Wip1 regulates proliferation of adult neural progenitors through p53 dependent g2 cell cycle control

WIP1 REGULATES PROLIFERATION OF ADULT NEURAL PROGENITORS THROUGH P53-DEPENDENT G2 PHASE CELL CYCLE CONTROL ZHU YUNHUA NATIONAL UNIVERSITY OF SINGAPORE 2009... WIP1 REGULATES PROLIFER

WIP1 REGULATES PROLIFERATION OF ADULT NEURAL

PROGENITORS THROUGH P53-DEPENDENT G2 PHASE

CELL CYCLE CONTROL

ZHU YUNHUA

NATIONAL UNIVERSITY OF SINGAPORE

2009

WIP1 REGULATES PROLIFERATION OF ADULT NEURAL PROGENITORS

THROUGH P53-DEPENDENT G 2 PHASE CELL CYCLE CONTROL

ZHU YUNHUA 2009

WIP1 REGULATES PROLIFERATION OF ADULT NEURAL

PROGENITORS THROUGH P53-DEPENDENT G2 PHASE

CELL CYCLE CONTROL

First, I would like to thank my supervisors: Associate Professor Xiao Cheng, for taking me as a graduate student in his lab, for his continuous guidance and supervision throughout the past three years I would like to express my gratitude to Associate Professor Dmitry Bulavin, for excellent mouse models provided and insightful discussions I feel much indebted and express my big thank to my supervisor Associate Professor Ng Yee Kong, for the key helps and encouragement during the last stage of my candidature and the critical reading and corrections on this thesis, without which I would not be able to submit this thesis

Zhi-I am especially thankful to the Head of the Department, Professor Bay Boon Huat, for his kind understanding and help in accepting me into the department, and the great assistance and support provided during my candidature Without him, I would neither be able to join the department nor to complete my Ph.D work and thesis

I am especially thankful to the administrative help and advice rendered by Ms Geetha Sreedhara Warrier and Madam Ang Lye Geck, Carolyne

I would like to thank the members of my Thesis Advisory Board, Associate Professor Yang Xiaohang, Dr He Beiping and Dr Gavins Stewart Dawe, for their invaluable discussions and suggestions towards the completion of this project

I would like also thank Associate Professor Tay Sam Wah, Samuel and Dr

Fu Jiang for their help in trouble shooting problems regarding culturing neural stem cells

Zhang Chengwu for providing assistance in Western blot technique, Xavier and Jocelyn who read and provided critical comments and suggestions to the thesis I would also thank my friend Dr Yu Faxing for the encouragement, the information and the pressure that make the completion of the thesis possible

Last but not least, I deeply appreciated my family, especially my father, mother and my wife, who support me, encourage me throughout the course of my study

1.2.5 Principle of common used methods in the study of adult neurogenesis

1.2.5.1 BrdU short- and long-term labeling 1.2.5.2 Neurosphere assays

1.2.5.3 Differentiation assay 1.3 Wip1 phosphatase

1.3.1 General features of Wip1 1.3.2 Wip1 knockout mice

i iii viii

x xiv

1.3.4 Wip1 in DNA repair and cell cycle regulation 1.3.5 Possible role of Wip1 in neurogenesis

1.4 Wip1 related cell cycle regulators in adult neurogenesis

1.4.1 A general theme of cell cycle regulation 1.4.2 Regulation of NPCs by DNA damage signaling molecules

1.4.2.1 ATM in hippocampal neurogenesis 1.4.2.2 p53-p21 in NPC proliferation and self-renewal 1.4.2.3 Bax in NPC spontaneous apoptosis

1.4.3 Regulation of NPCs by p38MAPK stress signaling molecules

1.4.3.1 p38MAPK in hippocampal neurogenesis 1.4.3.2 Bmi1-p16Ink4ap19Arf in proliferation and self-renewal 1.4.4 NPCs in aging and transformation

1.4.4.1 Wip1 related molecules in aging 1.4.4.2 NPC transformation

1.5 Aims, hypothesis and strategies

1.5.1 Aims and perspectives 1.5.2 Hypothesis

1.5.3 Questions to address 1.5.4 Strategies

Chapter 2, Materials and Methods

2.1 In vivo methods

2.1.1 Animals 2.1.2 Antibodies 2.1.3 BrdU injection

2.1.4 Immunohistochemistry 2.1.5 Counting of immuno-positive cells 2.1.6 TUNEL staining

2.1.7 Brain dissection

2.2 In vitro methods

2.2.1 Neurosphere experiments 2.2.2 Immunocytochemistry and quantification 2.2.3 CFSE pulse labeling

2.2.4 PI labeling and flow cytometry 2.3 Molecular and biochemical methods

2.3.1Real-time PCR 2.3.2 Western blot 2.4 Data analysis

Chapter 3, Results

3.1 Wip1 deficiency decreases new neuron formation and NPC activity in vivo

3.1.1 Wip1 deficiency decreased new neuron formation in olfactory bulb

3.1.2 Wip1 deficiency reduced the number of SVZ NPCs in vivo

3.1.3 Wip1 deficiency reduced SGZ neurogenesis in hippocampus

3.2 Functional analysis of Wip1 ko NPCs in vitro

3.2.1 Wip1 deficiency decreased amplification of NPCs 3.2.2 Wip1 deficiency impaired self-renewal of NPCs

3.2.3 Wip1 deficiency diminished neuronal differentiation in vitro

3.3.1 Wip1 ko NPCs exhibited prolonged cell cycle 3.3.2 G2 to M phase transition was compromised upon knocking out Wip1

3.4 Molecular abnormalities in Wip1 ko NPCs

3.4.1 Phosphorylation of p53 and expression of cell cycle inhibitors are both elevated in neurospheres and SVZ

3.4.2 The elevation of cell cycle inhibitors was p53-dependent 3.5 p53 mediates the repressed cell cycle abnormalities in Wip1 ko NPCs

3.5.1 Knocking out p53 released the restriction on M phase entry 3.5.2 Knocking out p53 drove NPC cell cycle shorter than normal

3.6 Wip1 modulates NPC amplification in a p53-dependent manner in vitro

3.7.1 Wip1 expression decreased while phospho-p53 increased during aging

3.7.2 Transgenic expression of Wip1 increased neurogenesis in aged mice

Chapter 4, Discussion and Conclusion

4.1 G2/M transition is important in mammalian NPCs

4.2 Wip1-p53 pathway is critical for adult hippocampal neurogenesis

4.3 ATM is probably not involved in NPC proliferation

4.4 p38MAPK may not mediate the function of Wip1 in NPCs

4.5 Implications of the Wip1/p53 pathway on NPC aging and transformation

4.5.1 The change of Wip1-p53 balance during aging may contribute to age related functional decline of NPCs

4.5.2 A proposed model of NPC aging 4.5.3 A possible role of Wip1 in brain tumor formation 4.6 Conclusion

4.7 Remaining questions

4.7.1 What are upstream regulators of p53?

4.7.2 How does Wip1/p53 regulate NPC self-renewal?

4.7.3 What regulates the activity of Wip1?

List of Figures

Figure 1.1: A general theme of adult neurogenesis

Figure 1.2: Identified Wip1 substrates and functions

Figure 1.3: Wip1 related molecules in the regulation of NPCs

Figure 3.1: Wip1 is essential for new cell formation in adult olfactory bulb

Figure 3.2: In vivo analysis of lineage specification, apoptosis and migration

Figure 3.3: Wip1 deficiency reduces the number of NPCs in adult SVZ

Figure 3.4: Characterization of effects of Wip1 ko on different cell types in SVZ

Figure 3.5: Knocking out Wip1 decreases neurogenesis in adult hippocampus

Figure 3.6: Knocking out Wip1 decreases long-term survival of new cells in adult

hippocampus

Figure 3.7: Wip1 deficiency reduces proliferation of NPCs in vitro

Figure 3.8: Neurosphere diameters of Wip1 wt and ko NPCs across different

developmental stages

Figure 3.9: Reduction of amplification of Wip1 ko NPCs is not dependent on specific

growth factors

Figure 3.10: BrdU labeling in vitro

Figure 3.11: Impaired self-renewal of Wip1 ko NPCs

Figure 3.12: Impaired neurogenesis of Wip1 ko NPCs

Figure 3.13: Cell cycle abnormalities of Wip1 ko NPCs

Figure 3.14: M phase entry is impaired in Wip1 deficient NPCs

Figure 3.15: Molecular abnormalities of Wip1 ko NPCs

Figure 3.16: Elevations of p21 and Reprimo are p53-dependent in Wip1 ko NPCs

Figure 3.17: NPC cell cycle regulation by Wip1 is dependent on p53

7

14

20

59 60-2

65 66-7

90

92

Figure 3.18: The compromised amplification of NPCs in Wip1 ko mice is

p53-dependent

Figure 3.19: Regulation by Wip1 on NPC self-renewal is p53-dependent

Figure 3.20: Regulation by Wip1 on neurogenesis in vitro is p53-dependent

Figure 3.21: Functional deficiency of ATM or p38MAPK does not affect SVZ NPC

numbers in vivo

Figure 3.22: Regulation of NPC number in vivo by Wip1 is p53-dependent

Figure 3.23: Reduction of the OB size in Wip1 ko mice is p53-dependent

Figure 3.24: p53 deficiency elevates the number of NPCs in SGZ of adult hippocampus

Figure 3.25: p53 knockout rescues the number of NPCs in adult SGZ

Figure 3.26: Wip1 regulation on the neuroblast formation of SGZ cells is p53

dependent

Figure 3.27: Characterization of the Wip1-p53 pathway in SVZ during aging

Figure 3.28: Transgenic expression of Wip1 increases the number of NPCs in aged

SVZ

Figure 3.29: Knocking out Chk2 rescues defects of Wip1 ko NPCs on stem cell

frequency and NPC amplification in vitro

Figure 4.1: The working model: NPC aging is partially mediated by age related

decline of Wip1 expression

113

114

132

Abbreviations

APC Adenomatous polyposis coli

ATM Ataxia telangiectasia mutated

ATR Ataxia telangiectasia mutated related

bp base pairs

Bax Bcl-2–associated X protein

BH3 Bcl-2 homology domain 3

BrdU 5-bromo-2-deoxyuridine

CDK Cyclin dependent kinase

cDNA complementary deoxyribonucleic acid

Cdkn2a Cyclin dependent kinase inhibitor 2a

CNS Central nervous system

CFSE 5(6)-carboxyfluorescein diacetate succininyl ester

DAPI 4, 6-diamidino-2-phenylindole

DCX Doublecortin

DG Dentate gyrus

dko Double knockout

DNA Deoxyribonucleic acid

DNA-PK DNA-dependent protein kinase

E14 Embryonic day 14

EGF Epidermal growth factor

EGFR Epidermal growth factor receptor

ENU N-ethyl-Nnitrosourea

FGF Fibroblast growth factor

FSH Forward scatter height

G1 Gap phase 1

G2 Gap phase 2

GBM Glioblastoma

GCL Granular cell layer

GFAP Glial fibrillary acidic protein

γ-H2AX histone H2A variant γ

HCl Hydrochloric acid

Het Heterozygous

Hmga2 High mobility group AT-hook 2

HSC Hematopoietic stem cells

IACUC: Institutional Animal Care and Use Committee

IMCB: Institute of Molecular and Cellular Biology

IR Ionizing radiation

ko Knockout

M Mitosis phase

MAPKAPK2 Mitogen-activated protein kinase-activated protein kinase 2

MEF Mouse embryonic fibroblast

mm milimeter

RNA Ribonucleic acid

mRNA Messenger ribonucleic acid

NaCl Sodium chloride

NaOH Sodium hydroxide

NeuN Neuronal nuclei

NPC Neural stem cell/progenitor

NS Not significant

NSC Neural stem cell

OB Olfactory bulb

P0 Postnatal day 0

p16Ink4a Inhibitor of cyclin-dependent kinase 4a

p19Arf Alternative reading frame

p21 CDK-interacting protein 1

p27 Cyclin-dependent kinase inhibitor p27

p38DM p38MAPK double mutant

p38MAPK p38 mitogen-activated protein kinase

p53 Tumor protein 53

p53BP1 p53 binding protein 1

p75NTR low-affinity p75 neurotropic receptor

PBS Phosphate buffered saline

PCNA Proliferating cell nuclear antigen

PCR Polymerase chain reaction

PF Paraformaldehyde

pH3 phospho-histone 3

PI Propidium iodide

PP2Cσ Protein phosphatase 2C isoform delta

PPM1D Protein phosphatase 1D magnesium-dependent, delta isoform

pRB Retinoblastoma protein

PSA-NCAM Polysialylated neural cell adhesion molecule

pSer18 Phospho-serine 18

pSer23 Phospho-serine 23

PTEN Phosphatase and tensin homolog

RC2 Radial Glial Cell Marker-2

RMS Rostral migratory stream

ROS Reactive oxygen species

RT-PCR reverse transcription-polymerase chain reaction

S DNA Synthesis phase

SD Standard deviation

SEM Standard error mean

SGH: Singapore General Hospital

SGZ Sub-granular zone

SVZ Sub-ventricular zone

Thr Threonine

Tuj1 Neuronal class III β-Tubulin

TUNEL Terminal deoxynucleotidyl transferase dUTP nick end labeling

UV Ultra violet

VZ Ventricular zone

Wip1 Wild-type p53-induced phosphatase 1

Wip1Tg Wip1 transgenic

wt Wild type

Publication and Awards:

Zhu Yunhua, Zhang Cheng-Wu, Lu Li, Demidov Oleg N., Sun Li, Yang Lan, Bulavin Dmitry V., Xiao Zhi-Cheng (2009) Wip1 Regulates the Generation of New Neural Cells in the Adult Olfactory Bulb through p53-Dependent Cell Cycle Control Stem Cells 27(6):1433-1422

Award

Young Investigator’s Award (Basic Science)

Singapore General Hospital 17th Annual Scientific Meeting 2008

Oral presentation

“Wip1 regulate proliferation of adult neural progenitors through p53 dependent G2 cell cycle control”

Singapore General Hospital 17th Annual Scientific Meeting 2008

Date: 25 April 08, Friday

Abstract published in SGH Proceedings 17(1):45-48

Poster Presentation

“Wip1 regulates neural stem cell/progenitor proliferation through p53 dependent G2

cell cycle control”

Institute of Molecular and Cell Biology Symposium 2008

Date: 13 August 08, Wednesday

Abstract published in the symposium proceedings

Summary

Adult neurogenesis, the continual generation of new neurons from adult neural stem/progenitor cells (NPCs) in brain, represents the life-long regeneration capacity and holds great promise for cell replacement therapy to neurodegenerative diseases in elderly The regenerative capacity of NPCs decline with aging, however, the intrinsic regulations of NPCs, especially those mediating aging of NPCs, remain poorly defined This study shows that Wip1 phosphatase, previously studied in oncogenesis, functions as a crucial physiological regulator in adult neurogenesis during aging

In the forebrain, Wip1 deficiency resulted in a 90% decrease in new cell formation in adult olfactory bulb, accompanied by aberrantly decreased amplification

of NPCs, stem cell frequency and self-renewal Similarly, adult neurogenesis in hippocampus was also diminished upon knocking out Wip1 At cellular level, Wip1 knockout (ko) NPCs exhibited a prolonged cell cycle, an accumulation at G2 to M phase transition and elevated expression of cell cycle inhibitors p21 and Reprimo Two observations suggested that Wip1 regulates NPCs through p53, which are the

phosphorylation level of p53 was up-regulated in Wip1 ko NPCs in vivo and in vitro,

and transcriptions of p21 and Reprimo was p53-dependent in NPCs

Furthermore, the impaired M-phase entry and amplification of Wip1-null NPCs were both reversed in Wip1/p53 double-null genotype to that comparable to p53-null genotype Functional rescues by Wip1/p53 double null genotype were

reproduced in vivo with the number of NPCs and neuroblasts Present data

demonstrate that Wip1 regulates the generation of new neural cells in adult olfactory bulb specifically through p53-dependent M phase entry of the cell cycle of NPCs

The implication of Wip1/p53 pathway on aging of NPCs was analyzed by

regulated, which correlated with a marked down-regulation of Wip1 and a drastic regulation of p53 phosphorylation Based on these data, Wip1/p53 regulation was proposed to mediate the aging process of NPCs This study led to the identification of the important function of Wip1/p53 pathway in adult neurogenesis which possibly mediates aging of NPCs

up-Key words: Neural stem cell, neural progenitor, Wip1, PPM1D, PP2Cσ, p53, adult neurogenesis, aging

Chapter 1, General Introduction

1.1 Preface

The mammalian central nervous system (CNS) has been traditionally believed

to be non-regenerative in adulthood Only during the last two decades, the existence

of multi-potent primary neural stem cells (NSCs) in adult mammalian brain was discovered, believed and studied (Ma et al 2009) These adult NSCs continuously proliferate and differentiate into neural progenitor cells, and further differentiate into neuroblasts and functionally integrate into existing neural circuits as mature neurons The whole process evolving from NSCs to mature neurons is collectively termed as adult neurogenesis (Ming and Song 2005) Adult neurogenesis represents an important form of long-life brain plasticity and regenerative capacity It also represents a promising resource for autonomous cell replacement therapy in neurodegenerative diseases (Lie et al 2004; Muraoka et al 2006; Arias-Carrion and Yuan 2009)

Like other tissues, NSCs and neural progenitors are subjectable to malfunctions For instance, the regenerative capacity of NSCs is progressively constricted by aging (Molofsky et al 2006) Moreover, it has been reported that dysregulation of the NSC/progenitor (NPC) pool size by alteration in proliferation and self-renewal by genomic mutations can lead to brain tumors (Bachoo et al 2002; Sanai et al 2005; Zheng et al 2008; Alcantara Llaguno et al 2009) In spite of these serious consequences, the underlying mechanisms remain poorly understood Therefore, regulatory mechanisms controlling the NSC/progenitor pool size are important to be understood and subjected to active investigation, both on niche factors and intrinsic signaling networks of NSCs

In recent years, some niche factors, such as growth factors, extracellular matrix and hormones have been reported to be important in the regulation of NPC

pool size (Ming and Song 2005) Comparatively, intrinsic pathways that regulate the NPC pool size are less well studied In this study, Wip1 is identified as a critical regulator of adult neurogenesis acting through modulating p53-dependent G2 progression in NSCs and progenitors

Three components of background information will be reviewed in this chapter:

a general theme of adult neurogenesis; a summary of known functions of wild-type p53-induced phosphatase 1 (Wip1); and current knowledge of NPC regulation by Wip1 related molecules At the end of the chapter, aims, hypothesis and strategies of the current study will be presented

1.2 Adult neurogenesis

Until the end of the last century, the central dogma of non-regenerative CNS was commonly believed At the end of the 20th century and the beginning of the 21stcentury, technical advances, especially the use of 5-bromo-2-deoxyuridine (BrdU) in birth dating and lineage tracing, as well as the identification of tropic growth factors including epidermal growth factor (EGF) and fibroblast growth factor (FGF) that

enabled the in vitro culture of neural stem and progenitor cells, led to rapid progress

in the research of regenerative cells in adult brain (Ma et al 2009) A whole scheme

of adult neurogenesis and methodology has been established This section will review various aspects of adult neurogenesis including a brief description on the ontology of adult NSCs, processes in adult neurogenesis, functional significance of adult neurogenesis as well as pathological changes of NSCs/progenitors during aging and tumor transformation Finally, the rationale behind a few commonly used technologies adopted by this study will also be explained

1.2.1 Ontology of adult NSCs

In the mammalian embryo, a major source of NSCs is initially established in the neural epithelium of ventricular zone (VZ) at the anterior of the neural tube Progressively radial glial cells in sub-ventricular zone (SVZ), which are the progenies

of neural epithelium cells, behave as NSCs (Merkle and Alvarez-Buylla 2006) Radial glial cells are bipolar in shape and express the radial glial cell marker-2 (RC2) During the early postnatal stage, a subset of radial glial cells lose RC2 expression, start to express glial fibrillary acidic protein (GFAP), and transform into adult NSCs (Doetsch

et al 1999; Merkle et al 2004) A relatively smaller pool of NSCs resides in the

sub-granular zone (SGZ) in the dentate gyrus (DG) of hippocampus, which also express GFAP (Ming and Song 2005)

1.2.2 Processes in adult neurogenesis

Primary NSCs in both SVZ and SGZ are relatively quiescent, with an estimated cell cycle length of about 15 days (Morshead et al 1998) These stem cells produce transient-amplifying progenitors expressing mash1 and EGF Receptor (EGFR) (Doetsch et al 2002), which divide much faster with a cell cycle length of about 14 hours for SVZ NPCs (Smith and Luskin 1998) and 25 hours for SGZ NPCs (Cameron and McKay 2001)

In the adult forebrain, transient-amplifying progenitors proliferate and further differentiate into neuroblasts, which express doublecortin (DCX) and polysialylated neural cell adhesion molecule (PSA-NCAM) These neuroblasts tangentially migrate along SVZ, and subsequently travel through the rostral migratory stream (RMS) into the olfactory bulb (OB), where they radiate from the center to outer layers, namely the granule cell layer and the periglomerular cell layer, and become mature neurons expressing protein marker neuronal nuclei (NeuN)

In the adult hippocampus, newly generated neuroblasts in SGZ area migrate a short distance to the hippocampal granular cell layer with a leading dendritic process projecting through the hippocampal granular layer towards the molecular layer These migrating cells transiently express calretinin which is an immature marker, and express calbindin and NeuN later in matured neurons (Ming and Song 2005) Newly generated neurons gradually extend their dendrites and spines and form active synapses with neighboring cells Only cells that have established proper synaptic

connections with surrounding neurons survive whereas those that fail in making sufficient and appropriate links undergo apoptosis (Tashiro et al 2006)

It is important to note that no specific molecular marker has been identified for primary NSCs A commonly used marker for NSCs is Nestin, but Nestin is also

expressed in neuroblasts (Fukuda et al 2003) Another marker commonly used in vivo

for NSCs is GFAP (Garcia et al 2004) However, not all GFAP positive cells in SVZ are NSCs (Raponi et al 2007) In fact, GFAP marks astrocytes in general in other areas of the central nervous system As both NSCs and progenitors are proliferative and can be labeled with short-term BrdU treatment, both populations will be collectively designated as NPCs in the following sections of this thesis, unless otherwise specified

Figure 1.1: A general scheme of adult neurogenesis

(A) An overview of neurogenic regions in the adult mouse brain The sub-ventricular

zone (SVZ) and the sub-granular zone (SGZ) are indicated in green SVZ neural stem

cells (NSCs) give rise to neuroblasts, which migrate through the rostral migratory

stream (RMS) to the olfactory bulb (OB) In the OB, the majority of neuroblasts

migrate to the granular cell layer (GCL) to become mature neurons

(B) A more detailed illustration of SVZ adult neurogenesis Numbers correspond to

different stages of differentiation Arrows indicate the direction of neuroblast

migration NSCs enter mitosis and proliferate to give rise to transit amplifying cells

The transit amplifying cells further differentiate into neuroblasts which migrate

through the RMS until they reach the OB Neuroblasts then migrate radially from the

center to the peripheral of the OB Finally, neuroblasts differentiate and integrate into

the GCL and periglomerular layers

(C) A detailed illustration of SGZ adult neurogenesis Numbers correspond to

different stages of differentiation Adult hippocampal NSCs reside in the SGZ area of

dentate gyrus (DG) SGZ NSCs give rise to transit amplifying cells which

differentiate into neuroblasts The cell bodies of neuroblasts migrate into the GCL and

gradually mature into interneurons These interneurons receive inputs in the molecular

layer (ML) from neuronal projections of the entorhinal cortex and send outputs to the

pyramidal neurons in the CA3 and hilus regions through the mossy fiber pathway

1.2.3 Functional significance of adult neurogenesis

Adult NPCs have important physiological and pathological functions in vivo

Under physiological condition, proper activity of these NPCs is important for odor discrimination (Gheusi et al 2000) and learning and memory (Ming and Song 2005)

In neurodegenerative diseases and injury, NPCs can respond to the environment and adjust their proliferation and migration, which makes them a potential target for therapeutic interventions (Curtis et al 2007) Dysregulation of NPCs has also been proposed and demonstrated as a source of cancer stem cells (Sanai et al 2005) The physiological and pathological importance of adult NPCs urges a deeper understanding of the mechanisms governing their behaviors (Gage 2000)

1.2.4 NPCs in aging and transformation

A major question of stem cell biology is how to control the regenerative capacity of stem cells for tissue repair The fact that neurodegenerative diseases predominate in the elderly, in tandem with the limited regenerative capacity of tissue stem cells upon aging (Rossi et al 2008) re-emphasize the need to understand how NPCs age, particularly with the intrinsic and niche mechanisms restricting the regenerative capacity of aged stem cells

On the other hand, it has become clearer that dysregulation of NPC proliferation and self-renewal can lead to tumorigenesis (Rossi et al 2008) Recent progress has demonstrated that knocking out specific tumor suppressors specifically

in NPCs generates certain forms of brain tumor at a much higher rate than in differentiated cells (Kwon et al 2008; Zheng et al 2008; Alcantara Llaguno et al 2009) These studies have highlighted NPCs can be a cell of origin for brain tumor development

Both NPC aging and NPC transformation are intrinsically linked to physiological proliferation and self-renewal In fact, both aging related tissue degeneration and cancer have been proposed to arise from dysregulation of stem cells (Beachy et al 2004; Dontu et al 2005; Rossi et al 2008) Study of mechanisms underlying proliferation and self-renewal under physiological condition will provide insights into how NPCs age and how the loss of control of these mechanisms can possibly lead to NPC transformation into cancer stem cells (Rossi et al 2008)

1.2.5 Principle of commonly used methods in the study of adult neurogenesis

To study adult neurogenesis, several established methods have been described (Ming and Song 2005; Chojnacki and Weiss 2008) A few key chemical employed in this study will be introduced in the following paragraphs

1.2.5.1 BrdU short-term and long-term labeling

BrdU is a thymidine analogue, an easily synthesized chemical that can be incorporated into genomic DNA in the DNA synthesis phase of proliferating cells As specific antibodies are available, it has become the most commonly used method to

label proliferative cells in vivo and in vitro

Different schemes in using BrdU in NPC research can be generally categorized into two types: short-term BrdU labeling and long-term BrdU labeling The short-term BrdU labeling normally takes less than two days, and marks dividing cells in the brain including primary NSCs and neural progenitors To the contrast, long-term BrdU labeling normally lasts for several weeks At the time of visualization, the BrdU labeled cells have either re-entered the quiescent state of primary stem cells,

or have differentiated, migrated and integrated into target positions as postmitotic mature cells

The method used in this study is adopted from the study by Vanderluit et al (2004) For short-term labeling, BrdU is injected intraperitoneally once every two hours for six times within 12 hours, during which the plasma level of BrdU is sufficient to label NPCs (Taupin 2007) Mice were sacrificed 12 hours after the last injection As the average length of cell cycle of NPCs in SVZ is 14 hours (Smith and Luskin 1998), almost all the proliferating SVZ cells should have been labeled during the 12-hour tandem injection and the labeled cells should have undergone one cell division during the 12 hours after labeling and before killing Whereas in the SGZ area, NPCs have an average cell cycle length of 25 hours (Cameron and McKay 2001) The 12-hour period of tandem BrdU injections should label only about half of the total proliferating population in SGZ area, and on average the labeled NPC population should have no time to undergo a second round cell division before mice were sacrificed Nevertheless, the number of BrdU short-term labeled cells should be proportionate to the number of total NPCs

In long-term BrdU labeling experiments, mice will be sacrificed 4 weeks after the last administration (Allen et al 2001; Vanderluit et al 2004; Kippin et al 2005) Long-term BrdU labeling is used to assess the number of cells that have successfully integrated and survived in the existing neural circuits within OB and hippocampus In the SVZ, long-term BrdU labeling positive cells are primarily NSCs as they do not

migrate out of SVZ area (Vanderluit et al 2004)

1.2.5.2 Neurosphere assays

NPCs can be isolated and maintained in a serum free medium (Chojnacki and Weiss 2008), which provide a method to examine proliferation and self-renewal properties As only NSCs can form neurospheres, the number of primary neurospheres generated from isolated SVZ cells reflects the number of primary NSCs

in vivo (Morshead et al 1998) Thereby the number of primary neurospheres has been

termed as stem cell frequency (Molofsky et al 2006) The rate of secondary neurosphere formation reflects the percentage of cells that retain properties of NSCs

in vitro The average diameter of neurospheres reflects the amount of cells in each

neurosphere and is a surrogate measure for the amplification rate (Kippin et al 2005; Gil-Perotin et al 2006; Groszer et al 2006)

1.2.5.3 Differentiation assay

In the adult forebrain, NPCs differentiate and migrate into OB where cells predominantly mature as neurons (Molofsky et al 2006) Only minority of SVZ cells differentiate into oligodendrocytes in the corpus callosum (Merkle et al 2004) In the hippocampus, NSCs also dominantly differentiate into neurons (Jessberger et al

2008) As neuronal differentiation predominates in vivo, it is difficult to analyze any change in differentiation potential in vivo

The in vitro differentiation assay provides a straightforward platform to access

the differentiation potential of NPCs Upon withdraw of growth factors, NSCs spontaneously differentiate into different lineages By counting the number of neurons and glia, the differentiation potential of NSCs can be retrospectively determined

1.3 Wip1 phosphatase

1.3.1 General features of Wip1

Wip1 has been initially identified as a protein induced in response to ionizing radiation (IR) in a p53-dependent manner (Fiscella et al 1997) Wip1 has been classified into the protein phosphatase 2C (PP2C) family based on the sequence homologue and called PP2Cσ (Fiscella et al 1997), PP2C is one of the four major classes of mammalian serine/threonine specific protein phosphatases, that is monomeric, shows broad substrate specificity and is dependent on divalent cations (mainly manganese and magnesium) for its activity Wip1 is also called the protein phosphatase magnesium-dependent 1 delta (PPM1D), as magnesium ions are essential for the phosphatase activity in its catalytic domain

The expression of Wip1 mRNA has been examined and found to be ubiquitous

in embryo and adult tissues, including brain, and especially high in testis (Choi et al 2000) Subsequently, Wip1 has been found inducible in respond to many environmental stresses and the expression of Wip1 is not always p53-dependent (Takekawa et al 2000) Under UV radiation, Wip1 mRNA expression is both p38MAPK- and p53-dependent and Wip1 functions as a negative feedback regulator

of p38MAPK-p53 pathway in the recovery phase of radiation treated cells Choi et al (2000) established the first substrate of Wip1 as the conserved threonine residue Thr180 of p38MAPK, which is essential for p38MAPK activation

The gene encoding Wip1, ppm1d, is located on chromosome 17 of the human

genome (Choi et al 2000), which is often amplified in human cancers (Li et al 2002; Hirasawa et al 2003; Saito-Ohara et al 2003; Sinclair et al 2003; Ehrbrecht et al 2006; Loukopoulos et al 2007; Castellino et al 2008) To date, Wip1 has been well characterized as an oncogene in various tissues (Bulavin et al 2002; Bulavin et al

2004; Belova et al 2005; Shreeram et al 2006; Demidov et al 2007; Demidov et al 2007) The physiological function of Wip1, however, has not been well studied

Several Wip1 downstream substrates, such as p53, p38MAPK and ATM, have been implicated in the regulation of NPCs, which provokes an intriguing question as

to whether Wip1 has a function in regulating NPCs

Figure 1.2: Identified Wip1 substrates and functions

The identified Wip1 substrates are underlined Positive regulations are indicated by red arrows and negative regulations by blue lines Related functional implications have been listed below each substrate

1.3.2 Wip1 knockout mice

A Wip1 ko mouse strain has been generated by replacing the exon 4 and 5 with a neomycin cassette (Choi et al 2002) This genomic mutation has created a severely truncated Wip1 mRNA, which is not detectable by Northern blot probably because of destabilization of the mRNA Wip1 ko mice exhibit increased susceptibility to pathogens and diminished T- and B- cell function Male Wip1 ko mice exhibit defective reproductive organs and reduced longevity Wip1 ko mouse embryonic fibroblasts (MEFs) have difficulties in entering mitosis from G2 phase under resting state and have more prominent G1 arrest after gamma radiation (Choi et

al 2002)

This Wip1 ko strain has been used in multiple studies which showed resistance

to tumorigenesis in multiple tissues (Bulavin et al 2004; Harrison et al 2004; Shreeram et al 2006; Demidov et al 2007)

1.3.3 Wip1 in tumorigenesis

DNA damage checkpoints in cell cycle progression preserve the genomic fidelity of proliferating cells, which prevent the pathological accumulation of mutations required for cellular transformation (Sancar et al 2004) Since Wip1 provides negative feedback to DNA damage checkpoints, theoretically over-expression of Wip1 should promote oncogenesis and knocking out Wip1 should stabilize the genome and prevent oncogenesis Based on the above postulation, functional studies of Wip1 have been almost exclusively focused on the oncogenic effect upon over-expressing Wip1 and the tumor resistance effect upon knocking out Wip1

In 2002, two groups independently reported the function of Wip1 in breast

cancer (Li et al 2002); (Bulavin et al 2002) PPM1D, the human gene encoding

Wip1, was identified to be localized within an epicenter of the region at 17q23, which contains several candidate oncogenes and is amplified in some breast cancers (Li et al

2002) Over-expression of Wip1 in vivo abrogated serum starvation induced apoptosis

and enhanced the transformation of primary cells Another group (Bulavin et al 2002) suggested that the mechanism of attenuating p53 activity by Wip1 is probably through dephosphorylation of the upstream p38MAPK, whose activity is induced by Ras over-expression By injecting Wip1 and Ras co- over-expressed mouse MEF cells into

nude mice, they demonstrated, for the first time, that Wip1 has oncogenic potential in

vivo

Since Wip1 amplification does have a function in promoting tumor formation, inhibiting Wip1 in these cancers could be an attractive strategy in therapeutic intervention Taking advantage of Wip1 ko mice, Bulavin and colleagues (2004) have found that Wip1 ko mice are resistant to tumor formation driven by multiple types of oncogenes through p38MAPK dependent up-regulation of inhibitor of cyclin-dependent kinase 4A (p16Ink4a)/Alternative reading frame (p19Arf) proteins (Bulavin et

al 2004; Harrison et al 2004) The opposing effect of p38MAPK towards the Wip1 oncogene has been further supported by a follow-up study, where the tumor promoting effect of Wip1 was eliminated in mammary gland by p38MAPK activation (Demidov et al 2007)

The value of inhibiting Wip1 in preventing cancer should not be limited only

to breast cancers, as it has been further demonstrated in lymphomagenesis (Shreeram

et al 2006) and colo-rectal cancer models (Demidov et al 2007) Wip1 deletion enhances ATM/p53 mediated apoptosis, thus delays the onset of tumor formation in

Myc driven lymphoma formation (Shreeram et al 2006) Likewise, Wip1 deficiency inhibits the polyps formation and increases mouse survival in adenomatous polyposis coli (APC) driven colo-rectal cancer model by increasing apoptosis Interestingly, apoptosis seems to occur in intestinal cells that express stem cell markers (Demidov et

al 2007) Thereby it was proposed that Wip1 deletion in intestinal stem cells lowers the threshold of committing cells to apoptosis

Since the identification and following characterization of the oncogenic

property of Wip1, the gene encoding Wip1, ppm1d has been found to be amplified in

numerous tumor types such as breast cancer (Li et al 2002; Sinclair et al 2003; Barlund et al 2004; Natrajan et al 2009), neuroblastoma (Saito-Ohara et al 2003), medulloblastoma (Mendrzyk et al 2005; Ehrbrecht et al 2006; Castellino et al 2008), ovarian clear cell tumor (Hirasawa et al 2003; Tan et al 2009), gastric carcinoma (Fuku et al 2007), adenocarcinoma (Hirasawa et al 2003), pancreatic adenocarcinoma (Loukopoulos et al 2007) and chronic lymphocytic leukemia (Lopez-Guerra et al 2008) These reports further support the general role of Wip1 as

an oncogene

1.3.4 Wip1 in DNA repair and cell cycle regulation

In terms of cellular processes, study of Wip1 has been focused mainly on its interaction with DNA damage responsive molecules, and it becomes increasingly clear that the primary role of Wip1 is to shut down various processes in DNA damage response to restore cells into cell cycle progression Under resting conditions, the function of Wip1 on cell cycle progression has also been reported

Wip1 inhibits various molecules in the DNA damage sensing mechanisms

(Takekawa et al 2000), Wip1 also suppresses Chk1 and p53 mediated DNA damage responses by dephosphorylation of Chk1 at Ser345 and p53 at Ser15 (Lu et al 2005) Wip1 dephosphorylates both the critical Ser1981 site on ATM (Shreeram et al 2006; Batchelor et al 2008) and Chk2 downstream of ATM (Fujimoto et al 2006) Furthermore, Wip1 has also been reported to have a role in promoting MDM2 stabilization by dephosphorylating MDM2 at Ser395 (Lu et al 2007) Remarkably, Wip1 also inactivates DNA repair machinery mediated by UNG2 (Lu et al 2004; Lu

et al 2004)

Functions of Wip1 in cell cycle control were firstly suggested by the elevation

of G2/M ratio in Wip1 ko MEFs under the resting condition and more robust G1 phase arrest upon radiation (Choi et al 2002) Conversely, over-expression of Wip1 decrease the proportion of cells arrested both intra-S and G2/M upon radiation (Lu et

al 2005)

1.3.5 Possible role of Wip1 in neurogenesis

Since the identification of Wip1, growing knowledge of Wip1 has been accumulated in various stress signaling pathways and in cancer biology In contrast, the physiological function of Wip1, however, largely remains to be uncovered Given the relatively normal development and body size of Wip1 ko mice, identification of the physiological function of Wip1 remains a challenge

Wip1 is expressed in the brain (Choi et al 2000) In situ hybridization has indicated that Wip1 is expressed in regions of adult neurogenesis, such as SVZ, hippocampus and cerebellum during both development and adult stages (Heintz 2004 NCBI, GENSAT database) By real-time PCR, the expression of Wip1 in the adult

SVZ (Fig 3.15I and Fig 3.27F) and neurospheres formed from SVZ cells (Fig 3.15I)

was also confirmed Furthermore, oncogene screening has also identified Wip1 as an over-expressed protein in neuronal tumors such as medulloblastomas (Castellino et al 2008) and neuroblastomas (Saito-Ohara et al 2003) All these expression data of Wip1 suggests that Wip1 may have a function in the adult neurogenesis

Multiple cell cycle proteins, including DNA damage responsive genes such as p38MAPK, ATM and p53, have been implicated in adult neurogenesis In addition, multiple cell cycle regulators downstream of these Wip1 substrates, such as p21 and Bax downstream of ATM and p53, and Bmi1, p16Ink4a and p19Arf downstream of p38MAPK, have been reported to have functions in adult neurogenesis These existing links between Wip1 related molecules and adult neurogenesis have suggested

a possible role of Wip1 in the regulation of adult neurogenesis

Figure 1.3: Wip1 related molecules in the regulation of NPCs

The Wip1 related molecules implicated in the proliferation of NPCs are summarized

in this figure Blue lines indicate inhibitory relations and red arrows represent stimulatory activities between molecules The discontinued arrow from p38MAPK to p53 indicates the fact that the direct phosphorylation of p53 by p38MAPK has been confirmed in human but not in mouse Direct substrates of Wip1 have been underlined and molecules with direct functional links with NPCs have been highlighted with green color

1.4 Wip1 related cell cycle regulators in adult neurogenesis

As NPCs are dividing cells in the intact adult CNS, proliferation represents a very distinct activity of NPCs in the adult brain, which correlates with NPC pool size and often the self-renewal property of stem cells Therefore, cell cycle progression is crucial to the function of NPCs

This section will provide a general theme on cell cycle regulation, a review in details of Wip1 related molecules in IR induced DNA damage response (ATM- p53-p21/Bax), and molecules in stress activated signaling pathways (p38MAPK-Bmi1-p16Ink4a/p19Arf)

1.4.1 A general theme of cell cycle regulation

The cell cycle of proliferating cells is generally regulated by critical time points named restriction points and checkpoints (Zetterberg et al 1995; Malumbres and Barbacid 2001; Sherr 2004; Machida et al 2005) After passing through mitosis phase (M), a cell could enter a state called G1A, a competent phase of Gap phase 1 (G1), during which the cell needs continuous growth factor stimuli to progress through the restriction point When environment surrounding a cell is not favoring for proliferation, the cell will enter and get arrested in a quiescent state (G0) Upon continuous growth factor stimuli, a quiescent cell accumulate intrinsic changes in signaling and eventually re-enters the cell cycle through a restriction point into gap phase 1 (G1B), where it is committed to proliferate through all the following phases namely, gap phase 1 (G1), DNA synthesis phase (S), gap phase 2 (G2) and mitosis phase (M) (Pardee 1974; Zetterberg and Larsson 1985) The length of each gap phase varies due to delays at checkpoints thus the total cell cycle length varies For instance,