JOURNAL TRANSCRIPT

---

Clusterin (CLU) and its interacting proteins in cell signalling and neuroblastoma

KORN-ANONG CHAIWATANASIRIKUL

A thesis submitted for the Degree of Doctor of Philosophy

Institute of Child Health University College London 2011

1

DECLARATION

I, Korn-Anong Chaiwatanasirikul, confirm that the work presented in this thesis is my own. Where work has been derived from other sources, I confirm that this has been indicated in the thesis.

Sign

Date

2

ACKNOWLEDGEMENTS I am grateful to the many wonderful people in my life who have helped transform me into a better and wiser person. First and foremost, I would like to thank my supervisor, Dr. Arturo Sala for giving me the opportunity to work on this PhD project. His vision and passion for science are the essence of my inspiration. I am equally thankful to my secondary supervisor, Dr. John Anderson and to my postgraduate tutors for their kind advice throughout my project.

I would like to express my appreciation to the Hodson family and the trustees of the Olivia Hodson Cancer Fund for their financial support. Your trust and encouragement are the reasons why I wish to work hard and keep doing what I do to make a difference.

I would like to give a special thank you to Dr. Olesya Chayka, Dr. Sandy Lee and all the members of the Molecular Haematology and Cancer Biology unit at the ICH. It has been an honour to work with so many great scientists and friends. I immensely enjoyed our daily scientific discussions and appreciate all the support you gave me.

I sincerely thank Mr. Akira Thomas Matsukawa, who has remained marvellously positive throughout. Your great sense of humour and understanding are indispensable for my work. Thank you for always being there for me.

Lastly, I would like to express my deepest gratitude to my family for their unconditional love, support and advice. I owe every step of my success to my parents, who taught me the value of education and encouraged the realization of my dreams from early in life. Thank you for allowing me to pursue my passion and for inspiring me to help others. Your guidance and praise have made me become a stronger and wiser person. I wish to dedicate this thesis to my parents, grandmother and the rest of my family, who have always believed in me.

3

ABSTRACT Neuroblastoma is the most common extracranial solid tumour in children, with a high mortality rate among patients with aggressive disease. In a previous study we showed that Clusterin (CLU) inhibits the transcription factor NFB in a neuroblastoma cell line by stabilizing the NF-B inhibitors (IBs). Moreover, suppression of CLU could elicit NF-B activation and increased the expression of markers for the epithelial-to-mesenchymal transition (EMT), a developmental process utilized by aggressive cancer cells for invasion, in a mouse neuroblastoma model. The expression of CLU is also negatively regulated by the proto-oncogene MYCN, which is associated with aggressive stages of neuroblastoma tumours. Thus, we hypothesised that CLU is a tumour suppressor gene, which negatively regulates NF-B and metastasis. In this study, we investigated the role of the different isoforms of CLU in the regulation of signalling pathways. We also aimed at identifying the precise mechanisms by which CLU regulates NF-B. The results show that intracellular, but not secreted CLU, inhibits NF-B activity. Interestingly, extracellular CLU (secreted CLU) positively regulates AKT and the Phosphoinositide-3 Kinase (PI3K) pathway. Mass-spectrometry analysis and coimmunoprecipitation experiments demonstrated that a chaperone protein named Heat Shock Protein 60 (HSP60) is bound to the N-terminal region of intracellular CLU in neuroblastoma cells. Suppression of HSP60 by shRNA knock down experiments caused decreased neuroblastoma cell proliferation and increased cell death. Our results suggest that HSP60 exert its oncogenic activity by inhibiting the function of CLU and promoting NF-κB activity. In summary, in this report we demonstrate that a direct interaction between intracellular CLU and HSP60 could play an important role in the regulation of NF-κB activity and neuroblastoma development.

4

TABLE OF CONTENTS DECLARATION……………………………………………………………………….2 ACKNOWLEDGEMENTS……………………………………………………………3 ABSTRACT…………………………………………………………………………….4 TABLE OF CONTENTS………………………………………………………………5 LIST OF FIGURES…………………………………………………………………...10 LIST OF TABLES…………………………………………………………………….12 ABBREVIATIONS……………………………………………………………………13 CHAPTER 1: Introduction…………………………………………………………...18 1.1. Neuroblastoma……………………………………………………………..18 1.2. Neural crest development………………………………………………….19 1.3. Classification………………………………………………………………25 1.4. Genetic abnormalities in neuroblastoma…………………………………..28 1.4.1. Ploidy…………………………………………………………….28 1.4.2. Gain of 17q………………………………………………………28 1.4.3. Loss of 1p………………………………………………………..28 1.4.4. Loss of 11q………………………………………………………30 1.4.5. Loss of 14q………………………………………………………31 1.4.6. Deletion of other chromosomal regions/alterations in known TSG and DNA repair genes……………………………………...31 1.4.7. MYCN amplification……………………………………………..34 1.4.8. Neurotrophin signalling pathways……………………………….40 1.4.9. Overexpression of multidrug resistance genes…………………..41 1.4.10. Expression of telomerase……………………………………….41 1.4.11. Apoptotic signalling pathway…………………………………..41 1.4.12. Hereditary predisposition to neuroblastoma……………………42 1.5. Treatment…………………………………………………………………..44 1.6. Clusterin (CLU)……………………………………………………………45 1.6.1. CLU synthesis and Glycosylation……………………………….46 5

1.6.2. Role of CLU in cancer…………………………………………...49 1.6.3. CLU and neuroblastoma…………………………………………53 1.6.4. Mechanism of tumour promotion or suppression used by CLU...55 1.7. NF-kB (Nuclear Factor-kappa B)………………………………………….59 1.7.1. NF-κB inhibitors…………………………………………………60 1.7.2. Mechanism of NF-κB activation………………………………...60 1.7.3. NF-κB and Cancer……………………………………………….63 1.8. Aim………………………………………………………………………...64 CHAPTER 2: Materials and Methods………………………………………………65 2.1. Reagents…………………………………………………………………....65 2.2. Cell Biology………………………………………………………………..65 2.2.1. Cell lines…………………………………………………………65 2.2.1.1. WI-38…………………………………………………..65 2.2.1.2. VA-13 (WI-38 subline 2RA)…………………………..65 2.2.1.3. SHSY-5Y………………………………………………65 2.2.1.4. LA-N-1………………………………………………...66 2.2.1.5. 293FT………………………………………………….66 2.2.1.6. HNB……………………………………………………66 2.2.2. Cell culture………………………………………………………67 2.2.2.1. Harvesting and maintenance of cell lines……………...67 2.2.3. Freezing cultured cells…………………………………………...67 2.2.4. Recovery of frozen cells…………………………………………68 2.2.5. Lentiviral production in 293FT cells…………………………….68 2.2.6. Lentiviral transduction …………………………………………..68 2.2.7. Clonal selection………………………………………………….68 2.2.8. Proliferation assay and live cell count…………………………...69 2.2.9. Cell treatments…………………………………………………...69 2.2.9.1. Sub-lethal heat shock treatment………………………..69 2.2.9.2. Doxorubicin treatment…………………………………70 6

2.2.10. Flow cytometry…………………………………………………70 2.2.10.1. SubG1 cell cycle analysis…………………………….70 2.2.10.2. Annexin-V staining…………………………………...71 2.2.11. Genetically Modified mice……………………………………..71 2.2.11.1. Preparation of adrenal glands cell lysate……………..71 2.3. Molecular Biology…………………………………………………………72 2.3.1. Preparation of protein lysates……………………………………72 2.3.1.1 Bradford protein assay………………………………….72 2.3.2. Western Blot analysis……………………………………………73 2.3.2.1. Acrylamide gel preparation……………………………73 2.3.2.2. Gel electrophoresis and transfer……………………….74 2.3.3. GST pull down………………………………………………..…77 2.3.3.1. pGEX4T-1 plasmids generation……………………….77 2.3.3.2. GST-fusion protein expression………………………...77 2.3.3.3. Purification of GST-fusion protein…………………….78 2.3.3.4. Pull down………………………………………………78 2.3.4. Pull down analyses………………………………………………79 2.3.4.1. Silver staining………………………………………….79 2.3.4.2. Colloidal Coomassie blue staining…………………….81 2.3.4.3. Mass-spectrometry……………………………………..81 2.3.5. Co-immunoprecipitation…………………………………………82 2.3.5.1. In vitro study of protein-protein interactions…………..82 2.3.5.2. In vivo study of protein-protein interactions of endogenous proteins…………………………………82 2.3.6. Generation of conditioned medium………...……………………83 2.3.7. Luciferase reporter assay...............................................................83 2.3.7.1. Single luciferase reporter assay......................................83 2.3.7.2. Dual luciferase reporter assay…………………………83

7

2.4. Molecular Cloning…………………………………………………………84 2.4.1. Primer design…………………………………………………….84 2.4.2. Polymerase Chain reactions (PCR)……………………………...86 2.4.3. Plasmid vectors…………………………………………………..88 2.4.3.1. pcDNA3………………………………………………..88 2.4.3.2. pGIPZ lentivial vector…………………………………88 2.4.4. Restriction digest………………………………………………...92 2.4.5. DNA precipitation……………………………………………….92 2.4.6. De-phosphorylation of DNA plasmids…………………………..92 2.4.7. Agarose gel electrophoresis……………………………………...93 2.4.8. Ligation of DNA fragments……………………………………...93 2.4.9. Bacterial transformation…………………………………………93 2.4.9.1. Competent Bacterial strains……………………………93 2.4.9.2. Transformation of competent Escherichia Coli (E.Coli)……………………………………………….. 95 2.4.10. Preparation of plasmid constructs………………………………95 2.4.11. Sequencing……………………………………………………...95 2.5. Statistics……………………………………………………………95 CHAPTER 3: Investigating the roles of different CLU isoforms in Different signaling pathway……………….………………………………………….97 3.1.Introduction………………………………………………………………...97 3.2. The role of extracellular (secreted) CLU in NF-κB activity……………..100 3.3. The role of secreted CLU in other signalling pathways………………….106 3.4. Alteration of pCLU and sCLU expressions during cell Transformation....109 3.5. Discussion………………………………………………………………...111 CHAPTER 4: Identification of CLU-interacting proteins………………………..113 4.1. Introduction………………………………………………………………113 4.2. Regulation of NF-B activity by full length precursor CLU (pCLU) and its truncated forms…………………………………………………...113 8

4.3. Generation of GST-fusion protein constructs for GST-pull down assays……………………………………………………………………..116 4.4. Identification of CLU-interacting proteins……………………………….116 4.5. Identification of CLU-interacting proteins using a large-scale GST-pull down assay……………………………………………………..118 4.6. Mass-spectrometry analysis………………………………………………120 4.7. The results of mass-spectrometry analysis of HSP60 in GST-pull down assay……………………………………………………..122 4.8. Co-immunoprecipitation of CLU and HSP60 in vivo……………………124 4.9. A complex containing endogenous CLU and HSP60 is detected in neuroblastoma cells………………………………………………………126 4.10. Discussion……………………………………………………………….129 CHAPTER 5: The roles of HSP60 and CLU in neuroblastoma…………………..131 5.1. Introduction………………………………………………………………131 5.2. Reduced expression of endogenous HSP60 decreased SHSY-5Y cell proliferation and increased cell death…………………………………….132 5.3. Reduced expression of endogenous HSP60 decreased primary human neuroblastoma (HNB) cell proliferation………………………………….140 5.4. HSP60 acts upstream of CLU…………………………………………….144 5.5. Discussion………………………………………………………………...150 CHAPTER 6: HSP60 is required for NF-κB activity and its high expression predicts poor survival in neuroblastoma patients…………………………………152 6.1. Introduction………………………………………………………………152 6.2. HSP60 is required for NF-κB activity and its high expression predicts poor survival in neuroblastoma patients………………………………….153 6.3. Discussion………………………………………………………………...161 CHAPTER 7: Conclusions…………………………………………………………..163 REFERENCES………………………………………………………………………170 APPENDIX I…………………………………………………………………………205 APPENDIX II………………………………………………………………………..211 9

LIST OF FIGURES Figure 1.1. Border induction and neurulation……………………………………………………….23 Figure 1.2. Schematic diagram illustrating the neural crest gene regulatory network and the genes involved at different times during neural crest development…………………..................................24 Figure 1.3. A schematic representation of the p53 pathway………………………………………...33 Figure 1.4. Synthesis of different CLU forms (sCLU and nCLU)………………………………….48 Figure 1.5. An association between CLU mRNA to some of the cytogenetic profiles associated with aggressive neuroblastoma………………………………………………………………………54 Figure 1.6. Survival signalling pathways............................................................................................57 Figure 1.7. NF-κB signal transduction pathways……………………………………………………62 Figure 2.1. A schematic presentation of gel and membrane assembly for protein transfer process in Western Blot analysis……………………………………………………………………………..75 Figure 2.2. Schematic representation of the Thermo Scientific mir-30 hairpin design……………..89 Figure 2.3. Schematic representation of the pGIPZ shRNAmir lentiviral vector…………………...89 Figure 3.1. Intracellular CLU negatively regulates NF-B activity in neuroblastoma cell lines…...99 Figure 3.2. Generation of sCLU conditioned media……………………………………………….101 Figure 3.3. Secreted CLU does not regulate NF-B activity in neuroblastoma cell lines…………103 Figure 3.4. Purified secreted CLU does not regulate NF-B activity in neuroblastoma cell lines...104 Figure 3.5. Titration test to determine a working concentration of purified sCLU………………..105 Figure 3.6. Secreted CLU activates the PI3K pathway…………………………………………….107 Figure 3.7. Purified secreted CLU activates the PI3K pathway…………………………………...108 Figure 3.8. Modulation of pCLU and sCLU during cell transformation…………………………..110 Figure 4.1. The alpha chain of CLU (CLU-α) negatively regulates NF-B activity………………115 Figure 4.2. Identification of CLU-interacting proteins by GST-pull down………………………..117 Figure 4.3. Potential CLU-interacting protein of approximately 60 kDa identified by large-scale pull down assays……………………………………………………………………………………119 Figure 4.4. HSP60 was identified as an interacting partner of CLU within its α chain……………121 Figure 4.5. HSP60 expressions in the pull down lysates of CLU…………………………………123 Figure 4.6. CLU interacts with HSP60 in vivo…………………………………………………….125 Figure 4.7. Sub-lethal heat shock can induce endogenous CLU expression………………………127 Figure 4.8. Endogenous CLU and HSP60 interact directly in vivo in LA-N-1 cells………………128 Figure 5.1. Endogenous expression of HSP60 in response to shRNAs knockdown………………135

10

Figure 5.2. Reduced expression of endogenous HSP60 decreased SHSY-5Y cell proliferation….136 Figure 5.3. Reduced expression of endogenous HSP60 increased cell death and sensitivity to doxorubicin-induced death………………………………………………………………………139 Figure 5.4. Primary human neuroblastoma cells (HNB) showed high level of endogenous HSP60 and low CLU expressions………………………………………………………………….141 Figure 5.5. Reduced expression of endogenous HSP60 decreased HNB cell proliferation and increased neuroblastoma cell death…………………………………………………………….143 Figure 5.6. Endogenous HSP60 and CLU expressions can be reduced by shRNAs knockdown…146 Figure 5.7. HSP60 acts upstream of CLU by promoting neuroblastoma cell survival and reduced expression of CLU can restore cells back to normal cell cycle…………………………...148 Figure 5.8. HSP60 is upstream of CLU. Proposed model for the mechanism of action of HSP60 in neuroblastoma cell survival……………………………………………………………………...149 Figure 6.1. Knockdown of HSP60 decreases NF-κB activity and induces accumulation of CLU...155 Figure 6.2. Expression of HSP60 in neuroblastoma is correlated with that of NF-κB target genes.157 Figure 6.3. Expression of HSP60 in neuroblastoma is correlated with MYCN amplification and poor survival………………………………………………………………………………………..158 Figure 6.4. Proposed model for the mechanism of action of HSP60 in the control of NF-κB signalling pathway………………………………………………………………………………….160

11

LIST OF TABLES Table 1.1. Internaltional Neuroblastoma Risk Group staging system (INRGSS)…...……………..27 Table 1.2. International Neuroblastoma Risk Group (INRG) Consensus Pretreatment Classification schema…………………………………….……………………………………….………………...27 Table 1.3. Summary for a selection of direct MycN targets………………………………………...39 Table 1.4. Summary of CLU expression in tumours………………………………………………..52 Table 2.1. Compositions of resolving and stacking gels for SDS-PAGE…………………………...73 Table 2.2. Compositions of 1xTris-glycine SDS running buffer and 1xTransfer buffer……………75 Table 2.3. All the primary and secondary antibodies used………………………………………….76 Table 2.4. Solutions for silver staining……………………………………………………………...80 Table 2.5. Primers used in generating pGEXT-4T1 (GST) and pcDNA3-based protein expression constructs………………………………………………………………………………...85 Table 2.6. PCR conditions to amplify different cDNA sequences used to generate protein expression vectors……………………………………………………………………………………87 Table 2.7. Information of pGIPZ lentiviral vectors purchased from Thermo Scientific (UK)……...91 Table 2.8. Genotypes of E.Coli bacterial strains for molecular cloning and protein expression……94

12

ABBREVIATIONS α

Alpha

ALCL

Anaplastic large cell lymphomas

ALK

Anaplastic lymphoma kinase

Å

Angstrom

ApoJ

Apolipoprotein J

AS-ODN

Antisense oligodeoxynucleotide

Asp

Aspartic acid

ATCC

American Type Culture Collection

BAD

BCL-2-associated agonist of cell death

BAFF

B-cell activating factor

BDNF

Brain-derived neurotrophic factor

β

Beta

bHLH-LZ

basic-helix-loop-helix/leucine zipper

BLAST

Basic Alignment Search Tool

BMP

Bone morphogenetic protein

bp

Base pair

BSA

Bovine serum albumin

o

Degree Celsius

CASZ1

Castor gene

CCR7

C-C chemokine receptor type 7

CHD5

Chromodomain, helicase, DNA-binding gene

ChIP

Chromatin immunoprecipitation

CIAP

Calf intestinal alkaline phosphatase

CLU

Clusterin

CypD

Cyclophillin D

Da

Dalton

DMEM

Dulbecco’s Modified Eagle Medium

DMSO

Dimethyl sulfoxide

C

13

DNA

Deoxyribonucleic acid

DNA-PK

DNA-dependent protein kinase

DSB

Double-strand breaks

DTT

Dithiothreitol

ε

epsilon

EGFR

Epidermal growth factor receptor

EMT

Epithelial to mesenchymal transition

ER

Endoplasmic reticulum

ETB

Epolactaene tertiary butyl ester

FBS

Foetal bovine serum

FGF

Fibroblast growth factor

γ

Gamma

GFP

Green fluorescent protein

GST

Glutathione S-transferase

HRP

Horseradish peroxidase-conjugated

HSPs

Heat shock proteins

HSP60

Heat shock protein 60

IAP

Inhibitors of apoptosis

IER3

Immediate early response 3

IGF

Insulin-like growth factor

IGFR

Insulin-like growth factor receptor

IκBs

Inhibitors of kappa B

IKK

IκB kinase

IL

Interleukin

Inp

Input

INSS

International Neuroblastoma Staging System

IPTG

Isopropyl β-D-1-thiogalactopyranoside

IR

Ionizing radiation

IRF4

Interferon regulatory factor 4 14

k

Kilo

KO

Knockout

L

Litre

LB

Luria-Bertani

LC-MS/MS

Liquid Chromatography with tandem mass spectrometry

LOH

Loss of heterozygosity

LP

Leader peptide

LPS

Lipopolysaccharides

LRP2

Low-density lipoprotein receptor-related protein 2

LTb

Lymphotoxin B

mA

milliampere

MAPK

Mitogen-Activated Protein Kinase

MEFs

Mouse embryonic fibroblasts

MET

Mesenchymal-to-epithelial transition

g

microgram

mg

milligram

MHC

Major histocompatibility complex

MIBG

meta-iodobenzylguanidine

miRNA

Micro ribonucleic acid

µm

micrometre

ml

millilitre

mm

millimetre

MRP

Multidrug resistance-associated protein

MYCN-ER

MYCN-oestrogen receptor

nCLU

Nuclear CLU

NDRG

N-Myc downstream regulated gene

NEAA

Non-essential amino acid

NEMO

NF-κB essential modulator

NF-κB

Nuclear Factor-kappa B 15

ng

nanogram

NGF

Nerve growth factor

NGFR

Nerve growth factor receptor

NIK

NF-κB-inducing kinase

NLSs

Nuclear localization signals

NOD

Non-obese diabetic

NS

No significance

NT

Non-transfected

NT3

Neurotrophin-3

NT4

Neurotrophin-4

OD

Optical density

ODC

Ornithine decarboxylase

PBS

Phosphate buffered saline

PCR

Polymerase chain reactions

PDGFR

Platelet-derived growth factor receptor

PDK1

Phosphoinositide-dependent kinase 1

PHOX2B

Paired-like homeobox 2B

PI

Propidium iodide

PIG

p53 induced gene

PIP2

Phosphatidylionositol-3,4 diphosphate

PIP3

Phosphatidylinositol-3,4,5 triphosphate

PI3K

Phosphatidylinositol-3 kinase

pnCLU

Precursor nuclear CLU

RE

Response element

RFP

Red fluorescent protein

RIP

Receptor interacting protein

RNaseA

Ribonuclease A

ROS

Reactive oxygen species

rpm

round per minute 16

RPMI

Roswell Park Memorial Institute

RTKs

Receptor tyrosine kinases

SCF

Skp1-Cullin-F-box

SCLIP

SCG10-liked protein

sCLU

Secreted CLU

SD

Standard Deviation

SDS

Sodium dodecyl sulphate

SDS-PAGE

Sodium dodecyl sulphate polyacrylamide gel electrophoresis

SGP2

Sulfated glycoprotein 2

shRNA

Short hairpin ribonucleic acid

SNS

Sympathetic nervous system

TAE

Tris-acetate-EDTA

TGFβ

Transforming growth factor beta

TLR

Toll-like receptor

TNF-α

Tumour necrosis factor alpha

TNF-R

Tumour necrosis factor-receptor

TRAF

TNF receptors associated factor

TRAIL

TNF-related apoptosis-inducing ligand

TRAP-1

Tumour necrosis-factor receptor-associated protein-1

TRPM2

Testosterone repressed prostate message 2

Trk

Neurotrophic tyrosine kinase receptor

TSA

Trichostatin A

TSG

Tumour suppressor gene

Tyr

Tyrosine

u

Unit

V

Volts

VEGF

Vascular endothelial growth factor

VHL

Von Hippel–Lindau disease

17

CHAPTER 1 Introduction 1.1. Neuroblastoma Neuroblastoma is the most common extracranial solid tumour in infants and children, which accounts up to 10% of all childhood malignancies (Gurney et al., 1997). Approximately 90% of children with the disease are diagnosed within the first 5 years of life (Schwab et al., 2003). The occurrence of neuroblastoma is approximately 1 in 7,000-10,000 live births and although the disease is largely sporadic, approximately 12% of the patients have a family history of the disease (Shojaei-Brosseau et al., 2004). The hereditary predisposition to neuroblastoma is discussed in more detail in section 1.4.12.

Common signs for early detection of neuroblastoma in children include fatigue, loss of appetite/weight, fever and joint pain (Schwab et al., 2003). Other symptoms often depend on the primary site of tumour and metastases present. A tumour that originates in the abdomen may cause a swollen belly or constipation. A tumour in the chest may result in breathing problems. In addition, a tumour in the spinal cord may cause muscle weakness or inability to stand, crawl or walk. A tumour in the bones around the eyes may cause swelling, bruising or black-eye effect. Lastly, bone lesions in the legs or hips may cause pain and limping. However, these signs and symptoms of neuroblastoma are often unclear, making early diagnosis difficult. Thus, by the time of diagnosis, neuroblastoma often spreads to other parts of the body (Mazur, 2010). Noninvasive methods to screen for neuroblastoma such as urinary catecholamine metabolites have relatively high sensitivity and specificity for detecting the disease (Mosse et al., 2009). The tumour is believed to arise as a consequence of deregulated proliferations of embryonal neural crest cells, which later form the sympathetic nervous system (SNS) (see section 1.2.).

The sympathetic and parasympathetic nervous systems are part of the autonomic nervous system and are important for homeostasis. The sympathetic nervous system acts primarily on the cardiovascular system increasing heart rate, widening bronchial passages, decreasing intestinal activity in response to stress, this is also known as a fight-or-flight response. The parasympathetic nervous system, on the other hand, relaxes 18

the body function and allows the body to rest after stress by decreasing the heart rate and increasing digestion (Sherwood, 2008).

1.2. Neural crest development In the earliest phase of neural development, neural tissue is induced in the ectodermal (outer) layer of the embryo. As a consequence of neural induction, the ectoderm becomes divided into three different regions; the neural ectoderm or neural plate, which will give rise to the central nervous system; the non-neural ectoderm, which will form the epidermis; and the cells at the border between neural and non-neural ectoderm, which for the most part will become the neural crest (Figure 1.1.). Neurulation occurs whereby the neural tissues fold inwards to form the neural tube. The neural plate border cells then bend to form the neural folds and eventually become the dorsal aspect of the neural tube. Depending on the organism and the axial level, neural crest cells initiate migration from the closing neural fold or from the dorsal neural tube (Bronner-Fraser, 2002).

Thus, neural crest cells are vertebrate-specific cells, which form during the early stages of embryonic development. They are migratory multipotent cells, which only become distinct cell type once they have migrated away from the central nervous system during embryogenesis (LeDouarin and Kalcheim, 1999). The multipsteps process of neural crest development include its initial induction, gain of the ability to respond to crestinducing cues at the neural plate border, maintenance of multipotency in the newly induced population, control of cell cycle, epithelial to mesenchymal transition (EMT), delamination from the neuroepithelium, migration and differentiation (Sauka-Spengler and Bronner-Fraser, 2008).

Moreover, the neural plate border and neural crest cells form in response to different signalling events between newly induced neural tissue and the neighbouring non-neural ectoderm. Extracellular secreted signalling molecules (e.g. Wnt, FGF, BMP, and Notch/Delta) have been shown to be important for the initiation of neural crest induction (Stottmann and Klingensmith, 2011, Lewis et al., 2004, Endo et al., 2002, Monsoro-Burq et al., 2003). For example, loss of BMP type 1 receptor (BMPR1A) led to a dramatic decreased in neural crest cells in mice (Stottmann and Klingensmith, 2011). 19

Moreover, overexpression of FGF8 transiently induced neural crest cells in frog (Monsoro-Burq et al., 2003). In zebrafish, an inducible Wnt inhibitor activated during early neurulation has shown to specifically interfere with neural crest cell formation (Lewis et al., 2004). Lastly, Endo and coworkers (2002) demonstrated that Notch signalling is indirectly required for neural crest induction by BMP4 at the epidermisneural plate boundary. Moreover, Notch activation in the epidermis inhibited neural crest formation in this tissue, thus, neural crest generation by BMP4 is restricted only at the border. Afer neural crest signalling induction, the first transcription factors to appear at the neural plate border that can respond to the neural crest-specifying signals to form neural crest cells are known as neural plate border specifiers. This includes Mxs1/2, Dlx5, Pax3/7, Gbx2 and zinc finger–containing Zic proteins. For example, Dlx5 is regulated by attenuated levels of BMP (Luo et al., 2001). These transcription factors, in turn, regulate the neural crest specifier genes, important for cell cycle control, EMT and migration process such as Snail/Slug, AP-2, FoxD3, Twist, Id, cMyc, and Sox9/10 (Figure 1.2.). To initiate migration, premigratory neural crest cells need to delaminate from the neuroepithelium via EMT process. Therfore, transcription factors acting on the neural crest precursors must maintain the precursors in a multipotent, proliferating state and also activate or repress effector genes involved in their EMT. Similar to FoxD3, Sox10 overexpression could induce β1 integrin expression and inhibit N-Cadherin expression whose orchestrated regulation is crucial for EMT to occur (Cheung et al., 2005). Moreover, Cano et al (2000) demonstrated that Snail1 was directly responsible for the negative regulation of E-cadherin (E-Cad), a cell adhesion molecule characteristic of epithelial cells. Once the neural crest cells have migrated and reached their final destinations, expressions of most early neural crest cell specifiers, including Snail/Snail2, FoxD3, Id, and AP2α, are normally downregulated. However, SoxE transcription factor family members Sox9 and Sox10 may persist in specific subpopulations of neural crest cell derivatives such as cartilage and neuron/glial/melanocyte lineages respectively, to regulate their terminal differentiation (Kelsh 2006, Sauka-Spengler and Bronner-Fraser 2008). 20

Downregulation of Id has been shown to be essential for initiating neural crest cell differentiation. One possibility is that the inhibition of Id activity reduces Sox10 expression and that low concentrations of Sox10 can sustain the multipotency of neural crest cells (Kim et al., 2003, Paratore et al., 2001). Neural crest cells can give rise to various derivatives ranging from melanocytes, glia, and neurons to skeletal components of the head. The type of derivative depends upon the axial level from which the neural crest cells originate and the time of their emigration from the neuroepithelium. For example, early migrating cranial neural crest cells populate at the pharyngeal arches can generate bone, cartilage and connective tissue (skeletal structures). However, the later wave stays close to the central nervous system and generates the neurons and glia of the cranial ganglia (Graham et al., 2004). In addition to its role in melanocyte differentiation, Sox10 also controls specification of glial and neuronal fates in neural crest derivative specification. Sox10 has been shown to participate further in the differentiation of glia, as its expression within this lineage persisted into terminal differentiation stages (Kelsh, 2006). During glial differentiation, Sox10 directly regulates the expressions of protein zero (P0), myelin basic protein (MBP), peripheral myelin protein 22 (PMP22) and the gap junction protein connexin 32 (Cx32). Thus, Sox10 has been shown to affect all major components of the myelination process (Bondurand et al., 2001, Peirano et al., 2000). Finally, evidence concerning the direct regulatory role of Sox10 in the control of multipotency and maintenance of stem cell properties in neural crest during differentiation of neural crest–derived neurons comes from studies of sensory and autonomic lineages in the trunk. In mouse neural crest cell cultures, Sox10 regulates the expression of mouse achaete-scute homologue 1 (MASH1) and the paired homeodomain (Phox2b), transcription factors that are essential for autonomic neurogenesis. Moreover, Sox10 acts to delay differentiation of sympathetic and enteric neurons allowing the progenitors to migrate to the correct embryonic locations (Kim et al., 2003).

In primary sympathetic ganglia, the combined expression of the transcription factors Ascl1, Phox2a, Phox2b, Hand2 and Gata2/3 lead to the specification and differentiation of sympathetic neurons (Goridis and Rohrer, 2002, Howard, 2005, Ernsberger and 21

Rohrer, 1996). The Sox10+/Phox2b− subpopulation is considered as undifferentiated and pluripotent neural crest progenitors (Tsarovina et al., 2008). In addition, analysis in zebrafish confirmed that a Wnt signal feeds directly to the Sox10 regulatory element during migration and that SoxE, nuclear factor κB (NF κB), and Notch signals can potentially drive Sox10 expression in neural crest cells (Dutton et al., 2008). Thus, neurogenesis in sympathetic ganglia is also characterized by the Notchmediated segregation and maintenance (Tsarovina et al., 2008). . In summary, each step of the neural crest development is crucial and alteration in any of the signalling events, for example Sox10 and its downstream target, Phox2B, may cause neural crest cells to stop migration and differentiation, which may subsequently form congenital disorders such as Hirschsprung disease (lack of ganglia in the colon) and Neuroblastoma (Kulesa et al., 2009).

22

Figure 1.1. Border induction and neurulation A) The neural plate border (green) is induced by signalling between neuroectoderm (purple) and non-neural ectoderm (blue) and from the underlying paraxial mesoderm (yellow). B) During neurulation, the neural plate borders (neural folds) elevate. C) This causes the neural plate to roll into a neural tube. D) Neural crest cells (green) delaminate from the neural folds or the dorsal neural tubes (shown) depending on the species and axial level. (Adapted from Gammill and Bronner-Fraser, 2003)

23

Figure 1.2. Schematic diagram illustrating the neural crest gene regulatory network and the genes involved at different times during neural crest development (Adapted from Sauka-Spengler and Bronner-Fraser, 2008b).

24

1.3. Classification Neuroblastoma tumour often originates in the adrenal glands (65%). However, the tumour origin can be situated at any location of the sympathetic nervous system, which extends from the neck (5%), chest (20%) to the pelvis (5%) (Janoueix-Lerosey et al., 2010).

The most widely used staging system currently is the International Neuroblastoma Risk Group Staging System (INRGSS) (Table 1.1.). This is developed from the post-surgical staging system (INSS), which divided the risk groups into stages 1 (L1), 2-3 (L2), 4 (M) and 4S (Special) (MS) (Brodeur et al., 1993, D’Angio et al., 1971, Cohn et al., 2009). Spontaneous regression is most commonly observed in stage MS patients. However, the mechanism of tumour regression is still unknown (Brodeur, 2003, Friedman and Castleberry, 2007).

For the INRGSS, stage L1 tumours are localized tumours that do not involve vital structures as defined by the list of image-defined risk factors (IDRFs) (Table 1.1.). The tumour must be confined within one body compartment e.g. neck, chest, abdomen, or pelvis. Stage L2 tumours are locoregional tumours with one or more IDRFs. The tumour may be ipsilaterally continuous within body compartments (i.e. a leftsided abdominal tumour with left-sided chest involvement should be considered stage L2). However, a clearly left-sided abdominal tumour with right-sided chest (or vice versa) involvement is defined as metastatic disease. Stage M is defined as distant metastatic disease (i.e. not contiguous with the primary tumour) except as defined for MS. Nonregional (distant) lymph node involvement is metastatic disease. However, an upper abdominal tumour with enlarged lower mediastinal nodes or a pelvic tumour with inguinal lymph node involvement is considered locoregional disease. Ascites and a pleural effusion, even with malignant cells, do not constitute metastatic disease unless they are remote from the body compartment of the primary tumour. Stage MS is metastatic disease in patients younger than 18 months (547 days) with metastases confined to skin, liver, and/or bone marrow. Bone marrow involvement should be limited to less than 10% of total nucleated cells on smears or biopsy. MIBG scintigraphy must be negative in bone and bone marrow. Provided there is MIBG uptake in the primary tumour, bone scans are not required. The primary tumour can be

25

L1 or L2 and there is no restriction regarding crossing or infiltration of the midline (Monclair et al., 2009).

Children with neuroblastoma exhibit marked variability in outcome depending on the age, stage and biological characteristics of the disease at the time of diagnosis (DuBois et al., 1999). The characteristic risk group of neuroblastoma is shown in table 1.2 where INRG stage, age, histologic category, grade of tumour differentiation, MYCN status, presence/absence of 11q aberrations and tumour cell ploidy are considered. Moreover, the categories were divided into 16 statistically and/or clinically different pretreatment groups of patients, designated as very low (A, B, C), low (D, E, F), intermediate (G, H, I,

J), or high (K, N, O, P, Q, R) pretreatment risk subsets (Cohn et al., 2009).

Low-risk neuroblastoma often observed in younger patients (18 months) often show tumour metastases by the time of diagnosis with various genetic abnormalities (i.e. MYCN amplification and chromosomal alterations including loss of 1p, 11q or gain of 17q). Moreover, 11q aberration is associated with worse outcome in patients with L2 or MS tumours that lack MYCN amplification (Cohn et al., 2009). The survival probability in the high risk cases is less than 40% (Maris et al., 2007).

26

Stage

Description

L1

Localized tumour not involving vital structures as defined by the list of image-defined risk factors and confined to one body compartment

L2

Locoregional tumour with presence of one or more image-defined risk factors

M

Distant metastatic disease (except stage MS)

MS

Metastatic disease in children younger than 18 months with metastases confined to skin, liver, and/or bone marrow

Patients with multifocal primary tumours should be staged according to the greatest extent of disease as defined in the table.

Table 1.1. Internaltional Neuroblastoma Risk Group Staging System (INRGSS) (Adapted from Monclair et al., 2009).

Table 1.2. International Neuroblastoma Risk Group (INRG) Consensus Pretreatment Classification schema. GN, ganglioneuroma; GNB, ganglioneuroblastoma; Amp, amplified; NA, not amplified; L1, localized tumor confined to one body compartment and with absence of image-defined risk factors (IDRFs); L2, locoregional tumor with presence of one or more IDRFs; M, distant metastatic disease (except stage MS); MS, metastatic disease confined to skin, liver and/or bone marrow in children < 18 months of age. Letters A to R represent the 16 statistically and/or clinically different pretreatment groups of patients (Cohn et al., 2009).

27

1.4. Genetic abnormalities in neuroblastoma. Like other cancers, the molecular biology of neuroblastoma is characterised by somatically acquired genetic events that result in the activation of oncogenes and suppression of tumour suppressor genes or alterations in gene expression. Various techniques have been used to identify potential prognostic markers, which may determine the response to therapy and clinical outcome. Tumours with MYCN gene amplification, chromosome 1 loss of heterozygosity (LOH) and gain of chromosome 17 are associated with aggressive neuroblastomas. Alteration in the expressions of neurotrophin and their receptors correlate with clinical behaviour and may reflect the degree of neuroblastic differentiation before malignant transformation (Maris and Matthay, 1999, Janoueix-Lerosey et al., 2009).

1.4.1. Ploidy Ploidy refers to the term that describes the number of chromosomes in a cell. A normal diploid cell has 46 chromosomes with a DNA content of 1. A triploid cell (3n), on the other hand, would have 69 chromosomes with a DNA content of 1.5. The majority of neuroblastoma cell lines and advanced primary tumours have either a near-diploid (2n) or near-tetraploid (4n) DNA content. Infant neuroblastomas (50% of primary tumours. This type of genetic alteration is associated with more aggressive neuroblastomas. Overexpression of survivin (inhibitor of apoptosis protein) has been proposed to occur in this region (Islam et al., 2000).

1.4.3. Loss of 1p In 1977, Brodeur and coworkers first recognized that the deletion of the short arm chromosome 1 (1p) was a common karyotypic feature of neuroblastoma cell lines and tumours. LOH of chromosome 1p is observed between 19-36% of all primary tumours. 28

Introduction of an intact chromosome 1p to 1p-deleted neuroblastoma cell lines induce cell differentiation and/or cell death (Bader et al., 1991). This suggests that this region harbours tumour suppressor genes. Most 1p deletions cover almost the entire chromosomal arm, with common deletions occurring between 1p36.2-36.3 (White et al., 1995). One study has postulated the existence of two, or even more neuroblastoma genes in 1p (Caron et al., 2001). The existence of two separate 1p regions that are associated with neuroblastoma is supported by a study that identified one region at 1p36 and one at 1p22 (Mora et al., 2000). Moreover, Hiyama and colleagues (2001) reported three regions at 1p36.1-2, 1p36.3, and 1p32-34, each associated with different subgroups of neuroblastoma. Caron et al (1993) also found that tumours with MYCN amplification generally had 1p deletions extending proximal to 1p36, but single-copy MYCN tumours more often had small terminal deletions of 1p36 only.

There is a strong correlation between 1p LOH and high-risk features such as age >1 year at diagnosis, metastatic disease, MYCN amplification and 1p LOH may be used as a prognostic marker for relapse (Maris and Matthay, 1999).

Candidate tumour suppressor genes located on chromosome 1p include TP53 homolog TP73; the CDK2 homolog CDC2L1 (p58); the transcription factors HKR3, DAN, PAX7, ID3, and E2F2; the transcription elongation factor TCEB3 (Elongin A); and two members of the tumour necrosis factor receptor family, TNFR2 and DR3. However, each of these genes except HKR3 and DR3 are located outside the current consensus region, and no mutations have been found in the non-deleted allele of any candidate (Maris and Matthay, 1999).

Two very important tumour suppressor genes often deleted (1pLOH) in neuroblastoma have recently been identified (Liu et al., 2011, Fujita et al., 2008). Firstly, human Castor gene (CASZ1) is a zinc finger transcription factor, mapped to chromosome 1p36.22, is structurally homologous to Drosophila castor. CASZ1 is induced when cells of neural or mesenchymal origins undergo differentiation, suggesting its role in the differentiation program of neuroblast cell lineages. Other roles of CASZ1 include enhancement of cell adhesion, inhibition of cell migration and suppression of tumourigenicity (Liu et al., 2011). There are two isoforms of human CASZ1 (hCasz5 and hCasz11) (Liu et al., 2006). Liu and coworkers (2011) demonstrated that both 29

forms of CASZ1 function to suppress growth in neuroblastoma tumours both in vitro and in vivo and that low expressions of CASZ1 is significantly associated with decreased survival probability of neuroblastoma patients. Treatments with retinoic acids and epigenetic modification agents (e.g. TSA) could induce neuroblastoma cell differentiation through the induction of CASZ1 expression (Liu et al., 2011). CASZ1expressing xenograft tumours also showed increased expressions of NGFR and TrkA, which are highly expressed in differentiated neuroblastoma tumours and markers of good prognosis, suggesting that CASZ1 may function like a haplo-insufficient tumoursuppressor gene, where a partial loss of CASZ1 may be sufficient to result in neuroblastoma development (Liu et al., 2011).

In addition, chromodomain, helicase, DNA-binding gene (CHD5), a chromatin remodelling gene, mapped to chromosome 1p36.31 is often deleted in human neuroblastomas (Thompson et al., 2003). CHD5 is a member of the CHD gene family and encodes a novel class of Swi/Snf proteins that contains a Swi/Snf-like helicase ATPase domain, a DNA-binding domain and a chromodomain motif, which can directly affect chromatin structure, altering the access of the transcriptional apparatus and gene transcription (Woodage et al., 1997). Clonogenicity and tumour growth were abrogated in neuroblastoma cell lines overexpressing CHD5 (Fujita et al., 2008). Moreover, reactivation of CHD5 expression resulted in the induction of neuroblatoma cell differentiation (Garcia et al., 2010). Thompson et al (2003) demonstrated that low CHD5 expression is correlated with 1p deletion, MYCN amplification, advanced stage and unfavourable neuroblastoma cases. Therefore, CHD5 may also function like a haplo-insufficient tumour-suppressor gene, where a partial loss of CHD5 may be sufficient to result in aggressive neuroblastoma development.

1.4.4. Loss of 11q Chromosome 11q deletions are observed in approximately 15-20% of primary neuroblastomas. This suggests that chromosome 11q is another important site of neuroblastoma tumour suppressor genes (Mertens et al., 1997). The most common region for deletion is at 11q23. LOH of 11q is directly correlated with 14q deletion and poor prognosis but inversely correlated with MYCN amplification.

30

1.4.5. Loss of 14q Deletion of the long arm of the chromosome 14 (14q) was first identified in 1989 by Suzuki and coworkers. The most common deletion region occurred within 14q23, in 23% of the 280 primary tumours studied (Thompson et al., 2001). This type of chromosomal alterations was highly correlated with 11q LOH and inversely correlated to MYCN amplification and present in all clinical risk groups, indicating that this abnormality may occur early in tumour development.

1.4.6. Deletion of other chromosomal regions/ alterations in known TSG and DNA repair genes Other LOH and/or allelic imbalance at chromosome arms 3p, 4p, 5q, 9p and 18q have also been reported (Meltzer et al., 1996, Takita et al., 2000, Hirai et al., 1999). However, their molecular characteristics are less defined as these genetic alterations occur at a much lower frequency than 1p LOH.

Unlike other cancers, p53, which is located on the short arm of chromosome 17 (17p13), is only rarely inactivated (