ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2018

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1404

Gene regulation by differentproteins of TGFβ superfamily

VARUN MATURI

ISSN 1651-6206ISBN 978-91-513-0172-3urn:nbn:se:uu:diva-334411

Dissertation presented at Uppsala University to be publicly examined in Room B42, BMC,Husargatan 3, Uppsala, Monday, 29 January 2018 at 09:15 for the degree of Doctor ofPhilosophy (Faculty of Medicine). The examination will be conducted in English. Facultyexaminer: Senior Researcher Anita Göndör (Karolinska Institutet).

AbstractMaturi, V. 2018. Gene regulation by different proteins of TGFβ superfamily. DigitalComprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1404.50 pp. Uppsala: Acta Universitatis Upsaliensis. ISBN 978-91-513-0172-3.

The present thesis discusses how gene regulation by transforming growth factor β (TGFβ) familycytokines is affected by post-translational modifications of different transcription factors. Thethesis also focuses on gene regulation by transcription factors involved in TGFβ signaling.

The importance of the poly ADP-ribose polymerase (PARP) family in controlling geneexpression in response to TGFβ and bone morphogenetic protein (BMP) is analyzed first.PARP2, along with PARP1, ADP-ribosylates Smad2 and Smad3, the signaling mediators ofTGFβ. On the other hand, poly ADP-ribose glycohydrolase (PARG) removes the ADP-ribosefrom Smad2/3 and antagonizes PARP1 and PARP2. ADP-ribosylation of Smads in turn affectstheir DNA binding capacity. We then illustrate how PARP1 and PARG can regulate geneexpression in response to BMP that signals via Smad1, 5. Over-expression of PARP1 suppressedthe transcriptional activity of Smad1/5. Knockdown of PARP1 or over-expression of PARGenhanced the transcriptional activity of BMP-Smads on target genes. Hence our data suggestthat ADP-ribosylation of Smad proteins controls both TGFβ and BMP signaling.

I then focus on elucidating novel genes that are regulated by ZEB1 and Snail1, two keytranscriptional factors in TGFβ signaling, known for their ability to induce EMT and cancermetastasis. Chromatin immunoprecipitation-sequencing (ChIP-seq) and targeted whole genometranscriptomics in triple negative breast cancer cells were used, to find binding regions and thefunctional impact of ZEB1 and Snail1 throughout the genome. ZEB1 binds to the regulatorysequences of a wide range of genes, not only related to cell invasion, pointing to new functionsof ZEB1. On the other hand, Snail1 regulated only a few genes, especially related to signaltransduction and cellular movement. Further functional analysis revealed that ZEB1 couldregulate the anchorage-independent growth of the triple negative breast cancer cells, whereasSnail1 could regulate the expression of BMP6 in these cells. We have therefore elucidated novelfunctional roles of the two transcription factors, Snail1 and ZEB1 in triple negative breast cancercells.

Keywords: EMT, Snail1, ZEB1, TGFβ, BMP, Gene regulation

Varun Maturi, Ludwig Institute for Cancer Research, Box 595, Uppsala University, SE-75124Uppsala, Sweden.

© Varun Maturi 2018

ISSN 1651-6206ISBN 978-91-513-0172-3urn:nbn:se:uu:diva-334411 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-334411)

To my family and my teachers

తృ భవ తృ భవ

ఆ ర భవ In a day when you don’t come across any problems, you can

be sure that you are going in a wrong path.

- Swamy Vivekananda

Main Supervisor Aristidis Moustakas, Professor Department of Medical Biochemistry and Microbiology Uppsala University, Sweden Associate supervisor Carl-Henrik Heldin, Professor Department of Medical Biochemistry and Microbiology, Uppsala Univeristy, Sweden Stefan Enroth, Senior Researcher Department of Immunology Genetics and pathology Uppsala University Sweden Opponent Anita Göndör, Associate Professor Department of Microbiology, Tumor and Cell Biology Karolinska Institute Sweden Committee Members Tanel Punga, Associate Professor Department of Medical Biochemistry and Microbiology Uppsala University Sweden Lars Forsberg, Associate Professor Department of Medical Biochemistry and Microbiology Uppsala University Sweden Lars-Gunnar Larsson, Professor Department of Microbiology, Tumor and Cell Biology Karolinska Institute Sweden

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Markus Dahl*, Maturi V*, Peter Lönn*, Panagiotis Papout-

soglou*, Agata Zieba, Michael Vanlandewijck, Lars P. van der Heide, Yukihide Watanabe, Ola Söderberg, Michael O. Hottiger, Carl-Henrik Heldin and Aristidis Moustakas. (2014). Fine tuning of Smad protein function by poly (ADP-ribose) polymerases and poly (ADP-ribose) glycohydrolase during TGFβ signaling PLoS One. Aug 18; 9(8):e103651).

II Watanabe, Y.*, Papoutsoglou, P.*, Maturi, V.*, Yutaro Tsu-bakihara, Hottiger, M.O., Heldin, C.-H. and Moustakas, A. (2016). ADP-ribosylating enzymes control bone morphogenetic protein signaling. Journal of Biological chemistry. Apr 21.

III Maturi, V. Enroth, S. Heldin, C.-H. and Moustakas, A. (2017) Genome-wide binding of transcription factor ZEB1 in triple neg-ative breast cancer cells. Manuscript.

IV Maturi, V. Enroth, S. Heldin, C.-H. and Moustakas, A. (2017) Genome-wide binding of transcription factor Snail1 in triple neg-ative breast cancer cells. Manuscript.

*Indicates authors contributed equally to the work

Reprints were made with permission from the respective publishers.

Other Papers not included in the thesis

I. Kahata K, Maturi V, Moustakas A, (2017). TGF-β Family Signaling in Ductal Differentiation and Branching Morphogenesis. Cold Spring Harb Perspect Biol. Mar.13 pii: a031997. 

II. Lehmann L, Aibara S, Hewitt G, Leitner A, Marklund E, Maturi V, Van der spoel D, Moustakas A, Boulton SJ, and Deindl S. (2017) Mechanistic insights into the auto-inhibition of the oncogenic chromatin remodeler alc1. Molecular Cell. (In Press).

Contents

Introduction ................................................................................................... 13 Overview .................................................................................................. 13 Gene regulation ........................................................................................ 14 PARP Family............................................................................................ 16 Gene regulation by signaling pathways .................................................... 19 TGFβ family ............................................................................................. 19 

TGFβ-dependent Smad signaling ........................................................ 20 Bone Morphogenetic Proteins ............................................................. 21 Negative regulation of TGFβ and BMP pathways ............................... 22 Non-Smad signaling ............................................................................ 23 

TGFβ in cancer ......................................................................................... 23 Epithelial-Mesenchymal Transition (EMT) ......................................... 24 Zinc finger E-box binding homeobox1 (ZEB1) .................................. 25 Snail1 ................................................................................................... 26 

Methods ........................................................................................................ 28 Proximity ligation assay ........................................................................... 28 Chromatin immunoprecipitation-sequencing ........................................... 29 Ampliseq transcriptomic array ................................................................. 30 CRISPR-Cas9 ........................................................................................... 30 

Present Investigation ..................................................................................... 32 Paper I ...................................................................................................... 32 

Aim ...................................................................................................... 32 Background .......................................................................................... 32 Methods ............................................................................................... 32 Results ................................................................................................. 32 Conclusions ......................................................................................... 33 

Paper II ..................................................................................................... 33 Aim ...................................................................................................... 33 Background .......................................................................................... 33 Methods ............................................................................................... 33 Results ................................................................................................. 34 Conclusions ......................................................................................... 34 

Paper III .................................................................................................... 34 Aim ...................................................................................................... 34 

Background .......................................................................................... 35 Methods & Results .............................................................................. 35 Conclusion ........................................................................................... 36 

Paper IV ................................................................................................... 36 Aim ...................................................................................................... 36 Background .......................................................................................... 36 Methods & Results .............................................................................. 36 Conclusion ........................................................................................... 37 

Future Perspectives ....................................................................................... 38 

Acknowledgements ....................................................................................... 40 

References ..................................................................................................... 43 

Abbreviations

3-AB 3-Aminobenzamide ALC1 Amplified in Liver Cancer 1 ALK Activin receptor-like kinase ADP Adenosine 5’-diphosphate AP-1 Activator protein 1 ARH3 ADP-ribosyl hydrolase 3 ARTD1 ADP-ribosyltransferase diphtheria toxin-like 1 BMP Bone morphogenetic protein CAS9 CRISPR-associated protein 9 CBD Co-activator binding domain CPED1 Cadherin like and PC-esterase domain containing 1 ChIP-Seq. Chromatin immune precipitation-Sequencing CRB1 Crumbs 1, cell polarity complex component CRISPR Clustered regularly interspaced palindromic repeats CSC Cancer Stem cells CTCF CCCTC-binding factor DLG2 Discs large MAGUK scaffold protein2 DNA Deoxyribonucleic acid DUB De-ubiquitinase EMT Epithelial-mesenchymal transition ERK Extracellular signal-regulated kinase FAT3 FAT atypical cadherin 3 GDF Growth differentiation factor GC-rich Guanine–cytosine rich H1 Histone 1 H2A Histone 2 subunit A H2B Histone 2 subunit B H3 Histone 3 H4 Histone 4 H3K4me3 Histone 3 Lysine 4 tri methylation HaCaT Human immortalized keratinocyte cell line HD Homeo domain HMGA High mobility group A I-smads Inhibitory-smads JNK c-Jun N-terminal kinase kDa KiloDalton KDM5B Lysine de-methylase 5B

lncRNA long non-coding ribonucleic acid LAP Latency-associated peptide LLC Large latency complex LTBP Latent TGF-β binding protein MAPK Mitogen activated protein kinase MH Mad-homology MET Mesenchymal-epithelail transition NAD Nicotinamide adenine dinucleotide PARG Poly(ADP-ribose) glycohydrolase PARP-1 Poly(ADP-ribose) polymerase-1 PARP-2 Poly(ADP-ribose) polymerase-2 PARylation Poly(ADP-ribosyl)ation PLA Proximity ligation assay PPFIA1 PTPRF interacting protein alpha 1 PPMIA Protein phosphatase magnesium-dependent IA PP2C Protein phosphostase 2A catalytic subunit PRRX1 Paired related homeobox 1 PTEN Phosphatase and tensin homolog RCA Rolling circle amplification RMR Arginine-methionine-arginine domain RNA Ribonucleic acid R-Smads Receptor Smads rRNA ribosomal ribonucleic acid miRNA micro ribonucleic acid ncRNA non-coding ribonucleic acid SBD Smad binding domain Ser Serine Smad Small-mothers against decapentaplegic Smurf Smad ubiquitylation regulatory factor tRNA transfer ribonucleic acid TAK1 TGFβ activated kinase 1 TIMP3 Tissue inhibitor of metalloproteinase 3 TEMPAI Transmembrane prostate androgen-induced TENM2 Tenurin transmembrane protein 2 TGFβ Transforming growth factor β TGFβR TGFβ receptor TSS Transcription start site Tyr Tyrosine ZEB1 Zinc finger E-Box binding homeobox 1 ZFD Zinc finger domain ZO1 Zonula occludins 1

13

Introduction

Overview Cells divide by following a highly controlled cyclical mechanism. Deoxyri-bonucleic acid (DNA) and proteins take a major share in the function of cells (Spellman et al., 1998). DNA carries the information that dictates the structure and function of RNAs and proteins. Besides, proteins can regulate the tran-scription of different genes. During these processes of controlling each other both DNA and proteins undergo different modifications. Histones are bound to DNA in the form of octamers called nucleosomes to help compact the long DNA into the tightly folded chromatin. Modifications in the DNA, such as DNA methylation, can affect the transcription of genes. This is achieved by proteins, which chemically alter the DNA, causing DNA methylation, or chemically alter histones causing epigenetic modifications which can persist through many generations of cell divisions without changing the actual genetic sequence (Greer and Shi, 2012). Post-translational modifications in histones, such as acetylation or methylation, can either open or wind up the nucleosome to allow or prohibit the gene expression machinery from accessing the region. Additional modifications on histones, such as phosphorylation, ADP-ribosyl-ation and ubiquitination are also responsible for gene regulation. In the major-ity of the cases, these modifications are reversible and this allows switching on or off the expression of different genes so that the fate of the cells is con-trolled.

Post-translational modifications play additional important roles in cell sig-naling, which are important for cell physiological processes, such as differen-tiation, apoptosis, migration and invasion, by regulating the differential ex-pression of many genes. ADP-ribosylation is one such post-translational mod-ification that is well studies in this context. The poly (ADP-ribose) polymer-ase (PARP) family refers to a group of 18 proteins that catalyze ADP-ribosylation of other proteins (Amé et al., 2004). PARP1 and PARP2 ADP-ribosylate and PARG de-ADP-ribosylates proteins. PARP1 and PARP2 func-tion can be activated upon DNA damage by radiation or by chemicals (Alvarez-Gonzalez and Althaus, 1989; Davidovic et al., 2001).

Epithelial mesenchymal transition (EMT) is a process where epithelial cells undergo a change in differentiation that alters the morphological and physio-logical properties of the cells. EMT occurs in different contexts, including embryonic development, wound healing and cancer metastasis. During this

14

process of transition, cells tend to lose their epithelial, sedentary features and attain mesenchymal, motility and solitary features (Kalluri and Weinberg, 2009). In the context of cancer, the primary epithelial tumor cells undergo EMT, invade new organs and manage to establish their secondary tumors by expressing some stem cell features, which helps these metastatic cells to attain tumor-initiating-capacity. These cells are named cancer stem cells.

Transforming growth factor β (TGFβ) isoforms and bone morphogenetic proteins (BMPs) are selected cytokine families that have important roles in EMT and embryonic development. Downstream of these cytokines, cell sur-face receptors activate signaling by Smads and other proteins, such as protein kinases. Certain Smads are phosphorylated, which in turn results in their nu-clear import and interaction with other transcription factors to either turn on or turn off the transcription of various genes specially those related to EMT, invasiveness, apoptosis and other processes. Snail1 and ZEB1 are two such transcription factors that cooperate with TGFβ signaling and are implicated in EMT.

This thesis will focus on how activated PARP1/PARG can affect TGFβ/BMP signaling by effecting the post-translational modifications of some of the transcription factors, like Smads, along with novel sites of gene regulation by Snail1 and ZEB1 in triple negative breast cancer cells.

Gene regulation Multicellular organisms consist of several types of cells. Each cell carries ge-netic information that is transferred to the next generation by cell division. The genetic information is contained in DNA, which is a discrete and long polymer of deoxyribonucleotides forming functional stretches or regions called genes. Functionally, a gene is an assembly of regulatory DNA se-quences linked to an RNA-coding sequence. The RNA-coding part of the gene may code for protein (mRNA) or not (rRNA, tRNA, lncRNA, miRNA etc). The RNA-coding part of the gene includes exons, which are included in the mature RNA copy (mRNA or lncRNA for example) and introns which are spliced out. The mRNA sequences can be divided into a 5’-untranslated region (5’ UTR) including a regulatory structure and a starting site for ribosomal translation during protein synthesis, a middle protein-coding sequence, and a 3’ UTR with additional regulatory function and a poly-adenylation signal. Many RNAs do not contain any translatable open reading frames and do not code for proteins; they are called non-coding RNAs (e.g, rRNA, tRNA, lncRNA, miRNA and ncRNA). Ahead of the RNA-coding sequence is the region where transcription starts, called as transcription start site (TSS), cor-responding to the first nucleotide of the RNA. Upstream of the TSS a DNA segment known as gene promoter is located. RNA polymerase II binds to this region initiating the transcription of mRNA, which is processed via RNA

15

splicing and serves as a template for protein synthesis in the case of mRNA. The gene promoter is also where transcription factors bind controlling the gene expression and communicating either initiation or termination signals to the RNA polymerase, starting or stopping transcription. The gene promoter usually lies very near the TSS, on the average 30-60 bp upstream of the TSS, and can also extend downstream of the TSS into the first exon and/or first intron. At various distances from the TSS and the gene promoter lie additional regulatory sequences known as enhancers or silencers which positively or neg-atively affect gene transcription.

Histone proteins comprise dimers of H2A, H2B, H3 and H4 packed as an octamer called nucleosome, around which the DNA is wrapped. Different nu-cleosomes are connected by a linker DNA, and Histone H1 locks the nucleo-some-bound DNA. Histone modifications play a vital role in gene regulation by altering the chromatin structure. Histone modification is a dynamic process carried out by several enzymes, e.g., histone acetyl-transferases, de-acety-lases, methyl-transferases and demethylases. H3 and H4 post-translational modifications play a vital role in gene silencing and gene activation and splic-ing (Andersson et al., 2009). Lysine (K)3 and (K)27 in the N-terminal of H3 undergo post-translational modifications that recruit other proteins which dif-ferentially regulate gene expression and the formation of heterochromatin. H3K27 methylation is considered as a marker for silencing of gene expression and H3K27 acetylation is an activating marker (Greer and Shi, 2012; Venneti et al., 2013). These modifications are often used as markers to define tran-scriptionally active or inactive areas during genome-wide studies (Ernst et al., 2011). Studies of the global levels of histone methylation have revealed an elevated level of methylation signature in cancer patients (Greer and Shi, 2012).

Transcription factors interact with genes via specific DNA sequence mo-tifs. These motifs are sometimes diverge in sequence and shared by many dif-ferent genes, whose expression is regulated by the same transcription factors. Transcription factors recruit many other proteins to the enhancer or silencer elements to help recruit or displace the RNA polymerase to or from the pro-moter and TSS. Specific transcription factor binding to enhancer elements can activate gene expression, and binding to silencer elements can repress gene expression. Transcription factors are classified into various families based on their structure and function; for example, ZEB1 belongs to the family of homeo-box and zinc-finger domain proteins consisting of C2H2 type zinc fin-gers responsible for DNA binding and a homeo-domain responsible for bind-ing to other transcription factors which control the fate of ZEBs function as a repressor or as an activator of transcription. Snail1 belongs to a family of zinc-finger transcription factors, which have a zinc-finger DNA-binding domain, a repressor domain known as SNAG and a serine-rich domain which is an acti-vator domain. In addition to transcription factors there are other proteins like chromatin factors that regulate gene expression, such as the PARP family of

16

proteins. PARP family members were demonstrated to regulate gene expres-sion both positively and negatively in a cell differentiation- and gene-depend-ent context (Huang et al., 2011; Lönn et al., 2010).

PARP Family The PARP family is a group of multifunctional proteins that can polymerize ADP-ribose into linear or branched polymers, which affect certain cellular processes, such as apoptosis or proliferation, by regulating gene expression and the response to DNA damage (Chaitanya et al., 2010; Kaufmann et al., 1993; Tewari et al., 1995). The family consists of 18 proteins (Figure 1) which have conserved functional domains, based on which, the family is divided into different groups (Schreiber et al., 2006). The DNA-dependent PARPs com-prise PARP1 and PARP2 which become activated by DNA damage and reg-ulate DNA repair (Dantzer et al., 1999). PARP1 and PARP2 auto-ADP-ribo-sylate and each other (Ogata et al., 1981). PARP3 is a well-studied protein in the context of the cell cycle. It is reported to have close functional relation to PARP1 and can auto-ADP-ribosylate and also ADP-ribosylate PARP1 (Augustin et al., 2003). PARP4 is the largest of all the PARP family proteins of about 193 kDa molecular mass. PAPR4 is reported to form complexes with telomere guarding proteins to maintain the telomere length (Liu et al., 2004). Tankyrases comprise PARP5a and PARP5b that guard the stability of telo-meres in association with telomere repeat binding factor 1 and ankyrin-related proteins. The CCCH-PARPs comprise PARP7, 12 and 13 which have a C-X8-C-X5-C-X3-H domain that regulates cell-cycle and proliferation-related genes (Petrucco, 2003). The macro-PARPs comprise PARP9, 14 and 15, and are capable of RNA binding via their arginine-methionine-arginine (RMR) do-main. There are more PARP family members that do not fall into the above subgroups, i.e., PARP6, 8, 10, 11 and 16. The common biochemical property of PARPs, the synthesis of a poly-ADP-ribose (PAR) polymer is often referred to as PARylation (Schreiber et al., 2006). PARP1 and PARP 2 interacts with each other and have some structural similarity. PARP1 is activated by DNA damage and PARP2 interacts with PARP1 and is involved in different single and double-stranded DNA repair processes (Swindall et al., 2013).

PARP2 lacks a DNA-binding domain and performs its functions in the nu-cleus by interacting with PARP1. During the cell cycle, PARP1 along with PARP2 are activated by the DNA damage signal and alter the function of pro-teins by attaching linear or branched PAR chains which hinder the DNA-bind-ing of the target proteins (Petrucco, 2003).

PARylation is the main function of DNA-dependent PARPs. The poly-ADP-ribosyl chains are derived from the donor NAD+; PARPs attach the first ADP-ribose to the target protein and then polymerize multiple ADP-ribose units to build polymers with 200 ADP-ribose units or longer. What defines the

17

length of these polymers remains unknown (Schreiber et al., 2006). Gene reg-ulation by PARP1 was initially reported when PARP1 was found to act as a repressor of DNA nick-dependent transcription (Petrucco, 2003). PARP1 can recruit several co-transcription factors acting as a scaffold protein (Kraus and Lis, 2003). Activated PARP1 is cleaved by Caspase 7 which releases PARP1 from chromatin resulting in chromatin de-condensation, a process that initi-ates apoptosis (Kaufmann et al., 1993). PARP1 co-localizes with histone H1 and PARylates it. PARylation of histone H1 results in H1 dislodging, which allows a conformational change in the chromatin, an important step in the pro-cess of gene regulation (Tulin, 2003). PARP1 also interacts with H2A and PARylation of H2A could induce conformational changes of the nucleosomes exposing H3 and H4 which are vital in organizing the chromatin structure and regulating the accessibility of other transcriptional factors on to the down-stream genes (Kraus and Lis, 2003; Tulin, 2003). PARylation of histone de-methylase enzymes such as lysine (K) de-methylase 5B (KDM5B) by PARP1 controls the de-methylation pattern on H3K4, which regulates transcription (Krishnakumar and Kraus, 2010; Verdone et al., 2015). PARP1 binds to some of the chromatin modulation factors like Amplified in Liver Cancer1 (ALC1) and CCCTC-binding factor (CTCF) to facilitate the insulation or exposure of diverse transcriptional sites along the genome (Yu et al., 2004). In our study, we have shown that PARP1 can regulate cells stimulated with TGFβ or BMP, and in cooperation with Smads target gene expression.

PARG, is a glycohydrolase that detects the poly-ADP-ribose chains and hydrolyzes them into monomeric ADP-ribose units (Amé et al., 2004). By counteracting the PARP activity, PARG constitutes an important factor in fine-tuning the DNA repair mechanism which is vital for cell cycle regulation (Alvarez-Gonzalez and Althaus, 1989). Besides PARG, ADP-ribosyl hydro-lase 3 (ARH3) is another enzyme that can perform the same function as PARG (Rosenthal et al., 2013). Some reports suggest that the glycohydrolase activity is always high which suggests that PARG is constantly active in the nucleus and cleaves off poly-ADP-ribose chains from various proteins (Kraus and Hottiger, 2013)

18

Figure1: Classification of PARP family members. The figure shows the structural and functional classification of 18 PARP family members. It also illustrates various domains present in individual proteins(Schreiber et al., 2006). Reprints were made with permission from the journal.

19

Gene regulation by signaling pathways Cells go through several checkpoints during the cell cycle to counter errone-ous mutations during cell division. If mutations occur, they could promote the development of cancer or other diseases. Cancer is characterized by an un-controlled division of cells. In a normal context, cells receive growth signals from the surrounding micro-environment via growth factors. Mutations in genes involved in growth control can enhance signaling pathways that could lead to both tumorigenesis; and such genes are known as oncogenes inactivate important signaling molecules that stop the cell cycle or induce apoptosis, thus promoting tumorigenesis. Mutations in genes known as tumor suppressors. There are several growth factors that promote tumorigenesis by affecting the regulation of certain genes. Such growth factors include TGFβ and BMP which affect genes controlling growth, apoptosis, metastasis, etc.

TGFβ family In multi-cellular organisms cells need to communicate with each other to carry out certain functions, such as gene regulation by signal transduction. Such cell signaling could be autocrine, paracrine or endocrine. The TGFβ family signal-ing occurs mainly in autocrine and paracrine manners. The TGFβ family is a group of 33 structurally related cytokines (Moustakas and Heldin, 2009). The prototype family member, TGFβ1, was identified as a secreted protein which induced anchorage-independent growth of cells (Roberts et al., 1981). Other family members are TGFβ 2, TGFβ 3, BMPs, inhibins, activins, nodal and growth differentiation factors (GDFs) (Wakefield and Hill, 2013). Members of the TGFβ family play major roles in embryogenesis and tissue homeostasis. Proteolytic cleavage of their N-terminal pro-domains results in the formation of mature ligands, most of which are disulfide-linked homodimers and some form heterodimers. TGFβ family members exert their cellular effects by bind-ing to transmembrane serine/threonine kinase receptors which are divided into type I and type II receptors (Figure 1). There are seven type I receptors, also known as activin receptor-like kinases (ALKs), e.g. TGFβRI/ALK5, ACTRIA/ALK2, ACTRIB/ALK4 and ACVRLI/ALK1, and five type II re-ceptors, e.g. TGFβRII and BMPRII. Each of the type I and type II receptors form homo-dimers and ligand binding induces hetero-tetrameric receptor complex. This thesis will focus mainly on TGFβ and BMP signaling.

20

Figure 2: Smad dependent gene regulation. The figure illustrates the binding of ligands i.e, BMP, TGFβ or activin /Nodal to their receptors and the activation of re-ceptor activated Smads by the activated type I receptors, followed by oligomerisa-tion with Smad4 and nuclear import of Smad complexes and gene regulation.

TGFβ-dependent Smad signaling Each of the TGFβ isoforms is synthesized as a latent form composed of the homo-dimeric N-terminal latency-associated peptide (LAP), which is non-co-valently linked to the C-terminal homo-dimeric mature polypeptide; often this small latent complex forms a larger complex including latent TGFβ binding protein (LTBP) which is disulfide-linked to LAP (ten Dijke and Arthur, 2007). A large latency complex (LLC) is then secreted into the matrix allowing the release of active TGFβ from LLC when the LTBP is pulled by mechanical forces like fibronectin and fibrillin fibers (Saharinen et al., 1999; Shi et al., 2011). TGFβ signaling is initiated by the binding of mature TGFβ to TGFβRII, which then forms a hetero-tetrameric complex with TGFβRI. Within this com-plex, TGFβRII activates the TGFβRI by phosphorylation at several serine res-idues located in a glycine/serine-rich domain (GS) in TGFβRI, causing a con-formational change in TGFRI, which exposes the protein loop known as L45, creating a binding site for receptor-activated (R)-Smads (receptor-activated Small-mothers against decapentaplegic) (ten Dijke & Hill, 2004). TGFβRI binds and phosphorylates the R-Smads, Smad2 and Smad3, in their C-terminal serine-X-serine motif. Activated R-Smads, then form complexes with co-Smad (Smad4), which translocate into the nucleus to function as transcrip-tional regulators. Smad4 is a common component in all TGFβ family Smad

21

pathways, including the BMP signaling pathways, where Smad4 associates with R-Smads, Smad1, Smad5 and Smad8. Smad4 lacks a C-terminal ser-X-ser motif, and is not phosphorylated by TGFβRI. Smad3 along with Smad4 bind to DNA directly via their N-terminal Mad homology 1 (MH1) domains, whereas Smad2 has a stretch of unique amino acids in its MH1 domain that hinders direct DNA binding (ten Dijke & Hill, 2004). The complex of Smad3 and Smad4 binds to other transcription factors to either repress or activate transcription of downstream genes. The DNA region to which Smads bind is called Smad-binding element (SBE), which is either a stretch of repeated 5’-AGAC-3’ sequences or a GC-rich motif (Massagué et al., 2005). Inhibitory Smads (I-Smads; Smad6 and Smad7) are upregulated by TGFβ stimulated transcription and act in a negative feedback loop by binding to the receptors and competing with binding of R-Smads, and by recruiting ubiquitin ligases and phosphatases to the receptors (Derynck and Zhang, 2003).

Bone Morphogenetic Proteins The first BMP ligands were identified as cytokine that could induce the for-mation of bone and cartilage tissue in rats (Urist, 1965). Later studies have shown that BMPs also have vital roles during embryonic development, as they are involved in the development of teeth, skin, kidney, hair, muscle, hemato-poietic cells and neurons, iron metabolism and vascular homeostasis (Wu and Hill, 2009). Knockout of different components of the BMP signaling machin-ery in mice resulted in various embryonic malfunctions, illustrating the im-portance of BMP signaling during embryogenesis (Wu and Hill, 2009). It is also striking that the BMP signaling pathways, like the TGFβ signaling path-ways, are well-conserved from invertebrates to vertebrates (Miyazono et al., 2010).

Similar to TGFβ, BMP signaling depends on a tetrameric complex of ser-ine/threonine kinase receptors. The type I receptors are ALK1, 2, 3, 6 and the type II receptors are ActRIIA, ActRIIB and BMPRII (Figure 2). The BMP ligands are classified into several groups based on their structural similarly, such as BMP-2/4, BMP-5/6/7/8 and BMP-9/10 (Wakefield and Hill, 2013). BMP1 is a metalloprotease and is not a member of the BMP family, despite its name. Upon BMP ligand binding to the receptor complex, a constitutively active type II receptor phosphorylates the type I receptor. Then, activated type I receptors, subsequently bind to Smad1, 5 and 8 and phosphorylate them at their C-terminal serine-X-serine motifs. Similar to TGFβ signaling, the phos-phorylated BMPR-Smads are translocated to the nucleus by forming a hetero-oligomeric complexes with Smad4 (Figure1). BMP R-Smads bind to a char-acteristic GC-rich sequence GCCGnCGC motif in DNA (Miyazono et al., 2010). Upon BMP stimulation, R-Smads and Co-Smad along with different co-transcriptional factors regulate expression of different genes, including the I-Smads that also regulate BMP signaling.

22

Negative regulation of TGFβ and BMP pathways Sequestering of signaling is a biologically important process to maintain the balance in signaling. As signal activation and transduction occur in various steps, signal abrogation also occurs in various steps. Post-translational modi-fications play important roles in negative regulation of the signaling pathways.

Figure3: Effect of PARP turnover on TGFβ and BMP signaling. The figure il-lustrating the activities of PARP and PARG in the nucleus and their effects on ca-nonical Smad based transcriptional activity of TGFβ and BMP signaling.

TGFβRII is a protein kinase that is not only phosphorylated on serine residues but, autophosphorylation on tyrosine residue is important for signaling. Mu-tation of the autophosphorylated residues Tyr259, Tyr336 and Tyr424 in the receptor blocks the activity of TGFβRII and signaling (Lawler et al., 1997). De-phosphorylation of TGFβRI by protein phosphatases, like PP1 and PP2A, leads to conformational changes that block binding of R-Smads to the receptor complex (Bengtsson et al., 2009). Ubiquitination of the TGFβR complex is an active process where Smad7, a member of the I-Smads, plays a vital role. Smad7 recruits E3 ligases like Smurf1, Smurf2, NEDD4-2 and WWP1 to the receptors and allows ubiquitination of TGFβRI and TGFβRII (Ebisawa et al., 2001; Hayashi et al., 1997; Kavsak et al., 2000; Komuro et al., 2004; Kuratomi et al., 2005).

There are two I-Smads, Smad6 and Smad7. Smad6 inhibits BMP signaling, whereas Smad7 inhibits both TGFβ and BMP signaling (Goto et al., 2007; Hanyu et al., 2001). Dephosphorylation of R-Smads is an active negative reg-

23

ulatory process in termination of TGFβ signaling. PPMIA/PP2Cα de-phos-phorylates Smad2 and 3 which leads to dissociation of the complex between R-Smads and Co-Smad.

BMP signaling is controlled in similar ways as TGFβ signaling. Phospha-tases like Dullard and PP2A dephosphorylate and degrade BMP receptors, which terminates the BMP signaling at the receptor level (Bengtsson et al., 2009).

Inhibitory Smads, i.e, Smad6 and Smad7, are responsible for negative sig-naling by their binding to receptor complexes, blocking Smad1/5/8 phosphor-ylation by BMP type I receptors. Smad4 is an important Co-Smad for both TGFβ and BMP signaling which is ubiquitinated at K507 and K519, leading to termination of transcription. Members of the family of deubiquitinases (DUB’s) also modulate the TGFβ pathway by removing mono- and poly-ubiq-uitination chains from the target proteins, including receptors and Smads (Zhang et al., 2014a).

PARPs can PARylate Smads which results in loss of DNA binding capacity of the Smads resulting in sequestering of Smad-mediated transcription (Dahl et al., 2014; Lönn et al., 2010). Our recent studies show that this is a common mode of post-translational modification which negatively regulates both TGFβ and BMP signaling (Figure 3).

Non-Smad signaling Downstream of TGFβR activation, several pathways are triggered in a non-Smad-dependent manner. Several non-Smad-dependent pathways include ac-tivation of AKT by PI3K through a TRAF6 dependent manner. This is trig-gered by ubiquitination of TRAF6 on Lys63 site (Hamidi et al., 2017; Yang et al., 2009) or by induced expression of miR 216a/217 by the TGFβRII (Kato et al., 2009). TRAF6 subsequently activates TAK1-p38/JNK pathways and promote apoptosis (Sorrentino et al., 2008). Both TGFβ and BMP are respon-sible in activating RhoA in a Smad independent manner which in turn acti-vates JNK and p38 (Bhowmick et al., 2001; Voorneveld et al., 2014). How-ever roles of JNK activation by RhoA is yet to be investigated for importance in TGFβ and BMP signaling.

TGFβ in cancer TGFβ has dual roles in cancer. During initial stages of cancer TGFβ acts as a tumor suppressor by promoting cell cycle arrest, including enhanced expres-sion of CDK inhibitors, such as p15INK4b, p21CIPI and p27Kip1, and downregula-tion of c-Myc via C/EBPβ (Gomis et al., 2006; Seoane et al., 2001). TGFβ also induces apoptosis by enhanced expression of DAPK, GADD45β and

24

Smad-mediated transcription of the KP2 gene resulting in activation of Bim (Jang et al., 2002; Ramesh et al., 2008; Takekawa et al., 2002).

At later stages of tumor progression TGFβ acts as a tumor promoter. TGFβ initiates the tumor enhancing function by suppressing the immune system. This is achieved by both, suppressing activation and proliferation of T-cells by autocrine secretion by T-cells or in a Smad2/3-ATF1 mode of transcription (Chen, 2005; Thomas and Massagué, 2005; Zhang et al., 2005). Investigations prove that change in function of TGFβ from tumor suppressor to tumor pro-moter is because of molecules such as TEMPAI and PTEN (Singha et al., 2014). TGFβ promotes cancer stem cell properties, stimulating dissemination of the primary tumor, proliferation and metastasis (Oskarsson et al., 2014). TGFβ’s enhances CSC properties of primary tumors by cooperating with other signaling pathways, like WNT (Scheel and Weinberg, 2011; Scheel et al., 2011). Additional contribution of TGFβ to cancer progression occurs by Epi-thelial Mesenchymal Transition (EMT), which is discussed below.

Epithelial-Mesenchymal Transition (EMT) Epithelial-mesenchymal transition (EMT) is a process of trans-differentiation and morphological adaptation of epithelial cells to a mesenchymal phenotype due to induction of certain transcriptional processes (Kalluri and Weinberg, 2009). Epithelial cells are sedentary and polar. After stimulation with certain cytokines, leading to epigenetic and transcriptional regulation of genes, epi-thelial cells loose polarity and become more migratory by changing their dif-ferentiation to become more mesenchymal-like (Figure 4). EMT occurs dur-ing several physiological and pathological occasions, e.g. during gastrulation, wound healing and cancer metastasis (Scanlon et al., 2013). During gastrula-tion, a primitive streak of cells undergo EMT and helps in forming mesoderm and endoderm. During wound healing, the cells that surround the wound un-dergo EMT to heal the disrupted tissue, which also can cause tissue fibrosis if the process is not controlled. EMT of cancer cells occurs when oncogenic mu-tations or epigenetic alterations promote epithelial cells to become more mes-enchymal by activating specific transcription factors (see below). EMT, pro-motes primary tumor cells to enter the blood-stream and establish secondary metastatic tumor in other organs (Scanlon et al., 2013).

EMT has been classified in three different types. The type 1 form of EMT is when epithelial cells during embryonal development undergo changes to produce mesenchymal cells which can in-turn generate epithelial cells by Mesenchymal-epithelial transition (MET). The type II form of EMT involves production of mesenchymal cells during wound healing, and fibrosis, a sub-sequent response to inflammation process. Type III EMT occurs during tu-morigenesis and promotes invasion of human cells into the blood stream and promotion of metastasis (Kalluri and Weinberg, 2009). Early EMT does not lead to a complete morphological and physiological change of epithelial cells

25

to mesenchymal cells but demonstrates changes such as suppressed E-Cad-herin but not enhanced N-Cadherin (Chu et al., 2006). This also causes change in functional aspects of epithelial-like cells exhibiting invasion and drug re-sistance; such changes in functionality inspired researchers to investigate more on the molecular pathways related to the transition.

Figure 4: Epithelial Mesenchymal Transition. The figure illustrates the sequential process of epithelial cells gaining mesenchymal features by losing apical-basal po-larity (Micalizzi, Farabaugh, & Ford, 2010). Reprints were made with permission form the journal.

Factors that promote EMT include TGFβ, WNT, and FGF and other tyrosine kinase receptor ligands (Zarzynska, 2014). These cytokines trigger the down-stream transcriptional regulation of various genes, the most important being a group of transcriptional factors, e.g. Snail1, Slug, ZEB1, ZEB2, Twist, Prrx1, and HMGA2 which mediate EMT, hence called as EMT-TFs (Heldin et al., 2012; Ocaña et al., 2012). Apart from regulating downstream genes, EMT-TF’s also regulate other EMT-TFs, such as co-regulation of Snail1 and Twist (Dave et al., 2011). In order of the EMT-TFs to trigger the mesenchymal phe-notype in epithelial cells, they bind to epithelial and mesenchymal genes to repress and enhance their expression of the respective genes. Upon activation by different stimuli EMT-TFs along with other co-transcriptional factors bind to target genes which are either suppressed or activated. Transcriptional regu-lation of EMT related genes results in production of cytokines and chemo-kines; dissolution of tight junction proteins, such as ZO-1 and occludin (Polette et al., 2007); loss of adherent junction proteins such as E-cadherin (Tan et al., 2015) and EpCam; and induction of mesenchymal components to facilitate motility of cells such as reorganization of actin filaments and release of fibronectin (Zeisberg and Neilson, 2009).

Zinc finger E-box binding homeobox1 (ZEB1) ZEB1 is a zinc finger E-box-binding homeobox 1 protein first identified in repressing the inter-leukin2 (IL2) gene in T-lymphocytes (Williams et al., 1991). ZEB1 consists of various domains, including an N-terminal co-activa-tor binding domain (CBD), an N-terminal zinc finger domain (ZFD), a Smad binding domain (SBD), a homeo domain (HD), a CtBP interaction domain

26

(CID) and a C-terminal zinc-finger domain. ZEB1 is known to bind to the sequence 5'-CANNTG-3’ in the E-box region of various genes (Figure 5) (Wellner et al., 2010). Upon TGFβ stimulation, ZEB1 along with Smads bind to the promoter region of various EMT-related genes and either suppresses or induces these genes.

Figure 5: ZEB1 gene and protein organization. The figure illustrates the location of domains in ZEB1 protein along with respective location of exons and amino acid number.

Given the importance of the E-Cadherin (CDH1) gene in the field of EMT, it is worth noting that ZEB1 is a major factor that represses CDH1 during the onset of EMT (Aigner et al., 2007). ZEB1 represses polarity genes such as Crumbs2 and Lgl2, and represses laminin-332 (LAMC2) and integrin-β4 (ITGB4) genes, contributing to the invasiveness in prostate cancer cells (Drake et al., 2010). ZEB1 also plays a key role in activating the mesenchymal phe-notype by cooperating with the YAP transcription factor of the Hippo path-way. By inducing antagonists of BMP ligands such as noggin, follistatin and chordin-like 1, ZEB1 contributes to bone-specific metastasis in breast carci-noma (Mock et al., 2015). Studies also show that knock-out of ZEB1 in mice results in a lethal phenotype with defects in embryogenesis resulting in skele-tal and thymic defects (Takagi et al., 1998). Despite the importance of ZEB1, genome-wide binding of ZEB1 has not yet been investigated in highly meta-static breast carcinoma.

Snail1 The Snail family of transcription factors comprises three vertebrate members, namely Snail (Snail1), SLUG (Snail2) and SMUC (Snail3) (Nieto, 2002). The structure of the proteins comprises a well conserved N-terminal motif, PRS-FLV, which is found in repressive transcription factors, followed by a nuclear localization signal and 3 typical and one atypical zinc-finger domains which bind E-box DNA sequences e.g. 5’-CACCTG-3’, or its complementary forms, 5’-CAGGTG-3’ (Figure 6).

27

Figure 6: Snail1 gene and protein organization. The figure shows different do-mains of Snail protein with their corresponding exons and respective amino acids.

In the context of EMT, upon TGFβ stimulation, Snail expression is first in-duced, Snail1 then binds to Smads and represses epithelial genes, including E-cadherin (CDH1), coxsackie virus and adenovirus receptor (CAR) and oc-cludin (OCD)(Vincent et al., 2009). In addition to CDH1, Snail1 is involved in modulating ECM by regulating MUC1, tight junction protein claudin1 and epithelial polarity gene crumbs3 (Guaita et al., 2002; Mar Inez-estrada et al., 2006; Whiteman et al., 2008). Snail1 represses various other genes, including PTEN, p53, HDAC and PRC2 (Escrivà et al., 2008). Snail1 expression in can-cer-associated fibroblast is observed to drive the plasticity of adjacent carci-noma cells (Baulida and García de Herreros, 2015), suggesting that Snail1 regulates cytokines. Snail1 is one of the hallmark of transcriptional regulators of EMT (De Herreros et al., 2010), which is activated in early EMT driven by TGFβ. Early activation on Snail1 in breast cancer cells also effects other EMT-TF’s like ZEB1, positively by forming a complex with Twist1. High expression of Snail1 in breast cancer is also correlated with high-grade malig-nancy, invasion and metastatic potential in breast cancers (Dave et al., 2011; Delhalle et al., 2002; Guaita et al., 2002).

28

Methods

In this section for better understanding of the methods, I discuss only the meth-ods whose background was not sufficiently explained in individual papers.

Proximity ligation assay Proximity ligation assay is an in situ detection technique used to detect ex-tremely low amount (40*10-21 mol) of proteins expressed intracellular (Fredriksson et al., 2002). PLA was later adapted to measure the interactions between intracellular or secreted proteins (Söderberg et al., 2006) and has been used to detect interactions in TGFβ signaling (Zieba et al., 2012). This is considered to be a sensitive assay considering the detection compared to other techniques (Gustafsdottir et al., 2005). Proximity ligation assays are de-signed to detect proteins by specific antibodies (either for same protein or for different proteins) with an attached aptamer DNA sequence. At least, two an-tibodies attached to DNA strands and a complementary circular DNA mole-cule with a methyl mismatch were introduced to facilitate the loop formation for rolling circle amplification (RCA) reaction using the antibody-attached ol-igonucleotides as primers (Schweitzer et al., 2000). Antibodies deployed on fixed cells recognize the proteins and once the proteins are in close proximity (30 nm), the oligonucleotides form a double stranded hybrid with the circular DNA template. Using a ligase and DNA polymerase allows the DNA primers to form a circular loop that is further amplified by RCA. Visualisation of the RCA product is carried by hybridizes complementary oligonucleotide with a fluorescent tag to the amplified DNA blob (Figure 7). Sensitivity and quanti-fication serves as a strength of the procedure, while antibody specificity re-mains the threshold of the quality of the assay.

I established and optimized this technique in my lab and used it in Paper I and Paper II to visualize the interactions between PARP family members and also to measure the post-translational modification of Smad3. This also al-lowed us to analyse the localization of the PARylated Smads and PARPs in the cell.

29

Figure 7: Proximity ligation assay. The figure illustrates the detection of interact-ing proteins by primary antibodies, probing with secondary antibodies covalently bound to oligonucleotides and detection of complexes by rolling circle amplification after hybridization of oligonucletoides with a fluorescent tag (red stars) to a unique DNA sequence incorporated into the circle (shown in red) (Zieba et al., 2012). Re-prints were made with permission form the respective journal.

Chromatin immunoprecipitation-sequencing Chromatin immunoprecipitation (ChIP) is an assay developed to analyse pro-teins that bind to chromatin, which are precipitated using antibodies. It was first designed to study the association of RNA-polymerase II with poised genes (Gilmour and Lis, 1984, 1985, 1986). Application of this protocol was later combined with several techniques including Southern blotting, Western blotting (Wells and Farnham, 2002), microarray (Kim and Ren, 2006; Ren, 2000) or high-throughput sequencing (Elnitski et al., 2006).

The ChIP-Seq protocol itself includes several chemical and mechanical steps that lead to the final step containing a population of DNA fragments of short length. Such DNA fragments were then processed to library preparation and sequencing. Sequencing after ChIP made it possible to explore the targets DNA binding proteins and genomic sites where specific nucleosomes exhibit various histone modifications. ChIP-seq today stand as one of the most pow-erful techniques to understand the regulation of gene expression.

The yield of the ChIP assay remains the main hurdle for the application. The strength of ChIP-seq is that the method affords a possibility to precisely locate of the DNA binding regions of proteins. For our studies in PAPER III and PAPER IV, we performed Chip-Seq in Hs578T cells using the SOLiD sequencing platform.

30

Ampliseq transcriptomic array Transcriptomic analysis has made it easy to understand the differential gene expression and also various gene-network interactions. The evolution of gene expression studies on large scale incorporated microarray and RT-qPCR tech-niques that measure expression of specific groups of genes (Wang et al., 2009). Today tools like RNA-sequencing have emerged with the advantage to profile transcript expression at the whole genome level quantitatively. RNA-Seq works on the principle of sequencing every transcript by single ended or double ended PCRs with high sensitivity (Love et al., 2014; Wang et al., 2009; Yaari et al., 2013).

Applying the same principles of RNA-sequencing after target RNA ampli-fication (RT-qPCR), Ampliseq transcriptomic array was developed with a tar-geted pool of oligonucleotides that measure expression of 20,000 human pro-tein-coding genes (Blomquist et al., 2013; Zhang et al., 2014). This assay has proved to be more reliable quantitatively and quantitatively (Li et al., 2015). In PAPER III and PAPER IV we used Ampliseq transcriptomic array for our studies.

CRISPR-Cas9 Clustered regularly interspaced short palindromic repeats (CRISPR)-CRISPR associated endonuclease CAS9, is a rapidly growing genome editing technol-ogy (Hsu et al., 2014). CRISPR was identified as a defensive mechanism against viral infection in bacteria (Bhaya et al., 2011; Horvath and Barrangou, 2010). Bacteria memorise the viral DNA and synthesize a cascade of repeats targeting it, which are accompanied by an associate endonuclease that makes a double-stranded break (Hsu et al., 2014). The CRISPR endonuclease is guided to the target DNA sequence by RNA molecule, known as guide (g) RNAs. Due to the increased complexity and required precision in previously used DNA editing techniques, CRISPR-CAS9 was applied to human cells (293T and K562) to achieve mutations using unique gRNA against specific regions in the genome (Cong et al., 2013; Mali et al., 2013).

31

Figure 8: CRISPR-Cas9 applications. The figure illustrates different applications of CRISPR-Cas9 genome editing such as insertion and deletion (Doetschman & Georgieva, 2017). Abbreviations: Proto-spacer adjacent motif (PAM), sub-genomic RNA (sgRNA), Double stranded break (DSB), Non-homologouos end joining (NHEJ), Homologous DNA recombination (HDR). Reprints were made with per-mission form the respective journal.

CRISPR-Cas9 has rapidly gained acceptance in the scientific community compared to other gene editing technologies (Barrangou and Doudna, 2016). Apart from deleting a DNA sequence to quench the expression of a target gene, CRISPR is now applied as CRISPRi (combined with siRNA technol-ogy) to silence the gene (Barrangou and Doudna, 2016), CRISPRa combined with activation of genes (like plasmid transfections) for activation of genes (Barrangou and Doudna, 2016) (Figure 8), CRISPR-x (combined with site-directed mutagenesis) to achieve point mutations on targeted sites (Barrangou and Doudna, 2016). Today, overcoming inherent difficulties a technological innovation in CRISPR research was made to edit RNA using CRISPR-Cas13 (Barrangou and Doudna, 2016). In my studies, we have used the CRISPR-Cas9-based knockout of ZEB1 and Snail1 in Hs578T cells in PAPER III and PAPER IV.

32

Present Investigation

Paper I Aim The aim of the study was to investigate the activity of PARP2 and PARG in Smad-mediated TGFβ signaling.

Background Previous investigations from our lab showed that attenuation of Smad-medi-ated transcription by PARP1 (Lönn et al., 2010). PARP1 and PARP2 comprise the DNA-dependent members of the PARP family (Amé et al., 2004; Schreiber et al., 2006; Swindall et al., 2013). PARG is a well-studied glyco-hydrolase protein from the PARP family (Le May et al., 2012), which has an opposite function of PARP1 and PARP2. This idea of dynamic turn-over by PARylation and dePARylation by PARP1/2 along with PARG inspired us to pursue the investigation of fine-tuning of Smad mediated signaling by PARP2 and PARG.

Methods To investigate the TGFβ signaling effects, a robust cell model, human HaCaT keratinocytes were chosen. To validate the interaction between Smads, PARPs and Poly (ADP-ribose) chains, the sensitive protein interaction detection tech-nique, the PLA was used (Fredriksson et al., 2002). For additional interaction studies, immuno-precipitation assays followed by western blotting were per-formed, after transient plasmid and siRNA transfections. To determine the ef-fect of PARP on Smad3/4 transcriptional regulation, a CAGA-luciferase re-porter assay was performed. The post-translational modification of poly-(ADP-ribosylation) was visualized on Smads and PARPs using an in vitro PARylation after GST-pulldown of various proteins.

Results Our work confirmed that Smad3 is ADP-ribosylated and this post-transla-tional modification could be visualized by PLA using antibodies against PAR

33

chains and Smad3. It is also observed that PARylation on Smad3 is very dy-namic and is induced rapidly after TGFβ stimulation. Using the PLA, we also showed that PARP1 and PARP2 form complexes with Smad3 and the com-plexes are promoted by TGFβ stimulation. Using PLA, we were able to also demonstrate the already established results, namely PARylation of PARP1 and PARP2 themselves. Furthermore CAGA-luciferase reporter assay showed that the transcriptional activity of Smad3 was enhanced with knockdown of either PARP1 or PARP2 or both PARP1 and PARP2. Knockdown of PARP1 and knockdown of PARG showed an opposite effect on mRNA levels of Smad3-regulated genes like PAI1 and FN1and on CAGA-luciferase assay, confirming the findings at the transcriptional level.

Conclusions Our study illustrates that different proteins of the PARP family PARP1/2 and PARG have opposite functions, i.e, PARylation and de-PARylation which de-cide the transcriptional activity of Smad3 in response to TGFβ stimulation (Dahl et al., 2014). In, human keratinocytes Smad3 transcription is negatively regulated by PARP1/2 and PARG acts a positive regulatory factor for Smad3 transcription.

Paper II Aim The aim of the study was to investigate the effect of PARP1 and PARG on the BMP signaling pathway.

Background Previous publications from our group show that PARP activity is important in deciding the transcriptional activity of Smad3 in response to TGFβ stimula-tion. Both BMP and TGFβ signaling share similarities in receptor activation, structure of R-Smads, Smad4 being a redundant Co-Smad and similar I-Smads. Such similarities made us investigate the activity of PARP1 on BMP signaling.

Methods Human keratinocytes HaCaT and mouse myoblasts C2C12 were used to in-vestigate the effect of BMP signaling. Expression of BMP-responsive genes were measured at the RNA level by RT-qPCR and at the protein level by west-ern-blotting. The effect of BMP on transcriptional activity was measured by

34

BRE-reporter luciferase assay. Immunoprecipitation followed by western-blotting was used to measure protein expression after transient transfections. Proximity ligation assay was performed to study the interaction between Smad and PARP1. In vitro PARylation studies were performed to measure the PARylation levels on target proteins after GST-pulldown assay.

Results Over-expressing of PARG in HaCaT cells led to an increased expression of BMP target genes like ID1 and Smad7 and conversely by knockdown of PARG by two independent siRNAs decreased expression of these genes. Our results also suggest that PARP1 opposes the activity of PARG at the Smad transcriptional level and the level of responsive gene expression, such as ID1, Smad6 and Smad7. Reporter activity assay by BRE-Luc assay in PARP1 knockout mouse embryonic fibroblasts suggest that there is reduced transcrip-tion by Smad1/5, suggesting an attenuating function of PARP1 on BMP sig-naling. To mimic the activity of PARylation in cells, in vitro radioactive PARylation assay was performed with recombinant PARP1, PARG and GST-Smad1/5 with a radioactive NAD+ donor, which proves the hypothesis of PARP1 auto-PARylation, ADP-ribosylation of Smad1 and Smad5 by PARP1 and de-PARylation of PARP1, Smad1 and Smad5 by PARG, suggesting op-posite functions of PARG and PARP1. By domain mapping, the Smad1 N-terminal MH1 domain was identified as an interacting partner of PARP1 and mutations in the conserved MH1 PARylation site of Smad1 suggests differen-tial binding to PARP1 when incubated with recombinant and endogenous cell lysates. Inhibition of PARP activity by inhibitors like PJ34 and 3AB also showed a similar effect as PARP1 depletion.

Conclusions Our study on effects of PARP family members on BMP signaling helped un-derstand that PARP1 and PARG act simultaneously in opposite direction de-ciding the fate of target gene-expression. We were also able to map the site of PARylation on Smad1 which interacts with PARP1, modulating the transcrip-tional activity of Smad1 on BMP responsive genes.

Paper III Aim Identify novel sites of gene regulation by the ZEB1 transcription factor in tri-ple negative breast cancer cells.

35

Background ZEB1 is an EMT transcriptional factor responsible in triggering and maintain-ing mesenchymal phenotype in cancerous cells. ZEB1 is well studied as a transcriptional repressor repressing the epithelial genes like CDH1. It is also studied in indirect regulation of EMT by mir-200 genes. Despite the extensive availability of the literature on ZEB1 regulation in different cancer cells, there is no study that describes the, genes regulated by ZEB1 in triple negative breast cancer, an aggressive type of breast cancer with loss of estrogen recep-tor (ER), progesterone receptor (PR) and human epidermal growth factor Re-ceptor (HER) expression and a mesenchymal differentiation. This study has now made an attempt to focus novel binding sites of ZEB1 in the triple nega-tive breast cancer cells.

Methods & Results In silico analysis of breast cancer cells shows a varied expression of ZEB1. Basal-B breast cancer cells show high levels of EMT TF’s including ZEB1 compared to Basal-A and Luminal subtypes, which is consistent with the the-ory of EMT transcription factors being highly expressed and helping to main-tain mesenchymal features. Further analysis of expression of EMT TFs at pro-tein level in cell lines from different subgroups show that Hs578T, a Basal-B cell line expresses high levels of EMT TFs. TGFβ is a driving factor for EMT and stimulation of Hs578T cells with TGFβ further enhanced the expression of mesenchymal markers like N-Cadherin and Fibronectin, while a TGFβ in-hibitor blocked the enhanced expression. By this we confirmed the Hs578T cells as a robust model for our study. Chromatin Immunoprecipitation fol-lowed by sequencing was performed for ZEB1 in Hs578T, cells and bio-in-formatics analysis resulted in 3,900 peaks or 1,939 genes of ZEB1 peaks at a distance of -10kbp to +2.5kbp from the TSS were analysed and known and novel sequence motifs were discovered. Gene ontology data shows that ZEB1 not only has a functional role in EMT but also in heart development, nervous development, signal transduction, cell cycle and cell communication.

Since ZEB1 knockout mice are lethal, in order to exclude the insufficient

transient knockdown, we established completely knockout cells of ZEB1 us-ing the CRISPR-CAS9 gene editing technique. ZEB1 knockout Hs578T shows differences in cell morphology, reduced cell proliferation, reduced cell migration and reduced anchorage-independent growth when compared to wild type Hs578T cells. This further motivated our study to focus on genes regu-lated directly by ZEB1. A gene-specific amplification array of the cDNA of WT and ZEB1 knockout cells showed about 1,142 genes were upregulated and 230 genes were downregulated by ZEB1 knockout, suggesting the dual transcriptional role of ZEB1. Comparison of ChIP-seq and Ampliseq results

36

showed that ZEB1 acts as a transcriptional enhancer in the case of 21 genes and as a repressor for 133 genes. This list of ZEB1 target genes includes genes responsible for ECM remodeling (TIMP3) and adhesion-related molecules (TENM2) which are worth examining further.

Conclusion Our study verifies previous findings on ZEB1 function being both enhancer and a repressor of transcription. Knockout of the single EMT-TF ZEB1, is not sufficient to revert the mesenchymal phenotype to an epithelial phenotype. This investigation reveals new functions of ZEB1’s contribution towards an-chorage-independence. Further, this work provides a useful resource for future research on ZEB1 and EMT.

Paper IV Aim The aim of the study is to investigate the function of the Snail1 transcription in mesenchymal breast cancer cells.

Background Following the activation of TGFβRs the signaling is carried out by activate Smads which transcriptionally regulate Snail1. This makes Snail1 one of the early responsive genes of TGFβ signaling which further regulates other genes related to TGFβ biology. Snail1 is a transcription factor whose function is regulated by post-translational modifications like PARylation by PARP1, phosphorylation by GSK3β and sumoylation. Snail1 depletion in some carci-nomas can trigger MET, but the knowledge on gene targets of Snail1 genome-wide is very limited. Based on the available literature, we postulate that Snai1 either regulates a wide variety of genes or it could regulate a small group of master regulators which in turn carry various functions.

Methods & Results Our parallel studies on expression of EMT-TF’s and in silico data mining sug-gests that Snail1 is highly expressed in triple negative breast cancer cells. From our studies we establish that Basal-B, Hs578T and MAD-MB-231 cell types are robust models for Snail1 studies. Availability of good antibodies and expression level of Snail1 led us to use Hs578T as our cell-system for our analysis. ChIP-Seq for Snail1 in Hs578T resulted 405 peaks in 185 genes. Motifs of Snail1 binding regions suggest a possible binding of GATA1,

37

RREB1, RUNX2 and PAX4 along with Snail1 to the same region. Target genes of Snail1 provides information about their functional relevance to heart development, other developmental processes, guanyl-nucleotide exchange factors and various enzyme regulatory activates. Snail1 can bind to genes such as CRB1, PPFIA1, PTK2, Sox6 and FAT1 which were validated using ChIP-qPCR analysis.

The available literature on the Snail1 knockdown leading to MET in some cells models, further motivated us to ask questions behind the genes that lead to MET genome-wide and that are regulated by Snail1. To avoid the incon-sistencies due to transient knockdown, we established a stable knockout of Snail1 in Hs578T cells using CRISPR-Cas9 gene editing technique targeting the 1st exon and 2nd exon-intron junction of the Snail1 gene. This resulted in several clones with a deletions of 10 bp length. For further analysis, we used two clones, one with double deletion and one with single deletion of the target regions. By deleting the Snail1 transcription factor we observed that cells tend to show a small but observable trend towards the epithelial phenotype. Snail KO cells also had shown significant reduction in their migration capa-bilities compared to WT Hs578T cells in a wound healing T-Scratch assay. To identify genes regulated by Snail1, which contribute to this phenotype, we analysed changes in gene transcription by Ampliseq in WT and Snail1 KO Hs578T cells.

Our results showed that effect of Snail1 KO can effect both transcriptional activation and repression of genes, postulating the dual role of Snail1 in tran-scription. Rigorous analysis shows that 363 genes were upregulated in Snail1 KO and 276 were downregulated when compared to WT cells. Correlation studies showed that downregulated genes show high significance of genes re-lated to chromatin organization, cellular process and protein biding, whereas upregulated genes show high co-relation to developmental processes. Further intersection of ChIP-seq and Ampliseq results confirmed 5 genes, were di-rectly regulated by Snail1 in both high-throughput studies. Genes that were conversely regulated by Snail1 in triple negative breast cancer were BMP6 and CPED1. These genes were validated by ChIP-qPCR and RT-PCR analy-sis.

Conclusion Our study brings out interesting facts about Snail1 function in breast cancer cells. We also show that KO of Snail1 in triple negative breast cancer cells cannot completely reverse the phenotype of mesenchymal to epithelial as re-ported in literature by transient knockdown of Snail1 in different cell models. Our Ampliseq analysis shows that Snail1 can act as a positive regulator on BMP6, whereas it acts as a negative regulator on CPED1 suggesting the dual role of Snail1 in transcriptional regulation.

38

Future Perspectives

Both gene regulation and cancer are very complex processes where several factors play important roles to affect the phenotypic outcome. Considering the scope for depth in this study, a strict path of investigation was followed to answer questions such as post-translational modifications and transcriptional regulation related to TGFβ signaling, and which were studied in the chosen context of cancer.

The PARP family consists of 18 members with different functions. Previ-ous studies have shown that PARP inhibitors play an important role in cancers with BRCA1 and BRCA2 mutation, where inhibition of PARP blocks DNA repair pushing cancer cells to apoptosis (Frizzell et al., 2009). Based on our investigation, PARG blocks the activity of PARP1 and PARP2 by removing ADP-ribose chains, PARG inhibitors can also be used for therapeutic purposes (Dahl et al., 2014; Gravells, Grant, Smith, James, & Bryant, 2017; Lönn et al., 2010; Watanabe et al., 2016). A complementary study of different PARP and PARG inhibitors could help us identify various targets for cancer therapy. At the same time different PARP family members with similar structural and functional similarities must be investigated for their involvement in TGFβ and BMP signaling.

Different post-translational modifications affect the functionality of pro-teins that lead to activation, inactivation or degradation of the protein. They play an important role in orchestrating the signaling pathways. Our investiga-tion suggests early activation of PARP upon TGFβ stimulation. There is a possibility of cooperation between different post-translational modifications, which can be further investigated for better understanding of the functional importance of PARP.

Switching gears to gene regulation by PARP1, genome-wide association of PARP1 in MDA-MB-231 and MCF 7 cells were investigated revealing PARP1 binding to hypo-methylated regions, which also correlate with the binding pattern of CTCF. The role of CTCF as a general regulator of chroma-tin organization is well-studied. This study opens percepts for further investi-gation on the role of CTCF in PARP1-DNMT1 competition in binding to the genome and regulating its methylation pattern, as hyper-methylation of the genome is reported in various cancers (Nalabothula et al., 2015).

We studied ZEB1 and Snail, two important transcription factors driving the EMT. Despite many investigations previously performed, very little is known on genome-wide binding regions of Snail and ZEB1 in breast cancer cells. An

39

attempt to answer this question at the complete genomic scale for Snail1 and ZEB1 was made in this study. A subset of genes regulated by Snail and ZEB1 was investigated using Ampliseq transcriptomic array of a CRISPR-Cas9 -based knockout model of triple negative breast cancer cells. Our study reveals no-redundancy in genes regulated by ZEB1 and Snail1 in Hs578T cells. A similar study for different transcription factors related to EMT could show redundancy, which could be an interesting subset of genes to study further in understanding better the EMT. We also demonstrate that single knockout out these transcription factors could not revert the mesenchymal cells to epithelial cells. An interesting study would be to perform either combinatorial knock-outs in cancer cells to look for a reversal to epithelial phenotype. Adding an in-depth layer to it, studying genes regulated by such transcription factors re-sponsible for the epithelial reversal could be a key finding for EMT studies.

EMT is an important phenomenon in both gastrulation and cancer metasta-sis. In our investigation, we observed that mesenchymal cells were character-ized with features like high proliferation and high invasion. Investigation of the target genes reported by us to study mesenchymal features and EMT will be an interesting step to know about novel effectors of EMT. Investigation of these targets, when performed in different cell lines and different cancer types will further our understanding of the common principles that govern the EMT. By performing individual tumor sequencing, large datasets of personalized cancer conditions can be investigated, by which treatments to cancer patients can be designed in a more personalized manner. By using transcriptomic array methodologies, investigating novel molecules has become possible in cancer research, with which one aims at making a substantial contribution to cancer research.

Opening possibilities for new investigations a proposal can be made as PARP can ADP-ribosylate Smads and Snail1. Bioinformatic modelling sug-gests that PARP1 and ZEB1 could exhibit protein-protein interactions (Szklarczyk et al., 2017). If PARP regulates Snail1 and ZEB1 function in the same way as Smads, it will be an interesting study to check the transcriptional activity of these transcription factors for the target genes proposed here in the presence of PARP inhibitors. Since Snail1 and ZEB1 attain clinical im-portance by showing high activity in malignant neoplasias, PARP family members can be used to modulate their activity in cancers. Such studies can be performed for other cancer-related transcription factors.

40

Acknowledgements

This work was funded by grants obtained by Prof. Aristidis Moustakas from Ludwig Cancer Research, Cancerfonden and Swedish Research Council. I feel extremely happy to thank everyone who is part of this long journey and helped me to end this positively at a high note. It has been a journey with constant learning and I thoroughly enjoyed the roller coaster ride. I apologize if I have missed mentioning someone.

My supervisor Prof. Aristidis Moustakas, thanks for choosing and trusting me for this task. Your tremendous guidance and support over the past years, scientifically and personally was impeccable. Thanks for giving me the free-dom to do experiments with my own ideas, this gave me first-hand experience on pursuing own ideas. Your scientific guidance and mentoring inspired me create my own identity in scientific research. Thanks for being considerate and tolerant with me. I extend my sincere gratitude to my co-supervisor and director of Ludwig Can-cer Research, Uppsala, Prof. Carl-Henrik Heldin. Your passion in science, meticulous mentoring, humble and kind nature, balance between work and personal life is a great inspiration. Thanks for giving time for all discussions whenever I needed. Dr. Stefan Enroth, thank you very much for being my co-supervisor and helping me dive through the adventure of bio-informatics. You thought me understand Greek and Latin (program scripts). You were always friendly and pleasant to talk, I am very happy to be associated with you. I would say it is where, it all started, Prof. Bengt Westermark and Dr. Umashankar Singh, thank you both for making me believe that I am a worthy candidate to pursue PhD studies. I thank all co-authors who took part in contributing to my publications from this lab. It was always fun to come to work with these people at work. Panos you deserve a special place in this acknowledgements for all the fun we had during PARP projects and all discussions we had on every topic on earth, it was always nice to meet you and Christina , please do invite me for the Greek-wedding. Anita thanks for being an encyclopedia of this lab and for

41

being so helpful with all practical problems I faced in lab. Laia, in one word, you are a super-mom-scientist, good to know you for these many years. Clau-dia, you are one of the most hard working woman I have ever seen , I see you as young, female, Aris (although you are way too, louder than him ). Costas and Kallia its lot of fun to be associated with you guys, I know at least few words in Greek language (Καλημέρα) and little bit about Greek traditions now. Mahsa (the dancing doll), thank god, I met you, at least one non-Euro-pean PhD student, who shared fun in struggling with bureaucracy here. Yutaro, thanks for making the work place calm, exciting and colorful with GFP and td tomato, good to know you. Thanks to all the group leaders at Ludwig Johan, Marene, Ingvar and Evi for your kind gesture during my stay here. I am also fortunate to know many seniors who just moved on when I am joining, I learnt a lot from you people Peter Lönn, E Jean Tan, Glenda, Vahid, Berit, Helena and Agata. Kaoru, thanks for teaching me the routines of the lab and giving a kick start in my PhD. My fellow PhD students Merima, Guilia, Ana Rosa and Chun thanks for great discussions in the corridors and parties. The group who wel-comed me to LICR with a loudest smile at lunch table: Markus, Erna, Yukihide, Takeshi, Jon, Gad, Linda and noisy Noopur as everyone says. Senior Post-Docs of the department Anders, Olex and Ria (most troubled person with my stupid questions), thanks for clarifying all the doubts that I used to ask everywhere in the corridor, you are quicker than internet for me to ask my doubts. Natalia, it was so nice to know you, lots of fun discussing with you, I will definitely miss you in my new office. It’s important to have good office colleagues, to share all the happiness and frustrations, I had great time with people, Ihor, Haisha, Andries, Masato, Sara, Melanie, Yanyu, Maria (you have the best baking skills ), you all made it easy for me. Masood (reminds me of discussions on tours & cars) and Anahita thanks for all the laughter we had and tips you gave with parental topics . I would specially thank the pillars of Ludwig Lotti, Aino, Anita, Mariya yakymovych, Ulla, Eva, Lasse, Uffe and Ingegard. Each one of you was very kind at me and you took care of most of the practical things in a profes-sional way. Huge turnover with guest students and project students is defi-antly an additional flavor of this department, it’s nice to meet all these young people Micheal, Natasa, Fredrick (Car nerd), George, Panagiotis, Oscar, Simmon, Anna, Nikki, Katia, Savvas & Per (cool combination ), Paris, Ashish, Sandra, Angelios, Rafailia, Andrew, Natsnet, Larina and Yukari. Sweden has become second home for me. It is pleasant to be here because of these people who made me feel this: Umash, my teacher (be it Immuno-Pre-cipitation or driving a car), friend and family, you are the person whom I ad-mire the most after my father; Noopur my mother in Sweden, friend and the

42

person who showed me that science can be done along with fun in life (not to mention the nail polishing during Western-blotting) thanks for all the argu-ments we had, thanks for caring me and Prathyusha so much; you are part of my family no matter where I am. I feel fortunate to meet you both and of course the guitar rockstar Nangu-pongu Daksh too. Thanks for all the support, love, care, fun and inspiration from my extended family Gayatri and Arvind (I will defiantly listen to your story one day), one amazing couple who always makes me look forward to spend time with, I am still looking forward for Norway road trip and skydiving; Sravani, my sister who supported me in my highs and much more in my lows; Rohit I will never forget, death by chocolate and lymphoma incidents; Roshan, best political analyst, one can never be as kind and friendly as you are; Gopi and Sandhya (rakhi sister who gave the most hilarious ride in Uppsala ever) meeting you is always a festive for me. Every day in Uppsala is a colorful and eventful day for me, every day I look forward to meet you guys, I call this as huge Janata Garage headed by Chandu anna followed by Geeta garu, Nandu & Ashritha; Sudarsan & Christin ; Sethu, Anusha & Keshvin; Srinu, Sony & Lohith; Kiran anna, Bhavya & Suryansh; Hari & Saranya; Praksh, Ammulu & Jatin; Srisailam anna, Swati & Vidathri; Venkat and Ram , I cherish every mo-ment I spent with you people. Fortunately I have many friends in Sweden who make my life smiling and pleasant strong Javeed, Huda & Maryam (Jai Ho couple , thanks for a memorable Venice trip); Naresh & Soundarya; Suri & Spandana; Sriram; Arun (I learnt a lot about publications form you) & Roopa; Phani & Sireesha; Manoj & Kanaka; Rahul, Shruthi & Aadhya; Sharath & Mounika; Mahesh & Saritha; Shimbu; Srinivas & Rekha; Santosh; Maruthi & Chandana; Archana, Chandra & Annika; Narendra Padhan, Sujatha & two little rockstars; Snehangshu & Soumi; Rajesh, Priyank & Munnulal; Satish & Priya; Avinash; Malli; Nikhil; Vivek; Mo-han; Pruthvi; Kalyani; Prasad; Raviteja Inturi and Deepesh. I thank my Mom and Dad for all the support you gave me to try new things and for supporting me to come to Sweden and pursue by interest in doing research. My brother Pavan and Thulasi vadina has been a continuous sup-port and inspiration. I received tremendous support from my family in India without which, this would have been not possible. Thanks to everyone from the families of Maturi Subbarao garu; Ponnaganti Satyanarayana garu and Pendekanti Ramarangaiah garu. Whatever I achieved, whatever I do and whoever I met, it is just because of you, your love and support, thanks for everything my dear wife Prathyusha

and my darling daughter Taanvika.

43

References

Aigner, K., Dampier, B., Descovich, L., Mikula, M., Sultan, A., Schreiber, M., Mikulits, W., Brabletz, T., Strand, D., Obrist, P., et al. (2007). The transcription factor ZEB1 (δEF1) promotes tumour cell dedifferentiation by repressing master regulators of epithelial polarity. Oncogene 26, 6979–6988.

Alvarez-Gonzalez, R., and Althaus, F. (1989). Poly (ADP-ribose) catabolism in mammalian cells exposed to DNA-damaging agents. Mutat. Res. Repair 218, 67–74.

Amé, J.C., Spenlehauer, C., and De Murcia, G. (2004). The PARP superfamily. BioEssays 26, 882–893.

Andersson, R., Enroth, S., Rada-Iglesias, A., Wadelius, C., and Komorowski, J. (2009). Nucleosomes are well positioned in exons and carry characteristic histone modifications. Genome Res. 19, 1732–1741.

Augustin, A. (2003). PARP-3 localizes preferentially to the daughter centriole and interferes with the G1/S cell cycle progression. J. Cell Sci. 116, 1551–1562.

Barrangou, R., and Doudna, J.A. (2016). Applications of CRISPR technologies in research and beyond. Nat. Biotechnol. 34, 933–941.

Baulida, J., and García de Herreros, A. (2015). Snail1-driven plasticity of epithelial and mesenchymal cells sustains cancer malignancy. Biochim. Biophys. Acta - Rev. Cancer 1856, 55–61.

Bengtsson, L., Schwappacher, R., Roth, M., Boergermann, J.H., Hassel, S., and Knaus, P. (2009). PP2A regulates BMP signalling by interacting with BMP receptor complexes and by dephosphorylating both the C-terminus and the linker region of Smad1. J. Cell Sci. 122, 1248–1257.

Bhaya, D., Davison, M., and Barrangou, R. (2011). CRISPR-Cas Systems in Bacteria and Archaea: Versatile Small RNAs for Adaptive Defense and Regulation. Annu. Rev. Genet. 45, 273–297.

Bhowmick, N. a, Ghiassi, M., Bakin, a, Aakre, M., Lundquist, C. a, Engel, M.E., Arteaga, C.L., and Moses, H.L. (2001). Transforming growth factor-beta1 mediates epithelial to mesenchymal transdifferentiation through a RhoA-dependent mechanism. Mol. Biol. Cell 12, 27–36.

Blomquist, T.M., Crawford, E.L., Lovett, J.L., Yeo, J., Stanoszek, L.M., Levin, A., Li, J., Lu, M., Shi, L., Muldrew, K., et al. (2013). Targeted RNA-Sequencing with Competitive Multiplex- PCR Amplicon Libraries. PLoS One 8.

Chaitanya, G., Alexander, J.S., and Babu, P. (2010). PARP-1 cleavage fragments: signatures of cell-death proteases in neurodegeneration. Cell Commun. Signal. 8, 31.

Chen, M.L. (2005). Regulatory T cells suppress tumor-specific CD8 T cell cytotoxicity through TGF-β signals in vivo. Proc. Natl Acad. Sci. USA 102, 419–424.

44

Chu, Y.S., Eder, O., Thomas, W.A., Simcha, I., Pincet, F., Ben-Ze’ev, A., Perez, E., Thiery, J.P., and Dufour, S. (2006). Prototypical type I E-cadherin and type II cadherin-7 mediate very distinct adhesiveness through their extracellular domains. J. Biol. Chem. 281, 2901–2910.

Cong, L., Ran, F.A., Cox, D., Lin, S., Barretto, R., Habib, N., Hsu, P.D., Wu, X., Jiang, W., Marraffini, L.A., et al. (2013). Multiplex Genome Engineering Using CRISPR/Cas Systems. Science. 339, 819 LP-823.

Dahl, M., Maturi, V., Lönn, P., Papoutsoglou, P., Zieba, A., Vanlandewijck, M., van der Heide, L.P., Watanabe, Y., Söderberg, O., Hottiger, M.O., et al. (2014). Fine-Tuning of Smad Protein Function by Poly(ADP-Ribose) Polymerases and Poly(ADP-Ribose) Glycohydrolase during Transforming Growth Factor β Signaling. PLoS One 9, e103651.

Dantzer, F., Schreiber, V., Niedergang, C., Trucco, C., Flatter, E., De La Rubia, G., Oliver, J., Rolli, V., Ménissier-de Murcia, J., and de Murcia, G. (1999). Involvement of poly(ADP-ribose) polymerase in base excision repair. Biochimie 81, 69–75.

Dave, N., Guaita-Esteruelas, S., Gutarra, S., Frias, À., Beltran, M., Peiró, S., and García De Herreros, A. (2011). Functional cooperation between snail1 and twist in the regulation of ZEB1 expression during epithelial to mesenchymal transition. J. Biol. Chem. 286, 12024–12032.

Davidovic, L., Vodenicharov, M., Affar, E.B., and Poirier, G.G. (2001). Importance of poly(ADP-ribose) glycohydrolase in the control of poly(ADP-ribose) metabolism. Exp. Cell Res. 268, 7–13.

Delhalle, S., Rie Deregowski, V., Rie Benoit, V., Merville, M.-P., and Bours, V. (2002). NF-kB-dependent MnSOD expression protects adenocarcinoma cells from TNF-a-induced apoptosis. Oncogene 21, 3917–3924.

Derynck, R., and Zhang, Y.E. (2003). Smad-dependent and Smad-independent pathways in TGF-beta family signalling. Nature 425, 577–584.

ten Dijke, P., and Arthur, H.M. (2007). Extracellular control of TGFβ signalling in vascular development and disease. Nat. Rev. Mol. Cell Biol. 8, 857–869.

ten Dijke, P., and Hill, C.S. (2004). New insights into TGF-β-Smad signalling. Trends Biochem. Sci. 29, 265–273.

Doetschman, T., and Georgieva, T. (2017). Gene Editing With CRISPR/Cas9 RNA-Directed Nuclease. Circ. Res. 120, 876–894.

Drake, J.M., Barnes, J.M., Madsen, J.M., Domann, F.E., Stipp, C.S., and Henry, M.D. (2010). ZEB1 coordinately regulates laminin-332 and β4 integrin expression altering the invasive phenotype of prostate cancer cells. J. Biol. Chem. 285, 33940–33948.

Ebisawa, T., Fukuchi, M., Murakami, G., Chiba, T., Tanaka, K., Imamura, T., and Miyazono, K. (2001). Smurf1 Interacts with Transforming Growth Factor-β Type I Receptor through Smad7 and Induces Receptor Degradation. J. Biol. Chem. 276, 12477–12480.

Elnitski, L., Jin, V.X., Farnham, P.J., and Jones, S.J.M. (2006). Locating mammalian transcription factor binding sites : A survey of computational and experimental techniques. 1455–1464.

Ernst, J., Kheradpour, P., Mikkelsen, T.S., Shoresh, N., Ward, L.D., Epstein, C.B., Zhang, X., Wang, L., Issner, R., Coyne, M., et al. (2011). Mapping and analysis of chromatin state dynamics in nine human cell types. Nature 473, 43–49.

Escrivà, M., Peiró, S., Herranz, N., Villagrasa, P., Dave, N., Montserrat-Sentís, B., Murray, S.A., Francí, C., Gridley, T., Virtanen, I., et al. (2008). Repression of PTEN Phosphatase by Snail1 Transcriptional Factor during Gamma Radiation-Induced Apoptosis. Mol. Cell. Biol. 28, 1528–1540.

45

Fredriksson, S., Gullberg, M., Jarvius, J., Olsson, C., Pietras, K., Gústafsdóttir, S.M., Östman, A., and Landegren, U. (2002). Protein detection using proximity-dependent DNA ligation assays. Nat. Biotechnol. 20, 473–477.

Frizzell, K.M., Kraus, W.L., Inbar-Rozensal, D., Castiel, A., Visochek, L., Castel, D., Dantzer, F., Izraeli, S., Cohen-Armon, M., Kraus, W., et al. (2009). PARP inhibitors and the treatment of breast cancer: beyond BRCA1/2? Breast Cancer Res. 11, 111.

Gilmour, D.S., and Lis, J.T. (1984). Detecting protein-DNA interactions in vivo: Distribution of RNA polymerase on specific bacterial genes. Biochemistry 81, 4275–4279.

Gilmour, D.S., and Lis, J.T. (1985). In vivo interactions of RNA polymerase II with genes of Drosophila melanogaster. Mol. Cell. Biol. 5, 2009–2018.

Gilmour, D.S., and Lis, J.T. (1986). RNA polymerase II interacts with the promoter region of the noninduced hsp70 gene in Drosophila melanogaster cells. Mol. Cell. Biol. 6, 3984–3989.

Gomis, R.R., Alarcón, C., Nadal, C., Van Poznak, C., and Massagué, J. (2006). C/EBPβ at the core of the TGFβ cytostatic response and its evasion in metastatic breast cancer cells. Cancer Cell 10, 203–214.

Goto, K., Kamiya, Y., Imamura, T., Miyazono, K., and Miyazawa, K. (2007). Selective inhibitory effects of Smad6 on bone morphogenetic protein type I receptors. J. Biol. Chem. 282, 20603–20611.

Gravells, P., Grant, E., Smith, K.M., James, D.I., and Bryant, H.E. (2017). Specific killing of DNA damage-response deficient cells with inhibitors of poly(ADP-ribose) glycohydrolase. DNA Repair (Amst). 52, 81–91.

Greer, E.L., and Shi, Y. (2012a). Histone methylation: a dynamic mark in health, disease and inheritance. Nat. Rev. Genet. 13, 343–357.

Guaita, S., Puig, I., Francí, C., Garrido, M., Domínguez, D., Batlle, E., Sancho, E., Dedhar, S., De Herreros, A.G., and Baulida, J. (2002). Snail induction of epithelial to mesenchymal transition in tumor cells is accompanied by MUC1 repression and ZEB1 expression. J. Biol. Chem. 277, 39209–39216.

Gustafsdottir, S.M., Schallmeiner, E., Fredriksson, S., Gullberg, M., Söderberg, O., Jarvius, M., Jarvius, J., Howell, M., and Landegren, U. (2005). Proximity ligation assays for sensitive and specific protein analyses. Anal. Biochem. 345, 2–9.

Hamidi, A., Song, J., Thakur, N., Itoh, S., Marcusson, A., Bergh, A., Heldin, C.-H., and Landström, M. (2017). TGF-β promotes PI3K-AKT signaling and prostate cancer cell migration through the TRAF6-mediated ubiquitylation of p85α. Sci. Signal. 10, eaal4186.

Hanyu, A., Ishidou, Y., Ebisawa, T., Shimanuki, T., Imamura, T., and Miyazono, K. (2001). The N domain of Smad7 is essential for specific inhibition of transforming growth factor-β signaling. J. Cell Biol. 155, 1017–1027.

Hayashi, H., Abdollah, S., Qiu, Y., Cai, J., Xu, Y.Y., Grinnell, B.W., Richardson, M. a, Topper, J.N., Gimbrone, M. a, Wrana, J.L., et al. (1997). The MAD-related protein Smad7 associates with the TGFbeta receptor and functions as an antagonist of TGFbeta signaling. Cell 89, 1165–1173.

Heldin, C.H., Vanlandewijck, M., and Moustakas, A. (2012). Regulation of EMT by TGFβ in cancer. FEBS Lett. 586, 1959–1970.

De Herreros, A.G., Peiró, S., Nassour, M., and Savagner, P. (2010). Snail family regulation and epithelial mesenchymal transitions in breast cancer progression. J. Mammary Gland Biol. Neoplasia 15, 135–147.

Horvath, P., and Barrangou, R. (2010). CRISPR/Cas, the Immune System of Bacteria and Archaea. Science (80-. ). 327, 167–170.

46

Hsu, P.D., Lander, E.S., and Zhang, F. (2014). Development and applications of CRISPR-Cas9 for genome engineering. Cell 157, 1262–1278.

Huang, D., Wang, Y., Wang, L., Zhang, F., Deng, S., Wang, R., Zhang, Y., and Huang, K. (2011). Poly(ADP-ribose) polymerase 1 is indispensable for transforming growth factor-β induced smad3 activation in vascular smooth muscle cell. PLoS One 6.

Jang, C.-W., Chen, C.-H., Chen, C.-C., Chen, J., Su, Y.-H., and Chen, R.-H. (2002). TGF-β induces apoptosis through Smad-mediated expression of DAP-kinase. Nat. Cell Biol. 4, 51–58.

Kalluri, R., and Weinberg, R.A. (2009a). The basics of epithelial-mesenchymal transition. J. Clin. Invest. 119, 1420–1428.

Kato, M., Putta, S., Wang, M., Yuan, H., Lanting, L., Nair, I., Gunn, A., Nakagawa, Y., Shimano, H., Todorov, I., et al. (2009). TGF-β activates Akt kinase through a microRNA-dependent amplifying circuit targeting PTEN. Nat. Cell Biol. 11, 881–889.

Kaufmann, S.H., Desnoyers, S., Ottaviano, Y., Davidson, N.E., and Poirier, G.G. (1993). Specific Proteolytic Cleavage of Poly(ADP-ribose) Polymerase: An Early Marker of Chemotherapy-induced Apoptosis. Cancer Res. 53, 3976–3985.

Kavsak, P., Rasmussen, R.K., Causing, C.G., Bonni, S., Zhu, H., Thomsen, G.H., and Wrana, J.L. (2000). Smad7 binds to Smurf2 to form an E3 ubiquitin ligase that targets the TGFβ receptor for degradation. Mol. Cell 6, 1365–1375.

Kim, T.H., and Ren, B. (2006). Genome-Wide Analysis of Protein-DNA Interactions. Annu. Rev. Genomics Hum. Genet 7, 81–102.

Komuro, A., Imamura, T., Saitoh, M., Yoshida, Y., Yamori, T., Miyazono, K., and Miyazawa, K. (2004). Negative regulation of transforming growth factor-beta (TGF-beta) signaling by WW domain-containing protein 1 (WWP1). Oncogene 23, 6914–6923.

Kraus, W.L., and Hottiger, M.O. (2013). PARP-1 and gene regulation: Progress and puzzles. Mol. Aspects Med. 34, 1109–1123.

Kraus, W.L., and Lis, J.T. (2003). PARP Goes Transcription. Cell 113, 677–683. Krishnakumar, R., and Kraus, W.L. (2010). PARP-1 Regulates Chromatin Structure

and Transcription through a KDM5B-Dependent Pathway. Mol. Cell 39, 736–749.

Kuratomi, G., Komuro, A., Goto, K., Shinozaki, M., Miyazawa, K., Miyazono, K., and Imamura, T. (2005). NEDD4-2 (neural precursor cell expressed, developmentally down-regulated 4-2) negatively regulates TGF-beta (transforming growth factor-beta) signalling by inducing ubiquitin-mediated degradation of Smad2 and TGF-beta type I receptor. Biochem. J. 386, 461–470.

Lawler, S., Feng, X.H., Chen, R.H., Maruoka, E.M., Turck, C.W., Griswold-Prenner, I., and Derynck, R. (1997). The type II transforming growth factor-beta receptor autophosphorylates not only on serine and threonine but also on tyrosine residues. J. Biol. Chem. 272, 14850–14859.

Li, W., Turner, A., Aggarwal, P., Matter, A., Storvick, E., Arnett, D.K., and Broeckel, U. (2015). Comprehensive evaluation of AmpliSeq transcriptome, a novel targeted whole transcriptome RNA sequencing methodology for global gene expression analysis. BMC Genomics 16, 1069.

Liu, Y., Snow, B.E., Kickhoefer, V.A., Erdmann, N., Zhou, W., Wakeham, A., Gomez, M., Rome, L.H., and Harrington, L. (2004). Vault poly(ADP-ribose) polymerase is associated with mammalian telomerase and is dispensable for telomerase function and vault structure in vivo. Mol. Cell. Biol. 24, 5314–5323.

Love, M.I., Huber, W., and Anders, S. (2014). Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15.

47

Lönn, P., van der Heide, L.P., Dahl, M., Hellman, U., Heldin, C.H., and Moustakas, A. (2010). PARP-1 attenuates smad-mediated transcription. Mol. Cell 40, 521–532.

Mali, P., Yang, L., Esvelt, K.M., Aach, J., Guell, M., DiCarlo, J.E., Norville, J.E., and Church, G.M. (2013). RNA-Guided Human Genome Engineering via Cas9. Science. 339, 823–826.

Mar Inez-estrada, O.M., Culle, A.E., Soriano, F.X., Peinado, H., Os, V.B., Mar Inez, F.O., Reina, M., Cano, A., Fabre, M., and Vila, S.O. (2006). The transcription factors Slug and Snail act as repressors of Claudin-1 expression in epithelial cells 1. Biochem. J 394, 449–457.

Massagué, J., Seoane, J., and Wotton, D. (2005). Smad transcription factors Smad transcription factors. Genes Dev. 19, 2783–2810.

Le May, N., Iltis, I., Amé, J.C., Zhovmer, A., Biard, D., Egly, J.M., Schreiber, V., and Coin, F. (2012). Poly (ADP-Ribose) Glycohydrolase Regulates Retinoic Acid Receptor-Mediated Gene Expression. Mol. Cell 48, 785–798.

Micalizzi, D.S., Farabaugh, S.M., and Ford, H.L. (2010). Epithelial-mesenchymal transition in cancer: Parallels between normal development and tumor progression. J. Mammary Gland Biol. Neoplasia 15, 117–134.

Miyazono, K., Kamiya, Y., and Morikawa, M. (2010). Bone morphogenetic protein receptors and signal transduction. J. Biochem. 147, 35–51.

Mock, K., Preca, B.-T., Brummer, T., Brabletz, S., Stemmler, M.P., and Brabletz, T. (2015). The EMT-activator ZEB1 induces bone metastasis associated genes including BMP-inhibitors. Oncotarget 6.

Moustakas, A., and Heldin, C.-H. (2009). The regulation of TGFbeta signal transduction. Development 136, 3699–3714.

Nalabothula, N., Al-Jumaily, T., Eteleeb, A.M., Flight, R.M., Xiaorong, S., Moseley, H., Rouchka, E.C., and Fondufe-Mittendorf, Y.N. (2015). Genome-wide profiling of PARP1 reveals an interplay with gene regulatory regions and DNA methylation. PLoS One 10.

Nieto, M.A. (2002). The snail superfamily of zinc-finger transcription factors. Nat. Rev. Mol. Cell Biol. 3, 155–166.

Ocaña, O.H., Córcoles, R., Fabra, Á., Moreno-Bueno, G., Acloque, H., Vega, S., Barrallo-Gimeno, A., Cano, A., and Nieto, M.A. (2012). Metastatic Colonization Requires the Repression of the Epithelial-Mesenchymal Transition Inducer Prrx1. Cancer Cell 22, 709–724.

Ogata, N., Ueda, K., Kawaichi, M., and Hayaishi, O. (1981). Poly(ADP-ribose) synthetase, a main acceptor of poly(ADP-ribose) in isolated nuclei. J. Biol. Chem.

Oskarsson, T., Batlle, E., and Massagué, J. (2014). Metastatic stem cells: Sources, niches, and vital pathways. Cell Stem Cell 14, 306–321.

Petrucco, S. (2003). Sensing DNA damage by PARP-like fingers. Nucleic Acids Res. 31, 6689–6699.

Polette, M., Mestdagt, M., Bindels, S., Nawrocki-Raby, B., Hunziker, W., Foidart, J.M., Birembaut, P., and Gilles, C. (2007). β-catenin and ZO-1: Shuttle molecules involved in tumor invasion-associated epithelial-mesenchymal transition processes. Cells Tissues Organs 185, 61–65.

Ramesh, S., Qi, X.-J., Wildey, G.M., Robinson, J., Molkentin, J., Letterio, J., and Howe, P.H. (2008). TGFb-mediated BIM expression and apoptosis are regulated through SMAD3-dependent expression of the MAPK phosphatase MKP2. EMBO Rep. 9.

Ren, B. (2000). Genome-Wide Location and Function of DNA Binding Proteins. Science. 290, 2306–2309.

48

Roberts, a B., Anzano, M. a, Lamb, L.C., Smith, J.M., and Sporn, M.B. (1981). New class of transforming growth factors potentiated by epidermal growth factor: isolation from non-neoplastic tissues. Proc. Natl. Acad. Sci. U. S. A. 78, 5339–5343.

Rosenthal, F., Feijs, K.L.H., Frugier, E., Bonalli, M., Forst, A.H., Imhof, R., Winkler, H.C., Fischer, D., Caflisch, A., Hassa, P.O., et al. (2013). Macrodomain-containing proteins are new mono-ADP-ribosylhydrolases. Nat. Struct. Mol. Biol. 20, 502–507.

Saharinen, J., Hyytiäinen, M., Taipale, J., and Keski-Oja, J. (1999). Latent transforming growth factor-beta binding proteins (LTBPs)--structural extracellular matrix proteins for targeting TGF-beta action. Cytokine Growth Factor Rev. 10, 99–117.

Scanlon, C.S., Van Tubergen, E. a, Inglehart, R.C., and D’Silva, N.J. (2013). Biomarkers of epithelial-mesenchymal transition in squamous cell carcinoma. J. Dent. Res. 92, 114–121.

Scheel, C., and Weinberg, R.A. (2011). Phenotypic plasticity and epithelial-mesenchymal transitions in cancer and normal stem cells? Int. J. Cancer 129, 2310–2314.

Scheel, C., Eaton, E.N., Hsin, S., Li, -Jung, Chaffer, C.L., Reinhardt, F., Kah, K.-J., Bell, G., Guo, W., Rubin, J., et al. (2011). Paracrine and Autocrine Signals Induce and Maintain Mesenchymal and Stem Cell States in the Breast.

Schreiber, V., Dantzer, F., Ame, J.-C., and de Murcia, G. (2006). Poly(ADP-ribose): novel functions for an old molecule. Nat. Rev. Mol. Cell Biol. 7, 517–528.

Schweitzer, B., Wiltshire, S., Lambert, J., O’Malley, S., Kukanskis, K., Zhu, Z., Kingsmore, S.F., Lizardi, P.M., and Ward, D.C. (2000). Immunoassays with rolling circle DNA amplification: A versatile platform for ultrasensitive antigen detection. Proc. Natl. Acad. Sci. 97, 10113–10119.

Seoane, J., Pouponnot, C., Staller, P., Schader, M., Eilers, M., and Massagué, J. (2001). TGFbeta influences Myc, Miz-1 and Smad to control the CDK inhibitor p15INK4b. Nat. Cell Biol. 3, 400–408.

Shi, M., Zhu, J., Wang, R., Chen, X., Mi, L., Walz, T., and Springer, T.A. (2011). Latent TGF-β structure and activation. Nature 474, 343–349.

Singha, P.K., Pandeswara, S., Geng, H., Lan, R., Venkatachalam, M.A., and Saikumar, P. (2014). TGF-β induced TMEPAI/PMEPA1 inhibits canonical Smad signaling through R-Smad sequestration and promotes non-canonical PI3K/Akt signaling by reducing PTEN in triple negative breast cancer. Genes Cancer 5, 320.

Sorrentino, A., Thakur, N., Grimsby, S., Marcusson, A., von Bulow, V., Schuster, N., Zhang, S., Heldin, C.-H., and Landström, M. (2008). The type I TGF-β receptor engages TRAF6 to activate TAK1 in a receptor kinase-independent manner. Nat. Cell Biol. 10, 1199–1207.

Spellman, P.T., Sherlock, G., Zhang, M.Q., Iyer, V.R., Anders, K., Eisen, M.B., Brown, P.O., Botstein, D., and Futcher, B. (1998). Comprehensive identification of cell cycle-regulated genes of the yeast Saccharomyces cerevisiae by microarray hybridization. Mol. Biol. Cell 9, 3273–3297.

Swindall, A.F., Stanley, J. a., and Yang, E.S. (2013). PARP-1: Friend or foe of DNA damage and repair in tumorigenesis? Cancers (Basel). 5, 943–958.

Szklarczyk, D., Morris, J.H., Cook, H., Kuhn, M., Wyder, S., Simonovic, M., Santos, A., Doncheva, N.T., Roth, A., Bork, P., et al. (2017). The STRING database in 2017: quality-controlled protein-protein association networks, made broadly accessible. Nucleic Acids Res. 45, D362–D368.

49

Söderberg, O., Gullberg, M., Jarvius, M., Ridderstråle, K., Leuchowius, K.-J., Jarvius, J., Wester, K., Hydbring, P., Bahram, F., Larsson, L.-G., et al. (2006). Direct observation of individual endogenous protein complexes in situ by proximity ligation. Nat. Methods 3, 995–1000.

Takagi, T., Moribe, H., Kondoh, H., and Higashi, Y. (1998). DeltaEF1, a zinc finger and homeodomain transcription factor, is required for skeleton patterning in multiple lineages. Development 125, 21 LP-31.

Takekawa, M., Tatebayashi, K., Itoh, F., Adachi, M., Imai, K., and Saito, H. (2002). Smad‐dependent GADD45β expression mediates delayed activation of p38 MAP kinase by TGF‐β. EMBO J. 21, 6473 LP-6482.

Tan, E.J., Kahata, K., Idås, O., Thuault, S., Heldin, C.H., and Moustakas, A. (2015). The high mobility group A2 protein epigenetically silences the Cdh1 gene during epithelial-to-mesenchymal transition. Nucleic Acids Res. 43, 162–178.

Tewari, M., Quan, L.T., O’Rourke, K., Desnoyers, S., Zeng, Z., Beidler, D.R., Poirier, G.G., Salvesen, G.S., and Dixit, V.M. (1995). Yama/CPP32β, a mammalian homolog of CED-3, is a CrmA-inhibitable protease that cleaves the death substrate poly(ADP-ribose) polymerase. Cell 81, 801–809.

Thomas, D.A., and Massagué, J. (2005). TGF-β directly targets cytotoxic T cell functions during tumor evasion of immune surveillance. Cancer Cell 8, 369–380.

Tulin, A. (2003). Chromatin Loosening by Poly(ADP)-Ribose Polymerase (PARP) at Drosophila Puff Loci. Science. 299, 560–562.

Urist, M.R. (1965). Bone: formation by autoinduction. 1965. Clin. Orthop. Relat. Res. 150, 4–10.

Venneti, S., Garimella, M.T., Sullivan, L.M., Martinez, D., Huse, J.T., Heguy, A., Santi, M., Thompson, C.B., and Judkins, A.R. (2013). Evaluation of Histone 3 Lysine 27 Trimethylation (H3K27me3) and Enhancer of Zest 2 (EZH2) in Pediatric Glial and Glioneuronal Tumors Shows Decreased H3K27me3 in H3F3A K27M Mutant Glioblastomas. Brain Pathol. 23, 558–564.

Verdone, L., La Fortezza, M., Ciccarone, F., Caiafa, P., Zampieri, M., and Caserta, M. (2015). Poly(ADP-ribosyl)ation affects histone acetylation and transcription. PLoS One 10.

Vincent, T., Neve, E.P.A., Johnson, J.R., Kukalev, A., Rojo, F., Albanell, J., Pietras, K., Virtanen, I., Philipson, L., Leopold, P.L., et al. (2009). A SNAIL1-SMAD3/4 transcriptional repressor complex promotes TGF-β mediated epithelial-mesenchymal transition. Nat. Cell Biol. 11, 943–950.

Voorneveld, P.W., Kodach, L.L., Jacobs, R.J., Liv, N., Zonnevylle, A.C., Hoogenboom, J.P., Biemond, I., Verspaget, H.W., Hommes, D.W., De Rooij, K., et al. (2014). Loss of SMAD4 Alters BMP Signaling to Promote Colorectal Cancer Cell Metastasis via Activation of Rho and ROCK. Gastroenterology 147, 196–208.e13.

Wakefield, L.M., and Hill, C.S. (2013). Beyond TGFβ: roles of other TGFβ superfamily members in cancer. Nat. Rev. Cancer 13, 328–341.

Wang, Z., Gerstein, M., and Snyder, M. (2009). RNA-Seq: a revolutionary tool for transcriptomics. Nat. Rev. Genet. 10, 57–63.

Watanabe, Y., Papoutsoglou, P., Maturi, V., Tsubakihara, Y., Hottiger, M.O., Heldin, C.H., and Moustakas, A. (2016). Regulation of bone morphogenetic protein signaling by ADP-ribosylation. J. Biol. Chem. 291, 12706–12723.

Wellner, U., Brabletz, T., and Keck, T. (2010). ZEB1 in pancreatic Cancer. Cancers (Basel). 2, 1617–1628.

Wells, J., and Farnham, P.J. (2002). Characterizing transcription factor binding sites using formaldehyde crosslinking and immunoprecipitation. Methods 26, 48–56.

50

Whiteman, E.L., Liu, C.-J., Fearon, E.R., and Margolis, B. (2008). The transcription factor snail represses Crumbs3 expression and disrupts apico-basal polarity complexes. Oncogene 27, 3875–3879.

Williams, T.M., Moolten, D., Burlein, J., Romano, J., Bhaerman, R., Godillot, a, Mellon, M., Rauscher, F.J., and Kant, J. a (1991). Identification of a zinc finger protein that inhibits IL-2 gene expression. Science 254, 1791–1794.

Wu, M.Y., and Hill, C.S. (2009). TGF-β Superfamily Signaling in Embryonic Development and Homeostasis. Dev. Cell 16, 329–343.

Yaari, G., Bolen, C.R., Thakar, J., and Kleinstein, S.H. (2013). Quantitative set analysis for gene expression: A method to quantify gene set differential expression including gene-gene correlations. Nucleic Acids Res. 41.

Yang, W.-L., Wang, J., Chan, C.-H., Lee, S.-W., Campos, A.D., Lamothe, B., Hur, L., Grabiner, B.C., Lin, X., Darnay, B.G., et al. (2009). The E3 ligase TRAF6 regulates Akt ubiquitination and activation. Science 325, 1134–1138.

Yu, W., Ginjala, V., Pant, V., Chernukhin, I., Whitehead, J., Docquier, F., Farrar, D., Tavoosidana, G., Mukhopadhyay, R., Kanduri, C., et al. (2004). Poly(ADP-ribosyl)ation regulates CTCF-dependent chromatin insulation. Nat. Genet. 36, 1105–1110.

Zarzynska, J.M. (2014). Two faces of TGF-beta1 in breast cancer. Mediators Inflamm. 2014.

Zeisberg, M., and Neilson, E.G. (2009). Biomarkers for epithelial-mesenchymal transitions. J. Clin. Invest. 119, 1429–1437.

Zhang, J., Zhang, X., Xie, F., Zhang, Z., van Dam, H., Zhang, L., and Zhou, F. (2014a). The regulation of TGF-β/SMAD signaling by protein deubiquitination. Protein Cell 5, 503–517.

Zhang, J., Schindler, T., Küng, E., Ebeling, M., and Certa, U. (2014b). Highly sensitive amplicon-based transcript quantification by semiconductor sequencing. BMC Genomics 15, 565.

Zhang, Q., Yang, X., Pins, M., Javonovic, B., Kuzel, T., Kim, S.J., Van Parijs, L., Greenberg, N.M., Liu, V., Guo, Y., et al. (2005). Adoptive transfer of tumor-reactive transforming growth factor-β-insensitive CD8+ T cells: Eradication of autologous mouse prostate cancer. Cancer Res. 65, 1761–1769.

Zieba, A., Pardali, K., Söderberg, O., Lindbom, L., Nyström, E., Moustakas, A., Heldin, C.-H., and Landegren, U. (2012). Intercellular variation in signaling through the TGF-β pathway and its relation to cell density and cell cycle phase. Mol. Cell. Proteomics 11, M111.013482.

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1404

Editor: The Dean of the Faculty of Medicine

A doctoral dissertation from the Faculty of Medicine, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofMedicine. (Prior to January, 2005, the series was publishedunder the title “Comprehensive Summaries of UppsalaDissertations from the Faculty of Medicine”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-334411

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2018