 fusion proteins were measured
with a BIAcore biosensor (18). The 12-mer phosphopeptide corresponding
to the sequence flanking Tyr317 within Shc was covalently
bound to the carboxymethylated dextran matrix of the sensor chip (CM5
sensorchip, Pharmacia). The chip surface was first activated with 50 mM N-hydroxysuccinimide and 200 mM
N-ethyl-N
-(3-diethylaminopropyl)-carbodiimide
hydrochloride for 3 min at a flow rate of 5 µl/min. The peptide
binding surface was prepared by injecting a solution of the
Tyr317 phosphopeptide (1 or 4 mg/ml) in 10 mM
HEPES, pH 7.4, and 150 mM NaCl over the activated matrix
for 5 min at a rate of 5 µl/min. Remaining active sites were blocked
with 1 M ethanolamine (pH 8.5) for 6 min. Several
concentrations of immobilized phosphopeptide were used for affinity
measurements. All kinetic measurements were performed in HBS buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 3.4 mM EDTA, 0.005% Nonidet P-20) at a flow rate of 5 µl/min. The affinity constants of the fusion proteins were determined by monitoring the binding interactions using the following
concentrations: GST-Grb2 SH2 (0, 100, 200, 400, 800, and 1200 nM); GST-SHIP SH2 (0, 400, 800, 1000, 2000, 3000, and 4000 nM); and GST-PLC-
1 amino-terminal SH2 (0, 1, 2, 4, 8, and 12 µM). After each binding interaction, the
phosphopeptide binding surface was regenerated with 0.1% SDS (5 µl)
with no loss of binding response. The apparent equilibrium dissociation
constants (KD values) were calculated from Scatchard
analysis of steady-state binding.
For DNA laddering studies, 2 × 106 DA-ER cells expressing HA-tagged WT- and mutant SHIPs were lysed in 50 µl of 50 mM Tris-HCl, pH 7.5, and 20 mM EDTA containing 0.5% Nonidet P-40, and protein and RNA were removed by digestion with RNase A (5 µg/µl) and proteinase K (2.5 µg/µl), respectively, and the ethanol-precipitated DNA was dissolved in gel-loading buffer and electrophoresed using 2% agarose gels containing 0.1 µg/ml EtBr (19). For propidium iodide and fluorescence-activated cell sorter analysis (FACS), 1 × 106 DA-ER cells expressing HA-tagged WT- and mutant SHIPs were washed in phosphate-buffered saline, incubated for 5 min in the dark in 1 ml of 50 mM sodium citrate, pH 8.0, 20 µg/ml propidium iodide, and 0.1% Triton X-100, and analyzed by FACS (20).
To
confirm our earlier studies, which suggested that p145 (SHIP)
interacted via its SH2 domain with Shc (2), lysates from B6SUtA1 cells, treated with and without IL-3, were
incubated with a bead-bound GST-fusion protein containing the SH2
domain of SHIP, and the bound proteins were subjected to Western
analysis using anti-Tyr(P) mAbs. As can be seen in the first two
lanes of Fig. 1A, the SH2 domain
of SHIP bound both a 70- and a 56-kDa tyrosine-phosphorylated protein
following IL-3 stimulation. To determine whether these phosphorylated
proteins bound directly or indirectly to this SH2 domain, lysates were
boiled in SDS before dilution and incubation with the beads (Fig.
1A, lanes 3 and 4). This resulted in
the disappearance of the 70-kDa protein (which reprobing experiments demonstrated to be the tyrosine phosphatase, SHP2 (also known as
Syp))2 but not the 56-kDa protein.
Reprobing this blot with anti-Shc antibodies demonstrated that the
56-kDa protein was Shc (Fig. 1A, lower panel) and
that the SH2 domain of SHIP binds directly to Shc in
vitro.
The SH2 Domain of SHIP Has an Affinity for the Tyr(P)317 Motif of Shc That Is Consistent with Other SH2/Tyr(P) Interactions
To determine if the SH2 domain of SHIP
was capable of binding to the Tyr(P)317 motif of Shc and
also to gain some insight into the affinity of the SH2 domain of SHIP
for this motif, we examined the relative binding affinities of
GST-fusion proteins containing the SH2 domains of SHIP, Grb2, and
PLC-1 for the Shc Tyr(P)317 phosphopeptide using surface
plasmon resonance. From three representative experiments, each with a
different concentration of the cross-linked phosphopeptide, titration
curves were carried out with the SH2 domains of Grb2 and SHIP, and the
apparent KD values calculated from equilibrium data
were determined to be approximately (2.9 ± 1.3) × 102 nM and (2.8 ± 0.3) × 101
nM for the SHIP-SH2 and Grb2-SH2 fusion proteins,
respectively. The PLC-
1 amino-terminal SH2 fusion did not show any
affinity for the binding surface up to a GST-SH2 concentration of 10 µM. Fig. 1B depicts the curves generated at 1 µM concentration of each SH2 domain for comparative
purposes. This confirms that the SH2 domain of SHIP is capable of
specifically interacting with the Tyr(P)317 motif of Shc
and that the affinity of this interaction is consistent with that
reported for other SH2 domains (21).
Lioubin et al.
(12) and Kavanaugh et al. (13) recently demonstrated that
the PTB domain of Shc is capable of binding to SHIP via the
NPXpY motifs of SHIP in vitro (and we have
confirmed this using bead-bound GST fusion proteins containing the PTB
domain of SHIP, data not shown). To determine whether the SHIP/Shc
association in vivo can be mediated solely via
SH2/Tyr(P)317 or NPXpY/PTB interactions, lysates
from B6SUtA1 cells, treated with and without IL-3 for 5 min, were subjected to anti-SHIP immunoprecipitation in the presence
and absence of phosphopeptides corresponding to the NPNpY and
Tyr(P)317 sequences in SHIP and Shc, respectively. Western
analysis, using anti-Shc mAbs revealed that both phosphopeptides, but
not a control phosphopeptide, inhibited the co-precipitation of Shc
with SHIP (Fig. 2, top panel). Probing this
blot with anti-Tyr(P) mAbs demonstrated that the tyrosine
phosphorylation level of SHIP was unaffected by the competing
phosphopeptides, suggesting that these phosphopeptides were not causing
SHIP/Shc dissociation by binding to and activating a SHIP or
Shc-associated SH2-containing tyrosine phosphatase (data not shown).
Reprobing this blot with anti-SHIP antibodies confirmed equal loading
(lower panel). Since both the NPNpY and
Tyr(P)317 phosphopeptides were capable of disrupting
SHIP·Shc complexes it suggests that the SH2 domain and the NPNpY
motif of SHIP are both required for a high affinity interaction with
Shc. This dual interaction may explain, in part, why we observed
substantial levels of both Shc·SHIP and Shc·Grb2 complexes in these
cells even though our BIAcore results indicated that the isolated SH2 domain of SHIP had a 10-fold lower affinity than the isolated SH2
domain of Grb2 for the Tyr(P)317 motif of Shc. In this
regard it should be pointed out that anti-Grb2 immunoprecipitates do
not contain SHIP (2), and anti-SHIP immunoprecipitates do not contain
Grb2 in B6SUtA1 cells (data not shown). Thus SHIP and Grb2
may compete in vivo for Shc.
SHIP Mutants Lacking a Functional SH2 Domain Do Not Bind Shc and Do Not Become Tyrosine Phosphorylated in Vivo in Response to IL-3
To
explore the relevance of this SH2 domain to downstream events in
vivo, DA-ER cells were retrovirally infected with cDNAs for
5-HA-tagged forms of WT-,
SH2- (lacking a functional SH2 domain),
and R34G- (lacking a critical arginine in the FLVR sequence of the SH2
domain) (9) SHIP. The constructs were shown to retain the proper coding
frame by sequencing the resulting cDNA and demonstrating that the
protein products were immunoprecipitable with anti-15-mer anti-SHIP
antiserum (5). Moreover, anti-HA antibody immunoprecipitates from these
infected cells, but not uninfected cells, possessed SHIP inositol
polyphosphate 5-phosphatase activity, and the specific activity
(i.e. inositol polyphosphate 5-phosphatase activity/mg of
SHIP) was similar in all constructs (data not shown). DA-ER cell clones
expressing HA-WT-SHIP, HA-
SH2-SHIP, and HA-R34G-SHIP were then
stimulated with IL-3, and total cell lysates were subjected to Western
analysis with anti-HA antibodies (Fig. 3A).
This revealed that while all three constructs gave rise to lower
molecular weight forms of SHIP, most likely through proteolytic
degradation at the carboxyl terminus (since they were HA-tagged at the
amino terminus), they clearly expressed the expected full-length forms. Aliquots of these same cell lysates were also immunoprecipitated with
anti-Shc antibodies prior to Western analysis with anti-HA antibodies.
This revealed that HA-WT-SHIP co-precipitated with Shc, as expected,
but HA-
SH2-SHIP and HA-R34G-SHIP did not (Fig. 3B). The
absence of the lower molecular weight forms of HA-WT-SHIP in this
anti-Shc immunoprecipitate most likely reflects the requirement of the
carboxyl-terminal NPXY motifs of SHIP for the in
vivo association of SHIP with Shc.
The same DA-ER cell clones were treated with and without IL-3, and
lysates were subjected to anti-Tyr(P) mAb immunoprecipitation and
Western analysis with anti-HA mAbs. As can be seen in Fig. 3C, the HA-SH2-SHIP and HA-R34G-SHIP were not detectably
tyrosine-phosphorylated, while HA-WT-SHIP was strongly phosphorylated
under the same conditions. The reciprocal experiment, in which anti-HA
immunoprecipitates were carried out and Western analysis performed with
anti-Tyr(P) mAbs, yielded identical results (data not shown). Thus the
SH2 domain of SHIP is also required for SHIP to become tyrosine
phosphorylated in response to IL-3 stimulation. Interestingly, this
effect on its tyrosine phosphorylation suggests that the SH2 domain of
SHIP may play a role either in the binding of SHIP to a tyrosine kinase (directly or through its interaction with Shc) or in the translocation of SHIP to a tyrosine kinase. Related to this, it has been reported that IL-3 stimulates not only the activation of the IL-3R-associated tyrosine kinase Jak2 (22) but also Fps/Fes (23) and various Src family
members (24, 25).
In preliminary studies to
investigate the biological role of SHIP we found we could not express
HA-WT-SHIP in DA-ER cells at more than twice endogenous SHIP levels,
perhaps reflecting a reduced ability of these cells to survive. Lioubin
et al. (12) also found they could not express exogenous
WT-SHIP at more than twice endogenous levels in FD-Fms cells.
Nonetheless, even at these slightly elevated levels, they observed an
inhibition of the macrophage colony-stimulating factor receptor-induced
colony size (12). To gain some insight into the biological consequences of overexpressing WT SHIP in DA-ER cells, suspension cultures containing parental DA-ER and two independently isolated clones infected with HA-tagged WT-SHIP were initiated with 2 × 105 cells/ml in 10% fetal calf serum plus 5 ng/ml IL-3,
and duplicate samples were counted following trypan blue-staining every
24 h for 4 days. As can be seen in Fig.
4A, the exponential growth rate of all the
cells was identical. However, once confluence was achieved the survival
of the two cell types was markedly different. Interestingly, cell
counts at various times during days 2 and 3 revealed that the total
number of cells (i.e. the trypan blue positive plus negative
cells) was actually the same in all the cultures, but the level of
trypan blue positive cells was substantially higher in the WT SHIP
cultures. These results suggested that overexpression of WT SHIP
reduced the viability of confluent DA-ER cells. This effect was not
observed with logarithmically growing DA-ER cells starved of growth
factor.
SHIP reduces the viability of IL-3-stimulated
DA-ER cells at confluence. A, parental DA-ER () and two
independently isolated clones of DA-ER cells expressing HA-WT-SHIP (
= WT6;
= WT12) were put into culture at 2 × 105
cells/ml in 10% fetal calf serum plus 5 ng/ml IL-3 and counted following trypan blue staining every 24 h for 4 days. Each point represents the mean ± S.E. of two replicates. Similar results were obtained in five independent experiments. B, parental
and two independently isolated clones of DA-ER cells expressing
HA-tagged WT-,
SH2-, and R34G-SHIP were put into culture at 2 × 105 cells/ml in 10% fetal calf serum plus 5 ng/ml IL-3,
and the number of trypan blue negative cells on day 3 were determined.
Each point represents the mean ± S.E. of three replicates with
parental cells assigned a value of 100%. Similar results were obtained
in five independent experiments. C, DNA fragmentation
analysis of parental cells and two clones each of HA-tagged WT-,
SH2-, and R34G-SHIP, taken on day 2. Lane 1, parental DA-ER; lanes 2 and 3, WT-SHIP6 and WT-SHIP12, respectively; lanes
4 and 5, two clones of
SH2-SHIP; lanes 6 and 7, two clones of R34G-SHIP.
We then compared the viability of two independently isolated clones of
HA-WT-SHIP, HA-SH2-SHIP, and HA-R34G-SHIP expressing DA-ER cells
with parental cells and found that the two HA-
SH2-SHIP-infected clones displayed a viability pattern in between that seen with parental
and WT cells while the HA-R34G-SHIP-infected clones consistently possessed a viability equal to or slightly greater than parental cells
(Fig. 4B).
To determine if the WT-SHIP-induced loss of viability was mediated by
an apoptotic pathway, DNA was extracted from the different cell clones
before differences in viability were detected by trypan blue exclusion,
i.e. on day 2, and DNA fragmentation was assessed by
electrophoresis on agarose gels (19). As can be seen in Fig. 4C, the extent of DNA laddering was far greater in the two
clones expressing HA-WT-SHIP than in the parental or HA-R34G-SHIP cell clones. The two HA-SH2-SHIP clones, on the other hand, showed an
intermediate pattern (consistent with our trypan blue studies). DNA
fragmentation was also assessed on days 2 and 3 by FACS using propidium
iodide staining (20), and the results were consistent with our trypan
blue staining and our DNA laddering results. For example, on day 2, HA-WT-SHIP-expressing cells showed the most DNA fragmentation
(e.g. 22.3 ± 0.5%), HA-
SH2-SHIP-expressing cells
showed less (15.8 ± 3.8%), and parental- (2.3 ± 0.5%) and R34G-SHIP- (3.5 ± 0.1%) expressing cells had the least. The SH2 domain of SHIP thus appears to be critical for the apoptotic effect of
SHIP since a very subtle single amino acid change (R34G) in full-length
SHIP, in an amino acid which has been shown previously to be critical
for the binding of SH2 domains to tyrosine-phosphorylated residues (9),
completely eliminates this effect. Interestingly, a complete
elimination of this SHIP effect was not observed with cells expressing
the HA-
SH2-SHIP. This may be due to the fact that a large section of
the SH2 domain was removed to generate this construct, and this may
have resulted in a significant conformational change and subsequent
unpredictable effects on signaling.
Mechanistically, this SHIP-induced loss of viability could occur through the ability of SHIP to compete with Grb2 for Shc and thus, potentially, reduce Ras activation. This is consistent with a recent report by Kinoshita et al. (26), in which they showed that the Ras pathway plays an important role in preventing apoptosis in IL-3-stimulated cells. Alternatively, since PI 3-kinase has been shown to prevent apoptosis in certain cell types (27), SHIP might reduce cell viability by hydrolyzing the primary in vivo product of PI 3-kinase, PI 3,4,5-P3.
We are currently generating inducible SHIP vectors so that the effects of WT and mutant forms of SHIP on intracellular signaling and various biological end points can be further investigated.
We thank Mark Ware for helpful discussions, Vivian Lam for excellent technical assistance, and Christine Kelly for typing the manuscript.