The Texas Medical Center Library DigitalCommons@TMC
The University of Texas MD Anderson Cancer Center UTHealth Graduate School of The University of Texas MD Anderson Cancer Biomedical Sciences Dissertations and Theses Center UTHealth Graduate School of (Open Access) Biomedical Sciences
8-2014
TARGETING THE MDM2-P53 AXIS FOR THE TREATMENT OF DEDIFFERENTIATED LIPOSARCOMA
Katelynn Bill
Follow this and additional works at: https://digitalcommons.library.tmc.edu/utgsbs_dissertations
Part of the Laboratory and Basic Science Research Commons, and the Translational Medical Research Commons
Recommended Citation Bill, Katelynn, "TARGETING THE MDM2-P53 AXIS FOR THE TREATMENT OF DEDIFFERENTIATED LIPOSARCOMA" (2014). The University of Texas MD Anderson Cancer Center UTHealth Graduate School of Biomedical Sciences Dissertations and Theses (Open Access). 498. https://digitalcommons.library.tmc.edu/utgsbs_dissertations/498
This Dissertation (PhD) is brought to you for free and open access by the The University of Texas MD Anderson Cancer Center UTHealth Graduate School of Biomedical Sciences at DigitalCommons@TMC. It has been accepted for inclusion in The University of Texas MD Anderson Cancer Center UTHealth Graduate School of Biomedical Sciences Dissertations and Theses (Open Access) by an authorized administrator of DigitalCommons@TMC. For more information, please contact [email protected]. TARGETING THE MDM2-P53 AXIS FOR THE TREATMENT OF DEDIFFERENTIATED LIPOSARCOMA by Kate Lynn Jane Bill, B.S. APPROVED:
Alexander Lazar, M.D., Ph.D. (Co-Advisory Professor)
Dina Chelouche-Lev, M.D. (Co-Advisory Professor)
David McConkey, Ph.D.
Elsa Flores, Ph.D.
Xiongbin Lu, Ph.D.
Rosemarie Schmandt, Ph.D.
APPROVED:
Dean, The University of Texas Graduate School of Biomedical Sciences at Houston TARGETING THE MDM2-P53 AXIS FOR THE TREATMENT OF DEDIFFERENTIATED LIPOSARCOMA
A
DISSERTATION
Presented to the Faculty of The University of Texas Health Science Center at Houston and The University of Texas MD Anderson Cancer Center Graduate School of Biomedical Sciences in Partial Fulfillment
of the Requirements for the Degree of DOCTOR OF PHILOSOPHY
by
Kate Lynn Jane Bill, B.S. Houston, Texas
August 2014
DEDICATION
I dedicate this work to all those who have been afflicted by cancer.
iii ACKNOWLEDGEMENTS
There are many people to whom I owe thanks for help in completion of my dissertation work. First and most importantly my advisor, Dr. Dina Lev, provided an ideal yet challenging environment for me to grow as a researcher and critical thinker, and has influenced my approach to science and research that I’ll carry to all future work. I cannot thank her enough.
Dr. Alexander Lazar, words cannot convey how thankful I am for your support and mentorship throughout my last year of graduate school.
I would like to thank Dr. Raphael Pollock; his vast experience, knowledge and dedication towards the development of my research is deeply appreciated.
I would like to thank the members of my supervisory committee (Dr. David McConkey,
Dr. Elsa Flores, Dr. Xiongbin Lu, and Dr. Rosamarie Schmandt) who dedicated their time to progress my research.
I’d also like to thank all the collaborators who lent me their knowledge, especially Isabelle Meaux for all of her help involving my SAR405838 studies.
I want to thank the Amschwand Foundation for providing my support during my graduate training.
To the teachers who have impacted my life, especially Dr. Richard Garner, you have inspired me to become the scientist I am today.
I owe thanks to each of the members of the Lev lab who I worked with every day. I am forever grateful for the lessons I have learned in science, life and friendship. Paul Cuevas, thank you for everything. You were the glue that held our lab together, and would go above and beyond when I needed you; I am forever grateful. I’d like to thank Dr. Jeannine Garnett
iv for her help and advice in my studies. A special thank you to Dr. Kristelle Lusby for all the memories, struggles and triumphs in lab. Devin Jones, thank you for your advice and friendship since day one of this journey.
Stephanie Hansen, you have been my best friend my entire life, thank you for always being there for me. You are an inspiration to not only me but to so many others, thank you for always reminding me to *be brave*.
Lastly, I’d like to thank my family for their unconditional support. My father, William
Bill, who instilled in me my work ethic and integrity. My mother, Mary Beth Bill, for teaching me compassion and strength, and is the role model for the woman I will always strive to be.
And my brothers, Billy and Ryan, for the years of training in physical and mental toughness.
v TARGETING THE MDM2-P53 AXIS FOR THE TREATMENT OF DEDIFFERENTIATED LIPOSARCOMA
Kate Lynn Jane Bill, Ph.D.
Advisory Professors: Dina Chelouche-Lev, M.D. and Alexander Lazar, M.D., Ph.D.
Dedifferentiated liposarcoma (DDLPS) is an aggressive malignancy characterized by a high rate of recurrence and dismal patient outcome. Minimal improvement in patient survival has been made in the last several decades, highlighting the crucial need for improved therapeutic strategies. A better understanding of the molecular deregulations underlying
DDLPS would facilitate the discovery of improved therapeutic approaches. MDM2 is a well characterized oncoprotein and the most known negative regulator of p53. MDM2 amplification is considered the “hallmark” of DDLPS. Additionally, these tumors are known to harbor wild- type p53. We sought to take advantage of this knowledge and evaluate the role of SAR405838, a novel small molecule inhibitor of MDM2, as a potential anti-DDLPS therapeutic strategy.
To facilitate these studies, we first developed a large set of DDLPS experimental tools, specifically human cell lines and xenograft mouse models. These cell lines were further characterized, confirming the presence of MDM2 amplification and lack of p53 mutations.
Other liposarcoma cell lines harboring p53 mutations, but no MDM2 amplification, were used as controls. Initial experiments demonstrated that SAR405838 was highly active in DDLPS vi cells but not in LPS cells that exhibit mutated p53. Cell culture based assays further demonstrated that SAR405838 selectively targeted MDM2 and led to p53 activation. In
DDLPS cells, SAR405838 resulted in exemplary anti-proliferation, with EC50 values in the low nanomolar range (0.13 to 0.49 nM), as well as successfully induced cell cycle (G1 and G2 phase) arrest and apoptosis in a dose-dependent manner. The anti-DDLPS effects of
SAR405838 were significantly more pronounced when compared to those of the older generation MDM2 inhibitors Nutlin-3a and MI-219. Most importantly, these cell based experiments were replicated in vivo where oral administration of SAR405838 resulted in potent anti-DDLPS effects in a dose-dependent manner (50-200 mg/kg).
Finally, to determine the mechanism of action of MDM2 inhibition in DDLPS cells, gene expression array experiments on SAR405838-treated cells and xenografts were conducted. We identified multiple pathways and genes that can further enhance our understanding of the MDM2-p53 pathway and can be studied as biomarkers of therapeutic response. Taken together, this study highlights the ability of SAR405838 to selectively inhibit a biologically relevant target, MDM2, and induce potent, anti-DDLPS effects in a pre-clinical investigation supporting further evaluation of this compound in the clinical context.
vii TABLE OF CONTENTS
APPROVAL PAGE ...... I
TITLE PAGE ...... II
DEDICATION ...... III
ACKNOWLEDGEMENTS ...... IV
TABLE OF CONTENTS ...... VIII
LIST OF FIGURES ...... XIV
LIST OF TABLES ...... XVII
LIST OF ABBREVIATIONS ...... XIX
CHAPTER 1. INTRODUCTION AND BACKGROUND ...... 1
Section 1. Dedifferentiated liposarcoma: an aggressive histological subtype of liposarcoma...... 1 Liposarcoma classification...... 1 Pathological and molecular differences in liposarcoma subtypes ...... 1 Well-differentiated/dedifferentiated liposarcoma ...... 1 Myxoid/Round cell liposarcoma ...... 4 Pleomorphic liposarcoma ...... 4 Dedifferentiated liposarcoma ...... 6 Incidence ...... 7 Symptoms and diagnosis...... 7 Therapeutic treatment options for dedifferentiated liposarcoma ...... 8 Surgical resection ...... 8 Conventional chemotherapy ...... 8 Radiation therapy ...... 9 Novel systemic therapies ...... 10 Inhibition of MDM2 ...... 11 Outcome ...... 12
viii Section 2. Molecular driving forces of dedifferentiated liposarcomagenesis ...... 13 Genomic aberrations associated with dedifferentiated liposarcoma ...... 13 Other deregulations: receptor tyrosine kinases ...... 17
Section 3. The MDM2:p53 axis ...... 18 P53: The guardian of the genome ...... 18 Stabilization and activation of p53 ...... 18 Functions of p53: Downstream responses to activation ...... 21 Structure ...... 21 The p53 pathway ...... 22 Cell cycle ...... 24 Apoptosis ...... 28 Loss of p53 activity in cancer ...... 31 Regulation of p53 function ...... 31 Cellular localization ...... 32 The MDM2:p53 negative feedback loop ...... 32 MDM2...... 34 Identification and biological function ...... 34 Structure ...... 34 Master regulator of p53 ...... 35 MDM2 functions as an E3 ubiquitin ligase ...... 36 MDM2 regulation ...... 36 Other proteins that interact with MDM2 ...... 37 MDM2 overexpression in cancer ...... 38
Section 4. Targeting the MDM2:p53 with small molecule inhibitors as a therapeutic strategy for DDLPS ...... 40 Reactivation of the p53 pathway with small molecule MDM2 inhibitors ....40 Small molecule inhibitors of the MDM2-p53 protein-protein interaction ...... 40 Mechanism of action ...... 42 Nutlins ...... 44 MI-219 ...... 45
ix Anti-tumorigenic responses of MDM2 inhibition ...... 46 SAR405838: a novel small molecule MDM2 inhibitor ...... 47
HYPOTHESIS AND SPECIFIC AIMS ...... 48
Hypothesis...... 48
Specific Aims ...... 48
SIGNIFICANCE ...... 49
CHAPTER 2. MATERIALS AND METHODS ...... 50 Isolation and establishment of human DDLPS cell lines from solid tumor ...... 50 Cell Culture ...... 51 Short Tandem Repeat (STR) ...... 51 Fluorescence in situ Hybridization (FISH) ...... 52 gDNA extraction ...... 53 RNA extraction and quantitative real-time PCR ...... 53 Western blotting ...... 56 Antibodies for western blot analyses ...... 56 P53 Mutational Analysis ...... 57 Hematoxylin and eosin (H&E) staining ...... 57 Immunohistochemistry ...... 58 Array CGH ...... 59 Gene expression array ...... 60 Ingenuity Pathway Analysis ...... 60 Reagents for experimental procedures ...... 60 Cell proliferation analysis ...... 61 Cell cycle analysis...... 62 Annexin V/ FITC FACs analysis for detection of apoptosis ...... 62 siRNA knockdown ...... 62 Pharmacokinetics ...... 63 DDLPS xenograft models ...... 64 Statistical analysis ...... 64 x CHAPTER 3. RESULTS: CREATION AND CHARACTERIZATION OF DEDIFFERENTIATED LIPOSARCOMA BIORESOURCES ...... 66
Section 1. Creation of dedifferentiated liposarcoma experimental tools ...... 66 Generation of patient derived DDLPS cell lines ...... 66 Conservation of molecular characteristics ...... 67 TP53 mutational analysis ...... 72 Development of DDLPS xenograft models ...... 73 Preservation of DDLPS histopathology in xenograft models ...... 73
Section 2. Characterization of dedifferentiated liposarcoma cell lines ...... 75 Genomic profiling by array CGH analysis of DDLPS cell lines ...... 75 Identifying differential gene expression patterns in DDLPS cell lines...... 81
CHAPTER 4. RESULTS: THE MDM2:P53 AXIS IS A TARGETABLE THERAPEUTIC STRATEGY IN DEDIFFERENTIATED LIPOSARCOMA ...... 84
Section 1. Targeting the MDM2:p53 axis in vitro ...... 84 SAR405838 inhibition stabilizes p53 and activates downstream targets ...... 84 SAR405838 inhibits DDLPS cell growth ...... 88 Wild type p53 is required for effective anti-DDLPS SAR405838 effects ...... 91 Activation of the p53 pathway halts cell cycle progression ...... 94 SAR405838 induces p53 apoptotic response ...... 97 Amplification of MDM2 corresponds to therapeutic response ...... 100
Section 2. Targeting the MDM2:p53 axis in vivo ...... 102 Pharmacokinetics of SAR405838 ...... 102 SAR405838 elicits potent anti-DDLPS tumor growth effects in xenograft models ...... 106 Therapeutic effects of SAR405838 in large DDLPS tumor burden ..110
xi CHAPTER 5. RESULTS: GENE EXPRESSION PROFILING IDENTIFIES IMPORTANT GENES MODULATED BY SAR405838 ...... 112
Section 1. Identifying molecular changes induced by SAR405838 for potential biomarkers of therapeutic response ...... 112 Identifying molecular deregulations induced by SAR405838 in dedifferentiated liposarcoma in vitro ...... 112 Identifying molecular deregulations induced by SAR405838 in dedifferentiated liposarcoma in vivo ...... 117
Section 2. Canonical pathways enriched by genes differentially expressed following treatment with SAR405838 ...... 119 Microarray gene profiling in DDLPS cells treated with SAR405838 demonstrated up- and down-regulated genes in a p53-dependent manner...... 122 SAR405838 modulates the expression of genes in the 12q13~15 chromosomal loci ...... 128
CHAPTER 6. DISCUSSION ...... 130 Establishment and characterization of novel DDLPS models ...... 130 SAR405838 restores the p53 pathway in DDLPS cells ...... 134 P53; a choice between cell life and death ...... 138 Anti-tumor response of SAR405838 in xenograft models ...... 140 Predictors and biomarkers of therapeutic response ...... 141 Therapeutic resistance ...... 144 Enhancing the therapeutic efficacy of MDM2 inhibitors with a combination setting ...... 146 Other deregulations contributing to dedifferentiated liposarcomagenesis ...... 147
xii CHAPTER 7. SUMMARY ...... 149
CHAPTER 8. FUTURE DIRECTIONS ...... 151
APPENDIX ...... 155
Appendix A. aCGH Data for Dedifferentiated Liposarcoma Cell lines...... 155
Appendix B: Differential Gene expression patterns in Dedifferentiated Liposarcoma...... 158
Appendix C. Top 50 differentially expressed genes in vitro and in vivo following SAR405838 treatment...... 162
Appendix D. The hepatocyte growth factor receptor as a potential therapeutic target for dedifferentiated liposarcoma...... 164
Appendix E. SAR405838: A novel and potent inhibitor of the MDM2:p53 axis for the treatment of dedifferentiated liposarcoma ...... 165
VITA ...... 166
REFERENCES ...... 167
xiii LIST OF FIGURES
Figure 1.1 WDLPS and DDLPS share a common genetic aberration:
12q13~15 amplification ...... 3
Figure 1.2 p53 influences multiple cellular stress pathways...... 21
Figure 1.3 Schematic representation of p53 structural domains...... 22
Figure 1.4 Schematic of the cell cycle...... 24
Figure 1.5 p53 cell cycle regulation...... 27
Figure 1.6 Schematic of extrinsic and intrinsic apoptotic pathways...... 30 Figure 1.7 Schema of the autoregulatory negative feedback loop between
MDM2 and p53 ...... 33
Figure 1.8 Structure of MDM2 protein...... 35 Figure 1.9 Small molecule inhibitors of MDM2 inhibit the MDM2-p53
complex...... 43
Figure 1.10 The chemical structure Nutlin-3a...... 44
Figure 1.11 The chemical structure MI-219...... 45
Figure 1.12 The chemical structure SAR405838...... 47 Figure 3.1 Fluorescence in situ hybridization (FISH) for the detection of
MDM2 amplification in cell panel...... 69
Figure 3.2 MDM2 is amplified and overexpressed in human DDLPS...... 71
Figure 3.3 MDM2 and p53 protein expression in LPS panel. Western ...... 72 Figure 3.4 Histological analysis of DDLPS tumors from patients and mouse
xenografts ...... 74
Figure 3.7 The 12q13~15 amplification in DDLPS cell lines...... 80 Figure 3.8 Differentially expressed genes between DDLPS cell lines and
controls...... 82
xiv Figure 4.1 MDM2 antagonists stabilize p53 and induce the p53 pathway in
DDLPS cells...... 86 Figure 4.2 SAR405838 stabilizes p53 and induces the p53 pathway in WT
p53 DDLPS cells...... 87
Figure 4.3 Small molecule MDM2 inhibitors produce anti-proliferative effects. .89 Figure 4.4 SAR405838 elicits anti-proliferative effects in DDLPS cell line
panel...... 90
Figure 4.5 SAR405838 efficacy is dependent on the presence of WT p53...... 93
Figure 4.6 Inhibition of the MDM2:p53 complex induces cell cycle arrest...... 95
Figure 4.7 SAR405838 induces cell-cycle arrest...... 96
Figure 4.8 Inhibition of the MDM2:p53 complex induces apoptosis...... 98
Figure 4.9 Apoptotic response of DDLPS cells to SAR405838...... 99
Figure 4.10 Time-concentration profiles ...... 103 Figure 4.11 Expression levels from DDLPS xenograft models following
SAR405838 treatment...... 105 Figure 4.14 The in vivo effects of SAR405838 on established large tumor
burdens...... 111
Figure 5.1 Schema of the experimental outline...... 113 Figure 5.2 Gene expression profiling in DDLPS cells in response to
SAR405838...... 115
Figure 5.3 Gene Ontology Enrichment in DDLPS with SAR405838 treatment. 116 Figure 5.4 Gene expression profiles in response to SAR405838 treatment in
vivo...... 118
Figure 5.5 Canonical pathways modulated by differentially expressed genes. ...120
xv Figure 5.6 Differentially expressed genes following SAR405838 treatment
involved in cell cycle control...... 123
Figure 5.7 Connections between p53 signaling and the mitotic pathway...... 125
Figure 5.8 SAR405838 induces apoptotic gene profiles in DDLPS...... 127
xvi LIST OF TABLES
Table 1.1 Important differences associated with liposarcoma subtypes...... 5 Table 1.2 Overview of current clinical trials for novel systemic therapies in
liposarcoma...... 11 Table 1.3 A summary of genomic aberrations and their functional
consequences in DDLPS...... 16
Table 1.4 Proteins influencing p53 stabilization and activation ...... 20
Table 1.5 p53 responses and associated target genes...... 23
Table 1.6. Proteins that interact with MDM2 function and regulation...... 38
Table 1.7 Small molecule inhibitors of the MDM2-p53 complex...... 41
Table 2.1 qRT-PCR primer sequences...... 55 Table 3.1 Characteristics of DDLPS cell lines frequently used for cell line
panel...... 67
Table 4.1 EC50 values following treatment with MDM2 ...... 89
Table 4.2 EC50 values following SAR405838 treatment...... 91
Table 4.3 Relative MDM2 amplification; MDM2HI and MDM2LO...... 101
Table 4.4 Summary of PK parameters...... 103
Table 4.5 PK parameters ratio versus dose...... 103
Table 4.6 Tumor volume of Lipo863 tumor bearing mice...... 107
Table 5.1 Top three Ingenuity canonical pathways...... 121 Table 5.2 Genes identified by both microarray analyses involved in cell
cycle regulation ...... 123 Table 5.3 Genes identified by both microarray analyses involved in
apoptosis ...... 127
xvii Table 5.4 Genes identified by both microarray analyses located within the
12q13~15 chromosomal interval...... 129
xviii LIST OF ABBREVIATIONS aCGH array comparative genomic hybridization ATM ataxia telangiectasia mutated ATR ataxia telangiectasia and Rad3 related Bcl-2 B cell lymphoma-2 BIW bi-weekly CDK Cyclin dependent kinase CDK4 cyclin dependent kinase 4 Chk 1 or 2 checkpoint kinase 1 or 2 CNV copy number variation CT computed tomography DISC death-inducing signaling complex DDLPS dedifferentiated liposarcoma DMEM Dulbecco’s modified Eagle’s medium DMSO dimethyl sulfoxide DOX doxorubicin E2F1 E2 promoter binding factor 1 EC50 Effective concentration (50) FACS fluorescence assisted cell sorting FADD Fas-associated death domain FBS fetal bovine serum GADD45 growth arrest and DNA damage-inducible 45 GDF-15 growth differentiation factor 15 GIST gastrointestinal stromal tumor HAUSP Herpes virus-associated ubiquitin specific protease hMSC-AT human mesenchymal stem cells-adipose tissue HMGA2 high mobility group AT-hook2 FISH fluorescence in situ hybridization FRS2 Fibroblast growth factor receptor substrate 2 IPA Ingenuity Pathway Analysis LPS liposarcoma MDM2 murine double minute 2 MI-219 MDM2 inhibitor-219 MLPS myxoid liposarcoma p.o. per os PBS phosphate buffered saline PBST PBS containing 0.05% tween 20 PCR polymerase chain reaction PK pharmacokinetics PLS pleomorphic liposarcoma
xix PUMA p53 up-regulated modulator of apoptosis qRT-PCR quantitative reverse transcriptase PCR RCLPS round cell liposarcoma RNA ribonucleic acid RT radiation therapy RTK receptor tyrosine kinase SCID severe combined immunodeficiency siRNA small interfering RNA SNP single-nucleotide polymorphism STR short tandem repeats STS soft tissue sarcoma P53 tumor protein 53 WB western blot WDLPS well-differentiated liposarcoma WHO World Health Organization wk week WT wild-type
xx CHAPTER 1. INTRODUCTION AND BACKGROUND
SECTION 1. DEDIFFERENTIATED LIPOSARCOMA: AN AGGRESSIVE HISTOLOGICAL SUBTYPE OF LIPOSARCOMA
Liposarcoma classification Soft tissue sarcomas (STS) are a heterogeneous group of malignancies consisting of over 50 different histological subgroups all of which show mesenchymal differentiation1.
Liposarcoma (LPS), as the name suggests, is a malignancy of adipocyte-derived cells1. The
World Health Organization (WHO) classifies liposarcomas into three distinct subtypes based on their unique characteristics and molecular aberrations: (1) Well- differentiated/dedifferentiated liposarcoma, (2) Myxoid/round cell liposarcoma, and (3)
Pleomorphic liposarcoma1. Well-differentiated/dedifferentiated liposarcoma is the most prevalent histological subtype, accounting for ~65% of all LPS; myxoid/round cell LPS account for ~30%; and pleomorphic LPS accounts for <5% of all liposarcomas1,2.
Pathological and molecular differences in liposarcoma subtypes
Well-differentiated/dedifferentiated liposarcoma Although grouped together, Well-differentiated (WDLPS) and dedifferentiated
(DDLPS) have very different tumorigenic phenotypes. WDLPS, sometimes referred as atypical lipomatous tumor, is classified as low grade tumors due to its local aggressiveness, but lack of metastatic potential1. This lesion is pathologically characterized by aberrantly enlarged and hyperchromatic nuclei and is comprised of malignant adipocytes and spindle cells, the latter thought to show fibroblastic/myofibroblastic differentiation1,2.
1 Originally described by Evans in 1979, DDLPS is a biphasic tumor consisting of high- grade cellular and non-lipogenic areas with prominent mitoses juxtaposed to a usually more adipocyte-rich WDLPS portion3. Although it is a malignant adipocytic neoplasm, the adipocytic differentiation of the high grade component is no longer recognizably adipocytic4.
On gross examination, these tumors show solid, non-lipomatous dedifferentiated areas sharply demarcated from the often multi-nodular and yellow mass that constitutes the well differentiated portion of the tumor4,5.
Together, the WD/DDLPS subgroup share a unique and defining molecular aberration, namely the presence of supernumerary circular ring and giant rod chromosomes that harbor amplification of the 12q13~15 chromosomal region6–8 (Figure 1.1). There are hundreds of genes located in this chromosomal region whose amplifications can be punctuated and/or discontinuous, and the copy number increase can extend over 100-fold9–11. Unlike other genes in this interval that are variably amplified from case to case, amplification of the murine double minute 2 (MDM2) gene is seen in virtually 100% of cases and is considered a molecular “hallmark” of this subgroup6–8. Further discussion of these genetic aberrations as well as other molecular derangements will be discussed later in this introduction.
Nevertheless, as our understanding of the molecular underpinnings of DDLPS have advanced,
WDLPS and DDLPS identifiably possess unmistakably different characteristics. It remains unclear if WDLPS and DDLPS represent a tumorigenic continuum that accumulates additional aberrations over time or if these diseases represent two separate and distinct lesions that arise from related yet distinct adipogenic cells of origin whilst sharing common molecular aberrations, such as the genetic amplification of the 12q13~15 chromosomal region8,10–13.
2
Figure 1.1 WDLPS and DDLPS share a common genetic aberration: 12q13~15 amplification. The shared molecular hallmark of WDLPS and DDLPS is the presence of one or more supernumerary circular ring and/or giant rod chromosomes containing highly amplified DNA sequences localized to the 12q13~15 region. Fluorescence in situ hybridization (FISH) shows the amplified copes of this region (red), as well as the 12q centromeric probe (green).
3 Myxoid/Round cell liposarcoma The second most prevalent subtype of LPS, myxoid/round cell (MLPS/RCLPS), is characterized by distinct chicken-wire vasculature, a sparse cellular component consisting of spindle or ovoid cells, mature adipocytes, and immature signet ring lipoblasts set in an abundant myxoid stroma1,14. The molecular hallmark of MLPS/RCLPS is the recurrent translocation of chromosomes 12 and 16 t(12;16)(q13;p11) that results in the FUS-CHOP gene fusion present in over 95% of cases1,2,14. Alternatively, translocation in
(t12;22)(q13;q12) results in an EWS-CHOP fusion gene14. Both scenarios are thought to interfere with adipocytic differentiation14,15.
Pleomorphic liposarcoma Lastly, pleomorphic liposarcoma (PLS) is the rarest subtype of LPS1.
Morphologically, PLS is a high grade neoplasm that contains varying numbers of multi- vacuolated lipoblasts and sometimes multi-nucleated giant cells admixed into a high grade spindle cell neoplasm1,16. PLS is molecularly displays a complex karyotype characterized by aneuploidy and complex chromosomal rearrangements1. The most common genetic aberration is deletion of the 13q14.2~14.3 chromosomal region that occurs in over 60% of PLS; however, due to the rarity of the disease, molecular studies are limited2. Of note, TP53 mutations have been observed in PLS, which rarely occurs in the other two LPS histological subtypes.
Inactivating mutations in NF1 have also been observed17.
4
TABLE 1.1 IMPORTANT DIFFERENCES ASSOCIATED WITH LIPOSARCOMA SUBTYPES.
Cytogenetic Molecular Histological Histology aberrations genetics features Well- Ring chromosomes Amplification of Adipocytes with differentiated and giant markers MDM2, CDK4, aberrant nuclei LPS (12q13~15) HMGA2 Ring chromosomes Amplification of Highly cellular area Dedifferentiated and giant markers MDM2, CDK4, juxtaposed to a WD LPS (12q13~15) HMGA2 portion Extracellular myxoid FUS-CHOP/DDIT3 Translocation material; mature Myxoid/Round EWS- (t12;16)(q13;p11) adipocytes and Cell LPS CHOP/DDITS (t12;22)(q13;q22) immature lypoblasts, fusion protein round cells Highly cellular; Pleomorphic TP53 mutations in Complex karyotype pleopmorphic LPS 60% lipoblasts
5 Dedifferentiated liposarcoma Although grouped together due to their shared genetic abnormalities, WDLPS and
DDLPS pursue two very different clinical courses. WDLPS represents the largest subgroup of LPS, accounting for nearly 40-45%1. WDLPS has a peak incidence in the 6th decade of life and affects males and females equally1. Clinically, WDLPS presents as a deep-seated, painless mass often found in the retroperitoneum; however, it can be found in deep soft tissue of the thigh, paratesticular area, as well as the mediastinum1. While WDLPS does not metastasize, it has a high recurrence rate (up to 90%) following surgical resection and repeated recurrences can lead to extreme morbidity and even death. The location of these tumors is an important prognostic factor. Tumors found in very deep sites are more capable of recurrence ( >40% in the retroperitoneum compared to <2% in the extremities) and, due to uncontrollable local effects, can cause serious problems for patients1. This location is also an important factor determining mortality; essentially no extremity WDLPS leads to death, whereas 80% of retroperitoneal WDLPS leads to disease specific mortality within 10-20 years of initial diagnosis. Overall, the five-year disease-specific survival rate for WDLPS is 85%.
Nevertheless, the most important prognostic characteristic of WDLPS is the risk of dedifferentiation upon recurrence (20%)2. The term “dedifferentiation” was first described in chondrosarcoma in 1991 to explain a phenomenon in which a relatively well-differentiated lesion accumulates a clearly demarcated component of high-grade tumor cells no longer showing the same differentiation as the tumor in which it arises18,19. This dedifferentiation results in an overall more aggressive phenotype (15% incidence of distant metastasis) and poor patient outcome; therefore, given the severity of this tumor, DDLPS is the histological
LPS subtype that will be the focus of this study.
6 Incidence Soft tissue sarcomas are rare malignancies that constitute approximately 1% of all adult cancers (10-12,000 new cases per year in the US)1,20; LPS is the most common adult histological subtype of STS, accounting for nearly 40%1 of all STS cases. The specific incidence of DDLPS is approximately 1/330,000 persons/year21. There are no gender predilections, nor are there currently any known risk factors for the development of DDLPS22.
This disease typically manifests during the sixth to seventh decade of life22. DDLPS most frequently occurs in the retroperitoneum; however, it can also occur in the deep soft tissue of the extremities, trunk, mediastinum, head and neck region and spermatic cord22.
Symptoms and diagnosis Patients with DDLPS found in the retroperitoneum can often present with very large masses (>30cm) because the retroperitoneal space is able to accommodate much larger volumes of a symptomatic disease than other areas of the body23. Symptoms are often associated with pressure caused by the lesion on the surrounding organs; e.g. abdominal pain, and constipation 24.
The diagnosis of DDLPS is usually confirmed by several techniques. Upon physical examination, a firm palpable mass is often detectable. Computed topography (CT) imaging or magnetic resonance imaging (MRI) scanning is critical for staging purposes and used to assess the size, involved organs and vasculature of the lesion25,26.
The next step in diagnostic work up consists of biopsy27. Usually, multiple core needle biopsies are performed; however, for small (<5 cm) superficial specimens, an excisional biopsy may be possible25. Pathologists consider various cytogenetic correlations associated with DDLPS specifically along with histological considerations to diagnose the tumor16. The genetic amplification of MDM2 has proven to be especially helpful as a diagnostic finding6,28.
7 Therapeutic treatment options for dedifferentiated liposarcoma Despite advances in the field, few new treatment options have emerged. Treatment of
DDLPS is often determined on the basis of certain factors such as age of patient, size and location of the lesion, and involvement of surrounding organs22. Consequently, prospective multidisciplinary treatment planning is crucial to provide the best possible outcomes.
Surgical resection The current standard of care for DDLPS is surgical removal of the tumor in its entirety29. However, surgery can often be difficult due to tumor location and ability of DDLPS to invade into surrounding organs. Patients often must undergo multiple surgeries due to tumor local recurrence, which can lead to significant post-surgical morbidity.
Conventional chemotherapy Even following complete surgical resection, more than 80% of DDLPS patients will develop unresectable recurrences30,31. For patients with unresectable tumors, chemotherapy is frequently used in an attempt to shrink the tumor to a resectable size. The most frequently used cytotoxic chemotherapeutic agents are doxorubicin (DOX) and ifosfamide, which is often used in combination to DOX32,33.
Doxorubicin-based chemotherapy has been the most commonly used treatment for adult STS since the 1970’s and is the standard first-line agent for treatment34. DOX acts by inhibiting topoisomerase II, resulting in DNA double-strand breaks35–37. Cells then activate the DNA damage response signaling cascade to guide recruitment of the repair machinery to address these breaks38. If this fails, the DNA repair program initiates apoptosis. DOX is usually administered intravenously for 96 hours once every three weeks; the threshold dose for optimal activity can be as high as 120mg/m2 per cycle in selected patients39.
8 A meta-analysis of 18 randomized prospective clinical trials of 1953 patients was performed to assess the impact of adjuvant doxorubicin-based chemotherapy. In this study it was observed that the odds ratio for local recurrence as well as distant and overall recurrence
(0.73 and 0.67, respectively) statistically favored chemotherapy; however, this was not associated with a significant improvement in survival40.
In the early 1990s, ifosfamide-based treatment was deployed as a new and possibly improved treatment for progressive disease32,41. Ifosfamide is a nitrogen mustard alkylating agent that is delivered intravenously; the threshold for optimal activity 6 g/m2 per cycle42,43; however, in selected patients doses as high as 10-14 g/m2/cycle may be feasible.
In one phase III clinical trial, Lorigan et al. (2007) compared the effects of ifosfamide to doxorubicin as a single agent regarding toxicity, response rate, progression-free and overall survival. They concluded that while both drugs exhibited similar anti-tumor effects, single- agent DOX had a better therapeutic impact than ifosfamide alone due to the comparatively increased toxicity in patients receiving ifosfamide44.
While these chemotherapeutic agents have been considered the front-line treatment option when surgery is not an option, DDLPS has proven to be resistant to such conventional chemotherapy22. Furthermore, neither single-agent nor combination therapies have proven survival benefits in clinical trials45,46.
Radiation therapy Use radiation therapy (RT) has been shown to have an effect on local control but no significant effect on the overall patient survival47,48. For patients with deep seated lesions, RT in conjunction with surgery appears to improve local control25.
9 RT has been used in combination with chemotherapeutic agents, with minimal improvement in recurrence-free and overall survival. In an 83 patient trial (27/83 patients diagnosed with DDLPS), Gronchi et al. (2014) administered a combination of ifosfamide
(three cycles; 14 g/m2) and radiotherapy (total dose up to 50.4Gy) prior to surgery as part of a phase I-II study for localized liposarcoma. Most patients were unable to complete the full course of pre-operative RT. Gronchi showed that in patients diagnosed with DDLPS, 33% had locoregional recurrence, 25.9% had distant recurrence and only one patient (3.7%) had both local and distant recurrence. Compared to other soft tissue sarcomas in this study (well- differentiated liposarcoma, leiomysarcoma, and “others”), DDLPS tumors responded the worst overall, with an overall progression-free survival in only 10/83 patients49. Currently, there is an on-going phase III clinical trial to study the effects of pre-operative RT followed by surgery (clinical trials identifier: NCT01344018) versus surgery alone, comparing recurrence-free, metastasis-free, and overall survival of these patient groups.
Novel systemic therapies Novel systemic therapies targeting the unique genetic aberrations specific to DDLPS have emerged in the past decade based on disease biology, including genetic and molecular aberrations; thereby, increasing the specificity and efficacy of a drug. A list of currently on- going clinical trials can be found in Table 1.2.
10 TABLE 1.2 OVERVIEW OF CURRENT CLINICAL TRIALS FOR NOVEL SYSTEMIC THERAPIES IN LIPOSARCOMA.
Mechanism of Therapy Author (year) Cohort Study action Cabazitaxel + Prolonged Hayward (2013 - interferes with DDLPS Phase II infusional NCT01913652) microtubules ifosfamide Anti-microtubule Eribulin Schöffski (2011) DDLPS Phase II agent Flavopiridol Luke (2012) pan-CDK inhibitor WDLPS/DDLPS Phase I CDK4 and CDK6 PD0332991 Dickson (2013) WDLPS/DDLPS Phase II inhibitor Tyrosine kinase All histologic Pazopanib Sleijfer (2009) Phase II receptor inhibitor subtypes of LPS Von Mehren Tyrosine kinase All histologic Sorafenib Phase II (2012) receptor inhibitor subtypes of LPS Tyrosine kinase All histologic Sunitinib Mahmood (2011) Phase II receptor inhibitor subtypes of LPS Troglitazone, Debrock (2003), PPAR-γ agonist All histologic Phase I
Inhibition of MDM2 Targeting a specific molecular aberration in a certain tumor type will hopefully better enable physicians to treat patients with drugs that will have improved effects on outcome. The amplification of MDM2 in DDLPS has proven to be reliable in clinical diagnosis8,28,50,51. As will be explained below, MDM2 as a negative regulator of the tumor protein p53 (p53) tumor suppressor gene implies that targeting of MDM2 may be a potential approach for DDLPS therapy. Recently, Ray-Coquard et al. (2012) has reported an initial WDLPS/DDLPS patient directed clinical trial using the MDM2 inhibitor RG711252. In brief, 20 untreated patients were enrolled in a phase 1 proof-of-mechanism study of RG7112. Resected tumors demonstrated that RG7112 restored p53 activity and elicited anti-tumorigenic properties52. Clinical trials specific to the molecular drivers of DDLPS pave the way for future novel therapeutic options;
11 hence a better understanding of molecular driving forces promoting tumorigenesis is critical to develop more effective therapeutic strategies.
Outcome Unfortunately, DDLPS often results in a dismal fate for afflicted patients. DDLPS may arise de novo (approximately 90% of cases) or in areas of recurrence from the WD precursor lesions (~10% of cases; i.e., secondary DDLPS)23. Various predictors of outcome have been investigated; it has been observed that in patients with primary lesions, the median progression-free survival and overall survival are 21.1 and 59 months, respectively. However, this high recurrence rate is one of the most notorious characteristics of the disease which has the propensity to recur locally at rates as high as 90%, most frequently in the retroperitoneum
(92%) or peritoneal cavity (26%)53. From the presentation of the initial primary lesion, the local- and distant recurrence-free survival is approximately 21.5 and 45.8 months53.
Furthermore, patients diagnosed with primary DDLPS may have as many as three or more episodes of local recurrences before metastatic dissemination occurs23. Unlike its
WDLPS counterpart, DDLPS has the potential to metastasize, observed in >20% of patients, typically to the lungs23. In patients with advanced disease, the average elapsed time between primary diagnosis and metastasis is 25 months, with a median survival time of 11.5 months for metastatic patients. Even worse is the grim five year disease free survival rate of 5.2% for advanced disease23.
The high rate of recurrence and disease progression, coupled with unacceptably low survival rates regardless of treatment, renders DDLPS a deadly disease.
12 SECTION 2. MOLECULAR DRIVING FORCES OF DEDIFFERENTIATED LIPOSARCOMAGENESIS
Genomic aberrations associated with dedifferentiated liposarcoma Identifying the molecular underpinnings of DDLPS is a critical step in understanding the biology of DDLPS. Advancements with high-throughput techniques in tissue based studies have propelled our understandings of the molecular deregulations in DDLPS54–56. These recent developments in genomic studies have provided critical information for genetic aberrations driving tumorigenesis, histological characterization, as well as identify new biomarkers for targeted drug therapy and treatment response in patients.
More than 90% of DDLPS tumors contain supernumery ring chromosomes and/or giant marker chromosomes derived from the 12q13~15 chromosome segment8,10–13,57. Within this amplicon lies the oncogene MDM258. However, because this is a discontinuous amplification with several hundred genes, it has been an ongoing challenge to distinguish which genes are vital to disease biology59. Two genes, for example, high mobility group AT- hook 2 (HMGA2), and fibroblast growth factor receptor substrate 2 (FRS2) which are amplified and highly expressed are hypothesized as essential regulators for the initiation and progression of liposarcomagenesis9,60. Furthermore, overexpression of HMGA2 and FRS2 have demonstrated important roles in cellular transformation and tumor establishment61.
HMGA2 has also demonstrated a direct adipogenic neoplastic involvement when spontaneous lipomas developed in mice expressing C-terminal truncated HMGA262. Another gene and more popularly studied, cyclin dependent kinase 4 (CDK4) amplification has been observed as frequently as 90% in DDLPS and is often used as a diagnostic marker alongside of MDM29–
11,13. Functionally, CDK4 is an important regulator of cell cycle progression in which it complexes with cyclin D to enable the G1-S transition by phosphorylating the retinoblastoma protein (Rb) and resulting in the activation of E2F target genes63–65. CDK4 is also a target for
13 therapy in an ongoing phase II clinical trial (ClinicalTrials.gov identifier: NCT01209598).
Genes located in the 12q13~15 chromosomal loci can also serve to differentiate between LPS subtypes. For example, differentially expressed genes that are involved in cell cycle control
(G1-S and G2-M transition) and growth factor receptors have been identified to molecularly differentiate between DDLPS and the more aggressive, higher grade PLS subtypes54. These examples demonstrate the importance for understanding the role of the genes located in this chromosomal amplification.
The progression from a WDLPS to the highly cellular DDLPS state acquires multiple genetic and epigenetic events that influence disease progression. In one recent study, Crago et al. (2012) identified genomic aberrations between 55 WDLPS and 52 DDLPS samples in an attempt to outline the molecular pathogenesis of liposarcoma. The most commonly observed alteration in disease progression was the loss of 11q23~24 chromosomal region in which multiple tumor suppressor genes reside. On the other hand, the loss of 19q13 was associated with tumor establishment, but did not correlate with tumor progression. To a lesser extent, other copy number variations were observed during progression including losses at
3p14~21, 3q29, 9p22~24, 10p15, 17q21 and gains at 17p11 and 20q119. Together, these amplified and lost regions result in the presumed expression of oncogenes and/or loss of tumor suppressor genes enhancing the malignant DDLPS phenotype.
Additionally, it has been observed that chromosomal aberrations associated with prognosis and survival correlated with amplification of 13q21 and loss of 19q139,66. Gain of
13q21 significantly correlated with poor overall survival (average of 78 months for patients without gain versus 35 months with 13q gain)66. Loss of the chromosome segment 19q13 resulted in a significantly shorter disease specific survival (27 months compared to >90
14 months for diploid retaining specimens) as well as negatively correlated to local recurrence rate9.
Recently, the amplification of chromosome 1p in DDLPS has been observed in multiple groups9,67. The JUN gene, located at 1p32~31 encodes the oncoprotein c-Jun. Jun is required for progression through the G1 phase of the cell cycle, as well as protects the cells from certain induced apoptosis68. Importantly, overexpression of c-Jun results in decreased levels of p53.
While these studies have unveiled new findings, deeper exploration of the associated genes within the affected amplicons must be investigated. Indeed, aberrations in DDLPS are highly prevalent; however, few are currently being targeted as a novel therapeutic modality.
15 TABLE 1.3 A SUMMARY OF GENOMIC ABERRATIONS AND THEIR FUNCTIONAL CONSEQUENCES IN DDLPS.
c Imbalance Functional Consequence First Author
Transcription factor, blocks adipocytic differentiation and increases 1p21~32 Amplification JUN (Mariani, 2007)69 aggressiveness
1q21~23 Amplification Gains in this region have been observed in tumor establishment (Szymanska.,1997)70
6q23~24 Amplification ASK1 Kinase, mediates differentiation and apoptosis signal transduction. (Chibon, 2004)71
This region contains key tumor suppressor genes that are lost in tumor progression 11q23~24 Loss (Crago, 2012)9 from WDLPS to DDLPS (ZBTB16,PP2R1B, E124) E3 ubiquitin ligase, targets the tumor suppressor protein p53 for MDM2 (Momand, 1992)72 proteasomal degradation Regulates cell cycle at the G /S phase, can result in chromosomal instability CDK4 1 (Segura-Sánchez, 2006)10 proliferation Amplification Transcription factor that regulates cell cycle (G /M) and role in 12q13~15 HMGA2 2 (Arlotta, 2000)62 adipogenesis and differentiation Role in FGR signaling; increases tumor formation, angiogenesis and tumor FRS2 (Zhang, 2013)61 progression Activator of apoptosis; interacts with pro-survival proteins BCL2 and Loss HRK (Bjorn Fritz, 2002)73 BCLXL Gains are associated with poor prognosis and worsened overall survival in DDLPS 13q21 Amplification (Schmidt, 2005)66 patients lost in tumor progression, regulates cell cycle at G /M and involved in 19q13 Loss CEBPa 2 (Y. V Wu et al., 2012)74 differentiation
16 Other deregulations: receptor tyrosine kinases A broad understanding of various deregulated pathways is crucial for understanding the pathobiology of DDLPS. With this in mind, we have strived to investigate common abnormalities observed in other cancers that are also specific to DDLPS. While there are many deregulations in DDLPS that have yet to be explored due to the rarity of this disease, it is critical to identify targetable deregulations as a key strategy for therapy. Receptor tyrosine kinases (RTKs) are aberrantly overexpressed and active in sarcomas, including DDLPS75,76.
RTKs are activated through ligand-dependent and ligand-independent mechanisms and, as a result, increased activation is able to activate and enhance multiple downstream tumorigenic pathways contributing to disease progression77. Of specific interest, we have observed that the
RTK c-Met is overexpressed and activated in DDLPS75. However, c-Met is but one of many targetable deregulations in this disease. Targeting c-Met in DDLPS demonstrated anti-DDLPS tumorigenic effects; however, it was not the focus of this dissertation. The results from this study can be found in Appendix D.
In summary, there are many potential molecular aberrations observed in DDLPS; however, amplification and expression of MDM2 is the most prominent deregulation and plays a critical role in tumorigenesis and progression of this disease.
17 SECTION 3. THE MDM2:P53 AXIS Together, MDM2 and p53 have established one of the most significant axes in all human cancers. In order to exemplify the importance of this axis, we must first understand their basic properties and functions.
P53: The guardian of the genome The p53 tumor suppressor protein is irrefutably the most heavily studied molecule in cancer biology since its identification in 197978,79. Often referred to as “the guardian of the genome” for its ability to prevent genome mutations, p53 activities are carefully controlled in normal cells. Consequently, p53 activity must be inhibited in order for cancer cells to flourish and survive. Therefore, it is no surprise that 50% of all human tumor cells harbor mutations or deletions in the TP53 gene, resulting in either partial or complete inactivation of p53 responses80. In response to cellular stress, p53 protein is stabilized and transformed into its active form functioning as a transcription factor to regulate the expression levels of downstream transcriptional targets genes which participate in various cellular processes78. In normal cells, p53 levels are negatively regulated by MDM2 which binds and targets p53 to the cytoplasm for degradation72. A complete review of p53 regulation and function in response to all of its various stimuli is beyond the depths of this study; therefore, we will focus on the specific functions in the p53 signaling network that relates to this body of work.
Stabilization and activation of p53 There are a variety of proteins that influence the stabilization and activation of p53 constitutively and in response to different stress stimuli. A few of these proteins, which include protein kinases, proteins influencing localization, as well as proteins that influence the p53-MDM2 interaction are listed in Table 1.4.
18 p53 stabilization occurs in response to DNA damage as well as by deregulated growth signals. One well characterized p53 stabilization process involves phosphorylation of p53.
The two checkpoint kinases, Chk1 and Chk2 are capable of phosphorylating p53 allowing p53 to deter degradation and be stabilized81. Chk2 phosphorylates p53 on Ser20 and, therefore, inhibits p53 degradation from MDM2. Double-stranded breaks in chromosomal DNA caused by ionizing radiation, transfer a signal to the ataxia telangiectasia mutated (ATM) kinase which passes along its signal to the ataxia telangiectasia and Rad3-related protein (ATR) kinase that is capable of directly phosphorylating p53 on Ser15 and deterring it from destruction82,83. Chemotherapeutic agents depend on the ability of ATR kinase to transfer its signal to casein kinase II which, in turn, phosphorylates p53 in another mechanism for p53 stabilization83. Lastly, in response to aberrant growth signals of the pRB-E2F cell cycle control deregulation, p53 activation can occur84.
Cells are constantly induced to environmental factors that cause DNA damage such as
X-rays, ultraviolet radiation, chemotherapeutic drugs, hypoxia, DNA synthesis inhibitors, etc. are just a few that cause a rapid increase in p53 protein levels and result in p53 stabilization and activation. p53 protects cells from such physiological stresses by activating DNA damage responses. In the event that a cell endures DNA damage, p53 will become activated by both increased protein expression levels and post-translational modifications that allow downstream activation of transcriptional targets. These cellular processes include DNA- repair, cell cycle arrest, senescence, inhibition of angiogenesis, and apoptosis, etc. (Figure
1.2)79,85–88. However, not all signals are equal, and in response to stress, p53 must mediate the decision of a cell to either repair the damage and resume proliferation or eliminate these irreparable cells through cell death mechanisms.
19
Table 1.4 Proteins influencing p53 stabilization and activation. Select proteins and their specific effect on p53. Inhibits or Protein Effect on p53 activates p53 Reference activity MDM2 Tags for ubiquitin degradation Inhibits Kubbutat (1997)89 Acetylates p53 during DNA Activates P300/CBP (Grossman, 2001)90 damage response ATM Phosphorylates p53 at ser15 Activates (Morgan, 1997)91 CHK2 Phosphorylates p53 at ser20 Activates (Kastan, 2004)92 De-ubiquitinates p53 and Activates HAUSP (Li, 2004)93 stabilizes p53 DAXX Enhances p53 ubiquitination Inhibits (Kruse, 2009)94
20
Figure 1.2 p53 influences multiple cellular stress pathways. Various physiological stresses increase and stabilize p53 levels. In turn, p53 activates a variety of different responses
Functions of p53: Downstream responses to activation
Structure The TP53 gene is localized on the short arm of chromosome 17 (17p13.1) and encodes for the 393 amino acid (a.a.) phosphoprotein95. The TP53 gene consists of 11 exons. The first exon, is non-coding; there are five highly conserved regions that extend from exons 2 to 8 that are essential for p53 function96. Like most transcription factors, p53 protein contains distinct domains which correspond to specific functions responsible for sequence-specific DNA binding and transcriptional activation. (1) the N-terminal transactivation domain (1-50 a.a.);
21 (2) the central DNA-binding domain (102-292 a.a.); (3) the C-terminal oligomerization domain (323-355 a.a.) containing a nuclear export signal; and the (4) regulatory domain (356-
393 a.a.) containing three nuclear localization signals and non-specific DNA binding domain that binds to damaged DNA97 (Figure 1.3).
FIGURE 1.3 SCHEMATIC REPRESENTATION OF P53 STRUCTURAL DOMAINS.
When stabilized, p53 forms a tetramer composed of two homodimers through the C- terminal oligomerization domain; tetramerization is required for p53 activity and growth suppression98. Within the C-terminal oligomerization domain is a basic region that is implicated in recognition of DNA damage, transcription, and apoptosis98. Furthermore, three nuclear localization signals which are required for nuclear localization of p5398.
Once stabilized, p53 is able to regulate gene transcription by binding in a sequence specific manner to the consensus sequence in p53 activated gene promoters on target genes99.
The p53 pathway Primarily, p53 initiates and controls a critical signal transduction network known as the p53 pathway100. As previously mentioned, upon activation p53 functions as a transcription factor to regulate various gene expressions that modulate p53 response pathways; these
22 responses include cell cycle arrest, differentiation, DNA repair, apoptosis, and senescence, etc. Importantly, when activated p53 must determine between life (cell cycle arrest) and death
(apoptosis). This decision is mediated by the ability of p53 to switch on or off certain genes through its transcriptional activity. There are over 4,000 putative p53-binding sites identified in the human genome; however, a short list of p53 target-genes tabland their associated p53 responses are listed in Table 1.5.
TABLE 1.5 P53 RESPONSES AND ASSOCIATED TARGET GENES.
p53 response p53 target genes Reference (Robles, Bemmels, Foraker, & APAF1 Harris, 2001)101 BAX (Toshiyuki & Reed, 1995)102 Apoptosis FAS (Owen-Schaub et al., 1995)103 NOXA (Oda, 2000)104 PUMA (Nakano & Vousden, 2001)105 CDKN1A (Broude et al., 2007)106 Cell cycle arrest GADD45 (Hollander et al., 1993)107 14-3-3-σ (Hermeking et al., 1997)108 (Dameron, Volpert, Tainsky, & Angiogenesis TSP1 Bouck, 1994)109 (X. Wu, Bayle, Olson, & Auto-regulation MDM2 Levine, 1993b)110
23 Cell cycle In normal cell biology, cell growth and proliferation is tightly regulated through a signal processing circuit known as the “cell cycle clock”. This “clock” decides the fate of a cell by integrating signals and choosing to either proliferate and continue through the cell cycle, or withdraw into a non-proliferative state. In order to implement this decision, cyclin- dependent kinases (CDKs) and their associated subunits (cyclin proteins) send signals to other molecules that enable and carry out the process of cell cycle111. CDK-cyclin complexes are well defined in each period of the cell cycle (Figure 1.4). Importantly, these complexes can be regulated by proteins known as CDK inhibitors. The best characterized CDK inhibitor is p21WIF/CP1 (p21)112. p21 is a widely acting, pan-CDK inhibitor that can arrest a cell in many different stages of the cell cycle and plays an important role in p53-dependent cell cycle arrest113.
Figure 1.4 Schematic of the cell cycle. The phases of the cell cycle and the major cyclin/CDK complexes associated with each phase.
24
p53 is involved in several different aspects of the cell cycle and transactivates or inhibits the expression of genes in order to mediate cell cycle progression114. A cell makes the decision to proliferate specifically during the G1 phase of the cell cycle, in which it is responsive to extracellular signals and mitogenic growth factors; whereas the following
92 phases are able to proceed on fixed schedule . Predominately, p53 influences the G1 phase of the cell cycle by inducing the expression of p21. In response to DNA damage, p21 mediates p53-dependent cell cycle arrest by binding to and inhibiting cyclin-Cdk complexes115. For example, in normal cellular context, pRB is bound to the E2F protein and upon nearing the end of the G1 phase, pRB is targeted for phosphorylation by Cyclin E-Cdk2 and disrupts the pRB-E2F interaction, thus, signaling the cell to continue to S phase116. However, in situations of cellular stress, p53 stimulates the transcription of p21 which inhibits the Cyclin E-Cdk2 function and holds the cell in the G1 phase until the stress is removed. p21 is active throughout the entire cell cycle, exemplifying the ability of p53 to control various points in cell cycle
113 progression . This is an important capability of p53 as inhibition of the G1 cyclin-Cdk complexes (cyclin E-Cdk2 and cyclin D-Cdk4/6) by p21 prevents a cell from copying un- repaired DNA damage in the S phase. However, if a cell has already progressed into S phase, p53 induced activation of p21 employs the DNA polymerase machinery at the replication fork resulting in inhibition of further replication of DNA template molecules. In addition, at the
G2/M checkpoint, p53 is able to induce a G2 arrest and inhibit entry into mitosis by decreasing cyclin B1 transcription and synthesis117. To do so, p53 induces the expression of other cell cycle inhibitors such as Gadd45a and 14-3-3σ. Gadd45a promotes cell cycle arrest by disassociation of the CDC2-cylin B1 complex formation118, while 14-3-3σ binds to and
25 108 inactivates Cdc25 and Cdc2 by sequestering them to the cytoplasm resulting in G2/M block .
In another method of control, p53 is able to stall the cell cycle by repressing Cdk4 and cyclin
E118. p53 has also been shown to play a role in the spindle checkpoint and avoid the re- replication of damaged DNA. Lastly, p53 is able to associate with centrosomes and prevent mitotic failure in centrosome duplication119. The ability for p53 to influence a cell at multiple checkpoints highlights the importance of its regulation and activity in cell biology.
26
Figure 1.5 p53 cell cycle regulation. Transcriptional activation and repression of target genes results in cell cycle arrest
27 Apoptosis
The cell death mechanism of apoptosis is a fundamental biological response that plays a critical role in normal cell development and homeostasis. p53’s role in apoptosis was first observed when imposed expression of p53 lead to features associated with apoptotic cell death such as DNA fragmentation caused by the cleavage of nuclear DNA in internucleosomal regions120. In response to environmental stresses, p53 employs several critical transcription targets that are essential for cell death and can trigger apoptosis through extrinsic or intrinsic signaling pathways, which, while differing on stimuli, converge at the level caspase activation.
The extrinsic pathway is initiated by the death receptor family of proteins, and activated through ligand-receptor binding on the cell surface. Cell surface receptors, such as
FAS, are transcriptionally up-regulated by p53 following DNA damage121. Following formation of the death-receptor-signaling-complex (DISC), a chain of events is triggered which results in caspase-8 and subsequent caspase activation. Caspsase-8 is the established initiator caspase for this specific pathway. p53 participates in this pathway by inducing the transcription of these death receptors and ligands.
In contrast, the intrinsic pathway is regulated by the pro-apoptotic and pro-survival
Bcl-2 family proteins. The Bcl-2 family is divided into three subclasses: (1) pro-survival proteins (BclXL and BCL-2); (2) pro-apoptotic proteins (BAX and BAK); and (3) BH3-only pro-apoptotic proteins (PUMA and NOXA)122.
In response to apoptotic stimuli, activation of the pro-apoptotic proteins bind to the pro-survival proteins, resulting in the activation of BAX/BAK. BAX and BAK oligomerize at the outer mitochondrial membrane which results in membrane permeabilization and promotes the release of cytochrome c and the resulting activation of effector caspases. The
28 pro-survival Bcl-2 family members inhibit this pathway by directly binding to BAX and
BAK122. However, this interaction is disrupted two other p53 transcriptional targets, PUMA and NOXA, which is up-regulated by p53 following cellular stress105,123. Together with
NOXA, these proteins mediate apoptosis liberating BAX and BAK. p53 is linked to the apoptotic pathway by directly inducing the transcription of several genes (PUMA, NOXA, and
BAX).
In summary, p53 is capable of inducing apoptosis through three mechanisms: (1) transcriptional regulation of apoptotic genes; (2) activating cytosolic BAX; and (3) inhibition and activation of apoptotic proteins at the mitochondria.
29
Figure 1.6 Schematic of extrinsic and intrinsic apoptotic pathways. The p53 targets are shown in red.
30 Loss of p53 activity in cancer p53 mediates critical cellular functions such as cell growth and proliferation; therefore, it is no surprise that p53 function is often disrupted in cancer. Various mechanisms contribute to the loss of p53 function in tumor cells, such as prevention of p53 activation, mutations in the p53 gene, as well as mutations in downstream target genes. The loss of p53 function due to mutations occurs in nearly 50% of all human cancers, wherein 90% of all mutations occur in the DNA-binding domain which manages the DNA-binding function of p53124,125. Most frequently (93% of mutations), the nature of p53 mutations is point-mutated alleles leading to amino acid substitutions. Because p53 function is a tetramer formation, the existence of a mutant allele could interfere with the function of the tetramer as a whole, giving the cell an advantage in compromising p53 function126. As a result, it is of particular interest as to how cancer cells are able to discard p53 function, as it appeared to be an anomaly to Knudson’s commonly accepted “two-hit hypothesis”127. Briefly, in this hypothesis Knudson described that the malignant benefits of silencing a tumor suppressor gene once it has lost both copies of the gene, and will only marginally benefit from the inactivation of one copy. Nevertheless, in the remaining 50% of cancers, such as DDLPS, the p53 retains its wild type (WT) genotype.
As a result, these tumors acquire additional defects in other ways that abrogate the stability and cellular responses of p53.
Regulation of p53 function p53 is regulated at both the mRNA and protein level through various mechanisms such as the control of p53 transcription and translation, modulation of p53s ability to function properly, and cellular localization98,128,129.
31 Cellular localization The localization of p53 between the cytoplasm and nucleus is a tightly regulated process; in order for p53 to function as a transcription factor it must be present in the nucleus.
As briefly mentioned, the nuclear localization signals and nuclear export signals control p53 location and can either import or export p53 to/from the nucleus, respectively129. On the other hand, in order for p53 to be degraded, it must be exported from the nucleus; however, the nuclear export of p53 also requires the ubiquitin ligase of MDM2129–132.
The MDM2:p53 negative feedback loop It is the p53 transcriptional target, MDM2 that is the best known regulator of p53 protein stability. MDM2 regulates the basal level and activity of p53 through direct protein- protein interaction126. Under normal conditions p53 levels are kept low, generally with a half- life of only 20 minutes; however, in response to physiological signals, p53 degradation is inhibited and p53 levels rapidly increased100 and p53 is free to transcriptionally regulate downstream target genes, one of which is MDM2. p53 binds to the P2 promoter of MDM2 and transcriptionally stimulates MDM2 expression133. In return, as MDM2 protein levels rise it binds to p53 and targets it for proteasomal degradation. Together, MDM2 and p53 regulate one another via an autoregulatory feedback loop100 (Figure 1.7).
32
Figure 1.7 Schema of the autoregulatory negative feedback loop between MDM2 and p53. p53 and MDM2 form an negative feedback loop. When activated, p53 is stabilized and binds to the transcriptional domain of MDM2. As MDM2 levels rise, it binds to p53 and tags it for degradation via the proteasome.
33 MDM2
Identification and biological function The MDM2 gene was originally identified in a spontaneously transformed BALB/c mouse cell line (3T3-DM). Later, it was found that the overexpression of the gene product was responsible for this transformation134,135. Initial experiments demonstrated that MDM2 binds to p53, thus, inhibiting p53-mediated transactivation136. Furthermore, the amplification of the MDM2 gene was observed in over 30% of sarcomas that retained WT p53137. These studies suggested that the overexpression of MDM2 was one mechanism in which p53 was regulated.
As previously described, MDM2 and p53 participate in a feedback loop regulating one another’s expression level; the physiological relevance of the p53-MDM2 relationship has been eloquently shown in mouse models. Two independent groups showed that the loss of
MDM2 lead to embryonic lethality in very early development; however, this can be rescued by simultaneous depletion of TP53138,139. Furthermore, Terzian and colleagues (2007) demonstrated that while MDM2 heterozygosity is adequate enough to maintain the level of p53 in adult mice under normal conditions, the exposure to even very low doses of ionizing radiation is lethal, and is therefore insufficient in response to DNA damage140. Taken together, these studies demonstrate the critical nature of MDM2 and p53 regulation and indicate that the primary function of MDM2 is to inhibit p53 activity in early development and when exposed to DNA damage.
Structure Structurally, MDM2 is a 491 a.a. protein that contains five domains for the basis of its oncogenic properties126 (Figure 1.8). Amino acids 1-118 comprise the p53 binding domain of the N-terminal portion and plays a critical role for inhibition of p53 transcriptional activity126.
34 The nuclear localization and nuclear export signal are located in the central region and are involved with the shuttling of MDM2 from the nucleus and cytoplasm141. The interaction between MDM2 and ribosomal proteins is regulated by the acidic and zinc finger domains142,143; while the C-terminus contains the RING finger domain responsible for the E3 ubiquitin ligase activity144,145.
FIGURE 1.8 STRUCTURE OF MDM2 PROTEIN.
Master regulator of p53 Although briefly mentioned, MDM2 is best known for its ability to regulate p53.
MDM2 inhibits p53 functions and negatively regulates the transcriptional activity and stability of p53 in at least three different ways. First, functioning as an E3 ubiquitin ligase, MDM2 binds to p53 and attaches ubiquitin groups to the carboxy-terminus of p53 resulting in ubiquitin-dependent degradation of p53 through the 26S proteasome100,126. Secondly, MDM2 and p53 also interact through the N-terminal transcription activation domain of p53, thus directly interrupting p53 transcriptional activity136,146. Thirdly, via its nuclear export signal sequence, MDM2 can promote p53 translocation through nuclear export as MDM2 removes p53 from its target genes and degrading the protein in the cytoplasm141,147,148
35 MDM2 functions as an E3 ubiquitin ligase The ubiquitin ligase activity of MDM2 is responsible for the attachment of an ubiquitin on the p53 protein. Ubiquitin, a 76 a.a. protein, was first identified in 1975 and can be covalently attached to proteins in a process called ubiquitinylation149. This is an inducible and reversible process in which ubiquitin tags a protein for degradation into its constituent amino acids by the 26S proteasome in an ATP dependent mechanism150,151. The sequential process for ubiquitinylation of p53 involves three enzymes. First, the ubiquitin activating enzyme (E1) binds to ubiquitin, this activated ubiquitin is then transferred to ubiquitin conjugating enzyme
(E2), and finally, MDM2 bound to p53 acts as an E3 ubiquitin ligase and causes ubiquitin to be directly transferred to p53130,131. Until recently, it was believed that p53 requires poly- ubiquitinylation of four or more ubiquitin subunits for effective degradation152; however, recent data reveals that MDM2 mediates monomeric ubiquitinylation on multiple lysine residues instead of polymeric ubiquitinylation153.
It is the E3 ubiquitin ligase that allows specificity in the ubiquitinylation process as there are over 100 different E3 ubiquitin ligases and only 20 different E2 ubiquitin ligases and only one E1 ubiquitin ligase154. Specifically, MDM2 is a Really Interesting New Gene (RING) domain type of E3 ubiquitin ligase154. The RING E3 ubiquitin ligases allow the direct transfer of ubiquitin from E2 to the targeted substrate; i.e. p53130,131,154. Furthermore, MDM2 undergoes self-ubiquitinylation, targeting itself for degradation154.
MDM2 regulation The MDM2 gene is transcriptionally controlled by two promoters: P1, which is located upstream of the first exon, and P2, which is located within the first intron. The basal expression of MDM2 transcription is regulated by the P1 promoter and is p53-independent. While there are few regulators of the P1 promoter, NF-kB and PTEN have been shown to positively and
36 negatively regulate MDM2, respectively. On the other hand, inducible expression of MDM2 is regulated by the P2 promoter which contains several response elements accountable for either inhibiting of activating MDM2 transcription155.
Other proteins that interact with MDM2 Several other proteins that interact with MDM2 either upstream or are specific downstream targets have been identified and characterized for their relevance in tumorigenesis. Generally, most upstream targets are kinases and phosphatases and can regulate MDM2 in one of three ways: (1) these effector proteins can regulate MDM2 transcriptionally by binding directly to the P1 or P2 promoter; (2) MDM2 can be regulated post-transcriptionally when these proteins bind to the MDM2 mRNA; (3) these proteins can post-translationally regulate MDM2 through ubiquitination, de-ubiquitination or phosphorylation155. On the other hand, because MDM2 functions as an E3 ubiquitin ligase, the majority of downstream proteins of MDM2 are targets for proteasomal degradation. A list of select proteins is summarized in Table 1.6.
37 TABLE 1.6. PROTEINS THAT INTERACT WITH MDM2 FUNCTION AND REGULATION.
Effect on MDM2/result of Protein Reference interaction
ARF Binds to MDM2 and inhibits (Tao & Levine, p19 MDM2:p53 interaction 1999)156 Phosphorylates MDM2 at (Maya et al., ATM ser395, inhibits its interaction 2001)157 Effector proteins with p53 of MDM2 (Terzian et al., MDMX Inhibits auto-polyubiquitination 2007)140 MDM2 blocks p300/CBP from (Clegg et P300/CBP interacting with p53 al.,2012)158 MDM2 activates transcriptional (Loughran & La E2F1/DP1 activity and antagonize E2F1 Thangue, 2000)159 apoptotic activity (Hsieh et al., RB RB binding to E2F1 is inhibited 1999)84 Affecter proteins MDM2 translocates p63 to of MDM2 (Su, Chakravarti, p63 cytoplasm and increases its & Flores, 2013)160 transcriptional activity MDM2 translocates p73 to (Zeng et al., p73 cytoplasm, P53 activity is 1999)161 decreased
MDM2 overexpression in cancer Based on its behavior in human malignancies, MDM2 is classified as an oncogene.
Over expression of the MDM2 gene has been found in many human malignancies either due to the amplification of the chromosomal region, or over expression of the protein without gene amplification58. The overall frequency of MDM2 amplification in human tumors is 7%; moreover, the highest frequency of MDM2 amplification (20%) is observed in STS58,125. Other mechanisms accounting for MDM2 overexpression include increased transcription and translation162, as well as a single nucleotide polymorphism (SNP), SNP309, which increases
38 the mRNA expression of MDM2163. Importantly, high expression of MDM2 is associated with tumor progression and worse prognosis164.
As previously mentioned, in DDLPS this high expression can be attributed to amplification of the 12q13~15 chromosomal segment. This apparent predilection to MDM2 amplification merits further investigation in these tumors, and, as a result, DDLPS not only provides a unique model for probing of the MDM2:p53 axis, but also an exemplary candidate for treatment with inhibitors of the MDM2:p53 complex.
39 SECTION 4. TARGETING THE MDM2:P53 WITH SMALL MOLECULE INHIBITORS AS A THERAPEUTIC STRATEGY FOR DDLPS
Reactivation of the p53 pathway with small molecule MDM2 inhibitors Reactivation of p53 as a therapeutic strategy is an attractive therapeutic target due to its powerful tumor suppressive functions that can be activated to eradicate tumors.
Approximately half of all human tumors retain WT p53; however, the p53 function can be abrogated by the overexpression of MDM2165. Frequently (~30%), this overexpression is due to amplification, such as seen in DDLPS (~90-100%)58. As a result, this therapeutic strategy is particularly advantageous for the treatment of DDLPS. MDM2 has proven itself to be a promising target for cancer therapies. There are two classes of MDM2 targeted compounds:
(1) inhibiting the E3 ubiquitin ligase activity of MDM2; (2) inhibits the protein-protein interactions between MDM2 and p53. For this study, we have focused on the latter.
Small molecule inhibitors of the MDM2-p53 protein-protein interaction Structural characterization of the p53-MDM2 interface revealed a deep hydrophobic cleft on the surface of MDM2 is occupied by the transactivation domain of p53166. Further analysis unveiled that deep within this “pocket” are four key hydrophobic residues crucial for the interaction between MDM2 and p53. Nevertheless, the ability to disrupt the protein- protein interactions has proven to be a difficult challenge. In an attempt to target this interaction, several classes of small molecule inhibitors have been described. Each inhibitor has a distinct chemical structure and includes: analogs of cis-imidazoline, spiro-oxindole, benzodiasepinedione, sulfanomide, and many others167. Currently, there are several small- molecule MDM2 inhibitors in various stages of clinical investigation (Table 1.7).
Specifically, in this study we investigated the effects of two previously reported small-
40 molecule inhibitors (Nutlin-3a and MI-219) and one novel MDM2 antagonist (SAR405838)
for the treatment of DDLPS.
Table 1.7 Small molecule inhibitors of the MDM2-p53 complex. An overview of
several MDM2 antagonists currently in clinical trials.
Mechanism Compound Study Nutlin-3 Phase I RG7112 Phase I/II MI-63 Phase I Inhibitor of the p53 binding pocket on SAR405838 Phase I MDM2 NU8231 Pre-clinical PXN-822 Pre-clinical RG7388 Phase I CGM097 Phase I Inhibitor of the p53 binding pocket to MK-8242 Phase I both MDM2 and MDMX RO-5963 Pre-clinical p53 inhibitor of MDM2 and MDMX binding pockets on the wild type and RITA Pre-clinical mutant p53 proteins
Initial peptide-based compounds were designed to target the MDM2 binding pocket
of MDM2; however, they demonstrated poor cell permeability and were not useful for cell
based studies. As a result, the pursuit was on to design a compound with enhanced
permeability, high-affinity, and resulted in marked cellular activity. Chemical library
41 screening, computational three-dimensional database screening, and structure-based design, all lead to the discovery and development of lead compounds168.
Mechanism of action While the core structure of MDM2 antagonists varies between each compound class, the overall mechanism of action remains the same. Targeting the protein-protein interaction of p53 and MDM2 with small molecule inhibitors has presented its challenges due to the large hydrophobic interface on MDM2169,170; however, if prevented from interacting, p53 is liberated from MDM2 degradation and capable of reactivating various tumor suppressor responses171–174. Four key hydrophobic residues are required for the interaction between
MDM2 and p53: Phe19, Leu22, Trp23 and Leu26. Disrupting the interaction of these amino acids inhibits the binding of MDM2 and p53. As a result, the ability to mimic this binding through use of selective, small-molecule MDM2 antagonists in cancers retaining WT p53 has been highly invested as an important therapeutic option171–174.
42 Figure 1.9 Small molecule inhibitors of MDM2 inhibit the MDM2-p53 complex.
MDM2 antagonists bind to the p53 binding pocket on the surface of MDM2 and inhibit the interaction of MDM2 and p53. As a result, p53 is liberated from MDM2 proteasomal degradation and is able to activate the p53 pathway.
43 Nutlins Nutlin-3a Discovered by Vassilev and colleagues, the Nutlins
were the first examples of potent and selective small-
molecule MDM2 antagonists. Experimental screenings of
synthetic compounds and extensive chemical
modifications ultimately resulted in the cis-imidazoline FIGURE 1.10 THE CHEMICAL 174 STRUCTURE NUTLIN-3A. core structure of the small-molecule MDM2 inhibitors .
Nutlin-1, Nutlin-2, and Nutlin-3 displaced p53 from MDM2 with IC50 values of 260nM,
175 140nM, and 90nM, respectively . However; it was, Nutlin-3a, an active enantiomer isolated from Nutlin-3 that provided the most promising therapeutic potential. Vassilev demonstrated that Nutlin-3a successfully activated the p53 pathway in cells retaining WT p53 in basic cell culture based assays, as well as provided the first in vivo proof-of-concept that the pharmacological inhibition of MDM2 and reactivation of WT p53 inhibited tumor-growth.
This early therapeutic pioneer opened the doors for increased selectivity and potency for
MDM2 antagonists.
Recently, the Nutlin-derivative RG7112 exemplified the use of MDM2 inhibition and reactivation of p53 in WDLPS and DDLPS52 and is currently in the most advanced stages of clinical development (Phase II) for a small-molecule MDM2 inhibitor 174. The importance of this study lies in the fact that this traditionally chemotherapeutic-resistant disease responded to treatment, as well as provide insight into biomarkers for therapeutic response174.
The design of small molecule inhibitors using a computational pharmacophore search strategy was performed by Galatin and Abraham176. NSC279827 was the one compound
44 found to inhibit MDM2-p53 binding; however, it demonstrated a weak affinity with only 20% increase in p53 transcriptional activity following 100μM.
MI-219 Nevertheless, these two screening approaches had MI-219 limitations based on the lead-core design. As a result, a new
class of compounds were developed using a structure-based de
novo strategy177. This design is based on the crystal structure of
the p53:MDM2 complex and heavily weighted on one specific
FIGURE 1.11 THE residue (Trp23) of p53 that is located in the hydrophobic pocket CHEMICAL STRUCTURE MI-219. of MDM2177. To inhibit this interaction, the oxindole ring system was utilized to mimic this interaction177. This resulted in the design of a new class of inhibitors based on the spiro(oxindole-3,3’-pyrrolidine) core structure. Spiro-oxindoles are able to bind to MDM2 with high affinity and activate the p53 pathway in cell lines with WT p53177,178. Following structure-based optimization, highly potent inhibitors were developed, such as MI-219179.
MI-219 (MDM2 inhibitor-219) is an optimized analog of MDM2-inihibitor-63 (MI-
63) and has been the leading compound in this class of inhibitors177. MI-219 mimics the four key amino acids on the MDM2 binding pocket. MI-219 showed initial pharmacological potential due to the oral bioavailability, specificity to MDM2, and disruption of the
MDM2:p53 complex. Furthermore, MI-219 displayed a high binding affinity and specificity for MDM2, disrupting MDM2:p53 binding and inducing stabilization of p53 in cells with WT p53177,178.
45
Anti-tumorigenic responses of MDM2 inhibition Small molecule inhibitors have certain desirable properties that are crucial for its effectiveness, these include: (1) high binding affinity and specificity to MDM2; (2) potent cellular activity in cancer cells with wild-type p53; and (3) a desirable pharmacokinetic (PK) profile167. Nutlin-3a and MI-219 provided valuable preliminary evaluations for MDM2 inhibitors. In studies comparing the therapeutic effects of these drugs, both Nutlin-3a and MI-
219 demonstrated p53 induced anti-tumorigenic results173. It was observed that in normal cells that these inhibitors induced cell cycle arrest, but not death; meaning that these inhibitors are not toxic to normal cells; therefore, highlighting their potential therapeutic use173. p53 pathway activation was observed by the G1 and G2 cell cycle arrest, as well as up regulation of the p53 transcriptional target, p21, a known mediator in cell cycle arrest. Both drugs induced p53-dependent cell cycle arrest and apoptosis; however, MI-219 demonstrated a more desirable pharmacokinetic profile than Nutlin-3a167,174,177,178. Together, these two inhibitors were important to the development for the rational and evidence of small-molecule MDM2 inhibitors and are often used as a standard in comparing the efficacy of new drugs.
46 SAR405838: a novel small molecule MDM2 inhibitor
Currently, optimization of spiro-oxindole SAR405838 compounds is being heavily pursued; consequently, many
of these inhibitors have only been disclosed in patents and
little is known about their cellular activity, pharmacological
properties and mechanism of action. Towards that end, a
novel spiro-oxindole small-molecule inhibitor of MDM2
FIGURE 1.12 THE CHEMICAL has been developed by Sanofi-Aventis: SAR405838. STRUCTURE SAR405838. SAR405838, an analog of MI-219, has shown superior pre- clinical results. Initial experiments demonstrated p53 stabilization and activation in cancer cell lines with WT p53. Osteosarcoma-bearing xenograft models showed anti-tumor dose- dependent responses following SAR405838 treatment, as well as demonstrated a promising pharmacokinetic profile as the plasma concentrations of SAR405838 were similar between a single day and 11 day repetitive oral administration.
Due to the encouraging initial results from SAR405838, we investigated the initial pre- clinical use of this drug in DDLPS, of whom it would be the most biologically relevant model.
47 HYPOTHESIS AND SPECIFIC AIMS
HYPOTHESIS
MDM2 targeted therapy is an effective anti-DDLPS therapeutic strategy.
SPECIFIC AIMS
Aim 1: To develop and characterize experimental models for DDLPS
1.1 Develop currently lacking human DDLPS cell lines and xenograft models
1.2 Determine MDM2 and p53 status in these cell lines
1.3 Unravel the molecular characteristics of these cell lines
Aim 2: To investigate MDM2 as a therapeutic target for the treatment of DDLPS
2.1 Determine the efficacy of the small molecule MDM2 inhibitor SAR405838 in
vitro
2.2 Evaluate the efficacy of SAR405838 in vivo
Aim 3: To define the molecular changes in DDLPS in response to MDM2 inhibition
3.1 Identify gene expression changes in response to SAR405838 treatment in vitro
3.2 Identify gene expression changes in response to SAR405838 in vivo
48 SIGNIFICANCE
MDM2 amplification in the presence of WT TP53 is a molecular hallmark of DDLPS.
In this body of work, we inhibited this oncogenic pathway as a node of therapeutic vulnerability. New generation small molecule MDM2 inhibitors, such as SAR405838 are highly effective in disinhibiting the p53 pathway in DDLPS cells resulting in significant anti- tumorigenic effects. These preclinical data strongly support the development of clinical trials for patients with this aggressive cancer and perhaps other cancers where MDM2 is pathogenic.
MDM2 amplification has commonly been observed in other malignancies that retain wild- type p53. Our findings suggest a basis for the utility of SAR405838 in these clinical contexts.
From the biological standpoint, MDM2-p53 pathway is of critical importance in both normal physiology and disease. The preclinical tools we developed here can be used to further unravel the functional consequences of inhibiting this pathway.
49 CHAPTER 2. MATERIALS AND METHODS
Isolation and establishment of human DDLPS cell lines from solid tumor Human DDLPS cell strains and lines were originally isolated and established in our lab were obtained from patient consented tumors. Following surgical resection, a representative sample was sectioned and placed into our tissue bank, as well as for pathological review. The rest of the DDLPS tumor tissue was aseptically placed into 50mL conical vials containing sterile PBS and brought over to our facility on dry ice immediately following excision. Tumor samples weighing 20-30 grams were removed from PBS and transferred to a glass 10cm petri dish. 10-20mL of DMEM media is added to the tumor, or, if it is not a human sample, RPMI1640 media is used. The tumor is cut into 1mm pieces with a sterile scalpel. The minced tumor pieces are then transferred in DMEM to a 50mL Falcon spinner flask and the petri dish is rinsed with DMEM to collect any remaining tissue. DMEM is added to the Falcon flask to reach a total volume of 25mL. 2.5mL of 3% collagenase Type
1 (Sigma), 5.0mL of 0.02% DNAse I Type II (Sigma), and 1.0mL of 1.5mg/mL hyaluronidase
(Sigma) is added to the vial. The spinner flask is placed on a Bell-Stir at speed of 8 in 37°C incubator for 2 hours to digest the tumor. The tumor samples are then strained through a sterile wire mesh screen (Fisher) into a new 50mL tube and centrifuged at 1500 rpm at room temperature for 5 minutes. The media is aspirated from the tissue and PBS is added to for a total volume of 40mL to wash the cell pellet. The cells are then centrifuged again at 1500rpm for 10 minutes at room temperature. The PBS is aspirated and the pellet is resuspended in
20mL of PBS. Histopaque tubes were set up for every 1mL of cell pellet. 10mL of 100%
Histopaque (Sigma) is added to the 50mL conical vial and carefully overlayed with 15mL of
75% histopaque in PBS The cell suspension is carefully added to the top of the histopaque so that the solutions are not mixed. The tubes were centrifuged at 4°C for 30 minutes at 1800
50 rpm. The first interface, which contains the tumor cells, is carefully pipetted out of the tube without disturbing the other layers. 50mL of PBS is added to the extracted cells and centrifuged at 1800rpm for 5 minutes at room temperatures. The PBS is aspirated after centrifugation and cells are in DMEM with high glucose, 10% FBS and 1% penicillin/streptomycin in a cell flask. Cells are incubated at 37°C.
Cell Culture SW872 was obtained from the American Type Culture Collection. LISA2 and LS2B were kindly given to us from Dr. Kiki Broccoli (The Curtis and Elizabeth Anderson Cancer
Institute, Savannah, GA, USA). Dr. Jonathan Fletcher (Brigham and Women’s Hospital,
Boston, Ma) generously provided us with LPS141 cells. All cells were cultured in Dulbecco's modified Eagle's medium (DMEM) and supplemented with 10% fetal bovine serum (FBS),
100 U/mL penicillin, and 100 U/mL streptomycin. Human mesenchymal stem cells derived from adipose tissue (hMSC-AT) were obtained from Zen-Bio and PromoCell and cultured in mesenchymal stem cell medium supplemented with manufacturer’s supplemental mix
(PromoCell). These cells were cultured under normal conditions in a humidified chamber delivering 5% CO2 at 37°C.
Short Tandem Repeat (STR) DNA fingerprinting via sequencing of short tandem repeats (STR) was conducted for all cell lines prior to use in studies in order to validate the cell line to the original tumor sample from which they were derived. DNA from each cell lines was obtained using a Qiagen Blood and Cell Culture DNA kit (Qiagen, Valencia, CA, USA) per manufacturer’s instructions. In order to amplify DNA for fingerprinting, AmpFISTR Identifier PCR Amplification kit
(Applied Biosystems) was used. This kit results in amplification of the amelogenin gender
51 determining marker and 15 tetranucleotide repeat loci. Cell strain/line fingerprints were compared with the original tumor or the earliest available passage of cells. Known cell line fingerprints were compared to those published by the American Type Culture Collection
(ATCC).
Fluorescence in situ Hybridization (FISH) FISH analysis was used to observe the genomic amplification of 12q13-15 in our panel of cells. Cultured cells were harvested, washed 2x in PBS and resuspended in 200μL of 3:1
Acetic Acid:Methanol. 200μL of cell suspension was then carefully placed onto slides and allowed to “age” over-night at RT. The following day, the cellular DNA is denatured by placing the slides in a Coplin jar warmed to 75°C containing the denaturing solution of 70% formamide/2XSSC for 5 minutes. The slides are then dehydrated by placing them in sequential solutions of 70%, 85% and 100% ethanol at room temperature for 2 minutes each. Each slide requires 5 µL of probe mix for the 12q15 chromosomal region (RP11-185H13, RP11-450G15,
RP11-816C9, RP11-630N19, RP11-717F7, RP11-1104N20, and RP11-426B12; probe was kindly given to us from Dr. Dolores Lopez-Terrada, Texas Children’s Hospital, Houston,
Texas) 5μL of probe mix containing centromeric chromosome-12 fluorescence-labeled probes (spectrum green; Abbott Molecular, DesPlaines, IL, USA) is applied to each slide and carefully covered with a cover slip. Cover-slipped slides are placed into HyBrite incubator to hybridize. The hybridization cycle is preset to melt for 5 minutes at 80°C, and then hybridized for 16 hours at 37°C. On the third day, the slides are quickly dipped into a wash solution containing 2XSSC/0.3% NP-40 to remove the coverslip. Excess probe is washed off by placing slides into a Coplin jar for 2 minutes containing 0.4XSSC/0.3% NP-40 at 70° C, and a second wash of 2xSSC/0.1% NP-40 for 1 minute at room temperature. Slides are then
52 allowed to air dry in the dark for 20-30 minutes. Once dry, mounting media containing DAPI
(ProLong Gold Antifade Reagent with DAPI; Life Technologies, Grand Island, NY, USA) is applied and slides are cover-slipped. MDM2 amplification was observed and considered at a ratio of MDM2/CEP12 ≥ 2.1 per 100 cells. gDNA extraction gDNA was extracted from DDLPS cell lines using the QIAamp DAN Mini Kit
(Qiagen) as per manufacturer’s instructions. Briefly, the cell pellet was resuspend in PBS to a final volume of 200 µl. 20 µl of proteinase K and 200 µl of lysis buffer were added to the sample, mixed by pulse-vortexing for 15 seconds and incubated at 56°C for 10 min. 200 µl ethanol (100%) was added to the sample, and mixed again by pulse-vortexing for 15 seconds.
The mixture was added to the column and centrifuged at 6000 x g for 1 min. The filtrate was discarded and 500 µl of buffer AW1 was added (removing any non-specific binding of inhibitors to the spin column membrane) and centrifuged at 6000 for 1 min. The filtrate was discarded and 500 µl Buffer AW2 was added (in order to wash away any salts that are present) and centrifuged at full speed (20,000 x g) for 3 min. The column was placed in the centrifuge at full speed for 1 min to eliminate the chance of possible Buffer AW2 carryover. The column was placed in 1.5 ml tube, and 200 µl of distilled water was added, incubated at room temperature (15–25°C) for 1 min, and then centrifuged at 6000 x g for 1 min.
RNA extraction and quantitative real-time PCR Cells were cultured at 75% confluence. mRNA was extracted from cell pellets using
Qiagen RNeasy Mini Kit (Qiagen, Valencia, CA, USA) and quantified using a NanoDrop
2000 instrument. Bio-Rad's iScript cDNA Synthesis Kit (Bio-Rad, Hercules, CA, USA) was used to perform reverse transcription and Light Cycler 480 SYBR Green 1 Master Mix
53 (Roche) was used for quantitative detection of transcripts. Second derivative max values were calculated by Roche’s Light Cycler software. Expression levels were normalized to β-2- microglobulin, and calculated using the 2ΔΔCT method; all experiments were performed in triplicate. Primers are detailed in Table 2.1.
54 Table 2.1 qRT-PCR primer sequences. Sequence of primers used in qRT-PCR experiments.
Primer Strand Sequence sense AAAGGGCCAGGTTAAATGGT MDM2 antisense GTGTGCCCCAGAACAAAGAT sense GGAAGACCATGTGGACCTGT CDKN1A antisense GGCGTTTGGAGTGGTAGAAA sense TCAAAGCAGGTTTTTGTGGA Nup107 antisense CGAGGAACAAGGAATGGAAA sense GAAGATGGGCACACTCATCA YEATS4 antisense TGGATTGCCATAGCTTTCATGT sense ATCACTGTCGGGGTGTACGA GADD45a antisense ATCTCTGTCGTCGTCCTCGT sense AGAGCCCAGGAGACTTCACA BUB1 antisense CAAAGTCGCCTGGGTACACT sense ACGCTACGAGGACCTGCTAA GDF15 antisense AGAGATACGCAGGTGCAGGT sense GTTCCGAGAGCTGAATGAGG P53 antisense TCTGAGTAGGCCCTTCTGT sense GTGGCAGCTGACATGTTTTC BAX antisense GGAGGAAGTCCAATGTCCAG sense GACGACCTCAACGCACAGTA PUMA antisense CACCTAATTGGGCTCCATCT sense GAATTCACCCCCACTGAAAA B2M antisense CCTCCATGATGCTGCTTACA sense GTGCAGGTTCGAGTCAGGAT AVIL antisense TAACTTCCACTGCTTTGGTG sense TACCCCTCAGCCACTGTTCT CYP27B1 Antisense TGTCCCACACGAGAATTTCC sense TGCGAGATCAACTTGGACAG RAB31P antisense CTGCTGTTGCCTGCTTGATA
55 Western blotting Western blot (WB) analyses were used to evaluate the protein expression. Cells were cultured as previously described prior to collection. Cells were then washed with ice-cold phosphate buffered saline (PBS), and resuspended in whole cell lysis buffer (0.5M EDTA,
200mM PMSF, 200mM Na3VO4) that contained phosphatase inhibitors (Sigma-Aldrich), then placed on ice for 30 minutes and centrifuged for 15 min and supernatant was collected.
Lysates were stored at -20°C. A Bradford assay was performed (Bio-Rad, Hercules, CA,
USA) using bovine serum albumin (BSA) in order to establish a standard curve to determine the protein concentration of each sample. 25-50µg of protein and 1x loading buffer in 10%
β-mercaptoethanol was prepared and loaded into a bis-acrylamide gel. The protein marker to determine the protein weight was added to the first well of the gel. The samples were run at
100 volts for 1.5 hours to separate the proteins. Protein was transferred onto a Nitrocellulose membrane and ran at 100 volts for 1 hour. All western blots were blocked with 5% milk in
PBS-T for 1 hour. Primary antibodies were diluted at 1:1000 in fresh 5% milk/PBS-T and applied and incubated overnight at 4°C. The following day, blots were washed 3 times with
PBS-T for 10 minutes and a secondary antibody, (goat anti-rabbit IgG and goat anti-mouse
IgG conjugated to horseradish peroxidase (Santa Cruz Biotechnology, Inc)) diluted 1:3000 in 5% milk/PBS-T was applied for 1 hour at room temperature. Blots were developed using
Western Lightning ECL kit (Perkin-Elmer, Inc.)
Antibodies for western blot analyses The following commercially available antibodies were used: MDM2, and Nup107, were purchased from Abcam (Cambridge, MA); Cleaved caspase 3, and Aurora A were purchased from Cell Signaling (Danvers, MA); p53, p21, GADD45a and β-actin antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Goat anti-rabbit IgG and
56 goat anti-mouse IgG antibodies conjugated to horseradish peroxidase (Santa Cruz
Biotechnology) were used as secondary antibodies for western blot experiments.
P53 Mutational Analysis p53 mutational analysis was conducted on all cell strains/lines as previously described by our lab180. In brief, QUIamp DNA Mini Kit (Qiagen, Valencia, CA, USA) was used to extract genomic DNA as per manufacturer’s instructions. The integrity and concentration of each sample was evaluated using a NanoDrop spectrophotometer and primers recognizing the intronic sequences of exons five through nine of p53 were evaluated. PCR amplification was performed on genomic DNA for all p53 and the sequence analysis from the PCR product was performed with Sequence Scanner (version 1.0, Applied Biosystems).
Hematoxylin and eosin (H&E) staining As previously explained, tissue samples were fixed in 10% phosphate-buffered formalin and embedded in paraffin. These formalin-fixed, paraffin-embedded sections were cut and mounted onto slides with a poly-l-lysine coat (Sigma Aldrich, St. Louis, MO. USA) and then allowed to dry at 37°C. Slides were then heated on a heating block (55-60°C) for 15-
20 minutes. Slides were deparaffanized and hydrated through the xylene and alcohol to water sequences and washed with running water 3-4 times. Slides were stained with Harris hematoxylin (Fisher Scientific, #SH26-500D) for 3 minutes, followed by 3 washes with water.
Slides were then quickly dipped in acid alcohol (700mL 95% EtOH + 250mL dH2O + 10mL hydrochloric acid) followed by 3 washes with water. Samples were then placed in PBS for 1 minute and again washed 3 times with water. Slides were dipped quickly 5-6 times in Eosin-
Y solution (Sigma Aldrich, # 046K4365) to stain the cytoplasm. The slides were decolorized and dehydrated through alcohols and xylenes: 4 changes in 95% EtOH (10 dips) 2 changes
57 100% EtOH (10 dips) and 1 change in xylene (10 dips). Following the last xylene dip, slides were coverslipped and mounted using Permount (Fischer) and observed under the microscope.
Immunohistochemistry The protocol used to perform immunohistochemistry on paraffin sections has been obtained and modified from the protocol established by Donna Reynolds (Chief, Histology
Laboratory, Department of Cancer Biology, The University of Texas as M.D. Anderson
Cancer Center, Houston, TX, USA) and varies slightly amongst each protein of interest. The following protocol is the standard modified protocol performed. Tissue samples were fixed in
10% phosphate-buffered formalin and embedded in paraffin. These formalin-fixed, paraffin- embedded sections were cut, mounted onto slides with a poly-l-lysine coat (Sigma Aldrich,
St. Louis, MO. USA) and then allowed to dry at 37°C. Prior to deparaffinization, sections were then placed on a slide warmer for 20 mins (55-60°C) to soften the paraffin and adhere the tissue to the slide. Slides were deparaffinized and hydrated to PBS using the following series: Xylene for 4 minutes, Xylene for 3 minutes, 2 100% EtOH washes for 2 minutes each,
2 washes 95% EtOH for 1 minute each, 80% EtOH for 1 minute and 2 washes of 1X PBS for
2 minutes each. Antigen retrieval was then performed to restore the immunoreactivity of the protein of interest after fixation in formalin. The optimal pH of the recovery buffer is a critical element of the recovery step and was established specifically to each protein. DIVA Decloaker
(BioCare Medical) pH 6.0 was diluted and used as the antigen retrieval. The retrieval was performed using a steamer (Vegetable Steamer, Black and Decker) for 40 minutes. After heating, the slides were cooled for 15 minutes using partial exchanges (4-8) of retrieval solution with PBS until reaching room temperature. Slides were then removed from PBS and a Pap Pen was used to draw a small circle around the sample. Slides were taped to a slide
58 holder and placed in a humidity chamber. 3% hydrogen peroxide/methanol solution to the slides was added for 12 minutes to block the endogenous peroxidase activity in the tissue and prevent it from interfering with the activity of the HRP conjugated to the secondary antibody that produces the color change when DAB is added. Slides were then washed 3 times in PBS
(3 minutes each) and protein block (5:1 normal horse serum:normal goat serum was dropped onto tissue for 15-20 minutes. Primary antibody was diluted at 1:100 in the protein block and
50µL of each primary was added to each slide and allowed to incubate overnight at 4°C. The following day, slides were washed 3 times with PBS, and 4+Biotinylated Goat Anti-Rabbit secondary antibody was added for 10 minutes and then washed again 3 times in PBS for 3 minutes. HRP conjugated 4+ Streptavidin was added and incubated for 10 minutes, followed by 3 PBS washes for 3 minutes. Slides were rinsed once with PBS/Brij (pH 7.6) and then incubated with DAB for 5-10 minutes while monitoring the reaction under the microscope for optimal color. Samples were then washed for dH2O 3 times for 3 minutes and once with dH2O/Brij. Slides were counterstained with Gill’s #3 hematoxylin for 5-15 seconds, followed by a drop of PBS for one minute to darken the hematoxylin. Slides were then washed and dried and mounted using Universal Mount to preserve the tissue.
Array CGH Genome wide profiling for chromosomal aberrations was investigated using
Affymetrix Human SNP Assay 6.0. Over 946,000 probes were used to detect the copy number variation in our cell panel. The gDNA was extracted as previously described and the array was performed as per manufacturer’s instructions. In brief, 500ng of total gDNA with a concentration of 50ng/μL was used per sample. DNA was digested with Nsp I and Sty I restriction enzymes, ligated to adaptors and a primer that recognizes the adaptor sequence
59 amplified the DNA fragments. PCR was used to amplify fragments in the 200 to 1100bp size and the PCR products for each restriction enzyme were combined and purified using activated beads. The amplified DNA was then fragmented, labeled and hybridized to the array. This array was then scanned by the GeneChip Scanner 3000 7G system.
Gene expression array Total RNA was isolated from DDLPS cell lines and xenograft tumor tissue as previously described in RNA isolation and purification. The total RNA was converted into cRNA (Illumina TotalPrep Amplification Kit, Ambion). Samples were hybridized to the HT-
12 Version 4.0 Illumina Chip. Gene expression analysis was performed on samples.
Ingenuity Pathway Analysis Ingenuity Pathway Analysis (IPA, Ingenuity Systems, www.ingenuity.com) was utilized to analyze the gene expression array results and reveal which pathways were affected by SAR405838 on the mRNA of DDLPS cells and DDLPS xenograft models. Only the significantly (p < 0.05) up- and down regulated expressed genes compared to controls (Fold change 2.0) were selected for consideration and imported into the software for analysis. These genes were associated to the canonical pathways in which they enriched.
Reagents for experimental procedures Nutlin-3a, an MDM2 small-molecule inhibitor, was purchased from Caymen
Biochemicals. Nutlin-3a was suspended in DMSO for a stock solution of 10mM and stored at
-20°C. MI-219, and SAR405838 were provided by Sanofi-Aventis (Paris, France). For in vitro studies, MI-219, and SAR405838 were suspended in DMSO for a stock solution of 10mM and stored at -20°C. For in vivo studies, the following calculations and protocol were performed to prepare SAR405838:
60 Preparation for 2mL solution at 20mg/mL
Density: TPGS=0.947-0.951 PEG=1.127
1. Weigh 40mg of drugs substance
2. (1960μL) 2208.92mg PEG200 then vortex 30seconds + ultrasonication 1 minute.
(vortex and sonicate to repeat if needed)
3. Add (40μL) 37.95mg TPGS (after melting at about 50°C) then vortex 30 seconds
+ ultra-sonication 1 min (vortex and sonicate to repeat if needed). If TPGS
becomes solid during mixing, heat the mixture.
Cell proliferation analysis Cell proliferation was measured via MTS assay using the CellTiter96 Aqueous Non-
Radioactive Cell Proliferation Assay Kit (Promega, Madison, WI, USA) per manufacturer’s instructions. Cells were seeded in 96-well plates at 1,500 cells per well in 100μL of media and allowed to adhere overnight. Media was then replaced in each well and cells were exposed to inhibitors as determined per experiment. At the end of each finalized incubation time point,
20μL of MTS (Sigma Aldrich) was added to each well and incubated for 2-4 hours at 37°C.
The absorbance of each well was read at 490 nm (Beckman Coulter, Miami, FL, USA). The cell proliferation was calculated by subtracting the absorbance level from the mock or control wells containing no drugs or additives from the treated cell wells. The effective concentration for which growth was inhibited by 50% by a certain drug (EC50) was determined using BioStat
Speed software. All experiments were performed at least three times to obtain final values for statistics.
61 Cell cycle analysis Cells were exposed to variable conditions as determined per experiment for 48 hours.
Following standard protocol, attached and floating cells in the media were collected and centrifuged at 1.2x103 RPM for 3 min. The media was aspirated and cells were washed twice with ice-cold PBS and centrifuged for 3 min at 1.5x103 RPM. The cell pellet was resuspended and fixed in 70% EtOH at 4°C overnight. The following day, cells were centrifuged at 2500
RPM for 5 minutes, and excess ethanol was aspirated. Cell pellets were resuspended in 400-
500μL of Propidium Iodide staining containing 10ug/mL RNase for 2 h in the dark prior to flow cytometry analysis of DNA content. The results were read using Flow Cytometry
(channel FL-2A, and histograms of cell numbers vs. linear integrated red fluorescence were recorded for a minimum of 10,000 nuclei at flow rates no greater than 30 to 50 events per second.
Annexin V/ FITC FACs analysis for detection of apoptosis Apoptosis was measured using the Apoptosis Detection kit 1 (BD Pharmagen, San
Diego, CA, USA) per manufacturer’s instructions. Following standard protocol, cells were exposed to variable time and dose conditions as indicated experiment for 96 hours. Floating and attached cells were harvested and centrifuged at 2x103 RPM for 2 minutes. Cell pellets were resuspended in 100μL of 10% binding buffer and fixed with 5 μL Annexin V/FITC (BD
Pharmagen, San Diego, CA, USA) and 5 μL propidium iodide (Sigma Aldrich Co.) per sample. FACScan (Becton Dickinson) was used to analyze samples by emitting excitation light at 488 nm. CellQuest software (Becton Dickinson) was used to analyze data. siRNA knockdown Small-interfering RNA (siRNA) targeting MDM2 (SMARTpool: ON-TARGETplus
MDM2 siRNA, #L-003279-00-005), p53 (SMARTpool: ON-TARGETplus p53 siRNA, # L-
62 003329-00-0005), and non-targeting control (ON-TARGETplus Non-targeting siRNA #1,
#D-001810-01-20) were purchased from Dharmacon. siRNA was resuspended in 5X siRNA
Buffer (Dharmacon) to a 20 nmol final concentration and kept at -80°C. In brief, 2.5x105 cells per well were seeded in 2mL of media in a 6-well plate and allowed to adhere overnight. The next day, the media was aspirated and siRNA was transfected into DDLPS cells using
Lipofectamine 2000 (Invitrogen, Grand Island, NY, USA) according to the manufacturer’s instructions. In brief, (described for a 24-well format and scaled per experiment as manufacturer describes) each transfection sample was prepared as followed: 20pmol siRNA oligomer was diluted in 50µL of Opti-MEM Reduced Serum Medium without serum (Opti)
(Life Technologies) and mixed gently. 1µL of Lipofectamine2000 was diluted in 50µL of
Opti and incubated for 5 minutes at room temperature. Following this incubation, the diluted
Lipofectamine/2000 and siRNA/Opti suspensions were mixed together and incubated for 20 minutes at room temperature. The final mixture was then added carefully to each well containing the cells and medium and incubated at 37°C for knockdown. siRNA knockdown was confirmed via western blot analysis.
Pharmacokinetics Groups of 6-week-old female BALB/c nude mice were given 100, or 200 mg/kg body weight of SAR405838 orally into Lipo863 tumor bearing xenografts. Blood samples were taken at 6, 24, 48 and 72 hours post treatment. Just after sacrifice of the animals with O2/CO2, the blood is taken by intraocular puncture and transferred into heparinazed tubes. The plasma is collected (BD Pharmagen, San Diego, CA, USA) after 3 minute centrifugation at 7600g at
4°C. The tubes with plasmas are frozen at -20°C. For each tumor a piece of 50mg is collected for the RNA measurements, this piece is sliced thinly and put in an eppendorf tube containing
63 1.5ml RNA later (Qiagen). After 24h at 4°C, these samples are stored at -20°C. The rest of the tumor is cut into 100mg pieces and placed into a Precelly’s tube (Precelly’s tubes ref:
03961-1-007) where they are immediately frozen at -80°C (on dry ice) and then stored at -
80°C. 1 tube was reserved for PK studies. Samples were shipped to Sanofi Aventis at -80°C for analysis. Plasma (100 µL) was submitted to an organic extraction with 300 µL ACN supplemented with 2H5-SAR405838 as internal standard.
DDLPS xenograft models Female Balb/c nude SCID mice were purchased from Charles River Laboratories
International, Inc. and housed in pathogen free conditions as per The University of Texas MD
Anderson Cancer Center and Institutional Animal Care and Use Committee (IACUC) regulations. Animals received humane care as per the Animal Welfare Act and the NIH
“Guide for the Care and Use of Laboratory Animals”. For tumor growth studies, 2-3.5x106 cells were injected subcutaneously and tumor size was monitored 2-3 times a week until the study was terminated. 5-10 mice per group were used per study. Once palpable tumors formed after approximately two weeks, mice were randomized into vehicle control or SAR405838 treatment arms (concentrations designated per experiment, ranging from 50-200 mg/kg).
Tumors were resected, weighed, and either snap-frozen in liquid nitrogen or paraffin embedded for sectioning and staining once the experiment was terminated.
Statistical analysis Each cell culture assay was performed in triplicate and standard error mean (SEM) was calculated. Statistical analysis was performed using the GraphPad Prism 6.01 Software
(La Jolla, Ca, USA). Specific tests of significance were detailed in figure legends. When t-
64 tests were performed, they were unpaired and two-tailed. Significance for cell culture based assays was set at: p<0.05 =*; p<0.005 = **; and p< 0.0001 = ***.
For in vivo experiments, the average volume (mm3) and SEM for all group variables per study were calculated and recorded. A two-sample t-test was used to assess the differences of an outcome at a single time point. Xenograft growth was assessed using a two-tailed
Student’s t-test. Significance was set at p ≤ 0.05.
65 CHAPTER 3. RESULTS: CREATION AND CHARACTERIZATION OF DEDIFFERENTIATED LIPOSARCOMA BIORESOURCES
SECTION 1. CREATION OF DEDIFFERENTIATED LIPOSARCOMA EXPERIMENTAL TOOLS
Perhaps the most impeding obstacle in DDLPS research is the lack of commercially available bioresources. Due to this limitation, we created stable cell lines and reproducible xenograft models in our lab that made the following studies possible.
Generation of patient derived DDLPS cell lines Cell lines that accurately represent DDLPS molecular features have been difficult to establish. Currently, there is only one ‘DDLPS’ cell line commercially available through the
American Type Culture Collection (ATCC): SW872. However, SW872 fails to accurately represent established DDLPS characteristics; SW872 lacks MDM2 amplification and has a mutated p53 (c.752 T > A). Consequently, our first goal towards studying DDLPS was to establish cell based models derived from patient samples that recapitulate clinical features.
Towards this goal, we have generated a protocol that efficiently generates DDLPS cell lines in culture as explained in the materials and methods. To date we have created >50 primary human DDLPS cell line cultures, of which 10 are successfully passaged beyond 50 times. It was these 10 cell lines that were used as our cell line panel to represent DDLPS for the following studies. The cell lines that make up our DDLPS cell line panel are listed in
Table 3.1.
66 Table 3.1 Characteristics of DDLPS cell lines frequently used for cell line panel. Cell lines created at the UTMDACC Sarcoma Research Center from primary patient tumors.
Name Histology Primary/Recurrent tumor Tumor location
Lipo224 DDLPS primary retroperitoneum Lipo224B DDLPS recurrent retroperitoneum Lipo246 DDLPS recurrent, secondary mesentery Lipo863 DDLPS recurrent retroperitoneum Lipo573 DDLPS recurrent, secondary mesentery Lipo815** DDLPS recurrent retroperitoneum Lipo615 DDLPS recurrent groin Lipo984 DDLPS recurrent, secondary retroperitoneum LipoDL635 DDLPS primary retroperitoneum Lipo318 DDLPS recurrent, secondary retroperitoneum **Indicates tumors cells believed to be isolated from a well-differentiated component of a DDLPS.
Conservation of molecular characteristics Once we were successful in growing disaggregated tumor cells in culture, it was critical to ensure that the molecular characteristics of the tumor from which the cell line was derived were retained. Therefore, prior to experimental use DDLPS cells were confirmed and characterized.
As previously explained, certain unique molecular characteristics are useful to classify each subtype of LPS, specifically, it is the MDM2 amplification in DDLPS that serves as the biomarker for histological classification13,51. In WD/DDLPS diagnosis, fluorescence in situ hybridization (FISH) studies are frequently performed on samples to reveal the amplification of the 12q13~15 region51 and, as a result, a differential diagnosis can be made between
67 WD/DDLPS and benign lipomas, as well as other high-grade sarcomas51. In normal cell biology, two red (fluorescent probe for 12q13~15 segment) and two green (fluorescent 12q centromeric probe) signals should be seen per cell representing a single copy of MDM2 per chromosome, as marked by the centromeric probe. In tumor cells, multiple red signals indicate multiple copies of the 12q13~15 chromosomal region and multiple green signals indicate cellular polysomy. Therefore, initial studies to confirm MDM2 amplification were performed on our primary cell cultures using FISH; MDM2 amplification was defined as a ratio of
MDM2/CEP12 ≥ 2.1 per 100 tumor cells. FISH images for each cell line in our LPS cell panel are represented in Figure 3.1. MDM2 amplification was observed in each of our DDLPS cell lines, but was not amplified in the three LPS cell lines SW872 (unclassified LPS), PLS1 (PLS) and LISA2 (PLS), deemed “control cells” in this body of work.
Although FISH was used primarily for conformational purposes, we observed that the percent of MDM2 amplified cells corresponded to the potential cell line establishment capability of the primary tissue culture. We observed that if a primary culture had >80%
MDM2 amplification it was able to progress from a primary cell strain to a reproducible cell line. While this is trend was not fail-proof, MDM2 amplification is associated with progression towards malignant phenotype; therefore, it is possible that this amplification influences a cell’s in vitro growth capability.
68
Figure 3.1 Fluorescence in situ hybridization (FISH) for the detection of MDM2 amplification in cell panel. The MDM2 gene is amplified in each DDLPS patient-derived cell.
Multiple copies of MDM2 (red probe) are seen compared to the CEP12-(green probe) per cell, therefore determined as positive for MDM2 amplification. Amplification is not seen in the three control liposarcoma cells SW872, PLS1 and LISA2.
69 Further characterizing these cells, we began with the most obvious molecular deregulation: MDM2 amplification and expression in our cell panel. Although MDM2 levels can be estimated using the FISH technique, the probe is not MDM2 specific and, instead, probes for a portion of the 12q15 chromosomal region that overlaps the MDM2 locus. In order to determine the level of MDM2 amplification, we performed qRT-PCR to quantify the genomic level of MDM2 in our cells. In this short panel, we compared 10 DDLPS cell lines to adipose-derived human mesenchymal stem cells (hMSC-AT) which was used as the control. Our data demonstrates that the genomic MDM2 levels, while variable (4.41 to
194.91), are without a doubt amplified in our DDLPS cell lines (Figure3.2A). Nevertheless, amplification does not necessarily mean that the gene will be expressed; therefore, our next step was to investigate the level of expression produced by the amplification. qRT-PCR and western blot assays were performed to determine the relative MDM2 expression levels at mRNA and protein level, respectively. As demonstrated in Figure 3.2B/C, MDM2 was not only amplified, but also highly expressed in our DDLPS cell lines. Interestingly, while the degree of amplification and expression varied in our panel, we observed a trend between the level of amplification and expression.
70
Figure 3.2 MDM2 is amplified and overexpressed in human DDLPS. A) qRT-PCR analysis of the genomic levels of MDM2 in DDLPS. B) MDM2 mRNA expression in LPS cell panel. C) Western blot of MDM2 protein expression in LPS panel.
71 TP53 mutational analysis DDLPS is a disease that retains WT p53; therefore, it was critical to determine the p53 status in our DDLPS cell line panel and confirm that p53 was not mutated. Sequencing was performed to examine mutations in exons 5-9 of the p53 gene. From the sequencing results, no mutations were found and; therefore, were deemed p53 WT. The LPS cell line SW872 was confirmed for p53 mutations; PLS1 was determined p53 null, and LISA2, retained a WT p53 status but lacked MDM2 amplification. These three non-DDLPS adipogenic-derived tumor cells (SW872, PLS1, and LISA2) were used concurrently throughout or studies to compare the results to cell lines with varying p53 and MDM2 status.
Figure 3.3 MDM2 and p53 protein expression in LPS panel. Western blot analysis of
MDM2 and p53 in our LPS panel.
72 Development of DDLPS xenograft models Developing xenograft models that faithfully recapitulate the clinical and molecular characteristics of DDLPS have proven to be a challenge; however, the ability to grow patient- derived cells in a xenograft model system offers a multitude of advantages to cell culture based assays. To that end, we generated a protocol to create DDLPS xenograft models to determine the tumor forming competence of our patient derived cell lines, which is well described in the materials and methods. To date, we have not successfully developed a WDLPS cell line derived xenograft model; however, while it was not relevant to this body of work, there have been reports that a WDLPS model is possible181.
For each cell line, we observed the latency period required to establish a palpable tumor. The two most reproducible DDLPS tumor-establishing cell lines are Lipo246 and
Lipo863. While all of our DDLPS cell lines successfully established tumors in mice, they grew much slower and did not generate consistent tumors as dependably as Lipo246 and
Lipo863. The latency of Lipo246 (14 days) and Lipo863 (24 days) allowed us the ability to estimate when an established tumor will be ready for further investigation, such as for evaluation of therapeutic response. The importance of establishing this type of model allowed us to determine which cells were able to be utilized as reproducible in vivo models providing us the ability for enhanced investigations to test tumor progression as well as test the efficacy of novel drugs.
Preservation of DDLPS histopathology in xenograft models It is important that our xenograft models preserved the histopathology of DDLPS; therefore, tumors were resected once they reached a size of 1.5cm in dimension in order to evaluate the morphology. From our resected tumor samples we found that our mouse models retained DDLPS features observed in patient sample (amplification of MDM2; Figure 3.4).
73
Figure 3.4 Histological analysis of DDLPS tumors from patients and mouse xenografts. This histology of patient tumors (Patient) is recapitulated in the xenograft tumors
(Xenografts). H&E staining representing the tumor morphology as well as immunohistological staining using antibodies against MDM2 demonstrate that the model represents the clinical features. (DDLPS tumor picture courtesy of Dr. Kristelle Lusby;
University of California, Irvine Medical Center.
74 SECTION 2. CHARACTERIZATION OF DEDIFFERENTIATED LIPOSARCOMA CELL LINES
Genomic profiling by array CGH analysis of DDLPS cell lines In order to investigate patterns of genomic aberrations in our DDLPS cell line panel, a genome wide, high resolution analysis of copy number variations (CNVs) was performed using array comparative genomic hybridization (aCGH:SNP Array 6.0 – Affymetrix platform). More than 946,000 probes were used for the detection of CNVs, thus, providing maximum panel power and the highest physical coverage of the genome. Our goal was to observe the CNVs acquired in order to enhance our understanding of genomic alterations associated with dedifferentiated liposarcomagenesis.
75 Figure 3.5 Copy number variations in DDLPS cell lines. The heatmap represents a genome wide, high-resolution analysis of CNV. More than 946,000 probes for the detection of CNV were utilized to provide maximum power and the highest physical coverage of the genome. Significant differences have been determined for DNA copy number fold changes between mesenchymal stem cells and 10 DDLPS cell lines. 95 genes where CNV is significantly different in >50% of DDLPS cell lines compared with mesenchymal stem cell controls. The blue coloring represents genes that are lost, while the red coloring represents genes that are amplified.
76 CNVs were analyzed in 10 DDLPS cell lines (Lipo224A, Lipo224B, Lipo246,
Lipo815, Lipo863, LPS141, Lipo318, Lipo573B, Lipo615, and Lipo984) and compared to two human adipocyte derived mesenchymal stem cells (hMSC-AT) from two different patients as controls. CNV changes were observed across the whole genome and the circos plot
(Figure 3.5) visualizes the chromosomal location of the amplified/deleted genes; for example, the amplifications seen in chromosome 12q are made obvious by the red coloration displayed on the plot. We then wanted to know which genes are the most significantly altered due to these genetic aberrations. In order to assess these differences numerically we assigned each gene a score. This score corresponds to the largest absolute copy number difference between a liposarcoma cell line and he mean of the two mesenchymal samples per gene. To identify genes with significant copy number differences, we implemented a two-sample t-test per genes. The heatmap in Figure 3.6 visually represents the copy numbers of the top 95 genes significantly differentiated in >50% of DDLPS cells compared to controls. Of these significantly up-regulated genes 39% of these were located in the 12q13~15 loci. The 12q21 loci harbored 17% of up-regulated genes, followed by the 5p chromosomal loci that harbored
13% of the most statistically significant genes. The most significant regions of CNVs lost were located at 11q23~24 (26%) and 11p15 (15%) (1.5 fold-change in at least 50% of cells).
77 Figure 3.6 Circos plot depicting copy number in the genome. Circos plot summarizes all CNV regions detected in our DDLPS cell lines and control samples; plot runs clockwise from chromosome 1 to X. The plot represents several tracks; the two innermost tracks represent the two hMSC-AT controls, while the following represent DDLPS samples. CNV regions are visualized by color, red representing multiple copies.
78 Significant gains in the chromosomal 12q13~15 loci were to be expected. The cumulative copy number plot in Figure 3.7 provides visualization of the physical location (x- axis) and copy number (y-axis) of the amplifications (green color) and deletions (red color) in chromosome 12 for all combined DDLPS cell lines. However, this is a discontinuous amplification in which the specific gene amplification frequency varies. The graph in Figure
3.7 represents genes located in chromosome 12q13~15 and their amplification frequency observed. MDM2, which is highlighted, is up-regulated in 100% our cells. The top 95 genes where CNV is significantly different in >50% of DDLPS cell lines compared with mesenchymal cell lines is listed in Appendix A.
79
Figure 3.7 The 12q13~15 amplification in DDLPS cell lines. The top panel demonstrates the frequency of amplified genes in chromosome bands 12q13~15 loci. The amplifications are distinguished by color as green represents amplification and red represents deletion in copy numbers. The bottom panel graphically visualizes the amplification frequency of genes located in the 12q13~15 loci detected in DDLPS cells. MDM2 has been highlighted.
80
While the focus of this body of work is mainly concerned with the amplification and expression of MDM2, this analysis suggests additional genomic regions that might be important for DDLPS tumor biology.
Identifying differential gene expression patterns in DDLPS cell lines The goal of this analysis was to detect differentially expressed mRNA in DDLPS cell lines compared to control cells. Data was acquired using the Illumina HT-12 v4 BeadChip assay from 10 DDLPS cells lines and 6 controls (2 hMSC-AT, 2 pre-adipocyte cell lines, and
2 adipocyte cell lines).
To identify differentially expressed genes between these two groups, modified 2- sample t-tests were performed. The obvious segregation of the differentially expressed genes between our panel of DDLPS cell lines and controls is visualized by the heatmap in Figure
3.8. This heatmap represents the 602 significantly differentially expressed genes (FDR 0.05, fold change >2.0, ± 2 STD). The red coloring represents genes that are over-expressed in
DDLPS cell lines, while the blue color represents the genes that are under-expressed. These results demonstrate that the differential probes can clearly differentiate between the two groups, as well as demonstrate consistency within each group. The differential genes with the strongest statistical significance between DDLPS and controls are listed in Appendix B.
81
Figure 3.8 Differentially expressed genes between DDLPS cell lines and controls.
Heatmap of the 602 most differentially expressed genes selected at a FDR 0.05 and fold change > 2.0. The expression values shown on the heatmap have been standardized ± 2 STD.
Red color represents genes over-expressed; blue color represents genes under-expressed in
DDLPS cell lines compared to controls.
82
We then wanted to investigate what pathways these gene expression profiles were modulating. To answer this question, Ingenuity Pathway Analysis (IPA; www.ingenuity.com) was utilized to analyze the up- and down-differentially expressed genes from our microarray.
Our results revealed that these genes enriched many signaling pathways involved in chromosomal instability, cell cycle control and proliferation. Figure 3.9 graphs the top 15 canonical pathways with the highest estimated probability of involvement from our gene set.
The red bars represent the percent of genes that are up-regulated in our cells in that specific pathway, while the green bars represent the percent of down-regulated genes.
Figure 3.9 Canonical signaling pathways modulated in DDLPS. The graph represents the pathways that were significantly different between DDLPS cell and controls. X axis, the name of the pathway; Y axis, the negative logarithm of the P value (-logP). The graph is distinguished by color; red represents genes in pathway that are up-regulated, green represents genes down-regulated
83 CHAPTER 4. RESULTS: THE MDM2:P53 AXIS IS A TARGETABLE THERAPEUTIC STRATEGY IN DEDIFFERENTIATED LIPOSARCOMA
The main goal of my study herein was to evaluate the potential anti-DDLPS effects of
SAR405838 in the pre-clinical context.
SECTION 1. TARGETING THE MDM2:P53 AXIS IN VITRO
SAR405838 inhibition stabilizes p53 and activates downstream targets Prior to determining the effects on DDLPS tumor growth and survival, we first confirmed that SAR405838 truly activated the p53 pathway and gave forth to formerly observed results. In previous studies, the small-molecule MDM2 antagonists Nutlin-3a and
MI-219 were successful in activating the p53 pathway in tumor cells that retain WT p53, but not in cells which the transcriptional properties of p53 have been mutated167. Therefore, these drugs were used to anticipate and confirm expected results elicited by SAR405838 in our LPS cell panel. Our cell panel consisted of 8 LPS cell lines: five DDLPS cells that exemplified
MDM2 amplification and WT p53 (Lipo224, Lipo246, Lipo863, LPS141, and Lipo815), and three PLS cell lines which had either WT p53, but lacked MDM2 amplification (LISA2), mutated p53 (SW872) or p53 null (PLS1). The latter three cell lines were deemed “controls” for this study.
To confirm that SAR405838 in fact hit its target (i.e. activated p53) we tested our cell panel first with Nutlin-3a and MI-219; DDLPS cells were incubated with incremental drug doses (0-10µM) of these for 24h. As shown by western blot analysis (Figure 4.1), p53 was stabilized and activated in a dose-dependent manner following incubation with these MDM2
84 inhibitors. Two p53 transcriptional targets, p21 and MDM2, confirmed the transcriptional activity induced by p53 stabilization in DDLPS cell lines (Figure 4.1). Following suit, we evaluated the activation of p53 when incubated with SAR405838. Again, we treated our cell panel with incremental doses (0-10µM) for 24h (Figure 4.2). Similar to our previous results,
SAR405838 treatment resulted in a dose-dependent effect on p53 and downstream targets; however, no effect was seen in the control cell lines, as expected. Compared to Nutlin-3a and
MI-219, SAR405838 induced p53 activity at lower doses (0.1nM versus 5µM and 1µM, respectively), as demonstrated in the increase in MDM2 and p21. Together, these results confirmed the initial evidence that SAR405838 induced p53 activity in DDLPS cell lines.
85
Figure 4.1 MDM2 antagonists stabilize p53 and induce the p53 pathway in DDLPS cells. Treatment of cultured LPS cells with MDM2 antagonists lead to concentration- dependent accumulation of p53 protein and its transcriptional targets, MDM2 and p21. Protein levels were analyzed by western blot analyses following 24h incubation with the indicated
MDM2 antagonist.
86
Figure 4.2 SAR405838 stabilizes p53 and induces the p53 pathway in WT p53 DDLPS cells. Treatment of cultured LPS cells with SAR405838 led to concentration-dependent accumulation of p53 protein and its transcriptional targets, MDM2 and p21. A) p53 activation was observed all DDLPS cell lines. B) No effect was seen in control cell lines SW872 or
PLS1. Protein levels were analyzed following 24 hour incubation with SAR405838.
87 SAR405838 inhibits DDLPS cell growth Further investigating the biological implications of SAR405838 in the cellular context of DDLPS, we evaluated the effects of MDM2 inhibition on cell growth. First, we incubated two exponentially proliferating DDLPS cells, Lipo246 and Lipo863, with each MDM2 inhibitor for 96 hours and evaluated the consequential effect on cell growth via MTS assay.
In order to institute the effective dose in which 50% of cell proliferation was abrogated (EC50), values of each inhibitor were calculated (Table 4.1). We found that while all drugs were capable of inhibiting cell proliferation, SAR405838 more potently abrogated cell growth as demonstrated by the lower EC50 values following SAR405838 compared to Nutlin-3a and MI-
219 in both cell lines (Figure 4.3).
88
Figure 4.3 Small molecule MDM2 inhibitors produce anti-proliferative effects.
MDM2 antagonists Nutlin-3a, MI-219, and SAR405838 result in a dose-dependent inhibition of cell growth and viability. Lipo246 (left panel) and Lipo863 (right panel) were treated with each inhibitor for 96 hours and cell viability was measured via MTS assay. (Graphs represent n=3 ± SEM).
TABLE 4.1 EC50 VALUES FOLLOWING TREATMENT WITH MDM2
Mean EC50 (µM)
Inhibitor Lipo246 Lipo863
Nutlin-3a 2.90 2.75
MI-219 2.34 3.03
SAR405838 0.31 0.49
89
SAR405838 was then tested in each cell line of our cell panel. Following 96 hour incubation with SAR405838, each DDLPS cell line exhibited a dose-dependent, anti- proliferative response; however, there was no effect on proliferation in the control cell lines at the concentrations tested (Figure 4.4). For each cell line, the EC50 value in response to
SAR405838 treatment at 96 hours was calculated (Table 4.2). Also from these results, we inferred that SAR405838 not only required functional, WT p53, but also amplified MDM2 expression as LISA2, which retains WT p53 but lacks MDM2 over-expression, was not affected by the inhibitor.
Figure 4.4 SAR405838 elicits anti-proliferative effects in DDLPS cell line panel. The cytotoxicity of SAR405838 depends on the p53 status and MDM2 amplification. Cell lines were incubated with SAR405838 for 96h to determine the anti-proliferative effects. Cellular viability was measured by MTS assay. (Graphs represent n=3 ± SEM).
90
Mean EC50 Cell line SAR405838 (µM) Lipo246 0.31 Lipo224 0.31 LPS141 0.13 Lipo863 0.49 Lipo815 0.37 SW872 >10 LISA2 >10 PLS1 >10
Table 4.2 EC50 values following SAR405838 treatment. Table representing the EC50 values of each cell line in our panel following 96h incubation with SAR405838 calculated with
BiostatSpeed software. Control cell lines had an unmeasurable EC50 value with the concentrations tested.
Wild type p53 is required for effective anti-DDLPS SAR405838 effects Following these initial experiments, we wanted to validate that WT p53 was mandatory for the effects induced by SAR405838. Thus far, our results revealed that
SAR405838 was not effective in inducing p53 activity effects in our cells with mutated/null p53 or when MDM2 was not amplified; therefore, to address this question, we used siRNA to transiently knockdown p53 in Lipo246 cells. Lipo246 parental mock and a non-targeting siRNA were used as controls. Following transient deletion (24h), Lipo246 cells were incubated with DMSO, SAR405383 (3µM) and Nutlin-3a (5µM) for 24 hours and protein expression level was evaluated. Compared to Lipo246 controls, the depletion of p53 resulted
91 in complete ablation of both MDM2 antagonists’ abilities to activate p53 and transcribe downstream targets, as seen in the western blot (Figure 4.5A). We then tested the anti- proliferative effects of SAR405838 in response of p53 silencing. Our results confirmed that when p53 is eliminated, SAR405838 is unable to induce the anti-proliferative effects (Figure
4.5B).
92
Figure 4.5 SAR405838 efficacy is dependent on the presence of WT p53. siRNA p53 knock out in Lipo246 depleted the ability for SAR405838 to stabilize p53 and activate p53 pathway functions. A) Western blot analysis confirmed p53 knockdown and showed that the expression of downstream p53 transcription targets, MDM2 and p21, was not increased when treated with MDM2 antagonists. B) p53 knock out eliminated the anti-proliferation effects of
SAR405838.
93 Activation of the p53 pathway halts cell cycle progression One of the main functions of p53 is to regulate cell cycle progression and protect cells from advancing through cell division. p21 is a transcriptional target of p53 and one of the main proteins responsible for inducing cell-cycle arrest. As previously observed (Figure 4.2), p21 was increased following SAR405838 treatment in DDLPS cells, suggesting that p53 is active and inducing target genes in the p53 pathway. Additionally, we observed that in all 8
DDLPS cells, p21 was highly expressed following incubation with 1µM of SAR405838.
Therefore, utilizing these results, we chose two concentrations for cell cycle analysis: 1µM and 3µM.
Exponentially proliferating cells were incubated for 48 hours with each inhibitor
(Figure 4.6) and the cell cycle distribution was assessed by flow cytometry. Our results revealed that SAR405838 successfully induced cell cycle arrest in all DDLPS cell lines. While the G1 and G2 arrest varied amongst cell lines, the average G1 fraction reduced from 57.29% to 53.28% and the G2 fraction increased from 16.55% to 33.55%. Also, the S-phase fraction decreased from an average of 26.15% to 13.15%. The variation observed between cells is likely due each cell’s innate proliferation rate and cycling time. Furthermore, there was a significant and sub-G1 fraction observed in DDLPS cells following SAR40838 treatment indicating possible apoptosis. Again, there was no effect on cell cycle progression in the control cell lines (Figure 4.7).
94
Figure 4.6 Inhibition of the MDM2:p53 complex induces cell cycle arrest. Treatment with Nutlin-3a (2.5, 5μM), MI-219 (1, 3μM), and SAR405838 (1, 3μM) for 48h resulted in
G1/G2 cell cycle arrest in DDLPS cells. Graphs represent the average of triplicate experiments
± SEM.
95
Figure 4.7 SAR405838 induces cell-cycle arrest. Treatment with SAR405838 (1,
3μM) for 48h resulted in G1/G2 cell cycle arrest in DDLPS cells. No change in the different cell cycle phases was seen in control cells (SW872 and PLS1). Graphs represent the average of triplicate experiments ± SEM.
96 SAR405838 induces p53 apoptotic response While the transcriptional activity of p53 controls both cell cycle arrest and induction of apoptosis, they are regulated by different downstream pathways and in certain circumstances p53 may opt to trigger apoptosis. Therefore, we wanted to see if SAR405838 potently induced apoptosis in DDLPS cells.
First, we evaluated the ability of SAR405838 to mimic Nutlin-3a and MI-219 induced apoptotic effects in Lipo246 and Lipo863 cell lines. We observed that SAR405838 was successful in penetrating cells and inducing cell cycle arrest at 1 and 3µM; however, due to the increase of MDM2 expression observed at previously with lower doses, as well as the
EC50 values previously established, we chose two lower concentrations (0.1 and 0.3µM) to determine the apoptotic response following SAR405838 (Figure 4.8). We treated exponentially proliferating cells with each inhibitor for 96 hours (SAR405838: 0, 0.1, 0.3µM;
Nutlin-3a: 0, 1, 2.5µM; and MI-219: 0, 0.1, 0.3µM). Coinciding with our results from the
MTS assay and EC50 vales, SAR405838 induced marked, dose-dependent apoptosis after 96 hour incubation and to a much higher degree than Nutlin-3a and MI-219 after the same amount of time as determined via Annexin V/PI staining FACS analyses. In Lipo246 cells, Nutlin-3a induced significant apoptosis (27.19%) after incubation with 2.5µM treatment; however, after only 0.3µM, SAR405838 induced 30.88% apoptosis (p<0.05). Lipo863 cells resulted in
28.99%, 11.92% and 24.12% apoptosis with the highest dose each drug of Nutlin-3a, MI-219 and SAR405838, respectively. Interestingly, we found that in both cell lines, MI-219 did not induce apoptosis, neither between control treatment and MI-219 treatment, nor between each dose (Figure 4.8).
97
Figure 4.8 Inhibition of the MDM2:p53 complex induces apoptosis. Treatment with
Nutlin-3a (1.0, 2.5μM), MI-219 (0.1, 0.3μM), and SAR405838 (0.1, 0.3μM) for 96h resulted in a dose-dependent increase in apoptosis following incubation with small molecule MDM2 inhibitors compared to control treatment (DMSO) in DDLPS cells. Apoptosis was measured via Annexin V/PI staining and FACS analysis. Graphs represent the average of triplicate experiments ± SEM; * denotes statistically significant effects (p<0.05)
We then investigated the apoptotic response in the rest of our cell panel. We incubated cells with DMSO, 0.1 or 0.3µM of SAR405838 for 96 hours and the percentage of apoptotic cells (Annexin V-positive portion) was determined via flow cytometry. Following 96 hours, all DDLPS cell lines displayed a dose-dependent apoptotic response as determined by the
Annexin V-positive cells. Although Annexin V staining is generally regarded as an indicator for early apoptosis, 96 hours of treatment proved to be an adequate time for cells to commit
98 to apoptosis and account for the possible delayed response in cells. To further confirm that these results are due to p53-induced apoptosis, cleaved caspase-3 expression was evaluated following 96 hours with SAR405838 treatment observed via western blot (Figure 4.9).
Taken together, these results demonstrate that MDM2 inhibition abrogated DDLPS cell proliferation through the induction of two p53 driven mechanisms: cell cycle arrest and apoptosis.
Figure 4.9 Apoptotic response of DDLPS cells to SAR405838. A) DDLPS cells were treated with 0 (DMSO control), 0.1 or 0.3 µM of SAR405838 for 96 hours. Graph represents the average number of Annexin V –positive cells (n=3, ± SEM, *=p < 0.05).
99 Amplification of MDM2 corresponds to therapeutic response Based on our results from these initial experiments, we observed that when subjected to MDM2 antagonists, our p53 WT DDLPS cells varied in response. Although all cells showed marked anti-proliferation, cell cycle arrest and apoptosis following treatment with
SAR405838, we wanted to further understand what might be the reason for these varied results. So, we looked into which cells had the strongest therapeutic response in each assay in an effort to find a common trend. We characterized these cells for MDM2 amplification and p53 mutational status (Table 4.3) and found that while each cell had amplified MDM2, the amplification varied between cell lines. Therefore we asked whether there was a potential correlation between MDM2 expression levels and therapeutic response. To address this question, we categorized our cell panel by their relative MDM2 expression level into two groups, MDM2HI or MDM2LO, and took a deeper look into our previous results. This analysis revealed that DDLPS cell lines expressing relatively higher amounts of MDM2 (Lipo246,
Lipo224, LPS141) had a lower overall EC50 compared to the relatively low MDM2 cell lines
(Lipo863, Lipo815) after 96h incubation with SAR405838 (0.25μM versus 0.43μM).
Furthermore, MDM2HI cells induced a higher percent of apoptosis than that of the MDM2LO group (30.88% versus 24.12%) following 0.3µM SAR405838.
100 TABLE 4.3 RELATIVE MDM2 AMPLIFICATION; MDM2HI AND MDM2LO.
MDM2 copy MDM2 mRNA HI/LO Cell line p53 Status number level MDM2
WT HI Lipo246 TP53 194.10 106.6 MDM2
WT HI Lipo224 TP53 65.25 44.81 MDM2
WT HI LPS141 TP53 131.22 112.7 MDM2
WT LO Lipo863 TP53 42.16 12.35 MDM2
WT LO Lipo815 TP53 50.15 14.23 MDM2
101 SECTION 2. TARGETING THE MDM2:P53 AXIS IN VIVO
Pharmacokinetics of SAR405838 To study the tumor penetration and stability of SAR405838, we performed a pharmacokinetics (PK) study in Lipo863 tumor bearing mice. Plasma concentrations were measured from 0 to 72 hours in mice following a single oral dose of 100 or 200 mg/kg of
SAR405838. The concentration-time data of SAR405838 is plotted (Figure 4.10); following administration, SAR405838 concentrations rapidly reached a maximum value (Cmax) at approximately 6 hours measuring 8200 ng/ML (at 100mg/kg) and 1630 ng/mL (at 200mg/kg)
(Table 4.4). After 72 hours, all data were below the lower limit of quantitation (LLOQ)
(5ng/mL) of the assay. The data demonstrates that drug accumulation was dose-dependent and at 200 mg/kg was almost 2-fold to 100 mg/kg administration. The AUC0-72 increased in a dose- dependent manner was approximately twice as high with 200mg/kg dose versus the 100mg/kg dose. These results suggest that the Cmax and AUClast increase proportionally with the dose
(Table 4.5).
102
Figure 4.10 Time-concentration profiles. Mean plasma concentration profiles following a single oral dose administration of SAR405838 (100 or 200 mg/kg).
TABLE 4.4 SUMMARY OF PK PARAMETERS.
TABLE 4.5 PK PARAMETERS RATIO VERSUS DOSE.
103
In order to better understand the concentration-effect relationship the relative gene and protein expression levels at each time point were assessed. Compared to vehicle, MDM2, p21 and PUMA mRNA all peaked following 6 hours of SAR405838 treatment; minimal difference between the 100 and 200 mg/kg treatment was observed at 6 hours. The MDM2 protein expression peaked for both 100 and 200 mg/kg at 6 hours after dose and continued to persist at 48 hours at the 200mg/kg dose whereas the levels for 100 mg/kg quickly dissipated. p21 followed a similar trend, between doses and retained its levels in 200mg/kg for 48 hours. proteolytic activation of caspase-3 increased in a dose and time dependent manner. Taken together, the effects of SAR405838 lead to p53 pathway activation in a as observed by the expression of p53 transcriptional targets (Figure 4.11).
104
Figure 4.11 Expression levels from DDLPS xenograft models following SAR405838 treatment. A) mRNA and B) protein expression levels of selected genes in response to time
(6-72 h) and a single dose (vehicle, 100 and 200 mg/kg, p.o.) of SAR405838. Data represents mean ± STD, n=3.
105 SAR405838 elicits potent anti-DDLPS tumor growth effects in xenograft models In response to our cell-culture based results, we sought to evaluate whether the anti-
DDLPS effects of SAR405838 could be replicated in vivo. Towards that end, multiple therapeutic experiments were performed. In all therapeutic experiments, two DDLPS cell lines, Lipo246 (2.5 x 106 cells injected s.c.) and Lipo863 (3.5 x 106 cells injected s.c.), were utilized and injected subcutaneously into female Balb/c nude SCID mice.
In our first therapeutic experiment we sought to evaluate the dose response following
SAR405838 treatment. DDLPS tumor bearing mice received either vehicle control
(TPGS/PEG200) or SAR405838 once a tumor was established (~5mm3). For Lipo863-tumor bearing mice, mice were grouped into four treatment arms and received oral administration of vehicle or bi-weekly (BIW) (100mg/kg) and weekly (100mg/kg and 200mg/kg) doses of
SAR405838. However, due to the unexpected death of two mice in the bi-weekly group, the dose was reduced to 50mg/kg BIW for the remainder of the experiment. Vehicle and 50 mg/kg/BIW treatment groups received a total of 7 weeks of treatment prior to termination, while the other two treatment groups continued dosage for an additional two weeks prior to termination as the tumor volume in these groups was very small. Upon termination, the tumor volume of each group was calculated. The average tumor volume of the vehicle group was
1932.84 mm3 whereas the average tumor volume of the 50mg/kg/BIW, 100mg/kg/wk and
200mg/kg/wk treatment groups were 137.66mm3, 2.55mm3 and 0.57mm3, respectively
(Figure 4.12).
106 Figure 4.12. Therapeutic effects of SAR405838 in vivo.
TABLE 4.6 TUMOR VOLUME OF LIPO863 TUMOR BEARING MICE.
Last day of First day of treatment Control mice sacrificed measurement
Avg. # of # of mice Avg. tumor Avg. tumor tumor # of mice with mice with volume volume volume tumor with tumor (±SEM) (±SEM) (±SEM) tumor 83.70 1932.84 1932.84 Vehicle 8/8 6/6 6/6 ± 10.71 ± 460.08 ± 460.08 84.91 137.66 137.66 50mg/kg/BIW* 8/8 4/6 4/6 ± 4.01 ± 80.77 ± 80.77 84.65 4.29 2.55 100mg/kg/wk 8/8 1/8 1/8 ± 10.12 ± 4.29 ± 2.55 86.57 0.51 0.57 200mg/kg/wk 8/8 1/7** 1/7 ± 10.18 ± 0.51 ± 0.57 * Dose changed from 100mg/kg/BIW to 50mg/kg/BIW after first week **One mouse died of unknown cause
107 In a similar manner and to provide an additional model for SAR405838 therapeutic response, Lipo246 tumor bearing mice were separated into four treatment groups and received oral administration of either vehicle, 50mg/kg, 100mg/kg or 200mg/kg once a week. On day
15, three mice from both vehicle and 50mg/kg treatment were terminated due to large tumor burden; the remaining mice from these two groups were terminated one week later and the mean tumor volume was calculated (vehicle: 2722.35 mm3; 50mg/kg: 1821.65 mm3). Mice receiving 100 and 200mg/kg doses continued treatment for an additional two weeks due to the anti-tumor response and small tumor sizes. Upon termination, there was a measurable tumor in 5/7 mice receiving 100mg/kg with an average tumor volume of 1047.80 mm3. More impressively, 0/8 mice that received the 200mg/kg dose had any measureable tumor when the experiment was terminated (Figure 4.13).
108
Figure 4.13 Therapeutic effects of SAR405838 in vivo.
Table 4.7 Tumor volume of Lipo246 tumor bearing mice.
First day of treatment Control mice sacrificed Last day of measurement
# of mice Avg. tumor # of mice Avg. tumor # of mice Avg. tumor with volume with volume with volume tumor (±SEM) tumor (±SEM) tumor (±SEM) 137.36 2722.35 2722.35 Vehicle 7/7 7/7 7/7 ± 22.75 ± 203.38 ± 203.38 137.13 1821.65 1821.65 50mg/kg/wk 7/7 7/7 7/7 ± 27.21 ± 377.11 ± 377.11 140.83 126.13 1047.80 100mg/kg/wk 7/7 6/7 5/7 ± 23.07 ± 75.83 ± 464.37 140.69 5.43 0.00 200mg/kg/wk 8/8 3/8 0/8 ± 26.73 ± 3.18 ± 0.00
Tables 4.6 and 4.7 list the tumor volume of each group on day one of treatment, the day the vehicle control group was sacrificed, as well as the last day of measurement in both
Lipo863 and Lipo246, respectively. Taken together, our data demonstrates that SAR405838 effectively inhibits the growth of DDLPS tumors in a dose-dependent manner in xenograft models.
109 Therapeutic effects of SAR405838 in large DDLPS tumor burden Patients with DDLPS often have very large, deep-seated, and invasive tumors that make complete surgical difficult; therefore, in an effort to recapitulate the large tumor burdens presented in patients, we allowed for DDLPS tumor bearing mice (n=5, Balb/c nude SCID mice) to form very large tumors (~400mm3) prior to treatment with SAR405838
(200mg/kg/wk p.o.) in our two xenograft models.
These results exemplified the anti-DDLPS effects of SAR405838 as treatment significantly abrogated the tumor growth in Lipo246 tumor bearing mice, showing a complete regression of tumor growth after only two treatments (412.48mm3 ± 47.9 average tumor volume prior to treatment, 0.00mm3 ± 0.00 after 14 days and termination on day 43) (Figure
4.14A). On the other hand, Lipo863 tumor bearing mice did not respond with such anti- tumorigenic enthusiasm. Following the first 3 doses, all tumors showed an initial regression, including one mouse with complete tumor burden elimination; however, tumors in the other mice began to show tumor regrowth and continued until termination of the experiment. The average tumor volume following six weeks of SAR405838 treatment was 492.56mm3 ±
171.95, these tumors began at a volume of 369.13mm3 ± 84.56 (Figure 4.14B).
110
Figure 4.14 The in vivo effects of SAR405838 on established large tumor burdens.
Therapeutic effects of SAR405838 in DDLPS xenografts. Mice were administered
200mg/kg/wk p.o. The effects on two different DDLPS models were observed. A) Lipo246 tumor bearing mice. B) Lipo863 tumor bearing mice. (n=5 per group)
111 CHAPTER 5. RESULTS: GENE EXPRESSION PROFILING IDENTIFIES IMPORTANT GENES MODULATED BY SAR405838
SECTION 1. IDENTIFYING MOLECULAR CHANGES INDUCED BY SAR405838 FOR POTENTIAL BIOMARKERS OF THERAPEUTIC RESPONSE
The goal of the studies within this chapter was to determine the SAR405838 induced effects on downstream targets in both cell culture and in xenograft models. This knowledge can potentially highlight and/or unravel novel p53 targets not previously known in the
MDM2-p53 regulatory pathway, as well as give insight into biomarkers of therapeutic response that can potentially be used clinically. To address this goal, we performed two microarray analyses using an Illumina Human ref-12 v4.0 bead chip and compared global gene expression. The first array investigated the differentially expressed genes following
SAR405838 treatment in four DDLPS cells in vitro; the second array investigated the modulated genes in vivo. In order to ensure consistency and significance of these expression changes, the two arrays were combined for a final analysis.
Identifying molecular deregulations induced by SAR405838 in dedifferentiated liposarcoma in vitro In the first stage, we performed a gene expression array based on in vitro results; the goal of this study was to detect mRNAs that are differentially expressed before and after treatment with SAR405838. As the schematic in Figure 5.1 shows, two independent biological replicates of four DDLPS cell lines (Lipo224, Lipo815, Lipo246 and Lipo863) were treated with 0.003% DMSO (control), 1µM, and 3µM of SAR405838 for 24 hours. Following
24 hours, cells were harvested and RNA was extracted.
112
Figure 5.1 Schema of the experimental outline. Schema of the experiments performed to investigate differentially expressed genes before and after SAR05838 treatment.
The dataset was acquired using the Illumina HT-12 v4 Bead Chip and for each cell line, the p-value and fold change for treatment (1µM + 3µM) profiles versus the DMSO control profiles were computed using the log-transformed data. Genes were selected that obtained a significant p-value (p<0.05) and a fold change >1.4 or <1/1.4 (treatment profiles
(1µM + 3µM) versus the DMSO profiles) for any three out of the four cell lines investigated;
113 a total of 858 probes were found to lay within these parameters for these profiles. Results showed that the expression patterns were quite similar from between cell lines. Furthermore,
1µM versus 3µM patterns were largely similar as well, although the gene patterns appeared to be slightly stronger in the 3µM group, as might be expected. Of these differentially expressed genes found, a total of 381 up-regulated and 477 down-regulated genes compared to DMSO controls in at least three out of four cell lines (p<0.05, FC>1.4 or <1/1.4).
114
Figure 5.2 Gene expression profiling in DDLPS cells in response to SAR405838. The heatmap of untreated and SAR405838-treated DDLPS cells demonstrates the 472 most differentially expressed significantly up- and down-regulated gene in DDLPS cells. Red indicates significantly up-regulated genes, and blue indicates down-regulated genes selected at FDR 0.05 and a fold change >2.0. The expression values shown on the heatmap have been standardized and at ± 3 STD.
115
Gene ontologies (GOs) of the differential expression genes were analyzed. The bar plot in Figure 5.3 represents the number of genes that were down-regulated in our microarray that enrich the top 20 GOs. These results show that the top down regulated are heavily involved in mitosis and cell cycle regulation. Cell cycle processes, kinetochore, spindle, regulation of mitosis, etc. are all processes that relate to cell division and mitosis, implying that if the genes enriching these processes are significantly down-regulated, it is likely that the process of proliferation is being slowed.
Figure 5.3 Gene Ontology Enrichment in DDLPS with SAR405838 treatment. Bar plot of the number of genes in the top 20 gene ontology groups.
116 Identifying molecular deregulations induced by SAR405838 in dedifferentiated liposarcoma in vivo In our second microarray, we wanted to perform a similar experiment identifying which genes are being modulated following SAR405838 treatment in vivo. Lipo246 tumor bearing xenografts were utilized to investigate the changes in gene expression. Once palpable tumors were established, mice (n=5) were given a single dose of either vehicle
(TPGS/PEG200) or SAR405838 (200mg/kg); following 24 hours, mice were sacrificed and tumors were extracted for RNA and protein evaluation. RNA was extracted from tumor samples of each group (n=3) for microarray analysis (Illumina Platform, HT v12).
From our microarray data, we detected 1074 (586 up-regulated, 488 down-regulated) genes to be significantly different following SAR405838 treatment (fold change >2.0, p<0.01). The differential gene expressions are clearly delineated by the heatmap in Figure
5.4. Gene expression was normalized to the controls and then separated between up-regulated
(yellow) and down-regulated (blue) following SAR405838 treatment.
Appendix C lists of the top 50 significantly affected genes following SAR405838 treatment in vitro and in vivo.
117
Figure 5.4 Gene expression profiles in response to SAR405838 treatment in vivo. The heatmap represents the 1074 most significant differentially expressed up- and down-regulated gene in DDLPS cells. Yellow indicates significantly up-regulated genes, and blue indicates down-regulated genes selected at FDR 0.05 and a fold change >2.0 or <0.5 and p<0.01.
118 SECTION 2. CANONICAL PATHWAYS ENRICHED BY GENES DIFFERENTIALLY EXPRESSED FOLLOWING TREATMENT WITH SAR405838 In an effort to determine which enriched pathways held significant biological relevance, we crossed our two independent microarray datasets. Intersecting these gene lists revealed that 237 genes were significant in both analyses, of which 115 were down-regulated in both analyses and 122 were up-regulated in both analyses. These differentially expressed genes were then selected for subsequent analysis in order to determine commonly affected pathways in DDLPS following SAR405838 treatment. We used Ingenuity Pathway Analysis to determine which signaling pathways were enriched from the differentially expressed genes of our microarray results (Figure 5.5).
119
Figure 5.5 Canonical pathways modulated by differentially expressed genes. Venn diagrams represent the matching differentially expressed genes that are up-regulated (left panel) and down-regulated (right panel) following SAR405838 treatment between in vitro and in vivo gene expression microarrays. These combined 237 genes (122 up- and 115 down-regulated) enrich multiple pathways determined by IPA; select genes are listed next to the associated pathway.
120
Our results demonstrated that the expressed genes enriched 171 canonical pathways;
Table 5.1 lists the top 3 canonical pathways with the highest estimated probability involvement. This analysis revealed that the p53 signaling pathway was the most affected pathway modified by both up and down-regulated genes from our two microarrays. Other pathways that were affected included pathways important in genomic stability such as cell cycle and checkpoint regulation (p53 signaling, G2/M and G1/S checkpoint regulation), and response to DNA damage (role of BRCA1 in DNA damage response, ATM signaling). In order to confirm these results, qRT-PCR was performed on select genes with RNA used in each microarray.
TABLE 5.1 TOP THREE INGENUITY CANONICAL PATHWAYS.
Ingenuity canonical Overexpressed genes in pathway P-value pathway
BCL2L1, TP53INP1, GADD45A, ADCK3, p53 Signaling CDKN1A, CCNK, TNFRSF10B, RRM2B, 1.17E-09 MDM2, BAX, FAS, DRAM1
Cell Cycle Control of MCM5, MCM3, MCM2, CDC7, MCM7 1.21E-05 Chromosomal Replication
Cell Cycle: G2/M DNA GADD45A, CDKN1A, TOP2A, MDM2, Damage Checkpoint 1.24E-04 CCNB1 Regulation
121 Microarray gene profiling in DDLPS cells treated with SAR405838 demonstrated up- and down-regulated genes in a p53-dependent manner Among the affected genes, we detected several p53-regulated targets. IPA transcription factor analysis revealed that p53 was the most significantly activated transcription factor (z-score of 5.643). qRT-PCR analysis validated several of these genes.
The data from the qRT-PCR was very similar to the microarray analyses. Together,
SAR405838 reactivated p53 activity and affected many p53 target genes in a p53-dependent manner.
One of the main cellular functions of p53 is regulation of the cell cycle control by preventing cell progression in G1/S and G2/M phases; we observed that treatment with
SAR405838 induced cell cycle arrest in DDLPS cells. As a transcription factor, p53 transcriptionally regulates the expression of multiple genes involved in this pathway. From our two arrays, our results revealed that pathways involved in cell cycle progression and genomic instability were significantly affected following SAR405838 treatment. To validate the induced effects of SAR405838 on cell cycle regulation we performed qRT-PCR (Figure
5.6). The selected genes are all involved in different checkpoint regulations of the cell cycle and were used to for confirmation purposes of the arrays (Table 5.2).
122
Figure 5.6 Differentially expressed genes following SAR405838 treatment involved in cell cycle control. qRT-PCR was used to confirm mRNA expression of differentially expressed genes . (*=p<0.05; student’s t-test compared to vehicle control; n=3 ± SEM).
Table 5.2 Genes identified by both microarray analyses involved in cell cycle regulation.
Select genes from our two microarrays and their protein function.
Gene symbol Function Pan cyclin-dependent kinase inhibitor; regulates cell cycle CDKN1A progression GADD45a DNA damage repair and promotes growth arrest at G2/M Kinase involved in spindle checkpoint function, inhibits BUB1 activation of the anaphase Microtubule-dependent processes: nuclear movement prior CDC20 to anaphase and chromosome separation. PBK Mitotic kinase, regulates cell cycle progression at G2/M
123 To gain further insight into the connection between p53 signaling and the mitotic pathway, we studied p53-induced genes from the expression arrays analyses. The increased expression of Aurora Kinase A, Aurora Kinase B and PLK1 was observed and the connection to p53 is demonstrated in Figure 5.7A. So our question became: does p53 regulate the aurora pathway? To answer this, we investigated our top down regulated gene list and crossed it with potential interactions using IPA and found that there were potential direct and indirect pathway interactions that deserved further exploration. We further confirmed these results with WB analysis (Figure 5.7B). Lipo863 cells were treated with increasing concentrations of SAR405838 for 48 hours and our results show that there is a decrease in Aurora A when treated with SAR405838.
124
Figure 5.7 Connections between p53 signaling and the mitotic pathway. A) The shortest connection between p53 and the mitotic pathway; red represents p53 affected genes.
Solid arrows represent direct transcriptional regulation. B) Western blot analysis of
SAR405838 treatment (48 h) and its effect on the expression of Aurora Kinase A.
125 While cell cycle arrest and the induction of apoptosis are both p53 driven processes, they involve very different downstream target genes. Ideally, the desired result of p53 based therapies is the induction of apoptosis. Therefore, to gain a better understanding of
SAR405838 induced apoptosis in DDLPS, we investigated the apoptotic genes modified from our expression arrays. Activation of p53 following SAR405838 displayed mutual apoptotic profiles in both arrays; increase in two p53 dependent apoptotic genes, BAX and PUMA were observed in both arrays and confirmed with qRT-PCR (Figure 5.9).
126
Figure 5.8 SAR405838 induces apoptotic gene profiles in DDLPS. Increase of apoptotic gene expressions from microarray analyses were confirmed with qRT-PCR. (n=3 ± SEM).
Table 5.3 Genes identified by both microarray analyses involved in apoptosis. Select genes from our two microarrays and their protein function.
Gene symbol Function BCL2L1 Inhibits activation of caspases in apoptosis BAX Antagonizes BCL2, promotes activation of caspase-3 FAS Cell surface death receptor, recruits caspase-8 to activated receptor PUMA Induces mitochondrial dysfunction and caspase activation
127 SAR405838 modulates the expression of genes in the 12q13~15 chromosomal loci Given that the 12q13~15 chromosomal region is amplified in DDLPS, we wanted to take a deeper look into which genes were being modulated in this amplicon following
SAR405838 treatment. To do this, we utilized all three of our microarray analyses. Genes that were over-expressed in DDLPS cells compared to controls (Figure 3.7) were then further analyzed following SAR405838 treatment. From this analysis we found 3 genes that were up- regulated and 5 that were down-regulated following MDM2 inhibition (Table 5.4). qRT-PCR was performed to validate the expression following (Figure 5.9).
Figure 5.9 Genes modulated in the 12q13~15 amplicon following SAR405838 treatment. qRT-PCR was performed to confirm (A) genes up-regulated and (B) down-regulated following SAR405838 treatment. (n=3 ± SEM).
128
Table 5.4 Genes identified by both microarray analyses located within the 12q13~15 chromosomal interval. Select genes from our two microarrays and their protein function.
Gene symbol Function AVIL Formation of actin-containing structure Role in metabolism and tissue Up-regulated CYP27B1 differentiation MDM2 E3 ubiquitin ligase activity Nuclear pore complex assembly and/or NUP107 maintenance Polymerase, synthesizes RNA primers PRIM1 made during DNA replication Down-regulated XRCC6BP1 DNA dependent protein kinase activity Structural constituent of cytoskeleton and YEATS4 sequence-specific DNA binding transcription factor
129 CHAPTER 6. DISCUSSION
Dedifferentiated liposarcoma is a devastating disease with very few treatment options.
More than 90% of DDLPS patients will suffer recurrent disease and, disturbingly, a mere
5.2% of patients with advanced stage disease will ultimately survive23. In this body of work we sought to create and characterize DDLPS bioresources as well as target MDM2 as a potential therapeutic strategy for the treatment of DDLPS.
Establishment and characterization of novel DDLPS models One of the major limitations to achieve an enhanced molecular understanding, as in many rare cancers, is the lack of experimental models that can be used to investigate this disease which hinders pre-clinical and clinical advances75. Consequently, creating a model system that mimics the human disease is a vital first step. The most common model system used in cancer research involves cell lines; however, there is only one commercially available cell line for the study of DDLPS (SW872)181. Cell lines are an effective way to initially study the molecular aberrations and mechanisms in DDLPS and for pre-clinical testing of potential therapeutic agents. For these reasons, we first developed novel DDLPS bioresources to pave the way for many future experiments. The DDLPS cell line panel used in these studies represents our 10 most successful and reproducible cell lines generated in our lab. Evaluation of DDLPS in vitro enables pre-clinical investigations, as well as enhances our understanding of the molecular underpinnings. However, this is complicated by the lack of relevant baseline or control cell lines in which to make comparative conclusions of significance. In an attempt to overcome these issues we have created an in depth DDLPS “encyclopedia” with our cell line model to understand the baseline properties that can be used to compare results and make
130 more meaningful conclusions. This initial characterization also provided information on the molecular phenotypes of each cell line that could then be used to predict responses, thereby making even better use of these rare resources.
Another common way to study cancer biology uses animal models. Animal models for well-differentiated liposarcoma have been recently developed. One group reported a mouse model that spontaneously developed a WDLPS when fed a high fat diet in the presence of IL-
22 over expression182 while another developed a zebrafish model that expressed a transactivation mutation in p53 and constitutively active AKT leading to WDLPS tumors183.
Nonetheless, no such animal model exists for the spontaneous development of DDLPS. The most common animal-based model to study to DDLPS is a xenograft system. Xenograft models are able to grow human tissue from tissue samples or cell cultures in an immune compromised mouse. We were able to generate xenografts from all 10 of our cell lines, and the molecular and morphological features in our xenografts mimic those of the tumors from which they were derived. With the development of this system, we were able to identify molecular aberrations as well as test the pre-clinical responses to therapeutic drugs.
Another limitation for studying DDLPS is the relatively small patient cohort in each
LPS histological subgroup; consequently, LPS is frequently lumped together as though it were one single disease. This broad and potentially misleading categorization has slowly dissipated over the years as subtype-specific therapies, based on in depth characterizations, have been developed67,184. However, minimal therapeutic advances and overall low success rates of these newer therapies highlight the urgent need for the development of new drugs based on specific molecular abnormalities and molecular driving forces unique to each subtype. With this in
131 mind, we characterized our panel of DDLPS cell lines to establish genomic- and expression- based profiles.
Cytogenetic studies are perhaps the most comprehensive investigative tool available to learn about the genomic alterations influencing dedifferentiated liposarcomagenesis. While limited, tissue-based aCGH and gene expression profiling studies have been performed to investigate the molecular differences regarding tumor establishment and progression as well as histological classification7,54,57,61,66,185,186. We hypothesized that these genetic aberrations were reproducible and consistent as per specific malignant phenotypes. To test this hypothesis, we used aCGH analysis to determine genomic aberrations acquired by DDLPS, perhaps from their mesenchymal stem cell precursors. DDLPS cell lines, as a group, clustered and segregated together compared to controls. The results suggest that although cancer is heterogeneous, DDLPS acquires certain common and reproducible genetic changes that distinguish it from other mesenchymal cells.
With this in mind, the genomic loci alterations and the associated gene expression in our DDLPS cell panel were further assessed. Our analyses revealed patterns of genomic imbalances, specifically gains in the chromosomal regions 12q13~15, 12q21 and 5p, as well as losses in 11q23~24 and 11p15.
FISH and aCGH studies have clearly visualized and demonstrated amplification of
12q13~15 material located within the ring and giant chromosomes of DDLPS. Therefore, amplification of 12q13~15 was expected and confirmed by our data. The delineated amplification of 12q13~15 corresponds to variable amplification of oncogenes involved in tumorigenesis. The most important amplification in the 12q13~15 region is MDM2. MDM2 amplification was observed in our cells via FISH as a common finding in our cells and used
132 to confirm diagnosis of a primary tumor tissue as well as to confirm that molecular characteristics were retained in subsequent passages of our cell lines.
The ability to efficiently and effectively diagnose the tumor histology provides valuable information relevant to treatment. In one study, Weaver and colleagues (2010) investigated the use of core needle biopsies as a reliable tool for diagnosis for WDLPS compared to benign lipomas28. Their results showed that FISH had higher sensitivity (100%) and specificity (100%) compared to the immunohistochemistry analysis of core needle biopsies (65% and 89%, respectively). However; in a recent study assessing the accuracy of biopsy techniques and resection specimen of 257 patients, the diagnostic accuracy of a core biopsy was only accurate in 78% of patient tumors187. These results vary from center to center, but the ability to confirm MDM2 amplification via a core needle biopsy can only be as accurate as the biopsy itself, mandating that tumor rather than normal tissues is sampled. Nevertheless, not only were we able to confirm the genomic amplification as relevant to histological purposes, but the percent of MDM2 amplification in our primary cell cultures also gave us insight into the ability of these cells to grow in vitro.
Genetic changes provide insight into the establishment and progression of DDLPS; the relationship between disease progression and amplification of chromosomal loci has been linked in previous studies. For example, it has been suggested that the initial amplification of chromosome 12q13~15 gives rise to the morphological appearance of WDLPS from an adipogenic cell56,188. While the 12q13~15 amplicon is amplified in the WDLPS/DDLPS subtype, the progression from a WDLPS to a DDLPS phenotype requires additional specific genetic changes.
133 Amplification of 1p31~32 and 6q23~24 amplicons have been reported to be mutually exclusive, resulting in the up-regulation of JUN or MAP3K5 which correspond to each loci, respectively. However, in our study, we did not observe significant amplification at these loci.
Additionally, gains within 13q have recently been shown to correlate with worse prognosis in
DDLPS patient samples66; however, there were no striking amplifications observed in our array studies.
It is critical to distinguish which aberrations are driving DDLPS transformation; therefore, by developing and characterizing novel model systems we were able to acquire new information about the biology of DDLPS that can be used to not only better understand the disease but also serve as a target for potential therapies.
SAR405838 restores the p53 pathway in DDLPS cells Small molecule MDM2 inhibitors have provided researchers with the opportunity to scrutinize the molecular mechanisms of p53 due to their potency and specificity. Inhibition of MDM2 leads to the canonical cellular responses attributable to p53 activation; as a result
MDM2 inhibition is an exciting possibility for DDLPS treatment. Two of these processes,
(cell cycle and apoptosis) were therefore investigated for their response to SAR405838 treatment.
The cell cycle is regulated by cyclin-CDK complexes that prevent transition through the cell cycle in a well-organized check and balances scenario: G1 progression is regulated by cyclin D-CDK4/6; cyclin E-CDK2 regulates the G1-S transition; S-phase progression is regulated by cyclin A-CDK2; cyclin A/B-CDK1 regulates the entry into M phase78,79. We observed that SAR405838 modulated the cell cycle profile in DDLPS cells in a dose-
134 dependent manner (0 to 3μmol/L) and induced cell-cycle arrest in both G1 and G2 phases, depending on the cell line. In a similar study using Nutlin-3a, this differential phase response was also observed depending on the cell line studied. It was observed that cell arrest in G1 was due to suppressed phosphorylation of the retinoblastoma protein (Rb), while G2 arrest was achieved following suppression of CDC2 and cyclin B189. Therefore, it is possible that suppression of Rb phosphorylation is greater in cells manifesting in G1 phase arrest, while cell arrest in G2 may be due to suppression of the CDC2/cyclin B complex.
p53 regulates cell cycle largely through transcriptional activation of CDKN1A (p21).
Indeed, the p53-dependent increase of p21 expression following MDM2 inhibition is well documented106,167,171,173,190–196. In agreement, we observed increased p21 expression throughout our experiments, and p21 levels were used as a standard to represent p53 activation in our studies. CCNB1 encodes an important regulator of the G2/M DNA damage checkpoint, cyclin-B1, and was down-regulated following SAR405838 treatment compared to controls as demonstrated in our microarrays. High expression levels of cyclin B1 provides not only evidence of unregulated cell cycle progression but is also associated with tumor progression and a worse prognosis197. Cyclin B1 forms a complex with Cdk1 and is predominately expressed during the early G2/M phase of the cell cycle. Following SAR405838 treatment,
CCNB1 expression was decreased (FC -2.14) and provides futher evidence for p53-induced cell cycle arrest.
Identifying which cell cycle-regulated genes are being modulated is important when investigating the proliferation signature that may be drug-induced. However, when investigating the genes involved in cell-cycle following treatment it is possible that the decrease in cell proliferation is due to cells are not entering the cell cycle. Consequently, these
135 genes may not be directly affected by SAR405838 and are not expressed because the cells are not proliferating at that particular time. In order to effectively identify cell-cycle-regulated genes, synchronization of the cells is important; while we did not synchronize the cells prior to our investigations, SAR405838 produced a strong cell cycle expression profile as the modulation of multiple cell cycle regulating genes were observed in our gene expression arrays.
Taken together, MDM2 antagonist drugs have shown to induce cell cycle arrest through reactivation of the p53 pathway; therefore, by identifying genes that are directly affected by MDM2 antagonists, this can potentially predict if a tumor is experiencing a cell cycle arrest, which often leads to cytostatic response, or is due to an effect on proliferation and therefore a possible more desirable cytotoxic response.
Apoptosis is an orchestrated and programmed process of cell-death. The cleavage of caspases in the cytoplasm and nucleus produces a proteolytic caspase cascade that triggers cell death198. This evolutionarily conserved process is characterized from other forms of cell death (i.e., necrosis) by morphological features observed including membrane blebbing, DNA fragmentation, as well as the formation of distinct apoptotic bodies198.
The cytotoxic effects of MDM2 inhibitors rely on the innate ability of p53 to induce apoptosis. Partly by promoting the specific downstream target genes, and partly by repressing the expression of anti-apoptotic proteins, p53 is able to initiate components of apoptosis. In our studies we have demonstrated the induction of apoptosis in multiple experiments. We observed a dose-dependent correlation of apoptosis following SAR405838 treatment in cell culture-based assays, including Annexin V/FITC, as well as by the increased expression of
136 apoptotic genes and proteins via qRT-PCR and western blot assays, respectively.
Additionally, the resistance to apoptosis correlated with either p53 mutations or a lack of
MDM2 amplification. In comparison to cell cycle arrest, apoptosis is a more desirable therapeutic response; therefore, it is gratifying that SAR405838 was capable of inducing these cytotoxic results more effectively than both Nutlin-3a and MI-219 as a means if gaining therapeutic leverage.
Following SAR405838 treatment, an apoptotic expression profile was observed in our microarray analyses. Specifically, the expression of two genes (PUMA and BAX) was increased in response to MDM2 inhibition. The pro-apoptotic protein PUMA interacts with anti-apoptotic Bcl-2 family members, thus inhibiting their interaction with Bax and Bak and as a result, Bax activates mitochondrial dysfunction leading to caspase activation and cell death86,105,199. PUMA is capable of inducing apoptosis rapidly200; however, at the earliest time point we examined its expression was following 24 hours of treatment; therefore, SAR405838 may be capable of inducing apoptosis at much earlier time points. It is very plausible that these genes have been selectively modulated due to their involvement in tumor development.
Additionally, SAR405838 could assist in identifying which of these aberrations are most prevalent in the p53 pathway and potentially aid in the prediction of patient selection and response.
Malignant progression and chemo-resistance are frequently seen when tumor cells no longer exhibit an apoptotic response; therefore, a better understanding of p53-induced apoptotic genes may be relevant to therapeutic advancements. Because p53 apoptotic targets frequently retain a WT status201, it suggests the possibility of targeted therapies that are not dependent on the status and activation of p53.
137
P53; a choice between cell life and death While the mechanisms of cell cycle and apoptosis have been well characterized, the question remains how does p53 determine a cell's fate and participate in either life (cell cycle arrest) or death (apoptosis), a process that is based on a multitude of different factors.
Depending on the physiologically relevant circumstance, p53 induces different genes that mediate different p53 functions.
It has been suggested that a p53 “affinity model” is feasible in which the low or high expression level of p53 is able to induce transcriptional targets involved in either cell cycle arrest or apoptosis, respectively. Recently, Kracikova and colleagues challenged this model by showing that p53 is able to induce apoptotic responses even at very low p53 expression levels202. Most importantly, they demonstrated that lowering the ‘apoptotic threshold’ by inhibiting the anti-apoptotic BCL2 gene sensitized the cells to apoptosis. In our results we have demonstrated that both MDM2 and p53 levels vary. MDM2 levels have been suggested as a possible biomarker for therapeutic efficacy; consistently, cells expressing high MDM2 levels had a more potent response to MDM2 inhibition, whereas ‘low’ MDM2 expressing cells did not elicit as strong a response. Our results also demonstrate that very low doses of
SAR405838 (0.1nM) induced the expression of the cell cycle-arrest associated gene p21; however, although not significant, we did notice apoptosis at this low concentration. As a result, it is a possible that by lowering the apoptotic threshold to switch from cell cycle arrest to apoptosis by concomitant BCL2 inhibition with an inhibitor of apoptosis, such as ABT-263
(an inhibitor of the BCL2 family genes), apoptosis could be triggered at extremely low doses in cells that would not otherwise do so. This not only highlights the use of MDM2 as a
138 biomarker of response, but also suggests a way to induce a more potent response. In support of this hypothesis, results from our differentially expressed gene arrays showed that the major pathways affected by SAR405838 treatment were p53 signaling, cell cycle: G2/M DNA damage checkpoint, and ATM signaling for which the associated genes were pro-arrest and pro-survival, such as CDKN1A, GADD45a, and BCL2. While this can be tested in vitro, the more pertinent results would be an in vivo model. Lipo863, which has a low MDM2 expression, would provide an excellent construct with which to test this hypothesis.
Another possibility that might influence p53 effects is due to the basal level and expression of certain proteins that modulate different downstream pathways of p53. It is an interesting and therapeutically applicable observation that in response to p53, tumor cells seem to have a greater propensity for death as opposed to normal tissues. Malignant transformation is supplemented with uncontrolled proliferation, and loss of normal cellular environment as well as stress can render cancer cells more sensitive to apoptotic signals203.
Alternatively, normal cells are not subject to the same conditions and their intact survival mechanisms may be better suited to override the death signals instituted by p53. The basal levels of p53 are constitutively expressed in normal cells, and as a result, p53 is able to transiently activate cellular responses in normal or stress situations. This activation is not sufficient to elicit apoptosis possibly because either the specific stimuli induce different transcriptional processes or perhaps the nature of the response is only temporary.
Nevertheless, this transient activation of p53 can possibly be enough to deter early tumorigenic events.
How p53 contributes to reversible cell cycle arrest versus irreversible apoptotic responses is of great relevance to cancer therapies. As we continue to investigate the biology
139 underlying these mechanisms, we may be able to develop enhanced therapeutics that selectively lead to a desired process.
Anti-tumor response of SAR405838 in xenograft models Although cellular-based assays are efficient to assess the anti-proliferative effects of a drug, the results from an animal model are more often considered as representative of clinical behavior. The effects of MDM2 inhibition in xenograft models have been considered using
Nutlin-3a and MI-219. Despite initial pharmacodynamic experiments demonstrating p53 activation; both drugs failed to achieve complete tumor regression173. Throughout our studies, we demonstrated that SAR405838 had exemplary results and induced the anticipated favorable effects of an MDM2 inhibitor better than either Nutin-3a or MI-219. As a result, we had hoped that in vivo SAR405838 treatment would result in strong anti-tumorigenic responses. Indeed, after a single oral dose of SAR405838, we did observe activation of p53 in our pharmacodynamic studies, as suggested by the accumulation of p53 downstream targets
(MDM2 and p21), and a dose-dependent response in these initial therapeutic experiments.
Interestingly, we observed two completely different responses between our two cell lines in large tumor burden studies. Mice that were injected with Lipo246 maintained complete responses following two weeks of treatment (200mg/kg), whereas Lipo863 appeared to have an initial response after only one week of treatment, but were no longer affected by treatment in that the tumors continued to grow. One possible explanation for this is that the MDM2 levels in Lipo863 are not expressed highly enough for SAR405838 to have sustained therapeutic effects after this treatment span. We noticed in the LISA2 cell line which retains
WT MDM2 and TP53 that SAR405838 did not induce anti-tumor responses, and a similar
140 situation may hold true for Lipo863. Although SAR405838 was able to elicit anti-tumorigenic effects when the tumor is very small, these may not be enough in a larger tumor setting where additional abrogations independent of MDM2 and p53 may be the primary driving force.
Predictors and biomarkers of therapeutic response Predictors of response to MDM2 inhibitors have been studied using Nutlin-3 and MI-
219 where it was shown that the major determinant of MDM2 inhibitor antitumor activity is the WT p53 status178. Nutlin-3a and MI-219 target the MDM2-p53 interaction with high specificity and as a result they elicit promising anti-tumorigenic responses in cancer cell lines in which MDM2 overexpression remains the dominant regulator inhibiting p53173. MDM2 overexpression may be due to several different mechanisms, including amplification, increased transcription and translation, and single-nucleotide polymorphism (SNP) of the
MDM2 gene204. This recently identified SNP is a T to G mutation at nucleotide 309 (SNP309); importantly, this mutation has been shown to correlate with poor overall survival as well as therapeutic resistance163. Interestingly, it has been suggested that the cohort of patients bearing this molecular signature may benefit from the use of MDM2 antagonists205.
There is inconclusive evidence regarding the relevance of MDM2 amplification as a benchmark biomarker of therapeutic response196,206–208. In a recent study, Pishas et al. reported that neither amplification nor expression correlated with apoptotic response following Nutlin-
3a as shown in a broad screening of sarcomas, perhaps due to the hypermethylation of
GADD45A as observed in cell lines that did not respond to treatment209. Nevertheless, they did observe effective cytostatic results in all cell lines and therefore might have possibly seen stronger apoptotic responses in these cells if the time utilized to assess apoptosis was longer
141 than 24 hours209. In light of this possibility, throughout our studies we evaluated the toxicity of SAR405838 in our panel of cells as based on the relative MDM2 amplification status assigned (MDM2HI and MDM2LO). Our results showed that SAR405838 successfully stabilized p53 and activated downstream transcriptional targets, effectively inhibited cell proliferation, induced cell-cycle arrest as well as induced apoptosis, all in a dose-dependent manner. MDM2HI cells were able to show these responses more consistently and at lower concentrations than MDM2LO cells.
Nonetheless, there is contradictory evidence regarding the clinical relevance of
MDM2 amplification which may possibly be due to several confounding factors. One possibility is that the particular antibody used to detect the MDM2 might only assess the total
MDM2 protein level and not take into account mutant or alternatively spliced MDM2.
However, the question still remains whether or not MDM2 amplification correlates with therapeutic outcome and if it can serve as a potential biomarker for patients receiving MDM2 antagonist therapies.
So the question remains, which processes controlled by p53 are critical for tumor suppression? To answer this question, we performed multiple high-throughput experiments to analyze the therapeutic responses evoked by p53 reactivation. The purpose of these high throughput screens was to identify molecular aberrations induced by inhibition of the MDM2- p53 complex that can either serve as biomarkers for therapeutic response, or can be potentially targeted in a combinational drug setting. We identified expression profiles associated with p53 pathway signaling, cell cycle deregulation, and apoptosis; however, it is the interpretation of these expression patterns and their associated functional deregulation that remains a
142 challenge. Because gene expression analyses are more objective and quantitative than cellular- based assays, their results can possibly be more directly translated to clinical use. For example, the first critically utilized gene expression array, performed only a few years ago, was used to predict the recurrence of tamoxifen-treated, node negative breast cancer210.
Studies like this may be even more crucial for rare cancers such as DDLPS current therapeutic options are much more limited.
SAR405838 is an MDM2 inhibitor whose antitumor effects depend on the activation of p53. As a result, the biomarkers of response of this compound must be specific indicators of p53 activation, as well as demonstrate specificity for the tumor and not a component of the surrounding normal tissue. This criterion can be difficult to establish as p53 has numerous downstream transcriptional targets. Therefore, we utilized a multi-microarray approach to analyze candidate target genes with hopes of identifying the most therapeutically and biologically relevant markers of response. By crossing the significantly expressed (both significantly up- or down-regulated) genes in both of our gene expression microarrays, we identified which signaling pathways were significantly enriched to ensure that only the genes detected in the representing pathways were significantly up or down regulated in both arrays.
Expression profiles showed enrichment in the canonical pathways that were involved in cell cycle regulation, DNA damage response, and apoptosis. As expected, p53 pathway signaling was the most up-regulated following treatment. Expression levels of the genes located in this pathway correlated mostly with cell cycle control and apoptosis. The genomic instability observed provides promising results from this MDM2 inhibitor.
143 From our data we identified growth differentiation factor-15 (GDF-15) to be the most statistically significant expressed gene. GDF-15 is a distant member of the transforming growth factor-β (TGF-β) family, regulates tissue differentiation and maintenance and is expressed at high levels in the placenta, prostate and skin, but at very low levels in other tissues211. Basal expression levels of GDF-15 are generally low until a stress stimuli induces its expression212 and functions as a growth factor most commonly in the P13K/AKT signaling pathway. This cytokine is induced during cellular stress, and high expression is associated with cancer progression213. Furthermore, studies have demonstrated that GDF-expression can be induced by p53 activation-mediated cell cycle arrest and apoptosis213. As a result, this cytokine could plausibly serve as a secreted biomarker of SAR450838 activation in DDLPS.
Identification of novel biomarkers, such as GDF-15, can progress our understanding of molecular underpinnings driving liposarcomagenesis, and provide insight into new therapeutic targets. Identifying genes that are altered following treatment with SAR405838 provides valuable information regarding the effects of this specific inhibitor in DDLPS. Our data supports further investigation of the clinical applicability of GDF-15 as a biomarker for therapeutic response.
Therapeutic resistance It is important to understand how apoptosis is induced in cells as a therapeutic response; it is equally important to understand how a tumor cell might become resistant to apoptotic stimuli. Perhaps the most common mechanism of therapeutic resistance is inactivation of the p53 pathway. Small molecule inhibitors of the MDM2-p53 interaction are based almost entirely on the anti-tumoral responses induced by p53. Consequently, therapeutically resistant clones may emerge from p53 mutant cells in a tumor or from an
144 acquired p53 mutation thereby rendering such therapeutic strategies as useless. It has recently been observed that repeated exposure of Nutlin-3a in SJSA-1 cells leads to somatic mutations in p53214, rendering these cells as no longer responsive to Nutlin treatment. This important study provided the initial demonstration of in vitro drug resistance to Nutlin-3a. One possibility to overcome such resistance would be to combine MDM2 antagonists with a drug that is capable of targeting p53 mutated cells.
Other mechanisms of resistance have been documented, including therapy-induced autophagy215. The concept of autophagic cell death remains controversial. Nevertheless, this dynamic process is fundamental in maintaining cellular homeostasis through degradation and recycling of cellular components via the lysosomal system198,216.
Identified by Crighton and colleagues (2006), DRAM1 is a direct transcriptional target of p53217. DRAM1 is located in the 12q13~15 chromosomal loci217. It has been demonstrated that DRAM1 is induced by genotoxic stress (DOX; 24h) in a p53-dependent manner in cancer cells217. Importantly, a role for DRAM1 in p53-mediated cell death has been suggested as siRNA knockdown of DRAM1 decreased the amount of cell death observed. It has been reported that DRAM1 is down-regulated in tumor cells, and although not capable of inducing apoptosis alone, it is a key factor in p53-dependent apoptosis. Furthermore, the relationship between DRAM1 expression and p53 corresponds to a WT p53 status217.
The increase of DRAM1 observed in our gene expression arrays following MDM2 inhibition is an interesting observation for several reasons. First, our data agrees with the observation that DRAM1 and WT p53 demonstrate a mutually exclusive relationship, supporting an additional form of p53-mediated cell control. Second, the location of DRAM1 is within the 12q13~15 chromosomal region of DDLPS. The discontinuous amplification of
145 this chromosomal loci suggests plausible evidence as to why certain genes are not amplified.
Thirdly, although the role of autophagy in cell death or survival remains controversial, this gene offers an interesting insight into DDLPS biology. The increase in DRAM1 (FC of 3.46) following SAR405838 was observed in our gene expression arrays. This amplification status of could possibly be working in the context of a cell survival mechanism of autophagy, or specifically, ‘productive autophagy’ as certain cell lines did not induce apoptosis as potently following SAR405838 treatment218. However, the addition of a stressor such as SAR405838 activates p53 and might thereby induce a switch from productive autophagy to autophagic cell death. While this hypothesis has yet to be investigated, it provides an interesting potential therapeutic target. If DRAM1 is increased in certain cell lines, targeting DRAM1 could potentially be capable of transitioning a cell from survival to apoptosis, especially if used in combination with an MDM2 inhibitor.
Enhancing the therapeutic efficacy of MDM2 inhibitors with a combination setting The hope of developing a singular “magic bullet” for cancer treatments is pervasive even if unrealistic. At one time it was proposed that Nutlin-3 would be an efficacious as a single-agent therapy due to the possibility of complete reactivation of p53 and its tumor suppression function208. Likewise, SAR405838 exhibits many highly desirable properties of a small-molecule inhibitor. However, the utility of SAR405838 as a single agent, while hopeful, is probably not realistic in the clinical setting. For example, we observed variable anti-tumor responses in our large tumor burden xenograft study when SAR405838 was used as a single agent, suggesting the possibility that SAR405838 combination regimens might offer enhanced results. Frequently, anticancer drugs are used in combination to produce more potent antitumor effects while sparing normal cells. MDM2 inhibitors, such as SAR405838,
146 are non-genotoxic drugs that result in the activation of p53 without causing normal cell damage, unlike many traditional genotoxic agents.
DOX is the current first-line chemotherapeutic agent in DDLPS therapy and is
22 included in most DDLPS treatment regimens . DOX activates p53, inducing G2/M cell cycle arrest and apoptosis219. In other cancer types, it has been demonstrated that the combination of DOX and Nutlin-3a results in synergistic apoptotic effects in wild-type p53 tumor cells
220 while sparing normal cells . We observed that SAR405838 arrested the cell cycle in G1-S and G2-M phases and significantly reduced the S-phase fraction of cells. Perhaps the use of
S-phase- and M-phase-targeted drugs (taxanes and gemcitabine, respectively) may provide protection to normal cells when used in combination and/or as a pre-treatment with
SAR405838221,222. Combination therapies generally involve either synergistic effects on a single pathway or target multiple pathways simultaneously; the combination of SAR405838 with traditional agents may reduce chemotoxicity or is possible that SAR405838 could be combined with standard chemotherapeutic agents to induce apoptosis in cancer cells, thereby creating a novel therapeutic option for DDLPS treatment.
Other deregulations contributing to dedifferentiated liposarcomagenesis While there are many deregulations in dedifferentiated liposarcoma that have yet to be investigated, we sought to identify therapeutically advantageous and targetable abnormalities that could positively impact patient outcome. One such study was the identification of the overexpression of the RTK c-MET. Identification of unique deregulations underlying the pathobiology and tumorigenesis has been useful in multiple sarcoma histologies.
Gastrointestinal stromal tumor (GIST) provides a useful example of identifying biologically
147 relevant and targetable genetic mutations and applying this knowledge in a clinically useful manner. In 1998, Hirota et al. identified the gain-of-function mutation of the KIT gene, which in turn encodes the proto-oncogenic receptor tyrosine kinas (RTK)223. This discovery led to the initial awareness of GIST as an entity separate from leiomyosarcoma while simultaneously providing a biologically relevant and targetable therapeutic option. Imatinib mesylate, an RTK inhibitor, has shown significant utility as a single-agent therapy, in combination with surgery for control of local and metastatic GIST224,225. The promising use of RTK inhibitors have gained much attention in the past decade76,226. Therefore, following identification of c-MET amplification in a previous study75 in our lab, we have investigated how inhibition of this receptor might prove to be therapeutically relevant (Appendix D). While only one example, it represents one small step toward developing novel therapeutic strategies for patients suffering from DDLPS.
148 CHAPTER 7. SUMMARY
This body of work encompassed multiple studies investigating dedifferentiated liposarcoma including molecular and cytogenetic aberrations in DDLPS tumor cells and the pre-clinical effects of a novel small-molecule inhibitor, SAR405838. In doing so, the following results were found:
1. DDLPS is a rare disease with limited experimental resources. The establishment of
cell lines and xenograft models was a critical first step in studying this disease.
Importantly, these models accurately represent the tumorigenic phenotype and
molecular alterations observed in patients.
2. DDLPS accumulates multiple cytogenetic aberrations; notably, amplification of the
12q13~15 chromosomal region.
3. DDLPS cell lines express genetic profiles associated with disease progression.
4. MDM2 amplification occurred in 100% of all DDLPS cell lines, albeit, to varying
levels; this genomic amplification demonstrated a linear mRNA and protein
expression. Furthermore, our results suggest that the amplification of MDM2 provides
possible therapeutic relevance as a biomarker for therapeutic efficacy.
5. DDLPS retains a functional and wild-type p53 genomic status which can be
reactivated, as proven by the re-activation of the p53 pathway following MDM2
inhibition.
6. MDM2 antagonists successfully activate the p53 pathway; SAR405838 possessed the
most effective results in comparison to Nutlin-3a and MI-219.
7. SAR405838 elicits a dose-dependent anti-DDLPS tumor growth response in vivo.
149 8. The modulated gene expressions following SAR405838 enriched p53 pathways
involved in genomic stability, cell-cycle arrest and apoptosis.
Taken together, our data highlights the significant contribution of the deregulated
MDM2:p53 axis in DDLPS tumorigenesis. We have shown that targeting MDM2 with the small molecule inhibitor SAR405838 activates the p53 pathway and elicits anti-DDLPS effects in vitro and in vivo. We have investigated pathway enrichment following MDM2 inhibition and suggested possible targetable therapeutic nodes as well as known biomarkers associated with p53 activation. Lastly, our data demonstrates that MDM2 in a critical manipulator in DDLPS tumorigenesis and targeting MDM2 is a highly effective therapeutic option for patients suffering from this disease.
150 CHAPTER 8. FUTURE DIRECTIONS
In this body of work we have shown the therapeutic efficacy of a novel small-molecule
MDM2 inhibitor for the treatment of DDLPS. While there are many deregulations in DDLPS that have yet to be investigated, we aimed at targeting the most therapeutically advantageous abnormality, MDM2. MDM2 is genomically amplified and highly expressed in DDLPS, establishing the biological relevance of MDM2 as a therapeutic target.
Following MDM2 inhibition, DDLPS cell lines exhibited variable p53- dependent responses, notably the level of apoptosis. Cell lines that elicited only minimal apoptosis may allude to a defective apoptotic pathway in these cells as a means of survival.
Cell lines that demonstrated a higher degree of MDM2 amplification responded with stronger apoptotic responses, and provide plausible evidence that MDM2 amplification in DDLPS is not just an initial tumorigenic driver, but continues to play an important role in cell survival.
Further analysis of the SAR405838-induced gene expression profiles may reveal differential patterns in apoptosis between the cells lines, as well as elucidate potential defective or hyperactive molecules that are implicated in apoptosis. Although only a very small panel of cells were used to perform this experiment, this can be expanded to other STS histologies that do not respond to p53-activating therapies as MDM2 amplification is a common aberration in many different tumor types. Additionally, the unique therapeutic responses to SAR405838 in our DDLPS model emphasizes the need to further dissect this distinction based on the level of MDM2 amplification. Our data implies that MDM2 amplification positively correlates with the efficacy of MDM2 inhibition. Future studies determining the level of MDM2 amplification in patient biopsies prior to treatment will possibly distinguish a correlation between MDM2 amplification and therapeutic response.
151 In addition, two very different responses were observed from the large tumor burden therapeutic experiment. Previously we observed that SAR405838 successfully penetrated the tumor in Lipo863 tumor-bearing mice and initiated p53 responses as demonstrated in our pharmacokinetics/pharmacodynamics studies. However, Lipo863 tumors did not respond to
SAR405838 when treatment began with large initial tumor burdens. This may be due to a defective apoptotic pathway in these cells in which an alternative pathway bears more influence than p53 does. These cells may have also acquired therapeutic resistance towards
MDM2 inhibition. Additional therapeutic experiments repeating this large tumor burden model with DDLPS cell lines that also demonstrate weak apoptotic responses should be performed. This knowledge can potentially demonstrate additional MDM2-p53 independent pathways that are dysregulated. Therapeutic resistance due to p53 mutations acquired from certain clones in a tumor population should also be investigated and mutational analysis should be performed to determine p53 status. Gene expression arrays on these tumors can further unravel pathways and processes that are enriched from such oncogenic genes.
Additionally, combination of SAR405838 with traditional therapies, such as with the topoisomerase-2 inhibitor DOX, should result in a stronger anti-tumorigenic response and may be more effective in tumors that do not respond with single-agent treatment and should be evaluated.
Consequences from p53 pathway reactivation were examined, mainly cell cycle arrest and apoptosis; however, there are many other outcomes that should be explored. Senescence is another result of p53 pathway activation and may be an alternative outcome of MDM2 inhibition. In DDLPS cells it was observed that the most enriched pathways from the gene expression arrays involved different cell cycle processes and spindle and mitotic instability.
152 Detection of SA-beta-Gal following SAR405838 in cells will determine if the tumor suppressive mechanism is senescence.
There are multiple regulators of MDM2 that were not evaluated in our studies that may impact this oncogene in liposarcomagenesis. For example, MDMX, a close homolog of
MDM2, is another negative regulator of p53. MDMX is capable of modulating the levels of
MDM2, thus, regulating the level of p53 protein and function. Nevertheless, this is just one of many alternative regulators of MDM2 function that might influence the effectiveness of
MDM2 inhibitors.
In order for p53 to be degraded, the nuclear export signal (NES) must export MDM2 and p53 to the cytosol where MDM2 will inhibit p53 function in at least two ways, by binding to the transactivation domain of p53 and inhibiting its transcriptional properties or targeting p53 for degradation by its E3 ubiquitination properties. Further investigation into the ability of MDM2 to bind to the transactivational domain of p53 should be confirmed through in vitro experiments involving the fusion of MDM2 to the DNA-binding domain of p53 to that the region is leading to inhibition of p53. Alternatively, MDM2 ubiquitination function may not be fully functional or sufficient to degrade p53 in DDLPS; therefore, the detection of p53 protein ubiquitination in DDLPS cells should be investigated. The high expression of p53 observed in DDLPS cells might also be due to the MDM2-p53 complex being held together.
Performing a pulse-chase experiment to track the localization of these proteins and expression levels. If MDM2 is bound to p53, levels will remain high as p53 will continue to increase
MDM2 levels, but neither will be degraded.
Within these studies multiple molecular aberrations in DDLPS were identified that warrant further evaluation. As a result, these proteins may reveal potential prognosticators,
153 biomarkers and therapeutic targets. For example, growth differentiation factor 15 (GDF-15) is a p53 transcriptional target and was the most up-regulated gene following SAR405838 treatment in the microarrays. GDF-15 protein is a cytokine that can be easily evaluated in patient blood samples. Translationally, this can not only reveal p53 activation in patient tumors, but potentially can be used to caliber the minimum dose of SAR405838 required for efficacy.
Hopefully in the very near future, the combined knowledgebase of specific histological characteristics and the unique genetic and molecular aberrations will lay the foundation clinical decisions. It is without any doubt that as we continue to increase our understandings of the molecular underpinnings of DDLPS, development of enhanced therapeutic strategies will certainly follow. The potential studies mentioned can further unveil the significant impact of the MDM2-p53 axis in DDLPS preclinical models and lead to more personalized, therapeutically relevant strategies and future DDLPS preclinical trials.
154 APPENDIX
APPENDIX A. ACGH DATA FOR DEDIFFERENTIATED LIPOSARCOMA CELL LINES. Symbol Cytoband p-values FDR Tspan31 15q24.3 0.0001475 1.4E-06 Cdk4 12q14.1 1.637E-10 1.4E-06 March9 9q33.1 2.328E-10 1.4E-06 LOC100128056 7q34 1.512E-09 5.4E-06 MIR26A2 13q31.3 1.375E-09 5.4E-06 CYP27B1 12q14.1 2.381E-09 7.1E-06 MDM2 12q15 9.858E-09 2.3E-05 slc35e3 12q15 1.012E-08 2.3E-05 NUP107 12q15 2.218E-06 0.00332 BEST3 12q15 0.0005657 0.12971 LRRC10 12q15 0.0005657 0.12971 CCT2 12q15 0.0004703 0.12971 RAB3IP 12q15 0.000279 0.09584 FRS2 12q15 0.0001338 0.09584 CNOT2 12q15 0.0007495 0.12971 KCNMB4 12q15 0.0006077 0.12971 CPM 12q15 6.149E-08 0.00011 Agap2 12q14.1 2.258E-08 4.5E-05 OS9 12q14.1 1.749E-05 0.01846 Yeats4 12q15 0.0013939 0.12971 LYZ 12q15 0.0013939 0.12971 Cpsf6 12q15 0.0015365 0.12971 ELK3 12q23.1 0.0019154 0.12971 C12orf28 12q24.31 0.0019378 0.12971 B4GALNT1 12q14.1 0.0013466 0.12971 GEFT 12q13.3 0.0013466 0.12971 TSFM 12q14.1 1.659E-05 0.01846 FAM119B 12q14.1 1.659E-05 0.01846 mettl1 Xp21.3 1.659E-05 0.01846 AVIL 12q14.1 1.659E-05 0.01846 xrcc6bp1 12q14.1 0.0013009 0.12971 MYF5 12q21.31 0.0310806 0.19857 MYF6 12q21.31 0.0310806 0.19857 ptprq 12q21.31 0.0296756 0.19658 Lin7a 12q21.31 0.0397626 0.19981 LOC552889 12q21.1 0.0461876 0.20616
155 TBC1D15 12q21.1 0.0188899 0.16909 rab21 12q21.1 0.0267993 0.19119 TPH2 12q21.1 0.0308744 0.19857 ACSS3 12q21.31 0.0131279 0.1609 alx1 12q21.31 0.0147234 0.16432 HMGA2 12q14.3 8.42E-08 0.00014 TRBV4-3 3q13.13 2.997E-05 0.0283 LOC100129940 19q13.2 0.0005047 0.12971 dyrk2 12q15 0.0046173 0.12971 TRHDE 12q21.1 0.002387 0.12971 Rap1b 12q15 0.065157 0.23022 CEP72 5p15.33 0.0117281 0.16065 EXOC3 5p15.33 0.0117281 0.16065 C5orf55 5p15.33 0.0117281 0.16065 LOC25845 5p15.33 0.0117281 0.16065 SLC9A3 5p15.33 0.0117281 0.16065 NKD2 5p15.33 0.0177186 0.16565 Trip13 5p15.33 0.0177186 0.16565 BRD9 5p15.33 0.0177186 0.16565 SLC12A7 5p15.33 0.0177186 0.16565 ZDHHC11 5p15.33 0.0148858 0.16432 TERT 5p15.33 0.0078586 0.16065 SLC6A19 5p15.33 0.0089416 0.16065 SLC6A18 5p15.33 0.006575 0.14813 Syt1 12q21.2 0.0226519 0.1845 PPFIA2 12q21.31 0.028728 0.19526 TRBV6-5 3q13.13 0.0010668 0.12971 TRBV6-1 3q13.13 0.0010668 0.12971 Trbv28 3q13.13 0.000136 0.09584 CFHR3 1q31.3 2.01E-05 0.02004 Fam19a2 12q14.1 0.0233896 0.18739 Tmbim4 12q14.3 0.2408784 0.44891 CDCA7L 7p15.3 0.2736632 0.49554 SNTG1 8q11.21 0.404485 0.61203 GBP3 1p22.2 0.3998699 0.60692 ADAM3A 8p11.23 0.2956098 0.52211 SNRPF 12q22 0.0226315 0.1845 LTA4H 12q23.1 0.0227504 0.18488 CCDC38 12q22 0.0223633 0.1845 amdhd1 12q23.1 0.0240796 0.18739 hal 12q23.1 0.027481 0.19231 PIP4K2C 12q13.3 0.0091986 0.16065 DTX3 12q13.3 0.0280448 0.19421 Cdk17 16q24.3 0.0038873 0.12971 C12orf55 12q24.13 0.0557562 0.21983
156 TMTC3 12q21.32 0.0457105 0.20484 PTPRB 12q15 0.0081241 0.16065 ptprr 12q15 0.0061592 0.14716 Kif5a 12q13.3 0.0193486 0.16962 Tspan8 12q15 0.009377 0.16065 RASSF9 12q21.31 0.0451902 0.20332 MGAT4C 12q21.31 0.0674155 0.23022 ITGA5 12q13.13 0.0250062 0.18739 TRBV24-1 3q13.13 0.7151719 0.86551 WWOX 16q23.1 0.6041573 0.77455 IGHV3-11 11q13.2 0.0156944 0.16432 gstm1 1p13.3 0.5434609 0.7201
157 APPENDIX B: DIFFERENTIAL GENE EXPRESSION PATTERNS IN DEDIFFERENTIATED LIPOSARCOMA
Name Lipo224 Lipo224B Lipo246 Lipo863 LPS141 Lipo573 Lipo815 Lipo615 Lipo984 Control diff.group foldchange p.value t.stat NUP107 12.126 13.515 12.673 10.758 14.357 13.689 13.077 12.290 12.496 8.781 3.995 15.939 9.39E-08 9.6302 CDC20 11.690 12.413 11.964 12.471 11.896 11.087 12.165 10.047 11.187 7.988 3.670 12.729 6.27E-07 8.282677 METTL1 12.129 12.022 12.354 11.761 8.048 12.434 11.377 11.591 11.096 7.924 3.499 11.308 7.99E-06 6.678477 TOP2A 10.135 11.751 11.490 12.506 11.502 11.376 12.402 11.680 11.449 8.162 3.426 10.746 2.74E-06 7.328429 TRIP13 13.341 10.578 12.779 12.211 9.890 10.412 10.956 9.808 10.105 7.763 3.357 10.248 4.34E-05 5.716167 UBE2C 10.763 11.014 11.400 10.815 10.219 10.301 10.981 10.275 10.087 7.414 3.236 9.423 4.01E-07 8.586087 ANLN 9.705 11.040 10.720 10.521 12.126 10.335 10.885 9.612 10.185 7.352 3.218 9.307 1.25E-06 7.827341 CDCA5 11.327 10.979 10.770 11.306 11.217 10.829 11.426 10.561 10.249 7.749 3.214 9.278 1.29E-08 11.20808 AURKB 9.760 10.280 10.414 9.965 10.006 9.985 9.939 9.158 9.282 6.705 3.161 8.944 5.82E-09 11.89527 CDC45L 10.497 10.625 10.995 9.657 10.414 10.232 10.356 8.773 9.260 7.039 3.051 8.285 7.31E-07 8.178961 CDK4 14.275 14.055 14.090 14.306 14.195 14.183 14.252 14.256 14.306 11.208 3.005 8.030 1.97E-14 29.00825 TSFM 10.975 10.234 11.950 10.304 7.019 11.486 10.583 10.780 8.571 7.256 2.955 7.756 0.000176 4.968256 UHRF1 10.816 10.945 10.347 10.687 10.859 10.440 10.340 9.427 9.334 7.430 2.925 7.594 3.73E-07 8.637617 PBK 9.759 10.822 10.060 10.999 8.996 9.696 10.429 9.371 9.581 7.063 2.905 7.492 4.09E-07 8.573916 SLC35E3 11.186 12.508 12.203 12.183 13.790 13.016 12.085 12.363 12.567 9.531 2.903 7.479 4.56E-08 10.18325 PRC1 11.248 11.556 11.365 11.460 11.389 11.240 11.491 10.945 11.633 8.479 2.891 7.417 5.05E-07 8.42825 DUSP1 10.848 12.112 11.567 11.108 11.257 12.730 11.464 11.631 11.467 8.730 2.846 7.191 0.013816 2.792843 RAD51AP1 10.346 10.527 10.900 10.342 9.831 10.064 10.111 8.973 11.153 7.405 2.845 7.183 2.74E-07 8.852177 CKAP2L 9.524 10.587 10.389 10.267 9.916 9.643 9.728 8.651 9.000 6.919 2.826 7.090 2.86E-07 8.821365 NCAPG 10.047 10.972 10.555 11.230 10.909 10.625 10.391 9.623 10.353 7.739 2.784 6.889 1.65E-06 7.648092 PTTG3P 11.209 11.950 11.277 11.392 9.832 10.674 11.623 10.637 10.873 8.270 2.782 6.877 6.77E-06 6.776799 TK1 9.618 10.418 9.899 10.102 10.294 9.511 10.999 8.950 9.715 7.191 2.754 6.748 4.22E-07 8.551707 B4GALNT1 10.244 9.400 8.458 9.452 7.513 7.363 10.460 8.559 10.844 6.396 2.747 6.715 6.08E-05 5.533025 AURKA 10.717 10.903 11.004 11.214 10.381 10.207 10.640 10.262 10.011 7.849 2.744 6.699 2.17E-07 9.016679 UBE2T 10.765 10.453 11.590 10.311 9.558 10.825 10.730 11.048 9.821 7.838 2.728 6.626 6.67E-07 8.240765
158 FAM119B 11.717 11.325 12.265 10.879 7.395 11.817 10.001 12.124 10.635 8.191 2.716 6.570 0.000348 4.618887 PTTG1 11.133 12.010 11.313 11.720 10.325 11.136 11.880 11.009 11.064 8.602 2.686 6.434 4.29E-06 7.051833 TYMS 11.390 12.039 11.591 11.067 12.106 11.971 12.341 10.456 11.233 8.904 2.673 6.376 2.85E-05 5.948184 CCNB2 10.028 10.883 9.897 10.499 9.981 9.859 10.411 9.573 9.634 7.428 2.657 6.307 5.69E-07 8.348144 BUB1 9.288 10.282 10.444 10.670 9.574 9.048 9.559 8.628 8.973 6.953 2.655 6.297 1.16E-06 7.877226 CDCA3 9.114 10.198 10.199 9.656 8.424 9.170 10.337 9.500 9.083 6.868 2.652 6.284 3.52E-07 8.676438 CDKN3 9.649 10.637 10.764 9.875 9.781 9.286 11.023 9.593 9.781 7.395 2.648 6.267 1.05E-06 7.938021 KIF20A 9.296 10.716 9.491 10.580 9.898 9.054 10.357 9.730 9.702 7.242 2.627 6.178 6.28E-07 8.280724 ASF1B 8.968 9.430 9.576 8.409 9.284 8.759 9.892 9.124 8.474 6.489 2.613 6.117 5.59E-09 11.93064 CDC2 9.571 10.385 9.726 10.301 9.110 9.845 9.525 9.164 8.980 7.017 2.606 6.087 9.85E-08 9.594813 CD24 10.909 11.923 8.972 6.604 7.068 11.832 11.941 9.744 7.384 6.995 2.603 6.074 0.009647 2.971166 NUSAP1 9.496 10.476 9.924 9.680 10.410 10.014 10.139 9.632 9.789 7.356 2.595 6.041 6.91E-08 9.861983 CEP55 9.918 10.236 10.393 10.325 9.305 10.132 11.002 9.707 10.333 7.557 2.593 6.035 5.99E-06 6.849507 FEN1 11.067 10.360 10.222 11.605 10.946 10.612 10.993 10.130 9.990 8.084 2.574 5.956 3.86E-07 8.613625 FAM83D 9.231 9.851 10.094 10.378 9.453 9.102 9.280 8.983 8.853 6.898 2.572 5.945 5.12E-08 10.09362 CD70 10.802 9.808 8.639 9.944 11.574 6.417 8.184 8.423 8.838 6.636 2.545 5.835 0.000806 4.195397 CCT2 14.357 14.275 14.131 14.357 14.446 11.571 14.146 11.003 10.804 10.694 2.538 5.807 0.001936 3.760962 KIFC1 8.885 9.873 9.246 9.517 9.113 9.536 9.895 8.711 8.752 6.761 2.520 5.734 5.22E-08 10.07862 TTK 9.475 10.018 9.825 9.727 9.201 9.052 10.014 8.918 9.366 7.011 2.500 5.656 3.38E-08 10.41813 ASPM 9.896 10.441 10.426 11.274 10.804 10.112 10.191 9.311 10.834 7.870 2.496 5.641 9.8E-06 6.55864 TPX2 9.272 10.150 9.820 10.989 9.987 9.411 10.378 9.559 9.404 7.393 2.493 5.630 5.01E-07 8.433756 DLGAP5 8.839 9.783 9.898 9.996 9.953 9.218 10.002 8.736 9.336 7.043 2.486 5.601 6.72E-07 8.235632 CDK2 9.449 9.528 9.944 9.280 8.220 9.345 12.448 8.512 12.012 7.400 2.459 5.500 0.000632 4.317029 LOC390557 10.279 9.105 9.992 11.228 9.977 7.453 8.321 9.652 8.152 6.897 2.454 5.480 0.000108 5.227146 NTF3 7.273 9.708 11.133 9.335 8.683 8.200 10.079 7.424 8.332 6.470 2.437 5.417 0.00022 4.853493 PLAU 11.832 11.618 13.910 12.810 12.545 11.613 11.720 9.143 12.278 9.534 2.407 5.303 0.001328 3.946847 MELK 10.348 11.416 10.603 10.325 10.990 10.543 10.148 9.696 10.233 8.078 2.400 5.279 2.22E-06 7.459821 CCNB1 9.562 10.248 9.714 10.591 9.236 8.868 9.643 9.362 9.424 7.229 2.399 5.274 2.16E-07 9.021686
159 MCM7 10.545 10.723 11.168 10.758 10.430 10.689 10.477 9.576 9.902 8.088 2.387 5.230 1.17E-07 9.464971 BIRC5 9.085 9.921 8.482 9.617 9.004 8.598 9.502 8.746 8.742 6.709 2.368 5.162 1.93E-07 9.10363 CCNA2 9.796 10.499 10.357 9.921 9.991 10.119 10.010 9.752 9.714 7.663 2.354 5.113 9.27E-08 9.640476 MCM4 9.356 9.976 9.989 10.580 11.142 10.282 10.161 8.869 9.419 7.635 2.340 5.062 4.13E-06 7.074286 HJURP 9.029 9.383 9.170 9.310 9.649 8.996 9.151 8.947 9.598 6.924 2.324 5.007 1.28E-08 11.21253 PRIM1 9.879 9.717 9.997 10.332 9.003 9.637 9.930 8.181 8.547 7.157 2.313 4.968 1.4E-06 7.754261 GINS2 9.711 10.684 9.737 8.947 9.117 10.110 9.642 9.136 9.783 7.356 2.296 4.911 1.65E-06 7.646397 CDCA8 9.251 9.502 9.610 9.841 9.147 8.944 9.695 7.991 8.694 6.893 2.293 4.902 3.58E-07 8.664526 CKS1B 11.529 11.883 11.612 11.956 10.445 11.166 11.452 11.436 10.606 9.051 2.292 4.897 2.55E-07 8.903431 KIF11 9.228 9.420 9.503 10.014 8.708 9.592 9.633 8.990 9.163 7.082 2.279 4.853 7.25E-08 9.825896 YEATS4 10.886 12.632 13.573 12.350 8.350 9.610 11.849 8.800 8.445 8.446 2.276 4.842 0.010528 2.927912 FBXO5 9.676 10.082 9.858 9.058 9.233 9.623 9.559 9.009 9.987 7.295 2.271 4.825 9.16E-08 9.648927 TACC3 9.510 9.649 9.028 9.802 9.028 9.176 9.507 9.149 9.219 7.072 2.269 4.819 3.04E-08 10.50249 XRCC6BP1 11.064 10.934 12.278 9.089 8.209 13.270 11.907 8.281 7.746 8.061 2.248 4.750 0.013008 2.822851 MAD2L1 9.938 10.186 9.776 9.708 9.606 10.124 9.803 9.183 8.853 7.450 2.237 4.713 1.64E-06 7.653115 MCM10 9.192 9.368 10.715 9.386 8.617 8.319 8.233 7.249 7.852 6.544 2.226 4.677 7.8E-05 5.398945 OIP5 9.145 9.333 9.299 9.018 9.313 9.366 9.453 8.402 8.916 6.925 2.214 4.639 3.99E-09 12.23267 ATAD2 9.664 9.541 9.782 9.873 11.116 9.478 8.979 8.676 8.910 7.345 2.213 4.635 6.58E-06 6.793873 RRM2 8.379 9.491 10.238 8.007 9.588 9.655 9.168 6.842 8.026 6.609 2.212 4.634 0.000152 5.044431 KIF2C 8.997 9.451 9.483 9.989 9.594 8.495 9.261 7.191 8.348 6.772 2.207 4.616 1.62E-05 6.266329 HMMR 9.484 10.239 10.606 10.128 10.145 9.392 10.034 9.430 9.485 7.687 2.196 4.581 1.33E-07 9.373581 APOBEC3B 8.492 9.836 9.134 8.122 8.963 8.087 9.450 8.127 7.031 6.389 2.193 4.574 1.31E-05 6.388713 KRT81 6.276 8.103 6.622 13.157 8.599 6.391 11.576 8.661 6.142 6.215 2.177 4.522 0.040208 2.249207 CENPM 8.730 9.190 9.222 8.178 9.034 8.472 9.160 8.332 8.591 6.593 2.175 4.515 5.95E-09 11.87552 E2F2 9.721 10.042 8.387 7.975 9.736 7.843 8.766 7.499 8.112 6.513 2.162 4.477 3.44E-05 5.84425 MCM5 10.536 10.175 9.881 8.971 9.411 9.369 9.779 8.611 9.174 7.396 2.150 4.437 2.71E-06 7.33487 TSPAN31 11.615 10.867 11.216 11.454 13.361 12.503 9.849 11.228 12.211 9.458 2.131 4.381 0.000148 5.061573 TSPAN8 10.735 8.877 6.255 10.350 11.565 6.323 10.766 6.770 8.340 6.765 2.122 4.354 0.021288 2.575827
160 CCNE2 7.665 8.957 8.736 8.862 9.922 8.821 8.649 7.738 7.398 6.411 2.117 4.337 9.98E-06 6.547631 NUF2 7.666 8.161 8.160 8.610 11.517 7.563 8.480 7.634 8.968 6.428 2.101 4.291 0.000577 4.362385 TROAP 8.869 8.868 8.462 8.839 8.243 8.409 9.322 9.043 8.320 6.623 2.085 4.244 1.38E-08 11.14824 CPA4 9.217 10.411 6.424 9.497 9.196 8.125 10.756 9.316 8.444 6.960 2.083 4.236 0.001746 3.811787 LOC399942 11.937 11.847 11.832 10.848 9.882 11.332 12.284 11.243 10.472 9.216 2.081 4.232 0.000736 4.240247 CENPF 8.792 9.110 8.646 10.281 9.495 8.573 9.339 8.452 9.472 7.058 2.071 4.203 1.72E-06 7.619297 LOC651816 10.787 11.874 11.294 10.943 9.718 10.894 11.703 10.771 10.460 8.873 2.065 4.185 6.94E-06 6.761898 DTL 9.370 9.532 8.181 8.708 8.701 9.337 9.152 7.664 8.429 6.736 2.050 4.140 2.43E-06 7.403018 CENPK 9.130 9.629 9.635 9.459 9.593 9.483 9.748 9.378 8.805 7.381 2.048 4.135 1.21E-06 7.849096 HIST1H4C 12.907 14.060 13.937 13.573 12.775 13.717 14.040 13.215 13.292 11.462 2.040 4.112 2.15E-07 9.025872 CDT1 8.819 9.484 8.604 8.428 8.411 8.574 8.428 7.441 7.971 6.441 2.021 4.059 2.46E-07 8.928118 OS9 10.685 11.258 10.894 10.560 12.670 7.378 10.466 8.006 10.735 8.294 2.001 4.003 0.007074 3.124248 MCM2 9.644 9.367 9.874 9.716 8.537 9.106 9.276 7.944 8.568 7.114 2.001 4.002 4.25E-06 7.056522 LOC731049 11.307 12.378 11.952 11.673 10.583 11.464 12.154 11.227 11.060 9.534 1.999 3.996 3.01E-06 7.269147 TMEM51 9.696 10.049 9.484 10.166 11.301 9.434 10.008 8.938 9.393 7.833 1.997 3.992 4.52E-05 5.693548 CCNF 8.087 10.185 9.511 10.515 9.839 9.273 9.527 8.802 9.014 7.428 1.989 3.969 2.49E-05 6.024545 STMN1 9.124 9.357 9.081 8.567 8.852 8.840 9.254 7.628 8.360 6.799 1.985 3.959 1.19E-06 7.859257 HBEGF 7.992 8.495 9.064 9.869 9.889 10.573 9.374 8.612 8.007 7.119 1.978 3.940 0.00085 4.168348
161 APPENDIX C. TOP 50 DIFFERENTIALLY EXPRESSED GENES IN VITRO AND IN VIVO FOLLOWING SAR405838 TREATMENT. Top up-regulated in vitro in vivo Gene Fold Change Gene Fold Change GDF15 16.2129 PADI4 23.8301 LOC100008589 9.9671 LCE1B 16.2649 C18ORF56 9.7888 MYBPHL 15.0798 CLCA2 9.6741 BTG2 9.7775 MDM2 5.5876 RN5S9 8.6088 MGC42367 5.4459 GADD45A 8.0443 BTG2 5.3475 PHLDA3 7.4409 PRAGMIN 5.2925 CLCA2 7.3535 C7ORF10 5.0681 EFCAB7 6.9302 SESN1 4.9502 FUCA1 6.5052 EFCAB7 4.8842 TXNIP 5.6006 CDH10 4.8820 SLC2A5 5.1532 TP53INP1 4.7794 LCE1C 4.9808 C2ORF55 4.5492 LCE1A 4.8335 AKR1B10 4.3901 HOOK1 4.7078 SPATA18 4.1305 KIAA1324 4.1808 LOC729989 4.1040 SESN2 3.8978 CDKN1A 3.8017 MR1 3.7296 HMOX1 3.5960 C1ORF68 3.6732 GADD45A 3.5772 ATF3 3.6688 RDH10 3.5767 RNPEP 3.5194 SNORD3D 3.3887 SYNC1 3.4818 ACTA2 3.3438 OVGP1 3.3910 RRM2B 3.2410 DHRS3 3.2688 HIST2H2BE 3.1726 C1ORF183 3.2535 Top down-regulated in vitro in vivo Gene Fold Change Gene Fold Change RAB3IP -1.3343 PAQR4 -6.35542 LOC100129362 -1.3519 HSPA8 -6.19031 UBE2N -1.3671 HIST1H4C -6.07239 LSM3 -1.4068 EMG1 -5.29068 C7ORF44 -1.4088 LOC399988 -5.11464
162 GSG2 -1.4129 MCM6 -4.75124 NRM -1.4147 GINS2 -4.54509 WDR67 -1.4180 TMEM126A -4.37410 ORC3L -1.4211 DCTPP1 -4.47037 NONO -1.4223 MRPL17 -4.34799 LRRC8C -1.4232 ARMCX2 -4.32155 APLN -1.4233 TM4SF1 -4.29947 CXORF57 -1.4237 COMMD7 -4.23540 C1ORF93 -1.4241 LOC493869 -4.14157 C6ORF125 -1.4258 USP13 -4.06063 MTF2 -1.4284 METTL1 -4.04629 PTBP1 -1.4298 REEP1 -4.04095 SHPK -1.4351 MCM2 -3.87347 HNRNPA2B1 -1.4377 TWIST1 -3.77542 CDCA7L -1.4381 FAM136A -3.84662 SFRS7 -1.4391 ATIC -3.82031 C1ORF24 -1.4393 LOC552889 -3.81067 POLE -1.4417 HS.213541 -3.82384 CCDC18 -1.4481 MT1X -3.75534 MRPS11 -1.4492 TST -3.73768
163 APPENDIX D. THE HEPATOCYTE GROWTH FACTOR RECEPTOR AS A POTENTIAL THERAPEUTIC TARGET FOR DEDIFFERENTIATED LIPOSARCOMA.
Manuscript briefly mentioned throughout this dissertation, which involved a small project investigating the receptor tyrosine kinase, Met, in dedifferentiated liposarcoma as a therapeutic target. For reference once published, the manuscript is titled: The hepatocyte growth factor receptor as a potential therapeutic target for dedifferentiated liposarcoma.
164 APPENDIX E. SAR405838: A NOVEL AND POTENT INHIBITOR OF THE MDM2:P53 AXIS FOR THE TREATMENT OF DEDIFFERENTIATED LIPOSARCOMA
Manuscript in preparation for submission and involves the data discussed throughout this dissertation. For reference once published, the manuscript is titled: SAR405838: a novel and potent inhibitor of the MDM2:p53 axis for the treatment of dedifferentiated liposarcoma. The following authors were involved in this manuscript:
Kate Lynn Bill, Jeanine Garnett, Isabelle Meaux, XiaoYen Ma, Chad J. Creighton, David
Pollock, Theresa Nguyen, Svetlana Bolshakov, Cedric Barriere, Alexander J. Lazar, Dina Lev,
Raphael E Pollock.
165 VITA
Kate Lynn Jane Bill was born in Elizabeth, New Jersey on January 19, 1987, the daughter of Mary Elizabeth Bill and William Raymond Bill. After completing her work at
Oak Ridge High School, in Conroe, Texas in 2005, she entered Hardin-Simmons University in Abilene, Texas. She received the degree of Bachelor of Science with a major in biology from Hardin-Simmons in May, 2009. In August 2009 she entered The University of Texas
Graduate School of Biomedical Sciences at Houston.
Permanent address:
Kate Lynn Bill
271 Frankfort Square
Columbus, OH 43206
166 REFERENCES
1. Fletcher CDM, Unni KK, Mertens F. World Health Organization Classification of
Tumours Pathology and Genetics of Tumours of Soft Tissue and Bone. Fourth Edi.
(Fletcher CDM, Unni KK, Mertens F, eds.). Lyon, France: IARC Press; 2013.
2. Conyers R, Young S, Thomas DM. Liposarcoma: molecular genetics and therapeutics.
Sarcoma. 2011:483154. doi:10.1155/2011/483154.
3. Evans HL. Liposarcoma a study of 55 cases with a reassessment of its classification.
Am J Surg Pathol. 1979;3(6).
4. Henricks WH, Chu YC, Goldblum JR, Weiss SW. Dedifferentiated Liposarcoma: A
Clinicopathological Analysis of 155 Cases with a Proposal for an Expanded Definition
of Dedifferentiation. Am J Clin Pathol. 1997;21(3):271-281.
5. Evans HL, Khurana KK, Kemp BL, Ayala AG. Heterologous Elements in the
Dedifferentiated Components of Dedifferentiated Liposarcoma. Am J Clin Pathol.
1994;18(11):1077-1182.
6. Pilotti S, Della Torre G, Lavarino C, Di Palma S, Sozzi G, Minoletti F, Rao S, Pasquini
G, Azzarelli A, Rilke F, Pierotti MA. Distinct mdm2/p53 expression patterns in
liposarcoma subgroups: implications for different pathogenetic mechanisms. J Pathol.
1997;181(1):14-24. doi:10.1002/(SICI)1096-9896(199701)181:1<14::AID-
PATH730>3.0.CO;2-O.
7. Horvai AE, DeVries S, Roy R, O’Donnell RJ, Waldman F. Similarity in genetic
alterations between paired well-differentiated and dedifferentiated components of
dedifferentiated liposarcoma. Mod Pathol. 2009;22(11):1477-88.
doi:10.1038/modpathol.2009.119.
167 8. Ware PL, Snow AN, Gvalani M, Pettenati MJ, Qasem S a. MDM2 Copy Numbers in
Well-Differentiated and Dedifferentiated Liposarcoma: Characterizing Progression to
High-Grade Tumors. Am J Clin Pathol. 2014;141(3):334-41.
doi:10.1309/AJCPLYU89XHSNHQO.
9. Crago AM, Socci ND, DeCarolis P, O’Connor R, Taylor BS, Qin L-X, Antonescu CR,
Singer S. Copy number losses define subgroups of dedifferentiated liposarcoma with
poor prognosis and genomic instability. Clin Cancer Res. 2012;18(5):1334-40.
doi:10.1158/1078-0432.CCR-11-2820.
10. Segura-Sánchez J, González-Cámpora R, Pareja-Megia MJ, García-Escudero A,
Galera-Ruiz H, López-Beltrán A. Chromosome-12 copy number alterations and
MDM2, CDK4 and TP53 expression in soft tissue liposarcoma. Anticancer Res.
2006;26(6C):4937-42. Available at: http://www.ncbi.nlm.nih.gov/pubmed/17214366.
11. Pilotti S, Della Torre G, Lavarino C, Minoletti F, Vergani B, Azzarelli A, Rilke F,
Pierotti MA. Molecular abnormalities in liposarcoma: role of MDM2 and CDK4-
containing amplicons at 12q13–22. J Pathol. 1998;185(2):188-190.
doi:10.1002/(SICI)1096-9896(199701)181:1<14::AID-PATH730>3.0.CO;2-O.
12. Gobble RM, Qin L-X, Brill ER, Angeles C V, Ugras S, O’Connor RB, Moraco NH,
Decarolis PL, Antonescu C, Singer S. Expression profiling of liposarcoma yields a
multigene predictor of patient outcome and identifies genes that contribute to
liposarcomagenesis. Cancer Res. 2011;71(7):2697-705. doi:10.1158/0008-5472.CAN-
10-3588.
13. Aleixo PB, Hartmann a a, Menezes IC, Meurer RT, Oliveira a M. Can MDM2 and
CDK4 make the diagnosis of well differentiated/dedifferentiated liposarcoma? An
168 immunohistochemical study on 129 soft tissue tumours. J Clin Pathol.
2009;62(12):1127-35. doi:10.1136/jcp.2009.070201.
14. Rubin BP, Fletcher CDM. The cytogenetics of lipomatous tumours. Histopathology.
1997;30:507-511.
15. Yang S, Warraich ST, Nicholson G a, Blair IP. Fused in sarcoma/translocated in
liposarcoma: a multifunctional DNA/RNA binding protein. Int J Biochem Cell Biol.
2010;42(9):1408-11. doi:10.1016/j.biocel.2010.06.003.
16. Mariño-Enríquez A, Hornick JL, Dal Cin P, Cibas ES, Qian X. Dedifferentiated
liposarcoma and pleomorphic liposarcoma. Cancer Cytopathol. 2013:n/a-n/a.
doi:10.1002/cncy.21362.
17. Barretina J, Taylor B, Banerji S. Subtype-specific genomic alterations define new
targets for soft-tissue sarcoma therapy. Nature. 2010;42(8):715-721.
doi:10.1038/ng.619.Subtype-specific.
18. Meis JM. “Dedifferentiation” in bone and soft-tissue tumors. A histological indicator
of tumor progression. Pathol Annu. 1991;26:37-62.
19. Henricks WH, Chu YC, Goldblum JR, Weiss SW. Dedifferentiated liposarcoma: a
clinicopathological analysis of 155 cases with a proposal for an expanded definition of
dedifferentiation. Am J Surg Pathol. 1997;3(21):271-281.
20. Lurkin A, Ducimetière F, Vince DR, Decouvelaere A-V, Cellier D, Gilly FN, Salameire
D, Biron P, de Laroche G, Blay JY, Ray-Coquard I. Epidemiological evaluation of
concordance between initial diagnosis and central pathology review in a comprehensive
and prospective series of sarcoma patients in the Rhone-Alpes region. BMC Cancer.
2010;10(1):150. doi:10.1186/1471-2407-10-150.
169 21. Singer S. Dedifferentiated liposarcoma. 2013. Available at: www.orpha.net.
22. Tseng W, Somaiah N, Lazar A, Lev DC, Pollock RE. Novel Systemic Therapies in
Advanced Liposarcoma: A Review of Recent Clinical Trial Results. Cancers (Basel).
2013.
23. Ghadimi MP, Al-Zaid T, Madewell J, Peng T, Colombo C, Hoffman A, Creighton CJ,
Zhang Y, Zhang A, Lazar AJ, Pollock RE, Lev D. Diagnosis, management, and
outcome of patients with dedifferentiated liposarcoma systemic metastasis. Ann Surg
Oncol. 2011;18(13):3762-70. doi:10.1245/s10434-011-1794-0.
24. Dalal KM, Antonescu CR, Singer S. Diagnosis and management of lipomatous tumors.
J Surg Oncol. 2008;97(4):298-313. doi:10.1002/jso.20975.
25. Lietman SA. Soft-tissue sarcomas: Overview of management, with a focus on surgical
treatment considerations. Cleve Clin J Med. 2010;77 Suppl 1(SUPPL. 1):S13-7.
doi:10.3949/ccjm.77.s1.03.
26. O’Donnell PW, Griffin AM, Eward WC, Sternheim A, White LM, Wunder JS,
Ferguson PC. Can Experienced Observers Differentiate between Lipoma and Well-
Differentiated Liposarcoma Using Only MRI? Sarcoma. 2013;2013:982784.
doi:10.1155/2013/982784.
27. Weaver J, Downs-Kelly E, Goldblum JR, Turner S, Kulkarni S, Tubbs RR, Rubin BP,
Skacel M. Fluorescence in situ hybridization for MDM2 gene amplification as a
diagnostic tool in lipomatous neoplasms. Mod Pathol. 2008;21(8):943-9.
doi:10.1038/modpathol.2008.84.
28. Weaver J, Rao P, Goldblum JR, Joyce MJ, Turner SL, Lazar AJF, López-Terada D,
Tubbs RR, Rubin BP. Can MDM2 analytical tests performed on core needle biopsy be
170 relied upon to diagnose well-differentiated liposarcoma? Mod Pathol.
2010;23(10):1301-6. doi:10.1038/modpathol.2010.106.
29. Grimer R, Judson I, Peake D, Seddon B. Guidelines for the management of soft tissue
sarcomas. Sarcoma. 2010;2010(15s):506182. doi:10.1155/2010/506182.
30. Park JO, Qin L-X, Prete FP, Antonescu C, Murray BF, Singer S. Predicting outcome
by growth rate of locally recurrent retroperitoneal liposarcoma: the one centimeter per
month rule. Ann Surg. 2009;250(6):977-982.
31. Singer S, Antonescu CR, Riedel E, Brennan MF. Histologic subtype and margin of
resection predict pattern of recurrence and survival for retroperitoneal liposarcoma.
Ann Surg. 2003;238(3):358-70; discussion 370-1.
doi:10.1097/01.sla.0000086542.11899.38.
32. Antman K, Crowley J, Balcerzak SP, Rivkin SE, Weiss GR, Elias A, Natale RB,
Cooper RM, Barlogie B, Trump DL. An intergroup phase III randomized study of
doxorubicin and dacarbazine with or without ifosfamide and mesna in advanced soft
tissue and bone sarcomas. J Clin Oncol. 1993;11(7):1276-1285. Available at:
http://jco.ascopubs.org/content/11/7/1276.abstract.
33. Judson I, Verweij J, Gelderblom H, Hartmann JT, Schöffski P, Blay J-Y, Kerst JM,
Sufliarsky J, Whelan J, Hohenberger P, Krarup-Hansen A, Alcindor T, Marreaud S,
Litière S, Hermans C, Fisher C, Hogendoorn PCW, Dei Tos AP, van der Graaf WTA.
Doxorubicin alone versus intensified doxorubicin plus ifosfamide for first-line
treatment of advanced or metastatic soft-tissue sarcoma: a randomised controlled phase
3 trial. Lancet Oncol. 2014. doi:10.1016/S1470-2045(14)70063-4.
171 34. Benjamin R, Wiernik P, Bachur N. Adriamycin: a new effective agent in the therapy of
disseminated sarcomas. Med Pediatr Oncol. 1975;1(1):63-76.
35. Pommier Y, Leo E, Zhang H, Marchand C. DNA topoisomerases and their poisoning
by anticancer and antibacterial drugs. Chem Biol. 2010;17(5):421-33.
doi:10.1016/j.chembiol.2010.04.012.
36. Walczak BE, Irwin RB. Sarcoma chemotherapy. J Am Acad Orthop Surg.
2013;21:480+. Available at:
http://go.galegroup.com/ps/i.do?id=GALE|A341937016&v=2.1&u=txshr_a_main&it
=r&p=HRCA&sw=w&asid=3c1ee97f57644ea45c095d70e12cc3d5.
37. Nitiss JL. Targeting DNA topoisomerase II in cancer chemotherapy. Nat Rev Cancer.
2009;9(5):338-350. doi:10.1038/nrc2607.Targeting.
38. Misteli T, Soutoglou E. The emerging role of nuclear architecture in DNA repair and
genome maintenance. Nat Rev Mol Cell Biol. 2009;10(4):243-254.
doi:10.1038/nrm2651.The.
39. Borden EC, Amato DA, Rosenbaum C, Enterline HT, Shiraki MJ, Creech RH, Lerner
HJ, Carbone PP. Randomized comparison of three adriamycin regimens for metastatic
soft tissue sarcomas. J Clin Oncol . 1987;5 (6 ):840-850. Available at:
http://jco.ascopubs.org/content/5/6/840.abstract.
40. Pervaiz N, Colterjohn N, Farrokhyar F, Tozer R, Figueredo A, Ghert M. A systematic
meta-analysis of randomized controlled trials of adjuvant chemotherapy for localized
resectable soft-tissue sarcoma. Cancer. 2008;113(3):573-81. doi:10.1002/cncr.23592.
41. Antman KH. Chemotherapy of advanced sarcomas of bone and soft tissue. Semin
Oncol. 1992;19(6):13-20.
172 42. Cerny T, Leyvraz S, von Briel T, Küpfer A, Schaad R, Hsu Schmitz S-F, Honegger P,
Sessa C, Brunner J, Boddy A V, (SAKK) SG for CCR. Saturable metabolism of
continuous high-dose ifosfamide with Mesna and GM-CSF: A pharmacokinetic study
in advanced sarcoma patients. Ann Oncol . 1999;10 (9 ):1087-1094.
doi:10.1023/A:1008386000547.
43. Patel SR, Vadhan-Raj S, Papadopolous N, Plager C, Burgess MA, Hays C, Benjamin
RS. High-dose ifosfamide in bone and soft tissue sarcomas: results of phase II and pilot
studies--dose-response and schedule dependence. J Clin Oncol . 1997;15 (6 ):2378-
2384. Available at: http://jco.ascopubs.org/content/15/6/2378.abstract.
44. Lorigan P, Verweij J, Papai Z, Rodenhuis S, Le Cesne A, Leahy MG, Radford JA, Van
Glabbeke MM, Kirkpatrick A, Hogendoorn PCW, Blay J-Y. Phase III Trial of Two
Investigational Schedules of Ifosfamide Compared With Standard-Dose Doxorubicin
in Advanced or Metastatic Soft Tissue Sarcoma: A European Organisation for Research
and Treatment of Cancer Soft Tissue and Bone Sarcoma Group Study. J Clin Oncol .
2007;25 (21 ):3144-3150. doi:10.1200/JCO.2006.09.7717.
45. Edmonson JH, Ryan LM, Blum RH, Brooks JSJ, Shiraki M, Frytak S, Parkinson DR.
Randomized Comparison of Doxorubicin Alone Versus Ifosfamide Plus Doxorubicin
or Mitomycin, Doxorubicin, and Cisplatin Against Advanced Soft Tissue Sarcomas. J
Clin Oncol. 1993;11(7):1269-1275.
46. Santoro A, Tursz T, Mouridsen H, Verweij J, Steward W, Somers R, Buesa J, Casali
P, Spooner D, Rankin E, Kirkpatrick A, Van Glabbeke M, van Oosteram A.
Doxorubicin versus CYVADIC versus doxorubicin plus ifosfamide in first-line
treatment of advanced soft tissue sarcomas: a randomized study of the European
173 Organization for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma
Group. J Clin Oncol. 1995;13(7):1537-1545.
47. Pisters PW, Harrison LB, Leung DH, Woodruff JM, Casper ES, Brennan MF. Long-
term results of a prospective randomized trial of adjuvant brachytherapy in soft tissue
sarcoma. J Clin Oncol . 1996;14 (3 ):859-868. Available at:
http://jco.ascopubs.org/content/14/3/859.abstract.
48. Yang JC, Chang a E, Baker a R, Sindelar WF, Danforth DN, Topalian SL, DeLaney T,
Glatstein E, Steinberg SM, Merino MJ, Rosenberg S a. Randomized prospective study
of the benefit of adjuvant radiation therapy in the treatment of soft tissue sarcomas of
the extremity. J Clin Oncol. 1998;16(1):197-203. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/9440743.
49. Gronchi A, De Paoli A, Dani C, Merlo DF, Quagliuolo V, Grignani G, Bertola G,
Navarria P, Sangalli C, Buonadonna A, De Sanctis R, Sanfilippo R, Dei Tos AP,
Stacchiotti S, Giorello L, Fiore M, Bruzzi P, Casali PG. Preoperative chemo-radiation
therapy for localised retroperitoneal sarcoma: A phase I-II study from the Italian
Sarcoma Group. Eur J Cancer. 2014;50(4):784-92. doi:10.1016/j.ejca.2013.11.021.
50. Cho J, Lee SE, Choi Y-L. Diagnostic Value of MDM2 and DDIT3 Fluorescence In
Situ Hybridization in Liposarcoma Classification: A Single-Institution Experience.
Korean J Pathol. 2012;46(2):115-22. doi:10.4132/KoreanJPathol.2012.46.2.115.
51. Weaver J, Downs-Kelly E, Goldblum JR, Turner S, Kulkarni S, Tubbs RR, Rubin BP,
Skacel M. Fluorescence in situ hybridization for MDM2 gene amplification as a
diagnostic tool in lipomatous neoplasms. Mod Pathol. 2008;21(8):943-9.
doi:10.1038/modpathol.2008.84.
174 52. Ray-Coquard I, Blay J-Y, Italiano A, Le Cesne A, Penel N, Zhi J, Heil F, Rueger R,
Graves B, Ding M, Geho D, Middleton S a, Vassilev LT, Nichols GL, Bui BN. Effect
of the MDM2 antagonist RG7112 on the P53 pathway in patients with MDM2-
amplified, well-differentiated or dedifferentiated liposarcoma: an exploratory proof-of-
mechanism study. Lancet Oncol. 2012;13(11):1133-40. doi:10.1016/S1470-
2045(12)70474-6.
53. Keung EZ, Hornick JL, Bertagnolli MM, Baldini EH, Raut CP. Predictors of outcomes
in patients with primary retroperitoneal dedifferentiated liposarcoma undergoing
surgery. J Am Coll Surg. 2014;218(2):206-17. doi:10.1016/j.jamcollsurg.2013.10.009.
54. Fritz B, Schubert F, Wrobel G, Liposarcoma P, Schwaenen C, Wessendorf S, Nessling
M, Korz C, Rieker RJ, Montgomery K, Kucherlapati R, Mechtersheimer G, Eils R,
Joos S, Lichter P. Microarray-based Copy Number and Expression Profiling in
Dedifferentiated and Pleomorphic Liposarcoma. Cancer Res esearch.
2002;62(6):2993-2998.
55. Singer S, Socci ND, Ambrosini G, Sambol E, Decarolis P, Wu Y, O’Connor R, Maki
R, Viale A, Sander C, Schwartz GK, Antonescu CR. Gene expression profiling of
liposarcoma identifies distinct biological types/subtypes and potential therapeutic
targets in well-differentiated and dedifferentiated liposarcoma. Cancer Res.
2007;67(14):6626-36. doi:10.1158/0008-5472.CAN-07-0584.
56. Matushansky I, Hernando E, Socci ND, Matos T, Mills J, Edgar MA, Schwartz GK,
Singer S, Cordon-Cardo C, Maki RG. A developmental model of sarcomagenesis
defines a differentiation-based classification for liposarcomas. Am J Pathol.
2008;172(4):1069-80. doi:10.2353/ajpath.2008.070284.
175 57. Weaver J, Goldblum JR, Turner S, Tubbs RR, Wang W-L, Lazar AJ, Rubin BP.
Detection of MDM2 gene amplification or protein expression distinguishes sclerosing
mesenteritis and retroperitoneal fibrosis from inflammatory well-differentiated
liposarcoma. Mod Pathol. 2009;22(1):66-70. doi:10.1038/modpathol.2008.153.
58. Momand J, Jung D, Wilczynski S, Niland J. The MDM2 gene amplification database.
Nucleic Acids Res. 1998;26(15):3453-3459. doi:10.1093/nar/26.15.3453.
59. Baird K, Davis S, Antonescu CR, Harper UL, Walker RL, Chen Y, Glatfelter A a,
Duray PH, Meltzer PS. Gene expression profiling of human sarcomas: insights into
sarcoma biology. Cancer Res. 2005;65(20):9226-35. doi:10.1158/0008-5472.CAN-05-
1699.
60. Hoffman A, Lazar AJ, Pollock RE, Lev D. New frontiers in the treatment of
liposarcoma, a therapeutically resistant malignant cohort. Drug Resist Updat.
2011;14(1):52-66. doi:10.1016/j.drup.2010.11.001.
61. Zhang K, Chu K, Wu X, Gao H, Wang J, Yuan Y-C, Loera S, Ho K, Wang Y, Chow
W, Un F, Chu P, Yen Y. Amplification of FRS2 and activation of FGFR/FRS2
signaling pathway in high-grade liposarcoma. Cancer Res. 2013;73(4):1298-307.
doi:10.1158/0008-5472.CAN-12-2086.
62. Arlotta P, Tai AK, Manfioletti G, Clifford C, Jay G, Ono SJ. Transgenic mice
expressing a truncated form of the high mobility group I-C protein develop adiposity
and an abnormally high prevalence of lipomas. J Biol Chem. 2000;275(19):14394-400.
Available at: http://www.ncbi.nlm.nih.gov/pubmed/10747931. Accessed May 19,
2014.
176 63. Ortega S, Malumbres M, Barbacid M. Cyclin D-dependent kinases, INK4 inhibitors
and cancer. Biochim Biophys Acta - Rev Cancer. 2002;1602(1):73-87.
doi:10.1016/S0304-419X(02)00037-9.
64. Harbour JW, Luo RX, Dei Santi A, Postigo AA, Dean DC. Cdk phosphorylation
triggers sequential intramolecular interactions that progressively block Rb functions as
cells move through G1. Cell. 1999;98(6):859-69. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/10499802. Accessed April 29, 2014.
65. Italiano A, Bianchini L, Gjernes E, Keslair F, Ranchere-Vince D, Dumollard J-M,
Haudebourg J, Leroux A, Mainguené C, Terrier P, Chibon F, Coindre J-M, Pedeutour
F. Clinical and biological significance of CDK4 amplification in well-differentiated
and dedifferentiated liposarcomas. Clin Cancer Res. 2009;15(18):5696-703.
doi:10.1158/1078-0432.CCR-08-3185.
66. Schmidt H, Bartel F, Kappler M, Würl P, Lange H, Bache M, Holzhausen H-J, Taubert
H. Gains of 13q are correlated with a poor prognosis in liposarcoma. Mod Pathol.
2005;18(5):638-44. doi:10.1038/modpathol.3800326.
67. Taylor BS, Barretina J, Maki RG, Antonescu CR, Singer S, Ladanyi M. Advances in
sarcoma genomics and new therapeutic targets. Nat Rev Cancer. 2011;11(8):541-57.
doi:10.1038/nrc3087.
68. McCarthy N. JUN steps in. Nat Rev Cancer. 2005;5(8):584. Available at:
http://dx.doi.org/10.1038/nrc1687.
69. Mariani O, Brennetot C, Coindre J-M, Gruel N, Ganem C, Delattre O, Stern M-H,
Aurias A. JUN oncogene amplification and overexpression block adipocytic
177 differentiation in highly aggressive sarcomas. Cancer Cell. 2007;11(4):361-74.
doi:10.1016/j.ccr.2007.02.007.
70. Szymanska J, Virolainen M, Tarkkanen M, Wiklund T, Asko-Seljavaara S, Tukiainen
E, Elomaa I, Blomqvist C, Knuutila S. Overrepresentation of 1q21-23 and 12q13-21 in
lipoma-like liposarcomas but not in benign lipomas: a comparative genomic
hybridization study. Cancer Genet Cytogenet. 1997;99(1):14-8. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/9352790. Accessed May 21, 2014.
71. Chibon F, Mariani O, Derré J, Mairal A, Coindre J-M, Guillou L, Sastre X, Pédeutour
F, Aurias A. ASK1 (MAP3K5) as a potential therapeutic target in malignant fibrous
histiocytomas with 12q14-q15 and 6q23 amplifications. Genes Chromosomes Cancer.
2004;40(1):32-7. doi:10.1002/gcc.20012.
72. Momand J, Zambetti GP, Olson DC, George D, Levine AJ. The mdm-2 oncogene
product forms a complex with the p53 protein and inhibits p53-mediated
transactivation. Cell. 1992;69(7):1237-1245. doi:10.1016/0092-8674(92)90644-R.
73. Fritz B, Schubert F, Wrobel G, Schwaenen C, Wessendorf S, Nessling M, Korz C,
Rieker RJ, Montgomery K, Kucherlapati R, Mechtersheimer G, Eils R, Joos S, Lichter
P. Microarray-based Copy Number and Expression Profiling in Dedifferentiated and
Pleomorphic Liposarcoma. Cancer Res. 2002;62(11):2993-2998. Available at:
http://cancerres.aacrjournals.org/content/62/11/2993.short. Accessed May 21, 2014.
74. Wu Y V, Okada T, DeCarolis P, Socci N, O’Connor R, Geha RC, Joy Somberg C,
Antonescu C, Singer S. Restoration of C/EBPα in dedifferentiated liposarcoma induces
G2/M cell cycle arrest and apoptosis. Genes Chromosomes Cancer. 2012;51(4):313-
27. doi:10.1002/gcc.21917.
178 75. Peng T, Zhang P, Liu J, Nguyen T, Bolshakov S, Belousov R, Young ED, Wang X,
Brewer K, López-Terrada DH, Oliveira AM, Lazar AJ, Lev D. An experimental model
for the study of well-differentiated and dedifferentiated liposarcoma; deregulation of
targetable tyrosine kinase receptors. Lab Invest. 2011;91(3):392-403.
doi:10.1038/labinvest.2010.185.
76. Torres KE, Zhu Q-S, Bill K, Lopez G, Ghadimi MP, Xie X, Young ED, Liu J, Nguyen
T, Bolshakov S, Belousov R, Wang S, Lahat G, Liu J, Hernandez B, Lazar AJ, Lev D.
Activated MET is a molecular prognosticator and potential therapeutic target for
malignant peripheral nerve sheath tumors. Clin Cancer Res. 2011;17(12):3943-55.
doi:10.1158/1078-0432.CCR-11-0193.
77. Graveel CR, Tolbert D, Vande Woude GF. MET: a critical player in tumorigenesis and
therapeutic target. Cold Spring Harb Perspect Biol. 2013;5(7).
doi:10.1101/cshperspect.a009209.
78. Lane DP. p53, guardian of the genome. Nature. 1992;358(6381):15-16.
doi:10.1038/358015a0.
79. Levine a J. P53, the Cellular Gatekeeper for Growth and Division. Cell.
1997;88(3):323-31. Available at: http://www.ncbi.nlm.nih.gov/pubmed/9039259.
80. Leach FS, Tokino T, Meltzer P, Burrell M, Oliner JD, Smith S, Hill DE, Sidransky D,
Kinzler KW, Vogelstein B. p53 Mutation and MDM2 Amplification in Human Soft
Tissue Sarcomas. Cancer Res. 1993;53:2231-2234.
81. Shieh SY, Ahn J, Tamai K, Taya Y, Prives C. The human homologs of checkpoint
kinases Chk1 and Cds1 (Chk2) phosphorylate p53 at multiple DNA damage-inducible
sites. Genes Dev. 2000;14(3):289-300. Available at:
179 http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=316358&tool=pmcentrez
&rendertype=abstract. Accessed May 30, 2014.
82. Gannon HS, Woda B a, Jones SN. ATM phosphorylation of Mdm2 Ser394 regulates
the amplitude and duration of the DNA damage response in mice. Cancer Cell.
2012;21(5):668-79. doi:10.1016/j.ccr.2012.04.011.
83. Reaper PM, Griffiths MR, Long JM, Charrier J-D, Maccormick S, Charlton PA, Golec
JMC, Pollard JR. Selective killing of ATM- or p53-deficient cancer cells through
inhibition of ATR. Nat Chem Biol. 2011;7(7):428-30. doi:10.1038/nchembio.573.
84. Hsieh J-K, Chan FS., O’Connor DJ, Mittnacht S, Zhong S, Lu X. RB Regulates the
Stability and the Apoptotic Function of p53 via MDM2. Mol Cell. 1999;3(2):181-193.
doi:10.1016/S1097-2765(00)80309-3.
85. Terzian T, Lozano G. Building p53. Genes Dev. 2010;24(20):2229-32.
doi:10.1101/gad.1988510.
86. Yu J, Zhang L. The transcriptional targets of p53 in apoptosis control. Biochem Biophys
Res Commun. 2005;331(3):851-8. doi:10.1016/j.bbrc.2005.03.189.
87. Hainaut P, Wiman KG. 30 years and a long way into p53 research. Lancet Oncol.
2009;10(9):913-9. doi:10.1016/S1470-2045(09)70198-6.
88. Aylon Y, Oren M. Living with p53, dying of p53. Cell. 2007;130(4):597-600.
doi:10.1016/j.cell.2007.08.005.
89. Kubbutat MH, Jones SN, Vousden KH. Regulation of p53 stability by Mdm2. Nature.
1997;387(6630):299-303. doi:10.1038/387299a0.
180 90. Grossman SR. p300/CBP/p53 interaction and regulation of the p53 response. Eur J
Biochem. 2001;268(10):2773-8. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/11358491. Accessed May 20, 2014.
91. Morgan SE, Kastan MB. p53 and ATM: cell cycle, cell death, and cancer. Adv Cancer
Res. 1997;71:1-25. Available at: http://www.ncbi.nlm.nih.gov/pubmed/9111862.
Accessed May 20, 2014.
92. Kastan MB, Bartek J. Cell-cycle checkpoints and cancer. Nature. 2004;432(7015):316-
23. doi:10.1038/nature03097.
93. Li M, Brooks CL, Kon N, Gu W. A dynamic role of HAUSP in the p53-Mdm2 pathway.
Mol Cell. 2004;13(6):879-86. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/15053880. Accessed May 13, 2014.
94. Kruse J-P, Gu W. Modes of p53 regulation. Cell. 2009;137(4):609-22.
doi:10.1016/j.cell.2009.04.050.
95. Joerger a C, Fersht a R. Structure-function-rescue: the diverse nature of common p53
cancer mutants. Oncogene. 2007;26(15):2226-42. doi:10.1038/sj.onc.1210291.
96. Soussi T, May P. Structural aspects of the p53 protein in relation to gene evolution: a
second look. J Mol Biol. 1996;260(5):623-37. doi:10.1006/jmbi.1996.0425.
97. Shan B, Li D-W, Brüschweiler-Li L, Brüschweiler R. Competitive binding between
dynamic p53 transactivation subdomains to human MDM2 protein: implications for
regulating the p53·MDM2/MDMX interaction. J Biol Chem. 2012;287(36):30376-84.
doi:10.1074/jbc.M112.369793.
98. Okorokov AL, Orlova E V. Structural biology of the p53 tumour suppressor. Curr Opin
Struct Biol. 2009;19(2):197-202. doi:10.1016/j.sbi.2009.02.003.
181 99. Horn HF, Vousden KH. Coping with stress: multiple ways to activate p53. Oncogene.
2007;26(9):1306-16. doi:10.1038/sj.onc.1210263.
100. Harris SL, Levine AJ. The p53 pathway: positive and negative feedback loops.
Oncogene. 2005;24(17):2899-908. doi:10.1038/sj.onc.1208615.
101. Robles AI, Bemmels NA, Foraker AB, Harris CC. APAF-1 is a transcriptional target
of p53 in DNA damage-induced apoptosis. Cancer Res. 2001;61(18):6660-4. Available
at: http://www.ncbi.nlm.nih.gov/pubmed/11559530. Accessed May 20, 2014.
102. Toshiyuki M, Reed JC. Tumor suppressor p53 is a direct transcriptional activator of the
human bax gene. Cell. 1995;80(2):293-299. doi:10.1016/0092-8674(95)90412-3.
103. Owen-Schaub LB, Zhang W, Cusack JC, Angelo LS, Santee SM, Fujiwara T, Roth JA,
Deisseroth AB, Zhang WW, Kruzel E. Wild-type human p53 and a temperature-
sensitive mutant induce Fas/APO-1 expression. Mol Cell Biol. 1995;15(6):3032-40.
Available at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=230534&tool=pmcentrez
&rendertype=abstract. Accessed May 13, 2014.
104. Oda E. Noxa, a BH3-Only Member of the Bcl-2 Family and Candidate Mediator of
p53-Induced Apoptosis. Science (80- ). 2000;288(5468):1053-1058.
doi:10.1126/science.288.5468.1053.
105. Nakano K, Vousden KH. PUMA, a Novel Proapoptotic Gene, Is Induced by p53. Mol
Cell. 2001;7(3):683-694. doi:10.1016/S1097-2765(01)00214-3.
106. Broude E, Demidenko Z, Vivo C, Swift M, Davis B, Blagosklonny, MV Roninson I.
p21 (CDKN1A) is a Negative Regulator of p53 Stability. Cell Cycle. 2007;6:1467-
1470.
182 107. Hollander MC, Alamo I, Jackman J, Wang MG, McBride OW, Fornace AJ. Analysis
of the mammalian gadd45 gene and its response to DNA damage. J Biol Chem.
1993;268(32):24385-93. Available at: http://www.ncbi.nlm.nih.gov/pubmed/8226988.
Accessed May 13, 2014.
108. Hermeking H, Lengauer C, Polyak K, He T-C, Zhang L, Thiagalingam S, Kinzler KW,
Vogelstein B. 14-3-3σ Is a p53-Regulated Inhibitor of G2/M Progression. Mol Cell.
1997;1(1):3-11. doi:10.1016/S1097-2765(00)80002-7.
109. Dameron KM, Volpert O V, Tainsky MA, Bouck N. Control of angiogenesis in
fibroblasts by p53 regulation of thrombospondin-1. Science (80- ).
1994;265(5178):1582-4. Available at: http://www.ncbi.nlm.nih.gov/pubmed/7521539.
Accessed May 13, 2014.
110. Wu X, Bayle JH, Olson D, Levine AJ. The p53-mdm-2 autoregulatory feedback loop.
Genes Dev. 1993;7(7A):1126-32. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/8319905. Accessed May 13, 2014.
111. Lim S, Kaldis P. Cdks, cyclins and CKIs: roles beyond cell cycle regulation.
Development. 2013;140(15):3079-93. doi:10.1242/dev.091744.
112. Rane C, Minden A. P21 activated kinases: Structure, regulation, and functions. Small
GTPases. 2014;5(1). doi:10.4161/sgtp.28003.
113. Cazzalini O, Scovassi AI, Savio M, Stivala LA, Prosperi E. Multiple roles of the cell
cycle inhibitor p21(CDKN1A) in the DNA damage response. Mutat Res.
2010;704(1):12-20. doi:10.1016/j.mrrev.2010.01.009.
114. Agarwal ML, Agarwal a, Taylor WR, Stark GR. p53 controls both the G2/M and the
G1 cell cycle checkpoints and mediates reversible growth arrest in human fibroblasts.
183 Proc Natl Acad Sci U S A. 1995;92(18):8493-7. Available at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=41183&tool=pmcentrez&
rendertype=abstract.
115. Ashcroft M, Taya Y, Vousden KH. Stress signals utilize multiple pathways to stabilize
p53. Mol Cell Biol. 2000;20(9):3224-33. Available at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=85616&tool=pmcentrez&
rendertype=abstract. Accessed May 7, 2014.
116. Nevins JR. The Rb/E2F pathway and cancer. Hum Mol Genet. 2001;10(7):699-703.
Available at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3131836&tool=pmcentre
z&rendertype=abstract.
117. Innocente SA, Abrahamson JLA, Cogswell JP, Lee JM. p53 regulates a G2 checkpoint
through cyclin B1. Proc Natl Acad Sci. 1999;96(5):2147-2152.
doi:10.1073/pnas.96.5.2147.
118. Loewer A, Batchelor E, Gaglia G, Lahav G. Basal dynamics of p53 reveal
transcriptionally attenuated pulses in cycling cells. Cell. 2010;142(1):89-100.
doi:10.1016/j.cell.2010.05.031.
119. Pucci B, Kasten M, Giordano A. Cell cycle and apoptosis. Neoplasia. 2000;2(4):291-
9. Available at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1550296&tool=pmcentre
z&rendertype=abstract.
184 120. Yonish-Rouach E, Resnitzky D, Lotem J, Sachs L, Kimchi A, Oren M. Wild-type p53
induces apoptosis of myeloid leukaemic cells that is inhibited by interleukin-6. Nature.
1991;352(6333):345-7. doi:10.1038/352345a0.
121. Bieging KT, Attardi LD. Deconstructing p53 transcriptional networks in tumor
suppression. Trends Cell Biol. 2012;22(2):97-106. doi:10.1016/j.tcb.2011.10.006.
122. Vaseva A V, Moll UM. The mitochondrial p53 pathway. Biochim Biophys Acta.
2009;1787(5):414-20. doi:10.1016/j.bbabio.2008.10.005.
123. Oda E. Noxa, a BH3-Only Member of the Bcl-2 Family and Candidate Mediator of
p53-Induced Apoptosis. Science (80- ). 2000;288(5468):1053-1058.
doi:10.1126/science.288.5468.1053.
124. Leroy B, Anderson M, Soussi T. TP53 mutations in human cancer: database
reassessment and prospects for the next decade. Hum Mutat. 2014:1-70.
doi:10.1002/humu.22552.This.
125. Leach FS, Tokino T, Meltzer P, Burrell M, Oliner JD, Smith S, Hill DE, Sidransky D,
Kinzler KW, Vogelstein B. p53 Mutation and MDM2 Amplification in Human Soft
Tissue Sarcomas. 1993:2231-2234.
126. Kussie PH, Gorina S, Marechal V, Elenbaas B, Moreau J, Levine AJ, Pavletich NP.
Structure of the MDM2 Oncoprotein Bound to the p53 Tumor Suppressor
Transactivation Domain. Science (80- ). 2014;274(5289):948-953. Available at:
http://www.jstor.org/stable/2891290.
127. Knudson a G. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad
Sci U S A. 1971;68(4):820-3. Available at:
185 http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=389051&tool=pmcentrez
&rendertype=abstract.
128. Kruse J-P, Gu W. SnapShot: p53 posttranslational modifications. Cell.
2008;133(5):930-30.e1. doi:10.1016/j.cell.2008.05.020.
129. O’Brate A, Giannakakou P. The importance of p53 location: nuclear or cytoplasmic zip
code? Drug Resist Updat. 2003;6(6):313-322. doi:10.1016/j.drup.2003.10.004.
130. Brooks CL, Gu W. p53 ubiquitination: Mdm2 and beyond. Mol Cell. 2006;21(3):307-
15. doi:10.1016/j.molcel.2006.01.020.
131. Shan J, Brooks C, Kon N, Li M, Gu W. Dissecting roles of ubiquitination in the p53
pathway. Ernst Schering Found Symp Proc. 2008;(1):127-36. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/19202598. Accessed May 7, 2014.
132. Ashcroft M, Vousden KH. Regulation of p53 stability. Oncogene. 1999;18(53):7637-
43. doi:10.1038/sj.onc.1203012.
133. Perry ME, Piette J, Zawadzki J a, Harvey D, Levine a J. The mdm-2 gene is induced in
response to UV light in a p53-dependent manner. Proc Natl Acad Sci U S A.
1993;90(24):11623-7. Available at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=48036&tool=pmcentrez&
rendertype=abstract.
134. Cahilly-Snyder L, Yang-Feng T, Francke U, George DL. Molecular analysis and
chromosomal mapping of amplified genes isolated from a transformed mouse 3T3 cell
line. Somat Cell Mol Genet. 1987;13(3):235-44. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/3474784. Accessed May 7, 2014.
186 135. Fakharzadeh SS, Trusko SP, George DL. Tumorigenic potential associated with
enhanced expression of a gene that is amplified in a mouse tumor cell line. EMBO J.
1991;10(6):1565-9. Available at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=452821&tool=pmcentrez
&rendertype=abstract. Accessed May 7, 2014.
136. J M, GP Z, Olson D, George D, Levine A. The mdm-2 oncogene product forms a
complex with the p53 protein and inhi- bits p53-mediated
transactivation.pdf.crdownload. Cell. 1992;69:1237-1245.
137. Oliner JD, Kinzler KW, Meltzer PS, George DL, Vogelstein B. Amplification of a gene
encoding a p53-associated protein in human sarcomas. Nature. 1992;358(6381):80-83.
Available at: http://dx.doi.org/10.1038/358080a0.
138. De Oca Luna RM, Wagner DS, Lozano G. Rescue of early embryonic lethality in
mdm2-deficient mice by deletion of p53. Nature. 1995;378(6553):203-206. Available
at: http://dx.doi.org/10.1038/378203a0.
139. Jones SN, Roe AE, Donehower LA, Bradley A. Rescue of embryonic lethality in
Mdm2-deficient mice by absence of p53. Nature. 1995;378(6553):206-208. Available
at: http://dx.doi.org/10.1038/378206a0.
140. Terzian T, Wang Y, Van Pelt CS, Box NF, Travis EL, Lozano G. Haploinsufficiency
of Mdm2 and Mdm4 in Tumorigenesis and Development. Mol Cell Biol.
2007;27(15):5479-5485. doi:10.1128/MCB.00555-06.
141. Roth J, Dobbelstein M, Freedman D a, Shenk T, Levine a J. Nucleo-cytoplasmic
shuttling of the hdm2 oncoprotein regulates the levels of the p53 protein via a pathway
187 used by the human immunodeficiency virus rev protein. EMBO J. 1998;17(2):554-64.
doi:10.1093/emboj/17.2.554.
142. Zhang Y, Lu H. Signaling to p53: ribosomal proteins find their way. Cancer Cell.
2009;16(5):369-77. doi:10.1016/j.ccr.2009.09.024.
143. Vichi A, Moss J, Vaughan M. ADP-ribosylation factor domain protein 1 (ARD1), a
multifunctional protein with ubiquitin E3 ligase, GAP, and ARF domains. Methods
Enzymol. 2005;404:195-206. doi:10.1016/S0076-6879(05)04019-X.
144. Poyurovsky M V, Priest C, Kentsis A, Borden KLB, Pan Z-Q, Pavletich N, Prives C.
The Mdm2 RING domain C-terminus is required for supramolecular assembly and
ubiquitin ligase activity. EMBO J. 2007;26(1):90-101. doi:10.1038/sj.emboj.7601465.
145. Martins CP, Brown-Swigart L, Evan GI. Modeling the therapeutic efficacy of p53
restoration in tumors. Cell. 2006;127(7):1323-34. doi:10.1016/j.cell.2006.12.007.
146. Chen J, Marechal V, Levine AJ. Mapping of the p53 and mdm-2 interaction domains.
Mol Cell Biol. 1993;13(7):4107-4114. doi:10.1128/MCB.13.7.4107.
147. Juven-gershon T, Oren M. Minireview Mdm2 : The Ups and Downs. Mol Med.
1999;5(12):71-83.
148. Wu X, Bayle JH, Olson D, Levine a J. The p53-mdm-2 autoregulatory feedback loop.
Genes Dev. 1993;7(7a):1126-1132. doi:10.1101/gad.7.7a.1126.
149. Schlesinger DH, Goldstein G, Niall HD. The complete amino acid sequence of
ubiquitin, an adenylate cyclase stimulating polypeptide probably universal in living
cells. Biochemistry. 1975;14(10):2214-8. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/1170880. Accessed May 7, 2014.
188 150. Hoeller D, Hecker C-M, Dikic I. Ubiquitin and ubiquitin-like proteins in cancer
pathogenesis. Nat Rev Cancer. 2006;6(10):776-88. doi:10.1038/nrc1994.
151. Fang S, Weissman AM. A field guide to ubiquitylation. Cell Mol Life Sci.
2004;61(13):1546-61. doi:10.1007/s00018-004-4129-5.
152. Chau V, Tobias JW, Bachmair A, Marriotr D, Ecker DJ, Gonda DK, Varshavsky A. A
multiubiquitin chain is confined to specific lysine in a targeted short lived protein.
Science (80- ). 1989;243(3):1576-83.
153. Lai Z, Ferry K V, Diamond MA, Wee KE, Kim YB, Ma J, Yang T, Benfield PA,
Copeland RA, Auger KR. Human mdm2 mediates multiple mono-ubiquitination of p53
by a mechanism requiring enzyme isomerization. J Biol Chem. 2001;276(33):31357-
67. doi:10.1074/jbc.M011517200.
154. Sun Y. E3 ubiquitin ligases as cancer targets and biomarkers. Neoplasia.
2006;8(8):645-54. doi:10.1593/neo.06376.
155. Riley MF, Lozano G. The Many Faces of MDM2 Binding Partners. Genes Cancer.
2012;3(3-4):226-39. doi:10.1177/1947601912455322.
156. Tao W, Levine AJ. P19(ARF) stabilizes p53 by blocking nucleo-cytoplasmic shuttling
of Mdm2. Proc Natl Acad Sci U S A. 1999;96(12):6937-41. Available at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=22020&tool=pmcentrez&
rendertype=abstract. Accessed May 21, 2014.
157. Maya R, Balass M, Kim ST, Shkedy D, Leal JF, Shifman O, Moas M, Buschmann T,
Ronai Z, Shiloh Y, Kastan MB, Katzir E, Oren M. ATM-dependent phosphorylation of
Mdm2 on serine 395: role in p53 activation by DNA damage. Genes Dev.
2001;15(9):1067-77. doi:10.1101/gad.886901.
189 158. Clegg H V, Itahana Y, Itahana K, Ramalingam S, Zhang Y. Mdm2 RING mutation
enhances p53 transcriptional activity and p53-p300 interaction. PLoS One.
2012;7(5):e38212. doi:10.1371/journal.pone.0038212.
159. Loughran O, La Thangue NB. Apoptotic and Growth-Promoting Activity of E2F
Modulated by MDM2. Mol Cell Biol. 2000;20(6):2186-2197.
doi:10.1128/MCB.20.6.2186-2197.2000.
160. Su X, Chakravarti D, Flores ER. p63 steps into the limelight: crucial roles in the
suppression of tumorigenesis and metastasis. Nat Rev Cancer. 2013;13(2):136-43.
doi:10.1038/nrc3446.
161. Zeng X, Chen L, Jost CA, Maya R, Keller D, Wang X, Kaelin WG, Oren M, Chen J,
Lu H. MDM2 suppresses p73 function without promoting p73 degradation. Mol Cell
Biol. 1999;19(5):3257-66. Available at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=84120&tool=pmcentrez&
rendertype=abstract. Accessed May 21, 2014.
162. Landers JE, Haines DS, Strauss JF, George DL. Enhanced translation: a novel
mechanism of mdm2 oncogene overexpression identified in human tumor cells.
Oncogene. 1994;9(9):2745-50. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/8058341. Accessed May 13, 2014.
163. Bond GL, Hu W, Bond EE, Robins H, Lutzker SG, Arva NC, Bargonetti J, Bartel F,
Taubert H, Wuerl P, Onel K, Yip L, Hwang S-J, Strong LC, Lozano G, Levine AJ. A
single nucleotide polymorphism in the MDM2 promoter attenuates the p53 tumor
suppressor pathway and accelerates tumor formation in humans. Cell.
2004;119(5):591-602. doi:10.1016/j.cell.2004.11.022.
190 164. Senturk E, Manfredi JJ. Mdm2 and tumorigenesis: evolving theories and unsolved
mysteries. Genes Cancer. 2012;3(3-4):192-8. doi:10.1177/1947601912457368.
165. Vousden KH. Activation of the p53 tumor suppressor protein. Biochim Biophys Acta -
Rev Cancer. 2002;1602(1):47-59. doi:10.1016/S0304-419X(02)00035-5.
166. Kussie PH, Gorina S, Marechal V, Elenbaas B, Moreau J, Levine AJ, Pavletich NP.
Structure of the MDM2 oncoprotein bound to the p53 tumor suppressor transactivation
domain. Science. 1996;274(5289):948-53. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/8875929. Accessed May 21, 2014.
167. Shangary S, Wang S. Small-molecule inhibitors of the MDM2-p53 protein-protein
interaction to reactivate p53 function: a novel approach for cancer therapy. Annu Rev
Pharmacol Toxicol. 2009;49:223-41.
doi:10.1146/annurev.pharmtox.48.113006.094723.
168. Wang S, Zhao Y, Bernard D, Aguilar A. Protein-Protein Interactions. Wendt MD, ed.
2012;8:57-80. doi:10.1007/978-3-642-28965-1.
169. Popowicz GM, Dömling A, Holak T a. The structure-based design of Mdm2/Mdmx-
p53 inhibitors gets serious. Angew Chemie Int Ed English. 2011;50(12):2680-8.
doi:10.1002/anie.201003863.
170. Vu BT. Small-molecule inhibitors of the p53-MDM2 interaction. Curr Top Microbiol
Immunol. 2011;348:151-72.
171. Chène P. Inhibition of the p53-MDM2 Interaction : Targeting a Protein-Protein
Interface Inhibition of the p53-MDM2 Interaction: Targeting a Protein-Protein
Interface. Mol cancer Res. 2004;2(1):20-28.
191 172. Chen J, Wang J, Xu B, Zhu W, Li G. Insight into mechanism of small molecule
inhibitors of the MDM2-p53 interaction: molecular dynamics simulation and free
energy analysis. J Mol Graph Model. 2011;30:46-53.
doi:10.1016/j.jmgm.2011.06.003.
173. Shangary S, Wang S. Targeting the MDM2-p53 interaction for cancer therapy. Clin
Cancer Res. 2008;14(17):5318-24. doi:10.1158/1078-0432.CCR-07-5136.
174. Vassilev LT, Vu BT, Graves B, Carvajal D, Podlaski F, Filipovic Z, Kong N, Kammlott
U, Lukacs C, Klein C, Fotouhi N, Liu E a. In vivo activation of the p53 pathway by
small-molecule antagonists of MDM2. Science (80- ). 2004;303(5659):844-8.
doi:10.1126/science.1092472.
175. Klein C, Vassilev LT. Targeting the p53-MDM2 interaction to treat cancer. Br J
Cancer. 2004;91(8):1415-9. doi:10.1038/sj.bjc.6602164.
176. Galatin PS, Abraham DJ. A nonpeptidic sulfonamide inhibits the p53-mdm2
interaction and activates p53-dependent transcription in mdm2-overexpressing cells. J
Med Chem. 2004;47(17):4163-5. doi:10.1021/jm034182u.
177. Ding K, Lu Y, Nikolovska-Coleska Z, Wang G, Qiu S, Shangary S, Gao W, Qin D,
Stuckey J, Krajewski K, Roller PP, Wang S. Structure-Based Design of Spiro-
oxindoles as Potent, Specific Small-Molecule Inhibitors of the MDM2−p53 Interaction.
J Med Chem. 2006;49(12):3432-3435. doi:10.1021/jm051122a.
178. Shangary S, Qin D, McEachern D, Liu M, Miller RS, Qiu S, Nikolovska-Coleska Z,
Ding K, Wang G, Chen J, Bernard D, Zhang J, Lu Y, Gu Q, Shah RB, Pienta KJ, Ling
X, Kang S, Guo M, Sun Y, Yang D, Wang S. Temporal activation of p53 by a specific
MDM2 inhibitor is selectively toxic to tumors and leads to complete tumor growth
192 inhibition. Proc Natl Acad Sci U S A. 2008;105(10):3933-8.
doi:10.1073/pnas.0708917105.
179. Yu S, Qin D, Shangary S, Chen J, Wang G, Ding K, McEachern D, Qiu S, Nikolovska-
Coleska Z, Miller R, Kang S, Yang D, Wang S. Potent and orally active small-molecule
inhibitors of the MDM2-p53 interaction. J Med Chem. 2009;52(24):7970-3.
doi:10.1021/jm901400z.
180. Bolshakov S, Walker CM, Strom SS, Selvan MS, Clayman GL, El-Naggar A, Lippman
SM, Kripke ML, Ananthaswamy HN. p53 Mutations in Human Aggressive and
Nonaggressive Basal and Squamous Cell Carcinomas. Clin Cancer Res . 2003;9 (1
):228-234. Available at: http://clincancerres.aacrjournals.org/content/9/1/228.abstract.
181. Stratford EW, Castro R, Daffinrud J, Skarn M, Lauvrak S, Munthe E, Myklebost O.
Characterization of Liposarcoma Cell Lines for Preclinical and Biological Studies.
Sarcoma. 2012. doi:10.1155/2012/148614.
182. Wang Z, Yang L, Jiang Y, Ling Z-Q, Li Z, Cheng Y, Huang H, Wang L, Pan Y, Wang
Z, Yan X, Chen Y. High fat diet induces formation of spontaneous liposarcoma in
mouse adipose tissue with overexpression of interleukin 22. PLoS One.
2011;6(8):e23737. doi:10.1371/journal.pone.0023737.
183. Gutierrez A, Snyder EL, Marino-enriquez A, Zhang Y, Sioletic S. Aberrant AKT
activation drives well-differentiated liposarcoma. 2011.
doi:10.1073/pnas.1106127108/-
/DCSupplemental.www.pnas.org/cgi/doi/10.1073/pnas.1106127108.
193 184. Tschoep K, Kohlmann A, Schlemmer M, Haferlach T, Issels R-D. Gene expression
profiling in sarcomas. Crit Rev Oncol Hematol. 2007;63(2):111-24.
doi:10.1016/j.critrevonc.2007.04.001.
185. Hostein I, Coindre J-M, Derré J, Mariani O, Chibon F, Aurias A. Comparative Genomic
Hybridization Study of Paraffin-Embedded Dedifferentiated Liposarcoma Fixed With
Holland Bouin’s Fluid. Diagnostic Mol Pathol. 2003;12(3):166-173.
doi:10.1097/00019606-200309000-00008.
186. Hostein I, Coindre JM, Derré J, Mariani O, Chibon F, Aurias A. Comparative genomic
hybridization study of paraffin-embedded dedifferentiated liposarcoma fixed with
Holland Bouin’s fluid. Diagnostic Mol Pathol. 2003;12(3):166-173. Available at:
http://www.scopus.com/inward/record.url?eid=2-s2.0-
0042388353&partnerID=tZOtx3y1.
187. Layfield LJ, Schmidt RL, Sangle N, Crim JR. Diagnostic accuracy and clinical utility
of biopsy in musculoskeletal lesions: A comparison of fine-needle aspiration, core, and
open biopsy techniques. Diagn Cytopathol. 2014:n/a-n/a. doi:10.1002/dc.23005.
188. Horvai AE, DeVries S, Roy R, O’Donnell RJ, Waldman F. Similarity in genetic
alterations between paired well-differentiated and dedifferentiated components of
dedifferentiated liposarcoma. Mod Pathol. 2009;22(11):1477-88.
doi:10.1038/modpathol.2009.119.
189. Kan CE, Patton JT, Stark GR, Jackson MW. p53-mediated growth suppression in
response to Nutlin-3 in cyclin D1 transformed cells occurs independently of p21.
Cancer Res. 2007;67(20):9862-8. doi:10.1158/0008-5472.CAN-07-0259.
194 190. Zhang Z, Wang H, Li M, Agrawal S, Chen X, Zhang R. MDM2 Is a Negative Regulator
of p21WAF1/CIP1, Independent of p53. J Biol Chem . 2004;279 (16 ):16000-16006.
doi:10.1074/jbc.M312264200.
191. Valentine JM, Kumar S, Moumen A. A p53-independent role for the MDM2 antagonist
Nutlin-3 in DNA damage response initiation. BMC Cancer. 2011;11(1):79.
doi:10.1186/1471-2407-11-79.
192. Ohnstad HO, Paulsen EB, Noordhuis P, Berg M, Lothe R a, Vassilev LT, Myklebost
O. MDM2 antagonist Nutlin-3a potentiates antitumour activity of cytotoxic drugs in
sarcoma cell lines. BMC Cancer. 2011;11(1):211:1-11. doi:10.1186/1471-2407-11-
211.
193. Manfé V, Biskup E, Johansen P, Kamstrup MR, Krejsgaard TF, Morling N, Wulf HC,
Gniadecki R. MDM2 inhibitor nutlin-3a induces apoptosis and senescence in cutaneous
T-cell lymphoma: role of p53. J Invest Dermatol. 2012;132(5):1487-96.
doi:10.1038/jid.2012.10.
194. Vassilev LT. MDM2 inhibitors for cancer therapy. Trends Mol Med. 2007;13(1):23-
31. doi:10.1016/j.molmed.2006.11.002.
195. Espinoza-Fonseca LM. Targeting MDM2 by the small molecule RITA: towards the
development of new multi-target drugs against cancer. Theor Biol Med Model.
2005;2:38. doi:10.1186/1742-4682-2-38.
196. Müller CR, Paulsen EB, Noordhuis P, Pedeutour F, Saeter G, Myklebost O. Potential
for treatment of liposarcomas with the MDM2 antagonist Nutlin-3A. Int J Cancer.
2007;121(1):199-205. doi:10.1002/ijc.22643.
195 197. Suzuki T, Urano T, Miki Y, Moriya T, Akahira J, Ishida T, Horie K, Inoue S, Sasano
H. Nuclear cyclin B1 in human breast carcinoma as a potent prognostic factor. Cancer
Sci. 2007;98(5):644-651. doi:10.1111/j.1349-7006.2007.00444.x.
198. Edinger AL, Thompson CB. Death by design: apoptosis, necrosis and autophagy. Curr
Opin Cell Biol. 2004;16(6):663-669. doi:10.1016/j.ceb.2004.09.011.
199. Vogelstein B, Kinzler KW. Cancer genes and the pathways they control. Nat Med.
2004;10(8):789-99. doi:10.1038/nm1087.
200. Yu J, Zhang L, Hwang PM, Kinzler KW, Vogelstein B. PUMA Induces the Rapid
Apoptosis of Colorectal Cancer Cells. Mol Cell. 2001;7(3):673-682.
doi:10.1016/S1097-2765(01)00213-1.
201. Vogelstein B, Lane D, Levine AJ. Surfing the p53 network. Nature.
2000;408(6810):307-310.
202. Kracikova M, Akiri G, George A, Sachidanandam R, Aaronson S a. A threshold
mechanism mediates p53 cell fate decision between growth arrest and apoptosis. Cell
Death Differ. 2013;20(4):576-88. doi:10.1038/cdd.2012.155.
203. Vousden KH, Lu X. Live or let die: the cell’s response to p53. Nat Rev Cancer.
2002;2(8):594-604. doi:10.1038/nrc864.
204. Zhao Y, Yu H, Hu W. The regulation of MDM2 oncogene and its impact on human
cancers. Acta Biochim Biophys Sin (Shanghai). 2014. doi:10.1093/abbs/gmt147.
205. Ohnstad HO, Castro R, Sun J, Heintz K-M, Vassilev LT, Bjerkehagen B, Kresse SH,
Meza-Zepeda L a, Myklebost O. Correlation of TP53 and MDM2 genotypes with
response to therapy in sarcoma. Cancer. 2013;119(5):1013-22.
doi:10.1002/cncr.27837.
196 206. Kojima K, Konopleva M, Samudio IJ, Shikami M, Cabreira-Hansen M, McQueen T,
Ruvolo V, Tsao T, Zeng Z, Vassilev LT, Andreeff M. MDM2 antagonists induce p53-
dependent apoptosis in AML: implications for leukemia therapy. Blood.
2005;106(9):3150-9. doi:10.1182/blood-2005-02-0553.
207. Gu L, Zhu N, Findley HW, Zhou M. MDM2 antagonist nutlin-3 is a potent inducer of
apoptosis in pediatric acute lymphoblastic leukemia cells with wild-type p53 and
overexpression of MDM2. Leukemia. 2008;22(4):730-9. doi:10.1038/leu.2008.11.
208. Tovar C, Rosinski J, Filipovic Z, Higgins B, Kolinsky K, Hilton H, Zhao X, Vu BT,
Qing W, Packman K, Myklebost O, Heimbrook DC, Vassilev LT. Small-molecule
MDM2 antagonists reveal aberrant p53 signaling in cancer: implications for therapy.
Proc Natl Acad Sci U S A. 2006;103(6):1888-93. doi:10.1073/pnas.0507493103.
209. Pishas KI, Neuhaus SJ, Clayer MT, Schreiber AW, Lawrence DM, Perugini M,
Whitfield RJ, Farshid G, Manavis J, Chryssidis S, Mayo BJ, Haycox RC, Ho K, Brown
MP, D’Andrea RJ, Evdokiou A, Thomas DM, Desai J, Callen DF, Neilsen PM. Nutlin-
3a Efficacy in Sarcoma Predicted by Transcriptomic and Epigenetic Profiling. Cancer
Res . 2014;74 (3 ):921-931. doi:10.1158/0008-5472.CAN-13-2424.
210. Paik S, Shak S, Tang G, Kim C, Baker J, Cronin M, Baehner FL, Walker MG, Watson
D, Park T, Hiller W, Fisher ER, Wickerham DL, Bryant J, Wolmark N. A multigene
assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. N Engl
J Med. 2004;351(27):2817-26. doi:10.1056/NEJMoa041588.
211. Bootcov MR, Bauskin AR, Valenzuela SM, Moore AG, Bansal M, He XY, Zhang HP,
Donnellan M, Mahler S, Pryor K, Walsh BJ, Nicholson RC, Fairlie WD, Por SB,
Robbins JM, Breit SN. MIC-1, a novel macrophage inhibitory cytokine, is a divergent
197 member of the TGF-beta superfamily. Proc Natl Acad Sci U S A. 1997;94(21):11514-
9. Available at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=23523&tool=pmcentrez&
rendertype=abstract. Accessed May 16, 2014.
212. Yang H, Filipovic Z, Brown D, Breit SN, Vassilev LT. Macrophage inhibitory
cytokine-1: a novel biomarker for p53 pathway activation. Mol Cancer Ther.
2003;2(10):1023-9. Available at: http://www.ncbi.nlm.nih.gov/pubmed/14578467.
Accessed May 16, 2014.
213. Bauskin AR, Brown DA, Kuffner T, Johnen H, Luo XW, Hunter M, Breit SN. Role of
macrophage inhibitory cytokine-1 in tumorigenesis and diagnosis of cancer. Cancer
Res. 2006;66(10):4983-6. doi:10.1158/0008-5472.CAN-05-4067.
214. Aziz MH, Shen H, Maki CG. Acquisition of p53 mutations in response to the non-
genotoxic p53 activator Nutlin-3. Oncogene. 2011;30(46):4678-4686.
215. Rufini A, Melino G. Cell death pathology: the war against cancer. Biochem Biophys
Res Commun. 2011;414(3):445-50. doi:10.1016/j.bbrc.2011.09.110.
216. Klionsky DJ. Autophagy. Curr Biol. 2005;15(8):R282-3.
doi:10.1016/j.cub.2005.04.013.
217. Crighton D, Wilkinson S, O’Prey J, Syed N, Smith P, Harrison PR, Gasco M, Garrone
O, Crook T, Ryan KM. DRAM, a p53-induced modulator of autophagy, is critical for
apoptosis. Cell. 2006;126(1):121-34. doi:10.1016/j.cell.2006.05.034.
218. Lopez G, Torres K, Liu J, Hernandez B, Young E, Belousov R, Bolshakov S, Lazar AJ,
Slopis JM, McCutcheon IE, McConkey D, Lev D. Autophagic Survival in Resistance
to Histone Deacetylase Inhibitors: Novel Strategies to Treat Malignant Peripheral
198 Nerve Sheath Tumors. Cancer Res . 2011;71 (1 ):185-196. doi:10.1158/0008-
5472.CAN-10-2799.
219. Radosevic N, Delmer A, Tang R, Marie J, Ajchenbaum-Cymbalista F. Cell cycle
regulatory protein expression in fresh acute myeloid leukemia cells and after drug
exposure. Leukemia. 2001;(4):559-66.
220. Del Poeta G, Venditti A, Del Principe MI, Maurillo L, Buccisano F, Tamburini A, Cox
MC, Franchi A, Bruno A, Mazzone C, Panetta P, Suppo G, Masi M, Amadori S.
Amount of spontaneous apoptosis detected by Bax/Bcl-2 ratio predicts outcome in
acute myeloid leukemia (AML): Presented in part at the 42nd annual meeting of the
American Society of Hematology, San Francisco, CA, December 1-5, 2000.46 . Blood
. 2003;101 (6 ):2125-2131. doi:10.1182/blood-2002-06-1714.
221. Kranz D, Dobbelstein M. Nongenotoxic p53 activation protects cells against S-phase-
specific chemotherapy. Cancer Res. 2006;66(21):10274-80. doi:10.1158/0008-
5472.CAN-06-1527.
222. Carvajal D, Tovar C, Yang H, Vu BT, Heimbrook DC, Vassilev LT. Activation of p53
by MDM2 antagonists can protect proliferating cells from mitotic inhibitors. Cancer
Res. 2005;65(5):1918-24. doi:10.1158/0008-5472.CAN-04-3576.
223. Hirota S, Isozaki K, Moriyama Y, Hashimoto K, Nishida T, Ishiguro S, Kawano K,
Hanada M, Kurata A, Takeda M, Muhammad Tunio G, Matsuzawa Y, Kanakura Y,
Shinomura Y, Kitamura Y. Gain-of-Function Mutations of c-kit in Human
Gastrointestinal Stromal Tumors. Science (80- ). 1998;279(5350):577-580.
doi:10.1126/science.279.5350.577.
199 224. Reichardt P, Joensuu H, Blay J-Y. New fronts in the adjuvant treatment of GIST.
Cancer Chemother Pharmacol. 2013:715-723.
225. Muler J, Baker L, Zalupski MM. Gastrointestinal stromal tumors: chemotherapy and
imatinib. Curr Oncol Rep. 2002;4(6):499-503.
226. Wardelmann E, Schildhaus H-U, Merkelbach-Bruse S, Hartmann W, Reichardt P,
Hohenberger P, Büttner R. Soft tissue sarcoma: from molecular diagnosis to selection
of treatment. Pathological diagnosis of soft tissue sarcoma amid molecular biology and
targeted therapies. Ann Oncol. 2010;21 Suppl 7(Supplement 7):265-9.
doi:10.1093/annonc/mdq381.
200