Regulation of E2F1 in Keratinocytes During UV-Damage and Differentiation

Western University Scholarship@Western

Electronic Thesis and Dissertation Repository

11-10-2016 12:00 AM

Randeep K. Singh The University of Western Ontario

Supervisor Dr. Lina Dagnino The University of Western Ontario

Graduate Program in Pharmacology and Toxicology A thesis submitted in partial fulfillment of the equirr ements for the degree in Doctor of Philosophy © Randeep K. Singh 2016

Follow this and additional works at: https://ir.lib.uwo.ca/etd

Part of the Cancer Biology Commons

Recommended Citation Singh, Randeep K., "Regulation of E2F1 in Keratinocytes During UV-Damage and Differentiation" (2016). Electronic Thesis and Dissertation Repository. 4202. https://ir.lib.uwo.ca/etd/4202

This Dissertation/Thesis is brought to you for free and open access by Scholarship@Western. It has been accepted for inclusion in Electronic Thesis and Dissertation Repository by an authorized administrator of Scholarship@Western. For more information, please contact [email protected].

Abstract

The E2F1 transcription factor regulates the expression of key genes involved in cell

proliferation and differentiation to maintain skin homeostasis. The expression of E2F1 is

tightly regulated during cell cycle progression and when cells are committed to

differentiate, as well as in response to DNA damage. In keratinocytes, E2F1 protein and

transcript levels increase following UV-induced DNA damage, whereas, in response to

Ca2+-induced differentiation, both E2F1 protein and transcript levels decrease. In this

thesis, I examined in detail the mechanism that modulates E2F1 stability following DNA

damage and during keratinocyte differentiation. I show that E2F1 associates with hHR23

and together these proteins participate in the nucleotide excision repair (NER) pathway.

hHR23A stabilizes E2F1 by inhibiting its proteasomal degradation. Moreover, E2F1

regulates hHR23A subcellular localization and recruits hHR23A to sites of DNA

photodamage to facilitate DNA repair. These results unveil a novel mechanism involving

E2F1 and hHR23A to enhance the repair of damaged DNA, contributing to the

maintenance of genomic stability following UV-induced DNA damage in epidermal

cells. In contrast to DNA damage, proper keratinocyte differentiation occurs when E2F1

is down regulated. Although the regulation of E2F1 that leads to its nuclear export and

degradation has been reported, the exact mechanisms are poorly characterized. When

E2F1 Ser403 and/or Thr433 are substituted to an alanine (E2F1 ST/A), a significant

increase in the half-life of this mutant in differentiated keratinocytes is observed.

Furthermore, the E2F1 ST/A mutant does not form appreciable levels of K11- or K48- linked polyubiquitylation, which is associated with degradation. Moreover, I show that

Cdh1, an activator protein of the anaphase-promoting complex/cyclosome (APC/C) complex, interacts with E2F1 in keratinocytes, and its inactivation prevents the turnover of E2F1 following Ca2+-induced differentiation. Collectively, these results reveal a new mechanism involving E2F1 and Cdh1 in coordinating its degradation through post- translational modification and nuclear export in differentiated keratinocytes. By elucidating the appropriate events required for the turnover of E2F1, we can better understand and treat various epidermal diseases such as non-melanoma skin carcinoma.

Keywords

Keratinocytes, epidermis, E2F, phosphorylation, ubiquitylation, hHR23, DNA photodamage, ubiquitin ligase, Cdh1, UBC13.

iii

Co-Authorship Statement

This thesis contains material from previously published and submitted manuscript co- authored by Randeep K. Singh and Lina Dagnino. All the experimental work presented in this thesis was performed by Randeep K. Singh. All the figures and quantifications presented in this thesis were generated by Randeep K. Singh. Part of this thesis was written by Randeep K. Singh and Lina Dagnino. A copyright release for the manuscript is shown in Appendix 3.

Acknowledgments

Throughout my entire graduate career, there are numerous people I would like to thank.

Without anyone of them, my experience at Western would not have been as rewarding

and memorable. First and foremost, I would like to thank my supervisor, my mentor, Dr.

Lina Dagnino, without whom this thesis would have been impossible. You have taught

me how to solve many puzzles and what is required to become a real scientist. With your

guidance and advice, my presentation and writing skills have improved tremendously.

Thank you for your continuous encouragement, patience, kindness, and support over the

past eight years. You have given me a great environment to thrive at. I would also like to

thank my advisory committee members: Dr. Nathalie Bérubé, Dr. Rommel Tirona, Dr.

Sean Cregan and Dr. Patricia Watson. The wealth of your guidance has been

immeasurable.

I would also like to take this opportunity to thank all the administrative staff members in the Department of Physiology and Pharmacology, including Susan McMillan for sending out committee meeting reminders and making sure I have completed all course

requirements, Bruce Arppe for knowing where everything is around the department, and

Chris Webb for delivering all our lab reagents and materials, especially when it is

something I have been waiting for weeks. Thank you all for being so nice to me over the

past years. I will miss you all so very much.

I would also like to thank both past and present Dagnino lab members. They are not

merely lab members but have become part of my second family. Throughout the eight

years, many have left and moved on to better things but I will never forget the great moments we shared in the lab, tea times in the office, lunch at the cafeteria or UCC concrete beach over the summer and of course our annual summer beach day at The

Pinery. First, I would like to thank Tames Irvine who has welcomed me into the lab when

I first joined. He has been a great friend. I have learned so much from you, from new techniques in the lab to teaching me real life experiences. I will always remember your quote, "There's always some truth behind a joke". I would like to thank Dr. Linda Vi who taught me to become a better scientist. Your love for research is so contagious, and because of your enthusiasm I enjoyed research and being in the lab. I will always remember our long talks from troubleshooting why our western blots did not work to coming up with new and exciting experiments to do. I would like to thank Stellar Boo for always being there for me when I needed someone to talk to the most. You are the sassy queen, though I took your crown when you left. I would like to thank Dr. Samar

Sayedyahossein with whom I share the deepest thoughts on life. The amount of knowledge you have never ceased to amaze me. I would also like to thank Dr. Dany

Ivanova, Dr. Kerry-Ann Nakrieko, Dr. Alena Rudkouskava Dr. Lylia Nini, Dr. Ernest Ho and Kevin Barr - the wise and mature ones in the lab, thank you for sharing your knowledge, experiences as a graduate student and what life entails after grad school.

Thank you Dr. David Judah for making the lab an enjoyable place to work. I will always remember your handmade miniature sculptures, rest in peace. I would like to thank

Meera Karajgikar, the lab mom you have been so great to me. Thanks for purifying DNA for me, feeding me, giving me amazing Indian food and a place to stay. I will never forget your kindness. I would like to thank Michelle Im who is like my gangster sister. I

thoroughly enjoyed the positivity you bring into the lab with your smile. You made the

lab livelier, especially during the days (or weeks) when none of my experiments worked.

And thank you Bradley Jackson for being the gentleman in the lab. I would also like to

thank the younglings of the lab, Maria Doubova, Shinthujah Arulanantham, Valerie

Leclerc, Melissa Crawford and Hannah Murphy-Marshman, you guys bring so much energy in the lab and I feel young at times, though there are moments when you guys remind me of how old I am. I would also like to thank Dr. John Di Guglielmo and his lab members for lending me reagents and antibodies when our lab forgets to order on time.

Special thanks to Eddie Chan for tolerating my rants, especially toward the end. I will always remember the DotBlot idea you gave me.

This thesis would not have been completed without the support and love from my family.

Thank you my brother, KoKo and my sister, Kuldeep for listening to my graduate life stories. Thank you Mum and Dad, words cannot express my gratitude towards everything you have done for me. I hope I have made you both very proud. And I am saving the best for last, thank you my husband, my love, my partner in crime, Harinder Singh Dhatt.

Thank you for spending numerous hours on editing my thesis and my presentation.

Without your support, I do not think I would have been able to overcome all the challenges. You have always reassured me that everything will work out in the end and it always does. You have made me so happy beyond words can describe. I LOVE YOU.

vii

Table of Contents

Abstract ...... ii

Co-Authorship Statement...... iv

Acknowledgments...... v

Table of Contents ...... viii

List of Tables ...... xii

List of Figures ...... xiii

List of Appendices ...... xvi

List of Abbreviations ...... xvii

Chapter 1 ...... 1

1 Introduction ...... 1

1.1 The Skin ...... 1

1.1.1 The epidermis...... 4

1.1.2 The dermis and hypodermis ...... 8

1.2 Regulation of keratinocyte proliferation and differentiation ...... 9

1.2.1 Keratinocyte stem cells ...... 9

1.2.2 Regulation of keratinocyte differentiation ...... 11

1.3 Keratinocyte responses to UV radiation ...... 13

1.3.1 Nucleotide excision repair ...... 14

1.4 The Ubiquitin-proteasome system ...... 18

1.4.1 Ubiquitylation ...... 18

1.5 hHR23 ...... 19

1.6 Cell Cycle Regulation ...... 26

1.6.1 Retinoblastoma family proteins ...... 29

viii

1.6.2 The E2F family of transcription factors ...... 30

1.7 Other functions of E2F factors ...... 37

1.7.1 Role of E2F in apoptosis ...... 37

1.7.2 Role of E2F in DNA damage ...... 38

1.8 Mechanisms of E2F1 Regulation ...... 40

1.8.1 Post-translational modifications...... 40

1.8.2 E2F1 subcellular localization ...... 46

1.8.3 Protein-protein interactions ...... 46

1.9 The role of E2F in the epidermis ...... 47

1.10 Rationale and Aims of this Study ...... 50

Chapter 2 ...... 52

2 Materials and Methods ...... 52

2.1 Reagents and materials ...... 52

2.2 Antibodies, enzymes and kits ...... 57

2.3 Vectors ...... 61

2.4 Molecular cloning ...... 63

2.4.1 Generation of vectors encoding E2F1 and hHR23A mutants ...... 63

2.4.2 Polymerase chain reaction (PCR) ...... 69

2.4.3 Isolation and purification of DNA and PCR products ...... 69

2.4.4 DNA ligation ...... 69

2.4.5 Preparation of competent DH5α and BL-21 (DE) cells ...... 70

2.4.6 Bacterial transformation...... 70

2.5 Recombinant Protein Purification ...... 71

2.5.1 GST fusion protein expression...... 71

2.5.2 TRX fusion protein expression ...... 72

2.6 Preparation of glass coverslips for cultured cells ...... 73

2.7 Cell culture and transfection ...... 73

2.7.1 Isolation of primary epidermal keratinocytes ...... 73

2.7.2 Culture of tumour cell lines ...... 75

2.7.3 Cell transfection and drug treatments ...... 75

2.8 RNA isolation and quantitative reverse-transcription PCR (qPCR) ...... 76

2.9 Immunoblot analysis and immunoprecipitation...... 77

2.10 Indirect immunofluorescence microscopy ...... 78

2.11 In vitro binding assays ...... 79

2.12 UV damage and repair assays ...... 80

2.13 Cdh1 silencing by RNA interference ...... 81

2.14 Statistical analyses ...... 82

Chapter 3 ...... 83

3 Results ...... 83

3.1 Interaction of hHR23A and E2F1 ...... 83

3.2 Modulation of hHR23A subcellular localization by E2F1 ...... 89

3.3 hHR23A domains involved in binding to E2F1 ...... 94

3.4 Inhibition of E2F1 degradation by hHR23 proteins ...... 101

3.5 Modulation of DNA damage repair by hHR23A and E2F1 ...... 109

3.6 Amino acid residues in E2F1 that contributes to its stabilization following drug-induced DNA damage ...... 118

3.7 Identification of E2F1 amino acid residues required for nuclear export and stabilization in differentiated keratinocytes ...... 119

3.8 Nuclear export is not the only mechanism required for E2F1 ubiquitylation and degradation following keratinocyte differentiation ...... 128

3.9 Role of E2F1 pseudo-phosphorylation at Ser403 and Thr433...... 136

3.10 Differentiation-specific role of Ser403 and Thr433 in E2F1 degradation ...... 141

3.11 K11- and K48-linkages on E2F1 promotes its degradation in differentiated keratinocytes ...... 145

3.12 Cdh1 regulates E2F1 degradation in keratinocytes ...... 146

3.13 E2 conjugating enzyme preferentially interacts with ST/A-E2F1 in differentiated keratinocytes ...... 161

Chapter 4 ...... 164

4 Discussion ...... 164

4.1 E2F1 regulation by hHR23 proteins in keratinocytes ...... 165

4.2 E2F1 induction following genotoxic stress in keratinocytes ...... 171

4.3 Regulation of E2F1 turnover in differentiated keratinocytes ...... 172

4.4 Future studies ...... 179

4.5 Concluding remarks ...... 184

References ...... 185

Appendices ...... 208

Curriculum Vitae ...... 213

List of Tables

Table 2.1: List of materials and catalogue numbers...... 52

Table 2.2: Reagents, the catalogue number, and the final concentration used...... 53

Table 2.3: Composition of routinely used solutions and media...... 56

Table 2.4: Antibodies and dilutions used...... 58

Table 2.5: Enzymes for molecular biology experiments...... 60

Table 2.6: List of kits used in this study...... 60

Table 2.7: Plasmids used in this study...... 61

Table 2.8: Forward (Fwd) and reverse (Rev) primers used to generate E2F1 point mutants...... 67

Table 2.9: Forward (Fwd) and reverse (Rev) primers used for deletion mutations in hHR23A...... 68

xii

List of Figures

Figure 1.1: The skin structure...... 1

Figure 1.2: The epidermal layers...... 5

Figure 1.3: Nuclear excision repair pathway...... 16

Figure 1.4: Schematic of ubiquitin and types of ubiquitylation formed...... 20

Figure 1.5: Model of HR23 regulation in proteasome degradation of proteins...... 24

Figure 1.6: Cell cycle regulation by CDK-cyclin complexes...... 27

Figure 1.7: The mammalian E2F family of a transcription factor...... 31

Figure 1.8: Cell cycle regulation of E2F family of transcription factors...... 34

Figure 1.9: Post-translational modifications on E2F1...... 41

Figure 1.10: Regulation and signaling pathway of E2F1 in differentiated keratinocytes. 48

Figure 2.1: Schematic representation of the three-step PCR protocol used for site-directed E2F1 mutagenesis...... 64

Figure 3.1: Expression of mHR23 and hHR23 proteins...... 84

Figure 3.2: Association of hHR23A and E2F1...... 87

Figure 3.3: Modulation of hHR23A subcellular distribution by E2F1...... 90

Figure 3.4: hHR23A domains involved in E2F1 binding...... 95

Figure 3.5: Regions in E2F1 involved in binding to hHR23A...... 99

Figure 3.6: Modulation of ubiquitylated E2F1 degradation by hHR23 proteins...... 102

xiii

Figure 3.7: Modulation of E2F1 t½ by hHR23 proteins...... 106

Figure 3.8: Effect of UVB radiation on E2F1 levels in keratinocytes...... 110

Figure 3.9: Co-localization of E2F1 and hHR23 at DNA damage sites following UV radiation...... 113

Figure 3.10: Increased repair of UV-induced DNA damage by E2F1 and hHR23A. .... 115

Figure 3.11: Changes in E2F1 levels in response to etoposide...... 120

Figure 3.12: Ser403 and Thr433 are required for E2F1 nuclear export during keratinocyte differentiation...... 123

Figure 3.13: Ser403 and Thr433 mediate E2F1 degradation during keratinocyte differentiation...... 126

Figure 3.14: E2F1 is polyubiquitylated in the nucleus and in the cytoplasm when keratinocytes are induced to differentiate with Ca2+...... 129

Figure 3.15: Ubiquitylation of E2F1 is not affected by mutation of E2F1 at Ser403 and Thr433 in keratinocytes...... 132

Figure 3.16: Effect of constitutive nuclear export on E2F1 abundance in differentiating keratinocytes...... 134

Figure 3.17: Subcellular localization of pseudophosphorylated E2F1 mutant proteins in keratinocytes...... 137

Figure 3.18: Ser403 and Thr433 are critical contributors to E2F1 phosphorylation...... 139

Figure 3.19: Ser403 and Thr433 regulated E2F1 protein half-life in differentiated keratinocytes...... 142

Figure 3.20: Ser403 and Thr433 regulate K11- and K48-linked ubiquitylation of E2F1 in differentiated keratinocytes...... 147

xiv

Figure 3.21: Cdh1 and not Skp2 reduce the abundance of E2F1 protein in keratinocytes...... 152

Figure 3.22: Inhibition of Cdh1 activity with hEmi1 increases E2F1 protein levels. .... 154

Figure 3.23: Cdh1 promotes E2F1 degradation in differentiating keratinocytes...... 157

Figure 3.24: E2F1 interacts with Cdh1 in keratinocytes...... 159

Figure 3.25: E2F1 interaction with E2 UBC13 conjugating enzyme is modulated by Ser403 and Thr433 in differentiated keratinocytes...... 162

Figure 4.1: Proposed model for hHR23A and E2F1 regulation of DNA repair in undifferentiated keratinocytes ...... 166

Figure 4.2: Proposed model of E2F1 phosphorylation-linked ubiquitylation modulated by ubiquitin ligases and deubiquitylating enzymes...... 176

Figure 4.3: Proposed model of E2F1 regulation and function in differentiated keratinocytes...... 180

List of Appendices

Appendix 1: Primary epidermal keratinocytes irradiated with UV...... 209

Appendix 2: Ubiquitylation of E2F1 proteins in keratinocytes...... 211

Appendix 3: Copyright permission...... 212

xvi

List of Abbreviations

a.a Amino acid

ANOVA Analysis of variance

Ap1 Activator protein 1

Apaf1 Apoptotic protease-activating factor 1

APC/C Anaphase-promoting complex/cyclosome

Aspp Apoptosis-stimulating of p53 protein

ATM Ataxia-telangiectasia mutated

ATR Ataxia telangiectasia and Rad3-related

Bax BCL2 Associated X, Apoptosis Regulator

BCL2 B-Cell CLL/Lymphoma 2

BER Base excision repair

BMP Bone morphogenetic proteins bp Base pair

BRCA1 Breast cancer type 1 susceptibility protein

BRD4 Bromodomain-containing protein 4

BRG1 Brahma-related gene 1

BRM Brahma

BSA Bovine Serum Albumin

C/EBP Transcription factor CCAAT/enhancer-binding protein

Ca2+ Calcium ions

CAK CDK-activating kinase

CBP p300/CREB-binding protein

xvii

CBP p300/CREB-binding protein

CD34 Hematopoietic progenitor cell antigen CD34

CD71 Transferrin receptor

Cdc Cell division cycle

Cdh1 Cdc20 homolog 1

CDK Cyclin dependent kinase

CETN2 Centrin-2

CHD8 Chromodomain-helicase-DNA-binding protein 8

ChIP Chromatin immunoprecipitation

Chk Checkpoint kinase

CHX Cycloheximide

CIP/KIP CDK interacting protein/Kinase inhibitory protein

CKI Cyclin kinase inhibitor

CPD Cyclobutane pyrimidine dimer

CRL RING E3 ligas

CRM1 Chromosome region maintenance 1

CSB Cockayne syndrome protein

CtBP C-terminus-binding protein

C-terminal Carboxyl-terminal

CtIP CtBP interacting protein

CUL Cullin

CUL4A-ROC1 Cullin 4A–regulator of cullins 1

DAG Diacylglycerol

xviii

D-box Destruction box

DDB DNA damage-binding protein

DEME-RS Dulbecco’s Modified Eagle Medium-Reduced Serum

Dkk1 Dickkopf-related protein 1

DMSO Dimethyl sulfoxide

DP Dimerization domain

DUB Deubiquitylating enzyme

EDTA Ethylenediaminetetraacetic acid

EMEM Eagle’s minimum essential medium

Emil1 Early mitotic inhibitor 1

ERCC1 Excision repair cross-complementation group 1

ERK Extracellular signal–regulated kinase

ESCRT Endosomal sorting complex required for transport

Fas Tumor necrosis factor receptor superfamily member 6

FBS Fetal bovine serum

Fen1 Flag structure-specific endonuclease

FGF Fibroblast growth factor

FZR1 Fizzy/cell division cycle 20 related 1

GADD45A DNA-damage-inducible alpha

GAPDH Glyceraldehyde 3-phosphodehydrogenase

GFP Green fluorescent protein

GG-NER Global genome NER

GST Glutathione

xix

HA Hemagglutinin

HCF1 Host cell factor 1

HDAC Histone deacetylases

HEK Human epidermal keratinocytes

Hes1 Transcription factor HES-1

His Histidine

IB Immunoblot

IF Immunofluorescence

IL-1β Interleukin-1β

IP Immunoprecipitation

IP3 Inositol 1,4,5-trisphosphate

IPTG Isopropyl β-D-1-thiogalactopyranoside

Jab1 c-Jun activation domain-binding protein 1

Jmy Junction-mediating and -regulatory protein

K1-15 Keratin 1-15

KCS Keratinocyte stem cells kDa Kilo Dalton

KDM1A Lysine-specific histone demethylase 1A

LB Luria Bertani

Lgr5 Leucine-rich repeat-containing G-protein coupled receptor 5

LMB Leptomycin B

MAP Mitogen-activated protein

Mdm2 Mouse double minute 2 homolog

MEF Mouse embryonic fibroblasts

MEK Mouse epidermal keratinocytes

MEK3 MAP kinase kinase 3

MEKK MAP kinase kinase kinase

MSH DNA mismatch repair protein

MTPBS Mouse tonicity phosphate-buffered saline

Myt1 Membrane-associated tyrosine- and threonine-specific cdc2- inhibitory kinase

NEM N-ethylmaleimide

NER Nucleotide excision repair

NES Nuclear export sequence

NLS Nuclear localization sequence

Notch 1/2 Neurogenic locus notch homolog protein 1/2

NPDC1 Neural proliferation differentiation and control protein 1

NTA Nitrilotriacetic acid

N-terminal Amino-terminal

P/CAF p300/CBP-associated factor

p63 Tumor protein 63

PARP1 Poly [ADP-ribose] polymerase 1

PBS Phosphate Buffered Saline

PCNA Proliferating cell nuclear antigen

PCR Polymerase chain reaction

xxi

PDA4 Protein arginine deiminases 4

PEI Polyethylenimine

PFA Paraformaldehyde

PI3K Phosphatidylinositol-4,5-bisphosphate 3-kinase

PIP2 Phosphatidylinositol 4,5-bisphosphate

PKC Protein kinase C

PLC Phospholipase C

PLL Poly-L-lysine

PMS2 Mismatch repair endonuclease

PMSF Phenylmethylsulfonyl- fluoride

PRMT Protein arginine methyltransferase

PSMD14 26S proteasome non-ATPase regulatory subunit 14

PSMD4 26S proteasome non-ATPase regulatory subunit 4

PVDF Polyvinylidene difluoride membranes

qPCR Quantitative polymerase chain reaction

R1 Restriction point

Rb Retinoblastoma

RBP-Jκ Recombination signal binding protein for immunoglobulin kappa J region

RIPA Modified radioimmunoprecipitation assay

ROS Reactive oxygen species

SCF Skp1-cullin-F-box

SD Standard deviation

SDS Sodium dodecyl sulfate

xxii

SEM Standard error of mean

SETD7 Histone-lysine N-methyltransferase

siRNA Small (or short) interfering RNA

Skp2 S-phase kinase-associated protein 2

SOB Super Optimal Broth

Sox9 Transcription factor SOX-9

Sp1 Specificity protein 1

SUMO-1 Small ubiquitin-like modifier 1

Suv39H1 Suppressor of variegation 3-9 homolog 1

SV40 Simian virus 40

SWI/SNF SWItch/Sucrose Non-Fermentable

t½ Half-life

TAC Transit amplifying cells

TAE Tris-acetate-EDTA

TAp73 Transcriptional domain-containing tumor protein 73

TB Transformation buffer

TBS Tris-Buffered Saline

TC-NER Transcription-coupled NER

TFIIH Transcription initiation factor IIH

TNFα Tumor necrosis factor–α

TopBP1 Topoisomerase IIβ-binding protein I

TPp53inp Tumor protein p53-inducible nuclear protein

TRAF2 TNF receptor-associated factor 2

xxiii

TRIM28 Transcription intermediary factor 1-β

TRX Thioredoxin

UBA Ubiquitin-associated

UBC13 Ubiquitin-conjugating enzyme 13

UBD Ubiquitin-binding domain

UbL Ubiquitin-like

UCHL5 Ubiquitin carboxyl-terminal hydrolase isozyme L5

UNG Uracil-DNA glycosylase

UPS Ubiquitin-proteasome system

USP37 Ubiquitin-specific protease 37

UV Ultraviolet

Wnt Wingless-type MMTV integration site family member

WT Wild type

XP Xeroderma pigmentosum

XPC XP complementation group C

ZBRK Zinc finger and BRCA1-interacting protein with a KRAB domain

xxiv 1

Chapter 1 1 Introduction 1.1 The Skin

The skin serves as the first line of defense against environmental and physical insults, and

excessive loss of water and electrolytes. A major role of the skin is to provide a barrier

protection between the organism and the external environment. Other functions of the

skin include maintenance of body temperature, synthesis of vitamin D, and tactile

sensation. Extensive damage to the skin can have substantial negative consequences,

including increased risk of infection and water-electrolyte imbalance. Therefore, proper maintenance of skin structure and function is crucial for keeping our body healthy.

The skin is composed of three main layers: the hypodermis, the dermis and the epidermis

(Figure 1.1). The skin also contains epidermal appendages including sweat glands,

sebaceous glands, apocrine glands, and hair follicles. Signaling events can lead to the

activation of differentiation processes, giving rise to the stratified epidermis, hair follicles

and sebaceous glands (Fuchs, 2007). The epidermis is derived from the embryonic

ectoderm and the dermis and hypodermis are mesodermal in origin (Hardy, 1992). The

epidermis and the dermis are attached to each other through a basement membrane.

Epidermal morphogenesis is regulated through many signaling pathways, including the

canonical wingless-related (Wnt), bone morphogenetic proteins (BMP), and Notch

signaling pathways (Fuchs, 2007). Wnt signaling events promote the activation of BMP

signaling in early ectoderm progenitor cells to initiate the progression towards epidermal

Figure 1.1: The skin structure.

The three distinct layers of the skin are shown: epidermis (orange), dermis (pink) and hypodermis (yellow). At the dermal-epidermal junction, a basement membrane lies between the epidermis and dermis. Dermal papillae are finger-like projections of dermis that project up to the epidermis, increasing the surface area. Nerve endings and blood vessels present in the dermis are also shown. Below the dermis is the hypodermis, composed mainly of adipose tissue. The skin also contains epithelial appendages structures including hair follicles, sweat glands, and sebaceous glands. The hair follicle bulge where epithelial stem cells reside is also shown.

morphogenesis. In chick embryos, expression of Wnt signaling represses fibroblast

growth factor (FGF) signals, allowing for the activation of BMP signaling, inducing

neural cell fate to switch into epidermal cell fate (Wilson et al., 2001). The Wnt signaling

pathway also plays an essential role in the initiation of epidermal appendages, such as

hair follicles (Fuchs, 2007). Constitutive expression of the Wnt inhibitor, Dickkopf- related protein 1 (Dkk1) prevents the formation of hair follicles in the epidermis (Andl et al., 2002). Finally, the activation of Notch signaling events promote stratification of the epidermis (Wilson and Radtke, 2006).

1.1.1 The epidermis

The epidermis is a keratinized stratified squamous epithelium and is the outermost

protective layer of the skin. The epidermis consists of four cell types: keratinocytes,

derived from surface ectoderm, Langerhans cells derived from the bone marrow,

melanocytes and Merkel cells, both derived from the neural crest (Carlson, 1994).

Keratinocytes constitute about 80% of all epidermal cells, whereas other cell types are

scarcely found throughout the epidermis (Rook and Burns, 2004).

In humans and other mammals, the epidermis is composed of several layers (Figure 1.2).

The lowermost layer, or stratum basale, contains a heterogeneous population of

proliferating and differentiating cells that are characterized into three categories:

keratinocyte stem cells (KSC), transit amplifying cells (TAC), and postmitotic

differentiating cells (Kaur and Li, 2000; Li et al., 1998). KSC divide symmetrically or

asymmetrically (Clayton et al., 2007). During symmetric cell division (delamination), cells expressing low levels of integrins differentiate and move out to the basal epidermal

Figure 1.2: The epidermal layers.

There are several distinct layers of the epidermis and keratinocytes are the dominant cells aligned through the layers. Other cells found in the epidermis include melanocytes, Merkel cells, and Langerhans cells. The stratum basale is the bottom single layer of cells attached to the basement membrane. Epidermal stem cells are located in the basal layer and during mitosis, it gives rise to two daughter cells. One retains the properties of a keratinocyte stem cell (KSC), and the other differentiates into transient amplifying cells (TAC) that migrate to the epidermal surface forming the protective layer of the skin. The stratum spinosum begins to synthesize intermediate filaments called keratins. These intermediate filaments are anchored to the desmosomes joining adjacent cells. In the stratum granulosum, the cells begin to lose their nuclei and the cytoplasm looks granular. The cells in this layer exocytose lamellar granules to generate a waterproof barrier between the outer dead cells and the lower layer of active epidermal cells. The stratum lucidum is a thin transparent layer present in thick skins such as palms and soles. The outermost layer of the epidermis is called the stratum corneum. This layer is made up of hexagonal-shaped, non-nucleated cells known as corneocytes. The corneocytes are filled with keratin fibers and lipids. This structure provides the protective barrier function of the skin.

layer and high integrin expressing cells remain attached to the basement membrane (Frye

et al., 2007). These cells also expresses Numb, an inhibitor of Notch signaling, which is

required for stratification of the epidermis (Clayton et al., 2007). During asymmetric cell

division, a stem cell divides into two daughter cells, giving rise to a new cell that retains

the stem cell capability, and a TAC, which divides rapidly and initiates the process

towards differentiation to give rise to suprabasal layers (Figure 1.2). Both modes of

division contribute to the balance of epidermal self-renewal, differentiation, and

regeneration after injury. However, the mechanisms involved in one versus the other are

poorly understood (Fuchs and Chen, 2013). Over the basal layer, the stratum spinosum

and stratum granulosum are found, in body areas with thicker skin, such as the foot sole,

the stratum lucidum is found above the granular layer in human epidermis. The outermost

layer of the epidermis is the stratum corneum, which protects the inner body from foreign

invasion (Fuchs, 2009; Fuchs and Raghavan, 2002).

The cells in the stratum spinosum are metabolically and transcriptionally active. This

layer is characterized by the formation of desmosomes, which provide strong adhesion

between adjacent cells by linking the outer aspect of two cells to each other, and their

cytoplasmic domains are linked to keratin filaments providing mechanical strength. Early

markers of keratinocyte differentiation, including involucrin, and keratin-1 and -10 (K1,

K10) are expressed in this layer (Ekanayake-Mudiyanselage et al., 1998). The stratum granulosum contains keratinocytes that are enriched in lipid-lamellar granules that function as a water repellent, contributing to the barrier function of the skin. The activation of transglutaminase in this layer irreversibly cross-links lysine- and glutamine- rich proteins to give rise to the cornified envelope (Koster and Roop, 2007; Steinert and

Marekov, 1999). As granular cells progress to form the stratum corneum, metabolic activity is halted and cytoplasmic organelles are lost. The keratinocytes in this layer, also termed corneocytes, exhibit the terminal stage of differentiation of all epidermal cells, and are constantly shed and replaced by new differentiated cells moving outwards (Mack et al., 2005).

Melanocytes are found in the basal layer of the interfollicular epidermis and in the hair follicle (Figure 2.2). Melanocytes primarily function to produce the pigment melanin and help in protecting against the harmful effects of ultraviolet (UV) radiation (Tsatmali et al., 2002). The Merkel cells are also found in the basal layer, and are specialized in the touch sensation (Moll et al., 2005). Langerhans cells are dendritic and are situated in the suprabasal layer. They are involved in immunity, and function in antigen presentation.

These cells are central components of cutaneous allergic reactions (Merad et al., 2008).

1.1.2 The dermis and hypodermis

The dermis sustains and provides support to the epidermis. The dermis is composed of two layers, papillary dermis, and reticular dermis. The elasticity and strength of the dermis are achieved through elastin and collagen fibers, which are secreted by dermal fibroblasts. Fibroblasts, mast cells, macrophages, and adipocytes are present within the dermis and hypodermis. The hypodermis is beneath the dermis, and its main function is for fat storage.

1.2 Regulation of keratinocyte proliferation and differentiation

A balance between epidermal proliferation and differentiation must be maintained to

ensure proper epidermal homeostasis. Defects in epidermal proliferation can lead to thin

and fragile skins (Raghavan et al., 2000). Abnormalities in epidermal differentiation and

disruption of the epidermal barrier function can lead to chronic skin inflammatory

disorders, such as atopic dermatitis, psoriasis, as well as to basal- or squamous-cell

carcinoma. In the following section, I will describe some of the main transcription factors

and mechanisms that regulate keratinocyte proliferation and differentiation.

1.2.1 Keratinocyte stem cells

The epidermis is constantly undergoing self-renewal to replace cells that have terminally differentiated and been shed. Self-renewal and regeneration after injury are possible through the presence of KSC. The stem cells present in the hair follicle bulge are multipotent, as they have the ability to give rise to keratinocytes in the epidermis, hair follicles, and sebaceous glands (Figure 1.1).

KSC express high levels of integrins. Integrins are essential for keratinocyte viability and adhesion to the basement membrane, and non-pathologic loss of adhesion is associated with epidermal differentiation (Carter et al., 1990). The main integrins found in basal

keratinocytes in intact epidermis are α2β1, α3β1 and α6β4 (Peltonen et al., 1989). KSC and

TAC express high levels of β1 and α6 integrins whereas suprabasal postmitotic

differentiated cells do not show detectable integrins (Kaur and Li, 2000; Li et al., 1998).

KSC population can be distinguished from TAC population through the expression of a

cell surface marker, the transferrin receptor (CD71), where the latter expresses high

levels of CD71 (Kaur and Li, 2000; Tani et al., 2000). Therefore, KSC can be enriched in

culture by selecting cells expressing high levels of α6 integrins and low levels of CD71.

Other cell surface markers of the KSC located in the hair follicle bulge include K15

(Morris et al., 2004), CD34 (Trempus et al., 2003), Lgr5 (Jaks et al., 2008) and Sox9

(Vidal et al., 2005).

The transcription factor p63 is preferentially expressed in the basal epidermis and is

essential for normal epidermal morphogenesis (Mills et al., 1999). The p63 protein is necessary for maintaining the epidermal proliferative capacity, commitment to KSC, and keratinocyte differentiation (Koster et al., 2004; Senoo et al., 2007). There are two isoforms of the p63 gene, the transactivating TAp63, and the non-transactivating ΔNp63 generated through the use of alternate promoters (Blanpain and Fuchs, 2007). Ectopic expression of TAp63 results in hyperplasia of keratinocytes, and inactivation of ΔNp63 leads to formation of a single-layered epidermis, which indicate defects in keratinocyte differentiation (Koster et al., 2004; Romano et al., 2012). In the current proposed model, the function of TAp63 is to maintain high proliferative potential during embryogenesis while limiting the differentiation process, and ΔNp63 functions as a dominant-negative protein to inhibit TAp63 to promote differentiation (McKeon, 2004). p63 also functions to prevent the activation of p16INK4A, an inhibitor of the G1-S phase transition of the cell

cycle (Su et al., 2009). Inhibition of p16INK4A allows for CDK4/6-Cyclin D activation,

consequently regulating the E2F/Rb pathway to promote cell proliferation. Thus, p63 is

necessary for epidermal proliferation. In addition to p63 transcriptional repression on

p16IK4A, p63 prevents the activation of Notch-targeted genes including p21Cip1/Waf1,

involucrin and Hes1. Reciprocally, Notch activation leads to suppression of p63

expression in differentiated keratinocytes. Hence, the cross-talk between p63 and Notch signaling is one of the mechanisms that maintain the balance between epidermal cell proliferation and differentiation (Nguyen et al., 2006).

1.2.2 Regulation of keratinocyte differentiation

The Notch signaling pathway is a key regulator of keratinocyte differentiation and is

activated in the suprabasal layer, where it induces p21Cip1/Waf1 expression in an RBP-Jκ- dependent manner (Rangarajan et al., 2001; Watt et al., 2008). Activation of p21Cip1/Waf1

expression leads to keratinocyte growth arrest and initiation of terminal differentiation.

Moreover, Neurogenic locus notch homolog protein 1 (Notch1) and Notch2 induce the expression of the early differentiation markers involucrin and K1 through an RBP-Jκ- independent pathway (Rangarajan et al., 2001). Notch activity is stimulated when cultured keratinocytes are induced to differentiate by increasing the extracellular calcium ion (Ca2+) concentration (Kolly et al., 2005).

A Ca2+ gradient is observed throughout the epidermal layers, with lower Ca2+ levels in

the basal layer and higher concentration in the granular compartment (Mascia et al.,

2012). This Ca2+ gradient plays a crucial role in epidermal homeostasis and

differentiation. Any disruption to the gradient leads to abnormal keratinocyte

differentiation and loss of epidermal barrier function (Elias et al., 2002). Cultured

keratinocytes can be induced to differentiate by modulating the extracellular Ca2+

concentration (Hennings et al., 1980; Yuspa et al., 1989). When keratinocytes are

cultured in low Ca2+ medium (0.02 to 0.09 mM), they are undifferentiated and maintain

their proliferative capacity. These cells model basal keratinocytes. An increase in the

extracellular Ca2+ concentration to >0.1 mM leads to cell cycle arrest and up regulation of

differentiation markers, such as involucrin (Xie et al., 2005). Ca2+ activates a variety of

signaling pathways, including Notch, phospholipase C (PLC), protein kinase C (PKC)

and Phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K) (Denning et al., 1995;

Rangarajan et al., 2001; Xie and Bikle, 1999; Xie et al., 2005).

Ca2+-induced keratinocyte differentiation occurs through the PLC/PKC signaling

pathway. PLCγ activation hydrolyses phosphatidylinositol 4,5-bisphosphate (PIP2) to

diacylglycerol (DAG) and inositol 1,4,5-trisphosphate (IP3). IP3 binds its receptor and

activates the release of intracellular Ca2+. Both Ca2+ and DAG subsequently activate PKC

to initiate differentiation (Mascia et al., 2012). There are nine PKC family members, of

those, the classical PKCα (activated by both Ca2+ and DAG), the novel PKCδ, PKCη, and

PKCε (stimulated by DAG and not Ca2+) and the atypical PKCζ (not activated by Ca2+ or

DAG) are present in keratinocytes (Denning et al., 1995).

PKCδ and PKCη are key for keratinocyte differentiation. The expression levels of PKCδ

and PKCη increase following Ca2+-induced keratinocyte differentiation (Denning et al.,

1995; Deucher et al., 2002; Kashiwagi et al., 2002). The increase in PKCη leads to

transcriptional activation of proteins involved in the formation of the cornified envelope

at the suprabasal layer of the epidermis including transglutaminase I and involucrin

(Efimova and Eckert, 2000; Ueda et al., 1996). The transcriptional activation of

involucrin occurs through the nPKC/Ras/MEKK/MEK3/p38δ-ERK1-2 signaling

pathway, resulting in an increase in transcription factor CCAAT/enhancer-binding

protein (C/EBP), activator protein 1 (AP1) and specificity protein 1 (Sp1) levels and

binding to the involucrin promoter sequence, thereby increasing involucrin gene

transcription (Agarwal et al., 1999; Crish et al., 2006; Efimova et al., 2002; Efimova et

al., 1998). PKCη activates Fyn and p21Cip1/Waf1, which lead to cell cycle arrest followed

by keratinocyte differentiation (Cabodi et al., 2000; Kashiwagi et al., 2000). PKCη

phosphorylates p21Cip1/Waf1 at serine 146 and inhibits CDK2 kinase activity, which leads

to G1 arrest through dephosphorylation of the retinoblastoma protein (pRb) (Kashiwagi

et al., 2000).

1.3 Keratinocyte responses to UV radiation

The epidermis of the skin is constantly exposed to environmental insults and therefore

has established endogenous processes to prevent, protect, and repair damage. There are a

number of signals that sense the presence of damaged DNA, which lead to cell cycle arrest, allowing time for the repair to occur, or force cells to undergo apoptosis if the damage is beyond repair. If the damage is not repaired properly, it will result in loss of

DNA integrity and accumulation of mutations that can lead to cancer.

Solar UV radiation is a major environmental factor that influences and affects the survival of epidermal keratinocytes, which can lead to the development of premature skin aging and skin cancer (Diffey, 1991). Nonmelanoma skin carcinomas are the most

frequent cancers in humans, accounting for about 40% of all diagnosed cancers in North

America (Christenson et al., 2005). UV radiation is composed of UVA (320-420nm),

UVB (280-320 nm), and UVC (200-280 nm). UVC radiation is highly bioactive,

however, most of the rays are absorbed by the atmospheric ozone layer, and therefore, we

are not exposed to significant levels of UVC (unless you live in Antarctica). UVB radiation causes direct DNA damage and reaches the epidermis. UVA radiation penetrates the dermis and increases levels of reactive oxygen species (ROS) that can induce DNA mutations.

1.3.1 Nucleotide excision repair

UV radiation causes specific mutations in epidermal keratinocytes. This occurs through

the reaction between UV photons and cellular DNA (Grossman and Leffell, 1997). In the

presence of UV light, two pyrimidines on the same DNA strand can undergo dimerization

and become covalently bound to each other, forming cyclobutane pyrimidine dimers

(CPD) and (6-4) pyrimidine pyrimidone ((6-4)PP) photoproducts (Sancar et al., 2004).

Approximately 85% of primary lesions caused by UV radiation give rise to CPD

photoproducts and (6-4)PP accounts for 10 to 20% (Grossman and Leffell, 1997). In

mammals, DNA repair of these lesions relies on nucleotide excision repair (NER)

pathways (Thiagarajan et al., 2011).

The importance of the NER pathway is demonstrated by the hereditary disorder

xeroderma pigmentosum (XP), an autosomal recessive genetic disorder that arises from

an impaired ability of epidermal cells to repair damaged DNA. Individuals with XP have

a mutation in one of the several genes involved in the NER pathway, and are

hypersensitive to UV radiation, exhibiting an increased risk of developing nonmelanoma

skin carcinoma (Rezvani et al., 2011). The NER pathway can occur through either global

genome NER (GG-NER), which repairs damage from the entire genome or through

transcription-coupled NER (TC-NER), which generally repairs lesions when RNA

polymerase II is stalled on actively transcribed DNA strands (Figure 1.3) (Marteijn et al.,

2014).

The first step in the NER pathway is recognition of damaged DNA. The damage sensor

protein, XP complementation group C (XPC) forms a complex with UV excision repair

protein Rad23 homolog (hHR23) and centrin 2 (Sugasawa et al., 1997; van der Spek et al., 1996a). When the XPC complex is recruited at the sites of DNA damage, hHR23 dissociates from the complex. In the NER pathway, the hHR23 protein functions to recruit XPC to sites of DNA damage and promote nuclear accumulation of XPC by preventing its proteasomal degradation (Ng et al., 2003). At the damaged site, complex formation also occurs between DNA damage-binding protein-1 (DDB1, also known as the XPE-binding factor) and DDB2, the UV-DDB complex, which further facilitates subsequent binding of XPC (Reardon and Sancar, 2003). Lesion recognition is followed by verification of DNA damage by recruitment of transcription initiation factor IIH

(TFIIH), composed of 10 protein subunits including XPA, XPB and XPD helicases. This complex verifies the presence of the lesion and unwinds approximately 20 base pairs (bp) around the damaged DNA site (Evans et al., 1997). This is followed by recruitment of

XPG, XPF and excision repair cross-complementation group 1 (ERCC1) endonucleases by the XPA subunit to make a′ and3 5′ incisions, respectively and excise 25-26 bp nucleotide strand surrounding the lesion site (Staresincic et al., 2009). In the final step,

DNA gap-filling is completed by the proliferating cell nuclear antigen (PCNA), which recruits DNA polymerases and the repaired DNA is ligated by DNA ligases (Figure 1.3)

(Moser et al., 2007; Ogi et al., 2010).

Figure 1.3: Nuclear excision repair pathway.

The two pathways, global genomic NER, and transcriptional coupled NER differ in the recognition of DNA damage but converge to a common mechanism from verification to DNA synthesis and ligation. In global genomic NER (left), the UV-DDB complex and the cullin 4A-regulator of cullins 1 (CUL4-ROC1) complex mediates XPC ubiquitylation to increase its binding affinity to damaged DNA and recruitment to lesion sites together with HR23 protein. Once XPC and centrin 2 (CETN2) is on lesion site, the UV-DDB complex is ubiquitylated and dissociates together with HR23, which is essential for the recruitment of TFIIH to verify the damaged site. In transcriptional couple NER (right), the stalled RNA polymerase II facilitates the recruitment of Cockayne syndrome protein (CSB) to sites of DNA damage. The recognition is followed by verification of DNA damage by TFIIH in both pathways. The XPC-CETN2 complex is ubiquitylated and targeted for degradation to enable binding of XPF-ERCC1 and XPG endonuclease to excise 20-25 nucleotide of the damaged DNA. In the final step, the gap is filled by DNA polymerase (DNA pol) to synthesize new DNA and the nick is sealed by DNA ligase. Ub, ubiquitylation.

1.4 The Ubiquitin-proteasome system

The ubiquitin-proteasome system (UPS) is a complex biological system that targets

proteins for degradation in the cytosol, the nucleus, or within the endoplasmic reticulum

of eukaryotic cells (Peters et al., 1994). The main function of the UPS is to degrade

misfolded, denatured, or improperly translated proteins, as well as participating in normal

protein turnover. Target proteins are first marked by ubiquitylation and the ubiquitylated

protein is then recognized, unfolded and hydrolyzed by the proteasome.

1.4.1 Ubiquitylation

Ubiquitin is a small 8.5-kDa protein with 76 amino acid (a.a.) residues that functions as a

protein degradation signal (Ciechanover et al., 1980). Ubiquitylation is also an inducible

event activated by extracellular stimuli, phosphorylation or during DNA damage

response (Hershko and Ciechanover, 1998). When a substrate protein is covalently linked

to polyubiquitin chains, it is frequently targeted for degradation through the 26S

proteasome. When the substrate protein in delivered to the proteasome, the ubiquitin is

recycled by deubiquitylating enzymes (DUB).

Ubiquitin is conjugated to a substrate in a three-step process catalyzed by specific E1, E2

and E3 enzymes (Scheffner et al., 1995). The carboxyl-terminal (C-terminal) end of ubiquitin contains two glycine residues (Gly75-Gly76) that are activated by the E1 enzyme in an ATP-dependent manner. This forms a covalent thioester linkage between ubiquitin and a cysteine residue of the E1 ubiquitin-activating enzyme. The activated ubiquitin is subsequently transferred to an E2 ubiquitin-conjugating enzyme, which acts

as a ubiquitin carrier protein. The E2 enzyme then transfers the ubiquitin to an active cysteine residue on the E3 ubiquitin-ligase bound to the substrate protein. Ubiquitin is subsequently transferred to the substrate protein through an amide isopeptide linkage between Gly76 of ubiquitin and the ε-amino group of a lysine residue on the substrate protein (Scheffner et al., 1995). A single ubiquitin molecule can be transferred to a lysine residue on the substrate protein, leading to monoubiquitylation or it can be attached to multiple lysine residues, which leads to multi-monoubiquitylation (Chan et al., 2006;

Hoege et al., 2002). Additional ubiquitins can be added to a lysine residue of ubiquitin, resulting in the formation of polyubiquitin chains (Pickart, 2000). Ubiquitin has seven lysine residues, all capable of forming polyubiquitin chains (Figure 1.4). There are proteins with ubiquitin-binding domains (UBD) that interact with different types of ubiquitin linkages, leading to numerous biological effects (Ravid and Hochstrasser,

2008). For example, K63-linked polyubiquitylation binds UBD-containing endosomal sorting complex required for transport (ESCRT) proteins, and targets a substrate for endosomal trafficking to lysosomes (Erpapazoglou et al., 2012; Nathan et al., 2013). On the other hand, K11- and K48-linked polyubiquitin chains preferentially interact with ubiquitin-binding receptors, such as hHR23, and target a substrate for degradation

(Bremm and Komander, 2011; Elsasser et al., 2004).

1.5 hHR23

The yeast Saccharomyces cerevisiae, Rad23 and its mammalian orthologue, hHR23 are evolutionary conserved scaffold proteins essential for NER and are also involved in

Figure 1.4: Schematic of ubiquitin and types of ubiquitylation formed.

The seven lysine residues on ubiquitin are shown. The K0 ubiquitin has all lysine residues substituted for an arginine, and therefore can only form mono-ubiquitin on one or more sites on proteins (multi-ubiquitylation). The K0 ubiquitin prevents the generation of polyubiquitin chains. The K11, K48, and K63 ubiquitins have all lysine substituted to an arginine expect for the indicated lysine residue. Therefore, K11, K48, or K63 ubiquitin can only form K11-, K48- or K63- linked polyubiquitin chains, respectively. Different ubiquitin linkages lead to different biological consequences, multi-ubiquitylation of substrate leads to endocytosis, K11- or K48-linked polyubiquitylation leads to substrate degradation through the proteasomal pathway, and K63-linked polyubiquitylation leads to DNA repair, endocytosis, or activation of kinase function. K, lysine; R, arginine; Ub, ubiquitylation.

modulating proteasomal degradation of variety of proteins (Dantuma et al., 2009;

Masutani et al., 1994). Depending on the cellular context, hHR23 can bind to and escort polyubiquitylated proteins to the proteasome, increasing their degradation, or,

paradoxically, protect XPC and other proteins from proteasomal degradation (Blount et

al., 2014; Fang et al., 2013; Ng et al., 2003; Zhu et al., 2001).

There are two homologs of the mammalian HR23: HR23A, and HR23B (Renaud et al.,

2011). XPC associates with hHR23B and hHR23A in vivo (Sugasawa et al., 1997; van

der Spek et al., 1996a). mHR23A knockout mice are viable, develop normally and show

no defects in NER, whereas inactivation of the gene that encodes mHR23B severely

impairs the development of mice (Ng et al., 2003; Ng et al., 2002). This indicates mHR23A can only partially compensate for the loss of mHR23B, and mHR23B has additional important functions. Moreover, inactivation of both genes leads to embryonic lethality, suggesting a role in pathways other than NER, most likely linked to the UPS.

Indeed, mHR23B associates with proteins involved in the UPS and knockout of mHR23B affects the proteasome function resulting in decreased cell proliferation (Bergink et al.,

2013). HR23 proteins are composed of an amino-terminal (N-terminal) ubiquitin-like

(UbL) domain, two ubiquitin-associated domains (UBA1 and UBA2), and linking the

UBA1 and UBA2 regions is the XPC domain (Jung et al., 2014). These four domains are connected through flexible linker regions ranging in length from 25 to 80 a.a. (Walters et al., 2003).

The UbL domain of hHR23 consists of 80 a.a. and shares ~23% sequence similarity with ubiquitin (van der Spek et al., 1996b). The UbL domain interacts with the 26S

proteasome non-ATPase regulatory subunit 4 (PSMD4) (Chen and Madura, 2006;

Hiyama et al., 1999). Deletion of the UbL domain of Rad23 leads to increased sensitivity

to UV radiation, which indicates the importance of the interaction between UbL and the proteasome and its function to efficiently repair damaged DNA (Lambertson et al.,

2003). Indeed, the NER recognition protein, XPC is stabilized by hHR23 (Ng et al.,

2003). The interaction between the UbL domain of hHR23 and the proteasome inhibits

the formation of multi-ubiquitin chains leading to increased protein stability (Ortolan et al., 2000).

The UBA1 and UBA2 domains in hHR23 contain 48 a.a. and 44 a.a., respectively. Both

UBA domains can interact with ubiquitin and multi-ubiquitin chains (Bertolaet et al.,

2001; Kang et al., 2006). The interaction between UBA and ubiquitin prevents the assembly and disassembly of multi-ubiquitin chains, and therefore can either inhibit or favor degradation, respectively (Chen et al., 2001; Raasi and Pickart, 2003). UBA1 and

UBA2 domains mediate inhibition of proteolysis and the UbL domain is dispensable for inhibiting substrate proteasomal degradation. The UbL domain can interact with both

UBA domains and PSMD4 (Kang et al., 2007; Lambertson et al., 1999). The interaction

between PSMD4 and UbL changes the conformation of hHR23A, so that it prevents UbL

from interacting with the UBA domain. The UBA domain can then interact with

polyubiquitylated substrates. This tertiary structure mediates the delivery of ubiquitylated

substrates to the proteasome for degradation where hHR23 act as an adaptor protein

linking the proteasome with the substrate (Figure 1.5A). Through biochemical and

structural analysis, many studies have proposed a working model for hHR23 proteins,

where high levels of hHR23 leads to protein stability and lower levels of hHR23 promote

Figure 1.5: Model of HR23 regulation in proteasome degradation of proteins.

A. In the first model, HR23 interacts with ubiquitylated substrates through its UBA domain and the UbL domain mediates its interaction with PSMD4 of the proteasome. This tertiary complex delivers ubiquitylated proteins to the proteasome for degradation. In this model, HR23 functions as a bridging factor to facilitate protein turnover. B. In a second proposed model, HR23 functions as a negative regulator of proteasome degradation. HR23 prevents the assembly of more than four ubiquitins on a substrate, which is crucial for the degradation of ubiquitylated proteins through the proteasome. Secondly, an excess amount of HR23 proteins binds to ubiquitylated proteins and the proteasome separately preventing the formation of a tertiary structure. Ub, ubiquitylation.

degradation of ubiquitylated substrates (Figure 1.5B) (Brignone et al., 2004; Chen and

Madura, 2002; Fang et al., 2013; Glockzin et al., 2003; Kim et al., 2004; Ng et al., 2003).

Following UV-induced DNA damage, hHR23 stabilizes XPC and recruits it to DNA damage site in the presence of DNA lesions (Baron et al., 2012; Kumar et al., 1999; Ng et al., 2003). In humans, cells lacking either of these proteins are sensitive to UV radiation and are at an increased risk of transformation to nonmelanoma skin carcinoma.

Thus, hHR23 and XPC are necessary to prevent loss of DNA integrity and inhibit accumulation of mutations leading to cancer. The signals that sense the damaged DNA will initiate cell cycle checkpoint signaling pathways to allow repair to occur, and constitute cellular surveillance strategies to coordinate DNA repair and cell cycle transitions.

1.6 Cell Cycle Regulation

A somatic cell can move through two phases, the mitosis phase (M) and interphase, which includes three sub-phases: Gap phase 1, (G1), DNA synthesis (S), Gap phase 2,

(G2) (Figure 1.6A). Both intrinsic and extrinsic factors control cell division (van den

Heuvel, 2005). However, when G1 cells pass a restriction point (R1), they are committed to go through the cell cycle. Cells can exit the cell cycle and enter the G0 quiescent phase.

Exposure to growth factors promotes entry into the G1 phase where cells prepare for

DNA replication in S phase. During S phase, the DNA is duplicated before cell division and in G2 phase, the cell ensures proper intracellular signals prior to entering into M phase where cells are divided into two daughter cells (van den Heuvel, 2005). Each of these cycles is tightly regulated by cyclins and cyclin-dependent kinases (CDK) (Figure

Figure 1.6: Cell cycle regulation by CDK-cyclin complexes.

A. The cell cycle is divided into four phases: G1, S, G2, and M phase. Different CDK- cyclin complexes regulate each of these phases. When cells are not prepared to progress

through the cell cycle, cells enters the G0 quiescent phase. The CDK interacting protein/Kinase inhibitory protein (CIP/KIP) family of CDK inhibitors involved in suppressing the activity of CDK are shown. B. CDK-cyclin is regulated through phosphorylation and dephosphorylation by membrane-associated tyrosine- and threonine- specific cdc2-inhibitory kinase (Myt1) and cell division cycle 25 (Cdc25) to form an inactive and active complex, respectively. The CDK-activating kinase (CAK) also phosphorylates CDK to form an active CDK-cyclin complex. P, phosphorylation.

1.6). CDK are serine/threonine protein kinases activated by association with cyclins. The

CDK-cyclin complex is inactivated through association with cyclin kinase inhibitors

(CKI) and phosphorylation of CDK by membrane-associated tyrosine- and threonine- specific cdc2-inhibitory kinase (Myt1). The inhibitor effect is lifted off when CKI are ubiquitylated and targeted for degradation and dephosphorylation of CDK by cell division cycle (Cdc25) phosphatase (Figure 1.6B). In order to ensure proper cell cycle progression, there are multiple cyclins and CDK governing each phase of the cell cycle.

During early G1 phase the CDK4 and CDK6 form a complex with cyclins D1, D2, and

D3. During transition and progression through S phase, CDK2-cyclin E/A are activated

and in G2/M phase, the CDK2/CDK1 and cyclins A/B are involved (Figure 1.6A). One of

the main functions of the CDK-cyclin complex is to phosphorylate the retinoblastoma

(Rb) proteins to activate transcription of E2F-dependent genes required for S phase

transition.

1.6.1 Retinoblastoma family proteins

There are three family members of the Rb proteins: pRb, p107, and p130 (DeCaprio et

al., 1988; Dyson et al., 1989; Hannon et al., 1993; Mayol et al., 1993). Mutations in the

RB gene lead to retinoblastoma (Knudson et al., 1975) and other cancers, including

osteosarcomas, breast cancer, and lung cancer (Ceccarelli et al., 1998; Kelley et al.,

1995; Toguchida et al., 1988). pRb is phosphorylated during late G1 phase of the cell

cycle and regulates G1 to S phase transition (DeCaprio et al., 1988). All three Rb family

members interact with E2F to inhibit its transcriptional activity.

The Rb family of proteins are often referred to as "pocket proteins" owing to their highly conserved carboxyl-domain that interacts with E2F called the pocket domain. The pocket domain is subdivided into the A and B domains separated by a spacer region. The p107 and p130 have two B domains. Many mutations on Rb proteins are found on the pocket domain (Giacinti and Giordano, 2006). The Rb family of proteins also contain an N- terminal and C-terminal domain, sixteen phosphorylation sites, and a highly conserved

LXCXE binding cleft to interact with proteins containing an LXCXE motif, such as histone deacetylases (HDAC) (Dahiya et al., 2000; Rubin, 2013). Mutations on the

LXCXE binding cleft do not prevent Rb proteins to bind E2F family proteins but inhibit

Rb proteins from recruiting repressor complexes to promoters of E2F target genes

(Dahiya et al., 2000). In addition to binding of E2F1 to the pocket domain of pRb, E2F1 binds pRb on a second site located at the end of the C-terminus of pRb (Dick and Dyson,

2003). This interaction is unique to E2F1 and prevents E2F1-induced apoptotic events

(Julian et al., 2008). Following DNA damage, the interaction of E2F1 to the pRb unique site is lost but the interaction with the pocket domain is enhanced (Dick and Dyson, 2003;

Inoue et al., 2007). This complex formation not only represses transcription of genes required for cell cycle progression, but also transcriptionally activates proapoptotic genes

(Inoue et al., 2007).

1.6.2 The E2F family of transcription factors

The E2F family of transcription factors regulates many functions including: proliferation, apoptosis, differentiation, metabolism and DNA repair. In mammals, the E2F family is composed of E2F 1-8 and three dimerization partners (DP), DP1, -2 and -4 (Figure 1.7).

E2F1 through -6 have a DNA binding domain, a leucine zipper, a marked box domain,

Figure 1.7: The mammalian E2F family of a transcription factor.

E2F family of transcription factors are comprised of ten members and three polypeptide DP1, -2, and -4 to form a heterodimer. E2F family members share several homologous regions including the DNA binding domain and a dimerization domain. E2F1 through -3 have a nuclear localization sequence and a CDK-cyclin binding domain in the N- terminus; and a transcriptional activation domain and Rb proteins binding site (specifically interacts with pRb, p107 or p130) at the C-terminus. E2F4 and -5 lack a CDK-cyclin binding sequence but still contain a pocket protein binding domain (preferentially interact with the other Rb family proteins, p107 and p130). E2F4 and -5 are exported out of the nucleus via their nuclear export signal. E2F6 shares only the DNA binding domain and the dimerization domain with the other E2F family members. Both E2F7 and -8 contain two conserved DNA binding domains allowing it to form heterodimer without DP proteins. E2F6 through -8 lack pRb-binding and a transcriptional activation domain. DP1, -2, and -4 proteins lack transcriptional activation domains and pRb-binding regions.

and a dimerization domain which associates with DP proteins to form a functional

heterodimer complex. In E2F1, the marked box domain interacts with c-Jun activation domain-binding protein 1 (Jab1) to enhance E2F1-induced apoptosis and not proliferation

(Hallstrom and Nevins, 2006). E2F1 through -5 have a transactivation domain and an Rb protein-binding region. Rb family proteins can modulate the transcriptional activity of

E2F proteins. E2F1 through -3 associate specifically with pRb, E2F4 preferentially interacts with p107 or p130 and E2F5 binds p130 (Beijersbergen et al., 1994; Hijmans et al., 1995; Ivey-Hoyle et al., 1993; Lees et al., 1993; Shan et al., 1992). The interaction

between E2F and Rb family proteins represses the transcriptional activity of E2F1

through -5 (Figure 1.8). E2F6 does not have a transactivation domain or an Rb-binding

region. E2F1 through -3 are structurally similar and have a cyclin A-binding domain, which is not present on E2F4 and -5. E2F7 and -8 have two DNA binding domains but lack a dimerization domain. Thus, they are not dependent on DP proteins for functional activity. E2F7 and -8 can interact with each other to form homodimer complexes (Li et al., 2008).

Binding of E2F family of proteins to hypophosphorylated Rb family proteins recruits histone deacetylases and DNA methyltransferases, such as HDAC1, suppressor of variegation 3-9 homolog 1 (Suv39H1) and SWItch/Sucrose Non-Fermentable (SWI/SNF)

(Figure 1.8). This complex functions to inhibit transcription of E2F-responsive genes

(Magnaghi-Jaulin et al., 1998; Nagl et al., 2007; Vandel et al., 2001). In an actively proliferating cell, Rb family proteins are sequentially phosphorylated by CDK4/6-cyclin

D and CDK2-cyclin E complexes resulting in dissociation from E2F, rendering E2F activators transcriptionally active to target genes required for G1/S phase progression

Figure 1.8: Cell cycle regulation of E2F family of transcription factors.

The activity of different E2F-DP-Rb protein complexes during each phase of the cell

cycle are shown. E2F1 through -3 function as activators during late G1 to initiation S

phase progression and G2 phase to promote activation of genes required for mitosis. E2F6 through -8 are involved in repressing E2F-target genes during S phase. E2F1 through -5

are involved in repressing E2F-dependent transcription during cell cycle arrest (G0

phase). pRb is preferentially phosphorylated by CDK4/6-cyclin D during late G1 phase which leads to subsequent phosphorylation at multiple sites on pRb by CDK2-Cyclin E causing full inactivation of pRb and dissociation with E2F1 through -3 to promote G1 to S phase transition. P, phosphorylation.

(Ezhevsky et al., 1997; Lundberg and Weinberg, 1998). The E2F family members function as both transcriptional repressors and activators, depending on cellular context.

E2F1 through -3 can switch from being transcriptional activators in proliferating cells to repressors in cells committed to differentiate and in response to DNA damage (Chong et al., 2009a; Ianari et al., 2009).

The best-characterized function of E2F1 through -3 is the activation of gene transcription to promote G1/S phase transition leading to proper cellular proliferation (Wu et al., 2001).

Constitutive expression of E2F1 through -3 induce S phase entry of quiescent cells and

triple knockout lead to defects in proliferation and increased cell death (Chong et al.,

2009a; Johnson et al., 1993; Wu et al., 2001). During G2 phase, E2F1 through -3 activate

transcription of genes required for G2/M progression (Zhu et al., 2004). The E2F3 locus

encodes two gene products, E2f3a and E2f3b, through the use of alternate promoters.

While E2F3a expression is highest during S phase, E2F3b is expressed constitutively

throughout the cell cycle (Leone et al., 2000). Both E2F3 isoforms can function as activators or repressors of E2F target genes (Chong et al., 2009b).

E2F4 and -5 function as transcriptional repressors by forming a complex with p107 or p130, which recruits chromatin remodeling proteins to repress E2F target genes and prevent G1/S phase transition (Takahashi et al., 2000). E2F6 acts as a transcriptional

repressor to inactivate E2F-responsive genes and inhibit S phase entry by forming a

repressor complex with enhancer of polycomb (Attwooll et al., 2005; Cartwright et al.,

1998; Trimarchi et al., 1998). E2F7 and -8 delay cell cycle progression during S phase

functioning as transcriptional repressors (Logan et al., 2005). E2F7 and -8 repress the

expression of E2F1 and their expression is dependent on E2F1 (Moon and Dyson, 2008).

This negative feedback loop serves to maintain E2F1 balance during cell cycle

progression.

1.7 Other functions of E2F factors

Inactivation of E2f genes in mice has demonstrated the existence of multiple roles for

these proteins. They include tumour formation in mice lacking the E2f1, E2f2 or E2f3

gene (Olson et al., 2007; Paulson et al., 2006; Scheijen et al., 2004), fetal erythrocyte

maturation defects in mice lacking the E2f4 gene (Kinross et al., 2006), prenatally lethal

hydrocephalus upon inactivation of the E2f5 gene (Lindeman et al., 1998), and

embryonic lethality with widespread apoptosis, vascular dilation and hemorrhage upon

joint inactivation of the E2f7 and E2f8 genes (Li et al., 2008).

1.7.1 Role of E2F in apoptosis

E2F1 through -6 play a role in activating apoptotic events in cultured cells and transgenic

mouse models (Chang et al., 2000; Chen et al., 2013; Johnson and Degregori, 2006;

Martinez et al., 2010; Yang et al., 2007). The involvement of E2F1 in activating

apoptosis has been extensively studied. Expression of E2F1 induces p53-dependent and -

independent apoptosis. E2F1 increases p53 expression levels by transcriptionally

activating the p14ARF gene. The p14ARF protein negatively regulates mouse double

minute 2 homolog (Mdm2), an E3 ubiquitin ligase, essential for ubiquitylating p53 and targeting it for degradation. Deregulated Mdm2 stabilizes p53 protein levels followed by activation of genes required for apoptosis, such as TAp73, Fas, Bax, and Noxa (Haupt et

al., 2003). E2F1 also stimulates expression of ataxia-telangiectasia mutated (ATM),

checkpoint kinase 1 (Chk1), and Chk2, which leads to p53 phosphorylation and

stabilization. Moreover, E2F1 up regulates transcription of several p53 co-factors,

including Aspp1, Aspp2, Jmy, and TP53inp, which are essential for induction of the pro-

apoptotic p53 target gene, Bax.

E2F1 can also stimulate apoptosis by inducing TAp73 expression (Wu et al., 2008).

Similarly, E2F1 is found on the promoter of genes important for the apoptotic pathway,

such as the Bcl-2 family of proteins, apoptotic protease-activating factor 1 (Apaf1),

caspase 3, and caspase 7 (Moroni et al., 2001; Wu et al., 2009). E2F1 is also involved in

transcriptionally inactivating genes important for cell survival, by disrupting the NF-κB

signaling pathway through down regulation of TNF receptor-associated factor 2 (TRAF2)

(Ginsberg, 2002).

1.7.2 Role of E2F in DNA damage

E2F1 up regulates expression of genes required for the initiation of the DNA damage

response (Polager et al., 2002). Some of these genes function in the mismatch repair

pathway, such as DNA mismatch repair protein 2 (MSH2), MSH6, mismatch repair

endonuclease (PMS2). Others include uracil-DNA glycosylase (UNG), flag structure-

specific endonuclease (Fen1), and PCNA, which participate in the base excision repair

(BER) pathway. E2F1 also induces the expression of genes involved in the NER

pathway, such as DDB2/XPE (Prost et al., 2007), and XPC (Lin et al., 2009). By acting

as a transcriptional repressor, E2F1 forms a complex with pRb, C-terminus-binding

protein (CtBP), and CtBP interacting protein (CtIP) corepressor to mediate transcriptional

inactivation of zinc finger and BRCA1-interacting protein with a KRAB domain 1

(ZBRK1) (Liao et al., 2010). ZBRK1 negatively regulates growth arrest and DNA-

damage-inducible alpha (GADD45A), an important DNA repair response protein that

functions to stimulate cell cycle arrest following DNA damage (Zheng et al., 2000).

Treatment with DNA damaging agents including UV or ionizing radiation, actinomycin

D, adriamycin, neocarzinostatin or etoposide increases E2F1 through -3 transcript and

protein levels (Castillo et al., 2015; Hofferer et al., 1999; Martinez et al., 2010; Meng et

al., 1999). Both transcriptional and post-translational modifications are required for the

induction of E2F1 through -3 in response to DNA damage. E2F1 and E2F3 are

phosphorylated at Ser364 and Ser124, respectively by Chk1/2 to activate transcription of

pro-apoptotic genes, such as TAp73 and Apaf1 (Martinez et al., 2010; Pediconi et al.,

2003; Stevens et al., 2003). E2F1 is also phosphorylated at Ser31 by ATM and/or ataxia

telangiectasia and Rad3-related (ATR) proteins following drug-induced DNA damage

(Lin et al., 2001). This leads to E2F1 stabilization and transcriptional activation of TAp73

and Apaf1. This residue is also necessary for E2F1's interaction with DNA topoisomerase

IIβ-binding protein I (TopBP1) to repress transcriptional activity of E2F1 by recruiting

Brahma (BRM) and Brahma-related gene 1 (BRG1), subunits of the SWI/SNF

chromatin-remodeling complex and to activate replication checkpoint surveillance

following DNA damage (Liu et al., 2003; Liu et al., 2004). UV-induced DNA damage

also phosphorylates E2F1, however under these circumstances, phosphorylation at Ser31

facilitates E2F1 recruitment to sites of UV-induced DNA damage or DNA double strand

breaks (DSB). Ser31 of E2F1 is also required for the recruitment of XPA, XPC, GCN5,

and accumulation of H3K9 acetylation at sites of UV-induced DNA damage but is

dispensable for E2F-dependent transcriptional activity and induction of apoptosis

(Biswas et al., 2014; Guo et al., 2011; Guo et al., 2010). Therefore, the stabilization of

E2F1 through phosphorylation at Ser31 could either increase its transcriptional activity or promote non-transcriptional functions of E2F1 (Biswas and Johnson, 2012).

1.8 Mechanisms of E2F1 Regulation 1.8.1 Post-translational modifications

E2F1 can be modified by acetylation, citrullination, methylation, neddylation,

ubiquitylation, and phosphorylation (Figure 1.9). E2F1 can be acetylated at Lys117,

Lys120, and Lys125 by the p300/CREB-binding protein (CBP) or by p300/CBP- associated factor (P/CAF) acetyltransferase (Martinez-Balbas et al., 2000; Marzio et al.,

2000). Acetylation of E2F1 allows for stabilization and increase in transcriptional

activation of E2F1 target genes (Farhana et al., 2002; Martinez-Balbas et al., 2000).

Ectopic expression of P/CAF or p300/CBP increases E2F1 ubiquitylation and it is

dependent on Lys117, Lys120, and Lys125 following DNA damage (Galbiati et al.,

2005). This could indicate that DNA damaging signals may prevent ubiquitylated E2F1

to be targeted for degradation through the proteasomal pathway (Galbiati et al., 2005).

Acetylation of E2F1 also promotes its interaction with bromodomain-containing protein 4

(BRD4). This interaction is dependent on E2F1 citrullination at Arg109, Arg111,

Arg113, and Arg127 by protein arginine deiminases 4 (PDA4) and enhances E2F1

binding to the promoter region of inflammatory genes tumor necrosis factor–α (TNFα)

and interleukin-1β (IL-1β) (Ghari et al., 2016).

Methylation of E2F1 can occur at Lys185 by histone-lysine N-methyltransferase

(SETD7), and demethylated by lysine-specific histone demethylase 1A (KDM1A) in

Figure 1.9: Post-translational modifications on E2F1.

The protein kinases involved in modifying the residues on E2F1 is shown. Ac, acetylation; Ci, citrullination; Me, methylation; NE, NEDDylation; P, phosphorylation.

vitro and in cultured cells (Kontaki and Talianidis, 2010; Xie et al., 2011a).

Demethylation stabilizes E2F1 and up regulates the transcription of TAp73 in response to

DNA damage. Importantly, methylation of E2F1 impairs its acetylation and

phosphorylation at Ser364, targeting E2F1 for ubiquitylation and degradation (Kontaki

and Talianidis, 2010). Methylation of E2F1 at Lys185 promotes E2F1 neddylation (also

at Lys185) and subsequent down regulation of E2F1 transcriptional activity and increases turnover of E2F1 (Loftus et al., 2012). E2F1 can also be methylated at arginine residues.

Asymmetric arginine methylation of E2F1 at Arg109 by protein arginine methyltransferase 1 (PRMT1) stabilizes E2F1 and symmetric arginine methylation at

Arg111 and Arg113 by PRMT5 promotes E2F1 ubiquitylation and degradation (Cho et al., 2012; Zheng et al., 2013). In response to DNA damage, the interaction between E2F1 and PRMT1 is enhanced with concomitant increase in asymmetric arginine methylation and this prevents PRMT5 binding and symmetric arginine methylation of E2F1.

Together, the E2F1-PRMT1 complex favors E2F1-induced apoptosis. On the other hand, when E2F1 is methylated by PRMT5, the binding of E2F1 with PRMT1 is inhibited and cell proliferation is stimulated (Zheng et al., 2013).

E2F1 regulation can also be controlled through ubiquitylation. Targeting of E2F1 for ubiquitylation can occur through the C-terminal activation domain. The physical association between pRb and E2F1 stabilizes E2F1 by protecting it from ubiquitylation

(Campanero and Flemington, 1997; Hateboer et al., 1996; Hofmann et al., 1996). At the

S/G2 boundary, E2F1 is regulated through ubiquitylation by S-phase kinase-associated

protein-2 (Skp2), the F-box protein of the Skp1–Cullin–F-box protein (SCF) E3 ubiquitin

ligase complex (Marti et al., 1999). Skp2 functions as a substrate recognition component

to specifically recognize phosphorylated substrates resulting in subsequent ubiquitylation and proteasomal degradation of the targeted protein (Tsvetkov et al., 1999). E2F1 interaction with 14-3-3τ (Wang et al., 2004) or Mdm2 (Zhang et al., 2005) displace Skp2 from E2F1 thereby inhibiting E2F1 ubiquitylation and degradation. Both 14-3-3τ and

Mdm2 enhance E2F1 transcriptional activity (Martin et al., 1995; Wang et al., 2004).

During mitosis, the interaction between E2F1 and activator cell division cycle20 (Cdc20) of the anaphase-promoting complex/cyclosome (APC/C) complex regulates E2F1 stability (Budhavarapu et al., 2012). Cdc20 homolog 1 (Cdh1), another activator subunit of the APC/C complex that preferentially interacts with destruction box (D-box) of substrates (Qin et al., 2016), specifically forms K11-linked polyubiquitin chains on E2F1 to target it for degradation during late S phase (Budhavarapu et al., 2012) However, DNA damage does not prevent Cdh1 from ubiquitylating E2F1, but rather inhibits its degradation. E2F1 can also be ubiquitylated in vitro by the multi-subunit Cullin (CUL)-

RING E3 ligase (CRL) independent of Skp2 and phosphorylation of E2F1 (Ohta and

Xiong, 2001). The presence of multiple E3 ligases that interact with and mediate degradation of E2F1 enables orderly control of E2F1 expression under different circumstances.

E2F1 can also be de-ubiquitylated by deubiquitylase 26S proteasome non-ATPase regulatory subunit 14 (PSMD14) and Ubiquitin carboxyl-terminal hydrolase isozyme L5

(UCHL5). PSMD14 stabilizes E2F1 by cleaving K11- and K63- linked polyubiquitin chains, which lead to increased expression of E2F1-induced pro-survival genes (Wang et al., 2015). UCHL5 is not involved in the stability of E2F1 but the cleavage of K63-linked

polyubiquitin chains lead to enhanced E2F1 transcriptional activity following DNA damage (Mahanic et al., 2015).

E2F1 is phosphorylated at multiple sites at the N- and C-terminus in vitro and in vivo by

a number of kinases, and most of these phosphorylation sites are conserved between

human and mouse orthologues. E2F1 phosphorylation at Ser332 and Ser337 residues by

CDK4-cyclin D attenuates E2F1-pRb association and phosphorylation of these residues

indirectly regulates E2F1 ubiquitylation (Fagan et al., 1994). E2F1 is also phosphorylated

by CDK2-cyclin A on Ser375 and reduces DNA binding (Dynlacht et al., 1994; Krek et

al., 1995; Xu et al., 1994). During S phase, E2F1 is also phosphorylated at Ser403 and

Thr433 by TFIIH-CDK7, which targets E2F1 for degradation (Vandel and Kouzarides,

1999). Disruption of the cyclin A binding domain or the TFIIH-CDK7 binding domain of

E2F1 increases its stability and can lead to apoptosis, emphasizing the importance of tight

regulation of E2F1 by both CDK2-cyclin A and TFIIH kinases (Krek et al., 1995; Vandel

and Kouzarides, 1999). The Ser403 and Thr433 of E2F1 are also phosphorylated in vitro

by p38β-MAPK. Activation of p38β-MAPK triggers E2F1 nuclear export and

degradation allowing for proper keratinocyte differentiation (Ivanova et al., 2009).

As mentioned earlier, E2F1 is phosphorylated at Ser31 and Ser364 residues by ATM and

Chk1/2, respectively, in response to DNA damage, and phosphorylation mediates E2F1

stabilization (Lin et al., 2001; Stevens et al., 2003). The Ser31 of E2F1 resides within the

Skp2 binding domain and has been suggested to modulate E2F1 degradation by inhibiting

Skp2 binding. It also mediates interactions with 14-3-3τ, which interfere with

ubiquitylation (Wang et al., 2004). Phosphorylation of E2F1 at Ser364 increases E2F1

transcriptional activity (Stevens et al., 2003).

1.8.2 E2F1 subcellular localization

E2F regulation can also be modulated by its subcellular localization (Verona et al., 1997).

The N-terminal domain of E2F1 through -3 has a nuclear localization signal (NLS)

(Muller et al., 1997). Differentiation leads to E2F1 nuclear export and degradation in keratinocytes (Ivanova and Dagnino, 2007). E2F1 has a nuclear export sequence (NES),

which allows chromosomal maintenance-1 (CRM1)-dependent export and is involved in nucleocytoplasmic shuttling in a variety of cells (Ivanova et al., 2007). Conversely, E2F4 and -5 are expressed predominantly in the cytoplasm owing to the presence of NES

(Magae et al., 1996). Both E2F4 and -5 are localized to the nucleus through their interactions with DP or p107/p130. However, the N-terminus of E2F5 has an intrinsic nuclear localization signal, which is required for its nuclear import independent of interactions with DP or p130 (Apostolova et al., 2002).

1.8.3 Protein-protein interactions

In addition to pRb and DP factors, E2F1 associates with other proteins. E2F1 interactions

with transcription intermediary factor 1-β (TRIM28) (Wang et al., 2007), MDM4

(Strachan et al., 2003), and neural proliferation differentiation and control protein 1

(NPDC1) (Sansal et al., 2000) reduces DNA binding and hence E2F1 transcriptional

activity. On the other hand, E2F1 binding with poly [ADP-ribose] polymerase 1 (PARP1)

(Simbulan-Rosenthal et al., 2003), host cell factor 1 (HCF1) (Tyagi et al., 2007), and

chromodomain-helicase-DNA-binding protein 8 (CHD8) (Subtil-Rodriguez et al., 2014)

promotes cell cycle progression by recruiting chromatin remodeling proteins to E2F1

target promoter regions.

1.9 The role of E2F in the epidermis

The E2F family of transcription factors modulate many cellular processes involved in

epidermal homeostasis, including cell proliferation, apoptosis, differentiation, DNA

replication, DNA damage response and DNA repair (Dicker et al., 2000; Hong et al.,

2008; Ivanova et al., 2009; Stevens and La Thangue, 2004; Wang et al., 2004). One of

the events that contribute to the abnormal regulation of proliferation and differentiation is the disruption of the pRb/E2F pathway, which is common in most cancers, including

NMSC (Ebihara et al., 2004; Yamazaki et al., 2005).

The members of the E2F family of transcription factors are differentially expressed in the epidermis. The E2f1 through -4 transcripts are expressed in the basal layer, whereas E2f5

transcripts are abundant in the suprabasal layers (Dagnino et al., 1997). Moreover, the

expression levels of E2F1 through -3 decrease following keratinocyte differentiation, with an increase in E2F5 protein levels (D'Souza et al., 2001; Wong et al., 2003).

E2F1 maintain normal proliferation in undifferentiated keratinocytes and modulates keratinocyte differentiation (Ivanova et al., 2009; Wong et al., 2003). In undifferentiated

keratinocytes E2F1 localizes to the nucleus, but differentiation induces E2F1 export to

the cytoplasm, where it is degraded through the proteasomal pathway. The signaling

pathways involved in E2F1 turnover during Ca2+-induced keratinocyte differentiation

involve the activation of PKCη and PKCδ. PKCη and PKCδ subsequently activate p38β,

which appears to phosphorylate E2F1 at Ser403 and Thr433 (Figure 1.10). E2F1 nuclear

Figure 1.10: Regulation and signaling pathway of E2F1 in differentiated keratinocytes.

In keratinocytes, E2F1 shuttles between the nucleus and cytoplasm. Upon keratinocyte differentiation through external stimuli, such as Ca2+, it triggers signaling events that lead to activation of protein kinase C eta and delta which phosphorylates p38β. p38β appears to phosphorylate E2F1 at Ser403 and Thr433. This leads to E2F1 nuclear export through the chromosome region maintenance 1 (CRM1)-nuclear protein export pathway with concurrent ubiquitylation of E2F1 and degradation through the proteasomal pathway. The sequence of events involving E2F1 phosphorylation, ubiquitylation, nuclear export, and subsequent degradation are required for proper keratinocyte differentiation. Ub, ubiquitylation; P, phosphorylation.

export is dependent on chromosome region maintenance 1 (CRM1) and requires E2F1

Ser403 and Thr433. When differentiated keratinocytes are exposed to a specific inhibitor of CRM1-nuclear export,leptomycin B (LMB), E2F1 accumulates in the nucleus and is stable. Further, forced nuclear retention by engineering an additional NLS protects E2F1 from degradation. This indicates that E2F1 nuclear export is required for its proteasomal degradation during keratinocyte differentiation. The sequence of events involving E2F1 phosphorylation, ubiquitylation, nuclear export, and subsequent degradation are therefore required for proper keratinocyte differentiation. In cultured keratinocytes, substitution of

Ser403 and Thr433 to Ala (hereafter termed ST/A) generated a mutant protein defective in nuclear export, and more stable than wild-type E2F1 following Ca2+-induced

keratinocyte differentiation. Exogenous expression of the ST/A in cultured keratinocytes

caused defects in differentiation (Ivanova et al., 2009). Abnormal increases in E2F1 expression and activity can lead to epidermal tumor formation by inducing proliferation- specific genes and suppressing squamous differentiation-specific genes (Dicker et al.,

2000). Constitutive expression of E2F1 in basal keratinocytes results in hyperkeratosis

and contributes to tumor development (Pierce et al., 1998b). Inactivation of the E2f1 gene

results in impaired keratinocyte migration, proliferation and significant delay in wound

healing (D'Souza et al., 2002).

1.10 Rationale and Aims of this Study

The E2F1 transcription factor regulates expression of key genes involved in cell cycle

progression and cell survival. The levels of E2F1 are tightly regulated during the cell

cycle, when keratinocytes and other cells differentiate, as well as in response to DNA

damage. Despite the fact that the regulation of E2F1 has been extensively studied, major

51 questions remain to be addressed. First, E2F1 is necessary for recruitment of XPC to damaged sites, however the recognition of DNA damage by XPC is dependent on hHR23. This begs the question: does E2F1 co-operate with hHR23 to recruit XPC and subsequent recruitment of XPA to DNA damage sites? Second, E2F1 degradation is necessary for proper keratinocyte differentiation. E2F1 Ser403 and Thr433 are required for its degradation, but whether other sites on E2F1 are necessary for proper keratinocyte differentiation is not known. Third, it is currently unclear whether mechanisms that regulate E2F1 stability during cell cycle progression also regulate E2F1 stability following DNA damage or differentiation. The objective of this study is to further understand the regulation of E2F1 in epidermal keratinocytes. I hypothesize that the regulation of E2F1 protein turnover plays a critical role during epidermal keratinocyte

DNA repair and differentiation. To test the hypothesis, I addressed the following aims:

Specific Aims:

1) Characterize the role of E2F1 and hHR23 following UV-induced DNA damage in

keratinocytes.

2) Identify residues essential for E2F1 stability following drug-induced DNA damage in

keratinocytes.

3) Investigate the mechanisms of E2F1 degradation following keratinocyte

differentiation.

Chapter 2 2 Materials and Methods 2.1 Reagents and materials

The materials used in this study are listed in Table 2.1. The reagents and, where

applicable, the concentrations used in this study are listed in Table 2.2. The composition

of routinely used solutions and the final concentrations of each reagent are presented in

Table 2.3.

Table 2.1: List of materials and catalogue numbers. Materials Source Catalogue number 3-µm pore polycarbonate EMD Millipore, Billerica, MA TSTP02500 membrane 8×10 inch Blue Ray Film Endoch Inc. Concord, ON, CA 810UPB Amicon Ultra-15 columns EMD Millipore, Billerica, MA UFC903024 Amersham Hybond-N+ Nylon GE Healthcare, Mississauga, RPN203B membrane ON Corning Incorporated, Corning, Falcon® 70 µm cell strainers 352350 NY VWR International, Radnor, Grade 703 Blotting paper 28298-020 PA GE Healthcare, Mississauga, PD10 desalting column 17-0851-01 ON Poly-L-lysine-coated glass Neuvitro Corporation, NEU-GG-12-PLL coverslips 12 mm Vancouver, WA Thermo Scientific, Waltham, Immu-mount mounting media 9990402 MA Polyvinylidene difluoride EMD Millipore, Billerica, MA IPVH00010 membranes (PVDF) Santa Cruz Biotechnology, Ultra Cruz autoradiography film SC-201697 Santa Cruz, CA

Table 2.2: Reagents, the catalogue number, and the final concentration used.

Reagents Source Catalogue Number Concentration 0.25% trypsin with Gibco, Burlington, ethylenediaminetetraacetic 25200-056 0.25% ON acid (EDTA) Gibco, Burlington, 2.5% Trypsin 15090-046 0.25% ON BioShop, Agar AGR001 1.5% Burlington, ON Albumin, Bovine Serum BioShop, ALB001 2% (BSA) Burlington, ON Amersham ECL Prime GE Healthcare, Western Blotting detection RPN2232 N/A Mississauga, ON reagent BioShop, Aprotinin APR200 1 µg/ml Burlington, ON Sigma, St Louis, Betaine B2629 1 M MO Bradford protein Bio-Rad, 500-0006 20% quantification solution Mississauga, ON Ca2+-free Eagle’s Lonza, minimum essential 06-174G N/A Walkersville, MD medium (EMEM) Chelex 100 resin (100-200 Bio-Rad, dry mesh size, 150–300 1422832 80 mg/ml Mississauga, ON µm wet bead size) Sigma, St. Louis, Cholera toxin C8052 9.5 ng/ml MO Sigma, St Louis, Cycloheximide (CHX) C7698 100 µg/ml MO Dimethyl sulfoxide Caledon Labs Ltd. 4100-1 N/A (DMSO) Georgetown, ON Roche Diagnostics, Dispase II 4942078001 8 mg/ml Indianapolis, IN Epidermal growth factor Sigma, St Louis, E4127 5 ng/ml (EGF) MO

Thermo Scientific, FBS for keratinocytes 12483-020 8% Waltham, MA

Gibco, Burlington, Fetal bovine serum (FBS) 12483 4% ON

Sigma, St Louis, Etoposide E1383 150 µM MO

Glutathione Sepharose™ GE Healthcare, 17-0756-01 50% 4B Mississauga, ON

Thermo Scientific, Hoechst 33342 H1399 1 µg/ml Waltham, MA

Sigma, St. Louis, Hydrocortisone H4001 74 ng/ml MO

HyQ-Dulbecco’s Modified Thermo Scientific, Eagle Medium-Reduced SH3056501 N/A Waltham, MA Serum (HyQ-DMEM-RS)

Thermo Scientific, Imidazole 03196-500 Varies Waltham, MA

Sigma, St. Louis, 5 µg/ml, pH Insulin H4001 MO 3.0

Isopropyl β-D-1- BioShop, thiogalactopyranoside 0.5 mM Burlington, ON (IPTG) IPT001.5

Enzo life sciences, Leptomycin B (LMB) ALX-380-100 5 ng/ml Plymouth, PA

BioShop, Leupeptin LEU001 1 µg/ml Burlington, ON

Sigma, St. Louis, L-Glutathione reduced G4251 10 mM MO

Lipofectamine 2000 Thermo Scientific, 11668029 15 µg Transfection Reagent Waltham, MA

Sigma, St. Louis, N-ethylmaleimide (NEM) E3876 10 mM MO

Qiagen, Louisville, Ni-NTA agarose slurry 30210 50% KY

Thermo Scientific, Paraformaldehyde (PFA) AC416780030 4% Waltham, MA 100 U/ml, Penicillin-Streptomycin, Gibco, Burlington, 15140-122 (Pen) 100 100× liquid ON µg/ml (Strep) BioShop, Pepstatin A PEP605 1 µg/ml Burlington, ON Phenylmethylsulfonyl- BioShop, PMS123 1 mM fluoride (PMSF) Burlington, ON

Pierce protein A/G Thermo Scientific, 400 µg or 88803 magnetic beads Waltham, MA 250 µg

Polyethylenimine (PEI, 25 Polysciences, 1 mg/ml, 23966 kDa linear) Warrington, PA pH 5.0 (Stock)

Poly-L-lysine Sigma, St. Louis, A00013 1 mg/ml hydrobromide MO

BD Biosciences, Rat tail collagen type I 354236 20 µg/cm2 Bedford, MA

Thermo Scientific, SimpleBlue™ Safe Stain LC6060 100% Waltham, MA

Sodium Pyrophosphate Sigma, St. Louis, tetrabasic decahydrate S9515 10 mM MO (Na P O ·10H2O)

₄ ₂ ₇ BDH Inc. Dubai, Sodium Azide (NaN ) B30111 10 mM 3 UAE

Sigma, St. Louis, Sodium Deoxycholate D6750 0.5% MO

Sigma, St. Louis, Sodium Fluoride (NaF) S-6521 10 mM MO

Sodium Orthovanadate BioShop, SOV664-10 1.2 mM (Na VO ) Burlington, ON

₃ ₄ Sigma, St. Louis, Triiodothyronine (T3) T6397 6.7 ng/ml MO

Z-Leu-Leu-Leu-al (MG- Sigma, St. Louis, C2211 10 µM 132) MO

Table 2.3: Composition of routinely used solutions and media.

Name Composition 1× Sodium dodecyl sulfate 25 mM Tris, 192 mM glycine, 0.1% (w/v) SDS (SDS) 50 mM Tris-HCl pH 6.8, 2% (w/v) SDS, 0.1% (w/v) 1× Laemmli loading buffer bromophenol blue, 10% (v/v) glycerol, 100 mM DTT

1× Phosphate Buffered 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM Saline (PBS) KaH2PO4, pH 7.4 1× Tris-acetate-EDTA (TAE) 40 mM Tris, 5 mM NaAc, 1 mM EDTA, 20 mM HAc 1× Tris-Buffered Saline 25 mM Tris-HCl, 137 mM NaCl, 5 mM KCl, pH 7.4 (TBS) 1× Tris-Glycine Transfer 39.2 mM glycine, 48.1 mM Tris, 20% (v/v) methanol Buffer 2× Yeast extract and 2% (w/v) bacto-tryptone, 1% (w/v) yeast extract, 85 Tryptone (2× YT) broth mM NaCl2, 17 mM MgSO4 Guanidine hydrochloride- 6 M GnHCl, 0.2% (v/v) Nonidet P-40 (NP-40), 20 mM based (GnHCl) stripping Tris-HCl pH 7.5 solution 1% (w/v) bacto-tryptone, 0.5% (w/v) yeast extract, 0.5% Luria Bertani (LB) Broth (w/v) NaCl2 Low pH stripping buffer 25 mM glycine pH 2.0, 1% (w/v) SDS LB Agar LB + 1.5% (w/v) Agar 50 mM Tris-HCl pH 7.4, 150 mM NaCl, 0.1% (v/v) Modified NP-40, 0.5% (w/v) sodium deoxycholate, 0.1% (w/v) radioimmunoprecipitation SDS, 1 µg/ml aprotinin, leupeptin, pepstatin A, 1 mM assay (RIPA) Buffer PMSF, 10 mM sodium pyrophosphate, 10 mM sodium fluoride, 1.2 mM sodium orthovanadate Mouse tonicity phosphate- 150 mM NaCl, 4 mM NaH PO , 16 mM Na HPO buffered saline (MTPBS) 2 4 2 4 Super Optimal Broth (SOB) 2% (w/v) bacto-tryptone, 0.5% (w/v) yeast extract, 10 medium mM NaCl, 2.5 mM KCl, 10 mM MgCl2, 10 mM MgSO4 TBST TBS, 0.1% (v/v) Tween-20 10 mM PIPES, 15 mM CaCl , 250 mM KCl, adjust to Transformation buffer (TB) 2 pH 6.7 with KOH, then add MnCl (final concentration, solution 2 55mM)

2.2 Antibodies, enzymes and kits

The antibodies used in the study are listed in Table 2.4. Table 2.5 shows the list of restriction endonucleases, DNA polymerases, and other modifying enzymes. Table 2.6 shows the list of commercially available kits used in this study. For immunoblotting, primary antibodies were diluted in TBST unless stated otherwise, and secondary antibodies were diluted in TBST containing 5% (w/v) non-fat milk. For immunofluorescence microscopy, primary antibodies were diluted in PBS containing 5%

(v/v) goat serum, and secondary antibodies were diluted in PBS containing 1% (w/v) non-fat milk and 5% (v/v) goat serum.

Table 2.4: Antibodies and dilutions used.

Antibodies Source Catalogue Number Dilution* Primary antibodies Anti-FZR1 [DH01 (DCS- Abcam, Cambridge, ab3242 IB: 1:2500 266)] UK Santa Cruz Anti-c-Myc (9E10) Biotechnology, SC-40 IB:1:1000 Santa Cruz, CA CosmoBio Co. IB: 1:5000 Anti-CPD (Clone TMD-2) NMDND001 Tokyo, Japan IF: 1:1000 Abcam, Cambridge, Anti-dsDNA Ab27156 IB: 1:1000 UK OriGene IB: 1:1000 Anti-E2F1 Technologies, TA308764 IP: 3 µg Rockville, MD Santa Cruz Anti-E2F1 (C-20) Biotechnology, SC-193 IB: 1:500 Santa Cruz, CA Sigma, St. Louis, Anti-FLAG (Rabbit) F7425 IP: 3 µg MO Sigma, St. Louis, Anti-FLAG M2 F1804 IB: 1:5000 MO Assay Designs, Ann Anti-GAPDH CSA-335 IB: 1:10 000 Arbor, MI Santa Cruz IB: 1:5000 Anti-HA (Y-11) Biotechnology, SC-805 IP: 3 µg Santa Cruz, CA IF: 1:1000 Cell Signaling, Anti-Lamin A/C 2032S IB: 1:2500 Danvers, MA Santa Cruz Anti-Rad23A (D-6) Biotechnology, SC-365669 IB: 1:1000 Santa Cruz, CA Abcam, Cambridge, Anti-V5 (Chicken) Ab9113 IF: 1:1000 UK IB: 1:5000 Thermo Scientific, Anti-V5 (Mouse) R960-25 IP: 3 µg Waltham, MA IF: 1:1000

Hybridoma Bank, E7 Anti-β-Tubulin IB: 1:50 University of Iowa RRID:AB_2315513 Sigma, St. Louis, Anti-γ-Tubulin T6557 IB: 1:50 000 MO AlexaFluor™ 488- Thermo Scientific, A12379 IF: 1:100 Phalloidin Waltham, MA AlexaFluor™ 564- Thermo Scientific, A12381 IF: 1:100 Phalloidin Waltham, MA AlexaFluor™ 647- Thermo Scientific, A22287 IF: 1:100 Phalloidin Waltham, MA Jackson ChromPure Mouse IgG ImmunoResearch, 015-000-003 IP: 3 µg Whole molecule West Grove, PA Jackson AffiniPure Goat anti- ImmunoResearch, 111-005-045 IP: 3 µg Rabbit IgG [H+L] West Grove, PA Secondary antibodies AlexaFluor™ 488- Thermo Scientific, conjugated Goat anti- A11008 IF: 1:1000 Waltham, MA Rabbit IgG AlexaFluor™ 594- Thermo Scientific, conjugated Goat anti- A11042 IF: 1:1000 Waltham, MA Chicken IgG AlexaFluor™ 594- Thermo Scientific, conjugated Goat anti- A11005 IF: 1:1000 Waltham, MA Mouse IgG AlexaFluor™ 647- Thermo Scientific, conjugated Goat anti- A21235 IF: 1:1000 Waltham, MA Mouse IgG AlexaFluor™ 647- Thermo Scientific, conjugated Goat anti- A21244 IF: 1:1000 Waltham, MA Rabbit IgG Jackson HRP-conjugated Goat anti- ImmunoResearch, 115-035-146 IB: 1:5000 Mouse IgG West Grove, PA Jackson HRP-conjugated Goat anti- ImmunoResearch, 111-035-003 IB: 1:5000 Rabbit IgG West Grove, PA *IB: immunoblot, IP: immunoprecipitation; IF: immunofluorescence.

Table 2.5: Enzymes for molecular biology experiments.

Enzymes Source Catalogue number

New England BioLabs Inc, ApaI R0114 Whitby, ON New England BioLabs Inc, HindIII-HF R3104 Whitby, ON New England BioLabs Inc, λ Protein Phosphatase P0753S Whitby, ON New England BioLabs Inc, NotI-HF R3189 Whitby, ON Quanta Biosciences, PerfeCTa® qPCR SuperMix 95064-100 Gaithersburg, MD Platinum™ Taq DNA Thermo Scientific, Waltham, 10966-018 polymerase MA SuperScript™ II Reverse Thermo Scientific, Waltham, 18064-014 Transcriptase MA New England BioLabs Inc, T4 DNA ligase M0202 Whitby, ON New England BioLabs Inc, XhoI R0146 Whitby, ON

Table 2.6: List of kits used in this study.

Kits Source Catalogue Number

RNeasy Mini Kit Qiagen, Louisville, KY 74104 DNeasy® Blood & Tissue Kit Qiagen, Louisville, KY 69506 QIAprep Spin Miniprep Kit Qiagen, Louisville, KY 27106 EndoFree Plasmid Maxi Kit Qiagen, Louisville, KY 12362 QIAquick Gel Extraction Kit Qiagen, Louisville, KY 28704 NE-PER™ Nuclear and Thermo Scientific, Waltham, 78833 Cytoplasmic Extraction Kit MA Macherey-Nagel, Düren, NucleoBond® Xtra Maxi Kit 740414 Germany QIAquick PCR Purification Qiagen, Louisville, KY 28104 Kit

2.3 Vectors

Table 2.7: Plasmids used in this study.

Plasmid References or source Thermo Scientific, Waltham, MA pCDNA 3.1 catalogue number V790-20 pCDNA 3.1 hHR23A-FLAG pCDNA 3.1 hHR23A-HA pCDNA 3.1 hHR23A-HA UBA2 Dr. Christine Blattner, Institute of Toxicology and Genetics, Karlsruher Institut für Technologie, pCDNA 3.1 hHR23A-HA ΔUBA2 Germany (Glockzin et al., 2003) pCDNA 3.1 hHR23A-HA ΔUbL pCDNA 3.1 hHR23B-FLAG pCDNA 3.1 V5-E2F1

K117R/K120R/K125R (KTR) pCDNA 3.1 V5-E2F1 K185R pCDNA 3.1 V5-E2F1 R111K/R113K (RDK) pCDNA 3.1 V5-E2F1 S31A pCDNA 3.1 V5-E2F1 S332A pCDNA 3.1 V5-E2F1 S337A This study pCDNA 3.1 V5-E2F1 S364A pCDNA 3.1 V5-E2F1 S375A pCDNA 3.1 V5-E2F1 S403D pCDNA 3.1 V5-E2F1 T433D pCDNA 3.1 V5-E2F1 S403D/T433D (ST/D) pCDNA 3.1 V5-E2F1 WT-NES pCDNA 3.1 V5-E2F1 ST/A-NES

62 pEGFP-C1 hHR23A-HA WT pEGFP-C1 hHR23A-HA UBA1 pEGFP-C1 hHR23A-HA UBA2 pEGFP-C1 hHR23A-HA XPC- UBA2 pEGFP-C1 hHR23A-HA ΔUBA1 This study pEGFP-C1 hHR23A-HA ΔUBA2 pEGFP-C1 hHR23A-HA ΔUbL pEGFP-C1 hHR23A-HA ΔXPC pET32b-TRX-hHR23A-FLAG pCDNA 3.1 V5-E2F1 WT pCDNA 3.1 V5-E2F1 T433A Dr. Iordanka A. Ivanova, Western University, pCDNA 3.1 V5-E2F1 S403A London, Ontario (Ivanova et al., 2007) pCDNA 3.1 V5-E2F1 S403A/T433A (ST/A) pCDNA 3.1 FLAG-Cdh1 Dr. Daniele Guardavaccaro, Hubrecht Institute, pCDNA 3.1 FLAG-Cdc20 Utrecht, Netherlands (Ping et al., 2012) pCMV2 FLAG-Skp2 Dr. Wenyi Wei, Harvard Medical School Boston, MA (Wei et al., 2004) Addgene, Cambridge, MA, deposited by Dr. pCMV2 FLAG-Ubc13 Dong-Er Zhang, plasmid #12460 (Zou et al., 2005)

Dr. Peter K. Jackson, Stanford University School pCS2-Myc-hEmi1 of Medicine, Stanford, CA (Hsu et al., 2002)

Clontech Laboratories, Inc, Mountain View, CA, pEGFP-C1 catalogue number 6084-1

Novagene®, EMD Millipore, Billerica, MA pET32b-TRX catalogue number 69016

Amersham Pharmacia Biotech, Piscataway, NJ pGEX-4T-1 catalogue number 28-9545-49

Dr. Paul Hamel, University of Toronto, Toronto, pGEX-GST-E2F1 Ontario

Dr. David Litchfield, Western University, pBluescriptSK--HA-Ubiquitin-WT London, Ontario (Treier et al., 1994)

Addgene, Cambridge, MA, deposited by Sandra pRK-HA-Ubiquitin-K11 Weller, plasmid #22901 (Livingston et al., 2009)

pRK-HA-Ubiquitin-K0 Plasmid #17603 Addgene, Cambridge, pRK-HA-Ubiquitin-K48 Plasmid #17605 MA, deposited by Dr. pRK-HA-Ubiquitin-K63 Plasmid #17606 Ted Dawson (Lim et al., 2005) pRK-HA-Ubiquitin-WT Plasmid #17608

2.4 Molecular cloning 2.4.1 Generation of vectors encoding E2F1 and hHR23A mutants

All E2F1 mutants were generated using a three-step site-directed mutagenesis strategy, as summarized in Figure 2.1. This approach allowed for introduction of mutations in the central regions of the E2F1 cDNA while simultaneously maintaining good amplicon yields, which proved challenging because the E2F1 cDNA has several GC-rich regions that interfered with optimal polymerase-chain reaction (PCR) amplification. The plasmid encoding V5-tagged wild type E2F1 was used as a template (Table 2.7). For each V5- tagged E2F1 mutant, the N- and the C-terminal halves were amplified separately. For the

N-terminal half, a forward primer containing a HindIII site for subsequent cloning, together with sequence encoding the V5 tag. In these reactions, the reverse primer contained the mutated nucleotide(s). For the C-terminal half, a forward primer containing

Figure 2.1: Schematic representation of the three-step PCR protocol used for site- directed E2F1 mutagenesis.

A three-step PCR amplification procedure was used to generate fragments F’, R’ and FL. Two complementary primers to the E2F1 cDNA, as indicated by a and d (primers E2F1 V5 WT Fwd and Rev, see Table 2.8) including, respectively HindIII and NotI sequences, together with primers containing the desired point mutations (red circle) were designed to generate the vectors encoding E2F1 mutants used in this thesis. For each mutant construct, fragment F’ was amplified using primers a and c (PCR 1), and fragment R’ was amplified using primers b and d (PCR 2). Amplicons F’ and R’ contain overlapping sequences, which allowed them to be used as templates for the third, and final PCR reaction to synthesize full-length FL fragments , using primes a and d (PCR 3). All FL fragments were digested with HindIII and NotI and cloned into pCDNA3.1.

the same mutated nucleotide(s) paired with a reverse primer containing the STOP codon

(TGA) and a NotI site for subsequent cloning (Figure 2.1). The N- and C- terminal amplicons were then combined and served as templates in a third PCR reaction, which generated full-length V5-tagged E2F1 cDNAs, using the HindIII-V5-tag forward and

STOP-NotI reverse primer (primers a and d, Figure 2.1). Full-length E2F1 mutant cDNAs thus obtained (1500 bp) were cloned into the expression plasmid pCDNA3.1

(4500bp) at the HindIII and NotI sites. The sequences of all the primers used to generate mutant E2F1 cDNAs are shown in Table 2.8.

To add a NES to the E2F ST/A mutant, the simian virus 40 (SV40) large T antigen NES

(LPPLERLTL) was inserted at the C-terminus of E2F1 ST/A by PCR. The sequences of all mutant constructs generated were verified by dideoxy sequencing at the Robarts

Research Institute sequencing facility.

To generate vectors encoding GFP-tagged hHR23A forms, cDNAs corresponding to wild type hHR23A and the deletion mutants ∆UbL, ∆UBA1, ∆UBA2, ∆XPC, XPC-UBA2,

UBA1 and UBA2 were amplified by PCR using reverse primers that included sequences corresponding to the HA tag (AGCATAATCAGGAACATCATA) at the C-terminus

(Table 2.9). The templates used for these reactions were vectors encoding the corresponding HA-tagged hHR23A proteins, which were generous gifts from Dr.

Christine Blattner (Glockzin et al., 2003). Amplicons thus generated were cloned into the

XhoI and ApaI sites of pEGFP-C1. The sequences of all mutant constructs generated were verified by dideoxy sequencing at the Robarts Research Institute sequencing facility.

Table 2.8: Forward (Fwd) and reverse (Rev) primers used to generate E2F1 point mutants. Primers Sequence* Fwd: 5'-CTTaagcttCCATGGGTAAGCCTATCCCTAACCCTCTCC TCGGTCTCGATTCTACGGCCTTGGCCGGGGCCCCTGCGGGC E2F1 V5 WT GGCCCA-3' Rev: 5'-AATAATgcggccgcTCAGAAATCCAGGGGGGTGAGGT CCCCAAAGTCACAGTCG-3' Fwd: 5'-CGGCTGCTCGACTCCGCGCAGATCGTCATCATC-3' E2F1 S31A Rev: 5'-GATGATGACGATCTGCGCGGAGTCGAGCAGCCG-3' Fwd:5'-CTGCCACCATAGTGGCACCACCACCATCATC-3' E2F1 S332A Rev: 5'-GATGATGGTGGTGGTGCCACTATGGTGGCAG-3' Fwd: 5'-CACCACCACCATCAGCTCCCCCCTCATCCC-3' E2F1 S337A Rev: 5'-GGGATGAGGGGGGAGCTGATGGTGGTGGTG-3' Fwd:5'-GTTGTCCCGGATGGGCGCCCTGCGGGCTCCCGTG-3' E2F1 S364A Rev: 5'-CACGGGAGCCCGCAGGGCGCCCATCCGGGACAAC- 3' Fwd: 5'-GACGAGGACCGCCTGGCCCCGCTGGTGGCGG-3' E2F1 S375A Rev: 5'-CCGCCACCAGCGGGGCCAGGCGGTCCTCGTC-3' Fwd: 5'-GCCAAGAAGTCCAGGAACCACATCCAGTGG -3' E2F1 K185R Rev: 5'- CCACTGGATGTGGTTCCTGGACTTCTTGGC -3' Fwd:5'- CCATCCAGGAAGAGGTGTGAGATCCCCGGGGGAG E2F1 AGGTCACGCTATG -3' K117R/K120R/K 125R Rev: 5'- CATAGCGTGACCTCTCCCCCGGGGATCTCACACCT CTTCCTGGATGG -3' Fwd: 5'-CCAGCTCGGGGCAAAGGCAAGCATCCAGGAAAAG E2F1 GTGTG-3' R111K/R113K Rev: 5'- CACACCTTTTCCTGGATGCTTGCCTTTGCCCCGAG CTGGC-3' Fwd: 5'-CCTGAGGAGTTCATCAGCCTTGACCCACCCCACGA GGCCC-3' E2F1 S403D Rev: 5'-GGCCTCGTGGGGTGGGTCAAGGCTGATGAACTCCT CAGG-3' Rev: 5'- CGAgcggccgcTCAGAAATCCAGGGGGTCGAGGTCC E2F1 T433D CCAAAGTCACAGTCG-3' Rev: 5'-gcggccgcTCAAAGAGTAAGTCTCTCAAGCGGTGGTA E2F1 ST/A NES GG AAATCCAGGGGGGCGAGGTCCCC 3' *Mutated nucleotides are underlined and shown in bold characters; the nucleotide sequence for the V5 tag is italicized. The HindIII and NotI sequences are written in lower case characters.

Table 2.9: Forward (Fwd) and reverse (Rev) primers used for deletion mutations in hHR23A. Primers Sequence* Fwd: 5'-CCATACGACGTCCCAGActcgagAGGGATCCATGGCC GTCACC-3' hHR23A WT Rev: 5'-GgggcccTTAAGCATAATCAGGAACATCATACTCGT CATCAAAGTTCTG-3' Fwd: 5'-CCATACGACGTCCCAGActcgagAGGGATCCATGCAG GG TACC-3' hHR23A UbL Rev: 5'-GgggcccTTAAGCATAATCAGGAACATCATAGGCAG GTGGGGGATGGG-3' Fwd: 5'-CATACGACGTCCCAGActcgagAGGGATCCATGGCGT ACCCATAC-3' hHR23A UBA2 Rev: 5'-GgggcccTTAAGCATAATCAGGAACATCATACTCGT CATCAAAGTTCTG-3' Fwd: 5'-CCCAGActcgagAGGGATCCATGGCCAGAGAGGACA AG-3' hHR23A UBA1 Rev: 5'-GgggcccTTAAGCATAATCAGGAACATCATACGGCT GCTCCGATACCTGG-3' Fwd:5'-GCCAGAGAGGACAAGAGCGCCACGGAAGCAGCAG G-3' hHR23A ∆UBA1 Rev: 5'-GCTCTTGTCCTCTCTGGCGGCAGGTGGGGGATGGG- 3' Fwd: 5'-GAGCAGCCGGCCACGCAGGTGACGCCGCAG-3' hHR23A ∆XPC Rev: 5'-CGTGGCCGGCTGCTCCGATACCTGGCTCTC-3' *The HA-tag is bolded and underlined; The XhoI and ApaI nucleotides are written in lower case characters.

To generate the construct encoding the thioredoxin (TRX) fusion protein of hHR23A for bacterial expression, a cDNA containing the FLAG tag sequence at the C-terminus was amplified by PCR using the following primers: Forward 5’-CCCAGAAAGCTTGCCGT

CACCATCACGCTC-3’ and reverse 5’-CCGTCTGGCGGCCG CTCATATGTCGTAA-

3’. The amplicon was digested with HindIII and NotI and cloned into the corresponding sites of the bacterial expression vector pET32b-TRX.

2.4.2 Polymerase chain reaction (PCR)

All PCR reactions were carried out using Platinum™ Taq DNA polymerase (Qiagen) in a

50-µl reaction containing 150 ng of DNA, 300 nM each forward and reverse primers, 200

µM dNTP mix, 1× PCR buffer (20 mM Tris-HCl pH 8.4 and 50 mM KCl), 1 M Betaine,

2 mM MgCl2, and 5 U Platinum™ Taq DNA polymerase. The reactions were conducted

in a PCR thermal cycler (Eppendorf Mastercycler EP Gradient S, Hauppauge, NY) under

the following cycling parameters: one cycle at 95ºC for 2 min, followed by 30 cycles of

denaturation at 94ºC for 30 sec, annealing at 58ºC for 30 sec, and elongation at 72ºC for

50 sec. The final step involved a post-elongation cycle at 72ºC for 7 min, followed by

cooling to 4ºC until sample retrieval from the thermal cycler. All PCR products were

analyzed on 1× TAE agarose gels.

2.4.3 Isolation and purification of DNA and PCR products

The PCR products were purified from agarose gels using gel purification kits (Qiagen),

following the manufacturers’ instructions. After gel purification, amplicons were digested

with the appropriate restriction enzymes and purified with PCR cleanup kits (Qiagen),

followed by ligation into appropriate vectors previously digested with the appropriate

restriction enzymes and gel purified.

2.4.4 DNA ligation

All ligation reactions were performed in 10-µl reaction volumes containing 1:3 molar

reactions of vector to insert, and using T4 DNA ligase. Ligation reactions were incubated

at 22ºC for 1 h before transformation into Escherichia coli (E. coli).

2.4.5 Preparation of competent DH5α and BL-21 (DE) cells

E. coli DH5α or BL-21(DE) cells were streaked on an LB agar plate and cultured for 16 h at 37ºC. Immediately afterwards, a single colony from the streak was used to inoculate 5 ml of LB medium and cultured for 16 h at 37ºC with shaking (250 rpm). The following day, bacteria were subcultured into 250 ml of SOB medium at a ratio of 1:3. The culture was incubated at 18ºC with shaking at 250 rpm. When the culture reached an optical density at 600 nm (OD600) of 0.6, it was cooled on ice for 10 min and centrifuged at

2500×g for 10 min at 4ºC. After removing the supernatant, cell bacterial pellets were resuspended in 80 ml ice-cold TB and incubated on ice for 10 min. The bacteria were then collected by centrifugation (2500×g, 10 min, 4ºC). The bacterial pellets were gently resuspended in 20 ml ice-cold TB and DMSO was added to a final concentration of 7%

(w/v). Two hundred-µl aliquots of competent cells were freeze using liquid nitrogen, and stored at -80ºC. The competence of the bacteria was determined by transforming 100 pg of DNA, and calculated using the following equation:

Number of colonies obtained × 106pg × 1000 µl total transformation volume × DF 100 pg transformed DNA µg 100 µl plated

= Number of colony-forming-units (cfu) per µg plasmid DNA

The transformation efficiency following this protocol was routinely > 1 × 106 cfu/µg supercoiled plasmid DNA.

2.4.6 Bacterial transformation

Plasmids were transformed into E. coli DH5α, to prepare vectors used for mammalian cell transfection. For bacterial expression of recombinant proteins, E. coli BL-21(DE)

was used because this strain generally attains higher yields of recombinant proteins. For

transformation, a 250-µl aliquot of competent bacteria was thawed on ice and mixed with

10 ng of purified plasmid DNA or 2 µl from a 10-µl ligation reaction. The bacterial suspension was incubated on ice for 30 min with intermittent mixing, followed by heat shock at 42ºC for 45 sec. Seven-hundred and fifty ml of LB medium were quickly added

and the bacteria were incubated at 37ºC for 1 h with shaking at 250 rpm. The

transformation mixture was centrifuged (3000×g, 3 min, 22ºC). The bacterial pellet was resuspended in 100 µl LB and plated on LB agar plates containing 100 μg/ml ampicillin or 50 µg/ml kanamycin, according to the antibiotic resistance associated with the plasmid transformed, and the plates were incubated at 37ºC for 16ºC to allow for colony formation.

2.5 Recombinant Protein Purification 2.5.1 GST fusion protein expression

Glutathione S-transferase (GST) fusion proteins were purified as described (Apostolova

et al., 2002; Ivanova et al., 2007) with slight modifications. GST and GST-E2F1

expression plasmids were transformed into E. coli BL-21(DE). A single colony was used

to inoculate 5 ml of 2× YT medium containing 200 µg/ml ampicillin, and cultured for 16

h at 37ºC with shaking at 250 rpm. One ml of this overnight culture was used to inoculate

200 ml of fresh 2× YZ with 200 µg/ml ampicillin, and was cultured at 37ºC until the

OD600 was 0.6 (approximately 4 h). To induce protein expression, isopropyl β-D-1-

thiogalactopyranoside (IPTG) was added to the bacterial culture to a final concentration

of 0.5 mM, and incubation continued at 18ºC for 18 h. Bacteria were harvested by

centrifugation (6000×g, 15 min, 4ºC) and resuspended in MTPBS buffer (Section 2.1,

Table 2.3) containing 1% (v/v) Triton X-100, 1 mM PMSF, 1 µg/ml each aprotinin,

leupeptin, and pepstatin A. The cells were lysed through a French-pressure cell press

(American instrument Co In, Silver Spring MD) twice at a pressure of 10,000-11,000 psi.

The lysate was then centrifuged (14000×g, 15 min, 4ºC) to remove cell debris. The

supernatant was incubated in 50% (v/v) slurry of GST sepharose beads rinsed with lysis

buffer prior to use. Incubation proceeded for 30 min at 4ºC with gentle mixing. The

sepharose beads were washed three times with MTPBS, and bound GST or GST fusion

proteins were eluted with 50 mM Tris-HCl pH 8.0 containing 10 mM reduced L-

glutathione. The purified GST fusion proteins were desalted using PD10 columns, and

concentrated to a final volume of about 500 µl, using Amicon Ultra-15 columns. The

protein concentrations in the eluates were determined using Bradford assays, and

recombinant proteins were stored in single-use 20-µl aliquots at -80ºC.

2.5.2 TRX fusion protein expression

His-TRX and His-TRX-hHR23A-FLAG proteins were induced as described in Section

2.5.1 and purified as described (Ho et al., 2009), with the following exceptions.

Harvested bacterial pellets were resuspended in lysis buffer containing imidazole (10 mM

imidazole, 300 mM NaCl, 50 mM NaH2PO4 (pH 8.0), 1% (v/v) Triton X-100, 1 mM

PMSF, 1 µg/ml each aprotinin, leupeptin, and pepstatin A), homogenized using a French- press, and cleared by centrifugation as above. Supernatants were mixed with 50% (v/v) slurry of Ni-NTA agarose previously rinsed with lysis buffer. Agarose slurry mixtures incubated for 30 min at 4ºC with gentle rocking, washed three times with wash buffer (40 mM imidazole, 300 mM NaCl, 50 mM NaH2PO4 (pH 8.0)), and eluted with wash buffer

containing 250 mM imidazole. The purified TRX fusion proteins were desalted using

PD10 columns, and concentrated to a final volume of about 500 µl, using Amicon Ultra-

15 columns. The protein concentrations in the eluates were determined using Bradford assays, and recombinant proteins were stored in single-use 20-µl aliquots at -80ºC.

2.6 Preparation of glass coverslips for cultured cells

Glass coverslips were washed in 1 N HCl at 60ºC for 4 h, and subjected to 2-3 washes in

double distilled H2O (ddH2O) at 50-60°C. The glass coverslips were allowed to cool to

room temperature, and were then rinsed with 95% (v/v) ethanol and allowed to air dry.

Dry coverslips were incubated in a sterile poly-L-lysine (PLL) solution (1 mg/ml) for 1 h

at 22°C, with gentle rocking. PLL-coated coverslips were washed with sterile ddH2O and allowed to air dry under sterile conditions in a tissue culture biosafety cabinet. Sterile

PPL-coated coverslips were then stored at 22ºC until needed. Prior to use to culture mammalian cells, PPL-coated coverslips were further coated with collagen type I as described (Nakrieko et al., 2008). Briefly, the coverslips were rinsed with 70% (v/v) ethanol, washed with PBS, and incubated in a solution of rat tail collagen type I (20

µg/cm2) dissolved in 0.02 N acetic acid for 4 h at 37°C. These coverslips were then washed three times with sterile PBS, and placed in tissue culture plates.

2.7 Cell culture and transfection 2.7.1 Isolation of primary epidermal keratinocytes

All experiments that involved the use of mice were approved by the University of

Western Ontario Animal Care Committee (Protocol No. 2015-021), in accordance with regulations and guidelines from the Canadian Council on Animal Care. Primary mouse

keratinocytes were isolated from 2d-old CD-1 mice and cultured in keratinocyte growth medium (Ca2+-free EMEM supplemented with 100U/ml penicillin, 100μg/ml

streptomycin, 74 ng/ml hydrocortisone, 5 μg/ml insulin, 9.5 ng/ml cholera toxin, 5 ng/ml

EGF, 6.7 ng/ml T3 and 8% (v/v) FBS pre-treated with chelex resin) (Dagnino et al.,

2010; Ivanova et al., 2006; Ivanova et al., 2009). Mice were euthanized by CO2/O2

inhalation and placed in 70% (v/v) isopropyl alcohol for sterilization. The limbs, tail and head were removed, and a longitudinal incision was made along the back, starting at the base of the skull. The skins were harvested using two forceps, washed with PBS, spread epidermis side up on 10-cm bacterial culture plates containing freshly diluted 0.25% (v/v) trypsin, and incubated for 16 h at 4°C. The trypsin was then removed by aspiration, and replaced with fresh 0.25% (v/v) trypsin, and the tissues were incubated at 37ºC for 15 min. The epidermis was mechanically separated from the dermis, and placed in 5 ml of keratinocyte growth medium. The epidermis was minced very finely with scissors, and 30 ml of keratinocyte growth medium was added. The minced epidermis was incubated with gentle rocking at 37ºC for 30-40 min. The cell suspension was filtered through a 70-μm

nylon mesh to remove the cornfield envelope, and the cell density in the suspension was

determined using a haemocytometer. Keratinocytes were plated at a density 1.6 × 105

cells/cm2. For microscopy experiments, cells were seeded on PPL- and collagen-coated sterile glass coverslips (Section 2.5). The day after plating, keratinocytes were washed once with PBS and cultured in fresh growth medium, which was subsequently replaced every other day. Keratinocyte differentiation was induced by in culture by adding growth

2+ medium containing 1 mM CaCl2 (“High-Ca medium”) for at least 24 h. Keratinocytes

were transfected 3 days after plating.

2.7.2 Culture of tumour cell lines

HeLa cervical carcinoma and RPMI-7951 human melanoma cells were purchased from the American Type Culture Collection. These cells were cultured in HyQ Reduced Serum

DMEM, supplemented with 4% (v/v) FBS at 37ºC in a humidified 5% (v/v) CO2

atmosphere. When cells reached a confluence of ~ 90%, they were subcultured at a ratio

of 1:4, and the growth medium was replaced 2-3 times per week. All experiments were

conducted with cells that had reached less than 80% confluence, and had been passaged

for 10 generations or less.

2.7.3 Cell transfection and drug treatments

Cells were transfected with PEI solution (Section 2.1, Table 2.2), as previously described

(Dagnino et al., 2010). Cells were transfected at a ratio of 1 µg of DNA per 3 µl PEI,

adjusted to 100 µl (24-well plates), 200 µl (6-well plates), 300 µl (6-cm plates), or 500 µl

(10-cm plates) with sterile 150 mM NaCl. The DNA/PEI mixture was vortexed, followed

by a pulse centrifugation, incubated at 22ºC for 10 min, and added dropwise onto the

culture medium, swirling gently to ensure thorough mixing. The cells were cultured in

the presence of DNA/PEI mix for 4 h, at which time they were washed twice with Ca2+- free PBS and cultured with fresh growth medium. Where indicated, transfected keratinocytes were induced to differentiate by culturing in growth medium containing 1 mM CaCl2 for a total of 24 h, which was initiated 4 h after transfection. For experiments

with proteasome inhibitors, cells were treated with MG132 (10 µM) for 6 h prior to

harvesting. Where indicated, LMB (5 ng/ml) was added 24 h after transfection and cells

were harvested 6 h later. To induce single strand DNA damage, etoposide (150 µM) was

added 24 h after transfection and cells were harvested 8 h later. For steady-state E2F1

protein levels, CHX was added to keratinocyte cultures 24 h after transfection for 30 min

prior to replacing the culture medium with growth medium containing either low Ca2+ or

high Ca2+ in the presence of CHX, and cell lysates were prepared 2.5 h later. For E2F1 t½

experiments, CHX (100 µg/ml) was added 24 h after transfection to keratinocytes

cultured in low Ca2+ or high Ca2+ for 4 h. Cell lysates were prepared at timed intervals following CHX addition and protein abundances were measured by immunoblot analysis

(Section 2.9).

2.8 RNA isolation and quantitative reverse-transcription PCR (qPCR)

Total RNA was isolated from cultured cells or from tissues harvested from 2d-old CD-1

mice, using RNeasy mini-kits (Qiagen), and following the manufacturer’s instructions.

Epidermis and dermis samples were obtained from skin treated with 8 mg/ml dispase for

15 minutes at 37°C, as described (Judah et al., 2012). RNA concentration and quality

were determined with an Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA). Only

samples with RNA integrity numbers ≥8.5 were used. RNA (200 ng/sample) was reverse-

transcribed using SuperScriptTM II RT, as per manufacturer’s protocol. qPCR reactions

were conducted on a CFX384™ Real-Time PCR system (Bio-Rad) operated by CFX

Manager™ software (version 1.6). cDNA samples corresponding to 20 ng of RNA were

amplified using PerfeCTa® qPCR SuperMix and the following primers (400 nM, final):

mHR23A forward 5’-AAGATCCGCATGGAACCTGA-3’ and reverse 5’-TGGTC

ACCATGACAACCACAA-3’; mHR23B forward 5’-CTTAGGCACCATGCAGGTCA-

3’ and reverse 5’-CGGAAAGGCATCTTTCCCCT-3’. The results were normalized to

the expression of the housekeeping genes Rpl6 and Rps29, which encode ribosomal

protein L6 and S29, respectively (de Jonge et al., 2007). Replicate cDNA samples were

amplified for 40 cycles (98°C for 10 sec, and 58°C for 30 seconds per cycle). Primer

specificity and amplicon sizes were confirmed, respectively, by analysis of melt curves

conducted between 65°C and 95°C, in 0.5°C increments, and agarose gel electrophoresis.

Relative mHR23A and mHR23B mRNA levels were calculated using the ΔΔCt method.

2.9 Immunoblot analysis and immunoprecipitation

Whole-cell lysates were prepared using a modified RIPA-lysis buffer (Section 2.1, Table

2.3). To prepare nuclear and cytoplasmic fractions, primary keratinocytes were lysed

using a NE-PER nuclear and cytoplasmic extraction kit, following the manufacturers’

protocol. For phosphatase treatments (Ivanova et al., 2009), cell lysates were prepared in

duplicate. One of the samples was prepared in modified RIPA-lysis buffer, and the other

in the same buffer, but without phosphatase inhibitors (Na3VO4, NaF and Na4P2O7). The lysates prepared in the absence of phosphatase inhibitors were further incubated with λ phosphatase (4 U/µg of lysate protein) at 30ºC for 30 min. In experiments assessing E2F1 ubiquitylation, the lysis buffer also contained 25 mM NEM. The protein concentration of lysates were measured using the Bio-Rad protein assay reagent (Table 2.2) on a DU®640

spectrophotometer (Beckman Coulter Inc. CA, USA). Proteins in lysates (30 µg/sample)

were resolved by denaturing polyacrylamide gel electrophoresis and transferred to PVDF

membranes, which were probed with antibodies indicated in individual experiments.

After antibody probing, protein signals were visualized using Amersham ECL Prime

Western Blotting Detection Reagent. Keratinocytes exhibits contact inhibition and therefore to obtain 1.5 mg of protein, five 10-cm tissue culture dishes (9.6 × 106 cells,

~2.5 skins from 3-day-old mice per plate) per sample were used for immunoprecipitation assays. Cell lysates prepared as described above (1.5 mg protein/sample, for Figures 3.2,

3.6, 3.14, 3.15, and 3.20 and 1 mg/protein/sample for Figures 3.4, 3.5, 3.21, and 3.23) were incubated with antibodies indicated in individual experiments (3 µg antibody/sample for Figures 3.2, 3.6, 3.14, 3.15, and 3.20 and 2 µg antibody/sample for

Figures 3.4, 3.5, 3.21, and 3.23) for 16 h at 4°C, with rotation. The immune complexes were captured by incubation with 400 µg of A/G magnetic beads previously rinsed with

PBS containing 0.05% (v/v) Tween-20 for 1 h at 22°C with rotation, followed by 5 washes with PBS containing 0.05% (v/v) Tween-20, and elution in 1× Laemmli sample buffer (Table 2.3) for 10 min at 25°C. Immune complexes were resolved and analyzed by immunoblot, as described above. Protein signals were visualized with either a 8×10 inch

Blue Ray Film and quantified using ImageJ (Schneider et al., 2012) or with a VersaDoc-

3000 imager (Bio-Rad Laboratories) and quantified with Quantity One software (Bio-Rad

Laboratories). In all experiments, E2F1 and hHR23A band intensities were normalized to

γ-Tubulin or GAPDH. For E2F1 turnover experiments, E2F1 to γ-Tubulin band intensity ratio for each time point was normalized to t=0. The apparent t½ of E2F1 protein was calculated by plotting the percentage of band intensity setting T=0 at 100% against time

(min) following CHX treatment and determined using GraphPad Prism software (version

5.00 for Windows, San Diego, CA) setting Y=0 at 100.0 and plateau at 0.0.

2.10 Indirect immunofluorescence microscopy

Twenty-four hours after transfection, cells were fixed and processed for microscopy, as previously described (Ivanova and Dagnino, 2007). Cells were fixed with 4% (w/v) PFA

on ice for 15 min and permeabilized with PBS containing 0.1% (v/v) Triton X-100 at

22ºC for 10 min. After washing with PBS buffer, cells were incubated with 1% (w/v)

non-fat milk and 5% (v/v) goat serum diluted in TBST for 1 h at 22ºC, and then

incubated with primary antibodies for 16 h at 4ºC (Table 2.4). Cells were washed three

times with 1× PBS and incubated with the appropriate AlexaFluor-conjugated secondary

antibody (Table 2.4) for 1 h at 22ºC. DNA was visualized using Hoechst 33342 (1 µg/ml, final; Table 2.2). The fraction of cells with nuclear or cytoplasmic E2F1 or hHR23A was determined from the analysis of at least 100 cells per experiment in replicates. To visualize CPD, cells were fixed in freshly diluted 4% (w/v) PFA in 1× PBS, followed by incubation in 0.2N HCl for 30 min at 22°C, to denature nuclear DNA. The cells were washed 5 times with PBS and incubated with primary antibodies, as described (Ivanova and Dagnino, 2007). All photomicrographs shown are representative of experiments

conducted on duplicate samples, and scoring > 100 transfected cells per experiment.

Images were obtained with a Leica DMIRBE fluorescence microscope equipped with an

Orca-ER digital camera (Hamamatsu Photonics, Hamamatsu City, Japan), using Volocity

6.1.1 software (Improvision-PerkinElmer, Waltham, MA).

2.11 In vitro binding assays

For in vitro protein binding assays, 2 µg of each recombinant protein indicated were

mixed in a final volume of 250-µl of PBS containing 1 mM PMSF, and 1 µg/ml each

aprotinin, leupeptin, and pepstatin A. Protein mixtures were incubated at 4ºC for 16 h,

followed by incubation in 50% (v/v) slurry of pre-equilibrated Ni-NTA agarose or GST sepharose beads at 4ºC for 30 min. Protein complexes captured by the resins were washed

5 times with PBS containing 0.05% (v/v) Tween-20 and eluted in 1× Laemmli sample buffer for 10 min at 70°C. The samples were resolved by electrophoresis, using polyacrylamide gels as described in Section 2.9.

2.12 UV damage and repair assays

The co-localization of proteins to sites of UVC radiation-induced DNA damage was

conducted as described (Guo et al., 2011; Guo et al., 2010), with minor modifications.

Keratinocyte monolayers were rinsed once with PBS, leaving a thin PBS layer. A sterile

3-µm pore polycarbonate membrane was gently placed onto the cells, followed by irradiation with 254 nm UVC light (200 J/m2), using a Bio-Rad GS Gene Linked UV

Chamber. The membrane was removed and normal growth medium was added to the

cells, which were processed for immunofluorescence microscopy 2 h later, as described

in Section 2.10.

Repair of UV-induced DNA damage was assessed as described (Guo et al., 2011), with

some modifications. HeLa or RPMI-7951 cells seeded in 10-cm culture dishes were

transfected with vectors encoding E2F1 and/or hHR23A. Sixteen hours after transfection,

the cells were trypsinized, re-seeded in quadruplicate 35-mm culture dishes, and cultured

for an additional 16-h interval to allow attachment. Duplicate samples were then mock-

irradiated or subjected to UVB treatment (302 nm, 100 J/m2), using a UVM-28 EL series8-watt lamp (UVP, Upland, CA). At timed intervals following UVB irradiation, genomic DNA was purified with DNeasy® Blood & Tissue kits (Qiagen) and denatured by boiling at 95ºC for 5 min. Replicate 200 ng samples were spotted onto Hybond N+

membranes (Amersham, NJ), which were baked at 80ºC for 1 h, and placed onto filter

paper pre-wetted with 0.4 N NaOH for 20 min. The membranes were blocked for 16 h at

4°C in 5% (w/v) non-fat milk diluted in TBST. After two TBST washes, membranes were incubated for 2 h at 22°C with anti-CPD antibodies. To normalize for DNA content, the membrane was blocked for 1 h at 22ºC in 2% (w/v) BSA diluted in TBST and probed with anti-dsDNA antibodies diluted in TBST containing 2% (w/v) BSA. CPD and dsDNA signals were visualized with a VersaDoc-3000 imager (Bio-Rad Laboratories) and quantified with Quantity One software (Bio-Rad Laboratories).

2.13 Cdh1 silencing by RNA interference

Primary keratinocytes were transfected 2 d after plating onto 60-mm culture dishes with a non-targeting Silencer Negative Control siRNA #1 (AM4611, Ambion® ThermoFisher

Scientific, Waltham, MA) or Cdh1-targeting siFZR1 siGENOME SMARTpool (M-

065289-00, Dharmaco™ GE Healthcare, Lafayette, CO) using Lipofectamine 2000

transfection reagent according to manufacturer's instruction. Specifically, 15 µl of

Lipofectamine 2000 (1 µg/µl stock) mixed with 1 ml of serum-free (SF) and Ca2+-free

EMEM were combined with 200 pmoles of siRNA that had been mixed with 1 ml SF and

Ca2+-free EMEM. The samples were vortexed and incubated at 22ºC for 5 min. This

solution was then used to replace the culture medium of the keratinocytes, which had

been previously rinsed once with Ca2+-free PBS. The cells were incubated at 37ºC for 4 h prior to the addition of 3 ml of keratinocyte growth medium (Ca2+-free EMEM supplemented with growth additives and 8% (v/v) FBS pre-treated with chelex resin).

The cells were cultured for 24 h, at which time they were washed twice with Ca2+-free

PBS and cultured with fresh growth medium for 72 h. The cells were transfected with

vectors encoding V5-tagged E2F1, as described in section 1.7.3, and cultured for 21 additional hours. Keratinocytes were then treated with CHX as described in section 1.7.3 prior to lysis and protein analysis (Section 1.9).

2.14 Statistical analyses

Data are expressed as mean ± SEM or mean ± SD, as indicated in individual experiments.

Student’s t-tests were used for comparisons between two groups. One-way analysis of variance (ANOVA) with Newman-Keuls post-hoc test was used for comparisons among multiple groups and one variable. Two-way ANOVA with Bonferroni post-hoc test was used for comparisons among multiple groups and two variables. Statistical analyses were conducted with GraphPad Prism software (version 5.00 for Windows, San Diego, CA), and significance was set at P<0.05. The results shown in all figures are representative of

3-8 experiments, each conducted with independent cell isolates.

Chapter 3 3 Results 3.1 Interaction of hHR23A and E2F1

The mammalian orthologues of yeast Rad23 (HR23) proteins play dual roles in DNA

NER and modulation of ubiquitin-mediated protein degradation (Dantuma et al., 2009).

To begin to address the role of these proteins in the epidermis, their expression levels in

mouse tissues were first assessed. Quantitative polymerase chain reaction (qPCR)

analysis revealed the presence of transcripts encoding mHR23A and mHR23B in the

epidermis and the dermis, at levels comparable to those found in a variety of other tissues

(Figure 3.1A). Further, mHR23A and mHR23B mRNA species were also detected in

primary cultured epidermal keratinocytes, irrespective of whether they were

undifferentiated or had been induced to differentiate (Figure 3.1A). mHR23A protein was

detected at low levels in mouse epidermis and in cultured keratinocytes, whereas it was

substantially more abundant in various human epithelial cell types and mouse tissues,

including the dermis (Figure 3.1B, 3.1C).

hHR23A and mHR23A proteins participate in NER processes through the formation of

multiprotein complexes that recognize sites of DNA damage (Dantuma et al., 2009). The

ability of the E2F1 transcription factor to recognize sites of DNA damage induced by UV

radiation and to promote NER (Guo et al., 2010) prompted me to investigate the

possibility that E2F1 and hHR23A might associate. To this end, exogenous V5-tagged

E2F1 together with FLAG-tagged hHR23A were expressed in primary mouse

Figure 3.1: Expression of mHR23 and hHR23 proteins.

A. mRNA was isolated from primary undifferentiated (Undiff. KT) or differentiated (Diff. KT) 3-day keratinocyte cultures, or from the indicated tissues harvested from 2-d- old CD-1 mice. mHR23A and mHR23B mRNA levels were assessed by qPCR, and normalized to the abundance of Rpl16 and Rps29 transcripts. The results are expressed as the mean + SEM (n=3), relative to mRNA abundance in undifferentiated keratinocytes, which is set to 1.0. The asterisks indicate P<0.05, relative to abundance in undifferentiated keratinocytes (ANOVA). B, C. The abundance of mHR23A or hHR23A was determined in protein lysates from the tissues harvested in (A) or from the indicated cultured cells. Glyceraldehyde-3-phosphatase (GAPDH) was used to normalize for protein loading. MEK and HEK indicate, respectively, primary cultures of undifferentiated mouse and human epidermal keratinocytes. Blot exposure times and the amount of protein on the blots (in μg) are indicated, to better illustrate differences in protein abundance (n=3).

keratinocytes that had been maintained undifferentiated, or had been induced to differentiate by 24 h of culture in medium containing 1 mM Ca2+ (High-Ca2+ medium).

When hHR23A immune complexes were isolated, V5-tagged E2F1 was detected, irrespective of the differentiation status of the keratinocytes (Figure 3.2A). Reciprocally, hHR23A fused to green fluorescent protein (GFP) and hemagglutinin (HA) tags was readily detected in V5-E2F1 immune complexes (Figure 3.2B). The larger molecular mass of the GFP-tagged hHR23A species avoided the interference of IgG heavy chains in this analysis. It was also found that endogenous E2F1 could associate with hHR23A proteins (Figure 3.2C), but I was unable to detect mHR23A in E2F1 immune complexes, likely due to its very low abundance in primary keratinocytes. hHR23 proteins can function as scaffolds between ubiquitylated substrates and the proteasome. To determine if hHR23A interactions with E2F1 required ubiquitylation of the latter, I examined if bacterially produced hHR23A and glutathione (GST)-tagged

E2F1 were able to interact. I readily detected hHR23A in GST-E2F1 immune complexes, indicating that interactions between these two proteins do not require the presence of post-translational modifications (Figure 3.2D). These observations do not exclude the possibility that binding of hHR23A to E2F1 also occurs through ubiquitylated residues on the latter. Bacterially produced hHR23A or GST-E2F1 could also interact with exogenously expressed V5-tagged E2F1 or HA-tagged hHR23A from mammalian cell lysates respectively (Figure 3.2E, 3.2F).

Figure 3.2: Association of hHR23A and E2F1.

A, B. Primary keratinocytes were transfected with vectors encoding V5-tagged E2F1 with or without FLAG-tagged hHR23A or HA- and GFP-tagged hHR23A and cultured for 24 h after transfection in Low Ca2+ or in High Ca2+ medium, to induce differentiation. Cell lysates were prepared and immunoprecipitated with anti-FLAG or anti-V5 antibodies, as indicated, or an irrelevant IgG. Immune complexes were resolved by denaturing gel electrophoresis, transferred to membranes, and the blots were probed with the indicated antibodies. Samples of lysates used for immunoprecipitation show expression levels of exogenous proteins. γ-Tubulin was used to normalize for protein loading, and the asterisks indicate a non-specific band. n=3. C. Lysates prepared from transfected keratinocytes as in (A, B), were immunoprecipitated with anti-E2F1 antibodies, to isolate endogenous and/or exogenous E2F1 immune complexes. Replicate lysate samples were also analyzed to show expression levels of endogenous and exogenous proteins. n=3. D. Bacterially produced GST, GST-E2F1, as well as His- and FLAG-tagged hHR23A (2 μg each) were used in immunoprecipitation experiments with anti-GST or anti-His antibodies, as indicated. The immune complexes were further analyzed by immunoblot, using anti-FLAG or anti-E2F1 antibodies. The lanes labelled “Input” contain 100 ng each of the indicated recombinant proteins. n=3. E. GST fusion E2F1 from bacterial extracts were incubated with cell lysates prepared from undifferentiated keratinocytes transfected with HA-tagged hHR23A for 16 h at 4ºC. The protein mixtures were subjected for GST pull down and immunecomplexes were analyzed by immunoblot with anti-HA antibodies, to detect hHR23A. n=3. F. His fusion FLAG-tagged hHR23A from bacterial extracts were incubated with cell lysates prepared from keratinocytes transfected with V5-tagged E2F1. The protein complexes were subjected for His pull down and analyzed by immunoblot using anti-V5 antibodies, to detect E2F1. Samples of the lysates were also analyzed by immunoblot with the indicated antibodies. n=3.

3.2 Modulation of hHR23A subcellular localization by E2F1

Given that hHR23 proteins and their orthologues are involved in both proteasomal degradation and DNA repair, the subcellular localization of exogenously expressed hHR23A in undifferentiated keratinocytes was investigated. I observed that hHR23A exhibited cytoplasmic distribution in about 90% of the cells (Figure 3.3A). This pattern markedly differs from that observed in transformed HeLa cells, which show nuclear concentration of hHR23A (Figure 3.3B).

In many cell types, including undifferentiated keratinocytes, E2F1 is predominantly found in the nucleus, although it exhibits nucleocytoplasmic shuttling (Ivanova et al.,

2007) (Figure 3.3A). Significantly, the joint expression of hHR23A and E2F1 resulted in the co-localization of both proteins in the nucleus of almost all keratinocytes (Figure

3.3A). Induction of differentiation in keratinocytes promotes E2F1 nuclear export and degradation (Ivanova and Dagnino, 2007). Thus, it was of interest to investigate if the presence of cytoplasmic E2F1 in these cells was associated with loss of nuclear pools of hHR23A. Keratinocytes were cultured in medium containing either 0.12 mM or 1.0 mM

Ca2+, and found that in both culture conditions, when E2F1 was cytoplasmic, hHR23A was also excluded from the nucleus (Figure 3.3C). These observations are consistent with the concept that E2F1 is capable of modulating hHR23A subcellular localization and recruitment into the nucleus, possibly mediated via complex formation between these two proteins.

Figure 3.3: Modulation of hHR23A subcellular distribution by E2F1.

A. Primary keratinocytes or B. HeLa cells were transfected with vectors encoding the indicated proteins, and 24 h later were processed for immunofluorescence microscopy using the indicated antibodies. The histograms represent the average fraction of cells with nuclear (N) or cytoplasmic (C) protein distribution + SEM (n=4). At least 100 transfected cells per condition were scored in each experiment. The asterisks indicate P<0.05, relative to cells transfected with single vectors (ANOVA). F-actin and DNA were visualized, respectively, with phalloidin and Hoechst 33342. Bar, 25 μm. C. Subcellular distribution of hHR23A and E2F1 in differentiated keratinocytes. Keratinocytes were transfected with vectors encoding the indicated proteins, and 4 h after transfection cells were induced to differentiate by culture in growth medium with the indicated Ca2+ concentration. The cells were processed for immunofluorescence microscopy using anti- V5 and anti-HA antibodies to detect, respectively, E2F1 and hHR23A. DNA was visualized with Hoechst 33342. Bar, 25 μm.

3.3 hHR23A domains involved in binding to E2F1

To begin to understand the biological significance of E2F1 binding to hHR23A, I next

undertook an analysis of the protein domains involved in these interactions. As a scaffold

protein, hHR23A has several functionally important domains that mediate interactions

with ubiquitin and/or various ubiquitin-unrelated proteins (Dantuma et al., 2009). To identify which of those regions are important for E2F1 binding, I exogenously expressed several GFP- and HA-tagged hHR23A deletion mutants (Figure 3.4A). The use of GFP-

tagged hHR23A allowed for the confirmation that the GFP moiety did not affect the

ability of hHR23A to bind E2F1 (Figure 3.4B, 3.4C), and allowed for efficient expression

of small hHR23A domains for analysis. The N-terminal ubiquitin-like domain (UbL) in hHR23A mediates binding to the proteasome, and is important for protein-protein interactions. The deletion of this domain (∆UbL hHR23A) did not affect the ability of hHR23A to interact with E2F1 (Figure 3.4B, 3.4C). hHR23A has two ubiquitin- associated domains, UBA1 and UBA2, which mediate interactions with ubiquitin moieties conjugated to various proteins. When I exogenously expressed two hHR23A fragments corresponding to UBA1 or UBA2, I observed that only the former was able to associate with E2F1 (Figure 3.4B). Thus, the contribution of UBA2 to hHR23A binding to E2F1 is likely negligible, if any. In a complementary approach, I also investigated the consequences of deleting UBA1 or UBA2 in the context of the entire hHR23A protein

(hHR23A ∆UBA1 or ∆UBA2), and observed efficient binding of these mutants to E2F1

(Figure 3.4B). This suggests that, in addition to UBA1, other hHR23A regions likely contribute to its association with E2F1. hHR23 proteins bind the XPC via their XPC

Figure 3.4: hHR23A domains involved in E2F1 binding.

A. Schematic of GFP- and HA- doubly tagged hHR23A proteins analyzed for their ability to bind E2F1. B. and C. Primary keratinocytes were transfected with vectors encoding V5-tagged E2F1 and the indicated hHR23A proteins. Twenty-four hours after transfection, cell lysates were prepared and hHR23A immune complexes were isolated and analyzed by immunoblot with the indicated antibodies, and γ-Tubulin was used to normalize for protein loading. A portion of the lysates from cells co-expressing wild type (WT) hHR23A and V5-tagged E2F1 were also used for immunoprecipitation with an unrelated IgG control. n=4.

domain (Dantuma et al., 2009). To determine if the XPC region contributes to E2F1

binding, an hHR23A mutant composed of only the XPC and UBA2 domains was

expressed (hHR23A XPC-UBA2). I reasoned that this mutant would provide information

on the role of the XPC domain in this process, given our previous observation that UBA2

does not bind E2F1. I found that this hHR23A form was still able to bind E2F1 fairly

efficiently (Figure 3.4B, 3.4C). However, the XPC region appears to be sufficient, but not

indispensable for E2F1 binding, given that an hHR23A mutant lacking these sequences

(∆XPC) also associates with E2F1 (Figure 3.4C). Together, my observations implicate

the UBA1 and XPC domains in the association of hHR23A with E2F1.

I undertook a similar analysis to determine the regions in E2F1 involved in binding to

hHR23A. The N-terminal domain in E2F1 mediates protein-protein interactions, such as

binding to cyclin A, and contains the nuclear localization and export domains responsible

for nucleocytoplasmic shuttling (Ivanova et al., 2007). A mutant, constitutively cytoplasmic E2F1 protein lacking these sequences (E2F1 126-437) associated with

hHR23A (Figure 3.5A, 3.5B), indicates that nuclear localization of E2F1 is not required for this interaction. The C-terminus in E2F1 contains the transcriptional activation domain, and mediates protein-protein interactions with other proteins, including those in the retinoblastoma family (reviewed in (Meng and Ghosh, 2014)), but it was also dispensable for hHR23A binding (Figure 3.5B). Similarly, E2F1 proteins incapable of both cyclin A and retinoblastoma binding also formed complexes with hHR23A (Figure

3.5B). Of note, E2F1 contains 14 lysine residues that can potentially be modified by

Figure 3.5: Regions in E2F1 involved in binding to hHR23A.

A. Schematic of V5-tagged E2F1 proteins analyzed. B. Primary keratinocytes were transfected with vectors encoding wild type hHR23A and the indicated V5-tagged E2F1 proteins. Twenty-four hours after transfection, cell lysates were prepared and hHR23A immune complexes were isolated using anti-HA antibodies. The immune complexes were resolved by denaturing gel electrophoresis, and analyzed with the indicated antibodies. γ- Tubulin was used to normalize for protein loading. Lysates from cells co-expressing E2F1 and wild type (WT) hHR23A were used for immunoprecipitations using anti-HA antibodies, or an unrelated, control IgG. n=3.

100

101

ubiquitylation and, given that the ubiquitin-binding UBA1 domain in hHR23A

participates in the interaction between these two proteins, it is conceivable that no single

region in E2F1 is indispensable.

3.4 Inhibition of E2F1 degradation by hHR23 proteins

E2F1 ubiquitylation includes K11-, K48- and K63-linked polyubiquitin chains, which can

promote its proteasomal degradation (Wang et al., 2015). To begin to characterize the

functional significance of the interactions between hHR23 proteins and E2F1,

keratinocytes were exogenously expressed with V5-tagged E2F1 and HA-tagged wild

type ubiquitin, in the presence or absence of hHR23 proteins. Next, HA-containing

immune complexes were isolated and investigated therein the presence of E2F1. This

approach examined the effects of hHR23 on the abundance of ubiquitylated E2F1. In the

absence of exogenous hHR23 proteins, and consistent with previous reports (Ivanova et

al., 2006), low levels of ubiquitylated E2F1 in both undifferentiated and differentiated

keratinocytes were detected, likely because it efficiently undergoes proteasomal

degradation (Figure 3.6A, 3.6B). The presence of either hHR23A or hHR23B

substantially increased the abundance of ubiquitylated E2F1, irrespective of the

differentiation status of the keratinocytes. Thus, hHR23 proteins appear to protect E2F1

from degradation, although this effect does not appear to involve interference with E2F1

ubiquitylation per se (Figure 3.6A, 3.6B). Next, the effect of proteasomal inhibition on

modulation by hHR23A of polyubiquitylated E2F1 was determined. To this end, I

exogenously expressed V5-tagged E2F1 together with HA-tagged wild type ubiquitin and

FLAG-tagged hHR23A as before, and assessed the effect of proteasomal inhibition by

Figure 3.6: Modulation of ubiquitylated E2F1 degradation by hHR23 proteins.

A. Primary undifferentiated keratinocytes were transfected with vectors encoding V5- tagged E2F1 and HA-tagged ubiquitin, in the presence or absence of hHR23-encoding vectors, as indicated. The cells were cultured for 24 h in Low Ca2+ (undifferentiated keratinocytes) or High Ca2+ medium (differentiated keratinocytes) in the absence of MG132. Cell lysates were prepared and ubiquitylated proteins were immunoprecipitated with anti-HA antibodies. HA immune complexes were analyzed with anti-V5 antibodies, to detect ubiquitylated E2F1 in the immune complexes. The asterisk indicates a lower mobility non-specific band. n=3. B. Primary undifferentiated keratinocytes were transfected with a vector encoding V5-tagged E2F1 and either wild type (WT) ubiquitin, or a ubiquitin mutant lacking all Lys residues (K0), in the presence or absence of FLAG- tagged hHR23A. Keratinocytes were treated with vehicle (dimethylsulfoxide) or MG132 (10 μM) for 6 h, and cell lysates were prepared, immunoprecipitated with anti-HA antibodies and analyzed for the presence of ubiquitylated V5-E2F1, as in (A). n=3.

103

104

MG132 on the abundance of E2F1. A robust increase in polyubiquitylated E2F1 in the presence of hHR23A was slightly enhanced in MG132-treated cells (Figure 3.6B),

indicating that exogenous expression of hHR23A does not impair proteasomal activity

but interferes with the ability of the proteasome to target E2F1 for degradation.

The UBA domains of hHR23A exhibit maximum affinity for chains composed of 4-6

ubiquitin residues (Raasi et al., 2004). Cellular proteins can be modified by polyubiquitin

chains linked through isopeptide bonds between the terminal glycine (G76) in ubiquitin

and one of the seven Lys residues on another ubiquitin moiety. To determine whether the

stabilizing effect of hHR32A required the presence of polyubiquitin chains on E2F1, I co-

expressed V5-tagged E2F1, FLAG-tagged hHR23A and an HA-tagged mutant ubiquitin form lacking all lysine residues (K0). This mutant is presumably incapable of forming polyubiquitin chains, although it can still contribute to monoubiquitylation (Lim et al.,

2005). In the absence of MG132, I found negligible E2F1 immunoreactivity in ubiquitin-

K0 immunoprecipitates (Figure 3.6B). Significantly, proteasomal inhibition was associated with the presence of abundant poly-ubiquitylated E2F1 species (Figure 3.6B).

These observations suggest that hHR23A preferentially modulates the degradation of poly-ubiquitylated E2F1, although monoubiquitin modification of the latter on one or more Lys residues likely also takes place in keratinocytes, without precluding its proteasomal degradation.

The outcomes of hHR23 modulation of proteasome-mediated proteolysis are complex. In some cases, they deliver polyubiquitylated substrates to the proteasome for destruction

(Verma et al., 2004). In others, they impede the interaction of bound proteins with the

105

proteasome, thus suppressing their degradation (Blount et al., 2014). First, I assessed the effect of increasing amount of hHR23A on the stability of E2F1 and showed that E2F1 levels gradually increased with increasing amount of hHR23A (Figure 3.7A). To further confirm that hHR23 negatively modulates E2F1 turnover, I used cycloheximide chase assays to determine E2F1 apparent half-life (t½) in undifferentiated keratinocytes expressing wild type or mutant hHR23A proteins (Figure 3.7B). The apparent t½ of V5-

E2F1 was 77±8 min, which was increased 3- and 2-fold, respectively, in the presence of similar levels of wild type hHR23A and hHR23B. Thus, although both hHR23 forms are able to stabilize E2F1, the effect of hHR23A would appear to be more pronounced.

Similar to full-length hHR23A, ∆UbL GFP-hHR23A, which can efficiently bind E2F1, increased its apparent t½ 3-fold to 185±5 min. In contrast, expression of UBA2 GFP- hHR23A did not significantly alter E2F1 t½, consistent with the notion that the stabilizing effect of hHR23A requires its association with E2F1, and that UBA2 does not associate with this transcription factor. Although the UBA2 domain does not participate in binding to or stabilizing E2F1, it is critical for the ability to hHR23A to escape proteasomal degradation itself, and its deletion markedly increases hHR23A turnover

(Heinen et al., 2011). Consistent with these characteristics, I observed that, whereas the apparent t½ of wild type hHR23A was > 6 h, the apparent t½ of ∆UBA2 GFP-hHR23A was only about 4 h (Figure 3.7B). Further, ∆UBA2 GFP-hHR23A was without effect on

E2F1 apparent t½, suggesting that insufficient pools of the former protein may have been available to protect exogenous E2F1 from degradation.

Figure 3.7: Modulation of E2F1 t½ by hHR23 proteins.

A. Undifferentiated keratinocytes were transfected with vectors encoding V5-tagged E2F1 with the indicated DNA concentration of FLAG-tagged hHR23A. Cell lysates were prepared 24 h later and analyzed by immunoblot with the indicated antibodies. The histograms represent quantification of E2F1 proteins (mean + SEM, n=3), relative to E2F1 abundance in the absence of exogenous hHR23A, which is set at 100%. The asterisks indicate P<0.05 relative to E2F1 levels in cells cultured in Low Ca2+ medium in the absence of exogenous hHR23A proteins (ANOVA). B. Primary undifferentiated keratinocytes were cotransfected with vectors encoding V5-tagged E2F1 and the indicated FLAG- or HA-tagged hHR23 proteins. Twenty-four hours after transfection, cycloheximide (CHX, 100 μg/ml, final) was added to the culture medium, and cell lysates were prepared at the indicated time intervals thereafter. The lysates were analyzed by immunoblot with the indicated antibodies, and γ-Tubulin was used to normalize for protein loading. The graphs represent the mean ± SD of E2F1 levels quantified in replicate immunoblots generated from lysates prepared from three independent cell isolates, and were used to calculate E2F1 apparent t½ (mean ± SD, n=3) in the presence of each hHR23 protein, as summarized in the chart. The asterisks indicate P<0.05 relative to E2F1 t½ in the absence of any exogenous hHR23 protein (ANOVA).

107

108

109

3.5 Modulation of DNA damage repair by hHR23A and E2F1

Following DNA damage, hHR23A participates in NER processes, whereas E2F1 is

stabilized, either promoting apoptosis or stimulating DNA repair (Biswas et al., 2014).

Given that a major source of DNA damage in keratinocytes is UV radiation, I examined

the responses of these two proteins to UV-induced DNA damage. I observed that

endogenous E2F1 levels are increased by UVB radiation or by expression of hHR23A,

and the presence of exogenous hHR23A further increased E2F1 levels in UVB-treated keratinocytes (Figure 3.8A). The kinetics of exogenous E2F1 and hHR23A proteins were also examined following UVB radiation. A time course of post- irradiation was carried out at 0, 3 and 6 h. The protein levels of hHR23A did not change 6 h post-irradiation, whereas E2F1 showed increased protein levels at 6 h (Figure 3.8B). Interestingly, when

E2F1 and hHR23 were co-expressed, the expression levels of both proteins peaked at 3 hr and returned to control levels by 6 hr (Figure 3.8B). Next, I determined if hHR23A subcellular localization was regulated by DNA damage in primary keratinocytes, using filtered UV radiation coupled with immunofluorescence microscopy (Guo et al., 2010;

Mone et al., 2001). Cells were irradiated with UV light through a filter (3-µm pore size), and nuclear regions containing UV-induced photoproducts were identified with an antibody specific for CPD. UVB radiation penetrated through the entire filter, giving rise to CPD-positive nuclei, rather than discrete CPD-positive areas with DNA damage

(Appendix 1A). UVC radiation did not penetrate through the entire filter, and was transmitted down the nuclear DNA only through the pores (Appendix 1B). This allowed

Figure 3.8: Effect of UVB radiation on E2F1 levels in keratinocytes.

Primary keratinocytes expressing HA-tagged hHR23A and/or V5-tagged E2F1 was exposed to A. 250 J/m2 of UVB, and whole cell lysates were prepared 24 h after irradiation, n=3 or B. 100 J/m2 of UVB, and cell lysates were prepared at indicated time point post-irradiation to analyze levels of the indicated proteins. GAPDH levels were used to normalize for protein loading. The graphs in (B) represent the mean ± SD (n=3) of hHR23A (left) and E2F1 (right) levels quantified in replicate immunoblots generated from lysates prepared from three independent cell isolates.

111

112

detection of discrete CPD-positive DNA damaged foci. Therefore, I used light in the

UVC spectrum for these experiments. Although, UVC light does not reach the earth surface because it is absorbed by the ozone layer, both UVB and UVC radiation induce the same type of DNA damage and activate the same DNA repair pathways, making the use of UVC irradiation physiologically relevant. It was observed that UV treatment of keratinocytes induced cytoplasm-to-nucleus translocation of hHR23A (Figure 3.9). In the majority of cells expressing exogenous E2F1, E2F1 was substantially enriched in nuclear areas that exhibited CPD immunoreactivity (Figure 3.9). Significantly, when E2F1 and hHR23A were co-expressed, both proteins concentrated in CPD-positive regions, indicating that E2F1 promotes enrichment of hHR23A at sites of UV-induced DNA

damage in primary keratinocytes (Figure 3.9).

Next, I investigated if the increased localization of hHR23A to DNA photolesions in the

presence of E2F1 was associated with changes in DNA repair. To this end, I transfected

cells with vectors encoding hHR23A and/or E2F1, subjected the cells to 100 J/m2 UVB

radiation, and measured CPD abundance in genomic DNA (indicative of NER) as a

function of time. Because UVB treatment of transfected primary keratinocytes resulted in

variable proportions of transfected cell death among different cell isolates, I conducted

this experiment on HeLa cervical carcinoma cells, in which no such variability was

observed. CPD levels in control cells transfected with empty vector decreased by about

30%-40% of initial values 3 h after UVB treatment, and did not decrease substantially

further 6 h after UVB treatment (Figure 3.10A). CPD levels in cells expressing E2F1 or

hHR23A had decreased by about 20%-25% 6 h after UVB treatment. In contrast, in cells

Figure 3.9: Co-localization of E2F1 and hHR23 at DNA damage sites following UV radiation.

Keratinocytes were transfected with vectors encoding the indicated proteins, and 24 h later were subjected to UVC irradiation through 3-μm polycarbonate isopore filters, and cultured for 2 h. The cells were processed for immunofluorescence microscopy to detect cyclobutane pyrimidine dimers (CPD), V5-tagged E2F1 and HA-tagged hHR23A. Nuclear DNA was visualized with Hoechst 33342. Bar, 25 μm. n=3

114

Figure 3.10: Increased repair of UV-induced DNA damage by E2F1 and hHR23A.

A. HeLa or B. RPMI 7951 human melanoma cells transfected with plasmids encoding the indicated proteins were subjected to UVB irradiation (100 J/m2). Genomic DNA was isolated at the indicated times after UVB treatment, and analyzed for the presence of CPD as described in “Materials and Methods”. CPD levels were normalized to the amount of dsDNA. The histograms show the fraction of normalized CPD signal remaining after 3 or 6 h, relative to CPD 0.1 h after irradiation (set to 100%), for each transfection group. The results are expressed as the mean + SD (n=3), and the asterisks indicate P<0.05 (ANOVA).

116

117

118

co-expressing E2F1 and hHR23A, CPD levels steadily decreased by about 40% and 60%,

respectively, 3 h and 6 h following UVB exposure. Similar results were observed in

RPMI 7951 human melanoma cells (Figure 3.10B). Thus, the joint expression of E2F1

and hHR23A accelerates CPD removal and DNA repair following UVB damage.

3.6 Amino acid residues in E2F1 that contributes to its stabilization following drug-induced DNA damage

E2F1 accumulates in mouse embryonic fibroblasts (MEF) and cancer cell lines when

treated with various chemotherapeutic drugs (Lin et al., 2001). Several residues on E2F1

are essential for its stabilization following DNA damage. They include Ser31, Arg111,

Arg113, Lys117, Lys120, Lys125, Lys185, Ser403, and Thr433 (Cho et al., 2012;

Kontaki and Talianidis, 2010; Lin et al., 2001; Pediconi et al., 2003; Real et al., 2010;

Xie et al., 2011b). The importance of these residues has been established by site-directed

mutagenesis studies in transformed cell lines. However, their importance in primary cells

treated with DNA damaging agents has not been studied. Therefore, to better understand

the role of each residue on E2F1 in response to DNA damage, the following constructs

were made: a single mutant with Ser to Ala substitution on a.a. site 31 (S31A); a single mutant with Lys to Arg substitution on a.a. site 185 (K185R); a triple mutant with Lys to

Arg substitution on a.a. sites 117, 120, and 125 (KTR); a double mutant with Arg to Lys substitution on a.a. sites 111 and 113 (RDK); and a double mutant with Ser to Ala substitution on a.a. sites 403 and 433 (ST/A). The various E2F1 mutants were exogenously expressed in primary mouse keratinocytes and subsequently DNA double strand breaks was induced by treating the cells with 150 µM etoposide.

119

Consistent with previous studies on transformed cell lines and MEFs, the protein levels of

wild-type E2F1 increased and the E2F1-S31A mutant failed to accumulate following etoposide treatment in undifferentiated and differentiated keratinocytes (Figure 3.11).

The KTR triple E2F1 mutant also showed impaired induction following treatment with etoposide in differentiated keratinocytes (Figure 3.11). These results indicate that

modifications at Ser31, Lys117, Lys120, and Lys125 play an essential role in stabilizing

E2F1 in response to DNA damage not only in transformed cell lines but also in primary

mouse keratinocytes. However, in undifferentiated keratinocytes, the KTR triple E2F1

mutant protein levels increased following etoposide treatment (Figure 3.11). This

indicates that multiple mechanisms contribute to E2F1 induction in response to DNA

damage. On the other hand, the K185R, RDK, or ST/A E2F1 mutant increased following

etoposide treatment in undifferentiated and differentiated keratinocytes (Figure 3.11).

This indicates that these residues are not essential for E2F1 accumulation and

stabilization following DNA damage.

3.7 Identification of E2F1 amino acid residues required for nuclear export and stabilization in differentiated keratinocytes

Following Ca2+-induced keratinocyte differentiation, E2F1 is exported to the cytoplasm

and substitutions of Ser403 and Thr433 to Ala (ST/A) generated a mutant protein that is

defective in nuclear export and is more stable than WT-E2F1 (Ivanova and Dagnino,

2007; Ivanova et al., 2009). Whether other amino acid residues on E2F1 are required for

its nuclear export and stabilization is yet to be elucidated. First, residues shown to be

required for the interaction between E2F1 and pRb were examined. During the cell cycle,

Figure 3.11: Changes in E2F1 levels in response to etoposide.

Vectors encoding V5-tagged wild type (WT) E2F1 or the indicated mutants were transfected in undifferentiated keratinocytes. Fours after transfection, the cells were cultured in Low (-) or High (+) Ca2+ medium, and 24 h later, they were incubated in the absence or presence of etoposide (150 µM). Cell lysates were prepared 8 h later and analyzed by immunoblot, using anti-V5 antibodies or GAPDH as loading control. The histograms represent normalized densitometric quantification of each E2F1 protein (mean + SEM, n=3), and are expressed as percentage of a given E2F1 form relative to its abundance in Low Ca2+ medium, which is set at 100%. The asterisks indicate P<0.05, and # indicate P<0.05 relative to the corresponding E2F1 mutant protein levels in cells cultured in Low Ca2+ medium in the absence of etoposide (ANOVA).

121

122

E2F1 is specifically regulated by associating with pRb. The interaction between pRb and

E2F1 protects E2F1 from ubiquitylation; therefore, events that affect the association

between pRb and E2F1 could regulate E2F1 stability. Phosphorylation at Ser332 and

Ser337 is required for the dissociation of E2F1 from pRb (Fagan et al., 1994), whereas

phosphorylation at Ser375 favors pRb binding to E2F1 (Peeper et al., 1995). To test

whether the association with pRb plays a role in E2F1 nuclear export and degradation,

three E2F1 single mutants with serine to alanine substitutions at sites 332 (S332A), 337

(S337A), or 375 (S375A) were generated and expressed in keratinocytes. These three mutant E2F1 showed no apparent defect in nuclear export and were efficiently degraded in differentiated keratinocytes in a similar manner as wild-type E2F1 (WT-E2F1), which is significantly different from the ST/A mutant E2F1 protein (Figure 3.12, 3.13). These findings indicated that the residues known to promote the association or dissociation between E2F1 and pRb may not be essential for the Ca2+-induced nuclear export and degradation of E2F1 in keratinocytes.

As mentioned earlier, during DNA damage, E2F1 is post-translationally modified, which contributes to its increased stability. Thus, residues involved in the stabilization of E2F1 were examined. The triple KTR E2F1 mutant showed increased cytoplasmic localization

(Figure 3.12) and decreased E2F1 abundance following Ca2+-induced keratinocyte differentiation (Figure 3.11, 3.13), similar to that observed with WT-E2F1. This indicates that these three lysine residues are dispensable for regulation of E2F1 abundance upon keratinocyte differentiation. Similarly, S31A, K185R, RDK and S364A E2F1 mutants also showed increased nuclear export (Figure 3.12) and decreased E2F1 protein levels

Figure 3.12: Ser403 and Thr433 are required for E2F1 nuclear export during keratinocyte differentiation.

Undifferentiated keratinocytes were transfected with vectors encoding V5-tagged wild type (WT) or the indicated E2F1 mutant proteins. Four hours after transfection, the culture medium was replaced with Low or High Ca2+ medium, and the cells were processed for immunofluorescence microscopy 24 h later, using anti-V5 antibodies. DNA was visualized with Hoechst 33342. The photomicrographs represent three replicas, and 100 cells per experiment were analyzed. The values in the histograms represent the percentage of cells (mean + SEM, n=5) that exhibited nuclear (N, green bar) or cytoplasmic (C, yellow bar) E2F1 distribution. The asterisks indicate P<0.05 relative to values in the corresponding subcellular compartments in Low Ca2+ medium (ANOVA). Bar, 16 μm. KTR: K117R/K120R/K125R; RDK: R109K/R113K; ST/A: S403A/T433A.

124

125

Figure 3.13: Ser403 and Thr433 mediate E2F1 degradation during keratinocyte differentiation.

Vectors encoding V5-tagged wild type (WT) or the indicated E2F1 mutants were transfected in undifferentiated keratinocytes. Twenty-four hours after transfection, cells were treated with cycloheximide (CHX, 100 μg/ml) and 30 min later the culture medium was replaced with Low or high Ca2+ medium supplemented with CHX. Cell lysates were prepared 2.5 h later and analyzed by immunoblot for the indicated proteins, using anti-V5 antibodies, or γ-Tubulin as loading control. The histograms represent normalized densitometric quantification of each E2F1 protein (mean + SEM, n=3), and are expressed as percentage of a given E2F1 form relative to its abundance in Low Ca2+ medium, which is set at 100%. The asterisks indicate P<0.05 (ANVOA). KTR: K117R/K120R/K125R; RDK: R109K/R113K; ST/A: S403A/T433A.

127

128

following Ca2+ induction, as seen with WT-E2F1 (Figure 3.11, 3.13). Therefore, these

data indicate that residues involved in the regulation of E2F1 levels following DNA

damage are not major contributors to E2F1 stability, whereas Ser403 and Thr433 are

involved in E2F1 nuclear export and degradation following Ca2+-induced differentiation.

3.8 Nuclear export is not the only mechanism required for E2F1 ubiquitylation and degradation following keratinocyte differentiation

E2F1 nuclear export and ubiquitylation are both required for the degradation of E2F1

following Ca2+-induced keratinocyte differentiation. When differentiated keratinocytes

are exposed to a specific inhibitor of CRM1-mediated nuclear export, LMB, E2F1

accumulates in the nucleus and its stability increases (Ivanova and Dagnino, 2007). This indicates that E2F1 nuclear export is required for E2F1 degradation following keratinocyte differentiation. However, whether nuclear export could inhibit E2F1 ubiquitylation has yet to be elucidated. First, the subcellular compartment of E2F1 ubiquitylation was assessed by exogenously expressing V5-tagged E2F1 and HA-tagged

wild type ubiquitin in keratinocytes. Subsequently, cells were treated with ethanol or

LMB to prevent nuclear export and lysates were harvested from whole cell, nuclear, and

cytoplasmic fractions. HA-containing immune complexes were isolated and the presence

of E2F1 was investigated. In differentiated keratinocytes, the abundance of ubiquitylated

E2F1 was indistinguishable in the presence or absence of LMB (Figure 3.14). Further, the

presence of polyubiquitylated E2F1 species was detected in both nuclear and cytoplasmic

fractions (Figure 3.14). These results indicate that nuclear export is not required for E2F1

ubiquitylation, and the ubiquitin enzymes that promote E2F1 ubiquitylation are present in

Figure 3.14: E2F1 is polyubiquitylated in the nucleus and in the cytoplasm when keratinocytes are induced to differentiate with Ca2+.

Undifferentiated keratinocytes were transfect with vectors encoding V5-tagged wild type (WT) E2F1 and HA-tagged ubiquitin. Twenty-four hours later, MG132 (10 µM) was added to the culture medium, followed 3 h later by leptomycin B (LMB, 10 ng/ml) or ethanol (vehicle). Cultures were incubated in the presence of both drugs for 30 min, at which time the Ca2+ concentration was adjusted to 1.0 mM. Lysates containing whole- cell (T), nuclear (N), and cytoplasmic (C) fractions were prepared 2.5 h later, HA- ubiquitin immunecomplexes were isolated, resolved by denaturing gel electrophoresis, and analyzed with anti-V5 antibodies to detect ubiquitylated V5-tagged E2F1. Antibodies against β-Tubulin and Lamin A/C were used to verify the purity of the fractionated extracts and as loading controls. n=3.

130

131

the nucleus and in the cytoplasm. The increase in stability of the ST/A mutant E2F1

protein could be due to alterations in E2F1 ubiquitylation. Therefore, the importance of

these residues for E2F1 ubiquitylation was examined. Consistent with previous accounts

(Ivanova and Dagnino, 2007), the abundance of polyubiquitylated WT- E2F1 species was

increased following Ca2+ induction (Figure 3.15). Interestingly, the ST/A E2F1 mutant

was polyubiquitylated in undifferentiated keratinocytes and differentiation increased the

abundance of the polyubiquitylated ST/A E2F1 mutant species (Figure 3.15). This

indicates that ST/A is still ubiquitylated but not degraded as seen with the WT-E2F1.

The increase in stability of the ST/A mutant suggested that nuclear export is essential for

degradation in differentiated keratinocytes (Ivanova et al., 2009). If nuclear export is the

only mechanism involved in E2F1 degradation, it is predicted that sequestrating the ST/A

mutant E2F1 to the cytoplasm should induce its degradation in differentiated

keratinocytes. To test this, an ST/A mutant containing a.a. encoding for the NES present

in Simian Virus 40 (SV40) large T antigen, which directs CRM1-mediated nuclear

export, was generated. The addition of NES to wild-type E2F1 (WT-E2F1-NES)

produced a protein that is 100% cytoplasmic in both undifferentiated and differentiated

keratinocytes (Figure 3.16A), and its protein levels decreased upon Ca2+-induced differentiation (Figure 3.16B). Similarly, the addition of NES sequences to E2F1 ST/A

(ST/A-NES) also produced a protein that is 100% cytoplasmic in both undifferentiated and differentiated keratinocytes (Figure 3.16A). However, the protein levels of ST/A-

NES remained the same in differentiated keratinocytes (Figure 3.16B). This indicates nuclear export is not the only mechanism involved in the turnover of E2F1 in differentiated keratinocytes.

Figure 3.15: Ubiquitylation of E2F1 is not affected by mutation of E2F1 at Ser403 and Thr433 in keratinocytes.

Undifferentiated keratinocytes were transfected with vectors encoding HA-tagged ubiquitin and V5-tagged wild type (WT) or ST/A E2F1. Four hours after transfection, the cultured medium was replaced with Low (-) or High (+) Ca2+ medium, and 24 h later, they were incubated in the presence of MG132 (10 µM). Cell lysates were prepared 6 h later, HA or unrelated IgG immunecomplexes were isolated and analyzed by immunoblot with anti-V5 antibodies. The abundance of exogenous E2F1 protein in lysates was analyzed by immunoblot, using γ-Tubulin as loading control. The asterisk indicates IgG heavy chain. n=5.

133

Figure 3.16: Effect of constitutive nuclear export on E2F1 abundance in differentiating keratinocytes.

A. Undifferentiated keratinocytes were transfected with vectors encoding V5-tagged wild type (WT) or the indicated E2F1 mutant proteins. Four hours after transfection, the culture medium was replaced with Low or High Ca2+ medium, and the cells were processed for immunofluorescence microscopy 24 h later, using anti-V5 antibodies. DNA was visualized with Hoechst 33342. The photomicrographs are representative of three indepenent cell isolates, and 100 cells per experiment were analyzed. Bar, 16 μm. n=3. B. 2+ Cells transfected as described in (A) were cultured for 24 h in Low Ca medium. Cycloheximide (100 μg/ml) was added, and 30 min later the cells were cultured for 2.5 h in Low or High Ca2+ medium supplemented with cycloheximide. Cell lysates prepared and analyzed by immunoblot with antibodies against V5 or γ-Tubulin used as loading control. The histograms represent normalized densitometric quantification of E2F1 protein level, and are expressed as the percentage of each E2F1 protein (mean + SEM, n=3), and are expressed as the percentage of a given E2F1 form relative to its abundance in Low Ca2+ medium, which is set to 100%. The asterisks indicate P<0.05. NES: nuclear export sequence.

135

136

3.9 Role of E2F1 pseudo-phosphorylation at Ser403 and Thr433

To further understand the role of E2F1 Ser403 and Thr433, three E2F1 mutants with serine and/or threonine substituted for aspartic acid at sites 403 and/or 433 (S403D,

T433D, and ST/D) were generated and expressed in keratinocytes. The E2F1 S403D,

T433D, and ST/D proteins are phospho-mimetic mutants, where the aspartic acid adds a negative charge, which mimics phosphorylation of a protein. First, their subcellular localization was assessed in the absence or presence of Ca2+. As shown in Figure 3.15,

WT-E2F1 shows nuclear localization in 90% of undifferentiated keratinocytes and Ca2+

treatments leads to a change in E2F1 localization to the cytoplasm in 40% of cells. The

S403D, T433D, and STD E2F1 mutants also showed 50-55% of cells with cytoplasmic

localization in differentiated keratinocytes compared to only 20% of cells with

cytoplasmic localization in undifferentiated keratinocytes (Figure 3.17). The result

suggests that the presence of Ca2+ contributes to the nuclear export of E2F1 and phosphorylation at Ser403 and Thr433 is required for this to occur.

To further characterize the contribution of Ser403 and Thr433 to overall E2F1 phosphorylation in keratinocytes, the mobility pattern in lysates from undifferentiated and differentiated keratinocytes of WT-E2F1, S403A, T433A, ST/A, S403D, T433D, and

ST/D was examined. The results showed at least three species of E2F1 with different mobility, which collapsed into a single band when treated with λ phosphatase in lysates from both undifferentiated and differentiated keratinocytes (Figure 3.18). This indicates that E2F1 is phosphorylated in keratinocytes. The S403A and S403D E2F1 mutants had a single predominant species that migrated at the same level as the unphosphorylated form

Figure 3.17: Subcellular localization of pseudophosphorylated E2F1 mutant proteins in keratinocytes.

Undifferentiated keratinocytes were transfected with vectors encoding V5-tagged wild type (WT) or the indicated E2F1 mutant proteins. Four hours after transfection, the culture medium was replaced with Low or High Ca2+ medium and processed for immunofluorescence microscopy 24 h later, using anti-V5 antibodies. DNA was visualized with Hoechst 33342. The values in the histograms represent the precentage of cells (mean + SEM, n=3) that exhibited nuclear (N) or cytoplasmic (C) E2F1 distribution. The asterisks indicate P<0.05 relative to values in the corresponding subcellular compartments in Low Ca2+ medium (ANOVA).Bar, 16 μm.

138

Figure 3.18: Ser403 and Thr433 are critical contributors to E2F1 phosphorylation.

Undifferentiated keratinocytes were transfected with vectors encoding V5-tagged wild type (WT) or the indicated E2F1 mutant proteins. Four hours after transfection, the culture medium was replaced with Low (Top) or High (Bottom) Ca2+ medium. After 24 h, cell lyates were prepared and incubated in the presence or absence of λ phosphatase (4 U/mg lysate protein), resolved by denaturing polyacrylamide gels, and analyzed by immunoblot, with antibodies against V5 or γ-Tubulin, used as loading control. n=3.

140

141

of E2F1, indicating that Ser403 likely contributes to the overall phosphorylation status of

E2F1. The T433A and T433D E2F1 mutants showed a similar banding pattern as WT-

E2F1 and treatment with λ phosphatase led to an increase in mobility. This suggests that other Ser/Thr residues on E2F1 can be phosphorylated when Thr433 is mutated (Figure

3.18). Finally, the ST/A and ST/D E2F1 mutants behaved similarly to S403A and S403D, and these E2F1 mutants migrated with a mobility that was indistinguishable from that observed in lysates treated with λ phosphatase (Figure 3.18). This indicates that Ser403 and to a lesser extent Thr433 are critical contributors to E2F1 phosphorylation in keratinocytes.

3.10 Differentiation-specific role of Ser403 and Thr433 in E2F1 degradation

To determine whether phosphorylation at Ser403 and Thr433 sites has a role in the

turnover of E2F1 following Ca2+ induction in keratinocytes, the apparent t½ of the

Ser403 and Thr433 phosphorylation site mutants was examined. In undifferentiated

keratinocytes, the apparent t½ of E2F1 was 74±0.4 min (Figure 3.19A) and following

Ca2+ induction, the apparent t½ of E2F1 was 63±3.5 min (Figure 3.19B). In the half-life

experiments, keratinocytes were treated with Ca2+ for 4 h prior to cycloheximide chase

assay, by 4 h the steady-state level of E2F1 protein had decreased by 30% (Figure 3.13).

Therefore, when the cycloheximide chase assay was performed, the change in E2F1

apparent t½ may not have been apparent. Interestingly, the apparent t½ of S403D

(124±27 min) and T433D (104±2.3 min) E2F1 was higher than WT-E2F1 in undifferentiated keratinocytes (Figure 3.19A), and following Ca2+ induction, the apparent

t½ were similar to WT-E2F1 (Figure 3.19B), but S403D and T433D E2F1 apparent t½

Figure 3.19: Ser403 and Thr433 regulated E2F1 protein half-life in differentiated keratinocytes.

Undifferentiated keratinocytes were transfected with vectors encoding V5-tagged wild type (WT) or the indicated E2F1 mutant proteins. Twenty-four hours after transfection, the culture medium was replaced with A. Low or B. High Ca2+ medium, followed by addition of cycloheximide (100 µg/ml). Cell lysates were prepared from replicate samples at the indicated times thereafter, and analyzed by immunoblot using anti-V5 antibodies, to detect E2F1, or γ-Tubulin used as loading control. Representative blots of each E2F1 protein are shown. The graphs represent the densitometric quantification of E2F1 protein levels (mean + S.D., n=3) calculated from experiments conducted on three independent keratinocyte isolates, and are expressed as a percentage of levels measured at t=0, set to 100%. Decay curves were used to calculate the apparent half-life of each E2F1 protein (t½), summarized in the chart. The asterisks indicates P<0.05 relative to t½ of a given E2F1 form proteins cultured in Low Ca2+ medium and # indicates P<0.05 relative to t½ of wild type E2F1 (ANOVA).

143

144

145

decreased by 2- and 1.6-fold, respectively, compared to undifferentiated keratinocytes.

Further, the apparent t½ of ST/D E2F1 was found to be similar to WT-E2F1 in undifferentiated keratinocytes (76±15 min). In differentiated keratinocytes, the apparent t½ of ST/D E2F1 decreased by 1.6-fold as compared to WT-E2F1 and by 1.8-fold as compared to undifferentiated keratinocytes (Figure 3.19). Thus, although the phospho- mimetic mutants have a similar t½ as WT-E2F1 in undifferentiated keratinocytes, Ca2+

induction decreased the t½ of phospho-mimetic E2F1 mutants, suggesting that they are

subjected to degradation at a higher rate. In contrast, the t½ of ST/A E2F1 was 59±8.9

min, which was increased by >10-fold in the presence of Ca2+ (Figure 3.19). This

suggests that E2F1 phosphorylation at Ser403 and Thr433 is required for E2F1

degradation in Ca2+-induced keratinocyte differentiation.

3.11 K11- and K48-linkages on E2F1 promotes its degradation in differentiated keratinocytes

The mechanisms of protein ubiquitylation are complex. They involve sequential catalytic

reactions to transfer ubiquitin onto a substrate protein, and a single ubiquitin monomer

can then be further ubiquitylated to form polyubiquitin chains through any of the seven-

lysine residues present in ubiquitin. Different ubiquitin linkages can lead to different

biological outcomes. For example, K48-linked polyubiquitin targets substrates for

degradation, whereas K63-linked polyubiquitylation targets substrates for changes in

subcellular localization or endocytosis. Since ST/A is ubiquitylated (Figure 3.15), the

different types of ubiquitin linkages were examined on WT-E2F1 and E2F1 ST/A mutant

in undifferentiated and differentiated keratinocytes. Keratinocytes were transfected with

vectors encoding V5-tagged WT-E2F1 or ST/A and HA-tagged mutant ubiquitin in

146

which all Lys residues were mutated to Arg except for one (K11, K48, or K63) or a

mutant in which all Lys residues were mutated to Arg (K0). In undifferentiated and

differentiated keratinocytes, the presence of WT-E2F1 immunoreactivity was detected in

ubiquitin-K0, -K11, -K48, and -K63 HA-immunoprecipitates (Figure 3.20A, 3.20B).

Similar to WT-E2F1, ST/D-E2F1 is detected in HA-containing immune complexes when

cells were exogenously expressing K11-, K48-, or K63-ubiquitin (Figure 3.20C). Further,

I found robust ubiquitylation of ST/A-E2F1 in undifferentiated keratinocytes when K11-

or K48- ubiquitin was expressed (Figure 3.20A). This indicates that K11- and K48- polyubiquitylation of ST/A-E2F1 can occur in keratinocytes. Strikingly, in differentiated

keratinocytes, E2F1 ST/A ubiquitylation through K11- and K48-linkages was barely detectable, but K63-associated modifications were readily detected (Figure 3.20B). These results suggest that phosphorylation at Ser403 and/or Thr433 is required for K11- and

K48-linked ubiquitylation and targeting of E2F1 to the proteasome for degradation when

keratinocytes are induced to differentiate with Ca2+.

3.12 Cdh1 regulates E2F1 degradation in keratinocytes

Three different E3 ubiquitin ligases, Skp2, Cdh1 and Cdc20, are known to ubiquitylate

E2F1 during S phase, late S phase, and mitosis, respectively (Budhavarapu et al., 2012;

Marti et al., 1999; Peart et al., 2010). In a high-throughput microarray analysis of the keratinocyte transcriptome during differentiation, SKP2 and CDC20 transcripts were reportedly down regulated, whereas FZR1 transcripts, which encode Cdh1, were up regulated (Toufighi et al., 2015). This indicates that Skp2 and Cdc20 might not be involved in regulating E2F1 ubiquitylation and that, potentially, Cdh1 may play a role in

Figure 3.20: Ser403 and Thr433 regulate K11- and K48-linked ubiquitylation of E2F1 in differentiated keratinocytes.

Keratinocytes cultured in Low Ca2+ medium were transfected with vectors encoding the indicated V5- tagged and HA-tagged ubiquitin proteins. After 4 h, the cultured medium was replaced with A. Low or B. C. High Ca2+ medium, and cells were cultured for 18 h, followed by incubation in the presence of MG132 (10 µM) for 6 h. Cell lysates were prepared and HA immunecomplexes were isolated and analyzed by immunoblot with anti-V5 antibodies. The abundance of exogenous E2F1 and HA-ubiquitin-containing proteins in the lysates was analyzed by immunoblot, using γ-Tubulin as loading control. n=3.

148

149

150

151

E2F1 turnover in differentiating mouse keratinocytes.

To test this hypothesis, I exogenously expressed V5-tagged E2F1 in the presence of

FLAG-tagged -Cdh1 and -Skp2. I then examined E2F1 levels in undifferentiated keratinocyte extracts and in extracts isolated from keratinocytes induced to differentiate by incubation in High Ca2+ medium for 24 h. Expression of Skp2 had a negligible effect

on E2F1 protein levels, whereas the presence of exogenous Cdh1 decreased E2F1

abundance in undifferentiated and in differentiated keratinocytes (Figure 3.21). These

data suggest that Cdh1 is likely involved in E2F1 degradation in keratinocytes.

I next conducted experiments to confirm the role of Cdh1 in E2F1 degradation using two

complementary approaches to interfere with Cdh1 function. First, Cdh1 activity was

repressed by exogenously expressing the human early mitotic inhibitor 1 (hEmi1), a

Cdh1 inhibitor (Hsu et al., 2002). To this end, keratinocytes were transfected with vectors

encoding E2F1 in presence of hEmi1 and/or Cdh1. The cells were maintained

undifferentiated or induced to differentiate by culture in High Ca2+ medium for 3 h in the

absence of de novo protein synthesis. This allowed me to assess changes in E2F1 protein

levels specifically associated with degradation during the early stages of Ca2+ induction

of differentiation. Exogenous expression of hEmi1 induced a two-fold increase in the

abundance of E2F1 in undifferentiated keratinocytes (Figure 3.22). Notably, hEmi1

prevented the decrease in E2F1 levels associated with differentiation, and E2F1

abundance in cells cultured in High Ca2+ medium was indistinguishable from that in undifferentiated keratinocytes (Figure 3.22). Exogenously expressed hEmi1 was also able to interfere with the reduction in E2F1 levels in both undifferentiated and

Figure 3.21: Cdh1 and not Skp2 reduce the abundance of E2F1 protein in keratinocytes.

Primary epidermal keratinocytes cultured in Low Ca2 medium were transfected with

vectors encoding the indicated FLAG-tagged ubiquitin ligases in the presence of V5-

tagged E2F1 at a ratio of 1:5 (E2F1 and each ubiquitin ligase). Four hours after

transfection, the culture medium was replaced with Low (-) or High (+) Ca2+ medium and

continued to culture for 24 h. Cell lysates were prepared and analyzed by immunoblot

with the indicated antibodies. n=3.

153

Figure 3.22: Inhibition of Cdh1 activity with hEmi1 increases E2F1 protein levels.

Undifferentiated keratinocytes were transfected with vectors encoding V5-tagged E2F1 in the presence or absence of vectors encoding FLAG-tagged Cdh1 and myc-tagged hEmi1, as indicated. Twenty-four hours later, the cells were treated with cycloheximide (100 μg/ml) and the cultured medium was replaced 30 min later with Low (-) or High (+) Ca2+ medium. Keratinocyte lysates were prepared 3 h later and analyzed by immunoblot with the indicated antibodies. The histograms represent quantification of E2F1 proteins (mean + SEM, n=3), relative to E2F1 abundance in Low Ca2+ medium in the absence of exogenous Cdh1 and hEmi1, which is set at 100%. The asterisks indicate P<0.05 relative to E2F1 levels in cells cultured in Low Ca2+ without other exogenous proteins except where otherwise indicated (ANOVA).

155

156

differentiated keratinocytes observed upon overexpression of Cdh1. In a complementary

approach, I silenced Cdh1 using RNA interference. A three-fold increase in E2F1 levels in undifferentiated keratinocytes treated with Cdh1-targeting siRNA was observed, which was unaffected by induction of differentiation following Ca2+ treatment (Figure 3.23).

Together, these observations indicate that Cdh1 plays key roles in decreasing E2F1

abundance in differentiating keratinocytes.

The E2F1 ST/A mutant is stable in differentiated keratinocytes, therefore I determined

whether Cdh1 could associate and promote the degradation of ST/A E2F1 in

keratinocytes. FLAG-tagged Cdh1 was co-transfected with V5-tagged WT-E2F1, ST/D, or ST/A in undifferentiated keratinocytes, or in cells induced to differentiate by 24 h of culture in medium containing 1.0 mM Ca2+. FLAG-Cdh1 immune complexes were

isolated and analyzed for the presence of E2F1. In these complexes, V5-tagged WT-

E2F1, STD, or ST/A were readily detected, irrespective of the differentiation status of the keratinocytes (Figure 3.24A, 3.24B). Unexpectedly, I observed similar decreases in all three E2F1 forms (WT, ST/D, or ST/A) in the presence of exogenous Cdh1 (Figure

3.24A, 3.24B). These results suggest that phosphorylation of E2F1 at Ser403 and/or

Thr433 is not essential for its interaction with Cdh1, and that a high level of Cdh1 is able to target the very stable ST/A E2F1 mutant for degradation. The results presented in section 3.11 show that E2F1 ST/A mutant is not defective in forming K11- and K48- linkages in undifferentiated keratinocytes, but following Ca2+-induced differentiation,

little or no detectable K11- and K48-associated modifications were observed. Since Cdh1

could interact with E2F1 ST/A, the lack of K11- and K48-linkages is not likely due to its

inability to interact with the ubiquitin ligase, but other aspects of the degradation process.

Figure 3.23: Cdh1 promotes E2F1 degradation in differentiating keratinocytes.

Undifferentiated keratinocytes were sequentially transfected with control, non-targeting (NC) or Cdh1-targeting siRNAs, followed by transfection with vectors encoding V5- tagged E2F1, as described in Materials and Methods (Chapter 2, Section 2.13). Cells were treated with cycloheximide (100 μg/ml) and the cultured medium was replaced 30 min later with Low (-) or High (+) Ca2+ medium. Keratinocyte lysates were prepared 3 h later and analyzed by immunoblot with the indicated antibodies. The histograms represent normalized densitometry quantification of E2F1 levels (mean + SEM, n=3), relative to those in cells cultured in Low Ca2+ medium transfected in the presence of NC siRNA, which is set to 100%. The asterisks indicate P<0.05 relative to cells transfected in the presence of NC siRNA, except where otherwise indicated (ANOVA).

158

Figure 3.24: E2F1 interacts with Cdh1 in keratinocytes.

Primary epidermal keratinocytes cultured in Low Ca2+ medium were transfected with vectors encoding FLAG-tagged Cdh1 and the indicated V5-tagged E2F1 proteins. Four hours after transfection, the cells were cultured in Low (-) or High (+) Ca2+ medium for 24 h. Cell lysates were prepared and subjected to immunoprecipitation with anti-FLAG. The immunecomplexes were analyzed by immunoblot with anti-V5 antibodies, to detect E2F1. Samples of the lysates were also analyzed by immunoblot with the indicated antibodies. n=3.

160

161

One likely possibility is that E2F1 ST/A associate with deubiquitylase that actively

remove K11- and K48-linkages. Alternatively, substitution of S403 and T433 with non-

phosphorylatable alanine could specifically associate with ubiquitin ligases that favors

K63-linkages and inhibits K11- and K48-modifications.

3.13 E2 conjugating enzyme preferentially interacts with ST/A-E2F1 in differentiated keratinocytes

Ubiquitin-conjugating enzyme 13 (UBC13) is the only known mammalian E2 ubiquitin enzyme that catalyzes the formation of K63-linked polyubiquitin chains (Hofmann and

Pickart, 1999; Thorslund et al., 2015). Since K63-polyubiquitylation on E2F1 was observed, the interaction between UBC13 and all three forms of E2F1 (WT-E2F1, ST/A, or ST/D) was investigated. In undifferentiated keratinocytes, WT-E2F1, ST/A, and ST/D were able to efficiently interact with UBC13 (Figure 3.25A, 3.25B). However, following

Ca2+ induction of differentiation, the interaction of WT-E2F1 and ST/D was no longer

detected, whereas that of ST/A was maintained (Figure 3.25A, 3.25B). This indicates that

modification at Ser403 and Thr433 (potentially phosphorylation) may prevent UBC13

from interacting with E2F1 when differentiation pathways are activated in keratinocytes.

Figure 3.25: E2F1 interaction with E2 UBC13 conjugating enzyme is modulated by Ser403 and Thr433 in differentiated keratinocytes.

Undifferentiated keratinocytes were transfected with vectors encoding FLAG-tagged UBC13 and the indicated V5-tagged E2F1 proteins. After 4 h, the cells were cultured in Low (-) or High (+) Ca2+ medium for 24 h. Cell lysates were prepared and subjected to immunoprecipitation with anti-FLAG antibodies. The immunecomplexes were analysed by immunoblot with anti-V5 antibodies, to detect E2F1. Samples of lysates were also analyzed by immunoblot with the indicated antibodies. n=3.

163

164

Chapter 4 4 Discussion

Epidermal keratinocytes provide a barrier function to protect against environmental and

physical damage. Keratinocytes undergo constant self-renewal to repair damaged tissues

and replace old cells that have gone through the normal process of differentiation.

Therefore, it is crucial to understand the regulation of repair mechanisms and the

processes that control keratinocyte differentiation. In this thesis, I explored the function

of E2F1 in these two processes and elucidated some of the mechanisms that regulate

E2F1 stability and turnover following DNA damage or during differentiation. This study

showed that the regulation of E2F1 stability plays a critical role during epidermal DNA

repair and differentiation. First, I identified a novel interaction between E2F1 and

hHR23A, whereby the stability of E2F1 is modulated by hHR23A. More importantly, I

showed that following UV-induced DNA damage, E2F1 recruits hHR23A to damage

sites, which leads to an increase in DNA repair efficiency. Secondly, I demonstrated that

multiple mechanisms are involved in the degradation of E2F1 following Ca2+-induced

keratinocyte differentiation, including nuclear export and ubiquitylation. This study

further shows that Ser403 and Thr433 of E2F1 are likely major phosphorylation sites in

keratinocytes, and appear to be essential for E2F1 turnover through the formation of K11-

or K48-linked ubiquitin chains in differentiated keratinocytes. Finally, I showed that the

ubiquitin ligase Cdh1 mediates E2F1 degradation in differentiated keratinocytes.

Elucidating the events required for E2F1 turnover in keratinocytes, brings us one step

165

closer to understanding various epidermal diseases such as non-melanoma skin

carcinoma.

4.1 E2F1 regulation by hHR23 proteins in keratinocytes

This study has uncovered a complex reciprocal regulation between hHR23 proteins and

E2F1. Specifically, hHR23 proteins can bind E2F1 and protect it from proteasomal

degradation. In turn, E2F1 can recruit hHR23A to the nucleus in the absence of DNA

damage. Upon photodamage, E2F1 is stabilized and is recruited to DNA lesion sites,

bringing hHR23A with it, and increasing the efficiency of DNA repair (Figure 4.1).

hHR23A and the yeast orthologous protein Rad23 function as shuttle proteins that deliver

ubiquitylated cargo to the proteasome for proteolysis (Chen and Madura, 2002).

However, hHR23A can also associate directly with unmodified proteins. For example,

the HIV-1 protein Vpr binds to the UBA2 and XPC domains of hHR23A (Jung et al.,

2014). Similarly, in vitro binding assays with bacterially produced proteins demonstrate

that hHR23A also binds unmodified E2F1. The association of hHR23A and E2F1 in vivo

is likely more complex, due to the presence of ubiquitylated E2F1 pools normally

targeted for proteasomal degradation in mammalian cells. Domain mapping analysis clearly indicates that the UBA2 domain in hHR23A is neither necessary nor sufficient for

E2F1 interaction. The data presented in this thesis further suggests that the UBA1 and

XPC domains of hHR23A are important regions involved in E2F1 binding in keratinocytes. Whereas the XPC domain mediates binding to unmodified E2F1 residues,

UBA1 also contacts ubiquitylated E2F1. An important area for future research will be to

Figure 4.1: Proposed model for hHR23A and E2F1 regulation of DNA repair in undifferentiated keratinocytes

A. In the absence of DNA damage, hHR23A remains mainly localized to the cytoplasm. E2F1 is known to undergo nucleocytoplasmic shuttling in these cells, with the equilibrium favouring nuclear concentration of this protein. Under these conditions, E2F1 turnover is promoted by polyubiquitylation and proteasomal degradation. B. DNA photodamage induces hHR23A nuclear translocation and E2F1 recruitment to sites of DNA lesions, with potential involvement of factors that serve as a bridge between DNA and E2F1. E2F1 can form complexes with hHR23A at those sites, promoting DNA repair. In a coordinate fashion, E2F1 is also protected from degradation through various mechanisms, likely including association with hHR23 proteins.

167

168

determine which domains in hHR23A and E2F1 bind to each other, independent of

ubiquitin modifications.

The diversity of domains and possible modes of interaction of hHR23A and E2F1 imply

that diverse cellular processes may be activated, depending on cell type, differentiation

status and presence of DNA-related stresses. For example, under normal conditions

hHR23A appears to be largely cytoplasmic in undifferentiated keratinocytes.

Significantly, E2F1 is capable of modulating hHR23A subcellular localization, as its

exogenous expression results in nuclear concentration of hHR23A in the absence of DNA

damage in these cells. Reciprocally, hHR23A is able to reduce E2F1 turnover in

undifferentiated keratinocytes. The ability of E2F1 to modulate hHR23A subcellular

localization is further emphasized by the observation that both E2F1 and hHR23A remain

cytoplasmic as a result of differentiation in keratinocytes. E2F1 is indispensable to

maintain proliferative capacity in undifferentiated keratinocytes and for their activation

during epidermal repair after injury (D'Souza et al., 2002). A potential role of hHR23A in

these cells might be to stabilize E2F1 to ensure maintenance of a proliferative cell

population. Further, it will be important to determine if hHR23A also plays a role in

mediating the increases in E2F1 abundance observed in the regenerating epidermis.

Similar stabilizing roles for hHR23A have been reported vis-à-vis p53 and ataxin-3

(Blount et al., 2014; Glockzin et al., 2003).

Because of their location at the skin surface, epidermal keratinocytes are uniquely exposed to UV-related damage, and have evolved several protective mechanisms, including the ability of these cells to repair damaged DNA through NER. The importance

169

of these pathways for epidermal homeostasis is underlined by the existence of

dermatological diseases such as xeroderma pigmentosum, which is associated with skin

photosensitivity. Individuals with this disorder exhibit hypersensitivity to solar UV light

and frequent development of skin cancers (Cleaver, 1968). Several genetic abnormalities

that cause XP have been identified, including alterations in XPC. In response to DNA

damage, various signaling proteins are recruited to areas with altered bases. Under these

conditions, hHR23 proteins play key roles facilitating recognition by XPC of DNA lesion

sites, contributing to the repair of transcriptionally silent genome regions or of non-

transcribed strands of transcriptionally active genes (Bergink et al., 2012; Dantuma et al.,

2009). Given that E2F1 is most abundant in undifferentiated basal keratinocytes with

proliferation potential (Dagnino et al., 1997), the cells of origin of nonmelanoma skin carcinoma, it is conceivable that a major role in vivo for the interaction between hHR23 proteins and E2F1 is as a protective mechanism against DNA damage and malignant transformation of undifferentiated keratinocytes with proliferative potential in the interfollicular epidermis and hair follicles.

The regulation of E2F1 in response to DNA damage is complex, depends on the damaging agent, and occurs through various mechanisms. The human ortholog of E2F1 is phosphorylated by the ATM and ATR kinases on Ser31 (Ser 29 in mouse E2F1) upon drug-induced DNA damage, promoting its binding to 14-3-3τ, which decreases its ubiquitylation and proteasomal degradation (Wang et al., 2004). Under these circumstances, stabilization of E2F1 leads to the transcriptional activation of its proapoptotic targets TAp73 and Apaf1. E2F1 can also be acetylated on several Lys residues in response to chemicals that cause DNA double-strand breaks, with similar outcomes on

170

transcriptional activation of E2F pro-apoptotic targets (Galbiati et al., 2005; Pediconi et

al., 2003). Efficient DNA repair in response to UV-induced DNA damage also requires

E2F1, although different mechanisms appear to be involved. Under those conditions,

E2F1 is also phosphorylated on Ser31 and stabilized (Berton et al., 2005; Pediconi et al.,

2003). Phospho-E2F1 can bind TopBP1, decreasing TAp73 and Apaf1 transcription.

TopBP1 also mediates recruitment of E2F1 to DNA double-strand breaks, where it colocalizes with BRCA1, and promotes NER through mechanisms that appear to involve formation of complexes containing TopBP1, E2F1 and GCN5 (Guo et al., 2011). Indeed,

E2F1-deficient cells are deficient in NER through mechanisms that involve impaired recruitment of DNA repair factors, such as GCN5, to sites of DNA damage. Further, a mouse E2F1 mutant protein lacking Ser29 is incapable of localizing to UV-induced sites of DNA damage, and epidermal keratinocytes expressing this protein exhibit increased susceptibility to UV-induced carcinogenic transformation (Biswas et al., 2014). This

study identifies a novel mechanism that contributes to NER, involving interactions

between hHR23 proteins and E2F1, and the recruitment of hHR23A by E2F1 to sites of

DNA photodamage. Key issues for future research are to determine if E2F1- hHR23

complexes also contain XPC, and to establish the relative importance of the E2F1-hHR23

interaction to the overall capacity of keratinocytes to repair UV-induced DNA. This is a

complex issue to address, given that individual depletion of hHR23 proteins or E2F1

impair cellular repair capacity. The identification of E2F1 mutants incapable of binding

to hHR23A may provide answers to this question.

E2F1 is a well-established promoter of cell proliferation, but its roles in tumour formation

and progression are complex, as it can either induce or suppress tumourigenesis. In the

171

context of the epidermis, E2F1 is necessary for normal wound healing (D'Souza et al.,

2002), and its overexpression leads to spontaneous tumour formation, exacerbated by the

loss of p53 (Pierce et al., 1998b). Paradoxically, E2F1 also protects the epidermis from

UV-induced transformation, and cooperates with the retinoblastoma protein, pRb, to

prevent spontaneous carcinogenic transformation of hair follicle keratinocytes through

modulation of β-catenin/Wnt pathways (Biswas et al., 2014; Costa et al., 2013). These studies now provide new insights on the role of E2F1 in maintenance of genomic stability, underlining the importance of this transcription factor in maintenance of epidermal homeostasis.

4.2 E2F1 induction following genotoxic stress in keratinocytes This study revealed that the residues involved in the stability of E2F1 following

genotoxic stress differ from the residues required for the degradation of E2F1 following

keratinocyte differentiation. Moreover, residues identified previously to be required for

E2F1 stability in response to DNA damage are similar in primary mouse keratinocytes

and cancer cells. E2F1 response following DNA damage has been extensively studied

over the years. Post-translational modifications on E2F1 lead to protein stability

following genotoxic stress. For the first time, the importance of E2F1 post-translational modification in primary mouse keratinocytes has been analyzed and shows that a similar mechanism of E2F1 accumulation in cancer cells exists in primary mouse keratinocytes.

Modifications at Ser31, Lys117, Lys120, and Lys125 are essential for the accumulation of E2F1 in response to genotoxic stress whereas Arg111, Arg113, and Lys185 are unmodified following DNA damage, which contributes to E2F1 accumulation. E2F1

172

plays a dual role in response to genotoxic insults, either leading to apoptosis or favoring cell survival (Poppy Roworth et al., 2015). In cancer cells, these modifications on E2F1

promote an increase in apoptotic events leading to cell death (Cho et al., 2012; Kontaki and Talianidis, 2010; Lin et al., 2001; Pediconi et al., 2003). On the other hand, E2F1 accumulates at DNA damage sites to stimulate NER pathway, which favors cell survival and resistance to genotoxic stress (Castillo et al., 2015; Guo et al., 2010). The increase in

DNA repair is dependent on E2F1 residue Ser31 to favor recruitment of E2F1 at sites of

DNA damage (Biswas et al., 2014). The Mer11 complex, containing Mer11, Nbs1, and

Rad50, is recruited to sites of DNA DSB during DNA replication. Mer11 is methylated by PRMT1 and methylation favors the binding of Rad51 to damage sites (Yu et al.,

2012). E2F1 interacts with Mer11 at the replication fork to recruit Rad51 in the presence of DNA DSB (Maser et al., 2001). E2F1 is also methylated by PRMT1 at residue Arg109 and increases E2F1 protein stability (Zheng et al., 2013). It will be of interest to determine whether Arg109 and other residues on E2F1 play a role in either accumulating

E2F1 at DNA lesion sites or promoting E2F1-induced apoptosis in primary keratinocytes.

4.3 Regulation of E2F1 turnover in differentiated keratinocytes The present study demonstrated multiple mechanisms of E2F1 degradation following

Ca2+-induced keratinocyte differentiation. First, the stability of E2F1 is associated with

E2F1 nuclear export through the CRM1-mediated pathway, as previously shown in our

laboratory (Ivanova and Dagnino, 2007). Second, phosphorylation of E2F1 at Ser403 and

Thr433 prevents the cleavage of K11- and K48- polyubiquitin chains in differentiated keratinocytes leading to recognition by the proteasome for degradation. Third, E2F1

173

protein stability is regulated through E3 ubiquitin ligase, Cdh1, in differentiated

keratinocytes.

The interaction between E2F1 and pRb plays an essential role in E2F1 transcriptional

activity and stability (Helin et al., 1993; Martelli and Livingston, 1999). In my study, I

established that modification at residues previously identified to be involved in the

interaction between pRb and E2F1 do not prevent E2F1 degradation. Therefore, although

pRb dissociation leads to increased transcription of genes through E2F1, this dissociation

may not favour E2F1 degradation in primary mouse keratinocytes as seen in late S phase.

E2F1 nuclear export plays an essential role leading to its degradation following Ca2+- induced keratinocyte differentiation. However, this is not the only step necessary for

E2F1 turnover under these circumstances. Modification at Ser403 and Thr433 of E2F1 also appears to contribute significantly to E2F1 degradation. When Ser403 and Thr433 are substituted for an alanine in a mutant also containing an NES to ensure proper

CRM1-mediated nuclear export, this mutant E2F1 protein is cytoplasmic but still protected from degradation. The use of pseudo-phosphorylation mutants further confirmed the involvement of Ser403 and Thr433 in keratinocytes. The negative charge of aspartic acid can mimic phosphorylation of a protein. Notably, when E2F1 Ser403 and

Thr433 are substituted for Asp, the half-life of this protein was significantly reduced in differentiated keratinocytes. Together, these findings strongly suggest that E2F1 can be phosphorylated at Ser403 and Thr433 and this post-translational event is required for its turnover in response to keratinocyte differentiation. The activity of p38β has been shown to be necessary for E2F1 degradation in keratinocytes, and in vitro kinase assays demonstrated that E2F1 Ser403 and Thr433 are substrates of p38β-MAPK

174

phosphorylation (Ivanova et al., 2009). However, whether E2F1 is phosphorylated by

p38β-MAPK in cultured differentiating keratinocytes is yet to be determined.

The regulation of E2F1 through ubiquitylation has been extensively studied. The APC/C

complex plays a pivotal role in targeting E2F1 for degradation during cell cycle

progression (Budhavarapu et al., 2012; Peart et al., 2010). E2F1 interacts with Cdh1, the

activator of the APC/C complex during early G1 phase to down regulate E2F1 through

K11 linked ubiquitin chain formation (Budhavarapu et al., 2012). My studies for the first

time show phosphorylation-linked ubiquitylation and E2F1 degradation is modulated by

Cdh1. In this thesis, I show that hEmi1-induced inhibition of Cdh1 and loss of Cdh1

interfered with E2F1 down regulation in keratinocytes. Moreover, K11-, K48, and K63-

polyubiquitylation of E2F1 occur in response to keratinocyte differentiation, and the

modification of E2F1 at Ser403 and Thr433 regulate K11- and K48-polyubiquitin linkages. These results could explain why the ST/A E2F1 mutant is more stable that wild type E2F1. Exogenous overexpression of Cdh1 can promote the degradation of ST/A

E2F1 mutant protein as efficiently as wild type E2F1. This suggests that modification at

Ser403 and Thr433 on E2F1 does not interfere with the ability of Cdh1 to target E2F1 for degradation. Moreover, phosphorylation of E2F1 also does not appear to directly interfere with the interaction with Cdh1, since all three forms of E2F1, WT, ST/A, and

ST/D can interact with Cdh1 in keratinocytes. The absence of E2F1 post-translational

modifications does not prevent its interaction with Cdh1 (Budhavarapu et al., 2012).

Therefore, it was not surprising to observe the interaction between Cdh1 and all three

forms of E2F1. If overexpressed Cdh1 can still interact with and mediate the degradation

of ST/A E2F1, why does the ST/A mutant show decrease K11- and K48-linked

175

polyubiquitylation? One possibility is that under physiological conditions in

differentiated keratinocytes, where Cdh1 is maintained at a normal level, both

phosphorylated E2F1 and substitution of Ser403 and Thr433 with non-phosphorylatable

alanine could interact with Cdh1 and catalyzes K11-linkage ubiquitin chain formation

(Figure 4.2A). The phosphorylated E2F1 is then targeted for 26S proteasomal

degradation. On the other hand, deubiquitylating enzymes (potentially ubiquitin-specific

protease (USP) 37) could specifically interact with ST/A E2F1 and actively remove K11-

or K48-linked ubiquitin molecules, thus protecting ST/A E2F1 from degradation (Figure

4.2A). Ectopic expression of Cdh1 decreases endogenous USP37 protein levels (Huang et

al., 2011). Therefore, in keeping with the notion that ST/A E2F1 could potentially be a

target of USP37, when USP37 is down regulated, ST/A E2F1 is no longer

deubiquitylated, thus could now be targeted for degradation (Figure 4.2B). Future study

is warranted to identify whether E2F1 is a USP37 substrate.

The increase in the half-life of E2F1-ST/A mutant and exclusive formation of K63-linked

polyubiquitin chains in differentiated keratinocytes prompted me to investigate the

possibility of interactions between E2F1 and the only known mammalian E2 ubiquitin

conjugating enzyme, UBC13, which specifically generates K63-linked polyubiquitin

chains. These studies identified a previously unrecognized interaction between E2F1 and

UBC13. The phosphorylation state of E2F1 does not affect the ability of UBC13 to

interact with E2F1 in undifferentiated keratinocytes. However, following Ca2+-induced differentiation, UBC13 preferentially interacts with E2F1-ST/A mutant and not with

Figure 4.2: Proposed model of E2F1 phosphorylation-linked ubiquitylation modulated by ubiquitin ligases and deubiquitylating enzymes.

A. Under normal conditions, Cdh1 and UBC13 interact with phosphorylatable E2F1 at Ser403 and Thr433 and non-phosphorylatable E2F1 (ST/A E2F1) to form specific K11- and K63-linked ubiquitin chains respectively. However, the K11-linkages on ST/A E2F1 could be specifically cleaved by deubiquitylating enzymes (DUB, potentially by USP37) which lead to its stabilization. The phosphorylatable E2F1 is on the other hand targeted for degradation. B. The deubiquitylating enzyme, USP37 is targeted for degradation when Cdh1 is overexpressed. Under this circumstance, the ST/A E2F1 is likely no longer deubiquitylated by USP37 leading to its degradation via the 26S proteasome just like wild type E2F1.

177

178

E2F1 ST/D or wild type E2F1. Modifications on residues Ser403 and Thr433 do not appear to be necessary for UBC13 to interact with E2F1, but differentiation might alter

E2F1 and/or UBC13 structure and thus prevent the physical interaction between phosphorylated E2F1 and UBC13. Therefore, one possibility of increased stability of the non-phosphorylated form of E2F1 mutant, ST/A could be due to its interaction with

UBC13, which promotes the formation of K63-linked polyubiquitin chains that are modifications that likely regulate aspects of E2F1 function different from degradation

(Figure 4.2A).

Polyubiquitylation of protein through K63-linked ubiquitin chains is critical for recruiting

DNA repair proteins upon DNA damage (Doil et al., 2009; Pinato et al., 2011;

Ramachandran et al., 2010). Studies have shown that RNF168, an E3 ubiquitin ligase together with UBC13 catalyzes K63-linked ubiquitylation on the chromatin that surrounds DNA lesions. This follows by recruitment of RAP80-ABRA1-BRCA1 complex to the damage sites (Stewart et al., 2009). RNF168 is able to ubiquitylate various histones and potentially other proteins found at sites of DNA damage to promote efficient DNA repair. E2F1 is important for recruitment of DNA repair proteins, such as

XPC, XPA and hHR23 to sites of double strand breaks (Biswas et al., 2014; Chen et al.,

2011). Whether Ser403 and Thr433 play a role in favoring the accumulation of DNA repair proteins to lesion sites is yet to be determined.

E2F1 activity is essential for keratinocyte proliferation and it is regulated through the pRb/E2F pathway. E2F1 down regulation is a prerequisite for keratinocytes to begin the differentiation process followed by entry into quiescence. The PKC-MAPK pathway is implicated in the epidermal differentiation process leading to E2F1 degradation (Ivanova

179

et al., 2006). This study has provided a mechanism of E2F1 turnover through

phosphorylation-linked ubiquitylation mediated by ubiquitin ligase Cdh1. Following

2+ extracellular Ca stimulation, the PLCγ enzyme is activated by PIP3, which hydrolyzes

2+ PIP2 to IP3 and DAG (Bikle et al., 2012). This leads to the release of Ca from

intracellular stores, and activation of PKC. The PKCη and PKCδ are specifically

activated by DAG leading to activation of p38β-MAPK through the Ras-MEKK1-

MEK3/6 pathway (Denning et al., 1995; Efimova et al., 2002; Ivanova et al., 2006). In

response to PKCδ activation, phosphorylated p38δ translocates into the nucleus (Efimova

et al., 2004). Therefore, activation of p38β by phosphorylation could cause its

translocation to the nucleus, where it could phosphorylate E2F1 at Ser403 and Thr433

(Ivanova et al., 2009). Phosphorylation of E2F1 at these sites may prevent its

deubiquitylation (likely occurring through USP37) and interaction with UBC13, which

catalyzes specific K63-linkage ubiquitin chains. Together, the data favour a model where

E2F1 degradation is mediated through its association with Cdh1 (Figure 4.3).

Understanding the signaling events that lead to the degradation of E2F1 is vital as

increased E2F1 expression in the epidermis results in hyperproliferation and hyperplasia

(Pierce et al., 1998a).

4.4 Future studies

This study identified a novel interaction between E2F1 and hHR23A proteins. Together,

these proteins promote DNA repair of UV-damaged keratinocytes. The proximal promoter region of Rad23A contains a consensus E2F1 binding site but putative binding sites for any of the other E2F family members were not detected. A future direction of

Figure 4.3: Proposed model of E2F1 regulation and function in differentiated keratinocytes.

2+ The activation of phospholipase C γ (PLCγ) through binding of extracellular Ca to cell surface receptor(s) catalyzes the hydrolysis of phosphatidylinositol 4,5-bisphosphate

(PIP2) to inositol 1,4,5-trisphosphate (IP3) and diacylglycerol (DAG). Through DAG, novel PKC (η/δ) are activated which leads to phosphorylation of p38β through the Ras- MEKK1-MEK3/6 pathway. Phosphorylated p38β translocate to the nucleus where it might phosphorylate E2F1 at Ser403 and Thr433. E2F1 phosphorylation at Ser403 and Thr433 might prevent DUB enzymes (USP37) from cleaving K11- and K48-linkages. Phosphorylation of E2F1 through Ca2+ induction also prevents its interaction with UBC13, which catalyzes specifically K63-linked ubiquitin chains. Together, this favors degradation of E2F1 mediated through its association with Cdh1 in differentiated keratinocytes.

181

182

this study would be to determine whether E2F1 binds to the promoter of Rad23A in a UV

damage induced-dependent manner by chromatin immunoprecipitation (ChIP)-qPCR assays.

Following DNA damage, hHR23A associates with XPC, and subsequently recruits XPC to DNA damage sites. The recruitment of hHR23A to DNA damage sites is transient and the protein does not accumulate at local DNA damage sites (Bergink et al., 2012). In this

study, I showed that exogenous expression of E2F1 favours hHR23A localization to

DNA damage sites. However, whether the interaction between XPC and hHR23A is

necessary for E2F1 to recruit hHR23A to these DNA lesions has yet to be determined. To

understand the dynamics between the E2F1-hHR23A-XPC complex, future studies should examine whether a mutant hHR23A, defective in interacting with XPC by deleting the XPC domain, can be recruited to DNA damage sites in the presence of E2F1 following UV irradiation. Given that E2F1 could maintain its interaction with a hHR23A mutant lacking the XPC sequences (∆XPC; Figure 3.14C), I would expect that the mutant hHR23A would be recruited to the damage sites. Moreover, I predict that the interaction between E2F1 and hHR23A ∆XPC will not promote efficient DNA repair, since XPC is no longer recruited to the damage sites. Together this would reveal whether E2F1 and hHR23A are necessary to recruit XPC to DNA damage sites to initiate DNA repair successfully.

E2F1 post-translational modifications are numerous and have been extensively studied; however, the specific ubiquitin-modified lysine residues on E2F1 have yet to be nonequivocally identified. By identifying sites of protein modification, one can better understand the biological role of such covalent modifications. E2F1 substitution at a.a.

183

sites 117, 120, and 125 or 185 were identified as acceptor lysines where ubiquitin

molecules are catalyzed (Galbiati et al., 2005; Kontaki and Talianidis, 2010). However, in keratinocytes, exogenous expression of E2F1 KTR or K185R did not prevent ubiquitylation of E2F1 (Appendix 2). Therefore, lysines at position 117, 120, and 125 or

185 may not be the only acceptor sites where ubiquitin molecules attach. E2F1 has 14

lysine residues (16 lysines in murine E2F1) and mutating each residue could be very

challenging. Moreover, this method only indirectly maps ubiquitylation sites (Kaiser and

Wohlschlegel, 2005). Future studies should utilize protein mass spectrometry analysis to

identify the direct location of these modifications. One could also identify the type of

ubiquitin linkages, using proteomic approaches, since E2F1 can be modified to form

K11-, K48-, or K63-linkages.

My study also provided new evidence for the involvement of modifications at Ser403 and

Thr433 in K11- and K48-linked polyubiquitylation of E2F1 and subsequent turnover of

E2F1 in differentiated keratinocytes. In denaturing polyacrylamide gels, the S403A and

S403D E2F1 mutant migrated at the same position as analogous samples treated with λ

phosphatase, indicating that Ser403 is an important contributor to the overall E2F1

phosphorylation state. Whether phosphorylation at Ser403 is necessary for the

phosphorylation of other serine/threonine sites will need to be examined in future

experiments. Phospho-specific antibodies could be used to determine whether

phosphorylation of Thr433 and other serine/threonine residues is dependent on

phosphorylation at Ser403. However, commercially available phospho-specific

antibodies could not detect phosphorylated E2F1 in my hands. The use of

phosphopeptide mapping could potentially help understand the interplay between

184

phosphorylation at Ser403 and Thr433 in vivo. The generation of a conditional knock-in

ST/A-E2F1 mouse model could further define the physiological role of the Ser403 and

Thr433 post-translational modification during epidermal development.

4.5 Concluding remarks

The maintenance of genomic integrity is a crucial requirement for cell viability, and its

disruption can result in aneuploidy and diseases such as cancer in mammals. The mechanisms of differentiation and DNA damage repair play crucial roles in the homeostasis of a cell. In this thesis, I investigated the regulation of E2F1 stability during epidermal DNA repair and differentiation. I showed that the interaction between E2F1 and hHR23A prevents E2F1 degradation and that together, E2F1 and hHR23A are recruited to UV-induced DNA damage sites to promote DNA repair. During keratinocyte differentiation, I found that E2F1 residues Ser403 and Thr433 are required for ubiquitin

K11- and K48-linkage leading to subsequent turnover of E2F1. Modification of these

same residues prevent E2 ubiquitin conjugating enzyme UBC13 from interacting with

E2F1 and potentially favour the association of Cdh1 in differentiated keratinocytes. Tight

control of E2F1 expression thus plays an essential role to ensure proper DNA repair

following genotoxic stress and differentiation.

185

References

Agarwal, C., T. Efimova, J.F. Welter, J.F. Crish, and R.L. Eckert. 1999. CCAAT/enhancer-binding proteins. A role in regulation of human involucrin promoter response to phorbol ester. J Biol Chem. 274:6190-6194.

Andl, T., S.T. Reddy, T. Gaddapara, and S.E. Millar. 2002. WNT signals are required for the initiation of hair follicle development. Dev Cell. 2:643-653.

Apostolova, M.D., I.A. Ivanova, C. Dagnino, S.J. D'Souza, and L. Dagnino. 2002. Active nuclear import and export pathways regulate E2F-5 subcellular localization. J Biol Chem. 277:34471-34479.

Attwooll, C., S. Oddi, P. Cartwright, E. Prosperini, K. Agger, P. Steensgaard, C. Wagener, C. Sardet, M.C. Moroni, and K. Helin. 2005. A novel repressive E2F6 complex containing the polycomb group protein, EPC1, that interacts with EZH2 in a proliferation-specific manner. J Biol Chem. 280:1199-1208.

Baron, Y., S. Corre, N. Mouchet, S. Vaulont, S. Prince, and M.D. Galibert. 2012. USF-1 is critical for maintaining genome integrity in response to UV-induced DNA photolesions. PLoS Genet. 8:e1002470.

Beijersbergen, R.L., R.M. Kerkhoven, L. Zhu, L. Carlee, P.M. Voorhoeve, and R. Bernards. 1994. E2F-4, a new member of the E2F gene family, has oncogenic activity and associates with p107 in vivo. Genes Dev. 8:2680-2690.

Bergink, S., A.F. Theil, W. Toussaint, D.I.M. Cuyper, D.I. Kulu, T. Clapes, R. van der Linden, J.A. Demmers, E.P. Mul, F.P. van Alphen, J.A. Marteijn, T. van Gent, A. Maas, C. Robin, S. Philipsen, W. Vermeulen, J.R. Mitchell, and L. Gutierrez. 2013. Erythropoietic Defect Associated with Reduced Cell Proliferation in Mice Lacking the 26S Proteasome Shuttling Factor Rad23b. Molecular and cellular biology. 33:3879-3892.

Bergink, S., W. Toussaint, M.S. Luijsterburg, C. Dinant, S. Alekseev, J.H. Hoeijmakers, N.P. Dantuma, A.B. Houtsmuller, and W. Vermeulen. 2012. Recognition of DNA damage by XPC coincides with disruption of the XPC-RAD23 complex. J Cell Biol. 196:681-688.

Bertolaet, B.L., D.J. Clarke, M. Wolff, M.H. Watson, M. Henze, G. Divita, and S.I. Reed. 2001. UBA domains of DNA damage-inducible proteins interact with ubiquitin. Nat Struct Biol. 8:417-422.

Berton, T.R., D.L. Mitchell, R. Guo, and D.G. Johnson. 2005. Regulation of epidermal apoptosis and DNA repair by E2F1 in response to ultraviolet B radiation. Oncogene. 24:2449-2460.

186

Bikle, D.D., Z. Xie, and C.L. Tu. 2012. Calcium regulation of keratinocyte differentiation. Expert Rev Endocrinol Metab. 7:461-472.

Biswas, A.K., and D.G. Johnson. 2012. Transcriptional and nontranscriptional functions of E2F1 in response to DNA damage. Cancer Res. 72:13-17.

Biswas, A.K., D.L. Mitchell, and D.G. Johnson. 2014. E2F1 responds to ultraviolet radiation by directly stimulating DNA repair and suppressing carcinogenesis. Cancer Res. 74:3369-3377.

Blanpain, C., and E. Fuchs. 2007. p63: revving up epithelial stem-cell potential. Nat Cell Biol. 9:731-733.

Blount, J.R., W.L. Tsou, G. Ristic, A.A. Burr, M. Ouyang, H. Galante, K.M. Scaglione, and S.V. Todi. 2014. Ubiquitin-binding site 2 of ataxin-3 prevents its proteasomal degradation by interacting with Rad23. Nat Commun. 5:4638.

Bremm, A., and D. Komander. 2011. Emerging roles for Lys11-linked polyubiquitin in cellular regulation. Trends Biochem Sci. 36:355-363.

Budhavarapu, V.N., E.D. White, C.S. Mahanic, L. Chen, F.T. Lin, and W.C. Lin. 2012. Regulation of E2F1 by APC/C Cdh1 via K11 linkage-specific ubiquitin chain formation. Cell Cycle. 11:2030-2038.

Cabodi, S., E. Calautti, C. Talora, T. Kuroki, P.L. Stein, and G.P. Dotto. 2000. A PKC- eta/Fyn-dependent pathway leading to keratinocyte growth arrest and differentiation. Mol Cell. 6:1121-1129.

Campanero, M.R., and E.K. Flemington. 1997. Regulation of E2F through ubiquitin- proteasome-dependent degradation: stabilization by the pRB tumor suppressor protein. Proc Natl Acad Sci U S A. 94:2221-2226.

Carlson, B.M. 1994. Human embryology and developmental biology. Mosby, St. Louis. xv, 447 p. pp.

Carter, W.G., E.A. Wayner, T.S. Bouchard, and P. Kaur. 1990. The role of integrins alpha 2 beta 1 and alpha 3 beta 1 in cell-cell and cell-substrate adhesion of human epidermal cells. J Cell Biol. 110:1387-1404.

Cartwright, P., H. Muller, C. Wagener, K. Holm, and K. Helin. 1998. E2F-6: a novel member of the E2F family is an inhibitor of E2F-dependent transcription. Oncogene. 17:611-623.

Castillo, D.S., A. Campalans, L.M. Belluscio, A.L. Carcagno, J.P. Radicella, E.T. Canepa, and N. Pregi. 2015. E2F1 and E2F2 induction in response to DNA damage preserves genomic stability in neuronal cells. Cell Cycle. 14:1300-1314.

187

Ceccarelli, C., D. Santini, P. Chieco, M. Taffurelli, M. Gamberini, S.A. Pileri, and D. Marrano. 1998. Retinoblastoma (RB1) gene product expression in breast carcinoma. Correlation with Ki-67 growth fraction and biopathological profile. J Clin Pathol. 51:818-824.

Chan, W.M., M.C. Mak, T.K. Fung, A. Lau, W.Y. Siu, and R.Y. Poon. 2006. Ubiquitination of p53 at multiple sites in the DNA-binding domain. Mol Cancer Res. 4:15-25.

Chang, Y.C., H. Nakajima, S. Illenye, Y.S. Lee, N. Honjo, T. Makiyama, I. Fujiwara, N. Mizuta, K. Sawai, K. Saida, Y. Mitsui, N.H. Heintz, and J. Magae. 2000. Caspase-dependent apoptosis by ectopic expression of E2F-4. Oncogene. 19:4713-4720.

Chen, D., Y. Chen, D. Forrest, and R. Bremner. 2013. E2f2 induces cone photoreceptor apoptosis independent of E2f1 and E2f3. Cell Death Differ. 20:931-940.

Chen, J., F. Zhu, R.L. Weaks, A.K. Biswas, R. Guo, Y. Li, and D.G. Johnson. 2011. E2F1 promotes the recruitment of DNA repair factors to sites of DNA double- strand breaks. Cell Cycle. 10:1287-1294.

Chen, L., and K. Madura. 2002. Rad23 promotes the targeting of proteolytic substrates to the proteasome. Mol Cell Biol. 22:4902-4913.

Chen, L., and K. Madura. 2006. Evidence for distinct functions for human DNA repair factors hHR23A and hHR23B. FEBS Lett. 580:3401-3408.

Chen, L., U. Shinde, T.G. Ortolan, and K. Madura. 2001. Ubiquitin-associated (UBA) domains in Rad23 bind ubiquitin and promote inhibition of multi-ubiquitin chain assembly. EMBO reports. 2:933-938.

Cho, E.C., S. Zheng, S. Munro, G. Liu, S.M. Carr, J. Moehlenbrink, Y.C. Lu, L. Stimson, O. Khan, R. Konietzny, J. McGouran, A.S. Coutts, B. Kessler, D.J. Kerr, and N.B. Thangue. 2012. Arginine methylation controls growth regulation by E2F-1. EMBO J. 31:1785-1797.

Chong, J.-L., P.L. Wenzel, T.M. Sáenz-Robles, V. Nair, A. Ferrey, J.P. Hagan, Y.M. Gomez, N. Sharma, H.-Z. Chen, M. Ouseph, S.-H. Wang, P. Trikha, B. Culp, L. Mezache, D.J. Winton, O.J. Sansom, D. Chen, R. Bremner, P.G. Cantalupo, M.L. Robinson, J.M. Pipas, and G. Leone. 2009a. E2f1–3 switch from activators in progenitor cells to repressors in differentiating cells. Nature. 462:930-934.

Chong, J.L., S.Y. Tsai, N. Sharma, R. Opavsky, R. Price, L. Wu, S.A. Fernandez, and G. Leone. 2009b. E2f3a and E2f3b contribute to the control of cell proliferation and mouse development. Mol Cell Biol. 29:414-424.

188

Christenson, L.J., T.A. Borrowman, C.M. Vachon, M.M. Tollefson, C.C. Otley, A.L. Weaver, and R.K. Roenigk. 2005. Incidence of basal cell and squamous cell carcinomas in a population younger than 40 years. Jama-J Am Med Assoc. 294:681-690.

Ciechanover, A., H. Heller, S. Elias, A.L. Haas, and A. Hershko. 1980. ATP-dependent conjugation of reticulocyte proteins with the polypeptide required for protein degradation. Proc Natl Acad Sci U S A. 77:1365-1368.

Clayton, E., D.P. Doupe, A.M. Klein, D.J. Winton, B.D. Simons, and P.H. Jones. 2007. A single type of progenitor cell maintains normal epidermis. Nature. 446:185-189.

Cleaver, J.E. 1968. Defective repair replication of DNA in xeroderma pigmentosum. Nature. 218:652-656.

Costa, C., M. Santos, M. Martinez-Fernandez, M. Duenas, C. Lorz, R. Garcia-Escudero, and J.M. Paramio. 2013. E2F1 loss induces spontaneous tumour development in Rb-deficient epidermis. Oncogene. 32:2937-2951.

Crish, J.F., R. Gopalakrishnan, F. Bone, A.C. Gilliam, and R.L. Eckert. 2006. The distal and proximal regulatory regions of the involucrin gene promoter have distinct functions and are required for in vivo involucrin expression. J Invest Dermatol. 126:305-314.

D'Souza, S.J., A. Pajak, K. Balazsi, and L. Dagnino. 2001. Ca2+ and BMP-6 signaling regulate E2F during epidermal keratinocyte differentiation. J Biol Chem. 276:23531-23538.

D'Souza, S.J., A. Vespa, S. Murkherjee, A. Maher, A. Pajak, and L. Dagnino. 2002. E2F- 1 is essential for normal epidermal wound repair. J Biol Chem. 277:10626-10632.

Dagnino, L., C.J. Fry, S.M. Bartley, P. Farnham, B.L. Gallie, and R.A. Phillips. 1997. Expression patterns of the E2F family of transcription factors during murine epithelial development. Cell Growth Differ. 8:553-563.

Dagnino, L., E. Ho, and W.Y. Chang. 2010. Expression and analysis of exogenous proteins in epidermal cells. Methods Mol Biol. 585:93-105.

Dahiya, A., M.R. Gavin, R.X. Luo, and D.C. Dean. 2000. Role of the LXCXE binding site in Rb function. Mol Cell Biol. 20:6799-6805.

Dantuma, N.P., C. Heinen, and D. Hoogstraten. 2009. The ubiquitin receptor Rad23: at the crossroads of nucleotide excision repair and proteasomal degradation. DNA Repair (Amst). 8:449-460.

189

de Jonge, H.J., R.S. Fehrmann, E.S. de Bont, R.M. Hofstra, F. Gerbens, W.A. Kamps, E.G. de Vries, A.G. van der Zee, G.J. te Meerman, and A. ter Elst. 2007. Evidence based selection of housekeeping genes. PLoS One. 2:e898.

DeCaprio, J.A., J.W. Ludlow, J. Figge, J.Y. Shew, C.M. Huang, W.H. Lee, E. Marsilio, E. Paucha, and D.M. Livingston. 1988. SV40 large tumor antigen forms a specific complex with the product of the retinoblastoma susceptibility gene. Cell. 54:275- 283.

Denning, M.F., A.A. Dlugosz, E.K. Williams, Z. Szallasi, P.M. Blumberg, and S.H. Yuspa. 1995. Specific protein kinase C isozymes mediate the induction of keratinocyte differentiation markers by calcium. Cell Growth Differ. 6:149-157.

Deucher, A., T. Efimova, and R.L. Eckert. 2002. Calcium-dependent involucrin expression is inversely regulated by protein kinase C (PKC)alpha and PKCdelta. J Biol Chem. 277:17032-17040.

Dick, F.A., and N. Dyson. 2003. pRB contains an E2F1-specific binding domain that allows E2F1-induced apoptosis to be regulated separately from other E2F activities. Mol Cell. 12:639-649.

Dicker, A.J., C. Popa, A.L. Dahler, M.M. Serewko, P.A. Hilditch-Maguire, I.H. Frazer, and N.A. Saunders. 2000. E2F-1 induces proliferation-specific genes and suppresses squamous differentiation-specific genes in human epidermal keratinocytes. Oncogene. 19:2887-2894.

Diffey, B.L. 1991. Solar ultraviolet radiation effects on biological systems. Phys Med Biol. 36:299-328.

Doil, C., N. Mailand, S. Bekker-Jensen, P. Menard, D.H. Larsen, R. Pepperkok, J. Ellenberg, S. Panier, D. Durocher, J. Bartek, J. Lukas, and C. Lukas. 2009. RNF168 binds and amplifies ubiquitin conjugates on damaged chromosomes to allow accumulation of repair proteins. Cell. 136:435-446.

Dynlacht, B.D., O. Flores, J.A. Lees, and E. Harlow. 1994. Differential regulation of E2F transactivation by cyclin/cdk2 complexes. Genes Dev. 8:1772-1786.

Dyson, N., P.M. Howley, K. Munger, and E. Harlow. 1989. The human papilloma virus- 16 E7 oncoprotein is able to bind to the retinoblastoma gene product. Science. 243:934-937.

Ebihara, Y., M. Miyamoto, T. Shichinohe, Y. Kawarada, Y. Cho, A. Fukunaga, S. Murakami, H. Uehara, H. Kaneko, H. Hashimoto, Y. Murakami, T. Itoh, S. Okushiba, S. Kondo, and H. Katoh. 2004. Over-expression of E2F-1 in esophageal squamous cell carcinoma correlates with tumor progression. Diseases of the esophagus : official journal of the International Society for Diseases of the Esophagus / I.S.D.E. 17:150-154.

190

Efimova, T., A.M. Broome, and R.L. Eckert. 2004. Protein kinase Cdelta regulates keratinocyte death and survival by regulating activity and subcellular localization of a p38delta-extracellular signal-regulated kinase 1/2 complex. Mol Cell Biol. 24:8167-8183.

Efimova, T., A. Deucher, T. Kuroki, M. Ohba, and R.L. Eckert. 2002. Novel protein kinase C isoforms regulate human keratinocyte differentiation by activating a p38 delta mitogen-activated protein kinase cascade that targets CCAAT/enhancer- binding protein alpha. J Biol Chem. 277:31753-31760.

Efimova, T., and R.L. Eckert. 2000. Regulation of human involucrin promoter activity by novel protein kinase C isoforms. J Biol Chem. 275:1601-1607.

Efimova, T., P. LaCelle, J.F. Welter, and R.L. Eckert. 1998. Regulation of human involucrin promoter activity by a protein kinase C, Ras, MEKK1, MEK3, p38/RK, AP1 signal transduction pathway. J Biol Chem. 273:24387-24395.

Elias, P., S. Ahn, B. Brown, D. Crumrine, and K.R. Feingold. 2002. Origin of the epidermal calcium gradient: regulation by barrier status and role of active vs passive mechanisms. J Invest Dermatol. 119:1269-1274.

Elsasser, S., D. Chandler-Militello, B. Muller, J. Hanna, and D. Finley. 2004. Rad23 and Rpn10 serve as alternative ubiquitin receptors for the proteasome. J Biol Chem. 279:26817-26822.

Erpapazoglou, Z., M. Dhaoui, M. Pantazopoulou, F. Giordano, M. Mari, S. Leon, G. Raposo, F. Reggiori, and R. Haguenauer-Tsapis. 2012. A dual role for K63-linked ubiquitin chains in multivesicular body biogenesis and cargo sorting. Mol Biol Cell. 23:2170-2183.

Evans, E., J.G. Moggs, J.R. Hwang, J.M. Egly, and R.D. Wood. 1997. Mechanism of open complex and dual incision formation by human nucleotide excision repair factors. EMBO J. 16:6559-6573.

Ezhevsky, S.A., H. Nagahara, A.M. Vocero-Akbani, D.R. Gius, M.C. Wei, and S.F. Dowdy. 1997. Hypo-phosphorylation of the retinoblastoma protein (pRb) by cyclin D:Cdk4/6 complexes results in active pRb. Proc Natl Acad Sci U S A. 94:10699-10704.

Fagan, R., K.J. Flint, and N. Jones. 1994. Phosphorylation of E2F-1 modulates its interaction with the retinoblastoma gene product and the adenoviral E4 19 kDa protein. Cell. 78:799-811.

Fang, D.F., K. He, J. Wang, R. Mu, B. Tan, Z. Jian, H.Y. Li, W. Song, Y. Chang, W.L. Gong, W.H. Li, and G.J. Wang. 2013. RAD23A negatively regulates RIG- I/MDA5 signaling through promoting TRAF2 polyubiquitination and degradation. Biochem Biophys Res Commun. 431:686-692.

191

Farhana, L., M. Dawson, A.K. Rishi, Y. Zhang, E. Van Buren, C. Trivedi, U. Reichert, G. Fang, M.W. Kirschner, and J.A. Fontana. 2002. Cyclin B and E2F-1 expression in prostate carcinoma cells treated with the novel retinoid CD437 are regulated by the ubiquitin-mediated pathway. Cancer Res. 62:3842-3849.

Frye, M., A.G. Fisher, and F.M. Watt. 2007. Epidermal stem cells are defined by global histone modifications that are altered by Myc-induced differentiation. PLoS One. 2:e763.

Fuchs, E. 2007. Scratching the surface of skin development. Nature. 445:834-842.

Fuchs, E., and T. Chen. 2013. A matter of life and death: self-renewal in stem cells. EMBO Rep. 14:39-48.

Galbiati, L., R. Mendoza-Maldonado, M.I. Gutierrez, and M. Giacca. 2005. Regulation of E2F-1 after DNA damage by p300-mediated acetylation and ubiquitination. Cell Cycle. 4:930-939.

Ghari, F., A.M. Quirke, S. Munro, J. Kawalkowska, S. Picaud, J. McGouran, V. Subramanian, A. Muth, R. Williams, B. Kessler, P.R. Thompson, P. Fillipakopoulos, S. Knapp, P.J. Venables, and N.B. La Thangue. 2016. Citrullination-acetylation interplay guides E2F-1 activity during the inflammatory response. Sci Adv. 2:e1501257.

Giacinti, C., and A. Giordano. 2006. RB and cell cycle progression. Oncogene. 25:5220- 5227.

Ginsberg, D. 2002. E2F1 pathways to apoptosis. FEBS letters. 529:122-125.

Glockzin, S., F.X. Ogi, A. Hengstermann, M. Scheffner, and C. Blattner. 2003. Involvement of the DNA repair protein hHR23 in p53 degradation. Mol Cell Biol. 23:8960-8969.

Grossman, D., and D.J. Leffell. 1997. The molecular basis of nonmelanoma skin cancer: new understanding. Arch Dermatol. 133:1263-1270.

Guo, R., J. Chen, D.L. Mitchell, and D.G. Johnson. 2011. GCN5 and E2F1 stimulate nucleotide excision repair by promoting H3K9 acetylation at sites of damage. Nucleic Acids Res. 39:1390-1397.

Guo, R., J. Chen, F. Zhu, A.K. Biswas, T.R. Berton, D.L. Mitchell, and D.G. Johnson. 2010. E2F1 localizes to sites of UV-induced DNA damage to enhance nucleotide excision repair. J Biol Chem. 285:19308-19315.

Hallstrom, T.C., and J.R. Nevins. 2006. Jab1 is a specificity factor for E2F1-induced apoptosis. Genes Dev. 20:613-623.

192

Hannon, G.J., D. Demetrick, and D. Beach. 1993. Isolation of the Rb-related p130 through its interaction with CDK2 and cyclins. Genes Dev. 7:2378-2391.

Hardy, M.H. 1992. The secret life of the hair follicle. Trends Genet. 8:55-61.

Hateboer, G., R.M. Kerkhoven, A. Shvarts, R. Bernards, and R.L. Beijersbergen. 1996. Degradation of E2F by the ubiquitin-proteasome pathway: regulation by retinoblastoma family proteins and adenovirus transforming proteins. Genes Dev. 10:2960-2970.

Haupt, S., M. Berger, Z. Goldberg, and Y. Haupt. 2003. Apoptosis - the p53 network. J Cell Sci. 116:4077-4085.

Heinen, C., K. Acs, D. Hoogstraten, and N.P. Dantuma. 2011. C-terminal UBA domains protect ubiquitin receptors by preventing initiation of protein degradation. Nat Commun. 2:191.

Helin, K., E. Harlow, and A. Fattaey. 1993. Inhibition of E2F-1 transactivation by direct binding of the retinoblastoma protein. Mol Cell Biol. 13:6501-6508.

Hennings, H., D. Michael, C. Cheng, P. Steinert, K. Holbrook, and S.H. Yuspa. 1980. Calcium regulation of growth and differentiation of mouse epidermal cells in culture. Cell. 19:245-254.

Hershko, A., and A. Ciechanover. 1998. The ubiquitin system. Annu Rev Biochem. 67:425-479.

Hijmans, E.M., P.M. Voorhoeve, R.L. Beijersbergen, L.J. van 't Veer, and R. Bernards. 1995. E2F-5, a new E2F family member that interacts with p130 in vivo. Mol Cell Biol. 15:3082-3089.

Hiyama, H., M. Yokoi, C. Masutani, K. Sugasawa, T. Maekawa, K. Tanaka, J.H. Hoeijmakers, and F. Hanaoka. 1999. Interaction of hHR23 with S5a. The ubiquitin-like domain of hHR23 mediates interaction with S5a subunit of 26 S proteasome. J Biol Chem. 274:28019-28025.

Ho, E., T. Irvine, G.J. Vilk, G. Lajoie, K.S. Ravichandran, S.J. D'Souza, and L. Dagnino. 2009. Integrin-linked kinase interactions with ELMO2 modulate cell polarity. Mol Biol Cell. 20:3033-3043.

Hoege, C., B. Pfander, G.L. Moldovan, G. Pyrowolakis, and S. Jentsch. 2002. RAD6- dependent DNA repair is linked to modification of PCNA by ubiquitin and SUMO. Nature. 419:135-141.

Hofferer, M., C. Wirbelauer, B. Humar, and W. Krek. 1999. Increased levels of E2F-1- dependent DNA binding activity after UV- or gamma-irradiation. Nucleic Acids Res. 27:491-495.

193

Hofmann, F., F. Martelli, D.M. Livingston, and Z. Wang. 1996. The retinoblastoma gene product protects E2F-1 from degradation by the ubiquitin-proteasome pathway. Genes Dev. 10:2949-2959.

Hofmann, R.M., and C.M. Pickart. 1999. Noncanonical MMS2-encoded ubiquitin- conjugating enzyme functions in assembly of novel polyubiquitin chains for DNA repair. Cell. 96:645-653.

Hong, S., Q.X. Paulson, and D.G. Johnson. 2008. E2F1 and E2F3 activate ATM through distinct mechanisms to promote E1A-induced apoptosis. Cell Cycle. 7:391-400.

Hsu, J.Y., J.D. Reimann, C.S. Sorensen, J. Lukas, and P.K. Jackson. 2002. E2F- dependent accumulation of hEmi1 regulates S phase entry by inhibiting APC(Cdh1). Nat Cell Biol. 4:358-366.

Huang, X., M.K. Summers, V. Pham, J.R. Lill, J. Liu, G. Lee, D.S. Kirkpatrick, P.K. Jackson, G. Fang, and V.M. Dixit. 2011. Deubiquitinase USP37 is activated by CDK2 to antagonize APC(CDH1) and promote S phase entry. Mol Cell. 42:511- 523.

Ianari, A., T. Natale, E. Calo, E. Ferretti, E. Alesse, I. Screpanti, K. Haigis, A. Gulino, and J.A. Lees. 2009. Proapoptotic function of the retinoblastoma tumor suppressor protein. Cancer Cell. 15:184-194.

Inoue, Y., M. Kitagawa, and Y. Taya. 2007. Phosphorylation of pRB at Ser612 by Chk1/2 leads to a complex between pRB and E2F-1 after DNA damage. EMBO J. 26:2083-2093.

Ivanova, I.A., S.J. D'Souza, and L. Dagnino. 2006. E2F1 stability is regulated by a novel- PKC/p38beta MAP kinase signaling pathway during keratinocyte differentiation. Oncogene. 25:430-437.

Ivanova, I.A., and L. Dagnino. 2007. Activation of p38- and CRM1-dependent nuclear export promotes E2F1 degradation during keratinocyte differentiation. Oncogene. 26:1147-1154.

Ivanova, I.A., K.A. Nakrieko, and L. Dagnino. 2009. Phosphorylation by p38 MAP kinase is required for E2F1 degradation and keratinocyte differentiation. Oncogene. 28:52-62.

Ivanova, I.A., A. Vespa, and L. Dagnino. 2007. A novel mechanism of E2F1 regulation via nucleocytoplasmic shuttling: determinants of nuclear import and export. Cell Cycle. 6:2186-2195.

Ivey-Hoyle, M., R. Conroy, H.E. Huber, P.J. Goodhart, A. Oliff, and D.C. Heimbrook. 1993. Cloning and characterization of E2F-2, a novel protein with the biochemical properties of transcription factor E2F. Mol Cell Biol. 13:7802-7812.

194

Jaks, V., N. Barker, M. Kasper, J.H. van Es, H.J. Snippert, H. Clevers, and R. Toftgard. 2008. Lgr5 marks cycling, yet long-lived, hair follicle stem cells. Nat Genet. 40:1291-1299.

Johnson, D.G., and J. Degregori. 2006. Putting the Oncogenic and Tumor Suppressive Activities of E2F into Context. Curr Mol Med. 6:731-738.

Johnson, D.G., J.K. Schwarz, W.D. Cress, and J.R. Nevins. 1993. Expression of transcription factor E2F1 induces quiescent cells to enter S phase. Nature. 365:349-352.

Judah, D., A. Rudkouskaya, R. Wilson, D.E. Carter, and L. Dagnino. 2012. Multiple roles of integrin-linked kinase in epidermal development, maturation and pigmentation revealed by molecular profiling. PLoS One. 7:e36704.

Julian, L.M., O. Palander, L.A. Seifried, J.E. Foster, and F.A. Dick. 2008. Characterization of an E2F1-specific binding domain in pRB and its implications for apoptotic regulation. Oncogene. 27:1572-1579.

Jung, J., I.J. Byeon, M. DeLucia, L.M. Koharudin, J. Ahn, and A.M. Gronenborn. 2014. Binding of HIV-1 Vpr protein to the human homolog of the yeast DNA repair protein RAD23 (hHR23A) requires its xeroderma pigmentosum complementation group C binding (XPCB) domain as well as the ubiquitin-associated 2 (UBA2) domain. J Biol Chem. 289:2577-2588.

Kaiser, P., and J. Wohlschlegel. 2005. Identification of ubiquitination sites and determination of ubiquitin-chain architectures by mass spectrometry. Methods Enzymol. 399:266-277.

Kang, Y., X. Chen, J.W. Lary, J.L. Cole, and K.J. Walters. 2007. Defining how ubiquitin receptors hHR23a and S5a bind polyubiquitin. J Mol Biol. 369:168-176.

Kang, Y., R.A. Vossler, L.A. Diaz-Martinez, N.S. Winter, D.J. Clarke, and K.J. Walters. 2006. UBL/UBA ubiquitin receptor proteins bind a common tetraubiquitin chain. J Mol Biol. 356:1027-1035.

Kashiwagi, M., M. Ohba, K. Chida, and T. Kuroki. 2002. Protein kinase C eta (PKC eta): its involvement in keratinocyte differentiation. J Biochem. 132:853-857.

Kashiwagi, M., M. Ohba, H. Watanabe, K. Ishino, K. Kasahara, Y. Sanai, Y. Taya, and T. Kuroki. 2000. PKCeta associates with cyclin E/cdk2/p21 complex, phosphorylates p21 and inhibits cdk2 kinase in keratinocytes. Oncogene. 19:6334-6341.

Kaur, P., and A. Li. 2000. Adhesive properties of human basal epidermal cells: an analysis of keratinocyte stem cells, transit amplifying cells, and postmitotic differentiating cells. J Invest Dermatol. 114:413-420.

195

Kelley, M.J., K. Nakagawa, S.M. Steinberg, J.L. Mulshine, A. Kamb, and B.E. Johnson. 1995. Differential inactivation of CDKN2 and Rb protein in non-small-cell and small-cell lung cancer cell lines. J Natl Cancer Inst. 87:756-761.

Kinross, K.M., A.J. Clark, R.M. Iazzolino, and P.O. Humbert. 2006. E2f4 regulates fetal erythropoiesis through the promotion of cellular proliferation. Blood. 108:886- 895.

Knudson, A.G., Jr., H.W. Hethcote, and B.W. Brown. 1975. Mutation and childhood cancer: a probabilistic model for the incidence of retinoblastoma. Proc Natl Acad Sci U S A. 72:5116-5120.

Kolly, C., M.M. Suter, and E.J. Muller. 2005. Proliferation, cell cycle exit, and onset of terminal differentiation in cultured keratinocytes: pre-programmed pathways in control of C-Myc and Notch1 prevail over extracellular calcium signals. J Invest Dermatol. 124:1014-1025.

Kontaki, H., and I. Talianidis. 2010. Lysine methylation regulates E2F1-induced cell death. Mol Cell. 39:152-160.

Koster, M.I., S. Kim, A.A. Mills, F.J. DeMayo, and D.R. Roop. 2004. p63 is the molecular switch for initiation of an epithelial stratification program. Genes Dev. 18:126-131.

Krek, W., G. Xu, and D.M. Livingston. 1995. Cyclin A-kinase regulation of E2F-1 DNA binding function underlies suppression of an S phase checkpoint. Cell. 83:1149- 1158.

Kumar, S., A.L. Talis, and P.M. Howley. 1999. Identification of HHR23A as a substrate for E6-associated protein-mediated ubiquitination. J Biol Chem. 274:18785- 18792.

Lambertson, D., L. Chen, and K. Madura. 1999. Pleiotropic defects caused by loss of the proteasome-interacting factors Rad23 and Rpn10 of Saccharomyces cerevisiae. Genetics. 153:69-79.

Lambertson, D., L. Chen, and K. Madura. 2003. Investigating the importance of proteasome-interaction for Rad23 function. Curr Genet. 42:199-208.

Lees, J.A., M. Saito, M. Vidal, M. Valentine, T. Look, E. Harlow, N. Dyson, and K. Helin. 1993. The retinoblastoma protein binds to a family of E2F transcription factors. Mol Cell Biol. 13:7813-7825.

Leone, G., F. Nuckolls, S. Ishida, M. Adams, R. Sears, L. Jakoi, A. Miron, and J.R. Nevins. 2000. Identification of a novel E2F3 product suggests a mechanism for determining specificity of repression by Rb proteins. Mol Cell Biol. 20:3626- 3632.

196

Li, A., P.J. Simmons, and P. Kaur. 1998. Identification and isolation of candidate human keratinocyte stem cells based on cell surface phenotype. Proc Natl Acad Sci U S A. 95:3902-3907.

Li, J., C. Ran, E. Li, F. Gordon, G. Comstock, H. Siddiqui, W. Cleghorn, H.Z. Chen, K. Kornacker, C.G. Liu, S.K. Pandit, M. Khanizadeh, M. Weinstein, G. Leone, and A. de Bruin. 2008. Synergistic function of E2F7 and E2F8 is essential for cell survival and embryonic development. Dev Cell. 14:62-75.

Liao, C.C., C.Y. Tsai, W.C. Chang, W.H. Lee, and J.M. Wang. 2010. RB.E2F1 complex mediates DNA damage responses through transcriptional regulation of ZBRK1. J Biol Chem. 285:33134-33143.

Lim, K.L., K.C. Chew, J.M. Tan, C. Wang, K.K. Chung, Y. Zhang, Y. Tanaka, W. Smith, S. Engelender, C.A. Ross, V.L. Dawson, and T.M. Dawson. 2005. Parkin mediates nonclassical, proteasomal-independent ubiquitination of synphilin-1: implications for Lewy body formation. J Neurosci. 25:2002-2009.

Lin, P.S., L.A. McPherson, A.Y. Chen, J. Sage, and J.M. Ford. 2009. The role of the retinoblastoma/E2F1 tumor suppressor pathway in the lesion recognition step of nucleotide excision repair. DNA Repair (Amst). 8:795-802.

Lin, W.C., F.T. Lin, and J.R. Nevins. 2001. Selective induction of E2F1 in response to DNA damage, mediated by ATM-dependent phosphorylation. Genes Dev. 15:1833-1844.

Lindeman, G.J., L. Dagnino, S. Gaubatz, Y. Xu, R.T. Bronson, H.B. Warren, and D.M. Livingston. 1998. A specific, nonproliferative role for E2F-5 in choroid plexus function revealed by gene targeting. Genes Dev. 12:1092-1098.

Liu, K., F.T. Lin, J.M. Ruppert, and W.C. Lin. 2003. Regulation of E2F1 by BRCT domain-containing protein TopBP1. Mol Cell Biol. 23:3287-3304.

Liu, K., Y. Luo, F.T. Lin, and W.C. Lin. 2004. TopBP1 recruits Brg1/Brm to repress E2F1-induced apoptosis, a novel pRb-independent and E2F1-specific control for cell survival. Genes Dev. 18:673-686.

Livingston, C.M., M.F. Ifrim, A.E. Cowan, and S.K. Weller. 2009. Virus-Induced Chaperone-Enriched (VICE) domains function as nuclear protein quality control centers during HSV-1 infection. PLoS Pathog. 5:e1000619.

Loftus, S.J., G. Liu, S.M. Carr, S. Munro, and N.B. La Thangue. 2012. NEDDylation regulates E2F-1-dependent transcription. EMBO Rep. 13:811-818.

Logan, N., A. Graham, X. Zhao, R. Fisher, B. Maiti, G. Leone, and N.B. La Thangue. 2005. E2F-8: an E2F family member with a similar organization of DNA-binding domains to E2F-7. Oncogene. 24:5000-5004.

197

Lundberg, A.S., and R.A. Weinberg. 1998. Functional inactivation of the retinoblastoma protein requires sequential modification by at least two distinct cyclin-cdk complexes. Mol Cell Biol. 18:753-761.

Magae, J., C.L. Wu, S. Illenye, E. Harlow, and N.H. Heintz. 1996. Nuclear localization of DP and E2F transcription factors by heterodimeric partners and retinoblastoma protein family members. J Cell Sci. 109 ( Pt 7):1717-1726.

Magnaghi-Jaulin, L., R. Groisman, I. Naguibneva, P. Robin, S. Lorain, J.P. Le Villain, F. Troalen, D. Trouche, and A. Harel-Bellan. 1998. Retinoblastoma protein represses transcription by recruiting a histone deacetylase. Nature. 391:601-605.

Mahanic, C.S., V. Budhavarapu, J.D. Graves, G. Li, and W.C. Lin. 2015. Regulation of E2 promoter binding factor 1 (E2F1) transcriptional activity through a deubiquitinating enzyme, UCH37. J Biol Chem. 290:26508-26522.

Marteijn, J.A., H. Lans, W. Vermeulen, and J.H. Hoeijmakers. 2014. Understanding nucleotide excision repair and its roles in cancer and ageing. Nat Rev Mol Cell Biol. 15:465-481.

Martelli, F., and D.M. Livingston. 1999. Regulation of endogenous E2F1 stability by the retinoblastoma family proteins. Proc Natl Acad Sci U S A. 96:2858-2863.

Marti, A., C. Wirbelauer, M. Scheffner, and W. Krek. 1999. Interaction between ubiquitin-protein ligase SCFSKP2 and E2F-1 underlies the regulation of E2F-1 degradation. Nat Cell Biol. 1:14-19.

Martin, K., D. Trouche, C. Hagemeier, T.S. Sorensen, N.B. La Thangue, and T. Kouzarides. 1995. Stimulation of E2F1/DP1 transcriptional activity by MDM2 oncoprotein. Nature. 375:691-694.

Martinez-Balbas, M.A., U.M. Bauer, S.J. Nielsen, A. Brehm, and T. Kouzarides. 2000. Regulation of E2F1 activity by acetylation. EMBO J. 19:662-671.

Martinez, L.A., E. Goluszko, H.Z. Chen, G. Leone, S. Post, G. Lozano, Z. Chen, and A. Chauchereau. 2010. E2F3 is a mediator of DNA damage-induced apoptosis. Mol Cell Biol. 30:524-536.

Marzio, G., C. Wagener, M.I. Gutierrez, P. Cartwright, K. Helin, and M. Giacca. 2000. E2F family members are differentially regulated by reversible acetylation. J Biol Chem. 275:10887-10892.

Mascia, F., M. Denning, R. Kopan, and S.H. Yuspa. 2012. The black box illuminated: signals and signaling. J Invest Dermatol. 132:811-819.

Masutani, C., K. Sugasawa, J. Yanagisawa, T. Sonoyama, M. Ui, T. Enomoto, K. Takio, K. Tanaka, P.J. van der Spek, D. Bootsma, and et al. 1994. Purification and

198

cloning of a nucleotide excision repair complex involving the xeroderma pigmentosum group C protein and a human homologue of yeast RAD23. EMBO J. 13:1831-1843.

Mayol, X., X. Grana, A. Baldi, N. Sang, Q. Hu, and A. Giordano. 1993. Cloning of a new member of the retinoblastoma gene family (pRb2) which binds to the E1A transforming domain. Oncogene. 8:2561-2566.

McKeon, F. 2004. p63 and the epithelial stem cell: more than status quo? Genes Dev. 18:465-469.

Meng, P., and R. Ghosh. 2014. Transcription addiction: can we garner the Yin and Yang functions of E2F1 for cancer therapy? Cell Death Dis. 5:e1360.

Meng, R.D., P. Phillips, and W.S. El-Deiry. 1999. p53-independent increase in E2F-1 expression enhances the cytotoxic effects of etoposide and of adriamycin. Int J Oncol. 14:5-14.

Mills, A.A., B. Zheng, X.J. Wang, H. Vogel, D.R. Roop, and A. Bradley. 1999. p63 is a p53 homologue required for limb and epidermal morphogenesis. Nature. 398:708- 713.

Mone, M.J., M. Volker, O. Nikaido, L.H. Mullenders, A.A. van Zeeland, P.J. Verschure, E.M. Manders, and R. van Driel. 2001. Local UV-induced DNA damage in cell nuclei results in local transcription inhibition. EMBO Rep. 2:1013-1017.

Moon, N.S., and N. Dyson. 2008. E2F7 and E2F8 keep the E2F family in balance. Dev Cell. 14:1-3.

Moroni, M.C., E.S. Hickman, E. Lazzerini Denchi, G. Caprara, E. Colli, F. Cecconi, H. Muller, and K. Helin. 2001. Apaf-1 is a transcriptional target for E2F and p53. Nat Cell Biol. 3:552-558.

Morris, R.J., Y. Liu, L. Marles, Z. Yang, C. Trempus, S. Li, J.S. Lin, J.A. Sawicki, and G. Cotsarelis. 2004. Capturing and profiling adult hair follicle stem cells. Nat Biotechnol. 22:411-417.

Moser, J., H. Kool, I. Giakzidis, K. Caldecott, L.H. Mullenders, and M.I. Fousteri. 2007. Sealing of chromosomal DNA nicks during nucleotide excision repair requires XRCC1 and DNA ligase III alpha in a cell-cycle-specific manner. Mol Cell. 27:311-323.

Muller, H., M.C. Moroni, E. Vigo, B.O. Petersen, J. Bartek, and K. Helin. 1997. Induction of S-phase entry by E2F transcription factors depends on their nuclear localization. Mol Cell Biol. 17:5508-5520.

199

Nagl, N.G., Jr., X. Wang, A. Patsialou, M. Van Scoy, and E. Moran. 2007. Distinct mammalian SWI/SNF chromatin remodeling complexes with opposing roles in cell-cycle control. EMBO J. 26:752-763.

Nakrieko, K.A., I. Welch, H. Dupuis, D. Bryce, A. Pajak, R. St Arnaud, S. Dedhar, S.J. D'Souza, and L. Dagnino. 2008. Impaired hair follicle morphogenesis and polarized keratinocyte movement upon conditional inactivation of integrin-linked kinase in the epidermis. Mol Biol Cell. 19:1462-1473.

Nathan, J.A., H.T. Kim, L. Ting, S.P. Gygi, and A.L. Goldberg. 2013. Why do cellular proteins linked to K63-polyubiquitin chains not associate with proteasomes? EMBO J. 32:552-565.

Ng, J.M., W. Vermeulen, G.T. van der Horst, S. Bergink, K. Sugasawa, H. Vrieling, and J.H. Hoeijmakers. 2003. A novel regulation mechanism of DNA repair by damage-induced and RAD23-dependent stabilization of xeroderma pigmentosum group C protein. Genes Dev. 17:1630-1645.

Ng, J.M., H. Vrieling, K. Sugasawa, M.P. Ooms, J.A. Grootegoed, J.T. Vreeburg, P. Visser, R.B. Beems, T.G. Gorgels, F. Hanaoka, J.H. Hoeijmakers, and G.T. van der Horst. 2002. Developmental defects and male sterility in mice lacking the ubiquitin-like DNA repair gene mHR23B. Mol Cell Biol. 22:1233-1245.

Nguyen, B.C., K. Lefort, A. Mandinova, D. Antonini, V. Devgan, G. Della Gatta, M.I. Koster, Z. Zhang, J. Wang, A. Tommasi di Vignano, J. Kitajewski, G. Chiorino, D.R. Roop, C. Missero, and G.P. Dotto. 2006. Cross-regulation between Notch and p63 in keratinocyte commitment to differentiation. Genes Dev. 20:1028-1042.

Ogi, T., S. Limsirichaikul, R.M. Overmeer, M. Volker, K. Takenaka, R. Cloney, Y. Nakazawa, A. Niimi, Y. Miki, N.G. Jaspers, L.H. Mullenders, S. Yamashita, M.I. Fousteri, and A.R. Lehmann. 2010. Three DNA polymerases, recruited by different mechanisms, carry out NER repair synthesis in human cells. Mol Cell. 37:714-727.

Ohta, T., and Y. Xiong. 2001. Phosphorylation- and Skp1-independent in vitro ubiquitination of E2F1 by multiple ROC-cullin ligases. Cancer Res. 61:1347- 1353.

Olson, M.V., D.G. Johnson, H. Jiang, J. Xu, M.M. Alonso, K.D. Aldape, G.N. Fuller, B.N. Bekele, W.K. Yung, C. Gomez-Manzano, and J. Fueyo. 2007. Transgenic E2F1 expression in the mouse brain induces a human-like bimodal pattern of tumors. Cancer Res. 67:4005-4009.

Ortolan, T.G., P. Tongaonkar, D. Lambertson, L. Chen, C. Schauber, and K. Madura. 2000. The DNA repair protein rad23 is a negative regulator of multi-ubiquitin chain assembly. Nature cell biology. 2:601-608.

200

Paulson, Q.X., M.J. McArthur, and D.G. Johnson. 2006. E2F3a stimulates proliferation, p53-independent apoptosis and carcinogenesis in a transgenic mouse model. Cell Cycle. 5:184-190.

Peart, M.J., M.V. Poyurovsky, E.M. Kass, M. Urist, E.W. Verschuren, M.K. Summers, P.K. Jackson, and C. Prives. 2010. APC/C(Cdc20) targets E2F1 for degradation in prometaphase. Cell Cycle. 9:3956-3964.

Pediconi, N., A. Ianari, A. Costanzo, L. Belloni, R. Gallo, L. Cimino, A. Porcellini, I. Screpanti, C. Balsano, E. Alesse, A. Gulino, and M. Levrero. 2003. Differential regulation of E2F1 apoptotic target genes in response to DNA damage. Nat Cell Biol. 5:552-558.

Peeper, D.S., P. Keblusek, K. Helin, M. Toebes, A.J. van der Eb, and A. Zantema. 1995. Phosphorylation of a specific cdk site in E2F-1 affects its electrophoretic mobility and promotes pRB-binding in vitro. Oncogene. 10:39-48.

Peltonen, J., H. Larjava, S. Jaakkola, H. Gralnick, S.K. Akiyama, S.S. Yamada, K.M. Yamada, and J. Uitto. 1989. Localization of integrin receptors for fibronectin, collagen, and laminin in human skin. Variable expression in basal and squamous cell carcinomas. J Clin Invest. 84:1916-1923.

Peters, J.M., W.W. Franke, and J.A. Kleinschmidt. 1994. Distinct 19 S and 20 S subcomplexes of the 26 S proteasome and their distribution in the nucleus and the cytoplasm. J Biol Chem. 269:7709-7718.

Pickart, C.M. 2000. Ubiquitin in chains. Trends Biochem Sci. 25:544-548.

Pierce, A.M., S.M. Fisher, C.J. Conti, and D.G. Johnson. 1998a. Deregulated expression of E2F1 induces hyperplasia and cooperates with ras in skin tumor development. Oncogene. 16:1267-1276.

Pierce, A.M., I.B. Gimenez-Conti, R. Schneider-Broussard, L.A. Martinez, C.J. Conti, and D.G. Johnson. 1998b. Increased E2F1 activity induces skin tumors in mice heterozygous and nullizygous for p53. Proc Natl Acad Sci U S A. 95:8858-8863.

Pinato, S., M. Gatti, C. Scandiuzzi, S. Confalonieri, and L. Penengo. 2011. UMI, a novel RNF168 ubiquitin binding domain involved in the DNA damage signaling pathway. Mol Cell Biol. 31:118-126.

Ping, Z., R. Lim, T. Bashir, M. Pagano, and D. Guardavaccaro. 2012. APC/C (Cdh1) controls the proteasome-mediated degradation of E2F3 during cell cycle exit. Cell Cycle. 11:1999-2005.

Polager, S., Y. Kalma, E. Berkovich, and D. Ginsberg. 2002. E2Fs up-regulate expression of genes involved in DNA replication, DNA repair and mitosis. Oncogene. 21:437-446.

201

Poppy Roworth, A., F. Ghari, and N.B. La Thangue. 2015. To live or let die - complexity within the E2F1 pathway. Mol Cell Oncol. 2:e970480.

Prost, S., P. Lu, H. Caldwell, and D. Harrison. 2007. E2F regulates DDB2: consequences for DNA repair in Rb-deficient cells. Oncogene. 26:3572-3581.

Qin, L., D.S. Guimaraes, M. Melesse, and M.C. Hall. 2016. Substrate Recognition by the Cdh1 Destruction Box Receptor Is a General Requirement for APC/CCdh1- mediated Proteolysis. J Biol Chem. 291:15564-15574.

Raasi, S., I. Orlov, K.G. Fleming, and C.M. Pickart. 2004. Binding of polyubiquitin chains to ubiquitin-associated (UBA) domains of HHR23A. J Mol Biol. 341:1367-1379.

Raasi, S., and C.M. Pickart. 2003. Rad23 ubiquitin-associated domains (UBA) inhibit 26 S proteasome-catalyzed proteolysis by sequestering lysine 48-linked polyubiquitin chains. J Biol Chem. 278:8951-8959.

Raghavan, S., C. Bauer, G. Mundschau, Q.Q. Li, and E. Fuchs. 2000. Conditional ablation of beta 1 integrin in skin: Severe defects in epidermal proliferation, basement membrane formation, and hair follicle invagination. Journal of Cell Biology. 150:1149-1160.

Ramachandran, S., R. Chahwan, R.M. Nepal, D. Frieder, S. Panier, S. Roa, A. Zaheen, D. Durocher, M.D. Scharff, and A. Martin. 2010. The RNF8/RNF168 ubiquitin ligase cascade facilitates class switch recombination. Proc Natl Acad Sci U S A. 107:809-814.

Rangarajan, A., C. Talora, R. Okuyama, M. Nicolas, C. Mammucari, H. Oh, J.C. Aster, S. Krishna, D. Metzger, P. Chambon, L. Miele, M. Aguet, F. Radtke, and G.P. Dotto. 2001. Notch signaling is a direct determinant of keratinocyte growth arrest and entry into differentiation. EMBO J. 20:3427-3436.

Ravid, T., and M. Hochstrasser. 2008. Diversity of degradation signals in the ubiquitin- proteasome system. Nat Rev Mol Cell Biol. 9:679-690.

Real, S., L. Espada, C. Espinet, A.F. Santidrian, and A. Tauler. 2010. Study of the in vivo phosphorylation of E2F1 on Ser403. Biochim Biophys Acta. 1803:912-918.

Reardon, J.T., and A. Sancar. 2003. Recognition and repair of the cyclobutane thymine dimer, a major cause of skin cancers, by the human excision nuclease. Genes Dev. 17:2539-2551.

Renaud, E., L. Miccoli, N. Zacal, D.S. Biard, C.T. Craescu, A.J. Rainbow, and J.F. Angulo. 2011. Differential contribution of XPC, RAD23A, RAD23B and CENTRIN 2 to the UV-response in human cells. DNA Repair. 10:835-847.

202

Rezvani, H.R., A.L. Kim, R. Rossignol, N. Ali, M. Daly, W. Mahfouf, N. Bellance, A. Taieb, H. de Verneuil, F. Mazurier, and D.R. Bickers. 2011. XPC silencing in normal human keratinocytes triggers metabolic alterations that drive the formation of squamous cell carcinomas. J Clin Invest. 121:195-211.

Romano, R.A., K. Smalley, C. Magraw, V.A. Serna, T. Kurita, S. Raghavan, and S. Sinha. 2012. DeltaNp63 knockout mice reveal its indispensable role as a master regulator of epithelial development and differentiation. Development. 139:772- 782.

Rook, A., and T. Burns. 2004. Rook's textbook of dermatology. Blackwell Science, Malden, Mass.

Rubin, S.M. 2013. Deciphering the retinoblastoma protein phosphorylation code. Trends Biochem Sci. 38:12-19.

Sancar, A., L.A. Lindsey-Boltz, K. Unsal-Kacmaz, and S. Linn. 2004. Molecular mechanisms of mammalian DNA repair and the DNA damage checkpoints. Annu Rev Biochem. 73:39-85.

Sansal, I., E. Dupont, D. Toru, C. Evrard, and P. Rouget. 2000. NPDC-1, a regulator of neural cell proliferation and differentiation, interacts with E2F-1, reduces its binding to DNA and modulates its transcriptional activity. Oncogene. 19:5000- 5009.

Scheffner, M., U. Nuber, and J.M. Huibregtse. 1995. Protein ubiquitination involving an E1-E2-E3 enzyme ubiquitin thioester cascade. Nature. 373:81-83.

Scheijen, B., M. Bronk, T. van der Meer, D. De Jong, and R. Bernards. 2004. High incidence of thymic epithelial tumors in E2F2 transgenic mice. J Biol Chem. 279:10476-10483.

Schneider, C.A., W.S. Rasband, and K.W. Eliceiri. 2012. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 9:671-675.

Senoo, M., F. Pinto, C.P. Crum, and F. McKeon. 2007. p63 Is essential for the proliferative potential of stem cells in stratified epithelia. Cell. 129:523-536.

Shan, B., X. Zhu, P.L. Chen, T. Durfee, Y. Yang, D. Sharp, and W.H. Lee. 1992. Molecular cloning of cellular genes encoding retinoblastoma-associated proteins: identification of a gene with properties of the transcription factor E2F. Mol Cell Biol. 12:5620-5631.

Simbulan-Rosenthal, C.M., D.S. Rosenthal, R. Luo, R. Samara, L.A. Espinoza, P.O. Hassa, M.O. Hottiger, and M.E. Smulson. 2003. PARP-1 binds E2F-1 independently of its DNA binding and catalytic domains, and acts as a novel

203

coactivator of E2F-1-mediated transcription during re-entry of quiescent cells into S phase. Oncogene. 22:8460-8471.

Staresincic, L., A.F. Fagbemi, J.H. Enzlin, A.M. Gourdin, N. Wijgers, I. Dunand- Sauthier, G. Giglia-Mari, S.G. Clarkson, W. Vermeulen, and O.D. Scharer. 2009. Coordination of dual incision and repair synthesis in human nucleotide excision repair. EMBO J. 28:1111-1120.

Stevens, C., and N.B. La Thangue. 2004. The emerging role of E2F-1 in the DNA damage response and checkpoint control. DNA Repair (Amst). 3:1071-1079.

Stevens, C., L. Smith, and N.B. La Thangue. 2003. Chk2 activates E2F-1 in response to DNA damage. Nat Cell Biol. 5:401-409.

Stewart, G.S., S. Panier, K. Townsend, A.K. Al-Hakim, N.K. Kolas, E.S. Miller, S. Nakada, J. Ylanko, S. Olivarius, M. Mendez, C. Oldreive, J. Wildenhain, A. Tagliaferro, L. Pelletier, N. Taubenheim, A. Durandy, P.J. Byrd, T. Stankovic, A.M. Taylor, and D. Durocher. 2009. The RIDDLE syndrome protein mediates a ubiquitin-dependent signaling cascade at sites of DNA damage. Cell. 136:420- 434.

Strachan, G.D., K.L. Jordan-Sciutto, R. Rallapalli, R.S. Tuan, and D.J. Hall. 2003. The E2F-1 transcription factor is negatively regulated by its interaction with the MDMX protein. J Cell Biochem. 88:557-568.

Su, X., M.S. Cho, Y.J. Gi, B.A. Ayanga, C.J. Sherr, and E.R. Flores. 2009. Rescue of key features of the p63-null epithelial phenotype by inactivation of Ink4a and Arf. EMBO J. 28:1904-1915.

Subtil-Rodriguez, A., E. Vazquez-Chavez, M. Ceballos-Chavez, M. Rodriguez-Paredes, J.I. Martin-Subero, M. Esteller, and J.C. Reyes. 2014. The chromatin remodeller CHD8 is required for E2F-dependent transcription activation of S-phase genes. Nucleic Acids Res. 42:2185-2196.

Sugasawa, K., J.M. Ng, C. Masutani, T. Maekawa, A. Uchida, P.J. van der Spek, A.P. Eker, S. Rademakers, C. Visser, A. Aboussekhra, R.D. Wood, F. Hanaoka, D. Bootsma, and J.H. Hoeijmakers. 1997. Two human homologs of Rad23 are functionally interchangeable in complex formation and stimulation of XPC repair activity. Mol Cell Biol. 17:6924-6931.

Takahashi, Y., J.B. Rayman, and B.D. Dynlacht. 2000. Analysis of promoter binding by the E2F and pRB families in vivo: distinct E2F proteins mediate activation and repression. Genes Dev. 14:804-816.

Tani, H., R.J. Morris, and P. Kaur. 2000. Enrichment for murine keratinocyte stem cells based on cell surface phenotype. Proc Natl Acad Sci U S A. 97:10960-10965.

204

Thiagarajan, V., M. Byrdin, A.P. Eker, P. Muller, and K. Brettel. 2011. Kinetics of cyclobutane thymine dimer splitting by DNA photolyase directly monitored in the UV. Proc Natl Acad Sci U S A. 108:9402-9407.

Thorslund, T., A. Ripplinger, S. Hoffmann, T. Wild, M. Uckelmann, B. Villumsen, T. Narita, T.K. Sixma, C. Choudhary, S. Bekker-Jensen, and N. Mailand. 2015. Histone H1 couples initiation and amplification of ubiquitin signalling after DNA damage. Nature. 527:389-393.

Toguchida, J., K. Ishizaki, M.S. Sasaki, M. Ikenaga, M. Sugimoto, Y. Kotoura, and T. Yamamuro. 1988. Chromosomal reorganization for the expression of recessive mutation of retinoblastoma susceptibility gene in the development of osteosarcoma. Cancer Res. 48:3939-3943.

Toufighi, K., J.S. Yang, N.M. Luis, S. Aznar Benitah, B. Lehner, L. Serrano, and C. Kiel. 2015. Dissecting the calcium-induced differentiation of human primary keratinocytes stem cells by integrative and structural network analyses. PLoS Comput Biol. 11:e1004256.

Treier, M., L.M. Staszewski, and D. Bohmann. 1994. Ubiquitin-dependent c-Jun degradation in vivo is mediated by the delta domain. Cell. 78:787-798.

Trempus, C.S., R.J. Morris, C.D. Bortner, G. Cotsarelis, R.S. Faircloth, J.M. Reece, and R.W. Tennant. 2003. Enrichment for living murine keratinocytes from the hair follicle bulge with the cell surface marker CD34. J Invest Dermatol. 120:501-511.

Trimarchi, J.M., B. Fairchild, R. Verona, K. Moberg, N. Andon, and J.A. Lees. 1998. E2F-6, a member of the E2F family that can behave as a transcriptional repressor. Proc Natl Acad Sci U S A. 95:2850-2855.

Tsvetkov, L.M., K.H. Yeh, S.J. Lee, H. Sun, and H. Zhang. 1999. p27(Kip1) ubiquitination and degradation is regulated by the SCF(Skp2) complex through phosphorylated Thr187 in p27. Curr Biol. 9:661-664.

Tyagi, S., A.L. Chabes, J. Wysocka, and W. Herr. 2007. E2F activation of S phase promoters via association with HCF-1 and the MLL family of histone H3K4 methyltransferases. Mol Cell. 27:107-119.

Ueda, E., S. Ohno, T. Kuroki, E. Livneh, K. Yamada, K. Yamanishi, and H. Yasuno. 1996. The eta isoform of protein kinase C mediates transcriptional activation of the human transglutaminase 1 gene. J Biol Chem. 271:9790-9794. van den Heuvel, S. 2005. Cell-cycle regulation. WormBook:1-16. van der Spek, P.J., A. Eker, S. Rademakers, C. Visser, K. Sugasawa, C. Masutani, F. Hanaoka, D. Bootsma, and J.H. Hoeijmakers. 1996a. XPC and human homologs

205

of RAD23: intracellular localization and relationship to other nucleotide excision repair complexes. Nucleic Acids Res. 24:2551-2559.

van der Spek, P.J., C.E. Visser, F. Hanaoka, B. Smit, A. Hagemeijer, D. Bootsma, and J.H. Hoeijmakers. 1996b. Cloning, comparative mapping, and RNA expression of the mouse homologues of the Saccharomyces cerevisiae nucleotide excision repair gene RAD23. Genomics. 31:20-27.

Vandel, L., and T. Kouzarides. 1999. Residues phosphorylated by TFIIH are required for E2F-1 degradation during S-phase. EMBO J. 18:4280-4291.

Vandel, L., E. Nicolas, O. Vaute, R. Ferreira, S. Ait-Si-Ali, and D. Trouche. 2001. Transcriptional repression by the retinoblastoma protein through the recruitment of a histone methyltransferase. Mol Cell Biol. 21:6484-6494.

Verma, R., R. Oania, J. Graumann, and R.J. Deshaies. 2004. Multiubiquitin chain receptors define a layer of substrate selectivity in the ubiquitin-proteasome system. Cell. 118:99-110.

Verona, R., K. Moberg, S. Estes, M. Starz, J.P. Vernon, and J.A. Lees. 1997. E2F activity is regulated by cell cycle-dependent changes in subcellular localization. Mol Cell Biol. 17:7268-7282.

Vidal, V.P., M.C. Chaboissier, S. Lutzkendorf, G. Cotsarelis, P. Mill, C.C. Hui, N. Ortonne, J.P. Ortonne, and A. Schedl. 2005. Sox9 is essential for outer root sheath differentiation and the formation of the hair stem cell compartment. Curr Biol. 15:1340-1351.

Walters, K.J., P.J. Lech, A.M. Goh, Q. Wang, and P.M. Howley. 2003. DNA-repair protein hHR23a alters its protein structure upon binding proteasomal subunit S5a. Proc Natl Acad Sci U S A. 100:12694-12699.

Wang, B., K. Liu, F.T. Lin, and W.C. Lin. 2004. A role for 14-3-3 tau in E2F1 stabilization and DNA damage-induced apoptosis. J Biol Chem. 279:54140- 54152.

Wang, B., A. Ma, L. Zhang, W.L. Jin, Y. Qian, G. Xu, B. Qiu, Z. Yang, Y. Liu, Q. Xia, and Y. Liu. 2015. POH1 deubiquitylates and stabilizes E2F1 to promote tumour formation. Nat Commun. 6:8704.

Wang, C., F.J. Rauscher, 3rd, W.D. Cress, and J. Chen. 2007. Regulation of E2F1 function by the nuclear corepressor KAP1. J Biol Chem. 282:29902-29909.

Watt, F.M., S. Estrach, and C.A. Ambler. 2008. Epidermal Notch signalling: differentiation, cancer and adhesion. Curr Opin Cell Biol. 20:171-179.

206

Wei, W., N.G. Ayad, Y. Wan, G.J. Zhang, M.W. Kirschner, and W.G. Kaelin, Jr. 2004. Degradation of the SCF component Skp2 in cell-cycle phase G1 by the anaphase- promoting complex. Nature. 428:194-198.

Wilson, A., and F. Radtke. 2006. Multiple functions of Notch signaling in self-renewing organs and cancer. FEBS Lett. 580:2860-2868.

Wilson, S.I., A. Rydstrom, T. Trimborn, K. Willert, R. Nusse, T.M. Jessell, and T. Edlund. 2001. The status of Wnt signalling regulates neural and epidermal fates in the chick embryo. Nature. 411:325-330.

Wong, C.F., L.M. Barnes, A.L. Dahler, L. Smith, M.M. Serewko-Auret, C. Popa, I. Abdul-Jabbar, and N.A. Saunders. 2003. E2F modulates keratinocyte squamous differentiation: implications for E2F inhibition in squamous cell carcinoma. J Biol Chem. 278:28516-28522.

Wu, L., C. Timmers, B. Maiti, H.I. Saavedra, L. Sang, G.T. Chong, F. Nuckolls, P. Giangrande, F.A. Wright, S.J. Field, M.E. Greenberg, S. Orkin, J.R. Nevins, M.L. Robinson, and G. Leone. 2001. The E2F1-3 transcription factors are essential for cellular proliferation. Nature. 414:457-462.

Wu, S., S. Murai, K. Kataoka, and M. Miyagishi. 2008. Yin Yang 1 induces transcriptional activity of p73 through cooperation with E2F1. Biochem Biophys Res Commun. 365:75-81.

Wu, Z., S. Zheng, and Q. Yu. 2009. The E2F family and the role of E2F1 in apoptosis. The international journal of biochemistry & cell biology. 41:2389-2397.

Xie, Q., Y. Bai, J. Wu, Y. Sun, Y. Wang, Y. Zhang, P. Mei, and Z. Yuan. 2011a. Methylation-mediated regulation of E2F1 in DNA damage-induced cell death. J Recept Signal Transduct Res.

Xie, Q., Y. Bai, J. Wu, Y. Sun, Y. Wang, Y. Zhang, P. Mei, and Z. Yuan. 2011b. Methylation-mediated regulation of E2F1 in DNA damage-induced cell death. J Recept Signal Transduct Res. 31:139-146.

Xie, Z., and D.D. Bikle. 1999. Phospholipase C-gamma1 is required for calcium-induced keratinocyte differentiation. J Biol Chem. 274:20421-20424.

Xie, Z., P.A. Singleton, L.Y. Bourguignon, and D.D. Bikle. 2005. Calcium-induced human keratinocyte differentiation requires src- and fyn-mediated phosphatidylinositol 3-kinase-dependent activation of phospholipase C-gamma1. Mol Biol Cell. 16:3236-3246.

Xu, M., K.A. Sheppard, C.Y. Peng, A.S. Yee, and H. Piwnica-Worms. 1994. Cyclin A/CDK2 binds directly to E2F-1 and inhibits the DNA-binding activity of E2F- 1/DP-1 by phosphorylation. Mol Cell Biol. 14:8420-8431.

207

Yamazaki, K., M. Hasegawa, I. Ohoka, K. Hanami, A. Asoh, T. Nagao, I. Sugano, and Y. Ishida. 2005. Increased E2F-1 expression via tumour cell proliferation and decreased apoptosis are correlated with adverse prognosis in patients with squamous cell carcinoma of the oesophagus. J Clin Pathol. 58:904-910.

Yang, W.W., Z.H. Wang, Y. Zhu, and H.T. Yang. 2007. E2F6 negatively regulates ultraviolet-induced apoptosis via modulation of BRCA1. Cell Death Differ. 14:807-817.

Yu, Z., G. Vogel, Y. Coulombe, D. Dubeau, E. Spehalski, J. Hebert, D.O. Ferguson, J.Y. Masson, and S. Richard. 2012. The MRE11 GAR motif regulates DNA double- strand break processing and ATR activation. Cell Res. 22:305-320.

Yuspa, S.H., A.E. Kilkenny, P.M. Steinert, and D.R. Roop. 1989. Expression of murine epidermal differentiation markers is tightly regulated by restricted extracellular calcium concentrations in vitro. J Cell Biol. 109:1207-1217.

Zhang, Z., H. Wang, M. Li, E.R. Rayburn, S. Agrawal, and R. Zhang. 2005. Stabilization of E2F1 protein by MDM2 through the E2F1 ubiquitination pathway. Oncogene. 24:7238-7247.

Zheng, L., H. Pan, S. Li, A. Flesken-Nikitin, P.L. Chen, T.G. Boyer, and W.H. Lee. 2000. Sequence-specific transcriptional corepressor function for BRCA1 through a novel zinc finger protein, ZBRK1. Mol Cell. 6:757-768.

Zheng, S., J. Moehlenbrink, Y.C. Lu, L.P. Zalmas, C.A. Sagum, S. Carr, J.F. McGouran, L. Alexander, O. Fedorov, S. Munro, B. Kessler, M.T. Bedford, Q. Yu, and N.B. La Thangue. 2013. Arginine methylation-dependent reader-writer interplay governs growth control by E2F-1. Mol Cell. 52:37-51.

Zhu, Q., G. Wani, M.A. Wani, and A.A. Wani. 2001. Human homologue of yeast Rad23 protein A interacts with p300/cyclic AMP-responsive element binding (CREB)- binding protein to down-regulate transcriptional activity of p53. Cancer Res. 61:64-70.

Zhu, W., P.H. Giangrande, and J.R. Nevins. 2004. E2Fs link the control of G1/S and G2/M transcription. EMBO J. 23:4615-4626.

208

Appendices

209

Appendix 1: Primary epidermal keratinocytes irradiated with UV.

Keratinocytes isolated from 3-day-old mice were plated at 70% confluence. Twenty-four hours later, keratinocytes were subjected to various dose of A. UVB or B. UVC irradiation through no filters or through 3-µm polycarbonate isopore filters, and cultured for 30 min. DNA was denatured with 2 N HCl and processed for immunofluorescence microscopy using anti-CPD antibodies. DNA was visualized with Hoechst 33342. Bar, 16 μm.

210

211

Appendix 2: Ubiquitylation of E2F1 proteins in keratinocytes.

Undifferentiated keratinocytes were transfected with vectors encoding HA-tagged ubiquitin and the indicated V5-tagged E2F1 proteins, cultured in Low (-) or High (+) Ca2+ medium, treated with MG132 (10 µM) for 6 h, and harvested to prepare whole-cell lysates. HA or unrelated IgG immunoprecipitates were isolated from the lysates and analyzed by immunoblot with anti-V5 antibodies. The abundance of exogenous E2F1 proteins in the lysates was analyzed by immunoblot, using γ-Tubulin as loading control.

212

Appendix 3: Copyright permission.

Open-Access License No Permission Required

Oncotarget applies the Creative Commons Attribution License (CCAL) to all works we publish (read the human-readable summary or the full license legal code). Under the CCAL, authors retain ownership of the copyright for their article, but authors allow anyone to download, reuse, reprint, modify, distribute, and/or copy articles in Oncotarget journal, so long as the original authors and source are cited.

No permission is required from the authors or the publishers.

In most cases, appropriate attribution can be provided by simply citing the original article. If the item you plan to reuse is not part of a published article (e.g., a featured issue image), then please indicate the originator of the work, and the volume, issue, and date of the journal in which the item appeared. For any reuse or redistribution of a work, you must also make clear the license terms under which the work was published. This broad license was developed to facilitate open access to, and free use of, original works of all types. Applying this standard license to your own work will ensure your right to make your work freely and openly available.

For queries about the license, please contact us at [email protected]

213

Curriculum Vitae Name

Randeep Kaur Singh

Post-secondary Education and Degrees

2010-2016 University of Western Ontario London, Ontario PhD in Physiology and Pharmacology • Regulation of E2F1 in keratinocytes during UV-damage and differentiation.

2006-2008 University of Western Ontario London, Ontario M.Sc in Biochemistry • The H-NS protein regulates the chemical steps in Tn10 transposition.

2001-2005 University of Western Ontario London Ontario B.MSc. Honors Biochemistry • Chromatin genes and double strand break repair of Drosophila melanogaster.

Honours and Awards:

2015 George W. Stavraky Teaching Scholarship 2013 10th Oncology Research & Education Day: Poster Presentation Winner 2012 Schulich Ontario Graduate Scholarship 2011-2013 Ontario Graduate Scholarship 2011 Graduate Research Thesis Award 2010-2012 The London Strategic Training Initiative in Cancer Research: PhD Award 2010-2013 Western Graduate Research Scholarship 2008 Society of Graduate Student Teaching Assistant Award 2007-2008 USC teaching Honor Roll: Award of Excellence 2006-2008 Western Graduate Research Scholarship 2005 Dean’s Honor List

214

Related Work Experience

Teaching Assistant (Pharmacology 3580Z, Pharmacology laboratory) University of Western Ontario September 2011 to April 2015

Teaching Assistant (Pharmacology & Toxicology 3560a, Introductory Toxicology) University of Western Ontario September 2010 to December 2010

Research Technician Lina Dagnino, University of Western Ontario October 2008 to November 2009

Teaching Assistant (Biochemistry 3387G, Clinical Biochemistry Laboratory) University of Western Ontario January 2007 to April 2008

Research Assistant David Haniford, University of Western Ontario August 2005 to April 2006

Peer-Reviewed Publications

Cdh1 regulates E2F1 degradation in response to differentiation signals in keratinocytes. Singh, R.K. and Dagnino, L. Oncotarget. 2016. In press.

E2F1 interactions with hHR23A inhibit its degradation and promote DNA repair. Singh, R.K. and Dagnino, L. Oncotarget. 2016. 7(18): 26275-26292.

Modulation of type II TGF-β receptor degradation by integrin-linked kinase. Vi, L., Boo, S., Sayedyahossein, S., Singh, R.K., McLean, S., DiGuglielmo, G.M., Dagnino, L., J Invest Dermatol. 2015. 135(3):885-894.

The nucleoid binding protein H-NS acts as an anti-channeling factor to favor intermolecular Tn10 transposition and Dissemination. Singh, R.K., Liburd, L., Wardle, S.J. and Haniford, D.H., J. Mol. Biol. 2008. 376: 950–962.

The global regulator H-NS binds to two distinct classes of sites within the Tn10 transpososome to promote transposition. Ward, C.M., Wardle, C.J., Singh, R.K. and Haniford D.B. Mol. Microbiol. 2007 64(4): 1000-1013.

215

Invited Book Chapter

Post-transcriptional regulation of E2F transcription factors: Fine-tuning DNA repair, cell cycle progression and survival in development and disease. Dagnino, L., Singh, R.K. and Judah, D. "DNA Repair/Book 4", (ed. Kruman, I.) InTech July 2011 (ISBN 978-953-307-687-3), Pp. 162-184.