Introduction

The initiation and progression of the developmental program of terminal myeloid differentiation involves the activation of positive regulators and suppression of negative regulators, the latter found among genes that control myeloid cell growth. Lesions in genes that control differentiation may lead to either anemia or leukemia (Hoffman et al., 2002). The autonomously proliferating M1 myeloblastic leukemia cell line is induced by interleukin-6 (IL-6) to undergo terminal macrophage differentiation with concomitant loss of leukemogenicity, providing an attractive model system to analyse the different regulators of terminal differentiation. Using the M1 cell line, it has been shown in our laboratory that the proto-oncogenes c-myc and c-myb and the transcription factor E2F-1 are negative regulators, and Egr-1 is a positive regulator of myeloid differentiation (Nguyen et al., 1993; Hoffman-Liebermann and Liebermann, 1991; Selvakumaran et al., 1992; Krishnaraju et al., 1998; Amanullah et al., 2000).

Previously, this laboratory has demonstrated that deregulated expression of E2F-1, a pivotal transcription factor for the transition through the G₁/S phase of the cell cycle, blocks terminal myeloid differentiation, and prevents the loss of leukemogenicity of IL-6-treated M1 cells in vivo (Amanullah et al., 2000).

The early growth response gene 1 (Egr-1), a member of the Egr family of genes that encodes for zinc-finger transcription factors, plays a role in the development, growth control and survival of several cell types, including T and B cells, neuronal cells and myeloid cells (Beckmann and Wilce, 1992; McMahon and Monroe, 1996; Krishnaraju et al., 1995, 1998). Egr-1 acts as a growth stimulator in human prostate tumors, but behaves as a tumor suppressor in other cells (Liu et al., 1999; Baron et al., 2003; Shafarenko et al., 2005). Egr-1, a macrophage differentiation primary response gene, has been shown to be essential for and to restrict differentiation along the macrophage lineage (Nguyen et al., 1993). Consistent with these findings, Laslo et al. (2006) have demonstrated that Egr-1 and Egr-2 are major positive modulators of macrophage differentiation under the direction of the primary transcription factor PU.1, although studies conducted by Carter and Tourtellotte (2007) claim that Egr proteins are dispensable for macrophage differentiation. Ectopic Egr-1 expression potentiated macrophage differentiation of the hematopoietic precursor cell line 32Dcl3, and stimulated the development of bone marrow-derived hematopoietic progenitor cells along the macrophage lineage at the expense of granulocyte and erythroid lineages (Krishnaraju et al., 1995, 2001). Deregulated expression of Egr-1 in the M1 cell line activated the macrophage differentiation program in the absence of differentiation inducer (Krishnaraju et al., 1998). Furthermore, in M1Myc cells, deregulated expression of Egr-1 abrogated the block in terminal differentiation imparted by c-Myc, resulting in cells that have the characteristics of functionally mature macrophages, diminished the aggressiveness of M1Myc leukemias and abrogated the leukemic potential of IL-6-treated M1Myc cells (Shafarenko et al., 2005), although the cells did not undergo G₀/G₁ arrest. These observations are consistent with Egr-1 behaving like a tumor suppressor gene in myeloid cells.

In this study, we asked if Egr-1 also functions as a tumor suppressor by abrogating the block in differentiation caused by deregulated expression of E2F-1, which blocks differentiation at an earlier stage than the proto-oncogene c-myc. This work shows that expressing Egr-1 in M1E2F-1 cells completely overrode the block in terminal myeloid cell differentiation, abrogated E2F-1-driven leukemogenicity, and, unlike in cells expressing both c-Myc and Egr-1, also underwent G₀/G₁ arrest. These findings raise the possibility that Egr-1 and/or Egr-1 target genes can provide important tools for differentiation therapy in certain types of leukemias.

Results

Establishment of M1E2F-1/Egr-1 cell lines

M1E2F-1/Egr-1 cell lines were established as described in the legend to Figure 1. The M1E2F-1/Egr-1 clones #1 and #2, with levels of transcripts and proteins comparable to parental M1E2F-1 and M1Egr-1 clones, were chosen for further analysis. In all experiments, M1E2F-1/Egr-1 cells were compared to M1, M1E2F-1 and M1Egr-1 cells, both without and following treatment with IL-6.

Figure 1.

Establishment of M1E2F-1/Egr-1 cell lines. M1E2F-1 cells were transduced with MSCV-neo retroviral vector, encoding for murine Egr-1 transgene. M1 clones were assessed for expression of transgenes by western blotting. Egr-1 expression was determined in untreated cells, whereas E2F-1 expression was assessed only after the cells were treated with interleukin-6 (IL-6) for 3 days to reduce endogenous E2F-1 levels.

Full figure and legend (45K)

IL-6-treated M1E2F-1/Egr-1 cells undergo terminal myeloid differentiation

To determine the effect of coexpression of the positive regulator Egr-1 and the negative regulator E2F-1 on M1 terminal differentiation program, M1, M1Egr-1, M1E2F-1 and M1E2F-1/Egr-1 cells were treated with the differentiation inducer IL-6 and assessed for differentiation. This assessment included the morphology of the cells, cell surface markers, differentiation-associated proteins and the ability to phagocytose.

May–Grunwald–Giemsa-stained cytospin smears of M1 and each of the established M1 variant cell lines, untreated or treated with IL-6 for 3 and 6 days, were scored for the percentage of cells at different stages of macrophage differentiation. Interestingly, induction of M1E2F-1/Egr-1 differentiation was similar to both M1 and M1Egr-1, assessed by the presence of mature cells (Figures 2a and b). In contrast, IL-6-treated M1E2F-1 cells remained predominantly in the immature blast stage, as previously reported by this laboratory (Amanullah et al., 2000). Thus, coexpression of the tumor suppressor Egr-1 with E2F-1 resulted in morphological maturation of M1 cells following treatment with IL-6.

Figure 2.

Analysis of differentiation-associated properties of M1, M1Egr-1, M1E2F-1 and M1E2F-1/Egr-1 cells. Cells seeded at 1 10⁵ cells/ml were induced to differentiate with interleukin-6 (IL-6; 50 ng/ml). (a) Photomicrographs of May–Grunwald–Giemsa-stained cytospin smears, untreated (0 day) and with IL-6 for 3 and 6 days. Images were viewed through a Zeiss Axioplan microscope using a 40/0.75 NA lens and captured using a Sensys camera and Spot software. (b) Morphology of cells treated with IL-6 for 3 days was determined. Results are the average of three experiments with s.d.<10%. Percentage of mature cells was similar between M1, M1Egr-1 and M1E2F-1/Egr-1 cells (P>0.1, not significant), but different between M1E2F-1 and other three lines (P<0.05, significant). (c) F4/80 expression was determined using anti-F4/80 antibody. The percentage is average of three experiments. Significant differences in expression between M1E2F-1 and M1E2F-1/Egr-1 (^*) were observed (P<0.05). (d) Representative northern blot, showing ferritin and lysozyme mRNA expression. (e and f) Cells were seeded in the absence or presence of IL-6, and at 4 days were analysed for phagocytosis of latex beads. (e) Photomicrographs of images captured using a Zeiss Axioplan microscope with a 40 objective and a Sensys camera assisted by Spot software. Each is representative of at least 20 fields. Scale bar, 20 m. (f) Phagocytic cells engulfing latex beads were analysed by flow cytometry. There is no difference for all untreated M1 cell variant. Differences between IL-6-treated M1E2F-1 cells and IL-6-treated M1E2F-1/Egr-1 #1 and #2 cells (^*) are significant with P<0.05.

Full figure and legend (247K)

M1E2F-1/Egr-1 cells were further examined for expression of the macrophage differentiation markers F4/80, ferritin and lysozyme. About 45% of IL-6-treated M1E2F-1/Egr-1 cells expressed the macrophage differentiation marker F4/80, comparable to similarly treated M1 and M1Egr-1 cells, and in contrast to less than 10% for the M1E2F-1 cells (Supplementary Figure 1; Figure 2c). Elevated levels of both ferritin and lysozyme were observed in M1, M1Egr-1 and M1E2F-1/Egr-1 cells treated with IL-6, but not in the M1E2F-1 cells (Figure 2d). These data further demonstrate the differentiation of the M1E2F-1/Egr-1 cells.

One functional characteristic of mature macrophages is its ability to phagocytose. M1, M1Egr-1, M1E2F-1 and M1E2F-1/Egr-1 cells, untreated or following 4 days with IL-6, were assayed for phagocytosis. Fluorescent images show that a large proportion of the IL-6-treated M1E2F-1/Egr-1 cells engulfed multiple latex beads, comparable to similarly treated M1 and M1Egr-1 cells (Figure 2e), and in contrast to IL-6-treated M1E2F-1 cells. Quantification by FACS analysis corroborated these observations (Figure 2f; Supplementary Figure 2).

Subsequent to the completion of the myeloid differentiation program, M1 and normal myeloid cells undergo programmed cell death, which starts between 4 and 5 days following treatment with IL-6 (Shafarenko et al., 2005). Using annexin/PI to assess apoptosis, a significant percentage (35%) of IL-6-treated M1E2F-1/Egr-1 cells shifted toward the lower right quadrant, comparable to what was observed for similarly treated M1 and M1Egr-1 cells (Figures 3a and b). Only a small percentage of the IL-6-treated M1E2F-1 cells (9%) were apoptotic under similar conditions; however, these cells continued to proliferate (see below, Figure 4a), and at later times, the percent of apoptotic cells was reduced (data not shown). After 8 days treatment with IL-6, no viable cells were detected in the M1, M1Egr-1 and M1E2F-1/Egr-1 cell populations, whereas the M1E2F-1 cell population was still viable and proliferating (see the section, M1E2F-1/Egr-1 become growth arrested following treatment with IL-6 and Figure 4a).

Figure 3.

Analysis of apoptosis associated with terminal differentiation. Cells were either untreated or treated with interleukin-6 (IL-6) for 5 days. (a) Flow cytometric analysis showing the percentages of Annexin V-positive cells. There is no difference in the profile shown for untreated M1 cells and each of the other untreated cell lines (data not shown). Results are representative of at least four independent experiments. (b) Summary of total percentages of early apoptotic cells. ^* Denotes the differences between M1E2F-1 and M1E2F/Egr-1 cells (P<0.05).

Full figure and legend (110K)

Figure 4.

Growth kinetics and cell cycle analysis of cells treated with interleukin-6 (IL-6). (a) At indicated time points, viable cell numbers were determined using a hemocytometer. Results are the average of three independent experiments with s.d. up to 15%. () M1; () M1Egr-1; (^*) M1E2F/Egr#1; () M1E2F/Egr#2. (b) Cell cycle analysis by flow cytometry on cells treated with or without IL-6 for 3 days. This is a representative experiment, done four times, showing distribution of cells in cell cycle. There is a significant reduction of cells in S phase of IL-6-treated M1E2F-1/Egr-1 cells compared to IL-6-treated M1E2F-1 (P<0.05). The profiles of IL-6-treated M1E2F-1 and M1E2F-1/Egr-1 cells are shown below.

Full figure and legend (78K)

Taken together, these data demonstrate that constitutive expression of Egr-1 abrogated the E2F-1-mediated block in differentiation, allowing the cells to differentiate into functional, mature macrophages that subsequently underwent apoptosis.

M1E2F-1/Egr-1 become growth arrested following treatment with IL-6

M1 cells constitutively expressing E2F-1 not only failed to differentiate but also did not undergo growth arrest when treated with IL-6 (Amanullah et al., 2000). Since concomitant expression of Egr-1 and E2F-1 in M1 cells resulted in terminally differentiated cells following treatment with IL-6, we asked if these differentiated cells also underwent normal growth arrest similar to parental M1 cells.

M1E2F-1/Egr-1 cells did not proliferate after 2 days (Figure 4a). To determine whether the impeded growth of the M1E2F-1/Egr-1 cells was the result of cell cycle arrest in G₀/G₁, the distribution of cells in the different phases of the cell cycle was assessed by flow cytometry. Following 3 days of treatment with IL-6, there was an accumulation of M1E2F-1/Egr-1 cells in the G₀/G₁ phase of the cell cycle (Figure 4b), with a concomitant loss of cells in the S phase.

To address the growth arrest mechanism of M1E2F-1/Egr-1 cells, analysis of regulatory genes of the cell cycle pathway was performed. These regulators, many being target genes of either E2F-1 or Egr-1, included the positive regulators of proliferation CDK2, CDK4, cyclin D1, cyclin E, cdc25A and c-Myc. Expression of all these proteins was substantially reduced or turned off in IL-6-treated M1E2F-1/Egr-1 cells, similar to M1 and M1Egr-1 cells (Figure 5a). In contrast, in IL-6-treated M1E2F-1 cells, CDK2, cyclin D1 and cyclin E did not appear to be regulated, CDK4 protein was reduced, and both myc and CDC25A expression were reduced after 1 day and then returned to the initial level; by 8 days, the expression levels of all these proteins resembled untreated M1E2F-1 (data not shown).

Figure 5.

Expression of cell cycle regulators in cells following treatment with interleukin-6 (IL-6). (a) Positive regulators of proliferation. (b) Negative regulators of proliferation. (c) Phosphorylation status of pRb. These are representative experiments done at least three times.

Full figure and legend (181K)

Expression analysis was also performed on negative regulators of proliferation. The previously reported induction of the cdk inhibitors p15INK4B and p16INK4A in IL-6-treated M1E2F-1 and M1 cells (Amanullah et al., 2000) was corroborated and extended to M1Egr-1 and M1E2F-1/Egr-1 cells (data not shown). The cdk inhibitor p27 is constitutively expressed in all the cell lines, and does not appear to be regulated in either M1E2F-1/Egr-1 cells or parental M1 and M1Egr-1 following IL-6 treatment. However, in M1E2F-1 cells, p27 expression is significantly reduced following IL-6 treatment. There was induction of the cdk inhibitor p21 in IL-6-treated M1E2F-1/Egr-1 cells as well as similarly treated parental M1 and M1Egr-1 (Figure 5b), compared to very low levels in M1E2F-1 cells.

It has been previously reported by us that deregulated E2F-1 overrode the p15/p16-Rb-E2F checkpoint in the G₁ phase of the cell cycle, since both p15INK4B and p16INK4A were induced in IL-6-treated M1E2F-1 cells yet pRb was not completely hypophosphorylated and the cells did not undergo G₀/G₁ arrest (Amanullah et al., 2000). To understand further how Egr-1 coexpression with E2F-1 promotes growth arrest in M1 cells following IL-6 treatment, the expression profile and phosphorylation status of pRb was analysed (Figure 5c). In M1E2F/Egr-1 cells, the level of pRb expression and hypophosphorylation resembled M1 and M1Egr-1 cells, unlike M1E2F-1 cells. Taken together, these data show that Egr-1 restored the expression levels of cell cycle regulators, consistent with the observed G₀/G₁ arrest following IL-6 treatment in M1E2F-1/Egr-1 cells.

Egr-1 expression suppresses the leukemic phenotype of M1E2F-1 cells treated with IL-6

Intravenous injection of untreated M1 cells into nude or syngeneic mice results in the rapid development of a leukemic phenotype that is similar to acute myelogenous leukemia (AML) in human patients (Selvakumaran et al., 1992). The ability of M1 cells to cause leukemia is lost after induction of differentiation in vitro or in vivo. Deregulated expression of E2F-1 not only blocked M1 differentiation but also prevented the loss of their leukemic phenotype (Amanullah et al., 2000). In contrast, deregulated expression of Egr-1 in M1 cells activated macrophage differentiation and decreased the aggressiveness of M1 leukemias (Krishnaraju et al., 1998). Therefore, it was imperative to assess the effect of concomitant expression of Egr-1 and E2F-1 on the leukemic potential of M1 cells, either untreated or following treatment with IL-6.

M1, M1Egr-1, M1E2F-1 and M1E2F-1/Egr-1 cells, with or without IL-6 treatment, were i.v. injected into CD-1 nu/nu mice to assay for leukemia. All mice injected with untreated cells died between 4 and 7 weeks (Figure 6a) and the M1E2F-1-injected mice died significantly faster than the mice injected with the other cell types. Mice injected with M1Egr-1 cells appeared to survive longer than mice injected with parental M1 cells, consistent with previously reported results (Krishnaraju et al., 1998). M1E2F-1/Egr-1 cells behaved similarly to parental M1 cells and were less aggressive than M1E2F-1 cells, demonstrating that Egr-1 reduced the oncogenic potential of M1E2F-1 cells.

Figure 6.

Leukemogenicity determination. For each cell type 10⁶ cells, with or without interleukin-6 (IL-6) for 5 days, were suspended in 100 l 1 phosphate-buffered saline (PBS) and i.v. (tail vein) injected into eight CD-1 nude mice. Control animals were injected with same volume of 1 PBS. Experiment was terminated after 8 weeks. (a) Kaplan–Meier survival curve of nude mice injected with untreated cells. Differences were observed between M1E2F-1-injected animals versus the other three groups (P<0.05, significant). (b) Survival curve analysis of nude mice injected with IL-6-treated cells. Surviving M1, M1Egr-1 and M1E2F-1/Egr-1-injected mice were asymptomatic.

Full figure and legend (18K)

Cells first treated with the differentiation inducer IL-6 for 5 days were inoculated into nude mice. All nude mice injected with M1E2F-1/Egr-1 cells survived up to 12 weeks, whereas all the nude mice injected with M1E2F-1 cells died by week 7 (Figure 6b). Thus, M1E2F-1 cells did not lose the ability to cause leukemia following in vitro treatment with IL-6, as previously reported (Amanullah et al., 2000), whereas M1E2F-1/Egr-1 cells behaved like parental M1, losing the ability to cause leukemia.

Using one mouse from each experimental group, bone marrow was examined 3 weeks after injection to assess for the presence of myeloid leukemic cells, determined by growth and differentiation characteristics. Myeloid leukemic cells were recovered from mice injected with each of the untreated M1 variants as well as the IL-6-treated M1E2F-1 cells, but not from mice injected with IL-6-treated M1, M1Egr-1 and M1E2F-1/Egr-1 cells. Taken together, these results demonstrate that Egr-1 diminished the aggressiveness of M1E2F-1 leukemia and abrogated the leukemic potential of IL-6 -treated M1E2F-1 cells.

Discussion

This work has shown that deregulated expression of Egr-1 in IL-6-treated M1E2F-1 cells alleviated the block in the myeloid differentiation program, resulting in functionally mature macrophages that underwent G₀/G₁ arrest. Apoptosis occurred following differentiation, as also observed in similarly treated parental M1 cells. In addition, Egr-1 expression abrogated the leukemic potential of IL-6-treated M1E2F-1 cells and diminished the aggressiveness of M1E2F-1 leukemias. These data show that Egr-1 functions as a tumor suppressor when myeloid differentiation is blocked by E2F-1 (Figure 7).

Figure 7.

Comparing tumor suppressor effect of Egr-1 in M1myc and M1E2F-1 cells. Note that deregulated c-myc blocks differentiation at a later stage than deregulated E2F-1, and M1Myc/Egr-1 cells do not undergo G₀/G₁ arrest after interleukin-6 (IL-6) treatment.

Full figure and legend (144K)

Recently, we showed that Egr-1 abrogates the block in terminal M1 myeloid differentiation imparted by deregulated c-myc, which blocks differentiation at a later stage than E2F-1, resulting in cells that have the characteristics of functionally mature macrophages that undergo apoptosis. However, in contrast to M1E2F-1/Egr-1 cells, M1Myc/Egr-1 cells did not undergo G₀/G₁ arrest (Figure 7). Egr-1 is dominant to E2F for the block in both growth arrest and terminal differentiation, whereas it is dominant to c-Myc only for the block in differentiation. These findings are consistent with Egr-1 behaving as a tumor suppressor against two oncogenes, each preventing myeloid differentiation by a different mechanism. Interestingly, the human EGR-1 gene was localized to the q region of chromosome 5, where either deletion or monosomy is frequently observed in patients with myelodysplastic syndromes or acute myeloid leukemia (Bram et al., 2004; Giagounidis et al., 2004). It would be worthwhile to assess Egr-1 expression in patients with either myelodysplasia or AML at time of diagnosis and during the course of treatment, and to correlate these expression levels to outcome.

It was suggested by Amanullah et al. (2000) that sufficient functional E2F-1 remained in IL-6-treated M1E2F-1 cells to promote cell cycle progression in spite of expression of p15/p16. We confirmed that E2F-1 protein levels in M1E2F-1/Egr-1 cells are similar to M1E2F-1 cells following IL-6 treatment, yet regulation of all the parameters of cell cycle control examined is consistent with the observed growth arrest. The restoration of induction of TGF- and p35 (data not shown), two putative Egr-1 target genes that are negative regulators of proliferation (Liu et al., 1999; Chen et al., 2004), to parental cell levels and downregulation of c-Myc, a positive regulator of cell cycle progression that remains elevated in IL-6-treated M1E2F-1 cells, takes place in IL-6-treated M1E2F/Egr-1 cells. These changes may counteract the effect of continued expression of E2F-1; blocking expression of TGF- and p35, separately and together, should give further insights into the regulation of the cell cycle during myeloid differentiation. Further mechanistic insights into how growth arrest associated with myeloid differentiation is regulated will be obtained by ongoing studies comparing M1E2F-1/Egr-1, which differentiate and undergo G₀/G₁ arrest, to M1Myc/Egr-1, which only differentiate.

It was proposed by Amanullah et al. (2000) that deregulated E2F-1 blocks terminal myeloid differentiation by preventing growth arrest, yet it was recently shown for M1Myc/Egr-1 cells that myeloid cell differentiation can occur in the absence of G₀/G₁ arrest (Shafarenko et al., 2005). Therefore, how deregulated E2F-1 expression in M1 cells perturbs the myeloid differentiation program is still an open question. In addition, how Egr-1 promotes M1E2F-1 myeloid cell differentiation is not understood.

Determination of changes in gene and protein expression and post-translation modifications using the established M1 variant cell lines should clarify how Egr-1 overrides the oncogenic blocks in myeloid differentiation. These studies should provide further insights into how Egr-1 and/or Egr-1 target genes may be used to treat or suppress oncogene-driven hematological malignancies.

Materials and methods

Cells, culture, cytokines and mice

The M1 murine myeloblastic leukemia cell line previously described (Shafarenko et al., 2005). Stably transduced M1Egr-1 and M1E2F-1 cell lines were similarly cultured with the addition of G418 (200 g/ml) or puromycin (2 g/ml) to maintain selection of the transgenes. To induce differentiation, cells were treated with 50 ng/ml of purified recombinant human IL-6 (Amgen, Thousand Oak, CA, USA) after being seeded at a concentration of 1 10⁵ cells/ml. Mice were 8-week-old CD-1 nu/nu mice (Charles River Laboratories, Wilmington, MA, USA).

Establishment of M1 cell line that ectopically expresses both Egr-1 and E2F-1 transgenes

Cell lines were established as described previously (Shafarenko et al. 2005). Several independent M1E2F-1/Egr-1 clones were characterized for the level of expression of their transgenes. Initially, five clones were found to behave similarly. More extensive analysis was done using two of the clones throughout this study, and gave comparable results. Control cell lines were indistinguishable from their respective parental controls.

General recombinant DNA techniques and expression vectors

Plasmids, DNA probes, MSCV-neo-Egr-1 and MSCV-puro-E2F-1 (Krishnaraju et al., 1998; Amanullah et al., 2000) were described previously.

Analysis of cell morphology

This was done as described previously (Krishnaraju et al., 1998; Shafarenko et al., 2005).

Phagocytosis analysis

Cells were seeded at 1 10⁵ cells/ml with or without IL-6 (50 ng/ml). The night before harvesting, latex beads (3.0 m size Fluoresbrite YG latex beads; Polyscience, Warrington, PA, USA; 150/cell) were added. At indicated times, cells were centrifuged (10 min, 2000 r.p.m.), cell pellets were fixed in ethanol, counterstained with Alexa Fluor 568 palloidin (Molecular Probes, Eugene, OR, USA) for F-actin and a nuclear dye, Hoechst (Molecular Probes). Cells, mounted on slides, were analysed by fluorescent microscopy using an Olympus AH-3 microscope ( 40/0.7 NA objective with filters for FITC, Texas Red and 4',6-diamidino-2-phenylindole (DAPI)). Images were acquired through Insight camera using imaging software SPOT. Latex beads are visualized with the FITC filter, F-actin with the Texas Red filter, and nuclei with the DAPI filter. To quantify phagocytic cells, cells were subjected to flow cytometry (BD FACSCalibur) using Cell Quest Pro 5.2 software (Becton Dickinson, San Jose, CA, USA; Steinkamp et al., 1982).

Analysis of apoptosis by annexin/PI

Apoptosis associated with terminal myeloid differentiation was measured using the annexin V-FITC apoptosis detection kit II along with propidium iodide (PI) staining for cell viability, according to the manufacturer's instructions (BD PharMingen, San Diego, CA, USA), and analysed by flow cytometry.

RNA extraction, northern blots and probes

This was performed as described previously (Krishnaraju et al., 1998; Shafarenko et al., 2005).

Protein extraction and immunoblotting (western blots)

Protein extraction and immunoblotting were described previously (Shafarenko et al., 2005). Primary antibodies against E2F-1, Egr-1, cdc25A, c-Myc, cyclin A, cyclin D1, pRB, cdk4, p21, p27 and -actin were from Cell Signaling Technology, Inc. (Danvers, MA, USA). Antibodies against cdk2 and cyclin E were from Abcam Inc. (Cambridge, MA, USA).

Flow cytometry to analyse cell cycle distribution and F4/80 cell surface markers

Cell cycle analysis was as previously described (Shafarenko et al., 2005). For F4/80 cell surface marker assessment, cells were labeled with PE-conjugated F4/80 antibody (eBioscience, San Diego, CA, USA) and analysed by flow cytometry on a BD FACSCalibur along with using Cell Quest Pro 5.2 software (Becton Dickinson).

Assay for leukemia

Eight-week-old CD-1 nu/nu mice were i.v. injected (tail vein) with 10⁶ cells suspended in 100 l phosphate-buffered saline (PBS). Control animals were injected with same volume of PBS.

Statistical analysis

Values are meanss.d. of n independent experiments. Statistical analysis was performed using Student's t-test and Kaplan–Meier survival curve with the aid of a statistical software SPSS 11.

References

1. Amanullah A, Hoffman B, Liebermann DA. (2000). Deregulated E2F-1 blocks terminal differentiation and loss of leukemogenicity of M1 myeloblastic leukemia cells without abrogating induction of p15^INH4B and p16^INK4A. Blood 96: 475–482. | PubMed | ISI | ChemPort |
2. Baron V, De Gregorio G, Krones-Herzig A, Virolle T, Calogero A, Urcis R et al. (2003). Inhibition of Egr-1 expression reverses transformation of prostate cancer cells in vitro and in vivo. Oncogene 22: 4194–4204. | Article | PubMed | ISI | ChemPort |
3. Beckmann AM, Wilce PA. (1992). Egr transcription factors in the nervous system. Neurochem Int 31: 477–510. | Article |
4. Bram S, Rodjer S, Swolin B. (2004). Several chromosomes involved in translocation with chromosome 5 shown with fluorescence in situ hybridization in patients with malignant myeloid disorders. Cancer Genet Cytogenet 155: 74–78. | Article | PubMed | ISI | ChemPort |
5. Carter JH, Tourtellotte WG. (2007). Early growth response transcriptional regulators are dispensable for macrophage differentiation. J Immunol 178: 3038–3047. | PubMed | ISI | ChemPort |
6. Chen F, Wang Q, Wang X, Studzinski GP. (2004). Up-regulation of Egr-1 by 1, 25-dihydroxyvitamin D3 contributes to increased expression of p35 activator of cyclin-dependent kinase 5 and consequent onset of the terminal phase of HL60 cell differentiation. Cancer Res 64: 5425–5433. | Article | PubMed | ISI | ChemPort |
7. Giagounidis AA, Germing U, Haase S, Hildebrandt B, Schlegelberger B, Schoch C et al. (2004). Clinical morphological, cytogenetic, and prognostic features of patients with myelodysplastic syndromes and del (5q) including band q31. Leukemia 18: 113–119. | Article | PubMed | ISI | ChemPort |
8. Hoffman B, Amanullah A, Shafarenko M, Liebermann DA. (2002). The proto-oncogene c-myc in hematopoietic development and leukemogenesis. Oncogene 21: 3414–3421. | Article | PubMed | ISI | ChemPort |
9. Hoffman-Liebermann B, Liebermann DA. (1991). Interleukin-6 and leukemia inhibitory factor-induced terminal differentiation of myeloid leukemia cells is blocked at an intermediate stage by constitutive c-myc. Mol Cell Bio 11: 2375–2381. | ChemPort |
10. Krishnaraju K, Hoffman B, Liebermann DA. (1998). The zinc finger transcription factor Egr-1 activates macrophage differentiation in M1 myeloblastic leukemia cells. Blood 92: 1957–1966. | PubMed | ISI | ChemPort |
11. Krishnaraju K, Hoffman B, Liebermann DA. (2001). Early growth response gene 1 stimulates development of hematopoietic progenitor cells along the macrophage lineage at the expense of the granulocyte and erythroid lineages. Blood 97: 1298–1305. | Article | PubMed | ISI | ChemPort |
12. Krishnaraju K, Nguyen HQ, Liebermann DA, Hoffman B. (1995). The zinc finger transcription factor Egr-1 potentiates macrophage differentiation of hematopoietic cells. Mol Cell Biol 15: 5499–5507. | PubMed | ISI | ChemPort |
13. Laslo P, Spooner CJ, Warmflash A, Lancki DW, Lee HJ, Sciammas R et al. (2006). Multilineage transcriptional priming and determination of alternate hematopoietic cell fates. Cell 126: 755–766. | Article | PubMed | ISI | ChemPort |
14. Liu C, Yao J, de Belle I, Huang RP, Adamson E, Mercola D. (1999). The transcription factor Egr-1 suppresses transformation of human fibrosarcoma HT1080 cells by coordinated induction of transforming growth factor-beta 1, fibronectin, and plasminogen activator inhibitor-1. J Biol Chem 274: 4400–4411. | Article | PubMed | ISI | ChemPort |
15. McMahon SB, Monroe JG. (1996). The role of early growth response gene1 (egr-1) in regulation of the immune response. J Leukoc Biol 60: 159–166. | PubMed | ISI | ChemPort |
16. Nguyen HQ, Hoffman-Liebermann B, Liebermann DA. (1993). The zinc finger transcription factor Egr-1 is essential for and restricts differentiation along the macrophage lineage. Cell 72: 197–209. | Article | PubMed | ISI | ChemPort |
17. Selvakumaran M, Liebermann DA, Hoffman-Liebermann B. (1992). Deregulated c-myb disrupts interleukin-6 or leukemia inhibitory factor-induced myeloid differentiation prior to c-myc: role in leukemogenesis. Mol Cell Biol 12: 2493–2500. | PubMed | ISI | ChemPort |
18. Shafarenko M, Liebermann DA, Hoffman B. (2005). Egr-1 abrogates the block imparted by c-Myc on terminal M1 myeloid differentiation. Blood 106: 871–878. | Article | PubMed | ISI | ChemPort |
19. Steinkamp JA, Wilson JS, Saunders GC, Stewart CC. (1982). Phagocytosis: flow cytometric quantitation with fluorescent microspheres. Science 215: 64–66. | Article | PubMed | ISI | ChemPort |

Acknowledgements

We thank Dr Arthur Balliet for his guidance in the FACS/Imaging and other technical insights. This work was supported by a grant from the National Institutes of Health (RO1 CA081168) to BH.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc).