Original Article

Leukemia (2007) 21, 1679–1690; doi:10.1038/sj.leu.2404778; published online 24 May 2007

MN1 overexpression is an important step in the development of inv(16) AML

C Carella1, J Bonten1, S Sirma1, T A Kranenburg1, S Terranova1, R Klein-Geltink1, S Shurtleff2, J R Downing2, E C Zwarthoff3, P P Liu4 and G C Grosveld1

  1. 1Department of Genetics and Tumor Cell Biology, St Jude Children's Research Hospital, Memphis, TN, USA
  2. 2Department of Pathology, St Jude Children's Research Hospital, Memphis, TN, USA
  3. 3Department of Pathology, Josephine Nefkens Institute, Erasmus MC, Rotterdam, The Netherlands
  4. 4Oncogenesis and Development Section, National Human Genome Research Institute, NIH, Bethesda, MD, USA

Correspondence: Dr GC Grosveld, Department of Genetics and Tumor Cell Biology, St Jude Children's Research Hospital, 332 North Lauderdale, Memphis, TN 38105-0318, USA. E-mail: gerard.grosveld@stjude.org

Received 27 April 2007; Accepted 30 April 2007; Published online 24 May 2007.



The gene encoding the transcriptional co-activator MN1 is the target of the reciprocal chromosome translocation (12;22)(p13;q12) in some patients with acute myeloid leukemia (AML). In addition, expression array analysis showed that MN1 was overexpressed in AML specified by inv(16), in some AML overexpressing ecotropic viral integration 1 site (EVI1) and in some AML without karyotypic abnormalities. Here we describe that mice receiving transplants of bone marrow (BM) overexpressing MN1 rapidly developed myeloproliferative disease (MPD). This BM also generated myeloid cell lines in culture. By mimicking the situation in human inv(16) AML, forced coexpression of MN1 and Cbfbeta–SMMHC rapidly caused AML in mice. These findings identify MN1 as a highly effective hematopoietic oncogene and suggest that MN1 overexpression is an important cooperative event in human inv(16) AML.


AML, MN1, inv(16), CBFbetaSMMHC



The MN1 gene was identified as the target of a unique-balanced t(4;22) in a patient with meningioma and was thought to be a prime candidate for the meningioma tumor-suppressor gene on chromosome 22 (meningioma 1; Lekanne Deprez, 19911), but its relation to meningioma remains unresolved. MN1 encodes a nuclear protein of 150 kDa, highly conserved among vertebrates and no homology to other proteins. Its amino acid (aa) sequence suggested a role in transcription,2 which was confirmed by the observation that MN1 activated transcription of the Moloney sarcoma virus long terminal repeat (MSV-LTR) in transient transcription assays.3 MN1 appeared to activate transcription of the LTR via direct repeat sequences (DR5) that bind RAR-RXR nuclear receptor dimers. MN1 interacts with RAR-RXR most probably via the protein intermediates p300 and RAC3 (also known as nuclear receptor co-activator 3, NCOA3).4 RAC3 and MN1 are transcription co-activators5, 6 and coexpression of MN1 with p300 or RAC3 synergistically activated the transcriptional activity of RAR-RXR dimers in the presence of retinoic acid.4 MN1's co-activation activity is not restricted to the RAR-RXR nuclear receptor, as MN1 expression inhibits proliferation of an osteoblast cell line via co-activation of the vitamin D receptor.7

MN1 is also the target of a balanced t(12;22) in myeloid leukemia,3 in which its first exon is fused to TEL, a member of the family of ETS transcription factors.8 Although TEL generally functions as a site-specific transcriptional repressor,9, 10, 11 MN1-TEL moderately activates transcription of TEL-responsive reporters in transient transfection experiments, and the fusion protein has transforming activity in both NIH3T3 fibroblasts3 and mouse bone marrow (BM).12 The transforming activity in NIH3T3 cells critically depends on DNA binding via the ETS domain of TEL and on the presence of the N-terminal 500 aa of MN1.3 In BM cells, the transforming activity also depends on the MN1 N-terminal 500 aa, but it is independent of DNA binding.12

Conditional MN1-TEL knock-in mice that expressed the gene under the control of Aml1 regulatory sequences developed lymphoid or myeloid malignancies depending on the nature of the cooperating mutations.13, 14 This confirmed MN1-TEL's role as a bona fide hematopoietic oncogene.

Interestingly, the association of MN1 with myeloid malignancy might go beyond MN1's involvement in the t(12;22), as the gene was found to be overexpressed in inv(16)(p13;q22) acute myeloid leukemia (AML)15, 16 in some AMLs overexpressing the transcription factor ecotropic viral integration 1 site (EVI1)16 and in some adult AMLs without karyotypic abnormalities.17 In the latter case, overexpression of MN1 was associated with a worse prognosis and a shorter survival rate.17

Inv(16) is the chromosomal hallmark of one of two core-binding factor (CBF) leukemias and encodes the CBFbeta–smooth muscle myosin heavy chain (CBFbeta–SMMHC) fusion protein.18 CBF consists of a CBFbeta/RUNX1 heterodimer that regulates genes associated with lymphoid and myeloid differentiation.19 The RUNX1 subunit is the target of the recurrent t(8;21) in AML, giving rise to a RUNX1-ETO fusion protein. Both CBFbeta–SMMHC and RUNX1-ETO have a dominant-negative effect on CBF function.20, 21 This was concluded from the observation that Runx1 and Cbfb knockout mouse embryos and heterozygous Runx1-ETO or Cbfb-MYH11 knock-in embryos all die at mid-gestation due to the inability to switch to definitive hematopoiesis21, 22, 23 Mice chimeric for Runx1-ETO or Cbfb-MYH11 show alterations in multilineage differentiation of hematopoietic cells in the BM, but do not spontaneously develop myeloid leukemia.24, 25 Consistent with the notion that leukemogenesis is a multistep process,26 such mice only developed myeloid disease after treatment with the chemical carcinogen N-ethyl-N-nitrosourea (ENU).24, 25 Retroviral mutagenesis of Cbfb-MYH11 chimeric knock-in mice identified the cooperating zinc finger genes Plag1 and PlagL2,27 which are also overexpressed in 20% of human AML samples with PLAGL2 preferentially increased in inv(16) leukemia samples.28

Here we report that mice receiving transplants with BM overexpressing MN1 rapidly developed a fatal myeloproliferative disease (MPD), while mice receiving transplants with Cbfb-MYH11 chimeric BM overexpressing MN1 developed AML. In addition, quantitative real-time (QRT)–PCR analysis of AML samples confirmed MN1 overexpression in inv(16) patient samples, but elevated expression was also found in other pediatric AML samples. Our data suggest that MN1 overexpression is an important secondary mutation in inv(16) AML, but its upregulation may also contribute to the development of other AML subtypes.


Materials and methods

Patient materials

All patients and patient materials used in this paper have been described previously.15 Informed consent for the use of the leukemic cells for research was obtained from parents, guardians or patients (as age-appropriate) in accordance with the Declaration of Helsinki, and study approval was obtained from the SJCRH Institutional Review Board (IRB).

Plasmids and retrovirus production

MN1 cDNA3 was cloned into the EcoRI site of MSCV-IRES-GFP and high-titer ecotropic virus (5 times 105–1 times 106 CFU/ml) was obtained as described.29 As a control, we used MSCV-IRES-GFP virus without cDNA insert.

BM transplantation

For transplantation of MSCV-MN1-IRES-GFP-transduced wild-type BM into C57Bl/6 females, BM was harvested from femurs and tibiae of 8-week-old male C57BL/6 or C57BL/6/129svJ mice treated with 5-fluorouracil (5FU). 5FU injection, isolation of Lin- cells, viral transduction and BM transplantation into lethally irradiated female recipients was performed as described previously.29 For transplantation of inv(16) chimeric BM, heterozygous Cbfb-MYH11knock-in ES cells20 were injected into C57BL/6 blastocysts and ten highly chimeric male mice were generated. After treatment with 5FU, BM of these mice was isolated, transduced with MSCV-MN1-IRES-GFP or MSCV-IRES-GFP retrovirus, yielding cell populations that were 30 and 60% GFP+, respectively. The MSCV-MN1-IRES-GFP-transduced cells were transplanted into 12 lethally irradiated female C57BL/6 recipients and the MSCV-IRES-GFP control cells into six such recipients.


Cytospin preparations of BM of two pediatric inv(16) patients from the St Jude Children's Research Hospital tumor bank and BM of a healthy donor were fixed with 4% paraformaldehyde for 4 min. Cells were permeabilized with 0.2% Triton X-100 and incubated with MN1 monoclonal antibody 2F2,3 followed by incubation with a CY3-labeled goat-anti-mouse secondary antibody (1/400 dilution). MN1 staining of cytospin preparations of MN1-transduced BM-derived cell lines was performed following the same procedure. Fluorescent images of cytospin preparations were obtained using a BX-50 microscope (equipped with a UPlanFI times 40/0.75 or times 100/1.30 numeric aperture objectives, Olympus, Tokyo, Japan) with a SPOT camera and SPOT Advanced imaging software (Diagnostic Instruments, Sterling Heights, MI, USA).

Cell cycle analysis by Fluorescence-activated cell sorting

Cell cycle analysis of MN1-transduced BM cell lines and vector-transduced control BM cells was performed as described previously.29

Secondary BM transplantation

Secondary BM transplantation of leukemic BM was performed as described previously.12

Hematopoietic progenitor assays

Sequential methylcellulose-based cultures (MC1–4) of BM cells was performed using MethoCult GFM3434 (StemCell Technologies, Vancouver, BC, Canada), containing mouse stem cell factor (SCF) (50 ng/ml), mouse interleukin (IL)-3 (10 ng/ml), human IL-6 (10 ng/ml) and human erythropoietin (3 units/ml) and were performed as described.12

Analysis of diseased mice and tissue preparation

All animal procedures were conducted in accordance with the US Public Health Service Policy on the Humane Care and Use of Laboratory Animals. Collection of blood, green fluorescent protein (GFP) analysis of peripheral blood (PB), BM, spleen and suspensions, euthanasia of diseased animals, tissue collection, fixation, paraffin embedding, sectioning and histological staining and May-Grunwald–Giemsa staining of PB were performed as described.12 Select tissues were also processed for immunohistochemical analysis with antibodies to hematopoietic phenotype markers: CD3 (Dako, Carpinteria, CA, USA), CD45R/B220 (PharMingen, San Diego, CA, USA), terminal deoxynucleotidyl transferase (TdT, Supertechs, Bethesda, MA, USA), myeloperoxidase (MPO, Dako, Carpinteria, CA, USA), TER 119 (PharMingen), GATA (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) and GFP (Clontech, Palo Alto, CA, USA).

Cell surface markers

Marker analysis by fluorescent-activated cell sorting (FACS) of single-cell suspensions of BM and spleen was performed as described.12 Cells were incubated with monoclonal antibodies (CD3c, CD4, CD8, CD11b/Mac1, CD19, CD34, B220, TER-119, Gr1, Sca1, c-Kit, Flt3, all from Pharmingen; anti-mouse IgM from Southern Biotechnology Associates, Birmingham, AL, USA) on ice for 30 min. Stained cells were analyzed using a BD Biosciences FACSCalibur flow cytometer (BD Biosciences, San Jose, CA, USA).

Sorting of mouse BM for stem cell and progenitor fractions

Following the methods of Akashi et al.,30 different mouse BM progenitor populations representing the hematopoietic stem cell (HSC), the myeloid/erythroid progenitor (MEP), the common myeloid progenitor (CMP), the granulocyte/monocyte progenitor (GMP) and the common lymphoid progenitor (CLP) were isolated and their identity verified by reverse trancriptase (RT)–PCR analysis for the expression of signature genes.30


The RT reactions of patient BM RNA were done with 50 ng total RNA in 20 mul total volume using TaqMan Reverse Transcription Reagent (Applied Biosystems, Foster City, CA, USA) following the manufacturer's recommendations.

Primers and probes for genes were chosen with the assistance of the computer program Primer Express (PE Applied Biosystems, version 2.0.0). We confirmed the gene specificity of the nucleotide sequences chosen for the primers using BLASTN searches. To avoid amplification of contaminating genomic DNA, one of the two primers was placed at the junction between two exons or in a different exon. The primers and probes are shown in Table 1. Each probe was synthesized with the 5'-end reporter dye (FAM: 6-carboxy fluorescene phosphoramidite) and 3'-end BHQ1 dark quencher dye at St Jude's Hartwell Center for Bioinformatics and Biotechnology.

Quantitative PCR was performed on an ABI Prism 7900HT Sequence Detection System (PE Applied Biosystems). All PCR mixtures contained 2 mul cDNA (corresponding to 5 ng reverse transcribed total RNA), 1 times TaqMan Universal PCR Master Mix (PE Applied Biosystems), 1 times Eukaryotic 18S rRNA endogenous control, 300 nM of each primer and 200 nM probe in 30 mul reaction volume in 96-well plates. The thermal cycling conditions were as follows: after incubation at 50°C for 2 min and an initial denaturation step at 95°C for 10 min, 40 cycles were performed at 95°C for 15 s and 1 min at 60°C. Standard curves were obtained using cDNA generated with human or mouse total BM RNA (pooled human normal BM from eight male/female from BD Bioscience Clontech, Mountain View, CA, USA and pooled mouse BM from three male C57Bl/6/129svj mixed-background mice). Each PCR run included the six points of the standard curve (fivefold serially diluted human or mouse BM cDNA), a non-template control with water, a calibrator cDNA and the unknown cDNA samples. Baseline and threshold CT value were analyzed manually with the ABI Prism SDS2.1 software. Values for each PCR product were normalized against 18S rRNA to compare expression in patient BM samples with that in human BM total RNA (BD Bioscience Clontech). Quantitative PCR assays were conducted in duplicate for each sample and a mean value was used to calculate mRNA levels.

Q–PCR reactions to compare the copy number of the MN1 gene in inv(16) patient BM DNA samples with that in normal BM DNA was performed using the TaqMan Reverse Transcription Reagent following the manufacturer's protocols. A dilution series of 1, 0.5, 0.25, 0.125, 0.025, 0.005, 0.001, 0.0002 mug genomic BM DNA of two different inv(16) patients (indicated in red in Figure 1) and a normal individual were subjected to one-step PCR with MN1 first exon 1 and GAPDH exon 8 primers (Table 1), using 40 cycles of amplification (10 s 95°C, 1 min 60°C) after incubation of the samples for 10 min at 25°C, 30 min at 48°C and 10 min at 95°C. The GAPDH 5' end reporter dye was tetrachlorofluorescein instead of FAM. Baseline and threshold CT value were analyzed manually with the ABI Prism SDS2.1 software. Values for the MN1 PCR product were normalized against that of the GAPDH product in inv(16) and normal BM DNA samples.

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

MN1 expression in acute myeloid leukemia (AML) patient samples and mouse bone marrow (BM). (a) Upper panel: BM RNA of pediatric AML FAB-M4 patients (16 in total, of which 10 with eosinophilia (E), including 9 samples with inv(16), 1 with t(8;21) and 6 with other karyotypic abnormalities (other)) were analyzed for expression of MN1 by quantitative real-time (QRT)–PCR. Expression levels are depicted as fold expression of MN1 in normal human BM. Results are the average of two experiments. The upper row of the nine fluorescence micrographs shows indirect immunofluorescence detection of MN1, using BM cytospin preparations stained with an MN1 monoclonal antibody of 2 inv(16) patients (indicated by red labels in the upper panel) and a healthy individual. The middle three micrographs show the 4'-6-diamidino-2-phenylindole (DAPI) nuclear staining of the corresponding samples in the upper row. The lower row shows control staining of the same cytospin preparations using secondary antibody only. (b) BM RNA of 41 pediatric patients with AML (1 FAB-M0 (olive green), 4 FAB-M1 (orange), 5 FAB-M2 (purple), 7 FAB-M3 (blue), 16 FAB-M4 (red), 6 FAB-M5 (green) and 2 FAB-M7 (black)) with the indicated chromosomal translocations or other karyotypic abnormalities (other) were analyzed for expression of MN1 by QRT–PCR. Expression levels are depicted as fold expression of MN1 in normal human BM. 'MLL' signifies samples with different chromosomal translocations involving the MLL gene. 'E' indicates eosinophilia. Results are the average of two experiments. (c) Mouse BM (FVB) was fluorescent-activated cell sorting (FACS) sorted into fractions representing the hematopoietic stem cells (HSC), the common myeloid progenitors (CMP), the common lymphoid progenitors (CLP), the granulocyte/macrophage progenitors (GMP) and the megakaryocyte/erythrocyte progenitors (MEP) using the forward scatter of differentiation lineage-negative cells, expressing or not of the cell surface markers c-Kit, Sca-1, Fcitalic gamma-receptor and CD34. RNA of each fraction was extracted and subjected to QRT–PCR to determine the level of Mn1 expression in relation to Hprt. Expression levels are depicted as fold expression of Mn1 in unfractionated mouse BM. Lin- BM represents a sample containing all progenitors depleted of cells expression lineage markers.

Full figure and legend (253K)

Analysis of Mn1 expression in mouse HSCs and BM progenitor populations was carried out using a two-step real-time RT–PCR protocol. FACS sorted HSCs and BM progenitor fractions were pelleted by centrifugation and subsequently lysed in 800 mul of TRIzol reagent (Invitrogen Corporation, Carlsbad, CA, USA). Total RNA was isolated using the PureLink Micro-to-Midi Total RNA Purification Kit (Invitrogen Corporation) following the manufacturers protocol, however, with addition of 10 mug of RNase-free glycogen before binding to the column. Genomic DNA was degraded by addition of 1 unit of DNase I for 15 min at room temperature, followed by inactivation of DNase I by addition of 1 mul of 25 mM ethylenediaminetetraacetic acid and heating at 65°C for 10 min. First-strand cDNA synthesis was carried out on the total RNA isolate using SuperScript III First-Stand Synthesis SuperMix (Invitrogen Corporation), following the manufacturers instructions. Real-time analysis was carried out using a modified version of the multiplexed tandem PCR approach.31 For first-round multiplexed amplification, 5 mul of the first-strand cDNA reaction was added to MN1 or HPRT outer primers (300 nM final of each), 200 nM dNTPs (Promega, Madison, WI, USA), 1.5 mM MgCl2, GoTaq buffer (Promega) and 0.5 U GoTaq (Promega) in a total volume of 20 mul. PCR was carried out on a PTC-200 thermocycler (Bio-Rad Laboratories, Hercules, CA, USA) for 15 cycles using the following conditions: 95°C, 10 min for 1 cycle followed by 20 cycles of 95°C for 10 s, 60°C for 20 s and 72°C for 20 s. The resulting products were mixed and first-round primers, dNTPs and Taq, were removed using the QIAGEN mini elute PCR clean-up kit (QIAGEN Inc., Valencia, CA, USA). The real-time PCR consisted of 5 mul of an appropriate dilution of the first-round amplified cDNA mixed with 15 mul of 2 times TaqMan Universal PCR Master Mix (Applied Biosystems, Foster City, CA, USA), 300 nM each of the inner Mn1 or Hprt forward and reverse primers and 200 nM probe in a final reaction volume of 30 mul. Real-time PCR was carried out in a 96-well plate on Bio-Rad iQ5 Multicolor Real-Time PCR Detection System using the following cycling conditions: one cycle each of 50°C for 2 min and 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. For all samples, the expression level of Mn1 was normalized to Hprt and expressed relative to the level in total BM.

Retroviral integration sites in leukemia samples

To determine the number of MN1 retroviral integration sites, Southern blots containing tumor DNA digested with EcoRI were hybridized with a GFP probe. This detects hybrid DNA fragments containing murine stem cell virus (MSCV) sequences downstream of the viral 3' EcoRI site (Figure 1a) to the first downstream EcoRI site in mouse genomic DNA.



Inv(16) leukemia cells overexpress MN1

Expression profiling of pediatric and adult patients with AML15, 16 has shown that BM from patients with inv(16) AML, who express the CBFbeta–SMMHC fusion protein, expressed up to 10 times more MN1 mRNA than normal BM. Inv(16) AML is often of the French–American–British classification32 (FAB)-M4 subtype with eosinophilia. Also AML patients overexpressing EVI1 showed upregulated expression of MN1.16 We therefore assessed the level of MN1 expression in 41 pediatric AML BM samples in comparison with expression in normal BM using QRT–PCR. This group included: 1 acute undifferentiated leukemia (FAB-M0), 4 acute myeloblastic leukemias (FAB-M1; 1 with t(8;21)), 5 acute myeloblastic leukemias with maturation (FAB-M2; 4 with t(8;21), 1 other), 7 acute promyelocytic leukemias (FAB-M3; all t(15;17)), 16 acute myelomonocytic leukemias (FAB-M4; 7 inv(16) with eosinophilia, 2 inv(16) without eosinophilia, 1 t(8;21) with eosinophilia, 6 other M4)), 6 acute monocytic leukemias (FAB-M5, all MLL translocations) and 2 megakaryoblastic leukemias (FAB-M7). All 9 inv(16) samples, with or without eosinophilia (E), showed between 17- to 112-fold higher MN1 expression than normal BM, whereas only one non-inv(16) FAB-M4 sample showed medium to high MN1 expression (Figures 1a and b and Supplementary Figure 1A). The six other FAB-M4 samples showed no (1 t(8;21)M4E) or only moderate upregulation (1.7- to 2.9-fold; Figures 1a and b). Also the four FAB-M1 leukemia samples showed significant MN1 upregulation (4.1- to 25.3-fold) (Figure 1b and Supplementary Figure 1B). Of the remaining 21 AML samples, 9 showed moderately increased expression (2.2- to 8-fold) of MN1, including all M2 samples, 4 of which carried the t(8;21), whereas 12 samples showed a lower than twofold increase in MN1 expression (Figure 1b). The latter group contained five of six FAB-M5 samples harboring MLL translocations, five of seven t(15;17) FAB-M3 samples, one of two FAB-M7 samples and one FAB-M0 sample.

These results not only confirmed overexpression of MN1 in inv(16) FAB-M4, but also indicated that the level of MN1 overexpression substantially exceeded that determined by expression profiling.

Because Plag1 and PlagL2 overexpression was shown to cooperate with Cbfb-MYH11 in a mouse model of inv(16) leukemia27 and PLAGL2 expression was preferentially upregulated in BM samples of inv(16) leukemia patients,28 we tested whether there was a direct correlation between the levels of MN1 and PLAGL2 expression in our 9 inv(16) patient samples. QRT–PCR for PLAGL2 mRNA showed that only 1 FAB-M4E sample showed a 4.4-fold higher expression of PLAGL2 than control BM, whereas the other 8 samples showed equal or lower PLAGL2 expression than control BM (Supplementary Figure 1A). This result suggested that upregulation of MN1 and PLAGL2 are not functionally linked and appear to be independent genetic events in inv(16) leukemia.

Because Valk and co-workers16 also reported overexpression of MN1 in leukemia samples with upregulated EVI1 expression, we repeated the QRT–PCR analysis of the same 41 AML samples for EVI1 expression. Only five samples showed increased expression of this gene (Supplementary Figure 1C), and two of those (1 FAB-M7 and 1 FAB-M1) also showed upregulated expression of MN1. These data suggest that upregulation of MN1 does occur in EVI1 leukemia, but is not an obligatory step.

Using an MN1 monoclonal antibody,3 immunofluorescence detection of MN1 in the BM cells of 2 inv(16) patients and a healthy subject (fold overexpression of MN1 mRNA patient1: patient2: normal BM=50:25:1; Figure 1a) indeed confirmed increased speckled staining in the nucleus of tumor cells of patient1 and slightly increased staining in cells of patient2 (Figure 1a). The same cells incubated with secondary antibody alone showed no staining (Figure 1a).

To determine whether MN1 upregulation might be the result of gene amplification, we employed QRT–PCR of DNA of the same 2 inv(16) patient samples that we used for MN1 immunofluorescence to determine whether their MN1 copy number was higher than that in normal BM DNA. However, no amplification could be detected (not shown).

Expression of mouse Mn1 in selected mouse BM progenitor populations

Because MN1 is upregulated in inv(16) leukemia, we wished to address which hematopoietic progenitor cells normally express this gene. To answer this question, we used mouse BM and followed the method of Akashi and co-workers30 to sort populations representing the HSC, CMP, CLP, MEP and GMP (Figure 1c). We then determined the amount of Mn1 mRNA present in these fractionated cell populations by using quantitative (Q)RT–PCR with whole mouse BM RNA as a control. This showed that Mn1 mRNA was present at a similar level in whole BM and in the HSC fraction, while Lin- BM cells were twofold enriched in Mn1-expressing progenitors. Analysis of the other fractions showed that this signal was derived from the GMP fraction in which Mn1 expression is 300-fold higher than that in whole BM, whereas there was no Mn1 expression in the CMP, CLP or MEP fractions. This result suggests a specific role for Mn1 in the GMP, which could involve expansion and/or differentiation of the progeny of this fraction and that forced overexpression of MN1 causes this progenitor population to expand abnormally.

Overexpression of MN1 in mouse BM stimulates outgrowth of myeloid cells and produces immortalized cell lines

We reported that MN1-TEL transforms NIH3T3 fibroblasts, an activity dependent on both the presence of MN1 N-terminal sequences and a functional TEL DNA-binding domain.3 Paradoxically, the self-renewal activity of mouse BM cells was equally stimulated by MN1-TEL as by an MN1-TEL mutant with a nonfunctional ETS DNA-binding domain, as measured by colony-forming assays in semisolid medium.12 To determine if MN1 alone also possessed this capacity, BM transduced with MSCV expressing MN1-IRES-GFP (30% GFP+) (Figure 2a) was plated in methylcellulose and its colony-forming activity compared with that of BM transduced with MSCV-IRES-GFP. In the first methylcellulose assay (MC1), MSCV-MN1-IRES-GFP-transduced cells gave twice as many colonies as vector-transduced BM (Figure 2a). Upon serial replating (MC2, MC3, MC3), the number of MN1+ BM and vector BM colonies dropped to equal numbers in the MC2, but the number of MN1+ colonies increased drastically in the MC3 and MC4, whereas vector-transduced cells produced no colonies. FACS analysis of cells recovered from the MC3 culture revealed that >95% of cells were GFP+ (not shown). This together with the observation that MN1-overexpressing cells recovered from the MC1 grew rapidly in liquid culture, and were 98% GFP+ within 4 weeks of culture, strongly suggested that MN1 overexpressing cells acquired a distinct growth advantage over that of non-transduced cells. Immunofluorescence analysis with an MN1 antibody showed mostly nuclear, punctate MN1 signals in all cells (Figure 2b), whereas vector-transduced cells showed only faint GFP fluorescence encoded by the retroviral vector. Cell surface marker analysis showed that the cells were c-Kit+/Sca1+/Mac1+ with 10% of the cells also expressing Gr1 (not shown). Cells cultured this way were immortalized but their growth and survival was strictly dependent upon addition of the cytokines IL3 and SCF to the culture (not shown). Cell cycle analysis using flow cytometry of two independent MN1 cell lines and vector-transduced BM, all cultured for 2 months after transduction, showed that the fraction of MN1-transduced cells in the S- and G2/M-phase of the cell cycle was drastically increased compared to that of vector-transduced cells (Figure 2c). This showed that MN1 overexpression strongly stimulates cell cycle traverse.

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

Mouse bone marrow (BM) transduced with MN1 retrovirus shows increased proliferation and generates cytokine-dependent cell lines. (a) The map at the top depicts the structure of the MSCV-MN1-IRES-GFP retroviral vector used to transduce mouse BM. The MN1 open reading frame (MN1) is fused to IRES-GFP (I-GFP), flanked by murine stem cell virus (MSCV) long terminal repeat (LTR) sequences. As control, we used the same vector not containing MN1 sequences. R1 indicates EcoRI sites flanking the insert. BM from C57Bl/6/129svJ mice, transduced with MSCV-MN1-IRES-GFP (MN1) or MSCV-IRES-GFP retrovirus (vector), were serially plated in methylcellulose semisolid medium (MC1 (20.000 BM cells plated), MC2 (1000 MC1 cells plated), MC3 (1000 MC2 cells plated), MC4 (1000 MC3 cells plated)) and at each step colonies were counted after 2 weeks of culture. Results are the average of two experiments. (b) Fluorescence micrograph of an MSCV-MN1-IRES-GFP-transduced cell line and MSCV-IRES-GFP-transduced BM stained with an MN1 antibody. (c) Flow cytometric analysis by fluorescent-activated cell sorting (FACS) of the DNA content of propidium iodide stained nuclei from two independent MSCV-MN1-IRES-GFP-transduced cell lines (MN1) and MSCV-IRES-GFP-transduced BM (vector), cultured for 2 months after transduction of the BM.

Full figure and legend (155K)

Mice receiving transplants of BM transduced with MN1 retrovirus rapidly develop MPD

We next tested the effect of MN1 overexpression on mouse BM in vivo. Lethally irradiated C57Bl/6/129svJ mice (n=14) died of fulminant hematopoietic disease 5–8 weeks after receiving transplants of 3–5 times 105 MSCV-MN1-IRES-GFP-transduced (Figure 3a) C57Bl/6 Lin- BM cells (60% GFP+), whereas mice receiving transplants with the same number of MSCV-IRES-GFP C57Bl/6 Lin- BM cells (60% GFP+) (n=5) remained healthy (Figure 3a). The average white blood cell (WBC) count of diseased mice was 1.5 times 107/ml, and their PB contained large numbers of neutrophils, neutrophil precursors and some blast-like cells (Figure 3b) all expressing GFP (Figure 3e). Southern blotting of the BM DNA of three of the diseased mice probed with IRES-GFP showed multiple integrations of the MSCV-MN1-IRES-GFP retrovirus (Figure 3c), suggesting that the disease was oligoclonal. Malignant cells invaded the spleen, liver and brain (Figure 3d), and most of the mice died of brain hemorrhage that was probably caused by the high WBC. FACS analysis showed that BM and spleen were each composed of a single population of Mac1+/Gr1+ cells (Figure 3e). Together, these features suggested that the mice suffered from an MPD.

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

Mice receiving transplants of MSCV-MN1-IRES-GFP-transduced bone marrow (BM) rapidly develop myeloproliferative disease (MPD). (a) Survival curve of lethally irradiated C57Bl/6 mice receiving transplants of syngeneic BM transduced with MSCV-MN1-IRES-GFP (MN1, n=14) or MSCV-IRES-GFP (Vector, n=5) retrovirus. Mice receiving MN1-overexpressing BM died of hematopoietic disease between 34 and 52 days after transplantation, whereas vector-transplanted mice remained healthy. (b) Peripheral blood (PB) smear of a representative moribund MN1-BM-transplanted mouse (average WBC is 150 times 107/ml) showing large numbers of partially and fully differentiated neutrophils and some blast-like cells (arrow), showing the mice suffered of an MPD. (c) Southern blot of BM DNA of three MPD mice overexpressing MN1, digested with EcoRI and hybridized with a green fluorescent protein (GFP) probe. All three samples contain multiple retroviral integrations. (d) Malignant cells invaded the spleen and liver. All diseased mice showed brain hemorrhages, often causing sudden death. Secondary recipients receiving transplants of diseased BM rapidly (18–23 days) developed the same disease (not shown). (e) fluorescent-activated cell sorting (FACS) analysis of cell surface markers of MPD BM showing that cells were GFP+/Gr1+/Mac1+ and some also expressed Sca-1 and/or Thy-1. Cells were negative for c-Kit.

Full figure and legend (277K)

Transplantation of pooled 5 times 105 BM cells from two of the diseased MN1 mice into five sublethally irradiated secondary recipients recreated the same fulminant MPD 18–23 days later (not shown), suggesting the disease was cell autonomous. We conclude that MN1 is a highly efficient oncogene that strongly promotes the growth of myelomonocytic cells, but does not substantially inhibit their differentiation.

Coexpression of Cbfbeta–SMMHC and MN1 causes AML in mice

Because MN1 is consistently overexpressed in inv(16) AML, we tested whether its overexpression in BM of mice expressing Cbfbeta–SMMHC cooperates to cause AML. Chimeric mice generated with Cbfb-MYH11knock-in ES cells do not spontaneously develop hematopoietic disease, and hematopoietic cells expressing the Cbfbeta–SMMHC fusion protein fail to differentiate and remain in the BM.24 We used Cbfb-MYH11 ES cells to generate 10 highly chimeric mice (>90% chimerism of the coat color). Lin- BM cells from these mice were transduced with MN1-IRES-GFP retrovirus (30% GFP+) and transplanted into 12 lethally irradiated C56Bl/6 recipients. As a control, we transplanted chimeric BM transduced with control IRES-GFP retrovirus (45% GFP+) into 6 irradiated recipients. Given that all our human inv(16) samples overexpressed MN1, it was possible that transcription of MN1 is regulated by Cbfbeta–SMMHC. Therefore, we determined by QRT–PCR whether the level of endogenous Mn1 mRNA in MSCV-IRES-GFP-transduced chimeric mouse BM was more abundant than in similarly transduced wild-type BM. As shown in Figure 4a Mn1 expression in both types of BM was similar, opposing the possibility that Cbfbeta–SMMHC directly or indirectly upregulated Mn1 expression in mouse BM.

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

Mice receiving transplants of chimeric Cbfb-MYH11 bone marrow (BM) transduced with MSCV-MN1-IRES-GFP develop acute myeloid leukemia (AML). (a) RNA from normal mouse BM (wt) and chimeric inv(16) BM (inv), both transduced with MSCV-IRES-GFP virus was analyzed by quantitative real-time (QRT)–PCR for the expression of endogenous Mn1. The values are the average of two independent experiments. Mn1 expression is not upregulated in inv(16) chimeric BM. (b) Survival curve showing that lethally irradiated C57Bl/6 mice receiving transplants of chimeric Cbfb-MYH11 BM transduced with MSCV-MN1-IRES-GFP (n=12) developed hematopoietic disease 58–68 days after transplantation, while those receiving transplants of the same BM transduced with MSCV-IRES-GFP (n=6) retrovirus remained healthy. The blood smear to the right shows the peripheral blood (PB) of a mouse with AML with a preponderance of blast cells, but also showing neutrophil progenitors. (c) Left panel: fluorescent-activated cell sorting (FACS) cell surface marker analysis of the BM of a mouse with a preponderance of blast cells in the PB, with 75.3% of cells expressing c-Kit, 9.6% of cells expressing Gr1 and 17% Mac1 (upper three plots) and of the BM of a mouse with a lower number of blast cells in the PB showing 19.6% of cells expressing c-Kit and 49.1% of cells expressing Gr1 and 60.9% of cells expressing Mac1 (lower three plots). More than 98% of cells express green fluorescent protein (GFP). Right panel: Southern blot of BM DNA of three leukemic MN1/Cbfbeta–SMMHC-transplanted mice (1–3), digested with EcoRI and hybridized with a GFP probe. All three samples contain multiple retroviral integrations. Owing to DNA overloading of sample 3, a shorter exposure of this lane is shown. (d) Fluorescence micrographs of BM cytospin preparations of a diseased mouse containing mainly c-Kit+ BM cells (left panel) and a mouse containing partial c-Kit+ BM cells (right panel), double stained with antibodies specific for the core-binding factor-beta encoding the smooth muscle myosin heavy chain (CBFbeta–SMMHC) fusion peptide (red) and MN1 (green). Most cells in the left panel are positive for both CBFbeta–SMMHC (cytoplasm) and MN1 signals (nucleus), whereas only part of the cells in the right panel are positive for both signals with the remainder of the cells only positive for the MN1 signal. The nuclei of the cells in the left panel were counter stained with 4'-6-diamidino-2-phenylindole (DAPI; blue).

Full figure and legend (415K)

All 12 mice receiving inv(16)/MN1 transplants died of hematopoietic disease 58–68 days after transplantation, whereas the 6 mice given vector-transduced inv(16) transplants remained well (Figure 4b). Leukemic cells infiltrated the spleen and lymph nodes and effaced their normal architecture (not shown). In addition, myeloid cells infiltrated the liver, brain, uterus, lungs, stomach and heart (not shown). Southern blot analysis of BM DNA of three of these mice with a GFP probe showed multiple integrations (Figure 4c), suggesting the disease was oligoclonal (see discussion). The diseased mice could be divided into those whose PB contained a predominance of blast cells (Figure 4b), and those whose PB contained partly blast cells and partly more differentiated myeloid cells (neutrophils and neutrophil precursors). Cell surface marker analysis by FACS (Figure 4c) showed that BM cells of the former type mostly expressed no markers other than c-Kit, with a small percentage of cells expressing Mac1 and Gr1, whereas the BM of the latter mice contained fewer c-Kit+ cells and more Mac1+ and Gr1+ cells. The expression of c-Kit in the MN1/Cbfbeta-SMMHC leukemic cells was in sharp contrast with the malignant cells of mice that received MN1-transduced BM, which expressed Mac1 and Gr1 but not c-Kit (Figure 3e). This result is similar to Cbfb-MYH11chimeric mice treated with ENU which also developed a c-Kit+ AML.24 GFP signal was present in 99% of all cells (Figure 4c). Immunofluorescense using an MN1 antibody and an antibody to the CBFbeta–SMMHC fusion break point33 to double label cytospin preparations of BM from both types of mice showed that in mice with mostly blasts in the PB, most BM cells expressed both MN1 and CBFbeta–SMMHC, whereas mice with fewer blasts and more differentiated cells in the PB also contained cells in the BM expressing MN1 only (Figure 4d). This suggested that mice whose PB contains more differentiated cells have a mixed AML/MPD in which the AML arose from inv(16) cells overexpressing MN1 and the MPD from wild-type cells overexpressing MN1, whereas mice with mostly blast cells in the PB have inv(16)/MN1 AML. This result showed that compared to MN1 overexpression alone, the combined expression of MN1 and CBFbeta–SMMHC neither changed the latency nor the penetrance of the disease but had a clear effect on the phenotype of the disease. We next assessed whether the level of MN1 expression in the BM of our MN1/Cbfbeta–SMMHC transplanted mice would be comparable with that in inv(16) patient BM. Using QRT–PCR with human MN1 primers (our mouse Mn1 primers do not amplify human MN1 cDNA, not shown), we determined the amount of MN1 mRNA in the leukemic BM of four Cbfbeta–SMMHC/MN1 transplanted mice, in BM of three inv(16) patients and in the inv(16) M4E cell line ME-1.34 This showed that MN1 expression in these 4 Cbfbeta–SMMHC/MN1 BM samples was 0.7- to 3.9-fold as abundant as in the patient sample with the highest MN1 expression (Figures 1 and 5a). Thus, MN1 expression in the BM of three of the mice was considerably higher than in inv(16) patients, but in one mouse comparable expression also appeared sufficient to provoke disease.

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

Comparison of MN1 expression in bone marrow (BM) of inv(16) patients and diseased mice transplanted with Cbfbeta–SMMHC/MN1-expressing BM. Quantitative real-time (QRT)–PCR with primers specific for human MN1 was used to determine the level of MN1 RNA expression in the human inv(16) cell line ME-1 (ME-1), in pooled BM of healthy individuals, in BM of three inv(16) patients (M4E) and in BM of four diseased Cbfbeta–SMMHC/MN1-transplanted mice (inv(16)/MN1). As a negative control, we used BM of a mouse transplanted with vector-transduced Cbfbeta–SMMHC chimeric BM (inv(16)). Expression levels are depicted as fold expression of MN1 in normal human BM. MN1 expression in BM of Cbfbeta–SMMHC/MN1-transplanted mice was between 0.7- to 3.9-fold higher than in BM of the inv(16) patient with the highest level of MN1 expression (speckled box M4E). Results are the average of two experiments.

Full figure and legend (37K)



Myeloid leukemogenesis is a multistep process,26 and it has been well documented that in AML, specified by recurrent chromosome translocations, cooperating mutations are essential for disease development.13, 24, 27, 35, 36, 37 In expression profiling experiments, the level of MN1 mRNA was specifically elevated in BM samples of pediatric and adult patients with inv(16) AML.15, 16 MN1 upregulation was also found in a subtype of AML associated with very poor prognosis,38 which is defined by overexpression of the immortalizing39 transcription factor EVI1.16 Given that MN1 sequences confer oncogenic properties to the MN1-TEL fusion protein3, 12 the product of the t(12;22) in AML, it opened the possibility that overexpression of MN1 in these AML subtypes actively contributed to the leukemic process. This suggestion was supported by the observation that retrovirus-mediated overexpression of MN1 in mouse BM boosted proliferation of myeloid cells, allowed the establishment of myeloid cell lines in vitro and caused rapid development of MPD after transplantation into lethally irradiated mice. Moreover, the MN1 effects were cell intrinsic as secondary recipients receiving transplants of BM of diseased primary recipients rapidly developed the same disease. Despite the rapid development of the primary MPD, Southern blot analysis of the malignant cells showed that the disease was oligoclonal rather than polyclonal, which implied that additional genetic changes must have occurred for the disease to develop. Currently we do not know the nature of these additional genetic changes.

Expression of endogenous Mn1 is present in the sorted HSC but not in the CMP, CLP, MEP progenitor compartments. Given the profound proliferative effect of human MN1 overexpression on GMP-derived cells in mouse BM, we speculate that expansion of this compartment might be dependent on endogenous upregulation of Mn1. Therefore, forced expression using the MN1 retroviral vector would lead to overexpansion. Why MN1 overexpression would specifically affect the proliferation of GMP-derived but not MEP-derived cells is currently unknown. Mn1 knockout mice have defects in the development of membranous bones of the cranial skeleton,40 but whether these mice also harbor hematopoietic defects is a question we are currently addressing.

We also do not know via which mechanism MN1 stimulates growth of myelomonocytic cells. We reported that in Hep3B cells, the protein is recruited to RAR/RXR dimers via the co-activators p300/CBP and RAC34 and stimulates the transcription activity of RAR/RXR in the presence of retinoic acid. It is possible that in myeloid cells, MN1 participates in a similar protein complex or in protein complexes with other transcription factors that recruit p300/CBP, such as MYB; RUNX1; GATA1, 2 and 3; C/EBP; PU.1 and MLL.41 Growth stimulation of myeloid cells by MN1 is opposite to its reported effects in an osteoblast cell line,7 in which MN1-mediated co-activation of the vitamin D receptor inhibited proliferation. Growth inhibition of several types of epithelial cells by TGF-beta, an inhibitor of epithelial cell proliferation, is also associated with induction of MN1 expression.42, 43 This profound difference in response to MN1 upregulation can only be explained if MN1 can be recruited into different transcription factor complexes whose effects are cell type specific.

Previously we reported that BM of MN1-TEL knock-in mice showed increased self-renewal activity of myeloid progenitors13 and produced myeloid cell lines in vitro. The same effects were observed with BM transduced with MN1-TEL retrovirus. In both scenarios, MN1-TEL cell lines displayed a more primitive phenotype (c-Kit+/Sca1+)12, 13 than the MN1 cell lines (c-Kit+/Sca1+/Mac1+) reported here. We speculate that this difference is caused by the presence of TEL ETS domain in the fusion protein, which might recruit MN1-TEL to TEL recognition sites at promoter/enhancer areas of genes that inhibit differentiation.

As determined by QRT–PCR, MN1 mRNA levels in inv(16) FAB-M4 samples were much greater than that determined by expression profiling. QRT–PCR analysis showed that MN1 mRNA expression in our 9 inv(16) patient samples was on average 43.7-fold higher than in normal BM, while estimations using array analysis suggested an average 4.6-15 or 9.2-fold16 increase. This discrepancy leads to a distinct underestimation of MN1 expression using expression arrays. Indeed, inclusion of other AML subtypes in our QRT–PCR analysis showed substantial upregulation of MN1 in FAB-M1 samples (14.1-fold average) and moderate (two- to eightfold) upregulation in all M2-, 1 M7-, 2 M3- and 1 M5 sample(s), not reported in the study by Ross et al.15 which included the very same patient samples. MN1 expression was also moderately increased in all but two non-inv(16) FAB-M4 samples. QRT–PCR of a much larger group of AML samples will have to determine whether it will support our initial results. Given the small numbers of patients analyzed it was impossible to obtain a statistically significant correlation between levels of MN1 expression and the survival rates of inv(16) and FAB-M1 patients. Nonetheless, it was reported that MN1 overexpression in adult patient samples with a normal karyotype correlated with a worse prognosis and a shorter survival rate.17

Immunofluorescence analysis with an MN1 monoclonal antibody of two of our inv(16) BM samples, expressing 50- and 25-fold more MN1 mRNA than normal BM, showed that the MN1 signal was elevated in leukemic cells, but the difference was much more modest than expected from the increase in MN1 mRNA levels. It is possible that not all MN1 mRNA is translated in these AML cells, or alternatively that MN1 overexpression increases its protein turnover. To date all inv(16) patient samples analyzed showed upregulated MN1 expression (Ross et al.15 and Valk et al.16) strongly suggesting that overexpression of MN1 is an obligatory step in the development of this disease.

An important question that remains to be answered is which molecular mechanism provokes MN1 overexpression. Based on the observation that expression of the endogenous mouse Mn1 gene was similar in Cbfbeta–MYH11chimeric and normal BM (Figure 4a), we do not think that the gene is a direct transcriptional target of the CBF transcription factor. Given that the promoter regions of the mouse and human MN1 genes show extensive sequence conservation, their transcriptional regulation is likely to be similar. Also, MN1 upregulation was reported in AML patients with a normal karyotype,17 further indicating that other genetic changes than expression of CBF fusion proteins are responsible for MN1 overexpression. Using PCR amplification of MN1, first exon sequences in the genomic DNA of the two patient samples that overexpressed MN1 protein (Figure 1a, indicated in red) and of normal BM samples revealed a diploid copy number in all three samples (not shown). Therefore, the increased MN1 expression in these two patients was not the result of MN1 gene amplification.

Cbfbeta–SMMHC chimeric knock-in mice do not develop myeloid leukemia,24, 28 whereas mice carrying a conditional Cbfbeta–SMMHC knock-in gene do develop disease within 3–6 months after induction. This difference is most likely caused by the increased size of the Cbfbeta–SMMHC+ preleukemic progenitor pool in the BM of the conditional knock-in mice.44 Thus, the increased proliferative capacity of MN1 overexpressing cells may enlarge the pool of inv(16) cells enough to promote additional mutations allowing the leukemia to emerge. We favor this possibility because Southern blotting suggested that mouse inv(16)/MN1 leukemia is oligoclonal rather than polyclonal, although our analysis cannot exclude that the oligoclonality derives from the concomitant MN1-induced MPD in these transplanted mice. Irrespective of whether the MN1/inv(16) disease is monoclonal or oligoclonal, the finding that it is not polyclonal strongly suggests that additional mutations must occur during AML development in the transplanted mice. Known candidate genes are Plag1 and PlagL2, which were identified as Cbfb-MYH11cooperating genes in mouse inv(16) AML,28 and were found to be upregulated in human inv(16) AML.16 We do not think Plag1 or PlagL2 are transcriptional targets of MN1 because the mRNA levels of these genes were not increased in QRT–PCR analysis of MN1 overexpressing mouse cell lines, MN1 MPD BM or inv(16)/MN1 BM. The same holds true for PLAGL2 in inv(16) leukemia as we did not find any correlation between the levels of PLAGL2 and MN1 expression in our patient samples (Supplementary Figure 1C), of which only one showed increased expression of PLAGL2. We also investigated the mutation status of the Npm gene in BM of the three MN1 MPD mice shown in Figure 3c and the three inv(16)/MN1 AML mice shown in Figure 4c. Mutation in codons 288 or 290 of NPM causes relocalization of the protein from the nucleolus to the cytoplasm in 35% of AML patients,45 which affects the p53 tumor suppressor pathway activity.46 Sequencing of PCR-amplified Npm cDNA of these eight mice showed that the gene was not mutated (not shown).

Given the cooperation between MN1 overexpression and CBFbeta–SMMHC in inv(16) AML, interference with MN1 function might provide a novel therapeutic approach for this CBF leukemia. In addition, it will be interesting to determine whether MN1 overexpression similarly promotes leukemia development in a mouse model for t(8;21) AML, which also targets the Cbf transcription factor. On the other hand, MN1 overexpression is much higher in inv(16) AML than in t(8;21) CBF leukemia, opening the possibility that the products of these two chromosome translocations function quite differently despite the fact that they target the same transcription factor complex.

Together our experiments suggest that the differentiation-inhibiting protein Cbfbeta–SMMHC and the proliferation-stimulating protein MN1 cooperate in the development of AML in this mouse model. Given that MN1 expression is upregulated in all inv(16) patients investigated (Ross et al.15 and Valk et al.16), it is reasonable to speculate that these two proteins play a similar cooperative role in human inv(16) AML.



  1. Lekanne Deprez RH, Groen NA, van Biezen NA, Hagemeijer A, van Drunen E, Koper JW et al. A t(4;22) in a meningioma points to the localization of a putative tumor-suppressor gene. Am J Hum Genet 1991; 48: 783–790. | PubMed | ChemPort |
  2. Lekanne Deprez RH, Riegman PH, Groen NA, Warringa UL, van Biezen NA, Molijn AC et al. Cloning and characterization of MN1, a gene from chromosome 22q11, which is disrupted by a balanced translocation in a meningioma 8. Oncogene 1995; 10: 1521–1528. | PubMed | ChemPort |
  3. Buijs A, van Rompaey L, Molijn AC, Davis JN, Vertegaal AC, Potter MD et al. The MN1-TEL fusion protein, encoded by the translocation (12;22)(p13;q11) in myeloid leukemia, is a transcription factor with transforming activity 24. Mol Cell Biol 2000; 20: 9281–9293. | Article | PubMed | ISI | ChemPort |
  4. van Wely KH, Molijn AC, Buijs A, Meester-Smoor MA, Aarnoudse AJ, Hellemons A et al. The MN1 oncoprotein synergizes with coactivators RAC3 and p300 in RAR-RXR-mediated transcription 5. Oncogene 2003; 22: 699–709. | Article | PubMed | ISI | ChemPort |
  5. Chen Z, Fisher RJ, Riggs CW, Rhim JS, Lautenberger JA. Inhibition of vascular endothelial growth factor-induced endothelial cell migration by ETS1 antisense oligonucleotides 10. Cancer Res 1997; 57: 2013–2019. | PubMed | ISI | ChemPort |
  6. Leo C, Chen JD. The SRC family of nuclear receptor coactivators 1. Gene 2000; 245: 1–11. | Article | PubMed | ISI | ChemPort |
  7. Sutton AL, Zhang X, Ellison TI, Macdonald PN. The 1,25(OH)2D3-regulated transcription factor MN1 stimulates vitamin D receptor-mediated transcription and inhibits osteoblastic cell proliferation 9. Mol Endocrinol 2005; 19: 2234–2244. | Article | PubMed | ISI | ChemPort |
  8. 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 2. Cell 1994; 77: 307–316. | Article | PubMed | ISI | ChemPort |
  9. Chakrabarti SR, Nucifora G. The leukemia-associated gene TEL encodes a transcription repressor which associates with SMRT and mSin3A 3. Biochem Biophys Res Commun 1999; 264: 871–877. | Article | PubMed | ISI | ChemPort |
  10. Fenrick R, Amann JM, Lutterbach B, Wang L, Westendorf JJ, Downing JR et al. Both TEL and AML-1 contribute repression domains to the t(12;21) fusion protein 10. Mol Cell Biol 1999; 19: 6566–6574. | PubMed | ISI | ChemPort |
  11. Lopez RG, Carron C, Oury C, Gardellin P, Bernard O, Ghysdael J. TEL is a sequence-specific transcriptional repressor 42. J Biol Chem 1999; 274: 30132–30138. | Article | PubMed | ISI | ChemPort |
  12. Carella C, Potter M, Bonten J, Rehg JE, Neale G, Grosveld GC. The ETS factor TEL2 is a hematopoietic oncoprotein 3. Blood 2006; 107: 1124–1132. | Article | PubMed | ISI | ChemPort |
  13. Kawagoe H, Grosveld GC. Conditional MN1-TEL knock-in mice develop acute myeloid leukemia in conjunction with overexpression of HOXA9 13. Blood 2005; 106: 4269–4277. | Article | PubMed | ISI | ChemPort |
  14. Kawagoe H, Grosveld GC. MN1-TEL myeloid oncoprotein expressed in multipotent progenitors perturbs both myeloid and lymphoid growth and causes T-lymphoid tumors in mice 13. Blood 2005; 106: 4278–4286. | Article | PubMed | ISI | ChemPort |
  15. Ross ME, Mahfouz R, Onciu M, Liu HC, Zhou X, Song G et al. Gene expression profiling of pediatric acute myelogenous leukemia 12. Blood 2004; 104: 3679–3687. | Article | PubMed | ISI | ChemPort |
  16. Valk PJ, Verhaak RG, Beijen MA, Erpelinck CA, Barjesteh van Waalwijk van Doorn-Khosrovani S, Boer JM et al. Prognostically useful gene-expression profiles in acute myeloid leukemia 16. N Engl J Med 2004; 350: 1617–1628. | Article | PubMed | ISI | ChemPort |
  17. Heuser M, Beutel G, Krauter J, Dohner K, von Neuhoff N, Schlegelberger B et al. High meningioma 1 (MN1) expression as a predictor for poor outcome in acute myeloid leukemia with normal cytogenetics. Blood 2006; 108: 3898–3905. | Article | PubMed | ISI | ChemPort |
  18. Liu PP, Hajra A, Wijmenga C, Collins FS. Molecular pathogenesis of the chromosome 16 inversion in the M4Eo subtype of acute myeloid leukemia 9. Blood 1995; 85: 2289–2302. | PubMed | ISI | ChemPort |
  19. Otto F, Lubbert M, Stock M. Upstream and downstream targets of RUNX proteins 1. J Cell Biochem 2003; 89: 9–18. | Article | PubMed | ISI | ChemPort |
  20. Castilla LH, Wijmenga C, Wang Q, Stacy T, Speck NA, Eckhaus M et al. Failure of embryonic hematopoiesis and lethal hemorrhages in mouse embryos heterozygous for a knocked-in leukemia gene CBFB-MYH11 4. Cell 1996; 87: 687–696. | Article | PubMed | ISI | ChemPort |
  21. Okuda T, van Deursen J, Hiebert SW, Grosveld G, Downing JR. AML1, the target of multiple chromosomal translocations in human leukemia, is essential for normal fetal liver hematopoiesis. Cell 1996; 84: 321–330. | Article | PubMed | ISI | ChemPort |
  22. Sasaki K, Yagi H, Bronson RT, Tominaga K, Matsunashi T, Deguchi K et al. Absence of fetal liver hematopoiesis in mice deficient in transcriptional coactivator core binding factor beta 22. Proc Natl Acad Sci USA 1996; 93: 12359–12363. | Article | PubMed | ChemPort |
  23. Wang Q, Stacy T, Binder M, Marin-Padilla M, Sharpe AH, Speck NA. Disruption of the Cbfa2 gene causes necrosis and hemorrhaging in the central nervous system and blocks definitive hematopoiesis 8. Proc Natl Acad Sci USA 1996; 93: 3444–3449. | Article | PubMed | ChemPort |
  24. Castilla LH, Garrett L, Adya N, Orlic D, Dutra A, Anderson S et al. The fusion gene Cbfb-MYH11 blocks myeloid differentiation and predisposes mice to acute myelomonocytic leukaemia 2. Nat Genet 1999; 23: 144–146. | Article | PubMed | ISI | ChemPort |
  25. Higuchi M, O'Brien D, Kumaravelu P, Lenny N, Yeoh EJ, Downing JR. Expression of a conditional AML1-ETO oncogene bypasses embryonic lethality and establishes a murine model of human t(8;21) acute myeloid leukemia 1. Cancer Cell 2002; 1: 63–74. | Article | PubMed | ISI | ChemPort |
  26. Look AT. Oncogenic transcription factors in the human acute leukemias 5340. Science 1997; 278: 1059–1064. | Article | PubMed | ISI | ChemPort |
  27. Castilla LH, Perrat P, Martinez NJ, Landrette SF, Keys R, Oikemus S et al. Identification of genes that synergize with Cbfb-MYH11 in the pathogenesis of acute myeloid leukemia 14. Proc Natl Acad Sci USA 2004; 101: 4924–4929. | Article | PubMed | ChemPort |
  28. Landrette SF, Kuo YH, Hensen K, Barjesteh van Waalwijk van Doorn-Khosrovani S, Perrat PN, Van de Ven WJ et al. Plag1 and Plagl2 are oncogenes that induce acute myeloid leukemia in cooperation with Cbfb-MYH11 7. Blood 2005; 105: 2900–2907. | Article | PubMed | ISI | ChemPort |
  29. Cardone M, Kandilci A, Carella C, Nilsson JA, Brennan JA, Sirma S et al. The novel ETS factor TEL2 cooperates with Myc in B lymphomagenesis 6. Mol Cell Biol 2005; 25: 2395–2405. | Article | PubMed | ISI | ChemPort |
  30. Akashi K, Traver D, Miyamoto T, Weissman IL. A clonogenic common myeloid progenitor that gives rise to all myeloid lineages 6774. Nature 2000; 404: 193–197. | Article | PubMed | ISI | ChemPort |
  31. Stanley KK, Szewczuk E. Multiplexed tandem PCR: gene profiling from small amounts of RNA using SYBR Green detection 20. Nucleic Acids Res 2005; 33: e180. | Article | PubMed | ChemPort |
  32. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR et al. Proposal for the recognition of minimally differentiated acute myeloid leukaemia (AML-MO) 3. Br J Haematol 1991; 78: 325–329. | PubMed | ISI | ChemPort |
  33. Wijmenga C, Gregory PE, Hajra A, Schrock E, Ried T, Eils R et al. Core binding factor beta-smooth muscle myosin heavy chain chimeric protein involved in acute myeloid leukemia forms unusual nuclear rod-like structures in transformed NIH 3T3 cells. Proc Natl Acad Sci USA 1996; 93: 1630–1635. | Article | PubMed | ChemPort |
  34. Yanagisawa K, Horiuchi T, Fujita S. Establishment and characterization of a new human leukemia cell line derived from M4E0 2. Blood 1991; 78: 451–457. | PubMed | ISI | ChemPort |
  35. Ayton PM, Cleary ML. Transformation of myeloid progenitors by MLL oncoproteins is dependent on Hoxa7 and Hoxa9 18. Genes Dev 2003; 17: 2298–2307. | Article | PubMed | ISI | ChemPort |
  36. He LZ, Bhaumik M, Tribioli C, Rego EM, Ivins S, Zelent A et al. Two critical hits for promyelocytic leukemia 5. Mol Cell 2000; 6: 1131–1141. | Article | PubMed | ISI | ChemPort |
  37. Largaespada DA. Genetic heterogeneity in acute myeloid leukemia: maximizing information flow from MuLV mutagenesis studies 7. Leukemia 2000; 14: 1174–1184. | Article | PubMed | ISI | ChemPort |
  38. 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 25. Oncogene 2005; 24: 4129–4137. | Article | PubMed | ChemPort |
  39. Du Y, Jenkins NA, Copeland NG. Insertional mutagenesis identifies genes that promote the immortalization of primary bone marrow progenitor cells. Blood 2005; 106: 3932–3939. | Article | PubMed | ISI | ChemPort |
  40. Meester-Smoor MA, Vermeij M, van Helmond MJ, Molijn AC, van Wely KH, Hekman AC et al. Targeted disruption of the Mn1 oncogene results in severe defects in development of membranous bones of the cranial skeleton 10. Mol Cell Biol 2005; 25: 4229–4236. | Article | PubMed | ISI | ChemPort |
  41. Blobel GA. CREB-binding protein and p300: molecular integrators of hematopoietic transcription. Blood 2000; 95: 745–755. | PubMed | ISI | ChemPort |
  42. Chen CR, Kang Y, Massague J. Defective repression of c-myc in breast cancer cells: A loss at the core of the transforming growth factor beta growth arrest program 3. Proc Natl Acad Sci USA 2001; 98: 992–999. | Article | PubMed | ChemPort |
  43. Kang Y, Chen CR, Massague J. A self-enabling TGFbeta response coupled to stress signaling: Smad engages stress response factor ATF3 for Id1 repression in epithelial cells 4. Mol Cell 2003; 11: 915–926. | Article | PubMed | ISI | ChemPort |
  44. Kuo YH, Landrette SF, Heilman SA, Perrat PN, Garrett L, Liu PP et al. Cbf beta-SMMHC induces distinct abnormal myeloid progenitors able to develop acute myeloid leukemia 1. Cancer Cell 2006; 9: 57–68. | Article | PubMed | ISI | ChemPort |
  45. Grisendi S, Pandolfi PP. NPM mutations in acute myelogenous leukemia 3. N Engl J Med 2005; 352: 291–292. | Article | PubMed | ISI | ChemPort |
  46. den Besten W, Kuo ML, Williams RT, Sherr CJ. Myeloid leukemia-associated nucleophosmin mutants Perturb p53-dependent and independent activities of the Arf tumor suppressor protein 11. Cell Cycle 2005; 4: 1593–1598. | PubMed | ISI | ChemPort |


We thank Dr Jerold Rehg for the pathological analysis of the mice, Dr Richard Ashmun and Ann-Mary Hamilton Easton for expert FACS analysis and Blake McGourty for the supply of C57Bl/6/129svJ mixed-background mice. This work was supported by NCI Grant CA72999, the Cancer Center (CORE) support grant CA021765, the Dutch Cancer Society Grants EUR98-1778 and DDHK2003-2869, the Intramural Research Programs of NHGRI, NIH and by the American Lebanese Syrian Associated Charities (ALSAC).

Supplementary information accompanies the paper on the Leukemia Web site (http://www.nature.com/leu)