Original Article

Leukemia (2006) 20, 1414–1421. doi:10.1038/sj.leu.2404266; published online 8 June 2006

FLT3 is fused to ETV6 in a myeloproliferative disorder with hypereosinophilia and a t(12;13)(p13;q12) translocation

H A Vu1,2, P T Xinh1,2, M Masuda3, T Motoji3, A Toyoda4, Y Sakaki4, K Tokunaga2 and Y Sato1

  1. 1Department of Pathology, Division of Ultrafine Structure, Research Institute of International Medical Center of Japan, Tokyo, Japan
  2. 2Department of Human Genetics, School of International Health, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan
  3. 3Department of Hematology, Tokyo Women's Medical University, Tokyo, Japan
  4. 4Genome Core Technology Facilities, RIKEN Genomic Sciences Center, Kanagawa, Japan

Correspondence: Dr Y Sato, Department of Pathology, Division of Ultrafine Structure, Research Institute of International Medical Center of Japan, Toyama 1-21-1, Shinjuku-ku, Tokyo 162-0052, Japan. E-mail: ysato@ri.imcj.go.jp

Received 3 February 2006; Revised 22 March 2006; Accepted 5 April 2006; Published online 8 June 2006.



The FMS-like tyrosine kinase 3 (FLT3) gene, belonging to the receptor tyrosine kinase (TK) subclass III family, plays an important role in normal hematopoiesis and is one of the most frequently mutated genes in hematologic malignancies as well as an attractive target for directed inhibition. Activating mutations of this gene, including internal tandem duplication in the juxtamembrane (JM) domain and point mutations in the TK domain, are found in approximately one-third of patients with acute myeloid leukemia and in a smaller subset of patients with acute lymphoblastic leukemia. We report here that FLT3 may contribute to leukemogenesis in a patient with myeloproliferative disorder and a t(12;13)(p13;q12) translocation through generating a fusion gene with the ETS variant gene 6 (ETV6) gene. ETV6 has been reported to fuse to various partner genes, including TK and transcription factors. Both ETV6/FLT3 and reciprocal FLT3/ETV6 transcripts were detected in the patient mRNA by reverse transcriptase-polymerase chain reaction. At the protein level, however, only ETV6/FLT3 products were expressed. Among them, one retains the helix–loop–helix (HLH) oligomerization domain of ETV6 and the JM as well as TK domain of FLT3. FLT3 receptor in leukemic cells might be inappropriately activated through dimerization by HLH domain of ETV6, which consequently interfered with proliferation and differentiation of hematopoietic cells.


FLT3, ETV6, FISH, MPD, t(12;13)(p13;q12), hypereosinophilia



The myeloproliferative disorders (MPD) are chronic malignant conditions characterized by the clonal expansion of hematopoietic cells from one or more myeloid lineages.1, 2 Constitutive activation of mutant tyrosine kinases (TK), which have been identified by cloning recurrent chromosomal translocation breakpoints, is a key element in the pathogenesis of MPD,3 making therapy targeted to specific TKs a crucial new approach to treatment of MPD. For example, imatinib is effective treatment in nine of 11 patients with hypereosinophilic syndrome, recently categorized as a rare MPD of unknown etiology.3, 4 Among them, however, only five had a FIP1L1-PDGFRA fusion, a well-known target for imatinib. Moreover, two of the 11 patients treated with imatinib exhibited no responses. Further information is needed for understanding molecular pathogenesis of MPD.

The ETV6 (ETS variant gene 6) gene, also known as TEL (translocation, ETS, leukemia), located at band 12p13, is frequently rearranged by chromosomal translocations in a broad spectrum of hematologic malignancies. A member of the ETS family of transcription factors, ETV6 is characterized by two functional domains, a C-terminal ETS DNA-binding domain and a N-terminal helix–loop–helix (HLH) oligomerization domain that plays a role in modulating transcriptional activities.5 Since ETV6 was cloned as a partner gene of platelet-derived growth factor receptor beta (PDGFRB) in a t(5;12)(q33;p13) translocation,6 it has been found to fuse to a growing number of different genes such as ABL,7 JAK2,8 NTRK3,9 AML1,10 MDS1/EVI1,11 MN1,12 CDX2,13 STL,14 BTL,15 ARG,16 ACS2,17 HLXB9,18 PAX519 and TTL.20 Among them, several are TK genes. In ETV6/TK fusion proteins, the HLH domain of ETV6 functions as a homodimerization motif that activates the TK domain of the partner genes,21 resulting in a definite phenotype depended on the partners. In MPD, ETV6 was reported to fuse to several TK genes such as ABL, JAK2 and PDGFRB.1

The FLT3 (FMS-like TK3) gene, located at band 13q12, is a member of the receptor TK (RTK) subclass III family genes. It is one of the most frequently mutated genes in hematologic malignancies.22 The most common mutation of FLT3 is an internal tandem duplication in exons 14 and 15, whereas other mutations have also been found at and around codon 835 of exon 20. These activating mutations are found in approximately one-third of patients with acute myeloid leukemia (AML) and in a smaller subset of patients with acute lymphoblastic leukemia (ALL)22, 23, 24 promoting constitutive RTK activity in the absence of ligand and leading to aberrant signaling including strong activation of signal transducers and activators of transcription 5,25, 26 although detailed signaling downstream following wild-type FLT3 activation has not yet been fully clarified.27 Many questions with regards to the biology of FLT3 and its role in leukemogenesis remain to be clarified. Despite its highly frequent mutations, FLT3 has never been reported to fuse to any partner genes. In this report, we show that FLT3 was fused to ETV6 in a patient with MPD and a t(12;13)(p13;q12), resulting in the expression of a fusion protein, which contained the HLH domain of ETV6 and the juxtamembrane (JM) as well as TK domain of FLT3. This is the first report showing that FLT3 was involved in leukemogenesis through forming a fusion gene.


Materials and methods


A 68-year-old Japanese woman was admitted and treated with interferon alpha (IFNalpha) at a hospital under the tentative diagnosis of chronic myeloid leukemia in March 2002. As her leukocytosis was not improved with IFNalpha, she was referred to Tokyo Women's Medical University Hospital for further evaluation and therapy in May 2002. The hematologic findings showed white blood cell (WBC) counts 33.6 times 109/l (3% myelocytes, 33.5% mature neutrophils, 54% eosinophils, 1.5% basophils, 2% monocytes and 7.0% lymphocytes), hemoglobin (Hb) 119 g/l and platelet counts 5450 times 109/l. The bone marrow (BM) was marked hypercellular with 0.9% blasts, 6.0% promyelocytes, 15.6% myelocytes, 8.1% immature eosinophils and 19.2% mature eosinophils. Karyotype of BM cells was 46,XX,t(12;13)(p13.1;q12.3–13)[28]/46,XX[2]. Southern blot analysis showed no rearrangement of BCR/ABL. As serum immunoglobulin (Ig)E was not elevated and parasite infection or allergic disorders, which would cause eosinophilia, were ruled out, she was suspected to have MPD with hypereosinophilia. Hydroxyurea was initiated in June 2002, and WBC was controlled under 15 times 109/l for more than a year. In September 2003, as WBC was increased to more than 200 times 109/l, imatinib (200 mg/day) was started. However, her leukocytosis was refractory to even to increased imatinib (400 mg/day). In October 2003, WBC was 293 times 109/l with 76% eosinophils. Hydroxyurea was re-started with no response. She moved to another hospital owing to her hope. The hematologic findings showed WBC counts 394.5 times 109/l, Hb 75 g/l and platelet counts 33 times 109/l. Although daunorubicine and cytarabine were initiated, she died of hemorrhage in December 2003. The patient's BM sample was obtained with informed consent.

Fluorescence in situ hybridization analysis

Bacterial artificial chromosome (BAC) probes located at 12p13.3 and 13q12.3, and LL12NCO1 ETV6 cosmid probes (184C4 and 148B6) located within the ETV6 (generously gifted by Professor Marynen), were used for Fluorescence in situ hybridization (FISH). Metaphase cell slides were prepared using a standard air-dry method. The probes, labeled with biotin-16-dUTP by using nick translation, were hybridized to metaphase cells by using a standard method. FISH signals were analyzed as described previously.28

PCR-based methods

Sequences of all polymerase chain reaction primers used are listed in the Table 1.

3' Rapid amplification of cDNA ends (3'RACE) was performed using forward primers located in ETV6 exons 4 and 5 and total RNA extracted from a cryopreserved BM sample, without separation of granulocytes from eosinophils, using ISOGEN reagent (Nippon Gene, Toyama, Japan). First-strand cDNA was synthesized from 2.5 mug of total RNA with Superscript first-strand synthesis system (Invitrogen, Carlsbad, CA, USA) with primer Q0–Q1–dT18 (5'-CCAGTGAGCAGAGTGACGAGGACTCGAGCTCAAGCTTTTTTTTTTTTTTTTTT-3'). The first PCR was performed with primer TEL.E4F and outer adaptor Q0 (CCAGTGAGCAGAGTGACG) using the first-strand cDNA as the template. Diluted product of the first PCR was used for the second nested PCR with primer TEL.E5F and inner adaptor Q1 (GAGGACTCGAGCTCAAGC). The thermal cycling profile was: 95°C for 5 min, followed by 35 cycles of 95°C for 30 s, 55°C for 30 s, 72°C for 2 min and a final extension at 72°C for 10 min. PCR products were separated by electrophoresis on a 1.5% agarose gel, cloned into pT7Blue T-Vector (Novagen, Darmstadt, Germany) and subjected to DNA sequencing.

Reverse transcriptase-PCR (RT-PCR) was performed to detect wild-type ETV6 and FLT3, ETV6/FLT3 and reciprocal FLT3/ETV6 transcripts. Total RNA (1 mug) was reverse transcribed with Moloney murine leukemia virus reverse transcriptase (MMLV-RT; Stratagene, La Jolla, CA, USA) and random hexamers. Seminested PCR was performed to demonstrate the presence of ETV6/FLT3 transcripts by using primers TEL.E4F, TEL.E5F and FLT3.E20R. To confirm full-length ETV6/FLT3 transcripts, we used primer sets TEL.1F1 and FLT3. E24R1 for the first PCR, and TEL.E1F2 and FLT3.E24R2 for the second nested PCR. To detect the reciprocal FLT3/ETV6 transcripts, primer sets FLT3.E12F and TEL.E8R in combination with FLT3.E13F and TEL.E7R were used. Wild-type ETV6 was amplified by TEL.E1F3 and TEL.E8R primers, whereas primers FLT3.E3F and FLT3.E22R were used to analyze wild-type FLT3.

To identify the exact breakpoints in ETV6 and FLT3, we performed genomic PCR using forward primer in ETV6 intron 4 (TEL.int4F) and reverse primer in FLT3 intron 14 (FLT3.int14R) for ETV6/FLT3 and forward primer in FLT3 exon 12 (FLT3.E12F) and reverse primer in ETV6 exon 6 (TEL.E6R) for FLT3/ETV6. The thermal cycling profile was the same as described above.

DNA sequencing and homology search

The amplified products were sequenced directly or after subcloning into pT7Blue T-Vector with an ABI PRISM 310 (PE Applied Biosystems, Foster City, CA, USA). DNA sequences were analyzed using the NCBI Blast server.

Northern blot analysis

Total RNA (5 mug) was separated by electrophoresis on a 1% agarose/formaldehyde gel and transferred to Hybond-N membrane (Amersham, Buckinghamshire, UK). Hybridization was performed as described previously16 with full-length ETV6 cDNA probe.

Plasmid construction

Full-length ETV6/FLT3 cDNAs were constructed from the patient sample by PCR using primers TEL.BamHI and FLT3. BamHI. The products were ligated into the BamHI site of pcDNA3.1(+) vector (Invitrogen) and sequenced to verify the open-reading frames. Constructions of pcDNA3.1(+)-EF1, -EF6 and -EF7 were generated.

Cell culture and transfection

Ba/F3 cells, maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum and 1 ng/ml IL-3, were transfected with pcDNA3.1(+)-ETV6/FLT3 constructs by lipofectamine (Invitrogen) according to the manufacturer's instructions. Transfected cells were selected in medium containing IL-3 and 1 mg/ml G418 (GIBCO BRL, Gaithersburg, MD, USA) for 2 weeks and then subjected to limiting dilution to isolate clones. Expressions of ETV6/FLT3 proteins were documented (Figure 5a).

Figure 5.
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

ETV6/FLT3 rendered Ba/F3 cells IL-3 independent. (a) Expression of EF1, EF7 and EF6 in Ba/F3 cells. Total cell lysates were immunoprecipitated by TEL N-19, followed by immunoblot analysis with Flt3/Flk2 C-20 (for EF1 and EF7) or TEL N-19 (for EF6). (b) EF1 was a constitutively tyrosine-phosphorylated protein. Total cell lysates were immunoprecipitated by TEL N-19, followed by immunoblot analysis with 4G10. (c) EF1 conferred IL-3 independence to Ba/F3 cells. Ba/F3 cells expressing the indicated proteins were cultured in IL-3-free medium at 1 times 105 cells/ml, and cell numbers were counted at the indicated time points. The data presented are the average of three separate experiments.

Full figure and legend (62K)

In survival experiment, cells were seeded at 1 times 104 cells/well (1 times 105/ml) in 96-well plate in the medium in the presence or absence of 1 ng/ml IL-3. The number of viable cells was determined every 24 h by trypan blue exclusion.

Western blot analysis and immunoprecipitation

Total cell lysate was isolated from the patient's BM cells or cell lines using RIPA buffer (Santa Cruz Biotechnology, Santa Cruz, CA, USA) according to the manufacturer's instructions. Immunoprecipitation (IP) was performed as described previously.29 Protein samples were separated on a 12% sodium dodecyl sulfate (SDS)-polyacrylamide gel and transblotted onto Hybond-P membranes (Amersham). After appropriate blocking with 5% bovine serum albumin (Sigma, St Louis, MO, USA) in phosphate-buffered saline-Tween 20 (pH 7.5) (PBST: 0.1 M NaCl, 0.08 M Na2HPO4, 0.02 M NaH2PO4dot2H2O, 0.1% Tween 20), immunoblot analysis was performed with specific primary antibodies Flt3/Flk2 C-20 (Santa Cruz) anti-FLT-3 (F-0550; Sigma, St Louis, MO, USA), TEL N-19 (Santa Cruz), and TEL C-20 (Santa Cruz). Membranes were washed and incubated with horseradish peroxidase (HRP)-conjugated secondary antibody (anti-goat IgG HRP or anti-rabbit IgG HRP; Santa Cruz). Immunodetection was performed by using enhanced chemiluminescence (ECL; Amersham). For the detection of tyrosine-phosphorylated proteins, the mouse monoclonal antibody against phosphotyrosine (4G10; Upstate Biotechnology, Lake Placid, NY, USA) was used.



Rearrangement of ETV6 and FLT3 in the patient with MPD and a t(12;13)(p13;q12)

FISH analysis using CITD-2635D8, which covers the last six exons of ETV6, showed split signals between the der(12) and der(13), suggesting the break of ETV6 in this t(12;13) (Figure 1a). Moreover, the signals obtained from 184C4 containing ETV6 exons 3–5 were found on the der(13), whereas the signals from 148B6 containing ETV6 exon 8 were on the der(12), indicating that the 12p13 breakpoint occurred between exons 5 and 8 of ETV6. The signals on the normal chromosome 12 were always observed with every BAC and cosmid probes, suggesting that ETV6 was not deleted on the normal allele.

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

FISH and Northern blot analyses showed the rearrangement of ETV6, FLT3 and CDX2. (a) FISH detected split signals between the der(12) and der(13) with CITD-2635D8 probe. Signals were found on the der(13) and normal 12p with 184C4 probe, and on the der(12) and normal 12p with 148B6 probe, indicating that the breakpoint at 12p13 occurred between ETV6 exons 5 and 8. Signals on the normal 12p were always observed with every BAC and cosmid probes, suggesting that the non-translocated allele of ETV6 was not deleted. (b) With RP11-9D14 and RP11-153M24 probes, FISH study showed split signals between the der(12) and der(13) in about a half number of metaphases analyzed, suggesting the coexistence of two clones, one with split FLT3 and the other with suspicion of split CDX2. (c) Northern blot analysis using a full-length ETV6 cDNA probe showed an abnormal transcript in the patient's leukemic cells in addition to a wild-type ETV6 transcript of 6.5 kb, whereas wild-type ETV6 transcripts of 6.5 and 4.5 kb were demonstrated in TCC-S cells used as a control.

Full figure and legend (142K)

ETV6 was reported to fuse to CDX2 in a patient with AML and a t(12;13)(p13;q12).13 To check whether our patient had a different fusion partner of ETV6, we first carried out FISH by using the contig BAC probes surrounding CDX2 (Figure 1b). In an approximately half number of analyzed metaphases, split signals were observed between the der(13) and der(12) with RP11-153M24, which spans CDX2 and the 3' portion of FLT3. Interestingly, RP11-9D14 which entirely belongs to FLT3 showed also split signals in an about half number of metaphases. Therefore, FISH results highly suggested the coexistence of two clones in the patient, one with split FLT3 and the other with suspicion of split CDX2.

Northern blot analysis hybridized with full-length ETV6 cDNA demonstrated wild-type ETV6 transcripts of 6.5 and 4.5 kb in control cell line, TCC-S.30 On the other hand, t(12;13) leukemic cells showed an abnormal transcript between 6.5 and 4.5 kb, besides a 6.5-kb wild-type ETV6 transcript, further indicating the rearrangement of ETV6 (Figure 1c).

Cloning and identification of ETV6 fusion cDNA

To identify a fusion partner of ETV6, we performed 3'RACE-PCR with nested primers located in ETV6 exons 4 and 5. PCR products were subcloned into pT7Blue T-Vector and 50 resultant clones were analyzed. Sequence analyses detected novel sequences fused to ETV6. BLAST search revealed that the novel sequences were identical to FLT3 (GenBank accession no. NM_004119), which maps to band 13q12. In the ETV6/FLT3 transcripts, nucleotide (nt) 1178 within ETV6 exon 5 (GenBank accession no. BC043399) was fused to nt 1778 within FLT3 exon 14, preserving the reading frame from ETV6 through FLT3 (data not shown). We were unable to detect any clones that included CDX2 sequences.

Detection of the ETV6/FLT3 and FLT3/ETV6 transcripts

To check whether the ETV6/FLT3 transcript was indeed expressed in the patient's cells, we performed seminested RT-PCR by using forward primers in ETV6 exons 4 and 5 (TEL.E4F and TEL.E5F) and a reverse primer in FLT3 exon 20 (FLT3.E20R). As shown in Figure 2a, four smaller bands besides a product with an expected size (EF1, 1241 bp) were observed. Sequence analysis revealed that they were five alternatively spliced ETV6/FLT3 transcripts (EF1–EF5) (Figure 2d). The EF1 had the identical junction between ETV6 and FLT3 as the 3'RACE clones did. The other four smaller bands were generated from the junctions between nt 953 of ETV6 and nt 2099 of FLT3 (EF2, 695 bp), nt 925 and nt 2207 (EF3, 559 bp), nt 1113 and nt 2416 (EF4, 538 bp) and nt 805 and nt 2136 (EF5, 510 bp), respectively. Among them, EF1 and EF2 are in-frame fusion transcripts.

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Identification of ETV6/FLT3 and reciprocal FTL3/ETV6 transcripts and genomic junctions by PCR and DNA sequencing. (a) Seminested RT-PCR by using forward primers in ETV6 exons 4 and 5 and reverse primer in FLT3 exon 20 detected several ETV6/FLT3 bands ranging from 510 to 1241 bp in the patient's leukemic cells (EF1–EF5), but not in TCC-S cell line used as a control. (b) Nested RT-PCR by using forward primers in ETV6 exon 1 and reverse primers in FLT3 exon 24 showed three types of ETV6/FLT3 fusion transcripts, ranging from 1580 to 2234 bp (EF1, EF6 and EF7). (c) Nested RT-PCR by using forward primers in FLT3 exons 12 and 13 and reverse primers in ETV6 exons 7 and 8 detected two reciprocal FLT3/ETV6 transcripts. (d) Nucleotide sequences and deduced amino acids around the junctions of seven types of ETV6/FLT3 transcripts and one reciprocal FLT3/ETV6 transcript are shown. Another type of reciprocal FLT3/ETV6 transcript had an additional 12-bp sequence insertion at the junction. An arrow indicates junction. (e) Nucleotide sequences around the genomic breakpoints are shown. On the der(13), A of ETV6 exon 5 (nt 1178) was fused to G of FLT3 exon 14 (nt 1778), whereas on the der(12), T of FLT3 intron 13 was fused to A of ETV6 intron 5, indicating that the t(12;13) in our patient was not a perfectly reciprocal translocation.

Full figure and legend (224K)

To determine whether the ETV6/FLT3 transcripts actually contained the sequences encoding the functional domains of ETV6 and FLT3, we performed RT-PCR to amplify the coding region of this fusion gene by using forward primers in ETV6 exon 1 (TEL.E1F1 and TEL.E1F2) and reverse primers in FLT3 exon 24 (FLT3.E24R1 and FLT3.E24R2). Three types of ETV6/FLT3 transcripts were found (Figure 2b). Sequence analyses confirmed that the longest band (2234 bp) was the full-length EF1, a fusion between ETV6 from the start codon ATG (nt 275–279) to nt 1178 and FLT3 from nt 1778 to the stop codon TAG (nt 3037–3039), without any mutations or deletions. Surprisingly, the other smaller bands did not have the junctions found in EF2 to EF5 transcripts. Instead, the 1644-bp band (EF6) had the same junction as found in EF1 with the lack of FLT3 exons 16–20, presumably owing to alternative splicing. The remaining 1580-bp band (EF7) was generated from an in-frame fusion between ETV6 exon 4 (nt 737) and FLT3 exon 16 (nt 2000) (Figure 2d).

Next, to detect the reciprocal FLT3/ETV6 transcript, we performed nested RT-PCR by using forward primers in FLT3 exons 12 and 13 (FLT3.E12F and FLT3.E13F) and reverse primers in ETV6 exons 7 and 8 (TEL.E7R and TEL.E8R). We obtained two bands, a smaller band with a fusion between FLT3 exon 13 and ETV6 exon 6 (FE, 275 bp) (Figure 2c and d), and a large band (287 bp) with an additional 12-bp sequence inserted between these same exons. As a result of BLAST search, the 12-bp sequence turned out to be the first 12 nucleotides of FLT3 intron 13. Both FLT3/ETV6 transcripts were out-of-frame.

Detection of the genomic junction

PCR using genomic DNA with ETV6 intron 4 forward primer (TEL.int4F) and FLT3 intron 14 reverse primer (FLT3.int14R) amplified a 695-bp product (data not shown). This was a fusion between A of ETV6 exon 5 (corresponding to nt 1178 of cDNA) and G of FLT3 exon 14 (nt 1778 of cDNA) on the der(13) (Figure 2e). On the other hand, PCR using FLT3 exon 12 forward primer (FLT3.E12F) and ETV6 exon 6 reverse primer (TEL.E6R) amplified a 1880-kb product resulted from a fusion between T of FLT3 intron 13 and A of ETV6 intron 5 on the der(12). Therefore, the t(12;13) in our patient turned out to be a non-perfect reciprocal translocation.

Expression of wild-type ETV6 and FLT3 transcripts

To check expression of ETV6 from the normal allele, we performed RT-PCR by using primers in ETV6 exons 1 and 8, followed by direct DNA sequencing. We detected a full length of wild-type ETV6, in addition to two alternative slicing transcripts, one lacking ETV6 exon 5 and the other lacking exons 5 and 7 (data not shown). A full length of wild-type FLT3 was also detected (data not shown).

Expression of ETV6/FLT3 fusion proteins

Each of the ETV6/FLT3 fusion transcripts was expected to encode a fusion protein, which has the intact HLH domain of ETV6 at N-terminus6 (Figure 3). The longest in-frame fusion product (EF1) contained almost complete JM domain and two intact TK domains of FLT3 at C-terminus.31, 32 The other two in-frame products (EF2 and EF7) lacked JM and 5' part of TK1 domain, whereas the alternative isoform of EF1 (EF6) retained JM and 5' part of TK1 domain. Reciprocal transcripts (FE) were expected to encode truncated FLT3 without the JM and TK domains.

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Structures of the putative fusion proteins. (a) Schematic structures of wild-type ETV6 and FLT3. Solid arrows indicate breakpoints in each protein. HLH, helix–loop–helix oligomerization domain; ETS, ETS DNA-binding domain; EC, extracellular domain; JM, juxtamembrane domain; and TK, tyrosine kinase domain. (b) Schematic presentation of seven ETV6/FLT3 isoforms generated through alternative splicing, and FLT3/ETV6. Each of the ETV6/FLT3 isoforms has a complete HLH domain of ETV6 at N-terminus. However, with respect to ETV6/FLT3 isoforms, only EF1 retained both JM and TK domains of FLT3 at C-terminus.

Full figure and legend (96K)

Western blot analysis by using an antibody directed against C-terminus of FLT3 (Flt3/Flk2 C-20) detected several strongly abnormal bands in the patient's cells, in addition to a weaker band corresponding to the normal FLT3 (150 kDa) (Figure 4a). Doublet of about 83 kDa would be products translated from EF1 using an additional nearby in-frame internal start codon in ETV6 exon 3,29, 33 whereas a strong band of about 60 kDa would be product corresponding to EF7 (58 kDa), because a full-length EF2 transcript was not detected by RT-PCR. Control TCC-S cells showed two bands of normal FLT3 of 130 and 150 kDa. The antibody directed against N-terminus of ETV6 (TEL N-19) recognized another 44-kDa band besides two abnormal bands detected by Flt3/Flk2 C-20 (Figure 4b). This would correspond to EF6 of 44 kDa. It is likely that the 44-kDa EF6 band was not observed by Flt3/Flk2 C-20 because it lacked C-terminal FLT3. Therefore, at least three types of ETV6/FLT3 fusion proteins were detected in the leukemic cells.

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Western blot analysis showed the expression of ETV6/FLT3 fusion proteins in the patient's leukemic cells. (a) C-terminal FLT3 antibody (Flt3/Flk2 C-20) detected normal 130- and 150-kDa FLT3 expressions in TCC-S and t(12;13) leukemic cells. On the other hand, 83- and about 60-kDa abnormal proteins were observed only in t(12;13) leukemic cells, presumably corresponding to EF1, EF2 and EF7. (b) N-terminal ETV6 antibody (TEL N-19) confirmed the presence of these two abnormal proteins and also recognized another 44-kDa protein, corresponding to EF6. The C-terminal FLT3 antibody (Flt3/Flk2 C-20) could not recognize EF6, because EF6 did not retain the C-terminus of FLT3. A normal 57-kDa ETV6 protein was detected in K-562 cell line used as a control.

Full figure and legend (41K)

Antibodies reacting specifically with either the extracellular domain of FLT3 (anti-FLT-3) or the C-terminus of ETV6 (TEL C-20) failed to detect FLT3/ETV6 reciprocal product (data not shown), indicating that very small amount, if any, of the FLT3/ETV6 protein was expressed in the patient's cells. TEL C-20 was also unable to detect the endogenous 57-kDa ETV6 in the leukemic cells (data not shown), suggesting that a normal ETV6 protein was not translated in spite of the expression of a wild-type ETV6 transcript.34

ETV6/FLT3 protein is a constitutively activated TK

To examine whether ETV6/FLT3 proteins were constitutively activated in Ba/F3 cells, total cell lysates from cells cultured in IL-3-free medium were immunoprecipitated by TEL N-19 followed by immunoblot analysis with 4G10. An approximately 83-kDa band was detected in Ba/F3-expressing EF1, indicating that EF1 protein was tyrosine-phosphorylated (Figure 5b). No band was observed at around 44 and 58 kDa corresponding to EF6 and EF7.

Transforming activity of ETV6/FLT3

The murine hematopoietic cell line Ba/F3 is IL-3 dependent. When cultured under the presence of IL-3, all stably transfected Ba/F3 cells grew at a comparable rate (data not shown). In the absence of IL-3, Ba/F3 cells expressing EF1 continued to proliferate at a comparable rate to cells grown in IL-3-containing medium, whereas cells expressing EF6 or EF7 did not proliferate and died within 3 days (Figure 5c). The result was in good agreement with the fact that only EF1 was tyrosine-phosphorylated (Figure 5b) and indicated that EF1 was an oncoprotein.



We have demonstrated that a fusion gene of ETV6 and FLT3 was formed in a patient with MPD and a t(12;13)(p13;q12). Reciprocal ETV6/FLT3 and FLT3/ETV6 transcripts were detected by RT-PCR; however, in protein level, only ETV6/FLT3 was expressed in the patient's leukemic cells. In an AML patient with the same t(12;13)(p13;q12) translocation, even though an ETV6/CDX2 fusion gene was detected, the ectopic expression of CDX2 instead of ETV6/CDX2 chimeric gene was proved to induce myeloid leukemogenesis.13, 35 In our patient, a half number of metaphases examined showed split FISH signals by using RP11-153M24, which includes CDX2 (Figure 1b). However, we were unable to detect either ETV6/CDX2 or wild-type CDX2 transcripts with 3'RACE and RT-PCR (data not shown) despite several attempts, suggesting that CDX2 was not involved in the disorder found in our patient.

FIP1L1/PDGFRA fusion, a result of micro-deletion on chromosome 4,3 has been identified as a therapeutic target for imatinib in hypereosinophilic syndrome. However, around one-third of responding patients lacked the fusion gene, suggesting genetic heterogeneity. More recently, a gain-of-function of JAK2 (V617F) has been found in a high proportion of patients with MPD.36 Using PCR and DNA sequencing, with the same primers and conditions as published, we excluded the presence of those two genetic aberrances in our patient (data not shown).

Very recently, three tyrosines among 22 cytoplasmic tyrosine residues of murine mutant Flt3 has been proven to be critical for the constitutive activation of that mutant Flt3.37 However, very little is known about the contribution of each tyrosine residue to human FLT3 signal transduction. It is reported that the JM domain of RTK subclass III family genes not only has an autoinhibitory effect on the catalytic activity of the kinase domain but also plays a more active role in leukemogenesis. For instance, not any other tyrosines in TK domain in ETV6/PDGFRB, but tyrosines 579 and 581 of the JM domain were proved to determine myeloproliferative phenotype in mouse.38 As EF1 contained HLH domain of ETV6 and TK domain and almost complete JM domain of FLT3, its FLT3 TK is expected to be constitutively activated through dimerization by the HLH domain exhibiting the oncogenic property, similar to a mechanism described by Tse et al.33 However, the oncogenic ability of EF6 and EF7 is questionable, because they lack almost whole part of TK1 and TK2 domains or JM domain, respectively. A similar, but not identical to EF6, chimeric fusion between 154 first amino acids of ETV6 and the JM and TK1 domains of FLT3 failed to transform Ba/F3 cell (Tse et al33). As expected, we have confirmed that EF1 is a constitutively activated TK that could confer IL-3-independent growth to Ba/F3 cells, whereas EF6 and EF7 did not (Figure 5b and c). Thus, it is likely that among several ETV6/FLT3 proteins, EF1 would play a crucial role in the leukemogenesis of the patient.

Of note is that none of these ETV6/FLT3 proteins retained tyrosine 314 in ETV6 exon 5, which is a direct binding site for Grb2 and required for the induction of MPD by ETV6/ABL.39 Grb2 is one of various signaling intermediates found in FLT3 signaling cascade.22 Thus, the myeloproliferative phenotype in our patient may be mainly determined by the behavior of aberrant FLT3.

On the other hand, we were unable to detect normal ETV6 protein in the patient' cells (data not shown); nevertheless, a full-length normal ETV6 transcript was found by RT-PCR. Because ETV6 is a transcription suppressor,5 the absence of normal ETV6 protein may favor oncogenic process as a secondary genetic event.34 Finally, because these ETV6/FLT3 proteins were simultaneously found in leukemic cells, the effect of their co-expression should be evaluated with respect to leukemogenesis and induction of MPD.

In summary, we have demonstrated that FLT3 was involved in leukemogenesis as a fusion partner of ETV6 in a patient with MPD and a t(12;13)(p13;q12). This is the first report of an FLT3 fusion gene involved in leukemia. Functional study of the ETV6/FLT3 proteins will help us to further understand the role of FLT3 in leukemogenesis, and to develop target therapy focusing on FLT3.



  1. Cross NC, Reiter A. Tyrosine kinase fusion genes in chronic myeloproliferative diseases. Leukemia 2002; 16: 1207–1212. | Article | PubMed | ISI | ChemPort |
  2. Cortes J, Kantarjian H. Beyond chronic myelogenous leukemia: potential role for imatinib in Philadelphia-negative myeloproliferative disorders. Cancer 2004; 100: 2064–2078. | Article | PubMed | ChemPort |
  3. Cools J, DeAngelo DJ, Gotlib J, Stover EH, Legare RD, Cortes J et al. A tyrosine kinase created by fusion of the PDGFRA and FIP1L1 genes as a therapeutic target of imatinib in idiopathic hypereosinophilic syndrome. N Engl J Med 2003; 348: 1201–1214. | Article | PubMed | ISI | ChemPort |
  4. Weller PF, Bubley GJ. The idiopathic hypereosinophilic syndrome. Blood 1994; 83: 2759–2779. | PubMed | ChemPort |
  5. Fears S, Gavin M, Zhang DE, Hetherington C, Ben-David Y, Rowley JD et al. Functional characterization of ETV6 and ETV6/CBFA2 in the regulation of the MCSFR proximal promoter. Proc Natl Acad Sci USA 1997; 94: 1949–1954. | Article | PubMed | ChemPort |
  6. Golub TR, Barker GF, Lovett M, Gilliland DG. Fusion of PDGF receptor beta to a novel ets-like gene, tel, in chronic myelomonocytic leukemia with t(5;12) chromosomal translocation. Cell 1994; 77: 307–316. | Article | PubMed | ISI | ChemPort |
  7. Golub TR, Goga A, Barker GF, Afar DE, McLaughlin J, Bohlander SK et al. Oligomerization of the ABL tyrosine kinase by the Ets protein TEL in human leukemia. Mol Cell Biol 1996; 16: 4107–4116. | PubMed | ISI | ChemPort |
  8. Peeters P, Raynaud SD, Cools J, Wlodarska I, Grosgeorge J, Philip P et al. Fusion of TEL, the ETS-variant gene 6 (ETV6), to the receptor-associated kinase JAK2 as a result of t(9;12) in a lymphoid and t(9;15;12) in a myeloid leukemia. Blood 1997; 90: 2535–2540. | PubMed | ISI | ChemPort |
  9. Knezevich SR, McFadden DE, Tao W, Lim JF, Sorensen PH. A novel ETV6-NTRK3 gene fusion in congenital fibrosarcoma. Nat Genet 1998; 18: 184–187. | Article | PubMed | ISI | ChemPort |
  10. Romana SP, Mauchauffe M, Le Coniat M, Chumakov I, Le Paslier D, Berger R et al. The t(12;21) of acute lymphoblastic leukemia results in a tel-AML1 gene fusion. Blood 1995; 85: 3662–3670. | PubMed | ISI | ChemPort |
  11. Peeters P, Wlodarska I, Baens M, Criel A, Selleslag D, Hagemeijer A et al. Fusion of ETV6 to MDS1/EVI1 as a result of t(3;12)(q26;p13) in myeloproliferative disorders. Cancer Res 1997; 57: 564–569. | PubMed | ISI | ChemPort |
  12. Buijs A, Sherr S, van Baal S, van Bezouw S, van der Plas D, Geurts van Kessel A et al. Translocation (12;22) (p13;q11) in myeloproliferative disorders results in fusion of the ETS-like TEL gene on 12p13 to the MN1 gene on 22q11. Oncogene 1995; 10: 1511–1519. | PubMed | ISI | ChemPort |
  13. Chase A, Reiter A, Burci L, Cazzaniga G, Biondi A, Pickard J et al. Fusion of ETV6 to the caudal-related homeobox gene CDX2 in acute myeloid leukemia with the t(12;13)(p13;q12). Blood 1999; 93: 1025–1031. | PubMed | ISI | ChemPort |
  14. Suto Y, Sato Y, Smith SD, Rowley JD, Bohlander SK. A t(6;12)(q23;p13) results in the fusion of ETV6 to a novel gene, STL, in a B-cell ALL cell line. Genes Chromosomes Cancer 1997; 18: 254–268. | Article | PubMed | ChemPort |
  15. Cools J, Bilhou-Nabera C, Wlodarska I, Cabrol C, Talmant P, Bernard P et al. Fusion of a novel gene, BTL, to ETV6 in acute myeloid leukemias with a t(4;12)(q11-q12;p13). Blood 1999; 94: 1820–1824. | PubMed | ISI | ChemPort |
  16. Iijima Y, Ito T, Oikawa T, Eguchi M, Eguchi-Ishimae M, Kamada N et al. A new ETV6/TEL partner gene, ARG (ABL-related gene or ABL2), identified in an AML-M3 cell line with a t(1;12)(q25;p13) translocation. Blood 2000; 95: 2126–2131. | PubMed | ISI | ChemPort |
  17. Yagasaki F, Jinnai I, Yoshida S, Yokoyama Y, Matsuda A, Kusumoto S et al. Fusion of TEL/ETV6 to a novel ACS2 in myelodysplastic syndrome and acute myelogenous leukemia with t(5;12)(q31;p13). Genes Chromosomes Cancer 1999; 26: 192–202. | Article | PubMed | ISI | ChemPort |
  18. Beverloo HB, Panagopoulos I, Isaksson M, van Wering E, van Drunen E, de Klein A et al. Fusion of the homeobox gene HLXB9 and the ETV6 gene in infant acute myeloid leukemias with the t(7;12)(q36;p13). Cancer Res 2001; 61: 5374–5377. | PubMed | ISI | ChemPort |
  19. Cazzaniga G, Daniotti M, Tosi S, Giudici G, Aloisi A, Pogliani E et al. The paired box domain gene PAX5 is fused to ETV6/TEL in an acute lymphoblastic leukemia case. Cancer Res 2001; 61: 4666–4670. | PubMed | ISI | ChemPort |
  20. Qiao Y, Ogawa S, Hangaishi A, Yuji K, Izutsu K, Kunisato A et al. Identification of a novel fusion gene, TTL, fused to ETV6 in acute lymphoblastic leukemia with t(12;13)(p13;q14), and its implication in leukemogenesis. Leukemia 2003; 17: 1112–1120. | Article | PubMed | ChemPort |
  21. Tomasson MH, Williams IR, Hasserjian R, Udomsakdi C, McGrath SM, Schwaller J et al. TEL/PDGFbetaR induces hematologic malignancies in mice that respond to a specific tyrosine kinase inhibitor. Blood 1999; 93: 1707–1714. | PubMed | ISI | ChemPort |
  22. Stirewalt DL, Radich JP. The role of FLT3 in haematopoietic malignancies. Nat Rev Cancer 2003; 3: 650–665. | Article | PubMed | ISI | ChemPort |
  23. Gilliland DG, Griffin JD. The roles of FLT3 in hematopoiesis and leukemia. Blood 2002; 100: 1532–1542. | Article | PubMed | ISI | ChemPort |
  24. Armstrong SA, Mabon ME, Silverman LB, Li A, Gribben JG, Fox EA et al. FLT3 mutations in childhood acute lymphoblastic leukemia. Blood 2004; 103: 3544–3546. | Article | PubMed | ChemPort |
  25. Mizuki M, Fenski R, Halfter H, Matsumura I, Schmidt R, Muller C et al. Flt3 mutations from patients with acute myeloid leukemia induce transformation of 32D cells mediated by the Ras and STAT5 pathways. Blood 2000; 96: 3907–3914. | PubMed | ISI | ChemPort |
  26. Choudhary C, Schwable J, Brandts C, Tickenbrock L, Sargin B, Kindler T et al. AML-associated Flt3 kinase domain mutations show signal transduction differences compared with Flt3 ITD mutations. Blood 2005; 106: 265–273. | Article | PubMed | ISI | ChemPort |
  27. Choudhary C, Muller-Tidow C, Berdel WE, Serve H. Signal transduction of oncogenic Flt3. Int J Hematol 2005; 82: 93–99. | Article | PubMed | ChemPort |
  28. Xinh PT, Tri NK, Nagao H, Nakazato H, Taketazu F, Fujisawa S et al. Breakpoints at 1p36.3 in three MDS/AML(M4) patients with t(1;3)(p36;q21) occur in the first intron and in the 5' region of MEL1. Genes Chromosomes Cancer 2003; 36: 313–316. | Article | PubMed | ChemPort |
  29. Iijima Y, Okuda K, Tojo A, Tri NK, Setoyama M, Sakaki Y et al. Transformation of Ba/F3 cells and Rat-1 cells by ETV6/ARG. Oncogene 2002; 21: 4374–4383. | Article | PubMed | ChemPort |
  30. Van PN, Xinh PT, Kano Y, Tokunaga K, Sato Y. Establishment and characterization of A novel Philadelphia-chromosome positive chronic myeloid leukemia cell line, TCC-S, expressing P210 and P190 BCR/ABL transcripts but missing normal ABL gene. Hum Cell 2005; 18: 25–33. | PubMed |
  31. Rosnet O, Schiff C, Pebusque MJ, Marchetto S, Tonnelle C, Toiron Y et al. Human FLT3/FLK2 gene: cDNA cloning and expression in hematopoietic cells. Blood 1993; 82: 1110–1119. | PubMed | ISI | ChemPort |
  32. Griffith J, Black J, Faerman C, Swenson L, Wynn M, Lu F et al. The structural basis for autoinhibition of FLT3 by the juxtamembrane domain. Mol Cell 2004; 13: 169–178. | Article | PubMed | ISI | ChemPort |
  33. Tse KF, Mukherjee G, Small D. Constitutive activation of FLT3 stimulates multiple intracellular signal transducers and results in transformation. Leukemia 2000; 14: 1766–1776. | Article | PubMed | ISI | ChemPort |
  34. Barjesteh van Waalwijk van Doorn-Khosrovani S, Spensberger D, de Knegt Y, Tang M, Lowenberg B, Delwel R. Somatic heterozygous mutations in ETV6 (TEL) and frequent absence of ETV6 protein in acute myeloid leukemia. Oncogene 2005; 24: 4129–4137. | Article | PubMed | ChemPort |
  35. Rawat VP, Cusan M, Deshpande A, Hiddemann W, Quintanilla-Martinez L, Humphries RK et al. Ectopic expression of the homeobox gene Cdx2 is the transforming event in a mouse model of t(12;13)(p13;q12) acute myeloid leukemia. Proc Natl Acad Sci USA 2004; 101: 817–822. | Article | PubMed | ChemPort |
  36. Kralovics R, Passamonti F, Buser AS, Teo SS, Tiedt R, Passweg JR et al. A gain-of-function mutation of JAK2 in myeloproliferative disorders. N Engl J Med 2005; 352: 1779–1790. | Article | PubMed | ISI | ChemPort |
  37. Ishiko J, Mizuki M, Matsumura I, Shibayama H, Sugahara H, Scholz G et al. Roles of tyrosine residues 845, 892 and 922 in constitutive activation of murine FLT3 kinase domain mutant. Oncogene 2005; 24: 8144–8153. | Article | PubMed | ChemPort |
  38. Tomasson MH, Sternberg DW, Williams IR, Carroll M, Cain D, Aster JC et al. Fatal myeloproliferation, induced in mice by TEL/PDGFbetaR expression, depends on PDGFbetaR tyrosines 579/581. J Clin Invest 2000; 105: 423–432. | PubMed | ISI | ChemPort |
  39. Million RP, Harakawa N, Roumiantsev S, Varticovski L, Van Etten RA. A direct binding site for Grb2 contributes to transformation and leukemogenesis by the Tel-Abl (ETV6-Abl) tyrosine kinase. Mol Cell Biol 2004; 24: 4685–4695. | Article | PubMed | ISI | ChemPort |


We thank Professor Peter Marynen, Center of Human Genetics, University of Leuven, Belgium, for kindly providing the LL12NCO1 ETV6 cosmid probes.This work was supported by The Japan Foundation for the Promotion of International Medical Research Cooperation (JF-PIMRC).