Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. 2006 Dec 5; 103(49): 18580–18585.
Published online 2006 Nov 28. doi:  10.1073/pnas.0604773103
PMCID: PMC1693705
Cell Biology

Smad4 cooperates with lymphoid enhancer-binding factor 1/T cell-specific factor to increase c-myc expression in the absence of TGF-β signaling


The c-myc protooncogene is a key regulator of cell proliferation whose expression is reduced in normal epithelial cells in response to the growth inhibitory cytokine TGF-β. Smad4 mediates this inhibitory effect of TGF-β by forming a complex with Smad3, E2F4/5, and p107 at the TGF-β inhibitory element (TIE) element on the c-myc promoter. In contrast, cell proliferation and c-myc expression are increased in response to Wnt ligands; this effect is mediated through the lymphoid enhancer-binding factor 1/T cell-specific factor (LEF/TCF) family of transcription factors on the c-myc promoter LEF/TCF-binding elements (TBE1 and TBE2). We report that a peptide aptamer designed to inhibit the binding between Smad4 and LEF/TCF reduced c-myc expression and the growth rate of HepG2 cells. Further analysis demonstrated that, in the absence of TGF-β, Smad4 was bound to the positive regulatory element TBE1 from the c-myc promoter and activated c-myc promoter activity. Smad4 binding to the positive TBE1 c-myc element was reduced by TGF-β, consistent with Smad4's inhibitory role on c-myc expression in response to TGF-β. Reduction of Smad4 levels by RNAi knockdown also reduced c-myc expression levels and sensitized hepatocytes to cell death by serum deprivation. Two tumor-derived mutant Smad4 proteins that fail to mediate TGF-β responses were still competent to cooperate with LEF1 to activate the c-myc promoter. These results support a previously unreported TGF-β-independent function for Smad4 in cooperating with LEF/TCF to activate c-myc expression.

Keywords: Lef 1, T cell factor, cell death, peptide aptamer

C-myc is a protooncogene regulating a diverse group of genes involved in cell growth, apoptosis, metabolism, and differentiation (1, 2). The deregulation of this gene is one of the most common abnormalities found in human cancers, including melanoma, leukemia, breast carcinoma, and gastrointestinal carcinoma (3). The importance of c-myc in carcinogenesis is also supported by animal studies; for example, overexpression of MYC in mouse liver induces hepatocellular carcinoma, which regresses if MYC expression is inactivated (4). In cancer cells, the c-myc gene is activated through several mechanisms, including chromosomal translocation, gene amplification, and increased mRNA stability (1).

Genetic and epigenetic alterations in the Wnt/β-catenin and the TGF-β/Smad pathway also up-regulate c-myc expression in tumors (5, 6). A β-catenin-LEF/TCF complex binds to the c-myc promoter and activates its transcription in human colon cancer cells. Two binding sites for LEF/TCF transcription factors were identified in the myc promoter and named LEF/TCF-binding elements (TBE; TBE1 and TBE2; ref. 6). Smads are involved in regulating c-myc promoter activity in a complex with E2F4/5 and p107 (5). Upon TGF-β signaling, this complex is recruited to the TGF-β-inhibitory element (TIE), another region of the c-myc promoter, and represses c-myc transcription.

Smad4 is essential for the antiproliferative effect of TGF-β signaling. It forms a heterocomplex with receptor-regulated Smads (R-Smads), such as Smad2 and Smad3, and this complex regulates transcription of target genes in a complex with other transcription factors and cofactors (7). Smad4 has been studied intensively in the context of TGF-β signaling, but its function in the absence of the signal is not well defined. Although ≈50% of pancreatic carcinomas and some colorectal carcinomas harbor homozygous deletions or inactivating mutations in the Smad4 gene, Smad4 mutation is rare in other types of cancers (8, 9), suggesting that many cancer cells express wild-type Smad4 protein.

Peptide aptamers are “proteins that contain a conformationally constrained peptide region of variable sequence displayed from a scaffold” (10). We have generated small focused libraries of Smad-binding domains from various Smad-interacting proteins displayed on a thioredoxin A scaffold (Trx). Several of these peptide aptamers bind to Smad proteins and inhibit TGF-β/Smad signaling selectively (11, 12). Among them, TrxLef1D, generated by inserting the Smad-binding domain from LEF1 into the Trx scaffold, was able to bind Smad1, -2, -3, -4, and -7, as does full length LEF1 (1317). TrxLef1D specifically inhibited a Smad-LEF/TCF complex-dependent reporter, Twntop-lux, without interfering with TGF-β activation of other Smad-dependent reporters (11).

In characterizing the effects of the TrxLef1D aptamer on cellular responses to TGF-β, we noted that expression of TrxLef1D slowed the growth of cells regardless of whether TGF-β was present. In trying to explain this effect, we found that expression of TrxLef1D reduced the level of c-myc expression in the cells. Examination of the c-myc promoter sequences previously defined as important to c-myc regulation revealed that Smad4 could bind and activate transcription through the TBE1 element. Addition of TGF-β to the cells, which represses c-myc expression, reduced Smad4 binding to the positive regulatory element TBE1. Expression of both Smad4 and LEF1 were needed to activate transcription optimally from the TBE1 element in the c-myc promoter. Interestingly, mutant Smad4 proteins that do not support a TGF-β response, because they fail to form the active trimeric protein complex with phosphorylated R-Smads, were still able to activate transcription together with LEF1 from the TBE1. We conclude that Smad4 has a TGF-β-independent function in positively regulating c-myc expression, a function that is maintained even by some Smad4 mutant proteins previously thought to be inactive.


Peptide Aptamer TrxLef1D Inhibits Proliferation of HepG2 Cells.

Previously we developed a peptide aptamer, TrxLef1D that binds to Smad proteins, disrupts the interaction between Smad4 and LEF1, and inhibits TGF-β-induced, Smad-mediated activation of transcription specifically at a reporter gene that requires a Smad-LEF/TCF complex (ref. 11 and Fig. 6a, which is published as supporting information on the PNAS web site). Because Smad is an essential mediator of the TGF-β signaling pathway, we decided to test whether TrxLef1D was able to alter the antiproliferative effect of TGF-β signaling. HepG2 cells from human hepatocellular carcinoma express endogenous Smads and LEF/TCF proteins, such as TCF1 and TCF4 (18). HepG2 cells stably expressing TrxLef1D were generated by retroviral infection and sorted for high level expression of GFP to verify the infection (H2-TrxLef1D cells). In addition, HepG2 cells expressing TrxGA, which has a tandem repeat of gly-ala in the Trx scaffold and has been used as a negative control aptamer in previous assays (11, 12), were also prepared by the same procedure (H2-TrxGA cells). When each cell type was cultured in the absence or presence of TGF-β, the proliferation of both H2-TrxLef1D and H2-TrxGA cells was inhibited by TGF-β treatment (Fig. 1). The expression of TrxLef1D did not interfere with the antiproliferative effect of TGF-β/Smad signaling. However, H2-TrxLef1D cells grew more slowly than H2-TrxGA cells regardless of whether the cells were treated with TGF-β.

Fig. 1.
Peptide aptamer TrxLef1D inhibits the proliferation of HepG2 cells, irrespective of TGF-β treatment. H2-TrxGA and H2-TrxLef1D cells were seeded in a 96-well plate and the next day treated by 0 or 30 pM TGF-β. Viable cell number was monitored ...

To confirm this observation and also exclude the possibility that HepG2 cell proliferation might be accelerated by Trx-GA rather than decelerated by TrxLef1D, each cell type was mixed 50:50 with parental HepG2 cells and cocultured. The aptamer-expressing GFP-positive cell ratio was monitored to determine whether the cell proliferation rate was different between parental HepG2 cells and the H2-TrxGA or the H2-TrxLef1D cells. Over a 3-day time course, the GFP-positive cell ratio gradually decreased in the coculture of H2-TrxLef1D and HepG2 cells, whereas it was not changed significantly in the H2-TrxGA and HepG2 cell mixture (Fig. 6b). We conclude that TrxLef1D-expressing cells have reduced proliferation compared with the parental HepG2 cells.

TrxLef1D Inhibits c-myc Transcription Independent of TGF-β Signaling.

Smad proteins are able to regulate cell proliferation by repressing c-myc and inducing cyclin-dependent kinase inhibitors such as p15Ink4b and p21Cip1/Waf1 through direct binding to the promoters of each gene (5, 19, 20). c-myc was a particularly interesting candidate gene that might be affected by TrxLef1D, because both Smads and LEF/TCF directly bind to the c-myc promoter (5, 6), and a known target of TrxLef1D is a Smad-LEF/TCF complex (11). To test whether c-myc expression might be changed by TrxLef1D expression, a reporter gene driven by the c-myc promoter was cotransfected in HepG2 cells with plasmids expressing either TrxLef1D or TrxGA. Another peptide aptamer, TrxSARA, which specifically binds Smad2 and Smad3 and inhibits TGF-β/Smad signaling by reducing the level of active Smad2/3–4 complex (12), was transfected to examine the involvement of TGF-β signaling. Expression of TrxLef1D inhibited c-myc promoter activity when compared with the TrxGA control (Fig. 2a). TrxSARA did not alter expression from the c-myc promoter-driven reporter gene, indicating that the Smad2/3-Smad4 complex was not regulating the c-myc promoter in this assay, which was done in the absence of TGF-β signaling. The competence of TrxSARA in HepG2 cells was confirmed by using the TGF-β-responsive reporter gene, SBE12 (Fig. 7a, which is published as supporting information on the PNAS web site). We conclude that TrxLef1D has an inhibitory effect on the c-myc reporter gene that is independent of the Smad2/3-Smad4 complex.

Fig. 2.
TrxLef1D inhibits c-myc transcription independent of TGF-β signaling. (a) TrxLef1D, but not TrxSARA, inhibits reporter gene activity driven by the c-myc promoter in HepG2 cells. Twenty-four hours after transfection, cells were lysed for luciferase ...

The TGF-β independence of the TrxLef1D inhibition was tested by using the mouse mammary gland cancer cell line, 4T1. In accordance with the previous report that 4T1 cells are refractory to the antiproliferative effect of TGF-β (21, 22), treatment with either TGF-β or the TGF-β type I receptor kinase inhibitor SB431542 did not affect expression from the c-myc reporter gene in 4T1 cells (Fig. 2b). However, TrxLef1D inhibited c-myc promoter activity and reduced the level of endogenous c-myc transcript in 4T1 cells (Fig. 2 b and c). As in HepG2 cells, cotransfection of TrxSARA did not affect c-myc reporter gene expression, whereas it inhibited a TGF-β-responsive reporter, SBE12, in 4T1 cells (Fig. 7 b and c). Therefore, we conclude that TrxLef1D inhibits c-myc promoter activity independent of TGF-β signaling, which is consistent with our observation that TrxLef1D inhibits cell proliferation independent of TGF-β signaling.

Smad4 Binds to the TBE1 of the c-myc Promoter in the Absence of TGF-β Signaling.

TGF-β signaling induces a Smad-E2F4/5-p107 complex to bind to a specific region of the c-myc promoter, called the TIE, and repress c-myc transcription (5). TrxLef1D binds to Smad proteins and inhibits Smad-dependent transcription at the Twntop luciferase reporter gene (11), but it seemed unlikely that TrxLef1D was acting on the Smad complex at the TIE of the c-myc promoter, especially because its effects on c-myc were observed in the absence of TGF-β signaling. This led us to hypothesize that a Smad protein might positively regulate c-myc transcription in a complex with LEF/TCF in the absence of TGF-β signaling. He et al. (6) reported that there are two LEF/TCF-binding elements on the human c-myc promoter, named TBE1 and TBE2. Examination of the c-myc promoter sequence revealed a Smad-binding element (GTCT) adjacent to TBE1 (Fig. 3a). To address whether any Smad proteins bind to this region, a DNA pull-down assay was performed with oligonucleotides comprising either TBE1 or TBE2. Smad4 associated with the TBE1 probe (Fig. 3b). Although R-Smads such as Smad1 and -3 have the same DNA-binding β-hairpin motif as Smad4, neither of them was detected in TBE1 oligonucleotide pull downs (Fig. 3 b and c). Smad4 was not detected with the TBE2 probe containing the other known LEF/TCF-binding element but no Smad-binding element (Fig. 3b). Mutation of the Smad-binding element in the TBE1 probe abolished the Smad4 binding to TBE1 probe and reduced the expression sensitive to TrxLef1D inhibition from the c-myc reporter gene (Fig. 8 a and b, which is published as supporting information on the PNAS web site).

Fig. 3.
Smad4 binds to the TBE1 of the c-myc promoter in the absence of TGF-β signaling (a) Diagram of human c-myc promoter (−1,239 to +1,221). Smad- (GTCT) and LEF/TCF-binding sequence (CCTTTGATT) are in capital letters. (b) HepG2 cell lysates ...

Because TrxLef1D inhibited c-myc expression in the absence of TGF-β signaling, we tested whether Smad4 binding to the TBE1 region was independent of TGF-β signaling. HepG2 cells were treated with SB431542 to block residual TGF-β signaling from the serum or endogenous expression. Even under these conditions, Smad4 was bound to TBE1, and in fact the amount of Smad4 bound to TBE1 probe was slightly increased (Fig. 3d). Addition of exogenous TGF-β significantly reduced Smad4 binding to TBE1 (Fig. 3d).

ChIP assays were used to determine whether Smad4 binds to the TBE1 in cells. HepG2 cells were pretreated with SB431542 or TGF-β, and chromatin was immunoprecipitated with anti-Smad4 antibody. As expected, Smad4 binding to the PAI-1 promoter required TGF-β signaling (23). In contrast, the TBE1 sequence was detected in the chromatin immunoprecipitated by Smad4 from both SB431542- and TGF-β-treated cells (Fig. 3e). Furthermore, Smad4 immunoprecipitated less TBE1 DNA from cells after TGF-β treatment. This ChIP result was consistent with the in vitro pull-down assay result in which lysates from TGF-β-treated cells had reduced Smad4 binding to the TBE1 oligonucleotide. Although TBE1 is 1.1 kb from the Smad-binding TIE, the detection of the TBE1 sequences in the Smad4 immunoprecipitate was not due to Smad4 binding to the TIE, because Smad4 binding to the TIE was not detected in SB431542-treated cells (Fig. 8c).

Smad4 Is Able to Activate the c-myc Promoter in the Absence of TGF-β Signaling.

To test whether Smad4 might positively regulate c-myc transcription by binding to the TBE1 element, in direct contrast to its negative regulator role on the TIE, the Smad- and E2F-binding elements of the TIE in the c-myc promoter were mutated so that Smad4 could not bind to the TIE (TIEm myc reporter; ref. 5). When this reporter was cotransfected with Smad4 in HepG2 cells, the promoter activity was slightly increased (Fig. 4a). In MDA-MB-468 cells, where Smad4 is null because of a homozygous deletion, expression of either exogenous Smad4 or exogenous LEF1 was not sufficient to activate the reporter gene, but when Smad4 and LEF1 were coexpressed, the c-myc reporter gene was activated (Fig. 4b). We conclude that Smad4 and LEF1 activate the c-myc promoter.

Fig. 4.
Smad4 activates the c-myc promoter in the absence of TGF-β signaling. (a) Smad4 overexpression activates TIEm myc reporter in HepG2 cells. TIEm myc reporter was cotransfected by Smad4 expression vector or control empty vector. Twenty-four hours ...

To confirm this new role of Smad4, a short hairpin RNAi to Smad4 was expressed in HepG2 cells, and the level of c-myc transcript was detected by RT-PCR. HepG2 cells infected with a retrovirus encoding shRNA against Smad4 had a reduced level of Smad4 protein and concurrently a reduced level of c-myc transcript (Fig. 4c). In addition, we noticed that knockdown of Smad4 sensitizes HepG2 cells to cell death induced by serum deprivation, which is executed in a caspase 3/7-independent manner (Fig. 4d and data not shown). HepG2 cells expressing TrxLef1D were also more vulnerable to cell death in the low-serum condition (data not shown). These observations are consistent with a previous report that the inhibition of c-myc expression increases susceptibility of hepatocytes to caspase 3/7-independent apoptosis and necrosis by tumor necrosis factor (24). In summary, both overexpression and knockdown experiments indicate that Smad4 activates the c-myc promoter in the absence of TGF-β signaling.

Prompted by a previous report that bone morphogenetic protein (BMP) signaling activates c-myc transcription in mouse renal tissue (25), we also considered the possibility that endogenous BMP signaling was activating c-myc transcription in HepG2 cells, and that the positive regulator Smad4 on TBE1 is in a complex with Smad1 activated by BMP signaling. However, c-myc promoter activity in HepG2 cells was inhibited by the activation of BMP signaling (Fig. 4e), which opposes the hypothesis that BMP-responsive Smad might be involved as a positive regulator of the c-myc reporter in HepG2 cells.

Tumor-Derived Smad4 Mutants, D351H and R361H, Have Lost TGF-β-Dependent Function but Maintain TGF-β-Independent Function.

Asp-351 and -361 in the loop–helix region of the Smad4 MH2 domain are important residues in forming a Smad4 heterocomplex with R-Smad (26). Mutations of either residue to histidine are found in tumor cells at higher frequency than other missense mutations and completely abolish the interaction with phosphorylated Smad2 and Smad1 (ref. 27 and data not shown). Although these Smad4 mutants lost TGF-β-dependent function, we tested whether they retained the ability to activate c-myc promoter activity. When wild-type Smad4, Smad4D351H, or Smad4R361H proteins were expressed in Smad4-null MDA-MB-468 cells with the TGF-β-responsive SBE12 reporter, only the wild-type Smad4 protein could restore the TGF-β response (Fig. 9, which is published as supporting information on the PNAS web site). However, all three proteins were capable of activating the TIEm myc reporter activity (Fig. 4b). We conclude that these two Smad4 mutations, which are associated with the loss of TGF-β response in human cancer cells, still retain a TGF-β-independent function of Smad4, activation of c-myc, which would be beneficial to the survival and proliferation of the cancer cells.


Negative regulation of c-myc expression by TGF-β is well established and is a key mechanism through which TGF-β causes G1 arrest and inhibition of cell proliferation in epithelial cells. Three studies identified the TIE sequence in the c-myc promoter as mediating the TGF-β effect on c-myc expression. A complex of Smad3-Smad4, E2F4/5, DP1, and p107 binds to the TIE in response to TGF-β to inhibit transcription of c-myc (5, 28, 29). This Smad-dependent repression of c-myc expression was previously the only known function of Smad4 in the regulation of c-myc. The data presented here provide evidence that Smad4 also can function as a positive regulator of c-myc expression in the absence of TGF-β signaling (Fig. 5).

Fig. 5.
Dual role of Smad4 on c-myc regulation.

Smad4 (DPC4) was found in the search for candidate tumor suppressor genes on human chromosome 18q, where ≈90% of pancreatic carcinomas show loss of heterozygosity (8, 30). It is ubiquitously expressed in most tissues, and homozygous Smad4 mutant mice are embryonic lethal, showing reduced cellular proliferation (31, 32). Although most Smad4 research focused on its role in mediating TGF-β signaling, several recent studies suggested TGF-β-independent functions for Smad4. For example, tumor-derived Smad4 proteins that harbor missense mutations in the loop–helix region, resulting in defective TGF-β signaling, showed strong nuclear localization implying alternative functions for the Smad4 protein (33). In prostatic adenocarcinoma, increased Smad4 expression positively correlated with tumor grade, stage, and DNA ploidy (34). Genetic studies from Drosophila reported that some tumor-derived Smad4 mutations, defective in the Dpp signaling pathway, exhibit neomorphic gain of function rather than loss of function, mimicking the phenotype of zw3 (a Wnt/wg antagonist) null mutant (35). Here we report that, in the absence of TGF-β signaling, Smad4 is able to bind to the TBE1 of the c-myc promoter. The interaction with R-Smad is not necessary for this alternative function of Smad4, because it does not require signaling from TGF-β and Smad4 mutants that cannot form a complex with R-Smads are still capable of activating the c-myc promoter. This new function of Smad4 is consistent with the previous observations that Smad4 shuttles between cytoplasm and nucleus, and that even in the absence of TGF-β signaling, a significant portion of Smad4 exists in the nucleus (12, 3638).

The first evidence that Smad and LEF/TCF functionally interact came from studies of the Xtwn promoter in Xenopus, which contains both SBE and TBE elements (13, 14). In mouse, the Smad-LEF/TCF complex is implicated in the gene expression of Emx2, MSX2, and gastrin and also in the process of palatal development (16, 17, 39, 40). The general model proposed on the basis of these studies is that in the presence of either Wnt or BMP/TGF signaling, the target genes are activated moderately, but when both signals are activated, the target genes are synergistically activated by the formation of R-Smad-Smad4-LEF/TCF-β-catenin complex. Furthermore, Hussein et al. (16) reported that, although LEF1 cooperates with Smad1 and Smad4 to activate the MSX2 promoter, Smad4 formed a complex with LEF1 detectable by ChIP in the presence of BMP antagonists, i.e., in the absence of active phosphorylated Smad1, showing a BMP-independent role for Smad4. The regulation of MSX2 and c-myc by Smad4 has a resemblance, in that Smad4 can bind to the promoter even without BMP or TGF-β signaling. A difference is that the TGF-β signal reduces the amount of Smad4 bound to the TBE1 of c-myc, whereas the BMP signal recruits Smad1 to the bound Smad4 to achieve higher levels of MSX2 expression.

A previous study from the Alk3QD (constitutively activated BMP receptor) transgenic mouse showed that BMP signaling can activate the c-myc promoter through a Smad1-TCF-β-catenin complex in mouse renal tissue (25). This raised a question that BMP signaling might be responsible for the positive function of Smad4 on TBE1; that is, the Smad1–Smad4 complex activated by BMP signaling is binding the TBE1 region. However, activation of BMP signaling inhibited c-myc promoter activity in HepG2 cells (Fig. 4e), indicating that BMP signaling may regulate c-myc in a tissue-specific manner. The observation disproves that BMP-responsive Smad is involved as a positive regulator of the c-myc reporter in HepG2 cells. Furthermore, that Smad4 mutants that cannot bind to R-Smads are still competent to activate the c-myc promoter (Fig. 4b) strongly supports the hypothesis that the positive function of Smad4 on c-myc regulation does not require interaction with R-Smads activated by TGF-β or BMP.

Smads can regulate diverse genes depending on the transcription factor they bind, which allows TGF-β to conduct two very opposite functions in cancer: tumor suppressor in the early stage and tumor promoter in the late stage (7, 41). Interestingly, Smad-binding proteins do not share a single Smad-binding domain, indicating there must be multiple binding sites on the surface of Smad. Our group has been developing Smad peptide aptamers to investigate the functions of each Smad complex in TGF-β signaling; the ultimate goal with respect to cancer therapy is finding a peptide aptamer that can specifically inhibit tumor-promoting Smad complexes while maintaining tumor suppressor Smad complexes. In the current study, TrxLef1D did not show any significant effect on the antiproliferative effect of TGF-β/Smad signaling, whereas it did repress the activation of the c-myc oncogene by Smad4 and the proliferation of human hepatocellular carcinoma cells, suggesting that the protein–protein interaction between Smad4 and LEF/TCF may be a good therapeutic target. Furthermore, the result that TrxLef1D sensitizes cells to cell death in low-serum conditions may provide a promising strategy to inhibit the survival of cancer cells.

Materials and Methods

Reporter Gene Assay.

Cells were transfected with FuGene6 (Roche, Indianapolis, IN) according to the manufacturer's instruction. Approximately 24 h later, cells were lysed, and luciferase activity was analyzed with Bright-Glo Luciferase (Promega, Madison, WI). pCMV-β-galactosidase (Fig. 2) or pSV40-β-galactosidase (Fig. 4) was cotransfected in each well and measured with the Galacto-Star kit (Applied Biosystems, Foster City, CA), and all luciferase values were normalized to the level of β-galactosidase activity. Data in each experiment are presented as the mean ± standard deviation of triplicates.

Retroviral Infection and Isolation of Cell Populations Stably Expressing Aptamers.

Peptide aptamers, TrxLef1D and TrxGA, were cloned into the NotI site of the pCMMV-IRES-GFP vector (42). pRetro-shNS (random sequence) and pRetro-shSmad4-b vectors were purchased from Cellogenetics (Ijamsville, MD). Retroviruses were generated as described (11), and then HepG2 or 4T1 cells were incubated with retrovirus supernatants at 4°C for ≈1.5 h with gentle rocking. To obtain HepG2 cells stably expressing aptamer, Fluorescence-Activated Cell Sorting was performed at the University of Wisconsin, Madison, Comprehensive Cancer Center Flow Cytometry Facility, and the top 10% GFP-positive cells were obtained.

WST-1 Assay.

Ten thousand cells of H2-TrxLef1D and H2-TrxGA were seeded in each well of a 96-well plate, incubated overnight and treated by 30 pM TGF-β in 0.2% FBS-containing RPMI-1640 medium. After starting the treatment, WST-1 (Roche) assay measuring the activity of mitochondrial dehydrogenases was performed following the manufacturer's instruction at 0-, 24-, 48-, and 72-h time points.


Human c-myc reporter was kindly provided by B. Vogelstein (Johns Hopkins University School of Medicine, Baltimore, MD) (Del2 construct; ref. 6). The TIEm myc reporter was generated by mutating GGCTTGGCGGG of the TIE region to TTAAAGGTTTC (5) by using the site-directed mutagenesis Quickchange Kit (Stratagene, La Jolla, CA). tttt TIEm myc reporter was generated by using Quickchange Kit. BMP responsive element-luc reporter was generated following the protocol described (43). pCMV5 FlagSmad4 was kindly provided by J. Massagué (Memorial Sloan–Kettering Cancer Center, New York, NY). Smad4D351H and Smad4R361H expression vectors were generated by the site-directed mutagenesis Quikchange Kit (Stratagene) by using pCMV5 FlagSmad4 as template. pCG LEF1 was kindly provided by R. Grosschedl (Max Planck Institute, Freiburg, Germany). pCI TrxLef1D, pCI TrxSARA, and pCI TrxGA were described (11, 12). pcDNA Alk3QD was kindly provided by C.-H. Heldin (Ludwig Institute for Cancer Research, Uppsala, Sweden).


RNA was isolated by RNeasy Mini Kit (Qiagen, Valencia, CA), and the reverse transcription reaction was performed by using SuperScript II Reverse Transcriptase (Invitrogen, Carlsbad, CA). PCR products after 20 (GAPDH) or 25 (myc) cycles were separated on agarose gel and visualized by ethidium bromide. Primer sequences are published as Table 1, which is published as supporting information on the PNAS web site.

DNA Pull-Down Assay.

The HepG2 cell lysate was incubated with 3 μg of biotinylated oligonucleotide at 4°C overnight. After an additional 1-hr incubation with streptavidin-conjugated beads (Pierce, Rockford, IL), the complex was pelleted by centrifugation, washed with lysis buffer, eluted by adding sample buffer and boiling, and subjected to SDS/PAGE. The oligonucleotide sequences are published as Table 1. Antibodies for Western blotting were purchased from Zymed (Smad3) and Santa Cruz (Smad4 B-8, β-catenin, and β-tubulin). SB431542 was purchased from Tocris Cookson, Bristol, U.K.).

ChIP Assay.

HepG2 cells were pretreated with SB431542 at 5 μM or TGF-β1 at 100 pM for 1.5 h, and the chromatin was isolated following procedures described (44). Anti-Smad4 antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA) (H-552). PCRs were done with Taq polymerase using 35 (PAI-1 and TBE1) or 37 cycles (TIE). Primer sequences are published as Table 1.

Supplementary Material

Supporting Information:


We thank Drs. Joan Massagué (Sloan–Kettering Institute, New York, NY), Carl-Henrik Heldin (Ludwig Institute for Cancer Research, Uppsala, Sweden), Bert Vogelstein (Johns Hopkins University School of Medicine, Baltimore, MD), Peter ten Dijke (Leiden University Medical Center, Leiden, The Netherlands), Rik Derynck (University of California, San Francisco, CA), and Rudolf Grosschedl (Max Planck Institute, Freiburg, Germany) for expression constructs and luciferase reporter constructs. This work was supported by R01 CA090875 (to F.M.H.) and Cancer Center Support Grant P30 CA014520. S.K.L. was supported in part as a Wisconsin Distinguished Shapiro Fellow.


TIETGF-β inhibitory element
TBELEF/TCF-binding element
R-Smadsreceptor-regulated Smads
Trxthioredoxin A scaffold.


The authors declare no conflict of interest.

This article is a PNAS direct submission.


1. Grandori C, Cowley SM, James LP, Eisenman RN. Annu Rev Cell Dev Biol. 2000;16:653–699. [PubMed]
2. Dang CV. Mol Cell Biol. 1999;19:1–11. [PMC free article] [PubMed]
3. Nesbit CE, Tersak JM, Prochownik EV. Oncogene. 1999;18:3004–3016. [PubMed]
4. Shachaf CM, Kopelman AM, Arvanitis C, Karlsson A, Beer S, Mandl S, Bachmann MH, Borowsky AD, Ruebner B, Cardiff RD, et al. Nature. 2004;431:1112–1117. [PubMed]
5. Chen CR, Kang Y, Siegel PM, Massagué J. Cell. 2002;110:19–32. [PubMed]
6. He TC, Sparks AB, Rago C, Hermeking H, Zawel L, da Costa LT, Morin PJ, Vogelstein B, Kinzler KW. Science. 1998;281:1509–1512. [PubMed]
7. Massagué J, Seoane J, Wotton D. Genes Dev. 2005;19:2783–2810. [PubMed]
8. Hahn SA, Schutte M, Hoque AT, Moskaluk CA, da Costa LT, Rozenblum E, Weinstein CL, Fischer A, Yeo CJ, Hruban RH, Kern SE. Science. 1996;271:350–353. [PubMed]
9. Schutte M, Hruban RH, Hedrick L, Cho KR, Nadasdy GM, Weinstein CL, Bova GS, Isaacs WB, Cairns P, Nawroz H, et al. Cancer Res. 1996;56:2527–2530. [PubMed]
10. Geyer CR, Colman-Lerner A, Brent R. Proc Natl Acad Sci USA. 1999;96:8567–8572. [PMC free article] [PubMed]
11. Cui Q, Lim SK, Zhao B, Hoffmann FM. Oncogene. 2005;24:3864–3874. [PubMed]
12. Zhao BM, Hoffmann FM. Mol Biol Cell. 2006;17:3819–3831. [PMC free article] [PubMed]
13. Labbe E, Letamendia A, Attisano L. Proc Natl Acad Sci USA. 2000;97:8358–8363. [PMC free article] [PubMed]
14. Nishita M, Hashimoto MK, Ogata S, Laurent MN, Ueno N, Shibuya H, Cho KW. Nature. 2000;403:781–785. [PubMed]
15. Hu MC, Piscione TD, Rosenblum ND. Development (Cambridge, UK) 2003;130:2753–2766. [PubMed]
16. Hussein SM, Duff EK, Sirard C. J Biol Chem. 2003;278:48805–48814. [PubMed]
17. Theil T, Aydin S, Koch S, Grotewold L, Ruther U. Development (Cambridge, UK) 2002;129:3045–3054. [PubMed]
18. Thievessen I, Seifert HH, Swiatkowski S, Florl AR, Schulz WA. Br J Cancer. 2003;88:1932–1938. [PMC free article] [PubMed]
19. Feng XH, Lin X, Derynck R. EMBO J. 2000;19:5178–5193. [PMC free article] [PubMed]
20. Moustakas A, Kardassis D. Proc Natl Acad Sci USA. 1998;95:6733–6738. [PMC free article] [PubMed]
21. Brown KA, Aakre ME, Gorska AE, Price JO, Eltom SE, Pietenpol JA, Moses HL. Br Cancer Res. 2004;6:R215–R231. [PMC free article] [PubMed]
22. McEarchern JA, Kobie JJ, Mack V, Wu RS, Meade-Tollin L, Arteaga CL, Dumont N, Besselsen D, Seftor E, Hendrix MJ, et al. Int J Cancer. 2001;91:76–82. [PubMed]
23. Dennler S, Itoh S, Vivien D, ten Dijke P, Huet S, Gauthier JM. EMBO J. 1998;17:3091–3100. [PMC free article] [PubMed]
24. Liu H, Lo CR, Jones BE, Pradhan Z, Srinivasan A, Valentino KL, Stockert RJ, Czaja MJ. J Biol Chem. 2000;275:40155–40162. [PubMed]
25. Hu MC, Rosenblum ND. Development (Cambridge, UK) 2005;132:215–225. [PubMed]
26. Shi Y, Hata A, Lo RS, Massague J, Pavletich NP. Nature. 1997;388:87–93. [PubMed]
27. Wu JW, Hu M, Chai J, Seoane J, Huse M, Li C, Rigotti DJ, Kyin S, Muir TW, Fairman R, et al. Mol Cell. 2001;8:1277–1289. [PubMed]
28. Yagi K, Furuhashi M, Aoki H, Goto D, Kuwano H, Sugamura K, Miyazono K, Kato M. J Biol Chem. 2002;277:854–861. [PubMed]
29. Frederick JP, Liberati NT, Waddell DS, Shi Y, Wang XF. Mol Cell Biol. 2004;24:2546–2559. [PMC free article] [PubMed]
30. Hahn SA, Seymour AB, Hoque AT, Schutte M., da Costa LT, Redston MS, Caldas C, Weinstein CL, Fischer A, Yeo CJ, et al. Cancer Res. 1995;55:4670–4675. [PubMed]
31. Sirard C, de la Pompa JL, Elia A, Itie A, Mirtsos C, Cheung A, Hahn S, Wakeham A, Schwartz L, Kern SE, Rossant J, Mak TW. Genes Dev. 1998;12:107–119. [PMC free article] [PubMed]
32. Yang X, Li C, Xu X, Deng C. Proc Natl Acad Sci USA. 1998;95:3667–3672. [PMC free article] [PubMed]
33. Iacobuzio-Donahue CA, Song J, Parmiagiani G, Yeo CJ, Hruban RH, Kern SE. Clin Cancer Res. 2004;10:1597–1604. [PubMed]
34. Sheehan GM, Kallakury BV, Sheehan CE, Fisher HA, Kaufman RP, Jr, Ross JS. Hum Pathol. 2005;36:1204–1209. [PubMed]
35. Takaesu NT, Herbig E, Zhitomersky D, O'Connor MB, Newfeld SJ. Development (Cambridge, UK) 2005;132:4883–4894. [PubMed]
36. Pierreux CE, Nicolas FJ, Hill CS. Mol Cell Biol. 2000;20:9041–9054. [PMC free article] [PubMed]
37. Xu L, Massague J. Nat Rev Mol Cell Biol. 2004;5:209–219. [PubMed]
38. De Bosscher K, Hill CS, Nicolas FJ. Biochem J. 2004;379:209–216. [PMC free article] [PubMed]
39. Nawshad A, Hay ED. J Cell Biol. 2003;163:1291–1301. [PMC free article] [PubMed]
40. Lei S, Dubeykovskiy A, Chakladar A, Wojtukiewicz L, Wang TC. J Biol Chem. 2004;279:42492–42502. [PubMed]
41. Feng XH, Derynck R. Annu Rev Cell Dev Biol. 2005;21:659–693. [PubMed]
42. Kennedy G, Sugden B. Mol Cell Biol. 2003;23:6901–6908. [PMC free article] [PubMed]
43. Korchynskyi O, ten Dijke P. J Biol Chem. 2002;277:4883–4891. [PubMed]
44. Wang J, Lindner SE, Leight ER, Sugden B. Mol Cell Biol. 2006;26:1124–1134. [PMC free article] [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem chemical compound records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records. Multiple substance records may contribute to the PubChem compound record.
  • Gene
    Gene records that cite the current articles. Citations in Gene are added manually by NCBI or imported from outside public resources.
  • GEO Profiles
    GEO Profiles
    Gene Expression Omnibus (GEO) Profiles of molecular abundance data. The current articles are references on the Gene record associated with the GEO profile.
  • HomoloGene
    HomoloGene clusters of homologous genes and sequences that cite the current articles. These are references on the Gene and sequence records in the HomoloGene entry.
  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem chemical substance records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records.
  • Taxonomy
    Taxonomy records associated with the current articles through taxonomic information on related molecular database records (Nucleotide, Protein, Gene, SNP, Structure).
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...