Biochimica Et Biophysica Acta 1813 (2011) 713–722

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Biochimica et Biophysica Acta 1813 (2011) 713–722

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

journal homepage: www.elsevier.com/locate/bbamcr

Ataxin-1 occupies the promoter region of E-cadherin in vivo and activates CtBP2-repressed promoter

Soyeon Lee a, Sunghoi Hong a,b, Sungsu Kim a,c, Seongman Kang a,⁎

a School of Life Sciences and Biotechnology, Korea University, 1, 5ka, Anam-dong, Sungbuk-gu, Seoul 136-701, Republic of Korea b Department of Biomedical Science, College of Health Sciences, Korea University, Seoul, Republic of Korea c Department of Developmental Biology, Washington University, St. Louis, MO 63110, USA

article info abstract

Article history: Ataxin-1 is a polyglutamine protein of unknown function that is encoded by the ATXN1 gene in humans. Received 13 July 2010 To gain insight into the function of ataxin-1, we sought to identify proteins that interact with ataxin-1 Received in revised form 25 January 2011 through yeast two-hybrid screening. In this study, transcriptional corepressor CtBP2 was identiﬁed as a Accepted 27 January 2011 protein that interacted with ataxin-1. CtBP2 and ataxin-1 colocalized in the nucleus of mammalian cells. Available online 23 February 2011 Since the E-cadherin promoter is a target of CtBP-mediated repression, the relationship between ataxin-1 and the E-cadherin promoter was investigated. Chromatin immunoprecipitation assays showed that CtBP2 Keywords: Ataxin-1 and ataxin-1 were recruited to the E-cadherin promoter in mammalian cells. Luciferase assays using E- CtBP2 cadherin promoter reporter constructs revealed that the luciferase activity was enhanced as the level of E-cadherin ataxin-1 protein expression increased. CtBP2 overexpression decreased E-cadherin expression, but Repression expression of ataxin-1 inversely increased the mRNA and protein levels of endogenous E-cadherin. Activation Interestingly, siRNA experiments showed that the transcriptional activation of ataxin-1 was associated with the presence of CtBP2. This study demonstrates that ataxin-1 occupies the promoter region of E- cadherin in vivo and that ataxin-1 activates the promoter in a CtBP2-mediated transcriptional regulation manner. © 2011 Elsevier B.V. All rights reserved.

1. Introduction subsequently been found in complexes with several known DNA- binding transcription factors that participate in a wide variety of Ataxin-1 is a foci-forming polyglutamine protein of unknown developmental and adult biological pathways and processes, including function, whose mutant form causes spinocerebellar ataxia type 1 Wnt and BMP/TGFβ signaling [8–10], GATA factor activity, cell–cell (SCA1), an autosomal-dominant neurodegenerative disorder caused adhesion, myogenesis, and vascularization [11,12]. CtBP proteins by the expansion of the polyglutamine tract within ataxin-1 [1]. It was mediate their transcriptional function through interaction with various reported that ataxin-1 interacts with several transcriptional co- DNA-binding repressors with PLDLS-like motifs and chromatin-modi- regulators, including polyQ-binding protein (PQBP1) [2], leucine- fying enzymes, such as class 1 histone deacetylases (HDAC) without rich acidic nuclear protein (LANP) [3,4], silencing mediator of such motifs. Although the detailed molecular function of CtBP recruited retinoid/thyroid hormone receptors (SMRT) [5], and Sp1 [6]. These to DNA by transcriptional factors is unknown, it is quite clear that CtBP findings support the idea that ataxin-1 is related to the regulation of acts as a transcriptional corepressor. Recent work through a combina- gene expression in the nucleus. To gain further insight into the tion of gene trapping and gene targeting in embryonic stem cells showed function of ataxin-1 as a transcriptional regulator, we sought to that Ctbp1 and Ctbp2 are necessary for the regulation of gene expression identify proteins that interact with ataxin-1, particularly those that in multiple developmental programs during mouse embryogenesis [13]. are involved in transcriptional regulation. C-terminal binding protein In addition, CtBPs promote epithelial–mesenchymal transition and 2 (CtBP2), a transcriptional corepressor, was identified as a protein mediate repression of several tumor suppressor genes. Previous studies interacting with ataxin-1 by a yeast two-hybrid screening. demonstrated that CtBPs are able to inhibit E-cadherin gene expression Members of the C-terminal binding protein (CtBP) family function as by targeting to the promoter via several repressors such as ZEB, and transcriptional corepressors. CtBP was originally identified based on its recruiting chromatin remodeling complexes [14–16]. ability to bind the C-terminus of the E1A oncoprotein [7].Ithas As described above, the CtBP family proteins are important molecules as modulators of several essential cellular processes. Interactions with specific modulators in the CtBP transcriptional ⁎ Corresponding author. Tel.: +82 2 3290 3448; fax: +82 2 927 9028. regulatory complex may alter the CtBP-mediated gene regulation E-mail address: [email protected] (S. Kang). function, and especially abnormal interactions may cause diseases due

0167-4889/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.bbamcr.2011.01.035 714 S. Lee et al. / Biochimica et Biophysica Acta 1813 (2011) 713–722 to the transcriptional dysregulation [17].Therefore,thisstudywasto was transferred to induction medium supplemented with galactose investigate whether the presence of ataxin-1 interacting with CtBP2 has and raffinose instead of glucose. At 5 h after induction, the optical effects on the CtBP2-mediated repression of the E-cadherin gene as one density of each culture was measured at 600 nm, and the quantitative of target genes repressed by CtBP2. The evidence linking ataxin-1 to the β-galactosidase assay was performed with o-nitrophenyl-β-D-galac- regulation of CtBP2-mediated transcription was examined. topyranoside (ONPG). β-Galactosidase activity was measured at 420 nm using a multi-well plate reader (Bio-Rad). Each assay was 2. Materials and methods performed in triplicate with separate colonies in two independent experiments. 2.1. Plasmid constructs 2.4. Cell culture and transfection 2.1.1. Ataxin-1 constructs Human full-length ataxin-1 constructs, pcDNAI amp/ataxin-1 HeLa and HEK293T cells were maintained in DMEM supplemented (30Q), and pcDNAI amp/ataxin-1(82Q) were generously provided with 10% fetal bovine serum, streptomycin (100 μg/ml), and penicillin by Dr. H.T. Orr (Institute of Human Genetics, University of Minnesota). (100 U/ml). MCF-7 cells were cultured in RPMI supplemented with Full-length and various deleted versions of ataxin-1 cDNAs were 10% fetal bovine serum, streptomycin (100 μg/ml), and penicillin polymerase chain reaction (PCR) amplified from pcDNAI amp/ataxin- (100 U/ml). Transfections were performed with Lipofectamine (Invi- 1 using Pfu polymerase (Stratagene). The PCR-amplified products trogen) or polyethyleneimine (Sigma-Aldrich) as previously de- were cloned into EcoRI and XhoI sites of pLexA (Clontech) or scribed [19]. Mixtures combined at the DNA to reagent ratio of 1:3 pcDNA3.1 HisC (Invitrogen), and EcoRI and XbaI sites of pCMV-3x were incubated in serum-free OPTI-MEM for 15 min. After incubation, FLAG (Sigma). DNA–reagent complexes were applied to the cells. Hela cells were used to study intranuclear IFA, and MCF cells expressing a lot of E- 2.1.2. CtBP2 constructs cadherin were employed for Western blot experiments. Human full-length CtBP2 constructs, pGAD10/CtBP2, were generously provided by Dr. A.P. Otte (E. C. Slater Institute, BioCentrum 2.5. Co-immunoprecipitation experiments Amsterdam, University of Amsterdam). The PCR-amplified full-length or truncated CtBP2 cDNA was cloned into either pcDNA3.1 HisC or HEK293T cells were transiently transfected with Xpress-tagged pCMV-3x FLAG. CtBP2 (pcDNA3.1 HisC/CtBP2) and either FLAG-tagged ataxin-1 (pcDNAI amp/ataxin-1) or empty vectors. At 48 h after the transfection, 2.1.3. Reporter constructs the cells were harvested and co-immunoprecipitation experiments The full-length E-cadherin promoter reporter construct, including were performed as previously described [20]. The immunoprecipitates the region extending from 427 bases upstream to 53 bases were subjected to 10% sodium dodecyl sulfate–polyacrylamide downstream of the E-cadherin transcriptional initiation site, and the gel electrophoresis (SDS-PAGE) and the separated proteins were truncated promoter constructs were kind gifts from Dr. Stephen P. transferred onto nitrocellulose. The blots were probed with either Sugrue (Department of Anatomy and Cell Biology, University of anti-FLAG (Sigma) or anti-Xpress antibody (Invitrogen), and detected Florida College of Medicine). This full-length cloned region included with the ECL system (Pharmacia). all of the putative regulatory elements of the E-cadherin promoter, including two palindromic elements (E-boxes), CAAT box, and CpG 2.6. Immunofluorescence experiments island [18]. HeLa cells split on a coverslip were transiently transfected with 2.1.4. Small interference RNA each construct as indicated in Fig. 2. After 24 h post-transfection, the Small interference RNA oligomers for a knock-down of endoge- cells were fixed with 3.7% formaldehyde in phosphate buffered saline nous CtBP2 were purchased from Dharmacon. The purchased (PBS) for 15 min, and then permeabilized with 1% Triton X-100 in PBS siGENOME SMARTpool (M-008962-03) contains a mixture of four for 5 min. The permeabilized cells were blocked with 2% bovine serum SMART selection-designed siRNAs targeting the CtBP2 gene. The albumin (BSA) in PBS for 1 h at 4 °C, and then incubated for 1 h at control siRNA oligomers, non-targeting negative siRNA control pools room temperature with mouse anti-FLAG antibody (1:500). The cells (D-001206-13) were purchased from Dharmacon. were rinsed with PBS three times, and incubated with tetramethylr- hodamine isothiocyanate (TRITC, Sigma-Aldrich)-conjugated anti- 2.2. Yeast two-hybrid screening mouse IgG antibodies (red) for 1 h. Then, the cells were stained with 4′,6′-diamidino-2-phenylindole dihydrochloride (DAPI, Sigma- The yeast expression construct pLexA-DBD/ataxin-1(82Q) was Aldrich) to observe the nucleus. Finally, the coverslips were rinsed sequentially cotransformed with a human fetal brain Matchmaker three times with PBS and mounted on glass slides with FluoroGuard™ two-hybrid library (Clontech) into yeast strain EGY48 (Clontech). For Antifade Reagent (Bio-Rad). Fluorescence images were obtained using the library selection, approximately 1×107 independent transfor- a fluorescence microscope (OLYMPUS) and processed with Adobe mants were plated on selective medium lacking the amino acids Photoshop. leucine, histidine, and tryptophan but containing 5-bromo-4-chloro- 3-indolyl-beta-D-galactoside (X-gal). Five days later, Leu+- and LacZ 2.7. Chromatin immunoprecipitation (ChIP) +-positive colonies were picked. The selected positive clones were sequenced by an ABI sequencer, and the sequence of each clone was A ChIP assay was performed as previously described [21]. MCF-7 cells analyzed using the NCBI BLASTN program. were transfected with 4 μg DNAs of FLAG-tagged-ataxin-1(30Q), FLAG- tagged CtBP2 or empty vectors. At 48 h after transfection, cells were cross- 2.3. Yeast quantitative β-galactosidase assays linked in 1% formaldehyde and the cross-linking reaction was quenched by adding 0.2 M glycine. Pellets collected by centrifugation were washed Transformants selected on histidine- and tryptophan-deficient twice with ice-cold Tris buffered saline, and then three times with MC lysis plates were cultured in histidine- and tryptophan-deficient liquid buffer (10 mM Tris–Cl [pH 7.5], 10 mM NaCl, 3 mM MgCl2,and0.5%NP- medium supplemented with glucose for 24 h. For the induction of the 40), which disrupted the cells and generated a nuclear pellet. The nuclear Lex-DBD fusion protein and B42AD fusion protein, the liquid culture pellet was resuspended with MNase buffer (10 mM Tris–Cl [pH 7.5], S. Lee et al. / Biochimica et Biophysica Acta 1813 (2011) 713–722 715

10 mM NaCl, 3 mM MgCl2,1mMCaCl2, 4% NP-40, and 1 mM phenyl- as follows: initial denaturation at 95 °C for 5 min, followed by 35–37 methanesulphonylfluoride (PMSF)) and then 2 mM PMSF, 1× protease cycles of 95 °C for 10 s, 59 °C for 10 s, and 72 °C for 20 s. Thresholds inhibitors, 1% SDS, and 200 mM NaCl were added in the order indicated cycles (Ct) were determined as recommended by the manufacturer's and mixed well. The resuspended pellet was sheared by sonication to software. The amplified DNA was separated on a 1.5% agarose gel and reduce the DNA fragment size to approximately 500 bp. After removing visualized with ethidium bromide. cellular debris, chromatin samples were diluted (1:4) by adding FA lysis buffer(50mMHEPES[pH7.5],150mMNaCl,1mMEDTA,1%TritonX- 2.10. Luciferase reporter assay 100, 0.1% sodium deoxycholate, and 0.1% SDS) containing 2 mM PMSF and 1× protease inhibitors, and precleared with protein G-Sepharose beads HEK293T cells were cotransfected with 0.2 μg of pGL3–E-cadherin (GE Healthcare Bio-Sciences). Ten percent of the precleared chromatin reporter gene constructs, 0.1 μg of pSV–β-galactosidase control vector was taken as an input, and the remaining supernatant was immunopre- (Promega), and the indicated amount of a given type of DNA per dish cipitated with anti-FLAG antibody at 4 °C for 4 h. Immunoprecipitated using the polyethylenimine precipitation method. The pSV–β-galac- samples were additionally incubated at 4 °C for 2 h with the addition of tosidase construct was used as an internal control for transfection protein G-sepharose. DNA and proteins non-specifically associated with efficiency. The amount of DNA in each transfection reaction was the protein G-Sepharose were removed by washing the beads twice with equalized to control for non-specific effects on the expression of FA lysis buffer/0.15 M NaCl, once with FA lysis buffer/0.5 M NaCl, ChIP luciferase using an empty expression vector. After transfection, the washing buffer (10 mM Tris–Cl [pH 8.0], 0.25 M LiCl, 1 mM EDTA, 0.5% cells were incubated for 24 h and assayed for luciferase expression NP-40, and 0.5% sodium deoxycholate), and finally with TE buffer (10 mM using Promega's Luciferase Assay System according to the manufac- Tris–Cl [pH 8.0], and 1 mM EDTA). The beads were then resuspended in turer's instructions. The luciferase reporter activity was measured ChIP elution buffer (50 mM Tris–Cl [pH7.5], 10 mM EDTA, and 1% SDS) at with the Luminoskan Ascent (Thermo Labsystems). Each experiment 65 °C for 10 min. The eluted protein–DNA complexes were incubated at was performed more than at least two independent experiments in 42 °C for 2 h in the presence of 2 mg/ml proteinase K to digest the protein, triplicate. Statistical analysis was performed using a t-test. followed by incubation at 65 °C overnight to reverse formaldehyde cross- linking. The DNA was phenol-extracted and ethanol-precipitated. 2.11. Reverse transcription-PCR (RT-PCR) analysis Precipitated DNA was PCR-amplified with CDH1 (E-cadherin) primers as previously described [22,23] to detect the human E-cadherin promoter MCF-7 cells were transfected with DNA as indicated in Fig. 4B. region. The sequences of the CDH1 primers were (forward) 5′- After 36 h, total RNA was isolated from transfected MCF-7 cells using ACTCCAGGCTAGAGGGTCAC-3′ and (reverse) 5′-CCGCAAGCTCA- TRIzol reagent (Invitrogen). RT-PCR reactions were carried out using CAGGTGCTTTGCAGTTCC-3′. The human TK gene was used as a negative the SuperScript III First-Strand Synthesis System (Invitrogen) accord- control. The TK gene sequences were (forward) 5′-TCCCGGATTCCTCC- ing to the manufacturer's instructions. cDNA was synthesized from

CACGAG-3′ and (reverse) 5′-TGCGCCTCCGGGAAGTTCAC-3′ [19,24].The 1 μg of total RNA using the oligo(dT)20 primer. The cDNA obtained in PCR cycling conditions were as follows: 95 °C for 5 min; then 35 cycles of the synthesis reaction was amplified using PCR, which was carried out 94°Cfor20s,56°Cfor20s,and72°Cfor20s;followedby72°Cfor at an annealing temperature of 60 °C for 30 cycles for E-cadherin and 5 min. The amplified DNA was separated on a 1.5% agarose gel and at an annealing temperature of 57 °C for 25 cycles for β-actin. The visualized with ethidium bromide. primer sequences used in the PCR experiments were described previously [27]. The target sequences for E-cadherin were (forward) 2.8. Sequential chromatin immunoprecipitation (SeqChIP) 5′-TTCCTCCCAATACATCTCCCTTCACAG-3′ and (reverse) 5′-CGAA- GAAACAGCAAGAGCAGCAGAATC-3′ [28]. The target sequences for β- SeqChIP assay was performed as described previously [25,26]. actin were (forward) 5′-CCAACTGGGACGACATGGAG-3′ and (reverse) MCF-7 cells were cotransfected with FLAG-tagged-CtBP2 and Xpress- 5′-GCACAGCTTCTCCTTAATGTC-3′. tagged-ataxin-1(30Q). Twenty-four hours later, transfected cells were treated according to conventional ChIP procedures by first performing 3. Results IP with anti-FLAG antibody. Protein–DNA complexes obtained by the first IP were subjected to an additional IP with anti-Xpress antibody, 3.1. Identification of human CtBP2 as a protein interacting with ataxin-1 anti-CtBP2 antibody (C-16, Santa Cruz Biotechnology) or IgG anti- mouse in SeqChIP. First, the precipitated complexes were eluted via a To identify proteins that interact with ataxin-1, we performed a yeast 10 min incubation with 100 μl of ChIP elution buffer (10 mM Tris–Cl two-hybrid screening using full-length mutant ataxin-1(82Q) as bait [pH 7.5], 10 mM EDTA, and 1% SDS) at 65 °C. At this point, 10 μl as the (Fig. 1A). Yeast expressing ataxin-1 was sequentially transformed with a first IP sample was removed from the sample for subsequent analysis human fetal brain cDNA library. The partial cDNA encoding the of the first immunoprecipitation elutes (90 μl). For SeqChIP, the C-terminal region (amino acids 203–445) of the human C-terminal elutants were incubated in the presence of 5 mg/ml BSA, 25 μg/ml binding protein 2 (CtBP2) was isolated from independent positive clones. phage λ DNA, 50 μg/ml yeast tRNA, an appropriate amount of each To confirm the interaction between ataxin-1 and the CtBP2 C-terminal antibody, and 50 μl of protein G-Sepharose in a total volume of 1 ml FA region (CtBP2203–445), we performed β-galactosidase assays using the lysis buffer (150 mM NaCl, without SDS) to make the second yeast two-hybrid system. Strong β-galactosidase activity was observed in immunoprecipitation similar to the initial one. The washes, elution, yeast cells expressing LexA-ataxin-1 and pB42AD-CtBP2203–445,whereas and crosslink reversal following the second IP were carried out as weak activity was shown in cells containing LexA-ataxin-1(82Q) and described above. pB42AD, or LexA-BD and pB42AD-CtBP2203–445 (Fig. 1B), indicating that a specific interaction exists between CtBP2203–445 and ataxin-1, and that the 2.9. Quantitative real-time PCR (qPCR) polyglutamine tract of ataxin-1 does not modulate the interaction

between ataxin-1 and CtBP2203–445. qPCR of the DNA samples described above was used to assess the To deﬁne the protein region responsible for the interaction extent of co-occupancy. Each PCR reaction was performed in triplicate between ataxin-1 and CtBP2, deleted versions of ataxin-1 were in a total volume of 20 μl in each well, containing one-tenth of each IP expressed as LexA-BD fusion proteins in yeast. As shown in Fig. 1C, sample or the ﬁrst IP sample as a template DNA, 1 μM CDH1 primers LacZ reporter genes were strongly activated in cells co-expressing and 2× SYBR Green Master mix (Takara Bio). qPCR was performed on pB42AD-CtBP2 and LexA-ataxin-1N-terminus (a.a. 1–196). However, a Light Cycler 480 (Roche Applied Science). The PCR conditions were β-galactosidase activities were weak in cells expressing the protein 716 S. Lee et al. / Biochimica et Biophysica Acta 1813 (2011) 713–722

Fig. 1. Identiﬁcation of hCtBP2 as a protein that interacts with ataxin-1(82Q). (A) Schematic representation of ataxin-1 and CtBP2. Full-length human mutant ataxin-1(82Q) fused to

LexA-DBD was used as bait in a yeast two-hybrid screening. Clone number 38, represented as CtBP2203–445, corresponded to the C-terminal region of hCtBP2. (B) CtBP2203–445 specifically interacts with ataxin-1. β-Galactosidase activity in liquid culture was quantified after yeasts were cotransformed with the indicated constructs. LexA-ataxin-1(82Q) and pB42AD, or LexA-BD and pB42AD-CtBP2203–445, and LexA-p53 and pB42AD-CtBP2203–445 were used as negative and positive controls, respectively. (C) Schematic representation of various ataxin-1 deletion mutants fused to LexA-DBD. β-Galactosidase activity was quantified after yeasts were cotransformed with the indicated constructs and full-length pB42AD-CtBP21–445. Yeast cells expressing LexA-atrophin-1 and pB42AD-CtBP2 were served as a negative control. Mean values and standard deviations from two independent experiments performed in triplicate are shown.

pairs, LexA-ataxin-1 (a.a. 275–587) and pB42AD-CtBP2, or LexA- 3.2. Ataxin-1 colocalizes with CtBP2 in the nucleus ataxin-1C-terminus (a.a. 537–816) and pB42AD-CtBP2. These results suggest that the N-terminal (a.a. 1–196) region of ataxin-1 is essential CtBP has been known to be diffusely distributed both in the nucleus for the protein–protein interaction. In control experiments, co- and cytoplasm [7,29–31]. On the other hand, it was reported that expression of pB42AD-CtBP2 and LexA-atrophin-1 (the DRPLA gene ataxin-1 proteins form nuclear inclusions (NIs) in mammalian cells. To product) C-terminus did not show β-galactosidase activity, while examine the distribution of CtBP2 in the presence of ataxin-1 in activation did occur with pB42AD-TAg and LexA-p53 performed as the mammalian cells, HeLa cells were cotransfected with the expression positive control. plasmid of pEGFP-ataxin-1(30Q) or ataxin-1(82Q), and FLAG-tagged Co-immunoprecipitation experiments were carried out in mam- CtBP2, and immunofluorescence experiments were performed using malian cells to verify the interaction between CtBP2 and ataxin-1. anti-FLAG antibodies. In cotransfected HeLa cells, CtBP2 were observed Constructs containing Xpress-tagged CtBP2 and either FLAG-tagged in the same loci with both ataxin-1(30Q) and ataxin-1(82Q) (Fig. 2B), ataxin-1(30Q or 82Q), or empty vectors were cotransfected into indicating that CtBP2 and ataxin-1 colocalize in the nucleus. HEK293T cells. After 48 h, total cell extracts were prepared and Additionally, to investigate the distribution of endogenous CtBP in immunoprecipitated with anti-FLAG antibodies. Fig. 2A shows that mammalian cells, HeLa cells expressing exogenous ataxin-1 was CtBP2 proteins were co-immunoprecipitated with ataxin-1(30Q) or analyzed using immunofluorescence confocal microscopy (Supple- ataxin-1(82Q) (Fig. 2A, lanes 1 and 2) but not with pcDNA1 empty mentary Fig. 2). Exogenous ataxin-1 was localized in nuclear vector (Fig. 2A, lane 3). In addition, PLDLS-binding motifs of CtBP2 inclusions (NIs) and endogenous CtBP was shown to be diffusely did not appear to be necessarily required for interaction with distributed throughout the cell. In HeLa cells expressing exogenous ataxin-1 (Supplementary Fig. 1). These results are indicative of an ataxin-1, however, CtBP was redistributed into NIs of ataxin-1. interaction between CtBP2 and ataxin-1 in vivo, and indicate that the length of the polyglutamine tract does not affect the interaction 3.3. Both ataxin-1 and CtBP2 occupy the endogenous E-cadherin promoter between the two molecules. Additional reciprocal co-immunopre- in vivo cipitation assays using anti-Xpress antibodies also confirmed the specific interaction between ataxin-1 and CtBP2 in mammalian cells Several transcriptional factors, such as TBP and CBP, contain (data not shown). glutamine tracts. Polyglutamine disease proteins, including androgen S. Lee et al. / Biochimica et Biophysica Acta 1813 (2011) 713–722 717

Fig. 2. Ataxin-1 interacts and colocalizes with CtBP2 in the nucleus. (A) HEK293T cells were transiently cotransfected with FLAG-tagged ataxin-1 and Xpress-tagged CtBP2. After 48 h, total cell extracts were prepared and immunoprecipitated with anti-FLAG antibodies. The resulting immunoprecipitates were Western blotted and analyzed with the anti-Xpress antibody to detect CtBP2 (top). This blot was stripped and probed with anti-FLAG antibody to detect immunoprecipitated FLAG-ataxin-1 (middle). Each whole cell extract was resolved by SDS-PAGE and immunoblotted with anti-Xpress antibody to detect exogenous CtBP2 (bottom). (B) HeLa cells were cotransfected with expression plasmids of FLAG- tagged CtBP2 and either pEGFP control or pEGFP-ataxin-1. At 24 h post-transfection, cells were immunolabeled with anti-FLAG antibody and TRITC-conjugated anti-mouse IgG (red). Fluorescence images were obtained using a fluorescence microscope (OLYMPUS). Merged signals are shown in yellow. receptor and atrophin-1, have been reported to function as transcrip- chromatin extracts prepared from MCF-7 cells expressing the two tional factors. The interaction between ataxin-1 and CtBP2 strongly proteins, followed by a second IP with anti-Xpress, anti-CtBP2 suggests that ataxin-1 functions as a transcriptional regulator within a antibodies, or anti-mouse IgG. DNAs eluted through the SeqChIP CtBP corepressor complex. CtBP proteins exist as corepressor com- process were quantified by real-time PCR. E-cadherin promoter plexes consisting of a variety of transcriptional regulators [32].We occupancy was determined by the amount of PCR product yielded tried to identify a target gene(s) regulated by ataxin-1 in terms of after triplicate amplifications. PCR products amplified by real-time CtBP2-mediated transcriptional regulation. It was previously reported PCR were normalized to products acquired by the non-specific anti- that the transcriptional regulation of E-cadherin might involve several IgG IP. Western blot analysis (Fig. 3C) indicated that CtBP2 and ataxin- repressor complexes, such as the SIN3 repressor complex and CtBP 1 proteins were sufficiently overexpressed for the SeqChIP assays to corepressor complex [33]. Therefore, we used ChIP assays to examine be performed. Fig. 3D shows that two proteins might simultaneously whether CtBP2 and ataxin-1 are recruited to the E-cadherin promoter. co-occupy a stretch of E-cadherin promoter, but it also indicates that MCF-7 cells were transfected with FLAG-tagged CtBP2 or ataxin-1. the two proteins associate with different populations of the E- Formaldehyde cross-linked chromatin extracts from transfected MCF- cadherin promoter sequence and ataxin-1 is a part of transcriptional 7 cells were immunoprecipitated with anti-FLAG antibodies, and the complexes occupying specific sites on the E-cadherin promoter. presence of CDH1 (E-cadherin) promoter sequences was detected by Ataxin-1 proteins appear to partially occupy the E-cadherin promoter PCR with specific primers. Anti-IgG antibody IP and TK promoter relative to CtBP2 proteins. These results suggest that CtBP2 and primer amplification were used as non-specific controls and negative ataxin-1 are co-recruited on the E-cadherin promoter to form controls, respectively. The Western blot analysis shown in Fig. 3A was transcriptional complexes. carried out to evaluate the overexpression of each protein for the ChIP assay. The CDH1 promoter region was enriched when either CtBP2 or ataxin-1 was precipitated with anti-FLAG antibody, while the TK 3.4. Overexpression of ataxin-1 enhances E-cadherin expression at the promoter sequences were not (Fig. 3B). These results show that both protein and mRNA levels in MCF-7 cells CtBP2 and ataxin-1 occupy the endogenous E-cadherin promoter in vivo. To determine whether ataxin-1 can affect the expression of Furthermore, sequential ChIP (SeqChIP) was used to qualitatively endogenous E-cadherin, Western blot analysis was performed to detect address whether two proteins can simultaneously co-occupy a stretch alterations in the E-cadherin protein levels after transfection with of E-cadherin promoter sequences in vivo. MCF-7 cells were Xpress-tagged ataxin-1 or CtBP2. For the Western blot analysis, cotransfected with FLAG-tagged CtBP2 and Xpress-tagged ataxin-1 duplicate experiments were performed. The diagram in Fig. 4A(lower constructs. First, IP with anti-FLAG antibody against CtBP2 proteins panel) shows the relative amounts of E-cadherin proteins normalized by was performed with DNAs eluted from formaldehyde cross-linked β-actin proteins using Multi-Gauge (FUJIFILM) software. Interestingly, 718 S. Lee et al. / Biochimica et Biophysica Acta 1813 (2011) 713–722

Fig. 3. Both ataxin-1 and CtBP2 occupy the endogenous E-cadherin promoter in vivo. (A) FLAG-tagged CtBP2 and -ataxin-1(30Q) were transfected into MCF-7 cells, respectively. Western blot analysis was performed to evaluate each protein overexpression for the ChIP assay. (B) Chromatin extracts eluted from MCF-7 cells expressing CtBP2 and ataxin-1 were immunoprecipitated using anti-FLAG antibody, and PCR-amplified with specific primers, such as CDH1 (E-cadherin) or TK promoter primer. Chromatin DNA extract from empty vector transfected cells, anti-IgG IP, and TK promoter primer amplification were used as non-specificor‘non-target’ controls to detect the E-cadherin promoter sequence. (C) MCF-7 cells were cotransfected with FLAG-tagged CtBP2 and Xpress-tagged ataxin-1(30Q) constructs. Western blot analysis was performed to detect overexpressed-proteins for the SeqChIP assay. The IgG control lane in Fig. 2D could be a control of Fig. 2C and D. (D) IP with anti-FLAG antibody against CtBP2 proteins was performed with DNAs eluted from formaldehyde cross-linked chromatin extracts prepared from MCF-7 cells co-expressing the two proteins, followed by a second IP with anti-Xpress, anti-CtBP2 antibodies, or anti- mouse IgG. DNAs eluted through the SeqChIP process were quantified by real-time PCR. E-cadherin promoter occupancy was determined by the amount of PCR product yielded after triplicate amplifications. The amplified PCR products were normalized to products yielded by an anti-IgG IP. The SeqChIP products amplified by real-time PCR were separated on a 1.5% agarose gel and visualized with ethidium bromide (lower panel). the overexpression of the ataxin-1 proteins increased the levels of E- was isolated from the lysed cells and cDNAs were synthesized using cadherin proteins. RT-PCR. The cDNAs were amplified with E-cadherin or β-actin Then, real-time PCR was carried out to quantify the mRNA levels of primers by real-time PCR. PCR products amplified by the E-cadherin E-cadherin. MCF-7 cells were transfected with FLAG-tagged CtBP2 primer were normalized to those amplified by the β-actin primer. PCR and/or Xpress-tagged ataxin-1(30Q) as indicated in Fig. 4B. Total RNA was performed in triplicate. E-cadherin mRNA level was increased in

Fig. 4. Ataxin-1 increases the levels of E-cadherin proteins and mRNAs in MCF-7 cells. (A) MCF-7 cells were transiently transfected with Xpress-tagged ataxin-1(30Q) or CtBP2 plasmids and harvested 24 h later. Western blot analysis was performed in duplicate experiments, and the histogram (lower panel) shows the relative amounts of E-cadherin proteins normalized by β-actin proteins using Multi-Gauge (FUJIFILM) software. (B) Transfected MCF-7 cells with each expression vector for ataxin-1 and CtBP2 were harvested 36 h later. Total RNA was isolated from the lysed cells and cDNAs were synthesized using RT-PCR employing oligo(dT)20 primer. The cDNAs obtained in the synthesis reaction were amplified with E-cadherin or β-actin primers by real-time PCR. PCR products amplified by the E-cadherin primer were normalized to those amplified by the β-actin primer. Each PCR reaction was performed in triplicate. Two independent experiments were performed and representative one was shown. The error bars indicate standard deviations. S. Lee et al. / Biochimica et Biophysica Acta 1813 (2011) 713–722 719 the case of overexpressing ataxin-1. Our results clearly show that promoter and overexpression of an increased amount of ataxin-1 overexpressed ataxin-1 affects the transcriptional regulation of the consistently maintained its activity level. However, in cells expressing E-cadherin gene in MCF-7 cells. endogenous CtBP2 proteins (Fig. 5D, black bars), ataxin-1 increased its activity in a dose-dependent manner. These results suggest that 3.5. Ataxin-1 activates the E-cadherin gene promoter in a CtBP2-dependent ataxin-1 might effectively regulate the promoter of the E-cadherin manner gene in the presence of CtBP2, indicating a functional linkage between ataxin-1 and CtBP2 on the regulation of E-cadherin promoter. Taken The relationship between ataxin-1 and CtBP2 in transcriptional together, it is assumed that ataxin-1 proteins might be recruited to the regulation of E-cadherin gene was investigated. First, luciferase assays CtBP2 transcriptional regulatory complex, resulting in activating the were performed using either Xpress-tagged CtBP2 or ataxin-1, and E-cadherin promoter in a CtBP2-dependent manner. full-length E-cadherin promoter reporter constructs (−427 +53) [18]. Overexpression of the CtBP2 protein enhanced the repression 3.6. Full-length E-cadherin promoter is essential for the recruitment of activity of the E-cadherin promoter in a dose-dependent manner, ataxin-1 and the transcriptional activation by ataxin-1 whereas overexpression of ataxin-1 increased the activity of the E-cadherin promoter in a dose-dependent manner (Fig. 5A and B). CtBP2 proteins are recruited to promoter by sequence-specific To further understand the functional mechanism of interaction DNA-binding transcription factors such as ZEB (zinc-finger E-box between ataxin-1 and CtBP2 in E-cadherin gene regulation, we binding homeobox) and snail, either through direct interaction or investigated the activity of reporter genes mediated by ataxin-1 in the indirectly through bridging protein. These transcription factors block presence or absence of the CtBP2 proteins. HEK293T cells were E-cadherin expression by binding to specific E-box in the E-cadherin transfected with CtBP2 siRNA oligomers as much as amounts promoter [28,34]. Therefore, we performed the luciferase assay using indicated in Fig. 5C. As predicted, depletion of endogenous CtBP2 several truncated reporter constructs to examine whether recruit- increased the luciferase activity of E-cadherin promoter. It was ment of ataxin-1 on the E-cadherin promoter depends on CtBP2 examined whether ataxin-1 overexpression could affect the tran- proteins. The E-cadherin promoter contains several important cis- scriptional activity of the E-cadherin promoter in a knock-down regulatory elements, including an upstream E-box 1 (CAGGTG; −78), condition of endogenous CtBP2. Xpress-tagged ataxin-1 as indicated conserved CCAAT box (−65), CpG island (−52 −32), and down- in Fig. 5D was transiently cotransfected with either 20 pM of CtBP2 stream E-box 2 (CACCTG; −28) [18,35]. These cis-regulatory elements siRNA or control siRNA oligomers into HEK293T cells. The over- are often binding sites of one or more trans-acting factors. expressed cells were harvested and assayed 24 h later. When Cotransfection was performed with full-length ataxin-1(30Q) and endogenous CtBP2 proteins were depleted by siRNA oligomers each truncated E-cadherin promoter construct as shown in Fig. 6. The targeting the CtBP2 gene (Fig. 5D, gray bars), overexpression of full-length cloned region, including all of the putative regulatory ataxin-1 slightly enhanced the luciferase activity of the E-cadherin elements of the E-cadherin promoter, was the most responsive to

Fig. 5. Ataxin-1 activates the E-cadherin gene promoter in a CtBP2-dependent manner. (A and B) Luciferase assays were performed using the full-length E-cadherin promoter reporter gene (−427 +53). HEK293T cells were transfected with two different concentrations of Xpress-tagged ataxin-1(30Q) or CtBP2 as indicated, along with the E-cadherin promoter reporter gene, pSV–β-galactosidase constructs. (C) CtBP2 siRNA oligomers were transfected in three different concentrations to assess silencing effects of CtBP2 in HEK293T cells. (D) HEK293T cells were transiently cotransfected with Xpress-tagged ataxin-1(30Q) at two different concentrations and either CtBP2 siRNA or control siRNA oligomers (20 pM). Transfected cells were harvested and assayed 24 h later. Luciferase reporter activity was normalized with β-galactosidase. Each experiment was performed three independent experiments in triplicate. Statistical analysis was performed using a t-test. 720 S. Lee et al. / Biochimica et Biophysica Acta 1813 (2011) 713–722

A promoter and could affect the regulation of E-cadherin promoter. However, the depletion of CtBP2 with CtBP2 siRNA might block the access of ataxin-1 to the E-cadherin promoter region as drawn in the bottom panel of Fig. 7. However, the promoter activity in the bottom panel might be higher than that in the top panel because the depletion of CtBP2 by CtBP2 siRNA oligomers can relieve the basal repression activities on the E-cadherin promoter, and alternatively a small quantity of CtBP2 that is not depleted, can recruit ataxin-1 proteins into the transcriptional complexes, resulting in increasing the promoter activity. Conclusively, this study identified the E-cadherin promoter as a novel molecular target that can be regulated by ataxin- 1, suggesting that ataxin-1 might enhance E-cadherin expression by B relieving transcriptional repression by CtBP2. Alternatively, CtBP2 could be recruited to ataxin-1 NIs, which would decrease the available pool of CtBP2, resulting in the increase of E-cadherin expression. Further study is required to understand the mechanism by which ataxin-1 enhances E-cadherin expression. There are an increasing number of reports suggesting that polyglutamine diseases may result from the transcriptional dysregulation caused by mutant proteins with expanded polyglutamine tracts in specific neuronal populations. Similarly, the connection of ataxin-1 with the regulation of a gene expression provides a fascinating hypothesis regarding the onset mechanism of SCA1. Since the gene expression process is a very complicated phenom- enon controlled by the coordinated action of transcriptional Fig. 6. Full-length E-cadherin promoter is essential for the recruitment of ataxin-1 and the transcriptional activation by ataxin-1. (A) Schematic representation of truncated E- regulators, transcriptional dysregulation mediated by mutant cadherin promoter constructs. (B) Luciferase assay was carried out using full-length or ataxin-1 might be an important feature of SCA1 pathogenesis by several truncated E-cadherin promoter reporter constructs. HEK293T cells were altering general functions of a variety of regulatory nuclear proteins transfected with each reporter construct and Xpress-tagged ataxin-1(30Q) or empty through the disruption of interactions or creation of new ones with vectors. The bars show the luciferase activity of each reporter gene in response to ataxin-1 or empty vectors. Each experiment was performed two independent other regulatory proteins. Indeed, Fernandez-Funez et al. reported experiments in triplicate. that transcriptional regulation cofactors including Sin3A, Rpd3, and dCtBP are one group of genes that modify ataxin-1-induced neurodegeneration [17]. ataxin-1-mediated transcriptional activation. However, serial reduc- Changes in gene expression during the early stage of development tions of the human E-cadherin promoter activity were observed in have been observed in SCA1 transgenic mice [36]. At the same time, cells transfected with truncated E-cadherin promoter constructs. the fact that CtBP2 is presently expressed during the early stages of Especially, the E-box 1 deletion mutant drastically reduced the development and is necessary for the regulation of gene expression in luciferase activity of the reporter gene, indicating that this box multiple developmental programs suggests that CtBP2 may mediate might be critical for the recruitment of the ataxin-1 on the E-cadherin the transcriptional dysregulation in SCA1. In addition, E-cadherin promoter and thus ataxin-1-induced activation. genes that are known to be repressed by CtBP are transiently expressed in restricted regions of mouse embryonic brain, and the 4. Discussion expression is in part associated with local pattern formation of brain tissues [37,38]. Therefore, we have focused on E-cadherin. The Ataxin-1 is predominantly located in the nuclei of neurons [1], and abnormal alteration of E-cadherin expression level by mutant is known to interact with several transcriptional co-regulators, such as ataxin-1 and CtBP2 might affect cerebellar Purkinje cells, severe and PQBP1, LANP, SMRT, and Sp1. Mutant ataxin-1 enhances the frequent pathology sites of SCA1. repression of transcription in the presence of PQBP1 [2]. However, In this study, a relief of CtBP2-mediated repression with mutant ataxin-1 relieves the transcriptional repression induced by the LANP- ataxin-1(82Q) (data not shown) as well as wild-type ataxin-1(30Q) E4F complex by competing with E4F for LANP [4]. Thus, ataxin-1 has was found. However, overexpressed ataxin-1(82Q) proteins in- an important role in transcriptional regulation. The function of ataxin- creased more the level of E-cadherin mRNA and the luciferase 1 might be changeable between repressor and activator depending on activity of E-cadherin promoter than ataxin-1(30Q) proteins. We specific cellular contexts. Our study reveals that ataxin-1 is capable of could not detect functional differences regarding the E-cadherin up-regulating the expression of E-cadherin by reducing the effect of gene regulation between wild-type and mutant ataxin-1. The result CtBP2-mediated repression on the E-cadherin gene. might not be surprising, given that overexpressed wild-type proteins We wondered about the mechanism by which ataxin-1 affects the in cell lines can induce similar effects as mutant proteins. function of CtBP2. ChIP assays in Fig. 3 indicate that the ataxin-1 and Nevertheless, the studies of the expression profiles of specific target CtBP2 proteins co-occupy the E-cadherin promoter. Furthermore, the genes that are regulated by ataxin-1 might contribute to the results of the siRNA experiments summarized in Fig. 5 imply that the elucidation of SCA1 pathology and the development of new and transcriptional activation of ataxin-1 is associated with the presence potent therapeutic strategies. Although the transcriptional molecular of CtBP2. On the basis of these results, a hypothesis of the mechanism mechanism of the interaction between ataxin-1 and CtBP2 in the by which ataxin-1 activates the E-cadherin promoter is proposed. As pathogenesis of SCA1 remains to be investigated, we propose that shown in Fig. 7, overexpression of CtBP2 decreases the E-cadherin interference of CtBP-mediated transcriptional repression due to the promoter activity, and overexpression of ataxin-1 in the presence of binding of ataxin-1 to CtBP2 might be involved in the SCA1 endogenous CtBP2 inversely increases the promoter activity. Ataxin-1 pathogenic mechanisms. proteins in the presence of endogenous CtBP2 might be recruited into Supplementary materials related to this article can be found online the CtBP2-mediated transcriptional complexes on the E-cadherin at doi:10.1016/j.bbamcr.2011.01.035. S. Lee et al. / Biochimica et Biophysica Acta 1813 (2011) 713–722 721

Fig. 7. Ataxin-1 activates the promoter in a CtBP2-mediated transcriptional regulation manner. The overexpression of the CtBP2 repressor reduces the E-cadherin promoter activity (top). The addition of ataxin-1 in the presence of endogenous CtBP2 enhances the promoter activity by directly interacting with CtBP2 (middle). The depletion of CtBP2 with siRNA CtBP2 blocks the access of the ataxin-1 to the promoter (bottom).

Acknowledgements [11] U. Dressel, P.J. Bailey, S.C. Wang, M. Downes, R.M. Evans, et al., A dynamic role for HDAC7 in MEF2-mediated muscle differentiation, J. Biol. Chem. 276 (2001) 17007–17013. We thank Jae-Hyun Yoon (Department of Molecular and Cellular [12] C.L. Zhang, T.A. McKinsey, J.R. Lu, E.N. Olson, Association of COOH-terminal- Biology, University of Arizona) for helpful discussions. This work was binding protein (CtBP) and MEF2-interacting transcription repressor (MITR) supported by a grant funded by the Korea Ministry for Health, contributes to transcriptional repression of the MEF2 transcription factor, J. Biol. Chem. 276 (2001) 35–39. Welfare and Family Affairs [A101254] and Korea University grant [13] J.D. Hildebrand, P. Soriano, Overlapping and unique roles for C-terminal binding [K0821341]. protein 1 (CtBP1) and CtBP2 during mouse development, Mol. Cell. Biol. 22 (2002) 5296–5307. [14] M. Grooteclaes, Q. Deveraux, J. Hildebrand, Q. Zhang, R.H. Goodman, et al., C- References terminal-binding protein corepresses epithelial and proapoptotic gene expression programs, Proc. Natl Acad. Sci. USA 100 (2003) 4568–4573. [1] A. Servadio, B. Koshy, D. Armstrong, B. Antalffy, H.T. Orr, et al., Expression analysis [15] Y. Shi, J. Sawada, G. Sui, B. Affar el, J.R. Whetstine, et al., Coordinated histone of the ataxin-1 protein in tissues from normal and spinocerebellar ataxia type 1 modifications mediated by a CtBP co-repressor complex, Nature 422 (2003) 735–738. individuals, Nat. Genet. 10 (1995) 94–98. [16] L.J. Zhao, M. Kuppuswamy, S. Vijayalingam, G. Chinnadurai, Interaction of ZEB and [2] H. Okazawa, T. Rich, A. Chang, X. Lin, M. Waragai, et al., Interaction between histone deacetylase with the PLDLS-binding cleft region of monomeric C-terminal mutant ataxin-1 and PQBP-1 affects transcription and cell death, Neuron 34 binding protein 2, BMC Mol. Biol. 10 (2009) 89. (2002) 701–713. [17] P. Fernandez-Funez, M.L. Nino-Rosales, B. de Gouyon, W.C. She, J.M. Luchak, et al., [3] A. Matilla, B.T. Koshy, C.J. Cummings, T. Isobe, H.T. Orr, et al., The cerebellar leucine- Identification of genes that modify ataxin-1-induced neurodegeneration, Nature rich acidic nuclear protein interacts with ataxin-1, Nature 389 (1997) 974–978. 408 (2000) 101–106. [4] M. Cvetanovic, R.J. Rooney, J.J. Garcia, N. Toporovskaya, H.Y. Zoghbi, et al., The role [18] R. Alpatov, G.C. Munguba, P. Caton, J.H. Joo, Y. Shi, et al., Nuclear speckle- of LANP and ataxin 1 in E4F-mediated transcriptional repression, EMBO Rep. associated protein Pnn/DRS binds to the transcriptional corepressor CtBP and 8 (2007) 671–677. relieves CtBP-mediated repression of the E-cadherin gene, Mol. Cell. Biol. 24 [5] C.C. Tsai, H.Y. Kao, A. Mitzutani, E. Banayo, H. Rajan, et al., Ataxin 1, a SCA1 (2004) 10223–10235. neurodegenerative disorder protein, is functionally linked to the silencing [19] D. Choi, S.J. Lee, S. Hong, I.H. Kim, S. Kang, Prohibitin interacts with RNF2 and mediator of retinoid and thyroid hormone receptors, Proc. Natl Acad. Sci. USA regulates E2F1 function via dual pathways, Oncogene 27 (2008) 1716–1725. 101 (2004) 4047–4052. [20] S. Hong, S. Lee, S.G. Cho, S. Kang, UbcH6 interacts with and ubiquitinates the SCA1 [6] R. Goold, M. Hubank, A. Hunt, J. Holton, R.P. Menon, et al., Down-regulation of the gene product ataxin-1, Biochem. Biophys. Res. Commun. 371 (2008) 256–260. dopamine receptor D2 in mice lacking ataxin 1, Hum. Mol. Genet. 16 (2007) 2122–2134. [21] O. Aparicio, J.V. Geisberg, E. Sekinger, A. Yang, Z. Moqtaderi, et al., Chromatin [7] U. Schaeper, J.M. Boyd, S. Verma, E. Uhlmann, T. Subramanian, et al., Molecular immunoprecipitation for determining the association of proteins with specific cloning and characterization of a cellular phosphoprotein that interacts with a genomic sequences in vivo, Curr. Protoc. Mol. Biol. (2005)8 Chapter 21: Unit 21 23. conserved C-terminal domain of adenovirus E1A involved in negative modulation [22] S. Guaita, I. Puig, C. Franci, M. Garrido, D. Dominguez, et al., Snail induction of of oncogenic transformation, Proc. Natl Acad. Sci. USA 92 (1995) 10467–10471. epithelial to mesenchymal transition in tumor cells is accompanied by MUC1 [8] M. Brannon, J.D. Brown, R. Bates, D. Kimelman, R.T. Moon, XCtBP is a XTcf-3 co- repression and ZEB1 expression, J. Biol. Chem. 277 (2002) 39209–39216. repressor with roles throughout Xenopus development, Development 126 (1999) [23] S. Peiro, M. Escriva, I. Puig, M.J. Barbera, N. Dave, et al., Snail1 transcriptional 3159–3170. repressor binds to its own promoter and controls its expression, Nucleic Acids Res. [9] K. Izutsu, M. Kurokawa, Y. Imai, K. Maki, K. Mitani, et al., The corepressor CtBP 34 (2006) 2077–2084. interacts with Evi-1 to repress transforming growth factor beta signaling, Blood [24] S. Wang, B. Zhang, D.V. Faller, Prohibitin requires Brg-1 and Brm for the repression 97 (2001) 2815–2822. of E2F and cell growth, EMBO J. 21 (2002) 3019–3028. [10] T.A. Melhuish, D. Wotton, The interaction of the carboxyl terminus-binding [25] J.V. Geisberg, K. Struhl, Analysis of protein co-occupancy by quantitative protein with the Smad corepressor TGIF is disrupted by a holoprosencephaly sequential chromatin immunoprecipitation, Curr. Protoc. Mol. Biol. (2005)8 mutation in TGIF, J. Biol. Chem. 275 (2000) 39762–39766. Chapter 21: Unit 21 28. 722 S. Lee et al. / Biochimica et Biophysica Acta 1813 (2011) 713–722

[26] J.V. Geisberg, K. Struhl, Quantitative sequential chromatin immunoprecipitation, a [32] G. Chinnadurai, CtBP, an unconventional transcriptional corepressor in develop- method for analyzing co-occupancy of proteins at genomic regions in vivo, ment and oncogenesis, Mol. Cell 9 (2002) 213–224. Nucleic Acids Res. 32 (2004) e151. [33] H. Peinado, E. Ballestar, M. Esteller, A. Cano, Snail mediates E-cadherin repression [27] S. Lee, S. Hong, S. Kang, The ubiquitin-conjugating enzyme UbcH6 regulates the by the recruitment of the Sin3A/histone deacetylase 1 (HDAC1)/HDAC2 complex, transcriptional repression activity of the SCA1 gene product ataxin-1, Biochem. Mol. Cell. Biol. 24 (2004) 306–319. Biophys. Res. Commun. 372 (2008) 735–740. [34] G. Chinnadurai, Transcriptional regulation by C-terminal binding proteins, Int. J. [28] E. Batlle, E. Sancho, C. Franci, D. Dominguez, M. Monfar, et al., The transcription Biochem. Cell Biol. 39 (2007) 1593–1607. factor snail is a repressor of E-cadherin gene expression in epithelial tumour cells, [35] J. Comijn, G. Berx, P. Vermassen, K. Verschueren, L. van Grunsven, et al., The two- Nat. Cell Biol. 2 (2000) 84–89. handed E box binding zinc finger protein SIP1 downregulates E-cadherin and [29] Y. Nibu, H. Zhang, M. Levine, Interaction of short-range repressors with induces invasion, Mol. Cell 7 (2001) 1267–1278. Drosophila CtBP in the embryo, Science 280 (1998) 101–104. [36] X. Lin, B. Antalffy, D. Kang, H.T. Orr, H.Y. Zoghbi, Polyglutamine expansion down- [30] R. Weigert, M.G. Silletta, S. Spano, G. Turacchio, C. Cericola, et al., CtBP/BARS regulates specific neuronal genes before pathologic changes in SCA1, Nat. induces fission of Golgi membranes by acylating lysophosphatidic acid, Nature Neurosci. 3 (2000) 157–163. 402 (1999) 429–433. [37]K.Shimamura,S.Hirano,A.P.McMahon,M.Takeichi,Wnt-1-dependent [31] P. Criqui-Filipe, C. Ducret, S.M. Maira, B. Wasylyk, Net, a negative Ras-switchable regulation of local E-cadherin and alpha N-catenin expression in the embryonic TCF, contains a second inhibition domain, the CID, that mediates repression mouse brain, Development 120 (1994) 2225–2234. through interactions with CtBP and de-acetylation, EMBO J. 18 (1999) [38] K. Shimamura, M. Takeichi, Local and transient expression of E-cadherin involved 3392–3403. in mouse embryonic brain morphogenesis, Development 116 (1992) 1011–1019.