Deregulation and Mislocalization of the Cytokinesis Regulator ECT2 Activate the Rho Signaling Pathways Leading to Malignant Transformation
2004; Elsevier BV; Volume: 279; Issue: 8 Linguagem: Inglês
10.1074/jbc.m306725200
ISSN1083-351X
AutoresShinichi Saito, Xiu-fen Liu, Keiju Kamijo, Razi Raziuddin, Takashi Tatsumoto, Isamu Okamoto, Xiaohong Chen, Chong-Chou Lee, Matthew V. Lorenzi, Naoya Ohara, Toru Miki,
Tópico(s)Pluripotent Stem Cells Research
ResumoThe human ECT2 protooncogene encodes a guanine nucleotide exchange factor for the Rho GTPases and regulates cytokinesis. Although the oncogenic form of ECT2 contains an N-terminal truncation, it is not clear how the structural abnormality of ECT2 causes malignant transformation. Here we show that both the removal of the negative regulatory domain and alteration of subcellular localization are required to induce the oncogenic activity of ECT2. The transforming activity of oncogenic ECT2 was strongly inhibited by dominant negative Rho GTPases, suggesting the involvement of Rho GTPases in ECT2 transformation. Although deletion of the N-terminal cell cycle regulator-related domain (N) of ECT2 did not activate its transforming activity, removal of the small central domain (S), which contains two nuclear localization signals (NLSs), significantly induced the activity. The ECT2 N domain interacted with the catalytic domain and significantly inhibited the focus formation by oncogenic ECT2. Interestingly, the introduction of the NLS mutations in the S domain of N-terminally truncated ECT2 dramatically induced the transforming activity of this otherwise non-oncogenic derivative. Among the known Rho GTPases expressed in NIH 3T3 cells, RhoA was predominantly activated by oncogenic ECT2 in vivo. Therefore, the mislocalization of structurally altered ECT2 might cause the untimely activation of cytoplasmic Rho GTPases leading to the malignant transformation. The human ECT2 protooncogene encodes a guanine nucleotide exchange factor for the Rho GTPases and regulates cytokinesis. Although the oncogenic form of ECT2 contains an N-terminal truncation, it is not clear how the structural abnormality of ECT2 causes malignant transformation. Here we show that both the removal of the negative regulatory domain and alteration of subcellular localization are required to induce the oncogenic activity of ECT2. The transforming activity of oncogenic ECT2 was strongly inhibited by dominant negative Rho GTPases, suggesting the involvement of Rho GTPases in ECT2 transformation. Although deletion of the N-terminal cell cycle regulator-related domain (N) of ECT2 did not activate its transforming activity, removal of the small central domain (S), which contains two nuclear localization signals (NLSs), significantly induced the activity. The ECT2 N domain interacted with the catalytic domain and significantly inhibited the focus formation by oncogenic ECT2. Interestingly, the introduction of the NLS mutations in the S domain of N-terminally truncated ECT2 dramatically induced the transforming activity of this otherwise non-oncogenic derivative. Among the known Rho GTPases expressed in NIH 3T3 cells, RhoA was predominantly activated by oncogenic ECT2 in vivo. Therefore, the mislocalization of structurally altered ECT2 might cause the untimely activation of cytoplasmic Rho GTPases leading to the malignant transformation. The ECT2 oncogene has been isolated in a search for mitogenic signal transducers in epithelial cells, where a murine keratinocyte expression cDNA library was introduced into fibroblasts to induce foci of morphologically transformed cells (1Miki T. Fleming T.P. Bottaro D.P. Rubin J.S. Ron D. Aaronson S.A. Science. 1991; 251: 72-75Crossref PubMed Scopus (362) Google Scholar). The ECT2 transfectants exhibit anchorage-independent cell growth and efficient tumor formation in nude mice. The transforming ECT2 cDNA encodes the C-terminal half of the full-length protein containing Dbl-homology (DH) 1The abbreviations used are: DH, Dbl homology; AP-1, activator protein-1; BRCT, BRCA1 C-terminal; CLB6, cyclin B6; Cut5, cells untimely torn 5; DAPI, 4′, 6-diamidino-2′-phenylindole; ECT2, epithelial cell transforming gene 2 (human); ect2, mouse ECT2; Erk, extracellular signal regulated kinase; GAP, GTPase activating protein; GFP, green fluorescent protein; GST, glutathione S-transferase; JNK, c-jun N-terminal kinase; MAPK, mitogen-activated protein kinase; NLS, nuclear localizing signal; PH, pleckstrin homology; RFP, red fluorescent protein; SRE, serum response element; SRF, serum response factor; ffu, focus-forming units; DN, dominant negative; WT, wild type. 1The abbreviations used are: DH, Dbl homology; AP-1, activator protein-1; BRCT, BRCA1 C-terminal; CLB6, cyclin B6; Cut5, cells untimely torn 5; DAPI, 4′, 6-diamidino-2′-phenylindole; ECT2, epithelial cell transforming gene 2 (human); ect2, mouse ECT2; Erk, extracellular signal regulated kinase; GAP, GTPase activating protein; GFP, green fluorescent protein; GST, glutathione S-transferase; JNK, c-jun N-terminal kinase; MAPK, mitogen-activated protein kinase; NLS, nuclear localizing signal; PH, pleckstrin homology; RFP, red fluorescent protein; SRE, serum response element; SRF, serum response factor; ffu, focus-forming units; DN, dominant negative; WT, wild type. and pleckstrin homology (PH) domains, which are now found in a number of molecules involved in regulation of the Rho family GTPases. The N-terminal half of ECT2 contains domains related to cell cycle control and repair proteins, including Clb6 and Rad4/Cut5 (2Tatsumoto T. Xie X. Blumenthal R. Okamoto I. Miki T. J. Cell Biol. 1999; 147: 921-927Crossref PubMed Scopus (345) Google Scholar, 3Saito S. Tatsumoto T. Lorenzi M.V. Chedid M. Kapoor V. Sakata H. Rubin J. Miki T. J. Cell. Biochem. 2003; 90: 819-836Crossref PubMed Scopus (52) Google Scholar). CLB6 encodes a B-type cyclin of the budding yeast, which promotes the transition from G1 into S phase (4Schwob E. Nasmyth K. Genes Dev. 1993; 7: 1160-1175Crossref PubMed Scopus (406) Google Scholar). Fission yeast cut5, which is identical to the repair gene rad4, is required for both the onset of S phase and the restraint of M phase before the completion of S phase (5Saka Y. Yanagida M. Cell. 1993; 74: 383-393Abstract Full Text PDF PubMed Scopus (195) Google Scholar). The Cut5-related domain of ECT2 consists of two repeats (6Saka Y. Fantes P. Sutani T. McInerny C. Creanor J. Yanagida M. EMBO J. 1994; 13: 5319-5329Crossref PubMed Scopus (122) Google Scholar, 7Bork P. Hofmann K. Bucher P. Neuwald A.F. Altschul S.F. Koonin E.V. FASEB J. 1997; 11: 68-76Crossref PubMed Scopus (661) Google Scholar), designated BRCT (BRCA1 C-terminal) repeats, which are widespread in a number of cell-cycle checkpoint control and DNA repair proteins (7Bork P. Hofmann K. Bucher P. Neuwald A.F. Altschul S.F. Koonin E.V. FASEB J. 1997; 11: 68-76Crossref PubMed Scopus (661) Google Scholar). These cell-cycle regulator-related domains of ECT2 play essential roles on the regulation of cytokinesis (2Tatsumoto T. Xie X. Blumenthal R. Okamoto I. Miki T. J. Cell Biol. 1999; 147: 921-927Crossref PubMed Scopus (345) Google Scholar, 3Saito S. Tatsumoto T. Lorenzi M.V. Chedid M. Kapoor V. Sakata H. Rubin J. Miki T. J. Cell. Biochem. 2003; 90: 819-836Crossref PubMed Scopus (52) Google Scholar). ECT2 catalyzes guanine nucleotide exchange in vitro on three representative Rho GTPases; RhoA, Rac1, and Cdc42 (2Tatsumoto T. Xie X. Blumenthal R. Okamoto I. Miki T. J. Cell Biol. 1999; 147: 921-927Crossref PubMed Scopus (345) Google Scholar). The Rho family of small GTPases function as molecular switches of diverse biological functions, including cytoplasmic actin reorganization, cell motility, and cell scattering (8Hall A. Science. 1998; 279: 509-514Crossref PubMed Scopus (5218) Google Scholar). Activation of the Rho proteins is promoted by guanine nucleotide exchange factors, which catalyze the replacement of bound GDP by GTP. The GTP-bound form of Rho proteins can specifically interact with their effectors or targets and transmit signals to downstream molecules. Rho proteins are inactivated through the hydrolysis of bound GTP to GDP by the intrinsic GTPase activity assisted by GTPase-activating proteins (GAPs). RhoA, Rac1, and Cdc42 induce the formation of actin stress fibers, lamellipodia, and filopodia, respectively (9Olson M.F. Ashworth A. Hall A. Science. 1995; 269: 1270-1272Crossref PubMed Scopus (1058) Google Scholar). Among the known guanine nucleotide exchange factors for Rho GTPases, ECT2 shows several unique characteristics. ECT2 expression is induced in S phase and reaches the highest level in G2 and M phases in regenerating mouse liver (10Sakata H. Rubin J.S. Taylor W.G. Miki T. Hepatology. 2000; 32: 193-199Crossref PubMed Scopus (32) Google Scholar). ECT2 protein is specifically phosphorylated in G2 and M phases (2Tatsumoto T. Xie X. Blumenthal R. Okamoto I. Miki T. J. Cell Biol. 1999; 147: 921-927Crossref PubMed Scopus (345) Google Scholar). ECT2 exhibits nuclear localization in interphase, disperses throughout the cytoplasm in prometaphase, and is condensed in the midbody during cytokinesis. Expression of a dominant negative ECT2 or microinjection of anti-ECT2 antibody strongly inhibits cytokinesis, indicating that ECT2 is a critical regulator of cytokinesis (2Tatsumoto T. Xie X. Blumenthal R. Okamoto I. Miki T. J. Cell Biol. 1999; 147: 921-927Crossref PubMed Scopus (345) Google Scholar). Furthermore, the Drosophila pbl gene, whose mutation results in the inhibition of cytokinesis in mitotic cycle 14 during embryogenesis, was found to encode the fly homologue of human ECT2 (11Prokopenko S.N. Brumby A. O'Keefe L. Prior L. He Y. Saint R. Bellen H.J. Genes Dev. 1999; 13: 2301-2314Crossref PubMed Scopus (227) Google Scholar). Although the transforming activity of several DBL family oncogenes is stimulated by N-terminal alterations (12Horii Y. Beeler J.F. Sakaguchi K. Tachibana M. Miki T. EMBO J. 1994; 13: 4776-4786Crossref PubMed Scopus (185) Google Scholar, 13Chan A.M. McGovern E.S. Catalano G. Fleming T.P. Miki T. Oncogene. 1994; 9: 1057-1063PubMed Google Scholar, 14Chan A.M. Takai S. Yamada K. Miki T. Oncogene. 1996; 12: 1259-1266PubMed Google Scholar, 15Cerione R.A. Zheng Y. Curr. Opin. Cell Biol. 1996; 8: 216-222Crossref PubMed Scopus (466) Google Scholar), the activation mechanisms are still obscure. Because Rho GTPases play a critical role in cell transformation (16Prendergast G.C. Khosravi-Far R. Solski P.A. Kurzawa H. Lebowitz P.F. Der C.J. Oncogene. 1995; 10: 2289-2296PubMed Google Scholar, 17Qiu R.G. Chen J. McCormick F. Symons M. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 11781-11785Crossref PubMed Scopus (488) Google Scholar, 18Qiu R.G. Chen J. Kirn D. McCormick F. Symons M. Nature. 1995; 374: 457-459Crossref PubMed Scopus (813) Google Scholar), ECT2 may display its transforming activity through the activation of Rho proteins. However, ECT2 is predominantly expressed in the nucleus where no expression of Rho GTPases is reported. In the present study, we examined the activation mechanism of the transforming activity of ECT2. We identified the small central domain containing two tandem nuclear localization signals as a negative regulator of the transforming activity. We show that elimination of these signals and a negative regulatory domain from ECT2 resulted in the activation of Rho GTPases in the cytoplasm, leading to malignant transformation of the cells. DNA Constructs—Full-length and N-terminally truncated ECT2 cDNAs were amplified by PCR using ECT2 clone 1M (2Tatsumoto T. Xie X. Blumenthal R. Okamoto I. Miki T. J. Cell Biol. 1999; 147: 921-927Crossref PubMed Scopus (345) Google Scholar, 3Saito S. Tatsumoto T. Lorenzi M.V. Chedid M. Kapoor V. Sakata H. Rubin J. Miki T. J. Cell. Biochem. 2003; 90: 819-836Crossref PubMed Scopus (52) Google Scholar) as template and subcloned between the BamHI and EcoRI sites of the mammalian expression vector pCEV29 or its derivative pCEV29F3, which contains three tandem copies of FLAG sequence (19Lorenzi M.V. Castagnino P. Chen Q. Hori Y. Miki T. Oncogene. 1999; 18: 4742-4755Crossref PubMed Scopus (22) Google Scholar). ECT2 N-terminal derivatives, ECT2-N1 (amino acids 1-421), ECT2-N2 (amino acids 1-378), ECT2-N3 (amino acids 1-360), and ECT2-N4 (amino acids 1-333) have been described previously (3Saito S. Tatsumoto T. Lorenzi M.V. Chedid M. Kapoor V. Sakata H. Rubin J. Miki T. J. Cell. Biochem. 2003; 90: 819-836Crossref PubMed Scopus (52) Google Scholar). ECT2-ΔS mutant lacking the S domain (amino acids 329-420) was created from two PCR products using the same template and primers I-IV with the following sequences: I, 5′-CTC GGA TCC ATG GCT GAA AAT AGT GTA TTA-3′; II, 5′-CAG ACT CGC GGA GTA TTT GCC TTT TCA TA-3′; III, 5′-TCA CTC CGC GGT GGC AAG TTG CAA AAG AG-3′; and IV, 5′-ACT GAA TTC GGT AAC GCT TCA TAT CAA ATG-3′. The PCR products synthesized using primer pairs I and II were digested with BamHI and BstUI. The PCR products generated by primers III and IV were digested with BstUI and EcoRI. These products were ligated together with the pCEV29 or pCEV29F3 vector, which had been digested with BamHI and EcoRI, to create ECT2-ΔS. Two ECT2 mutants, S1 and S2, containing RRR to AAA and R to A mutations in the NLS sequence of the S domain, respectively, were generated by the similar procedure, but following oligonucleotides were used for PCR to introduce mutations: V, 5′-CAG ACT GCG GCC GCT TTG CGA TTG CTG TTA GGG GT-3′; VI, 5′-TCA CTC GCG GCC GCT TTA AAA GAA ACA CTT GCT CAG-3′; VII, 5′-TTT GGC GCG CCC GGG GTG GAA ATG GTG ACA C-3′; and VIII, 5′-TTT GGC GCG CCC ATC AGC TGA GCA TTC CCT T-3′. NotI and AscI were used to create S1 and S2, respectively, instead of BstUI. S3 was created by the similar procedure, but S1 was used as a PCR template instead of ECT2-F. ECT2-F, S1, S2, S3, ΔS, and ΔN5 were also cloned into pEGFP-C1 (BD Biosciences/Clontech) to express ECT2 as green fluorescent protein (GFP) fusion proteins. All constructs generated by the use of PCR were sequenced to ensure that no PCR mutation was generated except the desired mutations. An ECT2-ΔN5 derivative containing PVQR to AAAA mutations (amino acids 564-567) in the DH domain was generated by amplifying two PCR fragments. Primers for the first fragment were as follows: a forward primer with a BamHI restriction site, 5′-CCC GGA TCC GCC ACC ATG GTT CCT TCA AAG CAG TCA GCA-3′, and a reverse primer with SfiI site, 5′-CAG ACT GGC CGC TGC GGC CCG GAT AAG AAG TTC AAC AAG-3′. Primers for the second fragment were: a forward primer with a SfiI site, 5′-TCA CTC GGC CGC AGC GGC CTT ACC CAG TGT TGC ATT ACT-3′, and a reverse primer with an EcoRI site, 5′-ACT GAA TTC GGT AAC GCT TCA TAT CAA ATG-3′. PCR products were then digested with the indicated restriction enzymes and simultaneously ligated with the pCEV29F3 vector digested with BamHI and EcoRI. A new SfiI site was generated as a result of the introduction of PVQR to AAAA mutations. To introduce the SV40 NLS (nuc) into ECT2-ΔN5, ECT2-ΔN5 fragment was subcloned into BamHI site of pECFP-Nuc vector (BD Biosciences/Clontech), and then a DNA fragment containing the triple repeats of SV40 large T antigen NLS sequence (GAT CCA AAA AAG AAG AGA AAG GTA GAT CCA AAA AAG AAG AGA AAG GTA GAT CCA AAA AAG AAG AGA AAG GTA) and ECT2-ΔN5 sequence was excised and then cloned in pEGFP-C1 and pCEV29F3 vectors. Expression of fusion proteins of expected sizes were confirmed by Western blotting. ECT2 NLS sequence (amino acids 336-378) was attached to the 3′ end of EGFP sequence (BD Biosciences/Clontech) by PCR and subcloned between BamHI and EcoRI sites of the mammalian EGFP expression vector pCAGGFP (20Isaacs J.S. Saito S. Neckers L.M. J. Biol. Chem. 2001; 276: 18497-18506Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar) to create an EGFP-EGFP-ECT2 NLS fusion protein. The expression of 60-kDa ECT2 NLS-tagged tandem GFP protein was confirmed by Western blotting in U-2 OS cells. Focus Forming Assays—NIH 3T3 cells were transfected with various amounts (0.01-1.0 μg) of the eukaryotic expression vector pCEV29 or pCEV29F3 (19Lorenzi M.V. Castagnino P. Chen Q. Hori Y. Miki T. Oncogene. 1999; 18: 4742-4755Crossref PubMed Scopus (22) Google Scholar) containing ECT2 cDNAs or vector alone by the calcium phosphate transfection method. Focus formation was observed in unselected plates ∼14 days after transfection and quantified after Giemsa staining. FLAG-tagged ECT2 variants (in pCEV29F3) showed slightly lower transforming activity than non-tagged versions (in pCEV29). Transforming activity was expressed as the number of foci per picomoles of DNA (ffu/pmol). Comparative efficiency of transfection was confirmed by G-418-resistant colony formation in duplicated plates. Expression levels of the ECT2 variants were examined using anti-FLAG M2 antibody (Sigma, St. Louis, MO) and anti-GFP (BD Biosciences/Clontech) 48 h after transfection with 10:1 mixture of FLAG-tagged ECT2 expression vector and pEGFP-C1. Transient Expression Reporter Gene Assays—The construction of SRF-, and AP-1-luciferase reporter plasmids in pGL2Luc containing a minimal c-fos promoter (-56 to +109) has been described previously (21Tolkacheva T. Feuer B. Lorenzi M.V. Saez R. Chan A.M. Oncogene. 1997; 15: 727-735Crossref PubMed Scopus (21) Google Scholar). The SRF binding sequence is derived from the serum response element of the c-fos gene and corresponds to SRE.mutL (22Hill C.S. Wynne J. Treisman R. Cell. 1995; 81: 1159-1170Abstract Full Text PDF PubMed Scopus (1206) Google Scholar). Firefly luciferase reporter and TK-Renilla luciferase control plasmids were cotransfected with each expression vector into COS cells. Total amount of DNA was adjusted by the addition of vector DNA. 36-48 h post-transfection, cells were lysed and luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega, WI) according to the accompanied protocol. Firefly luciferase activity in the lysates was normalized to Renilla luciferase activity and expressed as a relative luciferase activity. No significant increase in luciferase activity was observed following transfection of each expression vector DNA with a reporter plasmid containing only a luciferase cassette and c-fos minimal promoter. In Vitro Invasion Assays—Invasion assays were performed using Biocoat Matrigel invasion chambers (BD Biosciences/Clontech) essentially as described in the manufacturer's protocol. Matrigel invasion chambers (24-well) were rehydrated with Dulbecco's modified Eagle's medium containing 0.1% bovine serum albumin for 2 h at room temperature. NIH 3T3 cells transfected with ECT2 or control vectors were seeded at 5 × 105/0.4 ml of medium containing 0.1% bovine serum albumin into the inner well of invasion chambers. The outer chambers were filled with 0.4 ml of medium containing 10% calf serum. Cells were allowed to invade Matrigel matrices for 10-12 h at 37 °C in a CO2 incubator. To count cells that migrated onto the membrane at lower surface, the cells on the upper side of the membrane were scraped off with cotton swipe, then the inserts with membrane were stained with Diff-Quick (Dade Diagnostics). Cells on the lower side of membrane were photographed and counted. Activation Assays for Rho GTPases—COS cells were transfected by GFP or GFP-ECT2-ΔN5 expression vectors together with pEF4/Myc-His (Invitrogen, CA) carrying inserts for AU5-tagged Rho GTPases. Lysates were prepared 24 h after transfection, and the GTP-bound forms of Rho GTPases were determined by pull-down assays using GST-Rhotekin (for RhoA, RhoB, and RhoC) or GST-PAK PBD according to the manufacturer's protocol (Cytoskeleton, Denver, CO). Subcellular Localization and Cell Morphology—NIH 3T3 cells were transfected with the GFP- or FLAG-tagged ECT2 expression vectors using the LipofectAMINE Plus reagent (Invitrogen, Carlsbad, CA). GFP-expressing cells were identified by green fluorescence. Actin and DNA were stained with rhodamine-conjugated phalloidin (Sigma) and 4′,6′-diamidino-2-phenylindole (DAPI, Sigma), respectively, as reported previously (2Tatsumoto T. Xie X. Blumenthal R. Okamoto I. Miki T. J. Cell Biol. 1999; 147: 921-927Crossref PubMed Scopus (345) Google Scholar, 19Lorenzi M.V. Castagnino P. Chen Q. Hori Y. Miki T. Oncogene. 1999; 18: 4742-4755Crossref PubMed Scopus (22) Google Scholar). FLAG-tagged ECT2 derivatives (0.5 μg each) were transfected into NIH 3T3 cells using the LipofectAMINE Plus reagent. Cells were fixed with a freshly prepared mixture of methanol:acetone (1:1) for 2 min at room temperature 40 h after transfection. Expressed proteins were visualized using anti-FLAG M2 monoclonal antibody-Cy3 conjugate (Sigma) in the presence of 1 μg/ml DAPI. U-2 OS cells were also transiently transfected with the expression vectors for GFP-tagged ECT2 mutants using FuGENE 6 reagent (Roche Applied Science) to confirm their subcellular localization. In this case, Hoechst 33342 dye was added to culture medium at a final concentration of 10 μm, and cells were directly observed under the fluorescence microscope. Images were acquired using a Zeiss Axiovert microscope equipped with a Photometrics digital camera and processed with IPLab software (Signal Analytics). Time-lapse Video Microscopy—The RFP-ECT2-ΔN5 expression vector was constructed by inserting ECT2-ΔN5 into pRsRed2-C1 (BO Biosciences/Clontech) at the BglII and EcoRI sites. pEGFP-actin was obtained from Clontech. NIH 3T3 cells were transfected with the equal mixture of both the plasmids using FuGENE 6 transfection reagent (Roche Applied Science). Cells were cultured on 35-mm plates in an environmental chamber on a stage of Zeiss Axiovert S-100 microscope equipped with motorized X-Y-Z stages. Images were taken at 3-min intervals by using a photometric digital camera controlled by OpenLab software (Improvision, Lexington, MA). Deletion of the N-terminal Half of Human ECT2 Induces Cell Transformation and Invasiveness—The mouse ect2 cDNA, ect2-T, which carries an N-terminal truncation, exhibits a high transforming activity, whereas the full-length clone does not significantly induce transformation (23Miki T. Smith C. Long J. Eva A. Fleming T. Nature. 1993; 362: 462-465Crossref PubMed Scopus (258) Google Scholar). To test if N-terminal truncation can also activate the transforming activity of human ECT2, we generated a FLAG epitope-tagged full-length human ECT2 and its derivative ECT2-ΔN5, which has a similar N-terminal deletion to mouse ect2-T. Like ect2-T, ECT2-ΔN5 exhibited a high transforming activity in NIH 3T3 cells, whereas the full-length ECT2, ECT2-F, did not show any detectable activity (Fig. 1A). Both mouse and human ECT2 similarly induced tiny foci of transforming cells with stellate morphology, which was distinct from ras- or sis-induced foci (Fig. 1, A and B). We previously reported that ect2-T stimulates anchorage-independent growth of NIH 3T3 cells and tumorigenicity in nude mice (23Miki T. Smith C. Long J. Eva A. Fleming T. Nature. 1993; 362: 462-465Crossref PubMed Scopus (258) Google Scholar). To examine additional oncogenic activities of ECT2, we established NIH 3T3 clones expressing FLAG-tagged derivatives of ECT2-F, ECT2-ΔN5, and ECT2-N2. ECT2-N2 carries the region from the N terminus to the S domain (see Fig. 6A). Western blot analysis with anti-FLAG antibody showed that these stable transfectants expressed FLAG-tagged ECT2-F, ECT2-N2, and ECT2-ΔN5 at comparable levels (Fig. 1C). Upon plating, ECT2-ΔN5-expressing cells formed secondary foci with stellate morphology, whereas the morphology of ECT2-F and ECT2-N2 clones was indistinguishable from the vector alone transfectants (Fig. 1B). When cultured in the presence of 10% serum, all of the stable ECT2 clones exhibited similar growth properties (data not shown). However, in medium containing 1% serum, vector alone, ECT2-N2 and ECT2-F transfectants did not grow well and the number of viable cells gradually decreased (Fig. 2A). In contrast, ECT2-ΔN5 transfectants continued to grow for at least 48 h under these conditions, suggesting that these transfectants acquired low serum dependence.Fig. 2Characterization of NIH 3T3 cells transfected with ECT2 variants. A, growth of ECT2 transfectants in low serum conditions. Cells were cultured in Dulbecco's modified Eagle's medium containing 1% serum, and viable cells were scored at the indicated time points. B, induction of cell invasiveness by ECT2. Stable NIH 3T3 transfectants expressing the indicated cDNA were used for Matrigel assays to estimate cell invasiveness in vitro. The cells that invaded the Matrigels were stained and photographed (left half). The number of cells passed through the Matrigel is summarized (right half). ect2-T is a mouse cDNA with an N-terminal deletion similar to human ECT2-ΔN5.View Large Image Figure ViewerDownload Hi-res image Download (PPT) To test if ECT2 can induce cell invasiveness, NIH 3T3 cells expressing ECT2-F, ECT2-ΔN5, or vector alone were placed on the surface of an artificial basement membrane, Matrigel, and the number of the cells that had migrated through the membrane was counted. Interestingly, ECT2-ΔN5 transfectants exhibited a strong invasion activity, whereas ECT2-F or the vector alone transfectants did not show significant activity (Fig. 2B). Additionally, mouse ect2-T, which corresponds to human ECT2-ΔN5, also exhibited a high activity of cell invasiveness. These results indicate that oncogenic ECT2 is an efficient activator of cell invasiveness. Oncogenic ECT2 Activates Rho Signaling Pathways—Because Rho GTPases are known to regulate the JNK and p38 MAPK pathways (24Coso O.A. Chiariello M. Yu J.C. Teramoto H. Crespo P. Xu N. Miki T. Gutkind J.S. Cell. 1995; 81: 1137-1146Abstract Full Text PDF PubMed Scopus (1567) Google Scholar, 25Minden A. Lin A. Claret F.X. Abo A. Karin M. Cell. 1995; 81: 1147-1157Abstract Full Text PDF PubMed Scopus (1446) Google Scholar), we examined whether these pathways are activated in ECT2 transfectants. We first analyzed endogenous JNK activity in these ECT2 transfectants using an antibody specific to the activated form of c-Jun, which is phosphorylated at serine 63. In vector alone transfectants, a very low level of JNK activity was observed (Fig. 3A, top). The activity of JNK in these cells was increased by the stimulation with sorbitol, an activator of the JNK pathway. The activity of JNK was also elevated in cells expressing ECT2-ΔN5. In contrast, ECT2-F or ECT2-N2 expression did not significantly affect the JNK activity. The JNK activity was induced by sorbitol to a similar level in all the transfectants (data not shown), indicating that all of the transfectants maintained the ability to induce JNK activity. In contrast, we did not observe significant activation of p38 or Erk MAPKs by ECT2 and its derivatives (Fig. 3A, middle and bottom). These results indicate that ECT2-ΔN5 preferentially activates the JNK signaling pathway. Rho proteins can stimulate the transcriptional activity regulated by serum response factor (SRF) (22Hill C.S. Wynne J. Treisman R. Cell. 1995; 81: 1159-1170Abstract Full Text PDF PubMed Scopus (1206) Google Scholar). To examine whether ECT2 can stimulate SRF-regulated transcription, a serum response element (SRE)-luciferase plasmid was used as a reporter. Upon coexpression of the reporter plasmid with either the full-length or truncated ECT2 expression vector, luciferase activity was estimated. As shown in Fig. 3B, upper panel, expression of ECT2-ΔN5 potently induced the transcriptional activity of the SRE reporter plasmid. In contrast, expression of ECT2-F exhibited the activity slightly higher than the vector alone control. The SRE-regulated transcriptional activity induced by ECT2-ΔN5 was efficiently inhibited by either of dominant negative RhoA, Rac1, or Cdc42. Moreover, either of constitutively active RhoA, Rac1, or Cdc42 efficiently enhanced SRE-mediated transcription in this system. These results suggest that ECT2-ΔN5 can stimulate SRE-mediated transcription through the activation of Rho GTPases. We previously showed that Ost, a guanine nucleotide exchange factor for RhoA and Cdc42, activates the transcriptional activity regulated by activator protein-1 (AP-1) (19Lorenzi M.V. Castagnino P. Chen Q. Hori Y. Miki T. Oncogene. 1999; 18: 4742-4755Crossref PubMed Scopus (22) Google Scholar). To examine whether ECT2 can also stimulate AP-1-regulated transcriptional activity, an AP-1-binding site-luciferase plasmid was utilized as a reporter. As shown in Fig. 3B, lower panel, expression of ECT2-ΔN5 moderately elevated AP-1-regulated transcriptional activity, whereas ECT2-F or ECT2-N2 failed to stimulate the activity. Coexpression of dominant negative RhoA, Rac1, or Cdc42 reduced the ECT2-ΔN5-mediated stimulation of AP-1-regulated transcription, albeit at lower levels as compared with their effects on SRE-regulated transcription. We also found that constitutively active RhoA, Rac1, or Cdc42 efficiently stimulated AP-1-mediated transcription. Among these GTPases, Rac1 displayed the highest level of stimulation of AP-1-regulated transcriptional activity. All of these results indicate that ECT2 can regulate the transcriptional events mediated by SRE and AP-1 through the activation of Rho GTPases. Oncogenic ECT2 Induces Cell Rounding in NIH 3T3 Fibroblasts—Rho family proteins are involved in the organization of actin-based cytoskeletal structures. In fibroblasts, RhoA activates actin stress fiber formation, whereas Rac1 and Cdc42 induce lamellipodia and filopodia formation, respectively (8Hall A. Science. 1998; 279: 509-514Crossref PubMed Scopus (5218) Google Scholar). To test which actin-based structures ECT2 can induce, we transiently expressed GFP-tagged ECT2-F, ECT2-ΔN5, or GFP vector alone in NIH 3T3 cells. A population (∼20%) of cells expressing GFP-ECT2-ΔN5 showed a flat phenotype with moderately enhanced actin stress fiber formation (Fig. 4A, GFPECT2-ΔN5, left panels), suggesting that Rho was preferentially activated by ECT2-ΔN5 in these cells. However, the majority of NIH 3T3 cells expressing GFP-ECT2-ΔN5 exhibited a compacted structure with saturated F-actin staining
Referência(s)