SP100 expression modulates ETS1 transcriptional activity ... - Nature

3 downloads 0 Views 519KB Size Report
Jul 12, 2004 - protein SP100, Clone 15 (representing amino acids. 1–280) and Clone 18 ...... body (AC-15, ab6276) was purchased from Abcam. EMSA.
Oncogene (2004) 23, 6654–6665

& 2004 Nature Publishing Group All rights reserved 0950-9232/04 $30.00 www.nature.com/onc

SP100 expression modulates ETS1 transcriptional activity and inhibits cell invasion John S Yordy1, Runzhao Li1, Victor I Sementchenko1,6, Huiping Pei2,3, Robin C Muise-Helmericks4 and Dennis K Watson*,1,5 1

Department of Pathology and Laboratory Medicine, Medical University of South Carolina, Charleston, SC 29425, USA; Department of Medicine, Medical University of South Carolina, Charleston, SC 29425, USA; 3Hollings Cancer Center, Medical University of South Carolina, Charleston, SC 29425, USA; 4Department of Cell Biology and Anatomy, Medical University of South Carolina, Charleston, SC 29425, USA; 5Department of Biochemistry and Molecular Biology, Medical University of South Carolina, Charleston, SC 29425, USA 2

The ETS1 transcription factor is a member of the Ets family of conserved sequence-specific DNA-binding proteins. ETS1 has been shown to play important roles in various cellular processes such as proliferation, differentiation, lymphoid development, motility, invasion and angiogenesis. These diverse roles of ETS1 are likely to be dependent on specific protein interactions. To identify proteins that interact with ETS1, a yeast two-hybrid screen was conducted. Here, we describe the functional interaction between SP100 and ETS1. SP100 protein interacts with ETS1 both in vitro and in vivo. SP100 is localized to nuclear bodies and ETS1 expression alters the nuclear body morphology in living cells. SP100 negatively modulates ETS1 transcriptional activation of the MMP1 and uPA promoters in a dose-dependent manner, decreases the expression of these endogenous genes, and reduces ETS1 DNA binding. Expression of SP100 inhibits the invasion of breast cancer cells and is induced by Interferon-a, which has been shown to inhibit the invasion of cancer cells. These data demonstrate that SP100 modulates ETS1-dependent biological processes. Oncogene (2004) 23, 6654–6665. doi:10.1038/sj.onc.1207891 Published online 12 July 2004 Keywords: Ets; transcription; invasion; SP100; MMP; UPA Introduction Ets proteins constitute a family of conserved sequencespecific transcription factors and share an 85 amino-acid DNA-binding domain, the ETS domain (Graves and Petersen, 1998; Watson et al., 2002; Hsu et al., 2004). Ets factors bind to the GGAA/T core motif (Ets-binding site, EBS) in the enhancer or promoter regions of their *Correspondence: DK Watson, Hollings Cancer Center, Room 332, Medical University of South Carolina, 86 Jonathan Lucas St, Charleston, SC 29425, USA; E-mail: [email protected] 6 Current address: Affymetrix, Inc., 3380 Central Expressway, Santa Clara, CA 95051, USA Received 2 March 2004; revised 3 May 2004; accepted 11 May 2004; published online 12 July 2004

target genes (Sementchenko and Watson, 2000; Dittmer, 2003; Obika et al., 2003). ETS1 is involved in cell differentiation, proliferation, lymphoid cell development, transformation, apoptosis, angiogenesis, motility and invasion (Li et al., 1999; Kavurma et al., 2002; Watson et al., 2002; Xu et al., 2002). Further studies are required to elucidate how ETS1 participates in and is modulated by the various pathways involved in controlling these diverse biological processes. Cellular signaling and specific protein interactions control multiple ETS1 functions and may contribute to divergent ETS1 roles in various biological settings. Protein interaction(s) influence ETS1 response to cellular signals, DNA binding, target gene selection, transcriptional activity, and protein degradation (Wasylyk et al., 1997; Li et al., 2000a; Yordy and Muise-Helmericks, 2000; Verger and Duterque-Coquillaud, 2002). These protein interactions are influenced by the convergence of multiple intracellular and extracellular signals in conjunction with cell-type-specific protein expression. Consequently, the complex and often contrasting cellular functions of ETS1 may arise from intricate combinatorial controls resulting from unique protein interactions within a specific cellular context (Wasylyk et al., 1998; Yordy and Muise-Helmericks, 2000). To identify proteins that may modulate ETS1 function, we isolated proteins that interact with ETS1. In this study, we identify and functionally characterize the interaction between ETS1 and SP100. SP100 was first identified as a nuclear autoantigen in patients with the autoimmune disease Primary Biliary Cirrhosis (Szostecki et al., 1990, 1992), and its mRNA and protein is potently induced by interferon (IFN) (Guldner et al., 1992). SP100 is a constituent member of the nuclear body (NB), a subnuclear organelle, also referred to as Kr-bodies, nuclear domain 10 (ND10) or PML oncogenic domains (PODs) (Lamond and Earnshaw, 1998; Seeler et al., 1998; Seeler and Dejean, 1999). The function of SP100, as well as NBs, remains unknown, although nuclear bodies have been implicated in the development of neoplasia and the progression of viral infection (Lamond and Earnshaw, 1998; Seeler et al., 1998; Seeler and Dejean, 1999).

SP100 modulates ETS1 transactivation and cell invasion JS Yordy et al

6655

We show that SP100 negatively modulates ETS1 transcriptional activity on the MMP1 and uPA promoters and reduces the expression of endogenous MMP1 and uPA mRNA. Consistent with downregulation of MMP1 and uPA, expression of SP100 inhibits cell invasion. Thus, the functional interaction between ETS1 and SP100 directly modulates ETS1-dependent biological processes, further elucidating a biological function for SP100.

Results SP100 interacts with ETS1 in yeast, in vitro and in vivo A yeast two-hybrid interaction screen (Gyuris et al., 1993) was used to identify ETS1-interacting proteins. Of the 15 clones encoding known proteins, two independent overlapping clones coded for the nuclear autoantigen protein SP100, Clone 15 (representing amino acids 1–280) and Clone 18 (representing amino acids 10–480)

(Figure 1a). To verify the interaction, b-galactosidase activity assays were performed. Coexpression of either Clone 15 or Clone 18 with LexA-ETS1-N139 resulted in elevated b-galactosidase activity of the reporter gene. No significant activity was measured when the LexAETS1-N139 vector was coexpressed with either the empty vector or the BICOID-encoding negative control vector (Figure 1b). To confirm this interaction, GST-pulldown experiments were performed. A GST-SP100 fusion protein containing amino acids 10–480 of SP100 (GST-18470) was incubated with [35S]methionine-labeled in vitro translation proteins p42-ETS1, p51-ETS1, Fli1, Luciferase and SP100–18470. GST-SP100 interacted with p42ETS1, p51-ETS1 and Fli1 but not with Luciferase. SP100 also showed homodimerization, consistent with previous reports (Sternsdorf et al., 1999) (Figure 1c). Coimmunoprecipitation studies were performed to determine if ETS1 and SP100 are associated in cell extracts prepared from COS-1 cells transiently cotransfected with pSG5-ETS1 and pcDNA3.1-SP100

Figure 1 SP100 interacts with ETS1 in vitro and in vivo. (a) Schematic representation of the portion of Ets1 used in the yeast two hybrid and the identified SP100 clones (Clones 15 and 18 in plasmid pJG-45). (b) The EGY48 yeast strain containing the pSH18-34 bgalactosidase reporter and SP100-encoding Clone 15 or Clone 18 was transformed with the empty vector pEG202, the negative control vector pRFHM1, and the LexA-ETS1-N139 vector, respectively. Transformants were grown in Gal/CM ura his leu trp media. Galactosidase activity was measured and normalized to equivalent amounts of protein. (c) GST pull down assay. In vitro translated [35S]methionine-labeled p42-ETS1, p51-ETS1, FLI1, Luciferase and SP100-18470 were incubated with 10 ml of immobilized GST-SP100 fusion protein (GST 18470). The sepharose beads were washed and the bound proteins were separated on a 10% SDS-acrylamide gel and visualized by autoradiography (panel a2). In all, 20% of the input was analysed for comparison (panel a1). The interaction autoradiogram (panel a2) was exposed twice as long as the input autoradiogram (panel a1). (d) COS-1 cells were transiently transfected with the expression plasmids pSG5-ETS1 and pcDNA3.1-SP100. At 24 h after transfection, total cellular protein was collected and 200 mg was immunoprecipitated with a rabbit polyclonal antibody against SP100 (98159) and the captured immunoprecipitate was Western blotted for ETS1. Cell lysate and immunoprecipitation supernatant (20 mg) were also run in the Western blot for comparison. (e) MDA-MB-231 cell lysates (200 mg) were analysed for endogenous protein association between ETS1 and SP100. ETS1 monoclonal antibody E44 was used to coimmunoprecipitate ETS1 and SP100, and a monoclonal GAL4 DNA-binding domain antibody was used as a control. The immunoprecipitates were Western blotted for SP100. Total cellular lysate (20 mg) was analysed for SP100 level Oncogene

SP100 modulates ETS1 transactivation and cell invasion JS Yordy et al

6656

expression plasmids. Immunoprecipitation of SP100 followed by Western blotting for ETS1 demonstrated an interaction between ETS1 and SP100 (Figure 1d). To further characterize this SP100-ETS1 association, coimmunoprecipitation studies using cellular extracts from the breast carcinoma cell line MDA-MB-231, which expresses endogenous ETS1 and SP100, were conducted. Immunoprecipitation of ETS1 followed by Western blotting for SP100 confirmed that SP100 interacts with ETS1 (Figure 1e). The apparent differences in the level of interaction between exogenous and endogenous protein immunoprecipitations may be due to the overexpression of the transfected proteins or to the directionality of the immunoprecipitation. ETS1 alters SP100 nuclear body morphology Our data demonstrate an interaction between ETS1 and SP100. To determine if these proteins are colocalized in intact living cells, we created fusion proteins of ETS1 with YFP and SP100 with RFP to visualize the localization of these two proteins in HeLa cells using fluorescent confocal microscopy. YFP does not affect known ETS1 function, as both YFP-C1-ETS1 and YFP-N1-ETS1 fusion proteins transcriptionally activate the MMP1 promoter (data not shown) and are properly localized to the nucleus (Figure 2a and d). The RFP-N1SP100 fusion protein consistently and exclusively targeted to the NB, a hallmark of SP100 subnuclear localization (Figure 2h and k). Coexpression of SP100 with either ETS1 fusion protein (Figure 2m and n, p and q) resulted in a substantial increase in NB size and reduction of NB number (Figure 2n and q compared to Figure 2h and k). SP100 and ETS1 expression levels vary widely between different cell types and SP100 protein expression is induced by interferon To characterize endogenous ETS1 and SP100 protein expression in cell lines of different origin, we examined the levels of ETS1 and SP100 protein in multiple cell lines by Western blot analysis. Comparison of protein abundance demonstrated a wide variation of expression levels for ETS1 and SP100 across these cell lines (Figure 3a). SP100 is induced by IFN (Guldner et al., 1992). To examine SP100 induction in cells of different tissue origin and to determine if ETS1 expression is modulated by IFN, we examined SP100 and ETS1 levels in DLD-1, HeLa, SK-HEP-1, DU145, PC-3 and MDA-MB-231 cell lines grown in the absence or presence of IFN-a. In all cell lines tested, IFN-a significantly induced SP100 expression without affecting the levels of ETS1 protein expression (Figure 3b). Since DAXX also resides in the NB (Ishov et al., 1999) and interacts with ETS1 (Li et al., 2000b), we also determined if DAXX is induced by IFNa. Although it has been shown that DAXX is induced by type I interferons in lymphoid cells (Gongora et al., 2001), we did not detect any change in expression of Oncogene

Figure 2 ETS1 alters SP100 nuclear body morphology in living HeLa cells. The control vectors YFP-C1, YFP-N1 and RFP-N1, along with the corresponding ETS1 and SP100 fusion constructs YFP-C1-ETS1, YFP-N1-ETS1 and RFP-N1-SP100, were used to transiently transfect HeLa cells. YFP-ETS1 (a–f) vectors were cotransfected with RFP-N1 to serve as a control for ETS1 localization. RFP-N1-SP100 (g–l) was cotransfected with YFP control vectors to serve as a control for SP100 localization. YFPC1-ETS1 (m–o) and YFP-N1-ETS1 (p–r) were cotransfected with RFP-N1-SP100. YFP expression and RFP expression were pseudocolored blue and gray, respectively, and the first two images in each row were merged to create the third image in each row. Each panel is labeled with the indicated construct, and the scale bar represents 10 mm

DAXX in response to IFN-a treatment in the epithelial cell lines tested (Figure 3b). SP100 represses the transcriptional activity of ETS1 To assess the functional significance of the interaction between SP100 and ETS1, the effect of SP100 on ETS1dependent transcriptional activity was measured utilizing a luciferase reporter gene driven by the matrix metalloproteinase 1 (MMP1) promoter (Rutter et al., 1998). MMP1 is a well-characterized target for ETS1 (Sementchenko and Watson, 2000) and is involved in the processes of tumor invasion and metastasis (Iwasaka et al., 1996; Oda et al., 1999). This promoter contains a polymorphism that creates a second EBS that correlates

SP100 modulates ETS1 transactivation and cell invasion JS Yordy et al

6657

Figure 3 (a) Characterization of endogenous SP100 and ETS1 protein expression. The distribution of ETS1 and SP100 protein from cell lines of different tissue origins were examined by Western analyses. The cell lines used included HUVEC (human umbilical vein endothelial cells), CEM (T lymphoblast acute lymphoblastic leukemia), K562 (chronic myelogenous leukemia), H9 (cutaneous T lymphocyte lymphoma), Jos (acute T-cell leukemia), HeLa (cervical adenocarcinoma), MDA-MB-231 (breast adenocarcinoma), MCF7 (breast adenocarcinoma), SK-HEP-1 (liver adenocarcinoma), PC-3 (prostate adenocarcinoma), DU145 (prostate carcinoma) and DLD-1 (colorectal adenocarcinoma). Westerns using both monoclonal and polyclonal antibodies for b-Actin are shown as loading controls. (b) IFN-a induction of SP100 protein. The designated cell lines were maintained in the absence or presence of IFN-a for 24 h, as indicated. SP100, Daxx and ETS1 protein levels were monitored by Western analyses. b-Actin was used as a loading control

with increased tumor aggressiveness (Westermarck et al., 1997; Rutter et al., 1998). We tested three variants of the MMP1 promoter: MMP1 1G has one EBS at 87, MMP1 2G, which contains a known polymorphism, that creates a second EBS at 1606, and a truncated MMP1 promoter containing the region 517/ þ 63 that has one EBS at 87 (Figure 4a). Luciferase reporter constructs were transiently cotransfected in COS-1 cells either with constructs expressing ETS1 or SP100 alone or with constant amounts of ETS1 and increasing amounts of SP100. As expected, expression of p51-ETS1 and the p42-ETS1 isoform increased the transcriptional activity of the MMP1 2G reporter 20- and 16-fold over background (Figure 4b, upper left panel and lower left panel). Cotransfection of increasing amounts of SP100 inhibited the transcriptional activity of p51-ETS1 and p42-ETS1 in a dose-dependent manner. Similarly, expression of SP100 reduced ETS1 transcriptional activity on the truncated MMP1 517/ þ 63 reporter (Figure 4b, upper right panel and lower right panel). Similar results were also obtained with the MMP1 1G

promoter (data not shown). These assays were also repeated in HeLa cells with similar results (data not shown). These findings demonstrate that SP100 represses ETS1 transactivation of the MMP1 promoter. To independently confirm the inhibition observed with the MMP1 promoter, we assessed the role of SP100 modulation of another ETS1-regulated promoter, uPA. Consistent with the MMP1 transient reporter data, increasing the level of SP100 protein expression reduced ETS1-mediated transcriptional activation of the uPA promoter (Figure 4c). Cooperation among Ets family members and other transcription factors has been demonstrated for a large number of enhancers or promoters containing adjacent or overlapping controlling elements. One well-characterized example is the collaboration between ETS1 and AP-1 in the transcriptional regulation of MMP expression (Wasylyk et al., 1990; Bassuk and Leiden, 1995). To determine if SP100 could affect ETS1 and AP-1 synergy on the MMP1 promoter, cotransfection studies were performed. p51-ETS1 or p42-ETS1 alone resulted in 15to 20-fold activation of the MMP1 promoter (lane 2, Figure 5a and b, respectively), while addition of AP-1 resulted in over 80- to 120-fold activation in the presence of ETS1 (lane 4, Figure 5a and b). This synergistic activation was reduced by the expression of SP100 (lanes 6–8, Figure 5a and b), although the observed plateau of this inhibition may indicate saturation in this context. To examine whether the ETS1 and SP100 expression levels following transient transfection were physiologically relevant, we compared the amounts of ETS1 and SP100 protein expressed in transfected HeLa cells to the endogenous levels found in HeLa, MDA-MB-231 and SK-HEP-1 cells. The level of ETS1 protein expression is within the range observed for endogenous proteins. The amount of SP100 found in cells transfected with the highest amount of SP100 expression vector is comparable to the endogenous SP100 levels in MDA-MB-231 and HeLa cells treated with IFN-a (Figure 5c). SP100 inhibits ETS1 DNA binding To determine whether SP100 modulates the ability of ETS1 to bind to DNA, in vitro translated p42-ETS1, p51-ETS1 and SP100 proteins (Figure 6a) were used in electrophoretic mobility shift assays (EMSA) with an oligonucleotide that contains the consensus ETS1binding sequence. p51-ETS1 and p42-ETS1 bound to this oligonucleotide, as expected, while full-length SP100 protein did not (Figure 6b). To determine if SP100 is able to modulate ETS1 DNA binding, EMSAs were performed using SP100 protein in combination with either p51-ETS1 or p42-ETS1. Addition of increasing amounts of SP100 protein inhibited ETS1 DNA binding in a dose-dependent manner (Figure 6c, lanes 4–6 and 10–12). The specificity of p51-ETS1 and p42-ETS1 binding to this oligonucleotide was demonstrated by competition analysis (Figure 6c, lanes 2 and 8) and by supershift using an ETS1 antibody (Figure 6c, lanes 1 and 7). Oncogene

SP100 modulates ETS1 transactivation and cell invasion JS Yordy et al

6658

Figure 4 SP100 negatively modulates the transcriptional activity of ETS1. (a) Schematic representation of the MMP1 promoter constructs used in this study. The polymorphism that creates a second EBS in the MMP1 2G is indicated. (b) SP100 represses ETS1dependent transcriptional activation of MMP1. The MMP1 2G or MMP1 517/ þ 63 promoter construct (0.1 mg) was cotransfected in COS-1 cells with either p51-ETS1 or p42-ETS1 (0.25 mg) alone or plus increasing amounts of SP100 (0.1, 0.25, 0.5 and 0.75 mg), as indicated. The expression of ETS1, SP100 and b-Actin protein was determined by Western blot using equivalent amounts of protein and is shown below the graphs. (c) SP100 modulates ETS1 transcriptional activity on the uPA promoter. The uPA promoter was cotransfected into COS-1 cells (0.1 mg) with p51-ETS1 (0.25 mg) alone or p51-ETS1 plus increasing amounts of SP100 (0.1, 0.25, 0.5 and 0.75 mg), as indicated

SP100 inhibits MDA-MB-231 breast cancer cell invasion ETS1 has been shown to positively contribute to cell motility and invasion (Iwasaka et al., 1996; Chen et al., 1997; Delannoy-Courdent et al., 1998; Kita et al., 2001). To determine possible biological consequences of the ETS1-SP100 interaction, we examined the effect of SP100 expression on the invasion of the breast adenocarcinoma cell line MDA-MB-231. MDA-MB-231 cells infected with an adenovirus expressing SP100 showed a 60% reduction in invasion relative to cells infected with the control adenovirus expressing GFP (Figure 7a, lane 4 vs lane 1). Similarly, an adenovirus expressing dominant negative ETS protein (ETS-DN) reduced invasion of MDA-MB-231 cells (lane 3). This inhibition is also observed in cells infected with an adenovirus expressing

Figure 5 SP100 modulates ETS1 and AP1 synergistic transcriptional activity. (a) p51-ETS1 or (b) p42-ETS1 (0.25 mg), AP1(c-fos plus c-jun; 0.25 mg of each) and SP100 (0.125, 0.25 and 0.5 mg) were cotransfected in COS-1 with the MMP1 2G promoter construct (0.1 mg), as indicated. Cells were harvested after 24 h and equivalent amounts of protein extract were used for the luciferase assay. The expression level of SP100, ETS1, c-Jun and b-Actin using equivalent amounts of protein was monitored by Western blot and is shown below the graphs. (c) Comparison of transfected and endogenous protein levels. HeLa cells were transfected under the same conditions as used for the transient reporter assays. ETS1 (0.25 mg) and SP100 (0.25 and 0.75 mg) along with the MMP1 2G promoter construct (0.1 mg) were transfected as indicated and the cells were harvested after 24 h. Endogenous protein was isolated from HeLa, MDA-MB-231 and SK-HEP-1 cells with and without 24 h induction by IFN-a. Equivalent amounts of protein were used for Western blot and the expression level of SP100, ETS1 and bActin protein was monitored Oncogene

SP100 modulates ETS1 transactivation and cell invasion JS Yordy et al

6659

Figure 6 SP100 inhibits ETS1 DNA binding. (a) Representative [35S]methionine-labeled in vitro translated p51-ETS1, p42-ETS1 and SP100 proteins used in the gel shift experiments. In all, 5 ml of the cold in vitro translation reaction was removed and 0.5 ml of ProMix was added to assess the quality of the in vitro translated proteins. (b) SP100 does not bind to an oligonucleotide that contains an ETS1consensus binding site. In vitro translated p51-ETS1, p42-ETS1 and SP100 protein, as well as control TNT, were used for EMSA to determine DNA binding. (c) DNA binding by in vitro translated p51-ETS1 (open arrowhead) and p42-ETS1 (closed arrowhead) protein is inhibited in the presence of increasing amounts of in vitro translated SP100 protein. ETS1 (1 ml) and SP100 (2, 4 and 6 ml as indicated) were coincubated for 15 min at room temperature prior to the addition of the labeled oligonucleotide. ETS1–DNA complex formation was competed by excess cold oligonucleotide and supershifted by the ETS1-C20 antibody as indicated (ss). Representative autoradiogram from four experiments using independently prepared in vitro translated proteins

Figure 7 SP100 inhibits invasion of breast cancer cells. Breast cancer MDA-MB-231 cells were infected with Ad-GFP, AdantisenseEts1, Ad-DN, Ad-SP100 and Ad-GFP þ IFN-a at an MOI of 100, which consistently yield infection efficiencies greater than 99%. (a) Quantification of cells invading through Matrigel-coated filters. Cells were plated in serum-free medium onto Matrigel invasion chambers and allowed to invade for 24 h. (b) Western blot analysis of SP100, ETS1, ETS-DN and b-Actin expression from concurrently infected cells using equivalent amounts of protein. (c) Viable cell number was monitored using the MTT assay. Readings were taken 24 h prior to plating for the invasion assay, at the time of plating and 24 h after plating, which corresponded to the end of the invasion experiment. For the invasion assays, each point was performed in triplicate and the results are representative of two independent experiments. For the viability assay, each point was averaged from 16 replicates Oncogene

SP100 modulates ETS1 transactivation and cell invasion JS Yordy et al

6660

Figure 8 IFN-a treatment and SP100 expression reduce the promoter activity and expression of endogenous ETS1 target genes. Transient reporter assays using the MMP1 2G and uPA promoters were performed with increasing concentrations of IFN-a. In total, 0.1 mg of the indicated promoter was used in each assay. Luciferase activity was assessed using equivalent amounts of protein. Both HeLa (a) and MDA-MB-231 (b) cells were used. The expression level of endogenous SP100 and b-Actin (HeLa) and SP100, ETS1 and b-Actin (MDA-MB-231) protein was monitored by Western blot using equivalent amounts of protein and are shown below the graphs. (c) Northern analysis of uPA mRNA in MDA-MB-231 and SK-HEP-1 cell lines cells that were untreated or treated with IFN-a for 24 h. Total RNA (5 mg) was used for Northern blot analysis of uPA gene expression. Ethidium bromide image of the 18S rRNA is used as a loading control. Relative expression of this autoradiogram was quantified in relation to 18S rRNA (lower panel). (d) Real-time RT–PCR was used to measure the relative gene expression of MMP1, uPA and uPAR in MDA-MB-231 cells infected with Ad-SP100 compared to Ad-GFP. Relative gene expression for each gene in the experimental Ad-SP100 and control Ad-GFP-infected cells was normalized to the expression of S26, and the relative gene expression for cells infected with the Ad-SP100 was divided by the relative gene expression for cells infected with the control Ad-GFP to determine the ratio of SP100/GFP gene expression for MMP1, uPA and uPAR, depicted as percent expression SP100/GFP

antisense Ets1 (lane 2) and cells infected with the control GFP adenovirus and treated with IFN-a (lane 5). For each infection, the expression levels of ETS1, ETS-DN and SP100 were monitored by Western blot analysis (Figure 7b). MTT assays were simultaneously performed using cells with identical treatments. None of the experimental conditions affected cell viability over the time course of the invasion experiment (Figure 7c). IFN-a reduces MMP1 and uPA promoter activity, and IFN-a and SP100 expression reduce the transcription of endogenous ETS1-responsive genes in breast cancer cells Based on the functional interaction between SP100 and ETS1, and the induction of SP100 by IFN-a, we assessed the ability of IFN-a to modulate gene expression. Transient reporter assays were performed to assess the effect of IFN-a on the activation of the MMP1 and uPA promoters. IFN-a was able to repress MMP1 (Figure 8a and b, left panels) and uPA (Figure 8a and b, right panels) promoter activities at all concentrations tested in both HeLa and MDA-MB-231 cells (Figure 8a and b, respectively). The expression level of endogenous ETS1 and/or SP100 was monitored by Western blot analysis. The increased expression of SP100 protein following Oncogene

IFN-a treatment correlated with a reduction in the activity of these promoters. Since IFN-a is able to repress the promoter activity of several ETS1-responsive genes in transient reporter assays, we next determined whether IFN-a could repress the transcription of endogenous ETS1-target genes. MDA-MB-231 and SK-HEP-1 cells were treated for 24 h without or with IFN-a and Northern blot analysis of uPA gene expression was performed, revealing a 30– 50% decrease in the endogenous uPA transcript expression (Figure 8c). To determine if SP100 expression could also reduce the expression of endogenous genes, the relative expression of MMP1, uPA and uPAR in MDA-MB-231 cells infected with adenoviruses expressing SP100 was measured by real time RT–PCR. As compared to the control GFP-expressing virus, the endogenous transcript levels of MMP1, uPA and uPAR were reduced approximately by 45, 37 and 23%, respectively (Figure 8d).

Discussion In these studies, we identify and characterize the functional interaction between ETS1 and SP100. This

SP100 modulates ETS1 transactivation and cell invasion JS Yordy et al

6661

interaction was first identified in a yeast two-hybrid screen (Figure 1) and was further confirmed by coimmunoprecipitation and colocalization experiments (Figures 1 and 2), indicating that SP100 and ETS1 physically associate in vitro and in vivo. A previous report demonstrated that SP100 interacts with ETS1 at both the amino and the carboxy terminus of ETS1 (Wasylyk et al., 2002), which contain the PNT domain and the ETS (DNA-binding) domain, respectively. Since the ETS and PNT domains are the most conserved regions in the Ets family (Ghysdael and Boureux, 1997; Graves and Petersen, 1998), we predicted that SP100 might interact with other Ets family members. Consistent with this hypothesis, at least one other Ets factor, FLI1, associates with SP100 in vitro (Figure 1). We show that the expression of ETS1 affects NB size and number (Figure 2). The NB is a complex subnuclear organelle that contains a wide variety of different proteins with divergent and sometimes opposing function, as demonstrated by the identification of the residence of both transcriptional activators and repressors (Sternsdorf et al., 1997; Hodges et al., 1998; Seeler and Dejean, 1999; Maul et al., 2000; Negorev et al., 2001), and different cellular states dramatically affect nuclear body morphology (Hodges et al., 1998; Maul et al., 2000). We hypothesize that ETS1 may interconnect independent SP100 dimeric and/or multimeric complexes, or may cause a conformational change in SP100, leading to the observed increase in size and decrease in number of the SP100-containing NBs. It has also been hypothesized that the nuclear body serves as a dynamic nuclear depot (Negorev et al., 2001), providing a mechanism to control the activity of the proteins that localize to the NB. Although ETS1 does colocalize with SP100 in the NB, the NB does not appear to provide a significant storage depot for ETS1 relative to total ETS1 protein, as both previous indirect immunofluorescence analysis (Wasylyk et al., 2002) and the studies described here show ETS1 is found throughout the entire nucleus. Further experimentation is needed to determine if the functional interaction between ETS1 and SP100 is geographically restricted to the NB. Cell invasion and metastasis has been shown to require the expression and activity of various proteolytic proteins (Cairns et al., 2003; Quaranta and Giannelli, 2003; Stamenkovic, 2003). SP100 reduces both p42ETS1 and p51-ETS1 transcriptional activity on the MMP1 and uPA promoters (Figure 4). In these studies, increasing SP100 concentration in the presence of ETS1 proportionately decreased ETS1 transcriptional activity. SP100 has been shown to interact with heterochromatin protein 1 (HP1), which functions as a transcriptional repressor (Seeler et al., 1998), and SP100 acts as repressor of transcription when tethered to DNA (Lehming et al., 1998; Seeler et al., 1998; Bloch et al., 1999). However, SP100 and ETS1 have recently been reported to coactivate the MMP3 (stromelysin) promoter as well as a heterologous promoter containing multimerized EBS (Wasylyk et al., 2002). Although no direct evidence was provided, it was proposed that ETS1 and SP100 are able to form a complex on the EBS

present on the MMP3 promoter. The MMP3 promoter contains palindromic EBS, both of which are essential for activation (Wasylyk et al., 1991; Baillat et al., 2002), while both the uPA and MMP1 promoters contain single EBS. The presence of a palindromic EBS may allow ETS1 to bind DNA in the presence of SP100. In contrast, SP100 is able to inhibit ETS1 binding to DNA containing a single EBS in the absence of ternary complex formation (Figure 6). Similarly, SP100 represses transcriptional activity of Bright by inhibiting DNA binding (Zong et al., 2000). Inhibition of ETS1 transcriptional activity by interfering with DNA binding is also similar to the previously reported inhibitory mechanism of another ETS1-interacting protein and component of the NB, DAXX (Li et al., 2000b). As SP100 has now been reported to both repress and activate ETS1 function, these apparently opposite outcomes suggest that the functional relationship between ETS1 and SP100 may depend upon promoter context, and/or other as-yet undefined factors. We show that IFN-a negatively modulates both MMP1 and uPA promoters in transient reporter assays, and that this repression was correlated with IFNmediated increases in SP100 protein expression (Figure 8). Furthermore, increased SP100 protein expression was found to reduce endogenous MMP1, uPA and uPAR gene expression (Figure 8). Consistent with a biological consequence of this SP100-ETS1 interaction, we found that expression of SP100 inhibits the invasion of MDA-MB-231 cells (Figure 7). Inhibition of migration by IFN-a was greater than that by increased SP100 expression, implying that SP100 contributes to, but does not replace, the overall biological effect of IFN-a. This model is consistent with the observation that IFNs, which strongly induce SP100 expression, reduce angiogenesis and the metastatic spread of certain tumors (Kerbel and Folkman, 2002; von Marschall et al., 2003; Wang et al., 2003). These observations suggest that SP100 may modulate late tumorigenic events such as metastasis, and as such may facilitate the antineoplastic effects of IFN-a.

Materials and methods Two-hybrid system The yeast interaction trap assay reagents and a human fetal brain cDNA library were a gift from by Dr Roger Brent (The Molecular Sciences Institute, Berkeley, CA, USA). Bait construction and library screening have been previously described (Li et al., 2000b). Briefly, the N-terminal 139 amino acids of ETS1 were used as bait and a human fetal brain cDNA library was constructed using the yeast vector pJG-45, which contains a B42 activation domain and provides the ability to directly fuse proteins to its COOH terminus. From approximately 1  106 transformants obtained in the primary screen, 23 unique clones were identified by restriction mapping of PCR products. Eight of the 23 clones were found to encode novel proteins. Two of these, EAP1 (ETS1-associated Protein1), subsequently identified as Daxx (Yang et al., 1997; Li et al., 2000b), and EAPII (Pei et al., 2003) have been previously described. All experiments were performed in the yeast Oncogene

SP100 modulates ETS1 transactivation and cell invasion JS Yordy et al

6662 reporter strain EGY48 (MATa leu2 his trp upa LEU2H plexaop6LEU2[DUASLEU2]). Isolated clones were sequenced using an ABI 373 automated sequencer (Applied Biosystems). Plasmid construction A 2.2 kb EcoRI/NotI fragment from Clone 18, which contains a fragment encoding amino acids 10–480 of the SP100 protein, was inserted into pCDNA3.1 vector (Invitrogen) to generate pcDNA3.1/1800 (SP100–18470). PCR (Pfu, Stratagene) was used to amplify a 764 base pair fragment from human T lymphocyte Jurkat cell cDNA using the forward primer AGGGCCTAGGGTaccAAGATGGCAG (Asp718 site underlined) and reverse primer TCGCAGGACTCTGTTGGCT CAGCCT. The product was digested by Asp718 and CelII and inserted into pcDNA3.1/1800 at these same sites to generate the pcDNA3.1-SP100 plasmid that allows expression of the full-length SP100 protein in mammalian cells. GST-SP100 fusion proteins were constructed as follows. GST-SP100-273 (amino acids 10–283) was generated by inserting a 820 bp FspI/ NcoI fragment from Clone 18 into pGEX-2TK2 (Li et al., 2000b), which had been filled-in after digestion with XbaI and subsequently digested with NcoI. GST-SP100-470 (amino acids 10–480 of SP100) was generated by inserting a 987 bp EspI/ XhoI fragment from Clone 18 into GST-SP100–273 (above) at EspI/SalI. For the fluorescent fusion protein construction, PCR (Pfu, Stratagene) was used to amplify the full-length coding regions of ETS1 and SP100 from pSG5-p51-ETS1 and pcDNA3.1-SP100, respectively, using primers that incorporate restriction sites as indicated for each construct to allow cloning into the respective vectors at the same restriction sites. As outlined in the table below, pYFP-C1-ETS1 was created by inserting a full-length ETS1 PCR product (containing the ETS1 stop codon) generated with primers ETS1 50 Asp718 and ETS1 30 Sma1 into pEYFP-C1. pYFP-N1-ETS1 was created by inserting a full-length ETS1 PCR product (not containing the ETS1 stop codon) using primers ETS1 50 Eco47III and ETS1 30 HindIII into pEYFP-N1. pRFP-N1-SP100 was created by inserting a full-length SP100 PCR product (not containing the SP100 stop codon) using primers SP100 50 Eco47III and SP100 30 HindIII into pDsRed-N1. Fluorescent fusion vectors were obtained from Clontech, and all constructs were verified by sequencing. Ets1 50 Asp718

TTAGGTACCATGAAGGCGGCCG TCGATCTCA Ets1 30 SmaI (STOP TTACCCGGGTCACTCGTCGGCAT codon) CTGGCT Ets1 50 Eco47III TTAAGCGCTATGAAGGCGGCCG TCGATCTC Ets1 30 HindIII (no TTCGAAGCTTCTCGTCGGCATCT stop codon) GGCTTGA SP100 50 Eco47III TTAAGCGCTATGGCAGGTGGGG GCGGCG SP100 30 HindIII (no GCAAAGCTTATCTTCTTTACCTG stop codon) ACCCTCTTC

GST pull-down assays The GST fusion proteins were purified as described in the manufacturer’s protocol (Amersham Pharmacia Biotech). 35Slabeled proteins were obtained by in vitro translation in the presence of complete amino-acid mix with RedivuePRO-MIX (Amersham Pharmacia Biotech) using T7 RNA Polymerase as per the manufacturer’s instructions (TNT coupled reticulocyte lysate system, Promega). GST fusion protein (2–5 mg) was bound to glutathione–agarose beads and incubated with 35Slabeled protein (5 ml), in 0.2 ml binding buffer (20 mM HEPES pH 7.6, 3 mM MgCl2, 0.1 mM EDTA, 0.1% Tween-20, 10% glycerol, 1 mM DTT, 0.2 M KCl, 2% BSA) by rocking for 2 h at 41C. The glutathione–agarose beads were washed four times in binding buffer (without BSA), resuspended in 20 ml of SDS sample buffer, and the proteins were released from the beads by boiling for 5 min. The proteins were resolved on 10% SDS– PAGE and visualized by autoradiography. Cell culture, transfections and transient luciferase assays All the cells used in this study were maintained at 371C under 5% CO2 in Dulbecco’s modified Eagle’s minimum essential medium supplemented with 10% fetal bovine serum (GibcoBRL) except for HUVEC, which were grown in EGM medium (Clonetics, Cambrex Corporation). IFN-a was purchased from Calbiochem. COS-1 and HeLa cells were used in the transient reporter experiments. Transient transfections were carried out using Fugene VI (Roche) as per the manufacturer’s directions. Transfection efficiency assays using GFP reporters under these conditions result in 70–80% transfection efficiency in these cells. For reporter assays, cells were seeded the day before transfection in 24-well plates. Equal amounts of DNA were maintained in each transfection by addition of pSG5 and pcDNA3.1 vectors to balance pSG5ETS1 and pcDNA3.1-SP100, respectively. Preparation of cell lysates and reporter assays were carried out as per the manufacturer’s protocols. Cells were harvested 24 h after transfection and luciferase activity was measured by a luciferase assay kit (Promega) using equivalent volumes of protein extract. All results were normalized to total protein concentration, because dual reporter vectors are modulated by transcription factors and are Ets responsive (Farr and Roman, 1992; Bergeron et al., 1995; Oettgen et al., 2000). All transient reporter assays were performed at least three times in triplicate. GFP and RFP fusion protein confocal microscopy HeLa cells were grown in 24-well plates (Falcon, Becton Dickinson). Cells were seeded at a density of 50 000 cells/well the night before transfection. The next day, 0.5 mg of each plasmid encoding both fusion protein(s) or one fusion protein and the control plasmid were transfected as indicated. At 24 h after transfection, the living cells were observed using an Olympus IX70 inverted microscope withanOlympus FluoViewt 300 confocal imaging system (Olympus America). Western blots, immunoprecipitations and antibodies

pSG5-p51-ETS1 and pSG5-p42-ETS1 have been previously described (Huang et al., 1997). pGL3-MMP1-2G, pGL3MMP1-1G (Rutter et al., 1998), MMP1-517/ þ 63-coll-luciferase (Schneikert et al., 1996) and pGL2-uPA (pPR99, which contains the 90 bp fragment (2446 to 2356) fused to the uPA proximal promoter (114 to þ 398) (Rorth et al., 1990; Stacey et al., 1995) were used as reporter genes in the transient transfection assays. Oncogene

For Western analyses, cells were lysed in RIPA buffer, containing protease inhibitors (Complete Protease Inhibitors, Roche). Equal amounts of total protein (20 mg) were resolved by 10% SDS–PAGE and subjected to Western blot analyses using SuperSignal West Pico (Pierce). For immunoprecipitation, proteins were extracted from transfected cells using a NP40 buffer (50 mM Tris pH 7.4, 150 mM NaCl, 1% NP-40). Endogenous proteins were isolated by three freeze–thaw cycles

SP100 modulates ETS1 transactivation and cell invasion JS Yordy et al

6663 using a sucrose buffer (20 mM Tris pH 7.4, 5% sucrose, 0.5 mM EDTA, 0.1% Triton-X-100, 15 mM NaCl, 140 mM KCl, 1 mM MgCl plus protease inhibitors) after which the fractured cells were centrifuged at 5000 g for 2 min at room temperature. The supernatant was removed and glycerol was added to 5% of the volume of the recovered supernatant, which was then centrifuged at 14 000 g for 1 min at room temperature. For immunoprecipitations, protein extracts were incubated overnight at 41C in the presence of the appropriate antibody. Protein A or protein G antibody complexes were washed three times with the sucrose buffer plus 5% glycerol. The immunoprecipitated protein complexes were boiled, resolved by 10% SDS–PAGE and subjected to Western blot analyses as above. Anti-SP100 rabbit polyclonal antibodies (98159 and 98160) were generated against amino acids 10–480 of SP100, prepared from GST-SP100-470 after thrombin cleavage. The E44 ETS1 monoclonal antibody and the anti-DAXX rabbit polyclonal antibodies have been previously described (Bhat et al., 1994; Li et al., 2000b). Polyclonal antibodies against ETS1 (C-20, sc-350), ETS2 (C-20, sc-351), FLI1 (C-19, sc-356), c-Jun (H-79, sc-1694) and b-Actin (C-11, sc-1615) and the monoclonal antibody GAL4-DBD (RK5C1, sc-510) were purchased from Santa Cruz Biotechnology. ETS1 monoclonal antibody (610362) was purchased from BD Transduction Laboratories, Becton Dickinson. b-Actin monoclonal antibody (AC-15, ab6276) was purchased from Abcam. EMSA Oligonucleotides containing the ETS1-consensus sequence GATCTCGAGCCGGAAGTTCGA (Fisher et al., 1991) and its complement were synthesized, annealed, gel purified and labeled with [a-32P]dCTP using the Exo () Klenow enzyme (Stratagene, La Jolla, CA, USA). For each EMSA, 5–7 ml of in vitro translated protein(s) (TNT Quick coupled reticulocyte lysate system, Promega) were incubated for 15 min at room temperature (with the addition of unlabeled oligonucleotide for the competition analysis) prior to the addition of labeled duplex oligonucleotide (10–20 fmol) in 25 ml 1.2  gel shift buffer. Specifically, 6 ml of 5  reaction buffer (100 mM HEPES pH 7.9, 50% glycerol, 2.5 mM dithiothreitol, 1 mM EDTA pH 7.5, 5 mM MgCl2) was added to 1.5 mg of poly(dIdC), 0.75 ml 1 M KCl and 10–20 fmol labeled oligonucleotide, and the volume was brought to 25 ml with water. Upon addition of the oligonucleotide to the preincubated in vitro translated proteins, the reactions were incubated at room temperature for an additional 20 min (antibody was added for the last 10 min for the supershift analysis). Reaction products were separated by nondenaturing 5% polyacrylamide gel electrophoresis for 3 h in 0.25  Tris-borate-EDTA (TBE) running buffer at 200 V and visualized by autoradiography. Recombinant adenoviral constructs The pAdTrack-CMV adenoviral vectors were constructed and replication-deficient adenoviruses were made as previously described (He et al., 1998). The pAdTrack-CMV vector was a gift from Dr Bert Vogelstein (The Johns Hopkins Oncology Center). pAdTrack-CMV contains bicistronic cytomegalovirus (CMV) promoters and expresses green fluorescent protein (GFP) under the control of one of the CMV promoters. AdGFP, which expresses only GFP, was used as control virus. All other adenoviruses expressed the gene of interest and GFP. Ad-ETS1 has been previously described (Czuwara-Ladykowska et al., 2002). Ad-DN (which contains the DNA-binding domain of ETS2) was constructed from the pSGneoSK-DNEts2 vector described previously (Sementchenko et al., 1998).

Specifically, the Ets2 fragment was amplified with Pfu polymerase using the pSGneoKS-DN-Ets2 plasmid as template and using 50 primer GAGAGGATCCATGAAT GATCCCGGAATCAAGAAT (with BamHI restriction site) and 30 primer GAGACTCGAGGCGACCTCAGTCCTCCG TGTC (with a XhoI restriction site). This PCR product was then digested with the respective enzymes and cloned into BamHI–XhoI sites of the modified pcDNA3 vector (Foos et al., 1998), containing a FLAG sequence (provided by Dr Craig Hauser, Burnham Institute). The Ets2 DN sequence was verified by automated sequence analysis. The HindIII–XhoI fragment was then cloned into the HindIII–XhoI sites of pAdTrack shuttle vector. Ad-antisense ETS1 was created by inserting an inverted full-length ETS1 cDNA into pAdTrack-CMV. Ad-SP100 was constructed by inserting the full-length SP100 cDNA obtained from pcDNA3.1-SP100 by a HindIII/XbaI double digestion into pAdTrack-CMV, and the resulting shuttle vector (pAdTrack-CMV-SP100) was linearized by PmeI digestion and cotransformed with adenovirus backbone vector pAd-Easy-1 into recombination-competent Escherichia coli strain BJ5183. Recombinant adenovirus vectors containing SP100 coding sequence were identified by enzyme digestion and Western blotting. The recombinant viruses were prepared as high-titer stocks through propagation in 293 cells. Cell invasion and proliferation assays Cell invasion experiments were carried out using 8-mm pore size invasion chambers precoated with matrigel (Becton Dickinson). MDA-MB-231 cells (5  104) in 500 ml of serumfree medium were added to each invasion chamber. Media (750 ml) containing 10% FBS as the chemoattractant was added to the underside of the invasion chamber. Cells were allowed to invade for 24 h at 371C in the presence of 5% CO2. Cells that did not migrate were removed by wiping the top of the membrane with a cotton swab. The migrated cells were fixed and stained with Diff-Quiks as per the manufacturer’s protocol (Dade Behring). All invading cells in each invasion chamber were counted. Cell viability was carried out using MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) according to the manufacturer’s instructions (Sigma-Aldrich). Cells were assayed 24 h before the start of the invasion experiments, at the time of cell plating, and 24 h postplating, to coincide with the end of the invasion experiment. RNA isolation, Northern analysis and real-time RT–PCR Total RNA (5 mg) was isolated using Trizol (Invitrogen) and fractionated on 1.2% agarose gels containing 0.66 M formaldehyde. RNA was transferred to nylon filters (Duralon) in 0.1 M sodium phosphate (pH 6.8), UV crosslinked and hybridized in Quik-Hyb using random primed labeled uPA cDNA as probe (Stratagene). Northern quantification was performed using Kodak 1D image analysis software. Real-time PCR was conducted using a LightCycler with the LightCycler FastStart DNA Master SYBR Green I reaction mix according to the manufacturer’s directions (Roche). Total RNA was purified from MDA-MB-231 cells infected with either AdSP100 or Ad-GFP using TRIzol (Invitrogen), treated with DNase and purified a second time using RNeasys minicolumns (Qiagen). cDNA was then made from 2 mg of the purified total RNA using the SuperScriptt First-Strand Synthesis System for RT–PCR (Invitrogen) according to the manufacture’s protocol. A 1 : 10 dilution of the resulting Oncogene

SP100 modulates ETS1 transactivation and cell invasion JS Yordy et al

6664 cDNA was used for all real-time RT–PCR experiments. The primers used for real-time PCR are as follows: Primer

Primer sequence

MMP1 forward MMP1 reverse uPA forward uPA reverse uPAR forward uPAR reverse S26 forward S26 reverse

TGGACCTGGAGGAAATCTTGC

Amplicon size 151 bases

TCCAAGAGAATGGCCGAGTTC ATTTGTGAGGCCCATGGTTG

102 bases

at 951C for 10 s, annealing at 601C for 5 s and extension at 721C for 8 s, with a single data acquisition at the end of each extension. All ramping was carried out at 201C/s. The cycling for s26 was identical to the above except that the annealing was carried out at 551C and the extension was for 25 s. Melting curve analysis was carried out following the PCR cycling, using denaturation at 951C for 10 s, annealing at 651C for 15 s and stepping 0.11C/s from 65 to 951C with continuous data acquisition. Relative expression analysis was conducted using the program LinRegPCR according to the suggested specifications (Ramakers et al., 2003).

AAACCGCTGCTCCCACATT GGCTCCAATGGTTTCCACAAC

102 bases

CGGCAGATTTTCAAGCTCCA GCGAGCGTCTTCGATGC

192 bases

CTCAGCTCCTTACATGGGC

All primers were used at a concentration of 0.5 mM. The cycling conditions MMP1, uPA and uPAR were: preincubation at 951C for 10 min, followed by 40 cycles of denaturation

Acknowledgements We thank Dr R Brent (The Molecular Sciences Institute, Berkeley, CA) for providing the yeast two-hybrid reagents, Dr R Tjian (University of California, Berkeley) for the pCMV-cfos and pCMV-c-jun plasmid, Drs J Rutter and C Brinckerhoff (Dartmouth Medical School, NH) for the pGL3-MMP12G and pGL3-MMP1-1G plasmids, and Dr Andrew CB Cato for the MMP1-517/ þ 63 luciferase plasmid. This work was supported in part by grants from the National Institutes of Health Grant P01CA78582 (DKW) and American Cancer Society Grant IRG97-151 (RL), as by the Wachovia Hollings Cancer Center scholarship and a Cato scholarship (JSY).

References Baillat D, Begue A, Stehelin D and Aumercier M. (2002). J. Biol. Chem., 277, 29386–29398. Bassuk AG and Leiden JM. (1995). Immunity, 3, 223–237. Bergeron D, Barbeau B, Leger C and Rassart E. (1995). Cell Mol. Biol. Res., 41, 155–159. Bhat NK, Romano-Spica V, Georgiou P, Chen SL, Kui PG and Suzuki H. (1994). Hybridoma, 13, 1–8. Bloch DB, Chiche JD, Orth D, de la Monte SM, Rosenzweig A and Bloch KD. (1999). Mol. Cell. Biol., 19, 4423–4430. Cairns RA, Khokha R and Hill RP. (2003). Curr. Mol. Med., 3, 659–671. Chen Z, Fisher RJ, Riggs CW, Rhim JS and Lautenberger JA. (1997). Cancer Res., 57, 2013–2019. Czuwara-Ladykowska J, Sementchenko VI, Watson DK and Trojanowska M. (2002). J. Biol. Chem., 277, 20399–20408. Delannoy-Courdent A, Mattot V, Fafeur V, Fauquette W, Pollet I, Calmels T, Vercamer C, Boilly B, Vandenbunder B and Desbiens X. (1998). J. Cell Sci., 111 (Part 11), 1521–1534. Dittmer J. (2003). Mol. Cancer, 2, 29. Farr A and Roman A. (1992). Nucleic Acids Res., 20, 920. Fisher RJ, Mavrothalassitis G, Kondoh A and Papas TS. (1991). Oncogene, 6, 2249–2254. Foos G, Garcia-Ramirez JJ, Galang CK and Hauser CA. (1998). J. Biol. Chem., 273, 18871–18880. Ghysdael J and Boureux A. (1997). The ETS Family of Transcriptional Regulators. Birkhauser Verlag Basel: Switzerland. Gongora R, Stephan RP, Zhang Z and Cooper MD. (2001). Immunity, 14, 727–737. Graves BJ and Petersen JM. (1998). Adv. Cancer Res., 75, 1–55. Guldner HH, Szostecki C, Grotzinger T and Will H. (1992). J. Immunol., 149, 4067–4073. Gyuris J, Golemis E, Chertkov H and Brent R. (1993). Cell, 75, 791–803. He TC, Zhou S, da Costa LT, Yu J, Kinzler KW and Vogelstein B. (1998). Proc. Natl. Acad. Sci. USA, 95, 2509–2514. Oncogene

Hodges M, Tissot C, Howe K, Grimwade D and Freemont PS. (1998). Am. J. Hum. Genet., 63, 297–304. Hsu T, Trojanowska M and Watson D. (2004). J. Cell. Biochem., 91, 896–903. Huang CC, Papas TS and Bhat NK. (1997). Oncogene, 15, 851–856. Ishov AM, Sotnikov AG, Negorev D, Vladimirova OV, Neff N, Kamitani T, Yeh ET, Strauss III JF and Maul GG. (1999). J. Cell Biol., 147, 221–234. Iwasaka C, Tanaka K, Abe M and Sato Y. (1996). J. Cell Physiol., 169, 522–531. Kavurma MM, Bobryshev Y and Khachigian LM. (2002). J. Biol. Chem., 277, 36244–36252. Kerbel R and Folkman J. (2002). Nat. Rev. Cancer, 2, 727–739. Kita D, Takino T, Nakada M, Takahashi T, Yamashita J and Sato H. (2001). Cancer Res., 61, 7985–7991. Lamond AI and Earnshaw WC. (1998). Science, 280, 547–553. Lehming N, Le Saux A, Schuller J and Ptashne M. (1998). Proc. Natl. Acad. Sci. USA, 95, 7322–7326. Li R, Pei H and Papas T. (1999). Proc. Natl. Acad. Sci. USA, 96, 3876–3881. Li R, Pei H and Watson DK. (2000a). Oncogene, 19, 6514–6523. Li R, Pei H, Watson DK and Papas TS. (2000b). Oncogene, 19, 745–753. Maul GG, Negorev D, Bell P and Ishov AM. (2000). J. Struct. Biol., 129, 278–287. Negorev D, Ishov AM and Maul GG. (2001). J. Cell Sci., 114, 59–68. Obika S, Reddy SY and Bruice TC. (2003). J. Mol. Biol., 331, 345–359. Oda N, Abe M and Sato Y. (1999). J. Cell Physiol., 178, 121–132. Oettgen P, Finger E, Sun Z, Akbarali Y, Thamrongsak U, Boltax J, Grall F, Dube A, Weiss A, Brown L, Quinn G, Kas K, Endress G, Kunsch C and Libermann TA. (2000). J. Biol. Chem., 275, 1216–1225.

SP100 modulates ETS1 transactivation and cell invasion JS Yordy et al

6665 Pei H, Yordy JS, Leng Q, Zhao Q, Watson DK and Li R. (2003). Oncogene, 22, 2699–2709. Quaranta V and Giannelli G. (2003). Tumori, 89, 343–348. Ramakers C, Ruijter JM, Deprez RH and Moorman AF. (2003). Neurosci. Lett., 339, 62–66. Rorth P, Nerlov C, Blasi F and Johnsen M. (1990). Nucleic Acids Res., 18, 5009–5017. Rutter JL, Mitchell TI, Buttice G, Meyers J, Gusella JF, Ozelius LJ and Brinckerhoff CE. (1998). Cancer Res., 58, 5321–5325. Schneikert J, Peterziel H, Defossez PA, Klocker H, Launoit Y and Cato AC. (1996). J. Biol. Chem., 271, 23907–23913. Seeler JS and Dejean A. (1999). Curr. Opin. Genet. Dev., 9, 362–367. Seeler JS, Marchio A, Sitterlin D, Transy C and Dejean A. (1998). Proc. Natl. Acad. Sci. USA, 95, 7316–7321. Sementchenko VI, Schweinfest CW, Papas TS and Watson DK. (1998). Oncogene, 17, 2883–2888. Sementchenko VI and Watson DK. (2000). Oncogene, 19, 6533–6548. Stacey KJ, Fowles LF, Colman MS, Ostrowski MC and Hume DA. (1995). Mol. Cell. Biol., 15, 3430–3441. Stamenkovic I. (2003). J. Pathol., 200, 448–464. Sternsdorf T, Grotzinger T, Jensen K and Will H. (1997). Immunobiology, 198, 307–331. Sternsdorf T, Jensen K, Reich B and Will H. (1999). J. Biol. Chem., 274, 12555–12566. Szostecki C, Guldner HH, Netter HJ and Will H. (1990). J. Immunol., 145, 4338–4347. Szostecki C, Will H, Netter HJ and Guldner HH. (1992). Scand J. Immunol., 36, 555–564.

Verger A and Duterque-Coquillaud M. (2002). BioEssays, 24, 362–370. von Marschall Z, Scholz A, Cramer T, Schafer G, Schirner M, Oberg K, Wiedenmann B, Hocker M and Rosewicz S. (2003). J. Natl. Cancer Inst., 95, 437–448. Wang L, Wu WZ, Sun HC, Wu XF, Qin LX, Liu YK, Liu KD and Tang ZY. (2003). J. Gastrointest. Surg., 7, 587–594. Wasylyk B, Hagman J and Gutierrez-Hartmann A. (1998). Trends Biochem. Sci., 23, 213–216. Wasylyk B, Wasylyk C, Flores P, Begue A, Leprince D and Stehelin D. (1990). Nature, 346, 191–193. Wasylyk C, Bradford AP, Gutierrez-Hartmann A and Wasylyk B. (1997). Oncogene, 14, 899–913. Wasylyk C, Gutman A, Nicholson R and Wasylyk B. (1991). EMBO J., 10, 1127–1134. Wasylyk C, Schlumberger SE, Criqui-Filipe P and Wasylyk B. (2002). Mol. Cell. Biol., 22, 2687–2702. Watson DK, Li R, Sementchenko VI, Mavrothalassitis G and Seth A. (2002). Encyclopedia of Cancer Vol. 2. Bertino JR (ed). Academic Press: New York, pp 751–758. Westermarck J, Seth A and Kahari VM. (1997). Oncogene, 14, 2651–2660. Xu D, Wilson TJ, Chan D, De Luca E, Zhou J, Hertzog PJ and Kola I. (2002). EMBO J., 21, 4081–4093. Yang X, Khosravi-Far R, Chang HY and Baltimore D. (1997). Cell, 89, 1067–1076. Yordy JS and Muise-Helmericks RC. (2000). Oncogene, 19, 6503–6513. Zong RT, Das C and Tucker PW. (2000). EMBO J., 19, 4123–4133.

Oncogene