Role of Nuclear Speckle Protein PNN/DRS in Transcriptional Regulation

Material Information

Role of Nuclear Speckle Protein PNN/DRS in Transcriptional Regulation
Copyright Date:


Subjects / Keywords:
Antibodies ( jstor )
Cadherins ( jstor )
Co repressor proteins ( jstor )
Gene expression ( jstor )
Journalism ( jstor )
Messenger RNA ( jstor )
Proteins ( jstor )
Repression ( jstor )
RNA ( jstor )
Splicing ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright the author. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:


This item is only available as the following downloads:

Full Text




Copyright 2005 by Roman Alpatov


I dedicate this work to the Florida Gators.


ACKNOWLEDGMENTS I would like to give my special thanks to my mentor, Dr. Stephen P. Sugrue for shaping my scientific curiosit y, for the great opportunity to work in his laboratory, and for the invaluable experience I gained in the field of biology. I also would like to thank my committee members Dr. Brian Burke, Dr. Na ohiro Terada, Dr. Gerard P.J. Shaw, and Dr. Jorg Bungert for their critical input a nd helpful comments. I thank my friends and coworkers Dr. Moira Jackson, Jong Hoon J oo, Marguerite Hunt and Mr. Gustabo Munguba and his girlfriend Bozena for f un times in the lab and stimulating conversations. I would like to thank Todd Barnash for his great computer support and Linda Hanssen and Kimberly Hodges for admini strative help. I also would like to thank our Department of Anatomy and Cell Biology as well as my classmates for their support during the school years. I am thankful to my girlfriend Penny Lim for encouragement and understanding. And, finally, I want to extend my gratitude to my family for their love and support. iv


TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF FIGURES..........................................................................................................vii ABSTRACT.....................................................................................................................viii CHAPTER 1 INTRODUCTION........................................................................................................1 2 NUCLEAR SPECKLE-ASSOCIATED PR OTEIN PNN/DRS BINDS TO THE TRANSCRIPTIONAL CO-REPRESSOR CTBP AND RELIEVES CTBP MEDIATED REPRESSION OF THE E-CADHERIN GENE....................................8 Introduction................................................................................................................... 8 Materials and Methods...............................................................................................11 Cell lines, Cell Culture and Transfections...........................................................11 Expression Vectors and Reporter Constructs......................................................12 In vitro Binding Assays.......................................................................................13 Pnn and CtBP1-Flag Complex Isolation.............................................................14 Isolation of the Endogenous Pnn-CtBP Complex...............................................15 Co-immunoprecipitaton and Immunoblotting.....................................................15 Cell Fixation and Immunofluorescence...............................................................16 RNA Interference................................................................................................16 Dual Luciferase Reporter Assays........................................................................17 Results.........................................................................................................................18 Pnn Associates with CtBP1 in vitro and in vivo ..................................................18 Nuclear Distribution of Pnn is Altered in Response to CtBP1 Overexpression..20 Pnn Upregulates the E-cadhe rin Promoter Activity............................................22 Full-length E-cadherin Promoter is Re quired for Pnn-dependent Upregulation of Transcription................................................................................................25 Pnn De-represses CtBP1-mediated Sile ncing of the E-cadherin Promoter.........27 Discussion...................................................................................................................30 Note.............................................................................................................................35 3 PNN/DRS, A CTBP AND RNA PO LYMERASE II-ASSOCIATED FACTOR, REGULATES E-CADHERIN GENE TRANSCRIPTION AND MRNA PROCESSING............................................................................................................36 v


Introduction.................................................................................................................36 Materials and Methods...............................................................................................38 Cell lines, Cell Culture and Transfections...........................................................38 Expression Vectors..............................................................................................39 RNA Polymerase II Antibodies...........................................................................39 Co-immunoprecipitaton and Immunoblotting.....................................................40 Chromatin Immunoprecipitation (ChIP) Assays.................................................40 Isolation of the Endogenous Pnn-Pol II Complex...............................................41 Pnn and RNA Polymerase II Complex Isolation.................................................41 In vitro Binding Assays.......................................................................................41 Immunofluorescence...........................................................................................42 BromoUridine Incorporation...............................................................................42 Deconvolution Microscopy.................................................................................43 Splicing Reporters and Splicing Assays..............................................................43 Results.........................................................................................................................44 Pnn is Recruited to CtBP-associated Repressor Complexes...............................44 Pnn Can Be Recruited to ZEB1/CtBP and mSin3A/CtBP Complexes via CtBP.................................................................................................................46 Pnn Is Present at the E-cadherin Promoter..........................................................47 Pnn Interacts with the Transcriptionally Competent RNA Polymerase II and Can Differentially Associate with CtBP and Pol II in a NADH-dependent Manner.............................................................................................................48 Distribution of Pnn Overlaps with the Transcriptionally Competent Forms of Pol II.................................................................................................................50 Pnn Positively Affects E-cadherin mRNA Splicing............................................52 Discussion...................................................................................................................53 4 DISCUSSION.............................................................................................................58 LIST OF REFERENCES...................................................................................................66 BIOGRAPHICAL SKETCH.............................................................................................76 vi


LIST OF FIGURES Figure page 1 Pnn interacts with co-repressor CtBP1 in vitro and in vivo .....................................19 2 Pnn exhibits nuclear redistribution in response to CtBP1 overexpression...............21 3 Nuclear distribution of Pnn is al tered in response to CtBP1 RNAi.........................23 4 Pnn is capable of upregulation of the E-cadherin promoter activity, independent of cell type origin.....................................................................................................25 5 Identification of the Pnn-responsive regions in the E-cadherin promoter................26 6 Wild type Pnn, but not the truncated or mutant form of Pnn, is capable of relieving CtBP1-mediated repression of the E-cadherin promoter..........................28 7 Pnn can recruit CtBP1 to a heterologous promoter, leading to a CtBP1-mediated repression of the promoter activity...........................................................................29 8 Pnn, but not Pnn mutant carrying a mu tation in the CtBP1 binding motif, is capable of relieving CtBP1-mediated re pression of a heterologous promoter activity......................................................................................................................30 9 Pnn was recruited to the CtBP-a ssociated repressor complexes..............................45 10 Pnn may be recruited to the ZEB1/ CtBP and mSin3A/CtBP complexes via CtBP.........................................................................................................................46 11 Pnn was present at the E-cadherin promoter, where it modulated transcriptional state of chromatin.....................................................................................................47 12 Pnn interacted with the transcriptionally competent Pol II and can differentially associate with CtBP and Pol II in a NADH-dependent manner...............................49 13 Pnn partially overlaped with the transcriptionally competent Pol II........................51 14 Pnn enhanced E-cadherin mRNA splicing efficiency..............................................53 15 Diagram depicting possible mechanism of Pnn involvement in multiple steps of regulation of E-cadherin gene expression................................................................56 vii


Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ROLE OF NUCLEAR SPECKLE PROTEI N PNN/DRS IN TRANSCRIPTIONAL REGULATION By Roman Alpatov December 2005 Chair: Stephen P. Sugrue Major Department: Anatomy and Cell Biology Pinin/DRS (Pnn), a 140-kDa nuclear and cell adhesion-related protein, is involved in the regulation of cell a dhesion and modulation of the activity of multiple tumor suppressor genes. In the nucleus Pnn is concentrated in the "nuclear speckles," zones of accumulation of transcriptiona l and mRNA splicing factors, where Pnn is involved in mRNA processing. Alternatively, other roles of Pnn in gene regulation have not yet been established. By utilizing in vitro pull-down assays, in vivo interaction studies, and immunofluorescence in comb ination with overexpression and RNA interference experiments, we presented evidence that Pnn interacts with th e known transcriptional corepressor CtBP in a NADH-dependent manner. As a consequence of this interaction Pnn was capable of relieving the CtBP-mediated repression of E-cadherin promoter activity. We also demonstrated that Pnn is recruited to CtBP-associated repressor complexes, which target E-cadherin promot er, where it may attenuate CtBP-mediated viii


repression. In addition, by u tilizing chromatin immunoprecipita tions we showed that Pnn expression affects chromatin modifications at the E-cadherin prom oter region, driving chromatin to a more transcriptionally favorable state. Interestingly, recently reported data indicat ed the presence of Pnn in complexes, associated with the basal tr anscriptional machinery such as Mediator and CA150. We showed, that Pnn preferentially associates with transcriptionally competent RNA polymerase II. In contrast to Pnn/CtBP in teraction, increased NADH levels result in diminished Pnn/Pol II association, sugge sting the existence of NADH-dependent mechanisms which trigger reorganiza tion of transcrip tional complexes. And finally, using E-cadherin splicing reporter co nstruct, we demonstrated, that Pnn positively modulates E-cadherin mRNA splic ing efficiency in a promoter specific manner, thus possibly coordinating promote r-associated and mRNA processing events. In summary, our findings implicate the exis tence of Pnn-dependent protein machineries capable of controlling gene expression at the promoter level as well as through transcription-coupled mRNA processing. ix


CHAPTER 1 INTRODUCTION The unique proteomic composition of any organism is a contribution of the multiple regulatory determinants operati ng during gene expression. This multistep process is composed of intertwined networks of regulatory complexes responsible for coordinated processes allowing the informa tion from specific genomic regions to be processed into mRNA and ultimately to protein. As a result of the coordinated events which occur during gene expression, it is likely that tr anscriptional and posttranscriptional mechanisms function toge ther in concerted fashion allowing tight spatial and temporal regulation gene expres sion. The existence of multiple interconnected checkpoint mechanisms within gene e xpression machinery ensures coordinated surveillance of promoter-associated events , transcription, mRNA processing, as well as translational mechanisms, which contributes to production of the final gene product. It is becoming increasingly clear that this cohesi ve network requires the presence of factors which function as integrators of multiple processes during gene expression and which are capable of influencing gene expression at the gene promoters as well as during transcription and mRNA processing stages. Therefore, the molecular interactions invol ved in the coupling of transcription and splicing-associated events probably arise from the ability of a subset of transcriptional coregulators to interact with components of the transcriptional machinery and splicing apparatus. It often involves a promoter de pendent combination of factors, which are capable of recruiting bridging proteins to the proximity of the basal transcriptional 1


2 machinery. Subsequently, the interaction between coupling proteins and elongating RNA polymerase II establishes the necessary functional links between transcription and splicing. Thus, it is important to recognize the essential role of promoter-related events as well as transcription-coupled mRNA pro cessing as crucial determinants during the production of a gene product. An example of a coupling protein that is involved in promoter-dependent gene activation and mRNA splicing is the coactiv ator activator (CoAA). Identified as a thyroid hormone receptor transcriptional regul ator (TRBP) interacting protein (Iwasaki et al., 2001), CoAA represents heterogeneous nuclear ribonuclear pr otein (hnRNP)-like factor known to engage in sp licing activities. Importantly, the engagement of CoAA in splicing processes can be depende nt on the nature of the promot er it targets. In addition, involvement of CoAA in coupling transcription and splicing is likely to reflect the composition of the transcriptional complexes capable of recruiting CoAA to the promoter (Auboeuf et al., 2004). Another example of a multifunctional pr otein involved in transcription and splicing is a heterogeneous nuclear ribonucle ar protein K (hnRNP K) which interacts with diverse groups of molecu lar partners including chromatin remodeling, transcriptionassociated splicing, mRNA stability, as well as translation factors. The modular structure of this protein allows th e network of coordinated a ssociations within hnRNP Kassociated complexes to be carried out in a controlled manner, creating a compact structure of intracellular co mmunication circuitry, where hnR NP K might represent a hub (Bomsztyk et al., 2004). Fact ors involved in the interac tion with hnRNP K include a chromatin remodeling factor from the polycom b group of proteins Eed, which exists in


3 the complex with H3K9 and H3K27-specific histone methyltransferases (Cao et al., 2002). hnRNP K also interacts with DNA methy ltransferases (Bomsztyk et al., 2004) and factors involved in mRNA processing, such as 9G8 and Srp20 (Shnyreva et al., 2000). In addition, hnRNP K regulates mRNA stability via its associati on with 3’ UTR (Skalweit et al., 2003), and is involved in translational regulation (Bomsztyk et al., 2004). Interestingly, modulators of chromatin structure at the gene promoters can also be involved in mRNA processing steps. C HD1 is a member of the chromodomain containing nuclear proteins involved in chroma tin remodeling (Simic et al., 2003). It was suggested that a novel function for CHD is as a coupling protein that bridges chromatin machinery and splicing apparatus. CHD1 wa s shown to interact with nuclear receptor corepressor NCoR as well as RNA splici ng proteins Srp20, SAF-B, and SR splicing protein p54 (Kelley et al., 1999; Tai et al., 2003). Splici ng assays using a splicing reporter minigene showed that CHD1 modulat es splice site selection, suggesting that CDH1 might be involved in mRNA processing (Kelley et al., 1999; Tai et al., 2003). In addition these findings imply the link between NCoR-dependent chromatin reorganization at the gene promoters and mRNA splicing. Steroid hormone coregulators modulate expression of genes involved in hormonal response networks and control the splicing products of their target genes in a promoter specific manner. Importantly, individual hor monal ligands can have opposite effects on promoter activity and splicing decisions of target genes, presumably through ligand dependent rearrangements of th e transcriptional regulatory co mplexes at the vicinity of their target promoters. ASC-1 and ASC-2 co regulators and their associated proteins stimulate the production of progesterone regul ated gene transcripts. However, they


4 mediated the opposite effect on CD44 pre-mR NA processing in a promoter dependent manner implicating the existence of transcri ption dependent decisi ons that might have differential effects on the na ture of the final gene product (Auboeuf et al., 2004). Pinin (Pnn/DRD) was first identified as cell adhesion related mo lecule, capable of enhancing cell junction formation and stabi lization of the desmosome-intermediate filament complex in epithelial cells (O uyang and Sugrue, 1996). Examination of the distribution of Pnn within co rneal epithelia l cells as they under go transitions from quiescence to migratory revealed that the asso ciation of Pnn with de smosomes is greatly reduced in migrating cells i llustrating the importance of Pnn’s involvement in formation of epithelial phenotype (Shi et al., 2000). Im munohistochemical experiments and analysis of Pnn mRNA expression demonstrated ab sence or reduction of Pnn expression in different tumor lines, which is consistent with Pnn’s contribution to the reinforcement of intracellular adhesion (Shi et al., 2000). Later studies revealed that Pnn exists simultaneously at the desmosomal plaques where it is associated with keratin filaments (Shi and Sugrue, 2000), and in nuclear speckles, known zones of accumulation of bot h splicing and a number of transcriptional factors (Ouyang, 1999; Zimowska et al., 2003) . Speckles are considered to be the storage/assembly sites for splicing factors, such as SR family of proteins, before they are recruited to the locations of nascent premRNA transcript. In addition it has been demonstrated that Pnn interacts with SRrp 130, SRp75, SRm300 splicing factors and affects mRNA processing (Wang et al., 2002; Zimowska et al., 2003). Interestingly, experiments utilizing Pnn-inducible cell lines in combination with Atlas Human Cancer cDNA arrays demonstrated change in the e xpression profiles of a number of genes in


5 response to Pnn expression, suggesting that P nn has a functional role in transcriptional regulation (Shi, 2001). A considerable amount of evidence supports the roles of transc riptional regulators functioning as linkers between the basal transcriptional machinery and splicing complexes. The unique characteristic of these bridging elements is their ability to coordinate the activity of promoter-asso ciated components of RNA Polymerase II complex and at the same time to modulate sp licing efficiency, thus ensuring effective coupling between the production of mRNA and its subsequent processing. These prospects of the involvement of P nn in transcription were substantiated when, in addition to splicing proteins, it was also shown to interact with transcriptional regulators associated with the basal transc riptional machinery such as Mediator and CA150 (Sato et al., 2004; Smith et al., 2004). Mediator complex plays a crucial role in transcriptional initi ation, whereas CA150 modulates el ongation rate of RNA polymerase II and affects splicing. Therefore, it is po ssible that Pnn represents a multifunctional protein capable of coordinating tr anscriptional and splicing events. Indeed, preliminary data indicated that Pnn may interact with nuclear coactivator kDa (NcoA62/SKIP) a known vitamin D recep tor (VDR) interacting partner that augments VDR and other nuclear receptor-m ediated transcription. NcoA62/SKIP also interacts with Ski protein, a potent tran scriptional repressor implicated in TGF signaling, that is involved in re gulation of the differentiation program in the muscle cells (MacDonald et al., 2004). Evidence accumulate d in several laboratories implicates the role of NcoA62/SKIP as a transcriptional activator and modulato r of mRNA splicing. Consistent with NR and VDR coactivator, it enhances 1,25-(OH) 2 D 3 -, retinoic acid-,


6 estrogen-, and glucocorticoid-activated transcription (Zhang et al., 2003) as well as Notch-dependent genes (Bres et al., 2005). In addition, the predominant nuclear proteins identified as NcoA62/SKIP interacting factors include the components of U5 snRNP, a major snRNP involved in splicing of mRNA transcripts. Furthermore, a functional evidence for the role of NcoA62/SKIP in sp licing also exists, as NcoA62/SKIP can influence mRNA splicing of various splici ng reporter minigenes (MacDonald et al., 2004; Nagai et al., 2004; Zhang et al., 2003) . Significantly, NcoA62/SKIP-mediated effect on transcription might arise from its ability to interact with a CDK9/P-TEFb, responsible for the phosphorylation of the C-terminal doma in (CTD) of RNA polymerase II during transcriptional elongati on (Bres et al., 2005). These findings suggested a novel role of Pnn as a part of multi-protein complexes perhaps involved in transcriptioncoupled splicing and/or othe r aspects of mRNA maturation, or spatial organization of these processes in the nucleus. In addition, recent studies using an inte grated proteomic approach identified nuclear receptor coactivator CAPER-associat ed complex to contain Pnn (Jung et al., 2005). U2AF-family related proteins CAPER and CAPER regulate both steroid hormone receptor-mediated transcription and alternative splicing decisions. The model for CAPER mediated coordina tion of transcription and splicing implies that coupling factors can be recruited to the transcriptio nal machinery in a promoter specific manner where they mediate appropriate interactions essential for the expression of a particular gene. It also provides a di rect link between the basal transcriptional machinery and coregulators involved in coordi nation of transcription and sp licing. Promoter specific interactions might mediate the recruitment of CAPER proteins to the specific target


7 promoters where CAPER will be available for incorporation into the initiation complex to further influence pre-mRNA processing during the production of nascent RNA transcript (Dowhan et al., 2005). These findings carry an important implication for the possible role of Pnn in the transcriptio n-coupled mRNA processing network. Furthermore, experiments involving anal ysis of the chromatin remodeling and pre-mRNA splicing activities revealed the presence of Pnn in PRP4 kinase complex, thought to operate during both promoter-associ ated chromatin remodeling and splicing stages of gene expression (Dellaire et al ., 2002). Mammalian PRP4K has been shown to associate with nuclear hormone-regulated ch romatin remodeling factors such as BRG1 and NCoR. Interestingly, PRP4K also copurifie s with splicing relate d proteins such as PRP6, SWAP, U5 snRNP and pinin (Dellaire et al., 2002). This information provided novel insight into the potential ro le of Pnn as part of the promoter-associated regulatory complexes. Our own preliminary findings of the interaction between Pnn and transcriptional corepressor CtBP that target s E-cadherin gene promot er (Grooteclaes and Frisch, 2000; Nitta et al., 2005; Postigo and Dean, 1999; Srinivasan and Atchison, 2004) provided another evidence for existence of the “molecular interface” between Pnn and the transcriptional regulatory machiner y operating at the gene promoters. Significantly, experiments describing Pnndependent associations at the gene control regions as well as in context of basal transcriptional machinery would ultimately elucidate its role in transcription-coupled spli cing and promoter-related events. This will also provide an attractive opportunity fo r a more comprehensive look into Pnn’s contribution to the transcripti onal regulation, an important st ep in our understanding of multiple roles protein factors can play in the cellular organization and function.


CHAPTER 2 NUCLEAR SPECKLE-ASSOCIATED PR OTEIN PNN/DRS BINDS TO THE TRANSCRIPTIONAL CO-REPRESSOR CTBP AND RELIEVES CTBP MEDIATED REPRESSION OF THE E-CADHERIN GENE. Introduction We have studied a molecule, pinin (Pnn/DRS/memA), a 140 kDa phosphoprotein associated with the desmosome and localized in the nucleus of various tissues and cultured cell lines (Bran dner et al., 1997; Ouyang, 1999; Ouyang and Sugrue, 1996; Ouyang and Sugrue, 1992; Shi and Sugrue, 2000; Shi et al., 2000; Wang et al., 2002), which seems to play a key role in the estab lishment and maintenance of epithelia (Shi et al., 2000; Shi et al., 2000). The expression of exogenous Pnn in transformed cells dramatically altered th e recipient cells’ morphology, drivi ng them to a more epithelial phenotype. Expression of Pnn wa s linked to the expression of genes such as E-cadherin, p21cip/waf, MIC-1 and Rho-A, which impact epithelial adhesion, pr oliferation, and cell motility (Shi, 2001). These observations suggest that Pnn may play an integral role in epithelial-specific gene expression. Transfection experiments revealed that the majority of expressed Pnn is found within the nucleus, exhibiting a diffuse nucle oplasmic and nuclear speckle distribution. However, little, if any, exogenous Pnn was locali zed to cell-cell adhe sion sites. This raises the possibility that Pnn may be exer ting its effect predominantly via interaction with nuclear components, perhaps involving mRNA transcription and processing (Sakashita et al., 2004; Wang et al., 2002). The proposed functions of nuclear speckles include a role as storage compartments fo r molecular components involved in gene 8


9 transcription, such as a subpopulation of polymerase II, and mRNA processing machinery, such as the SR family of proteins , distinguished by thei r serine/arginine rich motifs. SR proteins are required for cons tituitive mRNA splicing and regulation of alternative splice site selec tion (Caceres et al., 1997; Longm an et al., 2000; Mintz and Spector, 2000; Misteli et al., 1997; Sacco-B ubulya and Spector, 2002; Spector, 2001). Previous two-hybrid and co-localization experi ments revealed that Pnn indeed binds to, and co-localizes with, the SR protei n family members SRp75 and SRm300, known components of the spliceosome machinery, and a novel SR protein (De llaire et al., 2002; Sacco-Bubulya and Spector, 2002; Wang et al ., 2002; Zimowska et al., 2003). In addition, Pnn was demonstrated to contribute to th e alternative spice site selection in splicing reporter assays, specifically the regula tion of 5’ splice site choice (Wang et al., 2002). Recently, the Mayeda group describe d a role for Pnn along with RNPS1 in alternative pre-mRNA splicing regu lation (Sakashita et al., 2004). In addition to its role in RNA processing, expression of Pnn is linked to increased activity of a subset of tu mor suppressor genes, incl uding p21 and E-cadherin. In particular, p21 promoter activity was si gnificantly enhanced in response to the exogenously expressed Pnn (Shi et al., 2001), tempting speculation th at Pnn’s effect on gene expression may also be conferred th rough specific protein-protein interactions governing transcriptional regu lation. The PEDLS sequence mo tif found in the proximity of C-terminus of Pnn was show n to interact with the transcriptional co-repressor BS69 by two hybrid analyses (data not shown), rais ing the possibility that Pnn may bind to transcriptional proteins. Inte restingly, the PEDLS motif is known to be involved in the interactions of various transc riptional factors with C-term inal binding protein (CtBP), a


10 transcriptional co-repressor with a pronounced function in developmental processes as well as tumorgenesis. Like BS69, CtBP was fi rst identified through the interaction with the adenoviral E1A protein via E1A’s PLDL S motif (Boyd et al., 1993; Hateboer et al., 1995; Ladendorff et al., 2001; Masselink and Bernards, 2000; Schaeper et al., 1995; Schaeper et al., 1998) . CtBP has been found to associate with multiple transcriptional repressors, which contain the basic CtBP-b inding signature sequence motif PXDLS, although flexibility in the sequence specificity for the CtBP binding has been reported (Chinnadurai, 2002; Izutsu et al., 2001; Koip ally and Georgopoulos, 2000; Melhuish and Wotton, 2000; Meloni et al., 1999; Phippen et al., 2000; Postigo and Dean, 2000; Postigo and Dean, 1999; Sewalt et al., 1999). Mammalian CtBP family members include CtBP1 and CtBP2 isoforms, which carry diverse f unctions in embryogenesis and vertebrate development (Hildebrand and Soriano, 2002). The most well-documented function of CtBP proteins is as short range transcriptional repressors, when CtBP proteins are recruited to promoters by sequence-specific DNA-binding transcription factors, either through direct physical intera ction or indirectly thr ough bridging proteins. CtBPmediated repression is believed to involve multiprotein associations, which exhibit histone methyltransferase and deacetylase activities, although many details pertaining to the exact mechanism of CtBP-med iated transcriptional silenci ng remain to be elucidated. One important target for CtBP-mediated repression is the E-cadherin promoter, where CtBP is able to interact with transcriptional repressors and thereby repress Ecadherin gene expression (Gr ooteclaes et al., 2003; Grootec laes and Frisch, 2000; Shi et al., 2003; Zhang et al., 2002). On the basis of these findings it is possible that CtBP participates in coordinated biochemical and en zymatic events resulting in the inhibition


11 of E-cadherin transcription, including complex interaction with promoter targeting transcriptional repressors and chromatin rem odeling complexes. Given the potential of a Pnn-CtBP interaction through the PEDLS binding motif, it is of intere st to explore the possibility of a Pnn-CtBP binding and its eff ect on E-cadherin gene expression. In this study, we investigate the func tional interactions between Pnn and CtBP1. We suggest that Pnn interacts with CtBP 1 and relieves CtBP1-mediated silencing in context of Ecadherin promoter by interfering with Ct BP1-dependent gene silencing events. Materials and Methods Cell lines, Cell Culture and Transfections HEK293, MuM-2C, and MDCK cells were cultured in Dulbecco’s modified Eagle medium (DMEM; Bio Whittaker) contai ning 10% FBS (Cellgro; Mediatech) 2mM L-Glutamine, and 200U/ml each streptomycin a nd penicillin (Cellgro; Mediatech). Cells were passed with 0.1 % trypsin and 0.04% EDTA in Hanks medium. Human corneal epithelial cells (HCET, RCB 1384, K. Araki-Sasaki) were cultured in DMEM/F12 (Bio Whittaker) containing 5% FBS, 5g/ml insulin (Sigma-Aldrich), 0.1g/ml cholera toxin (Sigma-Aldrich), 10ng/ml human epidermal gr owth factor (hEGF; Invitrogen) and 0.5% dimethyl sulfoxide (DMSO; Sigma-Aldrich). Suspensory HeLa cells (s-HeLa) stably expressing Pnn-Flag-HA fusion protein (POZ-N-Pnn-Flag-HA), s-HeLa cells stably expressing CtBP1-Flag-HA fusion protein at 1:1 ratio to the endogenous CtBP1 (Shi et al., 2003). (POZ-CtBP1-Flag-HA), and sHeLa containing empty POZ vector, we re propagated in suspension at 2X10 5 confluence in Joklik’s media (Cambrex) containing 10% FBS (Cellgro;Mediatech) and 200U/ml each streptomycin and penicillin.


12 Cells were transfected at 70-90% confluence, utilizing 3l of 1mg/ml 25kDa branched PEI (Sigma-Aldrich) per 1g of D NA. PEI and DNA were incubated in serum free DMEM for 10 minutes in separate tubes. Following incubation the contents of two tubes were combined, incuba ted for additional 10 minutes, and applied onto cells. Expression Vectors and Reporter Constructs pcI-neo-GFP vector was previously de scribed (Wang et al., 1996), pcI-neoPnnGFP and pcI-neo-PnnGFP1-421 deletion vect or, which expressed hPnn, were based on pcI-neo-GFP. pCMV-CtBP1-Flag, expressing hCtBP1 and pcDNA3.1-Pnn-myc/His, expressing hPnn were based on pCMV-Fla g (Stratagene) and pcDNA3.1-myc/His (Invitrogen) respectively. Pnn-GALBD vect or was based on pBIND construct (Promega) and expressed hPnn and yeast GAL4 DNA binding domain as a fusion protein. CtBP1GALBD vector expressing hCtBP1 and yeast GAL4 DNA binding domain as a fusion protein, was a generous gift of Catharina Svensson (Uppsala University, Sweden). gal4SV-40-luciferase reporter vect or carried binding sites for th e DNA binding domain of the yeast GAL4 protein upstream of the constitutively active SV-40 promoter/enhancer (a generous gift from Dr. Liao, University of Florida). CtBP1-GST fusion construct expressed hCtBP1 as glutathione S-transferase fusion protein. To create the E-cadherin promoter reporte r construct, the region extending from 427 bases upstream of the E-cadherin transcri ptional initiation site down to 53 bases post transcriptional initiat ion site was PCR amplified usi ng adaptor primers coding for Kpn1 and Bgl2 restriction sites on both ends of the amplified product with human genomic DNA serving as a template. The polymerase chain reaction was performed at a denaturing temperature of 96 o C for 45s, annealing temperature of 60 o C for 45s, and extension temperature of 70 o C for 1min. After the amp lification, the PCR product was


13 digested with Kpn1 and Bgl2 and cloned into the pGL-3 basic luciferase reporter construct (Promega) in Bgl2 and Kpn1 cl oning sites. The promoter sequence was validated by automated DNA sequencing. The cloned region included all the putative regulatory elements of the E-cadherin promot er including two palindromic elements (Eboxes), CAAT box, and CpG island. E-cadherin promoter truncation mutants were created by PCR using specific adapter primers, carrying Kpn1 and Bgl2 restriction sites at the termini and E-cadherin +53 pGL-3 basic luciferase reporter vector as a template. Truncations were constructed so th at each subsequent truncation step towards the transcriptional start site eliminated a single cis-regulatory element. Mutations included E-BOX1 deletion; E-BOX1, CAAT box deletion; EBOX1, CAAT box, CpG island deletion; E-BOX1, C AAT BOX, CpG island, and E-BOX2 deletion. Truncation mutants were digested with Bgl2 and Kpn1 and cloned into the pGL-3 basic luciferase reporter vector. Mutations w ithin the promoter region were subjected to automated DNA sequencing and confirmed by sequence analysis. A Pnn-myc/His PEDLS AADLS mutation of the CtBP1 binding motif was c onstructed using Stratagene’s QuikChange mutagenesis kit according to the manufactur er instructions usi ng a wild type Pnnmyc/His expression vector as a template . Mutations were confirmed by sequence analysis. In vitro Binding Assays Glutathione S-transferase (GST) and GST-CtBP1 proteins expressed in Escherichia coli strain BL21-Gold(DE3) (Stratag ene) were immobilized on 30l of agarose-conjugated glutathione in the presence of HEMGN buffer (25mM HEPES, pH 7.4, 0.1mM EDTA, 100mM KCl, 12.5mM MgCl 2 , 0.1% NP-40, 10% Glycerol, 2mM


14 PMSF and Complete cocktail of protease inhibitors (Roche)). Glutathione immobilized GST and GST-CtBP1 proteins were then inc ubated with 50l of s-HeLa nuclear extract enriched for Pnn-Flag-HA protein in the presence of 500l of binding buffer (20mM HEPES, pH7.9, 180mM NaCl, 1.5mM MgCl 2 , 0.2mM EDTA, 25% Glycerol, 2mM PMSF and Complete cocktail of protease inhibitors (Roche)) for 3 hours with rotation at 4 o C. After washing with binding buffer 3 times, beads were resuspended in 20l of SDS loading buffer and the associated proteins were resolved on 8%SDS-PAGE followed by western blotting analysis using mouse anti-Flag (Sigma-Aldrich) at 1:4000 dilution and mouse anti-GST (Zymed) at 1:4000 dilution. For experiments using NADH (Sigma-Aldrich 340-125), the compound was added to the binding buffer before addition of the nuclear lysate. Pnn and CtBP1-Flag Complex Isolation To determine if Pnn interacts with nuclea r CtBP1 we utilized sHeLa cells stably expressing human CtBP1 tagged with both Flag and hemagglutinin (HA) epitopes at its amino terminus (POZ-CtBP1-Flag-HA). The e xpression levels of the tagged CtBP1 were comparable to that of endogenous CtBP1. The detailed procedure for complex isolation was described previously (Ogawa et al., 2002). Nuclear extract was made from 4L of cells, from which the CtBP1 complex was part ially purified by using anti-Flag M2 mAbconjugated agarose beads followed by washes until no CtBP1 is released, and elution with SDS loading buffer. For western bl otting mouse anti-Flag antibody (Covance) at 1:4000 dilution and mouse anti-Pnn 143 undiluted were used.


15 Isolation of the Endogenous Pnn-CtBP Complex To determine if Pnn and CtBP are capable of interaction at the endogenous level 200l of sHeLa nuclear lysate was brought up to 400l using equilibration buffer (20mM HEPES, pH7.9, 1.5mM MgCl 2 , 0.2mM EDTA, 10mM KCl, 25% Glycerol (v/v) (SigmaAldrich) and incubated with mouse anti-P nn 143 (hybridoma supernatant) or control supernatant for 2hrs followed by 1hr incuba tion with protein G-Se pharose Fast Flow (Sigma-Aldrich). Beads were then washed with 180mM NaCl equilibration buffer until no Pnn is released, resuspended in SD S loading buffer, and incubated at 95 0 C for 5min. Proteins were resolved on 8% SDS-PAGE followed by western blotting using anti-Pnn 143 mAb undiluted, anti-CtBP1 and anti-Ct BP2 mAb (Signal Transduction) at 1:4000 dilution. Co-immunoprecipitaton and Immunoblotting HEK293 cells were grown to confluence in 10 cm dishes and transiently transfected with 10g of the expression pl asmids Pnn-myc/His or Pnn-AADLS-myc/His and CtBP1-Flag using Lipofectamine r eagent (Life Technologies) according to manufacturer’s instructions. Cells were fu rther lysed in plates us ing 1ml of RIPA buffer (150mM NaCl, 1% NP-40, 0.5% DOC, 0.1% SD S, 50mM Tris pH 8.0, 2mM PMSF and Complete cocktail of protease inhibitors (Roc he)) and collected in microcentrifuge tubes where cell lysates were purged 10 times through the 27 gauge syringe, followed by a 30 minute incubation with the gentle rotation at 4 o C. After centrifugation at 4 o C for 30 min at 14.000g, the supernatant was used fo r co-immunoprecipitation using 30l of antimyc beads (Sigma-Aldrich). Following overn ight incubation with rotation, the beads were washed 4 times with 500l of RIPA buf fer using micro-bio spin chromatography columns; immunoprecipitates were eluted with SDS sample buffer, incubated for 5min at


16 95 0 C, and fractionated by electrophoresis on 8%SDS-PAGE, followed by western blotting. Antibodies for immunodetection we re anti-myc mAb at 1:4000 dilution and anti-Flag mAb at 1:4000 dilution. Cell Fixation and Immunofluorescence Cells grown on coverslips were washed th ree times for 5 minutes in PBS pH 7.4. Cells then were fixed in acetone at –20 0 C for 2 min followed by three 5 minute washes in PBS. After the washes coverslips were in cubated with primary antibodies for 1 hour at 25 0 C followed by three 5 minute PBS washes. Secondary antibodies conjugated to Cy3 (Chemicon) or Alexa 488 (Chemicon) diluted to 1:2000 were applied to coverslips for1 hour followed by three 5 minute washes in PB S. Coverslips were then mounted on the Vectashield mounting media with DAPI (Vector Laboratories) and cells were visualized using a fluorescent microscope (DM IRBE; Leica). For the immunostaining, the following primary antibodies were used: mouse anti-Flag (Sigma-Aldrich) at 1:500 dilution, rabbit anti-Flag (Sig ma-Aldrich) at 1:500 diluti on, mouse anti-SR proteins (Zymed) at 1:500 dilution, mouse anti-Pnn143 (hybridoma supernatant) undiluted, mouse anti-CtBP1 (Transduction laboratorie s) at 1:500 dilution, mouse anti-NuMa (Transduction laboratories) at 1:500 d ilution, mouse anti-PCNA (Transduction laboratories) at 1:500 dilution, mouse anti-s m proteins (NeoMarkers) at 1:500 dilution. RNA Interference Pnn and CtBP1 RNA interference were carried out using targeting vectors expressing short hairpin RNAs directed agai nst Pnn or CtBP1 mRNA coding regions as previously described (Shi et al., 2003). Pnn shRNA expressing vect or was constructed by inserting an inverted repeat of Pnn speci fic 21 nucleotide sequences into a plasmid BS/U6 harboring PolIII U6 promoter as de scribed (Sui et al ., 2002). The sequences


17 corresponding to the shRNA were analyzed by BLAST Search to eliminate any possibility of sequence homology to other genes. The first fragment of an inverted repeat was cloned into a BS/U6 vector as an ApaI-HindIII insert. The second fragment, also containing six nucleotides as a loop and five T’s as the transc riptional termination signal, was then inserted into the HindIII and Eco RI sites. The coding sequences were GGTAAGGTGGCTCAGCGAGAGA (sense) and AGCTTCTCTCGCTGAGCCACCTTA CC CTTTTTG (antisense). CtBP1 shRNA vector was based on shRNA vector harboring the PolIII U6 promoter (Shi et al., 2003). Cells were seeded on coverslips at 40% confluence one day prior to transfection in 6 well dishes. Transfections were carri ed out using 3l of 1mg/ml 25kDa branched PEI (Sigma-Aldrich) and 1g of shRNAi v ector per well. 48 hours post-transfection protein expression and distribut ion was assayed by immunofl uorescence as described in the previous section. Dual Luciferase Reporter Assays The day prior to transfection cells were trypsinized and plated at 60-70% density in a 24 or 6 well plate using 500l or 2ml of standard DMEM media without antibiotics, respectively. Optimal quanti ties of the reporter DNA were 0.1g of E-cadherin and gal4SV-40 reporters per well and the Renilla re porter construct at 0.01g per well, which were empirically determined in the pilot experi ments. The Renilla construct served as an internal control for transfection efficiency. The concentration of the expression vectors varied depending on the experimental design. The amount of DNA in each transfection reaction was equalized using an empty expres sion vector to cont rol for non-specific effects on the luciferase expr ession. After transfection cell s were incubated for 24 hrs and assayed for the luciferase e xpression using the Dual-Luciferase reporter assay


18 system (Promega). All results are shown as the average +/standard deviation of three independent experiments. Results Pnn Associates with CtBP1 in vitro and in vivo In order to assay the inte raction of Pnn with CtBP1, s-HeLa POZ-N-Pnn-Flag-HA nuclear lysates containing epitope-tagged Pnn (Pnn-Flag-HA) were incubated with a CtBP1-GST fusion protein or with GST alone. Immobilized glutathi one was utilized to capture CtBP1 and associated proteins. West ern blots of pulled-dow n material revealed the presence of Pnn in the CtBP1-GST containing fractions but not in the fractions containing GST alone (Figure 1A). Consistent with many other PXDLS mo tif-containing CtBP1-binding proteins (Kumar et al., 2002; Zhang et al., 2002) , a ssociation of Pnn and CtBP1 was noticeably enhanced with increasing concentrations of NADH. These in vitro data suggest that Pnn is indeed capable of binding to CtBP1 and NADH facilitated the binding. To assess whether the Pnn-CtBP1 interaction could be detected in vivo , nuclear extracts from either POZ-CtBP1-Flag-HA sHeLa cells or from sHeLa cells carrying empty POZ vector were incuba ted with anti-Flag antibody-c onjugated agarose beads. After washing, the CtBP1-Flag-HA and associated proteins were eluted from the beads. Pnn was found associated with the Flag elut es in sHeLa-CtBP1-Flag-HA nuclear extracts and not sHeLa control extracts (Figure1B). Furthermore, endogenous CtBP1 and CtBP2 can be co-immunoprecipitated with anti-Pnn antibody from sHeLa nuclear extracts (Figure1C). Together these data suggest that Pnn interacts w ith CtBP1 and CtBP2 in vivo .


19 Next, we asked whether Pnn interacts with CtBP1 through the putative CtBP1 binding motif PEDLS. Co-transfections of HEK293 cells with CtBP1-Flag along with myc-tagged Pnn or myc-tagged Pnn mutated to contain the seque nce AADLS instead of the PEDLS, revealed that the Pnn-CtBP1 interaction was indeed dependent on the PEDLS motif (Figure 1D). Figure 1. Pnn interacts with co-repressor CtBP1 in vitro and in vivo . A. CtBP1-GST fusion protein or GST alone immobilized on glutathione agarose beads were incubated with POZ-N-Pnn-Flag-HA nuclear extracts. The bound material was eluted and evaluated for the presen ce of Pnn via western blotting with the anti-Flag antibody. Pnn was found associated with CtBP1-GST beads but not GST-alone (center panel). CtBP1GST fusion protein immobilized on glutathione agarose beads was incubated with s-HeLa POZ-N-Pnn-Flag-HA nuclear extract in the presence of incr easing concentrations of NADH. Bound material was evaluated for the presence of Pnn using anti-Flag antibody. Increased concentration of NADH resulte d in the increased association of CtBP1 with Pnn (right panel). B. CtBP1associated complex of proteins was obtained by passing POZ-CtBP 1-Flag-HA s-HeLa nuclear extract or sHeLa cell extract over a Flag-M2 column, follo wed by washes and elution with SDS loading buffer. Flag bound material from sHeLa CtBP1-Flag-HA nuclear extracts but not sHeLa cont rol nuclear extracts showed the presence of Pnn. C. Pnn-associated proteins were obt ained by co-immunoprecipitation from sHeLa nuclear lysate us ing anti-Pnn 143 mouse hybridoma supernatant or control supernatant. Co-immunopreciita tes were immunostained for CtBP1 and CtBP2. Both, CtBP1 and CtBP2 we re found to associate with Pnn. D. HEK293 cells were transiently co-trans fected with CtBP1-Flag expression


20 vector and Pnn-myc/His construct or Pnn-AADLS-myc/His mutant. Coimmunoprecipitations were performed using anti-myc agarose, followed by western blotting with anti-myc and antiFlag antibodies. Wild type but not Pnn-AADLS was capable of interacting with CtBP1. In addition, the western blots of Pnn-associated material revealed the presence of a higher molecular weight species of CtBP1, perhaps indicative of a post-translationally modified CtBP1. The higher molecular weight CtBP1 has been detected in other pulldown experiments raising the possibility that the modified CtBP1 may exhibit increased affinity for PXDLS-containing proteins (Bar nes et al., 2003; Kagey et al., 2003; Lin et al., 2003). Nuclear Distribution of Pnn is Altered in Response to CtBP1 Overexpression In the nucleus, Pnn has been shown to concentrate in nuclear speckles and localize more diffusely throughout the in terchromatin space (Brandner et al., 1997; Ouyang, 1999; Zimowska et al., 2003), while CtBP1 is found throughout the nucleoplasm. Epithelial cells (HCET) transf ected with CtBP1 exhibited a more diffuse nuclear distribution of Pnn and increased cy tosolic immunostaining for Pnn (Figure 2). Similar observations were made in s-HeLa cells constructed to express exogenous CtBP1 at 1:1 ratio to the endogenous CtBP1 as described (Shi et al., 2003). Surprisingly, immunostaining with an anti body that reacts with many SR -proteins revealed that exogenous CtBP1 expression in HCET’s and He La’s resulted in a dispersal of SR immunostaining from the nuclear speckles (F igure 2). These data suggest that the perturbation of the interac tion of Pnn with CtBP1 may have significant consequences on not only Pnn but also on the overall or ganization of SR-containing proteins. We next asked whether reducing the e xpression of either Pnn or CtBP1 by utilizing vector based RNA interference resulte d in a change in the distribution of the


21 other component. While Pnn RNAi effectivel y lowered the levels of Pnn, little change was observed in distribution of CtBP1 (Figure 3B). Figure 2. Pnn exhibits nuclear redistribution in response to CtBP1 overexpression.(1,2) sHeLa cells containing empty POZ vector (1) or s-HeLa cells stably expressing CtBP1-Flag-HA fusion protein (POZ-CtBP1-Flag-HA) (2) were immunostained for CtBP1 using rabb it anti-Flag antibody (1a,2a) and 143 antibody against the endogenous Pnn (1b,2b). (1c,2c) merged DAPI nuclear stain (blue), Pnn (red), CtBP1 (green). (5,6) HCET cells were transiently transfected with the CtBP1-Flag expre ssion vector and localization of CtBP1 and Pnn was assessed by immunofluorescence using rabbit an ti-Flag antibody (5a,6a) and 143 antibody against the e ndogenous Pnn (5b,6b). (5c,6c) merged DAPI nuclear stain (blue), Pnn (green), CtBP1 (red). (3,4) Same as (1,2) and (7,8) same as (5,6), except, instead of Pnn, cells were immunostained for the endogenous SR proteins using mouse an ti-SR proteins antibody. Both Pnn and SR proteins demonstrated nuclear re distribution in response to the CtBP1 overexpression. Bar 10m.


22 In contrast, lowering CtBP1 levels result ed in decreased CtBP1 expression and a dramatic change in Pnn’s di stribution. In CtBP1-knockdown cells, sHeLa (Figure3A), as well as HCET cells (Figure 3B), Pnn exhibite d a diffuse nucleoplasmic distribution with concurrent increase in cytoplasmic levels. Interestingly and s eemingly paradoxical, both increasing and decreasing CtBP1 expression resulted in a si milar distribution change of Pnn, perhaps reflective of th e importance of balanced expression of CtBP1 and Pnn. To verify the specificity of the CtBP1 RNAi-dependent Pnn redistribution and to eliminate the possibility of general RNAi-m ediated cellular toxicity, which might result in Pnn redistribution, we performed CtBP1 R NAi and immunostained for several nuclear antigens including SR proteins, sm splicing factors, NuMa and PCNA antigens. In contrast to Pnn, CtBP1 RNAi did not exert dr astic effect on the di stribution of SR, sm, NuMa, and PCNA antigens, supporting the spec ificity of CtBP1 RNAi on Pnn nuclear redistribution (3C). Interestingly, in some instances we observed accumulation of SR proteins in the nuclear speckles in response to CtBP1 RNAi bringing the possibility that levels of CtBP1 might, directly or indirectly, regulat e splicing events. Pnn Upregulates the E-cadherin Promoter Activity The effect of Pnn on E-cadherin gene act ivity was originally described using EcR293 hPnnGFP inducible cell lines, in whic h induction of Pnn expression resulted in upregulation of E-cadherin at bo th message and protein levels (Shi, 2001). In addition, we have demonstrated that RNAi-mediated Pnn kn ockdown results in the decreased levels of E-cadherin, leading to the disruption of adherence junctions (Joo et al, 2005). These findings raised the possibility that Pnn may pa rticipate in functional interaction(s) with the transcriptional machinery, which contribut e to the control of the E-cadherin gene expression, such as interaction with the co-repressor CtBP1.


23 Because E-cadherin promoter is a k nown target for the CtBP1-mediated repression (Grooteclaes et al., 2003; Grooteclaes and Frisch, 2000; Shi et al., 2003; Zhang et al., 2002), we wished to determin e if Pnn-CtBP1 interaction has functional consequences on the E-cadherin promoter activity. Figure 3. Nuclear distribution of Pnn is alte red in response to CtBP1 RNAi. A. (1-4) sHeLa cells were co-transfected with a CtBP1 RNAi vector and the GFP vector, (2,4), or the GFP vector al one (1,3). CtBP1 staining (1b,2b) Pnn staining (3b,4b). (1c-4c) merged DAPI nuc lear stain (blue), GFP (green), Pnn or CtBP (red). B. (1,2) HCET cells were co-transfected with a CtBP1 RNAi vector and GFP. Cells were immunost ained for CtBP1 (1b) and Pnn (2b). Cells, which received CtBP1 RNAi vector exhibited drastic Pnn redistribution from nuclear speckles. (3,4) HCET cells were co-transfected with Pnn RNAi vector and a GFP vector. Cells were immunostained for Pnn (3b) or CtBP (4b). (1c-4c) merged DAPI nuclear stai n (blue), GFP (green), Pnn or CtBP (red). Pnn knockdown did not result in CtBP1 redistribution. C. HCET cells were co-transfected with a CtBP1 RNAi vector and GFP vector. Cells were immunostained with anti-Pnn (1b), anti-SR proteins (2b), anti-sm proteins (3b), anti-NuMa (4b), and anti-PCNA (5b). (1c-5c) merged DAPI nuclear


24 stain (blue), GFP (green), Pnn, SR, sm, NuMa, PCNA (red). Arrowheads indicate the transfecte d cells. Bar 10m. E-cadherin luciferase promoter (-427+53) reporter vector was constructed and assayed in the presence of ove rexpressed Pnn in the HEK293 cell line. Co-transfection of the Pnn expression vector resulted in upreg ulation of the E-cadherin promoter reporter activity, suggesting that Pnn can activate the E-cadherin promoter (Figure 4A). Figure 3. Continued. To explore the effect of Pnn expressi on on the E-cadherin promoter activity in different cellular contexts, we co-transfect ed Pnn expression vector and E-cadherin reporter construct into inva sive uveal melanoma cells, Mu M-2C, (Figure 4B) and the epithelial cell line MDCK (Figure 4C). Pnn was able to upregulate E-cadherin prom oter reporter activity in all the cell lines, thus providing evidence that Pnn possesses a general positive regulatory function.


25 Interestingly, the greatest relative increase was seen in MDCK cells which exhibit more robust morphological and physiological characte ristics of epithelial cells than do the epithelial derived HEK293’s or the MuM-2C’s, suggestive of a context-dependent degree of response. Figure 4. Pnn is capable of upr egulation of the E-cadherin pr omoter activity, independent of cell type origin. MDCK, uveal melanoma MuM-2C, and HEK293 cells were transiently co-transfected with 100ng of E-cadherin luciferase reporter construct with either 500ng of the Pnn-GFP expression vector or the GFP vector alone. The numerical data obtain ed from the luciferase readings was normalized to the values of the GFP v ector alone control transfectons, which represented a value of 1. In HEK293 and MuM-2C cells E-cadherin promoter exhibited 2 and 2.5-fold increase in lu ciferase activity in response to Pnn expression respectively, whereas in MD CK’s 4-fold increase was observed. Full-length E-cadherin Promoter is Requi red for Pnn-dependent Upregulation of Transcription In order to assess the requirement of the immediate region of the E-cadherin promoter for the Pnn-induced activation, we created E-cadherin promoter truncations and utilized them in the luciferase report er assays in context of exogenous Pnn overexpression (Figure 5). E-cadherin promoter contains several key cis -regulatory elements implicated in modulation and cell type specifi c patterns of E-cadherin gene expression. They include E-boxes 1 CAGGTG (-78) and 2 CACCTG (-28) , conserved CCAAT box (-65) and CpG island (-52 ) which contribute to the epithelial specific E-cadherin gene expression


26 (Comijn et al., 2001; Giroldi et al., 1997; Graff et al., 2000; Henni g et al., 1995; Hennig et al., 1996; Hosono et al., 2000). Figure 5. Identification of the Pnn-responsive regions in the E-cadherin promoter. Each truncation of the E-cadherin promoter luci ferase reporter is represented in the diagram, with indicated distances from the transcriptional initiation site. Bars on the right represent luciferase activity of each of the truncation construct in response to Pnn-myc/His expression (Pnn) or expression of the myc/His vector alone (vector). 100ng of each pr omoter construct was transiently cotransfected into HEK293 cells along with 500ng of the Pnn-myc/His expression vector or myc/His vector alone. The nume rical data obtained from the luciferase readings was normalized to the values of the full-length Ecadherin promoter (-427+53) activity when co-transfected with the myc/His vector alone, which represented a valu e of 1. E-box1 and E-box2 regions, as well as the region adjacent to the transc riptional start site , demonstrated an increased activity in the presence of Pnn. The modular structure of the E-cad herin gene control region allows interdependent regulation by trans -acting transcriptional factors, which utilize common or adjacent cis -regulatory regions of the promoter. When co-transfected with the Pnn expression vector, +53 E-cadherin promoter was the most responsive to th e Pnn-induced transc riptional activation, suggesting that the increase in the transcriptional output trigge red by Pnn requires an


27 intact, immediate E-cadherin promoter region (F igure 5). In particular, deletion of the distal promoter region, which includes E-box1, abolished Pnn-depende nt upregulation of the E-cadherin promoter activity, indicating that this region may be involved in the enhancement of the E-cadherin promoter transcriptional output by Pnn. We are currently investigating the possibility that Pnn is a part of transcriptional complexes occupying specific sites on the E-cadherin promoter. Pnn De-represses CtBP1-mediated Sile ncing of the E-cadherin Promoter In order to establish the functional relationship between CtBP1 and Pnn at the Ecadherin promoter region, HEK 293 cells were co-transfected with the E-cadherin reporter construct and the CtBP1 vector, with or without the wild type (wt) Pnn construct. As predicted (Grooteclaes et al ., 2003; Grooteclaes and Frisch, 2000; Shi et al., 2003; Zhang et al., 2002), CtBP1 expression resulted in dose dependent repr ession of E-cadherin promoter activity suggesting that CtBP1 serves as a transcriptional co-repressor in the Ecadherin promoter context (Figure 6-A). While CtBP1 was capable of repressing Ecadherin expression, increasing c oncentrations of the Pnn expr ession vector resulted in a dose-dependent reversal of Ct BP1-mediated repression of the reporter activity. These data are consistent with the proposal that Pnn may function as a transcriptional activator indirectly by binding to nuclear CtBP1, re sulting in transactiv ation of E-cadherin transcription. Pnn1-421, a Pnn truncation that expresses Pnn lacking it’s C-terminal portion, which includes the re gion containing the PEDLS motif , did not display reversal of CtBP1mediated re pression (Figure 6-B). Furthermore, cells co-transfected with Pnn-AADLS also exhibited moderate relief of CtBP1-mediated repression. However th e effect of Pnn-AADLS was less dramatic


28 than the full deletion of the C-terminus of Pnn, raising the possibility that other regions of Pnn may have effects on transcrip tional regulation of E-cadherin. Figure 6. Wild type Pnn, but not the truncated or mutant form of Pnn, is capable of relieving CtBP1-mediated repression of the E-cadherin promoter.6a. HEK293 cells were co-transfected with E-cadhe rin luciferase repo rter vector and increasing amounts of the CtBP1-Flag vector (.1g and 1g). Parallel transfection reactions cont ained increasing concentra tions of Pnn-GFP (.01g, .1g, 1g) in the presence of 1g of the CtBP1-Flag vector. The same experiment was performed using Pnn-G FP1-421, a Pnn truncation mutant that lacks CtBP1 binding site (6b). 6c. HEK293 cells were transiently cotransfected with E-cadherin luciferase reporter vector a nd increasing amounts of the CtBP1-Flag vector (.1g and 1g). Parallel transfection reactions contained 1g of the CtBP1-Flag vector and 1g of the Pnn-myc/His or PnnAADLS-myc/His vector. The numerical da ta obtained from the luciferase readings was normalized to the values of full-length E-cadherin reporter in the absence of CtBP1 or Pnn constructs, which represented a value of 1. In contrast to wild type Pnn, Pnn1-421 de letion construct was not capable of relieving CtBP1-mediated repression of the E-cadherin promoter activity. Pnn PE AA mutant only partially re lieved the promoter activity. Further evidence of Pnn f unctioning by interacting with CtBP1 was obtained in the reporter assays in which Pnn-GALBD was tethered to Gal4 binding sites immediately upstream of the constitutively expressed viral SV-40 promoter/enhancer. Only in the presence of Pnn-GALBD could CtBP1 repre ss SV40 promoter activity (Figure 7 left panel). However, co-expression of wtPnn, but not Pnn-AADLS, effectively blocked the Pnn-GALBD/CtBP1 repression (Fi gure 7 right panel). In a r eciprocal experiment CtBP1GALBD, when tethered upstream of the SV40 pr omoter, was able to repress the activity


29 of the SV40 promoter (Figure 8). However, in this scenario, only the co-expression of wtPnn, but not Pnn-AADLS, was capable of blocking CtBP1-mediated repression. Figure 7. Pnn can recruit CtBP1 to a heterol ogous promoter, leading to a CtBP1-mediated repression of the promoter activity. left panel. HEK293 cells were cotransfected with the luciferase repo rter construct gal4-SV40-luc and the following constructs: 100ng of the P nn-GALBD fusion construct alone, 1g of the CtBP1-Flag expression vector in the presence or absence of the PnnGALBD fusion construct (100ng). gal4 -SV-40-luc activity was subject to CtBP1-mediated repression only in th e presence of the Pnn-GALBD fusion construct, indicating that Pnn-CtBP1 interaction was res ponsible for this effect. right panel. HEK293 cells were co-transfected with the gal4-SV40-luc reporter, 100ng of the Pnn-GALBD fusion construct, and increasing concentrations of CtBP1-Flag (100ng, 1g). Transfection reactions also contained 1g of the wild type Pnn-myc/ His or mutant form of this vector Pnn-AADLS-myc/His, carrying PE AA substitution in the CtBP1 binding motif. The numerical data obtained from the luciferase readings was normalized to the value of gal4-SV40luc reporter in the presence of PnnGALBD only, which represented a value of 1. Wild type, but not mutant form of Pnn, was capable of competition with Pnn-GALBD, thereby relieving CtBP1-mediated repression of the reporter activity. We speculate that wtPnn may bind to th e tethered CtBP1 and occupy or block a site on CtBP1 that is required for the recru itment of additional protein(s) that enable CtBP1 to repress transcription.


30 Figure 8. Pnn, but not Pnn mutant carrying a mutation in the CtBP1 binding motif, is capable of relieving CtBP1-mediated re pression of a heterologous promoter activity. HCET cells were transfected with the gal4-SV40-luc reporter construct carrying GAL4 binding sites upstream of the SV40 promoter/enhancer. Transfection reac tions also included 1g of CtBP1GALBD fusion vector alone, and in th e presence of 1g of Pnn-myc/His construct (upper panel) or its mu tant Pnn-AADLS-myc/His, carrying PE AA substitution in the PEDLS CtBP1 binding motif (lower panel). The numerical data obtained from the luciferase readi ngs was normalized to the value of the gal4-SV40-luc reporter alone, which repres ented a value of 1. Wild type, but not mutant form of Pnn, was capable of relieving CtBP1-mediated repression of the promoter activity, as depicted in the diagram. Discussion In our experiments we demonstrated that Pnn interacted with transcriptional corepressor CtBP1 in vitro and in vivo . As a consequence of th eir interaction, exogenously expressed Pnn was capable of reversing CtBP1 mediated repr ession of E-cadherin promoter, providing some mechanistic insights into the functional in teraction of Pnn and CtBP1 as well as its effect on gene transcription. Our data regarding Pnn-CtBP1 interacti on places Pnn as a promising candidate modifier of the CtBP1 silencing function. This is in accord with similar reports describing adenoviral E1A protein, responsib le for the acquisiti on of the epithelial phenotype by invasive tumors by binding CtBP and preventing its repressive influence on the target genes (Frisch, 1994; Schaeper et al., 1995; Zhang et al., 2002; Zhang et al.,


31 2000). It is tempting to speculate that by binding to CtBP E1A exploits preexisting functional protein-protein interactions of the host cellular transcriptional machinery. CtBP family members interact with, and m odify, the activity of a large number of proteins with key roles in development, di fferentiation, oncogenesis, cell cycle control, and apoptosis. Interaction of CtBP with the C-term inus of E1A antagonizes a transcriptional activation mediated by the N-terminus of E1A (Chinnadurai, 2002; Schaeper et al., 1998; Sundqvist et al., 2001). In addition, co -repressor CtBP has been shown to associate with the DNA binding tr anscriptional repressors ZEB1, and SIP1 (ZEB2), histone deacetylases HDAC 1 and 2, re lated histone methyltransferases G9a and Eu-HMTase1 (EuHMT), two chromodomain containing proteins HPC2 and CDYL, transcriptional repressor Co REST, and its related prot ein KIAA1343 (Comijn et al., 2001; Postigo and Dean, 1999; Shi et al., 2003). Taken together these data suggest that CtBP-mediated repression encompasses combinatorial events involving DNA binding repressors and coordinated activity of the histone deacetylases and methyltransferases. Consistent with the fact that CtBP binds a diverse array of transcriptional regulators, CtBP deficiency dramatically compromised mouse embryo-wide development as a result of genetic and biochemical perturbations (Hildebrand and Soriano, 2002). Therefore, by binding to CtBP1, Pnn is likely to alter CtBP1-dependent sile ncing functions of transcriptional repression machinery, resulti ng in activation of the CtBP1 target genes. In our experiments Pnn interacted with Ct BP1 in the nucleus and exhibited drastic nuclear redistribution in re sponse the CtBP1 overexpression or RNAi mediated CtBP1 knockdown, supportive of Pnn/CtBP1 functi onal association. Although the data presented here pertains to CtBP1, we expect the same effect on P nn nuclear distribution


32 in response to perturbations of the CtBP2 le vels. Both, CtBP1 and CtBP2 are found to be associated with Pnn in the nucleus and, ther efore, can possibly exert similar effect on Pnn’s nuclear localization. Interestingly, Ct BP1 and CtBP2 can be SUMOylated (Kagey et al., 2003; Lin et al., 2003) and, as in cas e of CtBP1, exhibit SUMOylation-dependent subcellular redistribution. However, in teraction of CtBP1 w ith PXDLS-containing proteins is not affected by the CtBP1 SUMO ylation status (Lin et al., 2003), suggesting that SUMOylation does not affect CtBP1 asso ciation with the transcriptional proteins, although it can possibly dictate the amount of CtBP1 present in the nucleus. As with the majority of the interactions between CtBP and its interacting partners, Pnn’s interaction with CtBP1, is mediated through the PXDLS motif within the Cterminus of Pnn. In addition, NADH facilitates Pnn-CtBP1 binding. It has been demonstrated that NAD + / NADH-induced conformational change in CtBP family members results in an increase in their a ffinity for PXDLS motif-containing proteins (Chinnadurai, 2002; Kumar et al., 2002; Marmorstein, 2002; Zhang et al., 2002). It is possible that the ability of NADH to stimulat e CtBP oligomerization could contribute to the enhanced binding to PXDLScontaining proteins such as Pnn. Indeed, Nardini et al. recently determined the structural basis of the increased affinity of CtBP for NADH (Nardini et al., 2003). They proposed that nucleotide binding indu ces a conformational change that promotes CtBP dimerization, which is essential for co-repressor activity. Our finding that NADH stimulated Pnn and CtBP1 binding implies that physiological state of the cell can modulate Pnn-CtBP1 interaction, which is consis tent with the previously reported data describing CtBP interactions with E1A and the transcriptional repressor ZEB1 (Kumar et al., 2002; Zhang et al., 2002).


33 We show here that Pnn is capable of the reversal of CtBP1-me diated repression of the E-cadherin promoter activity. Accord ing to the pull-down experiments it appears that approximately 1-2% of nuclear CtBP is bound to Pnn, suggesting that change in Pnn levels is unlikely to drastically affect availa ble nuclear CtBP pool. Therefore, rather than acting as a CtBP sink we suggest that Pnn may be recrui ted to the transcriptional regulatory complexes and, through CtBP, alter CtBP-mediated repressi on, contributing in a positive manner to E-cadherin transcripti onal regulation. Thus, it is tempting to speculate that Pnn-CtBP interaction reflect the existence of the bridging machinery, a framework for coupling of transcription and mRNA processing. It remains to be established how Pnn-CtBP interaction with th e transcriptional machinery will affect Pnndependent splicing events. The E-cadherin promoter is a well-cha racterized target for CtBP-mediated transcriptional repression, which may be manifested in the course of epithelialmesenchymal transitions associated with ma lignant transformation of tumors and cell migration during embryogenesis. Recent attemp ts to elucidate the nature of CtBP’s repressive functions resulted in a series of crucial findings linking CtBP-mediated Ecadherin silencing to the local chromatin stat e at the promoter region as well as to the intranuclear NAD + /NADH ratio, establishing CtBP’s role as an important regulator of Ecadherin activity in context of the local meta bolic and biochemical environment (Shi et al., 2003; Zhang et al., 2002). Therefore, it has been speculated that because CtBP appears to play important roles in normal development and in oncogenesis, expression of CtBP target genes such as Ecadherin may be influenced by the cellular redox status. In this context, NADH-dependent transcriptional properties of CtBP1, as well as its Pnn


34 binding specificities, imposes significant c onsequences of physiological changes on the nature of Pnn-CtBP1 interaction as relate d to transcriptional regulation. Significantly, Pnn was first characterized as a component of the complex containing dehydrogenase activity (Bellavite et al ., 1990), which may provide the foundation for the underlying mechanism governing the Pnn and CtBP1 func tional relationship. Given the assumption that Pnn and CtBP1 can be responsive to the common and unique factor, which defines patterns of cellular metabolism, such as NAD + /NADH, it is plausible to speculate that the general metabolic spectrum of enzymatic activities can evok e opposite responses depending on the innate characteris tics of transcriptional regulator. While the role of Pnn in mRNA splicing activities has been described (Li et al., 2003; Wang et al., 2002; Zimowska et al., 2003), its function as a transcriptional regulator has not been explored in detail. Our data identified Pnn as a transcriptional regulator, which functionally antagonizes the effect of co-repressor-mediated transcriptional silenci ng. Interestingly, Dellaire et al. have reported that PRP4 kinase, which binds to Pnn, is part of an N-CoR-c ontaining deacetylase complex (Dellaire et al., 2002), again potentially linking Pnn to transc riptional regulation. We suggest that Pnn may belong to an emerging family of prot eins involved in nuclear functions of transcription and mRNA processing. Taken t ogether these data raise an interesting possibility in connection with the transcriptional regulation and splicing activity. It is known that splicing activities are coupled to transcripti on through the interaction of transcriptional factors and constituents of the splicing complexes (Goldstrohm et al., 2001; Goldstrohm et al., 2001; Reed, 2003). As a result of the functional interplay between the factors responsible for orchestrated gene activation, such as chromatin


35 remodeling enzymes, transcriptional regulat ors and splicing machinery, gene expression acquires the necessary flexibility and synchr onization of regulatory events due to the possibility of multiple control points (Dellaire et al., 2002; Martinez et al., 2001; Shin and Manley, 2002). It remains to be establis hed whether Pnn-associated modulation of transcriptional output and splicing activities represents two distinct mechanisms, which Pnn is a part of, or the effect of a single functional regulatory unit. In any case, PnnCtBP1 interaction at the E-cadhe rin promoter provides a novel insight into the nature of gene regulation through the combinatorial as sociation with repression factors and mRNA splicing machinery, which supplements transcriptional mechanisms with the necessary means for a flexible, multilayered control of gene expression. Note The work presented here was published in an article “Alpatov, R., Munguba, G. C., Caton, P., Joo, J. H., Shi, Y., Hunt, M. E., and Sugrue, S. P. (2004). Nuclear speckleassociated protein Pnn/DRS binds to the tran scriptional corepressor CtBP and relieves CtBP-mediated repression of the E-cadherin gene. Mo l. Cell. Biol. 24, 10223-35.” We thank Dr. Liao (University of Flor ida) for the gal4-SV40-luc reporter and Catharina Svensson (Uppsala University, Sweeden) for the CtBP1-GALBD expression vector. We also thank Todd Barnash a nd Lynda Hanssen for graphics support.


CHAPTER 3 PNN/DRS, A CTBP AND RNA POLYMERASE II-ASSOCIATED FACTOR, REGULATES E-CADHERIN GENE TRANSCRIPTION AND MRNA PROCESSING. Introduction Gene expression is a complex process involving coordinate d events at the promoter region and pre-mRNA pr ocessing. A number of studi es have demonstrated the connection between co-activator or co -repressor complexes operating at the gene promoters and mRNA splicing (A uboeuf et al., 2005; Auboeuf et al., 2004; Auboeuf et al., 2002; Dellaire et al., 2002; Dowhan et al., 2005; Nagai et al., 2004; Tai et al., 2003; Zhang et al., 2003). Futhermore, the coupling mechanisms of promoter-associated regulators and mRNA splicing complexes depe nd on the promoter architecture and may involve RNA polymerase II-dependent spec ific interactions (Auboeuf et al., 2002; Cramer et al., 1999; Cramer et al., 1997; Della ire et al., 2002; Paga ni et al., 2003; Simic et al., 2003; Smith et al., 2004; Tai et al., 2003; Tasic et al., 2002). Therefore promoterassociated protein interac tions can be important dete rminants in mRNA processing events. CtBP is a transcriptional co-repressor that is targeted to the E-cadherin promoter. CtBP associates with transcri ptional repressor complexes invo lved in gene regulation in various developmental, oncogenic, and apoptotic contexts (Barnes et al., 2003; Chinnadurai, 2002; Grooteclaes et al., 2003; H ildebrand and Soriano, 200 2; Izutsu et al., 2001; Koipally and Georgopoulos, 2000; Melh uish and Wotton, 2000; Meloni et al., 1999; Phippen et al., 2000; Postigo and Dean, 2000; Postigo and Dean, 1999; Sewalt et 36


37 al., 1999; Zhang et al., 2005; Zh ang et al., 2003). CtBP is targeted to promoters via sequence-specific DNA-binding tran scription factors, and contributes to gene silencing by recruiting complexes with histone methyltr ansferase, demethylase, and deacetylase activities (Grooteclaes and Fr isch, 2000; Nitta et al., 2005; Postigo and Dean, 1999; Shi et al., 2003; Srinivasan and At chison, 2004). For example, CtBP is recruited to the Ecadherin gene promoter by repressor ZEB (Chinnadurai, 2002; Comijn et al., 2001; Grooteclaes et al., 2003; Grooteclaes and Frisch, 2000; Pos tigo and Dean, 1999; Shi et al., 2003; Zhang et al., 2002). Similarly, co-rep ressor mSin3A has been demonstrated to interact with CtBP in vivo (Koipally and Georgopoulos, 200 0) and target E-cadherin promoter via its interaction with Snail (Pei nado et al., 2004) and as part of the CoREST complex (Grimes et al., 2000; Shi et al., 2003). Many studies have shown that CtBP possesses a conserved NADH binding motif, characteristic for D-hydroxyacid dehydrogena ses, indicating that NAD(H) may have a role in function of CtBP. Indeed, as a consequence of the NAD(H) binding, CtBPmediated repression and interact ion with the transcriptional pr oteins is regulated by levels of NAD(H), linking CtBP function to the loca l metabolic state (Balasubramanian et al., 2003; Barnes et al., 2003; Fjeld et al., 2003; Kim et al., 2005; Kumar et al., 2002; Nardini et al., 2003; Sutrias-Grau and Arnosti, 2004; Thio et al., 2004; Zhang et al., 2002). However, CtBP is also capable of tran scriptional silencing in NAD(H) independent manner (Grooteclaes et al., 2003). Pinin/DRS (Pnn/DRS) is a multifunctional protein that has been initially recognized by virtue of its ability to promote epithelia l adhesion properties of the transformed cell lines (Joo et al., 2005; Ouyang and Sugrue, 1996; Ouyang and Sugrue,


38 1992; Shi and Sugrue, 2000; Shi et al., 2000). Additional studies revealed that Pnn also influences the expression of a number of genes, hence in troducing Pnn as a potential transcriptional regulato r (Shi, 2001; Shi et al., 2000). I ndeed, later experi ments indicated that in the nucleus Pnn interacts with mR NA processing factors such as SRm300, SRp75, SRrp130, and RNPS1 at the nuclear speckles and is involved in mRNA splicing (Li et al., 2003; Sakashita et al., 2004; Wang et al., 2002; Zimowska et al ., 2003). In addition, recent data indicates that Pnn associates w ith components of the basal transcriptional machinery such as Mediator and CA150 (Sato et al., 2004; Smith et al., 2004). We have also demonstrated that Pnn modulates Ecadherin promoter activity through its NADHdependent interaction with the co-repressor CtBP (Alpatov et al., 2004). These findings connect Pnn to transcript ion-specific mechanisms. Here, we investigated the potential mechanism by which Pnn can influence Ecadherin gene expression. We demonstrate th at Pnn associates with CtBP-dependent silencing complexes and affects chromatin remodeling at the E-cadheirn promoter. We also demonstrate that Pnn interacts with the transcriptionally competent form of Pol II in a NADH-dependent manner. Finally, we show th at Pnn can positively affect E-cadherin mRNA splicing efficiency. These findings carry an intriguing implication that similar to coupling factors, such as CHD1 and nuclear receptor coregulators (Auboeuf et al., 2004; Auboeuf et al., 2002; Tai et al., 2003), Pnn is involved in promoter-dependent regulatory events as well as post-transc riptional mRNA processing. Materials and Methods Cell lines, Cell Culture and Transfections NIH-3T3hPnn-GFP cells stably expressi ng human Pnn-GFP and HEK293 cells were cultured in Dulbecco’s modified Eagle medium (DMEM; Bio Whittaker) containing


39 10% FBS (Cellgro; Mediatech) 2mM L-Glutam ine, and 200U/ml each streptomycin and penicillin (Cellgro; Mediatech ). Cells were passed with 0.1 % trypsin and 0.04% EDTA in Hanks medium. Suspensory HeLa cells (sHeLa) stably expressing Pnn-Flag-HA fusion protein (POZ-Pnn-Flag-HA) at 1:1 ratio to the endogenous Pnn were designed as described previously (54). POZ-Pnn-Flag-HA, and sHeLa cells containing empty POZ vector (POZ), were propagated in suspension at 2X10 5 confluence in Joklik’s media (Cambrex) containing 10% FBS (Cellgro;Mediatech) and 200U/ml each streptomycin and penicillin. Cells were transfected utilizing 3l of 1mg/ml 25kDa branched PEI (SigmaAldrich) per 1g of DNA. PEI and DNA we re incubated in serum free DMEM for 10 minutes in separate tubes. Following in cubation the contents of two tubes were combined, incubated for additional 10 minutes, and applied onto cells. Expression Vectors pcI-neo-PnnGFP, which expressed hPnn, was based on pcI-neo-GFP, which was previously described (Wang et al., 1996). pCMV-CtBP1-Flag, expressing hCtBP1 and pcDNA3.1-Pnn-myc/His, expressing hPnn were based on pCMV-Flag (Stratagene) and pcDNA3.1-myc/His (Invitrogen) respectively. Pnn-His-GST fusion construct was based on Pet 42b(+) vector (Novagen). ZEB1-myc vect or was a gift from Dr. Postigo and Dr. Dean (Washington University). mSin3A-myc vector was a gift from Dr. Seto (University of South Florida). RNA Polymerase II Antibodies Mouse anti-RNA polymerase II H14 antibody recognizes phosphoserine 5 version of initiation specific Pol II (Covance), m ouse anti-RNA polymerase II H5 antibody recognizes phosphoserine 2 version of elongation specific Pol II (Covance) mouse anti


40 RNA polymerase II 8WG16 antibody recognizes hypophosphorylated form of preinitiation Pol II (Covance), rabbit N-20 antibody recognizes N-terminus of Pol II (Santa Cruz). Co-immunoprecipitaton and Immunoblotting HEK293 cells were transfected with 4g of the expression plasmids ZEB1-myc or mSin3A-myc, and CtBP1-Flag in the presence or absence of 4g of Pnn-GFP. 24 hours post-transfection cells were washed with cold PBS and lysed in IP buffer (20mM HEPES, pH7.9, 200mM NaCl, 1.5mM MgCl 2 , 0.2mM EDTA, 10mM KCl, 25% Glycerol (v/v), 2mM PMSF and Complete cocktail of protease inhibitors (Roche)). Co-immunoprecipitations were performed usi ng anti-Flag (Sigma-Aldrich) or anti-myc agarose beads (Sigma-Aldrich). Immunoprecipita tes were eluted with SDS sample buffer, and fractionated by electrophoresis on 8%SDS-PAGE, followed by western blotting. Antibodies for immunodetection were an ti-myc 9B11 mAb (Cell Signalling) (1:4000 dilution), anti-Flag M2 mAb (Sigma) (1: 4000 dilution) and 143 anti-Pnn mAb undiluted. Chromatin Immunoprecipitation (ChIP) Assays ChIP was performed according to guidelines from Upstate Biotechnology. In brief, chromatin from HeLa POZ or HeLa POZ-Pnn-Flag-HA cells was sheared to contain on average 500 bp DNA fr agments. ChIPs were perfomed using the following antibodies: rabbit anti-HA (Cova nce), mouse anti-acetylated histone H4 (Covance), N20 anti-Pol II (Santa Cruz), rabbit anti-dim ethylated histone H3 at lysine9 (H3K9) (Covance). For PCR reactions primers sp anning E-cadherin prom oter region or GAPDH promoter region were utilized. PCR products were visualized on 0.5% TAE agarose gel.


41 Isolation of the Endogenous Pnn-Pol II Complex sHeLa nuclear (200l) lysates were brought up to 500l with equilibration buffer (20mM HEPES, pH7.9, 150 mM NaCl, 1.5mM MgCl 2 , 0.2mM EDTA, 10mM KCl, 25% Glycerol (v/v) and incubated with rabbit N-20 (Santa Cruz) or control IgG for 2hrs followed by 1hr incubation with protein A-Sepharose Fast Flow (Amersham Biosciences). Beads were then washed with equilibration buffer and resuspened in SDS loading buffer. Proteins were resolved on 8% SDS-PAGE followed by western blotting using anti-Pnn 143 mAb undiluted and N-20 an ti-Pol II antibody (1:500 dilution). Pnn and RNA Polymerase II Complex Isolation Nuclear extracts from sHeLa POZ-Pnn-Flag -HA cells were prepared as described previously (Ogawa et al., 2002). Nuclear extr acts from 4L of cells were incubated with ANTI-FLAG M2 mAb-conjugated agarose be ads (Sigma-Aldrich) followed by washes, until no Pnn is released, and elution with SD S loading buffer. For western blotting the following antibodies were utilized: m ouse anti-Flag antibody (Covance) (1:4000 dilution), Pol II H14 antibody (1:500 dilution) , Pol II H5 antibody (1:500 dilution) and Pol II 8WG16 (1:500 dilution). In vitro Binding Assays Glutathione S-transferase (GST ) and GST-Pnn expressed in Escherichia coli strain BL21-Gold(DE3) (Stratagene) were immobilized on agarose-conjugated glutathione in the presence of HEMGN buffer (25mM HEPES, pH 7.4, 0.1mM EDTA, 100mM KCl, 12.5mM MgCl 2 , 0.1% NP-40, 10% Glycerol , 2mM PMSF and Complete cocktail of protease inhibito rs (Roche)). Glutathione im mobilized GST and GST-Pnn were then incubated with 200l of sHeLa nuc lear extract in the presence of 300l of binding buffer (20mM HEPES, pH 7.9, 150mM NaCl, 1.5mM MgCl 2 , 0.2mM EDTA,


42 25% Glycerol, 2mM PMSF and Complete cocktail of protease inhibitors (Roche)) for 3 hours with rotation at 4 o C. After washing with binding buffer, beads were resuspended in SDS loading buffer and the associated proteins were resolved on 8%SDS-PAGE followed by western blotting using mouse an ti-CtBP1 (Signal Transduction) (1:4000 dilution), Pol II H14 antibody (1:500 diluti on) and Pol II 8WG16 (1:500 dilution). NADH (Sigma-Aldrich 340-125) was added to the binding buffer before addition of the nuclear lysate. It was also present in the binding buffer during the washing steps. Immunofluorescence Cells grown on coverslips were washed with PBS pH 7.4 and fixed in methanol at 0 C for 2 min. After the washes coverslips were incubated with primary antibodies for 1 hour at 25 0 C followed by PBS washes. Secondary antibodies conjugated to Rhodamine (Chemicon) or Alexa 488 (Chemicon) diluted to 1:2000 were applied to coverslips for1 hour followed by additional washes in PBS. Coverslips were then mounted on the Vectashield mounting media with DAPI (Vector Laboratories) and cells were visualized using a fluorescent microscope (DM IRBE; Leica). For the immunostaining, the following primary antibodies were us ed: Rabbit anti-GFP (Molecular Probes) (1:1000), Pol II H14 antibody (1:500 dilution) , Pol II H5 antibody (1:100 dilution) and Pol II 8WG16 (1:100 dilution). BromoUridine Incorporation Cells were permeabilized for 3 minutes in Permeabilization Buffer (20mM TrisHCl, pH 7.4, 5mM MgCl2, 0.5mM EGTA , 25% glycerol, 0.1% Triton X-100, 1mM PMSF, RNasin 20 units/ml) followed by incuba tion in transcription buffer (100mM KCl, 50mM Tris-HCl, pH 7.4, 10mM MgCl2, 0.5mM EGTA, 25% glycerol, 2mM ATP,


43 0.5mM CTP, 0.5mM GTP, 0.5mM BrUTP, 1mM PMSF, RNasin 20 units/ml) for 5 minutes at 37 C. Cells were then washed with 1X PBS and fixed in -20 C methanol for 2 minutes. Deconvolution Microscopy Volocity 3-Dimensional Models were ge nerated using a Leica Microscope in conjunction with OpenLab (Improvision) and Vo locity (Improvision) imaging software. Using a 100X magnificati on, a Z-Stack with a 0.2 m step-width was acquired in red/green/blue channels. Z-Stacks were then cropped and calibrated in Openlab before the files were transferred to Volocity. In Volocity the files were employed for PhotonReassignment deconvolution. Splicing Reporters and Splicing Assays E-cadherin Exon4-intron-Exon5 segm ent was PCR amplified using Pfu polymerase (Stratagene) and genomic DNA as a template. Splicing reporter cons tructs were based on the pGL-3 basic luciferase reporter (Promega), carrying +53 E-cadherin basal promoter or SV40 promoter. The luciferase cDNA was exised and replaced with the cassette, which included inta ct exons 4 (144 bp) and 5 (156 bp) of E-cadherin gene linked by the entire intronic sequence (124 bp). E-cadherin[E-cadEx4-E x5] or SV40[E-cadEx4Ex5] splicing reporters contained E-cadhe rin or SV40 promoters driving E-cadherin exon4-intron-exon5 cassette which is followed by SV40 polyA signal sequence carried over with the pGL-3 backbone. For splicing assays HEK293 cells were co -transfected with 500 ng of splicing reporters and 1 and 2 g of Pnn expression vector or 1g of Pnn RNAi short hairpin vector (Joo et al., 2005; Ouyang and Sugr ue, 1996; Ouyang and Sugrue, 1992; Shi and


44 Sugrue, 2000; Shi et al., 2000). Total am ount of DNA was equalized using unrelated vector. 24 hrs post transfection total RNA was isolated using NucleoSpin kit (BD Biosciences), including 1hr DNAse treatment. To further eliminate DNA contamination isolated RNA was additionally treated w ith RQ1 DNAse (Promega) for 1hour followed by phenol-cholophorm extraction. Equal amou nt of RNA was subjected to reverse transcription using cMaster RT kit (Eppendorf) primed with SV40-specific primer. Resulting cDNA was 26 cycle-PCR amplifie d utilizing E-cadherin exon4 and exon5specific primers. The linear amplification ra nge of the PCR products was determined in pilot experiments. PCR products were re solved utilizing 5% TBE polyacrylamide Criteriongel system followed by staining for detection using SYBR Green dye (Molecular Probes). Gels were sc anned on Typhoon 8600 (Amersham Pharmacia Biotech) and quantitated utilizing ImageQua nt software (Molecular Dynamics). In addition, PCR products corresponding to spliced and unspliced E-cadherin message were excised from the gel and verified by sequencing. Results Pnn is Recruited to CtBP-associated Repressor Complexes In order to investigate the mechanisms behind Pnn-dependent modulation of CtBP-mediated silencing, we entertained tw o possible models, which may account for the Pnn-mediated de-repression of the E-cadherin promoter (Fig.9AC). Firstly, Pnn may sequester CtBP from CtBP -associated proteins, which target the E-cadherin promoter (Fig.9C). Alternativel y, Pnn may itself be br ought to the silencing complexes operating in the vicinity of the Ecadherin promoter via its interaction with CtBP, and attenuate CtBP-mediated repressi on (Fig.9B). To explore these possible


45 outcomes we focused on two transcriptional silencers ZEB and mSi n3A, which interact with CtBP and target E-cadherin promoter. Fig9. Pnn was recruited to the CtBP-associated repressor complexes. Pnn may sequester CtBP from the CtBP-associated represso r, alternatively P nn can be recruited to the repressor complex via CtBP (A-C ). HEK 293 cells were co-transfected with the vectors expressing ZEB1-myc (D) or mSin3A-myc (E) and CtBP1Flag in the absence or presence of hPnn-GFP. Immunoprecipitations were performed using Flag affinity agarose, followed by western blotting using anti-Pnn 143, anti-myc, and anti-Flag antibodies. Exogenous hPnnGFP is indicated by an arrowhead, and endogenous Pnn by an arrow. Increase in Pnn levels did not appreciably affect the amount of ZEB1 or mSin3A coprecipitated with CtBP, (D) and (E) respectively. Exogenous and endogenous Pnn could be detected in CtBP-Flag immunoprecipitates, indicating that Pnn may be present in ZEB1/CtBP1 and mSin3A/CtBP1 complexes. HEK 293 cells were co-transfected with vectors expressing ZEB1-myc (Fig.9D) or mSin3A-myc (Fig.9E) and CtBP1-Flag in the absence or presence of hPnn-GFP vector. Immunoprecipitati ons were performed using anti-Flag agarose. Immunoprecipitated material was subjected to western blotting using anti-Pnn 143, antimyc, and anti-Flag antibodies. hPnn-GFP e xpression did not result in changes of the amount of ZEB1 and mSin3A a ssociated with CtBP1 leading us to conclude that Pnn does not drastically affect the interaction of CtBP with ZEB or mSin3A. Interestingly, endogenous, as well as exogenous Pnn c ould be detected in CtBP-Flag


46 immunoprecipitates, suggesting that Pnn can be present in the CtBP-dependent silencing complexes (Fig.9D and 9E). Pnn Can Be Recruited to ZEB1/CtBP and mSin3A/CtBP Complexes via CtBP Next we determined whether binding to CtBP is essential for Pnn’s presence in the silencing complexes. Fig10. Pnn may be recruited to the ZEB1/Ct BP and mSin3A/CtBP complexes via CtBP. HEK293 cells were co-transfected with ZEB1-myc (A) or mSin3A-myc (B) in the presence of short hairpin-mediated RNAi vector for CtBP1 or GFP RNAi. Immunoprecipitations were performed using anti-myc agarose, followed by western blotting using anti-Pnn 143 and an ti-myc antibodies. Amount of Pnn in ZEB1 and mSin3A precipitates was diminished in the presence of CtBP RNAi, indicating that Pnn association with the ZE B1/CtBP and mSin3A/CtBP repressor complexes was CtBP-dependent. HEK293 cells were co-transfected wi th ZEB1-myc (Fig.10A) or mSin3A-myc (Fig.10B) in the presence of short hairpinmediated RNAi vector for CtBP1 or GFP RNAi as a control. Transient transfecti on of RNAi vector for CtBP1 resulted in approximately 60% reduction in CtBP1 protein levels (data not shown). Immunoprecipitations were performed using anti-myc agarose, followed by western blotting using anti-Pnn 143 and anti-myc an tibodies. Endogenous Pnn was detected in anti-ZEB1-myc and anti-mSin3A-myc precipitat es. However, amount of Pnn in the antiZEB-myc and anti-mSin3A-myc precipitates was diminished in the presence of CtBP RNAi, indicating that Pnn a ssociation with the ZEB1/CtB P and mSin3A/CtBP repressor complexes may be CtBP-dependent as depicted in the diagram (Fig. 9B). This suggests


47 that CtBP might play a role of a bridgi ng element capable of recruiting regulatory proteins to the silencing complexes. Pnn Is Present at the E-cadherin Promoter Because the expression of Pnn leads to the enhanced activity of E-cadherin promoter (Alpatov et al., 2004) we investigat ed whether Pnn can be detected at the promoter region and whether expression of Pnn affects local chromatin modification. Fig11. Pnn was present at the E-cadherin pr omoter, where it modulat ed transcriptional state of chromatin. (A) POZ-Pnn-FlagHA cells or POZ cells were subjected to ChIP utilizing anti-HA antibody. ChIPs were amplified using either Ecadherin or GAPDH promoter specific primers. Pnn was found to be associated with E-cadherin but not GAPDH promoter. (B) HeLa POZ-PnnFlag-HA cells or HeLa POZ cells were subjected to chromatin immunoprecipitation (ChIP) utilizing anti-acetylated histone 4 (AcH4), antidimethylated histone H3 at lysine9 (diMeH3K9), and anti-Polymerase II (N20). ChIPs were amplified using E-ca dherin or GAPDH promoterspecific primers. Cells expressing Pnn-Flag-HA exhibited enhanced H4 acetylation, diminished diMeH3K9, and increased pr esence of Pol II at the E-cadherin promoter. POZ-Pnn-Flag-HA or POZ cells were subjected to chromatin immunoprecipitation (ChIP) utilizing anti-HA antibody followed by PCR using Ecadherin promoter specific primers or GAPDH pr omoter primers as a control (Fig. 11A). Pnn was found to be associated with Ecadherin but not GAPDH promoter. Next, POZPnn-Flag-HA or POZ cells were subjected to ChIP analysis utili zing antibodies against acetylated histone 4 (AcH4) as a marker for transcriptionally active chromatin, anti


48 dimethylated histone H3 at lysine9 (diMeH3K9) as a marker for silenced chromatin, and anti-Polymerase II (N-20) (Fig. 11B). Cells expressing hPnn-Flag-HA exhibited enhanced H4 acetylation , diminished dimethylation at H3K9 and increased presence of Pol II at the E-cadherin promoter. These results suggest that presence of Pnn may affect local chromatin remodeling in the vicinity of the E-cadherin promoter, driving promoter to transcriptionally favorable state. Pnn Interacts with the Transcriptionally Competent RNA Polymerase II and Can Differentially Associate with CtBP a nd Pol II in a NADH-dependent Manner Observation that Pnn expression correlates with the increased presence of RNA polymerase II at the E-cadherin promoter pr ompted us to investigate the potential involvement of Pnn in basal transcriptiona l machinery. In order to determine if endogenous Pnn and Pol II can interact, we ut ilized HeLa nuclear extracts to perform immunoprecipitations with anti-Pol II (N20) antibody, followed by western blotting with anti-Pol II(N-20) and anti-Pnn 143 antibodies. Pnn was detected in Pol II precipitates indicating that these two factors associate in vivo (Fig. 12A). Next, we assessed whether Pnn can preferentially associate with a particular form of Pol II. Functionally une ngaged, preinitiation Pol II is hypophosphorylated at its Cterminal domain (CTD), initiation-specific Pol II is predominantly phosphorylated at Ser5 of CTD, and elongation-specific Pol II is predominantly phoshorylated at Ser2 of CTD (Cramer et al., 2001; Hirose and Manley, 2000; Maniatis and Reed, 2002; Proudfoot et al., 2002; Reed, 2003; Zorio and Be ntley, 2004). Nuclear extracts from the HeLa cells expressing hPnn-Flag-HA were s ubjected to immunoprecipitation using antiHA antibody, followed by western blotting using anti-HA antibody and antibodies directed against different forms of Pol II (Fig. 12B). Pnn immunoprecipitates contained


49 both, initiation and elongation-sp ecific Pol II but not the preinitiation form of Pol II suggesting that Pnn is predominantly associates with RNA polymerase II complex during the process of active transcription. Fig12. Pnn interacted with th e transcriptionally competent Pol II and can differentially associate with CtBP and Pol II in a NADH-dependent manner. (A) HeLa nuclear extracts were utilized for i mmunoprecipitation using anti-Pol II (N20) antibody, followed by western blotting using anti-Pol II(N-20) and antiPnn 143 antibodies. Endogenous Pnn and Po l II were found to be associated. (B) Nuclear extracts from the HeLa POZ-Pnn-Flag-HA cells were used for immunoprecipitation using anti-HA antibody, followed by western blotting for HA, preinitiation form of Pol II (CTD), initiation specific Pol II (Pol II PSer5), and elongation specific Pol II (Pol II P-Ser2). Pnn was found to be associated with transcriptionally compet ent forms of Pol II, but not with the preinitiation form of Pol II. (C) Bact erially expressed Pnn-GST, immobilized on glutathione beads was incubated w ith HeLa nuclear extracts in the presence or absence of 1mM NA DH. Western blotting for CtBP1, preinitiation Pol II (CTD), and initiat ion specific Pol II (Pol II P-Ser5) revealed that in the presence of NA DH Pnn pull-down material was enriched for CtBP1, however , Pol II P-Ser5 a mmount was diminished, suggesting the reorganization of Pnn/Pol II and Pnn/ CtBP complexes in a NADH-dependent manner. Variation of the levels of NAD(H) can aff ect the degree of Pnn/CtBP interaction (Alpatov et al., 2004) raising th e possibility that higher conc entration of NADH may also have an effect on Pnn/Pol II association. To test this bacteria lly expressed Pnn-GST


50 immobilized on glutathione beads was incuba ted with sHeLa nuclear extracts in the presence or absence of 1m M NADH (Fig. 12C). In the pr esence of NADH Pnn pull-down material was enriched for CtBP1, however, amount of initiation-specific Pol II was diminished. This suggests that NADH can ha ve an opposite effect on Pnn’s interaction with the transcriptionally competent form of Pol II and co-repressor CtBP. Distribution of Pnn Overlaps with the Transcriptionally Competent Forms of Pol II To explore a potential correlation between nuclear distribution of different forms of Pol II and Pnn, NIH3T3 cells stably e xpressing GFP-tagged Pnn were immunostained for GFP, initiation-specific, elongation-specific, and pre-initiation forms of Pol II (Fig. 13A). Pnn immunoreactivity partially overlapped with initiation-spec ific Pol II (upper panel), elongation-specific Pol II (middle panel) but less so with the preinitiation form of Pol II (lower panel) as indicated by the arro wheads. Next, we reasoned that if Pnn preferentially associates with the transcrip tionally active forms of Pol II, these factors may exhibit similar behavior in response to the transcriptional block. Therefore, we performed immunolabeling of actinomycin D-tr eated NIH3T3 cells stably expressing GFP-tagged Pnn for GFP and differe nt forms of Pol II (Fig. 13B). In support to Pnn/Pol II interaction data , initiation-specifi c (upper panel) and elongation-specific (middle panel) but not pr einitiation form of Pol II (lower panel) exhibited similar to Pnn redistribution dyna mics concentrating as enlarged speckles (arrowheads) in response to th e transcriptional inhibition. For detailed analysis of Pnn/Pol II co-localizati on in context of active transcription we utilized image deconvolu tion of NIH3T3 hPnnGFP cells pulse-labeled with BrU and immunostained with anti-GFP, anti-initiation-specifi c Pol II, and antielongation-specific Pol II (Fig. 13C). Activ e transcription often takes place at the


51 periphery of the nuclear speckles where co -transcriptional mRNA splicing is likely to occur (Bregman et al., 1995; Lamond and Spect or, 2003; Mintz et al., 1999; Mortillaro et al., 1996). Fig13. Pnn partially overlaped with the tran scriptionally competent Pol II. (A) NIH3T3 hPnnGFP cells were immunostained for G FP (green), Pol II P-Ser5, Pol II PSer2, and Pol II CTD (all red). Localizat ion of Pnn partially overlapped with Pol II P-Ser5 (upper panel), Pol II P-Se r2 (middle panel), but less so with CTD staining (lower pannel ) as indicated by the arrowh eads. (B) Same as (A) only after actinomycin D treatment. Po l II P-Ser5 and Pol II P-Ser2 but not transcriptionally inactive form of Po l II (CTD) exhibited similar to Pnn immunostaining pattern concentrating as enlarged speckle aggregates (arrowheads). (C) NIH3T3 hPnnGFP cells pulse-labeled with BrU and immunostained with anti-BrdU and anti -GFP antibodies (upper pannel), antiGFP and anti-Pol II P-Ser5 (middle pa nel), and anti-GFP and antiPol II PSer2 antibodies (lower panel). HPnnGFP is green, and BrU, Pol II-Ser5, Pol II P-Ser2 are red. BrU staini ng as well as staining for Pol II P-Ser5 and Pol II PSer2 appear to overlap with Pnn immunoreactivity at the Pnn “speckle” periphery (arrowheads). Bar 10 m.


52 We observed BrU staining as well as staining for initiation and elongationspecific Pol II to overlap with Pnn imm unoreactivity at the “speckle” periphery (arrowheads). These data indicates that P nn can be present in the zones of active transcription. Pnn Positively Affects E-cadherin mRNA Splicing Because Pnn is involved in pre-mRNA pro cessing (Li et al., 2003; Wang et al., 2002) we considered the possibility that Pnn can affect E-cadherin mRNA splicing efficiency. We created splicing reporter c onstructs which carried the intact E-cadherin exon4-intron-exon5 cassette followed by SV40 pol yA signal, which were driven by Ecadherin or SV-40 promoter as a control (Fig. 14). HEK293 cells were co-transfected with E-cadherin splicing reporter or SV-40 splicing reporter and increased amount of Pnn. Significan tly, increased amounts of Pnn resulted in the greater spliced/total mRNA ratio in context of E-cadherin (Fig. 14A left panel) but not SV40 promoter (Fig.14A right pa nel). We next, determined if reduction of Pnn levels would have an affect on th e E-cadherin mRNA splicing efficiency. Splicing assays were conducted in the pres ence or absence of P nn RNA interference (RNAi). Pnn RNAi resulted in reduction of spliced/total mRNA ratio in case of Ecadherin (Fig. 14B left panel) but not SV40 pr omoter (Fig. 14B right panel). These data suggest that Pnn is capable of modulati ng E-cadherin mRNA splicing efficiency in a promoter-specific manner. Because co-transcriptional splicing involves positioning mRNA processing factors along the transcribed ge ne (Bentley, 2005), we determined if Pnn is present within E-cadherin gene. POZ-Pnn-Flag-HA or PO Z cells were subjected to chromatin immunoprecipitation utilizing anti-HA antibody followed by PCR using E-cadherin


53 intragenic primers (Fig. 14D). Pnn was found to be associated with E-cadherin intragenic region, which supports its role in E-cadherin mRNA processing. Fig14. Pnn enhanced E-cadherin mRNA splicing efficiency. (A) HEK293 cells were cotransfected with E-cadherin[E-cadEx4Ex5] (left pannel) or SV40[E-cadEx4Ex5] (right pannel) splici ng reporters and increasing amount of Pnn. Spliced and unspliced RNAs were amplified by RT-PCR followed by quantitation of band intensity. Histograms on the right represent ratio spliced/total mRNA obtained from 3 independent assays. Bars represen t standard deviations. Increasing Pnn expression resulted in the increased spliced/total mRNA ratio in context of E-cadherin promoter but not SV40 promoter. (B) Transfection of Pnn RNAi vector resulted in the decreased spliced/total mRNA ratio in context of E-cadherin promoter but not SV40 promoter. (C) HeLa POZ-PnnFlag-HA cells or HeLa POZ cells were subjected to chromatin immunoprecipitation (ChIP) utilizi ng anti-HA antibody. ChIPs were amplified using E-cadherin intragenic primers. Pnn was found to be associated with E-cadherin intragenic region. Discussion The coupling mechanisms governing transc ription include syne rgistic action of regulatory complexes at the pr omoter regions as well as complexes involved in later stages of gene expression such as mRNA processing. In this work we explored the mechanis ms involved in Pnn-dependent modulation of CtBP-mediated silencing of the E-cadherin gene and connection between transcription


54 and splicing. We demonstrate that Pnn can be recruited to the CtBP-associated repressor complexes where it may relieve CtBP-mediated re pression. Furthermore, we showed that Pnn interacts with transcri ptionally competent RNA Poly merase II and enhances Ecadherin mRNA splicing efficiency. To investigate the possible effect of Pnn on transcriptional silencing mediated by CtBP, we focused on two transcriptiona l repressors ZEB1 and mSin3A, which are known to associate with CtBP, and ta rget E-cadherin promoter. Using coimmunoprecipitation assays we showed that Pnn does not sequester CtBP from CtBPassociated repressor proteins, but rather, gets recruited to the silencing complexes via CtBP. Whether the presence of Pnn triggers structural reorganization of the CtBPdependent complexes or so mehow modulates the function of CtBP needs further investigation. Our findings provide indir ect support for the contex t-dependent effect of regulatory complexes on a target gene. For example a protein complex can deliver silencing or activating effect on the gene promoter dependi ng on its factor composition. This assumption parallels a new exciting m odel describing a novel, cofactor-mediated specificity for histone demethylation exerted by the newly discovered factor LSD1(Shi et al., 2004). Intriguingly, it is po ssible that LSD1 may act as an activator or repressor depending on LSD1-associated protein comp lex composition (Wysocka et al., 2005). By using ChIP assays we demonstrated that Pnn is present at the E-cadherin promoter. The presence of Pnn correlated with increased histone H4 acetylation, decrease in histone H3K9 dimethylation, as well as the increa sed presence of RNA polymerase II, which corresponds to the transc riptionally active chromatin. This is


55 consistent with our data describing Pnn-de pendent enhancement of E-cadherin promoter activity (Alpatov et al., 2004). By utilizing co-immunoprecipitations along with immunofluorecense we demonstrated that Pnn associates with the transcriptionally competent forms of Pol II, such as initiation and elongatio n-specific Pol II but not with the preinitiation Pol II. The fact that Pnn associates with the transcriptionally competent forms of Pol II is in agreement with previously reported data on Pnn’s presence in Pol II-associated complexes involved in transcri ptional initiation a nd elongation, such as Mediator and CA150 (Sato et al., 2004; Smith et al., 2004). According to our 3D reconstruction studies Pnn overlaps with Pol II at the periphery of the Pnn speckles where transcription-coupled mRNA processing is thought to occur. Theref ore it is conceivable th at Pnn is recruited from the nuclear speckles to sites of tr anscription where Pnn engages in splicing activities. Intriguingly, our in vitro pull-down studies utilizing nucle ar extracts revealed that NADH levels could differentiall y modulate the degree of Pnn-de pendent interactions. In contrast to the Pnn/CtBP asso ciation, increased levels of NADH resulted in the decrease in Pnn/Pol II association. These findings imply the existence of a possible novel mechanism of NAD(H)-dependent reorganizatio n of transcriptional regulatory complexes in response to changes in the local metabolic state. Indeed, the role of NAD derivatives in transcriptional regulation has been recently described. The reduced forms of the redox co-factors, NAD(H) and NADP(H) enhan ce DNA binding of the Clock:BMAL1 and NPAS2:BMAL1 heterodimeric transcription factor s, whereas oxidized forms inhibit. In addition, histone H2B promoter activation by OCA-S coactivator complex has been


56 shown to be NAD(H) dependent (Rutter et al., 2001; Zheng et al., 2003). Therefore, it is tempting to speculate that nuclear NAD cofactors might dictate the degree of association between Pnn and its binding partners affec ting downstream transcriptional events. Finally, by utilizing a spli cing reporter construct driven by E-cadherin basal promoter containing E-cadherin exon4-intron-exon5 cassette, we showed that Pnn can modulate E-cadherin mRNA splicing efficien cy in a promoter-specific manner. Therefore, Pnn might affect E-cadherin gene transcription through Pnn-dependent associations with the regulatory complexes at the E-cadherin promot er as well as through its affect on splicing, thus li nking promoter related events and transcription-coupled mRNA processing. It is possible that depe nding on the cellular and promoter context, transcription-associated complexes recruit proteins capable of coordinating promoter specific as well as mRNA splicing events, en suring efficient produc tion of a processed gene product. Fig15. Diagram depicting possible mechanism of Pnn involvement in multiple steps of regulation of E-cadherin ge ne expression. (A) CtBP pa rticipates in repression of E-cadherin gene. (B) Pnn may be r ecruited to CtBP-associated repressor complexes via CtBP, where Pnn can attenuate CtBP-mediated repression. (C) Pnn may also regulate E-cadherin gene expression at the mRNA level via its association with Pol II, where Pnn mi ght couple transcription and splicing. Importantly NADH can modulate the degree of association of Pnn with CtBP and Pol II. In conclusion, we propose the concept of Pnn-dependent functional coupling of processes governing gene expression, wher e Pnn serves as an example of a unique protein factor, which contribu tes to promoter related co mplex interactions in a NADH


57 sensitive manner as well as impacts mRNA processing (Fig. 15). This idea provides an exiting opportunity for pharmacological targe ting of various cancers, whereby carefully designed targeting strategy directed towards a si ngle key regulatory molecule will have a cumulative and hopefully beneficial effect at multiple control points.


CHAPTER 4 DISCUSSION In our work we explored the underl ying mechanisms of Pnn-dependent modulation of the E-cadherin ge ne expression. Our findings s uggest that Pnn is capable of regulating E-cadherin gene expression via interaction with CtBP-dependent repressor complexes which target E-cadherin promoter as well as through its effect on E-cadherin mRNA splicing efficiency. These data highlights the role of Pnn as a multifunctional protein involved in multiple stages of ge ne expression, possibly coordinating promoter related events and mechan isms of mRNA processing. To exert the coordinated pattern of gene expres sion DNA-encoded genetic information has to be deciphered and organized in a readable manner, such that it can be transcribed and further translated into the final gene product. The driving force behind the transmission of the genetic information from DNA to protein consists of multiple protein factories, which heavily influence the unique proteomic composition from cell to an individual organism. It takes the coor dinated effort of exceedingly complicated machines to “unwrap” the tightly packaged chromatin at the gene promoter regions to allow the transcriptional machinery to be loaded on the promoter triggering transcriptional initi ation. Conversely, silencing comple xes function to drive chromatin to condensed conformation to prevent transcriptio nal initiation. During the later stages of gene expression transcription-coupled mRNA sp licing, export and translation take place, all subject to regulation by ex tensive protein networks. Th erefore, if one decides to categorize gene regulatory machinery, it w ould include activator complexes, which 58


59 activate gene expression and repressor complexes, which repress transcription, thus counterbalancing effect of activators. However, this clear cut picture dissipat es giving way to a emerging complexity when experimenters embark to establish the nature of driving forces behind a transcriptional regulatory pathway. One of the examples of the complexity of the machines involved in gene regulation is Ct BP-dependent transcriptional repression. Initially identified as a factor that attenuates the E1A-dependent oncogenic transformation, CtBP was later shown to act as a transcriptional corepessor in multiple developmental and oncogenic contexts. Functi oning mainly as a co-factor, it is recruited to its target promoters through various repr essor proteins where CtBP silences gene expression. CtBP-associated complex of pr oteins includes factors carrying enzymatic activities such as histone deace tylases (HDAC1 and HDAC2), histone methyltransfereases (G9a, and EuHMT), as well as recently discovered histone demethylase LSD1, all of them implicated in gene silencing due to their ability to covalently modify histone tails, thus creating compact or “silenced” chromatin conformation. Our findings that Pnn, a factor carrying no known connection to transcriptional silenci ng, is recruited to CtBP-associated complexes brings a possibility that in fact, protein complexe s such as those involved in repression are transient moieties that can accommodate factors with various functions including re pressors as well as activators. Therefore, depending on the factor co mposition “repressor” complex can also exert a positive effect on ge ne expression. Intriguingly, Pnn does not seem to be a component of the core CtBP complex, but rather , gets recruited to the complex via CtBP.


60 The presence of Pnn may somehow modulate the nature of CtBP-a ssociated functional assembly resulting in local chromatin re modeling and enhancement of E-cadherin promoter activity. A vivid example of this kind of contex t dependent gene regulation is LSD1, a monoamine oxidase found in the complex w ith CtBP and corepressor Co-REST. The repressive effect of this enzyme arises from its ability to specifically demethylate histone marks for the active chromatin such as me thylated H3K4 but not methylated H3K9, which is a mark for silenced chromatin. Surprisingly, LSD1 can also associate with androgen receptor and intriguingly act as a transcriptional coactiv ator to promote androgen receptor-dependent transcription. Therefore, the recent model proposes that LSD1 represses transcripti on of genes silenced by Co -REST by demethylating active histone marks mono-and di-methyl H3K4. Ho wever, in complex with the androgen receptor LSD1 activates gene transcription by demethylating the repressive histone marks mono-and di-methyl H3K9. Thus, depending on the LSD1-dependent complex composition LSD1 can be either coactivator or corepressor (Metzger et al., 2005; Shi et al., 2004; Wysocka et al., 2005). This kind of variation in the control of gene expression might reflect context dependent nature of the transcriptional comple xes in general. For example, depending on the local metabolic status the mechanisms for complex rearrangements are initiated, resulting in the factor reshuffling. Ou r observations that NADH can differentially regulate Pnn-dependent interactions with Pol II and CtBP might, indeed. reflect the existence of such metabolism-sensitive tran scriptional reorganizat ions. So, why having transcription factors interacting in various comb inations is advantageous to a cell? It is


61 possible that evolutionary e xpansion of diversity is the matter of accumulation of novel functional networks and not the number of ge nes, so that increased complexity of regulatory elements and protein interaction in terfaces allowed more flexible control of gene expression utilizing limited number of effector molecules. In this way with a minimal number of protein components you can create the func tional diversity of complexes involved in transcri ptional control. In addition, “complex reorganization” approach in gene control mi ght provide quick and efficien t change in gene expression profile in response to intracellular signal. It is much more effective to modulate gene expression by merely reorganizing the constitu ents of transcriptional complexes already present at the promoter or pr omoter vicinity. This approach is not entirely novel, as nuclear hormone receptor (NR)-mediated tran scriptional machinery utilizes similar principle through ligand-dependent reorganization of the transc riptional complexes present on the NR-response elements of the gene promoters. Interestingly, complex rearrangements may depend on the presence of molecules such as CtBP, which play a role of bridging elements capable of recruiting an array of factors involved in transcript ional control, whether represso rs or activators, depending, for example, on the levels of nuclear NAD(H) . On the other hand, the nature of CtBPdependent recruitment of Pnn might refl ect the state of Pnn posttranslational modifications such as phoshporylation or hydr oxylation. Postransla tional modifications of CtBP such as phosphorylati on and ubiquitination may also have a role in th e ability of CtBP to serve as a bridging factor. It appear s that synergy of coex isting gene regulatory networks reflects the hierarchy of recruitm ent of various regulatory complexes in a coordinated manner to gene promoters, de pending on the physiologi cal context of the


62 cell. In general, the existe nce of the transcriptional modul es capable of carrying out the transcriptional program would allow tight and efficient cont rol of gene expression at multiple control points utilizing multifunctional key regulatory molecules, such as CtBP. For example, by modulating CtBP posttranscri ptional modifications it would be possible to trigger changes in its bindi ng preferences for a subset of the transcriptional factors at the given promoter, resulting in change of gene expression. Th erefore, the recruitment of Pnn to CtBP-associated complexes via CtBP may represent an example of a common pathway for a number of transcri ptional factors invol ved in regulation of gene expression. As previously stated, one mechanistic e xplanation for the presence of Pnn at the E-cadherin promoter is through the CtBP-depe ndent interactions, where Pnn might affect the function of the CtBP-associated protein assemblies. However, Pnn has also been shown to be part of the Mediator complex, involved in modulation of the transcriptional initiation (Sato et al., 2004). Furthermor e, Pnn is found in complex with CA150, a multifunctional protein linking transcription a nd mRNA processing (Smith et al., 2004). In addition, we demonstrated that Pnn inter acts with the transcri ptional competent RNA polymerase II. Therefore, the presence of P nn at the promoter may be modulated through different pathways, including RNA polymerase II-dependent associations. This argument adds additional complex ity to the Pnn-dependent regulatory networks which may orchestrate gene expressi on. Does the presence of Pnn in CtBPdependent complex influence its function as a basal transcription-related constituent? Or are these events not functionally linked? The interdependence of these interactions is unclear; however, we have demonstrated that the levels of NAD(H) may have an effect on the Pnn’s presence in Pol II and CtBP-associated complexes. It is possible that


63 depending on the levels of NA D(H) Pnn might preferentially associate with Pol II or CtBP, influencing transcription through diffe rent unrelated mechanisms. On the other hand Pnn might represent a link between the repressor complexes and Pol II machinery, thus invoking an intriguing possibility that the role of transcrip tional repressors can extend to the level of attenua tion of RNA polymerase II associ ated complexes. Because higher NADH levels enhance Pnn/CtBP intera ction, mechanism of the CtBP-mediated repression might include sequesterization of RNA polymerase II-associated factors resulting in a compro mised transcription. How can corepressor CtBP attenuate transcriptional events via its interaction with Pnn? Our data indicate that Pnn can positively influence E-cadherin mRNA splicing efficiency, presumably through Pol II depende nt interactions. In this manner Pnn is capable of modulating CtBP-mediated repres sion and at the same time affecting transcription-linked mRNA processing. However , it is tempting to speculate that, under certain circumstances, CtBP-dependent sequest erization of Pnn might in turn inflict the uncoupling of transcription and splicing events due to the loss of a coupling factor. This would result in inefficient coordination between transcription and mRNA processing, potentially leading to the accumulation of unprocessed gene product. In reality, the assumption that known transcriptional repressors might have an effect on the splicing decisions is not entirely far fetched. The example would be PRP4 kinase, which is found in complex with the nuclear receptor-associ ated corepressor NcoR and splicing-related factors (Dellaire et al., 2002). Although the functional consequences of this organization need to be determined, PRP4K complex is a clear representation of the increasingly evident link between the promoter associ ated regulatory complexes and mRNA


64 processing machinery. This argument may provide a novel and unexpected connection between the function of CtBP in gene regulation and mRNA processing, where Pnn might represent a functiona l switch that mediates these two processes. The realization that transcription factors which regulate gene promoters, in fact, might also influence mRNA splicing is an im portant consideration which needs to be addressed experimentally. Assays such as lu ciferase reporter analys is or real time PCR are all based on the amount of the final mRNA product and do not take into consideration the possibility that a factor of intere st might actually f unction during the mRNA processing stage in addition to it s role at the gene promoter. Therefore, determination of mRNA levels as computational means to monitor gene expression might not be sufficiently comprehensive to functionally cate gorize a given transcriptional regulator. By designing E-cadherin splicing reporter dr iven by endogenous E-cadherin promoter we created a useful tool for sc reening known transcriptional regulators which target Ecadherin promoter in order to determine if th ey have an additional effect on E-cadherin mRNA splicing. In summary, in our experiments we addressed a question of multifunctionality of transcriptional regulators. Pnn may represent an example of a factor capable of affecting CtBP-mediated functional interactions at the E-cadherin promoter as well as affecting transcription through the enha ncement of the transcription-coupled mRNA processing. These findings imply the existence of the machineries capable of controlling gene expression through multiple mechanisms, including promoter related events as well as splicing processes. By having a limited number of key multifunctional regulators affecting multiple stages of gene expression a cell is able to achieve necessary


65 coordination of numerous functional networks in the process of creating a functional gene product. In addition the quality control of gl obal gene expression can be easily expedited through surveying a limited number of elemen ts located at the control junctions of transcriptional communication infrastructure.


LIST OF REFERENCES Alpatov, R., Munguba, G. C., Caton, P., Joo, J. H., Shi, Y., Hunt, M. E., and Sugrue, S. P. (2004). Nuclear speckle-associated protein Pnn/DRS binds to the transcriptional corepressor CtBP and relieves CtBP-mediate d repression of the E-cadherin gene. Mol. Cell. Biol. 24 , 10223-35. Auboeuf, D., Dowhan, D. H., Dutertre, M., Ma rtin, N., Berget, S. M., and O'Malley, B. W. (2005). A subset of nuclear receptor co regulators act as coupling proteins during synthesis and maturation of RNA transcripts. Mol. Cell. Biol. 25, 5307-16. Auboeuf, D., Dowhan, D. H., Kang, Y. K., Lark in, K., Lee, J. W., Berget, S. M., and O'Malley, B. W. (2004). Differential recruitm ent of nuclear receptor coactivators may determine alternative RNA splice site choice in target genes. Proc. Natl. Acad. Sci. U S A 101, 2270-4. Auboeuf, D., Dowhan, D. H., Li, X., Larkin, K., Ko, L., Berget, S. M., and O'Malley, B. W. (2004). CoAA, a nuclear receptor coactivator protein at the interfac e of transcriptional coactivation and RNA splicing. Mol.Cell. Biol. 24, 442-53. Auboeuf, D., Honig, A., Berget, S. M., and O'Malley, B. W. (2002). Coordinate regulation of transcription and splicing by steroid receptor coregulators. Science 298, 416-9. Balasubramanian, P., Zhao, L. J., and Chinnadurai, G. (2003). Nicotinamide adenine dinucleotide stimulates oligomerization, inter action with adenovirus E1A and an intrinsic dehydrogenase activity of CtBP. FEBS Lett. 537, 157-60. Barnes, C. J., Vadlamudi, R. K., Mishra, S. K., Jacobson, R. H., Li, F., and Kumar, R. (2003). Functional inactivation of a transcriptio nal corepressor by a signaling kinase. Nat. Struct. Biol. 10, 622-8. Bellavite, P., Bazzoni, F., Cassa tella, M. A., Hunter, K. J., and Bannister, J. V. (1990). Isolation and characterization of a cDNA clone for a novel se rine-rich neutrophil protein. Biochem. Biophys. Res. Commun. 170, 915-22. Bentley, D. L. (2005). Rules of engagement: co-transcriptional re cruitment of pre-mRNA processing factors. Cu rr. Opin. Cell Biol. 17, 251-6. Bomsztyk, K., Denisenko, O., and Ostrowski, J. (2004). hnRNP K: one protein multiple processes. Bioessays 26, 629-38. 66


67 Boyd, J. M., Subramanian, T., Schaeper, U., La Regina, M., Bayley, S., and Chinnadurai, G. (1993). A region in the C-terminus of adenovirus 2/5 E1a protein is required for association with a cellular phosphoprotein and important for the negative modulation of T24-ras mediated transformation, tumo rigenesis and metastasis. Embo J. 12, 469-78. Brandner, J. M., Reidenbach, S., and Fra nke, W. W. (1997). Evidence that "pinin", reportedly a differentiation-specific desmosomal protein, is actually a widespread nuclear protein. Differentiation 62 , 119-27. Bregman, D. B., Du, L., van der Zee, S ., and Warren, S. L. (1995). Transcriptiondependent redistribution of the large subunit of RNA polymerase II to discrete nuclear domains. J. Cell. Biol. 129 , 287-98. Bres, V., Gomes, N., Pickle, L., and Jones, K. A. (2005). A human splicing factor, SKIP, associates with P-TEFb and enhances transc ription elongation by HI V-1 Tat. Genes Dev. 19, 1211-26. Caceres, J. F., Misteli, T., Screaton, G. R., Spector, D. L., and Krainer, A. R. (1997). Role of the modular domains of SR proteins in subnuclear localization a nd alternative splicing specificity. J. Cell Biol. 138 , 225-38. Cao, R., Wang, L., Wang, H., Xia, L., Erdjumen t-Bromage, H., Tempst, P., Jones, R. S., and Zhang, Y. (2002). Role of histone H3 lysine 27 methylation in Polycomb-group silencing. Science 298, 1039-43. Chinnadurai, G. (2002). CtBP, an unconven tional transcriptiona l corepressor in development and oncogenesis. Mol. Cell 9, 213-24. Comijn, J., Berx, G., Vermassen, P., Verschueren, K., van Grunsven, L., Bruyneel, E., Mareel, M., Huylebroeck, D., and van R oy, F. (2001). The two-handed E box binding zinc finger protein SIP1 downregulates E-cadherin and induces invasion. Mol. Cell 7, 1267-78. Cramer, P., Caceres, J. F., Cazalla, D., Kade ner, S., Muro, A. F., Baralle, F. E., and Kornblihtt, A. R. (1999). Coupling of transcri ption with alternativ e splicing: RNA pol II promoters modulate SF2/ASF and 9G8 effects on an exonic splicing enhancer. Mol. Cell 4, 251-8. Cramer, P., Pesce, C. G., Baralle, F. E ., and Kornblihtt, A. R. (1997). Functional association between promoter structure and tr anscript alternative splicing. Proc. Natl. Acad. Sci. U S A 94, 11456-60. Cramer, P., Srebrow, A., Kadener, S., Werbaj h, S., de la Mata, M., Melen, G., Nogues, G., and Kornblihtt, A. R. (2001). Coordina tion between transcription and pre-mRNA processing. FEBS Lett. 498, 179-82.


68 Dellaire, G., Makarov, E. M., Cowger, J. J., Longman, D., Sutherland, H. G., Luhrmann, R., Torchia, J., and Bickmore, W. A. (2002). Mammalian PRP4 kinase copurifies and interacts with components of both the U5 snRNP and the N-CoR deacetylase complexes. Mol. Cell. Biol. 22, 5141-56. Dowhan, D. H., Hong, E. P., Auboeuf, D., Denni s, A. P., Wilson, M. M., Berget, S. M., and O'Malley, B. W. (2005). Steroid hormone receptor coactivation and alternative RNA splicing by U2AF65-related proteins CA PERalpha and CAPERbeta. Mol. Cell 17, 42939. Fjeld, C. C., Birdsong, W. T., and Goodman, R. H. (2003). Differential binding of NAD+ and NADH allows the transcrip tional corepressor carboxyl-term inal binding protein to serve as a metabolic sensor. Proc. Natl. Acad. Sci. U S A 100 , 9202-7. Frisch, S. M. (1994). E1a induces the expression of epithelia l characteristics. J. Cell. Biol. 127, 1085-96. Giroldi, L. A., Bringuier, P. P., de We ijert, M., Jansen, C., van Bokhoven, A., and Schalken, J. A. (1997). Role of E boxes in the repression of E-cadherin expression. Biochem. Biophys. Res. Commun. 241, 453-8. Goldstrohm, A. C., Albrecht, T. R., Sune, C ., Bedford, M. T., and Garcia-Blanco, M. A. (2001). The transcription elonga tion factor CA150 interacts w ith RNA polymerase II and the pre-mRNA splicing factor SF1. Mol. Cell. Biol. 21, 7617-28. Goldstrohm, A. C., Greenleaf, A. L., and Garcia-Blanco, M. A. (2001). Cotranscriptional splicing of pre-messenger R NAs: considerations for the mechanism of alternative splicing. Gene 277, 31-47. Graff, J. R., Gabrielson, E., Fujii, H., Ba ylin, S. B., and Herman, J. G. (2000). Methylation patterns of the E-cadherin 5' CpG island are unstable and reflect the dynamic, heterogeneous loss of E-cadherin ex pression during metast atic progression. J. Biol. Chem. 275, 2727-32. Grimes, J. A., Nielsen, S. J., Battaglioli, E., Mi ska, E. A., Speh, J. C., Berry, D. L., Atouf, F., Holdener, B. C., Mandel, G., and Kouzarid es, T. (2000). The co-repressor mSin3A is a functional component of the REST-CoR EST repressor complex. J. Biol. Chem. 275, 9461-7. Grooteclaes, M., Deveraux, Q., Hildebrand, J., Zhang, Q., Goodman, R. H., and Frisch, S. M. (2003). C-terminal-binding protein core presses epithelial a nd proapoptotic gene expression programs. Proc. Natl. Acad. Sci. U S A 100, 4568-73. Grooteclaes, M. L., and Frisch, S. M. (2000). Evidence for a function of CtBP in epithelial gene regulation and anoikis. Oncogene 19, 3823-8.


69 Hateboer, G., Gennissen, A., Ramos, Y. F., Kerkhoven, R. M., Sonntag-Buck, V., Stunnenberg, H. G., and Bernards, R. (1995) . BS69, a novel adenovirus E1A-associated protein that inhibits E1A transactivation. Embo J. 14, 3159-69. Hennig, G., Behrens, J., Truss, M., Frisch, S., Reichmann, E., and Birchmeier, W. (1995). Progression of carcinoma cells is associated with alterations in chromatin structure and factor binding at the E-cadheri n promoter in vivo. Oncogene 11, 475-84. Hennig, G., Lowrick, O., Birchmeier, W., and Be hrens, J. (1996). Mechanisms identified in the transcriptional control of epit helial gene expression. J. Biol. Chem. 271, 595-602. Hildebrand, J. D., and Soriano, P. (2002). Ov erlapping and unique roles for C-terminal binding protein 1 (CtBP1) and CtBP2 during mouse development. Mol. Cell. Biol. 22 , 5296-307. Hirose, Y., and Manley, J. L. (2000). RNA pol ymerase II and the inte gration of nuclear events. Genes Dev. 14, 1415-29. Hosono, S., Gross, I., English, M. A., Hajra, K. M., Fearon, E. R., and Licht, J. D. (2000). E-cadherin is a WT1 target gene. J. Biol. Chem. 275, 10943-53. Iwasaki, T., Chin, W. W., and Ko, L. (2001) . Identification and ch aracterization of RRMcontaining coactivator activat or (CoAA) as TRBP-interac ting protein, and its splice variant as a coactivator modul ator (CoAM). J. Biol. Chem. 276, 33375-83. Izutsu, K., Kurokawa, M., Imai, Y., Maki, K., Mitani, K., and Hirai, H. (2001). The corepressor CtBP interacts with Evi-1 to repress transforming gr owth factor beta signaling. Blood 97, 2815-22. Joo, J. H., Alpatov, R., Munguba, G. C., Jackson, M. R., Hunt, M. E., and Sugrue, S. P. (2005). Reduction of Pnn by RNAi induces loss of cell-cell adhe sion between human corneal epithelial cells. Mol. Vis. 11 , 133-42. Jung, S. Y., Malovannaya, A., Wei, J., O'Ma lley B, W., and Qin, J. (2005). Proteomic Analysis of Steady-State Nuclear Hormone Receptor Coactivator Complexes. Mol. Endocrinol. 28, 28. Kagey, M. H., Melhuish, T. A., and Wotton, D. (2003). The polycom b protein Pc2 is a SUMO E3. Cell 113, 127-37. Kelley, D. E., Stokes, D. G., and Perry, R. P. (1999). CHD1 interacts with SSRP1 and depends on both its chromodomain and its ATPase/helicase-like domain for proper association with chromatin. Chromosoma 108, 10-25.


70 Kim, J. H., Cho, E. J., Kim, S. T., and Youn, H. D. (2005). CtBP represses p300mediated transcriptional activ ation by direct association with its bromodomain. Nat. Struct. Mol. Biol. 12, 423-8.. Koipally, J., and Georgopoulos, K. (2000). Ikar os interactions with CtBP reveal a repression mechanism that is independent of histone deacetylase activity. J. Biol. Chem. 275, 19594-602. Kumar, V., Carlson, J. E., Ohgi, K. A., Edwards, T. A., Rose, D. W., Escalante, C. R., Rosenfeld, M. G., and Aggarwal, A. K. (2002) . Transcription corepressor CtBP is an NAD(+)-regulated dehydrogenase. Mol. Cell 10 , 857-69. Ladendorff, N. E., Wu, S., and Lipsick, J. S. (2001). BS69, an adenovirus E1A-associated protein, inhibits the transcripti onal activity of c-Myb. Oncogene 20, 125-32. Lamond, A. I., and Spector, D. L. (2003). Nuclear speckles: a model for nuclear organelles. Nat. Rev. Mol. Cell. Biol. 4, 605-12. Li, C., Lin, R. I., Lai, M. C., Ouyang, P ., and Tarn, W. Y. (2 003). Nuclear Pnn/DRS protein binds to spliced mRNPs and partic ipates in mRNA processing and export via interaction with RNPS1. Mol. Cell. Biol. 23, 7363-76. Lin, X., Sun, B., Liang, M., Liang, Y. Y., Ga st, A., Hildebrand, J., Brunicardi, F. C., Melchior, F., and Feng, X. H. (2003). Oppos ed regulation of corepressor CtBP by SUMOylation and PDZ binding. Mol. Cell 11, 1389-96. Longman, D., Johnstone, I. L., and Caceres, J. F. (2000). Functional characterization of SR and SR-related genes in Caenorhabditis elegans. Embo J. 19, 1625-37. MacDonald, P. N., Dowd, D. R., Zhang, C., and Gu, C. (2004). Emerging insights into the coactivator role of NCoA62/SKIP in Vita min D-mediated transc ription. J. Steroid.. Biochem. Mol. Biol. 89-90, 179-86. Maniatis, T., and Reed, R. (2002). An ex tensive network of coupling among gene expression machines. Nature 416, 499-506. Marmorstein, R. (2002). Dehydrogenases, NAD, a nd transcription-what's the connection? Structure (Camb.) 10, 1465-6. Martinez, E., Palhan, V. B., Tjernberg, A., Lymar, E. S., Gamper, A. M., Kundu, T. K., Chait, B. T., and Roeder, R. G. (2001). Human STAGA complex is a chromatinacetylating transcription coactivator that interacts with pre-mRNA splicing and DNA damage-binding factors in vivo. Mol. Cell. Biol. 21, 6782-95. Masselink, H., and Bernards, R. (2000). The adenovirus E1A binding protein BS69 is a corepressor of transcription through recruitment of N-CoR. Oncogene 19, 1538-46.


71 Melhuish, T. A., and Wotton, D. (2000). The interaction of the carboxyl terminus-binding protein with the Smad corepressor TGIF is disrupted by a holoprosencephaly mutation in TGIF. J. Biol. Chem. 275 , 39762-6. Meloni, A. R., Smith, E. J., and Nevins , J. R. (1999). A mechanism for Rb/p130mediated transcription repre ssion involving recruitment of the CtBP corepressor. Proc. Natl. Acad. Sci. U S A 96 , 9574-9. Metzger, E., Wissmann, M., Yin, N., Muller, J. M., Schneider, R., Peters, A. H., Gunther, T., Buettner, R., and Schule, R. (2005). LSD1 demethylates repressive histone marks to promote androgen-receptor-dependent transcription. Nature 437 , 436-9. Mintz, P. J., Patterson, S. D., Neuwald, A. F ., Spahr, C. S., and Spector, D. L. (1999). Purification and biochemical ch aracterization of interchromatin granule clusters. Embo J. 18, 4308-20. Mintz, P. J., and Spector, D. L. (2000). Co mpartmentalization of RNA processing factors within nuclear speckle s. J. Struct. Biol. 129, 241-51. Misteli, T., Caceres, J. F., and Spector, D. L. (1997). The dynamics of a pre-mRNA splicing factor in living cells. Nature 387, 523-7. Mortillaro, M. J., Blencowe, B. J., Wei, X ., Nakayasu, H., Du, L., Warren, S. L., Sharp, P. A., and Berezney, R. (1996). A hyperphosphor ylated form of the large subunit of RNA polymerase II is associated with splicing co mplexes and the nuclear matrix. Proc. Natl. Acad. Sci. U S A 93, 8253-7. Nagai, K., Yamaguchi, T., Takami, T., Kawa sumi, A., Aizawa, M., Masuda, N., Shimizu, M., Tominaga, S., Ito, T., Tsukamoto, T., and Osumi, T. (2004). SKIP modifies gene expression by affecting both transcription and splicing. Biochem. Biophys. Res. Commun. 316, 512-7. Nardini, M., Spano, S., Cericola, C., Pesce, A., Massaro, A., Millo, E., Luini, A., Corda, D., and Bolognesi, M. (2003). CtBP/BARS: a dual-function protein involved in transcription co-repression and Golgi membrane fission. Embo J. 22, 3122-30. Nitta, E., Izutsu, K., Yamaguchi, Y., Imai, Y ., Ogawa, S., Chiba, S., Kurokawa, M., and Hirai, H. (2005). Oligomeriz ation of Evi-1 regulated by th e PR domain contributes to recruitment of corepressor CtBP. Oncogene 16, 16. Ogawa, H., Ishiguro, K., Gaubatz, S., Livings ton, D. M., and Nakatani, Y. (2002). A complex with chromatin modifiers that occupies E2Fand Mycresponsive genes in G0 cells. Science 296, 1132-6.


72 Ouyang, P. (1999). Antibodies differentiate desmosome-form and nucleus-form pinin: evidence that pinin is a moonlighting protei n with dual location at the desmosome and within the nucleus. Biochem. Biophys. Res. Commun. 263, 192-200. Ouyang, P., and Sugrue, S. P. (1996). Ch aracterization of pinin, a novel protein associated with the desmosome-intermed iate filament complex. J. Cell. Biol. 135, 102742. Ouyang, P., and Sugrue, S. P. (1992). Identifica tion of an epithelial pr otein related to the desmosome and intermediate filament network. J. Cell. Biol. 118, 1477-88. Pagani, F., Stuani, C., Zuccato, E., Kornblihtt, A. R., and Baralle, F. E. (2003). Promoter architecture modulates CFTR exon 9 skipping. J. Biol. Chem. 278, 1511-7. Peinado, H., Ballestar, E., Esteller, M., a nd Cano, A. (2004). Snail mediates E-cadherin repression by the recruitment of the Si n3A/histone deacetylase 1 (HDAC1)/HDAC2 complex. Mol. Cell. Biol. 24, 306-19. Phippen, T. M., Sweigart, A. L., Moniwa, M., Krumm, A., Davie, J. R., and Parkhurst, S. M. (2000). Drosophila C-terminal binding pr otein functions as a context-dependent transcriptional co-factor and interferes with both mad and groucho transcriptional repression. J. Biol. Chem. 275, 37628-37. Postigo, A. A., and Dean, D. C. (1999). ZEB represses transcripti on through interaction with the corepressor CtBP. Pr oc. Natl. Acad. Sci. U S A 96, 6683-8. Postigo, A. A., and Dean, D. C. (2000). Diffe rential expression and function of members of the zfh-1 family of zinc finger/homeodoma in repressors. Proc. Natl. Acad. Sci. U S A 97, 6391-6. Proudfoot, N. J., Furger, A., and Dye, M. J. (2002). Integrating mRNA processing with transcription. Cell 108, 501-12. Reed, R. (2003). Coupling transcription, sp licing and mRNA export. Curr. Opin. Cell Biol. 15, 326-31. Rutter, J., Reick, M., Wu, L. C., and McKni ght, S. L. (2001). Regul ation of clock and NPAS2 DNA binding by the redox stat e of NAD cofactors. Science 293 , 510-4. Sacco-Bubulya, P., and Spector, D. L. (2002). Disassembly of interchromatin granule clusters alters the coordination of transcri ption and pre-mRNA splicing. J. Cell. Biol. 156, 425-36. Sakashita, E., Tatsumi, S., Werner, D., Endo, H., and Mayeda, A. (2004). Human RNPS1 and its associated factors: a versatile alternative pre-mR NA splicing regulator in vivo. Mol. Cell. Biol. 24, 1174-87.


73 Sato, S., Tomomori-Sato, C., Parmely, T. J., Florens, L., Zybailov, B., Swanson, S. K., Banks, C. A., Jin, J., Cai, Y., Washburn, M. P., Conaway, J. W., and Conaway, R. C. (2004). A set of consensus mammalian mediator subunits identified by multidimensional protein identification technology. Mol. Cell 14, 685-91. Schaeper, U., Boyd, J. M., Verma, S., Uhlm ann, E., Subramanian, T., and Chinnadurai, G. (1995). Molecular cloning and characteri zation of a cellular phosphoprotein that interacts with a conserved C-terminal doma in of adenovirus E1A involved in negative modulation of oncogenic transformati on. Proc. Natl. Acad. Sci. U S A 92, 10467-71. Schaeper, U., Subramanian, T., Lim, L., Boyd, J. M., and Chinnadurai, G. (1998). Interaction between a cellular protein that binds to the C-terminal region of adenovirus E1A (CtBP) and a novel cellular protein is disrupted by E1A through a conserved PLDLS motif. J. Biol. Chem. 273 , 8549-52. Sewalt, R. G., Gunster, M. J., van der Vla g, J., Satijn, D. P., and Otte, A. P. (1999). CTerminal binding protein is a transcriptional re pressor that interacts with a specific class of vertebrate Polycomb proteins. Mol. Cell. Biol. 19, 777-87. Shi, J., and Sugrue, S. P. (2000). Dissecti on of protein linkage between keratins and pinin, a protein with dual location at desmos ome-intermediate filament complex and in the nucleus. J. Biol. Chem. 275, 14910-5. Shi, S. (2001). Change in gene expression subsequent to induc tion of Pnn/DRS/memA: increase in p21. Oncogene. Shi, Y., Lan, F., Matson, C., Mulligan, P., Whetstine, J. R., Cole, P. A., and Casero, R. A. (2004). Histone demethylation mediated by the nuclear amine oxidase homolog LSD1. Cell 119, 941-53. Shi, Y., Ouyang, P., and Sugrue, S. P. ( 2000). Characterization of the gene encoding pinin/DRS/memA and evidence for its poten tial tumor suppressor function. Oncogene 19, 289-97. Shi, Y., Sawada, J., Sui, G., Affar el, B., Wh etstine, J. R., Lan, F., Ogawa, H., Luke, M. P., and Nakatani, Y. (2003). Coordinated hi stone modifications mediated by a CtBP corepressor complex. Nature 422, 735-8. Shi, Y., Simmons, M. N., Seki, T., Oh, S. P., and Sugrue, S. P. (2001). Change in gene expression subsequent to i nduction of Pnn/DRS/memA: increase in p21(cip1/waf1). Oncogene 20, 4007-18. Shi, Y., Tabesh, M., and Sugrue, S. P. (2000) . Role of cell adhesion-associated protein, pinin (DRS/memA), in corneal epithelial mi gration. Invest. Ophthalmol. Vis. Sci. 41 , 1337-45.


74 Shin, C., and Manley, J. L. (2002). The SR protein SRp38 represses splicing in M phase cells. Cell 111 , 407-17. Shnyreva, M., Schullery, D. S., Suzuki, H., Higaki, Y., and Bomsztyk, K. (2000). Interaction of two multifuncti onal proteins. Heterogeneous nuclear ribonucleoprotein K and Y-box-binding protein. J. Biol. Chem. 275, 15498-503. Simic, R., Lindstrom, D. L., Tran, H. G., Roinick, K. L., Costa, P. J., Johnson, A. D., Hartzog, G. A., and Arndt, K. M. (2003). Ch romatin remodeling protein Chd1 interacts with transcription elongation factors and localizes to transcribed genes. Embo J. 22, 1846-56. Skalweit, A., Doller, A., Huth, A., Kahne, T ., Persson, P. B., and Thiele, B. J. (2003). Posttranscriptional control of renin synthesis: identification of proteins interacting with renin mRNA 3'-untranslated region. Circ. Res. 92 , 419-27. Smith, M. J., Kulkarni, S., and Pawson, T. (2004). FF domains of CA150 bind transcription and splicing factors through mu ltiple weak interacti ons. Mol. Cell. Biol. 24 , 9274-85. Spector, D. L. (2001). Nuclear domains. J. Cell Sci. 114, 2891-3. Srinivasan, L., and Atchison, M. L. (2004). YY1 DNA binding and PcG recruitment requires CtBP. Genes Dev. 18, 2596-601. Sui, G., Soohoo, C., Affar el, B., Gay, F., Sh i, Y., and Forrester, W. C. (2002). A DNA vector-based RNAi technology to suppress ge ne expression in mammalian cells. Proc. Natl. Acad. Sci. U S A 99 , 5515-20. Sundqvist, A., Bajak, E., Kurup, S. D., So llerbrant, K., and Svensson, C. (2001). Functional knockout of the corepressor Ct BP by the second exon of adenovirus E1a relieves repression of tr anscription. Exp. Cell Res. 268, 284-93. Sutrias-Grau, M., and Arnosti, D. N. (2004). CtBP contributes quantitatively to Knirps repression activity in an NAD binding-de pendent manner. Mol. Cell. Biol. 24, 5953-66. Tai, H. H., Geisterfer, M., Bell, J. C., M oniwa, M., Davie, J. R., Boucher, L., and McBurney, M. W. (2003). CHD1 associates wi th NCoR and histone deacetylase as well as with RNA splicing proteins . Biochem. Biophys. Res. Commun. 308, 170-6. Tasic, B., Nabholz, C. E., Baldwin, K. K., Kim, Y., Rueckert, E. H., Ribich, S. A., Cramer, P., Wu, Q., Axel, R., and Maniatis, T. (2002). Promoter choice determines splice site selection in protocadherin alpha and gamma pre-mRNA splicing. Mol. Cell 10, 2133.


75 Thio, S. S., Bonventre, J. V., and Hsu, S. I. (2004). The CtBP2 co-repressor is regulated by NADH-dependent dimerization and possesses a novel N-terminal repression domain. Nucleic Acids Res. 32, 1836-47. Wang, D. S., Miller, R., Shaw, R., and Shaw, G. (1996). The pleckstrin homology domain of human beta I sigma II spectrin is targeted to the plasma membrane in vivo. Biochem. Biophys. Res. Commun. 225, 420-6. Wang, P., Lou, P. J., Leu, S., and Ouyang, P. (2002). Modulation of alternative premRNA splicing in vivo by pinin. Biochem. Biophys. Res. Commun. 294, 448-55. Wysocka, J., Milne, T. A., and Allis, C. D. (2005). Taking LSD1 to a New High. Cell 122, 654-8. Zhang, C., Dowd, D. R., Staal, A., Gu, C., Lian, J. B., van Wijnen, A. J., Stein, G. S., and MacDonald, P. N. (2003). Nuclear coactivator-6 2 kDa/Ski-interacting protein is a nuclear matrix-associated coactivator that may coupl e vitamin D receptor-mediated transcription and RNA splicing. J. Biol. Chem. 278 , 35325-36. Zhang, Q., Nottke, A., and Goodman, R. H. (2005). Homeodomain-i nteracting protein kinase-2 mediates CtBP phosphorylation a nd degradation in UVtriggered apoptosis. Proc. Natl. Acad. Sci. U S A 102, 2802-7. Zhang, Q., Piston, D. W., and Goodman, R. H. (2002). Regulation of corepressor function by nuclear NADH. Science 295, 1895-7. Zhang, Q., Yao, H., Vo, N., and Goodman, R. H. (2000). Acetylation of adenovirus E1A regulates binding of the transcriptional core pressor CtBP. Proc. Natl. Acad. Sci. U S A 97, 14323-8. Zhang, Q., Yoshimatsu, Y., Hildebrand, J., Fr isch, S. M., and Goodman, R. H. (2003). Homeodomain interacting protein kinase 2 promotes apop tosis by downregulating the transcriptional corepressor CtBP. Cell 115, 177-86. Zheng, L., Roeder, R. G., and Luo, Y. (2003). S phase activation of the histone H2B promoter by OCA-S, a coactiv ator complex that contains GAPDH as a key component. Cell 114, 255-66. Zimowska, G., Shi, J., Munguba, G., Jackson, M. R., Alpatov, R., Simmons, M. N., Shi, Y., and Sugrue, S. P. (2003). Pinin/DRS/memA interacts with SRp75, SRm300 and SRrp130 in corneal epithelial cells. Invest. Ophthalmol. Vis. Sci. 44, 4715-23. Zorio, D. A., and Bentley, D. L. (2004). The link between mRNA processing and transcription: communication works both ways. Exp. Cell Res. 296, 91-7.


BIOGRAPHICAL SKETCH Roman Alpatov was born in Ukraine and la ter moved to Moldova where he earned his bachelor’s degree in agronomy and plan t biology from the University of Moldova. He moved to the United States in 1994 and attended the University of Michigan from 1997 to 2000. In the year 2000 he graduated from the University of Michigan-Dearborn with a bachelor’s degree in biology. He started his doctorate studies at the University of Florida College of Medicine in 2000 and join ed Dr. Stephen P. S ugrue’s laboratory in 2001. 76