THE MOLECULAR BASIS OF SS18-SSX ACTION
IN SYNOVIAL SARCOMA
by
Le Su
B.Sc., Yantai University, 2005
A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
in
THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES
(Anatomy and Cell Biology)
THE UNIVERSITY OF BRITISH COLUMBIA
(Vancouver)
August 2014
© Le Su, 2014
Abstract
Synovial sarcoma is a highly invasive soft tissue malignancy of young people, for which current treatment provides limited benefit. A chromosomal translocation t(X;18) is characteristic of this disease, and the resulting fusion SS18-SSX acts as an oncoprotein that promotes synovial sarcoma oncogenesis through unclear mechanisms. The work presented in this thesis aims to understand the molecular basis of SS18-SSX oncogenic action. Using patient-derived synovial sarcoma cells, we characterized a multicomponent
SS18-SSX complex and discovered a scaffolding role for SS18-SSX in connecting two previously unlinked molecules, activating transcription factor 2 (ATF-2) and transducin- like enhancer of split 1 (TLE1). The latter recruits Polycomb-group (PcG)/histone deacetylase (HDAC) repressors to SS18-SSX transcriptional complex, to silence ATF-2 target genes. In accordance with this model, HDAC inhibitors effectively reverse
PcG/HDAC-mediated epigenetic repression, rescue the expression of SS18-SSX target genes, and suppress synovial sarcoma tumor cell survival. Beyond this finding, we further demonstrated an unexpected transcription-independent mechanism whereby treatment with HDAC inhibitors triggers SS18-SSX oncoprotein degradation, thereby disrupting this aberrant transcriptional complex. HDAC inhibitor-induced SS18-SSX downregulation involves two critical steps: 1) upregulation of the E3 ligase Mule for
SS18-SSX ubiquitination, and 2) SS18-SSX trafficking into the cytoplasm for Mule recognition. Collectively, these findings advance our understanding of the molecular mechanisms underlying SS18-SSX action in synovial sarcoma pathogenesis. This thesis also provides further support for the clinical evaluation of HDAC inhibitors alone or in combination with other agents to treat synovial sarcoma.
! ii! Preface
Chapter 1 is a comprehensive introduction, which reviews the critical discoveries in the synovial sarcoma field and defines the concepts essential for the development of my thesis studies. This chapter also provides the rationale for specific research questions, which are following addressed in three individual chapters.
Chapter 2 is modified from a published manuscript – “Su L, Cheng H, Sampaio
AV, Nielsen TO, Underhill TM, 2010. EGR1 reactivation by histone deacetylase inhibitors promotes synovial sarcoma cell death through the PTEN tumor suppressor.
Oncogene 29, 4352-4361.” My personal contributions include: 1) developing the concept,
2) performing the majority of experiments (except the Cellomics analysis in Figures 2.1
A, 2.5 B and 2.6 B, and the apoptosis assay in Figure 2.1 B), and 3) writing the entire manuscript.
Chapter 3 is modified from a published manuscript – “Su L, Sampaio AV, Jones
KB, Pacheco M, Goytain A, Lin S, Poulin N, Yi L, Rossi FM, Kast J, Capecchi MR,
Underhill TM, Nielsen TO, 2012. Deconstruction of the SS18-SSX fusion oncoprotein complex: insights into disease etiology and therapeutics. Cancer Cell 21, 333-347.” My personal contributions include: 1) developing the concept, 2) performing the majority of experiments (except mass spectrometry analysis in Figure 3.1 A, apoptosis analysis in
Figure 3.3 C, immunohistochemistry analysis of patient specimens in Figure 3.7 B, and microarray analysis in Figure 3.8 A), and 3) writing the entire manuscript.
Chapter 4 is modified from a manuscript, which is being prepared for publication.
My personal contributions include: 1) developing the concept, 2) performing the majority of experiments (except transgenic mouse work in Figures 4.3 C-D, mass spectrometry
! iii! analysis in Figures 4.4 F, 4.5 A and 4.6 A, and immunofluorescent analysis in Figure 4.5
E), and 3) writing the entire manuscript.
Chapter 5 summarizes the critical findings described in this thesis, and builds a novel mechanistic model for SS18-SSX fusion protein action in tumorigenesis. This chapter also discusses the strength and limitations of current thesis studies, and proposes some potential directions for the future investigation on molecular mechanisms of SS18-
SSX regulation.
This thesis research was performed following the protocols approved by the UBC
Research Ethics Board (#H08-0717 and #H06-00013). Mouse synovial sarcoma studies described here were independently conducted by the collaborators, Drs. Kevin B. Jones and Mario R. Capecchi (Huntsman Cancer Institute, Department of Genetics and Howard
Hughes Medical Institute, University of Utah).
! iv! Table of Contents
Abstract ...... ii!
Preface ...... iii!
Table of Contents ...... v!
List of Tables ...... x!
List of Figures ...... xi!
List of Abbreviations ...... xiv!
Acknowledgements ...... xviii!
Dedication ...... xix!
Chapter 1: Introduction ...... 1!
1.1! Chromosomal Translocation and Cancer ...... 1!
1.2! The t(X;18) Translocation in Synovial Sarcoma ...... 3
1.3! SS18-SSX Fusion Protein ...... 5!
1.4! Epigenetic (de)Regulation in Synovial Sarcoma ...... 7
1.5! Rationale and Overview ...... 10!
1.6! Central Theme and Specific Aims ...... 12
Chapter 2: EGR1 Reactivation by Histone Deacetylase Inhibitors Promotes Synovial
Sarcoma Cell Death through the PTEN Tumor Suppressor ...... 19!
2.1! Introduction ...... 19!
2.2! Materials and Methods ...... 22
2.2.1! Cells and chemicals ...... 22!
2.2.2! Western blots ...... 23!
v 2.2.3! RNA interference (RNAi) ...... 23!
2.2.4! Plasmids and transfection ...... 23!
2.2.5! Cell death and apoptosis assay ...... 24!
2.2.6! Chromatin immunoprecipitation (ChIP) analysis ...... 24!
2.2.7! Establishment of stable EGR1-knockdown cell lines ...... 25!
2.3! Results ...... 25!
2.3.1! HDAC inhibitors induce synovial sarcoma apoptosis ...... 25
2.3.2! EGR1 is specifically up-regulated in HDAC inhibitor-mediated synovial
sarcoma apoptosis ...... 26!
2.3.3! EGR1 knockdown reduces HDAC inhibitor-mediated apoptosis ...... 27!
2.3.4! EGR1 mediated synovial sarcoma apoptosis involves PTEN ...... 28!
2.3.5! HDAC inhibitors increase PTEN expression through EGR1 ...... 30!
2.3.6! PTEN is critical for HDAC inhibitor-induced EGR1-dependent apoptosis ...... 31!
2.4! Discussion ...... 32!
2.4.1! EGR1 and Synovial Sarcoma Biology ...... 33!
2.4.2! EGR1-PTEN Pathway and Regulation of Apoptosis in Synovial Sarcoma ...... 34!
Chapter 3: Deconstruction of the SS18-SSX Fusion Oncoprotein Complex: Insights into Disease Etiology and Therapeutics ...... 48!
3.1! Introduction ...... 48!
3.2! Materials and Methods ...... 49
3.2.1! Cells, Tissues and Chemicals ...... 49!
3.2.2! Plasmid DNA Constructs ...... 50!
3.2.3! Immunoprecipitation (IP) and Western Blots ...... 51!
vi 3.2.4! Mass Spectrometry ...... 51!
3.2.5! Immunofluorescence and Immunohistochemistry ...... 52!
3.2.6! Glycerol-Gradient Sedimentation ...... 52!
3.2.7! RNA Interference (RNAi) ...... 53!
3.2.8! Cell Growth and Colony Formation Assay ...... 53!
3.2.9! Cell Death and Apoptosis Assay ...... 54!
3.2.10! Chromatin Immunoprecipitation (ChIP) ...... 54!
3.2.11! Electrophoretic Mobility Shift Assay (EMSA) ...... 55!
3.2.12! Luciferase Reporter Assay ...... 55!
3.2.13! Real-time quantitative PCR ...... 56!
3.2.14! Antibodies ...... 56
3.3! Results ...... 57
3.3.1! Identification of ATF2 and TLE1 within an SS18-SSX complex ...... 57!
3.3.2! Copurification of ATF2 and TLE1 requires SS18-SSX ...... 59!
3.3.3! SS18-SSX/TLE1 functions to repress ATF2 target gene expression ...... 60!
3.3.4! HDAC inhibitors impact SS18-SSX target gene expression through modulation
of TLE1 complex recruitment ...... 65!
3.4! Discussion ...... 67!
Chapter 4: Histone Deacetylase Inhibitors Induce Mule-mediated Ubiquitination and
Degradation of the Synovial Sarcoma SS18-SSX Oncoprotein ...... 90!
4.1! Introduction ...... 90!
4.2! Materials and Methods ...... 91
4.2.1! Cell Culture and Chemicals ...... 91!
vii 4.2.2! Mouse Tumor Models ...... 92!
4.2.3! Tissue and Histological Analysis ...... 92!
4.2.4! Cell Viability Assays ...... 93!
4.2.5! Plasmid DNA Constructs ...... 93!
4.2.6! Chromatin Immunoprecipitation (ChIP) ...... 94!
4.2.7! Electrophoretic Mobility Shift Assays (EMSA) ...... 94!
4.2.8! Immunoprecipitation (IP) and Western Blots ...... 95!
4.2.9! Mass Spectrometry (MS) ...... 95!
4.2.10! RNA Interference (RNAi) ...... 96!
4.2.11! Real-time qPCR (RT-qPCR)...... 96!
4.2.12! Subcellular Fractionation ...... 96!
4.3! Results ...... 97
4.3.1! HDAC inhibitor treatment triggers SS18-SSX complex disruption ...... 97!
4.3.2! HDAC inhibitors accelerate SS18-SSX degradation ...... 98!
4.3.3! Mule targets SS18-SSX for ubiquitination and degradation ...... 101!
4.3.4! Mule-mediated SS18-SSX degradation occurs in the cytoplasm ...... 102!
4.4! Discussion ...... 104!
Chapter 5: Conclusions and Discussion ...... 117!
5.1! General Conclusion ...... 117!
5.2! The Molecular Basis of SS18-SSX-mediated Repression ...... 118
5.3! Therapeutic Targeting of SS18-SSX Fusion Protein ...... 120!
5.4! Clinical Significance ...... 122!
5.5! Regulation of SS18-SSX Intracellular Trafficking ...... 124!
viii References ...... 134!
ix List of Tables
Table 1.1 Chromosomal Translocation and Cancer ...... 17
Table 1.2 Biological Function of Translocation-Associated Gene Products ...... 17
Table 1.3 The Major Components of TrxG and PcG Complexes ...... 18
Table 1.4 Classification of HDAC Proteins ...... 18
x List of Figures
Figure 1.1 Chromosomal Translocation ...... 13
Figure 1.2 The Fusion Product of SS18 and SSX Genes ...... 13
Figure 1.3 A Diagram of SS18-SSX Fusion Protein and Its Partners ...... 14
Figure 1.4 Histone Code Hypothesis ...... 15
Figure 1.5 Transcriptional Regulation by SS18 and SS18-SSX ...... 16
Figure 1.6 A Proposed Model for SS18-SSX Action ...... 16
Figure 2.1 EGR1 is specifically increased in synovial sarcoma during HDAC inhibitor- mediated apoptosis ...... 37
Figure 2.2 HDAC inhibitors activate caspase-3 in synovial sarcoma cells ...... 38
Figure 2.3 The protein level of EGR1 in human embryonic kidney HEK293 and synovial sarcoma SYO-1 and FUJI cells ...... 38
Figure 2.4 EGR1 is specifically induced by romidepsin in stable HEK293 cells expressing the SS18-SSX2 oncoprotein ...... 39
Figure 2.5 Knockdown of EGR1 decreases HDAC inhibitor-mediated apoptosis ...... 40
Figure 2.6 EGR1 knockdown reduces FUJI cell killing by romidepsin treatment ...... 41
Figure 2.7 Knockdown of EGR1 in the synovial sarcoma cell lines SYO-1 ...... 41
Figure 2.8 EGR1 knockdown reduces the sensitivity of SYO-1 cells to MS-275-induced apoptosis ...... 42
Figure 2.9 PTEN is activated in EGR1-induced apoptosis ...... 43
Figure 2.10 Effect of enforced PTEN expression on synovial sarcoma cells ...... 44
Figure 2.11 EGR1 is required for HDAC inhibitor-mediated PTEN induction ...... 44
xi Figure 2.12 Romidepsin directly induces EGR1 but not PTEN in synovial sarcoma cells ... 45
Figure 2.13 Suppression of EGR1 or PTEN attenuates HDAC inhibitor-stimulated cell death
...... 46
Figure 2.14 The effect of EGR1 and PTEN on AKT phosphorylation in synovial sarcoma cells ...... 47
Figure 3.1 SS18-SSX associates with ATF2 and TLE1 in synovial sarcoma ...... 71
Figure 3.2 Related to Figure 3.1 ...... 72
Figure 3.3 Disruption of the SS18-SSX complex reduces synovial sarcoma cell growth ... 74
Figure 3.4 Related to Figure 3.3 ...... 75
Figure 3.5 Molecular association of SS18-SSX with ATF2 and TLE1 ...... 76
Figure 3.6 Related to Figure 3.5 ...... 77
Figure 3.7 ATF2 is recruited to the EGR1 promoter along with TLE1 and SS18-SSX, and is nuclear localized in synovial sarcoma ...... 78
Figure 3.8 SS18-SSX is recruited to genes with an ATF/CRE element ...... 79
Figure 3.9 Related to Figure 3.8 ...... 80
Figure 3.10 ATF2 is critical for DNA-binding of the SS18-SSX complex ...... 81
Figure 3.11 Related to Figure 3.10 ...... 82
Figure 3.12 TLE1 contributes to SS18-SSX-mediated repression ...... 83
Figure 3.13 Related to Figure 3.12 ...... 85
Figure 3.14 Effect of HDAC inhibitors on TLE1 recruitment and SS18-SSX-mediated gene silencing ...... 86
Figure 3.15 Related to Figure 3.14 ...... 87
Figure 4.1 HDAC inhibitor treatment triggers SS18-SSX complex disruption ...... 106
xii Figure 4.2 Related to Figure 4.1 ...... 107
Figure 4.3 HDAC inhibitors induce proteasomal degradation of SS18-SSX ...... 108
Figure 4.4 Related to Figure 4.3 ...... 110
Figure 4.5 SS18-SSX degradation requires the Mule-associated E3 ligase activity ...... 112
Figure 4.6 Related to Figure 4.5 ...... 113
Figure 4.7 SS18-SSX associates with Mule in the cytoplasm ...... 114
Figure 4.8 Related to Figure 4.7 ...... 115
Figure 5.1 Schematic for the Mechanism of HDAC Inhibitor Action ...... 127
Figure 5.2 Mule Tissue Microarray on Patient Specimens ...... 128
Figure 5.3 MDM2 Regulation of Mule in Synovial Sarcoma Cells ...... 129
Figure 5.4 Schematic for HDAC Inhibitor-Induced SS18-SSX Turnover ...... 130
Figure 5.5 HDAC Inhibitor Regulation of SS18-SSX Intracellular Trafficking ...... 130
Figure 5.6 Identification of SS18-SSX N-terminal NES Signal ...... 131
Figure 5.7 SS18-SSX Interaction with CRM1 ...... 132
Figure 5.8 HDAC Inhibitor Regulation of SS18-SSX Action in Synovial Sarcoma ...... 133
xiii List of Abbreviations
ALL Acute lymphoblastic leukemia
ATF2 Activating transcription factor 2
ATF3 Activating transcription factor 3
BCR-ABL Breakpoint cluster region-Abelson
BMI1 B Lymphoma Mo-MLV insertion region 1 homolog bp Base pair
BSA Bovine serum albumin
CBP CREB-binding protein
CCND2 Cyclin D2
CDKN2D Cyclin-dependent kinase inhibitor 2D cDNA Complementary DNA
ChIP Chromatin immunoprecipitation
ChIP-seq ChIP-sequencing
CHX Cycloheximide
CML Chronic myeloid leukemia
CRE cAMP-responsive element
CRM1 Chromosome Region Maintenance 1
CT Cancer-testis
DAB Diaminobenzidine dn Dominant-negative dsDNA Double-stranded deoxyribonucleic acid
! xiv! DTT Dithiothreitol
EED Embryonic ectoderm development
EGR1 Early Growth Response 1
EMSA Electrophoretic mobility shift assay
EtBr Ethidium bromide
EZH2 Enhancer of Zeste Homolog 2
FA Formic acid
FBS Fetal bovine serum
FISH Fluorescence in situ hybridization
GFP Green fluorescence protein
H3K4me3 Lysine 4 trimethylation on histone H3
H3K27me3 Lysine 27 trimethylation on histone H3
HA Hemagglutinin
HAT Histone acetyltransferase hBRM Brahma Homolog
HDAC Histone deacetylase
HEK293 Human embryonic kidney 293
H&E Hematoxylin and eosin
IAA Iodoacetamide
IGH Immunoglobulin heavy chain
IHC Immunohistochemistry
IP Immunoprecipitation
IRES Internal ribosome entry site
! xv! KRAB Kruppel-associated box
MDM2 Mouse Double Minute 2 Homolog mRNA Messenger RNA
MS Mass Spectrometry
MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
Mule MCL-1 ubiquitin ligase E3
NES Nuclear export signal
NUPR1 Nuclear Protein 1
Ph Philadelphia
PI Propidium iodide
PcG Polycomb group
PTEN Phosphatase and tensin homolog deleted in chromosome 10
PTM Post-translational modification
QPGY Glycine, proline, glutamine and tyrosine
RNA Ribonucleic acid
RNAi RNA interference rRNA Ribosomal RNA
RT-qPCR Real-time quantitative PCR
SDS-PAGE Sodium dodecyl sulfate-polyacrylamide gel electrophoresis
SH2 Src homolog 2
SH3 Src homolog 3 shRNA Small hairpin RNA siRNA Small interfering RNA
! xvi! SSXRD SSX repressor domain
SUZ12 Suppressor of zeste 12 homolog
TGFβ1 Transforming growth factor β1
TLE1 Transducin-like enhancer of split 1
TMA Tissue microarray
TrxG Trithorax group
TSS Transcription start site
Ub Ubiquitin
UPS Ubiquitin-proteasome system wt Wild type
! xvii! Acknowledgements
First, I wish to thank my supervisor Dr. T. Michael Underhill and my co- supervisor Dr. Torsten O. Nielsen for their instruction and direction during the course of my doctoral research. Importantly, Drs. Underhill and Nielsen allowed me to ask and address my own questions. This greatly motivated me to continue doing research as my future career.
I also love to extend my thanks to the members of my supervisory committee,
Drs. Fabio Rossi, Calvin Roskelley and Michel Roberge, for their guidance in my graduate studies. I must also express my gratefulness to Drs. Mario Capecchi and Kevin
Jones (Huntsman Cancer Institute, University of Utah) for enabling me to participate in what proved to be a fruitful collaboration.
This thesis would never be feasible if not for the efforts and support of several important individuals. I would like to thank Drs. Makoto Endo and Marina Pacheco for processing patient specimens. I also wish to thank Petra Schreiner, Kimberly Snyder and
Dr. Kelly McNagny for their help with animal experiments and primary cell cultures, which have served as a valuable tool in my research. I appreciate the aid and assistance provided by the past and current members in Underhill and Nielsen laboratories.
Finally, I love to thank my parents and in-laws for believing in me and supporting me throughout my PhD training. Most importantly, I wish to thank my wife, Jue-Hong
Wang, who always stays together with me to face all challenges.
! xviii! Dedication
To my grandmother,
Xue-Po Yang
! xix! Chapter 1: Introduction
1.1 Chromosomal Translocation and Cancer
Cancer is a deadly disease that is often associated with well-defined genetic defects, a very severe type of these being chromosomal translocation. Theoretically, a translocation is formed through two steps: firstly, double-stranded deoxyribonucleic acid
(dsDNA) breaks on two non-homologous chromosomes; and secondly, the broken ends of two partner chromosomes associate with each other and fuse into one chromosome
(Figure 1.1 A). At the molecular level, such rearrangements lead to two different consequences. In some cases, the two translocated genes exchange their promoter regions without changes in their coding sequences (Figure 1.1 B). For example, in Burkitt’s lymphoma, the t(8;14) translocation results in abnormal expression of the MYC oncogene which is now placed under the control of the immunoglobulin heavy chain (IGH) gene promoter (Taub et al., 1982). However, in many instances, the coding sequences of translocated genes are disrupted prior to their fusion to create a novel chimeric gene
(Figure 1.1 C). A good example of this, is the famous Philadelphia (Ph) chromosome
(also known as the t(9;22) translocation), which encodes the breakpoint cluster region-
Abelson (BCR-ABL) fusion oncoprotein which is associated with chronic myeloid leukemia (CML) (Groffen et al., 1983; Heisterkamp et al., 1985; Rowley, 1973;
Shtivelman et al., 1985).
Although abnormal chromosomes were observed in tumor cells more than a century ago, there was long-term disagreement about whether these abnormalities are the cause of oncogenic transformation. In 1973, Janet D. Rowley identified the Ph translocation as the
! 1! crucial event for the development of CML (Rowley, 1973). The Ph or t(9;22) results in an exchange between chromosome 9 and chromosome 22, by which the 5’ coding sequence of BCR is fused to the 3’ region of ABL. The resulting fusion protein BCR-ABL exerts an aberrant tyrosine kinase activity, which is mechanistically responsible for the oncogenic action of BCR-ABL (Konopka et al., 1984). Along these same lines, animal studies further demonstrated that BCR-ABL is sufficient to induce CML-like myeloproliferative disorders in mice (Daley et al., 1990; Kelliher et al., 1990). These early discoveries provided the first definitive evidence supporting the concept that chromosome translocation could cause cancer, and significantly inspired biologists to identify and characterize new translocations involved in malignant transformation. At present, a great number of fusion genes (more than 300) have been identified in human cancer patients (Mitelman et al., 2007). Interestingly, the majority of them occur in hematological malignancies (such as, leukemia and lymphoma) and childhood sarcomas
(Table 1.1). Consistent with the role of chromosomal translocations in tumor development, most of the partner genes involved in fusions are oncogenes. Their protein products broadly function to transmit cellular signals, bind to DNA or ribonucleic acid
(RNA), thereby modifying global chromatin structure and consequently gene expression
(Table 1.2).
The discovery of chromosomal translocations has had an important impact on disease detection and identification, particularly because they are closely associated with specific tumor types. Firstly, translocations could serve as a powerful tool for diagnosis.
With the advent of fluorescence in situ hybridization (FISH) technique, pathologists can recognize specific translocation products and therefore provide a more accurate
! 2! determination of tumor type even within tumors with similar histological characteristics.
Secondly, translocations are also useful for prognosis. For example, in acute lymphoblastic leukemia (ALL) patients, the t(12;21) translocation indicates a good prognosis, while the Ph is commonly associated with more aggressive leukemia progression and a poorer outcome (Alvarez et al., 2011; Kovacsovics and Maziarz, 2006).
Finally, and perhaps more importantly, translocations could be used as the target to develop effective therapies. In fact, the success of using the BCR-ABL kinase inhibitors
STI-571 in CML treatment (Druker et al., 2001; O’Dwyer and Druker, 2000) has motivated scientists in diverse fields to characterize additional chromosomal translocations and to understand the mechanisms of their actions, in order to design and develop more efficacious tumor-specific therapeutics.
1.2 The t(X;18) Translocation in Synovial Sarcoma
Sarcomas are mesenchymal malignancies that frequently present in bone, cartilage, fat, muscle and other connective tissues (Frith et al., 2013). They are significantly different from carcinomas, which are cancers arising from transformed cells of epithelial origin. In human, sarcomas are generally rare, whereas more common malignancies such as breast or colon cancer are most typically carcinomas. Sarcomas are normally named according to the their tissue of origin. For instance, osteosarcoma forms in and resembles bone, while liposarcoma is adipogenic in nature. In addition to multiple origins, sarcomas are also associated with diverse malignant behaviors – from low grade that slowly grows to high grade that is extremely invasive and metastatic. However, compared to carcinomas, sarcomas have a relatively simple biology with limited genetic
! 3! abnormalities. Chromosomal translocation is the most common mechanism involved in sarcomagenesis (Taylor et al., 2011). This molecular aberration often creates specific chimeric transcription factors, which leads to transcriptional dysregulation.
Synovial sarcoma is a high-grade soft tissue sarcoma, which is unrelated to the synovium and known to arise almost anywhere in the body (Davicioni et al., 2008). Most patients are diagnosed in the second decade of life. Males and females are equally affected, and the tumors typically originate in the extremities (especially the lower ones), but these tumors often metastasize to other sites including the lungs and lymph nodes
(Davicioni et al., 2008; Loya et al., 2007; Rong et al., 2009). Traditional treatment typically involves wide surgical excision with adjuvant radiation. But, the 5-year and 10- year survival rates have been reported as low as 36% and 20%, respectively (Mullen et al., 1994). Unfortunately, current cytotoxic chemotherapies provide limited benefit for the treatment of this disease.
Pathohistological analysis of synovial sarcoma revealed that it includes mesenchymal spindle-shaped cells with variable epithelial differentiation (Fisher, 1986).
Based on the morphology, synovial sarcoma can be divided into three main subtypes: monophasic (mesenchymal cells only), biphasic (spindle-shaped cells with focal epithelial compartments) and poorly differentiated (small round cells). This feature has also been observed in other forms of sarcomas (such as, fibrosarcoma and leiomyosarcoma). However, synovial sarcoma carries a pathognomic chromosomal translocation t(X;18)(p11.2;q11.2), which does not exist in other tumor types. As a direct consequence of this abnormality, the SS18 (also known as SYT) gene on chromosome 18 is fused with one of three highly similar genes (SSX1, SSX2, or SSX4) on chromosome
! 4! X (Figure 1.2 A) (Clark et al., 1994). Given its specificity for synovial sarcoma, the t(X;18) translocation has widely been accepted as an extremely reliable diagnostic marker, greatly improving on the use of strictly morphological-based diagnosis
(Davicioni et al., 2008). Since the t(X;18) occurs as a sole cytogenetic abnormality in synovial sarcoma, its formation thus raises intriguing questions of whether and how this translocation leads to malignancy.
1.3 SS18-SSX Fusion Protein
Synovial sarcoma-associated translocation t(X;18) involves two chromosomes (18 and X), which may theoretically result in the formation of two reciprocal rearrangements, der(X) and der(18) (Figure 1.2 A). However, in primary tumors, only the der(X) rather than der(18) chromosome has been consistently detected (Smith et al., 1997). In addition, loss of der(18) but not der(X) has been observed in culture of synovial sarcoma-derived cell lines such as, FUJI and HS-SY-II (Clark et al., 1994; Sonobe et al., 1992). This indicates that the creation of the der(X) is a key event during tumor development. Indeed, the synovial sarcoma-specific SS18-SSX fusion gene is a direct result of this event. Colin
S. Cooper and colleagues have provided molecular details of the fusion gene SS18-SSX in which the exons 1-10 from the 5’ part of SS18 are associated with exons 5-6 from the
3’ part of SSX (Figure 1.2 A) (Crew et al., 1995). Consequently, in the SS18-SSX fusion protein the sequences of SS18 are almost entirely included, except for the last eight amino acids, which are replaced by the seventy-eight amino acids of the SSX carboxy- terminus (Figure 1.2 B).
! 5! SS18 exhibits a near ubiquitous expression pattern and can be found in most tissues and cell lines in both rodents and humans (de Bruijn et al., 2001). Contrary to this, the expression of human SSX genes (SSX1-9) is restricted to the testis and certain tumors, and is part of the family of cancer-testis (CT) associated antigens (Gure et al., 2002).
Both SS18 and SSX gene products are localized to the nucleus, while their biological functions still remain unresolved. The SS18 protein lacks apparent DNA-binding motifs, but contains a transactivation domain rich in glycine, proline, glutamine and tyrosine
(QPGY domain), which has been shown when combined with a DNA-binding domain to activate a promoter in luciferase reporter assays (Bett et al., 1997; Thaete et al., 1999).
Additionally, multiple SH2-/SH3-binding domains have been identified within the carboxy end of SS18 (Figure 1.3 A). Considering that these domains are well known to be involved in protein-protein interactions (Pawson and Gish, 1992), the SS18 protein most likely expresses its transcriptional regulatory function through association with unknown DNA-binding factors. In contrast, SSX appears to behave as a transcriptional co-repressor possessing two significant repression domains, the amino-terminal Kruppel- associated box (KRAB) and carboxy-terminal SSX repressor domain (SSXRD) (Figure
1.3 B) (Crew et al., 1995; Lim et al., 1998). Although these two domains are disrupted in the formation of SS18-SSX fusion protein, the latter SSXRD is entirely retained (Figure
1.3 C) and proven to have a dominant-negative effect on SS18-mediated transcriptional activation (Bett et al., 1997; Lim et al., 1998; Thaete et al., 1999). The control of gene silencing by SSXRD is believed to involve chromatin remodeling, due to its association with core histones and Polycomb repressive proteins (Kato et al., 2002; Lim et al., 1998;
Soulez et al., 1999). Therefore, the fusion oncoprotein and its components have no DNA-
! 6! binding activity and the major protein-protein interaction domains occur in the SS18 region, leading to a long-standing hypothesis that SS18-SSX may function as a transcriptional co-repressor to inhibit the expression of SS18 target genes. If this is the case, an important issue to resolve is defining the mechanisms by which this chimeric protein dysregulates “normal” cell function and contributes to tumor development.
1.4 Epigenetic (de)Regulation in Synovial Sarcoma
Chromatin is a highly ordered structure with the central nucleosomal component being comprised of DNA and core histone proteins. Post-translational modifications
(PTMs) on the amino-termini of histones can generate differential affinities for transcriptional regulator recruitment and therefore regulate gene expression without changes in the DNA sequence. These reversible alterations usually remain inheritable during cell division, and have opened up a relatively new concept named the “histone code” (Berger, 2007; Jenuwein and Allis, 2001). In mammals, this epigenetic system is closely governed by two groups of proteins, Polycomb (PcG) and Trithorax (TrxG)
(Table 1.3), which trimethylate histone H3 at Lysine 27 (H3K27me3) and Lysine 4
(H3K4me3), respectively. These protein complexes are central to almost all biological processes, but display opposite functions: H3K27me3 is associated with transcriptional repression, while H3K4me3 is coupled to activation of gene expression (Figure 1.4)
(Berge, 2007; Mills, 2010). The balance between TrxG and PcG activities is predominantly regulated by a specific enzyme family termed histone deacetylase
(HDAC). The HDAC family contains eighteen members classified into four groups
(Table 1.4) (Witt et al., 2009). HDAC proteins partially diminish TrxG transactivation
! 7! activity through reducing the level of HAT-mediated histone acetylation. Meanwhile, these enzymes also recruit PcG repressors through protein-protein interactions, to establish repressive histone marks for chromatin condensation and transcriptional inhibition.
In synovial sarcoma, the fusion protein SS18-SSX has been shown to associate with both PcG and TrxG members, including B Lymphoma Mo-MLV insertion region 1 homolog (BMI1) and Protein Brahma Homolog (hBRM) (Kato et al., 2002; Nagai et al.,
2001; Soulez et al., 1999; Thaete et al., 1999). This suggests a potential role for SS18-
SSX in governing the epigenetic balance of activating versus repressive histone marks at specific gene loci. It is of interest to note that like the coactivators SS18, the fusion protein has also been found to interact with the histone acetyltransferase (HAT) protein p300 and its homolog CREB-binding protein (CBP), both of which can diminish PcG- mediated H3K27me3 repressive signals through acetylating the H3K27 residues (Eid et al., 2000; Pretto et al., 2006; Tie et al., 2009). However, identification of histone deacetylase (HDAC) complex components as SS18-SSX-interacting cofactors (Ito et al.,
2004) increases the possibility that in the SS18-SSX complex HDAC proteins may inhibit
HAT activity. Collectively, these observations support the concept that the SS18-SSX fusion protein functions as an epigenetic switch in combination with PcG and HDAC factors to override the TrxG/HAT activity in tumor cells (Figure 1.5). Considering that gain-of-PcG and loss-of-TrxG function activity underlies many pathological processes, in particular malignant transformation (Mills, 2010), an interesting issue is whether the fusion protein SS18-SSX alone is sufficient to promote tumorigenesis.
! 8! To address the importance of SS18-SSX in tumorigenesis, Kazuo Nagashima and colleagues engineered 3Y1 rat fibroblast cells to express human SS18, SSX1, and SS18-
SSX1, and found that only SS18-SSX1 was able to promote cell growth in culture, colony formation in soft agar, and tumor growth in xenografted mice (Nagai et al., 2001).
These experiments demonstrated that the SS18-SSX fusion protein has transforming activity, and emphasized the need for the intact fusion oncoprotein for transformation, as neither SS18 nor SSX alone could change the behavior of 3Y1 cells. More direct evidence, supporting the causative role of SS18-SSX in tumorigenesis, comes from recent studies from Mario R. Capecchi’s laboratory. This group used gene-targeting technologies to engineer mice that expressed an SS18-SSX2 fusion gene in skeletal muscle progenitors, thereby producing the first synovial sarcoma mouse model (SSM2)
(Halder et al., 2007; Halder et al., 2009). All of the transgenic mice present tumors, which are highly similar to human synovial sarcomas, according to their morphology, immunohistology and gene expression profiles. Therefore, the evidence from the SSM2 mouse model strongly supports a fundamental role for the SS18-SSX fusion protein in tumor development. To date, the mechanism by which SS18-SSX functions as an oncoprotein is still far from understood, but repression of HDAC activity by the inhibitors (such as, romidepsin and SB939) has been found to effectively suppress synovial sarcoma in vivo and in vitro (Ito et al., 2005; Lubieniecka et al., 2007; Su et al.,
2010). The sensitivity of synovial sarcoma to HDAC inhibitors implies an essential role of epigenetic regulation in this pathological process. This is certainly consistent with the hypothesis that loss of HDAC activity will allow TrxG and HAT coactivators to abrogate the action of PcG repressive complexes, culminating in a reversal of SS18-SSX-mediated
! 9! transcriptional repression. However, the lack of detectable DNA-binding activity for
SS18-SSX obviously increases the complexity of identifying its target genes, and discovering a direct link between SS18-SSX-mediated repression and tumor formation.
1.5 Rationale and Overview
A long-standing puzzle in the field of synovial sarcoma is how the molecular events required for tumorigenesis are regulated by the fusion oncoprotein SS18-SSX. A good example is the apoptotic response of human synovial sarcoma cells to small interfering
RNA (siRNA)-based depletion of SS18-SSX transcripts, which connects SS18-SSX protein action to control of cell growth and apoptosis (Peng et al., 2008; Takenaka et al.,
2010). To examine this aspect, we established an inducible human embryonic kidney
(HEK293) cell line for expressing SS18-SSX2, and used complementary DNA (cDNA) microarrays to analyze the change of gene expression patterns before and after SS18-
SSX2 induction. Consequently, under these conditions one of the most significantly downregulated genes was the tumor suppressor Early Growth Response 1 (EGR1), which initiates programmed cell death mainly through activation of PTEN and/or p53 signaling cascades (Lubieniecka et al., 2007). In fact, loss of EGR1 expression has widely been reported in human cancers (such as, hematopoietic malignancies and non-small-cell lung cancers) (Calogero et al., 1996; Huang et al., 1997; Le Beau et al., 1993; Levin et al.,
1995). In agreement with these observations, we discovered a consistent downregulation of EGR1 and its target genes in synovial sarcoma, based upon the microarray expression profiles prepared from patient specimens (Lubieniecka et al., 2007). The importance of
EGR1 in regulating synovial sarcoma survival was further investigated using two
! 10! experimental paradigms. In the first series of experiments, we transiently transfected two human synovial sarcoma cell lines (SYO-1 and FUJI) with EGR1 cDNA or control empty vector, and found that rather than the control, EGR1 overexpression triggers
Caspase-3 activation and apoptotic cell death in both cell models. In the second set of experiments, EGR1 derepression appears to be essential for HDAC inhibitor-induced apoptosis in synovial sarcoma. Firstly, we discovered EGR1 as an HDAC inhibitor- sensitive target since its transcript and protein levels were significantly increased after drug treatment. This study was then extended by the RNA interference (RNAi) experiment, in which siRNA-dependent knockdown of EGR1 protects synovial sarcoma cells from the apoptotic activity of HDAC inhibitors. In order to gain more detailed insights into the mechanism of EGR1’s anti-tumor action in synovial sarcoma, future investigation should concentrate on identifying HDAC inhibitor-induced EGR1- dependent tumor suppressive pathways. In addition, what is also required is further investigation into the mechanism(s) regulating EGR1 repression in synovial sarcoma.
To further understand SS18-SSX function, I focused on the relationship of SS18-
SSX to EGR1 repression, and subsequently established a direct link between SS18-SSX and EGR1 by demonstrating recruitment of the fusion oncoprotein to the human EGR1 promoter. Along the same lines, we further showed that the core PcG repressors (such as,
Enhancer of Zeste Homolog 2 (EZH2) and BMI1) also associate with the same EGR1 promoter region through SS18-SSX (Lubieniecka et al., 2007). Consistent with this,
EGR1 repression by SS18-SSX is faithfully correlated with the establishment of PcG- mediated H3K27me3 silencing marks. These data strongly suggest that EGR1 downregulation in synovial sarcoma is a direct consequence of SS18-SSX-mediated
! 11! chromatin remodeling (Figure 1.6). Therefore, it would be of interest to follow the behavior of SS18-SSX and its interacting cofactors (such as, transcription factors and chromatin modifiers) in future studies, to test exactly how the fusion oncoprotein SS18-
SSX binds to DNA and how its repressive actions in synovial sarcoma are regulated in response to HDAC inhibitor treatment.
1.6 Central Theme and Specific Aims
My overarching goal of the current thesis is to understand the molecular mechanisms underlying the function of SS18-SSX fusion oncoprotein in synovial sarcoma. Toward this end, the following chapters will focus on three principal objectives:
Chapter 2 – To uncover the mechanism of anti-tumor action for EGR1 in synovial sarcoma, by investigating the essential signal transduction pathway(s) downstream of
EGR1 in response to HDAC inhibitor stimulation;
Chapter 3 – To uncover the underlying mechanism for SS18-SSX-mediated transcriptional repression, by identifying novel SS18-SSX-interacting cofactors and characterizing the fusion oncoprotein complex(es);
Chapter 4 – Using the model of SS18-SSX transcriptional complex(es) (established in Chapter 3), to uncover the molecular mechanism underlying the anti-synovial sarcoma tumor activity of HDAC inhibitors.
! 12!
Figure 1.1 – Chromosomal Translocation
Schematic for formation of chromosomal translocation (A) and its possible consequences (B-C). Arrow indicates the break point.
Figure 1.2 – The Fusion Product of SS18 and SSX Genes
(A) Reciprocal fusion genes caused by the t(X;18) translocation. (B) Schematic for the fusion gene product SS18-SSX.
! 13!
Figure 1.3 – A Diagram of SS18-SSX Fusion Protein and Its Partners
Protein sequences of SS18 (A), SSX (B) and SS18-SSX (C). SNH, SYT N-terminal Homolog domain; QPGY, Glutamine/Proline/Glycine/Tyrosine-rich domain; SH2-BM, Src Homolog 2- binding motif; SH3-BM, Src Homolog 3-binding motif; KRAB, Kruppel-associated box domain; SSXRD, SSX repressor domain.
! 14!
Figure 1.4 – Histone Code Hypothesis
(A) Transcriptional Activation: TrxG proteins function to trimethylate the Lysine 4 residues on histone H3 (H3K4me3), which recruits other modifiers (such as HAT proteins) to acetylate multiple lysine residues on core histones. Histone acetylation increases chromatin accessibility for RNA Polymerase II (RNAPII) elongation. (B) Gene Silencing: HDAC proteins remove HAT- mediated acetylation, while PcG Repressive Complex 2 (PRC2) specifically trimethylates the Lysine 27 residues on histone H3 (H3K27me3), which recruits PcG Repressive Complex 1 (PRC1) to ubiquitylate histone H2A. These posttranslational modifications mediate chromatin condensation, thereby preventing RNAPII elongation and removing RNAPII from DNA.
! 15!
Figure 1.5 – Transcriptional Regulation by SS18 and SS18-SSX
SS18 (A) has transactivation activity, whereas SS18-SSX fusion protein (B) functions to repress gene expression.
Figure 1.6 – A Proposed Model for SS18-SSX Action
In the human EGR1 promoter, the fusion protein SS18-SSX associates with specific transcription factor(s) for DNA-binding, while its C-terminal SSXRD domain cooperates with HDAC/PcG proteins to epigenetically silence transcription.
! 16! Table 1.1 – Chromosome Translocation and Cancer
Table 1.2 – Biological Function of Translocation-Associated Gene Products
! 17! Table 1.3 – The Major Components of TrxG and PcG Complexes
Table 1.4 – Classification of HDAC Proteins
! 18! Chapter 2: EGR1 Reactivation by Histone Deacetylase Inhibitors
Promotes Synovial Sarcoma Cell Death through the PTEN Tumor
Suppressor
2.1 Introduction
Early growth response-1 (EGR1) is a zinc-finger transcription factor rapidly induced by a wide range of mitogenic stimuli (Sukhatme et al., 1987) with important roles in the regulation of cell growth, differentiation, and death (Gashler and Sukhatme, 1995;
Sukhatme, 1990). It has been suggested that EGR1 is a tumor suppressor, as its overexpression can dramatically inhibit tumor cell growth in vitro and in xenografted mice (Huang et al., 1994; Liu et al., 1996). EGR1 appears to act as a key regulator of cell death and growth through regulation of a number of downstream targets including, transforming growth factor β1 (TGFβ1), phosphatase and tensin homolog deleted in chromosome 10 (PTEN), p53, p73, FAS, cyclin D2 (CCND2) and cyclin-dependent kinase inhibitor 2D (CDKN2D) (Krones-Herzig et al., 2005; Liu et al., 1999; Virolle et al., 2001; Virolle et al., 2003; Yu et al., 2007). Indeed, loss of EGR1 expression is observed in many human tumors, including breast carcinomas, non-small cell lung cancers, hematopoietic malignancies, gliomas, and sarcomas (Calogero et al., 1996;
Calogero et al., 2004; Huang et al., 1997; Le Beau et al., 1993; Levin et al., 1995;
Lubieniecka et al., 2008). RNA interference (RNAi)-mediated EGR1 suppression was shown to enhance the tumorigenic phenotype of v-sis-transformed NIH/3T3 cells (Huang et al., 1995), suggesting that reduced endogenous EGR1 expression may correlate with
! 19! tumor formation. However, the mechanisms underlying EGR1 down-regulation in tumorgenesis have not been fully elucidated.
We recently demonstrated that the oncogenic protein SS18-SSX, characteristic of synovial sarcoma, directly participates in EGR1 repression (Lubieniecka et al., 2008).
Synovial sarcoma is a high-grade soft tissue malignancy with a peak incidence in young adults, which exhibits a pathognomonic t(X;18)(p11.2;q11.2) chromosomal translocation
(Okcu et al., 2003). The resulting fusion protein SS18-SSX contains a transcriptional activation domain from SS18 and a repressor domain from SSX (Brett et al., 1997; Lim et al., 1998). Lacking DNA-binding domains, SS18-SSX dysregulates a number of genes through interacting with SNF/SWI and polycomb-group (PcG) chromatin remodeling factors (Nagai et al., 2001; Soulez et al., 1999; Thaete et al., 1999). In previous investigations, enforced SS18-SSX expression was shown to repress EGR1 transcription and promote growth of human fibroblast cells (de Bruijn et al., 2006; Nagai et al., 2001).
In synovial sarcoma, EGR1 expression is maintained at low levels, whereas it is markedly up-regulated by SS18-SSX RNAi (Lubieniecka et al., 2008). We have found that the SS18-SSX protein is directly associated with the EGR1 promoter, and that two
PcG repressor factors, BMI-1 and EZH2, interact with the same EGR1 promoter region in an SS18-SSX-dependent manner. Notably, PcG-mediated histone H3 Lysine27 trimethylation (H3K27me3), a key mark of genomic silencing, is tightly correlated with
EGR1 repression. These data suggest that EGR1 down-regulation occurs as a result of
SS18-SSX-mediated chromatin remodeling. Indeed, this mechanism has been further supported by the observation that histone deacetylase (HDAC) inhibitors efficiently remove H3K27me3 marks, recruit acetyl groups to histone H3 tails, and thereby
! 20! reactivate expression of EGR1 in synovial sarcoma. However, the precise contribution of
EGR1 to synovial sarcoma or HDAC-inhibitor responsiveness of these cells is unclear.
HDAC inhibitors are a group of small molecules that can promote hyperacetylation of histones to alter gene expression and particularly to activate genes that are usually silenced in tumors (Bolden et al., 2006). In fact, several HDAC inhibitors, such as FK228
(also known as romidepsin and depsipeptide), MS-275, SAHA, and PXD101 are being tested in clinical trials (Moradei et al., 2008; Stimson et al., 2009), and SAHA has recently been approved for use in T-cell lymphomas (Khan and La Thangue, 2008). At the molecular and cellular levels, it is widely accepted that HDAC inhibitors suppress tumor growth by inducing cyclin-dependent kinase inhibitor p21/WAF1/CIP1-mediated cell cycle arrest, as well as mitochondrial/cytochrome c-dependent apoptotic cell death
(Archer et al., 1998; Medina et al., 1997). There is also evidence to suggest that HDAC inhibitors impact PTEN expression (Gan and Zhang, 2009; Pan et al., 2007), however, the underlying mechanisms have not been resolved.
PTEN is an extremely important phosphatase that regulates the activity of the
PI3K/AKT survival pathway (Downes et al., 2001). Furthermore, in response to a variety of cellular insults, EGR1 has been shown to regulate PTEN expression (Virolle et al.,
2001). We therefore investigated the possibility that EGR1 initiates HDAC inhibitor- mediated apoptosis by serving as an upstream activator of PTEN. Here we show a critical role of EGR1 reactivation in HDAC inhibitor-induced apoptosis of synovial sarcoma.
The importance of EGR1 for this process is confirmed by the observations that repression of EGR1 by RNAi inhibits HDAC inhibitor-induced PTEN activation and apoptosis, and that EGR1 protein directly binds to the PTEN promoter region after HDAC inhibitor
! 21! addition. Knockdown of PTEN is able to counteract the effects of both HDAC inhibitors and heterologous expression of EGR1. Furthermore, an EGR1-PTEN network was found to influence synovial sarcoma cell survival, demonstrating that SS18-SSX-mediated attenuation of this pathway enables these cells to escape cell death.
2.2 Materials and Methods
2.2.1 Cells and chemicals
Synovial sarcoma cell lines SYO-1 and FUJI (kindly provided by Dr. Kazuo
Nagashima, Hokkaido University School of Medicine, and Dr. Akira Kawai, National
Cancer Centre Hospital, Tokyo, Japan) and human embryonic kidney HEK293 cells were cultured in RPMI-1640 and DMEM media (Invitrogen), respectively. Stable HEK293 cell lines, expressing Myc-tagged His, SS18, or SS18-SSX2, were grown in DMEM with 400
µg/mL Zeocin (Invitrogen). All media were supplemented with 10% fetal bovine serum
(Invitrogen), and all cells were maintained under 37°C, 95% humidity, 5% CO2.
HDAC inhibitors romidepsin (FK228, depsipeptide, NSC-630176), MS-275,
PXD101, and SAHA were obtained through the Developmental Therapeutic Branch of the National Cancer Institute (Bethesda, MD). Cycloheximide and z-VAD-fmk were purchased from BioVision (Mountain view, CA) and Promega (Madison, WI), respectively. All reagents were dissolved in DMSO prior to further application.
Real-time quantitative PCR (RT-qPCR) analysis. Total RNA was isolated and then transcribed to cDNA using the RNeasy Mini kit (QIAGEN) and the high capacity cDNA reverse transcription kit (Applied Biosystems). Taqman gene expression assays were carried our using the ABI 7500 Fast Real-Time PCR system with primer/probe sets
! 22! specific for EGR1, PTEN, or 18S ribosomal RNA (rRNA) (Applied Biosystems). All
RT-qPCR data were normalized to 18S rRNA.
2.2.2 Western blots
Cells were lysed with the SDS lysis buffer (Millipore). Proteins were separated on
10-12% SDS-PAGE gels, and transferred to nitrocellulose membranes (Bio-Rad
Laboratories). Blots were incubated with antibodies specific to caspase-3, AKT, p-AKT
(Serine 473), EGR1 (Cell Signaling Technology), PTEN, and actin (Santa Cruz
Biotechnology). Signals were visualized using the Odyssey Infrared System (LI-COR
Biosciences).
2.2.3 RNA interference (RNAi)
The EGR1-specific and control random small interfering RNAs (siRNAs) were purchased from Dharmacon, and the PTEN-specific siRNA from Cell Signaling
Technology. At 60% confluence, cells were transfected with the EGR1, PTEN, or control random siRNA using Lipofectamine RNAiMAX transfection reagent (Invitrogen) according to the manufacturer’s instructions. After culturing for 24 hr, cells were treated with HDAC inhibitors, and analyzed by western blotting or RT-qPCR to determine knockdown efficiency.
2.2.4 Plasmids and transfection
The EGR1 and PTEN cDNA (Invitrogen) were inserted into a Gateway pENTR1A vector followed by recombineering into the internal ribosome entry site (IRES)-EGFP
! 23! expression constructs by the LR site-specific recombination reaction (Invitrogen). Cells were grown on 24-well plates to 60% confluence, and transfected with EGR1, PTEN, or empty vector using FUGENE 6 transfection reagent (Roche). 48 hr post transfection,
EGFP-expressing cells were visualized under epifluorescence, and EGR1 or PTEN proteins were detected by western blotting analysis.
2.2.5 Cell death and apoptosis assay
Cells were grown on 96-well plates, and stained with propidium iodide (PI) at the concentration of 500 ng/mL. After indicated treatments, PI-positive cells were detected and counted over time by a high-content screening system (KineticScan HCS Reader,
Cellomics, Inc.). In other gene knockdown or overexpression assays, cells were cultured with PI on 24-well plates, and transfected with the indicated siRNA or plasmid. PI- positive cells were counted under a fluorescence microscope, and then normalized to
EGFP-positive cells (at least 500 cells scored for each experiment) to give the percentage of cell death. For apoptosis assays, following treatment, cells were harvested and resuspended in Annexin V-FITC-PI dye (Invitrogen). After addition of the Annexin V- binding buffer, the samples were run through a FACScan flow cytometer (Becton
Dickinson). Summit for MoFlo Acquisition and Sort Control Software (Cytomation, Inc.) were used to quantify apoptosis.
2.2.6 Chromatin immunoprecipitation (ChIP) analysis
ChIP assays were performed using the Active Motif ChIP-IT kit as described
(Lubieniecka et al., 2008). EGR1-chromatin complexes were immunoprecipitated using
! 24! the EGR1 (588) antibody (Santa Cruz Biotechnology). Specific sequences of the human
PTEN promoter were amplified by PCR using the following primers: 5’-
CTCGGTCTTCCGAGGC-3’ and 5’-CCGAGCGCGTATCCTG-3’ for P1, 5’-
AAACGAGCCGAGTTACCG-3’ and 5’-GACTGCATTCGCTCTTTCCT-3’ for P2.
2.2.7 Establishment of stable EGR1-knockdown cell lines
The EGR1 and control random small hairpin RNA (shRNA) lentiviral vectors were purchased from Open Biosystems. After transfection of the plasmids into the packaging cell line, TLA-HEK293T, the lentiviral supernatant was transduced in the biphasic synovial sarcoma cell line SYO-1. Cells were maintained in puromycin (2 µg/mL)- containing RPMI-1640 media with 10% fetal bovine serum. Puromycin-resistant colonies were isolated, and following HDAC inhibitor treatment, EGR1 expression was analyzed by western blot.
2.3 Results
2.3.1 HDAC inhibitors induce synovial sarcoma apoptosis
Romidepsin is an HDAC inhibitor that effectively suppresses synovial sarcoma in vitro and in vivo (Ito et al., 2005). To test if HDAC inhibitors induce synovial sarcoma apoptosis, the biphasic synovial sarcoma cell line SYO-1 was cultured in the presence of the membrane-impermeant dye propidium iodide (PI), and then exposed to romidepsin or vehicle. The Cellomics KineticScan Reader was used to count PI-positive cells in a 48-hr time course at 2-hr intervals. As shown in Figure 2.1 A, treatment with 1 nM romidepsin, but not DMSO, induced a time-dependent increase in PI-positive cell number, and higher
! 25! doses of 5 and 10 nM romidepsin further enhanced cell death. Similar results were also obtained using other structurally different HDAC inhibitors (MS-275, PXD101, and
SAHA), and using the monophasic synovial sarcoma cell line FUJI (data not shown).
To assess whether HDAC inhibitor-induced cell death is via apoptosis, we next carried out western blots to examine activation of the apoptotic marker, caspase-3, as reflected by an increase in its active cleaved form. After 16 hr exposure to three different
HDAC inhibitors romidepsin, MS-275, and PXD101, both SYO-1 and FUJI cells revealed a marked elevation of activated caspase-3 (Figure 2.2), indicating these cells undergo apoptosis following HDAC inhibitor treatment. Consistent with this result, flow cytometry of SYO-1 cells labeled with Annexin V and PI shows that romidepsin results in 80-90% of cells positive for Annexin V within 16 hr (Figure 2.1 B, top). Caspase-3 activation is inhibited by the broad-spectrum caspase inhibitor z-VAD-fmk (Fig. 2.1 B, bottom; compare lanes 2 and 4), and consistent with a caspase-dependent process, romidepsin-induced apoptosis is dramatically decreased in the presence of z-VAD-fmk.
Taken together, these data suggest that apoptosis is a key mechanism by which HDAC inhibitors suppress synovial sarcoma.
2.3.2 EGR1 is specifically up-regulated in HDAC inhibitor-mediated synovial sarcoma apoptosis.
To further characterize HDAC inhibitor action on tumor cells, synovial sarcoma cell lines, SYO-1 and FUJI, and human embryonic kidney HEK293 cells were stained with
Annexin V and PI, followed by 16-hr exposure to DMSO, romidepsin, or MS-275, and then subjected to flow cytometry analysis. As illustrated in Figure 2.1 C, treatment with
! 26! romidepsin and MS-275 strongly induced apoptosis in SYO-1 and FUJI cells, but not
HEK293 cells, relative to DMSO controls. To explore the molecular basis for this finding, we initially focused on the EGR1 tumor suppressor gene because (a) EGR1 functions as an early response factor in multiple apoptotic pathways, (b) EGR1 expression is markedly decreased in synovial sarcoma cell lines and tumors (Figure 2.3), and (c) EGR1 is one of the targets directly inhibited by the synovial sarcoma-associated
SS18-SSX fusion protein (Lubieniecka et al., 2008). Consistently, relative to controls,
HEK293 cells engineered to express SS18-SSX revealed a clear decrease in EGR1 transcripts, an effect reversed by romidepsin (Figure 2.4). In light of these findings, we performed real-time quantitative PCR (RT-qPCR) to monitor the alteration of EGR1 mRNA levels, and found that treatment with romidepsin or MS-275 increases EGR1 transcription in SYO-1 and FUJI cells, but not in HEK293 cells (Figure 2.1 D).
Furthermore, a similar approach was applied to stable HEK293 cell lines expressing native SS18, SS18-SSX2, or none, showing that romidepsin only stimulates activation of
EGR1 in SS18-SSX2-expressing HEK293 cells, rather than the other two HEK293- derived cell lines (Figure 2.4). Thus, EGR1 is specifically induced in synovial sarcoma during HDAC inhibitor-stimulated apoptosis. Importantly, we noticed that the elevated
EGR1 level appears to correlate with the degree of HDAC inhibitor-induced apoptosis, raising the possibility that EGR1 activation might mediate the apoptotic activity of
HDAC inhibitors.
2.3.3 EGR1 knockdown reduces HDAC inhibitor-mediated apoptosis
To evaluate the role of EGR1 in the apoptotic response to HDAC inhibitors, we performed transient transfection experiments of SYO-1 and FUJI cells with control
! 27! random or EGR1-specific siRNA. 24 hr post transfection, cells were treated with romidepsin or DMSO, and then subjected to Cellomics kinetic cell death analysis. With
EGR1 knockdown (Figure 2.5 A), si-EGR1-expressing SYO-1 cells showed a significant reduction of romidepsin-induced apoptosis (Figure 2.5 B). Similar effects were also seen using the FUJI synovial sarcoma cell line (Figure 2.6), supporting a requirement for
EGR1 reactivation in HDAC inhibitor induction of cell death.
To further confirm the aforementioned findings, we established two EGR1- knockdown SYO-1 cell lines by stably expressing EGR1-specific small hairpin RNA
(shRNA). There was no significant difference in cell morphology or growth rate between control and EGR1-knockdown cells (data not shown). Relative to control cells, EGR1 expression was maintained at low levels in both sh-EGR1-expressing cell lines after romidepsin addition, as determined by western blotting analysis (Figure 2.7). For further studies, the sh-EGR1-1 cell line was selected due to its more efficient knockdown. The time course of lentiviral-mediated EGR1 knockdown was examined by RT-qPCR analysis, which revealed a more than 70% decrease in EGR1 transcripts by 12 hr after romidepsin addition compared to control cells (Figure 2.5 C). Expression of EGR1 shRNA significantly inhibited the apoptotic effects of romidepsin (Figure 2.5 D). Similar results obtained using MS-275 (Figure 2.8) further implicate EGR1 as a key regulator of
HDAC inhibitor-mediated apoptosis.
2.3.4 EGR1 mediated synovial sarcoma apoptosis involves PTEN
To evaluate whether EGR1 is alone sufficient to induce apoptosis in synovial sarcoma, we subcloned the EGR1 cDNA upstream of an EGFP expression vector,
! 28! separated by the internal ribosome entry site (IRES)-EGFP cassette. The control (IRES-
EGFP) or EGR1-IRES-EGFP construct was then transiently transfected into SYO-1 and
FUJI cells, and the fate of EGFP-expressing cells were followed. Overexpression of
EGR1 dramatically enhanced caspase-3 activation in both cell lines (Figure 2.9 A), indicating that EGR1-expressing synovial sarcoma cells undergo apoptotic cell death. In addition, we also identified an increase in PTEN protein levels in the cultures expressing
EGR1. Given the fact that PTEN has been identified as a direct target of EGR1 in keratinocytes (Virolle et al., 2001), we sought to explore if a similar EGR1-PTEN pathway was operating in synovial sarcoma. A short region between –1118 and –858 of the human PTEN promoter has been reported to be sufficient for inducing transcriptional activation of PTEN (Ma et al., 2005). Importantly, this sequence contains three canonical
EGR1-binding sites (Virolle et al., 2001), which are highly conserved in human and mouse (Figure 2.9 B). To determine if EGR1 expressed in synovial sarcoma cells binds to the PTEN promoter, chromatin immunoprecipitation (ChIP) assays were carried out using control and EGR1-expressing SYO-1 cells. The sequence spanning EGR1-binding sites (P1) and a control upstream sequence (P2) of the human PTEN promoter were amplified by PCR. As shown in Figure 2.9 C, the sequence P1, but not P2, was strongly enriched in the immunoprecipitates obtained with EGR1 antibody from cells expressing
EGR1. Thus, EGR1 directly regulates PTEN in synovial sarcoma cells, however, the biological significance of this relationship was unclear.
To assess the functional importance of EGR1-induced activation of PTEN, we next performed transient transfection experiments of SYO-1 cells with PTEN-specific siRNA.
As shown in Figure 2.9 D, PTEN knockdown effectively depleted the ability of expressed
! 29! EGR1 to induce apoptosis. To further characterize the role of PTEN, we also overexpressed PTEN in SYO-1 and FUJI cells. Heterologous expression of PTEN led to caspase-3 activation in both tested synovial sarcoma cell lines (Figure 2.10 A). No apparent change in EGR1 protein levels was detected in cells transfected with or without the PTEN expression vector, suggesting that PTEN-induced apoptosis is downstream of
EGR1. In support of this idea, expression of PTEN resulted in ~50% apoptosis of synovial sarcoma cells in both EGR1 and control shRNA experiments (Figure 2.10 B).
Together, these results demonstrate that both EGR1 and PTEN expression are sufficient to induce synovial sarcoma cell apoptosis, and that PTEN serves as an important effector of EGR1-mediated apoptotic cell death.
2.3.5 HDAC inhibitors increase PTEN expression through EGR1
We next tested the possibility of HDAC inhibitors promoting EGR1-dependent activation of PTEN. First, RT-qPCR analysis using SYO-1 cells revealed a time- dependent increase in PTEN mRNA levels after 9-hr exposure to romidepsin. By contrast, EGR1 knockdown strongly inhibited romidepsin-induced PTEN activation, as did PTEN knockdown (Figure 2.11 A). This finding was also confirmed as the protein levels by western blotting analysis (data not shown). Furthermore, the effect of romidepsin on EGR1 expression appears to be direct, because it is unaffected by the protein synthesis inhibitor cycloheximide (CHX) (Figure 2.12 A). Conversely, romidepsin-induced activation of PTEN is sensitive to CHX (Figure 2.12 B), consistent with a requirement for de novo synthesis of EGR1.
! 30! To investigate this possibility, we performed ChIP assays using SYO-1 cells. As shown in the left panel of Figure 2.11 B, stimulation of these cells with romidepsin resulted in enhanced binding of EGR1 to EGR1-binding sites of the PTEN promoter in a time-dependent way, which was correlated with elevated EGR1 and, by 9 hr, elevated
PTEN protein levels (Figure 2.11 B, right). The same experiment was performed in FUJI cells with similar results (data not shown).
2.3.6 PTEN is critical for HDAC inhibitor-induced EGR1-dependent apoptosis
Finally, the biological importance of PTEN in the EGR1-dependent response to
HDAC inhibitors was examined. Previous studies have shown that enforced PTEN expression can reduce the levels of active phosphorylated AKT (Downes et al., 2001), and that the AKT signaling pathway is active in synovial sarcoma (Bozzi et al., 2008).
We therefore performed western blots for phosphorylated AKT at Serine 473 (p473-
AKT) to monitor the changes in AKT activity following HDAC inhibitor treatment. To this end, SYO-1 cells were transiently transfected with control random, EGR1, or PTEN siRNA. After exposure to romidepsin for 9 hr, control synovial sarcoma cells revealed a marked decrease in cellular p473-AKT levels, suggesting inactivation of the AKT pathway. However, romidepsin-mediated suppression of AKT phosphorylation was dramatically abolished by knockdown of either EGR1 or PTEN (Figure 2.13 A).
Because EGR1 transactivates PTEN expression in response to HDAC inhibitors (Figure
2.11 A), these findings implicate HDAC inhibitor-mediated expression of EGR1 inducing
PTEN, which in turn blocks the AKT survival pathway. Consistent with this, in general p473-AKT levels were observed to closely mirror that of PTEN expression in both loss
! 31! and gain-of-function experiments (Figures 2.13 A and 2.14). Similar to EGR1 silencing, knockdown of PTEN prolongs survival of SYO-1 cells treated with HDAC inhibitors
(Figure 2.13 B). Together, these results demonstrate an important role for HDAC- mediated activation of an EGR1-PTEN network that culminates in decreased activity of
AKT survival signaling and consequently synovial sarcoma apoptosis.
2.4 Discussion
Previous studies have characterized EGR1 as a tumor suppressor that functions as a transcription modulator of downstream target genes, in particular those involved in cell cycle arrest and apoptosis (Krones-Herzig et al., 2005; Liu et al., 1999; Virolle et al.,
2001; Yu et al., 2007; Yu et al., 2009). With the exception of prostate cancer, in which
EGR1 expression has been reported to be increased, most of the examined tumors typically exhibit reduced EGR1 expression; some of these include glioblastoma, lymphoma, breast and lung cancer, and more recently rhabdomyosarcoma (Huang et al.,
1997; Roeb et al., 2007; Virolle et al., 2003). In many of these tumors, the mechanism(s) underlying EGR1 down-regulation is poorly defined. Recently, we reported that the synovial sarcoma-associated SS18-SSX fusion oncoprotein dominantly inhibits EGR1 transcription through polycomb-mediated chromatin remodeling (Lubieniecka et al.,
2008). Notably, HDAC inhibitors, a new class of anti-cancer drugs (Moradei et al., 2008;
Stimson et al., 2009), can reverse this process. Herein, we demonstrate that EGR1 reactivation by HDAC inhibitors has a causal function in the induction of synovial sarcoma cell death by these compounds. Furthermore, these analyses have revealed a
! 32! critical role for EGR1 suppression by the SS18-SSX fusion oncoprotein in the etiology of synovial sarcoma.
2.4.1 EGR1 and Synovial Sarcoma Biology
EGR1 expression is reduced or absent in a large number of tumours and cell lines.
Re-expression of EGR1 in some of these cells leads to growth arrest, differentiation, or cell death, indicating that EGR1 plays a fundamental role in the biology of these tumors.
Recent studies of the PAX3-FOXO1 fusion oncoprotein in alveolar rhabdomyosarcoma, have shown that PAX3-FOXO1 inhibits cell cycle arrest by attenuating the expression of p57kip2 (CDKN1C), an EGR1 target gene (Roeb et al., 2007; Roeb et al., 2008).
Interestingly, in this tumor EGR1 expression is not affected, but the fusion oncoprotein interacts with EGR1 and promotes its degradation; neither of the wild-type constituents of the fusion oncoprotein interact with EGR1. The importance of EGR1 suppression in this tumour was highlighted by the findings that re-expression of EGR1 in a rhabdomyosarcoma-derived cell line enhanced exit from the cell cycle and subsequent cell differentiation. In synovial sarcoma, a similar picture is beginning to emerge with respect to the importance of EGR1 in the biology of this tumor type. The analysis of
HDAC inhibitor action in synovial sarcoma has provided unparalleled insights into the underlying tumorgenic mechanisms of the SS18-SSX fusion oncoprotein. In particular, reactivation of EGR1 with HDAC inhibitors or subsequent studies involving specific heterologous expression of EGR1 or its downstream target PTEN significantly increased cell death. Together these findings indicate that the SS18-SSX fusion oncoprotein through repression of EGR1 enables synovial sarcoma cells to escape normal apoptotic
! 33! cues. Thus, the fusion oncoproteins in these disparate sarcomas converge on a common target, EGR1, to subvert normal cell differentiation and death pathways.
2.4.2 EGR1-PTEN Pathway and Regulation of Apoptosis in Synovial Sarcoma
Although activation of PTEN greatly contributes to the apoptotic response of cancer cells to HDAC inhibitor treatment, the exact mechanism underlying this process has not been defined. Because PTEN expression can be directly regulated by the EGR1 transcription factor (Virolle et al., 2001), we speculated that activation of EGR1 by
HDAC inhibitors might elevate expression of PTEN and in turn stimulate apoptosis. In the current study, we show that HDAC inhibitors effectively up-regulate the expression of PTEN in wild-type, but not EGR1-knockdown, synovial sarcoma cells. Of note, after addition of HDAC inhibitors, the EGR1 protein time-dependently binds to a functional cis-acting element in the PTEN promoter, which tightly correlates with an increase in
PTEN mRNA and protein levels. As in the case of EGR1, a significant decrease in
HDAC inhibitor-induced apoptosis is observed in PTEN-knockdown synovial sarcoma cells. Thus, direct regulation of PTEN by EGR1 at least partly explains the ability of
HDAC inhibitors to induce apoptosis. Along with activation of PTEN, we also found that
AKT phosphorylation is down-regulated by HDAC inhibitor treatment, but is strongly protected by repression of EGR1. Given that the PI3K/AKT survival pathway is biologically active in synovial sarcoma (Bozzi et al., 2008), this observation further supports the functional significance of PTEN in an EGR1-dependent apoptotic response to HDAC inhibitors. In fact, a similar regulatory mechanism whereby EGR1 directly transactivates PTEN has been discovered under stressors such as radiation and etoposide
! 34! in other model systems (Kim et al., 2006; Virolle et al., 2001; Yu et al., 2009). It is tempting to speculate that the connection between EGR1 and PTEN may be a common mechanism underlying the apoptotic response of many human tumors to radiation and chemotherapy.
Although we identify an HDAC inhibitor-induced pathway for PTEN induction by
EGR1, we cannot rule out the possibility that EGR1 may also suppress tumor growth through one or more PTEN-independent pathways following HDAC inhibitor treatment.
Consistent with this, efficient knockdown of PTEN (Fig. 5A) only leads to an ~ 50% reduction in cell death (Fig. 6B), indicating that other pathways in addition to PTEN are likely contributing to HDAC-induced cell death. Earlier studies have shown that EGR1 functionally activates the p53-mediated cell death pathway both in vitro and in vivo
(Krones-Herzig et al., 2005; Nair et al., 1997). However, under our conditions, the
HDAC inhibitor romidepsin does not markedly affect the expression of either p53 or its negative regulator MDM2 in synovial sarcoma cells (data not shown). This could be consistent with the finding that HDAC inhibitor-mediated acetylation of the EGR1 protein reduces its capacity for p53 transactivation (Yu et al., 2004). Interestingly, we noticed a clear increase in C-terminal modifications of p53 after addition of romidepsin
(data not shown). In fact, six lysine residues have been identified at the C-terminus of human p53, all of which can be acetylated by many HDAC inhibitors (Condorelli et al.,
2008; Roy and Tenniswood, 2007; Zhao et al., 2006). Growing evidence suggests that acetylated p53 physically associates with EGR1, which in turn enhances the stability and transcriptional activation capacity of p53 (Habold et al., 2008; Liu et al., 2001). Future studies will assess whether p53 is involved in HDAC inhibitor-induced EGR1-dependent
! 35! apoptosis of synovial sarcoma and how p53 regulates downstream target genes in the
EGR1network.
In summary, our present studies demonstrate a central role of the transcription factor
EGR1 in synovial sarcoma cell survival and HDAC inhibitor-induced apoptotic effects on synovial sarcoma. Given that current doxorubicin/ifosphamide-based chemotherapy has limited benefit in synovial sarcoma patients (Italiano et al., 2009), our findings suggest that HDAC inhibitors are worth evaluating as a targeted inhibitor of the oncogenic effects of SS18-SSX through restoring EGR1 and PTEN expression.
! 36!
Figure 2.1 – EGR1 is specifically increased in synovial sarcoma during HDAC inhibitor- mediated apoptosis
(A) Cellomics kinetic cell death analysis of SYO-1 cells treated with romidepsin (1, 5, or 10 nM) or 0.1% DMSO. Using PI staining, non-viable cells were detected and counted at 2-hr intervals. Data are representative of three individual experiments; bars, 95% CI. (B, top) Average percentage of apoptotic SYO-1 cells incubated with or without 40 µM z-VAD- fmk and treated with 10 nM romidepsin or 0.1% DMSO for 16 hr from three individual experiments; bottom, western blots of activated caspase-3 expression in SYO-1 cells in the treatment as described above. (C) Flow cytometry analysis of Annexin V-positive HEK293, SYO-1, and FUJI cells treated with 10 nM romidepsin, 10 µM MS-275, or 0.1% DMSO for 16 hr. Data are representative of three independent assays; bars, 95% CI. (D) Real-time quantitative PCR (RT-qPCR) analysis of the EGR1 transcription in HEK293, SYO-1, and FUJI cells treated with 10 nM romidepsin, 10 µM MS-275, or DMSO for 12 hr. Columns, mean of two independent experiments; bars, 95% CI.
! 37!
Figure 2.2 – HDAC inhibitors activate caspase-3 in synovial sarcoma cells
Western blot analysis was performed to assess the levels of full and cleaved caspase-3 in SYO-1 and FUJI cells treated with 10 nM romidepsin, 10 µM MS-275, 1 µM PXD101, or DMSO for 16 hr. Actin is used as a loading control.
Figure 2.3 – The protein level of EGR1 in human embryonic kidney HEK293 and synovial sarcoma SYO-1 and FUJI cells
Western blots against EGR1 were done using HEK293, SYO-1, and FUJI cells; the fusion oncoprotein SS18-SSX2 was detected as a marker for synovial sarcoma. Actin is a loading control.
! 38!
Figure 2.4 – EGR1 is specifically induced by romidepsin in stable HEK293 cells expressing the SS18-SSX2 oncoprotein
RT-qPCR analysis of EGR1 transcription was done using stable HEK293 cell lines expressing Myc-tagged His, SS18, or SS18-SSX2 with 9 hr romidepsin treatment. 18S rRNA is applied as an internal control.
! 39!
Figure 2.5 – Knockdown of EGR1 decreases HDAC inhibitor-mediated apoptosis
(A) RT-qPCR analysis of the EGR1 transcription in SYO-1 cells transfected with EGR1-specific or control random siRNA for 24 hr. (B) Cellomics kinetic cell death analysis of SYO-1 cells after siRNA transfection followed by 48- hr incubation with 10 nM romidepsin or DMSO. Dots, mean of three individual assays; bars, 95% CI. (C) Relative amount of the EGR1 mRNA as determined by RT-qPCR in control and EGR1- knockdown SYO-1 cell lines treated with 10 nM romidepsin or DMSO for 12 hr. (D) Average percentage of PI-positive control and EGR1-knockdown SYO-1 cells after 32-hr incubation with 10 nM romidepsin or DMSO from three independent experiments.
! 40!
Figure 2.6 – EGR1 knockdown reduces FUJI cell killing by romidepsin treatment
(A) EGR1 mRNA levels were determined by RT-qPCR using control or EGR1-knockdown FUJI cells. (B) Cellomics analysis of FUJI cells transfected with control or EGR1 siRNA in the 48-hr treatment of romidepsin or DMSO. Data are representative of three independent experiments; bars, 95% CI.
Figure 2.7 – Knockdown of EGR1 in the synovial sarcoma cell lines SYO-1
SYO-1 cells were transduced with or without sh-Control or sh-EGR1 lentiviral construct as described in Materials and Methods. EGR1 and actin protein levels were assessed by western blot analysis 6 hr after romidepsin addition.
! 41!
Figure 2.8 – EGR1 knockdown reduces the sensitivity of SYO-1 cells to MS-275-induced apoptosis
(A) RT-qPCR analysis of EGR1 transcripts in sh-Control and sh-EGR1 SYO-1 cells treated with 10 µM MS-275 or 0.1% DMSO for 12 hr. (B) average percentage of control or EGR1-knockdown cells positive for PI after 32 hr post addition of MS-275 or DMSO.
! 42!
Figure 2.9 – PTEN is activated in EGR1-induced apoptosis
(A) Western blots of EGR1, PTEN, and cleaved caspase-3 expression in both SYO-1 and FUJI cells after 48-hr transfection with or without the empty or EGR1 expression vector. Actin is used as a loading control. (B) Schematic of the human PTEN promoter, and location of the PCR primers specific for EGR1- binding sites (P1, -978 to -867) and an upstream sequence (P2, -1703 to -1477). The arrow indicates the translational start site. (C) Chromatin immunoprecipitation (ChIP) assays were performed using SYO-1 cells after 48-hr transfection with or without the EGR1 vector. Chromatin was immunoprecipitated with anti- EGR1 or control rabbit IgG antibody as indicated. P1 and P2 sequences within the PTEN promoter were amplified by PCR. (D) Average percentage of apoptotic SYO-1 cells expressing control or PTEN siRNA after 48-hr transfection with or without EGR1 cDNA. Columns, mean of three individual assays; bars, 95% CI.
! 43!
Figure 2.10 – Effect of enforced PTEN expression on synovial sarcoma cells
(A) Western blot analysis of SYO-1 and FUJI cells transfected with or without the empty vector or PTEN expression construct for 48 hr. (B, top) Average percentage of apoptosis in control and EGR1-knockdown SYO-1 cells transfected with or without PTEN cDNA for 32 hr; bottom, the EGR1, PTEN, and actin protein levels were determined by western blots in SYO-1 cultures as described above.
Figure 2.11 – EGR1 is required for HDAC inhibitor-mediated PTEN induction
(A) RT-qPCR analysis of the PTEN transcription (normalized to 18S rRNA) in SYO-1 cells expressing control, EGR1, or PTEN siRNA after 9 hr romidepsin addition. Shown is the average of two independent experiments. (B, left) ChIP assay using the P1 primers on SYO-1 cells treated with 10 nM romidepsin for 9 hr; right, western blots of EGR1 and PTEN expression in SYO-1 treated as described above.
! 44!
Figure 2.12 – Romidepsin directly induces EGR1, but not PTEN, in synovial sarcoma cells
(A) RT-qPCR analysis of EGR1 mRNA levels was performed using SYO-1 cells incubated with or without cycloheximide (CHX) for 30 min followed by romidepsin treatment. (B) RT-qPCR was done in the same cultures as described above.
! 45!
Figure 2.13 – Suppression of EGR1 or PTEN attenuates HDAC inhibitor-stimulated cell death
(A) The levels of phosphorylated AKT (Ser 473) and total AKT were assessed by western blots using control, EGR1-, or PTEN-knockdown SYO-1 cells incubated with romidepsin for 12 hr. (B) Average percentage of PI-positive SYO-1 cells transfected with control, EGR1, or PTEN siRNA followed by 32-hr romidepsin treatment. Shown is mean of three individual experiments.
! 46! ! ! ! ! ! !
Figure 2.14 – The effect of EGR1 and PTEN on AKT phosphorylation in synovial sarcoma cells
(A) The levels of p473-AKT and total AKT were assessed by western blots using control, EGR1-, and PTEN-expressing SYO-1 cells. The transfection efficiency was also examined by western blots, and β-Actin was indicated as a loading control. (B) Western blotting analysis of PTEN protein levels was carried out on SYO-1 samples transfected with the indicated siRNAs and following treatment with romidepsin.
! 47! Chapter 3: Deconstruction of the SS18-SSX Fusion Oncoprotein
Complex: Insights into Disease Etiology and Therapeutics
3.1 Introduction
Synovial sarcoma is an aggressive soft-tissue tumor of adolescents and young adults
(Haldar et al., 2008). Histologically, these tumors can display monophasic (spindle shaped mesenchymal cells), biphasic (similar but with focal epithelial differentiation) or poorly differentiated (small blue round cells generic with some other translocation- associated sarcomas) morphology. Treatment consists of wide local tumor excision and radiation, which cures local disease. Metastatic disease is usually fatal despite treatment with conventional chemotherapy agents such as doxorubicin and ifosphamide, which confer at best a temporary response.
Almost all synovial sarcomas carry a demonstrable, pathognomonic t(X;18) reciprocal translocation fusing SS18 to an SSX gene. Clinical diagnosis can be molecularly confirmed by the identification of this event by karyotyping, RT-PCR or
FISH techniques, although recently TLE1 has emerged as a useful immunohistochemical marker that may obviate the need to resort to molecular testing (Jagdis et al., 2009). A variety of studies have shown that the resulting SS18-SSX fusion functions as an oncoprotein; heterologous expression induces transformation of rat fibroblasts, and continued expression is needed for tumor cell survival (Nagai et al., 2001). Most convincingly, in transgenic mice conditional overexpression of SS18-SSX2 in the myogenic progenitor compartment, but not other compartments, leads to the appearance of both monophasic and biphasic synovial sarcoma tumors with full penetrance (Haldar et
! 48! al., 2007). Together, these studies indicate that the SS18-SSX fusion protein exhibits oncogenic activity and is both necessary and sufficient for tumorigenesis.
The SS18-SSX fusion protein retains a C-terminal repressor domain from either of two highly similar cancer-testis antigens, SSX1 or SSX2 (SSX4 has also been reported in rare cases), which is fused to the N-terminus of SS18, a transcriptional coactivator
(Ladanyi, 2001). The resulting fusion proteins SS18-SSX1 and SS18-SSX2 have no apparent DNA-binding motif, yet appear to function predominantly in transcriptional regulation (Lim et al., 1998). The control of gene expression by SS18-SSX is believed to involve chromatin remodeling, due to its colocalization with both Trithorax (TrxG) and
Polycomb group (PcG) complexes, maintaining chromatin in a poised bivalent state (de
Bruijn et al., 2006; Lubieniecka et al., 2008; Soulez et al., 1999). Similar to other sarcoma-associated fusion oncoproteins, expression of SS18-SSX contributes to aberrant transcriptional activity and dysregulated gene expression. Since SS18-SSX itself lacks direct DNA-binding domains or activity, it has been challenging to identify target genes or to decipher its mechanism of action. In this report, we explore the mechanism of
SS18-SSX-mediated repression and its connection with the anti-tumor action of HDAC inhibitors by identifying the key constituents of SS18-SSX transcriptional complexes in synovial sarcoma.
3.2 Materials and Methods
3.2.1 Cells, Tissues and Chemicals
Human synovial sarcoma cell lines SYO-1 and FUJI were kindly provided by Dr.
Akira Kawai (National Cancer Centre Hospital, Tokyo, Japan) and Dr. Kazuo Nagashima
! 49! (Hokkaido University School of Medicine, Sapporo, Japan) and maintained in RPMI-
1640 medium with 10% fetal bovine serum (FBS) (Invitrogen). Human embryonic kidney HEK293 cells stably expressing Myc-tagged SS18 or SS18-SSX2 were grown in
DMEM medium with 10% FBS and 400ug/ml Zeocin (Invitrogen). Primary mouse synovial sarcoma cells were isolated from tumors of female Myf5-Cre/SSM2 mice as described previously (Haldar et al., 2007), and cultured in DMEM medium with 10%
FBS. All cells were maintained at 37°C, 95% humidity, and 5% CO2.
Human subjects in this study provided informed consent for use of tissues for research purposes following procedures approved by the Clinical Research Ethics Board of the University of British Columbia (projects H08-0717 "Sarcoma tissue bank" and
H06-00013 "Molecular targets for therapy of sarcoma")
HDAC inhibitors romidepsin (FK228, Depsipeptide, or NSC-630176) and SB939 were obtained from the Developmental Therapeutic Branch of the National Cancer
Institute (Bethesda, MD, USA) and S*BIO Pte Ltd (Singapore), respectively. DMSO was purchased from Sigma-Aldrich.
3.2.2 Plasmid DNA Constructs
To define the domains within SS18-SSX2 that interact with ATF2 and TLE1, variants missing the SNH or QPGY domain of SS18 and the SSXRD domain of SSX2 were generated via gene synthesis (Integrated DNA Technologies) and sub-cloned as
EcoR1 – Not1 fragments into the mammalian expression vector pcDNA4/myc-HisA
(Life Technologies). All genes were engineered to remove the stop, allowing read- through to generate a C-terminal Myc-6XHis tag.
! 50!
3.2.3 Immunoprecipitation (IP) and Western Blots
For immunoprecipitation, cells were washed twice with ice-cold PBS, and incubated with RIPA buffer (Santa Cruz Biotechnology) for 35 min on ice. Whole cell lysates were centrifuged at 4°C, at full speed in a microcentrifuge for 15 min, and the supernatants were mixed with 15 ul of protein A/G agarose beads (Santa Cruz Biotechnology) for 45 min at 4°C for pre-clearing. For immunoprecipitation, 500 ug of pre-cleared proteins were incubated with 1.5 ug of indicated antibody at 4°C overnight, followed by the addition of 25 ul of protein A/G agarose beads. After a 3-hr incubation at 4°C, the beads were precipitated, washed once with RIPA buffer and twice with ice-cold PBS, and boiled in 2×loading dye for 5 min. Samples were separated by 10-12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and transferred to nitrocellulose membranes (Bio-Rad Laboratories). Blots were incubated with indicated antibodies (see below for details). Signals were visualized using the Odyssey Infrared System (LI-COR
Biosciences).
3.2.4 Mass Spectrometry
Coomassie blue stained bands were excised from the gel and reduced with dithiothreitol (DTT), followed by alkylation with iodoacetamide (IAA). Gel bands were digested with Trypsin at 37°C overnight as described (Shevchenko et al., 1996).
Proteolytically digested peptides were then extracted from the gel pieces, reconstituted in formic acid (FA), and analyzed on a QStar XL LC-MS/MS (Applied Biosystems). The
MS/MS peaks were submitted to Mascot and gpmDB for peptide sequence database
! 51! search (Wong et al., 2009). Both of these databases were employed to confirm peptide/protein identification in this study.
3.2.5 Immunofluorescence and Immunohistochemistry
SYO-1 cells were cultured on glass coverslips, fixed with 3:1 acetone-methanol at -
20°C for 7 min, and blocked with 5% bovine serum albumin (BSA) for 30 min. Cells were then incubated with a polyclonal ATF2 rabbit antibody (Santa Cruz Biotechnology) at 4°C overnight, followed by three washes with ice-cold PBS. After incubation with
AlexFluor Conjugated anti-Rabbit secondary antibody (New England Biolabs), the coverslips were mounted in 50% glycerol and 2% DABCO (Sigma-Aldrich). The cellular localization of ATF2 was analysed under a fluorescence microscope (Zeiss).
Primary synovial sarcoma, malignant peripheral nerve sheath and breast cancer samples were embedded in paraffin, and stained with the same ATF2 antibody used in immunofluorescence. All immunostainings were performed with avidin-biotin- peroxidase complex technique (VectaStain) in combination with diaminobenzidine
(DAB), and counterstained with hematoxylin and eosin (H&E) on surgical pathology specimens as described previously (Pacheco et al., 2010; Terry et al., 2007). Negative controls were carried out with rabbit IgG.
3.2.6 Glycerol-Gradient Sedimentation
Nuclear extracts were prepared from SYO-1 and HEK293 stable cell lines using the Pierce NE-PER Nuclear Extraction kit. Samples were then subjected to a 10%-40% glycerol gradient in 4.8 ml buffer (150 mM NaCl, 10 mM HEPES (pH 7.5), 2 mM DTT,
! 52! 1 mM EDTA, and 0.1% Triton X-100), and centrifuged at 40,000 rpm for 16 hr at 4°C in a SW50.1 rotor (Beckman). Fractions (160 ul) were collected starting from the top of the gradient, followed by SDS-PAGE and western blot analysis. To determine molecular weight of fractions, marker proteins thyroglobulin (M. Wt. 669 kDa), β-amylase (M.Wt.
200 kDa) and bovine serum albumin (M.Wt. 66 kDa) were spiked into the gradients and detected using their respective antibodies.
3.2.7 RNA Interference (RNAi)
The small interfering RNAs (siRNAs) specific for ATF2/mATF2, TLE1/mTLE1, and HDAC1 were purchased from Dharmacon and Santa Cruz Biotechnology, respectively. Two different SS18-SSX2 siRNAs were synthesized by Integrated DNA
Technologies (IDT) as described in previous studies (Garcia et al., 2011; Lubieniecka et al., 2008). At 60% confluence, cells were transfected with the indicated siRNA using
Lipofectamine RNAiMAX transfection reagent (Invitrogen) according to the manufacturer’s instructions. Except where indicated, lysates or RNA were harvested 48 hr post-transfection, and used for IP, glycerol gradients, reporter gene assays, RT-qPCR and western blots. Knockdown efficiency was determined by RT-qPCR.
3.2.8 Cell Growth and Colony Formation Assay
To measure cell growth rate, human and mouse synovial sarcoma cells were cultured at 60% confluence on 48-well plates, and transfected with the indicated siRNA using Lipofectamine RNAiMAX transfection reagent (Invitrogen). At various times after transfection, cell growth was monitored by MTT assay (Life Technologies), and
! 53! normalized to control cells to give relative growth rate of cells. For colony formation assay, control and knockdown cells were replated on 6-well plates at a density of 1×103 cells per well. After 8 days of incubation, cells were fixed with 10% formalin and stained with 0.1% crystal violet, and the colonies counted by using Image J software as described
(Junttila et al., 2007).
3.2.9 Cell Death and Apoptosis Assay
Cells were cultured with propidium iodide (PI) at a concentration of 500 ng/ml, followed by transfection with the indicated siRNA. Cell death was indicated by PI- positive cells, and visualized under a fluorescence microscope (Zeiss). For analysis of apoptosis, cells were harvested 72 hr post siRNA transfection, and suspended in
Annexin-V-FITC-PI dye (Invitrogen). After adding Annexin-V binding buffer, the samples were run through a FACS scan flow cytometer (Becton Dickinson) as described
(Kawase et al., 2009). Summit for MoFlo Acquisition and Sort Control Software was used to quantify apoptosis (Annexin-V positive cells).
3.2.10 Chromatin Immunoprecipitation (ChIP)
ChIP experiments were performed following the Active Motif protocol as described (Su et al., 2010). Briefly, 5×107 cells or 150 mg synovial sarcoma tissues were cross-linked with 1% formaldehyde prior to lysis and homogenization. Cross-linked
DNA was sheared using a Bioruptor-UCD300 sonicator (Diagenode) for 15 × 25 second pulses (60 second pause between pulses) at 4°C. After centrifugation, the supernatants were pre-cleared with Protein G beads for 30 min at 4°C, and incubated with the
! 54! indicated antibody at 4°C overnight. After 4-hr incubation with Protein G beads, the precipitates were washed four times with different washing buffers (Active Motif), eluted with 1% SDS, and incubated at 65°C overnight to reverse cross-linking. ChIP-enriched
DNA was purified using the Qiagen PCR Purification kit, and subjected to SYBR Green qPCR analysis (Roche) using various primer sets (supplemental information).
3.2.11 Electrophoretic Mobility Shift Assay (EMSA)
The ATF/CRE probe was purchased from LI-COR, and labeled on the 5’-end of each strand with Infrared Dye-700nm. The wild-type and mutant ATF/CRE competitor probes were obtained from Santa Cruz Biotechnology. Binding reactions were peformed in the dark at room temperature for 30 min in 25 ul of EMSA buffer (250 mM NaCl, 20 mM HEPES (pH 7.9), 2 mM DTT, 20% glycerol, 0.5% Tween 20) as described before
(Boyle et al., 2009). Samples were separated on 4% polyacrylamide gels (29.2:0.8 acrylamide-bisacrylamide in 100 mM Tris, 100 mM borate, and 10 mM EDTA). The extent of gel shift was then visualized on the Odyssey Infrared scanner (LI-COR).
3.2.12 Luciferase Reporter Assay
For luciferase reporter assays, transient transfections were performed using
FuGENE 6 transfection reagent (Roche). SYO-1 cells were subcultured in 24-well plates, and transfected with the wild-type or mutant human ATF3 promoter-firefly luciferase reporter plasmid together with renilla luciferase expression vector. After 24-hr incubation, the control or SS18-SSX2-specific siRNA was introduced using
Lipofectamine RNAiMAX transfection reagent (Invitrogen). Cells were harvested at 48
! 55! hrs post siRNA transfection, and analyzed using the Dual-Luciferase Reporter Assay system (Promega). Firefly luciferase was normalized to renilla luciferase activity to control for differences in transfection efficiency and to generate relative luciferase activity.
3.2.13 Real-time quantitative PCR
Total RNA was isolated and then transcribed to cDNA using the Qiagen RNeasy
Mini kit and the high-capacity cDNA reverse transcription kit (Applied Biosystems), respectively, as described previously (Su et al., 2010). Taqman gene expression assays were performed by using the ABI-7500 Fast Real-Time PCR System with specific primer/probe sets (Applied Biosystems). All transcript levels were normalized to 18s ribosomal (rRNA) expression.
3.2.14 Antibodies
The rabbit polyclonal antibody (RA2009) against SS18-SSX was kindly provided by Dr. Diederik R.H. de Bruijn (Radbound University Nijmegen Medical Centre,
Nijmegen, the Netherlands). The antibodies for SS18 (H-80), SSX (C-9), ATF2 (C-19),
TLE1 (M-101 and N-18), HDAC1 (10E2), β-Actin (N-21), Caspase-3 (H-277), and
EGR1 (588) were purchased from Santa Cruz Biotechnology. The H3K4me3 (Upstate
05-745) and H3k27me3 (Upstate 07-449) antibodies were used for chromatin immunoprecipitation. For immunoprecipitation and western blots, we also used the following antibodies: Myc (Cell Signaling #2278), HDAC1 (Abcam ab1767), EZH2
! 56! (Active Motif #39639), EED (Abcam ab4469), SUZ12 (Abcam ab12073), Caspase-3
(Cell Signaling #9668), and GFP (Cell Signaling #2555).
3.3 Results
3.3.1 Identification of ATF2 and TLE1 within an SS18-SSX complex
To study transcriptional regulation governed by SS18-SSX, we used a validated antibody (RA2009, Figure 3.2 A) to isolate endogenous SS18-SSX2 and its interactants from human synovial sarcoma SYO-1 cells (Figure 3.1 A). Mass spectroscopy further confirmed the presence of SS18-SSX2 (Figure 3.2 B) and identified several known cofactors, including histone deacetylases (Figure 3.2 C). This approach also allowed us to capture multiple peptides corresponding to two previously uncharacterized components,
ATF2 and TLE1 (Figure 3.2 C). Both of these are master transcriptional regulators that are highly conserved across different species. ATF2 is a DNA-binding protein that recognizes the cAMP-responsive element (CRE) via its leucine zipper domain and recruits histone acetyltransferases (HATs) to increase transcription (Kawasaki et al.,
2000). However, the other component TLE1 is a Groucho co-repressor that usually interacts with transcriptional activators and functions in a dominant-negative manner to inhibit transcription (Ali et al., 2010). TLE1 is known to be highly expressed in synovial sarcoma (Terry et al., 2007) and has recently been demonstrated to be a robust diagnostic marker for synovial sarcoma, although its biological function in this disease has been unclear (Foo et al., 2011; Jagdis et al., 2009; Knosel et al., 2010).
To validate the proteomic data, immunoprecipitation (IP) was performed in two human synovial sarcoma cell lines (SYO-1 and FUJI), and this shows that both ATF2 and
! 57! TLE1 are specifically precipitated with anti-SS18-SSX, but not with rabbit IgG (Figures
3.1 B and 3.2 D). Interaction of SS18-SSX2 with both ATF2 and TLE1 was preserved in the presence of ethidium bromide (EtBr, Figure 3.1 B), which suggests that this fusion oncoprotein complex forms independently of DNA. ATF2 and TLE1 association with both SS18-SSX1 and SS18-SSX2 fusion proteins was verified by reciprocal IP using
RA2009, ATF2, and TLE1 antibodies (Figure 3.1 C) using patient primary tumors confirmed to express SS18-SSX1 and SS18-SSX2 (Figure 3.2 E). Importantly, we find that the mouse homologs of ATF2 (mATF2) and TLE1 (mTLE1/Grg1) are also bound to the human fusion protein in cell cultures derived from tumors from SS18-SSX2 conditional overexpression mice (Figure 3.1 C) (Haldar et al., 2007). The specificity of
ATF2-TLE1 association was confirmed by reciprocal IP using a clear cell sarcoma cell line (DTC-1) where, in the absence of the SS18-SSX fusion oncoprotein, ATF2 and
TLE1 no longer co-immunoprecipitated (Figure 3.2 F). This raised the possibility that
SS18-SSX serves as a scaffold to link ATF2 and TLE1. Indeed, glycerol-gradient fractionation on human synovial sarcoma SYO-1 cells revealed a co-elution profile of
ATF2 and TLE1 with SS18-SSX2 (Figures 3.1 D and 3.2 G), indicating that ATF2 and
TLE1 occur in the same SS18-SSX complex. We also observed several fundamental chromatin-remodeling factors (SMARCA2, HDAC1 and EZH2) in a major overlapping peak with SS18-SSX2 (Figure 3.2 G). To obtain further evidence of this disease-specific abnormal association of ATF2 with TLE1, a small interfering RNA (siRNA)-based method was used to deplete endogenous SS18-SSX2 and its complex components ATF2 and TLE1 in human SYO-1 and mouse synovial sarcoma cells (Figures 3.4 A and B), and the consequences on cell survival was assessed. Similar to SS18-SSX2 knockdown, both
! 58! ATF2 and TLE1 silencing reduces synovial sarcoma cell growth (Figures 3.3 A and 3.4
C) and impairs the ability of human and mouse tumor cells to form colonies (Figures 3.3
B and 3.4D). These knockdown cells appear to undergo apoptosis since depletion of either ATF2 or TLE1 induces an enrichment in the Annexin-V+ fraction (Figure 3.3 C) and also stimulates Caspase-3 activation (Figure 3.4 E). Together, these data demonstrate that ATF2 and TLE1 functionally associate with SS18-SSX to form an endogenous complex in synovial sarcoma important for tumor cell survival.
4.3.2 Copurification of ATF2 and TLE1 requires SS18-SSX
To gain molecular insights into SS18-SSX complex assembly, reciprocal IP was performed on human SYO-1 cells transfected with non-specific, SS18-SSX2, ATF2, or
TLE1 siRNA. Western blot analysis shows that ATF2 and TLE1 co-immunoprecipitation is dependent upon SS18-SSX2 (Figure 3.5 A). By contrast, recruitment of ATF2 and
TLE1 to SS18-SSX2 seems to be independent of each other because depletion of ATF2
(or TLE1) has no significant impact on SS18-SSX2 association with TLE1 (or ATF2)
(Figure 3.5 B). To further confirm binding specificity, HEK293 cell lines stably expressing Myc-tagged SS18, SS18-SSX2, or empty vector were generated (Figure 3.6
A). Analysis of the anti-Myc-tag precipitates reveals the co-existence of ATF2 and TLE1 with recombinant SS18-SSX2 (Figure 3.5 C). The lack of TLE1, but not ATF2, in the
Myc-SS18 precipitates (Figure 3.5 C) indicates that ATF2 and TLE1 recruitment involves different protein domains of SS18-SSX2. Reciprocal IP of ATF2 and TLE1 also supports this concept by showing that their connection depends on the presence of SS18-
SSX2 and does not occur with SS18 alone (Figures 3.6 B and C). Consistent with these
! 59! data, we find that compared to control cells, ATF2 and TLE1 migrate as individual glycerol-gradient peaks in SS18-SSX2-knockdown cells (Figure 3.5 D), implying that they are not found in a shared complex in the absence of SS18-SSX. The shared change in ATF2 and TLE1 distribution in glycerol-gradient sedimentation was also observed in
HEK293 stable cell lines with and without the fusion oncoprotein (Figure 3.6 D). To address which domains of SS18-SSX are responsible for ATF2 and TLE1 binding, we next generated SS18-SSX2 deletion mutants (Figures 3.5 E and 3.6 E) (Nagai et al.,
2001) and performed reciprocal IP using the antibodies specific to Myc-tag, ATF2, and
TLE1 in HEK293 cells. The results suggest that the N-terminal SNH (SYT N-terminal homolog) domain is responsible for the interaction of SS18-SSX2 with ATF2, whereas
TLE1 specifically interacts with the repressor domain (SSXRD) of SS18-SSX2 (Figure
3.5 F). In aggregate, these data further reinforce that S18-SSX fusion oncoprotein serves as a scaffold protein to bridge the Groucho corepressor TLE1 to transcription factor
ATF2 in synovial sarcoma (Figure 3.5 G).
4.3.3 SS18-SSX/TLE1 functions to repress ATF2 target gene expression
Recent studies have identified the tumor suppressor Early Growth Response 1
(EGR1) as a direct target of SS18-SSX (Lubieniecka et al., 2008). This gene was used to study the mechanisms underlying SS18-SSX occupancy of its targets. Chromatin IP
(ChIP) with antibodies to ATF2, TLE1 and SS18-SSX identifies a common occupied
DNA region around 100 bp upstream of the transcription start site of the human EGR1 locus (Figure 3.7 A). Sequence analysis of this promoter area reveals a consensus CRE site (5’-TCACGTCA-3’), which has been well defined in previous studies as a putative
! 60! ATF2-binding element, and is phylogenetically conserved across diverse species (Faour et al., 2005; Hayakawa et al., 2004). This indicates that the transcription factor ATF2 may have a critical role in the recruitment of the SS18-SSX complex to target promoters.
To test this possibility, we first examined the ATF2 cellular location because ATF2 has been shown to dynamically shuttle between the nucleus and cytoplasm in a context- dependent manner (Bhoumik et al., 2008; Liu et al., 2006; Maekawa et al., 2007).
Immunohistochemical and immunofluorescent analysis of ATF2 in patient synovial sarcoma specimens and SYO-1 cells, respectively, shows that ATF2 is predominantly located in the nucleus (Figures 3.7 B and C). To examine the transcriptional activity of
ATF2, published and in-house microarray expression profiles of patient specimens were interrogated (Baird et al., 2005; Nakayama et al., 2010; Nielsen et al., 2002) for the expression of known ATF2 target genes. In addition to two known SS18-SSX targets
EGR1 and Nuclear Protein 1 (NUPR1, or Candidate of Metastasis 1, COM1) (Ishida et al., 2007), a set of seven more genes (Figure 3.8 A) was chosen for further investigation because their promoters contain validated CRE sites for ATF2 binding (Figure 3.9 A)
(Hayakawa et al., 2004). These CRE sites are also conserved between humans and mice, and their protein products are involved in controlling cell cycle, apoptosis, and other cellular signaling pathways (Lopez-Bergami et al., 2010). To validate these candidate genes, ChIP was performed on SS18-SSX1- and SS18-SSX2-positive clinical tumor frozen tissue specimens. Site-specific qPCR shows both SS18-SSX1 and -SSX2 fusion proteins, together with ATF2 and TLE1, bind to the CRE-containing regions (Figures 3.8
B and C, and Figures 3.9 B and C). However, we were unable to detect any non-specific recruitment of these factors (Figures 3.8 B and C, and Figures 3.9 B and C), implying a
! 61! possible predominant role for the ATF2-binding element in directing SS18-SSX promoter occupancy. To further evaluate SS18-SSX DNA binding activity, nuclear proteins were extracted from HEK293 stable cell lines with and without the fusion oncoprotein and incubated with infrared dye-labeled CRE oligonucleotides. An electrophoretic mobility shift assay (EMSA) identifies a specific protein-DNA complex, which is supershifted by the antibody against Myc-tag in Myc-SS18-SSX2-expressing cells, but not in control cells (Figure 3.8 D). Consistently, a similar protein-DNA complex is also observed in human SYO-1 cells where it is supershifted by the antibodies to SS18-SSX, ATF2 and
TLE1 (Figure 3.8 E).
To further establish a direct link between ATF2 and recruitment of SS18-SSX, we used a specific siRNA to reduce the expression of ATF2 in human SYO-1 cells (Figure
3.11 A). ChIP analyses reveal that loss of ATF2 significantly compromises the association of SS18-SSX2 and TLE1 with target gene promoters (Figure 3.10 A).
Furthermore, RT-qPCR analysis shows that transcript abundance of multiple ATF2 targets is increased after depleting ATF2 or SS18-SSX2 (Figure 3.10 B). Consistent with this, in the mouse model of synovial sarcoma, SS18-SSX2 and TLE1 binding to target gene promoters is abrogated after ATF2 depletion (Figures 3.10 C and 3.11 B). Notably, an increased transcript level of either Egr1 or Atf3 was also observed in mATF2- and
SS18-SSX2-knockdown mouse synovial sarcoma cells (Figure 3.10 D). To confirm the specificity of this effect, wild-type (wt) or dominant-negative (dn) ATF2 was transfected into HEK293 cells in the presence or absence of the fusion protein SS18-SSX2. As shown in Figure S5C, compared with the dn form, overexpression of wt ATF2 in control cells significantly increases EGR1 and ATF3 transcript levels. However, this effect is no
! 62! longer observed in SS18-SSX2-expressing cells, indicating in the presence of SS18-
SSX2 ATF2 transactivational activity is reduced. In agreement with the RT-qPCR data, transfection of an ATF3 reporter gene in human SYO-1 cells shows that the promoter activity for this ATF2 target gene is increased ~ 4-fold after SS18-SSX2 depletion, while this stimulation is not seen in a construct with two point mutations in the CRE site of the
ATF3 promoter (Figure 3.10 E). Thus, these experiments demonstrate that the SS18-SSX complex occupies ATF2 target genes and this is dependent upon its interaction with
ATF2.
TLE1 also appears to be a functionally-important component of the SS18-SSX complex (Figures 3.3 A, B and C). To assess whether TLE1 influences SS18-SSX transcriptional activity, TLE1 was knocked down in synovial sarcoma cells. Unlike ATF2 knockdown, depletion of TLE1 affects neither SS18-SSX2 nor ATF2 recruitment to target promoters EGR1 and ATF3 (Figure 3.12 A). However, a significant increase in transcript levels for both tested target genes is detected by RT-qPCR in TLE1 knockdown cells, compared to control cells (Figure 3.12 B). The specificity of this effect was further confirmed by showing that TLE1 depletion only induces EGR1 and ATF3 transcription in HEK293 cells in the presence of Myc-SS18-SSX2 (Figures 3.13 A and B). These results indicate that SS18-SSX negatively regulates the transcription of its target genes via collaborating with TLE1. To gain molecular insights into the role of TLE1 in SS18-
SSX-mediated repression, histone modifications were analyzed as previous work linked
SS18-SSX recruitment to histone H3 lysine 27 trimethylation (H3K27me3), a key mark of gene repression (Lopez-Bergami et al., 2010). TLE1 knockdown in SYO-1 cells results in a pronounced reduction in H3K27me3 levels at the same EGR1 and ATF3
! 63! promoter regions occupied by SS18-SSX, whereas the levels of trimethylated histone H3 at lysine 4 (H3K4me3), used as controls, are unchanged (Figure 3.12 C). Given that
H3K27me3 is a hallmark of PcG-dependent gene silencing (Cao et al., 2002; Muller et al., 2002), we asked whether TLE1 serves to link the PcG complex to SS18-SSX, thereby promoting repression of target genes. TLE1 has previously been shown to have a close relationship with the catalytic subunits of the PcG complex (Chen et al., 1999; Dasen et al., 2001; Higa et al., 2006). To test this directly in synovial sarcoma, human SYO-1 and primary SS18-SSX2 mouse model tumor cells were used for reciprocal IP analysis. In both cases, co-precipitation of TLE1 leads to enrichment of the PcG component, enhancer of zeste 2 (EZH2), and its functional co-factor, histone deacetylase 1 (HDAC1)
(Figure 3.13 C). Similar interactions were obtained with other core PcG subunits, such as the embryonic ectoderm development (EED) protein and suppressor of zeste 12 homolog
(SUZ12) (Figures 3.13 D and E). These same components of the HDAC/PcG complex had also been identified in mass spectrometric analysis of SS18-SSX2 enriched proteins
(Figure 3.2 C), and not surprisingly, EZH2 interactions are maintained in the absence of
ATF2, but require TLE1 (Figure 3.12 D). Consistent with these findings, ChIP analysis demonstrates that depletion of TLE1 is associated with a concomitant decrease in
HDAC1 and EZH2 occupancy on both EGR1 and ATF3 target promoter regions (Figure
3.12 E), suggesting that TLE1 functionally regulates HDAC/PcG recruitment to SS18-
SSX target promoters. Reciprocal IP of TLE1, HDAC1, and EZH2 in normal human and mouse fibroblast cells (CCL153 and NIH/3T3) shows association of these three proteins
(Figure 3.12 F), and also raises the possibility that TLE1-containing complexes are shared between cancerous and normal cells. In accordance with observations in human
! 64! tumor cells, TLE1 also has a critical role in assembling HDAC1 and EZH2 into the SS18-
SSX complex and maintaining H3K27me3 levels and transcriptional repression on the
SS18-SSX-bound promoter regions in the mouse synovial sarcoma model (Figures 3.12
G, H and I, and Figure 3.13 F). Taken together, these results indicate that TLE1 is responsible for SS18-SSX-mediated gene silencing by an HDAC/PcG-directed epigenetic mechanism.
4.3.4 HDAC inhibitors impact SS18-SSX target gene expression through modulation of TLE1 complex recruitment
The involvement of HDAC/PcG components in the repressor activity of the SS18-
SSX complex suggests that HDAC or PcG proteins could be therapeutic targets for the treatment of synovial sarcoma. Indeed, it has been shown that repression of HDAC activity by small-molecule inhibitors can effectively suppress synovial sarcoma by reversing SS18-SSX-mediated epigenetic silencing (Lubieniecka et al., 2008; Su et al.,
2010). To further examine the importance of HDAC proteins in regulating SS18-SSX activity, HDAC1 (identified as a core SS18-SSX complex subunit, Figure 3.2 C) was knocked down. Similar to the published effects of HDAC inhibitors, depletion of HDAC1 from SYO-1 cells results in EGR1 reactivation (Figure 3.15 A), and is also associated with decreased cell growth and increased cell death (Figures 3.15 B and C). Transcript levels for the SS18-SSX target genes EGR1 and ATF3, as well as other identified targets, increase following addition of romidepsin or SB939 (clinical-grade HDAC inhibitors)
(Figure 3.14 A and 3.15 D). H3K27me3 repressive marks are decreased on the EGR1 and
ATF3 promoters during HDAC inhibitor treatment (Figure 3.14 B). Further analysis
! 65! under the same conditions reveals a concomitant reduction in localization of HDAC1 and
EZH2 to these ATF2 target promoters, even though their protein levels are unaffected
(Figures 3.14 B and 3.15 E). Consistent with this observation, HDAC inhibitor treatment of SYO-1 cells reduces HDAC1 and EZH2 appearance in the SS18-SSX complex (Figure
3.14 C). Interestingly, HDAC1 and EZH2 remain bound to TLE1 before and after exposure to romidepsin (Figure 3.15 F). As noted above, HDAC/PcG components are recruited to SS18-SSX through TLE1 (Figures 3.12 D and E), and we thus hypothesized that HDAC inhibitors block HDAC/PcG activity by altering the behavior of TLE1. In support of this possibility, glycerol-gradient sedimentation was performed using vehicle
(DMSO)- or romidepsin-treated SYO-1 cell extracts. Western blot analysis shows that
SS18-SSX2 and TLE1 are located in two separate elution peaks after HDAC inhibitor treatment, whereas the co-elution of ATF2 with the fusion protein appears to be stable under both conditions (Figure 3.14 D). To directly test the effect of HDAC inhibitors on
TLE1 recruitment, the interaction of TLE1 with SS18-SSX was assessed in the presence of romidepsin or SB939 (Figure 3.14 C). Under these conditions, both HDAC inhibitors block the association of TLE1 with SS18-SSX and its DNA-binding partner ATF2. ChIP analysis in HDAC inhibitor-treated SYO-1 cells demonstrates the removal of TLE1 from target promoters while SS18-SSX2 and ATF2 remain resident (Figure 3.15 G). A similar abrogation of TLE1, HDAC1, and EZH2 occupancy on the Egr1 and Atf3 promoters is observed in mouse synovial sarcoma cells following HDAC inhibitor treatment (Figure
3.14 E). Congruent with this, romidepsin and SB939 induce a time-dependent increase in the expression of both Egr1 and Atf3 (Figure 3.14 F), in accordance with a significant decrease in H3K27me3 levels on their promoter regions (Figure 3.14 E). Taken together,
! 66! these findings suggest that HDAC inhibitors derepress SS18-SSX target genes, at least in part through disrupting the recruitment of TLE1 and its associated HDAC/PcG proteins to the SS18-SSX complex, thus leading to loss of the repressive H3K27me3 mark and restored gene expression.
3.4 Discussion
The nature through which SS18-SSX dysregulates transcription is a long-standing question in the synovial sarcoma field. Previous studies have shown that SS18-SSX can interact with components of the TrxG transcriptional activator complexes (Nagai et al.,
2001; Thaete et al., 1999), as well it has been found to co-localize with PcG repressor factors (Soulez et al., 1999). Although these observations suggest a potential role for chromatin remodeling in SS18-SSX-mediated gene silencing, it remained unclear how
SS18-SSX controls the TrxG-PcG balance and, more importantly, how this fusion oncoprotein regulates gene expression in the absence of any known DNA-binding domain. In this study, we identify a core SS18-SSX transcriptional complex that is required for epigenetic silencing of tumor suppressor genes in synovial sarcoma. For assembly of this complex, the SS18-SSX fusion oncoprotein serves as a scaffolding protein to connect together two important transcriptional regulators, ATF2 and TLE1.
SS18-SSX alone cannot bind to DNA, and its recruitment to target promoters is dependent on the sequence-specific transcriptional activator ATF2. In this manner, SS18-
SSX recruitment of a TLE1-containing repressor complex functions to silence ATF2 target genes.
! 67! Modulation of ATF2 has also been identified in other cancers and interestingly, both activation and inhibition of ATF2 have been linked to tumorigenesis indicating that
ATF2 function in cancer is context-dependent (Lopez-Bergami et al., 2010). For instance, in melanoma, activation of ATF2 which is associated with predominantly nuclear localization appears to be important for tumorigenesis and metastasis, while suppression of ATF2 leads to increased susceptibility to various cell stressors. Conversely, in other cancers, loss or decreased expression of ATF2 is associated with an increased incidence of tumorigenesis and metastasis (Maekawa et al., 2008; Maekawa et al., 2007). Putative
ATF2 inactivating mutations in lung cancer have been identified, and in melanoma increased ATF2 cytoplasmic localization is associated with reduced tumorigenic potential and a better prognosis (Berger et al., 2003; Woo et al., 2002). Atf2 heterozygous mice exhibit an increased incidence of breast cancer after a long latency period (> 60 weeks) suggesting that an additional hit(s) is required for tumor progression in this background
(Maekawa et al., 2007). However, this second hit may in part involve loss of ATF2, as in all tumors examined ATF2 was undetectable. In a skin cancer model in mice, deletion of
Atf2 was sufficient to increase the appearance of precancerous lesions (Bhoumik et al.,
2008). However, in this model loss of ATF2 appears to promote tumorigenesis and is unlikely involved in initiation. Together, these studies support a fundamental role for
ATF2 in tumorigenesis, and highlight the varied mechanisms employed to inactivate its function.
In synovial sarcoma ATF2 function is disrupted through a mechanism in which the fusion oncoprotein couples ATF2 to a TLE1-containing complex. Interestingly, while
ATF2 shows predominantly nuclear localization, the presence of the fusion oncoprotein
! 68! in turn leads to repression of ATF2 target genes. Repression of ATF2 targets is observed in both synovial sarcoma-derived cell lines and in primary tumors. Furthermore, restoration of ATF2 transcriptional activity and/or the expression of some ATF2 target genes leads to growth suppression and apoptosis in synovial sarcoma cells, indicating that loss of ATF2 function is important for the maintenance of the tumor cell phenotype.
Importantly, loss of TLE1 phenocopies loss of ATF2, and leads to up-regulation of several ATF2 target genes. In this regard, TLE1 appears to function in a dominant- negative manner on ATF2-mediated transactivation, by mediating HDAC/PcG-directed gene silencing of ATF2 targets. It should be noted that TLE1 expression is an important clinical feature for distinguishing synovial sarcoma from other soft-tissue tumors (Jagdis et al., 2009; Knosel et al., 2010; Terry et al., 2007). However, the relevance of TLE1 as a specific biomarker for this disease remains controversial (Foo et al., 2011;
Kosemehmetoglu et al., 2009), in part due to an absence of supportive functional data.
Herein, we define a fundamental role for TLE1 in the etiology of synovial sarcoma and provide a biological rationale for its use as a diagnostic biomarker and potential therapeutic target in synovial sarcoma.
Synovial sarcomas have been shown to be highly sensitive to HDAC inhibitors in preclinical models (Ito et al., 2005; Liu et al., 2008), and herein we find that the interaction between SS18-SSX and TLE1 is critical in regulating the epigenetic reprogramming that occurs following HDAC inhibitor treatment. Based on these findings, we propose a model (Figure 3.14 G) wherein HDAC inhibitors relieve SS18-
SSX-mediated repression at ATF2 target genes most likely through removal of TLE1 and its associated HDAC/PcG factors from SS18-SSX. Consequently, as was also observed
! 69! with TLE1 knockdown, HDAC/PcG complexes are no longer recruited to ATF2 target promoters. In support of this concept, TLE1 depletion similarly results in diminished
H3K27me3 signals and elevated transcript levels for ATF2/SS18-SSX target genes. The mechanisms underlying HDAC inhibitor induced disruption of SS18-SSX-TLE1 interaction are currently unknown, but are being investigated. In summary, our findings provide fundamental insights into the nature of the SS18-SSX transcriptional complex, including a DNA-binding partner protein (ATF2) and abnormal recruitment of enzymatic epigenetic corepressors via TLE1. This information provides a biological rationale for including synovial sarcoma in clinical trials of HDAC inhibitors (NCT01112384,
NCT00918489, NCT00878800) and a framework for identifying therapeutic strategies to treat this deadly disease.
! 70!
Figure 3.1 –!SS18-SSX associates with ATF2 and TLE1 in synovial sarcoma
(A) Coomassie-stained gel of the SS18-SSX complex in SYO-1 cells. ATF2 and TLE1 were identified by mass spectrometry. Asterisk indicates IgG bands. (B) Western blot analysis of the SS18-SSX precipitates (in the presence or absence of ethidium bromide, EtBr) in SYO-1 cells. Rabbit IgG was used as a negative control. (C) Reciprocal immunoprecipitation (IP) of SS18-SSX, ATF2, and TLE1 showing their interactions in human and mouse synovial sarcoma (SS) tumors. (D) Glycerol-gradient fractionation profile of SS18-SSX2, ATF2 and TLE1 in SYO-1 cells.
! 71!
! 72! Figure 3.2 – Related to Figure 3.1
(A) Validation of S18-SSX IP using the RA2009 antibody in SYO-1 cells. SS18-SSX2 was detected by western blot with the SSX antibody, indicated by an arrowhead. (B) Unique peptide sequences of the SS18-SSX2 fusion oncoprotein identified by mass spectroscopy in SYO-1 cells. Arrow indicates the break point of SS18 and SSX2. (C) Putative SS18-SSX2-binding proteins and their unique peptide sequences identified by tandem mass spectrometric analysis in SYO-1 cells. Components of the TrxG and PcG complexes have recently been shown to interact and/or co-localize with the SS18-SSX fusion oncoprotein (Kato et al., 2002; Lubieniecka et al., 2008; Nagai et al., 2001; Soulez et al., 1999). (D) SS18-SSX IP analysis confirming its interaction with endogenous ATF2 and TLE1 in human synovial sarcoma FUJI cells. Rabbit IgG was used as the negative control. (E) RT-PCR analysis was used to define the fusion type in the primary synovial sarcoma patient tumor specimens shown in Figure 1C. SYO-1 cells are positive for the SS18-SSX2 isotype, and a melanoma sample (negative control) is negative for both SS18-SSX1 and SS18-SSX2. GAPDH indicates a loading control. (F) Reciprocal IP analysis of human clear cell sarcoma DTC-1 cells using the SS18-SSX, ATF2 and TLE1 antibodies. (G) Glycerol-gradient sedimentation profile of SS18-SSX2 and its complex components in human SYO-1 cells.
! 73!
Figure 3.3 –!Disruption of the SS18-SSX complex reduces synovial sarcoma cell growth
(A) The effect of SS18-SSX2, ATF2 and TLE1 knockdown on SYO-1 cell growth. Data represent mean ±S.D. of three experiments. (B) Colony formation assays on SYO-1 cells at 8 days after indicated siRNA transfections. Representative images for crystal violet stain, and quantitation of number of colonies by Image J software. (C) Flow cytometric analysis of apoptotic SYO-1 cells transfected with indicated siRNA following 72 hr. Percentages of Annexin-V+ cells are shown (n = 3). Bar charts are mean +/- SD.
! 74!
Figure 3.4 – Related to Figure 3.3
(A and B) human SYO-1 (A) and mouse synovial sarcoma cells (B) were transfected with control, SS18-SSX2, ATF2/Atf2, or TLE1/Tle1 siRNA. The knockdown efficiency was confirmed by western blot assays, and β-actin was used as a loading control. (C) MTT assay performed on mouse SS tumor cells transfected with control, Atf2 or Tle1/Grg1 siRNA. Data represent mean ±S.D. of three experiments. (D) Colonies were enumerated in mouse synovial sarcoma cell cultures at 8 days post- transfection with control, SS18-SSX2, mAtf2, or mTle1(Grg1) siRNA. Error bars indicate S.D. (n = 6). (E) Western blot analysis of Caspase-3 (CASP3) activation in SYO-1 and mouse synovial sarcoma cells transfected with indicated siRNA. β-actin was used as a loading control.
! 75!
Figure 3.5 - Molecular association of SS18-SSX with ATF2 and TLE1
(A) Reciprocal IP of ATF2 and TLE1 in control and SS18-SSX2 knockdown SYO-1 cells. Western blot analysis of whole cell lysates following SS18-SSX2 knockdown are shown on the left. (B) Western blot analysis of the extracts of control, ATF2 and TLE1 knockdown SYO-1 cells immunoprecipitated by the anti-SS18-SSX antibody. (C) Myc IP analysis of HEK293 cells stably expressing empty vector, Myc-tagged wild-type SS18 or SS18-SSX2.
! 76! (D) Sedimentation profile of control and SS18-SSX2 knockdown SYO-1 cell extracts by 10-40% glycerol gradients. (E) Schematic representing C-terminal Myc-tagged SS18-SSX2 truncation and deletion constructs. FL, full-length fusion oncoprotein; SNH, SYT N-terminal homolog; QPGY, glycine- /proline-/glutamine-/tyrosine-domain; SSXRD, SSX repressor domain. (F) Mapping the interface in SS18-SSX2 for its association with ATF2 and TLE1 by reciprocal IP experiments with the Myc, ATF2 and TLE1 antibodies in HEK293 cells expressing the SS18- SSX2 constructs as described in (E). (G) Schematic model illustrating the scaffolding role of SS18-SSX in ATF2 and TLE1 association.
Figure 3.6 – Related to Figure 3.5
(A) Western blot analysis of whole cell lysates extracted from control, ATF2 and TLE1 knockdown SYO-1 cells. (B and C) Reciprocal IP assay of ATF2 and TLE1 in HEK293 cell lines expressing Myc-tagged wild-type SS18 (B) or SS18-SSX2 (C). HEK293 cells expressing empty vector were used as a negative control for ATF2-TLE1 interaction. (D) Sedimentation profile of ATF2 and TLE1 through 10-40% glycerol gradients in the presence or absence of Myc-tagged SS18-SSX2 in HEK293 cells. (E) Western blot analysis of the HEK293 extracts expressing full-length (FL) or deletion mutants of Myc-SS18-SSX2. β-actin was used as a loading control.
! 77!
Figure 3.7 - ATF2 is recruited to the EGR1 promoter along with TLE1 and SS18-SSX, and is nuclear localized in synovial sarcoma
(A) Binding of SS18-SSX, ATF2, and TLE1 across the human EGR1 locus was assessed by chromatin immunoprecipitation (ChIP) and site-specific qPCR. Data shown are the mean ±S.D. of three experiments where values are expressed relative to rabbit IgG. (B) Endogenous ATF2 protein is localized to the nucleus in primary synovial sarcoma tissues by immunohistochemistry. Malignant peripheral nerve sheath tumor (MPNST) was used as a ATF2- negative control, and cytoplasmic ATF2 staining is shown in a breast cancer case for comparison. (C) Immunofluorescence analysis of ATF2 nuclear localization in SYO-1 cells. Hoechst staining defines the nuclei.
! 78!
Figure 3.8 – SS18-SSX is recruited to genes with an ATF/CRE element
(A) Heat map from meta-analysis of Affymetrix HG-U133_Plus_2 arrays, from Gene expression omnibus accessions GSE21050 (PMID 20581836) and GSE20196 (PMID: 20975339). Quantile normalized GCRMA expression values are given in log base 2, and calculated relative to the median expression for each gene. SS: synovial sarcoma; LPS liposarcoma; LMS: leiomyosarcoma. (B and C) ChIP results of primary synovial sarcoma specimens showing SS18-SSX, ATF2, and TLE1 recruitment to the promoter regions of indicated genes. B and C, represent synovial sarcomas containing either SS18-SSX1 or SS18-SSX2 fusion oncoproteins, respectively. The ChIP enrichment was normalized to Rabbit IgG, anti-GFP ChIP was used as the negative control
! 79! and ChIP assays were also carried out using non-CRE (NC) containing portions of the respective gene promoters. Bar charts are mean +/- SD. (D and E) Electrophoretic mobility shift assay competition and supershift assays showing SS18- SSX2 DNA-binding activity in HEK293 cell models (D) and SYO-1 cells (E). SS, supershift; SC, SS18-SSX:CRE complex; FP, free probe.
Figure 3.9 – Related to Figure 3.8
(A) Alignment of ATF2-binding sites in indicated gene promoters. Conserved nucleotide residues are highlighted in black, and variant residues in grey. (B and C) ChIP analysis of indicated gene promoters in primary synovial sarcoma specimens. Binding of SS18-SSX, ATF2, and TLE1 was assessed by qPCR assay specific for non-CRE (NC) regions of indicated promoters. GFP were used as the negative control.
! 80!
Figure 3.10 – ATF2 is critical for DNA-binding of the SS18-SSX complex
(A and C) Binding of SS18-SSX, ATF2 and TLE1 to representative target promoters was determined by ChIP-qPCR in human SYO-1 (A) and mouse SS tumor cells (C) transfected with control or ATF2/mATF2 siRNA. (B and D) RT-qPCR analysis of indicated ATF2 target gene expression in human SYO-1 cells (B) and mouse SS tumor cells (D) transfected with control, ATF2/mATF2, or SS18-SSX2
! 81! siRNA. Transcript levels were normalized to 18s rRNA, and depicted as a fold change between control and knockdown cells. (E) Luciferase reporter assays showing the human ATF3 promoter activity in control and SS18- SSX2 knockdown SYO-1 cells. The reporter constructs were made with the wild-type ATF3 promoter regions with or without indicated base substitutions in the ATF/CRE site. All panels, bar charts are mean +/- SD.
Figure 3.11 – Related to Figure 3.10
(A) RT-qPCR analysis of SS18-SSX2, ATF2 and TLE1 gene transcripts in human SYO-1 cells transfected with control, SS18-SSX, ATF2 or TLE1 siRNA for 48 hr. Error bars represent S.D. from three assays. (B) ChIP-qPCR analysis of the Cdkn2a, Dusp1, Fos, Meis2 and Nupr1 promoter regions in mouse SS tumor cells transfected with control or Atf2 siRNA. (C) RT-qPCR analysis of EGR1 and ATF3 gene transcripts in control and SS18-SSX2- expressing HEK293 cells transfected with wild-type or dominant-negative ATF2 cDNA.
! 82!
Figure 3.12 – TLE1 contributes to SS18-SSX-mediated repression (A, C, and E) ChIP-qPCR analysis of the human EGR1 and ATF3 promoters in control and TLE1 knockdown SYO-1 cells. The GFP antibody was used as a negative control for ChIP assays. Columns represent mean ± S.D. (n = 3). (B) RT-qPCR analysis for human EGR1 and ATF3 gene transcripts in SYO-1 cells before and after TLE1 depletion. Transcript levels were normalized to 18s rRNA, and depicted as a fold change between control and TLE1 knockdown SYO-1 cells. (D) SS18-SSX IP assay showing its association with HDAC1 and EZH2 in control, ATF2 and TLE1 knockdown SYO-1 cells. Asterisk indicates IgG bands. (F) Reciprocal IP with the TLE1, HDAC1 and EZH2 antibodies showing their interaction in normal human (left) and mouse (right) fibroblast cells.
! 83! (G) ChIP-qPCR analysis of control and mTLE1 knockdown mouse synovial sarcoma cells using the indicated antibodies. (H) Association of SS18-SSX2 with HDAC1 and EZH2 was determined by IP in mouse SS tumor cells transfected with control or mTLE1 siRNA. (I) RT-qPCR analysis of Egr1 and Atf3 gene transcripts in mouse tumor cells transfected with control or mTLE1 siRNA for 48 hr. Changes in expression were normalized to control cells. Columns represent mean ±S.D. (n = 3). Bar charts are mean +/- SD.
! 84!
Figure 3.13 – Related to Figure 3.12
(A) Control and Myc-SS18-SSX2 expressing HEK293 cells transfected with nonspecific, SS18- SSX2, or TLE1 siRNA. Western blot analysis was performed using the indicated antibodies. β- actin was used as a loading control. (B) RT-qPCR analysis of human EGR1 and ATF3 transcripts in control, SS18-SSX2, and TLE1 knockdown HEK293 cells expressing empty vector or Myc-SS18-SSX2. Columns represent mean ±S.D. from three independent experiments. (C) Reciprocal IP of TLE1, HDAC1, and EZH2 showing their interactions in SYO-1 and mouse SS tumor cells. Rabbit IgG was used as the negative control. (D and E) Association of the SS18-SSX complex with HDAC/PcG components in SYO-1 cells was determined by reciprocal IP assays using the antibodies indicated at the top of each panel. Rabbit IgG was used as the negative control. (F) Western blot analysis of whole cell lysates extracted from mouse synovial sarcoma cells transfected with control or Tle1/Grg1 siRNA for 48 hr. β-actin was used the loading control.
! 85!
! 86! Figure 3.14 – Effect of HDAC inhibitors on TLE1 recruitment and SS18-SSX-mediated gene silencing (A) RT-qPCR analysis of human EGR1 and ATF3 gene transcripts in SYO-1 cells treated with DMSO, romidepsin, or SB939 for 8 hr. Changes in expression were normalized to the 0-hr time point. (B and E) ChIP-qPCR analysis of the EGR1 and ATF3 promoters in SYO-1 and mouse synovial sarcoma cells treated with DMSO, romidepsin or SB939 for 8 hr. The antibodies used in ChIP assays are shown at the bottom of each panel. (C) SS18-SSX IP assay in DMSO-, romidepsin- and SB939-treated SYO-1 cells. ATF2 and TLE1 protein levels were determined by western blot analysis of whole cell lysates (Fig. S7E). Asterisk indicates IgG bands. (D) Glycerol-gradient sedimentation analysis of DMSO- and romidepsin-treated SYO-1 cell extracts 8h following treatment. (F) Transcript levels for Egr1 and Atf3 were measured by RT-qPCR in mouse SS tumor cells treated with DMSO, romidepsin, or SB939 for 8 h, and depicted as a fold change relative to the 0- hr time point. (G) Model of how the SS18-SSX complex regulates transcription before and after HDAC inhibitor treatment. Bar charts are mean +/- SD.
! 87!
Figure 3.15 – Related to Figure 3.14
(A) Western blot analysis showing an increased EGR1 protein level in HDAC1 knockdown SYO- 1 cells, compared to control cells. β-actin was used as the loading control. (B) Quantification of cell growth for SYO-1 cells transfected with control or HDAC1 siRNA. Error bars indicate S.D. from four independent experiments. (C) Control and HDAC1 knockdown SYO-1 cells co-stained by Hoechst and Propidium Iodide (PI, marking nuclei of dead cells) 48 hr after siRNA transfection. Representative images are shown and images were merged using Image J software. (D) RT-qPCR analysis of human CDKN2A, DUSP1, FOS, MEIS2 and NUPR1 transcripts in DMSO-, romidepsin- or SB939-treated SYO-1 cells. Error bars represent S.D. from three independent experiments.
! 88! (E) Western blot analysis of whole cell lysates prepared from DMSO-, romidepsin- or SB939- treated SYO-1 cells. β-actin was used as a loading control. (F) TLE1 IP assay showing its interaction with HDAC/PcG components HDAC1 and EZH2 in human SYO-1 (left) and mouse SS tumor cells (right) before and after DMSO and romidepsin treatment. (G) Recruitment of SS18-SSX2, ATF2 and TLE1 to the EGR1 and ATF3 promoters was assessed by ChIP-qPCR in DMSO-, romidepsin- or SB939-treated SYO-1 cells. Data were normalized to rabbit IgG, and GFP ChIP was used as a negative control. Columns represent mean ±S.D. from three experiments.
! 89! Chapter 4: Histone Deacetylase Inhibitors Induce Mule-mediated
Ubiquitination and Degradation of the Synovial Sarcoma SS18-SSX
Oncoprotein
4.1 Introduction
Synovial sarcoma is an invasive soft tissue malignancy, which predominantly targets adolescents and young adults (Haldar et al., 2008). Like other sarcomas, this disease is histologically composed of mesenchymal cells, but also exhibits variable epithelial-like differentiation and carries a unique chromosomal translocation t(X;18) (Ladanyi, 2001) creating the fusion oncoprotein SS18-SSX. Depleting SS18-SSX by small interfering
RNA (siRNA) results in apoptotic cell death of synovial sarcoma cells (Cai et al., 2011;
Carmody Soni et al., 2013; Peng et al., 2008). Conversely, overexpression of SS18-SSX in noncancerous rat fibroblasts shows transforming activity in a xenograft model (Nagai et al., 2001). Notably, mice conditionally expressing SS18-SSX in certain cell lineages develop tumors that are pathologically indistinguishable from human synovial sarcoma
(Haldar et al., 2007), confirming the critical role for SS18-SSX in synovial sarcomagenesis.
It has been challenging to fully address how SS18-SSX promotes tumorigenesis, but one proposed mechanism involves aberrant epigenetic repression (de Bruijn et al., 2007).
Indeed, SS18-SSX has the ability to set trimethylated silencing marks on histone H3 lysine 27 residues (H3K27me3) through Polycomb-group (PcG)/histone deacetylase
(HDAC) repressor complexes (de Bruijn et al., 2006; Lubieniecka et al., 2008).
Consistent with these findings, purification of SS18-SSX protein complexes has clarified
! 90! its role as a scaffold that links Activating Transcription Factor 2 (ATF2), a promoter
DNA-binding histone acetyltransferase, to TLE1 (Jones et al., 2013; Su et al., 2012), an evolutionarily ancient corepressor recruiting PcG/HDAC proteins. In accordance with this model, ablating HDAC activity through small molecule inhibitors impairs the repressive properties of SS18-SSX and suppresses synovial sarcoma growth in both cell cultures and in xenografted mice (Ito et al., 2005; Su et al., 2012). Given that chemoresistance is a major impediment to synovial sarcoma treatment, these studies highlight the potential utility of HDAC inhibitors as an effective therapy to reverse the oncogene’s transcriptional impact.
In this report, we characterize the effect of HDAC inhibitors on SS18-SSX complex assembly and identify a novel, transcription-independent mechanism underlying HDAC inhibitor-mediated tumor suppression. Our studies indicate that the ubiquitin system is important for stimulating SS18-SSX degradation, and disrupting its associated transcriptional complexes in response to HDAC inhibitor treatment thereby leading to cessation of tumor growth.
4.2 Materials and Methods
4.2.1 Cell Culture and Chemicals
Human synovial sarcoma cell lines FUJI and SYO-1 were kindly provided by Drs.
Kazuo Nagashima (Hokkaido University School of Medicine, Sapporo, Japan) and Akira
Kawai (National Cancer Centre Hospital, Tokyo, Japan) and maintained in RPMI-1640 medium with 10% fetal bovine serum (FBS) (Invitrogen). Human breast cancer cell lines
MCF-7 and T47D were purchased from ATCC and cultured in DMEM medium with
! 91! 10% FBS. Primary mouse synovial sarcoma cells were isolated from tumors of female
Myf5-Cre/SSM2 mice (Haldar et al., 2007), and maintained in DMEM medium with 10%
FBS. All cells were grown at 37°C, 95% humidity, and 5% CO2.
HDAC inhibitors romidepsin (FK228, depsipeptide and NSC-630176) and PXD101 were obtained through the Developmental Therapeutic Branch of the National Cancer
Institute (Bethesda, MD, USA). HDAC inhibitors MS-275 and SB939 were obtained from Cayman Chemical (Ann Arbor, MI, USA) and S*BIO Pte Ltd (Singapore), respectively. Cycloheximide was purchased from BioVision, MG-132 from Santa Cruz
Biotechnology, and DMSO from Sigma-Aldrich.
4.2.2 Mouse Tumor Models
All mouse experiments were approved by University of Utah and University of
British Columbia Committees on Animal Care. To generate mouse synovial sarcoma, the conditional SSM2 mice were bred to Myf5-Cre mice as described previously (Haldar et al., 2007), to express SS18-SSX2 in myoblasts and mesenchymal precursor cells. The resulting Myf5-Cre/SSM2 progenies at age 14 weeks were subjected to HDAC inhibitor administration. Mice harboring the MMTV-PyMT transgene (mammary gland tumor model) were purchased from Jackson Laboratory, and sacrificed at age 16 weeks for tumor collection.
4.2.3 Tissue and Histological Analysis
Mouse synovial sarcoma tissues were fixed in formalin and paraffin-embedded in blocks. Blocks were serially sectioned at 6-µm thickness onto Fisher Plus microscope
! 92! slides, and deparaffinized using standard procedures. Slides were stained with hematoxylin and eosin, and subsequently dehydrated in ethanol and xylene. Cytoseal-60
(Thermo Scientific) was used to mount specimens prior to microscopy analysis. Images were taken with an Olympus BX63 microscope and an Olympus DP72 camera.
4.2.4 Cell Viability Assays
Human synovial sarcoma SYO-1 cells were cultured at 60% confluence, and transfected with the indicated construct using Lipofectamine 2000 (Invitrogen). After 72 hours, cell viability was examined by 3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyltetrazolium bromide (MTT) assay as described previously (Su et al., 2012). For alamarBlue assays, transfected cells were seeded into triplicate wells of a 96-well plate.
After overnight incubation, AB (Invitrogen) was directly added to the medium at a final concentration of 10%. Absorbance was read at 570 and 600 nm with a colorimeter. The number of viable cells correlates with the magnitude of dye reduction, and cell viability is calculated as normalized absorbance (570/600 nm). Reduced viability of transfected cells is expressed as percentage of control cells.
4.2.5 Plasmid DNA Constructs
Plasmids constructs containing either wild-type (pRK5-HA-Ub) or K48R mutated ubiquitin (pRK5-HA-Ub-K48R) were obtained from Addgene (plasmids 17608 and
17604) (Lim et al., 2005). SS18-SSX2 cDNA was synthesized via Integrated DNA
Technologies (IDT) and sub-cloned into a Gateway pENTR1A vector. To generate GFP-
! 93! tagged SS18-SSX2, recombineering with the LR site-specific recombination (Invitrogen) was used to transfer SS18-SSX into the pcDNA6.2-GFP-DEST backbone.
4.2.6 Chromatin Immunoprecipitation (ChIP)
ChIP was performed following the Active Motif protocol. Briefly, 5 × 107 SYO-1 cells were cross-linked with 1% formaldehyde prior to lysis and homogenization. DNA was sheared with a Diagenode Bioruptor-UCD300 sonicator (Denville, NJ, USA). After centrifugation, the supernatants were pre-cleared with Protein-G beads, and incubated with the indicated antibody at 4°C overnight. The precipitates were harvested with
Protein-G beads, washed with buffers, and eluted with 1% SDS. After 65°C incubation to reverse cross-linking, ChIP-enriched DNA was purified using the Qiagen PCR
Purification kit, and subjected to SYBR Green qPCR analysis (Fermentas) using specific primer sets (for EGR1: 5’-TAG GGT GCA GGA TGG AGG T-3’, 5’-AAG CAG GAA
GCC CTA ATA TGG CAG-3’; for ATF3_P1: 5’-AGA GGC AGG TGG AAA GAA
GCA GGT-3’, 5’-ACC CCA ACA ATT TCT GCT CAG AGA-3’; for ATF3_P2: 5’-TTC
TCC CGG GAA GCT ATT AAT-3’, 5’-GAC TGT GGC TTG GAG AGC GTT-3’).
4.2.7 Electrophoretic Mobility Shift Assays (EMSA)
The cAMP-responsive element (CRE) probe was purchased from LI-COR, and labeled on the 5’-end of each strand with an Infrared Dye (700 nm). Binding reactions were performed in the dark at room temperature for 30 min in 25 ul of EMSA buffer (250 mM NaCl, 20 mM HEPES (pH 7.9), 2 mM DTT, 20% glycerol, 0.5% Tween-20) as described before (Su et al., 2012). Samples were separated on 5% polyacrylamide gels
! 94! (29.2:0.8 acrylamide-bisacrylamide in 100 mM Tris, 100 mM Borate, 10 mM EDTA).
Gel shift was visualized on the Odyssey Infrared scanner (LI-COR Biosciences).
4.2.8 Immunoprecipitation (IP) and Western Blots
Cells were washed with ice-cold PBS and incubated with RIPA buffer (Santa
Cruz Biotechnology). Lysates were pre-cleared by the addition of 15 ul of Protein-A/G beads and centrifuged. Subsequently, 500 ug of pre-cleared supernatant was incubated with 3.0 ug of indicated antibodies at 4°C overnight, followed by the addition of 20 ul of
Protein-A/G beads. After a 3-hr incubation, the beads were pelleted, washed once with
RIPA buffer and twice with PBS, and boiled in 2 × loading dye. Samples were separated by 6% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and transferred to nitrocellulose membranes (Bio-Rad Laboratories). Blots were incubated with indicated antibodies, and visualized using the Odyssey Infrared System (LI-COR
Biosciences). The antibodies used in this study include: ATF-2 (C-19, Santa Cruz
Biotech); β-Actin (C4, Santa Cruz Biotech); GAPDH (6C5, Santa Cruz Biotech); Lamin
B (B-10, Santa Cruz Biotech); SS18 (SYT, H-80, Santa Cruz Biotech); TLE1 (C-19,
Santa Cruz Biotech); Ubiquitin (P4D1, Santa Cruz Biotech); Mule (A300-486A, Bethyl laboratories); Flag (M2, Sigma-Aldrich); α-Tubulin (DM1A, Sigma-Aldrich); HA
(12CA5, Roche); and Myc (9E10, Abcam).
4.2.9 Mass Spectrometry (MS)
For analysis of immunoprecipitated complexes by mass spectrometry, samples were run on an SDS-PAGE gel, and subsequently stained with Coomassie blue. Stained
! 95! bands were cut out and reduced with dithiothreitol (DTT), prior to alkylation with iodoacetamide (IAA). Gel bands were subsequently digested with trypsin at 37°C overnight. Digested peptides were extracted and reconstituted in formic acid (FA) before analysis on a QStar XL LC-MS/MS (Applied Biosystems). MS/MS spectra were used to confirm protein identity through the Mascot database.
4.2.10 RNA Interference (RNAi)
Small interfering RNAs (siRNAs) for human and mouse Mule were purchased from IDT Pre-designed DsiRNA collection. At 60% confluence, cells were transfected with the indicated siRNA using Lipofectamine RNAiMAX transfection reagent
(Invitrogen) according to the manufacturer’s instructions. Knockdown efficiency was determined by RT-qPCR and western blots.
4.2.11 Real-time qPCR (RT-qPCR)
Total RNA was isolated and then transcribed to cDNA using the Qiagen RNeasy kit and the high-capacity cDNA reverse transcription kit (Applied Biosystems), respectively. Taqman gene expression assays were performed on a ABI-7500 Fast Real-
Time PCR System with specific primer/probe sets. All transcript levels were normalized to 18S ribosomal RNA (rRNA) expression.
4.2.12 Subcellular Fractionation
Cells were washed with ice-cold PBS, followed by incubation in lysis buffer (10 mM Tris pH 7.5, 10 mM NaCl, 3 mM MgCl2, 1 mM EGTA, 0.05% NP-40). Cell lysates
! 96! were scraped from plates and pelleted at 5000 rpm for 10 min. The supernatant was transferred into a new tube and centrifuged at 14000 rpm for 15 min, to give the cytoplasmic fraction. The pellet was washed twice with nuclear wash buffer (10 mM
PIPES pH 6.8, 300 mM Sucrose, 25 mM NaCl, 3 mM MgCl2, 1 mM EGTA), and subsequently resuspended in the same buffer. This mixture was layered on to the top of 1
M sucrose solution and centrifuged at 5000 rpm for 10 min. The pellet was washed with nuclear wash buffer, and suspended in nuclear extraction buffer (20 mM HEPES pH 7.9,
300 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA), to generate the nuclear fraction.
4.3 Results
4.3.1 HDAC inhibitor treatment triggers SS18-SSX complex disruption
To clarify the mechanism of HDAC inhibitor action in synovial sarcoma, patient- derived SYO-1 cells (confirmed to express SS18-SSX2) were exposed to romidepsin (a clinically applicable HDAC inhibitor) or DMSO (vehicle control) for 16 hours. SS18-
SSX complex assembly was evaluated by immunoprecipitation (IP) using an ATF2 antibody. Consistent with previous work (Su et al., 2012), SS18-SSX and its repressive cofactor TLE1 were efficiently co-purified with ATF2 in control cells (Figure 4.1 A).
However, these interactions were deficient in romidepsin-treated ATF2 immunoprecipitates.
Next, we analyzed the functional relevance of SS18-SSX complex disruption in response to HDAC inhibitor treatment. Experiments used a tagged SS18-SSX construct to ensure antibody specificity, in a synovial sarcoma cell model to ensure transfected proteins are expressed in a cellular background modeling the disease. Specifically, a
! 97! Myc-tagged SS18-SSX2 construct was introduced into SYO-1 cells, and its DNA- binding activity was monitored by chromatin IP (ChIP) and real-time quantitative PCR
(RT-qPCR). In untreated cells, anti-Myc antibodies pulled down the known SS18-SSX- binding sequences of the EGR1 promoter (containing the CRE cAMP responsive element recognized by ATF2) at levels well above the IgG background (Figure 4.1 B). The lack of ChIP signals from mock transfection confirmed the specificity of Myc-SS18-SSX2 recruitment to DNA. Importantly, this protein-DNA interaction was significantly decreased upon romidepsin stimulation. Similar results (Figure 4.2 A) were obtained in the locus of ATF3, another known SS18-SSX target gene (Hayakawa et al., 2013; Su et al., 2012). This observation was further supported by experiments using siRNA knockdown of SS18-SSX (Figure 4.2 B) in electrophoretic mobility shift assays. Using
CRE as the binding template, we observed a specific SS18-SSX/CRE complex in SYO-1 cells (Figure 4.2 C). The analogous protein-DNA complex was also detected in untreated cells expressing Myc-SS18-SSX2, and its identity was confirmed through an anti-Myc antibody-dependent electrophoretic mobility shift (Figure 4.1 C). In the same cellular context, romidepsin caused near-complete loss of the supershift response (Figure 4.1 C, compare lanes 6 and 9). Together with the IP analyses, these results demonstrate that assembly of the SS18-SSX transcriptional complex is disrupted in response to HDAC inhibitor treatment.
4.3.2 HDAC inhibitors accelerate SS18-SSX degradation
To better understand the mechanism of HDAC inhibitor-induced SS18-SSX complex dissociation, protein lysates from control and romidepsin-treated SYO-1 cells were
! 98! separated by electrophoresis on a polyacrylamide-SDS gel. Western blot analysis revealed that amounts of SS18-SSX protein decreased in the presence of romidepsin, whereas the levels of its wild-type SS18 isoform and major complex partners (ATF2 and
TLE1) remained constant (Figure 4.1 A). Importantly, this finding could be extended to three additional HDAC inhibitors with structures distinct from romidepsin (Figure 4.3 A).
In contrast, the chemotherapeutic drug, doxorubicin, had no effect on normalized SS18-
SSX protein levels (Figure 4.3 A, compare lanes 5 and 6), confirming the specificity of
HDAC inhibitor-stimulated SS18-SSX reduction in synovial sarcoma cells.
To investigate whether HDAC inhibitors control SS18-SSX levels in vivo, a previously established synovial sarcoma mouse model (SSM2) (Haldar et al., 2007) was weekly injected with either 5% glucose vehicle or romidepsin at different doses for 3 weeks (Figure 4.3 B). Tumors were dissected from the thoracic cage and processed for standard hematoxylin and eosin (H&E) staining (Figure 4.3 C), to confirm their histological features. Consistent with the cultured human synovial sarcoma cell results,
SS18-SSX protein in mouse tumor tissues was selectively decreased after romidepsin administration (Figure 4.3 D). The SS18-SSX reduction consistently correlated with a significantly decreased tumor mass (Figure 4.3 C). Likewise, primary cultures of SSM2 tumor cells also exhibited a specific decrease in SS18-SSX protein levels and an impairment of SS18-SSX complex assembly with murine TLE1 after romidepsin addition
(Figure 4.4 A).
To study how HDAC inhibitors regulate SS18-SSX levels, total RNA was isolated from human SYO-1 and mouse SSM2 cells treated with DMSO or romidepsin.
RT-qPCR analysis showed no changes in SS18-SSX mRNA abundance (Figures 4.2 D
! 99! and 4.4 B), raising the possibility that SS18-SSX reduction is due to post-transcriptional mechanisms. In support of this notion, cycloheximide (CHX)-chase analysis revealed a clear reduction in SS18-SSX’s half-life (Figures 4.3 E and 4.4 C), suggesting that HDAC inhibitors might promote post-translational SS18-SSX down-regulation. Indeed, SS18-
SSX degradation by romidepsin was greatly abrogated in the presence of a proteasome inhibitor (MG-132) in both SYO-1 and SSM2 cells (Figure 4.3 F). Polyubiquitinated
SS18-SSX conjugates accumulated after stimulation with romidepsin (Figure 4.3 G) and two other structurally different HDAC inhibitors (Figure 4.4 D). To further dissect the ubiquitination of SS18-SSX, romidepsin-treated SS18-SSX immunoprecipitates were analyzed by multidimensional chromatography coupled with tandem mass spectrometry
(LC/LC-MS/MS). Although approximately 90% of the human ubiquitin sequence was covered, SS18-SSX only carried the branched Glycine-Glycine signature peptide with the lysine 48 (K48) linkage (Figures 4.4 E and F). Given the well-known role of K48-linked ubiquitination in proteasomal degradation (Ravid and Hochstrasser, 2008), constructs containing the K48 mutation in which the lysine 48 residue was replaced with arginine
(Lim et al., 2005) were expressed in the SYO-1 synovial sarcoma cell background. Upon romidepsin stimulation, SS18-SSX was strongly polyubiquitinated by hemagglutinin
(HA)-tagged wild-type ubiquitin. Under the same conditions, however, the HA-tagged
K48R mutant lacked virtually all polyubiquitination products (Figure 4.4 G, compare lanes 2 and 4). Collectively, these results indicate that HDAC inhibitors induce SS18-
SSX ubiquitination and degradation through the ubiquitin-proteasome system.
! 100! 4.3.3 Mule targets SS18-SSX for ubiquitination and degradation
To identify the ubiquitin ligase responsible for SS18-SSX degradation, a proteomics approach was taken to reveal SS18-SSX interactants. Multiple peptides from MCL-1 ubiquitin ligase E3 (Mule) were detected specifically in romidepsin-treated human and mouse synovial sarcoma cells (Figures 4.5 A and 4.6 A). Mule is an essential HECT- family E3 ligase that targets diverse substrates for ubiquitin-mediated degradation and controls cellular fate during development and disease pathogenesis (Adhikary et al.,
2005; Chen et al., 2005; Hall et al., 2007; Herold et al., 2008; Leboucher et al., 2012;
Ross et al., 2011; Zhang et al., 2011; Zhao et al., 2009; Zhong et al., 2005). Reciprocal IP experiments confirmed that endogenous SS18-SSX-Mule interactions occur in SYO-1 cells following romidepsin or SB939 treatment (Figures 4.5 B and 4.6 B). Furthermore, decreasing Mule expression via two distinct pairs of siRNA duplexes greatly attenuated
SS18-SSX protein degradation without any significant effect on SS18-SSX transcript levels (Figure 4.5 C), suggesting an important role for Mule in HDAC inhibitor-induced
SS18-SSX turnover.
Consistent with this view, Mule protein levels were very low or undetectable in human synovial sarcoma cell lines and in SSM2 mouse tumor specimens, under conditions that readily detect Mule in human breast cancer control cell lines and mouse tumor tissue (Figure 4.6 C). Upon stimulation with romidepsin, however, Mule protein accumulates in human synovial sarcoma cells (Figure 4.5 C, compare lanes 1 and 4), whereas its transcript levels are unaffected. Furthermore, both western blot and immunofluorescence analyses confirmed analogous results in SSM2 mouse tumors
(Figures 4.5 D and E). Considering that the proteasome inhibitor MG-132 caused an
! 101! increase in Mule protein levels in the absence of HDAC inhibitors (Figure 4.6 D, compare lanes 1 and 4), the data suggest that when synovial sarcoma cells are treated with HDAC inhibitors, Mule is post-transcriptionally up-regulated to bind and ubiquitinate targets such as SS18-SSX leading to their degradation.
4.3.4 Mule-mediated SS18-SSX degradation occurs in the cytoplasm
To further test the role of Mule in SS18-SSX turnover, Flag-tagged wild-type Mule
(Flag-Mule) and a mutated (C4341A) kinase-deficient Mule were transfected into SYO-1 cells. Surprisingly, compared with the C4341A negative control, overexpression of Flag-
Mule failed to promote SS18-SSX degradation (Figure 4.8 A), or reduce the viability of
SYO-1 cells (Figures 4.7 A and 4.8 B). These results indicate that Mule decreases SS18-
SSX stability in an HDAC inhibitor-dependent manner.
To explore the mechanistic link between Mule-mediated SS18-SSX degradation and
HDAC inhibitor stimulation, SYO-1 cells were transiently transfected with an expression vector encoding green fluorescence protein (GFP)-tagged SS18-SSX2. Fluorescence microscopy revealed ectopic fusion oncoprotein localization to the nucleus, with a previously described punctate pattern (dos Santos et al., 1997; Soulez et al., 1999; Wang et al., 2013), when compared to the diffuse localization of GFP alone (Figure 4.7 B, compare top and middle panels). Strikingly, addition of romidepsin resulted in the redistribution of GFP-SS18-SSX2 to the cytoplasm (Figure 4.7 B, bottom panel). This observation was also verified by subcellular fractionation assays in which SYO-1 cell lysates were separated into nuclear and cytoplasmic fractions: Western blot analysis showed that endogenous SS18-SSX protein remains nuclear, whereas a large fraction of
! 102! SS18-SSX (but not native SS18) protein relocates to the cytoplasm in response to romidepsin exposure (Figures 4.7 C and 4.8 C, top panel).
In support of SS18-SSX cytoplasmic re-localization and degradation following
HDAC inhibitor treatment, we found that in SYO-1 cells Mule was exclusively cytoplasmic throughout the romidepsin treatment (Figure 4.8 C). Consistent with this,
Mule protein was also observed predominantly within the cytoplasm in romidepsin- treated mouse tumors (Figure 4.7 E). Reciprocal IP experiments confirmed that romidepsin-stimulated SS18-SSX-Mule interactions occur only in the cytoplasm (Figures
4.7 D and 4.8 D). Given that poly-ubiquitinated SS18-SSX conjugates were detected only within the cytoplasm (Figure 4.7 D), we speculated that nuclear export of SS18-SSX might be a key regulatory step for HDAC inhibitor-induced Mule-mediated degradation.
To rule out a role for Mule in regulating the intracellular distribution of SS18-SSX, siRNA-dependent knockdown experiments were carried out to deplete Mule expression in SYO-1 cells in the absence or presence of romidepsin. Although decreasing Mule virtually ablated poly-ubiquitination of SS18-SSX (Figure 4.8 E, compare lanes 6 and 8), a comparable fraction of SS18-SSX localized to the cytoplasm in both control and knockdown cells (Figures 4.8 E and F), indicating that romidepsin-directed SS18-SSX shuttling to the cytoplasm is independent of Mule. In summary, we conclude that in response to HDAC inhibitor exposure, SS18-SSX exhibits nuclear-cytoplasmic shuttling, where the fusion oncoprotein interacts with the E3 ligase Mule for ubiquitination and proteolysis.
! 103! 4.4 Discussion
With the current study, we show that HDAC inhibitors not only reverse SS18-SSX transcriptional silencing (Lubieniecka et al., 2008; Su et al., 2012), but also regulate the level of oncoprotein expression in both patient-derived cell lines and in a synovial sarcoma mouse model. This effect is post-translational, and relies on ubiquitin-mediated proteolysis. We further show that the E3 ligase, Mule, is responsible for SS18-SSX degradation, and that HDAC inhibitors stimulate ubiquitination of the fusion oncoprotein.
Strikingly, Mule-associated ubiquitin ligase activity is specific for SS18-SSX, as we were not able to detect any significant effect on wild-type SS18 or other major interacting partners (ATF2 and TLE1).
It is noteworthy that in response to HDAC inhibitor stimulation, Mule recognition may not be the only step regulating SS18-SSX degradation. In synovial sarcoma cells, cell fractionation studies indicate that Mule accumulates in the cytoplasm throughout drug treatment. Consistent with this, poly-ubiquitinated SS18-SSX conjugates appear selectively in the cytoplasm. Although the precise mechanism remains undefined, we have found that SS18-SSX becomes acetylated in response to HDAC inhibitor stimulation (L.S., T.O.N. and T.M.U., unpublished data). It will be important to map acetylated lysine residue(s) in the fusion oncoprotein, to test further whether and how acetylation regulates SS18-SSX action in HDAC inhibitor-induced tumor suppression.
At present, the treatment for primary synovial sarcoma is largely dependent on surgical excision with adjuvant radiation. Use of cytotoxic chemotherapy is variable and is not considered curative, leaving little in the way of truly effective systemic medical options – in particular, none exist that directly target our emerging mechanistic
! 104! understanding of SS18-SSX biology. Although it is a transcriptional regulator, this fusion oncoprotein itself has no apparent capability to associate directly with DNA. SS18-SSX- containing transcriptional complexes have been identified in synovial sarcoma cells that both positively and negatively regulate target gene expression (de Bruijn et al., 2008;
Garcia et al., 2012; Kadoch and Crabtree, 2013; Kato et al., 2002; Su et al., 2012). Such complexities in SS18-SSX action obviously make therapeutic development more challenging, in the absence of strategies targeting the oncoprotein itself. Here, however, we present evidence for a novel mechanism (Figure 4.7 E) by which HDAC inhibitors induce degradation of the SS18-SSX oncoprotein itself that, mechanistically, supports the use of HDAC inhibitors as a targeted therapy to treat synovial sarcoma.
! 105!
Figure 4.1 – HDAC inhibitor treatment triggers SS18-SSX complex disruption
(A) Immunoprecipitations (IPs) were performed in DMSO- or romidepsin (10nM)-treated human SYO-1 cells with anti-ATF2 antibodies. Whole cell lysates were analyzed by western blots under the same conditions. (B) Anti-Myc chromatin immunoprecipitation (ChIP) experiments were carried out in Myc-SS18- SSX2-expressing SYO-1 cells in the absence or presence of romidepsin. Mouse IgG was used as an internal control, and mock transfection with an empty vector served as an external control. (C) Electrophoresis mobility shift assays were performed in Myc-SS18-SSX2-expressing SYO-1 cells treated with or without romidepsin. Mock transfection was used as a negative control for the supershift assay. S, supershift; S/C, SS18-SSX/CRE complexes.
! 106!
Figure 4.2 – Related to Figure 4.1
(A) Anti-Myc ChIP-qPCR analysis of the human ATF-3 promoter was performed in control or in Myc-SS18-SSX2-expressing SYO-1 cells treated with DMSO vehicle or romidepsin. Primer sets for ChIP-qPCR are shown in the ATF-3 promoter diagram; mouse IgG was used as a negative control. NS, not significant. *, p < 0.01. (B) Western blots of whole cell lysates extracted from human SYO-1 cells transfected with nonspecific or SS18 siRNA. (C) EMSA experiments were performed in the nuclear extracts (NE) from SYO-1 cells under the condition as shown in (B). cAMP-responsive elements (CRE) were used as the SS18-SSX- binding probe; no NE input was used as a negative control. S/C, SS18-SSX/CRE complexes. (D) RT-qPCR analysis of SS18-SSX mRNA levels was performed in human SYO-1 cells treated with DMSO or romidepsin.
! 107!
Figure 4.3 – HDAC inhibitors induce proteasomal degradation of SS18-SSX
(A) Western blot analysis of whole cell lysates extracted from human SYO-1 cells treated with DMSO (vehicle), MS-275 (1uM), PXD101 (1uM), SB939 (1uM), or doxorubicin (1uM). HDAC inhibitors are highlighted in red. The anti-SS18 antibody recognizes both SS18-SSX fusion (65 kDa) and native SS18 (50 kDa) proteins. (B) HDAC inhibitor injection schema. SSM2 mice at age 14 weeks were injected with 5% glucose (control) or romidepsin weekly for 3 weeks. (C) Effect of romidepsin treatment at indicated doses on total tumor mass (N = 5-8 per group ± standard deviation). Representative H&E staining confirms monophasic synovial sarcoma tumor histology. Scale bar, 50 µm. (D) Western blot analysis was performed in two groups of control and romidepsin-treated tumor samples (at the dose of 3.0 mg/kg).
! 108! (E) Cycloheximide (CHX)-chase assays were carried out in human SYO-1 cells with or without romidepsin treatment. The levels of SS18 and SS18-SSX protein were quantified on the basis of western blot analysis (see Figure S2C). (F) Western blot analysis of whole cell lysates from control or romidepsin-treated SYO-1 and SSM2 cells in the absence or presence of proteasome inhibitor MG-132. (G) Polyubiquitinated SS18-SSX was examined by anti-ubiquitin western blotting in DMSO- or romidepsin-treated SYO-1 and SSM2 cells. Rabbit IgG was used as a negative control.
! 109!
Figure 4.4 – Related to Figure 4.3
(A) IP experiments were performed in DMSO- or romidepsin-treated mouse SSM2 cells with anti-ATF2 antibodies. Whole cell lysates were analyzed by western blots under the same conditions. (B) RT-qPCR analysis of SS18-SSX mRNA levels was performed in mouse SSM2 cells treated with DMSO or romidepsin. (C) Cycloheximide (CHX)-chase assay was carried out in DMSO- or romidepsin-treated SYO-1 cells. SS18-SSX and SS18 protein levels are shown in western blots, with GAPDH as a loading control.
! 110! (D) SS18-SSX was immunoprecipitated from SYO-1 cells treated with DMSO, MS-275, or SB939. Polyubiquitinated SS18-SSX products were detected in western blots with use of anti- ubiquitin antibodies. (E) Amino-acid sequence of human ubiquitin protein. The peptides identified in mass spectrometry (MS) are highlighted in red. (F) MS/MS spectrum of the ubiquitin peptide containing a Gly-Gly modified Lysine 48 (K48) residue. (G) Anti-SS18 IP assays were performed in DMSO- or romidepsin-treated SYO-1 cells expressing either HA-tagged ubiquitin or its K48R mutation. Polyubiquitinated SS18-SSX products were examined by western blots using anti-HA antibodies.
! 111!
Figure 4.5 – SS18-SSX degradation requires the Mule-associated E3 ligase activity
(A) Mass spectrometry (MS) identified Mule in the SS18-SSX immunoprecipitates prepared from romidepsin-treated SYO-1 cells. The unique peptides of Mule are listed in the right panel. (B) Reciprocal IPs using anti-SS18 and anti-Mule antibodies were performed in control or romidepsin-treated SYO-1 cells. Asterisk, rabbit IgG heavy chain. (C) Real-time quantitative PCR (RT-qPCR) and western blots were performed in control or romidepsin-treated SYO-1 cells transfected with nonspecific or two individual Mule siRNAs. **, p < 0.005. (D) Western blot analysis of Mule protein levels was performed in the tumor tissues from two individual groups of SSM2 mice treated with 5% glucose (control) or 3.0 mg/kg/dose romidepsin. (E) Immunofluorescence staining of Mule in vehicle and romidepsin-treated mouse tumor tissues. DAPI was used to indicate nuclei; scale bar, 20 µm. The insert image highlights Mule accumulation and cytoplasmic localization under romidepsin treatment (scale bar, 5 µm).
! 112!
Figure 4.6 – Related to Figure 4.5
(A) Tandem MS identified Mule in the SS18-SSX immunoprecipitates prepared from mouse SSM2 cells. Western blot analysis was carried out to confirm the identity of Mule, and unique peptides of Mule are listed in the right panel. (B) Reciprocal IPs of SS18-SSX and Mule were performed in human SYO-1 cells treated with DMSO vehicle or HDAC inhibitor SB939. (C) Cell (lanes 1-4) and tissue (lanes 5-8) lysates were prepared from cultured human cell lines and mouse tumor tissues, and analyzed by western blots. Synovial sarcoma models are highlighted in red, and breast cancer models in black. (D) Whole cells lysates were prepared from DMSO- or SB939-treated SYO-1 cells in the absence or presence of proteasome inhibitor MG-132. Western blots were performed to examine endogenous Mule protein levels, and GAPDH was used as a loading control.
! 113!
Figure 4.7 –!SS18-SSX associates with Mule in the cytoplasm
(A) MTT assay was performed to examine the effect of Flag-Mule overexpression on the viability of human SYO-1 cells. The C4341A mutation was used as a negative control, and the tumor suppressor EGR1 as a positive control (Lubieniecka et al., 2008; Su et al., 2012). **, p < 0.005. (B) Fluorescence microscopy showed the cellular distribution of GFP-SS18-SSX2 in SYO-1 cells treated with or without romidepsin. GFP vector alone was used as a negative control. Scale bar, 10 µm. (C) Cellular fractionation assays were performed in DMSO- or romidepsin-treated SYO-1 cells in the presence of MG-132. Lamin B and α-Tubulin were used as markers for nuclear (Nuc) and cytoplasmic (Cyto) fractions, respectively. (D) Anti-SS18 IP assays were carried out in nuclear and cytoplasmic fractions of SYO-1 cells treated with romidepsin at indicated time-points. Polyubiquitinated SS18-SSX was detected by anti-ubiquitin western blots. (E) A dual mechanism proposed to illustrate how HDAC inhibitors suppress SS18-SSX action in synovial sarcoma cells.
! 114!
Figure 4.8 – Related to Figure 4.7
(A) Western blots of whole cell lysates extracted from human SYO-1 cells transfected with Flag- tagged Mule or its C4341A mutation. Mock transfection with an empty vector was used as a negative control. (B) AlamarBlue assay was performed to examine the effect of Flag-Mule overexpression on the viability of human SYO-1 cells. The C4341A mutation was used as a negative control, and the tumor suppressor EGR1 as a positive control. (C) Western blot analysis of nuclear and cytoplasmic fractions was carried out in romidepsin- treated SYO-1 cells in the absence or presence of MG-132. Number indicates the time of drug treatment. (D) Anti-Mule IPs were performed in both nuclear and cytoplasmic fractions of SYO-1 cells treated with romidepsin for either 8 or 20 hours.
! 115! (E) SYO-1 cells were transfected with nonspecific (control) or Mule siRNA. SS18-SSX was immunoprecipitated from nuclear and cytoplasmic fractions of these cells in the absence or presence of romidepsin. Polyubiquitinated SS18-SSX products were detected in western blots using anti-ubiquitin antibodies. (F) Effect of Mule knockdown on SS18-SSX distribution in control or romidepsin-treated SYO-1 cells. The indicated fractions are the same as in panel E.
! 116! Chapter 5: Conclusions and Discussion
5.1 General Conclusion
Synovial sarcoma is a highly aggressive cancer in children and young adults. Its molecular mechanism of pathogenesis has remained elusive for more than 25 years. The creation of the SS18-SSX fusion protein is a unique feature of this disease (Clark et al.,
1994), and it is also necessary and sufficient for the development of synovial sarcoma
(Carmody Soni et al., 2014; Peng et al., 2008; Su et al., 2012; Takenaka et al., 2010). Its oncogenic effect is believed to involve aberrant repression of tumor suppressor genes (de
Bruijn et al., 2006; Lubieniecka et al., 2008). For example, EGR1 is one of the known
SS18-SSX targets, which functions as a master regulator of cellular growth and death pathways (such as, PTEN/AKT and Tp53) (Sarver et al., 2010; Su et al., 2010).
This thesis focused on a long-standing and fundamental question – how the fusion protein SS18-SSX functions – and through the use of a combination of proteomic and cell biological strategies novel insights were revealed into the chromatin remodeling- dependent activity of SS18-SSX in tumorigenesis. Indeed, a number of previous and recent studies have pointed out a potential relationship between SS18-SSX and chromatin remodeling complexes (de Bruijn et al., 2006; Garcia et al., 2012; Ito et al., 2004; Kadoch and Crabtree, 2013; Kato et al., 2002; Lubieniecka et al., 2008; Nagai et al., 2001; Thaete et al., 1999). Here, my work led to purifying distinct chromatin modifiers (such as,
HDAC and PcG proteins) in addition to novel cofactors for SS18-SSX, further demonstrating that aberrant epigenetic alterations (in particular, the H3K27me3 repressive chromatin marks) play a direct role in SS18-SSX oncogenic action. In
! 117! addition, my work also provided strong evidence suggesting that specific small-molecule inhibitors can reverse these epigenetic abnormalities and thus effectively suppress synovial sarcoma growth both in vitro and in vivo. Considering the fact that epigenetic inhibitors are increasingly being used in the clinic (Campbell and Tummino, 2014; West and Johnstone, 2014), the present study may not only uncover the molecular basis of synovial sarcoma pathogenesis, but more importantly provide rationale for pursuing the therapeutic use of these small molecules either alone or in combination to treat this deadly disease.
5.2 The Molecular Basis of SS18-SSX-mediated Repression
Our understanding of SS18-SSX-mediated repression was significantly challenged by the fact that the fusion protein itself does not have any detectable DNA-binding or enzymatic activity. Using biochemical and proteomics approaches, I sought to address how the fusion oncoprotein targets gene promoters and modifies transcription in cancer cells. My thesis work particularly focused on two previously unrelated molecules –
Activating Transcription Factor 2 (ATF2) and Transducin-like Enhancer of Split 1
(TLE1) – because they both associate with SS18-SSX in human and mouse synovial sarcoma models, and have DNA-binding and/or histone modifying activity. Surprisingly, these novel SS18-SSX cofactors are known tumor suppressors; silencing, deletion, or mutation of the ATF2 and TLE1 genes commonly occurs in many human cancers
(Bhoumik et al., 2008; Dayyani et al., 2008; Fraga et al., 2008; Maekawa et al., 2007;
Sakai et al., 1991). Although the mechanism of their increased expression in synovial sarcoma is still unclear, ATF2 and TLE1 are physically assembled into one complex in
! 118! synovial sarcoma. Importantly, the fusion protein SS18-SSX turns out to be the scaffolding molecule responsible for assembly of this tumor-associated complex.
In subsequent studies, it was important to define how the fusion oncoprotein SS18-
SSX utilizes the tumor suppressors ATF2 and TLE1 to induce tumor formation. Toward this end, I took advantage of the nature of SS18-SSX in gene silencing, to address mechanistically how ATF2 and TLE1 contribute to SS18-SSX-mediated repression. The fact that SS18-SSX specifically recognizes a response element 200 base pair (bp) upstream of the EGR1 transcription start site (TSS) led us to identify a consensus ATF2- binding site (also known as cAMP-responsive element or CRE). Gel shift assays clearly showed that SS18-SSX interacts with in vitro synthesized ATF2-binding sequences.
Along the same line, in vivo chromatin immunoprecipitation (ChIP) assays further demonstrated that SS18-SSX no longer binds to the EGR1 promoter in the absence of
ATF2, thus confirming ATF2 as an essential factor for SS18-SSX recruitment to DNA.
This finding was greatly extended by validation at multiple ATF2 regulatory genes as well at novel SS18-SSX targets. In the future, ChIP-sequencing (ChIP-seq) analyses would be necessary to globally map SS18-SSX-associated locations in the genome. Since
ATF2 usually functions as a homodimer or heterodimer (Benbrook and Jones, 1990; Hai and Curran, 1991; Huguier et al., 1998), it would be also worthwhile to search for other potential transcription factors coupled with ATF2 and/or SS18-SSX. These analyses would greatly aid further delineation of the DNA-binding specificity of the fusion oncoprotein complex and the extent of ATF2 target gene involvement.
In contrast to ATF2, TLE1 does not bind to DNA on its own, however we have found that it is consistently present at SS18-SSX target gene promoters. Given that ATF2
! 119! activates and TLE1 inhibits transcription (Chen and Courey, 2000; Kawasaki et al., 2000;
Palaparti et al., 1997), it was speculated that TLE1 might contribute to SS18-SSX- mediated repression. Consistent with this view, TLE1 was found to recognize ATF2- binding sites through SS18-SSX and to act in a dominant manner to repress gene expression over ATF2 transactivation activity. Mechanistically, the dominant-negative effect of TLE1 is the result of HDAC/PcG-mediated H3K27me3 silencing. Importantly, the connection between TLE1 and HDAC/PcG repressors seems to be completely conserved in normal and cancer cells. Indeed, as our manuscript (described in Chapter 3) was in press, Gregory R. Dressler and colleagues showed that other TLE family molecules induce gene silencing in a similar H3K27me3-dependent manner, but through association with distinct PcG cofactors (Patel et al., 2012). These independent studies raise significant questions for future investigations. For example, why does SS18-SSX select TLE1 (rather than other TLEs) as its partner in tumorigenesis, and what determines the functional specificity of TLE family members in diverse biological processes?
Together with earlier studies, the current research strongly supports a novel oncogenic mechanism (Figure 5.1 A) through which two unrelated tumor suppressors are connected by the fusion oncoprotein SS18-SSX to antagonize each other culminating in deregulation of cell survival and death pathways.
5.3 Therapeutic Targeting of SS18-SSX Fusion Protein
The success of elucidating the components within the SS18-SSX transcriptional complex led us to address whether this information could be applied to identification and validation of disease-modifying therapeutics. For this purpose, HDAC inhibitors were an
! 120! attractive candidate, as numerous compounds have been used in clinical trials, but more importantly because they are able to directly target the core components of the SS18-SSX complex (HDAC/PcG). Indeed, our laboratory together with other groups has provided direct evidence that synovial sarcoma is highly sensitive to HDAC inhibitor treatment both in vitro and in vivo (Cassier et al., 2013; Ito et al., 2005; Lubieniecka et al., 2008;
Nguyen et al., 2009; Su et al., 2010; Su et al., 2012). In further support of this concept, my work demonstrated that reactivation of SS18-SSX target genes is a critical step for
HDAC inhibitor-mediated suppression of synovial sarcoma survival. For example, siRNA-based depletion of EGR1 remarkably rescued synovial sarcoma cells from HDAC inhibitor-induced cell death. Consistently, HDAC inhibitor-induced reactivation of SS18-
SSX target genes is tightly correlated with a corresponding reduction of H3K27me3 repressive marks. These data therefore indicate that HDAC inhibitors reverse SS18-SSX- mediated repression at least in part through inhibition of the HDAC/PcG activity.
I went on to investigate the detailed mechanism through which HDAC inhibitors function to interfere with SS18-SSX action. Unexpectedly, ChIP assays showed that
HDAC/PcG proteins were removed from the SS18-SSX target promoters in response to
HDAC inhibitor stimulation. Furthermore, biochemical analyses showed that the inhibitory effect of HDAC inhibitors is the result of TLE1/HDAC/PcG dissociation from the SS18-SSX complex (Figure 5.1 B). These findings strongly support a novel, anti- cancer mechanism through which HDAC inhibitors target the SS18-SSX transcriptional complex and promote its disassembly. More surprisingly, long-term treatment of synovial sarcoma cells with HDAC inhibitors led to a significant decrease in SS18-SSX protein abundance but not its messenger RNA (mRNA). This was subsequently found to be a
! 121! result of increased ubiquitination of SS18-SSX in an HDAC inhibitor-dependent fashion.
These findings indicated that HDAC inhibitors are likely working through a novel mechanism to regulate SS18-SSX stability. Congruent with this, identification of MCL-1
Ubiquitin Ligase E3 (Mule) as the E3 ligase for regulating SS18-SSX turnover provides a molecular mechanism for the anti-cancer action of HDAC inhibitors (Figure 5.1 C), and emphasizes their potential significance as a targeted therapy to effectively treat synovial sarcoma. Future investigations should focus on two particularly important directions – how SS18-SSX dissociates from its cofactors (such as, ATF2 and TLE1) prior to its degradation, and how HDAC inhibitors determine Mule association with SS18-SSX for its turnover. Further analysis of post-translational modifications (PTMs) on SS18-SSX fusion protein (and possibly other complex components) would provide more detailed insights into these fundamental questions.
5.4 Clinical Significance
In addition to its scientific contributions, my thesis research is also expected to have an important impact on the human condition. Firstly, these studies provided mechanistic support for disease diagnosis using TLE1. For instance, gene expression profiling and tissue microarray (TMA) analyses of patient specimens suggested that TLE1 could consistently distinguish synovial sarcoma from other types of tumors with similar histology (Foo et al., 2011; Jagdis et al., 2009; Knosel et al., 2010; Lino-Silva et al.,
2011; Mao et al., 2011; Rekhi et al., 2012; Terry et al., 2007; Valente et al., 2013).
However, for a variety of reasons the use of TLE1 as a diagnostic has also recently emerged (Kosemehmetoglu et al., 2009), and this was likely aided by delineation of its
! 122! biological role in synovial sarcoma pathogenesis. My current research contributed to filling this gap in our understanding by showing that TLE1 functions as a critical SS18-
SSX cofactor for its oncogenic activity. Together with the SS18-SSX fusion, TLE1 now has been widely accepted and utilized as a sensitive and specific marker for synovial sarcoma diagnosis. Secondly, the studies described in this thesis could contribute to improving our prognosis of synovial sarcoma and possibly other mesenchymal tumors. A good example is the transcription factor ATF2, which serves as a DNA-binding molecule for SS18-SSX recruitment and repression. It has been reported that ATF2 expression is negatively correlated to clinical outcomes in melanoma patients (Berger et al., 2003;
Gould Rothberg et al., 2009). Consistent with this finding, our immunohistochemistry
(IHC) analysis of multiple mesenchymal tumors (including synovial sarcoma) revealed that ATF2 expression is higher in malignant tumors, compared to benign lesions and normal tissues (Endo et al., 2014). Given the fact that synovial sarcoma is relatively rare, more clinical cases should be collected and analyzed in the future studies, to carefully examine the prognostic value of ATF2. Beside the histological analysis, additional investigations with larger sample sizes would be also necessary to establish a more direct relationship between ATF2 expression level and patient outcomes.
Finally, and maybe most importantly, the molecular mechanisms (see Figure 5.1) proposed in this thesis could provide a biological rationale to aid development of more effective therapeutics against synovial sarcoma, for which current radio/chemotherapies have limited benefit. In particular, HDAC inhibitors have already attracted tremendous clinical attention for use in a number of indications (Campbell and Tummino, 2014; West and Johnstone, 2014). Our studies support show that HDAC inhibitors specifically
! 123! disrupt SS18-SSX function, and effectively suppress synovial sarcoma growth in vitro and in vivo. In addition, it should be noted that in addition to their transcriptional regulatory activities, HDAC inhibitors also influence SS18-SSX protein stability through induction of Mule E3 ligase. Importantly, the clinical relevance of this finding has been recently confirmed by TMA analysis of patient specimens (Figure 5.2; credited to Drs.
Makoto Endo and Torsten O. Nielsen, UBC Genetic Pathology Evaluation Centre).
Although the exact mechanism of Mule downregulation in synovial sarcoma is still unclear, my ongoing research highlights an essential role for the E3 ligase Mouse Double
Minute 2 Homolog (MDM2) in regulation of Mule protein stability (Figure 5.3).
According to these mechanistic findings, use of both HDAC and MDM2 inhibitors might provide a more efficacious therapeutic treatment for synovial sarcoma.
5.5 Regulation of SS18-SSX Intracellular Trafficking
The major body of my thesis research was focused on two aspects: 1) the molecular basis of SS18-SSX action, and 2) the mode of action of HDAC inhibitors in synovial sarcoma. The latter study led to us identifying a previously unexpected mechanism
(Figure 5.4) whereby epigenetic compounds (HDAC inhibitors) post-translationally downregulate SS18-SSX fusion oncoprotein through the ubiquitin proteasomal system.
As discussed in sections 5.3 and 5.4, Mule upregulation appears to be an intermediate step critical for HDAC inhibitor-induced SS18-SSX degradation. However, overexpression of Mule E3 ligase was insufficient to promote SS18-SSX ubiquitination and degradation, suggesting that HDAC inhibitors have an additional function(s) that was missing in the original model proposed in Figure 5.4 (Step 2).
! 124! A possible interpretation to reconcile these findings is that Mule is located within the cytoplasm of synovial sarcoma cells, whereas SS18-SSX is a nuclear transcriptional regulator. Indeed, the evidence from biochemical and immunofluorescence experiments clearly demonstrated that SS18-SSX translocates from the nucleus into the cytoplasm in response to HDAC inhibitor stimulation. Therefore, these data raised a more complicated model (see Figure 5.5) where cytoplasmic accumulation of SS18-SSX is an indispensable step for the anti-cancer action of HDAC inhibitors in synovial sarcoma. Delineation of the mechanism underlying SS18-SSX intracellular trafficking would be an exciting, challenging and important question to address in future investigations. In my preliminary studies, a Leucine-rich nuclear export signal (NES) sequence has been identified in the amino-terminus of SS18-SSX (Figure 5.6 A). The observation that in the absence of this motif SS18-SSX no longer has its shuttling activity (Figures 5.6 B and C) further supports the existence of a previously unknown NES signal during the process of SS18-
SSX nucleocytoplasmic transport. However, given the fact that the native isoform SS18 contains the same NES sequence, it would be necessary to address why only SS18-SSX is exported into the cytoplasm following HDAC inhibitor treatment. This issue has also been raised from reciprocal immunoprecipitation (IP) experiments that show that the nuclear export receptor Chromosome Region Maintenance 1 (CRM1) only associates with SS18-SSX (rather than SS18) after addition of HDAC inhibitors (Figure 5.7). In this regard, future investigations should be directed towards identifying novel cofactors specific for either SS18-SSX or SS18, to further address how the addition of SSX modifies SS18 function. Meanwhile, proteomics approaches should be applied to systematically characterize the PTMs (particularly the lysine acetylation) on SS18-SSX,
! 125! to provide additional molecular insights into mechanisms regulating fusion oncoprotein stability and activity. In summary (Figure 5.8), SS18-SSX fusion oncoprotein functions as a scaffold molecule, which connects multiple transcription regulators and chromatin modifiers to induce gene silencing. However, addition of HDAC inhibitors quickly reverses this action through removal of the TLE1/HDAC/PcG repressor complex and
H3K27me3 repressive marks at fusion oncoprotein-occupied targets. Over a longer term,
HDAC inhibitors play a novel unexpected transcription-independent role in regulating
SS18-SSX stability. This is mediated through modulation of the intracellular localization of the fusion oncoprotein, and its subsequent ubiquitination and proteasome-mediated degradation. Taken altogether, my thesis studies have provided a completely new understanding of synovial sarcoma etiology, but have also identified additional layers of complexity that govern SS18-SSX intracellular activities during tumorigenesis.
! 126!
Figure 5.1 – Schematic for the Mechanism of HDAC Inhibitor Action
(A) Transcription activator ATF2 recruits the SS18-SSX complex to specific gene promoters, whereas the corepressor TLE1 dominant-negatively silences transcription through HDAC/PcG- mediated H3K27me3 repression. (B) Short-term (6-8 hours) exposure to HDAC inhibitors triggers dissociation of TLE1 and its repressive cofactors (HDAC/PcG), and thus reactivates SS18-SSX target gene expression. (C) Long-term (up to 12 hours) exposure to HDAC inhibitors causes a complete disruption of SS18-SSX transcriptional complex through proteasomal degradation of the fusion oncoprotein itself.
! 127!
Figure 5.2 – Mule Tissue Microarray on Patient Specimens
(A) Tissue microarray analysis shows the distribution of Mule protein expression in synovial sarcoma specimens. Histogram of carcinomas serves as a positive control for comparison. (B) Representative images of Mule immunostaining in synovial sarcoma and breast carcinoma. IHC, immunohistochemistry.
! 128!
Figure 5.3 – MDM2 Regulation of Mule in Synovial Sarcoma Cells
(A) Human synovial sarcoma SYO-1 cells were transiently transfected with control or MDM2- specific siRNA. Cell lysates were subjected to western blotting analysis, to measure Mule protein levels. α-Tubulin was used as a loading control. (B) Immunoprecipitation analysis shows that MDM2 interaction with Mule is disrupted in response to HDAC inhibitor stimulation.
! 129!
Figure 5.4 – Schematic for HDAC Inhibitor-Induced SS18-SSX Turnover
HDAC inhibitor treatment shows a dual role in regulation of SS18-SSX stability: (1) upregulating Mule protein levels, and (2) promoting Mule recognition of SS18-SSX.
Figure 5.5 – HDAC Inhibitor Regulation of SS18-SSX Intracellular Trafficking
A proposed model shows that the nucleocytoplasmic shuttling of SS18-SSX is a critical step for HDAC inhibitor-induced Mule-mediated fusion oncoprotein degradation.
! 130!
Figure 5.6 – Identification of SS18-SSX N-terminal NES Signal
(A) netNES (Server 1.1) analysis (la Cour et al., 2004) predicts a leucine-rich nuclear export signal at the amino-terminus of SS18-SSX fusion protein. (B) Human HEK293 cells were transiently transfected with empty vector or Myc-tagged SS18- SSX2 constructs (FL, full-length; ΔN-term, amino-terminal deletion). After addition of vehicle (DMSO) or romidepsin, the nuclear fraction was isolated and subjected to western blotting analysis. Lamin-B was used as a control for nuclear proteins. (C) Under the same conditions described in (B), the cytoplasmic fraction was separated and subjected to western blotting analysis. α-Tubulin was used a control for enrichment of cytoplasmic proteins.
! 131!
Figure 5.7 – SS18-SSX Interaction with CRM1
Human synovial sarcoma SYO-1 cells were treated with vehicle (DMSO) or HDAC inhibitors (such as, SB939 and romidepsin) for 12 hours, followed by 1-hour proteasomal inhibition (with MG-132). Cell lysates were subjected to reciprocal immunoprecipitation using antibodies specific for CRM1 and SS18. Western blotting analysis showed that the nuclear export receptor CRM1 specifically recognizes the fusion protein SS18-SSX in response to HDAC inhibitor treatment. Asterisk (*) indicates the IgG bands.
! 132! ! ! ! ! !
Figure 5.8 – HDAC Inhibitor Regulation of SS18-SSX Action in Synovial Sarcoma
A comprehensive schematic illustrates the molecular mechanisms underlying SS18-SSX- mediated transcriptional repression, and the anti-cancer effect of HDAC inhibitors on SS18-SSX complex assembly and its protein stability in synovial sarcoma cells.
! 133! References
Adhikary S, Marinoni F, Hock A, Hulleman E, Popov N, Beier R, Bernard S, Quarto M, Capra M, Goettig S, Kogel U, Scheffner M, Helin K, Eilers M, 2005. The ubiquitin ligase HectH9 regulates transcriptional activation by Myc and is essential for tumor cell proliferation. Cell 123, 409-421.
Ali SA, Zaidi SK, Dobson JR, Shakoori AR, Lian JB, Stein JL, van Wijnen AJ, Stein GS, 2010. Transcriptional corepressor TLE1 functions with Runx2 in epigenetic repression of ribosomal RNA genes. Proceedings of the National Academy of Sciences of the United States of America 107, 4165-4169.
Alvarez Y, Coll MD, Ortega JJ, Bastida P, Dastugue N, Robert A, Cervera J, Verdeguer A, Tasso M, Aventín A, Guitart M, Caballín MR, 2011. Genetic abnormalities associated with the t(12;21) and their impact in the outcome of 56 patients with B-precursor acute lymphoblastic leukemia. Cancer Genetics and Cytogenetics 162, 21-9.
Baird K, Davis S, Antonescu CR, Harper UL, Walker RL, Chen Y, Glatfelter AA, Duray PH, Meltzer PS, 2005. Gene expression profiling of human sarcomas: insights into sarcoma biology. Cancer Research 65, 9226-9235.
Benbrook D M, Jones N C, 1990. Heterodimer formation between CREB and JUN proteins. Oncogene 5, 295-302.
Berger AJ, Kluger HM, Li N, Kielhorn E, Halaban R, Ronai Z, Rimm DL, 2003. Subcellular localization of activating transcription factor 2 in melanoma specimens predicts patient survival. Cancer Research 63, 8103-8107.
Bett D, Whitehouse S, Antonson P, Shipley J, Cooper C, Goodwin G, 1997. The SYT protein involved in the t(X;18) synovial sarcoma translocation is a transcriptional activator localised in nuclear bodies. Human Molecular Genetics 6, 1559-1564.
Bhoumik A, Fichtman B, Derossi C, Breitwieser W, Kluger HM, Davis S, Subtil A, Meltzer P, Krajewski S, Jones N, Ronai Z, 2008. Suppressor role of activating transcription factor 2 (ATF2) in skin cancer. Proceedings of the National Academy of Sciences of the United States of America 105, 1674-1679.
! 134! Cai W, Sun Y, Wang W, Han C, Ouchida M, Xia W, Zhao X, Sun B, 2011. The effect of SYT-SSX and extracellular signal-regulated kinase (ERK) on cell proliferation in synovial sarcoma. Pathology and Oncology Research 17, 357-367.
Calogero A, Cuomo L, D’Onofrio M, de Grazia U, Spinosanti P, Mercola D, Faggioni A, Frati L, Adamson ED, Ragona G, 1996. Expression of Egr-1 correlates with the transformed phenotype and the type of viral latency in EBV genome positive lymphoid cell lines. Oncogene 13, 2105-2110.
Campbell RM, Tummino PJ, 2014. Cancer epigenetics drug discovery and development: the challenge of hitting the mark. Journal of Clinical Investigation 124, 64-69.
Cao R, Wang L, Wang H, Xia L, Erdjument-Bromage H, Tempst P, Jones RS, Zhang Y, 2002. Role of histone H3 lysine 27 methylation in Polycomb-group silencing. Science 298, 1039-1043.
Carmody Soni EE, Schlottman S, Erkizan HV, Uren A, Toretsky JA, 2014. Loss of SS18- SSX1 inhibits viability and induces apoptosis in synovial sarcoma. Clinical Orthopaedics and Related Research 472, 874-882.
Cassier PA, Lefranc A, Amela EY, Chevreau C, Bui BN, Lecesne A, Ray-Coquard I, Chabaud S, Penel N, Berge Y, Dômont J, Italiano A, Duffaud F, Cadore AC, Polivka V, Blay JY, 2013. A phase II trial of panobinostat in patients with advanced pretreated soft tissue sarcoma. A study from the French Sarcoma Group. British Journal of Cancer 109, 909-914.
Chen D, Kon N, Li M, Zhang W, Qin J, Gu W, 2005. ARF-BP1/Mule is a critical mediator of the ARF tumor suppressor. Cell 121, 1071-1083.
Chen G, Courey AJ, 2000. Groucho/TLE family proteins and transcriptional repression. Gene 249, 1-16.
Chen G, Fernandez J, Mische S, Courey AJ, 1999. A functional interaction between the histone deacetylase Rpd3 and the corepressor groucho in Drosophila development. Genes and Development 13, 2218-2230.
! 135! Clark J, Rocques PJ, Crew AJ, Gill S, Shipley J, Chan AM, Gusterson BA, Cooper CS, 1994. Identification of novel genes, SYT and SSX, involved in the t(X;18)(p11.2;q11.2) translocation found in human synovial sarcoma. Nature Genetics 7, 502-508.
Crew AJ, Clark J, Fisher C, Gill S, Grimer R, Chand A, Shipley J, Gusterson BA, Cooper CS, 1995. Fusion of SYT with two genes, SSX1 and SSX2, encoding proteins with homology to the Kruppel-associated box in human synovial sarcoma. EMBO Journal 14, 2333-2340.
Daley GQ, Van Etten RA, Baltimore D, 1990. Induction of chronic myelogenous leukemia in mice by the P210bcr/abl gene of the Philadelphia chromosome. Science 16, 824-830.
Dasen JS, Barbera JP, Herman TS, Connell SO, Olson L, Ju B, Tollkuhn J, Baek SH, Rose DW, Rosenfeld MG, 2001. Temporal regulation of a paired-like homeodomain repressor/TLE corepressor complex and a related activator is required for pituitary organogenesis. Genes and Development 15, 3193-3207.
Davicioni E, Wai DH, Anderson MJ, 2008. Diagnostic and prognostic sarcoma signitures. Molecular Diagnosis and Therapy 12, 359-274.
Dayyani F, Wang J, Yeh JR, Ahn EY, Tobey E, Zhang DE, Bernstein ID, Peterson RT, Sweetser DA, 2008. Loss of TLE1 and TLE4 from the del(9q) commonly deleted region in AML cooperates with AML1-ETO to affect myeloid cell proliferation and survival. Blood 111, 4338-4347. de Bruijn DR, Allander SV, van Dijk AH, Willemse MP, Thijssen J, van Groningen JJ, Meltzer PS, van Kessel AG, 2006. The synovial-sarcoma-associated SS18-SSX2 fusion protein induces epigenetic gene (de)regulation. Cancer Research 66, 9474-9482. de Bruijn DR, Kater-Baats E, Eleveld M, Merkx G, Geurts Van Kessel A, 2001. Mapping and characterization of the mouse and human SS18 genes, two human SS18-like genes and a mouse SS18 pseudogene. Cytogenetics and Cell Genetics 92, 310-319. de Bruijn DR, Nap JP, van Kessel AG, 2007. The (epi)genetics of human synovial sarcoma. Genes, Chromosomes and Cancer 46, 107-117.
! 136! de Bruijn DR, van Dijk AH, Willemse MP, van Kessel AG, 2008. The C terminus of the synovial sarcoma-associated SSX proteins interacts with the LIM homeobox protein LHX4. Oncogene 27, 653-662. dos Santos NR, de Bruijn DR, Balemans M, Janssen B, Gartner F, Lopes JM, de Leeuw B, van Kessel GA, 1997. Nuclear localization of SYT, SSX and the synovial sarcoma- associated SYT-SSX fusion proteins. Human Molecular Genetics 6, 1549-1558.
Druker BJ, Talpaz M, Resta DJ, Peng B, Buchdunger E, Ford JM, Lydon NB, Kantarjian H, Capdeville R, Ohno-Jones S, Sawyers CL, 2001. Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. New England Journal of Medicine 344, 1031-1037.
Eid JE, Kung AL, Scully R, Livingston DM, 2000. p300 interacts with the nuclear proto- oncoprotein SYT as part of the active control of cell adhesion. Cell 102, 839-848.
Endo M, Su L, Nielsen TO, 2014. Activating transcription factor 2 in mesenchymal tumors. Human Pathology 45, 276-284.
Faour WH, Alaaeddine N, Mancini A, He QW, Jovanovic D, Di Battista JA, 2005. Early growth response factor-1 mediates prostaglandin E2-dependent transcriptional suppression of cytokine-induced tumor necrosis factor-alpha gene expression in human macrophages and rheumatoid arthritis-affected synovial fibroblasts. Journal of Biological Chemistry 280, 9536-9546.
Fisher C, 1986. Synovial sarcoma: ultrastructural and immunohistochemical features of epithelial differentiation in monophasic and biphasic tumors. Human Pathology 17, 996- 1008.
Foo WC, Cruise MW, Wick MR, Hornick JL, 2011. Immunohistochemical staining for TLE1 distinguishes synovial sarcoma from histologic mimics. American Journal of Clinical Pathology 135, 839-844.
Fraga MF, Berdasco M, Ballestar E, Ropero S, Lopez-Nieva P, Lopez-Serra L, Martín- Subero JI, Calasanz MJ, Lopez de Silanes I, Setien F, Casado S, Fernandez AF, Siebert R, Stifani S, Esteller M, 2008. Epigenetic inactivation of the Groucho homologue gene TLE1 in hematologic malignancies. Cancer Research 68, 4116-4122.
! 137! Frith AE, Hirbe AC, Van Tine BA, 2013. Novel pathways and molecular targets for the treatment of sarcoma. Current Oncology Reports 15, 378-385.
Garcia CB, Shaffer CM, Eid JE, 2012. Genome-wide recruitment to Polycomb-modified chromatin and activity regulation of the synovial sarcoma oncogene SYT-SSX2. BMC Genomics 13, 189-202.
Gould Rothberg BE, Berger AJ, Molinaro AM, Subtil A, Krauthammer MO, Camp RL, Bradley WR, Ariyan S, Kluger HM, Rimm DL, 2009. Melanoma prognostic model using tissue microarrays and genetic algorithms. Journal of Clinical Oncology 27, 5772-5780.
Groffen J, Stephenson JR, Heisterkamp N, de Klein A, Bartram CR, Grosveld G, 1984. Philadelphia chromosomal breakpoints are clustered within a limited region, BCR, on chromosome 22. Cell 36, 93-99.
Gure AO, Wei IJ, Old LJ, Chen YT, 2002. The SSX gene family: characterization of 9 complete genes. International Journal of Cancer 10, 448-453.
Hai T, Curran T, 1991. Cross-family dimerization of transcription factors Fos/Jun and ATF/CREB alters DNA binding specificity.!Proceedings of the National Academy of Sciences of the United States of America 88, 3720-3724.
Haldar M, Hancock JD, Coffin CM, Lessnick SL, Capecchi MR, 2007. A conditional mouse model of synovial sarcoma: insights into a myogenic origin. Cancer Cell 11, 375- 388.
Haldar M, Hedberg ML, Hockin MF, Capecchi MR, 2009. A CreER-based random induction strategy for modeling translocation-associated sarcomas in mice. Cancer Research 69, 3657-3664.
Hall JR, Kow E, Nevis KR, Lu CK, Luce KS, Zhong Q, Cook JG, 2007. Cdc6 stability is regulated by the Huwe1 ubiquitin ligase after DNA damage. Molecular Biology of the Cell 18, 3340-3350.
Hayakawa K, Ikeya M, Fukuta M, Woltjen K, Tamaki S, Takahara N, Kato T Jr, Sato S, Otsuka T, Toguchida J, 2013. Identification of target genes of synovial sarcoma-
! 138! associated fusion oncoprotein using human pluripotent stem cells. Biochemical and Biophysical Research Communications 432, 713-719.
Heisterkamp N, Stam K, Groffen J, de Klein A, Grosveld G, 1985. Structural organization of the bcr gene and its role in the Ph' translocation. Nature 315, 758-761.
Herold S, Hock A, Herkert B, Berns K, Mullenders J, Beijersbergen R, Bernards R, Eilers M, 2008. Miz1 and HectH9 regulate the stability of the checkpoint protein, TopBP1. EMBO Journal 27, 2851-2861.
Higa LA, Wu M, Ye T, Kobayashi R, Sun H, Zhang H, 2006. CUL4-DDB1 ubiquitin ligase interacts with multiple WD40-repeat proteins and regulates histone methylation. Nature Cell Biology 8, 1277-1283.
Huang RP, Fan Y, de Belle I, Niemeyer C, Gottardis MM, Mercola D, Adamson ED, 1997. Decreased Egr-1 expression in human, mouse and rat mammary cells and tissues correlates with tumor formation. International Journal of Cancer 72, 102-109.
Huguier S, Baguet J, Perez S, van Dam H, Castellazzi M, 1998. Transcription factor ATF2 cooperates with v-Jun to promote growth factor-independent proliferation in vitro and tumor formation In Vivo. Molecular and Cellular Biology 18, 7020-9.
Ishida M, Miyamoto M, Naitoh S, Tatsuda D, Hasegawa T, Nemoto T, Yokozeki H, Nishioka K, Matsukage A, Ohki M, Ohta T, 2007. The SYT-SSX fusion protein down- regulates the cell proliferation regulator COM1 in t(x;18) synovial sarcoma. Molecular and Cellular Biology 27, 1348-1355.
Ito T, Ouchida M, Ito S, Jitsumori Y, Morimoto Y, Ozaki T, Kawai A, Inoue H, Shimizu K, 2004. SYT, a partner of SYT-SSX oncoprotein in synovial sarcomas, interacts with mSin3A, a component of histone deacetylase complex. Laboratory Investigation 84, 1484-1490.
Ito T, Ouchida M, Morimoto Y, Yoshida A, Jitsumori Y, Ozaki T, Sonobe H, Inoue H, Shimizu K, 2005. Significant growth suppression of synovial sarcomas by the histone deacetylase inhibitor FK228 in vitro and in vivo. Cancer Letter 224, 311-319.
! 139! Jagdis A, Rubin BP, Tubbs RR, Pacheco M, Nielsen TO, 2009. Prospective evaluation of TLE1 as a diagnostic immunohistochemical marker in synovial sarcoma. American Journal of Surgical Pathology 33, 1743-1751.
Jenuwein T, Allis CD, 2001. Translating the histone code. Science 293, 1074-1080.
Jones KB, Su L, Jin H, Lenz C, Randall RL, Underhill TM, Nielsen TO, Sharma S, Capecchi MR, 2013. SS18-SSX2 and the mitochondrial apoptosis pathway in mouse and human synovial sarcomas. Oncogene 32, 2365-2371.
Kadoch C, Crabtree GR, 2013. Reversible disruption of mSWI/SNF (BAF) complexes by the SS18-SSX oncogenic fusion in synovial sarcoma. Cell 153, 71-85.
Kato H, Tjernberg A, Zhang W, Krutchinsky AN, An W, Takeuchi T, Ohtsuki Y, Sugano S, de Bruijn DR, Chait BT, Roeder RG, 2002. SYT associates with human SNF/SWI complexes and the C-terminal region of its fusion partner SSX1 targets histones. Journal of Biological Chemistry 277, 5498-5505.
Kawasaki H, Schiltz L, Chiu R, Itakura K, Taira K, Nakatani Y, Yokoyama KK, 2000. ATF-2 has intrinsic histone acetyltransferase activity which is modulated by phosphorylation. Nature 405,195-200.
Kelliher MA, McLaughlin J, Witte ON, Rosenberg N, 1990. Induction of a chronic myelogenous leukemia-like syndrome in mice with v-abl and BCR/ABL. Proceedings of the National Academy of Sciences of the United States of America 87, 6649-6653.
Knösel T, Heretsch S, Altendorf-Hofmann A, Richter P, Katenkamp K, Katenkamp D, Berndt A, Petersen I, 2010. TLE1 is a robust diagnostic biomarker for synovial sarcomas and correlates with t(X;18): analysis of 319 cases. European Journal of Cancer 46, 1170- 1176.
Konopka JB, Watanabe SM, Witte ON, 1984. An alteration of the human c-abl protein in K562 leukemia cells unmasks associated tyrosine kinase activity. Cell 37, 1035-1042.
Kosemehmetoglu K, Vrana JA, Folpe AL, 2009. TLE1 expression is not specific for synovial sarcoma: a whole section study of 163 soft tissue and bone neoplasms. Modern Pathology 22, 872-878.
! 140!
Kovacsovics T, Maziarz RT, 2006. Philadelphia chromosome-positive acute lymphoblastic leukemia: impact of imatinib treatment on remission induction and allogeneic stem cell transplantation. Current Oncology Reports. 2006 8, 343-351. la Cour T, Kiemer L, Mølgaard A, Gupta R, Skriver K, Brunak S, 2004. Analysis and prediction of leucine-rich nuclear export signals. Protein Engineering, Design and Selection 17, 527-536.
Ladanyi M, 2001. Fusions of the SYT and SSX genes in synovial sarcoma. Oncogene 20, 5755-5762.
Le Beau MM, Espinosa 3rd R, Neuman WL, Stock W, Roulston D, Larson RA, Keinanen M, Westbrook CA, 1993. Cytogenetic and molecular delineation of the smallest commonly deleted region of chromosome 5 in malignant myeloid diseases. Proceedings of the National Academy of Sciences of the United States of America 90, 5484-5488.
Leboucher GP, Tsai YC, Yang M, Shaw KC, Zhou M, Veenstra TD, Glickman MH, Weissman AM, 2012. Stress-induced phosphorylation and proteasomal degradation of mitofusin 2 facilitates mitochondrial fragmentation and apoptosis. Molecular Cell 47, 547-557.
Levin WJ, Press MF, Gaynor RB, Sukhatme VP, Boone TC, Reissmann PT, Figlin RA, Holmes EC, Souza LM, Slamon DJ, 1995. Expression patterns of immediate early transcription factor in human non-small cell lung cancer. Oncogene 11, 1261-1269.
Lim FL, Soulez M, Koczan D, Thiesen HJ, Knight JC, 1998. A KRAB-related domain and a novel transcription repression domain in protein encoded by SSX genes that are disrupted in human synovial sarcomas. Oncogene 17, 2013-2018.
Lim KL, Chew KC, Tan JM, Wang C, Chung KK, Zhang Y, Tanaka Y, Smith W, Engelender S, Ross CA, Dawson VL, Dawson TM, 2005. Parkin mediates nonclassical, proteasomal-independent ubiquitination of synphilin-1: implications for Lewy body formation. The Journal of Neuroscience : the official journal of the Society for Neuroscience 25, 2002-2009.
! 141! Lino-Silva LS, Flores-Gutiérrez JP, Vilches-Cisneros N, Domínguez-Malagón HR, 2011. TLE1 is expressed in the majority of primary pleuropulmonary synovial sarcomas. Virchows Arch 459, 615-621.
Liu H, Deng X, Shyu YJ, Li JJ, Taparowsky EJ, Hu CD, 2006. Mutual regulation of c- Jun and ATF2 by transcriptional activation and subcellular localization. EMBO Journal 25, 1058-1069.
Lopez-Bergami P, Lau E, Ronai Z, 2010. Emerging roles of ATF2 and the dynamic AP1 network in cancer. Nature Reviews Cancer 10, 65-76.
Loya AC, Prayaga AK, Arora A, Sundaram C, Rao IS, Uppin SG, Raju GS, Surath A, Rajappa RS, 2007. Lymph node metastasis of soft tissue tumors: a cytomorphologic sutdy. Acta Cytologica 51, 153-160
Lubieniecka JM, de Bruijn DR, Su L, van Dijk AH, Subramanian S, van de Rijn M, Poulin N, van Kessel AG, Nielsen TO, 2008. Histone deacetylase inhibitors reverse SS18-SSX-mediated polycomb silencing of the tumor suppressor early growth response 1 in synovial sarcoma. Cancer Research 68, 4303-4310.
Maekawa T, Sano Y, Shinagawa T, Rahman Z, Sakuma T, Nomura S, Licht JD, Ishii S, 2008. ATF-2 controls transcription of Maspin and GADD45 alpha genes independently from p53 to suppress mammary tumors. Oncogene 27, 1045-1054.
Maekawa T, Shinagawa T, Sano Y, Sakuma T, Nomura S, Nagasaki K, Miki Y, Saito- Ohara F, Inazawa J, Kohno T, Yokota J, Ishii S, 2007. Reduced levels of ATF-2 predispose mice to mammary tumors. Molecular and Cellular Biology 27, 1730-1744.
Mao RJ, Li QM, Fang HQ, Han FL, Huang XF, Wu YX, Zeng M, 2011. Application of TLE1 expression and fluorescence in-situ hybridization in diagnosing poorly differentiated synovial sarcoma. Zhonghua Bing Li Xue Za Zhi 40, 403-405.
Mills AA, 2010. Throwing the cancer switch: reciprocal roles of polycomb and trithorax proteins. Nature Reviews Cancer 10, 669-682.
Mitelman F, Johansson B, Mertens F, 2007. The impact of translocations and gene fusions on cancer causation. Nature Reviews Cancer 7, 233-245.
! 142!
Mullen JR, Zagars GK, 1994. Synovial sarcoma outcome following conservation surgery and radiotherapy. Radiotherapy and Oncology 33, 23-30.
Muller J, Hart CM, Francis NJ, Vargas ML, Sengupta A, Wild B, Miller EL, O'Connor MB, Kingston RE, Simon JA, 2002. Histone methyltransferase activity of a Drosophila Polycomb group repressor complex. Cell 111, 197-208.
Nagai M, Tanaka S, Tsuda M, Endo S, Kato H, Sonobe H, Minami A, Hiraga H, Nishihara H, Sawa H, Nagashima K, 2001. Analysis of transforming activity of human synovial sarcoma-associated chimeric protein SYT-SSX1 bound to chromatin remodeling factor hBRM/hSNF2 alpha. Proceedings of the National Academy of Sciences of the United States of America 98, 3843-3848.
Nakayama R, Mitani S, Nakagawa T, Hasegawa T, Kawai A, Morioka H, Yabe H, Toyama Y, Ogose A, Toguchida J, Nakayama T, Yoshida T, Ichikawa H, 2010. Gene expression profiling of synovial sarcoma: distinct signature of poorly differentiated type. American Journal of Surgical Pathology 34, 1599-1607.
Nguyen A, Su L, Campbell B, Poulin NM, Nielsen TO, 2009. Synergism of heat shock protein 90 and histone deacetylase inhibitors in synovial sarcoma. Sarcoma 2009, 794901.
Nielsen TO, West RB, Linn SC, Alter O, Knowling MA, O'Connell JX, Zhu S, Fero M, Sherlock G, Pollack JR, Brown PO, Botstein D, van de Rijn M, 2002. Molecular characterisation of soft tissue tumours: a gene expression study. Lancet 359, 1301-1307.
O'Dwyer ME, Druker BJ, 2000. STI571: an inhibitor of the BCR-ABL tyrosine kinase for the treatment of chronic myelogenous leukaemia. Lancet Oncology 1, 207-211.
Palaparti A, Baratz A, Stifani S, 1997. The Groucho/transducin-like enhancer of split transcriptional repressors interact with the genetically defined amino-terminal silencing domain of histone H3. Journal of Biological Chemistry 272, 26604-26610.
Patel SR, Bhumbra SS, Paknikar RS, Dressler GR, 2012. Epigenetic mechanisms of Groucho/Grg/TLE mediated transcriptional repression. Molecular Cell 45, 185-195.
! 143! Pawson T, Gish GD, 1992. SH2 and SH3 domains: from structure to function. Cell 71, 359-362.
Peng C, Guo W, Yang Y, Zhao H, 2008. Downregulation of SS18-SSX1 expression by small interfering RNA inhibits growth and induces apoptosis in human synovial sarcoma cell line HS-SY-II in vitro. European Journal of Cancer Prevention 17, 392-398.
Pretto D, Barco R, Rivera J, Neel N, Gustavson MD, Eid JE, 2006. The synovial sarcoma translocation protein SYT-SSX2 recruits beta-catenin to the nucleus and associates with it in an active complex. Oncogene 25, 3661-3669.
Ravid T, Hochstrasser M, 2008. Diversity of degradation signals in the ubiquitin- proteasome system. Nature Reviews Molecular Cell Biology 9, 679-690.
Rekhi B, Basak R, Desai SB, Jambhekar NA, 2012. Immunohistochemical validation of TLE1, a novel marker, for synovial sarcomas. Indian Journal of Medical Research 136, 766-775.
Rong R, Doxtader EE, Tull J, de la Roza G, Zhang S, 2009. Metastatic poorly differentiated monophasic synovial sarcoma to lung with unknown primary: a molecular genetic analysis. International Journal of Clinical and Experimental Pathology 25, 217- 221.
Ross AJ, Li M, Yu B, Gao MX, Derry WB, 2011. The EEL-1 ubiquitin ligase promotes DNA damage-induced germ cell apoptosis in C. elegans. Cell Death and Differentiation 18, 1140-1149.
Rowley JD, 1973. A new consistent chromosomal abnormality in chronic myelogenous leukaemia identified by quinacrine fluorescence and Giemsa staining. Nature 243, 290- 293.
Sakai T, Ohtani N, McGee TL, Robbins PD, Dryja TP, 1991. Oncogenic germ-line mutations in Sp1 and ATF sites in the human retinoblastoma gene. Nature 353, 83-86.
Sarver AL, Li L, Subramanian S, 2010. MicroRNA miR-183 functions as an oncogene by targeting the transcription factor EGR1 and promoting tumor cell migration. Cancer Research 70, 9570-9580.
! 144!
Shtivelman E, Lifshitz B, Gale RP, Canaani E, 1985. Fused transcript of abl and bcr genes in chronic myelogenous leukaemia. Nature 315, 550-554.
Smith S, Reeves BR, Wong L, Fisher C, 1987. A consistent chromosomal translocation in synovial sarcomas. Cancer Genetics and Cytogenetics 26, 179-180.
Sonobe H, Manabe Y, Furihata M, Iwata J, Oka T, Ohtsuki Y, Mizobuchi H, Yamamoto H, Kumano O, Abe S, 1992. Establishment and chararcterization of a new human synovial sarcoma cell line, HS-SY-II. Laboratory Investigation 67, 498-505.
Soulez M, Saurin AJ, Freemont PS, Knight JC, 1999. SSX and the synovial sarcoma- specific chimaeric protein SYT-SSX co-localize with the human Polycomb group complex. Oncogene 18, 2739-2746.
Su L, Cheng H, Sampaio AV, Nielsen TO, Underhill TM, 2010. EGR1 reactivation by histone deacetylase inhibitors promotes synovial sarcoma cell death through the PTEN tumor suppressor. Oncogene 29, 4352-4361.
Su L, Sampaio AV, Jones KB, Pacheco M, Goytain A, Lin S, Poulin N, Yi L, Rossi FM, Kast J, Capecchi MR, Underhill TM, Nielsen TO, 2012. Deconstruction of the SS18-SSX fusion oncoprotein complex: insights into disease etiology and therapeutics. Cancer Cell 21, 333-347.
Takenaka S, Naka N, Araki N, Hashimoto N, Ueda T, Yoshioka K, Yoshikawa H, Itoh K, 2010. Downregulation of SS18-SSX1 expression in synovial sarcoma by small interfering RNA enhances the focal adhesion pathway and inhibits anchorage- independent growth in vitro and tumor growth in vivo. International Journal of Oncology 36, 823-831.
Taub R, Kirsch I, Morton C, Lenoir G, Swan D, Tronick S, Aaronson S, Leder P, 1982. Translocation of the c-myc gene into the immunoglobulin heavy chain locus in human Burkitt lymphoma and murine plasmacytoma cells. Proceedings of the National Academy of Sciences of the United States of America 79, 7837-7841.
Taylor BS, Barretina J, Maki RG, Antonescu CR, Singer S, Ladanyi M, 2011. Advances in sarcoma genomics and new therapeutic targets. Nature Reviews Cancer 11, 541-557.
! 145!
Terry J, Saito T, Subramanian S, Ruttan C, Antonescu CR, Goldblum JR, Downs-Kelly E, Corless CL, Rubin BP, van de Rijn M, Ladanyi M, Nielsen TO, 2007. TLE1 as a diagnostic immunohistochemical marker for synovial sarcoma emerging from gene expression profiling studies. American Journal of Surgical Pathology 31, 240-246.
Thaete C, Brett D, Monaghan P, Whitehouse S, Rennie G, Rayner E, Cooper CS, Goodwin G, 1999. Functional domains of the SYT and SYT-SSX synovial sarcoma translocation proteins and co-localization with the SNF protein BRM in the nucleus. Human Molecular Genetics 8, 585-591.
Tie F, Banerjee R, Stratton CA, Prasad-Sinha J, Stepanik V, Zlobin A, Diaz MO, Scacheri PC, Harte PJ, 2009. CBP-mediated acetylation of histone H3 lysine 27 antagonizes Drosophila Polycomb silencing. Development 136, 3131-3141.
Valente AL, Tull J, Zhang S, 2013. Specificity of TLE1 expression in unclassified high- grade sarcomas for the diagnosis of synovial sarcoma. Applied Immunohistochemistry and Molecular Morphology 21, 408-413.
Wang J, Wang H, Hou W, Liu H, Zou Y, Zhang H, Hou L, McNutt MA, Zhang B, 2013. Subnuclear distribution of SSX regulates its function. Molecular and Cellular Biochemistry 381, 17-29.
West AC, Johnstone RW, 2014. New and emerging HDAC inhibitors for cancer treatment. Journal of Clinical Investigation 124, 30-39.
Witt O, Deubzer HE, Milde T, Oehme I, 2009. HDAC family: What are the cancer relevant targets? Cancer Letter 277, 8-21.
Wong JP, Reboul E, Molday RS, Kast J, 2009. A carboxy-terminal affinity tag for the purification and mass spectrometric characterization of integral membrane proteins. Journal of Proteome Research 8, 2388-2396.
Zhang J, Kan S, Huang B, Hao Z, Mak TW, Zhong Q, 2011. Mule determines the apoptotic response to HDAC inhibitors by targeted ubiquitination and destruction of HDAC2. Genes and Development 25, 2610-2618.
! 146! Zhao X, D' Arca D, Lim WK, Brahmachary M, Carro MS, Ludwig T, Cardo CC, Guillemot F, Aldape K, Califano A, Iavarone A, Lasorella A, 2009. The N-Myc-DLL3 cascade is suppressed by the ubiquitin ligase Huwe1 to inhibit proliferation and promote neurogenesis in the developing brain. Developmental Cell 17, 210-221.
Zhong Q, Gao W, Du F, Wang X, 2005. Mule/ARF-BP1, a BH3-only E3 ubiquitin ligase, catalyzes the polyubiquitination of Mcl-1 and regulates apoptosis. Cell 121, 1085-1095.
! 147!