the j b c vol. 275, no. 38, issue of september 22, pp ... · role of smad proteins and...

Role of Smad Proteins and Transcription Factor Sp1 in p21Waf1/Cip1

Regulation by Transforming Growth Factor-b*

Received for publication, November 30, 1999, and in revised form, May 25, 2000Published, JBC Papers in Press, June 30, 2000, DOI 10.1074/jbc.M909467199

Katerina Pardali, Akira Kurisaki‡, Anita Moren, Peter ten Dijke§, Dimitris Kardassis¶, andAristidis Moustakasi

From the Ludwig Institute for Cancer Research, Box 595, SE-751 24 Uppsala, Sweden, §The Netherlands Cancer Institute,Division of Cellular Biochemistry, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands, and ¶Department of BasicSciences, University of Crete Medical School and Institute of Molecular Biology and Biotechnology, F.O.R.T.H.,Heraklion GR-711 10, Greece

Transforming growth factor-b (TGF-b) inhibits cell cy-cle progression, in part through up-regulation of geneexpression of the p21WAF1/Cip1 (p21) cell cycle inhibitor.Previously we have reported that the intracellular effec-tors of TGF-b, Smad3 and Smad4, functionally cooperatewith Sp1 to activate the human p21 promoter in hepa-toma HepG2 cells. In this study we show that Smad3 andSmad4 when overexpressed in HaCaT keratinocyteslead to activation of the p21 promoter. Activation re-quires the binding sites for the ubiquitous transcriptionfactor Sp1 on the proximal promoter. Induction of theendogenous HaCaT p21 gene by TGF-b1 is further en-hanced after overexpression of Smad3 and Smad4,whereas dominant negative mutants of Smad3 andSmad4 and the inhibitory Smad7 all inhibit p21 induc-tion by TGF-b1 in a dose-dependent manner. We showthat Sp1 expressed in the Sp1-deficient Drosophila SL-2cells binds to the proximal p21 promoter sequences,whereas Smad proteins do not. In support of this find-ing, we show that DNA-binding domain mutants ofSmad3 and Smad4 are capable of transactivating the p21promoter as efficiently as wild type Smads. Co-expres-sion of Smad3 with Smad4 and Sp1 in SL-2 cells or co-incubation of phosphorylated Smad3, Smad4, and Sp1 invitro results in enhanced binding of Sp1 to the p21 prox-imal promoter sequences. We demonstrate that Sp1physically and directly interacts with Smad2, Smad3,and weakly with Smad4 via their amino-terminal (Mad-Homology 1) domain. Finally, by using GAL4 fusion pro-teins we show that the glutamine-rich sequences in thetransactivation domain of Sp1 contribute to the cooper-ativity with Smad proteins. In conclusion, Smad pro-teins play important roles in regulation of the p21 geneby TGF-b, and the functional cooperation of Smad pro-teins with Sp1 involves the physical interaction of thesetwo types of transcription factors.

Transforming growth factor-b (TGF-b)1 is the prototype of afamily of multifunctional cytokines that regulate many aspectsof cell physiology, including cell growth, differentiation, motil-ity and death, and play important roles in many developmentaland pathological processes (1, 2). TGF-b inhibits cell prolifer-ation of various cell types of epithelial origin by repressing theexpression of the proto-oncogene c-myc and by inhibiting theactivity of cyclin-dependent kinases (CDKs) which leads to thearrest of the cell cycle at an early G1 phase (3–6). The mecha-nism of action of TGF-b in modulating the activity of the CDKsinvolves the regulation of the CDK inhibitors (CKIs) p15Ink4B,p21Waf1/Cip1, and p27Kip1 and the repression of the CDK phos-phatase cdc25A (4, 7–10). This regulation consists of both tran-scriptional induction of the genes for p15 and p21 and partition-ing of the CKIs between different complexes with CDKs (4, 7, 8).Thus, the transcriptional induction of the genes for the two CKIs,p15 and p21, has been postulated at least as partially determin-ing the anti-proliferative action of TGF-b (11).

The p21 gene is regulated by a rapidly growing list of phys-iological and pathological factors, such as tumor suppressors ofthe p53 family, differentiation factors, growth factors, cyto-kines, and stress factors (Ref. 12 and references therein). How-ever, the detailed transcriptional mechanisms involved in p21gene regulation by the above factors still remain poorly under-stood. In the cases of p53, vitamin D3, interferon g, and othersignals, factor-specific DNA motifs scattered in the region be-tween 22,300 and 2210 base pairs upstream from the tran-scriptional initiation site of the p21 gene have been shown tomediate the response of this gene to the above stimuli (13–15).In contrast, for a large number of other signaling factors suchas, TGF-b1, progesterone, phorbol esters, and Rho GTPases,the proximal region of the p21 promoter (base pairs 2210 to 11)is the major site for reception of the inducing signal (Ref. 12 andreferences therein). This short, proximal promoter contains sev-eral closely spaced G/C-rich motifs that serve as binding sites formembers of the Sp1 family of transcription factors (16).

Sp1 is a ubiquitously expressed transcription factor with azinc finger DNA-binding domain that recognizes G/C-rich DNA

* This research was supported in part by a grant from the HumanFrontier Science Program (to D. K. and A. M.), Institute of MolecularBiology and Biotechnology for internal funds (to D. K.), and a grantfrom the Dutch Cancer Society (to P. t. D.). The costs of publication ofthis article were defrayed in part by the payment of page charges. Thisarticle must therefore be hereby marked “advertisement” in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.

‡ Recipient of a STINT post-doctoral fellowship from the SwedishFoundation for International Cooperation in Research and HighEducation.

i To whom correspondence should be addressed: Ludwig Institute forCancer Research, Box 595, SE-751 24 Uppsala, Sweden. Tel.: 46-18-160411/2; Fax: 46-18-160420; E-mail: [email protected].

1 The abbreviations used are: TGF-b, transforming growth factor b;CKI, cyclin-dependent kinase inhibitor; Smad, Sma and Mad-relatedprotein; MH, Mad-Homology; SBE, Smad-binding element, ALK, ac-tivin receptor-like kinase; CA, constitutively active; HaCaT, humankeratinocyte cell line; HepG2, human hepatoma cell line; m.o.i., multi-plicity of infection; GEMSA, gel electrophoretic mobility shift assay;PAGE, polyacrylamide gel electrophoresis; CAT, chloramphenicolacetyltransferase; GST, glutathione S-transferase; DMEM, Dulbecco’smodified Eagle medium; PMSF, phenylmethylsulfonyl fluoride; CDKs,cyclin-dependent kinases; HA, hemagglutinin; DN, dominant negative;DTT, dithiothreitol.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 275, No. 38, Issue of September 22, pp. 29244–29256, 2000© 2000 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.org29244

by guest on August 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

sequences (17, 18). Sp1 is required for early embryogenesis andregulates the terminal differentiation state of cells by affectingthe methylation of DNA CpG islands (19). The transcriptionalactivity of Sp1 is regulated by phosphorylation in a cell cycle-specific manner, acetylation by the co-activator p300, and gly-cosylation, which protects this factor from proteasome-depend-ent degradation (20–22). Sp1 exerts its transcriptional propertiesby interacting directly with factors of the basal transcriptionmachinery and by cooperating with several transcriptional ac-tivators (23–29). Thus, although Sp1 traditionally appeared asa ubiquitous factor primarily serving the core activity of pro-moters, recent evidence increasingly implicates this protein inseveral instances of regulated gene transcription.

It is established that transcription factor Sp1 participates inthe regulation of the p21 gene by TGF-b (16, 30). The TGF-bsignaling pathway utilizes plasma membrane serine/threoninekinase receptors and their cytoplasmic effectors termed theSmad proteins (2, 31). Smad proteins are transcriptional acti-vators that bind to DNA and cooperate with a fast growing listof transcription factors in regulating target gene expression(32–34). In a previous report (35) we have provided evidence forthe role of Smad3 and Smad4 in mediating the induction of thep21 gene by TGF-b1. The Smads were shown to cooperatefunctionally with Sp1 and depend on the Sp1-binding sites ofthe p21 promoter for their action. In addition, we have demon-strated that Jun family members, which themselves are in-duced and activated by TGF-b, can also regulate the p21 pro-moter by physically interacting with Sp1 and utilizing thesame G/C-rich motifs of the proximal promoter (12). In thepresent work we demonstrate that Smad3 and Smad4 enhancethe level of endogenous HaCaT p21 gene induction by TGF-b,whereas dominant negative Smads and the inhibitory Smad7block p21 induction by TGF-b in a dose-dependent manner.These results correlate very well with the p21 promoter trans-activation studies in the same cells. We show that Smadsmediate enhancement of the Sp1 affinity for the p21 promoter,independent from a direct association of Smads to DNA, andSmad2, Smad3, and Smad4 physically interact with Sp1. Wehave mapped the domain of Smad3 and Smad4 required forthis interaction and provide evidence for the involvement ofspecific Sp1 sequences in the functional cooperation betweenthese two classes of transcription factors.

EXPERIMENTAL PROCEDURES

Materials—The purified baculoviral Smad3, TGF-b type I receptor-phosphorylated Smad3 and Smad4 proteins were a generous gift fromF. M. Hoffman and A. Comer (36). Restriction enzymes and modifyingenzymes (T4 DNA ligase, T4 polynucleotide kinase, Klenow fragment ofDNA polymerase I, and calf intestinal alkaline phosphatase) were pur-chased from Minotech, New England Biolabs, or Life Technologies, Inc.Vent DNA polymerase was from New England Biolabs. The Sequenaseversion 2 kit, poly(dI/dC), acetyl-CoA, dNTPs, protein A, protein G-Sepharose beads, and the GST purification kit were from AmershamPharmacia Biotech. Isopropyl-b-D-thiogalactopyranoside was from Cal-biochem. [g-32P]ATP, [a-32P]dCTP, and [14C]chloramphenicol were fromAmersham Pharmacia Biotech or NEN Life Science Products. All re-agents for cell culture (Dulbecco’s modified Eagle medium (DMEM),fetal bovine serum, trypsin-EDTA, and phosphate-buffered saline) werefrom Life Technologies, Inc. O-Nitrophenyl galactopyranoside and themonoclonal anti-FLAG (M5, F-4042) antibody were from Sigma. Thetransfection reagent Fugene-6 and the monoclonal anti-hemagglutinin(HA) (12CA5) antibody were from Roche Molecular Biochemicals. Theluciferase assay system, the consensus Sp1 oligonucleotide, and puri-fied Sp1 protein were from Promega. All oligonucleotides were synthe-sized at the microchemical facility of the IMBB, Heraklion, Greece. Themonoclonal anti-Myc antibody was produced by the 9E10 hybridomacell clone. The monoclonal anti-GAL4 DNA-binding domain antibody(RK5C1), the monoclonal anti-Smad1/2/3 (H-2) antibody, the rabbitpolyclonal anti-Sp1 (PEP-2) antibody, and the polyclonal anti-hexahis-tidine (His-probe H-15) antibody were from Santa Cruz Biotechnology.The monoclonal anti-b-catenin (C19220) and the monoclonal anti-p21

(C24420) antibodies were from Transduction Laboratories. The anti-phosphoserine rabbit polyclonal antibody (Poly-Z-PS1) was from ZymedLaboratories Inc. The anti-mouse horseradish peroxidase-conjugatedsecondary antibody was from Amersham Pharmacia Biotech. All otherchemicals were obtained from the usual commercial sources at thepurest grade available.

Plasmid Constructions—The p21 promoter plasmid 22,300/18 p21luc has been described previously (35). The p21 promoter deletionconstruct 2143/18 p21 luc was constructed by transferring the XbaI toHindIII promoter fragment from the 2143/18 p21 CAT plasmid (35) topGL2-basic after digestion with NheI and HindIII. The expressionvectors pCDNA3–6myc-Smad2, pCDNA3–6myc-Smad3, pCDNA3–6myc-Smad4, and pCDNA3-HA-CA-ALK5 and the reporter constructpGL3-CAGA12-MLP-luc were generously provided by Dr. S. Itoh of theLudwig Institute, Uppsala, Sweden. The expression vector pCDNA1/amp encoding the FLAG-tagged human Smad3 was described previ-ously (12). The expression vector encoding the FLAG-tagged Smad4protein was the generous gift of Dr. J. Massague (Memorial Sloan-Kettering Institute, New York). The expression vectors encoding thethree deletion mutants of Smad3, Smad3-(1–122), Smad3-(1–248), andSmad3-(122–424) were constructed by polymerase chain reaction am-plification of the specified fragments using appropriate primers andsubsequent subcloning of the amplified fragments into the expressionvector pCDNA1/amp. The expression vectors encoding the FLAG-tagged human Smad3 with the double mutation (R74K and K81R) andthe FLAG-tagged human Smad4 with the double mutation (R81K andK88R) in the Smad DNA-binding domain will be described in detailelsewhere.2 The GAL4(DBD)-Sp1 fusion constructs pSG424/GAL4-Sp1A1B, pSG424/GAL4-Sp1B, pSG424/GAL4-Sp1Bn, pSG424/GAL4-Sp1Bc, and the pBXG1 plasmid containing the GAL4 DBD portion onlywere the generous gifts of Dr. G. Gill, Harvard Medical School, Boston.The pG5B-CAT reporter containing five tandem GAL4-binding sites infront of the E1B minimal promoter and the CAT reporter gene was thegenerous gift of Dr. G. Mavrothalassitis, University of Crete MedicalSchool, Heraklion, Greece. The bacterial expression vectors pGEX-Sp1(GST-Sp1 83–778), pGEX-Sp1 516C (GST-Sp1 DA), pGEX-Sp1 N619(GST-Sp1 DD), and pGEX-Sp1 Dint 349 (GST-Sp1 DB1C) were thegenerous gifts of Dr. E. Flavey, Section of Molecular Genetics, BostonUniversity Medical Center, Boston. The original Sp1 mutants were thegenerous gifts of Dr. R. Tjian, University of California, Berkeley. Thebacterial expression vectors pGEX-Smad3, pGEX-Smad3DMH1, pGEX-Smad3DMH2, pGEX-Smad3MH1, pGEX-Smad3MH2, pGEX-Smad3-Linker, pGEX-Smad4, and pGEX-Smad4DMH2 were the generous giftsof Dr. S. Itoh of the Ludwig Institute, Uppsala, Sweden. The Drosophilaexpression vectors pPac-Sp1 and pPacO were the generous gifts of Dr.J. M. Horowitz, North Carolina State University, Raleigh, and J. Noti,Guthrie Research Institute, Sayre, PA, respectively. The Drosophilaexpression vectors pRactH-Smad3 and pRactH-Smad4 were con-structed by transferring the corresponding Smad cDNA that includesonly the protein-coding region from pCDNA1/amp-Smad3 andpCDNA3-Smad4, respectively, into the BamHI and HindIII sites of thepolycloning region of the basic Drosophila expression vector pRactH.The pRactH and hsp-lacZ expression vector used for normalization oftransfections in Drosophila SL2 cells were the generous gifts of Dr. C.Delidakis, University of Crete, and IMBB, Heraklion, Greece. The qual-ity of all new DNA constructs was verified by DNA sequencing.

Cell Cultures, Transient Transfections, Adenoviral Infections, Re-porter, and Western Blot Assays—Human HaCaT keratinocytes, humanhepatoma HepG2 cells, and monkey kidney COS-7 cells were culturedin DMEM supplemented with 10% fetal bovine serum, L-glutamine, andpenicillin/streptomycin at 37 °C in a 5% CO2 atmosphere. DrosophilaSchneider’s SL2 cells were cultured in Schneider’s insect medium sup-plemented with 10% insect culture-tested fetal bovine serum and pen-icillin/streptomycin at 27 °C. Transient transfections of COS-7 cells forco-immunoprecipitation assays and of Schneider’s SL2 cells for nuclearextract isolation were performed using the liposome reagent Fugene-6according to the manufacturer’s protocol (Roche Molecular Biochemi-cals). Transient transfections using the calcium phosphate co-precipi-tation method, chloramphenicol acetyltransferase, luciferase, and b-ga-lactosidase assays were performed as described previously (35).

Adenoviral stocks were maintained, and infections were performedas described previously (37). The dominant negative (DN) Smad3 andDN-Smad4-encoding adenoviruses were the generous gifts of Dr. The-odore Fotsis, University of Ioannina, Ioannina, Greece. Under optimal

2 Moren, A., Itoh, S., Moustakas, A., ten Dijke, P., and Heldin, C.-H.(2000) Oncogene 19, in press.

p21 Regulation by Smads 29245


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

conditions, more than 90% of the cells were infected as determined bythe blue, b-galactosidase-positive staining using a lacZ control virus orthe green fluorescent protein autofluorescence (for the DN-Smad3 andDN-Smad4 viruses). Routine infections were performed at a multiplic-ity of infection (m.o.i.) of 50 with single viruses. HaCaT cells wereseeded at a density of 5 3 104 cells/cm2 in 24-well tissue culture plates.The next day the cells were transfected with the reporter constructsusing the Fugene-6 reagent. Twelve hours later the culture mediumwas changed to DMEM containing 5% fetal bovine serum. Cells wereinfected 1 h later at the appropriate m.o.i. for 12 h and then washed andfed fresh 5% fetal bovine serum/DMEM. Twenty four hours later thereporter assays were performed, which corresponds to 49 h post-trans-fection and 36 h post-infection. For p21 protein analysis, HaCaT cellswere seeded at a density of 106 cells/well in 6-well tissue culture plates.Twelve hours later the culture medium was changed to DMEM contain-ing 5% fetal bovine serum and infected with different doses of eachvirus (see Fig. 2) 1 h later. Twenty four hours later cells were treatedwith 10 ng/ml TGF-b1 for 20 h, and 20 mg of detergent-soluble cellextracts were analyzed by Western blotting with the p21-specific anti-body. FLAG, HA, and Smad1/2/3-specific antibodies served as controlsto measure expression of the co-infected proteins, and b-catenin anti-body served as control to verify equal protein amount loading. Westernanalysis of transiently transfected SL2 cell extracts or purified baculo-viral Smad3 proteins was performed in the same fashion using therelevant antibodies as described under “Results” and figure legends.

Relative protein expression levels were quantified using the scan-ning densitometric software of the PhosphorImager Fujix BAS 2000.Ratios of band intensities of the tested protein (p21) over the controlprotein (b-catenin) were calculated, and the ground condition ratio wasset to 1 or 100% relative to which all other conditions are expressed.

Co-immunoprecipitation Assays—Forty hours post-transfection ofCOS-7 cells, total detergent extracts were prepared by lysing the cells in20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 10% (v/v) glycerol, 1% (v/v)Triton X-100, 1 mM phenylmethylsulfonyl fluoride (PMSF), 10 mM apro-tinin at 4 °C. The cell extracts were clarified from the insoluble materialby a brief centrifugation at 10,000 rpm and were pre-cleared by incu-bation with protein A-Sepharose at 4 °C for 30 min. The pre-clearedextracts were incubated with the FLAG or GAL4-DBD antibody at 4 °Cfor 2 h, and the immunocomplexes were precipitated with protein-GSepharose, washed with lysis buffer four times, and dissolved in Lae-mmli SDS-polyacrylamide gel electrophoresis (PAGE) loading buffer.After 7% SDS-PAGE the resolved proteins were transferred to Hy-bond-C extra nitrocellulose (Amersham Pharmacia Biotech), and therelevant proteins were detected after incubation with the anti-Myc(9E10) antibody followed by anti-mouse horseradish peroxidase-conju-gated secondary antibody and homemade enhanced chemiluminescentassay on x-ray film (Fuji).

In Vitro and Bacterial Expression of Proteins—Expression of proteinsin vitro was performed using the coupled in vitro transcription/trans-lation system (TNT) of Promega as described previously (12). The qual-ity of the synthesized proteins was verified by SDS-PAGE and autora-diography or PhosphorImager (Fujix BAS 2000) detection. Theglutathione S-transferase (GST) fusion proteins were expressed inEscherichia coli strain DH-10b and purified as described previously(12). The solubilization of the expressed proteins was monitored bySDS-PAGE and Coomassie Brilliant Blue staining. All proteins wereobtained at rather high levels and in relatively pure form as only theprimary protein species were detectable without significant degrada-tion products (Fig. 6, B, E, and H).

GST Protein Interaction Assays—Interaction assays of GST-Sp1 pro-teins with in vitro synthesized Smad proteins were performed as de-scribed previously (12). For the interaction assays of GST-Smad pro-teins with endogenous HaCaT Sp1, total cell extracts were prepared bylysis in 20 mM Tris-HCl, pH 7.4, 100 mM NaCl, 1 mM DTT, 0.5% NonidetP-40, 1 mM PMSF, 1% aprotinin, 50 mM NaF, 25 mM b-glycerophos-phate, 1 mM NaOVO3. Aliquots of the extracts corresponding to approx-imately 107 cells were incubated with the glutathione-Sepharose beadscarrying 10 mg of the GST-Smad fusions for 5 h at 4 °C. The boundproteins were washed with the lysis buffer, dissolved in Laemmli SDSloading buffer, resolved by 8% SDS-PAGE, and analyzed by Westernblotting using the Sp1 antibody and enhanced chemiluminescence.

Nuclear Extract Preparation and Gel Electrophoretic Mobility ShiftAssays—Nuclear extracts from transiently transfected Schneider’s SL2cells were prepared 40 h post-transfection by hypotonic lysis of the cellsin 10% (v/v) glycerol, 20 mM Hepes, pH 7.9, 0.1 mM EDTA, 50 mM KCl,2 mM DTT, 0.5 mM spermidine, 50 mM NaF, 1 mM NaOVO3, 1 mM

PMSF, 10 mM aprotinin by three sequential freeze (liquid nitrogen)-thaw (4 °C) cycles. The nuclear pellets obtained by centrifugation at

4,000 3 g were solubilized by slow rotation for 45 min in hypertonicbuffer, 20% glycerol, 20 mM Hepes, pH 7.9, 0.1 mM EDTA, 600 mM KCl,2 mM DTT, 50 mM NaF, 1 mM NaOVO3, 1 mM PMSF, 10 mM aprotininat 4 °C. The soluble nuclear extracts were collected after centrifuga-tion at 12,000 3 g, were aliquoted, flash-frozen, and stored at 280 °C.The corresponding cytoplasmic extracts were used for b-galactosidaseassays to calibrate the extracts for transfection efficiency. The totalprotein in the nuclear extracts was measured by Bradford assay(Bio-Rad protein assay kit). The abundance of the transfected pro-teins in the nuclear extracts was estimated by Western blot analysisafter 7% SDS-PAGE, using the relevant antibodies and enhancedchemiluminescence.

Gel electrophoretic mobility shift assays (GEMSAs) were performedas described previously (12). The sequences of the oligonucleotides fromthe p21 promoter and the consensus Sp1 site used in the GEMSAexperiments were described by Kardassis et al. (12). The sequence of thesense strand of the mutant double-stranded 286/270 p21 promoteroligonucleotide is 59-GGGTCGACCCTCCTTGA-39, where the underlineindicates the mutated nucleotides. For comparison, the wild type oligo-nucleotide sequence is 59-GGGTCCCGCCTCCTTGA-39 (Fig. 1A). Thequadruple Smad-binding element (SBE4) oligonucleotide was describedby Jonk et al. (38). Oligonucleotides corresponding to the p21 promoterregions were synthesized, annealed, labeled with Klenow and[a-32P]dCTP, and 10 fmol were incubated with 200 ng of each purifiedprotein or with aliquots of nuclear extracts from the transfected SL2cells that were calibrated for total protein content, b-galactosidaseactivity, and specific protein abundance based on the Western blotsignals. For the Smad-binding depletion experiment, purified Smadproteins were first incubated with excess (4 pmol) cold SBE4 oligonu-cleotide for 10 min at 4 °C, and then the Sp1 protein was added andincubated for another 10 min, and finally the labeled p21 promoteroligonucleotide was included in the reaction that proceeded for 30 minprior to 4% PAGE.

Relative bandshift intensities were quantified using the scanningdensitometric software of the PhosphorImager Fujix BAS 2000. Theground condition (Sp1 alone) was set to 1 or 100% relative to which allother conditions are expressed.

RESULTS

Smad3 and Smad4 Can Transactivate the p21 Promoter inHuman HaCaT Keratinocytes—The proximal (2124/242) re-gion of the human p21 promoter is G/C-rich and contains 5sequence motifs that resemble or match exactly the recognitionsequence of the ubiquitous transcription factor Sp1 (59-GGGCGG-39, Fig. 1A, double underline). The constitutive ac-tivity and induction of this promoter by extracellular signals indifferent cell systems depends on some of the Sp1-like motifs(16, 35, 39). Critical for TGF-b-mediated induction of the p21promoter is one of these Sp1 sites (designated TbRE in Fig. 1A)(16). Smads were found capable of transactivating the proximalp21 promoter in HepG2 cells (35). However, similar experi-ments performed in HaCaT cells failed to demonstrate Smad-mediated transactivation of the p21 promoter (data not shownand see Refs. 11 and 40). A reason for this could be the lowSmad expression levels achieved after the inefficient transienttransfection of HaCaT cells with various protocols (data notshown). Adenovirus-mediated gene transfer is highly efficient(80–95% infection rate), and relatively high levels of expressionof the encoded protein can be obtained (37, 41). We thus com-bined transient transfections of HaCaT cells with three differ-ent p21 promoter-luciferase reporters (22,300/18 p21 Luc,22,300/18DSp1 p21 Luc, and 2143/18 p21 Luc) with transientinfections with Smad3- and Smad4-encoding viruses. To stim-ulate the TGF-b pathway we utilized another adenoviral vectorencoding the constitutively active (CA) TGF-b type I receptor(also termed activin receptor-like kinase 5, ALK-5). As shownin Fig. 1B, CA-ALK-5 overexpression led to a moderate butdistinct 2.2-fold activation of the 22,300/18 p21 promoter.Overexpression of either Smad3 or Smad4 by means of theadenovirus system also resulted in the same level (2.2-fold) oftransactivation in the absence of activated ALK-5 receptor(Fig. 1B). In the presence of the activated receptor the effect of

p21 Regulation by Smads29246


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

the Smads was further augmented. Co-expression of Smad3and Smad4 led to a higher level (3.7-fold) of transactivation,which reached levels as high as 5.5-fold after activated receptorco-expression. The transactivation potential of the adenovirallyencoded Smads is specific as a control virus encoding for b-ga-lactosidase resulted in transactivation levels similar to theuninfected condition. Thus, Smad proteins can transactivatethe 22,300/18 p21 promoter in HaCaT cells, and their poten-tial is significantly increased by TGF-b receptor activation.

To pinpoint the importance of the Sp1 G/C-rich motifs of theproximal p21 promoter, we used a deletion mutant of the22,300/18 p21 promoter that lacks sequences between 2124and 261 of the promoter (Fig. 1C). This deletion removes fourout of five of the Sp1 motifs of the proximal promoter and hasbeen previously shown to support very low basal activity thatcould not be regulated by TGF-b or the Smads (16, 35). Indeed,under the present conditions of transient transfection coupledto adenoviral infection, the activity of this promoter remainedlow and not significantly altered by Smad3/4 or the constitutivereceptor. This reconfirms the importance of the Sp1 motifsbetween 2124 and 261 of the proximal promoter for TGF-band Smad-mediated transactivation of the promoter.

A similar analysis was performed with the proximal2143/18 p21 promoter (Fig. 1D). In this case, overexpression ofthe CA-ALK-5 receptor transactivated the proximal promoter10-fold, almost to the same extent as the transactivationachieved by the overexpression of Smad3 and Smad4 proteinsalone (11-fold for Smad3 and 7-fold for Smad4). Co-expressionof Smad3 and Smad4 with the CA-ALK-5 receptor further

increased p21 promoter transactivation by Smads, whereasco-expression of Smad3 with Smad4 and CA-ALK-5 led to evenhigher level of promoter activation (18-fold). It is worth notingthat the CA-ALK-5-stimulated activity of the proximal2143/18 p21 promoter is reproducibly higher than the recep-tor-stimulated activity of the 22,300/18 p21 promoter (com-pare Fig. 1, B and D). This also stands true for the stimulatoryeffects of TGF-b on these promoters (35) and implies the pres-ence of negative regulatory elements in the distal p21 promoterregion (see “Discussion”).

As an independent control for the specificity of Smad-de-pendent transcriptional activation measured by the transientcoupled transfection-adenovirus infection assay, we used thewell established reporter 123(CAGA) Luc (43) whose transac-tivation by TGF-b depends solely on the Smads (Fig. 1E). Asexpected, overexpressed Smad3 and Smad4 synergized withthe CA-ALK-5 signal resulting in a robust 14-fold activationrelative to the basal promoter activity.

In conclusion, by using the adenovirus system we were ableto demonstrate Smad-mediated transactivation of the humanp21 promoter in HaCaT keratinocytes thus firmly establishingthe importance of these factors in p21 gene regulation byTGF-b in this and other cell types.

Smad Proteins Contribute to the Regulation of the Endoge-nous p21 Gene by TGF-b in HaCaT Cells—To evaluate furtherthe involvement of Smad proteins in the regulation of theendogenous p21 gene by TGF-b in HaCaT cells, we monitoredp21 protein expression levels by Western blotting using ap21-specific antibody (Fig. 2). Treatment of HaCaT cells with

FIG. 1. A–E, Smad3 and Smad4 trans-activate the human p21 promoter via itsG/C-rich proximal sequences in HaCaTcells. A, schematic representation of thehuman p21 promoter region 22300/18.The region between nucleotides 2124/242,which is important for both the constitu-tive and the inducible activity of the pro-moter, is shown as a black bar and its nu-cleotide sequence is expanded below. Theoval represents a Smad-binding element(SBE) identified in the distal part of thepromoter (42, 50). In the sequence, heavyunderlines mark the GEMSA oligonucleo-tide probes (A, p21Pr(286/270)) and (B,p21Pr(2122/284)) used in Fig. 3. Bindingsites for the ubiquitous transcription factorSp1 are double-underlined. TbRE indi-cates the Sp1 motif shown previously to beimportant for the stimulation of the pro-moter by TGF-b (16). The TATA box isboxed. B–E, HaCaT cells were transientlytransfected with the 22,300/18 p21 luc(B), the 22,300/18DSp1 p21 luc (C), the2143/18 p21 luc (D), the 123(CAGA) luc,and control b-galactosidase reporter con-structs and the next day were infectedwith adenoviruses expressing Smad3,Smad4, CA-ALK-5 or control, LacZ.Thirty six hours post-infection cell lysateswere assayed for luciferase and b-galacto-sidase activities. The luciferase activitynormalized over the b-galactosidase ac-tivity was plotted in a bar graph relativeto the mock transfection control, whichwas arbitrarily set at 1. The data repre-sent measurements from two independ-ent experiments that include triplicatesamples each.



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

10 ng/ml TGF-b1 for 20 h resulted in the robust accumulationof p21 protein (12-fold compared with untreated control, Fig.2A, lanes 1 and 2) as previously reported (7). The TGF-b1 effectcould be mimicked by the adenovirus-encoded constitutivelyactive type I receptor ALK-5 in a dose-dependent manner. Aselected m.o.i. of 50 of this virus resulted to a 3-fold activation(Fig. 2A, lane 3). Overexpression of adenovirus-encoded Smad3and Smad4 also resulted in a dose-dependent increase of en-dogenous p21 protein levels. m.o.i. of 50 for each Smad-virus

resulted in a low but reproducible 2-fold increase of p21 proteinlevels (Fig. 2A, lane 6). Combination of overexpressed Smad3and Smad4 together with CA-ALK-5 or TGF-b1 treatmentshowed further increase of the p21 protein levels, which corre-spond to 5- and 14-fold, respectively, under the infection con-ditions shown in Fig. 2A (lanes 4 and 5).

To enhance the evidence that the Smad signaling pathway isinvolved in p21 protein accumulation in response to TGF-b1 inHaCaT cells, we also made use of dominant negative and in-hibitory Smad proteins (Fig. 2, B–D). A Smad3 carboxyl-termi-nal truncated mutant (DN-Smad3), the equivalent truncationof Smad4 (DN-Smad4), and the inhibitory Smad7 all resultedin a dose-dependent decrease of p21 accumulation in responseto TGF-b1. The Smad3 and Smad4 mutants have been previ-ously shown to interfere with TGF-b signaling in a dominantnegative fashion with respect to specific target gene responses(44, 45), and we have shown that the DN-Smad4 mutant inter-feres with p21 promoter regulation in HepG2 cells (35). Sim-ilarly, the Smad7 inhibitor has also been previously reportedto inhibit the induction of the p21 promoter-luciferase re-porter by TGF-b1 in HaCaT cells (46). The combined p21Western blotting results strongly support the involvement ofthe Smad signaling pathway in endogenous cell p21 regula-tion by TGF-b1.

Smad Proteins Do Not Associate with the Proximal p21 Pro-moter DNA—Smad3 contains DNA binding activity with lowaffinity toward 59-TCTGAGAC-39 (termed the Smad-bindingelement (SBE)), whereas Smad4 toward both the SBE andG/C-rich motifs (42, 43, 47–49). The SBE is absent from theproximal p21 promoter (Fig. 1A) and exists in an upstreamdistal segment of the promoter (50) with no apparent functionalsignificance (16, 35). On the other hand the p21 proximalpromoter is G/C-rich. We thus tested the hypothesis of directDNA binding of Smads to the proximal p21 promoter, whichconfers the inducibility of this gene to TGF-b and Smads (16,35). Since mammalian cells express high levels of Sp1 protein,it has been rather difficult to identify TGF-b-specific nucleo-protein complexes by using GEMSA on the p21 or other Sp1-containing promoter sequences (16, 30). For this reason weused the well established Drosophila Schneider’s SL2 cell linethat lacks endogenous Sp1 activity (51). Fig. 3A shows repre-sentative GEMSA data produced from nuclear extracts of tran-siently transfected SL2 cells with the indicated combinations ofSp1 and Smad proteins and a radioactively labeled oligonucleo-tide corresponding to the TGF-b-responsive element (TbRE)(p21Pr(286/270)) of the proximal p21 promoter. Similar re-sults were obtained when the upstream oligonucleotide con-taining two Sp1 motifs (p21Pr(2122/284)) as shown in Fig. 1Awas tested (data not shown). Our results showed the following.(a) Under the conditions used, endogenous SL2 nuclear pro-teins do not recognize the p21 oligonucleotides (Fig. 3A, lane 1).(b) Overexpression of Smad3 or Smad4 or both gives the samenegative result as mock-transfected cells (Fig. 3A, lanes 2–4).(c) Overexpression of Sp1 results in a specific nucleoproteincomplex as expected (Fig. 3A, lane 8). (d) Overexpression ofSmad3 together with Sp1 or Smad4 together with Sp1 resultsin a small 2- and 2.5-fold enhancement of the Sp1 nucleoproteincomplex, respectively (Fig. 3A, lanes 5 and 6). (e) Overexpres-sion of both Smad3 and Smad4 together with Sp1 results in asignificant (5-fold) enhancement of the Sp1 nucleoprotein com-plex on both p21 probes (Fig. 3A, lane 7 and data not shown). Inorder to prove the specificity of the obtained Sp1 nucleoproteincomplex in the SL2 nuclear extracts, we used competition ex-periments with excess amounts of cold oligonucleotides corre-sponding to the wild type p21 proximal promoter (Fig. 3B, lanes3–5), the same oligonucleotides harboring point mutations in

FIG. 2. A–D, the Smad signaling pathway regulates endogenous p21responsiveness to TGF-b. A, Western blot analysis of HaCaT p21 ex-pression in cells transiently infected with Mock (LacZ, m.o.i. of 200),Smad3 plus Smad4 and CA-ALK-5 (each at m.o.i. of 50) adenovirusesand treated or not with 10 ng/ml TGF-b1 as indicated. A representativechemiluminogram with only the relevant part is shown (WB, for West-ern blot). Relative arbitrary densitometric values of the chemilumino-gram are boxed below the p21 blot. The same results were obtained intwo independent experiments. Control blots showing expression of theinfected proteins and equal loading of the protein gel (b-catenin) arebelow. The Smad3, Smad4, and CA-ALK-5 receptor protein species areshown with arrows to the right of the panel. Nonspecific protein speciesare indicated by an asterisk. B–D, p21 expression levels in HaCaT cellstransiently infected with increasing m.o.i.s (indicated) of dominantnegative Smad3 mutant (B), dominant negative Smad4 mutant (C), orinhibitory Smad7 (D) and treated or not with 10 ng/ml TGF-b1 asindicated. Representative chemiluminograms with only the relevantparts are shown. Relative arbitrary densitometric values of the chemi-luminogram are boxed below each p21 blot. The same results wereobtained in two independent experiments. Control blots showing ex-pression of the infected protein and equal loading of the protein gel(b-catenin) are below. The endogenous Smad3 and overexpressed DN-Smad3 or Smad7 protein species are shown with arrows on the left ofthe panel.



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

the Sp1-like elements (Fig. 3B, lanes 6–8) or a consensus Sp1sequence (Fig. 3B, lane 9). These experiments confirmed thatthe nucleoprotein complexes formed with p21 promoter DNAprobes are specific and contain only Sp1. In addition, unre-lated, non-Sp1 sequence containing oligonucleotides includingthe SBE4 did not show any competition (data not shown). Itmust be noted that co-expression of Smads with Sp1 resultsonly in relative enhancement of the nucleoprotein complexwithout any additional higher size complexes. Finally, the en-hanced binding of Sp1 to the p21 proximal promoter afterco-expression of Smad proteins in the SL2 cells could not resultfrom nonspecific effects of differential protein expression in thetransfected cells, as the levels of nuclear Sp1 and Smad pro-teins appeared rather comparable (Fig. 3C).

Thus the combined data demonstrate that Smad proteins donot associate with the proximal p21 sequences tested and co-expression of Smad3, Smad4, and Sp1 results in an enhancednucleoprotein complex that retains the binding characteristicsof Sp1.

GEMSA using nuclear extracts of transiently transfected DrosophilaSchneider SL2 cells with the indicated Sp1 and Smad expression con-structs and the radiolabeled 286/270 p21 promoter oligonucleotide asa probe. The nucleoprotein complexes were resolved by 4% PAGE anddetected by autoradiography. A representative autoradiogram isshown. The migrating positions of the free probe and the bandshiftproduced by Sp1 (lanes 5–8) are shown with arrows on the left side ofthe panel. Relative arbitrary densitometric values of the autoradiogramare boxed on top of the panel. B, GEMSA using the same nuclearextracts as above from SL2 cells transiently transfected with Sp1 alone(SL2/Sp1 NE) and radiolabeled p21Pr(286/270) (lanes 1–9) or thecorresponding point mutant (mut) (lane 10) oligonucleotides as probes.A representative autoradiogram is shown. The migrating positions ofthe free probe and the bandshift produced by Sp1 are shown witharrows on the left side of the panel. In lane 1 half the amount of nuclearextract compared with lanes 2–10 was used. Competitions with 50-foldexcess (lanes 3, 6, and 9), 100-fold excess (lanes 4 and 7), and 200-foldexcess (lanes 5 and 8) of the wild type (wt, lanes 3–5), the point mutant(mut, lanes 6–8) 286/270 p21, or the consensus Sp1 (con, lane 9)oligonucleotide are shown on the top of the panel. C, expression levels ofthe indicated transiently transfected Smad and Sp1 proteins in thenuclear extracts of the Drosophila Schneider SL2 cells used in theGEMSA shown in A. Aliquots of the nuclear extracts were resolved by7% SDS-PAGE, and Western blot analysis was performed using aSmad3-, FLAG- (for Smad4), and Sp1-specific antibody as indicated onthe left side of each chemiluminogram. Only the relevant part of thechemiluminogram is shown (WB, for Western blot). The migratingpositions of the relevant proteins are shown with arrows on the rightside of the panel. Nonspecific protein species are indicated by asterisks.D, GEMSA using purified bacterially expressed Sp1 and baculovirallyexpressed Smad proteins and radiolabeled 286/270 p21 promoter oli-gonucleotide. The nucleoprotein complexes were resolved by 4% PAGEand detected by autoradiography. A representative autoradiogram isshown. The migrating positions of the free probe and the two distinctbandshifts are shown with arrows on the left of the panel. Competitionswith 20-fold excess (lanes 6 and 10), 100-fold excess (lanes 7 and 11),and 200-fold excess (lanes 8 and 12) of the 286/270 p21 oligonucleotideare shown on the top of the panel. The exposure time for lanes 1–8 is 8 hand for lanes 9–16 is 16 h. E, GEMSA using the same reagents as in C,except that the two Smad proteins were first preincubated with excess(400-fold, lanes 2 and 3) cold SBE4 and then allowed to interact with(lane 2) or without (lane 3) Sp1 and finally with radiolabeled p21promoter oligonucleotide (286/270). Only the relevant part of a repre-sentative autoradiogram is shown. The migrating positions of the band-shifts are shown by arrows on the left side of the panel. Note thesignificant enhancement of the Sp1-specific bandshift when excess coldSBE4 oligonucleotide is provided in the reaction (lane 2). The exposuretime of the autoradiogram was 3 h and must be compared with the 8-hexposure of lane 5 in D. F, quality and phosphorylation status of the twoforms of baculovirally expressed Smad3 proteins used in the GEMSAshown in D and E. Aliquots of the purified proteins were resolved by 7%SDS-PAGE, and Western blot analysis was performed using a phospho-serine (P-Ser)- and histidine (His)-specific antibody as indicated on theleft side of each chemiluminogram. Only the relevant part of the chemi-luminogram is shown (WB, for Western blot). The migrating positions ofthe relevant proteins are shown with arrows on the right side of thepanel.

FIG. 3. A–F, Smad proteins do not bind directly to the proximal p21promoter sequences but enhance binding of Sp1 to these sequences. A,



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Smad Proteins Induce an Increase in Sp1 Affinity for theDNA Sequences of the p21 Proximal Promoter—To examinethoroughly whether Smad proteins alter the Sp1 affinity forDNA, we relied on purified protein factors and in vitro GEMSA.By using purified unphosphorylated or phosphorylated (by CA-ALK-5 receptor) Smad3 and unphosphorylated Smad4 from abaculovirus system (36) and purified Sp1 from bacteria, weconfirmed that Smad3, phospho-Smad3, Smad4 alone, or incombinations could not exhibit stable complexes with the p21promoter DNA (Fig. 3D, lanes 2–4 and 13–15). The same pro-teins strongly and stably associated with the SBE DNA (datanot shown and see Ref. 36). Purified Sp1 formed a strongcomplex with both p21 promoter probes (12); however, for thepurpose of the experiments presented in this figure, a lowconcentration of Sp1 was used that reproducibly gave a ratherweak binding (Fig. 3D, lane 9). Interestingly co-incubation ofphospho-Smad3, Smad4, and Sp1 with the p21 probe resultedin a strong nucleoprotein complex of higher molecular mass(labeled Sp19) when compared with the complex obtained bySp1 alone (Fig. 3D, lane 5). This effect was only obtained whenphospho-Smad3 was used, as unphosphorylated Smad3 in com-bination with Smad4 and Sp1 resulted in the same weak band-shift as Sp1 alone (compare Fig. 3D, lanes 9 and 16). Westernblot analysis of the two Smad3 preparations using anti-phos-phoserine-specific antibodies confirmed that only phospho-Smad3 contained phosphorylated serines (Fig. 3F). The en-hanced Sp19 bandshift was also followed by a lower and weakertrailing bandshift. Competition experiments with excess coldoligonucleotides demonstrated that the higher mass Sp19complex is specific for the p21 probe used, whereas the trail-ing bandshift possibly represents an experimental artifact(Fig. 3D, lanes 6–8). In addition, these competition assaysproved that the Sp19 bandshift represented a high affinitycomplex of Sp1 to the p21 promoter DNA (Fig. 3D, lanes 5–12).A 20-fold excess cold oligonucleotide could easily compete mostof the Sp1 binding under the conditions used (lane 10), whereascomplete competition of the higher mass complex was obtainedonly when 200-fold excess of cold oligonucleotide (lane 8) wasused. Similar in vitro GEMSA results to those shown in Fig. 3Dfor the p21Pr(286/270) probe were also obtained for the up-stream p21Pr(2122/284) probe (data not shown). Identicalcompetition profiles were observed by the consensus Sp1 oligo-nucleotide, whereas the mutant p21Pr(286/270) oligonucleo-tide failed to compete like in the SL2 nuclear extract GEMSAs(Fig. 3B), and the unrelated SBE4 oligonucleotide not onlyfailed to compete but exhibited enhancement of Sp1 binding tothe p21 promoter (data not shown and see below). These ex-periments suggest that Smad proteins increase the affinity ofSp1 for its cognate G/C-rich-binding motif.

To understand further the nature of the induced enhance-ment of the Sp1 affinity for DNA by the Smads, we performeda preincubation-competition experiment using a quadrupleconcatamer of the consensus SBE (SBE4) (Fig. 3E). In thisexperiment the phospho-Smad3 and Smad4 were preincubatedwith excess (400-fold) cold SBE4 oligonucleotide to saturatethe intrinsic binding of the Smad proteins for DNA, followedby addition of the Sp1 and the labeled p21 promoter probe(Fig. 3E, lane 2). The resulting complex was compared withthat obtained when all proteins were mixed together in theabsence of excess SBE4 oligonucleotide (lane 1) or in the ab-sence of Sp1 (lane 3). Surprisingly, this method of treatment

FIG. 4. A and B, Smad3 and Smad4 mutant proteins that cannotassociate with consensus Smad-binding elements are fully capable oftransactivating the proximal p21 promoter. Transient transfection ex-periments of HepG2 cells with the indicated wild type (wt) and doublemutant (dm) Smad and CA-ALK-5 expression constructs together withthe 2143/18 p21 luciferase (A) or the Smad-sensitive 123(CAGA) lu-ciferase and control b-galactosidase reporter constructs. Forty hourspost-transfection cell lysates were assayed for luciferase and b-galacto-sidase activities. The luciferase activity normalized over the b-galacto-sidase activity is plotted in a bar graph relative to the mock transfectioncontrol, which is arbitrarily set at 1. The data represent measurements

from three independent experiments that included triplicate sampleseach. The analysis in B serves as control experiment for the results ofA. The scale of the relative luciferase activity in B is broken to fit in thefigure.



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

resulted in a further 3-fold enhancement of the Sp19 bandshift(compare lanes 1 and 2) which suggests that when the Smadproteins are provided with their DNA substrate they can stillenhance the affinity of Sp1 for the p21 promoter DNA. Thesefindings suggest that the Smad3-Smad4 complex bound orunbound to DNA leads to a significant increase in the affinityof Sp1 for its DNA-binding sequences.

Smad Binding to DNA Is Not Essential for the Activation ofthe p21 Proximal Promoter—To examine whether Smad bind-ing to DNA is required for the activation of the p21 promoter,we made use of point mutants in the conserved b-hairpin of theMH1 domain of Smad3 and Smad4 which is the DNA-bindingdomain of Smads (47). We used an R74K/K81R double mutantof Smad3 (Smad3dm) and the corresponding R81K/K88R dou-ble mutant of Smad4 (Smad4dm). These mutations in theb-hairpin of the MH1 domain, as predicted from the crystalstructure, completely abolish binding of the Smads to theSBE.2 The prediction in our experiment was that these mu-tants should have wild type transactivation activity on thep21 promoter if DNA binding of Smads to the p21 proximalpromoter is not important. Indeed, the experiments shown inFig. 4A, performed in HepG2 cells, demonstrate that bothSmad3 and Smad4 mutants can transactivate the proximal p21promoter as efficiently as the wild type proteins. The constitu-tive effect of overexpression of the Smad proteins on the prox-imal p21 promoter was retained by the mutants, as was theinducible effect stimulated by the co-expression of CA-ALK-5.As a control experiment we tested the transactivation of anartificial promoter that contains 12 copies of the SBE (CAGAelement) in front of the adenovirus major late promoter andthat is very sensitive to activation by TGF-b in a Smad-depend-ent way (43). As previously reported, the overexpression ofSmad3 with Smad4 significantly increased the basal activity ofthe promoter, and activation by means of CA-ALK-5 furtheraugmented this response (Fig. 4B). In contrast, Smad3dmfailed to increase the basal promoter levels when co-expressedwith wild type Smad4, suggesting that Smad3 is the primaryDNA-binding factor on this promoter. Smad4dm gave wild typelevels of constitutive activation. Importantly, the receptor-induced superactivation showed a 2–3-fold decrease by bothmutants. Finally, co-expression of both mutants resulted inunaffected basal promoter levels and reduced receptor-super-activated levels. The negative effects of the mutants on basallevel and receptor-induced promoter activation seen with the123(CAGA) reporter stand in contrast to the p21 proximalpromoter experiments, demonstrating that the transactivationof the p21 promoter is independent from mutations that dis-rupt Smad-SBE DNA binding and interfere with optimal trans-

activation of SBE-dependent promoters.Smad2 and Smad3 Co-immunoprecipitate with Sp1 in

Transfected COS-7 Cells—Our previous work demonstrated astrong functional interaction between Smads and Sp1 (35). Totest whether Smads and Sp1 physically interact, we appliedseveral complementary experimental approaches. Fig. 5 showsthe results of co-immunoprecipitation experiments performedwith protein extracts from COS-7 cells transiently transfectedwith epitope-tagged Smad and GAL4-Sp1 fusion proteins. As apositive control we used the receptor activation-dependent in-teraction of Smad3 with Smad4 (Fig. 5A, lanes 2 and 3). Smad3was found to co-precipitate with Sp1 at measurable levels evenin the absence of stimulation of the signaling pathway (Fig. 5B,lane 2). However, co-expression of the CA-ALK-5 resulted insignificant enhancement of co-precipitating Smad3 (Fig. 5B,lane 3). Essentially the same results were obtained for theco-precipitation of Smad2 with Sp1 (Fig. 5C, lanes 2 and 3). Incontrast, we failed to detect co-precipitation of Smad4 withGAL4-Sp1, suggesting that the interaction of these two pro-teins might be either indirect or rather weak compared withthe Smad3-Sp1 interaction (Fig. 5C, lanes 4 and 5).

The combined data presented in Fig. 5 demonstrate thatSmad2 and Smad3 can co-precipitate with Sp1, whereasSmad4 is unable to do so under the conditions used.

Smad2, Smad3, and Smad4 Directly Associate with Sp1 viaTheir Conserved MH1 Domain—In order to map the domains ofSmad and Sp1 proteins involved in their physical interaction,we used an in vitro interaction assay with GST-Sp1 and wildtype or mutated Smad proteins synthesized in vitro. Fig. 6Ashows schematically various Smad and GST-Sp1 constructsused in these experiments. The relative expression levels ofaffinity-purified Sp1, negative control GST, and in vitro syn-thesized Smad proteins are shown in Fig. 6B. When the in vitrosynthesized Smad2, Smad3, and Smad4 were allowed to inter-act with GST-Sp1, all three Smad proteins were found capablefor the direct interaction (Fig. 6C, lanes 4–6). However, amongthe three, Smad3 showed the strongest potential for interac-tion, whereas Smad2 and Smad4 interacted rather weakly (6and 10% relative to Smad3 which is arbitrarily set to 100%,Fig. 6A). The interaction was specific as the GST moiety of thefusion protein when tested alone failed to support productiveinteraction (Fig. 6C, lanes 1–3). It must be noted here thatthese interactions are constitutive and do not depend on anactivated TGF-b signaling pathway. In order to map the do-main in Smad3 that is responsible for the interaction with Sp1,we used three different deletion mutants of Smad3 (Fig. 6A).Smad3-(1–248) contains the MH1 and linker sequences andshowed detectable, albeit weak, (17.8%) interaction with Sp1

FIG. 5. A–C, Smad3 and Smad2 co-immunoprecipitate with Sp1 in mammalian cells. A, Smad3 co-precipitates with Smad4 in transientlytransfected COS-7 cells. COS-7 cells were transiently transfected with the indicated combinations of epitope-tagged expression constructs,including the CA-ALK-5, which mimics ligand stimulation. Forty hours post-transfection cell lysates were immunoprecipitated with antibodiesagainst the indicated epitopes (IP for immunoprecipitation), resolved by 7% SDS-PAGE, transferred to membranes, and probed with the indicatedepitope-specific antibodies (WB, for Western blot) prior to detection by enhanced chemiluminescence. Total cell lysates were analyzed in parallelwith the same Western blots (bottom two rows). The detected Smad proteins are labeled on the right of the panels (S3 for Smad3 and S4 for Smad4).B, Smad3 co-precipitates with Sp1 in transiently transfected COS-7 cells. COS-7 cells were transiently transfected and analyzed as in A. Total celllysates were in parallel analyzed in the same Western blots (bottom row). The detected Smad proteins are shown with arrows on the right (S3 forSmad3). C, Smad2 co-precipitates with Sp1 in transiently transfected COS-7 cells. COS-7 cells were transiently transfected and analyzed as in A.Total cell lysates were in parallel analyzed in the same Western blots (bottom row). The detected Smad proteins are shown with arrows on the right(S2 for Smad2 and S4 for Smad4).



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

FIG. 6. A–I, Smad2, Smad3, and Smad4 directly interact with transcription factor Sp1 via their MH1 domain. A, diagrammatic representationof the various Smad3, Smad2, and Smad4 proteins used for the in vitro interaction assays shown in B and C. The Smad amino-terminally conservedMad homology (MH) 1 domain, the linker, and the conserved carboxyl-terminal MH2 domain are indicated with boxes of different shading. Theb-hairpin loop of the MH1 domain, which is the DNA-binding domain of Smads, is shown as a small box with two pinheads representing thepositions of the two point mutations described in Fig. 4. The amino- and carboxyl-terminal amino acids are numbered. The Sp1 functional domainsA–D (23) are shown with brackets along with the conserved repeated Ser/Thr-rich and Gln-rich subdomains of the transactivation domains A andB, the zinc-finger DNA-binding motifs, and the transactivation modulatory region (2/1). The relative sizes of the different proteins are not in scale.The column on the right-hand side summarizes the interaction results of C after densitometric analysis, and the values are presented aspercentages relative to Smad3 that was set arbitrarily to 100. The actual arbitrary densitometric units (adu) of the reference sample (Smad3) areshown in parentheses. B, left, Coomassie Brilliant Blue staining of the input GST-Sp1 and control GST proteins coupled to glutathione beads (lanes1 and 2) that were used in the interaction assay of C. Arrows indicate the two protein species. Molecular mass markers (expressed in kilodaltons)are shown to the left of the gel. Right, autoradiogram of the input 35S-labeled in vitro synthesized Smad proteins (lanes 3–8). Smad proteins weresynthesized in vitro by a rabbit reticulocyte lysate. Each protein band represents 5% of the total amount of protein used in each interaction assayof C. Arrows indicate the positions of the relevant protein bands. Molecular mass markers (expressed in kilodaltons) are shown to the left of theautoradiogram. C, left, interaction of in vitro translated Smad proteins with GST-Sp1 immobilized on glutathione beads. The in vitro synthesizedSmad2, Smad4, and Smad3 proteins (B, right) were allowed to interact with glutathione beads carrying GST alone (lanes 1–3) or GST-Sp1 (lanes4–6), washed thoroughly, resolved by 7% SDS-PAGE, and detected by autoradiography. The positions of each relevant protein are marked witharrows. Molecular mass markers (expressed in kilodaltons) are shown to the left of the autoradiogram. Right, interaction of in vitro translatedSmad3 deletion mutant proteins with GST-Sp1 immobilized to glutathione beads proves that the Smad3 MH1 domain is required for theinteraction. Smad3 deletion mutants were synthesized in vitro by a rabbit reticulocyte lysate (input, B) and then allowed to interact withglutathione beads carrying GST alone (lanes 7–9) or GST-Sp1 (lanes 10–12), washed thoroughly, resolved by 12% SDS-PAGE, and detected by



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

compared with the wild type Smad3 (Fig. 6C, lane 10). Smad3-(1–122) contains the MH1 domain only, which is also truncatedat its carboxyl terminus and exhibited similar weak interaction(23.9%) as the previous MH1-linker mutant (Fig. 6C, lane 11).Finally, Smad3-(122–424), which contains very few amino ac-ids from the MH1 domain, the linker, and the MH2 domains,failed to support detectable interactions (,1.5%) with Sp1 (Fig.6C, lane 12). The same Smad3 mutants failed to show nonspe-cific retention to the GST affinity columns (Fig. 6C, lanes 7–9).

We conclude from the in vitro experiments that all threeSmad proteins of the TGF-b signaling pathway, Smad2, Smad3and Smad4, are capable of direct physical interaction withtranscription factor Sp1, although Smad3 shows a more pro-nounced interaction (10–16-fold, based on densitometric anal-ysis) in the in vitro pull-down assays. The MH1 domain ofSmad3 is the primary determinant for this interaction, al-though the MH2 domain is necessary for fully productive in-teraction, as MH2 truncation significantly decreased (by 4–5-fold) the interaction potential of the residual Smad3 domains.

Although the previous experiments provide strong evidencefor the physical association of Smad proteins and in particularSmad3 with Sp1, in these assays chimeric Sp1 proteins werealways used (GAL4-Sp1 and GST-Sp1). To obtain evidence thatthe natural Sp1 molecule also interacts with Smad proteins, weused a series of GST-Smad fusion proteins that included full-length Smad3 and Smad4 as well as deletion mutants of thesetwo proteins (Fig. 6D). Total detergent extracts from HaCaTcells were passed through the GST-Smad affinity columns(Fig. 6E) and washed, and the proteins bound to the columnswere analyzed by Western blotting using an Sp1-specific poly-clonal antibody (Fig. 6F). Endogenous Sp1 was readily detect-able in the total HaCaT cell extract (Fig. 6F, lane 9). Sp1 wasfound to interact with GST fusions of full-length Smad4 (lane 1)and Smad3 (lane 3) but not GST alone (lane 8). The interactionwith Smad3 was more efficient than with Smad4, which is inagreement with the in vitro experiments of Fig. 6A–C. Exper-iments with deletion mutants of the two Smad proteins alsocorroborated the previous results as they showed that the MH1plus linker domains of Smad4 were capable of sustaining in-teraction with Sp1 but less efficiently than full-length protein(Fig. 6F, lane 2). In addition, the MH1 domain of Smad3 wasthe primary determinant for the interaction with Sp1 (Fig. 6F,lane 6); however, the presence of the linker (lane 5) and theMH2 domain (lane 3) enhances the interaction considerably.Furthermore, the isolated linker plus MH2 (DMH1, lane 4),MH2 (lane 7), and linker (not shown) domains did not supportany detectable interaction, excluding the possibility that thesedomains could provide primary specificity to the intermolecularassociations studied.

Therefore, these experiments are in full agreement withthose shown in Figs. 5 and 6, A–C, and establish that thenatural Sp1 molecule is capable of interacting with Smad pro-teins via their MH1 domains.

The Glutamine-rich Region of the Sp1 Transactivation Do-main Can Sustain Functional Synergism with Smad Pro-teins—Since a first map of the interaction domain on Smadproteins was established, we were interested in defining theSp1 sequences that were participating in the Smad-Sp1 inter-action. For this reason we relied on the in vitro interactionassay by using a panel of GST-Sp1 deletion mutants (Fig. 6, Gand H) and in vitro synthesized full-length Smad3 protein sincethis showed the best potential in the interaction assays (Fig. 6,C and F). The Sp1 mutants included deletions of the conservedand duplicated subdomain A of the major transactivation do-main of this protein (23), the second subdomain B of the majortransactivation domain together with domain C, which alsoconfers transactivation potential to Sp1 and a carboxyl-termi-nal deletion of domain D (Fig. 6G). The three mutants andfull-length Sp1 were produced as GST fusions (Fig. 6H). Smad3exhibited direct interaction with full-length Sp1 (Fig. 6I, lane3) as described above, which was not seen with GST alone (lane2). Smad3 binding to Sp1 was decreased when subdomain Awas deleted (lane 4). However, Smad3 binding was not affectedat all by deletion of the subdomains B plus C (lane 5) and wasmildly decreased by deletion of domain D (lane 6). Since none ofthe Sp1 mutants showed complete loss of interaction potentialwith Smad3, this suggested that the conserved and duplicatedglutamine and serine/threonine-rich sequences of subdomainsA and B and/or the DNA-binding domain of Sp1 (domain Zn21

in Fig. 6G), which are included in all the Sp1 mutants, might beinvolved in the interaction. Alternatively, it is possible that Sp1contains multiple sequence motifs that contribute to the inter-action with Smad proteins.

To characterize in more detail sequences within the con-served domains A and B that might contribute to the functionalsynergism with Smads, we made use of another panel of dele-tion mutants of Sp1, which included shorter truncations of themajor transactivation domain fused to GAL4 DNA-binding do-main (GAL4-DBD, Fig. 7A). Fig. 7B shows that co-expression ofSmad3 and Smad4 together with full-length Sp1 fused to GAL4resulted in almost 20-fold activation of a reporter containingfive concatamerized GAL4-binding sites in front of a minimalthymidine kinase promoter, relative to the constitutive level ofGAL4-Sp1 alone. Mutants Sp1 A 1 B, Sp1 B, and Sp1 Bc allexhibited significant transactivation (34–37-fold) by theSmad3-Smad4 complex, which was comparable with each otherand higher than GAL4-Sp1. Mutant Sp1 Bc retains only theglutamine-rich region of domain B (amino acids 424–542). In

autoradiography. The positions of each relevant protein are marked with arrows. Molecular mass markers (expressed in kilodaltons) are shownto the left of the autoradiogram. D, diagrammatic representation of the various GST-Smad fusion proteins used for the interaction assays shownin F. The same drawing conventions as in A are used. The relative sizes of the different proteins are not to scale. E, Coomassie Brilliant Bluestaining of the GST-SMAD proteins used for the interaction assay in F. The relevant protein bands are marked with a dot to distinguish them fromminor degradation or aberrant protein fragments associated with the affinity columns. The migrating position of molecular mass markers(expressed in kilodaltons) is indicated to the left of the panel. F, interaction of endogenous Sp1 from HaCaT cells with Smad3 and Smad4. Totaldetergent extracts of subconfluent HaCaT cultures were incubated with the indicated GST-Smad affinity columns (lanes 1–7) or with a control GSTcolumn (lane 8), washed thoroughly, and the interacting proteins were resolved by 8% SDS-PAGE and Western blotting (WB) with a specificantibody against human Sp1. Total lysate of HaCaT cells was analyzed in lane 9. G, diagrammatic representation of the GST-Sp1 fusion proteindeletion mutants used in the interaction assay of I. The same drawing conventions as in A are used. The relative sizes of the different proteins arenot to scale. H, Coomassie Brilliant Blue staining of the wild type and three deletion mutants of Sp1 fused to GST after 7% SDS-PAGE.Odd-numbered lanes represent the first bacterial supernatant (see “Experimental Procedures”), and even-numbered lanes represent the secondsupernatant, which was actually used (relevant protein species marked by a dot) for the interaction assays of I. The migrating position of molecularmass markers (expressed in kilodaltons) is indicated to the left of the panel. I, Smad3 interacts with all tested Sp1 deletion mutants.Autoradiogram of the interaction assay of in vitro translated Smad3 with GST alone (lane 2), GST-Sp1 wild type (wt, 83–778), and mutant (asindicated, lanes 3–6) affinity columns. Lane 1 represents an aliquot of the input in vitro translated Smad3 used for the assays which is equivalentto 20% of the amount used for the interactions. In lane 5, migration of the Smad3 protein band is distorted due to the excess amount of unlabeledGST-Sp1 D(B1C), which co-migrates with Smad3. The position of Smad3 is marked by an arrow. The migrating position of molecular massmarkers (expressed in kilodaltons) is indicated to the left of the panel.



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

contrast, mutant Sp1 Bn that retains only the serine/threo-nine-rich region of domain B (amino acids 263–424) was notresponsive to the transactivation exerted by the Smad complex(1.2-fold). These data suggest that the glutamine-rich subdo-main of Sp1 might be involved in the Smad-Sp1 interaction.Furthermore, a GAL4-Sp1 Bc fusion protein, which contains atriple glutamine to alanine amino acid substitution (GAL4-Sp1Bc (Q3A), Fig. 7A) showed decreased transactivation (7.5-fold)in the presence of Smad3/Smad4, and the activity of this mu-tant alone decreased considerably relative to that of GAL4-Sp1(Fig. 7B). This result enhances the hypothesis that the gluta-mine residues of this domain are important for the cooperation

with Smad proteins, and of course, since not all glutamineswere mutated in the triple point mutant, the residual transac-tivation potential (7.5-fold) can be attributed to the remainingglutamine-rich protein surfaces.

Finally, we tested the functional interaction of the glu-tamine-rich domain of Sp1 with the three Smad proteins ofthe TGF-b signaling pathway, Smad2, Smad3, and Smad4(Fig. 7C). The results showed that Smad3 exhibits the strong-est cooperative activation (8-fold) together with Sp1-Bc, whichcan be enhanced further (13-fold) by co-expression of Smad4.Smad2 and Smad4 alone did not show any appreciable level oftransactivation (1.1–1.4-fold) in agreement with the low orundetectable constitutive interactions described above. How-ever, the Smad2 plus Smad4 combination resulted in measur-able but small transactivation (2.9-fold), which is consistentalso with the physical interaction data, since this combinationpartially mimics the activation of Smad2 by ligand.

In conclusion, these experiments suggest that the glutamine-rich subdomain of the transactivation domain of Sp1 playsimportant roles in the functional synergism between Sp1 andSmad proteins and further strengthen the finding that Smad3under all experimental conditions tested exhibits the bestphysical and functional cooperativity with transcription factorSp1.

DISCUSSION

The experiments of Fig. 1 demonstrate the positive regula-tory effect of Smad proteins on the activity of the p21 promoterin human keratinocytes HaCaT. This result is of particularsignificance as this cell line has been extensively analyzed forthe mechanism of growth inhibition by TGF-b (4, 11, 52) andexhibits a rather dramatic induction of its endogenous p21 genein response to TGF-b (7 and Fig. 2). The Smad effect alsodepends on the integrity of the G/C-rich, Sp1-occupied proximalpromoter (Fig. 1C). It must be noted that the adenovirus-mediated expression of Smad3 and Smad4 proteins positivelyup-regulates the 22,300/18 and the 2143/18 p21 promoters,to a lower extent than TGF-b (16). Thus, additional regulatoryfactors may be required for the maximal activation of the p21promoter by TGF-b (see below). In addition, the promoter anal-ysis of Fig. 1 supports our previous analysis in HepG2 cellswhere a distal inhibitory and a proximal stimulatory promotersegment were functionally defined (35). Removal of the distalsequences results in significant increase of the responsivenessof this promoter to the TGF-b signal and Smad proteins (35).For this reason the transcriptional activity of the proximal2143/18 p21 promoter in response to TGF-b is relativelyhigher than the activity of the 22,300/18 promoter (Fig. 1 andRef. 35).

The p21 promoter studies in HaCaT cells (Fig. 1) and inHepG2 cells (35) are in strong agreement with the HaCaTexperiments of Fig. 2 in which exogenous Smad3 and Smad4were found to potentiate the response of the endogenous p21gene to TGF-b1. Overexpression of Smad3 and Smad4 bymeans of adenovirus infection could significantly enhance en-dogenous p21 accumulation, an effect that could be furtheraugmented by co-expression of the constitutively active type Ireceptor for TGF-b (CA-ALK5) or TGF-b1. This implies thatactivation of the Smads by receptors leads to more efficient p21gene activation. These effects on endogenous p21 protein accu-mulation are dose-dependent for all tested activators, i.e.the ligand TGF-b1, the constitutively active type I receptor,and the Smads, and thus, conditions where the cell can tolerateexcessive amounts of p21 accumulation can be obtained(Fig. 2A and data not shown). In addition, carboxyl-termi-nally truncated dominant negative mutants of Smad3 andSmad4 both inhibited p21 accumulation in response to

FIG. 7. A–C, the glutamine-rich regions of the transactivation domainof Sp1 are important for the functional cooperation with Smad proteins.A, diagrammatic representation of the GAL4-Sp1 fusion protein con-structs used in the transactivation assays of B and C. The same draw-ing conventions as in Fig. 6A are used. For mutants Sp1 Bc and itsderivative Sp1 Bc (Q3A) the respective wild type and triple glutamineto alanine substitutions in the amino acid sequence of the pertinentregion are shown with the substituted amino acids boxed. The relativesizes of the different proteins are not to scale. B, the glutamine-richregion of the transactivation domain B of Sp1 is important for func-tional cooperation with Smad proteins. HepG2 cells were transientlytransfected with the indicated Smad expression constructs and variousGAL4-Sp1 deletion mutant fusions (described in detail in A), togetherwith the 5 3GAL4 CAT and control b-galactosidase reporter constructs.Forty hours post-transfection cell lysates were assayed for CAT andb-galactosidase activities. The CAT activity normalized over the b-ga-lactosidase activity is plotted in a bar graph relative to the mocktransfection control, which is arbitrarily set at 1. The data representmeasurements from two independent experiments. Relative fold differ-ences between the plus and minus Smad3/Smad4 groups of data areshown on top of each pair of bars. C, the glutamine-rich region of Sp1exhibits stronger functional cooperation with Smad3. HepG2 cells weretransiently transfected, analyzed, and depicted as in B using the mu-tant GAL4-Sp1 Bc that contains primarily the glutamine-rich region ofrepeat B (see A) and the indicated combinations of Smad proteins.



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

TGF-b1 (Fig. 2, B and C), in a dose-dependent manner. Sincethese mutants are known to interfere with Smad activationby receptors, oligomerization, nuclear translocation, and co-operation with transcription factors (34, 44, 45), the Smadsignaling pathway must be required for endogenous p21 geneinduction by TGF-b. The same results were obtained withincreasing doses of the inhibitory Smad7. The combined dataof Figs. 1 and 2 strongly confirm that activation of Smads bythe type I receptor is a critical step in endogenous p21 re-sponsiveness to TGF-b1.

Smad proteins are known to associate directly with DNAelements containing the TCGTAGAC or G/C-rich sequences,although with relatively low affinity (36, 42, 43, 47–49). Thep21 proximal promoter lacks any obvious SBEs but contains aG/C-rich region between nucleotides 2124 and 242 (Fig. 1A).GEMSA experiments (Fig. 3) showed that Smad3 or Smad4does not bind to the G/C-rich region of the p21 promoter. Thetransactivation experiments using Smad3 and Smad4 proteinscontaining point mutations in their DNA-binding domains,which cannot recognize the SBE (Fig. 4), confirmed that p21promoter regulation by Smads does not require the DNA-bind-ing function of Smads. Control experiments using the multim-erized SBE promoter confirmed the defective nature of themutant Smads in HepG2 cells; however, their absolute nega-tive effects could not be estimated since HepG2 cells containendogenous Smad3 and Smad4. The mutant Smad3 andSmad4 proteins failed to transactivate the same SBE promoterin cells that lack the genes for Smad3 or Smad4.2 Finally, it isworth noting that despite the lack of SBE sequences in theproximal p21 promoter, such elements have been described inthe distal segment of the promoter (42, 50). However, previousdeletion analyses have shown that both TGF-b and Smad-de-pendent activation of the p21 promoter do not require thisdistal SBE (16, 35). The exact role of the distal SBE on thebasal and inducible activity of the p21 promoter remains to beelucidated.

The GEMSA analyses illustrated in Fig. 3 showed thatSmad3 and Smad4 proteins enhanced the formation of a nucle-oprotein complex between Sp1 and p21 promoter oligonucleo-tides. This observation could support a mechanism of cooperat-ivity between Smads and Sp1 in p21 promoter transactivation.On the other hand, binding of Smads to SBE sequences re-sulted in a stronger cooperativity with Sp1 (Fig. 3E). Such amechanism would imply that TGF-b-responsive promoters thatcontain both SBE sequences and G/C-rich Sp1-binding motifswould provide more optimal substrates for a cooperative func-tion between Smads and Sp1 in transcriptional regulation.Such examples might be the p15Ink4B, the Smad7, the TGF-b1,and the TGF-b type I and type II receptor promoters (55–57).

In contrast to previously published examples of nucleopro-tein complex formation between Smads and other transcriptionfactors (53, 58), we observed the presence of a distinct nucleo-protein complex with only slightly slower mobility than theSp1-DNA complex (Fig. 3D). This implies that the Sp1-Smadnucleoprotein complex cannot withstand GEMSA conditions orthat alternatively high affinity Sp1-DNA complex formationinduced by Smad proteins may rapidly lead to Smad dissocia-tion from the complex. A qualitatively similar result has beenobserved in studies of interaction and cooperation between theco-activator p300 and Sp1 (21). Thus, Smads could induce thefollowing: (a) oligomerization of Sp1, which results in higheraffinity binding to the p21 promoter; (b) recruitment of addi-tional cooperating factors (such as p300 or c-Jun) that couldpossibly stabilize or enhance the transcriptional activity of Sp1;and (c) modulation of the phosphorylation or acetylation statusof Sp1 with concomitant effects in DNA binding and transac-

tivation potencies. Recent reports indicate that the mitogen-activated protein kinase (MAPK) pathway is also activated byTGF-b and contributes positively to the regulation of the p21gene (59–61). Thus, it is of interest to examine whetherMAPKs directly modulate the transcriptional activity of Sp1,which is phosphorylated in a cell cycle-specific manner (20).Alternatively, transcription factor targets which themselvesare activated by the MAPK cascade and which are known tobind to the p21 promoter sequences, such as the Ets-like factorE1AF (62) might also cooperate with the Sp1-promotercomplex.

The obvious corollary of the above results has been thatSmads may physically interact with Sp1. This hypothesis wastested by co-immunoprecipitation and GST pull-down analyses(Figs. 5 and 6), which led to the conclusion that Smad2, Smad3,and Smad4 proteins all can directly interact with Sp1. Smad3showed the strongest constitutive association with Sp1 (Fig. 6,C and F). This result, combined with the strong transactivationpotential of this protein on various TGF-b-inducible promoters,can explain a series of previous data attesting to a dominantand ligand-independent function of Smad3 in activating Sp1-dependent transcriptional events (35). However, activation ofthe TGF-b signaling pathway by means of a constitutivelyactive type I receptor (ALK-5) increased the levels of Smad3species that co-immunoprecipitated with Sp1 in transientlytransfected COS-7 cells (Fig. 5B). In contrast, the constitutiveassociation of Smad2 and Smad4 with Sp1 is much weaker(Fig. 5B, 6C, and 6F). Type I receptor activation leads to muchstronger enhancement of Smad2 association with Sp1 whencompared with the enhancement seen for Smad3 (Fig. 5, Bversus C). Although the constitutive association of Smad4 withSp1 was readily detectable with two independent techniques(Fig. 6), the co-immunoprecipitation assay failed to measureconstitutive or ligand-dependent association (Fig. 5C). Sincethe constitutive Smad-Sp1 interactions were detected using invitro synthesized Smad proteins and bacterially purified Sp1(Fig. 6C), these interactions must be direct and do not requireany additional intermediates. However, this result does notexclude the possibility that additional factors participate in theSmad-Sp1 nuclear complex in vivo.

The same set of experiments resulted in mapping the Smadinteraction domain with Sp1 as the amino-terminal conservedMH1 domain (Fig. 6, C and F). Interestingly, although the MH1domain is the primary determinant for interactions with Sp1,the linker and MH2 domains contribute positively to the inter-action in a progressive manner, making the full-length proteinmore capable of associating with Sp1 (Fig. 6, C and F). TheMH1 domain of Smad proteins is known to associate specifi-cally with several other transcription factors, such as Jun fam-ily members, ATF-2, and vitamin D receptor (34). In addition,the MH1 domain contains the DNA-binding domain of theSmad proteins (47). In our efforts to finely map the interactiondomain in Smad3, we have collected preliminary data suggest-ing that sequences proximal to the Smad3 DNA-binding b-hair-pin domain may be responsible for the specific interaction withSp1 (data not shown). Thus, the structural model proposed byShi et al. (47) for the cooperation between Smad3 and Junfamily members (leucine zipper proteins) might also apply fortranscription factor Sp1 (a prototype for zinc finger proteins).

On the other hand, one possible domain of Sp1 that couldconfer specificity to the Sp1-Smad cooperativity maps to theconserved and duplicated glutamine-rich region of the Sp1transactivation domains A and B (Fig. 7). This hypothesis is inagreement with previous results showing the importance of theglutamine-rich sequences in TGF-b1-mediated activation ofGAL4-Sp1 fusion proteins (30). However, additional sequences



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

in Sp1 might play roles in the interaction with Smad proteins(Fig. 6I), a possibility that deserves more detailed analysis.Recently, we reported on the direct association of transcriptionfactors c-Jun and Sp1 (12). Similar to Smad proteins, the Sp1glutamine-rich segment exhibits functional cooperativity withc-Jun. On the other hand, c-Jun and Smad proteins interactdirectly via the MH1 domain of the latter (53, 54). Thus, onecan envision complex intermolecular interactions betweenthese three classes of transcription factors, whereby complexesof all three factors might simultaneously occur as discussed inour previous work (12). The importance of such higher ordercomplexes in the regulation of the p21 promoter by TGF-brequires future investigation.

In conclusion, the present data provide strong evidence forthe involvement of the Smad signaling pathway in both p21promoter and endogenous p21 gene regulation by TGF-b inHaCaT cells. In addition, we demonstrate the physical associ-ation of Smad proteins with transcription factor Sp1. One of theconsequences of such interactions is the apparent enhancementof the affinity of Sp1 for its cognate G/C-rich DNA element.Finally, the protein-protein interactions between Smads andSp1 can account for the synergistic regulation of the p21 pro-moter by these factors in response to TGF-b only partially, asadditional nuclear cofactors are most probably participating.Thus, the complexity of the mechanism by which the TGF-bsignal is integrated on the p21 promoter is gradually uncoveredand deserves further attention.

Acknowledgments—We are grateful to M. Fujii, K. Miyazono, and T.Fotsis for the adenoviruses; F. M. Hoffman and A. Comer for thebaculoviral Smad proteins; Santa Cruz Biotechnology for the anti-Sp1and anti-Smad1/2/3 (H2) antibodies; B. Vogelstein, R. Derynck, J.Massague, X.-F. Wang, G. Gill, G. Mavrothalassitis, E. Flavey, R. Tjian,J. Noti, C. Delidakis, C. Rorsman, and S. Itoh for various reporterplasmids, bacterial and mammalian expression vectors, and antibodiesused in this study. We also thank S. Itoh for technical advice; K.Shiraishi and G. Koutsodontis for excellent technical assistance; S.Grimsby for automated sequencing of plasmids; and C.-H. Heldin and J.Ericsson for discussions and a critical reading of this manuscript.

REFERENCES

1. Roberts, A. B., and Sporn, M. B. (1990) Handb. Exp. Pharmacol. Part 1,419–472

2. Massague, J. (1998) Annu. Rev. Biochem. 67, 753–7913. Alexandrow, M. G., and Moses, H. L. (1997) J. Cell. Biochem. 66, 427–4324. Reynisdottir, I., Polyak, K., Iavarone, A., and Massague, J. (1995) Genes Dev.

9, 1831–18455. Laiho, M., DeCaprio, J. A., Ludlow, J. W., Livingston, D. M., Massague, J.

(1990) Cell 62, 175–1856. Polyak, K., Kato, J. Y., Solomon, M. J., Sherr, C. J., Massague, J., Roberts,

J. M., and Koff, A. (1994) Genes Dev. 8, 9–227. Datto, M. B., Li, Y., Panus, J. F., Howe, D. J., Xiong, Y., and Wang, X.-F. (1995)

Proc. Natl. Acad. Sci. U. S. A. 92, 5545–55498. Hannon, G. J., and Beach, D. (1994) Nature 371, 257–2619. Iavarone, A, and Massague, J. (1997) Nature 387, 417–422

10. Iavarone, A, and Massague, J. (1999) Mol. Cell. Biol. 19, 916–92211. Hu, P. P., Datto, M. B., and Wang, X.-F. (1998) Endocr. Rev. 19, 349–36312. Kardassis, D., Papakosta, P., Pardali, K., and Moustakas, A. (1999) J. Biol.

Chem. 274, 29572–2958113. Liu, M., Iavarone, A., and Freedman, L. P. (1996) J. Biol. Chem. 271,

31723–3172814. Chin, Y. E., Kitagawa, M., Su, W.-C. S., You, Z.-H., Iwamoto, Y., and Fu, X.-Y.

(1996) Science 272, 719–72215. El-Deiry, W. S., Tokino, T., Waldman, T., Oliner, J. D., Velculescu, V. E.,

Burrell, M., Hill, D. E., Healy, E., Rees, J. L., Hamilton, S. R., Kinzler,K. W., and Vogelstein, B. (1995) Cancer Res. 55, 2910–2919

16. Datto, M. B., Yu, Y., and Wang, X.-F. (1995) J. Biol. Chem. 270, 28623–2862817. Berg, J. M. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 11109–1111018. Lania, L., Majello, B., and De Luca, P. (1997) Int. J. Biochem. Cell Biol. 29,

1313–132319. Marin, M., Karis, A., Visser, P., Grosveld, F., and Philipsen, S. (1997) Cell 89,

619–62820. Black, A. R., Jensen, D., Lin, S. Y., and Azizkhan, J. C. (1999) J. Biol. Chem.

274, 1207–121521. Suzuki, T., Kimura, A., Nagai, R., and Horikoshi, M. (2000) Genes Cells 5,

29–4122. Han, I., and Kudlow, J. E. (1997) Mol. Cell. Biol. 17, 2550–255823. Gill, G., Pascal, E., Tseng, Z. H., and Tjian, R. (1994) Proc. Natl. Acad. Sci.

U. S. A. 91, 192–19624. Saluja, D., Vassallo, M. F., and Tanese, N. (1998) Mol. Cell. Biol. 18,

5734–574325. Lee, J. S., Galvin, K. M., and Shi, Y. (1993) Proc. Natl. Acad. Sci. U. S. A. 90,

6145–614926. Udvadia, A. J., Templeton, D. J., and Horowitz, J. M. (1995) Proc. Natl. Acad.

Sci. U. S. A. 92, 3953–395727. Lin, S. Y., Black, A. R., Kostic, D., Pajovic, S., Hoover, C. N., and Azizkhan,

J. C. (1996) Mol. Cell. Biol. 16, 1668–167528. Hirano, F., Tanaka, H., Hirano, Y., Hiramoto, M., Handa, H., Makino, I., and

Scheidereit, C. (1998) Mol. Cell. Biol. 18, 1266–127429. Naar, A. M., Beaurang, P. A., Robinson, K. M., Oliner, J. D., Avizonis, D.,

Scheek, S., Zwicker, J., Kadonaga, J. T., and Tjian, R. (1998) Genes Dev. 12,3020–3031

30. Li, J. M., Datto, M. B., Shen, X., Hu, P. P., Yu, Y., and Wang, X.-F. (1998)Nucleic Acids Res. 26, 2449–2456

31. Heldin, C.-H., Miyazono, K., and ten Dijke, P. (1997) Nature 390, 465–47132. Derynck, R., Zhang, Y., and Feng, X. H. (1998) Cell 95, 737–74033. Attisano, L., and Wrana, J. L. (2000) Curr. Opin. Cell Biol. 12, 235–24334. ten Dijke, P., Miyazono, K., and Heldin, C.-H. (2000) Trends Biochem. Sci. 25,

64–7035. Moustakas, A., and Kardassis, D. (1998) Proc. Natl. Acad. Sc.i U. S. A. 95,

6733–673836. Johnson, K., Kirkpatrick, H., Comer, A., Hoffmann, F. M., and Laughon, A.

(1999) J. Biol. Chem. 274, 20709–2071637. Piek, E., Moustakas, A., Kurisaki, A., Heldin, C.-H., and ten Dijke, P. (1999)

J. Cell Sci. 122, 4557–456838. Jonk, L. J., Itoh, S., Heldin, C.-H., ten Dijke, P., and Kruijer, W. (1998) J. Biol.

Chem. 273, 21145–2115239. Biggs, J. R., Kudlow, J. E., and Kraft, A. S. (1996) J. Biol. Chem. 271, 901–90640. Shen, X., Hu, P. P., Liberati, N. T., Datto, M. B., Frederick, J. P., and Wang,

X.-F. (1998) Mol. Biol. Cell 9, 3309–331941. Fujii, M., Takeda, K., Imamura, T., Aoki, H., Sampath, T. K., Enomoto, S.,

Kawabata, M., Kato, M., Ichijo, H., and Miyazono, K. (1999) Mol. Biol. Cell10, 3801–3813

42. Zawel, L., Dai, J. L., Buckhaults, P., Zhou, S., Kinzler, K. W., Vogelstein, B.,and Kern, S. E. (1998) Mol. Cell 1, 611–617

43. Dennler, S., Itoh, S., Vivien, D., ten Dijke, P., Huet, S., and Gauthier, J. M.(1998) EMBO J. 17, 3091–3100

44. Zhang, Y., Feng, X.-H., Wu, R.-Y., and Derynck, R. (1996) Nature 383, 168–17245. Lagna, G., Hata, A., Hemmati-Brivanlou, A., and Massague, J. (1996) Nature

383, 832–83646. Nakao, A., Afrakhte, M., Moren, A., Nakayama, T., Christian, J. L., Heuchel,

R., Itoh, S., Kawabata, M., Heldin, N. E., Heldin, C.-H., and ten Dijke, P.(1997) Nature 389, 631–635

47. Shi, Y., Wang, Y. F., Jayaraman, L., Yang, H., Massague, J., and Pavletich,N. P. (1998) Cell 94, 585–594

48. Labbe, E., Silvestri, C., Hoodless, P. A., Wrana, J. L., and Attisano, L. (1998)Mol. Cell 2, 109–120

49. Kusanagi, K., Inoue, H., Ishidou, Y., Mishima, H. K., Kawabata, M., andMiyazono, K. (2000) Mol. Biol. Cell 11, 555–565

50. Hunt, K. K., Fleming, J. B., Abramian, A., Zhang, L., Evans, D. B., and Chiao,P. J. (1998) Cancer Res. 58, 5656–5661

51. Courey, A. J., and Tjian, R. (1988) Cell 55, 887–89852. Moses, H. L. (1992) Mol. Reprod. Dev. 32, 179–18453. Zhang, Y., Feng, X. H., and Derynck, R. (1998) Nature 394, 909–913;

Correction (1998) Nature 396, 49154. 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

55. Li, J. M., Nichols, M. A., Chandrasekharan, S., Xiong, Y., and Wang, X.-F.(1995) J. Biol. Chem. 270, 26750–26753

56. Nagarajan, R. P., Zhang, J., Li, W., and Chen, Y. (1999) J. Biol. Chem. 274,33412–33418

57. Kim, Y., Ratziu, V., Choi, S. G., Lalazar, A., Theiss, G., Dang, Q., Kim, S. J.,and Friedman, S. L. (1998) J. Biol. Chem. 273, 33750–33758

58. Yeo, C. Y., Chen, X., and Whitman, M. (1999) J. Biol. Chem. 274, 26584–2659059. Bottazzi, M. E., Zhu, X., Bohmer, R. M., and Assoian, R. K. (1999) J. Cell Biol.

146, 1255–126460. Kivinen, L., and Laiho, M. (1999) Cell Growth Differ. 10, 621–62861. Hu, P. P., Shen, X., Huang, D., Liu, Y., Counter, C., and Wang, X.-F. (1999)

J. Biol. Chem. 274, 35381–3538762. Funaoka, K., Shindoh, M., Yoshida, K., Hanzawa, M., Hida, K., Nishikata, S.,

Totsuka, Y., and Fujinaga, K. (1997) Biochem. Biophys. Res. Commun. 236,79–82



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Aristidis MoustakasKaterina Pardali, Akira Kurisaki, Anita Morén, Peter ten Dijke, Dimitris Kardassis and

βby Transforming Growth Factor- RegulationWaf1/Cip1Role of Smad Proteins and Transcription Factor Sp1 in p21

doi: 10.1074/jbc.M909467199 originally published online June 30, 20002000, 275:29244-29256.J. Biol. Chem.

10.1074/jbc.M909467199Access 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/275/38/29244.full.html#ref-list-1

This article cites 61 references, 36 of which can be accessed free at


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M909467199

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;275/38/29244&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/275/38/29244

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=275/38/29244&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/275/38/29244

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/275/38/29244.full.html#ref-list-1

http://www.jbc.org/

the j b c vol. 275, no. 38, issue of september 22, pp ... · role of smad proteins and...

Documents