The Tumor Suppressor p53 Abrogates Smad-dependent Collagen Gene Induction in Mesenchymal Cells

The pleiotropic cytokine transforming growth factor-β (TGF-β) is a potent inducer of collagen synthesis and is implicated in the pathogenesis of fibrosis. Acting in concert with transcriptional coactivators p300/CBP, the Smads mediate TGF-β stimulation of collagen synthesis in human dermal fibroblasts. Little information exists regarding positive and negative modulation of physiological TGF-β responses. Because the tumor suppressor p53 is implicated in connective tissue homeostasis, here we examined the regulation of collagen gene expression by p53. Forced expression of ectopic p53 in dermal fibroblasts repressed basal and TGF-β-stimulated collagen gene expression, whereas the absence of cellular p53 was associated with significantly enhanced transcriptional activity of the Type I collagen gene (COL1A2) and collagen synthesis. Ectopic expression of p53 also repressed TGF-β stimulation of promoter activity driven by minimal Smad-binding elements, suggesting that p53 modulated Smad-dependent intracellular signaling. Inhibition was not due to altered levels, phosphorylation, or nuclear translocation of cellular Smads. Treatment of fibroblasts with etoposide, a potent inducer of cellular p53, abrogated TGF-β stimulation of COL1A2 promoter activity and collagen synthesis in a p53-dependent manner. Overexpression of the transcriptional coactivator p300 rescued TGF-β stimulation of COL1A2 promoter activity in fibroblasts overexpressing p53. Furthermore, the ligand-induced interaction of cellular Smad3 with p300 or with its cognate Smad-binding DNA element and recruitment of p300 to the DNA-protein complex assembled on the Smad-binding element were markedly reduced in p53-overexpressing fibroblasts. Collectively, these results indicate, for the first time, that p53 is a potent and selective endogenous repressor of TGF-β-regulated collagen gene expression in dermal fibroblasts. The ligand-dependent interaction of Smad3 with p300 may be one of the targets of p53-mediated inhibition of TGF-β responses. These findings suggest that a novel and important physiologic function for the tumor suppressor p53 is the regulation of fibrotic cellular responses. The pleiotropic cytokine transforming growth factor-β (TGF-β) is a potent inducer of collagen synthesis and is implicated in the pathogenesis of fibrosis. Acting in concert with transcriptional coactivators p300/CBP, the Smads mediate TGF-β stimulation of collagen synthesis in human dermal fibroblasts. Little information exists regarding positive and negative modulation of physiological TGF-β responses. Because the tumor suppressor p53 is implicated in connective tissue homeostasis, here we examined the regulation of collagen gene expression by p53. Forced expression of ectopic p53 in dermal fibroblasts repressed basal and TGF-β-stimulated collagen gene expression, whereas the absence of cellular p53 was associated with significantly enhanced transcriptional activity of the Type I collagen gene (COL1A2) and collagen synthesis. Ectopic expression of p53 also repressed TGF-β stimulation of promoter activity driven by minimal Smad-binding elements, suggesting that p53 modulated Smad-dependent intracellular signaling. Inhibition was not due to altered levels, phosphorylation, or nuclear translocation of cellular Smads. Treatment of fibroblasts with etoposide, a potent inducer of cellular p53, abrogated TGF-β stimulation of COL1A2 promoter activity and collagen synthesis in a p53-dependent manner. Overexpression of the transcriptional coactivator p300 rescued TGF-β stimulation of COL1A2 promoter activity in fibroblasts overexpressing p53. Furthermore, the ligand-induced interaction of cellular Smad3 with p300 or with its cognate Smad-binding DNA element and recruitment of p300 to the DNA-protein complex assembled on the Smad-binding element were markedly reduced in p53-overexpressing fibroblasts. Collectively, these results indicate, for the first time, that p53 is a potent and selective endogenous repressor of TGF-β-regulated collagen gene expression in dermal fibroblasts. The ligand-dependent interaction of Smad3 with p300 may be one of the targets of p53-mediated inhibition of TGF-β responses. These findings suggest that a novel and important physiologic function for the tumor suppressor p53 is the regulation of fibrotic cellular responses. Transforming growth factor-β (TGF-β) 1The abbreviations used are: TGF-β, transforming growth factor-β; CAT, chloramphenicol acetyltransferase; MEF, mouse embryo fibroblast.1The abbreviations used are: TGF-β, transforming growth factor-β; CAT, chloramphenicol acetyltransferase; MEF, mouse embryo fibroblast. is the prototype of a large superfamily of multifunctional cytokines that control cellular growth and differentiation. In a variety of mesenchymal cells, TGF-β induces the synthesis of collagens and other extracellular matrix components and is thus a pivotal contributor to pathological fibrosis (1Blobe G.C. Schiemann W.P. Lodish H.F. N. Engl. J. Med. 2000; 342: 1350-1358Crossref PubMed Scopus (2171) Google Scholar). The molecular mechanisms that govern the regulation of major extracellular matrix gene expression in response to TGF-β are the subject of intense investigation (2Ghosh A.K. Exp. Biol. Med. 2002; 227: 301-314Crossref PubMed Scopus (214) Google Scholar). The recent discovery of Smads as novel TGF-β signal transducers opens a new avenue for fibrosis research (see Ref. 3Varga J. Arthritis Rheum. 2002; 46: 1703-1713Crossref PubMed Scopus (113) Google Scholar and references therein). The Smad family consists of eight members that can be classified into three subgroups based on their structure and function. Smad1, Smad2, Smad3, Smad5, and Smad8 are receptor-activated Smads, or R-Smads; Smad4 is a co-Smad; and Smad6 and Smad7 are inhibitory (4Shi Y. Massague J. Cell. 2003; 113: 685-700Abstract Full Text Full Text PDF PubMed Scopus (4764) Google Scholar). We have shown previously that Smad3 was responsible for mediating TGF-β-induced stimulation of collagen synthesis in skin fibroblasts, whereas Smad7 abrogated this response (5Chen S-J. Yuan W. Mori Y. Levenson A. Trojanowska M. Varga J. J. Invest. Dermatol. 1999; 112: 49-57Abstract Full Text Full Text PDF PubMed Scopus (370) Google Scholar). Furthermore, we demonstrated that the transcriptional coactivator p300 physically and functionally interacted with Smad3 in a ligand-dependent manner and played a major role in Smad-dependent stimulation of collagen gene expression (6Ghosh A.K. Yuan W. Mori Y. Varga J. Oncogene. 2000; 19: 3546-3555Crossref PubMed Scopus (197) Google Scholar, 7Ghosh A.K. Yuan W. Mori Y. Chen S-J. Varga J. J. Biol. Chem. 2001; 276: 11041-11048Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar). The mechanisms that control the magnitude or duration of Smad-dependent TGF-β-mediated cellular responses are only partially understood. Because deregulated TGF-β signaling is implicated in the pathogenesis of fibrosis in scleroderma and related conditions (8Varga J. Mori Y. Takagawa S. Semin. Clin. Immunol. 2001; 2: 15-29Google Scholar), identification and characterization of transcriptional cofactors that modulate this process is important. The tumor suppressor p53 is a short lived nuclear phosphoprotein that is mutated in a majority of human cancers. It plays important roles in the regulation of cell growth, apoptosis, differentiation, and senescence (9Somasundaram K. El-Deiry W.S. Front. Biosci. 2000; 5: 424-437Crossref PubMed Google Scholar, 10Gudkov A.V. Nat. Med. 2002; 8: 1196-1198Crossref PubMed Scopus (43) Google Scholar). These activities of p53 are mediated through inhibition or stimulation of target gene expression during cell cycle (11Levine A. J. Cell Sci. 1992; 113: 1661-1670Google Scholar, 12Yu J. Zhang L. Hwang P.M. Rago C. Kinzler K.W. Vogelstein B. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 14517-14522Crossref PubMed Scopus (414) Google Scholar). Whereas activation of transcription by p53 is generally due to its direct interaction with regulatory cis-elements of target genes, repression involves protein-protein interaction between p53 and other transcription factors, modulation of the activity of the transcriptional coactivator p300, or recruitment of histone deacetylase in the transcriptional complex (9Somasundaram K. El-Deiry W.S. Front. Biosci. 2000; 5: 424-437Crossref PubMed Google Scholar, 13Kunz C. Pebler S. Otte J. von der Ahe D. Nucleic Acids Res. 1995; 23: 3710-3717Crossref PubMed Scopus (121) Google Scholar, 14Kern S.E. Pietenpol J.A. Thiagalingam S. Seymour A. Kinzler K.W. Vogelstein B. Science. 1992; 256: 827-830Crossref PubMed Scopus (888) Google Scholar, 15El-Deiry W.S. Kern S.E. Pientenpol J.A. Kinzler K.W. Vogelstein B. Nat. Genet. 1992; 1: 45-49Crossref PubMed Scopus (1738) Google Scholar, 16Gu W. Roeder R.G. Cell. 1997; 90: 595-606Abstract Full Text Full Text PDF PubMed Scopus (2159) Google Scholar, 17Gu W. Shi X-L. Roeder R.G. Nature. 1997; 387: 819-823Crossref PubMed Scopus (520) Google Scholar, 18Lill N.L. Grossman S.R. Ginsberg D. Decaprio J. Livingston D.M. Nature. 1997; 387: 823-827Crossref PubMed Scopus (594) Google Scholar, 19Juan L.-J. Shia W.-J. Chen M.-H. Yang W.-M. Seto E. Lin Y.-S. Wu C-W. J. Biol. Chem. 2000; 275: 20436-20443Abstract Full Text Full Text PDF PubMed Scopus (359) Google Scholar, 20Grossman S.R. Eur. J. Biochem. 2001; 268: 2773-2778Crossref PubMed Scopus (190) Google Scholar). Recent studies suggest that the tumor suppressor p53 plays an important role in regulating extracellular matrix homeostasis. Ectopic expression of p53 in fibroblasts inhibited the formation of fibronectin fibrils (21Alexandrova A. Ivanov A. Chumakov P. Kopnin B. Vasiliev J. Oncogene. 2000; 19: 5826-5830Crossref PubMed Scopus (39) Google Scholar), whereas inhibition of p53 expression in HeLa cells resulted in increased fibronectin synthesis (22Iotsova V. Stehelin D. Cell Growth & Differ. 1996; 7: 629-634PubMed Google Scholar). Furthermore, p53 has been shown to repress plasminogen activator (PA) and stimulate plasminogen activator inhibitor-1 (PAI-1) gene expression in mesenchymal cells (13Kunz C. Pebler S. Otte J. von der Ahe D. Nucleic Acids Res. 1995; 23: 3710-3717Crossref PubMed Scopus (121) Google Scholar). In mouse skin fibroblasts, wild type p53, but not its mutants, repressed the synthesis of matrix metalloproteinase (MMP)-1 and MMP-13, while stimulating the synthesis of MMP-2 (23Bian J. Sun Y.I. Mol. Cell. Biol. 1997; 17: 6330-6338Crossref PubMed Scopus (249) Google Scholar, 24Sun Y. Sun Y. Wegner L. Rutter J.L. Brinckerhoff C.E. Cheung H.S. J. Biol. Chem. 1999; 274: 11535-11540Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar, 25Sun Y. Cheung J.M. Martel-Pelletier J. Pelletier J.P. Wegner L. Altman R.D. Howell D.S. Cheung H.S. J. Biol. Chem. 2000; 275: 11327-11332Abstract Full Text Full Text PDF PubMed Scopus (84) Google Scholar, 26Ala-aho R. Grenman R. Seth P. Kahari V.-M. Oncogene. 2002; 21: 1187-1195Crossref PubMed Scopus (62) Google Scholar). The transcription of the tissue inhibitor of metalloproteinase-3 (TIMP-3) gene was repressed by p53 (27Loging W.T. Reisman D. Oncogene. 1999; 18: 7608-7615Crossref PubMed Scopus (35) Google Scholar). These results, indicating that p53 can positively or negatively modulate the expression of multiple extracellular matrix genes, suggest a novel function for p53 in physiological regulation of connective tissue homeostasis. The significance and molecular mechanisms underlying p53 regulation of extracellular matrix gene expression are largely unknown. The p53 protein undergoes activation via site-specific phosphorylation, dephosphorylation, and acetylation in response to different forms of cellular stresses (9Somasundaram K. El-Deiry W.S. Front. Biosci. 2000; 5: 424-437Crossref PubMed Google Scholar, 28Sakaguchi K. Herrera J.F. Saito S. Miki T. Bustin M. Vassilev A. Anderson C.W. Appella E. Genes Dev. 1998; 12: 2831-2841Crossref PubMed Scopus (1018) Google Scholar). In addition, the activity of p53 is also regulated via its degradation and subcellular localization (29Ryan K.M. Phillips A.C. Vousden K.H. Curr. Opin. Cell Biol. 2001; 13: 332-337Crossref PubMed Scopus (574) Google Scholar). Complex interactions between p53 and the TGF-β signaling pathway have been demonstrated. It has been shown that p53 modulates TGF-β-mediated stimulation or inhibition of cellular proliferation (30Miyazaki M. Ohashi R. Tsuji T. Mihara K. Gohda E. Namba M. Biochem. Biophys. Res. Commun. 1998; 246: 873-880Crossref PubMed Scopus (58) Google Scholar, 31Dkhissi F. Raynal S. Jullien P. Lawrence D.A. Oncogene. 1999; 18: 703-711Crossref PubMed Scopus (20) Google Scholar). Although p53 is known to modulate the expression of extracellular matrix protein genes and regulate TGF-β responses, the physiologic role of p53 in Smad-dependent TGF-β regulation of collagen gene expression has not been examined. We now report that expression of ectopic p53 in normal dermal fibroblasts resulted in repression of TGF-β-stimulated collagen gene transcription and Smad-dependent responses. In contrast, the TGF-β-induced PAI-1 promoter activity was enhanced by overexpressed p53, in agreement with a previous report (32Cordenonsi M. Dupont S. Maretto S. Insinga A. Imbriano C. Piccolo S. Cell. 2003; 113: 301-314Abstract Full Text Full Text PDF PubMed Scopus (322) Google Scholar). Type I collagen gene (COL1A2) promoter activity and collagen synthesis were significantly increased in murine embryonic fibroblasts lacking p53, indicating the physiological significance of cellular p53 in regulation of collagen gene expression. Etoposide, an inducer of p53 expression, repressed collagen synthesis in wild type but not in p53-null fibroblasts. Repression of Smad-dependent TGF-β responses was not due to alteration in the expression levels, phosphorylation, or nuclear translocation of Smads. Transient overexpression of p300 rescued TGF-β-induced COL1A2 promoter stimulation in the presence of p53. Furthermore, induction of cellular p53 resulted in repression of TGF-β-induced Smad3 interaction with the Smad-binding element and with p300. Collectively, these results demonstrate, for the first time, that cellular p53 is a potent negative modulator of TGF-β signaling, with potentially important role in regulating fibrotic responses in normal dermal fibroblasts. These results extend the range of biological activities attributed to p53. Cell Culture—Primary cultures of human dermal fibroblasts were established from anonymous newborn foreskin specimens (5Chen S-J. Yuan W. Mori Y. Levenson A. Trojanowska M. Varga J. J. Invest. Dermatol. 1999; 112: 49-57Abstract Full Text Full Text PDF PubMed Scopus (370) Google Scholar). Human fibroblasts and mouse embryonic fibroblasts (MEFs) derived from p53-null and wild type mice were maintained in Eagle's minimum essential medium or Dulbecco's modified Eagle's medium (Biowhittaker, Walkersville, MD), respectively, supplemented with 10% fetal bovine serum (Invitrogen), 1% vitamins, 1% penicillin/streptomycin, and 2 mm l-glutamine. Primary fibroblasts were used at passages 3–8. Plasmids—The COL1A2-CAT reporter constructs containing human COL1A2 promoter fragments with 5′ end points at –772, –353, –186, and –108 have been described previously (33Ihn H. LeRoy E.C. Trojanowska M. J. Biol. Chem. 1997; 272: 24666-24672Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar). The p53-luc reporter construct contains 14 copies of the minimal p53 response element in front of a basic promoter element and luciferase reporter gene (Stratagene, La Jolla, CA). The –460 PAI-1-CAT construct contains –460 bp of the human PAI-1 promoter (34Vulin A.G. Stanley F.M. J. Biol. Chem. 2002; 277: 20169-20176Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar). The SBE4-TK-luc construct contains four copies of the consensus Smad-binding elements in front of TK and luciferase reporter (35Zawel L. Dai J.L. Buckhaults P. Zhou S. Kinzler K.W. Vogelstein B. Kern S.E. Mol. Cell. 1998; 1: 611-617Abstract Full Text Full Text PDF PubMed Scopus (889) Google Scholar). The COL1A2-CAGA-TK-luc construct contains six copies of the COL1A2 CAGACA element in front of the TK promoter and luciferase reporter (36Dennler S. Itoh S. Vivien D. ten Dijke P. Huet S. Gauthier J-M. EMBO J. 1998; 17: 3091-3100Crossref PubMed Scopus (1578) Google Scholar). The pCI p300 construct contains the full-length p300 cDNA (37Boyes J. Byfield P. Nakatani Y. Ogryzko V. Nature. 1998; 396: 594-598Crossref PubMed Scopus (633) Google Scholar). The pEGFP-N-FLAG-Smad3 encodes full-length Smad3 in the pEGFPN1 expression vector (38Liu X. Sun Y. Constantinescu S.N. Karam E. Weinberg R.A. Lodish H.F. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 10669-10674Crossref PubMed Scopus (331) Google Scholar). Wild type CMV-pC53-SN3 and mutant CMV-p53–175 (Arg to His at 175), CMV-p53–248 (Arg to Trp at 248), and CMV-p53–273 (Arg to His at 273) p53 constructs were made by inserting wild type or mutant p53 cDNA in the pCMVNeo-Bam expression vector (14Kern S.E. Pietenpol J.A. Thiagalingam S. Seymour A. Kinzler K.W. Vogelstein B. Science. 1992; 256: 827-830Crossref PubMed Scopus (888) Google Scholar). The p53–213 (Arg to stop codon at 213) and p53–239 (Asp to Ser at 239) constructs were made by inserting rheumatoid synovium-derived mutant p53 cDNA in pCI vector (39Han Z. Boyle D.L. Shi Y. Green D.R. Firestein G.S. Arthritis Rheum. 1999; 42: 1088-1192Crossref PubMed Scopus (80) Google Scholar). Transient Transfections and Reporter Assays—Dermal fibroblasts or MEFs grown in 6-well or 12-well clusters were transiently transfected using Superfect Reagent (Qiagen, Valencia, CA) following the manufacturer's instructions, and CAT or luciferase assays were performed as described previously (40Ghosh A.K. Bhattacharyya S. Lakos G. Chen S.-J. Mori Y. Varga J. Arthritis Rheum. 2004; 50: 1305-1318Crossref PubMed Scopus (180) Google Scholar). The reporter constructs were transfected along with expression vectors or appropriate empty vectors and Renilla luciferase expression vector pRLTK-LUC as an internal standard. Transfected cells were incubated in fresh Eagle's minimum essential medium containing 10% fetal bovine serum with or without TGF-β2 (12.5 ng/ml) (Genzyme Corp., Framingham, MA) or etoposide (0.1–1 μm) (Sigma). Following a further 48-h incubation, fibroblasts were harvested, and cell lysates were prepared using 1× passive lysis buffer (Promega Corp, Madison, WI) and used for determination of CAT or luciferase activities. The results were normalized with protein concentrations, and transfection efficiency was monitored by measuring Renilla luciferase activity. Experiments were performed in triplicates and repeated two or three times with consistent results. Immunoprecipitation and Immunoblot Analysis—Whole cell lysates were prepared from dermal fibroblasts or MEFs using lysis buffer (Promega) or M-Per mammalian protein extraction buffer (Pierce) and centrifuged at 4 °C for 5 min at 10,000 rpm. Nuclear extracts or cytoplasmic extracts were prepared from transfected or etoposide-treated dermal fibroblasts as described previously (41Andrews N.C. Faller D.V. Nucleic Acids Res. 1991; 19: 2499Crossref PubMed Scopus (2210) Google Scholar). Equal amounts of proteins (10–25 μg) or conditioned media (40 μl) were resolved by electrophoresis on 4–20% Tris/glycine gradient gels (Bio-Rad). For immunoprecipitation, cell lysates were prepared and immunoprecipitated as described previously (7Ghosh A.K. Yuan W. Mori Y. Chen S-J. Varga J. J. Biol. Chem. 2001; 276: 11041-11048Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar). Immunoprecipitated proteins were subjected to electrophoresis and transferred to polyvinylidene difluoride membranes. The polyvinylidene difluoride membranes were blocked with 10% fat-free milk in TBST buffer and incubated with antibodies against human Type I collagen (Southern Biotechnology Associates, Birmingham, AL), PAI-1 (H-135), p53 (DO-1), Smad7 (N-19), Smad4 (B8), Smad1/2/3 (H2), or actin (C-2) or p300 (C-20) (all from Santa Cruz Biotechnology, Inc., Santa Cruz, CA) or Smad 3 (Zymed Laboratories Inc.) or p-Smad2 (Cell Signaling, Beverly, MA). The membranes were washed with TBST buffer and incubated with appropriate secondary antibodies for 45 min at room temperature. After washing with TBST, membranes were treated with ECL reagent (Amersham Biosciences) and exposed to Eastman Kodak Co. XAR5 film. Electrophoretic Mobility Shift Assays—Confluent fibroblasts were incubated with TGF-β in the presence or absence of etoposide (1 μm). At the end of incubation, nuclear extracts were prepared and analyzed by electrophoretic mobility shift assays as described (7Ghosh A.K. Yuan W. Mori Y. Chen S-J. Varga J. J. Biol. Chem. 2001; 276: 11041-11048Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar). For this purpose, equal aliquots (5 μg) were incubated for 30 min in binding buffer in the presence or absence of a 100-fold molar excess of cold competitors. Antibody super-shift assays were performed using anti-Smad3 antibody (I-20) (Santa Cruz Biotechnology) or control IgG. Radiolabeled Smad-binding element (SBE) probes were added to the reaction mixture and incubated for another 30 min, and DNA-protein complexes were resolved in 6% polyacrylamide gels. Gels were then dried and exposed for autoradiographs. DNA Affinity Precipitation Assays—The presence of Smad3 and p300 in the DNA-binding complex assembled on the SBE probe was examined by DNA-protein interaction assays (42Deng W.-G. Zhu Y. Montero A. Wu K.K. Anal. Biochem. 2003; 323: 12-18Crossref PubMed Scopus (66) Google Scholar). Briefly, fibroblasts incubated with TGF-β for 1 h in the presence or absence of etoposide (1 μm) were harvested, nuclear extracts were prepared, and equal amounts of protein (∼200 μg) were incubated with biotin-labeled double-stranded SBE probes for 30 min. At the end of incubation, 40 μl of streptavidin-agarose beads (4%) with 50% slurry (Sigma) were added, and mixtures were incubated at 4 °C for 45 min. The streptavidin-agarose beads were precipitated by centrifugation, pellets were washed three times with cold phosphate-buffered saline, and bead-bound proteins were resuspended in SDS-loading buffer followed by electrophoresis on 4–20% denaturing gels. Gels were processed for immunoblot analysis using antibodies against Smad3 and p300 as above. The p53 levels in nuclear extract were determined by immunoblot analysis using antibodies against p53 or Smad4. Statistical Analysis—The data are presented as means ± S.D. Statistical differences between experimental and control groups were determined by analysis of variance, and a value of p < 0.05 by Student's t test was considered significant. Forced Expression of p53 Selectively Inhibits Type I Collagen Synthesis and Abrogates Its Stimulation by TGF-β—In addition to its well characterized role as a tumor suppressor, p53 is increasingly recognized as a regulator of extracellular matrix synthesis. The molecular mechanisms underlying these important effects of p53 are largely unknown. To characterize the effects of p53 on collagen synthesis, normal dermal fibroblasts were transiently transfected with expression vector for p53 or empty vector, followed by incubation with TGF-β (12.5 ng/ml). At the end of the 48-h incubation, whole cell lysates were prepared and subjected to immunoblot analysis. The results showed that whereas TGF-β induced a marked increase in collagen, as expected, ectopic expression of p53 resulted in significantly reduced cellular levels of collagen both in the presence and absence of TGF-β (Fig. 1A). The cellular levels of actin remained unaltered, and no effect on cell viability and protein concentration was seen. To investigate the mechanism of collagen gene regulation by p53, fibroblasts were transiently transfected with expression vector for p53 or appropriate empty vector, along with the 772COL1A2-CAT or 460PAI-1-CAT reporter constructs or p53-luc as positive controls. After a 48-h incubation with TGF-β, cultures were harvested, and CAT or luciferase activities were determined. The results of transient transfection assays showed that ectopic expression of p53 resulted in repression of basal COL1A2 promoter-driven transcriptional activity and prevented its stimulation induced by TGF-β (Fig. 1B, left panel). The inhibitory effect of p53 appeared to be specific for the COL1A2 promoter, since under identical experimental conditions, the basal and TGF-β-induced activities of PAI-1 promoter (containing a consensus p53 binding element) were stimulated by p53 (Fig. 1B, middle panel). The activity of p53-luc, a minimal reporter construct driven by p53 response elements, was increased by 2-fold (Fig. 1B, right panels), and no effect on TK promoter activity was noted (data not shown). In order to investigate the mechanistic basis for p53-mediated transcriptional repression, well characterized mutants of p53 were used in transiently transfected fibroblasts. The results showed that forced expression of tumor-derived p53 mutants unable to bind to DNA (14Kern S.E. Pietenpol J.A. Thiagalingam S. Seymour A. Kinzler K.W. Vogelstein B. Science. 1992; 256: 827-830Crossref PubMed Scopus (888) Google Scholar) resulted in a substantial decrease in TGF-β-induced stimulation of COL1A2 promoter activity (Fig. 1C, left panel). In contrast, ectopic expression of dominant negative mutants derived from rheumatoid synovium (39Han Z. Boyle D.L. Shi Y. Green D.R. Firestein G.S. Arthritis Rheum. 1999; 42: 1088-1192Crossref PubMed Scopus (80) Google Scholar) failed to abrogate TGF-β stimulation (Fig. 1C, right panel), despite comparable expression levels of wild type and mutant p53. Because these p53 mutants are also unable to bind to DNA (43Kaku S. Albor A. Kulesz-Martin M. Biochem. Biophys. Res. Commun. 2001; 280: 204-211Crossref PubMed Scopus (6) Google Scholar), the results indicate that repression of COL1A2 by p53 is independent of direct DNA interaction. Etoposide-induced Cellular p53 Abrogates TGF-β Stimulation of COL1A2 Promoter Activity and Collagen Synthesis— Etoposide, a topoisomerase II inhibitor, is a potent stimulus for p53 accumulation in a variety of cell types. Etoposide has been shown to regulate the expression of MMP-1, MMP-2, p21, and other p53 target genes (23Bian J. Sun Y.I. Mol. Cell. Biol. 1997; 17: 6330-6338Crossref PubMed Scopus (249) Google Scholar, 24Sun Y. Sun Y. Wegner L. Rutter J.L. Brinckerhoff C.E. Cheung H.S. J. Biol. Chem. 1999; 274: 11535-11540Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar, 44Tishler R.B. Calderwood S.K. Coleman C.N. Price B.D. Cancer Res. 1993; 53: 2212-2216PubMed Google Scholar, 45Whitacre C.M. Hashimoto H. Tsai M-L. Chatterjee S. Berger S.J. Berger N.A. Cancer Res. 1995; 55: 3697-3701PubMed Google Scholar, 46Ding H. Duan W. Zhu W.G. Ju R. Srinivasan K. Otterson G.A. Villalona-Calero M.A. Biochem. Biophys. Res. Commun. 2003; 305: 950-956Crossref PubMed Scopus (31) Google Scholar). In order to examine the role of cellular p53 in regulation of collagen gene expression, fibroblasts transiently transfected with 772COL1A2-CAT reporter construct were incubated with etoposide (0.1–1 μm) in the presence and absence of TGF-β. After 48 h, fibroblasts were harvested, and CAT activities were determined. Etoposide by itself had only minimal effect on the activity of transfected 772COL1A2-CAT, but it abrogated TGF-β-induced stimulation in a dose-dependent manner (Fig. 2A). Next, the effect of etoposide on cellular collagen levels was examined. The results showed that treatment of fibroblasts with etoposide for 48 h substantially attenuated TGF-β-induced stimulation of collagen synthesis (Fig. 2B). In contrast, the levels of basal as well as TGF-β-induced PAI-1 were elevated in fibroblasts incubated with etoposide (Fig. 2B). The expression of p53 was induced by etoposide, as expected. At the concentrations used, etoposide had no effect on cellular viability, protein levels, or pRLTK promoter activity (data not shown). Together, these results indicated that etoposide-induced stimulation of cellular p53 was associated with selective repression of basal and TGF-β-stimulated collagen gene expression in dermal fibroblasts. Etoposide has multiple effects in addition to enhancing cellular p53. To further investigate the physiological significance of cellular p53 in the regulation of COL1A2 promoter activity, MEFs lacking p53 were used. For this purpose, wild type or p53-null MEFs were transiently transfected with 772COL1A2-CAT and incubated with etoposide (1 μm) in the absence or presence of TGF-β. At the end of a 48-h incubation, cells were harvested, and CAT activities were determined. The results of transient transfection assays revealed that in unstimulated p53-null MEFs, COL1A2 promoter activity was 3-fold higher than in wild type controls (Fig. 3A). Forced expression of ectopic p53 repressed basal and TGF-β-induced promoter activity to a comparable degree in both wild type and p53-null MEFs (data not shown). In contrast, whereas etoposide caused significant inhibition of TGF-β-stimulated COL1A2 promoter activity in wild type MEFs (∼90%), in p53-null MEFs etoposide was completely unable to reduce TGF-β stimulatory response (Fig. 3A). The regulation of endogenous collagen synthesis by cellular p53 was next studied in wild type and p53-null MEFs. The MEFs were incubated with etoposide in the presence and absence of TGF-β for 48 h, and culture supernatants were analyzed by Western immunoblot. The results showed that levels of basal and TGF-β-induced Type I collagen were significantly higher in p53-null MEFs compared with wild type controls. Furthermore, incubation of MEFs with etoposide failed to repress the basal and TGF-β-stimulated collagen synthesis in the absence of cellula