A Study of Photodynamic Damage to the Dna Replication System

Home , Phenanthroline, Photodynamic therapy

A STUDY OF PHOTODYNAMIC DAMAGE TO THE DNA REPLICATION SYSTEM

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of

Philosophy in the Graduate School of The Ohio State University

Ran Zhao, M.S.

Ohio State Biochemistry Program

The Ohio State University

2009

Dissertation Committee

Dr. Robert M. Snapka, Advisor

Dr. Marshall V. Williams

Dr. Donald H. Dean

Dr. Charles E. Bell

Ran Zhao

2009

ABSTRACT

Photodynamic therapy (PDT) is a promising clinical modality for killing unwanted

cells, especially cancer cells by the combined use of light, photosensitizers and molecular

oxygen. In the process of PDT, energy flows from light-activated triplet state photosensitizers to the triplet ground state molecular oxygen by the type II pathway, converting it to singlet oxygen, which can damage cancer cells by directly reacting with nearby biomolecules, or indirectly by destroying tumor vasculature and invoking systemic immune responses. To improve the efficiency of PDT in clinical practice, we studied photodynamic damage to important nuclear proteins involved in DNA replication/repair, and identified the [Ru(tpy)(pydppn)]2+complex as a very promising

photosensitizer.

PCNA, p53, SV40 large T-antigen, topoisomerase I and lamin B were identified as

cellular protein targets for photodynamic damage caused by proflavine and visible light.

Following photodynamic damage, p53 was detected in a distinct covalent crosslinking

profile, with p53 tetramers and dimers being the predominant forms; SV40 large

T-antigen was mainly crosslinked into a hexamer while lamin B was crosslinked to dimers, trimers and tetramers.

2+ [Ru(tpy)n(pydppn)2-n] (n = 0, 1) complexes have been reported to be capable of

photosensitizing the generation of singlet oxygen with near 100% efficiency in vitro. In

our study, [Ru(tpy)(pydppn)]2+ caused efficient covalent oligomerization of cellular

2+ 2+ PCNA and p53 in light. In the cell lysates, both [Ru(tpy)(pydppn)] and [Ru(pydppn)2] were able to produce efficient PCNA and p53 photodynamic crosslinking. Cellular

PCNA photocrosslinking caused by [Ru(tpy)(pydppn)]2+ increased linearly with

[Ru(tpy)(pydppn)]2+ concentration, time of uptake, or visible light exposure. The

inclusion of azide, a singlet oxygen quencher, led to significant suppression of p53 photocrosslinking, suggesting that singlet oxygen is the reactive agent causing p53

photocrosslinking. [Ru(tpy)(pydppn)]2+ caused efficient photodynamic protein-DNA

crosslinking in cells, which increased with increasing levels of photodynamic damage.

Photodynamic damage by [Ru(tpy)(pydppn)]2+ resulted in inhibition of DNA replication

in a biphasic manner in mammalian cells. Another ruthenium containing complex,

[Ru(bpy)2(CH3CN)2]Cl2 inhibited SV40 viral DNA replication to a great extent in a light-

dependent manner, but no protein damage of the types found for photodynamic

ruthenium complexes (protein-protein crosslinking, protein-DNA crosslinking) was

detected by Western blotting.

We also studied the contribution of photodynamic properties of certain topoisomerase poisons in improving their cytotoxicity. It was shown that the topoisomerase II poison

ellipticine reduced cell survival to a significantly greater extent when irradiated with

visible light than in the dark conditions. However, the topoisomerase-DNA-ellipticine

cleavable complex appears not to be the relevant cytotoxic target.

iii

Dedicated to my parents

ACKNOWLEDGMENTS

I owe my deepest gratitude to all people who have helped and encouraged me during

my graduate study in the Ohio State University.

I would like to give sincere thanks to my advisor, Dr. Robert M. Snapka, for his

guidance and inspiration, for teaching me how to think critically and deeply on scientific

questions, as well as for his encouragement during my research and study. His perpetual

enthusiasm in research had motivated all the advisees, including me. Special thanks are

given to my current committee members, Dr. Marshall V. Williams, Dr. Donald H. Dean

and Dr. Charles E. Bell, and two of my former committee members, Dr. Duxin Sun and

Dr. Gary E. Means for their valuable suggestions and comments.

I want to thank Mrs. Edie Yamasaki for her great help in both research and personal matters. I want to thank the former and current members in Snapka laboratory for their company and help. I am particularly grateful to Dr. SooIn Bae and Dr. Ragu Kanagasabai for being a constant source of encouragement and support for my study and life. The discussions and interactions with you had inspired me deeply both in my research and personal life, and made the long hours spent in the lab fruitful and memorable. Thanks are also given to the members of Radiobiology Division including Dr. Qianzheng Zhu,

Dr. Qien Wang and Dr. Haiming Ding for their help, kindness and valuable suggestions.

I thank Dr. Randolph P. Thummel (University of Houston) and Dr. Claudia Turro

(Department of Chemistry, the Ohio State University) for providing ruthenium complexes and help with the revision and submission of the ruthenium paper,

Dr. Altaf A. Wani (Radiology Department, the Ohio State University) for providing mouse anti-p53 antibodies, ts 85 cell lines, Dr. M. Yoshida (University of Tokyo, Japan)

for providing Leptomycin B.

Last but not least, I want to thank my family for their unflagging love and support. I

am deeply grateful to my husband, Xiaoqiang Liu, for his encouragement, understanding and support. I am deeply indebted to my dad for his support and encouragement. I can’t

find suitable words to express my gratitude for my mom, whose unconditional love has

been my greatest strength all along. The constant love and support from my sister is

sincerely acknowledged. Without all the support and help, I could not be able to finish

this degree at the Ohio State University.

VITA

1998-2002 ...... B.S., Biotechnology Shandong Agricultural University, China

2002-2005 ...... M.S., Cell Biology Institute of Botany the Chinese Academy of Sciences, China

2005-2009 ...... PhD Graduate Student, Biochemistry The Ohio State University

PUBLICATIONS

1. Ran Zhao, Richard Hammitt, Randolph P. Thummel, Yao Liu, Claudia Turro, Robert M. Snapka. (2009). Nuclear targets of photodynamic tridentate ruthenium complexes. Dalton Trans, DOI: 10.1039/b913959a 2. SooIn Bae, Ran Zhao, Robert. M. Snapka. (2008). PCNA damage caused by antineoplastic drugs. Biochem Pharmacol, 76, 1653-1668

3. Xiaoqiu Du, Qiying Xiao, Ran Zhao, Feng Wu, Qijiang Xu, Kang Chong, Zheng Meng. (2008). TrMADS3, a new MADS-box gene, from a perennial species Taihangia rupestris (Rosaceae) is up-regulated by cold and experiences seasonal fluctuation in expression level. Dev Genes Evol, 218, 281-292

FIELDS OF STUDY

Major Field: Biochemistry

vii

TABLE OF CONTENTS

Abstract ...... ii

Dedication ...... iv

Acknowledgments...... v

Vita……………...... vii

List of Tables ...... x

List of Schemes ...... xi

List of Figures ...... xii

Chapters:

1.Introduction

1.1 Photodynamic damage ...... 1 1.2 Proteins important for DNA replication ...... 10 1.3 Transition metal complexes as photosensitizers(PSs) ...... 14 1.4 Simian virus 40 (SV40) as a model system for mechanistic studies of DNA replication inhibiting drugs ...... 19 2. Identification of protein targets of photodynamic damage in cell culture system

2.1 Introduction ...... 26 2.2 Materials and methods ...... 30

viii

2.3 Results ...... 33 2.4 Discussion ...... 41 3. A noval singlet oxygen producer:[Ru(tpy)(pydppn)]2+

3.1 Introduction ...... 70 3.2 Materials and methods ...... 71 3.3 Results ...... 76 3.4 Discussion ...... 84 4. Analysis of the inhibitory effect on DNA replication by light-activated [Ru(bpy)2(CH3CN)2]Cl2

4.1 Introduction ...... 107 4.2 Materials and methods ...... 109 4.3 Results ...... 111 4.4 Discussion ...... 113 5. Analysis of the cytotoxic effect of photodynamic topoisomerase poisons on mammalian cells

5.1 Introduction ...... 120 5.2 Materials and methods ...... 122 5.3 Results ...... 126 5.4 Discussion ...... 128

Bibliography ...... 145

LIST OF TABLES

Table Pages

1.1 p53 post-translational modifications ...... 22

LIST OF SCHEMES

Scheme Pages

3.1 Tridentate ruthenium complexes ...... 86

LIST OF FIGURES

Figure Page

1.1 Reactions of excited triplet state photosensitizers by type I and type II

pathways ...... 23

1.2 Structure of some transition metal complexes and ligands ...... 24

1.3 The pattern of pulse-labeled SV40 viral DNA replication intermediates by

one-dimensional gel electrophoresis ...... 25

2.1 Photodynamic crosslinking and chemical crosslinking of p53 protein ...... 47

2.2 The effect of p53 accumulation on detection of p53 photodynamic crosslinking .... 49

2.3 Tests of the role of singlet oxygen in p53 photodynamic crosslinking ...... 51

2.4 Ubiquitination is not involved in p53 photodynamic crosslinking ...... 53

2.5 Tests for light-dependent covalent crosslinking of PCNA trimers ...... 55

2.6 Proflavine, hypericin and NPe6 photo-crosslinking of PCNA trimers

in CV-1 cells lytically infected with simian virus 40 ...... 57

2.7 Test for type I pathway and other ROS ...... 59

xii

2.8 Ubiquination is not involved in photodynamic PCNA crosslinking ...... 61

2.9 High molecular weight PCNA forms produced by glutaraldehyde ...... 63

2.10 Band depletion of topoisomerase I on SDS PAGE in response

to photodynamic damage ...... 65

2.11 Photodynamic crosslinking of SV40 large T-antigen ...... 67

2.12 Photodynamic crosslinking of lamin B ...... 69

3.1 PCNA photodynamic crosslinking by the ruthenium complexes ...... 88

3.2 Photocrosslinking of cellular PCNA by the ruthenium complexes ...... 90

3.3. PCNA crosslinking in cell lysates ...... 92

3.4. Linearity of PCNA photocrosslinking ...... 94

3.5 Ligands alone cause only negligible crosslinking of PCNA in light ...... 96

3.6 Covalent photocrosslinking of p53 subunits in GM639 cells ...... 98

3.7 Photodynamic crosslinking of p53 in cell lysates ...... 100

3.8 Protein-DNA photocrosslinking ...... 102

3.9 Chromosomal crosslinking assay ...... 104

3.10 Photodynamic inhibition of DNA replication by [Ru(tpy)(pydppn)]2+ ...... 106

4.1 PCNA can not be photodynamically crosslinked by

xiii

[Ru(bpy)2(CH3CN)2]Cl2 and the controls ...... 115

4.2 GF/C assay of SV40 viral DNA-protein crosslinking by

[Ru(bpy)2(CH3CN)2]Cl2 in light ...... 117

4.3 3H-Tdr fluorography of SV40 viral DNA replication intermediates

in response to proflavine in light ...... 119

5.1 Topoisomerase I/II status in the drug resistant mutant cell lines ...... 132

5.2 PCNA crosslinking caused by camptothecin in light ...... 134

5.3 The cytotoxicity of camptothecin and ellipticine for CV-1 cells in the dark by

MTT assay ...... 136

5.4 The cytotoxicity of camptothecin and ellipticine for CV-1 cells

in the dark and light by MTT assay ...... 138

5.5 The cytotoxicity of camptothecin and ellipticine for CV-1 cells

and mutant cells in light by MTT assay ...... 140

5.6 The cytotoxicity of ellipticine for CV-1 cells in light by

methylene blue growth inhibition assay ...... 142

5.7 The cytotoxicity of camptothecin for CPTCV10c22 cells in light

by clonogenic survival assay ...... 144

xiv

CHAPTER 1

INTRODUCTION

1.1 Photodynamic damage

Photodynamic damage refers to photo-induced damage caused by light-activated photosensitizers (a light-absorbing chemical compound that readily undergoes photo- excitation when exposed to light) and molecular oxygen. Three factors are required to produce photodynamic damage: compounds with photochemical properties

(photosensitizers), molecular oxygen and light (Castano, et al., 2004; Oleinick and Evans,

1998). In the past three decades, photodynamic damage has received more and more attention and has been actively exploited in clinical practice in the form of photodynamic therapy (PDT), a promising anti-cancer treatment modality (Detty, et al., 2004;

Dougherty, 2002). Besides an anti-tumor treatment, PDT has been extensively used in dermatology, age-related macular degeneration, anti-virus infection, and many other non- cancer diseases (Brown, et al., 2004; Detty, et al., 2004; Dougherty, 2002; Wainwright,

1998; Triesscheijn, et al., 2006)

Singlet oxygen is believed to be the major player in PDT (Ochsner, 1997; Weishaupt, et al., 1976). In the process of PDT, photosensitizers will be activated to the excited state

from the ground state by absorbing energy from light. In the presence of molecular

oxygen, energy transfer can take place from the activated photosensitizer to the ground

state molecular oxygen by the type II reaction to allow the photosensitizer to relax to the

ground state with the concomitant generation of singlet oxygen (Castano, et al., 2004;

Sharman et al., 1999). Singlet oxygen is a highly reactive and cytotoxic species as it can

rapidly react with nearby bimolecules, which in turn perturbs various cellular functions

and pathways, ultimately leading to necrotic and/or apoptotic cell death (Castano, et al.,

2004; Castano, et al., 2005; DeRosa and Crutchley, 2002). The efficacy of PDT results from the interaction of all three factors: the light energy level, the properties of photosensitizers, and the availability of molecular oxygen. Among them, the first two factors can be readily controlled in the practice of PDT.

Light

Cellular endogenous chromophores (hemes, melanin, etc) and photo-reactive proteins

can both interfere with the efficiency of PDT (Sibata, et al., 2000; Wilson and Patterson,

2008). The most abundant protein circulating in the blood capable of absorbing visible light is hemoglobin, whose absorption peaks are near 425, 544, and 577 nm (Sibata, et al.,

2000). Therefore, light with wavelengths above 600 nm is required as the irradiation source for PDT to bypass the absorption of hemoglobin and thus penetrate into the targeted tissue. However, water molecules absorb strongly at wavelengths beyond 1200 nm, while photochemical reactions might not occur at the wavelengths from 850 nm to

900 nm as the photons are not adequately energetic. Therefore, to strike a balance for efficient photodynamic effect, the ideal photosensitizer should be able to absorb at

wavelengths between 600 nm and 800 nm, which is referred as “the PDT therapeutic

window” (Sibata, et al., 2000). Photosensitizers themselves might also absorb light, therefore, the fraction of light absorbed by photosensitizers needs to be subtracted from the total amount of irradiance. This phenomenon is known as photosensitizer self-

shielding (Henderson and Dougherty, 1992). Photobleaching, a phenomenon resulting

from self-destructive reactions of photosensitizers in the excited-state, also leads to

reduced light absorbance and subdued function of photosensitizers (Henderson and

Dougherty, 1992).

Photosensitizers

1. First and second generation photosensitizers

From the clinical point of view, the currently available photosensitizers can be

broadly classified as first generation, second generation and third generation (DeRosa and

Crutchley, 2002; Josefsen and Boyle, 2008). The first generation photosensitizers are basically hematoporphyrin derivatives (HPD) (Henderson and Dougherty, 1992; Lipson

and Baldes, 1960; Lipson, et al., 1967). Photofrin (porfimer sodium), the first and the only photosensitizer approved by FDA (Food and Drug Administration) in the United

States, is a semi-purified HPD and a typical example of first generation photosensitizers

(Dougherty, et al., 1998). Up to now, it is still the most widely used photosensitizer in

clinical trials (Sharman, et al., 1999). Photofrin has been used in limiting such diseases as

non-small cell lung cancer, esophageal cancer and Barrett’s Esophagus (Baert, et al.,

1993; Dougherty, 2002; Furuse, et al., 1993; Sharman, et al., 1999). Compared to the first

generation photosensitizers, the second generation photosensitizers are advantageous in 3

one or more of the following aspects: (1) they are usually pure and their chemical

structures are known; (2) they usually absorb in the range between 630-800 nm, which

lies in the preferred 600-800 nm “PDT therapeutic window”; (3) they usually do not have

a prolonged subcutaneous photosensitivity upon completion of PDT (Sharman, et al.,

1999; Sibata, et al., 2000). Third generation photosensitizers are second generation

photosensitizers bound to carriers for selective tumor accumulation (DeRosa and

Crutchley, 2002). The following are several examples of photosensitizers used in

preclinical and clinical trials.

(1) δ-aminolevulinic acid (ALA)

Among the known photosensitizers, ALA is unique in that it is the precursor of the

naturally occurring photosensitizer PPIX (protoporphyrin IX) in the heme biosynthetic pathway. ALA-induced PDT, one of the most active areas in the clinical practice of PDT, has been successfully applied in various medical fields (Ladner, et al., 2001; Peng, et al.,

1997). However, due to the fact that ALA has little absorbance beyond 630 nm, it is

mainly used for treating superficial lesions.

(2) Benzoporphyrin Derivative-Monoacid Ring A (BPD-MA)

BPD-MA is in a liposomal formulation and can absorb up to 692 nm, which

facilitates cellular uptake and light penetration during PDT. Currently, it has been used

for treating age-related macular degeneration, choroidal melanoma, atherosclerotic

plaques and psoriasis (Allison, et al., 1997; Husain, et al., 1996; Taylor, et al.,1999).

(3) Tin ethyl etiopurpurin (SnET2 or Purlytin)

SnET2 is a chlorin photosensitizer and it has approval for clinical trials from the FDA

to treat cutaneous metastatic breast cancer and Kaposi’s sarcoma in patients with 4

acquired immunodeficiency syndrome as well as age-related macular degeneration and

prostate cancer (Razum, et al., 1996).

(4) Lutetium texaphyrin/Lutex

Lutetium texaphyrin/Lutex is hydrophilic and can be enriched in the malignant structures which bear more lipoprotein receptors. Lutex-based PDT has been used in treating cardiovascular diseases and skin lesions (Woodburn, et al., 1998).

(5) Tetra (m-hydroxyphenyl)chlorin (mTHPC or Foscan) and NPe6

Tetra (m-hydroxyphenyl)chlorin has been tested in clinical trials for treating recurrent

head and neck cancers (Grosjean, et al., 1996). mTHPC mediated PDT has been shown to

be effective in killing cells with multidrug resistance phenotypes (Teiten, et al., 2001).

Mono-L-aspartyl chlorin e6 (NPe6), which absorbs at 664 nm, has been tested for

treating superficial malignancies of the skin and nasopharynx (Taber, et al., 1998).

(6) Silicon phthalocyanine (Pc 4)

Pc 4 has emerged as a promising photosensitizer for PDT. It has a high extinction

coefficient at 672 nm, which greatly facilitates light penetration into the tissue and it can

be quickly cleared from tissue. It has been tested for treating subcutaneous T-cell

lymphoma in phase I clinical trials ( Lee, et al., 2008; Oleinick, et al., 1993).

2. The structure and subcellular localization of photosensitizers

As singlet oxygen can not diffuse far (the radius of diffusion is around 20 nm) from

the site where it is generated during photodynamic reaction due to its high reactivity, the

potency of photodynamic damage is related to the subcellular localizations of photosensitizers (Triesscheijn, et al., 2006). Due to the differences in their chemical 5

structures, photosensitizers can be roughly classified as comparatively hydrophilic (easily

bind to albumin), amphiphilic (have the tendency to insert into the phospholipids and the

apoprotein layer of lipoprotein particles), or hydrophobic (possibly insert into the lipid

core of lipoproteins) (Castano, et al., 2005). Following uptake into cells, photosensitizers

can localize at mitochondria (Photofrin and others), lysosomes (Npe6 and others), plasma

membranes (monocationic porphyrin based photosensitizer), Golgi (Foscan and others) and endoplasmic reticulum (hypericine, Foscan, and others), nuclei (photosensitive dyes

like methylene blue) as well as tumor vasculature. The properties of photosensitizers to localize in different cellular structures allow different site specific photodynamic damage for different drugs (Castano, et al., 2004). Generally, photosensitizers located or produced in mitochondria (for instance PPIX) are more likely to induce apoptosis; plasma membrane targeting photosensitizers tend to result in disruption of membrane structure and necrosis; lysosome locating photosensitizers tend to cause rupturing of lysosome and the subsequent release of lysosomal enzymes into the cytosol (Dougherty, et al., 1998).

The tumor vasculature targeting photosensitizer can damage the structure of vasculature and thus deprive the nutrient and oxygen supply to the tumor, leading to the shrinkage or disappearance of tumors (Castano, et al., 2005; Henderson and Dougherty, 1992).

Singlet oxygen and other reactive oxygen species

Photodynamic reaction is a light-activated process that involves a series of

biophysical and photochemical steps. By absorbing photons from light, the

photosensitizers will be activated and then the exited state will decay by different routes,

depending on such factors as the microenvironment and the chemical structure of 6

photosensitizers, leading to energy transfer between different molecules (see Fig 1.1).

The involved reactions can be categorized into three classes: type I, type II and type III

(Sibata, et al., 2000). Type I reactions are radical mediated. The radical is generated by the exchange of an electron or hydrogen atom between the activated photosensitizers and nearby molecules. A mixture of reactive oxygen species including superoxide anion, hydroxyl radical and peroxide can be produced in this way. Superoxide anion is frequently the first reactive oxygen species to be generated in type I reactions, and superoxide can subsequently lead to the formation of other reactive species via different routes (Castano, et al., 2004). First, superoxide can be catalyzed by SOD (superoxide dismutase) to generate hydrogen peroxide (H2O2) and oxygen. Second, superoxide can

react with metal ions as reducing agents, and the reduced metal ions can serve as catalyst

• in breaking up the oxygen-oxygen bond in H2O2 to produce one hydroxyl radical (HO )

and one hydroxide ion (HO-). As a free radical, hydroxyl radical is highly dangerous to

the cell as it can trigger a free radical chain reaction and pose massive oxidative assault to the lipids, fatty acids, DNA as well as other molecules. Third, superoxide can interact with HO• to form singlet oxygen, or with nitric oxide (NO-) to produce peroxynitrite

(OONO−), which is also a highly reactive species (Castano, et al., 2004). Type II

1 reactions are singlet oxygen ( O2) mediated (Castano, et al., 2004; Sibata, et al., 2000). In

this process, energy is transferred from the activated triplet state photosensitizer to

molecular oxygen to generate singlet oxygen (DeRosa and Crutchley, 2002). In contrast

to type I and II, type III is not a photo-oxidation reaction, instead, it is the interaction

between the activated triplet photosensitizer and the existing free radicals (Laustriat,

1986). These three types of reactions can occur simultaneously, with certain type of 7

reactions being the predominate form. At the molecular level, by absorbing the energy from light, the photosensitizer in the ground state can be raised to the singlet state, which is very unstable and short-lived. Photosensitizers in this state can go back to ground state by emitting fluorescence or heat, or decay through Type I and Type III reactions, or they can undergo intersystem crossing to the triplet state, which is also unstable but

comparatively longer-lived than the singlet state. The photosensitizer in the triplet state

can decay to the ground state by emission of phosphorescence or Type I and Type III

reactions, or react with molecular oxygen to produce singlet-oxygen by type II reactions

(Sibata, et al., 2000). Although the involvement of other forms of reactive oxygen species

can not be easily excluded, singlet oxygen has been accepted as the main functional agent

in PDT (Ochsner, 1997; Weishaupt, et al., 1976)

Due to its high reactivity and short life-time (less than 40 ns in the biological system),

only molecules within 20 nm of the location of the photosensitizer molecule can be

targeted by singlet oxygen. Proteins and DNA are two most important targets for singlet

oxygen. Among the 20 naturally occurring amino acids, cysteine, methionine, tyrosine,

histidine and tryptophan are most reactive to singlet oxygen. Through oxidation, free

cystenine can be easily oxidized to disulphide, methionine to sulfoxides, histidine to

endoperoxide, tryptophan to N-formylkynurenine and kynurenine, and long-lived

peroxides have also been reported to form on tyrosines (Davies, 2003; Wright, et al.,

2002). In DNA, either the bases (mostly guanine) or the sugars in the ribose ring can be

oxidatively damaged by singlet oxygen (Castano, et al., 2004). DNA can also be

crosslinked to proteins (Castano, et al., 2004). This form of damage can greatly interferes

with normal DNA metabolism as the bulky DNA-protein crosslinking is much more 8

difficult for the cell to remove and repair than mutations at the nucleotide level. Lipid can also be targeted by singlet oxygen and be oxidized to lipid hydroperoxides (Bachowski et al., 1994; Bachowski, et al., 1991) .

The cellular responses to photodynamic damage at the molecular level

Current studies on photodynamic damage can be divided into two major directions: one is on the improvement of PDT efficiency in clinical or pre-clinical trials by developing better photosensitizers, using more advanced light sources and seeking more effective methods for delivering photosensitizers into the target tissues; the other focuses

on the mechanism and effect of PDT. At the molecular level, photodynamic reactions

might damage cells through the following three routes: 1) direct damage to subcellular

targets, which in turn leads to the death of cells of interest; 2) damage to the tumor

vasculature; 3) induction of a systemic immune response (Castano, et al., 2005).

The accumulated data have shown that the photodynamic effect can influence the

whole cellular machinery: signal transduction (such as activation of MAPK family, release second messengers like ceramide and sphingomyelin), transcription (such as activation of transcription factors like NF-κB), cell adhesion, stress response (such as

over-expression of heat shock proteins), angiogenesis, apoptosis and necrosis (Castano, et

al., 2005).

The majority of recent studies focuses more on the photodynamic effect on

mitochondria-dependent apoptosis than that on DNA metabolism. As the proper regulation of DNA replication and DNA repair machinery is the key to the prevention of cancer development, the study of normal/stalled DNA replication fork in the setting of 9 photodynamic treatments holds great potential for us to gain a deeper understanding of carcinogenesis at the molecular level. Therefore, we conducted a study to examine the photosensitivity of several key proteins important for the DNA replication system.

1.2 Proteins important for DNA replication

Proliferating cellular nuclear antigen (PCNA)

PCNA was originally identified as a DNA elongation factor for in vitro SV40 DNA replication (O'Donnell and Kuriyan, 2006). Similar to the β clamp in prokaryotes, PCNA is an indispensable component in the mammalian DNA replication machinery which functions as a processivity factor for DNA polymerase δ and polymerase ε (Essers, et al.,

2005). The functional form of PCNA is a ring-shaped trimer (Krishna, et al., 1994). In the course of its normal metabolism, PCNA shifts between two states: chromatin-bound state and nucleoplasmic state. In the process of DNA replication, PCNA encircles the DNA duplex and slides along it with the tethering of polymerases (Essers, et al., 2005). In mammalian cells, a clamp loader called RFC (Replication factor C) is required to load the

PCNA ring onto the primed DNA template at the expense of ATP hydrolysis (Moldovan, et al., 2007). Recent investigations have proposed that PCNA might also be involved in the following activities: cell-cycle progression, gene transcription, DNA methylation, chromatin assembly and remodeling, sister chromatid cohesion, cell cycle regulation and

DNA repair of different types (Moldovan, et al., 2007). Taken together, PCNA is an essential protein for cell growth and cell proliferation.

p53

p53 is the well known “Genome guardian”, as emphasized by the fact that at least

half of human cancers contain a deletion or mutation of TP53 gene (Hainaut and

Hollstein, 2000; Hollstein, et al., 1991; Liu and Kulesz-Martin, 2001; Wang, et al.,

2009). In response to stress signals, p53 is stabilized and accumulated in the nucleus to

regulate the expression of various proteins involved in cell cycle arrest, senescence, or

apoptosis. The diverse cellular response to stress of p53 is regulated, at least in part, by

post-translational modifications (see Table 1.1).

p53 is negatively regulated by MDM2 which functions as a ubiquin-E3 ligase

targeting p53 for degradation in the cytoplasm (Haupt, et al., 1997; Kubbutat et al., 1997).

In response to DNA damage or other stress signals, p53 is phosphorylated at its N-

terminus and dissociates from MDM2, which in turn leads to the stabilization and

activation of p53. p53 is a transcription factor, and it functions by regulating the

transcription of various proteins in diverse cellular pathways. It can also work by directly

interacting with downstream binding partners to facilitate cell-cycle arrest or apoptosis

(Vogelstein et al.,2000).

As the enhancement of cellular sensitivity to chemotherapy or radiotherapy is

promising for better cancer treatment, much research has been pursued for this end, with

p53 being one of the most studied targets at the molecular level. However, there is still no

definitive conclusion for the contribution of p53 in this regard. Some studies demonstrate

that wild-type p53 plays an important role in keeping cells sensitive to chemotherapy and/or radiotherapy; however, others have come to opposite conclusions (Gallardo, et al.,

1996; Huang, et al., 1996; Lowe, et al., 1994). Recently, studies have been extended to 11 exploring the sensitivity of p53 to photodynamic therapy. SV40-transformed IMR-90 cells in which p53 function had been disrupted were reported to be more sensitive to hematoporphyrin derivative (HPD)-mediated PDT than the parent cell lines (Denstman et al., 1986). Wild-type p53 has been reported to enhance the anti-tumor efficiency of PDT in CaSki cells (a human cervical cancer cell line) and nude mice implanted with CaSki

(Lim, et al., 2006). Similar phenomena have also been observed in human promyelocyte leukemia HL60 cells (Fisher, et al., 1997). However, another report demonstrated that p53 had no influence on the sensitivity of colon or breast carcinoma cells to PDT based on the colony survival assay (Fisher, et al., 1999). It has also been reported that p53 may not be important in PDT-mediated cell killing or induction of apoptosis in osterosarcoma cell lines in response to hypericin-based PDT (Lee, et al., 2006). The response of p53 to

PDT has been studied in p53-null HT29 human colorectal carcinoma cell line (Zhang, et al., 1999). Growth inhibition was observed in the transfected cells without the induction of apoptosis when wild-type p53 was introduced into these cells; however, following

PDT, these cells were more sensitive to apoptosis initiation than their untransfected counterparts (Zhang, et al., 1999). Taken together, it is still an open question as to the significance of p53 in the cellular sensitivity to PDT, and there is no report of photodynamic damage to p53 protein itself to date.

Topoisomerase I/II

In the course of DNA metabolism, the supercoiled DNA templates must be rearranged properly before proceeding to the next steps. Topoisomerase I/II play essential roles in this process (Champoux, 2001). They work on DNA strands by introducing 12

transient DNA strand breaks for other strands to pass through, and then re-ligating the

strands. In the process of strand breaking and resealing, topoisomerase I/II are covalently

attached to the DNA template by forming a phosphodiester bond between the active

tyrosine residue in topoisomerase I/II and the DNA break site. Topoisomerase I relaxes

DNA by cutting on a single DNA strand while topoisomerase II can cleave and rejoin

double-stranded DNA simultaneously at the cost of ATP hydrolysis (Osheroff, 1989).

The activity of topoisomerase I is crucial for the initiation of DNA replication and transcription in that the highly supercoiled DNA duplex must be relaxed to allow for the following protein-DNA interactions. For topoisomerase II, not only can it bring in or

remove supercoils, it also helps in the decatenation of daughter chromosomes and

interlocked DNA duplexes. Topoisomease I and II can substitute one another in DNA replication and transcription, however, only topoisomerase II can decatenate double-

stranded DNA (Wang, 2002). In mammalian cells, there exist two topoisomerase II

isozymes: topoisomerase IIα (p170) and topoisomerase IIβ (p180).

Lamin B

Laminar filaments are important components of nuclear skeleton. There are A, B, C types of lamins. While the expression of A- and C- type lamins are subject to tissue specificity or development, B- type lamins are ubiquitous (Goldman, et al., 2002;

Gruenbaum, et al., 2005). Besides their role in supporting cellular structure, lamina are

required for proper cell cycle regulation, chromatin organization, DNA replication, cell

differentiation and apoptosis. Inside the nucleus, lamins can be found in the laminar layer

next to the inner surface of the nuclear envelope or throughout the nucleoplasm. During 13

mitosis, the association and dissociation of lamins is regulated by

phosphorylation/dephosphorylation on serine and threonine residues, which drives the

assembly and disassembly of the structure of nuclear envelope (Goldman, et al., 2002;

Gruenbaum, et al., 2005). By using DTHe (dimethyl tetrahydroxyhelianthrone) as the

photosensitizer, Lavie et al have reported the covalent polymerization of lamin B and the

release of an 86 kDa lamin protein into the cytosol, which might be related to the loss of

structure of nucleus during photodynamic damage (Lavie, et al., 1999).

1.3 Transition metal complexes as photosensitizers (PSs)

The discovery of the great anti-tumor effectiveness of cisplatin

(cis-diamminedichloroplatinum(II)) (See Fig 1.2A) around 40 years ago has spawned great interest in the tumor-curbing properties of transition metal-based complexes (Ang and Dyson, 2006; Gianferrara, et al., 2009). In the periodic table, metals placed in group

8 to 10B and feature an incompletely filled d or f subshell are classified as transition

metals. The partially filled subshell lends multiple distinct chemical properties to this

family of metals, which include their facile bonding capability, variable valence, and a

better ionic than covalent bonding nature. The transition metal family can be further

divided into three series: the first series include iron, cobalt and nickel, which contain a

partially filled 3d subshell; the second series is comprised of ruthenium, rhodium and

palladium, in which the 4d subshell is incompletely filled; the third series include

osmium, iridium and platinum, which possess a partly filled 5d subshell. In an alternative

way, the transition metal family can be subdivided into iron group (series 1) and platinum

group( series 2 and 3) as the extended 4d and 5d orbitals contribute most to the chemical

bonding capability of this family.

Cisplatin is a platinum based drug (Rosenberg, et al.,1965; Rosenberg, et al., 1969). It

has been widely used in treating such tumors as sarcomas, small cell lung cancer, ovarian

cancer, head and neck cancer, lymphomas and several others (Alderden et al., 2006;

Hambley, 1997; Wong and Giandomenico, 1999). Previous studies have proven that cisplatin works by chloride to water ligand exchange, which facilitates its binding to

DNA duplex (Reedijk, 2008). The binding generates a covalent intra-stand d (GpG) crosslinking in DNA duplex which distorts the normal conformation and structure of

DNA and therefore impairs the regular metabolism of DNA, ultimately leading to the apoptosis of cancerous cells (Jamieson and Lippard, 1999). Other members of the cisplatin family, including carboplatin and oxaliplatin have also shown anti-cancer effectiveness (Lebwohl and Canetta, 1998). Though highly effective, the side effects such as high toxicity to normal cells and drug resistance have become more and more a limiting factor for their usage, which impels the search for non-platinum based complexes in the platinum group that can override these limitations (Perez, 1998; Stordal,

et al., 2007). Ruthenium based complexes, among other transition metal-based complexes

that have been tested, have surfaced as a family of highly promising drugs in the light of

their comparatively lower cytotoxicity, high DNA binding affinity and intrinsic

photodynamic properties (for certain family members) (Ang and Dyson, 2006; Brabec and Nováková, 2006; Clarke, 2003; Herman, et al., 2008).

Endowed with a wide range of oxidation states (from -2 to +8), ruthenium based

complexes are highly redox-active and versatile in ligand coordination. In coordinated

ruthenium complexes, electronic transitions can be ligand centered (LC), metal centered

(MC), or on a charge transfer from the metal to a specific ligand (MLCT), where the п*

antibonding orbital lies in the lowest possible energy level (Meyer, 1986). Ruthenium

based complexes with a myriad of ligands have been synthesized and tested during the

last thirty years. The ruthenium containing complexes can be grouped into the following

families: polypyridyl-Ru complexes, polyaminocarboxylate-Ru complexes, DMSO-Ru

complexes, arylazopyridine-Ru complexes, organometallic arene-Ru complexes (Ang

and Dyson, 2006; Clarke et al., 1999). Two ruthenium-based anticancer drugs (NAMI-A

and KP1019) have passed phase 1 clinical trials and will soon enter phase 2 trials (See

Fig 1.2B) (Gianferrara, et al., 2009). It has been proposed that the observed anti-tumor

effectiveness of ruthenium based complexes derive mainly from the following two

mechanisms: the first is “activation by reduction”, which takes advantage of the local

characteristics of tumor microenvironment (reductive and more acidic than normal tissues)

to reduce the less active ruthenium (III) complexes to the more active ruthenium (II) oxidation state (Clarke, 2003). In this manner, ruthenium based complexes are selectively toxic toward tumor targets. Another mechanism is the mimicry for iron in occupying the iron-binding sites on transferrins. The higher demand of nutrient and iron for the growth of fast-dividing cancer cell results in a higher number of transferrin receptors on their

cellular surface. The ruthenium carrying transferrin might thus gain the selective

accumulation in tumors (Ang and Dyson, 2006; Clarke, 2003). It has been suggested that

the toxicity of ruthenium complexes is related to their DNA- binding property, though

their interaction to components of mitochondria and extracellular matrix might also 16

contribute to their antineoplastic effectiveness (Clarke, 2003; Fruhauf and Zeller, 1991;

Gallori, et al., 2000; Zhang and Lippard, 2003)

With the development of photodynamic therapy and increased knowledge of

ruthenium complexes, the application of ruthenium complexes as photosensitizers in

anti-cancer treatment has raised much interest in recent years (Clarke, 2003). The

photoluminescence and DNA binding properties of certain polypyridyl-ruthenium

complexes have been recognized much earlier, which lead to their extensive use as DNA

molecular probes (Lo, 2007; Metcalfe and Thomas, 2003; Pierard and

Kirsch-De Mesmaeker, 2006). Among the numerous polypyridyl ligands, the most

common ones are 2, 2’-bipyridine (bpy), 1,10-phenanthroline (phen) and 2,2’:6’2’’-

terpyridine (terpy) (See Fig 1.2C) (Ang and Dyson, 2006). cis-(Cl,Cl)-[RuII(terpy)-

(NO)-Cl2]Cl was reported to be more toxic to A2780 human ovarian carcinoma cell lines than mer-[Ru(terpy)Cl3], cisplatin or carboplatin (Karidi, et al., 2005). It was also

reported that this complex undergoes photodynamic liberation of the NO group, but no

further work was followed to test the relationship between the observed photodynamic properties and anti-tumor effect. Among the DMSO-Ru complexes, cis- and trans-Ru-

(DMSO)4Cl2 were found to be cytotoxic against human and murine melanoma cell lines

(SK-MEL 188, S91) in a photo-activated manner with the trans isomer being more potent

(Brindell, et al., 2005). They also exhibited markedly elevated reactivity towards oligonucleotides under UVA irradiation, implying a photodynamic mechanism. To date

there have been several reports of the photoinduced cleavage of DNA by ruthenium

complexes. Irradiation can lead to the photoactivation of

2+ cis-[Ru(bpy)2(NH3)2] ((bpy = 2,2’-bipyridine) through the loss of ammonia ligands and

subsequently covalent binding to DNA (Singh and Turro, 2004). Though the ligand loss

mechanism is reminiscent of cisplatin, this mechanism is light-activated. An ROS-

dependent DNA photocleavage pathway was found in [(TL)2Ru(dpp)]Cl2 complexes

(TL = 2,20-bipyridine, 1,10-phenanthroline, or 4,7-diphenyl-1,10-phenanthroline)

(Mongelli, et al., 2006). The excited MLCT (metal-to-ligand charge transfer) state of these complexes transfers energy to oxygen and generate reactive oxygen species, which lead to lesions in supercoiled pUC18 circular plasmid DNA. One porphyrin-ruthenium was reported to be able to cleave DNA and cause melanoma cell death upon light irradiation (Davia et al., 2008). Inhibition of cell growth was observed for

[{(bpy)2Ru(dpp)}2RhCl2]Cl5 and [{(bpy)2Os(dpp)}2RhCl2]Cl5 in African green monkey

kidney epithelial (Vero) cells (Holder, et al., 2007) . Through the combined use of such

techniques as electronic absorption titration, viscosity measurements, thermal

denaturation, circular dichroism, Liu et al reported that a polypyridyl complex

[Ru(phen)2(ipbd)](ClO4)2 (ipbd = 3-(1H-Imidazo[4,5-f] phenanthrolin-2-yl)-1-

benzodioxane, phen = 1,10-phenanthroline} is able to bind calf thymus DNA by

intercalating into the base pairs (Liu, et al., 2007). It has been shown that this complex can cleave plasmid pBR322 DNA upon irradiation in a dose-dependent manner.

1.4 Simian virus 40 (SV40) as a model system for mechanistic studies of

DNA replication inhibiting drugs

Simian virus 40 belongs to the subfamily of Polyomaviridae of the Papovaviridae

family. This virus has a covalently closed, superhelical and double stranded DNA with

the genomic size of around 5 kb (Fiers, et al., 1978). The genome of SV40 virus can be

divided into early region and late regions. The products of early region are regulatory

proteins named large T-antigen, small T-antigen. Both proteins share the same N-

terminus, though the C-terminus varies as a result of alternative splicing of their mRNA

transcripts. The late region encodes three structural proteins VP1, VP2 and VP3, which

are the components for the capsid. These three capsid proteins are also the products of

alternative splicing. A fourth protein, agnoprotein, is also encoded by the late region, but

the function of this protein still largely remains unknown (Fiers, et al., 1978). SV40 virus

has a narrow host range. Monkey cells are permissive for SV40 virus, rodent cells are

non-permissive hosts while human cells are semi-permissive (Ozer, et al.,1981; Zouzias

et al., 1981). CV-1, vero and BSC1 cells lines, which are derived from the African green

monkey kidney, are permissive for SV40 virus and have been used extensively to support

SV40 viral DNA replication and gene expression in cancer research.

To complete its life cycle, SV40 virus depends heavily on the cellular DNA

replication machinery, which provides us an excellent tool for probing into the dynamics

of the much more complicated mammalian DNA replication system. In addition, the viral

DNA of SV40 is packaged with histones derived from the host into minichromosomes,

which greatly facilitate our study of mammalian chromosomes. In comparison with

mammalian chromosomes, SV40 virus systems are superior in the following aspects for

the study of DNA replication: high copy number, small size of genome (5243bp),

established DNA sequence and defined genetic elements, uniformity of chromatin

structure, and the ease of viral chromatin or DNA extraction. Therefore, since its

discovery, SV40 virus has been intensively used as a model system for the study of

mammalian DNA replication, especially in studying the effect of chromatin structure on

DNA replication and transcription. Much of our current knowledge of DNA replication

and chromatin structure has been obtained from the use of this simplified system. The

first in vitro SV40 replication system was established in 1984 (Li and Kelly, 1984).

The SV40 replication system has found wide application in the study of the

mechanism of drugs/chemicals that interfere with DNA replication. Due to its small size,

the SV40 genome can be selectively extracted from cellular DNA and subjected to further analysis. When coupled with high resolution gel electrophoresis, the altered SV40

DNA replication intermediates resulting from the perturbation of DNA replication by

drugs/chemicals can be captured as “signature patterns”, which can provide clues as to

the cellular targets and working mechanisms of the drugs of interest. The well-known

anticancer drugs such as camptothecin, VM26 (teniposide), m-AMSA(amsacrine),

hydroxyurea have been shown to generate aberrant SV40 intermediates (Snapka and

Permana, 1993). It also can be used to display topological problems caused by certain

drugs which are otherwise hard to detect on mammalian cell chromosomes. Furthermore,

the SV40 virus system can be used to test anticancer drugs in crude extracts and is

insensitive to various cellular products that might cause false positives in routine

cytotoxicity assays. 20

The pattern of pulse-labeled SV40 viral DNA replication intermediates as detected by

one-dimensional gel electrophoresis is well defined (See Fig 1.3). The fully replicated

forms of viral DNA exist in three forms: form I (superhelical circular DNA, I), form II

(nicked circular DNA, II) and form III (double stranded linear DNA, III). Partially replicated viral genomes with bubbles are called intermediate Cairns structures (IC). The bubble enlarges as viral replication goes on and the size of the vial genome doubles. The ever-growing “θ-form” intermediates will make a continuous “smear” in the gel from the form I band to the LC (Late Cairns) band, in which only the terminal 5% of the viral

DNA remains to be replicated. There is a marked pause of viral replication with the appearance of late Cairns structure, which might be explained by the switch from topoisomerase I to topoisomerase II as a replication swivel (Snapka, 1996). For the newly generated daughter viral DNA, they are usually linked in the form of catenated dimers, which can be grouped into A-C families. The daughter chromosomes in the A -family dimers (A1-An) are both relaxed due to nicks in their structure. In the B-family dimers

(B1-Bn), one is nicked while the other is superhelical. In the C-family dimers (C1-Cn), both are superhelical. These catenated dimers can be separated by gel electrophoresis as higher catenation will lead to increased compactness and higher electrophoretic mobility.

On the one-dimensional gel, both A- and B- family dimers are separated in the form of ladders while the C- family members are not well-resolved and stay on the far top of the

DNA form II bands.

Modification Ref Modification Ref

(Lambert, et al., 1998; Ser15-P Tibbetts, et al., 1999) Ser315-P (Li, et al., 2000)

(Sakaguchi, et al., The18-P (Sakaguchi, et al., 2000) K320-Ac 1998)

Ser20-P (Chehab, et al., 2000) K373-Ac (Liu, et al., 1999) (Lu, et (Bulavin, et al., 1999; al.,1997;Waterman, et Ser33-P Sakaguchi, et al., 1998; Ser376-P al., 1998)

(Sakaguchi, et al., 1998; (Takenaka, et al., Ser37-P Tibbetts, et al., 1999) Ser378-P 1995)

(Sakaguchi, et al., Ser46-P (Bulavin, et al., 1999) K382-Ac 1998) (Kapoor and Lozano, 1998; Sakaguchi, et al., Ser392-P 1997) Ser315-P (Bischoff, et al., 1990)

Table 1.1 p53 post-translational modifications

Fig 1.1 Reactions of excited triplet state photosensitizers by type I and type II pathways

A B

Fig 1.2 Structure of some transition metal complexes and ligands

Fig 1.3 The pattern of pulse-labeled SV40 viral DNA replication intermediates by one-dimensional gel electrophoresis

CHAPTER 2

IDENTIFICATION OF PROTEIN TARGETS

OF PHOTODYNAMIC DAMAGE

IN CELL CULTURE SYSTEM

2.1 Introduction

Photodynamic therapy (PDT) is a clinical modality that harnesses visible light and photosensitizers (PSs) to kill unwanted cells through either apoptosis or necrosis

(Castano, et al., 2004). In medical practice, light, PSs and oxygen supply are all controllable factors that can be adjusted for optimal outcome. As the cytotoxicity of photodynamic damage is mainly determined by the activity of singlet oxygen, which is produced via the energy transfer from the excited PSs to the molecular oxygen, the quantum yield of singlet oxygen thus becomes a significant factor for the potency of photodynamic damage, though such factors as the location of the administered PSs also make their contributions to the ultimate outcome. Since the quantum yield, cellular location, etc are all related to the chemical nature of PSs, it is not surprising that much effort has been put in the synthesis, development and optimization of second generation or even third generation PSs in the past decade to overcome the limitations of the first

generation PSs and further improve the efficiency of photodynamic therapy in cancer

treatment (Castano, et al., 2005; DeRosa and Crutchley, 2002; Josefsen and Boyle,

2008). Nowadays, dozens of PSs have appeared at different stages of clinical or pre-

clinical trials, with varying photodynamic effectiveness. With the large number of

available PSs at hand, it has become more and more important to identify protein targets

for photodynamic damage as biomarkers to facilitate the comparison of the efficiency of

various PSs on the one hand and delineate the working mechanisms for PSs of interest on

the other hand. With this in mind, we used a simple but highly efficient photodynamic

dye, proflavine (a DNA intercalator), as the main photosensitizer in our in vitro cell

culture system to identify proteins targets in the DNA replication system.

The tumor suppressor protein p53 plays a pivotal role in orchestrating multiple

cellular pathways including cell cycle arrest, DNA damage/repair and apoptosis

(Vogelstein, et al., 2000). p53 was named after its apparent molecular mass of 53 kDa by

SDS polyacrylamide gel electrophoresis (SDS PAGE). Two p53 monomers bind together

to form a stable dimer, and a pair of these dimers associate to form the p53 tetramer,

which is the functional form of p53 in the cell (Joerger and Fersht, 2008). Strong detergents like SDS can disrupt the non-covalent inter-subunit binding of the tetramer so that only monomers are detected on SDS PAGE. As there are a certain level of p53 proteins in the cell that are ubiquitinated for degradation during the regular course of its life cycle, ubiquitinated p53 (in the range of 70 kDa) might also be detected (Vogelstein, et al., 2000; Vousden, 2002). Chemical agents like glutaraldehyde can crosslink p53 monomer to dimers, trimers, and tetramers that migrate at positions corresponding to p53 multiples on SDS PAGE, while higher levels of covalent crosslinking only produce a 27

tetramer (Friedman, et al., 1993; Wang, et al., 1994). When the concentration of

glutaraldehyde reaches an even higher level, it will result in additional intra-molecular

crosslinking in p53, so that the formed tetramer will become even more compact and

moves faster on SDS PAGE (Bell, et al., 2002; Wang, et al., 1994). p53 dimers are

frequently resolved into doublets (a closely spaced pair of bands), which might be caused by additional intra-peptide crosslinking leading to a more compact structure, or two different populations of p53, or two different determinates of dimerization (Friedman, et

al., 1993; Nagaich, et al., 1997) . Although there has been much research on the

significance of p53 in enhancing cellular sensitivity to photodynamic therapy, there is still no report on the response of p53 per se or p53 pathway components to photodynamic damage (Denstman, et al., 1986; Fisher, et al., 1997; Fisher, et al., 1999; Lee, et al., 2006;

Lim, et al., 2006; Zhang, et al., 1999). In this study, we chose p53 as a candidate and characterized the crosslinking pattern of p53 in response to photodynamic damage.

Proliferating cell nuclear antigen (PCNA) plays a critical role in DNA replication and repair and it is sometimes called the "ringmaster" or "maestro" of the genome (Moldovan, et al., 2007; Paunesku, et al., 2001). To perform its biological functions in DNA replication/ repair, three PCNA monomers assemble in a non-covalent way into a ring that encircles the DNA template, which allows it to "clamp" DNA polymerases δ and ε to the template strand (O'Donnell and Kuriyan, 2006). It has been reported that chemical crosslinking agents can covalently crosslink the subunits of the PCNA trimer in vitro so that a high molecular weight form of PCNA can be detected on SDS gels (Wenz et al.,

1998). Crosslinking studies also argued that there might exist PCNA double homotrimers in the cell (Naryzhny, et al., 2005; Naryzhny, et al., 2006). PCNA was crosslinked into a 28

covalent PCNA trimer following UVA irradiation in cells which have incorporated 6-

thioguanine into their genome (Montaner, et al., 2007). By sequence comparison and

mutant analysis, they were able to show that the covalent crosslinking was through a

histidine residue under acute oxidative stress. They proposed that singlet oxygen was the

agent that caused PCNA crosslinking, however, no test was done to prove the role of

singlet oxygen in this regard.

Lamin B protein belongs to the lamin family, which plays an important role in

maintaining the normal nuclear structure (Goldman, et al., 2002; Gruenbaum, et al.,

2005). SV40 large T-antigen serves as the helicase for the SV40 virus replication fork,

like MCM helicase for the mammalian DNA replication (Sclafani, et al., 2004). To

perform its helicase function, six SV40 large T antigens form a circular non-covalently

bound hexamer, which then binds to the DNA strand (Fanning and Knippers, 1992).

Topoisomerse I is the enzyme that relaxes supercoiled DNA template at the initiation step

of DNA replication and transcription (Champoux, 2001).

We also studied the behavior of other nuclear proteins including retinoblastoma

protein (Rb), Claspin, and others in response to photodynamic damage, but could not detect damage by Western blotting. Only a few cellular proteins were altered by photodynamic damage as detected by 2-dimensional gel electrophoresis (Bae, et al., 2008;

Grebeňová, et al., 2000). Here we propose that PCNA, p53, SV40 large T-antigen and topoisomerase I and lamin B are sensitive protein targets of photodynamic damage in mammalian cells.

2.2 Material and Methods

Cell lines

African green monkey kidney fibroblasts (CV-1) were obtained from the American

Type Culture Collection (ATCC, Manassas, VA) and cultured in MEM medium

(Invitrogen, Carlsbad, CA) containing 10% calf serum (Invitrogen) and 14 mM Hepes

(Sigma-Aldrich, St. Louis, MO). For SV40 infection, CV-1 cells were infected with

SV40 virus (strain 777, 2.3 x 107 pfu/mL) for 1 hr at 37 °C and cultured with complete

media for appropriate time periods (36-48 hr) before drug treatment. GM639 cells (SV40 transformed human fibroblasts) were obtained from the Coriell Institute (Camden, NJ)

and maintained in DMEM (Invitrogen) with 10% fetal bovine serum (FBS, Invitrogen).

MCF -7 (human breast cancer) cells were obtained from ATCC and cultured in DMEM

with 10% FBS. For experiments in D2O media, powdered MEM media was prepared in

99.9% D2O (Sigma-Aldrich) without serum. Mouse mammary carcinoma cells ts85, the

temperature sensitive ubiquitin conjugation mutant and its parental cells FM3A cells

were kindly provided by Dr. Altaf A Wani (Radiology Department, the Ohio State

University). The ts85 cells were cultured in DMEM medium with 10% FBS at the

permissive temperature, 32 °C (Pickart, 2004). Cells in DMEM were maintained in

water-jacketed incubators at 37 °C, in a humidified atmosphere with 5% CO2. Cells in

MEM/Hepes were maintained at 37 °C in water-jacketed incubators with humidified

room air. Experiments were done with cells in late log phase (80-90% confluence) to

ensure a high level of replicating cells.

Drugs

Proflavine (PF), acridine orange (AO), methylene blue (MB), 9-aminoacridine (9AA),

doxorubicin (DOX), hypericin (HY), ethidium bromide (EB), chloroquine (CH), and

ellipticine (EL) were from Sigma. Sanguinarine chloride (SAN), (S, R)-noscapine, 97%

(NOS), and berberine hydrochloride hydrate, 99% (BER) were from Aldrich (Milwaukee,

WI). Mono-L-aspartyl chlorin e6 tetrasodium salt (NPe6) was from Frontier Scientific

(Logan, UT). Camptothecin (CPT), m-amsacrine (m-AMSA, 4΄-(9-acridinylamino) methanesulfon-m-anisidide) (AMSA), and nitidine (NIT) were obtained from the

National Cancer Institute, Developmental Therapeutics Program (Frederick, MD).

Proflavine, acridine orange, methylene blue, ethidium bromide, and NPe6 were prepared as stock solutions in distilled water. 9-aminoacridine, camptothecin, doxorubicin, ellipticine, and m-AMSA were dissolved in DMSO and hypericin was dissolved in methyl alcohol. Stock solutions were aliquoted and stored frozen in the dark at -20 °C.

The E1 ubiquitin-activating enzyme inhibitor, PYR-41 was from BioGenova Corp.

(Frederick, MD). Other biochemicals from Sigma were phenylmethylsulfonyl fluoride

(PMSF), dithiothreitol (DTT), desferoxamine (DFO), superoxide dismutase (SOD), dimethyl sulfoxide (DMSO), glycerol (GL), mannitol (MAN), and ascorbic acid (ASC).

The protease inhibitors, aprotinin, leupeptin, and pepstatin were from USB (Cleveland,

OH). Leptomycin B (a gift from Dr. Yoshida, University of Tokyo, Japan) was prepared as a 10 μg/mL stock in 95% alcohol and stored frozen in the dark at -70 °C.

Photodynamic treatment

A specially constructed visible light irradiator was used for the photodynamic

experiments. An array of seven F8T5/D Sylvania Daylight fluorescent 8 watt 12” bulbs

(Osram Sylvania Products) assembled on top of the irradiator consistently delivers the irradiation at 0.45 J/cm2/minute. Irradiance was measured with a Li-Cor LI-185B

radiometer. To avoid the heating effect resulting from irradiation, a 12 watt cooling fan

was on during the irradiation, which controlled the temperature within 1.5 °C, as

recorded by an electronic probe. For the treatment, cells near confluence were loaded

with photosensitizers in the serum free medium for appropriate time periods to reach the

uptake equilibrium before irradiation. Before and after irradiation, cells harboring

photosensitizers were handled under dim red darkroom lights. In several experiments,

room light irradiation (Luxline Terra-Lux F32T8/735 3500K) was used. Radiant exposures are expressed as J cm-2. UV irradiation (254 nm) was measured with a

calibrated UVP J-225 meter (UVP, LLC, Upland, CA). In certain experiments, room

fluorescent lighting (recessed ceiling units: irradiance at the bench top: 1.3 W m-2) was also used as light source.

Western blot analysis

Whole cell protein extracts were prepared in SDS lysis buffer (62.5 mM Tris-HCl

( pH 6.8), 2% SDS, 10% glycerol, 1 mM phenylmethylsulfonylfluoride, 10 μg/mL aprotinin, 10 μg/mL leupeptin, and 10 μg/mL pepstatin ), separated by SDS PAGE

(polyacrylamide gel electrophoresis) and transferred to a nitrocellulose membrane

(Schleicher and Schuell, Keene, NH ) using a semi-dry transfer system (BioRad, 32

Hercules, CA). Membranes were blocked with 5% non-fat dry milk solution at room temperature for 1 hr, and then incubated with primary antibody overnight at 4 °C or 2 hr at room temperature. After rinsing with Tris buffered saline with Tween 20, membranes were incubated with horseradish peroxidase-conjugated secondary antibody for 1 hr at room temperature. Protein was detected by SuperSignal West Pico chemiluminescent substrate (Pierce, Rockford, IL) using either x-ray film or the BioRad ChemiDoc XRS imaging system. Primary antibodies used were monoclonal p53 Ab-6 (DO-1) (Lab Vision,

Fremont, CA), p53 Ab-1(PAb240) (kindly provided by Dr. Altaf A Wani), mouse PC10

(Santa Cruz Biotech, Santa Cruz, CA) directed against PCNA, mouse anti-SV40 large T antigen (Lab Vision, Fremont, CA), goat anti-lamin B (Santa Cruz), scleroderma patient serum anti-human topoisomerase I (Topogen), mouse anti-actin (Sigma), secondary antibodies were goat anti-mouse HRP conjugate (BioRad), and donkey anti-goat HRP conjugate (Santa Cruz Biotech). Precision Plus Protein Dual color standards (BioRad) and EZ-Run pre-stained Rec Protein ladder (Fisher Scientific, Pittsburgh, PA) were used as protein markers.

2.3 Results p53

Detection of p53 crosslinking following photodynamic damage

We first studied photo-damage to p53 using the SV40-transformed fibroblast cell line

GM639 employing the DNA intercalating dye proflavine as the photodynamic agent. As shown in Fig 2.1A, anti-p53 western blotting analysis revealed multiple anti-p53

antibody reactive species with three well-defined high molecular weight bands. We asked

whether or not this p53 crosslinking pattern was unique to GM639, so we did the same

study in SV40 virus infected CV-1 cells. As shown in Fig 2.1B, multiple high molecular

weight p53 bands were seen in SV40 infected CV-1 cells, and the bands at the higher molecular weight range from 100 kDa to 250 kDa matched those in GM639 cells. To further characterize this photodynamic crosslinking pattern, we compared it with that produced by glutaraldehyde, which is known to crosslink p53 monomers to produce high molecular weight p53 oligomers. As seen in Fig 2.1C, the bands resulting from p53 photodynamic crosslinking closely match the p53 dimer produced by glutaraldehyde crosslinking. It should be noted that the p53 dimer is not always resolved into a doublet,

as it is in the proflavine lane of Fig 2.1C (Reviewed in: Zhao, et al., 2009). Also, different

methods of protein-protein crosslinking can result in slight electrophoretic mobility shifts

for the crosslinked forms (Bae, et al., 2008; Montaner, et al., 2007).

p53 photodynamic crosslinking is dependent on the p53 level, not the functional type

As it is well known that p53 is frequently mutated with loss of function or altered function in cancerous cells, it could be argued that the p53 crosslinking pattern produced in SV40 transformed GM639 cells and SV40 infected CV-1 cells might be related to the abnormality either in the conformation or the function of p53 in these two cell lines. To clarify this issue, we tested the response of p53 in CV-1 cells which presumably express wild-type p53 (Zambetti, et al.,1992). When low amounts of protein lysates were loaded, the p53 monomer band was greatly reduced compared to the control, and there was no detectable higher molecular weight p53 bands (Fig 2.2A). However, a higher loading of 34

CV-1 lysates (Fig 2.2B) revealed the same pattern of p53 photodynamic crosslinking as seen in Fig 2.1A and B, which argues that the pattern of p53 photodynamic crosslinking is same for cells in which the function of p53 is normal or might have been mutanted, however, the low p53 level in wild type cells complicated the detection of high molecular weight p53 oligomers.

It has been reported that UV irradiation can effectively stabilize p53 by inhibiting the expression of MDM2, which functions as an ubiquin-E3 ligase targeting p53 for degradation in the cytoplasm (Haupt, et al., 1997; Kubbutat, et al., 1997). Leptomycin B

(LMB) has been reported to induce the accumulation of p53 in the nucleus by blocking its nuclear export (Freedman and Levine, 1998). We then subjected CV-1 cells to either

UV treatment or LMB treatment to stabilize p53, in an effort to examine the p53 photodynamic crosslinking profile more closely. As indicated in Fig 2.2C, at a UV dose of 20 J m-2, two bands between 100 kDa to 250 kDa were easily detected (other bands in

the pattern as seen above were also observed after longer film exposure, data not shown);

when cells were treated with LMB and photodynamic treatment, a well-defined crosslinking pattern (Fig 2.2C) was observed for p53, which was consistent with that in

Fig 2.1A and B. To determine whether other drugs, also known to be singlet oxygen generators, could produce this p53 crosslinking pattern, we included NPe6 (mono-L- aspartyl chlorin e6) and HY (hypericin) in our system. HY localizes at the endoplasmic reticulum while Npe6 localizes in the lysosome (Castano, et al., 2004). As indicated in

Fig 2.2D, HY can effectively produce p53 crosslinking, and the pattern is similar to that generated by PF, while NPe6 can not. These experiments suggest that the efficiency and

extent of p53 crosslinking vary among drugs, but the crosslinking pattern stays largely

the same.

The role of singlet oxygen in p53 crosslinking

Photodynamic drugs such as proflavine can produce other reactive oxygen species including superoxide, hydrogen peroxide and hydroxyl radicals in addition to singlet oxygen. To test whether or not singlet oxygen is the major species for producing p53 photodynamic crosslinking, we included a singlet oxygen enhancer and a quencher into our system. Heavy water (D2O) potentiates the biological effect of singlet oxygen by

significantly increasing its half-life (Ito, 1978). Cells treated with proflavine in D2O medium displayed a more pronounced p53 crosslinking pattern compared to that in regular medium, as seen in Fig 2.3. Histidine is a potent singlet oxygen quencher

(Matheson, et al., 1975). The p53 crosslinking was completely suppressed when His was added to the system (See Fig 2.3).

High molecular weight p53 bands are not due to polyubiquitination

The ts85 temperature-sensitive mouse mammary carcinoma cell line harbors a thermolabile mutant E1 ubiquitin activating enzyme. In our study, ts85 cells were subjected to proflavine-mediated photodynamic treatment at permissive temperature

(32 °C) and restrictive temperature (39 °C) (Pickart, 2004). Compared to the control at the permissive temperature, p53 was stabilized at the restrictive temperature (Fig 2.4), which confirmed that the temperature shift blocked the ubiquitin-proteosome mediated p53 degradation pathway in ts85 cells. When photodynamic treatment was applied to ts85 36

cells at the restrictive temperature, a similar p53 crosslinking profile as that in Fig 2.2B

was seen, proving that the high molecular weight p53 bands are not due to

polyubiquitination.

PCNA

PCNA photodynamic crosslinking caused by cytoplasmically localizing

photosensitizers

We previously reported that PCNA was covalently crosslinked as a trimer in

mammalian cells by many chemotherapeutic and chemopreventive drugs that act as PSs

with visible light (Bae, et al., 2008). There is no crosslinking of PCNA when CV-1 cells

were treated either with light only or compounds (in the dark) only, as shown in Fig 2.5.

Among all the compounds tested in this study, hypericin and NPe 6 are not localized

to the nucleus: the former is localized in the endoplasmic reticulum, while the latter is

localized to lysosomes (Castano, et al., 2004). However, both drugs caused efficient

PCNA crosslinking after irradiation. To rule out the possibility that this was caused by access to PCNA of these cytoplasmically locating compounds in a fraction of mitotic cells in our system, we did our assay in SV40 infected CV-1 cells whose DNA replication

machinery has been “hijacked” by SV40 virus (a multiplicity of infection as high as 20

plaque-forming units per cell was used to ensure that all the host CV-1 cells were

infected) and the cells are thus kept in S phase and can not enter mitosis (Gershey, 1979;

Lehman, et al., 2000; Okubo, et al., 2003). As shown in Fig 2.6, both hypericin and

NPe 6 were still able to cause efficient photodynamic crosslinking of PCNA in the SV40

system, which strongly argues that drugs which localize to the cytoplasm can also cause

photodynamic damage to nuclear proteins.

Tests for type I pathway ROS involved in photodynamic crosslinking of PCNA

As it might be argued that reactive oxygen species other than singlet oxygen are

involved in producing PCNA photodynamic crosslinking via the type I pathway, specific

ROS quenchers were used to test this possibility. As shown in Fig 2.7, PCNA

crosslinking by proflavine and light was not affected by hydroxyl radical scavengers

including mannitol, ethanol, DMSO, or ascorbate. In addition to quenching hydroxyl

radicals, ascorbate can also scavenge alkoxyl, thiyl, sulfenyl, and other radicals (Carr and

Frei, 1999). These results argued against PCNA crosslinking by hydrogen peroxide or

hydroxyl radicals. The inclusion of superoxide dismutase did not affect PCNA

crosslinking by proflavine and light, indicating that superoxide was not involved in

PCNA crosslinking. Desferoxamine, the iron chelator and hypoxia inducing agent, also

had no effect. The thiol reducing compound DTT (dithiothreitol) showed no effect on

PCNA photo-crosslinking, indicating that crosslinking was not due to disulfide bonds. It

also ruled out the possibility of His-His crosslinking, as this type of crosslinking is sensitive to DTT (Shen, et al., 1996). Taken together, the above results strongly argues against the involvement of type I pathway in the photodynamic crosslinking of PCNA.

Ubiquitination is not involved in the photodynamic crosslinking of PCNA

It is known that PCNA will undergo monoubiquitination when there are unrepaired lesions in the DNA template (Ulrich, 2006; Watts, 2006). To find out whether or not 38

PCNA is ubiquitintated in response to photodynamic damage, a temperature shift experiment was done with the E1-labile mutant ts85 cells and there was no change of

PCNA photodynamic crosslinking (Fig 2.8A).We also included a specific E1 ubiquitin- activating enzyme inhibitor PYR-41(4[4-(5-nitro-furan-2-ylmethylene)-3,5-dioxo- pyrazolidin-1-yl]-benzoic acid ethyl ester) in our system. PYR-41 is able to inhibit ubiquitin-activating enzyme (E1), which is the first enzyme for initiating the sequential action of ubiquitylation, thus blocking the entire process (Yang, et al., 2007). When this agent was added to our system, there was no alteration of the PCNA crosslinking pattern, as shown in Fig 2.8C. Another piece of evidence came from a glutaraldehyde crosslinking experiment. As shown in Fig 2.9B, an identical pattern of high molecular weight PCNA profile was produced by both glutaraldehyde and photodynamic treatment, suggesting that these bands represent PCNA chemically crosslinked to PCNA, or PCNA binding to heterogenous proteins, not ubiquitinated PCNA.

Other proteins

Band depletion reaction of topoisomerase I in response to photodynamic damage

Proflavine and visible light treatment leads to the depletion of the topoisomerase I band as detected by Western blotting in both CV-1 cells and SV40 infected CV-1 cells

(Fig 2.10). Later, our laboratory found that topoisomerase I covalently crosslinked to

DNA under this treatment (Ragu Kanagasabai, unpublished data). The crosslinking renders topoisomerse I unable to enter the stacking gel of SDS PAGE, thus resulting in the phenomenon of “band depletion” in anti-topoisomerase I Western blots. Actin was

used as a loading control, and it showed that actin was not affected by photodynamic

damage under the same treatment.

Photodynamic crosslinking of SV40 large T antigen and lamin B

A high molecular weight form of SV40 large T antigen was detected by Western

blotting when proflavine-treated SV40 transformed human cells were exposed to

laboratory room lighting (0.039 J cm-2) (Fig 2.11A left). When a radiant dose of

4.5 J cm-2 was used, there was a marked increase of the high molecular weight species and a simultaneous reduction of the SV40 large T-antigen monomer band. At an intermediate irradiation level of 3.4 J cm-2 , a faint band with the molecular weight of the

SV40 large T-antigen dimer was resolved in SV40 infected CV-1 cells which had been

pre-loaded with proflavine (Fig 2.11A, right). Glutaraldehyde can covalently crosslink

large T antigen to hexamers, causing them to migrate at a position corresponding to 500 kDa on SDS PAGE. Glutaraldehyde treatment of GM639 cell lysate produced a high molecular weight large T antigen band corresponding to that produced by proflavine and light (Fig 2.11B). Hypericin also caused substantial covalent crosslinking of large T antigen, though NPe6 did not (Fig 2.11C).

Lamin B monomer was efficiently crosslinked to oligomers of different orders when

MCF-7 cells were exposed to proflavine and light (Fig 2.12A). Hypericin also caused strong lamin B photo-crosslinking, as reported by others (Lavie, et al., 1999), while the efficiency of NPe6 is much lower (Fig 2.12B).

2.4 Discussion

Only few proteins (7 out of a total of ~ 300 protein spots resolved) are affected

(change of band intensity) by the treatment of proflavine and light by 2D gel

electrophoresis (Bae, et al., 2008). In another study of 5-aminolevulinic acid mediated photodamage to cellular proteins by two-dimensional gel analysis, 24 protein spots (7 increased and 17 decreased ) were affected among ~1350 spots resolved (Grebeňová, et al., 2000 ). Both results indicate that only a limited number of proteins were targeted for

damage by singlet oxygen, which might be related to such factors as the amino acid

composition of the proteins, the generation site of singlet oxygen. To get a better

understanding of photodynamic damage to DNA replication, we focused our search of

protein targets of photodynamic damage on important proteins for DNA replication. In

the present study, we have shown that p53 is a target of photodynamic damage, which is

of particular significance. Mutations in p53 are associated with poor response to cancer

therapy, driving more searches for drugs that are able to stabilize or re-activate p53

(Bykov and Wiman, 2003). p53 has also been studied for its role in the cytotoxicity of

photodynamic damage (Zawacka-Pankau, et al.,2008). It has been suggested that a

photodynamic drug protoporphyrin IX may enhance photodynamic therapy by binding

and stabilizing p53. However, the possibility of the involvement of photodynamic

damage to p53 has not been discussed (Zawacka-Pankau, et al., 2008). Most p53

mutations, which in many cases are caused by single nucleotide alterations, either

activate p53 as an oncogene or disrupt its tumor suppressor function (Whibley, et al.,

2009). Covalent p53 protein damage may have similar effects. The photodynamic damage to proteins manifest themselves in multiple forms. Besides protein-protein 41 crosslinking, other forms of protein damage, such as covalent modification of amino acid side chains, might also occur under the same conditions. Such protein damage has the potential to affect the interaction of p53 with proteins involved in its post-translational modification, degradation and transport, as well as its many interactions with other proteins and promoter regions of p53 regulated genes. Damage to oncogenic p53 mutant proteins might be beneficial, while damage to wild-type p53 might work against cytotoxicity. p53 is crucial to DNA replication, genomic stability, and the cellular response to DNA damage. High level damage to proteins important for cellular repair and survival might contribute to the cytotoxicity of photodynamic therapy. On the other hand, low level damage to these proteins might leave enough functionality to support cellular survival, but with reduced damage signaling and repair. This could contribute to tissue damage and photo-carcinogenesis. Others have discussed the implications of oxidative damage to proteins of DNA replication, DNA damage signaling, cell cycle checkpoint control and DNA repair, including damage to p53 by reactive oxygen species (Halliwell and Whiteman, 2004; Hussain, et al., 2003). As a highly reactive oxygen species, the half-life of singlet oxygen in the biological system is less than 40 ns and the sphere of action is within 20 nm, therefore, molecules in close proximity to the generation point of singlet oxygen (the location of the photosensitizer) are most easily attacked. By applying photodynamic treatment in different cells with differing p53 status and level, we were able to show that the threshold for detecting the p53 crosslinking pattern resulting from photodynamic damage is determined by the amount, not the functional type of p53. We have shown that p53 crosslinking is significantly enhanced by the D2O effect and completely quenched by histidine, which lends strong support that singlet oxygen is the 42

ROS responsible for p53 photocrosslinking. Furthermore, we have shown that the high

molecular weight p53 bands caused by photodynamic damage are not due to

polyubiquitination.

PCNA was crosslinked to high molecular weight forms in response to photodynamic

damage. The fact that a chemical agent like glutaraldehyde produces an identical PCNA

pattern strongly supports the idea that PCNA photodynamic crosslinking forms are

composed of PCNA oligomers or PCNA binding to heterogeneous proteins. The addition

of specific type I ROS quenchers had no effect on alternation of PCNA damage produced

by PSs of interest plus visible light, thus no type I ROS is involved in causing the photo-

induced damage to PCNA.It is a noteworthy finding that the cytoplasmically localizing

drugs, hypericin and NPe6 can produce significant photodynamic damage to proteins

localized in the nucleus in mammalian cells and SV40 infected mammalian cells which

have been locked into the DNA synthesis stage. It is generally accepted that

photodynamic damage occurs around the binding site of PSs. Since the localization of

PSs might not be absolute site-specific, it is conceivable that traces of these cytoplasmically localizing drugs might be in the nucleus. When irradiated, the singlet oxygen generated by this nuclear fraction of drugs will have access to nuclear PCNA, causing efficient PCNA oligomerization.

Like p53 and PCNA, SV40 large T-antigen and lamin B showed similar tendencies to be crosslinked to oligomers by photodynamic damage, while topoisomerase I was found to crosslink to DNA under the same treatment. Considering the fact that p53, PCNA,

SV40 large T-antigen and lamin B are known to perform their biological functions in the

form of oligomers while topoisomerase I functions as a monomer, we may explain the 43

differences in their response to the same photodynamic treatment from the perspective of

the difference in their functional forms. It suggests that singlet oxygen generated from

photodynamic drugs tends to crosslink proteins existing in the form of non-covalent

oligomers into covalent oligomers, thus depriving them of their normal biological

function. SV40 large T antigen and PCNA function as circular oligomers and were easily

photocrosslinked by short exposures to low level irradiation from room lighting.

Photocrosslinking of non-circular oligomers (p53 and lamin B) was detected only with much higher levels of photodynamic damage, as produced by the specially constructed irradiator. This suggests that the circular oligomers are exceptionally sensitive to photocrosslinking and may be the best biomarkers for photodynamic damage in the nucleus.

Topoisomerse I is important for both DNA replication and transcription as both processes require unwinding DNA double helix (Champoux, 2001). This unwinding requires a helix and a “swivel function”, and the latter is provided by DNA single strand nicking-closing activity of topoisomerse I. Topoisomerse I is the target of a group of

important anticancer drugs named as topoisomerase I poisons which stabilize the single

strand DNA passing intermediate of topoisomerse I resulting in a single strand DNA

break at the site of covalent topoisomerse I-DNA attachment (Liu, et al., 2000). Lethal

double strand DNA breaks might form when advancing DNA replication forks collide

with topoisomerse I poison-topoisomerse I-DNA complex (Rothenberg, 1997).

Topoisomerse I is targeted by photodynamic damage by crosslinking to DNA (data not

shown). It is possible that photodynamic damage to DNA facilitates crosslinking of

topoisomease I to DNA, or that topoisomerse I itself is photodynamically damaged 44 during DNA strand passing step, which may render it unable to complete its activity to reverse the topoisomerse I-DNA crosslink, thus stabilizing topoisomerse I-DNA attachment, similar to the effect of topoisomerse I poisoning.

We report here that multiple important nuclear proteins including p53, SV40 large T- antigen are targets for photodynamic damage. It has been reported that Bcl-2, STAT3 are targeted by photodynamic damage (Henderson, et al., 2007; Kim, et l., 1999; Xue, et al.,

2001). This work extended our previous study and provided new information on the protein targets in the DNA repair and replication system.

Figure 2.1 Photodynamic crosslinking and chemical crosslinking of p53 protein

The whole cell lysates of each sample were separated by SDS gel electrophoresis (7.5%

acrylamide). Western blotting was performed with anti-p53 Ab-6 (DO-1) antibody.

(A) SV40-transformed human fibroblasts (GM639) were exposed to PF (40 μM, 30 min,

37 °C) plus light (3.4 J cm-2) treatment. Cells treated with irradiation alone were used as a

light control. p53 dimer, trimer and tetramer are marked. PF: proflavine, Lt: 3.4 J cm-2.

(B) CV-1 monkey kidney cells were infected or mock-infected by SV40 virus and exposed to PF (40 μM, 30 min, 37 °C) plus light (3.4 J cm-2) treatment at the time point

of 48 hr post-infection. PF: proflavine, Lt: 3.4 J cm-2, Ctrl: a control which is infected,

but not treated with PF or irradiation. p53 dimer, trimer and tetramer are marked. (C)

Glutaraldehyde (GA) crosslinking of p53 (GA was added to the RIPA cellular extract at a

final percentage of 0.0125% and incubated for 10 min at 37 °C) compared to p53

crosslinking by PF (40 μM) plus light (4.5 J cm-2) in CV-1 cells. p53 dimer is marked.

PF: proflavine, Lt: 4.5 J cm-2, GA: glutaraldehyde.

A Lt PF/Lt B PF/Lt Ctrl 250- Tetramer Tetramer -250 150- Trimer Trimer -150

Dimer Dimer -100 100-

75- -75

50- p53 -50 p53

GM639 Cell SV40 infected CV-1 Cell

C GA PF/Lt

250-

150- Dimer 100-

75-

p53 50-

CV-1 Cell

Fig 2.1 Photodynamic crosslinking and chemical crosslinking of p53 protein

Figure 2.2 The effect of p53 accumulation on detection of p53 photodynamic

crosslinking

Cells were lysed in SDS lysis buffer after the treatments, and the whole cell lysates

were separated by SDS gel electrophoresis (7.5% acrylamide). Western blotting was

performed with anti-p53 Ab-6 (DO-1) antibody. (A) CV-1 cells were exposed to PF (40

μM, 30 min, 37 °C) plus light (3.15 J cm-2) treatment. A low amount of protein lysates

(12 µg) was loaded for each well. (B) CV-1 cell was exposed to PF (40 μM, 30 min,

37 °C) plus light (3.15 J cm-2) treatment. A high amount of protein lysates (80 µg) was

loaded for each well. High MWT refers to high molecular weight species. p53 dimer,

trimer and tetramer as well as High MWT are marked. (C) CV-1 cells were pre-incubated

with 5ng/mL leptomycin B (LMB) in serum free medium at 37 °C for 20 hr or pre-treated

with UV irradiation (20 J cm-2) and incubated for 20 hr at 37 °C, followed by PF (40 μM,

30 min, 37 °C) plus light (3.15 J cm-2) treatment without removing leptomycin B from the medium. p53 dimer and trimer are marked. (D) CV-1 cells were treated with 5 μM proflavine (PF), NPe6 or hypericin (HY) for 1 hr in serum-free medium at 37 °C before irradiation (3.15 J cm-2). p53 dimer and trimer are marked. Asterisks represent

nonspecific bands.

PF/Lt

Ctrl PF/Lt LMB LMB UV UV A C Trimer -150 p53 Dimer -100

CV-1 Cell -75

-50 p53

CV-1 Cell

Ctrl PF/Lt B D High MWT 250- PF NPe6 NPe6 Ctrl HY Tetramer 150- Trimer Trimer -150 Dimer

100- Dimer -100

75-

-75

50- p53 p53 -50

CV-1 Cell

Figure 2.2 The effect of p53 accumulation on detection of p53 photodynamic crosslinking

Figure 2.3 Tests of the role of singlet oxygen in p53 photodynamic crosslinking

GM639 cells were pre-treated with 100 mM L-histidine for 1.5 hr, followed by PF

-2 (40 μM, 30 min, 37 °C) plus light (3.15 J cm ) treatment, or treated either in D2O alone or in D2O medium made up in 99.9% D2O containing PF (40 μM) for 30 min at 37 °C

before irradiation (3.15 J cm-2), or treated with PF (40 μM, 30 min, 37 °C) before

irradiation (3.15 J cm-2) in the regular medium. Cells were lysed in SDS lysis buffer after

the treatments, and the whole cell lysate was separated by SDS gel electrophoresis (7.5%

acrylamide). Western blotting was performed with anti-p53 Ab-6 (DO-1) antibody.

PF/Lt

Ctrl D2O D2O His

250- Tetramer Trimer 150-

100- Dimer

75-

50- p53

GM 639 Cell

Figure 2.3 Tests of the role of singlet oxygen in p53 photodynamic crosslinking

Figure 2.4 Ubiquitination is not involved in p53 photodynamic crosslinking

ts85 cells were shifted from 32 oC to 39 oC for 16 hr, loaded with or without PF (40

μM) and maintained at 39 °C for 0.5 hr, followed by irradiation (3.15 J cm-2) without removing drug from the medium. 160 µg of the whole cell lysates were separated by SDS gel electrophoresis (7.5% acrylamide). Western blotting was performed with anti-p53

Ab-1 (PAb240) antibody.

Temp ( oC ) 39 32 39 32 PF/Lt + + - -

-250

-150

Dimer -100 -75

p53 -50

ts85 Cell

Figure 2.4 Ubiquitination is not involved in p53 photodynamic crosslinking

Fig 2.5 Tests for light-dependent covalent crosslinking of PCNA trimers

CV-1 cells were treated in the dark with the drugs at concentrations as indicated. The cellular proteins were separated by SDS PAGE and detected by Western blotting with anti-PCNA antibody PC10.

Fig 2.5 Tests for light-dependent covalent crosslinking of PCNA trimers

Fig 2.6 Proflavine, hypericin and NPe6 photo-crosslinking of PCNA trimers in

CV-1 cells lytically infected with simian virus 40

CV-1 cells were infected with SV40 (20 plaque-forming units per cell). At the peak of viral DNA replication (36 hr post-infection), the cells were treated with the drugs as indicated (10 μM) and exposed to light in the irradiator (Lt, 7 min, 3.15 J cm−2) or were kept in the dark for 7 min (Dk). Cellular proteins were analyzed by SDS PAGE and

Western blotted with anti-PCNA antibody PC10.

Fig 2.6 Proflavine, hypericin and NPe6 photo-crosslinking of PCNA trimers in CV-1 cells lytically infected with simian virus 40

Fig 2.7 Test for type I pathway and other ROS

(A) CV-1 cells were treated with 40 μM proflavine and light (3.15 J cm−2) as a control (Ctrl). Other samples were identically treated but included 300 U/mL superoxide dismutase added immediately before treatment with PF and light (SOD) or immediately after treatment (SOD*), 0.1 mM ascorbate (Asc) added for 1 hr (37 °C) before treatment, and 50 mM mannitol (Man) added 1 hr before treatment. (B) Proflavine control and duplicates of the SOD, SOD*, ascorbate and mannitol experiments in A were included with additional antioxidant tests: 5 mM dithiothreitol (DTT) added 15 min before

treatment, 0.3 mM desferoxamine (DFO) added 6 hr before treatment, 4% dimethyl

sulfoxide (DMSO) added 30 min before treatment, and 10 mM ethanol (ETOH) added

30 min before treatment.

Fig 2.7 Test for type I pathway and other ROS

Fig 2.8 Ubiquination is not involved in photodynamic PCNA crosslinking

(A) ts85 cells, capable of ubiquitination at 32 °C but not at 39 °C, were incubated at

each temperature for 16 hr, then treated with 40 μM proflavine (PF) and light

(3.15 J cm−2) and analysed by PCNA Western blotting. (B) The parental cells of ts85

cells, FM3A cells, were subject to the same treatment as in (A). (C) CV-1 cells treated with the E1 ubiquitin-activating enzyme inhibitor, PYR-41 (25 μM), proflavine (40 μM)

and light (3.15 J cm−2), were analyzed by PCNA Western blotting.

A B FM3A

CL PFL PFL CL 39°C 32°C 39°C 39°C

Fig 2.8 Ubiquination is not involved in photodynamic PCNA crosslinking

Fig 2.9 High molecular weight PCNA forms produced by glutaraldehyde

(A) A RIPA lysate of MCF-7 cells was treated with glutaraldehyde (GA, 0.025%, 10 min, 37 °C), glycine was added to 0.15 M to stop protein crosslinking, and the sample was Western blotted (anti-PCNA). Minor high molecular weight PCNA bands are

indicated (minor bands) with 93 and 154 kDa bands. (B) The RIPA lysates of CV-1 cells

and GM639 cells, respectively, were treated with glutaraldehyde as in “A” and the

samples were labeled as “GA” in the figure. CV-1 cells and GM639 cells, respectively,

were treated with proflavine (PF, 40 μM) and exposed to light in the irradiator (Lt, 7 min,

3.15 J cm−2).

A B CV-1 cell GM639 cell PF/LT GA PF/LT GA

Fig 2.9 High molecular weight PCNA forms produced by glutaraldehyde

Fig 2.10 Band depletion of topoisomerase I on SDS PAGE in response to

photodynamic damage

CV-1 cells and SV40 infected CV-1 cells (40 hr post SV40 virus infection),

respectively, were treated with proflavine (PF, 40 μM) and exposed to visible light

irradiation at 1.70 J cm−2. The whole cell lysates extracted by SDS lysis buffer were

separated by 7.5% SDS PAGE and then Western blotted by anti-human topoisomerase I antibody. Actin was used as a loading control.

CV-1 cell SV40 infected CV-1 cell

1.70 J cm-2 + + + + PF (40 µM) - + - +

Topoisomerase I

Actin

Fig 2.10 Band depletion of topoisomerase I on SDS PAGE in response to

photodynamic damage

Fig 2.11 Photodynamic crosslinking of SV40 large T antigen

(A) SV40-transformed human fibroblasts (GM639) were briefly exposed to

laboratory room lighting (0.039 J cm-2) and also exposed to more intense light in the

irradiator (4.5 J cm-2). SV40-infected CV-1 cells, with or without proflavine (40 µM, 30

min, 37 °C), were also exposed to light in the irradiator as indicated. SDS PAGE and

Western blotting with anti-SV40 large T antigen (LT) were performed.

(B) Glutaraldehyde (GA) crosslinking of large T antigen in a RIPA extract, compared to

-2 large T antigen crosslinked by proflavine (40 µM) plus light (PFL, 0.234 J cm ) in

GM639 cells. Large T antigen monomer was run off the 6% acrylamide SDS gel to

achieve higher resolution of the high molecular weight crosslink forms before Western

blotting with anti-large T antigen antibody. (C) GM639 cells were treated with 5 µM

proflavine, NPe6, or Hypericin for 1 hr at 37 °C, and then were irradiated (3.15 J cm-2).

Anti-SV40 large T antigen Western blotting was done. Controls (C) were GM639 cells without drug or light exposure.

Fig 2.11 Photodynamic crosslinking of SV40 large T-antigen

Fig 2.12 Photodynamic crosslinking of lamin B

(A) MCF-7 cells were treated with proflavine (PF, 30 min, 37 °C) and irradiation (3.4

J cm-2 ). Western blotting was done with anti-lamin B antibody. Controls (C) were MCF-

7 cells without drug or light exposure. (B) GM639 cells were treated with hypericin (1 hr) or NPe6 (30 min) at 37 °C and the indicated concentrations before irradiation

(3.4 J cm-2) and then Western blotted with anti-lamin B antibody. Controls (C) were

GM639 cells without drug or light exposure.

A PF C B 40µM

Fig 2.12 Photodynamic crosslinking of lamin B

CHAPTER 3

A NOVAL SINGLET OXYGEN PRODUCER:

[RU(TPY)(PYDPPN)]2+

3.1 Introduction

Transition metal complexes of ruthenium have emerged as promising candidates for clinical use in the field of cancer photodynamic therapy, as they are able to photosensitize oxygen production in a highly efficient way, absorb relatively strongly in the UV-VIS spectrum and intercalate into DNA (Ang and Dyson, 2006; Clarke, 2003; DeRosa and

Crutchley, 2002; Schmitt, et al., 2008). In the Ru (II) complexes family, pydppn ligands

2+ possessing [Ru(tpy)n(pydppn)2-n] (n = 0, 1) (tpy = [2,2'; 6',2'']-terpyridine,

pydppn = 3-(pyrid-2'-yl)-4,5,9,16-tetraaza-dibenzo [a,c] naphthacene) complexes have been reported to be the most efficient singlet oxygen producing photosensitizers available

1 with their near 100% efficiency in generating O2 in vitro (Liu, et al., 2009). Their quantum yields of singlet oxygen even surpass that of Photofrin, the most widely used

2+ 2+ photosensitizers in clinical trials to date. Both [Ru(tpy)(pydppn)] and [Ru(pydppn)2] bind and photocleave DNA upon irradiation in vitro, while the former is about 10 fold

70 more efficient in binding to DNA than the latter (Liu, et al., 2009). However, their biological activities remain largely unknown.

Since we have identified p53 and PCNA as sensitive protein targets of photodynamic damage in the cell culture system, we used these two proteins as biomarkers to test the potency of [Ru(tpy)(pydppn)]2+ as a photosensitizer. The results showed that

[Ru(tpy)(pydppn)]2+ is highly efficient in damaging nuclear protein targets. Furthermore, we found that [Ru(tpy)(pydppn)]2+ is able to cause extensive protein-DNA crosslinking and biphasic DNA replication inhibition.

3.2 Materials and Methods

Cell lines

African green monkey kidney fibroblasts (CV-1) were obtained from ATCC and cultured in MEM medium (Invitrogen) containing 10% calf serum and 14 mM Hepes.

Human fibroblasts (GM639) were from the Human Genetic Mutant Cell Repository, and were cultured in DMEM (Invitrogen) containing 10% fetal bovine serum.

Compounds

Proflavine (PF) was purchased from Sigma. Stock solution (40 mM) was aliquoted and stored frozen in the dark at -20 °C. [Ru(tpy)(pydppn)]2+ (14.04 mM),

2+ 2+ [Ru(pydppn)2] (19.1 mM) and [Ru(tpy)2] (9.87 mM) stock solutions made in distilled water were kindly provided by Dr. Turro (Department of Chemistry, the Ohio State

University) and kept in the dark at room temperature. Ligands tpy and pydppn stock

solution (10 mM) in DMSO were also provided by Dr. Turro. Stock solutions were

aliquoted and stored in the dark at room temperature.

Photodynamic treatment

Irradiation was done with a bank of seven F8T5/D Sylvania Daylight fluorescent 8

watt 12 inch bulbs (Osram Sylvania Products) as described with slight modification (Bae, et al., 2008). For cell experiments, cells in 35 mm culture plates (Corning Inc, Corning,

NY) were irradiated at 3.15 J cm-2 with plate lids in place. For RIPA lysate experiments,

aliquoted RIPA lysates in opaque black 96 well plate (Corning) were irradiated at

3.15 J cm-2 without plate lids; or the RIPA lysates were irradiated in inverted caps

removed from 1.5 mL micro tubes (USA Scientific Inc, Ocala, FL). In several

experiments a GG475 filter (Andover Corporation, Salem, NH) with a 475 nm cut-off

was placed either on top of the 96 well plate or the 35 mm culture plate. Before and after irradiation, samples containing photodynamic drugs were handled under dim red darkroom light.

Western blot analysis

Whole cell protein extracts, RIPA lysis of cells, SDS PAGE, and Western blotting were done as described (Bae, et al., 2008). In certain experiments, the SDS lysis was carried out 15 min after completion of irradiation. For RIPA lysates, confluent cells in

100 mm culture plates were extracted with RIPA lysis buffer, aliquoted either to opaque black 96 well plates or caps of 1.5 mL microcentrifuge tubes. Compounds or ligands

were then added in the dark. After irradiation, the contents in plate wells or caps were

transferred to 1.5 mL amber tubes (United Scientific Products, San Leandro, CA).

Quantitation of PCNA covalent trimerization

Covalent crosslinking of the subunits of the PCNA trimer was quantitated as

described (Bae, et al., 2008) using a BioRad ChemiDocTM XRS imaging system and

QuantityOne software (Bio-Rad). The quantification data was plotted with SigmaPlot

software.

GF/C filter based cellular protein-DNA crosslinking assay

The GF/C glass fiber filter assay is based on the selective binding of proteins to glass

fiber filters in 0.4 M GuHCl (Shin, et al., 1990). The cellular GF/C assay was used to

measure the extent of protein-DNA crosslinking. Cells in 35mm culture plates (Corning) were labeled with 3H-thymidine at a low level (1.0 µCi/mL) for 24 hr, then the labeling

medium was replaced with 5% calf serum containing MEM/Hepes medium loaded with

2+ ([Ru(tpy) n(pydppn)2-n] (n = 0, 1) complexes for another 24 hr. The medium was then

replaced with serum free medium (200-300 µL per 35 mm culture plates) and cells were

irradiated. Cells were lysed with 500 µL/plate Hirt lysis buffer (10 mM Tris buffer, pH

7.5, 10 mM EDTA, pH 7.5, 6% SDS) (Hirt, 1967). Lysates were then transferred to 1.5

mL microcentrifuge tubes containing a small (3 mm in diameter) stainless steel nut.

Tubes were securely capped and vigorously sheared with vortexing for 1 min, then heated

at 65°C for 15 min to ensure denaturation and dissociation of non-covalently attached

proteins from DNA. After cooling to room temperature, duplicate aliquots of the Hirt 73

extract were mixed with 0.4 M and 4 M GuHCL (guanidinium chloride), respectively,

then filtered through 0.4 M or 4 M GuHCL pre-wetted glass fiber filters. The 0.4 M

GuHCL is the protein binding condition. Only DNA covalently crosslinked proteins binds to the filters. The 4M GuHCL is the DNA binding condition. The radioactivity retained on the filter under this condition gives the value for the total labeled DNA in the aliquot. The ratio of the radioactivity retained on the GF/C filters in the 0.4 M GuHCL to

the total radioactivity retained on the filters in 4.0 M GuHCL thus provides the

percentage of labeled DNA that is crosslinked to protein. It has been reported that a

single crosslinked protein is sufficient to cause the retention of a DNA molecule on the

filter in 0.4M GuHCL (Shin, et al., 1990).

Inhibition of DNA Replication

The incorporation of 3H-Tdr into cellular DNA was used to determine the effect of photodynamic damage on DNA replication in CV-1 cells. Cells were seeded into 48 well tissue culture plates (10,000 cells per well per 0.5 mL growth medium) for 24 hr at 37 °C.

6 hr later, after replacing the medium with fresh growth medium, the cells were loaded

with [Ru(tpy)(pydppn)]2+ (prepared in MEM/Hepes containing 5% calf serum, 200 µL

per well) for another 24 hr at 37 °C. The cells were irradiated at 3.15 J cm-2, then labeled

with 5 μCi/mL 3H-Tdr (0.2 mL per well, prepared in growth medium) for 1 hr. Cells were rinsed with cold PBS, and then covered with 5% cold trichloroacetic acid (0.5 mL/well).

After 30 min on ice, the cells were rinsed once with 5% cold trichloroacetic acid, and

once with methanol. NaOH (0.2 mL, 0.3 M) was added to each well and the plates were

left overnight at room temperature. The solubilized material from each well was pipetted 74 to reduce viscosity, transferred to Mini PolyQ scintillation vials (Beckman Coulter) and assayed with 5 mL Ready Gel scintillation cocktail (Beckman Coulter) before counting

(Beckman LS6500 counter). Six wells were used for each point.

Chromosomal Crosslinking assay

CV-1 cells in 100 mm culture plates grown to confluence were treated with 40 µM

[Ru(tpy)(pydppn)]2+ in MEM/Hepes supplemented with 5% calf serum (3 mL per plate), then incubated at 37 °C for 24 hr. The drug medium was replaced with MEM/Hepes

(1 mL per plate) and irradiated (3.15 J cm-2) before lysis with 8 M GuHCL/1% sarkosyl

(700 µL per plate). The lysates from three plates were collected and transferred to a 50 mL sterile polypropylene centrifuge tube containing 3-4 stainless steel nuts (3 mm in diameter). The lysates were sheared with vortexing for 2 min and then heated at 65 °C for

15 min. The lysates were allowed to cool down to room temperature and then loaded on

CsCl step gradients (Rosenstein, et al., 1997). The CsCl step gradients were prepared by the successive layering of 2 mL volumes of solutions A through D (solution A, 1.75 g/mL CsCl; solution B, 1.63 g/mL; solution C, 1.42 g/mL; solution D, 1.32 g/mL) into an ultracentrifuge tube (14 × 89 mm, Beckman Coulter). Light mineral oil (Sigma) was used to fill the tube to the top. The tubes were centrifuged in an SW41 rotor (Beckman Coulter) at 30,000 rpm for 20 hr at 20 °C. After ultracentrifugation, the DNA from the lysates banded on the CsCl step gradient was eluted from the bottom of the tube. Free proteins were left at the top of the gradient, and proteins covalently attached to DNA were positioned with the DNA. The fractions containing protein-DNA crosslinks were pooled and dialyzed against distilled water to remove CsCl. The dialyzed sample was then 75

precipitated with 2 volumes of 100% ethanol and 2 M ammonium acetate. The

precipitated sample was solubilized in Omnicleave buffer, then treated with Omnicleave nuclease (Epicentre Technologies, Madison, WI) to remove DNA, followed by lyophilization to reduce the sample volume. The samples were subsequently subjected to

SDS PAGE and the gel was silver stained (Bio-Rad).

3.3 Results

2+ PCNA photodynamic crosslinking by [Ru(tpy)n(pydppn)2-n] (n = 0, 1) complexes in

CV-1 cells

2+ [Ru(tpy)n(pydppn)2-n] (n = 0, 1) complexes are highly efficient singlet oxygen

producers upon light irradiation, we first tested the response of PCNA to

2+ [Ru(tpy)n(pydppn)2-n] (n = 0, 1) in intact CV-1 cells. As the lowest energy excited state

2+ 3 in the [Ru(tpy)n(pydppn)2-n] (n = 0, 1) complex is ligand-centered (LC) with ππ*

localized on the pydppn ligand, not the tpy ligand. The photodynamically inactive

2+ homoleptic complex [Ru(tpy)2] was included as a negative control (see Scheme 3.1 for

the chemical structures of three Ru(II) complexes).When cells were treated with

[Ru(tpy)(pydppn)]2+, a high molecular weight anti-PCNA reactive species around

93 kDa was produced in a dose-dependent manner (Fig 3.1A). A very weak anti-PCNA

2+ reactive species around 93 kDa was produced by [Ru(pydppn)2] at the concentration as

2+ high as 80 µM in light (Fig 3.1B). [Ru(tpy)2] was completely inert under the same

experimental setting (Fig 3.1C). The molecular weight of covalently crosslinked PCNA

trimer has been reported to be around 93 kDa (Bae, et al., 2008).

As we only detected one form of crosslinked PCNA by [Ru(tpy)(pydppn)]2+ at concentrations as high as 20 μM and only a trace amount of crosslinked PCNA by

2+ [Ru(pydppn)2] , we asked whether this low level PCNA crosslinking by these two complexes was caused by their poor transport across cell membranes during the 1 hr

2+ incubation period, so we extended the cellular incubation time of [Ru(tpy)n(pydppn)2-n]

(n = 0, 1) complexes to as long as 24 hr. As we can see from Fig 3.2, [Ru(tpy)(pydppn)]2+

produced an identical PCNA photodynamic crosslinking pattern with that of proflavine.

Therefore, we are able to conclude that the band around 93 kDa is the PCNA trimer, and

the 154 kDa oligomer and minor bands probably represent a PCNA double trimer and

heterologous proteins that bind to PCNA. Compared to [Ru(tpy)(pydppn)]2+,

2+ [Ru(pydppn)2] caused markedly reduced PCNA photocrosslinking under the same

treatment, but the pattern remained the same. There was no photodynamic crosslinking of

2+ PCNA from [Ru(tpy)2] , the photodynamically inactive control.

2+ PCNA photodynamic crosslinking by [Ru(tpy)n(pydppn)2-n] (n = 0, 1) complex in

CV-1 RIPA lysates

To further test whether or not there is a cellular uptake difficulty for

2+ [Ru(tpy)n(pydppn)2-n] (n = 0, 1) complexes, we used the RIPA lysate of CV-1 cells as

our new system, since there is no cellular membrane barrier in a cell lysate. Multiple

PCNA crosslinked species were produced in nearly equal efficiency by both

2+ 2+ [Ru(tpy)(pydppn)] and [Ru(pydppn)2] complexes in the RIPA lysates of CV-1 cells

(Fig 3.3A). The extent of PCNA photocrosslinking generated from these two complexes also seems to be higher than that of proflavine, which is the reverse of the situation in 77 intact cells. What is more remarkable is that even a significant fraction of the PCNA monomer pool was depleted into the crosslinked forms, which has never been observed in our previous cell experiments. Irradiation of similar samples in sealed micro tubes placed horizontally resulted in almost complete loss of the monomer (Fig 3.3B). The 154 kDa

PCNA covalent oligomer was unusually pronounced in comparison to the 93 kDa PCNA trimer in the RIPA lysates. The 154 kDa form has been argued to be a PCNA double trimer based on chemical crosslinking experiments (Naryzhny, et al., 2006; Naryzhny, et al., 2005). The authors suggested that the double trimer may only exist in the non-DNA bound intracellular PCNA pool and carry out biological activities different from those of its DNA bound counterparts.

Quantitation of PCNA trimerization

The extent of PCNA photodynamic trimerization can be quantitated by analysis of anti-PCNA Western blots using electronic imaging equipment capable of integrating photons produced by ECL reagents (Bae, et al., 2008). When we changed only one of the three variables (cellular uptake time, [Ru(tpy)(pydppn)]2+ concentration, or radiant light dose), PCNA photocrosslinking was linear with the changed variable (Fig 3.4). However, the response of PCNA photocrosslinking by proflavine and acridine orange is a non- linear one, starting with a steep increase first with increasing dose, followed by a plateau

(Bae, et al., 2008). We reasoned that the absence of a plateau in the case of

[Ru(tpy)(pydppn)]2+ might come from its resistance to photodynamic self destruction, or because [Ru(tpy)(pydppn)]2+ is less efficient in exhausting the local concentration of oxygen (the source material of singlet oxygen) than proflavine and acridine orange. 78

Ligands alone caused only negligible crosslinking of PCNA in light

2+ The [Ru(tpy)n(pydppn)2-n] (n = 0, 1) complexes are very stable, however, it can be argued that the resulted PCNA crosslinking is a function of dissociated ligands (pydppn or tpy) but not the intact complexes, we further tested the photodynamic activity of individual ligands. First, we carried out a cell experiment in which CV-1 cells were exposed to either ligand at concentrations as high as 80 μM, with proflavine as a positive control. We also included control cells which have been subjected to the same treatment but lysed 15 minutes later, following completion of irradiation to allow more time for ligands to enter cells whose membranes might have been photo-damaged during the irradiation. As demonstrated in Fig 3.5A, there was no detectable PCNA crosslinking for either ligand, which ruled out the possibility of PCNA crosslinking by the activity of free

ligands in cells, at least at our detection limit. Next, based on the absorption differences between the intact complexes and ligands (Liu, et al., 2009), we applied a glass filter with an absorption cutoff at 475 nm to filter out the absorption spectrum of both ligands, while ruthenium complexes still maintain significant absorption. As shown in Fig 3.5B, there was no detection of PCNA crosslinking in CV-1 cell RIPA lysates treated with light filtered ligands (pydppn* or tpy*) . We can still detect efficient PCNA crosslinking resulting from both complexes under the same experimental condition and the crosslinking pattern after light filtering was same as that of proflavine (Fig 3.5C). This indicated that PCNA was damaged by singlet oxygen generated by

2+ [Ru(tpy)n(pydppn)2-n] (n = 0, 1) complexes as a whole, not the ligands. It should be

pointed out that the total PCNA crosslinking after light filtering was less than that before

filtering, which was expected as the former received a lesser amount of total light 79

irradiation. It is of note that the cut-off of the filter can not be 100% perfect, and the

absorption spectrum of the ligands tail off into the above 475 nm region. Therefore, it is

reasonable that there might be trace PCNA crosslinking resulting from free ligand

activity, but they must be at least orders of magnitude weaker than the intact complexes.

Taken together, we can exclude the interference of ligands in producing PCNA

crosslinking in light, at least to a significant extent.

Photodynamic crosslinking of p53 by ruthenium complexes

Treatment of GM639 cells with [Ru(tpy)(pydppn)]2+ or proflavine and light resulted

in covalent crosslinking of p53 to dimers, trimers and tetramers (Fig 3.6). The

2+ 2+ photodynamically inactive [Ru(tpy)2] and [Ru(pydppn)2] did not produce detectable p53 photocrosslinking in cells. In the GM639 RIPA lysates, there was no p53 covalent crosslinking in the absence of compounds, or treated with compounds in the dark (Fig 3.7

A). When lysates were exposed to light, [Ru(tpy)(pydppn)]2+ and

2+ [Ru(pydppn)2] complexes produced comparable p53 photocrosslinking while the

2+ inactive compound [Ru(tpy)2] remained inert (Fig 3.7A). The photodynamic crosslinking pattern of p53 matched those produced by glutaraldehyde (Fig 3.7A), which

further confirmed the identity of p53 oligomers. Azide, an efficient quencher of singlet

oxygen (Ito, 1978), greatly reduced or prevented the p53 photocrosslinking by

[Ru(tpy)(pydppn)]2+ and light (Fig.3.7B), again confirming the role of singlet oxygen in

p53 photocrosslinking.

Protein-DNA crosslinking caused by [Ru(tpy)(pydppn)]2+

Having identified PCNA and p53 as targets of singlet oxygen produced by

[Ru(tpy)(pydppn)]2+ , we then used GF/C filter based assay to find out if

[Ru(tpy)(pydppn)]2+ can cause photodynamic protein-DNA crosslinking. As shown in

Fig 3.8A and B, [Ru(tpy)(pydppn)]2+ caused marked photodynamic protein-DNA

crosslinking that increased with increasing doses of the compound or radiant light dose.

Both curves show increased slopes at higher levels of protein-DNA crosslinking, which might be caused by expedited DNA photodynamic damage that further facilitated protein-DNA crosslinking. We also tested the effect on protein-DNA crosslinking caused

2+ by two other ruthenium complexes. As shown in Fig 3.8C, [Ru(tpy)2] did not cause any

2+ protein-DNA crosslinking, and [Ru(pydppn)2] caused only marginal protein-DNA

crosslinking as compared to [Ru(tpy)(pydppn)]2+. To rule out the possibility that the

detected protein-DNA crosslinking is a false positive result of the binding of single- stranded DNA under protein binding conditions, we included proteinase K in our system.

The reversal of photodynamic protein-DNA crosslinking by proteinase K digestion, as

detected in Fig 3.8C, argued again the production of single strand DNA binding and supported the idea that it was protein-DNA crosslinking that was measured in these

assays.

Analysis of proteins crosslinked to DNA by [Ru(tpy)(pydppn)]2+

At 40 µM [Ru(tpy)(pydppn)]2+ and 3.15 J cm-2 of visible light irradiation, we

detected as high as ~40% protein-DNA crosslinking (Fig 3.8). The identification of those

proteins crosslinked to DNA may provide insight into the chromatin localization of the complex at the molecular level. Therefore, we employed a chromosomal crosslinking assay for a preliminary identification study. As shown in Fig 3.9, silver staining showed a dramatically different profile of proteins between [Ru(tpy)(pydppn)]2+ photodynamic

crosslinking and HCHO chemical crosslinking. A weak band with a molecular weight

similar to that of histones was detected for the [Ru(tpy)(pydppn)]2+ and light treated

sample, which was the first clue to the proteins crosslinked to DNA. The high level of

photodynamic protein-DNA crosslinking shown in Fig 3.8 suggested that either numerous protein bands with weak intensity or a few (or only one) protein bands with strong intensity might be detected by this chromosomal crosslinking assay. The detection of this single weak band is unexpected. However, although very sensitive, silver staining shows the most dramatic protein-to-protein variation among protein staining methods.

Some proteins can be intensively stained even in low amounts while others might not stain well or not stain at all with silver. In this study, a single protein might suffice to crosslink to DNA at a very high level, but might not stain well with silver. Alternately singlet oxygen might have damaged the specific amino acid residues or functional groups in the proteins that react with silver. Finally, this was a single pilot experiment that has not been repeated. Loss of sample during gel electrophoresis cannot be ruled out. All might lead to the detection of the only one weak band by our chromosomal crosslinking assay. Further work is needed in this direction.

Inhibition of DNA Replication

As we have detected nuclear protein damage and extensive protein-DNA crosslinking

caused by [Ru(tpy)(pydppn)]2+ in light, we were interested in probing the relationship

between [Ru(tpy)(pydppn)]2+ mediated photodynamic damage to DNA replication in

mammalian cells. Exposure of CV-1 cells to [Ru(tpy)(pydppn)]2+ at various

concentrations in light caused marked inhibition of DNA replication in a biphasic manner

as monitored by 3H- thymidine incorporation. The biphasic curve is the classic pattern

for inhibition of DNA replication by ionizing radiation and other DNA damaging agents

in mammalian cells (Makino and Okada, 1975; Painter and Young, 1975; Watanabe,

1974). It indicates that there are two components of DNA replication with different

sensitivity to DNA damage. The more sensitive component is the initiation of new

replicons, while the movement of the DNA replication forks is the component that is less

sensitive to DNA damage. Although different DNA damaging agents all cause biphasic

inhibition of DNA replication, the DNA damage caused can be different for each agent.

Ionizing radiation such as x-rays causes mainly single strand DNA breaks, topoisomerase

poisons damage DNA by forming “cleavable complexes”, while alkylating agents make

bulky DNA lesions. DNA damage inhibits both components of DNA replication

indirectly by way of trans acting factors that are components of cell cycle checkpoint

pathways (Miao, et al., 2003; Seiler, et al., 2007). The biphasic inhibition of DNA

replication seen in Fig 3.10 implied that checkpoint signaling pathways leading to DNA

replication arrest operate, even under extensive attack from [Ru(tpy)(pydppn)]2+ mediated photodynamic damage.

3.4 Discussion

In this work, we studied the photodynamic effect of a family of novel ruthenium

complexes. This class of ruthenium complexes is remarkable in that they have longer-

lived excited states and ~ 100% singlet oxygen production efficiency. Both complexes bind DNA through intercalation, however, the heteroleptic [Ru(tpy)(pydppn)]2+ can bind

2+ DNA with 10-fold higher efficiency than the homoleptic [Ru(pydppn)2] , which in turn

leads to a stronger DNA photocleavage ability upon irradiation (Liu, et al., 2009). In our

cell experiments, we also noticed dramatic differences between [Ru(tpy)(pydppn)]2+ and

2+ [Ru(pydppn)2] in producing PCNA and p53 oligomers, with the former being much

more efficient than the latter. We reasoned that the structural differences between these

two complexes might result in differences in their membrane permeability and thus

cellular uptake and accumulation, which ultimately lead to different levels of protein-

protein crosslinking. However, once the barrier of cell plasma membrane had been

disrupted, both complexes were able to produce protein crosslinking in a highly efficient

2+ 2+ way, which showed that both [Ru(tpy)(pydppn)] and [Ru(pydppn)2] are effective in

photo-damaging nuclear proteins like PCNA and p53. The azide quench results suggested that singlet oxygen plays a major role in producing protein crosslinking. Our optical filter

experiments lend strong support for the idea that the PCNA damage detected is from

2+ singlet oxygen generated from light-excited [Ru(tpy)n(pydppn)2-n] (n = 0, 1) complexes,

not individual tpy or pydppn ligands.

We have also shown that [Ru(tpy)(pydppn)]2+ is able to produce substantial protein-

DNA crosslinking in cells. In contrast to linear protein-protein crosslinking in the case of

PCNA, the protein-DNA crosslinking is non-linear, which suggests that protein-DNA

photocrosslinking becomes more efficient with increasing photodamage. Photodamage

from [Ru(tpy)(pydppn)]2+ caused pronounced cell cycle arrest, in a classic biphasic response that is known to be controlled by DNA damage and cell cycle checkpoint pathways, suggesting that these pathways were still operative after photodamage that

caused strong nuclear protein-protein crosslinking and protein-DNA crosslinking.

Acridine orange, a weaker photodynamic crosslinker of PCNA than proflavine under our

conditions has been used for photodynamic therapy of solid tumors with good results

(Kusuzaki, et al., 2007). Our finding of efficient nuclear protein damage, extensive

protein-DNA crosslinking, and strong inhibition of DNA replication suggests that

photodynamic ruthenium complexes may have potential for photodynamic therapy.

N N N N

N N N N N N N N N N Ru Ru Ru N N N N N N N N N N

2+ 2+ 2+ [Ru(tpy)2] (I) [Ru(tpy)(pydppn)] (II) [Ru(pydppn)2] (III)

Scheme 3.1 Tridentate ruthenium complexes

Fig 3.1 PCNA photodynamic crosslinking by the ruthenium complexes

(A) CV-1 cells (~90% confluence) were treated with [Ru(tpy)(pydppn)]2+

(37 °C, 1 hr), at the concentrations indicated, in serum-free medium (SFM), and then

were exposed to visible light irradiation (3.15 J cm-2). Western blotting was carried out

using anti-PCNA (PC10) antibody. PCNA monomer (PCNA) and trimer are indicated.

II: [Ru(tpy)(pydppn)]2+. (B) CV-1 cells (~90% confluence) were treated with

2+ 2+ [Ru(pydppn)2] (37 °C, 1 hr), other steps are same with A. III: [Ru(pydppn)2] .

2+ (C) CV-1 cells (~90% confluence) were treated with [Ru(tpy)2] (37 °C, 1 hr), other

2+ steps are same with A. I: [Ru(tpy)2] .

-2 A Lt (3.15 J cm ) - + - + + + + II (μM ) - - 80 10 20 40 80 93 kDa PCNA trimer

37 kDa PCNA

B -2 Lt (3.15 J cm ) - + - + + + + III (μM) - - 80 10 20 40 80 93 kDa PCNA trimer

37 kDa PCNA

C Lt (3.15 J cm-2 ) - + - + + + + I (μM) - - 80 10 20 40 80

37 kDa PCNA

Fig 3.1 PCNA photodynamic crosslinking by the ruthenium complexes

Fig 3.2 Photocrosslinking of cellular PCNA by the ruthenium complexes

CV-1 cells treated with the ruthenium complexes (40 μM, 24 hr) or proflavine

(40 μM, 30 min) were irradiated (3.15 J cm-2). After cell lysis with SDS, cellular proteins were separated by 10% SDS PAGE, and Western blotted for PCNA. The positions of the

PCNA monomer, the PCNA covalently crosslinked oligomers at 93 kDa and 154 kDa,

and minor high MWt PCNA bands are indicated. Ctr, control irradiated cells without

2+ 2+ 2+ drugs; I, [Ru(tpy)2] ; III, [Ru(pydppn)2] ; II: [Ru(tpy)(pydppn)] ; PF, proflavine.

Fig 3.2 Photocrosslinking of cellular PCNA by the ruthenium complexes

Fig 3.3 PCNA crosslinking in cell lysates

(A) The ruthenium complexes, proflavine, and glutaraldehyde were compared for

2+ PCNA covalent crosslinking in CV-1 cell RIPA lysates. [Ru(tpy)2] (I),

2+ 2+ [Ru(tpy)(pydppn)] (II), [Ru(pydppn)2] (III) and proflavine were added to the cell

lysates and then irradiated (3.15 J cm-2) before analysis by 10% SDS PAGE and PCNA

Western blotting. Another lysate was treated with 0.0125% glutaraldehyde to provide a

marker for the covalently crosslinked PCNA trimer (93 kDa band). (B) The same

2+ 2+ experiment was done with compounds [Ru(tpy)2] (I), [Ru(tpy)(pydppn)] (II),

2+ [Ru(pydppn)2] (III) with the cell lysates in sealed micro tubes placed on their sides during the irradiation to provide a greater surface area and less depth for the lysate during irradiation.

Fig 3.3 PCNA crosslinking in cell lysates

Fig 3.4 Linearity of PCNA photocrosslinking

(A) CV-1 cells (90% confluence, MEM/Hepes, 5% CS, 37°C) were treated with 40

μM [Ru(tpy)(pydppn)]2+ for the times indicated, rinsed with PBS, and exposed to light

(3.15 J cm-2) in serum-free medium (SFM). The cells were then lysed with SDS and

aliquots of the lysates were subjected to SDS PAGE and Western blotted for PCNA. The

93 kDa PCNA-positive band, representing the covalently crosslinked PCNA trimer was

quantitated using a Chemidoc XRS system. (B) CV-1 cells (80% confluence) were

treated with different concentrations of [Ru(tpy)(pydppn)]2+ for 24 hr (37°C,

MEM/Hepes, 5% calf serum), and were then irradiated (SFM, 3.15 J cm-2) and subjected

to SDS PAGE and PCNA Western blotting.The crosslinked PCNA trimer form was

quantitated as in “A”. (C) CV-1 cells (~80% confluence) were treated with 40 μM

[Ru(tpy)(pydppn)]2+ in MEM/Hepes, 5% CS, 37 °C for 24 hr. The medium was replaced

with SFM, and the cells were irradiated with different radiant light doses before being lysed for analysis of PCNA trimer crosslinking as in “A”.

hr 1 3 3 6 6 9 9 12 12 18 18 24 24 1

PCNA trimer

PCNA

µM 5 5 10 10 20 20 25 25 30 30 40 40 80 80

PCNA trimer

PCNA

Jcm-2 0.45 0.45 0.9 0.9 1.35 1.35 1.8 1.8 2.25 2.25 2.7 2.7 3.15 3.15

PCNA trimer

PCNA

Fig 3.4 Linearity of PCNA photocrosslinking

Fig 3.5 Ligands alone causes only negligible crosslinking of PCNA in light

(A) CV-1 cell were treated with PF or ligands (tpy and pydppn) at 80 μM in SFM at

37 °C for 1 hr, exposed to visible light irradiation (3.15 J cm-2), and lysed by SDS lysis

immediately or 15 min post irradiation. Control (CL) received the identical irradiation in

the absence of proflavine or ligands. D: Cells were treated with 80 μM ligands but kept in

the dark; L: cells were treated with 80 μM ligands, irradiated and immediately lysed. L15: cells were treated with 80 μM ligands, irradiated and lysed 15 min later. (B) Aliquots of

CV-1 cell RIPA lysates were treated with 80 μM ligands (tpy or pydppn) and irradiated

-2 (3.15 J cm ) with or without filter. “*”: filtered samples. Control (CL) received the

identical irradiation in the absence of ligands. (C) Aliquots of CV-1 cell RIPA lysates

2+ 2+ were treated with 80 μM [Ru(tpy)(pydppn)] (II), [Ru(pydppn)2] (III),

2+ -2 [Ru(tpy)2] (I) or proflavine (PF), and irradiated (3.15 J cm ) with or without filter.

“*”: filtered samples.

A tpy PF pydppn C L D L L15 D L L15

93 kDa

37 kDa PCNA

B pydppn* tpy* pydppn tpy CL* CL 93 kDa

37 kDa PCNA

C PF* III II I III* II * I* 250 kDa 150 kDa 100 kDa

75 kDa

50 kDa

37 kDa PCNA

Fig 3.5 Ligands alone causes only negligible crosslinking of PCNA in light

Fig 3.6 Covalent photocrosslinking of p53 subunits in GM639 cells

2+ Cells (~80% confluence) were treated with compounds I ([Ru(tpy)2] ),

2+ 2+ II ([Ru(tpy)(pydppn)] ) or III ([Ru(pydppn)2] ) (all 40 μM, 24 hr) or proflavine (40 μM,

30 min) in DMEM, 5% fetal bovine serum, 37 °C. After drug removal, the cells were

rinsed with phosphate buffered saline, covered with 300 μL DMEM, and irradiated

(3.15 J cm-2). Immediately after irradiation the cells were lysed with SDS lysis solution.

Cellular proteins were separated by 8% SDS PAGE and Western blotted for p53. Ctr,

2+ cells not treated with compounds; D, dark; L, light; PF, proflavine; I, [Ru(tpy)2] ;

2+ 2+ II, [Ru(tpy)(pydppn)] ; III, [Ru(pydppn)2] . Positions of the p53 monomer, dimer

doublet, trimer, tetramer and poorly resolved higher molecular weight p53 oligomers are

indicated.

Fig 3.6 Covalent photocrosslinking of p53 subunits in GM639 cells

Fig 3.7 Photodynamic crosslinking of p53 in cell lysates

(A) Drugs were added with mixing to 100 μL of GM639 RIPA lysate under dark conditions. The lysates were then either irradiated (3.15 J cm-2) or left in the dark.

Aliquots (5 μL, ~10 μg protein) were then separated by 8% SDS PAGE and Western

blotted for p53. Positions of the p53 monomer, dimer, trimer, tetramer and unresolved

high molecular weight p53 oligomers are indicated. An un-irradiated sample treated with

glutaraldehyde (GA, 0.0125%) was included to confirm the identity of the covalent p53

oligomers. Proflavine was also included. Lt, light irradiated (3.15 J cm-2); Ctr, control cell

2+ 2+ lysate irradiated without drugs; PF, proflavine; I, [Ru(tpy)2] ; II, [Ru(tpy)(pydppn)] ; III,

2+ [Ru(pydppn)2] ; GA, lysate treated with glutaraldehyde, 0.0125% final concentration.

(B) Effect of sodium azide on p53 photocrosslinking. GM639 cell RIPA lysates were irradiated (3.15 J cm-2) with [Ru(tpy)(pydppn)]2+ (Cpd II) (40 μM) and/or sodium azide as indicated.

Fig 3.7 Photodynamic crosslinking of p53 in cell lysates

100

Fig 3.8 Protein-DNA photocrosslinking

(A) CV-1 cells (~70% confluence) were grown in conditioned medium with 1 μCi/mL

3H-Tdr to radiolabel DNA. The labeling medium was then replaced with medium containing different concentrations of [Ru(tpy)(pydppn)]2+ for 24 hr (MEM/Hepes, 5%

CS, 37 °C). The medium was then removed; the cells were rinsed with PBS, and then covered with SFM before irradiation with visible light (3.15 J cm-2). Immediately after

irradiation, the cells were lysed with Hirt lysis solution. Aliquots of the lysate were

removed for quantitation of protein-DNA crosslinking. The points are averages of

triplicates, and the error bars are ± SD. (B) CV-1 cells (~80% confluence) were

radiolabeled and processed as in “A”, but all samples were treated with 40 μM

[Ru(tpy)(pydppn)]2+ for 24 hr. After irradiation to different radiant light doses, the

samples were analyzed for protein-DNA crosslinking as in “A”. (C) CV-1 cells, 90%

confluence, labeled with 1 μCi/mL/plate with 3H-Tdr in conditioned growth medium for

24 hr, then replaced with 0.5 mL 5% serum containing MEM/Hepes loaded with 40 μM

2+ 2+ 2+ II ([Ru(tpy)(pydppn)] ), I ([Ru(tpy)2] ), III ([Ru(pydppn)2] ) in for another 24 hr, then

replaced with 300 µL MEM/Hepes and irradiated at 3.15 J cm-2 , then lysed with

0.5 mL Hirt lysis buffer. Aliquots of the lysate were removed for quantitation of protein-

DNA crosslinking before and after proteinase K digestion (37 °C, overnight). The points

are averages of triplicates, and the error bars are ± SD.

2+ 2+ 2+ -2 I: [Ru(tpy)2] , II: [Ru(tpy)(pydppn)] , III: [Ru(pydppn)2] , dk: dark, lt: 3.15 J cm .

101

45.0% 40.0% 35.0% C 30.0% without proteinase K digest 25.0% 20.0% with proteinaseK digest 15.0% 10.0% 5.0% 0.0% % protein-DNA crosslinking

Fig 3.8 Protein-DNA photocrosslinking

102

Fig 3.9 Chromosomal crosslinking assay

CV-1 cells (100% confluence, in 100 mm culture plates) were treated with 40 µM

[Ru(tpy)(pydppn)]2+ at 37 °C for 24 hr before irradiation at 3.15 J cm-2. Then the cells were lysed with 8M GuHCL/1% sarkosyl, vortexed and heated at 65 °C for 15 min. The cell lysates were banded on CsCl step gradients before ultracentrifugation. The fractions containing crosslinked protein-DNA were pooled and dialyzed. The dialyzed sample was then precipitated with 2 volumes of 100% ethanol and 2 M ammonium acetate and treated with Omnicleave nuclease to remove DNA, followed by lyophilization to reduce the sample volume. The samples were subsequently separated by SDS PAGE and visualized by silver staining.

103

[Ru(tpy)(pydppn)]2+ HCHO Marker Dark irradiated

200 kDa

116 kDa

97 kDa

66kDa

45 kDa

31 kDa

22 kDa

14 kDa

Fig 3.9 Chromosomal crosslinking assay

104

Fig 3.10 Photodynamic inhibition of DNA replication by [Ru(tpy)(pydppn)]2+

CV-1 cells were seeded into a 48 well plate (10,000 cells/well). 24 hr later, the medium was replaced with fresh medium. At 30 hr post-seeding, [Ru(tpy)(pydppn)]2+

was added at different concentrations in MEM/Hepes, 5% CS. After 24 hr of exposure to

[Ru(tpy)(pydppn)]2+, the medium was removed, the cells were rinsed with PBS, and then

exposed to light (3.15 J cm-2, in SFM). The SFM was then replaced with labeling medium

(5 μCi/mL 3H-Tdr, MEM/Hepes, 10% CS) for 1 hr. 3H-Tdr incorporation into DNA was

then measured as described above under Materials and Methods. There were six

duplicates per point. Error bars are ± SD.

105

Fig 3.10 Photodynamic inhibition of DNA replication by [Ru(tpy)(pydppn)]2+

106

CHAPTER 4

ANALYSIS OF THE INHIBITORY EFFECT ON

DNA REPLICATION BY LIGHT-ACTIVATED

[RU(BPY)2(CH3CN)2]CL2

4.1 Introduction

In the ongoing search for non-platinum transition metal containing complexes to

overcome the side effects of cisplatin and its derivatives, much effort has been devoted to

the synthesis and characterization of Ru(II) complexes as potential alternatives (Brabec

and Nováková, 2006; Clarke, et al., 1999; Gianferrara, et al., 2009).

Ru(II) complexes are able to react to different microenvironments by changing their

absorption and emission behavior. Certain Ru(II) complexes with extended planar ligands

display a “light-switch” behavior which features recovery of luminescence when they

bind to DNA helix by ligand intercalation (Herman, et al., 2008; Liu, et al.,2007). The unique photochemical and photophysical properties of Ru(II) complexes have transformed them into efficient photosensors or photoprobes, and more recently, photosensitizers for photodynamic therapy (Clarke, 2003; Lo, 2007; Metcalfe and

Thomas, 2003; Pierard and Kirsch-De Mesmaeker, 2006).

107

Earlies studies focused on the interaction between Ru(II) complexes and DNA from

the spectroscopic and electrochemical perspectives, the more recent studies concentrated

more on in vitro DNA photocleavage resulting from photo-activated Ru(II) complexes

(Davia, et al., 2008; Liu, et al., 2007; Mongelli, et al., 2006; Singh and Turro, 2004).

However, the effect of Ru(II) complexes on DNA replication in the biological system

remains largely unknown.

[Ru(bpy)2(CH3CN)2]Cl2 (bpy = 2,2’-bipyridine) has been reported to undergo stepwise exchange of CH3CN ligand with H2O during irradiation, which is reminiscent of

the working mechanism of cisplatin (cis-Pt(NH3)2Cl2) (Liu, et al., 2008). In cisplatin, the

ligand replacement occurs between chloride ion and water molecule, and the resulting

2+ cation cis-[Pt(NH3)2(H2O)] is the species that is able to crosslink to the DNA helix and

produce the antitumor effect observed in clinical practice. Although the mechanism of

ligand loss and recombination are shared by both complexes, the photo-induced activity

of [Ru(bpy)2(CH3CN)2]Cl2 provides an additional layer of specificity and is therefore

more targeted.

High resolution agarose gel electrophoresis coupled with gel fluorography can be

used to detect fine differences (compactness, topology, shape, etc) in viral DNA replication intermediates resulting from compounds that interfere with specific steps in viral DNA replication (Snapka, 1996). Here we used the well-established SV40 virus replication system and high resolution agarose gel electrophoresis to study the relationship between photo-activated [Ru(bpy)2(CH3CN)2]Cl2 and viral DNA replication

by analyzing the structures of aberrant viral DNA replication intermediates, in an effort to

gain some insight into the mechanism of disruption of mammalian DNA replication 108

caused by light-activated [Ru(bpy)2(CH3CN)2]Cl2. We found that proteins might not the

primary cellular targets of [Ru(bpy)2(CH3CN)2]Cl2 , however, DNA replication was

significantly inhibited by light-activated [Ru(bpy)2(CH3CN)2]Cl2.

4.2 Materials and Methods

Cell lines and Virus

African green monkey kidney fibroblasts (CV-1) were obtained from the American

Type Culture Collection (ATCC) and cultured in MEM medium (Invitrogen) containing

10% calf serum (Invitrogen) and 14 mM Hepes (Sigma). For SV40 infection, CV-1 cells

were infected with SV40 virus (strain 777, 2.3 x 107 pfu/mL) for 1 hr at 37 °C and

cultured with complete media for appropriate times before drug treatment.

Complexes and drugs

Proflavine (PF) was purchased from Sigma. Stock solution(40 mM) was aliquoted

and stored frozen in the dark at -20 °C. [Ru(bpy)2(CH3CN)2]Cl2 and its two controls

[Ru(bpy)2(en)]Cl2, [Ru(tpy)(AN)3]Cl2 in the powder form were kindly provided by

Dr. Turro (Department of Chemistry, the Ohio State University) and the stock solutions

(10 mM) were prepared in distilled water immediately before use and kept in the dark at room temperature.

GF/C assay of 3H-Tdr radiolabeled SV40 DNA

At the time point of 12 hr post SV40 virus infection, CV-1 cells were treated with

109

[Ru(bpy)2(CH3CN)2]Cl2 at the desired concentration in the dark and then incubated at

37 0 C for another 24 hr, followed by 3H thymidine (250 μCi/mL) labeling for a total

length of 30 min. For those cells to be irradiated, the labeling was divided into two stages

(the first stage was in the dark at 37 0 C, the second stage was in the irradiator at room

temperature) but the total labeling length was still 30 min. The cells were then lysed with

0.3 mL Hirt lysis buffer (10 mM Tris Cl, pH 7.5, 10 mM EDTA, 0.6% SDS) per plate for

15 min at room temperature. The lysates were transferred to 1.5 mL microcentrifuge

tubes that contain 75 µL 5 M NaCl, gently mixed by inversion for 40 times, and then

store at 4 0 C overnight. The SV40 viral minichromosomes were then separated from

cellular chromosomes by centrifugation as the viral minichromosomes will remain in the

supernatant and the cellular chromosomes in the pellet. An aliquot of the supernatant was

then used for GF/C assay at 0.4 M and 4 M GuHCl to measure SV40 viral DNA-protein

crosslinking as described (Shin, et al., 1990).

Fluorography of SV40 replicating DNA intermediates

SV40 viral minichromosomes were digested with 1/10 vol proteinase K (1 mg/mL

prepared in Hirt lysis solution) at 37 0 C overnight. The digested sample was then extracted with an equal volume of CHCl3: isopropyl alcohol (24:1 vol/vol) by mixing and

placed in wet ice for ~ 15 min. The sample was then centrifuged at 14,000 rpm (4 0 C) for

15 min to isolate SV40 viral DNA and the top aqueous phase was transferred to fresh tubes. The SV40 viral DNA was precipitated by adding 2.5 vol ice-cold 95% ethanol on crushed dry ice for 15 min. Then the sample was centrifuged at 14,000 rpm (4 0 C) for 30 min to pellet the viral DNA. After drying, the viral DNA was solubilized in 12 μL 1X 110

sample loading buffer (3.75% Ficoll, 0.1% SDS, 0.025% bromophenol blue, 0.025%

Xylene Cyanol FF) at 37 0 C for 1 hr before loading onto the 0.8% agarose gel. Following

gel electrophoresis, the gel was immersed overnight in 600 mL fluor solution (600 mL of

glacial acetic acid with 12 g of PPO (2,5-diphenyloxazole)). Then the fluor solution was removed and the gel dried. In the dark room, XAR( Kodak) film was pre-flashed and then placed on top of the plastic-wrapped gel and the film was exposed at -75 0 C for 15 days

before development.

Photodynamic treatment and Western blot analysis

The photodynamic treatment, whole cell protein extracts, RIPA lysis of cells, SDS

PAGE, and Western blotting were done as described (Bae, et al., 2008).

4.3 Results

[Ru(bpy)2(CH3CN)2]Cl2 can not photodynamically crosslink PCNA

As we have identified PCNA as a highly sensitive target to photodynamic damage in the cell culture system, we first tested the response of this protein to

[Ru(bpy)2(CH3CN)2]Cl2 in light in CV-1 cells. Two other complexes, [Ru(bpy)2(en)]Cl2 and [Ru(tpy)(AN)3]Cl2 were included as controls. As shown in Fig 4.1, there was no

change of PCNA monomer or appearance of high molecular weight species in response

to the treatment of these complexes in light as detected by Western blotting. Protein

degradation or protein-protein crosslinking would cause an electrophoretic mobility shift,

111

but the possibility of formation of protein-adducts or single amino acid residue damage

can not be ruled out as they may or may not lead to altered electrophoretic mobility.

GF/C filter binding assay for protein-SV40 viral DNA crosslinking

We then used the SV40 viral DNA replication system and the GF/C binding assay to

find out whether or not proteins crosslinked to DNA in response to

[Ru(bpy)2(CH3CN)2]Cl2 in light. As shown in Fig 4.2, there was no significant protein-

viral DNA crosslinking for those cells treated with [Ru(bpy)2(CH3CN)2]Cl2 under either

dark or light conditions, which strengthened the idea that proteins might not be major

targets by light activated [Ru(bpy)2(CH3CN)2]Cl2. However, a closer examination of the

result showed that the radiolabelled DNA in the [Ru(bpy)2(CH3CN)2]Cl2 plus light treated

cells was less than 40% of that in control cells, implying that the DNA replication was

significantly inhibited. The result also showed that light or [Ru(bpy)2(CH3CN)2]Cl2 alone caused inhibition of SV40 viral DNA replication, but the inhibitory effect was more significant with [Ru(bpy)2(CH3CN)2]Cl2 in light.

Gel electrophoresis and fluorography of SV40 replicating DNA intermediates

To get a mechanistic view of how the SV40 viral DNA replication was affected by light-activated [Ru (bpy)2(CH3CN)2]Cl2, we employed high-resolution agarose gel

electrophoresis and tritium fluorography. First, we used proflavine as the photosensitizer

to test whether or not this technique can be used for detecting the differences in DNA

replication intermediates between dark and light conditions. As shown in Fig 4.3, the

addition of proflavine under room light conditions, or dark conditions, caused an increase 112

of catenated dimers as proflavine is a catalytic topoisomerase II inhibitor (Shin, et al.,

1990). The catenated dimers are topologically linked, unresolved newly replicated SV40

daughter chromosomes. Under irradiated conditions, however, both form I and form II

(the completely replicated forms) were significantly reduced compared to the dark

condition, suggesting that the generation of daughter DNA molecules was significantly

inhibited by the photodynamic damage of proflavine. More work is needed for the study

of [Ru(bpy)2(CH3CN)2]Cl2 on the vial DNA replication in this system.

4.4 Discussion

In this preliminary work, we studied the properties of [Ru(bpy)2(CH3CN)2]Cl2, a ruthenium containing complex that interacts with DNA through ligand to water exchange under the activation of light. The non-reactivity of PCNA to light-activated

[Ru(bpy)2(CH3CN)2]Cl2 supported the idea that PCNA is not greatly altered by this light-

activated complex. The lack of PCNA peptide backbone cleavage, or electrophoretic

mobility shift by light and [Ru(bpy)2(CH3CN)2]Cl2, and the lack of protein-DNA

crosslinking, argue against major protein damage. However, the possibility of amino acid

residue damage is not ruled not. SV40 Viral DNA replication was greatly inhibited by

[Ru(bpy)2(CH3CN)2]Cl2 in light. As both [Ru(bpy)2(CH3CN)2]Cl2 alone or light alone

inhibit SV40 DNA replication as measured by 3H-Tdr incorporation, our current data of

the greatly reduced SV40 replication with [Ru(bpy)2(CH3CN)2]Cl2 plus light may be the

additive inhibition effect of these two factors. More work is needed to provide deeper

insight into the inhibition mechanism of DNA replication at the biochemical level.

113

Figure 4.1 PCNA can not be photodynamically crosslinked by

[Ru(bpy)2(CH3CN)2]Cl2 and the controls

CV-1 cells (~80% confluence) were treated with [Ru(bpy)2(CH3CN)2]Cl2 and the two

controls [Ru(bpy)2(en)]Cl2, [Ru(tpy)(AN)3]Cl2 at 40 µM (37 °C, 18 hr) in MEM/Hepes supplemented with 5% serum, then replaced with serum-free medium (SFM) and exposed to visible light irradiation (3.15 J cm-2). Western blotting was carried out using anti-PCNA (PC10) antibody. PCNA monomer (PCNA) was marked.

-2 Ctrl: control; lt: 3.15 J cm ; A: [Ru(bpy)2(en)]Cl2, B: [Ru(bpy)2(CH3CN)2]Cl2,

C: [Ru(tpy)(AN)3]Cl2

114

lt dark lt lt dark lt lt dark lt lt ctrl A A A B B B C C C

PCNA

Figure 4.1 PCNA can not be photodynamically crosslinked by

[Ru(bpy)2(CH3CN)2]Cl2 and the controls

115

Fig 4.2 GF/C assay of SV40 viral DNA-protein crosslinking by

Ru(bpy)2(CH3CN)2]Cl2 in light

CV-1 cells (around 100% confluence) were infected by SV40 virus, 14 hr later,

[Ru(bpy)2(CH3CN)2]Cl2 was added to the final concentration of 40 µM for another 24 hr.

The cells were then pulse-labeled with 250 µCi/mL 3H-Tdr (prepared in 200 µL

SFM/plate) at 37 °C for 30 min. For those to be irradiated, the irradiation was 15 min

(6.75 J cm-2) in the irradiator without removing the labeling medium. Then the cells were lysed and processed for GF/C assay to measure the level of protein-DNA crosslinking

(Shin, et al., 1990) as described in Materials and Methods. Triplet samples were assayed for each condition. cpx: [Ru(bpy)2(CH3CN)2]Cl2.

116

SV40 DNA 3H-Tdr incorporation

120.0% protein-DNA crosslinking

100.0%

80.0%

60.0%

40.0%

20.0% SV40 DNA Replication 0.0% dark light dark light ctrl ctrl cpx cpx

Fig 4.2 GF/C assay of SV40 viral DNA-protein crosslinking by

[Ru(bpy) (CH CN) ]Cl in light 2 3 2 2

117

Fig 4.3 3H-Tdr fluorography of SV40 viral DNA replication intermediates in response to proflavine in light

CV-1 cells (around 100% confluence) were infected by SV40 virus, 36 hr later, the

cells were pulse labeled with 250 µCi/mL of 3H-Tdr at 37 °C. At the time point of 15 min,

proflavine (40 µM) was added to the infected cells either under room light, or in the dark.

For those to be irradiated, irradiation was done at the last 7 min of the labeling period

(3.15 J cm-2) in the irradiator without removing the labeling medium. Then the viral DNA

wad processed as described in the Materials and Methods.

ctrl: control; PF: proflavine (40 µM); RL: room light; light: 3.15 J cm-2.

118

dark RL dark light ctrl PF PF PF

Ori

A1 A2 LC A3 A4 A5 B1 B2 B3 B4 B5 Form II Form III Highly catenated dimers

Form I

Fig 4.3 3H-Tdr fluorography of SV40 viral DNA replication intermediates

in response to proflavine in light

119

CHAPTER 5

ANALYSIS OF THE CYTOTOXIC EFFECT OF

PHOTODYANMIC TOPOISOMERASE

POISONS ON MAMMALIAN CELLS

5.1 Introduction

Topoisomerases I and II are essential enzymes for DNA replication and transcription,

since the topology of DNA templates can not be properly regulated without them

(Champoux, 2001, Wang, 2002). Because of their vital roles, much effort has been put into the study of compounds that target topoisomerase I/II , thus providing an anti-cancer effect in clinical practices. Among those topoisomerases-targeting drugs, topoisomerase

I/II poisons belong to a unique class as they are believed to be cytotoxic to cells by trapping a “topoisomerase I/II poisons-topoisomerase I/II-DNA” ternary structure (also called “cleavable complex”) instead of directly inhibiting the catalytic activity of topoisomerases (Froelich-Ammon and Osheroff, 1995). CPT (Camptothecin), a plant alkaloid, is the best known poison for mammalian topoisomerase I (Liu, et al., 2000;

Rothenberg, 1997). CPT interferes in the rejoining step of the cleavage/religation reaction in the course of topoisomerase I activity, resulting in an arrested DNA strand passing

120

intermediate in which topoisomerase I is covalently attached to DNA at the site of the

DNA strand break (the cleavable complex). CPT cytotoxicity is primarily derived from

collisions of moving DNA replication forks with CPT-stabilized topoisomerase I-DNA

cleavable complexs, resulting in complex double strand DNA breaks (Snapka, 1986). In

addition, the collision between the forward-moving transcription machinery and the

“topoisomerase I cleavable complexes” can also contribute to CPT cytotoxicity (Liu, et al., 2000). Currently, CPT-based derivatives, including irinothecan, topotecan, and 9- amino camptothecin, have been widely used in clinical practice (Rothenberg, 1997). For mammalian topoisomerase II, etoposide (VP-16), daunorubicin, doxorubicin

(Adriamycin), m-AMSA(4′-(9-acridinylamino) methanesulfon-m-anisidide), ellipticine are specific poisons that have been extensively studied (Burden and Osheroff, 1998).

Both topoisomerase I and topoisomerase II poisons are widely used in cancer chemotherapy.

Studies have shown that some of the widely used topoisomerase I/II poisons possess photochemical properties, which make it possible to bridge topoisomerase I/II poisoning and photo-damage. The photochemical properties of CPT in the presence of copper (II) ions was studied, and it was suggested that free radicals and singlet oxygen generated by

CPT upon irradiation play a central role in DNA damage (Brezová, et al., 2007). The combination of CPT and photodynamic treatment produced a slight synergistic effect at low CPT concentrations and high PDT doses in V79 and NHIK 3025 cells (Gaullier, et al., 1996). Doxorubicin (DOX) and other anthracyclines have also been shown to be weak photosensitizers with low quantum yields (Andreoni, et al., 1991). It was reported that irradiation significantly increased the sensitivity of several cell lines to daunomycin, 121 as well as its imino or amino derivatives (Andreoni, et al., 1993; Andreoni, et al., 1989;

Andreoni, et al., 1991). Much effort has also been exerted in the study of the photochemical properties of anthracyclines (Andreoni, et al., 1992; Andreoni, et al.,

1989). The cytotoxicity of doxorubicin have been reported to be enhanced by high- energy laser in several cell lines, and a certain degree of DNA damage was detected, which was thought to be involved in the molecular mechanism for the light-increased cytotoxicity (Gao, et al., 1997).

Taken together, these studies provide evidence that the photochemical properties of these chemotherapeutic drugs such as camptothecin might be useful for improving the efficiency of the corresponding chemotherapies. We have further studied the photodynamic damage of several well-established topoisomerase I/II poisons in relation to their effect on cell survival.

5.2 Materials and Methods

Cell lines

Drug sensitive (wild-type) parental CV-1 cells (African green monkey kidney cell) were obtained from ATCC and cultured in MEM medium (Invitrogen) containing 10% fetal calf serum and 14 mM Hepes. Two drug resistant CV-1 cell lines

(CPTCV10c22cells resistant to 1.5 μM CPT and AMCV1cells resistant to 3 μM m-

AMSA) were cultured in 10% fetal calf serum and 14 mM Hepes under the constant selective pressure of 1.5 μM CPT or 3 μM m-AMSA. CPTCV10c22 was from CV-1 cells selected stepwise for CPT resistance starting from 8 nM CPT following a single

122

exposure to germicidal UV light irradiation (254 nm, 30 J m-2). AMCV1 cells were from

CV-1 cells that were selected stepwise for resistance to m-AMSA, starting from 20 nM

m-AMSA following a single exposure to germicidal UV light irradiation (254 nm, 30 J

m-2) (Gao, et al., 1999).

Compounds

Camptothecin was obtained from the National Cancer Institute, Developmental

Therapeutics Program (Frederick, MD). Ellipticine was from Sigma. Camptothecin and

ellipticine were dissolved in DMSO. Stock solutions (10 mM) were aliquoted and stored frozen in the dark at -20 °C. Phenylmethylsulfonyl fluoride (PMSF), dithiothreitol (DTT), dimethyl sulfoxide (DMSO), glycerol were from Sigma. The protease inhibitors, aprotinin, leupeptin, and pepstatin were from USB (Cleveland, OH).

MTT assay

The MTT reduction assay was used to measure the cytotoxicity of drugs of interest

(Mosmann, 1983). MTT (3-(4,5-dimethylthiazol-2-Yl)-2,5-diphenyltetrazolium bromide),

a tetrazolium salt, is reduced to purple formazan in living cells and the conversion ratio

can be used as a measure of the cell viability. Cells were seeded at an appropriate density

in 96-well tissue culture plates and incubated in MEM medium supplemented with 10% fetal calf serum. After 24 hr incubation at 37 °C, camptothecin or ellipticine at different concentrations were added to the cells in the dark and the incubation continued in the

dark for another 2 hr before irradiation. At the end of irradiation, the old medium was

replaced with fresh growth medium after PBS rinsing, and the cells were incubated at 123

37 °C for another 2 days. MTT was then added to a final concentration of 0.5 mg/mL and

the incubation continued in the dark for 4 hr. The medium was then removed, DMSO

(100 µL per well) added, and the incubation continued for another 1 to 2 hr with gentle

rocking at a low speed of 50 rpm. The absorbance of each well was measured in a Model

EL800 automated EIA plate reader (Bio-Tek Instruments Inc, Winooski, VT) at 490 nm.

Values in the drug solvent alone were used as the untreated control.

Methylene blue based growth inhibition assay

The methylene blue based assay can be used to measure the cellular growth inhibition level (Finlay, et al.,1984). Cells were seeded at an appropriate density in the 24-well tissue culture plates to reach ~70% confluency. Then ellipticine was added to the cells in serum-free medium in the dark for 2 hr before irradiation at 27 J cm-2 , and incubated in

the dark for another 1 hr. Then the plates were rinsed with PBS twice and fresh growth medium added. The plates were kept at 37 °C for another 3 days. The cells were rinsed once with PBS and fixed with acetic acid: methyl alcohol (1:3) at room temperature for

30 min and then stained with 1% methylene blue (dissolved in water). The staining solution was then removed and the wells dried, 500 µL acetic acid: methyl alcohol: water

(1:5:4) was added to each well. The absorbance of each well was determined in a Model

EL 800 automated EIA plate reader (Bio-Tek Instruments Inc) at 660 nm.

Clonogenic survival assay

The clonogenic survival assay is used to measure the effect of drugs of interest on the survival and proliferation of cells (Munshi, et al., 2005). CPTCV10c22 cells (resistant to 124

1.5 µM CPT) were seeded at a density of 1× 106 cell for each 100 mm tissue culture

plates in MEM medium supplemented with 10% fetal calf serum for overnight, then

treated with 10 µM CPT and kept in the dark for 2 hr before irradiation (27 J cm-2). Then

the cells were trypsinized and immediately reseeded in triplicate at a density of 5, 000

cells per plate in MEM medium supplemented with 10% fetal calf serum and 1.5 µM

CPT. The cells were cultured at 37 °C for another 14 days, and then rinsed with PBS

once and fixed with methanol: acetic acid (1:3) and stained with 2% crystal violet

(dissolved in distilled water). After removal of staining solution, the plates were rinsed

thoroughly with water and completely dried. The stained colonies (>= 50 cells) were

counted.

Western blot analysis

The Western blotting procedure was as described (Bae, et al., 2008). Primary

antibodies used were scleroderma patient serum anti-human topoisomerase I (Topogen,

Columbus, OH), mouse anti-actin (Sigma), polyclonal rabbit anti-human topoisomerase

IIα (Topogen), polyclonal rabbit anti-human topoisomerase IIß was a gift from

Danial M. Sullivan (H. Lee Moffitt Hospital, Tampa, FL).

Photodynamic treatment

The photodynamic irradiation was conducted as described (Bae, et al., 2008). For cell

experiments, cells in 35 or 100 mm tissue culture plates (Corning) were irradiated with plate lids in place. For cytotoxicity assays, cells in tissue culture plates (Corning) were

125 irradiated with plate lids in place. Before and after irradiation, samples containing photodynamic drugs were handled under dim red darkroom light.

5.3 Results

The topoisomerase status in two topoisomerase I/II poison-resistant cell lines

Two drug-resistant CV-1 cell lines with greatly reduced topoisomerase I or topoisomerase II levels have been developed and characterized (Gao, et al., 1999), which have been cultured continuously in the drug-containing growth medium. CPTCV10c22, resistant to 1.5 μM CPT, is a cell line in which the topoisomerase I expression level has been reduced to about 10% of the wild-type (Gao, et al., 1999). AMCV1, resistant to

3 μM m-AMSA, is a cell line which has only 10% of normal topoisomerase IIα function and no detectable topoisomerase IIβ (Gao, et al., 1999). We first examined the topoisomerase I and topoisomerase II status in these two mutant cell lines, using parental

CV-1 cells as the positive control. The Western blotting result was shown in Fig 5.1. It was clear that topoisomerase IIα and IIβ were below detection in AMCV1 cells and topoisomerase I was below detection in CPTCV10c22 cells. Owing to the detection limit, the remaining low level of topoisomerase I or topoisomerase IIα is difficult to detect in our western blot. Topoisomerase I is not essential for cell survival while topoisomerase II is (Wang, 2002). Although levels of topoisomerase I (in CPTCV10c22 cells) and topoisomerase II (in AMCV1) is hard to detect, the low express levels of both enzymes can not be ruled out.

126

The weak photodynamic property of camptothecin

As stated earlier, CPT is used as a chemotherapeutic drug to inhibit the DNA

replication of cancer cells by specifically poisoning topoisomerase I. However, when

we combined the visible light irradiation and CPT to treat CV-1 cells, a weak but clear

high molecular weight band of PCNA was detected in the 10% SDS-PAGE, as shown

in Fig 5.2. Since we have identified PCNA as a sensitive target for photodynamic

damage, this result is consistent with the reports that CPT is a weak photosensitizer.

The cytotoxicity of camptothecin in light

The next question we wanted to address was how irradiation alters the cellular toxicity

of CPT. We first monitored the cellular growth behavior in response to CPT in the dark

and the result is shown in Fig 5.3A. CV-1 cells were very sensitive to the effect of CPT,

and the cell survival rate was reduced by half at a concentration at 0.15 µM. Next we

compared the cell survival rate by CPT treatment both in the dark and low level light

irradiation (3.15 J cm-2,visible light) conditions. Not much difference was observed for these two conditions (Fig 5.4A), which might be related to the fact that CPT is a comparatively weak photosensitizer. As we have confirmed that CPTCV10c22 cells are highly resistant to CPT, we included this cell line in our study to test the photodynamic effect of CPT at higher concentrations and higher irradiation dose. As we can see from

Fig 5.5A, with the increase of CPT dose, the cell survival rate reduced gradually, but not very significantly, and there was a weak reduction in cell survival for the irradiated cells.

We then turned to a more accurate clonogenic survival assay to get a better idea of the response of CPTCV10c22 cells to CPT in light (Fig 5.7). The colony formation capability 127

of those cells treated with CPT under light was reduced by about 10%, which suggested

that the combination of photodynamic properties and topoisomerase I poisoning worked

together to inhibit cellular survival to a greater extent.

The cytotoxicity of ellipticine in light

Ellipticine (EL) is a specific topoisomerase II poison (Burden and Osheroff, 1998)

and also a photosensitizer (Bae, et al., 2008) . Here we were interested in whether or not

the cytotoxicty of EL can be influenced by light. First we examined the cell survival to

EL in the dark. As we can see from 5.3B, the cell survival rate was reduced by half at the concentration of around 2.5 µM. Next we compared the cell survival rate caused by EL under both the dark and low level irradiation settings (3.15 J cm-2 visible light). We

observed a growth inhibition for those irradiated at 2.5 µM EL (Fig 5.4B) .We then

included AMCV1 cells to test the photodynamic effect of EL at a higher irradiation dose.

As we can see from Fig 5.5B, when the irradiation dose went higher, the cell survival rate

decreased sharply at 2.5 µM EL for both cell lines. The overall survival rate for AMCV1

cells was a little higher, but it also showed a dramatic reduction of cell survival rate at 2.5

µM EL. The growth inhibition assay also confirmed this observation (Fig 5.6).

5.4 Discussion

Topoisomerase I/II poisons have been widely applied in cancer treatment because of their high anti-tumor effectiveness. Although the photodynamic properties of certain topoisomerase I/II drugs have come to be recognized, it remains elusive whether or not

128 their photodynamic activities can add to their anti-cancer effectiveness. With this in mind, we conducted this study in an effort to make better use of their dual property.

The oligomerization of PCNA in response to CPT in light (Fig 5.2) was consistent with the idea that CPT is a photosensitizer, although its photodynamic effect is very weak, as evidenced by the weak PCNA crosslinking band detected by Western blotting, and the small difference in terms of cell survival rate for both parental CV-1 cells and CPT- resistant CPTCV10c22 cells under light activation conditions (Fig 5.4A, 5.5A, 5.7).

The clonogenic survival assay data pointed out the involvement of the photodynamic effect of CPT in contributing to the lower cell survival rate observed for cells. But the difference is not very significant, which might be due to such factors as the low quantum yield of singlet oxygen for CPT, or the reversible diffusion of CPT from the “cleavable complex” (a short residence time) which would lead to less singlet oxygen damage to the topoisomerase-DNA complex. The lack of a significant differences in CPT phototoxicity between wild type CV-1 cells and topoisomerase I deficient CPTCV10c22 cells suggests that topoisomerase I-DNA cleavable complexes are not the key cytotoxic target for CPT generated photodamage. A greater cytotoxic effect was observed for ellipticine in light

(Fig 5.4B, 5.5B, 5.6). Both the MTT assay and the methylene blue-based growth inhibition assay showed that the photo-activated ellipticine was significantly more potent in reducing cell survival, in both parental CV-1 cells and AMCV1 cells. The fact that the cytotoxicity caused by ellipticine plus light treatment is similar for both parental CV-1 cells (wild-type topoisomerase IIα and topoisomerase IIβ) and AMCV1 (greatly reduced

IIα and undetectable topoisomerase IIβ) suggests that photodynamic damage to the ellipticine stabilized “topoisomerase II-DNA-ellipticine” complex is not a significant 129

contributor to the higher cytotoxicity caused by ellipticine in light. There might exist

more critical molecular targets for ellipticine photodynamic damage. These results are promising in that it might open up the possibility of exploiting the intrinsic photodynamic properties of ellipticine to increase the anti-cancer effectiveness to a higher level.

130

Fig 5.1 Topoisomerase I/II status in the drug resistant mutant cell lines

(A) CV-1 cells and CPTCV10c22 cells in exponential growth were lysed by SDS lysis buffer and Western blotted with scleroderma patient serum anti-human topoisomerase I antibody. Actin was the loading control. (B) CV-1 cells and AMCV1 cells in exponential growth were lysed by SDS lysis buffer and Western blotted with rabbit anti-human topoisomerase IIα or rabbit anti-human topoisomerase IIß antibody.

Actin was the loading control.

131

CPTCV10c22 CV-1

topoisomerase I

actin

B AMCV1 CV-1

topoisomerase II α

topoisomerase II β

actin

Fig 5.1 Topoisomerase I/II status in the drug resistant mutant cell lines

132

Fig 5.2 PCNA crosslinking caused by camptothecin in light

Confluent CV-1 cells were subjected to camptothecin (CPT, 200 μM ) plus visible light treatment (3.15 J cm-2) as indicated, then the cells were lysed with SDS lysis buffer and Western blotted against mouse anti-PCNA PC10 antibody. PCNA monomer

(PCNA) and PCNA trimer are marked in the figure.

133

CPT - + irradiation - +

PCNA trimer

PCNA

Fig 5.2 PCNA crosslinking caused by camptothecin in light

134

Fig 5.3 The cytotoxicity of camptothecin and ellipticine for CV-1 cells in the dark by MTT assay

(A) CV-1 cells were seeded at a density of 20,000 cells per well into the 96-well culture plate and incubated at 37 °C for overnight. Then the cells were subjected to CPT in the dark at different concentrations as indicated for 1 hr, then the drug-containing medium was removed and the incubation continued at 37 °C for another 2 days. The cells were then analyzed by the MTT assay. Values in the drug solvent alone were used as the untreated control. At least three replicates were used for the calculation. Standard deviation is represented as the error bars. (B) The procedures are the same as in A except that the compound applied is ellipticine.

135

A 120% 100% 80% 60% 40% 20%

Survival (% Survival (% if cibtrik) 0%

B 140% 120% 100% 80% 60% 40% 20%

Survival (% of control ) control of (% Survival 0%

Fig 5.3 The cytotoxicity of camptothecin and ellipticine for CV-1 cells in the dark by MTT assay

136

Fig 5.4 The cytotoxicity of camptothecin and ellipticine for CV-1 cells in the dark and light by MTT assay

(A) CV-1 cells were seeded at a density of 20,000 cells per well into the 96-well culture plate and incubated at 37 °C for overnight. Then the cells were subjected to CPT in the dark at different concentrations for 1 hr as indicated, then the drug-containing medium was removed before irradiation at 3.15 J cm-2. The incubation continued at 37 °C for another 2 days. The cells were then analyzed by the MTT assay. Values with the drug solvent alone were used as the untreated control. At least three replicates were used for calculation. Standard deviation was represented as the error bars. (B) The procedures are the same as in A except that the compound applied is ellipticine.

137

A 120%

100% CV-1 CPT DARK CV-1 CPT LT 80%

60%

40%

20% Survival (% of control) Survival 0%

120%

100%

80%

60%

40% CV-1 EL DARK

20% CV-1 EL LT Survival (% of control) Survival 0%

Fig 5.4 The cytotoxicity of camptothecin and ellipticine for CV-1 cells in

the dark and light by MTT assay

138

Fig 5.5 The cytotoxicity of camptothecin and ellipticine for CV-1 cells and mutant cells in light by MTT assay

(A) CPTCV10c22 cells were seeded at a density of 10,000 cells per well to the 96- well tissue culture plates and incubated at 37 °C for overnight. The cells were then treated with CPT at different concentrations as indicated in serum-free medium in the dark for 2 hr, then irradiated at 27 J cm-2. Then the incubation continued at 37 °C for another 2 days.

The cells were then analyzed by the MTT assay. Values in the drug solvent alone were used as the untreated control. Standard deviation was represented as the error bars. (B)

The procedures are the same as in A except that CV-1 cells and AMCV1 cells were used and the compound applied was ellipticine.

139

120% CPTCV10c22 DARK 100% CPTCV10c22 LT 80%

60%

40%

20%

0% Survival (% of control cell) Survival CPT 0.0 CPT 5.0 CPT 10.0 CPT 20.0 µM µM µM µM

120% CV-1 DARK 100% CV-1 LT AMCV1 DARK 80% AMCV1 LT 60%

40%

20% Survival (% of control) Survival 0%

Fig 5.5 The cytotoxicity of camptothecin and ellipticine for CV-1 cells and mutant cells in light by MTT assay

140

Fig 5.6 The cytotoxicity of ellipticine for CV-1 cells in light by methylene blue

growth inhibition assay

CV-1 cells were seeded at a density of 10,000 cells per well in 24-well tissue culture

plates and incubated at 37 °C for overnight. The cells were then treated with ellipticine

(EL) at different concentrations as indicated in the serum-free medium in the dark for 2

hr, then irradiated at 27 J cm-2 . The plates were then rinsed with PBS twice and MEM

medium supplemented with 10% fetal calf serum was added. The incubation continued

for another 2 days. The cells were then analyzed by methylene blue assay. The absorbance of each well was determined at 660 nm.

141

140%

120% EL DARK

100% EL LT

80%

60%

40% survival (% of control) 20%

0% EL 0.0 µM EL 2.5 µM

Fig 5.6 The cytotoxicity of ellipticine for CV-1 cells in light by methylene blue growth inhibition assay

142

Fig 5.7 The cytotoxicity of camptothecin for CPTCV10c22 cells in light by

clonogenic survival assay

CPTCV10c22 cells (resistant to 1.5 µM CPT) were seeded at a density of 1× 106 cell for each 100 mm tissue culture plates. Then the cells were treated with 10 µM CPT and kept in the dark for 2 hr before irradiation (27 J cm-2). The cells were trypsinized immediately and reseeded in triplicate at a density of 5, 000 cells per plate. The cells were then cultured at 37 °C for another 14 days, and then rinsed with PBS once, fixed

with methanol: acetic acid (1:3) and stained with 2% crystal violet. The stained colonies

(>= 50 cells) were counted.

143

120% CPTCV10c22 DARK 100% CPTCV10c22 LT 80%

60%

40%

20% Survival(% of control)

0% CPT 0.0 µM CPT 10 µM

Fig 5.7 The cytotoxicity of camptothecin for CPTCV10c22 cells in light by clonogenic survival assay

144

BIBLIOGRAPHY

Alderden, R. A., Hall, M. D., & Hambley, T. W. (2006). The discovery and development of cisplatin. J. Chem. Edu, 83(5), 728-734.

Allison, B. A., Crespo, M. T., Jain, A. K., Richter, A. M., Hsiang, Y. N., & Levy, J. G. (1997). Delivery of benzoporphyrin derivative, a photosensitizer, into atherosclerotic plaque of Watanabe heritable hyperlipidemic rabbits and balloon-injured New Zealand rabbits. Photochem Photobiol, 65(5), 877-883.

Andreoni, A., Colasanti, A., Kisslinger, A., Malatesta, V., Mastrocinque, M., & Roberti, G. (1991). Antitumor drugs as photochemotherapeutic agents. Proc. SPIE, 1525, 351-366.

Andreoni, A., Colasanti, A., Kisslinger, A., Mastrocinque, M., Portella, G., Riccio, P., & Roberti, G. (1993). Enhanced response to daunomycin of normal, tumor and metastatic cell lines via drug photo activation. Photochem Photobiol, 57(5), 851-855.

Andreoni, A., Colasanti, A., Malatesta, V., Mastrocinque, M., Roberti, G., & von Kisslinger, A. (1989). Cell photosensitization by 5-iminodaunomycin activated with red light. Biochim Biophys Acta Mol Cell Res, 1014(1), 8-13.

Andreoni, A., Colasanti, A., Malatesta, V., Riccio, P., & Roberti, G. (1991). Enhancement of antitumor drug cytotoxicity via laser photoactivation. Photochem Photobiol, 53(6), 797- 805.

Andreoni, A., Colasanti, A., & Roberti, G. (1992). Excited singlet state properties of anthracenedione photosensitizers. J Photochem Photobiol B, 14(4), 319-328.

Andreoni, A., Land E, J., Malatesta, V., McLean, A. J., & Truscott, T. G. (1989). Triplet state characteristics and singlet oxygen generation properties of anthracyclines. Biochim. Biophys. Acta, 990(2), 190-197.

Ang, W. H., & Dyson, P. J. (2006). Classical and non-classical ruthenium-based anticancer drugs: towards targeted chemotherapy. Eur J Inorg Chem, 2006(20), 4003-4018.

Bachowski, G., Korytowski, W., & Girotti, A. (1994). Characterization of lipid hydroperoxides generated by photodynamic treatment of leukemia cells. Lipids, 29(7), 449-459.

Bachowski, G. J., Pintar, T. J., & Girotti, A. W. (1991). Photosensitized lipid peroxidation and enzyme inactivation by membrane-bound merocyanine 540: reaction mechanisms in the absence and presence of ascorbate. Photochem Photobiol, 53(4), 481-491.

145

Bae, S., Zhao, R., & Snapka, R. M. (2008). PCNA damage caused by antineoplastic drugs. Biochem Pharmacol, 76(12), 1653-1668.

Baert, L., Berg, R., Van Damme, B., D'Hallewin, M.A., Johansson, J., Svanberg, K., & Svanberg, S. (1993). Clinical fluorescence diagnosis of human bladder carcinoma following low- dose Photofrin injection. Urology, 41(4), 322-330.

Bell, S., Hansen, S., & Buchner, J. (2002). Refolding and structural characterization of the human p53 tumor suppressor protein. Biophys Chem, 96(2-3), 243-257.

Bischoff, J. R., Friedman, P. N., Marshak, D. R., Prives, C., Beach, D. (1990). Human p53 is phosphorylated by p60-cdc2 and cyclin B-cdc2. Proc Natl Acad Sci USA,87(12),4766- 4770.

Brabec, V., & Nováková, O. (2006). DNA binding mode of ruthenium complexes and relationship to tumor cell toxicity. Drug Resist Updat, 9(3), 111-122.

Brezová, V., Dvoranová, D., Žúbor, V., Breza, M., Mazúr, M., & Valko, M. (2007). Photochemical properties of camptothecin in the presence of copper(II) ions: the role of radicals as prospective species in photodynamic therapy. Mol Biotechnol, 37(1), 48-51.

Brindell, M., Kuliś, E., Elmroth, S. K., Urbańska, K., & Stochel, G. (2005). Light-induced anticancer activity of [RuCl2(DMSO)4] complexes. J Med Chem, 48(23), 7298-7304.

Brown, S. B., Brown, E. A., & Walker, I. (2004). The present and future role of photodynamic therapy in cancer treatment. Lancet Oncol, 5(8), 497-508.

Bulavin, D. V., Saito, S., Hollander, M. C., Sakaguchi, K., Anderson, C. W., Appella, E., & Fornace, A. J. (1999). Phosphorylation of human p53 by p38 kinase coordinates N- terminal phosphorylation and apoptosis in response to UV radiation. EMBO J, 18(23), 6845-6854.

Burden, D. A., & Osheroff, N. (1998). Mechanism of action of eukaryotic topoisomerase II and drugs targeted to the enzyme. Biochim Biophys Acta, 1400(1-3), 139-154.

Bykov, V. J., & Wiman, K. G. (2003). Novel cancer therapy by reactivation of the p53 apoptosis pathway. Ann Med, 35(7), 458-465.

Carr, A., & Frei, B. (1999). Does vitamin C act as a pro-oxidant under physiological conditions? FASEB J., 13(9), 1007-1024.

Castano, A. P., Demidova, T. N., & Hamblin, M. R. (2004). Mechanisms in photodynamic therapy: part one-photosensitizers, photochemistry and cellular localization. Photodiagnosis Photodyn Ther, 1(4), 279-293.

Castano, A. P., Demidova, T. N., & Hamblin, M. R. (2005). Mechanisms in photodynamic therapy: part two-cellular signaling, cell metabolism and modes of cell death. Photodiagnosis Photodyn Ther, 2(1), 1-23.

146

Castano, A. P., Demidova, T. N., & Hamblin, M. R. (2005). Mechanisms in photodynamic therapy: Part three-Photosensitizer pharmacokinetics, biodistribution, tumor localization and modes of tumor destruction. Photodiagnosis Photodyn Ther, 2(2), 91-106.

Champoux, J. J. (2001). DNA topoisomerases: structure, function, and mechanism. Annu Rev Biochem, 70(1), 369-413.

Chehab, N. H., Malikzay, A., Appel, M., & Halazonetis, T. D. (2000). Chk2/hCds1 functions as a DNA damage checkpoint in G1 by stabilizing p53. Genes Dev, 14(3), 278-288.

Clarke, M. J. (2003). Ruthenium metallopharmaceuticals. Coord Chem Rev, 236(1-2), 209-233.

Clarke, M. J., Zhu, F., & Frasca, D. R. (1999). Non-platinum chemotherapeutic metallopharmaceuticals. Chem Rev, 99(9), 2511-2534.

Davia, K., King, D., Hong, Y., & Swavey, S. (2008). A porphyrin-ruthenium photosensitizer as a potential photodynamic therapy agent. Inorg Chem Commun, 11(5), 584-586.

Davies, M. J. (2003). Singlet oxygen-mediated damage to proteins and its consequences. Biochem Biophys Res Commun, 305(3), 761-770.

Denstman, S. C., Dillehay, L. E., & Williams, J. R. (1986). Ehanced susceptibility to HPD- sensitized phototoxicity and correlated resistance to trypsin detachment in SV-40 transformed IMR-90 cells. Photochem Photobiol, 43(2), 145-147.

DeRosa, M. C., & Crutchley, R. J. (2002). Photosensitized singlet oxygen and its applications. Coord Chem Rev, 233-234, 351-371.

Detty, M. R., Gibson, S. L., & Wagner, S. J. (2004). Current clinical and preclinical photosensitizers for use in photodynamic therapy. J Med Chem, 47(16), 3897-3915.

Dougherty, T. J. (2002). An update on photodynamic therapy applications. J Clin Laser Med Surg, 20(1), 3-7.

Dougherty, T. J., Gomer, C. J., Henderson, B. W., Jori, J., Kessel, D., Korbelik, M., Moan, J., & Peng, Q. (1998). Photodynamic therapy. J Natl Cancer Inst, 90, 889-905.

Essers, J., Theil, A. F., Baldeyron, C., van Cappellen, W. A., Houtsmuller, A. B., Kanaar, R., et al. (2005). Nuclear dynamics of PCNA in DNA replication and repair. Mol. Cell. Biol., 25(21), 9350-9359.

Fanning, E., & Knippers, R. (1992). Structure and function of simian virus 40 large tumor antigen. Annu Rev Biochem, 61, 55-85.

Fiers, W., Contreras, R., Haegeman, G., Rogiers, R., Van de Voorde, A., Van Heuverswyn, H., Volckaert, G., & Ysebaert, M. (1978). Complete nucleotide sequence of SV40 DNA. Nature, 273(5658), 113-120.

147

Finlay, G., Baguley, B., & Wilson, W. (1984). A semiautomated microculture method for investigating growth inhibitory effects of cytotoxic compounds on exponentially growing carcinoma cells. Anal Biochem, 139(2), 272-277.

Fisher, A. M. R., Danenberg, K., Banerjee, D., Bertino, J. R., Danenberg, P., & Gomer, C. J. (1997). Increased photosensitivity in HL60 cells expressing wild-type p53. Photochem Photobiol, 66(2), 265-270.

Fisher, A. M. R., Ferrario, A., Rucker, N., Zhang, S., & Gomer, C. J. (1999). Photodynamic therapy sensitivity is not altered in human tumor cells after abrogation of p53 function. Cancer Res, 59(2), 331-335.

Freedman, D. A., & Levine, A. J. (1998). Nuclear export is required for degradation of endogenous p53 by Mdm2 and human papillomavirus e6. Mol. Cell. Biol., 18(12), 7288- 7293.

Friedman, P. N., Chen, X., Bargonetti, J., & Prives, C. (1993). The p53 protein is an unusually shaped tetramer that binds directly to DNA. Proc Natl Acad Sci USA, 90(8), 3319-3323.

Froelich-Ammon, S. J., Osheroff, N.F. (1995). Topoisomerase poisons: harnessing the dark side of enzyme mechanism. J Biol Chem, 270(37), 21429-21432.

Fruhauf, S., & Zeller, W. J. (1991). New platinum, titanium, and ruthenium complexes with different patterns of DNA damage in rat ovarian tumor cells. Cancer Res, 51(11), 2943- 2948.

Furuse, K., Fukuoka, M., Kato, H., Horai, T., Kubota, K., Kodama, N., et al. (1993). A prospective phase II study on photodynamic therapy with photofrin II for centrally located early-stage lung cancer. The Japan Lung Cancer Photodynamic Therapy Study Group. J Clin Oncol, 11(10), 1852-1857.

Gallardo, D., Drazan, K. E., & McBride, W. H. (1996). Adenovirus-based transfer of wild-type p53 gene increases ovarian tumor radiosensitivity. Cancer Res, 56(21), 4891-4893.

Gallori, E., Vettori, C., Alessio, E., Vilchez, F. G., Vilaplana, R., Orioli, P., et al. (2000). DNA as a possible target for antitumor Ruthenium(III) complexes: a spectroscopic and molecular biology study of the interactions of two representative antineoplastic Ruthenium(III) complexes with DNA. Arch Biochem Biophys, 376(1), 156-162.

Gao, H. L., Huang, K. C., Yamasaki, E. F., Chan, K. K., Chohan, L., Snapka, R. M. (1999). XK469, a selective topoisomerase IIβ poison. Proc Natl Acad Sci USA, 96(21), 12168- 12173.

Gao, J. P., Lanks, K. W., Rosen, M., & Lai, B.T. (1997). Mechanism of action and spectrum of cell types susceptible to doxorubicin photochemotherapy. Cancer Chemother Pharmacol, 40(2), 138-142.

148

Gaullier, J.-M., Rodal, S. K., Moan, J., & Berg, K. (1996). Combined treatment of photodynamic therapy and the topoisomerase I inhibitor camptothecin in growing V79 and NHIK 3025 cells. Proc. SPIE, 2924,243-251

Gershey, E. L. (1979). Simian virus 40-host cell interaction during lytic infection. J. Virol, 30(1), 76-83.

Gianferrara, T., Bratsos, I., & Alessio, E. (2009). A categorization of metal anticancer compounds based on their mode of action. Dalton Trans, 37, 7588-7598.

Goldman, R. D., Gruenbaum, Y., Moir, R. D., Shumaker, D. K., & Spann, T. P. (2002). Nuclear lamins: building blocks of nuclear architecture. Genes Dev, 16(5), 533-547.

Gostissa, M., Hengstermann, A., Fogal, V., Sandy, P., Schwarz, S. E., Scheffner, M., & Del Sal, G. (1999). Activation of p53 by conjugation to the ubiquitin-like protein SUMO-1. EMBO J, 18(22), 6462-6471.

Grebeňová, D., Halada, P., Stulík, J., Havlíček, V., Hrkal, Z. (2000). Protein changes in HL60 leukemia cells associated with 5-aminolevulinic acid-based photodynamic therapy. Early effects on endoplasmic reticulum chaperones. Photochem Photobiol. 72(1),16-22.

Grosjean, P., Savary, J. F., Wagnières, G., Mizeret, J., Woodtli, A., Theumann, J. F., Fontolliet, C., Van Den Bergh, H., &Monnier, P. (1996). Tetra(m-hydroxyphenyl)chlorin clinical photodynamic therapy of early bronchial and oesophageal cancers. Lasers Med Sci, 11(4), 227-235.

Gruenbaum, Y., Margalit, A., Goldman, R. D., Shumaker, D. K., & Wilson, K. L. (2005). The nuclear lamina comes of age. Nat Rev Mol Cell Biol, 6(1), 21-31.

Hainaut, P., & Hollstein, M. (2000). p53 and human cancer: the first ten thousand mutations. Adv Cancer Res, 77, 81-137.

Halliwell, B., & Whiteman, M. (2004). Measuring reactive species and oxidative damage in vivo and in cell culture: how should you do it and what do the results mean? Br J Pharmacol, 142(2), 231-255.

Hambley, T. W. (1997). The influence of structure on the activity and toxicity of Pt anti-cancer drugs. Coord Chem Rev, 166, 181-223.

Haupt, Y., Maya, R., Kazaz, A., & Oren, M. (1997). Mdm2 promotes the rapid degradation of p53. Nature, 387(6630), 296-299.

Henderson, B. W., Daroqui, C., Tracy, E., Vaughan, L. A., Loewen, G. M., Cooper, M. T., & Baumann, H. (2007). Cross-linking of signal transducer and activator of transcription 3-a molecular marker for the photodynamic reaction in cells and tumors. Clin Cancer Res, 13, 3156-3163.

Henderson, B. W., & Dougherty, T. J. (1992). How does photodynamic therapy work? Photochem Photobiol, 55(1), 145-157.

149

Herman, L., Ghosh, S., Defrancq, E., & Mesmaekera, A. K.-D. (2008). Ru(II) complexes and light: molecular tools for biomolecules. J Phys Org Chem, 21(7-8), 670-681.

Hirt, B. (1967). Selective extraction of polyoma DNA from infected mouse cell cultures. J Mol Biol, 26(2), 365-369.

Holder, A. A., Zigler, D. F., Tarrago-Trani, M. T., Storrie, B., & Brewer, K. J. (2007). Photobiological Impact of [{(bpy)2Ru(dpp)}2RhCl2]Cl5 and [{(bpy)2Os(dpp)}2RhCl2]Cl5 [bpy = 2,2‘-Bipyridine; dpp = 2,3-Bis(2-pyridyl)pyrazine] on Vero Cells. Inorg Chem, 46(12), 4760-4762.

Hollstein, M., Sidransky, D., Vogelstein, B., & Harris, C. (1991). p53 mutations in human cancers. Science, 253(5015), 49-53.

Huang, H., Li, C. Y., & Little, J. B. (1996). Abrogation of P53 function by transfection of HPV16 E6 gene does not enhance resistance of human tumour cells to ionizing radiation. Int J Radiat Biol, 70, 151-160.

Husain, D., Miller, J. W., Michaud, N., Connolly, E., Flotte, T. J., & Gragoudas, E. S. (1996). Intravenous infusion of liposomal benzoporphyrin derivative for photodynamic therapy of experimental choroidal neovascularization. Arch Ophthalmol, 114(8), 978-985.

Hussain, S. P., Hofseth, L. J., & Harris, C. C. (2003). Radical causes of cancer. Nat. Rev. Cancer, 3, 276-285.

1 Ito, T. (1978). Cellular and subcellular mechanisms of photodynamic action: the O2 hypothesis as a driving force in recent research. Photochem. Photobiol, 28, 493-508.

Jamieson, E. R., & Lippard, S. J. (1999). Structure, recognition, and processing of cisplatin-DNA adducts. Chem Rev, 99(9), 2467-2498.

Joerger, A. C., & Fersht, A. R. (2008). Structural biology of the tumor suppressor p53. Annu Rev Biochem, 77(1), 557-582.

Josefsen, L. B., & Boyle, R. W. (2008). Photodynamic Therapy and the development of metal- based photosensitisers. Met Based Drugs, 2008, 1-24.

Kapoor, M., & Lozano, G. (1998). Functional activation of p53 via phosphorylation following DNA damage by UV but not γadiation. Proc Natl Acad Sci USA, 95(6), 2834-2837.

Karidi, K., Garoufis, A., Tsipis, A., Hadjiliadis, N., den Dulk, H., & Reedijk, J. (2005). Synthesis, characterization, in vitro antitumor activity, DNA-binding properties and electronic + structure (DFT) of the new complex cis-(Cl,Cl)[RuIICl2(NO )(terpy)]Cl. Dalton Trans, 7, 1176-1187.

Kim, H.-R. C., Luo, Y., Li, G., & Kessel, D. (1999). Enhanced apoptotic response to photodynamic therapy after bcl-2 transfection. Cancer Res, 59, 3429-3432.

Krishna, T. S. R., Kong, X.-P., Gary, S., Burgers, P. M., & Kuriyan, J. (1994). Crystal structure of the eukaryotic DNA polymerase processivity factor PCNA. Cell, 79(7), 1233-1243. 150

Kubbutat, M. H. G., Jones, S. N., & Vousden, K. H. (1997). Regulation of p53 stability by Mdm2. Nature, 387(6630), 299-303.

Kusuzaki, K., Murata, H., Matsubara, T., Satonaka, H., Wakabayashi, T., Matsumine, A., & Uchida, A. (2007). Acridine orange could be an innovative anticancer agent under photon energy. In Vivo, 21, 205-214.

Ladner, D. P., Steiner, R. A., Allemann, J., Haller, U., & Walt, H. (2001). Photodynamic diagnosis of breast tumours after oral application of aminolevulinic acid. Br J Cancer, 84(1), 33-37.

Lambert, P. F., Kashanchi, F., Radonovich, M. F., Shiekhattar, R., & Brady, J. N. (1998). Phosphorylation of p53 Serine 15 Increases Interaction with CBP. J Biol Chem, 273(49), 33048-33053.

Laustriat, G. (1986 ). Molecular mechanisms of photosensitization. Biochimie, 68(6), 771-778.

Lavie, G., Kaplinsky, C., Toren, A., Aizman, I., Meruelo, D., Mazur, Y., & Mandel, M. (1999). A photodynamic pathway to apoptosis and necrosis induced by dimethyl tetrahydroxyhelianthrone and hypericin in leukaemic cells: possible relevance to photodynamic therapy. Br J Cancer, 79(3/4), 423-432.

Lebwohl, D., & Canetta, R. (1998). Clinical development of platinum complexes in cancer therapy: an historical perspective and an update. Eur J Cancer, 34(10), 1522-1534.

Lee, H., Ho, A., & Teo, S. (2006). p53 Status does not affect photodynamic cell killing induced by hypericin. Cancer Chemother Pharmacol, 58(1), 91-98.

Lee, T., Baron, E., & Foster, T. (2008). Monitoring Pc 4 photodynamic therapy in clinical trials of cutaneous T-cell lymphoma using noninvasive spectroscopy. J Biomed Opt, 13(3), 030507.

Lehman, J. M., Laffin, J., & Friedrich, T. D. (2000). Simian virus 40 induces multiple s phases with the majority of viral DNA replication in the g2 and second s phase in cv-1 cells. Exp Cell Res, 258(1), 215-222.

Li, J. J., & Kelly, T. J. (1984). Simian virus 40 DNA replication in vitro. Proc. Natl. Acad. Sci. USA, 81(22), 6973-6977.

Li, L., Ljungman, M., & Dixon, J. E. (2000). The human Cdc14 phosphatases interact with and dephosphorylate the tumor suppressor protein p53. J Biol Chem, 275(4), 2410-2414.

Lim, D. S., Bae, S. M., Kwak, S. Y., Park, E. K., Kim, J. K., Han, S. J., Oh, C. H., Lee, C. H., Lee, W. Y., &Ahnet al, W.S. (2006). Adenovirus-mediated p53 treatment enhances photodynamic antitumor response. Hum Gene Ther, 17(3), 347-352.

Lipson, R. L., & Baldes, E. J. (1960). The photodynamic properties of a particular hematoporphyrin derivative. Arch Dermatol, 82(4), 508-516.

151

Lipson, R. L., Baldes, E. J., & Gray, M. J. (1967). Hematoporphyrin derivative for detection and management of cancer. Cancer, 20(12), 2255-2257.

Liu, L., Scolnick, D. M., Trievel, R. C., Zhang, H. B., Marmorstein, R., Halazonetis, T. D., & Berger, S. L. (1999). p53 sites acetylated in vitro by pcaf and p300 are acetylated in vivo in response to DNA damage. Mol. Cell. Biol., 19(2), 1202-1209.

Liu, L. F., Desai, S. D., Li, T. K., Mao, Y., Sun, M., & Sim, S. P. (2000). Mechanism of action of camptothecin. Ann N Y Acad Sci, 922, 1-10.

Liu, Y., Hammitt, R., Lutterman, D. A., Joyce, L. E., Thummel, R. P., & Turro, C. (2009). Ru(II) 1 complexes of new tridentate ligands: unexpected high yield of sensitized O2. Inorg Chem, 48(1), 375-385.

Liu, Y., Hammitt, R., Lutterman, D. A., Thummel, R. P., & Turro, C. (2007). Marked differences in light-switch behavior of Ru(II) complexes possessing a tridentate DNA intercalating ligand. Inorg Chem, 46(15), 6011-6021.

Liu, Y., & Kulesz-Martin, M. (2001). p53 protein at the hub of cellular DNA damage response pathways through sequence-specific and non-sequence-specific DNA binding. Carcinogenesis, 22(6), 851-860.

Liu, Y., Turner, D. B., Singh, T. N., Angeles-Boza, A. M., Chouai, A., Dunbar, K. R., & Turro, C. (2008). Ultrafast ligand exchange: detection of a pentacoordinate Ru(II) intermediate and product formation. J Am Chem Soc, 131(1), 26-27.

Liu, Y. J., Wang, N., Mei, W. J., Chen, F., He, L. X., Jian, L. Q., Wang, R. J ., & Wu, F. H. (2007). Photoinduced cleavage and DNA-binding of the Ruthenium(II) polypyridyl complex [Ru(phen)2(ipbd)](ClO4)2. Trans Met Chem, 32(3), 332-337.

Lo, K. (2007). Luminescent transition metal complexes as biological labels and probes. Struct Bond, 123, 205-245.

Lowe, S., Bodis, S., McClatchey, A., Remington, L., Ruley, H., Fisher, D., Housman, D. E., & Jacks, T. (1994). p53 status and the efficacy of cancer therapy in vivo. Science, 266(5186), 807-810.

Lu, H., Fisher, R. P., Bailey, P., & Levine, A. J. (1997). The CDK7-cycH-p36 complex of transcription factor IIH phosphorylates p53, enhancing its sequence-specific DNA binding activity in vitro. Mol Cell Biol, 17(10), 5923-5934.

Makino, F., & Okada, S. (1975). Effects of ionizing radiations on DNA replication in cultured mammalian cells. Radiat Res, 62(1), 37-51.

Matheson, I. B., Etheridge, R. D., Kratowich, N. R., & Lee, J. (1975). The quenching of singlet oxygen by amino acids and proteins. Photochem. Photobiol, 21, 165-171.

Metcalfe, C., & Thomas, J. A. (2003). Kinetically inert transition metal complexes that reversibly bind to DNA. Chem. Soc. Rev., 2003, 32, 215-224.

152

Meyer, T. J. (1986). Photochemistry of metal coordination complexes: metal to ligand charge transfer excited states. Pure Appl. Chem, 58( 9), 1193-1206.

Miao, H. Y., Seiler, J. A., & Burhans, W. C. (2003). Regulation of cellular and SV40 virus origins of replication by chk1-dependent intrinsic and uvc radiation-induced checkpoints. Journal of Biological Chemistry, 278(6), 4295-4304.

Moldovan, G.L., Pfander, B., & Jentsch, S. (2007). PCNA, the maestro of the replication fork. Cell, 129(4), 665-679.

Mongelli, M. T., Heinecke, J., Mayfield, S., Okyere, B., Winkel, B. S. J., & Brewer, K. J. (2006). Variation of DNA photocleavage efficiency for [(TL)2Ru(dpp)]Cl2 complexes where TL = 2,2'-bipyridine, 1,10-phenanthroline, or 4,7-diphenyl-1,10-phenanthroline. J Inorg Biochem, 100(12), 1983-1987.

Montaner, B., O'Donovan, P., Reelfs, O., Perrett, C., Zhang, X., Xu, Y. Z., Ren, X., Macpherson, P., Frith, D., &Karran, P. (2007). Reactive oxygen-mediated damage to a human DNA replication and repair protein. EMBO reports, 8(11), 1074-1079.

Mosmann, T. (1983). Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J Immunol Methods, 65(1-2), 55-63.

Munshi, A., Hobbs, M., & Meyn, R. E. (2005). Clonogenic cell survival assay. Methods Mol Med, 110, 21-28.

Nagaich, A. K., Zhurkin, V. B., Sakamoto, H., Gorin, A. A., Clore, G. M., Gronenborn, A. M., Appella, E., & Harrington, R. E. (1997). Architectural accommodation in the complex of four p53 DNA binding domain peptides with the p21/waf1/cip1 DNA response element. J Bio Chem, 272(23), 14830-14841.

Naryzhny, S., Desouza, L., Siu, K., & Lee, H. (2006). Characterization of the human proliferating cell nuclear antigen physico-chemical properties:aspects of double trimer stability. Biochem.Cell Biol, 84, 669-676.

Naryzhny, S. N., Zhao, H., & Lee, H. (2005). Proliferating cell nuclear antigen (PCNA) may function as a double homotrimer complex in the mammalian cell. J Bio Chem, 280(14), 13888-13894.

O'Donnell, M., & Kuriyan, J. (2006). Clamp loaders and replication initiation. Curr Opin Struct Biol, 16(1), 35-41.

Ochsner, M. (1997). Photophysical and photobiological processes in the photodynamic therapy of tumours. J Photochem Photobiol B, 39(1), 1-18.

Okubo, E., Lehman, J. M., & Friedrich, T. D. (2003). Negative regulation of mitotic promoting factor by the checkpoint kinase chk1 in simian virus 40 lytic infection. J. Virol, 77(2), 1257-1267.

153

Oleinick, N. L., Antunez, A. R., Clay, M. E., Rihter, B. D., & Kenney, M. E. (1993). New phthalocyanine photosensitizers for photodynamic therapy. Photochemistry and Photobiology. Photochem Photobiol, 57(2), 242-247.

Oleinick, N. L., & Evans, H. H. (1998). The photobiology of photodynamic therapy: cellular targets and mechanisms. Radia Res, 150(5), S146-S156.

Osheroff, N. (1989). Biochemical basis for the interactions of type I and type II topoisomerases with DNA. Pharmacol Ther, 41(1-2), 223-241.

Ozer, H. L., Slater, M. L., Dermody, J. J., & Mandel, M. (1981). Replication of simian virus 40 DNA in normal human fibroblasts and in fibroblasts from xeroderma pigmentosum. J Virol, 39(2), 481-489

Painter, R. B., & Young, B. R. (1975). X-Ray-induced inhibition of DNA synthesis in Chinese Hamster Ovary, human HeLa, and mouse L Cells. Radiat Res, 64(3), 648-656.

Paunesku, T., Mittal, S., Protić, M., Oryhon, J., Korolev, S., Joachimiak, A., & Woloschak, G. E. (2001). Proliferating cell nuclear antigen(PCNA): ringmaster of the genome. Int J Radiat Biol, 77, 1007-1021.

Peng, Q., Berg, K., Moan, J., Kongshaug, M., & Nesland, J. M. (1997). 5-Aminolevulinic Acid- based photodynamic therapy:principles and experimental research. Photochem Photobiol, 65 (2), 235-251.

Perez, R. P. (1998). Cellular and molecular determinants of cisplatin resistance. Eur J Cancer, 34(10), 1535-1542.

Pickart, C. M. (2004). Back to the future with ubiquitin. Cell, 116, 181-190.

Pierard, F., & Kirsch-De Mesmaeker, A. (2006). Bifunctional transition metal complexes as nucleic acid photoprobes and photoreagents. Inorg Chem Commun, 9(1), 111-126.

Razum, N. J., Snyder, A. B., & Doiron, D. R. (1996). SnET2: clinical update. Proc. SPIE, 2675, 43-46

Reedijk, J. (2008). Metal-ligand exchange kinetics in platinum and ruthenium complexes. Platinum Metals Rev, 52(1), 2-11.

Rodriguez, M., Desterro, J., Lain, S., Midgley, C., Lane, D., & Hay, R. (1999). SUMO-1 modification activates the transcriptional response of p53. EMBO J, 18(22), 6455-6461.

Rosenber.B, Vancamp, L., & Krigas, T. (1965). Inhibition of cell division in Escherichia coli by electrolysis products from a platinum electrode. Nature, 205(4972), 698-699.

Rosenberg, B., Vancamp, L., Trosko, J. E., & Mansour, V. H. (1969). Platinum compounds: a new class of potent antitumour agents. Nature, 222(5191), 385-386.

154

Rosenstein, B. S., Subramanian, D., & Muller, M. T. (1997). The involvement of topoisomerase I in the induction of DNA-protein crosslinks and DNA single-strand breaks in cells of ultraviolet-irradiated human and frog cell lines. Radiat Res, 148(6), 575-579.

Rothenberg, M. L. (1997). Topoisomerase I inhibitors: review and update. Ann Oncol, 8, 837-855.

Sakaguchi, K., Herrera, J. E., Saito, S., Miki, T., Bustin, M., Vassilev, A., Anderson, C. W., & Appella, E.(1998). DNA damage activates p53 through a phosphorylation-acetylation cascade. Genes Dev, 12(18), 2831-2841.

Sakaguchi, K., Saito, S., Higashimoto, Y., Roy, S., Anderson, C. W., & Appella, E. (2000). Damage-mediated phosphorylation of human p53 threonine 18 through a cascade mediated by a Casein 1-like kinase. J Bio Chem, 275(13), 9278-9283.

Sakaguchi, K., Sakamoto, H., Lewis, M. S., Anderson, C. W., Erickson, J. W., Appella, E., & Xie, D. (1997). Phosphorylation of serine 392 stabilizes the tetramer formation of tumor suppressor protein p53. Biochemistry, 36(33), 10117-10124.

Schmitt, F., Govindaswamy, P., Süss-Fink, G., Ang, W. H., Dyson, P. J., Juillerat-Jeanneret, L., & Therrien, B. (2008). Ruthenium porphyrin compounds for photodynamic therapy of cancer. J Med Chem, 51(6), 1811-1816.

Sclafani, R. A., Fletcher, R. J., & Chen, X. S. (2004). Two heads are better than one: regulation of DNA replication by hexameric helicases. Genes Dev, 18(17), 2039-2045.

Seiler, J. A., Conti, C., Syed, A., Aladjem, M. I., & Pommier, Y. (2007). The intra-S-phase checkpoint affects both DNA replication initiation and elongation: single-cell and -DNA fiber analyses. Mol. Cell. Biol, 27(16), 5806-5818.

Sharman, W. M., Allen, C. M., & van Lier, J. E. (1999). Photodynamic therapeutics: basic principles and clinical applications. Drug Discov Today, 4(11), 507-517.

Shen, H.-R., Spikes, J. D., Kopeckov, P., & Kopecek, J. (1996). Photodynamic crosslinking of proteins II. Photocrosslinking of a model protein-ribonuclease A. J Photochem Photobiol B, 35(3), 213-219.

Shin, C.G., Strayer, J. M., Wani, M. A., & Snapka, R. M. (1990). Rapid evaluation of topoisomerase inhibitors: Caffeine inhibition of topoisomerases in vivo. Teratog Carcinog Mutagen, 10(1), 41-52.

Sibata, C. H., Colussi, V. C., Oleinick, N. L., & Kinsella, T. J. (2000). Photodynamic therapy: a new concept in medical treatment. Braz J Med Biol Res, 33, 869-880.

2+ Singh, T. N., & Turro, C. (2004). Photoinitiated DNA binding by cis-[Ru(bpy)2(NH3)2] . Inorgc Chem, 43(23), 7260-7262.

Snapka, R. M. (1986). Topoisomerase inhibitors can selectively interfere with different stages of simian virus 40 DNA replication. Mol Cell Biol., 6(12), 4221-4227.

155

Snapka, R. M. (1996). Analysis of SV40 DNA replication intermediates:the SV40 replicon model for analysis of anticancer drugs (pp. 16-22). San Diego: Academic Press, Inc.

Snapka, R. M., & Permana, P. A. (1993). SV40 DNA replication intermediates: Analysis of drugs which target mammalian DNA replication. BioEssays, 15(2), 121-127.

Stordal, B., Pavlakis, N., & Davey, R. (2007). A Systematic review of platinum and taxane resistance from bench to clinic: an inverse relationship. Cancer Treat Rev, 33(8), 688-703.

Taber, S. W., Fingar, V. H., Coots, C. T., & Wieman, T. J. (1998). Photodynamic therapy using mono-L-aspartyl chlorin e6 (Npe6) for the treatment of cutaneous disease: a Phase I clinical study. Clin Cancer Res, 4, 2741-2746.

Takenaka, I., Morin, F., Seizinger, B. R., & Kley, N. (1995). Regulation of the sequence-specific DNA binding function of p53 by protein kinase c and protein phosphatases. J Bio Chem, 270(10), 5405-5411.

Taylor, C. R., Kwangsukstith, C., Wimberly, J., Kollias, N., & Anderson, R. R. (1999). Turbo- PUVA: dihydroxyacetone-enhanced photochemotherapy for psoriasis: a pilot study. Arch Dermatol, 135(5), 540-544.

Teiten, M. H., Bezdetnaya, L., Merlin, J. L., Bour-Dill, C., Pauly, M. E., Dicato, M., et al. (2001). Effect of meta-tetra(hydroxyphenyl)chlorin (mTHPC)-mediated photodynamic therapy on sensitive and multidrug-resistant human breast cancer cells. J Photochem Photobiol B, 62(3), 146-152.

Tibbetts, R. S., Brumbaugh, K. M., Williams, J. M., Sarkaria, J. N., Cliby, W. A., Shieh, S.Y. Taya, Y., Prives, C., & Abraham, R. T. (1999). A role for ATR in the DNA damage-induced phosphorylation of p53. Genes Dev, 13(2), 152-157.

Triesscheijn, M., Baas, P., Schellens, J. H. M., & Stewart, F. A. (2006). Photodynamic therapy in oncology. Oncologist, 11(9), 1034-1044.

Ulrich, H. D. (2006). Deubiquitinating PCNA: a downside to DNA damage tolerance. Nat Cell Biol, 8(4), 303-305.

Vogelstein, B., Lane, D., & Levine, A. J. (2000). Surfing the p53 network. Nature, 408(6810), 307-310.

Vousden, K. H. (2002). Activation of the p53 tumor suppressor protein. Biochim. Biophys. Acta, 1602, 47-59.

Wainwright, M. (1998). Photodynamic antimicrobial chemotherapy (PACT). J Antimicrob Chemother, 42, 13-28.

Wang, J. C. (2002). Cellular roles of DNA topoisomerases: a molecular perspective. Nat Rev Mol Cell Biol, 3, 430-440.

156

Wang, P., Reed, M., Wang, Y., Mayr, G., Stenger, J. E., Anderson, M. E., Schwedes, J. F., & Tegtmeyer, P. (1994). p53 domains: structure, oligomerization, and transformation. Mol. Cell. Biol, 14(8), 5182-5191.

Wang, Y. V., Leblanc, M., Wade, M., Jochemsen, A. G., & Wahl, G. M. (2009). Increased radioresistance and accelerated b cell lymphomas in mice with Mdmx mutations that prevent modifications by DNA-damage-activated kinases. Cancer Cell, 16(1), 33-43.

Watanabe, I. (1974). Radiation effects on DNA chain growth in mammalian cells. Radiat Res, 58(3), 541-556.

Waterman, M. J. F., Stavridi, E. S., Waterman, J. L. F., & Halazonetis, T. D. (1998). ATM- dependent activation of p53 involves dephosphorylation and association with 14-3-3 proteins. Nat Genet, 19(2), 175-178.

Watts, F. Z. (2006). Sumoylation of PCNA: Wrestling with recombination at stalled replication forks. DNA Repair, 5(3), 399-403.

Weishaupt, K. R., Gomer, C. J., & Dougherty, T. J. (1976). Identification of singlet oxygen as the cytotoxic agent in photo-inactivation of a murine tumor. Cancer Res, 36, 2326-2329.

Wenz, F., Azzam, E. I., & Little, J. B. (1998). The response of proliferating cell nuclear antigen to ionizing radiation in human lymphoblastoid cell lines is dependent on p53. Radiat Res, 49(1), 32-40.

Whibley, C., Pharoah, P. D. P., & Hollstein, M. (2009). p53 polymorphisms: cancer implications. Nat Rev Cancer, 9(2), 95-107.

Wilson, B. C., & Patterson, M. S. (2008). The physics, biophysics and technology of photodynamic therapy. Phys Med Biol, 53(9), R61-R109.

Wong, E., & Giandomenico, C. M. (1999). Current status of platinum-based antitumor drugs. Chem Rev, 99(9), 2451-2466.

Woodburn, K. W., Fan, Q., Kessel, D., Luo, Y., & Young, S. W. (1998). Photodynamic therapy of B16F10 murine melanoma with Lutetium texaphyrin. J Invest Dermatol, 110(5), 746- 751.

Wright, A., Bubb, W. A., Hawkins, C. L., & Davies, M. J. (2002). Singlet oxygen-mediated protein oxidation: evidence for the formation of reactive side chain peroxides on tyrosine Residues. Photochem Photobiol, 76(1), 35-46.

Xue, L. Y., Chiu, S. M., & Oleinick, N. L. (2001). Photochemical destruction of the Bcl-2 oncoprotein during photodynamic therapy with the phthalocyanine photosensitizer Pc 4. Oncogene, 20(26), 3420-3427.

Yang, Y., Kitagaki, J., Dai, R. M., Tsai, Y. C., Lorick, K. L., Ludwig, R. L., Pierre, S. A., Jensen, J. P., Davydov, I.V., Oberoi, P., Li, C. C, Kenten, J. H., Beutler, J. A., Vousden, K. H., & Weissman, A.M. (2007). Inhibitors of ubiquitin-activating enzyme (E1), a new class of potential cancer therapeutics. Cancer Res, 67(19), 9472-9481. 157

Zambetti, G. P., Bargonetti, J., Walker, K., Prives, C., & Levine, A. J. (1992). Wild-type p53 mediates positive regulation of gene expression through a specific DNA sequence element. Genes Dev, 6(7), 1143-1152.

Zawacka-Pankau, J., Krachulec, J., Grulkowski, I., Bielawski, K. P., & Selivanova, G. (2008). The p53-mediated cytotoxicity of photodynamic therapy of cancer: Recent advances. Toxicol Appl Pharmacol, 232(3), 487-497.

Zhang, C. X., & Lippard, S. J. (2003). New metal complexes as potential therapeutics. Curr Opin Chem Biol, 7(4), 481-489.

Zhang, W. G., Li, X. W., Ma, L. P., Wang, S. W., Yang, H. Y., & Zhang, Z. Y. (1999). Wild-type p53 protein potentiates phototoxicity of 2-BA-2-DMHA in HT29 cells expressing endogenous mutant p53. Cancer Lett, 138(1-2), 189-195.

Zhao, R., Hammitt, R., Thummel, R. P., Liu, Y., Turro, C., Snapka, R. M. (2009). Nuclear targets of photodynamic tridentate ruthenium complexes. Dalton Trans, DOI: 10.1039/b913959a

Zouzias, D., Bertelsen, A. H., & Rush, M. G. (1981). DNA mediated gene transfer using Simian virus 40 or Polyoma virus morphological transformation as selective markers. Arch Virol, 70(3), 291-295.

158