Interaction with Corepressors

Three Smad corepressors have been identified: TGIF (Massague and Wotton 2000), Ski and the related SnoN (Massague and Wotton, 2000; Liu et al., 2001b; Frederick and Wang, 2002; Luo, 2004). In addition, PIASy can also act as a corepressor for Smads (Long et al, 2003).


TGIF is a member of the evolutionarily conserved three-amino-acid loop extension (TALE) family of atypical homeodomain proteins. TGIF exhibits several modes of repression. TGIF can repress transcription by competing with retinoid receptors for the DNA binding sites, the retinoid X receptor (RXR) responsive element, in the promoter regions of the regulated genes (Massague and Wotton, 2000). TGIF can also repress transcription through recruiting HDAC (Wotton et al, 1999a). TGIF can interact directly with the paired amphipathic a-helix 2 domain of the mSin3 corepressor (Wotton et al, 1999b; Wotton et al, 2001). TGIF can also recruit CtBP (Melhuish and Wotton, 2000).

TGIF is highly conserved in mammals. In contrast, Drosophila TGIF proteins, achintya and vismay, share homology with human TGIF only in the TALE homeodomain. The Drosophila TGIF proteins are transcriptional activators (Hyman et al, 2003). Obviously, regions outside of the TALE domain specify activation or repression of transcription.

TGIF interacts with the C-terminal domain of Smad2 and Smad3, and the interactions are increased in the presence of TGF-P (Wotton et al., 1999a). TGIF represses Smads-mediated transcriptional activation in part by interacting with histone deacetylases (HDACs) (Wotton et al, 1999a). In addition, TGIF recruits mSin3 to a TGF-P activated Smad complex to inhibit transcriptional responses (Wotton et al, 1999b; Wotton et al., 2001). Efficient repression of TGF-P transcriptional responses by TGIF is also dependent on the interaction with CtBP (Melhuish and Wotton, 2000). Thus, TGIF uses several modes to repress Smads-mediated transcription.

TGIF is a short-lived protein. Small changes in the physiological levels of TGIF can result in profound effects on human development, as shown by the devastating brain and craniofacial developmental defects in heterozygotes carrying a hypomorphic TGIF

mutant allele (Gripp et al, 2000). EGF signaling can lead to the phosphorylation of TGIF at two Erk MAP kinase sites. This stabilizes TGIF and increases Smad2 TGIF complex in response to TGF-p (Lo et al., 2001).

Recent studies have shown that TGIF knockout mice are viable and fertile (Shen and Walsh, 2005). In addition, there were no discernible derangements in all the major organs (Shen and Walsh, 2005). TGIF2 shows an expression pattern very similar to that of TGIF. TGIF2 and TGIF share 77% identity in the homeodomain and 49% similarity outside of the homeodomain. It is possible that TGIF and TGIF2 have redundant or at least overlapping roles, and that TGIF2 can compansate for the loss of TGIF. Germline knockout or conditional knockout both TGIF and TGIF2 will help to identify genes that are repressed by TGIF under natural setting.


Ski was first identified as a viral oncogene (v-Ski) from the Sloan-Kettering avian retrovirus that transforms chicken embryonic fibroblasts (Liu et al., 2001b; Luo, 2004). SnoN is a member of the Ski proto-oncogene family. In addition to being an oncogene, SnoN appears to act as a tumor suppressor, at least in certain cells (Liu et al., 2001b; Luo, 2004). Ski and SnoN directly binds to the N-CoR and mSin3A that form a complex with HDAC (Luo et al., 1999; Nomura et al., 1999). In addition, Ski has been shown to be able to bind directly to the corepressors HIPK2 and MeCP2 (Kokura et al., 2001; Harada et al., 2003). Ski is also required for transcriptional repression by several other proteins, including the Mad, the thyroid hormone receptor-P, Rb protein and the Gli3 repressor (Liu et al., 2001b; Luo, 2004). Thus, Ski appears to be an integral part of the transcriptional repression machinery.

Ski and SnoN interact with Smad2 and Smad3 in a TGF-P dependent manner, whereas Ski and SnoN associate with Smad4 constitutively (Akiyoshi et al., 1999; Luo et al., 1999; Stroschein et al., 1999b; Sun et al., 1999a; Sun et al., 1999b; Xu et al., 2000). The MH2 domains of Smad2, 3, 4 are essential for these interactions (Akiyoshi et al., 1999; Luo et al., 1999; Stroschein et al., 1999b; Sun et al., 1999a; Sun et al., 1999b; Xu et al., 2000), and Ski has been shown to recognize trimeric Smad3 (Moustakas and Heldin, 2002; Qin et al., 2002). Ski and SnoN can inhibit TGF-P transcriptional responses through multiple mechanims. Ski and SnoN can recruit HDAC to TGF-P activated Smad complexes through direct interaction with N-CoR and mSin3A (Luo et al., 1999; Liu et al., 2001b; Luo, 2004). Ski also competes with the coactivator p300/CBP for binding to the activated Smad3 (Akiyoshi et al., 1999). In addition, Ski can inhibit Smad2 and Smad3 binding to the L3 loop of the Smad4 MH2 domain (Frederick and Wang, 2002; Wu et al., 2002). Ski also stabilizes inactive Smad complexes on SBE (Suzuki et al., 2004). Finally, Ski has been reported to inhibit TGF-P signaling through inhibition of Smad2 phosphorylation by the TGF-P receptor (Prunier et al., 2003). The transforming activity of Ski and SnoN is dependent on their interaction with Smads (He et al., 2003). In addition to inhibition of the TGF-P pathway, Ski but not SnoN, also associates with BMP specific Smad complex in a ligand-dependent manner and blocks BMP transcriptional responses (Luo, 2004).

SnoN, and to a lesser extent, Ski, are degraded upon TGF-P treatment (Stroschein et al., 1999b; Sun et al., 1999b; Liu et al, 2001b; Luo, 2004). Therefore, SnoN and Ski have been proposed as nuclear corepressors for Smad4 to maintain TGF-P responsive genes in a repressed state in the absence of ligand. TGF-p also induces the expression of SnoN, which likely to function in a negative feedback control to turn off TGF-P signaling at late stages (Stroschein et al, 1999b; Liu et al, 2001b; Luo, 2004).

Studies on the natural Smad7 promoter have provided evidence that Ski is indeed a corepressor for Smad4 at basal state. As described in the DNA binding section, the Smad7 promoter contains the perfect 8 bp palindromic SBE. Smad4 binds to this SBE. Ski is recruited to the Smad7 promoter through interaction with Smad4 and represses Smad7 expression at basal state (Denissova and Liu, 2004).


PIASy also inhibits TGF-P-inducible transcriptional responses (Imoto et al, 2003; Long et al, 2003). It interacts most strongly with Smad3 and also associates with other receptor-regulated Smads and Smad4. Smad3, Smad4 and PIASy can form a complex. PIASy can associate with HDAC1, and the inhibitory effect of PIASy can be disrupted by treatment with TSA, a HDAC inhibitor (Long et al, 2003). Thus, PIASy can inhibit TGF-P-mediated transcription by recruiting HDAC to Smads. Similarly, PIASxP can also interact with HDAC3 (Tussie-Luna et al, 2002). PIASxa and PIASxP have also been implicated to recruit HDAC to inhibit IL12-mediated and STAT4-dependent gene activation (Arora et al, 2003). Thus, some PIAS proteins can act as corepressors through association with HDAC.

PIAS family members have SUMO E3 ligase activities (Schmidt and Muller, 2003). HDAC1 itself may be a target for sumoylation by PIAS proteins. A

study has demonstrated that HDAC1 is sumoylated both in vivo and in vitro (David et al., 2002). Moreover, mutation of the sumoylation sites in HDAC1 reduces its transcriptional repression activity in reporter gene assays (David et al., 2002). Further studies are necessary to fully explore this interesting possibility.

Related to the issue of sumoylation targets of PIASy, it is worth pointing out that Smad4 is modified by SUMO and sumoylation represses its transcriptional activity (Long et al., 2004a). This occurs through Daxx (Chang et al., 2005), which regulates apoptosis and represses transcription through its interaction with various cytoplasmic and nuclear proteins. Daxx interacts with Smad4 and represses its transcriptional activity. Binding of Daxx to Smad4 is dependent on Smad4 sumoylation. Mutation of the major Smad4 sumoylation site not only disrupted Smad4-Daxx interaction but also relieved Daxx-mediated repression of Smad4 transcriptional activity. Thus, Daxx represses Smad4-mediated transcriptional activity by direct interaction with the sumoylated Smad4 (Chang et al., 2005). Interestingly, sumoylation of Smad4 also increases its stability and nuclear accumulation (Lee et al., 2003; Lin et al., 2003a; Lin et al., 2003b). Thus, the net effect of sumoylation of Smad4 can be either stimulatory or inhibitory, depending on the target promoter that is analyzed.

B4: Why Multiple Corepressors?

With the presence of TGIF, Ski/SnoN and PIASy, a natural question is whether each one can inhibit all TGF-(3-responisve genes. Several studies indicate that TGIF, Ski/SnoN and PIASy have distinct promoter specificities. For example, TGF-P induction of pl5 is greatly inhibited by overexpression of PIASy (Long et al., 2003), partially inhibited by overexpression of TGIF (Lo et al., 2001), but not inhibited by overexpression of Ski (Sun et al., 1999a). Conversely, the early responsiveness of JunB to TGF-P is only moderately inhibited by overexpression of PIASy (Long et al., 2003) but is significantly inhibited by overexpression of Ski (Luo et al., 1999). One possibility is that TGIF, Ski/SnoN and PIASy are present in different corepressor complexes that inhibit distinct sets of promoters. Both TGIF and PIASy may inhibit TGF-P activated Smad complex, whereas Ski/SnoN may function as corepressors for Smad4 to maintain TGF-P responsive genes in a repressed state in the absence of TGF-P and contribute to TGF-P signal termination. The presence of multiple regulators that share overlapping functions in the same cells confers ample flexibility under complex circumstances. In addition, TGIF, Ski/SnoN and PIASy are differentially expressed in various tissues, which may confer cell-type specific functions.


TGF-P regulates a wide variety of biological activities through transcriptional regulation of distinct target genes. Extensive research on Smads in the past nine years has provided many fundamental insights into the basic mechanisms of Smads-mediated transcriptional control as well as the correlation with developmetal status or disease conditions. Still, many interesting questions remain to be answered. For instrance, Smads have been shown to recruit two classes of coactivators: histone acetyltransferases and the ARC complex. The role of SWI/SNF complex, a distinct class of coactivator that contains ATP-dependent DNA unwinding activities, remains to be addressed for Smads-mediated transcriptional activation. Post-translational modifications, such as phosphorylation and sumoylation, regulate Smad DNA binding and transcriptional activities. Do other types of modification, such as methylation, affect Smad activity? Genome-wide gene expression profiling coupled with conditional knockout, knockin and RNAi approaches has emerged, and will continue to be extremely useful in delineating cell specific TGF-P responsive genes with disease states. The findings may lead to the development of effective drugs for therapeutic applications.


I apologize to those colleagues whose work were not cited due to space constrain. I thank Y. Shi, Y. E. Zhang, and many colleagues for stimulating discussions. I also thank the American Association for Cancer Research-National Foundation for Cancer Research, the Burroughs Wellcome Fund, the Sidney Kimmel Foundation for Cancer Research, the Pharmaceutical Research and Manufacturer of America Foundation, the Emerald Foundation, the New Jersey Commission on Cancer Research, the Department of Defense Breast Cancer Research Program, and the National Institutes of Health for support of our research.

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook

Post a comment