Characterization of Novel Substrates of the Tankyrase and RNF146 Destruction Complex and Mechanisms of Its Regulation

Characterization of Novel Substrates of the Tankyrase and RNF146 Destruction Complex and Mechanisms of its Regulation

Arun Chandrakumar

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Medical Biophysics University of Toronto

Characterization of Novel Substrates of the Tankyrase and RNF146 Destruction Complex and Mechanisms of its Regulation

Arun Chandrakumar

Doctor of Philosophy

Medical Biophysics University of Toronto

2020 Abstract

Poly-ADP-ribose is a post-translational modification that was first described over 50 years ago as a polymer derived from nicotinamide adenine dinucleotide or NAD. Since then, a family of 17 enzymes responsible for generating Poly-ADP-ribosylation or PARylation post translational modification has been identified and characterized. A unique member of this family of enzymes, called Poly-ADP-ribose Polymerases (PARPs), are Tankyrases. There are two mammalian Tankyrases, Tankyrase 1 and 2, whose domain organization include ankyrin repeat clusters which mediate enzyme-substrate interactions and a SAM domain which regulates oligomerization of Tankyrase into large macromolecular complexes. Tankyrases bind to proteins through a ‘RxxxxG’ peptide motif which facilities these proteins to undergo PARylation. A subset of Tankyrase substrates are recognized by an E3 ligase, RNF146, which facilities protein degradation through PARylation dependent ubiquitylation. The studies summarized in this thesis have focused on the identification of new proteins targets involved in the Tankyrase:RNF146 degradation pathway and investigates how Tankyrase PARylation is regulated through FIH dependent hydroxylation.

I have identified SH3BP5 and SH3BP5L as new Tankyrase substrates that are targets of RNF146. I have shown that both substrates are guanine exchange factors for the small GTPase Rab11a. I have defined the minimal catalytic core of SH3BP5/L and shown it to be composed of a novel two α-helix GEF domain. My work has demonstrated that SH3BP5 and SH3BP5L are both required for optimal activation of Rab11a in epithelial cells during lumenogenesis and that their activities are repressed by Tankyrase-mediated PARylation. RNF146 regulates Rab11a

activity by controlling Tankyrase protein abundance, marking the first time that RNF146 has been described as antagonistic towards Tankyrase function.

I have demonstrated that FIH-mediated hydroxylation of Tankyrase affects the ability of substrates to bind to the ankyrin repeat clusters and inhibits autoPARylation and substrate PARylation yet does not affect protein degradation. Since PARylation dependant degradation is unaffected, it suggests that hydroxylation could affect the degradation independent functions of Tankyrase.

iii

Acknowledgments

Foremost, I am extremely grateful for the opportunity given to me to pursue graduate work and realize my dream of becoming a scientist by my supervisor Dr. Robert Rottapel. Thank you for taking a chance on a 21-year-old who did not have a clue what he wanted to do in life but knew he enjoyed lab research. Thank you for your constant encouragement through this up and down journey and always being available whenever I really needed your help. I will always appreciate the support you have given me and I’m sure you will continue to provide as I go into the next phase of my career. Many thanks to my committee members, Dr. Raught and Dr. Wouters, for their helpful feedback and encouragement during my graduate work.

Secondly, I have been so fortunate to have amazing, helpful, hilarious, and enjoyable lab colleagues. There have been so many people that have come and gone through the lab over the years and someway every individual has had an impact on me that has made working in the Rottapel lab an enjoyable experience.

Thank you to my collaborators who have contributed critical pieces of data that made completion of this thesis possible. Dr. Etienne Coyaud, you were such a pleasure to work with and cannot thank you enough for your help and advice and willingness to let me bother you so many times. Dr. Chris Marshall, I learned so much from you over the past couple years and it has been truly a pleasure to work with you. Can’t thank you enough for all the time you spent collecting data, brainstorming and working through road-bumps. Also, a big thank you to Dr. Sebastian Guettler and Dr. David Bryant for their advice and providing valuable reagents.

A big thank you to my family and friends for your support and encouragement for the last seven years as I navigated through this challenging yet satisfying journey.

Lastly, huge thanks to the Toronto Raptors for putting the cherry on top of 2019, one of the best years of my life.

WHAT IT DOOOO BAYYBEEEE!!!!!!

Table of Contents

Acknowledgments...... iv

Table of Contents ...... v

List of Tables ...... xi

List of Figures ...... xii

List of Abbreviations ...... xiv

Chapter 1 ...... 1

Introduction ...... 1

1.1 Post translational modification ...... 1

1.2 Poly-ADP-Ribose ...... 1

1.3 Poly-ADP-Ribose Polymerases (PARPs) ...... 2

1.3.1 Revised classification system based on the PARP domain ...... 2

1.3.2 PARP biology defined by PAR binding domains ...... 3

1.4 Tankyrases ...... 4

1.4.1 Tankyrase family ...... 4

1.4.2 Tankyrases bind to substrates through a specific amino-acid motif ...... 5

1.4.3 Tankyrases oligomerize and form higher order assemblies ...... 8

1.4.4 Role of Tankyrases at human telomeres ...... 8

1.4.5 Tankyrase in mitosis ...... 9

1.4.6 Tankyrase controls glucose metabolism ...... 9

1.4.7 Role of Tankyrase in WNT signaling and cancer ...... 10

1.4.8 Tankyrase promotes tumour growth independently of WNT signaling ...... 10

1.4.9 Tankyrase regulates bone metabolism and dysregulation of 3BP2 is the underlying cause of Cherubism ...... 11

1.4.10 Tankyrase promotes YAP signaling and cell polarity ...... 11

1.5 Regulation of Tankyrase ...... 12

1.5.1 Factors that regulate catalytic activity ...... 12

1.5.2 Protein-protein interactions regulate Tankyrase stability ...... 12

1.5.3 Tankyrase stability is regulated by PAR dependent Ubiquitylation (PARdU) ...... 13

1.6 RNF146 ...... 14

1.6.1 RNF146 regulates DNA damage and PAR-mediated cell death ...... 15

1.6.2 RNF146 regulates bone metabolism ...... 15

1.7 SH3BP5...... 16

1.7.1 Role of SH3BP5 in stress kinase signaling ...... 16

1.7.2 SH3BP5 is a guanine nucleotide exchange factor for Rab11 ...... 18

1.7.3 SH3BP5 in cancer ...... 20

1.8 Rab11 GTPases ...... 20

1.8.1 Rab11 family ...... 20

1.8.2 Rab11 function at recycling endosomes ...... 21

1.8.3 Rab11 in epithelial polarity ...... 21

1.8.4 Rab11 in disease ...... 22

1.8.5 Rab11 in host-pathogen relations...... 22

1.9 Factor Inhibiting HIF1α (FIH) and the regulation of Tankyrase Function ...... 23

1.9.1 Regulation of FIH catalytic activity ...... 24

1.9.2 Hydroxylation of ankyrin-repeat containing proteins ...... 24

1.9.3 Hydroxylation affects protein catalytic function ...... 24

1.9.4 Hydroxylation affects protein-protein interactions ...... 25

1.9.5 Role of FIH in metabolism ...... 26

Chapter 2 ...... 27

Characterization of the RNF146 Proteome ...... 27

2.1 Abstract ...... 27

2.2 Introduction ...... 28 vi

2.3 Experimental Procedures ...... 30

2.3.1 Cell lines ...... 30

2.3.2 Plasmids ...... 30

2.3.3 siRNA transfection...... 31

2.3.4 Immunoblotting and immunoprecipitation ...... 31

2.3.5 In vitro ADP-ribosylation assay...... 32

2.3.6 Antibodies and reagents ...... 33

2.3.7 BioID Assay and sample preparation ...... 33

2.3.8 Mass spectrometry analysis ...... 34

2.3.9 Interaction network analysis ...... 35

2.4 Results ...... 36

2.4.1 Systematic elucidation of the RNF146 interactome ...... 36

2.4.2 SH3BP5 and SH3BP5L are novel Tankyrase substrates ...... 44

2.4.3 SH3BP5 and SH3BP5L are RNF146 interactors and substrates ...... 48

2.4.4 Endogenous SH3BP5 and SH3BP5L are not regulated by RNF146 ...... 52

2.5 Discussion ...... 55

Chapter 3 ...... 57

Tankyrase-mediated Poly-ADP Ribosylation Regulates Lumen Formation Through the Rab11 GEFs SH3BP5 and SH3BP5L ...... 57

3.1 Abstract ...... 57

3.2 Introduction ...... 58

3.3 Experimental Procedures ...... 60

3.3.1 Cell lines ...... 60

3.3.2 Plasmids ...... 60

3.3.3 Retroviral and lentiviral (shRNA and CRISPR) plasmids ...... 63

3.3.4 Retrovirus and lentivirus generation and transduction ...... 64

vii

3.3.5 MDCK stable expression cell line generation ...... 64

3.3.6 MDCK knockout lines created by CRISPR ...... 65

3.3.7 Immunoblotting and immunoprecipitation ...... 65

3.3.8 MDCK Cyst culture ...... 66

3.3.9 Immunofluorescence ...... 66

3.3.10 Confocal microscopy ...... 67

3.3.11 Single lumen quantification ...... 67

3.3.12 Protein expression and purification ...... 67

3.3.13 Rab11-GTP pulldown assay ...... 69

3.3.14 Antibodies and reagents ...... 69

3.3.15 Rab11a nucleotide exchange assay by NMR ...... 70

3.3.16 SH3BP5 BioID Assay ...... 70

3.3.17 Mass spectrometry analysis ...... 71

3.3.18 Interaction network analysis ...... 71

3.4 Results ...... 72

3.4.1 Identification of SH3BP5 interacting proteins ...... 72

3.4.2 SH3BP5 and SH3BP5L are Rab11a GEFs that mediate exchange through a novel GEF domain ...... 76

3.4.3 Tankyrase alter localization of SH3BP5/SH3BP5L and Rab11a ...... 80

3.4.4 SH3BP5 and SH3BP5L promote lumen formation in MDCK cysts ...... 84

3.4.5 Tankyrase inhibits lumen formation through negative regulation of SH3BP5 and SH3BP5L ...... 88

3.4.6 RNF146 regulates lumen formation by degrading Tankyrase ...... 90

3.5 Discussion ...... 94

Chapter 4 ...... 98

Investigating the role of asparaginyl hydroxylase FIH towards Tankyrase activity and function ...... 98

viii

4.1 Abstract ...... 98

4.2 Introduction ...... 99

4.3 Experimental Procedures ...... 101

4.3.1 Cell lines ...... 101

4.3.2 Expression plasmids...... 101

4.3.3 Immunoblotting and immunoprecipitation ...... 102

4.3.4 In vitro PARylation assay ...... 102

4.3.5 Antibodies and reagents ...... 103

4.3.6 Retroviral constructs ...... 103

4.3.7 Retrovirus and lentivirus production and transduction ...... 103

4.3.8 Isolation of primary bone marrow macrophages ...... 104

4.4 Results ...... 105

4.4.1 FIH hydroxylates Tankyrase-2 within Ankyrin Repeat Cluster (ARC) domains 105

4.4.2 FIH interacts with TNKS2 Ankyrin Repeat Clusters ...... 105

4.4.3 FIH inhibits TNKS2 activity ...... 107

4.4.4 FIH impairs TNKS2-mediated 3BP2 PARylation ...... 110

4.4.5 Depletion of FIH does not affect protein stability of Tankyrase and its substrates ...... 111

4.4.6 FIH expression has no effect on protein stability of Tankyrase substrates ...... 113

4.5 Discussion ...... 114

Chapter 5 ...... 116

Discussion and Future Directions ...... 116

5.1 Overall perspective ...... 116

5.2 Role of WWE domain E3 ligases in PARdU...... 116

5.3 Rab11a activation during lumenogenesis ...... 117

5.4 Tissue specificity and function of Tankyrase hydroxylation ...... 119

References ...... 121

List of Tables

Table 1.1: Classification of mammalian ADP-ribosyltransferases Table 1.2: List of identified Tankyrase interacting proteins Table 2.1: List of constructs and cloning primers Table 2.2: Significant RNF146 interactors from BioID Table 3.1: List of constructs and cloning primers Table 3.2: Received plasmids Table 3.3: shRNA sequences Table 3.4: pLKO.1 cloning primers Table 3.5: Guide RNA (sgRNA) Sequences Table 3.6: pLCKO.1 cloning primers Table 3.7: List of significant SH3BP5 interacting proteins Table 4.1: List of constructs and cloning primers

List of Figures

Figure 1.1: Tankyrases are Poly-ADP-Ribose Polymerases (PARPs) Figure 1.2: Crystal structure of ARC4 with Tankyrase binding motif peptides Figure 1.3: Tankyrases and substrates are degraded through PARylation dependent Ubiquitylation (PARdU) Figure 1.4: Domain organization of SH3BP5 and SH3BP5L Figure 1.5: Conservation of SH3BP5 family across organisms Figure 2.1: Generation of RNF146 BirA cell lines. Figure 2.2: Characterization of the RNF146 Interactome Figure 2.3: Classification of RNF146 interactors Figure 2.4: SH3BP5 and SH3BP5L are Tankyrase binders through multiple motifs. Figure 2.5: SH3BP5 and SH3BP5L are Tankyrase substrates. Figure 2.6: SH3BP5 and SH3BP5L are RNF146 binders. Figure 2.7: SH3BP5 and SH3BP5L are RNF146 substrates. Figure 2.8: SH3BP5 and SH3BP5L are degraded by the proteasome Figure 2.9: Endogenous SH3BP5 and SH3BP5L stability are differentially regulated by RNF146 Figure 2.10: Endogenous SH3BP5 stability is regulated by Tankyrase. Figure 3.1: Generation of SH3BP5 BirA cell line Figure 3.2: SH3BP5 interactome reveals association to Rab11 trafficking. Figure 3.3: SH3BP5 and SH3BP5L interact with Rab11a-GDP. Figure 3.4: SH3BP5 and SH3BP5L are Rab11a GEFs Figure 3.5: Entire SH3BP5 N-terminus is not required for Rab11a binding Figure 3.6: SH3BP5 and SH3BP5L have a novel two-helix catalytic domain. Figure 3.7: SH3BP5 and SH3BP5L alter localization of Rab11a. Figure 3.8: Tankyrase alters SH3BP5/L localization into punctate structures Figure 3.9: Tankyrase alters Rab11a localization through SH3BP5 and SH3BP5L. Figure 3.10: SH3BP5 and SH3BP5L promote lumen formation by activating Rab11a Figure 3.11: SH3BP5 and SH3BP5L knockdown leads to minor impairment of lumen formation Figure 3.11: SH3BP5 and SH3BP5L have necessary and redundant functions during lumen formation Figure 3.13: Tankyrase inhibits lumen formation through SH3BP5/L and other additional factors Figure 3.14: Tankyrase inhibits lumen formation and regulates SH3BP5 stability xii

Figure 3.15: RNF146 controls Tankyrase stability during lumen formation. Figure 3.16: Tankyrase inhibition rescues RNF146 impairment of lumen formation Figure 4.1: FIH hydroxylates TNKS2 within its Ankyrin Repeat Cluster domains Figure 4.2: FIH interacts with TNKS2 Ankyrin Repeat Clusters Figure 4.3 TNKS2 hydroxylation inhibits its catalytic activity Figure 4.4 FIH inhibits 3BP2 PARylation by Tankyrase. Figure 4.5 FIH knockdown and depletion do not affect the stability of Tankyrase and its substrates Figure 4.6: FIH expression has no effect on protein stability of Tankyrase substrates.

xiii

List of Abbreviations

2D – Two-dimensional 3D – Three-dimensional ABRO1 – Abraxas brother protein 1 ADP – adenosine diphosphate AIF – apoptosis inducible factor AKT – v-AKT murine thymoma viral oncogene homolog α-MEM – Minimum Essential Medium – Alpha Modification AMOT – Angiomotin AMOTL1 – Angiomotin Like 1 AMOTL2 – Angiomotin Like 2 AMPA – α-amino-3-hydroxy-5-methylisoxaole-4-propionic acid ANK - ankyrin APC – adenomatous polyposis coli ARC – ankyrin repeat cluster ARTD –ADP-ribosyltransferase Diptheria toxin-like ASB4 – Ankyrin Repeat and SOCS box containing 4 ASPP2 – Apoptosis-stimulating of p53 protein 2 ATG9 – autophagy-related protein 9 ATP – adenosine triphosphate AXIN1 – axis inhibition protein 1 BAR – Bin, Amphiphysin, Rvs BCA – bicinchoninic acid BioID – proximity-based biotinylation BMM – bone marrow macrophage BRISC – BRCC36 isopeptidase complex BSA – bovine serum albumin BTK – Bruton’s tyrosine kinase CAD – C-terminal transactivation domain Cas9 – CRISPR associated protein 9 CDC42 – cell division control protein 42 homolog CPAP – centrosomal P4.1-associated protein CBP – CREB-binding protein CRISPR – Clustered Regularly Interspaced Short Palindromic Repeats DAPI – 4,6-diamidine-2-phenylinodole-dihydrochloride DLG1 – discs large homolog 1 DMEM – Dulbecco’s Modified Eagles Medium DMSO – dimethyl sulfoxide DMOG – dimethyloxallyl Glycine DNA – deoxyribonucleic acid DTT – dithiothreitol xiv

DTX - Deltex ECL – enhanced chemiluminescence E. coli – Escherichia coli EDTA – ethylenediaminetetraacetic acid EGFR – epidermal growth factor receptor EMT – epithelial-mesenchymal transition ER – endoplasmic reticulum ERK – extracellular signal-regulated kinase FACS – fluorescence activated cell sorting FBS – fetal bovine serum FIH – Factor Inhibiting HIF-1α GAP – GTPase activating protein GDP – guanosine diphosphate GEF – guanine nucleotide exchange factor GFP – green fluorescent protein GLUT1 – glucose transporter type 1 GLUT4 – glucose transporter type 4 GMD – GDP-Mannose 4,6-Dehydratase GSK – glycogen synthase kinase GST – glutathione S-transferase GTP – guanosine triphosphate GTPγS – guanosine 5′-[γ-thio]triphosphate HA – Human influenza hemagglutinin HECT – Homologous to the E6-AP Carboxyl Terminus HEK – human embryonic kidney HEPES – 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid HIF-1α – hypoxia-inducible factor 1-alpha HLTF – helicase like transcription factor HRP – horseradish peroxidase HSQC – heteronuclear single quantum coherence HUWE1 – HECT, UBA and WWE domain-containing protein 1 IκB – NF-kappa-B inhibitor IRAP – insulin-responsive aminopeptidase IPTG – isopropyl β-D-1 thiogalactopyranoside JNK – c-Jun N-terminal kinase KO – knock-out Ku70 – 70kDa subunit of Ku antigen Ku86 – 86kDa subunit of Ku antigen LB – Luria-Bertani LDH-A – lactate dehydrogenase A LIG3 – DNA ligase 3 LRRK1 – leucine rich repeat kinase 1 xv

MAPK – mitogen-activated protein kinase MAR – mono ADP-ribose Mcl1 – induced myeloid leukemia cell differentiation protein Mcl-1 MDCK – Madine-Darby Canine Kidney MEF – mouse embryonic fibroblast MEK – MAPK/ERK Kinase MPP7 – membrane palmitoylated protein 7 NAD – nicotinamide adenine dinucleotide NEDD – neural precursor cell expressed developmentally down-regulated NF-κB – nuclear factor kappa-light-chain-enhancer of activated B-cells NMDA – N-methyl-D-aspartate NMR – Nuclear magnetic resonance Ni-NTA – Nickel nitrilotriacetic acid Notch – Neurogenic locus notch homolog protein 1 NuMA – Nuclear mitotic apparatus protein OTUB1 – ovarian tumor domain containing ubiquitin aldehyde binding protein 1 PALS1 – Protein associated with Lin seven PAR – poly (adenosine diphosphate ribose) PARdU – PAR dependent ubiquitination PARP – poly-ADP-ribose polymerase PBS – Phosphate Buffered Saline PBST – Phosphate Buffered Saline Tween PCR – polymerase chain reaction PEX14 – peroxisomal biogenesis factor 14 PFA - paraformaldehyde PHD – prolyl hydroxylase Plk1 – Polo like kinase 1 PMSF – phenylmethylsulfonyl fluoride PODXL - podocalyxin PTEN – phosphatase and tensin homolog PTM – posttranslational modification PVDF – polyvinylidene fluoride qPCR – quantitative PCR Rab11a – Ras-related protein Rab-11a Rab11b – Ras-related protein Rab-11b Rab25 – Ras-related protein Rab-25 Rab11-FIP – Rab11-family interacting protein RANKL – Receptor activator of nuclear factor kappa-B ligand RBD – Rab11 binding domain RFP – Red fluorescent protein RING – Really Interesting New Gene RNA – Ribonucleic acid xvi

RNAi – RNA interference RNF8 – RING finger protein 8 RNF146 – RING finger protein 146 ROS – reactive oxygen species SAM – sterile alpha motif SDS – sodium dodecyl sulfate SDS-PAGE – SDS-polyacrylamide gel electrophoresis SH3 – SRC homology 3 SH3BP5 – SH3-domain Binding Protein 5 SH3BP5L – SH3-domain Binding Protein 5 Like shRNA – short hairpin RNA sgRNA – single guide RNA SRC – v-src avian sarcoma viral oncogene homolog SUMO – small ubiquitin like modifier TAB182 – Tankyrase 1 binding protein 1, 182kDa TAZ – WW domain containing transcription regulator 1 TBM – Tankyrase binding motif TCEP – tris(2-carboxyethyl)phosphine TfR – Transferrin receptor TNF-α – tumor necrosis factor alpha TNKL – Tankyrase-like protein TNKS – TRF1-Interacting Ankyrin Related ADP-Ribose Polymerase TNKS2 – TRF1-Interacting Ankyrin Related ADP-Ribose Polymerase 2 TRAPP – Transport protein particle TRC – The RNAi Consortium TRF1 – Telomere repeat-binding factor 1 TRIP12 – Thyroid receptor-interacting protein 12 Tris – tris(hydroxymethyl)aminomethane TRPV3 – Transient receptor potential vanilloid 3 USP25 – ubiquitin specific peptidase 25 VEGF – vascular endothelial growth factor VSV-G – vesicular stomatitis virus-glyoprotein WNT – Wingless-type MMTV integration site WT – wild-type WWE – tryptophan tryptophan glutamate XRCC1 – x-ray repair cross-complementing protein 1 YAP – yes associated protein ZO-1 – zonula occludens- 1

xvii

Chapter 1

Introduction 1.1 Post translational modification

Post-translational modifications (PTMs) are the addition of chemical moieties that regulate protein stability, localization and function. PTMs are highly varied and include phosphorylation, hydroxylation, adenosine diphosphate (ADP)-ribosylation, acetylation, glycosylation, prenylation, myristoylation and palmitoylation. These modifications are usually reversible with the addition or removal of PTMs catalyzed by specialized enzymes. PTMs can also include covalent linkages of peptides through ubiquitylation, SUMOylation, or NEDDylation. Non- enzymatic PTMs include oxidation and glycation and these modifications occur upon release of the polypeptide chain from ribosomes (Knorre et al., 2009).

1.2 Poly-ADP-Ribose

Poly-ADP-ribose (PAR) was originally discovered in 1963 as a misidentified product of an enzyme whose activity was increased by a NAD+ precursor and generated the product referred to as polyA (Chambon et al., 1963). PolyA was later revealed to be a polymer of ADP-ribose linked at the ribose moiety (Chambon et al., 1966; Sugimura et al., 1967; Fujimura et al., 1967; Hasegawa et al., 1967; Nishizuka et al., 1967; Reeder et al., 1967). Bacterial pathogens have also evolved to utilize ADP-ribosylation to modulate host organisms. These bacteria have enzymes or exotoxins that facilitate transport through a host cell’s membrane and then use NAD as a substrate to generate ADP-ribose. Each pathogen has specific target proteins modified by ribosylation (Krueger and Barbieri 1995). Diphtheria toxin from C.diphtheriae and exotoxin A from P.aeruginosa PARylates elongation factor-2 (EF-2) and inhibit protein synthesis. Cholera toxin from V.cholerae targets the alpha subunit of G proteins and inhibits their GTPase activity. The C3 toxin from C.botulinum targets and inhibits Rho GTPases (Krueger and Barbieri 1995).

1.3 Poly-ADP-Ribose Polymerases (PARPs)

Poly-ADP Polymerases (PARPs) are post-translational modifying enzymes that transfer adenosine diphosphate ribose (ADP-ribose) onto target proteins. These enzymes use nicotinamide adenine dinucleotide (NAD+) as an ADP-ribose source and construct polymers of up to 200 ADP-ribose units in length onto glutamate, aspartate, lysine, asparagine, cysteine, arginine and tyrosine residues (Hottiger et al., 2010; Gibson and Kraus 2012; Leslie Pedrioli et al., 2018). The mammalian PARP family consists of 17 known members with some functioning solely as mono ADP-ribose (MAR) polymerases (Hottiger et al., 2010). The PARP domain is responsible for the catalytic activity of PARPs and was identified based on the high homology to the catalytic domain of Diphtheria toxin. PARP1 was the first identified member and all other proteins that contained the conserved PARP domain were given the nomenclature PARP1-17 (Hottiger et al., 2010).

1.3.1 Revised classification system based on the PARP domain

A new nomenclature to describe the family was proposed to aid in organizing ADP- ribosyltransferases to differentiate between mono and poly-ADP-ribosyltransferases. Enzymes were categorized by the conservation of key residues within the catalytic core of the PARP domain. Poly-ADP-ribosylation (PARylation) requires a histidine, tyrosine for orientation of NAD and the glutamate enables transfer of additional ADP-ribose units onto the initial acceptor site. This “HYE” triad, which is conserved in 6 PARPs is also conserved in Diphtheria toxin which led to the revised name of ADP-ribosyltransferase Diphtheria-like or ARTD for mammalian PARPs (see Table 1). 11 PARPs do not contain glutamate but have isoleucine, leucine, threonine, valine or tyrosine and were numbered based on phylogenetic analysis of the PARP domain of all 17 members (Hottiger et al., 2010).

Table 1.1: Classification of mammalian ADP-ribosyltransferases

Revised Name PARP Gene Triad Sequence Modification ARTD1 PARP1 PARP1 HYE PAR ARTD2 PARP2 PARP2 HYE PAR ARTD3 PARP3 PARP3 HYE MAR ARTD4 PARP4 PARP4 HYE PAR ARTD5 PARP5a TNKS HYE PAR ARTD6 PARP5b TNKS2 HYE PAR ARTD7 PARP15 PARP15 HYL MAR ARTD8 PARP14 PARP14 HYL MAR ARTD9 PARP9 PARP9 QYT MAR ARTD10 PARP10 PARP10 HYI PAR ARTD11 PARP11 PARP11 HYI MAR ARTD12 PARP12 PARP12 HYI MAR ARTD13 PARP13 ZC3HAV1 HYV MAR ARTD14 PARP7 TIPARP HYI MAR ARTD15 PARP16 PARP16 HYI MAR ARTD16 PARP8 PARP8 HYI MAR ARTD17 PARP6 PARP6 HYY MAR

1.3.2 PARP biology defined by PAR binding domains

PARPs are involved in a wide range of processes including DNA repair, protein turnover, transcription, mitochondrial function and signal transduction. These functions are achieved by either modulating protein-protein or protein-DNA interaction, protein localization, or protein ubiquitylation (Gibson and Kraus 2012). There are four types of PAR-recognition motifs/domains that facilitate regulation of proteins by ADP-ribose including: PAR-binding motif (PBM), PAR-binding zinc finger (PBZ), WWE domain and the macrodomain. Each PAR binding motifs/domains recognize a different aspect of the PAR chain. Whereas the WWE domain binds to iso-ADP-ribose, the PBZ domain recognizes the ADP-ribose junction or distal adenine ring, and the macrodomain binds to the terminal ADP-ribose of a PAR chain (Gibson and Kraus 2012).

1.4 Tankyrases

1.4.1 Tankyrase family

A unique member of the PARP family, TRF1-Interacting Ankyrin Related ADP-Ribose Polymerase (Tankyrase or TNKS), was originally identified as a binding partner of telomere repeat-binding factor 1 (TRF1) by yeast two-hybrid (Smith et al., 1998). It contains a series of repetitive histidine, serine and proline residues termed HPS region, 24 ankyrin repeats, a sterile alpha motif (SAM) domain, and a poly-ADP-ribose polymerase (PARP) domain. The domain organization of TNKS is shown in Figure 1.1. Another protein that shared homology to TNKS was identified through separate yeast two-hybrid screens of Grb14 and IRAP and named Tankyrase-2 (Figure 1.1) and was also identified through a serological screen of breast tumor cDNA and named TNKL or Tankyrase-2 (TNKS2) (Chi and Lodish 2000; Lyons et al., 2001; Kuimov et al., 2001). The 24 ankyrin repeats in both Tankyrases could be grouped into five conserved and distinct clusters termed ankyrin (ANK) repeat clusters or ARCs (Seimiya and Smith 2002). The ARCs are the defining characteristic of Tankyrases that make them unique amongst the PARP family. The ARC domains in TNKS and TNKS2 function to bind substrates and bring them in proximity to the catalytic PARP domain. Both TNKS and TNKS2 undergo autoPARylation upon binding of their substrates (Smith et al., 1998) (Lyons et al., 2001; Kuimov et al., 2001; Cook et al., 2002; Sbodio et al., 2002).

Figure 1.1: Tankyrases are Poly-ADP-Ribose Polymerases (PARPs). Domain organization of Tankyrase 1 and Tankyrase 2. Sequence homology is indicated in percent amino acid identity below each region. ARC - ankyrin repeat cluster; SAM - sterile alpha motif; PARP – Poly-ADP- ribose polymerase

The conservation in behavior between Tankyrase 1 and 2 suggests they may have redundant functions, which was confirmed by generation of single knockout and double knockout mice. Double knockout mice are embryonic lethal while TNKS or TNKS2 knockout mice are viable and showed no defects in development or telomere maintenance. TNKS2 deficient mice have decreased body weight and metabolic defects (Chiang, Y.J., et al 2005; 2008).

1.4.2 Tankyrases bind to substrates through a specific amino-acid motif

A specific region in the Tankyrase substrate TAB182 was identified to bind to the ankyrin regions of both Tankyrase 1 and 2 (Seimiya and Smith 2002). This region was later discovered to contain a six amino acid sequence RxxPDG that was common to multiple Tankyrase binding proteins including IRAP, TAB182 and human TRF1 (Sbodio and Chi 2002). Mutagenesis analysis demonstrated the arginine residue and the 4th position proline was essential for Tankyrase binding. Interestingly, mouse TRF1 lacks this motif and is unable to interact with Tankyrase revealing the specific nature of binding to Tankyrases (Sbodio and Chi 2002). Our laboratory and others have shown that there are specific rules that determine Tankyrase binding. A high-throughput fluorescence polarization-based screen was used to assay a library of eight amino acid peptides to identify the optimized binding sequence in each position of the peptide. This in vitro binding experiment revealed that the 1st position arginine and 6th position glycine were the two most critical elements required for Tankyrase interaction (Guettler et al., 2011) and highlighted an optimized Tankyrase binding sequence REAGDGEE that has a 600nM binding affinity to Tankyrase ARC4. An algorithmic scoring system was used to predict novel Tankyrase binding proteins from which new substrates have been validated. This binding motif has led to the identification of numerous Tankyrase binding proteins (Table 1.2).

Our laboratory has demonstrated that the adapter protein 3BP2 is a TNKS substrate. Mutation in the 3BP2 hexapeptide sequence occurs in the rare autosomal dominant disorder Cherubism and uncouples it from TNKS binding and subsequent PARylation. We have shown that the hexapeptide sequence is both necessary and sufficient to bind to the ARC domains of TNKS. Moreover, we demonstrated that ARC1,2, 4, and 5 are competent to bind to the hexapeptide motif and that ARC3 appears to have a structural non-peptide binding function. We have shown that mutations in the R, P or G of the Cherubism hexapeptide sequence uncouples 3BP2 from

TNKS binding. Crystal structure of the 3BP2 hexapeptide peptides bind within a pre-existing groove in the ARCs where the 1st position arginine is engaged by an “arginine cradle” which is formed by two salt bridges by aspartic and glutamic acid residues in Tankyrase and the 6th position glycine fits into a “glycine sandwich” which is the result of two parallel tyrosine residues (Figure 1.2A) (Guettler et al., 2011).

Axin1 and Axin2 are Tankyrase substrates identified through a small molecule screen for Wnt signaling inhibitors (Huang et al., 2009). The crystal structure of Tankyrase ARC2/ARC3 and Tankyrase binding region of Axin1 revealed two distinct binding motifs that bound simultaneously to each ARC2 domain within the ARC2/ARC3 homodimer (Morrone et al., 2012). Interestingly the Axin1 crystal structure revealed a unique mechanism by which substrates can bind Tankyrases. Instead of the traditional RxxPDG motif, the more C-terminal Tankyrase binding motif (TBM) in Axin1 has an extended motif where the 1st position arginine is anchored into the “arginine cradle” and is 9 residues upstream from the remaining peptide sequence that fits into the ARC binding pocket (Figure 1.2B). Another crystal structure showing the complex of Tankyrase and its E3 ligase RNF146 reveals another variation of the TBM. The RNF146 motifs have 1 or 2 residues that extend out of the ARC substrate pocket while the remaining residues reside within similarly to Axin1 (DaRosa et al., 2018). The variability in TNKS binding motifs has significantly expanded the number of substrates it can regulate and accordingly the complexity of TNKS biology (Table 1.2).

Figure 1.2: Crystal Structure of ARC4 with Tankyrase binding motif peptides: (A) Space- filling model of Tankyrase 2 ARC4 with Tankyrase binding motif of 3BP2. (B) Space-filling model of Tankyrase 2 ARC2 with Tankyrase binding motifs of Axin1. Numbers indicate positioning of each amino acid of the motif within substrate binding pocket of ARC4. Adapted from Guettler et al., 2011 and Morrone et al., 2012.

Table 1.2: List of identified Tankyrase interacting proteins Canonical Binding Motif PARylated ID Method Reference Binders TRF1 RGCADG Yes Y2H Smith et al., 1998 Grb14 Not determined Not determined Y2H Lyons et al., 2001 IRAP RQSPDG Yes Y2H Chi and Lodish 2000 NuMa1 RTQPDG Yes Y2H Sbodio and Chi, 2000 Mcl1 RPPPIG No Y2H Bae et al., 2003 TAB182 RPQPDG Yes Y2H Seimiya et al., 2004 3BP2 RSPPDG Yes Y2H Levaot et al., 2011 BLZF1 RGAGDG Not determined IP-MS Zhang et al., 2011 CASC3 RQSGDG Not determined IP-MS Zhang et al., 2011 Disc1 RGEAEG Yes TBM conservation Guettler et al., 2011 Striatin RSAGDG Yes Fat4 RKQPEG/RNPADG Yes RAD54 RPPPDG No BCR RPDGEG Yes Merit40 RSNPEG/RSEGEG Weakly CPAP REYPDG Yes TBM conservation Kim et al., 2012 Miki Not determined Yes Literature search Ozaki et al., 2012 GMD RGSGDG No IP-MS Bisht et al., 2012 PTEN RYQEDG Yes TBM conservation Li et al., 2015 AMOT RQEPQG Yes IP-MS; inhibitor, Wang et al., 2015 AMOTL1 RQEPQG Yes siRNA and Campbell et al., 2016 AMOTL2 RQEPQG Yes CRISPR screens Wang et al., 2016 USP25 RTPADG No IP-MS Xu et al., 2017 PEX14 RMEVQG/RRGGDG Yes IP-MS Li et al., 2017 ATG9 RLPGLG Not determined IP-MS Li et al., 2017 PrxII RLSEDYG No Literature search Kang et al., 2017 ABRO1 RPQAVG Not determined IP-MS Tripathi and Smith, 2017 NKD1 RESPEG Not determined MS Bhardwaj et al., 2017 NKD2 RESPEG Not determined Notch2 RREPVG Yes MS Bhardwaj et al., 2017 Non- Binding Motif PARylated ID Method Reference Canonical Binders Axin1 RPPVPG/ Yes Inhibitor Huang et al., 2009 Axin2 RRSDLDLGYEPEG Yes Screen Morrone et al., 2011 RPPVPG/ RRNEDGLGEPEG RNF146 RESSADG No IP/ Crystal Structure DaRosa et al., 2018 RPLTSVDG RSHRGEG RSRRPDG RSVAGG NELFE RELGPDG Not determined Sequence homology DaRosa et al., 2018 I4FA1 RDNGPDG Not determined Sequence homology DaRosa et al., 2018 FIH None No IP-MS Cockman et al., 2009

1.4.3 Tankyrases oligomerize and form higher order assemblies

Tankyrase 1 can bind to Tankyrase 2 to form heterodimers and co-localize in punctate structures within the cytoplasmic compartment of cells (Sbodio et al., 2002). Both proteins contain a SAM domain, which enables protein to undergo homo- and heterotypic interactions and suggests that Tankyrases can form large protein assemblies. To test this hypothesis, cross-linking of the SAM domain of either Tankyrase 1 or 2 revealed their ability to form large oligomers (De Rycker et al., 2003). To eliminate the possibility that cross-linking could artificially cause oligomerization, recombinant TNKS SAM domain was purified by gel filtration and revealed that stable polymers of the SAM domain could form without cross-linking (De Rycker and Price 2004). These polymers formed rod shaped structures when image by electron microscopy. Full length Tankyrase could also oligomerize and maintain its ability to bind to TRF1 (De Rycker and Price 2004). The authors also showed that Tankyrase auto-PARylation triggered the disassembly these polymers and suggest that Tanykrase can act as large scaffolding proteins that facilitate reversible protein complexes. These findings were followed up by additional publications showing crystal structures of the SAM domain which demonstrated the ability of these enzymes to form multivalent scaffolds for substrates in both homo and hetero-oligomer configurations. These structures also identified critical residues within the SAM domain that mediated Tankyrase oligomerization and showed that SAM domain mediated Tankyrase oligomerization resulted in the large vesicular distribution of Tankyrase in cells (DaRosa et al., 2016; Mariotti et al., 2016; Riccio et al., 2016).

1.4.4 Role of Tankyrases at human telomeres

Tankyrases positively regulate human telomere length through interaction with TRF1, which acts as an inhibitor of telomerase by binding telomere ends. Tankyrase binding to TRF1 results in its PARylation, which reduces the affinity of TRF1 binding to telomere DNA (Smith et al., 1998; Cook et al., 2002). Partial knockdown of human Tankyrase 1 caused slight telomere shortening while overexpression completely excluded TRF1 from telomeres (Donigian and de Lange 2007). Neither knockout of Tankyrase 1 nor Tankyrase 2 was able to affect telomere length but double knockouts in human cells had shortened telomeres (Bhardwaj et al., 2017). Interestingly, mouse telomere length is unaffected by Tankyrases since TRF1 lacks the ‘RxxPDG’ binding motif (Donigian and de Lange 2007). Tankyrase is also important for the separation of sister 8

chromatids during mitosis and particularly for the separation at telomere ends (Dynek and Smith 2004). The molecular mechanism of this process has been characterized where the E3 ligase RNF8 confers K63 ubiquitylation of Tankyrase 1 which stabilizes the protein and promotes binding to telomeres during late S phase and G2 phase through mitosis. ABRO1, part of the deubiquitinase complex BRISC, removes the K63 ubiquitin chains during G1 phase to prevent early separation of sister chromatids (Tripathi and Smith, 2016).

1.4.5 Tankyrase in mitosis

Tankyrase 1 is important for bipolar spindle formation during mitosis and is recruited there through its interaction with NuMa (Chang et al., 2005; Chang et al., 2005). Interestingly, knockdown of Tankyrase 1 can lead to mitotic arrest caused by misalignment of mitotic spindles and failure of sister chromatids to separate (Dynek and Smith 2004). Single knockouts of each enzyme also resulted in similar frequency of cells with defective spindles while double knockout cells frequency was even higher and had increased multipolar spindles (Bhardwaj et al., 2017). The requirement of both Tankyrases to rescue the mitotic defects shows that there is a non- redundant function of Tankyrase during mitosis. Tankyrase also regulates centrosome function during cell cycle progression through modulation of centrosome marker CPAP protein levels. Tankyrase 1 knockdown leads to increased stabilization of CPAP causing centrosome duplications and multipolar spindles (Kim et al., 2012). Miki, a Golgi-associated protein, colocalizes at centrosomes during prometaphase following PARylation by Tankyrase 1. Tankyrase 1 and PARylated Miki function together to ensure centrosome maturation during prometaphase to facilitate chromosome segregation during mitosis (Ozaki et al., 2012).

1.4.6 Tankyrase controls glucose metabolism

Tankyrase has been linked to glucose metabolism through its association with IRAP and becomes phosphorylated and highly active upon insulin stimulation (Chi and Lodish 2000). Tankyrase 1 promotes glucose uptake since knockdown inhibits insulin-stimulated glucose uptake and is reversed once Tankyrase protein levels are restored (Yeh et al., 2007). Tankyrase impairs the localization of GLUT4 and IRAP at the plasma membrane, which disrupts glucose uptake (Yeh et al., 2007). Tankyrase 1 deficient mice display increased energy consumption, glucose consumption and muscle insulin sensitivity yet their primary adipocytes have increased

GLUT4 transcription but normal localization (Yeh et al., 2009). Mice with specific deletion of the PARP domain of Tankyrase 1 in adipocytes show a similar phenotype but do not exhibit muscle insulin sensitivity (Zhong, L et al., 2016). In addition, treatment of mice with a Tankyrase specific inhibitor, G007-LK, increased glucose tolerance and insulin sensitivity suggesting that PARylation is involved in regulating glucose metabolism. Tankyrase inhibitors may have therapeutic utility to improve glucose tolerance (Zhong et al., 2016).

1.4.7 Role of Tankyrase in WNT signaling and cancer

Tankyrase positively regulates WNT signaling by downregulating protein levels of Axin which results in β-catenin stabilization. Pharmacological inhibition of Tankyrase ablates this effect through stabilization of Axin and β-catenin degradation (Huang et al., 2009). Tankyrase inhibition showed promise for WNT driven cancers such as β-catenin-dependent colorectal cancer, breast cancer, hepatocellular carcinoma, osteosarcoma and demonstrated synergy with EGFR inhibitors in non-small cell lung cancer. This has led to the development of several Tankyrase inhibitors by several biotechnology companies which have been shown to block cancer cell proliferation in vitro and tumor growth in vivo (Huang et al., 2009; Bao et al., 2012; Lau et al., 2013; Stratford et al., 2014; Ma et al., 2015; Arqués et al., 2016; Okada-Iwasaki et al., 2016; Scarborough et al., 2017; Gustafson et al., 2017; Mizutani et al., 2018; Martins-Neves et al., 2018). A new inhibitor was discovered that inhibits Wnt dependent signaling in prostate cancer. Here the authors describe a small molecule, C44, that blocks the interaction between Tankyrase and USP25 which promotes Tankyrase degradation and stabilizes Axin protein levels. This in turn decreases β-catenin levels and attenuates signaling that leads to decreased prostate cancer cell line proliferation, colony formation and tumor xenograft growth (Cheng et al., 2019). The vast number of studies indicates the great interest towards the promise of Tankyrases as a novel therapeutic agent against cancer.

1.4.8 Tankyrase promotes tumour growth independently of WNT signaling

Previous work identified Axin1 as a Tankyrase substrate and XAV939, a Tankyrase inhibitor, inhibited proliferation and tumor growth of β-catenin dependent cell lines (Huang et al., 2009). Follow up investigation in colon cancer cells that were either deficient in β-catenin or had a degradation resistant mutation led to the identification of PTEN as a novel Tankyrase substrate.

In both β-catenin dependent and independent cells, Tankyrase inhibition increased PTEN protein levels and decreased Akt phosphorylation, which resulted in decreased colon cancer cell proliferation and tumor growth in a PTEN dependent manner (Li et al., 2015). Comparison of protein expression of Tankyrases with PTEN between normal and tumor samples revealed Tankyrase up-regulation correlated with lower PTEN expression (Li et al., 2015).

1.4.9 Tankyrase regulates bone metabolism and dysregulation of 3BP2 is the underlying cause of Cherubism

Our lab discovered the pathogenic mechanism underlying the human genetic disease Cherubism. Cherubism arises from single missense mutations that are located within the ‘RxxPDG’ Tankyrase binding motif of the adapter protein 3BP2 (Ueki et al., 2001; Levaot et al., 2011). These mutations uncouple 3BP2 from Tankyrase regulation leading to its stabilization within myeloid lineage cells, which results in enhanced osteoclastogenesis, bone loss and cytokine production. In addition, Tankyrase inhibitors increase osteoclastogenesis in vitro as a result of increased steady-state expression of 3BP2. Elevated levels of 3BP2 in turn lead to increased activation of Src and Syk (Levaot et al., 2011). Similarly, further studies have shown that Tankyrase inhibition in mice leads to increased bone loss due to increased osteoclast numbers and poses a significant side-effect if used clinically (Fujita et al., 2018). Knockout studies of 3BP2 show defects in osteoblasts as well demonstrating a dominant role of 3BP2 and Tankyrase in osteoblasts.

1.4.10 Tankyrase promotes YAP signaling and cell polarity

Tankyrases positively regulate the hippo pathway through YAP by targeted degradation of the angiomotin proteins AMOT, AMOTL1, AMOTL2. Angiomotins sequester YAP in the cytoplasm which inhibits YAP’s oncogenic activity and Tankyrase inhibitors help maintain this inhibition by angiomotin stabilization (Wang et al., 2015; Campbell et al., 2016; Wang et al., 2016). Tankyrase inhibition of YAP synergizes with EGFR inhibition and increases the potency of Erlotinib to inhibit proliferation of non-small cell lung cancer cells and similarly with MEK or AKT inhibition can inhibit hepatocellular carcinoma cell proliferation (Wang et al., 2016; Jia et al., 2017). Overexpression of angiomotin has previously been shown to be important for epithelial polarity by regulating the localization of PALS1 at the apical membrane, which is

required for Crumbs complex assembly (Wells et al., 2006). Tankyrase has been shown to be a critical factor for tight junction integrity by regulating AMOTL2 stability, which alters PALS1 localization from tight junctions to internal puncta (Campbell et al., 2016).

1.5 Regulation of Tankyrase

1.5.1 Factors that regulate catalytic activity

Several mechanisms that regulate the stability and activity of Tankyrase have been described. Tankyrase 1 phosphorylation by MAPK and Plk1 has been shown to increase PARP activity under insulin stimulation and during mitosis (Chi and Lodish 2000; Ha et al., 2012). Since Tankyrase requires NAD+ as a co-factor to generate PAR, it was hypothesized that global levels of NAD could regulate its activity. ATP is required to create NAD during oxidative phosphorylation in the mitochondria. Cells with elevated NAD+ levels manifested increased Tankyrase enzymatic activity (Zhong et al., 2015). This finding suggested that cellular energy states could control Tankyrase PARylation dependent processes. The SAM domain has also been shown to be a necessary component for PARP activity. Point mutations in the SAM domain that disrupt Tankyrase oligomerization lead to decreased Tankyrase PARylation and consequent decrease in WNT signaling (Mariotti et al., 2016; Riccio et al., 2016). Most recently, Tankyrase PARP activity can be inhibited by oxidation upon exposure to hydrogen peroxide. The antioxidant protein, peroxiredoxin II, binds to Tankyrase and protects it from oxidation-mediated inactivation (Kang et al., 2017). In APC mutant colorectal cancer, excess hydrogen peroxide is produced and peroxiredoxin II promotes WNT signaling by maintaining Tankyrase activity. Inhibition or loss of peroxiredoxin in a colon xenograft cancer results in Tankyrase inactivation, Axin accumulation and degradation of β-catenin and inhibition of tumor growth (Kang et al., 2017). There are therefore multiple regulatory mechanisms that control cellular Tankyrase activity.

1.5.2 Protein-protein interactions regulate Tankyrase stability

Modulation of Tankyrase activity by its binding partners has also been reported. GDP-Mannose- 4,6-Dehydratase (GMD) binds through a single TBM and inhibits catalytic activity which results in Tankyrase protein stabilization (Bisht et al., 2012). Interestingly, GMD is not PARylated by Tankyrase and the authors suggest that GMD sequesters Tankyrase-1 in the cytoplasm away 12

from its substrates. The deubiquitinating enzyme USP25 binds to Tankyrase through a TBM, which antagonizes RNF146 mediated ubiquitylation and proteasomal degradation of Tankyrase. USP25 is thus a positive regulator of WNT signaling through Axin by stabilizing Tankyrase steady state protein levels (Xu et al., 2017).

1.5.3 Tankyrase stability is regulated by PAR dependent Ubiquitylation (PARdU)

Tankyrase protein abundance is regulated by the WWE domain of the E3 ligase RNF146 through a novel proteasomal degradation pathway called PARylation dependent ubiquitylation or PARdU (Zhang et al., 2011; Callow et al., 2011; DaRosa et al., 2014). PARdU is initiated by the recognition of the iso-ADP-ribose moiety that is found within PAR chains by the WWE domain of the RING E3 ligase, RNF146 (Wang et al., 2012; DaRosa et al., 2014). Engagement of the WWE domain to ADP-ribose causes a conformational change in the RING domain of RNF146, which facilitates ubiquitylation of substrates by lysine 48 linkages. RNF146, Tankyrase and the substrate form a degradation complex that is destroyed by the proteasome (Figure 1.3). This moiety has been shown to bind to other WWE containing E3 ligases, which include HECT domain proteins HUWE1 and TRIP12 and the RING domain containing DTX1, at varying affinities (Wang et al., 2012). Additionally, DTX2 and DTX4 are the other known RING domain E3’s to contain a WWE domain.

Figure 1.3: Tankyrases and substrates are degraded through PARylation dependent Ubiquitylation (PARdU). Tankyrase binding and PARylation of substrates engages RNF146 binding to iso-ADP-ribose by its WWE domain. This induces a conformational change in the RING domain which facilities ubiquitylation of the entire complex and targets all proteins for proteasomal degradation. Orange oval represents any Tankyrase substrate. White oval and green parallelogram represent the RING and WWE domains of RNF146 respectively. Red circles are PAR chains while yellow diamonds are lysine-48 ubiquitin linkages.

1.6 RNF146

To date, RNF146 is the only demonstrated Tankyrase E3 ligase. It remains unclear whether other WWE domain E3 ligases or other E3 ligases can regulate PARdU. RNF146 in concert with Tankyrase regulate WNT signaling, bone metabolism, Akt signaling, YAP signaling and cell polarity through its substrates Axin, 3BP2, PTEN and Angiomotins respectively (Zhang et al., 2011; Callow et al., 2011; Levaot et al., 2011; Li et al., 2014; Wang et al., 2015; Campbell et al., 2016; Wang et al., 2016; Matsumoto et al., 2017).

1.6.1 RNF146 regulates DNA damage and PAR-mediated cell death

RNF146 has been described to regulate PARylation dependant processes that are independent of Tankyrases including PAR-dependent cell death (parthanatos), DNA damage and DNA repair (Andrabi et al., 2011; Kang et al., 2011). RNF146 is neuroprotective against N-methyl-D- aspartate (NMDA) induced excitotoxicity and acts as a survival gene that becomes upregulated upon NMDA stimulation (Andrabi et al., 2011). During parthanatos, RNF146 binding to PAR affects translocation of apoptosis inducing factor (AIF) from mitochondria to the nucleus and reduces cell death (Wang et al., 2009; Andrabi et al., 2011). Proteomic studies searching for novel RNF146 interactors have identified DNA damage and DNA repair related proteins. DNA damage response proteins PARP1 and PARP2 and DNA repair pathway proteins Ku70, Ku86, LIG3 and XRCC1 were identified as RNF146 interactors. DNA damage triggers PARP1 degradation by RNF146 and protects cells from cell death and enhances DNA repair (Kang et al., 2011). It is still unknown whether RNF146 can function independently of PARylation and perhaps behave as a scaffold molecule.

1.6.2 RNF146 regulates bone metabolism

RNF146 plays a major role in bone dynamics as it regulates the differentiation and function of both osteoclasts and osteoblasts. Our lab has shown that RNF146 is required for osteoclast differentiation. RNF146 suppresses osteoclastogenesis through degradation of its substrates 3BP2 and Axin1, which in turn regulate SRC activation and β-catenin activity. Degradation of 3BP2 lowers SRC activation, which is required for osteoclast differentiation and Axin1 degradation increases β-catenin transcriptional activity which suppresses osteoclast differentiation. RANKL, a critical osteoclast differentiation factor suppresses RNF146 transcription and protein expression which potentiates osteoclastogenesis (Matsumoto et al., 2017). RNF146 is also important in osteoblast differentiation through its regulation of AXIN1 and β-catenin. Activated β-catenin promotes osteoblast differentiation while simultaneously repressing adipocyte differentiation. Loss of RNF146 leads to increased adipogenesis and suppression of osteoblastogenesis (Matsumoto et al., 2017). Collectively, these findings emphasize the importance of RNF146 as a switch that governs bone dynamics through regulation of both osteoblast and osteoclast differentiation. RNF146 knockout mice are also embryonic

lethal which reiterates the importance of this protein in regulating multiple biological functions through the diversity of its substrates (Matsumoto et al., 2017).

1.7 SH3BP5

1.7.1 Role of SH3BP5 in stress kinase signaling

SH3 Binding Protein 5 (SH3BP5) is an adaptor protein originally identified as a direct interactor with the SH3-domain of Bruton’s Tyrosine Kinase (BTK) through a yeast two-hybrid screen (Matsushita et al., 1998). By sequence homology, SH3BP5L has been identified as an SH3BP5 paralog (Figure 1.4). SH3BP5 was shown to be a unique interactor among BTK related kinases and inhibited its kinase activity (Yamadori et al., 1999).

Figure 1.4: Domain organization of SH3BP5 and SH3BP5L. Graphical representation of domains and structured regions within SH3BP5 and SH3BP5L. BTK-Binding - region required for BTK-SH3 domain binding; KIM - Kinase interaction motif for JNK; Green shading – alpha helical region.

Another yeast two-hybrid screen with c-Jun N-terminal Kinase (JNK) revealed SH3BP5 as an interactor and this binding was facilitated through two kinase interacting motifs (KIM) within its C-terminus. SH3BP5 was also shown to be a substrate of JNK and other JNK family kinases in subsequent studies. SH3BP5 localized with mitochondria by immunofluorescence analysis (Wiltshire et al., 2002). Further investigation have implicated a scaffolding role for SH3BP5 to recruit JNK to the mitochondrial outer membrane where JNK can respond to a variety of cell stresses including endoplasmic reticulum (ER) stress, anisomycin treatment, acetaminophen and tumor necrosis factor (TNF) induced-toxicity (Win et al., 2011; Win et al., 2014). ER stress induced by tunicamycin or brefeldin A can trigger apoptosis through release of cytochrome c and

caspases from the mitochondria and produce reactive oxygen species (ROS). Similarly, anisomycin, an antibiotic that inhibits protein synthesis, can induce cell death by apoptosis and result in increased mitochondrial ROS production. JNK phosphorylation increases under each of these stress conditions and mediates both ROS production and apoptosis. In both these instances, SH3BP5 knockdown led to the decreased localization and activation of JNK within mitochondria of cells and resulted in decreased ROS production and cell death (Chambers and LoGrasso 2011, Win et al., 2011; Win et al., 2014).

Increased doses of acetaminophen (APAP) leads to liver toxicity in vivo and in hepatocytes which is the result of sustained JNK activation and inhibition of JNK has a significant protective effect (Gunawan et al., 2006). APAP treatment leads to JNK activation, its recruitment to the mitochondria, increased ROS production and cytochrome c release which results in apoptotic cell death (Hanawa et al., 2008). SH3BP5 is the required factor that recruits JNK to the mitochondria and facilitates the sustained JNK activation seen during APAP and TNF induced hepatotoxicity as SH3BP5 knockdown significantly ablates these effects (Win et al., 2011). These findings were tested in mouse liver injury models after treatment with either APAP or TNF. Mice were treated with short hairpins targeting SH3BP5 before dosing with APAP or TNF and in both cases JNK was unable to localize to the mitochondria and sustain JNK activation and was protective against liver toxicity (Win et al., 2014). A subsequent studied characterized the molecular mechanism of this process by utilizing a conditional knockout system where sh3bp5 was genetically excised from the liver. The authors discovered that SH3BP5 binding and phosphorylation by JNK leads to dephosphorylation of SRC at tyrosine 419 by the phosphatase SHP1, which increases ROS production and creates a positive feedback loop that maintains JNK activation (Win et al., 2016).

JNK is highly expressed and active in the brain and the isoforms JNK1, JNK2, and JNK3 are expressed there as well (Gupta et al., 1996). To address whether SH3BP5 has a role in regulating JNK activity in the brain, a study looked at the expression of SH3BP5 in the brain and found it to be expressed at varying levels throughout the adult mouse brain. SH3PB5 was highly expressed in the hippocampus, cerebellum and ventral midbrain (Sodero et al., 2017). The authors showed that SH3BP5 localizes to mitochondria within neurons of the hippocampus. Additionally,

SH3BP5 is important for neuron function as knockdown by shRNA resulted in decreased firing and amplitude of cultured neurons (Sodero et al., 2017).

1.7.2 SH3BP5 is a guanine nucleotide exchange factor for Rab11

A yeast two-hybrid screen in C. elegans identified REI-1 and REI-2 as preferential binders to the GDP-locked form of Rab11a (Sakaguchi et al., 2015). These proteins are homologous to mammalian SH3BP5 and SH3BP5L and the authors demonstrated that human and mouse SH3BP5 are able to bind to Rab11a. The authors revealed that REI-1, REI-2 and SH3BP5 facilitated the exchange of Rab11a-GDP to Rab11a-GTP and had specific activity towards Rab11a in comparison to Rab5 (Sakaguchi et al., 2015). Similarly, another GEF for Rab11 and SH3BP5 ortholog was recently discovered in D. melanogaster called parcas (Riedel et al., 2018). Sakaguchi et al narrowed the binding region to the N-terminus of the SH3BP5 family of proteins but did not provide specificity beyond that (Figure 1.5) (Sakaguchi et al., 2015). A recent crystal structure of SH3BP5 showed that two α-helices in the N-terminus form the interface with Rab11a (Jenkins et al., 2018). The authors demonstrated that SH3BP5 and SH3BP5L facilitate nucleotide exchange exclusively for Rab11 proteins. The helical nature of the GEF domain shares homology with the Rab8 GEFs, Rabin8 and GRAB (Guo et al., 2013). In distinction to SH3BP5, the catalytic domains of both Rabin8 and GRAB are dimeric coiled-coiled folds and form a parallel interface with Rab11 while SH3BP5 is a monomeric two-helix fold that forms a perpendicular interface with Rab11 (Guo et al., 2013; Jenkins et al., 2018).

Figure 1.5: Conservation of SH3BP5 family across organisms. Graphical representation of domains and structured regions within SH3BP5 family of proteins. BTK-Binding- region required for BTK-SH3 binding; KIM - Kinase interaction motif for JNK; Green shading – alpha helical region that mediates Rab11a interaction.

1.7.3 SH3BP5 in cancer

There are no well-characterized associations between SH3BP5 mutations, deletions or amplifications with genetic disease. However, several studies have shown a correlation between SH3BP5 expression and cancer. In a study of gene expression in CD5, a T-cell surface marker, positive diffuse large B-cell lymphoma (DLBCL) which is associated with poor clinical outcome, SH3BP5 was one of the highest expressed genes that correlated with CD5 positive DLBLC samples in comparison to CD5 negative samples (Miyazaki et al., 2015). Immunohistochemistry of SH3BP5 was performed in CD5 positive, CD5 negative and activated B-cell like (ABC) DLBCL samples and SH3BP5 protein expression correlated with CD5 positive and ABC samples. In addition, SH3BP5 positive DLBCL patients correlated with advanced stage disease and poorer overall survival. The authors concluded that SH3BP5 expression could be used as a prognostic marker of the aggressiveness of DLBCL (Kobayashi et al., 2016).

Examination of SH3BP5 gene expression in ovarian cancer samples revealed that on average it has a five-fold decrease in expression when compared to normal samples (Paudel et al., 2018). Furthermore, expression level of SH3BP5 dictates the capacity of ovarian cancer cells to undergo apoptosis upon cisplatin or paclitaxel treatments. The authors demonstrated that the SKOV3 cell line which has low SH3BP5 expression is more chemoresistant and ectopic expression could increase chemosensitivity to platinum agents. Conversely, PA-1 an ovarian cancer cell line which expresses high levels of SH3BP5 is chemosensitive and became chemoresistant upon SH3BP5 knockdown. These results suggest that SH3BP5 signaling at the mitochondrial membrane affects the concentration of pro-apoptotic proteins at the membrane and boosting SH3BP5 signaling could be a useful strategy to target chemoresistant ovarian cancer (Paudel et al., 2018).

1.8 Rab11 GTPases

1.8.1 Rab11 family

The Rab11 family of proteins belongs to the Rab class of the small GTPase superfamily, which consists of three distinct genes known as Rab11a, Rab11b, and Rab25 (Rab11c). These proteins are localized on recycling endosomes and regulate late recycling of cargo through these vesicles 20

(Welz et al., 2014). Like other GTPases, Rab proteins cycle between GDP and GTP and the GTP bound form is necessary for downstream function. The nucleotide states are regulated by activating guanine nucleotide exchange factors (GEFs) and deactivating GTPase activating proteins (GAPs) where GEFs facilitate GDP to GTP exchange and GAPs regulate disassociation of GTP to GDP (Ullrich et al., 1996; Schlierf et al., 2000). Rab11 proteins are targeted to vesicle membranes by geranylgeranylation on two cysteine residues located in their C-terminus, which is catalyzed by geranylgeranyltransferase II (Pfeffer and Aivazian 2004). Rab11 proteins facilitate several biological processes including cytokinesis, neurite formation, and ciliogenesis (Wilson et al., 2005; Shirane and Nakayama 2006; Knödler et al., 2010). Thus far, several proteins have shown to have GEF activity towards Rab11 including D. melanogaster protein Crag, TRAPPII complex; C. elegans proteins REI-1 and REI2; and mammalian proteins SH3BP5 and SH3BP5L (Xiong et al., 2012; Riedel et al., 2018; Sakaguchi et al., 2015; Jenkins et al., 2018). The D. melanogaster protein Evi5 has been shown to act as a GAP for Rab11 (Laflamme et al., 2012).

1.8.2 Rab11 function at recycling endosomes

Rab11 proteins are resident markers of recycling endosomes and facilitate recycling of proteins to the plasma membrane including transferrin receptor (TfR), β1 integrin, epidermal growth factor receptor (EGFR), α5β1 integrin, GLUT4, TLR4 and α-amino-3-hydroxy-5- methylisoxaole-4-propionic acid receptor (AMPAR) (Ullrich et al., 1996; Powelka et al., 2004; Palmieri et al., 2006; Caswell et al., 2007; Welsh et al., 2007; Husebye et al., 2010; Kelly et al., 2011). Recycling by Rab11 proteins is facilitated through interaction with its effector proteins called Rab11-family interacting proteins (Rab11-FIPs). This interaction is mediated by a Rab11 binding domain (RBD), which is a 20 amino acid motif that binds to Rab11-GTP (Prekeris et al., 2001). Each Rab11-FIP binds unique signaling proteins that mediate distinct pathways and downstream processes.

1.8.3 Rab11 in epithelial polarity

Rab11 regulates apical recycling endosomes and apical transport in polarized epithelial cells and facilitates trafficking of polymeric IgA receptor, E-cadherin and Podocalyxin (PODXL) (Lock and Stow 2005; Desclozeaux et al., 2008; Prekeris et al., 2000; Bryant et al., 2010). Rab11-GTP

is required for proper basolateral membrane targeting of E-cadherin in both polarized monolayers as well as in cysts during lumen formation (Lock and Stow 2005; Desclozeaux et al., 2008). Furthermore, in a screen for Rabs required for apical transport of PODXL, a sialoglycoprotein necessary for single lumen formation, Rab11a and Rab25 but not Rab11b are necessary for single lumen formation (Bryant et al., 2010; Mrozowska and Fukuda 2016). PODXL undergoes a polarity inversion by being trafficked from the basolateral membrane to the apical membrane in Rab11a-positive endosomes and active Rab11a is required for this process (Bryant et al., 2010). Rab11-mediated trafficking is cell-type dependent and the diversity of downstream effectors and upstream regulators likely work in tandem to orchestrate protein trafficking.

1.8.4 Rab11 in disease

Rab11b mutations have been described in individuals with neurodevelopmental symptoms and both mutations V22M and A68T alter GDP/GTP binding. In particular, V22M mutant Rab11b has high affinity to SH3BP5 and results in mislocalization (Lamers et al., 2017). Mutations in SH3TC2, an adaptor-like protein are associated with the nerve disorder Charcot-Marie-Tooth disease type 4C, disrupt interaction with Rab11-GTP and cause mislocalization of SH3TC2 by Rab11 recycling endosomes (Roberts et al., 2010). Both Rab11a and Rab11b have additionally been implicated in playing roles in Alzheimer’s and Huntington disease (Kelly et al., 2012). Rab25 (a.k.a. Rab11c) expression is associated with increased tumor aggressiveness and lower overall survival in ovarian and breast cancer (Cheng et al., 2004). Rab25 has an inherit mutation Q71L that is similar to known oncogenic Ras activating mutations and suggests Rab25 is constitutively locked in its GTP active form (Goldenring et al., 1993; Kessler et al., 2012). In ovarian cancer Rab25 regulates the recycling of integrin α5β1, a known promoter of tumor cell survival and metastasis, maintains it at the tips of pseudopodia, which promotes cancer cell invasion and migration (Caswell et al., 2007).

1.8.5 Rab11 in host-pathogen relations

Several bacterial pathogens and viruses subvert Rab11 regulated processes in order to invade and escape host cells (Guichard et al., 2014). Both Chlamydia trachomatis and Chlamydia pneumoniae replication inclusions interact with Rab11a and silencing of Rab11a decreases the

number of infectious particles (Rzomp et al., 2003; Rejman Lipinski et al., 2009). Additionally, bacterial toxins from Bacillus anthracis and Vibrio cholerae inhibit both Rab11-mediated endocytic recycling and affect cell-cell adhesion of epithelial cells (Guichard et al., 2014). Viruses also utilize Rab11-positive vesicles to help with releasing replicated particles from cells. Andes virus preferentially binds to Rab11-positive vesicles within the Golgi compartment after virus assembly to help with transport to the cell surface. Respiratory syncytial virus, HIV-1 and Mason-Pfizer monkey virus all require Rab11-positive apical recycling vesicle to transport viral components to the apical surface for particle assembly (Guichard et al., 2014). Rab11 family members thus regulate normal vesicle trafficking critical for normal cell function but have been targeted by during evolution to mediate pathogen infection.

1.9 Factor Inhibiting HIF1α (FIH) and the regulation of Tankyrase Function

The asparaginyl hydroxylase, factor inhibiting HIF-1α (FIH) was originally identified as a binding partner to HIF-1α and repressed its transcriptional activation and is a member of the JmjC class of dioxygenases (Mahon et al., 2001). FIH-mediated asparaginyl hydroxylation of Asn-803 within the HIF-1α C-terminal transactivation domain (CAD) blocks the interaction between transcriptional coactivators p300/CBP (CREB-binding protein) and HIF-1α (Lando et al., 2002; Lando et al., 2002; Hewitson et al., 2002). Hypoxia suppresses HIF-1α prolyl and asparaginyl hydroxylation, which results in increased protein stability and transcriptional activity (Lando et al., 2002). Mass spectrometry based proteomic approaches have identified many FIH binding proteins including Tankyrase1/2, Notch, IκB, ASB4, Rabankyrin-5, RNase L, AnkyrinR, TRPV3, and ASPP2 which are subject to asparaginyl hydroxylation (Cockman et al., 2006; Coleman et al., 2007; Ferguson 3rd et al., 2007; Cockman et al., 2009; Yang et al., 2011; Janke et al., 2013; Karttunen et al., 2015). Additionally, FIH has been shown to mediate histidinyl hydroxylation on Tankyrase-2 and aspartate hydroxylation on cytoskeleton ankyrin proteins, ankyrinR and ankyrinB (Yang et al., 2011; Yang et al., 2011). Chapter 4 elucidates the role of FIH in controlling Tankyrase function. The following sections summarize what is presently known about FIH function and regulation and form the basis of many of the experiments conducted in the context of FIH control of Tankyrase activity.

1.9.1 Regulation of FIH catalytic activity

The crystal structure of FIH revealed it forms a homodimer to mediate catalysis under basal conditions. Leucine 340 forms an important hydrophobic interaction between FIH subunits and a critical mutation (L340R) disrupts dimerization and FIH activity (Lancaster et al., 2004). An alanine mutation to aspartic acid 201 (D201A) within the catalytic site failed to inhibit expression of the HIF target genes GLUT1, VEGF and LDH-A (Stolze et al., 2004). Phosphorylation at threonine 796 on HIF blocks FIH binding and promotes interaction with the coactivators p300/CBP (Lancaster et al., 2004). FIH activity can also be regulated by peroxide as treatment with H2O2 can block hydroxylation of the HIF-CAD as well as ankyrin repeats without affecting FIH protein stability (Masson et al., 2012).

1.9.2 Hydroxylation of ankyrin-repeat containing proteins

Multiple studies have revealed that FIH has a large variety of substrates in addition to its well- known target HIFα. Thus far, most additional substrates and interactors contain ankyrin repeat domains (ARD). The deubiquitinase OTUB1 a non-ankyrin repeat containing protein is an FIH substrate (Scholz et al., 2016). The biological significance of these modifications on protein function has been elucidated. It has been suggested that ankyrin repeat hydroxylation can act as a natural inhibitor towards HIF hydroxylation as FIH has a higher affinity for ARD than the HIF- CAD (Coleman et al., 2007). A common paradigm has emerged where FIH hydroxylation of proteins generally does not influence protein stability but can have an inhibitory effect on enzymatic or adaptor-like functions.

1.9.3 Hydroxylation affects protein catalytic function

There are two examples in the literature demonstrating that FIH hydroxylation can affect enzymatic activity of proteins. The ion channel TRPV3 is subject to asparagine hydroxylation which downmodulates current amplitude within the channel. In all cases where FIH is inhibited including hypoxia, DMOG and FIH catalytic mutant, current amplitude was increase in comparison to normal conditions (Karttunen et al., 2015) This study marked the first time where there was clear evidence showing FIH-mediated hydroxylation inhibiting substrate activity. Notch proteins, a family of four transmembrane protein important for cell differentiation, are another group of FIH substrates. However, only Notch1-3 are substrates and their hydroxylation 24

occur within the intracellular domain which is important for downstream transcription during signaling (Coleman et al., 2007). Subsequent work demonstrated FIH hydroxylation of Notch2 represses its activity. Depletion of FIH results in increased apoptosis of endothelial cells due to increased Notch2 activity that results in suppression of the apoptotic inhibitor survivin (Kiriakidis et al., 2015). The authors suggest that FIH is important for endothelial cell survival and growth and could have a role in angiogenic dependent disease.

1.9.4 Hydroxylation affects protein-protein interactions

FIH hydroxylation has also been shown to be important in regulating protein-protein interactions that modulate downstream signaling. Ankyrin repeat and SOCS box protein 4 (ASB4) which is an adaptor protein for the elongin B/elongin C/ cullin/ Roc ubiquitin ligase complex is also subject to asparagine hydroxylation. ASB4 regulates vascular lineage differentiation in an oxygen-dependent manner and the authors propose that FIH-mediated hydroxylation may influence ASB4 binding and degradation of target proteins during this process (Ferguson 3rd et al., 2007). Hydroxylation of apoptosis-stimulating p53-binding protein 2 (ASPP2) is required for its interaction with Par3 and localizing this protein complex to cell-cell junctions (Janke et al., 2013). FIH binding to LRRK1 disrupts its interaction with EGFR and inhibits EGFR turnover during endosomal trafficking. This results in increased EGFR phosphorylation and MAPK signaling downstream which increases keratinocyte migration (Peng et al., 2014). OTUB1, a deubiquitinating enzyme, is another substrate and hydroxylation at asparagine 22 influences OTUB1 protein interactions as an alanine mutation led to an increase in its protein interactome. Many of the new interactors were involved with cellular metabolism and expression of the OTUB1 mutant led to an increase in cell nutrient deprivation as indicated by high AMPK phosphorylation. These results suggest that OTUB1 hydroxylation plays a role in cellular energy metabolism (Scholz et al., 2011). The ankyrin repeats of p105 a precursor subunit of NF-κB and its related protein IκBα were identified as FIH interactors and subject to asparagine hydroxylation. In vitro studies showed hydroxylation of IκBα was important for its inhibitory function towards NF-κB DNA binding, but this could not be capitulated in vivo (Cockman et al., 2006). Taken together, these studies suggest FIH hydroxylation has context dependent functions which influence protein-protein interactions or enzymatic activity in target proteins. While in

vitro biochemical work shows that hydroxylation influences ankyrin repeat stability, there is little which supports FIH-mediated hydroxylation controlling protein stability in vivo.

1.9.5 Role of FIH in metabolism

FIH deficient mice surprisingly have no phenotype based on HIF regulated processes including angiogenesis, development or erythropoiesis as FIH null cells had no measurable differences in HIF target gene expression. Instead, these mice have significant metabolic phenotypes displayed by high metabolic rates, hyperventilation, increased insulin sensitivity and lipid metabolism. These mice have reduced body weight and are resistant to weight gains from high-fat diets (Zhang et al., 2010). The phenotypes of these mice strongly suggest FIH regulates metabolism through HIF independent processes. FIH also plays a role in oxidative metabolism and adaptation to hypoxia. Loss of FIH in skeletal muscle initially increases oxygen consumption and oxidative metabolism under normal oxygen conditions but then a switch occurs in hypoxic muscle tissue where decreases are seen in both oxygen consumption and oxidative metabolism (Sim et al., 2018). FIH works in tandem with prolyl hydroxylase (PHD) enzymes to regulate the metabolic response to hypoxia in mice.

Chapter 2

Characterization of the RNF146 Proteome 2.1 Abstract

PARylation dependent ubiquitination (PARdU) is a recently described mechanism for targeting proteins towards the proteasome. This pathway is triggered by the recognition of PAR by the WWE domain of the RING domain containing E3 ligase RNF146. Currently, several proteins have been identified that regulated by this pathway including Tankyrases, AXIN1, 3BP2, PTEN, and Angiomotins. The PARdU recognition molecule on substrates is facilitated by Tankyrases, a PARP family member. The RNF146 proteome has not been fully characterized and it is unclear whether RNF146 can bind to proteins in a non-PARylation dependent manner. We utilized BioID to classify the types of RNF146 interactors and discover novel substrates of RNF146 regulated PARdU. Through this approach we identified known PARdU targets Tankyrase and AMOTL1 and validated a novel target SH3BP5. Interestingly, SH3BP5 is regulated by PARdU in certain cell-types and we show that another substrate, 3BP2, is also differentially regulated across cell-types. This work reveals that PARdU has cell-type specificity and that other WWE domain E3 ligases may be involved in this mechanism and provide specificity and selectivity of targets.

2.2 Introduction

Ubiquitin-mediated proteasomal degradation is commonly regulated by post-translational modification. There are numerous examples of phosphorylation being a critical factor for targeting proteins for proteasomal degradation including β-catenin and TAZ/YAP (Orford et al., 1997; Liu et al., 2010; Zhao et al., 2010). Hydroxylation is another modification that regulates this process where prolyl-hydroxylation of HIF1α results in degradation by the von Hippel Lindau (pVHL) E3 ligase (Ivan et al., 2001; Jaakkola et al., 2001). More recently, poly-ADP- ribosylation (PARylation) has emerged as another modification that regulates proteasomal degradation. PARylation dependent ubiquitination (PARdU) is a new mechanism of targeting proteins where PAR recognition triggers activation of the E3 ligase (Wang et al., 2012, DaRosa et a., 2014).

The E3 ligase responsible for PARdU, RNF146, which has a RING domain, contains a WWE domain that recognizes the ADP-ribose moiety. In mammalian cells, RNF146 belongs to a unique subset of E3 ligases that contain this domain which include the RING domain ligases Deltex1, Deltex1 and Deltex4 as well as the HECT domain ligases HUWE1 and TRIP12 (Aravind 2001; Wang et al., 2012). To date, only RNF146 has been shown to regulate and function through PARdU (Zhang et al., 2011; Callow et al., 2011). The WWE domain of RNF146 binds iso-ADP-ribose, which is the smallest internal unit of PAR chains (Wang et al., 2012; He et al., 2012). The other E3 WWE domains have been shown or predicted to bind to iso- ADP-ribose as well, creating the possibility that all E3 WWE ligases can participate in PARdU (Wang et al., 2012). RNF146 has been shown to be in an inactive conformation until engagement of the WWE domain with ADP-ribose which results in a conformational change that activates the ligase activity of the RING domain (DaRosa et al., 2014).

Tankyrases have been shown as the source of PAR that is recognized by RNF146. Several proteins have been identified as targets of the RNF146/Tankyrase axis including Axin, 3BP2 and AMOT/AMOTL1/AMOTL2 (Huang et al., 2009; Zhang et al., 2011; Callow et al., 2011; Levaot et al., 2011; Wang et al., 2015; Campbell et al., 2016; Wang et al., 2016). However, the specificity of ADP-ribose recognition by RNF146 is unclear. It remains unknown whether RNF146 is the sole reader of Tankyrase-mediated PARylation or if it competes with other WWE

domain E3’s for substrates. Tankyrase-mediated degradation of the centromere protein CPAP and GDP-Mannose-4,6-Dehydratase (GMD) regulation of Tankyrase-1 stability are examples where the E3 ligase downstream has not been identified (Kim et al., 2012; Bisht et al., 2012).

Roles for RNF146 in other PARylation dependent processes have been described during PAR- dependent cell death (parthanatos), DNA damage and DNA repair (Andrabi et al., 2011; Kang et al., 2011). RNF146 is neuroprotective against N-methyl-D-aspartate (NMDA) induced excitotoxicity (Andrabi et al., 2011). RNF146 acts as a survival gene that becomes upregulated upon NMDA stimulation. Its ability to bind PAR affects the translocation of apoptosis inducing factor (AIF) from mitochondria to the nucleus and reduces cell death (Andrabi et al., 2011; Wang et al., 2009). Tandem affinity purification and mass spectrometry experiments using RNF146 as bait revealed interactions with components of the DNA damage response pathway and DNA repair. The DNA damage components PARP1 and PARP2 and DNA repair pathway proteins Ku70, Ku86, LIG3 and XRCC1 were identified as RNF146 interactors. RNF146 degrades PARP1 through PARdU upon DNA damage and protects cells from cell death and enhances DNA repair (Kang et al., 2011). It remains unclear whether RNF146 can mediate protein-protein interactions and function independently of PARylation separate from binding Tankyrases through its C-terminus (DaRosa et al., 2018).

To achieve a better understanding of the RNF146 interactome and identify new substrates of PARdU, we used a BioID mass spectrometry based RNF146 proximity assay. BioID involves creating a fusion protein with a biotin ligase, BirA, that biotinylates proteins within an estimated 10nm radius. Biotin labeled proteins represent either direct interactors, proteins that are in complex with direct interactors or proteins that are expressed within the neighborhood of RNF146 but not themselves interactors. Labeled proteins are then affinity purified using streptavidin beads, and trypsin-digested peptides are identified by mass spectrometry (Roux et al., 2012; Kim et al., 2014).

Significant RNF146 interactors from BioID were classified into three categories: PARylation dependent substrates, PARylation independent interactors and PARdU substrates. SH3BP5 was identified as a new target of PARdU that is dependent on being a Tankyrase substrate and

additionally found its paralog SH3BP5L as a PARdU target as well. Surprisingly, even though these proteins were regulated by RNF146 in an overexpression model, genetic perturbation studies suggest that they are differentially regulated by RNF146. These observations raise questions regarding the rules determining the specificity behind substrate recognition and targeting by PARdU by RNF146 and other WWE E3 ligase family members.

2.3 Experimental Procedures

2.3.1 Cell lines

Cell lines used in this study were HEK293T, HEK293 T-Rex (Invitrogen), and MEFs. HEK293T cells and MEFs were grown in DMEM (WISENT) with 10% FBS (WISENT) and 100U/ml penicillin and 100µg/ml streptomycin (WISENT). HEK293 T-Rex cells were additionally maintained with 5µg/ml blasticidin (BioShop) and 100µg/ml zeocin (Invitrogen). Cells were cultured at 37⁰C in a humidified incubator with 5% CO2.

2.3.2 Plasmids

Human RNF146 plasmids, pcDNA4-TO HA-RNF146, R163A, and ΔRING were kind gifts from Dr. Feng Cong (Novartis, Boston, U.S.A.). Plp dMyc TNKS/TNKS2 plasmids were previously described in (Guettler et al.,2011, Levaot et al., 2011). pcDNA4-TO HA-RNF146, R163A, and ΔRING were kind gifts from Dr. Feng Cong (Novartis, Boston, U.S.A.). RNF146 and R163A were cloned into the BioID plasmid pcDNA5 FRT.TO FlagBirA R118G by PCR and using AscI and NotI restriction sites. SH3BP5 cDNA was isolated from HEK293T RNA and subcloned into pCMV10 (Sigma) by PCR using HindIII and EcoRI restriction sites. Human SH3BP5L cDNA was obtained from OpenFreezer (Mount Sinai Hospital) and cloned into pCMV10 with EcoRI and XbaI sites. Tankyrase binding domain deletion mutants were generated using QuikChange Lightning XL (Agilent).

Table 2.1: List of constructs and cloning primers Plasmid Name Primer Sequence Primer Name Target pcDNA5 FRT.TO Flag ttaggcgcgcctatggctggctgtggtgaa hRNF146 flag BirAAscI for Human RNF146 BirAR118G RNF146 ttagcggccgcttaaacttcagttac hRNF146 flag BirANotI rev pcDNA5 FRT.TO Flag ttaggcgcgcctatggctggctgtggtgaa hRNF146 flag BirAAscI for Human RNF146 BirAR118G RNF146 R163A ttagcggccgcttaaacttcagttac hRNF146 flag BirANotI rev R163A pCMV10 3BP5 ttaaagcttgacgcggcactgaa Hu_3BP5HindIIIFor Human ttagaattcgtcagccaatctgca Hu_3BP5 EcoRI Rev SH3BP5 30

pCMV10 3BP5ΔTBM1 gtgccatggggcctggcagcagcac Hu3BP5 TBM1 For RGCGVGAE gtgctgctgccaggccccatggcac Hu3BP5 TBM1 Rev (269-276aa) pCMV10 3BP5ΔTBM2 gttcccagtgttgggcccttcccctgaatg Hu3BP5 TBM2 For RSECSGAS cattcaggggaagggcccaacactgggaac Hu3BP5 TBM2 Rev (368-375aa) pCMV10 gttcccagtgttgggcccttcccctgaatg Hu3BP5 TBM2 For RGCGVGAE 3BP5ΔTBM1ΔTBM2 cattcaggggaagggcccaacactgggaac Hu3BP5 TBM2 Rev RSECSGAS ΔTBM1 used as template pCMV10 3BP5L ttagaattcagctgagctca Hu3BP5L EcoRI5' Human SH3BP5L ttatctagactacaggctga Hu3BP5L Xba3' pCMV10 3BP5LΔTBM1 ttagaattcagcagaattgaccccacagg Hu35L TBM1EcoRI5 RQVPGGRE ttatctagactacaggctga Hu3BP5L Xba3' (5-12aa)

pCMV10 3BP5LΔTBM2 caggttccaggagggcggcctgaagttgta Hu3BP5L delTBM2 F RETPQGEL (11- tacaacttcaggccgccctcctggaacctg Hu3BP5L delTBM2 R 18aa) pCMV10 3BP5LΔTBM3 ctgggccctcgggcaggacccgag Hu3BP5L delTBM3 F RSSPVGAE ctcgggtcctgcccgagggcccag Hu3BP5L delTBM3 R (288-295aa) pCMV10 ttagaattcagcagaattgcggcctgaagt Hu3BP5L RQVPGGRE 3BP5LΔTBM1ΔTBM2 ttatctagactacaggctga delTBM1.2EcoRI5' RETPQGEL Hu3BP5L Xba3' pCMV10 ctgggccctcgggcaggacccgag Hu3BP5L delTBM3 F Used ΔTBM1 as 3BP5LΔTBM1ΔTBM3 ctcgggtcctgcccgagggcccag Hu3BP5L delTBM3 R template pCMV10 ctgggccctcgggcaggacccgag Hu3BP5L delTBM3 F Used ΔTBM2 as 3BP5LΔTBM2ΔTBM3 ctcgggtcctgcccgagggcccag Hu3BP5L delTBM3 R template pCMV10 3BP5L ctgggccctcgggcaggacccgag Hu3BP5L delTBM3 F Used ΔTBM1 ΔTBM1ΔTBM2ΔTBM3 ctcgggtcctgcccgagggcccag Hu3BP5L delTBM3 R ΔTBM2 as template

2.3.3 siRNA transfection

Silencer select siRNAs were purchased from Thermo Fisher Scientific. Human RNF146 specific siRNAs (ID: s37821, s37822) were transfected into 293T cells in 12-well plates. 10nM of siRNA was transfected for 72 hours using Lipofectamine RNAiMAX at a reagent ratio of 3:1 (Invitrogen). Human TNKS (ID: s16482) and TNKS2 (ID: s37281) siRNAs were transfected into 293T cells in 12 well-plates at 10nM and 20nM respectively, for 72 hours using RNAiMAX. Cells were lysed and analyzed by immunoblotting.

2.3.4 Immunoblotting and immunoprecipitation

Cell lysates were prepared by lysing in buffer containing 1% Triton X-100, 25mM Tris pH7.5, 100mM NaCl, and 1mM EDTA with 1x Halt protease and phosphatase cocktails (ThermoFisher Scientific). Lysates were cleared by centrifugation at 13,300 rpm for 10 minutes at 4⁰C, quantified using the 660nm protein assay kit (ThermoFisher Scientific) and boiled with 1x Laemmli buffer for 5 minutes at 100⁰C.

Immunoblotting was performed by separating samples by SDS-PAGE, transferring to PVDF membrane (Millipore), blocking with 5% skim milk (BioShop) in 1x PBS + 0.1% Tween 20 (PBST) (BioShop) and incubating with indicated antibodies overnight at 4⁰C. Membranes were washed three times with PBST and incubated with horseradish peroxidase conjugated mouse and rabbit antibodies for one hour in 5% milk. Membranes were washed three times with PBST and developed using ECL or ECL prime (GE Healthcare) and SuperSignal West Pico PLUS or SuperSignal West Femto (ThermoFisher) HRP (Horse-radish peroxidase) substrates. Membranes were imaged using a MicroChemi 2.0 chemiluminescent imager (DNR Bio-Imaging Systems).

2.3.5 In vitro ADP-ribosylation assay

HEK293T were transfected in 6-well plates for 24-48 hours with LipoD293 (FroggaBio). For Tankyrase inhibition, cells were treated 24 hours post-transfection with 1μM of TNKS656 for another 24 hours. Cells were lysed in PAR lysis buffer (50mM HEPES pH7.5, 150mM NaCl, 1% Triton X-100, 0.1% SDS, 0.5% sodium deoxycholate, 5mM EDTA, 1x Halt protease/phosphatase cocktails, 5mM dithiothreitol (DTT) and cleared lysates were immunoprecipitated with 12μl of a 50% slurry solution of Flag M2 agarose beads for an hour at 4⁰C. The beads were washed three times with lysis buffer then washed three times with in vitro assay buffer (50mM HEPES pH7.5, 100mM NaCl, 0.01% Triton X-100, 2mM tris(2- carboxyethyl)phosphine (TCEP). Beads were incubated in 50µl of assay buffer plus 10µM biotin-NAD+ and shaken at 25⁰C for one hour. The reaction was stopped with Laemmli buffer, boiled for 5 minutes at 100⁰C and subjected to SDS-PAGE and immunoblot analysis. To detect 32

ADP-ribosylation, blocked membranes were incubated with streptavidin HRP (Cell Signaling Technologies) for one hour at room temperature followed by three washes with PBST. Membranes were then imaged using ECL prime or SuperSignal West Femto HRP substrates.

2.3.6 Antibodies and reagents

The following antibodies were used for immunoblotting: Flag (clone M2, 1:2000; Sigma), Myc (clone 9E10, 1:250; Santa Cruz sc40), HA (1:1000; Cell Signaling Technologies #3724), α- tubulin (1:1000; Santa Cruz sc69969), streptavidin HRP (1:2000; Cell Signaling Technologies #3999), ubiquitin Lys48 (clone Apu2, 1:1000; Millipore), SH3BP5 (1:300; Sigma HPA036445), SH3BP5L (1:500; Novus Biologics NBP2-38385), Tankyrase1/2 (1:500; Santa Cruz sc365897), RNF146 (1:1000; Sigma SAB1408054). The reagents used: TNKS656 inhibitor was a kind gift from Dr. Feng Cong (Novartis), Biotin-NAD+ was purchased from Trevigen (cat# 4670-500-01), MG132 was purchased from Enzo Life Sciences (cat# BML-PI102-0005), D-Biotin (BB0078) was purchased from BioBasic Inc. and tetracycline (cat# 87128) was purchased from Sigma- Aldrich.

2.3.7 BioID Assay and sample preparation

BioID (Roux et al., 2012) was performed as previously described (Coyaud et al., 2015). HEK293 Flp-In T-Rex cell lines stably expressing Flag-BirA-RNF146 and Flag-BirA-RNF146 R163A were generated using the FlpIn system (Invitrogen). HEK293 T-Rex cells in 6-well plates were transfected 100ng of BirA construct and 900ng of pOG44 which contains Flp-recombinase. Culture media was changed 24 hours post-transfection for another 24 hours. Cells were placed under selection with 5µg/ml blasticidin and 200µg/ml hygromycin B for 2-3 weeks until visible colonies emerged, and cells were expanded. For the BioID experiment, five 150cm2 plates of cells were grown to 80% confluency prior to induction of protein expression with 1µg/ml tetracycline (Sigma) and 50µM biotin (BioBasic) for 24 hours. For WT and R163A lines, an additional five plates were treated with 5µM of MG132 in addition to the induction media for 24 hours. Cells were scraped, washed three times with 1x PBS by centrifugation (1,000 rpm for 5 minutes) and pellets were stored at -80⁰C. The cell pellet was resuspended in 10 ml of lysis buffer (50mM Tris-HCl pH 7.5, 150mM NaCl, 1mM EDTA, 1mM EGTA, 1% Triton X-100, 0.1% SDS, 1:500 protease inhibitor cocktail (Sigma-Aldrich), 1:1,000 benzonase nuclease

(Novagen) and incubated on an end-over-end rotator at 4°C for 1 hour, briefly sonicated to disrupt any visible aggregates, then centrifuged at 45,000 x g for 30 minutes at 4°C. Supernatant was transferred to a fresh 15ml conical tube. 30μl of packed, pre-equilibrated Streptavidin sepharose beads (GE) were added and the mixture incubated for 3 hours at 4°C with end-over- end rotation. Beads were pelleted by centrifugation at 2,000 rpm for 2 min and transferred with 1ml of lysis buffer to a fresh Eppendorf tube. Beads were washed once with 1ml lysis buffer and twice with 1ml of 50mM ammonium bicarbonate (pH=8.3). Beads were transferred in ammonium bicarbonate to a fresh centrifuge tube and washed two more times with 1 ml ammonium bicarbonate buffer. Tryptic digestion was performed by incubating the beads with 1 µg MS-grade TPCK trypsin (Promega, Madison, WI) dissolved in 200μl of 50 mM ammonium bicarbonate (pH 8.3) overnight at 37°C. The following morning, 0.5μg MS-grade TPCK trypsin was added, and beads were incubated 2 additional hours at 37°C. Beads were pelleted by centrifugation at 2,000 x g for 2 min, and the supernatant was transferred to a fresh eppendorf tube. Beads were washed twice with 150µl of 50mM ammonium bicarbonate, and these washes were pooled with the first eluate. The sample was lyophilized and resuspended in buffer A (0.1% formic acid). 1/5th of the sample was analyzed per MS run.

2.3.8 Mass spectrometry analysis

High performance liquid chromatography was conducted using a 2 cm pre-column (Acclaim PepMap 50 mm x 100 um inner diameter (ID)), and 50 cm analytical column (Acclaim PepMap, 500 mm x 75 um diameter; C18; 2 um; 100 Å, Thermo Fisher Scientific, Waltham, MA), running a 120 min reversed-phase buffer gradient at 225nl/min on a Proxeon EASY-nLC 1000 pump in-line with a Thermo Q-Exactive HF quadrupole-Orbitrap mass spectrometer. A parent ion scan was performed using a resolving power of 60,000, then up to the twenty most intense peaks were selected for MS/MS (minimum ion count of 1,000 for activation) using higher energy collision induced dissociation (HCD) fragmentation. Dynamic exclusion was activated such that MS/MS of the same m/z (within a range of 10 ppm; exclusion list size = 500) detected twice within 5 sec were excluded from analysis for 15 sec. For protein identification, Thermo .RAW files were converted to the .mzXML format using Proteowizard (Kessner et al., 2008) then searched using X!Tandem (Craig and Beavis 2004) and COMET (Eng et al., 2013) against the human Human RefSeq Version 45 database (containing 36,113 entries). Data were analyzed

using the trans-proteomic pipeline (TPP) via the ProHits software suite (v3.3) (Pedrioli 2010; Deutsch et al., 2010; Liu et al., 2010). Search parameters specified a parent ion mass tolerance of 10 ppm, and an MS/MS fragment ion tolerance of 0.4 Da, with up to 2 missed cleavages allowed for trypsin. Variable modifications of +16@M and W, +32@M and W, +42@N-terminus, and +1@N and Q were allowed. Proteins identified with an iProphet cut-off of 0.9 (corresponding to ≤1% FDR) and at least two unique peptides were analyzed with SAINT Express v.3.6. Ten control runs (from cells expressing the FlagBirA* epitope tag) were collapsed to the two highest spectral counts for each prey and compared to two technical runs of each of the two biological replicates of RNF146 WT/R163A and one run of two biological replicates of SH3BP5. High confidence interactors were defined as those with BFDR≤0.01.

2.3.9 Interaction network analysis

The RNF146 interaction network was created using Cytoscape v3.2 and importing known interaction data with the GeneMANIA plugin (Shannon et al., 2003; Warde-Farley et al., 2010). Proteins were grouped according to biological function through gene ontology (http://geneontology.org) and by manual literature search curation (Ashburner et al., 2000; Gene Ontology Consortium, 2015).

2.4 Results

2.4.1 Systematic elucidation of the RNF146 interactome

To identify the ensemble of proteins that interact with and are regulated by RNF146, a BioID mass spectrometry proximity assay was used with RNF146 as the bait. To identify PARylated proteins, which bound to RNF146 through the WWE domain we generated a WWE domain loss of function RNF146 mutant (R163A) unable to PAR. Lastly, to identify potential RNF146 substrates we used the proteasome inhibitor MG132 to identify candidate proteins that were actively turned over by ubiquitin mediated proteolysis. Therefore, four experimental conditions were interrogated by BioID: wild type RNF146 in the absence or presence of MG132 versus RNF146-R163A in the absence or presence of MG132. HEK293 Flp-In cells were used to express wildtype RNF146 or RNF146-R163A. The two lines were tested for expression of Flag- BirA-RNF146 and the mutant RNF146 protein. Each cell line was plated in 12-well plates and treated with 1μg/ml of tetracycline (to induce protein expression) and 50μM biotin (substrate for BirA) was added to the cultures for 24 hours. Cells were lysed in a 1% Triton X-100 lysis buffer and the lysates were subjected to immunoblot analysis. Membranes were probed with streptavidin-HRP to detect biotinylation of RNF146 and RNF146 associated proteins (Figure 2.1A, B). Fifty-one statistically significant proteins were identified in at least one of the four samples (Figure 2.2A). Gene ontology analysis and literature annotation of the significant interactors revealed a broad range of biological processes that RNF146 engages (Figure 2.2B). These processes include established roles in WNT signaling, DNA damage and cell polarity but also revealed potential new roles in regulating transcription, apoptosis, metabolism and vesicular trafficking.

Figure 2.1: Generation of RNF146 BirA cell lines. 293T Flp-In line validation for expression of Flag-BirA RNF146 (A) and RNF146 R163A (B). Flag-BirA expression was induced with 1µg/ml tetracycline and 50µM biotin for 24 hours. Whole cell lysates were subjected to immunoblot analysis for Streptavidin HRP to detect protein biotinylation and Flag.

Figure 2.2: Characterization of the RNF146 Interactome. A) Volcano plot of all interacting proteins assigned to each RNF146 BioID sample. All proteins are plotted based on their fold change value over a control Flag-BirA sample. Red points represent significant interactors above the q value threshold of 0.5. B) Representation of all significant RNF146 interactors from each BioID sample. The interaction network was created using Cytoscape and grouped by biological function using the Gene Ontology database (http://geneontology.org). The hexagon size indicates the spectrum sum of each BioID run. Solid black lines indicate known protein interactions imported from the GeneMANIA plugin and dotted lines represent new interactions identified by BioID.

Differential analysis of the spectral counts of each the candidate proteins were used to categorize RNF146 interactors as potential substrates (proteins stabilized by MG132), PARylation dependent binders (binding was lost with the RNF146-R163A mutant), and PARylation independent binders (bound to both wild type and mutant form of RNF146). Comparison between the log2 fold change (FC) of each interactor’s spectral sum across the four samples was used (See Table 2.2).

Three comparative analyses were used to interrogate the RNF146 interactome and identify PARylation dependent substrates: [(WT vs. R163A), (WT+MG132 vs. WT), (WT+MG132 vs. R163A+MG132)]. A log2FC>1 was used as a cutoff to classify proteins (Figure 2.3A). The first comparison between WT vs. R163A samples revealed the 18 PARylation dependent interactions including known RNF146 targets; Tankyrase, AMOTL, PARP1 and PARP2 (Figure 2.3B). Comparison of RNF146 vs. RNF146+MG samples identified PAR-dependent binders that are also likely RNF146 substrates (Figure 2.3C). Fourteen proteins are potential RNF146 substrates of which nine had putative Tankyrase binding motifs based on our previously developed consensus sequence scoring system (Figure 2.3D) (Guettler et al., 2011). The presence of established PARdU targets, AMOTL1 and TNKS provided a bench mark for the validity of our approach to identify new potential PARdU regulated proteins. Nine of the fourteen proteins are novel RNF146/Tankyrase regulated candidates while the remaining proteins (LAMC1, DBTB41, DCAFL82, KIAA0284 and APOBEC3C) could be targets of other PARPs. SH3BP5 was chosen for further validation since it had the highest scoring consensus motif (Figure 2.3D) and was observed in several independent Tankyrase mass spectrometry experiments (Guettler et al., 2011; Wang et al., 2015; Li et al., 2017; Huttlin et al., 2015; Bhardwaj et al., 2017).

Table 2.2: Significant RNF146 interactors from BioID

log2 RNF146 RNF146_MG WT MG Gene Control counts Counts Sum SAINT Counts Sum SAINT WT

PARP1 88|77|72|75|77|77|135|133 464|460|437|407 1768 1.00 500|508|492|479 1979 1.00 0.2 HSPA1B 209|213|203|225|200|210|242|232 261|274|237|214 986 0.00 419|435|458|446 1758 0.00 0.8 HSPA6 17|13|0|0|0|0|0|0 0|0|0|0 0 0.00 121|121|117|120 479 1.00 8.9 DLG1 13|7|6|10|12|18|30|33 116|117|101|96 430 1.00 107|96|9|99 311 0.75 -0.5 BAG3 2|0|4|6|10|7|4|8 5|4|6|5 20 0.00 36|42|32|29 139 1.00 2.8 CASK 0|0|0|0|0|0|0|0 20|14|29|28 91 1.00 29|27|25|30 111 1.00 0.3 TNKS 0|0|0|0|0|0|0|0 3|4|4|3 14 0.97 18|13|18|15 64 1.00 2.2 HLTF 2|0|3|0|0|0|0|0 0|0|2|0 2 0.00 17|18|11|10 56 0.94 4.8 MPP7 0|0|0|0|0|0|0|0 16|8|18|11 53 1.00 14|15|14|13 56 1.00 0.1 MAGI1 0|0|0|0|3|0|3|4 16|20|19|13 68 0.97 15|15|12|11 53 0.78 -0.4 TMPRSS11A 0|0|0|0|0|0|0|0 13|17|14|14 58 1.00 9|16|13|14 52 1.00 -0.2 ALMS1 0|2|2|2|0|0|0|0 5|5|2|8 20 0.19 10|15|10|13 48 0.97 1.3 KIAA0247 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 7|7|13|10 37 1.00 5.2 LAMC1 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 9|9|10|7 35 1.00 5.1 KIF1B 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 13|5|3|10 31 0.99 5.0 ZBTB41 0|0|0|0|0|0|0|0 0|0|2|0 2 0.19 2|10|8|8 28 0.94 3.8 DCAF8L2 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 6|7|6|9 28 1.00 4.8 PLCB1 0|0|0|0|0|0|0|0 0|3|0|0 3 0.24 3|8|13|3 27 0.97 3.2 IRS2 2|2|0|0|3|2|2|0 2|0|2|0 4 0.00 6|10|5|6 27 0.22 2.8 APOBEC3C 0|0|0|0|0|0|0|0 3|0|0|0 3 0.24 7|6|6|6 25 1.00 3.1 SH3BP5 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 5|7|5|6 23 1.00 4.5 LIN7C 0|0|0|0|0|0|0|0 8|7|7|5 27 1.00 7|5|5|6 23 1.00 -0.2 TK1 0|0|2|0|0|0|0|0 0|2|0|0 2 0.01 4|5|7|5 21 0.37 3.4 KIAA0284 0|0|0|0|0|0|0|0 0|2|3|0 5 0.42 7|7|3|3 20 0.97 2.0 DLG5 0|0|0|0|0|0|0|0 3|2|2|2 9 0.80 7|6|4|2 19 0.93 1.1 SMG5 0|0|0|0|0|0|0|0 2|0|2|3 7 0.61 7|3|4|4 18 0.98 1.4 AMOTL1 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 3|6|4|4 17 0.98 4.1 FOXO1 0|0|0|0|0|0|0|0 0|0|0|2 2 0.19 5|4|5|3 17 0.98 3.1 HERC1 0|0|0|0|0|0|0|0 12|7|10|12 41 1.00 4|6|3|4 17 0.98 -1.3 PARP2 0|0|0|0|0|0|0|0 9|12|10|11 42 1.00 4|5|5|3 17 0.98 -1.3 RAB2B 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 4|4|4|4 16 0.99 4.0 MAGI3 0|0|0|0|0|0|0|0 6|6|4|5 21 1.00 4|4|5|3 16 0.98 -0.4 FAR1 0|0|0|0|0|0|0|0 4|3|4|4 15 0.98 3|2|3|3 11 0.90 -0.4 CEP85 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 2|3|2|0 7 0.61 2.8 ASS1 2|0|5|3|0|3|0|0 5|2|2|4 13 0.00 2|3|0|2 7 0.00 -0.9 SNX27 0|0|0|0|0|0|0|0 5|9|5|9 28 1.00 2|4|0|0 6 0.43 -2.2 GSK3B 0|0|0|0|0|0|0|0 3|3|2|4 12 0.91 3|2|0|0 5 0.42 -1.3 KDELR1 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 2|0|0|2 4 0.37 2.0 ZFAND5 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0|0|2|0 2 0.19 1.0 GGA2 0|0|0|0|0|0|0|0 2|0|3|3 8 0.66 0|2|0|0 2 0.19 -2.0 CAPN2 0|0|0|0|0|0|0|0 3|4|3|3 13 0.96 2|0|0|0 2 0.19 -2.7 BAX 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0|0|0|0 0 0.00 na CDS2 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0|0|0|0 0 0.00 na ITIH3 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0|0|0|0 0 0.00 na PLK1S1 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0|0|0|0 0 0.00 na C21orf33 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0|0|0|0 0 0.00 na FDFT1 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0|0|0|0 0 0.00 na MYEOV2 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0|0|0|0 0 0.00 na KRTCAP2 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0|0|0|0 0 0.00 na AIFM1 0|0|0|0|0|0|0|0 4|2|2|0 8 0.62 0|0|0|0 0 0.00 na HDAC6 0|0|0|0|0|0|0|0 4|2|5|3 14 0.92 0|0|0|0 0 0.00 -3.8

log2 RNF146_R163A WT Gene Control counts Counts Sum SAINT R163A

PARP1 88|77|72|75|77|77|135|133 43|37|40|46 166 0.00 3.4 HSPA1B 209|213|203|225|200|210|242|232 352|372|327|306 1357 0.00 -0.5 HSPA6 17|13|0|0|0|0|0|0 16|21|13|15 65 0.00 -6.0 DLG1 13|7|6|10|12|18|30|33 132|123|118|124 497 1.00 -0.2 BAG3 2|0|4|6|10|7|4|8 8|6|5|5 24 0.00 -0.3 CASK 0|0|0|0|0|0|0|0 32|36|30|28 126 1.00 -0.5 TNKS 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 3.8 HLTF 2|0|3|0|0|0|0|0 0|0|0|0 0 0.00 1.0 MPP7 0|0|0|0|0|0|0|0 2|4|0|2 8 0.62 2.7 MAGI1 0|0|0|0|3|0|3|4 13|15|11|16 55 0.82 0.3 TMPRSS11A 0|0|0|0|0|0|0|0 21|20|17|26 84 1.00 -0.5 ALMS1 0|2|2|2|0|0|0|0 7|10|7|10 34 0.65 -0.8 KIAA0247 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0.0 LAMC1 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0.0 KIF1B 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0.0 ZBTB41 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 1.0 DCAF8L2 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0.0 PLCB1 0|0|0|0|0|0|0|0 0|2|0|0 2 0.19 0.6 IRS2 2|2|0|0|3|2|2|0 2|2|0|5 9 0.00 -1.2 APOBEC3C 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 1.6 SH3BP5 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0.0 LIN7C 0|0|0|0|0|0|0|0 4|4|3|4 15 0.98 0.8 TK1 0|0|2|0|0|0|0|0 0|4|2|2 8 0.07 -2.0 KIAA0284 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 2.3 DLG5 0|0|0|0|0|0|0|0 2|2|0|0 4 0.37 1.2 SMG5 0|0|0|0|0|0|0|0 3|0|0|6 9 0.49 -0.4 AMOTL1 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0.0 FOXO1 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 1.0 HERC1 0|0|0|0|0|0|0|0 6|4|5|4 19 0.99 1.1 PARP2 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 5.4 RAB2B 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0.0 MAGI3 0|0|0|0|0|0|0|0 8|8|8|7 31 1.00 -0.6 FAR1 0|0|0|0|0|0|0|0 4|7|6|6 23 1.00 -0.6 CEP85 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0.0 ASS1 2|0|5|3|0|3|0|0 17|18|17|18 70 1.00 -2.4 SNX27 0|0|0|0|0|0|0|0 5|6|6|10 27 1.00 0.1 GSK3B 0|0|0|0|0|0|0|0 3|4|3|3 13 0.96 -0.1 KDELR1 0|0|0|0|0|0|0|0 2|2|4|2 10 0.81 -3.3 ZFAND5 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0.0 GGA2 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 3.0 CAPN2 0|0|0|0|0|0|0|0 0|2|2|0 4 0.37 1.7 BAX 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0.0 CDS2 0|0|0|0|0|0|0|0 0|0|3|4 7 0.49 -2.8 ITIH3 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0.0 PLK1S1 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0.0 C21orf33 0|0|0|0|0|0|0|0 7|10|9|5 31 1.00 -5.0 FDFT1 0|0|0|0|0|0|0|0 3|3|4|5 15 0.97 -3.9 MYEOV2 0|0|0|0|0|0|0|0 4|3|3|3 13 0.96 -3.7 KRTCAP2 0|0|0|0|0|0|0|0 3|3|3|3 12 0.95 -3.6 AIFM1 0|0|0|0|0|0|0|0 10|15|17|19 61 1.00 -2.9 HDAC6 0|0|0|0|0|0|0|0 4|4|4|3 15 0.98 -0.1

log2 RNF146_R163A_MG WT MG Gene Control counts Counts Sum SAINT R163A MG

PARP1 88|77|72|75|77|77|135|133 71|72|78|72 293 0.00 2.8 HSPA1B 209|213|203|225|200|210|242|232 862|844|779|784 3269 1.00 -0.9 HSPA6 17|13|0|0|0|0|0|0 123|113|96|120 452 1.00 0.1 DLG1 13|7|6|10|12|18|30|33 85|85|89|85 344 0.97 -0.1 BAG3 2|0|4|6|10|7|4|8 18|19|24|22 83 0.05 0.7 CASK 0|0|0|0|0|0|0|0 22|22|20|22 86 1.00 0.4 TNKS 0|0|0|0|0|0|0|0 3|2|0|2 7 0.61 3.2 HLTF 2|0|3|0|0|0|0|0 0|0|0|0 0 0.00 5.8 MPP7 0|0|0|0|0|0|0|0 0|3|3|3 9 0.71 2.6 MAGI1 0|0|0|0|3|0|3|4 7|11|7|7 32 0.11 0.7 TMPRSS11A 0|0|0|0|0|0|0|0 2|5|0|3 10 0.67 2.4 ALMS1 0|2|2|2|0|0|0|0 26|26|29|32 113 1.00 -1.2 KIAA0247 0|0|0|0|0|0|0|0 15|11|7|9 42 1.00 -0.2 LAMC1 0|0|0|0|0|0|0|0 3|6|3|4 16 0.97 1.1 KIF1B 0|0|0|0|0|0|0|0 0|6|2|0 8 0.44 2.0 ZBTB41 0|0|0|0|0|0|0|0 0|2|4|3 9 0.67 1.6 DCAF8L2 0|0|0|0|0|0|0|0 2|3|3|4 12 0.91 1.2 PLCB1 0|0|0|0|0|0|0|0 4|0|3|0 7 0.49 1.9 IRS2 2|2|0|0|3|2|2|0 16|15|16|18 65 1.00 -1.3 APOBEC3C 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 4.6 SH3BP5 0|0|0|0|0|0|0|0 2|2|3|3 10 0.85 1.2 LIN7C 0|0|0|0|0|0|0|0 6|4|6|2 18 0.93 0.4 TK1 0|0|2|0|0|0|0|0 11|11|11|12 45 0.96 -1.1 KIAA0284 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 4.3 DLG5 0|0|0|0|0|0|0|0 4|3|0|0 7 0.49 1.4 SMG5 0|0|0|0|0|0|0|0 3|0|2|0 5 0.42 1.8 AMOTL1 0|0|0|0|0|0|0|0 0|2|0|2 4 0.37 2.1 FOXO1 0|0|0|0|0|0|0|0 2|3|2|2 9 0.80 0.9 HERC1 0|0|0|0|0|0|0|0 6|5|6|2 19 0.94 -0.2 PARP2 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 4.1 RAB2B 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 4.0 MAGI3 0|0|0|0|0|0|0|0 5|7|4|3 19 0.98 -0.2 FAR1 0|0|0|0|0|0|0|0 5|7|3|5 20 0.99 -0.9 CEP85 0|0|0|0|0|0|0|0 6|3|2|4 15 0.92 -1.1 ASS1 2|0|5|3|0|3|0|0 4|5|6|6 21 0.00 -1.6 SNX27 0|0|0|0|0|0|0|0 2|3|2|3 10 0.85 -0.7 GSK3B 0|0|0|0|0|0|0|0 0|0|2|0 2 0.19 1.3 KDELR1 0|0|0|0|0|0|0|0 4|6|2|3 15 0.92 -1.9 ZFAND5 0|0|0|0|0|0|0|0 3|5|3|3 14 0.96 -2.8 GGA2 0|0|0|0|0|0|0|0 3|4|3|5 15 0.97 -2.9 CAPN2 0|0|0|0|0|0|0|0 0|3|4|0 7 0.49 -1.8 BAX 0|0|0|0|0|0|0|0 4|4|3|5 16 0.98 -4.0 CDS2 0|0|0|0|0|0|0|0 4|5|4|3 16 0.98 -4.0 ITIH3 0|0|0|0|0|0|0|0 3|3|5|4 15 0.97 -3.9 PLK1S1 0|0|0|0|0|0|0|0 5|3|4|4 16 0.98 -4.0 C21orf33 0|0|0|0|0|0|0|0 3|3|3|5 14 0.96 -3.8 FDFT1 0|0|0|0|0|0|0|0 0|0|0|4 4 0.25 -2.0 MYEOV2 0|0|0|0|0|0|0|0 3|4|3|2 12 0.91 -3.6 KRTCAP2 0|0|0|0|0|0|0|0 0|0|0|0 0 0.00 0.0 AIFM1 0|0|0|0|0|0|0|0 4|8|5|6 23 1.00 -4.5 HDAC6 0|0|0|0|0|0|0|0 0|3|3|0 6 0.47 -2.6

Figure 2.3: Classification of RNF146 interactors. A) Bar plots of log2 fold change analysis of the spectrum sums of the four BioID samples. Left: WT vs. R163A, middle: WT+MG vs. WT, right: WT+MG vs. R163A+MG. Genes with log2FC>1 were chosen as substrate or WWE dependent candidates. B) Simplified interaction network displaying significant WWE dependent interactors. Proteins with log2FC>1 between WT and R163A samples were chosen as WWE dependent. C) Interaction network displaying WWE dependent interactors that are also substrates. Proteins with log2FC>1 between WT and WT+MG samples were overlaid with WWE dependent interactors. D) Ranking of WWE dependent substrates of RNF146 based on Tankyrase binding motif scores developed by Guettler et al., 2011.

2.4.2 SH3BP5 and SH3BP5L are novel Tankyrase substrates

SH3BP5 was originally identified as a negative regulator of Bruton’s Tyrosine Kinase (BTK), which bound to its SH3 domain (Matsushita et al., 1998; Yamadori et al., 1999). SH3BP5 is JNK stress kinase a substrate and binds to JNK through two kinase interacting motifs (KIM) (Wiltshire et al., 2002). SH3BP5L is an SH3BP5 paralog, that has a conserved N-terminus (green region Figure 2.4A). To validate the interaction between Tankyrases, SH3BP5, and SH3BP5L, HEK293T cells were co-transfected with Flag-SH3BP5/Flag-SH3BP5L and Myc- TNKS/TNKS2. Cell lysates were immunoprecipitated with Myc-agarose beads and probed for Flag-3BP5/3BP5L by immunoblot analysis. SH3BP5 and SH3BP5L bound to both TNKS or TNKS2 when co-expressed in HEK293T cells (Figure 2.4B, C). We observed that TNKS but not TNKS2 protein level increased dramatically upon co-expression of either SH3BP5 or SH3BP5L suggesting a unique regulatory function of SH3BP5 proteins in controlling the steady levels of TNKS but not TNKS2. Both SH3BP5 and SH3BPL proteins contain consensus Tankyrase binding motifs based on the prediction scoring system described in Guettler et al. (Guettler et al., 2011). Two Tankyrase binding motifs were identified in SH3BP5 while SH3BP5L contains three binding (Figure 2.4A) To determine the motifs required for binding, each of the binding motifs were deleted alone or in combination in SH3BP5/L and then the mutant proteins were tested for interaction with Tankyrase 1/2. Both motifs in SH3BP5 had to be deleted to abolish binding suggesting that each motif contributed to Tankyrase binding (Figure 2.4D). Loss of either of the two motifs in the N-terminus together with deletion of the third TBM was sufficient to uncouple SH3BP5L from Tankyrase (Figure 2.4E). Loss of TNKS interaction with SH3BP5/L resulted in decreased steady state protein abundance of Tankyrase demonstrating that interaction with SH3BP5/L proteins stabilizes TNKS possibly by protection from ubiquitin mediated degradation.

SH3BP5/L were next tested for their ability to serve as Tankyrase substrates using an in vitro PARylation assay. HEK293T cells were transfected for 24 hours with Flag-3BP5/Flag-3BP5L and Myc-TNKS/TNKS2 and lysates were immunoprecipitated with Flag agarose beads. Biotin- NAD+ was incubated with Flag immune complexes, subjected to immunoblot analysis and PARylation was detected by streptavidin-HRP. Both proteins showed strong PARylation signals in addition to auto-PARylation of Tankyrase (Figure 2.5A, B). To ensure the PAR modification

was specific to the PARP catalytic activity of Tankyrase, a Tankyrase specific inhibitor (TNKS656) was added to HEK293T cells for 24 hours after transfection. Comparison of treated and untreated samples showed the inhibitor abolished the PARylation signal (Figure 2.5C, D). These results demonstrate SH3BP5 and SH3BP5L are bona fide Tankyrase binders and substrates in transfected cellular model systems.

Figure 2.4: SH3BP5 and SH3BP5L are Tankyrase binders through multiple motifs. A) Protein domain maps of SH3BP5 and SH3BP5L. Green region indicates the conserved helical region from secondary structure prediction server PSIPRED. BTK Binding – Region that binds to BTK SH3 domain; KIM – Kinase Interaction Motif. TBM – Tankyrase Binding Motif. B,C) HEK293T lysates co-expressing Flag-3BP5 or Flag-3BP5L and Myc-TNKS or Myc-TNKS2 were immunoprecipitated with Myc (9E10) antibody. Immune complexes were subjected to immunoblot analysis for Flag and Myc. Whole cell lysates were analyzed for Flag, Myc and α- Tubulin (loading control). D,E) Lysates from HEK293T co-expressing Flag-3BP5 or Flag- 3BP5L or respective combinatorial TBM deletions (1=TBM1, 2=TBM2, 3=TBM3) with Myc- TNKS or Myc-TNKS2. Lysates were immunoprecipitated with Flag antibody and complexes were subjected to western blot analysis for Myc and Flag. Whole cell lysates were probed with Myc, Flag and α-Tubulin (loading control).

Figure 2.5: SH3BP5 and SH3BP5L are Tankyrase substrates. A,B) HEK293T were co- expressed with Flag-3BP5 or Flag-3BP5L with Myc-TNKS or Myc-TNKS2. Lysates were immunoprecipitated with Flag antibody and immune complexes were washed and incubated with 10µM Biotin-NAD for one hour. This mixture was analyzed by western blotting for Streptavidin-HRP to detect PARylation and Myc and Flag. Whole cell lysates were analyzed with Myc, Flag and α-Tubulin or GAPDH (loading control). C,D) HEK293T co-expressing Flag-3BP5/Flag-3BP5L and Myc-TNKS or Myc-TNKS2 were treated with and without TNKS656 inhibitor and lysates were analyzed by the in vitro PARylation assay.

2.4.3 SH3BP5 and SH3BP5L are RNF146 interactors and substrates

The established mechanism of PARdU involves recognition of iso-ADP-ribose, conjugated by Tankyrase on substrates by the WWE domain of RNF146. This interaction induces a conformational change in the RING domain to facilitate the transfer of ubiquitin molecules from the E2 enzyme to the PARylated target (DaRosa et al., 2014). To validate the interaction between SH3BP5 and RNF146, HEK293T cells were co-transfected with combinations of Flag- SH3BP5/SH3BP5L, Myc-TNKS/TNKS2 and HA-RNF146. Flag-SH3BP5 or Flag-SH3BP5L was immunoprecipitated using Flag agarose beads and subjected to immunoblotting for RNF146. Both SH3BP5 and SH3BP5L bound to RNF146 only in the presence of TNKS or TNKS2. These interactions were also contingent on a functional WWE domain as, the RNF146-R163A mutation, abolished binding (Figure 2.6 A-D). The protein abundance of SH3BP5 and SH3BP5L were decreased in the presence of Tankyrase and RNF146. Tankyrase expression was also decreased with RNF146 overexpression. However, protein levels of SH3BP5/L and Tankyrases were unaffected when expressed with the RNF146 WWE domain mutant (R163A). To determine whether attenuation of SH3BP5 protein expression was due to ubiquitin mediated proteolysis, cells were treated with 10μM MG132 for four hours to block proteasome degradation and stabilize ubiquitin conjugated proteins. Flag-SH3BP5/Flag-SH3BP5L was immunoprecipitated from each condition, subjected to western blotting and probed with an antibody specific for ubiquitin linkages at lysine 48. Expression of RNF146 strongly enhanced SH3BP5 lysine-48 ubiquitylation in the presence of either TNKS or TNKS2 whereas there was no lysine-48 ubiquitylation detected in the absence of Tankyrase expression. SH3BP5L lysine-48 ubiquitylation increased with TNKS/TNKS2 and RNF146 co-expression but notably less than SH3BP5. Deletion of the RNF146 catalytic RING domain impaired SH3BP5/L ubiquitylation demonstrating the specific role of RNF146 as an SH3BP5/L specific E3 ubiquitin ligase. Importantly, the RNF146 R163A mutant was unable to ubiquitylate SH3BP5/L indicating the requirement of PAR recognition by RNF146 to execute ubiquitylation of its substrate (Figure 2.7 A-D). These results demonstrate that RNF146 binds and ubiquitylates SH3BP5 and SH3BP5L in a Tankyrase and ADP-ribose dependent manner.

Figure 2.6: SH3BP5 and SH3BP5L are RNF146 binders. A-D) Lysates from HEK293T co- expressing Flag-3BP5 or Flag-3BP5L with Myc-TNKS or Myc-TNKS2 and HA-RNF146 or R163A mutant were immunoprecipitated with Flag antibody. Immune complexes were analyzed by immunoblot for HA, Myc and Flag while whole cell lysates were analyzed for HA, Myc, Flag and α-Tubulin (loading control).

Figure 2.7: SH3BP5 and SH3BP5L are RNF146 substrates. A-D) HEK293T cells co- expressing Flag-3BP5 or Flag-3BP5L with Myc-TNKS or Myc-TNKS2 and HA-RNF146 WT, ΔRING (ΔR) and R163A (RA) were treated with 10µM MG132 four hours prior to lysis. Lysates were immunoprecipitated with Flag antibody and immune complexes were analyzed by western blot for ubiquitin lysine 48 to detect ubiquitin chains as well as HA, Flag, and Myc. Whole cell lysates were additionally analyzed for HA, Flag, Myc and α-Tubulin (loading control). 50

To determine whether SH3BP5/SH3BP5L were degraded by the proteasome, lysates from HEK293T cells expressing all three proteins was compared to MG132 treated cells prior to cell lysis. Flag-SH3BP5/Flag-SH3BP5L protein levels were examined under four conditions: SH3BP5/SH3BP5L expressed alone, co-expressed with TNKS, co-expressed with RNF146 and TNKS, and co-expressed with RNF146 and TNKS with MG132 treatment. Both SH3BP5 and SH3BP5L proteins stabilized when comparing MG132 treated cells versus untreated cells when RNF146 and TNKS were overexpressed (Figure 2.8 A-D) These results reveal that SH3BP5 and SH3BP5L are novel PARdU regulated proteins whose steady state levels are controlled by the proteasome. Interestingly, TNKS2 levels were not rescued efficiently by proteasome inhibition.

Figure 2.8: SH3BP5 and SH3BP5L are degraded by the proteasome. A-D) HEK293T cells co-expressing Flag-3BP5 or Flag-3BP5L with Myc-TNKS or Myc-TNKS2 and HA-RNF146 were treated with and without 10µM MG132 four hours prior to lysis. Cell lysates were analyzed by western blot for HA, Flag, Myc and α-Tubulin (loading control).

2.4.4 Endogenous SH3BP5 and SH3BP5L are not regulated by RNF146

RNF146 knock out cells or cells lacking both Tankyrases results in stabilization of PARdU target proteins (Huang et al., 2009; Callow et al., 2011; Zhang et al., 2011; Levaot et al., 2011; Wang et al., 2015; Li et al., 2015; Campbell et al., 2016; Bhardwaj et al., 2017). Since overexpression data demonstrated SH3BP5/L as novel PARdU substrates, endogenous regulation was next examined. Knockdown of RNF146 by siRNA in HEK293T cells showed no stabilization of either SH3BP5 or SH3BP5L. However, RNF146 knockdown resulted in strong stabilization and accumulation of both TNKS/TNKS2 and AXIN1 (Figure 2.9A). We next analyzed SH3BP5 levels in RNF146 knockout MEFs and observed that SH3BP5 was significantly stabilized as were TNKS/TNKS2 and AXIN1 protein (Figure 2.9B). 3BP2, which our laboratory has previously shown to be stabilized in RNF146 KO macrophages and osteoclasts, was not stabilized in RNF146 KO MEFs suggesting cell-type specificity of PARdU targeting (Levaot et al., 2011; Matsumoto et al., 2017). To assess whether the abundance of SH3BP5/L proteins were controlled by Tankyrase catalytic activity, HEK293T cells were treated with TNKS656 for 24 and 48 hours. Western blot showed the accumulation of SH3BP5 protein in TNKS656 treated cells in addition to the known stabilization effects of this inhibitor on both Tankyrases and AXIN1 protein levels (Figure 2.10A). Knockdown of both Tankyrases in HEK293T cells and deletion in MEFs showed protein stabilization of AXIN1 and SH3BP5 (Figure 2.10B, C). In distinction to SH3BP5, neither the RNF146 knockout nor Tankyrase inhibition had a measurable effect on SH3BP5L protein levels (Figure 2.9, 2.10). These data show that Tankyrases and AXIN1 are predominant substrates of RNF146 in HEK293T cells and MEFs while SH3BP5 is a RNF146 PARdU target in fibroblasts but not HEK293T cells. Under physiological conditions endogenous SH3BP5L is not regulated by PARdU in HEK293T cells or fibroblasts.

Figure 2.9: Endogenous SH3BP5 and SH3BP5L stability are differentially regulated by RNF146. A) Knockdown of RNF146 by transfecting HEK293T cells with RNF146 siRNA (silencer select – Thermo Fisher) in HEK293T cells. Lysates were analyzed by immunoblotting for the indicated antibodies. B) Western blot analysis of lysates derived from RNF146 knockout MEFs.

Figure 2.10: Endogenous SH3BP5 stability is regulated by Tankyrase. Western blot analysis of cell lysates derived from A) HEK293T cells treated with 1µM TNKS656 for 24 and 48 hours. B) Double knockdown of Tankyrase1/2 in HEK293T cells by siRNA (Silencer Select – Thermo Fisher). C) Double knockout Tankyrase1/2 MEFs.

2.5 Discussion

The work in this chapter describes a proteomics-based approach to identify new components of the RNF146 interactome. Consistent with other immunoprecipitation mass spectrometry data which used RNF146 as the bait, we identified PARP1, PARP2, TNKS, HLTF, MPP7, DLG1, AMOTL1 as RNF146 interactors (Callow et al., 2011; Zhang et al., 2011; Huttlin et al., 2015; Campbell et al., 2016). PARP1 was previously shown to be PARdU target of RNF146 but analysis failed to show that PARP1 levels increased in response to MG132. This may be due to the requirement of DNA damage to induce PARP1 degradation (Kang et al., 2011). Gene ontology analysis of the significant interactors revealed a diversity of biological processes that RNF146 may regulate. These include cell polarity, WNT signaling, DNA damage, transcription, metabolism, RNA regulation, vesicle trafficking, apoptosis and cell adhesion. To date, RNF146 function has only been described through its substrates in cell polarity (Angiomotins), DNA damage (PARP1), WNT signaling (Axin1), osteoclastogenesis (AXIN1 and 3BP2) and osteoblastogenesis (AXIN1) (Kang et al., 2011; Zhang et al., 2011; Callow et al., 2011; Matsumoto et al., 2017; Matsumoto et al., 2017). Interestingly, AXIN1 and 3BP2 were not found amongst the significant hits in our dataset and are also not found in previous RNF146 datasets revealing some limitations in our approach as false negatives exist. It also suggests there may be host range determinants that control substrate engagement by RNF146.

Differential analysis of RNF146 mutants and proteasome inhibitor has revealed 14 new putative targets of PARdU including established targets TNKS and AMOTL1. Five of these new targets do not have consensus Tankyrase binding motifs which suggests other PARPs could be modifying these proteins or they have a cryptic Tankyrase binding motif. Since RNF146 can recognize PARylated PARP1 there is a strong possibility that RNF146 could be a reader of PARylation from other PARP family members (Kang et al., 2011).

As a proof of concept, SH3BP5 was validated as a new target of the RNF146/Tankyrase mediated PARdU in an overexpression system. However, its paralog, SH3BP5L was not identified in the BioID experiment, which may be a reflection of endogenous SH3BP5L levels not being regulated by PARdU in HEK293T cells. SH3BP5 and SH3BP5L overexpression dramatically increased TNKS stability which was dependent on the interaction between the

proteins. Further investigation is required to determine whether SH3BP5/L binding inhibits TNKS PARP activity like GMD or is protective against RNF146-mediated degradation under conditions where RNF146 expression is limiting (Bisht et al., 2012).

We provide data, which demonstrates that under physiological conditions SH3BP5 but not SH3BP5L is a RNF146 target in fibroblasts. SH3BP5 stabilization by Tankyrase inhibition or knockdown coupled with no effect by RNF146 silencing suggests RNF146 may not be the primary E3 ligase for SH3BP5 in certain cell-types. This finding suggests the possibility that other WWE domain-containing E3 ligases could have similar roles in PARdU and compete with RNF146 for substrates. Alternatively, upstream signaling events that regulate Tankyrase function and substrate selection may be cell type dependent as has been observed in the regulation of PARP1 by RNF146 during DNA damage (Kang et al., 2011)..

The lack of SH3BP5L degradation following PARylation conforms to other Tankyrase binding proteins that have been reported in the literature including TFR1, IRAP, TAB182, NuMa and Miki1. This point emphasizes that PARylation by Tankyrase per se does not invariably trigger PARdU (Smith et al., 1998; Chi and Lodish, 2000; Seimiya and Smith, 2002; Chang et al., 2005; Chang et al., 2005; Ozaki et al., 2012). Despite the homology between SH3BP5 and SH3BP5L, additional factors controlling their fate following TNKS mediated PARylation may be operative.

The role of RNF146 and Tankyrase in regulating the Rab11 GEF function of SH3BP5 and SH3BP5L function is investigated in chapter 3.

Chapter 3

Tankyrase-mediated Poly-ADP Ribosylation Regulates Lumen Formation Through the Rab11 GEFs SH3BP5 and SH3BP5L 3.1 Abstract In chapter 2, SH3BP5 and SH3BP5L were identified as novel RNF146 and Tankyrase substrates that were regulated by PARdU with evidence of cell-specific regulation of these proteins. In this chapter, these proteins are characterized as GEFs for Rab11a and a unique GEF domain comprising two α-helices is described. A role for these GEFs in mammalian cells is described for the first time during lumenogenesis in epithelial cells. These proteins have redundant functions and are both required for Rab11a activation and lumen formation. The function of RNF146 and Tankyrases are also examined during lumen formation and display antagonistic regulation of SH3BP5 and SH3BP5L. This is the first example where RNF146 and Tankyrase do not function in tandem to regulate substrate stability but rather RNF146 primarily regulates Tankyrase abundance in MDCK cells. These data demonstrate that Tankyrase-mediated PARylation of SH3BP5 and SH3BP5L inhibit Rab11a activation and shows the first example of Tankyrase modification directly inhibiting enzymatic activity.

3.2 Introduction

In chapter 2, SH3BP5 was identified as a PARylation-dependent substrate of RNF146 using mass spectrometry based BioID technology. Subsequent biochemical validation demonstrated that SH3BP5 and its paralog SH3BP5L were Tankyrase substrates and new targets of PARdU. SH3BP5 was originally identified as a binding partner to the SH3-domain of Bruton’s tyrosine kinase (Btk) and shown to inhibit Btk kinase activity by decreasing B cell antigen receptor tyrosine phosphorylation (Matsushita et al., 1998; Yamadori et al., 1999). SH3BP5 was later identified as a binding partner to c-Jun N-terminal kinase (JNK) by yeast two-hybrid screening and that it localized to the mitochondria with active JNK upon cell stress (Wiltshire et al., 2002). Further contributions to this finding showed SH3BP5 was enriched at the mitochondria and was required for maintaining JNK activation and reactive oxygen species (ROS) production during cell stress induced by anisomycin or endoplasmic reticulum (ER) stress by tunicamycin (Chambers and LoGrasso 2011; Win et al., 2014). SH3BP5 is enriched within the mitochondria of murine livers and its depletion is protective against acute liver injury induced by acetaminophen and tumor necrosis factor alpha (TNF-α) (Win et al., 2011). Upon cellular stress SH3BP5 is phosphorylated on the mitochondria outer membrane by JNK, which induces the displacement of the SHP-1 phosphatase and the inactivation of SRC. SRC inactivation in turn leads to impaired electron transport accumulation of ROS within the mitochondrial inner membrane (Win et al., 2016).

The C. elegans homolog of SH3BP5, REI-1, was shown to be a guanine nucleotide exchange factor (GEF) for the Rab11 GTPase (Sakaguchi et al., 2015). The authors also demonstrated that the mammalian SH3BP demonstrated exchange activity for Rab11. The C. elegans homologs REI-1 and REI-2 regulate the localization of Rab11 during cytokinesis of early embryos. In D. melanogaster there is a single SH3BP5 homolog called parcas, which also has Rab11 GEF activity (Riedel et al., 2018). There is an additional GEF in D. melanogaster called the TRAPPII complex and knockout flies of parcas and TRAPPII are viable while double knockouts are lethal. This suggests a redundant role for these Rab11 GEFs, as observed in C. elegans. SH3BP5 and SH3BP5L are thus a Rab11 family GEFs. SH3B5PLcatalyzes nucleotide exchange faster than SH3BP5 (Jenkins et al., 2018). The biological function of these two new GEFs in mammalian cells remains unknown.

The Rab11 family of small GTPases consist of Rab11a, Rab11b and Rab25 and are closely related genes. These proteins function within recycling endosomes and like other GTPases cycle between GDP and GTP bound states where the GTP bound form (Ullrich et al., 1996; Schlierf et al., 2000). Rab11 proteins facilitate recycling of proteins to the plasma membrane including transferrin receptor (TfR), epidermal growth factor receptor (EGFR), β1 integrin, α5β1 integrin, GLUT4 and α-amino-3-hydroxy-5-methylisoxaole-4-propionic acid receptor (AMPAR) (Palmieri et al., 2006; Powelka et al., 2004; Caswell et al., 2007; Welsh et al., 2007; Kelly et al., 2011). Rab11 is also involved in apical transport in polarized epithelial cells and help facilitates trafficking of E-cadherin, polymeric IgA receptor and Podocalyxin (PODXL) (Lock and Stow 2005; Desclozeaux et al., 2008; Prekeris et al., 2000; Bryant et al., 2010). Rab11 proteins mediate protein recycling through interaction with its effector proteins called Rab11-family interacting proteins (Rab11-FIPs). The interaction with Rab11-FIPs is mediated by a 20 amino acid Rab11 binding domain (RBD), which binds to the GTP bound form of Rab11 (Prekeris et al., 2001). Each Rab11-FIP specifies its own cargo to mediate distinct pathways and cellular processes. Rab11a is critical for lumen formation in epithelial cells by transporting PODXL from the basement membrane to the apical membrane. Rab11a activation is also necessary for Rab8 activation by recruiting its GEF, Rabin8 (Bryant et al., 2010). It remains unknown what proteins activate Rab11a during lumenogenesis.

In this chapter, a biological function for SH3BP5 and SH3BP5L in mammalian cells is described for the first time. SH3BP5 and SH3BP5L both activate Rab11, which is required for lumen formation in epithelial cells. A minimal catalytic GEF domain of SH3BP5 and SH3BP5L consisting of two α-helixes was characterized and shown to be highly active when expressed alone. Tankyrase affects SH3BP5 and SH3BP5L function through PARylation which resulted in impaired lumen formation and Rab11 activation. RNF146 however does not affect SH3BP5 or SH3BP5L protein stability in epithelial cells. Tankyrase expression induces SH3BP5 and SH3BP5L localization to large puncta structures that is dependent on an intact SAM domain. Tankyrase affects Rab11 localization through its interaction with SH3BP5 and SH3BP5L. I show that this effect is independent of PARylation. Collectively, the data in this chapter show that SH3BP5 and SH3BP5L are not targets of PARdU in epithelial cells but are subjected to negative regulation by Tankyrase mediated by PARylation.

3.3 Experimental Procedures

3.3.1 Cell lines

Cell lines used in this study were HEK293T, HEK293A, HEK293 T-Rex (Invitrogen), MDCKII (WT, GFP-Rab11a, GFP-Rab11a S25N, GFP-Rab11a Q70L; Gift from Dr. David Bryant, Beatson Institute, Scotland). HEK293 cells were grown in DMEM (WISENT) with 10% FBS (WISENT) and 100U/ml penicillin and 100µg/ml streptomycin (WISENT). MDCKII were grown in DMEM with 5% FBS. HEK293 T-Rex cells were maintained with 5µg/ml blasticidin (BioShop) and 100µg/ml zeocin (Invitrogen). Cells were cultured at 37⁰C in a humidified incubator set at 5% CO2.

3.3.2 Plasmids

SH3BP5 cDNA isolated from HEK293T RNA and SH3BP5L cDNA purchased from OpenFreezer (Mount Sinai Hospital) were cloned into pEBG using KpnI and NotI restriction sites. Deletions for both proteins in pEBG were created by overlap extension PCR. N-terminal and C-terminal fragments were cloned with inner primers flanking the deleted region and terminal full-length cloning primers. α-helix regions were determined based on the boundaries suggested by secondary structure prediction server PSIPRED (Ward et al., 2003). Full-length constructs were generated by PCR with KpnI and NotI cloning primers by using each fragment as template DNA. Human SH3BP5 and SH3PB5L were subcloned into pcDNA3.1 mCherry using AscI and EcoRI restriction sites. TNKS/TNKS2 mutant plasmids were kind gifts from Dr. Sebastian Guettler (Institute of Cancer Research, England). pEGFPC1-Rab11a WT and S25N were kind gifts from Dr. Richard Pagano (Addgene plasmids #12674, #12678 respectively). pEGFPC1-Rab11a Q70L was generated using QuikChange Lightning XL (Agilent). Rab11b cDNA was purchased from OpenFreezer and cloned into pEGFPC1 using XhoI and EcoRI restriction sites. Pet28a Rab11a 1-183 was subcloned using NheI/XhoI sites and kept in frame to generate a C-terminal 6xHis Tag. Human SH3BP5 and SH3BP5L were cloned into the bacterial expression vector pGEX-4T1 were cloned using EcoRI and NotI restriction sites. To generate Tag-RFP-T constructs using pTag-RFP-T-C1 plasmid (kind gift from Dr. David Bryant, University of Glasgow, Scotland), Human SH3BP5 and SH3BP5L were subcloned using

EcoRI/KpnI sites, human TNKS and TNKS2 were subcloned with the InFusion HD cloning kit by EcoRI/KpnI sites and human RNF146 was cloned using EcoRI/BamHI sites.

Table 3.1: List of constructs and cloning primers Plasmid Name Primer Sequence Primer Name Target pcDNA5 FRT.TO Flag ttaggcgcgcctgacgcggca Hu3BP5 FBirA AscI F Human BirA R118G 3BP5 ttagcggccgctcagccaatct Hu3BP5 FBirA NotI R pEBG 3BP5 ttaggtaccgacgcggcact Hu3BP5 Kpn 5' Human ttagcggccgctcagccaatct Hu3BP5 Not 3' pEBG 3BP5 101-455 ttaggtacctactgggaggca Hu3BP5 del100 Kpn5' A.A ttagcggccgctcagccaatct Hu3BP5 Not 3' 101-455 pEBG 3BP5 1-313 ttaggtaccgacgcggcact Hu3BP5 Kpn 5' A.A 1-313 ttagcggccgctcattcagaca Hu3BP5 del313 not3' pEBG 3BP5 162-313 ttaggtaccactcagagggtc Hu3BP5 163 kpn5' A.A 162-313 ttagcggccgctcattcagaca Hu3BP5 del313 not3' pEBG 3BP5Δ105-162 gaccctctgagttgcctcccagta Hu3BP5 delBAR inR A.A. 105-162 tactgggaggcaactcagagggtc Hu3BP5 delBAR outF pEBG 3BP5 1-163 ttaggtaccgacgcggcact Hu3BP5 Kpn 5' A.A 1-163 ttagcggccgcagtggcgtgat Hu3BP5 163 kpn5' pEBPG 3BP5 1-207 ttaggtaccgacgcggcact Hu3BP5 Kpn 5' A.A. 1-207 ttagcggccgccttgttgatg Hu3BP5 207 not3' pEBG 3BP5 1-260 ttaggtaccgacgcggcact Hu3BP5 Kpn 5' A.A 1-260 ttagcggccgcccgctcgtg Hu3BP5 260 not3' pEBG 3BP5 1-280 ttaggtaccgacgcggcact Hu3BP5 Kpn 5' A.A 1-280 ttagcggccgctctcatgtgctgctg pGEX4T1 Hu3BP5 280 Not3' pEBG 3BP5ΔBTK cgccatgggccaactgaaaa Hu3BP5 delBTK outF BTK Binding ttttcagttggcccatggcg Hu3BP5 delBTK inR region (A.A.193-223) pEBG 3BP5Δ100-207 gaagactccaagtccaagcctt Hu3BP5 del100-207 outF Alpha helix aaggcttggacttggagtcttc Hu3BP5 del100-207 inR 2&3 pEBG 3BP5Δ45-91 cacagctttgccccggggatc 3BP5 del45-91 inR Alpha helix 1 gtggatccccggggcaaagctgt 3BP5 del45-91 outF pEBG 3BP5Δ163-207 atgctgaattccaagccttaaggcttggaattca Hu3BP5 del163-207 outF A.A. 163-207 gcat Hu3BP5 del163-207 inR pEBG 3BP5L ttaggtaccgctgagctca Hu3BP5L Kpn5' Human ttagcggccgcctacaggctga Hu3BP5L Not3' pEBG 3BP5L 1-299 ttaggtaccgctgagctca Hu3BP5L Kpn5' A.A. 1-299 ttagcggccgcctactcgggtcct Hu3BP5L 4T1 299 Not3' pEBG 3BP5L 1-221 ttaggtaccgctgagctca Hu3BP5L Kpn5' A.A. 1-221 ttagcggccgcctacttgccgat Hu3BP5L 1-221 Not3' pEBG 3BP5LΔ58-106 aagaactggatgggagctgca Hu3BP5L del58-106 F Alpha helix 1 tgcagctcccatccagttctt Hu3BP5L del58-106 R pEBG 3BP5LΔ115-221 gaaagcccggagccgcccct Hu3BP5L del115-221 F Alpha helix aggggcggctccgggctttc Hu3BP5L del115-221 R 2&3 pGEX4T1 3BP5 ttagaattcgacgcggcactg pGEX4T1 hu3bp5 Eco5' Human ttagcggccgctctcagccaatct pGEX4T1 hu3bp5 Not3' pGEX4T1 3BP5 ttagaattcgacgcggcactg pGEX4T1 hu3BP5 Eco5' Alpha Helix 1-280Δ100-207 ttagcggccgctctcatgtgctgctg pGEX4T1 hu3BP5 280 Not3' 1&4 pGEX4T1 3BP5L ttagaattcgctgagctca Hu3BP5L 4T1 Eco5' Human ttagcggccgcctacaggctga Hu3BP5L Not3' pGEX4T1 3BP5L ttagaattcgctgagctca Hu3BP5L 4T1 Eco5' Alpha Helix 1-299Δ115-221 ttagcggccgcctactcgggtcct Hu3BP5L 4T1 299 Not3' 1&4

pEGFPC1 Rab11a Q70L ggacacagcagggctagagcgatatcgag huRab11a Q70L For Human Rab11a ctcgatatcgctctagccctgctgtgtcc huRab11a Q70L Rev Q70L pET28a Rab11a 1-183 ttagctagcggcacccgc pET28 huRab11a Nhe5' Human Rab11a Cterm 6xHis gcgctcgaggtctgacatt Hrab11a Xho3' 183R A.A. 1-183-6x- His Tag pET28a Rab11a 1-173 ttagctagcggcacccgc pET28 huRab11a Nhe5' Human Rab11a cccgctagcctagtaaatctc pET28 hurab11a Nhe3' A.A. 1-173 pcDNA3.1 mcherry 3BP5 ttaggcgcgccagacgcggcact Hu3BP5 AscI 5' Human ttagaattctcagccaatctgcac Hu3BP5 Eco 3' pcDNA3.1 mcherry 3BP5 ttaggcgcgccagacgcggcact Hu3BP5 AscI 5' TBM 1&2 ΔTBM1ΔTBM2 ttagaattctcagccaatctgcac Hu3BP5 Eco 3' pcDNA3.1 mcherry 3BP5 ttaggcgcgccagacgcggcact Hu3BP5 AscI 5' Alpha Helix 1 Δ45-91 ttagaattctcagccaatctgcac Hu3BP5 Eco 3' pcDNA3.1 mcherry ttaggcgcgccagctgagctca Hu3BP5L Asc5' Human 3BP5L ttagaattcctacaggctga Hu3BP5L Ecori3' pcDNA3.1 mcherry ttaggcgcgcca gcagaattgc H35L delTBM1.2 Asc5' TBM1&2&3 3BP5L ttagaattcctacaggctga Hu3BP5L EcoRI3' ΔTBM1ΔTBM2ΔTBM3 pcDNA3.1 mcherry ttaggcgcgccagctgagctca Hu3BP5L Asc5' Alpha Helix 1 3BP5LΔ58-106 ttagaattcctacaggctga Hu3BP5L EcoRI3' pTagRFP-T C1 3BP5 ttagaattctgacgcggcact Hu3BP5 EcoR15' Human ttaggtacctcagccaatctg Hu3BP5 Kpn3' pTagRFP-T C1 3BP5L ttagaattcagctgagctca Hu3BP5L EcoRI5' Human ttaggtaccctacaggctga Hu3BP5L Kpn3' pTagRFP-T C1 TNKS ctcaagcttcgaattctgcggcgtcgcgtcgctctga InFu TNKS Eco5' Human tcccgggcccgcggtaccctaggtcttctgct InFu TNKS Kpn3' pTagRFP-T C1 TNKS2 ctcaagcttcgaattcttcgggtcgcc InFu TNKS2 Eco5' Human gatcccgggcccgcggtaccttatccatcg InFu TNKS2 Kpn3' pTagRFP-T C1 RNF146 ttagaattctatggctggctgt huRNF146 EcoRI 5' Human ttaggatccttaaacttcagt huRNF146 BamHI 3' pMXs RFPT 3BP5 ggtggtacgggaattcaccatggtgt InFu TagRFP Eco5' RFPT 3BP5 atttacgtagcggccgctcagccaattc InFu 3BP5 Not3' pMXs RFPT 3BP5L ggtggtacgggaattcaccatggtgt InFu TagRFP Eco5' RFPT 3BP5L atttacgtagcggccgcctacaggctga InFu 3BP5L Not3' pMXs RFPT TNKS ggtggtacgggaattcaccatggtgt InFu TagRFP Eco5' RFPT TNKS atttacgtagcggccgctcaggtcttct InFu TNKS Not3' pMXs RFPT TNKS2 ggtggtacgggaattcaccatggtgt InFu TagRFP Eco5' RFPT TNKS2 atttacgtagcggccgctcatccatcga InFu TNKS2 Not3' pMXs RFPT RNF146 ggtggtacgggaattcaccatggtgt InFu TagRFP Eco5' RFPT RNF146 atttacgtagcggccgcttaaacttcagt InFu huRNF146 Not3' pGEX4T1 FIP3RBD ttagaattcggcgccaagagc pGEX4T1 rbdfip3 695 Eco5' Rab11FIP3 ttactcgagctacttgacctc pGEX4T1 rbdfip3 Xho 3' 695-756

Table 3.2: Received plasmids Plasmid Name Target Sent by pLP Myc TNKS Human TNKS Dr. Sebastian Guettler, Institute of pLP Myc TNKS E1291A Cancer Research, London, England pLP Myc TNKS G1185W pLP Myc TNKS V1056W/Y1073A pLP Myc TNKS2 E1138A Human TNKS2 Dr. Sebastian Guettler, Institute of plp Myc TNKS2 G1032W Cancer Research, London, England pLP Myc TNKS V903W/Y920A

3.3.3 Retroviral and lentiviral (shRNA and CRISPR) plasmids

TagRFP-T SH3BP5, SH3BP5L, TNKS, TNKS2 and RNF146 fusions were subcloned into the retroviral plasmid pMXs-IRES-Puro (Clontech) using NotI/EcoRI restriction sites and InFusion HD cloning kit (Clontech).

Two shRNA's against dog Rab11a and one shRNA against SH3BP5 were obtained from the TRC library (Moffat Lab). Hairpin sequences targeting dog SH3BP5 and SH3BP5L were identified using the siRNA Wizard (Invivogen). Hairpins were cloned into either pLKO.1 (Addgene #10878) or pLKO.1-hygro (Addgene #24150).

Table 3.3: shRNA sequences shRNA Sequence TRC Clone ID Plasmid RefSeq ID shscm CCTAAGGTTAAGTCGCCCTCG N.A. pLKO.1 puro N.A. Dog_Rab11a sh91 CGAGCTATAACATCAGCATAT TRCN0000303291 pLKO.1 puro NM_001003276.3 Dog_Rab11a sh44 CAGAGATATACCGCATTGTTT TRCN0000100344 pLKO.1 puro NM_001003276.3 Dog_3BP5 sh25 CATCAACAAGTCCAAGCCTTA TRCN0000122325 pLKO.1 puro XM_005634401.3 Dog_3BP5 sh3 GGCAGAAGGAGCAGAGAATAA N.A. pLKO.1 puro XM_005634401.3 Dog_3BP5L sh2 GGAAGAGCTAGAGCATCTGAA N.A. pLKO.1 hygro XM_854895.5 Dog_3BP5L sh4 GGAAGAGCTAGAGCATCTGAA N.A. pLKO.1 hygro XM_854895.5

Table 3.4: pLKO.1 cloning primers shRNA Cloning Primer Primer Sequence Dog_3BP5 sh3 Dog3BP5 sh3 5' ccggggcagaaggagcagagaataactcgagttattctctgctccttctgcctttttg Dog3BP5 sh3 3' aattcaaaaaggcagaaggagcagagaataactcgagttattctctgctccttctgcc Dog_3BP5L sh2 Dog3BP5L sh2 5' ccggggaagagctagagcatctgaactcgagttcagatgctctagctcttcctttttg Dog3BP5L sh2 3' aattcaaaaaggaagagctagagcatctgaactcgagttcagatgctctagctcttcc Dog_3BP5L sh4 Dog3BP5L sh4 5' ccggggaagagctagagcatctgaactcgagttcagatgctctagctcttcctttttg Dog3BP5L sh4 3' aattcaaaaaggaagagctagagcatctgaactcgagttcagatgctctagctcttcc

Single guide RNA (sgRNA) sequences targeting dog 3BP5, 3BP5L, TNKS, TNKS2 and RNF146 were selected from the CRISPOR design tool (www.crispor.tefor.net) by entering the cDNA sequence of specific exons (Haeussler et al., 2016). Guide sequences were cloned into the plasmid pLCKO using BfuAI and NsiI sites.

Table 3.5: Guide RNA (sgRNA) Sequences sgRNA Sequence Refseq ID and Target Exon Dog_3BP5 sg5 AGCGACAGTTCGACTCCGCC XM_005634401.3 Exon4 Dog_3BP5L sg3 GCAAGAAAGCTCAACACGCA XM_854895.5 Exon3 Dog_TNKS sg2 GACCTCCATCATCACGAGCA XM_844295.5 Exon2

Dog_TNKS2 sg3 CTGTTCGAGGCGTGCCGCAA XM_003639982.4 Exon1 Dog_RNF146 sg2 GAATATGCGTGGTATTACGA XM_533493.6 Exon3 Dog_RNF146 sg3 TCTCGCTAGGTTGACCGTGC XM_533493.6 Exon3

Table 3.6: pLCKO.1 cloning primers shRNA Cloning Primer Primer Sequence Dog_3BP5 sg5 Dog3BP5 sg#5 For ACCGAGCGACAGTTCGACTCCGCC Dog3BP5 sg#5 Rev AAACGGCGGAGTCGAACTGTCGCT Dog_3BP5L sg3 Dog3BP5L sg#3 For ACCGGCAAGAAAGCTCAACACGCA Dog3BP5L sg#3Rev AAACTGCGTGTTGAGCTTTCTTGC Dog_TNKS sg2 DogTNKS sg#2 For ACCGGACCTCCATCATCACGAGCA DogTNKS sg#2Rev AAACTGCTCGTGATGATGGAGGTC Dog_TNKS2 sg3 DogTNKS2 sg#3 For ACCGCTGTTCGAGGCGTGCCGCAA DogTNKS2 sg#3 Rev AAACTTGCGGCACGCCTCGAACAG Dog_RNF146 sg2 DogRNF146 sg#2 For ACCGGAATATGCGTGGTATTACGA DogRNF146 sg#2 Rev AAACTCGTAATACCACGCATATTC Dog_RNF146 sg3 DogRNF146 sg#3 For ACCGTCTCGCTAGGTTGACCGTGC DogRNF146 sg#3 Rev AAACGCACGGTCAACCTAGCGAGA

3.3.4 Retrovirus and lentivirus generation and transduction

Viruses were generated by transfecting 5x105 HEK293T cells in 6-well plates with 50ng envelope, 450ng packaging plasmids along with 500ng viral plasmids using X-tremegene 9 (Roche). VSV-G and psPAX2 were used as envelope and packaging plasmids for lentiviral particles while VSV-G and Gag-pol were used for retroviral particles. After 24 hours of transfection, the media was replaced with target cell media. Viral supernatants were collected 48 hours post-transfection and cleared through 0.45μm filters. Target cells were transduced with viral supernatants along with 8µg/ml polybrene for 24 hours. Antibiotic selection was started 24 hours post-infection for a further 48 hours in fresh media. For MDCK cells, 5µg/ml puromycin, 500µg/ml hygromycin, and 5µg/ml blasticidin were used for selection.

3.3.5 MDCK stable expression cell line generation

Stable MDCK cells expressing Tag-RFP-T-SH3BP5, Tag-RFP-T-SH3BP5L, Tag-RFP-T-TNKS, Tag-RFP-T-TNKS2 and Tag-RFP-T-RNF146 were generated by retroviral infection. Surviving cells that had gone through selection with puromycin, were sorted by fluorescence activated cell sorting (FACS) for medium and high expression of RFP. Cells from each population were single cell cloned in 96-well plates and uniform expressing cell lines were selected. These cell lines were expanded and used for downstream experiments.

3.3.6 MDCK knockout lines created by CRISPR

MDCKII-Cas9 cells were generated by infection of Lenti-Cas9-2A lentiviral particles, expressed from the Lenti-Cas9-2A plasmid (Hart et al., 2015) with 8μg/ml polybrene. 24 hours after infection, viral media was replaced with fresh culture media. Cells were placed under selection with 5μg/ml blasticidin for 48 hours. Surviving cells were checked for Flag-Cas9 expression by immunoblot analysis and then seeded into 96-well plates to achieve single cell clones. Clones were expanded and again tested for Cas9 expression. Clonal lines that maintained Cas9 were subsequently used for gene knockout studies.

To generate 3BP5/3BP5L and TNKS/TNKS2 double knockout lines, equal volumes of guide RNA lentiviruses for each gene were used to infect MDCKII-Cas9 cells with 8μg/ml polybrene for 24 hours. Single guide RNA viruses were used to create RNF146 knockout lines using the same infection parameters. Cells were placed under selection with 5μg/ml puromycin and 5μg/ml blasticidin for 48 hours. Surviving cells were seeded into 96-well plates to achieve single cell clones. Clones were analyzed by western blotting and clones exhibiting complete deletion of protein were expanded and used for further study.

3.3.7 Immunoblotting and immunoprecipitation

For immunoblotting, cells were lysed in Triton X-100 buffer containing 1% Triton X-100, 25mM Tris pH 7.5, 100mM NaCl, 1mM EDTA) with 1x Halt protease and phosphatase cocktails (ThermoFisher Scientific). Lysates were cleared by centrifugation and quantified using the 660nm protein assay kit (ThermoFisher Scientific).

Immunoblotting was performed by separating samples by SDS-PAGE, transferring to PVDF membrane (Millipore), blocking with 5% skim milk (BioShop) in 1x PBS + 0.1% Tween 20 (BioShop) and incubating with indicated primary antibodies. Primary antibodies were washed off with three washes of 1x PBS + 0.1% Tween 20. Membranes were incubated with horseradish peroxidase conjugated mouse and rabbit antibodies (BioShop) for one hour, washed thrice with 1x PBS + 0.1% Tween 20 and developed. ECL or ECL prime (GE Healthcare) and SuperSignal West Pico PLUS or SuperSignal West Femto (ThermoFisherScientific) were used as HRP substrates. Membranes were imaged using a MicroChemi 2.0 chemiluminescent imager (DNR Bio-Imaging Systems).

3.3.8 MDCK Cyst culture

8-well chamber slides (BD Falcon; #354108) were coated with 10µL of 100% matrigel (Corning; #356230). 300µl of a 1.5x104 cells/ml suspension with 2% matrigel was plated on top of solidified matrigel. Cysts were grown for four days before immunofluorescence analysis. In experiments using TNKS656 inhibitor, the compound was added during plating and kept for 48 hours before being replenished for another 48 hours.

3.3.9 Immunofluorescence

Cysts grown for four days were fixed with 4% paraformaldehyde (PFA) at room temperature for 25 minutes. Cysts were washed twice with 1x PBS with magnesium and calcium and blocked in 1x PBS (Mg2+, Ca2+), 1% BSA (BioShop), 5% goat serum (Abcam ab7481) and 0.025% saponin (Sigma S4521) for an hour. Cysts were incubated with primary antibodies overnight at room temperature. Chamber slides were washed three times with 1x PBS (Mg2+, Ca2+) and incubated with Alexa Fluor 488, 568, or 647 secondary antibody conjugates (Invitrogen) for 1 hour at room temperature. Slides were washed three times with 1x PBS and incubated with DAPI (Molecular Probes) for 10 minutes at room temperature. Slides were washed twice with 1x PBS and mounted with ProLong Gold Antifade (Molecular Probes).

For localization studies, 2.0x105 293A cells were plated on 1mm coverslips in a 12-well plate and transfected with 500ng of DNA using LipoD293 for 48 hours. Cells were fixed with 4% paraformaldehyde at room temperature for 10 minutes. Cells expressing GFP and mCherry were washed with PBS and permeabilized with 0.1% Triton X-100 for 5 minutes followed by 66

incubation with DAPI before being mounted on glass slides with ProLong Gold Antifade (Invitrogen). For cells expressing Myc-tagged Tankyrase, cells were blocked after PFA fixation for one hour with 0.1% Triton X-100, 1% BSA and 5% goat serum, followed by incubation with anti-Myc 9E10 (Santa Cruz) for one hour. Cells were washed thrice with 1x PBS and incubated with Alexa Fluor 647 for one hour in 5% goat serum. Cells were washed thrice again, incubated with DAPI and mounted on glass slides.

3.3.10 Confocal microscopy

MDCK cysts were imaged with a Leica DMi8 spinning-disk confocal microscope using a 63x (1.4NA) oil objective and Volocity software (Perkin Elmer). Z-stacks were taken using 0.3μm sections for each cyst.

293A cells were imaged with an Olympus IX81 inverted microscope using a 60x oil (1.4NA) Plan Apo objective (Nikon) and Fluoview software (Olympus).

3.3.11 Single lumen quantification

MDCK cysts were quantified as previously described (Martin-Belmonte et al., 2007). At least 100 cysts per condition were counted and classified as having single or multiple lumens. Cysts that were derived from a single cell and displayed correct basolateral polarity (via DAPI and b- catenin staining) were counted using the epifluorescence mode on the Olympus IX81 inverted microscope using a 20x air (0.75) Plan Apo objective (Nikon). The single lumen formation rate was calculated by determining the percentage of cysts with a single lumen and normalizing to the control percentage for each condition. Single lumen rates are displayed as the mean ± standard deviation of three independent experiments. Statistical significance was calculated by using a two-tailed paired t-test using GraphPad Prism 8. * indicates p<0.05, ** indicates p<0.01, *** indicates p<0.001, and **** indicates p<0.0001.

3.3.12 Protein expression and purification 6xHis-Rab11a 1-183-6xHis was expressed in E. coli BL21 codon plus cells. A 15N labeled Rab11a 1-183-6xHis was prepared as follows: a 50ml culture in LB broth was grown at 37⁰C overnight, pelleted the next day and resuspended in two litres of M9 minimal media (42mM

Na2HPO4, 22mM KH2PO4, 9mM NaCl, 1% glucose, 0.3mM CaCl2, 1mM MgSO4, 1mg/ml 67

biotin, 1mg/ml thiamine, and 1x trace elements). M9 labeling media was supplemented with 19mM 15N ammonium chloride. Cells were grown until an optical density at 600nm wavelength of 0.6, cooled to 15⁰C and protein expression induced with 0.25mM IPTG overnight at 15⁰C. Cells were centrifuged and stored at -80⁰C. Frozen pellet was resuspended in ice-cold lysis buffer (50mM Tris pH8, 150mM NaCl, 15mM imidazole, 2mM MgCl2, 1mM PMSF, 0.2mg/ml lysozyme, and 10mM β-mercaptoethanol). Cells were lysed by sonication and lysate was cleared by centrifugation at 20,000 rpm for 30 minutes at 4⁰C. Cleared lysate was passed through a 0.45μm filter before incubation with 5ml Ni-NTA resin (Qiagen) for 1 hour at 4⁰C. The resin was washed with 25 column volumes of lysis buffer (without PMSF and lysozyme) and protein was eluted with six column volumes of elution buffer (50mM Tris pH8, 150mM NaCl, 250mM imidazole, 2mM MgCl2, 10mM β-mercaptoethanol). The N-terminal 6xHis tag was removed by incubating eluted protein with thrombin (10u/ml of protein) overnight by dialysis in buffer containing 50mM Tris pH8, 150mM NaCl, 5mM MgCl2and 10mM β-mercaptoethanol. Protein was purified by size exclusion on a Superdex S75 26/60 column in dialysis buffer and elution peak fractions were pooled and concentrated to 6.5mg/ml and flash frozen in liquid nitrogen and stored at -80⁰C. Rab11a copurifies with GDP derived from E. coli. pGEX-4T1 SH3BP5 (1-280Δ100-207) and SH3BP5L(1-299Δ115-221) helix 1 and 4 constructs were expressed in BL21 codon plus cells. Cells were grown in LB until O.D. 600 reading of 0.6, cooled to 15⁰C and protein expression induced with 0.25mM IPTG overnight at 15⁰C. Cultures were pelleted by centrifugation at 6000 rpm for 20 minutes and kept at -80⁰C. Cell pellets were resuspended with ice-cold lysis buffer (50mM Tris pH8, 150mM NaCl, 10% glycerol, 0.2mg/ml lysozyme, 1mM PMSF, and 1mM DTT). Cells were lysed by sonication and lysates were cleared by centrifugation. Filtered lysates were incubated with 5ml glutathione agarose (ThermoFisher Scientific) overnight. Resin was washed with 25 column volumes of lysis buffer and the GST tag was cleaved off the resin with thrombin overnight. Cleaved protein was recovered, purified by gel filtration on a Superdex S75 26/60 column, concentrated and flash frozen with liquid nitrogen and stored at -80⁰C. For full length 3BP5 and 3BP5L, the GST tag was left on and proteins were eluted with 5 column volumes of elution buffer (50mM Tris pH8, 100mM NaCl, 10mM reduced glutathione). Eluted protein was dialyzed overnight in buffer (50mM Tris pH8, 100mM NaCl, 1mM DTT), concentrated and frozen the next day. 68

3.3.13 Rab11-GTP pulldown assay Rab11-GTP levels were detected through immunoprecipitation of cell lysates with the RBD of Rab11FIP3. Amino acids 695-756 of Rab11FIP3 were cloned into pGEX4T1 using EcoRI and XhoI restriction sites. GST-FIP3RBD was expressed in E. coli BL21 codon plus cells (Eathiraj et al., 2006). Cells were grown in LB until O.D. 600 reading of 0.6, cooled to 15⁰C and protein expression induced with 0.25mM IPTG overnight at 15⁰C. Cultures were pelleted by centrifugation at 6000 rpm for 20 minutes and kept at -80⁰C. Cell pellets were resuspended Cells were resuspended in lysis buffer (50mM Tris pH7.5, 100mM NaCl, 5mM MgCl2, 1mM DTT, 1mM PMSF, and 0.2mg/ml lysozyme). Cells were lysed using sonication and the lysate was cleared by centrifugation. Cleared lysate was incubated with glutathione agarose for one hour at 4⁰C and the resin was subsequently washed with 25 column volumes of lysis buffer (without PMSF and lysozyme). Protein was eluted using three column volumes of buffer containing 50mM Tris pH8, 100mM NaCl, and 10mM reduced glutathione. The eluted protein was dialyzed overnight to remove the reduced glutathione using buffer containing 50mM Tris pH7.5, 100mM

NaCl and 5mM MgCl2.

For Rab11-GTP pulldown experiments, cells were lysed in 50mM Tris pH8, 100mM NaCl, 1%

Triton X-100, 5mM MgCl2, 0.5mM EDTA, Halt protease and phosphatase inhibitor cocktails for 25 minutes on ice. Cleared lysates were quantified using the Pierce 660nm protein assay reagent and 500μg of cell lysate was incubated with 20μg of GST-FIP3RBD and 20μl of glutathione agarose for one hour. The agarose was washed thrice with lysis buffer then resuspended in 20μl of 2x Laemmli buffer, boiled for 5 minutes at 100⁰C then subjected to immunoblot analysis. Rab11a-GTP levels were quantified with ImageJ using blots from three independent experiments. Rab11a-GTP levels were normalized to total Rab11a protein and each experiment was normalized to the control.

3.3.14 Antibodies and reagents

The following antibodies were used for immunoblotting: GFP (1:2000; Abcam ab290), GST (1:1000; Santa Cruz sc138), SH3BP5 (1:300; Sigma HPA036445), SH3BP5L (1:500; Novus Biologics NBP2-38385), Tankyrase1/2 (1:500; Santa Cruz sc365897), RNF146 (1:1000; Sigma SAB1408054), Rab11 (1:500; ThermoFisher Scientific 71-5300), Cas9 (1:500; Santa Cruz 69

sc517386). For immunofluorescence, antibodies used were: β-catenin (1:500; Cell Signaling Technologies #9562), Podocalyxin (1:1000, clone 3F2, Millipore MABS1327), Rab11 (1:250, ThermoFisher Scientific 71-5300). The reagents used were: TNKS656 inhibitor was a kind gift from Dr. Feng Cong (Novartis), Biotin-NAD+ was purchased from Trevigen (cat# 4670-500-01).

3.3.15 Rab11a nucleotide exchange assay by NMR

Rab11a nucleotide exchange was monitored using a real-time NMR-based assay that was described previously (Gasmi-Seabrook et al., 2010; Marshall et al., 2009). A series of 1H-15N heteronuclear single quantum coherence (HSQC) spectra of 15N-labeled Rab11a were collected sequentially during the exchange reaction, and the heights of peaks that are specific to the GDP- or GTP-bound forms of the GTPase protein were used to determine exchange rates. A 40μl sample was prepared by mixing 300μM GDP-loaded 15N-Rab11a and 4.5mM excess of GTPγS (Guanosine 5′-[γ-thio] triphosphate tetralithium salt, Sigma-Aldrich) in a of 1.7 mm NMR tube. Sequential 1H-15N HSQC NMR experiments (10 min each) were collected throughout the time course of the exchange reaction using a Bruker 600 MHz Avance III NMR spectrometer equipped with a 1.7 mm cryogenic TCI MicroCryoProbe. Data were analyzed using Bruker TopSpin. The half-life of the exchange reaction was determined as the point at which GDP- and GTP-specific peaks exhibited equal heights, and nucleotide exchange rates were calculated from the half-life based on single-phase exponential decay (rate = ln 2 / half-life). To determine whether SH3BP5 and SH3BP5L possessed GEF activity and investigate which parts of the protein are required to mediate nucleotide exchange, the experiment was repeated with the addition of a series of purified recombinant SH3BP5 fragments as indicated.

3.3.16 SH3BP5 BioID Assay

Full length human SH3BP5 cDNAs were cloned into pcDNA5 FRT.TO FlagBirA R118G using AscI and NotI restriction enzymes. HEK293 Flp-In T-Rex cell lines stably expressing Flag-BirA SH3BP5 were created by selection in normal media supplemented with 5µg/ml blasticidin and 200µg/ml hygromycin B. Five 15cm plates of cells per condition were grown to 80% confluency prior to induction of protein expression with 1µg/ml tetracycline (Sigma) and 50µM biotin (BioBasic) for 24 hours. Cells were scraped, pelleted by centrifugation and washed with 1x PBS three times and pellets were stored at -80⁰C. Samples were prepared as described in Chapter 2.

3.3.17 Mass spectrometry analysis

Mass spectrometry analysis was performed as described in Chapter 2. Ten control runs (from cells expressing the FlagBirA* epitope tag) were collapsed to the two highest spectral counts for each prey and compared to one run of two biological replicates of SH3BP5 BioID.

3.3.18 Interaction network analysis

The SH3BP5 interaction network was created using Cytoscape v3.2 and importing known interaction data with the GeneMANIA plugin (Shannon et al., 2003; Warde-Farley et al., 2010). Proteins were grouped according to biological function through gene ontology (http://geneontology.org) and by manual literature search curation (Ashburner et al., 2000; Gene Ontology Consortium, 2015).

3.4 Results

3.4.1 Identification of SH3BP5 interacting proteins

To elucidate the mechanistic basis of SH3BP5 in controlling biologic processes, BioID was used to identify the ensemble of SH3BP5 interacting proteins. The HEK293 Flp-In line stably expressing Flag-BirA-SH3BP5 was generated. HEK293 harboring the Flag-BirA-SH3BP5 bait was treated with 1μg/ml of tetracycline (to induce protein expression) and 50μM biotin (substrate for BirA) for 24 hours. Cells were lysed in a 1% Triton X-100 lysis buffer and the lysates were subjected to immunoblot analysis. Membranes were probed with Flag antibody to check for expression of the bait protein and streptavidin-HRP to detect biotinylation of the SH3BP5 fusion protein (Figure 3.1). Once validated, the cell lines were used for the BioID experiment.

Figure 3.1: Generation of SH3BP5 BirA cell line. 293T Flp-In line validation for expression of Flag-BirA SH3BP5. Flag-BirA expression was induced with 1µg/ml tetracycline and 50µM biotin for 24 hours. Whole cell lysates were analyzed by immunoblotting for Streptavidin HRP to detect protein biotinylation and Flag expression.

SAINT score and spectral sum were used to rank order SH3BP5 interactors. Both Tankyrases were the top significant interactors. Thirty-six significant interactors were identified using a cutoff SAINT threshold of 0.9 (Table 3.7). Of these, both TNKS and TNKS2 were within the top hits, ten interactors were related to vesicle trafficking and specifically protein involved with the GTPase Rab11a (Figure 3.2). This result corroborates a recent publication showing that SH3BP5 functions as a guanine exchange factor for Rab11a (Sakaguchi et al. 2015). Given the unknown function of SH3BP5 towards Rab11a in mammalian cells, the roles of SH3BP5 and the RNF146/Tankyrase axis were explored as regulators of Rab11a. To validate the interactions between SH3BP5 and Rab11a, Rab11a mutants that stabilize GDP or GTP bound forms were tested for SH3BP5 binding. We additionally tested its paralog, SH3BP5L, for Rab11a interaction. GST-SH3BP5 or GST-SH3BP5L were expressed with either GFP-Rab11a, or GFP- Rab11a-S25N (GDP locked form) or GFP-Rab11a-Q70L (GTP locked form) (Ullrich et al., 1996) in HEK293T cells. Rab11a immune complexes were immunoblotted for GST-SH3BP5 or GST-SH3BP5L. Both SH3BP5 and SH3BP5L bound with greater affinity to the Rab11a-GDP form (Rab11a-S25N), which recapitulated previous findings (Figure 3.3) (Sakaguchi et al. 2015).

Table 3.7: List of significant SH3BP5 interacting proteins SH3BP5 Gene Controls Counts Sum SAINT TNKS 0|0|0|0|0|0|0|0 101|97 198 1.00 RAB11FIP1 5|3|3|4|10|11|15|15 89|91 180 1.00 RAB11FIP5 0|0|0|0|0|2|3|2 60|58 118 1.00 DST 20|18|15|15|4|3|17|10 52|65 117 0.93 EXOC3 0|0|2|0|4|4|17|17 59|53 112 1.00 TNKS2 0|0|0|0|0|0|0|0 39|37 76 1.00 EPHA2 7|5|8|5|4|5|8|7 33|36 69 1.00 ARL13B 0|0|0|0|2|0|2|3 36|28 64 1.00 MIS18BP1 0|0|0|0|0|0|0|0 32|32 64 1.00 N4BP2 2|2|0|2|0|0|0|0 23|23 46 1.00 RAB11A 0|0|0|0|0|0|0|0 12|11 23 1.00 LRP8 0|0|0|0|0|0|0|0 11|10 21 1.00 UHRF1BP1L 0|0|0|0|0|0|0|0 10|10 20 1.00 STEAP3 0|0|0|0|0|0|0|0 9|10 19 1.00 KIAA1549 0|0|0|0|0|0|0|0 10|7 17 1.00 PARP14 0|0|0|0|0|0|0|0 12|5 17 1.00 TMEM2 0|0|0|0|0|0|0|0 6|7 13 1.00 GSK3A 0|0|0|0|0|0|0|0 6|7 13 1.00 COX4I1 0|0|0|0|0|0|0|0 8|3 11 0.98 TBC1D12 0|0|0|0|0|0|0|0 7|4 11 1.00 HIP1 0|0|0|0|0|0|0|0 3|6 9 0.98 GSK3B 0|0|0|0|0|0|0|0 5|3 8 0.98 LRP2 0|0|0|0|0|0|0|0 2|5 7 0.89 RAB3IL1 0|0|0|0|0|0|0|0 5|2 7 0.89 TBC1D14 0|0|0|0|0|0|0|0 4|3 7 0.98 LNPEP 0|0|0|0|0|0|0|0 4|3 7 0.98 HLA-A 0|0|0|0|0|0|0|0 3|4 7 0.98 KDELR1 0|0|0|0|0|0|0|0 3|4 7 0.98 RNF114 0|0|0|0|0|0|0|0 4|3 7 0.98 RAC1 0|0|0|0|0|0|0|0 3|3 6 0.96 APC 0|0|0|0|0|0|0|0 3|3 6 0.96 ARFGEF2 0|0|0|0|0|0|0|0 3|3 6 0.96 SLC44A4 0|0|0|0|0|0|0|0 3|3 6 0.96 METAP2 0|0|0|0|0|0|0|0 3|3 6 0.96 ATP9A 0|0|0|0|0|0|0|0 3|3 6 0.96 LMTK2 0|0|0|0|0|0|0|0 3|3 6 0.96

Figure 3.2: SH3BP5 interactome reveals association to Rab11 trafficking. Interaction network of significant SH3BP5 interactors identified by BioID. The interaction network was created using Cytoscape and grouped by biological function using the Gene Ontology database. Hexagon size indicates the spectrum sum of each BioID run. Solid black lines indicate known protein interactions imported from the GeneMANIA plugin and dotted lines represent new interactions identified by BioID.

Figure 3.3: SH3BP5 and SH3BP5L interact with Rab11a-GDP. HEK293T cells were co- expressed with GFP-Rab11a WT, S25N (SN), Q70L (QL) and GST-3BP5 (left panel) or GST- 3BP5L (right panel) and lysates were immunoprecipitated with GFP antibody. Immune complexes were analyzed by immunoblotting for GST and GFP. Whole cell lysates were analyzed for GST, GFP and α-Tubulin (loading control).

3.4.2 SH3BP5 and SH3BP5L are Rab11a GEFs that mediate exchange through a novel GEF domain

To validate SH3BP5/L as GEFs, a real-time exchange assay by NMR was used to study the kinetics of Rab11a nucleotide exchange. 15N labeled Rab11a was monitored by collecting 1H-15N HSQC spectra during the time course of nucleotide exchange, and two specific pairs of peaks that are specific to Rab11a-GDP or Rab11a-GTP were used to monitor the exchange reaction and determine exchange rates (Figure 3.4A). Recombinant SH3BP5 and SH3BP5L were added to determine whether they accelerate nucleotide exchange. Both proteins were observed to be potent GEFs as a GTPase:GEF molar ratio of 6000:1 was sufficient to enhance nucleotide exchange rate ~4-fold (Figure 3.4B). Serial deletions of SH3BP5 and SH3BP5L were made to identify the minimal regions within the N-terminus that mediated the Rab11a interaction. Each deletion was co-expressed with GFP-Rab11a-S25N in HEK293T cells and immunoprecipitated Rab11a complexes were subjected to immunoblot analysis. These results revealed that regions within the first 100 amino acids and BTK domain were required for binding while other intermediate regions were not required (Figure 3.5A, B).

Figure 3.4: SH3BP5 and SH3BP5L are Rab11a GEFs. A) Overlay of HSQC spectra of 1H- 15N Rab11a in its GDP (black) and GTP (red) bound forms. The two peaks monitored during exchange are highlighted in black boxes. B) Comparison of nucleotide exchange rates of Rab11a between its intrinsic activity and addition of recombinant full length 3BP5 or 3BP5L. The GTPase:GEF molar ratio used was 6000:1.

Secondary structure prediction of the N-terminal region suggested it is highly helical, comprising four separate α-helices that are structurally similar or homologous to known α-helical GEF domains like Sec2p and Rabin8/GRAB, which form dimeric coiled-coil domains (Dong et al., 2007; Guo et al., 2013). To understand how these helices contributed to catalysis, deletion analysis was performed. These studies showed that that helix 1 and helix 4 in SH3BP5 and SH3BP5L are both required for interaction to Rab11a-GDP (S25N) (Figure 3.6A). The crystal structure of SH3BP5 with Rab11a was recently solved and confirmed that these two helices form the interface with Rab11a (Jenkins et al., 2018). The minimal binding unit containing recombinant helix 1 and helix 4 (1-280Δ100-207 for SH3BP5; 1-299Δ115-221 for SH3BP5L) also mediated exchange at slightly faster rates in comparison to full length proteins (Figure 3.6B). These results demonstrate that SH3BP5 and SH3BP5L are Rab11a GEFs and mediate catalysis through a novel two-helix GEF domain.

Figure 3.5: Entire SH3BP5 N-terminus is not required for Rab11a binding. A,B) Left panel: GST-tagged full length and deletions of 3BP5 co-transfected with GFP-Rab11a S25N in HEK293T cells. Lysates were immunoprecipitated with GFP antibody and analyzed by western blot for GST and GFP. Right panels: illustration of each expression construct with domains highlighted. Red boxes 1-4 represent alpha-helices 1-4 respectively.

Figure 3.6: SH3BP5 and SH3BP5L have a novel two-helix catalytic domain. A) Left panels: GST-tagged full length and deletions of SH3BP5/L co-transfected with GFP-Rab11a S25N in HEK293T cells. Lysates were immunoprecipitated with GFP antibody and analyzed by western blot for GST and GFP. Right panels: illustration of each expression construct with domains highlighted. Red boxes 1-4 represent alpha-helices 1-4 respectively. B) Nucleotide exchanges rates of Rab11a comparing recombinant full length 3BP5 and 3BP5L with their respective minimal two-helix catalytic domain.

3.4.3 Tankyrase alter localization of SH3BP5/SH3BP5L and Rab11a

To assess the biological significance of how SH3BP5 and SH3BP5L regulate Rab11 function we performed localization studies. mCherry SH3BP5 and SH3BP5L fusion proteins were made to examine their subcellular localization. Both proteins displayed cytoplasmic distributions in HEK293A cells (Figure 3.7A). GFP-Rab11a localization changed from punctate vesicular structures to a diffuse cytoplasmic distribution upon overexpression of either mCherry SH3BP5 or SH3BP5L. The shift in the distribution of Rab11a is due to physical interaction with SH3BP5/L since deletion of one of the α-helices that is required for binding impaired the ability of SH3BP5/L to perturb Rab11a localization (Figure 3.7A, B).

Figure 3.7: SH3BP5 and SH3BP5L alter localization of Rab11a. A) Immunofluorescence analysis of GFP-Rab11a, Cherry 3BP5 and 3BP5L transfected into 293A cells for 48 hours prior to fixation. Scale bar = 10µM and 60x magnification. B) Co-localization of GFP-Rab11a with Cherry SH3BP5 or SH3BP5L that is dependent on binding through α-helix 1. Constructs were transfected in 293A cells for 48 hours prior to fixation. Scale bar = 10µM and 60x magnification.

To identify a role in which Tankyrases regulate SH3BP5/L, we examined the impact on their subcellular localization when Tankyrase was expressed. Tankyrases are known to form large aggregates in cells due to its ability to form oligomers through its SAM domain (De Rycker and Price 2004; Mariotti et al., 2016). Myc-TNKS formed large puncta in HEK293A cells when ectopically expressed (Figure 3.8A). Cherry SH3BP5 or SH3BP5L both shift from cytoplasmic to punctate localizations with Tankyrase coexpression. The ability of Tankyrase to trigger the relocalization of SH3BP5/L was contingent on Tankyrase binding, since SH3BP5/L-ΔTBM mutants were impervious to Tankyrase expression (Figure 3.8B, C). To determine whether the Tankyrase induced puncta was PARylation or SAM domain dependent, the localization of SH3BP5 and SH3PB5L was observed after co-expression of PARP domain or SAM domain Tankyrase mutants. The PARP domain mutant caused larger punctate aggregates to form while the SAM domain mutant abrogated the formation of punctate structures (Figure 3.8B, C). These results reveal that Tankyrase induced relocalization of SH3BP5/L is dependent on Tankyrase binding and SAM domain mediated oligomerization. Taken together these data demonstrate that Tankyrases can sequester binding partners into large macromolecular protein complexes in discrete subcellular structures.

To assess the effect of Tankyrase on Rab11a localization, Myc-TNKS, GFP-Rab11a, and Cherry SH3BP5 or SH3BP5L were co-expressed HEK293A. We observed that all proteins colocalized into large punctate structures distinct from the small vesicular structures where GFP-Rab11a normally resides (Figure 3.9B, C first column). SH3BP5 and SH3BP5L are required components for Tankyrase triggered relocalization of Rab11a since expression of Tankyrase alone has no effect on Rab11a localization (Figure 3.9A). Tankyrase PARP domain and SAM domain mutants were required to mediated Rab11a relocalization. The SH3BP5/L α-helix mutants were unable to trigger Rab11a relocation demonstrating that this process was contingent on direct interaction between Rab11a and SH3BP5/L. Similarly, the Tankyrase effect was dependent on direct binding of SH3BP5/L as the SH3BP5/SH3BP5L-ΔTBMs mutants interfered with Tankyrase driven relocalization of Rab11a (Figure 3.9B, C). These data demonstrate that Tankyrase, SH3BP5/L and Rab11a form a stable tertiary complex when overexpressed in cells with SH3BP5/L serving as a scaffold function in linking Rab11a with Tankyrase.

Figure 3.8: Tankyrase alters SH3BP5/L localization into punctate structures. A) Immunofluorescence analysis of Myc-TNKS transfected into 293A cells for 48 hours. B, C) Localization analysis of SH3BP5 and SH3BP5L or TNKS binding mutant co-expressed with Myc-TNKS wildtype, SAM mutant and PARP mutant. Scale bar = 10µM and 60x magnification.

Figure 3.9: Tankyrase alters Rab11a localization through SH3BP5 and SH3BP5L. A) Immunofluorescence analysis of 293A cells expressing GFP-Rab11a with Myc-TNKS. B, C) Co-expression of GFP-Rab11a and Cherry 3BP5 or 3BP5L with Myc-TNKS and indicated mutants. 293A cells were transfected for 48 hours prior to fixation and immunofluorescence analysis. Scale bar = 10µM and 60x magnification.

3.4.4 SH3BP5 and SH3BP5L promote lumen formation in MDCK cysts

Rab11a is required for the transport of Podocalyxin (PODXL) from the basolateral membrane to the apical membrane during lumen formation (Bryant et al., 2010). The role of SH3BP5 and SH3BP5L during lumenogenesis is unknown. Since Rab11a-GTP is critical for lumen formation, we hypothesized that SH3BP5 and SH3BP5L are the Rab11a exchange factors that promote epithelial polarity during lumen formation. We first tested whether SH3BP5/L could stimulate Rab11a activation by creating MDCK cells stably expressing Tag-RFP-T SH3BP5 and SH3BP5L. Both proteins showed cytoplasmic localization when grown in 2D (Figure 3.10A). Rab11-GTP levels were evaluated using a Rab11FIP3-derived Rab11 binding domain pull down assay (Figure 3.10B, C). Examination of MDCK cysts cultured for four days in matrigel, showed expression of either SH3BP5 or SH3BP5L increased the frequency of single lumen cysts compared to control cells comparable to overexpressed Rab11a or constitutively active Q70L mutant (Figure 3.10E, F). SH3BP5 showed a stronger strong apical membrane localization than SH3BP5L and neither co-localized with PODXL (Figure 3.10D).

To determine whether these GEFs were critical for lumen formation, shRNA knockdowns of each gene and in combination were performed. Both single knockdowns and double knockdown of each GEF showed minimal effect on lumen formation (Figure 3.11A-H). To address the issue of redundancy, CRISPR was used to generate double knockout MDCKII cell lines as shown by immunoblot analysis (Figure 3.12A). The double knockout cells had decreased Rab11a-GTP loading in comparison to control cells and single knockout lines and the residual active Rab11a indicates the possibility of additional GEFs and/or mechanisms by which Rab11a becomes activated (Figure 3.12B, C). Double knockout cysts had a higher decrease in single lumen percentage in comparison to single knockout cysts (Figure 3.12D, E). These data show that

SH3BP5 and SH3BP5L have redundant function and are both necessary for Rab11a activation during lumenogenesis.

Figure 3.10: SH3BP5 and SH3BP5L promote lumen formation by activating Rab11a. A) 2D confocal images of MDCK cells expressing Tag-RFP-T-3BP5 and Tag-RFP-T 3BP5L. 63x magnification and scale bar represents 10µm. B) Rab11a-GTP loading analysis by immunoprecipitating 500µg of lysate from each cell type with 20µg GST-FIP3RBD and analyzing immune complexes by western blot for Rab11a. Whole cell lysates were analyzed for SH3BP5, SH3BP5L and Rab11a expression. C) Quantification of Rab11a-GTP pulldowns by ImageJ. Rab11a-GTP levels for each condition were normalized to total Rab11a protein and then normalized to pMX control. Data represents mean ± SD of three independent experiments. D) Representative confocal images of 4-day cysts that were incubated with PODXL and β- catenin antibodies. Scale bar = 10µm and 63x magnification. E) Quantification of single lumen cysts normalized to pMX control single lumen rate. Data represents the mean ± SD of three independent experiments. * indicates p>0.05. pMX n=397, RFPT 3BP5=395, RFPT 3BP5L=367. F) Quantification of single lumen cysts in MDCK GFP Rab11a WT, S25N and Q70L cell lines normalized to MDCK parental line. Data represents the mean ± SD of three independent experiments. * indicates p<0.05, ** indicates p<0.01, ns indicates not significant. MDCK n=346, WT n=342, S25N n=317, Q70L n=391.

Figure 3.11: SH3BP5 and SH3BP5L knockdown leads to minimal impairment of lumen formation. A) Western blot analysis of 3BP5 knockdown by shRNA in MDCK cells. α-Tubulin serves as a loading control. B) Representative confocal images of 3BP5 knockdown 4-day cysts stained with PODXL and β-catenin antibodies. C) Proportion of single lumen cysts in 3BP5 knockdown cells normalized to shSCM control. Data represents the mean ± SD of three independent experiments. * indicates p<0.05. shSCM n=453, sh25 n=416, sh3 n=300. D) Western blot analysis of 3BP5L knockdown by shRNA in MDCK cells. α-Tubulin serves as a loading control. E) Representative confocal images of 3BP5L knockdown 4-day cysts stained with PODXL and β-catenin antibodies. F) Proportion of single lumen cysts in 3BP5L knockdown cells normalized to shSCM control. Data represents the mean ± SD of three independent experiments. ** indicates p<0.01, ns indicates not significant. shSCM n=453, sh2 n=305, sh4 n=346. G) Western blot analysis of double knockdown 3BP5 and 3BP5L by shRNA in MDCK cells. α-Tubulin serves as a loading control. H) Proportion of single lumen cysts in 3BP5/3BP5L double knockdown cells normalized to shSCM control. Data represents the mean ± SD of three independent experiments. ns indicates not significant. shSCM n=385, sh25 +sh22 n=364. Scale bars = 10µm, 63x magnification.

Figure 3.12: SH3BP5 and SH3BP5L are required during lumen formation. A) Western blot analysis of MDCK CRISPR generated knockout single and double knockout cell lines for 3BP5 and 3BP5L. α-Tubulin serves as a loading control.B) Rab11a-GTP pulldown analysis by immunoblotting. Whole cell lysates were analyzed for SH3BP5, SH3BP5L and Rab11a expression. C) Quantification of Rab11a-GTP levels normalized total Rab11a protein for each condition and then normalized to pLCKO control cells. Data represents mean ± SD of three independent experiments. D) Representative confocal image of 3BP5/3BP5L double knockout 4-day cysts that were incubated with PODXL and β-catenin antibodies. Scale bar = 10µm, 63x magnification. E) Quantification of cysts with single lumens between 3BP5/3BP5L double knockout and single knockout cells. Data represents the mean ± SD of three independent experiments. * indicates p<0.05, ** indicates p<0.01, ns indicates not significant. pLCKO n=359, cl14 n=356, cl26 n=314, 3BP5 KO n=315, 3BP5L KO n=314. 87

3.4.5 Tankyrase inhibits lumen formation through negative regulation of SH3BP5 and SH3BP5L

We established SH3BP5 and SH3BP5L as GEFs required for Rab11a activation during lumenogenesis and characterized them as Tankyrase and RNF146 substrates. We hypothesized that Tankyrase and RNF146 are upstream regulators of Rab11a activation. However, treatment of MDCK cells with the inhibitor TNKS656 did not alter the protein abundance of either SH3BP5 or SH3BP5L (Figure 3.13A). However, TNKS656 treated cells displayed increased Rab11a-GTP levels and induced increased frequency of single lumen cysts comparable with elevated levels of activated Rab11a and similar to cysts observed with the expression of constitutively active mutant Rab11a Q70L (Figure 3.13B-E, 3.9F). This observation suggested that SH3BP5/L PARylation impaired their GEF activity. To assess whether this increase was a direct effect of Tankyrase inhibition on SH3BP5/L we examined the effect of TNKS656 on SH3BP5/L single or double knockout MDCK cells. The inhibitor had a modest effect on lumen formation rates in SH3BP5/L double knockout cysts suggesting that Tankyrase may regulate other protein targets and pathways involved in epithelial polarity that are required for lumen formation (Figure 3.13F). TNKS656 treatment of Rab11a-depleted cells partially rescued the lumen formation defect, emphasizing that Tankyrase may have multiple targets that control cell polarity (Figure 3.13G).

We observed that SH3BP5 protein levels were elevated in TNKS/TNKS2 double knockout MDCK cells emphasizing that SH3BP5 is a bone fide substrate of Tankyrase. In distinction, genetic ablation of TNKS/TNKS2 had little effect on SH3BP5L protein expression showing a regulatory hierarchy by Tankyrases in controlling SH3BP5/L expression (Figure 3.14A). Double TNKS/TNKS2 knockout cells had increased Rab11a-GTP activity, which phenocopied what was observed with the TNKS656 inhibitor (Figure 3.14A, B). The increase in Rab11a-GTP levels observed in the TNKS/TNKS2 knockout cells led to increased single lumen formation (Figure 3.14C, D). These data demonstrate that Tankyrase negatively regulates lumenogenesis through PARylation of SH3BP5/L with its attendant downstream effect on Rab11 in addition to Tankyrases’ control of other proteins that operate independent of Rab11a during lumen formation.

Figure 3.13: Tankyrase inhibits lumen formation through SH3BP5/L and other additional factors A) Rab11a-GTP loading analysis of MDCK cells treated with 1µM TNKS656 for 48 hours. Lysates were analyzed by western blot for indicated proteins. B) Quantification of Rab11a-GTP levels normalized to DMSO treated cells. Data represents mean ± SD of three independent experiments. C) Representative confocal images of MDCK 4-day cysts treated with 1µM TNKS656 that were stained with PODXL and β-catenin antibodies. Scale bar = 10µm, 63x magnification. D) Proportion of cysts with single lumens normalized to DMSO control. Data represents the mean ± SD of three independent experiments. * indicates p<0.05. DMSO n=355, TNKS656 n=403. E) Proportion of cysts with single lumens of 3BP5/3BP5L knockout lines without and with four days treatment of 1µM TNKS656. Data represents the mean ± SD of three independent experiments. * indicates p<0.05, ** indicates p<0.01, ns indicates not significant. pLCKO n=359, dKO cl14 n=356, dKO cl26 n=314, 3BP5 KO n=315, 3BP5L KO n=314, pLCKO+TNKS656 n=424, dKO cl14+TNKS656 n=309, dKO cl26+TNKS656 n=333, 3BP5 KO+TNKS656 n=310, 3BP5L KO+TNKS656 n=333. F) Proportion of single lumen cysts in Rab11a knockdown cells without and with 1µM TNKS656 normalized to shSCM control. Cysts were treated for 4 days with inhibitor. Data represents the mean ± SD of three independent experiments. ** indicates p<0.01, *** indicates p<0.001, **** indicates p<0.0001, ns indicates not significant. shSCM n=340, sh44 n=300, sh91 n=316, SCM+TNKS656 n=323, sh44+TNKS656 n=300, sh91+TNKS656 n=327.

Figure 3.14: Tankyrase inhibits lumen formation and regulates SH3BP5 stability. A) Rab11a-GTP loading analysis of TNKS/TNKS2 double knockout MDCK cells. Whole cell lysates were analyzed by immunoblotting for the indicated antibodies. B) Quantification of Rab11a-GTP levels normalized to total Rab11a protein and then normalized to pLCKO control cells. Data represents mean ± SD of three independent experiments. C) Representative confocal images of TNKS/TNKS2 double knockout 4-day cysts that were incubated with PODXL and β-catenin antibodies. Scale bar = 10µm, 63x magnification. D) Proportion of cysts with single lumens normalized to pLCKO control. Data represents the mean ± SD of three independent experiments. * indicates p<0.05. pLCKO n=308, dKO n=336.

3.4.6 RNF146 regulates lumen formation by degrading Tankyrase

RNF146 functions as a reader of PARylated marks on Tankyrase substrates to trigger ubiquitin mediated proteolysis for pathways affecting WNT signaling, bone metabolism, Hippo signaling and polarity (Zhang et al., 2011; Levaot et al., 2011; Li et al., 2014; Wang et al., 2015; Campbell et al., 2016; Matsumoto et al., 2017). Since Tankyrases differentially regulate SH3BP5 and SH3BP5L stability in MDCK cells, this suggested that RNF146 may have specific targets in epithelial cells that are regulated independently of Tankyrase. To characterize the function of RNF146 in these cells during lumen formation, RNF146 localization was examined first using MDCK cells stably expressing Tag-RFP-T-RNF146. RNF146 had a predominantly cytoplasmic localization in 2D culture (Figure 3.15A). Expression of RNF146 resulted in reduced Tankyrase protein levels consistent with a model that Tankyrase is itself an RNF146 substrate (Figure 3.15B). Importantly, ectopic expression of RNF146 had no effect on SH3BP5 or SH3BP5L protein levels (Figure 3.15B). However, Rab11a-GTP levels were nevertheless increased and

cysts had higher single lumen rates in RNF146 expressing cells than control cells (Figure 3.15C, E). RNF146 localization in cysts had uniform apical and basolateral localization (Figure 3.15D).

Figure 3.15: RNF146 controls Tankyrase stability during lumen formation. A) 2D confocal images representative of MDCK cells stably expressing Tag-RFP-T-RNF146. Scale bar =10µM and 63x magnification. B) Rab11a-GTP loading analysis of RNF146 expressing MDCK cells. Lysates were analyzed by western blot for indicated proteins. C) Quantification of Rab11a-GTP levels normalized to pMX control cells. Data represents mean ± SD of three independent experiments. D) Representative confocal images of 4-day cysts from RNF146 expressing cells that were stained with PODXL and β-catenin antibodies. Scale bar =10µM, 63x magnification. E) Quantification of single lumen cysts of MDCK cells expressing RNF146 normalized to pMX control cells. Data represents mean ±SD of three independent experiments. * indicates p<0.05. pMX n=351, RNF146 n=332.

RNF146 CRISPR mediated knockout MDCK cells stabilization of Tankyrase protein but no differences in SH3BP5 or SH3BP5L protein abundance (Figure 3.16A). These cells had decreased Rab11a-GTP levels and decreased capacity to form single lumen cysts (Figure 3.16B- D). The defect in lumen formation could be rescued with TNKS656 treatment suggesting Tankyrase abundance and PARylation are key negative regulators of this process (Figure 3.16E). These results show that Tankyrases are the predominate target of RNF146 in epithelial cells and regulates lumen formation by modulating Tankyrase stability. These findings also emphasize that there is cell-type specificity for substrate recognition during PARdU.

Figure 3.16: Tankyrase inhibition rescues RNF146 impairment of lumen formation. A) Rab11a-GTP loading analysis of RNF146 knockout MDCK cells. Lysates were analyzed by western blot for indicated proteins. B) Quantification of Rab11a-GTP levels normalized to pLCKO control cells. Data represents mean ± SD of three independent experiments. C) Representative confocal images of RNF146 knockout 4-day cysts that were incubated with PODXL and β-catenin antibodies. Scale bar = 10µm, 63x magnification. D) Proportion of RNF146 knockout cysts with single lumens normalized to pLCKO control. Data represents the mean ± SD of three independent experiments. ** indicates P<0.01. pLCKO n=383, KO sg2 n=312, KO sg3 n=334. E) Proportion of single lumens of RNF146 knockout 4-day cysts without and with 1µM TNKS656 treatment. Cysts were treated for 4 days with inhibitor. Data represents the mean ± SD of three independent experiments. * indicates p<0.05, ** indicates p<0.01, ns indicates not significant. pLCKO n=309, KO sg2 n=321, KO sg3 n=301, pLCKO+TNKS656 n=357, KO sg2+TNKS656 n=318, KO sg3+TNKS656 n=321.

3.5 Discussion

In this chapter, we have described a proteomic approach to identify new SH3BP5 interacting proteins. Since over 25% of the significant interactors were related to vesicle trafficking and Rab11a and the recent discovery that SH3BP5 and its paralog SH3BP5L are Rab11a GEFs, we focused on establishing a biological role of SH3BP5/L function in controlling Rab11a mediated cyst formation in epithelial cells (Sakaguchi et al., 2015; Jenkins et al., 2018). Of note we did not identify SH3BP5 interactors that support a role for its role in the mitochondria in mediating JNK signaling during ER stress (Wiltshire et al., 2002; Win et al., 2011; Win et al., 2014). The BioID assay allows for labeling of both physical interactors and proteins within a 10 nm range across all areas within a cell so mitochondria associated proteins should have been labeled in our proteomic screen (Roux et al., 2012; Kim et al., 2014). It is possible that a stress response is required to stimulate JNK/SH3BP5 pathway. Future experiments may probe the question of how the SH3PB5 interactome is contingent on responding to proteotoxic stress.

Through combinatorial deletions of different regions and α-helices of SH3BP5, a minimal catalytic core comprising helix 1 and 4 was identified. This finding has been corroborated with the recent publication of the crystal structure of Rab11a bound to full length SH3BP5 (Jenkins et al., 2018). The direct contact points between SH3BP5 and Rab11a identified within the structure are located on helix 1 and 4 and reside within the boundaries of the minimal catalytic domain. We found that expression of this minimal catalytic unit was sufficient for nucleotide exchange. We observed that helix 1 and 4 was a more efficient exchanger than the full length SH3BP5 protein suggesting that non-catalytic sequences may have negative regulatory function on enzyme activity.

In chapter 2, we showed that endogenous protein levels of SH3BP5 and SH3BP5L were regulated by RNF146 and Tankyrase in a cell-type specific manner. Here we show that Tankyrase has an inhibitory effect on lumen formation as cysts treated with the inhibitor TNKS656 had increased ability to form single lumens and deletion of both SH3BP5 and SH3BP5L could largely ablate this effect. This suggested that Tankyrase-mediated PARylation partially inhibits lumen formation. We anticipate that Tankyrase controls other polarity genes in addition to SH3BP5 and SH3BP5L that govern lumenogenesis (Mrozowska and Fukuda 2016).

The best characterized Tankyrase cell polarity substrates to date are the Angiomotins AMOT, AMOTL1, and AMOTL2. Striatin can be PARylated by Tankyrase (Wang et al., 2015; Campbell et al., 2016; Wang et al., 2016; Guetter et al., 2011; Lahav-Ariel et al., 2018). Stabilization of AMOTL2 by Tankyrase inhibition or RNF146 depletion results in mislocalization of Crumbs complex member PALS1 at the apical membrane and mislocalization of ZO-1 and PALS1 at tight junctions (Campbell et al., 2016). The Crumbs complex is important for single lumen formation as it is a critical component for maintaining apical membrane integrity during cytokinesis (Roh et al., 2003; Schlüter et al., 2009). Angiomotin stabilization would likely have a negative impact on lumen formation and is likely not contributing to the increased lumen phenotype we observed in Tankyrase inhibited conditions or Tankyrase knockout cells.

We hypothesized that RNF146 and Tankyrase worked in tandem and inhibition/depletion of either protein would stabilize SH3BP5/SH3BP5L and have similar phenotypes. However, the data reveal that RNF146 and Tankyrase have apparent opposing functions in MDCK cells. Tankyrase inhibition and deletion increases Rab11a-GTP levels in MDCK cells resulting in cysts with increased frequency of single lumen formation. Only Tankyrase deletion increased SH3BP5 protein levels as detected by western blot which suggests that the Tankyrase/3BP5 interaction is more critical for regulating protein stability. Depletion of RNF146 stabilizes Tankyrase protein levels and RNF146 overexpression decreases Tankyrase protein levels while SH3BP5/SH3BP5L levels were unaltered. These data show Tankyrase is the predominant RNF146 substrate in MDCK cells and RNF146 regulates Tankyrase protein expression. These data also suggest that PARylation of SH3BP5/SH3BP5L inhibits their GEF activities. The GEF exchange activities of PARylated SH3BP5/SH3BP5L will need to be examined directly. The principal enzyme substrate relationship between RNF146 and Tankyase in MDCK cells is distinct from the model where RNF146 targets Tankyrase mediated PARylation of third party substrates for PARdU as observed in mouse epithelial cells, breast cancer cells, colorectal cancer cells and macrophages/osteoclasts (Huang et al., 2009; Zhang et al., 2011; Levaot et al., 2011; Wang et al., 2015; Campbell et al., 2016; Matsumoto et al., 2017).

However, it is also possible that Tankyrase is regulating the localization of these GEFs as shown in Figure 3. Tankyrases are known to form oligomeric structures through its SAM domain and 95

bound substrates can also become part of this large protein complex (De Rycker et al., 2003). Tankyrase oligomers result in large vesicular aggregates in cells and contribute to PARylation activity (DaRosa et al., 2016; Mariotti et al., 2016; Riccio et al., 2016). There is a possibility that SH3BP5/SH3BP5L bound to Tankyrase form these structures and sequestration impacts Rab11a activation. A new study revealed a role for Tankyrase and PAR in the assembly of TDP-43 phase separation foci during stress. Tankyrase regulates the cytoplasmic localization of TDP-43 while PAR mediates TDP-43 localization into stress granules (McGurk et al., 2018). A similar mechanism could be regulating SH3BP5/SH3BP5L localization that is PAR dependent. The ability of Tankyrase to affect SH3BP5/SH3BP5L localization in MDCK cells will need to be examined to conclude whether there is a PAR or SAM dependent function that principally controls downstream Rab11a activation. Collectively, these findings raise the importance of understanding PARdU substrate selectivity and specificity in different cell-types.

Rab11a initiates a signaling cascade to facilitate the apical transport of PODXL from the basement membrane to the apical membrane during lumenogenesis and primary ciliogenesis (Bryant et al., 2010; Knödler et al., 2010; Westlake et al., 2011). Rab11a-GTP binds the Rab8 GEF Rabin8 and stimulates its GEF activity towards Rab8. Even though active Rab11a has been established for facilitating Rab8 activation, the proteins required for Rab11a activation during these processes is unknown. Knockdown of GEFs for Rab8 or Cdc42 result in decreased lumen formation (Bryant et al., 2010). The data shown in this chapter provides evidence that SH3BP5 and SH3BP5L are GEFs that activate Rab11a during apical transport of PODXL. In addition, a role for PARdU in the regulation of Rab11a GTPase activity through the SH3BP5 family of GEFs has been described. The apical localization of SH3BP5 during lumenogenesis is a new finding and suggests that SH3BP5 is also transported to the apical surface to maintain Rab11a activation during this process. These GEFs display a redundant function since complete knockout of both GEFs leads to a greater phenotype than either alone.

In mammalian cells, only SH3BP5 and SH3BP5L have been characterized as Rab11 GEFs while the TRAPPII complex has been described in yeast and drosophila (Jones et al., 2000; Riedel et al., 2018; Jenkins et al., 2018). TRAPPC9 and TRAPPC10 were identified as unique components differentiating TRAPPI and TRAPPII complexes in fruit flies and deletion of the SH3BP5 paralog parcas together with TRAPPC9 proved to be lethal (Riedel et al., 2018). It is unclear 96

whether the mammalian TRAPPC9 functions as a Rab11 GEF and whether it has a role during lumen formation.

Several studies have shown the efficacy of Tankyrase inhibitors as alternative or synergistic therapies in colon, breast, non-small cell lung and prostate cancers (Huang et al., 2009; Bao et al., 2012; Wang et al., 2016; Cheng et al., 2019). PODXL has been examined as a prognostic marker in a myriad of cancers including breast, prostate, pancreatic and hepatocellular cancers (Neilsen and McNagny 2009). PODXL overexpression is correlated with invasive breast cancer and poor outcome and enables cell migration leading to tumour metastasis (Somasiri et al., 2004; Snyder et al., 2015). A recent study demonstrated PODXL is upregulated during epithelial- mesenchymal transition (EMT) and promotes extravasation of breast and pancreatic cancer cells through the endothelium to promote metastasis (Fröse et al., 2018). Tankyrase deficiency leads to increased apical PODXL transport during lumen formation. This raises concerns of potential untoward effects with the clinical application of Tankyrase inhibitors since PODXL is associated with metastatic behavior. Further long-term studies should examine whether Tankyrase inhibition can promote metastasis in epithelial cancers.

Chapter 4

Investigating the role of asparaginyl hydroxylase FIH towards Tankyrase activity and function 4.1 Abstract Factor Inhibiting HIF-1α originally identified as an enzyme, which hydroxylates HIF-1α is now recognized to have many substrates that typically contain ankyrin repeats. Tankyrase-2 was previously demonstrated to an FIH substrate though the significance of hydroxylation on regulating Tankyrase function was not characterized. The work presented in this chapter demonstrates that FIH binds to Tankyrase Ankyrin Repeat Clusters (ARCs) and FIH-mediated hydroxylation inhibits PARP activity. FIH expression decreased Tankyrase auto-PARylation and substrate PARylation. However, FIH did not affect Tankyrase-2 stability nor affect stability of Tankyrase substrates. This work demonstrates that FIH regulates Tankyrase PARP activity that affects PARylation mediated processes that are independent of protein degradation.

4.2 Introduction

The asparaginyl hydroxylase, factor inhibiting HIF-1α (FIH) was originally identified as a binding partner to HIF-1α and repressed its transcriptional activation (Mahon et al., 2001). FIH catalyzes asparaginyl hydroxylation of Asn-803 within the C-terminal transactivation domain of HIF-1α and blocks the interaction with the transcriptional coactivator p300 (Lando et al., 2002; Lando et al., 2002; Hewitson et al., 2002). HIF-1α prolyl and asparaginyl hydroxylation is blocked under hypoxia resulting in increased stability and transcriptional activity (Lando et al., 2002). FIH hydroxylates many substrates in addition to HIFα. Most FIH substrates and interactors contain ankyrin repeat domains (ARD). Proteomic approaches have identified new FIH binding proteins including Notch, IκB, ASB4, Rabankyrin-5, RNase L, AnkyrinR, TRPV3, ASPP2 and Tankyrase1/2 (Cockman et al., 2006; Coleman et al., 2007; Ferguson, J.E. et al., 2007; Cockman et al., 2009; Yang et al., 2011; Janke et al., 2013; Kiriakidis et al., 2015; Karttunen et al., 2015). The deubiquitinase OTUB1 has also been shown to be hydroxylated but it does not contain ankyrin repeat (Scholz et al., 2016). Additionally, FIH has been shown to facilitate histidinyl hydroxylation on Tankyrase-2 and aspartate hydroxylation on the cytoskeleton ankyrin proteins, ankyrinR and ankyrinB (Yang et al., 2011; Yang et al., 2011). The ankyrin repeat containing kinase LRRK1 binds directly to FIH but it is unknown whether it is hydroxylated (Peng et al., 2014). The current paradigm suggests that FIH mediated hydroxylation of proteins has an inhibitory effect on enzymatic activity or interferes with the scaffold function of adaptor proteins.

The dynamics of Tankyrase activation, substrate recognition and regulation remain to be elucidated. Tankyrases are known to bind to some proteins but not PARylate them. Some Tankyrase substrates are PARylated but not ubiquitylated by RNF146, suggesting that are determinants of targeting and degradation that remain to be understood. Zhong et al showed that global cellular NAD+ levels regulates the PARylation capacity of Tankyrase (Zhong et al., 2015). Phosphorylation of Tankyrase-1 by MAPK following insulin signaling increases its catalytic activity (Chi and Lodish 2000). Tankyrase-1 is phosphorylated during mitosis by GSK3 but the functional consequences of this modification remain unclear (Yeh et al., 2006). Previous work examining the effect of hydroxylation on ankyrin repeats suggested hydroxylation was important for stabilizing the ankyrin fold containing proteins and thus critical for mediating 99

protein-protein interactions (Hardy et al., 2009). Since Tankyrases are heavily hydroxylated by FIH, it raised the possibility that this modification could be a critical factor in mediating substrate recognition by the ARCs and further suggests that Tankyrase could be regulated by oxic states. In addition, hydroxylation could be a mechanism of regulating PARdU since Tankyrase substrate binding often leads to PARylation which triggers binding and activation of the E3 ligase RNF146 (DaRosa et al., 2014).

In this chapter, another example of FIH expression regulates Tankyrase PARP activity and ability to interact with its substrate 3BP2. FIH did not affect Tankyrase stability or the steady state protein levels of its substrates 3BP2 and Axin1. Future study will be aimed at determining the biological context in which endogenous FIH regulates Tankyrase activity.

100

4.3 Experimental Procedures 4.3.1 Cell lines

HEK293T, RAW264.7, and FIH WT and KO mouse embryonic fibroblasts (MEFs) were cultured in DMEM (WISENT) with 10% FBS (WISENT) and supplemented with 100μg/ml streptomycin and 100U/ml penicillin (WISENT). Primary bone marrow derived macrophages (BMMs) were culture in α-MEM (WISENT) with 10% FBS and 2% CMG-14 supernatant. Cells were cultured at 37⁰C in a humidified incubator set at 5% CO2. FIH MEFs were a kind from from Dr. Randall Johnson (University of Cambridge, England)

4.3.2 Expression plasmids

Human TNKS2 ARC1, ARC2, ARC3, ARC4, ARC5, ARC4 L560W, and TNKS2ΔARC were cloned into pEF6 MycHis B (ThermoFisher Scientific) using BamHI and XbaI restriction sites. pEF6 TNKS2ΔSAM, TNKS2ΔPARP, GST-3BP2 were previously described in Levaotet al. 2011. pCDNA3 SPA-3xFlag-FIH was a kind gift from Dr. Peter Ratcliffe (University of Oxford, England) and pEF6 HA-FIH, D201A and L340R were gifts from Dr. Matthew Cockman (University of Oxford, England).

Table 4.1: List of constructs and cloning primers Plasmid Name Primer Sequence Primer Name Target pEF6 TNKS2ΔARC ttaggatccatgtctgctctgccctctt huTNKS2 noARCs Bam F TNKS2 ARCs 1-5 ttatctagacttccatcgaccatacct huTNKS2 noARCs Xba R (801-1167) pEF6 TNKS2 ARC1 ttaggatccatggccgtggagccggccg huTNKS2 ARC1 Bam F TNKS2 ARC1 ttatctagactcttatattcaccagta huTNKS2 ARC1 Xba R (20-178) pEF6 TNKS2 ARC2 ttaggatccatggtgctgttacagcatg huTNKS2 ARC2 Bam for TNKS2 ARC2 ttatctagactgcctttaaattcatat huTNKS2 ARC2 Xba Rev (173-332) pEF6 TNKS2 ARC3 ttaggatccatggcatatgaatttaaag huTNKS2 ARC3 Bam F TNKS2 ARC3 ttatctagacttaatgagataccctct huTNKS2 ARC3 Xba R (327-487) pEF6 TNKS2 ARC4 ttaggatccatgggtaattcagaggcag huTNKS2 ARC4 Bam for TNKS2 ARC4 ttatctagactcaaagctgcatctcc huTNKS2 ARC4 Xba Rev (488-649) pEF6 TNKS2 ARC5 ttaggatccatggatctgcttaggggagat huTNKS2 ARC5 Bam F TNKS2 ARC5 ttatctagacttgggggcatggctgctgt huTNKS2 ARC5 Xba R (641-800) pEF6 TNKS2 ARC4 ttaggatccatgggtaattcagaggcag huTNKS2 ARC4 Bam for TNKS2 ARC4 L560W ttatctagactcaaagctgcatctcc huTNKS2 ARC4 Xba Rev L560W as template

101

4.3.3 Immunoblotting and immunoprecipitation

For immunoblotting, RAW264.7 cells, MEFs, and BMMs were lysed in RIPA buffer (1% Triton X-100, 50mM HEPES pH7.5, 150mM NaCl, 5mM EDTA, 0.1% SDS, 0.5% sodium deoxycholate) plus 1x Halt protease and phosphatase cocktails. Lysates were cleared of cell debris by centrifugation at 4⁰C at 13,3000 rpm for 10 minutes. Cleared lysate was quantified using the BCA protein assay (ThermoFisher Scientific). 20-25μg of lysate was boiled with 1x Laemmli buffer for 5 minutes at 100⁰C before SDS-PAGE and western blot analysis.

For immunoprecipitation experiments, 8x105 HEK293T cells in 6-well plates were transfected for 24 hours using LipoD293 (Froggabio). Cells were lysed directly on ice for 10 minutes using 350μl of lysis buffer containing 1% Triton X-100, 25mM Tris pH7.5, 100mM NaCl, 1mM EDTA and 1x Halt protease and phosphatase cocktails (Thermo Scientific). Lysates were cleared by centrifugation at 13,300 rpm for 10 minutes at 4⁰C and 300μl was incubated with either EZviewTM Red HA-agarose or Flag M2 agarose (Sigma) for one hour. Beads were washed three times with lysis buffer (without inhibitor cocktails), boiled with 2x Laemmli buffer for 5 minutes at 100⁰C. For immunoblotting, samples were separated by SDS-PAGE, transferred to PVDF membrane (Millipore), blocked with 5% skim milk (BioShop) in 1x PBS + 0.1% Tween 20 (BioShop) and incubated with indicated primary antibodies. Primary antibody was washed off with three washes of 1x PBS + 0.01% Tween 20 and membranes were incubated with horseradish peroxidase conjugated mouse and rabbit antibody (BioShop) for one hour. Membranes were washed three times with 1x PBS + 0.1% Tween 20 and developed using ECL or ECL prime (GE Healthcare), SuperSignal West Pico PLUS and SuperSignal West Femto (ThermoFisher Scientific) HRP substrates. Membranes were imaged using a MicroChemi 2.0 chemiluminescent imager (DNR Bio-Imaging Systems).

4.3.4 In vitro PARylation assay

1.2x106 HEK293T were transfected in 6-well plates for 24 hours with LipoD293 (FroggaBio). For experiments using DMOG (Cayman Chemical), culture media was replaced 6 hours after transfection and cells were treated with DMOG for 18 hours. Tankyrase inhibitor, TNKS656, was added onto cells 6 hours post-transfection and treated for 18hrs where indicated. Cells were lysed in PAR lysis buffer (50mM HEPES pH7.5, 150mM NaCl, 1% Triton X-100, 0.1% SDS,

102

0.5% sodium deoxycholate, 5mM EDTA, 1x Halt protease/phosphatase cocktails, 5mM DTT) and cleared lysates were immunoprecipitated for one hour with Myc agarose (Sigma) for Tankyrase auto-PARyation and glutathione agarose for 3BP2 PARylation. The agarose beads were washed three times with lysis buffer then washed three times with in vitro assay buffer (50mM HEPES pH7.5, 100mM NaCl, 0.01% Triton X-100, 2mM TCEP). Agarose beads were incubated in 50µl of assay buffer with 10µM biotin-NAD+ and shaken at 25⁰C for one hour. The reaction was stopped with 1x Laemmli buffer, boiled for 5 minutes at 100⁰C and subjected to SDS-PAGE and immunoblot analysis. Blocked membranes were incubated with streptavidin HRP for one hour, washed three times with 1x PBS + 0.1% Tween 20 and biotin signal was detected with ECL prime or SuperSignal West Femto to detect the biotin signal.

4.3.5 Antibodies and reagents

For western blotting the antibodies used were: Myc (9E10, 1:250; Santa Cruz sc40), Flag (clone M2, 1:2000; Sigma), HA (clone 3F10, 1:500; Roche), α-Tubulin (1:1000; Santa Cruz sc69969), HIF1α (1:500; Novus Biologicals NB100-479), FIH (1:500; Novus Biologics NB100- 428),Streptavidin HRP (1:2000; Cell Signaling Technologies #3999), 3BP2 (clone 1E9, 1:1000; Abnova H00006452-M01), Axin1 (1:1000; Cell Signaling Technologies #2087), Tankyrase (1:500; Abcam ab13587).

TNKS656 inhibitor was a gift from Dr. Feng Cong (Novartis), DMOG was purchased from Cayman Chemical (cat#71210) and Biotin-NAD+ was purchased from Trevigen (cat# 4670-500- 01).

4.3.6 Retroviral constructs

FIH wildtype, D201A and L340R cDNAs were subcloned into pMXs using BamHI and EcoRI restrictions sites using the following cloning primers: (Hu_FIH BamHI For – TTAGGATCCACCATGGCGGCGACAG) and (Hu_FIH EcoRI Rev – TTAGAATTCCTAGTTGTATCGGCCCT).

4.3.7 Retrovirus and lentivirus production and transduction

Short hairpin RNA’s targeting murine FIH were obtained from the TRC library (Dr. Jason Moffat). 103

Table 4.2: shRNA sequences shRNA Sequence TRC Clone ID Plasmid RefSeq ID shGFP GCAAGCTGACCCTGAAGTTCAT N.A. pLKO.1 puro Mouse FIH shA1 CAACGGAGATTTCTCTGTGTA TRCN0000173449 pLKO.1 puro NM_176958 Mouse FIH shA2 CCTGCAAGAGAATATTGGCAA TRCN0000173743 pLKO.1 puro NM_176958 Mouse FIH shA4 CCAGCACCCATAAGTTCTTAT TRCN0000194381 pLKO.1 puro NM_176958

Lentiviral particles were generated by transfecting 5x105 HEK293T cells in 6-well plates with 50ng envelope (VSV-G), 450ng packaging (psPAX2) and 500ng mouse FIH shRNA (pLKO.1) plasmids using X-tremegene 9 (Roche). Retroviral particles were produced by transfection of 2x106 HEK293T cells in 6cm dishes with 1μg pMXs-HA-FIH constructs and 1μg Ecopak using the CalPhos transfection kit (Clontech). After 5 hours of transfection, the media was replaced with target cell media and once more the following morning. Viral supernatants were collected 48 hours post-transfection and cleared through 0.45μm filters. Target cells were transduced with viral supernatants along with 8µg/ml polybrene for 24 hours. Antibiotic selection was started 24 hours post-infection for a further 48 hours. For selection, 10µg/ml puromycin was used for RAW264.7 cells, and 2μg/ml puromycin for NIH3T3 cells.

4.3.8 Isolation of primary bone marrow macrophages

FIH knockout mice were a kind gift from Dr. Peter Ratcliffe. Bone marrow from 6-8-week-old FIH wildtype and knockout mice was flushed from the tibia and femur with α-MEM and 10% FBS. Cells from the bone marrow were pelleted at 1000 rpm for 5 minutes and resuspended in α- MEM with 10% FBS, 2% CMG-14-12 conditioned supernatant, 100μg/ml streptomycin and 100U/ml penicillin and passed through a 0.7-micron cell-strainer. The suspension was plated onto petri dishes and cells were cultured for 4 days before harvesting for immunoblot analysis.

104

4.4 Results

4.4.1 FIH hydroxylates Tankyrase-2 within Ankyrin Repeat Cluster (ARC) domains

FIH had been previously shown to hydroxylate Tankyrase-2 (TNKS2) at ten sites. Eight of the sites occurred on asparagine residues while two modifications occurred on histidine residues (Cockman et al., 2009; Yang et al., 2011). All ten sites were mapped on our crystal structure of the Tankyrase ARC domain (Guettler et al., 2011) to occur within the ankyrin repeat clusters of TNKS2 (Figure 4.1A). Interestingly, the modifications target at least one ankyrin repeat in each ARC except for ARC1. Molecular modelling of the hydroxylation sites onto ribbon structures of each ARC reveal that the modifications are on the opposite face of substrate binding (Figure 4.1B) (Guettler et al., 2011). This indicates that FIH may stabilize the substrate binding pocket in each ARC and facilitating substrate binding.

4.4.2 FIH interacts with TNKS2 Ankyrin Repeat Clusters

FIH was previously shown to bind directly to TNKS2 but the specific regions of interaction were not determined (Cockman et al., 2009). To determine where FIH binds to TNKS2, constructs deleting the entire ARC region, SAM domain and PARP domain as well as each individual ARC were used in immunoprecipitation studies. HEK293T cells were transfected with equal amounts of HA-FIH and indicated TNKS2 plasmids for 24 hours and lysates were immunoprecipitated with anti-HA-agarose beads. Each immune complex was subjected to western blotting and probed for Myc-antibody to identify which regions of TNKS2 were required for FIH interaction. These studies revealed that FIH bound to ARCs 1, 2, 4, and 5 directly which exhibited the same binding behavior as other known substrates and binding proteins (Figure 4.2A). To determine whether FIH could bind within the substrate recognition pocket or to a different face of the ARC, FIH binding was compared between wildtype and a binding pocket mutant ARC4-L560W, a mutant unable to bind to the RXPPDG hexapeptide found in Tankyase substrates (Figure 4.2B). Although FIH does not have a canonical Tankyrase hexapeptide binding motif, binding to ARC4-L560W was reduced suggesting that FIH binding might compete with other substrates for the same binding pocket despite not having a consensus motif.

105

106

Figure 4.1: FIH hydroxylates TNKS2 within its Ankyrin Repeat Cluster domains. A) Sequence alignment of human Tankyrase 2 ARCs 1-5. Blue shaded residues indicate level of conservation between each cluster while red boxes indicated hydroxylated residues by FIH. B) Ribbon models of ARC2-3, ARC4 and ARC5 with each hydroxylated amino acid labelled.

To assess if FIH underwent PARylation, an in vitro PARylation experiment was performed. Lysates from HEK293T cells expressing Flag-FIH with and without Myc-TNKS2 were immunoprecipitated with Flag agarose beads in the presence of biotin-NAD. Flag-FIH protein was then immunoblotted with Streptavidin HRP to detect poly-ADP-ribose. In distinction to 3BP2 and Tankyrase, there was no evidence that FIH became PARylated in the presence of Tankyrase (Figure 4.2C).

4.4.3 FIH inhibits TNKS2 activity

Since FIH is a binder but not a TNKS2 substrate, the function of the interaction was interrogated. TNKS2 can PARylate itself and was used as a measure of catalytic activity (Cook et al., 2002). Myc-TNKS2 immune complexes from HEK293T cells were incubated with biotin-NAD+ to examine auto-PARylation. FIH was overexpressed in the presence of TNKS2 and this led to a large decrease in TNKS2 auto-PARylation. We used the 2-oxoglutarate dependent dioxygenase inhibitor DMOG as a potent inhibitor of FIH to address if FIH mediated hydroxylation of Tankyrase regulated the catalytic activity of Tankyrase. The decreased catalytic activity of Tankyrase observed in FIH overexpressing cells was reversed by increasing concentrations of DMOG, suggesting FIH mediated hydroxylation inhibited TNKS2 PARP activity (Figure 4.3A). To determine the specificity of FIH inhibition on TNKS2 we examined the effect of FIH expression on PARP1 catalytic activity. FIH coexpression with PARP1 had no effect PARP1 auto-PARylation nor did expression of the catalytically inactive FIH mutants FIH-D201A (Figure 4.3B). Flag-FIH has additional amino acid sequences adjacent to Flag resulting in its slower migration in comparison to Ha-FIH. The inhibition of TNKS2 activity was also examined using the FIH catalytic mutant D201A. The D201A mutant had no effect on TNKS PARP activity and similarly had reduced binding affinity to Tankyrase compared to wildtype FIH (Figure 4.3C). Collectively, these results demonstrate that FIH catalytic activity has an inhibitory effect on TNKS2 PARP activity.

107

108

Figure 4.2: FIH interacts with TNKS2 Ankyrin Repeat Clusters. A) HEK293T cells were co-transfected with HA-FIH and full length TNKS2 and deletion mutants. Lysates were immunoprecipitated with HA affinity gel and immune complexes were subjected to western blot analysis for HA and Myc. B) HEK293T cells were co-transfected with HA-FIH and TNKS2 ARC4 and pocket mutant L560W. Lysates were immunoprecipitated with Flag agarose beads and immune complexes were analyzed by western blot for Myc and Flag. C) HEK293T cells co-transfected with Myc-TNKS2 and Flag-FIH or Flag-3BP2 (as a positive control) and lysates were immunoprecipitated with Flag agarose beads. Immune complexes were washed and subjected to the in vitro PARylation assay. Reaction mixtures were analyzed by immunoblotting for Streptavidin HRP, Flag, and Myc and whole cell lysates were analyzed for Flag, Myc and α-Tubulin (loading control).

109

Figure 4.3 TNKS2 hydroxylation inhibits its catalytic activity. A) HEK293T cells co- transfected with Flag-FIH and Myc-TNKS2 for 24 hours were treated with increasing doses of DMOG for an additional 24 hours prior to lysis. Immune complexes were examined for TNKS2 PARylation while whole cell lysates were analyzed for Myc, Flag, HIF1α as a marker of DMOG treatment and α-Tubulin (loading control). B) HEK293T cells were co-transfected with Myc-PARP1, Flag-FIH, HA-FIH and mutant. PARP1 immune complexes were examined for PARylation. α-Tubulin serves as a loading control. C) HEK293T cells co- transfected with Myc-TNKS2, Ha-FIH and mutant were examined for TNKS2 PARylation. α- Tubulin serves as a loading control.

4.4.4 FIH impairs TNKS2-mediated 3BP2 PARylation

Since FIH impacted TNKS2 autoPARylation, the ability of TNKS2 to modify substrates was examined. The in vitro PARylation assay was used to assess the ability of TNKS2 to modify exogenous substrates in the presence of FIH. We observed that expression of FIH impaired the TNKS2 mediated PARylation of GST-3BP2 in HEK293T cells. Expression of FIH impaired the ability of TNKS2 to PARylate 3BP2 indicating TNKS2 is also unable to modify substrates (Figure 4.4A). In addition, FIH expression reduced the interaction between 3BP2 and TNKS2. DMOG treatment reversed the effect of FIH expression on TNKS2 mediated 3BP2 PARylation (Figure 4.4A). We next examined the effect of the FIH catalytic mutants on 3BP2 PARylation. FIH-D201A had no effect on 3BP2 PARylation and D201A did not affect the 3BP2/TNKS2 interaction (Figure 4.4B). These results demonstrate hydroxylation of the ARC domains by FIH impairs substrate binding and suggests that FIH competes with substrate binding to TNKS2.

110

Figure 4.4 FIH inhibits 3BP2 PARylation by Tankyrase. A) HEK293T cells were co- transfected with GST-3BP2, Flag-FIH and Myc-TNKS2 for 24 hours and treated with DMSO or 1mM DMOG for a further 24 hours as indicated. Lysates were immunoprecipitated with glutathione agarose beads and immune complexes were examined for 3BP2 PARylation, GST, Flag and Myc expression by immunoblot analysis. Whole cell lysates were analyzed for GST, Flag, Myc, HIF1α as a positive marker for DMOG treatment, and α-Tubulin (loading control). B) HEK293T cells were co-transfected with GST-3BP2, Myc-TNKS2, HA-FIH and mutant for 24 hours prior to lysis. Lysates were immunoprecipitated with glutathione agarose beads and immune complexes were examined for 3BP2 PARylation. α-Tubulin serves as a loading control.

4.4.5 Depletion of FIH does not affect protein stability of Tankyrase and its substrates

Inhibition of Tankyrase PARP activity usually results in stabilization of its protein levels (Bisht et al., 2012). Since expression of FIH inhibited the TNKS2 PARP catalytic activity, this suggested FIH could also influence Tankyrase stability in cells. Inhibition of Tankyrase activity is known to increase its own protein stability as well as its substrates (Huang et al., 2009; Levaot et al., 2011; Wang et al., 2015). Since our lab established a critical role for Tankyrases in regulating 3BP2 levels in macrophages and osteoclasts, this was an ideal system to determine if

111

FIH regulates Tankyrase function. In addition, Tankyrase regulates SRC activation in these cells and dysregulation of Tankyrase results in increased SRC kinase activity by modulation of 3BP2 protein stability (Levaot et al., 2011). To determine whether FIH affected Tankyrase function and subsequent stability of its substrates, FIH gene expression was inhibited using short-hairpin RNAs (shRNA), in a mouse macrophage cell-line RAW264.7, to examine levels of 3BP2 and active SRC. Knockdown of FIH by three different shRNAs showed no appreciable difference in 3BP2 stability and SRC activity (measured by phosphorylation of tyrosine 416) in comparison to shGFP control (Figure 4.5A). To account for residual FIH expression after gene knockdown, primary bone marrow macrophages from FIH knockout mice were examined for 3BP2 levels and SRC activation. Immunoblot analysis of lysates showed no differences in both 3BP2 protein level or SRC activation between knockout and wildtype macrophages (Figure 4.5B). To determine if FIH has a cell-type effect, FIH knockout embryonic fibroblasts were also examined for differences in Tankyrase stability as well as its substrates 3BP2 and AXIN1. Immunoblot analysis of cell lysates showed no differences in protein stability between knockout and wildtype fibroblasts (Figure 4.5C).

Figure 4.5 FIH knockdown and depletion do not affect the stability of Tankyrase and its substrates. A) FIH was depleted by transducing RAW264.7 cells with FIH specific shRNA. Lysates were obtained 72 hours after transduction and subjected to western blot analysis for 3BP2, SRC phosphorylation (Tyrosine 416), FIH and α-Tubulin (loading control). B) Western blot analysis of bone marrow derived macrophages (BMM) from FIH wildtype and knockout mice for indicated proteins. α-Tubulin serves as a loading control. C) Western blot analysis of primary mouse embryonic fibroblasts derived from FIH wildtype and knockout mice for indicated proteins. α-Tubulin serves as a loading control.

112

4.4.6 FIH expression has no effect on protein stability of Tankyrase substrates

Biochemically, FIH expression decreased TNKS2 activity, so this system was used to examine substrate stability. FIH and its mutants D201A and L340R were overexpressed in HEK293T cells and the levels of 3BP2 and Axin1 were examined by immunoblot analysis. Comparison of wildtype FIH expressing cells to cells expressing empty vector showed no differences in Axin1 and 3BP2 levels. Similarly, FIH D201A and L340R expressing cells showed no differences in protein abundance to the vector control or to wildtype FIH (Figure 4.6A). These constructs were expressed in NIH3T3 fibroblasts by retroviral infection and similarly no differences in 3BP2 and Axin1 protein abundance was observed (Figure 4.6B). These results demonstrate that FIH has no impact on endogenous protein stability of Tankyrase and its substrates. This suggests the biochemical downregulation of TNKS2 activity is likely uncoupled from protein degradation.

Figure 4.6: FIH expression has no effect on protein stability of Tankyrase substrates. A) HEK293T cells were transfected with HA-FIH and mutants D201A and L340R. Lysates were collected 24 hours post-transfection and analyzed by western blot for indicated proteins. GAPDH serves as a loading control. B) NIH3T3 cells were transduced with retroviral particles containing wildtype FIH and mutants. Lysates were collected 72 hours after transduction and 48 hours after puromycin selection and analyzed by western blot for indicated proteins. GAPDH serves as a loading control.

113

4.5 Discussion

The work presented in this chapter is our attempt to find a physiologic function to the experimentally established FIH mediated hydroxylation of Tankyrase. Previous work has demonstrated that hydroxylation frequently increases the stability of ankyrin repeat containing proteins. Polyhydroxylation at multiple sites can potently increase the stability of the isolated ankyrin fold and lead to stabilization of the holo-protein (Hardy et al., 2009). Published accounts that have mapped multiple hydroxylation sites across ARCs 2, 3, 4, and 5 in TNKS2 suggested that hydroxylation may have a role in stabilizing Tankyrase or regulate substrate binding (Cockman et al., 2009; Yang et al., 2011). We substantiated that FIH and Tankyrase interact in HEK293T cells and that FIH binds to the Tankyrase ARC domains. We demonstrated using a Tankyrase loss of function mutation in the hexapeptide substrate binding groove that FIH may be binding to the substrate binding face of the ARC domain despite not having a consensus Tankyrase binding motif. These results suggest FIH may compete with Tankyrase substrates for ARC binding. In agreement with this hypothesis we found that FIH expression decreased the interaction between 3BP2 and TNKS2. Whereas FIH maps to the Tankyrase substrate binding groove, hydroxylation maps to the loops on the opposing face of the ARC domain. FIH is among a unique subset of Tankyrase interactors, including GDP-mannose-4,6-dehydratase (GMD) and Mcl1, which bind but are not PARylated (Bae et al., 2003; Bisht et al., 2012). GMD binding has also been shown to regulate Tankyrase PARP activity (Bisht et al., 2012).

FIH expression inhibited TNKS2 PARP activity, which led to decreased auto-PARylation and 3BP2 PARylation. However, the inhibition in PARP activity did not lead to increased stabilization of Tankyrase or 3BP2 protein levels. FIH expression in HEK293T and NIH3T3 cells also did not lead to stabilization of Axin1 and 3BP2. Similarly, FIH depletion by shRNA or FIH knockout MEFs and macrophages did not affect Tankyrase, 3BP2, or Axin1 stability. FIH deficient cells had no effect in activating SRC kinase signaling downstream of 3BP2. These findings are consistent with other studies of FIH substrates, namely IκB, LRRK1, ASPP2, Notch, OTUB1 and TRPV3, where FIH does not affect their protein stability (Cockman et al., 2006; Peng et al., 2014; Janke et al., 2013; Kiriakidis et al., 2015; Scholz et al., 2016; Karttunen et al., 2015). Thus far, FIH has been shown to inhibit the enzymatic activity of Notch2 and TRPV3 114

while affecting the downstream functions of LRRK1, OTUB1, ASPP2, and IκB proteins by disrupting protein-protein interactions. FIH interferes with protein-protein interactions by both physical interaction and by hydroxylation. In the case of LRRK1, FIH binding blocks the LRRK1/EGFR interaction and promotes EGFR signaling in keratinocytes. Similarly, we observed that FIH dependent hydroxylation led to decreased TNKS2 3BP2 interaction.

Tankyrase mediated PARylation does not always lead to proteasomal degradation of its substrates including IRAP, NuMa and Miki but can also act as a scaffold in the case of bridging PEX14 and ATG9 to promote pexophagy (Chi and Lodish 2000; Chang et al., 2005; Ozaki et al., 2012; Li et al., 2017). Collectively, the data presented in this chapter suggests FIH mediated TNKS PARP inhibition affects PARdU independent functions. Further investigation on the effect of FIH on other Tankyrase substrates will be required. FIH could play a role in segregating Tankyrase between PARdU and other PARylation-dependent functions.

DMOG mimics hypoxic conditions by inhibiting FIH function and rescues Tankyrase PARP activity upon FIH expression. This suggests hypoxic environments may promote Tankyrase PARylation and TNKS activity could thus be suboptimal under normoxia. Future work should interrogate the state of PARP activity and differences in PARylation of substrates under normoxic and hypoxic conditions.

115

Chapter 5

Discussion and Future Directions 5.1 Overall perspective

The overall theme of my PhD thesis involves broadening the landscape of Tankyrase biology and understanding new modes of its regulation. The studies described in chapters 2-4 have addressed some of these questions and emphasizes how much more knowledge is required regarding the intricacies of Tankyrase function. My work reveals that there is significant complexity surrounding the selectivity and regulation of these enzymes. Of the proteins that behave as substrates, only some proteins are subjected to PARdU-mediated degradation while others are seemingly exempt from this pathway. It remains unclear the determinants which control Tankyrase substrates selection into degradation pathways. Furthermore, little is known about the host range effects that control specificity for PARdU of RNF146 and Tankyrase.

5.2 Role of WWE domain E3 ligases in PARdU

WWE domains of Deltex1, HUWE1, TRIP12 to bind to PAR PTMs (Wang et al., 2012). There are several HUWE1 substrates that undergo proteasome mediated destruction including N-Myc, p53, Mcl-1 and unassembled soluble proteins while histone H1 and c-Myc ubiquitylation by HUWE1 controls their function without contributing to protein degradation (Chen et al., 2005; Zhong et al., 2005; Adhikary et al., 2005; Zhao et al., 2009; Xu et al., 2016; Mandemaker et al., 2017). TRIP12 has been identified as the E3 ligase for ARF as well as the ubiquitin fusion degradation pathway (Park et al., 2009; Chen et al., 2010). Thus far, only the C-terminal catalytic fragment of MEKK1 (MEKK1C) in T-cells has been identified as a Deltex substrate (Liu and Lai 2005). However, there are no published studies associating these E3 ligases with PARylation. Further work is necessary to determine whether these E3’s can participate in PARdU and compete with RNF146 for substrates or recognize a completely different subset of proteins. It is also interesting to investigate whether these other WWE domains can recognize Tankyrase-mediated PARylation or if they recognize PAR molecules conjugated by other PARP proteins.

116

Future work will utilize the BioID method as well as corresponding WWE domain mutations to determine the PARylation-dependent proteome of all WWE domain E3 ligases. The same experimental parameters used in Chapter 2 will be applied to the other E3’s. The equivalent R163A mutation in RNF146 will be made in the other proteins and used in the BioID assay. This data will hopefully provide a global overview of PARylation-dependent proteome of WWE E3 ligases and provide insight into any similarities or differences between the function of these enzymes.

The differential regulation of SH3BP5 by RNF146 across different cell types suggests that PARdU and the WWE domain E3 ligases operate in a cell-type context. It will be interesting to study whether Tankyrase substrates are regulated by different E3 ligases across tissues or if there is a PAR reading E3 ligase per tissue type. It remains unclear whether PARdU can be triggered as a response to external stimuli such as cell stress, DNA damage, hypoxia, bacterial toxins, or nutrient starvation. Different stimuli can be considered as an additional experimental parameter to the BioID experiment for each E3 in monitoring changes in substrate profiles. Once a core set of PARdU substrates has been established, validation on a single substrate basis can be interrogated across multiple tissue types and stimuli. These sets of experiments will hopefully provide a comprehensive and systematic overview of the specificity and selectivity of PARdU and provide significant understanding of mechanistic rules governing this system. These analyses may contribute to the design of therapeutics cancer and genetic inherited diseases.

5.3 Rab11a activation during lumenogenesis

In chapter 3, I demonstrated that the Rab11 GEFs SH3BP5 and SH3BP5L are both required for lumen formation in MDCK cysts. These GEFs have redundant functions towards Rab11 activation. This was the first time a mammalian function was described for these proteins in terms of Rab11 biology. Previous work has shown that a Rab cascade initiated by Rab11-GTP is important for activation of Rab8 through recruitment of its GEF Rabin8 for both ciliogenesis and lumenogenesis (Knödler et al., 2010; Bryant et al., 2010). However, the proteins that are responsible for activating Rab11 were unknown. My work provides mechanistic understanding for Rab11 activation during these processes. Using the double knockout 3BP5/3BP5L MDCK cells, I can examine if Rab8-GTP levels are decreased in these cells. I can also examine using

117

advanced microscopy techniques if the localization of Rabin8 and Rab8 are perturbed. It will be also interesting to look at the localization of Rab11 and Rab8 GEFs during lumen formation from the initial polarity inversion stage to the apical membrane initiation site to the final expansion of the lumen and assess PODXL localization simultaneously. It will be also interesting to observe when SH3BP5/L localize with Rab11 itself and if they are constantly co-localized during lumen formation.

Since Tankyrase inhibition partially rescued single lumen formation in Rab11a knockdown cells, it suggests that Tankyrase is also involved in Rab11a independent or other downstream signaling. Tankyrase could be regulating the GEFs for Rab8 which have been shown to be more critical than Rab11a, the exocyst complex, or other polarity complexes (Bryant et al., 2010; Mrozowska and Fukuda 2016). Since our data suggests PARylation of SH3BP5 and SH3BP5L inhibits their GEF function, a direct investigation of this idea will be performed. These GEFs can be modified in vitro by incubation with recombinant Tankyrase and NAD+ and subjected to the Rab11a NMR exchange assay. Exchange rates between GEF incubated with wildtype and catalytic dead Tankyrase will be compared and the expectation is wildtype Tankyrase will inhibit GEF activity while catalytic dead Tankyrase will have a minimal effect. The caveat of this experiment is that the stoichiometry of PARylated GEF will be difficult to quantify.

Through ectopic expression experiments, I have shown that Tankyrase can affect the localization of both SH3BP5/L and Rab11a into SAM domain dependent aggregates. To address whether Tankyrase altered localization affects Rab11a activity, Rab11a-GTP pulldowns with different Tankyrase mutants can be examined. Based on the MDCK model, Tankyrase overexpression should decrease Rab11a activity upon Tankyrase overexpression and sequester SH3BP5/L into Tankyrase aggregate structures. The SAM domain mutant should have a positive effect on Rab11a activity. Additionally, a sucrose gradient and differential centrifugation between control and RNF146 deficient MDCK cells could also show that SH3BP5/L localize to Tankyrase associated assemblies. The increased abundance of Tankyrase in the knockout cells could cause increased sequestration of the Rab11a GEFs and explain why Rab11a-GTP levels are decreased when there are high protein levels of Tankyrase.

118

SH3BP5 has distinct functions in cells whether that is functioning as a positive regulator of JNK at mitochondria or as a GEF for Rab11 at recycling endosomes and it remains unclear how SH3BP5 is regulated. It is possible SH3BP5 exists in isolated pools that regulate each pathway independently or if there is a regulatory switch that directs SH3BP5 molecules to either pathway as needed. Further study is required to determine if Tankyrase affects these GEFs by PARylation dependent or independent mechanisms to explain the regulatory mechanism that alters SH3BP5 localization and its downstream function.

5.4 Tissue specificity and function of Tankyrase hydroxylation

In chapter 4, I demonstrated that Tankyrase 2 PARP activity is inhibited by FIH in a catalytic dependent manner. However, I observed no changes in Tankyrase 2 protein stability nor effects on several select substrates in multiple cell types under FIH depleted or deficient conditions. These findings suggest that FIH could be regulating Tankyrase 2 activity that is not involved with PARdU. Hydroxylation sites were also identified on Tankyrase 1 and similar experiments will need to be performed to determine if FIH also inhibits Tankyrase 1 activity (Cockman et al., 2009). To achieve a better understanding of what Tankyrase-regulated biology FIH regulates, a more global and expansive investigation needs to occur. These experiments have been limited by the absence of analytic tools such as hydroxyl-specific monoclonal antibodies or mass spectrometry-based approaches to monitor hydroxylation status of FIH targets such as Tankyrase. Development of these analytic reagents would be enabling for future experiments.

Future experiments should examine initially the endogenous hydroxylation status of Tankyrase in a more global setting. Previous studies of FIH substrates like Notch and IκB have performed proteomic analyses of hydroxylation status across multiple cell-types (Cockman et al., 2006; Coleman et al., 2007). Accordingly, the activity of Tankyrase can be assessed across tissues and cells using newly developed ADP-ribose binding reagents (Gibson et al., 2017). Observations from these experiments could be used to correlate PARylation activity with hydroxylation status. This analysis would help identify if there is selectivity and specificity of FIH regulation of Tankyrase. Ideally, cell types or tissues with higher or lower correlations between hydroxylation and PARylation will be examined and cross-referenced with known Tankyrase substrates to identify biological relationships. Alternatively, proteins with significant roles in particular cell

119

types can be mined for Tankyrase binding motifs to identify a regulatory link between FIH and Tankyrase.

120

References

Adhikary, S., Marinoni, F., Hock, A., Hulleman, E., Popov, N., Beier, R., Bernard, S., Quarto, M., Capra, M., Goettig, S., Kogel, U., Scheffner, M., Helin, K., and Eilers, M. (2005) The Ubiquitin Ligase HectH9 Regulates Transcriptional Activation by Myc and is Essential for Tumor Cell Proliferation. Cell 123, 409-421.

Andrabi, S.A., Kang, H.C., Haince, J.F., Lee, Y.I., Zhang, J., Chi, Z., West, A.B., Koehler, R.C., Poirier, G.G., Dawson, T.M., and Dawson, V.L. (2011) Iduna protects the brain from glutamate excitotoxicity and stroke by interfering with poly(ADP-ribose) polymer-induced cell death. Nat Med 17, 692-699.

Aravind, L. (2001) The WWE domain: a common interaction module in protein ubiquitination and ADP ribosylation. Trends Biochem Sci 26, 273-275. Arqués, O., Chicote, I., Puig, I., Tenbaum, S.P., Argilés, G., Dienstmann, R., Fernández, N., Caratù, G., Matito, J., Silberschmidt, D., Rodon, J., Landolfi, S., Prat, A., Espín, E., Charco, R., Nuciforo, P., Vivancos, A., Shao, W., Tabernero, J., and Palmer, H.G. (2016) Tankyrase Inhibition Blocks Wnt/β-Catenin Pathway and Reverts Resistance to PI3K and AKT Inhibitors in the Treatment of Colorectal Cancer. Clin Cancer Res 22, 644-656.

Ashburner, M., Ball, C.A., Blake, J.A., Botstein, D., Butler, H., Cherry, J.M., Davis, A.P., Dolinski, K., Dwight, S.S., Eppig, J.T., Harris, M.A., Hill, D.P., Issel-Tarver, L., Kasarskis, A., Lewis, S., Matese, J.C., Richardson, J.E., Ringwald, M., Rubin, G.M., and Sherlock, G. (2000) Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet 25, 25-29.

Bae, J., Donigian, J.R., and Hsueh, A.J.W. (2003) Tankyrase 1 interacts with Mcl-1 Proteins and Inhibits their Regulation of Apoptosis. J Biol Chem 278, 5195-5204.

Bao, R., Christova, T., Song, S., Angers, S., Yan, X., and Attisano, L. (2012) Inhibition of tankyrases induces Axin stabilization and blocks Wnt signalling in breast cancer cells. PLoS One 7:e48670 doi: 10.1371/journal.pone.0048670.

Bhardwaj, A., Yang, Y., Ueberheide, B., and Smith, S. (2017) Whole proteome analysis of human tankyrase knockout cells reveals targets of tankyrase-mediated degradation. Nat Commun 8, doi:10.1038/s41467-017-02363-w

Bisht, K.K., Dudognon, C., Chang, W.G., Sokol, E.S., Ramirez, A., and Smith S. (2012) GDP- Mannose-4,5-Dehydratase is a Cytosolic Partner of Tankyrase 1 That Inhibits its Poly(ADP- Ribose) Polymerase Activity. Mol Cell Biol 32, 3044-3053.

Bryant, D.M., Datta, A., Rodríguez-Fraticelli, A.E., Peränen, J., Martín-Belmonte, F., and Mostov, K.E. (2010) A molecular network for de novo generation of the apical surface and lumen. Nat Cell Biol 12, 1035-1045.

Callow, M.G., Tran, H., Phu, L., Lau, T., Lee, J., Sandoval, W.N., Liu, P.S., Bheddah, S., Tao, J., Lill, J.R., Hongo, J.A., Davis, D., Kirkpatrick, D.S., Polakis, P., and Costa, M. (2011)

121

Ubiquitin Ligase RNF146 regulates tankyrase and axin to promote Wnt signaling. PLoS One 6 e22595. doi:10.1371/journal.pone.0022595.

Campbell, C.I., Samavarchi-Tehrani, P., Barrios-Rodiles, M., Datti, A., Gingras, A.C., and Wrana, J.L. (2016) The RNF146 and Tankyrase pathway maintains the junctional Crumbs complex through regulation of angiomotin. J Cell Sci 129, 3396-3411.

Caswell, P.T., Spence, H.J., Parson, M., White, D.P., Clark, K., Cheng, K.W., Mills, G.B., Humphries, M.J., Messent, A.J., Anderson, K.I., McCaffrey, M.W., Ozanne, B.W., and Norman, J.C. (2007) Rab25 associates with alpha5beta1 integrin to promote invasive migration in 3D microenvironments. Dev Cell 13, 496-510.

Chambers, J.W., and LoGrasso, P.V. (2011) Mitochondria c-Jun N-terminal kinase (JNK) Signaling Initiates Physiological Changes Resulting in Amplification of Reactive Oxygen Species Generation. J Biol Chem 286, 16052-16062.

Chambon, P., Weill, J.D., and Mandel, P. (1963) Nicotinamide mononucleotide activation of new DNA-dependent polyadenylic acid synthesizing nuclear enzyme. Biochem Biophys Res Commun 11, 39-43.

Chambon, P., Weill, J.D., Doly, J., Strosser, M.T., and Mandel, P. (1966) On the formation of a novel adenylic compound by enzymatic extracts of liver nuclei. Biochem Biophys Res Commun 25, 638-643.

Chang, P., Coughlin, M., and Mitchison, T.J. (2005) Tankyrase-1 polymerization of poly(ADP- ribose) is required for spindle structure and function. Nat Cell Biol 7, 1133-1139.

Chang, W., Dynek, J.N., and Smith, S. (2003) TRF1 is degraded by ubiquitin-mediated proteolysis after release from telomeres. Genes Dev 17, 1328-1333.

Chang, W., Dynek, J.N., and Smith, S. (2005) NuMa is a major acceptor of poly(ADP- ribosyl)ation by tankyrase 1 in mitosis. Biochem J 391, 177-184.

Chen, D., Kon, N., Li., M., Zhang, W., Qin, J., and Gu, W. (2005) ARF-BP1/Mule is a Critical Mediator of the ARF Tumor Suppressor. Cell 121, 1071-1083.

Chen, D., Shan, J., Zhu, W.G., Qin, J., and Gu, W. (2010) Transcription-independent ARF regulation in oncogenic stress-mediated p53 responses. Nature 464, 624-627.

Cheng, H., Li, X., Wang, C., Chen, Y., Li, S., Tan, J., Tan, B., and He, Y. (2019) Inhibition of tankyrase by a novel small molecule significantly attenuates prostate cancer cell proliferation. Cancer Lett 443, 80-90.

Cheng, K.W., Lahad, J.P., Kuo, W.L., Lapuk, A., Yamada, K., Auersperg, N., Liu, J., Smith- McCune, K., Lu, K.H., Fishman, D., Gray, J.W., and Mills, G.B. (2004) The RAB25 small GTPase determines aggressiveness of ovarian and breast cancers. Nat Med 10, 1251-1260.

122

Chi, N.W., and Lodish, H. (2000) Tankyrase is a Golgi-associated MAP kinase substrate that interacts with IRAP in Glut4 vesicles. J Biol Chem 275, 38437-38444.

Chiang, Y.J., Nguyen, M.L., Gurunathan, S., Kaminker, P., Tessarollo, L., Campisi, J., and Hodes, R.J. (2005) Generation and characterization of telomere length maintenance in tankyrase 2-deficient mice. Mol Cell Biol 26, 2037-2043.

Chiang, Y.J., Hsiao, S.J., Yver, D., Cushman, S.W., Tessarollo, L., Smith, S., and Hodes, R.J. (2008) Tankyrase 1 and tankyrase 2 are essential but redundant for mouse embryonic development. PLoS One 3 e2639. doi: 10.1371/journal.pone.0002639.

Cockman, M. E., Lancaster, D. E., Stolze, I. P., Hewitson, K. S., Mc-Donough, M. A., Coleman, M. L., Coles, C. H., Yu, X., Hay, R. T., Ley, S. C., Pugh, C. W., Oldham, N. J., Masson, N., Schofield, C. J., and Ratcliffe, P. J. (2006) Posttranslational hydroxylation of ankyrin repeats in IkB proteins by the hypoxia-inducible factor (HIF) asparaginyl hydroxylase,factor inhibiting HIF (FIH). Proc. Natl. Acad. Sci. U. S. A. 103, 14767–14772

Cockman, M.E., Webb, J.D., Kramer, H.B., Kessler, B.M., and Ratcliffe, P.J. (2009) Proteomics- based identification of novel factor inhibiting hypoxia-inducible factor (FIH) substrates indicates widespread asparaginyl hydroxylation of ankyrin repeat domain-containing proteins. Mol Cell Proteomics 8, 535-546.

Coleman, M. L., McDonough, M. A., Hewitson, K. S., Coles, C., Mecinovic,J., Edelmann, M., Cook, K. M., Cockman, M. E., Lancaster, D. E., Kessler, B. M., Oldham, N. J., Ratcliffe, P. J., and Schofield, C. J. (2007) Asparaginyl hydroxylation of the Notch ankyrin repeat domain by factor inhibiting hypoxia-inducible factor. J Biol Chem 282, 24027–24038

Cook, B.D., Dynek, J.N., Chang, W., Shostak, G., and Smith, S. (2002) Role for the Related Poly(ADP-Ribose) Polymerases Tankyrase 1 and 2 at Human Telomeres. Mol Cell Biol 22, 332- 342.

Coyaud, E., Mis, M, Laurent, E.M., Dunham, W.H., Couzens, A.L., Robitaille, M., Gingras, A.C., Angers, S., and Raught, B. (2015) BioID-based Identification of Skp Cullin F-box (SCF) β-TrCP1/2 E3 Ligase Substrates. Mol Cell Proteomics 14, 1781-1795.

Craig, R., and Beavis, R.C. (2004) TANDEM: matching proteins with tandem mass spectra. Bioinformatics 20, 1466-1467.

DaRosa, P.A., Wang, Z., Jiang, X., Pruneda, J.N., Cong, F., Klevit, R.E., and Xu, W. (2014) Allosteric activation of the RNF146 ubiquitin ligase by a poly(ADP-ribosyl)ation signal. Nature 517, 223-226.

DaRosa, P.A., Ovchinnikov, S., Xu, W., and Klevit, R.E. (2016) Structural insights into SAM domain-mediated tankyrase oligomerization. Protein Sci 25, 1744-1752.

123

DaRosa, P.A., Klevit, R.E., and Xu, W. (2018) Structural basis for tankyrase-RNF146 interaction reveals noncanonical tankyrase-binding motifs. Protein Sci 27, 1057-1067.

De Rycker, M., Venkatesan, R.N., Wei, C., and Price, C.M. (2003) Vertebrate tankyrase domain structure and sterile alpha motif (SAM)-mediated multimerization. Biochem J 372, 87-96.

De Rycker, M., and Price, C.M. (2004) Tankyrase polymerization is controlled by its sterile alpha motif and poly(ADP-ribose). Mol Cell Biol 24, 9802-9812.

Desclozeaux, M., Venturato, J., Wylie, F.G., Kay, J.G., Joseph, S.R., Le, H.T., and Stow, J.L. (2008) Active rab11 and functional recycling endosomes are required for E-cadherin trafficking and lumen formation during epithelial morphogenesis. Am J Physiol Cell Physiol 295, 545-556.

Deutsch, E.W., Mendoza, L., Shteynberg, D., Farrah, T., Lam, H., Tasman, N., Sun, Z., Nilsson, E., Pratt, B., Prazen, B., Eng, J.K., Martin, D.B., Nesvizhskii, A, I., and Aebersold, R. (2010) A guided tour of the Trans-Proteomic Pipeline. Proteomics 10, 1150-1159.

Dong, G., Medkova, M., Novick, P., and Reinisch, K.M. (2007) A catalytic coiled coil: structural insights into the activation of the Rab GTPase Sec4p by Sec2p. Mol Cell 25, 455-462.

Donigian, J.R., and de Lange, T. (2007) The Role of the Poly(ADP-ribose) Polymerase Tankyrase1 in Telomere Length Control by the TRF1 Component of the Shelterin Complex. J Biol Chem 282, 22662-22667.

Dynek, J.N., and Smith, S. (2004) Resolution of Sister Telomere Association Is Required for Progression through Mitosis. Science 304, 97-100.

Eathiraj, S., Mishra, A., Prekeris, R., and Lambright, D.G. (2006) Structural basis for Rab11- mediated Recruitment of FIP3 to Recycling Endosomes. J Mol Biol 364, 121-135.

Eng, J.K., Jahan, T.A., and Hoopmann, M.R. (2013) Comet: an open-source MS/MS sequence database search tool. Proteomics 13, 22-24.

Fan, C., Yarravarapu, N., Chen, H., Kulak, O., Dasari, P., Herbert, J., Yamaguchi, K., Lum, L., and Zhang, X. (2018) Regulation of tankyrase activity by a catalytic domain dimer interface. Biochem Biophys Res Commun 503, 1780-1785.

Ferguson, J.E. 3rd, Wu, Y., Smith, K., Charles, P., Powers, K., Wang, H., and Patterson, C. (2007) ASB4 is a hydroxylation substrate of FIH and promotes vascular differentiation via an oxygen-dependent mechanism. Mol Cell Biol 27, 6407-6419.

Fröse, J., Chen, M.B., Hebron, K.E., Reinhardt, F., Hajal, C., Zijlstra, A., Kamm, R.D., and Weinberg, R.A. (2018) Epithelial-Mesenchymal Transition Induces Podocalyxin to Promote Extravasation via Ezrin Signaling. Cell Rep 24, 962-972.

Fujimura, S., Hasegawa, S., Shimizu, Y., and Sugimura, T. (1967) Polymerization of the adenosine 5’-diphosphate-ribose moiety of nicotinamide-adenine dinucleotide by nuclear 124

enzyme. I. Enzymatic reactions. Biochim Biophys Acta 145, 247-259.

Fujita, S., Mukai, T., Mito, T., Kodama, S., Nagasu, A., Kittaka, M., Sone, T., Ueki, Y., and Morita, Y. (2018) Pharmalogical inhibition of tankyrase induces bone loss in mice by increasing osteoclastogenesis. Bone 106, 156-166.

Gálvez-Santisteban, M., Rodriguez-Fraticelli, A.E., Bryant, D.M., Vergarajauregui, S., Yasuda, T., Bañón-Rodríguez, I., Bernascone, I., Datta, A., Spivak, N., Young, K., Slim, C.L., Brakeman, P.R., Fukuda, M., Mostov, K.E., and Martín-Belmonte, F. (2012) Synaptotagmin-like proteins control the formation of a single apical membrane domain in epithelial cells. Nat Cell Biol 14, 838-849.

Gasmi-Seabrook, G.M., Marshall, C.B., Cheung, M., Kim, B., Wang, F., Jang, Y.J., Mak, T.W., Stambolic, V., and Ikura, M. (2010) Real-time NMR study of guanine nucleotide exchange and activation of RhoA by PDZ-RhoGEF. J Biol Chem 285, 5137-5145.

Gene Ontology Consortium. (2015) Gene Ontology Consortium: going forward. Nucleic Acids Res 43, D1049-1056.

Gibson, B.A., and Kraus, W.L. (2012) New insights into the molecular and cellular functions of poly(ADP-ribose) and PARPs. Nat Rev Mol Cell Biol 13, 411-424.

Gibson, B.A., Conrad, L.B., Huang, D., and Kraus, W.L. (2017) Generation and Characterization of Recombinant Antibody-like ADP-Ribose Binding Proteins. Biochemistry 56, 6305-6316.

Goldenring, J.R., Shen, K.R., Vaughan, H.D., and Modlin, I.M. (1993) Identification of a small GTP-binding protein, Rab25, expressed in the gastrointestinal mucosa, kidney, and lung. J Biol Chem 268, 18419-18422.

Guettler, S., LaRose, J., Petsalaki, E., Gish, G., Scotter, A., Pawson, T., Rottapel, R., and Sicheri, F. (2011) Structural basis and sequence rules for substrate recognition by Tankyrase explain the basis for cherubism disease. Cell 147, 1340-1354.

Guichard, A., Nizet, V., and Bier, E. (2014) RAB11-mediated trafficking in host-pathogen interactions. Nat Rev Microbiol 12, 624-634.

Gunawan, B.K., Liu, Z.X., Han, D., Hanawa, N., Gaarde, W.A., and Kaplowitz, N. (2006) c-Jun N-terminal kinase plays a major role in murine acetaminophen hepatotoxicity. Gastroenterology 131, 165-178.

Guo, Z., Hou, X., Goody, R.S. and Itzen, A. (2013) Intermediates in the Guanine Nucleotide Exchange Reaction of Rab8 Catalyzed by Rabin8/GRAB. J Biol Chem 288, 32466-32474.

Gupta, S., Barrett, T., Whitmarsh, A., Cavanagh, J., Sluss, H., Derijard, B., and Davis, R. (1996) Selective interaction of JNK protein kinase isoforms with transcription factors. EMBO J 15, 2760-2770. 125

Gustafson, C.T., Mamo, T., Shogren, K.L., Maran, A., and Yaszemski, M.J. (2017) FH535 Suppresses Osteosarcoma Growth In Vitro and Inhibits Wnt Signaling through Tankyrases. Front Pharmacol 8, doi: 10.3389/fphar.2017.00285.

Ha, G.H., Kim, H.S., Go, H., Lee, H., Seimiya, H., Chung, D.H., and Lee, C.W. (2012) Tankyrase-1 function at telomeres and during mitosis is regulated by Polo-like kinase-1- mediated phosphorylation. Cell Death Differ 19, 321-332.

Haeussler, M., Schönig, K., Eckert, H., Eschstruth, A., Mianné, J., Renaud, J.B., Schneider- Maunoury, S., Shkumatava, A., Teboul, L., Kent, J., Joly, J.S., and Concordet, J.P. (2016) Evaluation of off-target and on-target scoring algorithms and integration into the guide RNA selection tool CRISPOR. Genome Biol 17, doi: 10.1186/s13059-016-1012-2.

Hanawa, N., Shinohara, M., Saberi, B., Gaarde, W.A., Han, D., and Kaplowitz, N. (2008) Role of JNK translocation to mitochondria leading to inhibition of mitochondria bioenergetics in acetaminophen-induced liver injury. J Biol Chem 283, 13565-13577.

Hardy, A.P., Prokes, I., Kelly, L., Campbell, I.D., Schofield, C.J. (2009) Asparaginyl beta- hydroxylation of proteins containing ankyrin repeat domains influences their stability and function. J Mol Biol 392, 994-1006.

Hart, T., Chandrashekhar, M., Aregger, M., Steinhart, Z., Brown, K.R., MacLeod, G., Mis, M., Zimmerman, M., Fradet-Turcotte, A., Sun, S., Mero, P., Dirks, P., Sidhu, S., Roth, F.P., Rissland, O.S., Durocher, D., Angers, S., and Moffat, J. (2015) High-Resolution CRISPR Screens Reveal Fitness Genes and Genotype-Specific Cancer Liabilities. Cell 163, 1515-1526.

Hasegawa, S., Fujimura, S., Shimizu, Y., and Sugimura, R. (1967) The polymerization of the adenosine 5’-diphosphate-ribose moiety of nicotinamide-adenine dinucleotide by nuclear enzyme. II. Properties of the reaction product. Biochim Biophys Acta 149, 369-376.

He, F., Tsuda, K., Takahashi, M., Kuwasako, K., Terada, T., Shirouzu, M., Watanabe, S., Kigawa, T., Kobayashi, N., Guntert, P., Yokoyama, S., and Muto, Y. (2012) Structural insight into the interaction of ADP-ribose with PARP WWE domains. FEBS Lett 586, 3858-3864.

Hewitson, K.S., McNeill, L.A., Riordan, M.V., Tian, Y.M., Bullock, A.N., Welford, R.W., Elkins, J.M., Oldham, N.J., Bhattacharya, S., Gleadle, J.M., Ratcliffe, P.J., Pugh, C.W., and Schofield, C.J. (2002) Hypoxia-inducible Factor (HIF) Asparagine Hydroxylase is Identical to Factor Inhibiting HIF (FIH) and is Related to the Cupin Structural Family. J Biol Chem 277, 26351-26355.

Hottiger, M.O., Hassa, P.O., Lüscher, B., Schüler, H., and Koch-Nolte, F. (2010) Towards a unified nomenclature for mammalian ADP-ribosyltransferases. Trends Biochem Sci 35, 208-219.

Huang, S.M., Mishina, Y.M., Liu, S., Cheung, A., Stegmeier, F., Michaud, G.A., Charlat, O., Wiellette, E., Zhang, Y., Wiessner, S., Hild, M., Shi, X., Wilson, C.J., Mickanin, C., Myer, V., Fazal, A., Tomlinson, R., Serluca, F., Shao, W., Cheng, H., Schultz, M., Rau, C., Schirle, M., 126

Schlegl, J., Ghidelli, S., Fawell, S., Lu, C., Curtis, D., Kirschner, M.W., Lengauer, C., Finan, P.M., Tallarico, J.A., Bouwmeester, T., Porter, J.A., Bauer, A., and Cong, F. (2009) Tankyrase inhibition stabilizes axin and antagonizes Wnt signalling. Nature 461, 614–620.

Husebye, H., Aune, M.H., Stenvik, J., Samstad, E., Skjeldal, F., Halaas, O., Nilsen, N.J., Stenmark, H., Latz, E., Lien, E., Mollnes, T.E., Bakke, O., and Espevik, T. (2010) The Rab11a GTPase controls Toll-like receptor 4-induced activation of interferon regulatory factor-3 on phagosomes. Immunity 33, 583-596.

Huttlin, E.L., Ting, L., Bruckner, R.J., Gebreab, F., Gygi, M.P., Szpyt, J., Tam, S., Zarraga, G., Colby, G., Baltier, K., Dong, R., Guarani, V., Vaites, L.P., Ordureau, A., Rad, R., Erickson, B.K., Wühr, M., Chick, J., Zhai, B., Kolippakkam, D., Mintseris, J., Obar, R.A., Harris, T., Artavanis-Tsakonas, S., Sowa, M.E., De Camilli, P., Paulo, J.A., Harper, J.W., and Gygi, S.P. (2015) The Bioplex Network: A Systemic Exploration of the Human Interactome. Cell 162, 425- 440.

Ivan, M., Kondo K., Yang, H., Kim, W., Valiando, J., Ohh, M., Salic, A., Asara, J.M., Lane,W.S., and Kaelin, W.G. (2001) HIFalpha targeted for VHL-mediated destruction by proline hydroxylation: implications for O2 sensing. Science 292, 464-468.

Jaakkola, P., Mole, D.R., Tian, Y.M., Wilson, M.I., Gielbert, J., Gaskell, S.J., von Kriegsheim, A., Hebestreit, H.F., Mukherji, M., Schofield, C.J., Maxwell, P.H., Pugh, C.W., and Ratcliffe, P.J. (2001) Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by O2- regulated prolyl hydroxylation. Science 292, 468-372.

Janke, K., Brockmeier, U., Kuhlmann, K., Eisenacher, M., Nolde, J., Meyer, H.E., Mairbäurl, H., and Metzen, E. (2013) Factor inhibiting HIF-1 (FIH-1) modulates protein interactions of apoptosis-stimulating p53 binding protein 2 (ASPP2). J Cell Sci 126, 2629-2640.

Jenkins, M.L., Margaria, J.P., Stariha, J.T.B., Hoffmann, R.M., McPhail, J.A., Hamelin, D.J., Boulanger, M.J., Hirsch, E., and Burke, J.E. (2018) Structural determinants of Rab11 activation by the guanine nucleotide exchange factor SH3BP5. Nat Commun 9, doi:10.1038/s41467-018- 06196-z.

Jia, J., Qiao, Y., Pilo, M.G., Cigliano, A., Liu, X., Shao, Z., Calvisi, D.F., and Chen, X. (2017) Tankyrase inhibitors suppress hepatocellular carcinoma cell growth via modulating the Hippo cascade. PLoS One 12:e0184068. doi: 10.1371/journal.pone.0184068.

Jing, J., and Prekeris, R. (2009) Polarized endocytic transport: The roles of Rab11 and Rab11- FIPs in regulating cell polarity. Histol Histolpathol 24, 1171-1180.

Jones, S., Newman, C., Liu, F., and Segev, N. (2000) The TRAPP Complex is a Nucleotide Exchanger for Ypt1 and Ypt31/32. Mol Biol Cell 11, 4403-4411.

Kang, D.H., Lee, D.J., Lee, S., Lee, S.Y., Jun, Y., Kim, Y., Kim, Y., Lee, J.S., Lee, D.K., Lee, S., Jho, E.H., Yu, D.Y., and Kang, S.W. (2017) Interaction of tankyrase and peroxiredoxin II is

127

indispensable for the survival of colorectal cancer. Nat Commun 8, 40. doi: 10.1038/s41467-017- 00054-0.

Kang, D.H., Lee, D.H.S., and Kang, S.W. (2017) Survival of APC-mutant colorectal cancer cells requires interaction between tankyrase and a thiol peroxidase, peroxiredoxin II. BMB Rep 50, 391-392.

Kang, H.C., Lee, Y.I., Shin, J.H., Andrabi, S.A., Chi, Z., Gagné, J.P., Lee, Y., Ko, H.S., Lee. B.D., Poirier, G.G., Dawson, V.L., and Dawson, T.M. (2011) Iduna is a poly(ADP-ribose) (PAR)-dependent E3 ubiquitin ligase that regulates DNA damage. Proc Natl Acad Sci 108, 14103-14108.

Karttunen, S., Duffield, M., Scrimgeour, N.R., Squires, L., Lim, W.L., Dallas, M.L., Scragg, J.L., Chicher, J., Dave, K.A., Whitelaw, M.L., Peers, C., Gorman, J.J., Gleadle, J.M., Rychkov, G.Y., and Peet, D.J. (2015) Oxygen-dependent hydroxylation by FIH regulates the TRPV3 ion channel. J Cell Sci 128, 225-231.

Kelly, E.E., Horgan, C.P., McCaffrey, M.W. and Young, P. (2011) The role of endosomal- recycling in long-term potentiation. Cell Mol Life Sci 68, 185-194.

Kelly, E.E., Horgan, C.P., and McCaffrey, M.W. (2012) Rab11 proteins in health and disease. Biochem Soc Trans 40, 1360-1367.

Kessler, D., Gruen, G.C., Heider, D., Morgner, J., Reis, H., Schmid, K.W., and Jendrossek, V. (2012) The Action of Small GTPases Rab11 and Rab25 in Vesicle Trafficking During Cell Migration. Cell Physiol Biochem 29, 647-656.

Kessner, D., Chambers, M., Burke, R., Agus, D., and Mallick, P. (2008) ProteoWizard: open source software for rapid proteomics tools development. Bioinformatics 24, 2534-2536.

Kim, D.I., Birenda, K.C., Zhu, W., Motamedchaboki, K., Doye, V., and Roux, K.J. (2014) Probing nuclear pore complex architecture with proximity-dependent biotinylation. Proc Natl Acad Sci 111, 2453-2461.

Kim, M.K., Dudognon, C., and Smith, S. (2012) Tankyrase 1 regulates centrosome function by controlling CPAP stability. EMBO Rep 13, 724-732.

Kiriakidis, S., Henze, A.T., Kruszynska-Ziaja, I., Skobridis, K., Theodorou, V., Paleolog, E.M., and Mazzone, M. (2015) Factor-inhibiting HIF-1 (FIH-1) is required for human vascular endothelial cell survival. FASEB J 29, 2814-2827.

Knödler, A., Feng, S., Zhang, J., Zhang, X., Das, A., Peränen, J., and Guo, W. (2010) Coordination of Rab8 and Rab11 in primary ciliogenesis. Proc Natl Acad Sci 107, 6346-6351.

Knorre, D.G., Kudryashova, N.V., and Godovikova, T.S. (2009) Chemical and Functional Aspects of Posttranslational Modification of Proteins. Acta Naturae 1, 29-51.

128

Kobayashi, K., Yamaguchi, M., Miyazaki, K., Imai, H., Yokoe, K., Ono, R., Nosaka, T., and Katayama, N. (2016) Expressions of SH3BP5, LMO3, and SNAP25 in diffuse large B-cell lymphoma cells and their association with clinical features. Cancer Med 5, 1802-1809.

Krueger, K.M., and Barbieri, J.T. (1995) The family of bacterial ADP-ribosylating exotoxins. Clin Microbiol Rev 8, 34-47.

Kuimov, A.N., Kuprash, D.V., Petrov, V.N., Vdovichenko, K.K., Scanlan, M.J., Jongeneel, C.V., Lagarkova, M.A., and Nedospasov, S.A. (2001) Cloning and characterization of TNKL, a member of tankyrase gene family. Genes Immune 2, 52-55.

Laflamme, C., Assaker, G., Ramel, D., Dorn, J.F., She, D., Maddox, P.S., and Emery, G. (2012) Evi5 promotes collective cell migration through its Rab-GAP activity. J Cell Biol 198, 57-67.

Lahav-Ariel, L., Caspi, M., Nadar-Ponniah, P.T., Zelikson, N., Hofmann, I., Hanson, K.K., Franke, W.W., Sklan, E.H., Avraham, K.B., and Rosin-Arbesfeld, R. (2018) Striatin is novel modulator of cell adhesion. FASEB J 33, doi: 10.1096/fj.201801882R

Lamers, I.J.C., Reijnders, M.R.F., Venselaar, H., Kraus, A., Study, D.D.D., Jansen, S., de Vries, B.B.A., Houge, G., Gradek, G.A., Seo, J., Choi, M., Chae, J.H., van der Burgt, I., Pfundt, R., Letteboer, S.J.F., van Beersum, S.E.C., Dusseljee, S., Brunner, H.G., Doherty, D., Kleefstra, T., and Roepman, R. (2017) Recurrent De Novo Mutations Disturbing the GTP/GDP Binding Pocket of RAB11B Cause Intellectual Disability and a Distinctive Brain Phenotype. Am J Hum Genet 101, 824-832.

Lancaster, D.E., McNeill, L.A., McDonough, M.A., Aplin, R.T., Hewitson, K.S., Pugh, C.W., Ratcliffe, P.J., and Schofield, C.J. (2004) Disruption of dimerization and substrate phosphorylation inhibit factor inhibiting hypoxia-inducible factor (FIH) activity. Biochem J 383, 429-437.

Lando, D., Peet, D.J., Whelan, D.A., Gorman, J.J., and Whitelaw, M.L. (2002) Asparagine Hydroxylation of the HIF Transactivation Domain: A Hypoxic Switch. Science 295, 858-861.

Lando, D., Peet, D.J., Gorman, J.J., Whelan, D.A., Whitelaw, M.L., and Bruick, R.K. (2002) FIH-1 is an asparaginyl hydroxylase enzyme that regulates the transcriptional activity of hypoxia-inducible factor. Genes Dev 16, 1466-1471.

Lau, T., Chan, E., Callow, M., Waaler, J., Boggs, J., Blake, R.A., Magnuson, S., Sambrone, A., Schutten, M., Firestein, R., Machon, O., Korinek, V., Choo, E., Diaz, D., Merchant, M., Polakis, P., Holsworth, D.D., Krauss, S., and Costa, M. (2013) A novel tankyrase small-molecule inhibitor suppresses APC mutation-driven colorectal tumor growth. Cancer Res 73, 3132-3144.

Leslie Pedrioli, D.M., Leutert, M., Bilan, V., Nowak, K., Gunasekara, K., Ferrari, E., Imhof, R., Malmström, L., and Hottiger, M.O. (2018) Comprehensive ADP-ribosylome analysis identifies tyrosine as an ADP-ribose acceptor site. EMBO Rep 19 e45310. doi: 10.15252/embr.201745310.

129

Levaot, N., Voytyuk, O., Dimitriou, I., Sircoulomb, F., Chandrakumar, A., Deckert, M., Krzyzanowski, P.M., Scotter, A., Gu, S., Janmohamed, S., Cong, F., Simoncic, P.D., Ueki, Y., La Rose, J., Rottapel, R. (2011) Loss of Tankyrase-mediated destruction of 3BP2 is the underlying pathogenic mechanism of cherubism. Cell 147, 1324–1339.

Li, N., Zhang, Y., Han, X., Liang, K., Wang, J., Feng, L., Wang, W., Songyang, Z., Lin, C., Yang, L., Yu, Y., and Chen, J. (2014) Poly-ADP-ribosylation of PTEN by tankyrases promote PTEN degradation and tumor growth. Genes Dev 29, 157-170.

Li, W., Han, H., Zhou, M., Yang, B., Ta, A.P., Li, N., Chen, J., and Wang, W. (2017) Proteomic Analysis of the Human Tankyrase Protein Interaction Network Reveals its Role in Pexophagy. Cell Rep 20, 737-749.

Liu, C.Y., Zha, Z.Y., Zhou, X., Zhang, H., Huang, W., Zhao, D., Li, T., Chan, S.W., Lim C.J., Hong, W., Zhao, S., Xiong, Y., Lei, Q.Y., and Guan, K.L. (2010) The hippo tumor pathway promotes TAZ degradation by phosphorylating a phosphodegron and recruiting the SCF{beta}- TrCP E3 ligase. J Biol Chem 285, 37159-37169.

Liu, G., Zhang, J., Larsen, B., Stark, C., Breitkreutz, A., Lin, Z.Y., Breitkreutz, B.J., Ding, Y., Colwill, K., Pasculescu, A., Pawson, T., Wrana, J.L., Nesvizhskii, A.I., Raught, B., Tyers, M., and Gingras, A.C. (2010) ProHits: integrated software for mass spectrometry-based interaction proteomics. Nat Biotechnol 28, 1015-1017.

Liu, W.H., and Lai, M.Z. (2005) Deltex Regulates T-cell activation by targeted degradation of active MEKK1. Mol Cell Biol 25, 1367-1378.

Lock, J.G., and Stow, J.L. (2005) Rab11 in recycling endosomes regulates the sorting and basolateral transport of E-cadherin. Mol Cell Biol 16, 1744-1755. Lyons, R.J., Deane, R., Lynch, D.K., Ye, Z.S., Sanderson, G.M., Eyre, H.J., Sutherland, G.R., and Daly, R.J. (2001) Identification of a novel human tankyrase through its interaction with the adaptor protein Grb14. J Biol Chem 276, 17172-17180.

Ma, L., Wang, X., Jia, T., Wei, W., Chua, M.S., and So, S. (2015) Tankyrase inhibitors attenuate WNT/β-catenin signaling and inhibit growth of hepatocellular carcinoma cells. Oncotarget 6, 25390-25401.

Mahon, P.C., Hirota, K.,and Semenza, G.L. (2001) FIH-1: A novel protein that interacts with HIF-1 alpha and VHL to mediate repression of HIF-1 transcriptional activity. Genes Dev 15, 2675-2686.

Mandemaker, I.K., van Cuijk, L., Janssens, R.C., Lans, H., Bezstarosti, K., Hoeijmakers, J.H., Demmers, J.A., Vermeulen, W., and Marteijn, J.A. (2017) DNA damage-induced histone H1 ubiquitylation is mediated by HUWE1 and stimulates RNF8-RNF168 pathway. Sci Rep 7, 15353 doi: 10.1038/s41598-017-15194-y.

130

Mariotti, L., Templeton, C.M., Ranes, M., Paracuellos, P., Cronin, N., Beuron, F., Morris, E., and Guettler, S. (2016) Tankyrase requires SAM domain-dependant polymerization to support Wnt-β-catenin signaling. Mol Cell 63, 498-513.

Marshall, C.B., Ho, J., Buerger, C., Plevin, M.J., Li, G.Y., Li, Z., Ikura, M., and Stambolic, V. (2009) Characterization of the intrinsic and TSC2-GAP-regulated GTPase activity of Rheb by real-time NMR. Sci Signal 2, doi: 10.1126/scisignal.2000029

Martin-Belmonte, F., Gassama, A., Datta, A., Yu, W., Rescher, U., Gerke, V., and Mostov, K. (2007) PTEN-mediated apical segregation of phosphoinositides controls epithelial morphogenesis through Cdc42. Cell 128, 383-397.

Martins-Neves, S.R., Paiva-Oliveira, D.I., Fontes-Ribeiro, C., Bovée, J.V.M.G., Cleton-Jansen, A.M., and Gomes, C.M.F. (2018) IWR-1, a tankyrase inhibitor, attenuates Wnt/β-catenin signaling in cancer stem-like cells and inhibits in vivo the growth of a subcutaneous human osteosarcoma xenograft. Cancer Lett 414, 1-15

Masson, N., Singleton, R.S., Sekirnik, R., Trudgian, D.C., Ambrose, L.J., Miranda, M.X., Tian, Y.M., Kessler, B.M., Schofield, C.J., Ratcliffe, P.J. (2012) The FIH hydroxylase is a cellular peroxide sensor that modulates HIF transcriptional activity. EMBO Rep 13, 251-257.

Matsumoto, Y., Larose, J., Kent, O.A., Lim, M., Changoor, A., Zhang, L., Storozhuk, Y., Mao, X., Grynpas, M.D., Cong, F., and Rottapel, R. (2017) RANKL coordinates multiple osteoclastogenic pathways by regulating expression of ubiquitin ligase RNF146. J Clin Invest 127, 1303-1315.

Matsumoto, Y., La Rose, J., Lim, M., Adissu, H.A., Law, N., Mao, X., Cong, F., Mera, P., Karsenty, G., Goltzman, D., Changoor, A., Zhang, L., Stajkowski, M., Grynpas, M.D., Bergmann, C., and Rottapel, R. (2017) Ubiquitin ligase RNF146 coordinates bone dynamics and energy metabolism. J Clin Invest 127, 2612-2625.

Matsushita, M., Yamadori, T., Kato, S., Takemoto, Y., Inazawa, Y., Baba, Y., Hashimoto, S., Sekine, S., Arai, S., Kunikata, T., Kurimoto, M., Kishimoto, T., and Tsukuda, S. (1998) Identification and characterization of a novel SH3-domain binding protein, Sab, which preferentially associates with Bruton’s tyrosine kinase (Btk). Biochem Biophys Res Commun 245, 337-343.

McGuffin, L.J., Bryson, K., and Jones, D.T. (2000) The PSIPRED protein structure prediction server. Bioinformatics 16, 404-405.

McGurk, L., Gomes, E., Guo, L., Mojsilovic-Petrovic, J., Tran, V., Kalb, R.G., Shorter, J., and Bonini, N.M. (2018) Poly(ADP-Ribose) Prevents Pathological Phase Separation of TDP-43 by Promoting Liquid Demixing and Stress Granule Localization. Mol Cell 71, 703-717.

Miyazaki, K., Yamaguchi, M., Imai, H., Kobayashi, K., Tamaru, S., Kobayashi, T., Shiku, H., and Katayama, N. (2015) Gene expression profiling of diffuse large B-cell lymphomas surpervised by CD5 expression. 131

Mizutani, A., Yashiroda, Y., Muramatsu, Y., Yoshida, H., Chikada, T., Tsumura, T., Okue, M., Shirai, F., Fukami, T., Yoshida, M., and Seimiya, H. (2018) RK-287107, a potent and specific tankyrase inhibitor, blocks colorectal cancer cell growth in a preclinical model. Cancer Sci 109, 4003-4014

Morrone, S., Cheng, Z., Moon, R.T., Cong, F., and Xu, W. (2012) Crystal structure of a Tankyrase-Axin complex and its implications for Axin turnover and Tankyrase substrate recruitment. Proc Natl Acad Sci USA 109, 1500-1505.

Mrozowska, P.S., and Fukuda, M. (2016) Regulation of podocalyxin trafficking by Rab small GTPases in 2D and 3D epithelial cell cultures. J Cell Biol 213, 355-369. Nielsen, J.S., and McNagny, K.M. (2009) Role of podocalyxin in health and disease. J Am Soc Nephrol 20, 1669-1676.

Nishizuka, Y., Ueda, K., Nakazawa, K., and Hayaishi, O. (1967) Studies on the polymer of adenosine diphosphate ribose. I. Enzymic formation from nicotinamide adenine dinucleotide in mammalian nuclei. J Biol Chem 242, 3164-3171.

Okada-Iwasaki,R., Takahashi,Y., Watanabe, Y., Ishida, H., Saito, J., Nakai, R., and Asai, A. (2016) The Discovery and Characterization of K-756, a Novel Wnt/β-Catenin Pathway Inhibitor Targeting Tankyrase. Mol Cancer Ther 15 ,1525-1534.

Orford, K., Crockett, C., Jensen, J.P., Weissman, A.M., and Byers, A.W. (1997) Serine phosphorylation-regulated ubiquitination and degradation of beta-catenin. J Biol Chem 272, 24735-24738.

Ozaki, Y., Matsui, H., Asou, H., Nagamachi, A., Aki, D., Honda, H., Yasunaga, S., Takihara, Y., Yamamoto, T., Izumi, S., Ohsugi, M., and Inaba, T. (2012) Poly-ADP Ribosylation of Miki by tankyrase-1 Promotes Centrosome Maturation. Mol Cell 47, 694-706.

Palmieri, D., Bouadis, A., Ronchetti, R., Merino, M.J., and Steeg, P.S. (2006) Rab11a differentially modulates epidermal growth factor-induced proliferation and motility in immortal breast cells. Breast Cancer Res Treat 100, 127-137.

Park, Y., Yoon, S.K., and Yoon, J.B. (2009) The HECT domain of TRIP12 ubiquitinates substrates of the ubiquitin fusion degradation pathway. J Biol Chem 284, 1540-1549.

Paudel, I., Hernandez, S.M., Portalatin, G.M., Chambers, T.P. and Chambers, J.W. (2018) Sab concentrations indicate chemotherapeutic susceptibility in ovarian cancer cell lines. Biochem J 475, 3471-3492.

Pedrioli, P.G. (2010) Trans-proteomic pipeline: a pipeline for proteomic analysis. Methods Mol Biol 604, 213-238.

Peng, H., Kaplan, N., Yang, W., Getsios, S., and Lavker, R.M. (2014) FIH-1 disrupts an 132

EGFR/LRKK1 complex to positively regulate keratinocyte migration. Am J Pathol 184, 3262- 3271.

Pfeffer, S., and Aivazian, D. (2004) Targeting Rab GTPases to distinct membrane compartments. Nat Rev Mol Cell Biol 5, 886-896.

Powelka, A.M., Sun, J., Li, J., Gao, M., Shaw, L.M., Sonnenberg, A., and Hsu, V.W. (2004) Stimulation-dependent recycling of integrin beta1 regulated by Arf6 and Rab11. Traffic 5, 20-36.

Prekeris, R., Davies, J.M., and Scheller, R.H. (2001) Identification of a Novel Rab11/25 Binding Domain Present in Eferin and Rip Proteins. J Biol Chem 276, 38966-38970. Reeder, R.H., Ueda, K., Honjo, T., Nishizuka, Y., and Hayaishi, O. (1967) Studies on the polymer of adenosine diphosphate ribose. II. Characterization of the polymer. J Biol Chem 242, 3172-3179.

Rejman Lipinski, A., Heymann, J., Meissner, C., Karlas, A., Brinkmann, V., Meyer, T.F., and Heuer, D. (2009) Rab6 and Rab11 regulate Chlamydia trachomitis development and golgin-84- dependent Golgi fragmentation. PLoS Pathog 5 e:1000615. doi: 10.1371/journal.ppat.1000615

Riccio, A.A., McCauley, M., Langelier, M.F., and Pascal, J.M. (2016) Tankyrase Sterile α Motif Domain Polymerization Is Required for its Role in Wnt Signaling. Structure 24, 1573-1581.

Riedel, F., Galindo, A., Muschalik, N., and Munro, S. (2018) The two TRAPP complexes of metazoans have distinct roles and act on different Rab GTPases. J Cell Biol 217, 601-617.

Roberts, R.C., Peden, A.A., Buss, F., Bright, N.A., Latouche, M., Reilly, M.M., Kendrick-Jones, J., and Luzio, P.J. (2010) Mistargeting of SH3TC2 away from the recycling endosome causes Charcot-Marie-Tooth disease type 4C. Hum Mol Genet 19, 1009-1018.

Roh, M.H., Fan, S., Liu, C.J., and Margolis, B. (2003) The Crumbs-Pals1 complex participates in the establishment of polarity in mammalian epithelial cells. J Cell Sci 116, 2895-2906.

Roux, K.J., Kim, D.I., Raida, M., and Burke, B. (2012) A promiscuous biotin ligase fusion protein identifies proximal and interacting proteins in mammalian cells. J Cell Bio 196, 801-810. Rzomp, K.A., Scholtes, L.D., Briggs, B.J., Whittaker, G.R., and Scidmore, M.A. (2003) Rab GTPases are recruited to chlamydial inclusions in both a species-dependent and species- independent manner. Infect Immun 71, 5855-5870.

Sakaguchi, A., Sato, M., Sato, K., Gengyo-Ando, K., Yorimitsu, T., Nakai, J., Hara, T., Sato, K., and Sato, K. (2015) REI-1 is a Guanine Nucleotide Exchange Factor Regulating RAB-11 Localization and Function in C. elegans Embryos. Dev Cell 35, 211-221.

Sbodio, J.I. and Chi, N.W. (2002) Identification of a Tankyrase-binding Motif shared by IRAP, TAB182, and Human TRF1 but Not Mouse TRF1. J Biol Chem 277, 31887-31892.

Sbodio, J.I., Lodish, H.F., Chi, N.W. (2002) Tankyrase-2 oligomerizes with tankyrase 1 and binds to both TRF1 (telomere repeat-binding-factor 1) and IRAP (insulin-responsive 133

aminopeptidase). Biochem J 361, 451-459.

Scarborough, H.A., Helfrich, B.A., Casás-Selves, M., Schuller, A.G., Grosskurth, S.E., Kim, J., Tan, A.C., Chan, D.C., Zhang, Z., Zaberezhnyy, V., Bunn, P.A., and DeGregori, J. (2017) AZ1366: An Inhibitor of Tankyrase and the Canonical Wnt Pathway that Limits the Persistence of Non-Small Cell Lung Cancer Cells Following EGFR Inhibition. Clin Cancer Res 23, 1531- 1541.

Schlierf, B., Fey, G.H., Hauber, J., Hocke, G.M., and Rosorius, O. (2000) Rab11b is Essential for Recycling of Transferrin to the Plasma Membrane. Exp Cell Res 259, 257-265.

Schlüter, M.A., Pfarr, C.S., Pieczynski, J., Whiteman, E.L., Hurd, T.W., Fan, S., Liu, C.J. and Margolis, B. (2009) Trafficking of Crumbs3 during cytokinesis is crucial for lumen formation. Mol Biol Cell 20, 4652-4663.

Scholz, C.C., Rodriguez, J., Pickel, C., Burr, S., Fabrizio, J., Nolan, K.A., Spielmann, P., Cavadas, M.A.S., Crifo, B., Halligan, D.N., Nathan, J.A., Peet, D.J., Wenger, R.H., Kriegsheim, A.V., Cummins, E.P., and Taylor, C.T. (2016) FIH regulates Cellular Metabolism through Hydroxylation of the Deubiquitinase OTUB1. PLoS Biol 14: e1002347.

Seimiya, H., and Smith, S. (2002) The telomeric poly(ADP-ribose) polymerase, tankyrase 1, contains multiple binding sites for telomeric repeat binding factor 1 (TRF1) and a novel acceptor, 182-kDa tankyrase-binding protein (TAB182). J Biol Chem 277, 14116-14126.

Seimiya, H., Muramatsu, Y., Smith, S., and Tsuruo, T. (2004) Functional Subdomain in the Ankyrin Domain of Tankyrase 1 Required for Poly(ADP-Ribosyl)ation of TRF1 and Telomere Elongation. Mol Cell Biol 24, 1944-1955.

Shannon, P., Markiel, A., Ozier, O., Baliga, N.S., Wang, J.T., Ramage, D., Amin, N., Schwikowski, B., and Ideker, T. (2003) Cytoscape: a software environment for integrated models of bimolecular interaction networks. Genome Res 13, 2498-2504.

Shirane, M., and Nakayama, K.I. (2006) Protrudin Induces Neurite Formation by Directional Membrane Trafficking. Science 314, 818-821.

Sim, J., Cowburn, A.S., Palazon, A., Madhu, B., Tyrakis, P.A., Macías, D., Bargiela, D.M., Pietsch, S., Gralla, M., Evans, C.E., Kittipassorn, T., Chey, Y.C.J., Branco, C.M., Rundqvist, H., Peet, D.J., and Johnson, R.S. (2018) The Factor Inhibiting HIF Asparaginyl Hydroxylase Regulates Oxidative Metabolism and Accelerates Metabolic Adaptation to Hypoxia. Cell Metab 27, 898-913.

Smith, S., Giriat, L., Schmitt, A., and de Lange, T. (1998) Tankyrase, a poly(ADP-ribose) polymerase at human telomeres. Science 282, 1484-1487.

Snyder, K.A., Hughes, M.R., Hedberg, B., Brandon, J., Hernaez, D.C., Bergqvist, P., Cruz, F., Po, K., Graves, M.L., Turvey, M.E., Nielsen, J.S., Wilkins, J.A., McColl, S.R., Babcook, J.S., Roskelley, C.D., and McNagny, K.M. (2015) Podocalyxin enhances breast tumor growth and

134

metastasis and is a target for monoclonal antibody therapy. Breast Cancer Res 17, doi: 10.1186/s13058-015-0562-7.

Sodero, A.O., Rodriguez-Silva, M., Salio, C., Sassoè-Pognetto, M., and Chambers, J.W. (2017) Sab is differentially expressed in the brain and affects neuronal activity. Brain Res 1670, 76-85.

Somasiri, A., Nielsen, J.S., Makretsov, N., McCoy, M.L., Prentice, L., Gilks, C.B., Chia, S.K., Gelmon, K.A., Kershaw, D.B., Huntsman, D.G., McNagny, K.M., and Roskelley, C.D. (2004) Overexpression of the anti-adhesin podocalyxin is an independent predictor of breast cancer progression. Cancer Res 64, 5068-5073.

Stratford, E.W., Daffinrud, J., Munthe, E., Castro, R., Waaler,J., Krauss, S., and Myklebost, O. (2014) The tankyrase-specfic inhibitor JW74 affects cell cycle progression and induces apoptosis and differentiation in osteosarcoma cell lines. Cancer Med 3, 36-46.

Stolze, I.P., Tian, Y.M., Appelhoff, R.J., Turley, H., Wykoff, C.C., Gleadle, J.M., and Ratcliffe, P.J. (2004) Genetic Analysis of the Role of the Asparaginyl Hydroxylase Factor Inhibiting Hypoxia-inducible Factor (HIF) in Regulating HIF Transcriptional Target Genes. J Biol Chem 279, 42719-42725.

Sugimura, T., Fujimura, S., Hasegawa, S., and Kawamura, Y. (1967) Polymerization of the adenosine 5’-diphosphate ribose moiety of NAD by rat liver nuclear enzyme. Biochim Biophys Acta 138, 438-441.

Tian, Y.M., Yeoh, K.K., Lee, M.K., Eriksson, T., Kessler, B.M., Kramer, H.B., Edelmann, M.J., Willam, C., Pugh, C.W., Schofield, C.J., Ratcliffe, P.J. (2011) Differential sensitivity of hypoxia inducible factor hydroxylation sites to hypoxia and hydroxylase inhibitors. J Biol Chem 286, 13041-13051.

Tripathi, E., and Smith, S. (2017) Cell cycle-regulated ubiquitination of tankyrase 1 by RNF8 and ABRO1/BRCC36 controls the timing of sister telomere resolution. EMBO J 36, 503-519.

Ueki, Y., Tiziani, V., Santanna, C., Fukai, N., Maulik, C., Garfinkle, J., Ninomiya, C., doAmaral, C., Peters, H., Habal, M., Rhee-Morris, L., Doss, J.B., Kreiborg, S., Olsen, B.R., and Reichenberger, E. (2001) Mutations in the gene encoding c-Abl-binding protein SH3BP2 cause cherubism. Nat Genet 28, 125-126.

Ullrich, O., Reinsch, S., Urbé, S., Zerial, M., and Parton, R.G. (1996) Rab11 Regulates Recycling through the Pericentriolar Recycling Endosome. J Cell Biol 135, 913-924.

Wang, H., Lu, B., Castillo, J., Zhang, Y., Yang, Z., McAllister, G., Lindeman, A., Reeces-Hoyes, J., Tallarico, J., Russ, C., Hoffman, G., Xu, W., Schirle, M., and Cong, F. (2016) Tankyrase inhibitor sensitizes lung cancer cells to endothelial growth factor receptor inhibition via stabilizing angiomotins and inhibiting YAP signaling. J Biol Chem 291, 15256-15266.

135

Wang, W., Li, N., Li, X., Tran, M.K., Han, X., and Chen, J. (2015) Tankyrase inhibitors target YAP by stabilizing Angiomotin family proteins. Cell Rep 13, 524-532.

Wang, Y., Dawson, V.L., and Dawson, T.M. (2009) Poly(ADP-ribose) signals to mitochondrial AIF: a key event in parthanatos. Exp Neurol 218, 193-202.

Wang, Z., Michaud, G.E., Cheng, Z., Zhang, Y., Hinds, T.R., Fan, E., Cong, F., and Xu, W. (2012) Recognition of the iso-ADP-ribose moiety in poly(ADP-ribose) by WWE domains suggests a general mechanism for poly(ADP-ribosyl)ation-dependent ubiquitination. Genes Dev 26, 235-240.

Warde-Farley, D., Donaldson, S.L., Comes, O., Zuberi, K., Badrawi, R., Chao, P., Franz, M., Grouios, C., Kazi, F., Lopes, C.T., Maitland, A., Mostafavi, S., Montojo, J., Shao, Q., Wright, G., Bader, G.D., and Morris, Q. (2010) The GeneMANIA prediction server: biological network integration for gene prioritization and predicting gene function. Nucleic Acids Res 38, W214- 220.

Wells, C.D., Fawcett, J.P., Traweger, A., Yamanaka, Y., Goudreault, M., Elder, K., Kulkarni, S., Gish, G., Virag, C., Lim, C., Colwill, K., Starostine, A., Metalnikov, P., and Pawson, T. (2006) A Rich1/Amot complex regulates the Cdc42 GTPase and apical-polarity proteins in epithelial cells. Cell 125, 535-548.

Welsh, G.I., Leney, S.E., Lloyd-Lewis, B., Wherlock, M., Lindsay, A.J., McCaffrey, M.W., and Tavare, M. (2007) Rip11 is a Rab11- and AS160-RabGAP-binding protein required for insulin stimulated glucose uptake in adipocytes. J Cell Sci 120, 4197-4208.

Welz, T., Wellbourne-Wood, J., and Kerkhoff, E. (2014) Orchestration of cell surface proteins by Rab11. Trends Cell Biol 24, 407-415.

Westlake, C.J., Baye, L.M., Nachury, M.V., Wright, K.J., Ervin, K.E., Phu, L., Chaluoni, C., Beck, J.S., Kirkpatrick, D.S., Slusarski, D.C., Sheffield, V.C., Scheller, R.H., and Jackson, P.K. (2011) Primary cilia membrane assembly is initiated by Rab11 and transport protein particle II (TRAPPII) complex-dependent trafficking of Rabin8 to the centrosome. Proc Natl Acad Sci USA 108, 2759-2764.

Wilson, G.M., Fielding, A.B., Simon, G.C., Yu, X., Andrews, P.D., Hames, R.S., Frey, A.M., Peden, A.A., Gould, G.W., and Prekeris, R. (2005) The FIP3-Rab11 Protein Complex Regulates Recycling Endosome Targeting to the Cleavage Furrow during Late Cytokinesis. Mol Biol Cell 16, 849-860.

Wiltshire, C., Matsushita, M., Tsukada, S., Gillespie, D.A.F., and May, G.H.W. (2002) A new c- Jun N-terminal kinase (JNK)-interacting protein, SAB (SH3BP5), associates with the mitochondria. Biochem J 367, 577-585.

Win, S., Than, T.A., Han, D., Petrovic, L.M., and Kaplowitz, N. (2011) c-Jun-N-terminal Kinase (JNK)-dependent Acute Liver Injury from Acetominophen or Tumor Necrosis Factor (TNF) Requires Mitochondrial Sab Protein Expression in Mice. J Biol Chem 286, 35071-35078. 136

Win, S., Than, T.A., Fernandez-Checa, J.C., and Kaplowitz, N. (2014) JNK interaction with Sab mediates ER stress induced inhibition of mitochondrial respiration and cell death. Cell Death Dis 5, e989; doi:10.1038/cddis.2013.522.

Win, S., Than, T.A., Min, R.W., Aghajan, M., and Kaplowitz, N. (2016) c-Jun N-terminal kinase mediates mouse liver injury through a novel SAB (SH3BP5)-dependent pathway leading to inactivation of intramitochondrial Src. Hepatology 63, 1987-2003.

Xiong, B., Bayat, V., Jaiswal, M., Zhang, K., Sandoval, H., Charng, W.L., Li, T., David, G., Duraine, L., Lin, Y.Q., Neely, G.G., Yamamoto, S., and Bellen, H.J. (2012) Crag is a GEF for Rab11 required for rhodopsin trafficking and maintenance of adult photoreceptor cells. PLoS Biol 10 e1001438. doi: 10.1037/journal.pbio.1001438.

Xu, D., Liu, J., Fu, T., Shan, B., Qian, L., Pan, L., and Yuan, J. (2017) USP25 regulates WNT signaling by controlling the stability of tankyrases. Genes Dev 31, 1024-1035.

Xu, Y., Anderson, D.E., and Ye, Y. (2016) The HECT domain ubiquitin ligase HUWE1 targets unassembled soluble proteins for degradation. Cell Discov 2, 16040. eCollection 2016.

Yamadori, T., Baba, Y., Matsushita, Y., Hashimoto, S., Kurosaki, M., Kurosaki, T., Kishimoto, T., and Tsukuda, S. (1999) Bruton’s tyrosine kinase activity is negatively regulated by Sab, the Btk-SH3 domain-binding protein. Proc Natl Acad Sci USA. 96, 6341-6346.

Yang, M., Chowdhury, R., Ge, W., Hamed, R.B., McDonough, M.A., Claridge, T.D., Kessler, B.M., Cockman, M.E., Ratcliffe, P.J., and Schofield, C.J. (2011) Factor-inhibiting hypoxia- inducible factor (FIH) catalyses the post-translational hydroxylation of histidinyl residues within ankyrin repeat domains. FEBS J 278, 1086-1097.

Yeh, T.Y., Sbodio, J.I., and Chi, N.W. (2006) Mitotic phosphorylation of tankyrase, a PARP that promotes spindle assembly, by GSK3. Biochem Biophys Res Commun 350, 574-79.

Yeh, T.Y., Sbodio, J.I., Tsun, Z.Y., Luo, B., and Chi, N.W. (2007) Insulin-stimulated exocytosis of GLUT4 is enhanced by IRAP and its partner tankyrase. Biochem J 402, 279-290.

Yeh, T.Y., Beiswenger, K.K., Li, P., Bolin, K.E., Lee, R.M., Tsao, T.S., Murphy, A.N., Hevener, A.L., and Chi, N.W. (2015) Hypermetabolism, hyperphagia, and reduced adiposity in tankyrase- deficient mice. Diabetes 58, 2476-2485.

Zhang, N., Fu, Z., Linke, S., Chicher, J., Gorman, J.J., Visk, D., Haddad, G.G., Poellinger, L., Peet, D.J., Powell, F., Johnson, R.S. (2010) The asparaginyl hydroxylase factor inhibiting HIF- 1alpha is an essential regulator of metabolism. Cell Metab 11, 364-378.

Zhang, Y., Liu, S., Mickanin, C., Feng, Y., Charlat, O., Michaud, G.A., Schirle,M., Shi, X., Hild, M., Bauer, A., et al. (2011) RNF146 is a poly(ADP-ribose)-directed E3 ligase that regulates axin degradation and Wnt signalling. Nat Cell Biol 13, 623–629.

137

Zhao, B., Li, L., Tumaneng, K., Wang, C.Y., and Guan, K.L. (2010) A coordinated phosphorylation by Lats and CK1 regulates YAP stability through SCF(beta-TrCP). Genes Dev 24, 72-85.

Zhao, X., Heng, J.I., Guardavaccaro, D., Jiang, R., Pagano, M., Guillemot, F., Iavarone, A., and Lasorella, A. (2009) The HECT-domain ubiquitin ligase Huwe1 controls neural differentiation and proliferation by destabilizing the N-Myc oncoprotein. Nat Cell Biol 10, 643-653.

Zhong, L., Yeh, T.Y., Hao, J., Pourtabatabaei, N., Mahata, S.K., Shao, J., Chessler, S.D., and Chi, N.W. (2015) Nutritional energy stimulates NAD+ production to promote tankyrase- mediated PARsylation in insulinoma cells. PLoS One 10:e0122948 doi: 10.1371/journal.pone.0122948.

Zhong, L., Ding, Y., Bandyopadhyay, G., Waaler, J., Börgeson, E., Smith, S., Zhang, M., Phillips, S.A., Mahooti, S., Mahata, S.K., Shao, J., Krauss, S., and Chi, N.W. (2016) The PARsylation activity of tankyrase in adipose tissue modulates systemic glucose metabolism in mice. Diabetologia 59, 582-591.

Zhong, Q., Gao, W., Du, F., and Wang, X. (2005) Mule/ARF-BP1, a BH3-only E3 ubiquitin ligase, catalyzes the polyubiquitination of Mcl-1 and regulates apoptosis. Cell 121, 1085-1095.

138