Advertisement
Tumor and Stem Cell Biology

ETS-1 Oncogenic Activity Mediated by Transforming Growth Factor α

Chet E. Holterman, Aleksandra Franovic, Josianne Payette and Stephen Lee
DOI: 10.1158/0008-5472.CAN-09-2090 Published January 2010

Abstract

Inappropriate expression of Ets-1 is observed in a variety of human cancers, and its forced expression in cultured cells results in transformation, autonomous proliferation, and tumor formation. The basis by which Ets-1 confers autonomous growth, one of the primary hallmarks of cancer cells and a critical component of persistent proliferation, has yet to be fully explained. Using a variety of cancer cell lines, we show that inhibition of Ets-1 blocks tumor formation and cell proliferation in vivo and autonomous growth in culture. A screen of multiple diffusible growth factors revealed that inhibition of Ets-1 results in the specific downregulation of transforming growth factor α (TGFα), the proximal promoter region of which contains multiple ETS family DNA binding sites that can be directly bound and regulated by Ets-1. Notably, rescuing TGFα expression in Ets-1–silenced cells was sufficient to restore tumor cell proliferation in vivo and autonomous growth in culture. These results reveal a previously unrecognized mechanism by which Ets-1 oncogenic activity can be explained in human cancer through its ability to regulate the important cellular mitogen TGFα. Cancer Res; 70(2); 730–40

Keywords
  • Ets-1
  • oncogene
  • autonomous growth
  • TGFα

Introduction

Self-sufficiency in growth signals, or autonomous growth, plays a critical role in tumor initiation and is recognized as one of the primary hallmarks of cancer cells (1). Other hallmarks include, but are not limited to, evasion of apoptosis, increased angiogenesis, and invasiveness or metastatic potential, each of which is thought to play a role in tumor promotion (1). Cancer cells acquire these unique characteristics in several ways, including altered gene expression patterns in response to microenvironmental cues, inactivation of tumor suppressor genes, and inappropriate activation of oncogenes.

First characterized by its homology to the avian E26 viral oncogene, Ets-1 is the founding member of the ETS family of transcription factors and is a classic example of an oncogene (2). Inappropriate expression of Ets-1 is an early event in a wide variety of cancers, and its forced expression results in a transformed phenotype as shown by the ability of modified cells to undergo autonomous growth in serum-free medium and to form tumors in xenograft assays (3–5).

A large body of work has shown a role for Ets-1 in regulating invasiveness and metastatic potential in a variety of cancer cell lines through the modulation of genes responsible for extracellular matrix remodeling, migration, and invasion (5–12). These genes include matrix metalloproteinase-1 (MMP-1), MMP-3, MMP-9, and urokinase-type plasminogen activator (uPA; refs. 7, 10, 11, 13). As well, Ets-1 has been shown to promote an invasive phenotype through the modulation of integrin expression and to play a critical role in tumor angiogenesis (6, 7, 12, 14–16). Given its ability to regulate several of the later hallmarks of cancer, it is not surprising that Ets-1 expression has been correlated with advanced tumor stage and poor patient prognosis (17–21).

Although these studies have proved critical to our understanding of Ets-1 function in tumorigenesis, we have yet to gain a full understanding of the mechanisms through which this oncogene drives cellular transformation and growth autonomy in human cancer. Self-sufficiency in growth signaling is perhaps the most critical aspect of aberrant tumor cell proliferation (1). Although normal cells require specific proliferative cues to escape quiescence as shown by their requirement for exogenous growth factors, inappropriate activation of proliferative pathways is frequently observed in human cancers (22, 23). One such pathway that frequently displays constitutive activation in human cancers is the epidermal growth factor receptor (EGFR) pathway (24). Multiple mechanisms allow for EGFR activation and persistent tumor cell proliferation, including activating mutations, receptor amplification, and ligand upregulation (25–27). Accordingly, overt expression of transforming growth factor α (TGFα), a cognate ligand of the EGFR, has also been implicated in human cancers (28–30). Indeed, recent studies suggest that targeting TGFα may prove to be a viable therapeutic strategy in several types of cancer (31–35).

Based on the observation that Ets-1 confers the ability to proliferate in an autonomous manner, we hypothesized that its oncogenic potential may be attributable to the regulation of specific mitogens that are capable of driving cellular proliferation in serum-free medium. Here, we show that inhibition of Ets-1 in a panel of genetically diverse cancer cell lines blocks persistent autonomous proliferation. We show that Ets-1 directly binds to and regulates TGFα, the only mitogen that was differentially regulated by Ets-1 inhibition. Critically, reestablishing TGFα expression following Ets-1 inhibition is sufficient to restore in vivo tumor cell proliferation and autonomous growth in culture. As a whole, our results implicate Ets-1 in the regulation of an autocrine signaling pathway and provide a mechanistical explanation for Ets-1 transforming and oncogenic activity in human cancers.

Materials and Methods

Cell culture and reagents

The human sporadic von Hippel-Lindau−/− (VHL−/−) renal carcinoma cell line 786-0 and U87MG glioma cell line were obtained from the American Type Culture Collection. The WT-7 (VHL+) cell line is a kind gift from William Kaelin (Harvard University). PC3 prostate carcinoma cells were a kind gift from John Bell (University of Ottawa). Cell lines were maintained in DMEM supplemented with fetal bovine serum and antibiotics as previously described (36). Serum-free medium [insulin/transferrin/selenium (ITS)] was supplemented with 1× ITS (Invitrogen).

Constructs, short hairpin RNAs, and transfections

Ets-1 full-length cDNA and Ets-1 dominant-negative (amino acids 307–441) constructs were generated by cloning PCR fragments into the pCDNA3.1(−)neo vector (Invitrogen) as previously described (37). The TGFα rescue plasmid was generated by cloning the full-length TGFα cDNA into pCDNA3.1(−)hygro vector (Invitrogen) in a similar manner. The Ets-1 evader construct and ETS family binding site (EBS) mutant were generated using the QuikChange Site-Directed Mutagenesis kit (Clontech). Primers used to generate mutations are listed in Supplementary Table S1. All constructs were verified by standard DNA sequencing. Three unique SureSilencing short hairpin RNA (shRNA) sequence plasmids against human ETS1 were obtained from SuperArray Bioscience Corp. Ets-1 SMARTpool small interfering RNA (siRNA) was obtained from Dharmacon. Transfections were carried out using Effectene transfection reagent (Qiagen) as per the manufacturer's instructions. Stable cell lines were generated using appropriate antibiotic selection (800 μg/mL neomycin/G418 or 200 μg/mL hygromycin). Stable Ets-1 knockdown was achieved with a relative frequency of 1 in 10 clones from a minimum of 96 clones for each shRNA sequences. Clones are identified by their shRNA sequence and clone number (i.e., sh2.3 represents shRNA sequence 2 clone 3).

Western blotting, antibodies, and densitometry

Cells were washed with PBS and harvested in 4% SDS in PBS as previously described (36). Protein concentrations were determined by bicinchoninic acid protein assay, and samples were separated on denaturing polyacrylamide gels. Membranes were incubated with primary antibodies anti–β-actin (Sigma), anti–Ets-1 (Santa Cruz Biotechnology), anti–Ets-2 (Santa Cruz Biotechnology), anti-GABPα (Santa Cruz Biotechnology), anti–Elf-1 (Santa Cruz Biotechnology), anti–Elk-1 (Santa Cruz Biotechnology), anti-FLAG M2 (Sigma), anti-EGFR (LabVision), or anti–pY1173-EGFR (Santa Cruz Biotechnology) overnight at 4°C. Secondary antibodies and enhanced chemiluminescence substrate are described elsewhere (25). For densitometry, a protein dilution series of parental and experimental extracts was subjected to Western blot analysis. Film was scanned at high resolution and saved as tiff files. Images were inverted using Adobe Photoshop. A set area was selected using the Marquee tool, and the mean value was determined for each band. An empty area of the film was used to calculate background values. Background was subtracted and experimental values were divided by the reference to provide a relative change in expression.

RNA isolation and real-time PCR

RNA was isolated using TriPure reagent (Sigma). Reverse transcription-PCR was performed using the High-Capacity cDNA Reverse Transcription kit (Applied Biosystems) using random hexamer primers. Real-time PCR was performed as previously described (38). Primers are listed in Supplementary Table S1.

Cell proliferation assays

Cell proliferation assays were performed as previously described with the following changes (36). Cells were incubated for the indicated period of time and then labeled with bromodeoxyuridine (BrdUrd) for 24 h. Cells were fixed and immunostained using the BrdUrd Labeling and Detection kit I (Roche). BrdUrd-positive cells were assessed using a Zeiss Axiovert S100TV microscope.

TGFα ELISA

Medium was harvested from cells grown for 72 h on six-well dishes. Samples were processed using the Human TGF-α Quantikine ELISA kit (R&D Systems, Inc.) as per the manufacturer's instructions.

In silico promoter analysis and promoter assays

Sequence of the 2.1-kb proximal promoter region of TGFα, including nucleotides 49596837 to 49598648 (minus strand), was obtained from the National Center for Biotechnology Information (accession NT_022184.14) and examined using MatInspector Software (Genomatix). The proximal region was amplified from bacterial artificial chromosome clones obtained from TCAG Genome Resources and cloned into the pGL3Basic vector (Promega). Luciferase assays were performed using the Promega Dual-Luciferase Assay kit as per the manufacturer's instructions.

Chromatin immunoprecipitation assays

Cells were grown to confluency on 10-cm plates and processed using the EZChIP Chromatin Immunoprecipitation kit (Millipore) as per the manufacturer's instructions. RNA polymerase II was used as a positive control and rabbit IgG was used as a negative control. Primers are described in Supplementary Table S1.

Initial mass and in vivo tumor formation assays

Female CD-1 nude mice (Charles River) were used for xenograft assays. Experiments were performed in a double-blind manner as previously described (39). Initial masses were excised 1 wk following initial injection.

Spheroids

Multicellular spheroids were prepared as described by our group elsewhere (39). Spheroids were fixed in 10% formaldehyde and processed for paraffin embedding and sectioning. Sections were cut at a thickness of 10 μm and stained for human-specific Ki67 or terminal deoxynucleotidyl transferase–mediated dUTP nick end labeling (TUNEL) by University of Ottawa Pathological Services.

Results

Ets-1 is required for tumor cell proliferation in vivo

To examine the role of Ets-1 in driving persistent proliferation in cancer cells, we used three unique shRNA sequences against Ets-1 to generate stable clones of 786-0 human renal clear cell carcinoma (RCC) and U87MG glioma cells with minimal Ets-1 expression. Parental and stable lines expressing nontargeting scrambled shRNA were used as controls. Specific Ets-1 knockdown was verified by Western blot (Fig. 1A). Ets-1 knockdown resulted in statistically significant (P < 0.01) reduction in expression of uPA, a known Ets-1 target gene, as assessed by quantitative PCR (Fig. 1B). No changes in the normal proliferative capacity or general cell cycle distribution were observed, indicating that no significant alterations in cell cycle progression occurred due to the knockdown of Ets-1 (Supplementary Fig. S1). To investigate the requirement for Ets-1 in an in vivo setting, we performed xenograft assays in nude mice. We have previously observed that following xenograft injection, initial masses persist for a short time period (∼1 week) followed by rapid tumor expansion. Therefore, we excised initial masses 1 week following injection and performed immunohistochemistry for Ki67 using a human-specific Ki67 antibody. Initial masses generated from parental and nontargeting controls displayed readily observable Ki67 staining throughout the mass, whereas staining for Ki67 was almost undetectable in initial masses generated from Ets-1 knockdown cell lines (Fig. 1C). To further investigate cell viability, we performed TUNEL staining to examine apoptosis. As expected, the periphery of the initial masses did not stain for TUNEL; however, the core was highly TUNEL positive, consistent with the notion that this region is hypoxic and not conducive to cellular proliferation (Fig. 1C). Importantly, although little difference was observed in apoptosis, the overall percentage of proliferating cells (Ki67 positive) in the knockdown samples was significantly lower than in parental and control samples (Fig. 1D). Collectively, these results provide preliminary evidence supporting a critical role for Ets-1 in the persistent proliferation of a variety of tumor cells.

Figure 1.
Figure 1.

Ets-1 is required for tumor cell proliferation and tumor formation in vivo. A, Western blot analysis of ETS family members in control and Ets-1 shRNA stable cell lines. β-Actin served as a loading control. B, quantitative PCR for uPA in control and stable clones. Values normalized to β-actin and expressed as relative fold change compared with parental. Columns, mean (n = 3); bars, SE. *, P < 0.01. C, representative images of Ki67- and TUNEL-stained initial mass sections derived from control or shEts-1 stable cell lines. Magnification, ×400. D, percent Ki67-positive nuclei in initial masses. Columns, mean (n = 3); bars, SE. Magnification, ×200. Xenograft tumor volumes as a function of time following injection of 786-0 or U87MG control or stable clones. Points, mean; bars, SE.

Inhibition of Ets-1 blocks tumor formation in vivo

On extended incubation in xenograft assays, parental 786-0 cells and controls formed large tumors over a 9-week time period in all 15 injections, whereas Ets-1 knockdown cell lines failed to form significant tumor masses in 16 of 17 injections over the same time period (Fig. 1D). The remaining injection formed a minimal mass that failed to increase in volume. Similar experiments using the U87MG stable cell lines returned analogous results (Fig. 1D). Taken together, these results clearly establish a critical role for Ets-1 in in vivo tumor cell proliferation in genetically diverse cancer cell lines.

ETS is required for autonomous proliferation

To separate the role of autonomous growth from other critical events in tumor development such as angiogenesis and tissue invasion, we performed three-dimensional spheroid assays. Spheroids were formed and maintained in growth medium, harvested, and processed for Ki67 staining. Ki67 staining was readily apparent in parental and scrambled spheroid sections (Fig. 2A), whereas spheroids from knockdown cell lines displayed reduced Ki67 staining, further supporting our hypothesis (Supplementary Fig. S2).

Figure 2.
Figure 2.

Blocking Ets-1 activity inhibits autonomous growth of cancer cells in vitro. A, representative images of Ki67- and TUNEL-stained spheroid sections derived from control or shEts-1 stable cell lines. Magnification, ×400. Percent Ki67-positive nuclei. Columns, mean (n = 3); bars, SE. Magnification, ×200. B, autonomous growth of serum-starved stable clones, controls, or cells transiently transfected with Ets-1 evader, measured by 24-h BrdUrd labeling of cells following 48 h in ITS. Percentage of BrdUrd-positive cells was determined in a minimum of three independent fields in triplicate. Columns, mean; bars, SE. Magnification, ×100. C, Western blots showing endogenous Ets-1 and FLAG-tagged dominant-negative Ets-1 (*) expression in transfected cell lines and controls and quantitative PCR analyzing uPA expression in cells expressing dominant-negative Ets-1 or controls. Values normalized to β-actin and expressed as relative fold change compared with controls. Columns, mean (n = 3); bars, SE. D, autonomous growth of serum-starved cells expressing dominant-negative Ets-1 or controls measured as in B.

To further examine the role of Ets-1 in autonomous growth under in vitro culture conditions, 786-0, U87MG, and PC3 stable cell lines were examined for their ability to incorporate BrdUrd in serum-free conditions. Parental and control cells showed similar rates of BrdUrd incorporation in both normal growth medium and serum-free medium supplemented with ITS (Fig. 2B). Strikingly, we observed a dramatic reduction in BrdUrd incorporation following serum withdrawal in knockdown cell lines (Fig. 2B). The inability of these cells to engage in autonomous growth in vitro strongly implicates Ets-1 in regulating the persistent proliferation of human cancer cells. To ensure that any differences observed were not due to selection of clonal variants, we performed transient transfections with a dominant-negative Ets-1 expression vector (amino acids 307–441). Cell lines expressing dominant-negative Ets-1 as determined by Western blot (Fig. 2C) showed significant reduction (P < 0.01) in uPA expression, verifying the functionality of our construct (Fig. 2C). Consistent with our previous results, cells expressing dominant-negative Ets-1 were unable to engage in autonomous growth in serum-free conditions (Fig. 2D). To strengthen our argument that Ets-1 plays a critical role in autonomous proliferation, we designed a cDNA to evade shRNA inhibition. Transient transfection of the synthetic Ets-1 in stable cell lines rescued the ability of the cells to undergo autonomous growth in serum-free conditions (Fig. 2B; functionality verified in Fig. 4D; Supplementary Fig. S3). Taken together, these results strongly implicate ETS family members, in particular Ets-1, in regulating the autonomous growth of genetically diverse cancer cells.

Ets-1 specifically regulates TGFα by directly binding to the proximal promoter region

Given that inhibition of Ets-1 only affected proliferation in serum-free conditions, we reasoned that an inability to synthesize one or more diffusible mitogens in the absence of Ets-1 was the most likely cause for the loss of autonomous growth. Therefore, we examined the expression levels of several well-established growth factors, including amphiregulin, epiregulin, heparin-binding EGF, EGF, TGFα, platelet-derived growth factor, acidic fibroblast growth factor (FGF), basic FGF, and vascular endothelial growth factor (VEGF), in response to Ets-1 inhibition. Strikingly, TGFα was the only ligand to show a statistically significant (P < 0.01) reduction in mRNA expression in all three cell types (Fig. 3A). These results specifically link Ets-1 to the expression of TGFα, a bona fide mitogen whose overexpression is observed in a wide variety of human cancers.

Figure 3.
Figure 3.

Ets-1 specifically regulates TGFα by directly binding to the proximal promoter. A, quantitative PCR analysis of ligands in cells expressing shEts-1 or controls. Values normalized to β-actin and expressed as relative fold change compared with parental. Columns, mean (n = 3); bars, SE. #, not detected. B, model showing the presence of EBS identified through in silico analysis of the 2.1-kb TGFα proximal promoter region. Dual-luciferase assays performed in cells cotransfected with the TGFα proximal promoter series and either full-length or dominant-negative Ets-1 or empty vector as a control. Empty vector set to 1 and relative fold change in reporter activity reported. Columns, mean (n = 3); bars, SE. C, chromatin immunoprecipitation performed in 786-0 and U87MG cells. IgG was used as a negative control and RNA polymerase II was used as a positive control. Primer position is shown in B.

Next, we performed computer-assisted analysis of a 2.1-kb upstream fragment of the TGFα locus and identified multiple EBSs (Fig. 3B). To determine if this region was capable of conferring Ets-1 responsiveness and to identify specific functional regions, we performed dual-luciferase assays using the full-length proximal promoter as well as a series of promoter truncations (Fig. 3B). Cotransfection of the full 2.1-kb proximal promoter region along with Ets-1 resulted in a greater than 2-fold relative increase in reporter activity (Fig. 3B). Conversely, cotransfection with dominant-negative Ets-1 repressed reporter activity. Deletion mapping identified a region located between nucleotides −1784 and −978 that contained two consensus EBS sites and showed Ets-1 responsiveness (Fig. 3B). Significantly, disruption of either of the consensus EBS resulted in almost complete loss of Ets-1 responsiveness. These results confirm that the TGFα proximal promoter is responsive to Ets-1.

To show direct and specific interaction of Ets-1 with the TGFα promoter, we performed chromatin immunoprecipitation assays for endogenous Ets-1 as well as Elf-1 and GABPα, ETS family members previously shown to have overlapping DNA binding activity (40). Using primers specific for the endogenous TGFα promoter, we were able to show enrichment of the TGFα promoter in Ets-1 but not Elf-1 or GABPα pull-downs from both U87MG and 786-0 cell lines, verifying the direct and specific interaction of Ets-1 with the TGFα promoter (Fig. 3C).

A critical role for the TGFα/EGFR growth-stimulatory pathway in driving autonomous growth of several cancer cell lines has previously been established (32, 36, 39, 41, 42). To verify that loss of TGFα expression was a common occurrence following inhibition of Ets-1, and to further address clonal selection, we performed quantitative PCR analysis on cells transiently transfected with dominant-negative Ets-1, siRNA against Ets-1 (Fig. 4A), and stable shRNA expressing cell lines. As expected, TGFα RNA levels were reduced following Ets-1 inhibition in all cell lines examined (Fig. 4A). ELISA assays verified a corresponding decrease in soluble TGFα levels following stable silencing of Ets-1 (Fig. 4B). The reduction in soluble ligand also correlated with a decrease in phosphorylated EGFR (pEGFR) as determined by a Western blot dilution series (Fig. 4C). As further proof of the specific role of Ets-1 in regulating TGFα, we found expression levels of uPA and TGFα to be upregulated on transient expression of the synthetic Ets-1 in stable knockdown cell lines (Fig. 4D). Taken together, these results provide strong evidence linking Ets-1 oncogenic activity to its ability to regulate TGFα expression.

Figure 4.
Figure 4.

The TGFα proximal promoter is directly regulated by Ets-1. A, Western blot showing Ets-1 knockdown in 786-0 cells transiently transfected with siEts-1. Quantitative PCR analysis of TGFα expression in cells expressing dominant-negative Ets-1, siEts-1, shEts-1, or controls. Columns, mean (n = 3); bars, SE. B, standard ELISA analysis of soluble TGFα levels in 786-0 stable cell lines. Columns, mean (n = 3); bars, SE. C, Western blot analysis of total EGFR (EGFR) and pEGFR levels in cell lysates from 786-0 cell lines. β-Actin served as a loading control in A and C. Densitometric analysis of pEGFR was determined by calculating the average ratio of normalized total EGFR to normalized pEGFR. D, quantitative PCR analysis of uPA and TGFα expression levels in stable cell lines transiently transfected with Ets-1 evader cDNA plasmid.

Reexpression of TGFα restores autonomous growth

To verify that the loss of autonomous growth was due to inhibition of TGFα expression and to show that the EGFR signaling pathway remained functional following Ets-1 inhibition, cells expressing dominant-negative Ets-1 or shRNA directed against Ets-1 were grown in ITS with the addition of exogenous TGFα. Importantly, the addition of exogenous TGFα to the ITS medium rescued autonomous proliferation in all cell lines examined, verifying that the TGFα/EGFR autonomous growth pathway remains intact in the absence of Ets-1 and that Ets-1 regulation of TGFα is the main proponent of autonomous proliferation (Fig. 5A and B). To further support these data, the full-length TGFα cDNA was cloned into the pCDNA3.1(−)hygro vector and subsequently transfected into the Ets-1 shRNA cell lines. Clones were screened by quantitative PCR for TGFα expression using primer sets that recognized both endogenous and reexpressed TGFα (data not shown). Clones were also subjected to ELISA to verify restoration of soluble TGFα levels (Fig. 5C). As expected, reintroduction of TGFα into Ets-1 shRNA stable clones resulted in rescue of autonomous growth in vitro as determined by BrdUrd incorporation (Fig. 5D).

Figure 5.
Figure 5.

Restoring TGFα rescues autonomous growth of cancer cell lines in vitro. A, autonomous growth of cells expressing shRNA against Ets-1 or scrambled controls exposed to exogenous TGFα measured by 24-h BrdUrd labeling of cells following 48 h of serum starvation with or without the addition of exogenous TGFα. Percentage of BrdUrd-positive cells was determined in a minimum of three independent fields in triplicate. Columns, mean; bars, SE. Magnification, ×100. B, autonomous growth of cells transiently transfected with Ets dominant-negative or controls as in A. C, standard ELISA analysis of soluble TGFα levels in 786-0 and U87MG control and stable clones or clones reexpressing TGFα. Columns, mean (n = 3); bars, SE. D, autonomous growth of cells reexpressing TGFα cDNA as measured in A.

Rescue of TGFα expression reestablishes persistent tumor cell proliferation

As previously observed, Ki67 staining is readily apparent in parental and scrambled spheroid sections, whereas spheroids formed from Ets-1 knockdown cells display a strong reduction in Ki67 staining. Importantly, spheroids generated from TGFα rescue cell lines displayed a restoration of Ki67 staining, clearly establishing that the loss of TGFα expression following Ets-1 knockdown plays a critical role in driving the persistent proliferation of cancer cells (Fig. 6A). As further proof and to address the issue of clonal selection, we infected U87MG Ets-1 knockdown cell lines with adenovirus designed to express TGFα. Consistent with previous results, reestablishment of TGFα expression through viral infection restored autonomous proliferation as shown by enhanced Ki67 staining (Fig. 6A).

Figure 6.
Figure 6.

Restoring TGFα expression following inhibition of Ets-1 rescues early proliferation but is insufficient to rescue tumor formation. A, representative images of Ki67-stained sections derived from spheroids generated using 786-0 or U87MG Ets-1 knockdown and TGFα rescue cell lines visualized. Magnification, ×400. Percentage of Ki67-positive nuclei in spheroids determined in at least three independent sections. Columns, mean; bars, SE. Magnification, ×200. B, representative images of Ki67-stained sections derived from initial masses generated using 786-0 or U87MG Ets-1 knockdown and TGFα rescue cell lines. Magnification, ×400. Percentage of Ki67-positive nuclei in initial masses determined as in A. C, representative images of TUNEL-stained sections derived from initial masses generated using 786-0 or U87MG Ets-1 knockdown and TGFα rescue cell lines. Magnification, ×400. D, tumor xenograft volumes as a function of time following injection of 786-0 shEts-1 and rescue cell lines reexpressing TGFα. Points, mean (n = 3); bars, SE.

Because restoring TGFα expression in the absence of Ets-1 rescued cell proliferation in culture and in spheroids, we hypothesized that it would also drive persistent proliferation in vivo. Consistent with our in vitro results, initial masses generated from TGFα rescue cell lines showed readily observable Ki67 staining around the periphery of the tumor, whereas knockdown controls expressing vector alone failed to stain for Ki67, verifying that reestablishing TGFα expression allows cells to proliferate persistently in vivo (Fig. 6B). TUNEL staining was prevalent throughout the TGFα rescue initial masses, indicating that whereas a subset of cells were proliferating, the majority of the cells were undergoing apoptosis (Fig. 6C). Hence, rescue cell lines were only able to form small tumors (Fig. 6D). Taken as a whole, these results are the first to show that Ets-1 oncogenic activity in human cancers is achieved in part via its ability to drive cellular proliferation through the expression of the well-known cellular mitogen TGFα.

Discussion

Inappropriate expression of Ets-1 has been linked to a variety of human cancers and is known to transform cell lines, allowing them to engage in autonomous growth in culture and form tumors in vivo (3–5). The mechanism through which Ets-1 drives the autonomous proliferation of cancer cells has remained elusive. Here, we show that inhibition of Ets-1 is sufficient to block the persistent proliferation and tumor-forming ability of human cancer cell lines in vivo and their ability to engage in autonomous growth in vitro. We reveal that Ets-1 binds to and regulates the proximal promoter region of TGFα and that silencing Ets-1 results in a significant and specific reduction in endogenous TGFα levels and not other growth factors. Importantly, rescuing TGFα expression restores autonomous growth in culture and in vivo proliferation in xenograft assays, establishing a link between Ets-1 and TGFα mitogenic activity. We believe this to be the first study to provide such mechanistical insight into the oncogenic potential of Ets-1 in human cancers.

Ets-1 has been implicated in several of the later cancer hallmarks, including invasiveness, through regulation of MMPs, and angiogenesis, through regulation of VE-cadherin and VEGF receptor (6, 7, 9, 14, 43, 44). Although TGFα expression was sufficient to rescue tumor cell proliferation, we were not surprised that it could not fully restore the remaining cancer traits and ultimately tumor formation in vivo. Based on this result, it is evident that not only does Ets-1 act as an oncogene, driving the autonomous growth of tumor cells through activation of TGFα, but that it also plays a key role in other rate-limiting tumorigenic processes.

Interestingly, several reports implicate ETS family members, particularly Ets-1, in conferring hypoxia-inducible factor 2α (HIF2α) target specificity (14, 43–46). Consistently, we observed that dominant-negative Ets-1 blocks HIF2α-driven expression of TGFα in 786-0 VHL−/− RCC cells, corroborating their functional interaction (39). More importantly, we show the generality of this model in other cancer cell lines, including U87MG glioma and PC3 prostate carcinomas, and show that overexpression of Ets-1 is sufficient to activate TGFα reporter expression. Taken together, these data imply that Ets-1 regulates TGFα expression independently of, as well as in cooperation with, HIF2α.

Ets-1 can be induced in response to hypoxia, raising the possibility that tumor microenvironment may be sufficient to drive its inappropriate expression, thereby enhancing the ability of cells to engage in persistent proliferation through activation of the TGFα/EGFR axis (47, 48). Alternatively, cell proliferation in response to spurious Ets-1 expression may be sufficient to generate a hypoxic microenvironment, which would result in stabilization of HIF2α. Such an event would provide a double-edged sword in that HIF2α and Ets-1 could synergize to activate expression of TGFα, whereas stabilization of HIF2α would also result in upregulation of EGFR, generating a constitutively activated growth-stimulatory axis.

It has recently been reported that overlapping occupancy of consensus Ets-1 binding sites by other ETS family members, in particular Elf-1, is not uncommon (40). Competitive binding between family members seems to be a common occurrence for several housekeeping genes, whereas more specialized target genes show Ets-1–specific binding (40). Consistent with this hypothesis, we find that Ets-1, but not Elf-1, directly interacts with the TGFα promoter. Furthermore, the region that confers Ets-1 responsiveness contains two adjacent Ets-1 consensus binding sites, suggesting a possible Ets-1–Ets-1 mechanism governing regulation (49, 50). Interestingly, deletion of nucleotides −977 to −1 of the promoter enhances reporter activity, suggesting that there may be repressive elements contained within this region.

Ets-1 has long been known to harbor oncogenic activity; however, the mechanism through which Ets-1 allows tumor cells to proliferate in an uncontrolled manner has remained unclear. Herein, we show that Ets-1 is required for the persistent proliferation of several human cancer cell lines, specifically through its ability to regulate TGFα expression. Importantly, our study suggests that this is a broadly exploited pathway, further supporting the view that targeting Ets-1 transcriptional activity may prove to be an effective anticancer therapy in human malignancies.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

Grant Support: National Cancer Institute of Canada (NCIC) and Canadian Institute of Health Research (S. Lee). S. Lee is a recipient of the NCIC Harold E. Johns Award. A. Franovic holds a Terry Fox Foundation Studentship through the NCIC and an NCIC Harold E. Johns Studentship Award.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Footnotes

  • Received June 8, 2009.
  • Revision received November 2, 2009.
  • Accepted November 5, 2009.

References

  1. ↵
    1. Hanahan D,
    2. Weinberg RA
    . The hallmarks of cancer. Cell 2000;100:57–70.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Watson DK,
    2. McWilliams-Smith MJ,
    3. Nunn MF,
    4. Duesberg PH,
    5. O'Brien SJ,
    6. Papas TS
    . The ets sequence from the transforming gene of avian erythroblastosis virus, E26, has unique domains on human chromosomes 11 and 21: both loci are transcriptionally active. Proc Natl Acad Sci U S A 1985;82:7294–8.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Seth A,
    2. Papas TS
    . The c-ets-1 proto-oncogene has oncogenic activity and is positively autoregulated. Oncogene 1990;5:1761–7.
    OpenUrlPubMed
    1. Topol LZ,
    2. Tatosyan AG,
    3. Ascione R,
    4. et al
    . C-ets-1 protooncogene expression alters the growth properties of immortalized rat fibroblasts. Cancer Lett 1992;67:71–8.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Seth A,
    2. Watson DK
    . ETS transcription factors and their emerging roles in human cancer. Eur J Cancer 2005;41:2462–78.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Hahne JC,
    2. Okuducu AF,
    3. Kaminski A,
    4. Florin A,
    5. Soncin F,
    6. Wernert N
    . Ets-1 expression promotes epithelial cell transformation by inducing migration, invasion and anchorage-independent growth. Oncogene 2005;24:5384–8.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Furlan A,
    2. Vercamer C,
    3. Desbiens X,
    4. Pourtier A
    . Ets-1 triggers and orchestrates the malignant phenotype of mammary cancer cells within their matrix environment. J Cell Physiol 2008;215:782–93.
    OpenUrlCrossRefPubMed
    1. Dittmer J
    . The biology of the Ets1 proto-oncogene. Mol Cancer 2003;2:29.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Sahin A,
    2. Vercamer C,
    3. Kaminski A,
    4. et al
    . Dominant-negative inhibition of Ets 1 suppresses tumor growth, invasion and migration in rat C6 glioma cells and reveals differentially expressed Ets 1 target genes. Int J Oncol 2009;34:377–89.
    OpenUrlPubMed
  10. ↵
    1. Sahin A,
    2. Velten M,
    3. Pietsch T,
    4. et al
    . Inactivation of Ets 1 transcription factor by a specific decoy strategy reduces rat C6 glioma cell proliferation and mmp-9 expression. Int J Mol Med 2005;15:771–6.
    OpenUrlPubMed
  11. ↵
    1. Sementchenko VI,
    2. Watson DK
    . Ets target genes: past, present and future. Oncogene 2000;19:6533–48.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Kita D,
    2. Takino T,
    3. Nakada M,
    4. Takahashi T,
    5. Yamashita J,
    6. Sato H
    . Expression of dominant-negative form of Ets-1 suppresses fibronectin-stimulated cell adhesion and migration through down-regulation of integrin α5 expression in U251 glioma cell line. Cancer Res 2001;61:7985–91.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Hahne JC,
    2. Fuchs T,
    3. El Mustapha H,
    4. Okuducu AF,
    5. Bories JC,
    6. Wernert N
    . Expression pattern of matrix metalloproteinase and TIMP genes in fibroblasts derived from Ets-1 knock-out mice compared to wild-type mouse fibroblasts. Int J Mol Med 2006;18:153–9.
    OpenUrlPubMed
  14. ↵
    1. Dutta D,
    2. Ray S,
    3. Vivian JL,
    4. Paul S
    . Activation of the VEGFR1 chromatin domain: an angiogenic signal-ETS1/HIF-2α regulatory axis. J Biol Chem 2008;283:25404–13.
    OpenUrlAbstract/FREE Full Text
    1. Lefter LP,
    2. Dima S,
    3. Sunamura M,
    4. et al
    . Transcriptional silencing of ETS-1 efficiently suppresses angiogenesis of pancreatic cancer. Cancer Gene Ther 2009;16:137–48.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Pourtier-Manzanedo A,
    2. Vercamer C,
    3. Van Belle E,
    4. Mattot V,
    5. Mouquet F,
    6. Vandenbunder B
    . Expression of an Ets-1 dominant-negative mutant perturbs normal and tumor angiogenesis in a mouse ear model. Oncogene 2003;22:1795–806.
    OpenUrlCrossRefPubMed
  17. ↵
    1. Behrens P,
    2. Mathiak M,
    3. Mangold E,
    4. et al
    . Stromal expression of invasion-promoting, matrix-degrading proteases MMP-1 and -9 and the Ets 1 transcription factor in HNPCC carcinomas and sporadic colorectal cancers. Int J Cancer 2003;107:183–8.
    OpenUrlCrossRefPubMed
    1. Span PN,
    2. Manders P,
    3. Heuvel JJ,
    4. et al
    . Expression of the transcription factor Ets-1 is an independent prognostic marker for relapse-free survival in breast cancer. Oncogene 2002;21:8506–9.
    OpenUrlCrossRefPubMed
    1. Pande P,
    2. Mathur M,
    3. Shukla NK,
    4. Ralhan R
    . Ets-1: a plausible marker of invasive potential and lymph node metastasis in human oral squamous cell carcinomas. J Pathol 1999;189:40–5.
    OpenUrlCrossRefPubMed
    1. Takai N,
    2. Miyazaki T,
    3. Nishida M,
    4. Nasu K,
    5. Miyakawa I
    . c-Ets1 is a promising marker in epithelial ovarian cancer. Int J Mol Med 2002;9:287–92.
    OpenUrlPubMed
  21. ↵
    1. Behrens P,
    2. Rothe M,
    3. Wellmann A,
    4. Krischler J,
    5. Wernert N
    . The Ets-1 transcription factor is up-regulated together with MMP 1 and MMP 9 in the stroma of pre-invasive breast cancer. J Pathol 2001;194:43–50.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Brattain MG,
    2. Howell G,
    3. Sun LZ,
    4. Willson JK
    . Growth factor balance and tumor progression. Curr Opin Oncol 1994;6:77–81.
    OpenUrlPubMed
  23. ↵
    1. Sporn MB,
    2. Roberts AB
    . Autocrine growth factors and cancer. Nature 1985;313:745–7.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Salomon DS,
    2. Brandt R,
    3. Ciardiello F,
    4. Normanno N
    . Epidermal growth factor-related peptides and their receptors in human malignancies. Crit Rev Oncol Hematol 1995;19:183–232.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Franovic A,
    2. Gunaratnam L,
    3. Smith K,
    4. Robert I,
    5. Patten D,
    6. Lee S
    . Translational up-regulation of the EGFR by tumor hypoxia provides a nonmutational explanation for its overexpression in human cancer. Proc Natl Acad Sci U S A 2007;104:13092–7.
    OpenUrlAbstract/FREE Full Text
    1. Normanno N,
    2. De Luca A,
    3. Bianco C,
    4. et al
    . Epidermal growth factor receptor (EGFR) signaling in cancer. Gene 2006;366:2–16.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Bache KG,
    2. Slagsvold T,
    3. Stenmark H
    . Defective downregulation of receptor tyrosine kinases in cancer. EMBO J 2004;23:2707–12.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Jhappan C,
    2. Stahle C,
    3. Harkins RN,
    4. Fausto N,
    5. Smith GH,
    6. Merlino GT
    . TGFα overexpression in transgenic mice induces liver neoplasia and abnormal development of the mammary gland and pancreas. Cell 1990;61:1137–46.
    OpenUrlCrossRefPubMed
    1. Sandgren EP,
    2. Luetteke NC,
    3. Palmiter RD,
    4. Brinster RL,
    5. Lee DC
    . Overexpression of TGFα in transgenic mice: induction of epithelial hyperplasia, pancreatic metaplasia, and carcinoma of the breast. Cell 1990;61:1121–35.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Tang P,
    2. Steck PA,
    3. Yung WK
    . The autocrine loop of TGF-α/EGFR and brain tumors. J Neurooncol 1997;35:303–14.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Borrell-Pages M,
    2. Rojo F,
    3. Albanell J,
    4. Baselga J,
    5. Arribas J
    . TACE is required for the activation of the EGFR by TGF-α in tumors. EMBO J 2003;22:1114–24.
    OpenUrlAbstract
  32. ↵
    1. Franovic A,
    2. Robert I,
    3. Smith K,
    4. et al
    . Multiple acquired renal carcinoma tumor capabilities abolished upon silencing of ADAM17. Cancer Res 2006;66:8083–90.
    OpenUrlAbstract/FREE Full Text
    1. Kenny PA,
    2. Bissell MJ
    . Targeting TACE-dependent EGFR ligand shedding in breast cancer. J Clin Invest 2007;117:337–45.
    OpenUrlCrossRefPubMed
    1. Merchant NB,
    2. Voskresensky I,
    3. Rogers CM,
    4. et al
    . TACE/ADAM-17: a component of the epidermal growth factor receptor axis and a promising therapeutic target in colorectal cancer. Clin Cancer Res 2008;14:1182–91.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Zheng X,
    2. Jiang F,
    3. Katakowski M,
    4. et al
    . Inhibition of ADAM17 reduces hypoxia-induced brain tumor cell invasiveness. Cancer Sci 2007;98:674–84.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Gunaratnam L,
    2. Morley M,
    3. Franovic A,
    4. et al
    . Hypoxia inducible factor activates the transforming growth factor-α/epidermal growth factor receptor growth stimulatory pathway in VHL(−/−) renal cell carcinoma cells. J Biol Chem 2003;278:44966–74.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. Groulx I,
    2. Lee S
    . Oxygen-dependent ubiquitination and degradation of hypoxia-inducible factor requires nuclear-cytoplasmic trafficking of the von Hippel-Lindau tumor suppressor protein. Mol Cell Biol 2002;22:5319–36.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Holterman CE,
    2. Le Grand F,
    3. Kuang S,
    4. Seale P,
    5. Rudnicki MA
    . Megf10 regulates the progression of the satellite cell myogenic program. J Cell Biol 2007;179:911–22.
    OpenUrlAbstract/FREE Full Text
  39. ↵
    1. Smith K,
    2. Gunaratnam L,
    3. Morley M,
    4. Franovic A,
    5. Mekhail K,
    6. Lee S
    . Silencing of epidermal growth factor receptor suppresses hypoxia-inducible factor-2-driven VHL−/− renal cancer. Cancer Res 2005;65:5221–30.
    OpenUrlAbstract/FREE Full Text
  40. ↵
    1. Hollenhorst PC,
    2. Shah AA,
    3. Hopkins C,
    4. Graves BJ
    . Genome-wide analyses reveal properties of redundant and specific promoter occupancy within the ETS gene family. Genes Dev 2007;21:1882–94.
    OpenUrlAbstract/FREE Full Text
  41. ↵
    1. Sasaki T,
    2. Nakamura T,
    3. Rebhun RB,
    4. et al
    . Modification of the primary tumor microenvironment by transforming growth factor α-epidermal growth factor receptor signaling promotes metastasis in an orthotopic colon cancer model. Am J Pathol 2008;173:205–16.
    OpenUrlCrossRefPubMed
  42. ↵
    1. Awwad RA,
    2. Sergina N,
    3. Yang H,
    4. et al
    . The role of transforming growth factor α in determining growth factor independence. Cancer Res 2003;63:4731–8.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Elvert G,
    2. Kappel A,
    3. Heidenreich R,
    4. et al
    . Cooperative interaction of hypoxia-inducible factor-2α (HIF-2α) and Ets-1 in the transcriptional activation of vascular endothelial growth factor receptor-2 (Flk-1). J Biol Chem 2003;278:7520–30.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    1. Le Bras A,
    2. Lionneton F,
    3. Mattot V,
    4. et al
    . HIF-2α specifically activates the VE-cadherin promoter independently of hypoxia and in synergy with Ets-1 through two essential ETS-binding sites. Oncogene 2007;26:7480–9.
    OpenUrlCrossRefPubMed
    1. Hu CJ,
    2. Sataur A,
    3. Wang L,
    4. Chen H,
    5. Simon MC
    . The N-terminal transactivation domain confers target gene specificity of hypoxia-inducible factors HIF-1α and HIF-2α. Mol Biol Cell 2007;18:4528–42.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    1. Salnikow K,
    2. Aprelikova O,
    3. Ivanov S,
    4. et al
    . Regulation of hypoxia-inducible genes by ETS1 transcription factor. Carcinogenesis 2008;29:1493–9.
  47. ↵
    1. Mikami S,
    2. Oya M,
    3. Mizuno R,
    4. Murai M,
    5. Mukai M,
    6. Okada Y
    . Expression of Ets-1 in human clear cell renal cell carcinomas: implications for angiogenesis. Cancer Sci 2006;97:875–82.
    OpenUrlCrossRefPubMed
  48. ↵
    1. Oikawa M,
    2. Abe M,
    3. Kurosawa H,
    4. Hida W,
    5. Shirato K,
    6. Sato Y
    . Hypoxia induces transcription factor ETS-1 via the activity of hypoxia-inducible factor-1. Biochem Biophys Res Commun 2001;289:39–43.
    OpenUrlCrossRefPubMed
  49. ↵
    1. Baillat D,
    2. Leprivier G,
    3. Regnier D,
    4. et al
    . Stromelysin-1 expression is activated in vivo by Ets-1 through palindromic head-to-head Ets binding sites present in the promoter. Oncogene 2006;25:5764–76.
    OpenUrlCrossRefPubMed
  50. ↵
    1. Baillat D,
    2. Laitem C,
    3. Leprivier G,
    4. Margerin C,
    5. Aumercier M
    . Ets-1 binds cooperatively to the palindromic Ets-binding sites in the p53 promoter. Biochem Biophys Res Commun 2009;378:213–7.
    OpenUrlCrossRefPubMed
View Abstract
Back to top
Cancer Research: 70 (2)
January 2010
Volume 70, Issue 2

Sign up for alerts

View this article with LENS

Open full page PDF
Article Alerts
Email Article
Citation Tools
ETS-1 Oncogenic Activity Mediated by Transforming Growth Factor α
Chet E. Holterman, Aleksandra Franovic, Josianne Payette and Stephen Lee
Cancer Res January 15 2010 (70) (2) 730-740; DOI: 10.1158/0008-5472.CAN-09-2090
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

Advertisement

Keywords