FLT3 receptors with internal tandem duplications promote cell ... - Nature

Oncogene (2004) 23, 3338–3349

& 2004 Nature Publishing Group All rights reserved 0950-9232/04 $25.00 www.nature.com/onc

FLT3 receptors with internal tandem duplications promote cell viability and proliferation by signaling through Foxo proteins Blanca Scheijen1,2, Hai T Ngo1,2, Hyun Kang1,2 and James D Griffin*,1,2 1

Department of Medical Oncology, Dana-Farber Cancer Institute, Mayer 540, 44 Binney Street, Boston, MA 02115, USA; Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Mayer 540, 44 Binney Street, Boston, MA 02115, USA

2

In about 30% of the patients with acute myeloid leukemia, activating FLT3 receptor mutations have been identified, often as in-frame internal tandem duplications (ITD) at the juxtamembrane domain of the receptor. FLT3-ITD receptors exhibit constitutive tyrosine kinase activity in the absence of FLT3 ligand (FL) binding, and when expressed in cytokine-dependent cell lines and primary hematopoietic cells suppress programmed cell death and increase cell division. However, the signaling pathways important for transformation, in particular the nuclear targets, are unknown. Here we demonstrate that FLT3ITD expression in Ba/F3 cells results in activation of Akt and concomitant phosphorylation of the Forkhead family member Foxo3a. Phosphorylation of Foxo proteins through FLT3-ITD signaling promotes their translocation from the nucleus into the cytoplasm, which requires the presence of conserved Akt phosphorylation sites in Forkhead transcription factors and PI3K activity. Induction of Foxo3a phosphorylation by FLT3-ITD receptors in Ba/ F3 cells correlates with the suppression of Foxo-target genes p27Kip1 and the proapoptotic Bcl-2 family member Bim. Specifically, FLT3-ITD expression prevents Foxo3a-mediated apoptosis and upregulation of p27Kip1 and Bim gene expression. These data indicate that the oncogenic tyrosine kinase FLT3 can negatively regulate Foxo transcription factors, thereby promoting cell survival and proliferation. Oncogene (2004) 23, 3338–3349. doi:10.1038/sj.onc.1207456 Published online 23 February 2004 Keywords: Akt; Bim; FLT3; forkhead transcription factors; p27Kip1

Introduction FMS-like tyrosine kinase-3 (FLT3/CD135) belongs to the type III tyrosine kinase receptor family, and is structurally *Correspondence: JD Griffin, Department of Medical Oncology, Dana-Farber Cancer Institute, Mayer 540, 44 Binney Street, Boston, MA 02115, USA; E-mail: James_Griffi[email protected] Received 21 April 2003; revised 14 December 2003; accepted 15 December 2003; Published online 23 February 2004

related to c-KIT/stem cell factor (SCF), macrophage colony-stimulating factor (M-CSF) and platelet-derived growth factor (PDGF) receptors (Scheijen and Griffin, 2002). FLT3 expression can be detected in gonads, placenta, peripheral and central nervous system, and on the surface of hematopoietic stem cells (HSC), uncommitted lymphoid and myeloid progenitors as well as CD14 þ monocytes (Rosnet et al., 1991; Lyman and Jacobsen, 1998). Accumulating data indicate that FLT3 represents the most frequently mutated gene in human acute myeloid leukemia (AML), with approximately one-third of the patients displaying somatic FLT3 gene alterations on chromosome 13q12, which result in FLT3-ligand (FL)independent tyrosine kinase activation of the FLT3 receptor (Gilliland and Griffin, 2002). In about threequarters of these patients, one or more internal tandem duplications (ITD) within the juxtamembrane (JM) region of the FLT3 receptor can be detected (Nakao et al., 1996), while point mutations, insertions or deletions involving codons 835/836 and 840/841 in the activation loop of the bipartite tyrosine kinase domain are present in the remaining cases (Abu-Duhier et al., 2001; Yamamoto et al., 2001; Spiekermann et al., 2002; Thiede et al., 2002). Interestingly, wild-type FLT3 receptors are also often highly expressed in pre-B acute lymphoblastic leukemia (ALL) (Rosnet et al., 1993; DaSilva et al., 1994; Meierhoff et al., 1995; Carow et al., 1996), and in pediatric leukemias carrying rearrangements involving the mixed-lineage leukemia (MLL) gene (Armstrong et al., 2002), suggesting that in vivo proliferation of these leukemic blast cells is augmented through autocrine or paracrine FL stimulation. It has been shown that FLT3-ITD receptors prevent apoptosis and mediate oncogenic transformation of cytokine-dependent cell lines (Zhao et al., 2000; Levis et al., 2002), block myeloid differentiation (Zheng et al., 2002), and induce a lethal myeloproliferative disease in mice, employing a bone marrow transplant assay (Kelly et al., 2002). Several studies have suggested that the signaling proteins STAT5, ERK and Akt are linked to the activation of FLT3 (Hayakawa et al., 2000; Mizuki et al., 2000; Zhang et al., 2000). The serine/threonine kinase Akt, also termed protein kinase B (PKB), is known to be a downstream target of phosphatidylinositol 3-kinase (PI3K) pathway and involved in mediating insulin, survival and cell proliferation reponses (Datta

FLT3-ITD receptors inhibit Foxo protein function B Scheijen et al

3339

et al., 1999; Lawlor and Alessi, 2001). Full activation of Akt requires recruitment to the plasma membrane through its N-terminal lipid-binding pleckstrin homology (PH) domain, where Ser473 phosphorylation occurs, followed by phosphoinositide-dependent kinase 1 (PDK1)-mediated phosphorylation of residue Thr308 (Scheid et al., 2002). Several important substrates of Akt have been identified (Lawlor and Alessi, 2001), including members of the Forkhead family of transcription factors (Brunet et al., 1999; Kops et al., 1999). In Caenorhabditis elegans, there is strong genetic evidence implicating Forkhead transcription factor DAF-16 as a critical target of the insulin receptor/ PI3K/PDK1/Akt pathway (Ogg et al., 1997; Paradis and Ruvkun, 1998). The Foxo proteins Foxo1 (FKHR), Foxo3a (FKHRL1) and Foxo4 (AFX) represent the mammalian orthologues of DAF-16, and phosphorylation of these Forkhead transcription factors by Akt inhibits their nuclear translocation and hence their ability to transactivate transcriptional target genes (Brunet et al., 1999; Kops et al., 1999; Rena et al., 1999; Tang et al., 1999). In mammals, Foxo1 is involved in insulin inhibition of hepatic glucose production, stimulation of b cell proliferation and adipocyte differentiation (Nakae et al., 2002, 2003), while Foxo3a has been implicated as a regulator of ovarian follicular growth activation (Castrillon et al., 2003). Foxo proteins, in their nonphosphorylated form, contribute to apoptosis by their ability to activate FasL (Brunet et al., 1999), Bim (Dijkers et al., 2000a), TRAIL (Modur et al., 2002) and TRADD (Rokudai et al., 2002) gene expression. Furthermore, Foxo proteins are able to control different cell cycle checkpoints by modulating the expression of pRb family member p130 (Kops et al., 2002), cyclin-dependent kinase (CDK) inhibitor p27Kip1 (Medema et al., 2000), cyclin D (Ramaswamy et al., 2002; Schmidt et al., 2002), cyclin B and polo-like kinase (Plk) (Alvarez et al., 2001). Given the growing evidence that the PI3K/Akt pathway transmits important signals downstream of oncogenic tyrosine kinases, we have examined the potential involvement of Foxo proteins in mediating the growthstimulating and antiapoptotic action of FLT3-ITD receptor signaling. Here, we demonstrate that activated FLT3 receptor signaling induces phosphorylation of Foxo3a in Ba/F3 cells. FLT3-ITD receptors suppress Foxo3a-mediated apoptosis and induction of p27Kip1 and Bim gene expression. In addition, constitutive activation of FLT3 signaling triggers nuclear exclusion of Foxo proteins and suppresses their transcriptional activity. Thus, inhibition of Foxo protein function may contribute to the oncogenic transformation of FLT3-ITD receptors in hematopoietic malignancies.

Results Expression of FLT3-ITD4 is sufficient to promote cell survival and proliferation in IL-3-deprived Ba/F3 cells To examine the signaling events that are strictly regulated by activated FLT3 receptors harboring ITD,

we generated Ba/F3 cells expressing a FLT3-ITD receptor under the control of a tetracycline-responsive promoter. To this end, one specific FLT3-ITD variant, termed ITD4 (FYVDFREYDEDFYVDFREY), with a hemagglutinin A (HA)-antigenic tag at the C-terminus (Figure 1a), was introduced by retroviral infection into the TonBaF.1 cell line (TonB) containing the reverseTet transactivator gene (Klucher et al., 1998), generating the polyclonal cell line TonB.FLT3-ITD4. Addition of

Figure 1 Doxycycline-dependent expression of FLT3-ITD4 in Tet-On Ba/F3 (TonB) cell line. (a) Schematic representation of the FLT3-ITD4-HA receptor that harbors one specific ITD (ITD4) in the JM domain and HA tag at the carboxy-terminus. The FLT3 receptor contains five immunoglobulin-like domains in the extracellular domain (ECD), followed by a transmembrane (T) region, a short intracellular JM domain and a split tyrosine kinase (TK1 and TK2) domain. (b) FLT3-ITD4-HA was cloned in pRevTRE-Hyg, and stable transduced in TonB cells to generate the TonB.FLT3ITD4 cell line. TonB.FLT3-ITD4 cells were cultured for 24 h either in the absence of doxycycline (dox), FLT3 ligand (FL) and FLT3 tyrosine kinase inhibitor PKC412 (lane 1), in the presence of 2 mg/ ml doxycycline (lane 2), with 2 mg/ml doxycycline, and stimulated for 10 min with 100 ng/ml FL (lane 3), or in the presence of doxycycline and 10 nM PKC412 (lane 4). Cells were collected and subjected to immunoprecipitation (IP) with HA antibody (Ab) and analysed by immunoblotting (IB) using either FLT3 or phosphotyrosine (p-Tyr) MoAb. (c) TonB.FLT3-ITD4 cells were cultured in medium with doxycycline for 24 or 72 h and cell surface expression of FLT3-ITD4 was analysed by flow cytometry using a PE-conjugated FLT3/CD135 MoAb Oncogene


3340

the tetracycline analog doxycycline induced expression of both the 135 and 160 kDa isoforms of FLT3-ITD4, with virtually undetectable FLT3-ITD4 protein levels in its absence (Figure 1b). Flow-cytometric analysis confirmed cell surface expression of FLT3-ITD receptors in the presence of doxycycline, with more selective outgrowth of higher FLT3-ITD-expressing cells 72 h after the addition of doxycycline (Figure 1c). As expected from previous studies, the constitutive tyrosine kinase activity of FLT3-ITD4 was readily inhibited by the addition of 10 nM PKC412, a potent small-molecule tyrosine kinase inhibitor of both wild-type and mutant FLT3 receptors (Weisberg et al., 2002). TonB cells, like the parental Ba/F3 cell line, require IL-3 both for proliferation as well as to overcome the default apoptotic program. As measured by annexin-V staining, cell viability of the parental TonB cell line decreased progressively following removal of IL-3, both in the presence and absence of doxycycline (Figure 2a). In contrast, apoptosis was significantly reduced in TonB cells expressing FLT3-ITD4, especially after prolonged IL-3 deprivation (Figure 2a). Similar results were obtained with FACS analysis on propidium iodidelabeled cells, where induction of FLT3-ITD4 expression was sufficient to reduce the fraction of apoptotic cells (sub-G1 content) (Figure 2b). Furthermore, cell division was actively stimulated by FLT3-ITD4 receptor signaling with more TonB.FLT3-ITD4 cells in S phase in the presence of doxycycline (Figure 2b), accompanied by an exponential increase in cell numbers (Figure 2c). These results demonstrate that the expression of FLT3-ITD4 receptors is sufficient to rescue IL-3-deprived Ba/F3 cells from programmed cell death and stimulate their cell division. FLT3 receptors with ITD4 mutation activate ERK, STAT5 and Akt signaling pathways We initially analysed the activation of several known signaling targets of FLT3, including ERK (MAP kinase), STAT5 and Akt. FLT3-ITD4 expression could be detected in IL-3-starved TonB.FLT3-ITD4 cells 5 h after the addition of doxycycline, and this was accompanied by phosphorylation and activation of ERK1/2, STAT5 and Akt (Figure 3a). Further increase in FLT3-ITD4 expression levels with prolonged

Figure 2 Induction of FLT3-ITD4 expression is sufficient to promote survival and cell proliferation in IL-3-deprived Ba/F3 cells. (a) IL-3-dependent parental Tet-On Ba/F3 (TonB) and TonB.FLT3-ITD4 cells were grown in the absence of IL-3 with or without 2 mg/ml doxycycline (dox). After 0, 16 and 64 h, cells were harvested and the percentage of Annexin-V positive cells was determined by flow-cytometric analysis. (b) TonB.FLT3-ITD4 cells were cultured in the absence of IL-3 with or without doxycycline. After 16 and 40 h, cells were collected, fixed in ethanol, and labeled with propidium iodide to determine their DNA profiles by flow cytometry. (c) 1 105 TonB.FLT3-ITD4 cells were cultured without IL-3 in the absence or presence of 2 mg/ml doxycycline for 4 days, and each day the viable cell count was determined by trypan blue exclusion Oncogene

doxycycline treatment had no significant additional effect on the level of ERK and STAT5 phosphorylation, but phosphorylation of Akt on Ser473 continued to


3341

increase. PKC412 treatment strongly inhibited the activation of ERK, STAT5 and Akt (Figure 3a). These data confirm previous findings that FLT3 receptors with ITD mutations efficiently trigger the activation of ERK, STAT5 and Akt in the absence of FL stimulation. Phosphorylation of Foxo3a is induced after FLT3-ITD4 receptor expression Akt is thought to mediate critical signals from several oncogenic tyrosine kinases, including BCR/ABL and TEL/platelet-derived growth factor receptor b (Skorski et al., 1997; Dierov et al., 2002). In turn, although Akt has a number of substrates, genetic studies in C. elegans have implicated the Forkhead transcription factors of particular importance for growth and viability downstream of the insulin receptor. Furthermore, negative regulation of Foxo3a has been implicated in cytokinemediated survival signaling downstream of PI3K/Akt pathway (Dijkers et al., 2000b, 2002), and phosphorylation of Foxo3a on residue Thr32 by Akt is critical for interaction with 14-3-3 proteins and its cytoplasmic retention (Brunet et al., 1999). Thus, we examined whether the unscheduled cell proliferation and anti-

apoptotic action of FLT3-ITD4 expression correlated with the status of Foxo3a phosphorylation. Indeed, immunoblot analysis indicated an increased level of Foxo3a phosphorylation on residue Thr32, as shown by the appearance of additional slower-migrating phosphospecies immediately after induction of FLT3-ITD4 expression (Figure 3a). To further verify that FLT3-ITD receptor signaling promotes Foxo3a phosphorylation, Ba/F3 cells were transiently transfected with expression vectors encoding FLT3-ITD4, wild-type FLAG-Foxo3a or triple mutant (TM) FLAG-Foxo3a, where all the three Akt phosphorylation sites (Thr32, Ser253 and Ser315) of Foxo3a have been mutated to Ala residues. Phosphorylation of Thr32 was monitored after immunoprecipitation with FLAG antibody in IL-3-deprived Ba/F3 cells. In the absence of FLT3-ITD expression, FLAG-Foxo3a was not phosphorylated on residue Thr32 (Figure 3b, lane 1). However, coexpression of FLT3-ITD4 induced the phosphorylation of wild-type FLAG-Foxo3a (Figure 3b, lane 3), but not FLAG-Foxo3a-TM (Figure 3b, lane 4). These data show that FLT3-ITD4induced stimulation of the Akt pathway is accompanied by enhanced Foxo3a phosphorylation in Ba/F3 cells deprived from IL-3.

Figure 3 Activation of FLT3 receptor signaling induces Akt and Foxo3a phosphorylation. (a) Distinct signaling pathways are activated by FLT3-ITD receptors. TonB.FLT3-ITD4 cells were deprived of IL-3 for 16 h, and thereafter 2 mg/ml doxycycline (dox) was added to the medium. Cells were collected 0, 5, 8 and 11 h after the addition of dox in the absence or presence of 20 nM PKC412. Parallel polyacrylamide gels loaded with 50 mg of protein extracts were blotted and probed with phospho-ERK, phospho-STAT5, phospho-Serine 473 (pS473)-Akt and phospho-threonine 32 (pT32)-Foxo3a Ab. Blots were stripped and reprobed with anti-ERK, anti-STAT5, anti-Akt and anti-Foxo3a, respectively. Equal protein loading was verified by analysing actin expression. (b) FLT3-ITDmediated Foxo3a phosphorylation requires intact Akt phosphorylation sites. Ba/F3 cells were electroporated with expression vectors encoding FLT3-ITD4, FLAG-tagged wild-type Foxo3a (FLAG-Foxo3a), or FLAG-tagged triple mutant Foxo3a, where each of the three Akt phosphorylation sites (Thr32, Ser253, Ser315) have been mutated to Ala (FLAG-Foxo3a-TM). After 20 h of IL-3 deprivation, cells were harvested and subjected to immunoprecipitation (IP) with FLAG Ab. Whole-cell lysates were probed with FLT3 Ab, while IPs were probed with pT32-Foxo3a and Foxo3a Ab. (c) FL stimulates Akt and Foxo3a phosphorylation in Ba/F3 cells expressing wild-type FLT3 receptors. TonB, TonB.FLT3-WT and TonB.FLT3-ITD4 cell lines were cultured for 20 h in the absence of IL-3 and serum and in the presence of 2 mg/ml doxycycline. Cells were harvested after incubation with or without 200 ng/ml FL for 5 or 10 min, and analysed by Western blot analysis using pS473-Akt, Akt, pT32-Foxo3a and Foxo3a antibodies Oncogene


3342

FL stimulation of wild-type FLT3 receptors promotes Akt and Foxo3a phosphorylation in IL-3- and serum-starved Ba/F3 cells It has been demonstrated that FLT3-ITD but not ligand-bound wild-type FLT3 receptors activate STAT5 signaling (Hayakawa et al., 2000; Mizuki et al., 2000; Spiekermann et al., 2003), and upregulate Bcl-XL expression (Minami et al., 2003), implying qualitative differences between wild-type FLT3 and mutant FLT3ITD signaling. Therefore, we assessed whether stimulation of wild-type FLT3 receptors with FL was able to trigger Akt activation and inhibitory Foxo3a phosphorylation similar to FLT3-ITD receptors. To this end, TonB.FLT3-WT cells were generated that expressed equal levels of FLT3 protein as TonB.FLT3-ITD4 cells (Figure 3c). Only in the absence of exogenous IL-3 and serum, parental TonB and TonB.FLT3-WT cells displayed complete reduction of phospho-Akt and phospho-Foxo3a levels. In the presence of 10% FCS, there was apparently sufficient FL in the medium, to trigger wild-type FLT3 receptor signaling (data not shown). In starved TonB.FLT3-WT cells, the addition of 200 ng/ml FL for 5 or 10 min resulted in similar phospho-Akt and phospho-Foxo3a levels, as observed in TonB.FLT3ITD4 cells in the absence of FL (Figure 3c). Addition of exogenous FL to TonB.FLT3-ITD4 cells did not result in a further increase of Akt or Foxo3a phosphorylation. In conclusion, these findings indicate that stimulation of wild-type FLT3 receptors by FL and mutant FLT3-ITD receptor signaling both induce Akt activation and inhibitory phosphorylation of Foxo3a to the same extent in Ba/F3 cells. ITD mutations in FLT3 receptors regulate nuclear-cytoplasmic shuttling of EGFP-Foxo3a Each of the Foxo protein family members contains three Akt consensus phosphorylation sites (Arg–Xaa–Arg– Xaa–Xaa–Ser/Thr), which play distinct roles in controlling intracellular localization of Forkhead factors (Brownawell et al., 2001; Rena et al., 2001, 2002). To determine the regulation of Foxo protein subcellular localization by constitutively active FLT3 signaling, we analysed the intracellular localization of a fusion protein composed of enhanced green fluorescence protein (EGFP) and Foxo3a (EGFP-Foxo3a), as well as EGFP-Foxo3a-TM, where the Akt phosphorylation sites Thr24, Ser256 and Ser319 of Foxo3a have been mutated to Ala. COS7 cells were transfected with either pEGFP-Foxo3a or pEGFP-Foxo3a-TM and cultured for 24 h in the absence of serum. Under these conditions, both EGFP-Foxo3a and EGFP-Foxo3a-TM were localized in the nucleus (Figure 4a and b). However, coexpression of FLT3-ITD4 resulted in a significant re-localization of EGFP-Foxo3a from the nucleus into the cytoplasm (Figure 4c), which required the presence of intact Akt sites in Foxo3a, as cytoplasmic retention was not observed with EGFP-Foxo3a-TM (Figure 4d). Nuclear exclusion of EGFP-Foxo3a was dependent on PI3K activity, since treatment of transfected COS7 cells Oncogene

Figure 4 Cytoplasmic retention of EGFP-Foxo3a by FLT3-ITD receptor expression. (a–f) COS7 cells were transfected with 2.5 mg of wild-type EGFP-Foxo3a (a, c and e) or mutant EGFP-Foxo3aTM (b, d and f) and 2.5 mg pEBB empty vector DNA (a and b) or pEBB-FLT3-ITD4-HA (c–f). Directly after transfection, cells were serum-starved by culturing them in the absence of FBS. At 22 h post-transfection, cells were treated with DMSO (a–d) or 10 mM LY294002 (e and f) for 5 h. Images were taken on living cells with an inverted microscope

for 5 h with the specific PI3K inhibitor LY294002 abrogated nuclear-cytoplasmic shuttling of EGFP-Foxo3a by FLT3-ITD4 signaling (Figure 4e). Similar results were obtained for EGFP-Foxo1 and EGFP-Foxo1-TM (data not shown). We conclude, therefore, that cytoplasmic retention of Foxo proteins is actively induced by FLT3-ITD4 receptor expression and requires signaling through the PI3K/Akt pathway. Transcription activation by Foxo transcription factors is inhibited by FLT3-ITD4 signaling Foxo proteins show strong transcription activation of a minimal promoter element containing three tandem copies of alternating IRS-A (CAAAACAA) and IRS-B (TTATTTTG) sequences derived from IGFBP-1 promoter (3 IRS). The observation that FLT3-ITD promotes Foxo3a phosphorylation, along with the localization data obtained with EGFP-Foxo3a, suggests that FLT3-ITD signaling could regulate transcription activation by Foxo proteins. To test this hypothesis,


3343

U2-OS cells were transfected with a luciferase reporter plasmid containing 3 IRS promoter element (3 IRSLuc), together with increasing concentrations of expression vectors encoding HA-tagged Foxo3a, FLT3ITD4-HA or HA-tagged Akt containing a myristolated amino-terminus (HA-Myr-Akt). Under serum-starved conditions, Foxo3a transfection resulted in a dose-dependent increase in transcription of the 3 IRS-Luc reporter (Figure 5a), whereas no enhancement was observed with a reporter plasmid lacking IRS elements (data not shown). Importantly, the transcriptional activity of Foxo3a was strongly reduced after cotransfection of FLT3-ITD4 (Figure 5a), indicating that FLT3-ITD receptor signaling significantly inhibits the ability of Foxo3a to act as a transcriptional activator. In the presence of constitutive active Akt through transfection of HA-Myr-Akt, Foxo3a-mediated transcription activation was almost completely abolished, as has been reported before (Brunet et al., 1999; Dijkers et al., 2000b). Similarly, transcriptional activity of Foxo1 (data not shown) and Foxo4 (Figure 5b) were diminished by co-expression of FLT3-ITD4. These results demonstrate that constitutively active FLT3 signaling through acquired ITD mutations negatively regulates Foxo protein function by prohibiting their ability to activate transcriptional target genes. Downregulation of Foxo target genes p27Kip1 and Bim upon induction of FLT3-ITD4 expression Cytokine-mediated proliferation and survival has been shown to correlate with phosphorylation of Foxo3a and suppression of p27Kip1 and Bim gene expression (Dijkers et al., 2000a, b; Stahl et al., 2002). High p27Kip1 levels are linked to cell cycle arrest in G0/G1 through interaction with CDK-cyclin complexes (Toyoshima and Hunter, 1994), while the proapoptotic Bcl-2 family member Bim acts as an important death activator in hematopoietic cells (Bouillet et al., 1999; Shinjyo et al., 2001; Villunger et al., 2003). Thus, we investigated whether activation of FLT3-ITD receptor signaling may regulate p27Kip1 and Bim expression in cytokine-starved Ba/F3 cells. Multiple isoforms of Bim have been identified (O’Connor et al., 1998; U et al., 2001), but in IL-3-deprived TonB.FLT3ITD4 cells we detected predominantly BimEL and BimL expression by immunoblot analysis (Figure 6a). Doxycycline treatment resulted in a progressive induction of FLT3-ITD4 expression and simultaneous downregulation of p27Kip1 and BimEL protein levels (Figure 6a). Importantly, this decrease in protein expression was preceded by repression of p27Kip1 2.2 kb and Bim 5.7 kb mRNA transcripts (Figure 6b), arguing that FLT3-ITD signaling contributes to transcriptional inhibition of p27Kip1 and Bim gene expression. Foxo3a-mediated induction of p27Kip1 and Bim expression is abrogated by FLT3 receptors with internal tandem duplications

Figure 5 FLT3-ITD receptor signaling regulates transcriptional activity of Foxo proteins. (a and b) U2-OS cells were transfected with 2 mg 3 IRS-Luciferase reporter plasmid and increasing concentrations of pSG5-HA-Foxo3a (a) or pMT2-HA-Foxo4 (b) (0.4, 1 or 2 mg of plasmid DNA) in the absence or presence of either 2 mg pEBB-FLT3-ITD4-HA or 0.5 mg pcDNA3.1-HA-Myr-Akt. Immediately after transfection, cells were cultured in 0.1% FCS and 36 h later luciferase activity was analysed. Relative luciferase activity indicates luciferase values corrected for transfection efficiency, and is representative of two independent experiments. Five percent of the lysates used to measure luciferase activity were loaded on 8% polyacrylamide gel, blotted and probed with HA MoAb

In addition to Foxo proteins, c-Myc (Yang et al., 2001) and BRCA1 (Williamson et al., 2002) have been Oncogene


3344

Foxo3a-mediated transcriptional regulation of p27Kip1 and Bim. To this end, doxycycline-inducible Ba/F3 cells were generated, which expressed FLAG-Foxo3a (TonB.Foxo3a). These cells were subsequently infected with empty pMSCV-IRES-GFP retrovirus (TonB.Foxo3a/control cells) or pMSCV-IRES-GFP encoding FLT3-ITD4-HA (TonB.Foxo3a/FLT3-ITD4 cells). Treatment of IL-3-deprived TonB.Foxo3a/control cells for 24 h with doxycycline resulted in a significant induction of Foxo3a expression, with moderate upregulation of p27Kip1 and BimEL protein levels (Figure 7a), as well as p27Kip1 and Bim mRNA (Figure 7b). However, in the presence of FLT3-ITD4, induction of p27Kip1 and Bim transcription by Foxo3a was significantly diminished (Figure 7c), even though Foxo3a protein levels were more abundant in the presence of FLT3-ITD4 expression. Downregulation of p27Kip1 and BimEL protein levels correlated with abrogation of Foxo3amediated apoptosis by FLT3-ITD4 receptor signaling (Figure 7d). In summary, these data indicate that, in hematopoietic cells, FLT3-ITD receptors inhibit Foxo3a-dependent transcriptional activation of p27Kip1 and Bim, which in turn are likely to play a major role in regulating proliferation and cell viability.

Discussion

Figure 6 Induction of FLT3-ITD receptor signaling results in downregulation of p27Kip1 and Bim expression levels. (a) TonB.FLT3-ITD4 cells were cultured for 16 h in the absence of IL-3 and subsequently treated for increasing time periods with 2 mg/ ml doxycycline. Cells were harvested at the indicated time points and 2 107 cells were lysed for protein analysis. Parallel polyacrylamide gels were loaded with 50 mg protein extracts and probed with FLT3, pT32-Foxo3a, p27Kip1, Bim and actin antibodies. (b) RNA was extracted from 6 107 TonB.FLT3-ITD4 cells collected at the same time points as described in (a), and subjected to Northern blot analysis using murine p27Kip1, Bim and b-actin cDNA probes

implicated in controlling p27Kip1 gene expression, while Bim transcription is also regulated through the JNK and Raf/MEK/ERK pathway (Harris and Johnson, 2001; Shinjyo et al., 2001; Whitfield et al., 2001). Therefore, we asked if FLT3-ITD signaling specifically altered Oncogene

Activating mutations in the tyrosine kinase receptor FLT3 have been detected in about one-third of patients with AML (Gilliland and Griffin, 2002), but the mechanisms of oncogenic transformation by these constitutively active FLT3 receptors have not yet been examined in detail. In this study, we have analysed the biological consequences of FLT3-ITD4 expression employing a tetracycline-inducible promoter in the IL3-dependent pre-B cell line Ba/F3 (TonB cells). Induction of FLT3-ITD4 expression is sufficient to inhibit apoptosis and promote cell cycle entry of IL-3-deprived TonB.FLT3-ITD4 cells, and leads to the activation of ERK, STAT5 and Akt. It is well established that Akt kinase activity is stimulated by growth factor-induced PI3K activation (Franke et al., 1995; Datta et al., 1996) and activation of PI3K/Akt pathway is essential for transformation by oncogenic tyrosine kinases BCR/ ABL (Skorski et al., 1997) and TEL/platelet-derived growth factor receptor b (Dierov et al., 2002). In various systems, antiapoptotic signaling by Akt has been shown to occur both through inhibitory phosphorylation of Bad (del Peso et al., 1997), caspase-9 (Cardone et al., 1998), or Foxo subfamily of Forkhead transcription factors (Brunet et al., 1999; Kops et al., 1999; Rena et al., 1999; Tang et al., 1999). Here, we provide evidence that Foxo transcription factors are likely to play important roles as mediators of cell proliferation and viability as part of the PI3K/Akt pathway downstream of activated FLT3 receptor signaling. The Foxo subclass of Forkhead transcription factors represents the vertebrate orthologues of C. elegans DAF-16, which include Foxo1, Foxo3a and Foxo4


3345

Figure 7 FLT3-ITD expression inhibits Foxo3a-mediated activation of p27Kip1 and Bim gene expression and apoptosis. (a) Doxycycline-inducible Ba/F3 cells expressing FLAG-Foxo3a (TonB.Foxo3a) were generated and infected with either empty or FLT3ITD4-HA-encoding pMSCV-IRES-GFP retroviral vector. After GFP sorting, stable TonB.Foxo3a/control and TonB.Foxo3a/FLT3ITD4 cells were obtained. TonB.Foxo3a/control and TonB.Foxo3a/FLT3-ITD4 cells were cultured for 24 h without IL-3 in the presence or absence of 2 mg/ml doxycycline (dox). Cells were collected, protein lysates of 2 107 cells were obtained and subjected to Western blot analysis using FLT3, Foxo3a, p27Kip1, Bim and actin antibodies. (b) Identical samples as in (a) were harvested to extract RNA and analysed by Northern blot analysis using Foxo3a, p27Kip1, Bim and b-actin cDNA probes. (c) Densitometric quantification of results as depicted in (b), where intensities were obtained with radiolabeled cDNA probes, were corrected for b-actin signal. (d) TonB.Foxo3a (TonB.F3) and FLT3-ITD4/TonB.Foxo3a cells (FLT3-ITD4 TonB.F3) were cultured without IL-3 and in the absence or presence of 2 mg/ml doxycycline for either 16 or 40 h. Thereafter, the cells were collected, fixed in ethanol, and labeled with propidium iodide to determine their DNA profiles by flow cytometry

(Kaestner et al., 2000). PI3K- and PDK1-dependent activation of Akt induces phosphorylation of Foxo proteins on conserved serine and threonine residues, which inhibits their ability to act as transcriptional regulators. Recent studies have shown that in hematopoietic cells phosphorylation of Foxo3a is a downstream event of the PI3K/Akt pathway in IL-2 (Stahl et al., 2002), IL-3 (Dijkers et al., 2000b), Kit ligand (Engstrom et al., 2003), erythropoietin and thrombopoietin (TPO) signaling (Kashii et al., 2000; Uddin et al., 2000; Tanaka et al., 2001), where the activity of this transcription factor has been linked to the induction of apoptosis.

Our data demonstrate that both ligand-bound wildtype FLT3 and FLT3-ITD signaling promotes Akt activation as well as Thr32-Foxo3a phosphorylation in cytokine- and serum-starved Ba/F3 cells, which relies on the presence of the consensus Akt phosphorylation sites in Foxo3a. We observed a basal level of Thr32-Foxo3a phosphorylation in IL-3-starved TonB.FLT3-ITD4 cells, but not in IL3-deprived Ba/F3 cells. This finding suggests that the very low level of FLT3-ITD4 expression that is present in the absence of doxycycline treatment may be sufficient to induce detectable Foxo3a phosphorylation but not detectable Ser473-Akt Oncogene


3346

phosphorylation. This differential sensitivity for Foxo3a phosphorylation has also been noted in TPO signaling, where saturated Thr32-Foxo3a phosphorylation levels are achieved with at least 10-fold lower concentration of TPO than Ser473-Akt phosphorylation (Tanaka et al., 2001). The nonphosphorylated forms of Foxo proteins localize to the nucleus, but Akt-mediated phosphorylation promotes their interaction with 14-3-3 proteins and exclusion from the cell nucleus, prohibiting their action as transcription factors (Brunet et al., 1999; Brownawell et al., 2001; Rena et al., 2001). Our data indicate that expression of FLT3-ITD receptors in serum-deprived COS7 cells triggers cytoplasmic retention of EGFPFoxo1 and EGFP-Foxo3a, which requires PI3K activity and intact Akt phosphorylation sites of Foxo proteins. Furthermore, FLT3-ITD signaling in U2-OS cells strongly suppresses transcriptional activation of Foxo transcription factors in a reporter assay. Moreover, FLT3-ITD diminishes Foxo3a-induced apoptosis in Ba/ F3 cells. Thus, through its ability to induce inhibitory phosphorylation of Forkhead transcription factors and their exclusion from cell nucleus, FLT3-ITD signaling rescues the hematopoietic cells from Foxo3a-induced apoptosis. At present, we have not firmly established whether Akt is the kinase that phosphorylates and negatively regulates Forkhead transcription factors after activation of FLT3-ITD signaling. In insulin receptor-deficient hepatocytes, residues Thr24 and Ser319 of Foxo1 are not phosphorylated by constitutively activated Akt (Nakae et al., 2000, 2001), and dominant-negative Akt fails to inhibit Thr32 of Foxo3a in HEK293 (Brunet et al., 2001), and only partially prohibits Foxo4 phosphorylation after insulin stimulation in A14 cells (Kops et al., 1999). Indeed, other members of the AGC kinase family, which, besides Akt, include the serum- and glucocorticoid-inducible kinase 1 (SGK1) and related kinase SGK3/CISK, are able to directly phosphorylate Foxo3a (Liu et al., 2000; Brunet et al., 2001). Moreover, AGC kinase-independent signaling pathways have been implicated in the regulation of Foxo subclass of Forkhead transcription factors. Phosphorylation of Foxo1 on Ser329 by dual-specificity tyrosinephosphorylated regulated kinase 1A (DYRK1A) regulates basal nuclear localization and transactivation (Woods et al., 2001). Furthermore, Ser322 and Ser325 of Foxo1 are substrates for casein kinase 1 (CK1), after Akt-catalysed phosphorylation of Ser319, regulating the nuclear exclusion of Foxo1 (Rena et al., 2002). In addition, the Ras/Ral pathway promotes Foxo4 phosphorylation on Thr447 and Thr451, which inhibits Foxo4 transcription activation (Kops et al., 1999; De Ruiter et al., 2001). Thus, although activation of Akt by FLT3ITD correlates closely with phosphorylation of Foxo proteins, future analysis has to provide more insight as to which intermediate signaling molecules play a critical role in the regulation of Foxo protein function upon FLT3 receptor activation during normal hematopoiesis and in myeloid leukemia. Oncogene

To further elucidate the mechanism by which FLT3ITD signaling inhibits apoptosis in hematopoietic cells, we investigated the potential involvement of the previously described Foxo target gene Bim (Dijkers et al., 2000a). Bim encodes an BH3-only pro-apoptotic Bcl-2 family member, which acts as an allosteric inhibitor of the antiapoptotic proteins Bcl-2 and BclxL. Activation of Bim does not require caspase cleavage, as observed for Bid (Li et al., 1998), and Bim is not regulated by phosphorylation, like Bad (del Peso et al., 1997; U et al., 2001). Instead, induction of Bim activity occurs mainly at the transcriptional level and Bim mRNA gives rise to multiple alternatively spliced isoforms (O’Connor et al., 1998). We found that both BimEL protein and Bim mRNA levels decrease after induction of FLT3-ITD receptor signaling in IL-3starved TonB.FLT3-ITD4 cells. Moreover, FLT3-ITD4 prohibits Foxo3a-mediated induction of Bim gene expression. The fact that Bim-deficient bone marrow HSC and granulocytes are relatively resistant to apoptosis after cytokine withdrawal or PI3K inhibition (Dijkers et al., 2002) suggests that inhibition of Bim expression is an important downstream event of FLT3ITD/Akt signaling pathway in hematopoietic cells. In addition, we analysed whether FLT3-ITD controls expression of the important regulator of cell cycle progression p27Kip1, which can be transcriptionally induced by Foxo proteins (Medema et al., 2000). The CDK-inhibitor p27Kip1 mainly acts as a negative regulator of cell division through its ability to inhibit cyclin E-associated CDK2, thereby preventing pRb phosphorylation at the G1/S phase of the cell cycle (Toyoshima and Hunter, 1994). Furthermore, high p27Kip1 levels sensitize Ba/F3 cells to programmed cell death, while p27Kip1-deficient HSC display increased survival upon cytokine-starvation (Dijkers et al., 2000b; Parada et al., 2001). We found that FLT3-ITD4 downregulates p27Kip1 and the presented data support the notion that FLT3-ITD signaling mainly reduces p27Kip1 protein levels through repression of transcription, although potential alternative mechanisms, including proteosome-dependent degradation of p27Kip1, have not been excluded. Specifically, FLT3-ITD4 could prevent Foxo3a-induced upregulation of p27Kip1 in Ba/ F3 cells. Thus, inhibition of p27Kip1 expression presents an important way for FLT3-ITD to induce mitogenic signaling and provide protection against apoptosis as well. In conclusion, the studies reported here identify for the first time downstream targets of the Akt pathway that are regulated by oncogenic FLT3 receptors. FLT3ITD signaling promotes inhibitory phosphorylation of Foxo Forkhead transcription factors and repression of p27Kip1 and Bim gene expression, thereby bypassing G1 cell cycle checkpoint and rescuing hematopoietic cells from default programmed cell death. Similarly, BCR/ ABL regulates phosphorylation of Foxo3a in chronic myeloid leukemia (CML) cell lines and Foxo3a regulates p27Kip1 expression downstream of BCR/ABL-signaling in CML cells (Komatsu et al., 2003). Thus, an emerging theme is that constitutively activated tyrosine kinases in


3347

leukemia signal through Foxo proteins and these transcription factors may play a key role in transformation of myeloid cells.

Materials and methods Antibodies and reagents FLT3 (C-20), phosphotyrosine (PY20) and actin (C-11) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Akt, ERK1/2 antibodies and phospho-antibodies for Thr202/Tyr204-ERK1/2 (E10), Ser473Akt and Tyr694-STAT5 were obtained from Cell Signaling Technology (Beverley, MA, USA). Foxo3a and phosphoThr32-Foxo3a were purchased from Upstate (Lake Placid, NY, USA), Bim antibodies from StressGen (Victoria, Canada) and HA (HA.11) antibodies from Covance (Richmond, CA, USA). STAT5 and p27Kip1 antibodies were obtained from BD Transduction Laboratories (San Diego, CA, USA). FLAG (M2) antibodies, doxycycline, LY294002, PMSF and polybrene were purchased from Sigma-Aldrich (St Louis, MO, USA). Human recombinant FLT3 ligand (FL) was obtained from PeproTech (Rocky Hill, NJ, USA). PKC412 (N-benzoyl staurosporine) was kindly provided by Novartis Pharma AG (Basel, Switzerland). DNA constructs FLT3-ITD4 and wild-type FLT3 cDNA were a kind gift from Dr Tomoki Naoe (Nagoya University School of Medicine, Nagoya, Japan). The expression vector pEBB-FLT3-ITD4HA was created by exchanging the amino-terminal signal sequence of FLT3 (1–27 aa) with the signal sequence of c-FMS (1–22 aa), providing FLT3-ITD4 with a carboxy-terminal hemagglutin A (HA) tag at 30 end of the open reading frame by PCR and cloning the assembled cDNA fragment into BamHI–NotI restriction sites of pEBB. pMSCV-FLT3-ITD4 was generated by cloning FLT3-ITD4-HA BamHI–NotI/blunt in restriction sites BglII–HpaI of pMSCV-IRES-GFP (pMIG). Plasmids pcDNA3.1-FLAG-Foxo3a, pcDNA3.1-FLAGFoxo3a-TM, pcDNA3.1-HA-Myr-Akt, pSG5-HA-Foxo3a, pMT2-HA-Foxo4 and pGL2-3 IRS-Luc were generously provided by Dr William Sellers (Harvard Medical School, Boston, USA). pEGFP-Foxo3a and pEGFP-Foxo3a-TM were obtained by cloning Foxo3a and Foxo3a-TM cDNAs in pEGFP-C3 (BD Biosciences Clontech, Palo Alto, CA, USA). H2B-GFP was a kind gift from Dr Reuven Agami and CMVRenilla from Dr Rene´ Bernards (Netherlands Cancer Institute, Amsterdam, Netherlands). Generation doxycycline-inducible cell lines and transient transfection by electroporation FLT3-ITD4-HA, FLT3-WT-HA and FLAG-Foxo3a were cloned in pRevTRE-Hyg, and 10 mg of each retroviral vector was transfected with Superfect (Qiagen) in 5 106 phoenix cells that were plated on a 10-cm dish and cultured in DMEM supplemented with 10% (v/v) fetal bovine serum (FBS), 50 mg streptomycin/ml and 50 U penicillin/ml. The viruscontaining medium was collected after 24 and 48 h and added to 2 106 TonBaF.1 (TonB) cells (Klucher et al., 1998), together with 10% (v/v) conditioned medium from WEHI-3B cells as a source of interleukin-3 (IL-3) and 8 mg/ml polybrene. After 24 h, virus-infected TonB cells were resuspended in RPMI 1640 medium with 10% FBS, 10% WEHI-3B, 50 mg streptomycin/ml and 50 U penicillin/ml (RPMI com-

plete), and selected for 7 days in the presence of 700 mg/ml hygromycin. Signal transduction studies were performed with polyclonal TonB.FLT3-ITD4 cells by growing them for 3–5 days in 75 cm2 flasks in RPMI complete. Cells were washed once with PBS and resuspended in 10% FBS/RPMI (in the absence of IL-3). After 16 h cytokine starvation, 2 107 cells were harvested for the first time point (0 h) and 2 mg/ml doxycycline was added to the remaining TonB.FLT3-ITD4 cells for subsequent time points. For transient transfections, Ba/F3 cells were electroporated (0.29 kV; capacitance 960 mF) with a total amount of 25 mg of the expression vectors pEBB or pEBB-FLT3-ITD4-HA and 25 mg of pcDNA3.1-FLAGFoxo3a or pcDNA3.1-FLAG-Foxo3a-TM, together with 500 ng H2B-GFP to monitor the transfection efficiency. Western blotting and immunoprecipitations Cell lysates were prepared in ice-cold lysis buffer (200 mM NaCl, 20 mM Tris.Cl. pH 8.0, 1% NP-40, 100 mM NaF, 5 mM EDTA and 1 mM Na3VO4) supplemented with Completet protease inhibitors (Roche Applied Science, Indianapolis, IN, US) and 1mM PMSF. Protein concentration was determined with Bradford and equal amounts of protein extracts were loaded on SDS-polyacrylamide gel. For immunoprecipitations, 1 mg of protein extract was incubated with 5 mg of antibody and 40 ml protein A-sepharose beads for 16 h at 41C. Beads were washed four times with lysis buffer and subjected to SDS–PAGE. After SDS–PAGE, proteins were transferred in blotting buffer for 2 h to immobilon-P membranes (Millipore, Billerica, MA, USA). Blots were blocked for 1 h at room temperature in 4% nonfat dry milk in Tris-buffered salineTween 20 (TBST: 0.15 M NaCl, 0.01 M Tris-HCl pH 7.4, 0.05% Tween 20), and incubated overnight at 41C with primary antibodies diluted 1 : 2000 in 1% non-fat dry milk/TBST. After washing, blots were incubated with 1 : 5000 dilution of horseradish peroxidase-conjugated secondary antibody (Amersham Biosciences, Piscataway, NJ, USA) in 1% nonfat dry milk/TBST for 45 min, followed by four times 10 min washes with TBST. Enhanced chemiluminescence was performed according to the manufacturer’s instructions (Perkin-Elmer Life Sciences, Boston, MA, USA). Flow-cytometric analyses To determine FLT3 cell surface expression, 2 106 cells were harvested, washed with PBS and incubated on ice with 100 ml CD135-PE solution (Immunotech). Cells were washed, resuspended in PBS and analysed on a FACSscan using CellQuest software (BD Biosciences). For annexin-V staining, cells were washed with PBS and incubated for 15 min in 100 ml fluorescein isothiocyanate (FITC)-conjugated annexin-V incubation buffer (Roche Applied Science) containing propidium iodide. Cells were washed, resuspended in 1 ml binding buffer and immediately analysed by flow cytometry. For cell cycle analysis, the harvested cells were washed with PBS and fixed in 70% ethanol for at least 3 h on ice. Cells were centrifuged for 5 min at 480 g, treated with 200 mg/ml RNAase A at 371C for 40 min, washed with PBS, resuspended in 20 mg/ml propidium iodide in PBS, incubated for 10 min in the dark and analysed by flow cytometry using ModFit software. Luciferase reporter assays Transfections for reporter assays were carried out in U2-OS cells plated on 6-cm tissue culture dishes using Superfect (Qiagen). In each transfection, 20 ng cytomegalovirus (CMV)driven Renilla luciferase plasmid was included as an internal Oncogene


3348 standard. Immediately after transfection, cells were cultured in 0.1% FCS/DMEM medium and analysed 36 h later. Cells were lysed on ice for 20 min in 1 reporter lysis buffer (Promega) supplemented with Completet protease inhibitors (Roche Applied Science). Cleared lysates were used for quantification of luciferase activities with a dual-luciferase reporter assay system in accordance with the manufacturer’s instructions (Promega). Analyses of mRNA expression Total mRNA was extracted from frozen cell pellets with RNA Trizol, as specified by the manufacturer (Invitrogen), and 15 mg RNA was subjected to paraformaldehyde-containing agarose gel electrophoresis and transferred to Hybond-N þ nylon

membrane (Amersham). Membranes were successively hybridized in Quick Hyb (Stratagene) at 681C with 32P-dCTP radiolabeled p27Kip1, Bim and b-actin mouse cDNA probes generated by random primed labeling. Filters were exposed to Kodak Biomax MS films and densitometric analysis was performed with Kodak Digital Science 1D Image Analysis Software. Acknowledgements We wish to thank Tomoki Naoe, William Sellers and George Daley for reagents. This work was supported by a Leukemia and Lymphoma Society SCOR grant and NIH grant R01 CA66996. BS is the recipient of a Fellowship of the Dutch Cancer Society (KWF/NKB).

References Abu-Duhier FM, Goodeve AC, Wilson GA, Care RS, Peake IR and Reilly JT. (2001). Br. J. Haematol., 113, 983–988. Alvarez B, Martinez AC, Burgering BM and Carrera AC. (2001). Nature, 413, 744–747. Armstrong SA, Staunton JE, Silverman LB, Pieters R, den Boer ML, Minden MD, Sallan SE, Lander ES, Golub TR and Korsmeyer SJ. (2002). Nat. Genet., 30, 41–47. Bouillet P, Metcalf D, Huang DC, Tarlinton DM, Kay TW, Kontgen F, Adams JM and Strasser A. (1999). Science, 286, 1735–1738. Brownawell AM, Kops GJ, Macara IG and Burgering BM. (2001). Mol. Cell. Biol., 21, 3534–3546. Brunet A, Bonni A, Zigmond MJ, Lin MZ, Juo P, Hu LS, Anderson MJ, Arden KC, Blenis J and Greenberg ME. (1999). Cell, 96, 857–868. Brunet A, Park J, Tran H, Hu LS, Hemmings BA and Greenberg ME. (2001). Mol. Cell. Biol., 21, 952–965. Cardone MH, Roy N, Stennicke HR, Salvesen GS, Franke TF, Stanbridge E, Frisch S and Reed JC. (1998). Science, 282, 1318–1321. Carow CE, Levenstein M, Kaufmann SH, Chen J, Amin S, Rockwell P, Witte L, Borowitz MJ, Civin CI and Small D. (1996). Blood, 87, 1089–1096. Castrillon DH, Miao L, Kollipara R, Horner JW and DePinho RA. (2003). Science, 301, 215–218. DaSilva N, Hu ZB, Ma W, Rosnet O, Birnbaum D and Drexler HG. (1994). Leukemia, 8, 885–888. Datta K, Bellacosa A, Chan TO and Tsichlis PN. (1996). J. Biol. Chem., 271, 30835–30839. Datta SR, Brunet A and Greenberg ME. (1999). Genes Dev., 13, 2905–2927. De Ruiter ND, Burgering BM and Bos JL. (2001). Mol. Cell. Biol., 21, 8225–8235. del Peso L, Gonzalez-Garcia M, Page C, Herrera R and Nunez G. (1997). Science, 278, 687–689. Dierov J, Xu Q, Dierova R and Carroll M. (2002). Blood, 99, 1758–1765. Dijkers PF, Birkenkamp KU, Lam EW, Thomas NS, Lammers JW, Koenderman L and Coffer PJ. (2002). J. Cell. Biol., 156, 531–542. Dijkers PF, Medema RH, Lammers JW, Koenderman L and Coffer PJ. (2000a). Curr. Biol., 10, 1201–1204. Dijkers PF, Medema RH, Pals C, Banerji L, Thomas NS, Lam EW, Burgering BM, Raaijmakers JA, Lammers JW, Koenderman L and Coffer PJ. (2000b). Mol. Cell. Biol., 20, 9138–9148. Engstrom M, Karlsson R and Jonsson JI. (2003). Exp. Hematol., 31, 316–323. Oncogene

Franke TF, Yang SI, Chan TO, Datta K, Kazlauskas A, Morrison DK, Kaplan DR and Tsichlis PN. (1995). Cell, 81, 727–736. Gilliland DG and Griffin JD. (2002). Blood, 100, 1532–1542. Harris CA and Johnson Jr EM. (2001). J. Biol. Chem., 276, 37754–37760. Hayakawa F, Towatari M, Kiyoi H, Tanimoto M, Kitamura T, Saito H and Naoe T. (2000). Oncogene, 19, 624–631. Kaestner KH, Knochel W and Martinez DE. (2000). Genes Dev., 14, 142–146. Kashii Y, Uchida M, Kirito K, Tanaka M, Nishijima K, Toshima M, Ando T, Koizumi K, Endoh T, Sawada K, Momoi M, Miura Y, Ozawa K and Komatsu N. (2000). Blood, 96, 941–949. Kelly LM, Liu Q, Kutok JL, Williams IR, Boulton CL and Gilliland DG. (2002). Blood, 99, 310–318. Klucher KM, Lopez DV and Daley GQ. (1998). Blood, 91, 3927–3934. Komatsu N, Watanabe T, Uchida M, Mori M, Kirito K, Kikuchi S, Liu Q, Tauchi T, Miyazawa K, Endo H, Nagai T and Ozawa K. (2003). J. Biol. Chem., 278, 6411–6419. Kops GJ, de Ruiter ND, De Vries-Smits AM, Powell DR, Bos JL and Burgering BM. (1999). Nature, 398, 630–634. Kops GJ, Medema RH, Glassford J, Essers MA, Dijkers PF, Coffer PJ, Lam EW and Burgering BM. (2002). Mol. Cell. Biol., 22, 2025–2036. Lawlor MA and Alessi DR. (2001). J. Cell. Sci., 114, 2903–2910. Levis M, Allebach J, Tse KF, Zheng R, Baldwin BR, Smith BD, Jones-Bolin S, Ruggeri B, Dionne C and Small D. (2002). Blood, 99, 3885–3891. Li H, Zhu H, Xu CJ and Yuan J. (1998). Cell, 94, 491–501. Liu D, Yang X and Songyang Z. (2000). Curr. Biol., 10, 1233–1236. Lyman SD and Jacobsen SE. (1998). Blood, 91, 1101–1134. Medema RH, Kops GJ, Bos JL and Burgering BM. (2000). Nature, 404, 782–787. Meierhoff G, Dehmel U, Gruss HJ, Rosnet O, Birnbaum D, Quentmeier H, Dirks W and Drexler HG. (1995). Leukemia, 9, 1368–1372. Minami Y, Yamamoto K, Kiyoi H, Ueda R, Saito H and Naoe T. (2003). Blood, 102, 2969–2975. Mizuki M, Fenski R, Halfter H, Matsumura I, Schmidt R, Muller C, Gruning W, Kratz-Albers K, Serve S, Steur C, Buchner T, Kienast J, Kanakura Y, Berdel WE and Serve H. (2000). Blood, 96, 3907–3914. Modur V, Nagarajan R, Evers BM and Milbrandt J. (2002). J. Biol. Chem., 277, 47928–47937. Nakae J, Barr V and Accili D. (2000). EMBO J., 19, 989–996.


3349 Nakae J, Biggs III WH, Kitamura T, Cavenee WK, Wright CV, Arden KC and Accili D. (2002). Nat. Genet., 32, 245–253. Nakae J, Kitamura T, Kitamura Y, Biggs WH, Arden KC and Accili D. (2003). Dev. Cell, 4, 119–129. Nakae J, Kitamura T, Ogawa W, Kasuga M and Accili D. (2001). Biochemistry, 40, 11768–11776. Nakao M, Yokota S, Iwai T, Kaneko H, Horiike S, Kashima K, Sonoda Y, Fujimoto T and Misawa S. (1996). Leukemia, 10, 1911–1918. O’Connor L, Strasser A, O’Reilly LA, Hausmann G, Adams JM, Cory S and Huang DC. (1998). EMBO J., 17, 384–395. Ogg S, Paradis S, Gottlieb S, Patterson GI, Lee L, Tissenbaum HA and Ruvkun G. (1997). Nature, 389, 994–999. Parada Y, Banerji L, Glassford J, Lea NC, Collado M, Rivas C, Lewis JL, Gordon MY, Thomas NS and Lam EW. (2001). J. Biol. Chem., 276, 23572–23580. Paradis S and Ruvkun G. (1998). Genes Dev., 12, 2488–2498. Ramaswamy S, Nakamura N, Sansal I, Bergeron L and Sellers WR. (2002). Cancer Cell., 2, 81–91. Rena G, Guo S, Cichy SC, Unterman TG and Cohen P. (1999). J. Biol. Chem., 274, 17179–17183. Rena G, Prescott AR, Guo S, Cohen P and Unterman TG. (2001). Biochem. J., 354, 605–612. Rena G, Woods YL, Prescott AR, Peggie M, Unterman TG, Williams MR and Cohen P. (2002). EMBO J., 21, 2263–2271. Rokudai S, Fujita N, Kitahara O, Nakamura Y and Tsuruo T. (2002). Mol. Cell. Biol., 22, 8695–8708. Rosnet O, Marchetto S, deLapeyriere O and Birnbaum D. (1991). Oncogene, 6, 1641–1650. Rosnet O, Schiff C, Pebusque MJ, Marchetto S, Tonnelle C, Toiron Y, Birg F and Birnbaum D. (1993). Blood, 82, 1110–1119. Scheid MP, Marignani PA and Woodgett JR. (2002). Mol. Cell. Biol., 22, 6247–6260. Scheijen B and Griffin JD. (2002). Oncogene, 21, 3314–3333. Schmidt M, de Mattos SF, van der Horst A, Klompmaker R, Kops GJ, Lam EW, Burgering BM and Medema RH. (2002). Mol. Cell. Biol., 22, 7842–7852. 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. Skorski T, Bellacosa A, Nieborowska-Skorska M, Majewski M, Martinez R, Choi JK, Trotta R, Wlodarski P, Perrotti D, Chan TO, Wasik MA, Tsichlis PN and Calabretta B. (1997). EMBO J., 16, 6151–6161. Spiekermann K, Bagrintseva K, Schoch C, Haferlach T, Hiddemann W and Schnittger S. (2002). Blood, 100, 3423–3425.

Spiekermann K, Bagrintseva K, Schwab R, Schmieja K and Hiddemann W. (2003). Clin. Cancer Res., 9, 2140–2150. Stahl M, Dijkers PF, Kops GJ, Lens SM, Coffer PJ, Burgering BM and Medema RH. (2002). J. Immunol., 168, 5024–5031. Tanaka M, Kirito K, Kashii Y, Uchida M, Watanabe T, Endo H, Endoh T, Sawada K, Ozawa K and Komatsu N. (2001). J. Biol. Chem., 276, 15082–15089. Tang ED, Nunez G, Barr FG and Guan KL. (1999). J. Biol. Chem., 274, 16741–16746. Thiede C, Steudel C, Mohr B, Schaich M, Schakel U, Platzbecker U, Wermke M, Bornhauser M, Ritter M, Neubauer A, Ehninger G and Illmer T. (2002). Blood, 99, 4326–4335. Toyoshima H and Hunter T. (1994). Cell, 78, 67–74. U M, Miyashita T, Shikama Y, Tadokoro K and Yamada M. (2001). FEBS Lett., 509, 135–141. Uddin S, Kottegoda S, Stigger D, Platanias LC and Wickrema A. (2000). Biochem. Biophys. Res. Commun., 275, 16–19. Villunger A, Scott C, Bouillet P and Strasser A. (2003). Blood, 101, 2393–2400. Weisberg E, Boulton C, Kelly LM, Manley P, Fabbro D, Meyer T, Gilliland DG and Griffin JD. (2002). Cancer Cell, 1, 433–443. Whitfield J, Neame SJ, Paquet L, Bernard O and Ham J. (2001). Neuron, 29, 629–643. Williamson EA, Dadmanesh F and Koeffler HP. (2002). Oncogene, 21, 3199–3206. Woods YL, Rena G, Morrice N, Barthel A, Becker W, Guo S, Unterman TG and Cohen P. (2001). Biochem. J., 355, 597–607. Yamamoto Y, Kiyoi H, Nakano Y, Suzuki R, Kodera Y, Miyawaki S, Asou N, Kuriyama K, Yagasaki F, Shimazaki C, Akiyama H, Saito K, Nishimura M, Motoji T, Shinagawa K, Takeshita A, Saito H, Ueda R, Ohno R and Naoe T. (2001). Blood, 97, 2434–2439. Yang W, Shen J, Wu M, Arsura M, FitzGerald M, Suldan Z, Kim DW, Hofmann CS, Pianetti S, Romieu-Mourez R, Freedman LP and Sonenshein GE. (2001). Oncogene, 20, 1688–1702. Zhang S, Fukuda S, Lee Y, Hangoc G, Cooper S, Spolski R, Leonard WJ and Broxmeyer HE. (2000). J. Exp. Med., 192, 719–728. Zhao M, Kiyoi H, Yamamoto Y, Ito M, Towatari M, Omura S, Kitamura T, Ueda R, Saito H and Naoe T. (2000). Leukemia, 14, 374–378. Zheng R, Friedman AD and Small D. (2002). Blood, 100, 4154–4161.

Oncogene