Snail promotes Wnt target gene expression and interacts with b-catenin

Oncogene (2008) 27, 5075–5080

& 2008 Macmillan Publishers Limited All rights reserved 0950-9232/08 $30.00 www.nature.com/onc

SHORT COMMUNICATION

Snail promotes Wnt target gene expression and interacts with b-catenin V Stemmer1, B de Craene2, G Berx2 and J Behrens1 1

Nikolaus-Fiebiger-Center for Molecular Medicine, University Erlangen-Nu¨rnberg, Erlangen, Germany and 2Department for Molecular Biomedical Research (DMBR), Molecular and Cellular Oncology, VIB-Ghent University, Ghent, Zwijnaarde, Belgium

The transcription factor snail represses epithelial gene expression and thereby promotes epithelial-mesenchymal transitions (EMT) and tumor invasion. The Wnt/ b-catenin pathway is also involved in EMT and was shown to activate snail. Here, we demonstrate that snail increases Wnt reporter gene activity induced by b-catenin, LRP6 or dishevelled, and also promotes transcription activated by GAL4-b-catenin fusion proteins. Snail mutants lacking the transcriptional repressor domain also stimulate b-catenin-dependent transcription indicating that downregulation of snail target genes is not required for this activity. Snail interacts with b-catenin in immunoprecipitation experiments at its N-terminus, which is required for activation by snail. In colorectal cancer cell lines, overexpression of snail leads to increased expression of Wnt target genes, whereas downregulation of endogenous snail by siRNA reduces target gene expression. Our data indicate a positive feedback stimulation of the Wnt pathway by activation of snail. Oncogene (2008) 27, 5075–5080; doi:10.1038/onc.2008.140; published online 12 May 2008 Keywords: b-catenin; colorectal cancer; mesenchymal transition; snail; Wnt

epithelial-

Snail is a transcription factor, which binds to E-box motives in target gene promoters through a carboxyterminal zinc-finger domain and represses transcription by recruiting histone deacetylase activity through its amino-terminal SNAG (Snail/Gfi-1) domain (Batlle et al., 2000; Peinado et al., 2004). Snail represses various epithelial marker genes including E-cadherin and claudins and thereby causes epithelial-mesenchymal transition (EMT), a process by which epithelial cells lose their cell contacts and are converted to a mesenchymal phenotype with migratory and invasive properties (Batlle et al., 2000; Cano et al., 2000; De Correspondence: Dr J Behrens, Nikolaus-Fiebiger-Center for Molecular Medicine, University Erlangen-Nu¨rnberg, Glueckstr. 6, Erlangen D-91054, Germany. E-mail: [email protected] Received 23 April 2007; revised 5 March 2008; accepted 2 April 2008; published online 12 May 2008

Craene et al., 2005). In addition to its role in EMT, snail has been implicated in the regulation of cell proliferation and survival (Barrallo-Gimeno and Nieto, 2005). The Wnt signaling pathway regulates target gene expression by stabilizing the cytoplasmic component b-catenin, which enters the nucleus and forms complexes with T-cell factor (TCF) transcription factors bound to specific promoter sequences. b-catenin has strong transcriptional activation domains in its N- and Cterminus (Hsu et al., 1998) and thereby induces transcription of a variety of target genes including regulators of cell survival, proliferation and metastasis formation. In the absence of Wnt signals, b-catenin is phosphorylated by GSK3b (glycogen synthase kinase 3b) in a multiprotein degradation complex leading to its ubiquitination and degradation in the proteasome. In the presence of Wnt signals or after mutations of the degradation complex components APC (adenomatous polyposis coli), axin/conductin or b-catenin, phosphorylation and degradation of b-catenin is blocked. Mutations of Wnt signaling components are the major cause of colorectal cancer and other tumor types (reviewed in (Behrens, 2005). Wnt signaling can also induce EMT and is shown to activate the transcription, protein stability as well as nuclear localization of snail through the inhibition of GSK3b (Zhou et al., 2004; Bachelder et al., 2005; Yook et al., 2006). Here, we demonstrate that snail can functionally interact with b-catenin to increase Wnt-dependent target gene expression. This activation is independent of snail’s transcriptional repressor activity, and in particular does not involve repression of the E-cadherin promoter. Our results establish a novel and unexpected positive regulatory function of the transcriptional repressor snail.

Snail stimulates TCF/b-catenin-dependent transcription To analyze whether snail affects TCF/b-catenin-dependent transcription, we used the TOP/FOP reporter system, which consists of a minimal promoter linked to either wild type (TOP reporter) or mutated TCFbinding sites (FOP reporter) driving expression of luciferase. In HEK293T cells, snail potently increased the TOP/FOP ratio induced by cotransfected b-catenin, whereas snail alone had no effect (Figure 1a). Snail specifically increased the TOP values and had no effect on the FOP reporter controls. Snail also increased TOP

Snail promotes Wnt signaling V Stemmer et al

5076

Figure 1 Transcriptional activation of TCF/b-catenin-dependent reporter expression (TOP/FOP ratio) by snail. Reporter assays were performed in HEK293T cells (a–c) or SW480 cells (d). See Supplementary information for DNA amounts and experimental procedures. Snail increases TOP/FOP ratio after activation by transfected b-catenin (a), dominant-active LRP6 (LRP6da) (b), dishevelled2 (c) or by endogenous b-catenin in SW480 colorectal carcinoma cells (d). TCF, T-cell factor.

reporter activity after upstream activation of the Wnt pathway by transfection of dominant-active LRP6 (lowdensity lipoprotein receptor-related protein) (Figure 1b), or dishevelled2 (Figure 1c), as well as the constitutive reporter activity in the SW480 and HCT116 colorectal carcinoma cell line in which b-catenin is activated due to mutations in APC and b-catenin, respectively (Figure 1d, Supplementary Figure 1A). Snail did not alter reporter activities of pGL2control and of an HIF1a (hypoxia-inducible factor 1a)-dependent reporter construct, which were used as negative controls (Supplementary Figure 1B). It is possible that snail exerts its effects on TOP/FOP reporters by downregulation of E-cadherin, which could lead to the release of b-catenin and increase of b-catenin-dependent transcription. To analyze this possibility, we determined the activity of a snail deletion mutant (snailDSNAG, Figure 2a), which lacks the N-terminal repressor domain and was previously shown to be unable to repress E-cadherin (Peinado et al., 2004). SnailDSNAG strongly increased b-catenin-dependent activation of the TOP/FOP ratio (Figure 2b) similar to full-size snail, although it failed to repress the E-cadherin promoter (Supplementary Figure 2A). Moreover, the carboxy-terminal zinc-finger domain of snail (snail152-264) sufficed for activation of b-catenindependent transcription, whereas the amino-terminal half of snail (snail1-151) had no effect (Figure 2c). Interestingly, a deletion mutant of snail lacking zinc fingers 3–4 (SnailDZF3-4) no longer activated but rather suppressed b-catenin-dependent transcription Oncogene

demonstrating that the zinc-finger domain is important for transcriptional activation by snail (Figure 2d). The data indicate that the activity of snail on b-catenin-dependent transcription does not require its repressor activity, and in particular does not involve the transcriptional downregulation of E-cadherin. As both snail and b-catenin are targets of GSK3binduced degradation, snail might stabilize b-catenin by substrate competition for GSK3b. However, snail did not increase the amounts of cotransfected exogenous b-catenin (Figure 2e) or of endogenous b-catenin (Figure 2f) stabilized by upstream activation of Wnt signaling. Moreover, snail did not alter phosphorylation of b-catenin at the critical serines S33 and S37 (Figure 2g). Gross alterations in b-catenin nuclear localization were not detected by immunofluorescence analysis (Supplementary Figure 2D). In addition, snail also increased the transcriptional activation by S33Y-bcatenin, which is a point mutant of b-catenin resistant to degradation by GSK3b (Figure 2h). The data show that snail does not affect the stability, phosphorylation or nuclear localization of b-catenin.

Snail interacts with b-catenin and stimulates its transcriptional activity independent of TCFs To determine whether the effects of snail on b-catenin are based on the interaction of both components, we analyzed whether snail associates with b-catenin in


5077

Figure 2 Transcriptional activation by snail is independent of E-cadherin repression and b-catenin stabilization. All experiments except of (g) were performed in HEK293 T cells. (a) Schematic structure of snail and the used deletion mutants. The N-terminal SNAG domain (transcriptional repression), a central serine/threonine-rich region (S/T) and the C-terminal zinc-finger domain (DNA binding) are marked. Flag-tagged snail constructs were obtained by PCR on the full-length human snail cDNA and subsequent cloning into pcDNA3.1 containing the flag peptide coding sequence. (b) SnailDSNAG increases TOP/FOP ratio in the presence of b-catenin. (c) The zinc-finger domain (snail152-264), but not the N-terminal half (snail1-151) of snail increases TOP/FOP ratio in the presence of b-catenin. (d) SnailDZF3-4 represses TOP/FOP ratio in the presence of b-catenin. This mutant did not repress the E-cadherin promoter and unrelated control promoters (Supplementary Figure 2B and C). Snail, snail1-151, snail152-264 and snailDZF3-4 were expressed at similar levels, whereas snailDSNAG showed twofold higher expression. (e–g) Western blot analysis of hypotonic cell extracts with the indicated antibodies. Snail does not increase the stability of cotransfected YFP-tagged b-catenin (YFP-b-catenin) (e) or of endogenous b-catenin after upstream activation by dominant active LRP6 (f) and does not alter the amounts or phosphorylation of endogenous b-catenin in SW480 cells (g). (h) Snail activates transcription by degradation-resistant S33Y-b-catenin. See Supplementary information for experimental procedures and antibody information.

immunoprecipitation experiments. Transiently expressed snail coimmunoprecipitated with endogenous b-catenin from lysates of SW480 cells (Figure 3a). Importantly, endogenous complexes of both proteins were also revealed by immunoprecipitation experiments using anti-snail or anti-b-catenin antibodies (Figure 3b).

The zinc-finger domain (snail152-264), the aminoterminal half of snail (snail1-151) and the snailDZF3-4 mutant coimmunoprecipitated with b-catenin after transient coexpression of the proteins in HEK293 T cells (Supplementary Figure 3A and B). Thus, despite the fact that snail1-151 did not affect Oncogene


5078

Figure 3 Functional interaction of snail with b-catenin. (a) Coimmunoprecipitation of flag-tagged snail (flag snail) with endogenous b-catenin in SW480 cells after immunoprecipitation with anti-b-catenin antibodies. (b) Coimmunoprecipitation of b-catenin with snail and of snail with b-catenin after immunoprecipitation of endogenous proteins with the indicated antibodies from lysates of SW480 cells. (c) YFP-tagged full-size b-catenin (YFP-bcatenin), but not YFP-tagged b-cateninDNT (YFP-b-cateninDNT) coimmunoprecipitates with flag-tagged snail after immunoprecipitation with anti-flag antibodies in HEK293T cells. (d) Snail activates GAL4-b-catenin, but not GAL4-b-cateninDNTdependent transcription of a GAL4-binding sites containing promoter reporter construct in HEK293T cells. For conditions of coimmunoprecipitations and antibodies see Supplementary methods.

b-catenin-dependent transcription (cf. Figure 1c), the fragment retained its ability to immunoprecipitate b-catenin. The snailDZF3-4 mutant, which repressed Oncogene

b-catenin-dependent transcription (cf. Figure 2d) might act in a dominant-negative manner, because it retained b-catenin binding but has lost a large part of the zinc-finger domain, which is required for stimulation of b-catenin’s transcriptional activity. A mutant of b-catenin lacking the first 89 amino acids (YFP-b-cateninDNT) failed to coimmunoprecipitate with snail, indicating that this part of b-catenin is required for interaction with snail (Figure 3c). In contrast, deletions of the C-terminus and within the central armadillo repeat domain of b-catenin had no effect on the snail/b-catenin association (Supplementary Figure 3B). We next determined whether the N-terminal 89 amino acids of b-catenin are required to allow transcriptional stimulation by snail. To avoid any interference by endogenous b-catenin, we analyzed whether snail would alter the transcriptional activation generated by fusion proteins of b-catenin or b-cateninDNT to a GAL4 DNA-binding domain. While activation of a GAL4-dependent reporter by GAL4-bcatenin was further increased after snail transfection, activation by GAL4-b-cateninDNT was not altered (Figure 3d). GAL4-b-Catenin-dependent reporter activity was also increased by snailDSNAG again demonstrating that the activation is independent of E-cadherin repression (Supplementary Figure 4). These data show that the N-terminus of b-catenin, which is required for complex formation with snail is also required for transcriptional stimulation by snail. Although the Nterminal domain of b-catenin is important for the control of b-catenin stability as well as nuclear localization (Brembeck et al., 2004; Krieghoff et al., 2006), snail did not notably increase b-catenin protein levels or promote its nuclear residence as shown above (Figures 2e, f and g and Supplementary Figure 2D). The fact that snail can activate the GAL4-b-catenin fusion proteins also shows that it acts independent of TCF transcription factors. Furthermore, snail probably does not need to bind to DNA to stimulate b-catenin-dependent transcription, as the tested reporters (TOP/FOP, GAL4 and Axin2; see below) do not contain obvious snail-binding sites. Altogether, a functional interaction between snail and b-catenin is a plausible scenario to explain our findings, although the molecular mechanism remains to be determined. Snail might change the transcriptional activation potential of b-catenin either by preventing the interaction of b-catenin with transcriptional repressors or promoting the recruitment of transcriptional activators.

Snail is required for strong Wnt target gene expression in colorectal tumor cells To determine the functional consequences of snail for Wnt signalling, we analyzed its effects on the expression of known Wnt target genes. The mRNA and protein levels of Axin2/conductin (Lustig et al., 2002; Jho et al., 2002), a specific target gene of Wnt signaling were increased in DLD1 colorectal tumor cells expressing


5079

Figure 4 Snail is required for Wnt target gene expression. (a) RT–PCR shows elevated Axin2 mRNA levels after induction of snail expression in DLD1 cells expressing doxycycline (TET)-inducible snail. (b) Transient expression of flag-snail or flag-snailDSNAG increases Axin2 or survivin protein levels in colorectal carcinoma cells as determined by western blotting. (c) Snail activates an Axin2 promoter reporter construct in SW480 cells. (d) siRNA-mediated knockdown of snail in SW480 cells decreases mRNA levels of the Wnt target genes Axin2, cyclinD1, survivin, TCF1 and c-Myc, as determined by quantitative real-time RT–PCR. (e) siRNA-mediated knockdown of snail in SW480 cells decreases Axin2 and survivin protein amounts, while b-catenin and E-cadherin levels remain unaltered, as determined by western blot. (f) Reduction of Axin2 promoter activity in SW480 cells after siRNA-mediated knockdown as indicated. Results in (f) represent the mean of three independent experiments. Significance (*) was calculated by the student’s T-test. See Supplementary information for sequences of PCR primers and siRNAs. RT–PCR, reverse transcriptase PCR.

Tet-inducible snail as well as in SW480 and HCT116 cells after transient expression of snail (Figures 4a and b). Survivin, another Wnt target gene was also unregulated in response to snail (Figure 4b). Moreover, an Axin2 promoter reporter construct was strongly stimulated by transient expression of snail in SW480 cells (Figure 4c). To analyze whether endogenous snail is important for Wnt target gene expression, we silenced its expression using siRNA oligonucleotides. Knockdown of snail in SW480 cells led to the reduced expression of Axin2 and of the other Wnt target genes cyclinD1, survivin, TCF1 and c-Myc, as determined by

quantitative reverse transcriptase PCR (Figure 4d). Protein amounts of Axin2 and survivin were also reduced by snail knockdown (Figure 4e). Notably, E-cadherin and b-catenin levels were unchanged upon silencing of snail in these cells, again indicating that suppression of E-cadherin by snail is not required for the stimulation of b-catenin-dependent transcription by snail (Figure 4e and Supplementary Figure 5A). Axin2 promoter activity was also significantly reduced by siRNA against snail (Figure 4f). Altogether, the data suggest that endogenous snail coactivates Wnt target gene expression. Oncogene


5080

It was recently shown that the upregulation of Axin2 by Wnt signaling increases snail levels by sequestering GSK3b leading to EMT (Yook et al., 2006). Conversely, our data suggest that snail can activate Wnt signaling by binding to b-catenin thereby establishing a positive feedback loop for Wnt-dependent transcription. In addition, through such a mechanism, snail might stimulate its own activity because Axin2 levels are increased which would further stabilize snail. Thus, Wnt signaling and snail-dependent induction of EMT are interconnected by multiple positive cross talks, possibly adding to the robustness of both signaling systems. There is evidence for a close relationship between both pathways in vivo. Both snail and TCF/b-catenin signaling can induce EMT in various cellular systems (Huber et al., 2005), and snail and Wnt activities are oscillating during segmental patterning of vertebrate somites (Dale et al., 2006). In the MIN (multiple intestinal neoplasia) mouse, a model, which spontaneously develops intestinal adenomas as a result of a germline truncation in APC, knockdown of snail increases cell death and

decreases incidence and proliferation rates of the intestinal tumors (Roy et al., 2004). The snail/b-catenin interplay could be of particular importance during tumor invasion and metastasis. In human colorectal tumors, EMT occurs at the invasive front where tumor cells dissociate from the tumor mass and infiltrate the mesenchyme. b-catenin nuclear activity and expression of target genes is markedly increased in these areas, while E-cadherin expression is downregulated (Brabletz et al., 2001). From our data, we can speculate that activation of snail in such lesions would act to downregulate E-cadherin and independent of that further stimulate Wnt pathway activity, both of which would increase the invasive capacity of the tumor cells. Acknowledgements We thank F Costantini for providing the Axin2 promoter reporter construct, E Krieghoff for LRP6da and b-catenin constructs and M van de Wetering for TOP and FOPglow constructs.

References Bachelder RE, Yoon SO, Franci C, de Herreros AG, Mercurio AM. (2005). Glycogen synthase kinase-3 is an endogenous inhibitor of Snail transcription: implications for the epithelial-mesenchymal transition. J Cell Biol 168: 29–33. Barrallo-Gimeno A, Nieto MA. (2005). The Snail genes as inducers of cell movement and survival: implications in development and cancer. Development 132: 3151–3161. Batlle E, Sancho E, Franci C, Dominguez D, Monfar M, Baulida J et al. (2000). The transcription factor snail is a repressor of E-cadherin gene expression in epithelial tumour cells. Nat Cell Biol 2: 84–89. Behrens J. (2005). The role of the Wnt signalling pathway in colorectal tumorigenesis. Biochem Soc Trans 33: 672–675. Brabletz T, Jung A, Reu S, Porzner M, Hlubek F, Kunz-Schughart LA et al. (2001). Variable beta-catenin expression in colorectal cancers indicates tumor progression driven by the tumor environment. Proc Natl Acad Sci USA 98: 10356–10361. Brembeck FH, Schwarz-Romond T, Bakkers J, Wilhelm S, Hammerschmidt M, Birchmeier W. (2004). Essential role of BCL9-2 in the switch between beta-catenin’s adhesive and transcriptional functions. Genes Dev 18: 2225–2230. Cano A, Perez-Moreno MA, Rodrigo I, Locascio A, Blanco MJ, del Barrio MG et al. (2000). The transcription factor snail controls epithelial-mesenchymal transitions by repressing E-cadherin expression. Nat Cell Biol 2: 76–83. Dale JK, Malapert P, Chal J, Vilhais-Neto G, Maroto M, Johnson T et al. (2006). Oscillations of the snail genes in the presomitic mesoderm coordinate segmental patterning and morphogenesis in vertebrate somitogenesis. Dev Cell 10: 355–366. De Craene B, Gilbert B, Stove C, Bruyneel E, van Roy F, Berx G. (2005). The transcription factor snail induces tumor cell invasion through modulation of the epithelial cell differentiation program. Cancer Res 65: 6237–6244.

Hsu SC, Galceran J, Grosschedl R. (1998). Modulation of transcriptional regulation by LEF-1 in response to Wnt-1 signaling and association with beta-catenin. Mol Cell Biol 18: 4807–4818. Huber MA, Kraut N, Beug H. (2005). Molecular requirements for epithelial-mesenchymal transition during tumor progression. Curr Opin Cell Biol 17: 548–558. Jho EH, Zhang T, Domon C, Joo CK, Freund JN, Costantini F. (2002). Wnt/beta-catenin/Tcf signaling induces the transcription of Axin2, a negative regulator of the signaling pathway. Mol Cell Biol 22: 1172–1183. Krieghoff E, Behrens J, Mayr B. (2006). Nucleo-cytoplasmic distribution of beta-catenin is regulated by retention. J Cell Sci 119: 1453–1463. Lustig B, Jerchow B, Sachs M, Weiler S, Pietsch T, Karsten U et al. (2002). Negative feedback loop of Wnt signaling through upregulation of conductin/axin2 in colorectal and liver tumors. Mol Cell Biol 22: 1184–1193. Peinado H, Ballestar E, Esteller M, Cano A. (2004). Snail mediates E-cadherin repression by the recruitment of the Sin3A/histone deacetylase 1 (HDAC1)/HDAC2 complex. Mol Cell Biol 24: 306–319. Roy HK, Iversen P, Hart J, Liu Y, Koetsier JL, Kim Y et al. (2004). Down-regulation of SNAIL suppresses MIN mouse tumorigenesis: modulation of apoptosis, proliferation, and fractal dimension. Mol Cancer Ther 3: 1159–1165. Yook JI, Li XY, Ota I, Hu C, Kim HS, Kim NH et al. (2006). A WntAxin2-GSK3beta cascade regulates Snail1 activity in breast cancer cells. Nat Cell Biol 8: 1398–1406. Zhou BP, Deng J, Xia W, Xu J, Li YM, Gunduz M et al. (2004). Dual regulation of Snail by GSK-3beta-mediated phosphorylation in control of epithelial-mesenchymal transition. Nat Cell Biol 6: 931–940.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

Oncogene