Activation of NRas signaling is frequently found in human myeloid leukemia and can be induced by activating mutations as well as by mutations in receptors or signaling molecules upstream of NRas. To study NRas-induced leukemogenesis, we retrovirally overexpressed wild-type NRas in a murine bone marrow transplantation (BMT) model in C57BL/6J mice. Overexpression of wild-type NRas caused myelomonocytic leukemias ∼3 months after BMT in the majority of mice. A subset of mice (30%) developed malignant histiocytosis similar to mice that received mutationally activated NRas(G12D)-expressing bone marrow. Aberrant Ras signaling was demonstrated in cells expressing mutationally active or wild-type NRas, as increased activation of Erk and Akt was observed in both models. However, more NRas(G12D) were found to be in the activated, GTP-bound state in comparison with wild-type NRas. Consistent with observations reported for primary human myelomonocytic leukemia cells, Stat5 activation was also detected in murine leukemic cells. Furthermore, clonal evolution was detected in NRas wild-type-induced leukemias, including expansion of clones containing activating vector insertions in known oncogenes, such as Evi1 and Prdm16. In vitro cooperation of NRas and Evi1 improved long-term expansion of primary murine bone marrow cells. Evi1-positive cells upregulated Bcl-2 and may, therefore, provide anti-apoptotic signals that collaborate with the NRas-induced proliferative effects. As activation of Evi1 has been shown to coincide with NRAS mutations in human acute myeloid leukemia, our murine model recapitulates crucial events in human leukemogenesis.
The human RAS genes HRAS, NRAS and KRAS encode for 21-kDa proteins which belong to the Ras superfamily of GTP-binding proteins. These proteins act as GDP/GTP-regulated switches and in the GTP-bound activated state, they transmit mitogenic signals from activated cell surface receptors including receptor tyrosine kinases, cytokine receptors and heterotrimeric G-protein-coupled receptors to the nucleus.1 Ras signaling is positively regulated by guanine exchange factors (GEFs) which promote formation of the GTP-bound state, whereas negative regulation is achieved by GTPase-activating proteins that stimulate the intrinsic GTPase activity. Ras proteins interact with a set of effector proteins, such as Rafs, MEKK, PI-3K and Ral-GEF, resulting in activation of divergent downstream signaling pathways. Thus, cellular responses to Ras activity are diverse including cell proliferation, survival, differentiation or malignant transformation.2
Activating RAS mutations are commonly detected in human cancers including hematological disorders.3 The most frequent mutations of the RAS genes are point mutations in codons 12, 13 and 61, each of which impairs GTP-hydrolysis and results in an increased half-life of activated Ras. Different Ras isoforms are preferentially associated with distinct types of cancer. For instance, NRAS mutations are frequently found in human myeloid disorders, such as acute myeloid leukemia (AML) (10–20%),4 myeloproliferative neoplasms, myelodysplastic syndrome, chronic myelomonocytic leukemia (CMML, 20–50%)5 and juvenile myelomonocytic leukemia (JMML, 20%).6, 7
The significance of Ras signaling in myeloid leukemia is underscored by alternative mechanisms of activation in the absence of RAS mutations, for example, caused by mutations in other proto-oncogenes, such as receptor and non-receptor tyrosine kinases, or by fusion proteins. For example, Ras can also be activated owing to point mutations in the genes encoding the colony stimulating factor-1 receptor (CSF-1R) (10–20% in myelodysplastic syndrome and AML) and cKit receptor (10% in MPD), or by tandem duplication of the FLT3 receptor (30% in AML)4 (reviewed in Downward2 and Reuter et al.7). Furthermore, the underlying molecular defect in a subgroup of CMML is a translocation t(5;12)(q33;p13), which results in formation of the TEL-PDGFRβ fusion protein and causes constitutive activation of Ras-MAPK signaling.8
Inactivating mutations in the tumor suppressor gene NF1 (neurofibromin), a Ras GTPase-activating protein, lead to hyperactivation of Ras signaling, and loss of NF1 increases the risk of developing JMML (15% in JMML). Other groups of JMML patients display mutations of the PTPN11 or CBL genes. PTPN11 encodes for the tyrosine phosphatase SHP2, which transmits growth factor receptor signals to Ras, whereas the CBL gene encodes an E3 ubiquitin ligase whose mutation can cause constitutive activation of ERK and AKT.9, 10 Mutations of NF1, SHP2 and CBL all involve deregulation of Ras signaling and are mutually exclusive in JMML.6, 11 JMML and CMML can progress to the myelomonocytic (M4) or monocytic (M5) subtypes of AML.12
The understanding of deregulated Ras activity and its downstream pathways in cancer may identify new therapeutic targets.13 Thus, murine models recapitulating Ras-induced human myeloid leukemia are important to study cancer progression and to test therapeutic strategies in vivo. Different murine models using transgenic14 or knock-in approaches,15 as well as retroviral overexpression16, 17 and tetracycline-inducible expression systems,18 have been described. All of these models expressed mutationally active forms of NRas. The induced phenotypes ranged from T-ALL, JMML, CMML, AML, myeloproliferative neoplasm to mastocytosis and histiocytosis, probably depending on the expression strategy and expression levels, as well as the mouse strains used.
Our aim was to investigate the consequences of hyperactive NRas signaling without mutational activation in a murine leukemia model. To accomplish this, we directly compared the leukemic potential of wild-type NRas with that of the activated NRas(G12D) mutant in the same model system. In addition, analysis of vector insertion sites allowed identification of cooperating genes in NRas-induced leukemias.
Ectopic NRas(wt) expression induces hematological disease
To investigate the leukemogenic potential of ectopic murine wild-type NRas (NRas(wt)) overexpression, we transplanted retrovirally-transduced bone marrow (BM) cells into lethally irradiated C57BL/6J mice. Gammaretroviral long-terminal repeat vectors were generated to co-express NRas(wt) or the activated G12D mutant (NRas(G12D)) together with the enhanced green fluorescent protein (GFP) under the control of the spleen focus-forming virus enhancer/promoter (RSF91.NRas(wt).I.GFP.pre, RSF91.NRas(G12D).I.GFP.pre and RSF91.I.GFP.pre, Figure 1a). Functionality of the vectors was confirmed by immunoblot analysis to detect NRas expression and by flow cytometric analysis to detect GFP expression in transduced lineage-negative (lin−) BM cells (Figures 1b and c).
We performed three independent experiments (Exp. 1–3, Table 1) with 3–6 mice per group. Mice that received NRas(wt)-expressing cells developed hematological disease with weight loss, hunched posture and anemia ∼3 months after BMT. In contrast, mice that received BM cells overexpressing NRas(G12D) presented disease symptoms 3 weeks after transplantation. Vector control (Ctrl) mice remained healthy throughout the experiments (Figure 2a, Table 1). Our results demonstrate that overexpression of NRas(wt) in hematopoietic cells is sufficient to induce hematological disorders in mice, albeit with longer disease latency than NRas(G12D).
NRas(wt) expression induced myelomonocytic leukemias or malignant histiocytosis
The majority of mice (7 of 10) receiving NRas(wt)-transduced cells developed a myeloid leukemia, characterized by moderate to massive hepatosplenomegaly (Figure 2b) and leukocytosis (ranging from 4.6 to 24 × 104/μl) in five of seven mice, anemia (median hematocrit of 18.8%; median RBC count of 2.73 × 106/μl), and thrombocytopenia (median platelet count of 66 × 103/μl) (Figure 2c, Supplementary Tables 1 and 2). Spleen, liver, BM, and to a lesser extent, kidney and gut were infiltrated with myeloid leukemic cells with at least 20% of immature/blast cells in spleen and BM (Figures 3a and b). NRas(wt)/GFP-positive cells were positive for monocytic/macrophage markers (CD11b, F4/80) and to varying extent for granulocytic markers (CD11b, Gr1, Figure 3f) but negative for B-cell, T-cell and erythroid markers (data not shown). Immunohistochemistry identified CD14-positive infiltrates in spleen and liver (Supplementary Figure 1). Co-existence of monocytic and granulocytic cells was demonstrated by cytological analysis, with different degrees of maturation (Figures 3c–e). Notably, most cells were also positive for ER-MP58, a marker for myeloid precursors,19 indicating the lower degree of differentiation (Figure 3f). In addition, sporadic mice displayed atypical megakaryocytic proliferation with increased numbers of immature megakaryocytes in the spleen, or mast cell proliferation in histopathological analysis (Supplementary Figure 2). Transplantation of BM from diseased mice into secondary recipients (three donors into two recipients each) readily induced hematological disease with a shorter latency (41–89 days) and a higher degree of maturation. The observation of high blast cell numbers in primary recipients along with the involvement of both monocytic and granulocytic lineages and the transplantability of the disease defined the diagnosis as AML similar to human AML M4.
Remarkably, 3/10 mice transplanted with NRas(wt)-transduced BM cells developed a hematological disorder with atypical histiocytic infiltrates and hemophagocytosis in spleen and BM (Figures 3g and h). Histiocytic cells were characterized by pleomorphic nuclei, large cytoplasm (Figures 3i and j) and the occurrence of multinucleated giant cells. Infiltrations were either diffuse (Figure 3h) or presented as solid granulomas (Figure 3g) and involved other organs, such as liver.
Flow cytometric analysis of NRas(wt)-positive spleen cells with histiocytic infiltrations confirmed expression of the macrophage and dendritic cell markers F4/80, CD11b, CD11c and MHCII (Figure 3k), characteristic for malignant histiocytosis (mHC)/histiocytic sarcoma in mice20, 21 and human patients.22 In addition, histiocytic cells were clearly negative for T-cell and B-cell markers, thus distinguishing diagnosis from anaplastic large cell or T-cell lymphoma.22 Morphology and the partially diffuse infiltration in combination with immune-phenotyping established the diagnosis of mHC. BM from 2 of the 3 mHC mice was transplantable and induced mHC in the secondary recipients with incomplete penetrance within 23–54 days.
Comparison of mHC induced by NRas(wt) or NRas(G12D)
In contrast to NRas(wt) expression, NRas(G12D) induced mHC with complete penetrance within 3 weeks and was rapidly fatal owing to severe symptoms of pancytopenia (Figure 2c and Supplementary Tables 1 and 2), leading to massive hemorrhages, particularly in the lung. Most of the histiocytic infiltrates in the spleen and liver presented as solid granulomas (Figure 4a). Flow cytometric analysis revealed that histiocytic cells expressed the same markers as the NRas(wt)-induced mHC, except that the percentage of cells co-expressing CD11c/MHCII was considerably higher (Figure 4e). Immunohistochemical co-staining for F4/80 and CD11c visualized the infiltrates in spleen and liver (Supplementary Figure 3). The histiocytic morphology was confirmed by cytological analysis (Figures 4c and d).
The BM of NRas(G12D) mice was hypocellular and infiltrations were minimal (Figure 4b). This was probably due to the short disease latency, which did not allow for complete recovery of the BM after transplantation and the impaired engraftment of BM containing NRas(G12D)-expressing cells. We observed increased apoptosis shortly after NRas(G12D) transduction of BM cells in vitro, as well as decreased transgene-positivity in the initial days of culture (Supplementary Figures 4a, b and d), indicating that high NRas(G12D) expression was not well tolerated. Compared with vector control-transduced cells, NRas(wt) and NRas(G12D) cells were less clonogenic in colony-forming assays. Because colony assays were performed shortly after transduction, the reduced replating may reflect early effects of NRas overexpression that can result in growth arrest.23, 24 An increased proportion of macrophage colonies (CFU-M) and decreased formation of erythroid BFU-E colonies was observed for NRas(wt) and NRas(G12D) cells (Supplementary Figure 4c), indicating a shift toward differentiation into the monocytic/macrophage lineage. Consistently, NRas(G12D)-induced mHC were not transplantable into secondary recipients (5 donors, 2 recipients each). However, no increased numbers of apoptotic cells were found in NRas(G12D) mice at the day of killing as analyzed by TUNEL (TdT-mediated dUTP nick end labeling (assay)) staining (Supplementary Figure 5).
In summary, NRas(wt) expression induced mostly myelomonocytic leukemias (70%), whereas indolent mHC was observed in a subset of mice (30%). In contrast, NRas(G12D) induced fulminating mHC with complete penetrance.
NRas(wt) overexpression activates the same signaling pathways as NRas(G12D)
Activation of wild-type NRas and its downstream effectors is tightly regulated. Thus, cells overexpressing NRas(wt) could potentially retain their ability to regulate NRas activity, for example, mediated by GTPase-activating proteins stimulating the intrinsic GTPase activity. We therefore evaluated whether NRas(wt) overexpression induced hyperactivation of downstream pathways in comparison with signals induced by NRas(G12D).
Immunoblot analysis of spleen cells from diseased mice overexpressing NRas(wt) and NRas(G12D) demonstrated that Erk1/2 and Akt were activated without further cytokine stimulation (Figure 5a) similar to the signaling detected in vitro in NRas(wt) and NRas(G12D)-transduced 32D cells (Figure 5b). To detect the amount of GTP-bound Ras in relation to the total Ras, we performed pull-down assays using a GST-Ras-binding domain. The proportion of GTP-bound Ras was significantly higher in NRas(G12D)-transduced primary mononuclear BM cells (Figure 5c) and 32D cells (Figure 5d) in comparison with NRas(wt)-transduced cells.
In vivo, the percentage of NRas(G12D)-expressing cells was much lower than that of NRas(wt)-expressing cells, partially due to reduced infiltrations in the spleens. To circumvent the bias in the detection of activated signaling pathways due to different proportions of NRas-expressing cells, we analyzed the signaling in spleen cells by phosphoprotein FACS (fluorescence-activated cell sorting). Using this analysis, we compared the signaling in GFP-positive cells versus the GFP-negative cells in the same sample. The analysis confirmed activation of Erk1/2 and Akt. Furthermore, p38, Stat3 and Stat5 were activated in NRas-positive cells in vivo (Figure 5e). The transgene expression of NRas(G12D) as reflected by the median fluorescence intensity (MFI) was lower than that of NRas(wt) (Figure 5e), indicating that higher levels of NRas(wt) may be necessary to achieve a similar activity as that induced by NRas(G12D). Stat3 and Stat5 are not direct downstream targets of NRas signaling but their activation might reflect secondary upregulation of cytokine and growth factor receptor signaling. As Stat5 activation due to hypersensitivity to GM-CSF stimulation is one characteristic of myelomonocytic precursor cells in JMML, CMML and M4/M5 AML patient cells,25 we tested whether GM-CSF stimulation of NRas(G12D) and NRas(wt)-transduced BM cells would activate Stat5 in vitro. We found Stat5 activation in NRas-transduced cells to be stronger than in control vector-transduced cells (Figure 5f).
Clonal selection for Evi1 activation in NRas(wt)-induced leukemia
In our BMT model, we transplanted lin− BM cells overexpressing NRas in a large population of BM cells. In human leukemia, however, the first gene mutations occur in single cells leading to clonal outgrowth of transformed cells. Therefore, we investigated whether the induced hematological disorders were due to polyclonal hyperproliferation or clonal expansion of transplanted cells.
As each retroviral insertion site marks an individual cell, investigation of these sites allowed us to estimate the clonality of the leukemias by Southern blot analysis. The hybridization pattern of DNA from NRas(wt) spleens showed several distinct bands in the myeloid leukemias and, to a lesser extent, mHC induced by NRas(wt), suggesting oligoclonal disease.
In contrast, no distinct bands were observed in NRas(G12D)-induced mHC, indicating the polyclonal origin of the disease (Figure 6a). These results, together with the rapid disease onset, suggest that mHC in NRas(G12D) mice developed primarily due to NRas(G12D)-induced proliferation. However, the diseases observed in NRas(wt) mice appeared to be due to clonal evolution probably aided through acquisition of cooperating secondary mutations. We further confirmed the absence of activating mutations in leukemia cells from mice transplanted with NRas(wt)-transduced lin− cells (Supplementary Table 3). Spectral karyotype analysis excluded the involvement of chromosomal instability in NRas-induced leukemias (Supplementary Table 4).
Retroviral vector integration can lead to insertional upregulation of expression of nearby genes, which can contribute to leukemia development. Investigation of proviral integrations in NRas(wt) mice, which might provide second hits explaining the clonal outgrowth revealed 21 insertion sites from 10 mice that developed NRas(wt)-induced hematological disorders. Leukemic clones from three of the seven mice that developed myeloid leukemia had insertions in Evi1 (12 347, 12 592 and 13 991 bp upstream of the ATG), and an insertion in the Evi1 homolog Prdm16 was detected in another leukemic mouse. Evi1 expression is frequently found in human AML, myelodysplastic syndrome and CML.26 Six additional insertions were identified in proximity to genes that were listed in the retrovirus tagged cancer gene database (RTCGD27) at least two times (Chd2, Fli1, Mycn, Runx2, Pik3r1 and Gdf7; Supplementary Table 5). Transcriptional upregulation of Evi1 and Prdm16 was confirmed by RT–PCR analysis in the four mice with vector insertions close to these genes (Figure 6b). In addition, Evi1 and Prdm16 insertions correlated with high leukocytosis.
NRas and Evi1 cooperate in vitro to support cell expansion
As Evi1 might support myeloid leukemic growth, we analyzed cell expansion of primary murine BM cells expressing NRas(wt), Evi1 or both in an in vitro competitive proliferation assay. NRas induced the most pronounced positive effect on proliferation 2 weeks after transduction (Figure 6c). This delay may reflect the negative effect of NRas expression shortly after transduction in the HSC-enriched lin− BM cells (Supplementary Figure 4). Co-expression of Evi1 positively affected proliferation after 3 weeks, whereas Evi1 expression alone did not markedly alter cell expansion (Figure 6c). The proportion of Evi1/NRas(wt) double-positive cells increased overtime and outcompeted the NRas(wt)-transduced cells (Figure 6d). Likewise, the Evi1 single-positive cell fraction increased remarkably (Supplementary Figure 6) as was reported earlier.28, 29
Evi1 was reported to induce expression of the anti-apoptotic protein Bcl-2, by regulating miR449A and Bcl-xL.30, 31 The effect of Evi1 overexpression may therefore be an improved survival of NRas(wt)-expressing cells. We generated Evi1-expressing clones by insertional mutagenesis in murine lin− BM cells using the RSF91.NRas(wt).I.GFP.pre or the RSF91.I.eGFP.pre control vectors. After cultivating 2 weeks, cells were replated in 96-well plates and five clones from positive wells of each vector were expanded.32 Real-time PCR analysis confirmed high expression of Evi1 in the control vector clones (1700–5700-fold compared with untransduced expanded cultures) and a lower, but still strongly increased Evi1 expression in NRas(wt) clones (90–470-fold, Figure 6e). Bcl-2 was upregulated in all clones and correlated positively with Evi1 expression levels (Figure 6e) whereas Bcl-xL expression was not increased. As NRas-induced cell proliferation is primarily mediated via CyclinD1 expression,33 we analyzed CyclinD1 expression in control vector clones (Evi1-positive) and NRas(wt) clones (Evi1/NRas-positive). CyclinD1 expression was increased up to threefold in NRas(wt) clones compared with untransduced, expanded cultures but only mildly in control vector clones (Figure 6e).
To confirm these findings, we sorted Evi1/NRas(wt) and NRas(wt)-positive cells from the double-transduced competitive proliferation assay cultures. CyclinD1 expression was increased in all NRas(wt)-expressing cells but reduced 30% upon Evi1 co-expression (Evi1/NRas 5.4±1.7 versus NRas 7.6±2.2-fold, n=3, Supplementary Figure 7).
The four AML mice with Evi1/Prdm16 insertions presented with less apoptotic cells in the spleen compared with the other AML and mHC mice (Supplementary Figure 5). Bcl-2 and CyclinD1 were upregulated in all AML mice independent of the Evi1/Prdm16 insertion (Supplementary Figure 8). However, all AML underwent clonal selection and will have selected for secondary events.
Our data suggest that Evi1-induced Bcl-2 upregulation supports cell survival whereas reduced levels of CyclinD1 lead to slower cell cycle progression that may counteract the negative effect of the growth factor hyperresponsiveness by NRas overexpression.
In this study, we investigated the consequences of hyperactive signaling through wild-type NRas without mutational activation. There are conflicting reports regarding the role of NRas(wt) overexpression in cellular transformation. NRas(wt) overexpression was reported to have tumor suppressor activity,34 while others provide evidence that NRas(wt) overexpression predisposes cells to malignant transformation.35, 36, 37, 38 Our results are in accordance with an earlier report that tumorigenesis by NRas(wt) overexpression and mutationally activated NRas occurs by activation of the same pathways.38 As many leukemias are induced by mutations in hematopoietic cytokine receptors or fusion proteins with kinase activity that mediate their transforming activity via Ras downstream signaling pathways,9, 39 our model mimics important features of human leukemogenesis.
Mice that were transplanted with NRas(wt)-expressing cells developed AML (70%) or mHC (30%). The cells of both phenotypes exhibited monocytic/macrophage cell surface markers (CD11b, F4/80). However, the leukemias were more immature with higher ER-MP58 expression, while the histiocytic cells expressed markers of macrophages/monocytes and myeloid dendritic cells (MHCII/CD11c). Histiocytes belong to the monocytic phagocyte system including macrophages, monocytes and myeloid dendritic cells that share as the common progenitor the monocyte/macrophage/DC precursor (MDP).40, 41 Owing to their large heterogeneity in histiocytic sarcoma, it is not possible to distinguish macrophages and dendritic cells based on conventional surface markers (F4/80, CD11b, CD11c, MHCII). In our model, histiocytes induced by NRas(G12D) had a higher CD11c/MHCII expression in comparison with histiocytes induced by NRas(wt). Nevertheless, it can be assumed that the transformation initiated by NRas occurred at the stage of a common progenitor in both models. The observation that NRas(wt) induced mainly clonal AML involving granulocytic and monocytic proliferation, but also mHC in other mice, and the many shared markers on leukemic and mHC cells suggest that transformation occurred at the stage of the myeloid progenitor or earlier.
Expression of the mutationally active NRas induced a much more aggressive form of mHC. Analysis of the downstream signaling activation by NRas(wt) and NRas(G12D) showed that both activated MAPK (Erk1/2, p38) and PI3K/Akt signaling, as well as Stat3 and Stat5 in vivo. Elevated proliferative and pro-survival signaling are critical components contributing to malignant transformation. Our observation of higher levels of activated NRas(G12D) as compared with NRas(wt) may indicate that prolonged NRas activation may have led to accelerated mHC development.
In murine models, infection with the replication-competent Harvey murine leukemia virus or the histiocytosis sarcoma virus which both express the HRas oncogene in hematopoietic cells, induced predominantly mHC.21, 42 In addition, Eμ-NRas(G12D) mice present with a late-onset mHC phenotype.14, 43 However, full penetrance was not observed in these models.
Retroviral expression of NRas(G12D) in a bone marrow transplantion (BMT) model with a Moloney murine leukemia virus-based vector caused myeloproliferative neoplasm with long latency and incomplete penetrance,16 whereas NRas(G12D) expression using a murine stem cell virus-based vector induced CMML or AML together with mastocytosis, depending on NRas expression levels.17 Expression from the spleen focus-forming virus enhancer/promoter is higher than from the Moloney murine leukemia virus and murine stem cell virus vectors44 as this vector was optimized for expression in hematopoietic cells.45, 46 Therefore, the high NRas(G12D) expression, in combination with a good transduction efficiency might have caused the complete penetrance of mHC in our model. The lack of tumor progression beyond mHC may be a consequence of the rapid and severe disease onset.
The late onset of NRas(wt)-induced myelomonocytic leukemia and the oligoclonal nature of the disease suggest that additional genetic alterations were required for leukemic transformation, similar to the two-class model for leukemic progression defined in human AML.47 Insertion site analysis identified Evi1 and Prdm16 as potential cooperating genes in NRas(wt)-induced AML. In addition, we detected an insertion close to NMyc, which was previously described as a cooperating gene in the Eμ-NRas(G12D) transgenic murine model.14 According to the two-class model of malignant transformation in AML,47 NRAS mutations belong to the class I mutations that confer proliferative and survival advantages to cells.47 Additional class II mutations in genes impairing hematopoietic differentiation (RUNX1, RARA and EVI1) are necessary to induce progression to AML. Interestingly, several studies show that AML patients with Evi1 activation due to inv(3)(q21q26) and t(3;3)(q21;q26), or with other balanced translocations involving chromosomal region 3q26 such as t(2;3)(p15-23;q26.2), t(3;12)(q26.2;p13) and t(3;21)(q26.2;q22.1), frequently carry NRAS mutations (28% and 25%, respectively) or NFI deletions (16.2%).48, 49, 50 Co-existence of PRDM16 overexpression with FLT3-ITD mutations was found in five of nine analyzed AML cases.51 Further evidence for a potential cooperation of Evi1 with NRas was demonstrated in a mouse model expressing NRas(G12D) from the endogenous locus using a replication-competent virus.52
Overexpression of the zinc-finger transcription factor Evi1 in mice induces myelodysplasia53 and the cooperation with other oncogenes is necessary for progression to AML.54, 55 Insertional activation of Evi1 causes leukemia.27, 56, 57, 58 Evi1 expression supports myeloid cell growth in vitro,28, 29 blocks differentiation,59 modulates microRNAs60 and modifies epigenetic regulation.61 Conversely, Evi1 was found to arrest cells in G0/G1 and to induce apoptosis in differentiating cells in vitro.62 Evi1 may also upregulate anti-apoptotic proteins such as Bcl-2 and Bcl-xL,30, 31 and Bcl-2 expression is a rate limiting step in the progression of NRas-induced leukemias.63
Our experiments show that co-expression of Evi1 in NRas-transduced murine lin− BM cells support cell expansion in vitro and cultures selected for NRas/Evi1 double-positive cells. We detected Bcl-2 upregulation in Evi1-positive in vitro cultures and downregulation of NRas-induced CyclinD1 expression. Evi1 may therefore collaborate with NRas-induced transformation by compensating negative effects of hyperactive NRas signaling, while NRas counteracts Evi1-mediated growth reduction.
In summary, we have shown that ectopic expression of NRas(wt) activates signaling pathways such as PI3K/Akt and Mek/Erk, which are commonly deregulated in human leukemia. Overexpression of NRas(wt) in mice induced myelomonocytic leukemias that reflect the phenotype of human AML M4 often found to be associated with hyperactive NRas signaling. Our findings provide further evidence that hyperactive NRas signaling can cooperate with Evi1 and Prdm16 expression to cause leukemic transformation. The murine NRas(wt) AML model is therefore suitable to study the mechanism of how NRas signaling cooperates with oncogenes such as EVI1 and PRDM16, or others, in human leukemia.
Materials and methods
The retroviral vectors RSF91.IRES.GFP.pre and RSF91.IRES.dTomato.pre were kindly provided by Axel Schambach (Institute of Experimental Hematology, Hannover Medical School, Germany).64 For co-expression of NRas, we inserted the cDNA of murine wild-type NRas (NRas(wt)) or the murine constitutively active G12D mutant NRas (NRas(G12D)) upstream of the ribosomal entry site element. The murine Evi1 gene lacking the PR domain28 was expressed by the gammaretroviral SIN vector SRS.SF.mEvi1.pre.mCMV.GFP, which co-expresses the eGFP marker gene.
Production of vector supernatant
Viral supernatants were generated by calcium phosphate transfection of 293T packaging cells by co-transfection of the backbone constructs, the gag/pol construct (MLV gag/pol) and the ecotropic envelope construct (provided by T Kitamura, Tokyo, Japan). Cells were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal calf serum, 100 U/ml penicillin/streptomycin and 2 mmol/l glutamine. Viral titers determined on murine SC-1 fibroblasts by flow cytometry were in the range of 106 to 107 IU/ml.
Isolation and transduction of lineage-negative BM cells
Lin− cells were isolated from BM of C57BL/6 J mice by magnetic sorting using lineage-specific antibodies (Lineage Cell depletion kit; Miltenyi, Bergisch Gladbach, Germany). Cells were cultivated in StemSpan medium (Stem Cell Technologies, Vancouver, Canada) supplemented with 20 ng/ml mTPO, 10 ng/ml mSCF, 10 ng/ml hFGF-1, 20 ng/ml mIGF-2 (PeproTech, Heidelberg, Germany), 1% penicillin/streptomycin and 2 mmol/l glutamine, and were transduced on 2 consecutive days with a multiplicity of infection of 3. Virus preloading was carried out on RetroNectin-coated (10 μg/cm; Takara, Otsu, Japan) suspension culture dishes by spinoculation for 30 min at 4 °C.
Murine BMT model
Retrovirally-transduced lin− BM cells (0.5–1 × 106 cells per mouse) were transplanted into lethally irradiated (10 Gy) syngeneic C57BL/6J recipient mice. For secondary transplantation, 106 BM cells from diseased mice were transplanted together with 105 BM cells isolated from healthy donors. Transplanted mice were monitored and euthanized upon disease manifestation. Animal experiments were approved by the local ethics committee and performed according to their guidelines.
In vitro competition proliferation assay
Lin− murine BM cells were transduced with the vectors RSF91.NRas(wt).I.dTomato.pre, RSF91.I.GFP.pre or SRS.SF.mEvi1.pre.mCMV.GFP. Cells were cultured in Iscove's modified Dulbecco's medium, 10% fetal calf serum, 20 ng/ml mTPO, 10 ng/ml mSCF, 10 ng/ml hFGF-1, 20 ng/ml mIGF-2 (PeproTech, Heidelberg, Germany), 1% penicillin/streptomycin and 2 mmol/l glutamine, and counted at indicated time points.
Immortalized cell clones
Cell clones were generated using the vectors RSF91.NRas(wt).I.GFP.pre and RSF91.I.GFP.pre. Immortalization was induced by insertional mutagenesis by the gammaretroviral long-terminal repeat as described in the In Vitro Immortalization Assay.29, 32
Immunoblotting was performed as previously described.65 The following antibodies were used for protein detection: NRas, GFP, secondary antibodies goat anti-rabbit-IgG (Santa Cruz Biotechnology, Heidelberg, Germany), pAkt (Ser473), Akt, pErk1/2 (Thr202/Thr204), Erk1/2 (Cell Signaling Technology, Danvers, MA, USA). Pull-down and detection of GTP-bound Ras was performed using a GST-Raf1-Ras binding domain according to the manufacturer’s instructions (Thermo Scientific, Bonn, Germany).
Flow cytometric analysis
Spleen cells (0.5–1 × 106) were stained with antibodies detecting Gr-1, CD41, CD11b, F4/80, CD11c, MHCII (eBioscience, Frankfurt, Germany and BD Biosciences, Heidelberg, Germany) and ER-MP58 (Santa Cruz Biotechnology). For analysis of intracellular signaling proteins, 0.5–1 × 106 cells were fixed and permeabilized according to the manufacturer’s instructions and stained with Alexa-Fluor-647-conjugated antibodies detecting pErk1/2 (pT202/pY204), pAkt (pS473), p38 MAPK (pT180/pY182), pStat5 (pY694) or pStat3 (pY705) (BD Biosciences). Flow cytometric data were acquired with FACSCalibur or LSRII (BD Biosciences) and analyzed with FlowJo Version 7.2.2.
Southern blot and ligation-mediated PCR analysis
DNA of spleen cells was purified using QIAmp Blood DNA Preparation Kit (Qiagen, Hilden, Germany). Genomic DNA (10 μg) was digested with BglII, which cleaves proviral sequences at a unique site and Southern blotting was performed according to standard protocols. Membranes were hybridized with a 32P-labeled probe corresponding to the PRE sequence of the retroviral vector or a 569 bp NRas(G12D) cDNA fragment. Ligation-mediated PCR was performed as described.66, 67
Quantitative PCR was performed on the StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) using the Quantitect SYBR Green PCR kit (Qiagen) and Quantitect primer assays. Evi1 was detected using primers as described.32
Bourne HR, Sanders DA, McCormick F . The GTPase superfamily: a conserved switch for diverse cell functions. Nature 1990; 348: 125–132.
Downward J . Targeting RAS signalling pathways in cancer therapy. Nat Rev Cancer 2003; 3: 11–22.
Schlenk RF, Dohner K, Krauter J, Frohling S, Corbacioglu A, Bullinger L et al. Mutations and treatment outcome in cytogenetically normal acute myeloid leukemia. N Engl J Med 2008; 358: 1909–1918.
Kohlmann A, Grossmann V, Klein HU, Schindela S, Weiss T, Kazak B et al. Next-generation sequencing technology reveals a characteristic pattern of molecular mutations in 72.8% of chronic myelomonocytic leukemia by detecting frequent alterations in TET2, CBL, RAS, and RUNX1. J Clin Oncol 2010; 28: 3858–3865.
Koike K, Matsuda K . Recent advances in the pathogenesis and management of juvenile myelomonocytic leukaemia. Br J Haematol 2008; 141: 567–575.
Reuter CW, Morgan MA, Bergmann L . Targeting the Ras signaling pathway: a rational, mechanism-based treatment for hematologic malignancies? Blood 2000; 96: 1655–1669.
Carroll M, Tomasson MH, Barker GF, Golub TR, Gilliland DG . The TEL/platelet-derived growth factor beta receptor (PDGF beta R) fusion in chronic myelomonocytic leukemia is a transforming protein that self-associates and activates PDGF beta R kinase-dependent signaling pathways. Proc Natl Acad Sci USA 1996; 93: 14845–14850.
Loh ML, Sakai DS, Flotho C, Kang M, Fliegauf M, Archambeault S et al. Mutations in CBL occur frequently in juvenile myelomonocytic leukemia. Blood 2009; 114: 1859–1863.
Niemeyer CM, Kang MW, Shin DH, Furlan I, Erlacher M, Bunin NJ et al. Germline CBL mutations cause developmental abnormalities and predispose to juvenile myelomonocytic leukemia. Nat Genet 2010; 42: 794–800.
Steinemann D, Arning L, Praulich I, Stuhrmann M, Hasle H, Stary J et al. Mitotic recombination and compound-heterozygous mutations are predominant NF1-inactivating mechanisms in children with juvenile myelomonocytic leukemia and neurofibromatosis type 1. Haematologica 2010; 95: 320–323.
Arico M, Biondi A, Pui CH . Juvenile myelomonocytic leukemia. Blood 1997; 90: 479–488.
Morgan MA, Ganser A, Reuter CW . Therapeutic efficacy of prenylation inhibitors in the treatment of myeloid leukemia. Leukemia 2003; 17: 1482–1498.
Haupt Y, Harris AW, Adams JM . Retroviral infection accelerates T lymphomagenesis in E mu-N-ras transgenic mice by activating c-myc or N-myc. Oncogene 1992; 7: 981–986.
Wang J, Liu Y, Li Z, Du J, Ryu MJ, Taylor PR et al. Endogenous oncogenic Nras mutation promotes aberrant GM-CSF signaling in granulocytic/monocytic precursors in a murine model of chronic myelomonocytic leukemia. Blood 2010; 116: 5991–6002.
MacKenzie KL, Dolnikov A, Millington M, Shounan Y, Symonds G . Mutant N-ras induces myeloproliferative disorders and apoptosis in bone marrow repopulated mice. Blood 1999; 93: 2043–2056.
Parikh C, Subrahmanyam R, Ren R . Oncogenic NRAS rapidly and efficiently induces CMML- and AML-like diseases in mice. Blood 2006; 108: 2349–2357.
Wiesner SM, Jones JM, Hasz DE, Largaespada DA . Repressible transgenic model of NRAS oncogene-driven mast cell disease in the mouse. Blood 2005; 106: 1054–1062.
Leenen PJ, Melis M, Slieker WA, Van Ewijk W . Murine macrophage precursor characterization. II. Monoclonal antibodies against macrophage precursor antigens. Eur J Immunol 1990; 20: 27–34.
Hao X, Fredrickson TN, Chattopadhyay SK, Han W, Qi CF, Wang Z et al. The histopathologic and molecular basis for the diagnosis of histiocytic sarcoma and histiocyte-associated lymphoma of mice. Vet Pathol 2010; 47: 434–445.
Leenen PJ, Bechan GI, Melis M, den Broeder BJ, Lohler J, Egeler RM . Heterogeneity in a mouse model of histiocytosis: transformation of Langerin+ dendritic cells, macrophages, and precursors. J Leukoc Biol 2010; 87: 949–958.
Pileri SA, Grogan TM, Harris NL, Banks P, Campo E, Chan JK et al. Tumours of histiocytes and accessory dendritic cells: an immunohistochemical approach to classification from the International Lymphoma Study Group based on 61 cases. Histopathology 2002; 41: 1–29.
Serrano M, Lin AW, McCurrach ME, Beach D, Lowe SW . Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16INK4a. Cell 1997; 88: 593–602.
Takaoka M, Harada H, Deramaudt TB, Oyama K, Andl CD, Johnstone CN et al. Ha-Ras(G12V) induces senescence in primary and immortalized human esophageal keratinocytes with p53 dysfunction. Oncogene 2004; 23: 6760–6768.
Kotecha N, Flores NJ, Irish JM, Simonds EF, Sakai DS, Archambeault S et al. Single-cell profiling identifies aberrant STAT5 activation in myeloid malignancies with specific clinical and biologic correlates. Cancer Cell 2008; 14: 335–343.
Lugthart S, van Drunen E, van Norden Y, van Hoven A, Erpelinck CA, Valk PJ et al. High EVI1 levels predict adverse outcome in acute myeloid leukemia: prevalence of EVI1 overexpression and chromosome 3q26 abnormalities underestimated. Blood 2008; 111: 4329–4337.
Akagi K, Suzuki T, Stephens RM, Jenkins NA, Copeland NG . RTCGD: retroviral tagged cancer gene database. Nucleic Acids Res 2004; 32: D523–D527.
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.
Modlich U, Navarro S, Zychlinski D, Maetzig T, Knoess S, Brugman MH et al. Insertional transformation of hematopoietic cells by self-inactivating lentiviral and gammaretroviral vectors. Mol Ther 2009; 17: 1919–1928.
De Weer A, Van der Meulen J, Rondou P, Taghon T, Konrad TA, De Preter K et al. EVI1-mediated down regulation of MIR449A is essential for the survival of EVI1 positive leukaemic cells. Br J Haematol 2011; 154: 337–348.
Pradhan AK, Mohapatra AD, Nayak KB, Chakraborty S . Acetylation of the proto-oncogene EVI1 abrogates Bcl-xL promoter binding and induces apoptosis. PLoS One 2011; 6: e25370.
Modlich U, Bohne J, Schmidt M, von Kalle C, Knob S, Schambach A et al. Cell culture assays reveal the importance of retroviral vector design for insertional genotoxicity. Blood 2006; 108: 2545–2553.
Pylayeva-Gupta Y, Grabocka E, Bar-Sagi D . RAS oncogenes: weaving a tumorigenic web. Nat Rev Cancer 2011; 11: 761–774.
Diaz R, Ahn D, Lopez-Barcons L, Malumbres M, Perez de Castro I, Lue J et al. The N-ras proto-oncogene can suppress the malignant phenotype in the presence or absence of its oncogene. Cancer Res 2002; 62: 4514–4518.
Diaz R, Lopez-Barcons L, Ahn D, Garcia-Espana A, Yoon A, Matthews J et al. Complex effects of Ras proto-oncogenes in tumorigenesis. Carcinogenesis 2004; 25: 535–539.
Mangues R, Seidman I, Gordon JW, Pellicer A . Overexpression of the N-ras proto-oncogene, not somatic mutational activation, associated with malignant tumors in transgenic mice. Oncogene 1992; 7: 2073–2076.
Mangues R, Kahn JM, Seidman I, Pellicer A . An overexpressed N-ras proto-oncogene cooperates with N-methylnitrosourea in mouse mammary carcinogenesis. Cancer Res 1994; 54: 6395–6401.
Mangues R, Symmans WF, Lu S, Schwartz S, Pellicer A . Activated N-ras oncogene and N-ras proto-oncogene act through the same pathway for in vivo tumorigenesis. Oncogene 1996; 13: 1053–1063.
Braun BS, Shannon K . Targeting Ras in myeloid leukemias. Clin Cancer Res 2008; 14: 2249–2252.
Geissmann F, Manz MG, Jung S, Sieweke MH, Merad M, Ley K . Development of monocytes, macrophages, and dendritic cells. Science 2010; 327: 656–661.
Cline MJ . Histiocytes and histiocytosis. Blood 1994; 84: 2840–2853.
Lohler J, Franz T, Fusco A, Pragnell I, Ostertag W . Murine retrovirus-induced malignant histiocytosis, an experimental model for the disease in humans. Leukemia 1987; 1: 58–68.
Braig M, Lee S, Loddenkemper C, Rudolph C, Peters AH, Schlegelberger B et al. Oncogene-induced senescence as an initial barrier in lymphoma development. Nature 2005; 436: 660–665.
Baum C, Hegewisch-Becker S, Eckert HG, Stocking C, Ostertag W . Novel retroviral vectors for efficient expression of the multidrug resistance (mdr-1) gene in early hematopoietic cells. J Virol 1995; 69: 7541–7547.
Wahlers A, Zipfel PF, Schwieger M, Ostertag W, Baum C . In vivo analysis of retroviral enhancer mutations in hematopoietic cells: SP1/EGR1 and ETS/GATA motifs contribute to long terminal repeat specificity. J Virol 2002; 76: 303–312.
Wahlers A, Kustikova O, Zipfel PF, Itoh K, Koester M, Heberlein C et al. Upstream conserved sequences of mouse leukemia viruses are important for high transgene expression in lymphoid and hematopoietic cells. Mol Ther 2002; 6: 313–320.
Gilliland DG, Griffin JD . The roles of FLT3 in hematopoiesis and leukemia. Blood 2002; 100: 1532–1542.
Haferlach C, Bacher U, Haferlach T, Dicker F, Alpermann T, Kern W et al. The inv(3)(q21q26)/t(3;3)(q21;q26) is frequently accompanied by alterations of the RUNX1, KRAS and NRAS and NF1 genes and mediates adverse prognosis both in MDS and in AML: a study in 39 cases of MDS or AML. Leukemia 2011; 25: 874–877.
Bacher U, Haferlach T, Schoch C, Kern W, Schnittger S . Implications of NRAS mutations in AML: a study of 2502 patients. Blood 2006; 107: 3847–3853.
Lugthart S, Groschel S, Beverloo HB, Kayser S, Valk PJ, van Zelderen-Bhola SL et al. Clinical, molecular, and prognostic significance of WHO type inv(3)(q21q26.2)/t(3;3)(q21;q26.2) and various other 3q abnormalities in acute myeloid leukemia. J Clin Oncol 2010; 28: 3890–3898.
Shing DC, Trubia M, Marchesi F, Radaelli E, Belloni E, Tapinassi C et al. Overexpression of sPRDM16 coupled with loss of p53 induces myeloid leukemias in mice. J Clin Invest 2007; 117: 3696–3707.
Li Q, Haigis KM, McDaniel A, Harding-Theobald E, Kogan SC, Akagi K et al. Hematopoiesis and leukemogenesis in mice expressing oncogenic NrasG12D from the endogenous locus. Blood 2011; 117: 2022–2032.
Buonamici S, Li D, Chi Y, Zhao R, Wang X, Brace L et al. EVI1 induces myelodysplastic syndrome in mice. J Clin Invest 2004; 114: 713–719.
Jin G, Yamazaki Y, Takuwa M, Takahara T, Kaneko K, Kuwata T et al. Trib1 and Evi1 cooperate with Hoxa and Meis1 in myeloid leukemogenesis. Blood 2007; 109: 3998–4005.
Cuenco GM, Ren R . Both AML1 and EVI1 oncogenic components are required for the cooperation of AML1/MDS1/EVI1 with BCR/ABL in the induction of acute myelogenous leukemia in mice. Oncogene 2004; 23: 569–579.
Kustikova O, Fehse B, Modlich U, Yang M, Dullmann J, Kamino K et al. Clonal dominance of hematopoietic stem cells triggered by retroviral gene marking. Science 2005; 308: 1171–1174.
Modlich U, Schambach A, Brugman MH, Wicke DC, Knoess S, Li Z et al. Leukemia induction after a single retroviral vector insertion in Evi1 or Prdm16. Leukemia 2008; 22: 1519–1528.
Calmels B, Ferguson C, Laukkanen MO, Adler R, Faulhaber M, Kim HJ et al. Recurrent retroviral vector integration at the Mds1/Evi1 locus in nonhuman primate hematopoietic cells. Blood 2005; 106: 2530–2533.
Laricchia-Robbio L, Premanand K, Rinaldi CR, Nucifora G . EVI1 Impairs myelopoiesis by deregulation of PU.1 function. Cancer Res 2009; 69: 1633–1642.
Gomez-Benito M, Conchillo A, Garcia MA, Vazquez I, Maicas M, Vicente C et al. EVI1 controls proliferation in acute myeloid leukaemia through modulation of miR-1-2. Br J Cancer 103: 1292–1296.
Vazquez I, Maicas M, Cervera J, Agirre X, Marin-Bejar O, Marcotegui N et al. Down-regulation of EVI1 is associated with epigenetic alterations and good prognosis in patients with acute myeloid leukemia. Haematologica 2011; 96: 1448–1456.
Konrad TA, Karger A, Hackl H, Schwarzinger I, Herbacek I, Wieser R . Inducible expression of EVI1 in human myeloid cells causes phenotypes consistent with its role in myelodysplastic syndromes. J Leukoc Biol 2009; 86: 813–822.
Omidvar N, Kogan S, Beurlet S, le Pogam C, Janin A, West R et al. BCL-2 and mutant NRAS interact physically and functionally in a mouse model of progressive myelodysplasia. Cancer Res 2007; 67: 11657–11667.
Schambach A, Mueller D, Galla M, Verstegen MM, Wagemaker G, Loew R et al. Overcoming promoter competition in packaging cells improves production of self-inactivating retroviral vectors. Gene Ther 2006; 13: 1524–1533.
Meyer J, Jucker M, Ostertag W, Stocking C . Carboxyl-truncated STAT5beta is generated by a nucleus-associated serine protease in early hematopoietic rogenitors. Blood 1998; 91: 1901–1908.
Schmidt M, Hoffmann G, Wissler M, Lemke N, Mussig A, Glimm H et al. Detection and direct genomic sequencing of multiple rare unknown flanking DNA in highly complex samples. Hum Gene Ther 2001; 12: 743–749.
Kustikova OS, Modlich U, Fehse B . Retroviral insertion site analysis in dominant haematopoietic clones. Methods Mol Biol 2009; 506: 373–390.
We thank Hans Grundtke, Jörg Frühauf and Martin Werner (Radiotherapy) for irradiation of the mice, Sabine Knöß, Johanna Krause (Institute of Experimental Hematology) and Andrea Schienke (Institute of Cell and Molecular Pathology) for their excellent technical help, Axel Schambach (Institute of Experimental Hematology) for providing the vector backbone and Sandra Ließem (Institute of Pathology) for help with the immunohistochemistry (all Hannover Medical School). This study was supported by grants from the Deutsche Forschungsgemeinschaft (DFG RU- 1476 and Excellence Cluster REBIRTH).
The authors declare no conflict of interest.
Supplementary Information accompanies the paper on the Oncogene website
About this article
Cite this article
Wolf, S., Rudolph, C., Morgan, M. et al. Selection for Evi1 activation in myelomonocytic leukemia induced by hyperactive signaling through wild-type NRas. Oncogene 32, 3028–3038 (2013). https://doi.org/10.1038/onc.2012.329
- myelomonocytic leukemia
- malignant histiocytosis
- murine BMT model
- retroviral vector
Chronic iron exposure and c-Myc/H-ras-mediated transformation in fallopian tube cells alter the expression of EVI1, amplified at 3q26.2 in ovarian cancer
The heteromeric transcription factor GABP activates the ITGAM/CD11b promoter and induces myeloid differentiation
Biochimica et Biophysica Acta (BBA) - Gene Regulatory Mechanisms (2015)
Cell Reports (2015)
Identification of Campath-1 (CD52) as Novel Drug Target in Neoplastic Stem Cells in 5q-Patients with MDS and AML
Clinical Cancer Research (2014)