Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Modification of RHAMM and TPX2 optimizes Aurora kinase A (AURKA) inhibition in malignant peripheral nerve… Mohan, Pooja 2013

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

Item Metadata

Download

Media
24-ubc_2013_spring_mohan_pooja.pdf [ 2.85MB ]
Metadata
JSON: 24-1.0073680.json
JSON-LD: 24-1.0073680-ld.json
RDF/XML (Pretty): 24-1.0073680-rdf.xml
RDF/JSON: 24-1.0073680-rdf.json
Turtle: 24-1.0073680-turtle.txt
N-Triples: 24-1.0073680-rdf-ntriples.txt
Original Record: 24-1.0073680-source.json
Full Text
24-1.0073680-fulltext.txt
Citation
24-1.0073680.ris

Full Text

MODIFICATION OF RHAMM AND TPX2 OPTIMIZES AURORA KINASE A (AURKA) INHIBITION IN MALIGNANT PERIPHERAL NERVE SHEATH TUMOURS  by Pooja Mohan B.Sc., the University of Toronto, 2010  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIRMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate Studies (Experimental Medicine)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) April 2013 © Pooja Mohan, 2013  Abstract Malignant peripheral nerve sheath tumours (MPNST) are rare, hereditary cancers associated with neurofibromatosis type I. MPNSTs lack effective treatments as they often resist chemotherapies and have high rates of disease recurrence. Published analysis of copy number variation identified hemizygous loss of Hyaluronan Mediated Motility Receptor (HMMR, encodes RHAMM) in half of the examined high-grade MPNST, but not in benign neurofibromas or low grade tumours. RHAMM is a molecular brake for the mitotic kinase Aurora A (AURKA), so this loss of HMMR in high-grade MPNST may cause tumours to rely on AURKA activity and sensitizes them to aurora kinase inhibitors (AKI). Three MPNST cell-lines were profiled for the expression and activity of AURKA, as well as their responses to three AKI. The sensitivity of cell-lines with amplification of AURKA was reliant upon kinase activity, which correlated with the expression of the regulatory gene products TPX2 and RHAMM. Silencing of RHAMM, but not TPX2, increased AURKA activity and sensitized MPNST cells to AKI. All three AKIs reduced kinase activity in a dose-dependent manner, and AKI treatment induced cellular responses such as apoptosis, endoreduplication and cellular senescence. Additionally, two primary human MPNSTs grown in vivo as xenotransplants were treated with the AURKA-specific inhibitor MLN8237. Treatment resulted in tumour cells exiting the cell cycle and undergoing endoreduplication, which cumulated in stabilized disease. The MPNST cell-line S462 has a population of tumorigenic stem-like cells that can be grown in sphere culture. AURKA activity was critical to the propagation and self-renewal of sphere-enriched MPNST stem-like cells. AKI treatment significantly reduced the formation of spheroids, attenuated the self-renewal of spheroid forming cells, and promoted their differentiation. Silencing of TPX2 decreased AURKA activity, while silencing of RHAMM was ii  sufficient to endow MPNST cells with an ability to form and maintain sphere culture. Collectively, our data indicate that AURKA is a rationale therapeutic target for MPNST, and tumour cell responses to AKI, which include differentiation, are modulated by the abundance of RHAMM and TPX2.  iii  Preface  Parts of this work were published: P Mohan, J Castellsague, J Jiang, K Allen, H Chen, O Nemirovsky, M Spyra, K Hu, L Kluwe, MA Pujana, A Villanueva, VF Mautner, J Keats, SE Dunn, C Lazaro and CA Maxwell. Genomic imbalance of HMMR/RHAMM regulates the sensitivity and response of malignant peripheral nerve sheath tumour cells to aurora kinase inhibition. Oncotarget 2013; (epub Jan 9 2013). Comparative genomic hybridization studies, cell-lines and in vivo work were provided by co-authors, but I was responsible for the majority of the in vitro work in this paper and the writing of the manuscript.  Parts of the work presented in Chapter 3 were done in collaboration with different laboratories. The comparative genomic hybridization experiment was done in collaboration with Dr. Jonathan Keats’ laboratory (TGEN, USA) by Kristi Allen. Immunofluorescence experiments in HeLa cells displaying aurora kinase A inhibition by VX680 were done by Helen Chen (Dr. Christopher Maxwell’s lab). The in vivo models in Chapter 3 were done in collaboration with Dr. Conxi Lazaro’s laboratory (IDIBELL, Spain) by Joan Castellsague. I was responsible for all analyses done on the data. Furthermore, the S462 in vivo models were done in collaboration with Dr. Gregor Reid (UBC). I was responsible for dissociation and profiling of the tumours collected. All animal handling was done by Dr. Reid.  iv  Table of Contents Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iv Table of Contents .......................................................................................................................... v List of Tables ............................................................................................................................... vii List of Figures ............................................................................................................................. viii List of Abbreviations ................................................................................................................... ix Acknowledgements ..................................................................................................................... xii Dedication ................................................................................................................................... xiii Chapter 1: Introduction ............................................................................................................... 1 1.1 Neurofibromatosis associated Malignant Peripheral Nerve Sheath Tumours (NF1-MPNST) ..................................................................................................................................................... 1 Neurofibromatosis ................................................................................................................... 1 Neurofibromin ......................................................................................................................... 2 Neurofibroma formation and progression ............................................................................... 2 MPNST .................................................................................................................................... 3 1.2 MPNST progression, Ras activation, and Aurora kinase activity ........................................ 4 1.3 AURKA .................................................................................................................................. 5 Roles of AURKA in mitosis .................................................................................................... 6 1.4 AURKA regulators ................................................................................................................ 7 Polo-like kinase 1 (PLK1) ....................................................................................................... 8 TPX2 ........................................................................................................................................ 8 RHAMM .................................................................................................................................. 9 1.5 Non-mitotic roles of AURKA ............................................................................................... 12 1.6 Aurora kinase A inhibitors .................................................................................................. 13 MLN8237/ Alisertib .............................................................................................................. 13 VX680 ................................................................................................................................... 14 1.7 Aim of study and hypothesis ................................................................................................ 15 Chapter 2: Material and Methods ............................................................................................. 16 v  Adherent culture and culture as spheres ................................................................................ 16 Array comparative genomic hybridization (a-CGH) ............................................................. 17 Genomic and reverse-transcriptase PCR and real-time PCR ................................................ 18 siRNA and small molecule reagents ...................................................................................... 19 Human MPNST tumour explant models ............................................................................... 20 Cell based assays ................................................................................................................... 20 Immunofluorescence, immunohistochemistry, and immunoblot analyses ............................ 22 S462 sphere xenotransplant models ...................................................................................... 25 Lentivirus mediated shRNA knockdown and generation of sub-lines .................................. 26 Statistics ................................................................................................................................. 27 Chapter 3: Results....................................................................................................................... 28 3.1 MPNST cell-lines have gene dose alterations in HMMR, TPX2 and AURKA .................... 28 Conclusion (Section 3.1) ........................................................................................................... 33 3.2 AURKA activity is necessary for the growth of MPNST cells in vitro ................................ 35 Conclusion (Section 3.2) ........................................................................................................... 38 3.3 Aurora kinase inhibition is an encouraging pre-clinical treatment for human MPNST .... 38 Conclusion (section 3.3) ............................................................................................................ 41 3.4 Cellular responses to AKI in MPNST cells in vitro ............................................................ 41 Conclusion (Section 3.4) ........................................................................................................... 44 3.5 AKI results in impaired self-renewal of MPNST cancer stem-like cells ............................. 44 Conclusion (section 3.5) ............................................................................................................ 49 3.6 Loss of RHAMM modifies the sensitivity to AKI and sphere formation for MPNST cells .. 51 Chapter 4: Discussion and Conclusions .................................................................................... 57 4.1 The RHAMM-TPX2-AURKA pathway is critical to MPNST growth and survival ............. 57 4.2 Cellular responses to AKI ................................................................................................... 59 4.3 AURKA is a molecular brake on differentiation programs................................................. 60 Bibliography ................................................................................................................................ 63 Appendices ................................................................................................................................... 74 Appendix A: S462 MT1 cells are 97% human........................................................................... 74 Appendix B: CGH of genomic region coding for TP53 ............................................................ 75  vi  List of Tables Table 1: NF1 diagnosis ....................................................................................................................1 Table 2: Primers, shRNA and siRNA sequences ...........................................................................17 Table 3: Origins and profiles of MPNST cell-lines .......................................................................27  vii  List of Figures Figure 1: The TPX2-AURKA-RHAMM pathway ........................................................................11 Figure 2: Growth kinetics of MPNST cell-lines ............................................................................15 Figure 3: Comparative genomic hybridization in MPNST cells ...................................................28 Figure 4: Copy number for HMMR, TPX2 and AURKA in MPNST cell lines ..............................29 Figure 5: Genomic alterations in S462 and 2884 cell-lines ...........................................................30 Figure 6: Expression of RHAMM and TPX2 reflect genomic alterations in their genes ..............31 Figure 7: S462 cells have increased kinase activity.......................................................................33 Figure 8: Growth of MPNST cell lines is dependent upon AURKA expression ..........................34 Figure 9: Aurora kinase inhibitors are effective in decreasing kinase activity ..............................35 Figure 10: Treatment with AKI reduced cell viability in MPNST cell lines .................................36 Figure 11: MLN8237 is effective against primary MPNSTs grown as xenotransplants in vivo ...38 Figure 12: MLN8237 treated human MPNSTs undergo endoreduplication .................................39 Figure 13: AKI treatment results in apoptosis and endoreduplication in MPNST cell-lines ........41 Figure 14: VX680 treatment causes MPNST cells to become senescent ......................................42 Figure 15: S462 cells have a population of stem-like cells that form spheres in vitro ..................44 Figure 16: S462 sphere cells are more tumorigenic than adherent S462 cells ..............................45 Figure 17: S462 sphere cells have elevated AURKA activity ......................................................46 Figure 18: AURKA is required for self-renewal of stem-like cells in MPNST ............................47 Figure 19: MLN8237 treatment results in neuronal differentiation in MPNST stem-like cells ....49 Figure 20: TPX2 and RHAMM silenced sub-lines have similar growth kinetics to non-hairpin controls ..........................................................................................................................................51 Figure 21: TPX2 down regulation did not significantly lower IC-50s of S462 cells ....................51 Figure 22: RHAMM down regulation increases sensitivity of cells to AKI and AURKA activity in 2884 cells ...................................................................................................................................52 Figure 23: Modulation of RHAMM and TPX2 affects sphere formation in MPNST cells ..........53 Figure 24: TPX2 silenced cells have reduced AURKA activity ...................................................54 Figure 25: Tumour cells recovered from mice are human .............................................................74 Figure 26: CGH of TP53 in MPNST cell-lines .............................................................................75  viii  List of Abbreviations AKI  Aurora Kinase A Inhibitor  AURKA  Aurora Kinase A  AURKB  Aurora Kinase B  BRCA1  Breast Cancer 1, early onset protein  BSA  Bovine Serum Albumin  cDNA  Complementary DNA  CGH  Comparative Genomic Hybridization  DAPI  4’,6-diamidino-2-phenylindol  DMEM  Dulbecco’s Modified Eagle Medium  DMSO  Dimethyl Sulfoxide  DNA  Deoxyribonucleic acid  EGF  Epidermal Growth Factor  ERK  Extracellular signal-regulated kinase  FACS  Fluorescence-activated cell sorting  FBS  Fetal Bovine Serum  FGF  Fibroblast Growth Factor  GAP  GTPase accelerating protein  GRD  GAP-related domain  GTP  Guanine triphosphate  HER-2  human epidermal growth factor receptor  HMMR  Hyaluronan Mediated Motility Receptor (gene)  MEK  Mitogen-activated protein kinase  mESC  mouse Embryonic Stem Cells  MPNST  Malignant Peripheral Nerve Sheath Tumours ix  MTT  3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2Htetrazolium  NF1  Neurofibromatosis type I (disease)  NF1  Neurofibromin 1 (gene)  NF1-MPNST Neurofibromatosis type I associated Malignant Peripheral Nerve Sheath Tumours NHP  non-hairpin  NOD-SCID  Non-obese diabetic mice with a severe combined immunodeficiency mutation  NSG  NOD-SCID gamma mice  PBS  Phosphate Buffered Saline  PI  Propidium Iodide  PLK1  Polo-like kinase 1  PNS  Peripheral Nervous System  qPCR  Quantitative Polymerase Chain Reaction  qRT-PCR  Quantitative Reverse-Transcriptase Polymerase Chain Reaction  RHAMM  Receptor for Hyaluronan Mediated Motility (protein)  RNA  Ribonucleic acid  SA-β-Gal  Senescence activated β-galactosidase  SD  Standard deviation  SDS  Sodium dodecyl sulfate  SEM  Standard error mean  shRNA  Small hairpin RNA  siRNA  Small interfering RNA  SNP  Single Nucleotide Polymorphism  SP-MPNST  Sporadic Malignant Peripheral Nerve Sheath Tumours  TBP  TATA-Binding Protein  x  TPX2  Targeting protein for Xklp2  Tuj1  Neuron-specific class III β-tubulin  X-Gal  5-bromo-4-chloro-indolyl-β-D-galactopyranoside  XRHAMM  Xenopus RHAMM  xi  Acknowledgements I was recently told that one of the most important things to have during a graduate degree is a good mentor. I am lucky to have a great one. I would like to thank Dr. Christopher Maxwell for his guidance, encouragement and support. Working with Chris has been a rewarding experience that has resulted in both, personal and professional growth. I am very grateful to have had this opportunity to learn from him. I would also like to thank my supervisory committee, Dr. Jan Friedman and Dr. Sandra Dunn for their advice and insight into my research. I am grateful to Dr. Conxi Lazaro and Joan Castellsague for their input and guidance in this research. They are an integral part of this work. I would also like to thank Dr. Jonathan Keats and Dr. Gregor Reid for their assistance with this research. There are some very special people in the Maxwell Lab who have been with me every step of the way. I am very grateful to Dr. Jihong Jiang for her help and advice with all the experimental problems I have faced during my degree. I would also like to acknowledge all the trainees in mine and in neighbouring labs for all their encouragement and much needed comic relief. Thank you for getting me through the rough days. Last but not least, I would like to thank my family for their unwavering support. I owe everything, especially my sanity, to the three of you.  xii  Dedication This thesis is dedicated to every person who has encouraged and motivated me during the last three years, especially my parents.  xiii  Chapter 1: Introduction 1.1 Neurofibromatosis associated Malignant Peripheral Nerve Sheath Tumours (NF1MPNST) Neurofibromatosis Neurofibromatosis type I (NF1) is an autosomal dominant disease that is caused by the loss of the Neurofibromin 1 gene.1 NF1 occurs in ~ 1/ 3000 people, and the incidence appears to be higher among children under 9 years of age, presumably due to the early death of NF1 patients.1 Clinical manifestations of the disease include skin pigmentation, iris Lisch nodules and the occurrence of benign peripheral nerve sheath tumours called neurofibromas.2 Other symptoms include learning disabilities, vascular disease, and tumours of the central nervous system. Diagnosis of NF1 is based on the presence of two or more of the features listed in table 1.2 Table 1: NF1 diagnosis criteria Feature Description Six or more café au lait These are dark spots on the skin. They have to be 5mm or macules greater in pre-pubertal and 15mm or greater in post pubertal individuals to count as diagnostic features of NF1. Two or more neurofibromas Neurofibromas arise from Schwann cells and can occur of any type or one plexiform throughout the peripheral nervous system (PNS). Localized neurofibroma shallow, skin-associated neurofibromas are called discrete cutaneous neurofibromas. These are only associated with one peripheral nerve or nerve branch.3 Plexiform neurofibromas are deeper and can be more diffuse tumours that involve multiple nerve fascicles.2,3 Freckling in the axillary or freckles in areas that are not exposed to the sun. inguinal regions An optic glioma Benign tumours affecting the optic nerve Two or more Lisch nodules Small yellow-brown raised lesions on the surface of the iris.4 A distinctive osseous lesion Focal lesions in the bone such as tibial dysplasia, vertebral dysplasia or sphenoid wing dysplasia 1  A first degree relative with NF1 Neurofibromin The neurofibromin 1(NF1) gene is located on chromosome 17q11.2 and codes for the protein neurofibromin.5,6 Neurofibromin is a tumour suppressor and prevents tumour formation by regulating the oncogenic Ras protein family.2 Neurofibromin has a GAP-related domain (GRD) that promotes the activity of GTPase accelerating proteins, GAPs.7,8 GAPs convert active GTP-Ras to inactive GDP-Ras to reduce downstream Ras signalling.9 The oncogenic Ras signalling pathway promotes cell growth and survival.10,11 Ras over-expression has been linked to enhanced tumour cell growth, invasiveness and angiogenesis.10 Therefore, the regulation of Ras by NF1 is vital to normal cellular processes and development.12 Indeed, the inactivation of NF1 and the resulting hyper-activation of Ras signalling is thought to be one of the causes of neurofibroma formation.13,14  Neurofibroma formation and progression Neurofibromas formation requires both genetic alterations and the recruitment of additional cell types such as perineural, fibroblast and mast cells15. Tumours may have loss of heterozygosity of NF1 but this is not necessary for tumour formation. Indeed, mouse models have shown that Nf1  -/-  status in Schwann cells is insufficient for tumour formation and  progression and additional environmental and genetic events need to occur. For example, without the recruitment of Nf1 +/- mast cells, and to some extent Nf1+/- fibroblasts and perineural cells, neurofibromas will not arise in vivo.16,17 Additionally, genetic events modulating oncogenes and tumour suppressors such as TP53 may drive neurofibroma formation and  progression to  2  malignant states.18–20 Therefore, the genetic and cellular environments both contribute to neurofibroma progression. In some cases, neurofibromas will progress to malignant peripheral nerve sheath tumours (MPNSTs). The majority (~70%) of these MPNSTs have deletions in both copies of NF1.15,21  MPNST The most frequent malignant neoplasm associated with NF1 is MPNST.2 These occur in about 10-13% of NF1 patients at some time during their life and typically arise from plexiform neurofibromas.22 MPNSTs are otherwise rare, aggressive tumours that are often found within nerve roots and bundles.23,24 While these tumours are primarily composed of Schwann cells, they are heterogeneous and also include mast and perineural cells. Approximately half of the MPNSTs diagnosed are associated with Neurofibromatosis I (NF1-MPNST), with the remaining cases arising sporadically (SP-MPNST).24 Tumours of either origin are similar in pathology and share similar expression signatures; however, age at onset is usually much earlier and prognosis is worse in NF1 patients.9  Accurate early diagnosis of MPNSTs is challenging as low-grade MPNSTs often resemble benign plexiform neurofibromas in the clinic.22,24 Unfortunately, high grade MPNSTs, which account for 85% of NF1-associated cases, are difficult to treat. These tumours tend to be resistant to chemotherapy. Current treatments include single agent doxorubicin (an intercalating agent) or a combination of doxorubicin and ifosfamide (an alkylating agent), but these treatments are not curative and, at best, the response rate is ~25-30%.22 Furthermore, surgical removal of 3  the tumour in NF1 patients is often incomplete due to its large and diffuse nature. While radiotherapy can provide local control for tumour edges that remain behind after surgery, longterm survival remains unimproved.22  1.2 MPNST progression, Ras activation, and Aurora kinase activity Multiple studies have profiled common genetic alterations in MPNSTs in order to identify molecular targets for new treatment strategies.18,20,21,25–27 One such study utilized a mouse model with a constitutively active, Schwann cell-specific allele of H-Ras.28 This allele of Ras has been shown to drive proliferation of Nf1-/-cells specifically and is expressed in both progenitor and mature Schwann cells. The transcriptome of these cells was compared to control cells, and 308 genes were found to be differentially expressed, highlighting a population of potential Schwann cell-specific Ras pathway genes. The expression of these candidates was then validated in mouse and human neurofibromas and MPNSTs. One candidate was amplification of the aurora kinase A (AURKA) pathway. The AURKA gene was amplified along with its major downstream targets, including polo-like kinase 1 (Plk1) and targeting protein for XKlp2 (TPX2), the major activating protein for AURKA.28,29 Using SNP and qPCR analysis, it was shown that 5/5 MPNST cell lines and 62% of primary MPNSTs (n=13) also had copy number gains in AURKA.28 Therefore, gene-dose and expression of AURKA are common changes in MPNSTs and may contribute towards MPNST progression.28 Consistent with these findings, the region that codes for AURKA and TPX2, chromosome 20q, has also been shown to be amplified in MPNST.25,26,30  4  The activation of AURKA is reliant upon its access to the activating protein TPX2, which is regulated through its partner protein, termed RHAMM (encoded by HMMR)31–34. Interestingly, 52% of aggressive, high-grade NF1-MPNSTs, but not benign neurofibromas or low-grade MPNSTs, contain deletions in the Hyaluronan Mediated Motility Receptor (HMMR) gene.21 As the gene product, RHAMM, has been described as a negative feedback regulator for AURKA,31 Ras induced up-regulation of AURKA combined with genetic changes to kinase regulators suggest MPNST cells may have increased AURKA activity and be sensitive to small-molecule AURKA inhibitors.  1.3 AURKA AURKA belongs to the family of serine-threonine kinases called aurora. These kinases regulate microtubule-associated processes throughout mitosis. Aurora was discovered in Drosophila through a genetic screen. Mutation of the kinase resulted in monopolar spindles that, when imaged, resembled the aurora borealis on the horizon, giving rise to the name “aurora”.35 Since then, homologues of the kinase have been found in vertebrates and invertebrates. In humans, the family consists of 3 kinases, aurora kinase A, B and C.36 AURKA and aurora kinase B (AURKB) are highly expressed at the G2-M phase of most normal cell types, but aurora kinase C is specifically expressed in the testis. Importantly, AURKA is a bonafide oncogene and its overexpression in NIH-3T3 cells was sufficient for tumour formation in nude mice.37–39 Moreover, AURKA is amplified and overexpressed in several human cancers such as breast,  5  pancreatic and ovarian.38,40–44 Overexpression of AURKA occurs early in tumour development44,45 and is thought to lead to genomic instability in cancer cells.36 Roles of AURKA in mitosis The disruption of AURKA function, through small-molecule inhibition and genetic or epigenetic manipulation, leads to a number of mitotic phenotypes, including delayed G2-M progression, aberrant mitotic spindle formation, mitotic failure or slippage, and aneuploidy. These phenotypes are related to the critical role that AURKA plays in the organization and assembly of microtubules during mitosis.  Mitosis is a fundamental process that ensures equal segregation of genetic material into two identical daughter cells. It is split into five phases based upon the dynamics of the chromosomes. First, in prophase, DNA condenses into chromosomes and the centrosomes (microtubule organizing centres) start to migrate to each pole of the cell in preparation for spindle formation.46 Second, the nuclear membrane breaks down and microtubules start to form a bipolar spindle structure during prometaphase. Next, in metaphase, chromosomes align between the two spindle poles. At this point, the mitotic spindle checkpoint is activated to ensure that all chromatids are attached to each spindle pole. In the last two phases, paired chromosomes separate and are pulled to each spindle pole, where nuclear membranes enclose each set of divided chromosomes to give two daughter nuclei.46 AURKA is vital to the preparation for mitosis and the first three phases of mitosis: prophase, pro-metaphase and metaphase.36  6  Centrosome maturation and spindle assembly To prepare for mitosis, protein complexes are recruited to the centrosome to increase its ability to nucleate microtubules and form a mitotic spindle.47 This process is termed centrosome maturation. AURKA is required to recruit these spindle proteins and ensure separation of centrosomes.48–50 AURKA is also necessary for mitotic entry and spindle assembly51–53, and the kinase must be activated to commit the cell to mitosis.50,53–55 Without these events the cell cannot complete bipolar spindle assembly and will undergo aberrant mitosis, which may result in either apoptosis or aneuploidy and genetic instability.47  Taken together, AURKA signalling is vital for cell division, including spindle assembly and the equal segregation of DNA.48,56–59 Therefore, when AURKA is depleted in the cell, expected phenotypes include failed centrosome separation resulting in monopolar or abnormal spindles, mitotic delays and either apoptosis or unequal segregation of DNA and consequent aneuploidy.35,48,53,60–62  1.4 AURKA regulators Since AURKA signalling is vital to mitosis, a key cellular process, it stands to reason that the kinase is tightly regulated to prevent genomic instability and mitotic failure.63–66 Briefly, AURKA is turned on by self-phosphorylation of its threonine-288 residue during the early stages of mitosis.67 Different activating proteins bind to the kinase in a location and phase specific manner to regulate its downstream functions. After metaphase, AURKA levels are decreased, 7  which allows the cell to disassemble the spindle and form two daughter cells. To turn the kinase off, protein phosphatases remove the protective phosphate group and increase the probability of AURKA degradation by the proteasome.27,68–73 AURKA has several regulators but only the three major ones, PLK1, TPX2 and RHAMM will be discussed here.  Polo-like kinase 1 (PLK1) PLK1 is an oncogenic serine/ threonine kinase that is required for mitotic progression.74,75 AURKA and PLK1 regulate each other through phosphorylation events. PLK1 recruits AURKA to centrosomes, and AURKA, in turn, phosphorylates and activates PLK1.29,75,76 To close the cycle, eventually, PLK1 inactivates hBora, an AURKA activating protein, to feed back on the pathway and decreases AURKA activity at the centrosome.  TPX2 TPX2 is the most characterized regulator of AURKA activity. It is both necessary and sufficient for optimal kinase activity, a capability no other co-factor has been shown to have.72,77 Normally, TPX2 is prevented from associating with AURKA through interactions at the nuclear membrane. However, during mitosis, a Ran-mediated event releases TPX2, allowing TPX2 to bind and form a heterodimeric complex with AURKA.78–82 This complex not only localizes AURKA to the spindle and spindle poles but also stabilizes and protects p-AURKA (activated kinase) from dephosphorylation and downstream degradataion.72,77,83–89 Therefore, TPX2 binding allows for the accumulation and continuous activation of the kinase to ensure bipolar spindle formation and, therefore, equal segregation of the cell’s DNA. Furthermore, studies have shown 8  that both proteins are over-expressed in cancer models, and that TPX2 binding may be required for the increase in AURKA activity in cancer, spurring the name oncogenic holo-enzyme for the heterodimer.90  Just as TPX2 binding is important to AURKA activation, its removal is important for AURKA homeostasis. In fact, in anaphase, TPX2 levels decrease so that AURKA can be degraded and the cell can exit mitosis.83,90 There are two possible ways that TPX2 can be removed from its interaction with AURKA: the first is degradation of the complex, and the second is capture of TPX2 so that it is no longer available to bind to the kinase. The protein, RHAMM is heavily implicated in the latter process.  RHAMM The Receptor for Hyaluronan Mediated Motility (RHAMM) is a microtubule-associated protein that is coded for by HMMR on chromosome 5q33.2-qter.91 During mitosis, RHAMM localizes along spindle fibres to maintain mitotic spindle integrity.32–34,91,92 Alteration of RHAMM expression leads to abnormal spindles, mitotic delays and inappropriate chromosome separation.  RHAMM is thought to regulate AURKA activity by sequestering TPX2, rendering it inaccessible to the kinase (Fig 1). There are several lines of evidence for this. First, RHAMM has been shown to co-localize and associate with TPX2.32–34 In Xenopus, RHAMM can both enable and block the correct localization of TPX2,34 and presumably alter AURKA activity 9  downstream. When RHAMM was removed from Xenopus extracts, TPX2 was not correctly localized during mitosis.34 Moreover, when a fragment of RHAMM was added in abundance to these extracts, TPX2 was not correctly localized, which implies feedback inhibition on TPX2 location through RHAMM.34 More recently, a study from our lab showed that phosphorylated RHAMM is an AURKA substrate that may participate in feedback regulation of AURKA.31 Depletion of RHAMM changed the location of TPX2 and increased AURKA activity in vitro, implicating RHAMM as a negative regulator for AURKA- TPX2 heterodimeric complexes and kinase activity (Fig 1).31  10  Figure 1: The TPX2-AURKA-RHAMM pathway. Normally, TPX2 is held in the nucleus but its nuclear transport enables a complex with AURKA and activates the kinase. In turn, AURKA phosphorylates downstream substrates, one of which is the protein RHAMM. Phosphorylated RHAMM is thought to prevent the nuclear export of TPX2, or to promote its import, which prevents further AURKA-TPX2 complexes, and turns the kinase off. The diagram and the model are being used with permission and are adapted from Maxwell et al (2011).31  Like AURKA and TPX2, RHAMM is also heavily implicated in cancer susceptibility and progression. RHAMM is a breast cancer susceptibility gene,93 and common genetic variation in HMMR modifies the risk to develop breast cancer in carriers of BRCA1 mutations.31 In addition, RHAMM overexpression is linked to a variety of tumours including multiple myeloma, where it was linked to poor prognosis.94,95 In MPNSTs, RHAMM may be playing a tumour suppressive role. In an array-CGH study of 35 MPNSTs, there was hemizygous loss of HMMR in 52% of high grade, but not benign or low grade, tumours.21 Combined with the genomic amplification of  11  regions containing AURKA and TPX2 25,26,28,30, the removal of the RHAMM feedback may allow unchecked kinase activity in MPNSTs31. Therefore, we hypothesize that these MPNSTs may be oncogene addicted to AURKA activity, rendering these tumours susceptible to small molecule inhibitors against the kinase.  1.5 Non-mitotic roles of AURKA Traditionally, AURKA is known as a regulator of mitosis, but recently, several nonmitotic roles have been found that suggest it may be playing roles in differentiated cells as well. For example, the kinase mediates the disassembly of the primary cilium, a sensory organelle that is present on differentiated, non-mitotic and non-motile cells.96,97 Furthermore, AURKA has been shown to play a role in early development processes such as neurite extension in immature neurons and synaptic differentiation.98–100 Finally, recent evidence suggests that AURKA may be playing critical roles in the maintenance of pluripotency in embryonic stem cells and/or the reprogramming of differentiated cells to induced-pluripotent stem cells.101  12  1.6 Aurora kinase A inhibitors Due to the oncogenic properties and high levels of expression of AURKA in a multitude of cancers, it has been highlighted as a therapeutic target. There are currently more than 30 small molecule inhibitors for AURKA in development, where one in particular, MLN8237, is in phase III trials for refractory peripheral T cell lymphoma.102,103  MLN8237/ Alisertib MLN8237 or alisertib was developed by Millennium pharmaceuticals and is 200 times more specific to AURKA than AURKB with an in vitro IC-50 of 1.2nM and an in vivo IC-50 ranging from ~ 6 - 400nM depending on the cell-line.104 Other kinases are only inhibited at concentrations of 1 μM or higher. MLN827 has successfully completed phase I studies against pediatric refractory solid tumours and phase I/ II trials against advanced adult solid tumours.105– 107  Studies have shown that the main dose-limiting toxicity of MLN8347 is grade three and four  neutropenia, followed by hypertension, mucositis and stomatitis.102 Furthermore, the drug by itself, and in combination with other chemotherapies, has been shown to be effective in mouse models of several cancer cell-lines such as breast cancer and multiple myeloma.28,104,108–113 In all these studies utilizing MLN8237, treatment has resulted in a mitotic delay at the G2/M checkpoint,46 which can result in p53 mediated apoptosis, or mitotic slippage. In mitotic slippage, the cell escapes the spindle assembly checkpoint, fails mitosis and remains polyploid in a process termed endoreduplication. This cell then goes through further mitosis or undergoes senescence.46,110  13  VX680 Another small molecule inhibitor, VX680, also showed promise as a cancer therapeutic and was the first Aurora kinase inhibitor (AKI) to enter clinical trials. VX680 is a pan-Aurora inhibitor with varying specificities for each family member. It targets both AURKA and AURKB,114 and caused tumour regression in mice models of solid tumours such as pancreatic and colon cancer. VX680, like many other AKIs, results in characteristic responses such as tumour regression or stabilized disease and cellular responses included apoptosis, senescence and endoreduplication or polyploidy. Endoreduplication and polyploidy are more common due to the critical role for AURKB at the transition between metaphase and anaphase, called the spindle assembly checkpoint.115 The loss of AURKB at this time enables slippage and promotes endoreduplication. However, VX680 never made it past phase I clinical trials due to severe heart related toxicities.102 Currently, a second generation version of this molecule, VE465 is undergoing preclinical trials.116  14  1.7 Aim of study and hypothesis Hypothesis: The hemizygous loss of RHAMM in MPNSTs may augment AURKA activity, which in turn, may drive the growth of tumours. Therefore, these tumours may be particularly sensitive to AURKA inhibition.  Aims 1. Determine the efficacy of AURKA inhibition on MPNST cell viability and downstream cellular responses in vitro and in vivo. 2. Determine if RHAMM and TPX2 modulate the efficacy of kinase inhibition in MPNSTs.  15  Chapter 2: Material and Methods Adherent culture and culture as spheres MPNST tumour cell-lines used in this study (S462, 2884, and 2885) were obtained from Drs. VF Mautner and L Kluwe (University Hospital Eppendorf, Germany) or the ATCC, and were cultured in DMEM High Glucose (Thermo-Fisher) with 10% Fetal Bovine Serum and 20 U/ml penicillin and 20μg/ml streptomycin (Life Technologies). Cells were grown in 37oC with 5% CO2. Unless noted otherwise, cells were seeded at 1.0 x 104 (S462 cell-line) or 3.0 x 104 (2884 and 2885 cell-lines) cells/well in 24-well culture plates and allowed to adhere and grow for 24 hours prior to the denoted treatment. These seeding densities resulted in equivalent growth kinetics for S462 and 2884 cells (Fig 2).  Figure 2: Growth kinetics of MPNST cell lines. A. MPNST cell lines S462 and 2884 undergo equivalent growth while 2885 cells experience slow proliferation. Seeding densities for S462 and 2884 were modified to enable equivalent proliferation. Cell viability was measured by MTT assays over a 4 day period and values were normalized to day 1.  16  To culture MPNST cell-lines as spheres, cells were plated in low attachment plates (Costar) at 3.0 x 104 cells/well in NeuroCult NS-A proliferation medium with 20 ng/ ml EGF, 10 ng/ml FGF and 0.0002% heparin (Stem Cell Technologies). To passage spheres, cells were collected by centrifugation at 1000 rpm for 7 minutes and then mechanically dissociated by pipetting 50 times gently, and aggregates were removed through filtration with a 40 μm cell strainer. For sphere formation assays, cells were seeded at 3000 cells/well in supplemented NeuroCult Medium and allowed to grow for 6 days (Passage 0). Spheres were counted, and then passaged with 3000 cells being seeded for another 6 days, and then spheres were counted again (Passage 1). For formation assays including AKIs, wells were treated with 100 nM MLN8237 or VX680 and an equivalent volume of DMSO (0.001%) as a vehicle-alone control. Array comparative genomic hybridization (a-CGH)i Genomic DNA was isolated using the Gentra Puregene Cell Kit (Qiagen). Purified DNA was digested with Bovine DNase I (Ambion). Test and control samples were labelled with Alexa5 and Alexa3 dyes, respectively, using the BioPrime Total Genomic Labeling Kit (Invitrogen). Labeled samples were competitively hybridized to 2 x 400k Human CGH Arrays (Agilent Technologies) as recommended by the manufacturer. Copy number estimates were extracted from the microarray image files using Feature Extraction 10.5 (Agilent), and the data were analyzed in Agilent Genomic Workbench 6.5 (Agilent) after centralization and fuzzy zero normalization matrixes were applied. Copy number abnormalities were identified using the ADM-2 algorithm with a threshold setting of 5.5, and regions were only considered significant if they were defined by a minimum of 3 consecutive probes, and the average log2 value of all three probes had to either exceed a threshold of 0.2 or less than -0.2 (grey shaded regions). i  The comparative genomic hybridization work was done by Kristi Allen in Dr. Jonathan Keats laboratory. I was responsible for genomic DNA extraction from each cell line.  17  Genomic and reverse-transcriptase PCR and real-time PCR Genomic DNA was extracted with the DNeasy extraction kit (Qiagen), and preparations were measured with a NanoDrop (Thermo-Fisher). For genomic PCRs and real-time, genomic PCR, conditions were: 5 cycles at 95oC for 5 minutes, 60oC for 60s and 72oC for 60s, followed by 30 cycles at 95oC for 60s, 60oC for 30s and 72oC for 30s. RNA was extracted from cells using the RNeasy kit (Qiagen), quantified on a NanoDrop and 1 μg was converted to cDNA using AccessQuick (VWR), in accordance with the manufacturer’s protocols. For reverse transcriptase PCR, conditions were: 35 cycles at 95oC for 60s, 60oC for 30s and 72oC for 30s. Quantitative PCRs reactions were run in triplicate in an Applied Biosystems 7000 series machine (Life Technologies), and analysis of results was done using the ΔΔ Ct method. Expression of transcript/ gene was normalized to TATA box binding protein (TBP) levels, which was then normalized to levels of transcript/ gene in 2885 cells. Primers are listed in Table 2. Table 2: Primers, shRNA and siRNA sequences Gene HMMR  Target Genomic Message  TPX2  Genomic  Region Intron 9-Exon 10  Primer pair 5’ GCTGAAAGGCTGGTCAAGC 3’ 5’CCAACCTAACACGCTCACAT 3’ Exon 9-Exon 10 5’ TGTGCTTCAGATCAAGTGG 3’ 5’ CGTTGTGTTCTCTATTCCTG 3’ Exon 1-Intron 2 5’ AAACCACAGGTAAGGCAGTGAC 3’ 5’ TCACCCACTATCCCACCTCT 3’  Message RHAMM  shRNA  TPX2  shRNA  AURKA  siRNA  Exon 5  5’ AGCCTTTCAACCTGTCCCAAGGA 3’ 5’ AGACAGGGTCTTGCTCCGTCA 3’ 5’ CGTCTCCTCTATGAAGAACTA 3’ 5’ GCCAACTCAAATCGGAAGTAT 3’ 5’ CCGAGCCTATTGGCTTTGATT 3’ 5’TCCCAGCGCATTCCTTTGCAA 3’ 5’ CAGGGCTGCCATATAACCTGA 3’ 5’ CACGTGCTCTACCTCCATTTA 3’ 5’ CACCTTCGGCATCCTAATATT 3’ 18  siRNA and small molecule reagents Four redundant siRNA (Qiagen) against AURKA were used (sequences in Table 2) and transfected at 10 pmol/ well using Lipofectamine2000™ according to manufacturer’s protocols (Invitrogen). That is, S462 cells were seeded at 80% density and left to adhere overnight. Media was then changed to OPTI-MEM (Gibco) and incubated for 8-9 hours, after which the cells were transfected. To transfect the cell-lines, 10 pmol of each siRNA construct was incubated in 250μl OPTI-MEM for 5 minutes at room temperature. At the same time, 5μl of lipofectamine was added to 250μl OPTI-MEM. After the incubation period, the siRNA-media mix was added to each lipofectamine mixture and incubated for 20 minutes at room temperature and then added drop-wise to wells. After 16 hours, cells were washed and passaged into A) 24 well plates for MTT assays at 24, 48, 72 and 96 hours and B) into 6 well plates for western blotting to confirm AURKA knockdown at 48 hours post transfection. Three small-molecule AKIs were used in this study, including the pan-aurora inhibitors C1368 (Sigma) and VX680 (Selleck Chemicals), and the AURKA-specific inhibitor MLN8237 (Selleck Chemicals). For in vitro studies, drugs were reconstituted to 100 mM concentrations in DMSO (Fischer Scientific) and stored at -20°C. For in vitro assays, AKIs were dissolved and diluted in DMSO/DMEM immediately prior to administration.  19  Human MPNST tumour explant modelsii Four-month-old male NOD/SCID mice were anesthetized with isofluorane and orthotopically implanted with 1 mm3 piece of explant tumour in both legs, near the sciatic nerve (Castellsagué et al. manuscript under preparation). Tumour volume was calculated at V= (W2*L (π/6)), where L is the longest diameter and W is the width. Mice were randomly assigned to treatment or control groups. The pharmacokinetics of dosage of MLN8237 in mice was previously determined104. Mice were treated by oral gavage once daily with 30mg/kg MLN8237 or vehicle (10% 2-hydroxypropyl-β-cyclodextrin and 1% sodium bicarbonate) for 22 days in NF1-MPNST and 28 days in sporadic-MPNST (SP-MPNST) implanted mice. Tumours were measured every 3 days. At the end of the treatment period, mice were euthanized and harvested tumours were fixed in formalin.  Cell based assays For cell viability assays, MPNST cell-lines were plated for 24 hours and then treated with carrier alone (0.001% DMSO) or AKIs at indicated concentrations. Viability was quantified after 72 hours by addition of 5 mg/ ml 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4sulfophenyl)-2H-tetrazolium (MTT, Invitrogen) for one hour at 37oC, followed by removal of MTT, and addition of 500 μl DMSO. Absorbance was read at 540 nm in an Enspire 2300 Multilabel plate reader (Perkin Elmer). Each condition was done in triplicate and repeated three times. For growth curves, cells were plated in the absence of AKIs and MTT assays were conducted at 24, 48, 72 and 96 hours after plating.  ii  This work was done by Joan Castellsague at the laboratory of Dr. Conxi Lazaro, Spain  20  For nuclear content analysis by FACS, cells were plated and allowed to adhere for 24 hours, treated with AKIs for 48 hours, harvested with 0.05% trypsin, filtered through a 40 μm mesh and fixed overnight in 70% ethanol at -20oC. Cells were then washed with phosphate-buffered saline (PBS) and stained with 40 μg/ml propidium iodide (PI) (Invitrogen) with 10 U/ml RNAse A (Roche) in PBS for 30 minutes prior to acquisition. Nuclear content was analyzed with FACS Calibur and CellQuestPro software, respectively (BD).  For polyploidy analysis, AKI treated cells were stained with DAPI at 300 nM for 5 minutes in the dark at room temperature. After three washes of 5 minutes each, wells were imaged. Five images per well were taken and nuclear area was measured using Image J software. Annexin V (BD Biosciences) staining followed manufacturer’s protocol. First, cells were incubated in 2 ng/ml Hoechst stain for 30 minutes at 37oC, and then 5 μl of Annexin V and 40 μg /ml PI were added to each well and allowed to incubate at room temperature for 15 minutes in the dark. Cells were washed briefly in PBS and images were collected with a High Content Analyzer (ArrayScan VTI, Cellomics).  Cellular senescence was measured by detection of senescence-activated β-galactosidase (SA-βgal) activity. Cells were plated, treated with AKIs for 72 hours, fixed and stained as described in Debacq-Chainiaux et al. (2009)117. Briefly, cells were seeded and allowed to adhere overnight, and then treated with IC-50 concentrations of AKI for 72 hours with DMSO controls. Cells were washed twice in PBS and then fixed with 2% formaldehyde and 0.2% gluteraldehyde in PBS for 5 minutes at room temperature. Fixed cells were washed twice for 5 minutes each with PBS and 21  incubated with the X-gal staining solution for 16 hours at 37oC without any CO2. The X-gal staining solution is made up of 40 mM citric acid/Na phosphate pH 7.4, 5 mM K4[Fe(CN)6].3H2O, 5 mM K3[Fe(CN)6], 150 mM NaCl, 2 mM MgCl2 and 1 mg/ ml X-gal in distilled water. After staining, cells were washed twice with PBS for 30 seconds each at room temperature and allowed to air-dry. Five images of each well were taken with a Zeiss Axiovert 40 microscope and proportions of SA-β-gal positive cells (blue cells) were counted against normal cells.  Immunofluorescence, immunohistochemistry, and immunoblot analyses Antibodies were sourced and used as follows: RHAMM, 1:1000 (Epitomics), TPX2 1:1000 (Novus), nestin 1:200 (Covance), Tuj1 1:500 (Covance), β-actin 1:2000 (Sigma), AURKA 1:500, phospho (p)-AURKA (Thr288) 1:200 , p-histone H3 (Ser10) 1:500, and caspase 9 1:500 (Cell Signalling). The p-RHAMM (Thr703) polyclonal antibody is characterized in (11) and was used at a 1:10 000 dilution.  Cell-lines were fixed and permeabilized in methanol for 20 minutes at -20oC. Cells were washed with PBS-0.5% Triton X-100 (PBS-T) (Sigma) two times for 5 minutes each before immunofluorescence staining. Cells were blocked in 3% BSA/ PBS-0.1% Tween (Sigma) (0.1% PBST) for one hour at room temperature. Primary and secondary antibodies were diluted in 3% BSA/0.1% PBST. Primary antibody incubations were for 2 hours at room temperature. Cells were washed three times in 0.1% PBST for 10 minutes each and then stained with secondary antibody for one hour at room temperature in the dark. After three 0.1% PBST washes,  22  coverslips were mounted in 90% glycerol/PBS with 4,6-diamidino-2-phenylindole (DAPI) and images were acquired and analyzed using a Olympus FV10i confocal microscope.  For immunofluorescence of spheres, coverslips were coated with 2% geltrex (Invitrogen) in Neurocult media (StemCell Technologies) for 2 hours at room temperature. The coating media was removed and cells were plated and allowed to adhere overnight. To stain cells, coverslips were fixed with 4% paraformaldehyde for 30 minutes at room temperature at defined time points and then washed three times with PBS for 5 minutes each. Cells were then permeabalized with 0.3% PBS-T for 5 minutes, washed with PBS three times, and then blocked in 10% normal donkey serum in PBS for 1 hour at room temperature. Primary antibodies were diluted in 10% normal donkey serum/PBS and incubated with coverslips for 2 hours at room temperatures. Coverslips were washed 3 times for 10 minutes each with PBS, and then secondary stained with Alexaflour-488 and 594 antibodies at 1:4000 dilutions in 2% normal goat serum in PBS for one hour in the dark at room temperature. Coverslips were washed thoroughly with PBS and then mounted as stated above.  Paraffin sections were de-paraffinized, hydrated, and boiled in 10 mM sodium citrate buffer at pH 6.5 for antigen retrieval. After cooling slides for 30 minutes at room temperature, slides were rinsed in water and PBS. Slides were then soaked in hydrogen peroxidase for 10 minutes and washed with PBS. Endogenous biotin was saturated with 5% goat serum in 0.1% Tween-PBS for 1 hour at room temperature. Sections were stained with primary antibody overnight at 4ºC and then washed in 0.1% Tween-PBS Solution. Secondary antibody staining was done with the  23  relevant antibodies (EnVision kit, DAKO) for 30 minutes at room temperature. Staining was visualized by 3,3-diaminobenzidine, with Hematoxylin as a counter-stain.  For analysis of nuclear size in tumour samples, sections were fixed and permeabilized in formaldehyde 4% for 10 min, blocked with PBS-20% FBS for 1 hour, and mounted in 90% glycerol/PBS with DAPI and images were acquired using an Olympus FV10i confocal microscope. Images were analyzed with Image J software by measuring nuclear area of each distinguishable nucleus. Analysis of the area of p-RHAMM positive nuclei in vehicle treated tumours provided the parameters for G2-M nuclei.  Western blot analyses were performed on protein lysates collected from sub-confluent MPNST cells lysed in modified RIPA buffer, containing 50 mM Tris-HCl pH 7.4, 150 mM NaCl, 0.5% NP40, 1 mM EDTA. This buffer was kept at 4oC. Before lysis, fresh protease inhibitor tablet (Roche), 2 mM Na3VO4 and 50 mM NaF were added. To lyse cells 100-300 μl of buffer was added to cell pellets and incubated for 30 minutes at 4oC. Then, cells were spun down at 21000 rpm for 10 minutes at 4oC. Protein concentrations for the respective lysates were quantified using a BCA Protein Assay (Bio-Rad) and 10 or 20 μg of total protein were loaded per lane. All western blots were done on 10% SDS-polyacrylamide gels except for blots probed with pHistone or Histone H3, which were done on 15% gels. Following electrophoresis and transfer to nitrocellulose, blots were blocked in 3% BSA in TBS-T and proteins were detected with the appropriate antibodies. Antibodies were diluted in 3% BSA/ TBS-T and incubated for 2 hours at room temperature or 4oC overnight. Blots were washed three times with TBS-T for 10 minutes  24  each and then protein were detected and quantified with the Odyssey IR imaging system (LICOR) using IRDye 800-conjugated anti-rabbit IgG or IRDye 680-conjugated anti-mouse IgG (Rockland). Alternatively, protein was detected with enhanced chemiluminescence detection of HRP-conjugated antibodies (Sigma). All secondary antibody incubations were done for one hour at room temperature.  S462 sphere xenotransplant modelsiii S462 spheres at passage 5 were dissociated by gentle pipetting and injected sub-cutaneously in NSG mice in 50% geltrex (Invitrogen). 0.5, 2 and 4 million cells were injected into 3 mice and then time to tumour formation was measured. As controls, 1 million adherent S462 cells were injected into one mouse as well. Four tumours were obtained. Three were in the original injection site, but one had an additional tumour in the abdomen (presumably the needle penetrated into the abdomen). Once tumours had formed, they were excised from mice, disaggregated into single cells and plated as S462 Mouse Tumour (S462 MT) cells. Resulting cell-lines were named S462 MT1-4. Disaggregation and dissociation of tumours was performed as follows: tumours were washed once in PBS, mechanically disaggregated with forceps and scissors and then incubated in 10 mls of liberase/blendzyme TM Research Grade enzyme (Roche) at 0.08 WU/ml for 1 hour, shaking at 37oC. The cellular mixture was then passed through a 70 µm cell strainer (BD Falcon) and the filtrate was spun down for 3 minutes at 1500 rpm. Cells were resuspended in DPBS and 10 mls of ammonium chloride (Stem Cell Technologies) was added to cells to lyse red blood cells and incubated at 4oC for 13 minutes. The tube was spun down at 1500 rpm for 3 minutes and the cells were washed with 10mls of DPBS.  iii  All animal handling, injections and tumour excision was done by Dr. Gregor Reid, UBC. I performed all post tumour processing and further in vitro experiments.  25  Cells were centrifuged again, and resuspended as single tumour cells, and plated in DMEM media with 10% FBS. Cells were grown and passaged as with the original S462 cell-line. Cells were grown and re-injected into mice for a second passage at 1 million cells/ mouse into 3 NSG mice and time to tumour formation was measured.  Lentivirus mediated shRNA knockdown and generation of sub-lines To produce lentivirus encoding shRNA, virus was packaged in HEK-293FT cells using psPAX2 (packaging plasmid), pMD2.G (envelope plasmid) (Addgene), and pLKO.1- based vectors targeting the genes-of-interest or a non-hairpin negative control (shRNA sequences are described in Table 2) (sourced from MISSION™ shRNA; Sigma). Plasmids were transfected into HEK293FT cells using Lipofectamine2000™ as noted above. After 16 hours, cells were washed and fresh media was added. Every 24 hours thereafter for 48 hours, media containing virus was harvested and kept at 4oC. After the harvest period, virus was concentrated with the Lenti-X concentrator (Clontech, Mountain View, A), aliquoted in PBS and stored at -20°C. To transduce cell lines, cells were seeded in DMEM/ FBS (no antibiotics) media and after 9 hours switched to OPTI-MEM for 24 hours. Next, 50 μl of virus was added drop wise to wells along with 8 µg/ml polybrene for 16 hours.  Virus containing media was removed and replaced with fresh  DMEM/FBS (no antibiotics) media. After 24 hours, 0.5 µg/ ml puromycin (GIBCO) was added to select for transduced cells. The pLKO.1 shRNA containing plasmids, also have a puromycin resistance gene. Therefore, puromycin treatment for about two weeks removed cells without stable transfection of the pLKO.1 plasmid. Transduced cell lines were maintained with 0.3µg/ml puromycin.  26  Statistics Statistical significance was evaluated by unpaired two-tailed Student’s t-tests. Two way ANOVA was used to determine significance between treatments for in vivo tumour volumes using GraphPad Prism.  27  Chapter 3: Results 3.1 MPNST cell-lines have gene dose alterations in HMMR, TPX2 and AURKA We have hypothesized that genomic alterations that accompany the transition of neurofibromas to MPNST may cause AURKA activity to drive these tumours. In order to study this, NF1-MPNST cell-lines were obtained from the ATCC and Drs. Kluwe and Mautner. These cell-lines stemmed from aggressive human NF1-MPNSTs, which may have gene dose alterations in HMMR, TPX2, and AURKA.21,25,26 Table 3: Origins and profiles of MPNST cell lines. All cell lines came from NF1-MPNSTs. Cell-line S462118  Gender Female  Age 19  2884119  Male  27  2885120,121  Male  35  Stage/ source Stage IV/ Recurrent mass in the thigh Recurrent mass in the leg Lung metastasis  To screen for genomic changes, DNA was isolated and analyzed on a high density oligonucleotide array based CGH assay (Fig 3). iv The 2885 cell-line had a relatively diploid genome (Fig. 3). Consistent with primary MPNST28, S462 and 2884 cells had an amplification in the coding area for AURKA (20q13.2) (Fig 4). Moreover, S462 and 2884 cell-lines differed in the genomic regions encoding for HMMR and TPX2, which are AURKA regulating gene products.31,77,85,90,122 TPX2 (20q11.2) was amplified in the S462 cell-line (Fig 4A), while HMMR (5q33.2) was amplified in the 2884 cell-line (Fig 4B). These results suggest that the levels of HMMR may be elevated in S462 relative to 2884, and thus may alter the activity of AURKA in these cells. iv  This work was done by Kristi Allen in the lab of Dr. Jonathen Keats, TGEN, USA  28  Figure 3: Comparative genomic hybridization in MPNST cells. Red lines indicate genomic levels of 2884 cells, blue lines indicate genomic levels of S462 cells and green lines indicate genomic levels in 2885 cells. The 2885 cell-line is relatively diploid whereas 2884 and S462 have copy number alterations as seen by shifts to the right or left indicating gains or losses respectively. Boxed regions are magnified in Fig. 4. 29  Figure 4: Copy number for HMMR, TPX2 and AURKA in MPNST cell lines. Copy number gains of 20q are found in S462 and 2884 cells, and a gain of 5q33.2-qter is found in 2884 cells. These regions contain the genes for AURKA, TPX2 and HMMR as shown. Grey shifts to the right indicate amplification whereas shifts to the left indicate deletions.  To verify the CGH results for HMMR, three primer pairs were designed against intron- exon boundaries of the HMMR gene to distinguish genomic DNA from message (Fig 5A). Consistently, genomic DNA from 2884 cells had an increased level of HMMR relative to S462 cells (Fig 5A). Quantitative PCR was then used to measure genomic HMMR and TPX2 levels; for these analyses, results were normalized to 2885 as a control for diploid levels. As seen in the CGH study, levels of HMMR were increased in 2884 cells compared to S462 and levels of TPX2 were increased in S462 cells relative to 2884 cells (Fig. 5B, C).  30  Figure 5: Genomic alterations in S462 and 2884 cell lines. A. top: Map of HMMR gene with exons indicated as blocks and introns indicated as lines. Primers are denoted with arrows. Bottom: increased levels of HMMR were seen in 2884 cells relevant to S462 cells by PCR analysis across the gene. B. Q- PCR confirms a genomic amplification of HMMR in 2884 cells. C. Q- PCR confirms a genomic amplification in TPX2 in S462 cells.  31  To determine whether genomic changes translate to changes in expression, the levels of messenger RNA and protein were measured in S462 and 2884 cell-lines. Since RHAMM, TPX2 and AURKA are expressed during the cell cycle91,122,123; the growth rates of cell-lines affect their expression. To control for this, seeding densities were varied to obtain equal doubling times between the S462 and 2884 cell-lines. While the 2885 cell-line was poorly proliferative, the S462 and 2884 cell-lines had comparable growth rates at seeding densities of 1.0x104 and 3.0x104 respectively (Fig 2). To quantitate expression levels by qRT-PCR experiments, primers were designed within neighbouring exons to distinguish from amplification of genomic DNA. As an internal control for expression in each cell line, the expression of each gene was normalized to that of the TATAbinding protein (TBP) and then again to expression of the gene in diploid 2885 cells. Transcript levels were consistent with the trends seen in qPCR and CGH experiments, in that S462 cells have relatively less RHAMM and more TPX2 expression compared to 2884 cells (Fig 6A, B).  Figure 6: Expression of RHAMM and TPX2 reflect genomic alterations in their genes. A. qRT-PCR over three regions of RHAMM cDNA shows consistent increases of RHAMM expression in 2884 cells compared to S462 cells. Error bars = SD, n=3. B. qRT-PCR shows increased expression of TPX2 in S462 cells compared to 2884 cells, error bars = SD, n=3.  32  Next, infra-red (IR) conjugated secondary antibodies were used to detect the levels of the respective protein in S462 and 2884 cells, as this method of detection gives a 1:1 ratio of signal to protein. To compare values between replicate experiments, the respective protein levels detected in S462 were divided by the corresponding expression in 2884 cells to give a S462:2884 ratio. These ratios revealed significantly increased expression of TPX2 (ratio: 2.59 +/- 0.52, p= 0.02 for Student’s t-test comparison of levels in S462 vs. 2884), no variation of AURKA (1.10 +/- 0.19, p= 0.92), and a trend towards reduced expression of RHAMM (0.86 +/- 0.18, p= 0.3) in S462 cells (Fig 7A). To infer the respective kinase activity, the expression levels of three downstream substrates, p-Aurora (Thr-288), p-RHAMM (Thr-703) and p-Histone H3 (Ser-10) were profiled in the S462 and 2884 cell-lines (Fig 7B). Each of the AURKA –mediated phosphorylation events was augmented in S462 cells relative to 2884 cells, while the absolute levels of the substrate proteins were relatively constant.  Conclusion (Section 3.1) Two MPNST cell-lines have been identified that have genomic alterations in AURKA, but differ in the genomic dose for TPX2 and RHAMM. These genomic changes are reflected in the respective messenger RNA and protein levels, which correlate with AURKA kinase activity in these cells. Thus, AURKA activity may be altered by genomic alterations that accompany neurofibroma progression to MPNST. If so, MPNST cells may be highly sensitive to the inhibition of AURKA.  33  Figure 7: S642 cells have increased kinase activity. A. Consistent with the qRT-PCR results, quantitative immunoblotting using infra-red antibodies shows increased expression of TPX2 and a trend towards decreased expression of RHAMM in S462 cells compared to 2884 cells. Expression of AURKA between the two cell lines is comparable. β-actin serves as a loading control. Representative image on the left, quantitation on the right, n=5. B. Augmented AURKA activity is detected in S462 cell lysates as measured by immunoblot detection of p-RHAMM, pS10-Histone H3B and pT288-AURKA. Apart from an increased level of RHAMM expression in 2884, levels of the un-phosphorylated proteins are relatively constant. β-actin serves as a loading control.  34  3.2 AURKA activity is necessary for the growth of MPNST cells in vitro To determine whether AURKA is required for MPNST cell growth, S462 cells were treated with siRNA against AURKA. As a control for the transfection procedure and nonspecific effects of siRNA processing, a scrambled siRNA was included in the experimental design. However, siRNA may recognize genes in a non-specific manner. So, four different siRNAs targeting AURKA were used to test for consistent phenotypes with each siRNA species. AURKA knockdown was achieved 48 hours after transfection as seen by immunoblotting (Fig 8, left). Two siRNA (A1 and A2) were slightly more efficient at reducing the levels of AURKA. Importantly, treatment with these two siRNA also reduced cell viability to the greatest extent, as measured by MTT assay, which suggests a dose effect between AURKA abundance and MPNST cell viability.  Figure 8: Growth of MPNST cell lines is dependent upon AURKA expression. (left) Immunoblot analysis of AURKA expression in lysates from untreated S462 cells and those treated with scrambled or AURKA targeted siRNA at 48 hours post transfection reveal specific reduction of AURKA. β-actin serves as a loading control. (Right) cell viability is decreased in a dose dependent manner in cells treated with siRNA targeting AURKA relative to untreated cells and those cells treated with scrambled siRNA (right). Error bars = SEM, n=5. 35  To verify the dose dependent reliance on AURKA for MPNST cell viability, three small molecule AURKA inhibitors, MLN8237, VX680 and C1368 were obtained. Each of these inhibitors is soluble in DMSO. First, their ability to inhibit AURKA was determined through immunoblot detection of kinase activity. S462 cells were treated with either DMSO (vehicle alone control), 0.1 µM or 1.0 µM of AKI, and lysates were probed for expression of the AURKA substrate p-RHAMM. Reduction of p-RHAMM was seen in a dose dependent manner in the AKI treated lysates (Fig 9A). This effect was confirmed by immunofluorescence analysis of three AURKA substrates following treatment with VX680 in mitotic HeLa cells (Fig 9B).v These experiments confirm that AKIs reduce AURKA activity.  Figure 9: Aurora kinase inhibitors are effective in decreasing kinase activity. A. S462 cells were treated with MLN8237 and VX680, at 0.1 M and 1.0 M. p-RHAMM, but not AURKA and RHAMM, was reduced in a dose-dependent manner. β-actin serves as a loading control. B. By immunofluorescence, VX680- treated HeLa cells have reduced levels of the active kinase (pAurora) and two substrates (p-RHAMM and p-Histone H3). Scale bars represent 5 μm. v  Staining was done by Helen Chen in Dr. Christopher Maxwell’s Lab.  36  Next, the ability of AKIs to impede MPNST cell growth was tested. Cell-lines were treated with various doses of drug, or vehicle control, and cell viability was measured after 72 hours. MTT assays demonstrated a marked and dose dependent decrease in MPNST cell growth in response to AKI treatment (Fig 10). Consistent with the negligible expression of AURKA in 2885 cells (in green), this line resisted AKI and required 400-725 fold higher concentrations to inhibit 50% of cellular growth (IC-50) (Fig. 10). However, the sensitivity to inhibition of each cell-line did not completely align with the levels of AURKA expression. That is, the cellular responses to AKI were significantly different between S462 and 2884 (p=0.02, Fig. 10) despite equivalent growth rates and AURKA abundance (Fig. 5 and 7). We found that S462 cells (in black) were 2.2 – 9.2 fold more sensitive to the pan-Aurora inhibitors VX680 (IC-50- 2884, 387 ±47 nM; S462, 42 ±10 nM) and C1368 (IC-50- 2884, 2925 ±135 nM; S462, 1302 ±264 nM) (Fig. 10). This result implied that factors intrinsic to these cell-lines may be influencing their responses to AKI, such as the activity of AURKA (Fig. 7).  Figure 10: Treatment with AKI reduced cell viability in MPNST cell lines. Treatment of MPNST cells with three inhibitors to Aurora kinases, MLN8237, VX680 and C1368, reveals marked and dose-dependent reduction in cell viability as measured by MTT after 72 hours of treatment. Error bars = SEM, n=3 replicate experiments.  37  Conclusion (Section 3.2) MPNST cell viability is reliant upon AURKA expression and can be reduced by either siRNA targeting AURKA or small-molecule inhibition of the kinase in vitro. The increased sensitivity of S462 cells to AKI impies that other factors may also be involved. These results are encouraging but the efficacy of these drugs against primary MPNSTs is not known.  3.3 Aurora kinase inhibition is an encouraging pre-clinical treatment for human MPNST In vitro studies of immortalized MPNST cell-lines do not model the mixture of Schwann, mast, fibroblast and perineural cells that are found in human tumours.24 One recent study reported encouraging results for MLN8237 against an MPNST cell-line xenografted into mice; however, this still involves an isolated model and does not reflect the heterogeneous nature of MPNSTs.28 We tested MLN8237 against two distinct in vivo explant models utilizing two human primary MPNST (Castellsagué et al. Manuscript under preparation) vi . One sporadic (SPMPNST) (n=44) and one hereditary NF1-associated MPNST (NF1-MPNST) (n=39) were transplanted sub-cutaeneously on either flank of NOD/SCID mice. Tumours were expanded to 2000mm3, and mice were randomly assigned to groups. Pharmacodynamics for MLN8237 in mice are known104, so mice in the treatment group received 30 mg/ kg MLN8237 per day for 22 days. MLN8237 treatment in both models resulted in stabilized disease, while vehicle treated tumours expanded (SP-MPNST, p<0.0001; NF1-MPNST, p=0.0011) (Fig 11A, B). After the treatment period, tumours were excised and weighed. Consistent with tumour volumes measured  vi  The in vivo work was carried out by Joan Castellsague at the lab of Dr. Conxi Lazaro, Barcelona, Spain  38  during the treatment period, tumour weight was significantly lower in both the MLN8237-treated NF1- and SP-MPNST models (p<0.01) compared to vehicle treated tumours (Fig 11A, B).  Figure 11: MLN8237 is effective against primary MPNSTs grown as xenotransplants in vivo A. SP-MPNST tumours were orthotopically transplanted into NOD/SCID mice and allowed to grow to 2000 mm3 before treatment of 30mg/kg MLN8237 or vehicle was delivered daily. MLN8237 treatment resulted in stable disease after 2 weeks, as determined by calliper measurements of tumour volumes, n=30 for MLN8237 treated and n=14 for vehicle treated, *p<0.01, error bars = SD. Representative images of SP-MPNST tumours after 28 days of MLN8237 or vehicle treatment are shown along with a box plot of tumour weights which are significantly reduced in AKI treated mice, *p<0.01, Error bars = SD, the top and bottom of the box represent the upper and lower quartiles and the band in the box represents the median .B. NF1-MPNST tumours also show stabilized tumour volumes and reduced tumour weight in the  39  MLN8237 compared to vehicle treated tumours, n=18 vehicle treated and n=21 for MLN8237 treated mice, error bars = SD. To determine the mechanism of MLN8237 action in vivo, excised SP-MPNSTs were analyzed by immunohistochemistry for proliferation marker ki-67 and for nuclear size by DAPI and p-RHAMM vii . MLN8237 treated tumour cells exit the cell cycle as evidenced by the reduction of ki-67 positive cells in drug versus vehicle treated tumours (Fig 12A). Furthermore, MLN8237 treated tumours undergo endoreduplication as evidenced by the increased level of polyploidy, as measured by the nuclear volumes detected by DAPI (Fig 12B) and within pRHAMM positive cells (Fig 12C). Therefore, MPNST cells exposed to AKI in vivo exit the cell cycle and undergo polyploidization.  Figure 12: MLN8237 treated human MPNSTs undergo endoreduplication. A. Decreased number of Ki67 positive nuclei in SP-MPNST after 28 days of MLN8237 treatment versus vehicle control. B. Quantitation of nuclei area demonstrate increased numbers of 4N nuclei in drug treated cells C. p-RHAMM positive G2/M cells were larger in drug treated tumours, indicating that these cells are undergoing endoreduplication, *p<0.01, error bars = SD. vii  Sectioning and staining was done by Joan Castellsague. I was responsible for the analysis of the images.  40  Conclusion (section 3.3) Treatment of primary human MPNST transplanted into NOD-SCID mice with MLN8237 resulted in stabilized disease in vivo, and tumour cells either underwent endoreduplication or exited the cell cycle.  3.4 Cellular responses to AKI in MPNST cells in vitro Numerous cancer cell-lines have been treated with AKIs, and their cellular responses include arrests at G2-M, apoptosis, endoreduplication and senescence.28,104,109,110,124 All four pathways were investigated in the MPNST cell-lines to determine the in vitro cellular responses to AKI. First, AKI or DMSO treated S462 and 2884 cell-lines were stained with Annexin V (marker for early apoptosis) and propidium iodide (PI) (marker for late apoptosis) and put through high-content analysis. After 48 hours of AKI treatment, MPNST cells experienced about 10% apoptotic cell death (Fig 13A). Next, PI staining/ FACS analysis revealed significant G2/M cell-cycle arrest in both cell-lines after 72 hours of vehicle (DMSO) or AKI treatment (Fig 13B). The level of polyploidy, however, that resulted from MLN8237 treatment was very low, except at a high dose of 1000 nM.  41  Figure 13: AKI treatment results in apoptosis and endoreduplication in MPNST cell-lines. A. AKI treatment in cells for 48 hours resulted in apoptosis as detected by Annexin V staining by high content cell screening (Cellomics). Representative images are shown along with a PI counter stain, *p<0.05 between AKI treated cells and DMSO treated cells, error bars = SEM, n=3. B. (top) Representative DAPI stained nuclei display the enlarged and multi-nuclear phenotypes seen after 72 hours of AKI treatment (M = MLN8237, V = VX680). (bottom) AKI treated S462 cells display larger G2/M and 8N fractions (arrows), indicative of an endoreduplication phenotype, n=3. 42  Lastly, the levels of cellular senescence were measured, and we found that both cell-lines displayed a classic senescence phenotype with larger, multinucleate cells that express senescence-associated -galactosidase (Fig 14).  However, our comprehensive analysis of  cellular death and viability pathways was unable to identify one mechanism that discriminated the responses of S462 and 2884 cells, and may account for the elevated sensitivity of S462 cells to AKI.  Figure 14: VX680 treatment causes MPNST cells to become senescent. Treatment of both, S462 and 2884 cells, with AKIs at IC-50 concentrations for 72 hours induces significant amounts of senescence in VX680 treated cells compared to DMSO controls. Senescent cells were those positive for senescence-activated β-gal (arrows). Representative images (above) were taken on an Axiovert 40 microscope (Zeiss), scale bars represent 10 μm, error bars = SD, n=4, *p<0.05 between DMSO and AKI treated cells. 43  Conclusion (Section 3.4) AKI induces G2-M arrests, apoptosis, endoreduplication and cellular senescence in both S462 and 2884 cell-lines in vitro. The increased sensitivity of S462 cells to AKIs, however, is not explained by these pathways.  3.5 AKI results in impaired self-renewal of MPNST cancer stem-like cells Aggressive tumours may contain a stem-like population of cells that has the capacity to self-renew and reconstitute all the cellular components of a tumour.125 These cells can be expanded in vitro under low-adherence conditions, such as spheroid culture. The presence of these cells has been confirmed in aggressive recurrent tumours in the brain, breast and blood.126– 130  Recently, a population of stem-cell like cells were found in the MPNST cell-line S462.131  These cells are CD133 positive, are able to grow in sphere culture and have the capacity to differentiate down multiple lineages. We postulated that AKI may affect the self-renewal and differentiation capacity of the subset of S462 stem like cells.  First, S462 and 2884 cells were seeded in anchorage independent stem cell conditions. While the 2884 cell-line did not form any spheres, 1.2% (168 ± 25 of the 3000 cells) of the S462 cells seeded were capable of forming spheres (Fig 15). After 6 days of growth, spheres were dissociated and re-seeded for a further 6 days of growth. An increased number of spheres was seen, indicating the enrichment of S462 stem-like cells in culture as well as their ability to selfrenew.  44  Figure 15: S462 cells have a population of stem-like cells that form spheres in vitro. 2884 and S462 cells were seeded in anchorage independent culture conditions, but only S462 cells were able to form spheres after 6 days (passage 0). When these spheres were dissociated and put back into culture, they were able to grow once again as spheres (passage 1). Representative images of spheres are shown on the left, scale bars = 50 µm, and quantitation on the right, error bars = SD, n=3.  Stem-like cancer cells have been thought to lead to tumour relapse due to, among other factors, their increased tumorigenicity compared to bulk tumour cells.132 S462 stem-like cells may represent such a population of highly tumorigenic cells, so we performed a pilot study to test whether S462 sphere cells are more tumorigenic than adherent S462 cells. 0.5 million dissociated S462 sphere cells and S462 cells were xenografted sub-cutaneously into five NSG mice, and time to tumour formation was notedviii (Fig 16). The sphere cells resulted in tumours within 15-20 weeks, whereas the injected S462 adherent cells had no detectable tumours even after four months.  viii  Animal handling, cell injections and tumour excision was done by Dr. Gregor Reid, Vancouver, Canada. I was responsible for all in vitro handling and experimentation with these tumour cells.  45  Cells from the first S462 sphere-derived tumour were dissociated, expanded in vitro and purified into a 97% human population (Appendix A). These cells were termed S462 Mouse Tumour 1 (S462 MT1) cells. The S462 sphere-derived tumour cells could represent an in vivo model for the study of stem-cell-like MPNST cells. However, these cells take very long (15-20 weeks) to form tumours, rendering them not only unattractive for a xenotransplant model but also indicating low tumorigenicity (Fig 16). To expand the tumorigenic population, 1 million S462 MT1 cells were sub-cutaneously injected into three NSG mice. In comparison, 1 million S462 adherent cells were injected into two NSG mice and time to tumour development was measured (Fig 16). S462 MT1 formed tumours considerably faster, after only 6-8 weeks. Even after 15 weeks, adherent S462 cells did not form tumours (Fig 16). Therefore, the S462 sphere cells represent a more tumorigenic, and possibly stem-like, population of cells.  Figure 16: S462 sphere cells are more tumorigenic than adherent S462 cells. S462 adherent (n=2) and sphere cells (n=3) were xenotransplanted into NSG mice, sub-cutaneously in 50% geltrex. Adherent S462 cells did not form tumours, whereas, all three mice injected with sphere cells formed tumours between 14 and 18 weeks post injection. The first tumour (S462 MT1) was dissociated and re-xenotransplanted into NSG mice (n=3). S462 MT1 cells formed tumours quickly in 6-7 weeks, error bars = SD.  46  Next, S462 sphere and adherent cell lysates were profiled for the expression of pAURKA as well as AURKA. While overall expression of AURKA remains unchanged, lysates from sphere cell contained increased levels of p-AURKA (Fig 17).  Figure 17: S462 sphere cells have elevated AURKA activity. S462 sphere cell lysates were profiled and found to have higher expression of p-AURKA than adherent S462 cells.  To determine whether AURKA activity was necessary to sphere cell growth in vitro, S462 cells were seeded at low densities in the presence of either vehicle control (DMSO) or IC50 concentrations of AKIs. When AURKA is inhibited the self-renewal of these sphere cells is limited (Fig 18 A, B). By Day 6 (passage 0, P0), sphere formation was significantly reduced in cells treated with MLN8237 in comparison to vehicle alone treated cells (166 to 29 spheres in DMSO and MLN8237 treated cells respectively) (Fig 18A). Similar reductions in sphere formation were seen with VX680 treatment (Fig 18B). Spheres were then dissociated and reseeded for a second passage (passage 1, P1) to determine if cells with decreased AURKA activity can self-renew. As expected, the number of spheres in the vehicle treatment increased in passage 1 compared to passage 0, consistent with these cells undergoing self-renewal. However, the AKI treated cells are not able to form spheres in secondary passage, suggesting that AURKA activity is necessary for the propagation and self-renewal of sphere-enriched stem-like cells.  47  Figure 18: AURKA is required for self-renewal of stem-like cells in MPNSTs. A, B. Adherent S462 cells were seeded in sphere enriching conditions along with IC-50 concentrations of AKIs and DMSO control. At P0, AKI treated spheres have significantly reduced sphere formation compared to DMSO controls. Sphere formation is further reduced when cells are treated with AKIs for another 6 days compared to DMSO control. Representative images are shown above the quantitation, scale bars = 50µm. Error bars = SD, *p<0.04, **p<0.01 (significance is between AKI-treated and DMSO-treated cells), n=3.  48  We proposed that AKI treatment may be engaging a differentiation pathway in S462 sphere cells. Consistent with this, the few spheres that did form in the presence of AKIs were significantly smaller in size and less uniform in shape, implying that these cells may have undergone differentiation (Fig 18). Thus, AKI treated S462 sphere cells were dissociated, grown for nine days on geltrex coated coverslips in the presence of MLN8237 or vehicle control (DMSO) and stained for the expression of nestin, a type IV intermediate filament protein and marker for neuroprogenitor cells, and for Tuj1, a neuron-specific class III β-tubulin and neural differentiation marker.133,134 After nine days, MLN8237 treatment, in comparison to vehicle alone, resulted in the loss of nestin staining and strong expression of Tuj1 (Fig 19). Furthermore, morphological changes were seen at day nine that were consistent with neurons and neurite-like extensions. In contrast, DMSO vehicle treated control cells maintained a round morphology and nestin expression. This suggests that adherent culture alone is not inducing these cells to differentiate. All in all, MLN8237 treatment at IC-50 doses may be retarding growth by both inducing programs for cell death and engaging neuronal differentiation programs in S462 stemlike cells. Conclusion (section 3.5) S462 stem-like cells are tumorigenic cells that require AURKA for sphere formation and self-renewal. Treatment of these cells with MLN8237 reduced activity of AURKA and engaged a neuronal differentiation pathway. This pathway may be specific to MPNST stem-like cells. As the S462 cell-line had a greater population of stem-like cells compared to 2884 cells, this pathway may explain the increased sensitivity of S462 cells to AKI.  49  Figure 19: MLN8237 treatment results in neuronal differentiation in MPNST stem-like cells. Dissociated sphere-enriched cells were fixed and stained for DAPI, nestin, and Tuj1 at day zero and following nine days of treatment with MLN8237 (50nM), or DMSO. Cells grew in both adherence and sphere phenotypes. While untreated cells maintain expression of nestin through the nine days of culture, MLN8237 treated spheres lost nestin expression and are positive for the neuronal marker, Tuj1 instead. Scale bars equal 50µm.  50  3.6 Loss of RHAMM modifies the sensitivity to AKI and sphere formation for MPNST cells Sensitivity of MPNST cell-lines to AKIs correlated with kinase activity, which in turn, correlated with the levels of RHAMM and TPX2. Thus, we proposed that the abundance of these proteins may control AURKA activity. To investigate whether modifying the expression of TPX2 and RHAMM is sufficient to modify kinase activity and MPNST cellular responses to AKI, stable S462 and 2884 sub-lines were created that express shRNA targeting RHAMM or TPX2 and a control non-hairpin (NHP) shRNA. After puromycin selection for transfected cells, growth kinetics of the sub-lines were profiled and found to be comparable to NHP control cells (Fig 20A, B). The S462 shTPX2 line had ~40% reduction in TPX2 expression (Fig 21A). However, the reduced expression of TPX2 did not alter the IC-50 for MLN8237 (Fig 21B). On the other hand, reduction in RHAMM expression in 2884 cell lines significantly reduced IC-50s to MLN8237 (~65% with redundant constructs: shR1 and shR2) (Fig 22A, B). Furthermore, the activity of AURKA was augmented with reduction of RHAMM, as measured by elevated levels of pAURKA immunofluorescence intensity at both spindle poles in shR1 cells (Fig 22C).  51  Figure 20: TPX2- and RHAMM-silenced sub-lines have similar growth kinetics to nonhairpin controls. A. RHAMM-silenced (shR1 and shR2) and NHP 2884 lines growth rates were measured by MTT assays. Error bars = SD. B. TPX2-silenced S462 cells also grew similarly to their NHP control cells. Error bars = SD.  Figure 21: TPX2 down regulation did not significantly lower IC-50s of S462 cells. A. TPX2 was stably knocked down in S462 cells using shRNA mediated silencing of TPX2 in comparison to control NHP construct. β-actin served as a loading control. B. shTPX2 S462 cells experienced no significant differences in IC-50 to MLN8237 compared to NHP cells, error bars = SD, n=3.  52  Figure 22: RHAMM down regulation increases sensitivity of cells to MLN8237 and AURKA activity in 2884 cells. A. RHAMM was silenced in 2884 cells with two redundant shRNAs, shR1 and shR2. β-actin served as a loading control. B. Both shR1 and shR2 cells experienced a 2 fold or higher decrease in IC-50 in response to MLN8237 treatment compared to NHP cells, *p value<0.05, error bars = SD, n=3. C. Representative images of NHP and shR1 cells at metaphase stained for AURKA and p-AURKA-T288. shR1 cells have significantly higher p-AURKA expression than NHP cells. Scale bar = 5 µm. Quantitation of fluorescent intensity at spindle poles was done using FV10-ASW software, *p value <0.05, error bars = SD.  53  Augmented AURKA activity following silencing of RHAMM (Fig 22C) raises the possibility that these cells could have enhanced sphere forming ability. All three 2884 sub-lines; shR1, shR2 (RHAMM silenced) and NHP (control) cells were plated in sphere enriching media and sphere formation was quantified after six days. Strikingly, the stable silencing of HMMR endowed 2884 cells with the capacity to both propagate and self-renew as sphere-forming cells (Fig 23A). The ability of TPX2 silenced cells to form spheres was also determined. While silencing of TPX2 did not significantly alter the sphere forming ability of S462 cells initially, it did limit their self-renewal capacity in the first passage compared to NHP controls (Fig 23B).  Figure 23: Modulation of RHAMM and TPX2 affects sphere formation in MPNST cells. A. The sub-lines characterized for stable knockdown of RHAMM in Figure 22 (NHP, shR1 and shR2) were grown in anchorage independent conditions for 6 days and formed significantly more spheres than NHP 2884 cells. Scale bars = 50µm, *p-value < 0.05, error bars = SD, n=3. B. Growth of the sub-line characterized for stable knockdown of TPX2 in Figure 20 (NHP, shTPX2) did not result in significantly different sphere formation in passage 0, but TPX2 54  knockdown did affect sphere self-renewal in passage 1, scale bars = 50µm, *p-value< 0.05, error bars = SD, n=3.  The reduction in self-renewal capacity of shTPX2 sphere cells could be a result of small decreases in AURKA activity. Lysates from NHP and shTPX2 sphere cells revealed that expression of active kinase and its downstream phosphorylated substrates was reduced in TPX2 silenced sphere cells (Fig 24). However, the absolute levels of the proteins were also affected. Furthermore, since TPX2 expression is reduced in a stable sub-line model, knockdown efficiency is limited to allow for continued cell viability. Therefore, at this point, the effect of TPX2 knockdown on AURKA activity remains inconclusive and further analysis is required.  Figure 24: TPX2 silenced S462 cells have reduced AURKA activity. Lysates from NHP and TPX2-silenced cells were collected and probed for expression of p-AURKA and its substrates, pRHAMM and p-Histone. All three phosphorylated proteins had lower expression in TPX2 silenced cells, indicating decreased kinase activity in the absence of TPX2. Overall levels of each protein were also determined. Levels of RHAMM remained comparable; however AURKA, and Histone H3 levels were decreased in TPX2-silenced cells. 55  Conclusion (Chapter 3) Taken together, in vitro and in vivo studies have highlighted AURKA inhibition as a promising therapy for MPNSTs. MPNST cells are sensitive to AKIs, and their responses are determined by AURKA activity, which in turn relies upon the abundance of kinase regulators, TPX2 and RHAMM. Cellular responses to AKI include apoptosis, G2-M arrests, endoreduplication and cellular senescence, but an additional pathway may explain the increased sensitivity of S462 cells. The S462 cell-line has a population of stem-like cells that have the ability to self-renew, to differentiate and to form tumours in vivo. The stem-like self-renewal ability of these cells relies on AURKA expression, and treatment of these cells with AKIs engaged a neuronal differentiation program. Interestingly, modulation of TPX2 and RHAMM alters the ability of these cells to form spheres and self-renew.  56  Chapter 4: Discussion and Conclusions  The purpose of this study was to take advantage of our knowledge of the molecular drivers needed for neurofibroma progression to MPNSTs in order to find a new preclinical therapy. This study has shown that MPNST cell-lines have common genomic alterations in the RHAMM-TPX2-AURKA pathway and that these cells are sensitive to AKIs in a manner that correlates with AURKA activity and RHAMM expression. Cellular responses to AKI included apoptosis, endoreduplication and senescence in vitro, and stabilized disease and polyploidy in vivo. Furthermore, AKIs attenuate the self-renewal of tumorigenic MPNST stem-like cells and promotes their differentiation down a neuronal pathway. Finally, depletion of RHAMM was able to increase kinase activity, sensitize cells to AURKA inhibition and induce sphere formation.  4.1 The RHAMM-TPX2-AURKA pathway is critical to MPNST growth and survival Progression of neurofibromas to MPNSTs may relate to genomic changes in AURKArelated genes. These changes include frequent amplifications of AURKA and TPX2 coding regions and loss of HMMR.21,25,26 This study hypothesized that the gain of TPX2 (activator) and the loss of RHAMM (a molecular brake for kinase activity) may augment AURKA activity and therefore, sensitize MPNSTs to AURKA inhibition. In primary MPNSTs, 52% of aggressive MPNSTs had hemizygous deletions of HMMR. This was not seen in the NF1-MPNST cell lines examined in this study. Array CGH examination of three NF1-MPNST cell-lines showed amplifications in AURKA in two cell-lines, S462 and 2884, but the 2884 cell-line contained an amplification in the 5q33.2-qter region surrounding HMMR, rather than a hemizygous loss. However, these cell-lines can still act as a representative model for differences in HMMR gene  57  dose that may accompany the progression of neurofibromas to MPNST. Indeed, consistent with our hypothesis, the S462 cell-line expressed less RHAMM than 2884 cells and, despite similar levels of AURKA expression, was more sensitive to AKI treatment. Furthermore, silencing of RHAMM in 2884 cells decreased the negative regulatory effect of RHAMM on kinase activity in these cells, and sensitized them to MLN8237 treatment.  Importantly, regulation of AURKA activity by RHAMM may act in a bimodal manner. In Xenopus laevis, Xenopus RHAMM (XRHAMM) was required for appropriate localization of TPX234, which would then lead to activation of AURKA. However, when XRHAMM was overexpressed, the localization of TPX2 was inhibited34, potentially through the occupation of TPX2 with RHAMM rather than with AURKA. The 2884 cell-line may be a representation of a RHAMM over-expression model. In 2884 cells, the depletion of RHAMM may release TPX2 to bind AURKA and up-regulate kinase activity. However, the role of RHAMM is hard to classify as either tumor suppressive or oncogenic. Indeed, RHAMM has been shown to be a breast cancer susceptibility gene, and different haplotypes that either elevate or reduce germline expression are associated with increased risk to disease.93 Thus, the regulation of AURKA-TPX2 by RHAMM is likely to be cell-type specific and/ or context dependent.  58  4.2 Cellular responses to AKI Treatment of human MPNST xenotransplanted in mice with MLN8237 resulted in stabilized disease with tumour cells exiting the cell cycle and undergoing endoreduplication. However, in vitro cellular responses to MLN8237 included only low levels of endoreduplication. Interestingly, cells treated with VX680 in vitro did have significantly higher levels of endoreduplication. These in vivo and in vitro results suggest that cellular responses to AKI (i.e., levels of endoreduplication) are reliant both upon the drug and dosage used. VX680 inhibits both AURKA and AURKB, and most likely the inhibition of AURKB with VX680 treatment, induces high levels of endoreduplication. However, MLN8237 is a more specific AURKA inhibitor that only cross-inhibits AURKB at very high doses; MLN8237 inhibits AURKA at 6.7nM and AURKB at 1.5 μM in HeLa cells.104 In our study, only the treatment of cells in vitro with high doses of MLN8237 (1000 nM) resulted in the induction of endoreduplication. Moreover, both this study and that in Patel et al28 used MLN8237 doses in vivo that result in blood plasma levels in excess of 2.5 μM, exceeding the MLN8237 IC-50 against AURKB.104 We propose that the endoreduplication observed in our treated tumours is the result of cross-inhibition of AURKB at the prescribed dose. Finally, S462 cells exhibited more endoreduplication than 2884 cells, and this response may be a result of the genomic loss of TP53 in S462 cells (Appendix B). AKI treated cells arrest at G2/M due to activation of the mitotic checkpoint. This delay can be resolved by p53 mediated apoptosis.46,63,126 However, in the p53 deficient S462 cell-line, the arrested cells may have escaped apoptosis and exited the cell cycle in a polyploid state. That is, these cells undergo additional endoreduplication that will not occur in 2884 cells with normal levels of p53.  59  Overall, AKI treatment is sufficient to induce a significant G2/M arrest. The resolution of this delay, be it through apoptosis or endoreduplication, may rely upon the action of p53 and the extent of AURKB inhibition. 4.3 AURKA is a molecular brake on differentiation programs This study is the first to use AKI treatment to drive the differentiation of sphere-enriched cancer stem-like cells. Not only is AURKA activity critical to the propagation and self-renewal of MPNST sphere cells, but these phenotypes were also responsive to the silencing of regulators for the kinase, such as TPX2 and RHAMM. A recent study screened mouse embryonic stem cells (mESCs) with short-hairpin RNAs (shRNA) to identify proteins involved in maintaining stem cell self-renewal.101 AURKA was one of the eleven candidates identified, and its loss upregulated developmental genes that triggered the differentiation of mouse embryonic stem cells (mESCs) down mesoderm and ectoderm lineages.101 Our results are consistent with these findings and suggest a role for AURKA in maintaining stem-cells.  It is currently believed that stem-like cells in cancer are responsible for aggressive and drug-resistant cancer and, therefore, novel treatments targeting these cells are required.125 This study has shown that MPNST stem-like cells are more tumorigenic than parental S462 cells in vivo. MLN8237 treatment of these cells results in decreased self-renewal and engagement of a neuronal differentiation pathway in vitro. A possible future direction for this work would be to determine the effect of MLN8237 upon these sphere-derived tumours in vivo in order to build upon the encouraging in vitro results and highlight AKIs as a treatment for aggressive and recurrent cancers.  60  Molecular therapies target certain pathways or proteins and therefore, will be the most effective against tumours which are driven by the targeted molecule. This study has shown that MPNST cells with up-regulated AURKA activity are particularly sensitive to AURKA inhibition in vitro. Therefore, p-AURKA or p-RHAMM may serve as biomarkers that highlight tumours, both MPNST and others, that will be particularly sensitive to AKI.  To further improve responses to AKI, a combination of AKI with either chemotherapy or inhibitors of other mitotic and non-mitotic kinases such as PLK1 or MEK/ERK could be utilized.108,111,112 Depletion of PLK1 leads to the same phenotypes as loss of AURKA, and small molecule inhibitors of PLK1 are in both preclinical and clinical studies against various cancers.29,135–139 Therefore, combinatorial inhibition of both kinases may prove more effective in the clinic. In addition, AKI treatment of MPNSTs with low expression of p53 could be done in combination with a small molecule activator, Nutlin-3.124 This molecule augments the proapoptotic actions of AKIs in cell-line models for carcinomas; therefore combined treatment may result in more significant tumour regression in MPNSTs.124  Lastly, the results of this study can also be applied to other tumour subtypes. More specifically, variation in the genomic region surrounding HMMR modifies the risk to develop breast cancer in carriers of germline BRCA1 mutations.31,140 These carriers often develop triple negative breast tumours (do not express estrogen receptor, progesterone receptor or human epidermal growth factor receptor (HER-2)) and resemble basal epithelia by gene expression profiles (i.e. basal subtype). Additionally, these tumours are resistant to anti-estrogen, hormone, and targeted therapies directed against HER-2, highlighting a need for novel therapies.141 In a  61  recent study, 2000 primary breast tumours were analysed at both a genomic and transcriptomic level, and results were associated with long term clinical outcomes.142 The loss of 5-qter, a genomic region surrounding HMMR, was a frequent event in basal subtype breast cancer143, and this genomic loss was associated with an increase in the expression of AURKA in trans142. Thus, genomic loss of HMMR may indicate a subset of breast tumours that are reliant upon AURKA and, therefore, highly susceptible to AKI in a manner similar to MPNSTs.  In conclusion, this study highlights AURKA as a promising candidate for the treatment of MPNSTs, in particular those with AURKA amplification and loss of HMMR or those with elevated kinase activity. Furthermore a pro-differentiation effect for AKI on a subset of sphereenriched cancer stem-like cells was described, suggesting roles for AKI against both differentiated and stem like MPNST cells.  62  Bibliography  1.  Friedman, J. M. Epidemiology of neurofibromatosis type 1. American journal of medical genetics 89, 1–6 (1999).  2.  Jett, K. & Friedman, J. M. Clinical and genetic aspects of neurofibromatosis 1. Genetics in medicine 12, 1–11 (2010).  3.  Rubin, J. B. & Gutmann, D. H. Neurofibromatosis type 1 - a model for nervous system tumour formation? Nature reviews. Cancer 5, 557–64 (2005).  4.  Richetta, A. et al. Lisch nodules of the iris in neurofibromatosis type 1. Journal of the European Academy of Dermatology and Venereology : JEADV 18, 342–4 (2004).  5.  Cawthon, R. M. et al. A major segment of the neurofibromatosis type 1 gene: cDNA sequence, genomic structure, and point mutations. Cell 62, 609 (1990).  6.  Wallace, M. R. et al. Type 1 Neurofibromatosis Gene : Identification of a Large Transcript Disrupted in Three NFL Patients. Science 249 181-186 (2012).  7.  Xu, G. F. et al. The neurofibromatosis type 1 gene encodes a protein related to GAP. Cell 62, 599–608 (1990).  8.  Xu, G. F. et al. The catalytic domain of the neurofibromatosis type 1 gene product stimulates ras GTPase and complements ira mutants of S. cerevisiae. Cell 63, 835–41 (1990).  9.  Rubin, J. B. & Gutmann, D. H. Neurofibromatosis type 1 - a model for nervous system tumour formation? Nature reviews. Cancer 5, 557–64 (2005).  10.  Le, L. Q. & Parada, L. F. Tumor microenvironment and neurofibromatosis type I: connecting the GAPs. Oncogene 26, 4609–16 (2007).  11.  Downward, J. Targeting RAS signalling pathways in cancer therapy. Nature reviews. Cancer 3, 11–22 (2003).  12.  Jacks, T., Shih, T. & Schmitt, E. Tumour predisposition in mice heterozygous for a targeted mutation in Nf1. Nature genetics 7, 353–361 (1994).  13.  Basu, T. et al. Aberrant regulation of ras proteins in malignant tumour cells from type 1 neurofibromatosis patients. Nature 356, 713–715 (1992).  14.  Bollag, G., Clapp, D., Shih, S. & Adler, F. Loss of NF1 results in activation of the Ras signaling pathway and leads to aberrant growth in haematopoietic cells. Nature (1996).  63  15.  Tucker, T. & Friedman, J. Pathogenesis of hereditary tumors: beyond the “two-hit” hypothesis. Clinical Genetics 62, 345–357 (2002).  16.  Zhu, Y., Ghosh, P., Charnay, P., Burns, D. K. & Parada, L. F. Neurofibromas in NF1: Schwann cell origin and role of tumor environment. Science 296, 920–2 (2002).  17.  Yang, F.-C. et al. Nf1-dependent tumors require a microenvironment containing Nf1+/-and c-kit-dependent bone marrow. Cell 135, 437–48 (2008).  18.  Mantripragada, K. et al. Genome •wide high •resolution analysis of DNA copy number alterations in NF1 •associated malignant peripheral nerve sheath tumors using 32K BAC array. Genes, chromosomes & cancer 907, 897–907 (2009).  19.  Spurlock, G. et al. Molecular evolution of a neurofibroma to malignant peripheral nerve sheath tumor (MPNST) in an NF1 patient: correlation between histopathological, clinical and molecular findings. Journal of cancer research and clinical oncology 136, 1869–80 (2010).  20.  Menon, a G. et al. Chromosome 17p deletions and p53 gene mutations associated with the formation of malignant neurofibrosarcomas in von Recklinghausen neurofibromatosis. Proceedings of the National Academy of Sciences 87, 5435–9 (1990).  21.  Mantripragada, K. K. et al. High-resolution DNA copy number profiling of malignant peripheral nerve sheath tumors using targeted microarray-based comparative genomic hybridization. Clinical cancer research 14, 1015–24 (2008).  22.  Ferner, R. & Gutmann, D. International consensus statement on malignant peripheral nerve sheath tumors in neurofibromatosis 1. Cancer research 62, 1573–1577 (2002).  23.  Katz, D., Lazar, A. & Lev, D. Malignant peripheral nerve sheath tumour (MPNST): the clinical implications of cellular signalling pathways. Expert reviews in molecular medicine 11, e30 (2009).  24.  Woodruff, J. M. Pathology of tumors of the peripheral nerve sheath in type 1 neurofibromatosis. American journal of medical genetics 89, 23–30 (1999).  25.  Schmidt, H. et al. Genomic imbalances of 7p and 17q in malignant peripheral nerve sheath tumors are clinically relevant. Genes, chromosomes & cancer 25, 205–11 (1999).  26.  Bache, M. et al. Gains in chromosomes 7 , 8q , 15q and 17q are characteristic changes in malignant but not in benign peripheral nerve sheath tumors from patients with Recklinghausen’s disease. Cancer Letters 155, (2000).  27.  Wang, W., Stukenberg, P. T. & Brautigan, D. L. Phosphatase inhibitor-2 balances protein phosphatase 1 and aurora B kinase for chromosome segregation and cytokinesis in human retinal epithelial cells. Molecular biology of the cell 19, 4852–62 (2008). 64  28.  Patel, A. V et al. Ras-driven transcriptome analysis identifies aurora kinase a as a potential malignant peripheral nerve sheath tumor therapeutic target. Clinical cancer research 18, 5020–30 (2012).  29.  De Luca, M., Lavia, P. & Guarguaglini, G. A functional interplay between Aurora-A, Plk1 and TPX2 at spindle poles: Plk1 controls centrosomal localization of Aurora-A and TPX2 spindle association. Cell cycle 5, 296–303 (2006).  30.  Pryor, J. G., Brown-Kipphut, B. a, Iqbal, A. & Scott, G. a Microarray comparative genomic hybridization detection of copy number changes in desmoplastic melanoma and malignant peripheral nerve sheath tumor. The American Journal of dermatopathology 33, 780–5 (2011).  31.  Maxwell, C. A. et al. Interplay between BRCA1 and RHAMM regulates epithelial apicobasal polarization and may influence risk of breast cancer. PLoS biology 9, e1001199 (2011).  32.  Maxwell, C. A., Keats, J. J., Belch, A. R., Pilarski, L. M. & Reiman, T. Receptor for hyaluronan-mediated motility correlates with centrosome abnormalities in multiple myeloma and maintains mitotic integrity. Cancer research 65, 850–60 (2005).  33.  Groen, A. C. et al. XRHAMM Functions in Ran-Dependent Microtubule Nucleation and Pole Formation during Anastral Spindle Assembly. Current Biology 14, 1801–1811 (2004).  34.  Joukov, V. et al. The BRCA1/BARD1 heterodimer modulates ran-dependent mitotic spindle assembly. Cell 127, 539–52 (2006).  35.  Glover, D. M., Leibowitz, M. H., McLean, D. a & Parry, H. Mutations in aurora prevent centrosome separation leading to the formation of monopolar spindles. Cell 81, 95–105 (1995).  36.  Marumoto, T., Zhang, D. & Saya, H. Aurora-A - a guardian of poles. Nature reviews. Cancer 5, 42–50 (2005).  37.  Wang, X. et al. Overexpression of aurora kinase A in mouse mammary epithelium induces genetic instability preceding mammary tumor formation. Oncogene 25, 7148–58 (2006).  38.  Bischoff, J. R. et al. A homologue of Drosophila aurora kinase is oncogenic and amplified in human colorectal cancers. The EMBO journal 17, 3052–65 (1998).  39.  Zhou, H. et al. Tumour amplified kinase STK15/BTAK induces centrosome amplification, aneuploidy and transformation. Nature genetics 20, 189–93 (1998).  65  40.  Miyoshi, Y., Iwao, K., Egawa, C. & Noguchi, S. Association of centrosomal kinase STK15/BTAK mRNA expression with chromosomal instability in human breast cancers. International journal of cancer 92, 370–3 (2001).  41.  Sakakura, C. et al. Tumour-amplified kinase BTAK is amplified and overexpressed in gastric cancers with possible involvement in aneuploid formation. British journal of cancer 84, 824–31 (2001).  42.  Sen, S. et al. Amplification/overexpression of a mitotic kinase gene in human bladder cancer. Journal of the National Cancer Institute 94, 1320–9 (2002).  43.  Yang, G. et al. Aurora kinase A promotes ovarian tumorigenesis through dysregulation of the cell cycle and suppression of BRCA2. Clinical cancer research 16, 3171–81 (2010).  44.  Gritsko, T. M. et al. Activation and overexpression of centrosome kinase BTAK/AuroraA in human ovarian cancer. Clinical cancer research 9, 1420–6 (2003).  45.  Goepfert, T. M. et al. Centrosome amplification and overexpression of aurora A are early events in rat mammary carcinogenesis. Cancer research 62, 4115–22 (2002).  46.  Janssen, a & Medema, R. H. Mitosis as an anti-cancer target. Oncogene 30, 2799–809 (2011).  47.  Meraldi, P. & Nigg, E. a The centrosome cycle. FEBS letters 521, 9–13 (2002).  48.  Hannak, E., Kirkham, M., Hyman, a a & Oegema, K. Aurora-A kinase is required for centrosome maturation in Caenorhabditis elegans. The Journal of cell biology 155, 1109– 16 (2001).  49.  Giet, R., Uzbekov, R., Cubizolles, F., Le Guellec, K. & Prigent, C. The Xenopus laevis aurora-related protein kinase pEg2 associates with and phosphorylates the kinesin-related protein XlEg5. The Journal of biological chemistry 274, 15005–13 (1999).  50.  Barr, A. R. & Gergely, F. Aurora-A: the maker and breaker of spindle poles. Journal of cell science 120, 2987–96 (2007).  51.  Hirota, T. et al. Aurora-A and an interacting activator, the LIM protein Ajuba, are required for mitotic commitment in human cells. Cell 114, 585–98 (2003).  52.  Marumoto, T. et al. Roles of aurora-A kinase in mitotic entry and G2 checkpoint in mammalian cells. Genes to cells  7, 1173–82 (2002).  53.  Marumoto, T. et al. Aurora-A kinase maintains the fidelity of early and late mitotic events in HeLa cells. The Journal of biological chemistry 278, 51786–95 (2003).  66  54.  Dutertre, S. et al. Phosphorylation of CDC25B by Aurora-A at the centrosome contributes to the G2-M transition. Journal of cell science 117, 2523–31 (2004).  55.  Cazales, M. et al. CDC25B phosphorylation by Aurora-A occurs at the G2/M transition and is inhibited by DNA damage. Cell cycle 4, 1233–8 (2005).  56.  Berdnik, D. & Knoblich, J. a Drosophila Aurora-A is required for centrosome maturation and actin-dependent asymmetric protein localization during mitosis. Current biology 12, 640–7 (2002).  57.  Peset, I. et al. Function and regulation of Maskin, a TACC family protein, in microtubule growth during mitosis. The Journal of cell biology 170, 1057–66 (2005).  58.  Barros, T. P., Kinoshita, K., Hyman, A. A. & Raff, J. W. Aurora A activates D-TACCMsps complexes exclusively at centrosomes to stabilize centrosomal microtubules. The Journal of cell biology 170, 1039–46 (2005).  59.  Kinoshita, K. et al. Aurora A phosphorylation of TACC3/maskin is required for centrosome-dependent microtubule assembly in mitosis. The Journal of cell biology 170, 1047–55 (2005).  60.  Mori, D. et al. NDEL1 phosphorylation by Aurora-A kinase is essential for centrosomal maturation, separation, and TACC3 recruitment. Molecular and cellular biology 27, 352– 67 (2007).  61.  Roghi, C. et al. The Xenopus protein kinase pEg2 associates with the centrosome in a cell cycle-dependent manner, binds to the spindle microtubules and is involved in bipolar mitotic spindle assembly. Journal of cell science 111 557–72 (1998).  62.  Liu, Q. & Ruderman, J. V Aurora A, mitotic entry, and spindle bipolarity. Proceedings of the National Academy of Sciences 103, 5811–6 (2006).  63.  Meraldi, P., Honda, R. & Nigg, E. A. Aurora-A overexpression reveals tetraploidization as a major route to centrosome amplification in p53-/- cells. The EMBO journal 21, 483–92 (2002).  64.  Anand, S., Penrhyn-Lowe, S. & Venkitaraman, A. R. AURORA-A amplification overrides the mitotic spindle assembly checkpoint, inducing resistance to Taxol. Cancer cell 3, 51–62 (2003).  65.  Zhang, D. et al. Cre-loxP-controlled periodic Aurora-A overexpression induces mitotic abnormalities and hyperplasia in mammary glands of mouse models. Oncogene 23, 8720– 30 (2004).  66.  Castro, A. et al. The D-Box-activating domain (DAD) is a new proteolysis signal that stimulates the silent D-Box sequence of Aurora-A. EMBO reports 3, 1209–14 (2002). 67  67.  Littlepage, L. E. et al. Identification of phosphorylated residues that affect the activity of the mitotic kinase Aurora-A. Proceedings of the National Academy of Sciences 99, 15440–5 (2002).  68.  Taguchi, S. I. et al. Degradation of human Aurora-A protein kinase is mediated by hCdh1. FEBS letters 519, 59–65 (2002).  69.  Zeng, K., Bastos, R. N., Barr, F. A. & Gruneberg, U. Protein phosphatase 6 regulates mitotic spindle formation by controlling the T-loop phosphorylation state of Aurora A bound to its activator TPX2. The Journal of cell biology 191, 1315–32 (2010).  70.  Horn, V. et al. Functional interaction of Aurora-A and PP2A during mitosis. Molecular biology of the cell 18, 1233–41 (2007).  71.  Walter, A. O., Seghezzi, W., Korver, W., Sheung, J. & Lees, E. The mitotic serine/threonine kinase Aurora2/AIK is regulated by phosphorylation and degradation. Oncogene 19, 4906–16 (2000).  72.  Eyers, P. A., Erikson, E., Chen, L. G. & Maller, J. L. A novel mechanism for activation of the protein kinase Aurora A. Current biology : CB 13, 691–7 (2003).  73.  Satinover, D. L., Leach, C. A., Stukenberg, P. T. & Brautigan, D. L. Activation of AuroraA kinase by protein phosphatase inhibitor-2, a bifunctional signaling protein. Proceedings of the National Academy of Sciences 101, 8625–30 (2004).  74.  Macůrek, L. et al. Polo-like kinase-1 is activated by aurora A to promote checkpoint recovery. Nature 455, 119–23 (2008).  75.  Seki, A., Coppinger, J. A., Jang, C.-Y., Yates, J. R. & Fang, G. Bora and the kinase Aurora a cooperatively activate the kinase Plk1 and control mitotic entry. Science 320, 1655–8 (2008).  76.  Chan, E. H. Y., Santamaria, A., Silljé, H. H. W. & Nigg, E. A. Plk1 regulates mitotic Aurora A function through betaTrCP-dependent degradation of hBora. Chromosoma 117, 457–69 (2008).  77.  Bayliss, R., Sardon, T., Vernos, I. & Conti, E. Structural basis of Aurora-A activation by TPX2 at the mitotic spindle. Molecular cell 12, 851–62 (2003).  78.  Gruss, O. J. et al. Ran induces spindle assembly by reversing the inhibitory effect of importin alpha on TPX2 activity. Cell 104, 83–93 (2001).  79.  Wiese, C. et al. Role of importin-beta in coupling Ran to downstream targets in microtubule assembly. Science 291, 653–6 (2001).  68  80.  Nachury, M. V et al. Importin beta is a mitotic target of the small GTPase Ran in spindle assembly. Cell 104, 95–106 (2001).  81.  Tsai, M.-Y. et al. A Ran signalling pathway mediated by the mitotic kinase Aurora A in spindle assembly. Nature cell biology 5, 242–8 (2003).  82.  Trieselmann, N., Armstrong, S., Rauw, J. & Wilde, A. Ran modulates spindle assembly by regulating a subset of TPX2 and Kid activities including Aurora A activation. Journal of cell science 116, 4791–8 (2003).  83.  Giubettini, M. et al. Control of Aurora-A stability through interaction with. Journal of Cell Science 1, (2011).  84.  Anderson, K. et al. Binding of TPX2 to Aurora A alters substrate and inhibitor interactions. Biochemistry 46, 10287–95 (2007).  85.  Eyers, P. A. & Maller, J. L. Regulation of Xenopus Aurora A activation by TPX2. The Journal of biological chemistry 279, 9008–15 (2004).  86.  Eyers, P. A., Churchill, M. E. A. & Maller, J. L. The Aurora A and Aurora B protein kinases: a single amino acid difference controls intrinsic activity and activation by TPX2. Cell cycle 4, 784–9 (2005).  87.  Dodson, C. A. & Bayliss, R. Activation of Aurora-A Kinase by Protein Partner Binding and Phosphorylation Are Independent and Synergistic. Journal of Biological Chemistry 287, 1150 –1157 (2012).  88.  Kufer, T. A. et al. Human TPX2 is required for targeting Aurora-A kinase to the spindle. The Journal of cell biology 158, 617–23 (2002).  89.  Sardon, T., Peset, I., Petrova, B. & Vernos, I. Dissecting the role of Aurora A during spindle assembly. The EMBO journal 27, 2567–79 (2008).  90.  Asteriti, I. A., Rensen, W. M., Lindon, C., Lavia, P. & Guarguaglini, G. The AuroraA/TPX2 complex: A novel oncogenic holoenzyme? Biochimica et biophysica acta 1806, 230–239 (2010).  91.  Maxwell, C. A. et al. RHAMM Is a Centrosomal Protein That Interacts with Dynein and Maintains Spindle Pole Stability. Molecular Biology of the Cell 14, 2262–2276 (2003).  92.  Assmann, V., Jenkinson, D., Marshall, J. F. & Hart, I. R. The intracellular hyaluronan receptor RHAMM/IHABP interacts with microtubules and actin filaments. Journal of cell science 112 3943–54 (1999).  93.  Pujana, M. A. et al. Network modeling links breast cancer susceptibility and centrosome dysfunction. Nature genetics 39, 1338–49 (2007). 69  94.  Maxwell, C. a et al. RHAMM expression and isoform balance predict aggressive disease and poor survival in multiple myeloma. Blood 104, 1151–8 (2004).  95.  Shi, Y. et al. Targeting aurora kinases as therapy in multiple myeloma. Blood 109, 3915– 21 (2007).  96.  Ishikawa, H. & Marshall, W. F. Ciliogenesis: building the cell’s antenna. Nature reviews. Molecular cell biology 12, 222–34 (2011).  97.  Pugacheva, E. N., Jablonski, S. a, Hartman, T. R., Henske, E. P. & Golemis, E. a HEF1dependent Aurora A activation induces disassembly of the primary cilium. Cell 129, 1351–63 (2007).  98.  Mori, D. et al. An essential role of the aPKC-Aurora A-NDEL1 pathway in neurite elongation by modulation of microtubule dynamics. Nature cell biology 11, 1057–68 (2009).  99.  Wells, D. G. et al. A role for the cytoplasmic polyadenylation element in NMDA receptorregulated mRNA translation in neurons. The Journal of neuroscience 21, 9541–8 (2001).  100. Frank, C. L. & Tsai, L.-H. Alternative functions of core cell cycle regulators in neuronal migration, neuronal maturation, and synaptic plasticity. Neuron 62, 312–26 (2009). 101. Lee, D., Su, J. & Ang, Y. Regulation of Embryonic and Induced Pluripotency by Aurora Kinase-p53 Signaling. Cell Stem Cell 11, 179–94 (2012). 102. Kollareddy, M., Zheleva, D., Dzubak, P., Lepsik, M. & Hajduch, M. Aurora kinase inhibitors : Progress towards the clinic. Invest New Drugs (2012) 103. Pharmaceuticals, M. Millennium: The Takeda Oncology Company - Our Pipeline. at <http://www.millennium.com/ourScience/OurPipeline.aspx> 104. Manfredi, M. G. et al. Characterization of Alisertib (MLN8237), an investigational smallmolecule inhibitor of aurora A kinase using novel in vivo pharmacodynamic assays. Clinical cancer research 17, 7614–24 (2011). 105. Mossé, Y. P. et al. Pediatric Phase I Trial and Pharmacokinetic Study of MLN8237, an Investigational Oral Selective Small-Molecule Inhibitor of Aurora Kinase A: A Children’s Oncology Group Phase I Consortium Study. Clinical cancer research 18, 6058–64 (2012). 106. Cervantes, A. et al. Phase I pharmacokinetic/pharmacodynamic study of MLN8237, an investigational, oral, selective aurora a kinase inhibitor, in patients with advanced solid tumors. Clinical cancer research 18, 4764–74 (2012).  70  107. Matulonis, U. A. et al. Phase II study of MLN8237 (alisertib), an investigational Aurora A kinase inhibitor, in patients with platinum-resistant or -refractory epithelial ovarian, fallopian tube, or primary peritoneal carcinoma. Gynecologic oncology 127, 63–9 (2012). 108. Fiskus, W. et al. Co-treatment with vorinostat synergistically enhances activity of Aurora kinase inhibitor against human breast cancer cells. Breast cancer research and treatment 135, 433–44 (2012). 109. Görgün, G. et al. A novel Aurora-A kinase inhibitor MLN8237 induces cytotoxicity and cell-cycle arrest in multiple myeloma. Blood 115, 5202–13 (2010). 110. Liu, Y. et al. Targeting aurora kinases limits tumour growth through DNA damagemediated senescence and blockade of NF-κB impairs this drug-induced senescence. EMBO molecular medicine 5, 149–66 (2013). 111. Mahadevan, D. et al. Aurora A inhibitor (MLN8237) plus vincristine plus rituximab is synthetic lethal and a potential curative therapy in aggressive B-cell non-Hodgkin lymphoma. Clinical cancer research 18, 2210–9 (2012). 112. Sehdev, V. et al. The combination of alisertib, an investigational Aurora kinase A inhibitor, and docetaxel promotes cell death and reduces tumor growth in preclinical cell models of upper gastrointestinal adenocarcinomas. Cancer 119 904-914 (2012) 113. Wen, Q. et al. Identification of regulators of polyploidization presents therapeutic targets for treatment of AMKL. Cell 150, 575–89 (2012). 114. Harrington, E. a et al. VX-680, a potent and selective small-molecule inhibitor of the Aurora kinases, suppresses tumor growth in vivo. Nature medicine 10, 262–7 (2004). 115. Keen, N. & Taylor, S. Aurora-kinase inhibitors as anticancer agents. Nature reviews. Cancer 4, 927–36 (2004). 116. Oliveira, T. M., Ahmad, R. & Engh, R. a VX680 binding in Aurora A: π-π interactions involving the conserved aromatic amino acid of the flexible glycine-rich loop. The journal of physical chemistry. A 115, 3895–904 (2011). 117. Debacq-Chainiaux, F., Erusalimsky, J. D., Campisi, J. & Toussaint, O. Protocols to detect senescence-associated beta-galactosidase (SA-betagal) activity, a biomarker of senescent cells in culture and in vivo. Nature protocols 4, 1798–806 (2009). 118. Frahm, S. et al. Genetic and phenotypic characterization of tumor cells derived from malignant peripheral nerve sheath tumors of neurofibromatosis type 1 patients. Neurobiology of disease 16, 85–91 (2004).  71  119. Perrin, G. Q. et al. An orthotopic xenograft model of intraneural NF1 MPNST suggests a potential association between steroid hormones and tumor cell proliferation. Laboratory investigation 87, 1092–102 (2007). 120. Li, Y. et al. Notch and Schwann cell transformation. Oncogene 23, 1146–52 (2004). 121. Fieber, L. A., González, D. M., Wallace, M. R. & Muir, D. Delayed rectifier K currents in NF1 Schwann cells. Neurobiology of Disease 13, 136–146 (2003). 122. Giubettini, M. et al. Control of Aurora-A stability through interaction with TPX2. Journal of cell science 124, 113–22 (2011). 123. Carmena, M. & Earnshaw, W. C. The cellular geography of aurora kinases. Nature reviews. Molecular cell biology 4, 842–54 (2003). 124. Cheok, C. F., Kua, N., Kaldis, P. & Lane, D. P. Combination of nutlin-3 and VX-680 selectively targets p53 mutant cells with reversible effects on cells expressing wild-type p53. Cell death and differentiation 17, 1486–500 (2010). 125. Clarke, M. F. et al. Cancer stem cells--perspectives on current status and future directions: AACR Workshop on cancer stem cells. Cancer research 66, 9339–44 (2006). 126. Prestegarden, L. & Enger, P. Cancer stem cells in the central nervous system--a critical review. Cancer research 70, 8255–8 (2010). 127. Reynolds, B. A., Tetzlaff, W. & Weiss, S. A multipotent EGF-responsive striatal embryonic progenitor cell produces neurons and astrocytes. The Journal of neuroscience 12, 4565–74 (1992). 128. Al-Hajj, M., Wicha, M. S., Benito-Hernandez, A., Morrison, S. J. & Clarke, M. F. Prospective identification of tumorigenic breast cancer cells. Proceedings of the National Academy of Sciences 100, 3983–8 (2003). 129. Jordan, C. T. & Guzman, M. L. Mechanisms controlling pathogenesis and survival of leukemic stem cells. Oncogene 23, 7178–87 (2004). 130. Dick, J. E. Acute myeloid leukemia stem cells. Annals of the New York Academy of Sciences 1044, 1–5 (2005). 131. Spyra, M. et al. Cancer stem cell-like cells derived from malignant peripheral nerve sheath tumors. PloS one 6, e21099 (2011). 132. Zhou, B.-B. S. et al. Tumour-initiating cells: challenges and opportunities for anticancer drug discovery. Nature reviews. Drug discovery 8, 806–23 (2009).  72  133. Park, D. et al. Nestin is required for the proper self-renewal of neural stem cells. Stem cells 28, 2162–71 (2010). 134. Menezes, J. R. & Luskin, M. B. Expression of neuron-specific tubulin defines a novel population in the proliferative layers of the developing telencephalon. The Journal of neuroscience 14, 5399–416 (1994). 135. Vose, J. M. et al. The Plk1 inhibitor BI 2536 in patients with refractory or relapsed nonHodgkin lymphoma: a phase I, open-label, single dose-escalation study. Leukemia & lymphoma (2012). 136. Frost, A. et al. Phase I study of the Plk1 inhibitor BI 2536 administered intravenously on three consecutive days in advanced solid tumours. Current oncology 19, e28–35 (2012). 137. Lens, S. M. A., Voest, E. E. & Medema, R. H. Shared and separate functions of polo-like kinases and aurora kinases in cancer. Nature reviews. Cancer 10, 825–41 (2010). 138. Valsasina, B. et al. NMS-P937, an orally available, specific small-molecule polo-like kinase 1 inhibitor with antitumor activity in solid and hematologic malignancies. Molecular cancer therapeutics 11, 1006–16 (2012). 139. Ellis, P. M. et al. A Phase I Open-Label Dose-Escalation Study of Intravenous BI 2536 Together With Pemetrexed in Previously Treated Patients With Non-Small-Cell Lung Cancer. Clinical lung cancer 14, 19–27 (2013). 140. Nathanson, K. L. et al. CGH-targeted linkage analysis reveals a possible BRCA1 modifier locus on chromosome 5q. Human molecular genetics 11, 1327–32 (2002). 141. Foulkes, W. D., Smith, I. E. & Reis-Filho, J. S. Triple-negative breast cancer. The New England journal of medicine 363, 1938–48 (2010). 142. Curtis, C. et al. The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature 486, 346–52 (2012). 143. Waddell, N. et al. Subtypes of familial breast tumours revealed by expression and copy number profiling. Breast cancer research and treatment 123, 661–77 (2010).  73  Appendices Appendix A: S462 MT1 cells are 97% human  Figure 25: Tumour cells recovered from mice are human. A. Total population of stained cells. Gate was drawn around viable cells. B. Lower right quadrant shows population of human cells pre74  sort. C. Post sort, cells are 97% human (lower right quadrant). Quantitation is shown on the right.  Appendix B: CGH of genomic region coding for TP53  Figure 26: CGH of TP53 in MPNST cell-lines. The genomic region surrounding TP53 is amplified in the S462 cell-line but not in the 2884 or 2885 cell-lines.  75  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0073680/manifest

Comment

Related Items