Radiation induced biomarkers of individual

sensitivity to radiation therapy

PhD Thesis

by

Sara Skiöld

Centre for Radiation Protection Research

Department of Molecular Biosciences,

The Wenner-Gren Institute, Stockholm University, Sweden

Stockholm, 2014

Cover illustrations by Eric Sageby

© Sara Skiöld

ISBN 978-91-7447-824-2

All previously published papers are reproduced by the permission of the

publishers.

Printed in Sweden by Universitetsservice AB, Stockholm 2014

Distributor: Stockholm University Library

Till mina älskade pojkar, Vilgot & Ale

Min man Peter

Familjen

Till minne av Lasse

“True wisdom comes to

each of us when we realize

how little we understand

about life, ourselves,

and the world around us.”

Socrates 469 BC–399 BC

Abstract One third of the population will develop a cancer before the age of 75 and for women in

Sweden breast cancer is the most common form of cancer [1]. Approximately 50 % of solid

cancers are treated with radiation therapy [2]. The dose used in radiation therapy is adjusted to

the most sensitive individuals so that not more than 5% of the patients will have severe

adverse healthy tissue effects [3, 4]. As a consequence, the majority of the patients will

receive a suboptimal dose, as they would have tolerated a higher total dose and received a

better tumor control, without getting severe side effects [3, 5]. Thus, if tumor therapy could be

individualized based on radiation sensitivity, more patients would be cured and the most

severe adverse reactions could be avoided [5].

At present the mechanisms behind radiation sensitivity that leads to acute or late adverse side

effects of the healthy tissue in response to radiation therapy are not known as well as the

percentage of the population affected [6]. These uncertainties are also considered as a

problem in radiation protection practices where the dose limits applied, aim to provide a

robust protection for the whole population.

One of the main concerns with radiation exposure is the possible cancer risk, which it thought

to increase linearly with increasing doses. However the epidemiological data for the risk

below 50-100 mGy are too limited to provide evidence for the hypotheses that the risk is

linear also in the low dose region. This lack of resolution in the low dose region is due to the

spontaneous high cancer incidence, which will mask the small addition of radiation induced

cancers after low dose exposures. Therefore the real effects of low doses of ionizing radiation

(IR) are not known and neither the fraction of the population that may be at higher risk due to

individual sensitivity. This underlines the importance of research in the low dose area.

The long term aim of this thesis is to develop diagnostic tools to assess the individual

radiation sensitivity of breast cancer patients and to better understand the mechanisms behind

the radiation sensitivity and radiation effects after low dose exposures. The approach is based

on the hypothesis that biomarkers of individual sensitivity, in terms of acute adverse skin

reactions after breast cancer radiation therapy, can be found in whole blood that has been

stressed by low doses of radiation. These biomarkers should serve the purpose to provide

insights in the mechanisms behind the radiation sensitivity as well as to serve as diagnostic

tools.

To reach this goal a protocol for the analysis of differential protein expression in response to

low dose in vitro irradiated whole blood was developed (paper I). This protocol was then used

to investigate the proteomic profile of radiation sensitive and normo-sensitive patients, using

isotope-coded protein labeled proteomics (ICPL). The results from the ICPL study (paper III)

show that the two patient groups have different protein expression profiles after IR and that

these profiles could be used to identify functional networks that differed between the two

groups studied. The unique protein profiles identified might also be useful as diagnostic tools

for predicting radiation sensitivity but this application needs to be validated. Additionally it

was investigated if the levels of 8-oxo-dG (a biomarker of oxidative stress) in serum after IR

could be used as a biomarker of radiation sensitivity in radiation sensitive and normo-

sensitive patients (paper II). It was found that the relative levels of IR induced 8-oxo-dG in

serum from radiation sensitive patients (RTOG 3–4) differ significantly from patients with no

visual skin reactions (RTOG 0). This indicates that the patients with severe acute skin

reactions differ in their cellular response to ionizing radiation at the level of induction of

oxidative stress and that 8-oxo-dG is a potential biomarker for radiation sensitivity.

It is probable that radiation sensitivity is multifactorial in its origin and that a range of

biomarkers will be needed to provide a diagnostic tool with the needed precision to be used in

the clinic. In this thesis two different approaches to identify biomarkers of radiation

sensitivity have been investigated but further studies are needed to validate if these have the

necessary diagnostic power.

List of publications

This thesis is based on the following publications:

I. Skiöld, S., Becker, S., Hellman, U., Auer, G., Naslund, I., Harms–Ringdahl,

M., Haghdoost, S. (2011)

Low doses of γ-radiation induce consistent protein

expression changes in human leukocytes Int. J. Low Radiation Vol. 8, Nos.

5/6, pp. 374-387. (Reproduced by permission from IJ Low Radiation, 2011, Vol. 8, Nos 5/6, © Inderscience Publishers.)

II. Skiöld S, Näslund I, Brehwens K, Andersson A, Wersall P, Lidbrink E,

Harms-Ringdahl M, Wojcik A, Haghdoost S. (2013) Radiation-induced stress

response in peripheral blood of breast cancer patients differs between

patients with severe acute skin reactions and patients with no side effects to

radiotherapy Mutat Res. 30; 756 (1-2): 152-7.

III. Sara Skiöld, Omid Azimzadeh

, Ingemar Näslund, Peter Wersall, Elisabet

Lidbrink, Soile Tapio, Mats Harms-Ringdahl and Siamak Haghdoost. (2013)

Unique proteomic signature for radiation sensitive patients; a comparative

study between normo-sensitive and radiation sensitive breast cancer

patients Manuscript.

Additional publications, not included in this thesis;

IV. Fotouhi A, Skiöld S, Shakeri-Manesh S, Osterman-Golkar S, Wojcik A,

Jenssen D, Harms-Ringdahl M, Haghdoost S. (2011) Reduction of 8-oxodGTP

in the nucleotide pool by hMTH1 leads to reduction in mutations in the

human lymphoblastoid cell line TK6 exposed to UVA. Mutat Res.1;715 (1-2).

Table of Contents

INTRODUCTION ........................................................................................................ 1

BACKGROUND ......................................................................................................... 3

IONIZING RADIATION ............................................................................................................... 3 REACTIVE OXYGEN SPECIES (ROS) ......................................................................................... 5 DNA DAMAGES, OXIDATIVE STRESS AND REPAIR .................................................................... 8

Base excision repair (BER) ................................................................................................ 8 Sanitization of 8-oxo-dGTP in the nucleotide pool .......................................................... 10 The difference between high and low doses ..................................................................... 11 Radiation induced endogenous stress .............................................................................. 12 Adaptive response and bystander effects ......................................................................... 12

INDIVIDUAL RADIATION SENSITIVITY .................................................................................... 13 Symptoms of acute and late radiation sensitivity in breast cancer patients .................... 14 Radiation therapy treatment of breast cancer .................................................................. 16

ROLE OF INFLAMMATION RESPONSES AND OXIDATIVE STRESS IN INDIVIDUAL RADIATION

SENSITIVITY........................................................................................................................... 17 Studies of radiation sensitivity ......................................................................................... 18 Radiation sensitivity and OMICS (RADIO-OMICS) ........................................................ 19

AIM AND STRATEGY OF THE STUDIES ............................................................... 21

MATERIALS AND METHOD ..................................................................................................... 21 Blood ................................................................................................................................ 21 Patient cohort ................................................................................................................... 21 Radiation facilities ........................................................................................................... 22

PAPER I ................................................................................................................................. 23 AIM AND INTRODUCTION ....................................................................................................... 23

Methods ............................................................................................................................ 23 Result and discussion ....................................................................................................... 23 Main findings in paper I ................................................................................................... 24

PAPER II ................................................................................................................................ 25 AIM AND INTRODUCTION ....................................................................................................... 25

Methods ............................................................................................................................ 25 Result and discussion ....................................................................................................... 26 Main findings in paper II ................................................................................................. 27

PAPER III ............................................................................................................................... 28 INTRODUCTION AND AIM ....................................................................................................... 28

Methods ............................................................................................................................ 28 Result and discussion ....................................................................................................... 30 Main findings in paper III ................................................................................................ 31

CONCLUDING REMARKS AND FUTURE STUDIES ............................................. 32

ONGOING PROJECTS ............................................................................................................... 33 MicroRNA and gene expression studies ........................................................................... 33 Candidate single nucleotide polymorphisms study of breast cancer patients ................. 33

ACKNOWLEDGEMENTS ........................................................................................ 34

REFERENCES ......................................................................................................... 36

Abbreviations 1D gel One dimensional gel electrophoresis

2D-PAGE Two dimensional polyacrylamide gel electrophoresis

8-oxo-dGTP 8-oxo-7,8-dihydro-2’-deoxyguanosine-5’triphosphate

8-oxo-dG 8-oxo-7,8-dihydro-2’-deoxyguanosine

8-oxo-Gua 8-oxo-7,8-dihydroguanine

A Adenine

AP-1 Activator protein 1

APEX1 Apurinic/apyrimidinic endonuclease 1

AP-site Apurinic/ apurmidinic site

AT Ataxia-telangiectasia

ATM Ataxia-telangiectasia mutated gene

ATR ATM and Rad3 related

BD Base damage

BER Base excision repair

BLVRB Biliverdin reductase B

C Cysteine

CDKN1A Cyclin-dependent kinase inhibitor 1A

CT-scan Computed tomography scan

Cs Cesium

DNA Deoxyribonucleic acid

DNA pol λ DNA polymerase lambda

dNTP 2’-deoxyribonucleoside triphosphate

DSB Double strand break

ELISA Enzyme-Linked Immunosorbent Assay

ESI-MS/MS Electrospray ionization with tandem mass spectrometry

ESTRO The European society of therapeutic radiology and oncology

eV Electron Volt

FANC protein Fanconi anaemia

FEN-1 Flap-endonuclease 1

G Guanine

γ-H2AX γ histone 2AX

GPX Glutathione peroxidase

GR Glutathione reductase

GSH Glutathione

GSSG Glutathione disulfide

GWAS Genome wide association study

Gy Gray

H2O Water

H2O2 Hydrogen peroxide

HER2+ Herceptin positive

hMTH1 human MutT Homologue 1

HR Homologous recombination

hTR Telomerase RNA component

ICPL Isotope-coded protein label

ICRP International commission on radiological protection

IFN-γ Interferon gamma

IPA Ingenuity pathway analysis

IL Interleukin

IR Ionizing radiation

LET Linear energy transfer

Lig Ligase

LNT Linear non-threshold hypothesis

Ki67 Proliferation marker

MMR Mismatch repair

MRE11 Meiotic recombination 11

MS Mass spectrometry

MUTYH MutY Homologue

NER Nucleotide excision repair

NHEJ Non-homologous end joining

NIR Nucleotide incision repair

NOX NADPH oxidase

NBS1 Nijmegen Breakage Syndrome 1

O2•- Superoxide radical

OGG1 MutM

OH• Hydroxyl radical

PARK7 Parkinson protein 7

PARP1 Poly(ADP-ribose) polymerase 1

PCNA Proliferating cell nuclear antigen

PRDX2 Peroxiredoxin 2

PRRs Pattern recognition receptors

PRX Peroxiredoxin

RBE Relative biological effectiveness

RNA Ribonucleic acid

ROC Receivers operating characteristic

ROS Reactive oxygen species

RTOG Radiotherapy oncology group

SiRNA Small interfering RNA

SNP Single nucleotide polymorphism

SOD Superoxide dismutase

SSB Single strand break

SSM Strålsäkerhetsmyndigheten (Swedish radiation safety authority)

Sv Sievert

T Thymidine

TGF-β Transforming growth factor beta

TK6 cell line Human B lymphoblastoid cells

TLR Toll-like receptors

TLS Translesion synthesis

TNF-α Tumor necrosis factor alpha

TR Thioredoxin reductase

Trx Thioredoxin

Trx-S-S-Trx Thioredoxin disulphide

WR Radiation weighting factor

WRN Werner Syndrome, RecQ Helicase-Like

XPC Xeroderma pigmentosum complementation group C

XRCC1 X-ray cross-complementing gene 1

1

Introduction It is predicted that approximately one third of the Swedish population will be diagnosed with

cancer before the age of 75 years. For women in Sweden breast cancer is the most common

form of cancer and it accounts for 30 percent of all cancers, and the incidence is increasing

with 1.3 percent every year [1]. Approximately 50 percent of solid cancers are treated with

radiation therapy [2].

One of the challenges in medicine is the search for biomarkers that will permit an

individualized therapy, and that takes in consideration the genetic and life style factors that

may modify the success of the therapy. Radiation therapy exemplifies this problem where the

total dose provided is adjusted to the most sensitive individuals so that not more than 5

percent of the patients will have severe adverse healthy tissue effects. For the same therapy

treatment approximately 20 percent will experience some adverse effects from their radiation

therapy [3, 4]. As a consequence, the majority of the patients will receive a suboptimal dose,

as they would have tolerated a higher total dose and gained a better tumor control, without

getting severe side effects. Furthermore, studies imply that patients with no acute tissue

reactions to therapy have a higher incidence of local recurrence, indicating that all cancer cells

were not eradicated [3, 5]. Thus, if tumor therapy could be individualized based on radiation

sensitivity, more patients would be cured and the most severe adverse reactions could be

avoided [5].

At present the causes of radiation sensitivity leading to acute or late adverse side effects of the

healthy tissue in response to radiation therapy are not known, except for a few genetically

inherited disorders such as; Ataxia telangiectasia (AT), Fanconi anaemia, Nijmegen breakage

syndrome, Werner’s syndrome, Ligase IV syndrome, Seckel syndrome [7], and chromosomal

fragility syndrome [4], which are associated with extreme radiation sensitivity (table 3). These

genetic syndromes are extremely rare and with a minimal contribution to the number of

radiation sensitive patients observed in the clinic [8]. However, based on short and long term

follow up studies of radiation therapy patients [9-12] and from studies of lymphocyte

radiation sensitivity, reviewed in [8], it is known that the degree of radiation sensitivity differs

in the population. It was suggested in one study by Safwat et al. that 80-90 percent of

variability in late normal tissue radiation sensitivity was due to patient-related factors [13]. In

a study by Turesson et al. 70 percent of the patient-to-patient variability could not be

2

explained by 20 treatment- and patient-related factors [11]. The genetic mechanisms behind

this type of radiation sensitivity as well as the magnitude and percentage of the population

affected are not known [6]. These uncertainties are considered as a problem in radiation

protection practices where the dose limits applied aim to provide a robust protection for the

whole population. It is not known either, if radiation sensitive individuals are more prone to

cancer and therefore are overrepresented in the group receiving radiation therapy.

We are all exposed to low doses of ionizing radiation (IR), originating from the decay of

isotopes in bedrock, cosmic radiation, and to a minor part from fall out from nuclear weapon

testing and the Chernobyl nuclear accident. The average effective background dose in Sweden

is 2.2 mSv/year (excluding the contribution of radon) [14]. However a number of individuals

will receive higher doses due to over-sea flights, high residential levels of radon, work related

exposure or medical screenings (table 1). Thus in developing countries the collective dose is

increasing. The dose from medical screenings is estimated to contribute with an increase of

0.2 mSv/year and medical applications add 0.9 mSv/year to the background level [14, 15].

One of the main concerns with radiation exposure is the possible cancer risk. Cancer risk

estimates are based on the linear non-threshold hypothesis (LNT). This hypothesis is mainly

based on epidemiological data from the survivors of the two atomic bomb explosions and

suggests that the cancer risk associated with radiation increases linearly with dose without a

threshold [16, 17]. However there are no epidemiological data that prove the linearity below

100 mGy [18-20], although Brenner et al. (2003) suggests that there is evidence for an

increased cancer risk after protracted exposures of 50-100 mSv [21]. Accordingly these

authors suggest that even a small dose can pose a healthcare problem to society (increased

cancer incidence) if many individuals are exposed [21]. This lack of resolution in the low

dose region is due to the spontaneous high cancer incidence (1/3 of the population will get

diagnosed with cancer before the age of 75), which will mask the small addition of radiation

induced cancers after low dose exposures [1]. The current risk estimate for cancer mortality is

5.5x10-2

Sv-1

, thus a dose of 100 mSv to 1000 subjects would be expected to induce ~5

additional fatal cancer cases in addition to the 300 expected from the other causes [17].

There is also an increasing concern for radiation induced non-cancerous diseases, such as

heart disease, stroke, digestive disorders and respiratory diseases. For the non-cancerous

effects the epidemiological studies only provide solid dose response relations for doses > 1 Sv

3

thus the contribution of IR below doses of 100 mGy are uncertain [17, 22]. In conclusion the

estimates of health effects caused by low doses of IR are not based on scientific facts and the

fraction of the population that may be at higher risk due to individual sensitivity is not known

either. This stresses the importance of research in the low dose area.

The long term aim of this thesis was to develop diagnostic tools to assess the individual

radiation sensitivity of breast cancer patients and to better understand the mechanisms behind

the radiation sensitivity and radiation effects after low dose exposures.

Dose Source

100 mSv Typical dose from an international Space Station mission

Upper limit of full-body spiral computed tomography (CT) scan

2.4 mSv Average background dose per year in Sweden

0.1 mSv Single chest X-ray

Single mammographic screening

One return flight between New York and London

0.005 mSv Normal X-ray at the dentist

Table 1, Sources of exposures and doses, modified from [15, 23, 24].

Background

Ionizing radiation Ionizing radiation represents a quality of radiation that carries energies high enough to cause

ionizations in exposed matter, leading to chemical alterations. IR includes photon-radiation

(γ- and X-ray) and particle radiations (alpha-, beta-particles, neutrons, heavy ions, etc.).

The energies of the different radiation types are measured in electron volt (keV-MeV).

Different types of IR can also be characterized by their linear energy transfer (LET), which is

measured in keV/µm. For example, α-particles have a higher LET than γ-radiation. Thus α-

4

particles have a denser ionization pattern and therefore deposit more energy per µm, causing

more clustered damages along its path than γ-radiation. Gamma-radiation travels deeper into

matter than α-particles and has a more scattered ionization pattern [25]. This thesis will focus

on biological effects associated with low LET photon radiation.

The absorbed dose, with the unit Gray (Gy), is defined as the energy deposited in a defined

mass of material, where one Gy is equal to 1 joule/ kg. However, the same dose can give rise

to different quantities and qualities of damages depending on the radiation quality. To be able

to compare the damages caused by radiation with different LET, the relative biological

effectiveness (RBE) can be used. RBE is calculated by dividing the X-ray or γ-radiation dose

with the dose of another type of radiation needed to produce the same biological effect [17].

For radiation protection purposes it is necessary to estimate the cancer risk caused by

exposure to different types of radiation. For this purpose the equivalent dose, measured in

Sievert (Sv), is calculated. The equivalent dose takes into consideration a radiation weighting

factor (WR) (table 2) approximating the relative cancer risk from different radiation qualities

per unit dose. The equivalent dose is calculated as shown below.

Sv = WR × Gy where WR is the radiation weighting factor and Gy is

the dose.

Radiation type Radiation weighting factor WR

Photons 1

Electrons 1

Protons 2

Αlpha particles and heavy ions 20

Table 2. Radiation weighting factors for different radiation qualities [17].

5

To assess the risks associated with different radiation qualities and organ sensitivities, the

effective dose is calculated. The effective dose considers the dose, the type of radiation and

the risk factor of the exposed organ(s). Thus, the effective dose is the sum of the equivalent

dose for each organ, multiplied with a weighting factor related to the risk for that organ [17].

For example, the average background effective dose of radiation in Sweden estimated by the

Swedish radiation safety authority (SSM), is 2.4 mSv/year for non-smokers and almost 4

mSv/year for smokers, due to the increased lung cancer risk associated with radon exposure

and smoking [15].

Radiation has two principal modes of effects for interaction with cellular targets, direct effects

and indirect effects. Direct effects take place when, for example, a photon, directly ionizes

DNA or other cellular components. Indirect effects are mediated through radiolysis of cellular

water which gives rise to reactive oxygen species (ROS). Reactive oxygen species are radicals

which cause damages to nucleic acid and other cellular components through oxidation. For γ-

radiation approximately 70 percent of the damage in cells is caused by the indirect effects of

radiation [26]-[27].

Reactive oxygen species (ROS) ROS are formed by the indirect effect of irradiation, where 1 Gy of γ-radiation is estimated to

results in approximately 150 000 ROS per cell [25, 28]. ROS include superoxide anion (O2•-),

hydroxyl radical (•OH), and hydrogen peroxide (H2O2).

Nonetheless, ROS are also produced constantly by the cellular metabolism, approximately

2x1010

molecules of O2•- and H2O2 per cell and day are formed during cellular respiration (in

rat cells) [29]. The majority of endogenously produced ROS is derived from the mitochondria

and peroxisomes. This production leads to approximately 104 DNA modifications per cell and

day [29, 30]. Thus ROS are highly reactive and can oxidize cellular components such as

nucleic acid.

Since ROS are continually produced in the cell a number of defense mechanisms (belonging

to the cellular antioxidant capacity) have evolved to reduce the damage caused by ROS

(figure 1). Oxidative stress is defined as the stage when ROS levels exceed the capacity of the

antioxidant defense systems causing increased levels of ROS. During oxidative stress the

6

damage to cellular targets such as nucleic acids in DNA or in the nucleotide pool increase and

give rise to different base modifications, for example 8-oxo-7,8-dihydroguanine (8-oxo-Gua)

(in DNA) or 8-oxo-7,8-dihydro-2’-deoxyguanosine-5’triphosphate (8-oxo-dGTP) (in the

nucleotide pool). All nucleotides in the 2’-deoxyribonucleoside triphosphate (dNTP) pool can

be oxidized by ROS. However, guanine bases are easily oxidized and are considered a good

marker of oxidative stress [30-35]. When 8-oxo-Gua / 8-oxo-dG is removed from the DNA or

the nucleotide pool it is excreted to the extracellular liquids such as cell culture media, blood

serum or urine.

7

Figure 1. Schematic figure of the production of free radicals and cellular defense mechanisms in the

cytoplasm. Superoxide is transformed into hydrogen peroxide by superoxide dismutase (SOD).

Hydrogen peroxide can be transformed into hydroxyl radicals through the Fenton reaction; however

hydrogen peroxide is rapidly converted into water by catalase, peroxiredoxin (PRX) and glutathione

peroxidases (GPX). dGTP oxidized to 8-oxo-dGTP in the nucleotide pool is dephosphorylated by

hMTH1 and 8-oxo-dGMPase to 8-oxo-dG, which is excreted to the extracellular liquid. Antioxidants

can scavenge radicals thus protecting the cells, modified from paper II and [36].

8

DNA damages, oxidative stress and repair

One Gy of γ-radiation causes approximately 20-40 double strand breaks (DSB), 1000 single

strand breaks (SSB), 2000 ionizations in DNA, 100 000 ionizations in the cell nucleus and

150 000 ROS in one mammalian cell [25, 28]. Damages located in close proximity, may

results in multiple damage sites or clustered lesions. These lesions are generally more difficult

to repair than DSB, SSB, and base damages (BD) [25].

Through evolution, all living organisms have developed systems to maintain genomic stability

and to minimize the damaging effects on DNA caused by endogenous or environmental

factors, DNA repair processes are fundamental for the genomic integrity and a brief

description of the major repair pathways are given below.

Double strand breaks are the most dangerous type of DNA damage and can be repaired by

homologous recombination (HR) and non-homologous end-joining (NHEJ), reviewed in [37].

Base damages and single strand breaks in DNA are primarily repaired by base excision repair

(BER), mismatch repair (MMR), nucleotide excision repair (NER) efficiently restoring the

fidelity of the DNA, reviewed in [37]. The nucleotide incision repair (NIR) pathway has been

suggested as a backup pathway for BER [38]. In case of damage to the nucleotide pool,

hMTH1 will dephosphorylate oxidized purines and prevent incorporation in DNA [39] (figure

2).

Base excision repair (BER)

BER is the major repair pathway for oxidative BD and SSB. There are two types of BER,

short patch and long patch repair. Briefly described, in short patch BER, a DNA glycosylase

identifies the damaged base and removes the base, creating an apurinic or apyrimidinic site

(AP-site). A DNA AP endonuclease or DNA AP lyase cleaves the DNA backbone, creating a

single nucleotide gap (a single strand break). XRCC1 is recruited to the site and is thought to

function as a scaffold protein. A DNA polymerase inserts a new nucleotide in the gap and

DNA ligase seals the gap, for review see [40]. In long patch BER, AP endonuclease APEX1,

recognizes and excise the damage and creates a 5’ incision of the AP-site. This recruits DNA

polymerase beta or delta, PCNA, FEN1 and ligase 1 to the site. DNA polymerase beta and

PCNA inserts a number of nucleotides, resulting in a long single stranded DNA flap. This flap

is then cleaved by FEN-1 and the remaining nucleotide ligated by ligase 1, for review see

9

[40]. For SSB the damage can be recognized by PARP1, which recruits XRCC1. PARP1 and

XRCC1 have been suggested to recruit the other BER proteins to the damage site [41, 42].

If 8-oxo-dG is formed in DNA it can be repaired by BER, in two ways. OGG1 can recognize

8-oxo-dG opposite to cytosine (C) and removes the damaged base, resulting in an AP site [43,

44]. The excised 8-oxo-Gua will be excreted from the cell. The AP site is then repaired by the

BER machinery and the DNA sequence is unchanged, reviewed in [43]. However, if

replication takes place before the 8-oxo-dG has been removed, 8-oxo-dG can be paired with

an adenine (A). MUTYH can recognize this mismatch between template 8-oxo-dG and

adenine and removes the adenine [35, 43]. This results in an AP-site opposite to the 8-oxo-dG

[35, 45]. DNA polymerase λ will insert a cytosine opposite to 8-oxo-dG [46]. This

combination of 8-oxo-dG and cytosine can be recognized by OGG1 and the BER machinery

as described above [43]. However if neither of the repair pathways remove the damaged base

it will lead to a G:C to T:A transversion (figure 2).

10

Figure 2. Depicts the formation and repair of 8-oxo-dG (G in red color) and 8-oxo-dGTP in DNA and

the nucleotide pool respectively. Dashed arrows symbolize one round of replication. Blue lines depict

the template strand. The first pathway from the left, describes the possible scenarios of repair when

guanine is oxidized in DNA. Oxidized guanine in DNA can be excised by OGG1 followed by BER

machinery repair, this protects the DNA fidelity. In case OGG1 doesn’t recognizes the mismatch

before replication a G:A mismatch can be formed. MUTYH can recognizes adenine miss-paired with

template 8-oxo-dG and removes it, a C is then inserted opposite the oxidized guanine, which can then

be repaired by OGG1 and the BER machinery, thus restoring the DNA fidelity. If the miss-paired A is

not discovered it will gives rise to a G:C to T:A transversion in the next replication. In the second

pathway from the left, an 8-oxo-dGTP from the nucleotide pool is incorporated opposite an A, during

the next replication 8-oxo-dG can be paired with a C, resulting in a A:T to C:G transversion. In the

third pathway from the left, dGTP is oxidized to 8-oxo-dGTP in the nucleotide pool. The 8-oxo-dGTP

is then dephosphorylated by hMTH1 to 8-oxo-dGMP followed by dephosphorylation of 8-oxo-dGMP

by 8-oxo-dGMPase to 8-oxo-dG which is excreted from the cell, modified from [31, 35, 43, 45].

Sanitization of 8-oxo-dGTP in the nucleotide pool

If 8-oxo-dGTP is formed in the nucleotide pool it can be dephosphorylated by the sanitization

enzyme hMTH1 to 8-oxo-dGMP. The resulting 8-oxo-dGMP is then dephosphorylated to 8-

oxo-dG by 8-oxo-dGMPase. The 8-oxo-dG (8-oxo-7,8-dihydro-2’-deoxyguanosine) is

excreted from the cell to the extracellular liquid [39]. However if hMTH1 doesn’t

11

dephosphorylate 8-oxo-dGTP it can be incorporated opposite to an adenine or cysteine in

DNA. During replication this can give rise to an A:T-C:G transversion (figure 2). MUTYH

doesn’t excise template adenine opposite 8-oxo-guanine in the nucleus, reviewed in [35]. In

the mitochondria MUTYH might remove adenine independent of replication, reviewed in

[35]. The resulting AP-site is repaired by DNA polymerase λ which incorporates a cytosine

opposite to the 8-oxo-dG [46], this base pair combination can be recognized by BER [43] and

results in an A:T – C:G transversion.

The difference between high and low doses

In this thesis a “low” dose is defined as a dose below 100 mGy, and a high dose is defined as

a dose above 1 Gy. A dose between 100 mGy and 1 Gy is considered as an intermediate dose.

High doses can give rise to both deterministic effects and stochastic effects. Deterministic

effects are harmful tissue effects that arise after a high dose due to cell killing which leads to

loss of organ function. Generally there is a threshold dose below which no deterministic effect

can be detected. According to ICRP publication 103, deterministic effects are not induced at

doses below 100 mGy [17]. Stochastic health effects are primarily manifested as induction of

cancer or hereditable effects. The classification of non-cancerous effects such as vascular

diseases, lens opacities or effects on the cognitive function, being stochastic or not, is debated

and is presently a high priority research area [47].

The dose response relation for stochastic effects is considered to be linear without a threshold,

(the LNT hypothesis) thus at low doses only stochastic effects are considered.

Besides the genotoxic effects related to DNA-damage, other response pathways may also be

important in the low dose range; such as oxidative stress responses [32, 33, 48], inflammation

responses [48-50], and cell signaling [48].

In the search for mechanisms involved in health effect of low doses, other endpoints than

DNA damages and mutations have to be analyzed, as these traditionally have not been

sensitive enough to measure the effects caused by low or intermediate doses.

In this thesis the main focus is stress response effects of low dose irradiation, investigating

protein expression and oxidative stress, and how these responses can provide a mechanistic

understanding of individual sensitivity. These two goals have challenged researchers in

12

radiation biology for decades and belong still to the high priority research areas in tumor

therapy and radiation protection research [47].

Radiation induced endogenous stress

It has been shown that low doses of γ-radiation induce a cellular stress response [32, 51]. This

stress response results in an increase of extracellular 8-oxo-dG and intra-cellular hMTH1

protein levels. This increase in 8-oxo-dG (30 fold higher than expected based on the delivered

dose) can’t be explained by the effect of radiation alone [32-34]. Haghdoost et al. proposed

that doses in the mGy range of low LET radiation increase the endogenous production of

ROS in addition to the immediate effect of IR [32, 33]. A similar conclusion was drawn in the

review by Spitz et al. (2004), who states that there is strong support for the fact that

endogenous ROS contributes to the radiation effects after IR [51]. One of the targets for the

increased levels of ROS is the nucleotide pool [33, 34].

The possible function of the radiation induced endogenous production of ROS is not known,

but it has been suggested that ROS (H2O2) could function as a secondary messenger in cells to

respond to extracellular changes [51-53]. The many enzymes controlling the rapid turnover of

H2O2 supports this hypothesis [53]. Sena et al. suggests that mitochondrial ROS is induced in

response to different stressors and functions as a signaling intermediate to help cells adapt to

the stress [52].

Adaptive response and bystander effects

There are two phenomena associated with radiation responses, adaptive response and

bystander effects that are much debated with regard to the effects they might pose in low

doses and dose rates. The definition of an adaptive response is that a low dose of radiation

prior to a challenging dose of radiation will reduce the toxicity or genomic instability caused

by the challenging dose if given alone [54], for review see [55].

The bystander effect postulates that a cell targeted by IR can generate stress signals which

induce damages to the neighboring untargeted cells, thus increasing the damage caused by

lower doses of radiation [56] [57] for review see [55, 58].

13

Both adaptive response and bystander effects are interesting and important aspects of IR

induced damages. Especially since at low dose exposures (below 100 mGy) all cells in a

tissue or cell culture are not hit by irradiation [59]. It was calculated by Bond et al. that at 1

mGy 137

Cs only 20 percent of the exposed cells were hit [59]. However, the mechanism

responsible for adaptive responses and bystander effects is still largely unknown, as well as

their possible impact on the shape of the dose response curve for cancer and non-cancer

diseases in the low dose range.

Individual radiation sensitivity It has long been known that rare cases of extreme sensitivity to irradiation can be found in

human populations and evidences have been presented that the sensitivities are linked to

genetic disorders. The first studies were published in the 1970-ties describing that the genetic

disease Ataxia-telangiectasia was associated with extreme radiation sensitivity [60, 61]. Since

then a large number of studies have investigated the genetic mechanisms behind

predisposition to radiation sensitivity, for reviews see [4, 62, 63-65]. Several single nucleotide

polymorphisms (SNPs) has been suggested to predict acute or late radiation sensitivity but

could not be validated in other cohorts [64]. There is a growing consensus that the genetics

behind radiation sensitivity should be regarded as a complex trait dependent on several

different sequence alterations [66]. Despite the efforts no general genetic mechanism has been

found except for the few rare genetic traits previously mentioned, (table 3).

Oxidative stress has been connected to radiation and radiation sensitivity [7, 34, 67].

However, many other factors could affect the cellular stress levels in humans, for example;

background diseases, diet and lifestyle factors [11, 30]. Most of the genetic syndromes

associated with radiation sensitivity are due to genetic deficiencies in DNA repair pathways

(table 3). Still a few diseases are associated with high levels of oxidative stress: Alzheimer’s

disease, Parkinsson’s disease, and Downs syndrome have also been associated with increased

radiation sensitivity in cell cultures [68-70]. Several autoimmune diseases such as rheumatoid

arthritis, systemic lupus erythematous and scleroderma have been shown to have increased

radiation sensitivity of lymphocytes when analyzed in cell cultures [71]. This indicates that

radiation sensitivity could be connected to oxidative stress and inflammation.

14

Name of disease Protein deficiency DNA repair pathway

Ataxia telangiectasia ATM DSB cell signaling

Fanconi anaemia FANC protein HR

Dyskeratosis congenita Dyskerin/hTR Telomere metabolism

Nijmegen breakage syndrome NBS1 DSB cell signaling

Werner’s syndrome WRN RecQ helicase HR/TLS

Ligase IV syndrome Lig IV NHEJ

Seckel syndrome ATR DSB cell signaling

At-like disorder MRE11 DSB cell signaling

Chromosomal fragility

syndrome

Not investigated High spontaneous chromosome

aberrations.

Table 3. A summary of genetic syndromes associated with radiation sensitivity, modified from [4, 7].

Symptoms of acute and late radiation sensitivity in breast cancer patients

The symptoms of radiation sensitivity in breast cancer patients can either occur as an acute

adverse effect or as a late adverse effect. Acute effects of radiation sensitivity arise during the

treatment or at the end of the treatment and are most often transient and will heal. The effects

are mainly seen in tissues with a rapid cell turnover and are due to cell killing. For example

desquamation and ulceration of the skin is due to cell death. One exception is acute effects

such as erythema and edema that are thought to be caused by inflammation responses and

increased vascularization, reviewed in [72, 73]. The acute adverse effects in skin can be

classified according to the Radiation Therapy Oncology Group, acute radiation morbidity

scoring criteria (RTOG-scale) [74];

RTOG 0; No change over baseline

RTOG 1; Faint or dull erythema, dry desquamation

RTOG 2; Bright erythema, patchy moist desquamation, moderate edema

RTOG 3; Confluent moist desquamation, pitting edema

RTOG 4; Ulceration, hemorrhage, necrosis

15

Late adverse effects arise months to years after the completion of the treatment. Late effects

are generally more persistent than acute effects and don’t heal. These effects are mainly seen

in tissues with low cellular turnover, reviewed in [72, 73]. In skin the main late effects are

telangiectasia, atrophy, and necrosis where the severity can be scored on a scale from 0-5.

Zero is no effect above base line and 5 is death directly associated with irradiation [75].

Another important late effect seen in breast cancer patients is radiation-induced fibrosis, it has

been extensively studied for review see [2].

There is no clear correlation between the risk for a patient to develop acute adverse effects

and late adverse effects [76]. An exception are consequential late effects where the severe

acute effects lead to a breakdown of protecting barriers against mechanical or chemical stress

leading to a late effect [76].

Of growing concern in radiation therapy, correlated with the increased survival time of the

patients, is the risk that radiation will induce secondary cancers in the surrounding healthy

tissue exposed during the therapy. There are numerous studies investigating the possible

induction of secondary cancers from radiation therapy treatment [77-81]. One main

conclusion from these studies is the need for long follow up periods to investigate the

incidence of secondary cancers.

On the other hand the great gain of postoperative radiation therapy leading to increased

survival of breast cancer patients has also been verified. A recent update of the “early breast

cancer trialists” by Sautter-Bihl et al. 2012, showed that postoperative radiation following

breast conserving surgery reduced the local re-occurrence with 50 percent after a 10 year

follow up period compared to breast cancer patient who didn’t receive postoperative radiation

therapy [82].

All these factors; acute radiation sensitivity, late radiation sensitivity, the risk of secondary

cancers and the risk for local recurrences, need to be carefully considered in order to achieve

an individualized treatment for the patient.

16

Radiation therapy treatment of breast cancer

The standard protocol for treatment of breast cancer in Sweden is briefly described below,

this is a general description and individual cases might be treated differently. The source of

this information is through personal communication with MD Elisabet Lidbrink (Karolinska

University Hospital, Radiumhemmet, oncology department) and information from the

Swedish health care advice.

Once a “lump” is discovered, either through self-examination or mammographic screening,

both breasts will be investigated with mammographic screening and clinical examination. If

there are indications that the “lump” could be cancer a cytology is performed. Patients with a

diagnosed tumor are scheduled for surgery. At the surgery the tumor and part of the breast (or

the whole breast) is removed. During the surgery the sentinel node is identified by injecting a

colored, isotope labelled liquid into the breast lymph. The sentinel node is the first lymph

node that filters the lymphatic fluid from the breast. This sentinel node is removed and

analyzed for tumor cells during the surgery. If it is negative for tumor cells, no more lymph

nodes are removed. If it is positive additional lymph nodes are removed to investigate the

spread of tumor cells. Lymph node positive patients and patients with aggressive tumors,

generally receive chemotherapy prior to the radiation therapy.

The tumor is investigated for hormonal markers (estrogen receptor and progesterone receptor

positive), Herceptin positive (HER2+) and a proliferation marker (Ki67). Patients with

HER2+ tumors receive antibody based therapy and estrogen or progesterone responsive

tumors are treated with targeted hormonal therapy.

Two out of three breast cancer patients receive radiation therapy. After breast conservatory

surgery (removing the tumor and part of the breast) all patients receive radiation therapy.

Lymph node positive patients are generally treated with radiation therapy as well. The

fraction schedule and total dose depend on several clinical factors, for example age, lymph

node status and the location of the tumor. After the completion of treatment, the patient is

subjected to control screenings every 6-12 months for the first two years and then more sparse

controls for up to 10 years depending on the aggressiveness of the tumor.

17

Role of inflammation responses and oxidative stress in individual radiation sensitivity It is known that radiation cause immune responses, where high doses trigger an acute phase

response [72]. This acute phase response can be detected with the classical signs of

inflammation; redness, swelling, and, tenderness of the irradiated area [72]. Lower doses of

radiation (0.3-1 Gy fractions to an accumulated dose of 3-12 Gy) have also been shown to

have an immunosuppressive function and have been used to treat inflammatory diseases,

reviewed in [83, 84]. These diverse and dose dependent responses of the immune system

demonstrates the complexity of irradiation induced inflammatory responses and immunity

responses in general. An acute phase response can be triggered by many stressors, for

example ROS and resembles an inflammation response [85]. A brief summary of

inflammation responses and radiation sensitivity will follow.

High doses of radiation are known to give rise to immediate inflammation responses as

described above [72]. This results in activation of several immune-responsive cells, leading to

a release of pro-inflammatory cytokines: TNF-α, IL-1, IL-8, IFN-γ, and IL-6 [85]. This first

stage of inflammation will be followed by a late stage of inflammation down-regulating the

inflammation response, reviewed in [85]. The down-regulation of inflammation responses is

due partly to the short half-life of pro-inflammatory cytokines and to production of anti-

inflammatory cytokines, IL-4, IL-19, IL-13, and TGF-β [85].

It has been suggested that radiation therapy might promote “wound repair” in the irradiated

tissue, possibly through activation of many cellular processes, such as the coagulation system,

epithelial regeneration, inflammation, matrix changes, and granulation tissue [86]. Pattern

recognition receptors (PRRs) including Toll-like receptors (TLR) have also been suggested to

be responsible for part of this first immune activation by irradiation. PRRs are receptors on

innate immune cells that recognize foreign microbial products. PRRs have been suggested to

have a role in maintenance of tissue homeostasis, wound healing, and regeneration. Thus, the

normal “danger” signal for damage control and wound repair has been suggested to be

triggered by irradiation [87, 88]. Oxidative stress has been shown to both be a trigger of

inflammation and increased during inflammation, as a result of ROS production by immune

cells [36].

18

Lower doses of radiation have been suggested to have an anti-inflammatory effect on

inflammatory diseases, reviewed in [83, 84]. On the contrary, the effect of low dose IR in

normal healthy subjects has been associated with an increased oxidative burst, which was

interpreted as an inflammatory reaction, still to reach measurable levels additional activation

of the blood was needed [49]. In a 3D skin model the low doses of 30 and 100 mGy induced a

mild increase in a few cytokines, to compare, at 2 Gy there where a significant increase of

many cytokines [50]. These studies highlight the need for a better understanding of the

mechanisms of low dose responses.

There are a number of diseases associated with increased oxidative stress levels,

autoimmunity and radiation sensitivity, which indicates that inflammation, oxidative stress,

and radiation sensitivity might be interconnected (mentioned in the section “Individual

radiation sensitivity”). Still, the actual connection between these factors is not completely

understood.

Studies of radiation sensitivity

Over the years a large number of studies investigating the mechanisms of radiation sensitivity

as well as diagnostic tools have been conducted. The methods applied and endpoints

measured have differed greatly. However, the mechanisms involved are still largely unknown

and for medical diagnosis the predictive power of the methods has not proven to be sufficient.

Historically DNA repair has been viewed as one on the most important factors [8]. There are

many methods used to monitor individual capacity for DNA damage repair, for example,

micronucleus assay, comet assay and analysis of chromosomal aberrations, and γ-H2AX foci.

No validated assay exists at present that meets the standards to be used as a diagnostic tool in

the clinic for classification of individual radiation sensitivity [89]. Many researchers have

tried to investigate the intrinsic radiation sensitivity by culturing cells from sensitive patients,

fibroblasts or lymphocytes [89]. While some of the studies indicate an association between

late adverse effects and the sensitivity of the cultured cells [90] [91-93] other studies didn’t

find an association [94-97].

Nonetheless, it should be mentioned that there are studies suggesting that a modified protocol

of chromatid aberration measurement (G2-assay) in lymphocytes form patients after in vitro

exposure could be used to predict radiation sensitivity [98, 99]. The level of radiation induced

19

8-oxo-dG has been suggested to have predictive power when comparing non-responding

patients with extremely sensitive patients (paper II, [67]).

No general genetic profiles or polymorphisms connected to radiation sensitivity have been

found, with the exception of a few known genetic traits (table 3). As mentioned previously a

large number of reviews have been published, which summarize the attempts to discover the

biological mechanisms behind radiation sensitivity with genomics, primarily investigating

candidate SNPs [62] [65, 100, 101]. In a study by Barnett et al. (2012), 92 candidate SNPs

previously associated with radiation sensitivity were investigated in 1613 patients, however

none of the investigate SNPs were significantly associated with radiation sensitivity after

correction for multiple testing [64]. Nevertheless, Talbot et al (2012) published a promising

result for associations between overall toxicity and polymorphisms close to the TNFα gene in

a study of 2036 breast cancer patients [102].

A general conclusion based on these studies is that individual sensitivity to radiation is likely

to be multifactorial in its origin and that several biomarkers/bioassays will be needed to gain

the predictability necessary for implementation in the clinic.

Radiation sensitivity and OMICS (RADIO-OMICS)

Polymorphisms in several genes have been associated with radiation sensitivity. However,

none of the identified SNPs can in general explain radiation sensitivity [64]. Barnett and Hirst

suggested that large collaborative studies are needed to screen the number of individuals

necessary for statistically significant results in genome wide association studies (GWAS

assays) [65, 100]. There are so far only a few such studies examining the association between

SNPs and adverse tissue reaction in radiation sensitive patients, the European Society of

Therapeutic Radiology and Oncology’s (ESTRO) GENEPI project [103], RAPPER [104,

105], Gene-PARE [106], the RadGenomic project [107], and radiogenomic consortium [108].

It is possible that these large collaborations will be successful in the search of specific SNPs

associated with radiation sensitivity.

Several genomic expression profiles linking gene expression to radiation sensitivity have been

published. These studies have generally investigated the gene expression after high doses of

IR. A number of gene expression profiles have been published that was suggested to be

20

associated with radiation sensitivity for acute adverse effects [94, 109-111] or late adverse

effects [112-114]. Two studies suggested the expression of two single genes to be potential

biomarkers for radiation sensitivity, XPC [115] and CDKN1A [116] respectively. As to my

knowledge, none of the mentioned studies has been validated.

MicroRNA is a growing area of research and will most certainly become important in

radiation sensitivity research in the future. A recent review was published by Metheetrairut

and Slack (2013) summarizing the progress in the field regarding microRNAs and IR [117].

However, there are no studies investigating the effects of microRNA on normal tissue

radiation sensitivity.

In the field of proteomics few articles have to my knowledge been published that investigate

radiation sensitivity in humans. A review summarizing the present status of radiation research

and proteomics was published in 2013, however no studies of radiation sensitivity was

mentioned [118]. Another review was recently published by Lacombe et al. summarizing the

attempts using proteomics to find biomarkers of radiation sensitivity [119]. The review states

that very few studies have been conducted and that only two methods exist that possibly could

predict late radiation sensitivity [119]. These two studies are: Cai et al. who published that

patients with non-small-cell lung carcinoma being diagnosed with radiation-induced lung

toxicity grade of 2 or higher had an up-regulation of vitronectin and C4b-binding protein α

chain [120, 121] and Ozsahin et al. who published a test based on radiation-induced

lymphocyte apoptosis (RILA), predicting late radiation sensitivity [122].

For acute radiation sensitivity there are to my knowledge no proteomic studies in humans that

have found possible biomarkers. Nonetheless, Fang et al. 2010 found in breast cancer patients

an association between low protein levels of ATM and a high risk of being radiation sensitive

[123].

There is no clear correlation between acute and late effects in clinical studies. Therefore it is

questionable if the profile of late radiation sensitivity would be predictive of acute radiation

sensitivity and vice versa.

21

Aim and strategy of the studies The long term aim of this thesis was to develop diagnostic tools to assess the acute individual

radiation sensitivity of breast cancer patients and to improve our knowledge about the

mechanisms behind radiation sensitivity.

For these purposes whole blood from patients with different radiation sensitivity were used.

Whole blood was chosen since it is easily accessible for diagnostic screening and since it is a

complete tissue. A bioassay with a proteomic protocol was designed (Paper I) followed by a

study of the proteomic profiles in breast cancer patients with different radiation sensitivity

after in vitro irradiation of whole blood (Paper III). The level and induction of the oxidative

stress marker, 8-oxo-dG, after IR was investigated in the same patient cohort (Paper II).

Materials and method

Blood

There are strong evidences that cell-to-cell communication is important for the responses in

vivo of blood cells [87, 124]. The reason for choosing whole blood as a model system is that it

is a complete tissue where cell-to-cell interactions are possible. The intrinsic stress of the

system can be reduced avoiding the purification steps such as gradient centrifugation.

From the whole blood, leukocytes were isolated after exposure according to the bioassays

developed. Leukocytes consists of several different cell types, neutrophils, lymphocytes,

monocytes, eosinophils and basophils [124].

All studies were approved by the local ethical committee.

Patient cohort

The patient cohort consist of 2914 breast cancer patients whose skin reactions have been

photographically documented shortly after completion of radiation therapy and used to

document tissue effects of the irradiated area (figure 3). The pictures have been assessed

according to the RTOG scale, (described under the section “Symptoms of acute and late

radiation sensitivity in breast cancer patients”).

Information about the patients smoking habits, diet, anti-oxidant intake and medications was

collected at the sampling. The location of the breast cancer, information on chemotherapy,

dose per fraction and accumulated dose for each patient was recorded from their medical

journals.

22

Figure 3. The distribution of breast cancer patients treated by radiation therapy according to the

standard RTOG classification for acute adverse skin reactions (n=2914).

Radiation facilities

Low dose exposures were conducted using a custom built cell-culture incubator equipped

with a 137

Cs source, providing dose rates ranging from 5 mGy/h- 30 mGy/h. All doses below

10 mGy were delivered at a dose rate of 15 mGy/h. For moderate doses a high dose rate (0.4

Gy/min) source equipped with a 137

Cs source (Scanditronix, Uppsala, Sweden) was used. All

irradiations were performed on ice.

20

58

18

3.5

0.35

0

10

20

30

40

50

60

70

RTOG 0 RTOG 1 RTOG 2 RTOG 3 RTOG 4

Severity of side effects

Perc

en

t ra

dio

thera

py-t

reate

d b

reast

can

cer

pati

en

ts

23

Paper I

Low doses of γ-radiation induce consistent protein expression

changes in human leukocytes

Aim and introduction This study uses a proteomic approach to investigate whether a consistent radiation response

can be detected in samples of whole blood from two healthy volunteers following in vitro

exposure to low and moderate doses of γ-radiation. The study also examines how consistent

the response is with time and the differences between the two donors. The long term aim of

the study was to develop a protocol to investigate if there are differences in the proteomic

profiles between radiation sensitive patients and normo-sensitive patients when whole blood

were exposed to low or moderate doses of radiation in vitro.

Methods

The sampling, exposures and analysis of gel data were conducted at the Department of

Genetics, Microbiology and Toxicology at Stockholm University. The protein isolation, two-

dimensional polyacrylamide gel electrophoresis (2D-PAGE) and cutting of protein spots were

conducted at the Department of Oncology and Pathology at Cancer Centre Karolinska. The

mass spectrometry was conducted at Ludwig Institute for Cancer Research at Uppsala

University.

Result and discussion

It was found that the doses of 1 mGy and 150 mGy followed by 3 hours post-irradiation

incubation at 37 ºC gave rise to the greatest change in the number of significantly deregulated

protein compared to the control. A consistent radiation response could be seen for both the

low and the moderate doses; however the response differs between the two individuals. There

was also a difference in the protein expression between the two control samples in experiment

one and two. However, mass spectrometry (MS) identified 12 proteins in donor A and 10 in

donor B that significantly changed expression with IR at both time points. In total 17 proteins

were identified with MS to be low dose radiation specific. These proteins can be divided into

several groups based on their functions, for example, general oxidative stress, neutrophil and

T-cell activation, fibrinolytic system, and cytoskeletal modifications.

24

Our result indicates that the protocol for radiation exposure of whole blood (figure 4) can be

applied for studies of radiation induced changes of proteomic profiles associated with

radiation response and sensitivity.

Main findings in paper I

- The optimal doses and post-irradiation time was determined.

- A protocol for proteomic investigation of radiation induced changes of in vitro

irradiated whole blood was established (figure 4).

- Radiation specific protein changes can be detected after the low and intermediate

doses of 1 mGy and 150 mGy.

- Despite large individual variations with time, consistent radiation specific protein

changes could be detected at the two sampling time points.

Figure 4. Schematic picture of the protocol developed in paper I.

25

Paper II

Radiation-induced stress in peripheral blood of breast cancer

patients differs between patients with severe acute skin reactions

and patients with no side effects to radiotherapy

Aim and introduction The aim of the study was to compare the radiation-induced oxidative stress response in blood

samples from breast cancer patients that developed severe acute skin reactions during

radiation therapy, with the response in blood samples from patients with no visual side

effects.

It has been suggested that urinary 8-oxo-dG levels could be used as a marker for individual

radiation sensitivity in breast cancer patients [67]. It was hypothesized that the observed

differences in urinary levels of 8-oxo-dG in breast cancer patients undergoing radiation

therapy may reflect individual capacity to handle oxidative stress induced by the radiation.

Leukocytes in whole blood from sensitive patients and normo-sensitive patients can be used

to monitor oxidative stress in response to radiation. To test if the two groups differ in

radiation-induced oxidative stress response, peripheral blood was collected from 12 breast

cancer patients showing no acute skin reactions after radiation therapy (RTOG grade 0) and

from 14 breast cancer patients who developed acute severe skin reactions (RTOG grade 3–4).

Whole blood was irradiated with 5 and 2000 mGy γ-radiation and serum was isolated after 1 h

post-irradiation incubation at 37°C. The biomarker for oxidative stress, 8-oxo-dG, was

analyzed in the serum by a modified ELISA assay [32, 33].

Methods

The level of oxidative stress was measured with the oxidative stress marker 8-oxo-dG in

serum from in vitro irradiated whole blood. Blood sampling was conducted at Karolinska

University Hospital. The irradiations and 8-oxo-dG measurements were performed at

Stockholm University (figure 5).

The measurement of 8-oxo-dG was analyzed with a modified ELISA assay. Briefly, serum

was collected after in vitro irradiation on ice and 1 hour post-irradiation incubation at 37°C of

26

whole blood. The serum was filtered twice to accumulate 8-oxo-dG, a spiked sample was

used as a control of the filtering. This was followed by an ELISA assay, where triplicates of

each sample were run. The samples from each patient were run on the same ELISA-plate and

for each plate a standard curve was established. The relative radiation induced 8-oxo-dG

levels were calculated. Extracellular 8-oxo-dG is primarily a biomarker of oxidative

nucleotide damage in this assay [33].

Figure 5. A schematic picture of the assay used in paper II.

Result and discussion

Significant radiation-induced increase of 8-oxo-dG levels was observed in serum of the

RTOG 0 patients, no increase was seen in serum of the RTOG 3–4 patients. The results show

that samples of whole blood from RTOG 3-4 patients differ significantly in their oxidative

stress response to ionizing radiation compared to samples of whole blood from RTOG 0

patients. This indicates that the RTOG 3-4 patients differ in their cellular response to ionizing

radiation at the level of induction of oxidative stress or at the level of repair or both. The

radiation-induced differences in 8-oxo-dG levels between RTOG 0 and RTOG 3-4 patients

indicate that the assay could be used to identify radiation sensitive patients. However for this

purpose the relative level of radiation induced 8-oxo-dG in patients classified as RTOG 1 and

2 needs to be investigated.

27

Main findings in paper II

- Radiation sensitive patients and normo-sensitive patients show different relative levels

of radiation induced 8-oxo-dG after low or high doses of in vitro irradiated whole

blood. Normo-sensitive patients show an increased relative level of 8-oxo-dG after

radiation and radiation sensitive patients show no induction or a decrease of relative 8-

oxo-dG levels.

- This study suggests that there are differences in oxidative stress responses between

radiation sensitive and normo-sensitive patients.

- The difference in relative 8-oxo-dG levels indicate that this assay might be useful in

screening for the most radiation sensitive patients. However the response of patients

classified as RTOG 1 and 2 needs to be investigated.

28

Paper III

Unique proteomic signature for radiation sensitive patients; a

comparative study between normo-sensitive and radiation

sensitive breast cancer patients

Introduction and aim The aim of this study was to investigate the proteomic profiles, before irradiation and after in

vitro irradiation of whole blood, in radiation sensitive and normo-sensitive patients. Radiation

therapy is adjusted to the most sensitive patients where 5 percent severe acute adverse healthy

tissue effects are accepted and twenty percent will get milder adverse effects from their

radiation therapy. There are indications in literature that the patients with no signs of adverse

effects have a higher probability of local reoccurrence of cancer within 5 years, indicating

they would have benefitted from a higher dose of IR [3, 5].

Previous studies, Haghdoost 2001 and paper II, have shown that there are differences in these

patient groups regarding the level of radiation induced 8-oxo-dG (an oxidative stress marker)

[67]. This study aims at comparing the two groups both after irradiation and at the basal level

in hope of identifying possible biomarkers of acute individual radiation sensitivity.

Methods

Sampling was conducted at Karolinska University Hospital. Irradiations, isolation of cells

(according to the protocol developed in paper I) and preparation of proteins was conducted at

Stockholm University. The protein samples from each group were pooled and sent to

Helmholtz Centre in Munich where the Isotope-coded protein label (ICPL) experiments, mass

spectrometry, selection of high confident peptides and statistical calculations were conducted.

A pooling approach was chosen to reduce the individual differences and to facilitate the

investigation due to limited material. The protein-protein interaction and pathway analysis

were performed at Stockholm University under the supervision of the Helmholtz Centre in

Munich.

Briefly, the pooled protein samples were acetone precipitated and resuspended in a

compatible labelling buffer (SERVA). Protein measurements were conducted with the

Bradford method. Fifty µg of the samples were then reduced and alkylated followed by

labelling according to the manufacture’s protocol (SERVA). ICPL_0 (1- ( 12

C61H4 )-

29

Nicotinoyloxy-succinimide Mr= 105.0215 Da) for control samples, ICPL_4 (1 - ( 12

C62D4 )-

Nicotinoyloxy-succinimide Mr= 109.0715 Da) and ICPL_6 (1 - ( 13

C62H4 )-Nicotinoyloxy-

succinimide Mr= 111.0419 Da) for radiated samples. In the duplex labeling ICPL_0 and

ICPL_6 were used (figure 6). To reduces the complexity of the samples, the labeled samples

were mixed and separated based on molecular weight on a 1D SDS gel. The gel lane was then

cut into 5 slices and digested with trypsin. The trypsin-generated peptides were separated with

a reverse phase chromatograph connected to the ESI-MS/MS. The resulting MS/MS spectra

were searched against Ensembl human Mascot data base and protein identification and

quantification was done with the software proteome discoverer 1.3 (figure 7). The heavy/light

ratio and ratio variability (CV%) were applied for protein quantification. Proteins identified

by at least two unique peptides in at least two of the three experiments, with a variability

below 30% were considered for further analysis. Proteins deregulated greater than 1.3 fold or

less than -1.3 fold were selected as significantly deregulated (proteins with CV% of 30-40%

were included if the p-value was below 0.05). Significantly deregulated proteins were

analyzed for protein-protein interaction and pathway analysis using Ingenuity pathway

analysis or STRING [125]. Ingenuity is a knowledge data base, based on peer-reviewed

scientific publications, where functions, signaling networks and pathway analysis can be

conducted [126]. STRING is a peer-reviewed knowledge database, where protein-protein

interactions, associations and signaling networks can be investigated [127].

Figure 6. The work flow from sampling to protein analysis. The collection and irradiation of blood

samples, the isolation of leukocytes and preparation of the cell lysates (left side) and the duplex- and

triplex-ICPL labelling (right side).

30

Figure 7. A schematic view of Triplex isotope-coded protein labelling (ICPL) procedure. The samples

are reduced and alkylated followed by ICPL labelling (SERVA). Differently labeled proteins are then

combined and separated based on molecular weight 1D gel electrophoresis. The gel is cut in five

pieces followed by LC-ESI-MS/MS analysis. The MS/MS spectra were searched against Ensembl

human Mascot database. The protein identification and quantification was done with Proteome

discoverer version 1.3. The protein-protein interaction analysis and pathway analysis was done with

Ingenuity pathway analysis (IPA) and STRING.

Result and discussion

Unique proteomic signatures separating the two groups after the low and intermediate doses

of 1 and 150 mGy of IR was found. Pathway analysis with STRING and Ingenuity of the

proteomic profile suggests a few possible mechanisms involved in radiation sensitivity

responses. The main differences involve oxidative stress response, acute phase response and

31

wound healing responses in which the two groups show markedly different response.

Radiation sensitive patients show an increased inflammation or wound healing response

already at the basal level and this response is increased following low dose irradiation. No

acute phase or wound healing response were detected in the normo-sensitive patients.

Additionally the groups differ in redox responses, where normo-sensitive patients have higher

basal levels of SOD1 and PARK7 than radiation sensitive patients. The levels of these anti-

oxidative stress proteins (SOD1 and PARK7) are then decreased following 1 mGy of

irradiation in normo-sensitive patients. This response is not seen in radiation sensitive

patients, who instead up-regulate PRDX2 and BLVRB following both doses of IR.

The difference in redox response between the two groups is of particular interest as it links

mechanistically to the difference in ROS response observed in paper II.

There are a few proteins which show a general radiation response. However these proteins

don’t seem to be involved in the radiation sensitivity response.

Differently expressed proteins and predicted upstream regulators need to be validated in the

individual patients. However a unique protein expression profile of sensitive patients was

identified.

Main findings in paper III

- Radiation sensitive and normo-sensitive patients show different protein expression

profiles at the basal level (before irradiation).

- Low doses of IR trigger different protein expression profiles in groups of radiation

sensitive and normo-sensitive patients.

- The identified significantly deregulated proteins might be useful as biomarkers of

radiation sensitivity in the future.

32

Concluding remarks and future studies

No general biomarker of acute radiation sensitivity has been identified yet. One probable

reason for this is that the genetics behind radiation sensitivity is complex [66]. It is also

probable that the biology behind radiation sensitivity is diverse and caused by many different

factors. These aspects need to be taken into consideration in the search for biomarkers. This is

one reason why 8-oxo-dG might be a good marker, as it measures the resulting oxidative

stress of different stress responses independent of the mechanisms behind this difference. In

the future it is possible that 8-oxo-dG in combination with additional marker such as certain

single nucleotide polymorphisms and a few unique proteins might be useful for predicting

acute radiation sensitivity.

A combination of different biomarkers will most certainly improve the diagnostic power.

However, many factors need to be carefully considered in order to achieve an individualized

treatment for the patient; acute radiation sensitivity, late radiation sensitivity, the risk of

secondary cancers and the risk for local recurrences. Once a combination of biomarkers is

identified, this panel of biomarkers needs to be validated in a larger cohort of patients,

preferably a prospective study. However due to the low number of extremely sensitive

patients in our cohort (only 0.35% is RTOG4) this might be difficult. Therefore, a

combination of a prospective study with a retrospective study should be carried out to

investigate a sufficient number of sensitive patients to gain statistical power.

Paper III suggests a few pathways which might be triggered in response to IR in sensitive

patients, i.e. acute phase responses/ wound healing (inflammation) and oxidative stress

responses. It would be interesting to investigate these pathways in different cellular systems to

further examine the mechanisms on a molecular level. It is possible that cell lines derived

from radiation sensitive patient lymphocytes could be established. SiRNA could be used to

knock down the expression of target proteins, in for example the lymphoblastoid TK6 cell

line, to investigate if the knocked down protein effect the radiation sensitivity.

The acute phase response (inflammation response) would be interesting to investigate further

since there is very limited data regarding low dose radiation and inflammation responses in

humans. If possible, it would be interesting to collect blood following CT-investigations

33

where the dose of radiation is low. It is possible that individuals with normal radiation

sensitivity don’t respond with biomarkers of an inflammation reaction following low dose IR,

but it needs to be investigated. However since there are several diseases connected to

inflammation, increased oxidative stress and radiation sensitivity these responses might be

mechanistically connected. It would be interesting to investigate how patients with these

different diseases respond in our 8-oxo-dG assay following low doses of IR.

Ongoing projects

MicroRNA and gene expression studies

The patients in paper II and III, have been asked to participate in additional studies

investigating the microRNA response as well as the gene expression after low and

intermediate doses of IR. This will probably add to our understanding of the mechanisms

behind radiation sensitivity in this cohort.

Candidate single nucleotide polymorphisms study of breast cancer patients

DNA has been sampled from 146 patients (RTOG 0-4) in the breast cancer cohort. A

candidate SNP analysis in these patients investigating 52 SNPs previously associated with

radiation sensitivity, in genes involved in; oxidative stress responses, DNA repair or

inflammation was conducted.

Although the number of SNPs investigated was large and the number of patients was limited

we found a few risk associated SNPs that seem promising. Our idea is to combine potential

risk SNPs with relative 8-oxo-dG levels after in vitro IR and the deregulation of certain

proteins after in vitro IR in a statistic model predicting acute radiation sensitivity. This

approach has been tested in our lab with promising results for head and neck cancer patients

that developed late adverse tissue reactions (Brehwens et al. unpublished). The statistic

modeling would enable us to find a good combination of biomarkers, increasing the

predictability.

34

Acknowledgements I would like to thank my supervisors, professor Mats Harms-Ringdahl and assoc. professor

Siamak Haghdoost for all their support, supervision and discussions. You have made my

PhD studies interesting, fun and exciting, and I want to thank you not only for your excellent

guidance but also for your kindness and all the fun we have had! If I ever become a “boss” I

will try to be just like you.

I would also sincerely like to thank;

Professor Andrzej Wojcik, for good discussions, honest opinions and fun and interesting

trips.

Dr. Soile Tapio for introducing me to ICPL and pathway analysis, for helpful discussions and

suggestions in writing the paper III.

Dr. Omid Azimzadeh for all your help and patience. Without your support I would never

have understood the world of ICPL and pathway analysis. You are brilliant at explaining

things!

Susanne Becker for all your helpful support and for all the fun times we have had at KI while

learning 2D-PAGE.

Arja Andersson for all your efforts in assessing, inviting and collecting patient samples and

for nice dinners.

Elisabet Lidbrink for personal communications about breast cancer treatment and for

assessments of the breast cancer cohort.

Thank you all in the radiation biology group! We have had a lot of fun during the years!

Thank you;

Ainars for all your support during my PhD and for all our discussions, scientific and not so

scientific. Sharing an office with you was never boring and I really miss you and your

company!

Karl for help and discussion. I really enjoy our discussions during our coffee breaks. It has

been great to share the frustrations of writing with someone in the same situation.

Sara for always being so nice and helpful and for chairing a strong coffee addiction, staying

late doing 2D was never tiresome in your company.

35

Elina for all you support with SIMCA and for your fantastic attitude. You are always helpful

and friendly.

Asal for being such a nice colleague and good travelling companion. My suitcase always felt

light and small in comparison.

Tai for all your help and especially for always being helpful and nice. You are one of the

corner stones in our lab. I hope you succeed in fulfilling your dreams.

Alice for all the 8-oxo-dG analysis and optimization trials and especially for all the fun times

and chairing the trials of life, I really appreciate your support!

Eliana for being such a nice person, it is good to have you back in Sweden.

Siv for all your help, effort and patience while proof-reading my manuscripts.

Ramesh for helping me understand the world of proteomics and for your honesty.

All the members of the radiation biology group for their enthusiasm, helpfulness and

friendship. You guys made going to work a joy!

I would also like to thank my old colleagues and friends, you all have made a large impact on

my life and I thank you for your friendship and support.

Maria and Sandra at IMM, I learned a lot from you and really enjoyed the time we worked

together.

Södertörngänget; Micke, Eva-Maria, Lova, Katrin, Johanna och Emma, tack för allt!

Alla mina underbara vänner, tack för att ni finns där i vått och tort! Speciellt tack till Sara,

Emelie, Karin, Maria och Micke som har pushat och stöttat under skrivprocessen.

Sist men inte minst vill jag tacka min familjen. Ni betyder allt! Tack för att ni finns och för

allt stöd ni har gett mig genom åren.

Tack Inger & Kalle för allt ert stöd och hjälp genom åren! Pia för att du är världens bästa

syster och för att du om och om igen bevisar att man klarar allt bara man kämpar!

Tack Lilian & Lennart för att ni ställt upp med barnvakt och stöd.

Tack Farmor för din härliga livsinställning och för att du påminner mig om att allt är möjligt.

Tack Peter för att du finns där för mig, utan dig skulle jag vara vilse. Vilgot & Ale ni är

fantastiska, jag är tacksam över att ni är mina söner och för att ni gör livet spännande varje

dag!

36

References 1. Socialstyrelsen, Cancer Incidence in Sweden 2011Cancerförekomst i Sverige 2011.

www.socialstyrelsen.se, 7 december 2012, 2012. 2012-12-19.

2. Bentzen, S.M., Preventing or reducing late side effects of radiation therapy:

radiobiology meets molecular pathology. Nat Rev Cancer, 2006. 6(9): p. 702-13.

3. Kuhnt, T., et al., Acute radiation reaction and local control in breast cancer patients

treated with postmastectomy radiotherapy. Strahlenther Onkol, 1998. 174(5): p. 257-

61.

4. Alsbeih, G., et al., Chromosomal fragility syndrome and family history of

radiosensitivity as indicators for radiotherapy dose modification. Radiother Oncol,

2003. 66(3): p. 341-4.

5. Dahl, O., A. Horn, and O. Mella, Do acute side-effects during radiotherapy predict

tumour response in rectal carcinoma? Acta Oncol, 1994. 33(4): p. 409-13.

6. Andreassen, C.N., Can risk of radiotherapy-induced normal tissue complications be

predicted from genetic profiles? Acta Oncol, 2005. 44(8): p. 801-15.

7. Bourguignon, M.H., et al., Genetic and epigenetic features in radiation sensitivity.

Part II: implications for clinical practice and radiation protection. Eur J Nucl Med

Mol Imaging, 2005. 32(3): p. 351-68.

8. ICRP publication 79, I.P., Genetic susceptibility to cancer. ICRP publication 79.

Approved by the Commission in May 1997. International Commission on Radiological

Protection. Ann ICRP, 1998. 28(1-2): p. 1-157.

9. Bentzen, S.M. and J. Overgaard, Patient-to-Patient Variability in the Expression of

Radiation-Induced Normal Tissue Injury. Semin Radiat Oncol, 1994. 4(2): p. 68-80.

10. Tucker, S.L., I. Turesson, and H.D. Thames, Evidence for individual differences in the

radiosensitivity of human skin. Eur J Cancer, 1992. 28A(11): p. 1783-91.

11. Turesson, I., et al., Prognostic factors for acute and late skin reactions in radiotherapy

patients. Int J Radiat Oncol Biol Phys, 1996. 36(5): p. 1065-75.

12. Turesson, I., Individual variation and dose dependency in the progression rate of skin

telangiectasia. Int J Radiat Oncol Biol Phys, 1990. 19(6): p. 1569-74.

13. Safwat, A., et al., Deterministic rather than stochastic factors explain most of the

variation in the expression of skin telangiectasia after radiotherapy. Int J Radiat

Oncol Biol Phys, 2002. 52(1): p. 198-204.

14. SSI Report 2007:14 Utvärdering av miljökvalitetsmålet Säker strålmiljö, 2007: SSI. p.

122.

15. SSI Raport 2007:02 Strålmiljö i Sverige, 2007, SSi: SSI. p. 144.

16. Wrixon, A.D., New ICRP recommendations. J Radiol Prot, 2008. 28(2): p. 161-8.

17. ICRP publication 103, I.P., ICRP Publication, 103. The 2007 recommendations of the

international commision on radiological protection. ICRP Publication, 1032007:

Elsevier Ltd.

18. Royal, H.D., Effects of low level radiation-what's new? Semin Nucl Med, 2008. 38(5):

p. 392-402.

19. Cardis, E., et al., The 15-Country Collaborative Study of Cancer Risk among

Radiation Workers in the Nuclear Industry: estimates of radiation-related cancer

risks. Radiat Res, 2007. 167(4): p. 396-416.

20. Heidenreich, W.F., H.G. Paretzke, and P. Jacob, No evidence for increased tumor

rates below 200 mSv in the atomic bomb survivors data. Radiat Environ Biophys,

1997. 36(3): p. 205-7.

21. Brenner, D.J., et al., Cancer risks attributable to low doses of ionizing radiation:

assessing what we really know. Proc Natl Acad Sci U S A, 2003. 100(24): p. 13761-6.

37

22. Preston, D.L., et al., Studies of mortality of atomic bomb survivors. Report 13: Solid

cancer and noncancer disease mortality: 1950-1997. Radiat Res, 2003. 160(4): p.

381-407.

23. Mullenders, L., et al., Assessing cancer risks of low-dose radiation. Nat Rev Cancer,

2009. 9(8): p. 596-604.

24. Brenner, D.J. and E.J. Hall, Computed tomography--an increasing source of radiation

exposure. N Engl J Med, 2007. 357(22): p. 2277-84.

25. Goodhead, D.T., Initial events in the cellular effects of ionizing radiations: clustered

damage in DNA. Int J Radiat Biol, 1994. 65(1): p. 7-17.

26. Sanner, T. and A. Pihl, Significance and mechanism of the indirect effect in bacterial

cells. The relative protective effect of added compounds in Escherichia coli B,

irradiated in liquid and in frozen suspension. Radiat Res, 1969. 37(1): p. 216-27.

27. Pouget, J.P. and S.J. Mather, General aspects of the cellular response to low- and

high-LET radiation. Eur J Nucl Med, 2001. 28(4): p. 541-61.

28. Feinendegen, L.E., Evidence for beneficial low level radiation effects and radiation

hormesis. Br J Radiol, 2005. 78(925): p. 3-7.

29. Ames, B.N., M.K. Shigenaga, and T.M. Hagen, Oxidants, antioxidants, and the

degenerative diseases of aging. Proc Natl Acad Sci U S A, 1993. 90(17): p. 7915-22.

30. Loft, S. and H.E. Poulsen, Cancer risk and oxidative DNA damage in man. J Mol

Med, 1996. 74(6): p. 297-312.

31. Evans, M.D., M. Saparbaev, and M.S. Cooke, DNA repair and the origins of urinary

oxidized 2'-deoxyribonucleosides. Mutagenesis, 2010. 25(5): p. 433-42.

32. Haghdoost, S., et al., Extracellular 8-oxo-dG as a sensitive parameter for oxidative

stress in vivo and in vitro. Free Radic Res, 2005. 39(2): p. 153-62.

33. Haghdoost, S., et al., The nucleotide pool is a significant target for oxidative stress.

Free Radic Biol Med, 2006. 41(4): p. 620-6.

34. Sangsuwan, T. and S. Haghdoost, The nucleotide pool, a target for low-dose gamma-

ray-induced oxidative stress. Radiat Res, 2008. 170(6): p. 776-83.

35. Nakabeppu, Y., et al., Programmed cell death triggered by nucleotide pool damage

and its prevention by MutT homolog-1 (MTH1) with oxidized purine nucleoside

triphosphatase. Mutat Res, 2010. 703(1): p. 51-8.

36. Zhao, W., D.I. Diz, and M.E. Robbins, Oxidative damage pathways in relation to

normal tissue injury. Br J Radiol, 2007. 80 Spec No 1: p. S23-31.

37. Hoeijmakers, J.H., Genome maintenance mechanisms for preventing cancer. Nature,

2001. 411(6835): p. 366-74.

38. Ishchenko, A.A., et al., Uncoupling of the base excision and nucleotide incision repair

pathways reveals their respective biological roles. Proc Natl Acad Sci U S A, 2006.

103(8): p. 2564-9.

39. Maki, H. and M. Sekiguchi, MutT protein specifically hydrolyses a potent mutagenic

substrate for DNA synthesis. Nature, 1992. 355(6357): p. 273-5.

40. Robertson, A.B., et al., DNA repair in mammalian cells: Base excision repair: the

long and short of it. Cell Mol Life Sci, 2009. 66(6): p. 981-93.

41. Parsons, J.L., et al., Poly(ADP-ribose) polymerase-1 protects excessive DNA strand

breaks from deterioration during repair in human cell extracts. FEBS J, 2005. 272(8):

p. 2012-21.

42. El-Khamisy, S.F., et al., A requirement for PARP-1 for the assembly or stability of

XRCC1 nuclear foci at sites of oxidative DNA damage. Nucleic Acids Res, 2003.

31(19): p. 5526-33.

43. David, S.S., V.L. O'Shea, and S. Kundu, Base-excision repair of oxidative DNA

damage. Nature, 2007. 447(7147): p. 941-50.

38

44. Leipold, M.D., et al., Recognition and removal of oxidized guanines in duplex DNA by

the base excision repair enzymes hOGG1, yOGG1, and yOGG2. Biochemistry, 2003.

42(38): p. 11373-81.

45. Michaels, M.L., et al., A repair system for 8-oxo-7,8-dihydrodeoxyguanine.

Biochemistry, 1992. 31(45): p. 10964-8.

46. Maga, G., et al., 8-oxo-guanine bypass by human DNA polymerases in the presence of

auxiliary proteins. Nature, 2007. 447(7144): p. 606-8.

47. High Level and Expert Group Report European low dose risk research Radiation

Protection, in European commission2009: Luxembourg: Office for Official

Publications of the European Communities.

48. Pernot, E., et al., Ionizing radiation biomarkers for potential use in epidemiological

studies. Mutat Res, 2012. 751(2): p. 258-86.

49. Vicker, M.G., et al., Ionizing radiation at low doses induces inflammatory reactions in

human blood. Radiat Res, 1991. 128(3): p. 251-7.

50. Varnum, S.M., et al., The effects of low-dose irradiation on inflammatory response

proteins in a 3D reconstituted human skin tissue model. Radiat Res, 2012. 178(6): p.

591-9.

51. Spitz, D.R., et al., Metabolic oxidation/reduction reactions and cellular responses to

ionizing radiation: a unifying concept in stress response biology. Cancer Metastasis

Rev, 2004. 23(3-4): p. 311-22.

52. Sena, L.A. and N.S. Chandel, Physiological roles of mitochondrial reactive oxygen

species. Mol Cell, 2012. 48(2): p. 158-67.

53. Rhee, S.G., et al., Cellular regulation by hydrogen peroxide. J Am Soc Nephrol, 2003.

14(8 Suppl 3): p. S211-5.

54. Olivieri, G., J. Bodycote, and S. Wolff, Adaptive Response of Human-Lymphocytes to

Low Concentrations of Radioactive Thymidine. Science, 1984. 223(4636): p. 594-597.

55. Matsumoto, H., et al., A new paradigm in radioadaptive response developing from

microbeam research. J Radiat Res (Tokyo), 2009. 50 Suppl A: p. A67-79.

56. Nagasawa, H. and J.B. Little, Induction of sister chromatid exchanges by extremely

low doses of alpha-particles. Cancer Res, 1992. 52(22): p. 6394-6.

57. Mothersill, C. and C.B. Seymour, Cell-cell contact during gamma irradiation is not

required to induce a bystander effect in normal human keratinocytes: Evidence for

release during irradiation of a signal controlling survival into the medium. Radiation

Research, 1998. 149(3): p. 256-262.

58. Prise, K.M., et al., New insights on cell death from radiation exposure. Lancet Oncol,

2005. 6(7): p. 520-8.

59. Bond, V.P., L.E. Feinendegen, and J. Booz, What is a 'low dose' of radiation? Int J

Radiat Biol Relat Stud Phys Chem Med, 1988. 53(1): p. 1-12.

60. Gotoff, S.P., E. Amirmokri, and E.J. Liebner, Ataxia telangiectasia. Neoplasia,

untoward response to x-irradiation, and tuberous sclerosis. Am J Dis Child, 1967.

114(6): p. 617-25.

61. Taylor, A.M., et al., Ataxia telangiectasia: a human mutation with abnormal radiation

sensitivity. Nature, 1975. 258(5534): p. 427-9.

62. Andreassen, C.N. and J. Alsner, Genetic variants and normal tissue toxicity after

radiotherapy: a systematic review. Radiother Oncol, 2009. 92(3): p. 299-309.

63. Alsner, J., C.N. Andreassen, and J. Overgaard, Genetic markers for prediction of

normal tissue toxicity after radiotherapy. Semin Radiat Oncol, 2008. 18(2): p. 126-35.

64. Barnett, G.C., et al., Independent validation of genes and polymorphisms reported to

be associated with radiation toxicity: a prospective analysis study. Lancet Oncol,

2012. 13(1): p. 65-77.

39

65. Barnett, G.C., et al., Normal tissue reactions to radiotherapy: towards tailoring

treatment dose by genotype. Nat Rev Cancer, 2009. 9(2): p. 134-42.

66. Andreassen, C.N., Searching for genetic determinants of normal tissue

radiosensitivity--are we on the right track? Radiother Oncol, 2010. 97(1): p. 1-8.

67. Haghdoost, S., et al., Can 8-oxo-dG be used as a predictor for individual

radiosensitivity? Int J Radiat Oncol Biol Phys, 2001. 50(2): p. 405-10.

68. Robbins, J.H., et al., Parkinson's disease and Alzheimer's disease: hypersensitivity to

X rays in cultured cell lines. J Neurol Neurosurg Psychiatry, 1985. 48(9): p. 916-23.

69. Chen, P., C. Kidson, and M. Lavin, Heterogeneity in Alzheimer's disease: evidence

from cellular radiosensitivity and complementation of this phenotype. Mutat Res,

1991. 256(1): p. 21-7.

70. Takeshita, T., et al., The effect of aging on cell-cycle kinetics and X-ray-induced

chromosome aberrations in cultured lymphocytes from patients with Down syndrome.

Mutat Res, 1992. 275(1): p. 21-9.

71. Harris, G., et al., Radiosensitivity of peripheral blood lymphocytes in autoimmune

disease. Int J Radiat Biol Relat Stud Phys Chem Med, 1985. 47(6): p. 689-99.

72. Hall, E.J. and A.J. Giaccia, Radiobiology for the radiologist. 7th ed2012, Philadelphia:

Wolters Kluwer Health/Lippincott Williams & Wilkins. p.

73. Stone, H.B., et al., Effects of radiation on normal tissue: consequences and

mechanisms. Lancet Oncol, 2003. 4(9): p. 529-36.

74. Trotti, A., et al., Common toxicity criteria: version 2.0. an improved reference for

grading the acute effects of cancer treatment: impact on radiotherapy. Int J Radiat

Oncol Biol Phys, 2000. 47(1): p. 13-47.

75. Cox, J.D., J. Stetz, and T.F. Pajak, Toxicity criteria of the Radiation Therapy

Oncology Group (RTOG) and the European Organization for Research and Treatment

of Cancer (EORTC). Int J Radiat Oncol Biol Phys, 1995. 31(5): p. 1341-6.

76. Dorr, W. and J.H. Hendry, Consequential late effects in normal tissues. Radiother

Oncol, 2001. 61(3): p. 223-31.

77. Welte, B., et al., Second malignancies in highdose areas of previous tumor

radiotherapy. Strahlenther Onkol, 2010. 186(3): p. 174-9.

78. Bartkowiak, D., et al., Second cancer after radiotherapy, 1981-2007. Radiother Oncol,

2012. 105(1): p. 122-6.

79. Johansen, S., T. Danielsen, and D.R. Olsen, Estimated risk for secondary cancer in the

contra-lateral breast following radiation therapy of breast cancer. Acta Oncol, 2008.

47(3): p. 391-6.

80. Suit, H., et al., Secondary carcinogenesis in patients treated with radiation: a review

of data on radiation-induced cancers in human, non-human primate, canine and

rodent subjects. Radiat Res, 2007. 167(1): p. 12-42.

81. Berrington de Gonzalez, A., et al., Second solid cancers after radiation therapy: a

systematic review of the epidemiologic studies of the radiation dose-response

relationship. Int J Radiat Oncol Biol Phys, 2013. 86(2): p. 224-33.

82. Sautter-Bihl, M.L., et al., One life saved by four prevented recurrences? Update of the

Early Breast Cancer Trialists confirms: postoperative radiotherapy improves survival

after breast conserving surgery. Strahlenther Onkol, 2012. 188(6): p. 461-3.

83. Rodel, F., et al., Radiobiological mechanisms in inflammatory diseases of low-dose

radiation therapy. Int J Radiat Biol, 2007. 83(6): p. 357-66.

84. Arenas, M., et al., Anti-inflammatory effects of low-dose radiotherapy. Indications,

dose, and radiobiological mechanisms involved. Strahlenther Onkol, 2012. 188(11): p.

975-81.

40

85. Koj, A., Termination of acute-phase response: role of some cytokines and anti-

inflammatory drugs. Gen Pharmacol, 1998. 31(1): p. 9-18.

86. Denham, J.W. and M. Hauer-Jensen, The radiotherapeutic injury--a complex 'wound'.

Radiother Oncol, 2002. 63(2): p. 129-45.

87. Schaue, D. and W.H. McBride, Links between innate immunity and normal tissue

radiobiology. Radiat Res, 2010. 173(4): p. 406-17.

88. McBride, W.H., et al., A sense of danger from radiation. Radiat Res, 2004. 162(1): p.

1-19.

89. Chua, M.L. and K. Rothkamm, Biomarkers of radiation exposure: can they predict

normal tissue radiosensitivity? Clin Oncol (R Coll Radiol), 2013. 25(10): p. 610-6.

90. Brock, W.A., et al., Fibroblast radiosensitivity versus acute and late normal skin

responses in patients treated for breast cancer. Int J Radiat Oncol Biol Phys, 1995.

32(5): p. 1371-9.

91. Burnet, N.G., et al., Prediction of normal-tissue tolerance to radiotherapy from in-

vitro cellular radiation sensitivity. Lancet, 1992. 339(8809): p. 1570-1.

92. Geara, F.B., et al., Prospective comparison of in vitro normal cell radiosensitivity and

normal tissue reactions in radiotherapy patients. Int J Radiat Oncol Biol Phys, 1993.

27(5): p. 1173-9.

93. Johansen, J., et al., Evidence for a positive correlation between in vitro

radiosensitivity of normal human skin fibroblasts and the occurrence of subcutaneous

fibrosis after radiotherapy. Int J Radiat Biol, 1994. 66(4): p. 407-12.

94. Greve, B., et al., Evaluation of different biomarkers to predict individual

radiosensitivity in an inter-laboratory comparison--lessons for future studies. PLoS

One, 2012. 7(10): p. e47185.

95. Peacock, J., et al., Cellular radiosensitivity and complication risk after curative

radiotherapy. Radiother Oncol, 2000. 55(2): p. 173-8.

96. Russell, N.S., et al., Low predictive value of intrinsic fibroblast radiosensitivity for

fibrosis development following radiotherapy for breast cancer. Int J Radiat Biol, 1998.

73(6): p. 661-70.

97. Barber, J.B., et al., Relationship between in vitro chromosomal radiosensitivity of

peripheral blood lymphocytes and the expression of normal tissue damage following

radiotherapy for breast cancer. Radiother Oncol, 2000. 55(2): p. 179-86.

98. Terzoudi, G.I., et al., G2-checkpoint abrogation in irradiated lymphocytes: A new

cytogenetic approach to assess individual radiosensitivity and predisposition to

cancer. Int J Oncol, 2009. 35(5): p. 1223-30.

99. Pantelias, G.E. and G.I. Terzoudi, A standardized G2-assay for the prediction of

individual radiosensitivity. Radiother Oncol, 2011. 101(1): p. 28-34.

100. Hirst, D.G. and T. Robson, Molecular biology: the key to personalised treatment in

radiation oncology? Br J Radiol, 2010. 83(993): p. 723-8.

101. Twardella, D. and J. Chang-Claude, Studies on radiosensitivity from an

epidemiological point of view - overview of methods and results. Radiother Oncol,

2002. 62(3): p. 249-60.

102. Talbot, C.J., et al., A replicated association between polymorphisms near TNFalpha

and risk for adverse reactions to radiotherapy. Br J Cancer, 2012. 107(4): p. 748-53.

103. Baumann, M., T. Holscher, and A.C. Begg, Towards genetic prediction of radiation

responses: ESTRO's GENEPI project. Radiother Oncol, 2003. 69(2): p. 121-5.

104. Barnett, G.C., et al., No association between SNPs regulating TGF-beta1 secretion

and late radiotherapy toxicity to the breast: results from the RAPPER study. Radiother

Oncol, 2010. 97(1): p. 9-14.

41

105. Burnet, N.G., et al., RAPPER: the radiogenomics of radiation toxicity. Clin Oncol (R

Coll Radiol), 2013. 25(7): p. 431-4.

106. Ho, A.Y., et al., Genetic predictors of adverse radiotherapy effects: the Gene-PARE

project. Int J Radiat Oncol Biol Phys, 2006. 65(3): p. 646-55.

107. Iwakawa, M., et al., [RadGenomics project]. Nippon Igaku Hoshasen Gakkai Zasshi,

2002. 62(9): p. 484-9.

108. West, C. and B.S. Rosenstein, Establishment of a radiogenomics consortium.

Radiother Oncol, 2010. 94(1): p. 117-8.

109. Henriquez Hernandez, L.A., et al., Constitutive gene expression profile segregates

toxicity in locally advanced breast cancer patients treated with high-dose

hyperfractionated radical radiotherapy. Radiat Oncol, 2009. 4: p. 17.

110. Mayer, C., et al., A radiation-induced gene expression signature as a tool to predict

acute radiotherapy-induced adverse side effects. Cancer Lett, 2011. 302(1): p. 20-8.

111. Rieger, K.E., et al., Toxicity from radiation therapy associated with abnormal

transcriptional responses to DNA damage. Proc Natl Acad Sci U S A, 2004. 101(17):

p. 6635-40.

112. Finnon, P., et al., Correlation of in vitro lymphocyte radiosensitivity and gene

expression with late normal tissue reactions following curative radiotherapy for

breast cancer. Radiother Oncol, 2012. 105(3): p. 329-36.

113. Rodningen, O.K., et al., Radiation-induced gene expression in human subcutaneous

fibroblasts is predictive of radiation-induced fibrosis. Radiother Oncol, 2008. 86(3): p.

314-20.

114. Svensson, J.P., et al., Analysis of gene expression using gene sets discriminates cancer

patients with and without late radiation toxicity. PLoS Med, 2006. 3(10): p. e422.

115. Wiebalk, K., et al., In vitro radiation-induced expression of XPC mRNA as a possible

biomarker for developing adverse reactions during radiotherapy. Int J Cancer, 2007.

121(10): p. 2340-5.

116. Badie, C., et al., Aberrant CDKN1A transcriptional response associates with

abnormal sensitivity to radiation treatment. Br J Cancer, 2008. 98(11): p. 1845-51.

117. Metheetrairut, C. and F.J. Slack, MicroRNAs in the ionizing radiation response and in

radiotherapy. Curr Opin Genet Dev, 2013. 23(1): p. 12-9.

118. Azimzadeh, O., M.J. Atkinson, and S. Tapio, Proteomics in radiation research:

present status and future perspectives. Radiat Environ Biophys, 2013.

119. Lacombe, J., et al., Proteomic approaches to identify biomarkers predictive of

radiotherapy outcomes. Expert Rev Proteomics, 2013. 10(1): p. 33-42.

120. Cai, X.W., et al., Plasma proteomic analysis may identify new markers for radiation-

induced lung toxicity in patients with non-small-cell lung cancer. Int J Radiat Oncol

Biol Phys, 2010. 77(3): p. 867-76.

121. Cai, X.W., et al., Baseline plasma proteomic analysis to identify biomarkers that

predict radiation-induced lung toxicity in patients receiving radiation for non-small

cell lung cancer. J Thorac Oncol, 2011. 6(6): p. 1073-8.

122. Ozsahin, M., et al., Rapid assay of intrinsic radiosensitivity based on apoptosis in

human CD4 and CD8 T-lymphocytes. Int J Radiat Oncol Biol Phys, 1997. 38(2): p.

429-40.

123. Fang, Z., et al., Low levels of ATM in breast cancer patients with clinical

radiosensitivity. Genome Integr, 2010. 1(1): p. 9.

124. Goldsby, R.A., et al., Kuby immunology. 4th ed2000, New York: W.H. Freeman. xxv,

670 p.

125. Muller, T., et al., Sense and nonsense of pathway analysis software in proteomics. J

Proteome Res, 2011. 10(12): p. 5398-408.

42

126. Mayburd, A.L., et al., Ingenuity network-assisted transcription profiling:

Identification of a new pharmacologic mechanism for MK886. Clin Cancer Res, 2006.

12(6): p. 1820-7.

127. Franceschini, A., et al., STRING v9.1: protein-protein interaction networks, with

increased coverage and integration. Nucleic Acids Res, 2013. 41(Database issue): p.

D808-15.