UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The FUS-DDIT3 interactome in myxoid liposarcoma Yu, Jamie Simin Emma 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2019_february_yu_jamie.pdf [ 22.47MB ]
Metadata
JSON: 24-1.0375714.json
JSON-LD: 24-1.0375714-ld.json
RDF/XML (Pretty): 24-1.0375714-rdf.xml
RDF/JSON: 24-1.0375714-rdf.json
Turtle: 24-1.0375714-turtle.txt
N-Triples: 24-1.0375714-rdf-ntriples.txt
Original Record: 24-1.0375714-source.json
Full Text
24-1.0375714-fulltext.txt
Citation
24-1.0375714.ris

Full Text

 THE FUS-DDIT3 INTERACTOME IN MYXOID LIPOSARCOMA by  Jamie Simin Emma Yu  B.Sc. Honours, The National University of Singapore, 2005 M.Sc., The University of British Columbia, 2012  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Interdisciplinary Oncology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  December 2018  © Jamie Simin Emma Yu, 2018  ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled: The FUS-DDIT3 interactome in myxoid liposarcoma  submitted by Jamie Simin Emma Yu in partial fulfillment of the requirements for the degree of Doctor of Philosophy in The Faculty of Graduate and Postdoctoral Studies  Examining Committee: Torsten Nielsen Supervisor  Michael Cox Supervisory Committee Member   Supervisory Committee Member Cathie Garnis University Examiner Calvin Roskelley University Examiner  Additional Supervisory Committee Members: Gregg Morin Supervisory Committee Member T. Michael Underhill Supervisory Committee Member  iii  Abstract  Myxoid liposarcoma is a malignant fatty tumour that develops in the deep soft tissue. While local control rates are good, current chemotherapy options remain ineffective against metastatic disease. Myxoid liposarcoma is characterized by a balanced translocation involving FUS (12q13) and DDIT3 (16p11). The resulting FUS-DDIT3 oncoprotein is proposed to function as an aberrant transcription factor but its exact mechanism of action has remained unclear, rendering it difficult to formulate rational strategies for targeted therapy. A current gap in knowledge behind the oncogenic functions of FUS-DDIT3 is the lack of comprehensive data on its interactome, which could identify the oncoprotein's key partners. This study utilized immunoprecipitation-mass spectrometry to identify the FUS-DDIT3 interactome in whole cell lysates of myxoid liposarcoma cells, and results showed an enrichment of RNA processing proteins. RNA-seq analysis was performed on myxoid liposarcoma cells after FUS-DDIT3 knockdown to look for changes in the alternative splicing profile, but no evidence of such changes was seen. Further TMT-labeled immunoprecipitation-mass spectrometry analyses in nuclear lysates of myxoid liposarcoma cells showed that members of several chromatin regulatory complexes were present in the FUS-DDIT3 interactome. These complexes included the NuRD, SWI/SNF, PRC1, PRC2 and MLL1 COMPASS-like complexes. Co-immunoprecipitation experiments validated the association of FUS-DDIT3 with BRG1/SMARCA4, BAF155/SMARCC1, BAF57/SMARCE1, HDAC2, KDM1A, and MTA1. Knockdown of FUS-DDIT3 also reduced H3K27ac levels at the promoter of a gene target, PTX3. Other sarcoma fusion oncoproteins have been reported to interact with chromatin regulators and affect histone modifications or chromatin remodeling as one of their oncogenic mechanisms. Data from this study suggests that FUS-DDIT3 may utilize a iv  similar epigenetic mechanism of action, providing potential candidates for targeted therapy as epigenetic aberrations are potentially reversible by existing and emerging epigenetic drugs.  v  Lay Summary  Myxoid liposarcoma is an intermediate to high-grade fatty tumour that develops in deep soft tissue. Symptoms do not typically manifest until the tumour has grown and presses on surrounding tissue, at which time the tumour is often fairly advanced and requires systemic treatment. Myxoid liposarcoma is known to be caused by a mutation named FUS-DDIT3, but treatment that specifically targets the resulting mutant protein is currently unavailable, and existing chemotherapy is ineffective in higher grade tumours. Drugs that inhibit FUS-DDIT3's function should provide better treatment options but have been technologically difficult to develop. FUS-DDIT3 functions through co-operating with other proteins, some of which may already have existing drugs that inhibit them, and as a result, disrupt FUS-DDIT3's function. This study identifies the complement of proteins surrounding FUS-DDIT3 and forms the basis for future work to select a FUS-DDIT3 protein partner for drug inhibition. vi  Preface Histology images from Chapter 1 were stained by Christine Chow and selected by Dr. Dongxia Gao for use in this thesis. Chapter 2 is based on work conducted in collaboration with Dr. Gregg Morin at the Genome Sciences Centre. Dr. Torsten Nielsen identified the research program. My contributions include the design, performance and analysis of majority of the experiments, and writing of the manuscript. Mass spectrometry analyses were carried out by Dr. Christopher Hughes with Gregg Morin. Dr. Neal Poulin assisted with bioinformatics analysis and selection of protein candidates for validation. Chapter 3 is based on work conducted in collaboration with Dr. Michael Underhill at the Biomedical Research Centre. The RNA-seq experiment was designed by myself with guidance from Dr. Michael Underhill. My contributions to this chapter include the design, performance and analysis of majority of the experiments, and writing of the manuscript. RNA-seq analysis was carried out by Ryan Vander Werff with Michael Underhill. Dr. Saeed Saberi performed the MISO analysis. Neal Poulin assisted with bioinformatics analysis and selection of candidates for gene expression validation. Chapter 4 is based on work conducted in collaboration with Gregg Morin. My contributions to this chapter include the design, performance and analysis of majority of the experiments, and writing of the manuscript. Malin Lindén with Dr. Pierre Aman at the University of Gothenburg and Gregg Morin assisted with design of the nuclear lysis experiment. TMT labeling and mass spectrometry analyses were carried out by Shane Colborne with Gregg Morin. Experiments involving patient tissues were carried out using protocols approved by the University of British Columbia Clinical Research Ethics Board (REB# H08-01411). vii  The contents of Chapters 2 and 4 are modified from the manuscript in preparation “J.S.E. Yu, A. Goytain, S. Colborne, C.S. Hughes, G.B. Morin, T.O. Nielsen. 2019. “The FUS-DDIT3 interactome in myxoid liposarcoma”.  viii  Table of Contents  Abstract ......................................................................................................................................... iii Lay Summary ................................................................................................................................ v Preface ........................................................................................................................................... vi Table of Contents ....................................................................................................................... viii List of Tables ............................................................................................................................... xv List of Figures ............................................................................................................................. xvi List of Abbreviations ................................................................................................................. xix Acknowledgements ................................................................................................................... xxii Dedication ................................................................................................................................. xxiii Chapter 1: Introduction ............................................................................................................... 1 1.1 Soft tissue sarcomas ........................................................................................................ 1 1.2 Translocation-associated sarcomas ................................................................................. 2 1.3 Clinical and histopathological features of myxoid liposarcoma ..................................... 5 1.4 Fusion of FUS to DDIT3 ................................................................................................. 7 1.4.1 t(12;16)(q13;p11) translocation .............................................................................. 7 1.4.2 FUS ......................................................................................................................... 8 1.4.3 DDIT3 ................................................................................................................... 10 1.4.4 FUS-DDIT3 structure and regulation ................................................................... 12 1.5 Molecular features of myxoid liposarcoma .................................................................. 13 1.5.1 Functions of FUS-DDIT3 ..................................................................................... 13 1.5.2 Secondary mutations and pathway activations ..................................................... 15 1.6 Targeted therapy for myxoid liposarcoma .................................................................... 18 ix  1.6.1 PPARγ agonists ..................................................................................................... 18 1.6.2 Pazopanib .............................................................................................................. 19 1.6.3 Olaratumab ............................................................................................................ 19 1.6.4 HDAC inhibitors ................................................................................................... 20 1.6.5 NY-ESO-1 immunotherapy .................................................................................. 20 1.7 Experimental models of myxoid liposarcoma .............................................................. 21 1.7.1 Cell lines and primary cultures ............................................................................. 22 1.7.2 Mouse models ....................................................................................................... 23 1.8 Thesis objective and chapter overview ......................................................................... 24 Chapter 2: Protein interaction network of FUS-DDIT3 ......................................................... 32 2.1 Introduction ................................................................................................................... 32 2.1.1 Targeting oncogenic transcription factors ............................................................ 32 2.1.2 FUS-DDIT3 interactors ........................................................................................ 36 2.1.3 Affinity purification followed by mass spectrometry ........................................... 38 2.2 Materials and methods .................................................................................................. 39 2.2.1 Mammalian cell culture and chemicals ................................................................. 39 2.2.2 Immunoprecipitation ............................................................................................. 39 2.2.3 Protein clean up, digestion, mass spectrometry and analysis ............................... 40 2.2.4 RNA interference .................................................................................................. 42 2.2.5 Western blots ........................................................................................................ 43 2.2.6 Proximity ligation assay ........................................................................................ 43 2.2.7 Cell viability assay ................................................................................................ 44 2.3 Results ........................................................................................................................... 44 2.3.1 Immunoprecipitation-mass spectrometry discovery of FUS-DDIT3 interactome 44 x  2.3.2 The FUS-DDIT3 interactome is enriched for RNA-processing proteins ............. 46 2.3.3 Validation of FUS-DDIT3 association with selected candidate proteins ............. 47 2.3.4 MAP4K4 knockdown or inhibition does not affect cell viability ......................... 48 2.3.5 Assessing MAP4K4 downstream activation of AMPK in myxoid liposarcoma .. 50 2.4 Discussion ..................................................................................................................... 51 2.4.1 Summary ............................................................................................................... 51 2.4.2 Insights from the FUS-DDIT3 interactome .......................................................... 52 2.4.3 Functional validation of FUS-DDIT3 interactors ................................................. 55 2.4.4 Future work ........................................................................................................... 57 Chapter 3: Assessing involvement of FUS-DDIT3 in alternative splicing ............................. 82 3.1 Introduction ................................................................................................................... 82 3.1.1 RNA processing .................................................................................................... 82 3.1.1.1 5’ capping .......................................................................................................... 82 3.1.1.2 Splicing ............................................................................................................. 83 3.1.1.3 3’ polyadenylation ............................................................................................ 85 3.1.2 RNA sequencing ................................................................................................... 87 3.1.3 Rationale and aims of Chapter 3 ........................................................................... 88 3.2 Materials and methods .................................................................................................. 90 3.2.1 Mammalian cell culture ........................................................................................ 90 3.2.2 RNA interference .................................................................................................. 90 3.2.3 RNA-seq and alternative splicing analysis ........................................................... 91 3.2.4 Quantitative reverse transcription polymerase chain reaction .............................. 92 3.2.5 Western blots ........................................................................................................ 92 3.2.6 Proliferation assay ................................................................................................. 93 xi  3.3 Results ........................................................................................................................... 94 3.3.1 RNA-seq analysis of myxoid liposarcoma cell line 402-91 after FUS-DDIT3 knockdown ............................................................................................................................ 94 3.3.2 Validation of RNA-seq gene expression results by qPCR .................................... 95 3.3.3 Alternative splicing analysis after FUS-DDIT3 knockdown ................................ 95 3.3.4 FUS-DDIT3 does not affect alternative last exon usage ...................................... 97 3.4 Discussion ..................................................................................................................... 99 3.4.1 Summary ............................................................................................................... 99 3.4.2 Lack of evidence of alternative splicing interference ........................................... 99 3.4.3 Conclusion and future work ................................................................................ 102 Chapter 4: Association of FUS-DDIT3 with chromatin regulatory complexes .................. 117 4.1 Introduction ................................................................................................................. 117 4.1.1 Histone modifications ......................................................................................... 117 4.1.2 Chromatin regulatory complexes ........................................................................ 121 4.1.3 Epigenetic deregulation in sarcoma .................................................................... 124 4.1.4 Rationale and aim of Chapter 4 .......................................................................... 125 4.2 Materials and methods ................................................................................................ 126 4.2.1 Mammalian cell culture ...................................................................................... 126 4.2.2 Patient samples .................................................................................................... 127 4.2.3 Transfection of siRNA and FUS-DDIT3 vector ................................................. 127 4.2.4 Protein extraction ................................................................................................ 128 4.2.5 Immunoprecipitation ........................................................................................... 129 4.2.6 Western blots ...................................................................................................... 130 4.2.7 Mass spectrometry analysis (Velos) ................................................................... 130 xii  4.2.8 TMT labelling ..................................................................................................... 132 4.2.9 Mass spectrometry analysis (Fusion) .................................................................. 132 4.2.10 Protein enrichment score calculation .................................................................. 133 4.2.11 Crude histone extraction ..................................................................................... 134 4.2.12 Chromatin immunoprecipitation ......................................................................... 135 4.3 Results ......................................................................................................................... 137 4.3.1 TMT-labeled IP-MS identification of a FUS-DDIT3 interactome following nuclear extraction ................................................................................................................ 137 4.3.2 FUS-DDIT3 interactome includes multiple chromatin regulatory complexes ... 139 4.3.3 Validation of FUS-DDIT3’s association with chromatin regulators .................. 140 4.3.4 Global H3K27 acetylation or trimethylation are not affected by FUS-DDIT3 .. 141 4.3.5 FUS-DDIT3 knockdown in 402-901 reduces H3K27ac and increases H3K27me3 at the PTX3 promoter .......................................................................................................... 142 4.4 Discussion ................................................................................................................... 143 Chapter 5: Discussion, Conclusion and Future Directions ................................................... 163 5.1 Summary ..................................................................................................................... 163 5.2 Insights from the FUS-DDIT3 interactome ................................................................ 163 5.3 Translocation-associated sarcomas as epigenetic diseases ......................................... 167 5.4 Future directions ......................................................................................................... 169 References .................................................................................................................................. 173 Appendices ................................................................................................................................. 203 Appendix A ............................................................................................................................. 203 A.1 C/EBPβ is expressed in myxoid liposarcoma cell lines ...................................... 203 xiii  A.2 List of putative FUS-DDIT3 interactors that were previously reported as DDIT3 interactors (from the BioGRID database), or FUS interactors (from published FUS interactomes) ....................................................................................................................... 204 A.3 Overlap of proteins detected in the FUS, DDIT3 and FUS-DDIT3 interactomes 207 A.4 Assessment of MAP4K4 inibition in myxoid liposarcoma cells DL-221 after GNE-495 treatment ............................................................................................................. 208 A.5 Overlap of common proteins identified in previously published FUS interactomes 209 A.6 Interpro analysis of FUS-DDIT3 (Type 1) amino acid sequence ....................... 210 A.7 MAP4K4 is overexpressed in myxoid liposarcoma ............................................ 211 A.8 PPARG expression is high in myxoid liposarcomas ........................................... 212 Appendix B ............................................................................................................................. 213 B.1 Proteins in the FUS-DDIT3 interactome that are present in the Spliceosome Database 213 B.2 List of qPCR primers used in Chapter 3 ............................................................. 216 B.3 Number and percentage of global alternative splicing events ............................ 217 B.4 Number and percentage of types of differential alternative splicing events between control and DDIT3 knockdown ............................................................................ 218 B.5 All differential alternative last exon events between control and FUS-DDIT3 knockdown at 24 hr or 48 hr ............................................................................................... 219 Appendix C ............................................................................................................................. 225 C.1 List of qPCR primers used in Chapter 4 ............................................................. 225 C.2 FUS-DDIT3 protein degrades over time during immunoprecipitation ............... 226 xiv  C.3 DDIT3 peptides identified from the quality control analysis of unlabeled peptide fractions with the Velos Orbitrap mass spectrometer ......................................................... 227 C.4 List of 174 FUS-DDIT3-interacting proteins identified in nuclear lysates of 402-91 myxoid liposarcoma cells .............................................................................................. 228 C.5 CORUM protein complexes found in FUS-DDIT3 nuclear interactome after applying PES z-score cutoff ................................................................................................ 238 C.6 Gene ontology cellular component classification of full list of FUS-DDIT3 nuclear interactome ............................................................................................................. 239  xv  List of Tables Table 1.1 Myxoid liposarcoma cell line characteristics. ............................................................... 31 Table 2.1 Selection of IP-MS-identified proteins for further analysis based on their enrichment in at least 2 cell lines DDIT3 immunoprecipitation .......................................................................... 75 Table 2.2 List of FUS-DDIT3-interacting proteins identified by IP-MS in at least two myxoid liposarcoma cell lines (402-91, 1765-92, DL-221) ....................................................................... 76 Table 2.3 Gene ontology classification of FUS-DDIT3 interactome for cellular component ...... 79 Table 2.4 Gene ontology classification of FUS-DDIT3 interactome for biological processes .... 80 Table 2.5 Gene ontology classification of FUS-DDIT3 interactome for molecular function ...... 81 Table 3.1 RNA-seq read mapping statistics ................................................................................ 112 Table 3.2 Differential expression of DDIT3 transcripts ............................................................. 113 Table 3.3 RNA-seq data for transcripts selected for qPCR validation ....................................... 114 Table 3.4 Common differential alternative splicing events between 24 hr and 48 hr FUS-DDIT3 knockdown .................................................................................................................................. 115 Table 3.5 Differential alternative last exon events selected for qPCR validation ...................... 116  xvi  List of Figures Figure 1.1 Recurrent sarcoma fusion genes involving the FET (FUS, EWSR1, TAF15) family genes ............................................................................................................................................. 27 Figure 1.2 Histology of myxoid liposarcoma ............................................................................... 28 Figure 1.3 Fusion variants of FUS-DDIT3 ................................................................................... 29 Figure 1.4 Structure and domains of FUS and DDIT3 ................................................................. 30 Figure 2.1 Wild type DDIT3 is not expressed in myxoid liposarcoma cell lines ......................... 59 Figure 2.2 Experimental outline of IP-MS experiment and confirmation of successful FUS-DDIT3 IP ...................................................................................................................................... 60 Figure 2.3 Amino acid sequences of FUS and DDIT3 tryptic peptides detected from mass spectrometry .................................................................................................................................. 62 Figure 2.4 Partial overlap in potential FUS-DDIT3 interactors identified from three myxoid liposarcoma cell line ..................................................................................................................... 63 Figure 2.5: FUS-DDIT3 interactome network .............................................................................. 64 Figure 2.6: Enriched protein complexes and protein classes in FUS-DDIT3 interactome ........... 65 Figure 2.7 Reciprocal co-immunoprecipitation validates FUS-DDIT3 association with NONO, PSPC1, SFPQ ................................................................................................................................ 66 Figure 2.8 Proximity ligation assay validations association of  FUS-DDIT3 with Drosophila behavior/human splicing proteins NONO, PSPC1 and SFPQ ...................................................... 67 Figure 2.9 Reciprocal co-immunoprecipitation validates interaction of selected IP-MS-identified proteins with FUS-DDIT3 in 402-91 myxoid liposarcoma cells .................................................. 68 Figure 2.10 Reciprocal co-immunoprecipitation validates interaction of selected IP-MS-identified proteins with FUS-DDIT3 in 1765-92 myxoid liposarcoma cells ............................... 69 xvii  Figure 2.11 Reciprocal co-immunoprecipitation validates interaction of selected IP-MS-identified proteins with FUS-DDIT3 in DL-221 myxoid liposarcoma cells ................................ 70 Figure 2.12 Proximity ligation assay validations association of FUS-DDIT3 with MAP4K4 ..... 71 Figure 2.13 MAP4K4 inhibition does not affect myxoid liposarcoma cell viability .................... 72 Figure 2.14 siRNA-mediated MAP4K4 knockdown does not affect viability of myxoid liposarcoma cells ........................................................................................................................... 73 Figure 2.15 MAP4K4 knockdown does not affect oligomycin-induced AMPK activation ......... 74 Figure 3.1 Selection of FUS-DDIT3 knockdown duration for RNA-seq experiments ............... 104 Figure 3.2 Validation of differential gene expression by qPCR ................................................. 106 Figure 3.3 Global alternative splicing profile does not change after 24 hr and 48 hr FUS-DDIT3 knockdown in 402-91 ................................................................................................................. 107 Figure 3.4 Differential alternative splicing events after 24 hr and 48 hr FUS-DDIT3 knockdown in 402-91 ..................................................................................................................................... 108 Figure 3.5 Confirmation of FUS-DDIT3 reduction for alternative last exon validation experiment ..................................................................................................................................................... 109 Figure 3.6 qPCR does not validate alternative last exon usage in FAT1, FLT1 and PML ......... 111 Figure 4.1 Experimental outline for immunoprecipitation-mass spectrometry using TMT labeling ..................................................................................................................................................... 147 Figure 4.2 Scatterplot of FUS-DDIT3 nuclear interactome after selection cutoff ..................... 149 Figure 4.3 Overlapping FUS-DDIT3 interactors identified in whole cell lysate and nuclear fraction ........................................................................................................................................ 150 Figure 4.4 CORUM protein complexes enriched in FUS-DDIT3 nuclear interactome ............. 151 Figure 4.5 Chromatin regulators are present in the FUS-DDIT3 nuclear interactome ............... 153 Figure 4.6 Gene ontology classification of FUS-DDIT3 nuclear interactome ........................... 154 xviii  Figure 4.7 Co-immunoprecipitation validates FUS-DDIT3 interaction with HDAC2 ............... 155 Figure 4.8 Reciprocal co-immunoprecipitation validates FUS-DDIT3 interaction with KDM1A ..................................................................................................................................................... 156 Figure 4.9 Co-immunoprecipitation validates FUS-DDIT3 interaction with SWI/SNF components ................................................................................................................................. 157 Figure 4.10 Myxoid liposarcoma tumors do not show loss of H3K27 trimethylation ............... 158 Figure 4.11 FUS-DDIT3 knockdown or exogenous expression for 48 hr does not consistently alter H3K4me3, H3K27me1/3 and H3K27ac histone marks ...................................................... 160 Figure 4.12 FUS-DDIT3 knockdown for 3 or 7 days does not consistently alter H3K27 acetylation or trimethylation ....................................................................................................... 161 Figure 4.13 FUS-DDIT3 knockdown increases trimethylation and decreases acetylation of H3K27 at the PTX3 promoter ..................................................................................................... 162 Figure 5.1 Potential mechanism of action for how FUS-DDIT3 may deregulate the function of chromatin regulatory complexes ................................................................................................. 172  xix  List of Abbreviations AMPK AMP-activated protein kinase ATCC American type culture collection ATF Activating transcription factor ATRA All-trans retinoic acid BAF BRG1-Associated Factor BET Bromodomain and extra-terminal domain BF Bayes factor BRD1/2/3/4 Bromodomain protein 1/2/3/4 bZIP basic zipper C/EBP CCAAT-enhancer-binding protein CDK Cyclin dependent kinase CHD4 chromodomain helicase DNA binding protein 4 ChIP Chromatin immunoprecipitation ChIP-seq Chromatin immunoprecipitation-sequencing CHOP C/EBP homologous protein CORUM Comprehensive resource of mammalian protein complexes CPSF Cleavage and polyadenylation stimulatory factor CREBL2 cAMP response element (CRE)-binding protein-like-2 CV Coefficient of variation DDIT3 DNA damage inducible transcript 3 DNA Deoxyribonucleic acid EF-1α Elongation factor 1α EGFR Epidermal growth factor receptor EWSR1 Ewing sarcoma breakpoint region 1 EZH2 Enhancer-of-zest homolog 2 FISH Fluorescence in situ hybridization FLI1 Friend leukemia integration 1 FOXO1 Forkhead Box O1 FUS Fused in sarcoma HDAC Histone deacetylase hnRNP Heterogeneous nuclear ribonucleoproteins  IGF1R Insulin-like growth factor 1 receptor IL8 Interleukin 8 IP-MS immunoprecipitation-mass spectrometry KDM2B  Lysine (K)-specific demethylase 2B LSD1 Lysine demethylase 1 MAP4K4 Mitogen-Activated Protein Kinase Kinase Kinase Kinase 4 MDM2 Mouse double minute 2 homolog MISO Mixture of isoforms mRNA messenger RNA xx  mTOR Mammalian target of rapamycin NFκB  Nuclear factor-κB NONO Non-POU domain containing octamer binding  NuRD Nucleosome remodeling and deacetylase NY-ESO-1 New York esophageal squamous cell carcinoma-1 PARP1 Poly (ADP-ribose) polymerase 1 PAX3 Paired box gene 3 PcG Polycomb group PDGFR Platelet-derived growth factor receptor PES Protein enrichment score PHD plant homeodomain PI3K Phosphatidylinositol 3-kinase PML Promyelocytic leukemia PPARγ Peroxisome proliferator-activated receptor gamma  PRC Polycomb repressive complex PSPC1 Paraspeckle component 1 PTEN Phosphatase and tensin homolog deleted in chromosome 10 PTX3  Pentraxin 3 PY Proline-tyrosine RAR Retinoic acid receptor RET Rearranged during transfection RGG Arginine-glycine-glycine RNA Ribonucleic acid RNAPII RNA polymerase II RNA-seq RNA-sequencing RRM RNA recognition motif RT-PCR Reverse transcription polymerase chain reaction SFPQ Splicing factor proline and glutamine rich siRNA Small interfering RNA SMARC SWI/SNF-related matrix-associated actin-dependent regulator of chromatin snRNP U1 small nuclear ribonucleoprotein SS18 Synovial sarcoma 18 SSX Synovial sarcoma X SUZ12 Suppressor Of Zeste 12 SWI/SNF SWItch/Sucrose Non-Fermentable SYGQ Serine/tyrosine/glycine/glutamine TAF15 TBP Associated Factor 15 TERT  Telomerase reverse transcriptase TLE1 transducin-like enhancer of split 1 TMT Tandem mass tag TrxG Trithorax group xxi  UTR Untranslated region VEGFR Vascular endothelial growth factor WT1 Wilms' tumor 1 YB-1 Y-box-binding protein 1 ZnF Zinc finger     xxii  Acknowledgements I would like to thank Dr. Torsten Nielsen for the opportunity to work on this research project, and for his mentorship, guidance and unwavering support. I would further like to thank my supervisory committee members Drs. Michael Cox, Gregg Morin and Michael Underhill for being so generous with their help, advice and encouragement. My heartfelt gratitude goes out to our international collaborators for being such a wonderful team to work with and providing helpful discussions and suggestions – Drs. Pierre Åman, Hannah Beird, Judith Bovée, Marieke de Graaf, Alexander Lazar, Neeta Somaiah, and Per Wu. I would like to acknowledge generous funding support from the Shaun G Foundation for my studies. Further funding support for this research came from the Liddy Shriver Sarcoma Initiative, Candian Cancer Society and the Terry Fox Foundation. From our research group, I’d like to thank Angela Goytain and Jenny Wang for their technical support and assistance in my work, Dr. Neal Poulin for guidance on bioinformatics and pathway analyses, and Dr. Dongxia Gao and Christine Chow for support on immunohistochemical and scoring work. I’d also like to thank members of the Nielsen lab for providing a positive and supportive working environment – it is such a pleasure to show up for work with you. Finally, special thanks are owed to my family – my mom, husband and children – for their unconditional love and support throughout this journey.  xxiii  Dedication     To my family 1  Chapter 1: Introduction 1.1 Soft tissue sarcomas Soft tissue sarcomas are a heterogeneous group of malignant solid tumors that arise in connective tissues such as fat, muscle, tendon, fascia, blood vessels, peripheral nerves, and subcutaneous tissues. This group of neoplasms accounts for less than 1% of all new cancer diagnoses in the US, totaling an estimated 15,650 new cases with approximately 6,540 annual deaths (Siegel et al., 2017). The exact cause of sarcomagenesis is unknown, but sarcomas can arise from any part of the body.  Most are diagnosed in the extremities (40%), followed by visceral or retroperitoneal areas (38%), and rates of local recurrence are high (Brennan et al., 2014). The World Health Organization (WHO) has classified over 100 distinct histological subtypes displaying varying degrees of mesenchymal differentiation (Fletcher et al., 2013). This diversity poses a huge challenge for diagnosis that at times requires support from ancillary molecular testing, in specialty centers where the required equipment and expertise are available. The standard treatment for many primary soft tissue sarcomas is surgery with or without radiation therapy which can be curative for some patients. In advanced soft tissue sarcomas, conventional chemotherapy remains a mainstay despite the biological and histological heterogeneity of sarcomas, but confers little proven curative benefit with the exception of Ewing sarcoma, rhabdomyosarcoma, and osteosarcoma (Young et al., 2014). A Phase 2 trial for olaratumab, a human anti-platelet-derived growth factor receptor α monoclonal antibody, showed promising results in soft tissue sarcomas (Tap et al., 2016), but will require Phase 3 confirmation (Judson and van der Graaf, 2016), as past Phase 3 trials in sarcomas have not successfully confirmed previous promising Phase 2 results (Constantinidou and van der Graaf, 2017; Tap et al., 2017). 2  Accurate diagnosis is the cornerstone of therapy. Recent advances have improved our understanding of the underlying molecular characteristics of sarcomas, and offer the opportunity to refine the current classification system to take into account the spectrum of prognostic and therapeutic implications of the different subtypes (Schaefer and Fletcher, 2018). By morphology, karyotype, and expression profile, sarcomas have been broadly categorized into two groups: Molecularly defined sarcomas often carry pathognomonic translocations with low mutational burden and disease-specific RNA expression profiles, whereas pleomorphic sarcomas have highly rearranged karyotypes and expression profiles that do not correlate consistently with histologic categorizations based on microscopic phenotype (Mertens et al., 2010; Nielsen and West, 2010).  Recent sequencing studies have provided more detailed insights into the biology of different subtypes and reveal an intermediate group of sarcomas carrying recurrent events in the context of complex changes (Brohl et al., 2017; Cancer Genome Atlas Research Network, 2017; Chudasama et al., 2018; W. Lee et al., 2014; Lewin et al., 2018; M. Zhang et al., 2014). Additional consideration for molecular classification can lead to different, but better and clinically relevant therapeutic strategies for each group (Lim et al., 2015).  1.2 Translocation-associated sarcomas About one-third of sarcomas are characterized by recurrent chromosomal translocations that result in aberrant fusion transcription factors. Since the first fusions were identified in sarcoma in the early 1990s, over 140 fusions have been discovered (Mertens et al., 2016), and this is likely just the tip of the iceberg, as a recent large scale transcriptomic study on undiagnosed blue round cell tumors reported 29 novel fusions out of the 55 detected (Watson et al., 2018). The rarity of most of these fusion-associated entities means that many have not been 3  investigated in sufficient detail to understand their contributions to biology, diagnosis or treatment.  The exact mechanisms behind recurrent chromosomal translocations are unknown, but they do not appear to be random events. One contributing factor is spatial proximity between translocation partners, which correlates with translocation frequency. Such proximity is specific to cell type and cell division stage, but can also be influenced by hormones or induced through recruitment into transcription factories (Mani and Chinnaiyan, 2010). The resulting chimeric oncoproteins encoded by the fusion transcripts often appear to represent the main oncogenic driver in translocation-associated sarcomas, as they occur in a genetically quiescent background with relatively few mutations and copy number aberrations (Barretina et al., 2010; Brohl et al., 2014; Cancer Genome Atlas Research Network, 2017; Chalmers et al., 2017; Crompton et al., 2014; Joseph et al., 2014; Tirode et al., 2014; Vlenterie et al., 2015). Certain genes such as EWSR1 and FUS are promiscuous fusion partners in a variety of translocation-associated sarcomas. Both genes are highly conserved members of what has been termed the FET (FUS, EWSR1, TAF15) family of genes consisting of FUS (Fused in Sarcoma) / TLS (Translocated in Liposarcoma), EWSR1 (Ewing Sarcoma breakpoint Region 1) and TAF15 (TBP Associated Factor 15) / TAF2N (TATA-Binding Protein-Associated Factor 2N) (Schwartz et al., 2014). EWSR1 is the most common 5' partner, being involved in recurrent fusions from nine different sarcomas, followed by FUS in five entities (Mertens et al., 2016) (Figure 1.1). Their C-terminus partners, which make up the transcription factor component of the oncoprotein, show more specificity for sarcoma type, and determine tumour morphology (Zinszner et al., 1994). Some gene fusions, however, are observed in more than one sarcoma subtype, such as EWSR1-ATF1 in the very distinct entities of angiomatoid fibrous histiocytoma and clear cell 4  sarcoma, or FUS-CREBL2 in the possibly related entities of low-grade fibromyxoid sarcoma and sclerosing epithelioid fibrosarcoma. Against a background of few secondary mutations, the morphologic and clinical differences between different sarcomas driven by the same fusion genes suggest the importance of the different cellular context in which these fusions occur. Indeed, a mouse model for conditional EWS-ATF1 expression in different cell types and differentiation stages resulted in tumours of different tumour phenotypes (Straessler et al., 2013). The majority of genes involved in translocation-associated sarcoma fusions encode transcription factors and regulators, and result in the production of aberrant chimeric transcription factors that alter the developmental gene expression repertoire in permissive cells. There is increasing evidence that these fusion oncoproteins drive sarcomagenesis through altered epigenetic control of developmentally regulated genes. In Ewing sarcoma, EWS-FLI1 was shown in separate studies to mediate transcriptional changes in its gene targets through recruitment of two epigenetic regulatory complexes: the SWItch/Sucrose Non-Fermentable (SWI/SNF) chromatin remodeling complex (Boulay et al., 2017) and the nucleosome remodeling and deacetylase (NuRD) complex (Sankar et al., 2013). Similarly, in synovial sarcoma, three different mechanisms of action have been reported for SS18-SSX that involve redistribution of chromatin regulators. Firstly, SS18-SSX assembles an aberrant SWI/SNF complex lacking SMARCB1/BAF47, that is targeted to the locus of the pluripotency factor SOX2 (Kadoch and Crabtree, 2013). SWI/SNF is also mistargeted by SS18-SSX1 via interaction with the histone demethylase KDM2B and polycomb repressive complex 1.1 (PRC1.1), to activate expression of developmentally regulated genes that are otherwise targets of polycomb-mediated repression (Banito et al., 2018). Finally, SS18-SSX functions as a bridging factor to aberrantly recruit polycomb repressor components and HDAC1 to repress ATF2 target genes (Su et al., 2012). In rhabdomyosaroma, the PAX3-FOXO1 interactome is 5  enriched for protein subunits of chromatin remodelling complexes, with chromodomain helicase DNA binding protein 4 (CHD4) being a crucial coregulator of PAX3-FOXO1 activity (Böhm et al., 2016).  1.3 Clinical and histopathological features of myxoid liposarcoma Liposarcomas are adipocytic malignancies and constitute one of the most common types of soft tissue sarcoma. They are further categorized by the WHO into three types of liposarcomas with distinct histologies, molecular biology, patient prognosis and treatment regimen: (1) well and dedifferentiated liposarcoma, (2) pleomorphic liposarcoma, and (3) myxoid liposarcoma. Myxoid liposarcoma is a translocation-associated sarcoma that constitutes 5% of all soft tissue sarcomas in adults, with peak incidence in young adults between 30-50 years old.  Myxoid liposarcoma tumors are usually slow growing, and mainly present in the deep soft tissues of the lower extremities, with two-thirds of cases occurring in the thigh (Fletcher et al., 2013), although the retroperitoneum is also a common site. Distant metastases develop in 13-15% of cases (Dürr et al., 2018; Hoffman et al., 2013), and are more challenging to detect with conventional imaging strategies (i.e. staging chest CT) due to their predominantly extrapulmonary sites (often requiring PET scans for detection). Most metastases occur in the abdominal cavity (49% of case), followed by the skeleton and spine (23%) (Hoffman et al., 2013), with subcutaneous metastases also occurring more frequently than in other types of sarcoma.   The histopathology of myxoid liposarcoma includes a mixture of uniform round to oval-shaped non-lipogenic cells with occasional lipobloasts. A prominent myxoid stroma composed of hyaluronic acid is present (Fukuda and Tsuneyoshi, 2000), in which is embedded a characteristic delicate plexiform capillary vasculature. These features are progressively lost during histological 6  progression to hypercellular round cell morphology, characterized by packed, uniform areas of primitive round cells with high nuclear to cytoplasmic ratio (Figure 1.2). Despite the morphological differences between low grade myxoid and high grade round cell liposarcoma, both variants bear the same cytological hallmark of t(12;16)(q13;p11) chromosomal translocation fusing fused in sarcoma (FUS) to DNA damage inducible transcript 3 (DDIT3), and are considered part of a histological continuum. The high grade round cell component, however, is strongly associated with risk of recurrence, metastasis and poorer disease specific survival, with a cutoff of 5% or more round cells being considered enough to confer this higher clinical risk (Haniball et al., 2011; Lemeur et al., 2015; Nishida et al., 2010). Diagnosis of myxoid liposarcoma is typically supported by fluorescence in situ hybridization (FISH) breakapart assays for FUS and/or DDIT3, or by reverse transcription polymerase chain reaction (RT-PCR) for the most common fusion transcript variants (Downs-Kelly et al., 2008; Narendra et al., 2011; M. P. Powers et al., 2010), although newer techniques based on sequencing (Szurian et al., 2017) or NanoString technology (K. T. E. Chang et al., 2018) are becoming available. Increasingly, molecular diagnostic support is being considered obligatory for definitive diagnosis, even after expert subspecialty histopathology assessment (Italiano et al., 2016). Local control rates in myxoid liposarcoma are excellent with a combination of radiation and surgery, and is largely a result of a particularly high radiation sensitivity in this tumor type (Chung et al., 2009; Guadagnolo et al., 2008; Pitson et al., 2004). For larger locally-advanced tumors, chemotherapy with doxorubicin (an anthracycline) and ifosfamide (a DNA-alkylating agent) is commonly added in some centers. Chemotherapy is also the mainstay of treatment for unresectable and metastatic myxoid liposarcomas; despite being more chemosensitive than other soft tissue sarcomas in terms of apparent response rate, disease-free survival is still poor in the 7  metastatic setting (D. Katz et al., 2012). A 2013 study from the University of Texas MD Anderson Cancer Center reported an 8.2% disease-specific survival rate for metastatic myxoid liposarcoma at five years, and no patient survived beyond nine years (Hoffman et al., 2013). Much needed alternative therapeutic options were made available when the FDA approved trabectedin (a DNA minor groove-binding agent) in 2015, and eribulin (microtubule polymerization inhibitor) in 2016, for use in unresectable liposarcomas, and as a second line treatment in metastatic liposarcoma. However, these newly approved agents act in much the same way as conventional cytotoxic agents, and confer only very modest survival benefits. Although trabectedin was reported to displace the pathognomonic FUS-DDIT3 fusion transcription factor from its target genes in myxoid liposarcoma (Forni et al., 2009; Uboldi et al., 2011), the compound does not specifically target FUS-DDIT3. The resulting DNA conformation change due to binding of trabectedin to the DNA minor groove appears to hinder transcription factor recognition as some transcription factors such as TBP, E2F, SRF and NF-Y show reduced DNA binding after trabectedin treatment (Bonfanti et al., 1999), and additional fusion transcription factors are also displaced following trabectedin treatment, including EWS-FLI1 in Ewing sarcoma (Grohar et al., 2011; Harlow et al., 2016), and EWS-WT1 in desmoplastic small round cell tumours (Uboldi et al., 2017).   1.4 Fusion of FUS to DDIT3 1.4.1 t(12;16)(q13;p11) translocation Myxoid liposarcoma is characterized by the t(12;16)(q13;p11) reciprocal translocation (Limon et al., 1986) which results in the in-frame fusion of FUS to DDIT3 (Aman et al., 1992). In about 3% of cases, an alternative t(12;22)(q13;q12) translocation creates an EWSR1-DDIT3 fusion instead (Panagopoulos et al., 1996; Powers et al., 2010). Twelve fusion variants of FUS-8  DDIT3 have been reported in the literature, each varying in the number of exons retained from FUS and DDIT3 ( Powers et al., 2010; Willems et al., 2010) (Figure 1.3). While the different fusion variants can retain between 3-13 FUS exons, all variants consistently include at least exons 3 and 4 of DDIT3 which contain the full DDIT3 open reading frame. Even in the Type 11 fusion that retains only part of DDIT3 exon 3, the fusion junction occurs upstream of the start codon (Powers et al., 2010). Most variants also include the 5' untranslated region from DDIT3 exon 2 and this sequence becomes translated in the fusion protein (Figures 1.3 and 1.4). Despite the large number of reported fusion variants, types 1, 2 and 3 alone make up 96% of all cases, with the Type 2 variant that fuses FUS exon 5 to DDIT3 exon 2 being reported at the highest frequency (65%) ( Powers et al., 2010). The type of fusion variant does not appear to have an impact on in vitro transforming capabilities or clinical outcome (Antonescu et al., 2001; Bode-Lesniewska et al., 2007; Schwarzbach et al., 2004).  1.4.2 FUS FUS has 15 exons that encode a 75 kDa protein with multiple functions in several cellular pathways (Schwartz et al., 2014). The gene is constitutively active and shows almost ubiquitous expression in human tissue, except in melanocytes, cardiac muscle and endothelium, although its expression level is heterogeneous among different tissues. The subcellular localization of FUS is cell-type dependent but is predominantly nuclear (Andersson et al., 2008). FUS has a domain organization that is common with the other FET proteins, EWSR1 and TAF15 (Figure 1.4). The N-terminal low complexity domain contains degenerate repeats of a serine/tyrosine/glycine/glutamine (SYGQ) motif, and is a powerful transcriptional activation domain. The other domains are a 90-amino acid RNA recognition motif (RRM), which folds into a secondary structure consisting of four β-strands and two α-helices, and binds both RNA and 9  DNA (Xuehui Liu et al., 2013). Three RNA-binding RGG domains with variable number of RGG (arginine-glycine-glycine) repeats are found throughout the protein after the SYGQ-rich domain. There is a zinc finger (ZnF) domain near the C-terminus that also mediates RNA-binding (Iko et al., 2004), and the C-terminus is rounded out by a non-classical proline-tyrosine (PY) nuclear localization signal. Other than the structured RRM and ZnF domains (Iko et al., 2004), FUS is an intrinsically disordered protein, which is advantageous for interacting with a large number of binding partners in a functionally promiscuous manner or to serve as a scaffold for assembly of multiple proteins. Such functional versatility in intrinsically disordered proteins can turn dangerous when normal function is perturbed and results in aggregate formation, leading to diseases such as neurodegeneration (Babu, 2016). This has been the case in subsets of frontotemporal lobar degeneration and amyotrophic lateral sclerosis where FUS aggregates in the cytoplasm of neuronal and glial cells contribute to neurodegenerative pathology (Dormann and Haass, 2013). FUS is a multifunctional protein involved in several cellular pathways, and the existence of at least one FET protein in all muticellular organisms and high conservation of the FET proteins across vertebrates from fish to mammals suggest important physiological roles for these proteins (Schwartz et al., 2014). Multiple studies have characterized the role of FUS in transcriptional regulation. The SYGQ domain is a potent transcriptional activation domain in oncogenic fusion proteins (Bertolotti et al., 1999; Prasad et al., 1994; Zinszner et al., 1994), and FUS appears to affect transcription of thousands of genes in knockdown experiments (Schwartz et al., 2012). FUS acts as a co-regulator of NFκB (Uranishi et al., 2001), RUNX2 (Li et al., 2010), Spi-1/PU.1 (Hallier et al., 1998), and CBP/p300 (Xiangting Wang et al., 2008), and also interacts with steroid hormones, thyroid hormones and retinoid receptors (Powers et al., 1998). 10  Biochemical studies revealed that the N-terminal low complexity SYGQ-rich domain of FUS interacts with the C-terminal domain of RNA polymerase II (RNAP II) in an RNA-dependent manner (I. Kwon et al., 2013; Schwartz et al., 2013). As with other FET proteins, there has been evidence for involvement of FUS in RNA-processing events, including RNA splicing, where FUS knockdown experiments showed changes to RNA splicing (Ishigaki et al., 2012; Lagier-Tourenne et al., 2012; Rogelj et al., 2012; Schwartz et al., 2012). Recent data suggests a role for FUS to couple the transcription machinery to the splicing machinery by mediating the interaction between RNAP II and U1 small nuclear ribonucleoprotein (snRNP) (Chi et al., 2018; Yu and Reed, 2015), which is one of the first splicing factors to bind pre-mRNA during spliceosome assembly. Besides regulating RNA splicing, FUS is also required for optimal microRNA-mediated silencing through its interaction with the core microRNA-induced silencing complex component argonaute 2 (T. Zhang et al., 2018). FUS is involved in DNA damage response, and has DNA homologous pairing activity to mediate annealing of complementary DNA strands (Baechtold et al., 1999; Bertrand et al., 1999). Further evidence from in vivo studies with FUS knockout mice demonstrate high genomic instability and increased sensitivity to radiation (Hicks et al., 2000; Kuroda et al., 2000). Adding to the plethora of FUS functions is the regulation of telomere length through interaction of the FUS RGG3 domain with human telomeric DNA and telomeric repeat-containing RNAs (Takahama et al., 2013).  1.4.3 DDIT3 DDIT3, also known as C/EBP homologous protein (CHOP) or growth arrest and DNA damage-inducible protein 153 (GADD153), is a basic leucine zipper (bZIP) transcription factor, 11  and one of six proteins identified in the CCAAT-enhancer-binding protein (C/EBP) family. The DDIT3 protein has 169 amino acids and is characterized by two functional domains - a transactivation or repression domain in the N-terminus, and a basic leucine zipper domain in the C-terminus (Figure 1.4). The basic region mediates sequence-specific DNA binding, while the leucine zipper motif is required for functional dimerization with C/EBP family members or other bZIP proteins (Tsukada et al., 2011). DDIT3 does not form stable homodimers (Ubeda et al., 1996). Unlike other C/EBP proteins, the DDIT3 basic region contains proline and glycine substitutions that are believed to disrupt the α-helical structure of the DNA binding domain. This structural change leads to increased heterodimerization with other C/EBP proteins, but also allows DDIT3 to function as a dominant-negative inhibitor under certain conditions by disrupting the ability of DDIT3-C/EBP dimers to recognize classic C/EBP sites (Ron and Habener, 1992). However, DDIT3-C/EBP heterodimers are capable of binding other unique DNA sequences (Cucinotta et al., 2008; Ubeda et al., 1996). DDIT3 also dimerizes with other bZIP transcription factors, including ATF3, ATF4, AP-1 family transcription factors (Reinke et al., 2013; N. Su and Kilberg, 2008). DDIT3 is primarily a nuclear transcription factor, but has been reported to be expressed in the cytoplasm under endoplasmic reticulum stress conditions (Jauhiainen et al., 2012). Unlike the near-ubiquitously expressed FUS, DDIT3 expression is limited and tightly regulated. DDIT3 functions as a cellular stress sensor that can be rapidly induced in response to endoplasmic reticulum stress, nutrient deprivation, DNA damage, cellular growth arrest and hypoxia (Carrière et al., 2004; Ma et al., 2002; Y. Yang et al., 2017; Zinszner et al., 1998). DDIT3 also has a role in blocking adipocytic terminal differentiation (Brenner et al., 2015; Han et al., 2013; Tang and Lane, 2000). Adipogenesis can be divided into three phases: specification, commitment and terminal differentiation (Ali et al., 2013). Specification occurs 12  when a multipotent mesenchymal precursor cell restricts its lineage in a reversible manner to adipocytes. During commitment, the possible alternate fates of the multipotent mesenchymal precursor cell becomes progressively restricted in order to irreversibly “commit” to the adipose lineage and turn into a preadipocyte. Terminal differentiation involves a transcriptional cascade that allows the preadipocyte to acquire the characteristics and functions of a mature adipocyte. Existing cellular models for studying adipogenesis, such as mouse 3T3-L1 cells, are pre-adipocytes already committed to the adipose lineage. As a result, much more is known about the process of terminal differentiation than the other phases. The temporal expression of several C/EBP members is involved in the transcriptional cascade of adipocytic terminal differentiation ( Tang and Lane, 2012). DDIT3 is expressed early in the cascade to transiently sequester C/EBPβ in inactive heterodimers before being downregulated at S-phase, thereby releasing C/EBPβ to activate expression of PPARγ, the “master regulator” of mature adipocyte function, as well as C/EBPα (Tang and Lane, 2000). A positive feedback loop between PPARγ and C/EBPα then maintains the terminal differentiated phenotype (Wu et al., 1999), while the decision to differentiate is reinforced by a second positive feedback loop between PPARγ and C/EBPβ (Park et al., 2012).  1.4.4 FUS-DDIT3 structure and regulation The FUS-DDIT3 fusion protein contains at least part of the FUS N-terminus SYGQ-rich low complexity domain and full length DDIT3 (Figure 1.4). It can be argued that only the N-terminus portion of the FUS SYGQ-rich domain is required for the oncogenic property of the fusion protein, since the smallest reported Type 9 variant contains just the first 3 FUS exons. Therefore, in the fusion protein, the RNA binding sequences of FUS are replaced by the DNA binding and bZIP domains of DDIT3. 13  At the transcript level, FUS provides the promoter regulating transcription of the fusion gene, but the DDIT3 sequences, including its 3’ UTR, have a larger influence on FUS-DDIT3 transcript stability. FUS-DDIT3 is degraded much faster than FUS (Aman et al., 2016), due to the DDIT3 portion of the transcript being targeted for rapid degradation under non-stress conditions by RNase L (Fabre et al., 2012; Jackman et al., 1994). At the protein regulatory level, FUS-DDIT3 was reported to bind other FET proteins through the FUS portion of the fusion (Thomsen et al., 2013), which results in stabilization of the fusion protein (Aman et al., 2016).  1.5 Molecular features of myxoid liposarcoma  1.5.1 Functions of FUS-DDIT3 The FUS-DDIT3 fusion protein is the central driver for myxoid liposarcoma. Early studies showed that FUS-DDIT3 is sufficient for in vitro transformation in ST-13 mouse preadipocytes and NIH3T3 mouse embryonic fibroblasts, a phenotype that requires both the DNA binding domain of DDIT3 as well as the N-terminal of FUS or EWSR1 (Kuroda et al., 1997; Zinszner et al., 1994). Transformation was also observed after FUS-DDIT3 expression in the more primitive bone marrow-derived mesenchymal progenitor cells from mice, which resulted in what were claimed to be myxoid liposarcoma-like tumors in vivo (Riggi et al., 2006), but which morphologically have a generic blue cell tumor appearance lacking most of the characteristic histological features of the human disease. It should be noted that when FUS-DDIT3 is co-expressed with C/EBPβ in the context of C/EBPβ-induced adipocytic differentiation, no transformation was seen in NIH3T3 cells (Adelmant et al., 1998), and FUS-DDIT3 expression in fibroblastic cells BALB/c-3T3, adipogenic 3T3-L1, or Rat-1 cells did not result in transformation either (Zinszner et al., 1994). 14  These observations highlight the importance of a permissive cellular context for FUS-DDIT3-driven transformation. FUS-DDIT3 has been reported to block adipocytic terminal differentiation induced by thiazolidinedion and insulin in mouse preadipocytes ST-13(Kuroda et al., 1997), by C/EBPβ in mouse fibroblasts NIH-3T3 (Adelmant et al., 1998), and by hormones in primary mouse embryonic fibroblasts (Pérez-Mancera et al., 2008). The blockade is achieved by FUS-DDIT3 binding to C/EBPβ, and sequestering it from inducing transcription of PPARγ2 and C/EBPα (Adelmant et al., 1998; Pérez-Mancera et al., 2008). This is in line with microarray expression profiles showing an immature adipogenic signature in myxoid liposarcoma (Cheng et al., 2009). However, it is unclear whether and how the differentiation block is linked to the transformative capacity of FUS-DDIT3, as the fusion protein clearly has transcription factor functions beyond sequestering C/EBPβ from activating PPARγ and C/EBPα expression. FUS-DDIT3 is a predominantly nuclear protein (Thelin-Järnum et al., 2002), and is believed to function as an aberrant transcription factor (Riggi et al., 2006; 2007). Differential regulation of IL8 expression was observed between DDIT3 and FUS-DDIT3 expression in HT1080 fibrosarcoma cells, where DDIT3 downregulates IL8, but FUS-DDIT3 upregulates IL8 (Göransson et al., 2005). A follow up study confirmed, via ChIP experiments, binding of FUS-DDIT3 to the IL8 promoter to induce transcription. The promoter binding occurs with FUS-DDIT3 in complex with a novel interactor, NFKBIZ, and requires the NFκB (and not the C/EBP) binding site (Göransson et al., 2009). Another report showed binding of FUS-DDIT3 to the promoters of PTX3 and FN1 to upregulate gene expression, albeit in one myxoid liposarcoma cell line (402-91), but not the other (1765-92) (Forni et al., 2009). Crucially, no ChIP-seq data is available to identify the full complement of FUS-DDIT3 gene targets, representing a major gap in current knowledge of the function behind FUS-DDIT3. 15  Similarly, at the protein level, there have been no published reports of unbiased identification of the FUS-DDIT3 interactome, although the wild type FUS interactome was published in a number of recent studies (Chi et al., 2018; Kamelgarn et al., 2016; Sun et al., 2015; Tao Wang et al., 2015; Yamazaki et al., 2012), and some FUS-DDIT3-interacting proteins were reported in separate studies. Besides C/EBPβ, NFKBIZ, and the FET proteins that were mentioned earlier, CDK2 was also reported to associate with FUS-DDIT3 (Bento et al., 2009). Additionally, FUS-DDIT3 retains the ability to bind RNAP II, but is unable to recruit Y-box binding protein-1 (YB-1) due to the replacement of C-terminal FUS with DDIT3, and interfering with YB-1-mediated splicing in reporter assays (Rapp et al., 2002). Another study reported localization of FUS-DDIT3 to the splicing factor compartments (Göransson et al., 2002). These observations suggest that FUS-DDIT3 may participate in RNA splicing. FUS-DDIT3 also does not have the capability of wild type FUS to promote DNA pairing (Baechtold et al., 1999). It is tempting to speculate that FUS-DDIT3 may interfere with normal FUS function in DNA damage repair, as is the case for EWS-FLI1, which blocks homologous recombination repair through EWSR1 loss of function (Gorthi et al., 2018).  1.5.2 Secondary mutations and pathway activations While FUS-DDIT3 is the main recurrent genomic aberration in myxoid liposarcoma, additional non-random chromosomal aberrations with uncertain roles in tumor progression have been reported in a subset of myxoid liposarcomas, including trisomy 8 and idic(7)(p11.2) (Sreekantaiah et al., 1991; Ohijmi et al., 1992; Mandahl et al., 1994; Schaad et al., 2006). More importantly, a limited number of unbiased exome sequencing studies have attempted to identify actionable secondary mutations have revealed a relatively quiescent genomic landscape (Barretina et al., 2010; Hofvander et al., 2018; Joseph et al., 2014). The rarity of myxoid 16  liposarcoma has unfortunately meant that the disease is often excluded from many large-scale sequencing studies, such as the recent sarcoma study from The Cancer Genome Atlas that analyzed one of the largest cohorts of sarcomas in this field (Cancer Genome Atlas Research Network, 2017), or grouped with other biologically distinct liposarcomas (Chalmers et al., 2017). As a result, the genomic landscape of myxoid liposarcoma is poorly characterized compared to some other sarcoma types. However, targeted sequencing studies have identified a number of low frequency mutations, as discussed below. TP53 is the most commonly mutated gene in human cancer (Kandoth et al., 2013), but is not frequently mutated in myxoid liposarcoma. Overexpression of p53, typically associated with underlying missense mutation, has been detected by immunohistochemistry in a minority of cases ranging from 0.05% to 17% (Antonescu et al., 2001; Pilotti et al., 1998). Although an earlier study that sequenced exons 4-9 of TP53 showed mutations in 6/21 (28.5%) cases (Dei Tos et al., 1997), later exome sequencing studies did not detect any TP53 mutations in myxoid liposarcoma (Barretina et al., 2010; Joseph et al., 2014). TP53 mutations were correlated with poor outcome in an earlier study by Antonescu et al., 2001, but not in a more recent study by Hoffman et al., 2013. The variation in mutation frequency may be attributed to the use of different antibodies for immunohistochemical studies and /or small number of cases in each study, but taken together, the existing data suggests that TP53 mutation is not a frequent occurrence in myxoid liposarcoma. There are recent reports of telomerase reverse transcriptase (TERT) C228T and C250T hotspot mutations in 23-79% of myxoid liposarcoma tumors (Ferreira et al., 2018; Killela et al., 2013; Koelsche et al., 2014; Saito et al., 2016). These mutations can result in increased TERT mRNA and telomerase reactivation (Vinagre et al., 2013). Increased telomerase activity is also correlated with disease progression, although those studies did not evaluate TERT promoter 17  mutations in their cohort (Jaeger et al., 1999). A recent study on nine myxoid liposarcoma cases does report significantly longer telomeres in tumors harboring TERT promoter mutations compared to those without (Ferreira et al., 2018). Phosphatidylinositol 3-kinases (PI3Ks) are crucial regulators of cell growth, metabolism and survival in response to extracellular stimuli, and are negatively regulated by phosphatase and tensin homologue (PTEN). Mutation and/or amplification in the PI3K catalytic subunit p110α (PIK3CA), and PTEN loss are common in human cancers (Thorpe et al., 2015). In myxoid liposarcoma, recurrent activating PIK3CA mutations are found in 14-18% of tumors, and is mutually exclusive to the complete loss of PTEN found in 12% of cases (Barretina et al., 2010; Demicco et al., 2012). Overexpression of the rearranged during transfection (RET) receptor tyrosine kinase and insulin-like growth factor 1 receptor (IGF1R) in myxoid liposarcoma (Cheng et al., 2009) alludes to activation of the Ras-RAf-ERK/MAPK and PI3K/AKT pathways. Activation of other receptor tyrosine kinases – PDGFRβ, EGFR, MET, RET and VEGFR2 – was also reported in seven treatment naive tumors, with downstream activation of AKT, ERK1/2 and/or mTOR, although none harbored mutations in PI3KCA, PTEN, KRAS or BRAF. While the sample size was small, downstream AKT activation was restricted to the more aggressive round cell variant, which may indicate a role for AKT activation in tumor progression (Negri et al., 2010). Coupled with the reports of recurrent PI3KCA mutation and PTEN loss, the existing data highlights the potential of targeting the PI3K/AKT pathway in a subset of myxoid liposarcomas, although this class of drugs has thus far not provided great benefits to patients (Fruman and Rommel, 2014). Another upstream activator of the PI3K/AKT signaling pathway are the Src family of kinases that regulate embryonic development, cell growth and cell survival, and were identified to not only be active, but also sensitive to inhibition by dasatanib in myxoid 18  liposarcoma (Sievers et al., 2015; Willems et al., 2010), although clinical trials in sarcomas with this agent have also not shown activity in liposarcomas (Schuetze et al., 2016). A number of gene expression studies have also been carried out on small cohorts of myxoid liposarcoma cases using DNA microarray or RNA sequencing (Barretina et al., 2010; Brunner et al., 2012; Künstlinger et al., 2015; Renner et al., 2013; Singer et al., 2007). Although most of these studies did not pursue any targets for further validation, the data provide a rich and ready resource for differential gene expression in myxoid liposarcoma. One of the studies did validate the overexpression of fibroblast growth factor 2, with inhibitors against the receptor inducing apoptosis, and reducing viability and migration of myxoid liposarcoma cells in vitro (Künstlinger et al., 2015).  1.6 Targeted therapy for myxoid liposarcoma  1.6.1 PPARγ agonists Since one of the histological features myxoid liposarcoma shows is incomplete adipocytic differentiation, PPARγ agonists have been suggested as a therapeutic option. This rationale is supported by in vitro evidence of FUS-DDIT3 blocking adipocytic terminal differentiation through repression of the PPARG promoter (Pérez-Mancera et al., 2008). Two Phase 2 trials have been attempted with PPARγ agonists in liposarcomas. Tumor biopsies from two of the myxoid liposarcoma patients in the troglitazone trial (NCT00003058) showed histological and molecular evidence of terminal differentiation after treatment, but the trial had only a short follow-up period (Demetri et al., 1999). The other Phase 2 trial with rosiglitazone (NCT00004180) had a longer follow-up period, but showed no signs of differentiation or anti-tumor effects, despite evidence of PPARγ activation in vivo (Debrock et al., 2003).  19  1.6.2 Pazopanib Pazopanib is an oral angiogenesis inhibitor that targets mainly VEGFR, PDGFR and c-kit (Schutz et al., 2011). In the Phase 3 PALETTE trial (van der Graaf et al., 2012) that led to FDA approval of pazopanib in soft tissue sarcomas, liposarcomas as a group were excluded due to disappointing results in the preceding Phase 2 trial (Sleijfer et al., 2009). Two retrospective studies in Asian patients again lumped the molecularly distinct liposarcomas together, and showed shorter progression free survival in liposarcoma compared to other histologic types of soft tissue sarcomas (Nakamura et al., 2016; Yoo et al., 2015). As mentioned above, myxoid liposarcoma is molecularly distinct and should not be lumped together with well differentiated / dedifferentiated liposarcoma, meaning these studies missed an important opportunity to assess pazopanib in myxoid liposarcoma, with its unique plexiform vasculature (Endo and Nielsen, 2012). Myxoid liposarcoma shows activation of PDGFR-β and VEGFR2 (Negri et al., 2010), and has a distinct pathognomonic vascular pattern that is worthy of assessment with pazopanib as a molecularly disparate tumor type from other liposarcomas, but this is a challenge to achieve given its relative rarity to accrue enough trial patients for statistically meaningful analyses. However, there continues to be ongoing investigations of pazopanib in liposarcomas (NCT01506596, NCT01692496) that hopefully will stratify the potential for activity in myxoid liposarcoma. Nevertheless, this therapeutic strategy does not target the FUS-DDIT3 oncoprotein directly, nor any of its known direct oncogenic mechanisms.  1.6.3 Olaratumab Olaratumab is a monoclonal antibody that selectively binds to PDGFRα, and prevents the PDGF ligands from attaching to and activating the receptor. In 2016, the US FDA approved 20  olaratumab for the treatment of advanced soft tissue sarcomas when used in combination with doxorubicin based on data from a randomized Phase 2 trial (NCT01185964) (Tap et al., 2016). Such a combination could be particularly useful to a doxorubicin-sensitive disease such as myxoid liposarcoma, in which PDGFRα expression appears to be induced by FUS-DDIT3 (Riggi et al., 2006). A Phase 3 study is now underway to evaluate the doxorubicin/olaratumab combination and assess benefit to overall survival (NCT02451943).  1.6.4 HDAC inhibitors HDAC inhibition using quisinostat, dacinostat, and panobinostat have shown to be effective at reducing cell viability of myxoid liposarcoma cells in a recent in vitro drug screen (de Graaff et al., 2017), although the transcriptomic and epigenomic changes induced by HDAC inhibition have not been investigated in this tumour type. In a Phase 2 clinical trial of the HDAC inhibitor pracinostat (SB-939) for translocation-associated sarcomas (NCT01112384), stable disease was achieved in three of the four assessable myxoid liposarcoma patients (Chu et al., 2015). Although the trial was terminated prematurely due to prolonged unavailability of pracinostat, the initial data warrants further investigation into HDAC inhibition for myxoid liposarcoma.  1.6.5 NY-ESO-1 immunotherapy The cancer-testis antigen New York Esophageal Squamous Cell Carcinoma-1 (NY-ESO-1) / cancer-testis angtigen 1B (CTAG1B) is a highly immunogenic protein with unknown functions, but its expression is restricted to germ cells, placental cells and a wide range of tumor types (Thomas et al., 2018). The vast majority (88-100%) of myxoid liposarcomas express NY-ESO-1 (Endo et al., 2015; Hemminger and Iwenofu, 2013; Iura et al., 2015; Pollack et al., 2012). 21  NY-ESO-1 is widely believed to be an excellent candidate target for augmentation immunotherapy strategies due to its restricted expression in normal tissues, and highly immunogenic nature. One approach in myxoid liposarcoma utilizes a dendritic cell-targeted lentiviral NY-ESO-1 vaccine, LV305, to introduce NY-ESO-1 into dendritic cells for peptide processing and presentation by class II MHC molecules to CD4+ and CD8+ T cells. The vector has been proven safe and well tolerated in Phase 1 trials (NCT02122861, NCT02387125), with signs of clinical activity from interim results (Somaiah et al., 2016). A Phase 2 trial (NCT02609984) combining the vaccine regimen with PD-L1 inhibitor, atezolizumab, is ongoing and interim results suggest that the combination may be associated with better survival (Somaiah et al., 2017). Adoptive immunotherapy is also being explored in myxoid liposarcoma in a current Phase 1 / 2 pilot study (NCT02992743) for adoptive transfer of autologous T-cells transduced with the T cell receptor against NY-ESO-1. The most impressive advances in immunotherapy, however, have come in highly antigenic tumors carrying a high mutational burden, such as melanoma, non-small cell lung carcinoma and mismatch repair-deficient or microsatellite unstable cancers (Chang et al., 2018). Unlike these tumors, myxoid liposarcoma harbors a low mutational burden (Barretina et al., 2010; Joseph et al., 2014), and current immunotherapies in clinical use do not directly target the causative FUS-DDIT3 oncoprotein or its interacting partners.   1.7 Experimental models of myxoid liposarcoma While targeted therapies are urgently needed for myxoid liposarcoma, none of the current therapies discussed in the above sections convincingly target FUS-DDIT3 specifically. There is also a major gap in our current knowledge of immediate effectors that interact with FUS-DDIT3, which prevents the design of strategies that target these interactors instead. The discovery of 22  targeted therapies will rely entirely on the availability of accurate model systems to find and test key druggable effector interactors of FUS-DDIT3. In this area, a limited number of models are available for myxoid liposarcoma as detailed below.  1.7.1 Cell lines and primary cultures There are currently a limited number of immortalized human myxoid liposarcoma cell lines that have been published, with none available from ATCC. Only two SV40 large T antigen-transformed cell lines (402-91 and 1765-92) have been made widely available to researchers (Aman et al., 1992). With contributions from our group, a new, spontaneously immortalized cell line (DL-221) was recently characterized and made available to the research community (de Graaff et al., 2016). Both 402-91 and DL-221 harbor the Type 1 fusion (fusing FUS exon 7 to DDIT3 exon 2), whereas 1765-92 expresses the Type 8 fusion (fusing FUS exon 13 to DDIT3 exon 2) (Figure 1.3). The characteristics of all three cell lines used in this thesis, along with their reported molecular biomarkers (de Graff et al. 2016), are summarized in Table 1.1. Cell lines are easily grown and manipulated, and are an invaluable complement to in vivo experiments. However, there are inherent disadvantages in modelling the disease from which they are derived after adapting to growth in culture. Lack of tissue architecture and tumor microenvironment often abolishes tissue-based cell-cell and cell-matrix interactions, secretion and other functions, and cell lines may become selected to acquire a molecular phenotype quite different from tumor cells in vivo. Short term culture of primary cells circumvents some of these problems, and a few studies have successfully made use of short term cultures of myxoid liposarcoma primary cells for validation of drug inhibition findings (de Graaff et al., 2017; Sievers et al., 2015; Willems et al., 2010). However, it has been technically challenging to isolate and culture solid tumour cells 23  in an in vitro environment similar to the microenvironment of the original tumor (Mitra et al., 2013). Viable cells are also difficult to come by as most surgically excised primary tumors have been pre-treated with radiation.  1.7.2 Mouse models A number of myxoid liposarcoma mouse models have been reported in the literature. Ubiquitous expression of FUS-DDIT3 driven by the elongation factor 1α (EF-1α) promoter produces tumors exhibiting features of human liposarcomas, including lipoblasts with round nuclei, accumulation of intracellular lipid, and induction of adipocyte-specific genes, but do not contain the characteristic myxoid matrix and vascular pattern of myxoid liposarcoma (Pérez-Losada et al., 2000). Indeed, the images presented in this paper resemble hibernoma (a benign tumor of brown fat) much more than myxoid liposarcoma. Another transgenic mouse model is driven by the aP2 (FABP4) promoter, which expresses FUS-DDIT3 only after the expression of PPARγ, since FABP4 is a downstream target of PPARγ. These mice did not develop liposarcomas, suggesting that transformation has to occur at an earlier stage of mesenchymal development (Pérez-Mancera et al., 2007). The most recent published attempt at making a myxoid liposarcoma mouse model expressed FUS-DDIT3 under the mesoderm-specific promoter Prrx1 but required a p53-null background to produce tumors (Charytonowicz et al., 2012), despite TP53 mutations being uncommon in myxoid liposarcoma. Human myxoid liposarcoma tumors display nuclear PPARγ IHC stains (Cheng et al., 2009; Hoffman et al., 2013), but these were not observed in the Prrx1-driven mouse tumors. None of these transgenic mouse models have been able to accurately model myxoid liposarcoma, once again highlighting the importance of expressing FUS-DDIT3 in the correct, 24  permissive cellular background. Lack of knowledge of the cell of origin and/or most appropriate contextual Cre-driver representative of myxoid liposarcoma remains a stumbling block in this respect. Considering the predisposition of myxoid liposarcoma for presentation in young adults, it can be speculated that expression of the fusion protein at a specific developmental stage may also be important. Other possible reasons for the difficulties in making a myxoid liposarcoma mouse model could be due to inherent biological differences between human and mice. For example, alternative splicing of conserved exons is frequently species-specific (Pan et al., 2005), and may be a roadblock to generating myxoid liposarcoma in mice if FUS-DDIT3 functions partially through alternative splicing interference. Fusion proteins that function as aberrant transcription factors may also operate as pioneer factors for critical regulatory elements at DNA sites without evolutionary conservation. This was illustrated by the neomorphic binding of EWS-FLI1 at enhancer GGAA repeats that do not have normal regulatory activity or show evolutionary conservation, and might explain the failure to generate a mouse model of Ewing sarcoma despite major efforts from multiple research groups (Minas et al., 2017).   1.8 Thesis objective and chapter overview Despite the presence of a driver fusion oncoprotein FUS-DDIT3 in myxoid liposarcoma, the exact mechanisms of action behind its capacity for transformation are still unclear and represents part of the challenge in finding targeted therapies against this cancer of young adults. FUS-DDIT3 is not targeted by any existing drugs, suggesting a drug repurposing strategy might be worth pursuing, but to date, genomic profiling at the expression or exome level has not revealed consistent alternative targets. The lack of comprehensive data on the FUS-DDIT3 25  interactome represents one of the major gaps in knowledge behind the oncogenic functions of the fusion protein. My hypothesis is that application of new proteomic technologies to myxoid liposarcoma will identify key interactors and effectors of FUS-DDIT3’s oncogenic function. The ultimate goal of the work presented in this thesis is to identify the network of proteins in the FUS-DDIT3 interactome against which rational targeted therapeutic strategies could be designed. Chapter 2 employed immunoprecipitation-mass spectrometry (IP-MS) to isolate and identify FUS-DDIT3 interactors from whole cell lysates of three myxoid liposarcoma cell lines expressing endogenous FUS-DDIT3. Top-ranked proteins identified from the screen were independently validated to be in complex with FUS-DDIT3. Bioinformatics analyses of the FUS-DDIT3 interactome showed the presence of surprisingly few transcription factors and targetable proteins, but an enrichment of RNA processing and splicing proteins. Chapter 3 followed up on the observation of multiple RNA processing and splicing proteins in the FUS-DDIT3 interactome. This chapter focused on assessing the potential role for FUS-DDIT3 in deregulating alternative splicing. RNA-seq was performed on 402-91 myxoid liposarcoma cells after FUS-DDIT3 knockdown to look for changes in the alternative splicing profile, but no evidence of such changes were ultimately seen.  Chapter 4 attempted to increase the identification of transcription factors and co-regulators in the FUS-DDIT3 interactome that may have more functional relevance. This was achieved by (1) performing IP-MS specifically on nuclear lysates of 402-91 to increase enrichment of relatively less abundant transcription factors, and (2) by reducing sample background for MS identification by using Tandem Mass Tag (TMT)-labeling for quantitative analysis. Components of several epigenetic regulatory complexes were identified in this nuclear extract interactome. 26  Finally, Chapter 5 summarizes the findings of this work in the context of understanding FUS-DDIT3 biology and function in myxoid liposarcoma, and also provides future directions for extending the findings of this work. 27    Figure 1.1 Recurrent sarcoma fusion genes involving the FET (FUS, EWSR1, TAF15) family genes The homologous EWSR1 and FUS are promiscuous 5’ fusion gene partners in multiple sarcomas, while the 3’ gene partners exhibit more specificity for sarcoma types.   FUSEWSR1TAF15Myxoid liposarcomaDDIT3CREB3L1Sclerosing epithelioid fibrosarcomaCREB3L2Low-grade fibromyxoid sarcoma,sclerosing epithelioid fibrosarcomaCREB1ATF1 Angiomatoid fibrous histiocytoma,clear cell sarcomaMyoepithelioma/mixed tumourPBX1POU5F1ZNF444KLF17Extraskeletal myxoid chondrosarcomaNR4A3WT1ERGETV1ETV4FEVFLI1Ewing sarcomaCREB3L2Low-grade fibromyxoid sarcomaATF1 Angiomatoid fibrous histiocytomaDesmoplastic small round cell tumour28   Figure 1.2 Histology of myxoid liposarcoma Hematoxylin and eosin stain showing uniform, round to oval, primitive non-lipogenic mesenchymal cells and a variable number of small signet-ring lipoblasts. The prominent myxoid stroma and characteristic branching, plexiform vasculature pattern are lost in the high grade “round cell” variant of the disease (inset), which is defined as highly cellular areas with back to back primitive round cells with increased nucleocytoplasmic ratio and prominent nucleoli.   29    Figure 1.3 Fusion variants of FUS-DDIT3 The t(12;16)(q13;p11) chromosomal translocation results in the fusion of the first 3-13 N-terminus FUS to the last 2-3 C-terminus DDIT3 exons. Twelve FUS-DDIT3 variants have been reported in the literature as a result of different breakpoints, all of which include the entire coding region of DDIT3 and a portion of the DDIT3 5’ untranslated region (UTR). The most common variant (Type 2) occurs in 65% of cases, and fuses FUS exon 5 to DDIT3 exon 2 (5-2). The first 3 fusion types (Types 1, 2 and 3) together constitute 96% of all reported cases in a meta-analysis (Powers et al., 2010) . Exons that are only partially retained are indicated by white areas in the rectangles. Size of exons are not to scale.  30               Figure 1.4 Structure and domains of FUS and DDIT3 Wild type FUS contains the following protein domains: a low complexity serine/tyrosine/glycine/glutamine (SYGQ)-rich domain, three arginine- and glycine-rich RGG motif domains, an RNA recognition motif (RRM) domain, a zinc finger (ZnF) domain, and a non-classical proline-tyrosine (PY) nuclear localization signal. The domain organization of FUS is similar to homologues EWSR1 and TAF15. Wild type DDIT3 contains a transactivation/repression domain in the N-terminus followed by a basic leucine zipper in its C-terminus. The two most common FUS-DDIT3 fusion variants retain the SYGQ-rich domain of FUS, and also includes the amino acid sequence of a portion of the previously untranslated region (UTR) from DDIT3 exon 2. Schematic illustration of protein domain structure was generated with the tool Illustrator of Biological Sequences (IBS) (Liu et al., 2015).    31  Table 1.1 Myxoid liposarcoma cell line characteristics. Cell line 402-91 1765-92 DL-221 Source Primary; myxoid only (no round cell component) Unknown Pleural metastasis; with round cell component Gender M Unknown M Age 33 Unknown 42 Karyotype of original tumour 46,XY,t(1;7)(p21;p11),t(12:16)(q13;p11) Unknown Unknown Reported cell line karyotypes in literature 46-47,XY,t(1;7)(p21;p11),t(12;16)(q13;p11),inc (Aman et al. 1992)  46, X, der(Y)t(Y;19) (q11;p11), t(1;7)(p12;p12), der(8)t(8;21)(p11;p11) [7], der (8)t(8;9)(p11;p11) [7], del(8)(p11) [4], del(10)(p11), t(12;16)(q13;p11), del(18)(p11), -19,+20, -21 [7] [cp20] (Willems et al. 2010) 90-99, XX, der(1)inv(1)(p32q31)t(1;10)(p33;p12), der(1)inv(1) (p32q31)t(1;10) (p33;p12), -1, del(2)(p11), -3, +5, der(6)t (4;6)(4q, 6q), der(6)t(6;10)(p;q), +der(6)t(6;10) (p;q), der (8)t(3;8), i(8)(q10), +i(8)(q10), +9, der(10)t(1;10)(1p32, p12), der(10)t(1;10) (1p32, p12), -10, +11, t(12;16)(q13; p11), t(12;16)(q13;p11), -13, der(13)t(6;13)(q;q), +14, +15, +18, +20, +20 [cp20] (Willems et al. 2010) 71 ~ 75<3n ± >,der(X)t(X;15)(q;q)x2,-Y, der(2)t(X;2)(q;p)t(2;3)(q;p)x2,-4,+5,der(7)t(7;X)(q;q),+i(7)(q10), +7,+8,+8,-9,+11,t(12;16)(q13;p11), − 13,+14, − 15,der(17)t(10;17) (p;p),+18,+19,+20,+21,idic(22)[10]/ 71 ~ 75<3n ± >,der(X) t(X;15)(q;q)x2, − Y,der(2)t(X;2)(q;p)t(2;3)(q;p)x2,der(4)t(2;4)q(q; p), − 4,+5,der(7)t(5;7)(p;q),der(7)t(7;X)(q;q),+8,+8, − 9,+10,+11, t(12;16)(q13;p11)x2, − 13,+14, − 15,+18,+19,+20,+21,+22[6] (de Graff et al. 2016) Transformation Large T antigen Large T antigen Spontaneous immortalization Molecular biomarkers Fusion variant Type 1 (FUS exon 7 to DDIT3 exon 2) Type 8 (FUS exon 13 to DDIT3 exon 2) Type 1 (FUS exon 7 to DDIT3 exon 2) TP53 Wild-type Wild-type T125R, N239D P53 (western blot) Positive Positive Positive PIK3CA status Wild-type Wild-type Wild-type TERT status C228T mutation C228T mutation C228T mutation p-AKT473/Total AKT (western blot) +/++ +/++ -/++ AXL (western blot) + +++ ++ HDAC1/2/3 (western blot expression) +/++/+ +/+/- -/+/+ NY-ESO-1 (IHC) + ++ ++ 32  Chapter 2: Protein interaction network of FUS-DDIT3 2.1  Introduction 2.1.1 Targeting oncogenic transcription factors The driver mutation in myxoid liposarcoma is a reciprocal chromosomal translocation that results in the in-frame fusion of FUS to DDIT3. The resulting FUS-DDIT3 fusion protein functions as an aberrant transcription factor that represents the most logical and promising therapeutic target in myxoid liposarcoma. Unfortunately, there are currently no pharmaceutical options available for targeting FUS or DDIT3, let alone any drugs that can specifically target their chimaeric fusion oncoprotein. Transcription factor deregulation is a common theme in human cancers. There are an estimated 294 transcription factors and co-regulators (18.7%) out of a list of 1571 known and candidate oncogenic proteins (Lambert et al., 2018). With the exception of ligand-inducible nuclear receptors, transcription factors were for a long time considered "undruggable" because they lack enzymatic activity suitable for chemical intervention. There was also the perception that transcription is a nuclear event that is not readily accessible to therapeutic agents. However, recent work has revealed that transcription factors are amenable to modulation through various strategies, some of which have been successfully applied to driver fusion transcription factors (presented here with non-exhaustive examples):  (1) Blocking DNA binding The bromodomain and extra-terminal domain (BET) proteins, BRD2, BRD3, BRD4 and BRDT, are epigenetic readers that recognize and bind N-acetylated-lysine residues on histone tails. The bound BET proteins induce an open-chromatin structure and serve as a scaffold to recruit transcriptional complexes and RNA polymerases (Prinjha et al., 2012). Recently 33  developed BET inhibitors, such as JQ1, work by displacing BRD2/3/4/T from chromatin, and have been successfully used to block the BRD4-NUT fusion transcription factor from binding its gene targets in NUT midline carcinoma (Filippakopoulos et al., 2010). An early clinical proof-of-concept trial of drug-like derivatives of JQ1 has shown rapid and impressive antitumor activity in advanced NUT midline carcinoma (Stathis et al., 2016). Trabectedin, recently approved for the treatment of metastatic sarcomas with smooth muscle or fatty differentiation, is a chemotherapeutic agent that binds the DNA minor groove, resulting in a DNA conformation change that might hinder transcription factor recognition, as suggested by reduced DNA binding of some transcription factors such as TBP, E2F, SRF and NF-Y after trabectedin treatment (Bonfanti et al., 1999). Trabectedin has also been suggested to affect transcription in sarcomas by displacing fusion transcription factors from target promoters including FUS-DDIT3 in myxoid liposarcoma (Forni et al., 2009; Uboldi et al., 2011), EWS-FLI1 in Ewing sarcoma (Grohar et al., 2011; Harlow et al., 2016), and EWS-WT1 in desmoplastic small round cell tumors (Uboldi et al., 2017).  (2) Direct modulation of expression or degradation A second strategy for targeting transcription factors involves modulating their expression or degradation. Specifically, JQ1 has been shown to displace BRD4 from the promoter of the EWS-FLI1 fusion gene in Ewing sarcoma, with corresponding downregulation of the fusion transcript and protein, as well as antitumor effects in vitro and in vivo (Jacques et al., 2016). There are also examples of compounds modulating degradation of oncogenic transcription factors as one of their antitumor effects. In acute promyelocytic leukemia, which is driven by the PML (promyelocytic leukemia)-RARα (retinoic acid receptor α) oncogenic fusion transcription factor, the RARα ligand all-trans retinoic acid (ATRA) can reverse PML-RARα-34  mediated repression, and induce differentiation (Martens et al., 2010). However, together with arsenic trioxide, ATRA can also induce proteasome-dependent degradation of PML-RARα, which leads to p53-mediated senescence and eradication of leukemia stem cells (Ablain et al., 2014; Nasr and De Thé, 2010). Indeed, the combination of both ATRA and arsenic trioxide has led to >90% complete remission of acute promyelocytic leukemia patients in a major Phase 3 clinical trial (Lo-Coco et al., 2013). Further work confirmed that PML-RARα degradation (not ATRA-mediated differentiation) is the primary mechanism responsible for curing acute promyelocytic leukemia (Ablain et al., 2013; Nasr and De Thé, 2010). In recent years, targeted protein degradation using the PROTAC (PROteolysis Targeting Chimeras) technology has been developed against “undruggable” proteins (Neklesa et al., 2017). PROTACs are heterobifunctional molecules consisting of two recruiting ligands connected via a linker, where one ligand is designed to recognize the protein of interest, while the other ligand recruits an E3 ligase to facilitate poly-ubiquitination and subsequent proteasome-mediated degradation of the target protein (Toure and Crews, 2016). This approach has been successfully applied to induce degradation of a number of proteins, including BRD4 (Lu et al., 2015), BRD9 (Remillard et al., 2017), and BCR-ABL (Lai et al., 2016).   (3) Blocking crucial protein-protein interactions There are several examples of drugs in development that function through disrupting crucial protein-protein interactions with transcription factors. The p53 tumor suppressor protein was the first transcription factor targeted with this approach. Binding of p53 to its negative regulator MDM2 leads to proteasome-mediated degradation of p53 (B. P. Zhou et al., 2001). Major advances have been made in the design of inhibitors against the p53-MDM2 interaction, 35  and several, such as RG7112 and idasanutlin, have moved into clinical trials (Lemos et al., 2016; Shaomeng Wang et al., 2017). In the case of fusion proteins, the interaction of the Ewing sarcoma fusion transcription factor EWS-FLI1 with RNA helicase A inhibits the helicase activity of RNA helicase A in vitro (Erkizan et al., 2015), and is critical for transformation (Toretsky et al., 2006). A subsequent empirical screen using full-length EWS-FLI1 identified a small molecule YK-4-279 that can prevent RNA helicase A from binding EWS-FLI1, resulting in apoptosis of Ewing sarcoma cells, growth reduction of orthotopic xenografts (Erkizan et al., 2009) and sustained complete response in 2 of 6 Ewing sarcoma tumors in a rat xenograft model (Hong et al., 2013). The FDA granted YK-4-279 orphan drug status for Ewing sarcoma in 2013 under the name Efdispro® (Lambert et al., 2018), although research to improve its bioavailability is ongoing (Lamhamedi-Cherradi et al., 2015). A derivative of YK-4-279, TK216, was since developed and has become the first-in-class inhibitor of the ETS-family transcription factors in Phase 1 for relapsed or refractory Ewing sarcoma that includes a combination with vincristine in the expansion cohort (NCT02657005). In synovial sarcoma, investigations into the SS18-SSX interactome revealed that, as one of its mechanisms of action, the oncoprotein functions as a bridge between transducin-like enhancer of split 1 (TLE1) (which recruits the polycomb group repressor components and HDAC1) and activating transcription factor 2 (ATF2), resulting in the repression of ATF2 target genes, and a proliferative, anti-apoptotic phenotype ( Su et al., 2012). The use of histone deacetylase (HDAC) inhibitors was able to disrupt the critical SS18-SSX/TLE1/ATF2 association and reactivate expression of tumor suppressors otherwise repressed by SS18-SSX (Laporte et al., 2017; Su et al., 2012). These findings provided the biological rationale for including synovial sarcoma in clinical trials of HDAC inhibitors (NCT01112384, NCT00918489, NCT00878800). 36  (4) Functional inhibition of cofactors The net regulatory function of transcription factors is dependent on the distinct assortment of cofactors that they are in complex with. The identification and functional inhibition of these cofactors, mainly associated with epigenetic control, has been one of the strategies pursued in several types of translocation-associated sarcomas. For example, EWS-FLI1 mediated transcriptional repression is facilitated through direct interaction with the nucleosome remodeling and deacetylase (NuRD) - lysine demethylase 1 (LSD1) complex (Sankar et al., 2013), and can be reversed in Ewing sarcoma cell lines using the LSD1 inhibitor HCI-2509, resulting in significantly delayed tumorigenesis in vivo (Sankar et al., 2014). In alveolar rhabdomysoarcoma, a recent interactome screen for partners of its pathognomonic PAX3-FOXO1 fusion transcription factor revealed that multiple components of chromatin remodeling complexes were functionally important for alveolar rhabdomyosarcoma cell proliferation (Böhm et al., 2016). Specifically, chromodomain helicase DNA binding protein 4 (CHD4), an ATP-dependent chromatin remodeler, is a crucial coregulator of a subset of PAX3-FOXO1 target genes, and CHD4 depletion caused regression of xenograft tumors in vivo (Böhm et al., 2016). While CHD4 is not currently targetable, it contains 2 plant homeodomain (PHD) fingers and tandem chromodomains, which could be targets for a similar drug design strategy to the one described above that was successful against the BET bromodomains (Filippakopoulos et al., 2010; Morra et al., 2012).  2.1.2 FUS-DDIT3 interactors Several proteins have been reported to associate with FUS-DDIT3, the earliest of which was C/EBPβ as the major dimerization partner (Crozat et al., 1993; Göransson et al., 2005; Ron 37  and Habener, 1992) critical to the ability of FUS-DDIT3 to block adipocytc differentiation (Adelmant et al., 1998; Pérez-Mancera et al., 2008). Cyclin-dependent kinase 2 also associates with the DDIT3 part of FUS-DDIT3, whereby exogenous FUS-DDIT3-GFP is able to cause relocation of CDK2 from a general nuclear distribution into FUS-DDIT3 containing subnuclear structures, in HT1080 fibrosarcoma cells (Bento et al., 2009). Another reported FUS-DDIT3 interactor is NFKBIZ, a NFκB transcription cofactor that localizes with FUS-DDIT3 at the NFκB binding site in the IL8 promoter to induce transcription (Göransson et al., 2009). The other known interactors of FUS-DDIT3 associate through the retained N-terminal FUS sequences. An early study reported co-immunoprecipitation of RNAP II with both FUS and FUS-DDIT3, and confirmed that the RNAP II association was through the FUS (and not DDIT3) portion of the fusion protein (Rapp et al., 2002). Wildtype FUS was subsequently found to bind directly to RNAP II and prevent its hyperphosphorylation by CDK12 (Schwartz et al., 2012), but it is unclear whether and how this function might be altered by FUS-DDIT3. Finally, a conserved N-terminal motif in the FET proteins mediates robust binding with various C-terminal regions within the proteins, providing a model for homo- and hetero-complexes as major binding partners (Kato et al., 2012; Thomsen et al., 2013). Since the N-terminal of FUS is retained, FUS-DDIT3 was also shown to co-immunoprecipitate with FUS, EWSR1 and TAF15 (Thomsen et al., 2013). Despite identification of these individual protein partners, there has been no systematic investigation into the FUS-DDIT3 interactome using modern proteomic methods. It would be rational to assume that the fusion oncoprotein’s interactome would include a combination of FUS and DDIT3 interactors as dictated by cellular context. While an unbiased study of the DDIT3 interactome has not been published, a search of The Biological General Repository for Interaction Datasets (BioGRID) (https://thebiogrid.org) (Stark et al., 2006) returned 76 unique 38  interactors of DDIT3, most of which were bZIP transcription factors or other transcription coregulators. The FUS interactome, on the other hand, has been published in five recent studies, mainly due to interest in understanding the role of FUS in amyotrophic lateral sclerosis and frontotemporal lobar degeneration. Consistent with the role of FUS as a multifunctional RNA-binding protein, all five of these publications reported significant enrichment of RNA processing proteins (Chi et al., 2018; Kamelgarn et al., 2016; Sun et al., 2015; Tao Wang et al., 2015; Yamazaki et al., 2012).  2.1.3 Affinity purification followed by mass spectrometry Affinity purification requires enrichment of the endogenous or tagged protein of interest (known as the bait protein) and its complexes with specific antibodies or affinity tags. The captured bait protein and its complexes are then processed and analyzed by mass spectrometry for protein identification (Gingras et al., 2007). This approach has the advantage of identifying the interactome of a protein in a high throughput and unbiased way, and under near physiological conditions, compared to in vitro techniques such as the yeast two-hybrid system.  A number of fusion transcription factor targeting strategies discussed earlier stemmed from the identification of novel key interactors in translocation-associated sarcomas that could be targeted - TLE1 and ATF2 with SS18-SSX ( Su et al., 2012), RNA helicase A or NuRD/LSD1 with EWS-FLI1 (Sankar et al., 2013; Toretsky et al., 2006), and CHD4 with PAX3-FOXO1 (Böhm et al., 2016). This leads to the hypothesis that in myxoid liposarcoma, FUS-DDIT3 also exerts its oncogenic function through specific key interactors that are detectable with proteomic technologies. Since FUS-DDIT3 is not targeted by any existing drugs and its interactome has not been systematically studied, the aim behind this chapter was to employ a similar strategy by first identifying the FUS-DDIT3 interactome using an affinity purification - mass spectrometry 39  approach, then evaluating candidate interactors for functional significance in myxoid liposarcoma, to identify those that might best be targeted.  2.2 Materials and methods 2.2.1 Mammalian cell culture and chemicals The following cell lines were kindly provided: myxoid liposarcoma cell lines 402-91 and 1765-92 by Dr. Pierre Aman (University of Gothenburg, Sweden) (Aman et al., 1992), and DL-221 by Dr. Keila Torres (MD Anderson Cancer Center, Houston, TX, USA) (de Graaff et al., 2016). Myxoid liposarcoma cell lines were cultured in RPMI-1640 medium with 10% fetal bovine serum (Life Technologies), with routine verification of FUS-DDIT3 protein expression by western blot analysis. Non-sarcoma control cell lines HeLa and HEK293T were purchased from ATCC, and maintained in DMEM medium with 10% fetal bovine serum. All cells were cultured at 37°C, 95% humidity, and 5% CO2. Regular testing was carried out on all cell lines for mycoplasma infection. Tunicamycin (MilliporeSigma), GNE-495 (MedChemExpress) and oligomycin (Selleckchem) were reconstituted in DMSO.  2.2.2 Immunoprecipitation 402-91, 1765-92 and DL-221 cells were harvested and lysed in lysis buffer (1% Triton X-100, 1 mM MgCl2, 15 mM Tris pH 8.0, 100 mM NaCl) supplemented with Roche cOmplete EDTA-free protease inhibitors and 1 U/mL benzonase (MilliporeSigma) for background reduction, with rotation for 1 hr in the cold room. Whole cell lysates were clarified by centrifugation at maximum speed for 10 min at 4°C, quantified with BCA protein assay (Thermofisher Scientific), and stored on ice until ready for immunoprecipitation. 40  Antibodies used for immunoprecipitations were: normal mouse or rabbit IgG (Santa Cruz), DDIT3 (L63F7) (Cell Signaling Technology, #2895), GNL3 (Abcam, ab70346), MAP4K4 (Bethyl, A3011-502A), NONO (Abcam, ab70335), PSPC1 (Abcam, ab104238), SFPQ (Novus Biologicals, NB-100-61045), SPAG5 (Proteintech, 14726-1-AP), ZFR (Abcam, ab90865), and ZNF638 (Bethyl, A301-548A). Antibodies were first incubated with Dynabeads™ Protein G (ThermoFisher Scientific) for 10 min at room temperature.  For non-mass spectrometry immunoprecipitations, the non-crosslinked antibody-Dynabead mixture was incubated with whole cell lysates overnight with rotation in the cold room. Beads were washed 3 times with cold lysis buffer, followed by 2 times with cold PBS, then eluted by boiling for 5 min in sample loading buffer. For samples destined for mass spectrometry analysis, antibodies were crosslinked to Dynabeads with BS3 (bis(sulfosuccinimidyl)suberate) crosslinker (Thermofisher Scientific) according to manufacturer protocol. The antibody-bead mixture was then pre-blocked with 10 mg/mL BSA in casein blocker (ThermoFisher Scientific) with rotation at room temperature for 30 min, before being incubated with whole cell lysates for 1 hour at room temperature, with rotation. Beads were washed 3 times with cold lysis buffer, followed by 2 times with cold PBS. Proteins were eluted by boiling for 5 min in the SP3 elution buffer (4% SDS, 2% β-mercaptoethanol, 40 mM Tris pH 6.8). Eluted immunoprecipitates were incubated at 45°C for 30 min, then alkylated with 400 mM iodoacetamide for 30 min at 24°C. Reactions were quenched with addition of 200 mM dithiothreitol.  2.2.3 Protein clean up, digestion, mass spectrometry and analysis Eluted proteins were prepared for trypsin digestion using the SP3 cleanup protocol as previously described (Hughes et al., 2014). Acetonitrile was added to the SP3 bead-protein 41  mixture to a final 50% vol/vol, and incubated for 8 min at room temperature. Using a magnetic rack, beads were washed two times with 200 μL 70% ethanol for 30 sec, and once with 180 μL 100% acetonitrile for 15 sec. For digestion, beads were reconstituted in 5 μL 50 mM HEPES pH 8.0 buffer containing trypsin/rLys-C enzyme mix (Promega) at a 1:25 enzyme to protein ratio, and incubated for 14 hr at 37°C. Digested peptides were recovered by removing the supernatant on a magnetic rack. Analysis of peptide samples was carried out on an Orbitrap Fusion Tribrid MS platform (Thermo Scientific). Samples were introduced using an Easy-nLC 1000 system (Thermo Scientific). Columns used for trapping and separations were packed in-house. Trapping columns were packed in 100 μm internal diameter capillaries to a length of 25 mm with C18 beads (Reprosil-Pur, Dr. Maisch, 3 μm particle size). Trapping was carried out for a total volume of 10 μL at a pressure of 400 bar. After trapping, gradient elution of peptides was performed on a C18 (Reprosil-Pur, Dr. Maisch, 1.9 μm particle size) column packed in-house to a length of 15 cm in 100 μm internal diameter capillaries with a laser-pulled electrospray tip and heated to 45°C using AgileSLEEVE column ovens (Analytical Sales & Service). Elution was performed with a gradient of mobile phase A (water and 0.1% formic acid) to 25% B (acetonitrile and 0.1% formic acid) over 32 min, and to 40% B over 4 min, with final elution (80% B) and equilibration (5% B) using a further 4 min at a flow rate of 350 nL/min. Data acquisition on the Orbitrap Fusion was carried out using a data-dependent method. Survey scans covering the mass range of 350 – 1500 Da were acquired at a resolution of 120,000 (at m/z 200), with quadrupole isolation enabled, an S-Lens RF Level of  60%, a maximum fill time of 50 ms, and an automatic gain control (AGC) target value of 5e5. For MS2 scan triggering, monoisotopic precursor selection was enabled, charge state filtering was limited to 2 – 4, an intensity threshold of 5e3 was employed, and dynamic exclusion of previously selected masses 42  was enabled for 60 sec with a tolerance of 20 ppm. MS2 scans were acquired in the ion trap in Rapid mode after HCD fragmentation with a maximum fill time of 150 ms, quadrupole isolation, an isolation window of 1 m/z, collision energy of 35%, activation Q of 0.25, injection for all available parallelizable time turned OFF, and an AGC target value of 4e3. The total allowable cycle time was set to 4 sec. MS1 scans were acquired in profile mode, and MS2 in centroid format. Detected protein abundance was normalized across samples from each cell line. Abundance from each DDIT3 IP triplicate was expressed as a fold change over the IgG IP sample, and the mean of the triplicates was calculated. Only proteins that were detected in all triplicate samples, with a statistically significant (Student’s t-test, p < 0.05) mean fold change of > 1 were selected for further analysis. A second set of proteins that were detected in all the triplicate IP samples, but not in the IgG control were also included for further analysis. The raw abundance of both sets of proteins were normalized across all three cell lines, and combined for ranking in order of relative abundance.  2.2.4 RNA interference siRNA transfection was performed using RNAiMAX (Thermo) via reverse transfection according to the manufacturer’s instructions. Briefly, 10 nM final siRNA concentration was mixed with 0.12 μL RNAiMAX for every 20 μL OptiMEM, and scaled to appropriate volumes to incubate in 96-well plates at room temperature for 20 min. Cells in full culture media were then added to each well and incubated for 72 hr before performing the cell viability assay. MAP4K4 siRNAs used were SMARTpool ON-TARGETplus Human MAP4K4 purchased from Dharmacon (Cat: L-003971-00-0005) and FlexiTube GeneSolution for MAP4K4 from Qiagen (Cat: GS9448). The siGENOME non-targeting siRNA control pool #2 purchased from Dharmacon (Cat: D-001206-14-05) was used as a control. 43  2.2.5 Western blots Protein lysates were separated by 10% SDS-PAGE and transferred to nitrocellulose membranes. Blots were incubated with the indicated primary antibodies at 1:1000 dilution unless otherwise stated: C/EBPβ (Abcam, ab32358), FUS (Santa Cruz, sc-373698), DDIT3 (Abcam, ab10444), GAPDH (Santa Cruz, sc-25778) at 1:2000, GNL3 (Abcam, ab70346), MAP4K4 (Bethyl, A3011-502A), NONO (Abcam, ab70335), PSPC1 (Abcam, ab104238), SFPQ (Novus Biologicals, NB-100-61045), SPAG5 (Proteintech, 14726-1-AP), ZFR (Abcam, ab90865), ZNF638 (Bethyl, A301-548A). Secondary antibodies used for duplexing were: goat anti-mouse IgG (H+L) Alexa Fluor 790 (ThermoFisher Scientific) and goat anti-rabbit IgG (H+L) Alexa Fluor 680 (ThermoFisher Scientific). Western blot signals were visualized on the Odyssey Infrared System (LI-COR Biosciences).   2.2.6 Proximity ligation assay Cells were grown in culture treated chamber slides for 24 hr, then fixed with 2% paraformaldehyde and permeabilized with 0.1% Triton X-100. Cells were blocked with Duolink blocking solution and incubated overnight at 4°C in the cold room with the following primary antibodies: DDIT3 (L63F7) (Cell Signaling Technology, #2895), MAP4K4 (Bethyl, A3011-502A). Proximity ligation was performed with the Duolink® In Situ Red Starter Kit Mouse/Rabbit (MilliporeSigma) according to the manufacturer’s instructions. Goat anti-rabbit Alexa Fluor 488 secondary antibody (Life Technologies) was added during signal amplification to confirm the presence of MAP4K4. Fluorescence was detected using a Zeiss Axiovert microscope at 63X.  44  2.2.7 Cell viability assay Cells were grown in 96 well plates and treated in triplicates with siRNAs or varying concentrations of GNE-495 for 72 hr. Cell viability was assessed as compared with control siRNA or vehicle DMSO control using MTS reagent (Promega) and colorimetric measurements were read with a spectrophotometer at 490 nm. The dose-response curve for GNE-495 treatment was plotted and IC50 values calculated using Graphpad Prism.     2.3 Results 2.3.1 Immunoprecipitation-mass spectrometry discovery of FUS-DDIT3 interactome Endogenous FUS-DDIT3 was immunoprecipitated from three myxoid liposarcoma cell lines (402-91, 1765-92 and DL-221) using an anti-DDIT3 antibody that recognizes the C-terminal portion of DDIT3 that is retained in both FUS-DDIT3 variants from these cell lines. The myxoid liposarcoma cell lines do not express detectable wild type DDIT3 under regular cell culture conditions (Figure 2.1A). Even after treatment with tunicamycin to induce endoplasmic reticulum stress and DDIT3 expression, the levels of FUS-DDIT3 expression far exceed that of induced DDIT3 (Figure 2.1A). Therefore, immunoprecipitation with a DDIT3 antibody should only enrich for FUS-DDIT3 and its interactors, such as C/EBPβ (Figure 2.1B). Triplicate immunoprecipitations with the DDIT3 antibody were performed in benzonase-treated whole cell lysates from each cell line, then processed and analyzed by mass spectrometry (Figure 2.2A). Western blot analysis on the post-IP supernatant showed a reduction in FUS-DDIT3 levels in the IP:DDIT3 triplicates compared to the no antibody or mouse IgG control (Figure 2.2B). To further confirm the immunoprecipitation of FUS-DDIT3, multiple DDIT3 peptides were detected in all triplicate samples and in all cell lines (Figure 2.3A.) FUS peptides were not used as an indication for successful pull-down of the fusion protein due to the presence 45  of abundant wild type FUS in all cell lines. Moreover, the first 7 exons of FUS that are retained in the fusion variant present in 402-91 and DL-221 are poor in arginine and lysine amino acid residues, and would not generate appropriately sized tryptic peptides for MS identification (Figure 2.3B). In fact, FUS peptides (from exons 6 to 14) were only identified in the 1765-92 myxoid liposarcoma cell line (Figure 2.3B), wherein the fusion variant retains the first 13 FUS exons, but these could be attributed to either FUS-DDIT3 or wild type FUS as the fusion protein has been reported to interact with the FET proteins (Thomsen et al., 2013). To curate and reduce the large number of MS-identified proteins for downstream data analysis, proteins were selected from each cell line that were (1) detected only in all IP:DDIT3 triplicates but not in the IP:IgG control, or (2) were present at a statistically significant (Student’s t-test, p < 0.05) positive average fold change in all IP:DDIT3 triplicates over IP:IgG (Table 2.1). Among these proteins, 75 proteins were common in at least 2 cell lines, and were selected for further analysis (Figure 2.4 and Table 2.1). Twelve proteins (including DDIT3) were found to be common in all 3 cell lines (Figure 2.4). These are ranked in Table 2.2 according to their total normalized abundance from all cell lines, followed by the remaining 63 proteins that were detected in any two of the three cell lines. C/EBPβ was detected, serving as a positive control for the presence of FUS-DDIT3 interactors that validates the methodology being used (Table 2.2, Figure 2.1B). However, C/EBPβ was only detected in 402-91 and 1765-92, likely due to the much lower expression of C/EBPβ in DL-221 (Appendix A.1). In conclusion, the IP-MS screen successfully identified proteins that are potentially in the FUS-DDIT3 interactome.  46  2.3.2 The FUS-DDIT3 interactome is enriched for RNA-processing proteins For further analysis of the FUS-DDIT3 interactome, the STRING database (Search Tool for the Retrieval of Interacting Genes/Proteins, https://string-db.org/) (Szklarczyk et al., 2015) was used to search for known functional associations within the network of 75 proteins (Figure 2.5). FUS and DDIT3 (circled by solid lines) are linked by frequent co-mentions in PubMed abstracts in the context of their fusion in myxoid liposarcoma. As expected, there were multiple associations between DDIT3 and its major interacting partner C/EBPβ. Besides FUS and C/EBPβ, PARP1 was the only other protein associated with DDIT3, but only through co-mentions in PubMed abstracts, and not through any existing evidence that they are actually in the same complex. Among the published DDIT3 interactors found on the BioGRID database (https://thebiogrid.org) (Stark et al., 2006), only C/EBPβ and C11orf30 (Figure 2.5, circled by dotted blue lines) were found in the FUS-DDIT3 interactome (Appendix A.2, A.3). Cross referencing the FUS-DDIT3 interactome with data from five recent FUS interactome studies revealed that 21 FUS-DDIT3 interactors are common with the FUS interactome (Appendix A.2 and A.3) (Chi et al., 2018; Kamelgarn et al., 2016; Sun et al., 2015; Tao Wang et al., 2015; Yamazaki et al., 2012), including six that were not found in the STRING analysis (CAPRIN1, MKI67, RBM6, TRA2A, ZFR, ZNF326) (Figure 2.5, circled by dotted red lines). From the STRING analysis and interactome comparisons (Figure 2.5, Appendix A.2, A.3), a larger proportion of proteins in the FUS-DDIT3 interactome have known associations with FUS than with DDIT3.  Two main interaction clusters in the STRING analysis appeared to form around the Drosophila behavior/human splicing proteins splicing factor proline and glutamine rich (SFPQ), paraspeckle component 1 (PSPC1) and non-POU domain containing octamer binding (NONO), as well as around the cleavage and polyadenylation specificity factor (CPSF) 1/6/7 proteins 47  (Figure 2.5). The CORUM database (comprehensive resource of mammalian protein complexes, http://mips.gsf.de/genre/proj/corum/index.html) is a collection of experimentally verified mammalian protein complexes (Ruepp et al., 2010), and was used to identify potential protein complexes within the FUS-DDIT3 interactome. The CORUM analysis confirmed that the main complexes include SFPQ (PSF) and NONO (P54), as well as two complexes involved in 3'-mRNA processing – snRNP-free U1A and CFIIAm complexes (Figure 2.6A). Analysis with the PANTHER database (Protein ANalysis THrough Evolutionary Relationships, http://pantherdb.org) (Mi et al., 2017) showed that the largest protein class in the interactome are the nucleic acid binding proteins (24 proteins), and within this category, two-thirds are RNA-binding proteins (16 proteins) (Figure 2.6B). Despite FUS-DDIT3 being a transcription factor, only five other transcription factors (CEBPB, DIDO1, FUS, PPRC2A, TCF20) were identified in the current screen. Based on gene ontology (GO) classification, most of the FUS-DDIT3 interactome are nuclear proteins (Table 2.3) that are involved in mRNA processing and splicing (Tables 2.4 and 2.5). Thus, results of these bioinformatics analyses performed using the PANTHER and CORUM databases, as well as the GO classifications, suggest that the FUS-DDIT3 interactome is enriched in RNA processing proteins.  2.3.3 Validation of FUS-DDIT3 association with selected candidate proteins The top ranked proteins, NONO and SFPQ (Table 2.2), were selected for additional experimental work to validate their associations with FUS-DDIT3. Additionally, even though it was detected further down in the interactome ranking, PSPC1 was also included in the validation as a member of the Drosophila behavior/human splicing protein family that is often in complex with NONO and SFPQ (Knott et al., 2016). Importantly, results from reciprocal co-48  immunoprecipitation and proximity ligation assays validated the association of FUS-DDIT3 with NONO, PSPC1 and SFPQ (Figures 2.7 and 2.8).  Additional top-ranked proteins (GNL3, SPAG5, ZFR, and ZNF638) were selected for validation. MAP4K4 was also included in these studies as it was one of the few enzymes on the list of putative FUS-DDIT3-interacting proteins that could potentially be targeted with existing agents. Co-immunoprecipitation experiments validated association of FUS-DDIT3 with these proteins in myxoid liposarcoma cells 402-91 (Figure 2.9), 1765-92 (Figure 2.10) and DL-221 (Figure 2.11). Furthermore, proximity ligation assay signals (red) between FUS-DDIT3 and MAP4K4 were observed in the nuclei of all three myxoid liposarcoma cell lines, with only background signals observed outside the nuclei in the biological negative control cell line (HeLa) (Figure 2.12). In summary, all eight selected proteins (GNL3, MAP4K4, NONO, PSPC1, SFPQ, SPAG5, ZFR, ZNF638) were validated through co-immunoprecipitation as part of the FUS-DDIT3 interactome. Additional validation of the native proteins in situ via proximity ligation assay provided further confirmation for all of the four proteins that were also tested using this technique: MAP4K4, NONO, PSPC1 and SFPQ. This indicates that the proteins detected by the IP-MS screen, following further selection on the basis of (1) their consistent identification in at least two out of three cell lines, and (2) high ranking in terms of relative enrichment in the DDIT3 IPs, are likely real interactors of FUS-DDIT3.  2.3.4 MAP4K4 knockdown or inhibition does not affect cell viability MAP4K4 has been reported to function as a negative regulator of adipogenesis (Tang et al., 2006). MAP4K4 can also be targeted by several small molecule inhibitors, although none of the compounds have yet been tested for activity against cell viability or proliferation (Gao et al., 49  2016). To investigate the functional significance for MAP4K4 in myxoid liposarcoma, the response of myxoid liposarcoma cell lines was tested after treatment with GNE-495, a potent and selective inhibitor of MAP4K4 (Ndubaku et al., 2015). A cell viability assay was performed with increasing doses of GNE-495 in 402-91, 1765-92 and DL-221, using HEK293T as a non-malignant control cell line (Figure 2.13). The IC50 values (20.0 µM in 402-91, 61.2 µM in 1765-92, 25.6 µM in DL-221 and 21.4 µM in HEK293T) indicated that these cell lines were not particularly sensitive to MAP4K4 inhibition with GNE-495. Since MAP4K4 has been reported to be a negative regulator of PPARγ expression (Guntur et al., 2010; X. Tang et al., 2006), PPARγ expression was assessed after GNE-495 treatment, and observed to be reduced, similar to MAP4K4 knockdown (Appendix A.4), suggesting successful inhibition of MAP4K4 activity. While GNE-495 had been previously used in functional studies, these were not in the context of cancer, but against retinal neovascularization by inhibiting endothelial cell migration in vitro and angiogenesis in vivo (Ndubaku et al., 2015; Vitorino et al., 2015). As a result, no positive control cell line was available that is known to respond to GNE-495 in a cell proliferation assay. Therefore, an alternative means for targeting MAP4K4 in myxoid liposarcoma was sought, to confirm that the lack of apparent activity against cell viability was not simply an issue with the GNE-495 reagent preparation. To confirm the lack of impact on myxoid liposarcoma proliferation after MAP4K4 inhibition, MAP4K4 expression was reduced for 72 hr using two different MAP4K4 siRNAs (Figure 2.14A), but again no significant change to cell viability was observed (Figure 2.14B). The detection of multiple MAP4K4 bands on the western blots is in line with previous reports of multiple MAP4K4 splice isoforms (Wright et al., 2003). The knockdown, however, was incomplete, and was likely due to much higher levels of MAP4K4 expression in myxoid liposarcoma cells, when compared to HeLa cells (Figure 2.14A). While it cannot be ruled out 50  that residual MAP4K4 expression may allow normal MAP4K4 function through reduced, but sustained cell signaling dynamics, the lack of any apparent impact on cell proliferation by MAP4K4 inhibition would suggest that this was not the case. Together, the results suggest that MAP4K4 does not play a critical functional role in supporting cell proliferation in myxoid liposarcoma.  2.3.5 Assessing MAP4K4 downstream activation of AMPK in myxoid liposarcoma Since MAP4K4 inhibition and knockdown had no effect on myxoid liposarcoma cell proliferation, it is possible that relevant downstream pathways of MAP4K4 may not be active in myxoid liposarcoma. MAP4K4 has been reported in Danai et al. to be upstream of AMP-activated protein kinase (AMPK) following oligomycin treatment in mouse 3T3-L1 adipocytes (Danai et al., 2013). Therefore, myxoid liposarcoma cell lines were treated with oligomycin to induce AMPK phosphorylation on amino acid Thr172 with or without MAP4K4 knockdown (Figure 2.15). Knockdown of MAP4K4 with two different siRNAs resulted in a significant reduction of MAP4K4 expression in all three myxoid liposarcoma cell lines (Figure 2.15A, C, E). In control siRNA-treated 402-91 cells, the western blot intensity of AMPK Thr172 was normalized to total AMPK, and densitometric measurement showed an increased in pAMPK (Thr172) after a 15 min, 500 nM oligomycin treatment. This oligomycin-induced AMPK activation was reduced after MAP4K4 knockdown (Figure 2.15A for western blot, and Figure 2.15B for corresponding densitometry analysis). This suggests that, consistent with the Danai et al. study, MAP4K4 is a positive regulator of AMPK signaling, and that this is true in 402-91 myxoid liposarcoma cells. 51  A similar AMPK activation was induced in 1765-92 after oligomycin treatment in the control siRNA condition, albeit to a lesser degree compared to 402-91. However, unlike in 402-91, MAP4K4 knockdown did not reduce oligomycin-induced AMPK activation, but either had no effect on, or further increased AMPK activation (Figure 2.15C and D). The incongruent results observed from the different MAP4K4 siRNAs suggests that AMPK may not be an active downstream pathway of MAP4K4 in 1765-92. Finally, in DL-221, oligomycin treatment appeared to have no effect on pAMPK in the control siRNA-treated cells, but MAP4K4 knockdown slightly increased oligomycin-induced AMPK activation, suggesting that MAP4K4 was blocking oligomycin-induced AMPK activation in DL-221 (Figure 2.15E and F). The inconsistency among myxoid liposarcoma cell lines in MAP4K4 downstream activation of AMPK signaling suggests that the functional impact of MAP4K4 signaling through the AMPK pathway could be cell line specific, but not disease specific. Although MAP4K4 represents a targetable pathway, this inconsistency across models, combined with the lack of effect on proliferation described in the previous section, suggests it may not be worth further investigation as a targeted therapeutic strategy in myxoid liposarcoma.  2.4 Discussion 2.4.1 Summary Current cytotoxic therapies have limited activity and are ultimately ineffective for metastatic myxoid liposarcoma, and new options are urgently needed. The identification of functionally important and targetable interactors of the fusion proteins has been successfully applied to uncover potential new therapeutic options in other translocation-sarcomas (Erkizan et al., 2009; Laporte et al., 2017; Su et al., 2012; Toretsky et al., 2006). Therefore, in this chapter a 52  similar approach was applied to myxoid liposarcoma, with a hypothesis that FUS-DDIT3 exerts its oncogenic function through specific key interactors detectable with proteomic technologies. The IP-MS screen identified a list of 75 proteins in the putative FUS-DDIT3 interactome (Table 2.2). Co-immunoprecipitation and proximity ligation assays validated eight of these proteins: GNL3, MAP4K4, NONO, PSPC1, SFPQ, SPAG5, ZFR, ZNF638 (Figures 2.7 – 2.11). Further bioinformatic analyses of the FUS-DDIT3 interactome revealed an enrichment of proteins involved in RNA binding and processing (Figure 2.6B, Tables 2.4 and 2.5). Although inhibition or knockdown of one of the interactors, MAP4K4, did not affect myxoid liposarcoma cell viability, the FUS-DDIT3 interactome presented in this chapter provides multiple candidates for future functional validation and identification of key effector partners of FUS-DDIT3.  2.4.2 Insights from the FUS-DDIT3 interactome While a combination of the interactomes from both parent proteins (FUS and DDIT3) would be expected in the FUS-DDIT3 interactome, a disproportionately large number of FUS interactors were detected in the current IP-MS screen. Sixteen (21%) of the 75 proteins are RNA binding proteins, in keeping with the role of FUS as an RNA binding protein. A similar distribution was observed in the EWS-FLI1 interactome in Ewing sarcoma, where close to half (43%) of the proteins in the interactome were involved in RNA splicing and processing, presumably through the EWSR1 portion of the fusion protein (Selvanathan et al., 2015). The disproportionate representation between the FUS and DDIT3 portions of the interactome may be explained by a combination of several possibilities: (1) The first possibility could simply be due to FUS occupying a larger proportion of the amino acid sequences in the fusion protein (Figure 2.3A). 53  (2) FUS-DDIT3 has also been shown to interact with wild type, full-length FUS (Thomsen et al., 2013) (Figure 2.1B), so it would not be surprising for proteins associated with full-length FUS to be detected in the FUS-DDIT3 interactome.  (3) Parental proteins in fusions tend to be interaction-prone and form central nodes in interaction networks (Latysheva et al., 2016), which is certainly the case for the multifunctional FET proteins that are present in many protein complexes. Consequently, it would not be surprising for the FUS-associated interactome to be far larger than that of DDIT3, and hence, result in a larger representation in the FUS-DDIT3 interactome. The relatively large size of the FUS interactome was also alluded to with the combined identification of 413 unique interactors from five previously published interactome studies (Appendix A.3) (Chi et al., 2018; Kamelgarn et al., 2016; Sun et al., 2015; Tao Wang et al., 2015; Yamazaki et al., 2012), although there is relatively little overlap observed between some of those FUS interactomes and the FUS-associated proteins presented here for the FUS-DDIT3 interactome in the myxoid liposarcoma cell context (Appendix A.3 and A.5). The minimal overlap can also be attributed to the different complement of proteins present in different cellular backgrounds, and/or technical limitations due to incomplete proteome coverage, and dynamic range and parameters of different mass spectrometers (Kalli and Hess, 2012; Tsiatsiani and Heck, 2015; Zubarev, 2013). DDIT3 binds bZIP proteins from the C/EBP, CREB/ATF and AP-1 family of transcription factors (Ron and Habener, 1992; N. Su and Kilberg, 2008; Ubeda et al., 1999), and a recent in vitro screen showed direct interaction of DDIT3 with up to 14 other bZIP proteins (Reinke et al., 2013). A total of 76 unique DDIT3 interactors are present in the BioGRID database – besides the bZIP proteins, many of the remaining interactors are transcription factors or coregulators. The lack of known DDIT3 binding partners in the FUS-DDIT3 interactome is likely due to C/EBPβ being the major and preferred bZIP interactor for FUS-DDIT3 in myxoid 54  liposarcoma, and/or higher C/EBPβ expression in myxoid liposarcoma compared to other DDIT3 interactors. Transcription factors are generally expressed at much lower levels compared to other protein classes (Beck et al., 2011), especially the highly abundant RNA processing proteins, and may only associate with FUS-DDIT3 transiently (Swift and Coruzzi, 2017). Consequently, the abundance of other DDIT3 interactors may be too low in the immunoprecipitates for their peptides to be prioritized and sampled in the mass spectrometer for identification. The physical fusion of FUS to DDIT3 may also affect the binding affinity of normal DDIT3 interactors to the fusion protein.  The interactome presented here is by no means exhaustive, and did not include known FUS-DDIT3 interactors such as CDK2 and NFKBIZ (Bento et al., 2009; Göransson et al., 2009). A major reason is technical, as mentioned earlier, due to limits in protein coverage and dynamic sensitivity in the mass spectrometry workflow (Ponomarenko et al., 2016). The antibody-epitope binding could also affect binding of some proteins to FUS-DDIT3, a limitation that could be mediated by using different antibody clones that recognize other epitopes. Another limitation is the use of benzonase for background reduction in the IP-MS data, due to the promiscuous binding of FUS to RNA and DNA (Wang et al., 2015), which may increase non-specific RNA-mediated binding to other proteins, but comes as a trade-off for the loss of important interactors that may rely on specific DNA-protein contact. An example for such an interaction would be the DNA-dependent binding of the PAX3-FOXO1 in alveolar rhabdomyosarcoma to its functionally important epigenetic coregulator CHD4 (Böhm et al., 2016). Finally, there is the important caveat that proteins identified in any high throughput interactome screens should be considered only as representing potential interactors until validated by a separate experimental techniques in additional studies. 55  The fusion junction of Type 1 FUS-DDIT3 in two of the cell lines (402-91 and DL-221) contains 26 newly translated amino acids from the DDIT3 5'UTR (Crozat et al., 1993), and could theoretically provide a binding site for novel non-FUS or DDIT3 interactors, but this seemed unlikely as an InterPro database (http://www.ebi.ac.uk/interpro/) analysis did not find any known binding domain or sites within the fusion junction sequence (Appendix A.6). Moreover, the Type 11 FUS-DDIT3 fusion variant misses this section of 5'UTR sequence from DDIT3 exon 2 (as illustrated in Chapter 1, Figure 1.3), which argues against any critical biological function for this region in myxoid liposarcoma oncogenesis, as the disease can occur in its absence.  2.4.3 Functional validation of FUS-DDIT3 interactors Poly (ADP-ribose) polymerase 1 (PARP1) was one of the few proteins in the FUS-DDIT3 interactome that is targetable by drugs (PARP inhibitors) that are already approved or currently in clinical trials (Tangutoori et al., 2015). However, published data suggests that PARP inhibitors are ineffective against myxoid liposarcoma cells, although the data did not test PARP inhibitor sensitivity in combination with DNA damaging agents (Garnett et al., 2012). MAP4K4 was subsequently evaluated in this study for functional significance in myxoid liposarcoma for several reasons. Firstly, it is a kinase that can be inhibited by existing compounds (Ammirati et al., 2015; Ndubaku et al., 2015; Schröder et al., 2015). MAP4K4 has been suggested to promote transformation and invasion, and is associated with poor prognosis in colorectal cancer, hepatocellular carcinoma, pancreatic ductal adenocarcinoma, lung adenocarcinoma, and prostate cancer (Gao et al., 2016). Finally, of particular interest is the role of MAP4K4 as a negative regulator of adipocytic differentiation by functioning in a negative feedback loop that regulates PPARγ expression (Guntur et al., 2010; Tang et al., 2006). 56  Although MAP4K4 knockdown is anti-proliferative in hepatocellular carcinoma, colorectal carcinoma and gastric cancer (Han et al., 2010; Liu et al., 2016; Wang et al., 2015), knockdown or inhibition of MAP4K4 with the selective compound GNE-495 had no effect on cell proliferation in myxoid liposarcoma cells (Figures 2.13A and 2.14). Nevertheless, since MAP4K4 is also a positive regulator of invasion, migration and angiogenesis (Han et al., 2010; Liu et al., 2016; Ndubaku et al., 2015; Wang et al., 2015), it could be interesting to follow up on whether this would apply to myxoid liposarcoma through in vitro migration and invasion assays, or even in vivo xenograft model work to assess angiogenesis after MAP4K4 knockdown. Myxoid liposarcoma expresses more MAP4K4 compared to HeLa (Figure 2.14A), and public microarray data has shown MAP4K4 overexpression in myxoid liposarcoma (Appendix A.7) (Singer et al., 2007). This is in line with reports of MAP4K4 overexpression in multiple cancer types (Wright et al., 2003). PPARγ expression is also relatively high in myxoid liposarcoma tissues based on immunohistochemical analyses (Cheng et al., 2009), which seems paradoxical to existing reports of MAP4K4 as a negative regulator of PPARγ expression (Guntur et al., 2010; X. Tang et al., 2006). Adding to the challenge of studying MAP4K4 is the fact that the cell line models express very little PPARγ transcript or protein compared to the high expression observed in myxoid liposarcoma tissues (Cheng et al., 2009; Renner et al., 2013; Singer et al., 2007) (Appendix A.8), suggesting that the cell lines may not faithfully recapitulate the human in vivo disease in this aspect, and may not be an appropriate model for studying the function of MAP4K4 in myxoid liposarcoma.  57  2.4.4 Future work The large number of RNA processing proteins identified in the FUS-DDIT3 interactome begs the question of whether FUS-DDIT3 plays any role in deregulating normal RNA processing, a topic that will be addressed in Chapter 3. Furthermore, a likely mechanism of action is for FUS-DDIT3 to bridge the large variety of N-terminal FUS interactors, which includes RNAPII (Rapp et al., 2002) and possibly other functionally critical transcriptional regulators, to DNA targets of DDIT3-C/EBPβ dimers that would otherwise be regulated differently in the absence of FUS-DDIT3. This bridging (or scaffolding) function is seen in other translocation-associated sarcomas, where TLE1 (through the SSX portion of SS18-SSX) recruits polycomb repressors to inappropriately silence ATF2 targets (through the SS18 portion) in synovial sarcoma (Su et al., 2012), or where EWS-FLI1 recruits the SWI/SNF chromatin remodeling complex (through the EWSR1 portion) to neomorphic GGAA repeats in gene enhancers that have no regulatory potential in other cell types (Boulay et al., 2017; Riggi et al., 2014). The FUS-DDIT3 interactome identified from whole cell lysates of myxoid liposarcoma cell lines provides a valuable starting point for systematic evaluation of whether and how any of the interactors contribute to the oncogenic function of FUS-DDIT3. For example, SFPQ, one of the top hits in the interactome, is known to function as a transcriptional repressor by recruiting Sin3A and HDAC (reviewed in Knott et al., 2016). Incidentally, HDAC inhibition has shown promise as a targeted therapy in myxoid liposarcoma - a HDAC inhibitor (pracinostat) trial that included five myxoid liposarcoma patients showed stable disease in three of the four assessable patients, although the trial had to be ended prematurely due to prolonged unavailability of the drug (Chu et al., 2015). 58  Identification of the transcription regulatory complexes surrounding FUS-DDIT3 may provide a clearer starting point to identifying functionally important and targetable interactors. Besides exploring the extended network of the current interactome (for example, finding Sin3A and HDAC through SFPQ), another approach would be to increase the direct identification of transcription factors and coregulators in the FUS-DDIT3 interactome by reducing sample complexity for downstream mass spectrometry analyses. This will be addressed in Chapter 4.  59  A    B   Figure 2.1 Wild type DDIT3 is not expressed in myxoid liposarcoma cell lines (A) Western blot analysis for expression of FUS-DDIT3 and wild-type DDIT3 in myxoid liposarcoma cells 402-91, 1765-92 and DL-221 after treatment with 5 mg/mL tunicamycin for 4 hr. Cyan colour seen bands indicate saturation of signal during blot scanning. (B) Immunoprecipitation of FUS-DDIT3 from whole cell lysates of 402-91 myxoid liposarcoma cells using an anti-DDIT3 antibody (clone L63F7). The top blot is a merged image of bands detected by both FUS (green) and DDIT3 (red) antibodies, showing the FUS-DDIT3 band in yellow.    –5 mg/mL tunicamycin + – + – +DL-221IB: DDIT3Wild type DDIT3Type 1 FUS-DDIT3Type 8 FUS-DDIT3Myxoid liposarcoma cell lines50403570100150MW (kDa)1765-92402-91Wild type FUSInputIP: mouse IgGIP: DDIT3IB: FUSIB: FUS (green), DDIT3 (red)402-91IB: C/EBPβLAP*LAPIB: DDIT360  A  B  Figure 2.2 Experimental outline of IP-MS experiment and confirmation of successful FUS-DDIT3 IP (A) Workflow for each cell line. The same experiment is repeated across three myxoid liposarcoma cell lines 402-91, 1765-92 and DL-221 (i.e. each cell line was run in triplicate). (B) Western blot of post-IP supernatant to assess immunoprecipitation efficiency. The top blot is a merged image of bands detected by both FUS (green) and DDIT3 (red) antibodies, showing FUS-DDIT3 bands in yellow.   Lyse with 1% Triton X-100 buffer, get whole cell lysate.Treat with Benzonase 1hr @ 4°C. Spin down. IP: 1hr @ room temperatureCrosslink DDIT3 Ab to Dynabeads.Block with BSA and casein.Myxoid liposarcoma cell lineWash beadsElute proteins with SDS and boilingMass spectrometry analysis(Orbitrap Fusion)Reduction, alkylation, on-bead trypsin digestion, peptide clean-upIP: mo-IgG IP: DDIT3Triplicate (1)IP: DDIT3Triplicate (2)IP: DDIT3Triplicate (3)IB: FUSIB: DDIT3IB: FUS (green), DDIT3 (red)InputNo Antibody ControlMouse IgGControlDDIT3 (Replicate 1)DDIT3 (Replicate 2)DDIT3 (Replicate 3)Post-IP SupernatantInputNo Antibody ControlMouse IgGControlDDIT3 (Replicate 1)DDIT3 (Replicate 2)DDIT3 (Replicate 3)Post-IP SupernatantInputNo Antibody ControlMouse IgGControlDDIT3 (Replicate 1)DDIT3 (Replicate 2)DDIT3 (Replicate 3)Post-IP Supernatant402-91 1765-92 DL-22161  A       62  B   Figure 2.3 Amino acid sequences of FUS and DDIT3 tryptic peptides detected from mass spectrometry Red bars indicate lysine (K) or arginine (R) residues for trypsin digestion. Green arrows indicate corresponding amino acid sequence detected from different cell lines. (A) Amino acid sequence of Type 1 FUS-DDIT3. (B) Amino acid sequence of full length FUS.   Indicates trypsin digestion site: Arginine (R) or Lysine (K) amino acidIndicates peptides identified through mass spectrometry63  A B   Figure 2.4 Partial overlap in potential FUS-DDIT3 interactors identified from three myxoid liposarcoma cell line (A) Venn diagram visualization of the number of potential FUS-DDIT3 interactors identified from the IP-MS screen in myxoid liposarcoma cells 402-91 (n=387, pink circle), 1765-92 (n=202, green circle) and DL-221 (n=121, blue circle). Only proteins that fit the specified cutoff criteria were included. The venn diagram indicates the number of identified proteins that were common between cell lines. For example, of the 387 proteins identified in the 402-91 myxoid liposarcoma cell line, 316 proteins were identified only in 402-91 (indicated by the pink section of the venn diagram without any overlap with the green or blue circles). 46 proteins were common in both 402-91 and 1765-92 but not DL-221 (pink and green circle overlap, without any overlap from blue circle for DL-221); 13 proteins were common in 402-91 and DL-221 (purple region where the pink and blue circles overlap but excludes overlap from green circle for 1765-92); 12 proteins were commonly identified in all three cell lines (middle region where all 3 cricles overlap). The dotted triangle indicates the 75 proteins that were common in at least 2 cell lines. (B) The same data from the venn diagram is visualized in an UpSet plot to indicate the number of proteins that are common between any two or three cell lines as indicated by black dots joined by a vertical line (for the first four bars on the left) or were only found in one of the cell lines (last three bars on the right). The brown horizontal line at the bottom of the plot highlights total number of proteins identified in at least two cell lines.  64    Figure 2.5: FUS-DDIT3 interactome network STRING database analysis showing interactions among putative proteins within the FUS-DDIT3 interactome based on co-mentions in PubMed abstracts, experimental data, evidence for specific binding, co-expression or association in curated databases. Only 74 proteins are shown here as RECQL4 was not found in the STRING database. Bait proteins FUS and DDIT3 are circled with solid lines. Proteins reported in other published FUS interactomes but not linked in the current network are circled by red dotted lines. The blue dotted line indicates a DDIT3 interactor found in the BioGRID database, but not connected in the STRING network here.    Co-mentioned in PubMed AbstractsExperimental / biochemical dataEvidence for specific bindingCo-expressionAssociation in curated databasesLegend65  A  B  Figure 2.6: Enriched protein complexes and protein classes in FUS-DDIT3 interactome (A) Five protein complexes identified through the CORUM database are presented with their names as assigned in the database – PSF (SFPQ), p54 (NONO). Numbers after each bar indicate the number of complex members found in interactome / total number of complex members in database. (B) The PANTHER protein classification of FUS-DDIT3 interactome shows percentage of each protein class against the total number of proteins with a class hit (n=47). The nucleic acid binding protein class is further broken down into three different sub-classes (DNA-binding, RNA-binding and nuclease) (n=24).  0 1 2 3 4 5CF IIAm complex (Cleavage factor IIAm complex)snRNP-free U1A (SF-A) complexp54(nrb)-PSF-matrin3 complexTOP1-PSF-P54 complexPSF-p54(nrb) complex-log10 (padj-value)CORUM protein complexes total n = 724/162/32/42/32/266    Figure 2.7 Reciprocal co-immunoprecipitation validates FUS-DDIT3 association with NONO, PSPC1, SFPQ Whole cell lysates from two myxoid liposarcoma cell lines, 402-91 and 1765-92, each harboring a different variants of the FUS-DDIT3 fusion oncogene, were used to test for reciprocal co-immunoprecipitation of the Drosophila behavior/human splicing proteins NONO, PSPC1 and SFPQ with the FUS-DDIT3 oncoprotein. For color blots, FUS was detected in the green channel and DDIT3 in the red channel.    67   Figure 2.8 Proximity ligation assay validations association of  FUS-DDIT3 with Drosophila behavior/human splicing proteins NONO, PSPC1 and SFPQ Proximity ligation assay, a technique that amplifies a red signal when two proteins colocalize within 40 nm of each other in situ, was performed in myxoid liposarcoma cell lines 402-91, 1765-92, DL-221, and negative control cell line HeLa. Red signals indicate proximity between proteins of interest. Nuclei were counter-stained with DAPI (blue). Scale bar = 5 µm.   402-911765-92DL-221HeLaDDIT3 + SFPQ DDIT3 + NONO DDIT3 + PSPC168      Figure 2.9 Reciprocal co-immunoprecipitation validates interaction of selected IP-MS-identified proteins with FUS-DDIT3 in 402-91 myxoid liposarcoma cells Whole cell lysates from myxoid liposarcoma cells 402-91 were used for reciprocal co-immunoprecipitation between FUS-DDIT3 and nucleostemin (GNL3), HGK (MAP4K4), astrin (SPAG5), ZFR, or ZNF638. For color blots, FUS was detected in the green channel and DDIT3 in the red channel.  69     Figure 2.10 Reciprocal co-immunoprecipitation validates interaction of selected IP-MS-identified proteins with FUS-DDIT3 in 1765-92 myxoid liposarcoma cells Whole cell lysates from myxoid liposarcoma cells 1765-92 were used for reciprocal co-immunoprecipitation between FUS-DDIT3 and nucleostemin (GNL3), HGK (MAP4K4), astrin (SPAG5), ZFR, or ZNF638. For color blots, FUS was detected in the green channel and DDIT3 in the red channel. 70    Figure 2.11 Reciprocal co-immunoprecipitation validates interaction of selected IP-MS-identified proteins with FUS-DDIT3 in DL-221 myxoid liposarcoma cells Whole cell lysates from myxoid liposarcoma cells DL-221 were used for reciprocal co-immunoprecipitation between FUS-DDIT3 and HGK (MAP4K4). The same lysates were also used for co-immunoprecipitation of nucleostemin (GNL3), astrin (SPAG5) or ZNF638 with FUS-DDIT3. For color blots, FUS was detected in the green channel and DDIT3 in the red channel. A smaller number of proteins were selected for validation in the DL-221 cell line compared to 402-91 (Figure 2.9) and 1765-92 (Figure 2.10) for methodological reasons – the DL-221 cell line has a much slower growth rate and insufficient protein yield from lysates to provide enough material for the immunoprecipitation experiments.   71     Figure 2.12 Proximity ligation assay validations association of FUS-DDIT3 with MAP4K4 Proximity ligation assay was performed in myxoid liposarcoma cells 402-91, 1765-92, DL-221, and biological negative control cell HeLa. Red signals indicate proximity between MAP4K4 and DDIT3. Immunofluorescence signal for MAP4K4 is in green, nuclei are stained blue with DAPI.  402-911765-92DL-221MAP4K4 MAP4K4 / DDIT3 DAPI MergedMAP4K4 MAP4K4 / DDIT3 DAPI MergedMAP4K4 MAP4K4 / DDIT3 DAPI MergedHeLaMAP4K4 MAP4K4 / DDIT3 DAPI Merged72     Figure 2.13 MAP4K4 inhibition does not affect myxoid liposarcoma cell viability Treatment of myxoid liposarcoma cell lines 402-91, 1765-92, DL-221, and negative control cell line HEK293T with increasing concentrations of GNE-495 for 72 hr. IC50 values (µM) for each cell are indicated. Assessment of MAP4K4 inhibition is in further shown in Appendix A.4.   -12 -10 -8 -6 -4406080100120140log[GNE-495], M% Cell ViabilityGNE-495log-dose vs response402-911765-92DL-221HEK293TIC50 (uM) 402-91 19.951765-92 61.17DL-221 25.59HEK293T 21.3773  A   B   Figure 2.14 siRNA-mediated MAP4K4 knockdown does not affect viability of myxoid liposarcoma cells (A) Western blot to assess reduction of MAP4K4 in myxoid liposarcoma cell lines 402-91, 1765-92, DL-221, and negative control cell line HeLa. Cells were treated for 72 hr with 10 nM of control or pooled MAP4K4 siRNA from Dharmacon (D) or Qiagen (Q). GAPDH serves as loading control. (B) Cell viability was assessed by MTS assay after 72 hr MAP4K4 knockdown.   10 nM siRNA402-91ControlMAP4K4 (D)ControlControlControl1765-92 DL-221 HeLaIB: MAP4K4IB: GAPDHMAP4K4 (Q)MAP4K4 (D)MAP4K4 (Q)MAP4K4 (D)MAP4K4 (Q)MAP4K4 (D)MAP4K4 (Q)4 02 -9 11 76 5-92D L-22 1H eL a05 01 0 01 5 0M A P 4 K 4  % v ia b ilityC e ll L in e% Cell ViabilityC tr l s iR N AM A P 4 K 4  s iR N A  (D )M A P 4 K 4  s iR N A  (Q )74     Figure 2.15 MAP4K4 knockdown does not affect oligomycin-induced AMPK activation Western blot analysis of AMPK phosphorylation at Thr172 (red) normalized against total AMPK (green) after oligomycin treatment, with or without MAP4K4 knockdown using two different MAP4K4 siRNAs from Dharmacon (D) or Qiagen (Q). MAP4K4 protein reduction was assessed by western blot, with GAPDH serving as loading control. Blots are shown for myxoid liposarcoma cells (A) 402-91, (C) 1765-92 and (E) DL-221. Densitometric analysis was performed on the amount of phospho-AMPK signal relative to total AMPK in (B) 402-91, (D) 1765-92 and (F) DL-221. – + – + – +Ctrl MAP4K4 (D) MAP4K4 (Q) 10 nM siRNA500 nM oligomycin (15 mins)IB: GAPDHIB: MAP4K4IB: AMPK (green), pAMPK (red)IB: AMPKIB: pAMPK (Thr172)402-91A BControlMAP4K4 (D)MAP4K4 (Q)01234402-91siRNAFold change in pAMPK expression relative to total AMPKDMSOOligomycin– + – + – +Ctrl MAP4K4 (D) MAP4K4 (Q) 10 nM siRNA500 nM oligomycin (15 mins)IB: GAPDHIB: MAP4K4IB: AMPK (green), pAMPK (red)IB: AMPKIB: pAMPK (Thr172)1765-92C DControlMAP4K4 (D)MAP4K4 (Q)01231765-92siRNAFold change in pAMPK expression relative to total AMPKDMSOOligomycinDL-221– + – + – +Ctrl MAP4K4 (D) MAP4K4 (Q) 10 nM siRNA500 nM oligomycin (15 mins)IB: GAPDHIB: MAP4K4IB: AMPK (green), pAMPK (red)IB: AMPKIB: pAMPK (Thr172)E FControlMAP4K4 (D)MAP4K4 (Q)0.00.51.01.52.0DL-221siRNAFold change in pAMPK expression relative to total AMPKDMSOOligomycin75  Table 2.1 Selection of IP-MS-identified proteins for further analysis based on their enrichment in at least 2 cell lines DDIT3 immunoprecipitation Ticks indicates presence of a protein at high (✔) or low (✓) abundance.  402-91 1765-92 DL-221 Present in no. of cell lines No. of proteins Total IgG DDIT3 1 DDIT3 2 DDIT3 3 IgG DDIT3 1 DDIT3 2 DDIT3 3 IgG DDIT3 1 DDIT3 2 DDIT3 3 Present in IP:DDIT3 only   ✔  ✔  ✔    ✔  ✔  ✔    ✔  ✔  ✔  3 12 75 Also present in IP:IgG ✓ ✔  ✔  ✔  ✓ ✔  ✔  ✔  ✓ ✔  ✔  ✔  Present in IP:DDIT3 only 	 ✔  ✔  ✔  	 ✔  ✔  ✔      2 63 Also present in IP:IgG ✓ ✔  ✔  ✔  ✓ ✔  ✔  ✔      Present in IP:DDIT3 only   ✔  ✔  ✔      	 ✔  ✔  ✔  Also present in IP:IgG ✓ ✔  ✔  ✔      ✓ ✔  ✔  ✔  Present in IP:DDIT3 only 	 	 	   	 ✔  ✔  ✔  	 ✔  ✔  ✔  Also present in IP:IgG 	 	 	 	 ✓ ✔  ✔  ✔  ✓ ✔  ✔  ✔  76  Table 2.2 List of FUS-DDIT3-interacting proteins identified by IP-MS in at least two myxoid liposarcoma cell lines (402-91, 1765-92, DL-221) Rank1 Accession Gene Name Detected in no. of cell lines No. of peptides Normalized abundance2 402-91 1765-92 DL-221 402-91 1765-92 DL-221 Total from all cell lines 1 Q15233 NONO 3 16 19 19 48.0 80.9 15.6 143.6 2 P23246 SFPQ 3 27 28 25 0.8 100.0 15.7 116.4 3 Q96KR1 ZFR 3 21 27 22 2.5 58.2 6.3 66.9 4 P35638 DDIT3 3 4 4 1 19.0 44.0 3.9 66.9 5 Q96R06 SPAG5 3 23 28 10 7.7 7.7 0.7 16.1 6 Q14966 ZNF638 3 52 58 13 0.7 13.5 0.4 14.6 7 Q9Y4B5 SOGA2 3 74 52 40 2.0 6.3 1.8 10.1 8 Q5BKZ1 ZNF326 3 15 19 5 1.3 7.8 0.6 9.6 9 Q7L590 MCM10 3 17 18 4 1.1 6.3 0.4 7.7 10 Q9UH17 APOBEC3B 3 6 8 3 2.4 3.9 0.5 6.8 11 Q69YN4 KIAA1429 3 13 34 7 0.7 4.5 0.6 5.8 12 Q9BVP2 GNL3 3 10 5 5 3.3 1.6 0.4 5.2 13 Q15058 KIF14 2 25 28 4 91.2 2.3  93.5 14 P35637 FUS 2 2 9 2  50.4 0.2 50.6 15 Q96PK6 RBM14 2 20 22 17 0.4 41.9  41.2 16 J3QR09 RPL19 2 3 3 5 15.4  19.1 34.6 17 Q5T9A4 ATAD3B 2 10 17 12 1.8 23.3  25.0 18 P02545 LMNA 2 8 12 21 22.8 1.2  23.0 19 Q9UIG0 BAZ1B 2 27 19 5 19.5 1.9  21.4 20 P14866 HNRNPL 2 15 13 13 0.9 16.6  17.5 21 Q9HAU0 PLEKHA5 2 36 34 15 14.2  1.0 15.2 22 Q13595 TRA2A 2 2 5 2 4.9 8.0  13.9 23 Q8WUT1 POLDIP3 2 6 7 6 2.0  10.5 12.5 24 P17676 CEBPB 2 5 3  4.9 5.2  10.1 25 Q14444 CAPRIN1 2 6 14  0.8 9.0  9.8 26 Q86SQ0 PHLDB2 2 25 40 22 3.4 5.5  8.8 77  27 A0A087WV66 MKI67 2 40 10 15 5.8  0.4 6.2 28 Q8WXF1 PSPC1 2 10 18 6 2.4 3.1  5.5 29 Q8N684 CPSF7 2 7 10 5 2.8 2.6  5.3 30 Q5SW79 CEP170 2 54 45 39 0.2  5.0 5.2 31 P09874 PARP1 2 18 19 30 0.7 3.7  4.4 32 Q16537 PPP2R5E 2 3 2 4 4.0  0.2 4.2 33 P49790 NUP153 2 15 15  3.7 0.2  4.0 34 Q13415 ORC1 2 7 3  3.1 0.6  3.8 35 Q9BTC0 DIDO1 2 18 32 3 1.8 1.9  3.7 36 Q8IX01 SUGP2 2 19 21  0.3 3.2  3.6 37 H3BQZ7 HNRNPUL2-BSCL2 2 5 12  1.4 1.8  3.3 38 Q15678 PTPN14 2 16 15 5 1.5 1.7  3.2 39 P51608 MECP2 2 9 15 9 1.7  1.2 2.9 40 Q8WUU5 GATAD1 2 3 4  1.4 1.3  2.7 41 M0R1G4 NAV1 2 5 2  2.4 0.2  2.6 42 A0A0A0MTL4 NAV2 2 20 14  1.9 0.6  2.6 43 P78332 RBM6 2 10 24  0.5 1.9  2.3 44 Q99569 PKP4 2 16 26 2 0.9 1.4  2.3 45 F8WJN3 CPSF6 2 3 1  1.7 0.5  2.2 46 Q9NPG3 UBN1 2 9 17  0.8 1.2  2.0 47 O14578 CIT 2 10 19  0.9 1.1  2.0 48 Q9UPN4 AZI1 2 19 11  0.8 1.1  1.9 49 Q6UN15 FIP1L1 2 4 6  0.7 1.2  1.9 50 Q6DN90 IQSEC1 2 5 2  1.6 0.1  1.7 51 P48634 PRRC2A 2 55 55 22 1.0  0.8 1.7 52 Q7Z589 C11orf30 2 23 24 14 1.4  0.1 1.5 53 Q10570 CPSF1 2 2 6  0.4 1.0  1.4 54 O00411 POLRMT 2 5 4 2 1.2 0.2  1.4 78  55 Q9Y678 COPG1 2 4 6 8 0.6  0.8 1.4 56 J3QT28 BUB3 2 4 5 3 1.0  0.3 1.3 57 Q9Y2U8 LEMD3 2 5 4  0.7 0.6  1.3 58 O95232 LUC7L3 2 2 3  0.4 0.8  1.2 59 V9GY01 KNSTRN 2 1 3  0.4 0.7  1.1 60 Q96ST2 IWS1 2  4 5  0.3 0.8 1.1 61 A0A087WZ30 RECQL4 2 10 3  0.7 0.4  1.1 62 A0A087WTX2 SLC39A7 2  1 2  0.1 0.8 1.0 63 Q63ZY3 KANK2 2 6 7 9 0.5 0.4  0.9 64 Q96QT6 PHF12 2 7 13  0.6 0.2  0.9 65 O14974 PPP1R12A 2 3 8 31 0.0 0.7  0.8 66 E7EN19 MAP4K4 2 5 8  0.3 0.4  0.8 67 A0A5E8 GAS2L1 2 5 4  0.3 0.5  0.8 68 P17812 CTPS1 2 5 2 4 0.3  0.3 0.7 69 Q5JR04 MOV10 2 4 7 1 0.3 0.4  0.7 70 Q9UGU0 TCF20 2 23 16  0.5 0.1  0.6 71 Q8N556 AFAP1 2 1 1 7 0.5 0.0  0.5 72 Q6UXN9 WDR82 2 1  2 0.3  0.3 0.5 73 Q2M1P5 KIF7 2 5 7 1  0.2 0.0 0.3 74 Q8N0Z3 SPICE1 2 11 1  0.2 0.0  0.2 75 Q9UHI6 DDX20 2 2 2 5 0.1 0.1  0.2 1 Proteins are first ranked by the number of cell lines there were detected in the IP-MS screen, followed by the normalized total abundance. 2 The relative abundance was normalized by transforming values to the same scale across all three cell lines.   79  Table 2.3 Gene ontology classification of FUS-DDIT3 interactome for cellular component   STRING Cellular Component (GO)     #pathway ID pathway description observed gene count false discovery rate GO.0005654 nucleoplasm 38 4.68E-12 GO.0043232 intracellular non-membrane-bounded organelle 39 1.99E-10 GO.0031981 nuclear lumen 37 1.47E-09 GO.0044428 nuclear part 38 3.02E-09 GO.0070013 intracellular organelle lumen 38 4.51E-08 GO.0044451 nucleoplasm part 16 1.23E-07 GO.0042382 paraspeckles 4 3.85E-07 GO.0044446 intracellular organelle part 48 8.31E-06 GO.0005849 mRNA cleavage factor complex 4 1.04E-05 GO.0016604 nuclear body 10 2.80E-05 GO.0005634 nucleus 43 9.34E-05 GO.0005694 chromosome 13 0.000103 GO.0016363 nuclear matrix 6 0.000104 GO.0044427 chromosomal part 12 0.000197 GO.0016607 nuclear speck 7 0.00033     80  Table 2.4 Gene ontology classification of FUS-DDIT3 interactome for biological processes   STRING Biological Processes (GO)     #pathway ID pathway description observed gene count false discovery rate GO.0006397 mRNA processing 15 8.55E-08 GO.0016071 mRNA metabolic process 16 5.17E-07 GO.0008380 RNA splicing 13 5.81E-07 GO.0007017 microtubule-based process 14 2.93E-06 GO.0006396 RNA processing 16 4.18E-06 GO.0016070 RNA metabolic process 31 0.000135 GO.0090304 nucleic acid metabolic process 33 0.000148 GO.0000226 microtubule cytoskeleton organization 10 0.000291 GO.0070507 regulation of microtubule cytoskeleton organization 7 0.000291 GO.0006139 nucleobase-containing compound metabolic process 34 0.000416 GO.0000398 mRNA splicing, via spliceosome 8 0.000778 GO.0010564 regulation of cell cycle process 11 0.0013 GO.0051726 regulation of cell cycle 14 0.00192 GO.0010467 gene expression 30 0.00224 GO.0051983 regulation of chromosome segregation 5 0.00224     81  Table 2.5 Gene ontology classification of FUS-DDIT3 interactome for molecular function   STRING Molecular Function (GO)     #pathway ID pathway description observed gene count false discovery rate GO.0044822 poly(A) RNA binding 30 1.74E-15 GO.0003676 nucleic acid binding 38 2.52E-08 GO.0000166 nucleotide binding 27 4.78E-06 GO.0003677 DNA binding 23 0.000834  82  Chapter 3: Assessing involvement of FUS-DDIT3 in alternative splicing 3.1 Introduction 3.1.1 RNA processing During and after transcription, all messenger RNAs undergo tightly coupled processing events that add a layer of regulation that can affect export, localization, translation and stability of the transcript. The major processing events include capping, splicing and polyadenylation, each of which is described below.  3.1.1.1 5’ capping The first modification made to any RNA polymerase II (RNAPII)-transcribed RNA is 5' capping. The formation of the 7-methylguanosine cap at the 5' end of nascent transcripts takes place co-transcriptionally as soon as the first 25-30 nucleotides are incorporated by RNA polymerase II (RNAPII) (Moteki and Price, 2002; Shatkin and Manley, 2000). This cap structure protects mRNA transcripts from exoribonucleolytic degradation (Murthy et al., 1991), but also interacts with cap-binding proteins that mediate additional functions. The most well characterized nuclear cap-binding protein is the cap-binding complex, which recruits a spectrum of factors to 7-methylguanosine capped transcripts to mediate efficient pre-mRNA splicing, poly(A) site cleavage at the 3' end, nuclear export, and the pioneer round of translation (Gonatopoulos-Pournatzis and Cowling, 2014). In the cytoplasm, after completion of the pioneer round of translation, the cap-binding complex is then replaced by the eIF4E complex, which starts steady-state rounds of translation that provide the bulk of cellular protein synthesis (Maquat et al., 2010).  83  3.1.1.2 Splicing Splicing of pre-mRNA is required for the maturation of almost all mammalian mRNAs, and is the process of intron removal and concurrent ligation of the flanking exons in the order that they appear in a gene. Alternative splicing is a deviation from this preferred sequence through the inclusion or exclusion of specific exons, resulting in various forms of mature mRNA. Pre-mRNA splicing is mediated by the spliceosome, which is a dynamic and flexible macromolecular machine mainly composed of the U1, U2, U4, U5 and U6 small nuclear ribonucleic proteins (snRNP) in conjunction with a large number of accessory proteins (Will and Lührmann, 2011). The spliceosome has a catalytic center composed of RNA, and can be considered a ribozyme (Fica et al., 2013). There are two major steps in the splicing process: de novo assembly of the spliceosome around each intron in a stepwise manner, with intermediate splicing complexes designated as E (early), A (pre-spliceosome), B (fully assembled) and C (catalytic), and is followed by the actual splicing of the pre-mRNA (Staley and Woolford, 2009). Recognition and binding by the spliceosome requires three essential structural elements in all pre-mRNA introns that contain loosely defined consensus sequences: the 5' splice site, the intron branch point, and the 3' splice site (Lee and Rio, 2015). The strength of the splice sites depends on how far their sequences diverge from the consensus sequence, which determines their affinity for the spliceosome and the full usage of the sites (Kornblihtt et al., 2013). Additional control of splicing events is provided by four types of cis-acting splicing regulatory elements including the exonic splicing enhancers, exonic splicing silencers, intronic splicing enhancers, and intronic splicing silencers ( Wang and Burge, 2008). The binding of these regulatory elements by specific auxiliary RNA-binding proteins that are not part of the spliceosomal machinery can enhance or repress alternative splicing by promoting or hindering 84  spliceosome activity on the adjacent splice sites (Lee and Rio, 2015). These trans-acting splicing factors can be divided into three classes – the (1) Ser/Arg-rich (SR) family of proteins, (2) heterogeneous nuclear ribonucleoproteins (hnRNP) family of proteins and (3) tissue-specific RNA-binding splicing factors. Depending on the cellular context and/or their positions on the pre-mRNA, all three classes of splicing factors can act as splicing activators or repressors (Erkelenz et al., 2013; Lee and Rio, 2015). All these regulatory activities result in seven main types of alternative splicing events that generate functionally-distinct transcripts, of which the cassette-type alternative exons (or "skipped exons") are the most common, accounting for at least one-third of known alternative splicing events in humans (Galante et al., 2004; Kim et al., 2007). The other alternative splicing events are intron retention, alternative 5' splice sites, alternative 3' splice sites, mutually exclusive alternative exons, alternative first exon and alternative last exon (Blencowe, 2006). The overall function of alternative splicing is hypothesized to increase the diversity of mRNA expressed from the genome, and is presumably a major source of cellular complexity. In fact, the prevalence and extent of alternative splicing correlates with increasing organism complexity ( Chen et al., 2014; Kim et al., 2007; Nilsen and Graveley, 2010). Over 90% of human genes are estimated to undergo alternative splicing (Pan et al., 2008; Wang et al., 2008), generating multiple isoforms at an average of 4 transcripts per gene (de Klerk and 't Hoen, 2015), although one transcript is usually dominant (Gonzàlez-Porta et al., 2013). Alternative splicing contributes to proteome complexity, but not all alternative splicing events result in the production of functional proteins. Only one-third of annotated human genes generate multiple protein isoforms (Kim et al., 2014), as some transcripts may be non-coding, or are subject to changes in RNA stability and localization that affect translation. Ultimately, alternative splicing 85  contributes to regulation of tissue-identity acquisition including adipocytic differentiation (Lin, 2015), organ development, and tissue and organ physiology (Baralle and Giudice, 2017). Misregulation of alternative splicing due to mutations in the core splicing machinery, or changes to splicing regulatory factors can lead to production of aberrant protein isoforms that contribute to diseases including cancer (Zhang and Manley, 2013). The link between aberrant alternative splicing and cancer has been well established, and affects splice variants of genes involved in almost every aspect of cancer cell biology (Oltean and Bates, 2014).  3.1.1.3 3’ polyadenylation The final process in pre-mRNA maturation involves two reactions at the 3' end: (1) endonucleolytic cleavage of the nascent RNA, followed by (2) polyadenylation, which is the synthesis of a poly(A) tail on the 3' terminus of the cleaved product by a poly(A) polymerase.  Together, the two coupled reactions are known collectively as polyadenylation, and is carried out by the polyadenylation machinery consisting of subunits of cleavage and polyadenylation stimulatory factor (CPSF), cleavage stimulatory factor, and cleavage factor Im and IIm complexes, which in turn recruit other auxiliary factors (Neve et al., 2017; Shi et al., 2009).  A gene may generate transcripts with multiple polyadenylation sites, and differential usage of these sites results in distinct mature mRNA isoforms, a phenomenon termed alternative polyadenylation. Most alternative polyadenylation sites are located in 3' UTRs downstream of the terminal exon, but some are located upstream of the last exon, mostly in introns (Elkon et al., 2013). The polyadenylation site is defined by RNA sequence elements, the most prominent of which is the polyadenylation signal with the canonical AAUAAA sequence (Proudfoot and Brownlee, 1976). Weaker variants of the hexameric polyadenylation signal exist that are utilized with varying frequencies throughout the genome (Derti et al., 2012). Cleavage by the CPSF 86  subunit CPSF73 occurs 15-30 nucleotides downstream of the polyadenylation signal (Mandel et al., 2006). This is followed by the non-templated addition of about 200 adenosines by the poly(A) polymerase (Kühn et al., 2009; Wahle, 1991). Recent genomic studies show that alternative polyadenylation is frequent, with at least 70% of mammalian genes expressing alternative polyadenylation isoforms (Derti et al., 2012; Hoque et al., 2013). Auxiliary RNA elements upstream and downstream of the polyadenylation signal also serve as binding platforms for the assembly of the polyadenylation machinery to regulate polyadenylation site selection (Neve et al., 2017). Usage of one polyadenylation site over another is dependent on the similarity of the RNA sequence elements to the consensus sequence, the overall expression and activity of the polyadenylation machinery, and recruitment of associated regulatory RNA binding proteins (Tian and Manley, 2017).  Alternative polyadenylation usage in 3' UTRs has an effect on post-transcriptional gene regulation through modulation of mRNA stability, translation, nuclear export, and cellular localization (Eckmann et al., 2011; Neve et al., 2016), while alternative polyadenylation occurring in the upstream coding regions will affect protein identity. These changes can have a significant impact on major biochemical and physiological processes, including tissue development and cellular differentiation (Hollerer et al., 2014; Lianoglou et al., 2013), proliferation (Elkon et al., 2012), and activation of immune cells (Sandberg et al., 2008; Singh et al., 2018) and neurons (Berg et al., 2012; Flavell et al., 2008). As in alternative splicing, deregulation of alternative polyadenylation has been implicated in cancer. Global 3' UTR shortening is associated with cell transformation and cancer, and typically leads to ten-fold more protein expression, in part through loss of microRNA-mediated repression on 3' UTR cis-acting elements (Mayr and Bartel, 2009).  87  3.1.2 RNA sequencing Past high-throughput technologies used for the interrogation of RNA sequences such as expressed sequence tag identification (EST) (Adams et al., 1992), serial analysis of gene expression (SAGE) (Velculescu et al., 1995), and DNA microarray technology (Lockhart et al., 1996; Schena et al., 1995) suffered from several limitations. These include high sequencing cost, data being semi-quantitative and containing only short sequences per cDNA, and in the case of DNA microarrays, requiring the availability of a priori sequence information or reference genomes/transcriptomes for probe design. In the past decade or so, massive parallel sequencing or Next Generation Sequencing (Wheeler et al., 2008) has emerged as a revolution in biology and medicine due to its ability to acquire an unprecedented amount of data in a relatively short time. RNA-sequencing, or RNA-seq, is now the method of choice for studying gene expression and identifying novel RNA species. This technology offers less background noise, a greater dynamic range of detection, and most importantly, directly reveals sequence identity which is crucial for analyzing novel genes or transcript isoforms. Even though direct sequencing of RNA molecules is possible (Ozsolak et al., 2009), the instruments designed for DNA-based sequencing are more technically mature and routinely used for RNA-seq. However, an additional cDNA library preparation step is required. The RNA-seq library is composed of cDNA inserts of a certain size that are flanked by adapter sequences to allow for amplification and sequencing (Boone et al., 2018). Millions of small cDNA inserts are then sequenced in parallel in a single experiment that yields millions of short sequence reads for alignment to a known reference sequence. The number of reads affects coverage of the reference sequence in terms of breath and depth, where depth refers to the number of times a given nucleotide in the genome has been read. 88  Despite recent sharp falls in price (van Dijk et al., 2014), sequencing costs remain substantial. In order to achieve cost effectiveness, there is a constant balancing act between sequencing at the lowest necessary depth (and hence cost) without sacrificing data quality.  RNA-seq technologies are versatile and not only used for the general analysis of gene expression. With the use of specialized cDNA library preparation and bioinformatics algorithms, various aspects of RNA processing can be interrogated  through specific applications such as isoform and gene fusion detection, and 3' end sequencing (Hrdlickova et al., 2017).  3.1.3 Rationale and aims of Chapter 3 In Chapter 2, the FUS-DDIT3 interactome was found to be enriched for proteins involved in RNA processing and splicing. Thirty out of the 75 proteins (40%) in the FUS-DDIT3 interactome were found in the Spliceosome Database that tracks components of the spliceosome (http://spliceosomedb.ucsc.edu) (Cvitkovic and Jurica, 2013) (Appendix B.1). This observation was not surprising given the prominent role of FUS in coupling transcription with splicing (Yu and Reed, 2015), and its association with the spliceosome (Rappsilber et al., 2002; Z. Zhou et al., 2002) and splicing factors (Meissner et al., 2003; L. Yang et al., 1998). FUS also plays a well-documented role in regulating alternative splicing, especially in the context of neurodegenerative diseases caused by FUS mutations and aberrant function (Ishigaki et al., 2012; Lagier-Tourenne et al., 2012; Rogelj et al., 2012).  Even though many FUS-DDIT3 variants have lost the RNA-binding domains of FUS (Figures 1.3 and 1.4), the fusion protein retains the ability to bind wild type FUS (Thomsen et al., 2013), and by extension, the ability to associate with FUS interactors. Indeed, FUS-DDIT3 has been reported to localize to the splicing factor compartments (Göransson et al., 2002). 89  The functional implication of RNA processing proteins in the FUS-DDIT3 interactome is unknown. In myxoid liposarcoma, results from splicing reporter assays suggest that FUS-DDIT3 interferes with YB1-mediated splicing of the adenovirus E1A pre-mRNA (Rapp et al., 2002). The findings from the FUS-DDIT3 interactome in Chapter 1 are also reminiscent of the large number of RNA splicing and processing proteins found in the Ewing sarcoma EWS-FLI1 interactome in which the spliceosomal network of EWS-FLI1 differed from EWSR1, and that EWS-FLI1 interfered with the alternative splicing profile (Selvanathan et al., 2015). The hypothesis in this chapter is that interaction of FUS-DDIT3 with the RNA splicing and processing proteins interferes with the alternative splicing profile. Therefore, the aim of this chapter is to assess whether FUS-DDIT3 regulates alternative splicing, using RNA-seq to analyze differential alternative splicing after knockdown of FUS-DDIT3 in the myxoid liposarcoma cell line 402-91.     90  3.2 Materials and methods 3.2.1 Mammalian cell culture The following cell lines were kindly provided: myxoid liposarcoma cell lines 402-91 and 1765-92 by Dr. Pierre Aman (University of Gothenburg, Sweden) (Aman et al., 1992), and DL-221 by Dr. Keila Torres (MD Anderson Cancer Center, Houston, TX, USA) (de Graaff et al., 2016). Myxoid liposarcoma cell lines were cultured in RPMI-1640 medium with 10% fetal bovine serum (Life Technologies), with routine verification of FUS-DDIT3 protein expression by western blot analysis. Non-sarcoma control cell line HeLa was purchased from ATCC, and maintained in DMEM medium with 10% fetal bovine serum. All cells were cultured at 37°C, 95% humidity, and 5% CO2. Regular testing was carried out on all cell lines for mycoplasma infection.  3.2.2 RNA interference siRNA transfection was performed using RNAiMAX (Thermo Fisher Scientific) via reverse transfection according to the manufacturer’s instructions. Briefly, siRNA at a 10 nM final concentration was mixed with 0.12 μL RNAiMAX for every 20 μL OptiMEM, and scaled to appropriate volumes to incubate in 24-well plates (for RNA extraction) or 6 cm plates (for cell lysis and protein extraction) at room temperature for 20 min. Cells in full culture media were then added to each well and incubated for 24 hr or 48 hr before cell lysis for RNA or protein extraction. The FlexiTube GeneSolution GS1649 for DDIT3 (Qiagen) was used for RNA-seq and gene expression validation; two DDIT3 siRNAs, Hs_DDIT3_1 FlexiTube (labelled #1 in figures) and Hs_DDIT3_5 FlexiTube (#5) (Qiagen), were used for alternative last exon validation experiments. The siGENOME non-targeting siRNA control pool #2 purchased from Dharmacon (Cat: D-001206-14-05) was used as a control in all experiments. 91   3.2.3 RNA-seq and alternative splicing analysis Total RNA was isolated from 402-91 myxoid liposarcoma cells in triplicates using the RNeasy Mini kit (Qiagen) after 24 hr or 48 hr treatment with control or DDIT3 siRNA. RNA quality control was performed using the Agilent 2100 Bioanalyzer and confirmed with a minimum RNA Integrity Number of 10 for all samples. For the preliminary RNA-seq analysis, single replicates from each time point were then prepared following the standard protocol for the TruSeq stranded mRNA library kit (Illumina) on the Illumina Neoprep automated nanofluidic library prep instrument. Sequencing was performed on the Illumina NextSeq 500 with Paired End 42bp × 42bp reads at an average read depth of 84.6 million (ranging between 66-109 million). De-multiplexed read sequences were then aligned to the Homo sapiens (PAR-masked/hg19) reference sequence using TopHat splice junction mapper with Bowtie 2 (http://ccb.jhu.edu/software/tophat/index.shtml) aligner. Assembly and differential expression was estimated using Cufflinks (http://cole-trapnell-lab.github.io/cufflinks/) through bioinformatics apps available on Illumina Sequence Hub. Regulation of alternative splicing was investigated using the Mixture of Isoform (MISO) package (Y. Katz et al., 2010). MISO applies a statistical framework to distinguish eight different types of annotated alternative splicing and processing events - skipped exon, mutually exclusive exon, retained intron, alternative 3’ and 5’ splice sites, alternative first and last exon, and tandem 3’ UTR. The algorithm takes a pair of samples as input and reports a ΔΨ value between the two samples for each annotated event. The Ψ (Percent Spliced In) value represents the percentage of inclusion of a given splicing event when considering two isoforms, and is scaled according to the number of possible reads that could be generated from each isoform. A 92  corresponding Bayes Factor (BF) value is generated to express the statistical significance of a Ψ value being altered between the two samples. For each time point (24 hr or 48 hr), the DDIT3 and control siRNA-treated samples were paired for comparison, and events with BF values ≥ 20 and |ΔΨ| values ≥ 0.1 were selected for analysis.  3.2.4 Quantitative reverse transcription polymerase chain reaction Total RNA was isolated from cell lines treated with DDIT3 or control siRNA for 24 hr or 48 hr using the RNeasy Mini kit (Qiagen), then reverse transcribed to cDNA with oligo(dT) 12-18 Primer (Thermo Fisher Scientific) and Superscript III (Thermo Fisher Scientific). SYBR Green reagent (Roche) was used for qPCR expression analysis on an ABI ViiA7 qPCR system (Thermo Fisher Scientific). The list of primers can be found in Appendix B.2. All transcript levels were normalized to SDHA or HPRT1 expression (Vandesompele et al., 2002). For gene expression validation experiments, the normalized transcript levels from DDIT3 siRNA treated samples were compared to control siRNA treated samples to calculate relative fold induction of expression using the comparative CT (∆∆CT) method (Livak and Schmittgen, 2001). For the alternative last exon validation experiments, the normalized transcript levels of the proximal terminal exon isoforms were compared to that of the distal terminal exon isoforms to calculate relative fold change between both isoforms using the ∆∆CT method.  3.2.5 Western blots Cells were lysed in lysis buffer (1% Triton X-100, 1 mM MgCl2, 15 mM Tris pH 8.0, 100 mM NaCl) supplemented with cOmplete EDTA-free protease inhibitors (Roche). Whole cell lysates were clarified by centrifugation at maximum speed for 10 min at 4°C, and quantified with BCA protein assay (Thermo Fisher Scientific). Protein lysates were separated by 10% SDS-93  PAGE and transferred to nitrocellulose membranes. Blots were incubated with the indicated primary antibodies at 1:1000 dilution except as follows: FUS (Santa Cruz, sc-373698), DDIT3 (Abcam, ab10444), and GAPDH (Santa Cruz, sc-25778; a protein commonly used as loading control in whole cell lysate experiments such as those performed in this chapter) at 1:2000 dilution. Secondary antibodies used for duplexing were: goat anti-mouse IgG (H+L) Alexa Fluor 790 (Thermo Fisher Scientific) and goat anti-rabbit IgG (H+L) Alexa Fluor 680 (Thermo Fisher Scientific). Western blot signals were visualized on the Odyssey Infrared System (LI-COR Biosciences).   3.2.6 Proliferation assay The proliferation rate of control or DDIT3 siRNA-treated 402-91 cells cultured in 48-well plates was assessed by measuring cell confluency over a period of 84 hours utilizing the IncuCyte Zoom® live cell imaging software (Essen BioScience).     94  3.3 Results 3.3.1 RNA-seq analysis of myxoid liposarcoma cell line 402-91 after FUS-DDIT3 knockdown To investigate whether FUS-DDIT3 influences alternative splicing, RNA-seq was performed on myxoid liposarcoma cell line 402-91 after treatment with control or DDIT3 siRNA to reduce FUS-DDIT3 expression in the cell line. The DDIT3 siRNA was chosen over a FUS siRNA as the cells do not express wild type DDIT3 (as assessed in Chapter 2, Figure 2.1A). The proliferation rate of 402-91 was assessed by monitoring cell confluency after FUS-DDIT3 knockdown to determine the best duration of knockdown (Figure 3.1A). With FUS-DDIT3 knockdown, the proliferation curve started deviating from that of the control knockdown at 24 hr, and became more pronounced at 48 hrs, suggesting underlying transcriptomic changes have occurred. Therefore, these two time points were selected for the RNA-seq experiment. Treatment of control or DDIT3 siRNA in 402-91 was performed for 24 hr and 48 hr in triplicates. While RNA was collected from all triplicate samples, a preliminary RNA-seq run was performed on only one replicate from each condition to assess evidence of differential alternative splicing, before committing further resources to performing RNA-seq on the remaining replicates. Protein collected from the same knockdown experiment confirmed successful reduction of FUS-DDIT3 protein expression, with only 8.6% and 5.5% of FUS-DDIT3 remaining in the 24 hr and 48 hr siRNA-treated samples, respectively (Figure 3.1B). The siRNA treatment did not alter expression levels of wild type FUS. Sequencing depth for the single replicates was achieved at 66 to 109 million reads, as past studies on differential alternative splicing events have used between 78 – 100 million reads (Katz et al., 2010; Selvanathan et al., 2015; Tien et al., 2017), and over 90% of the reads were aligned (Table 3.1). For the purpose of differential gene expression analysis between control and 95  siRNA-treated conditions in this initial RNA-seq analysis, data from the 24 hr and 48 hr time points were considered as duplicates. Multiple DDIT3 transcripts were detected that were significantly reduced in the DDIT3 siRNA-treated samples (Table 3.2). This reduction in transcript levels can be attributed to FUS-DDIT3 as the cells do not express wild type DDIT3.  3.3.2 Validation of RNA-seq gene expression results by qPCR For the preliminary RNA-seq study, the quality of the data was first assessed by validating the differential gene expression estimated by Cufflinks (http://cole-trapnell-lab.github.io/cufflinks/). After combining both time points as duplicates for statistical analysis, 648 transcripts were assessed to be differentially expressed between control and knockdown samples. After applying a selection criteria of log2 fold change > 1 or < -1, six transcripts with reported involvement in cancer (CDKN1A, CTGF, CYR61, FOXN3, MYC, SOX11) (Beekman et al., 2018; Gabay et al., 2014; Georgakilas et al., 2017; Li et al., 2015; Li et al., 2017) were selected for validation of their differential expression by qPCR (Table 3.3). After FUS-DDIT3 knockdown in three myxoid liposarcoma cell lines, which included 402-91 that was used for generating the baseline RNA-seq data (Figure 3.2A), changes to gene expression of all six transcripts were validated in 402-91 (Figure 3.2B-G), suggesting good reliability of the RNA-seq data. CYR61 showed the most consistent differential expression in all myxoid liposarcoma cell lines, and no effect in negative control HeLa cells (Figure 3.2D), indicating that CYR61 could be a biologically relevant direct or indirect FUS-DDIT3 target.  3.3.3 Alternative splicing analysis after FUS-DDIT3 knockdown To identify changes in alternative splicing, the mixture of isoforms (MISO) package (Katz et al., 2010) was used. MISO applies a statistical framework to distinguish different 96  alternative splicing and processing events. These 8 events are skipped exon, mutually exclusive exon, retained intron, alternative 3' splice site, alternative 5' splice site, alternative first exon, alternative last exon, and tandem alternative 3’ UTR (Figure 3.3A). The MISO algorithm was applied individually to each of the four samples (control or DDIT3 siRNA treatment for 24 hr or 48 hr) to estimate the abundance of each splice isoforms and the frequency of each type of splicing event. The first analysis looked at the global alternative splicing profile of the different splicing events in each of the four conditions. The frequency of each type of event was expressed as a percentage of all detected alternative splicing events, with skipped exons being the most common at around 38% for all samples (Figure 3.3B and Appendix B.3). However, the data revealed no indication whatsoever of there being any major changes to the alternative splicing profile between control and knockdown conditions at either time point (Figure 3.3B and Appendix B.3). To identify differential alternative splicing usage between control and FUS-DDIT3 knockdown samples, MISO was used to estimate the Ψ (Percent Spliced In, PSI) value for each annotated event. The Ψ value denotes the proportion of inclusion between two isoforms for that event, and incorporates cDNA insert length information for accurate estimation of Ψ. The analysis also utilizes a Bayes Factor (BF) value to represent the statistical significance of a Ψ value being altered between the two samples. A selection criteria of BF ≥ 20 and |∆Ψ| ≥ 0.1 was applied to the dataset. A total of 76 alternative splicing events were identified as significantly altered after 24 hr of FUS-DDIT3 knockdown, and this number increased to 283 after 48 hr of knockdown (Figure 3.4A), suggesting an accumulation of differential alternative splicing events over time. However, only five differential alternative splicing events were common between both time points (Figure 3.4B, Table 3.4), suggesting that these events might be spurious. Indeed, a closer look at the Ψ 97  values of these five common events showed that the ∆Ψ, or the increase/decrease in proportion of one isoform over the other after FUS-DDIT3 knockdown, was inconsistent between 24 hr and 48 hr of knockdown for four of the transcripts (Table 3.4). When the differential alternative splicing events were split into the eight types of splicing and processing events, there appeared to be an increase in differential alternative last exon usage from 12% of events at 24 hr, to 33% at 48 hr of FUS-DDIT3 knockdown (Figure 3.4C and Appendix B.4). Of a total of 103 differential alternative last exon events across both time points, 39 saw an increase in proximal terminal exon usage after FUS-DDIT3 knockdown (Appendix B.5, positive ∆Ψ value), while 64 of these events recorded an increase in the distal proximal usage after knockdown (Appendix B.5, negative ∆Ψ value).  3.3.4  FUS-DDIT3 does not affect alternative last exon usage To evaluate whether there was a real increase in differential alternative last exon usage after FUS-DDIT3 knockdown in myxoid liposarcoma, three of the transcripts called by MISO to undergo differential alternative last exon usage at 48 hr were selected for qPCR validation in 402-91 (the myxoid liposarcoma cell line used in RNA-seq experiments), DL-221 (another myxoid liposarcoma cell line), and HeLa (control). Western blot analysis confirmed knockdown of FUS-DDIT3 by two different DDIT3 siRNAs (#1 and #5) in the validation experiment (Figure 3.5). As negative results were considered a likely result from this experiment, two different DDIT3 siRNAs (instead of one) were used for validation. The transcripts FAT1, FLT1, and PML were chosen for their relatively high Bayes Factor and |∆Ψ| values (Table 3.5 and Appendix B.5), and also for the presence of unique exons against which primers could be designed (Figure 3.6A, D, G). Differential alternative last exon usage was called by MISO for multiple different FLT1 isoforms at the 48 hr time point (Appendix B.5). 98  The FLT1 alternative last exon event selected here was different from the one found that was common at both 24 hr and 48 hr knockdown time points and reported in section 3.3.3 (Table 3.4), as that event showed inconsistent ∆Ψ between 24 hr and 48 hr. There were no other alternative last exon events that were common between two time points to aid in selection of events for validation (see Discussion). For each of the three selected transcripts, the Ψ value in Table 3.5 represents the proportion of the isoform utilizing a proximal (earlier) terminal exon relative to the isoform possessing a distal (later) terminal exon, and is visualized in Figure 3.6B, E, H. The relative expression of both isoforms from each transcript was then analyzed by qPCR (Figure 3.6C, F, I) and compared with the trend of the Ψ value, i.e. whether the proportion of the proximal (early) terminal exon isoform increased or decreased after FUS-DDIT3 knockdown (Figure 3.6B, E, H).  The MISO analysis for FAT1 showed a decrease in the proximal terminal exon usage after FUS-DDIT3 knockdown (Figure 3.6B), but the qPCR analysis showed that there was no change in the proportion of proximal compared to the distal terminal exon isoform after FUS-DDIT3 knockdown (Figure 3.6C). Similarly, for FLT1 and PML, FUS-DDIT3 knockdown resulted in either no change, or a decrease in the proximal terminal exon isoform abundance by qPCR (Figure 3.6F, I), which did not validate the MISO result of an increase in the proportion of the proximal terminal exon isoform (Figure 3.6E, H). Taken together, the qPCR results suggest that the detection of these three differential alternative last exon events in the preliminary RNA-seq analysis after FUS-DDIT3 knockdown were likely to be false positives.  99  3.4 Discussion 3.4.1 Summary Changes to the alternative splicing profile contribute to oncogenic traits acquired by tumors during transformation, progression and metastasis (Oltean and Bates, 2014). This chapter aimed to address the large number of RNA processing proteins in the FUS-DDIT3 interactome with the hypothesis that the fusion protein interferes with alternative splicing in myxoid liposarcoma.  FUS-DDIT3 expression was reduced in 402-91 myxoid liposarcoma cells with control or DDIT3 siRNA for 24 hr or 48 hr. RNA was extracted and isolated from triplicate samples, and a preliminary RNA-seq analysis was first performed on one replicate from each sample. While there was good knockdown of FUS-DDIT3 at the transcript (Table 3.2) and protein levels (Figure 3.1B), relatively deep sequencing reads (Table 3.1), and validated gene expression data in 402-91 (Figure 3.2), there was no convincing evidence of FUS-DDIT3 interfering with global alternative splicing to justify further expenditure of time and resources to sequence the remaining replicates (discussed further below).  3.4.2 Lack of evidence of alternative splicing interference There was no significant change to the global alternative splicing profile after knockdown of FUS-DDIT3 for 24 hr or 48 hr (Figure 3.3B and Appendix B.3). While an increase in the number of differential alternative splicing events from 76 events at 24 hr to 283 events at 48 hr (Figure 3.4A) suggested an accumulation of differential alternative splicing events with increasing duration of FUS-DDIT3 knockdown, the fact that only 5 events were common between both time points (Figure 3.4B), and that the ∆Ψ trend was inconsistent at both time points (Table 3.4) would argue that the detected differential alternative splicing results were 100  spurious. This was confirmed by the failure to validate any of the tested differential alternative last exon splicing events using qPCR (Figure 3.6). These particular events were chosen because they were alternative last exon events, the occurrence of which appeared to increase with the duration of FUS-DDIT3 knockdown (Figure 3.4C), and also for methodological reasons because they possessed unique exon sequences for qPCR primer design. The lack of overlapping events could also be an indication of insufficient read depth to detect and quantify all differential alternative splicing events. Past studies that reported detection of differential alternative splicing ranged between 78 - 100 million reads (Katz et al., 2010; Selvanathan et al., 2015; Tien et al., 2017). The average number of reads in this study was 84.6 million (ranging between 66 - 109 million) and is thus likely to be sufficient for detecting differential alternative splicing. A previous study by Liu et al. on adipose tissue before and after systemic endotoxin treatment reported a 76% detection rate of alternative splicing events at 100 million reads. However, only 9% of differential alternative splicing events were detected between different conditions at 100 million reads, and this increased to ~80% with at least 400 million reads (Liu et al., 2013). There is, of course, the caveat that the necessary sequencing depth for detecting differential alternative splicing is cell type and tissue-specific (Stephan-Otto Attolini et al., 2015), and these read number precedents may not apply to myxoid liposarcoma. Another reason for the lack of overlapping differential alternative splicing events between 24 hr and 48 hr of FUS-DDIT3 knockdown could be the duration of knockdown. While the time points in this study were chosen based on phenotypic changes in proliferation rates (Figure 3.1A), it is possible that an equilibrium may not have been reached at 24 hr between degradation of existing FUS-DDIT3 protein and the turnover of mRNA transcripts through synthesis and degradation. The half-life of the FUS-DDIT3 protein in 402-91 is reported to be 5 101  hr (Aman et al., 2016) and there was good reduction of FUS-DDIT3 by 24 hr (Figure 3.1B). Previous estimates of mRNA half-lives in mammalian cells have varied widely from 20 min to 10 h, which could be due to differences in experimental procedures and data analysis (Dölken et al., 2008; Rabani et al., 2011; Schwanhäusser et al., 2011; Yang et al., 2003). Based on the higher end of mRNA half-life estimates, it is then possible that not all expected changes in alternative splicing events may have occurred by 24 hr of FUS-DDIT3 knockdown, and that a comparison between two later time points may reveal a larger overlap in differential alternative splicing events. Nevertheless, my results suggest that FUS-DDIT3 does not affect alternative splicing on a global scale (Figure 3.3B), and any effect might be on low level alternative splicing events that could only be detected at later time points and/or with much deeper (realistically, cost-prohibitive) sequencing. Although interpretation of results may be difficult, an indirect and less expensive approach might be to look for differences in the spliceosome network between FUS-DDIT3 and FUS to infer any potential effects on alternative splicing. This was seen in Ewing sarcoma, where EWS-FLI1 and EWSR1 bind associate with slightly different spliceosomal components with different RNA-dependencies (Selvanathan et al., 2015).  The finding of increased differential alternative last exon events was interesting, since FUS knockdown has been shown to increase alternative last exon usage in a study where loss of FUS results in premature transcript termination in HEK293T/17 human kidney cells (Schwartz et al., 2012). The knockdown of FUS-DDIT3 resulted in more events increasing distal (64 events) than proximal (39 events) terminal exon usage at both knockdown time points (Appendix B.5), i.e. there are more events resulting in longer isoforms, suggesting an opposing function to wild-type FUS. Even though the three events selected in this study for more detailed analysis did not validate via qPCR, 100 other differential alternative last exon events were detected (Appendix 102  B.4), and it is possible that future use of replicates and longer knockdown durations may narrow down true positive events for validation. The use of alternative last exons may also suggest (but not confirm) alternative polyadenylation usage, which is dependent on the usage of variant polyadenylation sites. FUS has been shown to interact with the polyadenylation machinery to regulate alternative polyadenylation (Masuda et al., 2015; Sun et al., 2015). Moreover, the FUS-DDIT3 interactome does contain components of the cleavage factor IIm complex that is involved in polyadenylation (Figure 2.6A). To assess whether FUS-DDIT3 plays a role in regulating alternative polyadenylation, several different RNA-seq pipelines developed specifically for 3' end targeted deep sequencing (Neve et al., 2017) might be better options for detecting differential alternative polyadenylation in future studies.  3.4.3 Conclusion and future work In this chapter, I found no evidence of global alternative splicing changes after 24 hr or 48 hr FUS-DDIT3 knockdown. However, addition of independent replicates and sequencing depth are huge determinants in the ability to detect differential alternative splicing between conditions (Liu et al., 2013; Stephan-Otto Attolini et al., 2015). While it is still possible that FUS-DDIT3 might affect alternative splicing on a smaller scale, any future work would require additional replicates, later time points and more sequencing depth. The current data, however, does not provide strong justification to further pursue differential alternative splicing as a major oncogenic mechanism of FUS-DDIT3 in myxoid liposarcoma. However, the enrichment of cleavage factor IIm complex components in the FUS-DDIT3 interactome, and the detection of increased alternative last exon usage after FUS-DDIT3 knockdown suggests that it might be 103  worthwhile to utilize 3' end targeted deep sequencing to assess whether FUS-DDIT3 has an impact on alternative polyadenylation.  104  A    B    Figure 3.1 Selection of FUS-DDIT3 knockdown duration for RNA-seq experiments  (A) Rate of proliferation of myxoid liposarcoma 402-91 cells after treatment with control or DDIT3 siRNA was monitored over 84 hours by measuring confluency in a 48-well plate using the Incucyte Zoom®. (B) Confirmation of FUS-DDIT3 knockdown in samples prepared for RNA-seq experiments after treatment with control or DDIT3 siRNA for 24 hr or 48 hr. Western blot analysis of FUS (green) and DDIT3 (red) shows the FUS-DDIT3 band in yellow (arrowed). Wild type FUS can be seen below the FUS-DDIT3 band. Densitometry analysis of FUS-DDIT3 was normalized to GAPDH to calculate the percentage of FUS-DDIT3 remaining after knockdown.   24 hrs48 hrsControl siRNADDIT3 siRNAIB: FUS (green), DDIT3 (red)IB: FUSIB: DDIT3IB: GAPDHDDIT3 siRNA+– +–24 h 48 h% FUS-DDIT3 remaining8.6% 5.5%402-91105                   402-911765-92DL-2210.00.20.40.60.81.01.21.41.61.8Cell LineRelative expression(fold change)FUS-DDIT324h Control siRNA24h DDIT3 siRNA48h Control siRNA48h DDIT3 siRNAA FUS-DDIT3402-911765-92DL-2210.00.20.40.60.81.01.21.41.61.8Cell LineRelative expression(fold change)FUS-DDIT324h Control siRNA24h DDIT3 siRNA48h Control siRNA48h DDIT3 siRNA hr Control siRNA hr Control siRNA hr DIT3 siRNA hr DIT3 siRNA402-911765-92DL-221HeLa0123456Cell LineRelative expression(fold change)CDKN1A24h Control siRNA24h DDIT3 siRNA48h Control siRNA48h DDIT3 siRNAValidated in RNA-seqcell lineInconsistent trend in other MLS cell linessiRNA has an unspecific effect on non-MLS lineCDKN1A402-911765-92DL-2210.00.20.40.60.81.01.21.41.61.8Cell LineRelative expression(fold change)FUS-DDIT324h Control siRNA24h DDIT3 siRNA48h Control siRNA48h DDIT3 siRNA hr Control siRNA hr Control siRNA hr DIT3 siRNA hr DIT3 siRNABRelative expression(foldchange)402-911765-92DL-221HeLa051015Cell LineRelative expression(fold change)CTGF24h Control siRNA24h DDIT3 siRNA48h Control siRNA48h DDIT3 siRNAValidated in all MLS linessiRNA has an unspecific effect on non-MLS lineCCTGF402-911765-92DL-2210.00.20.40.60.81.01.21.41.61.8Cell LineRelative expression(fold change)FUS-DDIT324h Control siRNA24h DDIT3 siRNA48h Control siRNA48h DDIT3 siRNA hr Control siRNA hr Control siRNA hr DIT3 siRNA hr DIT3 siRNARelative expression(foldchange)402-911765-92DL-221HeLa012345Cell LineRelative expression(fold change)CYR6124h Control siRNA24h DDIT3 siRNA48h Control siRNA48h DDIT3 siRNAValidated in all MLS cell linesCYR61402-911765-92DL-2210.00.20.40.60.81.01.21.41.61.8Cell LineRelative expression(fold change)FUS-DDIT324h Control siRNA24h DDIT3 siRNA48h Control siRNA48h DDIT3 siRNA hr Control siRNA hr Control siRNA hr DIT3 siRNA hr DIT3 siRNADRelative expression(foldchange)402-911765-92DL-221HeLa0.00.51.01.5Cell LineRelative expression(fold change)FOXN324h Control siRNA24h DDIT3 siRNA48h Control siRNA48h DDIT3 siRNAValidated in all MLS linessiRNA has an unspecific effect on non-MLS lineFOXN3402-911765-92DL-2210.00.20.40.60.81.01.21.41.61.8Cell LineRelative expression(fold change)FUS-DDIT324h Control siRNA24h DDIT3 siRNA48h Control siRNA48h DDIT3 siRNA hr Control siRNA hr Control siRNA hr DIT3 siRNA hr DIT3 siRNAERelative expression(foldchange)402-911765-92DL-221HeLa0.00.51.01.5Cell LineRelative expression(fold change)MYC24h Control siRNA24h DDIT3 siRNA48h Control siRNA48h DDIT3 siRNAValidated in RNA-seqcell lineInconsistent trend in other MLS cell linessiRNA has an unspecific effect on non-MLS lineMYC402-911765-92DL-2210.00.20.40.60.81.01.21.41.61.8C ll LineRelative expression(fold change)FUS-DDIT324h Control siRNA24h D IT3 siRNA48h Control siRNA48h DDIT3 siRNA r trol siRNA hr Control siRNA hr DIT3 siRNA r I 3 siRNAFRelative expression(foldchange)402-911765-92DL-221HeLa0.00.51.01.52.0Cell LineRelative expression(fold change)SOX1124h Control siRNA24h DDIT3 siRNA48h Control siRNA48h DDIT3 siRNAValidated in RNA-seqcell lineInconsistent trend in other MLS cell linesSOX11402-911765-92DL-2210.00.20.40.60.81.01.21.41.61.8C ll LineRelative expression(fold change)FUS-DDIT324h Control siRNA24h D IT3 siRNA48h Control siRNA48h DDIT3 siRNA r trol siRNA hr Control siRNA hr DIT3 siRNA r I 3 siRNAGRelative expression(foldchange)106  Figure 3.2 Validation of differential gene expression by qPCR Expression of selected genes listed in Table 3.3 was analyzed by qPCR in myxoid liposarcoma cells 402-91, 1765-92, DL-221 and in negative control HeLa cells after treatment with control or DDIT3 siRNA for 24 hr or 48 hr. Relative expression of each transcript after FUS-DDIT3 knockdown was expressed as fold change over control siRNA treatment. (A) Assessment of FUS-DDIT3 transcript reduction after DDIT3 siRNA treatment. Transcripts analyzed for differential gene expression were (B) CDKN1A, (C) CTGF, (D) CYR61, (E) FOXN3, (F) MYC, (G) SOX11. Error bars = standard deviation.   107   A B    Figure 3.3 Global alternative splicing profile does not change after 24 hr and 48 hr FUS-DDIT3 knockdown in 402-91 (A) The types of alternative splicing and processing events estimated by MISO: alternative 3’ splice site (A3SS), alternative 5’ splice site (A5SS), alternative first exon (AFE), alternative last exon (ALE), mutually exclusive exon (MXE), retained intron (RI), skipped exon (SE), or tandem 3’ untranslated region (TandemUTR). (B) Each type of alternative splicing event was expressed as a percentage of total alternative splicing events called by MISO in 402-91 for the following conditions: treatment for 24 hr with 10 nM control siRNA for 24 hr (n=75,702) or 48 hr (n=75,193), and treatment with 10 nM DDIT3 siRNA for 24 hr (n=74,717) or 48 hr (n=73,382). The actual numbers of detected events are listed in Appendix B.3.    Skipped exon (SE) Retained intron (RI)Alternative 5’ splice site (A5SS)Alternative 3’ splice site (A3SS)Mutually exclusive exon (MXE)Alternative first exon (AFE)Alternative last exon (ALE)Tandem 3’ UTRA3SSA5SS AFEALEMXE RI SETandemUTR010203040Alternative splicing profile% of events402-91Control siRNA 24hDDIT3 siRNA 24hControl siRNA 48hDDIT3 siRNA 48hA3SSA5SS AFEALEMXE RI SETandemUTR010203040Alternative splicing profile% of events402-91Control siRNA 24hDDIT3 siRNA 24hControl siRNA 48hDDIT3 siRNA 48h24 hr Control siRNA48 hr Control siRNA48 hr DDIT3 siRNA24 hr DDIT3 siRNA108  A        B            C    Figure 3.4 Differential alternative splicing events after 24 hr and 48 hr FUS-DDIT3 knockdown in 402-91 (A) Number of differential alternative splicing events between FUS-DDIT3 knockdown and control after 24 hr and 48 hr of siRNA treatment. (B) Overlap of common differential alternative splicing events between both time points. (C) Percentage of each type of differential alternative splicing events expressed over the total number of events at each time point. The events are alternative 3’ splice sites (A3SS), alternative 5’ splice sites (A5SS), alternative first exon (AFE), alternative last exon (ALE), mutually exclusive exons (MXE), retained intron (RI), skipped exon (SE), or tandem untranslated region (TandemUTR). The actual numbers of differential alternative splicing events are found in Appendix B.4.    24 hr48 hrAll events5278 71r r24 hr48 hrDifferential alternative splicing events109    Figure 3.5 Confirmation of FUS-DDIT3 reduction for alternative last exon validation experiment Myxoid liposarcoma cell 402-91 and DL-221 were treated with control or two different DDIT3 siRNAs (#1 and #5) for 48 hr. For color blots, FUS was detected in the green channel and DDIT3 in the red channel. GAPDH was used as a loading control.   CtrlsiRNA #1 #5DDIT3 siRNA CtrlsiRNA #1 #5DDIT3 siRNA402-91 DL-221IB: FUS (green), DDIT3 (red)IB: GAPDH110      uc010isn.1uc003izf.1uc010iso.1mRNA transcript: FAT1FAT10.00.10.20.30.40.5ALE eventposterior mean ControlKnockdown402-91DL-221HeLa0.00.10.20.30.4FAT1Expression ratio(isoform 1 / isoform 2)Control siRNADDIT3 siRNA (#1)DDIT3 siRNA (#5)Data from MISO analysis qPCR validation from independent experimentAB CFAT1Fold change(proximal / distal terminal exon isoform)Distal terminal exon isoform primersProximal terminal exon isoform primersΨvalueuc010aap.1 uc001usc.2mRNA transcript: FLT1uc010aaq.1uc001usa.2uc001usb.2FLT10.00.20.40.60.8posterior meanALE eventControlKnockdown402-91DL-221HeLa05101520100200300400500Expression ratio(isoform 1 / isoform 2)FLT1Control siRNADDIT3 siRNA (#1)DDIT3 siRNA (#5)Proximal terminal exon isoform primersData from MISO analysis qPCR validation from independent experimentDE FFLT1Fold change(proximal / distal terminal exon isoform)Distal terminal exon isoform primersΨvalue111    Figure 3.6 qPCR does not validate alternative last exon usage in FAT1, FLT1 and PML Data is shown for 3 transcripts (A, B, C) FAT1, (D, E, F) FLT1, (G, H, I) PML. (A, D, G) UCSC genome browser track (hg19 assembly) showing the affected isoforms (Table 3.5) called by MISO to exhibit alternative last exon usage in the 402-91 RNA-seq data. Primers (arrowed) were designed against sequences unique to each isoform that would differentiate between isoforms utilizing a proximal (earlier) terminal exon or a distal (later) terminal exon. (B, E, H) The Ψ value obtained from MISO from each transcript is plotted out to show an increase or decrease in the proportion of the isoform using the proximal terminal exon isoform after FUS-DDIT3 knockdown.  (C, F, I) Myxoid liposarcoma cells 402-91, DL-221 and negative control HeLa were treated with control or two different DDIT3 siRNA (#1 and #5) for 24 hr or 48 hr. To assess whether the proportion of proximal terminal exon isoform increases or decreases after FUS-DDIT3 knockdown, relative transcript expression was expressed as fold change of the proximal terminal exon isoform over the distal terminal exon isoform.   PML0.00.20.40.60.8posterior meanALE eventControlKnockdown402-91DL-221HeLa0.00.51.01.52.0Expression ratio(isoform 1 / isoform 2)PMLControl siRNADDIT3 siRNA (#1)DDIT3 siRNA (#5)mRNA transcript: PMLuc002awv.1uc002aww.1uc002awu.1Data from MISO analysis qPCR validation from independent experimentGH IPMLFold change(proximal / distal terminal exon isoform)Distal terminal exon isoform primersProximal terminal exon isoform primersΨvalue112  Table 3.1 RNA-seq read mapping statistics Data is obtained from RNA-seq analysis on 4 samples – 402-91 cells were treated with control or DDIT3 siRNA for 24 hr or 48 hr.  Sample Number of Reads % Total Aligned % Abundant % Unaligned Median CV Coverage Uniformity Control siRNA (24 hr) 109,316,669 93.6% 3.2% 6.4% 0.58 DDIT3 siRNA (24 hr) 87,404,339 92.4% 3.1% 7.6% 0.54 Control siRNA (48 hr) 75,448,066 93.3% 2.7% 6.7% 0.55 DDIT3 siRNA (48 hr) 66,345,123 92.2% 3.4% 7.7% 0.58    113  Table 3.2 Differential expression of DDIT3 transcripts Transcripts from RNA-seq data after FUS-DDIT3 knockdown. FPKM = Fragments Per Kilobase Million.  Transcript ID Gene Control siRNA (FPKM) DDIT3 siRNA (FPKM) Log2 Fold Change NM_004083 DDIT3 89.000 14.3000 -2.6 NM_001195054 DDIT3 0.25 0.048 -2.4 NM_001195055 DDIT3 0.42 0.134 -1.7 NM_001195057 DDIT3 6.95 3.631 -0.9 NM_001195056 DDIT3 4.54 3.305 -0.5 NM_001195053 DDIT3 0.13 0.052 -1.3    114  Table 3.3 RNA-seq data for transcripts selected for qPCR validation  Gene Control FPKM DDIT3 FPKM Log2 (fold change) p value significant CDKN1A 30.5 60.70 0.99 5x10-5 yes CTGF 11000 30900 1.58 5x10-5 yes CYR61 40700 124000 1.60 5x10-5 yes FOXN3 11.0 3.78 -1.541 5x10-5 yes MYC 50.3 27.10 -0.894 5x10-5 yes SOX11 17.2 3.92 -2.134 5x10-5 yes    115  Table 3.4 Common differential alternative splicing events between 24 hr and 48 hr FUS-DDIT3 knockdown Alternative splicing events called by MISO to be differentially processed between control and DDIT3 siRNA treatment. Events include skipped exons (SE), alternative 5’ splice sites (A5SS), alternative first exon (AFE) and alternative last exon (ALE). The Ψ value denotes relative proportion of one isoform over the other. ∆Ψ denotes change in Ψ from FUS-DDIT3 knockdown to control. CI = confidence interval. Bayes factor denotes statistical significance of the Ψ value between knockdown and control samples: the larger, the more significant.  Transcript Time point Alternative splicing event DDIT3 siRNA Ψ DDIT3 siRNA CI (low) DDIT3 siRNA CI (high) Ctrl siRNA Ψ Ctrl siRNA CI (low) Ctrl siRNA CI (high) ∆Ψ Bayes factor SDHAP1 48 hr SE 0.74 0.46 0.97 0.13 0.06 0.23 0.61 2x109 SDHAP1 24 hr SE 0.16 0.08 0.28 0.79 0.48 0.97 -0.63 15,609 FAM21C 48 hr SE 0.45 0.24 0.72 0.93 0.83 0.98 -0.48 12,639 FAM21C 24 hr SE 0.04 0 0.14 0.54 0.35 0.76 -0.5 7x1011 WASH7P 48 hr A5SS 0.96 0.92 0.99 0.61 0.39 0.82 0.35 55 WASH7P 24 hr A5SS 0.51 0.26 0.77 0.95 0.89 0.99 -0.44 29 ZFAND5 48 hr AFE 0.54 0.48 0.61 0.4 0.35 0.45 0.14 21 ZFAND5 24 hr AFE 0.39 0.32 0.46 0.55 0.49 0.62 -0.16 23 FLT1 48 hr ALE 0.16 0.11 0.22 0.32 0.25 0.4 -0.16 45 FLT1 24 hr ALE 0.41 0.33 0.5 0.2 0.16 0.25 0.21 282   116  Table 3.5 Differential alternative last exon events selected for qPCR validation Transcripts selected for qPCR validation showing significant differential alternative last exon usage between control and FUS-DDIT3 knockdown at 48 hrs. Event name description: first four digits are gene ID, the “@” sign serves as a separator between proximal (early) terminal exon isoforms followed by the distal (later) terminal exon isoforms. The Ψ value denotes relative proportion of the proximal over the distal isoforms. ∆Ψ denotes change in Ψ from FUS-DDIT3 knockdown to control. CI = confidence interval. Bayes factor denotes statistical significance of Ψ value between knockdown and control samples; the larger, the more significant.  Transcript Event name DDIT3 siRNA Ψ DDIT3 siRNA CI (low) DDIT3 siRNA CI (high) Ctrl siRNA Ψ Ctrl siRNA CI (low) Ctrl siRNA CI (high) ∆Ψ Bayes factor FAT1 2195 @uc010iso.1 @uc010isn.1, uc003izf.1 0.24 0.22 0.26 0.41 0.39 0.43 -0.17 1x1012 FLT1 2321 @uc001usc.2 @uc010aap.1, uc010aaq.1, uc001usa.2, uc001usb.2 0.39 0.32 0.46 0.16 0.13 0.2 0.23 1x1012 PML 5371 @uc002aww.1 @uc002awu.1, uc002awv.1 0.69 0.64 0.74 0.51 0.46 0.56 0.18 2x107  117  Chapter 4: Association of FUS-DDIT3 with chromatin regulatory complexes  4.1 Introduction Epigenetics is the study of heritable but reversible changes in gene expression that are not accompanied by changes in DNA sequence (Bird, 2007). It is dictated by the accessibility of the chromatin environment for transcription factor binding, and is regulated by DNA methylation, chromatin regulation (through histone modification and chromatin remodeling), and non-coding RNA (Allis and Jenuwein, 2016). Although each mechanism plays an important role and there is significant crosstalk between mechanisms, chromatin regulation through histone modifications and chromatin remodeling have been the main mechanisms implicated in multiple sarcomas (Wojcik and Cooper, 2017).  4.1.1 Histone modifications DNA is organized into tightly regulated chromatin structures which begin with the nucleosome as the basic unit of chromatin. The nucleosome is composed of 145-147 base pairs of DNA wound around histone octamers (two copies each of the core histones H2A, H2B, H3 and H4) (Bentley et al., 1984; Kornberg, 1974; Luger et al., 1997). An astonishing number of histone post-translational modifications have been reported, including methylation, acetylation, phosphorylation, ubiquitination, sumoylation, biotynylation, citrullination, ADP-ribosylation, beta-N-glycosylation, isomerization, crotonylation, formylation, propionylation and butyrylation (Sadakierska-Chudy and Filip, 2015; Zhao and Garcia, 2015). The most well-studied modifications are acetylation and methylation on the N-terminal "tail" (as opposed to the "globular" region) of histones that project outward from the nucleosome, and are easily accessible (Luger et al., 1997). These modifications are in turn regulated by a plethora of enzymes responsible for depositing (the "writers") or removing (the "erasers") the histone marks. 118  The most versatile of all histone post-translational modifications is methylation, as two residues – lysine and arginine – can be mono-, di- and trimethylated. Although methylation does not affect overall charge, it does alter the size and hydrophobicity of the modified residue. H3K4me3 and H3K27ac/me3 are key histone modifications that are involved in transcriptional regulation. Gene expression is associated with the location and degree of methylation of H3K4 – H3K4me1 is enriched at enhancers, H3K4me2 at sequences immediately flanking the transcription start sites, and H3K4me3 is the hallmark of promoters or transcription start sites of both actively transcribed and poised genes (Barski et al., 2007; Ernst et al., 2011; Heintzman et al., 2007; Kim and Buratowski, 2009; Mikkelsen et al., 2007; Santos-Rosa et al., 2002). H3K27me3, and to a lesser extent H3K27me2, are associated with transcriptional repression, although H3K27me1 signals have been found to be higher at active than at silent promoters, particularly downstream of the transcription start sites (Barski et al., 2007; Lee et al., 2006). The lysine methyltransferases and demethylases show relatively high specificity for the lysine residue and for the degree of methylation. For example, KMT2A (also known as mixed lineage leukemia 1, MLL1) is responsible for methylating H3K4, but the degree of methylation (mono-, di-, or trimethylation) catalyzed by KMT2A is dependent on interactions with binding partners such as absent, small or homeotic-2 like (ASH2L) and WD repeat domain 5 (WDR5) (P. Ernst and Vakoc, 2012). Another enzyme, enhancer-of-zest homolog 2 (EZH2), is the H3K27 methyltransferase responsible for all three states of methylation (O'Meara and Simon, 2012). Histone lysine methylation was thought to be relatively permanent until the landmark discovery of the first histone demethylase, lysine-specific demethylase 1 (LSD1, also known as KDM1A) (Shi et al., 2004), that can demethylate H3K4me1 and H3K4me2. Since then, several members 119  of the LSD family of demethylases, as well as the larger family of Jumonji C demethylases, have been discovered and characterized for different substrates (Kooistra and Helin, 2012). Histone acetyltransferases catalyze the transfer of the acetyl group from the acetyl-CoA cofactor to a lysine side chain within histones (Marmorstein and Zhou, 2014). Acetylation of H3K27 is one of the most widely documented indicators of active enhancers (Creyghton et al., 2010; Rada-Iglesias et al., 2011) and promoters (Ernst et al., 2011; Hawkins et al., 2011), and is catalyzed by the acetyltransferase p300 (Hilton et al., 2015; Rada-Iglesias et al., 2011).  Unlike methyltransferases, the acetyltransferases have varying substrate specificities. For example, KAT8 (also known as MYST1 or MOF) is specifically responsible for H4K16 acetylation (Taipale et al., 2005; Thomas et al., 2008). By contrast, p300 is much more promiscuous in its acetyltransferase activity. Besides H3K27, p300 acetylates multiple residues on all four core histones (Henry et al., 2013; Liu et al., 2008; Luebben et al., 2010; McManus and Hendzel, 2003; Schiltz et al., 1999), as well as non-histone proteins such as p53 (Wang et al., 2013).  Histone acetyltransferase activity is balanced by histone deacetylases (HDACs) that belong to either the histone deacetylase family or the Sir2 regulator family. The better characterized histone deacetylase family members, such as HDAC1 and HDAC2, also show low substrate specificity for both histone and non-histone proteins (Seto and Yoshida, 2014). Notably, HDAC1 and HDAC2 form the catalytic core of at least three different stable chromatin regulatory complexes – Sin3A  (Laherty et al., 1997), NuRD (Tong et al., 1998; Xue et al., 1998), and CoREST (Hakimi et al., 2002; You et al., 2001), that possibly target different substrates. HDAC1 and HDAC2 share 83% sequence identity and functional redundancy (Jurkin et al., 2011). Together, these two proteins are the predominant histone deacetylation enzymes in 120  mammalian cells, as double knock-out of HDAC1 and HDAC2 in T-cells or embryonic stem cells leads to loss of half of the total deacetylase activity in the cells (Dovey et al., 2013). There are two functional consequences of histone modifications. The first is the direct structural perturbation of chromatin. For example, acetylation reduces the positive charge of histones, presumably leading to less compact chromatin structure and facilitating DNA access by transcriptional machinery and regulators (Shogren-Knaak et al., 2006). The second function of histone modifications is to recruit effector proteins harboring specialized "reader" domains that recognize distinct histone marks. Methyl-lysine readers are the most thoroughly studied group of readers with 10 domains identified to date, some of which show high degrees of specificity while others bind promiscuously (Hyun et al., 2017). Modifications such as H3K27me3 can be recognized by both chromodomain- and WD40 domain-containing proteins in the polycomb repressive complexes (Hyun et al., 2017). Among acetyl-lysine readers, the bromodomain is the most well characterized (Marmorstein and Zhou, 2014), and was regarded as the only acetyl-lysine reader until the plant homeodomain was reported to interact with acetylated as well as methylated histone H3 lysine residues (Lange et al., 2008). There is high diversity even within bromodomains – 42 bromodomain-containing proteins exist in humans, with a total of 61 unique bromodomain sequences (Sanchez et al., 2014). The various reader domains can be found in a wide range of proteins that mediate many fundamental processes including transcription regulation, DNA replication and recombination, DNA damage response, histone modifications (the "writers" and "erasers"), and chromatin remodeling (Musselman et al., 2012).  121  4.1.2 Chromatin regulatory complexes Cellular identity is grounded in the "transcriptional memory" of silent and activated gene expression states that are maintained over many cell generations (Steffen and Ringrose, 2014). This is achieved through one of the most prominent epigenetic regulatory systems involving the evolutionarily conserved Polycomb group (PcG) and Trithorax group (TrxG) components that act antagonistically as chromatin modifiers to regulate gene expression in cell differentiation and developmental processes (Schuettengruber et al., 2017). PcG proteins form histone modifying multiprotein complexes that are highly diverse and function as transcriptional repressors. There are two broad groups of PcG enzymatic complexes – polycomb repressive complex 1 (PRC1) and polycomb repressive complex 2 (PRC2). Each group contains a distinct set of core functional proteins, but multiple complexes exist within each group based on the composition of additional accessory proteins. The core of PRC1 consists of a RING1 protein (RING1A or RING1B) that mediates ubiquitination on H2AK119 for chromatin compaction, and one of six Polycomb group ring-finger domain proteins (PCGF1-6) that are required for H2AK119ub but also function as a scaffold for PRC1 assembly. Additional PRC1 accessory components interact with both core proteins to form various permutations of PRC1 complexes, including a chromobox protein (CBX2, CBX4, or CBX6-8) or RYBP/YAF2 in a mutually exclusive manner, and a selection of WDR5, polyhomeotic homologous protein (PHC1-3), B-cell lymphoma 6 corepressor (BCOR), and/or the demethylase KDM2B (Chittock et al., 2017; Gao et al., 2012). PRC2 has a different set of core proteins, namely the H3K27 methyltransferase EZH1/2, SUZ12, EED, and RBBP4/7. Like PRC1, compositional diversity of the PRC2 complex is further conferred by additional accessory proteins that modulate enzymatic activities or chromatin target sites (Chittock et al., 2017; Ciferri et al., 2012). 122  As the PcG complexes have relatively poor DNA sequence specificity, their recruitment to or eviction from chromatin is dependent on interaction with CpG islands, specific transcription factors, nascent RNAs, non-coding RNAs, and chromatin modifications from PcG complexes themselves or from other chromatin modifying complexes (Blackledge et al., 2015; Schuettengruber et al., 2017). Rather than actively driving transcriptional repression, recent evidence suggests that PcG complexes are recruited to sites at which transcriptional silencing has already been achieved, and are thus important for the maintenance, but not the establishment, of gene repression states (Hosogane et al., 2013; Riising et al., 2014). The TrxG complexes are a heterogeneous group with a widespread role in transcriptional activation, often in opposition to PcG-mediated repression (Poynter and Kadoch, 2016). These complexes can have either ATP-dependent chromatin remodeling or histone modification activities. The SWI/SNF (switch/sucrose non-fermentable) family of proteins are the most extensively studied TrxG chromatin remodeling complexes. While nucleosomes are stable with limited mobility, chromatin remodeling complexes may reposition and/or eject nucleosomes to regulate chromatin accessibility and gene transcription (Bai and Morozov, 2010). All chromatin remodeling complexes contain a DNA helicase belonging to one of four families depending on the sequence of the ATPase domain. Besides SWI/SNF, the other three families are CHD (chromodomain helicase DNA-binding), ISWI (imitation SWI) and INO80 (inositol requiring 80) (Tyagi et al., 2016).  SWI/SNF complexes are large - containing up to 15 subunits, including a single central ATPase BRG or BRM. The underlying mechanism for transcription activation by SWI/SNF was thought to be a result of nucleosome eviction at gene promoters to allow transcription factor binding (Kwon et al., 1994). A clearer picture for SWI/SNF antagonism with polycomb complexes has recently emerged, as SWI/SNF was shown to directly and rapidly evict PRC1 123  from chromatin in an ATPase-dependent manner, which was followed by the loss of H2AK119ub1, H3K27me3, and a consequent gradual increase in chromatin accessibility (Kadoch et al., 2017; Stanton et al., 2017). Additional recent evidence points to SWI/SNF playing a major role in regulating H3K27 acetylation through direct interaction with and recruitment of the p300/CBP histone acetyltransferase to enhancers of genes involved in differentiation and development (Alver et al., 2017; Mathur et al., 2017; Wang et al., 2017). A second group of TrxG complexes is involved in histone modifications. The family of COMPASS-like complexes contain different histone methyltransferases with unique functionalities and specificity for H3K4 or H3K36 (Piunti and Shilatifard, 2016; Poynter and Kadoch, 2016), but all share the same core proteins of WDR5, ASH2L, RBBP5 and DPY30 that are required to enhance the otherwise weak activities of the weak histone methyltransferases (Dou et al., 2006; Patel et al., 2009). For example, the MLL1/2 COMPASS-like complexes additionally contain HCF1 and MENIN, as well as either KMT2A (MLL1) and KMT2B (MLL2) methyltransferases in a mutually exclusive manner, with both enzymes having non-redundant functions (Denissov et al., 2014; Hu et al., 2013). Beyond PcG and TrxG, there is also the NuRD (nucleosome remodeling and histone deacetylation) complex. Initially defined as a transcriptional repressor, NuRD combines two distinct chromatin modifying enzymatic activities (Tong et al., 1998; Xue et al., 1998; Y. Zhang et al., 1998). NuRD couples histone deacetylation by HDAC1/2 with chromatin remodeling by CHD3/4 from the chromodomain helicase family. The rest of the core complex is made up of the histone chaperone proteins RBBP4 and RBBP7, one zinc-finger protein GATAD2A/B, two MTA proteins (MTA1-3), CDK2AP1 and MBD2/3. With the exception of RBBP4/7 and HDAC1/2 being present in other corepressor complexes, the remaining proteins are relatively specific to NuRD (Allen et al., 2013). Direct interaction of LSD1/KDM1A with NuRD has been 124  observed in some contexts (Wang et al., 2009; Whyte et al., 2012), adding a third enzymatic activity to the complex, although LSD1 is not a core NuRD component (Kloet et al., 2014; Smits et al., 2013). NuRD activity removes H3K27ac from target genes, and increases substrate availability for PRC2 recruitment and PRC2-mediated H3K27 trimethylation and gene silencing (Reynolds et al., 2012). Despite the initial description of NuRD as a corepressor, recent data shows the presence of NuRD at all sites of active transcription in embryonic stem cells, and that NuRD can act as both a transcriptional activator or repressor (Bornelöv et al., 2018). This transcriptional control is achieved by the NuRD chromatin remodeling activity to exert fine control over protein access to enhancers during lineage commitment (Bornelöv et al., 2018).  4.1.3 Epigenetic deregulation in sarcoma Epigenetic and chromatin deregulation has been well established in oncogenesis (Nebbioso et al., 2018). Chromatin regulators such as the KMT2 family members and SWI/SNF components are among the most frequently mutated genes in human cancer (Kadoch et al., 2013; Kandoth et al., 2013). Specifically in sarcomas, a range of mechanisms involving chromatin deregulation have been reported.  Malignant rhabdoid tumor is driven by the loss of SWI/SNF core member SMARCB1 (Biegel et al., 1999; Sévenet et al., 1999). The loss of SWI/SNF function disrupts the delicate antagonistic balance between SWI/SNF and PcG, leading to increased recruitment of EZH2 to Polycomb targets and H3K27me3-mediated repression ( Kim and Roberts, 2014; Versteege et al., 1998). Similarly, the comparatively understudied epithelioid sarcoma is also defined by SMARCB1 loss (Hornick et al., 2009; Modena et al., 2005), and likely to have an underlying epigenetic oncogenic mechanism.  125  The opposite is observed in malignant peripheral nerve sheath tumor, where a subset of tumors are driven by deletion of or loss of function mutations in PRC2 core components SUZ12 or EED ( Lee et al., 2014; Zhang et al., 2014). The resulting loss of PRC2 function leads to global H3K27me3 loss, as well as genes expressed in earlier progenitors (De Raedt et al., 2014; Lee et al., 2014).  In synovial sarcoma, three distinct models of SS18-SSX-driven epigenetic deregulation involving SWI/SNF and PcG have been reported: (1) gene repression through bridging of PcG components (via SSX) to ATF2 targets (via SS18) ( Su et al., 2012), (2) eviction of wildtype SS18 and BAF47 from SWI/SNF, earmarking both proteins for degradation and impacting SWI/SNF complex formation (Kadoch and Crabtree, 2013), and (3) gene activation through bridging of the SWI/SNF complex (via SS18) to KDM2B targets (via SSX) (Banito et al., 2018). Another fusion protein, EWS-FLI1 in Ewing sarcoma, was reported to possess the neomorphic property of recruiting SWI/SNF and p300 to GGAA repeats in tumor-specific enhancers for aberrant gene activation (Boulay et al., 2017; Riggi et al., 2014).  4.1.4 Rationale and aim of Chapter 4 In Chapter 2, immunoprecipitation-mass spectrometry (IP-MS) analysis of the FUS-DDIT3 interactome from whole cell lysates of myxoid liposarcoma cell lines did not reveal many transcription factors and regulators. The identification of transcriptional regulatory complexes around FUS-DDIT3 would be vital to understanding the role of FUS-DDIT3 as an aberrant transcription factor, and also to provide potential therapeutic targets that are important to the oncogenic function of FUS-DDIT3. However, transcription factors are notoriously difficult to detect via mass spectrometry due to their low abundance within highly complex eukaryotic proteomes, and the limited dynamic range of current mass spectrometers (Simicevic and 126  Deplancke, 2017). To bypass these limitations, two strategies were adopted in this chapter to increase identification of transcription factors and regulators in the FUS-DDIT3 interactome: target enrichment through the immunoprecipitation of FUS-DDIT3 from nuclear fractions instead of whole cell lysates (Simicevic and Deplancke, 2017), and label-based relative quantification using tandem mass tags (TMT) to reduce variances and improve quantification accuracy for lower abundance proteins (Lindemann et al., 2017). The isobaric TMT reagents include a mass reporter region that is released from labeled peptides during MS2 or MS3 fragmentation, allowing the ratio of different reporter groups to be used for relative quantification of peptides from different experimental conditions (Ankney et al., 2018). The increasing evidence of chromatin deregulation in other translocation-associated sarcomas leads to the hypothesis that FUS-DDIT3 associates with chromatin regulators in myxoid liposarcoma. Therefore, the aim of this chapter is to identify the transcriptional and/or chromatin regulatory complexes that associate with FUS-DDIT3 through TMT-labeled IP-MS specifically on nuclear fractions isolated from the myxoid liposarcoma cell line 402-91.   4.2 Materials and methods 4.2.1 Mammalian cell culture The following cell lines were kindly provided: myxoid liposarcoma cell lines 402-91 and 1765-92 by Dr. Pierre Aman (University of Gothenburg, Sweden) (Aman et al., 1992), and DL-221 by Dr. Keila Torres (MD Anderson Cancer Center, Houston, TX, USA) (de Graaff et al., 2016). Routine verification of FUS-DDIT3 protein expression was carried out with the myxoid liposarcoma cell lines using western blot analysis. Synovial sarcoma cell line SYO-1 was provided by Dr. Akira Kawai (National Cancer Centre Hospital, Tokyo, Japan) (Kawai et al., 2004). Sarcoma cell lines were cultured in RPMI-1640 medium with 10% fetal bovine serum 127  (Life Technologies). Non-sarcoma control cell line HeLa was purchased from ATCC, and maintained in DMEM medium with 10% fetal bovine serum. All cells were cultured at 37C, 95% humidity, and 5% CO2. Regular testing was carried out on all cell lines for mycoplasma infection.  4.2.2 Patient samples Human patients in this study provided informed consent for use of tissues for research purposes following procedures approved by the University of British Columbia Clinical Research Ethics Board (REB# H08-01411). Patient anonymity was maintained with the use of a sample code assigned by the Bone & Soft Tissue Tumor Bank as the only sample identifier. Surgically excised primary tumours including 19 myxoid liposarcomas, six malignant peripheral nerve sheath tumours, and two schwannomas, frozen and archived at the Vancouver General Hospital, were accessed from the Bone & Soft Tissue Tumor Bank for this study.   4.2.3 Transfection of siRNA and FUS-DDIT3 vector siRNA transfection was performed using RNAiMAX (Thermofisher Scientific) via reverse transfection according to the manufacturer’s instructions. Briefly, siRNA at a 10 nM final concentration was mixed with 0.12 μL RNAiMAX for every 20 μL OptiMEM, and scaled to appropriate volumes to incubate in 6 cm or 15 cm plates at room temperature for 20 min. Cells in full culture media were then added to each plate and incubated for 48 hr, 72 hr or 7 days before protein or histone extraction. For 7-day knockdown experiments, cells were re-transfected with siRNA on the 4th day. Two DDIT3 siRNAs were used: Hs_DDIT3_1 FlexiTube (labelled #1 in figures) and Hs_DDIT3_5 FlexiTube (#5) (Qiagen). The siGENOME non-targeting siRNA control pool #2 purchased from Dharmacon (Cat: D-001206-14-05) was used as a control. 128  The FUS-DDIT3 plasmid was obtained from Addgene (plasmid #21831) (Crozat et al., 1993). Transfection of the FUS-DDIT3 plasmid was performed with FuGENE HD (Promega) according to the manufacturer’s instructions. Briefly, HeLa cells were grown in 6 cm plates until 60-80% confluent. FuGENE HD was mixed with FUS-DDIT3 or control plasmid at a 3:1 (µL:µg) ratio in OptiMEM for 15 min. The mixture was then added directly to cells for 48 hr before protein extraction.  4.2.4 Protein extraction To obtain whole cell lysates, cells were harvested and lysed in Triton X-100 lysis buffer (1% Triton X-100, 1 mM MgCl2, 15 mM Tris pH 8.0, 100 mM NaCl) supplemented with Roche cOmplete EDTA-free protease inhibitors. Whole cell lysates were clarified by centrifugation at maximum speed for 10 min at 4°C, quantified with BCA protein assay (Thermofisher Scientific), and stored on ice until ready for immunoprecipitation or mixed with sample loading buffer for SDS-PAGE and western blot analysis. For preparation of nuclear extracts, adherent myxoid liposarcoma 402-91 cells were rinsed once with PBS on 15 cm dishes before trypsinization. Trypsinized cells were swelled with rotation for 15 min in the cold room with 5X packed cell volume of hypotonic lysis buffer (20 mM HEPES-KOH pH7.5, 5 mM MgCl2, 5 mM CaCl2, 1 mM EDTA) supplemented with Roche cOmplete EDTA-free protease inhibitors, then lysed with 5 strokes of the dounce homogenizer on ice. Cell nuclei were washed once with PBS and collected by centrifugation at 2000 RPM at 4°C for 5 min, then lysed with Triton X-100 lysis buffer supplemented with Roche cOmplete EDTA-free protease inhibitors and 1 U/mL benzonase (MilliporeSigma) for 1 hr in the cold room with rotation. Nuclear lysates were clarified by centrifugation at 16,000 RPM at  4°C for 129  10 min, then quantified with BCA protein assay (Thermofisher Scientific), and stored on ice until ready for immunoprecipitation.  4.2.5 Immunoprecipitation Antibodies used for immunoprecipitations were: normal mouse IgG (Santa Cruz, sc-2025), normal rabbit IgG (Cell Signaling Technology, #2729), DDIT3 (L63F7) (Cell Signaling Technology, #2895), HDAC2 (Abcam, ab12169), and KDM1A (Abcam, ab17721). Antibodies were first incubated with Dynabeads™ Protein G (ThermoFisher Scientific) for 10 min at room temperature.  For non-mass spectrometry immunoprecipitations, the non-crosslinked antibody-Dynabead mixture was incubated with whole cell lysates or protein extracts overnight with rotation in the cold room. Beads were washed three times with cold lysis buffer, twice with cold PBS, then eluted by boiling for 5 min in sample loading buffer. For samples destined for mass spectrometry analysis, antibodies were crosslinked to Dynabeads with BS3 (bis(sulfosuccinimidyl)suberate) crosslinker (Thermofisher Scientific) according to manufacturer protocol. The antibody-bead mixture was then pre-blocked with 10 mg/mL BSA in casein blocker (ThermoFisher Scientific) with rotation at room temperature for 30 min, before being incubated with nuclear extracts overnight in the cold room with rotation. Beads were washed three times with cold lysis buffer, followed by twice with cold PBS.  Proteins were eluted by boiling for 5 min in SP3 elution buffer (4% SDS, 2% β-mercaptoethanol, 40 mM Tris pH 6.8). Eluted immunoprecipitates were incubated at 45°C for 30 min, then alkylated with 400 mM iodoacetamide for 30 min at 24°C. Reactions were quenched with addition of 200 mM dithiothreitol. Eluted proteins were prepared for trypsin digestion using the SP3 cleanup protocol as previously described (Hughes et al., 2014). Acetonitrile was added 130  to the SP3 bead-protein mixture to a final 50% vol/vol, and incubated for 8 min at room temperature. Using a magnetic rack, beads were washed two times with 200 μL 70% ethanol for 30 sec, and once with 180 μL 100% acetonitrile for 15 sec. For digestion, beads were reconstituted in 5 μL 50 mM HEPES pH 8.0 buffer containing trypsin/rLys-C enzyme mix (Promega) at a 1:25 enzyme to protein ratio, and incubated for 14 hr at 37°C. Digested peptides were recovered by removing the supernatant on a magnetic rack.  4.2.6 Western blots Protein lysates were separated by 10% or 4-15% SDS-PAGE and transferred to PVDF membranes. Blots were incubated with the indicated primary antibodies: BAF57/SMARCE1 (Bethyl, A300-810A), BAF155/SMARCC1 (Cell Signaling Technology, #11956), BRG1/SMARCA4 (Bethyl, A300-813A), FUS (Santa Cruz, sc-373698), DDIT3 (Abcam, ab10444), GAPDH (Santa Cruz, sc-25778), Histone H3 (Abcam, ab10799), H3K4me3 (Cell Signaling Technology, #9751), H3K27ac (Cell Signaling Technology, #8173), H3K27me1 (Abcam, ab194688), H3K27me3 (Cell Signaling Technology, #9733), HDAC2 (Abcam, ab12169), KDM1A (Abcam, ab17721), or MTA1 (Abcam, ab71153). Secondary antibodies used for duplexing were: goat anti-mouse IgG (H+L) Alexa Fluor 790 (Thermofisher Scientific) and goat anti-rabbit IgG (H+L) Alexa Fluor 680 (Thermofisher Scientific). Western blot signals were visualized on the Odyssey Infrared System (LI-COR Biosciences).  4.2.7 Mass spectrometry analysis (Velos) Quality control of unlabeled peptide fractions was carried out on an Orbitrap Velos LTQ MS Platform (Thermofisher Scientific). Samples were introduced using an Easy-nLC 1000 system (Thermofisher Scientific). Columns used for trapping and separations were packed in-131  house. Trapping columns were packed in 100 μm internal diameter capillaries to a length of 25 mm with C18 beads (Reprosil-Pur, Dr. Maisch, 3 μm particle size). Trapping was carried out for a total volume of 10 μL at a pressure of 400 bar. After trapping, gradient elution of peptides was performed on a C18 (Reprosil-Pur, Dr. Maisch, 1.9 μm particle size) column packed in-house to a length of 15 cm in 100 μm internal diameter capillaries with a laser-pulled electrospray tip and heated to 45°C using AgileSLEEVE column ovens (Analytical Sales & Service). Elution was performed with a gradient of mobile phase A (water and 0.1% formic acid) to 8% B (acetonitrile and 0.1% formic acid) over 6 min, to 25% B over 15 min, to 40% over 5 min, with final elution (80% B) and equilibration (5% B) using a further 4 min at a flow rate of 375 nL/min.  Data acquisition on the Orbitrap Velos (control software version 2.1.0.SP1.1160) was carried out using a data-dependent MS2 method. Survey scans covering the mass range of 380 – 1600 were acquired at a resolution of 30,000 (at m/z 200), with quadrupole isolation enabled, an S-Lens RF Level of 60%, a maximum fill time of 100 ms and an automatic gain control (AGC) target value of 1e6. For MS2 scan triggering, monoisotopic precursor selection was enabled, charge state filtering was limited to 2 – 4, an intensity threshold of 5e2 was employed, and dynamic exclusion of previously selected masses was enabled for 60 seconds with a tolerance of 20 ppm. MS2 scans were acquired in the ion trap in Rapid mode after CID fragmentation with a maximum fill time of 10 milliseconds, quadrupole isolation, an isolation window of 2 m/z, collision energy of 35%, activation Q of 0.25 and an AGC target value of 1e4. MS1 was acquired in profile mode, and MS2 in centroid format. Successful quality control was determined by the detection of multiple DDIT3 peptides in the samples immunoprecipitated with the DDIT3 antibody, but not the IgG control antibody.  132  4.2.8 TMT labelling TMT 10-plex labeling kits were obtained from Pierce. Each TMT label (5 mg per vial) was reconstituted in 500 μL of acetonitrile and frozen at -80°C. Prior to labeling, TMT labels were removed from the freezer and allowed to equilibrate at room temperature. Labeling reactions were carried out through addition of TMT label in two volumetrically equal steps to achieve a 2:1 (μg:μg) TMT label to peptide final concentration, 30 min apart. All incubations were carried out at room temperature. Reactions were quenched by addition of glycine. Labeled peptides were concentrated in a SpeedVac centrifuge to remove acetonitrile, combined, and desalted before MS analysis on the Orbitrap Fusion.  4.2.9 Mass spectrometry analysis (Fusion) Analysis of TMT labeled peptide fractions was carried out on an Orbitrap Fusion Tribrid MS platform (Thermofisher Scientific). Samples were introduced using an Easy-nLC 1000 system (Thermofisher Scientific). Columns used for trapping and separations were packed in-house. Trapping columns were packed in 100 μm internal diameter capillaries to a length of 25 mm with C18 beads (Reprosil-Pur, Dr. Maisch, 3 μm particle size). Trapping was carried out for a total volume of 10 μL at a pressure of 400 bar. After trapping, gradient elution of peptides was performed on a C18 (Reprosil-Pur, Dr. Maisch, 1.9 μm particle size) column packed in-house to a length of 25 cm in 100 μm internal diameter capillaries with a laser-pulled electrospray tip and heated to 45°C using AgileSLEEVE column ovens (Analytical Sales & Service). Elution was performed with a gradient of mobile phase A (water and 0.1% formic acid) to 8% B (acetonitrile and 0.1% formic acid) over 5 min, to 25% B over 88 min, to 40% over 20 min, with final elution (80% B) and equilibration (5% B) using a further 7 min at a flow rate of 375 nL/min. 133  Data acquisition on the Orbitrap Fusion (control software version 2.1.1565.20) was carried out using a data-dependent method with multi-notch synchronous precursor selection MS3 scanning for TMT tags. Survey scans covering the mass range of 350 – 1500 were acquired at a resolution of 120,000 (at m/z 200), with quadrupole isolation enabled, an S-Lens RF Level of 60%, a maximum fill time of 50 ms, and an automatic gain control (AGC) target value of 4e5. For MS2 scan triggering, monoisotopic precursor selection was enabled, charge state filtering was limited to 2 – 4, an intensity threshold of 5e3 was employed, and dynamic exclusion of previously selected masses was enabled for 60 seconds with a tolerance of 20 ppm. MS2 scans were acquired in the ion trap in Rapid mode after CID fragmentation with a maximum fill time of 20 ms, quadrupole isolation, an isolation window of 1 m/z, collision energy of 30%, activation Q of 0.25, injection for all available parallelizable time turned OFF, and an AGC target value of 1e4. Fragment ions were selected for MS3 scans based on a precursor selection range of 400-1600 m/z, ion exclusion of 20 m/z low and 5 m/z high, and isobaric tag loss exclusion for TMT. The top 10 precursors were selected for MS3 scans that were acquired in the Orbitrap after HCD fragmentation (NCE 60%) with a maximum fill time of 90 ms, 50,000 resolution, 120-750 m/z scan range, ion injection for all parallelizable time turned OFF, and an AGC target value of 1e5. The total allowable cycle time was set to 4 seconds. MS1 and MS3 scans were acquired in profile mode, and MS2 in centroid format.  4.2.10 Protein enrichment score calculation The calculation of the protein enrichment score (PES) is adapted from the PSEA-Quant method which reflects the abundance and reproducibility of the abundance measurements across IP replicates (Lavallée-Adam et al., 2014). The average peptide mass spectra value for each peptide is obtained from Proteome Discoverer to obtain the peptide-level signal value. These 134  values are then normalized to the same median values within the control and DDIT3 IPs separately. For each peptide in each DDIT3 IP replicate, the ratio of the peptide-level signal value over the average value from all control IPs is obtained. The peptide-level ratios and coefficients of variation (CV) from each DDIT3 IP replicate are then aggregated to the protein-level. The protein level CV is transformed to the same scale as the average protein ratios (CV(transformed)) for easy visualization on a scatterplot. PES is calculated by: 	"#$ = &'()&*(	)&+,- − /0(+)&234-)5(6) + )&+,-9:; − 1(2 × )&+,-9:;) − 2  where ratiomax is the largest average ratio in the dataset. Proteins with an average ratio below 1 (ie. higher signal in control IPs than DDIT3 IPs) are then removed from the dataset. The bait protein should show the highest PES which reflects a high average protein ratio and low CV of the ratio across triplicate IPs. The PES values are then converted to z-scores to apply candidate selection cutoffs based on the number of standard deviations away from the mean.  4.2.11 Crude histone extraction Crude histone extracts were obtained using the reagents from the histone purification mini kit from Active Motif and adapted from the manufacturer protocol.  Each frozen patient tumor was cut into small pieces and placed in a Precellys CKMix tissue homogenizing tube (Bertin Instruments). Histone extraction buffer from the histone purification mini kit (Active Motif) was added at a volume that just covers the tissue pieces and beads completely. The homogenizing tube was then loaded onto the Precellys 24 tissue homogenizer (Bertin Instruments) to run at 5,500 RPM through four cycles of 20 sec each, with a 15 sec pause between cycles 1 and 2, and cycles 3 and 4. Samples were placed on ice for 2 min between the 2nd and 3rd cycles. The homogenized tissue was further diluted with histone 135  extraction buffer and transferred to a fresh tube for overnight incubation with the histone extraction buffer on a rotating platform in the cold room. For histone extraction from cell lines, the cells were first rinsed with pre-warmed PBS. Ice-cold histone extraction buffer was added to the cell culture dishes before scraping the cells off for collection in a microfuge tube. Cells were left to incubate overnight with the histone extraction buffer on a rotating platform in the cold room. For both tissues and cell lines, after the overnight histone extraction buffer incubation, extracts were transferred to fresh tubes and centrifuged at maximum speed, 5 min, 4°C. The supernatant containing crude histones was transferred to fresh tubes for neutralization and equilibration with ¼ volume of 5X Neutralization Buffer from the histone purification mini kit (Active Motif). Crude histone concentration was estimated on the nanodrop using the 230 nm wavelength. Sample loading buffer was added to the crude histones before proceeding with SDS-PAGE and western blotting.   4.2.12 Chromatin immunoprecipitation Chromatin immunoprecipitation (ChIP) was carried out using the ChIP-IT® Express Enzymatic kit from Active Motif according to the manufacturer’s instructions. Briefly, cells were fixed with 1% formaldehyde (in PBS) for 10 min with gentle shaking, followed by quenching of the reaction with 130 mM glycine for 5 min. Cell were rinsed twice with PBS, then scraped off cell culture plates and collected by centrifugation at 1,000 RPM for 4 min. Cells were incubated with the kit’s lysis buffer on ice for 30 min and lysed with 40 strokes of the dounce homogenizer. The nuclei were collected by centrifugation at 2,400 RCF for 10 min at 4°C, and resuspended in the kit’s digestion buffer before addition of the enzymatic cocktail for 6.5 min at 37°C to shear the DNA. The reaction was stopped by addition of EDTA to a final concentration 136  of 10 mM and chilling samples on ice for 10 min. Samples were centrifuged at 18,000 RCF for 10 min at 4°C, and the supernatant containing sheared chromatin was collected. Immunoprecipitation from 12 µg sheared chromatin was carried by overnight incubation with 1.8 µg antibody, ChIP Buffer 1, Protein G magnetic beads and protease inhibitor cocktail, with rotation in the cold room. Antibodies used for ChIP were: normal mouse IgG (Santa Cruz, sc-2025), normal rabbit IgG (Cell Signaling Technology, #2729), DDIT3 (L63F7) (Cell Signaling Technology, #2895), H3K27ac (Cell Signaling Technology, #8713), and H3K27me3 (Diagenode, C15410195). Beads were washed three times with ChIP Buffer 1, then twice with ChIP Buffer 2 before elution with elution buffer (50 mM Tris-HCl, pH 8.0, 10 mM EDTA, 1.0% SDS, 200 mM NaCl) for 1 hr at room temperature with rotation. Samples were reverse crosslinked with the kit’s reverse cross-linking buffer for 15 min at 95°C, then treated with Proteinase K from the kit for 90 min at 37°C. The PCR purification kit from Qiagen was used to clean up the samples before quantitative real-time polymerase chain reaction (qPCR). SYBR Green reagent (Roche) was used for qPCR analysis on an ABI ViiA7 qPCR system (Thermo Fisher Scientific). The sequences of the primers used can be found in Appendix C.1. For each ChIP sample, the percentage of input was calculated from the CT values, and used to obtain the fold enrichment over the normal IgG controls.    137  4.3 Results 4.3.1 TMT-labeled IP-MS identification of a FUS-DDIT3 interactome following nuclear extraction Immunoprecipitation (IP) of endogenous FUS-DDIT3 was performed on nuclear lysates from the 402-91 myxoid liposarcoma cell line. The IPs were carried out in four sets of replicates, with the first three sets of replicates used for TMT labeling and mass spectrometry analysis (Figure 4.1A).  The fourth set of replicates was subjected to western blot analysis to confirm immunoprecipitation of FUS-DDIT3 in the experiment (Figure 4.1B). The appearance of a doublet for the FUS-DDIT3 band (Figure 4.1B) was unexpected and not previously observed in IPs from whole cell lysates (Figure 2.1B). The smaller band is likely a degradation product that occurred as a result of the nuclear extraction protocol, which included a hypotonic swelling step that was not used the preparation of whole cell lysates. The degradation product can be visualized more clearly in the single channel DDIT3 western blot image in Appendix C.2A, where it was only present in minute amounts in the input, but was enriched considerably after overnight immunoprecipitation. In a separate, unrelated immunodepletion experiment on nuclear lysates where three successive IPs using normal mouse IgG were carried out over three days for 24 hr each, the amount of the smaller FUS-DDIT3 bands increased over time, with a corresponding decrease in the larger FUS-DDIT3 band that was dominant in the initial input sample (Appendix C.2B). This suggests that the smaller FUS-DDIT3 band observed in the fourth IP replicate in Figure 4.1B was a degradation product. However, it is unclear why the nuclear lysis protocol, but not the whole cell lysis protocol, would result in the degradation of FUS-DDIT3 over time, especially with the supplementation of fresh protease inhibitors in all lysis buffers and the use of cold temperatures for all steps in the protocols.  138  Nevertheless, quality control on unlabeled peptide fractions from the first three IP replicates identified three DDIT3 peptides in the all DDIT3 IP triplicates, and little to none detected in the triplicate mouse IgG IPs (Appendix C.3). This suggested successful enrichment of FUS-DDIT3 in all DDIT3 IP replicates, and provided the impetus to proceed with TMT labeling of the remaining peptide fractions. Following TMT labeling and mass spectrometry analysis, 174 proteins were selected for further analysis by having an average ratio of enrichment over control IPs greater than 1 (Appendix C. 4). The selected proteins were visualized on a scatterplot showing their relative enrichment ratio in the DDIT3 IPs and the coefficient of variation (CV) of this ratio across triplicates (Figure 4.2). A protein with a higher protein ratio and lower CV would be considered a higher confidence FUS-DDIT3 interactor. This measure of confidence can be quantified by calculating the protein enrichment score (PES). The PES value reflects both the abundance of a protein and the reproducibility of the abundance measurement (through consideration of CV) across triplicate IPs. This value was used to rank the candidate interactors (Appendix C.4) with a higher ranking protein showing a higher protein ratio and a lower CV on the scatterplot (Figure 4.2A). Using the z-score of the PES (Figure 4.2B), a cut-off of 0.5 standard deviations away from the mean was used to identify the top 43 proteins on the list of FUS-DDIT3 interactors (Appendix C.4), and highlighted in blue on Figure 4.2A. Despite the enrichment of DDIT3 peptides in the unlabeled peptide fractions of DDIT3 IPs during the quality control run (Appendix C.3), none were detected in the TMT-labeled peptide fractions (Appendix C.4). However, the top-ranked protein in the TMT-labeled nuclear interactome was C/EBPβ, the preferred dimerization partner of FUS-DDIT3 (Crozat et al., 1993; Göransson et al., 2009; 2005; Ron and Habener, 1992). Another known FUS-DDIT3 interactor 139  previously reported in the literature, wild type FUS (Thomsen et al., 2013), was also highly ranked (11/174) in the interactome (Appendix C.4).  Seven proteins from the nuclear interactome were common with the whole cell lysate interactome reported in Chapter 2 (Figure 4.3). Besides C/EBPβ and FUS, the other five proteins common between both nuclear and whole cell lysate interactomes were highly ranked in the current nuclear interactome list: SFPQ (rank: 3/174), NONO (4/174), RBM14 (5/174), PSPC1 (6/174), and SPAG5 (15/174) (Figure 4.3 and Appendix C.4). With the exception of RBM14, these proteins had all been validated using reciprocal co-immunoprecipitation and/or proximity ligation assay as FUS-DDIT3 interactors in Chapter 2 (Figures 2.7, 2.9, and 2.10). Taken together, these observations suggest the successful immunoprecipitation of FUS-DDIT3 and identification of putative FUS-DDIT3 interactors from the nuclear fraction of 402-91 myxoid liposarcoma cells.  4.3.2 FUS-DDIT3 interactome includes multiple chromatin regulatory complexes To look for protein complexes that were enriched in the FUS-DDIT3 nuclear interactome, the CORUM database (comprehensive resource of mammalian protein complexes, http://mips.gsf.de/genre/proj/corum/index.html) was used to analyze the 43 proteins that made a selection cut-off of PES z-score ≥ 0.5 (Appendix C.5). This shorter list of 43 proteins will henceforth be referred to as the post-cut-off list. The enriched complexes in the post-cut-off list were similar to the ones found in Chapter 2 from the whole cell lysate interactome (Figure 2.6A), and included the SFPQ/NONO complex, mRNA processing and miRNA processing complexes (Appendix C.5). Analysis of the full (pre-cut-off) list of 174 nuclear interactors without applying the PES z-score cut-off generated a longer list of CORUM protein complexes (Figure 4.4), which included several chromatin 140  regulatory complexes such as nucleosome remodeling and deacetylation (NuRD) complex, mixed-lineage leukemia 1 (MLL1) COMPASS-like complex, and polycomb repressive complex 1 (PRC1) (arrowed in Figure 4.4). A closer look at the full list of 174 nuclear interactors (Appendix C.4) found 16 proteins belonging to various chromatin regulatory complexes, including NuRD, SWI/SNF, PRC1, PRC2 and MLL1 (Figure 4.5).  While the top gene ontology biological process in the full list was mRNA processing, which was also observed in the whole cell lysate interactome in Chapter 2 (Table 2.4), several chromatin modification processes such as histone deacetylation, protein methylation and peptidyl-lysine modification were additionally enriched in the nuclear interactome (Figure 4.6A). Similarly, gene ontology cellular component analysis identified a number of chromatin regulatory complexes (Figure 4.6B) among a longer list of enriched cellular components (Appendix C.6). Therefore, the results of the CORUM and gene ontology analyses suggest that the FUS-DDIT3 interactome includes multiple chromatin regulatory complexes.  4.3.3 Validation of FUS-DDIT3’s association with chromatin regulators To validate the association of FUS-DDIT3 with chromatin regulators, co-immunoprecipitation experiments were carried out. Since enrichment of histone deacetylase activity was identified in the bioinformatics analyses (Figure 4.4 and 4.6A), HDAC2 was immunoprecipitated from whole cell lysates of 402-91 and 1765-92 myxoid liposarcoma cell lines, and FUS-DDIT3 was found to co-IP with HDAC2 in both cell lines (Figure 4.7A). A subsequent reciprocal co-IP was performed in 1765-92 to pull down FUS-DDIT3 and this confirmed co-IP of HDAC2 (Figure 4.7B). The core NuRD component MTA1 was also observed to co-IP with FUS-DDIT3 (Figure 4.7B). 141  An additional NuRD interactor, the KDM1A lysine demethylase (Wang et al., 2009; Whyte et al., 2012), was selected for validation due to its presence in the post-cut-off list. Reciprocal co-IP experiments confirmed the association of FUS-DDIT3 with KDM1A (Figure 4.8). Finally, co-IP of SWI/SNF core components BRG1 (SMARCA4), BAF155 (SMARCC1), and accessory protein BAF57 (SMARCE1) was shown with FUS-DDIT3 (Figure 4.9). The co-IP data here thus validate the association of FUS-DDIT3 with chromatin regulators including HDAC2, as well as other components of the NuRD (MTA1, KDM1A), and SWI/SNF (BRG1/SMARCA4, BAF155/SMARCC1, and BAF57/SMARCE1) complexes.  4.3.4 Global H3K27 acetylation or trimethylation are not affected by FUS-DDIT3 To address whether there are functional implications for the association of FUS-DDIT3 with various chromatin regulatory complexes, changes to global histone modifications in myxoid liposarcoma tumors and cell lines were investigated. The levels of H3K27 trimethylation from frozen tissues of 19 myxoid liposarcomas, 6 malignant peripheral nerve sheath tumors, and 2 schwannomas was analyzed by western blot (Figure 4.10). The malignant peripheral nerve sheath tumors were included as a biological negative control as a subset of these tumors are known to exhibit loss of H3K27me3 (Asano et al., 2017; Cleven et al., 2016; Prieto Granada et al., 2016; Röhrich et al., 2016; Schaefer et al., 2016), and this was indeed observed in three of five assessable tumors (Figure 4.10). Similarly, schwannomas were included as a positive control that rarely loses H3K27 trimethylation (Asano et al., 2017; Cleven et al., 2016; Röhrich et al., 2016; Schaefer et al., 2016). Among the 19 myxoid liposarcoma tumors, none showed complete loss of H3K27me3. Three myxoid liposarcomas (VSA4, VSA182 and VSA211) showed lower H3K27 trimethylation, but the levels were not as low as the two malignant peripheral nerve sheath 142  tumors (VSA94, VSA116) showing H3K27me3 loss. In general, global H3K27 trimethylation appeared to be intact in myxoid liposarcomas. The global levels of other well characterized histone marks – H3K4me3, H3K27me1/3 and H3K27ac – was investigated by western blot after 48 hr knockdown of FUS-DDIT3 in 402-91, 1765-92 and DL-221 myxoid liposarcoma cell lines, as well as after 48 hr transfection of FUS-DDIT3 into HeLa cells (Figure 4.11A). The “writers” or “erasers” of these histone modifications are also found in the FUS-DDIT3 interactome – KMT2A (writer for H3K4me3), SUZ12 (writer for H3K27me1/3), HDAC1 (eraser for H3K27ac) (Figure 4.5). Western blot confirmed FUS-DDIT3 knockdown in the three myxoid liposarcoma cell lines, and exogenous expression of FUS-DDIT3 in HeLa cells (Figure 4.11A). Densitometric analysis of the level of histone modifications normalized against total histone H3 expression showed no clear pattern in the consequent changes to histone modification levels among the different DDIT3 siRNAs or cell lines (Figure 4.11B). Knockdown of FUS-DDIT3 for longer durations of 3 or 7 days in 402-91 and 1765-92 also showed similarly inconsistent trends in changes to H3K27 acetylation and trimethylation (Figure 4.12). Therefore, the data suggests that FUS-DDIT3 does not regulate global histone modifications for H3K4me3, H3K27me1/3 and H3K27ac, but does not rule out potentially important effects on specific genes that could be functionally important in myxoid liposarcoma.  4.3.5 FUS-DDIT3 knockdown in 402-901 reduces H3K27ac and increases H3K27me3 at the PTX3 promoter Since changes to global histone modifications were not observed for H3K4me3, H3K27me1/3 and H3K27ac after FUS-DDIT3 knockdown, the next question was whether FUS-DDIT3 could be regulating specific local changes to histone modifications that would not be 143  evident at the global level. As of this writing, there are no known ChIP-seq studies on FUS-DDIT3 targets. One of the few reported gene targets to be regulated by FUS-DDIT3 is pentraxin 3 (PTX3) (Forni et al., 2009), so this locus was chosen for further investigation. After a 48 hr knockdown of FUS-DDIT3 in 402-91 myxoid liposarcoma cells, chromatin immunoprecipitations (ChIP) of H3K27ac and H3K27me3 were performed to investigate changes to their occupancy at the PTX3 promoter, as well as at sequences upstream and downstream of this promoter (Figure 4.13). As wild type DDIT3 was not expressed in 402-91 under regular culture conditions (Figure 2.1A), ChIP of DDIT3 was performed to check for occupancy of FUS-DDIT3 along PTX3, and to confirm knockdown of FUS-DDIT3 removes this occupancy pattern (Figure 4.13A). Knockdown of FUS-DDIT3 also resulted in a reduction of H3K27ac and increase of H3K27me3 at the PTX3 promoter (Figure 4.13B-C), suggesting that FUS-DDIT3 is able to regulate local histone modifications at its gene targets.  4.4 Discussion As an aberrant transcription factor, the identification of transcriptional regulatory complexes surrounding FUS-DDIT3 is crucial to understanding its mechanism of action, and would also identify the functionally relevant targets that mediate the oncogenic function of FUS-DDIT3, potentially opening new therapeutic opportunities given that FUS-DDIT3 itself is not specifically targeted by existing agents. The FUS-DDIT3 interactome identified from whole cell lysates in Chapter 2 did not contain many transcription factors or regulators. Due to emerging evidence of chromatin deregulation as a major oncogenic mechanism in other translocation-associated sarcomas that affect a similar patient population, a hypothesis was made that FUS-DDIT3 will also associate with chromatin regulators. In this chapter, an attempt was made to increase the identification 144  transcriptional and chromatin regulators through enriching them in nuclear lysates of 402-91 myxoid liposarcoma cells before performing IP-MS. In addition, relative quantification with TMT labeling was used to improve quantification accuracy for these typically low abundance proteins. Among the 174 proteins identified in the FUS-DDIT3 nuclear interactome (Appendix C.4) were members of a number of chromatin regulatory complexes including NuRD, MLL1, PRC1 and SWI/SNF (Figures 4.4, 4.5 and 4.6). Co-immunoprecipitation experiments validated the association of FUS-DDIT3 with some specific chromatin regulators – HDAC2, and members of NuRD (MTA1, KDM1A) (Figures 4.7 and 4.8) and SWI/SNF (BRG1/SMARCA4, BAF155/SMARCC1, BAF57/SMARCE1) (Figure 4.9).  While FUS-DDIT3 did not appear to regulate global histone modifications for H3K4me3, H3K27me1/3 and H3K27ac (Figures 4.10, 4.11 and 4.12), knockdown of FUS-DDIT3 for 48 hr reduced H3K27ac and increased H3K27me3 at the promoter of PTX3 (Figure 4.13), suggesting there is repression of PTX3 transcription in the absence of FUS-DDIT3. This observation is in line with a previous report of FUS-DDIT3 binding the PTX3 promoter to activate gene transcription (Forni et al., 2009). PTX3 has been characterized as a soluble pattern recognition receptor for innate immune system, and has been implicated in various aspects of cancer progression (Giacomini et al., 2018). Unfortunately, the protein is not currently targetable with any known agents. It would be interesting to confirm FUS-DDIT3 regulation of histone modifications on the promoters or cis regulatory elements of any other potentially important target genes that might be revealed in future epigenomic studies using newer technologies in this fast-moving field. More importantly, further investigation into the occupancy of other chromatin regulators will determine whether the changes to histone modifications are the result of FUS-DDIT3 modulating the activity or recruitment of chromatin regulatory complexes at its targets.  145  The identification of multiple chromatin regulatory complexes in the FUS-DDIT3 interactome, even those that possess opposing activities such as SWI/SNF and PRC, was not surprising. Other sarcoma fusion proteins have been reported to interact with and function through more than one chromatin regulatory complex — EWS-FLI1 with NuRD/KDM1A and SWI/SNF (Boulay et al., 2017; Sankar et al., 2013), and SS18-SSX with PRC1.1/PRC2 and SWI/SNF (Banito et al., 2018; Kadoch and Crabtree, 2013; Su et al., 2012). Extensive chromatin colocalization of TrxG and PcG complexes have been observed in both flies and mammals, regardless of the state of activity for the target gene (Beisel et al., 2007; Enderle et al., 2011; Papp and Müller, 2006), suggesting a model whereby regulatory mechanisms tip the balance between opposing activities of both complexes by regulating their recruitment or activity levels (Schuettengruber et al., 2017). In fact, SWI/SNF was recently shown to constantly be involved in a dynamic competition with PRC1 by evicting PRC1 with its ATPase subunit BRG1/SMARCA4 (Kadoch et al., 2017). Therefore, FUS-DDIT3 could be upsetting the balance between different chromatin regulatory complexes at its targets through inappropriate retargeting or alteration of their activity levels. However, it is also possible that FUS-DDIT3 interacts with different complexes at different targets in a context-dependent manner. The association of FUS-DDIT3 with SWI/SNF is particularly interesting. Recent evidence points to strong SWI/SNF regulation of lineage-specific distal enhancers through direct interactions with p300 and acetylation of H3K27 (Alver et al., 2017; Mathur et al., 2017; Wang et al., 2017). This has implications for tumorigenesis as deregulation of SWI/SNF function would likely disrupt the tightly regulated expression of genes involved in differentiation and development. Since global changes in H3K27 acetylation were not observed with knockdown of FUS-DDIT3 (Figures 4.11 and 4.12), it is possible that FUS-DDIT3 does not affect SWI/SNF 146  function, but rather redirects its distribution to FUS-DDIT3 targets. Such a mechanism of action was reported in Ewing sarcoma, where EWS-FLI1 recruits and redistributes SWI/SNF to tumor-specific enhancers that otherwise have no regulatory functions (Boulay et al., 2017), and also in synovial sarcoma where SS18-SSX redirects SWI/SNF to KDM2B targets (Banito et al., 2018). Finding out whether FUS-DDIT3 also redistributes SWI/SNF will require future, detailed ChIP-seq analyses to identify the DNA regulatory elements or genes targeted by FUS-DDIT3, along with any changes to the distribution of SWI/SNF or other chromatin regulators, and assessment of histone modifications such as H3K27ac/me3 in the presence/absence of the fusion protein. Such data would aid in further clarifying the details of how FUS-DDIT3 affects chromatin regulators to drive tumorigenesis, and perhaps identify key components whose activity can be reversed by epigenetic drugs. 147      Figure 4.1 Experimental outline for immunoprecipitation-mass spectrometry using TMT labeling (A) Nuclear lysates from myxoid liposarcoma cells 402-91 were used for immunoprecipitation of endogenous FUS-DDIT3 with an anti-DDIT3 antibody (clone L63F7). Immunoprecipitations were carried out in four replicates. The first three replicates were used for TMT labeling and mass spectrometry analysis. (B) Immunoprecipitates from the fourth set of replicates were subjected to western blot analysis for detection of FUS (green) and DDIT3 (red) to assess immunoprecipitation of FUS-DDIT3 (merged, yellow). Trypsinize cells, swell in hypotonic buffer, then lyse plasma membrane with dounce homogenizerIP: overnight @ 4 °CCrosslink DDIT3 Ab to Dynabeads.Block with BSA and casein.Myxoid liposarcoma cell line 402-91Wash beadsPool replicates 1-3 for TMT labelingMass spectrometry analysis(Orbitrap Fusion)Reduction, alkylation, on-bead trypsin digestion, peptide clean-upIP: mo-IgGReplicate (1)IP: DDIT3Replicate (1)IP: DDIT3Replicate (2)IP: DDIT3Replicate (3)Discard cytoplasmic fraction, collect nuclei.Lyse with 1% Triton X-100 and treat with benzonase to collect nuclear lysates.IP: mo-IgGReplicate (2)IP: mo-IgGReplicate (3)IP: mo-IgGReplicate (4)IP: DDIT3Replicate (4)For replicate (4), elute proteins to assess IP by western blotAIB: FUS + DDIT33%InputMoIgG DDIT3IPFUS-DDIT3402-91nuclear extractB148  A    149  B  Figure 4.2 Scatterplot of FUS-DDIT3 nuclear interactome after selection cutoff Proteins with an average enrichment ratio of >1 in DDIT3 vs control IPs are presented (n=174). (A) Each circle represents a putative FUS-DDIT3 nuclear interactor, with circle size corresponding to the number of peptides (pepNum) detected in the mass spectrometry analysis. Each protein is plotted by the log2 values of its average enrichment ratio (log2(Protein Ratio)) and the transformed coefficient of variation for the average ratio across triplicate IPs. The protein enrichment score (PES) for each protein is calculated to take into account the abundance measurement of the protein (enrichment ratio) and the reproducibility of this measurement through the CV values (Appendix C.3). PES values provide a way of ranking the putative interactors – a higher ranking FUS-DDIT3 interactor would have a higher protein ratio and lower CV. A cutoff score of PES z-score ≥ 0.5 was applied to identify 43 top ranking-proteins in the FUS-DDIT3 nuclear interactome, and these proteins are highlighted in blue. (B) The same proteins are plotted by the z-scores and rank of their PES values (higher PES value = higher rank). Proteins previously validated from the whole cell lysate IP-MS screen are highlighted in red. The PES z-score = 0.5 is indicated by a dotted line.  150   Figure 4.3 Overlapping FUS-DDIT3 interactors identified in whole cell lysate and nuclear fraction A comparison was made between the FUS-DDIT3 interactome as identified in Chapter 2 from whole cell lysates of three myxoid liposarcoma cell lines (402-91, 1765-92, DL-221), and from the nuclear fraction of 402-91 that met a cutoff of PES z-score ≥ 0.5. Seven proteins were found to be common between both IP-MS experiments, and are listed with their ranking in the nuclear interactome in parenthesis. Among these, two (CEBPB and FUS) are known interactors from the literature and four (SFPQ, NONO, PSPC1 and SPAG5) were validated FUS-DDIT3 interactors from Chapter 2.Common proteins (rank in nuclear interactome)CEBPB (1)SFPQ (3)NONO (4)RBM14 (5)PSPC1 (6)FUS (11)SPAG5 (15)68(61.3%)7(6.3%) 36(32.4%)Blue text = known FUS-DDIT3 interacting partners from literatureRed text = validated from whole cell lysate interactome in Chapter 2151   Figure 4.4 CORUM protein complexes enriched in FUS-DDIT3 nuclear interactome The full pre-cutoff list of 174 proteins in the FUS-DDIT3 nuclear interactome was used to query the CORUM database to identify enriched protein complexes which are presented here with their names as assigned in the database. Numbers after each bar indicate the number of complex members found in interactome / total number of complex members in database. Chromatin regulatory complexes nucleosome remodeling and deacetylation (NuRD), mixed-lineage leukemia 1 (MLL1), and polycomb repressive complex 1 (PRC1) are arrowed. 0 5 10 15  MWRAD complex (MLL1, WDR5, RBBP5, ASH2L, DPY30)  HDAC1-associated protein complex  18S U11/U12 snRNP  RNA pol II containing coactivator complex Tat-SF  12S U11 snRNP  MeCP1 complex  MTA2 complex  NuRD.1 complex  microtubule binding  SMN1-SIP1-SNRP complex  6S methyltransferase and RG-containing Sm proteins complex  structural constituent of ribosome  MLL1 core complex  ATAC complex, GCN5-linked  steroid hormone receptor binding  Polycomb repressive complex 1 (PRC1, hPRC-H)  CDC5L complex  Mi2/NuRD complex  RNA polymerase II distal enhancer sequence-specific DNA binding  LARC complex (LCR-associated remodeling complex)  MLL1 complex  HCF-1 complex  20S methylosome and RG-containing Sm protein complex  Anti-HDAC2 complex  zinc ion binding  MLL-HCF complex  PID complex  histone deacetylase binding  snRNP-free U1A (SF-A) complex  ALL-1 supercomplex  TOP1-PSF-P54 complex  p54(nrb)-PSF-matrin3 complex  histone deacetylase activity  NRD complex (Nucleosome remodeling and deacetylation complex)  histone-lysine N-methyltransferase activity  PSF-p54(nrb) complex  Nop56p-associated pre-rRNA complex  Large Drosha complex  17S U2 snRNP  DGCR8 multiprotein complex  Spliceosome  C complex spliceosome-log10 (padj-value)CORUM protein complexes total n = 17423/7927/1418/1111/338/2016/1047/472/27/445/79/1074/53/38/282/34/72/422/8126/174/64/86/186/197/724/77/304/127/894/103/49/1654/84/89/2174/84/84/85/153/56/244/93/5152  A     153  B  Figure 4.5 Chromatin regulators are present in the FUS-DDIT3 nuclear interactome  Proteins with an average enrichment ratio of >1 in DDIT3 vs control IPs are presented (n=174). Components of various chromatin regulatory complexes are indicated on the scatterplots. (A) Each circle represents a putative FUS-DDIT3 nuclear interactor, with circle size corresponding to the number of peptides detected in the mass spectrometry analysis. Each protein is plotted by the log2 values of its average enrichment ratio (log2(Protein Ratio)) and the transformed coefficient of variation for the average ratio across triplicate IPs. (B) The same proteins are plotted by the z-scores and rank of their PES values (higher PES value = higher rank. The PES z-score = 0.5 is indicated by a dotted line.   154  A  B  Figure 4.6 Gene ontology classification of FUS-DDIT3 nuclear interactome Enriched gene ontology (A) biological processes or (B) cellular components in the full pre-cutoff list of the FUS-DDIT3 nuclear interactome (n=174). Numbers after each bar indicate the number of group members found in interactome / total number of group members in database. 0 20 40 60  rRNA metabolic process  osteoblast differentiation  cellular response to chemical stimulus  SRP-dependent cotranslational protein targeting to membrane  regulation of signal transduction by p53 class mediator  peptidyl-lysine modification  protein methylation  histone deacetylation  RNA export from nucleus mRNA processing-log10 (padj-value)GO Biological Processes64/495total n = 17417/14112/8514/17817/38511/1758/9325/29377/20412/2920 1 2 3 4 5   NuRD complex  CHD-type complex  SWI/SNF superfamily-type complex  H4 histone acetyltransferase complex   PcG protein complex    histone deacetylase complex     MLL1 complex     MLL1/2 complex  PRC1 complex    histone methyltransferase complex-log10 (padj-value)GO Cellular Component(chromatin regulators) total n = 1748/694/125/295/297/606/465/377/754/164/16155     Figure 4.7 Co-immunoprecipitation validates FUS-DDIT3 interaction with HDAC2 (A) Immunoprecipitation (IP) of HDAC2 in whole cell lysates of myxoid liposarcoma cells 402-91 and 1765-92, and synovial sarcoma cells SYO-1. Unbound lane shows post-IP supernatant. FUS bands were detected in the green channel, and DDIT3 in red. FUS-DDIT3 bands are indicated by the merged yellow bands. The different size of FUS-DDIT3 in 402-91 and 1765-92 is due to the different fusion variants present in both cell lines – the smaller Type 1 variant is found in 402-91, and the longer Type 8 variant is present in 1765-92 (Figure 1.3). (B) IP of FUS-DDIT3 in whole cell lysates of 1765-92 to assess co-IP with HDAC2 (arrowed) and MTA1.InputIP: Mouse IgGIP: HDAC2UnboundSYO-1InputIP: Mouse IgGIP: HDAC2UnboundInputIP: Mouse IgGIP: HDAC2Unbound402-91 1765-92IB: DDIT3 (red)IB: FUS (green)IB: HDAC2A156    Figure 4.8 Reciprocal co-immunoprecipitation validates FUS-DDIT3 interaction with KDM1A IPs of FUS-DDIT3 and KDM1A were carried out in whole cell lysates of 402-91 and 1765-92 myxoid liposarcoma cells. For detection of FUS-DDIT3 in the DDIT3 IP (2nd row of blots), FUS was detected in the green channel, DDIT3 in the red channel. A merged yellow band represents FUS-DDIT3. The KDM1A band is arrowed in the first row of blots. For the KDM1A IP, FUS-DDIT3 (arrowed in the 3rd row of blots) was detected using an anti-DDIT3 antibody at the correct molecular size range. The much smaller wild type DDIT3 is not expressed in these cell lines under regular culture conditions (Figure 2.1A).157    Figure 4.9 Co-immunoprecipitation validates FUS-DDIT3 interaction with SWI/SNF components Nuclear lysates of 402-91 and 1765-92 myxoid liposarcoma cells were prepared for immunoprecipitation (IP) of FUS-DDIT3, which was confirmed by immunoblotting of FUS (green channel) and DDIT3 (red channel) through the appearance of a yellow (merged) FUS-DDIT3 band (last row of blots). Co-immunoprecipitation was observed for SWI/SNF components BRG1 (SMARCA4), BAF155 (SMARCC1) (arrowed in 2nd row of blots) and BAF57 (SMARCE1) (arrowed in 3rd row of blots).158  A   B   Figure 4.10 Myxoid liposarcoma tumors do not show loss of H3K27 trimethylation (A) Western blot analysis of H3K27me3 (red channel) and total H3 (green channel) in histones extracted from frozen tumors consisting of 19 myxoid liposarcomas (MLS), 6 malignant peripheral nerve sheath tumours (MPNST), and 2 schwannomas. Insufficient amounts of total histones were extracted from MPNST sample VSA260. (B) Densitometric analysis of H3K27 trimethylation levels normalized to total H3 expression.IB: H3 + H3K27me3IB: H3IB: H3K27me3VSA 298VSA 111VSA 135VSA 250VSA 260VSA 94VSA 146VSA 116VSA 14VSA 265VSA 249VSA 340VSA 18VSA 305VSA 81VSA320VSA 202VSA 280VSA 192VSA 175VSA 97VSA 275VSA 185VSA 286VSA 182VSA 4VSA 211Schwannoma MPNST Myxoid liposarcomaVSA 298VSA 111VSA 135VSA 250VSA 260VSA 94VSA 146VSA 116VSA 14VSA 265VSA 249VSA 340VSA 18VSA 305VSA 81VSA 320VSA 202VSA 280VSA 192VSA 175VSA 97VSA 275VSA 185VSA 286VSA 182VSA 4VSA 2110.00.20.40.60.8H3K27me3 expression in clinical specimensSpecimen numberRelative H3K27me3 WB signal(normalized to total H3)SchwannomaMPNSTMLS159  A    IB: H3 + H3K27me1IB: H3 + H3K27me3IB: H3 + H3K27acIB: H3 + H3K4me3IB: FUS + DDIT3IB: GAPDHCtrlDDIT3#1DDIT3#548 h10 nM siRNA CtrlDDIT3#1DDIT3#5 CtrlDDIT3#1DDIT3#5 VectorFUS-DDIT3HeLa402-91 1765-92 DL-221160  B   Figure 4.11 FUS-DDIT3 knockdown or exogenous expression for 48 hr does not consistently alter H3K4me3, H3K27me1/3 and H3K27ac histone marks (A) Western blot analysis of histone modifications in myxoid liposarcoma cells 402-91, 1765-92, and DL-221 after transfection for 48 hr with 10 nM control or DDIT3 siRNA, and 48 hr transfection of HeLa cells with vector control or FUS-DDIT3 plasmid. FUS-DDIT3 levels were assessed by western blot in whole cell lysates (30 µg) to confirm knockdown or exogenous expression. FUS was detected in the green channel, and DDIT3 in red to identify FUS-DDIT3 as a merged yellow band. Total histone and levels of histone modifications were assessed from histone extracts (7.5 µg). (B) Densitometric analysis of histone modification levels normalized against total histone H3 expression from the western blots.4 02 -9 11 76 5-92D L-22 10 .00 .10 .20 .30 .40 .5H 3 K 2 7 m e 1Relative H3K27me1 WB signal(normalized to total H3)C o n tro l s iR N AD D IT 3  s iR N A  # 1D D IT 3  s iR N A  # 54 02 -9 11 76 5-92D L-22 10 .00 .20 .40 .60 .81 .0H 3 K 2 7 m e 3Relative H3K27me3 WB signal(normalized to total H3)C o n tro l s iR N AD D IT 3  s iR N A  # 1D D IT 3  s iR N A  # 54 02 -9 11 76 5-92D L-22 10 .00 .20 .40 .6H 3 K 2 7 a cRelative H3K27ac WB signal(normalized to total H3)C o n tro l s iR N AD D IT 3  s iR N A  # 1D D IT 3  s iR N A  # 54 02 -9 11 76 5-92D L-22 10 .00 .10 .20 .30 .4H 3 K 4 m e 3Relative H3K4me3 WB signal(normalized to total H3)C o n tro l s iR N AD D IT 3  s iR N A  # 1D D IT 3  s iR N A  # 5H 3K 27me 1H 3K 27me 3H 3K 27 acH 3K 4me 30 .00 .20 .40 .60 .8HeLaRelative H3 mod WB signal(normalized to total H3)V e c to rF U S -D D IT 3 -H A161  A   B  C  D   Figure 4.12 FUS-DDIT3 knockdown for 3 or 7 days does not consistently alter H3K27 acetylation or trimethylation Western blot analysis of histone extracts for (A) H3K27me3 or (C) H3K27ac after 3 days or 7 days of treatment with 10 nM control or DDIT3 siRNA (#1 and #5) in myxoid liposarcoma cells 402-91 and 1765-92. (B, D) Densitometric analysis of H3K27me (A) and H3K27ac (B) levels normalized to total H3.CtrlDDIT3#1DDIT3#53-day knockdownCtrlDDIT3#1DDIT3#53-day knockdown402-91 1765-92IB: H3 + H3K27me3IB: H3IB: H3K27me310 nM siRNA CtrlDDIT3#1DDIT3#57-day knockdownCtrlDDIT3#1DDIT3#57-day knockdownDay 3Day 70.00.51.01.52.02.5FUS-DDIT3 knockdown in 402-91Relative H3K27me3 WB signal(normalized to total H3)Control siRNADDIT3 siRNA #1DDIT3 siRNA #5Day 3Day 70.00.51.01.52.02.5FUS-DDIT3 knockdown in 1765-92Relative H3K27me3 WB signal(normalized to total H3)Control siRNADDIT3 siRNA #1DDIT3 siRNA #5CtrlDDIT3#1DDIT3#53-day knockdownCtrlDDIT3#1DDIT3#53-day knockdown402-91 1765-92IB: H3 + H3K27acIB: H3IB: H3K27ac10 nM siRNA CtrlDDIT3#1DDIT3#57-day knockdownCtrlDDIT3#1DDIT3#57-day knockdownDay 3Day 70.00.51.01.5FUS-DDIT3 knockdown in 402-91Relative H3K27ac WB signal(normalized to total H3)Control siRNADDIT3 siRNA #1DDIT3 siRNA #5Day 3Day 70.00.51.01.5FUS-DDIT3 knockdown in 1765-92Relative H3K27ac WB signal(normalized to total H3)Control siRNADDIT3 siRNA #1DDIT3 siRNA #5162   Figure 4.13 FUS-DDIT3 knockdown increases trimethylation and decreases acetylation of H3K27 at the PTX3 promoter Myxoid liposarcoma cells 402-91 were treated with 10 nM control or DDIT3 siRNA (#1 and #5) for 48 hr. Chromatin immunoprecipitation was carried out for (A) FUS-DDIT3 using an anti-DDIT3 antibody (clone L63F7), (B) H3K27ac, and (C) H3K27me3, to investigate their occupancy of the PTX3 promoter, as well as at regions upstream and downstream of the promoter. Error bars = standard deviation.  UpstreamPromoterDownstream051015ChIP: DDIT3PTX3Fold changeover IgG controlControl siRNADDIT3 siRNA #1DDIT3 siRNA #5UpstreamPromoterDownstream0246ChIP: H3K27me3PTX3Fold changeover IgG controlControl siRNADDIT3 siRNA #5DDIT3 siRNA #1UpstreamPromoterDownstream051015ChIP: H3K27acPTX3Fold changeover IgG controlControl siRNADDIT3 siRNA #5DDIT3 siRNA #1ABC163  Chapter 5: Discussion, Conclusion and Future Directions 5.1 Summary Myxoid liposarcoma is driven by the fusion oncoprotein FUS-DDIT3, but the exact mechanism of action behind its capacity for transformation is unclear, presenting a challenge for designing new targeted therapies specific for this cancer. Identification of the FUS-DDIT3 interactome is crucial to gaining a better understanding of its mechanism of action, and would also provide functionally relevant secondary targets against which rational targeted therapeutic strategies could be designed, given the lack of any available agents that target FUS-DDIT3 directly. While a previous study identified the interactome of N-terminal portions of the FET proteins (Thomsen et al., 2013), this thesis is the first report on the FUS-DDIT3 interactome in myxoid liposarcoma based on comprehensive proteomics, and presents the following novel findings: (1) A large portion of the interactome consists of proteins involved in RNA processing and splicing, but FUS-DDIT3 does not appear to regulate alternative splicing; (2) FUS-DDIT3 associates with members of multiple chromatin regulatory complexes, and (3) FUS-DDIT3 alters H3K27 acetylation and trimethylation at the promoter of one of its direct gene targets, PTX3.  5.2 Insights from the FUS-DDIT3 interactome One of the largest groups of proteins in the FUS-DDIT3 interactome identified from whole cell lysates in Chapter 1 is involved in RNA processing (Table 2.4 and Figure 4.6A), but the RNA-seq data from Chapter 3 showed no evidence of global changes to the alternative splicing profile after 24 hr or 48 hr FUS-DDIT3 knockdown (Figures 3.3 and 3.4). However, it is possible that FUS-DDIT3 may still affect low level alternative splicing events. Such events could 164  be detected through the use of additional replicates, a longer FUS-DDIT3 knockdown duration, and sequencing RNA at greater depth. Since the RNA-binding domains of FUS are not retained in most FUS-DDIT3 variants (Figures 1.3 and 1.4), the fusion protein is not expected to retain the same splicing functions of wild type FUS. While this was not the research question addressed by this thesis, it is possible that the loss of one normal FUS allele due to the translocation event could impact normal FUS-mediated splicing functions in the cell. However, based on the demonstration of an increased mRNA and protein expression of the remaining FUS allele from an unknown compensatory mechanism in myxoid liposarcoma primary cell lines (Spitzer et al., 2011), loss of splicing function due to reduced wild type FUS expression appears unlikely. The next question is whether FUS-DDIT3 exerts a dominant negative effect (at the protein level) over wild type FUS splicing activities. How FUS-DDIT3 would effect a dominant negative function is unclear, but it might occur through the loss of C-terminal FUS interactors, such as YB-1. YB-1 was reported to mediate alternative splicing of adenovirus E1A pre-mRNA by FUS, but this activity was blocked by the presence of FUS-DDIT3, which does not interact with YB-1 (Rapp et al., 2002). Alternatively, the interaction of FUS-DDIT3 with multimers of wild type FUS (Thomsen et al., 2013) might result in improper sequestration and functional inactivation of FUS. Even if this was the case, the lack of global changes to the alternative splicing profile in my data suggests that any dominant negative loss of function in FUS-DDIT3 will be limited to a small set of mRNA targets, and is unlikely to be a major contributor to the oncogenic function of FUS-DDIT3. Therefore, the relatively large presence of RNA processing proteins in the FUS-DDIT3 interactome is probably the by-product of FUS-DDIT3 binding to wild type FUS (Thomsen et al., 2013) (Figure 2.1A), resulting in the inclusion of RNA processing proteins that associate 165  with FUS. The identification of functionally relevant FUS-DDIT3 interactors is then complicated by the fact that a large number of RNA binding and processing proteins are “housekeeping proteins” that are constitutively expressed at relatively high levels (Eisenberg and Levanon, 2013; Lehner and Fraser, 2004). In standard IP-MS experiments from whole cell lysates, such housekeeping proteins will mask detection of the comparatively less abundant, but functionally critical transcriptional / chromatin regulatory proteins in the FUS-DDIT3 interactome, by virtue of massive overrepresentation of the housekeeping protein-derived peptides. Therefore, nuclear lysates were used in Chapter 4 for a TMT-labeled IP-MS experiment designed to improve detection and quantification accuracy of these less abundant proteins. The FUS-DDIT3 interactome derived from nuclear lysates contained protein members from multiple chromatin regulatory complexes including the NuRD, SWI/SNF, PRC1, PRC2 and MLL1 COMPASS-like complexes (Figures 4.4, 4.5 and 4.6B). Co-immunoprecipitation experiments further validated associations of FUS-DDIT3 with some of the specific proteins in these complexes including BRG1/SMARCA4, BAF155/SMARCC1, BAF57/SMARCE1, HDAC2, KDM1A, and MTA1.  There are, however, limitations to the techniques used in this study. Affinity pull downs, such as the immunoprecipitation approach employed in Chapters 2 and 4 prior to mass spectrometry analyses, are widely used and generally reliable for detection of stable protein-protein interactions, but might miss transient (but potentially still important) interactors of FUS-DDIT3. This limitation could potentially be overcome by employing more recent proximity-dependent labeling technologies such as biotin identification (BioID) and engineered ascorbate peroxidase (APEX) that are highly sensitive and have a small labeling radius (Chen and Perrimon, 2017). Despite their advantage for detecting transient interactors, BioID and APEX experiments have additional complexities that bring their own limitations. These include the 166  need to exogenously express a bait protein that is fused to biotin ligase (for BioID) or ascorbate peroxidase (for APEX) at a non-physiological level, the potential for mislocalization of the fusion bait protein that results in false positives, the proper delivery of enzyme substrates to the location of the bait protein, and the critical need for multiple replicates as the high sensitivity of these technologies can result in labeling due to random collisions (Chen and Perrimon, 2017; P. Li et al., 2017). Nevertheless, the current identification of multiple chromatin regulatory complexes in the FUS-DDIT3 interactome provides a relevant and important basis for future functional validation studies. Without comprehensive ChIP-seq data on the identities of FUS-DDIT3 targets, it will be challenging to study whether and how FUS-DDIT3 might exert its oncogenic functions by associating with chromatin regulators and altering the chromatin state of its targets. However, evidence from recent studies supports the hypothesis that epigenetic deregulation could be an important oncogenic mechanism in myxoid liposarcoma. A study on the DNA methylation pattern of myxoid liposarcoma tumors revealed a pattern that was distinct to myxoid liposarcoma compared to the other sarcoma types tested (Renner et al., 2013). One of a short literature-supported list of potential target genes for epigenetic regulation by FUS-DDIT3 is PTX3, reported by Forni et al. (2009) to be upregulated by FUS-DDIT3. Although my experimental knockdown of FUS-DDIT3 did reduce acetylation of H3K27 at the promoter of PTX3 (Figure 4.13), there were no global changes to H3K27 acetylation and H3K4 trimethylation (Figures 4.11 and 4.12), as was observed after reduction of EWS-FLI1 in Ewing sarcoma (Tomazou et al., 2015).  HDAC inhibitors, in particular, have shown promising results against myxoid liposarcoma. A recent drug screen reported that myxoid liposarcoma cell lines were highly responsive to the HDAC inhibitors quisinostat, dacinostat, and panobinostat (de Graaff et al., 167  2017). In a Phase II clinical trial for another HDAC inhibitor, pracinostat, stable disease was achieved in three out of four assessable myxoid liposarcoma patients with recurrent / metastatic disease (Chu et al., 2015). Since HDAC inhibitors should increase acetylation levels, the efficacy of HDAC inhibitors appears counterintuitive to the observation of FUS-DDIT3 increasing acetylation of H3K27 on the PTX3 promoter (Figure 4.13). However, the effect of FUS-DDIT3 on histone acetylation on other targets is unknown, and could be different. Moreover, the HDAC inhibitors could be acting on multiple pathways owing to the promiscuity of HDACs in targeting different histone residues as well as non-histone proteins (Seto and Yoshida, 2014). Ultimately, characterizing the FUS-DDIT3 interactome provides a basis to guide subsequent evaluations about which chromatin regulators are critical to FUS-DDIT3, and how their oncogenic functions could be modulated by existing or future therapies.  5.3 Translocation-associated sarcomas as epigenetic diseases The possibility of myxoid liposarcoma being an epigenetic disease is interesting and fits with emerging concepts in other translocation-associated sarcomas such as Ewing sarcoma, synovial sarcoma and alveolar rhabdomyosarcoma (Banito et al., 2018; Boulay et al., 2017; Böhm et al., 2016; Kadoch and Crabtree, 2013; Riggi et al., 2014; Sankar et al., 2014; Su et al., 2012). The relatively recent discovery that a differentiated cell can be reprogrammed with a defined set of transcription factors into an induced pluripotent stem cell (Hanna et al., 2010; Takahashi and Yamanaka, 2006) highlighted the potentially powerful consequences of altering gene regulatory mechanisms in abnormal developmental contexts. Oncogenic transformation often involves disruptions to differentiation and developmental programs that are analogous to 168  cellular reprogramming, as many oncogenes and tumors suppressors are also effectors and modulators of cellular reprogramming (Suva et al., 2013).  Similarly, chromatin regulators also have established roles in cellular reprogramming and oncogenesis (Nebbioso et al., 2018; Orkin and Hochedlinger, 2011). It has been proposed that all classical hallmarks of human cancer (Hanahan and Weinberg, 2011; 2000) can be fulfilled entirely through chromatin-mediated epigenetic deregulation, providing a possible mechanistic explanation for tumors that arise with few or no recurrent mutations (Flavahan et al., 2017). This is well demonstrated by rhabdoid tumor, which is an aggressive pediatric tumor characterized by biallelic inactivation of SMARCB1, and displays one of the lowest mutational burden of all tumors (Jamshidi et al., 2016; Lee et al., 2012). Similarly, the reported genomic landscapes for translocation-associated sarcomas such as myxoid liposarcoma, Ewing sarcoma, synovial sarcoma all show a relatively low mutational burden (Barretina et al., 2010; Brohl et al., 2014; Cancer Genome Atlas Research Network, 2017; Chalmers et al., 2017; Crompton et al., 2014; Joseph et al., 2014; Tirode et al., 2014; Vlenterie et al., 2015). It is, therefore, tempting to speculate that epigenetic deregulation could be a common and unifying feature among most or all translocation-associated, fusion oncoprotein sarcomas, resulting in large-scale epigenetic reprogramming that disrupts normal developmental processes and promotes tumorigenesis. While their mechanisms of action are not fully understood, the better characterized fusion oncoproteins appear to share a similar mechanism: recruiting and mistargeting chromatin regulatory complexes through direct protein or DNA interactions. Since the interaction of EWSR1 with SWI/SNF is mediated through EWSR1's N-terminal prion-like domain (Boulay et al., 2017), a motif which is also found in the FUS or EWSR1 portions of many FET-associated fusion oncoproteins (Figure 1.1), it is highly possible that these fusion oncoproteins could also be recruiting SWI/SNF to their specific DNA targets. 169  Identifying a common underlying epigenetic mechanism of action for translocation-associated sarcomas is clinically relevant because epigenetic aberrations, unlike genetic mutations, are potentially reversible by existing and emerging epigenetic drugs, many of which are already in advanced preclinical studies or phase 1-3 clinical trials (Simó-Riudalbas and Esteller, 2015). However, clinical trial design and accrual for translocation-associated sarcomas have proven challenging due to the rarity of each individual sarcoma type. A shared mechanism of action would allow the design of basket trials for these individually rare cancers, where the focus is not on patients with a single disease histology, but on a common genomic alteration or class of alterations, increasing the number of trial patients available for statistical analyses while maintaining a strong biological rationale for possible efficacy (Simon, 2017).  5.4 Future directions The development of FUS-DDIT3 targeted therapies is beyond the scope of this study, but the lack of data about its key protein interactions represents a major gap in our knowledge of this fusion oncoprotein's sarcomagenic mechanism, and presents a major roadblock to evaluating the functional implications of FUS-DDIT3's association with different chromatin regulatory complexes. Does the fusion protein recruit different complexes to different target loci? What is the impact of this recruitment on histone modifications and chromatin accessibility? Which of these changes are crucial to the fusion protein’s oncogenic function? One hypothesis could be that FUS-DDIT3 exists in different protein complexes in which the specific composition of protein partners influences which genes are targeted and how they are regulated (Figure 5.1). Some of these chromatin regulatory components might not otherwise be recruited in the absence of FUS-DDIT3. 170  Acquisition of relevant granular details to address this hypothesis will require ChIP-seq studies before and after FUS-DDIT3 knockdown and/or induction, not only to identify the genomic targets of FUS-DDIT3, but also the specific chromatin regulatory complexes that co-localize with FUS-DDIT3 at its different targets, and how that may affect local histone modifications and, by extension, chromatin structure. Any changes to chromatin accessibility or even higher order chromatin organization as a result of FUS-DDIT3-mediated alterations of chromatin remodeling activities can be assessed with newer technologies in this fast-moving field that can map open chromatin and perform chromosome conformation capture to determine longer distance co-associations (Jia et al., 2017). Combining data from chromosome states with RNA-seq analyses before and after FUS-DDIT3 knockdown / induction should provide insights into the role of FUS-DDIT3 in chromatin regulation and its relation to gene expression. Information from such studies would also guide the identification of functionally crucial chromatin regulators and, ultimately, the selection of epigenetic drugs that can target them. To accelerate research in myxoid liposarcoma, it will be important to also identify the cell of origin for myxoid liposarcoma and/or a physiologically-appropriate Cre-driver in order to generate a conditional mouse model that recapitulates the human disease, as current models of myxoid liposarcoma are unsatisfactory in many ways. The successful generation of mouse models is also dependent on a better understanding of the underlying FUS-DDIT3 biology, as evidenced by the difficulty multiple laboratories have found in their attempts to generate EWS-FLI-driven Ewing sarcoma mouse models due to the lack of evolutionary conservation in EWS-FLI1 target motifs between human and mouse (Minas et al., 2017). Nevertheless, improved models are needed for in vivo validation and drug testing in preparation for translation of the in vitro findings into the clinic. 171  Finally, beyond myxoid liposarcoma, it is likely that fusion proteins in other translocation-associated sarcomas involving FET proteins (including Ewing and clear cell sarcomas, extraskeletal myxoid chondrosarcoma and desmoplastic small round cell tumors, among others) employ a similar mechanism of action. This provides a compelling rationale to investigate the existence of similar epigenetic mechanism of action in these individually rare and under-investigated sarcomas that typically affect young people, where appropriately targeted therapies, administered in these diseases with low mutational burdens, could prove particularly effective.172   Figure 5.1 Potential mechanism of action for how FUS-DDIT3 may deregulate the function of chromatin regulatory complexes FUS-DDIT3 may exist in multiple distinct protein complexes. Targeting of different FUS-DDIT3 complexes to different gene regulatory elements is likely dependent on the specific basic zipper (bZIP) dimerization partner of the DDIT3 portion of the fusion protein. The different composition of protein partners assembled in each complex influences the chromatin regulatory complex that is recruited to the target loci and the resulting effect on gene regulation and expression. For example, (A) a particular bZIP partner (blue oval) may result in the localization of a FUS-DDIT3-containing complex to the regulatory element of Gene A, as well as the recruitment of the SWI/SNF chromatin remodeling complex that ultimately results in the activation and expression of Gene A. (B) Conversely, a different bZIP partner (red oval) may localize a different FUS-DDIT3-containing complex to the regulatory element of another gene (Gene B), that also results in the recruitment of a different chromatin regulatory complex, in this case, the NuRD repressor complex, with the final effect of repressing the expression of Gene B. Some common partners may be present in both complexes which could include some of the interactors identified in Chapter 2.     bZIPdimerization partnerNuRDSWI/SNFbZIPdimerization partnerGene AFUS-DDIT3Gene BGene regulatory elementGene regulatory elementFUS-DDIT3AB173  References Ablain, J., Leiva, M., Peres, L., Fonsart, J., Anthony, E., De Thé, H., 2013. Uncoupling RARA transcriptional activation and degradation clarifies the bases for APL response to therapies. J. Exp. Med. 210, 647–653. Ablain, J., Rice, K., Soilihi, H., de Reyniès, A., Minucci, S., De Thé, H., 2014. Activation of a promyelocytic leukemia-tumor protein 53 axis underlies acute promyelocytic leukemia cure. Nat. Med. 20, 167–174. Adams, M.D., Dubnick, M., Kerlavage, A.R., Moreno, R., Kelley, J.M., Utterback, T.R., Nagle, J.W., Fields, C., Venter, J.C., 1992. Sequence identification of 2,375 human brain genes. Nature 355, 632–634. Adelmant, G., Gilbert, J.D., Freytag, S.O., 1998. Human translocation liposarcoma-CCAAT/enhancer binding protein (C/EBP) homologous protein (TLS-CHOP) oncoprotein prevents adipocyte differentiation by directly interfering with C/EBPbeta function. J. Biol. Chem. 273, 15574–15581. Ali, A.T., Hochfeld, W.E., Myburgh, R., Pepper, M.S., 2013. Adipocyte and adipogenesis. Eur. J. Cell Biol. 92, 229–236. Allen, H.F., Wade, P.A., Kutateladze, T.G., 2013. The NuRD architecture. Cell. Mol. Life Sci. 70, 3513–3524. Allis, C.D., Jenuwein, T., 2016. The molecular hallmarks of epigenetic control. Nat. Rev. Genet. 17, 487–500. Alver, B.H., Kim, K.H., Lu, P., Wang, X., Manchester, H.E., Wang, W., Haswell, J.R., Park, P.J., Roberts, C.W.M., 2017. The SWI/SNF chromatin remodelling complex is required for maintenance of lineage specific enhancers. Nat Commun 8, 14648. Aman, P., Dolatabadi, S., Svec, D., Jonasson, E., Safavi, S., Andersson, D., Grundevik, P., Thomsen, C., Ståhlberg, A., 2016. Regulatory mechanisms, expression levels and proliferation effects of the FUS-DDIT3 fusion oncogene in liposarcoma. J. Pathol. 238, 689–699. Aman, P., Ron, D., Mandahl, N., Fioretos, T., Heim, S., Arheden, K., Willén, H., Rydholm, A., Mitelman, F., 1992. Rearrangement of the transcription factor gene CHOP in myxoid liposarcomas with t(12;16)(q13;p11). Genes Chromosomes Cancer 5, 278–285. Ammirati, M., Bagley, S.W., Bhattacharya, S.K., Buckbinder, L., Carlo, A.A., Conrad, R., Cortes, C., Dow, R.L., Dowling, M.S., El-Kattan, A., Ford, K., Guimarães, C.R.W., Hepworth, D., Jiao, W., LaPerle, J., Liu, S., Londregan, A., Loria, P.M., Mathiowetz, A.M., Munchhof, M., Orr, S.T.M., Petersen, D.N., Price, D.A., Skoura, A., Smith, A.C., Wang, J., 2015. Discovery of an in Vivo Tool to Establish Proof-of-Concept for MAP4K4-Based Antidiabetic Treatment. ACS Med Chem Lett 6, 1128–1133. Andersson, M.K., Ståhlberg, A., Arvidsson, Y., Olofsson, A., Semb, H., Stenman, G., Nilsson, O., Aman, P., 2008. The multifunctional FUS, EWS and TAF15 proto-oncoproteins show cell type-specific expression patterns and involvement in cell spreading and stress response. BMC Cell Biol. 9, 37. Ankney, J.A., Muneer, A., Chen, X., 2018. Relative and Absolute Quantitation in Mass Spectrometry-Based Proteomics. Annu Rev Anal Chem (Palo Alto Calif) 11, 49–77. Antonescu, C.R., Tschernyavsky, S.J., Decuseara, R., Leung, D.H., Woodruff, J.M., Brennan, M.F., Bridge, J.A., Neff, J.R., Goldblum, J.R., Ladanyi, M., 2001. Prognostic impact of P53 status, TLS-CHOP fusion transcript structure, and histological grade in myxoid liposarcoma: a molecular and clinicopathologic study of 82 cases. 7, 3977–3987. 174  Asano, N., Yoshida, A., Ichikawa, H., Mori, T., Nakamura, M., Kawai, A., Hiraoka, N., 2017. Immunohistochemistry for trimethylated H3K27 in the diagnosis of malignant peripheral nerve sheath tumours. Histopathology 70, 385–393. Babu, M.M., 2016. The contribution of intrinsically disordered regions to protein function, cellular complexity, and human disease. Biochem. Soc. Trans. 44, 1185–1200. Baechtold, H., Kuroda, M., Sok, J., Ron, D., Lopez, B.S., Akhmedov, A.T., 1999. Human 75-kDa DNA-pairing protein is identical to the pro-oncoprotein TLS/FUS and is able to promote D-loop formation. J. Biol. Chem. 274, 34337–34342. Bai, L., Morozov, A.V., 2010. Gene regulation by nucleosome positioning. Trends Genet. 26, 476–483. Banito, A., Li, X., Laporte, A.N., Roe, J.-S., Sanchez-Vega, F., Huang, C.-H., Dancsok, A.R., Hatzi, K., Chen, C.-C., Tschaharganeh, D.F., Chandwani, R., Tasdemir, N., Jones, K.B., Capecchi, M.R., Vakoc, C.R., Schultz, N., Ladanyi, M., Nielsen, T.O., Lowe, S.W., 2018. The SS18-SSX Oncoprotein Hijacks KDM2B-PRC1.1 to Drive Synovial Sarcoma. Cancer Cell 33, 527–541.e8. Baralle, F.E., Giudice, J., 2017. Alternative splicing as a regulator of development and tissue identity. Nat Rev Mol Cell Biol 18, 437–451. Barretina, J., Taylor, B.S., Banerji, S., Ramos, A.H., Lagos-Quintana, M., Decarolis, P.L., Shah, K., Socci, N.D., Weir, B.A., Ho, A., Chiang, D.Y., Reva, B., Mermel, C.H., Getz, G., Antipin, Y., Beroukhim, R., Major, J.E., Hatton, C., Nicoletti, R., Hanna, M., Sharpe, T., Fennell, T.J., Cibulskis, K., Onofrio, R.C., Saito, T., Shukla, N., Lau, C., Nelander, S., Silver, S.J., Sougnez, C., Viale, A., Winckler, W., Maki, R.G., Garraway, L.A., Lash, A., Greulich, H., Root, D.E., Sellers, W.R., Schwartz, G.K., Antonescu, C.R., Lander, E.S., Varmus, H.E., Ladanyi, M., Sander, C., Meyerson, M., Singer, S., 2010. Subtype-specific genomic alterations define new targets for soft-tissue sarcoma therapy. Nat. Genet. 42, 715–721. Barski, A., Cuddapah, S., Cui, K., Roh, T.-Y., Schones, D.E., Wang, Z., Wei, G., Chepelev, I., Zhao, K., 2007. High-resolution profiling of histone methylations in the human genome. Cell 129, 823–837. Beck, M., Schmidt, A., Malmstroem, J., Claassen, M., Ori, A., Szymborska, A., Herzog, F., Rinner, O., Ellenberg, J., Aebersold, R., 2011. The quantitative proteome of a human cell line. Mol. Syst. Biol. 7, 549–549. Beekman, R., Amador, V., Campo, E., 2018. SOX11, a key oncogenic factor in mantle cell lymphoma. Curr. Opin. Hematol. 25, 299–306. Beisel, C., Buness, A., Roustan-Espinosa, I.M., Koch, B., Schmitt, S., Haas, S.A., Hild, M., Katsuyama, T., Paro, R., 2007. Comparing active and repressed expression states of genes controlled by the Polycomb/Trithorax group proteins. Proc. Natl. Acad. Sci. U.S.A. 104, 16615–16620. Bentley, G.A., Lewit-Bentley, A., Finch, J.T., Podjarny, A.D., Roth, M., 1984. Crystal structure of the nucleosome core particle at 16 A resolution. 176, 55–75. Bento, C., Andersson, M.K., Aman, P., 2009. DDIT3/CHOP and the sarcoma fusion oncoprotein FUS-DDIT3/TLS-CHOP bind cyclin-dependent kinase 2. BMC Cell Biol. 10, 89. Berg, M.G., Singh, L.N., Younis, I., Liu, Q., Pinto, A.M., Kaida, D., Zhang, Z., Cho, S., Sherrill-Mix, S., Wan, L., Dreyfuss, G., 2012. U1 snRNP determines mRNA length and regulates isoform expression. Cell 150, 53–64. Bertolotti, A., Bell, B., Tora, L., 1999. The N-terminal domain of human TAFII68 displays transactivation and oncogenic properties. Oncogene 18, 8000–8010. 175  Bertrand, P., Akhmedov, A.T., Delacote, F., Durrbach, A., Lopez, B.S., 1999. Human POMp75 is identified as the pro-oncoprotein TLS/FUS: both POMp75 and POMp100 DNA homologous pairing activities are associated to cell proliferation. Oncogene 18, 4515–4521. Biegel, J.A., Zhou, J.Y., Rorke, L.B., Stenstrom, C., Wainwright, L.M., Fogelgren, B., 1999. Germ-line and acquired mutations of INI1 in atypical teratoid and rhabdoid tumors. Cancer Res. 59, 74–79. Bird, A., 2007. Perceptions of epigenetics. Nature 447, 396–398. Blackledge, N.P., Rose, N.R., Klose, R.J., 2015. Targeting Polycomb systems to regulate gene expression: modifications to a complex story. Nat Rev Mol Cell Biol 16, 643–649. Blencowe, B.J., 2006. Alternative splicing: new insights from global analyses. Cell 126, 37–47. Bode-Lesniewska, B., Frigerio, S., Exner, U., Abdou, M.T., Moch, H., Zimmermann, D.R., 2007. Relevance of translocation type in myxoid liposarcoma and identification of a novel EWSR1-DDIT3 fusion. Genes Chromosomes Cancer 46, 961–971. Bonfanti, M., La Valle, E., Fernandez Sousa Faro, J.M., Faircloth, G., Caretti, G., Mantovani, R., D'Incalci, M., 1999. Effect of ecteinascidin-743 on the interaction between DNA binding proteins and DNA. Anticancer Drug Des. 14, 179–186. Boone, M., De Koker, A., Callewaert, N., 2018. Capturing the “ome”: the expanding molecular toolbox for RNA and DNA library construction. Nucleic Acids Res. 46, 2701–2721. Bornelöv, S., Reynolds, N., Xenophontos, M., Gharbi, S., Johnstone, E., Floyd, R., Ralser, M., Signolet, J., Loos, R., Dietmann, S., Bertone, P., Hendrich, B., 2018. The Nucleosome Remodeling and Deacetylation Complex Modulates Chromatin Structure at Sites of Active Transcription to Fine-Tune Gene Expression. Molecular Cell 71, 56–72. Boulay, G., Sandoval, G.J., Riggi, N., Iyer, S., Buisson, R., Naigles, B., Awad, M.E., Rengarajan, S., Volorio, A., McBride, M.J., Broye, L.C., Zou, L., Stamenkovic, I., Kadoch, C., Rivera, M.N., 2017. Cancer-Specific Retargeting of BAF Complexes by a Prion-like Domain. Cell. Böhm, M., Wachtel, M., Marques, J.G., Streiff, N., Laubscher, D., Nanni, P., Mamchaoui, K., Santoro, R., Schäfer, B.W., 2016. Helicase CHD4 is an epigenetic coregulator of PAX3-FOXO1 in alveolar rhabdomyosarcoma. J Clin Invest 126, 4237–4249. Brennan, M.F., Antonescu, C.R., Moraco, N., Singer, S., 2014. Lessons learned from the study of 10,000 patients with soft tissue sarcoma. Ann. Surg. 260, 416–21. Brenner, S., Bercovich, Z., Feiler, Y., Keshet, R., Kahana, C., 2015. Dual Regulatory Role of Polyamines in Adipogenesis. J. Biol. Chem. 290, 27384–27392. Brohl, A.S., Kahen, E., Yoder, S.J., Teer, J.K., Reed, D.R., 2017. The genomic landscape of malignant peripheral nerve sheath tumors: diverse drivers of Ras pathway activation. Sci Rep 7, 14992. Brohl, A.S., Solomon, D.A., Chang, W., Wang, J., Song, Y., Sindiri, S., Patidar, R., Hurd, L., Chen, L., Shern, J.F., Liao, H., Wen, X., Gerard, J., Kim, J.-S., Lopez Guerrero, J.A., Machado, I., Wai, D.H., Picci, P., Triche, T., Horvai, A.E., Miettinen, M., Wei, J.S., Catchpool, D., Llombart-Bosch, A., Waldman, T., Khan, J., 2014. The genomic landscape of the Ewing Sarcoma family of tumors reveals recurrent STAG2 mutation. PLoS Genet. 10, e1004475. Brunner, A.L., Beck, A.H., Edris, B., Sweeney, R.T., Zhu, S.X., Li, R., Montgomery, K., Varma, S., Gilks, T., Guo, X., Foley, J.W., Witten, D.M., Giacomini, C.P., Flynn, R.A., Pollack, J.R., Tibshirani, R., Chang, H.Y., van de Rijn, M., West, R.B., 2012. Transcriptional profiling of long non-coding RNAs and novel transcribed regions across a diverse panel of archived human cancers. Genome Biol 13, R75. 176  Cancer Genome Atlas Research Network, 2017. Comprehensive and Integrated Genomic Characterization of Adult Soft Tissue Sarcomas. Cell 171, 950–965.e28. Carrière, A., Carmona, M.-C., Fernandez, Y., Rigoulet, M., Wenger, R.H., Pénicaud, L., Casteilla, L., 2004. Mitochondrial reactive oxygen species control the transcription factor CHOP-10/GADD153 and adipocyte differentiation: a mechanism for hypoxia-dependent effect. J. Biol. Chem. 279, 40462–40469. Chalmers, Z.R., Connelly, C.F., Fabrizio, D., Gay, L., Ali, S.M., Ennis, R., Schrock, A., Campbell, B., Shlien, A., Chmielecki, J., Huang, F., He, Y., Sun, J., Tabori, U., Kennedy, M., Lieber, D.S., Roels, S., White, J., Otto, G.A., Ross, J.S., Garraway, L., Miller, V.A., Stephens, P.J., Frampton, G.M., 2017. Analysis of 100,000 human cancer genomes reveals the landscape of tumor mutational burden. Genome Med 9, 34. Chang, K.T.E., Goytain, A., Tucker, T., Karsan, A., Lee, C.-H., Nielsen, T.O., Ng, T.L., 2018. Development and Evaluation of a Pan-Sarcoma Fusion Gene Detection Assay Using the NanoString nCounter Platform. J Mol Diagn 20, 63–77. Chang, L., Chang, M., Chang, H.M., Chang, F., 2018. Microsatellite Instability: A Predictive Biomarker for Cancer Immunotherapy. Appl. Immunohistochem. Mol. Morphol. 26, e15–e21. Charytonowicz, E., Terry, M., Coakley, K., Telis, L., Remotti, F., Cordon-Cardo, C., Taub, R.N., Matushansky, I., 2012. PPARγ agonists enhance ET-743-induced adipogenic differentiation in a transgenic mouse model of myxoid round cell liposarcoma. J Clin Invest. Chen, C.-L., Perrimon, N., 2017. Proximity-dependent labeling methods for proteomic profiling in living cells. Wiley Interdiscip Rev Dev Biol 6, e272. Chen, L., Bush, S.J., Tovar-Corona, J.M., Castillo-Morales, A., Urrutia, A.O., 2014. Correcting for differential transcript coverage reveals a strong relationship between alternative splicing and organism complexity. Mol. Biol. Evol. 31, 1402–1413. Cheng, H., Dodge, J., Mehl, E., Liu, S., Poulin, N., van de Rijn, M., Nielsen, T.O., 2009. Validation of immature adipogenic status and identification of prognostic biomarkers in myxoid liposarcoma using tissue microarrays. Hum. Pathol. 40, 1244–1251. Chi, B., O'Connell, J.D., Yamazaki, T., Gangopadhyay, J., Gygi, S.P., Reed, R., 2018. Interactome analyses revealed that the U1 snRNP machinery overlaps extensively with the RNAP II machinery and contains multiple ALS/SMA-causative proteins. Sci Rep 8, 8755. Chittock, E.C., Latwiel, S., Miller, T.C.R., Müller, C.W., 2017. Molecular architecture of polycomb repressive complexes. Biochem. Soc. Trans. 45, 193–205. Chu, Q.S.-C., Nielsen, T.O., Alcindor, T., Gupta, A., Endo, M., Goytain, A., Xu, H., Verma, S., Tozer, R., Knowling, M., Bramwell, V.B., Powers, J., Seymour, L.K., Eisenhauer, E.A., 2015. A Phase II study of SB939, a novel pan-histone deacetylase inhibitor, in patients with translocation-associated recurrent/metastatic sarcomas -NCIC-CTG IND 200. Ann Oncol. Chudasama, P., Mughal, S.S., Sanders, M.A., Hübschmann, D., Chung, I., Deeg, K.I., Wong, S.-H., Rabe, S., Hlevnjak, M., Zapatka, M., Ernst, A., Kleinheinz, K., Schlesner, M., Sieverling, L., Klink, B., Schröck, E., Hoogenboezem, R.M., Kasper, B., Heilig, C.E., Egerer, G., Wolf, S., Kalle, von, C., Eils, R., Stenzinger, A., Weichert, W., Glimm, H., Gröschel, S., Kopp, H.-G., Omlor, G., Lehner, B., Bauer, S., Schimmack, S., Ulrich, A., Mechtersheimer, G., Rippe, K., Brors, B., Hutter, B., Renner, M., Hohenberger, P., Scholl, C., Fröhling, S., 2018. Integrative genomic and transcriptomic analysis of leiomyosarcoma. Nat Commun 9, 144. Chung, P.W.M., Deheshi, B.M., Ferguson, P.C., Wunder, J.S., Griffin, A.M., Catton, C.N., Bell, R.S., White, L.M., Kandel, R.A., O'Sullivan, B., 2009. Radiosensitivity translates into 177  excellent local control in extremity myxoid liposarcoma: a comparison with other soft tissue sarcomas. Cancer 115, 3254–3261. Ciferri, C., Lander, G.C., Maiolica, A., Herzog, F., Aebersold, R., Nogales, E., 2012. Molecular architecture of human polycomb repressive complex 2. Elife 1, e00005. Cleven, A.H.G., Sannaa, G.A.A., Briaire-de Bruijn, I., Ingram, D.R., van de Rijn, M., Rubin, B.P., de Vries, M.W., Watson, K.L., Torres, K.E., Wang, W.-L., van Duinen, S.G., Hogendoorn, P.C.W., Lazar, A.J., Bovée, J.V.M.G., 2016. Loss of H3K27 tri-methylation is a diagnostic marker for malignant peripheral nerve sheath tumors and an indicator for an inferior survival. Mod. Pathol. 29, 582–590. Constantinidou, A., van der Graaf, W.T.A., 2017. The fate of new fosfamides in phase III studies in advanced soft tissue sarcoma. Eur. J. Cancer 84, 257–261. Creyghton, M.P., Cheng, A.W., Welstead, G.G., Kooistra, T., Carey, B.W., Steine, E.J., Hanna, J., Lodato, M.A., Frampton, G.M., Sharp, P.A., Boyer, L.A., Young, R.A., Jaenisch, R., 2010. Histone H3K27ac separates active from poised enhancers and predicts developmental state. Proc. Natl. Acad. Sci. U.S.A. 107, 21931–21936. Crompton, B.D., Stewart, C., Taylor-Weiner, A., Alexe, G., Kurek, K.C., Calicchio, M.L., Kiezun, A., Carter, S.L., Shukla, S.A., Mehta, S.S., Thorner, A.R., de Torres, C., Lavarino, C., Sunol, M., McKenna, A., Sivachenko, A., Cibulskis, K., Lawrence, M.S., Stojanov, P., Rosenberg, M., Ambrogio, L., Auclair, D., Seepo, S., Blumenstiel, B., DeFelice, M., Imaz-Rosshandler, I., Schwarz-Cruz Y Celis, A., Rivera, M.N., Rodriguez-Galindo, C., Fleming, M.D., Golub, T.R., Getz, G., Mora, J., Stegmaier, K., 2014. The Genomic Landscape of Pediatric Ewing Sarcoma. Cancer Discov CD–13–1037. Crozat, A., Aman, P., Mandahl, N., Ron, D., 1993. Fusion of CHOP to a novel RNA-binding protein in human myxoid liposarcoma. Nature 363, 640–644. Cucinotta, M., Visalli, M., Aguennouz, M., Valenti, A., Loddo, S., Altucci, L., Teti, D., 2008. Regulation of interleukin-8 gene at a distinct site of its promoter by CCAAT enhancer-binding protein homologous protein in prostaglandin E2-treated human T cells. J. Biol. Chem. 283, 29760–29769. Cvitkovic, I., Jurica, M.S., 2013. Spliceosome database: a tool for tracking components of the spliceosome. Nucleic Acids Res. 41, D132–41. Danai, L.V., Guilherme, A., Guntur, K.V., Straubhaar, J., Nicoloro, S.M., Czech, M.P., 2013. Map4k4 suppresses Srebp-1 and adipocyte lipogenesis independent of JNK signaling. Journal of Lipid Research 54, 2697–2707. de Graaff, M.A., Malu, S., Guardiola, I., Kruisselbrink, A.B., de Jong, Y., Corver, W.E., Gelderblom, H., Hwu, P., Nielsen, T.O., Lazar, A.J., Somaiah, N., Bovée, J.V.M.G., 2017. High-Throughput Screening of Myxoid Liposarcoma Cell Lines: Survivin Is Essential for Tumor Growth. Transl Oncol 10, 546–554. de Graaff, M.A., Yu, J.S.E., Beird, H.C., Ingram, D.R., Nguyen, T., Juehui Liu, J., Bolshakov, S., Szuhai, K., Aman, P., Torres, K.E., Lev, D., Nielsen, T.O., Bovée, J.V.M.G., Lazar, A.J., Somaiah, N., 2016. Establishment and characterization of a new human myxoid liposarcoma cell line (DL-221) with the FUS-DDIT3 translocation. Lab. Invest. de Klerk, E., 't Hoen, P.A.C., 2015. Alternative mRNA transcription, processing, and translation: insights from RNA sequencing. Trends Genet. 31, 128–139. De Raedt, T., Beert, E., Pasmant, E., Luscan, A., Brems, H., Ortonne, N., Helin, K., Hornick, J.L., Mautner, V., Kehrer-Sawatzki, H., Clapp, W., Bradner, J., Vidaud, M., Upadhyaya, M., Legius, E., Cichowski, K., 2014. PRC2 loss amplifies Ras-driven transcription and confers sensitivity to BRD4-based therapies. Nature 514, 247–251. 178  Debrock, G., Vanhentenrijk, V., Sciot, R., Debiec-Rychter, M., Oyen, R., Van Oosterom, A., 2003. A phase II trial with rosiglitazone in liposarcoma patients. Br J Cancer 89, 1409–1412. Dei Tos, A.P., Piccinin, S., Doglioni, C., Vukosavljevic, T., Mentzel, T., Boiocchi, M., Fletcher, C.D., 1997. Molecular aberrations of the G1-S checkpoint in myxoid and round cell liposarcoma. Am. J. Pathol. 151, 1531–1539. Demetri, G.D., Fletcher, C.D., Mueller, E., Sarraf, P., Naujoks, R., Campbell, N., Spiegelman, B.M., Singer, S., 1999. Induction of solid tumor differentiation by the peroxisome proliferator-activated receptor-gamma ligand troglitazone in patients with liposarcoma. Proc. Natl. Acad. Sci. U.S.A. 96, 3951–3956. Demicco, E.G., Torres, K.E., Ghadimi, M.P., Colombo, C., Bolshakov, S., Hoffman, A., Peng, T., Bovée, J.V.M.G., Wang, W.-L., Lev, D., Lazar, A.J., 2012. Involvement of the PI3K/Akt pathway in myxoid/round cell liposarcoma. Mod. Pathol. 25, 212–221. Denissov, S., Hofemeister, H., Marks, H., Kranz, A., Ciotta, G., Singh, S., Anastassiadis, K., Stunnenberg, H.G., Stewart, A.F., 2014. Mll2 is required for H3K4 trimethylation on bivalent promoters in embryonic stem cells, whereas Mll1 is redundant. Development 141, 526–537. Derti, A., Garrett-Engele, P., Macisaac, K.D., Stevens, R.C., Sriram, S., Chen, R., Rohl, C.A., Johnson, J.M., Babak, T., 2012. A quantitative atlas of polyadenylation in five mammals. Genome Res. 22, 1173–1183. Dormann, D., Haass, C., 2013. Fused in sarcoma (FUS): an oncogene goes awry in neurodegeneration. Mol. Cell. Neurosci. 56, 475–486. Dou, Y., Milne, T.A., Ruthenburg, A.J., Lee, S., Lee, J.W., Verdine, G.L., Allis, C.D., Roeder, R.G., 2006. Regulation of MLL1 H3K4 methyltransferase activity by its core components. Nat. Struct. Mol. Biol. 13, 713–719. Dovey, O.M., Foster, C.T., Conte, N., Edwards, S.A., Edwards, J.M., Singh, R., Vassiliou, G., Bradley, A., Cowley, S.M., 2013. Histone deacetylase 1 and 2 are essential for normal T-cell development and genomic stability in mice. Blood 121, 1335–1344. Downs-Kelly, E., Goldblum, J.R., Patel, R.M., Weiss, S.W., Folpe, A.L., Mertens, F., Hartke, M., Tubbs, R.R., Skacel, M., 2008. The utility of fluorescence in situ hybridization (FISH) in the diagnosis of myxoid soft tissue neoplasms. Am. J. Surg. Pathol. 32, 8–13. Dölken, L., Ruzsics, Z., Rädle, B., Friedel, C.C., Zimmer, R., Mages, J., Hoffmann, R., Dickinson, P., Forster, T., Ghazal, P., Koszinowski, U.H., 2008. High-resolution gene expression profiling for simultaneous kinetic parameter analysis of RNA synthesis and decay. RNA 14, 1959–1972. Dürr, H.R., Rauh, J., Baur-Melnyk, A., Knösel, T., Lindner, L., Roeder, F., Jansson, V., Klein, A., 2018. Myxoid liposarcoma: local relapse and metastatic pattern in 43 patients. BMC Cancer 18, 304. Eckmann, C.R., Rammelt, C., Wahle, E., 2011. Control of poly(A) tail length. Wiley Interdiscip Rev RNA 2, 348–361. Eisenberg, E., Levanon, E.Y., 2013. Human housekeeping genes, revisited. Trends Genet. 29, 569–574. Elkon, R., Drost, J., van Haaften, G., Jenal, M., Schrier, M., Oude Vrielink, J.A.F., Agami, R., 2012. E2F mediates enhanced alternative polyadenylation in proliferation. Genome Biol 13, R59. Elkon, R., Ugalde, A.P., Agami, R., 2013. Alternative cleavage and polyadenylation: extent, regulation and function. Nat. Rev. Genet. 14, 496–506. 179  Enderle, D., Beisel, C., Stadler, M.B., Gerstung, M., Athri, P., Paro, R., 2011. Polycomb preferentially targets stalled promoters of coding and noncoding transcripts. Genome Res. 21, 216–226. Endo, M., de Graaff, M.A., Ingram, D.R., Lim, S., Lev, D.C., Briaire-de Bruijn, I.H., Somaiah, N., Bovée, J.V.M.G., Lazar, A.J., Nielsen, T.O., 2015. NY-ESO-1 (CTAG1B) expression in mesenchymal tumors. Mod. Pathol. 28, 587–595. Endo, M., Nielsen, T.O., 2012. Pazopanib for metastatic soft-tissue sarcoma. Lancet 380, 801–author reply 801. Erkelenz, S., Mueller, W.F., Evans, M.S., Busch, A., Schöneweis, K., Hertel, K.J., Schaal, H., 2013. Position-dependent splicing activation and repression by SR and hnRNP proteins rely on common mechanisms. RNA 19, 96–102. Erkizan, H.V., Kong, Y., Merchant, M., Schlottmann, S., Barber-Rotenberg, J.S., Yuan, L., Abaan, O.D., Chou, T.-H., Dakshanamurthy, S., Brown, M.L., Uren, A., Toretsky, J.A., 2009. A small molecule blocking oncogenic protein EWS-FLI1 interaction with RNA helicase A inhibits growth of Ewing's sarcoma. Nat. Med. 15, 750–756. Erkizan, H.V., Schneider, J.A., Sajwan, K., Graham, G.T., Griffin, B., Chasovskikh, S., Youbi, S.E., Kallarakal, A., Chruszcz, M., Padmanabhan, R., Casey, J.L., Uren, A., Toretsky, J.A., 2015. RNA helicase A activity is inhibited by oncogenic transcription factor EWS-FLI1. Nucleic Acids Res. Ernst, J., Kheradpour, P., Mikkelsen, T.S., Shoresh, N., Ward, L.D., Epstein, C.B., Zhang, X., Wang, L., Issner, R., Coyne, M., Ku, M., Durham, T., Kellis, M., Bernstein, B.E., 2011. Mapping and analysis of chromatin state dynamics in nine human cell types. Nature 473, 43–49. Ernst, P., Vakoc, C.R., 2012. WRAD: enabler of the SET1-family of H3K4 methyltransferases. Brief Funct Genomics 11, 217–226. Fabre, O., Salehzada, T., Lambert, K., Boo Seok, Y., Zhou, A., Mercier, J., Bisbal, C., 2012. RNase L controls terminal adipocyte differentiation, lipids storage and insulin sensitivity via CHOP10 mRNA regulation. Cell Death Differ. 19, 1470–1481. Ferreira, M.S.V., Crysandt, M., Braunschweig, T., Jost, E., Voss, B., Bouillon, A.-S., Knuechel, R., Brümmendorf, T.H., Beier, F., 2018. Presence of TERT Promoter Mutations is a Secondary Event and Associates with Elongated Telomere Length in Myxoid Liposarcomas. Int J Mol Sci 19, 608. Fica, S.M., Tuttle, N., Novak, T., Li, N.-S., Lu, J., Koodathingal, P., Dai, Q., Staley, J.P., Piccirilli, J.A., 2013. RNA catalyses nuclear pre-mRNA splicing. Nature 503, 229–234. Filippakopoulos, P., Qi, J., Picaud, S., Shen, Y., Smith, W.B., Fedorov, O., Morse, E.M., Keates, T., Hickman, T.T., Felletar, I., Philpott, M., Munro, S., McKeown, M.R., Wang, Y., Christie, A.L., West, N., Cameron, M.J., Schwartz, B., Heightman, T.D., La Thangue, N., French, C.A., Wiest, O., Kung, A.L., Knapp, S., Bradner, J.E., 2010. Selective inhibition of BET bromodomains. Nature 468, 1067–1073. Flavahan, W.A., Gaskell, E., Bernstein, B.E., 2017. Epigenetic plasticity and the hallmarks of cancer. Science 357, eaal2380. Flavell, S.W., Kim, T.-K., Gray, J.M., Harmin, D.A., Hemberg, M., Hong, E.J., Markenscoff-Papadimitriou, E., Bear, D.M., Greenberg, M.E., 2008. Genome-wide analysis of MEF2 transcriptional program reveals synaptic target genes and neuronal activity-dependent polyadenylation site selection. Neuron 60, 1022–1038. Fletcher, C.D.M., Bridge, J.A., Hogendoorn, P.C.W., Mertens, F., 2013. WHO Classification of Tumours of Soft Tissue and Bone. World Health Organization. 180  Forni, C., Minuzzo, M., Virdis, E., Tamborini, E., Simone, M., Tavecchio, M., Erba, E., Grosso, F., Gronchi, A., Aman, P., Casali, P., D'Incalci, M., Pilotti, S., Mantovani, R., 2009. Trabectedin (ET-743) promotes differentiation in myxoid liposarcoma tumors. Molecular Cancer Therapeutics 8, 449–457. Fruman, D.A., Rommel, C., 2014. PI3K and cancer: lessons, challenges and opportunities. Nat Rev Drug Discov 13, 140–156. Fukuda, T., Tsuneyoshi, M., 2000. Adhesion proteins, cellular morphology and fibrous components around the cell/extracellular-matrix interface in myxoid liposarcomas. J. Cancer Res. Clin. Oncol. 126, 320–324. Gabay, M., Li, Y., Felsher, D.W., 2014. MYC activation is a hallmark of cancer initiation and maintenance. Cold Spring Harb Perspect Med 4, a014241–a014241. Galante, P.A.F., Sakabe, N.J., Kirschbaum-Slager, N., de Souza, S.J., 2004. Detection and evaluation of intron retention events in the human transcriptome. RNA 10, 757–765. Gao, X., Gao, C., Liu, G., Hu, J., 2016. MAP4K4: an emerging therapeutic target in cancer. Cell Biosci 6, 56. Gao, Z., Zhang, J., Bonasio, R., Strino, F., Sawai, A., Parisi, F., Kluger, Y., Reinberg, D., 2012. PCGF homologs, CBX proteins, and RYBP define functionally distinct PRC1 family complexes. Molecular Cell 45, 344–356. Garnett, M.J., Edelman, E.J., Heidorn, S.J., Greenman, C.D., Dastur, A., Lau, K.W., Greninger, P., Thompson, I.R., Luo, X., Soares, J., Liu, Q., Iorio, F., Surdez, D., Chen, L., Milano, R.J., Bignell, G.R., Tam, A.T., Davies, H., Stevenson, J.A., Barthorpe, S., Lutz, S.R., Kogera, F., Lawrence, K., McLaren-Douglas, A., Mitropoulos, X., Mironenko, T., Thi, H., Richardson, L., Zhou, W., Jewitt, F., Zhang, T., O'Brien, P., Boisvert, J.L., Price, S., Hur, W., Yang, W., Deng, X., Butler, A., Choi, H.G., Chang, J.W., Baselga, J., Stamenkovic, I., Engelman, J.A., Sharma, S.V., Delattre, O., Saez-Rodriguez, J., Gray, N.S., Settleman, J., Futreal, P.A., Haber, D.A., Stratton, M.R., Ramaswamy, S., McDermott, U., Benes, C.H., 2012. Systematic identification of genomic markers of drug sensitivity in cancer cells. Nature 483, 570–575. Georgakilas, A.G., Martin, O.A., Bonner, W.M., 2017. p21: A Two-Faced Genome Guardian. Trends Mol Med 23, 310–319. Giacomini, A., Ghedini, G.C., Presta, M., Ronca, R., 2018. Long pentraxin 3: A novel multifaceted player in cancer. Biochim. Biophys. Acta 1869, 53–63. Gingras, A.-C., Gstaiger, M., Raught, B., Aebersold, R., 2007. Analysis of protein complexes using mass spectrometry. Nat Rev Mol Cell Biol 8, 645–654. Gonatopoulos-Pournatzis, T., Cowling, V.H., 2014. Cap-binding complex (CBC). Biochem. J. 457, 231–242. Gonzàlez-Porta, M., Frankish, A., Rung, J., Harrow, J., Brazma, A., 2013. Transcriptome analysis of human tissues and cell lines reveals one dominant transcript per gene. Genome Biol 14, R70. Gorthi, A., Romero, J.C., Loranc, E., Cao, L., Lawrence, L.A., Goodale, E., Iniguez, A.B., Bernard, X., Masamsetti, V.P., Roston, S., Lawlor, E.R., Toretsky, J.A., Stegmaier, K., Lessnick, S.L., Chen, Y., Bishop, A.J.R., 2018. EWS-FLI1 increases transcription to cause R-loops and block BRCA1 repair in Ewing sarcoma. Nature 12, 1–391. Göransson, M., Andersson, M.K., Forni, C., Ståhlberg, A., Andersson, C., Olofsson, A., Mantovani, R., Aman, P., 2009. The myxoid liposarcoma FUS-DDIT3 fusion oncoprotein deregulates NF-kappaB target genes by interaction with NFKBIZ. Oncogene 28, 270–278. 181  Göransson, M., Elias, E., Ståhlberg, A., Olofsson, A., Andersson, C., Aman, P., 2005. Myxoid liposarcoma FUS-DDIT3 fusion oncogene induces C/EBP beta-mediated interleukin 6 expression. Int J Cancer 115, 556–560. Göransson, M., Wedin, M., Aman, P., 2002. Temperature-dependent localization of TLS-CHOP to splicing factor compartments. Exp. Cell Res. 278, 125–132. Grohar, P.J., Griffin, L.B., Yeung, C., Chen, Q.-R., Pommier, Y., Khanna, C., Khan, J., Helman, L.J., 2011. Ecteinascidin 743 interferes with the activity of EWS-FLI1 in Ewing sarcoma cells. Neoplasia 13, 145–153. Guadagnolo, B.A., Zagars, G.K., Ballo, M.T., Patel, S.R., Lewis, V.O., Benjamin, R.S., Pollock, R.E., 2008. Excellent local control rates and distinctive patterns of failure in myxoid liposarcoma treated with conservation surgery and radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 70, 760–765. Guntur, K.V.P., Guilherme, A., Xue, L., Chawla, A., Czech, M.P., 2010. Map4k4 negatively regulates peroxisome proliferator-activated receptor (PPAR) gamma protein translation by suppressing the mammalian target of rapamycin (mTOR) signaling pathway in cultured adipocytes. J. Biol. Chem. 285, 6595–6603. Hakimi, M.-A., Bochar, D.A., Chenoweth, J., Lane, W.S., Mandel, G., Shiekhattar, R., 2002. A core-BRAF35 complex containing histone deacetylase mediates repression of neuronal-specific genes. Proc. Natl. Acad. Sci. U.S.A. 99, 7420–7425. Hallier, M., Lerga, A., Barnache, S., Tavitian, A., Moreau-Gachelin, F., 1998. The transcription factor Spi-1/PU.1 interacts with the potential splicing factor TLS. J. Biol. Chem. 273, 4838–4842. Han, J., Murthy, R., Wood, B., Song, B., Wang, S., Sun, B., Malhi, H., Kaufman, R.J., 2013. ER stress signalling through eIF2α and CHOP, but not IRE1α, attenuates adipogenesis in mice. Diabetologia 56, 911–924. Han, S.X., Zhu, Q., Ma, J.-L., Zhao, J., Huang, C., Jia, X., Zhang, D., 2010. Lowered HGK expression inhibits cell invasion and adhesion in hepatocellular carcinoma cell line HepG2. World J. Gastroenterol. 16, 4541–4548. Hanahan, D., Weinberg, R.A., 2000. The hallmarks of cancer 100, 57–70. Hanahan, D., Weinberg, R.A., 2011. Hallmarks of cancer: the next generation. 144, 646–674. Haniball, J., Sumathi, V.P., Kindblom, L.-G., Abudu, A., Carter, S.R., Tillman, R.M., Jeys, L., Spooner, D., Peake, D., Grimer, R.J., 2011. Prognostic factors and metastatic patterns in primary myxoid/round-cell liposarcoma. Sarcoma 2011, 538085–10. Hanna, J.H., Saha, K., Jaenisch, R., 2010. Pluripotency and cellular reprogramming: facts, hypotheses, unresolved issues. Cell 143, 508–525. Harlow, M.L., Maloney, N., Roland, J., Guillén-Navarro, M.J., Easton, M.K., Kitchen-Goosen, S.M., Boguslawski, E.A., Madaj, Z.B., Johnson, B.K., Bowman, M.J., D'Incalci, M., Winn, M.E., Turner, L., Hostetter, G., Galmarini, C.M., Aviles, P.M., Grohar, P.J., 2016. Lurbinectedin inactivates the Ewing sarcoma oncoprotein EWS-FLI1 by redistributing it within the nucleus. Cancer Res. 76, 6657–6668. Hawkins, R.D., Hon, G.C., Yang, C., Antosiewicz-Bourget, J.E., Lee, L.K., Ngo, Q.-M., Klugman, S., Ching, K.A., Edsall, L.E., Ye, Z., Kuan, S., Yu, P., Liu, H., Zhang, X., Green, R.D., Lobanenkov, V.V., Stewart, R., Thomson, J.A., Ren, B., 2011. Dynamic chromatin states in human ES cells reveal potential regulatory sequences and genes involved in pluripotency. Cell Res. 21, 1393–1409. Heintzman, N.D., Stuart, R.K., Hon, G., Fu, Y., Ching, C.W., Hawkins, R.D., Barrera, L.O., Van Calcar, S., Qu, C., Ching, K.A., Wang, W., Weng, Z., Green, R.D., Crawford, G.E., Ren, B., 182  2007. Distinct and predictive chromatin signatures of transcriptional promoters and enhancers in the human genome. Nat. Genet. 39, 311–318. Hemminger, J.A., Iwenofu, O.H., 2013. NY-ESO-1 is a sensitive and specific immunohistochemical marker for myxoid and round cell liposarcomas among related mesenchymal myxoid neoplasms. Mod. Pathol. Henry, R.A., Kuo, Y.-M., Andrews, A.J., 2013. Differences in specificity and selectivity between CBP and p300 acetylation of histone H3 and H3/H4. Biochemistry 52, 5746–5759. Hicks, G.G., Singh, N., Nashabi, A., Mai, S., Bozek, G., Klewes, L., Arapovic, D., White, E.K., Koury, M.J., Oltz, E.M., Van Kaer, L., Ruley, H.E., 2000. Fus deficiency in mice results in defective B-lymphocyte development and activation, high levels of chromosomal instability and perinatal death. Nat. Genet. 24, 175–179. Hilton, I.B., D'Ippolito, A.M., Vockley, C.M., Thakore, P.I., Crawford, G.E., Reddy, T.E., Gersbach, C.A., 2015. Epigenome editing by a CRISPR-Cas9-based acetyltransferase activates genes from promoters and enhancers. Nat Biotechnol 33, 510–517. Hoffman, A., Ghadimi, M.P.H., Demicco, E.G., Creighton, C.J., Torres, K., Colombo, C., Peng, T., Lusby, K., Ingram, D., Hornick, J.L., Wang, W.-L., Ravi, V., Lazar, A.J., Lev, D., Pollock, R.E., 2013. Localized and metastatic myxoid/round cell liposarcoma: clinical and molecular observations. Cancer 119, 1868–1877. Hofvander, J., Viklund, B., Isaksson, A., Brosjö, O., Vult von Steyern, F., Rissler, P., Mandahl, N., Mertens, F., 2018. Different patterns of clonal evolution among different sarcoma subtypes followed for up to 25 years. Nat Commun 9, 3662. Hollerer, I., Grund, K., Hentze, M.W., Kulozik, A.E., 2014. mRNA 3'end processing: A tale of the tail reaches the clinic. EMBO Mol Med 6, 16–26. Hong, S.-H., Youbi, S.E., Hong, S.P., Kallakury, B., Monroe, P., Erkizan, H.V., Barber-Rotenberg, J.S., Houghton, P., Uren, A., Toretsky, J.A., 2013. Pharmacokinetic modeling optimizes inhibition of the “undruggable” EWS-FLI1 transcription factor in Ewing Sarcoma. Oncotarget 5, 338–350. Hoque, M., Ji, Z., Zheng, D., Luo, W., Li, W., You, B., Park, J.Y., Yehia, G., Tian, B., 2013. Analysis of alternative cleavage and polyadenylation by 3' region extraction and deep sequencing. Nat Methods 10, 133–139. Hornick, J.L., Dal Cin, P., Fletcher, C.D.M., 2009. Loss of INI1 expression is characteristic of both conventional and proximal-type epithelioid sarcoma. Am. J. Surg. Pathol. 33, 542–550. Hosogane, M., Funayama, R., Nishida, Y., Nagashima, T., Nakayama, K., 2013. Ras-induced changes in H3K27me3 occur after those in transcriptional activity. PLoS Genet. 9, e1003698. Hrdlickova, R., Toloue, M., Tian, B., 2017. RNA-Seq methods for transcriptome analysis. Wiley Interdiscip Rev RNA 8, e1364. Hu, D., Garruss, A.S., Gao, X., Morgan, M.A., Cook, M., Smith, E.R., Shilatifard, A., 2013. The Mll2 branch of the COMPASS family regulates bivalent promoters in mouse embryonic stem cells. Nat. Struct. Mol. Biol. 20, 1093–1097. Hughes, C.S., Foehr, S., Garfield, D.A., Furlong, E.E., Steinmetz, L.M., Krijgsveld, J., 2014. Ultrasensitive proteome analysis using paramagnetic bead technology. Mol. Syst. Biol. 10, 757–757. Hyun, K., Jeon, J., Park, K., Kim, J., 2017. Writing, erasing and reading histone lysine methylations. Exp. Mol. Med. 49, e324–e324. 183  Iko, Y., Kodama, T.S., Kasai, N., Oyama, T., Morita, E.H., Muto, T., Okumura, M., Fujii, R., Takumi, T., Tate, S.-I., Morikawa, K., 2004. Domain architectures and characterization of an RNA-binding protein, TLS. J. Biol. Chem. 279, 44834–44840. Ishigaki, S., Masuda, A., Fujioka, Y., Iguchi, Y., Katsuno, M., Shibata, A., Urano, F., Sobue, G., Ohno, K., 2012. Position-dependent FUS-RNA interactions regulate alternative splicing events and transcriptions. Sci Rep 2, 529. Italiano, A., Di Mauro, I., Rapp, J., Pierron, G., Auger, N., Alberti, L., Chibon, F., Escande, F., Voegeli, A.-C., Ghnassia, J.-P., Keslair, F., Laé, M., Ranchere-Vince, D., Terrier, P., Baffert, S., Coindre, J.-M., Pedeutour, F., 2016. Clinical effect of molecular methods in sarcoma diagnosis (GENSARC): a prospective, multicentre, observational study. Lancet Oncol. 17, 532–538. Iura, K., Kohashi, K., Hotokebuchi, Y., Ishii, T., Maekawa, A., Yamada, Y., Yamamoto, H., Iwamoto, Y., Oda, Y., 2015. Cancer-testis antigens PRAME and NY-ESO-1 correlate with tumour grade and poor prognosis in myxoid liposarcoma. J Pathol Clin Res 1, 144–159. Jackman, J., Alamo, I., Fornace, A.J., 1994. Genotoxic stress confers preferential and coordinate messenger RNA stability on the five gadd genes. Cancer Res. 54, 5656–5662. Jacques, C., Lamoureux, F., Baud'huin, M., Calleja, L.R., Quillard, T., Amiaud, J., Tirode, F., Rédini, F., Bradner, J.E., Heymann, D., Ory, B., 2016. Targeting the epigenetic readers in Ewing Sarcoma inhibits the oncogenic transcription factor EWS/Fli1. Oncotarget. Jaeger, V., Schneider-Stock, R., Gerresheim, F., Epplen, J.T., Serra, M., Lippert, H., Roessner, A., 1999. Myxoid liposarcoma with transition to round-cell lesion-cell cycle regulator genes and telomerase activity characterizing tumor progression: a case report. Hum. Pathol. 30, 1515–1519. Jamshidi, F., Bashashati, A., Shumansky, K., Dickson, B., Gokgoz, N., Wunder, J.S., Andrulis, I.L., Lazar, A.J., Shah, S.P., Huntsman, D.G., Nielsen, T.O., 2016. The genomic landscape of epithelioid sarcoma cell lines and tumours. J. Pathol. 238, 63–73. Jauhiainen, A., Thomsen, C., Strömbom, L., Grundevik, P., Andersson, C., Danielsson, A., Andersson, M.K., Nerman, O., Rörkvist, L., Ståhlberg, A., Aman, P., 2012. Distinct cytoplasmic and nuclear functions of the stress induced protein DDIT3/CHOP/GADD153. PLoS ONE 7, e33208. Jia, R., Chai, P., Zhang, H., Fan, X., 2017. Novel insights into chromosomal conformations in cancer. Mol. Cancer 16, 173. Joseph, C.G., Hwang, H., Jiao, Y., Wood, L.D., Kinde, I., Wu, J., Mandahl, N., Luo, J., Hruban, R.H., Diaz, L.A., He, T.-C., Vogelstein, B., Kinzler, K.W., Mertens, F., Papadopoulos, N., 2014. Exomic analysis of myxoid liposarcomas, synovial sarcomas, and osteosarcomas. Genes Chromosomes Cancer 53, 15–24. Judson, I., van der Graaf, W.T., 2016. Sarcoma: Olaratumab - really a breakthrough for soft-tissue sarcomas? Nat Rev Clin Oncol 13, 534–536. Jurkin, J., Zupkovitz, G., Lagger, S., Grausenburger, R., Hagelkruys, A., Kenner, L., Seiser, C., 2011. Distinct and redundant functions of histone deacetylases HDAC1 and HDAC2 in proliferation and tumorigenesis. Cell Cycle 10, 406–412. Kadoch, C., Crabtree, G.R., 2013. Reversible Disruption of mSWI/SNF (BAF) Complexes by the SS18-SSX Oncogenic Fusion in Synovial Sarcoma 153, 71–85. Kadoch, C., Hargreaves, D.C., Hodges, C., Elias, L., Ho, L., Ranish, J., Crabtree, G.R., 2013. Proteomic and bioinformatic analysis of mammalian SWI/SNF complexes identifies extensive roles in human malignancy. Nat. Genet. 45, 592–601. 184  Kadoch, C., Williams, R.T., Calarco, J.P., Miller, E.L., Weber, C.M., Braun, S.M.G., Pulice, J.L., Chory, E.J., Crabtree, G.R., 2017. Dynamics of BAF-Polycomb complex opposition on heterochromatin in normal and oncogenic states. Nat. Genet. 49, 213–222. Kalli, A., Hess, S., 2012. Effect of mass spectrometric parameters on peptide and protein identification rates for shotgun proteomic experiments on an LTQ-orbitrap mass analyzer. Proteomics 12, 21–31. Kamelgarn, M., Chen, J., Kuang, L., Arenas, A., Zhai, J., Zhu, H., Gal, J., 2016. Proteomic analysis of FUS interacting proteins provides insights into FUS function and its role in ALS. Biochim. Biophys. Acta 1862, 2004–2014. Kandoth, C., McLellan, M.D., Vandin, F., Ye, K., Niu, B., Lu, C., Xie, M., Zhang, Q., McMichael, J.F., Wyczalkowski, M.A., Leiserson, M.D.M., Miller, C.A., Welch, J.S., Walter, M.J., Wendl, M.C., Ley, T.J., Wilson, R.K., Raphael, B.J., Ding, L., 2013. Mutational landscape and significance across 12 major cancer types. Nature 502, 333–339. Kato, M., Han, T.W., Xie, S., Shi, K., Du, X., Wu, L.C., Mirzaei, H., Goldsmith, E.J., Longgood, J., Pei, J., Grishin, N.V., Frantz, D.E., Schneider, J.W., Chen, S., Li, L., Sawaya, M.R., Eisenberg, D., Tycko, R., McKnight, S.L., 2012. Cell-free Formation of RNA Granules: Low Complexity Sequence Domains Form Dynamic Fibers within Hydrogels 149, 753–767. Katz, D., Boonsirikamchai, P., Choi, H., Lazar, A.J., Wang, W.-L., Xiao, L., Park, M.S., Ravi, V., Benjamin, R.S., Araujo, D.M., 2012. Efficacy of first-line doxorubicin and ifosfamide in myxoid liposarcoma. Clin Sarcoma Res 2, 2. Katz, Y., Wang, E.T., Airoldi, E.M., Burge, C.B., 2010. Analysis and design of RNA sequencing experiments for identifying isoform regulation. Nat Methods 7, 1009–1015. Kawai, A., Naito, N., Yoshida, A., Morimoto, Y., Ouchida, M., Shimizu, K., Beppu, Y., 2004. Establishment and characterization of a biphasic synovial sarcoma cell line, SYO-1. Cancer Lett. 204, 105–113. Killela, P.J., Reitman, Z.J., Jiao, Y., Bettegowda, C., Agrawal, N., Diaz, L.A., Friedman, A.H., Friedman, H., Gallia, G.L., Giovanella, B.C., Grollman, A.P., He, T.-C., He, Y., Hruban, R.H., Jallo, G.I., Mandahl, N., Meeker, A.K., Mertens, F., Netto, G.J., Rasheed, B.A., Riggins, G.J., Rosenquist, T.A., Schiffman, M., Shih, I.-M., Theodorescu, D., Torbenson, M.S., Velculescu, V.E., Wang, T.-L., Wentzensen, N., Wood, L.D., Zhang, M., McLendon, R.E., Bigner, D.D., Kinzler, K.W., Vogelstein, B., Papadopoulos, N., Yan, H., 2013. TERT promoter mutations occur frequently in gliomas and a subset of tumors derived from cells with low rates of self-renewal. Proc. Natl. Acad. Sci. U.S.A. 110, 6021–6026. Kim, E., Magen, A., Ast, G., 2007. Different levels of alternative splicing among eukaryotes. Nucleic Acids Res. 35, 125–131. Kim, K.H., Roberts, C.W.M., 2014. Mechanisms by which SMARCB1 loss drives rhabdoid tumor growth. Cancer Genet 207, 365–372. Kim, M.-S., Pinto, S.M., Getnet, D., Nirujogi, R.S., Manda, S.S., Chaerkady, R., Madugundu, A.K., Kelkar, D.S., Isserlin, R., Jain, S., Thomas, J.K., Muthusamy, B., Leal-Rojas, P., Kumar, P., Sahasrabuddhe, N.A., Balakrishnan, L., Advani, J., George, B., Renuse, S., Selvan, L.D.N., Patil, A.H., Nanjappa, V., Radhakrishnan, A., Prasad, S., Subbannayya, T., Raju, R., Kumar, M., Sreenivasamurthy, S.K., Marimuthu, A., Sathe, G.J., Chavan, S., Datta, K.K., Subbannayya, Y., Sahu, A., Yelamanchi, S.D., Jayaram, S., Rajagopalan, P., Sharma, J., Murthy, K.R., Syed, N., Goel, R., Khan, A.A., Ahmad, S., Dey, G., Mudgal, K., Chatterjee, A., Huang, T.-C., Zhong, J., Wu, X., Shaw, P.G., Freed, D., Zahari, M.S., Mukherjee, K.K., Shankar, S., Mahadevan, A., Lam, H., Mitchell, C.J., Shankar, S.K., Satishchandra, P., Schroeder, J.T., Sirdeshmukh, R., Maitra, A., Leach, S.D., Drake, C.G., 185  Halushka, M.K., Prasad, T.S.K., Hruban, R.H., Kerr, C.L., Bader, G.D., Iacobuzio-Donahue, C.A., Gowda, H., Pandey, A., 2014. A draft map of the human proteome. 509, 575–581. Kim, T., Buratowski, S., 2009. Dimethylation of H3K4 by Set1 recruits the Set3 histone deacetylase complex to 5' transcribed regions. Cell 137, 259–272. Kloet, S.L., Irem Baymaz, H., Makowski, M., Groenewold, V., Jansen, P., Berendsen, M., Niazi, H., Kops, G.J., Vermeulen, M., 2014. Towards elucidating the stability, dynamics and architecture of the NuRD complex using quantitative interaction proteomics. FEBS J. n/a–n/a. Knott, G.J., Bond, C.S., Fox, A.H., 2016. The DBHS proteins SFPQ, NONO and PSPC1: a multipurpose molecular scaffold. Nucleic Acids Res. 44, 3989–4004. Koelsche, C., Renner, M., Hartmann, W., Brandt, R., Lehner, B., Waldburger, N., Alldinger, I., Schmitt, T., Egerer, G., Penzel, R., Wardelmann, E., Schirmacher, P., Deimling, von, A., Mechtersheimer, G., 2014. TERT promoter hotspot mutations are recurrent in myxoid liposarcomas but rare in other soft tissue sarcoma entities. J. Exp. Clin. Cancer Res. 33, 33. Kooistra, S.M., Helin, K., 2012. Molecular mechanisms and potential functions of histone demethylases. Nat Rev Mol Cell Biol 13, 297–311. Kornberg, R.D., 1974. Chromatin structure: a repeating unit of histones and DNA. Science 184, 868–871. Kornblihtt, A.R., Schor, I.E., Alló, M., Dujardin, G., Petrillo, E., Muñoz, M.J., 2013. Alternative splicing: a pivotal step between eukaryotic transcription and translation. Nat Rev Mol Cell Biol 14, 153–165. Kuroda, M., Ishida, T., Takanashi, M., Satoh, M., Machinami, R., Watanabe, T., 1997. Oncogenic transformation and inhibition of adipocytic conversion of preadipocytes by TLS/FUS-CHOP type II chimeric protein. Am. J. Pathol. 151, 735–744. Kuroda, M., Sok, J., Webb, L., Baechtold, H., Urano, F., Yin, Y., Chung, P., de Rooij, D.G., Akhmedov, A., Ashley, T., Ron, D., 2000. Male sterility and enhanced radiation sensitivity in TLS(-/-) mice. EMBO J 19, 453–462. Kühn, U., Gündel, M., Knoth, A., Kerwitz, Y., Rüdel, S., Wahle, E., 2009. Poly(A) tail length is controlled by the nuclear poly(A)-binding protein regulating the interaction between poly(A) polymerase and the cleavage and polyadenylation specificity factor. J. Biol. Chem. 284, 22803–22814. Künstlinger, H., Fassunke, J., Schildhaus, H.-U., Brors, B., Heydt, C., Ihle, M.A., Mechtersheimer, G., Wardelmann, E., Büttner, R., Merkelbach-Bruse, S., 2015. FGFR2 is overexpressed in myxoid liposarcoma and inhibition of FGFR signaling impairs tumor growth in vitro. Oncotarget. Kwon, H., Imbalzano, A.N., Khavari, P.A., Kingston, R.E., Green, M.R., 1994. Nucleosome disruption and enhancement of activator binding by a human SW1/SNF complex. Nature 370, 477–481. Kwon, I., Kato, M., Xiang, S., Wu, L., Theodoropoulos, P., Mirzaei, H., Han, T., Xie, S., Corden, J.L., McKnight, S.L., 2013. Phosphorylation-Regulated Binding of RNA Polymerase II to Fibrous Polymers of Low-Complexity Domains. 155, 1049–1060. Lagier-Tourenne, C., Polymenidou, M., Hutt, K.R., Vu, A.Q., Baughn, M., Huelga, S.C., Clutario, K.M., Ling, S.-C., Liang, T.Y., Mazur, C., Wancewicz, E., Kim, A.S., Watt, A., Freier, S., Hicks, G.G., Donohue, J.P., Shiue, L., Bennett, C.F., Ravits, J., Cleveland, D.W., Yeo, G.W., 2012. Divergent roles of ALS-linked proteins FUS/TLS and TDP-43 intersect in processing long pre-mRNAs. Nat. Neurosci. 15, 1488–1497. 186  Laherty, C.D., Yang, W.M., Sun, J.M., Davie, J.R., Seto, E., Eisenman, R.N., 1997. Histone deacetylases associated with the mSin3 corepressor mediate mad transcriptional repression. Cell 89, 349–356. Lai, A.C., Toure, M., Hellerschmied, D., Salami, J., Jaime-Figueroa, S., Ko, E., Hines, J., Crews, C.M., 2016. Modular PROTAC Design for the Degradation of Oncogenic BCR-ABL. Angew. Chem. Int. Ed. Engl. 55, 807–810. Lambert, M., Jambon, S., Depauw, S., David-Cordonnier, M.-H., 2018. Targeting Transcription Factors for Cancer Treatment. Molecules 23, 1479. Lamhamedi-Cherradi, S.-E., Menegaz, B.A., Ramamoorthy, V., Aiyer, R.A., Maywald, R.L., Buford, A.S., Doolittle, D.K., Culotta, K.S., O'Dorisio, J.E., Ludwig, J.A., 2015. An Oral Formulation of YK-4-279: Preclinical Efficacy and Acquired Resistance Patterns in Ewing Sarcoma. Molecular Cancer Therapeutics 14, 1591–1604. Lange, M., Kaynak, B., Forster, U.B., Tönjes, M., Fischer, J.J., Grimm, C., Schlesinger, J., Just, S., Dunkel, I., Krueger, T., Mebus, S., Lehrach, H., Lurz, R., Gobom, J., Rottbauer, W., Abdelilah-Seyfried, S., Sperling, S., 2008. Regulation of muscle development by DPF3, a novel histone acetylation and methylation reader of the BAF chromatin remodeling complex. Genes Dev. 22, 2370–2384. Laporte, A.N., Barrott, J.J., Yao, R.J., Poulin, N.M., Brodin, B.A., Jones, K.B., Underhill, T.M., Nielsen, T.O., 2017. HDAC and Proteasome Inhibitors Synergize to Activate Pro-Apoptotic Factors in Synovial Sarcoma. PLoS ONE 12, e0169407. Latysheva, N.S., Oates, M.E., Maddox, L., Flock, T., Gough, J., Buljan, M., Weatheritt, R.J., Babu, M.M., 2016. Molecular Principles of Gene Fusion Mediated Rewiring of Protein Interaction Networks in Cancer. Molecular Cell 63, 579–592. Lavallée-Adam, M., Rauniyar, N., McClatchy, D.B., Yates, J.R., 2014. PSEA-Quant: a protein set enrichment analysis on label-free and label-based protein quantification data. J Proteome Res 13, 5496–5509. Lee, R.S., Stewart, C., Carter, S.L., Ambrogio, L., Cibulskis, K., Sougnez, C., Lawrence, M.S., Auclair, D., Mora, J., Golub, T.R., Biegel, J.A., Getz, G., Roberts, C.W.M., 2012. A remarkably simple genome underlies highly malignant pediatric rhabdoid cancers. J Clin Invest 122, 2983–2988. Lee, T.I., Jenner, R.G., Boyer, L.A., Guenther, M.G., Levine, S.S., Kumar, R.M., Chevalier, B., Johnstone, S.E., Cole, M.F., Isono, K.-I., Koseki, H., Fuchikami, T., Abe, K., Murray, H.L., Zucker, J.P., Yuan, B., Bell, G.W., Herbolsheimer, E., Hannett, N.M., Sun, K., Odom, D.T., Otte, A.P., Volkert, T.L., Bartel, D.P., Melton, D.A., Gifford, D.K., Jaenisch, R., Young, R.A., 2006. Control of developmental regulators by Polycomb in human embryonic stem cells. Cell 125, 301–313. Lee, W., Teckie, S., Wiesner, T., Ran, L., Prieto Granada, C.N., Lin, M., Zhu, S., Cao, Z., Liang, Y., Sboner, A., Tap, W.D., Fletcher, J.A., Huberman, K.H., Qin, L.-X., Viale, A., Singer, S., Zheng, D., Berger, M.F., Chen, Y., Antonescu, C.R., Chi, P., 2014. PRC2 is recurrently inactivated through EED or SUZ12 loss in malignant peripheral nerve sheath tumors. Nat. Genet. 46, 1227–1232. Lee, Y., Rio, D.C., 2015. Mechanisms and Regulation of Alternative Pre-mRNA Splicing. Annu. Rev. Biochem. 84, 291–323. Lehner, B., Fraser, A.G., 2004. Protein domains enriched in mammalian tissue-specific or widely expressed genes. Trends Genet. 20, 468–472. 187  Lemeur, M., Mattei, J.-C., Souteyrand, P., Chagnaud, C., Curvale, G., Rochwerger, A., 2015. Prognostic factors for the recurrence of myxoid liposarcoma: 20 cases with up to 8 years follow-up. Orthop Traumatol Surg Res 101, 103–107. Lemos, A., Leão, M., Soares, J., Palmeira, A., Pinto, M., Saraiva, L., Sousa, M.E., 2016. Medicinal Chemistry Strategies to Disrupt the p53-MDM2/MDMX Interaction. Med Res Rev 36, 789–844. Lewin, J., Garg, S., Lau, B.Y., Dickson, B.C., Traub, F., Gokgoz, N., Griffin, A.M., Ferguson, P.C., Andrulis, I.L., Sim, H.-W., Kamel-Reid, S., Stockley, T.L., Siu, L.L., Wunder, J.S., Razak, A.R.A., 2018. Identifying actionable variants using next generation sequencing in patients with a historical diagnosis of undifferentiated pleomorphic sarcoma. Int J Cancer 142, 57–65. Li, J., Ye, L., Owen, S., Weeks, H.P., Zhang, Z., Jiang, W.G., 2015. Emerging role of CCN family proteins in tumorigenesis and cancer metastasis (Review). Int. J. Mol. Med. 36, 1451–1463. Li, P., Li, J., Wang, L., Di, L.-J., 2017. Proximity Labeling of Interacting Proteins: Application of BioID as a Discovery Tool. Proteomics 17, 1700002. Li, W., Zhang, Z., Liu, X., Cheng, X., Zhang, Y., Han, X., Zhang, Y., Liu, S., Yang, J., Xu, B., He, L., Sun, L., Liang, J., Shang, Y., 2017. The FOXN3-NEAT1-SIN3A repressor complex promotes progression of hormonally responsive breast cancer. J Clin Invest 127, 3421–3440. Li, X., Decker, M., Westendorf, J.J., 2010. TEThered to Runx: novel binding partners for runx factors. Blood Cells Mol. Dis. 45, 82–85. Lianoglou, S., Garg, V., Yang, J.L., Leslie, C.S., Mayr, C., 2013. Ubiquitously transcribed genes use alternative polyadenylation to achieve tissue-specific expression. Genes Dev. 27, 2380–2396. Lim, J., Poulin, N.M., Nielsen, T.O., 2015. New Strategies in Sarcoma: Linking Genomic and Immunotherapy Approaches to Molecular Subtype. Clin. Cancer Res. 21, 4753–4759. Limon, J., Turc-Carel, C., Dal Cin, P., Rao, U., Sandberg, A.A., 1986. Recurrent chromosome translocations in liposarcoma. Cancer Genet Cytogenet. 22, 93-94. Lin, J.-C., 2015. Impacts of Alternative Splicing Events on the Differentiation of Adipocytes. Int J Mol Sci 16, 22169–22189. Lindemann, C., Thomanek, N., Hundt, F., Lerari, T., Meyer, H.E., Wolters, D., Marcus, K., 2017. Strategies in relative and absolute quantitative mass spectrometry based proteomics. Biol. Chem. 398, 687–699. Liu, Wenzhong, Xie, Y., Ma, J., Luo, X., Nie, P., Zuo, Z., Lahrmann, U., Zhao, Q., Zheng, Y., Zhao, Y., Xue, Y., Ren, J., 2015. IBS: an illustrator for the presentation and visualization of biological sequences. Bioinformatics 31, 3359–3361. Liu, Xin, Wang, L., Zhao, K., Thompson, P.R., Hwang, Y., Marmorstein, R., Cole, P.A., 2008. The structural basis of protein acetylation by the p300/CBP transcriptional coactivator. Nature 451, 846–850. Liu, Xuehui, Niu, C., Ren, J., Zhang, J., Xie, X., Zhu, H., Feng, W., Gong, W., 2013. The RRM domain of human fused in sarcoma protein reveals a non-canonical nucleic acid binding site. Biochim. Biophys. Acta 1832, 375–385. Liu, Yichuan, Ferguson, J.F., Xue, C., Silverman, I.M., Gregory, B., Reilly, M.P., Li, M., 2013. Evaluating the impact of sequencing depth on transcriptome profiling in human adipose. PLoS ONE 8, e66883. 188  Liu, Yuan Fei, Qu, G.-Q., Lu, Y.-M., Kong, W.-M., Liu, Y., Chen, W.-X., Liao, X.-H., 2016. Silencing of MAP4K4 by short hairpin RNA suppresses proliferation, induces G1 cell cycle arrest and induces apoptosis in gastric cancer cells. Mol Med Rep 13, 41–48. Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402–408. Lo-Coco, F., Avvisati, G., Vignetti, M., Thiede, C., Orlando, S.M., Iacobelli, S., Ferrara, F., Fazi, P., Cicconi, L., Di Bona, E., Specchia, G., Sica, S., Divona, M., Levis, A., Fiedler, W., Cerqui, E., Breccia, M., Fioritoni, G., Salih, H.R., Cazzola, M., Melillo, L., Carella, A.M., Brandts, C.H., Morra, E., Lilienfeld-Toal, von, M., Hertenstein, B., Wattad, M., Lübbert, M., Hänel, M., Schmitz, N., Link, H., Kropp, M.G., Rambaldi, A., La Nasa, G., Luppi, M., Ciceri, F., Finizio, O., Venditti, A., Fabbiano, F., Döhner, K., Sauer, M., Ganser, A., Amadori, S., Mandelli, F., Döhner, H., Ehninger, G., Schlenk, R.F., Platzbecker, U., Gruppo Italiano Malattie Ematologiche dell'Adulto, German-Austrian Acute Myeloid Leukemia Study Group, Study Alliance Leukemia, 2013. Retinoic acid and arsenic trioxide for acute promyelocytic leukemia. N. Engl. J. Med. 369, 111–121. Lockhart, D.J., Dong, H., Byrne, M.C., Follettie, M.T., Gallo, M.V., Chee, M.S., Mittmann, M., Wang, C., Kobayashi, M., Horton, H., Brown, E.L., 1996. Expression monitoring by hybridization to high-density oligonucleotide arrays. Nat Biotechnol 14, 1675–1680. Lu, J., Qian, Y., Altieri, M., Dong, H., Wang, J., Raina, K., Hines, J., Winkler, J.D., Crew, A.P., Coleman, K., Crews, C.M., 2015. Hijacking the E3 Ubiquitin Ligase Cereblon to Efficiently Target BRD4. Chem. Biol. 22, 755–763. Luebben, W.R., Sharma, N., Nyborg, J.K., 2010. Nucleosome eviction and activated transcription require p300 acetylation of histone H3 lysine 14. Proc. Natl. Acad. Sci. U.S.A. 107, 19254–19259. Luger, K., Mäder, A.W., Richmond, R.K., Sargent, D.F., Richmond, T.J., 1997. Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature 389, 251–260. Ma, Y., Brewer, J.W., Diehl, J.A., Hendershot, L.M., 2002. Two distinct stress signaling pathways converge upon the CHOP promoter during the mammalian unfolded protein response. J. Mol. Biol. 318, 1351–1365. Mandahl, N., Mertens, F., Aman, P., Rydholm, A., Brosjo, O., Willén, H., Mitelman, F., 1994. Nonrandom secondary chromosome-aberrations in liposarcomas with t(12, 16). Int. J. Oncol. 4, 307–310. Mandel, C.R., Kaneko, S., Zhang, H., Gebauer, D., Vethantham, V., Manley, J.L., Tong, L., 2006. Polyadenylation factor CPSF-73 is the pre-mRNA 3'-end-processing endonuclease. Nature 444, 953–956. Mani, R.-S., Chinnaiyan, A.M., 2010. Triggers for genomic rearrangements: insights into genomic, cellular and environmental influences. Nat. Rev. Genet. 11, 819–829. Maquat, L.E., Hwang, J., Sato, H., Tang, Y., 2010. CBP80-promoted mRNP rearrangements during the pioneer round of translation, nonsense-mediated mRNA decay, and thereafter. Cold Spring Harb. Symp. Quant. Biol. 75, 127–134. Marmorstein, R., Zhou, M.-M., 2014. Writers and readers of histone acetylation: structure, mechanism, and inhibition. Cold Spring Harb Perspect Biol 6, a018762–a018762. Martens, J.H.A., Brinkman, A.B., Simmer, F., Francoijs, K.-J., Nebbioso, A., Ferrara, F., Altucci, L., Stunnenberg, H.G., 2010. PML-RARalpha/RXR Alters the Epigenetic Landscape in Acute Promyelocytic Leukemia. Cancer Cell 17, 173–185. 189  Masuda, A., Takeda, J.-I., Okuno, T., Okamoto, T., Ohkawara, B., Ito, M., Ishigaki, S., Sobue, G., Ohno, K., 2015. Position-specific binding of FUS to nascent RNA regulates mRNA length. Genes Dev. 29, 1045–1057. Mathur, R., Alver, B.H., San Roman, A.K., Wilson, B.G., Wang, X., Agoston, A.T., Park, P.J., Shivdasani, R.A., Roberts, C.W.M., 2017. ARID1A loss impairs enhancer-mediated gene regulation and drives colon cancer in mice. Nat. Genet. 49, 296–302. Mayr, C., Bartel, D.P., 2009. Widespread shortening of 3'UTRs by alternative cleavage and polyadenylation activates oncogenes in cancer cells. Cell 138, 673–684. McManus, K.J., Hendzel, M.J., 2003. Quantitative analysis of CBP- and P300-induced histone acetylations in vivo using native chromatin. Mol. Cell. Biol. 23, 7611–7627. Meissner, M., Lopato, S., Gotzmann, J., Sauermann, G., Barta, A., 2003. Proto-oncoprotein TLS/FUS is associated to the nuclear matrix and complexed with splicing factors PTB, SRm160, and SR proteins. Exp. Cell Res. 283, 184–195. Mertens, F., Antonescu, C.R., Mitelman, F., 2016. Gene fusions in soft tissue tumors: Recurrent and overlapping pathogenetic themes. Genes Chromosomes Cancer 55, 291–310. Mertens, F., Panagopoulos, I., Mandahl, N., 2010. Genomic characteristics of soft tissue sarcomas. Virchows Arch. path Anat. 456, 129–139. Mi, H., Huang, X., Muruganujan, A., Tang, H., Mills, C., Kang, D., Thomas, P.D., 2017. PANTHER version 11: expanded annotation data from Gene Ontology and Reactome pathways, and data analysis tool enhancements. Nucleic Acids Res. 45, D183–D189. Mikkelsen, T.S., Ku, M., Jaffe, D.B., Issac, B., Lieberman, E., Giannoukos, G., Alvarez, P., Brockman, W., Kim, T.-K., Koche, R.P., Lee, W., Mendenhall, E., O'Donovan, A., Presser, A., Russ, C., Xie, X., Meissner, A., Wernig, M., Jaenisch, R., Nusbaum, C., Lander, E.S., Bernstein, B.E., 2007. Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature 448, 553–560. Minas, T.Z., Surdez, D., Javaheri, T., Tanaka, M., Howarth, M., Kang, H.J., Han, J., Han, Z.-Y., Sax, B., Kream, B.E., Hong, S.-H., Çelik, H., Tirode, F., Tuckermann, J., Toretsky, J.A., Kenner, L., Kovar, H., Lee, S., Sweet-Cordero, E.A., Nakamura, T., Moriggl, R., Delattre, O., Uren, A., 2017. Combined experience of six independent laboratories attempting to create an Ewing sarcoma mouse model. Oncotarget 8, 34141–34163. Mitra, A., Mishra, L., Li, S., 2013. Technologies for deriving primary tumor cells for use in personalized cancer therapy. Trends Biotechnol. 31, 347–354. Modena, P., Lualdi, E., Facchinetti, F., Galli, L., Teixeira, M.R., Pilotti, S., Sozzi, G., 2005. SMARCB1/INI1 tumor suppressor gene is frequently inactivated in epithelioid sarcomas. Cancer Res. 65, 4012–4019. Morra, R., Lee, B.M., Shaw, H., Tuma, R., Mancini, E.J., 2012. Concerted action of the PHD, chromo and motor domains regulates the human chromatin remodelling ATPase CHD4. 586, 2513–2521. Moteki, S., Price, D., 2002. Functional coupling of capping and transcription of mRNA. Molecular Cell 10, 599–609. Murthy, K.G., Park, P., Manley, J.L., 1991. A nuclear micrococcal-sensitive, ATP-dependent exoribonuclease degrades uncapped but not capped RNA substrates. Nucleic Acids Res. 19, 2685–2692. Musselman, C.A., Lalonde, M.-E., Cote, J., Kutateladze, T.G., 2012. Perceiving the epigenetic landscape through histone readers. Nat. Struct. Mol. Biol. 19, 1218–1227. Nakamura, T., Matsumine, A., Kawai, A., Araki, N., Goto, T., Yonemoto, T., Sugiura, H., Nishida, Y., Hiraga, H., Honoki, K., Yasuda, T., Boku, S., Sudo, A., Ueda, T., 2016. The 190  clinical outcome of pazopanib treatment in Japanese patients with relapsed soft tissue sarcoma: A Japanese Musculoskeletal Oncology Group (JMOG) study. Cancer 122, 1408–1416. Narendra, S., Valente, A., Tull, J., Zhang, S., 2011. DDIT3 gene break-apart as a molecular marker for diagnosis of myxoid liposarcoma--assay validation and clinical experience. Diagn. Mol. Pathol. 20, 218–224. Nasr, R., De Thé, H., 2010. Eradication of acute promyelocytic leukemia-initiating cells by PML/RARA-targeting. Int. J. Hematol. 91, 742–747. Ndubaku, C.O., Crawford, T.D., Chen, H., Boggs, J.W., Drobnick, J., Harris, S.F., Jesudason, R., McNamara, E., Nonomiya, J., Sambrone, A., Schmidt, S., Smyczek, T., Vitorino, P., Wang, L., Wu, P., Yeung, S., Chen, J., Chen, K., Ding, C.Z., Wang, T., Xu, Z., Gould, S.E., Murray, L.J., Ye, W., 2015. Structure-Based Design of GNE-495, a Potent and Selective MAP4K4 Inhibitor with Efficacy in Retinal Angiogenesis. ACS Med Chem Lett 6, 913–918. Nebbioso, A., Tambaro, F.P., Dell'Aversana, C., Altucci, L., 2018. Cancer epigenetics: Moving forward. PLoS Genet. 14, e1007362. Negri, T., Virdis, E., Brich, S., Bozzi, F., Tamborini, E., Tarantino, E., Jocollè, G., Cassinelli, G., Grosso, F., Sanfilippo, R., Casalini, P., Greco, A., Pierotti, M.A., Pilotti, S., 2010. Functional mapping of receptor tyrosine kinases in myxoid liposarcoma. 16, 3581–3593. Neklesa, T.K., Winkler, J.D., Crews, C.M., 2017. Targeted protein degradation by PROTACs. Pharmacol. Ther. 174, 138–144. Neve, J., Burger, K., Li, W., Hoque, M., Patel, R., Tian, B., Gullerova, M., Furger, A., 2016. Subcellular RNA profiling links splicing and nuclear DICER1 to alternative cleavage and polyadenylation. Genome Res. 26, 24–35. Neve, J., Patel, R., Wang, Z., Louey, A., Furger, A.M., 2017. Cleavage and polyadenylation: Ending the message expands gene regulation. RNA Biology 14, 865–890. Nielsen, T.O., West, R.B., 2010. Translating gene expression into clinical care: sarcomas as a paradigm. J Clin Oncol 28, 1796–1805. Nilsen, T.W., Graveley, B.R., 2010. Expansion of the eukaryotic proteome by alternative splicing. Nature 463, 457–463. Nishida, Y., Tsukushi, S., Nakashima, H., Ishiguro, N., 2010. Clinicopathologic prognostic factors of pure myxoid liposarcoma of the extremities and trunk wall. Clin. Orthop. Relat. Res. 468, 3041–3046. O'Meara, M.M., Simon, J.A., 2012. Inner workings and regulatory inputs that control Polycomb repressive complex 2. Chromosoma 121, 221–234. Ohjimi, Y., Iwasaki, H., Kaneko, Y., Ishiguro, M., Ohgami, A., Fujita, C., Shinohara, N., Yoshitake, K., Kikuchi, M., 1992. Chromosome abnormalities in liposarcomas. Cancer Genet. Cytogenet. 64, 111–117. Oltean, S., Bates, D.O., 2014. Hallmarks of alternative splicing in cancer. Oncogene 33, 5311–5318. Orkin, S.H., Hochedlinger, K., 2011. Chromatin connections to pluripotency and cellular reprogramming. Cell 145, 835–850. Ozsolak, F., Platt, A.R., Jones, D.R., Reifenberger, J.G., Sass, L.E., McInerney, P., Thompson, J.F., Bowers, J., Jarosz, M., Milos, P.M., 2009. Direct RNA sequencing. Nature 461, 814–818. Pan, Q., Bakowski, M.A., Morris, Q., Zhang, W., Frey, B.J., Hughes, T.R., Blencowe, B.J., 2005. Alternative splicing of conserved exons is frequently species-specific in human and mouse. Trends Genet. 21, 73–77. 191  Pan, Q., Shai, O., Lee, L.J., Frey, B.J., Blencowe, B.J., 2008. Deep surveying of alternative splicing complexity in the human transcriptome by high-throughput sequencing. Nat. Genet. 40, 1413–1415. Panagopoulos, I., Höglund, M., Mertens, F., Mandahl, N., Mitelman, F., Aman, P., 1996. Fusion of the EWS and CHOP genes in myxoid liposarcoma. Oncogene 12, 489–494. Papp, B., Müller, J., 2006. Histone trimethylation and the maintenance of transcriptional ON and OFF states by trxG and PcG proteins. Genes Dev. 20, 2041–2054. Park, B.O., Ahrends, R., Teruel, M.N., 2012. Consecutive positive feedback loops create a bistable switch that controls preadipocyte-to-adipocyte conversion. Cell Rep 2, 976–990. Patel, A., Dharmarajan, V., Vought, V.E., Cosgrove, M.S., 2009. On the mechanism of multiple lysine methylation by the human mixed lineage leukemia protein-1 (MLL1) core complex. J. Biol. Chem. 284, 24242–24256. Pérez-Losada, J., Pintado, B., Gutiérrez-Adán, A., Flores, T., Bañares-González, B., del Campo, J.C., Martín-Martín, J.F., Battaner, E., Sánchez-García, I., 2000. The chimeric FUS/TLS-CHOP fusion protein specifically induces liposarcomas in transgenic mice. Oncogene 19, 2413–2422. Pérez-Mancera, P.A., Bermejo-Rodríguez, C., Sánchez-Martín, M., Abollo-Jiménez, F., Pintado, B., Sánchez-García, I., 2008. FUS-DDIT3 prevents the development of adipocytic precursors in liposarcoma by repressing PPARgamma and C/EBPalpha and activating eIF4E. PLoS ONE 3, e2569. Pérez-Mancera, P.A., Vicente-Dueñas, C., González-Herrero, I., Sánchez-Martín, M., Flores-Corral, T., Sánchez-García, I., 2007. Fat-specific FUS-DDIT3-transgenic mice establish PPARgamma inactivation is required to liposarcoma development. Carcinogenesis 28, 2069–2073. Pilotti, S., Lavarino, C., Mezzelani, A., Torre, Della, C., Minoletti, F., Sozzi, G., Azzarelli, A., Rilke, F., Pierotti, M.A., 1998. Limited role of TP53 and TP53-related genes in myxoid liposarcoma. Tumori 84, 571–577. Pitson, G., Robinson, P., Wilke, D., Kandel, R.A., White, L., Griffin, A.M., Bell, R.S., Catton, C.N., Wunder, J.S., O'Sullivan, B., 2004. Radiation response: an additional unique signature of myxoid liposarcoma. Int. J. Radiat. Oncol. Biol. Phys. 60, 522–526. Piunti, A., Shilatifard, A., 2016. Epigenetic balance of gene expression by Polycomb and COMPASS families. Science 352, aad9780. Pollack, S.M., Jungbluth, A.A., Hoch, B.L., Farrar, E.A., Bleakley, M., Schneider, D.J., Loggers, E.T., Rodler, E., Eary, J.F., Conrad, E.U., Jones, R.L., Yee, C., 2012. NY-ESO-1 is a ubiquitous immunotherapeutic target antigen for patients with myxoid/round cell liposarcoma. Cancer. Ponomarenko, E.A., Poverennaya, E.V., Ilgisonis, E.V., Pyatnitskiy, M.A., Kopylov, A.T., Zgoda, V.G., Lisitsa, A.V., Archakov, A.I., 2016. The Size of the Human Proteome: The Width and Depth. Int J Anal Chem 2016, 7436849–6. Powers, C.A., Mathur, M., Raaka, B.M., Ron, D., Samuels, H.H., 1998. TLS (translocated-in-liposarcoma) is a high-affinity interactor for steroid, thyroid hormone, and retinoid receptors. Mol. Endocrinol. 12, 4–18. Powers, M.P., Wang, W.-L., Hernandez, V.S., Patel, K.S., Lev, D.C., Lazar, A.J., López-Terrada, D.H., 2010. Detection of myxoid liposarcoma-associated FUS-DDIT3 rearrangement variants including a newly identified breakpoint using an optimized RT-PCR assay. Mod. Pathol. 23, 1307–1315. 192  Poynter, S.T., Kadoch, C., 2016. Polycomb and trithorax opposition in development and disease. Wiley Interdiscip Rev Dev Biol 5, 659–688. Prasad, D.D., Ouchida, M., Lee, L., Rao, V.N., Reddy, E.S., 1994. TLS/FUS fusion domain of TLS/FUS-erg chimeric protein resulting from the t(16;21) chromosomal translocation in human myeloid leukemia functions as a transcriptional activation domain. Oncogene 9, 3717–3729. Prieto Granada, C.N., Wiesner, T., Messina, J.L., Jungbluth, A.A., Chi, P., Antonescu, C.R., 2016. Loss of H3K27me3 Expression Is a Highly Sensitive Marker for Sporadic and Radiation-induced MPNST. Am. J. Surg. Pathol. 40, 479–489. Prinjha, R.K., Witherington, J., Lee, K., 2012. Place your BETs: the therapeutic potential of bromodomains. Trends Pharmacol. Sci. 33, 146–153. Proudfoot, N.J., Brownlee, G.G., 1976. 3' non-coding region sequences in eukaryotic messenger RNA. Nature 263, 211–214. Rabani, M., Levin, J.Z., Fan, L., Adiconis, X., Raychowdhury, R., Garber, M., Gnirke, A., Nusbaum, C., Hacohen, N., Friedman, N., Amit, I., Regev, A., 2011. Metabolic labeling of RNA uncovers principles of RNA production and degradation dynamics in mammalian cells. Nat Biotechnol 29, 436–442. Rada-Iglesias, A., Bajpai, R., Swigut, T., Brugmann, S.A., Flynn, R.A., Wysocka, J., 2011. A unique chromatin signature uncovers early developmental enhancers in humans. Nature 470, 279–283. Rapp, T.B., Yang, L., Conrad, E.U., Mandahl, N., Chansky, H.A., 2002. RNA splicing mediated by YB-1 is inhibited by TLS/CHOP in human myxoid liposarcoma cells. J. Orthop. Res. 20, 723–729. Rappsilber, J., Ryder, U., Lamond, A.I., Mann, M., 2002. Large-scale proteomic analysis of the human spliceosome. Genome Res. 12, 1231–1245. Reinke, A.W., Baek, J., Ashenberg, O., Keating, A.E., 2013. Networks of bZIP protein-protein interactions diversified over a billion years of evolution. Science 340, 730–734. Remillard, D., Buckley, D.L., Paulk, J., Brien, G.L., Sonnett, M., Seo, H.-S., Dastjerdi, S., Wühr, M., Dhe-Paganon, S., Armstrong, S.A., Bradner, J.E., 2017. Degradation of the BAF Complex Factor BRD9 by Heterobifunctional Ligands. Angew. Chem. Int. Ed. Engl. 56, 5738–5743. Renner, M., Wolf, T., Meyer, H., Hartmann, W., Penzel, R., Ulrich, A., Lehner, B., Hovestadt, V., Czwan, E., Egerer, G., Schmitt, T., Alldinger, I., Renker, E.K., Ehemann, V., Eils, R., Wardelmann, E., Büttner, R., Lichter, P., Brors, B., Schirmacher, P., Mechtersheimer, G., 2013. Integrative DNA methylation and gene expression analysis in high-grade soft tissue sarcomas. Genome Biol 14, R137. Reynolds, N., Salmon-Divon, M., Dvinge, H., Hynes-Allen, A., Balasooriya, G., Leaford, D., Behrens, A., Bertone, P., Hendrich, B., 2012. NuRD-mediated deacetylation of H3K27 facilitates recruitment of Polycomb Repressive Complex 2 to direct gene repression. EMBO J 31, 593–605. Riggi, N., Cironi, L., Provero, P., Suvà, M.-L., Stehle, J.-C., Baumer, K., Guillou, L., Stamenkovic, I., 2006. Expression of the FUS-CHOP fusion protein in primary mesenchymal progenitor cells gives rise to a model of myxoid liposarcoma. Cancer Res. 66, 7016–7023. Riggi, N., Cironi, L., Suva, M.L., Stamenkovic, I., 2007. Sarcomas: genetics, signalling, and cellular origins. Part 1: The fellowship of TET. J. Pathol. 213, 4–20. 193  Riggi, N., Knoechel, B., Gillespie, S.M., Rheinbay, E., Boulay, G., Suvà, M.L., Rossetti, N.E., Boonseng, W.E., Oksuz, O., Cook, E.B., Formey, A., Patel, A., Gymrek, M., Thapar, V., Deshpande, V., Ting, D.T., Hornicek, F.J., Nielsen, G.P., Stamenkovic, I., Aryee, M.J., Bernstein, B.E., Rivera, M.N., 2014. EWS-FLI1 Utilizes Divergent Chromatin Remodeling Mechanisms to Directly Activate or Repress Enhancer Elements in Ewing Sarcoma. Cancer Cell 26, 668–681. Riising, E.M., Comet, I., Leblanc, B., Wu, X., Johansen, J.V., Helin, K., 2014. Gene silencing triggers polycomb repressive complex 2 recruitment to CpG islands genome wide. Molecular Cell 55, 347–360. Rogelj, B., Easton, L.E., Bogu, G.K., Stanton, L.W., Rot, G., Curk, T., Zupan, B., Sugimoto, Y., Modic, M., Haberman, N., Tollervey, J., Fujii, R., Takumi, T., Shaw, C.E., Ule, J., 2012. Widespread binding of FUS along nascent RNA regulates alternative splicing in the brain. Sci Rep 2, 603. Ron, D., Habener, J.F., 1992. CHOP, a novel developmentally regulated nuclear protein that dimerizes with transcription factors C/EBP and LAP and functions as a dominant-negative inhibitor of gene transcription. Genes Dev. 6, 439–453. Röhrich, M., Koelsche, C., Schrimpf, D., Capper, D., Sahm, F., Kratz, A., Reuss, J., Hovestadt, V., Jones, D.T.W., Bewerunge-Hudler, M., Becker, A., Weis, J., Mawrin, C., Mittelbronn, M., Perry, A., Mautner, V.-F., Mechtersheimer, G., Hartmann, C., Okuducu, A.F., Arp, M., Seiz-Rosenhagen, M., Hänggi, D., Heim, S., Paulus, W., Schittenhelm, J., Ahmadi, R., Herold-Mende, C., Unterberg, A., Pfister, S.M., Deimling, von, A., Reuss, D.E., 2016. Methylation-based classification of benign and malignant peripheral nerve sheath tumors. Acta Neuropathol 131, 877–887. Ruepp, A., Waegele, B., Lechner, M., Brauner, B., Dunger-Kaltenbach, I., Fobo, G., Frishman, G., Montrone, C., Mewes, H.W., 2010. CORUM: the comprehensive resource of mammalian protein complexes--2009. Nucleic Acids Res. 38, D497–501. Sadakierska-Chudy, A., Filip, M., 2015. A comprehensive view of the epigenetic landscape. Part II: Histone post-translational modification, nucleosome level, and chromatin regulation by ncRNAs. Neurotox Res 27, 172–197. Saito, T., Akaike, K., Kurisaki-Arakawa, A., Toda-Ishii, M., Mukaihara, K., Suehara, Y., Takagi, T., Kaneko, K., Yao, T., 2016. TERT promoter mutations are rare in bone and soft tissue sarcomas of Japanese patients. Mol Clin Oncol 4, 61–64. Sanchez, R., Meslamani, J., Zhou, M.-M., 2014. The bromodomain: from epigenome reader to druggable target. Biochim. Biophys. Acta 1839, 676–685. Sandberg, R., Neilson, J.R., Sarma, A., Sharp, P.A., Burge, C.B., 2008. Proliferating cells express mRNAs with shortened 3' untranslated regions and fewer microRNA target sites. Science 320, 1643–1647. Sankar, S., Bell, R., Stephens, B., Zhuo, R., Sharma, S., Bearss, D.J., Lessnick, S.L., 2013. Mechanism and relevance of EWS/FLI-mediated transcriptional repression in Ewing sarcoma. 32, 5089–5100. Sankar, S., Theisen, E.R., Bearss, J., Mulvihill, T., Hoffman, L.M., Sorna, V., Beckerle, M., Sharma, S., Lessnick, S.L., 2014. Reversible LSD1 inhibition interferes with global EWS/ETS transcriptional activity and impedes Ewing sarcoma tumor growth. Santos-Rosa, H., Schneider, R., Bannister, A.J., Sherriff, J., Bernstein, B.E., Emre, N.C.T., Schreiber, S.L., Mellor, J., Kouzarides, T., 2002. Active genes are tri-methylated at K4 of histone H3. Nature 419, 407–411. 194  Schaad, K., Strömbeck, B., Mandahl, N., Andersen, M.K., Heim, S., Mertens, F., Johansson, B., 2006. FISH mapping of i(7q) in acute leukemias and myxoid liposarcoma reveals clustered breakpoints in 7p11.2: implications for formation and pathogenetic outcome of the idic(7)(p11.2). Cytogenet. Genome Res. 114, 126–130. Schaefer, I.-M., Fletcher, C.D., Hornick, J.L., 2016. Loss of H3K27 trimethylation distinguishes malignant peripheral nerve sheath tumors from histologic mimics. Mod. Pathol. 29, 4–13. Schaefer, I.-M., Fletcher, C.D.M., 2018. Recent advances in the diagnosis of soft tissue tumours. Pathology 50, 37–48. Schena, M., Shalon, D., Davis, R.W., Brown, P.O., 1995. Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science 270, 467–470. Schiltz, R.L., Mizzen, C.A., Vassilev, A., Cook, R.G., Allis, C.D., Nakatani, Y., 1999. Overlapping but distinct patterns of histone acetylation by the human coactivators p300 and PCAF within nucleosomal substrates. J. Biol. Chem. 274, 1189–1192. Schröder, P., Förster, T., Kleine, S., Becker, C., Richters, A., Ziegler, S., Rauh, D., Kumar, K., Waldmann, H., 2015. Neuritogenic militarinone-inspired 4-hydroxypyridones target the stress pathway kinase MAP4K4. Angew. Chem. Int. Ed. Engl. 54, 12398–12403. Schuettengruber, B., Bourbon, H.-M., Di Croce, L., Cavalli, G., 2017. Genome Regulation by Polycomb and Trithorax: 70 Years and Counting. Cell 171, 34–57. Schuetze, S.M., Wathen, J.K., Lucas, D.R., Choy, E., Samuels, B.L., Staddon, A.P., Ganjoo, K.N., Mehren, von, M., Chow, W.A., Loeb, D.M., Tawbi, H.A., Rushing, D.A., Patel, S.R., Thomas, D.G., Chugh, R., Reinke, D.K., Baker, L.H., 2016. SARC009: Phase 2 study of dasatinib in patients with previously treated, high-grade, advanced sarcoma. Cancer 122, 868–874. Schutz, F.A.B., Choueiri, T.K., Sternberg, C.N., 2011. Pazopanib: Clinical development of a potent anti-angiogenic drug. Crit. Rev. Oncol. Hematol. 77, 163–171. Schwanhäusser, B., Busse, D., Li, N., Dittmar, G., Schuchhardt, J., Wolf, J., Chen, W., Selbach, M., 2011. Global quantification of mammalian gene expression control. Nature 473, 337–342. Schwartz, J.C., Cech, T.R., Parker, R.R., 2014. Biochemical Properties and Biological Functions of FET Proteins. Annu. Rev. Biochem. 84, 141210135300003. Schwartz, J.C., Ebmeier, C.C., Podell, E.R., Heimiller, J., Taatjes, D.J., Cech, T.R., 2012. FUS binds the CTD of RNA polymerase II and regulates its phosphorylation at Ser2. Genes Dev. 26, 2690–2695. Schwartz, J.C., Wang, X., Podell, E.R., Cech, T.R., 2013. RNA Seeds Higher-Order Assembly of FUS Protein. Cell Rep. Schwarzbach, M.H.M., Koesters, R., Germann, A., Mechtersheimer, G., Geisbill, J., Winkler, S., Niedergethmann, M., Ridder, R., Buechler, M.W., Knebel Doeberitz, von, M., Willeke, F., 2004. Comparable transforming capacities and differential gene expression patterns of variant FUS/CHOP fusion transcripts derived from soft tissue liposarcomas. Oncogene 23, 6798–6805. Selvanathan, S.P., Graham, G.T., Erkizan, H.V., Dirksen, U., Natarajan, T.G., Dakic, A., Yu, S., Liu, X., Paulsen, M.T., Ljungman, M.E., Wu, C.H., Lawlor, E.R., Uren, A., Toretsky, J.A., 2015. Oncogenic fusion protein EWS-FLI1 is a network hub that regulates alternative splicing. Proc. Natl. Acad. Sci. U.S.A. 112, 201500536–E1316. Seto, E., Yoshida, M., 2014. Erasers of histone acetylation: the histone deacetylase enzymes. Cold Spring Harb Perspect Biol 6, a018713–a018713. 195  Sévenet, N., Sheridan, E., Amram, D., Schneider, P., Handgretinger, R., Delattre, O., 1999. Constitutional mutations of the hSNF5/INI1 gene predispose to a variety of cancers. Am. J. Hum. Genet. 65, 1342–1348. Shatkin, A.J., Manley, J.L., 2000. The ends of the affair: capping and polyadenylation. Nat. Struct. Biol. 7, 838–842. Shi, Yongsheng, Di Giammartino, D.C., Taylor, D., Sarkeshik, A., Rice, W.J., Yates, J.R., Frank, J., Manley, J.L., 2009. Molecular architecture of the human pre-mRNA 3' processing complex. Molecular Cell 33, 365–376. Shi, Yujiang, Lan, F., Matson, C., Mulligan, P., Whetstine, J.R., Cole, P.A., Casero, R.A., Shi, Y., 2004. Histone demethylation mediated by the nuclear amine oxidase homolog LSD1. Cell 119, 941–953. Shogren-Knaak, M., Ishii, H., Sun, J.-M., Pazin, M.J., Davie, J.R., Peterson, C.L., 2006. Histone H4-K16 acetylation controls chromatin structure and protein interactions. Science 311, 844–847. Siegel, R.L., Miller, K.D., Jemal, A., 2017. Cancer Statistics, 2017. CA Cancer J Clin 67, 7–30. Sievers, E., Trautmann, M., Kindler, D., Huss, S., Gruenewald, I., Dirksen, U., Renner, M., Mechtersheimer, G., Pedeutour, F., Aman, P., Nishio, J., Schildhaus, H.U., Kirfel, J., Schirmacher, P., Wardelmann, E., Buettner, R., Hartmann, W., 2015. SRC inhibition represents a potential therapeutic strategy in liposarcoma. Int J Cancer 137, n/a–n/a. Simicevic, J., Deplancke, B., 2017. Transcription factor proteomics-Tools, applications, and challenges. Proteomics 17, 1600317. Simon, R., 2017. Critical Review of Umbrella, Basket, and Platform Designs for Oncology Clinical Trials. Clin. Pharmacol. Ther. 102, 934–941. Simó-Riudalbas, L., Esteller, M., 2015. Targeting the histone orthography of cancer: drugs for writers, erasers and readers. Br. J. Pharmacol. 172, 2716–2732. Singer, S., Socci, N.D., Ambrosini, G., Sambol, E., Decarolis, P., Wu, Y., O'Connor, R., Maki, R., Viale, A., Sander, C., Schwartz, G.K., Antonescu, C.R., 2007. Gene expression profiling of liposarcoma identifies distinct biological types/subtypes and potential therapeutic targets in well-differentiated and dedifferentiated liposarcoma. Cancer Res. 67, 6626–6636. Singh, I., Lee, S.-H., Sperling, A.S., Samur, M.K., Tai, Y.-T., Fulciniti, M., Munshi, N.C., Mayr, C., Leslie, C.S., 2018. Widespread intronic polyadenylation diversifies immune cell transcriptomes. Nat Commun 9, 1716. Sleijfer, S., Ray-Coquard, I., Papai, Z., Le Cesne, A., Scurr, M., Schöffski, P., Collin, F., Pandite, L., Marreaud, S., De Brauwer, A., van Glabbeke, M., Verweij, J., Blay, J.-Y., 2009. Pazopanib, a multikinase angiogenesis inhibitor, in patients with relapsed or refractory advanced soft tissue sarcoma: a phase II study from the European organisation for research and treatment of cancer-soft tissue and bone sarcoma group (EORTC study 62043). J Clin Oncol 27, 3126–3132. Smits, A.H., Jansen, P.W.T.C., Poser, I., Hyman, A.A., Vermeulen, M., 2013. Stoichiometry of chromatin-associated protein complexes revealed by label-free quantitative mass spectrometry-based proteomics. Nucleic Acids Res. 41, e28–e28. Somaiah, N., Block, M.S., Kim, J.W., Shapiro, G., Do, K.T., Hwu, P., Eder, J.P., Jones, R.L., Gnjatic, S., Lu, H., Meulen, Ter, J., Hsu, F.J., Pollack, S., 2016. Single-agent LV305 to induce anti-tumor immune and clinical responses in patients with advanced or metastatic sarcoma and other cancers expressing NY-ESO-1. JCO 34, 3093–3093. Somaiah, N., Chawla, S.P., Block, M.S., Morris, J.C., Do, K.T., Kim, J.W., Druta, M., Sankhala, K.K., Hwu, P., Gnjatic, S., Lu, H., Kenney, R.T., Bohac, G.C., Pollack, S., 2017. Immune 196  response, safety, and survival impact from CMB305 in NY-ESO-1+ recurrent soft tissue sarcomas (STS). JCO 35, 11006–11006. Spitzer, J.I., Ugras, S., Runge, S., Decarolis, P., Antonescu, C., Tuschl, T., Singer, S., 2011. mRNA and protein levels of FUS, EWSR1, and TAF15 are upregulated in liposarcoma. Genes Chromosomes Cancer 50, 338–347. Sreekantaiah, C., Karakousis, C.P., Leong, S.P., Sandberg, A.A., 1991. Trisomy 8 as a nonrandom secondary change in myxoid liposarcoma. Cancer Genet. Cytogenet. 51, 195–205. Staley, J.P., Woolford, J.L., 2009. Assembly of ribosomes and spliceosomes: complex ribonucleoprotein machines. Curr. Opin. Cell Biol. 21, 109–118. Stanton, B.Z., Hodges, C., Calarco, J.P., Braun, S.M.G., Ku, W.L., Kadoch, C., Zhao, K., Crabtree, G.R., 2017. Smarca4 ATPase mutations disrupt direct eviction of PRC1 from chromatin. Nat. Genet. 49, 282–288. Stark, C., Breitkreutz, B.-J., Reguly, T., Boucher, L., Breitkreutz, A., Tyers, M., 2006. BioGRID: a general repository for interaction datasets. Nucleic Acids Res. 34, D535–9. Stathis, A., Zucca, E., Bekradda, M., Gomez-Roca, C., Delord, J.-P., La Motte Rouge, de, T., Uro-Coste, E., de Braud, F., Pelosi, G., French, C.A., 2016. Clinical Response of Carcinomas Harboring the BRD4-NUT Oncoprotein to the Targeted Bromodomain Inhibitor OTX015/MK-8628. Cancer Discov 6, 492–500. Steffen, P.A., Ringrose, L., 2014. What are memories made of? How Polycomb and Trithorax proteins mediate epigenetic memory. Nat Rev Mol Cell Biol 15, 340–356. Stephan-Otto Attolini, C., Peña, V., Rossell, D., 2015. Designing alternative splicing RNA-seq studies. Beyond generic guidelines. Bioinformatics 31, 3631–3637. Straessler, K.M., Jones, K.B., Hu, H., Jin, H., van de Rijn, M., Capecchi, M.R., 2013. Modeling clear cell sarcomagenesis in the mouse: cell of origin differentiation state impacts tumor characteristics. Cancer Cell 23, 215–227. Su, L., Sampaio, A.V., Jones, K.B., Pacheco, M., Goytain, A., Lin, S., Poulin, N., Yi, L., Rossi, F.M., Kast, J., Capecchi, M.R., Underhill, T.M., Nielsen, T.O., 2012. Deconstruction of the SS18-SSX fusion oncoprotein complex: insights into disease etiology and therapeutics. Cancer Cell 21, 333–347. Su, N., Kilberg, M.S., 2008. C/EBP homology protein (CHOP) interacts with activating transcription factor 4 (ATF4) and negatively regulates the stress-dependent induction of the asparagine synthetase gene. J. Biol. Chem. 283, 35106–35117. Sun, S., Ling, S.C., Qiu, J., Albuquerque, C.P., Zhou, Y., Tokunaga, S., Li, H., Qiu, H., Bui, A., Yeo, G.W., Huang, E.J., Eggan, K., Zhou, H., Fu, X.-D., Lagier-Tourenne, C., Cleveland, D.W., 2015. ALS-causative mutations in FUS/TLS confer gain and loss of function by altered association with SMN and U1-snRNP. Nat Commun 6, 6171. Suva, M.L., Riggi, N., Bernstein, B.E., 2013. Epigenetic Reprogramming in Cancer. Science 339, 1567–1570. Swift, J., Coruzzi, G.M., 2017. A matter of time - How transient transcription factor interactions create dynamic gene regulatory networks. Biochim. Biophys. Acta 1860, 75–83. Szklarczyk, D., Franceschini, A., Wyder, S., Forslund, K., Heller, D., Huerta-Cepas, J., Simonovic, M., Roth, A., Santos, A., Tsafou, K.P., Kuhn, M., Bork, P., Jensen, L.J., Mering, von, C., 2015. STRING v10: protein-protein interaction networks, integrated over the tree of life. Nucleic Acids Res. 43, D447–52. 197  Szurian, K., Kashofer, K., Liegl-Atzwanger, B., 2017. Role of Next-Generation Sequencing as a Diagnostic Tool for the Evaluation of Bone and Soft-Tissue Tumors. Pathobiology 84, 323–338. Taipale, M., Rea, S., Richter, K., Vilar, A., Lichter, P., Imhof, A., Akhtar, A., 2005. hMOF histone acetyltransferase is required for histone H4 lysine 16 acetylation in mammalian cells. Mol. Cell. Biol. 25, 6798–6810. Takahama, K., Takada, A., Tada, S., Shimizu, M., Sayama, K., Kurokawa, R., Oyoshi, T., 2013. Regulation of telomere length by G-quadruplex telomere DNA- and TERRA-binding protein TLS/FUS. Chem. Biol. 20, 341–350. Takahashi, K., Yamanaka, S., 2006. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126, 663–676. Tang, Q.-Q., Lane, M.D., 2012. Adipogenesis: from stem cell to adipocyte. Annu. Rev. Biochem. 81, 715–736. Tang, Q.Q., Lane, M.D., 2000. Role of C/EBP homologous protein (CHOP-10) in the programmed activation of CCAAT/enhancer-binding protein-beta during adipogenesis. Proc. Natl. Acad. Sci. U.S.A. 97, 12446–12450. Tang, X., Guilherme, A., Chakladar, A., Powelka, A.M., Konda, S., Virbasius, J.V., Nicoloro, S.M.C., Straubhaar, J., Czech, M.P., 2006. An RNA interference-based screen identifies MAP4K4/NIK as a negative regulator of PPARgamma, adipogenesis, and insulin-responsive hexose transport. Proc. Natl. Acad. Sci. U.S.A. 103, 2087–2092. Tangutoori, S., Baldwin, P., Sridhar, S., 2015. PARP inhibitors: A new era of targeted therapy. Maturitas 81, 5–9. Tap, W.D., Jones, R.L., Van Tine, B.A., Chmielowski, B., Elias, A.D., Adkins, D., Agulnik, M., Cooney, M.M., Livingston, M.B., Pennock, G., Hameed, M.R., Shah, G.D., Qin, A., Shahir, A., Cronier, D.M., Ilaria, R., Conti, I., Cosaert, J., Schwartz, G.K., 2016. Olaratumab and doxorubicin versus doxorubicin alone for treatment of soft-tissue sarcoma: an open-label phase 1b and randomised phase 2 trial. Lancet 388, 488–497. Tap, W.D., Papai, Z., Van Tine, B.A., Attia, S., Ganjoo, K.N., Jones, R.L., Schuetze, S., Reed, D., Chawla, S.P., Riedel, R.F., Krarup-Hansen, A., Toulmonde, M., Ray-Coquard, I., Hohenberger, P., Grignani, G., Cranmer, L.D., Okuno, S., Agulnik, M., Read, W., Ryan, C.W., Alcindor, T., Del Muro, X.F.G., Budd, G.T., Tawbi, H., Pearce, T., Kroll, S., Reinke, D.K., Schöffski, P., 2017. Doxorubicin plus evofosfamide versus doxorubicin alone in locally advanced, unresectable or metastatic soft-tissue sarcoma (TH CR-406/SARC021): an international, multicentre, open-label, randomised phase 3 trial. Lancet Oncol. 18, 1089–1103. Thelin-Järnum, S., Göransson, M., Burguete, A.S., Olofsson, A., Aman, P., 2002. The myxoid liposarcoma specific TLS-CHOP fusion protein localizes to nuclear structures distinct from PML nuclear bodies. Int J Cancer 97, 446–450. Thomas, R., Al-Khadairi, G., Roelands, J., Hendrickx, W., Dermime, S., Bedognetti, D., Decock, J., 2018. NY-ESO-1 Based Immunotherapy of Cancer: Current Perspectives. Front Immunol 9, 947. Thomas, T., Dixon, M.P., Kueh, A.J., Voss, A.K., 2008. Mof (MYST1 or KAT8) is essential for progression of embryonic development past the blastocyst stage and required for normal chromatin architecture. Mol. Cell. Biol. 28, 5093–5105. Thomsen, C., Grundevik, P., Elias, P., Ståhlberg, A., Aman, P., 2013. A conserved N-terminal motif is required for complex formation between FUS, EWSR1, TAF15 and their oncogenic fusion proteins. FASEB J. 198  Thorpe, L.M., Yuzugullu, H., Zhao, J.J., 2015. PI3K in cancer: divergent roles of isoforms, modes of activation and therapeutic targeting. Nature Reviews Cancer 15, 7–24. Tian, B., Manley, J.L., 2017. Alternative polyadenylation of mRNA precursors. Nat Rev Mol Cell Biol 18, 18–30. Tien, J.F., Mazloomian, A., Cheng, S.-W.G., Hughes, C.S., Chow, C.C.T., Canapi, L.T., Oloumi, A., Trigo-Gonzalez, G., Bashashati, A., Xu, J., Chang, V.C.-D., Shah, S.P., Aparicio, S., Morin, G.B., 2017. CDK12 regulates alternative last exon mRNA splicing and promotes breast cancer cell invasion. Nucleic Acids Res. 45, 6698–6716. Tirode, F., Surdez, D., Ma, X., Parker, M., Le Deley, M.C., Bahrami, A., Zhang, Z., Lapouble, E., Grossetête-Lalami, S., Rusch, M., Reynaud, S., Rio-Frio, T., Hedlund, E., Wu, G., Chen, X., Pierron, G., Oberlin, O., Zaidi, S., Lemmon, G., Gupta, P., Vadodaria, B., Easton, J., Gut, M., Ding, L., Mardis, E.R., Wilson, R.K., Shurtleff, S., Laurence, V., Michon, J., Marec-Bérard, P., Gut, I., Downing, J., Dyer, M., Zhang, J., Delattre, O., St. Jude Children's Research Hospital–Washington University Pediatric Cancer Genome Project and the International Cancer Genome Consortium, 2014. Genomic landscape of Ewing sarcoma defines an aggressive subtype with co-association of STAG2 and TP53 mutations. Cancer Discov 4, 1342–1353. Tomazou, E.M., Sheffield, N.C., Schmidl, C., Schuster, M., Schönegger, A., Datlinger, P., Kubicek, S., Bock, C., Kovar, H., 2015. Epigenome mapping reveals distinct modes of gene regulation and widespread enhancer reprogramming by the oncogenic fusion protein EWS-FLI1. Cell Rep 10, 1082–1095. Tong, J.K., Hassig, C.A., Schnitzler, G.R., Kingston, R.E., Schreiber, S.L., 1998. Chromatin deacetylation by an ATP-dependent nucleosome remodelling complex. Nature 395, 917–921. Toretsky, J.A., Erkizan, V., Levenson, A., Abaan, O.D., Parvin, J.D., Cripe, T.P., Rice, A.M., Lee, S.B., Uren, A., 2006. Oncoprotein EWS-FLI1 activity is enhanced by RNA helicase A. Cancer Res. 66, 5574–5581. Toure, M., Crews, C.M., 2016. Small-Molecule PROTACS: New Approaches to Protein Degradation. Angew. Chem. Int. Ed. Engl. 55, 1966–1973. Tsiatsiani, L., Heck, A.J.R., 2015. Proteomics beyond trypsin. FEBS J. 282, 2612–2626. Tsukada, J., Yoshida, Y., Kominato, Y., Auron, P.E., 2011. The CCAAT/enhancer (C/EBP) family of basic-leucine zipper (bZIP) transcription factors is a multifaceted highly-regulated system for gene regulation. Cytokine 54, 6–19. Tyagi, M., Imam, N., Verma, K., Patel, A.K., 2016. Chromatin remodelers: We are the drivers!! Nucleus 7, 388–404. Ubeda, M., Vallejo, M., Habener, J.F., 1999. CHOP enhancement of gene transcription by interactions with Jun/Fos AP-1 complex proteins. Mol. Cell. Biol. 19, 7589–7599. Ubeda, M., Wang, X.Z., Zinszner, H., Wu, I., Habener, J.F., Ron, D., 1996. Stress-induced binding of the transcriptional factor CHOP to a novel DNA control element. Mol. Cell. Biol. 16, 1479–1489. Uboldi, S., Bernasconi, S., Romano, M., Marchini, S., Fuso Nerini, I., Damia, G., Ganzinelli, M., Marangon, E., Sala, F., Clivio, L., Chiorino, G., Digiandomenico, S., Rocchi, M., Capozzi, O., Margison, G., Watson, A., Caccuri, A., Pastore, A., Fossati, A., Mantovani, R., Grosso, F., Tercero, J., Erba, E., D'Incalci, M., 2011. Characterization of a new trabectedin resistant myxoid liposarcoma cell line that shows collateral sensitivity to methylating agents. Int J Cancer. 199  Uboldi, S., Craparotta, I., Colella, G., Ronchetti, E., Beltrame, L., Vicario, S., Marchini, S., Panini, N., Dagrada, G., Bozzi, F., Pilotti, S., Galmarini, C.M., D'Incalci, M., Gatta, R., 2017. Mechanism of action of trabectedin in desmoplastic small round cell tumor cells. BMC Cancer 17, 107. Uranishi, H., Tetsuka, T., Yamashita, M., Asamitsu, K., Shimizu, M., Itoh, M., Okamoto, T., 2001. Involvement of the pro-oncoprotein TLS (translocated in liposarcoma) in nuclear factor-kappa B p65-mediated transcription as a coactivator. J. Biol. Chem. 276, 13395–13401. van der Graaf, W.T.A., Blay, J.-Y., Chawla, S.P., Kim, D.-W., Bui-Nguyen, B., Casali, P.G., Schöffski, P., Aglietta, M., Staddon, A.P., Beppu, Y., Le Cesne, A., Gelderblom, H., Judson, I.R., Araki, N., Ouali, M., Marreaud, S., Hodge, R., Dewji, M.R., Coens, C., Demetri, G.D., Fletcher, C.D., Dei Tos, A.P., Hohenberger, P., EORTC Soft Tissue and Bone Sarcoma Group, PALETTE study group, 2012. Pazopanib for metastatic soft-tissue sarcoma (PALETTE): a randomised, double-blind, placebo-controlled phase 3 trial. Lancet 379, 1879–1886. van Dijk, E.L., Auger, H., Jaszczyszyn, Y., Thermes, C., 2014. Ten years of next-generation sequencing technology. Trends Genet. 30, 418–426. Vandesompele, J., De Preter, K., Pattyn, F., Poppe, B., Van Roy, N., De Paepe, A., Speleman, F., 2002. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3, RESEARCH0034. Velculescu, V.E., Zhang, L., Vogelstein, B., Kinzler, K.W., 1995. Serial analysis of gene expression. Science 270, 484–487. Versteege, I., Sévenet, N., Lange, J., Rousseau-Merck, M.F., Ambros, P., Handgretinger, R., Aurias, A., Delattre, O., 1998. Truncating mutations of hSNF5/INI1 in aggressive paediatric cancer. Nature 394, 203–206. Vinagre, J., Almeida, A., Pópulo, H., Batista, R., Lyra, J., Pinto, V., Coelho, R., Celestino, R., Prazeres, H., Lima, L., Melo, M., da Rocha, A.G., Preto, A., Castro, P., Castro, L., Pardal, F., Lopes, J.M., Santos, L.L., Reis, R.M., Cameselle-Teijeiro, J., Sobrinho-Simões, M., Lima, J., Máximo, V., Soares, P., 2013. Frequency of TERT promoter mutations in human cancers. Nat Commun 4, 2185. Vitorino, P., Yeung, S., Crow, A., Bakke, J., Smyczek, T., West, K., McNamara, E., Eastham-Anderson, J., Gould, S., Harris, S.F., Ndubaku, C., Ye, W., 2015. MAP4K4 regulates integrin-FERM binding to control endothelial cell motility. Nature 519, 425–430. Vlenterie, M., Hillebrandt-Roeffen, M.H.S., Flucke, U.E., Groenen, P.J.T.A., Tops, B.B.J., Kamping, E.J., Pfundt, R., de Bruijn, D.R.H., Geurts van Kessel, A.H.M., van Krieken, H.J.H.J.M., van der Graaf, W.T.A., Versleijen-Jonkers, Y.M.H., 2015. Next generation sequencing in synovial sarcoma reveals novel gene mutations. Oncotarget 6, 34680–34690. Wahle, E., 1991. A novel poly(A)-binding protein acts as a specificity factor in the second phase of messenger RNA polyadenylation. Cell 66, 759–768. Wang, Bo, Shen, Z.-L., Gao, Z.-D., Zhao, G., Wang, C.-Y., Yang, Y., Zhang, J.-Z., Yan, Y.-C., Shen, C., Jiang, K.-W., Ye, Y.-J., Wang, S., 2015. MiR-194, commonly repressed in colorectal cancer, suppresses tumor growth by regulating the MAP4K4/c-Jun/MDM2 signaling pathway. Cell Cycle 14, 1046–1058. Wang, Eric T, Sandberg, R., Luo, S., Khrebtukova, I., Zhang, L., Mayr, C., Kingsmore, S.F., Schroth, G.P., Burge, C.B., 2008. Alternative isoform regulation in human tissue transcriptomes. Nature 456, 470–476. 200  Wang, Feng, Marshall, C.B., Ikura, M., 2013. Transcriptional/epigenetic regulator CBP/p300 in tumorigenesis: structural and functional versatility in target recognition. Cell. Mol. Life Sci. 70, 3989–4008. Wang, Shaomeng, Zhao, Y., Aguilar, A., Bernard, D., Yang, C.-Y., 2017. Targeting the MDM2-p53 Protein-Protein Interaction for New Cancer Therapy: Progress and Challenges. Cold Spring Harb Perspect Med 7, a026245. Wang, Tao, Jiang, X., Chen, G., Xu, J., 2015. Interaction of amyotrophic lateral sclerosis/frontotemporal lobar degeneration-associated fused-in-sarcoma with proteins involved in metabolic and protein degradation pathways. Neurobiology of Aging 36, 527–535. Wang, Xiangting, Arai, S., Song, X., Reichart, D., Du, K., Pascual, G., Tempst, P., Rosenfeld, M.G., Glass, C.K., Kurokawa, R., 2008. Induced ncRNAs allosterically modify RNA-binding proteins in cis to inhibit transcription. 454, 126–130. Wang, Xiaofeng, Lee, R.S., Alver, B.H., Haswell, J.R., Wang, S., Mieczkowski, J., Drier, Y., Gillespie, S.M., Archer, T.C., Wu, J.N., Tzvetkov, E.P., Troisi, E.C., Pomeroy, S.L., Biegel, J.A., Tolstorukov, M.Y., Bernstein, B.E., Park, P.J., Roberts, C.W.M., 2017. SMARCB1-mediated SWI/SNF complex function is essential for enhancer regulation. Nat. Genet. 49, 289–295. Wang, Xueyin, Schwartz, J.C., Cech, T.R., 2015. Nucleic acid-binding specificity of human FUS protein. Nucleic Acids Res. 43, 7535–7543. Wang, Yan, Zhang, H., Chen, Y., Sun, Y., Yang, F., Yu, W., Liang, J., Sun, L., Yang, X., Shi, L., Li, R., Li, Y., Zhang, Y., Li, Q., Yi, X., Shang, Y., 2009. LSD1 is a subunit of the NuRD complex and targets the metastasis programs in breast cancer. 138, 660–672. Wang, Zefeng, Burge, C.B., 2008. Splicing regulation: from a parts list of regulatory elements to an integrated splicing code. RNA 14, 802–813. Watson, S., Perrin, V., Guillemot, D., Reynaud, S., Coindre, J.-M., Karanian, M., Guinebretière, J.-M., Fréneaux, P., Le Loarer, F., Bouvet, M., Galmiche-Rolland, L., Larousserie, F., Longchampt, E., Ranchere-Vince, D., Pierron, G., Delattre, O., Tirode, F., 2018. Transcriptomic definition of molecular subgroups of small round cell sarcomas. J. Pathol. 245, 29–40. Wheeler, D.A., Srinivasan, M., Egholm, M., Shen, Y., Chen, L., McGuire, A., He, W., Chen, Y.-J., Makhijani, V., Roth, G.T., Gomes, X., Tartaro, K., Niazi, F., Turcotte, C.L., Irzyk, G.P., Lupski, J.R., Chinault, C., Song, X.-Z., Liu, Y., Yuan, Y., Nazareth, L., Qin, X., Muzny, D.M., Margulies, M., Weinstock, G.M., Gibbs, R.A., Rothberg, J.M., 2008. The complete genome of an individual by massively parallel DNA sequencing. Nature 452, 872–876. Whyte, W.A., Bilodeau, S., Orlando, D.A., Hoke, H.A., Frampton, G.M., Foster, C.T., Cowley, S.M., Young, R.A., 2012. Enhancer decommissioning by LSD1 during embryonic stem cell differentiation. Nature 482, 221–225. Will, C.L., Lührmann, R., 2011. Spliceosome structure and function. Cold Spring Harb Perspect Biol 3, a003707–a003707. Willems, S.M., Schrage, Y.M., Bruijn, I.H.B.-D., Szuhai, K., Hogendoorn, P.C.W., Bovée, J.V.M.G., 2010. Kinome profiling of myxoid liposarcoma reveals NF-kappaB-pathway kinase activity and casein kinase II inhibition as a potential treatment option. Mol. Cancer 9, 257. Wojcik, J., Cooper, K., 2017. Epigenetic Alterations in Bone and Soft Tissue Tumors. Advances In Anatomic Pathology 24, 362–371. 201  Wright, J.H., Wang, X., Manning, G., LaMere, B.J., Le, P., Zhu, S., Khatry, D., Flanagan, P.M., Buckley, S.D., Whyte, D.B., Howlett, A.R., Bischoff, J.R., Lipson, K.E., Jallal, B., 2003. The STE20 kinase HGK is broadly expressed in human tumor cells and can modulate cellular transformation, invasion, and adhesion. Mol. Cell. Biol. 23, 2068–2082. Wu, Z., Rosen, E.D., Brun, R., Hauser, S., Adelmant, G., Troy, A.E., McKeon, C., Darlington, G.J., Spiegelman, B.M., 1999. Cross-regulation of C/EBP alpha and PPAR gamma controls the transcriptional pathway of adipogenesis and insulin sensitivity. Molecular Cell 3, 151–158. Xue, Y., Wong, J., Moreno, G.T., Young, M.K., Côté, J., Wang, W., 1998. NURD, a novel complex with both ATP-dependent chromatin-remodeling and histone deacetylase activities. Molecular Cell 2, 851–861. Yamazaki, T., Chen, S., Yu, Y., Yan, B., Haertlein, T.C., Carrasco, M.A., Tapia, J.C., Zhai, B., Das, R., Lalancette-Hebert, M., Sharma, A., Chandran, S., Sullivan, G., Nishimura, A.L., Shaw, C.E., Gygi, S.P., Shneider, N.A., Maniatis, T., Reed, R., 2012. FUS-SMN protein interactions link the motor neuron diseases ALS and SMA. Cell Rep 2, 799–806. Yang, E., van Nimwegen, E., Zavolan, M., Rajewsky, N., Schroeder, M., Magnasco, M., Darnell, J.E., 2003. Decay rates of human mRNAs: correlation with functional characteristics and sequence attributes. Genome Res. 13, 1863–1872. Yang, L., Embree, L.J., Tsai, S., Hickstein, D.D., 1998. Oncoprotein TLS interacts with serine-arginine proteins involved in RNA splicing. J. Biol. Chem. 273, 27761–27764. Yang, Y., Liu, L., Naik, I., Braunstein, Z., Zhong, J., Ren, B., 2017. Transcription Factor C/EBP Homologous Protein in Health and Diseases. Front Immunol 8, 1612. Yoo, K.H., Kim, H.S., Lee, S.J., Park, S.H., Kim, S.J., Kim, S.H., La Choi, Y., Shin, K.-H., Cho, Y.J., Lee, J., Rha, S.Y., 2015. Efficacy of pazopanib monotherapy in patients who had been heavily pretreated for metastatic soft tissue sarcoma: a retrospective case series. BMC Cancer 15, 154. You, A., Tong, J.K., Grozinger, C.M., Schreiber, S.L., 2001. CoREST is an integral component of the CoREST- human histone deacetylase complex. Proc. Natl. Acad. Sci. U.S.A. 98, 1454–1458. Young, R.J., Natukunda, A., Litière, S., Woll, P.J., Wardelmann, E., van der Graaf, W.T.A., 2014. First-line anthracycline-based chemotherapy for angiosarcoma and other soft tissue sarcoma subtypes: pooled analysis of eleven European Organisation for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group trials. Eur. J. Cancer 50, 3178–3186. Yu, Y., Reed, R., 2015. FUS functions in coupling transcription to splicing by mediating an interaction between RNAP II and U1 snRNP. Proc. Natl. Acad. Sci. U.S.A. 112, 8608–8613. Zhang, J., Manley, J.L., 2013. Misregulation of pre-mRNA alternative splicing in cancer. Cancer Discov 3, 1228–1237. Zhang, M., Wang, Y., Jones, S., Sausen, M., McMahon, K., Sharma, R., Wang, Q., Belzberg, A.J., Chaichana, K., Gallia, G.L., Gokaslan, Z.L., Riggins, G.J., Wolinksy, J.-P., Wood, L.D., Montgomery, E.A., Hruban, R.H., Kinzler, K.W., Papadopoulos, N., Vogelstein, B., Bettegowda, C., 2014. Somatic mutations of SUZ12 in malignant peripheral nerve sheath tumors. Nat. Genet. 46, 1170–1172. Zhang, T., Wu, Y.-C., Mullane, P., Ji, Y.J., Liu, H., He, L., Arora, A., Hwang, H.-Y., Alessi, A.F., Niaki, A.G., Periz, G., Guo, L., Wang, H., Elkayam, E., Joshua-Tor, L., Myong, S., Kim, J.K., Shorter, J., Ong, S.-E., Leung, A.K.L., Wang, J., 2018. FUS Regulates Activity of MicroRNA-Mediated Gene Silencing. Molecular Cell 69, 787–801.e8. 202  Zhang, Y., LeRoy, G., Seelig, H.P., Lane, W.S., Reinberg, D., 1998. The dermatomyositis-specific autoantigen Mi2 is a component of a complex containing histone deacetylase and nucleosome remodeling activities. Cell 95, 279–289. Zhao, Y., Garcia, B.A., 2015. Comprehensive Catalog of Currently Documented Histone Modifications. Cold Spring Harb Perspect Biol 7, a025064. Zhou, B.P., Liao, Y., Xia, W., Zou, Y., Spohn, B., Hung, M.C., 2001. HER-2/neu induces p53 ubiquitination via Akt-mediated MDM2 phosphorylation. Nature Cell Biology 3, 973–982. Zhou, Z., Licklider, L.J., Gygi, S.P., Reed, R., 2002. Comprehensive proteomic analysis of the human spliceosome. Nature 419, 182–185. Zinszner, H., Albalat, R., Ron, D., 1994. A novel effector domain from the RNA-binding protein TLS or EWS is required for oncogenic transformation by CHOP. Genes Dev. 8, 2513–2526. Zinszner, H., Kuroda, M., Wang, X., Batchvarova, N., Lightfoot, R.T., Remotti, H., Stevens, J.L., Ron, D., 1998. CHOP is implicated in programmed cell death in response to impaired function of the endoplasmic reticulum. Genes Dev. 12, 982–995. Zubarev, R.A., 2013. The challenge of the proteome dynamic range and its implications for in-depth proteomics. Proteomics 13, 723–726.   203  Appendices  Appendix A     A.1 C/EBPβ is expressed in myxoid liposarcoma cell lines   LAP*LAPLIPIB: C/EBPβHEK293T402-911765-92DL-221IB: GAPDHWhole cell lysates204  A.2 List of putative FUS-DDIT3 interactors that were previously reported as DDIT3 interactors (from the BioGRID database), or FUS interactors (from published FUS interactomes) Bait protein (DDIT3) indicated in red text; Previously reported DDIT3 or FUS interactors indicated in blue text. Rank Accession Gene Name DDIT3 (BioGRID) (n=76) FUS interactome studies Yamazaki et al. (n=158) Wang et al. (n=120) Sun et al. (n=98) Kamelgarn et al. (n=67) Chi et al. (n=156) 1 Q15233 NONO      ✓        2 P23246 SFPQ      ✓        3 Q96KR1 ZFR    ✓         ✓  4 P35638 DDIT3  ✓            5 Q96R06 SPAG5             6 Q14966 ZNF638             7 Q9Y4B5 SOGA2             8 Q5BKZ1 ZNF326          ✓   ✓  9 Q7L590 MCM10             10 Q9UH17 APOBEC3B             11 Q69YN4 KIAA1429             12 Q9BVP2 GNL3            ✓  13 Q15058 KIF14             14 P35637 FUS    ✓     ✓     ✓  15 Q96PK6 RBM14        ✓     ✓  16 J3QR09 RPL19             17 Q5T9A4 ATAD3B             18 P02545 LMNA             19 Q9UIG0 BAZ1B             20 P14866 HNRNPL      ✓   ✓     ✓  21 Q9HAU0 PLEKHA5             22 Q13595 TRA2A    ✓       ✓   ✓  23 Q8WUT1 POLDIP3             205  24 P17676 CEBPB  ✓            25 Q14444 CAPRIN1    ✓       ✓   ✓  26 Q86SQ0 PHLDB2             27 A0A087WV66 MKI67    ✓          28 Q8WXF1 PSPC1      ✓        29 Q8N684 CPSF7    ✓     ✓      30 Q5SW79 CEP170             31 P09874 PARP1          ✓    32 Q16537 PPP2R5E             33 P49790 NUP153             34 Q13415 ORC1             35 Q9BTC0 DIDO1             36 Q8IX01 SUGP2             37 H3BQZ7 HNRNPUL2-BSCL2             38 Q15678 PTPN14             39 P51608 MECP2             40 Q8WUU5 GATAD1             41 M0R1G4 NAV1             42 A0A0A0MTL4 NAV2             43 P78332 RBM6          ✓    44 Q99569 PKP4             45 F8WJN3 CPSF6      ✓        46 Q9NPG3 UBN1             47 O14578 CIT             48 Q9UPN4 AZI1             49 Q6UN15 FIP1L1             50 Q6DN90 IQSEC1             51 P48634 PRRC2A            ✓  206  52 Q7Z589 C11orf30  ✓            53 Q10570 CPSF1        ✓      54 O00411 POLRMT             55 Q9Y678 COPG1             56 J3QT28 BUB3             57 Q9Y2U8 LEMD3             58 O95232 LUC7L3             59 V9GY01 KNSTRN             60 Q96ST2 IWS1             61 A0A087WZ30 RECQL4             62 A0A087WTX2 SLC39A7             63 Q63ZY3 KANK2             64 Q96QT6 PHF12             65 O14974 PPP1R12A             66 E7EN19 MAP4K4             67 A0A5E8 GAS2L1             68 P17812 CTPS1             69 Q5JR04 MOV10          ✓   ✓  70 Q9UGU0 TCF20             71 Q8N556 AFAP1             72 Q6UXN9 WDR82            ✓  73 Q2M1P5 KIF7             74 Q8N0Z3 SPICE1             75 Q9UHI6 DDX20    ✓         ✓    Total 3 7 5 5 6 12   207  A.3 Overlap of proteins detected in the FUS, DDIT3 and FUS-DDIT3 interactomes The FUS interactome is a combination of datasets from 5 previously published FUS interactomes (Chi et al., 2018; Kamelgarn et al., 2016; Sun et al., 2015; Tao Wang et al., 2015; Yamazaki et al., 2012). The DDIT3 interactome was obtained from the BioGRID database. The two proteins common between the FUS and DDIT3 interactome are CSNK2A1 and KPNA2. Other common proteins are indicated in Appendix A.2.      23213905171208  A.4 Assessment of MAP4K4 inibition in myxoid liposarcoma cells DL-221 after GNE-495 treatment Assessed only in DL-221 as the other two myxoid liposarcoma cell lines (402-91 and 1765-92) do not express PPARy (see Appendix A.8). siRNA-mediated knockdown of MAP4K4 was included to corroborate effects of GNE-495 treatment.     209  A.5 Overlap of common proteins identified in previously published FUS interactomes   210  A.6 Interpro analysis of FUS-DDIT3 (Type 1) amino acid sequence    FUS-DDIT3211  A.7 MAP4K4 is overexpressed in myxoid liposarcoma Heatmap is generated from public microarray data (Singer et al., 2007). MLS = myxoid liposarcoma.  212  A.8 PPARG expression is high in myxoid liposarcomas Heat maps generated from two public microarray datasets (Renner et al., 2013; Singer et al., 2007) show that PPARG expression is high in myxoid liposarcomas (MLS) and normal adipocytes (ADNORM), but low in cell lines 402-91 (MLS402) and 1765-92 (MLS1765). Western blot: PPARγ protein expression is also low in myxoid liposarcoma cell lines compared to positive control bladder cancer cell line UC13. PPARG siRNA treatment was included to confirm location of PPARγ band on western blot.    Singer et al. 2007Renner et al. 2013- + - + - + - + - +402-91 1765-92 DL-221 HeLa UC13PPARG siRNA (20 nM, 48hr)IB: PPARγ+ve ctrl cell line. 213  Appendix B   B.1 Proteins in the FUS-DDIT3 interactome that are present in the Spliceosome Database Spliceosome Database (http://spliceosomedb.ucsc.edu). WCL = whole cell lysate. Second column indicates rank of protein in the FUS-DDIT3 interactome from Chapter 2 (Table 2.2).  No. WCL IP-MS Rank Protein name Gene Name Spliceosome class / family In database but unclassified Complex B Complex C hnRNP SR protein CPSF Other RNA binding domain Linked to spliceo-some Nuclear pore complex 1 1 Non-POU domain-containing octamer-binding protein NONO             Yes     2 2 Splicing factor, proline- and glutamine-rich SFPQ             Yes     3 3 Zinc finger RNA-binding protein ZFR           Yes       4 8 DBIRD complex subunit ZNF326 ZNF326           Yes       5 11 Protein virilizer homolog KIAA1429                 Yes 6 12 Guanine nucleotide-binding protein-like 3 GNL3                 Yes 7 14 RNA-binding protein FUS FUS     Yes             8 15 RNA-binding protein 14 RBM14           Yes       9 16 60S ribosomal protein L19 RPL19                 Yes 10 17 ATPase family AAA domain- ATAD3B                 Yes 214  containing protein 3B 11 18 Prelamin-A/C LMNA                 Yes 12 20 Heterogeneous nuclear ribonucleoprotein L HNRNPL     Yes             13 22 Transformer-2 protein homolog alpha TRA2A       Yes           14 23 Polymerase delta-interacting protein 3 POLDIP3                 Yes 15 29 Cleavage and polyadenylation specificity factor subunit 7 CPSF7         Yes         16 31 Poly [ADP-ribose] polymerase 1 PARP1                 Yes 17 33 Nuclear pore complex protein Nup153 NUP153               Yes   18 35 Death-inducer obliterator 1 DIDO1                 Yes 19 37 Heterogeneous nuclear ribonucleoprotein U-like 2 HNRNPUL2-BSCL2     Yes             20 41 Neuron navigator 1 NAV1 Yes                 21 45 Cleavage and polyadenylation specificity factor subunit 6 CPSF6         Yes         22 49 Pre-mRNA 3'-end-processing factor FIP1 FIP1L1                 Yes 23 51 Protein PRRC2A PRRC2A                 Yes 215  24 53 Cleavage and polyadenylation specificity factor subunit 1 CPSF1         Yes         25 55 Coatomer subunit gamma-1 COPG1                 Yes 26 56 Mitotic checkpoint protein BUB3 BUB3                 Yes 27 58 Luc7-like protein 3 LUC7L3                 Yes 28 69 Putative helicase MOV-10 MOV10                 Yes 29 72 WD repeat-containing protein 82 WDR82                 Yes 30 75 Probable ATP-dependent RNA helicase DDX20 DDX20                 Yes   Total 1 0 3 1 3 3 2 1 16   216  B.2 List of qPCR primers used in Chapter 3 Transcript Forward primer Reverse primer Gene expression validation CDKN1A CAGCAGAGGAAGACCATGTG GAGGCACAAGGGTACAAGACA CTGF CAGCATGGACGTTCGTCTG CCAACCACGGTTTGGTCCTT CYR61 GAGTGGGTCTGTGACGAGGAT GGTTGTATAGGATGCGAGGCT FOXN3 AAATGGAGCGCGGGTCCTGAG GCAGCTGGTGATGCCATTCCT FUS-DDIT3 (Type 1) TGAACCCAGAGGTCGTGGAG TGGTGCTGCTTTCAGGTGTG FUS-DDIT3 (Type 8) AGGCCCCTAAACCAGATGGC GGCTGCTTTGGTGCTGCTTT HPRT1 TGACACTGGCAAAACAATGCA GGTCCTTTTCACCAGCAAGTCT MYC GTCCTCGGATTCTCTGCTCT ACTCTGACCTTTTGCCAGGA SDHA TGGGAACAAGAGGGCATCTG CCACCACTGCATCAAATTCATG SOX11 CATGTAGACTAATGCAGCCATTGG CACGGAGCACGTGTCAATTG Alternative last exon validation FAT1 (proximal isoform) CGGGGAGGCAAAGTTGGTACT TGTCGGGAATCAGTCTAGCCG FAT1 (distal isoform) GGATGCCAAGCGTTCCTCTG TCGATGGCGTTTGGATCTGC FLT1 (proximal isoform) CGTCCCCTCCCTCCTCTCAT GCTGTGTGGTGCAGTCAGAG FLT1 (distal isoform) TGTTTGATGACTACCAGGGCGA ATCAGACAGCCCCGACTCCT PML (proximal isoform) GCTGAGGGAAGGAAGTGGGG AAGGCAGAAACACGGACCCA PML (distal isoform) TCCAAGGGGAAGCAGTGTGG TCTGACCGTGGAGGCCAATC   217  B.3 Number and percentage of global alternative splicing events Abbreviations: alternative 3’ splice sites (A3SS), alternative 5’ splice sites (A5SS), alternative first exon (AFE), alternative last exon (ALE), mutually exclusive exons (MXE), retained intron (RI), skipped exon (SE), or tandem untranslated region (TandemUTR).     24 hr 48 hr   Ctrl siRNA DDIT3 siRNA Ctrl siRNA DDIT3 siRNA  No. % No. % No. % No. % A3SS 6019 8.0% 5795 7.8% 5868 7.8% 5690 7.8% A5SS 10113 13.4% 9993 13.4% 10043 13.4% 9867 13.4% AFE 14119 18.7% 13989 18.7% 14149 18.8% 13669 18.6% ALE 7735 10.2% 7672 10.3% 7640 10.2% 7575 10.3% MXE 1925 2.5% 1888 2.5% 1900 2.5% 1862 2.5% RI 4529 6.0% 4501 6.0% 4513 6.0% 4397 6.0% SE 29040 38.4% 28669 38.4% 28861 38.4% 28111 38.3% Tandem UTR 2222 2.9% 2210 3.0% 2219 3.0% 2211 3.0% Total 75702 100.0% 74717 100.0% 75193 100.0% 73382 100.0%  218 B.4 Number and percentage of types of differential alternative splicing events between control and DDIT3 knockdown Abbreviations: alternative 3’ splice sites (A3SS), alternative 5’ splice sites (A5SS), alternative first exon (AFE), alternative last exon (ALE), mutually exclusive exons (MXE), retained intron (RI), skipped exon (SE), or tandem untranslated region (TandemUTR).     No. of events % of events AS event 24 hr 48 hr 24 hr 48 hr A3SS 5 5 6.58% 1.77% A5SS 4 9 5.26% 3.18% AFE 37 124 48.68% 43.82% ALE 9 94 11.84% 33.22% MXE 3 5 3.95% 1.77% RI 4 11 5.26% 3.89% SE 14 35 18.42% 12.37% Total 76 283 100.00% 100.00%    219 B.5 All differential alternative last exon events between control and FUS-DDIT3 knockdown at 24 hr or 48 hr The three events for FLT1, FAT1 and PML selected for validation are in bold. Event name description: first four digits are gene ID, the “@” sign serves as a separator between proximal (early) terminal exon isoforms followed by the distal (later) terminal exon isoforms. The Ψ value denotes relative proportion of the proximal over the distal isoforms. ∆Ψ denotes change in Ψ from FUS-DDIT3 knockdown to control. CI = confidence interval. Bayes factor denotes statistical significance of Ψ value between knockdown and control samples; the larger, the more significant.  Gene Condi-tion event_name DDIT3 siRNA Ψ DDIT3 siRNA CI (low) DDIT3 siRNA CI (high) Ctrl siRNA Ψ Ctrl siRNA CI (low) Ctrl siRNA CI (high) ∆Ψ Bayes factor MORF4L1 48 hr 10933@uc010bli.1@uc010blj.1 0.76 0.42 0.98 0.21 0.05 0.47 0.55 21.16 ATXN7 48 hr 6314@uc010hnu.1@uc010hnw.1 0.63 0.37 0.87 0.1 0.05 0.19 0.53 443.12 MLLT4 48 hr 4301@uc003qwf.2@uc003qwc.1 0.65 0.4 0.85 0.17 0.07 0.32 0.48 105918.84 MLLT4 48 hr 4301@uc003qwf.2@uc003qwg.1 0.68 0.42 0.89 0.21 0.08 0.36 0.47 34.14 UBAP2L 48 hr 9898@uc001feo.1@uc009wot.1 0.79 0.62 0.93 0.39 0.31 0.48 0.4 26374137 UBAP2L 48 hr 9898@uc001feo.1@uc001fep.2uc001fer.1 0.63 0.41 0.83 0.24 0.19 0.32 0.39 284.1 NA 48 hr 50809@uc001beb.1@uc001bea.1 0.49 0.35 0.63 0.23 0.17 0.3 0.26 34.15 HBS1L 48 hr 10767@uc003qfa.1@uc003qey.1uc003qez.1 0.73 0.66 0.79 0.49 0.43 0.54 0.24 1E+12 FLT1 48 hr 2321@uc001usc.2@uc010aap.1uc010aaq.1uc001usa.2uc001usb.2 0.39 0.32 0.46 0.16 0.13 0.2 0.23 1E+12 TRIM4 48 hr 89122@uc003usf.1@uc003usd.1uc003use.1 0.38 0.31 0.46 0.17 0.14 0.22 0.21 1.863E+09 FLT1 48 hr 2321@uc010aat.1@uc010aar.1 0.41 0.33 0.5 0.2 0.16 0.25 0.21 281.73 CEP350 48 hr 9857@uc001gnr.1@uc001gnt.1uc001gnv.1 0.61 0.5 0.71 0.4 0.34 0.47 0.21 39.32 NA 48 hr 400818@uc009wia.1uc2641.uc009wiq.1 0.53 0.44 0.61 0.33 0.28 0.38 0.2 40.35 EPRS 48 hr 2058@uc001hlz.1@uc001hly.1 0.55 0.5 0.61 0.36 0.32 0.41 0.19 912627.3 ASCC3 48 hr 10973@uc010kcv.1@uc003pqk.1 0.78 0.72 0.85 0.59 0.52 0.66 0.19 269.76  220 ABCC5 48 hr 10057@uc003fmh.1@uc003fmf.1uc010hxl.1uc003fmg.1 0.7 0.63 0.77 0.51 0.45 0.58 0.19 227.84 SPEG 48 hr 10290@uc010fwh.1uc002vlq.2@uc010fwg.1 0.54 0.46 0.63 0.35 0.29 0.41 0.19 37.31 PML 48 hr 5371@uc002aww.1@uc002awu.1uc002awv.1 0.69 0.64 0.74 0.51 0.46 0.56 0.18 20214299 ABCC5 48 hr 10057@uc010hxn.1@uc003fmf.1uc010hxl.1uc003fmg.1 0.68 0.61 0.75 0.5 0.44 0.56 0.18 840.22 KIF18A 48 hr 81930@uc001msd.2@uc001msc.2 0.74 0.68 0.8 0.56 0.5 0.63 0.18 67.95 HSPG2 48 hr 3339@uc009vqe.1@uc001bfi.1uc001bfj.1uc009vqd.1 0.59 0.55 0.63 0.44 0.41 0.48 0.15 2.068E+11 TDP1 48 hr 55775@uc001xyc.1@uc010atn.1 0.56 0.51 0.61 0.42 0.38 0.46 0.14 55.99 C11orf30 48 hr 56946@uc001oxo.1@uc001oxl.1uc001oxm.1uc001oxn.1uc001oxp.1 0.43 0.36 0.49 0.29 0.25 0.33 0.14 20.99 PRRC2C 48 hr 23215@uc001ghr.1@uc001ghs.1 0.53 0.5 0.57 0.4 0.37 0.42 0.13 1E+12 TMEM201 48 hr 199953@uc001apy.1@uc001apz.1 0.6 0.56 0.64 0.47 0.44 0.5 0.13 1E+12 ZFP106 48 hr 64397@uc001zpy.1@uc001zpu.1uc001zpv.1uc001zpw.1uc001zpx.1 0.61 0.58 0.64 0.48 0.45 0.5 0.13 1E+12 TJP2 48 hr 9414@uc004ahb.1@uc004ahe.1uc004ahf.1uc010mol.1 0.47 0.43 0.52 0.34 0.31 0.38 0.13 228800.9 FLT1 48 hr 2321@uc010aat.1@uc010aap.1uc010aaq.1uc001usa.2uc001usb.2 0.22 0.16 0.28 0.09 0.07 0.11 0.13 445.78 LRTOMT 48 hr 220074@uc001orr.1@uc001ors.2 0.84 0.8 0.88 0.71 0.66 0.76 0.13 310.09 WNK1 48 hr 65125@uc001qin.2@uc001qio.2uc001qip.2uc001qir.2 0.93 0.88 0.97 0.8 0.74 0.86 0.13 120.24 TDP1 48 hr 55775@uc001xyc.1@uc001xxy.1uc001xxz.1uc001xya.1uc010ato.1 0.5 0.45 0.56 0.37 0.33 0.41 0.13 66.07 SMG6 48 hr 23293@uc002fud.1@uc002fub.1 0.62 0.59 0.66 0.5 0.48 0.53 0.12 396538.48 TMEM209 48 hr 84928@uc003vpo.2@uc003vpn.2uc010lmc.1 0.9 0.86 0.94 0.78 0.73 0.83 0.12 42.14 CRISPLD2 48 hr 83716@uc002fik.2@uc002fip.1 0.92 0.89 0.94 0.81 0.76 0.86 0.11 200.62 ODZ3 48 hr 55714@uc003ive.1@uc003ivd.1 0.48 0.43 0.53 0.37 0.33 0.4 0.11 134.11  221 ADAR 48 hr 103@uc001ffl.1@uc001ffh.1uc001ffi.1uc001ffj.1uc001ffk.1 0.55 0.53 0.57 0.45 0.43 0.46 0.1 1E+12 NA 48 hr 3106@uc003ntg.1uc003nth.2@uc003ntf.2 0.53 0.51 0.55 0.43 0.41 0.45 0.1 1E+12 TJP2 48 hr 9414@uc004ahb.1@uc010mom.1 0.27 0.24 0.3 0.17 0.15 0.19 0.1 2861.76 MLLT4 48 hr 4301@uc003qwc.1@uc003qwd.1 0.14 0.09 0.21 0.04 0.02 0.06 0.1 62.1 DST 48 hr 667@uc003pdb.2@uc003pcv.2uc003pcw.2uc003pcx.2uc003pcy.2uc003pcz.2 0.09 0.07 0.11 0.19 0.17 0.22 -0.1 1E+12 SEC14L1 48 hr 6397@uc002jto.1@uc010dhc.1uc002jtp.1 0.28 0.26 0.3 0.38 0.36 0.41 -0.1 1E+12 SEC14L1 48 hr 6397@uc002jtm.1@uc010dhc.1uc002jtp.1 0.25 0.23 0.27 0.35 0.32 0.37 -0.1 1.036E+10 SEC14L1 48 hr 6397@uc002jtr.1@uc010dhc.1uc002jtp.1 0.23 0.21 0.26 0.33 0.3 0.35 -0.1 12890922 RGPD5,RGPD6 48 hr 729540@uc002tfy.1@uc002tfx.2 0.58 0.54 0.62 0.68 0.65 0.7 -0.1 5327.62 NA 48 hr 79027@uc003urc.1uc0561.uc010lgd.1 0.31 0.27 0.34 0.41 0.38 0.44 -0.1 1833.47 TNRC18 48 hr 84629@uc010ksx.1@uc003soi.2 0.14 0.11 0.17 0.24 0.21 0.27 -0.1 364.05 RGPD6,RGPD5 48 hr 729540@uc002tfe.1@uc002tfd.2uc010fjp.1 0.58 0.55 0.62 0.68 0.65 0.71 -0.1 189.25 SRGAP2 48 hr 23380@uc009xbt.1@uc001hdx.1 0.61 0.56 0.65 0.71 0.68 0.75 -0.1 62.68 MAPK8IP3 48 hr 23162@uc002cmi.1@uc002cmk.1uc002cml.1 0.2 0.16 0.24 0.3 0.26 0.35 -0.1 21.23 NA 48 hr 54856@uc001fmc.1uc03116.uc001fma.1 0.19 0.16 0.22 0.3 0.27 0.32 -0.11 6.995E+11 NA 48 hr 672@uc002ide.1uc0102528.uc002idb.1 0.35 0.32 0.38 0.46 0.43 0.49 -0.11 1.163E+11 AEN 48 hr 64782@uc010bnl.1@uc010bnm.1 0.23 0.2 0.26 0.34 0.31 0.37 -0.11 297243680 ZFP64 24 hr 55734@uc002xwl.1uc002xwm.1uc002xwn.1@uc002xwj.1uc002xwk.1 0.28 0.25 0.31 0.39 0.36 0.43 -0.11 876998.92  222 FBLN1 24 hr 2192@uc003bgh.1uc010gzz.1@uc003bgj.1 0.24 0.21 0.28 0.35 0.31 0.38 -0.11 1542.52 C1orf86 48 hr 199990@uc001aiy.1@uc001aix.1 0.16 0.13 0.19 0.27 0.22 0.32 -0.11 61.43 NA 48 hr 672@uc002idd.2@uc0022533.uc002idb.1 0.32 0.3 0.35 0.44 0.41 0.46 -0.12 7.918E+09 ZZZ3 48 hr 26009@uc009wbz.1@uc001dhp.1uc001dhq.1uc001dhr.1 0.59 0.54 0.63 0.71 0.67 0.74 -0.12 1345.61 SRSF6 48 hr 6431@uc002xki.1@uc002xkk.1 0.48 0.44 0.52 0.6 0.56 0.63 -0.12 237.95 TDP1 24 hr 55775@uc001xyc.1@uc001xyd.1 0.87 0.8 0.93 0.99 0.96 1 -0.12 38.86 ASCC3 48 hr 10973@uc003pqm.1@uc003pqk.1 0.07 0.05 0.1 0.2 0.16 0.24 -0.13 1E+12 PRDM2 48 hr 7799@uc009vod.1@uc001avh.1uc001avk.1 0.48 0.42 0.54 0.61 0.57 0.66 -0.13 40.81 CSNK1E 48 hr 1454@uc003avp.1@uc003avj.1uc003avk.1 0.63 0.57 0.69 0.76 0.72 0.81 -0.13 27.13 SETD2 48 hr 29072@uc003cqu.1uc003cqv.1@uc003cqr.1uc003cqs.1 0.29 0.26 0.32 0.43 0.39 0.46 -0.14 1E+12 ST5 48 hr 6764@uc001mgw.1@uc001mgt.1uc009yfr.1uc001mgu.1uc001mgv.1 0.2 0.17 0.23 0.34 0.31 0.37 -0.14 1E+12 NUP214 48 hr 8021@uc010mzh.1uc010mzi.1@uc004cag.1uc004cah.1uc004cai.1 0.17 0.14 0.2 0.31 0.26 0.36 -0.14 95440.87 ZC3HAV1 48 hr 56829@uc003vuo.1uc003vup.1@uc003vun.1 0.57 0.52 0.63 0.71 0.66 0.75 -0.14 66.64 BOD1L 48 hr 259282@uc010idr.1@uc003gmz.1 0.5 0.45 0.56 0.64 0.6 0.68 -0.14 60.21 TDP1 24 hr 55775@uc001xyd.1@uc001xxy.1uc001xxz.1uc001xya.1uc010ato.1 0.84 0.76 0.92 0.98 0.93 1 -0.14 57.4 LRRFIP1 48 hr 9208@uc002vxd.1uc002vxe.1uc002vxf.1@uc002vxc.1 0.74 0.7 0.78 0.89 0.87 0.91 -0.15 1E+12 NASP 48 hr 4678@uc001cok.1@uc001coi.1uc001coh.1uc001coj.1uc001col.1 0.49 0.45 0.53 0.64 0.6 0.67 -0.15 1E+12 MDC1 48 hr 9656@uc003rdu.1uc003rdv.1@uc003rds.2uc003rdt.2 0.36 0.32 0.4 0.51 0.47 0.54 -0.15 2.33E+09 MDC1 48 hr 9656@uc003rvx.1uc003rvy.1@uc003rvv.2uc003rvw.2 0.36 0.32 0.4 0.51 0.47 0.54 -0.15 138546970  223 ABCC5 48 hr 10057@uc010hxo.1@uc003fmf.1uc010hxl.1uc003fmg.1 0.22 0.18 0.27 0.37 0.32 0.42 -0.15 221590.47 MDC1 48 hr 9656@uc003nrh.1uc003nri.1@uc003nrf.2uc003nrg.2 0.36 0.32 0.4 0.51 0.47 0.54 -0.15 2531.45 USP34 24 hr 9736@uc002sbf.2@uc002sbd.1uc002sbe.1 0.79 0.72 0.86 0.94 0.9 0.98 -0.15 51.99 TNRC18 48 hr 84629@uc010ksx.1@uc003soj.2 0.25 0.19 0.32 0.4 0.35 0.46 -0.15 35.48 SPEG 48 hr 10290@uc002vln.1uc002vlp.1@uc010fwg.1 0.24 0.19 0.3 0.39 0.33 0.45 -0.15 21.5 FLT1 24 hr 2321@uc010aat.1@uc010aar.1 0.16 0.11 0.22 0.32 0.25 0.4 -0.16 44.94 ARSJ 48 hr 79642@uc010imv.1@uc010imu.1 0.36 0.31 0.43 0.52 0.45 0.59 -0.16 29.04 PRKAG2 24 hr 51422@uc003wkm.1@uc003wkl.1 0.7 0.62 0.77 0.86 0.81 0.92 -0.16 23.21 ASH1L 48 hr 55870@uc009wqr.1@uc001fkt.1uc009wqq.1 0.34 0.3 0.38 0.51 0.48 0.54 -0.17 1E+12 FAT1 48 hr 2195@uc010iso.1@uc010isn.1uc003ize.1uc003izf.1 0.24 0.22 0.26 0.41 0.39 0.43 -0.17 1E+12 ANKRD11 48 hr 29123@uc002fnf.1uc002fng.1@uc002fna.1 0.47 0.43 0.52 0.64 0.6 0.68 -0.17 9.285E+11 HEATR7A 48 hr 727957@uc003zbh.2uc003zbi.2@uc003zbj.1 0.07 0.04 0.11 0.24 0.15 0.35 -0.17 186.27 ANKRD11 48 hr 29123@uc002fne.1@uc002fna.1 0.4 0.33 0.47 0.58 0.5 0.67 -0.18 28.68 PML 48 hr 5371@uc002aww.1@uc002awq.1uc002awt.1 0.17 0.12 0.23 0.35 0.26 0.46 -0.18 28.43 NCOA6 48 hr 23054@uc010gew.1@uc002xav.1uc002xaw.1 0.3 0.26 0.34 0.49 0.45 0.53 -0.19 1E+12 KLHDC4 48 hr 54758@uc010chu.1@uc002fkk.2uc002fkl.2 0.55 0.45 0.64 0.75 0.68 0.83 -0.2 34.77 GOLGB1 48 hr 2804@uc010hrd.1@uc003eei.2uc003eej.2uc010hrc.1 0.47 0.43 0.52 0.68 0.64 0.72 -0.21 1E+12 TRAPPC12 48 hr 51112@uc002qxl.1@uc002qxm.1uc002qxn.1 0.36 0.3 0.43 0.58 0.5 0.65 -0.22 290.07 SRSF6 48 hr 6431@uc002xkj.1@uc002xkk.1 0.41 0.35 0.48 0.63 0.54 0.73 -0.22 91.71  224 PML 48 hr 5371@uc002aww.1@uc002awr.1uc002aws.1 0.17 0.12 0.23 0.4 0.29 0.54 -0.23 52.03 KLHDC4 48 hr 54758@uc010chu.1@uc010cht.1 0.35 0.27 0.44 0.59 0.49 0.69 -0.24 52.08 FGFR1OP2 48 hr 26127@uc001rhl.2@uc001rhm.1uc001rhn.1 0.21 0.14 0.31 0.45 0.34 0.58 -0.24 22.4 DPY19L4 48 hr 286148@uc003ygy.2@uc003ygx.2 0.42 0.33 0.51 0.68 0.6 0.77 -0.26 158.45 TMEM63B 24 hr 55362@uc003owq.1@uc003owr.1uc003ows.1 0.55 0.44 0.67 0.81 0.72 0.89 -0.26 35.41 FLT1 48 hr 2321@uc001usc.2@uc010aar.1 0.39 0.33 0.45 0.67 0.61 0.72 -0.28 1E+12 WDR27 48 hr 253769@uc010kkx.1@uc003qwz.1 0.31 0.18 0.45 0.66 0.47 0.83 -0.35 43.01 TSPAN10 24 hr 83882@uc002kaw.1@uc010die.1 0.12 0.05 0.2 0.49 0.33 0.69 -0.37 121174.57 ARHGAP28 48 hr 79822@uc002knc.1uc002knd.1uc002knf.1@uc002kne.1 0.46 0.29 0.67 0.84 0.66 0.96 -0.38 27.73 AKAP9 48 hr 10142@uc003ule.1@uc003ulj.2 0.08 0.02 0.19 0.48 0.27 0.71 -0.4 203.45 CHD2 48 hr 1106@uc002bsn.1@uc002bso.1 0.18 0.08 0.32 0.61 0.34 0.83 -0.43 26.22 ZMIZ1 48 hr 57178@uc001kae.2@uc001kah.1 0.23 0.11 0.4 0.75 0.51 0.94 -0.52 62.43    225 Appendix C   C.1 List of qPCR primers used in Chapter 4 Transcript Forward primer Reverse primer Upstream of PTX3 CCCCTGCACCTTCTATTCGGA TCATTTCATTGTTCAGGAGGACGA PTX3 promoter CCCACCAAATTCAGGGGAACT GCATTGCTGGAGAGACGCAAA Downstream of PTX3 promoter AGAATGGCTGCTGTGTGGGT ACTCTGCTCCTCCGGTCTCT    226 C.2 FUS-DDIT3 protein degrades over time during immunoprecipitation (A) Western blot analysis of fourth set of replicate for FUS-DDIT3 immunoprecipitation. Single channel images of FUS and DDIT3 blots from Figure 4.1B are shown here, with the DDIT3 blot clearly showing an increase of the second, smaller FUS-DDIT3 band in the IP compared to the input. (B) Immunodepletion experiment with mouse IgG control. Each successive IP was performed for 24 hr, and the recovered post-IP supernatant was used for the next IP. Western blot of DDIT3 shows gradual degradation of FUS-DDIT3, i.e. increased expression of smaller FUS-DDIT3 bands over 3 days of immunodepletion experiment. FUS-DDIT3 bands are arrowed in blue.   IB: FUS + DDIT33%InputMoIgG DDIT3IP402-91nuclear extractIB: DDIT3IB: FUS1°moIgG2°moIgG3°moIgG1°moIgG2°moIgG3°moIgGIP Post-IP supernatantInput InputIB: DDIT3AB 227 C.3 DDIT3 peptides identified from the quality control analysis of unlabeled peptide fractions with the Velos Orbitrap mass spectrometer Accession Gene Sequence Relative abundance Mo-IgG (1) Mo-IgG (1) Mo-IgG (1) DDIT3 (1) DDIT3 (2) DDIT3 (3) P35638 DDIT3 MVNLHQA NA NA NA 90,084,964 90,083,291 88,266,608 P35638 DDIT3 VAQLAEENER 4,463,917 1,744,011 NA 156,548,618 175,596,565 148,448,247 P35638 DDIT3 VAQLAEENERLK NA NA NA 33,039,520 6,392,745 24,068,780      228 C.4 List of 174 FUS-DDIT3-interacting proteins identified in nuclear lysates of 402-91 myxoid liposarcoma cells Relative enrichment of protein in triplicate immunoprecipitations of FUS-DDIT3 over IgG control is expressed as a ratio based on relative quantification of tandem mass tag labels. CV(T) = transformed coefficient of variation to be on the same scale as average ratio of relative enrichment. PES = protein enrichment score that takes into account both the ratio of relative enrichment of a protein and the CV of this ratio across triplicates. PES ranges from 0-1, with a higher score indicating higher enrichment of the protein in the FUS-DDIT3 interactome. PES z-score = number of standard deviations away from mean PES.  No. Gene Description No. of peptides Relative enrichment (ratio over IgG control) CV(T)  log2 Average log2 CV(T) PES PES z-score DDIT3 (1) DDIT3 (2) DDIT3 (3) Ave-rage 1 CEBPB CCAAT/enhancer-binding protein beta  4 8.94 8.58 9.82 9.11 1.54 3.19 0.62 1 5.7 2 CALD1 Caldesmon  2 5.43 6.55 7.43 6.47 3.4 2.69 1.76 0.71 2.47 3 SFPQ Splicing factor, proline- and glutamine-rich  22 3.97 4.3 3.91 4.06 1.14 2.02 0.19 0.7 2.36 4 NONO Non-POU domain-containing octamer-binding protein  20 3.8 4.6 4.35 4.25 2.1 2.09 1.07 0.65 1.82 5 RBM14 RNA-binding protein 14  8 3.91 4.86 4.68 4.49 2.47 2.17 1.31 0.65 1.71 6 PSPC1 Paraspeckle component 1  6 3.12 3.45 3.68 3.42 1.8 1.77 0.85 0.62 1.43 7 HNRNPM Heterogeneous nuclear ribonucleoprotein M  15 2.61 2.95 2.77 2.78 1.35 1.47 0.43 0.61 1.3 8 SIN3B Paired amphipathic helix protein Sin3b  3 2.31 2.53 2.47 2.44 1 1.28 0.01 0.61 1.3 9 RBMX RNA-binding motif protein, X chromosome  5 1.78 1.83 1.85 1.82 0.43 0.86 -1.21 0.61 1.27 10 PHF14 PHD finger protein 14  3 1.49 1.5 1.51 1.5 0.13 0.58 -2.98 0.6 1.25 11 FUS RNA-binding protein FUS  2 1.89 1.86 1.96 1.9 0.57 0.93 -0.82 0.6 1.23 12 HNRNPF Heterogeneous nuclear ribonucleoprotein F  4 1.97 1.85 1.97 1.93 0.82 0.95 -0.29 0.59 1.07  229 13 YLPM1 YLP motif-containing protein 1  12 2.62 3.06 2.99 2.89 1.83 1.53 0.87 0.58 1.03 14 ILF3 Interleukin enhancer-binding factor 3  7 2.46 2.84 2.8 2.7 1.71 1.43 0.78 0.58 0.98 15 SPAG5 Sperm-associated antigen 5  6 2.99 3.37 3.74 3.37 2.43 1.75 1.28 0.58 0.94 16 WDR33 pre-mRNA 3' end processing protein WDR33  2 1.3 1.35 1.31 1.32 0.43 0.4 -1.23 0.57 0.91 17 ZNF281 Zinc finger protein 281  2 1.29 1.33 1.34 1.32 0.45 0.4 -1.17 0.57 0.9 18 AKAP8 A-kinase anchor protein 8  2 2.03 2.28 2.12 2.14 1.28 1.1 0.36 0.57 0.89 19 RPL13 60S ribosomal protein L13  2 1.09 1.06 1.08 1.08 0.26 0.11 -1.96 0.57 0.86 20 TPX2 Targeting protein for Xklp2  3 1.64 1.71 1.79 1.71 0.93 0.78 -0.1 0.57 0.83 21 YEATS2 YEATS domain-containing protein 2  3 1.64 1.73 1.79 1.72 0.96 0.78 -0.06 0.57 0.82 22 NCOA5 Nuclear receptor coactivator 5  9 1.38 1.44 1.36 1.39 0.64 0.48 -0.65 0.57 0.81 23 HNRNPUL2 Heterogeneous nuclear ribonucleoprotein U-like protein 2  3 1.73 1.85 1.91 1.83 1.08 0.87 0.11 0.56 0.81 24 SYNCRIP Heterogeneous nuclear ribonucleoprotein Q  3 2.02 2.32 2.21 2.18 1.54 1.12 0.62 0.56 0.73 25 TRIM21 E3 ubiquitin-protein ligase TRIM21  2 1.73 1.84 1.94 1.84 1.24 0.88 0.31 0.56 0.7 26 HNRNPA3 Heterogeneous nuclear ribonucleoprotein A3  6 2.18 2.59 2.46 2.41 1.89 1.27 0.92 0.55 0.65 27 ZZZ3 ZZ-type zinc finger-containing protein 3  2 1.6 1.77 1.77 1.71 1.23 0.78 0.29 0.55 0.62 28 KDM1A Lysine-specific histone demethylase 1A  2 1.23 1.28 1.32 1.28 0.8 0.35 -0.33 0.55 0.61 29 SMARCD1 SWI/SNF-related matrix-associated actin-dependent regulator of chromatin subfamily D member 1  2 1.16 1.23 1.22 1.2 0.73 0.27 -0.46 0.55 0.61 30 STAU1 Double-stranded RNA-binding protein Staufen homolog 1  2 1.66 1.76 1.87 1.76 1.3 0.82 0.38 0.55 0.6 31 SF1 Splicing factor 1  2 1.11 1.18 1.18 1.16 0.7 0.21 -0.51 0.55 0.6  230 32 SUZ12 Polycomb protein SUZ12  2 1.14 1.11 1.18 1.14 0.69 0.19 -0.53 0.55 0.6 33 CBX5 Chromobox protein homolog 5  2 1.03 1.09 1.07 1.06 0.6 0.09 -0.73 0.55 0.6 34 HNRNPR Heterogeneous nuclear ribonucleoprotein R  6 1.91 2.19 2.18 2.09 1.65 1.07 0.72 0.55 0.59 35 SRSF9 Serine/arginine-rich splicing factor 9  2 2.14 2.46 2.53 2.38 1.96 1.25 0.97 0.54 0.57 36 AURKA Aurora kinase A  3 1.35 1.42 1.48 1.42 1 0.5 0.01 0.54 0.57 37 FAM98A Protein FAM98A  2 1.31 1.43 1.36 1.37 0.96 0.45 -0.06 0.54 0.57 38 ADAR Double-stranded RNA-specific adenosine deaminase  10 1.81 2.08 1.98 1.96 1.57 0.97 0.65 0.54 0.55 39 DNAJC7 DnaJ homolog subfamily C member 7  3 1.01 1.07 1.06 1.05 0.68 0.07 -0.56 0.54 0.54 40 THOC1 THO complex subunit 1  2 1.04 1.01 0.98 1.01 0.63 0.01 -0.66 0.54 0.54 41 RNF2 E3 ubiquitin-protein ligase RING2  3 1.53 1.65 1.72 1.63 1.27 0.71 0.35 0.54 0.53 42 SNRPB Small nuclear ribonucleoprotein-associated proteins B and B'  4 1.23 1.32 1.33 1.3 0.97 0.37 -0.05 0.54 0.51 43 NCL Nucleolin  13 1.04 1.11 1.09 1.08 0.76 0.11 -0.4 0.54 0.51 44 HNRNPD Heterogeneous nuclear ribonucleoprotein D0  4 1.73 1.99 1.92 1.88 1.57 0.91 0.65 0.54 0.49 45 DDX17 Probable ATP-dependent RNA helicase DDX17  11 1.71 1.92 1.95 1.86 1.58 0.89 0.66 0.54 0.47 46 GATAD1 GATA zinc finger domain-containing protein 1  3 1.49 1.59 1.42 1.5 1.24 0.58 0.31 0.53 0.46 47 PHC1 Polyhomeotic-like protein 1  2 1.98 2.31 2.04 2.11 1.88 1.08 0.91 0.53 0.44 48 FIP1L1 Pre-mRNA 3'-end-processing factor FIP1  2 1.44 1.47 1.61 1.51 1.28 0.59 0.36 0.53 0.44 49 SRSF1 Serine/arginine-rich splicing factor 1  2 1.27 1.33 1.4 1.33 1.11 0.42 0.15 0.53 0.44 50 TJP2 Tight junction protein ZO-2  3 0.99 1.04 1.06 1.03 0.8 0.04 -0.32 0.53 0.44  231 51 MATR3 Matrin-3  7 2.1 2.55 2.31 2.32 2.1 1.22 1.07 0.53 0.43 52 ZNF217 Zinc finger protein 217  3 1.85 2.05 2.19 2.03 1.82 1.02 0.87 0.53 0.42 53 WDR5 WD repeat-containing protein 5  4 1.23 1.28 1.35 1.29 1.1 0.36 0.14 0.53 0.41 54 HCFC1 Host cell factor 1  2 1.13 1.21 1.12 1.15 0.97 0.21 -0.05 0.53 0.41 55 HNRNPL Heterogeneous nuclear ribonucleoprotein L  5 1.69 1.95 1.78 1.81 1.63 0.85 0.71 0.53 0.4 56 RPL5 60S ribosomal protein L5  2 1.1 1.18 1.2 1.16 0.99 0.22 -0.01 0.53 0.39 57 SRSF7 Serine/arginine-rich splicing factor 7  5 1.37 1.37 1.51 1.42 1.29 0.5 0.36 0.53 0.37 58 RPS12 40S ribosomal protein S12  3 1.16 1.15 1.07 1.13 0.99 0.17 -0.02 0.53 0.37 59 MARK3 MAP/microtubule affinity-regulating kinase 3  3 1.25 1.39 1.36 1.34 1.21 0.42 0.28 0.52 0.36 60 DHX9 ATP-dependent RNA helicase A  20 2.23 2.78 2.58 2.53 2.42 1.34 1.28 0.52 0.35 61 RBM6 RNA-binding protein 6  5 1.9 2.21 2.26 2.12 2.02 1.09 1.01 0.52 0.35 62 KHDRBS1 KH domain-containing, RNA-binding, signal transduction-associated protein 1  6 1.49 1.71 1.6 1.6 1.5 0.68 0.58 0.52 0.35 63 ZNF512B Zinc finger protein 512B  5 1.19 1.28 1.32 1.26 1.15 0.34 0.2 0.52 0.35 64 CEP170 Centrosomal protein of 170 kDa  5 1.34 1.46 1.51 1.44 1.35 0.52 0.43 0.52 0.33 65 POLDIP3 Polymerase delta-interacting protein 3  14 1.3 1.46 1.45 1.4 1.34 0.49 0.42 0.52 0.32 66 HNRNPC Heterogeneous nuclear ribonucleoproteins C1/C2  2 1.62 1.9 1.8 1.77 1.77 0.83 0.82 0.52 0.28 67 CTCF Transcriptional repressor CTCF  4 0.98 1.07 1.02 1.03 1.02 0.04 0.03 0.52 0.28 68 HNRNPA1 Heterogeneous nuclear ribonucleoprotein A1  3 2.41 2.58 3.04 2.68 2.68 1.42 1.42 0.52 0.27 69 TRA2B Transformer-2 protein homolog beta  8 1.49 1.68 1.72 1.63 1.63 0.71 0.71 0.52 0.27  232 70 CHD4 Chromodomain-helicase-DNA-binding protein 4  2 1.07 1.15 1.04 1.09 1.09 0.12 0.12 0.52 0.27 71 RBMXL1 RNA binding motif protein, X-linked-like-1  3 1.65 1.85 1.96 1.82 1.86 0.86 0.9 0.51 0.24 72 SRSF10 Serine/arginine-rich splicing factor 10  6 1.47 1.65 1.71 1.61 1.68 0.69 0.75 0.51 0.22 73 CIT Citron Rho-interacting kinase  5 1.09 1.17 1.22 1.16 1.23 0.21 0.3 0.51 0.22 74 IMMT MICOS complex subunit MIC60  2 1.04 1.13 1.03 1.07 1.15 0.09 0.21 0.51 0.21 75 ACIN1 Apoptotic chromatin condensation inducer in the nucleus  3 1.05 1.16 1.15 1.12 1.24 0.16 0.31 0.51 0.19 76 HIRA Protein HIRA  6 2 2.45 2.13 2.19 2.32 1.13 1.21 0.51 0.18 77 PHC2 Polyhomeotic-like protein 2  4 1.36 1.53 1.57 1.49 1.61 0.57 0.69 0.51 0.18 78 SLTM SAFB-like transcription modulator  5 1.21 1.37 1.34 1.31 1.44 0.39 0.53 0.51 0.18 79 POLR2A DNA-directed RNA polymerase II subunit RPB1  2 1.21 1.37 1.34 1.31 1.43 0.39 0.52 0.51 0.18 80 MTCL1 Microtubule cross-linking factor 1  4 1.19 1.21 1.34 1.25 1.38 0.32 0.47 0.51 0.18 81 MTA1 Metastasis-associated protein MTA1  5 1.03 1.15 1.1 1.09 1.23 0.13 0.29 0.51 0.18 82 CIRBP Cold-inducible RNA-binding protein  3 1.66 1.91 1.98 1.85 1.99 0.89 0.99 0.51 0.17 83 U2AF1 Splicing factor U2AF 35 kDa subunit  3 0.97 1.08 1.02 1.02 1.19 0.03 0.25 0.51 0.15 84 CABIN1 Calcineurin-binding protein cabin-1  2 1.72 1.99 2.07 1.93 2.11 0.95 1.07 0.51 0.14 85 CBX4 E3 SUMO-protein ligase CBX4  5 1.62 1.88 1.94 1.81 2.04 0.86 1.03 0.5 0.11 86 SPEN Msx2-interacting protein  2 0.99 1.11 1.08 1.06 1.28 0.08 0.36 0.5 0.11 87 HADHB Trifunctional enzyme subunit beta, mitochondrial  2 0.95 1.05 1.04 1.01 1.23 0.01 0.3 0.5 0.11 88 ZNF638 Zinc finger protein 638  14 1.6 1.88 1.9 1.8 2.04 0.85 1.03 0.5 0.1  233 89 KIF20B Kinesin-like protein KIF20B  25 1.16 1.27 1.33 1.25 1.53 0.32 0.62 0.5 0.07 90 CAPRIN1 Caprin-1  2 1.09 1.18 1.24 1.17 1.45 0.23 0.54 0.5 0.07 91 HSPA9 Stress-70 protein, mitochondrial  7 0.94 1.05 1.03 1.01 1.32 0.01 0.4 0.5 0.05 92 ESCO2 N-acetyltransferase ESCO2  7 1.01 1.14 1.13 1.09 1.41 0.13 0.5 0.5 0.04 93 C1QBP Complement component 1 Q subcomponent-binding protein, mitochondrial  2 1.06 1.07 1.19 1.11 1.46 0.15 0.54 0.5 0.02 94 WTAP Pre-mRNA-splicing regulator WTAP  3 1.24 1.4 1.45 1.36 1.79 0.45 0.84 0.49 -0.03 95 MARK2 Serine/threonine-protein kinase MARK2  2 1.26 1.48 1.41 1.38 1.82 0.47 0.86 0.49 -0.04 96 THAP11 THAP domain-containing protein 11  4 1.04 1.05 1.18 1.09 1.52 0.12 0.61 0.49 -0.04 97 TRA2A Transformer-2 protein homolog alpha  3 1.37 1.6 1.62 1.53 1.99 0.61 0.99 0.49 -0.06 98 PLEKHA5 Pleckstrin homology domain-containing family A member 5  4 1.16 1.34 1.32 1.27 1.74 0.35 0.8 0.49 -0.06 99 DIDO1 Death-inducer obliterator 1  4 1.07 1.22 1.21 1.17 1.63 0.22 0.7 0.49 -0.06 100 H1FX Histone H1x  6 0.99 1.09 1.14 1.07 1.55 0.1 0.63 0.49 -0.07 101 RPS19 40S ribosomal protein S19  2 1.2 1.12 1.31 1.21 1.7 0.27 0.76 0.49 -0.08 102 ADNP Activity-dependent neuroprotector homeobox protein  4 1.02 1.18 1.12 1.11 1.59 0.15 0.67 0.49 -0.08 103 ASH2L Set1/Ash2 histone methyltransferase complex subunit ASH2  2 0.99 1.07 1.15 1.07 1.57 0.1 0.65 0.49 -0.09 104 NUMA1 Nuclear mitotic apparatus protein 1  46 1.06 1.18 1.24 1.16 1.71 0.21 0.77 0.48 -0.12 105 SNW1 SNW domain-containing protein 1  2 0.95 0.98 1.08 1 1.55 0.01 0.63 0.48 -0.12 106 MAP4 Microtubule-associated protein 4  23 1.26 1.49 1.46 1.4 1.96 0.49 0.97 0.48 -0.13 107 NUSAP1 Nucleolar and spindle-associated protein 1  11 0.98 1.13 1.04 1.05 1.62 0.07 0.69 0.48 -0.13  234 108 SUPT5H Transcription elongation factor SPT5  2 0.99 1.15 1.06 1.07 1.62 0.09 0.7 0.48 -0.13 109 MAP7D1 MAP7 domain-containing protein 1  3 0.97 1.05 1.12 1.05 1.61 0.07 0.69 0.48 -0.13 110 SNRPD3 Small nuclear ribonucleoprotein Sm D3  4 1.27 1.47 1.51 1.42 2 0.5 1 0.48 -0.15 111 KDM5A Lysine-specific demethylase 5A  4 1.51 1.81 1.85 1.72 2.33 0.79 1.22 0.48 -0.16 112 DDX5 Probable ATP-dependent RNA helicase DDX5  7 1.53 1.85 1.58 1.65 2.27 0.73 1.18 0.48 -0.17 113 CHTOP Chromatin target of PRMT1 protein  8 0.97 1.04 1.13 1.04 1.67 0.06 0.74 0.48 -0.17 114 POM121 Nuclear envelope pore membrane protein POM 121  2 1.09 1.25 1.28 1.2 1.87 0.27 0.91 0.47 -0.21 115 SCAF11 Protein SCAF11  2 1.02 1.07 1.19 1.09 1.76 0.13 0.82 0.47 -0.21 116 PRC1 Protein regulator of cytokinesis 1  12 0.96 1.11 1.09 1.05 1.74 0.08 0.8 0.47 -0.22 117 RBM12B RNA-binding protein 12B  5 1.65 2.09 2 1.91 2.67 0.94 1.42 0.47 -0.27 118 TADA3 Transcriptional adapter 3  5 0.96 1.02 1.14 1.04 1.86 0.06 0.89 0.47 -0.31 119 SNRPD2 Small nuclear ribonucleoprotein Sm D2  3 1.12 1.29 1.35 1.25 2.09 0.33 1.07 0.46 -0.33 120 RPS8 40S ribosomal protein S8  8 0.98 1.03 1.15 1.05 1.89 0.07 0.92 0.46 -0.33 121 HNRNPA0 Heterogeneous nuclear ribonucleoprotein A0  2 1.52 1.52 1.84 1.63 2.49 0.7 1.31 0.46 -0.34 122 SF3A1 Splicing factor 3A subunit 1  3 0.99 1.18 1.11 1.09 1.95 0.13 0.96 0.46 -0.34 123 KIF18B Kinesin-like protein KIF18B  2 1.04 1.24 1.23 1.17 2.07 0.23 1.05 0.46 -0.37 124 ATAD3B ATPase family AAA domain-containing protein 3B  4 0.98 1.02 1.16 1.05 1.95 0.07 0.96 0.46 -0.37 125 ERH Enhancer of rudimentary homolog  4 0.99 1.18 1.12 1.1 2.01 0.13 1.01 0.46 -0.38 126 EHMT1 Histone-lysine N-methyltransferase EHMT1  2 0.98 1.15 1.16 1.1 2.02 0.14 1.01 0.46 -0.39  235 127 MCM3 DNA replication licensing factor MCM3  2 0.93 1.12 1.03 1.02 1.96 0.03 0.97 0.46 -0.4 128 KMT2A Histone-lysine N-methyltransferase 2A  4 1.02 1.15 1.24 1.14 2.08 0.18 1.06 0.46 -0.41 129 HNRNPH1 Heterogeneous nuclear ribonucleoprotein H  10 1.45 1.85 1.7 1.67 2.65 0.74 1.4 0.45 -0.43 130 CPNE1 Copine-1  4 1.07 1.28 1.29 1.21 2.21 0.28 1.15 0.45 -0.44 131 THOC5 THO complex subunit 5 homolog  3 0.99 1.19 1.14 1.11 2.11 0.15 1.07 0.45 -0.44 132 U2AF2 Splicing factor U2AF 65 kDa subunit  6 0.96 1.14 1.14 1.08 2.08 0.11 1.06 0.45 -0.45 133 HNRNPK Heterogeneous nuclear ribonucleoprotein K  8 1.41 1.77 1.73 1.63 2.66 0.71 1.41 0.45 -0.46 134 SLC25A5 ADP/ATP translocase 2  3 1.27 1.59 1.41 1.42 2.47 0.51 1.3 0.45 -0.48 135 SF3B2 Splicing factor 3B subunit 2  3 1.02 1.18 0.99 1.06 2.11 0.09 1.08 0.45 -0.48 136 SLC25A6 ADP/ATP translocase 3  2 1.13 1.33 1.39 1.28 2.35 0.36 1.23 0.45 -0.49 137 POLR2D DNA-directed RNA polymerase II subunit RPB4  2 1.1 1.26 1.36 1.24 2.31 0.31 1.21 0.45 -0.5 138 RBM33 RNA-binding protein 33  4 1.03 1.26 1.15 1.15 2.24 0.2 1.16 0.45 -0.51 139 HIST1H1D Histone H1.3  2 1.11 1.37 1.31 1.26 2.39 0.34 1.26 0.45 -0.54 140 SPTY2D1 Protein SPT2 homolog  2 1.05 1.11 1.28 1.15 2.28 0.2 1.19 0.44 -0.54 141 RALY RNA-binding protein Raly  5 1.09 1.35 1.29 1.24 2.47 0.31 1.3 0.44 -0.61 142 CTNND1 Catenin delta-1  3 1.26 1.53 1.25 1.34 2.59 0.43 1.37 0.44 -0.62 143 GPX1 Glutathione peroxidase 1  2 1.04 1.29 1.22 1.18 2.43 0.24 1.28 0.44 -0.62 144 GRWD1 Glutamate-rich WD repeat-containing protein 1  2 0.89 1.09 1.05 1.01 2.27 0.02 1.18 0.44 -0.63 145 RPS11 40S ribosomal protein S11  8 0.92 1.14 1.07 1.05 2.35 0.06 1.23 0.43 -0.66 146 RPL7L1 60S ribosomal protein L7-like 1  2 1.06 1.33 1.28 1.22 2.55 0.29 1.35 0.43 -0.68  236 147 RBBP4 Histone-binding protein RBBP4  2 1.06 1.24 1.01 1.1 2.48 0.14 1.31 0.43 -0.72 148 TRIM28 Transcription intermediary factor 1-beta  2 0.9 1.09 1.09 1.02 2.41 0.04 1.27 0.43 -0.72 149 EFTUD2 116 kDa U5 small nuclear ribonucleoprotein component  4 0.9 1.11 1.08 1.03 2.45 0.04 1.29 0.43 -0.75 150 NSD1 Histone-lysine N-methyltransferase, H3 lysine-36 and H4 lysine-20 specific  2 1.01 1.28 1.14 1.15 2.59 0.2 1.37 0.43 -0.76 151 ANLN Anillin  2 0.98 1.25 1.15 1.13 2.7 0.17 1.43 0.42 -0.85 152 CSTF3 Cleavage stimulation factor subunit 3  2 1.3 1.71 1.65 1.55 3.16 0.64 1.66 0.41 -0.88 153 DDX28 Probable ATP-dependent RNA helicase DDX28  2 1.13 1.48 1.3 1.3 2.94 0.38 1.56 0.41 -0.9 154 PRDM2 PR domain zinc finger protein 2  4 0.96 1.21 1.24 1.14 2.91 0.19 1.54 0.4 -1 155 SAFB2 Scaffold attachment factor B2  2 1.5 1.53 1.99 1.67 3.63 0.74 1.86 0.39 -1.13 156 RPL34 60S ribosomal protein L34  4 0.92 1.22 1.13 1.09 3.05 0.12 1.61 0.39 -1.13 157 ZNF768 Zinc finger protein 768  2 1.32 1.01 1.07 1.14 3.11 0.18 1.64 0.39 -1.14 158 DDX3X ATP-dependent RNA helicase DDX3X  4 1.33 1.86 1.67 1.62 3.6 0.7 1.85 0.39 -1.15 159 RPL18A 60S ribosomal protein L18a  6 0.96 0.96 1.21 1.04 3.06 0.06 1.61 0.39 -1.17 160 PRRC2A Protein PRRC2A  2 1.37 1.84 1.46 1.56 3.58 0.64 1.84 0.39 -1.18 161 SMARCC1 SWI/SNF complex subunit SMARCC1  3 1.02 1.35 1.33 1.23 3.28 0.3 1.71 0.39 -1.19 162 ZNF664 Zinc finger protein 664  3 1.13 1.13 1.46 1.24 3.36 0.31 1.75 0.38 -1.25 163 SF3B3 Splicing factor 3B subunit 3  2 1.01 1.3 1.36 1.23 3.39 0.29 1.76 0.38 -1.27 164 LMNA Prelamin-A/C  3 1.07 1.11 1.4 1.19 3.34 0.25 1.74 0.38 -1.27 165 MTA2 Metastasis-associated protein MTA2  3 0.98 1.34 1.13 1.15 3.43 0.2 1.78 0.37 -1.36  237 166 FYTTD1 UAP56-interacting factor  9 0.96 1.17 1.32 1.15 3.45 0.2 1.79 0.37 -1.38 167 ALYREF THO complex subunit 4  6 1.17 1.57 1.19 1.31 3.83 0.39 1.94 0.36 -1.53 168 RBM15 Putative RNA-binding protein 15  3 0.9 1.23 1.2 1.11 3.67 0.15 1.88 0.35 -1.56 169 CHERP Calcium homeostasis endoplasmic reticulum protein  2 0.9 1.32 1.28 1.17 4.34 0.23 2.12 0.32 -2 170 SNRPD1 Small nuclear ribonucleoprotein Sm D1  2 1.02 1.41 0.97 1.14 4.7 0.18 2.23 0.29 -2.28 171 XRN2 5'-3' exoribonuclease 2  3 0.93 1.04 1.4 1.12 4.87 0.16 2.29 0.28 -2.42 172 HNRNPH2 Heterogeneous nuclear ribonucleoprotein H2  2 2.23 2.47 3.92 2.87 6.97 1.52 2.8 0.26 -2.67 173 VEZF1 Vascular endothelial zinc finger 1  2 0.82 1.26 1.34 1.14 5.38 0.19 2.43 0.25 -2.76 174 HDAC1 Histone deacetylase 1  2 0.92 1.64 2.25 1.6 9.11 0.68 3.19 0.04 -5.11  238 C.5 CORUM protein complexes found in FUS-DDIT3 nuclear interactome after applying PES z-score cutoff Numbers after each bar indicate the number of complex members found in interactome / total number of complex members in database.   0 1 2 3 4 5  Large Drosha complex  snRNP-free U1A (SF-A) complex  C complex spliceosome  p54(nrb)-PSF-matrin3 complex  TOP1-PSF-P54 complex  DGCR8 multiprotein complex PSF-p54(nrb) complex-log10 (padj-value)CORUM protein complexes total n = 434/112/27/792/32/32/43/20 239 C.6 Gene ontology cellular component classification of full list of FUS-DDIT3 nuclear interactome Table indicates enrichment p-value, T: number of all known genes in specific group, Q: number of genes in query list, Q&T: number of genes in query list belonging to specific group, Q&T/Q: proportion of genes in Q in group T, Q&T/T: proportion of genes in group T that are covered by query list (Q&T), t depth: level of hierarchical subgrouping for cellular component. Bolded cellular components were visualized in bar chart from Figure 4.7.   p-value T Q Q&T Q&T/Q Q&T/T term ID t name t depth 1.6E-60 3954 174 139 0.799 0.035 GO:0031981       nuclear lumen 4 4.89E-58 4301 174 141 0.81 0.033 GO:0044428      nuclear part 3 6.32E-57 3343 174 128 0.736 0.038 GO:0005654       nucleoplasm 4 8.68E-50 5058 174 142 0.816 0.028 GO:0031974    membrane-enclosed lumen 1 8.68E-50 5058 174 142 0.816 0.028 GO:0043233     organelle lumen 2 8.68E-50 5058 174 142 0.816 0.028 GO:0070013      intracellular organelle lumen 3 5.41E-44 7086 174 155 0.891 0.022 GO:0005634         nucleus 6 2.4E-33 8833 174 158 0.908 0.018 GO:0044446     intracellular organelle part 2 7.03E-32 9040 174 158 0.908 0.017 GO:0044422    organelle part 1 7.55E-30 825 172 55 0.32 0.067 GO:0030529       intracellular ribonucleoprotein complex 4 8.04E-30 826 172 55 0.32 0.067 GO:1990904     ribonucleoprotein complex 2 9.18E-27 10628 174 163 0.937 0.015 GO:0043231        intracellular membrane-bounded organelle 5 1.04E-26 1067 174 58 0.333 0.054 GO:0044451        nucleoplasm part 5 1.06E-23 95 170 23 0.135 0.242 GO:0071013        catalytic step 2 spliceosome 5 1.26E-23 4024 166 98 0.59 0.024 GO:0043228     non-membrane-bounded organelle 2  240 1.26E-23 4024 166 98 0.59 0.024 GO:0043232        intracellular non-membrane-bounded organelle 5 1.36E-23 12330 174 169 0.971 0.014 GO:0043229       intracellular organelle 4 7.42E-23 4998 162 106 0.654 0.021 GO:0032991    macromolecular complex 1 3.71E-22 182 170 27 0.159 0.148 GO:0005681       spliceosomal complex 4 1.8E-19 13308 174 170 0.977 0.013 GO:0043226    organelle 1 4.43E-19 12337 174 165 0.948 0.013 GO:0043227     membrane-bounded organelle 2 4.64E-18 717 168 40 0.238 0.056 GO:0016604         nuclear body 6 5.31E-18 14094 174 172 0.989 0.012 GO:0044424      intracellular part 3 2.16E-16 1362 174 52 0.299 0.038 GO:1902494     catalytic complex 2 3.31E-16 14446 174 172 0.989 0.012 GO:0005622     intracellular 2 7.98E-15 20 62 9 0.145 0.45 GO:0019013      viral nucleocapsid 3 2.35E-14 22 62 9 0.145 0.409 GO:0019028      viral capsid 3 4.18E-14 370 168 27 0.161 0.073 GO:0016607          nuclear speck 7 3.22E-13 28 62 9 0.145 0.321 GO:0019012    virion 1 3.22E-13 28 62 9 0.145 0.321 GO:0044423     virion part 2 3.57E-12 933 162 37 0.228 0.04 GO:0005730       nucleolus 4 1.43E-10 91 170 14 0.082 0.154 GO:0034708       methyltransferase complex 4 1.83E-10 980 174 37 0.213 0.038 GO:0005694         chromosome 6 2.25E-10 6 7 4 0.571 0.667 GO:0042382          paraspeckles 7 1.3E-08 344 174 21 0.121 0.061 GO:0000790        nuclear chromatin 5  241 1.41E-08 511 174 25 0.144 0.049 GO:0000785       chromatin 4 2.53E-08 3173 149 59 0.396 0.019 GO:0043234     protein complex 2 3.95E-08 536 174 25 0.144 0.047 GO:0044454       nuclear chromosome part 4 9.88E-08 865 174 31 0.178 0.036 GO:0044427      chromosomal part 3 1.47E-07 570 174 25 0.144 0.044 GO:0000228       nuclear chromosome 4 2.03E-07 103 7 5 0.714 0.049 GO:0016363       nuclear matrix 4 5.87E-07 127 7 5 0.714 0.039 GO:0034399       nuclear periphery 4 9.38E-07 312 131 16 0.122 0.051 GO:0005819     spindle 2 1.12E-06 761 134 24 0.179 0.032 GO:1990234      transferase complex 3 2.14E-06 16567 174 172 0.989 0.01 GO:0044464    cell part 1 2.87E-06 16597 174 172 0.989 0.01 GO:0005623    cell 1 1.37E-05 192 156 13 0.083 0.068 GO:0035770        ribonucleoprotein granule 5 2.61E-05 19 170 6 0.035 0.316 GO:0005686         U2 snRNP 6 4.42E-05 76 116 8 0.069 0.105 GO:0072686      mitotic spindle 3 7.38E-05 16 116 5 0.043 0.312 GO:0005671     Ada2/Gcn5/Ada3 transcription activator complex 2 9.33E-05 12 170 5 0.029 0.417 GO:0034709    methylosome 1 0.000107 115 161 10 0.062 0.087 GO:0022626     cytosolic ribosome 2 0.00013 69 147 8 0.054 0.116 GO:0035097         histone methyltransferase complex 6 0.00021 26 170 6 0.035 0.231 GO:0005689        U12-type spliceosomal complex 5 0.000439 12 86 4 0.047 0.333 GO:0035102    PRC1 complex 1  242 0.00045 70 170 8 0.047 0.114 GO:0030532       small nuclear ribonucleoprotein complex 4 0.000853 315 5 4 0.8 0.013 GO:0005667       transcription factor complex 4 0.00166 36 170 6 0.035 0.167 GO:0005684        U2-type spliceosomal complex 5 0.00177 50 122 6 0.049 0.12 GO:0005876      spindle microtubule 3 0.00225 21 170 5 0.029 0.238 GO:0071010        prespliceosome 5 0.00233 29 121 5 0.041 0.172 GO:0044665          MLL1/2 complex 7 0.00233 29 121 5 0.041 0.172 GO:0071339           MLL1 complex 8 0.00258 245 161 12 0.075 0.049 GO:0005840    ribosome 1 0.00274 60 174 7 0.04 0.117 GO:0000118         histone deacetylase complex 6 0.00291 10 170 4 0.024 0.4 GO:0005687         U4 snRNP 6 0.00294 12 138 4 0.029 0.333 GO:0000346       transcription export complex 4 0.00318 46 147 6 0.041 0.13 GO:0031519       PcG protein complex 4 0.00321 24 158 5 0.032 0.208 GO:0000803          sex chromosome 7 0.00328 63 170 7 0.041 0.111 GO:0097525        spliceosomal snRNP complex 5 0.00543 181 156 10 0.064 0.055 GO:0036464    cytoplasmic ribonucleoprotein granule 1 0.00666 37 116 5 0.043 0.135 GO:1902562    H4 histone acetyltransferase complex 1 0.00702 128 5 3 0.6 0.023 GO:0090575        RNA polymerase II transcription factor complex 5 0.00837 6 108 3 0.028 0.5 GO:0097427    microtubule bundle 1 0.00967 187 161 10 0.062 0.053 GO:0044391     ribosomal subunit 2 0.012 153 5 3 0.6 0.02 GO:0044798       nuclear transcription factor complex 4  243 0.012 1104 131 22 0.168 0.02 GO:0015630    microtubule cytoskeleton 1 0.0124 29 170 5 0.029 0.172 GO:0097526         spliceosomal tri-snRNP complex 6 0.0126 75 174 7 0.04 0.093 GO:0070603    SWI/SNF superfamily-type complex 1 0.0128 21 108 4 0.037 0.19 GO:0097431    mitotic spindle pole 1 0.0148 34 149 5 0.034 0.147 GO:0000791        euchromatin 5 0.0198 36 149 5 0.034 0.139 GO:0005720    nuclear heterochromatin 1 0.0209 16 49 3 0.061 0.188 GO:0045120          pronucleus 7 0.0265 16 174 4 0.023 0.25 GO:0090545     CHD-type complex 2 0.0265 16 174 4 0.023 0.25 GO:0016581      NuRD complex 3 0.0282 3 38 2 0.053 0.667 GO:0044530        supraspliceosomal complex 5 0.0314 17 170 4 0.024 0.235 GO:0034719    SMN-Sm protein complex 1 0.0328 6 170 3 0.018 0.5 GO:0034715    pICln-Sm protein complex 1   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0375714/manifest

Comment

Related Items