Role of the tumor suppressor ARF and the p53-pathway in retinoblastoma development

by

Kwong-Him To

A thesis submitted in conformity with the requirements

for the degree of Master of Science

Graduate Department of Molecular Genetics

University of Toronto

© Copyright by Kwong-Him To (2009)

ii

Role of the tumor suppressor ARF and the p53-pathway in retinoblastoma development Master of Science, 2009 Kwong-Him To Graduate Department of Molecular Genetics University of Toronto

Abstract

Retinoblastoma development is a multistep process, and inactivation of the RB1 gene is

not sufficient for tumorigenesis. Previous studies suggest that the p53-tumor suppressor is

inactivated due to overexpression of p53-antagonists MDM4 and MDM2. This thesis

evaluates the importance of ARF, a p53-activator that inhibits MDM2. In retinoblastomas,

ARF protein is nearly undetectable despite robust mRNA expression. Chemical inhibition of

the proteasome, which regulates ARF protein-turnover, did not result in ARF accumulation

in retinoblastoma cells, indicating that ARF protein was not aberrantly degraded by the

proteasome. During mouse retinoblastoma development, Arf protein was expressed at low

level, and p53-target genes involved in cell cycle arrest and autoregulation were not activated.

Overexpression of ARF in retinoblastoma cells led to growth inhibition, accompanied by

increased expression of p53 and p53-transcriptional targets. Taken together, our data

suggests that low ARF protein is an important factor in silencing of the p53-pathway during

retinoblastoma development.

iii

Acknowledgement

I would like to thank my supervisor Dr. Brenda Gallie for her guidance and support

throughout my graduate study. Her genuine interest in science and commitment to patient

care are truly inspiring. In addition, I am thankful to members of my supervisory committee,

Drs Rod Bremner, Michael Moran and Vuk Stambolic, for their valuable advices that were

instrumental to the completion of this thesis.

I would also like to thank members of the Gallie lab for the insightful discussions and

advices that shaped this Master thesis: Sanja Pajovic, Clarellen Spencer, Marija Orlic-

Milacic, Mellone Marchong, Tim Corson, Helen Dimaras, Ying Guo, Brigitte Theriault,

Stephanie Yee, Christine Yurkowski, Ghada Kurban, Sofia Khan, Michael Killian and

Harvey Lim.

Outside the lab, I am grateful to my parents Yuen-Chuen and Man-Ying To, my brother

Kai, and Winnie Lo for their supports.

Finally, I would like to thank Sandra and David Smith, and the Vision Science Research

Program for graduate student funding.

iv

Table of Contents

Abstract................................................................................................................................ ii

Acknowledgement............................................................................................................... iii

Table of Contents .................................................................................................................iv

List of Tables .......................................................................................................................ix

List of Figures .......................................................................................................................x

List of Abbreviations.......................................................................................................... xii

Chapter 1: Introduction..........................................................................................................1

1.1. Retinoblastoma: Disease manifestation ................................................................1

1.2. Current treatment for retinoblastoma....................................................................1

1.3. Retinal development ............................................................................................2

1.4. pRB molecular function .......................................................................................3

1.5. Retinal specific Rb-knockout in mice ...................................................................3

1.6. Rb-p107 and Rb-p130 compound mutants............................................................4

1.7. Multistep model of retinoblastoma development ..................................................5

1.8. Genomic changes in retinoblastoma .....................................................................7

1.8.1. 16q loss.........................................................................................................8

1.8.1.1. CDH11 ................................................................................................8

1.8.1.2. RBL2....................................................................................................8

1.8.2. 6p gain ..........................................................................................................9

1.8.2.1. DEK & E2F3 .......................................................................................9

1.8.3. 1q gain ..........................................................................................................9

v

1.8.3.1. KIF14 ..................................................................................................9

1.8.3.2. MDM4 ...............................................................................................10

1.9. ARF and the p53-pathway in retinoblastoma......................................................11

1.10. Role of ARF in the crosstalk between the pRB- and p53-pathway....................12

1.11. INK4a/ARF in cancer development ..................................................................12

1.12. ARF protein and functions ...............................................................................15

1.12.1. ARF protein ..............................................................................................15

1.12.2. p53-dependent functions ...........................................................................16

1.12.3. p53-independent functions ........................................................................17

1.12.3.1. Evidence from compound mutant studies .........................................17

1.12.3.2. Regulation of transcription factors....................................................18

1.12.3.3. Ribosome biogenesis........................................................................18

1.12.3.4. Link to ATM and ATR.....................................................................19

1.12.3.5. Sumoylation.....................................................................................19

1.12.3.6. smARF and autophagy .....................................................................19

1.13. Regulation of ARF expression..........................................................................20

1.13.1. Transcriptional repressors .........................................................................20

1.13.2. Translational repressor ..............................................................................21

1.13.3. Transcriptional activators ..........................................................................21

Chapter 2: Role of ARF and the p53-pathway in retinoblastoma development .....................23

2.1. Introduction .......................................................................................................23

2.2. Hypotheses ........................................................................................................23

2.3. Thesis aims & rationales ....................................................................................23

vi

2.3.1. Determine the cause of low ARF protein in retinoblastomas........................23

2.3.2. Examine the expression of Arf, p53 & p53-transcriptional targets during early

retinoblastoma development in a mouse retinoblastoma model ....................................24

2.3.3. Determine the effect of exogenous ARF expression in retinoblastoma cells.24

2.4. Material and methods.........................................................................................24

2.4.1. Tissue culture..............................................................................................24

2.4.2. Mouse model of retinoblastoma ..................................................................25

2.4.3. Tissue harvesting and fixation .....................................................................26

2.4.4. RT-PCR......................................................................................................27

2.4.5. Real-time PCR ............................................................................................28

2.4.6. Western Blotting .........................................................................................29

2.4.7. Immunofluorescence staining......................................................................30

2.4.8. Exogenous gene expression using adenoviruses ..........................................31

2.5. Results ...............................................................................................................32

2.5.1. Cause of low level of ARF protein in retinoblastomas .................................32

2.5.1.1. Analysis of ARF expression in retinoblastoma cell lines.....................32

2.5.1.2. Role of the proteasome pathway in ARF protein degradation in

retinoblastoma cell lines .................................................................................36

2.5.1.3. Analysis of miR-24 expression in retinoblastoma cell lines ................38

2.5.2. Expression of Arf, p53 & p53-downstream targets during early

retinoblastoma development in a mouse retinoblastoma model ....................................39

2.5.2.1. Expression of Arf during early mouse retinoblastoma development....39

vii

2.5.2.2. Expression of p53 and p53-transcriptional targets during early mouse

retinoblastoma development ...........................................................................45

2.5.3. Effect of exogenous ARF expression in a retinoblastoma cell line ...............46

2.6. Discussion .........................................................................................................53

2.6.1. Cause of low ARF protein in retinoblastomas..............................................53

2.6.1.1. Low ARF protein in retinoblastomas is unlikely due to aberrant protein

degradation.....................................................................................................53

2.6.1.2. Role of microRNAs in regulation of ARF translation..........................54

2.6.1.3. Block in ARF protein synthesis may be overcome by elevation in ARF

mRNA expression ..........................................................................................54

2.6.2. Expression of Arf, p53 & p53-transcriptional targets during early

retinoblastoma development in a mouse retinoblastoma model ....................................55

2.6.2.1. ARF protein is expressed at low level during early mouse

retinoblastoma development ...........................................................................55

2.6.2.2. Despite increased p53 expression, some p53-target genes remained

inactive upon Rb-inactivation .........................................................................56

2.6.2.3. Several factors might influence the transactivating activity of p53......58

2.6.3. Effect of exogenous ARF expression in retinoblastoma cell line..................60

2.6.3.1. The p53-pathway can be activated by exogenous ARF in

retinoblastoma cells ........................................................................................60

Chapter 3: Discussion..........................................................................................................61

3.1. Tumor surveillance by the ARF-p53-pathway in retinoblastoma development ...61

3.2. ARF-MDM2/4-p53 pathway: Therapeutic target in retinoblastomas?.................62

viii

3.2.1. Targeting MDM2 and MDM4 .....................................................................62

3.2.2. Targeting ARF ............................................................................................62

3.2.3. Targeting p53..............................................................................................64

3.3. Future directions ................................................................................................64

References...........................................................................................................................67

ix

List of Tables

Table 1. Primer sequences and PCR conditions for genotyping............................................26

Table 2. Primer sequences and PCR conditions for human and mouse gene expression

analyses...............................................................................................................................28

x

List of Figures

Figure 1. Multistep model of retinoblastoma development.....................................................6

Figure 2. Schematic of the INK4a/ARF locus.......................................................................13

Figure 3. Western Blot analysis of ARF and p21 in retinoblastoma and control cell lines ....33

Figure 4. ARF mRNA expression in retinoblastoma and control cell lines ...........................34

Figure 5. Western Blot analysis for ARF and p21 in retinoblastoma cell lines after MG132

treatment .............................................................................................................................37

Figure 6. Expression analysis of miR-24 in control and retinoblastoma cell lines.................38

Figure 7. Hematoxylin and Eosin staining of Rb-knockout retinas in p107+/+ or p107-/-

background at P9.................................................................................................................40

Figure 8. RT-PCR analysis for Arf in Rb-knockout retinas from P0 to P15 ..........................41

Figure 9. Immunostaining for Arf protein in Rb-knockout retinas in p107+/+ or p107-/-

background at P9.................................................................................................................43

Figure 10. Western Blot analysis of Arf in Rb-knockout retina in p107+/+ or p107-/-

background..........................................................................................................................45

Figure 11. Western Blot analysis of p53, and p53-transcriptional targets p21 and Bax in Rb-

knockout retinas in p107+/+ or p107-/- background ...............................................................47

Figure 12. mRNA expression of p53-transcriptional targets in Rb-knockout retinas in p107-/-

background..........................................................................................................................48

Figure 13. mRNA expression of p53-transcriptional targets in Rb-knockout retinas in p107+/+

background..........................................................................................................................49

Figure 14. Exogenous expression of ARF in the retinoblastoma cell line WERI...................51

Figure 15. Effect of ARF overexpression on gene expression and cell growth .....................52

xi

Figure 16. Role of the Arf-p53 pathway in early mouse retinoblastoma development ..........59

xii

List of Abbreviations

AML Acute myeloid leukemia

AP Alkaline phosphatase

ARF Alternate reading frame

ATM Ataxia telangiectasia mutated

ATR Ataxia telangiectasia and Rad3 related

Bax BCL2-associated X protein

BCIP 5-Bromo-4-Chloro-3'-Indolyphosphate p-Toluidine

BCL-2 B-cell CLL/lymphoma 2

BCL-XL BCL2-like 1

Bmi-1 Bmi1 polycomb ring finger oncogene

bp Base pair

BrdU 5-bromo-2-deoxyuridine

CBX Chromobox homolog

CDC6 Cell division cycle 6 homolog

CDH11 Cadherin-11

CDH13 Cadherin-13

CDK Cyclin-dependent kinase

cDNA Complementary DNA

CGH Comparative genomic hybridization

Chx10 C elegans ceh-10 homeo domain-containing homolog

Cre Cyclization recombinase

DAPI 4’, 6-diamidino-2-phenylindole

DAP-kinase Death-associated protein kinase

DEK DEK oncogene (DNA binding)

DMEM Dulbecco's modified eagles media

DNA Deoxyribonucleic acid

DR5/KILLER Tumor necrosis factor receptor superfamily, member 10b

E Embryonic day

xiii

E2F E2F transcription factor

Egr1 Early growth response 1

Eµ Immunoglobulin heavy chain enhancer

FBS Fetal bovine serum

FISH Fluorescence in situ hybridization

Gadd45 Growth arrest and DNA-damage-inducible alpha

GCL Ganglion cell layer

gfp Green fluorescence protein

H3 H3 histone

hnRNP K Heterogeneous nuclear ribonucleoprotein K

H-RAS v-Ha-ras Harvey rat sarcoma viral oncogene homolog

HRP Horseradish peroxidase

INL Inner nuclear layer

JunD Jun proto-oncogene related gene d

KIF14 Kinesin family member 14

KO Knockout

M1 Mutation 1

M2 Mutation 2

MDM Modified Dulbecco's Medium

Mdm2 Transformed mouse 3T3 cell double minute 2

Mdm4 Transformed mouse 3T3 cell double minute 4

MEFs Mouse embryonic fibroblasts

MEL-6 Polycomb group ring finger 2 (PCGF2)

min Minute

miR MicroRNA

MOI Multiplicity of infection

MYC v-myc myelocytomatosis viral oncogene homolog

NBT Nitro-blue tetrazolium chloride

NFκB Nuclear factor of kappa light polypeptide gene enhancer in B-cells

xiv

Noxa Phorbol-12-myristate-13-acetate-induced protein 1 (Pmaip1)

NPM Nucleophosmin (B23)

ONL Outer nuclear layer

P Postnatal

p107 Retinoblastoma-like 1

p130 Retinoblastoma-like 2

p15INK4b Cyclin-dependent kinase inhibitor 2B

p16INK4a Cyclin-dependent kinase inhibitor 2A

p21 Cyclin-dependent kinase inhibitor 1A

p53 Tumor suppressor protein p53

Pax-6 Paired box gene 6

PBS Phosphate buffered saline

PcG Polycomb group

PHC2 Polyhomeotic homolog 2

PMH/OCI Princess Margaret Hospital/Ontario Cancer Institute

pRB Retinoblastoma protein

Puma BCL2 binding component 3 (BCC3)

RB Retinoblastoma

RB1 Retinoblastoma tumor suppressor gene

RBL2 Retinoblastoma-like 2 (p130)

RING1b Ring finger protein 2 (RNF2)

RITA Reactivation of p53 and induction of tumor cell apoptosis

RNA Ribonucleic acid

RNAi RNA-interference

RNU6B RNA, U6 small nuclear 2

RPMI Roswell Park Memorial Institute

RT-PCR Reverse transcriptase-polymerase chain reaction

SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis

shRNA Short-hairpin RNA

xv

siRNA Short-interfering RNA

smARF Short mitochondrial ARF

STS Sequence tagged site

TBP TATA box binding protein

TBS Tris-buffered saline

TBX T-box

TP53 Tumor protein p53

U Units

UTR Untranscribed region

WERI Wills Eye Research Institute

WRN Werner Syndrome helicase

Wt Wild-type

ZBTB7A Zinc finger and BTB domain containing 7A, Pokemon

1

Chapter 1: Introduction

1.1. Retinoblastoma: Disease manifestation

Retinoblastoma, affecting approximately 1 in 18,000 live births (Devesa, 1975), is a

pediatric ocular cancer that originates from the developing retina. The disease is initiated by

two mutational events (M1 and M2), as originally postulated by Alfred Knudson in his Two-

Hit hypothesis (Knudson, 1971). Subsequent seminal discoveries showed that the two hits

consisted of inactivation of both alleles of the tumor suppressor gene, RB1 (Cavenee et al.,

1983; Comings, 1973; Friend et al., 1986; Fung et al., 1987; Lee et al., 1987). Patients with

retinoblastoma may have the first hit as a germline mutation in one allele of RB1, either

inherited from one of the parents or arising sporadically during early development, and

subsequent somatic mutation inactivates the second allele in the retina. Alternatively, both

hits may occur sporadically in the same cell within the developing retina. Patients with

germline RB1 mutation carry one mutant RB1 allele in every cell of the body, and are

predisposed to heritable retinoblastoma in both eyes (bilateral retinoblastoma) early in life

and second malignancies in other tissues later in life (Marees et al., 2008). Affected

individuals without germline mutation manifest nonheritable retinoblastoma in only one eye

(unilateral retinoblastoma).

1.2. Current treatment for retinoblastoma

Treatment for retinoblastoma varies depending on the laterality of the disease. In

developed countries, unilateral retinoblastoma is cured at a high rate (>95%) by enucleation

of the affected eye, while bilateral retinoblastoma requires multimodality therapy consisting

of local treatments, chemotherapy or external beam radiation therapy, in order to save vision

2

(Chintagumpala et al., 2007). Current treatment can be effective in curing retinoblastoma, but

has considerable side effects. In addition to compromised vision as a result of enucleation

and local treatments, retinoblastoma patients have to withstand the toxic effects of systemic

chemotherapy and in some cases, external beam radiation. Radiation markedly increases the

chance of second malignancies, especially in patients carrying germline RB1 mutation (Chan

et al., 2005). Therefore, treatment with minimal side effects is urgently needed in order to

maximize vision preservation, and to minimize the chance of inducing second malignancies

in retinoblastoma patients. The development of such targeted therapy requires a thorough

understanding of the genetic and molecular mechanisms governing retinoblastoma survival

and growth, and the identification of tumor specific markers that can be targeted by therapy.

Moreover, the early onset of retinoblastoma indicates that tumorigenesis is intertwined with

retinal development; therefore it is important to understand the role of the protein encoded by

RB1, pRB, in the developing retina, and to dissect the molecular response to RB1 inactivation.

1.3. Retinal development

Mammalian eye development begins with the bilateral evagination of the neural tube,

forming a pair of optic vesicles that induce the surface ectoderm to form the lens placodes.

Subsequent invagination of the lens placode and the optic vesicle results in the formation of

the lens vesicle and a double-layered optic cup. The inner layer of the optic cup eventually

forms the neural retina, while the outer layer of the optic cup gives rise to the retinal pigment

epithelium (Chow and Lang, 2001).

The early neural retina consists of a neuroblastic layer, in which retinal progenitor cells

divide and give rise to the seven main classes of retinal cells in a specific birth order that is

evolutionarily conserved. In the mature retina, cells are organized into three distinct layers.

3

The outer nuclear layer consists of rod and cone photoreceptors, while the inner nuclear layer

is made up of bipolar, Müller glia, amacrine and horizontal cells. The ganglion cell layer is

home for ganglion and amacrine cells (Dyer and Cepko, 2001).

1.4. pRB molecular function

The best characterized function of the retinoblastoma protein, pRB, is to control cell

cycle progression by interacting with and inhibiting the E2F-family of transcription factors,

which regulate the expression of genes required for cell proliferation (Cobrinik, 2005; Dyson,

1994). The affinity of pRB for E2Fs (E2F-1, -2, -3a, -3b) depends on its phosphorylation status,

which is modulated by cyclins and cyclin-dependent kinases (CDK). In the presence of

proliferative signals, elevated activity of cyclin-CDK complexes phosphorylates pRB, leading to

the dissociation of E2Fs from pRB. Consequently, E2Fs activate their transcriptional targets,

thereby promoting cell cycle entry and cell division. Given the pivotal role of pRB in cell cycle

control, RB1 mutations confer on cells the capacity to enter the cell cycle and divide

inappropriately, or allow cells destined for cell cycle exit to remain proliferative.

1.5. Retinal specific Rb-knockout in mice

The development of a valid mouse model for retinoblastoma has proven to be difficult. In

mice, germline homozygous deletion of Rb led to embryonic lethality (Jacks et al., 1992), and it

was necessary to circumvent the developmental defect by the generation of Rb-mutant chimeric

mice (Robanus-Maandag et al., 1998). However, generation of chimeric Rb-mutant mice was

laborious and low yield, leading to the demand for a mouse model with heritable retinal specific

Rb-inactivation. This demand was finally met in 2004, when several groups independently

developed mouse models for retinoblastoma using the Cre-lox technology (Macpherson, 2008;

Pacal and Bremner, 2006). Various promoters, including those of Nestin, Chx-10 and Pax-6,

4

were employed, with each promoter driving Cre expression that excises Rb in a unique spatial

and temporal pattern in the developing retina (Chen et al., 2004; MacPherson et al., 2004; Zhang

et al., 2004). Nevertheless, inactivation of Rb driven by the different promoters give rise to

common phenotypes during retinal development, consisting of ectopic cell division, increased

apoptosis and defects in differentiation. Chen et al later confirmed that E2f1 and E2f3a are the

main effectors of the aberrant phenotypes induced by Rb-inactivation, as the deletion of E2f1

rescues the abnormalities in proliferation and apoptosis of Rb-deficient retinas. Although ablation

of E2f3a does not rescue aberrant proliferation or apoptosis, it reverses the differentiation defect

of a subset of amacrine cells (Chen et al., 2007).

Despite causing ectopic division during mouse retinal development, Rb-inactivation does

not lead to retinoblastoma formation. In fact, all Rb-deficient cells eventually exit the cell

cycle and differentiate into one of the seven cell types in the mature retina (Chen et al., 2004;

MacPherson et al., 2004). Retinoblastoma development in mice requires the simultaneous

inactivation of Rb and either one of its homologs, p107 or p130 (Chen et al., 2004;

MacPherson et al., 2007; MacPherson et al., 2004; Robanus-Maandag et al., 1998).

1.6. Rb-p107 and Rb-p130 compound mutants

Collectively, pRb, p107 and p130 are known as the pocket proteins owing to the

conserved pocket domain they share, which is important for E2fs binding. Like pRb, p107

and p130 are important cell cycle regulators, whose activities are modulated by cyclin-CDK

complexes (Classon and Dyson, 2001; Cobrinik, 2005). In mice with p107 homozygous

deletion, retinal specific inactivation of Rb leads to exacerbated ectopic division and

apoptosis compared to that in mice with wild-type p107; about 60% of the compound mutant

animals develops retinoblastoma (Chen et al., 2004; MacPherson et al., 2004). Mice with Rb-

5

inactivation in a p130 mutant background are even more tumor-prone, with all compound

mutant animals developing retinoblastoma (MacPherson et al., 2007).

1.7. Multistep model of retinoblastoma development

Data from transgenic mice indicates that Rb-inactivation alone is not sufficient for

retinoblastoma development. Similar evidence from the characterization of the precursor to

human retinoblastoma, retinoma, also supports that RB1 inactivation alone is not sufficient

for malignant tumorigenesis (Dimaras et al., 2008). Retinoma is a benign lesion that can be

found adjacent to some malignant retinoblastomas. Dimaras et al showed that despite the

presence of mutations in both alleles of RB1, retinoma remains non-proliferative and

expresses markers of senescent cells. In addition, compared to its adjacent malignant

counterpart, retinoma possesses fewer genomic alterations, suggesting that these additional

changes are necessary for malignant transformation (Dimaras et al., 2008).

Therefore, data from transgenic mouse studies and retinoma supports that, like other

cancers, retinoblastoma development is a multistep process that requires additional molecular

alterations subsequent to the inactivation of RB1 (Vogelstein and Kinzler, 1993). Evidence

suggests that with pRB inactivation alone, tumor development does not progress beyond the

benign retinoma stage, perhaps due to the presence of additional tumor suppressors that

prevent malignant transformation. This model suggests that malignant retinoblastoma

develops only when these secondary tumor surveillance systems are abrogated due to their

inactivation (Figure 1).

A major goal in retinoblastoma research is to identify the alterations subsequent to RB1

mutations that promote malignant development. Retinoblastomas exhibit high level of

genomic instability presented as recurrent chromosomal gains and losses. Genomic losses

6

RB1+/+ RB1-/-

RetinomaMalignant development

Tumor suppressors

RB1+/+ RB1-/-

Retinoma

Tumor suppressors

B

A

oncogenes

RB1+/+ RB1-/-

Retinoma

Tumor suppressors

C

oncogenes

Malignant development

Figure 1. Multistep model of retinoblastoma development A. Inactivation of both alleles of RB1 alone is not sufficient for malignant retinoblastoma development due to the presence of other tumor suppressors, which prevent tumor progression beyond the benign retinoma stage. B. Malignant retinoblastoma develops only when these tumor suppressors are inactivated, or when gain of oncogenic function allows RB1-mutant cells to evade tumor surveillance. The exact number of oncogene and tumor suppressor involved, and the order in which the gene modifications occurs are not known. C. Retrospective review of archived paraffin-embedded sections of eyes enucleated for retinoblastoma revealed the presence of retinoma contiguous with retina and retinoblastoma in 15.6% (20/128) of cases examined (Dimaras et al, 2008). In the cases where no retinoma was detected, retinoblastoma development might have bypassed the benign retinoma stage.

7

may confer growth advantages to corrupted cells by deletions of tumor suppressor genes.

Alternatively, copy number gain of oncogenes due to genomic instability may enhance

proliferative capacity of corrupted cells, or enable mutant cells to evade clearance by tumor

surveillance mechanisms.

1.8. Genomic changes in retinoblastoma

Early karyotype analyses of retinoblastomas reported frequent gain of chromosome 1

and 6, and the loss of chromosome 16 (Benedict et al., 1983; Chaum et al., 1984; Kusnetsova

et al., 1982; Pogosianz and Kuznetsova, 1986; Squire et al., 1985). With the advent of the

comparative genomic hybridization (CGH) methodology, several groups were able to

characterize genomic changes in retinoblastomas at higher resolution (Chen et al., 2001;

Herzog et al., 2001; Lillington et al., 2003; Mairal et al., 2000; van der Wal et al., 2003;

Zielinski et al., 2005). Based on the six published CGH studies, the most

frequently gained regions in retinoblastomas are 1q (53%) and 6p (54%), with the most

frequently lost region being 16q (32%) (Corson and Gallie, 2007). Genomic changes in

retinoblastomas often encompass multiple genes, making it challenging to deduce the true

target that drives the gain or loss of a particular chromosomal region. Nevertheless, based on

the premise that the true target gene(s) is differentially expressed in retinoblastoma compared

to normal tissue, genes that showed similar levels of expression in both tumor and normal

tissues can be excluded and considered “passengers” because of their genomic proximity to

the true cancer genes. This approach of identifying tumor suppressors and oncogenes has

yielded several interesting candidates, which may be useful for future therapeutic

development.

8

1.8.1. 16q loss

1.8.1.1. CDH11

Marchong et al identified CDH11 and CDH13 as the most frequently lost genes on 16q

based on published CGH results and microsatellite markers on 16q (Marchong et al., 2004).

Expression analysis showed a reduction in protein expression for CDH11, but not CDH13 in

retinoblastomas compared to normal adult retinas, suggesting that CDH11 is a target for 16q

loss in retinoblastomas (Marchong et al., 2004). CDH11 belongs to the cadherin family of

membrane bound glycoproteins involved in cell-cell adhesion. Homozygous deletion of

Cdh11 led to faster growing retinoblastomas in the TAg-RB murine model (Marchong and

Yurkowski, personal communication).

1.8.1.2. RBL2

Although not as frequently lost as CDH11, another interesting gene located on 16q is

RBL2, encoding p130, which strongly cooperated with pRb in suppressing mouse

retinoblastoma development (MacPherson et al., 2007; MacPherson et al., 2004).

MacPherson et al showed that deletion of p130 in concert with Rb led to aggressive bilateral

retinoblastoma, while no tumor was observed with Rb inactivation alone (MacPherson et al.,

2007; MacPherson et al., 2004). In human, Dimaras et al showed that p130 protein was

strongly expressed in the benign retinoma, but its expression was decreased in adjacent

malignant retinoblastoma (Dimaras et al., 2008). Using similar techniques, Bellan et al also

showed a reduction of p130 in their retinoblastoma cohort (Bellan et al., 2002). Cautions

have to be taken while interpreting these data since p130 protein is cell cycle regulated and is

normally not present in proliferating cells (Tedesco et al., 2002). Further studies are required

9

to confirm that the reduced p130 expression observed in retinoblastomas is tumor specific

due to genomic loss, and not the result of proliferation.

1.8.2. 6p gain

1.8.2.1. DEK & E2F3

DEK and E2F3 are implicated to be the target genes of 6p gain in retinoblastomas

(Grasemann et al., 2005; Orlic et al., 2006). DEK is a phosphoprotein with suggested roles in

transcriptional regulation, chromatin architecture and mRNA splicing (Waldmann et al.,

2004). Its level of expression correlates with histological grade, aggressiveness or

invasiveness in several cancer types (Kondoh et al., 1999; Lu et al., 2005; Sanchez-Carbayo

et al., 2003; Savli et al., 2002). E2F3 encodes a transcription factor that regulates the cell

cycle and proliferation, and has been shown to be the target of 6p gain in bladder cancer

(Oeggerli et al., 2006; Olsson et al., 2007; Veltman et al., 2003). In a retinoblastoma cell line

with extra copies of DEK and E2F3, shRNA-knockdown of DEK or E2F3 led to decreased

proliferation, while DEK knockdown also led to cell death (Orlic et al, submitted). This data

indicates that DEK and E2F3 play important roles in proliferation, and survival in the case of

DEK, and may serve as useful therapeutic targets in retinoblastomas.

1.8.3. 1q gain

1.8.3.1. KIF14

Based on CGH results and analysis of sequence tagged site (STS) markers on 1q, Corson

et al identified the most frequently gained region to be within the chromosomal band 1q32

(Corson et al., 2005). Among the genes within that region, KIF14 is the only gene up-

10

regulated in retinoblastomas compared to normal adult retinas (Corson et al., 2005). KIF14

encodes a mitotic kinesin involved in cytokinesis (Carleton et al., 2006; Gruneberg et al.,

2006). KIF14 knockdown using siRNA in cervical cancer and non-small cell lung cancer cell

lines leads to reduced proliferation and colony formation on soft agar (Corson et al., 2007).

Similar results were also observed in retinoblastoma and ovarian cancer cell lines (Brigitte

Theriault, personal communication).

1.8.3.2. MDM4

Laurie et al identified MDM4 as another candidate oncogene of 1q gain in

retinoblastomas (Laurie et al., 2006). MDM4 is a negative regulator of the p53-tumor

suppressor, by inhibiting p53-transcriptional activity and stabilizing MDM2, which is an E3-

ligase that ubiquitinylates p53 for proteasomal degradation (Marine and Jochemsen, 2004).

Therefore, overexpression of MDM4 is one way to inactivate the p53-tumor suppressor

pathway in cancer development (Toledo and Wahl, 2006). In mice, inactivation of p53

accelerates the kinetics of retinoblastoma development in Rb-p107 compound mutants

(Zhang et al., 2004). In human, Laurie et al reported copy number gain of MDM4, using

fluorescence in situ hybridization (FISH), in 63% of retinoblastomas examined (Laurie et al.,

2006). MDM4 mRNA and protein are overexpressed in human retinoblastomas compared to

normal fetal retinas, and ectopic expression of Mdm4 leads to more aggressive tumors in

mice with Rb and p107 deletions (Laurie et al., 2006). Guo et al examined MDM4 expression

in another cohort of retinoblastomas using normal adult retinas as controls, and found no

differential expression in MDM4 mRNA or protein (Guo et al., 2008). This suggests that

MDM4 may be up-regulated during normal retinal development, which should be taken into

account when targeting MDM4 for therapy, since the therapeutic window for cancer-specific

11

killing may narrow as the patients age. Further analyses of MDM4 expression in normal

retinas of different ages relevant to retinoblastoma (eg. at birth and early childhood) are

required.

In another study, Dimaras et al examined both KIF14 and MDM4 copy number changes

on chromosome 1q simultaneously in retinoblastomas using FISH (Dimaras et al., 2008).

They found that MDM4 gain was observed only in combination with gain of KIF14, while

KIF14 showed genomic gain even when MDM4 copy number normal. This suggests that

KIF14 may be the more important 1q oncogene in retinoblastoma development.

1.9. ARF and the p53-pathway in retinoblastoma

The p53-pathway serves as an important tumor surveillance mechanism that responds to

oncogenic signals by triggering cell death or cell cycle arrest, and its inactivation is an

obligatory step in many cancers (Whibley et al., 2009). Retinoblastoma was believed to be an

exception since no mutation in TP53 has ever been reported. Laurie et al reported MDM4 and

MDM2 copy number gain in 63% and 10% of retinoblastomas, respectively, inferring that the

p53-pathway is functionally inactivated in these tumors (Laurie et al., 2006). A key

component of the p53-pathway is the upstream activator, ARF, which transmits oncogenic

signals to activate p53 by inhibiting the p53-antagonist MDM2. ARF is frequently inactivated

in cancers, resulting in a compromised p53-tumor surveillance pathway. The role of ARF in

retinoblastoma was dismissed by Laurie et al., who reported an elevation of ARF mRNA in

retinoblastomas compared to normal fetal retinas, and proposed that high ARF expression is

functionally irrelevant due to overexpression of MDM4 and MDM2 (Laurie et al., 2006).

However, Guo et al showed that despite robust mRNA expression, ARF protein was

undetectable in retinoblastomas (Guo et al., 2008). This observation raises the possibility that

12

low level of ARF protein may prevent p53 activation, and contribute to silencing of the p53-

pathway in retinoblastoma.

1.10. Role of ARF in the crosstalk between the pRB- and p53-pathway

In vitro and in vivo studies suggest that ARF is important in conveying the signal

initiated by pRB inactivation to trigger p53-mediated tumor surveillance during cancer

development. The immediate consequence of pRB inactivation is the derepression of E2F1

transcription factor, which, in certain contexts, is capable of inducing ARF transcription.

E2F1 regulates ARF through a novel E2F-responsive element that varies from the typical

E2F site (Komori et al., 2005). This element responds to the aberrant E2F1 activity resulting

from pRB inactivation or ectopic E2F1 expression, but not to the normal physiological

fluctuation of E2F1 during the cell cycle. Therefore, upon pRB inactivation, aberrant E2F1

activity may activate ARF, which in turn activates p53 to suppress tumorigenesis. Indeed,

ARF-mediated tumor suppression was demonstrated in a mouse pituitary tumor model

initiated by Rb mutation, in which Arf is activated upon Rb-inactivation (Tsai et al., 2002).

Moreover, Arf-knockout animals with Rb mutations developed more aggressive pituitary

tumors than those carrying normal copies of Arf.

1.11. INK4a/ARF in cancer development

ARF is encoded by the INK4a/ARF locus, which also codes for the important tumor

suppressor, p16INK4a (Figure 2). ARF and p16INK4a share common exon-2 and -3 while having

unique promoters and exon-1. The two proteins have no sequence homology, resulting in

entirely different molecular functions (Sherr, 2001). p16INK4a prevents phosphorylation of the

pocket proteins by inhibiting CDK4/6, thereby blocking cell cycle progression. Interestingly,

13

E1β E1α E2 E3

p16INK4aARF

Figure 2. Schematic of the INK4a/ARF locus ARF and p16INK4a share exon-2 (E2) and exon-3 (E3), while having unique promoters and exon-1 (E1) that recruit different sets of transcription factors. ARF and p16INK4a proteins have no sequence homology, since translation of the two genes utilizes different reading frames. Expression of ARF and p16INK4a can be repressed coordinately by PcG proteins and CDC6 through chromatin remodeling.

14

it is highly expressed in the benign retinoma, but not in malignant retinoblastoma, suggesting

a potential tumor suppressor role of p16INK4a in retinoblastoma development (Dimaras et al.,

2008). However, ectopic expression of p16INK4a in RB1-/- retinoblastoma cells did not induce

cell cycle arrest (unpublished data). This result is consistent with previous report, which

shows that p16INK4a function requires intact pRB (Bruce et al., 2000). Therefore, p16INK4a is

unlikely to play an important role in retinoblastoma development due to the lack of

functional pRB.

In many cancers, ARF is inactivated through co-deletion with p16INK4a (Rocco and

Sidransky, 2001). Deletions affecting only exon-1β (specific to ARF) have also been

observed, but not as frequently as co-deletion with p16INK4a (Freedberg et al., 2008). The

function of ARF can also be abolished through promoter methylation that shuts down ARF

transcription (Esteller et al., 2000; Freedberg et al., 2008). In some cancers, transcriptional

repressors of ARF, including BMI-1, CBX-7, POKEMON, TBX-2 and -3, are overexpressed

compared to their normal cellular counterparts, thereby preventing ARF activation during

tumor development (Bernard et al., 2005; Maeda et al., 2005; Scott et al., 2007; Silva et al.,

2006; Sinclair et al., 2002; Vance et al., 2005; Vonlanthen et al., 2001; Yarosh et al., 2008).

The functional relevance of ARF inactivation observed in human cancers has been

validated using transgenic mouse models. Mice with Arf-specific deletion in exon-1β, which

spares p16Ink4a function, are highly prone to developing lymphoma, soft tissue sarcoma, lung

carcinoma and osteosarcoma (Kamijo et al., 1997; Sharpless et al., 2004). Mouse embryonic

fibroblasts (MEFs) derived from Arf-/- mice exhibit defects in oncogenic- and culture-induced

growth arrest in vitro, and are highly susceptible to oncogenic transformation compared to

wild-type MEFs.

15

The roles of Arf in specific tissues were addressed by deleting Arf in various mouse

cancer models. In a melanoma model induced by expression of the activated H-RAS

oncogene, Arf deletion significantly decreased the latency of melanoma development

(Sharpless et al., 2003). In a mouse model of B-cell lymphoma initiated by expression of the

c-Myc oncogene driven by the immunoglobulin heavy chain enhancer (Eµ), Arf inactivation

drastically reduced the mean survival time of Eµ-Myc animals. Moreover, tumors in Eµ-Myc

animals that were Arf+/+ and Arf+/- sustained spontaneous deletions of Arf, mutations in p53,

or overexpression of Mdm2, indicating the strong selection for tumor cells with loss of

function in the Arf-p53-tumor surveillance pathway (Eischen et al., 1999).

1.12. ARF protein and functions

1.12.1. ARF protein

ARF is a small protein comprised of 173 amino acids in human and 169 amino acids in

mouse. The protein is highly basic, made up of 20% arginine residues that are widely spread

throughout the molecule. ARF primarily localizes to the nucleolus, an intranuclear organelle

mainly implicated in ribosome biosynthesis, where it interacts and forms complexes with

nucleophosmin (NPM, also known as B23) (Bertwistle et al., 2004; Brady et al., 2004;

Itahana et al., 2003). Despite the lack of lysine residue, a common substrate for

polyubiquitylation, ARF protein-turnover is regulated by the ubiquitin-proteasome pathway

through ubiquitinylation at its N-terminus (Kuo et al., 2004). The stability of ARF is highly

dependent on its interaction with NPM, as the disruption of ARF-NPM binding by the

introduction of ARF mutations, or by the reduction in NPM level resulting from RNAi-

knockdown, significantly accelerates the degradation of ARF by the proteasome (Kuo et al.,

16

2004). The protein segment encoded by exon-1β is required and sufficient for its binding to

NPM (Bertwistle et al., 2004), and also for its p53-dependent function, as enforced

expression of a synthetic ARF exon-1β mini-gene, excluding exon-2, was sufficient to elicit a

p53-dependent response (Weber et al., 2000).

Interestingly, translation of ARF mRNA initiated from an internal start site produces an

N-terminus truncated isoform. When overexpressed in vitro, this short form of ARF, lacking

all the amino acid residues required for NPM binding and p53-dependent activities, triggers

autophagy, a process usually initiated in response to nutrient deprivation in order to digest

cellular components for energy derivation (Reef et al., 2006).

1.12.2. p53-dependent functions

The best characterized function of ARF is to activate p53 in response to various

oncogenic signals (Gil and Peters, 2006). Once activated, ARF localizes to the nucleolus,

where it binds and sequesters MDM2, thereby enabling p53 to accumulate in the

nucleoplasm (Tao and Levine, 1999). An important function of p53 is to induce expression of

target genes by associating into homotetrameric transcription factor (Levine et al., 2006). The

functions of p53-target genes fall into several categories. A set of genes are involved in cell

cycle arrest, including p21, 14-3-3σ and GADD45 (Levine et al., 2006). Another set of genes

important for tumor suppression are those involved in apoptosis, which can be subdivided

into the intrinsic and the extrinsic apoptotic pathways. In the intrinsic pathway, p53 activates

the expression of pro-apoptotic genes BAX, NOXA and PUMA (Haupt et al., 2003). In the

extrinsic pathway, p53 up-regulates the cell surface transmembrane receptors Fas (Muller et

al., 1998) and DR5/KILLER (Wu et al., 1997), which transmit extrinsic death signals to

activate intracellular apoptotic machineries (Nicholson and Thornberry, 2003). In addition,

17

p53 can exit the nucleus to act upon the mitochondria and its associated proteins, BCL-2 and

BCL-XL, in order to promote apoptosis (Chipuk et al., 2004; Moll et al., 2005).

1.12.3. p53-independent functions

1.12.3.1. Evidence from compound mutant studies

Studies using transgenic mice establish that Arf has additional functions outside the p53-

pathway. In the pituitary tumor model driven by Rb mutation, either Arf or p53 cooperated

with Rb in tumor suppression; however, the types of lesions in Arf-null and p53-null mice did

not completely overlap, providing the first hint that Arf and p53 might not function in a strict

linear manner (Tsai et al., 2002). To address the dependency of Arf function on p53, mice

nullizygous for Arf, mdm2 and p53 (TKO) were generated, and these animals developed

multiple tumors at a frequency greater than those lacking both p53 and Mdm2 or p53 alone.

Moreover, ectopic expression of Arf in MEFs lacking p53 and Mdm2 was able to induce cell

cycle arrest, further supporting p53-independent functions of Arf (Weber et al., 2000). In

addition to its role in tumor suppression, Arf is also implicated in normal eye development,

independent of p53. Deletion of Arf leads to defects in developmental regression of the

hyaloid vascular system in the eye, and the condition persists even when p53 is co-deleted

with Arf (McKeller et al., 2002).

In addition to MDM2, ARF physically interacts with more than 25 other proteins, and

some of these interactions have been suggested to convey p53-independent functions (Sherr,

2006). Nonetheless, these studies were performed in vitro using supra-physiological levels of

ARF, and these findings, briefly discussed in the following sections, remain to be validated

in vivo under physiological conditions.

18

1.12.3.2. Regulation of transcription factors

Evidence suggests that ARF can directly regulate the activity of the proto-oncogene

E2F1, independent of p53. In p53-null cells, ARF binds and inhibits E2F1, and DP1, a

dimerization partner of E2F1 that is essential for transactivation of E2F1-target genes (Datta

et al., 2002; Datta et al., 2005; Eymin et al., 2001; Martelli et al., 2001). In addition, ARF can

also antagonize the proto-oncogene c-MYC in a number of ways. ARF binds and inhibits

HECTH9, an E3 ubiquitin ligase known to catalyze c-MYC polyubiquitinylation, a process

that enhances c-MYC transactivating activity (Chen et al., 2005). ARF has also been shown

to bind directly to the MYC BOX II domain of c-MYC (Amente et al., 2006), thereby

preventing c-MYC association with TIP60, a histone acetyl transferase essential for

transactivation mediated by c-MYC (Frank et al., 2003). ARF was also found to antagonize

other transcription factors implicated in cancers, including the forkhead box family member

FOXM1b (Costa et al., 2005; Kalinichenko et al., 2004), B-cell lymphoma-6 (BCL6) (Suzuki

et al., 2005) and MYCN (Amente et al., 2007).

1.12.3.3. Ribosome biogenesis

When activated, ARF protein primarily localizes to the nucleolus, an intranuclear

organelle mainly implicated in ribosome biosynthesis. In vitro, enforced expression of ARF

inhibits rRNA transcription (Ayrault et al., 2004) and processing (Itahana et al., 2003;

Sugimoto et al., 2003) in p53-null cells, and disrupts ribosome export from the nucleus to the

cytoplasm (Brady et al., 2004; Yu et al., 2006).

19

1.12.3.4. Link to ATM and ATR

Several studies suggest that ARF can influence the activity of ATR- and ATM- kinase

through mechanisms independent of the MDM2-p53 axis. Overexpression of ARF in cells

lacking functional p53 activates the ATR- and ATM-signaling cascade (Eymin et al., 2006).

In addition, expression of ARF triggers ATR and leads to phosphorylation of NFκB. This

promotes the association of NFκB with histone deacetylase 1, thereby inhibiting the

transactivating function of NFκB (Rocha et al., 2003; Rocha et al., 2005).

1.12.3.5. Sumoylation

When overexpressed, ARF promotes sumoylation of several of its binding partners

(Chen and Chen, 2003; Rizos et al., 2005; Tago et al., 2005; Woods et al., 2004; Xirodimas

et al., 2002). Although the functional relevance of ARF-induced sumoylation has not been

fully elucidated, sumoylation of the Werner Syndrome Helicase (WRN) by ARF is

associated with the relocalization of WRN from the nucleolus to the nucleoplasm (Woods et

al., 2004).

1.12.3.6. smARF and autophagy

Initiation of ARF translation from an internal start site produces a short truncated form

that lacks the N-terminus, termed short mitochondrial ARF (smARF). Overexpression of

smARF in p53-null cells leads to its localization in the mitochondria, and the reduction of

mitochondrial membrane potential, thereby triggering autophagy and cell death (Reef et al.,

2006).

20

1.13. Regulation of ARF expression

1.13.1. Transcriptional repressors

ARF expression is tightly regulated and its deregulation may lead to detrimental

physiological defects. For instance, Arf is normally repressed by Bmi-1 in the stem cell

compartment within the nervous system. Deletion of Bmi-1 leads to aberrant activation of Arf,

causing defects in stem cell renewal, forebrain proliferation and gut neurogenesis. These

phenotypes are in part rescued by deletion of Arf in Bmi-1-/- mice (Molofsky et al., 2005).

BMI-1 is a member of the Polycomb group (PcG) of transcriptional repressors. PcG

proteins silence transcription by participating in multi-component complexes, PRC1 and

PRC2, which generate and recognize histone modifications in order to form Polycomb bodies

at specific chromosomal locations (Gil and Peters, 2006; Lund and van Lohuizen, 2004; Otte

and Kwaks, 2003). Given their mode of action at the chromatin level, gene silencing by PcG

proteins often has a wide spread effect, affecting both ARF and p16INK4a. In addition to BMI-

1, other PcG proteins, CBX2, CBX7, MEL18, PHC2 and RING1b, are also implicated in

INK4a/ARF repression (Core et al., 2004; Gil et al., 2004; Isono et al., 2005; Voncken et al.,

2003). Besides repression by PcG proteins, INK4a/ARF and neighboring gene p15INK4b are

regulated coordinately through a DNA replication origin upstream of p15INK4b. Loading of

CDC6 to this origin recruits histone deacetylases and increases methylation of histone H3,

thereby promoting heterochromatin formation and gene silencing (Gonzalez et al., 2006).

In addition to PcG proteins and CDC6, ARF transcription is regulated by a number of

other molecules, demonstrated mainly by knockout studies in mice. MEFs derived from mice

deficient of Atm (Kamijo et al., 1999), JunD (Weitzman et al., 2000), Egr1 (Krones-Herzig et

al., 2003), Twist (Maestro et al., 1999) or E2f3 (Aslanian et al., 2004) exhibit elevated

21

expression of Arf compared to wild-type MEFs. In addition, Pokemon, encoded by the Zbtb7

gene, represses Arf promoter directly. MEFs lacking Zbtb7 are refractory to oncogene-

mediated cellular transformation, in part attributed to Arf hyperactivity in the absence of

Pokemon (Maeda et al., 2005). Other direct repressors of ARF include the T-box proteins

TBX2 and TBX3, which bind the T-box element within the ARF promoter. Ectopic

expression TBX2 or TBX3 in cultured cells enables the bypass of senescence in vitro, in part

by suppressing ARF activation (Brummelkamp et al., 2002; Jacobs et al., 2000; Lingbeek et

al., 2002).

1.13.2. Translational repressor

In addition to transcriptional regulation, ARF expression is also modulated at the

translational level by the microRNA miR-24. MicroRNAs are endogenous short RNAs

(approximately 23 nucleotides) that play important gene-regulatory roles in animals and

plants by pairing to the mRNAs of protein-coding genes in order to direct their post-

transcriptional repression (Bartel, 2009). Enforced expression of miR-24 in HeLa cells down-

regulates ARF protein production; conversely, knockdown of miR-24 using antisense

oligonucleotides leads to the accumulation of ARF protein (Lal et al., 2008).

1.13.3. Transcriptional activators

ARF expression can be activated by several upstream pathways with oncogenic

properties. Ectopic expression of c-Myc in MEFs activates Arf gene expression and apoptosis

(Zindy et al., 1998). Ras-signaling induces Arf expression through the activation of

transcription factor Dmp1 (Inoue et al., 1999; Palmero et al., 1998; Sreeramaneni et al.,

2005). Alternatively, Ras can activate the AP-1 family members, c-Jun and Fra-1, which

22

assemble into heterodimeric transcription factors that drive Arf gene expression (Ameyar-

Zazoua et al., 2005). β-catenin is also implicated in Arf activation; its ectopic expression in

MEFs leads to Arf accumulation, in part through E2f1, given that Arf induction is

significantly weakened in E2f1-null MEFs (Damalas et al., 2001). DAP-kinase, a pro-

apoptotic serine/threonine kinase, seems to link oncogenic signals to Arf activation. Ectopic

expression of DAP-kinase induces Arf-dependent apoptosis, while its inactivation

significantly weakens c-Myc or E2f1 induced apoptosis (Raveh et al., 2001).

23

Chapter 2: Role of ARF and the p53-pathway in retinoblastoma

development

2.1. Introduction

The p53-pathway is corrupted in many cancers, either by mutation in ARF or TP53, or

by overexpression of the p53-antagonists MDM2 or MDM4 due to genomic gain. In

retinoblastomas, mutation in TP53 has never been observed. Recent study suggests that the

p53-pathway is disrupted by genomic gain of MDM4 or MDM2 in a subset of

retinoblastomas (Laurie et al., 2006). However, the role of ARF, an upstream activator of the

p53-pathway, in retinoblastoma development remains unclear. ARF mRNA expression is

highly induced in retinoblastomas compared to normal retinas (Guo et al., 2008; Laurie et al.,

2006), but ARF protein is undetectable in neither tumors nor retinas (Guo et al., 2008).

2.2. Hypotheses

ARF mRNA transcription is activated in response to RB1 mutation; however, inhibition

of ARF translation, or instability of the ARF protein, prevents ARF-mediated activation of

the p53-tumor surveillance pathway during retinoblastoma development.

2.3. Thesis aims & rationales

2.3.1. Determine the cause of low ARF protein in retinoblastomas

Translation of ARF mRNA is known to be regulated by a microRNA, miR-24 (Lal et al.,

2008), and ARF protein-turnover is controlled by the ubiquitin-proteasome pathway (Kuo et

24

al., 2004). These mechanisms were assessed in order to determine the cause of low ARF

protein expression despite the abundance of ARF mRNA.

2.3.2. Examine the expression of Arf, p53 & p53-transcriptional targets during

early retinoblastoma development in a mouse retinoblastoma model

It is unclear whether ARF expression is consistently low during the entire course of

retinoblastoma development, or ARF expression is induced during early tumor development

but subsequently extinguished due to oncogenic insults. The expression of Arf, p53 and p53-

transcriptional targets were determined in a mouse model of retinoblastoma, in order to

evaluate the activity of this pathway during early retinoblastoma development.

2.3.3. Determine the effect of exogenous ARF expression in retinoblastoma cells

Expression studies show that MDM2 and MDM4 are expressed in retinoblastomas and

may even be overexpressed in a subset of the tumors, while the upstream activator of the

pathway, ARF, was undetectable at the protein level (Guo et al., 2008; Laurie et al., 2006). It

is unclear whether increased level of ARF protein could activate p53 despite high level of

MDM2 and MDM4. The effect of ARF overexpression on cell growth and gene expression of

retinoblastoma cells was determined.

2.4. Material and methods

2.4.1. Tissue culture

All cell lines were grown in a humidified 37 C incubator with 5% CO˚ 2 in their respective

tissue culture media (Tissue Culture Media Facility, PMH/OCI, ON, Canada) with the

addition of 100 U/ml penicillin and 0.1 mg/ml streptomycin (Wisent Bioproducts). Media

25

and supplements used for each cell line were as follows. Retinoblastoma cell lines: Iscove's

MDM with 15% FetalClone III (HyClone), 10 mg/L insulin (Sigma-Aldrich) and 0.0004%

(v/v) β-mercaptoethanol; HEK-293: alpha-MEM with 10% FBS (Wisent Bioproducts); HEK-

293T and HeLa: DMEM-H16 with 10% FBS; Saos-2 and U2OS: DMEM-H21 with 10%

FBS; OVCAR-3: RPMI-1640 with 20% FBS and 10 mg/L insulin; MEFs: DMEM-H16 with

15% FBS and 4 mM L-glutamine (Wisent Bioproducts); NIH-3T3: DMEM-H16 with 10%

calf serum (Wisent Bioproducts).

To study the role of proteasome-mediated protein degradation in retinoblastoma cell

lines, cells were incubated in culture media containing 50 µM of the proteasome inhibitor

MG132 (Sigma-Aldrich) for 10 hours and then harvested for Western Blot analysis.

2.4.2. Mouse model of retinoblastoma

Mice with retinal specific Rb-inactivation were used for studying early retinoblastoma

development. RbloxP/loxP, α-Cre; RbloxP/loxP and α-Cre; RbloxP/loxP; p107-/- mice were kind gifts

from Dr. Rod Bremner (Toronto Western Research Institute, ON, Canada). For simplicity,

animals with the genotype RbloxP/loxP and α-Cre; RbloxP/loxP were referred to as wild-type and

Rb-knockout, respectively. In Rb-knockout animals, Rb is flanked by loxP sites and is

excised by a Cre-recombinase driven by the α-enhancer of Pax-6 beginning from embryonic

day 10 in the peripheral retina. Inactivation of Rb in the p107-/- background leads to

retinoblastoma with 60% penetrance, while Rb-mutant retina in the p107+/+ background is

not tumor prone (Chen et al., 2004). Mice were maintained, handled and sacrificed using

protocols approved by the University Health Network Animal Resource Centre (ON,

Canada). For genotyping, genomic DNA was extracted from mouse tail. PCR was performed

in a RoboCycler Gradient 96 thermal cycler (Stratagene). Unless otherwise stated, PCR

26

began with 2 minutes at 94 C, followed by 30 cycles of amplification and ended with 10 ˚

minutes at 72 C, and the reaction ˚ mixture consisted of 200 µM dNTPs, 2.5 mM MgSO4, 0.5

µM each of forward and reverse primers, homemade Taq polymerase and reaction buffer, in

a final volume of 25 µl (Table 1).

Table 1. Primer sequences and PCR conditions for genotyping Gene Primer sequence PCR condition Product (bp)

Cre 5'-CCTGATGGACATGTTCAGGG-3' 5'-CTTCAGGTTCTGCGGGAAAC-3'

94 C 50 s, 55 C ˚ ˚30s, 72 C 50 s˚ 120

p107 5'-TCGTGAGCGGATAGAAAG-3' 5'-GTGTCCAGCAGAAGTTA-3'

5'-CCGCTTCCATTGCTCAGCGG-3'

94 C 50 s,˚ 54 C ˚50s, 72 C 1 min˚

386 (Knockout) 513 (Wild-type)

RbloxP 5'-GGCGTGTGCCATCAATG-3' 5'-CTCAAGAGCTCAGACTCATGG-3'

94 C 50 s, 54 C ˚ ˚50s, 72 C 1 min˚

235 (Wild-type) 283 (loxP)

2.4.3. Tissue harvesting and fixation

To generate tissue sections for immunofluorescence microscopy, heads of euthanized

neonatal mice were fixed overnight in 4% paraformaldehyde at 4 C and subsequently ˚

stored in 70% ethanol at 4 C. For mice over the age of five days, fixed heads were ˚

decalcified in 8% formic acid before transferring into 70% ethanol. Fixed mouse heads were

embedded in paraffin blocks and cut into 5 µm sections (Pathology, Hospital for Sick

Children, ON, Canada). To examine the histology of tissue sections, slides were stained with

hemotoxylin and eosin, and scanned using the Aperio ScanScope XT.

For extraction of retinal total RNA and protein, eyes were removed from euthanized

mice at different ages, and dissected to extract the retinas under a dissecting microscope.

Retinas from 2 to 3 mice of the same genotype were pooled and frozen in dry ice

immediately and subsequently stored at -70 C. Littermate controls ˚ were used for expression

analyses.

27

To embed MEFs in paraffin block, cells were first harvested from tissue culture flasks

using trypsin and then fixed in 4% paraformaldehyde for 15 minutes at room temperature,

followed by three washes in PBS. Fixed cells were resuspended in 1% agarose and allowed

to solidify. The agarose block was subsequently paraffin embedded and sectioned as

previously described.

To fix the retinoblastoma cell line WERI for immunofluorescence staining, cells grown

on a coverslip were incubated in 4% paraformaldehyde for 15 minutes at room temperature,

followed by three washes in PBS.

2.4.4. RT-PCR

Total RNA was extracted from cell lines and mouse retinas using TRIzol (Invitrogen)

according to the manufacturer’s instructions. Mouse retinas resuspended in TRIzol were

dissociated by passage through a gauge 20.5 needle prior to the addition of chloroform. The

concentration and quality of total RNA were determined using a Nanodrop-1000

spectrophotometer (Thermo Scientific). For cDNA synthesis, 1 µg of total RNA was reverse

transcribed using random primers (Invitrogen) and Superscript II Reverse Transcriptase

(Invitrogen). For gene expression analyses, PCR was performed using a RoboCycler

Gradient 96 thermal cycler (Stratagene). Thermal cycling conditions for the genes examined

were listed in Table 2. Unless otherwise stated, PCR began with 2 minutes at 94 C, followed ˚

by 30 cycles of amplification and ended with 10 minutes at 72 C, and the reaction mixture ˚

consisted of 200 µM dNTPs, 2.5 mM MgSO4, 0.5 µM each of forward and reverse primers,

0.5 U KOD hot start DNA polymerase (Novagen), 1x PCR buffer (Novagen) and 1 µl of

product from cDNA synthesis, in a final volume of 25 µl.

28

} Table 2. Primer sequences and PCR conditions for human and mouse gene expression analyses Human genes

Gene Primer sequence PCR condition Product (bp)

ARF 5'-TGGGTCCCAGTCTGCAGTTA-3' 5'-CCTGTAGGACCTTCGGTGAC-3'

94 C 50 s, 56 C ˚ ˚30 s, 72 C 1 mi˚ n,

0.2 M betaine 683

GADD45 5'-GAGAGCAGAAGACCGAAAGG-3' 5'-CCCTTGGCATCAGTTTCTGT-3'

94 C 50 s, 58 C ˚ ˚30 s, 72 C 1 min˚ 591

NOXA 5'-CACCGTGTGTAGTTGGCATC-3' 5'-TTCCATCTTCCGTTTCCAAG-3'

94 C 50 s, 58 C ˚ ˚30 s, 72 C 1 min˚ 474

TBP-1 5'-ACAACAGCCTGCCACCTTAC-3' 5'-GCTGGAAAACCCAACTTCTG-3'

94 C 50 s, 60 C ˚ ˚30s, 72 C 1 min˚ 743

Mouse genes

14-3-3σ 5'-AAGGCAGAGCCAGAGCAAG-3' 5'-GTAGGCGGCATCTCCTTCTT-3'

94 C 50 s, 63 C ˚ ˚30 s, 72 C 1 min˚ 581

Arf 5'-GTCGCAGGTTCTTGGTCACT-3' 5'-ACGTTCCCAGCGGTACAC-3'

94 C 50 s, 60 C ˚ ˚30 s, 72 C 1 min˚ 484

Bax 5'-GCACGTCCACGATCAGTCA-3' 5'-ATGGTCACTGTCTGCCATGT-3'

94 C 50 s, 59 C ˚ ˚30 s, 72 C 1 min, ˚

0.2 M betaine 599

Gadd45 5'-AGGGACTCGCACTTGCAATA-3' 5'-TGGGGAGTGACTGCTTGAGT-3'

94 C 50 s, 57 C ˚ ˚30 s, 72 C 1 min˚ 598

Mdm2 5'-TGCAAGCACCTCACAGATTC-3' 5'-CATCTCCCTCAGCTCACACA-3'

94 C 50 s, 57 C ˚ ˚30 s, 72 C 1 min˚ 555

Noxa 5'-GGTGCTGCCTACTGAAGCTC-3' 5'-CGAGCGTTTCTCTCATCACA-3'

94 C 50 s, 59 C ˚ ˚30s, 72 C 1 min˚ 562

p21 5'-TCCACAGCGATATCCAGACA-3' 5'-CCTGACCCACAGCAGAAGAG-3'

94 C 50 s, 57 C ˚ ˚30 s, 72 C 1 min˚ 558

Puma 5'-GCCTCAATAGCAACCCACTC-3' 5'-GGTGTCGATGCTGCTCTTC-3'

94 C 50 s, 59 C ˚ ˚30s, 72 C 1 min, ˚

0.2 M betaine 599

Tbp-1 5'-GGGAGAATCATGGACCAGAA-3' 5'-ATGATGACTGCAGCAAATCG-3'

94 C 50 s, 60 C ˚ ˚30 s, 72 C 1 min, ˚

28 cycles 544

2.4.5. Real-time PCR

RNA extraction and cDNA synthesis were performed as described in the previous

section. Real-time PCR was performed in a 7900HT Fast Real-Time PCR system using the

Universal PCR Master Mix in a 384-wells plate (Applied Biosystem). Taqman gene

29

expression assays (Applied Biosystem) were employed to measure the mRNA expression of

ARF (Hs99999189_m1), GAPDH (Hs99999905_m1), HPRT-1 (Hs99999909_m1) and TBP-

1 (Hs99999910_m1) in triplicate. Mean relative gene expression and standard deviation were

determined using the ΔΔCt method built-in to the SDS 2.2 software (Applied Biosystem).

GAPDH was used as the endogenous control gene and HeLa was used as the calibrator

sample. For expression analysis of microRNAs, cDNA was synthesized using the TaqMan

MicroRNA Reverse Transcription Kit according to the manufacturer’s instructions (Applied

Biosystem). Taqman gene expression assays (Applied Biosystem) were used for the detection

of miR-24 (hsa-miR-24), RNU6B (Part # 4373381) and U18 (Part # 4380904) in triplicate.

Mean relative gene expression and standard deviation were determined by the ΔΔCt method

using RNU6B or U18 as the endogenous control and HeLa as the calibrator sample.

2.4.6. Western Blotting

Cell lines and mouse retinas were resuspended in lysis buffer consisting of phosphate

buffered saline, 1% Nonidet P40, 5% sodium deoxycholate, 0.1% SDS, 1.0 µg/ml leupeptin,

0.1 mM PMSF, 1.0 µg/ml aproptinin and 100 µM sodium orthovanadate. Cells and tissues

were dissociated mechanically and lysed by three cycles of freezing and thawing, followed

by a 30 minutes incubation on ice, and centrifugation to remove cellular debris. Protein

concentration was determined using the BioRad Protein Assay (BioRad) in a Beckman

Coulter DU640B spectrophotometer. For SDS-PAGE, 50 µg of protein lysate were resolved

in 4-20% Tris-Glycine gradient gels (Lonza) and then transferred onto PVDF membranes

(BioRad). Membrane was blocked in 5% blotto (BioRad) in TBS overnight at 4 C and ˚

subsequently probed with primary antibody in TBS with 1% BSA and 0.05% Tween-20 at

room temperature for one hour, followed by three washes in TBS with 0.1% BSA and 0.05%

30

Tween-20. The primary antibodies and the respective dilutions used are as follows: ARF

(1:100, Abcam, ab3642), p21 (1:200, BD-Pharmingen, Cat. # 556430), p53 (1:200, Santa

Cruz, sc6243-G), BAX (1:200, Santa Cruz, sc7480), MDM2 (1:200, Calbiochem, Cat. #

OP115) and β-tubulin (1:1000, Sigma-Aldrich, T0198). After primary antibody incubation

and washes, blot was then probed with the appropriate secondary antibodies at the specified

dilution in TBS with 1% BSA and 0.05% Tween-20 at room temperature for one hour,

followed by three washes. Secondary antibodies used are as follows: anti-rabbit-HRP

(1:10000, Santa Cruz, sc2004), anti-goat-HRP (1:10000, Santa Cruz, sc2020) and anti-

mouse-AP (1:10000, Santa Cruz, sc2008). HyGLO Chemilluminescence Detection Reagent

(Denville) and HyBlot autoradiography film (Denville) were used to detect protein of interest.

Alternatively, proteins probed with AP-conjugated secondary antibodies were detected using

NBT/BCIP (Denville).

2.4.7. Immunofluorescence staining

For paraffin embedded sections, paraffin removal was performed as follows: two times

of 10 minutes each in xylene, two times of 5 minutes each in 100% ethanol, once for 2

minutes in 95%, 70%, and 50% ethanol, followed by a 5 minutes incubation in TBS. Heat-

mediated antigen retrieval was performed by heating microscope slides immersed in PBS-

citrate using a microwave pressure cooker. Slides were then incubated in 5% Triton-X for 10

minutes at room temperature. Blocking was carried out for 30 minutes at room temperature

in TBS with 10% DAKO Protein Block (DAKO-Cytomation), 1% BSA and 0.05% Tween-

20. Slides were then incubated with primary antibody against ARF (1:200, Abcam, ab3642)

in TBS with 1% BSA, 0.05% Tween-20 and 10% Antibody Diluent (DAKO-Cytomation),

followed by three washes in TBS with 0.1% BSA and 0.05% Tween-20. Subsequently, slides

31

were incubated in goat biotinylated anti-rabbit IgG secondary antibodies (1:200, Vector

Laboratories, BA-1000) at room temperature for 1 hour, followed by three washes.

Streptavidin-Alexa-594 was used for the detection of ARF. DAPI was used to visualize

nuclei of cells. Slides were mounted using the DAKO-Cytomation Fluorescent Mounting

Medium.

Staining of fixed cells on a cover-slip was performed as described above, with the

omission of paraffin removal and heat-mediated antigen retrieval.

2.4.8. Exogenous gene expression using adenoviruses

E1 and E3 early regions deleted adenovirus encoding gfp (adgfp) or human ARF (adARF)

(Huang et al., 2003) under the control of a CMV promoter are kind gifts from Dr. Erik

Knudsen (Kimmel Cancer Center, PA, USA) and Dr. Ruth Gjerset (Sidney Kimmel Cancer

Center, CA, USA), respectively. Adenoviruses were amplified in HEK-293 cells. Infected

HEK-293 cells were resuspended in PBS with 10% glycerol, and viral particles were released

by three cycles of freezing and thawing. Cellular debris was removed by centrifugation. Viral

titers were determined using the QuickTiter Adenovirus Titer Immunoassay Kit (Cell Biolabs)

according to the manufacturer’s instructions. Adenoviruses were stored at -72 C˚ in aliquots.

To assess the effect of ARF overexpression on gene expression and cell growth, WERI cells

were seeded at 200,000 cells per well in 6-well plates pretreated with poly-D-lysine and

incubated overnight under normal culture condition. For immunostaining, cells were seeded

on poly-D-lysine treated coverslips placed within 6-well plates. After the overnight

incubation, culture media was removed, and cells were rinsed once with PBS. Cells were

then incubated in 500 µl of serum-free Iscove’s media containing adgfp or adARF at 20

multiplicity of infection (MOI) for 1 hour in a 37 C incubator with 5% CO˚ 2. Complete

32

growth media (2 ml) was then added to each well and incubated for four days, after which

cells were harvested for gene expression analyses. Cell viability was determined by trypan

blue exclusion. Means and standard deviations were calculated based on three replicates per

treatment. Statistical significance was determined using the two-tailed t-test.

2.5. Results

2.5.1. Cause of low level of ARF protein in retinoblastomas

2.5.1.1. Analysis of ARF expression in retinoblastoma cell lines

Guo et al previously showed that ARF protein in primary retinoblastoma tumors is

disproportionally low compared to the amount of mRNA detected (Guo et al., 2008). To

confirm this result, mRNA and protein expression in four control cell lines (HeLa, HEK-

293T, OVCAR-3, Saos-2) with high ARF protein were compared to that in retinoblastoma

cell lines. All retinoblastoma cell lines examined expressed very low levels of ARF protein

relative to the control cell lines (Figure 3), yet three of the cell lines (RB381, RB383, RB247)

expressed comparable amount of mRNA relative to that in HeLa (Figure 4B). GAPDH was

used as the endogenous control for real-time PCR analysis; since compared to other

housekeeping genes, it exhibited the least variation between the different cell lines examined

(Figure 4A). Real-time PCR amplifies only 72 base pairs of the full length ARF mRNA and

therefore could detect fragmented mRNA. To ensure that ARF mRNA is intact in

retinoblastoma cell lines, RT-PCR analysis spanning all three ARF exons was performed,

confirming the presence of full length ARF mRNA (Figure 4C). Therefore, in three of six

retinoblastoma cell lines examined (RB381, RB383, RB247), the level of ARF protein was

very low despite high mRNA expression, suggesting attenuation of ARF protein synthesis

33

β-tubulin

ARFHeL

aRB24

7

RB383

RB1021

HEK-293T

OVCAR-3

Saos-2

RB381

p21

pRB status Inactivatedp53 status Inactivated Normal

Controls RB cell lines

1.0 4.3 0.0 21 28 106 13 176Normalized p21 intensity:

Figure 3. Western Blot analysis of ARF and p21 in retinoblastoma and control cell lines Cell lines expressing high level of ARF protein were used as controls (HeLa, HEK-293T, OVCAR-3, Saos-2). Retinoblastoma cell lines (RB247, RB381, RB383, RB1021) expressed very low level of ARF protein. In control cell lines, p53 and pRB are inactivated by mutations or expression of viral-oncoproteins, while p53 in the retinoblastoma cell lines appears normal (low mRNA levels, no mutations by PCR-SSCP, as shown in these and 25 other retinoblastomas, Duckett-Brown & Gallie, unpublished data). p21, a transcriptional target of p53, was not expressed, or expressed at low level in control cell lines (first four lanes), consistent with the p53 status in these cells. In contrast, p21 was highly expressed in retinoblastoma cell lines, further suggesting that p53 is intact in these cells. β-tubulin was used as a loading control. Intensity of the p21 bands were quantified using ImageJ, and normalized to that of β-tubulin.

34

5

10

15

20

25

30

HeLa

HEK-TOVCAR-3SAOS-2

WERIY79

RB247RB381RB383

RB1021

Ct

GAPDHHPRT-1TBP-1

HeLa

HEK-T

OVCAR-3SAOS-2

WERIY79

RB247

RB381

RB383

RB1021

B

0

1

2

3

4

5

6

SAOS-2HEK-T

OVCAR-3HeL

a

RB381

RB383

RB247

RB1021

WERIY79

AR

F m

RN

A e

xpre

ssio

n no

rmal

ized

to

GA

PDH

rel

ativ

e to

HeL

a

Saos-2 293

T

OVCAR-3HeL

a

RB381

RB383

RB247

RB1021

WERI

Y79

A

Controls RB cell lines

C

ARF

TBP-1

HeLa

HR WERIRB10

21

700bp

C33A

RB383

Y79 RB247

Controls RB cell lines

Figure 4. ARF mRNA expression in retinoblastoma and control cell lines Real-time PCR analysis was performed to determine if low ARF protein in retinoblastoma (RB) cell lines is due to low ARF mRNA. A. Expression of three housekeeping genes, GAPDH, HPRT-1 and TBP-1, was evaluated in order to identify the most suitable endogenous control for data analysis. Ct (y-axis) represents the cycle when a predetermined fluorescence signal was reached. GAPDH had the most consistent Ct-values in different cell lines, and was therefore used as the endogenous control for data analysis. B. Real-time PCR measurement of ARF mRNA in control and retinoblastoma cell lines. Normalized ARF

35

expression was presented relative to that of HeLa. The level of ARF mRNA in RB381, RB383 and RB247 was comparable to that found in HeLa, although ARF protein was very low (Figure 3), suggesting that ARF protein synthesis is attenuated, or ARF protein is unstable in these RB cell lines. Error bars represent the standard deviations of a triplicate experiment. C. To determine if full length ARF mRNA was expressed, RT-PCR was performed in control and retinoblastoma cell lines using PCR primers spanning the entire coding region of ARF. TBP-1 was used as an endogenous control. C33A, a cervical cancer cell line, and HeLa were used as positive controls, while normal human retina (HR) was used as a negative control.

36

or instability of the ARF protein. For the remaining cell lines (RB1021, WERI, Y79), low

ARF protein was at least in part due to low ARF mRNA.

In the four control cell lines (HeLa, HEK-293T, OVCAR-3, Saos-2), p53 is inactivated

either by mutation or by the expression of viral oncoproteins. This p53-status correlated with

the absence of p21, which is a p53-transcriptional target (Figure 3). In contrast, all examined

retinoblastoma cell lines expressed p21 protein, indicating that p53 may be intact in these

cells, as suggested by the lack of TP53 mutation found in these cell lines and 25 other

retinoblastomas (Duckett-Brown & Gallie, unpublished data).

2.5.1.2. Role of the proteasome pathway in ARF protein degradation in retinoblastoma

cell lines

ARF protein turnover is normally regulated by the ubiquitin-proteasome pathway (Kuo

et al., 2004). To determine if ARF protein is aberrantly degraded by this pathway, which

leads to low ARF expression, retinoblastoma cells were treated with a proteasome inhibitor,

MG132. After 10 hours of MG132 treatment, all retinoblastoma cell lines responded with the

accumulation of p21, a positive control whose turnover is regulated by the proteasome

(Figure 5). However, no accumulation of ARF protein was observed in response to MG132

treatment (Figure 5). This indicated that low ARF protein in retinoblastoma cells was not the

result of aberrant turnover by the ubiquitin-proteasome pathway.

37

ARF

p21

β-tubulin

MG132: - + - + - +RB247

HeLaRB383 RB1021

Figure 5. Western Blot analysis for ARF and p21 in retinoblastoma cell lines after MG132 treatment To determine if low ARF protein is the result of aberrant degradation by the ubiquitin-proteasome pathway, retinoblastoma cell lines were treated with the proteasome inhibitor MG132 for 10 hours. After treatment, retinoblastoma cell lines responded with the accumulation of the positive control, p21, which is known to be regulated by the ubiquitin-proteasome pathway. In contrast, ARF protein did not accumulate in retinoblastoma cells after MG132 treatment, indicating that low ARF protein in retinoblastomas is not due to aberrant degradation by the proteasome pathway. Untreated HeLa was used as a positive control for ARF expression.

38

2.5.1.3. Analysis of miR-24 expression in retinoblastoma cell lines

Translation of ARF mRNA is negatively regulated by the microRNA, miR-24 (Lal et al.,

2008). Expression of miR-24 was measured to determine if this microRNA is expressed at a

high level in retinoblastoma cell lines, thereby preventing efficient ARF protein synthesis.

Real-time PCR analysis showed that miR-24 expression was significantly lower in

retinoblastoma cell lines than that in HeLa or OVCAR-3, when using U18 or RNU6B as the

endogenous control for data analysis (Figure 6).

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

HeLa

OVCAR-3

HEK-293T

RB247

RB381

RB383

Nor

mal

ized

miR

-24

expr

essi

on re

lativ

e to

HeL

a

Normalized to U18Normalized to RNU6B

Controls: high ARF protein

RB cell lines: low ARF protein

Figure 6. Expression analysis of miR-24 in control and retinoblastoma cell lines The expression of miR-24 was significantly lower in retinoblastoma cell lines (RB247, RB381, RB383) relative to HeLa or OVCAR-3 when using either U18 or RNU6B as endogenous control. Error bars represent the standard deviations of a triplicate experiment.

39

2.5.2. Expression of Arf, p53 & p53-downstream targets during early

retinoblastoma development in a mouse retinoblastoma model

2.5.2.1. Expression of Arf during early mouse retinoblastoma development

To study the expression of Arf, p53 and p53-target genes in response to Rb-inactivation,

transgenic mice with retinal specific Rb-inactivation (termed Rb-knockout for simplicity)

were used. In this model, the Rb gene is excised in the peripheral retina by Cre-recombinase

under the control of the Pax-6 α-enhancer beginning from embryonic day 10. Rb-deficient

cells exhibit apoptosis and ectopic division during retinal development. These cells

eventually cease to divide and differentiate into mature retinal cells, and they are not prone to

retinoblastoma development (Figure 7B) (Chen et al., 2004; MacPherson et al., 2004). In

contrast, Rb-knockout animals in the p107-/- background develop retinoblastoma with

incomplete penetrance (Chen et al., 2004; MacPherson et al., 2004). During postnatal

development, Rb-p107 compound mutant retina exhibits retinal dysplasia (Figure 7C), which

progresses into tumor in approximately 60% of mutant animals (Chen et al., 2004;

MacPherson et al., 2007; MacPherson et al., 2004). In the remaining animals, the mutant

retinas eventually cease to proliferate and terminally differentiate by postnatal day 30 (P30)

without any tumor development (Chen et al., 2004). For this study, mutant retinas from

p107+/+ and p107-/- animals were collected at postnatal day 0 (P0), P5, P10 and P15 for RNA

and protein extractions.

RT-PCR analysis showed that Arf mRNA was increased in Rb-knockout retinas

compared to retinas with wild-type (Wt) Rb in both the p107+/+ (Figure 8A) and p107-/-

(Figure 8B) backgrounds, indicating that Arf transcription was activated in response to Rb-

40

Lens

RON KO region

Lens

ON R

ONL

KO region

Lens

ON R

A

B C

Wild-type

GCL INL ONLGCL INL

a b

c

Rb-knockout; p107-/-

Dysplastic retina

Rb-knockout

Figure 7. Hematoxylin and eosin staining of Rb-knockout retinas in p107+/+ or p107-/- background at P9 In Rb-knockout retina, the Rb gene is excised by Cre-recombinase controlled by the α-Pax6 enhancer in the peripheral retina beginning from embryonic day 10 of development. A & a. At P9, the formation of the three distinct retinal layers has already completed throughout the wild-type retina (R). B & b. This process is delayed in the peripheries of Rb-knockout retina where Rb is inactivated (KO region); the inner and outer nuclear layers have not yet completely separated. Nevertheless, in the p107+/+ background, cells in the developing Rb-knockout retina are not tumor prone, and eventually exit the cell cycle (Chen et al, 2004). C & c. In the p107-/- background, Rb-inactivation in the peripheral retina (KO region) leads to retinal dysplasia. In approximately 60% of Rb-knockout; p107-/- animals, the dysplastic retina progresses into tumor, while retinas of the remaining 40% undergo substantial apoptosis and terminally differentiate by P30 (Chen et al, 2004). R= Retina. L = Lens. ON = Optic nerve. GCL = Ganglion cell layer. INL = Inner nuclear layer. ONL= Outer nuclear layer.

41

MEFs

NTC

Arf

Tbp-1

+ - NIH-3T

3

P5P0 P10 P15+ - + - + -

NTCP5 P10 P15+ - + - + -

Arf

p107+/+

p107-/-

Tbp-1

Rb:

Rb:

A

B

Figure 8. RT-PCR analysis for Arf in Rb-knockout retinas from P0 to P15 A. Expression of Arf mRNA in Wt- (+) and Rb-knockout (-) retinas in p107+/+ background. Tbp-1 was used as an endogenous control. Late passage MEFs were used as the positive control, while NIH-3T3, in which the Ink4a/Arf locus is deleted, was used as the negative control. NTC = no cDNA control. B. Arf expression in Wt- (+) and Rb-knockout (-) retinas in p107-/- background. The low signal at P5 in the Rb+ retinas might be related to the resorbing hyaloid system.

42

inactivation. In Wt-retinas, Arf mRNA was detected at P5 but declined from P10 and beyond.

This was likely due to contamination during dissection by the resorbing hyaloid vascular

system, where Arf is normally expressed (McKeller et al., 2002).

To assess Arf protein expression, immunofluorescence staining was performed on tissue

sections of Rb-knockout retinas in p107+/+ and p107-/- backgrounds at P9. Rb is inactivated in

the peripheral retina, while the central retinal region possesses intact Rb, and therefore was

used as an internal control. In both the central and peripheral retinas of p107+/+ animal,

positive signal in the Arf channel was only detected in the outer plexiform layer (between the

inner and outer nuclear layer of the retina, Figure 9A and 9B). This signal was likely due to

non-specific staining since it did not overlap with nuclei, where Arf normally localizes. In the

Rb-deficient peripheral retina of p107-/- animal, in which the outer plexiform layer was

disrupted due to retinal dysplasia, it was clear that specific Arf staining was completely

absent (Figure 9C and 9D). Late passage wild-type MEFs, which express high level of Arf

(Krimpenfort et al., 2001; Sharpless et al., 2001), were used as the positive control.

To further assess Arf protein expression, Western Blotting was performed using Wt- and

Rb-knockout retinas in both p107 backgrounds. Arf was readily detected in MEFs (Figure

10A), but not in Wt- nor Rb-knockout retinas in either background (Figure 10A). To confirm

the presence or absence of Arf protein, increased amount of protein lysate and prolonged

chemilluminescence exposure were used. Under this condition, an elevation of Arf protein

was detected in Rb-knockout retina relative to Wt-retina at P5 and P10 in the p107+/+

background (Figure 10B). Taken together, Arf transcription was activated upon Rb-

inactivation in the developing retina, but the protein was expressed at very low level.

43

A

DAPI ARF Merge

CR

Lens

PR

Wt C

RR

b-kn

ocko

ut P

R

B

GCL INL ONL

GCL

ONL

INL

GCL INL ONL

GCL

ONL

INL

Rb-knockout; p107+/+

p107

+/+

Figure 9. Immunostaining for Arf protein in Rb-knockout retinas in p107+/+ or p107-/- background at P9 Central retina (CR) possesses wild-type Rb, and was used as an internal control. In both p107 +/+ (A & B) and p107-/- (C & D) backgrounds, positive Arf signal was present only in the outer plexiform layer, which lies between the INL and ONL in the Wt central retina (B & D, top panel). This signal was likely due to non-specific staining since it did not overlap with the nuclei, where Arf normally localizes, and Arf mRNA expression in Wt retinas at P9 was very low. A & B. In the p107 +/+ background, no nuclear staining was detected in the central Wt or peripheral Rb-knockout retina. As explained previously, staining between the INL and ONL in Wt retina was likely non-specific. C & D. In the p107-/- background, nuclear Arf staining was absent in both central and peripheral retinas. Late passage MEFs were used as

44

the positive control (bottom panel). DAPI was used to stain the nuclei of cells. GCL = Ganglion cell layer. INL = Inner nuclear layer. ONL = Outer nuclear layer.

`

CR

PR

Lens

C

DGCL INL ONL GCL INL ONL

DAPI ARF Merge

Wt C

RR

b-kn

ocko

ut P

RM

EFs

Rb-knockout; p107-/-

p107

-/-

45

MEF

s P5 P10 P15

+ - + - + -Arf

β-tubulin

+ - + - + -MEF

s3T

3

Arf

β-tubulin

MW

p107+/+

P5 P10 P15

p107+/+

MEF

s P5 P10 P15

+ - + - + -Arf

β-tubulin

p107-/-

Rb:

Rb:

Rb:

A

B

Figure 10. Western Blot analysis of Arf in Rb-knockout retina in p107+/+ or p107-/- background A. Protein expression of Arf was undetectable in Wt (+) and Rb-knockout (-) retinas in p107 +/+ (top) or p107-/- (bottom) background. MEFs were used as the positive control. β-tubulin was used as a loading control. MW = molecular weight markers. B. Increased amount of protein extract and prolonged chemilluminesence exposure were used in order to detect Arf protein (100 µg of retinal protein extract, instead of 50 µg). Under this condition, Arf protein showed an increase in Rb-knockout compared to Wt retinas at P5 and P10. 3T3 = NIH-3T3.

46

2.5.2.2. Expression of p53 and p53-transcriptional targets during early mouse

retinoblastoma development

The protein level of Arf was found to be low in Rb-knockout retinas, and therefore it was

questionable whether the attained level of Arf was sufficient for p53 activation. To assess the

activity of the p53 pathway in Rb-knockout retinas, Western Blotting was performed for p53

and its downstream targets p21 and Bax. Inactivation of Rb led to an increase in p53 and p21

protein in the p107+/+ or p107-/- background, but no increase in Bax protein was observed

(Figure 11). To further examine the extent of p53-target gene activation, RT-PCR analysis

was performed for genes involved in cell cycle arrest (p21, Gadd45, 14-3-3σ), apoptosis (Bax,

Puma, Noxa), and autoregulation (Mdm2). For Gadd45, 14-3-3σ and Mdm2, no induction

was observed upon Rb-inactivation in the p107-/- background (Figure12). Increase in mRNA

expression was observed for Bax, Puma, Noxa and p21 in response to Rb-inactivation.

Similarly, in the p107+/+ background, p21 was up-regulated upon Rb-inactivation; however,

no induction of Gadd45, 14-3-3σ and Mdm2 was observed (Figure 13).

2.5.3. Effect of exogenous ARF expression in a retinoblastoma cell line

To test whether enforced expression of ARF can activate p53 and p53-transcriptional

targets, an adenovirus encoding the human ARF cDNA (adARF) (Huang et al., 2003) was

employed to drive the overexpression of ARF in the retinoblastoma cell line WERI, which

expresses both MDM2 and MDM4 (Guo et al., 2008), and possesses extra copies of MDM4

due to genomic gain (Laurie et al., 2006). After viral infection, ARF expression was detected

by immunostaining as foci within the nucleus, characteristic of nucleolar localization where

ARF normally resides. Approximately 60% of treated cells were positive for ARF (Figure

47

+ -+ -P5

+ -P10 P15

p107-/-

p53

Bax

β-tubulin

p21

+ -+ -P5

+ -P10 P15

p107+/+

Rb:

Rb:p53

Bax

p21

β-tubulin

A

B

Figure 11. Western Blot analysis of p53, and p53-transcriptional targets p21 and Bax in Rb-knockout retinas in p107+/+ or p107-/- background Protein expression of p53 and its transcriptional target p21 was elevated in Rb-knockout retinas (-) compared to Wt-retinas (+) at P5, P10 and P15 in both p107+/+ (A) and p107-/- (B) backgrounds. However, no increase of the proapoptotic protein Bax was observed in Rb-knockout retinas. Note that protein lysates were extracted from whole retinas. Therefore, protein extracted from Rb-knockout retinas contained a mixture of central (intact Rb) and peripheral (mutant Rb) retinal materials.

48

P5+ -

P10 P15+ - + -

P5+ -

P10 P15+ - + -

1.0x 0.1x

Gadd45

p21

Tbp-1

Mdm2

Puma

Bax

Noxa

14-3-3σ

p107-/-

Rb:

Figure 12. mRNA expression of p53-transcriptional targets in Rb-knockout retinas in p107-/- background RT-PCR was performed to assess the expression of p53-target genes involved in autoregulation (Mdm2), cell cycle arrest (p21, 14-3-3σ, Gadd45) and apoptosis (Bax, Noxa, Puma) in Wt (+) and Rb-knockout (-) retinas at P5, P10 and P15. The response of p53-target genes to Rb-inactivation was heterogeneous. Mdm2, 14-3-3σ and Gadd45 were not induced upon Rb-inactivation, while the mRNA expression of p21, Bax, Noxa and Puma increased upon Rb-inactivation. Although Bax mRNA was elevated, its protein expression did not show an increase in Rb-knockout retinas (see Figure 11), suggesting that the magnitude of Bax mRNA induction was not sufficient to sustain detectable protein accumulation. Tbp-1 was used as an endogenous control. 0.1x = ten fold dilution of cDNA from 1.0x (to ensure that the log-phase of PCR amplification was captured)

49

Gadd45

p21

+ - +-Tbp-1

14-3-3σ

Mdm2

P5 P15 P5 P15+ - + -

0.1x1.0x

p107+/+

Rb:

Figure 13. mRNA expression of p53-transcriptional targets in Rb-knockout retinas in p107+/+ background Consistent with results obtained from the p107-/- strain, RT-PCR analysis showed a lack of induction for Mdm2, 14-3-3σ and Gadd45, and an elevation for p21 mRNA in Rb-knockout retinas (-) compared to Wt-retinas (+) at postnatal P5 and P15. Tbp-1 was used as an endogenous control. 0.1x = ten fold dilution of cDNA from 1.0x

50

14). Cells treated with adARF showed accumulation of p53 protein and p53-downstream

targets p21, MDM2 and BAX (Figure15A). RT-PCR analysis showed that ARF

overexpression also led to the induction of GADD45 and NOXA transcription (Figure 15B).

The changes in gene expression were accompanied by inhibition of cell growth (Figure 15C),

primarily due to inhibition of proliferation rather than cell death, given that no significant

increase in the number of dead cells was observed after treatment of adARF (Figure 15D).

Therefore, despite the presence of MDM2 and MDM4, enforced expression of ARF activated

p53 and p53-transcriptional targets, including those that remained inactive upon Rb-

inactivation during early mouse retinoblastoma development (Figure 11-13). This data

suggests that weak ARF protein expression, as observed in human retinoblastomas and in

pre-malignant Rb-/- retinas in mice, may prevent efficient activation of the p53-tumor

surveillance pathway during retinoblastoma development.

51

ARF DAPI/ ARF gfpad

AR

FA B C

D E F

adgf

p

Figure 14. Exogenous expression of ARF in the retinoblastoma cell line WERI Cells were infected with adenovirus encoding the human ARF cDNA (adARF) or gfp (adgfp). Immunostaining for ARF was performed four days after infection. Punctated stainings of ARF in adARF treated cells were consistent with nucleolar localization (E), indicating that exogenous ARF protein was properly localized.

52

p53

MDM2

BAX

β-tubulin

adgf

pad

ARF

GADD45

NOXA

TBP-1

A B

*

C

0

0.2

0.4

0.6

0.8

1

1.2

adgfp adARF

Rel

ativ

e N

umbe

r of V

iabl

e C

ells

0

0.2

0.4

0.6

0.8

1

1.2

adgfp adARF

Rel

ativ

e N

umbe

r of D

ead

Cel

ls*

** P < 0.05

**

* P > 0.05

D

p21

β-tubulin

adgf

pad

ARF

Figure 15. Effect of ARF overexpression on gene expression and cell growth A. Western Blot analysis showing protein expression of p53 and p53-transcriptional targets in cells treated with adARF or adgfp four days after infection. Expression of p53, MDM2, BAX and p21 proteins was elevated in adARF treated cells. B. RT-PCR analysis showing elevation in mRNA expression of the p53-transcriptional targets GADD45 and NOXA in cells treated with adARF. C. The number of viable cells was significantly reduced in adARF treated cells four days after infection compared to adgfp treated cells. D. No significant difference in the number of dead cells was observed between adgfp and adARF treated cells, suggesting that the growth inhibitory effect observed in C was caused by inhibition of proliferation rather than the induction of cell death. C & D are data from a representative experiment performed in triplicate. Statistical significance was determined by the t-test.

53

2.6. Discussion

2.6.1. Cause of low ARF protein in retinoblastomas

2.6.1.1. Low ARF protein in retinoblastomas is unlikely due to aberrant protein

degradation

Data from the present and previous studies (Guo et al., 2008) shows that ARF protein is

low in retinoblastoma despite robust expression of full length mRNA, suggesting that ARF

protein synthesis or stability may be deregulated. In this study, inhibition of the proteasome

by MG132 treatment in retinoblastoma cells did not lead to accumulation of ARF protein,

indicating that low ARF protein was not a result of aberrant turnover by this pathway.

Besides the proteasomal pathway, proteins may also be subjected to lysosomal degradation.

Studies in cell culture showed that chemical inhibition of the lysosome did not lead to ARF

protein accumulation (Kuo et al., 2004; Pollice et al., 2007), indicating ARF protein turnover

is primarily regulated by the proteasomal but not the lysosomal pathway.

Previous studies have shown that ARF protein stability is enhanced by its interaction

with TBP-1 (Pollice et al., 2004; Pollice et al., 2007) or NPM (Itahana et al., 2003;

Korgaonkar et al., 2005), therefore the absence of these proteins may shorten the half-life of

ARF. If ARF is aberrantly degraded in retinoblastoma cells due to the absence of TBP-1 or

NPM, inhibition of the proteasome by MG132 in this study would have resulted in ARF

accumulation. However, the lack of ARF accumulation in retinoblastoma cell lines after

MG132 treatment indicates that low ARF protein is not due to the absence of its binding

partners TBP-1 and NPM. Therefore, in retinoblastomas with high ARF mRNA expression,

54

low ARF protein is unlikely due to aberrant protein degradation. Instead, attenuation of

protein synthesis seems more plausible.

2.6.1.2. Role of microRNAs in regulation of ARF translation

Recently, much attention is focused on microRNAs and their role in regulation of gene

expression. ARF mRNA has been shown to be regulated by the microRNA, miR-24 (Lal et

al., 2008). In this study, we showed that miR-24 was expressed in retinoblastoma cell lines,

but at a lower level than in cell lines highly expressing ARF protein. Nevertheless, the level

of miR-24 attained in retinoblastoma cells may still be sufficient to repress ARF translation.

Future studies could employ antisense oligonucleotides to knockdown miR-24 in

retinoblastoma cells, and determine if cells respond by accumulating ARF protein and

activation of the p53-pathway.

In addition to miR-24, ARF translation may be regulated by other microRNAs. Using

the miRanda microRNA target detection software (Enright et al., 2003), ARF mRNA was

predicated to interact with 136 microRNAs in addition to miR-24 (Guo, personal

communication). In the future, expression analysis of candidate microRNAs could be

performed in retinoblastomas, followed by knockdown experiments to assess microRNA

functions.

2.6.1.3. Block in ARF protein synthesis may be overcome by elevation in ARF mRNA

expression

Data from this study indicates that the p53-pathway may be activated in retinoblastoma

cells by the infusion of extra copies of ARF mRNA, in this case, through the use of

adenovirus-mediated gene transfer. The adenovirus used encodes an ARF cDNA that lacks

55

the 3’ untranslated region. Since miR-24 interacts with binding sites residing in this region

(Lal et al., 2008), it is possible that the exogenous ARF mRNA escapes translational

regulation due to the absence of this segment and the binding site. Future experiments could

test ARF cDNA that encodes the full length mRNA containing the 3’ UTR, in order to

confirm whether the block in protein synthesis is simply overcome by extra copies of ARF

mRNA, or by the absence of the 3’ UTR in the exogenous cDNA.

A recent study reported that the chemotherapeutic agent, paclitaxel, could induce ARF

protein accumulation in a retinoblastoma cell line (Drago-Ferrante et al., 2008). The

expression of ARF mRNA was not measured in that study; therefore it is unclear whether

ARF protein accumulation was caused by an elevation in ARF mRNA expression or by other

mechanisms. Nevertheless, after paclitaxel treatment, increased E2F1 protein expression was

detected, which might induce ARF transcription and led to ARF protein accumulation. Future

studies could examine ARF mRNA expression in retinoblastoma cell lines after paclitaxel

treatment.

2.6.2. Expression of Arf, p53 & p53-transcriptional targets during early

retinoblastoma development in a mouse retinoblastoma model

2.6.2.1. ARF protein is expressed at low level during early mouse retinoblastoma

development

To assess the activity of Arf and the p53-pathway in response to Rb-inactivation,

expression analysis was performed in Rb-knockout retinas in p107+/+ and p107-/- mice during

early postnatal development. Arf mRNA was weakly detectable at P0 but became prominent

at P5, P10 and P15 in the Rb-knockout retinas. This pattern of Arf induction is parallel to the

56

E2f1 activity in Rb-knockout retinas, in which E2f1-dependent apoptosis elevates from P0 to

P8 (Chen et al., 2004; Chen et al., 2007), supporting that Arf transcription is E2f1-dependent.

Despite the induction of mRNA synthesis, Arf protein expression remained weak throughout

early retinoblastoma development. The functional relevance of Arf at this low level remains

to be elucidated. Downstream of Arf in the p53-pathway, several p53-transcriptional targets

remained inactive upon Rb-inactivation, suggesting that the attained Arf protein level might

not be sufficient to effectively activate p53. Nevertheless, the possibility of p53-independent

function of Arf cannot be excluded. The attained level of Arf protein during early

retinoblastoma development might not be sufficient to activate the p53-pathway; however, it

may be sufficient for Arf-functions outside the p53-pathway. This could be determined by

analyzing Rb or Rb-p107 mutant animals with Arf deletion. Precaution has to be taken when

using this genetic approach to assess Arf function in the mouse retina. Germline Arf

inactivation disrupts the regression of the hyaloid vascular system during postnatal

development, indirectly causing retinal dysplasia and detachment (McKeller et al., 2002).

These retinal phenotypes may confound tumor development, and it may be necessary to

conditionally inactivate Arf specifically in the retina using the Cre-lox system.

2.6.2.2. Despite increased p53 expression, some p53-target genes remained inactive

upon Rb-inactivation

The expression of p53 was elevated in Rb-knockout retinas (Figure 11). It is unclear

whether this p53 accumulation was Arf-dependent, since E2f1 has been shown to stabilize

p53 by promoting p53-phosphorylation independent of Arf (Berkovich and Ginsberg, 2003;

Powers et al., 2004; Rogoff et al., 2002; Rogoff et al., 2004). However, based on the low

expression of Arf protein, it is likely that p53 accumulation in Rb-knockout retinas is

57

mediated by E2f1. This prediction could be confirmed by analyzing p53 expression upon Rb-

inactivation in Arf-/- mice. Nevertheless, despite p53 accumulation in Rb-knockout retinas, no

increase in expression was observed for several of p53-transcriptional targets, including

Mdm2, 14-3-3σ and Gadd45 (Figure 12 and 13). Bax was induced at the mRNA level;

however, an elevation at the protein level was not observed (Figure 11). This data suggests

that the p53-pathway remains relatively inactive upon Rb-inactivation during early

retinoblastoma development in the Rb-p107 knockout murine model. Although a subset of

p53-targets, including p21, Bax, Noxa and Puma, were induced at the mRNA level, it is

unclear whether their activations were p53-dependent, since these genes are also direct

transcriptional targets of E2f1 (Gartel et al., 2000; Hershko and Ginsberg, 2004; Hiyama et

al., 1998). Analysis of expression for these p53 target genes in Rb-knockout retina with p53

deletion will resolve whether induction of p21, Bax, Noxa and Puma transcriptions is p53-

dependent.

14-3-3σ and Gadd45 are important effectors for p53-dependent cell cycle arrest. Failure

to activate these targets may compromise p53-mediated tumor surveillance and favor

malignant tumorigenesis. Gadd45 is an important tumor suppressor that regulates the cell

cycle through its interaction with the cyclin-CDK complex (Smith et al., 1994), and is also

implicated in maintaining genomic stability (Hollander et al., 1999). The 14-3-3σ protein is a

potent inhibitor of G2/M phase transition that sequesters cyclin B and CDC2 to prevent entry

into the nucleus (Chan et al., 1999). Therefore, the lack of 14-3-3σ and Gadd45 activation in

Rb-knockout retinas may favor cell cycle progression and subsequent acquisition of genomic

changes in Rb-deficient cells.

58

2.6.2.3. Several factors might influence the transactivating activity of p53

The response of p53 to particular stresses has been found to be highly cell-, tissue- and

stress-specific (Lu and Lane, 1993; MacCallum et al., 1996). p53 is at the hub of numerous

signaling pathways that are triggered by particular stresses, all of which may leave their

marks on p53 in the form of post-translational modifications, interactions with cofactors,

relocalization, and alteration in expression level. These various factors, collectively, dictate

the resulting cellular response to a particular stress (Murray-Zmijewski et al., 2008). In the

Rb-knockout retinas, the cellular response to p53 accumulation is presumably dictated by a

number of factors, including p53 localization, expression level, post-translational

modifications, and the availability of cofactors. A possible explanation for the lack of

activation of some p53-targets is that these factors are missing in the retina. The developing

retina may lack the modifications or cofactors that p53 needs in order to transactivate Mdm2,

14-3-3σ and Gadd45. Or, as the present study suggests, the attained level of p53 expression is

not sufficient for activation of those genes, perhaps due to the weak protein expression of its

upstream activator, Arf. We showed that exogenous expression of ARF in human

retinoblastoma cells led to p53 accumulation, and elevated protein expression of the p53-

targets p21, MDM2 and BAX. Enforced ARF expression also induced the transcription of

GADD45 and NOXA. This result suggests that during early retinoblastoma development, the

p53-pathway stays relatively inactive in part due to weak Arf expression (Figure 16). Future

experiments could validate this notion in vivo using a vector that overexpresses Arf in Rb-

p107 mutant retinas to determine if exogenous Arf leads to the activation of p53-target genes,

and reduce the frequency of malignant transformation.

59

Rb+/+ Rb-/-

p53

Mdm2

Gadd45

14-3-3σ

Weak Arf expression

p21

Bax

Puma

Noxa

E2f1

Figure 16. Role of the Arf-p53 pathway in early mouse retinoblastoma development Although p53 protein expression increased when Rb was inactivated, its transcriptional targets Gadd45, 14-3-3σ and Mdm2 were not activated. The lack of induction of Gadd45 and 14-3-3σ might compromise tumor surveillance, given their crucial roles in cell cycle arrest. This inactivity might be the result of weak Arf protein expression. In Rb-/- retinas, increased mRNA expression of p21, Bax, Puma and Noxa suggested that the p53-transactivating activity might be elevated. However, these genes are also targets of the E2f1 transcription factor, which is hyperactive in the absence of functional pRb. Further studies are required to determine whether the induction of these genes is p53- or E2f1-dependent.

60

2.6.3. Effect of exogenous ARF expression in retinoblastoma cell line

2.6.3.1. The p53-pathway can be activated by exogenous ARF in retinoblastoma cells

Data from this study suggests that the p53-pathway is not disrupted by MDM4 and

MDM2 in retinoblastomas, as previously postulated by Laurie et al (Laurie et al., 2006). The

retinoblastoma cell line used in this study, WERI, possesses extra copies of MDM4 due to

genomic gain, and it overexpresses MDM4 relative to normal fetal retina (Laurie et al., 2006).

Nevertheless, enforced expression of ARF was able to activate p53 and p53-target genes.

ARF physically interacts with MDM2 but not MDM4, and activates the p53-pathway by

relieving p53 from negative regulation by MDM2. Our data indicated that even in

retinoblastoma with MDM4 gain, p53 was also bound and regulated by MDM2, and was

available for activation by ARF. Therefore, in retinoblastoma, the lack of robust ARF protein

expression was an important factor contributing to the silencing of the p53-pathway.

61

Chapter 3: Discussion

3.1. Tumor surveillance by the ARF-p53-pathway in retinoblastoma

development

Data from the present and previous studies suggests that the p53-pathway does not play

a prominent role during early retinoblastoma development. MacPherson et al showed that the

deletion of p53 has no effect on apoptosis or ectopic division induced by Rb-inactivation

during early retinoblastoma development (MacPherson et al., 2004). Consistent with

previous observations in retinoblastoma, we now show that Rb-inactivation failed to induce

expression of p53-target genes, evidenced by the lack of induction of Mdm2, 14-3-3σ and

Gadd45. Our data suggests that weak Arf protein expression may contribute to the lack of

robust p53-response during early retinoblastoma development.

Despite the apparent low p53-activity during early tumor development, p53 may play an

important role later in retinoblastoma development. Inactivation of p53 in Rb-knockout retina

does not affect apoptosis nor does it initiate tumor formation (MacPherson et al., 2004). But

when inactivated in the tumor prone Rb-p107 mutant retina, p53 deletion leads to more

aggressive retinoblastoma development (Zhang et al., 2004). In human, expression analysis

by immunohistochemistry shows that p53 expression is only detectable in malignant

retinoblastoma but not in the benign retinoma (Dimaras et al., 2008), further supporting a

role of p53 in later stages of retinoblastoma development.

Genomic studies in mouse and human show that retinoblastomas exhibit genomic

instability (Corson and Gallie, 2007; MacPherson et al., 2007). Therefore, p53 may be

activated later during retinoblastoma development by the DNA damage and genomic

62

instability induced by loss of RB1. In the present study, analysis of p53 expression in Rb-/-

murine retinas only extended up to P15, at which point genomic instability might yet to

manifest, and therefore robust p53 activation could not be observed. To test this in the future,

retinas from older Rb-p107 mutant animals could be collected to examine the expression of

DNA damage markers, and determine if DNA damage coincides with p53-activation.

3.2. ARF-MDM2/4-p53 pathway: Therapeutic target in retinoblastomas?

3.2.1. Targeting MDM2 and MDM4

Previous studies show that the p53-pathway can be activated in retinoblastoma cells

using the small molecule Nutlin-3a (Vassilev et al., 2004), which inhibits MDM2 and MDM4

(Laurie et al., 2006), leading to the induction of apoptosis in retinoblastoma cell lines (Elison

et al., 2006; Laurie et al., 2006). The specificity of this approach is questionable. MDM4 and

MDM2 were reported to exhibit copy number gain on 1q and 12q in 63% and 10% of

retinoblastomas, respectively (Laurie et al., 2006). However, it is unclear whether MDM4

and MDM2 are the target genes of the respective chromosomal gains. Dimaras et al showed

that gain of MDM4 is not obligatory for gain of 1q (Dimaras et al., 2008), indicating that

either 1q contains multiple oncogenes, or MDM4 is a “passenger” of 1q gain in

retinoblastoma development. Future studies could determine the exact extent of genomic

copy number gain of MDM4 and MDM2, and their flanking genes on 1q and 12q,

respectively.

3.2.2. Targeting ARF

Data from the present and previous studies shows that the upstream activator of the p53-

pathway, ARF protein, is expressed at low level in retinoblastomas (Guo et al., 2008), and

63

enforced expression of ARF can activate the p53-pathway and inhibit cell growth in

retinoblastoma cell line. Therefore, ARF activation may be a feasible therapeutic strategy.

ARF mRNA expression is much higher in retinoblastomas than in normal fetal or adult

retinas (Guo et al., 2008; Laurie et al., 2006). Retinoblastomas possess a pool of ARF mRNA

that is not efficiently translated into protein. When the mechanism governing ARF protein

synthesis is elucidated, it may be modulated to utilize the ARF mRNA present in

retinoblastomas in order to synthesize ARF protein. In retinoblastomas, ARF translation

might be blocked by presently unknown microRNAs, which, when identified and

characterized, could be targeted using synthetic antisense antagonist.

Alternatively, the block in ARF protein synthesis in retinoblastomas may be overcome

simply by increasing ARF mRNA. In this case, modulating the upstream transcriptional

regulators of ARF may be a feasible therapeutic strategy. E2F3, which exhibits copy number

gain, and is overexpressed in retinoblastomas (Orlic et al., 2006), has been found to directly

repress Arf transcription (Aslanian et al., 2004). In retinoblastoma cell line with extra copies

of E2F3, shRNA-knockdown of E2F3 led to growth inhibition (Orlic-Milacic, submitted).

Therefore, future studies could determine whether the growth inhibition induced by E2F3

knockdown is ARF-mediated, by measuring ARF mRNA and protein expression after E2F3-

knockdown. In addition, Arf expression could be examined in Rb-knockout retinas with E2f3

inactivation, since E2f3 ablation may further activate Arf transcription. Future experiments

could also determine if other known ARF transcriptional repressors (BMI-1, CBX-7,

POKEMON, TBX-2 and -3), which are overexpressed in some cancers and prevent ARF

activation, are overexpressed in retinoblastomas relative to normal retinas. Subsequently,

64

RNAi experiment targeting the overexpressed repressor(s) could be performed, followed by

expression analysis of ARF, p53 and p53-target genes.

3.2.3. Targeting p53

A potential therapeutic strategy is to activate the endogenous wild-type p53 in order to

trigger apoptosis in retinoblastomas. Small molecules that activate p53 have been identified

and characterized. In addition to Nutlin-3a (Vassilev et al., 2004), RITA (reactivation of p53

and induction of tumor cell apoptosis) has been identified from a chemical screen that aimed

to discover novel p53-activating compounds (Issaeva et al., 2004). In comparison to Nutlin-

3a, which predominantly induced cell cycle arrest in tumor cells, RITA potently induced

apoptosis in a p53-dependent manner (Enge et al., 2009). RITA promoted the degradation of

p21, and hnRNP K, a cofactor that promotes p21 transcription, while Nutlin-3a did not

down-regulate either protein (Enge et al., 2009). p21, which was highly expressed in

retinoblastomas (Figure 3), possesses anti-apoptotic functions in addition to its role in cell

cycle arrest (Han et al., 2002). Therefore, RITA may be the more effective cancer killing

agent compared to Nutlin-3a, owing to its inhibitory effect on p21 (Enge et al., 2009). Future

studies could examine and compare the effect of RITA and Nutlin-3a on gene expression and

cell growth in retinoblastoma cell lines.

3.3. Future directions

This study has generated new insights into the role of ARF and the p53-pathway in

retinoblastoma development. Future studies will focus on answering the important questions

that remain, and validating the ideas and hypotheses that have arisen from this thesis. The

pending research questions (Q) and the proposed experiments (E) are listed below.

65

Q: What is the cause for low ARF protein in retinoblastomas despite high ARF mRNA

expression?

E: Elucidation of the functional role of miR-24, a known regulator of ARF translation (Lal et

al., 2008): Antisense oligonucleotides could be used to knockdown miR-24 in retinoblastoma

cell lines, and the response in ARF protein expression and p53-activity could be determined.

E: Identification of other microRNAs that inhibit ARF mRNA: The miRanda microRNA

target prediction software predicted 136 potential microRNAs (Guo, personal communication)

that bind the ARF mRNA. Expression analysis of these microRNAs could be performed in

retinoblastomas, and the functional relevance of candidates could then be assessed using

knockdown experiments in retinoblastoma cell lines.

Q: Can a further increase in ARF mRNA expression, which may lead to an excess amount of

ARF mRNA relative to the factors that inhibit ARF translation (eg. microRNAs), overcome

the blockade in protein synthesis in retinoblastomas?

E: Overexpression of ARF mRNA containing the 3’ UTR in retinoblastoma cell lines: The

ARF cDNA employed to transduce retinoblastoma cells in this study lacked the 3’ UTR,

therefore it is unclear whether the observed ARF protein accumulation was due to increased

ARF mRNA expression or the evasion of exogenous ARF mRNA from endogenous

translational regulation. To clarify the role of the 3’UTR, cells could be transduced with a

vector containing an ARF cDNA that possesses the entire ARF mRNA sequence including

the 3’ UTR, where many microRNAs are known to bind.

66

Q: Can inhibition of known ARF transcriptional repressors further increase ARF mRNA, and

induce ARF protein accumulation in retinoblastomas?

E: Knockdown of known ARF transcriptional repressors: E2F3 is a known transcriptional

repressor of ARF (Aslanian et al., 2004) and is overexpressed in human retinoblastomas

(Orlic et al., 2006). RNAi knockdown of E2F3 may increase ARF transcription in

retinoblastoma cells. In addition, expression of other known ARF transcriptional repressors

could be examined, and their roles in regulating ARF could subsequently be assessed by

knockdown experiments in retinoblastoma cell lines.

Q: Is the p53-pathway more active in malignant tumors, in which signals emanating from

genomic instability or other factors may trigger p53, than in pre-malignant lesions, in which

these signals have not yet manifested?

E: Determine the expression of p53 in concert with markers for genomic instability in

malignant tumors versus pre-malignant lesions in a mouse model of retinoblastoma (eg. Rb-

p107 knockout mice).

Q: Can the small molecule RITA activate p53, and lead to more potent cell death compared

to the chemotherapeutic agents currently used for retinoblastoma treatment?

E: Determine and compare the effect of RITA, Nutlin-3a, and the chemotherapeutic agents

currently used for retinoblastoma treatment, on cell proliferation and survival in

retinoblastoma cell lines.

67

References

Amente, S., Gargano, B., Diolaiti, D., Della Valle, G., Lania, L., and Majello, B. (2007). p14ARF interacts with N-Myc and inhibits its transcriptional activity. FEBS Lett 581, 821-825.

Amente, S., Gargano, B., Varrone, F., Ruggiero, L., Dominguez-Sola, D., Lania, L., and Majello, B. (2006). p14ARF directly interacts with Myc through the Myc BoxII domain. Cancer Biol Ther 5, 287-291.

Ameyar-Zazoua, M., Wisniewska, M. B., Bakiri, L., Wagner, E. F., Yaniv, M., and Weitzman, J. B. (2005). AP-1 dimers regulate transcription of the p14/p19ARF tumor suppressor gene. Oncogene 24, 2298-2306.

Aslanian, A., Iaquinta, P. J., Verona, R., and Lees, J. A. (2004). Repression of the Arf tumor suppressor by E2F3 is required for normal cell cycle kinetics. Genes Dev 18, 1413-1422.

Ayrault, O., Andrique, L., Larsen, C. J., and Seite, P. (2004). Human Arf tumor suppressor specifically interacts with chromatin containing the promoter of rRNA genes. Oncogene 23, 8097-8104.

Bartel, D. P. (2009). MicroRNAs: target recognition and regulatory functions. Cell 136, 215-233.

Bellan, C., De Falco, G., Tosi, G. M., Lazzi, S., Ferrari, F., Morbini, G., Bartolomei, S., Toti, P., Mangiavacchi, P., Cevenini, G., et al. (2002). Missing expression of pRb2/p130 in human retinoblastomas is associated with reduced apoptosis and lesser differentiation. Invest Ophthalmol Vis Sci 43, 3602-3608.

Benedict, W. F., Murphree, A. L., Banerjee, A., Spina, C. A., Sparkes, M. C., and Sparkes, R. S. (1983). Patient with 13 chromosome deletion: evidence that the retinoblastoma gene is a recessive cancer gene. Science 219, 973-975.

Berkovich, E., and Ginsberg, D. (2003). ATM is a target for positive regulation by E2F-1. Oncogene 22, 161-167.

Bernard, D., Martinez-Leal, J. F., Rizzo, S., Martinez, D., Hudson, D., Visakorpi, T., Peters, G., Carnero, A., Beach, D., and Gil, J. (2005). CBX7 controls the growth of normal and tumor-derived prostate cells by repressing the Ink4a/Arf locus. Oncogene 24, 5543-5551.

Bertwistle, D., Sugimoto, M., and Sherr, C. J. (2004). Physical and functional interactions of the Arf tumor suppressor protein with nucleophosmin/B23. Mol Cell Biol 24, 985-996.

Brady, S. N., Yu, Y., Maggi, L. B., Jr., and Weber, J. D. (2004). ARF impedes NPM/B23 shuttling in an Mdm2-sensitive tumor suppressor pathway. Mol Cell Biol 24, 9327-9338.

68

Bruce, J. L., Hurford, R. K., Jr., Classon, M., Koh, J., and Dyson, N. (2000). Requirements for cell cycle arrest by p16INK4a. Mol Cell 6, 737-742.

Brummelkamp, T. R., Kortlever, R. M., Lingbeek, M., Trettel, F., MacDonald, M. E., van Lohuizen, M., and Bernards, R. (2002). TBX-3, the gene mutated in Ulnar-Mammary Syndrome, is a negative regulator of p19ARF and inhibits senescence. J Biol Chem 277, 6567-6572.

Carleton, M., Mao, M., Biery, M., Warrener, P., Kim, S., Buser, C., Marshall, C. G., Fernandes, C., Annis, J., and Linsley, P. S. (2006). RNA interference-mediated silencing of mitotic kinesin KIF14 disrupts cell cycle progression and induces cytokinesis failure. Mol Cell Biol 26, 3853-3863.

Cavenee, W. K., Dryja, T. P., Phillips, R. A., Benedict, W. F., Godbout, R., Gallie, B. L., Murphree, A. L., Strong, L. C., and White, R. L. (1983). Expression of recessive alleles by chromosomal mechanisms in retinoblastoma. Nature 305, 779-784.

Chan, H. S., Gallie, B. L., Munier, F. L., and Beck Popovic, M. (2005). Chemotherapy for retinoblastoma. Ophthalmol Clin North Am 18, 55-63, viii.

Chan, T. A., Hermeking, H., Lengauer, C., Kinzler, K. W., and Vogelstein, B. (1999). 14-3-3Sigma is required to prevent mitotic catastrophe after DNA damage. Nature 401, 616-620.

Chaum, E., Ellsworth, R. M., Abramson, D. H., Haik, B. G., Kitchin, F. D., and Chaganti, R. S. (1984). Cytogenetic analysis of retinoblastoma: evidence for multifocal origin and in vivo gene amplification. Cytogenet Cell Genet 38, 82-91.

Chen, D., Gallie, B. L., and Squire, J. A. (2001). Minimal regions of chromosomal imbalance in retinoblastoma detected by comparative genomic hybridization. Cancer Genet Cytogenet 129, 57-63.

Chen, D., Kon, N., Li, M., Zhang, W., Qin, J., and Gu, W. (2005). ARF-BP1/Mule is a critical mediator of the ARF tumor suppressor. Cell 121, 1071-1083.

Chen, D., Livne-bar, I., Vanderluit, J. L., Slack, R. S., Agochiya, M., and Bremner, R. (2004). Cell-specific effects of RB or RB/p107 loss on retinal development implicate an intrinsically death-resistant cell-of-origin in retinoblastoma. Cancer Cell 5, 539-551.

Chen, D., Opavsky, R., Pacal, M., Tanimoto, N., Wenzel, P., Seeliger, M. W., Leone, G., and Bremner, R. (2007). Rb-Mediated Neuronal Differentiation through Cell-Cycle-Independent Regulation of E2f3a. PLoS Biol 5, e179.

Chen, L., and Chen, J. (2003). MDM2-ARF complex regulates p53 sumoylation. Oncogene 22, 5348-5357.

Chintagumpala, M., Chevez-Barrios, P., Paysse, E. A., Plon, S. E., and Hurwitz, R. (2007). Retinoblastoma: review of current management. Oncologist 12, 1237-1246.

69

Chipuk, J. E., Kuwana, T., Bouchier-Hayes, L., Droin, N. M., Newmeyer, D. D., Schuler, M., and Green, D. R. (2004). Direct activation of Bax by p53 mediates mitochondrial membrane permeabilization and apoptosis. Science 303, 1010-1014.

Chow, R. L., and Lang, R. A. (2001). Early eye development in vertebrates. Annu Rev Cell Dev Biol 17, 255-296.

Classon, M., and Dyson, N. (2001). p107 and p130: versatile proteins with interesting pockets. Exp Cell Res 264, 135-147.

Cobrinik, D. (2005). Pocket proteins and cell cycle control. Oncogene 24, 2796-2809.

Comings, D. E. (1973). A general theory of carcinogenesis. Proc Natl Acad Sci U S A 70, 3324-3328.

Core, N., Joly, F., Boned, A., and Djabali, M. (2004). Disruption of E2F signaling suppresses the INK4a-induced proliferative defect in M33-deficient mice. Oncogene 23, 7660-7668.

Corson, T. W., and Gallie, B. L. (2007). One hit, two hits, three hits, more? Genomic changes in the development of retinoblastoma. Genes Chromosomes Cancer 46, 617-634.

Corson, T. W., Huang, A., Tsao, M. S., and Gallie, B. L. (2005). KIF14 is a candidate oncogene in the 1q minimal region of genomic gain in multiple cancers. Oncogene 24, 4741-4753.

Corson, T. W., Zhu, C. Q., Lau, S. K., Shepherd, F. A., Tsao, M. S., and Gallie, B. L. (2007). KIF14 messenger RNA expression is independently prognostic for outcome in lung cancer. Clin Cancer Res 13, 3229-3234.

Costa, R. H., Kalinichenko, V. V., Major, M. L., and Raychaudhuri, P. (2005). New and unexpected: forkhead meets ARF. Curr Opin Genet Dev 15, 42-48.

Damalas, A., Kahan, S., Shtutman, M., Ben-Ze'ev, A., and Oren, M. (2001). Deregulated beta-catenin induces a p53- and ARF-dependent growth arrest and cooperates with Ras in transformation. Embo J 20, 4912-4922.

Datta, A., Nag, A., and Raychaudhuri, P. (2002). Differential regulation of E2F1, DP1, and the E2F1/DP1 complex by ARF. Mol Cell Biol 22, 8398-8408.

Datta, A., Sen, J., Hagen, J., Korgaonkar, C. K., Caffrey, M., Quelle, D. E., Hughes, D. E., Ackerson, T. J., Costa, R. H., and Raychaudhuri, P. (2005). ARF directly binds DP1: interaction with DP1 coincides with the G1 arrest function of ARF. Mol Cell Biol 25, 8024-8036.

Devesa, S. S. (1975). The incidence of retinoblastoma. Am J Ophthalmol 80, 263-265.

Dimaras, H., Khetan, V., Halliday, W., Orlic, M., Prigoda, N. L., Piovesan, B., Marrano, P., Corson, T. W., Eagle, R. C., Jr., Squire, J. A., and Gallie, B. L. (2008). Loss of RB1 induces

70

non-proliferative retinoma: increasing genomic instability correlates with progression to retinoblastoma. Hum Mol Genet 17, 1363-1372.

Drago-Ferrante, R., Santulli, A., Di Fiore, R., Giuliano, M., Calvaruso, G., Tesoriere, G., and Vento, R. (2008). Low doses of paclitaxel potently induce apoptosis in human retinoblastoma Y79 cells by up-regulating E2F1. Int J Oncol 33, 677-687.

Dyer, M. A., and Cepko, C. L. (2001). Regulating proliferation during retinal development. Nat Rev Neurosci 2, 333-342.

Dyson, N. (1994). pRB, p107 and the regulation of the E2F transcription factor. J Cell Sci Suppl 18, 81-87.

Eischen, C. M., Weber, J. D., Roussel, M. F., Sherr, C. J., and Cleveland, J. L. (1999). Disruption of the ARF-Mdm2-p53 tumor suppressor pathway in Myc-induced lymphomagenesis. Genes Dev 13, 2658-2669.

Elison, J. R., Cobrinik, D., Claros, N., Abramson, D. H., and Lee, T. C. (2006). Small molecule inhibition of HDM2 leads to p53-mediated cell death in retinoblastoma cells. Arch Ophthalmol 124, 1269-1275.

Enge, M., Bao, W., Hedstrom, E., Jackson, S. P., Moumen, A., and Selivanova, G. (2009). MDM2-dependent downregulation of p21 and hnRNP K provides a switch between apoptosis and growth arrest induced by pharmacologically activated p53. Cancer Cell 15, 171-183.

Enright, A. J., John, B., Gaul, U., Tuschl, T., Sander, C., and Marks, D. S. (2003). MicroRNA targets in Drosophila. Genome Biol 5, R1.

Esteller, M., Tortola, S., Toyota, M., Capella, G., Peinado, M. A., Baylin, S. B., and Herman, J. G. (2000). Hypermethylation-associated inactivation of p14(ARF) is independent of p16(INK4a) methylation and p53 mutational status. Cancer Res 60, 129-133.

Eymin, B., Claverie, P., Salon, C., Leduc, C., Col, E., Brambilla, E., Khochbin, S., and Gazzeri, S. (2006). p14ARF activates a Tip60-dependent and p53-independent ATM/ATR/CHK pathway in response to genotoxic stress. Mol Cell Biol 26, 4339-4350.

Eymin, B., Karayan, L., Seite, P., Brambilla, C., Brambilla, E., Larsen, C. J., and Gazzeri, S. (2001). Human ARF binds E2F1 and inhibits its transcriptional activity. Oncogene 20, 1033-1041.

Frank, S. R., Parisi, T., Taubert, S., Fernandez, P., Fuchs, M., Chan, H. M., Livingston, D. M., and Amati, B. (2003). MYC recruits the TIP60 histone acetyltransferase complex to chromatin. EMBO Rep 4, 575-580.

Freedberg, D. E., Rigas, S. H., Russak, J., Gai, W., Kaplow, M., Osman, I., Turner, F., Randerson-Moor, J. A., Houghton, A., Busam, K., et al. (2008). Frequent p16-independent inactivation of p14ARF in human melanoma. J Natl Cancer Inst 100, 784-795.

71

Friend, S. H., Bernards, R., Rogelj, S., Weinberg, R. A., Rapaport, J. M., Albert, D. M., and Dryja, T. P. (1986). A human DNA segment with properties of the gene that predisposes to retinoblastoma and osteosarcoma. Nature 323, 643-646.

Fung, Y. K., Murphree, A. L., T'Ang, A., Qian, J., Hinrichs, S. H., and Benedict, W. F. (1987). Structural evidence for the authenticity of the human retinoblastoma gene. Science 236, 1657-1661.

Gartel, A. L., Najmabadi, F., Goufman, E., and Tyner, A. L. (2000). A role for E2F1 in Ras activation of p21(WAF1/CIP1) transcription. Oncogene 19, 961-964.

Gil, J., Bernard, D., Martinez, D., and Beach, D. (2004). Polycomb CBX7 has a unifying role in cellular lifespan. Nat Cell Biol 6, 67-72.

Gil, J., and Peters, G. (2006). Regulation of the INK4b-ARF-INK4a tumour suppressor locus: all for one or one for all. Nat Rev Mol Cell Biol 7, 667-677.

Gonzalez, S., Klatt, P., Delgado, S., Conde, E., Lopez-Rios, F., Sanchez-Cespedes, M., Mendez, J., Antequera, F., and Serrano, M. (2006). Oncogenic activity of Cdc6 through repression of the INK4/ARF locus. Nature 440, 702-706.

Grasemann, C., Gratias, S., Stephan, H., Schuler, A., Schramm, A., Klein-Hitpass, L., Rieder, H., Schneider, S., Kappes, F., Eggert, A., and Lohmann, D. R. (2005). Gains and overexpression identify DEK and E2F3 as targets of chromosome 6p gains in retinoblastoma. Oncogene 24, 6441-6449.

Gruneberg, U., Neef, R., Li, X., Chan, E. H., Chalamalasetty, R. B., Nigg, E. A., and Barr, F. A. (2006). KIF14 and citron kinase act together to promote efficient cytokinesis. J Cell Biol 172, 363-372.

Guo, Y., Pajovic, S., and Gallie, B. L. (2008). Expression of p14(ARF), MDM2, and MDM4 in human retinoblastoma. Biochem Biophys Res Commun.

Han, Z., Wei, W., Dunaway, S., Darnowski, J. W., Calabresi, P., Sedivy, J., Hendrickson, E. A., Balan, K. V., Pantazis, P., and Wyche, J. H. (2002). Role of p21 in apoptosis and senescence of human colon cancer cells treated with camptothecin. J Biol Chem 277, 17154-17160.

Haupt, S., Berger, M., Goldberg, Z., and Haupt, Y. (2003). Apoptosis - the p53 network. J Cell Sci 116, 4077-4085.

Hershko, T., and Ginsberg, D. (2004). Up-regulation of Bcl-2 homology 3 (BH3)-only proteins by E2F1 mediates apoptosis. J Biol Chem 279, 8627-8634.

Herzog, S., Lohmann, D. R., Buiting, K., Schuler, A., Horsthemke, B., Rehder, H., and Rieder, H. (2001). Marked differences in unilateral isolated retinoblastomas from young and older children studied by comparative genomic hybridization. Hum Genet 108, 98-104.

72

Hiyama, H., Iavarone, A., and Reeves, S. A. (1998). Regulation of the cdk inhibitor p21 gene during cell cycle progression is under the control of the transcription factor E2F. Oncogene 16, 1513-1523.

Hollander, M. C., Sheikh, M. S., Bulavin, D. V., Lundgren, K., Augeri-Henmueller, L., Shehee, R., Molinaro, T. A., Kim, K. E., Tolosa, E., Ashwell, J. D., et al. (1999). Genomic instability in Gadd45a-deficient mice. Nat Genet 23, 176-184.

Huang, Y., Tyler, T., Saadatmandi, N., Lee, C., Borgstrom, P., and Gjerset, R. A. (2003). Enhanced tumor suppression by a p14ARF/p53 bicistronic adenovirus through increased p53 protein translation and stability. Cancer Res 63, 3646-3653.

Inoue, K., Roussel, M. F., and Sherr, C. J. (1999). Induction of ARF tumor suppressor gene expression and cell cycle arrest by transcription factor DMP1. Proc Natl Acad Sci U S A 96, 3993-3998.

Isono, K., Fujimura, Y., Shinga, J., Yamaki, M., J, O. W., Takihara, Y., Murahashi, Y., Takada, Y., Mizutani-Koseki, Y., and Koseki, H. (2005). Mammalian polyhomeotic homologues Phc2 and Phc1 act in synergy to mediate polycomb repression of Hox genes. Mol Cell Biol 25, 6694-6706.

Issaeva, N., Bozko, P., Enge, M., Protopopova, M., Verhoef, L. G., Masucci, M., Pramanik, A., and Selivanova, G. (2004). Small molecule RITA binds to p53, blocks p53-HDM-2 interaction and activates p53 function in tumors. Nat Med 10, 1321-1328.

Itahana, K., Bhat, K. P., Jin, A., Itahana, Y., Hawke, D., Kobayashi, R., and Zhang, Y. (2003). Tumor suppressor ARF degrades B23, a nucleolar protein involved in ribosome biogenesis and cell proliferation. Mol Cell 12, 1151-1164.

Jacks, T., Fazeli, A., Schmitt, E. M., Bronson, R. T., Goodell, M. A., and Weinberg, R. A. (1992). Effects of an Rb mutation in the mouse. Nature 359, 295-300.

Jacobs, J. J., Keblusek, P., Robanus-Maandag, E., Kristel, P., Lingbeek, M., Nederlof, P. M., van Welsem, T., van de Vijver, M. J., Koh, E. Y., Daley, G. Q., and van Lohuizen, M. (2000). Senescence bypass screen identifies TBX2, which represses Cdkn2a (p19(ARF)) and is amplified in a subset of human breast cancers. Nat Genet 26, 291-299.

Kalinichenko, V. V., Major, M. L., Wang, X., Petrovic, V., Kuechle, J., Yoder, H. M., Dennewitz, M. B., Shin, B., Datta, A., Raychaudhuri, P., and Costa, R. H. (2004). Foxm1b transcription factor is essential for development of hepatocellular carcinomas and is negatively regulated by the p19ARF tumor suppressor. Genes Dev 18, 830-850.

Kamijo, T., van de Kamp, E., Chong, M. J., Zindy, F., Diehl, J. A., Sherr, C. J., and McKinnon, P. J. (1999). Loss of the ARF tumor suppressor reverses premature replicative arrest but not radiation hypersensitivity arising from disabled atm function. Cancer Res 59, 2464-2469.

73

Kamijo, T., Zindy, F., Roussel, M. F., Quelle, D. E., Downing, J. R., Ashmun, R. A., Grosveld, G., and Sherr, C. J. (1997). Tumor suppression at the mouse INK4a locus mediated by the alternative reading frame product p19ARF. Cell 91, 649-659.

Knudson, A. G., Jr. (1971). Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci U S A 68, 820-823.

Komori, H., Enomoto, M., Nakamura, M., Iwanaga, R., and Ohtani, K. (2005). Distinct E2F-mediated transcriptional program regulates p14ARF gene expression. Embo J 24, 3724-3736.

Kondoh, N., Wakatsuki, T., Ryo, A., Hada, A., Aihara, T., Horiuchi, S., Goseki, N., Matsubara, O., Takenaka, K., Shichita, M., et al. (1999). Identification and characterization of genes associated with human hepatocellular carcinogenesis. Cancer Res 59, 4990-4996.

Korgaonkar, C., Hagen, J., Tompkins, V., Frazier, A. A., Allamargot, C., Quelle, F. W., and Quelle, D. E. (2005). Nucleophosmin (B23) targets ARF to nucleoli and inhibits its function. Mol Cell Biol 25, 1258-1271.

Krimpenfort, P., Quon, K. C., Mooi, W. J., Loonstra, A., and Berns, A. (2001). Loss of p16Ink4a confers susceptibility to metastatic melanoma in mice. Nature 413, 83-86.

Krones-Herzig, A., Adamson, E., and Mercola, D. (2003). Early growth response 1 protein, an upstream gatekeeper of the p53 tumor suppressor, controls replicative senescence. Proc Natl Acad Sci U S A 100, 3233-3238.

Kuo, M. L., den Besten, W., Bertwistle, D., Roussel, M. F., and Sherr, C. J. (2004). N-terminal polyubiquitination and degradation of the Arf tumor suppressor. Genes Dev 18, 1862-1874.

Kusnetsova, L. E., Prigogina, E. L., Pogosianz, H. E., and Belkina, B. M. (1982). Similar chromosomal abnormalities in several retinoblastomas. Hum Genet 61, 201-204.

Lal, A., Kim, H. H., Abdelmohsen, K., Kuwano, Y., Pullmann, R., Jr., Srikantan, S., Subrahmanyam, R., Martindale, J. L., Yang, X., Ahmed, F., et al. (2008). p16(INK4a) translation suppressed by miR-24. PLoS ONE 3, e1864.

Laurie, N. A., Donovan, S. L., Shih, C. S., Zhang, J., Mills, N., Fuller, C., Teunisse, A., Lam, S., Ramos, Y., Mohan, A., et al. (2006). Inactivation of the p53 pathway in retinoblastoma. Nature 444, 61-66.

Lee, W. H., Bookstein, R., Hong, F., Young, L. J., Shew, J. Y., and Lee, E. Y. (1987). Human retinoblastoma susceptibility gene: cloning, identification, and sequence. Science 235, 1394-1399.

Levine, A. J., Hu, W., and Feng, Z. (2006). The P53 pathway: what questions remain to be explored? Cell Death Differ 13, 1027-1036.

74

Lillington, D. M., Kingston, J. E., Coen, P. G., Price, E., Hungerford, J., Domizio, P., Young, B. D., and Onadim, Z. (2003). Comparative genomic hybridization of 49 primary retinoblastoma tumors identifies chromosomal regions associated with histopathology, progression, and patient outcome. Genes Chromosomes Cancer 36, 121-128.

Lingbeek, M. E., Jacobs, J. J., and van Lohuizen, M. (2002). The T-box repressors TBX2 and TBX3 specifically regulate the tumor suppressor gene p14ARF via a variant T-site in the initiator. J Biol Chem 277, 26120-26127.

Lu, X., and Lane, D. P. (1993). Differential induction of transcriptionally active p53 following UV or ionizing radiation: defects in chromosome instability syndromes? Cell 75, 765-778.

Lu, Z. L., Luo, D. Z., and Wen, J. M. (2005). Expression and significance of tumor-related genes in HCC. World J Gastroenterol 11, 3850-3854.

Lund, A. H., and van Lohuizen, M. (2004). Polycomb complexes and silencing mechanisms. Curr Opin Cell Biol 16, 239-246.

MacCallum, D. E., Hupp, T. R., Midgley, C. A., Stuart, D., Campbell, S. J., Harper, A., Walsh, F. S., Wright, E. G., Balmain, A., Lane, D. P., and Hall, P. A. (1996). The p53 response to ionising radiation in adult and developing murine tissues. Oncogene 13, 2575-2587.

Macpherson, D. (2008). Insights from mouse models into human retinoblastoma. Cell Div 3, 9.

MacPherson, D., Conkrite, K., Tam, M., Mukai, S., Mu, D., and Jacks, T. (2007). Murine bilateral retinoblastoma exhibiting rapid-onset, metastatic progression and N-myc gene amplification. Embo J 26, 784-794.

MacPherson, D., Sage, J., Kim, T., Ho, D., McLaughlin, M. E., and Jacks, T. (2004). Cell type-specific effects of Rb deletion in the murine retina. Genes Dev 18, 1681-1694.

Maeda, T., Hobbs, R. M., Merghoub, T., Guernah, I., Zelent, A., Cordon-Cardo, C., Teruya-Feldstein, J., and Pandolfi, P. P. (2005). Role of the proto-oncogene Pokemon in cellular transformation and ARF repression. Nature 433, 278-285.

Maestro, R., Dei Tos, A. P., Hamamori, Y., Krasnokutsky, S., Sartorelli, V., Kedes, L., Doglioni, C., Beach, D. H., and Hannon, G. J. (1999). Twist is a potential oncogene that inhibits apoptosis. Genes Dev 13, 2207-2217.

Mairal, A., Pinglier, E., Gilbert, E., Peter, M., Validire, P., Desjardins, L., Doz, F., Aurias, A., and Couturier, J. (2000). Detection of chromosome imbalances in retinoblastoma by parallel karyotype and CGH analyses. Genes Chromosomes Cancer 28, 370-379.

75

Marchong, M. N., Chen, D., Corson, T. W., Lee, C., Harmandayan, M., Bowles, E., Chen, N., and Gallie, B. L. (2004). Minimal 16q genomic loss implicates cadherin-11 in retinoblastoma. Mol Cancer Res 2, 495-503.

Marees, T., Moll, A. C., Imhof, S. M., de Boer, M. R., Ringens, P. J., and van Leeuwen, F. E. (2008). Risk of Second Malignancies in Survivors of Retinoblastoma: More Than 40 Years of Follow-up. J Natl Cancer Inst.

Marine, J. C., and Jochemsen, A. G. (2004). Mdmx and Mdm2: brothers in arms? Cell Cycle 3, 900-904.

Martelli, F., Hamilton, T., Silver, D. P., Sharpless, N. E., Bardeesy, N., Rokas, M., DePinho, R. A., Livingston, D. M., and Grossman, S. R. (2001). p19ARF targets certain E2F species for degradation. Proc Natl Acad Sci U S A 98, 4455-4460.

McKeller, R. N., Fowler, J. L., Cunningham, J. J., Warner, N., Smeyne, R. J., Zindy, F., and Skapek, S. X. (2002). The Arf tumor suppressor gene promotes hyaloid vascular regression during mouse eye development. Proc Natl Acad Sci U S A 99, 3848-3853.

Moll, U. M., Wolff, S., Speidel, D., and Deppert, W. (2005). Transcription-independent pro-apoptotic functions of p53. Curr Opin Cell Biol 17, 631-636.

Molofsky, A. V., He, S., Bydon, M., Morrison, S. J., and Pardal, R. (2005). Bmi-1 promotes neural stem cell self-renewal and neural development but not mouse growth and survival by repressing the p16Ink4a and p19Arf senescence pathways. Genes Dev 19, 1432-1437.

Muller, M., Wilder, S., Bannasch, D., Israeli, D., Lehlbach, K., Li-Weber, M., Friedman, S. L., Galle, P. R., Stremmel, W., Oren, M., and Krammer, P. H. (1998). p53 activates the CD95 (APO-1/Fas) gene in response to DNA damage by anticancer drugs. J Exp Med 188, 2033-2045.

Murray-Zmijewski, F., Slee, E. A., and Lu, X. (2008). A complex barcode underlies the heterogeneous response of p53 to stress. Nat Rev Mol Cell Biol 9, 702-712.

Nicholson, D. W., and Thornberry, N. A. (2003). Apoptosis. Life and death decisions. Science 299, 214-215.

Oeggerli, M., Schraml, P., Ruiz, C., Bloch, M., Novotny, H., Mirlacher, M., Sauter, G., and Simon, R. (2006). E2F3 is the main target gene of the 6p22 amplicon with high specificity for human bladder cancer. Oncogene 25, 6538-6543.

Olsson, A. Y., Feber, A., Edwards, S., Te Poele, R., Giddings, I., Merson, S., and Cooper, C. S. (2007). Role of E2F3 expression in modulating cellular proliferation rate in human bladder and prostate cancer cells. Oncogene 26, 1028-1037.

Orlic, M., Spencer, C. E., Wang, L., and Gallie, B. L. (2006). Expression analysis of 6p22 genomic gain in retinoblastoma. Genes Chromosomes Cancer 45, 72-82.

76

Otte, A. P., and Kwaks, T. H. (2003). Gene repression by Polycomb group protein complexes: a distinct complex for every occasion? Curr Opin Genet Dev 13, 448-454.

Pacal, M., and Bremner, R. (2006). Insights from animal models on the origins and progression of retinoblastoma. Curr Mol Med 6, 759-781.

Palmero, I., Pantoja, C., and Serrano, M. (1998). p19ARF links the tumour suppressor p53 to Ras. Nature 395, 125-126.

Pogosianz, H. E., and Kuznetsova, L. E. (1986). Nonrandom chromosomal changes in retinoblastomas. Arch Geschwulstforsch 56, 135-143.

Pollice, A., Nasti, V., Ronca, R., Vivo, M., Lo Iacono, M., Calogero, R., Calabro, V., and La Mantia, G. (2004). Functional and physical interaction of the human ARF tumor suppressor with Tat-binding protein-1. J Biol Chem 279, 6345-6353.

Pollice, A., Sepe, M., Villella, V. R., Tolino, F., Vivo, M., Calabro, V., and La Mantia, G. (2007). TBP-1 protects the human oncosuppressor p14ARF from proteasomal degradation. Oncogene 26, 5154-5162.

Powers, J. T., Hong, S., Mayhew, C. N., Rogers, P. M., Knudsen, E. S., and Johnson, D. G. (2004). E2F1 uses the ATM signaling pathway to induce p53 and Chk2 phosphorylation and apoptosis. Mol Cancer Res 2, 203-214.

Raveh, T., Droguett, G., Horwitz, M. S., DePinho, R. A., and Kimchi, A. (2001). DAP kinase activates a p19ARF/p53-mediated apoptotic checkpoint to suppress oncogenic transformation. Nat Cell Biol 3, 1-7.

Reef, S., Zalckvar, E., Shifman, O., Bialik, S., Sabanay, H., Oren, M., and Kimchi, A. (2006). A short mitochondrial form of p19ARF induces autophagy and caspase-independent cell death. Mol Cell 22, 463-475.

Rizos, H., Woodruff, S., and Kefford, R. F. (2005). p14ARF interacts with the SUMO-conjugating enzyme Ubc9 and promotes the sumoylation of its binding partners. Cell Cycle 4, 597-603.

Robanus-Maandag, E., Dekker, M., van der Valk, M., Carrozza, M. L., Jeanny, J. C., Dannenberg, J. H., Berns, A., and te Riele, H. (1998). p107 is a suppressor of retinoblastoma development in pRb-deficient mice. Genes Dev 12, 1599-1609.

Rocco, J. W., and Sidransky, D. (2001). p16(MTS-1/CDKN2/INK4a) in cancer progression. Exp Cell Res 264, 42-55.

Rocha, S., Campbell, K. J., and Perkins, N. D. (2003). p53- and Mdm2-independent repression of NF-kappa B transactivation by the ARF tumor suppressor. Mol Cell 12, 15-25.

77

Rocha, S., Garrett, M. D., Campbell, K. J., Schumm, K., and Perkins, N. D. (2005). Regulation of NF-kappaB and p53 through activation of ATR and Chk1 by the ARF tumour suppressor. Embo J 24, 1157-1169.

Rogoff, H. A., Pickering, M. T., Debatis, M. E., Jones, S., and Kowalik, T. F. (2002). E2F1 induces phosphorylation of p53 that is coincident with p53 accumulation and apoptosis. Mol Cell Biol 22, 5308-5318.

Rogoff, H. A., Pickering, M. T., Frame, F. M., Debatis, M. E., Sanchez, Y., Jones, S., and Kowalik, T. F. (2004). Apoptosis associated with deregulated E2F activity is dependent on E2F1 and Atm/Nbs1/Chk2. Mol Cell Biol 24, 2968-2977.

Sanchez-Carbayo, M., Socci, N. D., Lozano, J. J., Li, W., Charytonowicz, E., Belbin, T. J., Prystowsky, M. B., Ortiz, A. R., Childs, G., and Cordon-Cardo, C. (2003). Gene discovery in bladder cancer progression using cDNA microarrays. Am J Pathol 163, 505-516.

Savli, H., Aalto, Y., Nagy, B., Knuutila, S., and Pakkala, S. (2002). Gene expression analysis of 1,25(OH)2D3-dependent differentiation of HL-60 cells: a cDNA array study. Br J Haematol 118, 1065-1070.

Scott, C. L., Gil, J., Hernando, E., Teruya-Feldstein, J., Narita, M., Martinez, D., Visakorpi, T., Mu, D., Cordon-Cardo, C., Peters, G., et al. (2007). Role of the chromobox protein CBX7 in lymphomagenesis. Proc Natl Acad Sci U S A 104, 5389-5394.

Sharpless, N. E., Bardeesy, N., Lee, K. H., Carrasco, D., Castrillon, D. H., Aguirre, A. J., Wu, E. A., Horner, J. W., and DePinho, R. A. (2001). Loss of p16Ink4a with retention of p19Arf predisposes mice to tumorigenesis. Nature 413, 86-91.

Sharpless, N. E., Kannan, K., Xu, J., Bosenberg, M. W., and Chin, L. (2003). Both products of the mouse Ink4a/Arf locus suppress melanoma formation in vivo. Oncogene 22, 5055-5059.

Sharpless, N. E., Ramsey, M. R., Balasubramanian, P., Castrillon, D. H., and DePinho, R. A. (2004). The differential impact of p16(INK4a) or p19(ARF) deficiency on cell growth and tumorigenesis. Oncogene 23, 379-385.

Sherr, C. J. (2001). The INK4a/ARF network in tumour suppression. Nat Rev Mol Cell Biol 2, 731-737.

Sherr, C. J. (2006). Divorcing ARF and p53: an unsettled case. Nat Rev Cancer 6, 663-673.

Silva, J., Garcia, J. M., Pena, C., Garcia, V., Dominguez, G., Suarez, D., Camacho, F. I., Espinosa, R., Provencio, M., Espana, P., and Bonilla, F. (2006). Implication of polycomb members Bmi-1, Mel-18, and Hpc-2 in the regulation of p16INK4a, p14ARF, h-TERT, and c-Myc expression in primary breast carcinomas. Clin Cancer Res 12, 6929-6936.

78

Sinclair, C. S., Adem, C., Naderi, A., Soderberg, C. L., Johnson, M., Wu, K., Wadum, L., Couch, V. L., Sellers, T. A., Schaid, D., et al. (2002). TBX2 is preferentially amplified in BRCA1- and BRCA2-related breast tumors. Cancer Res 62, 3587-3591.

Smith, M. L., Chen, I. T., Zhan, Q., Bae, I., Chen, C. Y., Gilmer, T. M., Kastan, M. B., O'Connor, P. M., and Fornace, A. J., Jr. (1994). Interaction of the p53-regulated protein Gadd45 with proliferating cell nuclear antigen. Science 266, 1376-1380.

Squire, J., Gallie, B. L., and Phillips, R. A. (1985). A detailed analysis of chromosomal changes in heritable and non-heritable retinoblastoma. Hum Genet 70, 291-301.

Sreeramaneni, R., Chaudhry, A., McMahon, M., Sherr, C. J., and Inoue, K. (2005). Ras-Raf-Arf signaling critically depends on the Dmp1 transcription factor. Mol Cell Biol 25, 220-232.

Sugimoto, M., Kuo, M. L., Roussel, M. F., and Sherr, C. J. (2003). Nucleolar Arf tumor suppressor inhibits ribosomal RNA processing. Mol Cell 11, 415-424.

Suzuki, H., Kurita, M., Mizumoto, K., Moriyama, M., Aiso, S., Nishimoto, I., and Matsuoka, M. (2005). The ARF tumor suppressor inhibits BCL6-mediated transcriptional repression. Biochem Biophys Res Commun 326, 242-248.

Tago, K., Chiocca, S., and Sherr, C. J. (2005). Sumoylation induced by the Arf tumor suppressor: a p53-independent function. Proc Natl Acad Sci U S A 102, 7689-7694.

Tao, W., and Levine, A. J. (1999). P19(ARF) stabilizes p53 by blocking nucleo-cytoplasmic shuttling of Mdm2. Proc Natl Acad Sci U S A 96, 6937-6941.

Tedesco, D., Lukas, J., and Reed, S. I. (2002). The pRb-related protein p130 is regulated by phosphorylation-dependent proteolysis via the protein-ubiquitin ligase SCF(Skp2). Genes Dev 16, 2946-2957.

Toledo, F., and Wahl, G. M. (2006). Regulating the p53 pathway: in vitro hypotheses, in vivo veritas. Nat Rev Cancer 6, 909-923.

Tsai, K. Y., MacPherson, D., Rubinson, D. A., Nikitin, A. Y., Bronson, R., Mercer, K. L., Crowley, D., and Jacks, T. (2002). ARF mutation accelerates pituitary tumor development in Rb+/- mice. Proc Natl Acad Sci U S A 99, 16865-16870.

van der Wal, J. E., Hermsen, M. A., Gille, H. J., Schouten-Van Meeteren, N. Y., Moll, A. C., Imhof, S. M., Meijer, G. A., Baak, J. P., and van der Valk, P. (2003). Comparative genomic hybridisation divides retinoblastomas into a high and a low level chromosomal instability group. J Clin Pathol 56, 26-30.

Vance, K. W., Carreira, S., Brosch, G., and Goding, C. R. (2005). Tbx2 is overexpressed and plays an important role in maintaining proliferation and suppression of senescence in melanomas. Cancer Res 65, 2260-2268.

79

Vassilev, L. T., Vu, B. T., Graves, B., Carvajal, D., Podlaski, F., Filipovic, Z., Kong, N., Kammlott, U., Lukacs, C., Klein, C., et al. (2004). In vivo activation of the p53 pathway by small-molecule antagonists of MDM2. Science 303, 844-848.

Veltman, J. A., Fridlyand, J., Pejavar, S., Olshen, A. B., Korkola, J. E., DeVries, S., Carroll, P., Kuo, W. L., Pinkel, D., Albertson, D., et al. (2003). Array-based comparative genomic hybridization for genome-wide screening of DNA copy number in bladder tumors. Cancer Res 63, 2872-2880.

Vogelstein, B., and Kinzler, K. W. (1993). The multistep nature of cancer. Trends Genet 9, 138-141.

Voncken, J. W., Roelen, B. A., Roefs, M., de Vries, S., Verhoeven, E., Marino, S., Deschamps, J., and van Lohuizen, M. (2003). Rnf2 (Ring1b) deficiency causes gastrulation arrest and cell cycle inhibition. Proc Natl Acad Sci U S A 100, 2468-2473.

Vonlanthen, S., Heighway, J., Altermatt, H. J., Gugger, M., Kappeler, A., Borner, M. M., van Lohuizen, M., and Betticher, D. C. (2001). The bmi-1 oncoprotein is differentially expressed in non-small cell lung cancer and correlates with INK4A-ARF locus expression. Br J Cancer 84, 1372-1376.

Waldmann, T., Scholten, I., Kappes, F., Hu, H. G., and Knippers, R. (2004). The DEK protein--an abundant and ubiquitous constituent of mammalian chromatin. Gene 343, 1-9.

Weber, J. D., Jeffers, J. R., Rehg, J. E., Randle, D. H., Lozano, G., Roussel, M. F., Sherr, C. J., and Zambetti, G. P. (2000). p53-independent functions of the p19(ARF) tumor suppressor. Genes Dev 14, 2358-2365.

Weitzman, J. B., Fiette, L., Matsuo, K., and Yaniv, M. (2000). JunD protects cells from p53-dependent senescence and apoptosis. Mol Cell 6, 1109-1119.

Whibley, C., Pharoah, P. D., and Hollstein, M. (2009). p53 polymorphisms: cancer implications. Nat Rev Cancer 9, 95-107.

Woods, Y. L., Xirodimas, D. P., Prescott, A. R., Sparks, A., Lane, D. P., and Saville, M. K. (2004). p14 Arf promotes small ubiquitin-like modifier conjugation of Werners helicase. J Biol Chem 279, 50157-50166.

Wu, G. S., Burns, T. F., McDonald, E. R., 3rd, Jiang, W., Meng, R., Krantz, I. D., Kao, G., Gan, D. D., Zhou, J. Y., Muschel, R., et al. (1997). KILLER/DR5 is a DNA damage-inducible p53-regulated death receptor gene. Nat Genet 17, 141-143.

Xirodimas, D. P., Chisholm, J., Desterro, J. M., Lane, D. P., and Hay, R. T. (2002). P14ARF promotes accumulation of SUMO-1 conjugated (H)Mdm2. FEBS Lett 528, 207-211.

Yarosh, W., Barrientos, T., Esmailpour, T., Lin, L., Carpenter, P. M., Osann, K., Anton-Culver, H., and Huang, T. (2008). TBX3 is overexpressed in breast cancer and represses p14 ARF by interacting with histone deacetylases. Cancer Res 68, 693-699.

80

Yu, Y., Maggi, L. B., Jr., Brady, S. N., Apicelli, A. J., Dai, M. S., Lu, H., and Weber, J. D. (2006). Nucleophosmin is essential for ribosomal protein L5 nuclear export. Mol Cell Biol 26, 3798-3809.

Zhang, J., Schweers, B., and Dyer, M. A. (2004). The first knockout mouse model of retinoblastoma. Cell Cycle 3, 952-959.

Zielinski, B., Gratias, S., Toedt, G., Mendrzyk, F., Stange, D. E., Radlwimmer, B., Lohmann, D. R., and Lichter, P. (2005). Detection of chromosomal imbalances in retinoblastoma by matrix-based comparative genomic hybridization. Genes Chromosomes Cancer 43, 294-301.

Zindy, F., Eischen, C. M., Randle, D. H., Kamijo, T., Cleveland, J. L., Sherr, C. J., and Roussel, M. F. (1998). Myc signaling via the ARF tumor suppressor regulates p53-dependent apoptosis and immortalization. Genes Dev 12, 2424-2433.