HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Nishihara, A. Articles by Kawabata, M. Search for Related Content PubMed PubMed Citation Articles by Nishihara, A. Articles by Kawabata, M. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 40, 28716-28723, October 1, 1999

E1A Inhibits Transforming Growth Factor- Signaling through Binding to Smad Proteins^*,

Ayako Nishihara, Jun-ichi Hanai, Takeshi Imamura, Kohei Miyazono, and Masahiro Kawabata

From the Department of Biochemistry, The Cancer Institute of Japanese Foundation for Cancer Research, and Research for the Future Program, Japan Society for Promotion of Science, 1-37-1 Kami-ikebukuro, Toshima-ku, Tokyo 170-8455, Japan

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Smads form a recently identified family of proteins that mediate intracellular signaling of the transforming growth factor (TGF)- superfamily. Smads bind to DNA and act as transcriptional regulators. Smads interact with a variety of transcription factors, and the interaction is likely to determine the target specificity of gene induction. Smads also associate with transcriptional coactivators such as p300 and CBP. E1A, an adenoviral oncoprotein, inhibits TGF--induced transactivation, and the ability of E1A to bind p300/CBP is required for the inhibition. Here we determined the Smad interaction domain (SID) in p300 and found that two adjacent regions are required for the interaction. One of the regions is the C/H3 domain conserved between p300 and CBP, and the other is a nonconserved region. p300 mutants containing SID inhibit transactivation by TGF- in a dose-dependent manner. E1A inhibits the interaction of Smad3 with a p300 mutant that contains SID but lacks the E1A binding domain. We found that E1A interacts specifically with receptor-regulated Smads, suggesting a novel mechanism whereby E1A antagonizes TGF- signaling.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Members of the transforming growth factor (TGF)¹- superfamily constitute a major class of secreted polypeptides that mediate cell-cell communication in metazoan organisms (1). TGF-s, activins, bone morphogenetic proteins (BMPs), and other ligands belonging to this family govern the fate of cells of a variety of origins. TGF--related factors regulate cell proliferation, differentiation, adhesion, migration, and apoptosis through transcriptional regulation of a diverse array of genes such as cell cycle regulators, adhesion molecules, cytokines, transcription factors, and as yet unidentified targets. Receptors for the TGF- superfamily members are transmembrane serine-threonine kinases classified into two distinct groups, termed type I and type II. Type II receptors are constitutively active kinases, whereas type I receptors are dormant without stimulation. Upon ligand binding, type II receptors phosphorylate type I receptors at the juxtamembrane region. The activated type I receptors, in turn, phosphorylate intracellular signaling mediators with a generic name of Smad.

Smads are classified into three groups depending on the role in signaling: receptor-regulated Smads (R-Smads), common mediator Smads (Co-Smads), and antagonistic Smads (Anti-Smads) (2, 3). R-Smads are direct substrates of the type I receptor kinases. R-Smads exist as monomers in the absence of ligand stimulation (4). Activated type I receptors phosphorylate R-Smads at the C-terminal SSXS motif containing three serines, the last two of which serve as direct phosphorylation sites. R-Smads then form complexes with Co-Smads. The heteromeric oligomers translocate from the cytoplasm to the nucleus where they act as transcriptional regulators. Anti-Smads stably bind to the receptors, thereby interfering with the phosphorylation of R-Smads. The expression of Anti-Smads are induced by ligands, and Smads thus form an autoregulatory feedback loop inside the cell. Among R-Smads, Smads 2 and 3 respond to TGF-/activin stimulation, whereas Smads 1, 5, and 8 mediate BMP signals. While Smad4 is the only Co-Smad identified so far in mammals, two Co-Smads were recently found in Xenopus (5). Smads 6 and 7 belong to Anti-Smads. In Drosophila, Mad responds to Decapentaplegic (Dpp), a BMP-like ligand, whereas dSmad2 propagates TGF-/activin-like signals (6, 7). Medea is a Co-Smad, and Dad is an Anti-Smad. Signaling by the three classes of Smads is conserved in Drosophila (8).

R-Smads and Co-Smads share two conserved regions denoted the N-terminal MH1 and C-terminal MH2 domains. The MH1 and MH2 domains are tied by a linker region that is variable in the amino acid sequence. Anti-Smads contain the MH2 region, but their N-terminal regions are significantly diverged from the conserved MH1 region. R-Smads and Co-Smads directly bind to DNA through the MH1 domain. The MH2 region is the effector domain that has intrinsic transactivation activity. MH2 also mediates protein-protein association, including Smad-receptor, Smad-Smad, and Smad-nuclear protein interactions. Growth factors or Ras modulate the intracellular localization of R-Smads by phosphorylating the linker region through the mitogen-activated protein kinase cascade (9, 10).

Interaction of Smads with nuclear proteins is likely to form the basis of transcriptional regulation by members of the TGF- superfamily. FAST-1 was isolated as a transcription factor that binds to the activin-responsive element in the Mix.2 promoter in Xenopus (11). Activin-induced complex formation of FAST-1, Smad2, and Smad4 is required for the induction of the Mix.2 expression. AP-1 interacts with Smad3 and synergistically activates the expression of the human collagenase I gene (12). TGF- induces the expression of the plasminogen inhibitor-1 (PAI-1) gene, and AP-1 (12) and TFE3 (13) have been implicated in the induction, although the direct interaction of the latter with Smads has not been shown. Functional interaction of Smads with Sp1 has been reported (14). Smad3, on the other hand, binds to vitamin D receptor through the MH1 domain and acts as a transcriptional coactivator (15).

Several Smad binding motifs on DNA have been revealed. Screening of oligonucleotides by polymerase chain reaction-gel shift selection resulted in the identification of a palindromic sequence of GTCTAGAC as the consensus binding motif for Smad3 and Smad4 (16). Examination of the PAI-1 gene resulted in the identification of "CAGA box" as a sequence motif that binds Smads (17). An almost identical sequence was found as a Smad binding site in the junB promoter (18). Dpp induces the expression of various genes such as vestigial, and examination of various Dpp-responsive genes revealed GCCGnCGC as a Mad binding sequence. A similar sequence was found to recruit Mad and Medea from the analysis of a Dpp-responsive enhancer in tinman (19). An important indication deduced from the observations described above is that the DNA binding affinity of Smads may be relatively low and the binding specificity may not be rigid. Other sequence-specific DNA binding factors are likely to be required for the tight and specific Smad-DNA interaction in vivo (20).

Transcriptional coactivators such as p300 and CBP possess the histone acetyltransferase activity that loosens the condensed structure of chromatin and promotes the accessibility of transactivation machinery to target DNA (21). p300/CBP also directly interacts with RNA polymerase II in the basal transcription machinery. It has been shown that p300/CBP interacts with mammalian Smads and enhances transactivation by TGF- (22-26). In Drosophila, CBP is shown to be required for Dpp responses in vivo (27). p300 also bridges Smad1 and STAT3, thereby converging two distinct signaling pathways of BMP-2 and LIF (28). E1A, an adenoviral oncoprotein, interacts with p300 or CBP and directly inhibits the histone acetyltransferase activity (29). Here we determined the Smad interaction domain in p300 and have found that two adjacent regions are required for efficient association of the two proteins. One of them is the C/H3 domain, highly conserved between p300 and CBP, which binds E1A. The other domain is significantly diverged in the primary amino acid sequence between p300 and CBP. E1A interferes with the interaction between p300 and Smad3, but, the interaction of Smad3 with p300 lacking the E1A binding domain was also inhibited by E1A. We then found that E1A interacts with Smads, which seems to cause the inhibition of the p300-Smad interaction. E1A is thus likely to block TGF- signaling by directly inhibiting the activity of Smad proteins.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Plasmid Construction-- The construction of the expression plasmids of the wild type p300 and deletion mutant containing amino acids 1737-2414 was described previously (25). p300 deletion mutants shown in Fig. 1A were made using blunt cutting enzymes shown in the figure and pSKiMODs (25) that attach N-terminal EcoRI and C-terminal XhoI sites with a stop codon. The p300 fragments were subcloned into pJG4-5 (see below), FLAG-pcDNA3, or pcDEF3. p300 deletion mutants shown in Fig. 1B were made using polymerase chain reaction. The sequences of the primers are available upon request. The two-hybrid plasmids of mouse CBP were constructed in a similar manner. The internal EcoRI and XhoI sites were first removed, and N-terminal EcoRI and C-terminal XhoI sites were added, and the resulting fragments were subcloned into pJG4-5. The constructions of the other plasmids were described elsewhere (4, 25).

Yeast Two-hybrid Assay-- Yeast two-hybrid assays were done as described (25). Briefly, EGY48, the host yeast strain, was transformed with combinations of the reporter (pSH18-34), a bait (pEG202 or Smad3 in pEG202), and a prey (pJG4-5 or p300 in pJG4-5). Yeast was selected on appropriate growth media and then three independent colonies were assayed for the -galactosidase activity on 5-bromo-4-chloro-3-indolyl -D-galactopyranoside (X-gal) plates.

Protein-Protein Interaction in Vivo-- COS-7 cells were used for the detection of protein-protein interaction in vivo. Cells were transfected with an appropriate combination of expression plasmids, washed, scraped, and solubilized in a buffer containing 20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% Triton X-100, 0.5% deoxycholate, 1% aprotinin, and 1 mM phenylmethylsulfonyl fluoride. Lysates were cleared and incubated with anti-FLAG M2 antibody (Sigma) or anti-E1A M73 antibody (Santa Cruz), followed by incubation with protein G-Sepharose beads (Amersham Pharmacia Biotech). The beads were washed with solubilization buffer without deoxycholate, and the immunoprecipitates were eluted by boiling for 3 min in SDS sample buffer (100 mM Tris-HCl, pH 8.8, 0.01% bromphenol blue, 36% glycerol, 4% SDS) containing 10 mM dithiothreitol and subjected to SDS-gel electrophoresis. Proteins were electrotransferred to nitrocellulose filters, immunoblotted with anti-Myc 9E10 antibody, anti-E1A M73 antibody, or anti-FLAG M2 antibody. The bands were detected using the enhanced chemiluminescence detection system (Amersham Pharmacia Biotech). Some of the lysates were directly subjected to Western blotting without immunoprecipitation.

Luciferase Assay-- Luciferase assays were carried out using the p3TP-Lux reporter (30) and mink lung epithelial R mutant cells. Cells were transiently transfected with an appropriate combination of the reporter, expression plasmids, and pcDNA3 using FuGENE 6 (Roche Molecular Biochemicals). Total amounts of the transfected DNA were the same throughout the experiments, and luciferase activities were normalized using the sea-pansy luciferase activity under the control of the thymidine kinase promoter (25).

DNA Binding Assay Using Biotinylated Oligonucleotides-- Lysates of transfected COS-7 cells were precleared with streptavidin-agarose (Sigma) and then incubated with 30 pmol of biotinylated double-stranded oligonucleotides and 12 µg of poly(dI-dC) for overnight at 4 °C. DNA-bound proteins were precipitated with streptavidin-agarose for 30 min at 4 °C, washed, and detected by Western blotting. The sequences of the 3xCAGA probe are: 5'-TCGAGAGCCAGACAAGGAGCCAGACAAGGAGCCAGACACTCGAG-3' (sense strand) and 5'-CTCGAGTGTCTGGCTCCTTGTCTGGCTCCTTGTCTGGCTCTCGA-3' (antisense strand). The sense strand was biotinylated at the 5'-end (31).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Determination of Smad Interaction Domain in p300 in Yeast-- In our previous report (25), we identified the C-terminal region (amino acids 1737-2414) of p300 as the Smad-interacting region (Fig. 1A). This region interacted with Smad3 both in the yeast two-hybrid assay and in vivo. In the present study, we made an additional series of p300 deletion mutants and tested the interaction in the two-hybrid assay. The intensity of the interaction of each mutant is summarized in Fig. 1A. Amino acids 2247-2414 are dispensable because DEL-1 interacted with Smad3 as efficiently as the full-length p300. However, further C-terminal deletion (DEL-2 and DEL-3) progressively diminished the interaction. DEL-4 did not interact with Smad3 at all, whereas DEL-5 showed weak binding. We then tested three deletion mutants (DEL-6, -7, and -8) and found that DEL-8 interacts with Smad3 as efficiently as p300 (1737-2414). DEL-9 that overlaps with the C/H3 domain conserved between p300 and CBP did not interact with Smad3 in yeast.

View larger version (32K): [in this window] [in a new window]   Fig. 1.   Two-hybrid assay of Smad3-p300/CBP interaction. A, the interaction of various p300 deletion mutants and Smad3 was tested in the yeast two-hybrid assay. The deletion mutants used are shown with the result of the assay (ranging from negative, , to strongly positive, +++). The enzymes used for the construction of each deletion mutant are shown. The numbers indicate the locations in the amino acid sequence of p300. The hatched region is the C/H3 domain. SID stands for Smad interaction domain (see the text). B, the interaction of p300/CBP and Smad3 was studied in further detail. The amino acid sequences of human p300 and mouse CBP are compared. The DEL-11 mutants have amino acid change to alanines as indicated. Broken lines represent deletion. DEL-19 is the full-length p300 with an internal deletion indicated in the figure. CBP/DEL-1 and CBP/DEL-2 extend to the C-terminal end of the wild type CBP. The result of the two-hybrid assay is summarized on the right.

Most part of DEL-8 is not conserved between p300 and CBP (Fig. 1B). Therefore, we tested several CBP deletion mutants and found that CBP/DEL-3 corresponding to part of DEL-8 also interacts with Smad3. We then tried to determine the minimal requirement of the interaction. DEL-10 is almost identical to DEL-8, lacking only a short stretch of the C-terminal part, and interacted with Smad3 as well. Within the amino acid sequence of DEL-10, several clusters of amino acids are conserved between p300 and CBP. We mutated these residues to alanines (DEL-11-1 to DEL-11-4), but none of the mutations disrupted the interaction. DEL-13 lacking the N-terminal conserved region of DEL-10 still interacted with Smad3. However, two internal deletions within DEL-13 (DEL-14 and -15) abrogated the interaction. DEL-16 (amino acids 1884-1975) interacted with Smad3 as efficiently as the full-length p300. N-terminal deletion of DEL-13 (DEL-17) diminished the interaction, whereas N-terminal deletion of DEL-16 (DEL-18) lost the interaction. We made an internal deletion mutant of p300 missing the protein sequence of DEL-13 (DEL-19), and this mutant did not interact with Smad3. Taken together, we concluded that the region of DEL-13 is the Smad-interacting domain determined by the two-hybrid assay.

Two Adjacent Regions in p300 Are Required for the p300-Smad3 Interaction in Vivo-- We next tested the interaction in vivo using COS cells. p300 (1737-2414) was used as a positive control, which interacted with Smad3 in a ligand-dependent manner (Fig. 2A). As shown in the yeast assay, C-terminal deletion of the mutant p300 (1737-2414) progressively decreased the interaction (DEL-1 to -3). DEL-4 or -5 did not efficiently interact with Smad3 in vivo in consistent with the result of the yeast assay. In contrast to the two-hybrid assay, however, DEL-7 interacted with Smad3, much less efficiently than p300 (1737-2414) or DEL-1 in vivo, suggesting that a region N-terminal to DEL-7 may contribute to the interaction in vivo. We compared three mutants for the interaction with Smad3 (Fig. 2B). DEL-8 and DEL-9 interacted with Smad3, but DEL-2 containing the combined region of DEL-8 and DEL-9 interacted much more efficiently. We thus refer the regions of DEL-9, DEL-8, and DEL-2 to Smad-interacting domain-1 (SID-1), SID-2, and SID, respectively (Fig. 1A). SID-1 overlaps with the C/H3 domain, whereas SID-2 contains a nonconserved region between p300 and CBP. SID-1 and SID-2 synergistically contribute to the interaction of p300 with Smad3 in vivo.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Interaction of p300 with Smad3 in vivo. A, the interaction of various p300 deletion mutants with Smad3 in vivo was studied by immunoprecipitation (IP) followed by immunoblotting (Blot). TR-I(TD) is a constitutively active form of TGF- type I receptor. The lower two panels show the expression of Smad3 and p300 deletion mutants. B, the interaction of three Smad interaction domains of p300 with Smad3 was compared, as in A. p300 (1737-2414) is the positive control for the interaction. DEL-2/SID contains both DEL-8/SID-2 and DEL-9/SID-1. The lower two panels show the expression of Smad3 and p300 deletion mutants.

p300 Is Involved in the DNA Binding Complex of Smad3 and Required for Transactivation by TGF--- Feng et al. (23) showed that p300 is involved in the DNA binding complex of Smad3 and Smad4 using the PAI-1 promoter. We utilized the CAGA sequence previously shown to bind Smads (17). Biotinylated CAGA oligonucleotides were mixed with lysates of COS cells transfected with various combinations of expression plasmids and subjected to purification using streptavidin beads (Fig. 3). A relatively low level of Smad3 bound to the CAGA sequence in the absence of the constitutively active form of TGF- type I receptor, TR-I(TD), whereas p300 alone did not bind to CAGA at all. TGF- stimulation greatly increased the binding of Smad3 to CAGA, and cotransfection of Smad3 and p300 revealed that p300 is incorporated in the Smad3 DNA binding complex.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   DNA binding of p300 via Smad3. Lysates of COS cells transfected with the indicated combinations of plasmids were mixed with biotinylated oligonucleotides with the Smad binding motif (3xCAGA). The bound proteins were purified with streptavidin and analyzed by immunoblotting. The lower two panels show the expression of Smad3 and p300.

We showed previously that p300 (1737-2414) containing SID, but lacking the histone acetyltransferase domain, suppressed transcriptional activation by TGF- in a dominant-negative manner (25). The effect of various p300 deletion mutants on TGF--induced transactivation was studied in a luciferase assay using the p3TP-Lux reporter (30) (Fig. 4A). TR-I(TD) stimulated the activity of p3TP-Lux, and DEL-1 suppressed the activation. p300 (1737-2414) and DEL-1 showed similar dominant-negative effects (data not shown). In accordance with the result of the two-hybrid assay (Fig. 1A), C-terminal deletion of DEL-1 (DEL-2 and -3) mitigated the suppression. DEL-7 and DEL-8/SID-2 also suppressed the transactivation of the reporter. DEL-4 that does not interact with p300 either in the two-hybrid assay or in vivo did not suppress the transactivation at all. In contrast, DEL-9/SID-1 rather enhanced the activity of the reporter. The C/H3 domain contained in DEL-9/SID-1 interacts not only with Smad3 but with various proteins including E1A, which may complicate the result. DEL-9/SID-1 might inhibit the transcriptional repression activity of cellular E1A-like protein. Wild type p300 enhanced the transactivation of p3TP-Lux by TGF-, whereas DEL-1 and DEL-7 inhibited the reporter activity in a dose-dependent manner (Fig. 4B). The suppression of the TGF--induced p3TP-Lux activation by DEL-8/SID-2 and DEL-7 containing SID-2 provides another evidence that p300 contributes to transcriptional activation by TGF-.

View larger version (32K): [in this window] [in a new window]   Fig. 4.   Dominant-negative effect of p300 deletion mutants in transactivation by TGF-. A, the effect of p300 deletion mutants on transactivation by TGF- was examined using the p3TP-Lux luciferase reporter. Experiments were repeated several times, and one of the representative results is shown. The S.D. of duplicates is shown in a vertical line. B, dose-dependent suppression of the TGF- stimulated p3TP-Lux activity by two p300 deletion mutants was examined as in A. Expression plasmids encoding the wild type p300 (1.2 µg), and increasing amounts (0.01, 0.05, 0.1, or 0.2 µg) of DEL-1 and DEL-7 as indicated by triangles were used.

E1A Interacts with Smad3 and Interferes with the Interaction of Smad3 with p300-- E1A interferes with the transactivation by TGF- (22-26, 32). We studied the effect of E1A on the Smad3-p300 interaction in the presence of TR-I(TD) (Fig. 5A). E1A potently inhibited the interaction of Smad3 with the wild type p300 as well as with p300 (1737-2414). The interaction of E1A with the C/H3 domain in p300 may competitively inhibit the Smad3-p300 interaction. However, the interaction of Smad3 with DEL-8/SID-2 lacking the C/H3 domain was also inhibited by E1A. We tested the interaction of E1A with p300 (Fig. 5B). As reported previously, E1A interacted with the wild type p300 and p300 (1737-2414) containing the C/H3 domain. In contrast, E1A failed to interact with DEL-8/SID-2. These results prompted us to investigate the possibility of the association of E1A with Smad3.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Effect of E1A on the Smad3-p300 interaction. A, COS-7 cells were transfected with the indicated combinations of plasmids, and the effect of E1A expression on the interaction of p300 with Smad3 was examined. The lower three panels show the expression of Smad3, E1A, and p300. B, interaction of E1A with wild type and mutant p300 was studied. The lower two panels show the expression of E1A and p300.

E1A interacted with Smad3 both in the absence and presence of TGF- stimulation (Fig. 6A). The E1A interaction domain in Smad3 was determined using various Smad3 deletion mutants (Fig. 6B). The MH1 domain did not interact with E1A, whereas the MH2 domain strongly interacted with E1A. The linker region is not likely to interact with E1A because MH1+Linker did not interact with E1A. We next tested the interaction of E1A with various species of Smads (Fig. 6C). E1A interacted with Smad1, Smad2, and Smad3, but not with Smad4 or Smad6. Thus E1A specifically interacts with R-Smads, but not with Co-Smads or Anti-Smads. These results suggest that at least one of the mechanisms of the inhibition of the Smad3-p300 interaction by E1A is likely to be the direct association of E1A with Smad3.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Interaction of E1A with Smads. A, the interaction of E1A with Smad3 was studied in the absence or presence of TGF- stimulation. The lower two panels show the expression of Smad3 and E1A. B, the E1A interaction domain in Smad3 was determined. WT denotes the wild type Smad3. L stands for the linker region and MH1, MH1+L, L+MH2, and MH2 contain 1-145, 1-219, 146-425, and 220-425 amino acids of Smad3, respectively. The lower two panels show the expression of E1A and Smad3. C, the interaction of E1A with various Smads was examined. The lower two panels show the expression of E1A and Smads.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The recent identification of the Smad family proteins has enabled the investigation of the molecular mechanism of transcriptional regulation by members of the TGF- superfamily. Janknecht et al. (22) identified Smad3 as a CBP interacting protein using the two-hybrid screen. Smad4 with a C-terminal deletion of 42 amino acids also interacted with p300, whereas the full-length Smad4 did not interact. The Smad-interacting domain in CBP determined in GST pull-down assays was 285 amino acids from 1891 to 2175. The corresponding region in p300 is amino acids 1855-2171. The Smad-interacting domain reduced TGF--induced transactivation. Feng et al. (23) identified two Smad-interacting domains in CBP using the mammalian two-hybrid assay. One is a low affinity domain (amino acids 1-451), and the other is a high affinity domain (amino acids 1891-2441). We also observed weak interaction between Smad1 and the N-terminal part of p300 (amino acids 1-1030) in vivo (28). From the combination of the results of the yeast two-hybrid assay, Feng et al. (23) concluded that the Smad-interacting domain in CBP is amino acids 1891-1991, although they did not show the direct interaction of the region with Smad3. The corresponding region in p300 is amino acids 1855-1960. Topper et al. (24) also reported that the C-terminal region of CBP (amino acids 1892-2441) interacts with Smad2 and Smad4 using the mammalian two-hybrid assay. The interaction of Smad4 with CBP was TGF- dependent, suggesting that Smad4 may interact with CBP via endogenous R-Smads (23). Shen et al. (26) identified amino acids 1572-2414 in p300 as a Smad-interacting domain using GST pull-down assays. Waltzer and Bienz (27) identified amino acids 2413-2608 in Drosophila CBP as a minimal region required for Mad binding using GST pull-down assays. The corresponding region in p300 is amino acids 1741-1933. They observed that amino acids 2240 to 2507 of Drosophila CBP (amino acids 1675-1843 in p300) interacts with Mad in one of the GST pull-down assays, but not in the yeast two-hybrid assay. This region contains the C/H3 domain, but not SID-2 determined in our assay.

We identified two adjacent regions required for p300 to interact with Smad3. SID-1 overlaps with the C/H3 domain conserved between p300 and CBP, and SID-2 contains a nonconserved region between the two coactivators. DEL-7 and DEL-8 that strongly interacted with Smad3 in the two-hybrid assay did not efficiently bind to Smad3 in vivo. Furthermore, DEL-9/SID-1 did not show interaction in the yeast assay. This apparent discrepancy could have been caused by the fusion of a relatively large protein moiety containing a transactivation domain to the p300 deletion mutants, which may alter the structure of the N-terminal SID-1 portion of the p300 prey proteins. SID-2 in CBP mediates the interaction of CBP with Smad3 because one of the CBP deletion mutants, CBP/DEL-3, interacted with Smad3 in the two-hybrid assay (Fig. 1B). We showed that the p300-Smad3 interaction in vivo is augmented in a synergistic manner when both SID-1 and SID-2 are present (Fig. 2B). As the primary amino acid sequence of SID-2 is not conserved between p300 and CBP, three-dimensional conformation of this region may be conserved between the two proteins.

We showed that p300 is incorporated in the DNA binding complex of Smad3 using the TGF--responsive CAGA sequence (Fig. 3). In our previous study (25), we showed that p300 augments the transactivation of the p3TP-Lux reporter by TGF-, and the p300 (1737-2414) containing SID suppresses the activation in a dominant-negative manner. In the present study, DEL-7 containing SID-2 and the following C-terminal sequence inhibited TGF--induced transactivation most potently (Fig. 4A). As the progressive C-terminal deletion of DEL-1 diminished the dominant-negative action of DEL-1, not only SID-2 but its C-terminal following sequence may contribute to the suppression. It should be noted that the C-terminal part of DEL-2 is identical with that of DEL-8/SID-2. In contrast, DEL-9/SID-1 rather enhanced the transactivation by TGF-. The C/H3 domain has been shown to interact with various proteins including E1A. One of the explanations for the unexpected result is that DEL-9/SID-1 may sequester endogenous E1A-like proteins and relieves p300 from the transcriptional repression by the protein. This effect may have caused DEL-7 to be a more potent dominant-negative suppressor of TGF- activity than DEL-2/SID.

E1A has been shown to antagonize TGF- signaling in various systems. TGF- down-regulates c-myc expression, and E1A blocks the down-regulation (33). The retinoblastoma (Rb) binding domain in E1A was shown to be essential to this effect. E1A renders cells resistant to growth inhibition by TGF-. Missero et al. (34) showed that both Rb binding domain and p300 binding domain in E1A are required for the full suppression of TGF--induced growth arrest. TGF- represses the expression of the Cdc2 kinase, which is abrogated by the wild type E1A or a mutant E1A defective for Rb binding (35). The expression of junB is induced by TGF-, phorbol ester, and serum, and E1A selectively inhibits the induction by TGF-, but not by phorbol ester or serum (36). de Groot et al. (37) showed that E1A antagonizes both growth stimulatory and inhibitory effects of TGF-. p15 and p21 Cdk inhibitors mediate at least part of TGF--induced cell growth arrest. E1A inhibits the induction of p15 and p21 by TGF-, and the inhibition depends on E1A's ability to bind p300 (38). These observations argue that proteins that interact with E1A play important roles in TGF- signaling.

E1A binds to Rb and p300/CBP. The mechanisms proposed to date whereby E1A antagonizes TGF- can thus be summarized as 1) Rb-dependent and 2) p300/CBP-dependent ones. 1) Rb is a potent growth suppressor, and inactivation of Rb is required for cell cycle progression. Phosphorylation of Rb by the Cdk kinases relieves E2Fs that induce the expression of genes required for the G₁/S transition. DNA tumor viral oncoproteins such as E1A, SV40 large T antigen, and HPV E7 sequester and inactivate Rb, thereby disrupting the major cellular growth control, eventually leading to tumorigenesis. Growth arrest by TGF- is dependent on Rb and interfered with by direct binding of E1A to Rb (33, 34, 37). 2) TGF- down-regulates the expression of positive cell cycle regulators such as c-myc and Cdc25A and transactivates negative cell cycle regulators such as p15 and p21 cdk inhibitors. E1A suppresses the induction of p15 and p21, which is dependent on p300 (38). In Wnt signaling, CBP was shown to act as a transcriptional repressor by competing with -catenin in binding to TCF, a transcription factor activated by Wnt (39). p300/CBP may thus play a role in the down-regulation of c-myc or cdc25A and could be a target of E1A in transcriptional repression as well. In the present report, we propose the third mechanism of the E1A antagonization of TGF- signaling. We demonstrated that E1A associates with Smad3 and inhibits the interaction of Smad3 and p300. Therefore, E1A may directly inhibit TGF- signaling by binding to the signaling mediators of TGF-.

	ACKNOWLEDGEMENTS

We thank R. Eckner for human p300, R. Goodman for mouse CBP, M. Ikeda for E1A, R. Derynck for Smad3, J. Massagué for p3TP-Lux and mink R mutant cells, J. A. Langer for pcDEF3, and R. Brent for the two-hybrid system.

	FOOTNOTES

^* This work was supported by grants-in-aid for scientific research from the Ministry of Education, Science, Sports, and Culture of Japan and special coordination funds for promoting science and technology from the Science and Technology Agency.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The on-line version of this article (available at http://www.jbc.org) contains supplemental Fig. S1.

Supported by Princess Takamatsu Cancer Research Foundation and Sagawa Cancer Research Promotion Foundation. To whom correspondence should be addressed: Dept. of Biochemistry, The Cancer Institute of JFCR, 1-37-1 Kami-ikebukuro, Toshima-ku, Tokyo 170-8455, Japan. Tel.:/Fax: 81-3-3918-0342; E-mail: mkawabat-ind@umin.u-tokyo.ac.jp.

	ABBREVIATIONS

The abbreviations used are: TGF-, transforming growth factor-; SID, Smad interaction domain; BMP, bone morphogenetic protein; R-Smad, receptor-regulated Smad; Co-Smad, common mediator Smad; Anti-Smad, antagonistic Smad; Dpp, Decapentaplegic; PAI-1, plasminogen activator inhibitor-1; TR-I, TGF- type I receptor; Rb, retinoblastoma.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				D. L. Di Bartolo, M. Cannon, Y.-F. Liu, R. Renne, A. Chadburn, C. Boshoff, and E. Cesarman KSHV LANA inhibits TGF-{beta} signaling through epigenetic silencing of the TGF-{beta} type II receptor Blood, May 1, 2008; 111(9): 4731 - 4740. [Abstract] [Full Text] [PDF]

				X. Zhao, J. M. Nicholls, and Y.-G. Chen Severe Acute Respiratory Syndrome-associated Coronavirus Nucleocapsid Protein Interacts with Smad3 and Modulates Transforming Growth Factor-{beta} Signaling J. Biol. Chem., February 8, 2008; 283(6): 3272 - 3280. [Abstract] [Full Text] [PDF]

				A. W. Tu and K. Luo Acetylation of Smad2 by the Co-activator p300 Regulates Activin and Transforming Growth Factor beta Response J. Biol. Chem., July 20, 2007; 282(29): 21187 - 21196. [Abstract] [Full Text] [PDF]

				Y. Nagano, K. J. Mavrakis, K. L. Lee, T. Fujii, D. Koinuma, H. Sase, K. Yuki, K. Isogaya, M. Saitoh, T. Imamura, et al. Arkadia Induces Degradation of SnoN and c-Ski to Enhance Transforming Growth Factor-beta Signaling J. Biol. Chem., July 13, 2007; 282(28): 20492 - 20501. [Abstract] [Full Text] [PDF]

				S.-H. Jeon, B.-C. Chae, H.-A Kim, G.-Y. Seo, D.-W. Seo, G.-T. Chun, N.-S. Kim, S.-W. Yie, W.-H. Byeon, S.-H. Eom, et al. Mechanisms underlying TGF-{beta}1-induced expression of VEGF and Flk-1 in mouse macrophages and their implications for angiogenesis J. Leukoc. Biol., February 1, 2007; 81(2): 557 - 566. [Abstract] [Full Text] [PDF]

				S.-H. Choi and S. B. Hwang Modulation of the Transforming Growth Factor-beta Signal Transduction Pathway by Hepatitis C Virus Nonstructural 5A Protein J. Biol. Chem., March 17, 2006; 281(11): 7468 - 7478. [Abstract] [Full Text] [PDF]

				S.-X. Wang, P. K. Elder, Y. Zheng, A. R. Strauch, and R. J. Kelm Jr. Cell Cycle-mediated Regulation of Smooth Muscle {alpha}-Actin Gene Transcription in Fibroblasts and Vascular Smooth Muscle Cells Involves Multiple Adenovirus E1A-interacting Cofactors J. Biol. Chem., February 18, 2005; 280(7): 6204 - 6214. [Abstract] [Full Text] [PDF]

				H. Sowa, H. Kaji, G. N. Hendy, L. Canaff, T. Komori, T. Sugimoto, and K. Chihara Menin Is Required for Bone Morphogenetic Protein 2- and Transforming Growth Factor {beta}-regulated Osteoblastic Differentiation through Interaction with Smads and Runx2 J. Biol. Chem., September 24, 2004; 279(39): 40267 - 40275. [Abstract] [Full Text] [PDF]

				B. Sauer, R. Vogler, H. von Wenckstern, M. Fujii, M. B. Anzano, A. B. Glick, M. Schafer-Korting, A. B. Roberts, and B. Kleuser Involvement of Smad Signaling in Sphingosine 1-Phosphate-mediated Biological Responses of Keratinocytes J. Biol. Chem., September 10, 2004; 279(37): 38471 - 38479. [Abstract] [Full Text] [PDF]

				K. Kahata, M. Hayashi, M. Asaka, U. Hellman, H. Kitagawa, J. Yanagisawa, S. Kato, T. Imamura, and K. Miyazono Regulation of transforming growth factor-{beta} and bone morphogenetic protein signalling by transcriptional coactivator GCN5 Genes Cells, February 1, 2004; 9(2): 143 - 151. [Abstract] [Full Text] [PDF]

				H.-J. Lee, J. K. Lee, S. Miyake, and S.-J. Kim A Novel E1A-like Inhibitor of Differentiation (EID) Family Member, EID-2, Suppresses Transforming Growth Factor (TGF)-{beta} Signaling by Blocking TGF-{beta}-induced Formation of Smad3-Smad4 Complexes J. Biol. Chem., January 23, 2004; 279(4): 2666 - 2672. [Abstract] [Full Text] [PDF]

				E. Ogawa, W. M. Elliott, F. Hughes, T. J. Eichholtz, J. C. Hogg, and S. Hayashi Latent adenoviral infection induces production of growth factors relevant to airway remodeling in COPD Am J Physiol Lung Cell Mol Physiol, January 1, 2004; 286(1): L189 - L197. [Abstract] [Full Text] [PDF]

				M. di Pietro, G. Marra, P. Cejka, L. Stojic, M. Menigatti, M. S. Cattaruzza, and J. Jiricny Mismatch Repair-Dependent Transcriptome Changes In Human Cells Treated with the Methylating Agent N-Methyl-N'-Nitro-N-Nitrosoguanidine Cancer Res., December 1, 2003; 63(23): 8158 - 8166. [Abstract] [Full Text] [PDF]

K. Higashi, Y. Inagaki, K. Fujimori, A. Nakao, H. Kaneko, and I. Nakatsuka Interferon-{gamma} Interferes with Transforming Growth Factor-{beta} Signaling through Direct Interaction of YB-1 with Smad3 J. Biol. Chem., October 31, 2003; 278(44): 43470 - 43479. [Abstract] [Full Text] [PDF]

<