Tumor promoting properties of the ETS protein MEF in ovarian cancer

Article metrics

Abstract

We have previously shown that MEF (myeloid ELF1-like factor, also known as ELF4) functions as a transcriptional activator of the interleukin (IL)-8, perforin, granulocyte macrophage-colony stimulating factor (GM-CSF) and IL-3 genes in hematopoietic cells. MEF is also expressed in non-hematopoietic tissues including certain ovarian cancer cells. To define the function of MEF in these cells, we examined primary human ovarian epithelial tumors and found that MEF is expressed in a significant proportion of ovarian carcinomas, and in the CAOV3 and SKOV3 ovarian cancer cell lines, but not in normal ovarian surface epithelium. Manipulating MEF levels in these cell lines altered their behavior; reducing MEF levels, using short hairpin RNA expressing vectors, significantly inhibited the proliferation of SKOV3 and CAOV3 cells in culture, and impaired the anchorage-independent growth of CAOV3 cells. Overexpression of MEF in SKOV3 cells (via retroviral transduction) significantly increased their growth rate, enhanced colony formation in soft agar and promoted tumor formation in nude mice. The oncogenic activity of MEF was further shown by the ability of MEF to transform NIH3T3 cells, and induce their tumor formation in nude mice. MEF is an important regulator of the tumorigenic properties of ovarian cancer cells and could be used a therapeutic target in ovarian cancer.

Introduction

Myeloid ELF1-like factor (MEF) (ELF4), a member of the ETS family of transcription factors, is expressed in a variety of normal and malignant hematopoietic cells. Insights into its essential role in innate immunity have come from the analysis of MEF deficient mice, which have significant defects in natural killer (NK) (and NK-T) cell development and function (Lacorazza et al., 2002; Lacorazza and Nimer, 2003). MEF can also regulate key aspects of hematopoietic stem cell behavior, controlling quiescence and the resistance to myelosuppression (Lacorazza et al., 2006). Among its target genes, MEF regulates interleukin (IL)-8 and perforin expression in hematopoietic cells (Hedvat et al., 2004) (Lacorazza et al., 2002).

MEF is expressed in several tumor types, including leukemia, lymphoma and ovarian cancer cells (our data), and some studies have suggested a role for MEF in tumorigenesis. MEF was found to collaborate in tumor formation in several murine retroviral insertional mutagenesis screens, including Cdkn2a-deficient mice, which lack p16INK4a and p19ARF (Lund et al., 2002), the EμMyc transgenic mouse model that lacks expression of Pim1 and Pim2 (Mikkers et al., 2002) and the SOX4 transgenic bone marrow transplant model (Du et al., 2005).

We have shown that the activity and expression of MEF is regulated during the cell cycle by phosphorylation and ubiquitination and by proteasomal degradation, and its activity is greatest at the G1/S boundary (Miyazaki et al., 2001; Liu et al., 2006). The goal of the present study was to understand how MEF expression affects the behavior of tumor cells. Thus, we altered MEF levels in ovarian tumor cells using RNA interference and MEF overexpression and showed that MEF overexpression stimulates the aberrant growth properties of ovarian cancer cell lines and can transform NIH3T3 cell.

MEF is expressed in human ovarian surface epithelial tumors

MEF is expressed in normal hematopoietic cells, in primary acute myeloid leukemia (AML) patient samples (Fukushima et al., 2003) and in leukemia cell lines (Miyazaki et al., 1996). It is also expressed in other tissues, including liver, placenta and ovary (Miyazaki et al., 1996; Aryee et al., 1998). To investigate MEF expression in human ovarian epithelial tumors, we stained a series of 80 pathologic tissue specimens (detailed in Figure 1 and Table 1) by immunohistochemistry using an affinity-purified rabbit anti-MEF antiserum (Miyazaki et al., 2001). The highest incidence of nuclear staining was observed in the serous (48%) and endometrioid (43%) types of ovarian carcinoma, and to a lesser extent in those with clear cell (21%) or mixed (17%) histologies. Staining of normal ovarian tissue revealed MEF protein in scattered stromal cells, but no expression in ovarian surface epithelial cells or ovarian cysts (Figure 1a). When we attempted to correlate MEF protein expression with tumor stage, grade and patient outcome (overall survival), we found no significant associations (data not shown).

Figure 1
figure1

MEF is expressed in human primary ovarian epithelial tumors. Representative photomicrographs of normal ovary (a) and ovarian carcinoma with endometrioid (b), clear cell (c) and serous (d) histology from a tissue microarray section immunohistochemically stained with anti-MEF antiserum (previously generated) (18). Staining is observed in nuclei of tumor cells. Scale bar, 25 μ M (see Supplementary Information).

Table 1 Immunohistochemical expression of MEF in ovarian surface epithelial tumors

RNA interference targeting MEF inhibits the malignant potential of ovarian cancer cells

MEF is expressed in the SKOV3 and CAOV3 ovarian cancer cell lines, which are derived from human ovarian serous adenocarcinomas. To determine whether MEF was required for the tumorigenic properties of these cells, we diminished MEF expression using RNA interference. To do this we constructed a vector (pEGFP-H1) that expresses short hairpin RNAs (shRNA) from the H1 promoter with GFP as a selectable marker, and used shRNA inserts that target either MEF or the firefly luciferase (Luc) gene (as a control), to produce the pEGFP-H1-MEF1, pEGFP-H1-MEF2 and pEGFP-H1-Luc vectors. Transfection of SKOV3 and CAOV3 cells with the combination of pEGFP-HI-MEF1+2 significantly reduced the level of MEF protein expression (Figure 2a) and the combination of two shRNA vectors on MEF levels was greater than either single shRNA alone (Figure 2b). Therefore, we combined both shRNAs for all experiments. The reduction of MEF levels achieved by these shRNA vectors significantly decreased the proliferation of both the SKOV3 and CAOV3 cells (measured by the WST-1 assay) (Figure 2c), indicating the pro-growth effects of MEF. This effect was further demonstrated using propidium iodide (PI) staining, as a decrease in the S phase fraction was observed in both cell lines transfected with MEF-directed shRNA, with a corresponding increase in the G1-phase fraction (Figure 2d). Bromodeoxyuridine (BrdU) incorporation assays showed a decrease in proliferation, in cells expressing MEF-directed shRNA; a greater effect was seen in SKOV3 cells (Figure 2e). Interestingly, knockdown of MEF in SKOV3 (Figure 2f, upper panel), but not in CAOV3 cells (Figure 2f, lower panel), induced an increase in apoptotic cells (as shown by the increase in annexin V+ and 7-AAD+ cells).

Figure 2
figure2

shRNA targeting MEF decreases the proliferation of ovarian cancer cells and reduces anchorage-independent growth and contact inhibition. (a) MEF protein levels were determined by Western blot analysis of whole-cell extracts (50 μg protein per lane) from SKOV3 or CAOV3 cells transiently transfected with either pEGFP-H1-siLuc (−) or a combination of pEGFP-H1-siMEF1 and pEGFP-H1-siMEF2 (+) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). (b) Western blot analysis of SKOV3 cells transiently transfected with mock (lane 1), pEGFP-H1-siLuc (lane 2), pEGFP-H1-siMEF1 (lane 3), pEGFP-H1-siMEF2 (lane 4) or pEGFP-H1-siMEF1+2 (lane 5). The growth curve (c) is based on the WST-1 proliferation assay, performed according to the manufacturer's instructions (Roche Applied Science, Indianapolis, IN, USA). Data points represent the mean of triplicates from one of three independent experiments. Error bars represent s.d. (d) Flow cytometric analysis of propidium iodide stained SKOV3 (upper panel) or CAOV3 (lower panel) cells 48 h after transient transfection. The percent of cells in each phase of the cell cycle is shown. The value represents the mean of triplicates±s.d. (e) BrdU incorporation assays were performed on the same cells as shown in (d). The BrdU assay was carried out by ELISA using the BrdU Cell Proliferation Assay kit according to the manufacturer's protocol (EMD Biosciences, San Diego, CA, USA). (f) Cell death was assessed by flow cytometric analysis of SKOV3 (upper panel) or CAOV3 cells (lower panel) using the Annexin-V:PE Apoptosis detection kit I, as per the manufacturer's protocol (BD Biosciences Pharmingen, San Diego, CA, USA). (g) Soft agar (upper image) and focus formation (lower image) assays were performed using CAOV3 cells (see Supplementary Information).

CAOV3 cells have tumorigenic properties including lack of contact growth inhibition and loss of anchorage dependence; they efficiently form colonies in soft agar and foci in culture. When we transfected CAOV3 cells with the MEF shRNA vectors (pEGFP-H1-MEF1+2), the cells lost their ability to form colonies in soft agar (Figure 2g, upper panel), and had minimal focus forming capability (Figure 2g, lower panel). Thus, lowering MEF levels in these tumor cells impairs not only their proliferation but also their tumorigenic properties.

MEF overexpression increases the oncogenic properties of ovarian cancer cells

SKOV3 cells normally form colonies in soft agar and foci in culture, though to a lesser extent than CAOV3 cells. To examine whether MEF overexpression enhances the malignant potential of these cells, we overexpressed MEF using the pSRα retroviral vector, pSRα-MEF (Figure 3a). Increasing MEF protein levels in SKOV3 cells (SKOV3-MEF) significantly enhanced their proliferation (Figure 3b). Cell cycle analysis showed that the increased proliferation was reflected in the significant increase in cells in S phase (23.9–33.2%) (Figure 3c). The SKOV3-MEF cells more efficiently formed colonies in a soft agar assay than did SKOV3-pSRα cells (a six-fold increase) (Figure 3d). Furthermore, when injected subcutaneously into nude mice, the SKOV3-MEF cells formed larger tumors than the SKOV3-pSRα control cells (Figure 3e). Thus, increased levels of MEF enhanced the oncogenic properties of these ovarian cancer cells.

Figure 3
figure3

MEF Overexpression in SKOV3 cells increases proliferation and colony formation in soft agar. (a) Western blot analysis of whole-cell extracts from pSRα (lane 1) or pSRαMEF (lane 2) stably transfected SKOV3 cells. (b) WST-1 proliferation assay comparing the growth of pSRα and pSRαMEF stably infected SKOV3 cells. Data points represent the mean of triplicates, in one of three independent experiments. (c) Flow cytometric analysis of PI-stained SKOV3 cells 48 h after transient transfection with either the pSRα (upper panel) or pSRα-MEF (lower panel) plasmid. (d) Soft agar assay using pSRα or pSRαMEF-infected SKOV3 cells. Colonies were counted 2 weeks after infection. The columns represent the mean of triplicates. (e) Stably infected SKOV3 cells were injected subcutaneously into nude mice (see Supplementary Information).

MEF overexpression stimulates proliferation and transforms NIH3T3 cells

We recently showed that overexpression of MEF enhances the S-phase fraction of NIH3T3 cells (Liu et al., 2006), similar to what we observed for SKOV3 cells. However, to more clearly show the role of MEF in cellular transformation, we utilized the classic NIH3T3 transformation model. NIH3T3 cells normally express low levels of MEF (Figure 4a); overexpression of MEF in NIH3T3 cells using retroviral transduction (NIH3T3-MEF) resulted in increased anchorage-independent growth in soft agar with an approximately three-fold increase in colony formation (Figure 4b). NIH3T3-MEF cells formed numerous foci, in contrast to control cells, which formed no foci (Figure 4c). Subcutaneous injection of NIH3T3-MEF cells into nude mice led to significant tumor formation in all mice, whereas no tumors were seen in mice injected with control-transduced NIH3T3 cells (Figure 4d). Taken together, these data provide compelling evidence of the oncogenic activity of MEF.

Figure 4
figure4

Overexpression of MEF transforms NIH3T3 cells. (a) Western blot analysis of NIH3T3 cells infected with either pSRα (lane 1) or pSRα-MEF (lane 2). (b) NIH3T3 cells were plated in soft agar or (c) on 60 mm dishes (pSRαupper plate; pSRα-MEF, lower plate) and incubated for 2 weeks. (d) Tumor formation in nude mice injected with MEF-infected NIH3T3 cells. The columns represent the mean tumor size, measured in three mice. Error bars represent the s.d. (see Supplementary Information).

There are many examples of ETS factor overexpression in human tumors including carcinomas of the ovary, breast, colon, lung, thyroid and most recently prostate (reviewed in Seth and Watson, 2005). It was recently discovered that the ERG gene or the ETV1 gene is fused to the 5′ untranslated region of the TMPRSS2 gene in the majority of prostate cancers studied, resulting in overexpression of one of these ETS genes (Tomlins et al., 2005). The closely related ETS family member, ELF1, is expressed in ovarian cancer, and its protein expression correlates with clinical stage and a poor prognosis (Takai et al., 2003).

In some cases, ETS factors have a negative effect on growth and may act as tumor suppressors. MEF is located on Xq26 and has been posited to be a tumor suppressor gene in ovarian cancer, as MEF is contained in a minimal region of allelic loss in loss of heterozygosity studies of advanced stage ovarian cancer (Choi et al., 1997). There is a single report that MEF can act as negative regulator of the A549 lung cancer cell line, as its overexpression inhibited the ability of A549 cells to grow in soft agar and form tumors in nude mice (Seki et al., 2002). However, there is no direct evidence of MEF as a tumor suppressor as no mutations have been found in human tumors and MEF deficient mice do not spontaneously form tumors, even at an advanced age (Lacorazza et al., 2002).

How do the functions of MEF that we have described fit into existing models of ovarian tumorigenesis? The ability of MEF to enhance proliferation and promote the anchorage independence of ovarian tumor cells may cooperate with other oncogenes such as overexpressed AKT2 or mutant p53, which are also found in ovarian carcinomas (Yuan et al., 2000). The mechanism by which MEF positively affects the cell cycle is a focus of continuing studies. The physical interaction of MEF with the cell cycle regulatory cyclin A–CDK2 complex (Miyazaki et al., 2001) may contribute to these effects as could direct effects on the transcription of MEF target genes. Using Affymetrix oligonucleotide microarrays, we have identified several genes whose expression is reproducibly decreased in response to siMEF including CYCLIN D2, a D-type cyclin, which is expressed in some ovarian cancers (Milde-Langosch et al., 2003) and is known to play a role in G1–S progression. Thus, MEF has oncogenic effects in several tumor models; its cell cycle promoting and antiapoptotic functions demonstrate that MEF exerts multiple effects on cancer cell behavior. We continue to study the mechanism by which MEF regulates the growth of tumor cells. The identification of additional genes and pathways that may be regulated by MEF (ELF4) may be particularly important, if more specific antitumor therapies are to be developed.

References

  1. Aryee DN, Petermann R, Kos K, Henn T, Haas OA, Kovar H . (1998). Cloning of a novel human ELF-1-related ETS transcription factor, ELFR, its characterization and chromosomal assignment relative to ELF-1. Gene 210: 71–78.

  2. Choi C, Cho S, Horikawa I, Berchuck A, Wang N, Cedrone E et al. (1997). Loss of heterozygosity at chromosome segment Xq25–26.1 in advanced human ovarian carcinomas. Genes Chromosomes Cancer 20: 234–242.

  3. Du Y, Spence SE, Jenkins NA, Copeland NG . (2005). Cooperating cancer-gene identification through oncogenic-retrovirus-induced insertional mutagenesis. Blood 106: 2498–2505.

  4. Fukushima T, Miyazaki Y, Tsushima H, Tsutsumi C, Taguchi J, Yoshida S et al. (2003). The level of MEF but not ELF-1 correlates with FAB subtype of acute myeloid leukemia and is low in good prognosis cases. Leuk Res 27: 387–392.

  5. Hedvat CV, Yao J, Sokolic RA, Nimer SD . (2004). Myeloid ELF1-like factor is a potent activator of interleukin-8 expression in hematopoietic cells. J Biol Chem 279: 6395–6400.

  6. Lacorazza HD, Miyazaki Y, Di Cristofano A, Deblasio A, Hedvat C, Zhang J et al. (2002). The ETS protein MEF plays a critical role in perforin gene expression and the development of natural killer and NK-T cells. Immunity 17: 437–449.

  7. Lacorazza HD, Nimer SD . (2003). The emerging role of the myeloid Elf-1 like transcription factor in hematopoiesis. Blood Cells Mol Dis 31: 342–350.

  8. Lacorazza HD, Yamada T, Liu Y, Miyata Y, Sivina M, Nunes J et al. (2006). The transcription factor MEF/ELF4 regulates the quiescence of primitive hematopoietic cells. Cancer Cell 9: 175–187.

  9. Liu Y, Hedvat CV, Mao S, Zhu XH, Yao J, Nguyen H et al. (2006). The ETS protein MEF is regulated by phosphorylation-dependent proteolysis via the protein-ubiquitin ligase SCFSkp2. Mol Cell Biol 26: 3114–3123.

  10. Lund AH, Turner G, Trubetskoy A, Verhoeven E, Wientjens E, Hulsman D et al. (2002). Genome-wide retroviral insertional tagging of genes involved in cancer in Cdkn2a-deficient mice. Nat Genet 32: 160–165.

  11. Mikkers H, Allen J, Knipscheer P, Romeijn L, Hart A, Vink E et al. (2002). High-throughput retroviral tagging to identify components of specific signaling pathways in cancer. Nat Genet 32: 153–159.

  12. Milde-Langosch K, Hagen M, Bamberger AM, Loning T . (2003). Expression and prognostic value of the cell-cycle regulatory proteins, Rb, p16MTS1, p21WAF1, p27KIP1, cyclin E, and cyclin D2, in ovarian cancer. Int J Gynecol Pathol 22: 168–174.

  13. Miyazaki Y, Boccuni P, Mao S, Zhang J, Erdjument-Bromage H, Tempst P et al. (2001). Cyclin A-dependent phosphorylation of the ETS-related protein, MEF, restricts its activity to the G1 phase of the cell cycle. J Biol Chem 276: 40528–40536.

  14. Miyazaki Y, Sun X, Uchida H, Zhang J, Nimer S . (1996). MEF, a novel transcription factor with an Elf-1 like DNA binding domain but distinct transcriptional activating properties. Oncogene 13: 1721–1729.

  15. Seki Y, Suico MA, Uto A, Hisatsune A, Shuto T, Isohama Y et al. (2002). The ETS transcription factor MEF is a candidate tumor suppressor gene on the X chromosome. Cancer Res 62: 6579–6586.

  16. Seth A, Watson DK . (2005). ETS transcription factors and their emerging roles in human cancer. Eur J Cancer 41: 2462–2478.

  17. Takai N, Miyazaki T, Nishida M, Nasu K, Miyakawa I . (2003). The significance of Elf-1 expression in epithelial ovarian carcinoma. Int J Mol Med 12: 349–354.

  18. Tomlins SA, Rhodes DR, Perner S, Dhanasekaran SM, Mehra R, Sun XW et al. (2005). Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer. Science 310: 644–648.

  19. Yuan ZQ, Sun M, Feldman RI, Wang G, Ma X, Jiang C et al. (2000). Frequent activation of AKT2 and induction of apoptosis by inhibition of phosphoinositide-3-OH kinase/ Akt pathway in human ovarian cancer. Oncogene 19: 2324–2330.

Download references

Acknowledgements

We thank Irina Linkov and Marina Asherov from the Immunohistochemistry core laboratory of Memorial Sloan-Kettering Cancer Center for their assistance with immunohistochemical stains. The ϕ-A-MLV and pSRαMSVtkneo plasmids were kindly provided by Owen Witte. This work was supported by the National Heart, Lung, and Blood Institute Grant K08 HL4478 (CV Hedvat), National Cancer Institute Grant K01 CA099156 (HD Lacorazza) and the National Institute of Diabetes and Digestive and Kidney Diseases Grant R01 DK052208 (SD Nimer), Ovarian Cancer PPG #CA052477 (SD Nimer) and the Produce A Cure Fund (SD Nimer).

Author information

Correspondence to S D Nimer.

Additional information

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

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Yao, J., Liu, Y., Lacorazza, H. et al. Tumor promoting properties of the ETS protein MEF in ovarian cancer. Oncogene 26, 4032–4037 (2007) doi:10.1038/sj.onc.1210170

Download citation

Keywords

  • MEF
  • ELF4
  • RNA interference
  • ovarian cancer

Further reading