Background
Ras proteins are small GTPases that act as molecular switches, transducing signals from activated receptors to downstream effectors to regulate cell proliferation, survival and differentiation [
1]. Members of the Ras family include three cellular
Ras genes, which encode four highly homologous proteins: H-, N-, and K-Ras4A and 4B, the latter two being alternatively spliced isoforms differing only at the carboxyl terminus (with alternative 4
th exon) [
2]. H-, N-, and K-Ras proteins are widely expressed, with K-Ras expressing in almost all cell types [
3]. The four Ras proteins share an identical amino-terminal region. This region includes the nucleotide binding and effector loop domains, providing a basis for the ability of Ras proteins to interact with a common set of activators and effectors and to share many biochemical and biological functions. The next 80 amino acids are also highly homologous (85% identity) among all four Ras proteins. Conversely, the Ras proteins diverge highly at the carboxyl-terminal 23–24 amino acids, referred to as the hypervariable domain (HVR) [
4]. This HVR domain signals the post-translational modifications that enable the Ras proteins to attach to the inner surface of the plasma membrane, a prerequisite for Ras-mediated signal transduction. Different modifications and targeting mechanisms result in localization of the Ras proteins to functionally distinct microdomains of the plasma membrane, which may allow the Ras proteins access to different pools of Ras regulators and/or effectors and, therefore, generate distinct signal outputs [
3].
Gene knockout studies in mice have revealed functional differences as well as redundancies between Ras proteins. Mice that lack the expression of N-Ras or H-Ras or both are viable and have no obvious abnormal phenotype [
5,
6]. But the mice lacking the
K-Ras gene die during embryonic development between days 12 (E12) and 14 (E14), with fetal liver defects and evidence of anemia [
7,
8]. These results demonstrate that K-Ras provides a unique and essential function during mouse development. Interestingly, the mice heterozygous for K-Ras (K-Ras
+/-) and homozygous null for N-Ras (N-Ras
-/-) die between E10 and 12, while the K-Ras
+/- mice are normal. Furthermore, no viable double homozygous mutant (K-Ras
-/-; N-Ras
-/-) was recovered at E9.5. These results suggest that there is partial redundancy between Ras proteins [
7].
In addition to their normal cellular functions, Ras proteins play critical roles in tumorigenesis. Mutated
Ras genes are associated with approximately 30% of all human cancers, including both solid tumors and hematological malignancies [
9]. In addition to the direct activation by mutations, Ras can also be functionally activated by other oncogenic mutations, such as the BCR/ABL fusion protein.
BCR/ABL is produced when the breakpoint cluster region gene (
BCR) sequences on chromosome 22 are fused to
ABL sequences on chromosome 9 by a reciprocal translocation [
10]. The BCR/ABL fusion protein is present in nearly all patients with chronic myelogenous leukemia (CML) and in 20% of the adult and 2–5% of the pediatric patients with B-acute lymphoblastic leukemia (B-ALL). We and others have shown that expression of BCR/ABL in mouse bone marrow cells by retroviral transduction efficiently induces a myeloproliferative disease (MPD) resembling human CML [
11,
12]. In this model system, BCR-ABL with a mutation (Y177F) at the tyrosine-177 residue – a high affinity-binding site for the Grb2 SH2 domain when phosphorylated [
13,
14] – induced a T cell leukemia and lymphoma after a prolonged latent period [
15‐
17]. Grb2 is an SH2- and SH3-containing adapter protein that binds Sos – a guanine nucleotide exchange factor (GEF) of Ras – as well as the scaffolding adapter Gab2 [
18]. These interactions activate Ras, and recruits phosphotidylinositol-3 kinase (PI3-K) and the protein tyrosine phosphatase SHP2 [
13,
14,
18]. The importance of Y177 in the induction of CML-like MPD by BCR-ABL suggests that activation of Ras signaling pathways plays a critical role in the pathogenesis of CML. Consistent with this result, BCR/ABL transformation was shown to be blocked by interfering with Ras function using either a dominant negative mutant of Ras or the catalytic domain of Ras-GAP [
19].
The redundant function among Ras proteins makes it difficult to assess the importance of Ras in development and tumorigenesis. The Ras mutant that contains an asparagine at position 17 functions as a competitive inhibitor of the normal endogenous wild-type Ras for the binding of GEFs [
1]. The N17 H-Ras mutant has been shown to be the most effective in inhibiting the activation of all three isoforms of Ras in the cell, due to the wide distribution of this protein throughout the plasma membrane as compared to K-Ras and N-Ras [
20‐
22]. To assess the role of Ras in adult hematopoiesis and in BCR/ABL leukemogenesis, and to assess the importance of Ras as a therapeutic target, we examined the effect of N17 H-Ras mutant in bone marrow reconstitution and in the induction of CML-like MPD by BCR/ABL in mice.
Methods
DNA constructs
A vector capable of expressing N17 H-Ras and BCR/ABL fused to GFP on the 3' end of BCR/ABL in identical levels was made through a multistep cloning strategy. The MSCV-BCR/ABL-GFP-IRES-N17Ras was constructed as follows: The dominant negative Ras (N17 H-Ras) (generous gift provided by Dr. Hong Cai in Dana-Farber Cancer Institute) was amplified by PCR using the primers 5'CGC GGA TCC ATG ACA GAA TAC AAG CTT GTG GTC3' and 5'ACG CGT CGA CTC AGG AGA GCA CAC ACT TGC AGC T3' which contain a Bam HI and Sal I internal sites. The PCR product was digested with Bam HI and Sal I, cloned into pBluescript II/SK
- (Stratagene, La Jolla, CA) into the Bam HI/Sal I sites, and then sequenced to verify that there were no introduced mutations. This N17 H-Ras sequence was then compared with the H-Ras sequence of the Genbank database (Accession #J00277). To clone the N17 H-Ras fragment into the pCITE (Novagen, Madison, WI) vector that carries the IRES (internal ribosomal entry site) sequence necessary to allow the Ras expression, the N17 H-Ras fragment was ligated to the annealed adapters 5' CAT GGC GGC CGC TG 3' and 3' CGC CGG CGA CCT AG 5' to generate a Nco I site which was necessary for cloning into pCITE. This construct was then cut with Eco RI and Sal I to isolate the Ras insert to clone it into MSCV retroviral vector [
23]. This vector is capable of transducing genes with high efficiency into undifferentiated murine embryonic carcinomas and hematopoietic stem cells. GFP or BCR/ABL-GFP fusion protein was cloned into the EcoRI site of MSCV as described previously [
24]. In order to clone the IRES-neo fragment a similar strategy as described above was followed to clone the other construct MSCV-BCR/ABL-GFP-IRES-Neo. Another vector capable of expressing N17 H-Ras and GFP on the 5': MSCV-GFP-IRES-N17 H-Ras was constructed as follows: The GFP was amplified by PCR using the primers 5' AAG AAT TCG CGG CCG CCA TGA CAG AAT ACA AGC TTG TGG TGG TG3' AND 5'AAG AAT TCG CGG CCG CTC AGG AGA GCA CAC ATT GCA GCT CAT G3' which contain an Eco RI and Not I internal sites. The PCR product was digested with Eco RI, cloned into MSCV-IRES-N17 H-Ras into the Eco RI site. Orientation was checked by enzyme digestion. Finally, the MSCV-IRES-GFP vector control was constructed as described before [
12].
Cell culture, retrovirus preparation and viral titer quantification
NIH 3T3 and BOSC 23 cells were grown as previously described [
25,
26]. Helper-free retroviruses were generated by transient transfection of retroviral DNA constructs into BOSC 23 cells as previously described [
12,
25]. Retroviral transduction and tittering was performed as described [
12,
25]. All viruses were adjusted to equal titer based on GFP expression just before transduction of BM cells or other cell types.
NIH 3T3 cells (1 × 105) were plated on 60 mm Petri dish. After 18 hours the cells were infected for 4 hs with 1 ml, 0.5 ml, 0.1 ml, 0.01 ml, or 0.001 ml of retroviral supernatant per duplicate plates. Seven to ten days after infections the plates corresponding to the different retroviral dilutions were analyzed and the number of Foci quantified. For bone marrow soft agar colony assays, the retroviral-transduced cells were washed twice in excess phosphate-buffered saline (PBS) (GIBCO BRL, Grand Island, N.Y.), counted and plated into six-well plates at 105 cells per well (0.3% Bacto agar, 20% FBS, 200 U of penicillin/ml, 200 μg of streptomycin/ml, 50 M 2-mercaptoethanol in DMEM).
Infection, transplantation and colony assay of bone marrow cells
Bone marrow cell infection and transplantation were performed as described previously [
12]. Briefly, 5-fluoroucil (5-FU)- treated bone marrow cells (1 × 10
6 cells/ml) from male Balb/c (Taconic Farm, Germantown, NY) donor mice (5 to 6 weeks old) were infected for 24 hours in a cocktail containing DMEM, 15% FCS, 5% WEHI-conditioned medium, 7 ng/ml IL-3 (Genzyme, Cambridge, MA), 12 ng/ml IL-6 (R&D System, Minneapolis, MN), 56 ng/ml stem cell factor (SCF, R&D System), 2 mM L-glutamine (GIBCO BRL), 100 μg/ml streptomycin, 100 U/ml penicillin, 0.25 μg/ml amphotericin B (GIBCO BRL), 2 μg/ml polybrene and 30% of retroviral supernatant. After 24 hours, the bone marrow cells were re-infected with freshly made retroviruses. Two days post infection, the bone marrow cells were washed once with PBS and then injected into lethally irradiated (2 doses of 450 rads given 4 hours apart) syngeneic female BALB/c mice (5 to 6 weeks old) through tail vein injection of 4 × 10
5 cells per mouse. The secondary bone marrow transplant was performed with 5 × 10
6 whole bone marrow cells from mice reconstituted with MSCV-IRES-GFP or MSCV-GFP-IRES-N17 Ras bone marrow cells into lethaly irradiated BALB/c mice.
In vitro soft agar colony assays were performed as described previously [
26] with modifications. From the infected cells used for bone marrow transplantation, 10
5 infected BM cells were plated in 35-mm-diameter cells in DMEM, 20% FCS, 100 μg of streptomycin/ml, 100 U of penicillin/ml, 0.25 μg of amphoterecin B/ml, 50 μM 2-mercaptoethanol (Sigma), and 0.3% bacto-agar, on top of a layer of medium containing 0.6% bacto-agar. Colonies were counted on day 15. BM liquid cultures were grown in the absence of cytokines. Cytospin preparations of these cultures were stained with Hema-3.
In vitro methylcellulose colony assays
Bone marrow cells for
in vitro methylcellulose colony assays were either isolated from bone marrow of our retroviral transduced mice or were obtained from double retroviral infection of 5-FU enriched isolated bone marrow cells from 5–6 weeks male Balb/c mice (Taconic Farm, Germantown, NY) [
12]. Upon isolation, these cells were immediately sorted for GFP positive populations as described previously [
12,
25].
MethoCult™ CF M3434 and MethoCult™ M3334 (StemCell Technologies, Vancouver, BC, Canada) were used for myeloid and erythroid colony assays respectively. Sorted GFP positive cells were diluted to 2.0 × 105 cells/mL in Isccove's MDM (IMDM, StemCell Technologies, Inc.) with 2% FBS for MethoCult™ CF M3434 and 2 × 106 cells/mL for MethoCult™ M3334. 0.3 mL of cells were added to aliquoted 3 mL of MethoCult™ CF M3434 and MethoCult™ M3334 for duplicate cultures. Tubes of the MethoCult™ CF M3434 and MethoCult™ M3334 containing the cells were vortexed and bubbles were allowed to rise and dissipate for 5–10 minutes. 1.1 mL of the culture was dispensed into each of two 35 mm culture dishes using a 3 cc syringe and 16 g blunt end needle. Cultures in MethoCult™ CF M3434 were incubated for 10 days and cultures in MethoCult™ M3334 for 5 days at 37°C and 5% CO2a humidified incubator.
Analysis of disease phenotype in mice
Hematopathological analysis was performed as previously described [
12,
25]. Total blood cells and white blood cells (WBCs) were counted on Coulter Counter (Model Z1; Coulter Particle Characterization, Hialeah, Fla.). Immunophenotypes of hematopoietic cells and leukemic cells were determined by flow cytometry. Flow cytometry and cell sorting were performed as described [
12].
Western blot analysis
Total cell lysates of BM cells or NIH 3T3 cells were prepared as described [
25]. Proteins were separated on a 6–15% gradient SDS-polyacrilamide gel electrophoresis (SDS-PAGE) and electrophoretically transferred to nitrocellulose filters (Schleicher & Schuell, Keene, NH). The filters were probed with anti-Abl monoclonal antibody Ab-3 (Oncogene Research Products, Cambridge, MA), rat anti-Ras monoclonal antibody (Santa Cruz Biotechnology, Inc. Santa Cruz, CA), anti-dynamin (Transduction Laboratories, Lexington, KY), anti-actin (clone AC-40, Sigma, St. Louis, MO, USA), anti-phospho-p44/42 (Thr202/Tyr204) MAP kinases (Erk1 and Erk2), anti-p44/42 MAP kinases (Erk1 and Erk2) (New England Biolabs, Beverly, MA, USA). Bound antibodies were visualized by using horseradish peroxidase-conjugated anti-mouse, anti rabbit, or anti-rat IgG respectively (Southern Biotechnology Associates, Inc., Birmingham, AL, USA).
Mice perfusion
Mice were anesthetized and then cut from the lower abdomen to the chest to expose heart, lungs, liver and thymus. A 18-gauge needle was inserted through the left ventricle of the heart into the ascending aorta, and slowly perfused the animal with 50 ml room temperature PBS until the blood was replaced by the saline fluid and liver was blanched to a yellow color. The right atrium of the heart was cut to suction the outflow.
Discussion
Expression of the
N17 H-Ras transgene under the control of the Eμ enhancer and the
lck proximal promoter arrests B cell development at the pro-B cell stage [
29]. Our experiment did not show GFP
high B lymphoid cells, suggesting that high-levels of N17 H-Ras expression completely suppressed B-lymphopoiesis. There are two critical differences between our experiment and the previous one. One difference is the cells into which the N17 H-Ras was targeted. In previous experiment, the Eμ enhancer first becomes active in the pre-pro-B cells, whereas in our experiment, MSCV may drive the transgene expression in more primitive progenitor cells and completely block differentiation of such cells into B cell lineage. The other essential difference involves the expression levels of N17 H-Ras. Retrovirus-mediated transgene expression varies in target cells depending on its site of integration in the host chromosome. When a strong retroviral LTR promoter/enhancer is combined with certain integration sites, the transgene can be highly expressed in target cells. This may result in more effective suppression of Ras activation. Our finding that low expression of N17 H-Ras blocked B-lymphocyte development at the pro-B cell stage is consonant with, and builds upon, what was found previously. Taken together, these results offer two insights. First, certain levels of Ras activation are required for the earliest B-lymphocyte development. Second, the development of B-lymphocytes from pro-B to pre-B cell stage requires high levels of Ras activation.
Expression of the
N17 H-Ras transgene under the control of the
lck proximal promoter has been shown to nearly completely block the proliferation of thymocytes in response to T cell receptor (TCR) stimulation and severely impair the positive selection of thymocytes [
30]. In this report we found that the development of thymocytes was completely blocked by N17 H-Ras expression, at both high and low expression levels. As discussed above, the difference may due to targeting N17 H-Ras expression into different cells. Lck is first expressed in double negative (DN) thymocytes, whereas MSCV may target N17 H-Ras expression into earlier T lymphocyte precursors and/or lymphoid progenitors. Our results suggest that Ras activation is critical for early thymocyte development.
Similar to the effect of N17 H-Ras in B-lymphocyte development, high expression of N17 H-Ras completely suppressed erythopoiesis, whereas low expression of N17 H-Ras inhibited the expansion of BFU-E, but not CFU-E. A recent study also showed that dominant negative Ras does not reduce CFU-E colony formation [
31].
In contrast to the critical role of Ras activation in lymphopoiesis and erythropoiesis, myelopoiesis seemed not to be blocked by expression of N17 H-Ras. It is possible that there are redundant pathways downstream of myeloid growth receptors that regulate proliferation and survival of myeloid cells. It has been shown that both oncogenic Ras and STAT5 can induce myeloproliferative disease [
32‐
34], yet adult hematopoiesis appears normal in STAT5A/B-deficient mice [
35]. These two pathways may overlap in promoting myeloid cell proliferation and survival.
As mentioned in the introduction, BCR-ABL with a mutation at the tyrosine-177 residue – a high affinity-binding site for the Grb2 SH2 domain [
13,
14] – induced a T cell leukemia and lymphoma after a prolonged latent period [
15‐
17]. Consistent with the idea that Ras is a critical downstream target of Y177, N17 H-Ras greatly attenuates BCR/ABL's ability to induce the fatal CML-like MPD. However, unlike the BCR/ABL Y177F phenotype, BCR/ABL-GFP + N17 H-Ras mice developed B, instead of T, lymphoblast leukemia/lymphoma. These results suggest that the Y177 motif in BCR/ABL also regulates signaling pathway(s) other than Ras. Indeed, in addition to recruiting the Ras activator Sos, Grb2 also binds the scaffolding adapter Gab2, which in turn recruits SHP2 and PI3K [
18]. It has been shown that the ability of BCR/ABL to confer cytokine-independent growth of primary myeloid cells isolated from Gab2
-/- mice in vitro was significantly reduced [
18]. These results suggest the latter signaling pathways also play important roles mediating Y177 function in BCR/ABL.
Inhibiting Ras activation by N17 H-Ras has its limitations, as using dominant negative mutants in general. For example, Ras GEFs may have other substrates. Inhibiting Ras GEF, therefore, may affect molecules other than Ras. However, our findings show that the effect of N17 H-Ras in hematopoiesis is lineage, developmental stage, and dosage dependent. This differential role of Ras in the reconstitution and development of different blood lineages and in leukemogenesis warrants further studies.
Conclusion
In this study we assessed the importance of Ras in BCR/ABL leukemogenesis and the potential toxicity of inhibiting Ras in normal hematopoiesis and found that inhibition of Ras activation by N17 H-Ras suppresses the development of erythrocytes, T and B-lymphocytes, but not myeloid cells. In addition, the presence of N17 H-Ras greatly attenuated BCR/ABL-induced chronic myelogenous leukemia (CML)-like myeloproliferative disease. Most BCR/ABL + N17 H-Ras mice eventually developed B lymphoblastic leukemia/lymphoma. These results suggest that Ras activation is important for the development of lymphoid and erythroid cells but not myeloid cells, and that Ras is a critical target of BCR/ABL in the pathogenesis of CML, but not B-ALL.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
RR designed research, analyzed data, and drafted the manuscript. KB participated in the design of the study, performed research, analyzed data, and drafted the manuscript. All authors read and approved the final manuscript.