Regulation of TGF-Β Signaling by Post-Translational Modifications

To my parents

List of Papers

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

I Lönn, P., van der Heide, L.*, Dahl, M.*, Hellman, U., Heldin, C-H., Moustakas, A. PARP-1 attenuates Smad-mediated transcription. Accepted manuscript, Molecular Cell (2010) II Lönn, P., Heldin, C-H., Moustakas, A. TGFβ-mediated transcription is regulated by PARP-1/PARG-balanced ADP- ribosylation. Manuscript III Kowanetz, M.*, Lönn, P.*, Vanlandewijck, M., Kowanetz, K., Heldin, C-H., Moustakas, A. TGFβ induces SIK to negatively regulate type I receptor kinase signaling. The Journal of Cell Biology (2008) Aug 25;182(4):655-62. IV Lönn, P.*, Vanlandewijck, M.*, Kowanetz, M., Kowanetz, K., Heldin, C-H., Moustakas, A. SIK and Smurf2 cooperate to downregulate the TGF-β type I receptor. Manuscript

* Indicates that these authors contributed equally to the work

Reprints were made with permission from the respective publishers.

Related publications

I Lönn, P., Morén, A., Raja E., Dahl, M., Moustakas, A. Regu- lating the stability of TGFβ receptors and Smads. Cell Research (2009) Jan;19(1):21-35. Review

Contents

Introduction ...... 11 Transforming growth factor-β ...... 12 TGF-β/Smad signaling components ...... 12 TGF-β family members ...... 12 TGF-β receptors...... 13 Smad proteins ...... 14 TGF-β/Smad signaling – from cell membrane to gene promoters ...... 16 Ligand activation ...... 16 Receptor activation ...... 17 Smad activation ...... 18 Smad nuclear shuttling ...... 19 Smad-mediated transcriptional regulation ...... 21 Non-Smad signaling ...... 21 Physiological regulation by TGF-β ...... 22 Post-translational regulation of TGF-β/Smad signaling ...... 23 Post-translational regulation of TGF-β receptors ...... 23 Post-translational regulation of R-Smads ...... 24 Post-translational regulation of the Co-Smad ...... 26 Post-translational regulation of I-Smads ...... 27 Concluding remarks ...... 27 ADP-ribosylation ...... 29 Poly (ADP-ribose) polymerase-1 ...... 29 Salt-inducible kinase-1 ...... 34 Present Investigation ...... 36 Aim ...... 36 Paper I ...... 36 Paper II ...... 38 Paper III ...... 38 Paper IV ...... 39 Future perspectives ...... 40 Acknowledgments...... 44 References ...... 46

Abbreviations

3-AB 3-Aminobenzamide ALK Activin receptor-like kinase ADP Adenosine 5´ -diphosphate AMP Adenosine 5´ -monophosphate AMPK AMP-activated protein kinase AP-1 Activator protein 1 BMP Bone morphogenetic protein BRCT BRCA1-C-terminal CBP Creb-binding protein CDK Cyclin dependent kinase cPML cytoplasmic promyelocytic leukemia protein CRM1 Chromosome region maintenance-1 CTCF CCCTC-binding factor DBD DNA-binding domain EMT Epithelial to mesenchymal transition ERK Extracellular signal-regulated kinase FOX Forkhead box GDF Growth differentiation factors HaCaT Human keratinocyte cell line HMGA High mobility group A JNK c-Jun N-terminal kinase kDa kilodalton LAP Latency-associated peptide LTBP Latent TGF-β binding protein MAPK Mitogen activated protein kinase MH Mad-homology MIS Müllerian inhibiting substance NAD Nicotinamide adenine dinucleotide NES Nuclear export signal NLS Nuclear localization signal NMuMG Normal murine mammary gland OAZ Ornithine decarboxylase antizyme 1 PARP-1 Poly(ADP-ribose) polymerase-1

PARG Poly(ADP-ribose) glycohydrolase PARylation Poly(ADP-ribosyl)ation PKA Protein kinase A SARA Smad anchor for receptor activation SBE Smad binding element SIK Salt-inducible kinase siRNA Small interfering RNA Smad Small mothers against decapentaplegic Smurf Smad ubiquitylation regulatory factor SNF1 Sucrose-non-fermented 1 SNF1LK SNF1-like kinase TAK1 TGF-β activated kinase 1 TβRI/TβRII TGF-β type I/type II receptor TGF-β Transforming growth factor-β UBA Ubiquitin associated domain YY1 Yin yang 1 ZEB Zinc-finger E-box binding homeobox

Introduction

The competence of cells to give, receive and interpret information, builds the foundation for cell to cell communication. Signaling systems have developed between both neighbouring and distant cells, as well as between cells and their microenvironment. Targeted cells respond by initiating intracellular changes, such as protein re-localizations, protein interactions, protein degradation, altering enzyme kinetics or de novo protein synthesis, ultimately leading to desired physiological reactions. These stringently controlled interaction networks allow cells to act as a unit to regulate and synchronize ongo- ing cellular processes including proliferation, differentiation, migration and apoptosis. Such orchestrated actions are essential for complex multicellular life forms to develop and exist. The importance of cell communication is highlighted by the severe disease states that are linked to signaling malfunc- tion. Intracellular signaling relays are often regulated by post-translational modifications. These are enzyme-dependent additions of small chemical groups or polypeptides to specific amino acids of target proteins. A multitude of post-translational modifications usually participate during signal propagation, acting alone or in concert with each other to secure precise timing and action of cellular processes. Several types of stimuli can induce cellular responses. One important group is secreted factors, such as hormones, neurotransmitters and cytokines, which through blood vessels, lymph vessels or close cellular contacts allow cells to signal and respond. This thesis focuses and dwells on transforming growth factor-β (TGF-β), the prototype signaling polypeptide of the TGF-β family of cytokines, and its downstream activation of Smad signaling pathways. Although TGF-β/Smad signaling has acquired significant attention since its discovery three decades ago, many regulatory steps of this pathway remain elusive. The present investigation aims to expand the knowledge of how TGF-β and Smads regulate the initiation and termination of transcriptional responses. It highlights the importance of post-translational modifications in the pathway, such as phosphorylation, ubiquitination and ADP- ribosylation.

11 Transforming growth factor-β

TGF-β/Smad signaling components TGF-β family members TGF-β was isolated and described three decades ago as a secreted signaling polypeptide that can cause oncogenic transformation of certain specific cell types (Roberts et al. 1980; Moses et al. 1981). Today, TGF-β represents a family of numerous structurally related cytokines transcribed by 33 genes in humans and is known to regulate many different cellular processes (Shi and Massagué 2003; Feng and Derynck 2005; Moustakas and Heldin 2009). The family comprises TGF-βs, bone morphogenetic proteins (BMPs), ac- tivins/inhibins, nodal, growth differentiation factors (GDFs), Müllerian inhibiting substance (MIS) and few others (Figure 1) (Derynck and Miyazono 2008). The members are often divided into three subfamilies: TGF-βs, BMPs and activin/inhibin/nodal. Together they control cellular activities, including cell growth, cell migration, cell differentiation and cell survival (Massagué et al. 2000; Patterson and Padgett 2000; Moustakas 2002; Heldin et al. 2009; Wu and Hill 2009). TGF-β signaling is well conserved during evolution and components of their transduction pathways can be traced back to Trichoplax adhaerens and Caenorhabditis elegans (Huminiecki et al. 2009). TGF-β family member structures resemble a four-digit hand (Daopin et al. 1992; Schlunegger and Grutter 1992; Daopin et al. 1993; Lin et al. 2006). They have two pairs of β-strands that protrude out from an α-helix. Each pair of β-strands aligns next to each other but in opposite direction. The α- helix is thought of as the wrist and each β-strand as a finger. The structural integrity of the monomer is upheld by six cysteine residues forming three disulfide bonds. These disulfide bridges build a “cystine knot”, where one disulfide bond links through a ring formed by the two other bonds. Active TGF-β family ligands appear as homodimers or in rare cases as heterodimers (Lin et al. 2006). The dimers are, with few exceptions, linked via hydrophobic interactions and disulphide bonds. Interestingly, while the family ligand monomers are highly similar, the dimer conformations are quite diverse. The dimers are thought to be flexible and to adopt alternative conformations e.g. when binding to receptors (Bocharov et al. 2002; Thomp- son et al. 2003). The α-helix region is reported to be the most dynamic part

12 of the structure. The work presented in this thesis has been performed using the prototype member of the family, TGF-β1 (referred to as TGF-β in the papers).

TGF-β receptors TGF-β family ligands bind to type I and type II serine/threonine kinase receptors at cell surfaces (Shi and Massagué 2003; ten Dijke and Hill 2004; Feng and Derynck 2005). The human genome comprises twelve receptors devoted to these ligands (Figure 1): seven type I receptors (Activin receptor- like kinase, ALK, 1-7) and five type II receptors (TβRII, ActRII, ActRIIb, BMPRII and AMHR-II) (Manning et al. 2002; Moustakas and Heldin 2009). The receptors are glycoproteins with a ligand-interacting extracellular domain, a single membrane-spanning transmembrane domain and an intracellular domain upholding serine/threonine kinase activity (Figure 2). In addition, the type I receptors also carry a highly conserved glycine/serine- rich intracellular juxtamembrane stretch (GS-box) and a Smad-interacting L45 loop within the kinase domain. The type I receptors are approximately 55 kDa in size while type II receptors are around 70 kDa.

Figure 1. Simplified schematic presentation of TGF-β family ligands and the downstream receptors and R-Smads which they activate.

Figure 2. Illustration of a ligand-bound hetero-tetrameric receptor complex. Acti- vated TGF-β ligands (Tβ) are released from latency-associated peptides (LAP) and bind to type I and type II receptors. Multiple phosphorylations are represented by yellow dots. Transactivation of the type I receptor by the type II receptor is visua- lized by the black arrow.

Ligand binding shapes an active hetero-tetrameric receptor complex consist- ing of two type I receptors and two type II receptors (Lin et al. 2006; Groppe et al. 2008; Nickel et al. 2009) (Figure 2). Different superfamily ligands have different affinity for the two receptor types (Figure 1), which means that induction of complex-formation can occur in various orders. The assembly, configuration and orientation of the complex play important roles for which downstream signaling pathways that will become activated (Figure 1) (Shi and Massagué 2003). Thus, specific responses to TGF-β superfamily members are influenced by expression and availability of both ligands and receptors.

Smad proteins The most studied intracellular effector proteins of the TGF-β pathway are the Smads (Derynck et al. 1996; Feng and Derynck 2005; Massagué et al. 2005). The name Smad originates from the time when the proteins were first uncovered and annotated as small body size (Sma) in Caenorhabditis elegans and mothers against decapentaplegic (Mad) in Drosophila melanogaster

14 (Sekelsky et al. 1995; Savage et al. 1996). Eight Smads (Smad1 to 8) have been identified in the human genome (Massagué et al. 2005). These eight Smads are divided into three subclasses: receptor-regulated Smads (R- Smads), common mediator of Smad (Co-Smad) and inhibitory Smads (I- Smads). R-Smads comprise Smad1, 2, 3, 5 and 8, and are, as defined by their names, substrates of type I receptor serine/threonine kinases (Abdollah et al. 1997; Kretzschmar et al. 1997; Souchelnytskyi et al. 1997). Smad2 and 3 generally propagate signals from TGF-βs, activin and nodal ligand stimuli while Smad1, 5 and 8 mainly propagate signals by BMPs and GDFs (Figure 1) (Feng and Derynck 2005; Massagué et al. 2005; Sieber et al. 2009). Smad4 serves as a universal binding partner for all the R-Smads and is therefore called the common Smad (Co-Smad). The two Smads remaining, Smad6 and 7, are the inhibitory Smads. They are transcriptionally induced during TGF-β family signaling and serve as key negative regulators of the pathways (view later chapters) (Lonn et al. 2009). R-Smads and Co-Smad have similar domain structures which include an N-terminal Mad-homology 1 (MH1) domain, a C-terminal Mad-homology 2 (MH2) domain and a linker region (Figure 3). I-Smads have a MH2 domain but lack the features of an N-terminal MH1 domain. The three regions possess quite diverse characteristics (Massagué et al. 2005). The N-terminal MH1 domain is stabilized by a zinc atom and it has a nuclear localization signal (NLS) and a β-hairpin structure that enables DNA binding (Xiao et al. 2000; Kurisaki et al. 2001; Chai et al. 2003). However, Smad2 lacks DNA binding capability due to two short extra sequence inserts in the MH1 domain, inhibiting the function of the β-hairpin (Dennler et al. 1999; Kurisaki et al. 2001). In addition, a number of transcriptional co-regulators bind to the MH1 domain. The linker region is structurally flexible and contains a PPXY (PY) motif for WW-domain containing ubiquitin ligases (Zhu et al. 1999), and phosphorylation sites for kinases such as mitogen activated protein kinases (MAPKs) and cyclin dependent kinases (CDKs) (Kretzschmar et al. 1997; Mulder 2000; Pera et al. 2003; Liu 2006; Alarcon et al. 2009). How- ever, the linker region of Smad4 lacks the PY-motif but contains a classical nuclear export signal (NES) (Watanabe et al. 2000). The C-terminal MH2 domain possesses sites for interactions with type I receptors and other Smad molecules (Chen et al. 1998). In addition, several transcription factors and co-activators also bind to this region (Ross and Hill 2008). For this reason, the MH2 domain confers transactivation properties to the Smad proteins. The MH2 domain of the R-Smads also includes a highly conserved Ser-X- Ser motif at the very C-teminal end that can be phosphorylated by type I receptors (Abdollah et al. 1997; Kretzschmar et al. 1997; Souchelnytskyi et al. 1997).

Figure 3 . Overview of the R-Smad, Co-Smad and I-Smad domain structures. (MH1/MH2, mad-homology 1/2, NLS, nuclear localization signal; NES, nuclear export signal; SAD, Smad4 activation domain)

TGF-β/Smad signaling – from cell membrane to gene promoters The TGF-β family signals from cell membrane to nuclear gene promoters via a chain of molecular events that are stringently controlled. The subsequent discussion will focus on the cascades initiated by the prototype member of the family, TGF-β, which is the ligand used in the present investigation. Brief reflections on interesting differences or similarities with other family member signaling pathways will be highlighted.

Ligand activation TGF-β ligands are secreted as latent complexes. The precursor contains an N-terminal propeptide segment and a C-terminal mature TGF-β sequence (Annes et al. 2003; Keski-Oja et al. 2004; Rifkin 2005). After cleavage of the C-terminal disulfide-linked TGF-β homodimer it remains associated with the N-terminal propeptide, called the latency-associated peptide (LAP), and is therefore retained in an inactive state (Figure 2 and 4). The LAP/TGF-β complex interacts with latent TGF-β binding proteins (LTBPs) to form the- large latent complex (LLC), which associates with various extracellular matrix proteins. Activation of the latent TGF-β and subsequent release from extracellular matrix to enable receptor interaction occur through mechanisms

16 that disrupt the ligand binding of LAP and LTBP proteins. This can be executed by several different proteases and other interacting factors including thrombospondin-1, integrin αvß6, reactive oxygen species (ROS) or low pH (Katri Koli 2001). The system of synthesis, storage, latency and activation pinpoints how various tissues may quickly regulate the onset and activity of TGF-β signaling through ligand availability (Annes et al. 2003). Additional regulation of ligand availability is exerted by accessory receptors and ligand traps (Shi and Massagué 2003). Accessory receptors enhance signaling by delivering and elevating ligand concentration close to receptors. Known accessory receptors are betaglycan (also referred to as the TGF-β type III receptor), cripto and endoglin. Ligand traps, on the other hand, can bind to specific TGF-β family members and hinder them from reaching and activating their receptors. They include decorin, α2-macroglobulin, fol- listatin, DAN/Cerberus, Chordin/SOG, Noggin and the already discussed LAP (Shi and Massagué 2003).

Receptor activation TGF-β binds to the extracellular domain of the TGF-β type II receptor (TβRII) serine/threonine kinase (Figure 2 and 4) (Wrana et al. 1994; Shi and Massagué 2003; ten Dijke and Hill 2004; Feng and Derynck 2005; Mousta- kas and Heldin 2009). The TGF-β type I receptor (ALK5) is then recruited and a hetero-tetrameric complex is formed between two type II and two type I receptors. Other family member ligands may bind receptors in different order and whether receptors are recruited as single units or as preformed complexes is actively investigated. TβRII is constitutively active and autophosphorylation of serine residues 213 and 409 is particularly important for full kinase activity (Luo and Lodish 1997). TβRII transactivates ALK5 within the ligand-receptor complex by phosphorylating serine and threonine residues in the GS-box sequence close to the kinase domain (Wrana et al. 1994). Phosphorylation of the intracellular domain of ALK5 triggers the kinase activity and allows docking of the L3 loops of the R-Smads, Smad2 and 3, to the L45 loop of the type I receptor (Chen et al. 1998). Both activated and non-activated TGF-β receptors undergo endocytosis (Chen 2009). In essence, internalization is directed either via clathrin-coated pits to early endosomes, a route that enhances TGF-β signaling and enables re-cycling of receptors back to the cell-membrane by Rab proteins (Mitchell et al. 2004); or via caveolin-positive lipid rafts which promote degradation of the receptors and thereby stalls signaling (Lonn et al. 2009). The exact internalization signal and sorting mechanism are still not fully understood.

Figure 4 Simplified illustration of the TGF-β/Smad signaling pathway. Constitu- tively activated TβRIIs associate with TβRIs upon ligand binding, whereafter TβRIIs phosphorylate TβRIs. R-Smads (Smad2/3) are phosphorylated by activated TβRIs and translocate into the nucleus together with Smad4. Inside the nucleus, the R- Smad/Co-Smad complexes bind to DNA and regulate gene transcription (arrow). The Smad complex dissociate from DNA and its subunits recycle back to the cytoplasm. (LAP, Latency associated peptide; Tβ, TGF-β; S2/3, Smad2/3; S4, Smad4).

Smad activation When R-Smads bind the receptor complex, the type I receptor directly phosphorylates them at a highly conserved Ser-X-Ser motif situated at the very C-terminal of the protein (Figure 4) (Abdollah et al. 1997; Kretzschmar et al.

18 1997; Souchelnytskyi et al. 1997). This phosphorylation is a hallmark of Smad signaling since it promotes formation of the R-Smad/Co-Smad complex and subsequently induce interactions with other transcriptional co- regulators. The R-Smad/Co-Smad complex is formed as a trimer, with two phosporylated R-Smads binding to one Co-Smad (Chacko et al. 2001). The recruitment of each specific Smad to the activated ligand-receptor complex is dictated by the configuration of the complex together with different anchoring, scaffolding and chaperone proteins (Chen et al. 1998; Feng and Derynck 2005). A well studied anchoring protein important for Smad recruitment to activated receptors is the Smad anchor for receptor activation (SARA), which binds Smad2 and Smad3 and localizes to early endosomes (Tsukazaki et al. 1998; Xu et al. 2000; Itoh et al. 2002). SARA cooperates with cytoplasmic promyelocytic leukemia (cPML) which upon TGF-β stimulation is released from the TG-interacting factor (TGIF) in the nucleus by the PML competitor for TGIF association and relocates to the cytoplasm to join SARA (Lin et al. 2004; Faresse et al. 2008). In contrast, TMEPAI (transmembrane prostate androgen induced), was recently shown to bind R- Smads and compete with SARA and cPML to prevent Smads from reaching receptors (Watanabe et al. 2010). C-terminal R-Smad phosphorylation can be regulated by the I-Smads. For instance, I-Smads interact with the serine/threonine kinase receptors which prevent R-Smad phosphorylation (Hayashi et al. 1997; Imamura et al. 1997; Nakao et al. 1997; Hata et al. 1998; Mochizuki et al. 2004). The I-Smads can also interfere with the formation of the R-Smad/Co-Smad complex (Hata et al. 1998). Finally, I-Smads can promote degradation of key signaling proteins by recruiting ubiquitin E3 ligases to the receptors and to the Smads (Lonn et al. 2009). These mechanisms are further described in the chapter on Smads and post-translational modifications.

Smad nuclear shuttling R-Smads and Co-Smad are thought to constantly shuttle between cytoplasm and nucleus (Pierreux et al. 2000; Inman et al. 2002). In the non-signaling, resting state, R-Smads localize foremost to the cytosol. Smad4 has a more equally distributed pattern between cytosol and nucleus. Investigations by Nicolás and colleagues have shown that export rates for R-Smads are higher than the import rates, while Smad4 has a more even import/export rate (Nicolas et al. 2004; Schmierer and Hill 2005), which explains the Smad distributions in the unstimulated condition. Furthermore, R-Smad localization have been proposed to be affected by SARA and other cytoplasmic teth- ering proteins (Xu et al. 2000). R-Smads, when activated, accumulate in the nucleus and regulate transcription. Different nuclear retention factors, such as FoxH1/FAST and TAZ, have been suggested to keep selected Smads in the nucleus (Xu et al.

19 2002; Varelas et al. 2008). Collectively, experiments and modelling of Smad2 and Smad3 nucleocytoplasmic trafficking suggest that both import and export rates affect nuclear R-Smad accumulation (Nicolas et al. 2004; Schmierer and Hill 2005). This means that the most likely scenario is that a combination of events, involving different retention factors that sequester the proteins in the nucleus or cytoplasm together with other modifications and mechanisms that change the import and export rates of Smad proteins are responsible for proper shuttling. Regardless of the factors responsible for upholding Smad distribution, it is clear that the Smads need the capability to pass through the nuclear pores in both directions. The exact mechanism to cross these pores is still under examination, however, several different alternatives have been proposed and point towards the fact that numerous export and import mechanisms are involved in mediating proper Smad shuttling. Smad4, for example, has an NLS in its MH1 domain that can interact with importin-α1 to regulate its nuclear entry, and an NES motif in the linker region which interacts with the nuclear exporter CRM1 (Chromosome region maintenance-1) (Watanabe et al. 2000; Xiao et al. 2003). Smad3 also possesses an NLS signal in the MH1 domain that binds to importin-β and requires Ran (Xiao et al. 2000; Kurisaki et al. 2001). Smad2 contains the same NLS motif as Smad3 in its MH1 domain, but interestingly it is reported to use a different mechanism of shuttling, a cytosolic-factor-independent import activity that requires a region in the MH2 domain (Xu et al. 2000; Kurisaki et al. 2001). This difference between Smad2 and Smad3 could be due to the uniquely spliced-in exon 3 in the MH1 domain of Smad2, introducing an extra stretch of amino acids that affects the function of this domain (Kurisaki et al. 2001). In addition, recent investigations have linked TGF-β-mediated Smad1, 3 and 4 nuclear accumulation to importin7 and 8 (Xu et al. 2007; Yao et al. 2008). On the other hand, there are reports showing that Smad shuttling can also be mediated through direct interactions with nucleoporins, particularly Nup214 and Nup153 (Xu et al. 2002; Xu et al. 2003). Export of Smad3 is addressed by an exportin-4-dependent mechanism (Kurisaki et al. 2006). Cumulatively, all these data imply that a rather complex situation is at hand when it comes to regulation of Smad shuttling. A possible scenario could be that nuclear transport receptors promote Smad re-localization to specific nucleoporins, which then take over the transport by specific interactions with the Smads. Alternatively, Smads may need multiple, isolated, nuclear/cytoplasmic shuttling mechanisms to ensure correct localization. They could work as back-up for each other, or promote specific shuttling under certain cellular conditions. R-Smads to Smad4 interactions may form either in the cytoplasm before nuclear translocation or inside the nucleus. However, the R-Smad/Co- Smad complexes mainly reside in the nucleus under stimulatory conditions. The exact mechanism of shuttling is still an open question in the TGF-β field that needs further attention.

20 Smad-mediated transcriptional regulation Nuclear R-Smads/Co-Smad complexes bind to gene regulatory regions and orchestrate transcriptional responses. The β-hairpin structure in the MH1 domain of Smads contacts specific DNA sequences called Smad binding elements (SBE) (Yingling et al. 1997; Dennler et al. 1998; Zawel et al. 1998). The minimal site of binding is characterized as 5´-AGAC-3´. Visuali- zation of the DNA-bound Smad3 crystal structure uncovered the importance of hydrogen bonds between three bases of the SBE and three residues in the β-hairpin (Shi et al. 1998; Chai et al. 2003). In addition, Smads have been reported to bind to GC-rich promoter regions (Ishida et al. 2000). Acetyla- tion of lysine residues in the MH1 domain of Smad3 and of the Smad2 iso- form that lacks exon3 insert, has been shown to enhance their DNA binding in vitro and in vivo (Simonsson et al. 2006). Genes responsible for regulation of proliferation, differentiation, migration and apoptosis can be induced or repressed by TGF-β/Smad signaling. Selectivity for specific target promoters is most likely dictated through a joint venture between Smads and diverse processes and factors, such as chromatin modulation, post-translational modifications, additional DNA binding proteins and co-activators that cooperate with the Smads (Massagué et al. 2005; Ross and Hill 2008). Example of Smad transcriptional co- activators are ARC105, CBP/p300 and SMIF (Ross and Hill 2008). DNA- binding transcription factors that have been shown to affect Smad-mediated gene regulation include AP-1, p53, YY1, Sp1, OAZ, FoxO and FoxH1/FAST (Ross and Hill 2008). Cooperation with such transcription factors might require DNA-binding motifs for both Smads and the co-factors at a suitable distance and orientation. There are also several known co- repressors, such as c-Ski, SnoN and c-Myc (Ross and Hill 2008). Further- more, Smads have been shown to modulate histone and chromatin during transcription (Ross et al. 2006). It may therefore also be of importance where in the genome promoters are located as well as the state of the chromatin in that region. Together all these features highlight interesting possibilities for pathway-specific and cell type-specific transcriptional regulation.

Non-Smad signaling In addition to Smad signaling, which is the main topic of this thesis, TGF-β also activates several non-Smad pathways, including Rho, Rac and Cdc42 of the Rho-like GTPases, extracellular signal regulated kinases (ERKs), p38 and c-Jun N-terminal kinases (JNKs) of the mitogen-activated protein kinases (MAPKs) and the MAPK kinase kinase TGF-β activated kinase 1 (TAK1) (Engel et al. 1999; Yamaguchi et al. 1999; Edlund et al. 2002; Yu et al. 2002; Bhowmick et al. 2003; Itoh et al. 2003). Recently, one mechanism that links TβRI with non-Smad pathways via TAK1 has been described (Sor-

21 rentino et al. 2008). TAK1 was shown to be recruited to the receptor and activated via lysine 63-linked polyubiquitination, catalyzed by the ubiquitin E3 ligase TRAF6. Smad-independent TGF-β signals mediate important crosstalks with other major pathways and coordinate events such as epithelial-to-mesenchymal transition (EMT), apoptosis and migration (Derynck and Zhang 2003; Moustakas and Heldin 2005; Rahimi and Leof 2007).

Physiological regulation by TGF-β Cellular responses to TGF-β are diverse and include effects on proliferation, apoptosis and differentiation. Regulation of these events is important during embryonic development and in adult tissue homeostasis. Here is a brief summary of some important TGF-β-mediated physiological responses and the main mechanisms behind them. Cell cycle arrest at G1 can be induced by TGF-β in a wide range of cell types (Rahimi and Leof 2007; Heldin et al. 2009). One way that TGF-β promotes growth arrest is by Smad mediated induction of cyclin-dependent kinase (CDK) inhibitors, p15Ink4b and p21Cip1/WAF1 (Hannon and Beach 1994; Datto et al. 1995). CDK inhibition during TGF-β stimulation is further achieved by downregulation of a positive regulator of CDKs, the phosphatase Cdc25A (Iavarone and Massague 1997). Other important mechanisms in TGF-β-mediated cell cycle arrest are Smad-dependent transcriptional repression of growth related transcription factors like c-myc and Id family members (Pietenpol et al. 1990; Kang et al. 2003; Kowanetz et al. 2004). Induction of pro-apoptotic pathways by TGF-β is complex and appears to be cell-type specific (Rahimi and Leof 2007). Phosphoinositide-3-kinase (PI3K)/protein kinase B signaling and expression of Bcl-2 members are two examples of Smad-dependent mechanisms involved in TGF-β-mediated apoptosis. Non-Smad signaling that leads to downstream activation of p38 or JNK has also been linked to TGF-β-induced cell death. TGF-β is well studied for its role in driving epithelial-to-mesenchymal transition (EMT) (Rahimi and Leof 2007; Heldin et al. 2009). EMT is characterized by loss of cell-cell contacts, modulation of the extracellular matrix and reorganization of the actin cytoskeleton. It results in a less differentiated cell phenotype with better migratory capability. This process is critical during embryonic development but is also associated with tumor metastasis and fibrotic conditions. TGF-β induced EMT is propelled by a battery of upregulated transcription factors, including Snail1, Snail2, Twist, HMGA2, Zeb1 and Zeb2 (Heldin et al. 2009). In addition, TGF-β is known to regulate several other important cellular processes, such as modulation of immune responses, vascularisation and angiogenesis. Other TGF-β family members are known to regulate similar

22 but unique responses. For example, BMP is well known for its capability to induce bone and cartilage formation (Wagner et al. 2010). Although, studies on physiological responses to TGF-β family members are plentiful, a deeper discussion of all these mechanisms goes beyond the scope of this thesis which focuses mainly on mechanisms of signal transduction.

Post-translational regulation of TGF-β/Smad signaling Initiation and termination of signaling is particularly critical and needs to be sternly controlled at all levels of the cascade. Ligand and receptor availability has already been discussed in previous chapters and is of key importance for TGF-β on/off signaling, as well as for dose dependent stimuli important during embryonic development. Another way of fine-tuning cellular signaling pathways is via post-translational modifications, i.e. the enzymatic transfer of small chemical groups, like phosphorylation and acetylation, or bulky polypeptides, such as ubiquitination and sumoylation, to specific amino acids on substrate proteins. The functional consequences of such modifications are plentiful, which means that each situation needs to be studied and discussed carefully, both as isolated events and in relation to additional modifications.

Post-translational regulation of TGF-β receptors The constitutively active TβRII is phosphorylated at several serine and threonine residues (Luo and Lodish 1997). In ligand/receptor complexes, the TβRII is capable of phosphorylating TβRI serines and threonines located in the GS-box close to the intracellular kinase domain. Phosphorylations within this box lead to kinase activation and subsequent R-Smad phosphorylation. Knowledge about signaling regulation via type I and type II receptor dephosphorylation is limited. Studies in Drosphophila melanogaster and mammalian mink lung epithelial cells (MV1Lu) have suggested that phosphatase 1c (PP1c) can work together with SARA or Smad7 to dephosphorylate the TβRI, leading to loss of signaling (Bennett and Alphey 2002; Shi et al. 2004). Phosphatase 1α (PP1α) is reported to be able to dephosphorylate the ALK1 receptor in endothelial cells (Valdimarsdottir et al. 2006). Fur- thermore, protein phosphatase 2A (PP2A) subunits can bind to TGF-β receptors (Griswold-Prenner et al. 1998; Batut et al. 2008; Bengtsson et al. 2009). However, the mode of action of PP2A, in respect to receptor dephosphorylation, remains unclear as knock-down of different subunits have shown opposing effects on downstream signaling events. Sumoylation of TβRI lysines by Ubc9 has been reported (Kang et al. 2008). Sumoylated TβRIs show promoted recruitment and phosphorylation of Smad3. Thus, sumoylation by Ubc9 enhances receptor signaling.

23 TGF-β receptors undergo endocytosis both in their resting state and after activation (Chen 2009). Internalization via caveolin postitive lipid rafts is closely linked with receptor degradation. Therefore, as expected, components important for receptor ubiquitination and subsequent degradation have been shown to localize to the caveolae. For instance, TGF-β induces and recruits Smad7 and E3 ubiquitin ligases Smad ubiquitylation regulatory factor 1 and 2 (Smurf1and 2) to the rafts (Lonn et al. 2009). BMP stimulation upregulates Smad6 and Smurf1, which also regulate receptor degradation at lipid rafts. Smurf1 and Smurf2 are two of the most studied ubiquitin E3 ligases in the TGF-β pathway. These two proteins possess HECT domains that can transfer ubiquitin moieties onto target proteins, forming Lys-48 linked polyubiquitin chains which are recognized by proteasomes and lead to substrate degradation. The I-Smad/Smurf mediated receptor ubiquitination is rather well studied. An autoinhibitory binding between the C2 domain and the HECT domain of Smurf, restricts its E3-ligase activity under resting conditions (Wi- esner et al. 2007). After signal initiation, Smad7 and Smurf protein levels increase and Smurfs interact with the PY-motif of the I-Smads. Subse- quently, the proteins are exported to the cytoplasm by CRM1 where the C2 domain of Smurf is known to interact with lipid rafts at cell membranes and the Smad7/Smurf2 complex associates with the receptors (Kavsak et al. 2000; Suzuki et al. 2002; Tajima et al. 2003). Smad7 promotes Smurf2 activity by relieving the autoinhibition and by recruiting the E2 conjugating enzyme UbcH7 (Ogunjimi et al. 2005; Wiesner et al. 2007). This allows the Smurfs to ubiquitinate TβRIs which negatively influences TGF-β signaling. Other C2-WW-HECT containing E3 ligases have been shown to be able to regulate receptor stability in similar way as the Smurfs. These include WWP1 and NEDD4-2/NEDD4L (Neuronal precursor cell-expressed, devel- opmentally downregulated 4-2/L) (Komuro et al. 2004; Seo et al. 2004; Kuratomi et al. 2005 ). Finally, a deubiquitinating enzyme, UCH37, has also been linked to TβRI regulation (Wicks et al. 2005). UCH37 cooperates with Smad7 and deubiq- uitinates the type I receptor. This leads to receptor stabilization and promotes signaling.

Post-translational regulation of R-Smads R-Smads are phosphorylated at their C-terminal Ser-X-Ser motif by activated TGF-β type I receptors, which promotes R-Smad/Co-Smad interactions and is therefore essential for Smad transcriptional regulation. In addition, R-Smads can be regulated by kinases other than the type I receptors. Recently, CDK8 and CDK9 were shown to mediate agonist-induced linker phosphorylations inside the nucleus (Alarcon et al. 2009). These site specific phosphorylations enhance TGF-β signaling but, in addition, target

24 activated Smad2 and 3 for degradation by E3 ubiquitin ligase NEDD4- 2/NEDD4L (Alarcon et al. 2009; Gao et al. 2009). CDK2 and CDK4 have also been postulated to phosphorylate Smad3 linker residues which result in transcriptional inhibition (Matsuura et al. 2004). Ras, via the ERK pathway, is also able to mediate phosphorylation of specific serine residues in the linker domain of the R-Smads (Kretzschmar et al. 1997). This leads to per- turbed nuclear translocation and attenuated transcription. However, ERK can also phosphorylate the MH1 domain of Smad2 which then results in enhanced transcriptional activity (Funaba et al. 2002). In addition, TGF-ß- activated JNK can phosphorylate Smad3 to favor its nuclear accumulation and transcriptional engagements (Engel et al. 1999). Protein kinase C (PKC) is another protein shown to phosphorylate the Smads (Yakymovych et al. 2001). PKC abrogates Smad3 DNA-binding through phosphorylation at the MH1 domain. Other direct or indirect regulations of R-Smads by various kinases have been reported (Wrighton et al. 2009). C-terminal and linker dephosphorylations of R-Smads are important, and phosphatases have been found for the BMP R-Smad, Smad1, and the TGF-β R-Smads, Smad2 and 3 (Chen et al. 2006; Knockaert et al. 2006; Lin et al. 2006; Sapkota et al. 2006; Bengtsson et al. 2009). The identified phosphatases are protein phosphatase 2A (PP2A), pyruvate dehydrogenase phosphatase (PDP) and Small C-terminal Domain Phosphatases (SCPs) for Smad1, and PPM1A/PP2Cα for Smad2 and 3. C-terminal dephosphorylation of the Smads leads to dissociation of the R-Smad/Co-Smad complex, result- ing in their nuclear export and recycling to the cytoplasm (Pierreux et al. 2000; Inman et al. 2002; Lin et al. 2006). PP2A dephosphorylates the linker in Smad1, which is reported to upregulate Smad1 signaling (Bengtsson et al. 2009). In addition, SCPs can specifically dephosphorylate the linker regions of Smad1, 2 and 3 which result in enhanced TGF-β signaling while reducing BMP signaling (Sapkota et al. 2006). Ubiquitin mediated degradation is also a profound regulatory step at the level of Smad signaling. E3 ligases, Smurf1 and Smurf2, are in addition to regulating receptor stability also important for Smad degradation. The Smurfs bind to the PY motif situated within the linker region of the R-Smads (Zhu et al. 1999). Smurf1 can target Smad1 and 5 while Smurf2 ubiquitinates Smad1, 2 and 5 (Zhu et al. 1999; Kavsak et al. 2000; Zhang et al. 2001). Although Smurfs bind to Smad3, there are no reports that they modify or mediate degradation of it. Smad3 has, on the contrary, been reported to be ubiquitinated and degraded in a Smurf-independent way through the E3 ubiquitin ligase CHIP and the E3 ligase complex SCF/Roc1 (Fukuchi et al. 2001). In addition, there are other E3 ligases reported to act on R-Smads, like WWP1, PRAJA and Nedd4-2/NEDD4L (Komuro et al. 2004; Kuratomi et al. 2005 ; Saha et al. 2006). While many R-Smad E3 ligases promote degradation and signaling clo- sure, the HECT domain E3 ligase AIP4/Itch and Ring domain E3 ligase Cbl-

25 b both ubiquitinate Smad2 in order to promote type I receptor phosphorylation and signaling (Bai et al. 2004; Wohlfert et al. 2006). Another interesting E3 ligase that regulates R-Smads is the RING domain E3 ligase Arkadia. Arkadia can target phosphorylated Smad2 and Smad3 for ubiquitin-mediated degradation (Mavrakis et al. 2007). However, Arkadia also seems to play a role for transcription initiation since loss of Arkadia in embryonic cells has been connected with lowered activity of phosphorylated Smads (Mavrakis et al. 2007). Knowledge about sumoylation of R-Smads is, so far, limited to Smad3 (Imoto et al. 2003; Imoto et al. 2008). Sumoylation of Smad3 by PIASy is suggested to limit MH1 binding to DNA and to enhance Smad3 nuclear export. Acetylation is yet another proof of the broad spectrum of modifications targeting R-Smads. Smad2 is acetylated within its MH1 domain, more specifically at Lys-9, -20 and -39, by p300 and CBP (Simonsson et al. 2006; Inoue et al. 2007; Tu and Luo 2007). Smad3 is also acetylated at Lys-19 in the MH1 domain and Lys-378 in the MH2 domain. Acetylation of R-Smads favors their DNA binding and transcriptional activity.

Post-translational regulation of the Co-Smad ERK phosphorylates Thr276 in the mouse Smad4 activation domain (Roelen et al. 2003), and mutation of this site resulted in a more cytoplasmic distribution of Smad4. It is likely that Smad4 is phosphorylated on other residues as well, but this possibility remains to be explored. For receptors and R-Smads, the discussion on ubiquitination has centered on the classical ubiquitin-proteasomal degradation pathway. However, ubiquitination is a multi-faceted post-translational modification which is relevant when discussing Smad4. Ubiquitination can namely occur as monoubiquitination where one ubiquitin moiety is added to the substrate; multi- ubiquitination where multiple mono-ubiquitins are added to a protein; or as poly-ubiquitination which adds long chains of ubiquitin (Haglund and Dikic 2005). The chains can, in turn, be interlinked via different internal lysine residues. Smad4 has been reported to be mono-ubiquitinated at Lys-507 and Lys-519, which either enhances or attenuates Smad signaling through altering Co-Smad/R-Smad complex formation (Moren et al. 2003; Dupont et al. 2005; Wang et al. 2008; Dupont et al. 2009). E3 ligase Ectodermin/TIF1γ, deubiquitinase FAM/USP9x and p300 have all been linked to the modulation of Smad4 mono-ubiquitination (Moren et al. 2003; Wang et al. 2008; Dupont et al. 2009). Furthermore, a Smad4-ubiquitin fusion protein which mimicks the monoubiquitinated Smad4 localizes mainly to the cytoplasm (Wang et al. 2008). In addition, Smad4 is subjected to Lys-48-linked poly-ubiquitination by several E3-ligases which lead to proteasomal degradation. Morén and co- workers have shown that even though Smad4 lacks the PY motif and does

26 not directly bind Smurfs, Smad2, 6 and 7 can act as a bridges to bring Smurfs and other E3 ligases close to Smad4, which allows ubiquitination and subsequent Smad4 degradation (Morén et al. 2005). In addition, Smad4 is known to be degraded via Jab1 and SCFβTrCP (Wan et al. 2002; Wan et al. 2004), and specific cancer-derived mutants of Smad4 have been shown to be degraded by SCFSkp2 (Liang et al. 2004). Smad4 is also sumoylated (Lee et al. 2003; Lin et al. 2003; Ohshima and Shimotohno 2003; Long et al. 2004). Sumolylation occurs on Lys-113 and Lys-159 in the MH1 domain. However, the functional outcome of such modification is unclear since both activation and repression of signaling have been reported.

Post-translational regulation of I-Smads Although Smurfs bind to the PY motif of Smad7, Smad7 evades degradation and works as a scaffold protein that can direct Smurfs and other C2-WW- HECT domain E3 ligases to degrade the TGF-β receptors or Smad4 (Kavsak et al. 2000; Ebisawa et al. 2001; Suzuki et al. 2002; Morén et al. 2005). Smad7 is also targeted by the E3 ligase Arkadia (Koinuma et al. 2003). Arkadia ubiquitinates Smad7 which leads to its degradation and relieves the negative funtion it exerts on the pathway. In contrast, Smad7 can also be protected from degradation via p300 mediated acetylation of two lysine residues in the MH1 domain, Lys-64 and Lys- 70 (Grönroos et al. 2002). Smad7 acetylation is reversible, as HDAC1 and Sirt1 mediate Smad7 deacetylation (Simonsson et al. 2005; Kume et al. 2007). Smad6 is also dimethylated at arginine 74 by the protein arginine N- methyltransferase (Inamitsu et al. 2006). The functional relevance of Smad6 dimethylation is yet to be unveiled.

Concluding remarks With the growing body of different enzymes that work to regulate TGF- β/Smad pathways, it remains apparent that a lot of hard work remains before one can fully grasp the complex and sophisticated nature of post- translational modifications that control signal transduction. Timing, localization and orientation of known modifications and those that are yet to be revealed will be of utmost importance when trying to understand how, e.g., Smads can enter the nucleus, bind to DNA, regulate specific genes, and finally terminate transcription by recycling to the cytoplasm or by undergoing degradation. This sequence of events is most likely the result of well defined switches of post-translational modifications together with confined localization. The interplay between different modifications will require further attention in the future.

27 It is also interesting to experience how new functions can be assigned to already well studied molecules. For example, ubiquitination has been exam- ined for decades, but the multitude of ubiquitin regulation has not been understood until recently when mono-ubiquitination, multi-ubiquitination and various alternatives of polyubiquitination were described. This illustrates that cellular regulation can be very multi-faceted.

28 ADP-ribosylation

Poly(ADP-ribosyl)ation (PARylation) was uncovered in 1963 by Chambon and co-workers (Chambon et al. 1963). This negatively charged post- translational modification can be attached to protein substrates as either monounits or as linear or highly branched polymers of variable sizes (Hassa et al. 2006). Targeted amino acids include arginines, lysines, glutamic acids and aspartic acids. Linking negatively charged polymers onto proteins can affect structure, function and interactions with proteins or DNA. These properties have made PARylation a powerful and important modification found in organisms ranging from plants to mammals. Over the years it has been more and more evident that ADP-ribosylation is involved in several critical cellular processes, ranging from such diverse functions as DNA damage response and maintenance of genome stability to telomere dynamics, centro- somal function and transcriptional control. In recent years, inhibitors of ADP-ribosylation have emerged as potential anticancer drugs (Rouleau et al. 2010).

Poly (ADP-ribose) polymerase-1 Poly (ADP-ribose) polymerase-1 (PARP-1) is the best studied member of the PARP family of ADP-ribosylating enzymes (Ame et al. 2004; Kim et al. 2005; Schreiber et al. 2006). The C-terminal part of PARP-1 carries the “PARP signature” motif, a conserved catalytic domain which is found in all PARP family members. It provides PARylation capability which is the proc- essing of β-NAD (nicotinamide adenine dinucleotide) molecules to ADP- ribose which then is attached onto substrate proteins (Figure 5). PARP-1 is mainly known to modify target glutamic and aspartic acids, however, PARy- lation of lysines has recently been documented (Messner et al. 2010). Human PARP-1 is 1014 amino acids long and has an N-terminal part involved in DNA binding, and a C-terminal part with enzymatic activity (Fig- ure 6) (D'Amours et al. 1999 ; Ame et al. 2004). In detail, the protein has three nucleotide-interacting Zn-fingers localized in the first part of the polypeptide. Between Zn-finger number two and three is a nuclear localization signal (NLS). The third Zn-finger was just recently described and is suggested to link DNA binding to coordination of PARP-1 enzymatic activity (Langelier et al. 2008). The central part of the protein contains an automodi-

29 fication domain with a BRCA1-C-terminal (BRCT) motif. The C-terminal, as mentioned earlier, includes the enzymatic PARP signature domain (D'Amours et al. 1999 ; Kim et al. 2005).

Figure 5. Simplified illustration of PARP-1 mediated ADP-ribosylation. PARP-1 uses β-NAD (NAD+) to mediate poly(ADP-ribosyl)ation of a glutamic acid residue on a substrate protein. (Glu, Glutamic acid; NAM, Nicotinamide; ADP-R, ADP- ribose)

Figure 6. PARP-1 domain structure. (Zn, Zinc finger; NLS, nuclear localization signal; BRCT, BRCA1-C-terminal)

30 PARP-1 is known to modify itself in the automodification domain and in the DNA-binding domain. In addition, several other proteins, including histones, p53, Sp1, YY1 and numerous transcription factors, has been identified as substrates of PARP-1 (Ogata et al. 1981; Huletsky et al. 1989; Malanga et al. 1998; D'Amours et al. 1999 ; Mendoza-Alvarez and Alvarez-Gonzalez 2001; Oei and Shi 2001; Kraus and Lis 2003; Kim et al. 2005; Zaniolo et al. 2007; Kraus 2008). The DNA binding domain of PARP-1 recognizes a number of different DNA formations (D'Amours et al. 1999 ; Petrucco and Percudani 2008). Examples are single-strand and double-strand DNA breaks, cruciform and supercoiled DNA structures, and some sequence-specific sites located in gene regulatory regions. Normally, the activity of PARP-1 is well controlled and generally low. But several activators, such as the DNA structures mentioned above and protein interactors like CTCF (CCCTC-binding factor) or phosphorylated ERK2 (Cohen-Armon et al. 2007; Guastafierro et al. 2008), can trigger PARP-1 activity. PARP-1 is vital in DNA damage detection processes and participates in several different DNA repair pathways, including single-strand break repair, double strand break repair and base excision repair pathways (D'Amours et al. 1999 ; Ame et al. 2004). At DNA damaged sites, PARP-1 is activated and recruits several important components of the DNA repair system, such as the DNA-dependent protein kinase to double strand breaks and XRCC-1 to base excision repair sites (Masson et al. 1998; Ruscetti et al. 1998; Kim et al. 2005). But if the DNA damage is too severe, PARP-1 seems to promote cell death pathways, leading to apoptotis or necrosis. More recently, PARP-1 has also attracted attention for its role in regulating transcription (Kraus and Lis 2003; Kim et al. 2005; Schreiber et al. 2006; Kraus 2008). PARP-1 can regulate transcription in enzymatic-dependent and -independent ways and can act as enhancer or repressor of gene expression. Models that have been provided for the PARP-1 action in transcription inlude: regulation of chromatin organisation and involvement in gene- specific enhancer/promoter binding (Kim et al. 2005). The negative charged PARylation chains, added to substrate proteins, is thought to affect histones and chromatin structure as well as different transcriptional components, altering their activity or DNA-binding capability (Kim et al. 2005). Nucleo- tide-binding proteins usually utilize electrostatic interaction when connecting to DNA. Cationic amino acids contact the negatively charged phosphodiester backbone. Therefore, addition of negative charges, such as ADP-ribose, to a DNA-binding protein could affect such electrostatic interactions in various ways. For example, PARP-1-mediated ADP-ribosylation of transcription factors such as YY1, Sp1 and p53 is thought to restrict their DNA-binding activities (Malanga et al. 1998; Mendoza-Alvarez and Alvarez-Gonzalez 2001; Oei and Shi 2001; Kraus and Lis 2003; Zaniolo et al. 2007; Kraus 2008).

31 Links between PARP-1 and TGF-β family mediated transcription are few. Work by Ku et al. showed a connection between OAZ and PARP-1 (Ku et al. 2003). OAZ is a zinc-finger transcription factor known to be involved in BMP-regulated transcription. PARP-1 interacts with OAZ and works as a co-activator (Ku et al. 2003). In addition, PARP-1 has been found to bind a sequence-specific site within the connective tissue growth factor (CCN2/CTGF) promoter (Okada et al. 2008). This promoter responds to TGF-β stimulation. Removal of the PARP-1 binding sequence led to attenuated transcription both in general and under stimulatory conditions. The authors concluded that PARP-1 binding within this promoter works as a non- specific, fundamental enhancer rather than a modulator of specific transcription factors. However, they acknowledge that a possible link between PARP- 1 and Smads could be interesting to examine. In a similar way, the TβRII promoter also contain a PARP-1 DNA binding site which positively regulates the transcription of this gene in estrogen receptor-positive breast cancer cell lines (Sterling et al. 2006). Furthermore, blocking all ADP-ribosylation activities in bronchial epithelial cells, by using the PARylation inhibitor 3- aminobenzamide (3AB), reduced TGF-β mediated fibronectin expression (Beckmann et al. 1992). However, interpretations of such experiment in relation to PARP-1 and Smads have to be done with caution since it most likely involve effects attributed to the loss of function of all PARylation capable enzymes in the human genome, transcribed from a total of 22 genes (Hottiger et al. 2010). Poly(ADP-ribose) glycohydrolase (PARG) reverses PARylation (Figure 7) (Davidovic et al. 2001), by removing ADP-ribosyl units from substrate proteins through hydrolyzing ribosyl-ribose bonds. In the context of transcription, PARG has been reported to both cooperate with PARP-1 and antagonize its function (Frizzell et al. 2009). In summary, PARP-1 affects transcription by acting as chromatin modulator, corepressor, coactivator or sequence specific regulator. The combined effects of these assignments are most likely gene specific and determined by several different conditions such as where the promoter localizes and what factors bind to it.

Figure 7. Illustration of PARP-1 mediated PARylation and PARG mediated de- PARylation.

33 Salt-inducible kinase-1

Salt-inducible kinase-1 (SIK) received its name when it was observed to be highly upregulated in adrenal glands of rats that were fed on a high-salt diet (Wang et al. 1999). SIK has also been described as Sucrose non-fermented 1 like-kinase (SNF1LK), a name given due to its similarity with the yeast protein, SNF1, when found in a screen for protein kinases expressed during mouse heart development (Ruiz et al. 1994). Furthermore, SIK is induced during PC12 cell depolarization and has therefore also received the name Kinase induced by depolarization-2 (KID-2) in this studie (Feldman et al. 2000). SIK is one of the thirteen members of the AMP-activated protein kinase (AMPK) family of Ser/Thr kinases. The members of this family share high sequence similarities but display diverse functions, such as regulation of cell polarity, cellular transport, microtubules or stress responses (Drewes et al. 1998; Lefebvre and Rosen 2005). The AMPKs are regulated by the tumor suppressor LKB1 (Lizcano et al. 2004). LKB1 can activate the AMPK family members through phosphorylating residues in their activation loop (T- loop). SIK is a 783 amino acid protein. The protein possesses a serine/threonine kinase domain at the N-terminal (Figure 8). This domain is flanked by a ubiquitin associated (UBA) domain. The UBA domain is one out of at least sixteen known ubiquitin binding domains. It is 40 amino acids long and has proven to be able to regulate several processes through interacting with ubiquitinated proteins and altering their function or localization (Buchberger 2002). Serine 577 of SIK is a PKA-phosphorylation site which is important for SIK nuclear/cytoplasmic shuttling (Katoh et al. 2002). Mutating this residue localized SIK to the nucleus. Katoh and colleagues concluded that a nuclear localization signal must reside close to Serine 577 and that phosphorylation of this site allows SIK to translocate to the cytoplasm. SIK is known to phosphorylate and regulate the Transducer of regulated CREB activity (TORC) and class II histone deacetylases (HDAC) (Lin et al. 2001; Doi et al. 2002; Katoh et al. 2006; Berdeaux et al. 2007; van der Lin- den et al. 2007). SIK-mediated phosphorylation of TORC induces its nuclear export and relieves it from its function as a CREB co-activator. SIK has also been linked to endocytosis by a high-throughput RNA interference screen for kinases (Pelkmans et al. 2005). Annotated as SNF1LK in the screen, SIK

34 depletion altered morphology and distribution of caveolin-1 positive structures. Kin-29, the SIK homolog in Caenorhabditis elegans, is important for TGF-β signaling since it was shown to regulate the Sma pathway (Maduzia et al. 2005). When Maduzia and co-workers mutated Kin-29 it interestingly resulted in smaller body size of the nematodes. However, expression of Kin- 29 was also capable of inhibiting the phenotype of longer worms that are formed by overexpressing a C. elegans TGF-β-like ligand, dbl-1. These ob- servations support the possibility that there is interplay between TGF-β signaling pathways and Kin-29/SIK. The upstream kinase of SIK, LKB1, has been reported to be under the control of TAK1 (Xie et al. 2006). Thus, TGF-β, via the non-Smad signaling pathways that involves TAK1, could be an upstream regulator of LKB1, this hypothesis is yet to be directly tested.

Figure 8. SIK domain structure. (UBA, ubiquitin associated domain; NLS, nuclear localization signal).

35 Present Investigation

Aim The aim of the present investigation was to; screen for novel protein regulators of the TGF-β/Smad pathway, investigate the effect of such regulators on signaling outcome and finally examine their mechanism of action. A particu- lar aim was set at identifying Smad-interacting enzymes, such as protein kinases and ubiquitin E3-ligases, which could regulate components of the signaling cascade by post-translationally modifying them.

Specific aims:

• Identify novel Smad4-interacting enzymes and determine their functions in regulation of TGF-β/Smad signaling.

• Examine the role of ADP-ribosylation and de-ADP-ribosylation within the Smad signaling pathway.

• Characterize the functional role of the Smad4-dependent TGF-β-induced SIK protein.

• Investigate novel aspects of post-translational regulation within TGF- β signaling, such as ubiquitination and phosphorylation.

Paper I

PARP-1 attenuates Smad-mediated transcription How Smads regulate their binding to DNA is an exciting and still not completely understood process. In Paper I we have uncovered a novel mechanism that explains how DNA-binding of Smad proteins is affected by ADP- ribosylation. We present evidence that Smads, upon TGF-β stimulation, enter the nucleus and interact with PARP-1. We also confirmed the direct Smad to PARP-1 binding in vitro and provide a detailed mapping of the involved interaction domains. In addition, our data established that Smad3 and Smad4 can be ADP-ribosylated at glutamic acid residues situated in

36 their N-terminal MH1 domains. ADP-ribosylation of Smad3 and Smad4 leads to their release from DNA. We hypothesize that the negative charges from added ADP-ribose moieties either repel Smads from negatively charged DNA, or alternatively change the Smad structure to interfere with DNA-binding. Knocking-down PARP-1 leads to enhanced TGF-β/Smad- mediated transcription in human keratinocyte, HaCaT, cells and in normal murine mammary gland epithelial, NMuMG, cells and results in a more rapid and pronounced TGF-β-induced epithelial-to-mesenchymal transition in the NMuMG cells. In conclusion, we have identified a new Smad-interacting enzyme, PARP- 1, that negatively regulates Smad-mediated transcription through ADP- ribosylation. This paper provides new interesting insights to how Smads are modulated at sites of transcription. It opens up intriguing ideas about the interplay between different post-translational modifications, such as phosphorylation, ubiquitination, acetylation and ADP-ribosylation.

Figure 9. Model of PARP-1 action during TGFβ signaling. PARP-1 (P1) binds to TGF-β-induced Smad3/Smad4 (S3, S4) complexes. The Smads binds to DNA together with additional transcription factors (TF) and gene transcription starts. PARyla- tion of Smad3/Smad4 by PARP-1 leads to dissociation (double arrow) of the Smad3/Smad4/PARP-1 complex from DNA, while gene transcription decreases (dotted arrow). The fate of the interacting transcription factors remains unclear (question mark). Effects of PARP-1 on chromatin regulation and transcription factor regulation are shown in the inset. Red circles show Smad3 C-terminal phosphorylation and black squares represent PAR chains.

37 Paper II

TGFβ-mediated transcription is regulated by PARP-1/PARG-balanced ADP-ribosylation In Paper II we report a new series of findings which suggest that PARP-1- mediated ADP-ribosylation of Smad proteins can be de-ADP-ribosylated by the glycohydrolase PARG. PARG is one of few proteins known to counte- ract cellular ADP-ribosylation and has interestingly been shown capable to either cooperate with or antagonize PARP-1-regulated transcription (Frizzell et al. 2009). We showed, in vitro, that ADP-ribosylated Smads and auto- ADP-ribosylated PARP-1 can be de-ADP-ribosylated by incubation with recombinant PARG protein. We also provide evidence that modulation of PARP-1 or PARG expression levels have an impact on TGF-β-mediated gene transcription. In agreement with our earlier studies (Paper I), we found that TGF-β-induced PAI-1 and Fibronectin transcription is upregulated when PARP-1 is depleted by siRNA. However, we now also report, in Ha- CaT cells, that these two genes are downregulated after PARG siRNA knock-down. We hypothesize that silencing of PARG leads to enhanced cellular PARylation and that the balance in ADP-ribosylation is important during TGF-β signaling. We further demonstrate that the downregulation of TGF-β-mediated transcription, observed when depleting PARG, can to a large extent be rescued by also knocking-down PARP-1. This implies that the effect PARG has on TGF-β signaling is at least in part dependent on PARP-1-mediated ADP-ribosylation. In summary, we have found that PARG can mediate de-ADP-ribosylation of Smads and that PARG is important for TGF-β signaling. More experiments are needed to reveal how the PARG enzymatic action functions to oppose PARP-1-mediated effects on TGF-β-induced transcription. However, the PARP-1/PARG balanced regulation of Smad-mediated transcription opens up new possibilities on the dynamics of post-translational modifications within the TGF-β signaling pathway.

Paper III

TGFβ induces SIK to negatively regulate type I receptor kinase signaling Salt inducible kinase 1 (SIK) was identified in a screen for novel gene regulatory targets of TGF-β signaling that are dependent on Smad4 (Kowanetz et al. 2004). SIK belongs to the AMPK-family of signaling kinases and contain both a serine/threonine kinase domain and a UBA domain. We found that SIK is rapidly induced by TGF-β and that it plays an important role in regulating the stability of the TGF-β type I receptor, ALK5. Overexpression of

38 SIK downregulated the receptor potently, while depleting SIK led to a stabilization of the receptor. Both the kinase domain and the UBA domain of SIK are important during receptor turnover, which was demonstrated by overexpressing SIK deletion constructs. In addition, we found that the inhibitory Smad, Smad7, participates in the complex together with SIK and ALK5. Smad7 is also induced after TGF-β stimulation and serves as a key player for receptor degradation. In SIK, we have therefore established a new protein kinase that is upregulated by TGF-β and has a negative feed-back function by downregulating cell-surface TGF-β receptor levels. Since SIK has both kinase activity and ubiquitin binding ability, it potentially provides a link between ubiquitinating and phosphorylating processes during signal transduction.

Paper IV

SIK and Smurf2 cooperate to downregulate the TGF-β type I receptor The data in this paper add extra insights into the previously reported mechanism on Salt inducible kinase 1 (SIK)-mediated downregulation of the TGF-β type I receptor, ALK5 (Paper III). First, we highlighted and com- pared SIK and Smad7 mRNA up-regulation after TGF-β stimulation to better understand their position and organization in the negative feed-back loop of this pathway. Both genes are rapidly upregulated by TGF-β, but while Smad7 mRNA levels peaks and then falls back to a lower plateau, SIK1 mRNA levels stay at its highest induced level for all the time-points meas- ured. We found that both genes are direct targets of TGF-β signaling and that their inductions are dependent on Smad2, 3 and Smad4. We then also demonstrated the importance of the E3 ubiquitin ligase, Smurf2, in SIK promoted TGF-β type I receptor turnover. Smurf2 is upregulated by TGF-β and is known to interact with Smad7 to ubiquitinate and degrade ALK5 (Lonn et al. 2009). We further showed that SIK forms a functional complex with Smurf2 via Smad7 and/or ALK5. In addition, we revealed that the Smurf2 E3 ubiquitin-ligase activity is required for SIK to promote receptor turnover and that the kinase activity of SIK is crucial for Smurf2-mediated receptor degradation. Finally, we provide evidence that Smad7 can be phosphorylated by SIK in vitro. However, detailed phosphorylation mapping and phosphorylation experiments in vivo are needed to firmly establish the specific role of such modification during TGF-β signaling.

39 Future perspectives

Intracellular signaling pathways are initiated, terminated or fine-tuned by post-translational regulation of components in signaling cascades. Different post-translational modifications co-operate or compete with each other, making the alternatives of regulation very broad and complex. Numerous modifications have been described within the TGF-β/Smad-signaling routes, ranging from phosphorylation and methylation to ubiquitination and sumoylation. In addition, the fact that post-translational regulations can be cell-type specific add to the already complex picture. It is thus highly likely that there are yet unidentified post-translational modifications as well as new functions of known modifications within the pathways. Furthermore, the timing and possible co-operation between various modifications reported is an equally important task remaining to be inspected and fully understood. Which modifications happen where and when and which modifications are strong pre- requisites for additional modifications to take place?

Paper I and II ADP-ribosylation of Smad proteins opens up new ideas on how the function of nuclear Smads can be regulated. Additional studies will be needed, on both molecular and cellular levels, to fully understand the effects of Smad PARylation. Activation of PARP-1 during TGF-β signaling is a topic that needs further attention. Only a few proteins, like CTCF and phosphorylated ERK2 (Cohen-Armon et al. 2007; Guastafierro et al. 2008), have been shown to affect PARP-1 activity per se. Different conformations of DNA can also activate PARP-1 (D'Amours et al. 1999 ; Hassa and Hottiger 2008). We have shown that Smads interact with PARP-1 in the absence of DNA. However, if PARylation by PARP-1 can be triggered by Smad interactions or if alternative pathways via DNA or other protein partners are necessary during Smad signaling is still unsolved and requires a closer look. Another interesting open question is whether PARylation of Smads is an isolated event or if it acts in concert with other post-translational modifications known to initiate or terminate Smad-mediated transcription. As mentioned before, the molecular changes governing the start and stop mechanisms of Smad-promoted gene regulation remain an important step to deci- pher. We report that ADP-ribosylation of Smads negatively regulates Smad- activated transcription by releasing them from DNA (Paper I). This event

40 could work together with dephosphorylation or ubiquitin-mediated degradation, both being linked to signaling termination (Gao et al. 2009; Lonn et al. 2009; Wrighton et al. 2009). If so, timing and order of such orchestrated post-translational modifications would greatly contribute to the understand- ing of the various steps that Smads undertake during their nuclear residence. Our data so far suggest that PARylation is more efficient on R-Smads with C-terminal phosphorylation and that Smads in complex interact better with PARP-1 than single Smads (Paper I). However, it is still possible that these modifications are isolated events, working as backups for each other to make sure that signaling termination can be executed even if one system would fail. The answer to such questions could be resolved by combinations of protein knock-downs, different Smad point-mutants and/or the use of modification-specific antibodies. In addition to the Smads, several other DNA-binding proteins are also regulated by PARP-1-mediated PARylation, including Sp1, p53 and YY1 (Malanga et al. 1998; Mendoza-Alvarez and Alvarez-Gonzalez 2001; Oei and Shi 2001; Kraus and Lis 2003; Zaniolo et al. 2007; Kraus 2008). It has been proposed that the negative charge, from the addition of ADP-ribose moieties, might repel proteins from negatively charged DNA. However, PARylation of the Smad MH1 domain could also alter the structure to disfa- vor DNA binding. It would be interesting to visualize 3D-structures of non- PARylated versus PARylated Smads in order to observe possible changes in position of the β-hairpin. We have pointed out potential sites of PARylation in the MH1 domain of Smad3 (Paper I). Interestingly, they are positioned close to the preceding and initial part of the β-hairpin. In addition, the fact that a positively charged zinc ion is necessary to stabilize the MH1 domain implies that the domain structure could be highly sensitive to shifts in charges by PARylation. Since PARP-1 is able to relieve the binding of Smads to SBEs at gene regulatory regions, the prediction that PARP-1 is a negative regulator of Smad-activated transcription is logical. However, PARP-1 regulates a vast amount of cellular processes, including some important for gene regulation. PARP-1 is for instance recognized as a modifier of chromatin by PARylating chromatin regulators, such as histones and CTCF (Kraus 2008). Smads bound to PARP-1 may therefore also contribute to chromatin modulation in areas of Smad-activated transcription, something which has not yet been studied. Such regulation could contribute positively or negatively to the effects PARP-1 has on Smad DNA-binding. In addition, PARP-1 is known to modulate the function of several transcription factors, both in PARylation- dependent and non-PARylation-dependent ways, and can also bind to site- specific DNA sequences on gene promoters (Kraus 2008). Since transcription can be affected by both chromatin state and by the concerted actions of several transcription factors and co-regulators, gene to gene regulation va- riances by PARP-1 is expected, both in general and in respect to TGF-β sig-

41 naling. Such differences have already been reported by us and others (Paper I) (Frizzell et al. 2009). Furthermore, Sp1, p53 and YY1 are all known PA- Rylation targets of PARP-1 that also interact directly with the Smads (Ma- langa et al. 1998; Mendoza-Alvarez and Alvarez-Gonzalez 2001; Oei and Shi 2001; Kraus and Lis 2003; Zaniolo et al. 2007; Kraus 2008; Ross and Hill 2008). Genome-wide gene expression arrays under signaling conditions are therefore highly warranted in order to determine patterns of regulation of PARP-1 in TGF-β-induced transcription. The PARP family comprises 18 members (Ame et al. 2004). All of them have a highly similar PARP domain structure, capable of PARylating substrate proteins. We have found interactions between PARP-1 and Smads. However, it is still unknown whether any of the other PARP family members also interact with Smads. In addition, some of them have been reported to influence the activity and function of PARP-1. It would thus be worth ex- amining possible co-operation or competitions between different PARPs and examine their effect on TGF-β signaling. Furthermore, since the R-Smads are rather well conserved, possible events of PARylation of BMP-specific Smads and the functional outcome of such regulations is also worth pur- suing. In Paper II we report the importance of the glycohydrolase PARG in regulation of TGF-β signaling. PARG can de-PARylate proteins (Davidovic et al. 2001). We would like to test if PARG acts before Smad DNA-binding, keeping PARylation low to allow the interaction between Smads and nucleo- tides, or after, to balance PARP-1 mediated PARylation of Smads at active promoters and thus sustaining transcription. An additional interesting aspect of PARP-1/PARG regulation is the availability of β-NAD. Intracellular shifts in β-NAD concentrations could potentially control the enzymatic reactions. It is still also possible that PARG acts via yet undefined pathways which lead to enhanced TGF-β signaling.

Paper III and IV Paper III and IV place Salt inducible kinase 1 (SIK) into the negative feed- back loop of TGF-β signaling, together with Smad7 and Smurf2. This complex of proteins regulates ALK5 receptor turnover. We have shown that Smad7 is a substrate of SIK, but specific phosphorylation sites are lacking (Paper IV). Mapping of such sites would be important in order to allow generation of phospho-specific antibodies which would make mechanistical surveys feasible. A screen for additional substrates of SIK is also highly warranted. It is still not clear which stage of receptor downregulation SIK actually affects. Ubiquitin binding domains have been associated with receptor internalization, as well as binding and relocalization of ubiquitinated proteins (Buchberger 2002; Haglund and Dikic 2005). This would implicate that SIK, via its UBA domain, could bind to ubiquitinated receptors and aid important

42 steps of endocytosis, followed by sorting to sites of degradation. This notion is supported by the fact that overexpression of SIK led to lowered amounts of receptors at cell-surfaces, while knock-down of SIK enhanced cell surface receptor levels (Paper III). On the other hand, Smad7 and Smurf2 are closely linked with direct receptor ubiquitination which means that SIK could serve to enhance this modification. However, we have not been able to see enhanced effects on receptor ubiquitination by overexpressing SIK (Lönn, et al. unpublished data). The route of receptors for degradation is debated (Lonn et al. 2009). On one hand it is widely accepted that lysosomes mediate receptor turnover. On the other hand ubiquitination, E3 ligases and proteasomal inhibitors have been reported to be important for proper receptor downregulation. A reason- able conclusion is that both organelles could participate. Perhaps ubiquitinated receptors need to pass proteasomal structures and surrounding factors en route to lysosomes. Again, specific antibodies, inhibitors and highly sensitive microscopy techniques could be used to study these pathways and to determine if SIK influences sorting of ubiquitinated receptors to any of these compartments. Other components of the TGF-β/Smad signaling network, such as Smad4, are also targeted by I-Smad/Smurf-mediated degradation (Lonn et al. 2009). It would thus be interesting to investigate if SIK can play roles in these complexes as well, or if SIK is restrained to receptor endocytosis and turnover. Finally, the induction of SIK by TGF-β in HaCaT cells is rapid and mRNA levels remain for several hours (Paper III and IV). The sustained levels of SIK could imply that it functions in both early and later stages of signaling. For example, SIK could serve the purpose of receptor downregulation with Smad7 and Smurf at early time-points while it might fulfill other tasks later on when in complex with other factors. SIK mRNA is also induced by BMP7 (Paper III). The functional role of SIK during BMP signaling is would be interesting to explore further.

43 Acknowledgments

The work presented in this thesis was carried out at the Ludwig Institute for Cancer Research (LICR), Uppsala University. I am very grateful for hav- ing had the opportunity to grow as a scientist in the creative atmosphere that the LICR and the TGF-β signaling group offer. Over the years I have received great support from colleagues, family and friends. It would not have been possible for me to fulfill all this work without your contributions and encouragements. I will always be grateful to eve- ryone that has helped me and I would especially like to extend my sincere gratitude to the following people:

My supervisor, Dr. Aristidis Moustakas. You have a wonderful capability of encouraging people and helping them develop into good scientists. Your knowledge on experimental and scientific details is impressive. I still remember the first time I stepped into your office. After discussing Smad4 monoubiquitination for half an hour I was completely captivated by post- translational regulation of TGF-β signaling. From that day on, our meetings always inspire me to do new experiments. I thank you for constantly taking time to discuss and answer questions and for encouraging me to develop and try my own ideas.

Prof. Carl-Henrik Heldin. Thank you for giving me the opportunity to work at the LICR and for all the support and help you have offered me. Your passion for science and signal transduction seems endless. You are an inspi- ration to all aspiring young scientists.

All the present and past members of the TGF-β signaling group. First of all, Anita Morén, for guiding me when I first joined the TS-group as an under- graduate student, and for the continuous help throughout the years. All my in-group collaborators Markus Dahl, Michael Vanlandewijck, Marcin & Katarzyna Kowanetz and Lars van der Heide, thank you all, I think we made an excellent team. I will always remember all the help I have received from the whole lab; Erna Raja, E-Jean Tan, Kaoru Kahata, Stefan Termén, Katia Savary, Sylvie Thuault, Rosita Bergström, Hideki Niimi, Katerina Pardali and Ulrich Valcourt. Thank you all for the great atmosphere in the lab.

44 All LICR group-leaders, for creating a good environment to discuss and conduct science. Especially Prof. Ulf Hellman (without you my mass-spec projects would not have taken place and my bike would probably have been much slower), Prof. Johan Ericsson (who was always there to give experimental advices, day and night) and Prof. Maréne Landström (for valuable discussions). The LICR staff; Aino, Lotti, Aive, Eva H, Ingegärd, Lasse, Uffe, Ulla, Gullbritt and Christer for helping me in all possible ways. And finally all the PhD students and Post-docs that I have had the pleasure to meet and work with.

During my time at the Ludwig Institute I have had the fortune to get to know some of you better. Nimesh, I want to thank you for welcoming me into the lab when I arrived. You helped me a lot during my first years as a PhD student and I really enjoyed our discussions. We miss your warm smile (and your moustache) here at Ludwig. Michael, thank you for all the fun in the lab and for our collaborations. I think we developed the perfect lab-bench symbiosis, were we always managed to perfectly complement our buffer and glove need. Markus, collaborating with you has been great and I am en- dlessly thankful for your contribution. In addition, thanks to you I have also gained the utmost respect for Timrå Red Eagles and “Torsten”. Piotr, I have had many interesting discussions with you and also some of the best laughs. Good luck with your new job. I am looking forward to visit Wisla stadium. Aga, I am truly impressed by you. You always have new business ideas, excellent scientific knowledge and a great sense of humor. Spyridon, I miss our football discussions, training sessions and experiment updates. Let’s go for a Champions Leauge match again someday. Lars, you always encourag- es people to work hard and stay positive. Thank you for the impact you have had on my projects.

In addition, to the numerous people that I have crossed paths with and that inspired, enlightened and helped me. You all deserve a special mentioning; Maite, Eva G, Maria W, Marta, Ihor, Mariya, Ria, Anders, Vasyl, Anna and many more. My class-mates from Uppsala University: Tomas and Micke, for all the good times we had in and between classes. And to Stefan, Andreas and Mattias, for being the greatest friends and for reminding me of the world outside the eppendorf tubes.

Finally, to my wonderful family, my parents Caroline and Nils-Göran Lönn and my sister Jessica. Without your unconditional love and support I would never have been able to accomplish this. You are the foundation that made me reach this far. Ellen, I feel at best when I am with you. You en- courage me more than you can imagine. Thank you for always listening, for helping me and for constantly bringing new perspectives. I´ll fly with you any day.

45 References

Abdollah, S., M. Macias-Silva, T. Tsukazaki, H. Hayashi, L. Attisano and J. L. Wrana (1997). "TbetaRI phosphorylation of Smad2 on Ser465 and Ser467 is required for Smad2-Smad4 complex formation and signaling." J Biol Chem 272(44): 27678-85. Alarcon, C., A. I. Zaromytidou, Q. Xi, S. Gao, J. Yu, S. Fujisawa, A. Barlas, A. N. Miller, K. Manova-Todorova, M. J. Macias, G. Sapkota, D. Pan and J. Massague (2009). "Nuclear CDKs drive Smad transcriptional activation and turnover in BMP and TGF-beta pathways." Cell 139(4): 757-69. Ame, J. C., C. Spenlehauer and G. de Murcia (2004). "The PARP superfamily." BioEssays 26(8): 882-93. Annes, J. P., J. S. Munger and D. B. Rifkin (2003). "Making sense of latent TGFbeta activation." J Cell Sci 116(2): 217-224. Bai, Y., C. Yang, K. Hu, C. Elly and Y. C. Liu (2004). "Itch E3 ligase-mediated regulation of TGF-beta signaling by modulating smad2 phosphorylation." Mol Cell 15(5): 825-31. Batut, J., B. Schmierer, J. Cao, L. A. Raftery, C. S. Hill and M. Howell (2008). "Two highly related regulatory subunits of PP2A exert opposite effects on TGF-beta/Activin/Nodal signalling." Development 135(17): 2927-37. Beckmann, J. D., M. Illig, D. Romberger and S. I. Rennard (1992). "Induction of fibronectin gene expression by transforming growth factor beta-1 is attenuated in bronchial epithelial cells by ADP-ribosyltransferase inhibitors." J Cell Physiol 152(2): 274-80. Bengtsson, L., R. Schwappacher, M. Roth, J. H. Boergermann, S. Hassel and P. Knaus (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(Pt 8): 1248-57. Bennett, D. and L. Alphey (2002). "PP1 binds Sara and negatively regulates Dpp signaling in Drosophila melanogaster." Nat Genet 31(4): 419-23. Berdeaux, R., N. Goebel, L. Banaszynski, H. Takemori, T. Wandless, G. D. Shelton and M. Montminy (2007). "SIK1 is a class II HDAC kinase that promotes survival of skeletal myocytes." Nat Med 13(5): 597-603. Bhowmick, N. A., M. Ghiassi, M. Aakre, K. Brown, V. Singh and H. L. Moses (2003). "TGF-β-induced RhoA and p160ROCK activation is involved in the inhibition of Cdc25A with resultant cell-cycle arrest." Proc. Natl. Acad. Sci. U S A 100(26): 15548-15553. Bocharov, E. V., D. M. Korzhnev, M. J. Blommers, T. Arvinte, V. Y. Orekhov, M. Billeter and A. S. Arseniev (2002). "Dynamics-modulated biological activity of transforming growth factor beta3." J Biol Chem 277(48): 46273-9.

46 Buchberger, A. (2002). "From UBA to UBX: new words in the ubiquitin vocabu- lary." Trends in Cell Biology 12(5): 216-221. Chacko, B. M., B. Qin, J. J. Correia, S. S. Lam, M. P. de Caestecker and K. Lin (2001). "The L3 loop and C-terminal phosphorylation jointly define Smad protein trimerization." Nat Struct Biol 8(3): 248-53. Chai, J., J. W. Wu, N. Yan, J. Massague, N. P. Pavletich and Y. Shi (2003). "Fea- tures of a Smad3 MH1-DNA complex. Roles of water and zinc in DNA binding." J Biol Chem 278(22): 20327-31. Chambon, P., J. WEILL and P. MANDEL (1963). "Nicotinamide mononucleotide activation of new DNA-dependent polyadenylic acid synthesizing nuclear enzyme." Biochem Biophys Res Commun(11): 39-43. Chen, H. B., J. Shen, Y. T. Ip and L. Xu (2006). "Identification of phosphatases for Smad in the BMP/DPP pathway." Genes Dev. 20(6): 648-653. Chen, Y. G. (2009). "Endocytic regulation of TGF-beta signaling." Cell Res 19(1): 58-70. Chen, Y. G., A. Hata, R. S. Lo, D. Wotton, Y. Shi, N. Pavletich and J. Massague (1998). "Determinants of specificity in TGF-beta signal transduction." Genes Dev 12(14): 2144-52. Cohen-Armon, M., L. Visochek, D. Rozensal, A. Kalal, I. Geistrikh, R. Klein, S. Bendetz-Nezer, Z. Yao and R. Seger (2007). "DNA-independent PARP-1 activation by phosphorylated ERK2 increases Elk1 activity: a link to histone acetylation." Mol Cell 25(2): 297-308. D'Amours, D., S. Desnoyers, I. D'Silva and G. G. Poirier (1999 ). "Poly(ADP- ribosyl)ation reactions in the regulation of nuclear functions." Biochem. J. 342(2): 249-268. Daopin, S., M. Li and D. R. Davies (1993). "Crystal structure of TGF-beta 2 refined at 1.8 A resolution." Proteins 17(2): 176-92. Daopin, S., K. A. Piez, Y. Ogawa and D. R. Davies (1992). "Crystal structure of transforming growth factor-beta 2: an unusual fold for the superfamily." Science 257(5068): 369-73. Datto, M. B., Y. Li, J. F. Panus, D. J. Howe, Y. Xiong and X. F. Wang (1995). "Transforming growth factor beta induces the cyclin-dependent kinase inhibitor p21 through a p53-independent mechanism." Proc Natl Acad Sci U S A 92(12): 5545-9. Davidovic, L., M. Vodenicharov, E. B. Affar and G. G. Poirier (2001). "Importance of poly(ADP-ribose) glycohydrolase in the control of poly(ADP-ribose) metabolism." Exp Cell Res 268(1): 7-13. Dennler, S., S. Huet and J. M. Gauthier (1999). "A short amino-acid sequence in MH1 domain is responsible for functional differences between Smad2 and Smad3." Oncogene 18(8): 1643-8. Dennler, S., S. Itoh, D. Vivien, P. ten Dijke, S. Huet and J.-M. Gauthier (1998). "Direct binding of Smad3 and Smad4 to critical TGF β-inducible elements in the promoter of human plasminogen activator inhibitor-type 1 gene." EMBO J. 17(11): 3091-3100. Derynck, R., W. M. Gelbart, R. M. Harland, C.-H. Heldin, S. E. Kern, J. Massagué, D. A. Melton, M. Mlodzik, R. W. Padgett, A. B. Roberts, J. Smith, G. H.

47 Thomsen, B. Vogelstein and X.-F. Wang (1996). "Nomenclature: verte- brate mediators of TGFβ family signals." Cell 87(2): 173. Derynck, R. and K. Miyazono (2008). "The TGF-β family." Cold spring harbor laboratory press. Derynck, R. and Y. E. Zhang (2003). "Smad-dependent and Smad-independent pathways in TGF-β family signalling." Nature 425(6958): 577-584. Doi, J., H. Takemori, X.-z. Lin, N. Horike, Y. Katoh and M. Okamoto (2002). "Salt- inducible Kinase Represses cAMP-dependent Protein Kinase-mediated Ac- tivation of Human Cholesterol Side Chain Cleavage Cytochrome P450 Promoter through the CREB Basic Leucine Zipper Domain." J. Biol. Chem. 277(18): 15629-15637. Drewes, G., A. Ebneth and E.-M. Mandelkow (1998). "MAPs, MARKs and micro- tubule dynamics." Trends in Biochemical Sciences 23(8): 307-311. Dupont, S., A. Mamidi, M. Cordenonsi, M. Montagner, L. Zacchigna, M. Adorno, G. Martello, M. J. Stinchfield, S. Soligo, L. Morsut, M. Inui, S. Moro, N. Modena, F. Argenton, S. J. Newfeld and S. Piccolo (2009). "FAM/USP9x, a deubiquitinating enzyme essential for TGFbeta signaling, controls Smad4 monoubiquitination." Cell 136(1): 123-35. Dupont, S., L. Zacchigna, M. Cordenonsi, S. Soligo, M. Adorno, M. Rugge and S. Piccolo (2005). "Germ-layer specification and control of cell growth by Ec- todermin, a Smad4 ubiquitin ligase." Cell 121(1): 87-99. Ebisawa, T., M. Fukuchi, G. Murakami, T. Chiba, K. Tanaka, T. Imamura and K. Miyazono (2001). "Smurf1 interacts with transforming growth factor-beta type I receptor through Smad7 and induces receptor degradation." J. Biol. Chem. 276(16): 12477-12480. Edlund, S., M. Landström, C.-H. Heldin and P. Aspenström (2002). "Transforming growth factor-β-induced mobilization of actin cytoskeleton requires signaling by small GTPases Cdc42 and RhoA." Mol. Biol. Cell 13(3): 902-914. Engel, M. E., M. A. McDonnell, B. K. Law and H. L. Moses (1999). "Interdepen- dent SMAD and JNK signaling in transforming growth factor- β-mediated transcription." J. Biol. Chem. 274(52): 37413-37420. Faresse, N., F. Colland, N. Ferrand, C. Prunier, M. F. Bourgeade and A. Atfi (2008). "Identification of PCTA, a TGIF antagonist that promotes PML function in TGF-beta signalling." Embo J 27(13): 1804-15. Feldman, J. D., L. Vician, M. Crispino, W. Hoe, M. Baudry and H. R. Herschman (2000). "The salt-inducible kinase, SIK, is induced by depolarization in brain." J. Neurochem. 74(6): 2227-2238. Feng, X.-H. and R. Derynck (2005). "Specificity and versatility in TGF-β signaling through Smads." Annu. Rev. Cell Dev. Biol. 21: 659-693. Frizzell, K. M., M. J. Gamble, J. G. Berrocal, T. Zhang, R. Krishnakumar, Y. Cen, A. A. Sauve and W. L. Kraus (2009). "Global analysis of transcriptional regulation by poly(ADP-ribose) polymerase-1 and poly(ADP-ribose) glycohydrolase in MCF-7 human breast cancer cells." J Biol Chem 284(49): 33926-38. Fukuchi, M., T. Imamura, T. Chiba, T. Ebisawa, M. Kawabata, K. Tanaka and K. Miyazono (2001). "Ligand-dependent degradation of Smad3 by a ubiquitin

48 ligase complex of ROC1 and associated proteins." Mol. Biol. Cell 12(5): 1431-1443. Funaba, M., C. M. Zimmerman and L. S. Mathews (2002). "Modulation of Smad2- mediated signaling by extracellular signal-regulated kinase." J Biol Chem 277(44): 41361-8. Gao, S., C. Alarcon, G. Sapkota, S. Rahman, P. Y. Chen, N. Goerner, M. J. Macias, H. Erdjument-Bromage, P. Tempst and J. Massague (2009). "Ubiquitin ligase Nedd4L targets activated Smad2/3 to limit TGF-beta signaling." Mol Cell 36(3): 457-68. Griswold-Prenner, I., C. Kamibayashi, E. M. Maruoka, M. C. Mumby and R. De- rynck (1998). "Physical and functional interactions between type I transforming growth factor beta receptors and Balpha, a WD-40 repeat subunit of phosphatase 2A." Mol Cell Biol 18(11): 6595-604. Groppe, J., C. S. Hinck, P. Samavarchi-Tehrani, C. Zubieta, J. P. Schuermann, A. B. Taylor, P. M. Schwarz, J. L. Wrana and A. P. Hinck (2008). "Cooperative assembly of TGF-beta superfamily signaling complexes is mediated by two disparate mechanisms and distinct modes of receptor binding." Mol Cell 29(2): 157-68. Grönroos, E., U. Hellman, C.-H. Heldin and J. Ericsson (2002). "Control of Smad7 stability by competition between acetylation and ubiquitination." Mol. Cell 10(3): 483-493. Guastafierro, T., B. Cecchinelli, M. Zampieri, A. Reale, G. Riggio, O. Sthandier, G. Zupi, L. Calabrese and P. Caiafa (2008). "CCCTC-binding factor activates PARP-1 affecting DNA methylation machinery." J Biol Chem 283(32): 21873-80. Haglund, K. and I. Dikic (2005). "Ubiquitylation and cell signaling." Embo J 24(19): 3353-9. Hannon, G. J. and D. Beach (1994). "p15INK4B is a potential effector of TGF-beta- induced cell cycle arrest." Nature 371(6494): 257-61. Hassa, P. O., S. S. Haenni, M. Elser and M. O. Hottiger (2006). "Nuclear ADP- ribosylation reactions in mammalian cells: where are we today and where are we going?" Microbiol Mol Biol Rev 70(3): 789-829. Hassa, P. O. and M. O. Hottiger (2008). "The diverse biological roles of mammalian PARPS, a small but powerful family of poly-ADP-ribose polymerases." Front Biosci 13: 3046-82. Hata, A., G. Lagna, J. Massagué and A. Hemmati-Brivanlou (1998). "Smad6 inhibits BMP/Smad1 signaling by specifically competing with the Smad4 tumor suppressor." Genes Dev. 12(2): 186-197. Hayashi, H., S. Abdollah, Y. Qiu, J. Cai, Y. Y. Xu, B. W. Grinnell, M. A. Richard- son, J. N. Topper, M. A. Gimbrone, Jr., J. L. Wrana and D. Falb (1997). "The MAD-related protein Smad7 associates with the TGFβ receptor and functions as an antagonist of TGFβ signaling." Cell 89(7): 1165-73. Heldin, C. H., M. Landstrom and A. Moustakas (2009). "Mechanism of TGF-beta signaling to growth arrest, apoptosis, and epithelial-mesenchymal transition." Curr Opin Cell Biol 21(2): 166-76.

49 Hottiger, M. O., P. O. Hassa, B. Luscher, H. Schuler and F. Koch-Nolte (2010). "Toward a unified nomenclature for mammalian ADP-ribosyltransferases." Trends Biochem Sci 35(4): 208-19. Huletsky, A., G. de Murcia, S. Muller, M. Hengartner, L. Menard, D. Lamarre and G. Poirier (1989). "The effect of poly(ADP-ribosyl)ation on native and H1- depleted chromatin. A role of poly(ADP-ribosyl)ation on core nucleosome structure." J. Biol. Chem. 264(15): 8878-8886. Huminiecki, L., L. Goldovsky, S. Freilich, A. Moustakas, C. Ouzounis and C. H. Heldin (2009). "Emergence, development and diversification of the TGF- beta signalling pathway within the animal kingdom." BMC Evol Biol 9: 28. Iavarone, A. and J. Massague (1997). "Repression of the CDK activator Cdc25A and cell-cycle arrest by cytokine TGF-beta in cells lacking the CDK inhibitor p15." Nature 387(6631): 417-22. Imamura, T., M. Takase, A. Nishihara, E. Oeda, J. Hanai, M. Kawabata and K. Miyazono (1997). "Smad6 inhibits signalling by the TGF-β superfamily." Nature 389(6651): 622-626. Imoto, S., N. Ohbayashi, O. Ikeda, S. Kamitani, R. Muromoto, Y. Sekine and T. Matsuda (2008). "Sumoylation of Smad3 stimulates its nuclear export during PIASy-mediated suppression of TGF-beta signaling." Biochem Biophys Res Commun 370(2): 359-65. Imoto, S., K. Sugiyama, R. Muromoto, N. Sato, T. Yamamoto and T. Matsuda (2003). "Regulation of transforming growth factor-beta signaling by protein inhibitor of activated STAT, PIASy through Smad3." J Biol Chem 278(36): 34253-8. Inamitsu, M., S. Itoh, U. Hellman, P. Ten Dijke and M. Kato (2006). "Methylation of Smad6 by protein arginine N-methyltransferase 1." FEBS Lett 580(28- 29): 6603-11. Inman, G. J., F. J. Nicolas and C. S. Hill (2002). "Nucleocytoplasmic shuttling of Smads 2, 3, and 4 permits sensing of TGF- β receptor activity." Mol. Cell 10(2): 283-294. Inoue, Y., Y. Itoh, K. Abe, T. Okamoto, H. Daitoku, A. Fukamizu, K. Onozaki and H. Hayashi (2007). "Smad3 is acetylated by p300/CBP to regulate its transactivation activity." Oncogene 26(4): 500-8. Ishida, W., T. Hamamoto, K. Kusanagi, K. Yagi, M. Kawabata, K. Takehara, T. K. Sampath, M. Kato and K. Miyazono (2000). "Smad6 is a Smad1/5-induced smad inhibitor. Characterization of bone morphogenetic protein-responsive element in the mouse Smad6 promoter." J Biol Chem 275(9): 6075-9. Itoh, F., N. Divecha, L. Brocks, L. Oomen, H. Janssen, J. Calafat, S. Itoh and P. Dijke Pt (2002). "The FYVE domain in Smad anchor for receptor activation (SARA) is sufficient for localization of SARA in early endosomes and regulates TGF-beta/Smad signalling." Genes Cells 7(3): 321-31. Itoh, S., M. Thorikay, M. Kowanetz, A. Moustakas, F. Itoh, C.-H. Heldin and P. ten Dijke (2003). "Elucidation of Smad requirement in transforming growth factor-β type I receptor-induced responses." J. Biol. Chem. 278(6): 3751- 3761.

50 Kang, J. S., E. F. Saunier, R. J. Akhurst and R. Derynck (2008). "The type I TGF- beta receptor is covalently modified and regulated by sumoylation." Nat Cell Biol 10(6): 654-64. Kang, Y., C. R. Chen and J. Massague (2003). "A self-enabling TGFbeta response coupled to stress signaling: Smad engages stress response factor ATF3 for Id1 repression in epithelial cells." Mol Cell 11(4): 915-26. Katoh, Y., H. Takemori, J. Doi and M. Okamoto (2002). "Identification of the nuclear localization domain of salt-inducible kinase." Endocr Res 28(4): 315-8. Katoh, Y., H. Takemori, X.-z. Lin, M. Tamura, M. Muraoka, T. Satoh, Y. Tsuchiya, L. Min, J. Doi, A. Miyauchi, L. A. Witters, H. Nakamura and M. Okamoto (2006). "Silencing the constitutive active transcription factor CREB by the LKB1-SIK signaling cascade." FEBS Journal 273(12): 2730-2748. Katri Koli, J. S., Marko Hyytiäinen, Carita Penttinen , Jorma Keski-Oja, (2001). "Latency, activation, and binding proteins of TGF-?" Microscopy Research and Technique 52(4): 354-362. Kavsak, P., R. K. Rasmussen, C. G. Causing, S. Bonni, H. Zhu, G. H. Thomsen and J. L. Wrana (2000). "Smad7 binds to Smurf2 to form an E3 ubiquitin ligase that targets the TGF β receptor for degradation." Mol. Cell 6(6): 1365-75. Keski-Oja, J., K. Koli and H. von Melchner (2004). "TGF-β activation by traction?" Trends Cell Biol. 14(12): 657-659. Kim, M. Y., T. Zhang and W. L. Kraus (2005). "Poly(ADP-ribosyl)ation by PARP- 1: `PAR-laying' NAD+ into a nuclear signal." Genes Dev. 19(17): 1951- 1967. Knockaert, M., G. Sapkota, C. Alarcon, J. Massague and A. H. Brivanlou (2006). "Unique players in the BMP pathway: Small C-terminal domain phosphatases dephosphorylate Smad1 to attenuate BMP signaling." PNAS 103(32): 11940-11945. Koinuma, D., M. Shinozaki, A. Komuro, K. Goto, M. Saitoh, A. Hanyu, M. Ebina, T. Nukiwa, K. Miyazawa, T. Imamura and K. Miyazono (2003). "Arkadia amplifies TGF-β superfamily signalling through degradation of Smad7." EMBO J. 22(24): 6458-6470. Komuro, A., T. Imamura, M. Saitoh, Y. Yoshida, T. Yamori, K. Miyazono and K. Miyazawa (2004). "Negative regulation of transforming growth factor-β (TGF-β) signaling by WW domain-containing protein 1 (WWP1)." Onco- gene in press. Kowanetz, M., U. Valcourt, R. Bergstrom, C. H. Heldin and A. Moustakas (2004). "Id2 and Id3 define the potency of cell proliferation and differentiation responses to transforming growth factor beta and bone morphogenetic protein." Mol Cell Biol 24(10): 4241-54. Kraus, W. L. (2008). "Transcriptional control by PARP-1: chromatin modulation, enhancer-binding, coregulation, and insulation." Curr Opin Cell Biol 20(3): 294-302. Kraus, W. L. and J. T. Lis (2003). "PARP Goes Transcription." Cell 113(6): 677- 683. Kretzschmar, M., J. Doody and J. Massagué (1997). "Opposing BMP and EGF signalling pathways converge on the TGF-β family mediator Smad1." Nature 389(6651): 618-622.

51 Kretzschmar, M., F. Liu, A. Hata, J. Doody and J. Massague (1997). "The TGF-β family mediator Smad1 is phosphorylated directly and activated functional- ly by the BMP receptor kinase." Genes Dev. 11(8): 984-95. Ku, M.-c., S. Stewart and A. Hata (2003). "Poly(ADP-ribose) polymerase 1 interacts with OAZ and regulates BMP-target genes." Biochemical and Biophysical Research Communications 311(3): 702-707. Kume, S., M. Haneda, K. Kanasaki, T. Sugimoto, S. Araki, K. Isshiki, M. Isono, T. Uzu, L. Guarente, A. Kashiwagi and D. Koya (2007). "SIRT1 inhibits transforming growth factor beta-induced apoptosis in glomerular mesangial cells via Smad7 deacetylation." J Biol Chem 282(1): 151-8. Kuratomi, G., A. Komuro, K. Goto, M. Shinozaki, K. Miyazawa, K. Miyazono and T. Imamura (2005 ). "NEDD4-2 (neural precursor cell expressed, develop- mentally 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(3): 461-470. Kurisaki, A., S. Kose, Y. Yoneda, C.-H. Heldin and A. Moustakas (2001). "Trans- forming growth factor-β induces nuclear import of Smad3 in an importin- β1 and Ran-dependent manner." Mol. Biol. Cell 12(4): 1079-1091. Kurisaki, A., K. Kurisaki, M. Kowanetz, H. Sugino, Y. Yoneda, C.-H. Heldin and A. Moustakas (2006). "The Mechanism of Nuclear Export of Smad3 In- volves Exportin 4 and Ran." Mol. Cell. Biol. 26(4): 1318-1332. Langelier, M. F., K. M. Servent, E. E. Rogers and J. M. Pascal (2008). "A third zinc- binding domain of human poly(ADP-ribose) polymerase-1 coordinates DNA-dependent enzyme activation." J Biol Chem 283(7): 4105-14. Lee, P. S., C. Chang, D. Liu and R. Derynck (2003). "Sumoylation of Smad4, the common Smad mediator of transforming growth factor-beta family signaling." J Biol Chem 278(30): 27853-63. Lefebvre, D. L. and C. F. Rosen (2005). "Regulation of SNARK activity in response to cellular stresses." Biochimica et Biophysica Acta (BBA) - General Sub- jects 1724(1-2): 71-85. Liang, M., Y. Y. Liang, K. Wrighton, D. Ungermannova, X. P. Wang, F. C. Bruni- cardi, X. Liu, X.-H. Feng and X. Lin (2004). "Ubiquitination and Proteoly- sis of Cancer-Derived Smad4 Mutants by SCFSkp2." Mol. Cell. Biol. 24(17): 7524-7537. Lin, H. K., S. Bergmann and P. P. Pandolfi (2004). "Cytoplasmic PML function in TGF-beta signalling." Nature 431(7005): 205-11. Lin, S. J., T. F. Lerch, R. W. Cook, T. S. Jardetzky and T. K. Woodruff (2006). "The structural basis of TGF-beta, bone morphogenetic protein, and activin ligand binding." Reproduction 132(2): 179-90. Lin, X.-z., H. Takemori, Y. Katoh, J. Doi, N. Horike, A. Makino, Y. Nonaka and M. Okamoto (2001). "Salt-Inducible Kinase Is Involved in the ACTH/cAMP- Dependent Protein Kinase Signaling in Y1 Mouse Adrenocortical Tumor Cells." Mol Endocrinol 15(8): 1264-1276. Lin, X., X. Duan, Y. Y. Liang, Y. Su, K. H. Wrighton, J. Long, M. Hu, C. M. Davis, J. Wang, F. C. Brunicardi, Y. Shi, Y. G. Chen, A. Meng and X. H. Feng (2006). "PPM1A functions as a Smad phosphatase to terminate TGFβ signaling." Cell 125(5): 915-928.

52 Lin, X., M. Liang, Y. Y. Liang, F. C. Brunicardi, F. Melchior and X. H. Feng (2003). "Activation of transforming growth factor-beta signaling by SU- MO-1 modification of tumor suppressor Smad4/DPC4." J Biol Chem 278(21): 18714-9. Liu, F. (2006). "Smad3 phosphorylation by cyclin-dependent kinases." Cytokine & Growth Factor Reviews 17(1-2): 9-17. Lizcano, J. M., O. Göransson, R. Toth, M. Deak, N. A. Morrice, J. Boudeau, S. A. Hawley, L. Udd, T. P. Mäkelä, D. G. Hardie and D. R. Alessi (2004). "LKB1 is a master kinase that activates 13 kinases of the AMPK subfamily, including MARK/PAR-1." EMBO J. 23(4): 833-843. Long, J., G. Wang, D. He and F. Liu (2004). "Repression of Smad4 transcriptional activity by SUMO modification." Biochem J 379(Pt 1): 23-9. Lonn, P., A. Moren, E. Raja, M. Dahl and A. Moustakas (2009). "Regulating the stability of TGFbeta receptors and Smads." Cell Res 19(1): 21-35. Luo, K. and H. F. Lodish (1997). "Positive and negative regulation of type II TGF- beta receptor signal transduction by autophosphorylation on multiple serine residues." Embo J 16(8): 1970-81. Maduzia, L. L., A. F. Roberts, H. Wang, X. Lin, L. J. Chin, C. M. Zimmerman, S. Cohen, X.-H. Feng and R. W. Padgett (2005). "C. elegans serine-threonine kinase KIN-29 modulates TGFβ signaling and regulates body size formation." BMC Dev. Biol. 5(1): 8. Malanga, M., J. M. Pleschke, H. E. Kleczkowska and F. R. Althaus (1998). "Poly(ADP-ribose) binds to specific domains of p53 and alters its DNA binding functions." J Biol Chem 273(19): 11839-43. Manning, G., D. B. Whyte, R. Martinez, T. Hunter and S. Sudarsanam (2002). "The Protein Kinase Complement of the Human Genome." Science 298(5600): 1912-1934. Massagué, J., S. W. Blain and R. S. Lo (2000). "TGFβ signaling in growth control, cancer, and heritable disorders." Cell 103(2): 295-309. Massagué, J., J. Seoane and D. Wotton (2005). "Smad transcription factors." Genes Dev. 19(23): 2783-2810. Masson, M., C. Niedergang, V. Schreiber, S. Muller, J. Menissier-de Murcia and G. de Murcia (1998). "XRCC1 Is Specifically Associated with Poly(ADP- Ribose) Polymerase and Negatively Regulates Its Activity following DNA Damage." Mol. Cell. Biol. 18(6): 3563-3571. Matsuura, I., N. G. Denissova, G. Wang, D. He, J. Long and F. Liu (2004). "Cyclin- dependent kinases regulate the antiproliferative function of Smads." Nature 430(6996): 226-31. Mavrakis, K. J., R. L. Andrew, K. L. Lee, C. Petropoulou, J. E. Dixon, N. Navarat- nam, D. P. Norris and V. Episkopou (2007). "Arkadia enhances Nod- al/TGF-beta signaling by coupling phospho-Smad2/3 activity and turnover." PLoS Biol 5(3): e67. Mendoza-Alvarez, H. and R. Alvarez-Gonzalez (2001). "Regulation of p53 Se- quence-specific DNA-binding by Covalent Poly(ADP-ribosyl)ation." J. Bi- ol. Chem. 276(39): 36425-36430.

53 Messner, S., M. Altmeyer, H. Zhao, A. Pozivil, B. Roschitzki, P. Gehrig, D. Ruti- shauser, D. Huang, A. Caflisch and M. O. Hottiger (2010). "PARP1 ADP- ribosylates lysine residues of the core histone tails." Nucleic Acids Res. Mitchell, H., A. Choudhury, R. E. Pagano and E. B. Leof (2004). "Ligand-dependent and -independent transforming growth factor-beta receptor recycling regulated by clathrin-mediated endocytosis and Rab11." Mol Biol Cell 15(9): 4166-78. Mochizuki, T., H. Miyazaki, T. Hara, T. Furuya, T. Imamura, T. Watabe and K. Miyazono (2004). "Roles for the MH2 domain of Smad7 in the specific inhibition of transforming growth factor-β superfamily signaling." J. Biol. Chem. 279(30): 31568-31574. Moren, A., U. Hellman, Y. Inada, T. Imamura, C. H. Heldin and A. Moustakas (2003). "Differential ubiquitination defines the functional status of the tumor suppressor Smad4." J Biol Chem 278(35): 33571-82. Morén, A., T. Imamura, K. Miyazono, C.-H. Heldin and A. Moustakas (2005). "De- gradation of the tumor suppressor Smad4 by WW and HECT domain ubiquitin ligases." J. Biol. Chem. 280(23): 22115-22123. Moses, H. L., E. L. Branum, J. A. Proper and R. A. Robinson (1981). "Transforming growth factor production by chemically transformed cells." Cancer Res. 41(7): 2842-2848. Moustakas, A. and C.-H. Heldin (2005). "Non-Smad TGF-β signals." J. Cell Sci. 118(Pt 16): 3573-3584. Moustakas, A. and C. H. Heldin (2009). "The regulation of TGFbeta signal transduction." Development 136(22): 3699-714. Moustakas, A., Pardali, K., Gaal, A., and Heldin, C.-H. (2002). "Mechanisms of TGF-β signaling in regulation of cell growth and differentiation." Immunol. Lett. 82: 85-91. Mulder, K. M. (2000). "Role of Ras and Mapks in TGFbeta signaling." Cytokine Growth Factor Rev 11(1-2): 23-35. Nakao, A., M. Afrakhte, A. Morén, T. Nakayama, J. L. Christian, R. Heuchel, S. Itoh, M. Kawabata, N. E. Heldin, C.-H. Heldin and P. ten Dijke (1997). "Identification of Smad7, a TGFβ-inducible antagonist of TGF-β signalling." Nature 389(6651): 631-635. Nickel, J., W. Sebald, J. C. Groppe and T. D. Mueller (2009). "Intricacies of BMP receptor assembly." Cytokine Growth Factor Rev 20(5-6): 367-77. Nicolas, F. J., K. De Bosscher, B. Schmierer and C. S. Hill (2004). "Analysis of Smad nucleocytoplasmic shuttling in living cells." J. Cell Sci. 117(Pt 18): 4113-4125. Oei, S. L. and Y. Shi (2001). "Poly(ADP-Ribosyl)ation of Transcription Factor Yin Yang 1 under Conditions of DNA Damage." Biochemical and Biophysical Research Communications 285(1): 27-31. Ogata, N., K. Ueda, M. Kawaichi and O. Hayaishi (1981). "Poly(ADP-ribose) syn- thetase, a main acceptor of poly(ADP-ribose) in isolated nuclei." J. Biol. Chem. 256(9): 4135-4137. Ogunjimi, A. A., D. J. Briant, N. Pece-Barbara, C. Le Roy, G. M. Di Guglielmo, P. Kavsak, R. K. Rasmussen, B. T. Seet, F. Sicheri and J. L. Wrana (2005).

54 "Regulation of Smurf2 ubiquitin ligase activity by anchoring the E2 to the HECT domain." Mol Cell 19(3): 297-308. Ohshima, T. and K. Shimotohno (2003). "Transforming growth factor-beta-mediated signaling via the p38 MAP kinase pathway activates Smad-dependent transcription through SUMO-1 modification of Smad4." J Biol Chem 278(51): 50833-42. Okada, H., T. Inoue, T. Kikuta, N. Kato, Y. Kanno, N. Hirosawa, Y. Sakamoto, T. Sugaya and H. Suzuki (2008). "Poly(ADP-ribose) polymerase-1 enhances transcription of the profibrotic CCN2 gene." J Am Soc Nephrol 19(5): 933- 42. Patterson, G. I. and R. W. Padgett (2000). "TGF beta-related pathways. Roles in Caenorhabditis elegans development." Trends Genet 16(1): 27-33. Pelkmans, L., E. Fava, H. Grabner, M. Hannus, B. Habermann, E. Krausz and M. Zerial (2005). "Genome-wide analysis of human kinases in clathrin- and caveolae/raft-mediated endocytosis." Nature 436(7047): 78-86. Pera, E. M., A. Ikeda, E. Eivers and E. M. De Robertis (2003). "Integration of IGF, FGF, and anti-BMP signals via Smad1 phosphorylation in neural induction." Genes Dev. 17(24): 3023-3028. Petrucco, S. and R. Percudani (2008). "Structural recognition of DNA by poly(ADP- ribose)polymerase-like zinc finger families." FEBS J 275(5): 883-93. Pierreux, C. E., F. J. Nicolas and C. S. Hill (2000). "Transforming growth factor β- independent shuttling of Smad4 between the cytoplasm and nucleus." Mol. Cell. Biol. 20(23): 9041-54. Pietenpol, J. A., J. T. Holt, R. W. Stein and H. L. Moses (1990). "Transforming growth factor beta 1 suppression of c-myc gene transcription: role in inhibition of keratinocyte proliferation." Proc Natl Acad Sci U S A 87(10): 3758- 62. Rahimi, R. A. and E. B. Leof (2007). "TGF-beta signaling: a tale of two responses." J Cell Biochem 102(3): 593-608. Rifkin, D. B. (2005). "Latent transforming growth factor-β (TGF-β) binding proteins: orchestrators of TGF-β availability." J. Biol. Chem. 280(9): 7409- 7412. Roberts, A. B., L. C. Lamb, D. L. Newton, M. B. Sporn, J. E. De Larco and G. J. Todaro (1980). "Transforming growth factors: isolation of polypeptides from virally and chemically transformed cells by acid/ethanol extraction." Proc. Natl. Acad. Sci. U S A 77(6): 3494-3498. Roelen, B. A., O. S. Cohen, M. K. Raychowdhury, D. N. Chadee, Y. Zhang, J. M. Kyriakis, A. A. Alessandrini and H. Y. Lin (2003). "Phosphorylation of threonine 276 in Smad4 is involved in transforming growth factor-beta- induced nuclear accumulation." Am J Physiol Cell Physiol 285(4): C823- 30. Ross, S., E. Cheung, T. G. Petrakis, M. Howell, W. L. Kraus and C. S. Hill (2006). "Smads orchestrate specific histone modifications and chromatin remode- ling to activate transcription." Embo J 25(19): 4490-502. Ross, S. and C. S. Hill (2008). "How the Smads regulate transcription." Int J Bio- chem Cell Biol 40(3): 383-408.

55 Rouleau, M., A. Patel, M. J. Hendzel, S. H. Kaufmann and G. G. Poirier (2010). "PARP inhibition: PARP1 and beyond." Nat Rev Cancer 10(4): 293-301. Ruiz, J. C., F. L. Conlon and E. J. Robertson (1994). "Identification of novel protein kinases expressed in the myocardium of the developing mouse heart." Mech. Dev. 48(3): 153-164. Ruscetti, T., B. E. Lehnert, J. Halbrook, H. Le Trong, M. F. Hoekstra, D. J. Chen and S. R. Peterson (1998). "Stimulation of the DNA-dependent Protein Ki- nase by Poly(ADP-Ribose) Polymerase." J. Biol. Chem. 273(23): 14461- 14467. Saha, T., D. Vardhini, Y. Tang, V. Katuri, W. Jogunoori, E. A. Volpe, D. Haines, A. Sidawy, X. Zhou, I. Gallicano, R. Schlegel, B. Mishra and L. Mishra (2006). "RING finger-dependent ubiquitination by PRAJA is dependent on TGF-beta and potentially defines the functional status of the tumor suppressor ELF." Oncogene 25(5): 693-705. Sapkota, G., M. Knockaert, C. Alarcon, E. Montalvo, A. H. Brivanlou and J. Massa- gue (2006). "Dephosphorylation of the Linker Regions of Smad1 and Smad2/3 by Small C-terminal Domain Phosphatases Has Distinct Out- comes for Bone Morphogenetic Protein and Transforming Growth Factor- beta Pathways." J. Biol. Chem. 281(52): 40412-40419. Savage, C., P. Das, A. L. Finelli, S. R. Townsend, C. Y. Sun, S. E. Baird and R. W. Padgett (1996). "Caenorhabditis elegans genes sma-2, sma-3, and sma-4 define a conserved family of transforming growth factor β pathway components." Proc. Natl. Acad. Sci. U S A 93(2): 790-4. Schlunegger, M. P. and M. G. Grutter (1992). "An unusual feature revealed by the crystal structure at 2.2 A resolution of human transforming growth factor- beta 2." Nature 358(6385): 430-4. Schmierer, B. and C. S. Hill (2005). "Kinetic analysis of Smad nucleocytoplasmic shuttling reveals a mechanism for transforming growth factor beta- dependent nuclear accumulation of Smads." Mol. Cell. Biol. 25(22): 9845- 9858. Schreiber, V. r., F. o. Dantzer, J.-C. AmÃ© and G. de Murcia (2006). "Poly(ADP- ribose): novel functions for an old molecule." 7(7): 517-528. Sekelsky, J. J., S. J. Newfeld, L. A. Raftery, E. H. Chartoff and W. M. Gelbart (1995). "Genetic characterization and cloning of mothers against dpp, a gene required for decapentaplegic function in Drosophila melanogaster." Genetics 139(3): 1347-58. Seo, S. R., F. Lallemand, N. Ferrand, M. Pessah, S. L'Hoste, J. Camonis and A. Atfi (2004). "The novel E3 ubiquitin ligase Tiul1 associates with TGIF to target Smad2 for degradation." EMBO J. 23(19): 3780-3792. Shi, W., C. Sun, B. He, W. Xiong, X. Shi, D. Yao and X. Cao (2004). "GADD34- PP1c recruited by Smad7 dephosphorylates TGFbeta type I receptor." J Cell Biol 164(2): 291-300. Shi, Y. and J. Massagué (2003). "Mechanisms of TGF-β signaling from cell membrane to the nucleus." Cell 113: 685-700. Shi, Y., Y. F. Wang, L. Jayaraman, H. Yang, J. Massague and N. P. Pavletich (1998). "Crystal structure of a Smad MH1 domain bound to DNA: insights on DNA binding in TGF-beta signaling." Cell 94(5): 585-94.

56 Sieber, C., J. Kopf, C. Hiepen and P. Knaus (2009). "Recent advances in BMP receptor signaling." Cytokine Growth Factor Rev 20(5-6): 343-55. Simonsson, M., C. H. Heldin, J. Ericsson and E. Gronroos (2005). "The balance between acetylation and deacetylation controls Smad7 stability." J Biol Chem 280(23): 21797-803. Simonsson, M., M. Kanduri, E. Gronroos, C. H. Heldin and J. Ericsson (2006). "The DNA binding activities of Smad2 and Smad3 are regulated by coactivator- mediated acetylation." J Biol Chem 281(52): 39870-80. Sorrentino, A., N. Thakur, S. Grimsby, A. Marcusson, V. von Bulow, N. Schuster, S. Zhang, C. H. Heldin and M. Landstrom (2008). "The type I TGF-beta receptor engages TRAF6 to activate TAK1 in a receptor kinase- independent manner." Nat Cell Biol 10(10): 1199-207. Souchelnytskyi, S., K. Tamaki, U. Engström, C. Wernstedt, P. ten Dijke and C.-H. Heldin (1997). "Phosphorylation of Ser465 and Ser467 in the C terminus of Smad2 mediates interaction with Smad4 and is required for transforming growth factor-β signaling." J. Biol. Chem. 272(44): 28107-28115. Sterling, J. A., L. Wu and S. S. Banerji (2006). "PARP regulates TGF-beta receptor type II expression in estrogen receptor-positive breast cancer cell lines." Anticancer Res 26(3A): 1893-901. Suzuki, C., G. Murakami, M. Fukuchi, T. Shimanuki, Y. Shikauchi, T. Imamura and K. Miyazono (2002). "Smurf1 regulates the inhibitory activity of Smad7 by targeting Smad7 to the plasma membrane." J Biol Chem 277(42): 39919- 25. Tajima, Y., K. Goto, M. Yoshida, K. Shinomiya, T. Sekimoto, Y. Yoneda, K. Miya- zono and T. Imamura (2003). "Chromosomal region maintenance 1 (CRM1)-dependent nuclear export of Smad ubiquitin regulatory factor 1 (Smurf1) is essential for negative regulation of transforming growth factor- β signaling by Smad7." J. Biol. Chem. 278(12): 10716-10721. ten Dijke, P. and C. S. Hill (2004). "New insights into TGF-β-Smad signalling." Trends Biochem. Sci. 29(5): 265-273. Thompson, T. B., T. K. Woodruff and T. S. Jardetzky (2003). "Structures of an ActRIIB:activin A complex reveal a novel binding mode for TGF-beta ligand:receptor interactions." Embo J 22(7): 1555-66. Tsukazaki, T., T. A. Chiang, A. F. Davison, L. Attisano and J. L. Wrana (1998). "SARA, a FYVE domain protein that recruits Smad2 to the TGFβ receptor." Cell 95(6): 779-91. Tu, A. W. and K. Luo (2007). "Acetylation of Smad2 by the co-activator p300 regulates activin and transforming growth factor beta response." J Biol Chem 282(29): 21187-96. Wagner, D. O., C. Sieber, R. Bhushan, J. H. Borgermann, D. Graf and P. Knaus (2010). "BMPs: from bone to body morphogenetic proteins." Sci Signal 3(107): mr1. Valdimarsdottir, G., M. J. Goumans, F. Itoh, S. Itoh, C. H. Heldin and P. ten Dijke (2006). "Smad7 and protein phosphatase 1alpha are critical determinants in the duration of TGF-beta/ALK1 signaling in endothelial cells." BMC Cell Biol 7: 16.

57 van der Linden, A. M., K. M. Nolan and P. Sengupta (2007). "KIN-29 SIK regulates chemoreceptor gene expression via an MEF2 transcription factor and a class II HDAC." Embo J 26(2): 358-70. Wan, M., X. Cao, Y. Wu, S. Bai, L. Wu, X. Shi and N. Wang (2002). "Jab1 antago- nizes TGF-beta signaling by inducing Smad4 degradation." EMBO Rep 3(2): 171-6. Wan, M., Y. Tang, E. M. Tytler, C. Lu, B. Jin, S. M. Vickers, L. Yang, X. Shi and X. Cao (2004). "Smad4 protein stability is regulated by ubiquitin ligase SCF beta-TrCP1." J Biol Chem 279(15): 14484-7. Wang, B., H. Suzuki and M. Kato (2008). "Roles of mono-ubiquitinated Smad4 in the formation of Smad transcriptional complexes." Biochem Biophys Res Commun 376(2): 288-92. Wang, Z., H. Takemori, S. K. Halder, Y. Nonaka and M. Okamoto (1999). "Cloning of a novel kinase (SIK) of the SNF1/AMPK family from high salt diet- treated rat adrenal." FEBS Lett. 453(1-2): 135-139. Varelas, X., R. Sakuma, P. Samavarchi-Tehrani, R. Peerani, B. M. Rao, J. Dembo- wy, M. B. Yaffe, P. W. Zandstra and J. L. Wrana (2008). "TAZ controls Smad nucleocytoplasmic shuttling and regulates human embryonic stem- cell self-renewal." Nat Cell Biol 10(7): 837-48. Watanabe, M., N. Masuyama, M. Fukuda and E. Nishida (2000). "Regulation of intracellular dynamics of Smad4 by its leucine-rich nuclear export signal." EMBO Rep. 1(2): 176-182. Watanabe, Y., S. Itoh, T. Goto, E. Ohnishi, M. Inamitsu, F. Itoh, K. Satoh, E. Wier- cinska, W. Yang, L. Shi, A. Tanaka, N. Nakano, A. M. Mommaas, H. Shi- buya, P. Ten Dijke and M. Kato (2010). "TMEPAI, a transmembrane TGF- beta-inducible protein, sequesters Smad proteins from active participation in TGF-beta signaling." Mol Cell 37(1): 123-34. Wicks, S. J., K. Haros, M. Maillard, L. Song, R. E. Cohen, P. T. Dijke and A. Chan- try (2005). "The deubiquitinating enzyme UCH37 interacts with Smads and regulates TGF-beta signalling." Oncogene 24(54): 8080-4. Wiesner, S., A. A. Ogunjimi, H. R. Wang, D. Rotin, F. Sicheri, J. L. Wrana and J. D. Forman-Kay (2007). "Autoinhibition of the HECT-type ubiquitin ligase Smurf2 through its C2 domain." Cell 130(4): 651-62. Wohlfert, E. A., L. Gorelik, R. Mittler, R. A. Flavell and R. B. Clark (2006). "Cut- ting edge: deficiency in the E3 ubiquitin ligase Cbl-b results in a multifunc- tional defect in T cell TGF-beta sensitivity in vitro and in vivo." J Immunol 176(3): 1316-20. Wrana, J. L., L. Attisano, R. Wieser, F. Ventura and J. Massague (1994). "Mechan- ism of activation of the TGF-beta receptor." Nature 370(6488): 341-7. Wrighton, K. H., X. Lin and X. H. Feng (2009). "Phospho-control of TGF-beta superfamily signaling." Cell Res 19(1): 8-20. Wu, M. Y. and C. S. Hill (2009). "Tgf-beta superfamily signaling in embryonic development and homeostasis." Dev Cell 16(3): 329-43. Xiao, Z., R. Latek and H. F. Lodish (2003). "An extended bipartite nuclear localization signal in Smad4 is required for its nuclear import and transcriptional activity." Oncogene 22(7): 1057-1069.

58 Xiao, Z., X. Liu, Y. I. Henis and H. F. Lodish (2000). "A distinct nuclear localization signal in the N terminus of Smad3 determines its ligand-induced nuclear translocation." Proc. Natl. Acad. Sci. U S A 97(14): 7853-7858. Xie, M., D. Zhang, J. R. B. Dyck, Y. Li, H. Zhang, M. Morishima, D. L. Mann, G. E. Taffet, A. Baldini, D. S. Khoury and M. D. Schneider (2006). "A pivotal role for endogenous TGF-beta-activated kinase-1 in the LKB1/AMP- activated protein kinase energy-sensor pathway." PNAS 103(46): 17378- 17383. Xu, L., C. Alarcón, S. Cöl and J. Massagué (2003). "Distinct domain utilization by Smad3 and Smad4 for nucleoporin interaction and nuclear import." J. Biol. Chem. 278(43): 42569-42577. Xu, L., Y. G. Chen and J. Massagué (2000). "The nuclear import function of Smad2 is masked by SARA and unmasked by TGFβ-dependent phosphorylation." Nat. Cell Biol. 2(8): 559-562. Xu, L., Y. Kang, S. Cöl and J. Massagué (2002). "Smad2 nucleocytoplasmic shuttling by nucleoporins CAN/Nup214 and Nup153 feeds TGFβ signaling complexes in the cytoplasm and nucleus." Mol. Cell 10(2): 271-282. Xu, L., X. Yao, X. Chen, P. Lu, B. Zhang and Y. T. Ip (2007). "Msk is required for nuclear import of TGF-{beta}/BMP-activated Smads." J Cell Biol 178(6): 981-94. Yakymovych, I., P. ten Dijke, C.-H. Heldin and S. Souchelnytskyi (2001). "Regula- tion of Smad signaling by protein kinase C." FASEB J. 15(3): 553-555. Yamaguchi, K., S. Nagai, J. Ninomiya-Tsuji, M. Nishita, K. Tamai, K. Irie, N. Ue- no, E. Nishida, H. Shibuya and K. Matsumoto (1999). "XIAP, a cellular member of the inhibitor of apoptosis protein family, links the receptors to TAB1-TAK1 in the BMP signaling pathway." EMBO J. 18(1): 179-187. Yao, X., X. Chen, C. Cottonham and L. Xu (2008). "Preferential utilization of Imp7/8 in nuclear import of Smads." J Biol Chem 283(33): 22867-74. Yingling, J. M., M. B. Datto, C. Wong, J. P. Frederick, N. T. Liberati and X. F. Wang (1997). "Tumor suppressor Smad4 is a transforming growth factor beta-inducible DNA binding protein." Mol Cell Biol 17(12): 7019-28. Yu, L., M. C. Hebert and Y. E. Zhang (2002). "TGF-β receptor-activated p38 MAP kinase mediates Smad-independent TGF-β responses." EMBO J. 21(14): 3749-3759. Zaniolo, K., S. Desnoyers, S. Leclerc and S. L. Guerin (2007). "Regulation of poly(ADP-ribose) polymerase-1 (PARP-1) gene expression through the post-translational modification of Sp1: a nuclear target protein of PARP-1." BMC Mol Biol 8: 96. Zawel, L., J. L. Dai, P. Buckhaults, S. Zhou, K. W. Kinzler, B. Vogelstein and S. E. Kern (1998). "Human Smad3 and Smad4 are sequence-specific transcription activators." Mol Cell 1(4): 611-7. Zhang, Y., C. Chang, D. J. Gehling, A. Hemmati-Brivanlou and R. Derynck (2001). "Regulation of Smad degradation and activity by Smurf2, an E3 ubiquitin ligase." Proc. Natl. Acad. Sci. U S A 98(3): 974-979. Zhu, H., P. Kavsak, S. Abdollah, J. L. Wrana and G. H. Thomsen (1999). "A SMAD ubiquitin ligase targets the BMP pathway and affects embryonic pattern formation." Nature 400(6745): 687-693.