Title: Modulation of Transforming Growth Factor β (TGFβ)/Smad Transcriptional Responses through Targeted Degradation of TGFβ-inducible Early Gene-1 by Human Seven in Absentia Homologue
Abstract: Transforming growth factor β (TGFβ)-inducible early gene-1 (TIEG1) is a Krüppel-like transcription factor that is rapidly induced upon TGFβ treatment. TIEG1 promotes TGFβ/Smad signaling by down-regulating negative feedback through the inhibitory Smad7. In this report, we describe the identification of an E3 ubiquitin ligase, Seven in Absentia homologue-1 (SIAH1), as a TIEG1-interacting protein. We show that TIEG1 and SIAH1 interact through an amino-terminal domain of TIEG1. Co-expression of SIAH1 results in proteasomal degradation of TIEG1 but not of the related factor TIEG2. Importantly, co-expression of SIAH1 completely reverses repression of Smad7 promoter activity by TIEG1. Furthermore, overexpression of a dominant negative SIAH1 stabilizes TIEG1 and synergizes with TIEG1 to enhance TGFβ/Smad-dependent transcriptional activation. These findings suggest a novel mechanism whereby the ability of TGFβ to modulate gene transcription may be regulated by proteasomal degradation of the downstream effector TIEG1 through the SIAH pathway. In this manner, turnover of TIEG1 may serve to limit the duration and/or magnitude of TGFβ responses. Transforming growth factor β (TGFβ)-inducible early gene-1 (TIEG1) is a Krüppel-like transcription factor that is rapidly induced upon TGFβ treatment. TIEG1 promotes TGFβ/Smad signaling by down-regulating negative feedback through the inhibitory Smad7. In this report, we describe the identification of an E3 ubiquitin ligase, Seven in Absentia homologue-1 (SIAH1), as a TIEG1-interacting protein. We show that TIEG1 and SIAH1 interact through an amino-terminal domain of TIEG1. Co-expression of SIAH1 results in proteasomal degradation of TIEG1 but not of the related factor TIEG2. Importantly, co-expression of SIAH1 completely reverses repression of Smad7 promoter activity by TIEG1. Furthermore, overexpression of a dominant negative SIAH1 stabilizes TIEG1 and synergizes with TIEG1 to enhance TGFβ/Smad-dependent transcriptional activation. These findings suggest a novel mechanism whereby the ability of TGFβ to modulate gene transcription may be regulated by proteasomal degradation of the downstream effector TIEG1 through the SIAH pathway. In this manner, turnover of TIEG1 may serve to limit the duration and/or magnitude of TGFβ responses. transforming growth factor-β TGFβ-inducible early gene bone morphogenetic protein seven in absentia seven in absentia homologue glutathioneS-transferase dominant negative SIAH1 luciferase major late promoter Transforming growth factor-β (TGFβ)1-inducible early gene-1 (TIEG1) is a member of the Krüppel-like factor family of zinc finger transcription factors and is induced by several members of the TGFβ superfamily including TGFβ and bone morphogenetic protein-2 (BMP-2) (1Subramaniam M. Harris S.A. Oursler M.J. Rasmussen K. Riggs B.L. Spelsberg T.C. Nucleic Acids Res. 1995; 23: 4907-4912Crossref PubMed Scopus (220) Google Scholar, 2Hefferan T.E. Subramaniam M. Khosla S. Riggs B.L. Spelsberg T.C. J. Cell. Biochem. 2000; 78: 380-390Crossref PubMed Scopus (40) Google Scholar). The carboxyl-terminal three zinc finger DNA binding domain allows TIEG1 to directly associate with GC-rich sequences of DNA in target gene promoters (3Yajima S. Lammers C.H. Lee S.H. Hara Y. Mizuno K. Mouradian M.M. J. Neurosci. 1997; 17: 8657-8666Crossref PubMed Google Scholar, 4Tanabe A. Oshima K. Osada S. Nishihara T. Imagawa M. Biol. Pharm. Bull. 2001; 24: 144-150Crossref PubMed Scopus (2) Google Scholar, 5Johnsen S.A. Subramaniam M. Janknecht R. Spelsberg T.C. Oncogene. 2002; 21: 5783-5790Crossref PubMed Scopus (123) Google Scholar). The three zinc fingers are highly conserved between TIEG1 and the related transcription factor TIEG2, suggesting that they may regulate the same sequences in target gene promoters (6Cook T. Gebelein B. Mesa K. Mladek A. Urrutia R. J. Biol. Chem. 1998; 273: 25929-25936Abstract Full Text Full Text PDF PubMed Scopus (160) Google Scholar). However, outside of the DNA binding domain, TIEG1 and TIEG2 bear little sequence homology. Yet both factors are negative regulators of gene transcription and contain an α-helical repression domain (3Yajima S. Lammers C.H. Lee S.H. Hara Y. Mizuno K. Mouradian M.M. J. Neurosci. 1997; 17: 8657-8666Crossref PubMed Google Scholar, 5Johnsen S.A. Subramaniam M. Janknecht R. Spelsberg T.C. Oncogene. 2002; 21: 5783-5790Crossref PubMed Scopus (123) Google Scholar, 7Cook T. Gebelein B. Belal M. Mesa K. Urrutia R. J. Biol. Chem. 1999; 274: 29500-29504Abstract Full Text Full Text PDF PubMed Scopus (104) Google Scholar, 8Zhang J.S. Moncrieffe M.C. Kaczynski J. Ellenrieder V. Prendergast F.G. Urrutia R. Mol. Cell. Biol. 2001; 21: 5041-5049Crossref PubMed Scopus (154) Google Scholar). This domain is capable of repressing gene transcription when fused to a heterologous DNA binding domain and can directly interact with the transcriptional co-repressor mSin3A (7Cook T. Gebelein B. Belal M. Mesa K. Urrutia R. J. Biol. Chem. 1999; 274: 29500-29504Abstract Full Text Full Text PDF PubMed Scopus (104) Google Scholar, 8Zhang J.S. Moncrieffe M.C. Kaczynski J. Ellenrieder V. Prendergast F.G. Urrutia R. Mol. Cell. Biol. 2001; 21: 5041-5049Crossref PubMed Scopus (154) Google Scholar). Overexpression of TIEG1 mimics TGFβ action in several cell types by modulating differentiation markers, decreasing proliferation, and inducing apoptosis (9Hefferan T.E. Reinholz G.G. Rickard D.J. Johnsen S.A. Waters K.M. Subramaniam M. Spelsberg T.C. J. Biol. Chem. 2000; 275: 20255-20259Abstract Full Text Full Text PDF PubMed Scopus (67) Google Scholar, 10Ribeiro A. Bronk S.F. Roberts P.J. Urrutia R. Gores G.J. Hepatology. 1999; 30: 1490-1497Crossref PubMed Scopus (139) Google Scholar, 11Tachibana I. Imoto M. Adjei P.N. Gores G.J. Subramaniam M. Spelsberg T.C. Urrutia R. J. Clin. Invest. 1997; 99: 2365-2374Crossref PubMed Scopus (192) Google Scholar, 12Chalaux E. Lopez-Rovira T. Rosa J.L. Pons G. Boxer L.M. Bartrons R. Ventura F. FEBS Lett. 1999; 457: 478-482Crossref PubMed Scopus (90) Google Scholar). Accumulating in vitro and clinical data in the literature suggests that TIEG1 serves a tumor suppressor role because overexpression reduces proliferation and induces apoptosis (9Hefferan T.E. Reinholz G.G. Rickard D.J. Johnsen S.A. Waters K.M. Subramaniam M. Spelsberg T.C. J. Biol. Chem. 2000; 275: 20255-20259Abstract Full Text Full Text PDF PubMed Scopus (67) Google Scholar, 10Ribeiro A. Bronk S.F. Roberts P.J. Urrutia R. Gores G.J. Hepatology. 1999; 30: 1490-1497Crossref PubMed Scopus (139) Google Scholar, 11Tachibana I. Imoto M. Adjei P.N. Gores G.J. Subramaniam M. Spelsberg T.C. Urrutia R. J. Clin. Invest. 1997; 99: 2365-2374Crossref PubMed Scopus (192) Google Scholar, 12Chalaux E. Lopez-Rovira T. Rosa J.L. Pons G. Boxer L.M. Bartrons R. Ventura F. FEBS Lett. 1999; 457: 478-482Crossref PubMed Scopus (90) Google Scholar). Further support for this role arose from our studies demonstrating a decrease in TIEG1 protein levels that coincides with the increasing stages of cancer in breast tumor biopsies (13Subramaniam M. Hefferan T.E. Tau K. Peus D. Pittelkow M. Jalal S. Riggs B.L. Roche P. Spelsberg T.C. J. Cell. Biochem. 1998; 68: 226-236Crossref PubMed Scopus (76) Google Scholar). TGFβ is a pleiotropic cytokine known to regulate many cellular processes including differentiation, proliferation, apoptosis, and migration whereas BMPs have been shown to play critical roles in development and bone formation. TGFβ or BMP binding to its cognate transmembrane type II receptor initiates formation and activation of a heteromeric complex with the corresponding type I receptor (14Derynck R. Feng X.H. Biochim. Biophys. Acta. 1997; 1333: 105-150Crossref PubMed Scopus (505) Google Scholar). Upon activation, the type I receptor initiates intracellular signaling by phosphorylation of the receptor-regulated Smad (R-Smad) proteins. Smad2 and -3 function in TGFβ signaling, whereas Smad1, -5, and -8 function in BMP signaling (15Miyazono K. Cytokine Growth Factor Rev. 2000; 11: 15-22Crossref PubMed Scopus (226) Google Scholar, 16ten Dijke P. Miyazono K. Heldin C.H. Trends Biochem. Sci. 2000; 25: 64-70Abstract Full Text Full Text PDF PubMed Scopus (336) Google Scholar). Phosphorylated R-Smads complex with the common mediator Smad4, translocate to the nucleus, and modulate target gene transcription. Negative regulation of the Smad pathway occurs through the inhibitory Smads, Smad6 and -7. Smad7 blocks both TGFβ and BMP signaling by binding to the type I receptor and preventing phosphorylation of the R-Smads as well as increasing turnover of the receptor (17Hayashi H. Abdollah S. Qiu Y. Cai J., Xu, Y.Y. Grinnell B.W. Richardson M.A. Topper J.N. Gimbrone Jr., M.A. Wrana J.L. Falb D. Cell. 1997; 89: 1165-1173Abstract Full Text Full Text PDF PubMed Scopus (1145) Google Scholar, 18Imamura T. Takase M. Nishihara A. Oeda E. Hanai J. Kawabata M. Miyazono K. Nature. 1997; 389: 622-626Crossref PubMed Scopus (860) Google Scholar, 19Nakao A. Afrakhte M. Moren A. Nakayama T. Christian J.L. Heuchel R. Itoh S. Kawabata M. Heldin N.E. Heldin C.H. ten Dijke P. Nature. 1997; 389: 631-635Crossref PubMed Scopus (1534) Google Scholar, 20Kavsak P. Rasmussen R.K. Causing C.G. Bonni S. Zhu H. Thomsen G.H. Wrana J.L. Mol. Cell. 2000; 6: 1365-1375Abstract Full Text Full Text PDF PubMed Scopus (1080) Google Scholar, 21Ebisawa T. Fukuchi M. Murakami G. Chiba T. Tanaka K. Imamura T. Miyazono K. J. Biol. Chem. 2001; 276: 12477-12480Abstract Full Text Full Text PDF PubMed Scopus (684) Google Scholar). Smad6 specifically blocks BMP signaling by binding the TGFβ-activated kinase-1 and by binding Smad4, thereby preventing Smad4 heteromerization with Smad1, -5, and -8 (18Imamura T. Takase M. Nishihara A. Oeda E. Hanai J. Kawabata M. Miyazono K. Nature. 1997; 389: 622-626Crossref PubMed Scopus (860) Google Scholar, 22Kimura N. Matsuo R. Shibuya H. Nakashima K. Taga T. J. Biol. Chem. 2000; 275: 17647-17652Abstract Full Text Full Text PDF PubMed Scopus (208) Google Scholar). TIEG1 was recently shown to play a unique role in TGFβ/Smad signaling by down-regulating negative feedback through Smad7. By repressingSmad7 gene transcription, TIEG1 is able to enhance transcription of important TGFβ-regulated genes such as the cyclin-dependent protein kinase inhibitor p21and the plasminogen activator inhibitor-1 (PAI-1) (5Johnsen S.A. Subramaniam M. Janknecht R. Spelsberg T.C. Oncogene. 2002; 21: 5783-5790Crossref PubMed Scopus (123) Google Scholar). We have also observed that TIEG1 is able to repress Smad6promoter activity and may therefore serve an important role in BMP signaling as well. 2S. A. Johnsen, M. Subramaniam, and T. C. Spelsberg, unpublished data.2S. A. Johnsen, M. Subramaniam, and T. C. Spelsberg, unpublished data. Regulation of protein stability through the ubiquitin-proteasome pathway is now being recognized as a major mechanism of regulating a diverse array of cellular processes (23Weissman A.M. Nat. Rev. Mol. Cell. Biol. 2001; 2: 169-178Crossref PubMed Scopus (1247) Google Scholar). Ubiquitination is carried out in a three-step process requiring three different proteins referred to as E1, E2, and E3 (23Weissman A.M. Nat. Rev. Mol. Cell. Biol. 2001; 2: 169-178Crossref PubMed Scopus (1247) Google Scholar). The ubiquitin-activating enzyme (E1) undergoes a covalent intermediate in which the carboxyl terminus of ubiquitin forms a high energy thioester bond in an ATP-dependent process with an active cysteine residue in the E1 molecule. Another thioester intermediate is formed when ubiquitin is transferred from E1 to the E2 protein, also referred to as a ubiquitin-conjugating enzyme. Specificity of substrate recognition is provided by the E3 ubiquitin ligase, which facilitates the transfer of ubiquitin to the target protein (23Weissman A.M. Nat. Rev. Mol. Cell. Biol. 2001; 2: 169-178Crossref PubMed Scopus (1247) Google Scholar). The Seven in Absentia (Sina) protein was identified in a screen for mutants defective in the development of the R7 photoreceptor cell inDrosophila melanogaster (24Carthew R.W. Rubin G.M. Cell. 1990; 63: 561-577Abstract Full Text PDF PubMed Scopus (259) Google Scholar). It was subsequently shown to function in signaling downstream of the receptor tyrosine kinase Sevenless by targeting a transcriptional repressor, Tramtrack, for degradation by the ubiquitin-proteasome pathway (25Tang A.H. Neufeld T.P. Kwan E. Rubin G.M. Cell. 1997; 90: 459-467Abstract Full Text Full Text PDF PubMed Scopus (199) Google Scholar). Degradation of Tramtrack will then result in derepression of target genes such as the pair-rule genes even-skipped and fushi tarazu(26Read D. Levine M. Manley J.L. Mech. Dev. 1992; 38: 183-195Crossref PubMed Scopus (45) Google Scholar). The human homologues of Sina, SIAH1 and SIAH2, are able to target both nuclear and cytoplasmic proteins for proteasomal degradation (27Hu G. Zhang S. Vidal M. Baer J.L., Xu, T. Fearon E.R. Genes Dev. 1997; 11: 2701-2714Crossref PubMed Scopus (177) Google Scholar, 28Germani A. Bruzzoni-Giovanelli H. Fellous A. Gisselbrecht S. Varin-Blank N. Calvo F. Oncogene. 2000; 19: 5997-6006Crossref PubMed Scopus (77) Google Scholar, 29Boehm J., He, Y. Greiner A. Staudt L. Wirth T. EMBO J. 2001; 20: 4153-4162Crossref PubMed Scopus (78) Google Scholar, 30Tiedt R. Bartholdy B.A. Matthias G. Newell J.W. Matthias P. EMBO J. 2001; 20: 4143-4152Crossref PubMed Scopus (70) Google Scholar, 31Susini L. Passer B.J. Amzallag-Elbaz N. Juven-Gershon T. Prieur S. Privat N. Tuynder M. Gendron M.C. Israel A. Amson R. Oren M. Telerman A. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 15067-15072Crossref PubMed Scopus (84) Google Scholar, 32Wheeler T.C. Chin L.S., Li, Y. Roudabush F.L. Li L. J. Biol. Chem. 2002; 277: 10273-10282Abstract Full Text Full Text PDF PubMed Scopus (146) Google Scholar, 33Zhang J. Guenther M.G. Carthew R.W. Lazar M.A. Genes Dev. 1998; 12: 1775-1780Crossref PubMed Scopus (190) Google Scholar). We demonstrate that the stability of TIEG1 protein is regulated by interaction with the human homologue of Seven in Absentia, SIAH1. Thus we propose a pathway in mammalian cells that closely parallels theDrosophila pathway in which degradation of a transcriptional repressor (Tramtrack or TIEG1) by a ubiquitin ligase (Sina or SIAH) results in derepression of genes encoding important cellular proteins (even-skipped or Smad7). This novel interaction suggests a role for the regulation of TIEG1 protein stability in modulating the activity of the TGFβ signal transduction pathway. The bait plasmid was constructed by cloning the region containing amino acids 1–350 of TIEG1 (1Subramaniam M. Harris S.A. Oursler M.J. Rasmussen K. Riggs B.L. Spelsberg T.C. Nucleic Acids Res. 1995; 23: 4907-4912Crossref PubMed Scopus (220) Google Scholar) into pAS2–1 (CLONTECH, Palo Alto, CA) by PCR. This plasmid was transformed into the MATa yeast strain AH109 using the Yeastmaker Transformation System (CLONTECH). The yeast two-hybrid screen was performed using a pretransformed skeletal muscle library (CLONTECH) and interacting clones identified by the ability to grow on minimal SD agar medium lacking tryptophan, leucine, histidine, and adenine (CLONTECH). Prey plasmids were recovered and retransformed into AH109 with the bait or the GAL4 DNA binding domain alone to verify interactions by β-galactosidase filter lift assays. The identity of interacting clones was determined by DNA sequencing. The amino-terminal FLAG epitope-tagged TIEG1 and TIEG2 expression vectors were constructed by amplifying the respective open reading frames by PCR with primers that addEcoRI and BamHI or EcoRI and XbaI restriction sites at the 5′- and 3′-ends, respectively. The resulting fragment was cloned into pcDNA4/TO (Invitrogen) behind a FLAG epitope tag. The Renilla luciferase plasmid (pRL-CMV) was purchased from Promega (Madison, WI). The SIAH1 expression vector was constructed by cloning the entireSIAH1 coding region into pcDNA4/TO. FLAG-tagged dominant negative SIAH1 (dn-SIAH1), which contains a cysteine to serine mutation at residue 72, was constructed by PCR and cloned into pcDNA4/TO. Sequence integrity was verified by DNA sequencing. GST fusion proteins were generated by cloning portions of TIEG1 into pGEX-2T-6His-PL2. Fusion proteins were purified as previously described (34Janknecht R. Nordheim A. Oncogene. 1996; 12: 1961-1969PubMed Google Scholar). Recombinant35S-labeled SIAH1 was produced from pGBK-T7-SIAH1 using thetntin vitro transcription and translation system (Promega). GST fusion proteins were immobilized on glutathione-Sepharose 4B (Amersham Biosciences) and incubated with35S-labeled SIAH1 at 4 °C for 2 h in 50 mm NaH2PO4 (pH 7.2), 100 mm NaCl, and 0.05% (v/v) Triton X-100 with gentle mixing. The mixtures were then centrifuged, and pellets were washed extensively in the same buffer. Bound proteins were eluted by boiling in Laemmli buffer and separated by SDS-PAGE using precast 10% (w/v) Tris-HCl polyacrylamide gels (Bio-Rad). Radiolabeled SIAH1 was visualized using a Storm 840 PhosphorImager (Amersham Biosciences). C2C12 mouse myoblast cells were obtained from ATCC (Manassas, VA) and grown in Dulbecco's modified Eagle's medium/F12 (1:1) (Sigma) containing 10% (v/v) fetal bovine serum (Bio Whittaker, Walkersville, MD) and 1× antibiotic-antimycotic solution (Invitrogen). Cells were seeded in 12-well or 4-cm plates and transfected at 50% confluence with various combinations of plasmid DNA using LipofectAMINE Plus (Invitrogen) according to the manufacturer's directions. All transfections included a plasmid containing Renilla luciferase driven from the cytomegalovirus promoter (pRL-CMV) as an internal transfection control. Cell extracts were harvested in radioimmune precipitation assay buffer (phosphate-buffered saline, 1% (w/v) Nonidet P-40, 0.5% (w/v) sodium deoxycholate, 0.1% (w/v) SDS) containing 100 μg/ml phenylmethylsulfonyl fluoride, 2 μg/ml aprotinin, 10 μg/ml leupeptin, and 500 μm sodium orthovanadate. Protein was separated on a SDS-10% (w/v) polyacrylamide gel and blotted onto Protran nitrocellulose membranes (Schleicher & Schuell). TIEG1 protein was detected with a rabbit polyclonal antibody, TIEG-228, raised to a synthetic peptide corresponding to amino acids 133–154 of the human TIEG1 protein. Actin and FLAG epitope-tagged proteins were detected with anti-actin AC-40 and anti-FLAG M2 antibodies, respectively (Sigma). Primary antibodies were detected by enhanced chemiluminescence (Amersham Biosciences) using horseradish peroxidase-conjugated anti-rabbit or anti-mouse secondary antibodies (Sigma). Cell extracts were harvested 48 h after transfection using 300 μl of passive lysis buffer. Luciferase assays were performed using the dual luciferase reporter assay system (Promega), and samples were read on a Turner TD-20/20 luminometer. To correct for differences in transfection efficiency, firefly luciferase units were normalized relative toRenilla luciferase units of the same sample. Corrected luciferase values were then expressed as a ratio (-fold induction) relative to the vector control transfected cells. To identify proteins that interact with TIEG1, we utilized the yeast two-hybrid system. Because TIEG1 is a member of the Krüppel-like factor family of three zinc finger transcription factors, which bear a high degree of homology to one another in the carboxyl-terminal DNA binding domain, we excluded this domain by utilizing a bait construct containing only the first 350 amino acids of TIEG1 (Fig.1A) to screen a skeletal muscle library. This region contains several potential SH3 binding domains as well as three distinct repression domains (1Subramaniam M. Harris S.A. Oursler M.J. Rasmussen K. Riggs B.L. Spelsberg T.C. Nucleic Acids Res. 1995; 23: 4907-4912Crossref PubMed Scopus (220) Google Scholar, 7Cook T. Gebelein B. Belal M. Mesa K. Urrutia R. J. Biol. Chem. 1999; 274: 29500-29504Abstract Full Text Full Text PDF PubMed Scopus (104) Google Scholar). Four positive clones, which grew on selective media and were positive for β-galactosidase expression, were identified from 1.6 × 106 clones that were screened. All four clones contained inserts identified as the human homologues of the D. melanogaster seven in absentia gene. Three of the clones contained seven in absentia homologue (SIAH)-2 whereas the other contained SIAH1. SIAH1 and SIAH2 share 87% amino acid identity to one another and 76 and 68% to Sina, respectively (35Hu G. Chung Y.L. Glover T. Valentine V. Look A.T. Fearon E.R. Genomics. 1997; 46: 103-111Crossref PubMed Scopus (124) Google Scholar). Because these two proteins appear to be functionally redundant and similarly recognize substrate proteins (27Hu G. Zhang S. Vidal M. Baer J.L., Xu, T. Fearon E.R. Genes Dev. 1997; 11: 2701-2714Crossref PubMed Scopus (177) Google Scholar) and because SIAH1 has been more thoroughly characterized, we focused primarily upon the interaction between TIEG1 and SIAH1. As shown by growth on selective medium, both full-length TIEG1 as well as TIEG11–350specifically interact with SIAH1, whereas TIEG1 did not do so with a control protein, the estrogen receptor-α (Fig. 1B). Furthermore, as previously reported, SIAH1 also interacts with itself (25Tang A.H. Neufeld T.P. Kwan E. Rubin G.M. Cell. 1997; 90: 459-467Abstract Full Text Full Text PDF PubMed Scopus (199) Google Scholar, 36Hu G. Fearon E.R. Mol. Cell. Biol. 1999; 19: 724-732Crossref PubMed Scopus (195) Google Scholar). As shown in Fig. 1C, using in vitro binding assays TIEG1 appears to interact with SIAH1 through the amino-terminal 210 amino acids because all three constructs containing this region (TIEG11–480, TIEG11–350, and TIEG11–210) were able to pull SIAH1 down. However, a stronger interaction between TIEG1 and SIAH1 is observed when the entire amino-terminal domain remains intact (TIEG11–350; Fig. 1C). This suggests that some aspect of the TIEG1 tertiary structure is necessary for optimal interaction between TIEG1 and SIAH1 or that amino acids 211–350 augment binding. Based on previous data showing that both Sina and SIAH1 function as E3 ubiquitin ligases and target certain proteins for proteasomal degradation, we hypothesized that the interaction with SIAH1 may increase turnover of TIEG1 protein levels. Indeed, when a SIAH1 expression vector was co-transfected into C2C12 cells with a TIEG1 expression vector, a dramatic decrease in TIEG1 protein levels was observed. In fact, only small amounts of SIAH1 expression vector (75 ng) were necessary to decrease TIEG1 protein to undetectable levels, and even a very small amount (5 ng) had a visible effect (Fig.2A). As a control, actin protein levels were unaffected. Thus the effect of SIAH1 expression on TIEG1 protein levels is considerably more dramatic than that observed for other SIAH1 target proteins such as β-catenin, deleted in colorectal cancer (DCC), and synaptophysin, in which there is only a partial decrease in protein levels even with high amounts of SIAH1 expression (27Hu G. Zhang S. Vidal M. Baer J.L., Xu, T. Fearon E.R. Genes Dev. 1997; 11: 2701-2714Crossref PubMed Scopus (177) Google Scholar, 32Wheeler T.C. Chin L.S., Li, Y. Roudabush F.L. Li L. J. Biol. Chem. 2002; 277: 10273-10282Abstract Full Text Full Text PDF PubMed Scopus (146) Google Scholar, 37Liu J. Stevens J. Rote C.A. Yost H.J., Hu, Y. Neufeld K.L. White R.L. Matsunami N. Mol. Cell. 2001; 7: 927-936Abstract Full Text Full Text PDF PubMed Scopus (373) Google Scholar). Conversely, expression of a dominant negative SIAH1 (dn-SIAH1), in which the RING finger structure required for E3 ligase activity has been obstructed, appears to stabilize TIEG1 protein levels (Fig. 2A). In addition, the proteasome inhibitor MG132 completely blocks the decrease in TIEG1 protein levels upon SIAH1 co-expression, suggesting that the decrease in TIEG1 protein levels is proteasome-dependent (Fig. 2B). Furthermore, degradation of TIEG1 by SIAH1 is specific because SIAH1 has no effect on TIEG2 protein levels (Fig. 2B). Regulation of gene expression through targeted degradation of important transcription factors is a common theme in transcriptional control. For example, the protein levels and activities of p53, β-catenin, and NFκB are all regulated by targeted degradation of specific proteins through the ubiquitin-proteasome pathway (38Kubbutat M.H. Jones S.N. Vousden K.H. Nature. 1997; 387: 299-303Crossref PubMed Scopus (2798) Google Scholar, 39Haupt Y. Maya R. Kazaz A. Oren M. Nature. 1997; 387: 296-299Crossref PubMed Scopus (3629) Google Scholar, 40Aberle H. Bauer A. Stappert J. Kispert A. Kemler R. EMBO J. 1997; 16: 3797-3804Crossref PubMed Scopus (2122) Google Scholar, 41Palombella V.J. Rando O.J. Goldberg A.L. Maniatis T. Cell. 1994; 78: 773-785Abstract Full Text PDF PubMed Scopus (1908) Google Scholar). Therefore, based on the observation that SIAH1 targets TIEG1 for proteasomal degradation, we reasoned that SIAH1 may also influence the regulation of TIEG1 target genes. Simply stated, increased SIAH1 activity might decrease TIEG levels and reverse any TIEG protein effects on gene expression, whereas inhibition of SIAH1 activity may enhance these effects. Although we recently demonstrated that TIEG1 represses Smad7 expression by binding to a GC-rich sequence in the proximal Smad7 promoter (5Johnsen S.A. Subramaniam M. Janknecht R. Spelsberg T.C. Oncogene. 2002; 21: 5783-5790Crossref PubMed Scopus (123) Google Scholar), we now find that overexpression of SIAH1 by itself has very little effect onSmad7 promoter activity (Fig.3A). This observation may be explained by the fact that, in the absence of growth factor treatment, TIEG1 is expressed at levels too low for significant biological activity (1Subramaniam M. Harris S.A. Oursler M.J. Rasmussen K. Riggs B.L. Spelsberg T.C. Nucleic Acids Res. 1995; 23: 4907-4912Crossref PubMed Scopus (220) Google Scholar, 2Hefferan T.E. Subramaniam M. Khosla S. Riggs B.L. Spelsberg T.C. J. Cell. Biochem. 2000; 78: 380-390Crossref PubMed Scopus (40) Google Scholar, 9Hefferan T.E. Reinholz G.G. Rickard D.J. Johnsen S.A. Waters K.M. Subramaniam M. Spelsberg T.C. J. Biol. Chem. 2000; 275: 20255-20259Abstract Full Text Full Text PDF PubMed Scopus (67) Google Scholar). Thus a subsequent reduction of TIEG protein levels by SIAH1 would be predicted to have no observable effect on the regulation of TIEG target genes, whereas an increase in endogenous TIEG protein levels, either through the inhibition of SIAH activity or overexpression of TIEG, should have a noticeable effect. Indeed, overexpression of dn-SIAH1 or TIEG1 alone decreases Smad7promoter activity (Fig. 3A). The effect of SIAH1 is more evident when TIEG1 is co-expressed. As shown in Fig. 3B, TIEG1 dramatically repressesSmad7 promoter activity. However, co-expression of SIAH1 restores normal promoter activity. The reversal of Smad7repression by SIAH1 is also specific for TIEG1 because TIEG2 also represses Smad7 promoter activity (Fig. 3B), but SIAH1 does not abrogate this repression. The ubiquitin-proteasome pathway plays an important role in regulating TGFβ signal transduction through the Smad proteins. Most notably, activation of the TGFβ/Smad pathway stimulates an interaction between Smad2 and the Smad ubiquitination regulatory factor-2 (Smurf-2), an E3 ubiquitin ligase, and results in the ubiquitination and subsequent degradation of Smad2 (42Lin X. Liang M. Feng X.H. J. Biol. Chem. 2000; 275: 36818-36822Abstract Full Text Full Text PDF PubMed Scopus (398) Google Scholar, 43Zhang Y. Chang C. Gehling D.J. Hemmati-Brivanlou A. Derynck R. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 974-979Crossref PubMed Scopus (419) Google Scholar). Similarly, Smad3 is degraded through interaction with another ubiquitin ligase complex, the anaphase promoting complex, following ligand stimulation (44Fukuchi M. Imamura T. Chiba T. Ebisawa T. Kawabata M. Tanaka K. Miyazono K. Mol. Biol. Cell. 2001; 12: 1431-1443Crossref PubMed Scopus (176) Google Scholar). In cancer cells, expression of oncogenic Ras or the presence of specific destabilizing mutations promotes ubiquitin-proteasome-dependent turnover of Smad4, rendering cells insensitive to the effects of TGFβ (45Saha D. Datta P.K. Beauchamp R.D. J. Biol. Chem. 2001; 276: 29531-29537Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar, 46Xu J. Attisano L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 4820-4825Crossref PubMed Scopus (170) Google Scholar, 47Maurice D. Pierreux C.E. Howell M. Wilentz R.E. Owen M.J. Hill C.S. J. Biol. Chem. 2001; 276: 43175-43181Abstract Full Text Full Text PDF PubMed Scopus (79) Google Scholar). Thus the ubiquitin-proteasome pathway plays an important role in regulating the TGFβ/Smad pathway. Because repression of Smad7 promoter activity by TIEG1 enhances TGFβ/Smad signaling by relieving negative feedback imposed by Smad7, it is likely that modulation of TIEG1 protein levels would also have an effect on the entire Smad pathway. Therefore we hypothesized that SIAH1 or dn-SIAH1 overexpression may influence Smad signaling by regulating TIEG1 stability. This was tested using a TGFβ-responsive reporter construct (CAGA12-MLP-Luc) and a constitutively active type I TGFβ receptor (ALK5TD) along with co-expression of TIEG1 and SIAH1 or dn-SIAH1. As shown in Fig. 4 and similar to the effects observed with the Smad7 promoter, expression of wild-type SIAH1 has virtually no effect on CAGA12-MLP-Luc reporter induction by itself and slightly reverses the enhancement observed with TIEG1 co-expression. However, dn-SIAH1 overexpression enhances TGFβ/Smad signaling to a similar degree as TIEG1 overexpression by increasing ALK5TD- induced transcription ∼2-fold compared with vector transfected cells. Interestingly, co-expression of dn-SIAH1 and TIEG1 appears to synergistically enhance CAGA12-MLP induction to about 5.5-fold compared with vector transfected cells. This effect is probably because of stabilization of TIEG1 protein. These data suggest that activation of the TGFβ/Smad pathway may be regulated through the ubiquitin-proteasome pathway at an additional level by the regulation of the stability of the TGFβ-inducible early protein through SIAH1. By controlling TIEG1 protein levels a cell has an additional mechanism to control the amplitude or duration of TGFβ signaling. This also provides a mechanism that cancer cells may employ to evade cell cycle regulation by TGFβ. Mutations in the TGFβ pathway have been observed in certain types of cancer including pancreatic and colorectal cancers (48Akhurst R.J. Derynck R. Trends Cell Biol. 2001; 11 (suppl.): 44-51Google Scholar). Most of these mutations result in a complete blockage of TGFβ signaling through the Smad pathway and are believed to relieve the cancer cells from the growth inhibitory effects of TGFβ. However, in many cases there appears to be subtle perturbations in the activity of TGFβ signaling that are not because of mutations in the known signaling components (48Akhurst R.J. Derynck R. Trends Cell Biol. 2001; 11 (suppl.): 44-51Google Scholar). The effects of these subtle changes often increase the aggressiveness and/or metastatic potential of a tumor. Interestingly a decrease in TIEG1 protein levels correlates with the histological stage of breast cancer (13Subramaniam M. Hefferan T.E. Tau K. Peus D. Pittelkow M. Jalal S. Riggs B.L. Roche P. Spelsberg T.C. J. Cell. Biochem. 1998; 68: 226-236Crossref PubMed Scopus (76) Google Scholar). However, it is currently unknown whether the mRNA levels also correlate. It is possible that the decrease in TIEG1 protein levels may be because of post-translational regulation of TIEG1 protein levels, possibly by the SIAH proteins or another, as yet, unidentified ubiquitin ligase. Investigations into the expression of the SIAH proteins and the regulation of the SIAH genes may provide insights into the regulation of the TGFβ/Smad signaling pathway and may hint at a new area of exploration in cancer research. We thank the following for graciously providing plasmids: Drs. M. Kato and K. Miyazono (ALK5TD expression construct), Y. Chen (Smad7 promoter), C. Szpirer (TIEG2/KLF11 cDNA), and E. Fearon (SIAH expression constructs). We also thank K. Rasmussen and T. Ruesink for outstanding technical support and Drs. A. Tang and D. Rickard for critically reviewing the manuscript.