Chronic Myeloproliferative Disorders

p210BCR-ABL reprograms transformed and normal human megakaryocytic progenitor cells into erythroid cells and suppresses FLI-1 transcription


The BCR-ABL oncoprotein exhibits deregulated protein tyrosine kinase activity and is implicated in the pathogenesis of Philadelphia chromosome (Ph)-positive human leukemias. Here, we report that ectopic expression of p210BCR-ABL in the megakaryoblastic Mo7e cell line and in primary human CD34+ progenitors trigger erythroid differentiation at the expense of megakaryocyte (MK) differentiation. Clonal culture of purified CD41+CD42 cells, a population highly enriched in MK progenitors, combined with the conditional expression of p210BCR-ABL tyrosine kinase activity by imatinib identified a true lineage reprogramming. In both Mo7e or CD41+CD42 cells transduced with p210BCR-ABL, lineage switching was associated with a downregulation of the friend leukemia Integration 1 (FLI-1) transcription factor. Re-expression of FLI-1 in p210BCR-ABL-transduced Mo7e cells rescued the megakaryoblastic phenotype. Altogether, these results demonstrate that alteration of signal transduction via p210BCR-ABL reprograms MK cells into erythroid cells by a downregulation of FLI-1. In addition, our findings underscore the role of kinases in lineage choice and infidelity in pathology and suggest that downregulation of FLI-1 may have important implications in CML pathogenesis.


The BCR-ABL oncogene is a fusion gene that characterizes the neoplastic clone of patients with chronic myeloid leukemia (CML).1 It encodes a 210 kDa oncoprotein (p210BCR-ABL)2 with constitutive tyrosine kinase activity.3 CML has a triphasic clinical course with a chronic phase followed by an accelerated phase and then a terminal phase, known as blastic crisis, in which myeloid or lymphoid blasts fail to differentiate. In the chronic phase, one of the major biological abnormalities is an increased production of morphologically mature granulocytes.4 This phase is also characterized by an abnormal, unregulated expansion of Philadelphia chromosome-positive (Ph+) primitive erythroid progenitors, both in bone marrow and in peripheral blood.5 In addition, occasional examples of Ph+ patients presenting with an erythrocytosis have been reported.6 Further evidence for an abnormal erythroid differentiation is suggested by an unexpectedly frequent occurrence of blast cells with erythropoietic and/or megakaryocytic surface determinants in non-lymphoid blast crises.4 Moreover, most cell lines established from CML blast crisis patients exhibit some erythroid and/or megakaryocytic characters.7, 8, 9 These results strongly point to a critical role of BCR-ABL in acting upon erythroid and/or megakaryocytic development.

p210BCR-ABL perturbs many intracellular signaling pathways involved in cell proliferation, inhibition of apoptosis and adhesion, partly via protein–protein interactions and partly via tyrosine phosphorylation of target substrates.10 BCR-ABL may also regulate the expression of proteins that are involved in lineage commitment and differentiation.11, 12 These effects may explain previous observations showing that, by analogy with v-raf or Emu-myc, p210BCR-ABL may alter cell fate to another lineage in several systems. Overexpression studies have shown that most cells transduced with BCR-ABL acquire an erythroid or myeloid phenotype.13 BCR-ABL initiates an erythroid program in myeloid leukemic HL-60 cells14 or induces a reversible block in the granulocytic program of differentiation in 32D cells.15 In addition, mice engrafted with bone marrow cells infected with a BCR-ABL-encoding retrovirus develop multiple hematopoietic disorders, including a clonal hyperproliferation of erythroid and/or mast cells.13, 16, 17 Moreover, erythroid cells are generated from human cord blood CD34+-derived granulopoietic progenitors transduced with p210BCR-ABL.18 The precise molecular mechanisms that sustain the effect of p210BCR-ABL on cell fate lineage remain poorly understood.

During the hierarchical development of myelopoiesis, a close developmental relationship exists between the erythroid and the megakaryocytic (MK) lineages,19 as demonstrated by the identification of a bipotent BFU-E/MK progenitor in the mouse20, 21 and human bone marrow.22, 23 Furthermore, as revealed by gene targeting studies in mice, the development and differentiation of the erythroid and MK lineages are controlled by common transcription factors.24, 25, 26 Among these transcription factors, FLI-1 (Friend leukemia integration 1) plays an essential role in both erythropoiesis and megakaryopoiesis.27

In the present study, we report a novel mechanism whereby BCR-ABL overexpressing cells may undergo abnormal differentiation. p210BCR-ABL expression in the Mo7e megakaryoblastic cell line and in human primary CD34+ progenitors leads to erythroid differentiation at the expense of MK differentiation in both cell systems. This lineage reprogramming is associated with FLI-1 downregulation at the transcriptional level. These results suggest that BCR-ABL may interfere with lineage commitment and may directly deregulate differentiation.

Materials and methods

Purification of CD34+ progenitors, cell lines and culture conditions

Mo7e, transduced-Mo7e cells and 293-EBNA cells were cultured as described.28 CD34+ cells were isolated from apheresis products of mobilized individuals and cultured in serum-free medium supplemented with recombinant human (rh) fetal liver tyrosine kinase 3 ligand (rhFLT3L, 10 ng/ml), rh thrombopoietin (rhTPO), rhIL-3 and stem cell factor (rhSCF).29 E/MK medium allowing erythroid and MK differentiation was the same cytokine-supplemented medium in which rh erythropoietin (rhEPO) (1 U/ml) was added. Cytokines were obtained from Amgen Corp (Thousand Oaks, CA, USA), except for rhTPO, provided by Kirin Brewery (Tokyo, Japan).

Antibodies and flow cytometry

Phycoerythrin (PE)-conjugated anti-CD31, anti-CD42 and anti-GPA; allophycocyanin (APC)-conjugated anti-CD41; PE- and APC-conjugated IgG1 and IgG2a control MoAbs were from BD Biosciences (Le Pont de Claix, France). After staining, cells were analyzed on a FACsort with the Cell Quest software package or sorted on a FacsDIVA flow cytometer equipped with an automatic cell deposition unit (BD Biosciences, Becton Dickinson Biosciences, Le pont de Claix, France).

Constructs and retrovirus production

MIGR1-empty (control-GFP), MIGR1-p210BCR-ABL (p210BCR-ABL-GFP), MIGR1-K1172Rp210BCR-ABL and the MSCV-Neo-p210BCR-ABL retrovirus were as described.28 FLI-1 cDNA from the pCS3-FLI-1 plasmid (gift from F. Moreau-Gachelin, INSERM U580, Paris, France) was subcloned into the MIGR1-empty (FLI-1-GFP) retroviral vector. VSV-G pseudotyped retroviruses were produced and titrated as described.29 Titers ranged from 1 to 5 × 106 infectious particles/ml.

Retroviral transduction

Mo7e cells were transduced (MOI=10) as described.28 CD34+ cells were prestimulated during 72 h under conditions supporting cycling and MK differentiation before being transduced once a day with fresh virus stock during 3 days (MOI=50) in the presence of 2 μg/ml hexadimethrine bromide (Sigma-Aldrich, Saint-Quentin Fallavier, France). Cells were then transferred to E/MK medium. When stated, cultures were performed in the presence of 5 μ M imatinib mesylate (Novartis, Bale, Switzerland) or the same volume of dimethyl sulfoxide (DMSO) as a control. Experiments using imatinib were carried out in the absence of rhSCF.

Single-cell cloning and progeny analysis

One day after transduction, the GFP+CD41+CD42 cells were sorted at one cell/well into 96-well-plate and cultured under E/MK condition. One week later, the differentiation of the progeny from each individual clone was monitored by morphology. Four different types of clones were observed: pure MK colonies containing 4–30 MKs and raising from MK progenitors, clones containing 1–3 MKs deriving from promegakaryoblasts (pro-MK) with little potency for division, pure erythroid clones and mixed erythro/MK clones. Wells containing numerous homogenous small cells were cytospun and stained with benzidine to ascertain their erythroid nature. Suspicious MK clones were ascertained by CD41 staining and flow cytometry analysis.


Five days after transduction with either control-GFP or p210BCR-ABL-GFP viruses, GFP+ Mo7e cells were sorted and total RNA from each sample and from the parental Mo7e cells (used as reference) was extracted using the SV total RNA isolation system (Promega Co., Madison, WY, USA). cDNA labeling, hybridization and data analyses were as described,30 using oligonucleotide microarrays (oligos longs 22K Agilent Technologies, Palo Alto, CA, USA). The ratio of the two fluorescence intensities (from the sample and from the reference) provided a quantitative measurement of the relative gene modulation.

Data availability

Data obtained from the microarray analysis have been submitted to Array Express at the European Bioinformatics Institute ( (accession number E-TABM-173). Array Express is a public repository for microarray data aimed at storing well-annotated data in accordance with ‘Microarray Gene Expression Data’ recommendations (

Real-time reverse transcription-PCR (RT-PCR) analysis

Primers and internal probes for FLI-1 amplification were as follows: forward 5′-IndexTermCCACCAACGAGAGGAGAGTCA-3′, reverse 5′-IndexTermCATTGCCTCACATGCTCCTG-3′, probe 5′-IndexTermCGTCCCCGCAGACCCCACAC-3′. For each fraction, the expression level of FLI-1 was expressed relative to β-actin (Perkin-Elmer).

Single-cell RT–PCR analysis

One day after transduction in the presence of imatinib or DMSO, the GFP+CD41+CD42 cells were single cell sorted. cDNAs were directly synthesized in each well. Each RT product was divided into eight wells before performing nested PCR as described previously.31

Western blotting

Protein levels of BCR-ABL or phosphotyrosine were determined as described.28 FLI-1 and β-actin were revealed with an anti-FLI-1 (sc-356; Santa Cruz Biotechnology, Santa Cruz, CA, USA) or an anti-β-actin mouse (Sigma) antibody, respectively.


Results of experimental points obtained from multiple experiments were reported as the mean ±s.d.


BCR-ABL downregulates MK cell surface markers in Mo7e cells

The Mo7e cell line was derived from the transformation of an MK progenitor and expresses antigen determinants specific to bipotent erythro/MK precursors and to MK cells as well as antigen determinants common to multilineages. Mo7e cells were transduced with MIGR vectors, encoding either GFP alone or p210BCR-ABL and GFP.28 The phenotype of these transduced cells was evaluated by flow cytometry analyses at different time points. When compared to control GFP+ cells as early as 4 days after the end of the transduction protocol, Mo7e p210BCR-ABL−GFP+ cells displayed downregulated expression of the MK antigens CD31, CD41 and CD42, as evidenced by significant reduction of the corresponding percentages of positive cells (Figures 1a and b) and mean fluorescence intensities (MFI) (Figures 1c and b). Conversely, but later on at day 10 after transduction, the percentage of Mo7e p210BCR-ABL-GFP+ cells stained for the erythroid marker glycophorin A and, even more strikingly, the MFI of this marker were significantly increased. Similar increases in the expression of the other erythroid markers CD36 and CD71 were also observed (data not shown). Interestingly, the downregulation of CD31 and CD42 was observed 24 h after the first exposure of the cells to p210BCR-ABL retrovirus, suggesting that a cell selection is not responsible for this effect. No change in cell surface markers was detected when Mo7e cells were transduced with a retrovirus encoding a kinase-inactive BCR-ABL mutant, indicating that the p210BCR-ABL-specific kinase activity was necessary to mediate its effect on the MK phenotype (data not shown).

Figure 1

p210BCR-ABL expression in Mo7e cells induces MK to erythroid conversion. (a) Mean percentage ±s.d. of Mo7e cells stained for MK and erythroid markers in transduced cells (GFP+ cells) (n=4). (b) Example of flow cytometric analysis for surface expression of each marker. Mo7e cells were transduced with either control-GFP (thin line) or p210BCR-ABL-GFP (thick line). Labeling with isotype control MoAb is shown as broken line. These data are representative of four experiments. (c) Mean fluorescence intensity (MFI) ±s.d. of each marker (n=4). The phenotype of Mo7e cells transduced with a retrovirus encoding a kinase-inactive BCR-ABL mutant was identical to Mo7e control-GFP (thin line).

p210BCR-ABL induces erythroid differentiation to the detriment of MK differentiation in MK progenitors

To determine the effect of p210BCR-ABL on erythroid differentiation in primary MK progenitors, CD34+ cells were transduced with the p210BCR-ABL-GFP or control-GFP retroviruses, grown for 4 days under E/MK condition culture and phenotyped by flow cytometry. The percentage of cells stained for MK markers (CD31, CD41, CD42) was significantly decreased in the GFP+-p210BCR-ABL-transduced cells compared to GFP+-control cells, whereas the percentage of GFP+ cells expressing GPA was markedly increased in the p210BCR-ABL-transduced cells (Figures 2a and b). Similar results were observed when cells were stained with the anti-CD36 antibody (data not shown). These changes in the expression of MK and erythroid markers are associated with marked morphological modifications, as evidenced by May–Grünwald Giemsa staining of sorted control-GFP+ or p210 BCR-ABL-GFP+ cells (Figure 2c).

Figure 2

Induction of erythroid differentiation in CD34+ cells by p210BCR-ABL. CD34+ cells were transduced with either the control-GFP or p210BCR-ABL-GFP viruses. Cells were cultured in E/MK medium during 4 days. (a) Example of flow cytometric profiles for isotype control, each MK or erythroid marker. These data are representative of three experiments. (b) Mean percentage ±s.d. of cells stained for MK and erythroid markers in transduced cells (only GFP+ cells) (n=3). (c) Representative morphological pictures of transduced cells (only GFP+ sorted cells) stained by May–Grünwald Giemsa (magnification × 63).

To determine whether the p210BCR-ABL effect was due to an instructive role of the oncoprotein on the erythroid commitment or to the promotion of a proliferative advantage to committed erythroid progenitors, we performed single-cell progeny analyses of CD34+CD41+CD42 cell populations in association with the p210BCR-ABL kinase inhibitor imatinib.32 As shown in Figure 3a, addition of imatinib completely abolished p210BCR-ABL kinase activity in CD34+ cells transduced with p210BCR-ABL-GFP. The GFP+CD41+CD42 cell population, highly enriched in MK progenitors, was sorted after transduction, seeded at one cell per well in the presence of imatinib or DMSO (Figure 3b). In each of the three repeated experiments, the cloning efficiency was reproductively around 85% at day 7 in both the control and p210BCR-ABL groups and the presence of imatinib only slightly reduced it (around 75%) in both groups. The lineage differentiation of at least 100 individual clones in each experiment and for each culture conditions was monitored by morphology, immunostaining, and erythroid maturation was further ascertained by benzidine coloration. In experiments performed with the GFP+CD41+CD42 population transduced with the control virus, 11±0.5% of the wells contained pure erythroid cells (ascertained by benzidine coloration), 52±1% contained numerous MKs originating from MK progenitors, 36±0.5% contained a few MK deriving from a pro-MK, but no mixed erythroid/MK clones were seen (Figure 3c). The addition of imatinib did not modify the overall repartition of erythroid and MK clones. In contrast, the GFP+CD41+CD42 population transduced with the p210BCR-ABL-expressing virus gave rise to 42±13% of the wells containing pure erythroid cells (fourfold increase, P=0.02), 12±7% containing a great number of MKs (fourfold decrease, P=0.03), 35±7% deriving from pro-MKs and 11±11% of the wells containing mixed erythro/MK clones. The presence of imatinib in p210BCR-ABL culture restored the proportion of erythroid and MK clones to a level comparable to control-GFP-transduced cells.

Figure 3

Analysis of individual clones of MK progenitors. CD34+ cells were transduced with the control-GFP or p210BCR-ABL-GFP retroviruses in the presence of imatinib or DMSO. (a) Western blotting analysis performed on GFP+-sorted cells, cultured during 4 days in E/MK medium with imatinib or DMSO. The expression level of p210BCR-ABL and its phosphorylation were determined using an anti-phosphotyrosine antibody (lower panel). The membrane was then stripped and reprobed with an anti-c-Abl antibody (upper panel). K562 cells were used as a positive control. The endogenous p145c-Abl was used as endogenous loading control. (b and c) Phenotypic analysis of individual clones derived from CD41+CD42 cells. After transduction, the GFP+ CD41+CD42 progenitors were sorted at one cell per well and cultured for 7 days in E/MK medium with imatinib or DMSO. (b) Schematic representation of the experimental plan. (c) Differentiation of each clone was determined using microscopic observation and confirmed by benzidine coloration and/or cytometric analysis as described in Materials and methods section. Percentage of clones was calculated on the number of wells showing a cell growth, *P<0.001 between control and p210BCR-ABL-cells (n=3).

These single-cell experiments conclusively demonstrate that p210BCR-ABL modifies the lineage output of individual cells belonging to the purified CD41+CD42 cell population and thus strongly argues for a change in lineage programming, occurring at the level of both bipotent progenitors and MK-committed cells. However, these experiments do not formally exclude that some of these changes may be due to the selective outgrowth of CD41+CD42 erythroid progenitors occurring during the 3 days transduction protocol. To address this question, we performed the same experiments with imatinib added only during the transduction procedure and removed after cell sorting for time culture. As observed previously, a significant decrease in colonies deriving from MK progenitors was seen (32±17.2% versus 51±7% in the continuous presence of imatinib, P<0.001), whereas the percentage of pure erythroid clones was still significantly increased (21±5.4% versus 8±5.7% in the continuous presence of imatinib, P<0.015) and the frequency of clones arising from pro-MK was not statistically modified. Altogether, these single-cell experiments strongly argue for a modulation mediated by the kinase activity of p210BCR-ABL on lineage programming and likely occurring both at the level of bipotent E/MK progenitors and MK-committed cells.

Gene expression profiling reveals upregulation of erythroid-specific genes and downregulation of MK-specific transcription genes

To investigate which target genes were modulated by p210BCR-ABL, the profiles of transcripts of Mo7e cells transduced with control- or p210BCR-ABL-GFP viruses were compared using microarrays. Microarray data were analyzed using the Rosetta Resolver software to identify the genes significantly modulated by p210BCR-ABL (P<0.05) after exclusion of genes giving a very low intensity (less than 200). Our analysis was focused on about 300 hematopoietic genes related to the erythroid, and MK lineages, and genes whose expression was modified between control-GFP+ and p210BCR-ABL-GFP+ cells are presented in Table I given as Supplementary Data 1. In p210BCR-ABL-GFP+ cells, transcripts of genes highly expressed during MK differentiation, such as c-mpl, α-6 integrin, glycoproteins IX, platelet factor 4 and thromboxane A synthase, were significantly reduced, whereas various genes involved in erythropoiesis, such as erythrocyte membrane proteins or globin chains, were upregulated. In addition, transcripts of several transcription factors controlling erythroid and/or MK differentiation were also modulated (microarray analysis, accession number E-TABM-173). Changes in gene expression were validated by real-time RT-PCR, only 3 days after transduction. In three independent experiments, the expression levels of the hematopoietic transcription factors, p45 NF-E2, EKLF and GATA-1, were upregulated (3-, 6- and 13-fold, respectively; Supplementary Data 2) whereas FLI-1 was downregulated 15-fold in p210BCR-ABL-GFP+ cells (Figure 4a). In addition, the FLI-1 proteins were significantly decreased by p210BCR-ABL (Figure 4b). Altogether, the results confirm that expression of p210BCR-ABL in the megakaryoblastic Mo7e cell line upregulates several genes known to be involved in erythropoietic differentiation and downregulates several genes known to be involved in megakaryocytic differentiation.

Figure 4

Downregulation of FLI-1 mRNA and protein expression in p210BCR-ABL-Mo7e cells. p210BCR-ABL-GFP+ Mo7e cells were sorted on day 3 post-transduction. (a) Relative FLI-1 mRNA level in p210BCR-ABL-GFP+ Mo7e compared to Mo7e parental cells determined by real-time RT-PCR. β-Actin mRNA was used as the internal control. (b) Western blotting for FLI-1 expression on Mo7e p210BCR-ABL-GFP+ and nontransduced populations, a β-actin antibody was used for loading control. These data are representative of three experiments.

Enforced FLI-1 expression in p210BCR-ABL-Mo7e cells restores expression of MK markers

Since FLI-1 expression appears to be required for MK development,33, 34 we postulated that the erythroid switch induced by p210BCR-ABL could be a consequence of the FLI-1 downregulation. To test this hypothesis, Mo7e cells were first transduced with a Neo or a Neo-p210BCR-ABL retrovirus. Neo-resistant cells were then transduced with MIGR vectors encoding either GFP alone (control-GFP) or FLI-1 and GFP (FLI-1-GFP). After transduction with the FLI-1-GFP construct, the FLI-1 protein was overexpressed in both Neo and Neo-p210BCR-ABL cells (Figure 5a). No differences in erythroid and MK marker expression were observed between Mo7e-Neo/control-GFP+ and Mo7e-Neo/FLI-1-GFP+. Their phenotype was also identical to that of Mo7e control-GFP shown in Figure 1, demonstrating that FLI-I overexpression did not further increase the MK phenotype of this cell line (data not shown). As expected, Neo-p210BCR-ABL/control-GFP+ cells displayed an erythroid phenotype, as observed by the relative percentage of the population expressing MK or erythroid markers and by the MFI of these markers. When FLI-1 was expressed in Neo-p210BCR-ABL cell line (Neo-p210BCR-ABL/FLI-1-GFP), the MK phenotype was restored (Figures 5b and c). Thus, restoring FLI-1 expression in Mo7e-p210BCR-ABL cells restored their MK differentiation.

Figure 5

FLI-1 re-expression in p210BCR-ABL-Mo7e cells restores the MK phenotype. Mo7e cells were transduced with Neo-control or Neo-p210BCR-ABL retrovirus and selected on neomycin resistance, then transduced by control-GFP or FLI-1-GFP viruses. (a) Upper panel, Western blotting for FLI-1 expression performed on GFP+ sorted cells, 3 days post-transduction. Lower panel, after stripping off the FLI-1 antibody, a β-actin antibody was used for loading control. (b) Mean percentage ±s.d. of Mo7e cells stained for MK and erythroid markers in double transduced cells (Neo-p210BCR-ABL/control-GFP or Mo7e-Neo-p210BCR-ABL/FLI-1-GFP, only GFP+ cells) 5 days post-transduction (n=4). (c) Examples of flow cytometric analysis of Mo7e-Neo-p210BCR-ABL/control-GFP (thick line) and Mo7e-Neo-p210BCR-ABL/FLI-1-GFP (thin line) cells for the expression of MK or erythroid markers, 5 days post-transduction. Labeling with control MoAb is shown by the broken line. These data are representative of four experiments. The mean fluorescence intensity (MFI) ±s.d. of each marker expression is shown at the bottom of histograms (n=4).

Erythroid differentiation induced by p210BCR-ABL expression in MK progenitors occurs concomitant with FLI-1 downregulation

Semiquantitative single-cell RT–PCR assay was used to investigate FLI-1 mRNA expression in individual clones of GFP+CD41+CD42 cells transduced with p210BCR-ABL retroviruses in the presence of imatinib or DMSO. To quantify precisely FLI-1 mRNA expression, RT product of each cell was divided into eight wells before performing FLI-1 transcripts amplification. One of these eight wells for each cell was used to co-amplify β2-microglobulin (β2-M) transcripts to control the quality of the RT product. The number of FLI-1-positive PCRs was then taken as an indirect indication of the FLI-1 mRNA copy number. Fifty cells cultured with imatinib and 50 cells cultured with DMSO were analyzed. Out of the 50 cells cultured in the presence of imatinib, 46 cells demonstrated a β2-M-positive and 45 cells demonstrated an FLI-1 positive, amplification in at least one of the eight wells (Figure 6a). Out of the 50 cells cultured with DMSO, 47 cells demonstrated a cβ2-M-positive amplification, whereas only 31 cells demonstrated an FLI-1-positive amplification in at least one of the eight wells (Figure 6a). An illustration of representative cells grown in the presence of imatinib with eight (N° 1) and seven (N° 2) FLI-1-positive PCRs or DMSO with only 4 (N° 3) or 0 (N° 4) FLI-1 positive PCRs is presented Figure 6b. The average number of FLI-1 positive PCRs per cell showed that imatinib induced at least 2.24-fold higher expression level of FLI-1. Altogether, these data indicate that similar to Mo7e cells, the tyrosine kinase activity of p210BCR-ABL also downregulates FLI-1 expression in individual cells of the CD41+CD42 cell population (Figure 6c).

Figure 6

Semi-quantitative single-cell RT–PCR analysis of FLI-1 expression in primary cells transduced by p210BCR-ABL-GFP in presence of imatinib or DMSO. One day after transduction, the GFP+CD41+CD42 cells were sorted at one cell per well and RT was performed directly in the well. Each RT product was divided into eight wells. The β2-M, used as an internal control of mRNA integrity, was amplified in only one of eight wells and only cells positive for β2-M were analyzed for FLI-1. A ‘FLI-1-positive PCR’ represents at least one copy of FLI-1 transcript. Therefore, if ‘FLI-1 positive PCRs’ were seen in the eight wells of one cell, this was interpreted as a cell expressing at least eight mRNA copies. (a) Statistical analysis of experiment. (b) RT-PCR analysis in two representative cells for each group (N° 1, N° 2 with imatinib, N° 3, N° 4 with DMSO). ‘No cell’ well: RT-PCR was performed in well without sorted cell. (c) Average of the number of ‘FLI-1-positive PCRs per cell in the two groups.


Cumulative evidence supports the notion that BCR-ABL may interfere with differentiation.13, 14, 15, 16, 17, 18 In this study, we show that the human megakaryocytic Mo7e cell line, as well as primary MK progenitor cells switched from MK to erythroid character after transduction with a p210BCR-ABL-expressing retrovirus. Using a gene array strategy, we identify a strong downregulation of the FLI-1 transcription factor mediated by the kinase activity of p210BCR-ABL, which was associated with erythroid lineage switching. Finally, we show that enforced re-expression of FLI-1 in p210BCR-ABL-expressing cells may rescue the MK phenotype.

In both Mo7e and primary progenitors, the increased expression of erythroid markers induced by p210BCR-ABL occurred at the expense of the expression of megakaryocytic markers. This opposite regulation of erythrocytic and megakaryocytic markers raised the intriguing possibility that enforced expression of p210BCR-ABL might be able to reprogram MK progenitors toward erythroid differentiation. We addressed this question by the analysis of the single-cell progeny of a CD41+CD42 population sorted immediately after the 3 days transduction protocol and these have been shown to be highly enriched in MK progenitors. Even in this CD41+CD42-purified population, our results showed an increased number of pure erythroid clones and a concomitant decreased number of MK clones derived from MK progenitors (Figure 3c). However, although highly enriched in MK progenitors, this CD41+CD42 population still contains erythroid progenitors, as indicated by the 10% of pure erythroid clones observed in our single-cell progeny analysis of the CD41+CD42 population transduced by the control GFP virus (Figure 3c). These single-cell experiments thus could not exclude that the increased number of erythroid clones might be derived from the selective outgrowth of these erythroid progenitors during the 3 days transduction protocol. Having shown that the switching effect of p210BCR-ABL is completely suppressed by Imatinib (Figure 3c), we took advantage of the reversibility of the effect of Imatinib to repress the p210BCR-ABL kinase during the 3 days transduction period and to investigate the effect of de-repressing the kinase activity only after single-cell sorting of the transduced CD41+CD42 population. Even under these conditions, we still observed a significant increase in the number of pure erythroid clones and a significant decrease in the number of MK clones derived from MK progenitors. These data unambiguously demonstrate that enforced expression of p210BCR-ABL is indeed able to reprogram at least some already-committed MK progenitors present in the CD41+CD42 population. Interestingly, this reprogramming seems to be limited only to early MK progenitors (and most probably also to bipotent progenitors), since the proportion of clones harboring few MK (MK clones from pro-MK) was never seen to be modified in the transduced CD41+CD42 population (Figure 3c).

A gene array strategy using parental and p210BCR-ABL overexpressing Mo7e cells revealed that several transcription factors shared during erythroid and MK differentiation were regulated by p210BCR-ABL. Among these, FLI-1 and EKLF, which are more specific to the MK and erythroid lineages, were found to be down- or upregulated by p210BCR-ABL, respectively. We focused our work on FLI-1 as this transcription factor in association with the fact that GATA-1 activates the transcription of several MK-restricted genes such as gpIX (CD42a), gpIbα (CD42b), and c-mpl through cooperative DNA binding.35, 36, 37 FLI-1 has also the ability to inhibit the erythroid differentiation in several erythroleukemic cell lines.38, 39, 40 Altogether, these data strongly suggest that during normal hematopoiesis, FLI-1 not only activates the MK differentiation program but also simultaneously represses the erythroid program. In favor of this hypothesis, it was recently reported that FLI-1 knock-out induces a blockage in MK differentiation33, 34, 41 with an expansion of erythroid progenitors.42

A relationship between FLI-1 and p210BCR-ABL is also suggested by the observation that the CML derived erythroleukemic K562 cell is deficient in FLI-1. Re-expression of FLI-1 in K562 cells induces an MK differentiation36, 43 and inhibits the erythroid phenotype.38 Similarly, our results show that when FLI-1 expression was enforced in p210BCR-ABL-Mo7e cells, the cell line recovered its initial MK phenotype, demonstrating that FLI-1 was a key transcription factor responsible for the lineage switch induced by p210BCR-ABL. Together, these data strongly suggest that the signaling pathway induced by p210BCR-ABL is able to suppress or activate some transcription or repressor factors involved in the regulation of the erythro/MK commitment. Further experiments involving FLI-1 knockdown will be required to investigate whether FLI-1 downregulation is a direct consequence of BCR-ABL expression or a secondary effect. A similar regulation of transcription factors by p210BCR-ABL has been reported previously. For example, p210BCR-ABL is able to downregulate the transcript levels of the granulocyte colony stimulating factor receptor (G-CSFR) by inhibiting the translation of the transcription regulator C/EBPalpha, through hnRNP E2.11 More recently, it has been shown that BCR-ABL induces an aberrant splicing of IKAROS, suggesting that BCR-ABL may interfere with lineage commitment.12 In favor of this hypothesis, we have shown that FLI-1 expression was downregulated in Mo7e and in primary human hematopoietic cell by p210BCR-ABL. Investigation of this novel transdifferentiation response offers insights into the normal mechanisms of lineage restriction and commitment in primitive human cells.

It remains to determine if the downregulation of FLI-1 expression by p210BCR-ABL is relevant to the progression of the disease and to the lineage infidelity of the leukemic blasts frequently found in this disease. The downregulation of FLI-1 may participate to the frequency of erythroid markers on the blasts in CML.9 It has been reported that CML patients have elevated numbers of erythroid progenitor cells.44, 45 The present findings showing an upregulation of the generation of erythroid cells in the presence of p210BCR-ABL are in line with the pathologic features of human CML. However, differentiation beyond the CFU-E stage is blocked in CML and many patients have some degree of anemia, supporting the hypothesis of a discordant maturation of erythroid progenitors in addition to an erythroid progenitor amplification described by the present and previous studies.5, 18, 44, 46, 47

Finally, another important aspect in this study is the lineage switch was induced by an oncoprotein that is essentially a signaling molecule. Indeed, no effect on the lineage choice or on FLI-1 expression was found with a kinase-dead p210BCR-ABL or when imatinib was added to culture, suggesting that kinase activity of p210BCR-ABL is necessary to induce these effects. Whether other protein kinases that activate similar signaling pathway induce similar effects on E/MK commitment remains to be determined, but seem likely. In addition, the use of different p210BCR-ABL mutant proteins or pharmacologic inhibitors may help to link signaling pathway to transcription factor regulation.

Accession codes




  1. 1

    Groffen J, Stephenson JR, Heisterkamp N, de Klein A, Bartram CR, Grosveld G . Philadelphia chromosomal breakpoints are clustered within a limited region, BCR, on chromosome 22. Cell 1984; 36: 93–99.

    CAS  Article  Google Scholar 

  2. 2

    Ben-Neriah Y, Daley GQ, Mes-Masson AM, Witte ON, Baltimore D . The chronic myelogenous leukemia-specific P210 protein is the product of the BCR/ABL hybrid gene. Science 1986; 233: 212–214.

    CAS  Article  Google Scholar 

  3. 3

    Konopka JB, Witte ON . Detection of C-ABL tyrosine kinase activity in vitro permits direct comparison of normal and altered ABL gene products. Mol Cell Biol 1985; 5: 3116–3123.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  4. 4

    Bettelheim P, Lutz D, Majdic O, Paietta E, Haas O, Linkesch W et al. Cell lineage heterogeneity in blast crisis of chronic myeloid leukaemia. Br J Haematol 1985; 59: 395–409.

    CAS  Article  PubMed  Google Scholar 

  5. 5

    Eaves CJ, Eaves AC . Cell culture studies in CML. Baillieres Clin Haematol 1987; 1: 931–961.

    CAS  Article  PubMed  Google Scholar 

  6. 6

    Hoppin EC, Lewis JP . Polycythemia rubra vera progressing to Ph1-positive chronic myelogenous leukemia. Ann Intern Med 1975; 83: 820–823.

    CAS  Article  PubMed  Google Scholar 

  7. 7

    Eaves C, Cashman J, Eaves A . Defective regulation of leukemic hematopoiesis in chronic myeloid leukemia. Leuk Res 1998; 22: 1085–1096.

    CAS  Article  Google Scholar 

  8. 8

    Saikia T, Advani S, Dasgupta A, Ramakrishnan G, Nair C, Gladstone B et al. Characterisation of blast cells during blastic phase of chronic myeloid leukaemia by immunophenotyping-experience in 60 patients. Leuk Res 1988; 12: 499–506.

    CAS  Article  PubMed  Google Scholar 

  9. 9

    Klein E, Ben-Bassat H, Neumann H, Ralph P, Zeuthen J, Polliack A et al. Properties of the K562 cell line, derived from a patient with chronic myeloid leukemia. Int J Cancer 1976; 18: 421–431.

    CAS  Article  PubMed  Google Scholar 

  10. 10

    Deininger MW, Goldman JM, Melo JV . The molecular biology of chronic myeloid leukemia. Blood 2000; 96: 3343–3356.

    CAS  PubMed  Google Scholar 

  11. 11

    Perrotti D, Cesi V, Trotta R, Guerzoni C, Santilli G, Campbell K et al. BCR-ABL suppresses C/EBPalpha expression through inhibitory action of hnRNP E2. Nat Genet 2002; 30: 48–58.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  12. 12

    Klein F, Feldhahn N, Herzog S, Sprangers M, Mooster JL, Jumaa H et al. BCR-ABL1 induces aberrant splicing of IKAROS and lineage infidelity in pre-B lymphoblastic leukemia cells. Oncogene 2006; 25: 1118–1124.

    CAS  Article  PubMed  Google Scholar 

  13. 13

    Elefanty AG, Cory S . BCR-ABL-induced cell lines can switch from mast cell to erythroid or myeloid differentiation in vitro. Blood 1992; 79: 1271–1281.

    CAS  PubMed  Google Scholar 

  14. 14

    Fang G, Kim CN, Perkins CL, Ramadevi N, Winton E, Wittmann S et al. CGP57148B (STI-571) induces differentiation and apoptosis and sensitizes BCR-ABL-positive human leukemia cells to apoptosis due to antileukemic drugs. Blood 2000; 96: 2246–2253.

    CAS  PubMed  Google Scholar 

  15. 15

    Schuster C, Forster K, Dierks H, Elsasser A, Behre G, Simon N et al. The effects of BCR-ABL on C/EBP transcription-factor regulation and neutrophilic differentiation are reversed by the ABL kinase inhibitor imatinib mesylate. Blood 2003; 101: 655–663.

    CAS  Article  PubMed  Google Scholar 

  16. 16

    Daley GQ, Van Etten RA, Baltimore D . Induction of chronic myelogenous leukemia in mice by the P210bcr/abl gene of the Philadelphia chromosome. Science 1990; 247: 824–830.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  17. 17

    Kelliher MA, McLaughlin J, Witte ON, Rosenberg N . Induction of a chronic myelogenous leukemia-like syndrome in mice with V-ABL and BCR/ABL. Proc Natl Acad Sci USA 1990; 87: 6649–6653.

    CAS  Article  Google Scholar 

  18. 18

    Chalandon Y, Jiang X, Hazlewood G, Loutet S, Conneally E, Eaves A et al. Modulation of p210(BCR-ABL) activity in transduced primary human hematopoietic cells controls lineage programming. Blood 2002; 99: 3197–3204.

    CAS  Article  PubMed  Google Scholar 

  19. 19

    Adolfsson J, Mansson R, Buza-Vidas N, Hultquist A, Liuba K, Jensen CT et al. Identification of Flt3+ lympho-myeloid stem cells lacking erythro-megakaryocytic potential a revised road map for adult blood lineage commitment. Cell 2005; 121: 295–306.

    CAS  Article  Google Scholar 

  20. 20

    McDonald TP, Sullivan PS . Megakaryocytic and erythrocytic cell lines share a common precursor cell. Exp Hematol 1993; 21: 1316–1320.

    CAS  PubMed  Google Scholar 

  21. 21

    Vannucchi AM, Paoletti F, Linari S, Cellai C, Caporale R, Ferrini PR et al. Identification and characterization of a bipotent (erythroid and megakaryocytic) cell precursor from the spleen of phenylhydrazine-treated mice. Blood 2000; 95: 2559–2568.

    CAS  PubMed  Google Scholar 

  22. 22

    Debili N, Coulombel L, Croisille L, Katz A, Guichard J, Breton-Gorius J et al. Characterization of a bipotent erythro-megakaryocytic progenitor in human bone marrow. Blood 1996; 88: 1284–1296.

    CAS  PubMed  PubMed Central  Google Scholar 

  23. 23

    Papayannopoulou T, Brice M, Farrer D, Kaushansky K . Insights into the cellular mechanisms of erythropoietin-thrombopoietin synergy. Exp Hematol 1996; 24: 660–669.

    CAS  PubMed  Google Scholar 

  24. 24

    Perry C, Soreq H . Transcriptional regulation of erythropoiesis. Fine tuning of combinatorial multi-domain elements. Eur J Biochem 2002; 269: 3607–3618.

    CAS  Article  PubMed  Google Scholar 

  25. 25

    Cantor AB, Orkin SH . Transcriptional regulation of erythropoiesis: an affair involving multiple partners. Oncogene 2002; 21: 3368–3376.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  26. 26

    Shivdasani RA . Molecular and transcriptional regulation of megakaryocyte differentiation. Stem Cells 2001; 19: 397–407.

    CAS  Article  PubMed  Google Scholar 

  27. 27

    Wang X, Crispino JD, Letting DL, Nakazawa M, Poncz M, Blobel GA . Control of megakaryocyte-specific gene expression by GATA-1 and FOG-1: role of Ets transcription factors. EMBO J 2002; 21: 5225–5234.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  28. 28

    Geay JF, Buet D, Zhang Y, Foudi A, Jarrier P, Berthebaud M et al. p210BCR-ABL inhibits SDF-1 chemotactic response via alteration of CXCR4 signaling and downregulation of CXCR4 expression. Cancer Res 2005; 65: 2676–2683.

    CAS  Article  Google Scholar 

  29. 29

    Berthebaud M, Riviere C, Jarrier P, Foudi A, Zhang Y, Compagno D et al. RGS16 is a negative regulator of SDF-1-CXCR4 signaling in megakaryocytes. Blood 2005; 106: 2962–2968.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  30. 30

    Lacroix L, Lazar V, Michiels S, Ripoche H, Dessen P, Talbot M et al. Follicular thyroid tumors with the PAX8-PPARgamma1 rearrangement display characteristic genetic alterations. Am J Pathol 2005; 167: 223–231.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  31. 31

    Raslova H, Komura E, Le Couedic JP, Larbret F, Debili N, Feunteun J et al. FLI1 monoallelic expression combined with its hemizygous loss underlies Paris-Trousseau/Jacobsen thrombopenia. J Clin Invest 2004; 114: 77–84.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  32. 32

    Druker BJ, Tamura S, Buchdunger E, Ohno S, Segal GM, Fanning S et al. Effects of a selective inhibitor of the ABL tyrosine kinase on the growth of BCR-ABL positive cells. Nat Med 1996; 2: 561–566.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  33. 33

    Hart A, Melet F, Grossfeld P, Chien K, Jones C, Tunnacliffe A et al. FLI-1 is required for murine vascular and megakaryocytic development and is hemizygously deleted in patients with thrombocytopenia. Immunity 2000; 13: 167–177.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  34. 34

    Spyropoulos DD, Pharr PN, Lavenburg KR, Jackers P, Papas TS, Ogawa M et al. Hemorrhage, impaired hematopoiesis, and lethality in mouse embryos carrying a targeted disruption of the FLI-1 transcription factor. Mol Cell Biol 2000; 20: 5643–5652.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  35. 35

    Deveaux S, Filipe A, Lemarchandel V, Ghysdael J, Romeo PH, Mignotte V . Analysis of the thrombopoietin receptor (MPL) promoter implicates GATA and ETS proteins in the co-regulation of megakaryocyte-specific genes. Blood 1996; 87: 4678–4685.

    CAS  Google Scholar 

  36. 36

    Eisbacher M, Holmes ML, Newton A, Hogg PJ, Khachigian LM, Crossley M et al. Protein–protein interaction between FLI-1 and GATA-1 mediates synergistic expression of megakaryocyte-specific genes through cooperative DNA binding. Mol Cell Biol 2003; 23: 3427–3441.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  37. 37

    Jackers P, Szalai G, Moussa O, Watson DK . ETS-dependent regulation of target gene expression during megakaryopoiesis. J Biol Chem 2004; 279: 52183–52190.

    CAS  Article  PubMed  Google Scholar 

  38. 38

    Athanasiou M, Mavrothalassitis G, Sun-Hoffman L, Blair DG . FLI-1 is a suppressor of erythroid differentiation in human hematopoietic cells. Leukemia 2000; 14: 439–445.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  39. 39

    Starck J, Cohet N, Gonnet C, Sarrazin S, Doubeikovskaia Z, Doubeikovski A et al. Functional cross-antagonism between transcription factors FLI-1 and EKLF. Mol Cell Biol 2003; 23: 1390–1402.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  40. 40

    Starck J, Doubeikovski A, Sarrazin S, Gonnet C, Rao G, Skoultchi A et al. SPI-1/PU.1 is a positive regulator of the FLI-1 gene involved in inhibition of erythroid differentiation in friend erythroleukemic cell lines. Mol Cell Biol 1999; 19: 121–135.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  41. 41

    Kawada H, Ito T, Pharr PN, Spyropoulos DD, Watson DK, Ogawa M . Defective megakaryopoiesis and abnormal erythroid development in FLI-1 gene-targeted mice. Int J Hematol 2001; 73: 463–468.

    CAS  Article  PubMed  Google Scholar 

  42. 42

    Masuya M, Moussa O, Abe T, Deguchi T, Higuchi T, Ebihara Y et al. Dysregulation of granulocyte, erythrocyte, and NK cell lineages in FLI-1 gene-targeted mice. Blood 2005; 105: 95–102.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  43. 43

    Athanasiou M, Clausen PA, Mavrothalassitis GJ, Zhang XK, Watson DK, Blair DG . Increased expression of the ETS-related transcription factor FLI-1/ERGB correlates with and can induce the megakaryocytic phenotype. Cell Growth Differ 1996; 7: 1525–1534.

    CAS  PubMed  Google Scholar 

  44. 44

    Eaves AC, Eaves CJ . Abnormalities in the erythroid progenitor compartments in patients with chronic myelogenous leukemia (CML). Exp Hematol 1979; 7 (Suppl 5): 65–75.

    PubMed  Google Scholar 

  45. 45

    Goldman JM, Shiota F, Th'ng KH, Orchard KH . Circulating granulocytic and erythroid progenitor cells in chronic granulocytic leukaemia. Br J Haematol 1980; 46: 7–13.

    CAS  Article  PubMed  Google Scholar 

  46. 46

    Clarkson BD, Strife A, Wisniewski D, Lambek C, Carpino N . New understanding of the pathogenesis of CML: a prototype of early neoplasia. Leukemia 1997; 11: 1404–1428.

    CAS  Article  PubMed  Google Scholar 

  47. 47

    Marley SB, Lewis JL, Goldman JM, Gordon MY . Abnormal kinetics of colony formation by erythroid burst-forming units (BFU-E) in chronic myeloid leukaemia. Br J Haematol 1996; 93: 878–883.

    CAS  Article  PubMed  Google Scholar 

Download references


We are grateful to Frederic Larbret and Yann Lecluse (IFR54, Villejuif, France) for flow cytometric sorting experiments and François Delhommeau (INSERM U790) for cytological studies. We are indebted to Françoise Wendling for helpful discussions during the preparation of the manuscript. We thank Elizabeth Buchdunger (Novartis, Basel Switzerland) for providing Imatinib mesylate.

This work was supported by grants from the Institut National de la Santé et de la Recherche Médicale, the Institut Gustave Roussy, the Centre National de la Recherche Scientifique, the Ligue Nationale contre le Cancer (WV and FM ‘équipes labellisées LIGUE’) and the Association de Recherche contre le Cancer (grant 4309 to FL). DB was supported by PhD fellowships from the Ministère de la Recherche.

Author information



Corresponding author

Correspondence to F Louache.

Additional information

Supplementary Information accompanies the paper on the Leukemia website (

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Buet, D., Raslova, H., Geay, JF. et al. p210BCR-ABL reprograms transformed and normal human megakaryocytic progenitor cells into erythroid cells and suppresses FLI-1 transcription. Leukemia 21, 917–925 (2007).

Download citation


  • erythro/megakaryocytic lineage
  • reprogramming
  • FLI-1 transcription factor
  • hematopoietic commitment

Further reading