Email updates

Keep up to date with the latest news and content from BMC Medical Genomics and BioMed Central.

Open Access Research article

Gene profiling of the erythro- and megakaryoblastic leukaemias induced by the Graffi murine retrovirus

Veronique Voisin12, Philippe Legault1, Diana Paulina Salazar Ospina1, Yaacov Ben-David2 and Eric Rassart1*

Author Affiliations

1 Laboratoire de Biologie Moléculaire, Département des Sciences Biologiques, Centre BioMed, Université du Québec à Montréal, Case Postale 8888 Succursale Centre-ville, Montréal, QC, H3C-3P8, Canada

2 Sunnybrook Health Sciences Center, 2075 Bayview Ave. S223B, Toronto. ON, M4N 3M5, Canada

For all author emails, please log on.

BMC Medical Genomics 2010, 3:2  doi:10.1186/1755-8794-3-2

The electronic version of this article is the complete one and can be found online at: http://www.biomedcentral.com/1755-8794/3/2


Received:17 August 2009
Accepted:26 January 2010
Published:26 January 2010

© 2010 Voisin et al; licensee BioMed Central Ltd.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background

Acute erythro- and megakaryoblastic leukaemias are associated with very poor prognoses and the mechanism of blastic transformation is insufficiently elucidated. The murine Graffi leukaemia retrovirus induces erythro- and megakaryoblastic leukaemias when inoculated into NFS mice and represents a good model to study these leukaemias.

Methods

To expand our understanding of genes specific to these leukaemias, we compared gene expression profiles, measured by microarray and RT-PCR, of all leukaemia types induced by this virus.

Results

The transcriptome level changes, present between the different leukaemias, led to the identification of specific cancerous signatures. We reported numerous genes that may be potential oncogenes, may have a function related to erythropoiesis or megakaryopoiesis or have a poorly elucidated physiological role. The expression pattern of these genes has been further tested by RT-PCR in different samples, in a Friend erythroleukaemic model and in human leukaemic cell lines.

We also screened the megakaryoblastic leukaemias for viral integrations and identified genes targeted by these integrations and potentially implicated in the onset of the disease.

Conclusions

Taken as a whole, the data obtained from this global gene profiling experiment have provided a detailed characterization of Graffi virus induced erythro- and megakaryoblastic leukaemias with many genes reported specific to the transcriptome of these leukaemias for the first time.

Background

Human acute megakaryoblastic (FAB-AML7, [1]) and erythroleukaemias (FAB-AML6, [2]) are regarded as relatively rare entities of acute myeloid leukaemia but are associated with a very poor prognosis [3-7]. The poor outcome linked to these 2 types of leukaemias stems from a combination of failure to achieve complete remission, a high relapse rate and therapy-related toxicity, highlighting the need for more powerful therapies. Furthermore, AML6 or AML7 diagnosis represents a greater challenge than other types of acute myeloid leukaemia (AML) and additional markers are needed [8]. Furthermore, the blasts of patients with AML6 and AML7 share common markers [9] indicating that they originate from closely related haematopoietic lineages derived from a common bipotent progenitor [10,11].

We have recently shown that the murine retrovirus Graffi is able to induce a broad spectrum of leukaemias when inoculated into newborn mice. The leukaemias developed by these mice are of lymphoid (T-cell and B-cell) and non lymphoid (myeloid, erythroid and megakaryoblastic) origins. The incidence of erythro- and megakaryoblastic leukaemias is particularly high in NFS or FVB/n mice inoculated with the GV-1.4 variant of the Graffi virus [12]. The activation of the targeted proto-oncogene or the repression of tumor suppressor genes represents early events in the development of the murine leukaemia retrovirus (MuLV) induced leukaemia. It is then followed by a deregulation of numerous additional genes resulting in a cell, blocked at a very immature stage, which aggressively divides and escapes apoptosis. To analyze these cancerous signatures, we compared the gene profiles of each type of leukaemia (T-cell, B-cell, myeloid, erythroid, megakaryoblastic) induced by the Graffi virus. These analyses highlight many genes that may be potential oncogenes and may have a function related to erythropoiesis or megakaryopoiesis. The results support the importance of the known transcription factors Gata1, Fog1, Fli1, Scl and Lmo2 in both erythro- and megakaryoblastic leukaemias and the role of Runx1, Pbx1, Meis, Evi1 and Evi3 in the megakaryoblastic leukaemias. Moreover, numerous genes are being reported for the first time and some of these genes are candidate oncogenes: Fgf3, Nmyc, Fap, Myct1, Gucy1a3, Gulp1 and Fkbp9 specific to megakaryoblastic leukaemias and Ssx2ip, Rab11a, Ncoa3, Snca, Ltbp2, Rabgef1 and Btbd14a specific to erythroleukaemias. A screening for viral integrations was performed in mouse tumors. Several genes, amongst which Kit, Gata2, Irf8 and Itga1, were identified as potentially implicated in the onset development of the megakaryoblastic leukaemias.

Methods

Virus production and mice

GV-1.4 viral stock was made as previously described [12]. GV-1.4 viral particles (0.1 ml at a titer of 1.106 PFU/ml) were injected into 1 day newborn NFS mice. The mice were checked routinely for clinical signs of disease and moribund mice were sacrificed. Twenty-four diseased mice and 36 control mice were used for the microarray and RT-PCR experiments. Bone marrow cell suspension was prepared by flushing the femurs with IMDM 2% foetal bovine serum (FBS) and spleen cell suspension was prepared by mincing the spleen with scissors and aspirating the pieces up and down through a 1cc syringe in IMDM 2% FBS. The spleen and bone marrow cell suspensions were filtered through 70 μm cell strainers (Becton Dickson, Mississauga, Canada). All the experimental procedures are conformed to the Helsinki Declaration and were approved by the Animal Care and Use Committees of Université du Québec à Montréal.

Flow cytometry analyses and cell sorting

The flow cytometry staining procedure was performed as previously described [12]. The antibodies used were as follows: CD4, CD8a, CD3, CD90, CD19, B220, CD11b, Gr1, CD71, Ter119, CD41, Kit and Sca1 (BD Pharmingen, Mississauga, Canada). The leukaemic populations were isolated from the haematopoietic organs by positive selection using magnetic beads with the EasySep Kit (StemCell Technologies, Vancouver, Canada) according to the manufacturer's protocol. The rates of purity and viability of the sorted cells were fixed to be equal to or greater than 95%. Leukaemic T-cells were sorted from the thymus of leukaemic NFS mice, B-cell from the enlarged lymph nodes and erythro- and megakaryoblastic leukaemic cells were sorted from the infiltrated spleen. Control cells were sorted from the haematopoietic organs of 12 pooled non-infected NFS mice: T-cells were obtained from the thymus, B-cells from the spleen, and erythroblasts from the bone marrow.

RNA extraction and microarray processing

Total RNA was extracted from the sorted cell populations with Trizol reagent (Invitrogen, Burlington, Canada) followed by column purification using the Qiagen RNeasy Kit (QIAGEN, Mississauga, Canada) and processed for hybridization to Affymetrix GeneChip® Mouse Genome 430 2.0 arrays (Genome Quebec Innovation Centre, Montreal, Canada).

Data analysis

Data Set Normalization: Affymetrix MicroArray Suite version 5.0 was used to scan and quantify the arrays. Normalization of gene expression data were performed using the Bioconductor implementation of RMA (Robust Multi Array, B. Bolstad, University of California, Berkeley) available from the Flexarray software (1.2, R 2.7.2, [13]).

Unsupervised learning: Hierarchical clustering (complete linkage clustering, correlation uncentered, [14]) and Self-Organization Maps (SOM, parameters G 1-5, A 1-10, [15]) were constructed using GeneCluster software (M. Eisen). 3,000 transcripts were selected to be included in the analyses based on the differential expression from the mean. The deviation from the mean was calculated from the RMA values of the 45,000 probesets and the data were ranked in decreasing order to extract the first 3,000 genes. Only deviations equal or above 0.585 (1.5 fold change) and equal or below -0.585 (-1.5 fold change) were considered as significant.

Supervised learning: Significance Analysis of microarrays (SAM, [16]). SAM analyses were performed using Flexarray software using the normalized data of the 45,000 probesets. Data with p-values equal or below 0.01 and False Discover Rates (Benjamini Hochberg) equal or below 0.20 were included in further analyses. The data were ranked in decreasing order of the SAM d-score to obtain the list of the differentially expressed genes.

The NetAffx website (Affymetrix, Santa Clara, CA, USA) was used to retrieve gene ontology (GO) annotations, probe sequences, and was utilized as a link to Unigene (NCBI) for further functional studies.

The microarray dataset was deposited at Gene Expression Omnibus under accession number [GSE12581].

Cell line and differentiation assay

The murine erythroleukaemic cell line HB22.2 was obtained from murine erythroblasts infected with Friend Murine Leukaemia Virus (F-MuLV) [17]. This cell line was maintained in alpha minimum essential medium (α-MEM) supplemented with 10% (FBS) (Invitrogen, Frederick, MD) plus a penicillin/streptomycin cocktail. To induce differentiation, HB22.2 cells were incubated in the presence of hemin (Sigma-aldrich H5533) at a concentration of 100 μM. The cells were harvested 24 hours and 72 hours after addition of hemin. K562 (ATCC, USA), HEL (ATCC, USA), Jurkat (ATCC, USA) and Tk6 (ATCC, USA) cells were grown in RMPI supplemented with 10% FBS plus a penicillin/streptomycin cocktail. MEG-01 (ATCC, USA), CMK (DSMZ, Germany) and LAMA84 (DSMZ, Germany) cells were grown in RPMI supplemented with 20% FBS plus a penicillin/streptomycin cocktail with a concentration of 105 cells/ml.

RT-PCR

Oligo d(T) primed reverse transcription was performed using Omniscript Reverse Transcriptase (QIAGEN, Mississauga, Canada) in a 20 μl reaction volume (42°C, 1 h) by taking equal amounts of RNA (100 ng) from the Graffi-leukaemic cells and the murine HB22.2 cell line. cDNA (4 μl) was amplified using Taq polymerase (QIAGEN, Mississauga, Canada) at 94°C for 5 min, 72°C for 45 s, 57°C for 45 s, 72°C for 45 s, 72°C for 10 min. 25 and 28 cycles were used for the Graffi-leukaemic cells and 27 cycles were used for the HB22.2 cell line. 0.01 μl of cDNA and 25 cycles were used to amplify ubiquitously expressed β-actin and Gapdh genes. cDNA from the human haematopoietic cell lines was amplified using 500 ng of total RNA and the PCR reactions were performed using 4 μl of cDNA and 30 PCR cycles. Ubiquitously expressed human GAPDH gene was amplified using 0.01 μl of cDNA and 25 cycles. The primer sets are listed in supplementary data (Additional file 1). PCR products were analyzed on 2% agarose gels containing 0.5 μg/ml ethidium bromide. The gels were scanned (Molecular Dynamics Phosphorimager) and the density of the RT-PCR bands were assessed using the Quantity One software.

Additional file 1. Oligonucleotides utilized in the RT-PCR experiments. sequences of forward and reverse primers utilized in the RT-PCR experiments

Format: DOC Size: 47KB Download file

This file can be viewed with: Microsoft Word ViewerOpen Data

Amplification of retroviral integration sites

This protocol was adapted from A. Berns and colleagues [18]. Tumor DNA from the spleen (10 μg) was digested with BamHI (New England Biolabs, Pickering, Canada). A splinkerette adaptor was generated by annealing 2 oligonucleotides (5'CATGGGCTAAAGAGGACTAATAACAAGCGTGGCTGAATGAGACTGGTGTCGA CACTAGTGG3', 5'GATCCCACTAGTGTCGACACCAGTCTCTAATTTTTTTTTTC AAAAAAA3', 95°C 5 min, cool-down to room temperature). 1 μg of digested DNA was ligated to the splinkerette oligonucleotide (molar ratio DNA/splinkerette 1:10) using a T4 DNA ligase (New England Biolabs, Pickering, Canada). The ligated product was then digested with ClaI (New England Biolabs, Pickering, Canada). A PCR followed by a nested PCR (150 ng of the ligated product) were performed using a primer located in the Graffi virus U3 sequence (5'GGTCTCTTGAAACTGCTGAGGG 3' and 5'GACCTTGATCTGAACTTCCCTATTC3') and one corresponding to the splinkerette oligonucleotide. The PCR program was the following: 94°C for 1 min, 68°C for 30 s (2 cycles), 94°C for 15 s, 58°C for 30 s, 72°C 3 min (27 cycles), 72°C 10 min. The PCR products were cloned (PCR Cloning Kit, Qiagen, Mississauga, Canada,) and sent for sequencing (Genome Quebec Innovation Centre, Montreal, Canada). The sequences were subjected to BLAST analysis against the annotated mouse genome database using Ensembl Genome Browser (release 45).

Results

Erythro- and megakaryoblastic leukaemias induced by the murine Graffi retrovirus and hybridization on microarrays

NFS newborn mice inoculated with the Graffi murine retrovirus develop an average of 20% of erythroleukemia and 20% of megakaryoblastic leukemias with a latency of about 148 days [12]. These mice suffer from severe anaemia and hepatosplenomegaly. The erythroleukaemic cells, Ter119+CD71+, and the megakaryoblastic leukaemic cells, CD41+Kit+ or CD41-Kit+, are mainly found in the bone marrow and spleen of the diseased mice [12]. As opposed to Graffi-lymphoid leukaemias, the presence of blast cells is rare in blood smears of both erythro- and megakaryoblastic leukaemias, consistent with clinical data on human acute erythroleukaemia [8].

To gain insight into the cancerous signatures of the different leukaemias induced by Graffi MuLV, microarray experiments were designed to compare the expression signature of genes from each type of leukaemia. Cells from the infiltrated haematopoietic organs of the leukaemic mice were isolated (Additional file 2) and subjected to microarray analysis. Unsupervised learning methods, hierarchical clustering, and SOM analyses were used to uncover the primary pattern in the data (Figure 1). Altogether, four distinct gene clusters representing T-cells (T), B-cells (B), megakaryoblastic/myeloid cells (Mk/M) and erythroid cells (E) emerged from the clustering (Figure 1A).

thumbnailFigure 1. Hierarchical clustering and data scatter. (A) Heat map of the hierarchical clustering and SOM analyses of all data. (B) Heat map of the clustering and SOM analyses of the erythroid and megakaryoblastic over-expressed genes. Each column represents a leukaemia (T-cell: T1, T2, T3; B-cell: B1, B2, B3; Myeloid: M; Erythroid: E1, E2, E3; Megakaryoblastic: Mk1, Mk2, Mk3) and each row is assigned to a transcript. A red colour means a positive deviation of 0.595 and above from the mean (over-expression) and a green colour means a negative deviation of -0.595 and below from the mean (under-expression). The black colour corresponds to values comprised between 0.595 and -0.595. The result of the clustering on the arrays is shown as a dendrogram on the top of the figure. The nodes of the tree indicate the degrees of similarity between the leukaemias.

Additional file 2. Immunophenotype of the leukaemic samples selected for the microarray experiments. the table lists the leukaemias included in the microarray experiments including sample name, leukaemia type, immunophenotype, antibody used for sorting and tumor origins.

Format: DOC Size: 41KB Download file

This file can be viewed with: Microsoft Word ViewerOpen Data

A more detailed SOM analysis performed on the erythroid and megakaryoblastic genes further classified them into 3 major signatures: erythroid-megakaryoblastic EMk (19%), erythroid E (42%) and megakaryoblastic Mk (39%) (Figure 1B). E and Mk represent the genes over-expressed in the erythro- and megakaryoblastic leukaemias, respectively, and EMk represents the genes specifically over-expressed in both types. MkE1 indicates genes over-expressed in the 3 megakaryoblastic leukaemias (Mk1-Mk3) and the erythroleukaemia E1. Similarly, EMk3 corresponds to genes over-expressed in 3 erythroleukaemias (E1-E3) and the megakaryoblastic leukaemia Mk3. These results indicate that the leukaemias E1 and Mk3 are biphenotypic and express both erythroid and megakaryoblastic markers, which was previously observed in some Graffi-induced erythro- and megakaryoblastic leukaemias [12]. The MkMB signature includes genes over-expressed in megakaryoblastic leukaemias (Mk1-Mk3), myeloid leukaemia (M) and B-cell leukaemias (B1-B3). A detailed analysis of Mk, E and EMk signatures has revealed that many of the genes have not yet been reported in relation to the erythroid or the megakaryocytic lineages or to the corresponding leukaemias. The complete lists of genes detailing these non-lymphoid signatures are publicly available at http://www.biomed.uqam.ca/rassart/microarray.html webcite[19].

The lineage specific expression of genes involved in heme biosynthesis, the megakaryocytic fibrinogen receptors and the expression of well known transcription factors validate the true lineage of these erythro- and megakaryoblastic leukaemias (Table 1).

Table 1. Genes specific to the erythroid and megakaryoblastic leukaemias

The megakaryoblastic signature

The megakaryoblastic specific genes assigned with a functional annotation (GO terms) were divided into different functional classes. Table 2 lists some genes potentially implicated in the disease but the complete data are readily available [19]. Within this list of genes, the oncogenes Meis1 (Myeloid ecotropic viral integration site 1), Pbx1 (Pre B-cell leukaemia transcription factor 1), Evi1 (ecotropic viral integration site 1), Evi3 (Zfp521, zinc finger protein 521) and the co-repressor Cbfa2t3h (Core-binding factor, runt domain, alpha subunit 2, translocated to, 3 (human), Eto2) have already been related to megakaryoblastic leukaemias or megakaryocytic lineage [20-25].

Table 2. Genes over-expressed in the megakaryoblastic leukaemias

However, several other genes, for example Nmyc (Neuroblastoma myc-related oncogene 1), Fgf3 (Fibroblast growth factor 3) and Fap (Fibroblast activation protein), that are also known oncogenes [26-29], have never been reported with megakaryoblastic leukaemias. Myct1 (Myc target 1) is also potentially implicated in the disease as it is positively regulated by Myc and contains tumorigenic properties by itself [30].

Among the list of genes identified as the MkMB signature, we found Jun (Jun oncogene), Fosl2 (Fos-like antigen 2) and Fes (Feline sarcoma oncogene) (Table 2, [31,32]). The co-repressor Ctbp2 (C-terminal binding protein 2), Sfpi1 (SFFV proviral integration 1, PU.1) and Kit (Kit oncogene) are also MkMB signature elements. Ctbp2 is known to interact with Evi1 and Fog [33]. Sfpi1 was shown to regulate the expression of the integrin αIIb (Itga2B, CD41) in a TPO-induced Mpl-UT7 model [34,35] and it was reported as an insertional target of the Graffi MuLV [36].

Due to technical limitations, no megakaryoblastic control was present in our study. Normal megakaryocytes and therefore megakaryoblasts represent a minor population in normal mice and it was not possible to obtain enough purified cells with the technique utilized for other samples. We therefore compared our dataset to a study of murine megakaryocytic maturation indicating up- or down-regulation during differentiation (dataset GSE6593, [37]). According to the megakaryocytic GSE6593 dataset, Fgf3 Jun, Kit, Sfpi, Fes and Bmyc are down-regulated upon megakaryocytic differentiation (Table 2).

When genes within the MkMB signature were compared to GO annotations, one gene class was over-represented. Many of these genes were membrane receptors and extra-cellular factors known to be expressed by antigen presenting cells (APC) as well as implicated in inflammatory response [19]. For example, Tlr2 (Toll-like receptor 2), Tlr4 (toll-like receptor 4), Syk (Spleen tyrosine kinase) and Ebi3 (Epstein-Barr virus induced gene 3) are part of the Toll-like receptor signaling pathway to respond to microbial stimuli (LPS) and induce inflammation (Table 2). Confirming our data, Tlr2, Tlr4 and Syk are already known to be expressed by the megakaryocytic lineage [38,39].

To validate the microarray data, the expression of several megakaryoblastic specific genes was tested by semi-quantitative RT-PCR in samples different from those analyzed in the microarray experiments (Table 2 and Figure 2). Within these genes, Kit and Gata2 were tested due to their important roles in haematopoiesis. The other genes were selected for experimentation since they had no prior association with megakaryocytic lineages or with the corresponding leukaemia and also since their function remained poorly studied (Table 2, 'Genes selected for RT-PCR validation'). Within these genes, Gulp1 (Engulfment adaptor PTB domain containing 1) and Gucy1a3 (Guanylate cyclase 1, soluble, alpha 3) gave the highest specificity in the RT-PCR experiments with a strong expression in the 3 megakaryoblastic leukaemias (Figure 2). Most of the non-megakaryoblastic leukaemias showed very low or no expression of these genes. Ncf2 (Neutrophil cytosilic factor 2) is highly expressed in the B-cell and megakaryoblastic leukaemias in accordance with the microarray data (Table 2). Fkbp9 (FK506 binding protein 9) is strongly expressed in the megakaryoblastic leukaemias with a weaker but sustained expression in other types of leukaemias. Gata2 is strongly expressed in the megakaryoblastic leukaemias and to a lower level in the 3 erythroleukaemias. Finally, Kit was amplified in all leukaemias but with the strongest expression in the megakaryoblastic ones (Figure 2).

thumbnailFigure 2. Megakaryoblastic specific genes. Semi-quantitative RT-PCR analysis in 3 T-cell (T, 1-3), 3 B-cell (B, 4-7), 1 myeloid (M, 8), 3 erythroid (E, 8-10) and 3 megakaryoblastic leukaemias (Mk, 11-13). RT-PCRs have been performed in quadruplicate for each sample. Two out of 4 RT-PCR runs are shown for the leukaemic samples (lane a: 25 PCR cycles; lane b: 28 PCR cycles) and 1 out of 4 are shown for the β-actin internal control (lane c). The histograms represent the quantification of the density of the bands relative to the β-actin sample including all RT-PCR runs for each type of leukaemia (T, B, M, E, Mk).

The erythroid signature

The results of the erythroid signature were compared to the transcriptome analysis of G1E cells during GATA1 induced differentiation (dataset GSE628, [40]). Our results correlate neatly with this dataset although the Affymetrix Genechip MG-U74A used by Welch and colleagues contains only one third the probes of the MOE 430.2 used in our study. In the Welch et al study, the genes that displayed increased expression during differentiation tended to be under-expressed in the Graffi-induced erythroleukaemias compared to the control sample and vice-versa. This suggests that the Graffi-induced erythroleukaemias are blocked in an earlier stage than the control sample taken from a population of Ter119+CD71+ erythroblasts in the bone marrow. Table 3 provides examples of this correlation between the two studies. For example, Car1 (Carbonic anhydrase 1) is over-expressed in the leukaemias in comparison to the control (positive value, column 'E-CE') and its expression decreases during erythroid differentiation (descending arrow, column 'G1E'). Alas2 (Aminolevulinic acid synthase 1), involved in the heme biosynthesis, is under-expressed in the leukaemias in comparison to the control and its expression increases during erythroid differentiation.

Table 3. Genes specific to the erythroid leukaemias

Table 3 presents a summary of the erythroid specific genes over-expressed in comparison to the control sample and potentially implicated in the disease but the complete data are readily available [19]. Within these genes, Gdf3 (Growth differentiation factor 3), Podxl (Podocalyxin-like), Nupr1 (Nuclear protein 1), Pim1 (Proviral integration site 1) and Isr2 (Insulin receptor substrate 2) are known to be regulated by erythropoietin [41-43]. The oncogene Pim1 was found rearranged in Friend helper MuLV-induced erythroleukaemias and Graffi-induced leukaemias [36,44]. Ssx2ip (Synovial sarcoma, X breakpoint 2 interacting protein) was found over-expressed in some AML patients and is expressed by K562 erythroid cells [45]. Rab11a (RAB11a, member RAS oncogene family) was reported to regulate the recycling of the transferrin receptor [46]. This protein interacts with Evi5 [47] and has a potential role in cancer [48]. The oncogene Ncoa3 (Nuclear receptor coactivator 3) is over-expressed in numerous cancer types such as breast, prostate, ovarian, gastric, pancreatic and colorectal cancers [49].

Many genes have not yet been reported in relation to erythroid leukaemias and several others have a still unknown function and some of them have been selected for RT-PCR validation (Table 3 and Figure 3). Gata1 was tested due to its important role in haematopoiesis (Figure 3). Among the 7 tested erythroid genes (Table 3, 'Genes selected for RT-PCR validation'), both Cda (Cytidine deaminase) and Ltbp2 (Latent transforming growth factor beta binding protein 2) showed a very high and specific expression in the 3 erythroleukaemias (Figure 3). Slamf1 (Signaling lymphocytic activation molecule family member 1), Snca (Synuclein alpha) and Btbd14a (BTB/POZ domain containing 14A) are higher expressed in the erythroleukaemias and lower in the megakaryoblastic leukaemias but were also amplified in other types of leukaemias. Rabgef1 is specifically expressed in the erythroleukaemias at 25 PCR cycles (lane a) but was amplified in all samples at 28 PCR cycles (lane b). Gata1 is equally highly expressed in the erythroid and in the megakaryoblastic leukaemias (Figure 3).

thumbnailFigure 3. Erythroid specific genes. Semi-quantitative RT-PCR analysis in 3 T-cell (T, 1-3), 3 B-cell (B, 4-7), 1 myeloid (M, 8), 3 erythroid (E, 8-10) and 3 megakaryoblastic leukaemias (Mk, 11-13) was utilized for each of the 6 tested genes. RT-PCRs have been performed in quadruplicate for each sample. Two out of 4 RT-PCR runs are shown for the leukaemic samples (lane a: 25 PCR cycles; lane b: 28 PCR cycles) and 1 out of 4 are shown for the β-actin control sample (lane c). The histograms represent the quantification of the density of the bands relative to the β-actin sample including all RT-PCR runs for each type of leukaemia (T, B, M, E, Mk).

RT-PCR validation in a Friend virus murine erythroleukaemia cell line

The expression of the erythroid and megakaryoblastic specific transcripts validated by RT-PCR (Figures 2 and 3) was further assessed on a different erythroid model (Figure 4). The erythroleukaemia cell line HB22.2 has been derived from a leukaemia induced by the Friend Murine Leukaemia virus (F-MuLV) and it presents a very immature erythroid phenotype (Kit+CD71+Ter119-CD41-) ([17]). The 6 erythroid genes (Slamf1, Snca, Ltbp2, Rabgef1, Cda and Btbd14a) are expressed in HB22.2 but the intensity of the Ltbp2 and Btdb14a bands were weaker (Figure 4A). In accordance with our expectations, the megakaryoblastic genes, Ncf2, Gucy1a3 and Gulp1, could not be amplified. Fkbp9 is the only megakaryoblastic gene that gave a weak signal in this erythroid cell line. Indeed, Fkbp9 showed the strongest erythroid amplification in the RT-PCR validation experiment (Figure 2). Thus, these results show that, despite the close relationship between erythroid and megakaryoblastic leukaemias, this experiment's design enabled us to find genes that can distinguish these 2 types of murine leukaemias from each other.

thumbnailFigure 4. Expression of the erythroid and megakaryoblastic specific genes in the murine erythroleukaemia cell line HB22.2 during hemin-induced differentiation. (A) Semi-quantitative RT-PCR analysis of 6 erythroid specific genes (E) and 4 megakaryoblastic specific genes (Mk). The first lane (M) indicates the low-range DNA marker. (B) Hemin-induced differentiation assay of HB22.2 cells. Semi-quantitative RT-PCR analysis of Hba as differentiation control, Btbd14a and Ltbp2 was performed in non-induced (NI) and induced (Hemin) cells at day 1 (D1) and day 3 (D3) following hemin addition. GAPDH expression was used as internal control. Three independent RT-PCR reactions were performed on each sample using increasing amount of the RT reaction (0.75 volume, 1 volume, 1.25 volume). The band intensity was measured and indicated below each band. On the right half of the figure, the histograms show the average results of the band intensities relative to the Gapdh control sample.

We then induced HB22.2 differentiation with hemin and tested the expression of the erythroid specific genes at different time-points (Figure 4). Integration of F-MuLV upstream of Fli-1 is shown to block differentiation of this cell line [17]. However, these cells are able to undergo differentiation with hemin associated induction of alpha globin (Figure 4B). Among the 6 erythroid genes tested, both Btbd14a and Ltbp2 showed reproducible changes with a decrease and an increase with differentiation, respectively (Figure 4B). The increased expression of Ltbp2 indicates that it likely plays a role in mature erythroid cells whereas Btbd14a is likely to play a role in immature erythroid cells and this correlates well with a putative oncogenic role for this gene.

Validation in human leukaemic cell lines

The proteins encoded by these erythroid and megakaryoblastic specific genes have high homologies with their human counterparts. This makes it likely that these human and murine proteins have the same functional role. Gene expression was first tested in 2 human erythroid-like cell lines, HEL and K562, a human B-cell leukaemia cell line, TK6, and a human T-cell leukaemia cell line, Jurkat (Figure 5A). Because HEL and K562 are known to harbour mixed myeloid lineage phenotype, the genes were further tested in 2 human megakaryoblastic cell lines, MEG-01 and CMK, and 1 erythroid cell line LAMA-84 (Figure 5B).

thumbnailFigure 5. Validation in human cell lines. (A) RT-PCR analysis in the erythroid-like human cell lines K562 and HEL, the B- cell lymphoblastic cell line TK6 and the T-cell leukaemia cell line Jurkat. (B) RT-PCR analysis in the human megakaryoblastic cell lines MEG-01 and CMK and the human erythroid cell line LAMA-84. The left and right parts of each panel correspond to genes identified in the erythro- and megakaryoblastic leukaemias, respectively. GAPDH was used as internal control.

The results show that all tested erythroid and megakaryoblastic specific genes were amplified in some or all of the cell lines confirming their expression in human erythroid or megakaryoblastic cell lines. Moreover, several were also amplified in the lymphoid lines TK6 and/or in Jurkat. RABGEF1 and BTBD14A were the most ubiquitous genes with a strong amplification in lymphoid and non-lymphoid cell lines. GULP1 and SNCA were the most specific to the non-lymphoid lineage with no expression in TK6 or Jurkat cell lines.

Viral integrations in the megakaryoblastic leukaemias

We also identified retroviral integration sites (RIS) in the 3 megakaryoblastic leukaemias (Mk1-3) in order to search for genes that may have contributed to the oncogenic transformation. Several RIS were amplified, cloned and sequenced in these 3 tumors (11 in Mk1, 5 in Mk2 and 10 in Mk3) (Table 4). No common integration sites (CIS) that could clearly indicate the contribution to the oncogenic events were found. Therefore, the results were compared to the retroviral tagged cancer gene database (RTCGD, [50]) that compiles the RIS identified in different murine cancer models (underlined in Table 4). Genes identified in multiple screens have a high probability of involvement in oncogenic transformation. Eleven genes near the RIS were found in the RTCGD (underlined in Table 4). Some of these genes, such as Ccnd1 and Myc, are largely known to be involved in leukaemia. Foxf1 is a transcription factor known to regulate the megakaryocytic integrin β3 (CD61) [51]. Interestingly, Kit and Gata2 are also part of RIS. The presence of the RIS near Kit in Mk1 and near Gata2 in Mk3 has been validated by PCR in the sorted leukaemic megakaryoblastic population (not shown). Two other genes, Irf8 and Itga1 targeted by a RIS (Table 4) are also of interest: Irf8 is not included in the RTCGD database but is a known CIS [52] and the ITGA1 locus is repressed by methylation during megakaryopoiesis in humans [53].

Table 4. Graffi-virus integration sites in the megakaryoblastic leukaemias

Discussion

Characterization of genes specific to erythro- and megakaryoblastic leukaemias

Patient survey studies revealed that erythroleukaemias represent an average of 5% of all cases of acute myeloid leukaemias [3-5] and megakaryoblastic leukaemias have an incidence of approximately 1% in adults and 5-10% in children [6,7]. However, the overall survival rate is extremely poor and ranges from 6% to 17% [3,7]. Children suffering from Down Syndrome are an exception as they have a higher risk of developing megakaryoblastic leukaemias but respond better to therapy [54]. Acute erythro- and megakaryoblastic leukaemias are less studied than the more frequent types of leukaemias. Thus, genes involved in the development of these leukaemias remain insufficiently elucidated.

Our experimental design is based on the comparison of non-lymphoid versus lymphoid murine leukaemias and provides the whole picture of genes specific to these 2 groups and subgroups. The sole comparison of the non-lymphoid leukaemias and their respective controls without including the lymphoid group would not have provided such a dataset. Therefore, numerous genes not described previously or uncharacterized emerged from this study. We estimated that, within the identified gene signatures, there are oncogenes directly implicated in the disease and also genes related to the normal commitment of the cells toward the erythro- or megakaryoblastic lineages. To determine which genes are potential oncogenes, we first compared the erythroleukaemias to the erythroid control samples and to the study of Welch and colleagues [40]. The comparison with the Welch's study enabled to assess the differentiation state of the leukaemias and the control. Consequently, we could assume that the genes under-expressed in comparison to the control are late stage genes and that oncogenes are more likely to be within the over-expressed genes. We validated this hypothesis with Ltbp2 and Btbd14a in a differentiation assay in the HB22.2 erythroid cell line. Second, to gain more insights into the function of the megakaryoblastic genes and in the absence of a control sample, we compared our list of genes to the study of Shivdasani and colleagues [37]. The comparison with the studies of Welch for the erythroid leukaemias [40] and of Shivadasani [37] for the megakaryoblastic leukaemias provides valuable information about the behaviour of the genes during normal differentiation. However, their respective microarray chips contained less probesets than ours and we could not perform the comparison on the whole dataset. Therefore, further experiments are required to identify the unknown role played by these genes in erythro- and megakaryoblastic leukaemias.

Erythroid and megakaryoblastic genes

Erythroid and megakaryoblastic lineages, emerging from the same bipotent progenitor, are very closely related [11] and, as confirmed by our study, several transcription factors are commonly expressed. This strengthens the hypothesis that a very fine tuning of these factors influences the commitment toward the erythroid or megakaryocytic lineages.

Our microarray data indicate that Gata1, Gata2, Fog1, Scl and Lmo2 are expressed both in the Graffi-induced erythro- and megakaryoblastic leukaemias (Table 1). They are known to act on the promoter of their target as multimeric complexes. Our study highlights that PU.1 (Sfpi1), Ctbp2, Cbfa2t3h (Eto2), Evi1 and Runx1 have a strong megakaryoblastic pattern. PU.1 is a known determinant of erythroid versus megakaryoblastic differentiation and the Gata2 protein acts on PU.1 [55]. The Cbfa2t3h protein binds to the multimeric complex formed by Gata1, Fog1, Scl and Lmo2 and is known to repress the transcription of the target genes. The co-repressor Ctbp2 is known to bind to Evi1 and Fog1 [56]. Runx1 cooperates with Gata1 during megakaryocytic commitment [22,23] and the Runx1-Evi1 fusion protein leads preferentially to the development of megakaryoblastic leukaemias in transgenic mice [57]. In a model of in vitro differentiation, Evi1 is strongly induced and sustained upon thrombopoietin treatment of CD34+ cells in a pattern very similar to Gata2 and PU.1 but only weakly upon erythropoietin treatment [58]. Great evidence indicates that Evi1 is a direct activating target of Gata2 [59]. Thus, our study reinforces the importance of these genes in the megakaryoblastic leukaemias.

RT-PCR validated megakaryoblastic genes

The specific megakaryoblastic expression of several genes with poorly elucidated physiological roles was validated by RT-PCR. Our study reports for the first time Gucy1a3, Gulp1 and Fkbp9 as being specific to megakaryoblastic leukaemias. The function of these genes, related to the normal development or transformation of megakaryocytic cells, has yet to be elucidated. Insight into their physiological roles can be provided by their already known functions in other cell types. Gucy1a3 is known to heterodimerize with Gucy1b3, which gene is also specific to the Graffi-induced megakaryoblastic leukaemias. The Gucy1a3/b3 complex produces cGMP after activation by nitric oxide (NO) itself produced by the NADPH oxidase from reactive oxygen species. As expected, Ncf2 and other components of the NADPH oxidase (Ncf1, Ncf4, Cybb) are specifically over-expressed in the Graffi-induced megakaryoblastic leukaemias (Figure 2 and not shown). In human cancerous glioma cell lines, it is hypothesized that GUCY1a3/b3 may be responsible for VEGF over-expression resulting in an increased amount of NO [60]. NO is also known to play a role in platelet activation [61]. Gulp1 could be involved in the intracellular vesicular trafficking [62] which is of high importance in megakaryocytes for transporting the molecules in the storage organelles and during proplatelet formation. Fkbp9 is poorly studied and this present study reports its expression for the first time in cells of haematopoietic origin. It is strongly expressed in our megakaryoblastic leukaemias, in human non-lymphoid leukaemias HEL, K562, CMK, Meg-01 and LAMA84, and to a lesser extent, in the murine erythroid leukaemias and cell line.

RT-PCR validated erythroid genes

The selected erythroid genes with poorly elucidated physiological roles were Slamf1, Snca, Ltbp2, Rabgef1, Cda and Btbd1a. Slamf1 is known to be expressed by activated lymphocytes but not yet identified in relation to erythroid leukaemias. The expression of Slamf1 in Friend virus-induced erythroleukaemic cell line HB22.2 confirms the Graffi model. The gene was recently revealed as a marker of haematopoietic stem cells distinguishing these cells from more differentiated progenitors [63]. Snca has already been reported in erythrocytes [64,65]. Its over-expression in the control sample and its increased expression during erythroid differentiation (G1E dataset, Table 3) indicate that it may be implicated in normal erythroid cells function. We did not observe a significant increase during HB22.2 induced differentiation. Ltbp2 is strongly expressed in our erythroleukaemias and increases significantly during differentiation. It shows a non-lymphoid expression pattern in the tested human cell lines and is identified in relation to haematopoietic cells for the first time. Some studies suggest a role for Ltbp2 in cell adhesion and in cell migration [66]. Rabgef1 has never been reported in relation to erythroid lineage or leukaemia and the encoded protein is known to interact with Rab5, Rab21 or Rab22 [67]. Rab22a is indeed specifically over-expressed in the 3 tested erythroleukaemias (not shown). RAB proteins are implicated in the intra-cellular vesicular traffic regulation and Rabgef1 is expressed in mast cells where it acts on Kit internalization [68]. Even though Rabgef1 expression pattern is more erythroid, it was amplified in all other Graffi-induced leukaemias and in all the tested human cell lines, indicating its ubiquitous expression in haematopoietic cells. Cda, responsible for resistance to chemotherapy treatment, is highly expressed in our erythroleukaemias, in HB22.2, K562 and HEL. The GEO database shows that CDA is up-regulated during the differentiation of human CD34+ cells toward the erythroid lineage (NCBI GEO, GSE4655). As of this day, Btbd14a has never been studied. It is highly expressed in the Graffi-induced erythroleukaemias and in HB22.2. Finally, Btbd14a appears ubiquitously expressed in the haematopoietic system. The BTB/POZ domain of Btb14a is present in many oncogenes involved in the development of leukaemia and is often found at the N-terminus of transcription factors. Thus, this gene is particularly interesting to further study as it shows a decreased expression during HB22.2 induced differentiation and could represent a potential oncogene.

The MkMB Signature

Numerous genes were commonly over-expressed in the megakaryoblastic, the myeloid leukaemias and/or the B-cell leukaemias. These genes are, in most cases, already known to be expressed by cells implicated in innate immunity. Platelets function is too often considered limited to blood coagulation and formation of thrombosis but some studies now emphasize that the platelets' role is underestimated in innate immunity and inflammation response [39,69-72]. Upon activation, platelets can release microbicidal proteins, interleukins that trigger a general inflammation response and chemokines that recruit immune cells such as leukocytes. Pathogens such as bacteria and lentiviruses can be ingested by platelets [73]. The MkMB signature seems to reflect partly the complexity of the platelet function. Only very few examples are shown in Table 2 but the complete list is available in the supplementary data [19]. The unique design of this gene expression study that compared different types of leukaemias highlights this signature.

Retroviral integration and genes potentially implicated in the onset of the disease

The analysis of retroviral integration enables the identification of genes that may be responsible for malignant transformation. In this study, we screened the 3 megakaryoblastic leukaemias for viral integration sites as oncogenic transformation events leading to megakaryoblastic leukaemias remain unknown. No CIS were identified but some genes in the RTCGD have drawn our attention due to their known functions. Within these genes, Kit and Gata2 are of particular interest as accumulating evidences point at their role in megakaryopoiesis and megakaryoblastic leukaemias.

Gata2 was recently reported for the first time as a common integration site in leukaemias induced by the MOL4070LTR retrovirus in the NHD13 mouse [74]. In vitro studies showed that Gata2 over-expression redirects the haematopoietic differentiation from the macrophage lineage toward the erythroid or the megakaryocytic lineages or from the erythroid toward the megakaryocytic lineage [55,75]. The importance of Gata2 on megakaryopoiesis was also demonstrated in a differentiation study in which this gene is strongly induced and sustained upon thrombopoietin treatment of CD34+ cells but only weakly induced upon erythropoietin treatment [58]. The authors made the interesting hypothesis that GATA2 might repress the expression of the erythroid markers in maturing megakaryocytic cells since its activation inhibits erythroid differentiation in some systems.

Kit is involved in many cancers and is regulated by the SCL complex (Gata1/2, SCL, Lmo2) in haematopoietic cells [76]. More evidence begins to emerge for its role in very early stages of megakaryopoiesis [77,78] and in megakaryoblastic leukaemias [79,80]. Bourquin et al. reported increased levels of KIT, GATA2 and MYC in DS AMKL cells harbouring a GATA1 mutation compared to non-DS AMKL cells. They hypothesized that the mutated GATA1 in DS AMKL cells failed to repress the expression of these 3 genes [79].

The 3 megakaryoblastic leukaemias express very high levels of Kit and Gata2. We hypothesize that the viral integrations may block the repression of these genes by continuously activating the transcription or inhibiting the binding of repressor molecules. Kit and Gata2 levels normally decrease during megakaryocytic differentiation (Table 2 'GSE6593'). These 2 genes, affected by the retroviral integration, would send continuous signals of proliferation and survival to the cell. The analysis of a larger sample of Graffi-induced megakaryoblastic leukaemias would be required to prove more efficiently the involvement of Gata2 and Kit in this particular type of leukaemia.

Conclusions

In this report, we compared the gene profiles of the erythro- and megakaryoblastic leukaemias induced by the Graffi virus. Our study identifies genes that are highly expressed in the Graffi-induced erythro- and megakaryoblastic leukaemias. The complete dataset of this study is readily available (GSE12581 and [19]). Within the data, numerous genes have not yet been assigned with a known function and some of them could be used as markers for specific types of leukaemias and even the target of new therapies. We selected and RT-PCR validated genes those functions in normal cells are poorly characterized. For the majority, their expression in these lineages is shown for the first time and further functional characterization will surely complement the knowledge of erythroid and megakaryocytic lineages.

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

VV designed and performed the microarray, RT-PCR, differentiation assay and cloning of the viral insertion sites experiments, analyzed and interpreted the data, and wrote the manuscript. PL performed the RT-PCR experiments on the human haematopoietic cell lines. DPSO optimized the protocol for the cloning of viral integration sites. YBD provided the HB22.2, K562 and HEL cell lines and critically revised the manuscript. ER guided the project and wrote the manuscript. All authors read and approved the final manuscript.

Acknowledgements

We thank André Ponton, Michal Blazejczyk and Mathieu Miron from McGill University and Genome Quebec Innovation Centre (Montreal, Canada) for their help with the design and analyses of the microarray experiments. This work was supported by Canadian Institutes of Health Research grants MOP-37994 (ER) and MOP-84460 (YBD).

References

  1. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, Sultan C: Criteria for the diagnosis of acute leukemia of megakaryocyte lineage (M7). A report of the French-American-British Cooperative Group.

    Ann Intern Med 1985, 103(3):460-462. PubMed Abstract OpenURL

  2. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, Sultan C: Proposals for the classification of the acute leukaemias. French-American-British (FAB) co-operative group.

    British journal of haematology 1976, 33(4):451-458. PubMed Abstract | Publisher Full Text OpenURL

  3. Colita A, Belhabri A, Chelghoum Y, Charrin C, Fiere D, Thomas X: Prognostic factors and treatment effects on survival in acute myeloid leukemia of M6 subtype: A retrospective study of 54 cases.

    Annals of Oncology 2001, 12:451-455. PubMed Abstract | Publisher Full Text OpenURL

  4. Wells AW, Bown N, Reid MM, Hamilton PJ, Jackson GH, Taylor PR: Erythroleukaemia in the north of England: a population based study.

    Journal of Clinical Pathology 2001, 54:608-612. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  5. Domingo-Claros A, Larriba I, Rozman M, Irriguible D, Vallespi T, Aventin A, Ayats R, Milla F, Sole F, Florensa L, et al.: Acute erythroid neoplastic proliferations. A biological study based on 62 patients.

    Haematologica 2002, 87(2):148-153. PubMed Abstract | Publisher Full Text OpenURL

  6. Rogelio Paredes-Aguilera LR-GNL-SRAT: Biology, clinical, and hematologic features of acute megakaryoblastic leukemia in children.

    American Journal of Hematology 2003, 73:71-80. PubMed Abstract | Publisher Full Text OpenURL

  7. Garderet L, Labopin M, Gorin NC, Polge E, Baruchel A, Meloni G, Ortega J, Vossen J, Bunjes D, Leverger G, et al.: Hematopoietic stem cell transplantation for de novo acute megakaryocytic leukemia in first complete remission: a retrospective study of the European Group for Blood and Marrow Transplantation (EBMT).

    Blood 2005, 105(1):405-409. PubMed Abstract | Publisher Full Text OpenURL

  8. Park S, Picard F, Dreyfus F: Erythroleukemia: a need for a new definition.

    Leukemia 2002, 16(8):1399-1401. PubMed Abstract | Publisher Full Text OpenURL

  9. Linari S, Vannucchi AM, Ciolli S, Leoni F, Caporale R, Grossi A, Pagliai G, Santini V, Paoletti F, Ferrini PR: Coexpression of erythroid and megakaryocytic genes in acute erythroblastic (FAB M6) and megakaryoblastic (FAB M7) leukaemias.

    British journal of haematology 1998, 102(5):1335-1337. PubMed Abstract | Publisher Full Text OpenURL

  10. Debili N, Coulombel L, Croisille L, Katz A, Guichard J, Breton-Gorius J, Vainchenker W: Characterization of a bipotent erythro-megakaryocytic progenitor in human bone marrow.

    Blood 1996, 88(4):1284-1296. PubMed Abstract | Publisher Full Text OpenURL

  11. Nakorn TN, Miyamoto T, Weissman IL: Characterization of mouse clonogenic megakaryocyte progenitors.

    Proceedings of the National Academy of Sciences of the United States of America 2003, 100(1):205-210. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  12. Voisin V, Barat C, Hoang T, Rassart E: Novel insights into the pathogenesis of the Graffi murine leukemia retrovirus.

    J Virol 2006, 80(8):4026-4037. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  13. Blazejczyk M, Miron M, Nadon R: FlexArray: A statistical data analysis software for gene expression microarrays. Genome Quebec, Montreal, Canada, URL http://genomequebec.mcgill.ca/FlexArray webcite.

    2007. PubMed Abstract | Publisher Full Text OpenURL

  14. Eisen MB, Spellman PT, Brown PO, Botstein D: Cluster analysis and display of genome-wide expression patterns.

    Proceedings of the National Academy of Sciences of the United States of America 1998, 95(25):14863-14868. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  15. Tamayo P, Slonim D, Mesirov J, Zhu Q, Kitareewan S, Dmitrovsky E, Lander ES, Golub TR: Interpreting patterns of gene expression with self-organizing maps: methods and application to hematopoietic differentiation.

    Proceedings of the National Academy of Sciences of the United States of America 1999, 96(6):2907-2912. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  16. Tusher VG, Tibshirani R, Chu G: Significance analysis of microarrays applied to the ionizing radiation response.

    Proceedings of the National Academy of Sciences of the United States of America 2001, 98(9):5116-5121. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  17. Wong KS, Li YJ, Howard J, Ben-David Y: Loss of p53 in F-MuLV induced-erythroleukemias accelerates the acquisition of mutational events that confers immortality and growth factor independence.

    Oncogene 1999, 18(40):5525-5534. PubMed Abstract | Publisher Full Text OpenURL

  18. Mikkers H, Allen J, Knipscheer P, Romeijn L, Hart A, Vink E, Berns A: High-throughput retroviral tagging to identify components of specific signaling pathways in cancer.

    Nature genetics 2002, 32(1):153-159. PubMed Abstract | Publisher Full Text OpenURL

  19. Voisin V, Rassart E: Erythroid and megakaryoblastic leukemias induced by the murine retrovirus Graffi Gene expression profiling MOE 430.2 GEO:GSE12581 Additional analyses and Downloadable excel files URL: http://www.biomed.uqam.ca/rassart/microarray.html webcite.

    2009.

  20. Hao X, Shin MS, Zhou JX, Lee CH, Qi CF, Naghashfar Z, Hartley JW, Fredrickson TN, Ward JM, Morse HC: Histologic and molecular characterizations of megakaryocytic leukemia in mice.

    Leuk Res 2006, 30(4):397-406.

    Epub 2005 Oct 2010.

    PubMed Abstract | Publisher Full Text OpenURL

  21. Hamlett I, Draper J, Strouboulis J, Iborra F, Porcher C, Vyas P: Characterization of megakaryocyte GATA1-interacting proteins: the corepressor ETO2 and GATA1 interact to regulate terminal megakaryocyte maturation.

    Blood 2008, 112(7):2738-2749. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  22. Nucifora G, Laricchia-Robbio L, Senyuk V: EVI1 and hematopoietic disorders: history and perspectives.

    Gene 2006, 368:1-11. PubMed Abstract | Publisher Full Text OpenURL

  23. Kilbey A, Alzuherri H, McColl J, Cales C, Frampton J, Bartholomew C: The Evi1 proto-oncoprotein blocks endomitosis in megakaryocytes by inhibiting sustained cyclin-dependent kinase 2 catalytic activity.

    British journal of haematology 2005, 130(6):902-911. PubMed Abstract | Publisher Full Text OpenURL

  24. Bond HM, Mesuraca M, Carbone E, Bonelli P, Agosti V, Amodio N, De Rosa G, Di Nicola M, Gianni AM, Moore MA, et al.: Early hematopoietic zinc finger protein (EHZF), the human homolog to mouse Evi3, is highly expressed in primitive human hematopoietic cells.

    Blood 2004, 103(6):2062-2070. PubMed Abstract | Publisher Full Text OpenURL

  25. Hentges KE, Weiser KC, Schountz T, Woodward LS, Morse HC, Justice MJ: Evi3, a zinc-finger protein related to EBFAZ, regulates EBF activity in B-cell leukemia.

    Oncogene 2005, 24(7):1220-1230. PubMed Abstract | Publisher Full Text OpenURL

  26. Hu L, Sham JS, Xie D, Wen JM, Wang WS, Wang Y, Guan XY: Up-regulation of fibroblast growth factor 3 is associated with tumor metastasis and recurrence in human hepatocellular carcinoma.

    Cancer Lett 2007, 252(1):36-42. PubMed Abstract | Publisher Full Text OpenURL

  27. van Lohuizen M, Breuer M, Berns A: N-myc is frequently activated by proviral insertion in MuLV-induced T cell lymphomas.

    Embo J 1989, 8(1):133-136. PubMed Abstract | PubMed Central Full Text OpenURL

  28. Yancopoulos GD, Nisen PD, Tesfaye A, Kohl NE, Goldfarb MP, Alt FW: N-myc can cooperate with ras to transform normal cells in culture.

    Proc Natl Acad Sci USA 1985, 82(16):5455-5459. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  29. Cheng JD, Valianou M, Canutescu AA, Jaffe EK, Lee HO, Wang H, Lai JH, Bachovchin WW, Weiner LM: Abrogation of fibroblast activation protein enzymatic activity attenuates tumor growth.

    Mol Cancer Ther 2005, 4(3):351-360. PubMed Abstract | Publisher Full Text OpenURL

  30. Rogulski KR, Cohen DE, Corcoran DL, Benos PV, Prochownik EV: Deregulation of common genes by c-Myc and its direct target, MT-MC1.

    Proceedings of the National Academy of Sciences of the United States of America 2005, 102(52):18968-18973. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  31. Yates KE, Gasson JC: Role of c-Fes in normal and neoplastic hematopoiesis.

    Stem cells (Dayton, Ohio) 1996, 14(1):117-123. PubMed Abstract | Publisher Full Text OpenURL

  32. Emilia G, Donelli A, Ferrari S, Torelli U, Selleri L, Zucchini P, Moretti L, Venturelli D, Ceccherelli G, Torelli G: Cellular levels of mRNA from c-myc, c-myb and c-fes onc-genes in normal myeloid and erythroid precursors of human bone marrow: an in situ hybridization study.

    British journal of haematology 1986, 62(2):287-292. PubMed Abstract | Publisher Full Text OpenURL

  33. Fox AH, Liew C, Holmes M, Kowalski K, Mackay J, Crossley M: Transcriptional cofactors of the FOG family interact with GATA proteins by means of multiple zinc fingers.

    The EMBO journal 1999, 18(10):2812-2822. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  34. Hromas R, Orazi A, Neiman RS, Maki R, Van Beveran C, Moore J, Klemsz M: Hematopoietic lineage- and stage-restricted expression of the ETS oncogene family member PU.1.

    Blood 1993, 82(10):2998-3004. PubMed Abstract | Publisher Full Text OpenURL

  35. Doubeikovski A, Uzan G, Doubeikovski Z, Prandini MH, Porteu F, Gisselbrecht S, Dusanter-Fourt I: Thrombopoietin-induced expression of the glycoprotein IIb gene involves the transcription factor PU.1/Spi-1 in UT7-Mpl cells.

    The Journal of biological chemistry 1997, 272(39):24300-24307. PubMed Abstract | Publisher Full Text OpenURL

  36. Denicourt C, Edouard E, Rassart E: Oncogene activation in myeloid leukemias by Graffi murine leukemia virus proviral integration.

    J Virol 1999, 73(5):4439-4442. PubMed Abstract | PubMed Central Full Text OpenURL

  37. Chen Z, Hu M, Shivdasani RA: Expression analysis of primary mouse megakaryocyte differentiation and its application in identifying stage-specific molecular markers and a novel transcriptional target of NF-E2.

    Blood 2007, 109(4):1451-1459. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  38. Andonegui G, Kerfoot SM, McNagny K, Ebbert KV, Patel KD, Kubes P: Platelets express functional Toll-like receptor-4.

    Blood 2005, 106(7):2417-2423. PubMed Abstract | Publisher Full Text OpenURL

  39. Cognasse F, Hamzeh H, Chavarin P, Acquart S, Genin C, Garraud O: Evidence of Toll-like receptor molecules on human platelets.

    Immunology and cell biology 2005, 83(2):196-198. PubMed Abstract | Publisher Full Text OpenURL

  40. Welch JJ, Watts JA, Vakoc CR, Yao Y, Wang H, Hardison RC, Blobel GA, Chodosh LA, Weiss MJ: Global regulation of erythroid gene expression by transcription factor GATA-1.

    Blood 2004, 104(10):3136-3147. PubMed Abstract | Publisher Full Text OpenURL

  41. Fang J, Menon M, Kapelle W, Bogacheva O, Bogachev O, Houde E, Browne S, Sathyanarayana P, Wojchowski DM: EPO modulation of cell-cycle regulatory genes, and cell division, in primary bone marrow erythroblasts.

    Blood 2007, 110(7):2361-2370. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  42. Sathyanarayana P, Menon MP, Bogacheva O, Bogachev O, Niss K, Kapelle WS, Houde E, Fang J, Wojchowski DM: Erythropoietin modulation of podocalyxin and a proposed erythroblast niche.

    Blood 2007, 110(2):509-518. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  43. Sathyanarayana P, Dev A, Fang J, Houde E, Bogacheva O, Bogachev O, Menon M, Browne S, Pradeep A, Emerson C, et al.: EPO receptor circuits for primary erythroblast survival.

    Blood 2008, 111(11):5390-5399. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  44. Dreyfus F, Sola B, Fichelson S, Varlet P, Charon M, Tambourin P, Wendling F, Gisselbrecht S: Rearrangements of the Pim-1, c-myc, and p53 genes in Friend helper virus-induced mouse erythroleukemias.

    Leukemia 1990, 4(8):590-594. PubMed Abstract OpenURL

  45. Breslin A, Denniss FA, Guinn BA: SSX2IP: an emerging role in cancer.

    Biochemical and biophysical research communications 2007, 363(3):462-465. PubMed Abstract | Publisher Full Text OpenURL

  46. Ren M, Xu G, Zeng J, De Lemos-Chiarandini C, Adesnik M, Sabatini DD: Hydrolysis of GTP on rab11 is required for the direct delivery of transferrin from the pericentriolar recycling compartment to the cell surface but not from sorting endosomes.

    Proceedings of the National Academy of Sciences of the United States of America 1998, 95(11):6187-6192. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  47. Westlake CJ, Junutula JR, Simon GC, Pilli M, Prekeris R, Scheller RH, Jackson PK, Eldridge AG: Identification of Rab11 as a small GTPase binding protein for the Evi5 oncogene.

    Proceedings of the National Academy of Sciences of the United States of America 2007, 104(4):1236-1241. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  48. Palmieri D, Bouadis A, Ronchetti R, Merino MJ, Steeg PS: Rab11a differentially modulates epidermal growth factor-induced proliferation and motility in immortal breast cells.

    Breast cancer research and treatment 2006, 100(2):127-137. PubMed Abstract | Publisher Full Text OpenURL

  49. Yan J, Tsai SY, Tsai MJ: SRC-3/AIB1: transcriptional coactivator in oncogenesis.

    Acta pharmacologica Sinica 2006, 27(4):387-394. PubMed Abstract | Publisher Full Text OpenURL

  50. Akagi K, Suzuki T, Stephens RM, Jenkins NA, Copeland NG: RTCGD: retroviral tagged cancer gene database.

    Nucleic acids research 2004, (32 Database):D523-527. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  51. Malin D, Kim IM, Boetticher E, Kalin TV, Ramakrishna S, Meliton L, Ustiyan V, Zhu X, Kalinichenko VV: Forkhead box F1 is essential for migration of mesenchymal cells and directly induces integrin-beta3 expression.

    Molecular and cellular biology 2007, 27(7):2486-2498. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  52. Ma SL, Sorensen AB, Kunder S, Sorensen KD, Quintanilla-Martinez L, Morris DW, Schmidt J, Pedersen FS: The Icsbp locus is a common proviral insertion site in mature B-cell lymphomas/plasmacytomas induced by exogenous murine leukemia virus.

    Virology 2006, 352(2):306-318. PubMed Abstract | Publisher Full Text OpenURL

  53. Cheli Y, Kanaji S, Jacquelin B, Chang M, Nugent DJ, Kunicki TJ: Transcriptional and epigenetic regulation of the integrin collagen receptor locus ITGA1-PELO-ITGA2.

    Biochimica et biophysica acta 2007, 1769(9-10):546-558. PubMed Abstract | PubMed Central Full Text OpenURL

  54. Hitzler JK, Zipursky A: Origins of leukaemia in children with Down syndrome.

    Nat Rev Cancer 2005, 5(1):11-20. PubMed Abstract | Publisher Full Text OpenURL

  55. Kitajima K, Tanaka M, Zheng J, Yen H, Sato A, Sugiyama D, Umehara H, Sakai E, Nakano T: Redirecting differentiation of hematopoietic progenitors by a transcription factor, GATA-2.

    Blood 2006, 107(5):1857-1863. PubMed Abstract | Publisher Full Text OpenURL

  56. Chinnadurai G: CtBP, an unconventional transcriptional corepressor in development and oncogenesis.

    Molecular cell 2002, 9(2):213-224. PubMed Abstract | Publisher Full Text OpenURL

  57. Maki K, Yamagata T, Asai T, Yamazaki I, Oda H, Hirai H, Mitani K: Dysplastic definitive hematopoiesis in AML1/EVI1 knock-in embryos.

    Blood 2005, 106(6):2147-2155. PubMed Abstract | Publisher Full Text OpenURL

  58. Terui K, Takahashi Y, Kitazawa J, Toki T, Yokoyama M, Ito E: Expression of transcription factors during megakaryocytic differentiation of CD34+ cells from human cord blood induced by thrombopoietin.

    The Tohoku journal of experimental medicine 2000, 192(4):259-273. PubMed Abstract | Publisher Full Text OpenURL

  59. Yuasa H, Oike Y, Iwama A, Nishikata I, Sugiyama D, Perkins A, Mucenski ML, Suda T, Morishita K: Oncogenic transcription factor Evi1 regulates hematopoietic stem cell proliferation through GATA-2 expression.

    The EMBO journal 2005, 24(11):1976-1987. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  60. Saino M, Maruyama T, Sekiya T, Kayama T, Murakami Y: Inhibition of angiogenesis in human glioma cell lines by antisense RNA from the soluble guanylate cyclase genes, GUCY1A3 and GUCY1B3.

    Oncology reports 2004, 12(1):47-52. PubMed Abstract OpenURL

  61. Battinelli E, Willoughby SR, Foxall T, Valeri CR, Loscalzo J: Induction of platelet formation from megakaryocytoid cells by nitric oxide.

    Proceedings of the National Academy of Sciences of the United States of America 2001, 98(25):14458-14463. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  62. Kiss RS, Ma Z, Nakada-Tsukui K, Brugnera E, Vassiliou G, McBride HM, Ravichandran KS, Marcel YL: The lipoprotein receptor-related protein-1 (LRP) adapter protein GULP mediates trafficking of the LRP ligand prosaposin, leading to sphingolipid and free cholesterol accumulation in late endosomes and impaired efflux.

    The Journal of biological chemistry 2006, 281(17):12081-12092. PubMed Abstract | Publisher Full Text OpenURL

  63. Kiel MJ, Yilmaz OH, Iwashita T, Yilmaz OH, Terhorst C, Morrison SJ: SLAM family receptors distinguish hematopoietic stem and progenitor cells and reveal endothelial niches for stem cells.

    Cell 2005, 121(7):1109-1121. PubMed Abstract | Publisher Full Text OpenURL

  64. Nakai M, Fujita M, Waragai M, Sugama S, Wei J, Akatsu H, Ohtaka-Maruyama C, Okado H, Hashimoto M: Expression of alpha-synuclein, a presynaptic protein implicated in Parkinson's disease, in erythropoietic lineage.

    Biochemical and biophysical research communications 2007, 358(1):104-110. PubMed Abstract | Publisher Full Text OpenURL

  65. Scherzer CR, Grass JA, Liao Z, Pepivani I, Zheng B, Eklund AC, Ney PA, Ng J, McGoldrick M, Mollenhauer B, et al.: GATA transcription factors directly regulate the Parkinson's disease-linked gene alpha-synuclein.

    Proceedings of the National Academy of Sciences of the United States of America 2008, 105(31):10907-10912. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  66. Vehvilainen P, Hyytiainen M, Keski-Oja J: Latent Transforming Growth Factor-β-binding Protein 2 Is an Adhesion Protein for Melanoma Cells.

    The Journal of Biochemical Chemistry 2003, 278:24705-24713. OpenURL

  67. Mattera R, Tsai YC, Weissman AM, Bonifacino JS: The Rab5 guanine nucleotide exchange factor Rabex-5 binds ubiquitin (Ub) and functions as a Ub ligase through an atypical Ub-interacting motif and a zinc finger domain.

    The Journal of biological chemistry 2006, 281(10):6874-6883. PubMed Abstract | Publisher Full Text OpenURL

  68. Kalesnikoff J, Rios EJ, Chen C-C, Nakae S, Zabel BA, Butcher EC, Tsai M, Tam S-Y, Galli SJ: RabGEF1 regulates stem cell factor/c-Kit-mediated signaling events and biological responses in mast cells.

    Proceedings of the National Academy of Sciences of the United States of America 2006, 103:2659-2664. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  69. Weyrich AS, Zimmerman GA: Platelets: signaling cells in the immune continuum.

    Trends in immunology 2004, 25(9):489-495. PubMed Abstract | Publisher Full Text OpenURL

  70. Elzey BD, Sprague DL, Ratliff TL: The emerging role of platelets in adaptive immunity.

    Cellular immunology 2005, 238(1):1-9. PubMed Abstract | Publisher Full Text OpenURL

  71. Sprague DL, Sowa JM, Elzey BD, Ratliff TL: The role of platelet CD154 in the modulation in adaptive immunity.

    Immunologic research 2007, 39(1-3):185-193. PubMed Abstract | Publisher Full Text OpenURL

  72. von Hundelshausen P, Weber C: Platelets as immune cells: bridging inflammation and cardiovascular disease.

    Circulation research 2007, 100(1):27-40. PubMed Abstract | Publisher Full Text OpenURL

  73. Tanaka M, Aze Y, Fujita T: Adhesion molecule LFA-1/ICAM-1 influences on LPS-induced megakaryocytic emperipolesis in the rat bone marrow.

    Veterinary pathology 1997, 34(5):463-466. PubMed Abstract | Publisher Full Text OpenURL

  74. Slape C, Hartung H, Lin YW, Bies J, Wolff L, Aplan PD: Retroviral insertional mutagenesis identifies genes that collaborate with NUP98-HOXD13 during leukemic transformation.

    Cancer research 2007, 67(11):5148-5155. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  75. Ikonomi P, Rivera CE, Riordan M, Washington G, Schechter AN, Noguchi CT: Overexpression of GATA-2 inhibits erythroid and promotes megakaryocyte differentiation.

    Experimental hematology 2000, 28(12):1423-1431. PubMed Abstract | Publisher Full Text OpenURL

  76. Lecuyer E, Hoang T: SCL: from the origin of hematopoiesis to stem cells and leukemia.

    Experimental hematology 2004, 32(1):11-24. PubMed Abstract | Publisher Full Text OpenURL

  77. Antonchuk J, Hyland CD, Hilton DJ, Alexander WS: Synergistic effects on erythropoiesis, thrombopoiesis, and stem cell competitiveness in mice deficient in thrombopoietin and steel factor receptors.

    Blood 2004, 104(5):1306-1313. PubMed Abstract | Publisher Full Text OpenURL

  78. Zeuner A, Signore M, Martinetti D, Bartucci M, Peschle C, De Maria R: Chemotherapy-induced thrombocytopenia derives from the selective death of megakaryocyte progenitors and can be rescued by stem cell factor.

    Cancer research 2007, 67(10):4767-4773. PubMed Abstract | Publisher Full Text OpenURL

  79. Bourquin JP, Subramanian A, Langebrake C, Reinhardt D, Bernard O, Ballerini P, Baruchel A, Cave H, Dastugue N, Hasle H, et al.: Identification of distinct molecular phenotypes in acute megakaryoblastic leukemia by gene expression profiling.

    Proceedings of the National Academy of Sciences of the United States of America 2006, 103(9):3339-3344. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  80. Toki T, Kanezaki R, Adachi S, Fujino H, Xu G, Sato T, Suzuki K, Tauchi H, Endo M, Ito E: The key role of stem cell factor/KIT signaling in the proliferation of blast cells from Down syndrome-related leukemia.

    Leukemia 2009, 23(1):95-103. PubMed Abstract | Publisher Full Text OpenURL

Pre-publication history

The pre-publication history for this paper can be accessed here:

http://www.biomedcentral.com/1755-8794/3/2/prepub