Repression of Bone Morphogenetic Protein and Activin-inducibleTranscription by Evi-1*

Received for publication, December 20, 2004, and in revised form, April 1, 2005Published, JBC Papers in Press, April 22, 2005, DOI 10.1074/jbc.M414305200

Tamara Alliston‡§, Tien C. Ko‡§¶, Yanna Cao¶, Yao-Yun Liang�, Xin-Hua Feng�**,Chenbei Chang‡‡, and Rik Derynck‡§§

From the ‡Departments of Cell and Tissue Biology and Anatomy, University of California at San Francisco, SanFrancisco, California 94143-0512, the ¶Department of Surgery, Sealy Center for Cancer Cell Biology, University of TexasMedical Branch, Galveston, Texas 77555-0542, the �Michael E. DeBakey Department of Surgery, Department of Molecularand Cellular Biology, Biology of Inflammation Center, and Baylor College of Medicine Cancer Center, Baylor College ofMedicine, Houston, Texas 77030, and the ‡‡Department of Cell Biology, University of Alabama at Birmingham,Birmingham, Alabama 35294

Smads, key effectors of transforming growth factor(TGF)-�, activin, and bone morphogenetic protein(BMP) signaling, regulate gene expression and interactwith coactivators and corepressors that modulate Smadactivity. The corepressor Evi-1 exerts its oncogenic ef-fects by repressing TGF-�/Smad3-mediated transcrip-tion, thereby blocking TGF-�-induced growth arrest. Be-cause Evi-1 interacts with the highly conserved MH2domain of Smad3, we investigated the physical andfunctional interaction of Evi-1 with Smad1 and Smad2,downstream targets of BMP and activin signaling, re-spectively. Evi-1 interacted with and repressed the re-ceptor-activated transcription through Smad1 andSmad2, similarly to Smad3. In addition, Evi-1 repressedBMP/Smad1- and activin/Smad2-mediated induction ofendogenous Xenopus gene expression, suggesting a roleof repression of BMP and activin signals by Evi-1 invertebrate embryogenesis. Evi-1 also repressed the in-duction of endogenous Smad7 expression by TGF-� fam-ily ligands. In the course of these studies, we observedEvi-1 repression of Smad transactivation even whenSmad binding to DNA was kept constant. We thereforeexplored the mechanism of Evi-1 repression of TGF-�family-inducible transcription. Evi-1 repression did notresult from displacement of Smad binding to DNA or toCREB-binding protein but from the recruitment of Evi-1by Smad3 and CREB-binding protein to DNA. FollowingTGF-� stimulation, Evi-1 and the associated corepressorCtBP were recruited to the endogenous Smad7 pro-moter. Evi-1 recruitment to the promoter decreasedTGF-�-induced histone acetylation, coincident with itsrepression of Smad7 gene expression. In this way, Evi-1acts as a general Smad corepressor to inhibit TGF-�-,activin-, and BMP-inducible transcription.

TGF-�1 family ligands, such as TGF-�, BMPs, and activin,regulate a number of cellular functions, including cell prolifer-ation and differentiation, apoptosis, and matrix production.The actions of the TGF-� family ligands are particularly im-portant during embryogenesis and tumorigenesis (1–4). BMPand activin signaling are required for gastrulation, mesodermformation, tissue differentiation, and formation of extraembry-onic tissues (1). TGF-� acts as a tumor suppressor by inducingthe expression of the p15Ink4B or p21Cip1 Cdk inhibitor genes toarrest proliferation of many cell types (2–4). Interference withTGF-�-induced p15Ink4B and/or p21Cip1 expression, by onco-genic overexpression of c-Myc or c-Ski, leads to uncontrolledcell growth and tumorigenesis (5–7).

TGF-� family ligands signal through Smads, which regulatetranscription by cooperating with DNA sequence-specific tran-scription factors (8–10). Among the Smads, Smad1, -5, and -8relay signals from BMP-activated receptors, whereas Smad2and Smad3 are the primary effectors of activin and TGF-�signaling, respectively. Upon ligand binding, the receptor com-plex phosphorylates and activates these Smads, resulting indissociation of Smads from the receptor, heteromerization withSmad4, and nuclear translocation of the Smad complex. Withthe exception of Smad2, Smads bind DNA through their N-terminal MH1 domains. Stable interaction with select pro-moter sequences, however, occurs by association of the Smadcomplex with sequence-specific transcription factors. For ex-ample, in response to TGF-�, Smad2 and Smad3 interact withSp1 at the p15Ink4B or p21Cip1 promoters (11, 12). The receptor-activated Smads also bind directly, through their C-terminalMH2 domains, to the coactivators CBP/p300. This coactivatorrecruitment, together with the physical and functional inter-actions with sequence-specific transcription factors, confers li-gand-induced transcriptional activation (8–10). Smads can re-cruit additional coactivators, such as SMIF, MSG1, or Swift,that enhance the transcription response or corepressors, in-cluding c-Ski, TGIF, SIP1, SNIP1, or Tob, which decrease orinhibit ligand-induced transactivation (13).

Another Smad3 corepressor is the nuclear proto-oncogene,evi-1, which inhibits cellular responsiveness to TGF-�-induced growth arrest (14). The Evi-1 protein is a 145-kDa,10-zinc finger-containing polypeptide that can function as aDNA sequence-dependent transcription repressor (15). Tar-

* This research was supported by a Hulda Irene Duggan ArthritisInvestigator career development award from the Arthritis Foundation(to T. A.) and National Institutes of Health Grants RO1-GM63773 andRO1-CA108454 (to X.-H. F.), RO1-DK60105, P01-DK35608 and K08-CA64191 (to T. C. K.), RO1-HD43345 (to C. C.), and RO1-CA63101 andP60 DE13058 (to R. D.). The costs of publication of this article weredefrayed in part by the payment of page charges. This article musttherefore be hereby marked “advertisement” in accordance with 18U.S.C. Section 1734 solely to indicate this fact.

§ These authors contributed equally to this work.** A Leukemia and Lymphoma Society Scholar.§§ To whom correspondence should be addressed: Dept. of Cell and

Tissue Biology, University of California at San Francisco, San Fran-cisco, CA 94143-0512. Tel.: 415-476-7322; Fax: 415-502-7338; E-mail:derynck@itsa.ucsf.edu.

1 The abbreviations used are: TGF, transforming growth factor; BMP,bone morphogenetic protein; CBP, CREB-binding protein; CREB,cAMP-response element-binding protein; HA, hemagglutinin; RT, re-verse transcription; ZF1, zinc finger 1.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 280, No. 25, Issue of June 24, pp. 24227–24237, 2005© 2005 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org 24227

by guest on September 10, 2020

http://ww

w.jbc.org/

Dow

nloaded from

geted inactivation of the evi-1 gene results in embryoniclethality because of defective heart, brain, and paraxial mes-enchyme development and widespread hypocellularity (16).This phenotype is consistent with the expression of Evi-1during embryogenesis and the role of Evi-1 as a potent acti-vator of cellular proliferation (17). In adult tissues, Evi-1 isexpressed at very low levels with the exception of the promy-elocytic stage of myeloid differentiation, where it activatesmyeloid cell proliferation while inhibiting differentiation(18). The expression pattern and proliferative function ofEvi-1 explains the occurrence of myeloid leukemias in hu-mans with chromosomal rearrangement at the evi-1 locus.Most of these rearrangements cause Evi-1 overexpressionand uncontrolled myeloid cell proliferation (15). A recentstudy using microarray analysis of gene expression in AMLpatients revealed that overexpression of Evi-1 correlatedwith poor treatment outcomes, illustrating the importance oftightly regulated Evi-1 expression and function (19).

Evi-1-mediated transformation and transcriptional repres-sion require interaction with the corepressor CtBP (20, 21),which itself interacts with histone deacetylases (HDACs) (22).In addition to interacting with transcriptional corepressors,Evi-1 can also interact with the co-activators CBP and p300/CBP-associated factor, which have intrinsic histone acetylaseactivity (23). Coexpression of CBP reverses the transcriptionrepression by Evi-1 at an artificial promoter that allows forEvi-1 binding (23); however, the physiological importance ofthis interaction at a natural promoter that is regulated byEvi-1 has yet to be determined.

Evi-1 has been shown to suppress TGF-�-induced signalingthrough direct interaction with Smad3. Thus, Evi-1 inhibitsTGF-�-mediated Smad3 induction of a p15Ink4B Cdk inhibitorreporter gene (14). This inhibition is thought to prevent thenormal growth-inhibitory response to TGF-� and to enhanceproliferation. Full repression of Smad3 activity requires in-teraction of Evi-1 with CtBP (20). Evi-1 was proposed to actas a Smad3-specific corepressor that decreases Smad3 DNAbinding (14). The requirement of the Smad3 MH2 domain forEvi-1 repression of TGF-� inducible transcription led us toinvestigate the physical interaction of Evi-1 with Smads 1–4,which have MH2 domains that are highly homologous toSmad3. Since the MH2 domain of Smad3 also interacts withCBP/p300 (13), it was conceivable that Evi-1 binding to theMH2 domain might interfere with coactivator recruitment,thus providing a mechanism for repression of Smad3-medi-ated transcription. We therefore examined the ability ofEvi-1 to bind and repress the function of TGF-�-, BMP-, andactivin-receptor activated Smads.

We observed that Evi-1 interacted with each Smad tested.Evi-1 repressed BMP/Smad1-, activin/Smad2- and TGF-�/Smad3-induced transcription of reporter genes as well as en-dogenous Smad7 gene expression. Evi-1 also repressed BMP/Smad1- and activin/Smad2-mediated induction of endogenousXenopus genes required for cell fate specification, suggesting arole for Evi-1 in development. Evi-1 did not interfere with CBPbinding to Smad3; nor did it displace Smad3 from DNA.Rather, TGF-� stimulated recruitment of Evi-1 and CtBP tothe endogenous Smad7 promoter, resulting in decreased TGF-�-induced histone acetylation and transcription.

MATERIALS AND METHODS

Expression Plasmids—Coding sequences for C-terminally Myc tag-or VP16-fused Evi-1 proteins or defined regions were generated byoligonucleotide- or PCR-based techniques from pME-Evi-1 and pME-Evi-1�ZF1-7 (14), kindly provided by H. Hirai and inserted into theEcoRI-SalI sites of the mammalian expression plasmid pRK5 (24) orderivatives or into the retroviral vector LPCX (25). A neomycin-resist-

ant retroviral Evi-1 expression vector, p50MFLRpWT-neo, was gener-ously provided by C. Bartholomew (21). Retroviral vectors for express-ing siCtBP small interfering RNA and the vector control were providedby Y. Shi (61, 62). C-terminally HA-fused CBP proteins were generatedby restriction digestion of pRc-RSV-CBP-HA provided by R. Goodmanand inserted into pRK5 or derivatives. The pRK5-based expressionplasmid for constitutively active T�RI(T202D) has been described (26,27). Expression plasmids encoding Gal-Smad1 and GalL-Smad4 (28)were gifts from J. Massague. The expression plasmids encoding Gal-Smad2 and Gal-Smad3 were previously described (29). pFAST-1 wasprovided by M. Whitman. The HAT-defective point mutant CBP(F1541A) was provided by T. Kouzarides and co-workers (30). Consti-tutively active mutants of ActRIB and BMPRIB were gifts of L. Attisanoand J. Wrana (31, 32). Detailed information on the plasmids will beprovided upon request.

Reporter Plasmids—Plasmid p800-Luc containing the luciferase re-porter gene under control of the PAI-1 promoter (33) and (SBE)4-Luc(34) were used to measure TGF-�- and Smad-induced gene expression.The Xvent2-Luc reporter plasmid contains �250 bp of the Xvent2promoter that is specifically responsive to BMP signaling (35) and wasprovided by C. Niehrs. The �226gsc-Luc reporter plasmid, in whichluciferase is expressed from an activin/BVg1-specific response elementof the goosecoid promoter (36), was provided by K. W. Cho. The pFR-Lucreporter plasmid has five copies of a Gal4-binding element, followed bythe luciferase gene (Stratagene). Smad7Pro-Luc containing 0.5 kb ofSmad7 promoter sequence was previously described (37). Plasmid pSV-�-galactosidase was acquired from Promega.

Cell Culture and Transfections—COS-1 cells were maintained inDulbecco’s modified Eagle’s medium, 10% fetal bovine serum, whereasHepG2 and MDA-MB-468 cells were maintained in minimal essentialmedium, 10% fetal bovine serum, supplemented with nonessentialamino acids. Mouse embryonic Smad3�/� fibroblasts were grown inDulbecco’s modified Eagle’s medium, 10% fetal bovine serum. C2C12myoblasts were grown in Dulbecco’s modified Eagle’s medium, 20%fetal bovine serum. HEC-1b cells were grown in Eagle’s minimal essen-tial medium with Earle’s balanced salt solution supplemented withsodium pyruvate, nonessential amino acids, and 10% fetal bovine se-rum. COS-1 and Smad3�/� mouse embryonic fibroblasts were trans-fected using Lipofectamine (Invitrogen). HepG2, MDA-MB-468, andC2C12 cells were transfected using FuGENE 6 (Roche Applied Science).To generate C2C12 cells stably expressing Evi-1 or siCtBP, retroviralvectors were used as described (25). Evi-1 and CtBP expression levelswere verified using Western blotting.

Immunoprecipitations and Western Blotting—FLAG- or Myc-taggedproteins were immunoprecipitated from transfected cell lysates as de-scribed (38). Chemical cross-linking of interacting proteins using dithio-bis(succinimidyl propionate) was performed as described (39), followedby harvest of lysates in 50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.5%Triton X-100. Anti-FLAG (M2; Sigma-Aldrich) or anti-Myc (9E10; Co-vance Research Products) antibodies were used to immunoprecipitatetagged proteins. For subsequent Western blotting, the immunoprecipi-tated proteins were separated by SDS-PAGE, transferred onto polyvi-nylidene difluoride membrane, and detected with the appropriate pri-mary antibody. Antibodies used for Western blotting include anti-Evi-1(63), provided by J. Ihle, anti-Smad3 (Zymed Laboratories Inc.), anti-CtBP (H-440; Santa Cruz Biotechnology, Inc., Santa Cruz, CA), anti-FLAG (M2), anti-Myc (9E10), and anti-HA (HA-11; Covance ResearchProducts). Antibody-bound proteins were visualized with chemilumi-nescent reagents (Amersham Biosciences).

Transcription Reporter Assays—Transfections, TGF-� treatment,and reporter assays were done as described (38). The total amount oftransfected DNA was kept constant by adding vector DNA when nec-essary. Transfected cells were treated for 24 h with or without 400 pM

TGF-�, and luciferase activities were measured 48 h after transfection.All assays were performed at least three times. Values are shown asluciferase activity relative to the minimal promoter.

Specifically, in Gal4 transactivation assays, the plasmids encodingGal-Smads were used to keep DNA binding constant to allow evaluationof Smad transactivation function in the presence of other coexpressedproteins. Changes in Smad transactivation were quantified by meas-uring luciferase expression from the heterologous Gal4 promoter in thecotransfected FR-Luc reporter plasmid.

Mammalian two-hybrid assays were utilized to assess physicalinteractions among proteins. Plasmids encoding Gal-Smads or plas-mids for Evi-1 or Smads fused to the VP16 activation domain and theluciferase reporter plasmid pFR-Luc were transfected into HepG2,MDA-MB-468, or Smad3�/� cells, which were then treated with orwithout TGF-�.

Evi-1 Repression of TGF-� Family-inducible Transcription24228

by guest on September 10, 2020

http://ww

w.jbc.org/

Dow

nloaded from

GST Fusion Proteins and in Vitro Protein-binding Assays—PlasmidspGEX-Smad1–4 have been described (29, 40). Equal amounts of GST orGST-Smad fusion protein bound to glutathione-Sepharose beads wereincubated, as previously described, with 35S-labeled, in vitro translatedproteins (TNT translation kit; Promega) with similar specific radioac-tivity (29). Associated 35S-labeled proteins were detected by SDS-PAGEand autoradiography.

DNA Precipitation Assays—Biotinylated 5� oligonucleotides contain-ing two repeats of the consensus Smad binding element (GTCTAGAC)from the Smad7 promoter were synthesized and annealed with unbi-otinylated complementary oligonucleotides to generate probes for DNAprecipitation assays: SBE, 5�-ACTTGTCTAGACGTCTAGACTTGG-3�.Immobilization of biotinylated oligonucleotide probes and absorption ofnuclear extracts from transfected COS cells were carried out usingstreptavidin-coated Magna-sphere paramagnetic particles (Promega)following the manufacturer’s instructions and as described (11). Afterextensive washing, DNA-bound proteins were subjected to SDS-PAGE,followed by Western blotting using antibodies that recognized FLAG(M2)-, Myc (9E10)-, or HA (HA-11)-tagged proteins. In parallel, the celllysates were immunoblotted to demonstrate the expression of trans-fected proteins.

RNA Isolation and Quantitative Real Time PCR—RNA was isolatedusing RNAeasy reagents (Qiagen). cDNA was generated from equalamounts of RNA using NEB protoscript reagents with the dtVN23primer. Real time PCR analysis was performed to assess Smad7 ex-pression. Expression was normalized to ribosomal protein L19. Primersfor mouse Smad7 correspond to 5�-CAGAAAGTGCGGAGCAAGATC-3�and 5�-AAACCCACACGCCATCCA-3� and Taqman 5�-CGGCATCCAG-CTGACGCGG-3�. Primers for mouse ribosomal protein L19 correspondto 5�-GCTGGATGACAACCTGCTGTACT-3� and 5�-TGTGGGACCGGC-TTCATG-3� and Taqman 5�-CGCCCTTCCCGAGTACAGCACCTT-3�.

Chromatin Immunoprecipitations—LPCX and LPCX-Evi-1 C2C12cells were treated with or without TGF-� (1 ng/ml) in 0.2% serum for 60min to assess Evi-1 and CtBP binding and 120 min to assess acetylatedhistone 4. Cells were cross-linked with 1% formaldehyde at room tem-perature for 10 min. Following a 5-min 125 mM glycine quench and twoPBS washes, cells were lysed for 10 min in lysis buffer (50 mM HEPES-KOH, pH 7.9, 1 mM EDTA, 0.5 mM EGTA, 140 mM NaCl, 10% glycerol,1% Nonidet P-40, 0.5% Triton X-100) with protease inhibitor mixture(Roche Applied Science), harvested and spun 500 � g for 1 min. Nucleiwere washed for 10 min in wash buffer (10 mM Tris-HCl, pH 8.0, 1 mM

EDTA, 0.5 mM EGTA, and 200 mM NaCl), collected, and resuspended inradioimmune precipitation buffer (10 mM Tris-HCl, pH 8.0, 1 mM

EDTA, 0.5 mM EGTA, 140 mM NaCl, 1% Triton X-100, 0.1% sodiumdeoxycholate, 0.1% SDS). Soluble chromatin was prepared by sonica-tion and immunoprecipitated using anti-Myc (9E10), anti-acetyl his-tone H4 (Upstate Biotechnology, Inc.), or anti-CtBP (Santa Cruz Bio-technology) antibodies for 16 h. Immune complexes were collected usingProtein A-agarose beads that had been preincubated with 100 �g/mlsalmon sperm DNA and IgG. Sequential 10-min washes were per-formed in radioimmune precipitation buffer, radioimmune precipitationbuffer with 500 mM NaCl, buffer III (10 mM Tris-HCl, pH 8.0, 0.25 M

LiCl, 1% Nonidet P-40, 1% sodium deoxycholate, 1 mM EDTA), and TEbuffer. Reactions were incubated at 55 °C for 3 h with proteinase K andthen at 65 °C for 6 h to reverse the formaldehyde cross-linking. DNAfragments were purified by phenol chloroform extraction and ethanolprecipitation. PCR was performed using 5 �l of the 50-�l DNA solutionusing real-time quantitative PCR with primers corresponding to mouseSmad7 promoter sequence: 5�- AGCTTTTAGAAACCCGATCTGTTG-3�,5�-CGTCACGTGGCCGTCTAGA-3�, and Taqman 5�-TGCGAAACACA-ATCGCTTTTTT-3�. Each reaction was normalized using primers di-rected against ribosomal protein L19 sequence.

Xenopus Gene Expression Analyses—RNAs used for microinjectioninto Xenopus embryos were synthesized with SP6 RNA polymeraseusing mMessage mMachine kit (Ambion). The following DNA tem-plates were used: KpnI-linearized pRK5-Evi-1, AscI-linearizedpCS2��Smad1 and pCS2��Smad2, EcoRI-linearized pSP64T-activin,and XbaI-linearized pSP64T-BMP2. RNAs encoding the above geneswere injected alone or in combination into the animal poles of two-cellstage embryos. The doses of RNAs used are Evi-1 (1–2 ng), Smad1 (4ng), Smad2 (0.5 ng), activin (2 pg), and BMP2 (0.5 ng). Ectodermalexplants (animal caps) from injected embryos were obtained at blastulastages (stage 8–9) and incubated to gastrula stages (stage 10–11), atwhich time total RNA was extracted from these caps, and the geneexpression pattern was analyzed by reverse transcription (RT)-PCR.The primers used in the RT-PCR were as described (41), with theaddition of the Msx1 primers Msx1-U 5�-GCTAAAAATGGCTGCTAA-3�and Msx1-D 5�-AGGTGGGCTGTGTAAAGT-3� and the Mix1 primers

Mix1-U 5�-AATGTCTCAAGGCAGAGG-3� and Mix1-D 5�-GTGTCACT-GACACCAGAA-3�.

RESULTS

Evi-1 Represses Smad3-dependent Transcription Independ-ently of DNA Binding—Evi-1 has been shown to repress TGF-�-induced growth inhibition and transcription reporter activityfrom the p15Ink4B or 3TP promoters (14). At these promoters,TGF-�-activated Smad3/4 cooperates with sequence-specifictranscription factors (i.e. Sp1 or c-Jun/c-Fos, respectively) (11,42, 43). Therefore, we tested whether Evi-1 was able to repressTGF-�- and Smad3/4-mediated transcription independently ofits interacting, sequence-specific transcription factors. TandemSmad binding elements in the (SBE)4 promoter allow coopera-tive binding of the Smads and TGF-�-inducible transcriptionwithout the help of other sequence-specific transcription fac-tors (34). Similarly to the previously tested promoters, Evi-1repressed TGF-�- and Smad3/4-induced transcription from the(SBE)4 promoter (Fig. 1A). This result suggests that Evi-1directly targets the intrinsic transactivation function of Smad3in the absence of coactivation of an interacting, sequence-spe-cific transcription factor.

The zinc finger 1 (ZF1) domain of Evi-1 is required for Smad3interaction, and the repressor domain of Evi-1 confers tran-scriptional repression (14). However, the domains of Smad3required for physical interaction with Evi-1 were not fullydefined. We therefore examined the domains of Smad3 re-quired for interaction with Evi-1 using mammalian two-hybridassays, in which the transcriptional activation correlates withthe physical interaction in transfected cells (44). These assaysrequired deletion of the repressor domain of Evi-1, which wouldotherwise repress the transcription activity of the VP16 trans-activation domain (Fig. 1D). As shown in Fig. 1B, Evi-1 did notinteract with the MH1 domain and linker segment of Smad3 inGal-Smad3NL but showed a strong interaction with the MH2domain of Smad3 in Gal-Smad3C, which was further enhancedupon TGF-� stimulation. As expected, deletion of the ZF1 do-main abolished the interaction of Evi-1 with Smad3.

The repression of TGF-�/Smad3-induced transcription, evenwithout cooperation with other sequence-specific transcriptionfactors, and the interaction of Evi-1 with the MH2 domain ofSmad3 suggested that Evi-1 might repress TGF-�/Smad3-in-duced transcription through repression of the transactivationfunction of the Smad3 MH2 domain. We tested this hypothesisunder constant DNA binding conditions using the MH2 domainof Smad3 fused to a Gal4 DNA binding domain, using a Gal4binding promoter as a reporter. As shown in Fig. 1C, Evi-1repressed the transactivation function of Smad3, independentof DNA binding. Maximal repression required both the ZF1 andrepressor domains, since deletion of either domain reducedEvi-1 repression. We conclude that Evi-1 targets the transac-tivation function of the Smad3 MH2 domain, independently ofa possible effect of Evi-1 on DNA binding of Smad3.

Evi-1 Interacts with Smad1, -2, -3, and -4—The interactionof Evi-1 with the MH2 domain of Smad3 (Fig. 1, B and C), andthe high conservation of this Smad domain led us to testwhether Evi-1 interacts with other Smads in addition toSmad3. As shown using mammalian two-hybrid assays, Evi-1without its repressor domain interacted with the receptor-ac-tivated Smad1, -2, and -3 and had a lower level interaction withSmad4. These interactions were increased in the presence ofTGF-� and abolished when the ZF1 domain was deleted fromEvi-1 (Fig. 2A). In principle, the ability of the Smads to hetero-merize with each other and the endogenous presence of Smad2,-3, and -4 in HepG2 cells could contribute to these in vivointeractions. However, Evi-1 still interacted with Smad1 orSmad2 in Smad3-deficient cells and with Smad1, -2, and -3 in

Evi-1 Repression of TGF-� Family-inducible Transcription 24229

by guest on September 10, 2020

http://ww

w.jbc.org/

Dow

nloaded from

the Smad4-deficient MDA-MB-468 cells (Fig. 2B). Thus, Evi-1interacted with all Smads tested, and this interaction de-pended in all cases on the ZF1 domain in Evi-1.

In GST adsorption assays, the ZF1 domain of Evi-1 inter-acted with GST-fused Smads 1–4 and not GST alone (Fig. 2C),and all four Smads were able to interact with the GST-fusedZF1 domain of Evi-1 but not with GST alone (data not shown).These data are consistent with the two-hybrid interaction dataand confirm the role of the ZF1 domain in the in vivo interac-tions of Evi-1 with the Smads (Fig. 2A).

The interaction of Evi-1 with each of the four Smads was alsoobserved by coimmunoprecipitation from transfected cell ly-sates (Fig. 2D). In these assays, Evi-1 coprecipitated withSmad3 or Smad4, albeit to a different extent. Under the same

conditions, Evi-1 coprecipitated only weakly with Smad1 or -2,but reversible chemical cross-linking using dithiobis(succinimi-dyl propionate) sufficiently stabilized the Smad1-Evi-1 orSmad2-Evi-1 complexes to allow specific coprecipitation ofEvi-1 with Smad1 or -2 (Fig. 2D).

To test the interaction of endogenous Evi-1 and Smad pro-teins, we performed coimmunoprecipitation assays in HEC-1bcells, the only one of �60 cell lines screened to express highlevels of Evi-1 (60). Endogenous Evi-1 and Smad3 interactedonly following TGF-� stimulation (Fig. 2E). Together, datafrom mammalian two-hybrid, GST adsorption, and coimmuno-precipitation assays reveal that Evi-1 can interact with Smad1,-2, -3, and -4. The differing affinities presumably reflect thedifferential involvement of Evi-1 domains (data not shown), in

FIG. 1. Repression of TGF-�- and Smad3-inducible transcription by Evi-1 requires the MH2 domain of Smad3 and the ZF1 domainof Evi-1. A, HepG2 cells were transfected with the (SBE)4 promoter-luciferase reporter plasmids along with expression vectors for Evi-1, Smads,or pRK5 control plasmid, as indicated. B, mammalian two-hybrid assays using VP-fused Evi-1 deletion mutants, Gal-fused Smad segments, andthe Gal binding sequences of the FR-Luc reporter showed interaction of the N-terminal segment of Evi-1 (Evi-1N) with the MH2 domain of Smad3(Smad3C). C, Evi-1 represses transactivation of FR-Luc by the Gal-fused MH2 domain of Smad3 (Gal-Smad3C). D, schematic diagram of Evi-1 andSmad3 and the deletion mutants used in Fig. 1 and other experiments.

Evi-1 Repression of TGF-� Family-inducible Transcription24230

by guest on September 10, 2020

http://ww

w.jbc.org/

Dow

nloaded from

addition to ZF1, in Smad binding, as well as differences in theassays used.

Evi-1 Represses Smad1- and Smad2-mediated Transcrip-tion—Because of the physical interactions of Evi-1 with Smads1–4, we assessed whether Evi-1 could repress Smad1- andSmad2-mediated transactivation. We tested this hypothesisunder constant DNA binding conditions using the MH2 do-mains of Smad1, -2, and -3 fused to the Gal4 DNA bindingdomain, using a Gal4 binding promoter as a reporter. As shownin Fig. 3A, Evi-1 strongly repressed the basal and receptor-induced transactivation of Gal4-Smad1, -2, and -3. Consistentwith the previous report on Smad3 (14), the repression ofSmad1 and Smad2 transcription was decreased when the ZF1segment was deleted from Evi-1 (Fig. 3A). The requirement ofthe ZF1 domain for efficient repression is consistent with theinteraction of MH2 domains of Smads with the ZF1 domain ofof Evi-1 (Fig. 1B and data not shown).

We then tested the ability of Evi-1 to repress Smad1-,Smad2-, and Smad3-mediated transcription using BMP, ac-tivin, and TGF-�-responsive promoter-reporter genes. Wetested Smad1-mediated transcription using a BMP-respon-sive promoter segment derived from the Xenopus vent-2 gene.Consistent with previous reports (35), activation of ALK-6/BMP-RIB signaling enhanced transcription, and coexpres-sion of Smad1 or Smad1/4 strongly activated transcription. Ineach case, Evi-1 strongly repressed the transcriptional acti-vation of the Xvent-2 promoter (Fig. 3B). Similarly, we useda reporter plasmid containing a segment of the Xenopus

goosecoid promoter that binds the transcription factorFAST-1 to measure Smad2-activated transcription. Activinsignaling causes Smad2 to bind FAST-1, thereby activatingtranscription from the goosecoid promoter (45). In the pres-ence of FAST-1, activin signaling conferred a low level tran-scription, which was strongly enhanced by coexpression ofSmad2 or Smad2/4. Evi-1 strongly repressed the activin re-ceptor- and Smad2-induced transcription (Fig. 3C). Consist-ent with the ability of Evi-1 to repress TGF-�-activatedp15Ink4B reporter transactivation (14), Evi-1 also repressedTGF-�- and Smad3/Smad4-responsive induction of the PAI-1promoter (Fig. 3D). Therefore, Evi-1 repressed BMP/Smad1-,activin/Smad2-, and TGF-�/Smad3-mediated transactivationof their respective target gene promoters.

Evi-1 Represses Activin- and BMP-responsive EndogenousGene Expression in Xenopus Embryos—Although aberrantlyexpressed in tumorigenesis (15), Evi-1 expression is predomi-nantly restricted to embryonic stages and is down-regulated inadult tissues (17). During embryonic development, activin/nodal and BMP signals control patterning and cell fate deter-mination in multiple processes (1), raising the possibility thattheir activities may be regulated by Evi-1 during development.We therefore tested whether Evi-1 inhibits BMP- and activin-inducible endogenous gene expression during Xenopus embry-ogenesis using the Xenopus ectodermal explant system. Thechanges in gene expression and cell differentiation in responseto Smad1-mediated BMP signals and Smad2-mediated activin/nodal/Vg1 signals have been well studied in this system. In the

FIG. 2. Evi-1 interacts with Smad1, -2, -3, and -4. A and B, mammalian two-hybrid assays show interaction of Evi1N with Smad1, -2, -3, and-4. The interactions are enhanced by TGF-� in HepG2 cells (A) and also occur in Smad3�/� fibroblasts and Smad4-deficient MDA-MB-468carcinoma cells (B). C, the in vitro translated ZF1 domain of Evi-1 interacts with GST-Smads but not GST. Evi-1 ZF1, prior to adsorption, isvisualized in the 10% input lane. D, FLAG-tagged Smad1, -2, -3, and -4 interact with Myc-tagged Evi-1 in transfected COS cells. Western analysisusing Myc or FLAG antibodies detected the tagged proteins in immunoprecipitated complexes (IP) or cell lysates (IB). E, endogenous Evi-1interacts with endogenous Smad3 in HEC-1b cells following 1 h of TGF-� stimulation (1 ng/ml). Rabbit IgG was used as a negative control foranti-Smad3 coimmunoprecipitations.

Evi-1 Repression of TGF-� Family-inducible Transcription 24231

by guest on September 10, 2020

http://ww

w.jbc.org/

Dow

nloaded from

absence of growth factors, the ectodermal explants (i.e. animalcaps) develop into atypical epidermis. However, incubationwith TGF-� family ligands or increased expression of the cor-responding Smads induces mesoderm- and endoderm-specificgene expression. Specifically, activin/nodal/Vg1 and Smad2 in-duce dorsal mesoderm and endoderm, whereas BMP andSmad1 induce ventral mesodermal and endodermal markers(for a review, see Ref. 46).

We first analyzed the effects of Evi-1 on activin- and BMP-dependent mesendodermal marker induction. RNAs encodingactivin or BMP2 were injected into the animal poles of two-cellstage embryos either alone, or together with different doses ofEvi-1 RNA. Animal caps were dissected from the injected em-bryos at blastula stages and incubated to gastrula stages beforeprocessing them for RT-PCR assays for marker gene expres-sion. As shown in Fig. 4A, activin induced the ventrolateralmarker Xwnt8, the dorsal mesodermal marker chordin, and theendodermal marker Mix1 in animal caps (lanes 3). Coexpres-sion of Evi-1 with activin, however, suppressed the expressionof all of these genes (Fig. 4A, compare lanes 4 and 5 with lane3). Similarly, BMP2 induced the endodermal marker Mix1, aswell as the ventral markers Xhox3 and Msx1, and Evi-1 re-pressed the induction of these genes (Fig. 4A, compare lanes 7and 8 with lane 6). The data thus indicate that Evi-1 blocksboth activin and BMP signals in Xenopus animal caps.

We then examined the effects of Evi-1 on marker gene

induction by Smad1 or Smad2. RNAs coding for Smad1 orSmad2 were injected alone or with Evi-1 RNA into the animalpoles of two-cell stage embryos, and animal caps from in-jected embryos were obtained and analyzed as above. Smad2expression induced expression of Xwnt8, chordin, and Mix1,similarly to activin (Fig. 4B, lane 3). Coexpression of Evi-1repressed the induction of these mesendodermal genes bySmad2 (Fig. 4B, compare lanes 4 and 5 with lane 3). Simi-larly, the markers induced by Smad1 (i.e. Mix1, Xhox3 andMsx1) were inhibited by Evi-1 to different degrees (Fig. 4B,compare lanes 7 and 8 with lane 6). We also note that Evi-1alone blocked the endogenous expression of Msx1 in animalcaps (Fig. 4, A and B, lanes 2). Since Msx1 is a direct down-stream target of BMP signals (47, 48) and its expressionrelies on endogenously expressed BMP ligands in the animalcaps, our result demonstrates that Evi-1 blocks endogenousas well as ectopically activated BMP signals. Therefore, inaddition to its role in promoting cell proliferation in embry-ogenesis, Evi-1 may also regulate cell fate determination byits repression of activin and BMP signaling.

Evi-1 Represses Endogenous Smad7 Transactivation—Toevaluate the effect of Evi-1 on endogenous mammalian geneexpression, we generated retrovirally infected, control, andEvi-1-expressing C2C12 cells. In contrast to LPCX control cells,the LPCX-Evi-1 C2C12 cells expressed Myc-tagged Evi-1 (Fig.5A). Since Evi-1 repressed the transactivation of the synthetic

FIG. 3. Evi-1 represses Smad1-, Smad2- and Smad3-responsive transcription. A, Evi-1 represses the transactivation function of Gal-fusedSmad1, -2, and -3. Gal-Smad1 transactivation of FR-Luc was stimulated by constitutively activated ALK-6/BMP-RIB, Gal-Smad2 was activatedby constitutively activated ActRIB, and Gal-Smad3 was activated by constitutively activated T�RI. B, the effect of Evi-1 on Smad1-responsivetranscription was assessed using Xvent-Luc as reporter in the presence or absence of cotransfected activated ALK-6/BMP-RIB receptor. C, theeffect of Evi-1 on Smad2 was assessed using the gsc-luc reporter and cotransfected FAST-1, with or without cotransfected activated ActRIB. D, theeffect of Evi-1 on Smad3/4-responsive transcription was assessed using the PAI1-Luc reporter with or without TGF-� (1 ng/ml).

Evi-1 Repression of TGF-� Family-inducible Transcription24232

by guest on September 10, 2020

http://ww

w.jbc.org/

Dow

nloaded from

(SBE)4 promoter with Smad binding elements similar to thosein the Smad7 promoter (Fig. 1A), we first assayed the tran-scription activation from the 0.5-kbp Smad7 promoter segmentin the Smad7-Luc reporter in the retrovirally infected C2C12cells. Consistent with previous reports (49, 50), TGF-�, BMP,and activin receptor signaling activated the Smad7 reporter incontrol C2C12 cells. The activation by each of the three ligandswas absent or much reduced in cells expressing Evi-1 (Fig. 5B).We then evaluated the regulation of endogenous Smad7 ex-

pression by Evi-1. Smad7 mRNA expression was induced byBMP-2 in control C2C12 cells, but this induction was blockedby retroviral Evi-1 expression in C2C12 cells (Fig. 5C). Simi-larly, TGF-� induced Smad7 mRNA expression 4.3-fold after3 h, and this induction of Smad7 mRNA by TGF-� was nearlyabsent in Evi-1-expressing cells (1.39-fold induction over un-treated) (Fig. 5D). Together, the data from reporter and endog-enous gene expression experiments in Xenopus and mamma-lian systems demonstrate that Evi-1 represses BMP, activin,

FIG. 4. Evi-1 represses activin/Smad2 and BMP/Smad1 signals in vivo in Xenopus ectodermal explants. A, Evi-1 inhibits mesodermaland endodermal marker induction by activin and BMP2. Whereas activin (2 pg; lane 3) and BMP2 (0.5 ng; lane 6) induce several mesendodermalmarkers in animal caps, coexpression of 1 ng (lanes 4 and 7) or 2 ng (lanes 5 and 8) of Evi-1 RNA suppresses the marker induction by these TGF-�family ligands. B, Evi-1 blocks the activity of Smad2 and Smad1 in Xenopus animal caps. Mesendodermal gene induction by 0.5 ng of Smad2 (lane3) or 4 ng of Smad1 (lane 6) RNAs is inhibited by the presence of 1 ng (lanes 4 and 7) or 2 ng (lanes 5 and 8) of Evi-1 RNA. In all of these experiments,RNAs were injected into the animal poles of two-cell stage frog embryos. Animal caps from the injected embryos were dissected at blastula stagesand harvested at gastrula stages for the RT-PCR assay. The gene expression in uninjected control animal caps is shown in lane 1, and theexpression pattern in whole embryos processed in the absence or presence of reverse transcriptase during the RT-PCR is shown in lanes 9 and 10,respectively.

FIG. 5. Evi-1 represses induction ofSmad7 expression by TGF-�, BMP,and activin. A, expression of Myc-taggedEvi-1 was evaluated by Western analysisof lysates from C2C12 stably infectedwith LPCX or LPCX-Evi-1. B, Evi-1 re-presses TGF-� (T202D), BMP, or activinreceptor-dependent transactivation of theSmad7 promoter linked to luciferasegene. C and D, Evi-1 represses BMP- andTGF-�-inducible endogenous Smad7mRNA expression at 2 h as assessed byreal time PCR.

Evi-1 Repression of TGF-� Family-inducible Transcription 24233

by guest on September 10, 2020

http://ww

w.jbc.org/

Dow

nloaded from

and TGF-� receptor signaling through its physical and func-tional interactions with Smad1, -2, and -3.

Evi-1 Binding to DNA Is Enhanced by Smad3 and CBP—The ability of Evi-1 to repress Smad transactivation whenDNA binding was kept constant (Figs. 1C and 3A) suggestedthat Evi-1 may employ alternative strategies to repress tran-scription, in addition to the previously proposed displace-ment of Smad binding to DNA (14). We therefore investigatedpossible mechanisms of Evi-1 repression of Smad transacti-vation. The binding of Evi-1 to the MH2 domain and theconsequent repression of the transactivation function ofSmad3 (Fig. 1C) raised the possibility that Evi-1 might re-press transcription by blocking the interaction of CBP withthe C-terminal sequence of Smad3. Moreover, Evi-1 has beenshown to interact with CBP (23), and this was confirmed incoimmunoprecipitation assays (data not shown). We there-fore tested whether increasing levels of Evi-1 would interferewith the Smad3 association with CBP using coimmunopre-cipitation assays (Fig. 6A). Surprisingly, Evi-1 enhanced the

interaction of Smad3 with CBP, suggesting stabilization ofthe Smad3-CBP complex. Conversely, increasing levels ofCBP also enhanced the interaction of Evi-1 with Smad3 (Fig.6B). These data suggest that Evi-1 does not repress Smadactivity by displacing the coactivator CBP but rather en-hances the stability of the Smad3/CBP interaction.

The ability of Evi-1 to repress transcription when Smadbinding to DNA was kept constant (Figs. 1C and 3A) and theEvi-1 interaction with the MH2 domain, but not the DNA-binding MH1 domain, of Smad3 (Fig. 1B) led us to test theeffect of Evi-1 on binding of Smad3 to the Smad binding ele-ment from the Smad7 promoter. At this promoter, Smads bindto the SBE sequence independently of other transcription fac-tors (51). DNA precipitation assays showed the formation of aSmad3-DNA complex upon TGF-� receptor activation. Coex-pression of Evi-1, in the absence or presence of coexpressedCBP, did not decrease the binding of Smad3 to DNA (Fig. 6C).Identical results were observed using electrophoretic mobilityshift assays (data not shown). CBP was also recruited to the

FIG. 6. Smad3 and CBP stabilize recruitment of Evi-1 to DNA. A and B, Cell lysates were prepared from COS cells transfected withexpression plasmids for tagged CBP, Evi-1, or Smad3 as indicated. The presence of proteins in FLAG or Myc immunoprecipitation (IP) complexesor in cell lysates (IB) was detected by Western analysis with the indicated antibodies. A, increasing Evi-1 levels stabilize the interaction of Smad3with CBP. B, increasing CBP levels stabilize the Smad3/Evi-1 interaction. C, DNA precipitation assays evaluated the binding of each cotransfectedprotein in cell lysates to biotinylated SBE oligonucleotide (left panel). Western blotting of whole cell lysates shows the expression level of eachtagged protein (right panel). D, enzymatically inactive CBP F1541A is as effective as wild-type CBP in facilitating recruitment of Evi-1 toDNA-bound Smad3.

Evi-1 Repression of TGF-� Family-inducible Transcription24234

by guest on September 10, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Smad-binding DNA sequence and also did not displace Smad3from DNA. Interestingly, Evi-1 binding to the Smad-bindingDNA sequence was only observed in the presence of coex-pressed CBP and Smad3. Thus, the repression of Smad3-me-diated transcription by Evi-1 does not result from a displace-ment of Smad3 from the DNA or from competitive interferencewith CBP association with Smad3. Instead, recruitment ofEvi-1 to the Smad-binding DNA sequence is enhanced bySmad3 and CBP.

The ability of the coactivator CBP to acetylate Evi-1 (23) andthe surprising result that CBP enhanced recruitment of Evi-1to the Smad3-binding DNA sequence led us to evaluate the roleof the histone acetyltransferase activity in this recruitment(Fig. 6D). We therefore compared the effects of CBP and a CBPpoint mutant, CBP (F1541A), that lacks acetyltransferase ac-tivity in DNA binding assays (30). Both wild-type and mutantCBP were equally effective in recruiting Evi-1 to the Smad3-binding DNA sequence, indicating that the acetyltransferaseactivity is dispensable for this function of CBP.

Evi-1 Recruits CtBP to the Smad7 Promoter to Block TGF-�-induced Histone Acetylation—We then used chromatin immu-noprecipitation to evaluate the recruitment of Evi-1 to theendogenous Smad7 promoter in response to TGF-�. In theseassays, we used the LPCX-Evi-1 C2C12 cells that expressedMyc-tagged Evi-1 and LPCX control cells that were also used inthe gene expression assays in Fig. 5. Chromatin associatedwith Myc-tagged Evi-1 was isolated, and the Evi-1-boundSmad7 promoter sequences were quantified using real-timePCR analysis. As shown in Fig. 7A, Evi-1 bound to the Smad7promoter at 60 min after the addition of TGF-�. No bindingabove the background, using a mouse IgG antibody, was ob-served in the absence of added TGF-� or in control C2C12 cellsthat do not express Myc-Evi-1. The binding of Evi-1 to theSmad7 promoter following TGF-� stimulation is consistent

with the TGF-�-dependent interaction of Smad3 with Evi-1(Fig. 2E) and the observation that Evi-1 only binds to SBEsequences in the presence of TGF-�-activated Smad3 (Fig. 6C).

Because Evi-1 interacts with CtBP and its associatedHDACs to repress transcription of reporter genes (20–22), weevaluated the effect of Evi-1 on histone acetylation of the en-dogenous Smad7 promoter by chromatin immunoprecipitationusing an antibody specific for acetylated histone H4 (Fig. 7B).In control LPCX-C2C12 cells, TGF-� stimulation conferred a3.2-fold increase in histone H4 acetylation at the Smad7 pro-moter after 2 h, consistent with the induction of Smad7 mRNAexpression by TGF-�. However, the induction of histone H4acetylation by TGF-� was reduced to 1.5-fold in Evi-1-express-ing cells.

We also examined the binding of endogenous CtBP to theSmad7 promoter by chromatin immunoprecipitation. LikeEvi-1 (Fig. 7A), CtBP is recruited to the Smad7 promoterwithin 60 min of TGF-� treatment, resulting in a more than60-fold increase in CtBP binding (Fig. 7C). CtBP binding issubsequently lost following release of Evi-1 from the Smad7promoter in LPCX-Evi-1 cells (data not shown). We also foundthat, in the LPCX cells that do not express Evi-1, the CtBP thatbinds to the Smad7 promoter in the absence of TGF-� is rapidlyreleased upon TGF-� treatment (Fig. 7C, inset). These resultssuggest that the TGF-�-induced interaction of Evi-1 with theSmad7 promoter stabilizes and enhances CtBP recruitment.

It has previously been shown that mutation of the CtBPinteraction domain of Evi-1 partially reverses the repressionexerted by Evi-1 on TGF-�-induced transcription from the 3TPreporter. A similar, partial rescue was observed using tricho-statin A, an inhibitor of HDACs (20). Whereas these observa-tions implicate CtBP and associated HDACs in the repressionof TGF-�-induced gene expression by Evi-1, they also revealedthe role of additional repression domains in Evi-1 that are

FIG. 7. Evi-1 and CtBP binding to the Smad7 promoter prevents TGF-� induced histone acetylation. Myc (9E10) and acetyl histone4 chromatin immunoprecipitation of the endogenous Smad7 promoter in LPCX and LPCX-Evi-1 C2C12 stable cells showed TGF-�-dependentrecruitment of Evi-1 (A) and Evi-1 inhibition of TGF-�-inducible histone acetylation (B). C, chromatin immunoprecipitation revealed recruitmentof CtBP to the Smad7 promoter in Evi-1-expressing cells following TGF-� stimulation. Chromatin immunoprecipitation data are normalized foreach cell line to the untreated control. D, reduction of CtBP protein levels by stable expression of small interfering RNA for CtBP (left panel)reduced the repression of TGF-�-induced Smad7 mRNA expression by Evi-1 (right panel). Results were quantified using real time PCR. IB,immunoblot.

Evi-1 Repression of TGF-� Family-inducible Transcription 24235

by guest on September 10, 2020

http://ww

w.jbc.org/

Dow

nloaded from

CtBP- and HDAC-independent. To test the role of CtBP in therepression of TGF-�-induced, endogenous Smad7 mRNA ex-pression, we utilized small interfering RNA to reduce CtBPprotein levels (Fig. 7D, left panel). Although we achieved only apartial reduction of CtBP expression, it considerably decreasedthe repression of TGF-�-inducible Smad7 gene expression byEvi-1 (Fig. 7D, right panel). This partial reduction in repressionwas consistent with the partial reversal of the Evi-1-mediatedrepression by trichostatin A or in the presence of the mutantEvi-1 that lacks CtBP binding (20). Thus, consistent with thesereported findings (20), our data suggest that CtBP-dependentand -independent mechanisms are operative in Evi-1-mediatedrepression of TGF-�-inducible Smad7 expression.

DISCUSSION

Based on the reported physical interaction of Evi-1 with thehighly conserved MH2 domain of Smad3 (14), we investigatedthe interaction of Evi-1 with Smads 1–4. Evi-1 interacted witheach of the Smads tested, albeit with differing efficiencies de-pending on the assays. Consistent with these physical interac-tions, Evi-1 repressed Smad1- and Smad2-mediated transacti-vation by BMP and activin receptors, respectively. Therefore,Evi-1 repression is not restricted to Smad3 but affects Smad-mediated transactivation in response to other TGF-�-relatedproteins, such as BMP and activin.

The ability of Evi-1 to repress BMP and activin signalingraises the possibility that Evi-1 may function as a modulatorof Smad signaling in development. BMPs, activin, and re-lated TGF-� proteins are well known regulators of embryonicpatterning and cell fate specification (46). Deficiencies inBMP and activin signaling, analyzed by targeted gene inac-tivation, result in embryonic lethal phenotypes with mouseembryos possessing severe defects in multiple organs (1).Likewise, mice lacking Evi-1 die in utero with widespreadhypocellularity and multiple defects in various tissues (16,17) that partially overlap with those in mice that lack signal-ing molecules in TGF-� family pathways. In addition, thedevelopmental expression pattern of Evi-1 overlaps partiallywith that of TGF-�, activin, or BMP signaling components(16, 17). These data suggest that Evi-1 may regulate TGF-�-related signaling during early vertebrate embryogenesis,consistent with the direct repression of Smad1- and Smad2-stimulated induction of endogenous genes required for mesend-odermal cell fate specification by BMP and activin in Xenopusembryo assays (Fig. 4). Thus, in addition to the role of Evi-1 inthe repression of TGF-�/Smad3-mediated growth inhibition intumorigenesis (14), we propose a role for Evi-1 as a regulator ofSmad-mediated signaling during development.

We also showed that Evi-1 represses endogenous Smad7expression, the first demonstration that Evi-1 inhibits TGF-�family-induced expression of an endogenous gene. TGF-�,activin, and BMPs induce Smad7 expression through Smad-mediated transcription, and Smad7 inhibits effector Smadactivation, thus allowing Smad7 to exert a negative feedbackeffect on TGF-� signaling. Accordingly, ectopic Smad7 ex-pression in Xenopus explants inhibits mesoderm induction byactivin and ventralizing signals by BMPs in a dose-depend-ent manner (52, 53). The regulation of Smad7 expression byEvi-1, as shown in this report, may therefore have importantconsequences in development. Since Evi-1 represses Smadsignaling directly, and indirectly through inhibition of Smad7expression, the graded expression of inhibitory and inductivesignals (Evi-1, Smad7, and BMP and activin ligands) maycontribute to cell fate specification and differentiation duringembryonic development.

Our results also provide insight into the mechanism of Evi-1-mediated repression of effector Smads. Transcriptional re-

pression can result from several strategies. These include dis-placement of the DNA binding of a transcriptional activator,interference with coactivator recruitment, and direct func-tional repression through recruitment of HDACs. Based onstudies of the p15Ink4B or 3TP promoters, where Smad3 isrecruited to DNA by its interactions with Sp1 or c-Jun/c-Fos,respectively (11, 42, 43), Evi-1 was originally proposed to re-press TGF-�/Smad3-activated transcription by displacingSmad3 from DNA (14). However, we found that Evi-1 repressedSmad function even when DNA binding was kept constantusing a Gal4 DNA binding domain (Figs. 1C and 3A). We alsofound that Evi-1 repressed Smad3-mediated transactivation ofthe (SBE)4 promoter, where DNA binding is mediated only bythe Smad3 MH1 domain (34) instead of through the interactionof Smad3 with other sequence-specific transcription factors. Atthis promoter, Evi-1 did not displace Smad3 from the SBE DNAsequence but rather was recruited through its interaction withSmad3 to the SBE sequence (Fig. 6C). Therefore, Evi-1 usesalternative strategies to repress Smad transactivation.

Some reports provide evidence that competition between co-activators and corepressors for interaction with a sequence-specific transcription factor defines the level of transcription.For example, the coactivators GRIP, CBP, or p300 competewith the corepressor SMRT for binding to the same domain ofthe transcription factor HNF4� (54). Since Evi-1 interactedwith the MH2 domain of Smad3, which directly binds to thecoactivators CBP or p300 (29), we evaluated whether competi-tion of Evi-1 with CBP/p300 for Smad3 binding might contrib-ute to Evi-1-mediated repression. Contrary to our hypothesis,interactions of Evi-1 and CBP with Smad3 were not mutuallyexclusive; rather, CBP enhanced the Smad3 interaction withEvi-1, and Evi-1 enhanced Smad3 interaction with CBP. Ourdata suggest that physical interactions between Smad3, CBP,and Evi-1 allow formation of a stable complex that is presum-ably the basis for our observation that efficient recruitment ofEvi-1 to a Smad-binding DNA sequence required the partici-pation of Smad3 and CBP.

The participation of Evi-1 and CBP in the Smad-DNA com-plex suggests the involvement of acetylation and deacetylationactivities. CBP has intrinsic acetylation activity (55), whereasEvi-1 recruits CtBP (20, 21), which in turn interacts withHDACs (22). Acetylation of transcription factors has beenshown to regulate coactivator and corepressor recruitment. Forexample, acetylation of MyoD increases its affinity for coacti-vators (56), whereas acetylation of the orphan nuclear receptorRIP140 prevents corepressor recruitment (57). Regulatedacetylation may also impact Evi-1 function, since Evi-1 hasbeen proposed to be acetylated by CBP, P/CAF, and GCN5 (23,58). The acetylase activity of CBP activity is, however, notessential for efficient recruitment of Evi-1 to the Smad3-bind-ing DNA sequence, since both wild-type CBP and the acetylase-deficient CBP mutant are equally effective in recruiting Evi-1to DNA. This is similar to a recent observation that p300 canrecruit HDAC6 independently of its acetylase activity (59).Thus, CBP/p300 may act as a scaffold that stabilizes corepres-sor complex recruitment.

Whereas the acetylase function of CBP is not required forEvi-1 recruitment to the Smad-binding DNA sequence, TGF-�-induced acetylation is observed at the Smad7 promoter (Fig.7B), presumably a result of the recruitment of CBP by Smad3(29). In cells expressing Evi-1, however, Smad3-mediated re-cruitment of Evi-1 to the promoter resulted in decreased his-tone acetylation (Fig. 7B), probably due to the co-recruitment ofCtBP and associated HDACs with Evi-1 in response to TGF-�(Fig. 7C) (20, 21). Reduction of CtBP levels, through the use ofsmall interfering RNA, decreased the repression of TGF-�-

Evi-1 Repression of TGF-� Family-inducible Transcription24236

by guest on September 10, 2020

http://ww

w.jbc.org/

Dow

nloaded from

induced Smad7 expression by Evi-1. This observation suggeststhat reduced histone acetylation, concomitant with Evi-1 andCtBP recruitment, is an important determinant in the repres-sion of TGF-�-induced Smad7 expression by Evi-1. However,consistent with previous findings (20, 21), our results (Fig. 7D)additionally implicate CtBP-independent mechanisms in therepression of TGF-�/Smad3 signaling by Evi-1.

In conclusion, our results demonstrate that Evi-1 interactswith not only Smad3 but with all Smads tested, thus allowingEvi-1 to repress BMP-, activin-, and TGF-�-induced, Smad-mediated transcription. Upon TGF-� stimulation, Evi-1 andCtBP are recruited to the Smad7 promoter, where they inhibitTGF-�-induced histone acetylation and transcription activa-tion. This is the first report of Evi-1 recruitment to and repres-sion of an endogenous TGF-� inducible gene. Our in vitrostudies suggest that efficient Evi-1 recruitment requiresSmad3 and is stabilized by CBP. The repression of BMP/Smad1- and activin/Smad2-activated gene expression by Evi-1in Xenopus embryo explant assays suggests that the repressionof Smad function by Evi-1 may have a role in development inaddition to its previously proposed role in tumorigenesis.

Acknowledgments—We thank H. Hirai, R. Goodman, J. Massague,M.G. Rosenfeld, R. Tjian, K. Yamamoto, T. Kouzarides, M. Whitman,L. Attisano, J. Wrana, B. Vogelstein, C. Niehrs, J. Ihle,C. Bartholomew, Y. Shi, and K. Cho for providing essential plasmidsand Y. Zhang for critical review of the manuscript.

REFERENCES

1. Chang, H., Brown, C. W., and Matzuk, M. M. (2002) Endocr. Rev. 23, 787–8232. Siegel, P. M., and Massague, J. (2003) Nat. Rev. Cancer 3, 807–8213. Derynck, R., Akhurst, R. J., and Balmain, A. (2001) Nat. Genet. 29, 117–1294. Wakefield, L. M., and Roberts, A. B. (2002) Curr. Opin. Genet. Dev. 12, 22–295. Feng, X. H., Liang, Y. Y., Liang, M., Zhai, W., and Lin, X. (2002) Mol. Cell 9,

133–1436. Seoane, J., Le, H. V., and Massague, J. (2002) Nature 419, 729–7347. Sun, Y., Liu, X., Eaton, E. N., Lane, W. S., Lodish, H. F., and Weinberg, R. A.

(1999) Mol. Cell 4, 499–5098. Derynck, R., and Zhang, Y. E. (2003) Nature 425, 577–5849. ten Dijke, P., and Hill, C. S. (2004) Trends Biochem. Sci. 29, 265–273

10. Attisano, L., and Wrana, J. L. (2002) Science 296, 1646–164711. Feng, X. H., Lin, X., and Derynck, R. (2000) EMBO J. 19, 5178–519312. Pardali, K., Kurisaki, A., Moren, A., ten Dijke, P., Kardassis, D., and Mousta-

kas, A. (2000) J. Biol. Chem. 275, 29244–2925613. Moustakas, A., Souchelnytskyi, S., and Heldin, C. H. (2001) J. Cell Sci. 114,

4359–436914. Kurokawa, M., Mitani, K., Irie, K., Matsuyama, T., Takahashi, T., Chiba, S.,

Yazaki, Y., Matsumoto, K., and Hirai, H. (1998) Nature 394, 92–9615. Hirai, H. (1999) Int. J. Biochem. Cell Biol. 31, 1367–137116. Hoyt, P. R., Bartholomew, C., Davis, A. J., Yutzey, K., Gamer, L. W., Potter,

S. S., Ihle, J. N., and Mucenski, M. L. (1997) Mech. Dev. 65, 55–7017. Perkins, A. S., Mercer, J. A., Jenkins, N. A., and Copeland, N. G. (1991)

Development 111, 479–48718. Morishita, K., Parganas, E., Matsugi, T., and Ihle, J. N. (1992) Mol. Cell. Biol.

12, 183–18919. Valk, P. J., Verhaak, R. G., Beijen, M. A., Erpelinck, C. A., Barjesteh van

Waalwijk van Doorn-Khosrovani, S., Boer, J. M., Beverloo, H. B., Moor-house, M. J., van der Spek, P. J., Lowenberg, B., and Delwel, R. (2004)N. Engl. J. Med. 350, 1617–1628

20. Izutsu, K., Kurokawa, M., Imai, Y., Maki, K., Mitani, K., and Hirai, H. (2001)Blood 97, 2815–2822

21. Palmer, S., Brouillet, J. P., Kilbey, A., Fulton, R., Walker, M., Crossley, M.,and Bartholomew, C. (2001) J. Biol. Chem. 276, 25834–25840

22. Chinnadurai, G. (2002) Mol. Cell 9, 213–22423. Chakraborty, S., Senyuk, V., Sitailo, S., Chi, Y., and Nucifora, G. (2001) J. Biol.

Chem. 276, 44936–4494324. Graycar, J. L., Miller, D. A., Arrick, B. A., Lyons, R. M., Moses, H. L., and

Derynck, R. (1989) Mol. Endocrinol. 3, 1977–198625. Choy, L., Skillington, J., and Derynck, R. (2000) J. Cell Biol. 149, 667–68226. Feng, X. H., and Derynck, R. (1996) J. Biol. Chem. 271, 13123–1312927. Zhang, Y., and Derynck, R. (2000) J. Biol. Chem. 275, 16979–1698528. Liu, F., Hata, A., Baker, J. C., Doody, J., Carcamo, J., Harland, R. M., and

Massague, J. (1996) Nature 381, 620–62329. Feng, X. H., Zhang, Y., Wu, R. Y., and Derynck, R. (1998) Genes Dev. 12,

2153–216330. Martinez-Balbas, M. A., Bannister, A. J., Martin, K., Haus-Seuffert, P., Meis-

terernst, M., and Kouzarides, T. (1998) EMBO J. 17, 2886–289331. Hoodless, P. A., Haerry, T., Abdollah, S., Stapleton, M., O’Connor, M. B.,

Attisano, L., and Wrana, J. L. (1996) Cell 85, 489–50032. Attisano, L., Wrana, J. L., Montalvo, E., and Massague, J. (1996) Mol. Cell.

Biol. 16, 1066–107333. Keeton, M. R., Curriden, S. A., van Zonneveld, A. J., and Loskutoff, D. J. (1991)

J. Biol. Chem. 266, 23048–2305234. Zawel, L., Dai, J. L., Buckhaults, P., Zhou, S., Kinzler, K. W., Vogelstein, B.,

and Kern, S. E. (1998) Mol. Cell 1, 611–61735. Candia, A. F., Watabe, T., Hawley, S. H., Onichtchouk, D., Zhang, Y., Derynck,

R., Niehrs, C., and Cho, K. W. (1997) Development 124, 4467–448036. Watabe, T., Kim, S., Candia, A., Rothbacher, U., Hashimoto, C., Inoue, K., and

Cho, K. W. (1995) Genes Dev. 9, 3038–305037. Nagarajan, R. P., Zhang, J., Li, W., and Chen, Y. (1999) J. Biol. Chem. 274,

33412–3341838. Feng, X. H., Filvaroff, E. H., and Derynck, R. (1995) J. Biol. Chem. 270,

24237–2424539. Shum, L., Reeves, S. A., Kuo, A. C., Fromer, E. S., and Derynck, R. (1994)

J. Cell Biol. 125, 903–91640. Zhang, Y., Feng, X., We, R., and Derynck, R. (1996) Nature 383, 168–17241. Chang, C., Wilson, P. A., Mathews, L. S., and Hemmati-Brivanlou, A. (1997)

Development 124, 827–83742. Liberati, N. T., Datto, M. B., Frederick, J. P., Shen, X., Wong, C., Rougier-

Chapman, E. M., and Wang, X. F. (1999) Proc. Natl. Acad. Sci. U. S. A. 96,4844–4849

43. Zhang, Y., Feng, X. H., and Derynck, R. (1998) Nature 394, 909–91344. Feng, X. H., and Derynck, R. (2001) Methods Mol. Biol. 177, 221–23945. Labbe, E., Silvestri, C., Hoodless, P. A., Wrana, J. L., and Attisano, L. (1998)

Mol. Cell 2, 109–12046. Harland, R., and Gerhart, J. (1997) Annu. Rev. Cell Dev. Biol. 13, 611–66747. Su, M. W., Suzuki, H. R., Solursh, M., and Ramirez, F. (1991) Development

111, 1179–118748. Suzuki, A., Ueno, N., and Hemmati-Brivanlou, A. (1997) Development 124,

3037–304449. Afrakhte, M., Moren, A., Jossan, S., Itoh, S., Sampath, K., Westermark, B.,

Heldin, C. H., Heldin, N. E., and ten Dijke, P. (1998) Biochem. Biophys. Res.Commun. 249, 505–511

50. Ishisaki, A., Yamato, K., Hashimoto, S., Nakao, A., Tamaki, K., Nonaka, K.,ten Dijke, P., Sugino, H., and Nishihara, T. (1999) J. Biol. Chem. 274,13637–13642

51. Stopa, M., Anhuf, D., Terstegen, L., Gatsios, P., Gressner, A. M., and Dooley,S. (2000) J. Biol. Chem. 275, 29308–29317

52. Bhushan, A., Chen, Y., and Vale, W. (1998) Dev. Biol. 200, 260–26853. Casellas, R., and Brivanlou, A. H. (1998) Dev. Biol. 198, 1–1254. Torres-Padilla, M. E., Sladek, F. M., and Weiss, M. C. (2002) J. Biol. Chem.

277, 44677–4468755. Bannister, A. J., and Kouzarides, T. (1996) Nature 384, 641–64356. Polesskaya, A., and Harel-Bellan, A. (2001) J. Biol. Chem. 276, 44502–4450357. Vo, N., Fjeld, C., and Goodman, R. H. (2001) Mol. Cell. Biol. 21, 6181–618858. Senyuk, V., Sinha, K. K., Chakraborty, S., Buonamici, S., and Nucifora, G.

(2003) Biochem. Biophys. Res. Commun. 307, 980–98659. Ling, L., and Lobie, P. E. (2004) J. Biol. Chem. 279, 32737–3275060. Morishita, K., Parganas, E., Douglass, E. C., and Ihle, J. N. (1990) Oncogene 5,

963–97161. Shi, Y., Sawada, J., Sul, G., Affar, E. B., Whetstine, J. R., Lan, F., Ogawa, H.,

Luke, M. P., Nakatani, Y., and Shi, Y. (2003) Nature 422, 735–73862. Lin, X., Liang, Y. Y., Sun, B., Liang, M., Shi, Y., Brunicardi, F. C., Shi, Y., and

Feng, X. H. (2003) Mol. Cell. Biol. 23, 9081–909363. Matsugi, T., Morishita, K., and Ihle J. N. (1990) Mol. Cell. Biol. 10, 1259–1264

Evi-1 Repression of TGF-� Family-inducible Transcription 24237

by guest on September 10, 2020

http://ww

w.jbc.org/

Dow

nloaded from

and Rik DerynckTamara Alliston, Tien C. Ko, Yanna Cao, Yao-Yun Liang, Xin-Hua Feng, Chenbei Chang

Evi-1Repression of Bone Morphogenetic Protein and Activin-inducible Transcription by

doi: 10.1074/jbc.M414305200 originally published online April 22, 20052005, 280:24227-24237.J. Biol. Chem. 

  10.1074/jbc.M414305200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/280/25/24227.full.html#ref-list-1

This article cites 63 references, 32 of which can be accessed free at

by guest on September 10, 2020

http://ww

w.jbc.org/

Dow

nloaded from