Constitutive Activation of FLT3 Stimulates Multiple Intracellular Signal Transducers and Results in Transformation K-F Tse1, G Mukherjee2 and D Small1,3

Leukemia (2000) 14, 1766–1776  2000 Macmillan Publishers Ltd All rights reserved 0887-6924/00 $15.00 www.nature.com/leu Constitutive activation of FLT3 stimulates multiple intracellular signal transducers and results in transformation K-F Tse1, G Mukherjee2 and D Small1,3

Departments of 1Oncology, 3Pediatrics and 2Comparative Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA

Aberrant expression of FLT3 has been found in most cases of subunit of phosphatidylinositol 3Ј-kinase (PI3K), RAS/GTPase B-lineage ALL and AML, and subsets of T cell ALL, CML in blast activating protein, phospholipase C-␥, VAV, SHC, GRB2 and crisis and CLL. In 20% of patients with AML the receptor has 21–27 small internal tandem duplications of the juxtamembrane CBL. region which appear to contitutively activate the receptor. To c-FMS and c-KIT are over-expressed by many cases of investigate whether FLT3 activation could play a role in leuke- myeloid leukemia but not often in ALL. In contrast, data from mia, we generated a constitutively activated FLT3 by fusing its our previous studies and others have demonstrated aberrant cytoplasmic domain to the helix–loop–helix domain of TEL in expression of FLT3 in most cases of B-lineage ALL and AML analogy to the fusion that occurs with TEL-PDGFR in CMML. as well as subsets of T-ALL and CML samples.28–31 Over- In vitro translation assays demonstrated oligomerization and intrinsic tyrosine kinase activity of the TEL-FLT3 chimeric expression occurs at both RNA and protein levels and leads receptor. Constitutively activated TEL-FLT3 conferred IL-3 to a functional receptor as addition of FL in vitro causes phos- independence and long-term proliferation to transfected Ba/F3 phorylation of the receptor and a proliferative response in cells. Immunoblot analyses showed that JAK 2, STAT 3, STAT some cases of leukemic cells.28,32,33 5a, STAT 5b and CBL were tyrosine-phosphorylated in TEL- Expression of mRNA for the FLT3 ligand (FL) has also been FLT3 expressing Ba/F3 cells in the absence of IL-3. These data detected in virtually all leukemic-derived cell lines, bone mar- suggest a possible role for the JAK/STAT pathway in FLT3 sig- 9,34–36 naling. Transplantation of TEL-FLT3 expressing Ba/F3 cells into row stroma and endothelial cells. Coexpression of syngeneic mice caused mortality in all mice by 3 weeks after SCF/c-KIT and CSF-1/c-FMS pairs has also been reported in injection. Histopathologic analysis demonstrated a massive cases of leukemia.37–39 Coexpression of both FLT3 and its infiltration of mononuclear cells in the liver, spleen and bone ligand by the same leukemia cell suggest a possible marrow. The mimicking of naturally occurring TEL fusions pro- autocrine/paracrine/intracrine signaling loop which could vides an approach to assess aspects of the biology of activated contribute to the development or maintenance of FLT3, or other receptor-type tyrosine kinases (RTKs) in leu- 40,41 kemic transformation. Leukemia (2000) 14, 1766–1776. leukemia. Keywords: FLT3; STAT5; JAK2; CBL; transformation; leukemia In addition, small internal tandem duplications in the juxtamembrane portion of the receptor have been discovered in approximately 20% of patients with AML.42–45 The mutation Introduction appears to activate the kinase domain of the receptor through constitutive dimerization.46 While the consequence of FLT3 Signal transduction initiated by the interactions of hemato- activation is unproven, it is conceivable that constitutively poietic growth factors/cytokines with specific receptors is criti- activated FLT3 plays a role in the proliferation and block in cal for regulating normal hematopoietic cell growth and differ- differentiation that characterizes leukemia. entiation.1–3 Several of the hematopoietic growth factors In this report, we model an activated FLT3 receptor to study utilize members of the type III RTK family. This family is its possible role in leukemia. Our approach was to generate characterized by five immunoglobulin-like folds in the extra- a constitutively activated FLT3 receptor by fusion of the entire cellular domain, a single transmembrane domain and an inter- cytoplasmic domain of FLT3 to the helix–loop–helix (HLH; rupted tyrosine kinase domain in the intracellular region.4,5 also called ‘pointed’) domain of TEL. TEL is a member of the Members of this receptor family include two receptors which ETS family of transcription factors, and contains a highly conserved amino-terminal HLH domain which functions to play important roles in hematopoiesis, c-KIT and c-FMS, two 47,48 receptors for PDGF, and FLT3, the most recently discovered mediate protein–protein interaction. TEL is notable for its member of this family.6 involvement in several chromosomal translocations which During normal hematopoiesis, FLT3 is primarily expressed result in the generation of chimeric proteins such as TEL- + PDGF␤R in CMML, TEL-ABL in AML and TEL-JAK2 in within the CD34 hematopoietic stem/progenitor cell fraction 49,50 as well as a specific population of CD34− cells.7–10 FLT3 ALL. All of these fusions resulted in activated tyrosine kinase domains and cause transformation when expressed in appears to play a role in the maintenance of pluripotent hema- 51–54 topoietic stem cells7,11,12 and development of B cell progen- hematopoietic cells. itors,13–15 and dendritic cells.11,16–18 Similar to other class III RTKs, FLT3 is activated by ligand (FLT3 ligand, FL)-dependent dimerization and Materials and methods transphosphorylation of tyrosine residues.19,20 Subsequent signal transduction is mediated through association and/or Construction of chimeric receptors phosphorylation of cytoplasmic substrates including the p85 The construction of the TEL-FLT3 fusions was performed by polymerase chain reaction (PCR) as illustrated in Figure 1a. To generate DNA for the HLH domain of TEL, total RNA from Correspondence: D Small, The John Hopkins Oncology Center, Bunt- ing-Blaustein Cancer Research Building, Room 253, 1650 Orleans K562 cells was reverse transcribed and the resulting cDNA Street, Baltimore, Maryland 21231-1000, USA; Fax: 410-955-8897 was used as a template for 35 cycles of PCR: 94°C, 50°C and Received 30 March 2000; accepted 30 June 2000 72°C (1 min each) using Taq polymerase (GIBCO, Gaithers- Activated FLT3 results in leukemic transformation K-F Tse et al 1767

Figure 1 (a) Schematic illustration of the construction of TEL-FLT3 and kinase deletion mutant FLT3 (del) and TEL-FLT3 (del) constructs. (b) Oligomerization of TEL-FLT3 in vitro. TEL-FLT3 and TEL-FLT3 (del) cDNA constructs were expressed either alone or together by in vitro transcription and translation using radiolabeled 35S methionine. Labeled lysates were immunoprecipitated with anti-FLT3 antibody specific for the carboxyl-terminal of FLT3 (lanes 4–6) or with rabbit IgG as a control (lanes 7–9). Total reaction mixtures (lanes 1–3) and immunoprecipitates (lanes 4–9) were resolved by SDS-PAGE, transferred to a membrane, and exposed to film. T/F: TEL-FLT3; T/F (d): TEL-FLT3 (del). burg, MD, USA) with a primer (P1) containing a Sal1 site and (P4) containing FLT3 3Ј sequences: 5Ј-AGCTGTTGCGTTCAT- TEL 5Ј sequences: 5Ј-TGATCTCTCTCGCTGTGA-3Ј and a CAC-3Ј. The resulting TEL and FLT3 PCR products were primer (P2) containing sequences for both TEL and FLT3: 5Ј- mixed, boiled, allowed to reanneal and used as templates in CTTTTTGTACTTGTGACA (TEL)-TTCTTCATGGTTCTGATG a second round of PCR using the P1 and the P4 primers for (FLT3)-3Ј.7,49 To generate the cytoplasmic domain of FLT3, 30 cycles of amplification: 94°C (1 minute), 50°C (1 min) and FLT3 plasmid DNA was amplified by PCR for 30 cycles: 94°C, 72°C (2 min). The TEL-FLT3 fusion product was digested with 55°C and 72°C (1 min each) with a 5Ј primer (P3) containing Sal1 and BstEII and ligated to the Sal1 and BstEII digested full- both TEL and FLT3 sequences: 5Ј-CATCAGAACCATGAAGAA length FLT3 cDNA. The chimeric Tel-FLT3 product was then (TEL)-TGTCACAAGTACAAAAAG (FLT3)-3Ј and a 3Ј primer subcloned into the pCIneo expression vector (Promega, Madi-

Leukemia Activated FLT3 results in leukemic transformation K-F Tse et al 1768 son, WI, USA). Partial kinase domain deletion mutants for Cell culture and transfection both native FLT3 (FLT3 (del)) and chimeric TEL-FLT3 (Tel- FLT3 (del)) were created by digesting each plasmid with The murine IL-3-dependent pro-B cell line, Ba/F3, was main- EcoR1 to remove the region coding for the C-terminal 233 tained in RPMI 1640 medium (GIBCO) supplemented with amino acids of the kinase domain. The sequences of the 10% fetal calf serum (FCS; Gemini Bio-Products, Calabasas, recombinant constructs were conﬁrmed by the dideoxy CA, USA) and 1 ng/ml IL-3. Ba/F3 cells were transfected with sequencing method. vector, TEL (nucleotides 25–486),49 FLT3, FLT3 (del), TEL- FLT3 and TEL-FLT3 (del) plasmid DNA by electroporation at 300 mV/960 ␮F (Bio-Rad, Richmond, CA, USA). Transfected Cytokines and antibodies cells were cultured in IL-3 containing medium for 48 h and then selected in 1 mg/ml G418 (GIBCO) for a period of 2 Recombinant murine IL-3 (IL-3) and recombinant human FLT3 weeks. The ability of the cells to survive in the absence of IL- ligand were purchased from R & D Systems (Minneapolis, 3 was determined by trypan blue exclusion after plating MN, USA). Rabbit polyclonal antibodies against JAK 1, JAK 2, aliquots of the washed cells in 96-well plates. Transfected cell JAK 3, STAT 1, STAT 3, STAT 5a, STAT 5b, and CBL were lines were then subcloned by limiting dilution. For obtained from Santa Cruz Biotechnology (Santa Cruz, determination of growth properties of stably transfected Ba/F3 CA,USA). Polyclonal antibodies raised against the carboxyl cells, G418-resistant cells were seeded at 8 × 104 cells per terminus (amino acids 974–993, C-20) and the kinase insert plate in medium in the presence or absence of 1 ng/ml IL-3. region (amino acids 740–757, S-18) of FLT3 were from Santa Viable cell numbers were counted from triplicate cultures at Cruz Biotechnology. Mouse monoclonal antibody against the indicated time intervals. phosphotyrosine (4G10) was purchased from Upstate Biotech- nology (UBI, Lake Placid, NY, USA). Rabbit IgG was from Zymed (San Francisco, CA, USA). Cell starvation, stimulation and lysis

Parental and transfected Ba/F3 cells were washed free of In vitro transcription/translation and oligomerization residual IL-3 followed by cytokine deprivation for 16 h. Cells assay were then incubated with or without 1 ng/ml IL-3 or with or without 100 ng/ml FL for 15 min at 37°C under serum-free The cDNA constructs encoding TEL-FLT3 and TEL-FLT3 (del) conditions. Cells were washed once in cold phosphate-buf- were expressed using an in vitro transcription and translation fered saline (PBS) and lysed in cold NP-40 lysis buffer contain- rabbit reticulocyte lysate kit (TNT; Promega). Proteins were ing protease and phosphatase inhibitors (2 mmol/l sodium labeled by 35S methionine (Amersham, Arlington Heights, IL, orthovanadate, 50 ␮g/ml antipain, 5 ␮g/ml aprotinin, 1 ␮g/ml USA) incorporation. Labeled lysate (20 ␮l) was diluted in NP- leupeptin and 10 ␮g/ml phenylmethylsulfonyl fluoride). 40 lysis buffer (20 mmol/l Tris-HCl, pH 7.4, 150 mmol/l NaCl, Supernatants were collected after centrifugation at 16000 g 100 mmol/l NaF, 10% glycerol, 1% NP-40 and 10 mmol/l at 4°C and aliquots were used for immunoprecipitation with EDTA) and immunoprecipitated with 1 ␮g of anti-FLT3 anti- specific antibodies or boiled in sample buffer for later use as body (C-20) which is specific for the carboxyl-terminal total cellular lysate. domain of the FLT3. Immunoprecipitates and 5 ␮l of total reaction mixtures were boiled in sample buffer and resolved by SDS-polyacrylamide gel electrophoresis (PAGE). The pro- Immunoprecipitation and immunoblotting teins were transferred to Immobilon-P membranes (Millipore, Bedford, MA, USA) and analyzed by autoradiography. Clarified lysates were incubated for 16 h at 4°C with appropri- ate antibodies and then recombinant protein A-agarose (UBI) was added for 2 h at 4°C. Immunocomplexes were washed In vitro kinase assay in TBST (10 mmol/l Tris-HCI, pH 7.4, 100 mmol/l NaCl and 0.1% Tween-20), TBS (10 mmol/l Tris-NaCl, pH 7.4 and 100 Expression of the constructs was carried out using a non-radio- mmol/l NaCl), boiled in sample buffer, and resolved by SDS- labelled TNT reaction. Protein lysate (500 ␮g) was diluted in PAGE. For immunoblotting, proteins were transferred to NP-40 lysis buffer containing protease and phosphatase Immobilon membranes, blocked with 5% bovine serum inhibitors (described below) and immunoprecipitated with 1 albumin (BSA; Sigma, St Louis, MO, USA) in TBST and probed ␮g of anti-FLT3 antibody (S-18) which recognizes all of the with specific primary antibodies. Membranes were washed constructs containing a FLT3 portion. The immunoprecipitates and incubated with horseradish peroxidase-conjugated sec- were washed with wash buffer (10 mmol/l Tris-HCI, pH 7.4, ondary antibody (Amersham). The blots were then incubated 100 mmol/l NaCl, 0.1% Tween-20, 1 mmol/l dithiothreitol with enhanced chemiluminescence ECL (Amersham) and (DTT) and 10 ␮g/ml aprotinin) and kinase buffer (10 mmol/l exposed to film. In some cases, membranes were stripped of Tris-HCl, pH 7.4, 100 mmol/l NaCl, 0.1% Tween-20, 1 antibody for successive probing with different antibodies. ␮ mmol/l DTT, 10 g/ml aprotinin, 10 mmol/l MnCl2 and 10 ␮ mmol/l MgCl2). Kinase reactions were carried out in 50 lof kinase buffer at 30°C for 20 min with the addition of 10 ␮Ci Mice and histopathology of ␥32P-ATP (6000 Ci/mmol, Amersham). Enolase (6 ␮g) was also added as an exogenous substrate. The reactions were Six- to 8-week-old Balb/c mice were purchased from the Jack- terminated by the addition of sample buffer followed by boil- son Laboratory (Bar Harbor, ME, USA). Animals were housed ing for 5 min. The proteins were resolved by SDS-PAGE, in a temperature/humidity controlled environment and fed transferred to Immobilon membrane and subjected to commercial rodent chow and acidified water ad libitum. Mice autoradiography or immunoblotting. (five in each group) were injected in a tail vein with 3 × 106

Leukemia Activated FLT3 results in leukemic transformation K-F Tse et al 1769 cells from the transfected Ba/F3 cell lines. Survival of the precipitated with anti-FLT3 antibody which recognizes the injected mice was monitored until day 50. Tissues from the full-length and the kinase deletion constructs, and the precipi- sick mice including liver, spleen and bone marrow were sub- tates were subjected to in vitro kinase reactions with 32P-lab- jected to histopathological examination. Complete necropsies elled ATP with the addition of enolase as an exogenous sub- were performed on mice at the Division of Comparative Medi- strate. Immunoblotting analysis indicated that all proteins cine at the Johns Hopkins Hospital. Representative tissue were expressed in the lysates (although at varying levels) and samples were fixed by immersion in 10% neutral-buffered for- successfully immunoprecipitated (with the exception of TEL malin, embedded in paraffin, and 5 ␮m sections were stained as expected, Figure 2, panel a). Full-length FLT3, FLT3 (del) with hematoxylin and eosin (H&E) and photographed with a and TEL-FLT3 (del) proteins immunoprecipitated from the lys- Nikon microscope at 40–200 × magnification. ates did not exhibit tyrosine kinase activity (Figure 2, panel b, lanes 2, 3 and 5). This lack of kinase activity was seen even after 48 h of autoradiography. On the other hand, TEL-FLT3 Results immunoprecipitated from the lysate showed a high level of kinase activity, with only 30 min of exposure as measured by Construction and oligomerization of chimeric autophosphorylation of TEL-FLT3 protein and phosphorylation receptors of added enolase substrate (Figure 2, panel b, lane 4). To examine whether TEL-FLT3 also becomes constitutively Dimerization is known to be a prerequisite for activation of activated when expressed in a cell, we utilized the non-leuk- FLT3 and other receptor tyrosine kinases.55 Previous studies emic pro-B cell line, Ba/F3, derived from the Balb/c mouse. have implicated the amino-terminal domain of TEL as These cells do not express FLT3 receptor and are strictly responsible for the self-association and activation of the PDGF dependent on IL-3 for survival and proliferation. Ba/F3 cells receptor portion of the fusion that results from the t(5;12) stably transfected with each of the constructs were used for translocation that occurs between these two genes in a subset preparing cell lysates. Lysates were immunoprecipitated using of CMML patients.47,48,50 TEL-induced oligomerization has anti-FLT3 antibody followed by immunoblot analysis. The also been shown to induce constitutive activation of the pro- blots were probed sequentially with anti-phosphotyrosine tein kinase activity of non-receptor tyrosine kinases including antibody, stripped and then reprobed with anti-FLT3 antibody. ABL and JAK2.52–54 Figure 3a shows that transfected Ba/F3 cells expressed readily To investigate whether fusion of FLT3 to the same domain detectable levels of FLT3, FLT3 (del), TEL-FLT3 and TEL-FLT3 of TEL would result in oligomerization and activation of FLT3, (del) proteins. However, only TEL-FLT3 protein was strongly the HLH domain of TEL was fused to the cytoplasmic domain tyrosine-phosphorylated in IL-3-deprived cells (Figure 3b, lane of the wild-type FLT3 receptor. The schematic illustration of 3: bulk, lanes 5–10: independent clones). The faint band that the chimeric receptor and truncation mutant is shown in Fig- appears in each lane at the apparent molecular weight of 130 ure 1a. In Figure 1b, total reaction products form the in vitro kDa in Figure 3b is not FLT3 but is an artifact because Ba/F3 transcription and translation analysis confirmed that the cells do not express FLT3 and the antibody used does not fusions were of the expected size (lanes 1 and 2). Of note, all cross-react with murine FLT3. These results provide evidence transcription and translation products containing TEL fusions for constitutive activation of TEL-FLT3 in transfected Ba/F3 showed doublets that are thought to result from an additional nearby in frame internal translation start site and have been reported previously.51 To confirm the hypothesis that the TEL domain induces oligomerization of FLT3, the products were immunoprecipitated with FLT3 antibody specific for the carboxyl terminus of the FLT3 protein (Figure 1b, lanes 4–6), or with IgG control (Figure 1b, lanes 7–9). Because the TEL-FLT3 (del) does not contain the C-terminal epitope recognized by this antibody, the only way that it can be immunoprecipitated is by oligomerization with the full-length TEL-FLT3. The IgG control did not precipitate any of the translated proteins from the lysates (Figure 1b, lanes 7–9). Antibody directed against the C-terminus of FLT3 immunoprecipitated the TEL-FLT3 protein (Figure 1b, lane 4). Anti-FLT3 antibody failed to immuno- precipitate the TEL-FLT3 (del) as expected (Figure 1b, lane 5). However, when TEL-FLT3 and TEL-FLT3 (del) were cotrans- lated, the deletion mutant was co-immunoprecipitated by the antibody (Figure 1b, lane 6) demonstrating that the fusions oligomerized.

Figure 2 Activation of TEL-FLT3 in vitro. Constructs were in vitro Constitutive activation of FLT3 by TEL transcribed and translated with non-radioactive methionine and immunoprecipitated (IP) with an anti-FLT3 antibody that also recog- To investigate whether oligomerization of TEL-FLT3 would nizes deletion mutants. The immunoprecipitates were subjected to in ␥ 32 induce constitutive activation of the tyrosine kinase activity of vitro kinase assays using - P ATP with enolase added as an exgen- the receptor in the absence of FLT3 ligand, constructs, includ- ous substrate. The samples were analyzed by SDS-PAGE and the gel was transferred to membrane followed by autoradiography (panel b). ing full-length FLT3, FLT3 (del), TEL, TEL-FLT3 and TEL-FLT3 For immunoblotting analysis, the blot was probed with the same anti- (del), were in vitro transcribed and translated in the presence FLT3 antibody (panel a). Lane 1: TEL; lane 2: FLT3; lane 3: FLT3 (del); of non-radioactive methionine. The products were immuno- lane 4: TEL-FLT3; and lane 5: TEL-FLT3 (del).

Leukemia Activated FLT3 results in leukemic transformation K-F Tse et al 1770

Figure 4 Constitutive activation of FLT3 confers IL-3 independence to Ba/F3 cells. Ba/F3 cells transfected with vector, TEL, FLT3, FLT3 (del), TEL-FLT3, or TEL-FLT3 (del) and selected in G418 were seeded at 8 × 104 cells per plate in 10% FCS in the presence (a) or absence (b) of 1 ng/ml IL-3. Viable cell numbers were enumerated from triplicate cultures at daily intervals. The data shown are representative of three separate experiments.

Figure 3 Constitutive activation of TEL-FLT3 in Ba/F3 cells. Ba/F3 cells transfected with FLT3, TEL-FLT3 or kinase-inactive mutants FLT3 3 (Figure 4a). In contrast, when the transfectants were trans- (del) or TEL-FLT3(del) were grown in media without serum or IL-3 for ferred to IL-3-deprived medium supplemented only with 10% 16 h, lysed and immunoprecipitated with anti-FLT3 antibody. FCS, cells transfected with vector alone, TEL, FLT3, FLT3 (del) Immunoprecipitates were resolved by SDS-PAGE, transferred to mem- and TEL-FLT3 (del) did not proliferate and all cells died within brane and probed with anti-FLT3 antibody (a). The blot was stripped 4 days. Ba/F3 cells expressing TEL-FLT3, on the other hand, and then reprobed with anti-phosphotyrosine (4G10) antibody (b). were capable of continuous proliferation in the absence of IL- Lane 1: native FLT3; lane 2: FLT3 (del); lane 3: TEL-FLT3; lane 4: TEL- FLT3 (del); and lanes 5–10: TEL-FLT3 subclones. 3 at a rate comparable to parental or transfected cells grown in the presence of IL-3 (Figure 4b). These results demonstrate that constitutively activated FLT3 was able to abrogate IL-3 cells. Since there was no addition of IL-3 to the cultures, the dependence and elicit a proliferative response in Ba/F3 cells. activation is independent of IL-3 stimulation. This IL-3-independent phenotype was not observed in cells expressing the kinase-deﬁcient TEL-FLT3 (del). To examine the possibility of an autocrine mechanism evoked by TEL-FLT3 Transforming activity of TEL-FLT3 transfection, TEL-FLT3 expressing Ba/F3 cells were grown in IL-3-free medium and the conditioned medium was collected. To investigate the capacity of activated FLT3 to trigger a pro- Parental Ba/F3 cells were cultured in 10% FCS growth liferative response, the growth properties of stably transfected medium supplemented with 10–30% conditioned media in Ba/F3 cells were examined. Transfected bulk Ba/F3 cell the absence of IL-3. Parental Ba/F3 cells did not proliferate cultures were grown in the presence or absence of 1 ng/ml and died rapidly, suggesting that autocrine production of IL- IL-3 for 6 days. Viable cell numbers were determined from 3 or other proliferative factors did not account for IL-3 inde- triplicate cultures. In three separate experiments, all Ba/F3 pendent growth of TEL-FLT3 expressing Ba/F3 cells (data not transfectants grew at a comparable rate in the presence of IL- shown).

Leukemia Activated FLT3 results in leukemic transformation K-F Tse et al 1771 TEL-FLT3 expression induces constitutive tyrosine we examined the hypothesis that JAK-STAT proteins were phosphorylation of cellular proteins in Ba/F3 cells tyrosine-phosphorylated following TEL-FLT3 activation in Ba/F3 cells. To compare the spectrum of intracellular substrates phos- In IL-3 responsive cells, JAK 1 and JAK 2 are phosphorylated phorylated by constitutively activated TEL-FLT3 with that rapidly on tyrosine residues upon stimulation by IL-3. To evoked by FL or IL-3 stimulation, parental Ba/F3 cells and evaluate whether members of the JAK family were phos- cells expressing wild-type FLT3 or its kinase-deletion mutant phorylated in TEL-FLT3 expressing Ba/F3 cells in the absence FLT3 (del) were stimulated for 15 min with 100 ng/ml FL. of IL-3, cell lystates were prepared and subjected to immuno- Tyrosine-phosphorylated proteins were immunoprecipitated precipitation and immunoblotting with anti-JAK antibodies and then immunoblotted using antiphosphotyrosine anti- (Figure 6a). As expected, both JAK 1 and JAK 2 were phos- bodies. As illustrated in Figure 5, FL elicited an increase in tyrosine phosphorylation of an array of proteins in FLT3 expressing Ba/F3 cells (lane 5). TEL-FLT3 expresses a similar increase of tyrosine phosphorylated proteins in a pattern very similar to that of FL stimulated FLT3 (compare lanes 5 and 8). Neither TEL-FLT3 (del) nor FL stimulated FLT3 (del) showed a similar pattern of phosphorylation. Because TEL-FLT3 abrogated the requirement for IL-3, it was of interest to determine whether constitutive activation of FLT3 and activation of the IL-3 receptor resulted in phosphorylation of similar or distinct cellular substrates. Treatment of parental Ba/F3 cells with IL-3 resulted in phosphorylation of several prominent bands with some appearing similar but a number appearing different than the activated FLT3 pattern (compare lanes 2 to 5 or 8).

Participation of JAK and STAT proteins in FLT3- induced transformation

Given that binding of IL-3 to its cognate receptor induces tyrosine phosphorylation of proteins in the JAK-STAT signal transduction pathway56,57 and that the molecular masses of several prominent phosphoproteins in the lysate of TEL-FLT3 transfectant were similar to that of IL-3-treated parental cells,

Figure 6 Analysis of JAK and STAT tyrosine phosphorylation in Ba/F3 cells. Parental Ba/F3 cells and cells expressing the indicated constructs were grown in serum-free medium and IL-3 deprivation for Figure 5 Tyrosine phosphorylation of cellular proteins in Ba/F3 16 h. Parental cells were incubated with 1 ng/ml murine recombinant cells induced by constitutively activated TEL-FLT3. Parental Ba/F3 IL-3 for 15 min. (a) Lysates were prepared and immunoprecipitated cells and stable transfectants grown in serum and IL-3-free medium (IP) with either anti-JAK1 or anti-JAK2 antibody followed by SDS- for 16 h were incubated with or without murine IL-3 (1 ng/ml) or PAGE and consecutive immunoblotting (IB) analysis with with or without FL (100 ng/ml). Whole cell lysates were prepared and antiphosphotyrosine antibody 4G10 and anti-JAK1 or anti-JAK2 anti- subjected to immunoprecipitation (IP) with anti-phosphotyrosine antibody. (b) Analysis as per (a) above but lysates immunoprecipitated body 4G10 followed by immunoblotting (IB) with 4G10. Lanes 1 and with anti-STAT antibodies followed by blotting with 4G10 or anti- 2: parental cell line; lane 3: TEL; lanes 4 and 5: FLT3; lanes 6 and 7: STAT antibodies. Lane 1: parental cell line; lane 2: vector; lane 3: FLT3 (del); lane 8: TEL-FLT3; and lane 9: TEL-FLT3 (del). TEL; lane 4: FLT3; lane 5: TEL-FLT3; and lane 6: TEL-FLT3 (del).

Leukemia Activated FLT3 results in leukemic transformation K-F Tse et al 1772 phorylated in response to IL-3 treatment in parental Ba/F3 cells (lane 1). In the cell line expressing the TEL-FLT3 fusion, only JAK 2 (upper band) and a 120 kDa protein coprecipitated with JAK 2 but not Jak 1 were phosphorylated in the absence of IL-3 exposure (lane 5). The increase in JAK 2 phosphotyrosine content was not as great in the cells expressing TEL-FLT3 when compared to their IL-3 stimulated parental counterpart, despite an equal amount of JAK 2 detected by immunoblotting (lanes 1 and 5). Whether JAK 2 directly associates with activated FLT3 is currently under investigation. Ba/F3 cells transfected with the other constructs did not exhibit phosphorylation of either JAK 1 or JAK 2 upon IL-3 deprivation (lanes 2–4 and 6). In all transfected cells tested, there was no evidence of JAK 3 tyrosine phosphorylation although JAK3 protein was readily detectable (data not shown). The STAT proteins are substrates of activated JAKs,57 so it was of interest to determine whether any of these transcription factors were activated in TEL-FLT3 expressing Ba/F3 cells. Four of the known STAT proteins were immunoprecipitated Figure 7 Analysis of CBL phosphorylation in Ba/F3 cells from IL-3 stimulated parental and unstimulated transfected expressing constitutively activated TEL-FLT3. Cells were cultured in cells. Immunoblotting with an antiphosphotyrosine antibody IL-3-deprived, serum-free medium for 16 h followed by incubation was used to detect changes in phosphotyrosine content and with or without 1 ng/ml of IL-3 or 100 ng/ml of FL. Lysates were immunoprecipitated (IP) with anti-CBL antibody and immunoblotted blots were reprobed to assure adequate immunoprecipitation (IB) successively with antiphosphotyrosine antibody 4G10 (upper of each protein. As shown in Figure 6b, IL-3 stimulation panel) and anti-CBL antibody (lower panel). Lanes 1 and 2: parental resulted in phosphorylation of STAT 3, STAT 5a and STAT 5b cell line; lane 3: vector; lane 4: TEL; lanes 5 and 6: FLT3; lanes 7 and in parental Ba/F3 cells (lane 1). TEL-FLT3 also induced tyro- 8: FLT3 (del); lane 9: TEL-FLT3; and lane 10: TEL-FLT3 (del). sine phosphorylation of STAT 3, STAT 5a and STAT 5b (lane 5). The phosphorylation of STAT 3 was even greater in TEL- antibodies. As shown in Figure 7, CBL was not phosphorylated FLT3 transfectants when compared to IL-3 stimulated parental in unstimulated cells (lanes 3–5, 7, 10), but was tyrosine phos- cells. In comparison to the IL-3 stimulated parental cells, phorylated after 15 min in response to IL-3 or FL stimulation slightly lower levels of both STAT 5a and STAT 5b tyrosine (lanes 2 and 6). Notably, CBL was hyperphosphorylated in phosphorylation were seen in the TEL-FLT3 expressing cells. Ba/F3 cells expressing TEL-FLT3 (lane 9). The level of CBL STAT 1 phosphorylation was not detectable in either IL-3- phosphorylation varied much more than the level of CBL treated parental or TEL-FLT3 expressing cells despite readily protein which also appeared to be slightly increased with detectable levels of STAT 1 protein (data not shown). A low stimulation (Figure 7, lower panel). level of STAT 5a and STAT 5b phosphorylation was occassion- ally detected in cells expressing native FLT3 in the absence of ligand stimulation (Figure 6b, lane 4), suggesting partial Leukemic potential of the constitutively activated activation in Ba/F3 cells. This slight activation could be due FLT3 in mice to the overexpression of FLT3 receptor on the cell surface resulting in clustering and spontaneous activation, residual FL Because constitutively activated FLT3 conferred IL-3-inde- in the medium, or a combination of the two. These results pendent growth on the transfected Ba/F3 cells, we investigated demonstrate that transformation of Ba/F3 to IL-3 independ- the leukemic potential of this cell line in syngeneic Balb/c ence by TEL-FLT3 was accompanied by constitutive phos- mice. As illustrated in Figure 8, transfected Ba/F3 cells phorylation of JAK 2, STAT 3, STAT 5a, STAT 5b and a 120 kDa polypeptide that coprecipitated with JAK 2.

Tyrosine phosphorylation of CBL in Ba/F3 cells expressing activated TEL-FLT3

CBL is a 120 kDa protein originally isolated as the cellular homolog of the product of the transforming oncogene v-cbl of the Cas NS-1 retrovirus.58–60 Activation of v-cbl has been shown to induce pre-B cell lymphomas and myeloid leukemia in mice. Studies have shown that CBL is a target of tyrosine phosphorylation following cytokine stimulation by erythropoietin (EPO), IL-3 and FL among others.24,61 It is felt to modu- late signaling pathways in cells by recruiting ubiquitin-conju- gating enzymes to signaling substrates.62 To investigate whether CBL was tyrosine-phosphorylated in response to Figure 8 Survival of recipient mice after transplantation with Ba/F3 × 6 constitutive activation of FLT3, CBL was immunoprecipitated transfectants. Ba/F3 cells (3 10 ) expressing vector, TEL, FLT3, FLT3 (del), TEL-FLT3, or TEL-FLT3 (del) were injected intravenously into from lysates of untreated or cytokine-treated parental and syngeneic Balb/c mice on day 0. Survival of transplanted mice was Ba/F3 transfectants, resolved by SDS-PAGE and immunoblot- monitored daily for 50 days. Five mice were used for each experi- ted consecutively with anti-phosphotyrosine and anti-CBL mental group; a representative result is shown.

Leukemia Activated FLT3 results in leukemic transformation K-F Tse et al 1773

a b

c d

e f

g h

Figure 9 Histopathological examination of mice transplanted with vector/BaF3 or TEL-FLT3/BaF3 cells. Paraffin sections of the liver (a and b), spleen (c–f) and bone marrow (g and h) obtained from mice transplanted with vector/BaF3 (left column) or TEL-FLT3/BaF3 (right column) cells were stained with hematoxylin and eosin. The specimens shown are representative of five mice in each group. Original magnification: a and b: × 100, c and d: × 40, e and f: × 200, g and h: × 50. expressing vector, TEL, FLT3 (del) or TEL-FLT3 (del) have no IL-3-independent growth under our culture conditions. Since effect on the survival of the recipient mice over the experi- mouse FLT3 ligand is known to be active on human FLT3 mental period of 50 days. In contrast, TEL-FLT3/BaF3 cell- receptors, the FL-independent growth could be due to the injected mice became sick by 14 days following the injection combination of stimulation by FL expressed in the mice and and most died within 2–3 days after the onset of the illness overexpression of the receptor with subsequent self-associ- with none surviving beyond 22 days post-inoculation. Surpris- ation and activation. ingly, inoculation of Ba/F3 cells expressing wild-type FLT3 Histopathological analysis of the vector/BaF3-control and also caused mortality in mice, although, with delayed mor- TEL-FLT3/BaF3 inoculated mice revealed marked pathology in tality. This suggests that, as observed previously in cell culture, the latter group (Figure 9). Liver sections from the animals there was some degree of FLT3 activation in FLT3 expressing injected with vector/BaF3 cells demonstrated well-organized cells though this activation in vitro was not sufficient to cause hepatic lobules (Figure 9a). In the TEL-FLT3/BaF3 injected

Leukemia Activated FLT3 results in leukemic transformation K-F Tse et al 1774 mice, a massive infiltration of mononuclear cells was reports, JAK 1, JAK 2, STAT 1, STAT 3 and /or STAT 5 were observed in the sinusoids of the liver (Figure 9b). Mice trans- tyrosine-phosphorylated following receptor activation in planted with TEL-FLT3/BaF3 cells developed splenomegaly response to EGF, PDGF, CSF-1 or SCF stimulation.65–70 These with splenic weights ranging from 0.70 to 1.1 g compared studies, in conjunction with the findings reported here, imply with control mice having an average of 0.15 g. Microscopic a possible role for the JAK-STAT pathway in FLT3 receptor examination of the spleen from the control animals showed signaling. Whether this is a result of phosphorylation of JAKs distinct white and red pulp with well-defined architecture by the activated receptor via a direct interaction of the two (Figure 9c and e). In contrast, the architecture of the spleen proteins or is indirect and results from phosphorylation and was almost completely effaced in TEL-FLT3/BaF3 inoculated activation of other tyrosine kinases (eg SRC family members) mice (Figure 9d and f). The splenic red pulp in these mice or intermediate adaptor proteins is not known and will require was extensively infiltrated by mononuclear cells with a lym- further study. phocytic appearance. The femoral head of TEL-FLT3/BaF3 Analysis of the substrate specificity of the FLT3 receptor in inoculated mice were notable for increased cellularity in the cell lines of myeloid and lymphoid origin has shown that the marrow compartment and along the periosteum resulting from CBL proto-oncogene is prominently tyrosine-phosphorylated a massive mononuclear cell infiltration (Figure 9h). The in response to FL stimulation.24 CBL is also phosphorylated in expansion and infiltration of mononuclear cells observed in response to a broad range of other cytokines including IL-2, liver, spleen and bone marrow of TEL-FLT3/BaF3 inoculated IL-3, IL-15, GM-CSF, CSF-1 and EPO.24,61,71,72 It appears that mice confirm a leukemic phenotype for the Ba/F3 cells CBL functions as an adapter protein to direct the ubiquitin- expressing TEL-FLT3. mediated degradation of activated members of signaling pathways including RTKs and cytoplasmic kinases. In our experimental system, CBL was transiently tyrosine-phos- Discussion phorylated in IL-3 stimulated parental and FL-stimulated, FLT3 expressing Ba/F3 cells, confirming the previous findings.24,73 FLT3 has been shown by our laboratory and a number of Notably, CBL was constitutively hyperphosphorylated in cells others to be overexpressed in most cases of B-lineage ALL and expressing the TEL-FLT3 fusion. CBL is also constitutively AML as well as in subsets of T cell ALL and CML in blast 74–76 28–31 phosphorylated in cells lines transformed with BCR-ABL. crisis. The overexpressed receptor is functional and can Thus, CBL may play a role in TEL-FLT3-induced Ba/F3 cellular result in increased proliferation and block apoptosis in subsets transformation. Alternatively, it could represent a heroic yet of myeloid leukemia cells when stimulated with FL.32 It has unsuccessful effort on the part of the cell to destroy the offend- also been shown that mice in which FL is overexpressed are ing activated signaling protein(s). predisposed to the development of leukemia.63 The internal Our animal studies show that there is a good correlation tandem duplication of FLT3 that occurs in 20% of leukemic between the constitutive activation of FLT3 and leukemic patients also appears to result in constitutive dimerization development. Activating point mutations or chromosomal and activation.46,64 translocations involving FLT3 have not been reported thus far. In order to study the possible effects of an activated FLT3 However, internal tandem duplications of the juxtamembrane receptor and the signal transduction pathway induced by this activation, we generated a fusion of FLT3 with the oligomeriz- region of FLT3 gene reported in 20% of adult and childhood AML patients appears to constitutively activate the recep- ation domain of TEL, analogous to the TEL-PDGF receptor 42–45 fusion that occurs in CMML. Our results demonstrate that, tor. A recent publication has reported that 32D cells transfected with the mutated receptor are transformed and cause when expressed in Ba/F3 cells, the TEL-FLT3 fusion was able 77 to evoke a phosphorylation response similar to that elicited leukemia when injected into syngeneic mice. The frequent by the FL-stimulated wild-type FLT3. Furthermore, expression overexpression of FLT3 seen in leukemias with the possibility of TEL-FLT3 in Ba/F3 cells eliminated the IL-3 requirement for of self-activation or heightened sensitivity to FL also makes proliferation. These findings parallel previous observations the wild-type receptor a candidate for a potentially important that c-FMS/FLT3 chimeric receptors expressed in Ba/F3 cells role in this disease. were able to generate a proliferative response to CSF-1 in the Taken together, our data provide evidence that an activated absence of IL-3.21,22 We also detected a low level of wild-type FLT3 can lead to leukemic transformation through constitutive FLT3 receptor activation in a ligand-independent manner in activation of similar pathways utilized by FL-stimulated wild- Ba/F3 cells expressing high levels of FLT3. Perhaps this is due type FLT3. This model should be useful in defining the critical to overexpression of surface receptor leading to spontaneous elements of the pathway required for transformation and as a oligomerization and activation in the absence of ligand or to way to investigate possible therapeutic approaches in a sensitized response to small amounts of FL in serum. treating leukemia. The spectrum of tyrosine-phosphorylated proteins in TEL- FLT3 expressing cells partially overlaps that of parental Ba/F3 cells evoked by IL-3 stimulation. This was confirmed by immunoprecipitation and immunoblotting analyses of known components of the IL-3-mediated signaling cascade. In these Acknowledgements studies, JAK 2, STAT 3, STAT 5a and STAT 5b were all found to be constitutively tyrosine-phosphorylated by the activated TEL-FLT3 as they are in IL-3 stimulated parental cells. This We gratefully acknowledge Enrico Novelli for technical assist- suggests that constitutively activated FLT3 might overcome the ance. We also thank Zemin Wang and Gerry Hoehn for help- IL-3 dependence of the Ba/F3 cells by activating essential ful suggestions, and Linzhao Cheng and Joseph Ueland for components of the IL-3 signaling pathway. There is some evi- critical review of the manuscript. This work was supported by dence for the coupling of JAK-STAT-mediated signaling to a grant from the National Institute of Health (No. tyrosine kinase containing receptor activation. In these PO1CA70970).

Leukemia Activated FLT3 results in leukemic transformation K-F Tse et al 1775 References Dubreuil P. Substrate specificities and identification of a putative binding site for PI3K in the carboxy tail of the murine Flt3 receptor 1 D’Andrea A. Hematopoietic growth factors and the regulation of tyrosine kinase. Oncogene 1994; 9: 1755–1765. differentiative decisions. Curr Opin Cell Biol 1994; 6: 804–805. 23 Casteran N, Rottapel R, Beslu N, Lecocq E, Birnbaum D, Dubreuil 2 Miyajima A, Kitamura T, Haroda N, Yokota T, Arai K-I. Cytokine P. Analysis of the mitogenic pathway of the FLT3 receptor and receptors and signal transduction. Annu Rev Immunol 1992; 10: characterization in its C terminal region of a specific binding site 295–331. for the PI3Ј kinase. Cell Mol Biol 1994; 40: 443–456. 3 Metcalf D. Cellular hematopoiesis in the twentieth century. Semin 24 Lavagna-Sevenier C, Marchetto S, Birnbaum D, Rosnet O. FLT3 Hematol 1999; 36: 5–12. signaling in hematopoietic cells involves CBL, SHC and an 4 Yarden Y, Ullrich A. Growth factor receptor tyrosine kinases. Ann unknown P115 as prominent tyrosine-phosphorylated substrates. Rev Biochem 1988; 57: 443–478. Leukemia 1998; 12: 301–310. 5 Hanks SK, Quinn AM, Hunter T. The protein kinase family con- 25 Zhang S, Mantel C, Broxmeyer HE. Flt3 signaling involves tyrosyl- served features and deduced phylogeny of the catalytic domains. phosphorylation of SHP-2 and SHIP and their association with Science 1988; 241: 42–52. Grb2 and Shc in Baf3/Flt3 cells. J Leuk Biol 1999; 65: 372–380. 6 Rosnet O, Birnbaum D. Hematopoietic receptors of class III recep- 26 Zhang S, Broxmeyer HE. p85 Subunit of P13 kinase does not bind tor-type tyrosine kinases. Crit Rev Oncogen 1993; 4: 595–613. to human Flt3 receptor, but associates with SHP2, SHIP, and a 7 Small D, Levenstein M, Kim E, Carow C, Amin S, Rockwell P, tyrosine-phosphorylated 100-kDa protein in Flt3 ligand-stimulated Witte L, Burrow C, Ratajczak M, Gewirtz AM, Civin C. STK-1, the + hematopoietic cells. Biochem Biophys Res Commun 1999; 254: human homolog of Flk-2/FLT-3, is selectively expressed in CD34 440–445. human bone marrow cells and is involved in the proliferation of 27 Hayakawa F, Towatari M, Kiyoi H, Tanimoto M, Kitamura T, Saito early progenitor/stem cells. Proc Natl Acad Sci USA 1994; 91: H, Naoe T. Tandem-duplicated Flt3 constitutively activates STAT5 459–463. and MAP kinase and introduces autonomous cell growth in IL-3- 8 Rappold I, Ziegler BL, Kohler I, Marchetto S, Rosnet O, Birnbaum dependent cell lines. Oncogene 2000; 19: 624–631. D, Simmons PJ, Zannettino ACW, Hill B, Neu S, Knapp W, Alitalo 28 Carow CE, Levenstein M, Kaufmann SH, Chen J, Amin S, Rockwell R, Alitalo K, Ullrich A, Kanz L, Buhring H-J. Functional and pheno- P, Witte L, Borowitz MJ, Civin CI, Small D. Expression of the typic characterization of cord blood and bone marrow subsets hematopoietic growth factor receptor FLT3 (STK-1/Flk2) in human expressing FLT3 (CD135) receptor tyrosine kinase. Blood 1997; leukemias. Blood 1996; 87: 1089–1096. 90: 111–1125. 29 Birg F, Courcoul M, Rosnet O, Bardin F, Pebusque M, Marchetto 9 Brasel K, Escobar S, Anderberg R, de Vries P, Gruss H-J, Lyman S, Tabilio A, Mannoni P, Birnbaum D. Expression of the FMS/KIT- SD. Expression of the Flt3 receptor and its ligand on hematopoietic like gene FLT3 in human acute leukemias of the myeloid and cells. Leukemia 1995; 9: 1212–1218. lymphoid lineages. Blood 1992; 80: 2584–2593. 10 Gotze KS, Ramirez M, Tabor K, Small D, Matthews W, Civin CI. 30 Rosnet O, Buhring HJ, Marchetto S, Rappold I, Lavagna C, Sainty Flt3high and Flt3low CD34+ progenitor cells isolated from human bone D, Arnoulet C, Chabannon C, Kanz L, Hannum C, Birnbaum D. marrow are functionally distinct. Blood 1997; 91: 1947–1958. Human FLT3/FLK2 receptor tyrosine kinase is expressed at the sur- 11 Gabbianelli M, Pelosi E, Montesoro E, Valtieri M, Luchetti L, Sam- face of normal and malignant hematopoietic cells. Leukemia oggia P, Vitelli L, Barberi T, Testa U, Lyman S. Multi-level effects 1996; 10: 238–248. of Flt3 ligand on human hematopoiesis: expansion of putative stem cells and proliferation of granulomonocytic 31 Turner AM, Lin NL, Issarachai S, Lyman SD, Broudy VC. FLT3 progenitors/monocytic precursors. Blood 1995; 86: 1661–1670. receptor expression on the surface of normal and malignant 12 Shah AJ, Smogorzewska EM, Hannum C, Crooks GM. Flt3 ligand human hematopoietic cells. Blood 1996; 88: 3383–3390. induces proliferation of quiescent human bone marrow 32 Dehmel U, Zaborski M, Meierhoff G, Rosnet O, Birnbaum D, Lud- CD34+CD38− cells and maintains progenitor cells in vitro. Blood wig WD, Quentmeier H, Drexler HG. Effects of Flt3 ligand on 1996; 87: 3563–3570. human leukemia cells. I. Proliferative response of myeloid leuk- 13 Hirayama F, Lyman SD, Clark SC, Ogawa M. The flt3 ligand sup- emia cells. Leukemia 1996; 10: 261–270. ports proliferation of lymphohematopoietic progenitors and early 33 Lisovsky M, Estrov Z, Zhang X, Consoli U, Sanchez-Williams G, B-lymphoid progenitors. Blood 1995; 85: 1762–1768. Snell V, Munker R, Goodacre A, Savchenko V, Andreeff M. Flt3 14 Hunte BE, Hudak S, Campbell D, Xu Y, Rennick D. Flk2/Flt3 ligand stimulates proliferation and inhibits apoptosis of acute ligand is a potent cofactor for the growth of primitive B cell pro- myeloid leukemia cells: regulation of Bcl-2 and Bax. Blood 1996; genitors. J Immunol 1996; 156: 489–496. 88: 3987–3997. 15 Mackaretschian K, Hardin JD, Moore KA, Boast S, Goff SP, Lem- 34 Meierhoff G, Dehmel U, Gruss H-J, Rosnet O, Birnbaum D, ischka IR. Targeted disruption of the Flk2/Flt3 gene leads to Quentmeier H, Dirks W, Drexler HG. Expression of FLT3 receptor deficiencies in primitive hematopoietic progenitors. Immunity and FLT3-ligand in human leukemia–lymphoma cell lines. Leuk- 1995; 3: 146–161. emia 1995; 9: 1368–1372. 16 Hannum C, Culpepper J, Campbell D, McClanahan T, Zurawski S, 35 Lisovsky M, Braun M, Ge Y, Takahira H, Lu L, Savchenko VG, Bazan JF, Kastelein R, Hudak S, Wagner J, Mattson J, Luh J, Duda Lyman SD, Broxmeyer HE. Flt3-ligand production by human bone G, Martina N, Peterson D, Menon S, Shanafelt A, Muench M, marrow stromal cells. Leukemia 1996; 10: 1012–1018. Keiner G, Namikawa R, Rennick D, Roncarolo MG, Zlotnik A, Ros- 36 Solanilla AGC, Lemervier C, Dupouy M, Mahon FX, Schweitzer net O, Dubreuil P, Birnbaum D, Lee F. Ligand for FLT3/FLK2 recep- K, Reiffers J, Weksler B, Ripoche J. Expression of Flt3-ligand by tor tyrosine kinase regulates growth of haematopoietic stem cells the endothelial cell. Leukemia 2000; 14: 153–162. and is encoded by variant RNAs. Nature 1994; 368: 643–648. 37 Pietsch T, Kyas U, Yakisan E, Hadam M, Ludwig W, Zsebo K, 17 McKenna HJ, de Vries P, Brasel K, Lyman SD, Williams DE. Effect Welte K. Effects of human stem cell factor (c-kit ligand) on prolifer- of Flt3 ligand on the ex vivo expansion of human CD34+ hemato- ation of myeloid leukemia cells: heterogeneity in response and poietic progenitor cells. Blood 1995; 86: 3414–3420. synergy with other hematopoietic growth factors. Blood 1992; 80: 18 Lyman SD, Jacobsen SEW. c-kit ligand and Flt3 ligand: 1199–1206. stem/progenitor cell factors with overlapping yet distinct activities. 38 Torres H, Dubrueil P, Falzetti F, Courcoul M, Lopez M, Falcinelli Blood 1998; 91: 1101–1134. F, Birg F, Tabilio A, Mannoni P. c-fms expression in acute leuk- 19 van der Geer P, Hunter T, Lindberg R. Receptor protein-tyrosine emias with complex phenotypes. Leukemia 1990; 4: 673–677. kinases and their signal transduction pathways. Annu Rev Cell Biol 39 Parwaresch MR, Kriepe H, Felgner J, Heidorn K, Jaquet K, Bode- 1994; 10: 251–337. wadt-Radzun S, Radzun HJ. M-CSF and M-CSF receptor gene 20 Heldin C. Dimerization of cell surface receptors in signal trans- expression in acute myelomonocytic leukemias. Leukemia Res duction. Cell 1995; 80: 213–223. 1990; 14: 27–37. 21 Dosil M, Wang S, Lemischka I. Mitogenic signalling and substrate 40 Metcalf D. The roles of stem cell self-renewal and autocrine specificity of the Flk2/Flt3 receptor tyrosine kinase in fibroblasts growth factor production in the biology of myeloid leukemia. Can- and interleukin 3-dependent hematopoietic cells. Mol Cell Biol cer Res 1989; 49: 2305–2311. 1993; 13: 6572–6585. 41 Hawley R. Interleukin-6-type cytokines in myeloproliferative 22 Rottapel R, Turck, CW, Casteran N, Liu X, Birnbaum D, Pawson T, disease. Ann NY Acad Sci 1995; 762: 294–307.

Leukemia Activated FLT3 results in leukemic transformation K-F Tse et al 1776 42 Horiike S, Yokota S, Nakao M, Iwaai T, Sasai Y, Kaneko H, 59 Donovan J, Wange R, Langdon W, Samelson L. The protein pro- Taniwaki M, Kashima K, Fujii H, Abe T, Misawa S. Tandem duct of the c-cbl protooncogene is the 120-kDa tyrosine-phos- duplications of the FLT3 receptor gene are associated with leu- phorylated protein in Jurkat cells activated via the T cell antigen kemic transformation of myelodysplasia. Leukemia 1997; 11: receptor. J Biol Chem 1994; 269: 22921–22924. 1442–1446. 60 Langdon WY, Hartley JW, Klinken SP, Ruscetti SK, Morse HC. v- 43 Nakao M, Yokota S, Iwai T, Kaneko H, Horiike S, Kashima K, cbl, an oncogene from a dual-recombinant murine retrovirus that Sonoda Y, Fujimoto T, Misawa S. Internal tandem duplication of induces early B-lineage lymphomas. Proc Natl Acad Sci USA the Flt3 gene found in acute myeloid leukemia. Leukemia 1996; 1989; 86: 1168–1172. 10: 1911–1918. 61 Barber DL, Mason JM, Fukazawa T, Reedquist KA, Druker BJ, 44 Iwai T, Yokota S, Nakao M, Okamoto T, Taniwaki M, Onodera N, Band H, D’Andrea AD. Erythropoietin and interleukin-3 activate Watanabe A, Kikuta A, Tanaka A, Asami K, Sekine I, Mugishima tyrosine phosphorylation of CBL and association with CRK adaptor H, Nishimura Y, Koizumi S, Horikoski Y, Mimaya J, Ohta S, Nishi- proteins. Blood 1997; 89: 3166–3174. kawa K, Iwai A, Shimokawa T, Nakayama M, Kawakami K, 62 Joazeiro CAWS, Huang H, Leverson JD, Hunter T, Liu YC. The Gushiken T, Hyakuna N, Katano N, Tsurusawa M, Fujimoto T. tyrosine kinase negative regulator c-Cbl as a RING-type, E2- Internal tandem duplication of the FLT3 gene and clinical evalu- dependent ubiquitin-protein ligase. Science 1999; 286: 309–312. ation in childhood acute myeloid leukemia. Leukemia 1999; 13: 63 Hawley TS, Fong AZC, Griesser H, Lyman SD, Hawley RG. Leuk- 38–43. emic predisposition of mice transplanted with gene-modified hem- 45 Xu F, Taki T, Yang HW, Hanada R, Hongo T, Ohnishi H, Kobaya- atopoietic precursors expressing flt3 ligand. Blood 1998; 92: shi M, Bessho F, Yanagisawa M, Hayashi Y. Tandem duplication 2003–2011. of the FLT3 gene is found in acute lymphoblastic leukaemia as 64 Yokota S, Kiyoi H, Nakao M, Iwai T, Misawa S, Okuda T, Sonoda well as acute myeloid leukaemia but not in myelodysplastic syn- Y, Abe T, Kashima K, Matsuo Y, Naoe T. Internal tandem dupli- drome or juvenile chronic myelogenous leukaemia in children. Br cation of the FLT3 gene is preferentially seen in acute myeloid J Haematol 1999; 105: 155–162. leukemia and myelodysplastic syndrome among various hemato- 46 Kiyoi H, Towatari M, Yokota S, Hamaguchi M, Ohno R, Saito H, logical malignancies. A study on a large series of patients and cell Naoe T. Internal tandem duplication of the FLT3 gene is a novel lines. Leukemia 1997; 11: 1605–1609. modality of elongation mutation which causes constitutive acti- 65 Zhong Z, Wen Z, Darnell J. Stat3: a STAT family member activated vation of the product. Leukemia 1998; 12: 1333–1337. by tyrosine phosphorylation in response to epidermal growth fac- 47 Jousset C, Carron C, Oury C, Susanter-Fourt I, Charon M, Levin J, tor and interleukin-6. Science 1994; 264: 95–98. Bernard O, Ghysdael J. A domain of TEL conserved in a subset of 66 Sachsenmaier C, Sadowski HB, Cooper JA. STAT activation by the ETS proteins defines a specific oligomerization interface essential PDGF receptor requires juxtamembrane phosphorylation sites but to the mitogenic properties of the TEL-PDGFRB oncoprotein. not Src tyrosine kinase activation. Oncogene 1999; 18: 3583– EMBO J 1997; 16: 69–82. 3592. 48 Golub T, Goga A, Barker G, Afar D, McLaughlin J. Oligomeriz- 67 Novak U, Nice E, Hamilton JA, Paradiso L. Requirement for Y706 ation of the ABL tyrosine kinase by the Ets protein TEL in human of the murine (or Y708 of the human) CSF-1 receptor for STAT1 leukemia. Mol Cell Biol 1996; 16: 4107–4116. activation in response to CSF-1. Oncogene 1996; 13: 2607–2613. 49 Golub TR, Barker GF, Lovett M, Gilliland DG. Fusion of PDGF 68 Novak U, Mui A, Miyajima A, Paradiso L. Formation of STAT5- receptor ␤ to a novel ets-like gene, tel, in chronic myelomonocytic containing DNA binding complexes in response to colony-stimu- leukemia with t(5;12) chromosomal translocation. Cell 1994; 77: lating factor-1 and platelet-derived growth factor. J Biol Chem 307–316. 1996; 271: 18350–18354. 50 Golub TR, Barker GF, Stegmaier K, Gilliland DG. Involvement of 69 Deberry C, Mou S, Linnekin K. Stat1 associates with c-kit and is the TEL gene in hematologic malignancy by diverse molecular activated in response to stem cell factor. Biochem J 1997; 327: genetic mechanisms. Curr Top Microbiol Immunol 1996; 211: 78–80. 279–288. 70 Ryan JJ, Huang H, McReynolds LJ, Shelburne C, Hu-Li J, Huff TF, 51 Carroll M, Tomasson MH, Barker GF, Golub TR, Gilliland DG. The Paul WE. Stem cell factor activates STAT-5 DNA binding in IL-3- TEL/platelet-derived growth factor ␤ receptor (PDGF␤R) fusion in derived bone marrow mast cells. Exp Hematol 1997; 25: 357–362. chronic meylomonocytic leukemia is a transforming protein that 71 Wang Y, Yeung Y-G, Langdon WY, Stanley ER. c-cbl is transiently self-associates and activates PDGFBR kinase-dependent signaling tyrosine-phosphorylated, ubiquininated, and membrane-targeted pathways. Proc Natl Acad Sci USA 1996; 93: 14845–14850. following CSF-1 stimulation of macrophages. J Biol Chem 1996; 52 Papadopoulos P, Ridge SA, Boucher CA, Stocking C, Wiedermann 271: 17–20. LM. The novel activation of ABL by fusion to an ets-related gene, 72 Odai H, Sasaki K, Iwamatsu A, Hanazono Y, Tanaka T, Mitani K, TEL. Cancer Res 1995; 55: 34–38. Yazaki Y, Hirai H. The proto-oncogene product c-Cbl becomes 53 Peeters P, Raynaud SD, Cools J, Wlodarska I, Grosgeorge J, Philip tyrosine phosphorylated by stimulation with GM-CSF or Epo and P, Monpoux F, Van Rompaey L, Baens M, Van den Gerghe H, constitutively binds to the SH3 domain of Grb2/Ash in human Maryen P. Fusion of TEL, the ETS-variant gene 6 (ETV6), to the hematopoietic cells. J Biol Chem 1995; 270: 10800–10805. receptor-associated kinase JAK2 as a result of t(9;12) in a lymphoid 73 Lavagna-Sevenier C, Marchetto S, Birnbaum D, Rosnet O. The and t(9;15;12) in a myeloid leukemia. Blood 1997; 90: 2535–2540. CBL-related protein CBLB participates in FLT3 and interleukin-7 54 Lacronique V, Boureux A, Valle VD, Poirel H, Quang CT, Mau- receptor signal transduction in pro-B cells. J Biol Chem 1998; 273: chauffe M, Berthou C, Lessard M, Berger R, Ghysdael J, Bernard 14962–14967. OA. A TEL-JAK2 fusion protein with constitutive kinase activity in 74 Ribon V, Hubbell S, Herrera R, Saltiel AR. The product of the cbl human leukemia. Science 1997; 278: 1309–1312. oncogene forms stable complexes in vivo with endogenous Crk 55 Weiss A, Schlessinger J. Switching signals on or off by receptor in a tyrosine phosphorylation-dependent manner. Mol Cell Biol dimerization. Cell 1998; 94: 277–280. 1996; 16: 45–52. 56 Hara T, Miyajima A. Function and signal transduction mediated 75 Sattler M, Salgia R, Okuda K, Uemura N, Durstin MA, Pisick E, by the interleukin 3 receptor system in hematopoiesis. Stem Cells Xu G, Li J-L, Prasad KV, Griffin JD. The proto-oncogene product 1996; 14: 605–618. p120CBL and the adaptor proteins CRKL and c-CRK like c-ABL, 57 Darnell JE, Kerr IM, Stark G. Jak-STAT pathways and transcrip- p190 BCR/ABL and p210BCR/ABL to the phosphatidylinositol-3Ј tional activation in response to IFNs and other extracellular sig- kinase pathway. Oncogene 1996; 12: 839–846. naling proteins. Science 1994; 264: 1415–1420. 76 de Jong R, ten Hoeve J, Heisterkamp N, Groffen J. Crkl is com- 58 Fukazawa T, Reedquist K, Trub T, Soltoff S, Panchamoorthy G, plexed with tyrosine-phosphorylated Cbl in Ph-positive leukemia. Druker B, Cantley L, Shoelson S, Band H. The SH3 domain-bind- J Biol Chem 1995; 270: 21468–21471. ing T cell tyrosyl phosphoprotein p120. Demonstration of its ident- 77 Zhao M, Kiyoi H, Yamamoto Y, Towatari M, Omura S, Kitamura ity with the c-cbl protooncogene product and in vivo complexes T, Ueda R, Saito H, Naoe T. In vivo treatment of mutant FLT3- with Fyn, Grb2, and phosphatidylinositol 3-kinase. J Biol Chem transformed murine leukemia with a tyrosine kinase inhibitor. 1995; 270: 19141–19150. Leukemia 2000; 14: 374–378.

Leukemia