List of Papers. This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

Transcription

1

2

3 "A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects." R. A. Heinlein

4

5 List of Papers This thesis is based on the following papers, which are referred to in the text by their Roman numerals. I Qvarnström OF*, Simonsson M*, Johansson KA, Nyman J, Turesson I. (2004) DNA double strand break quantification in skin biopsies. Radiother Oncol, 72: II Simonsson M*, Qvarnström F*, Nyman J, Johansson KA, Garmo H, Turesson I. (2008) Low-dose hypersensitive H2AX response and infrequent apoptosis in epidermis from radiotherapy patients. Radiother Oncol, 88: III Qvarnström OF, Simonsson M, Nyman J, Hermansson I, Book M, Johansson KA, Turesson I. Low-dose hypersensitivity of initial and persistent DSB foci in epidermis throughout fractionated radiotherapy treatment. Manuscript IV Turesson I, Nyman J, Qvarnström OF, Simonsson M, Book M, Hermansson I, Sigurdardottir S, Johansson KA, A low-dose hypersensitive keratinocyte response to fractionated radiotherapy is mediated by growth arrest and apoptosis. Manuscript * These authors are considered as joint first authors. Reprints were made with permission from the respective publishers.

6

7 Contents 1. Introduction DNA-damage response DNA double strand breaks DNA damage response at a tissue level Cellular response to radiation Risks of low-dose radiation Hyperradiosensitivity and induced radiation resistance Radiotherapy Epidermal skin tissue Tissue structure of the epidermis Aims Material and Methods Skin biopsies Skin biopsy acquisition Dosimetry Histochemistry Tissue staining procedures Quantification procedures Microscopy procedures Quantification of DSB foci by digital image analysis Quantification of p21 by digital image analysis Manual quantifications of immunohistochemical stainings Statistics Results and Discussion Paper I: DNA double strand break quantification in skin biopsies Paper II: Low-dose hypersensitive H2AX response and infrequent apoptosis in epidermis from radiotherapy patients Paper III: Low-dose hypersensitivity of initial and persistent DSB foci in epidermis throughout fractionated radiotherapy treatment Paper IV: A low-dose hypersensitive keratinocyte response to fractionated radiotherapy is mediated by growth arrest and apoptosis Conclusions...41

8 6. Acknowledgements References...45

9 Abbreviations 53BP1 p53 binding protein 1. Alternative name: TP53BP1 ATM Ataxia telangiectasia mutated ATR Ataxia telangiectasia and Rad3 related BCD Basal cell density BKD Basal keratinocyte density CCD Charge coupled device CT Computed tomography DAB 3,3'-Diaminobenzidine DAPI 4',6-diamidino-2-phenylindole DDR DNA damage response DIA Digital image analysis DNA Deoxyribonucleic acid DNA-PK DNA protein kinase DSB Double strand break H2AX Histone H2AX, Alternative names: H2AFX or H2A.x H2AX Histone H2AX phosphorylated on serine 139 HIER Heat induced epitope retrieval HC Histochemistry HRP Horseradish peroxidase HRS Hyperradiosensitivity HTA28 Histone H3 phosphorylated on S28 HTX Hematoxylin IF Immunofluorescence IHC Immunohistochemistry IMRT Intensity-modulated radiotherapy IR Ionising radiation IRR Induced radiation resistance LET Linear energy transfer p21 Cyclin-dependent kinase inhibitor 1A. Alternative names: CDKN1A or CIP1 p53 protein 53, Alternative name: TP53 PARP-1 Poly (ADP-ribose) polymerase 1. Alternative names: ADPRT or PPOL

10 cparp-1 PAS RGB RT TA UV Cleaved PARP-1 Periodic acid-schiff Red, Green, Blue Radiotherapy Transient Amplifying Ultraviolet

11 1. Introduction 1.1 DNA-damage response The cellular response to DNA damage is essential for maintaining normal functions of cells and tissues. Cellular damage is frequently inflicted by both internal and external factors, requiring swift and accurate repair processes. The importance of cellular DNA damage response (DDR) is stressed by numerous clinical and in vitro observations. Many results demonstrate that inherited or introduced functional defects in DDR pathways can result in an increased risk of tumour development [Celeste et al. 2002, Celeste et al. 2003#1, Bassing et al. 2002, Bassing et al. 2003, Kastan et al. (rev) 2008]. Furthermore, an exaggerated DDR can be observed during the early stages of tumour development, suggesting that these mechanisms can function as an important anti-cancer barrier [Gorgoulis et al. 2005, Bartkova et al. 2005, Halazonetis et al. (rev) 2008]. New techniques and recently identified molecular markers have provided increasingly more detailed information about the cellular responses to DNA damage. It is now possible to detect and quantify specific effects; even following induction of very low amounts of damage [Bonner et al. (rev) 2008]. Surrogate markers for DNA double strand breaks (DSBs), such as H2AX and 53BP1, provide detailed information about how these serious DNA lesions are handled in individual cells. Molecular markers for cellular proliferation and cell cycle checkpoints further extend the possibility of studying the DDR course of events in cells, within their tissue, at clinically relevant damage levels. The papers of this thesis exemplify how these methodological approaches can be applied to determine DDR in epidermal skin samples taken from patients undergoing fractionated radiotherapy (Figure 1). 11

12 Figure 1. A summary of the main DDR events in proliferating compartments of normal tissues during fractionated radiotherapy treatments and how they were assessed in this thesis. Biopsies were taken from regions of the skin exposed to daily fractions of low doses in a range from Gy. DNA repair was assessed by quantification of DSB foci to determine the effect following delivery of individual fractions as well as overall changes throughout 7 weeks of treatment. If required, cell cycle checkpoints may be activated to provide extra time for repair. This step was assessed by quantification of the p21 kinase inhibitor, which is involved in activation of G 1 -arrest. The success or failure of the repair process will determine whether the individual cells will continue to proliferate or if they will be eliminated. This was assessed by markers for mitosis and apoptosis. The overall outcome could be detected as changes in tissue cell density, in our case quantified as the number of keratinocytes per mm of the epidermal basal membrane DNA double strand breaks Histones are nuclear proteins involved in regulation and compaction of eukaryotic DNA. In human cells, DNA is wrapped around octameres consisting of two copies of each of the histone proteins H2A, H2B, H3 and H4, forming fundamental chromatin units called nucleosomes [Alberts et al. (rev) 2002]. Some of the core histones are present in different variants. One of these, the H2A variant H2AX, has been subject to intensive research since 12

13 its involvement in DNA DSB-signalling first was discovered [Rogakou et al. 1998, Rogakou et al. 1999, Fernandez-Capetillo et al. (rev) 2004]. H2AX possesses a specific SQ-motif, situated close to its carboxyl terminus. The serine 139 residue within this motif is rapidly targeted for phosphorylation following DSB induction. The phosphorylation response involves large regions of the DNA flanking the break, providing an amplified signal that is detectable within minutes after damage induction [Rogakou et al. 1998, Rogakou et al. 1999]. The phosphorylated form of H2AX, referred to as H2AX, can be visualised within individual cell nuclei as distinct foci structures using antibodies specific for this phosphorylated form [Rogakou et al. 1999, Paull et al. 2000]. Since a single DSB appears to be sufficient for foci formation, H2AX has been widely employed as a marker for detection of very low levels of DSB damage [Chen et al. 2000, Sedelnikova et al. 2002, Rothkamm et al. 2003, Bonner et al. (rev) 2008, Nakamura et al. 2008]. The ataxia telangiectasia mutated (ATM) protein kinase is involved in several DNA damage activated pathways [Savitsky et al. 1995, Kastan et al. (rev) 2008]. Following induction of DSBs, ATM is also the main kinase responsible for the rapid phosphorylation of H2AX [Burma et al. 2001, Stiff et al. 2004]. Other kinases, such as ATR and DNA-PK, are however capable of phosphorylating H2AX [Andegeko et al. 2001, Ward et al. 2001, Stiff et al. 2004, Wang et al. 2005, Falck et al. 2005]. Phosphorylated H2AX is dispensable for the initial recognition of the break but appears to have a function as a platform for recruiting other factors involved in the repair process [Paull et al. 2000, Celeste et al. 2003#2, Kinner et al. (rev) 2008]. The kinetics of the removal of H2AX foci and their connection to DSB repair is somewhat unclear. It has however been suggested that H2AX foci, induced by clinically relevant doses, are removed from sites of DSBs shortly after the breaks are repaired [Rothkamm et al. 2003, Chowdhury et al. 2005, Bouquet et al. 2006, Bonner et al. (rev) 2008]. 53BP1 is another DSB-signalling protein with foci-kinetics similar to those observed for H2AX [Schultz et al. 2000, Anderson et al. 2001, Kao et al. 2003, Ward et al. 2003]. 53BP1 is not only hyperphosphorylated upon DSB induction, it also re-localises to the sites of the lesions within minutes following damage induction [Schultz et al. 2000, Anderson et al. 2001, Rappold et al. 2001]. Several studies have thus described co-localisation of H2AX and 53BP1 foci in a wide range of cell types [Schultz et al. 2000, Rappold et al. 2001, Lee et al. 2005, Markova et al. 2007]. 53BP1 does not require phosphorylation of H2AX for foci formation although it is not retained at the DSB sites if H2AX is absent [Celeste et al. 2003#2]. Thus, 53BP1 foci formation appears to follow a parallel but interacting pathway to H2AX foci formation [Chronis et al. (rev) 2007, Zgheib et al. 2009]. It has been suggested that 53BP1 has a dual role in the DSB damage response; as an initial DSB sensor involved in activation of ATM and further 13

14 downstream as a mediator, by contributing to a scaffold structure involved in recruiting additional ATM kinase to the DSB sites [Huyen et al. 2004, Chronis et al. (rev) 2007]. The ability of 53BP1 to relocate to sites of DSBs is facilitated by a series of histone methylations and ubiquitylations [Huyen et al. 2004, Botuyan et al. 2006, Huen et al. 2007, Kolas et al. 2007, Mailand et al. 2007]. It is possible that DSB-induced structural changes of the chromatin make these modified histone residues accessible to 53BP1 for binding [Huyen et al. 2004, Sanders et al. 2004, Chronis et al. (rev) 2007, Zgheib et al. 2009] DNA damage response at a tissue level Cell cycle checkpoints are an important part of the DDR mechanisms, capable of stalling cycling cells and preventing cells with unrepaired DNA damage from continued proliferation. Failure to trigger these checkpoints can, in normal tissues, lead to various types of cell death and will increase the risk for introduction of carcinogenic alterations in surviving cells. A multitude of checkpoints regulate cell cycle progression but the sensitive G 1 -checkpoint, governed by the critical p53 tumour suppressor, is suggested to be superior with regards to its ability to detect low levels of damage [Huang et al. 1996, Linke et al. 1997, Löbrich et al. (rev) 2007]. This G 1 -checkpoint is initially triggered by phosphorylation of p53 by ATM, resulting in an overall increase in transcriptional activity of p53 [Banin et al. 1998, Canman et al. 1998]. The phosphorylated p53 induces an increased expression of the cyclin dependent kinase inhibitor p21, which in turn directly inhibits the activity of cyclind-cdk4/6 and cycline-cdk2, resulting in cell cycle arrests [el-deiry et al. 1993, Harper et al. 1993, Weinberg et al. (rev) 2002]. Checkpoint-related nuclear p21 expression can then be readily detected in epidermal skin by immunohistochemistry (IHC) [Turesson et al. 2001]. Severe damage or failure to induce checkpoints may result in elimination of cells by cell death mechanisms such as apoptosis. The importance of apoptotic cell death in the overall DDR of normal tissues, such as the epidermis, is however unclear [Steel (rev) 2001]. Senescence and differentiation are additional possible outcomes of DNA damage. The frequency of apoptotic events in a tissue can be estimated by quantification of a subset of intensely H2AX-stained cells, signalling massive DNA fragmentation [Rougakou et al. 2000, Holubec et al. 2005, Mukherjee et al. 2006, Nowak et al. 2006]. Another important factor in the tissue DDR is the regulation of mitosis. The frequency of dividing cells provides information about how the damage affects proliferation rates and overall maintenance of the tissue integrity. Mitotic events can be visualised and quantified in epidermis by the monoclonal HTA28 antibody, which detects a phosphorylation of histone H3 at serine 28 [Hirata et al. 2004]. 14

15 Cell cycle checkpoints, apoptosis and mitosis are factors that have an influence on the overall tissue response to DNA damage. Regulations of these key mechanisms will influence, to varying degrees, tissue responses observed during clinical cancer treatments, such as fractionated radiotherapy where tissues are exposed to repeated insult. The combined effects of the damage regulatory mechanisms are exemplified in the normal skin epidermis where a decrease in cell density can be observed during a fractionated radiotherapy treatment course [Joiner et al. 2001, Harney et al. 2004#1]. The cellular depletion can be assessed by quantifications of basal cell density (BCD) or basal keratinocyte density (BKD). This further exemplifies how the effect of DNA damage responses can be studied at a tissue level. 1.2 Cellular response to radiation DNA damage induced by radiation can potentially cause disease and cancer. At the same time, radiation also provides powerful tools for treating cancers, such as in radiotherapy. This dual role is illuminated by the intricate molecular pathways involved in cellular response to DNA damage [Helleday et al. (rev) 2008, Kinner et al. (rev) 2008]. The response can be dependent on cell and tissue type as well as the nature and magnitude of the damage [van der Kogel (rev) 1997, Gudkov et al. (rev) 2003]. The complexity and dose dependencies of these DDR mechanisms are further emphasised by phenomena such as hyperradiosensitivity (HRS), observed following exposure to very low doses [Bonner et al. (rev) 2004, Marples et al. (rev) 2008#1, Marples et al. (rev) 2008#2]. Overall, the observed cell and tissue differences further stress the importance of studies where cellular DDR is observed directly in the tissues, investigating the dose response effects following exposure to clinically relevant dose levels Risks of low-dose radiation The risks and consequences associated with exposure to very low doses of radiation are difficult to assess on a human population level as well as at a molecular level. On a population level, a linear extrapolation based on effects observed for higher doses are generally applied when risks associated with low doses are estimated [Brenner et al. (rev) 2003]. A range of epidemiological studies are also available on population cohorts such as atomic bomb survivors and radiation workers in the nuclear industry [Cardis et al. 2007, Preston et al. 2008]. At a molecular level, the discovery of sensitive markers for DSBs and other DDR related proteins have provided greater insight into the biological responses to low doses of radiation [Rothkamm et al. 2003, Bonner et al. (rev) 2008, Kastan et al. (rev) 2008]. Improvements in cell survival assays have also been important for studying low dose phe- 15

16 nomena such as HRS [Marples et al. (rev) 2008#1]. Further understanding of the molecular mechanisms behind low-dose DDR will provide more detailed information, which will help to achieve a more accurate estimation of the risks associated with exposure to very low doses. Understanding the cellular response to low-dose exposure is further stressed by the increased use of clinical techniques involving exposure to low doses of radiation. An example is the increased use of computed tomography (CT), which has been beneficial from a diagnostics perspective but also brings up questions concerning potential risks of the associated lowdose exposures [Brenner et al. 2007]. In radiotherapy, the use of new techniques such as intensity-modulated radiotherapy (IMRT), is also associated with increased low-dose exposure of larger volumes of the body [Honoré et al. 2006]. Understanding the effects of low-dose radiation is not only important from a clinical perspective. Assessment of the possible risks of low-dose exposure is also applicable from an environmental perspective, including unusual events such as nuclear incidents and space travel [Brenner et al. (rev) 2003, Löbrich et al. (rev) 2007] Hyperradiosensitivity and induced radiation resistance HRS can be described as an exaggerated sensitivity to low levels of low- LET radiation damage and has been observed in a wide range of cell types and tissues [Joiner et al. 1986, Marples et al. 1993, Wouters et al. 1994, Joiner et al. 1996, Wouters et al. 1996, Wouters et al. 1997, Marples et al. (rev) 2008#1, Marples et al. (rev) 2008#2]. Clonogenic cell survival assays display HRS as an initial depression in the cell survival dose-response curve. This corresponds to elevated levels of cell-kill per dose unit in the HRS dose region. For slightly higher doses, in the range ~ Gy, an induced radiation resistance (IRR) effect can be observed. In clonogenic cell survival assays, IRR results in levelled-out or increased cell survival following an increase in dose. The mechanisms underlying HRS/IRR are not fully understood. It has however been suggested that a dose-dependent response of an early-g 2 cell cycle checkpoint could be of importance [Marples et al. (rev) 2004, Marples et al. (rev) 2008#2]. This is supported by observations suggesting that low levels of radiation damage are insufficient to fully activate the checkpointinvolved ATM kinase [Bakkenist et al. 2003, Buscemi et al. 2004, Löbrich et al. (rev) 2005, Wykes et al. 2006]. A failure to activate the specific G 2 checkpoint will result in mitotic entry of damaged cells and consequently lead to an increased cell death. This dose-dependent checkpoint response is a possible explanation for the suppressed clonogenic cell survival observed for cells exposed to doses in the HRS dose range. Larger doses will result in a transition from HRS to IRR, which also corresponds to the dose-dependent 16

17 activation of ATM [Bakkenist et al. 2003]. Recent studies have also shown that the HRS effect translates to increased apoptosis for low doses [Enns et al. 2004, Krueger et al. 2007#1, Krueger et al. 2007#2] (Figure 2). Apart from observations on cells in culture, HRS/IRR related effects have also been observed in various tissues [Joiner et al. 1986, Joiner et al. 1988, Joiner et al. 2001]. In human epidermis, previously published results revealed a HRS-like effect in BCD, for doses below ~0.4Gy [Joiner et al. 2001, Harney et al. 2004#1]. Throughout this thesis and in the included papers, HRS-related effects, referring to endpoints other than cell survival, are referred to as low-dose hypersensitivity. From a clinical perspective, HRS/IRR effects create an opportunity to improve radiotherapy treatments [Lin et al. 2005, Honoré et al. 2006]. Increased cell-kill in the tumours, as well as reduced side effects in surrounding tissues, are possibilities that must be considered. Side-effects in normal tissues are limiting factors in radiotherapy. They occur since normal tissues, surrounding the targeted tumours, will inevitably be exposed to small amounts of radiation during treatment. This exposure is often within the HRS/IRR dose range. Knowledge regarding normal tissue dose response, in relation to the dose response of tumour cells, could be exploited with the aim of protecting the normal tissue while maximising the cell-kill within the tumour. Clinical studies have been able to confirm possible advantages of treatment regimes utilising the effects of HRS/IRR [Joiner et al. 2001, Harney et al. 2004#1, Harney et al. 2004#2]. On the other hand, there are a number of animal studies where no significant advantages could be observed in tumour control [Krause et al. 2003, Krause et al. 2005#1, Krause et al. 2005#2]. The use of relatively large fraction sizes in these animal studies (0.4Gy or higher) could however be a reason for the lack of effect. 17

18 Figure 2. Possible mechanisms behind HRS/IRR. Even though ATM-kinase responds rapidly to DNA damage, recent findings suggest that doses under ~0.3Gy fail to fully activate this important DNA-damage-involved kinase. Reduced ATM activity leads to abrogated activation of a checkpoint in the early G 2 phase allowing damaged cells to enter mitosis. This will in turn lead to increased levels of apoptosis and decreased clonogenic cell survival within the HRS dose range. A recovery in clonogenic cell survival can be observed in the IRR dose range, interpreted as corresponding to an increased activation of ATM. The illustrated dose-response relationships are interpretations of data from recent publications, presented as ratios normalised to background levels [Enns et al. 2004, Marples et al. (rev) 2004, Krueger et al. 2007#1, Krueger et al. 2007#2] Radiotherapy Besides surgery, radiotherapy is considered as the most important cancer treatment [Steel (rev) 2002]. The basic strategy behind radiotherapy treatments is to kill tumour cells by ionising radiation (IR), which penetrates the tissues and induces DNA-damage. Some of the surrounding healthy normal tissues will inevitably be damaged in this process, limiting how much damage is possible to induce in the tumour tissue. Each radiotherapy treatment is therefore carefully planned to reduce the risk of inducing side effects in the normal tissues while keeping the cell-kill in the tumour as high as possible. Simulation and treatment planning procedures are employed to customise treatments to individual patients. In brief, the simulation procedure involves positioning of patients in relation to the treatment fields, ensuring that the positions are fixed and possible 18

19 to maintain throughout the treatment course. In this process, images of the tumour area are commonly achieved by CT or other imaging techniques. These images can be further used in a treatment planning system where tumour and surrounding structures of concern are outlined to determine the details of the radiation delivery. A common way of delivering radiation in the clinic is by external beam therapy. An internal target, such as the prostate, can in this way be targeted with radiation fields from several directions, merging in the tumour to achieve the desired target dose levels. By using several fields, the absolute dose levels delivered to the surrounding healthy normal tissues can be reduced. Megavoltage x-rays, a type of low-let radiation, produced in linear accelerators are a common source of radiation in external beam therapy. The total radiation dose aimed at a tumour can be divided and delivered over several weeks of treatment. This is referred to as fractionation and works by taking advantage of differences in DDR between normal tissues and tumour tissues. In theory, treatment regimes with daily fractions allow normal tissues to repair damage to a larger extent than rapidly dividing tumour cells with defective repair mechanisms. Fractionation related changes in cell cycle distributions may also favour the relative survival of normal cells over tumour cells [Mundt et al. (rev) 2003]. In the specific case of normal epidermal skin, fractionated radiotherapy results in general reductions in cell numbers, and a continuous thinning out of the epidermal layer. This is reflected by a reduction in proliferating cells in the basal layer. Several studies have described these tissue responses by quantification of BCD in skin biopsies from patients undergoing radiotherapy [Nyman et al. 1994, Joiner et al. 2001, Harney et al. 2004#1]. A similar but more specific approach is described in Paper IV, where quantification of the tissue response were performed by counting only the basal keratinocytes, BKD, thereby avoiding the inclusion of the melanocytic cell compartment. 1.3 Epidermal skin tissue Further insight into the morphology and function of normal tissues could prove crucial in understanding how abnormalities arise. Epidermal samples from patients undergoing radiotherapy offer an in vivo perspective of DDR, observable directly in the tissue itself. Furthermore, the structural subcompartments of the epidermis make it possible to study DDR in defined cell compartments as well as observing the damage response at a tissue level. The accessibility of the epidermis, as well as its structural properties, contribute to making this tissue an advantageous model for studying radiation induced damage responses in normal tissues. The skin model also provides a possibility to use careful clinical dosimetry to determine the amount of initial damage induction. 19

20 1.3.1 Tissue structure of the epidermis The epidermis is a multilayered epithelium serving as the outermost protective barrier of the skin (Figure 3). The layered structure is subject to a constant turnover, as terminally differentiated keratinocytes are continually shed from the skin surface. The turnover of keratinocytes, the major cell type in epidermal skin, is maintained by proliferation of epidermal stem cells [Watt et al. (rev) 2006, Blanpain et al. (rev) 2009]. These stem cells undergo controlled cell divisions, thereby maintaining the stem cell pool as well as repopulating the differentiating cell compartments. It is suggested that a population of transient amplifying (TA) cells also resides in the basal layer, providing extended possibilities for proliferation control. These cells are capable of a few cycles of cell division before progressing further through the differentiation process [Jones et al. (rev) 2007]. A population of melanocytes also reside in the basal layer. Melanocytes produce melanin, which serve as a UV-protecting agent in the skin. The epidermis is situated directly on top of a connective tissue skin layer referred to as the dermis. The epidermis relies on the supportive function of the dermis and is also dependent on the dermal blood vessels for nutrient supplies and waste removal. 20

21 Figure 3. Top layers of human skin. The sub-compartments (strata) of the epidermal layer reflect the differentiation pathways of keratinocytes. Epidermal stem cells, TAcells and melanocytes reside in the basal layer, stratum basale. Continuous cell divisions in the basal layer re-fill the differentiating cell pools further up. In a continuous movement towards the skin surface, the keratinocytes will finally be incorporated in the top layer, stratum corneum, where they eventually are shed. The upper part of the dermal layer consists mainly of connective tissues interspersed with fibroblasts, immune cells and small blood vessels coated by endothelial cells. 21

22 2. Aims Detection of foci structures within cell nuclei enable detailed studies of DSB damage where individual breaks can be visualised and quantified. Complex immunostaining patterns and background variations can however make the foci structures indistinct and difficult to interpret by eye. In such cases, it can prove advantageous to employ automated quantification approaches based on digital image analysis. The aim of the first study was to set up and evaluate a semi-automated foci quantification procedure for skin biopsies taken from patients undergoing radiotherapy. The procedure therefore had to be robust and sensitive enough to detect very low levels of foci in formalin fixed materials. The aim of the second study was to apply the semi-automated foci quantification procedure to record epidermal dose-response relationships in more detail. More specifically, to determine the H2AX foci dose response in normal epidermis during the first week of a radiotherapy treatment course. A second aim was to investigate whether H2AX can be employed as a marker for apoptosis. Furthermore, the experiments were set up to elucidate whether any of our end-points displayed low-dose hypersensitivity corresponding to the HRS/IRR effects observed in various cell lines [Marples et al. (rev) 2008#1]. The aim of the third study was to expand the results from the second study by investigating the DSB foci dose response throughout the treatment course and to quantify possible foci accumulation and occurrence of persistent foci between fractions. Furthermore, the aim was to investigate if DSB low-dose hypersensitivity is preserved throughout the treatment and between fractions. A quantitative and qualitative comparison of the DSB-markers H2AX and 53BP1 was an additional aim. The fourth study is a continuation of the previous studies, which described low-dose hypersensitivity in DSB foci dose response, preserved throughout a fractionated radiotherapy treatment course. The aim of the fourth study was to investigate epidermal dose-response relationships for an additional set of end-points. Four key parameters, downstream in the DDR pathways, were chosen: p21 induction, apoptosis, mitosis and basal keratinocyte density (BKD). A second aim of this study was to compare digital image analysis quantification of DAB/HTX stainings with traditional manual counting by eye. 22

23 The final outcome of cellular response to severe DNA damage is of crucial importance for the overall risk of introducing malignancies. A DDR system where severely damaged cells are eliminated by cell death could in many ways be advantageous from a carcinogenesis perspective. A response system with a higher degree of cell survival could be beneficial for tissue structure maintenance, but will at the same time increase the risk of repairrelated alterations of the genome and increase the risk of carcinogenesis. It is therefore of importance to elucidate how these DDR mechanisms interact and to determine which of these mechanisms is predominant in normal tissues such as the epidermis. The papers of this thesis highlight this issue and provide new insight into normal tissue DDR. 23

24 3. Material and Methods 3.1 Skin biopsies Skin biopsies taken in clinical practice offer a direct opportunity to study human normal tissue responses and cellular interactions in a tissue context. The ability to accurately determine dose distributions on the skin makes these biopsies highly suitable for studying normal tissue dose response during radiotherapy Skin biopsy acquisition The biopsy material in the included papers was acquired between 1996 and 2005, from prostate cancer patients undergoing fractionated radiotherapy treatment. The patients received daily fractions throughout a seven week treatment course with interruptions over weekends. Each fraction delivered 2Gy to the prostate target volume using 3- or 4-field techniques where the prostate is exposed to photon irradiation from several directions. During these treatments it is inevitable that surrounding normal tissues, such as the skin, also are exposed to varying degrees. Collecting skin punch biopsies during a radiotherapy treatment course can, in this way, provide samples of the in vivo skin dose response. Skin punch biopsies (3mm) corresponding to a range of low doses ( Gy) could in our case be collected by acquiring biopsies within and slightly outside the lateral fields (Figure 4). Biopsies were collected from a total of 115 patients throughout the treatment course. At all sampling time points, biopsies were taken within the fields, providing biopsies corresponding to doses of 0.45 and 1.1Gy. The majority of the results in this thesis were however based on extended biopsy sets, where four biopsies were collected per sampling occasion (Table 1). The extended biopsy sets were taken from 90 of the patients and collected from areas of the skin exposed to fraction doses of Gy providing a material where individual dose response can be determined. Each of the 90 patients was sampled at least once, providing a total of 140 biopsy sets. The biopsy sets were collected so that dose-response relationships throughout the treatment, as well as between fractions, could be investigated. To enable assessment of dose response in between fractions, the extended biopsy sets were collected at different time points following fractions: 30 24

25 minutes, 3 hours and 24 hours. Unexposed control biopsies were collected from all patients. Table 1. Biopsies acquired for dose response determinations. Group, based on weeks into treatment Total no. of collected biopsy sets Number of fractions 0 weeks week weeks weeks weeks weeks Days into treatment Dosimetry Accurate and reliable dose determinations are a prerequisite for dose response studies involving low-dose exposures. In this case, modifications of the available clinical dosimetry techniques were applied to ensure an optimal accuracy. Doses corresponding to specific biopsy locations were, for each of the patients, initially calculated in the radiotherapy treatment planning system. The dose determinations were further improved by adjustments derived from series of ionisation chamber measurements based on the individual field setups. Dose compensations were also made after the biopsies had been taken. These were based on the actual biopsy locations on the skin. The locations were compared with the dose plan to compensate for possible deviations from the intended biopsy locations. 25

26 Figure 4. Skin punch biopsy collection. Locations for biopsy collection corresponded to doses of 0.1Gy, 0.2Gy and 0.45Gy, taken within and just outside one of the lateral fields (shaded area). Regions of the skin within the opposed lateral field provided biopsies corresponding to skin doses of 1.1Gy, obtained using a 5mm bolus. 3.2 Histochemistry Visualisation of molecular targets in cells within tissues is possible using various histochemistry (HC) techniques. In the particular case of immunohistochemistry (IHC), the histochemical localisation of target specific antibodies are visualised in thin tissue sections by various staining procedures. The IHC techniques enable examination of cellular protein expression patterns in the context of their tissue localisation. Conventional bright field microscopy is, in many cases, used for examination of HC stained slides. Alternative techniques are also possible, such as immunofluorescence (IF) detection by fluorescence microscopy Tissue staining procedures Tissues collected in the clinic are commonly stored by immediate formalin fixation followed by dehydration and paraffin embedding. This procedure aims to preserve and protect proteins and other tissue structures by creating covalent bonds, anchoring the different structures to each other. To be able to analyse these archival samples by histochemical techniques, a range of 26

27 procedures are employed to restore the original tissue structures and protein conformations. As a first step, thin sections of the paraffin embedded tissues are cut in a microtome and mounted on microscopy slides. The slides are further deparaffinised, rehydrated and often subjected to an epitope retrieval procedure. Heat induced epitope retrieval (HIER) is a commonly used approach whereby covalent bonds created in the fixation process are resolved, thus restoring many of the tissues original epitopes. Appropriate HIERprocedures proved to be crucial for H2AX foci detection in epidermal skin. Initial attempts to retrieve epitopes by heating buffer-submerged sections by microwaves resulted in partially weak and uneven staining patterns. As described in Paper I, reproducible results could be achieved when the microwave HIER was replaced by 45 minutes of incubation in a 90 C waterbath. Restored epitopes can be detected by IHC using purified antibodies, specific for the particular targets. These primary binding antibodies can in turn be detected by secondary systems, which visualise and amplify the location of the primary signal so it can be observed in a microscope. Secondary detection by IF was utilised for visualisation of DSB-signalling foci and apoptosis in the included papers. These fluorescence techniques use fluorophore conjugated antibodies to allow for detection of several targets simultaneously, in individual cells, by fluorescence microscopy. For the conventional bright field microscopy detections, described in Paper III and IV, the common HRP/DAB enzymatic amplification system was employed. In the HRP/DAB system, secondary detection is enabled by HRPenzyme bound to secondary antibodies. HRP catalyses the reaction where DAB molecules are converted to a brown highly unsolvable dye at the specific binding sites. Detection of a specific marker is often accompanied by a subsequent counter-staining procedure of the cell nuclei. Immunofluorescence stainings in the included papers were accompanied by UV-light excitable DAPI stainings. DAPI molecules bind strongly to DNA and enable visualisation of cell nuclei in the blue spectra. All DAB-stained sections were counterstained by hematoxylin (HTX), which provides a blue nuclear staining suitable for bright field microscopy. Stained tissue sections are prepared for microscopy by application of mounting media and cover glasses. In our experiments, the mounting procedure, as well selection of mounting media and fluorophores, proved crucial for the intensity and stability of the immunofluorescence applications. Allowing the sections to air-dry prior to mounting further decreased the intensity variations. All IF-stainings, intended for apoptosis and foci analysis by H2AX and 53BP1, were conducted manually according to standardised protocols. A faster approach was employed for the DAB/HTX stainings since a semi automated Ventana Benchmark immunostaining system was available for 27

28 this purpose. The Ventana technique was applied to stain sections for p21, the mitosis marker HTA28 and some of the 53BP1 stainings. A third type of staining, Eosin-PAS, was employed in Paper IV to provide histochemical stainings for BKD quantifications in bright field microscopes. This staining visualises the epidermal basal membrane and provides a distinct contrast suitable for manual cell counting by eye. 3.3 Quantification procedures Quantification of histochemical stainings can be as straightforward as manually counting numbers of stained cells in defined areas. This is possible for distinct staining patterns where the main objective is to determine the frequency of stained cells. Ambiguous staining patterns or extended quantification requirements can require additional quantification techniques, such as digital image analysis (DIA). Quantifications by DIA were applied for DSB foci as well as p21 stainings (Figure 5). All digital image analysis procedures described in the included papers, were implemented in Java, as plug-ins to the open source ImageJ software [Abramoff et al. 2004]. Figure 5. Dose response determinations in epidermal skin by digital image analysis. Molecular targets in the cells were targeted and visualised by immunohistochemistry in skin biopsies taken from patients undergoing radiotherapy. Images of the stainings were acquired using a microscope equipped with a CCD-camera. Further processing by digital image analysis provided a dose response of the molecular target. 28

29 3.3.1 Microscopy procedures The choice of microscopy techniques and histochemical detection system will in several ways determine the quantification possibilities. Conventional bright field microscopy can often be advantageous for straight-forward quantifications of distinct staining patterns. The simplicity of the technique and the opportunity of preserving stained sections are significant advantages of bright field microscopy. Additional quantification possibilities are provided by fluorescence microscopy. The use of fluorophores with specific excitation and emission spectra provide a sensitive detection system where several targets in a tissue can be detected simultaneously. This allows for co-localisation analysis or other staining pattern comparisons of multiple targets in individual cells. Quantification using fluorescence microscopy can however suffer from intensity variations due to fluorophore fading or light source inaccuracies. Staining protocol optimisation and microscope calibration procedures can help to reduce these inter- and intra-staining variations. For this reason, a calibration procedure was set up based on fluorescent microspheres as intensity standards. This provided information about fluctuations in the light field intensity and evenness. Exposure times for all individual experiments were calculated based on the microsphere intensities. Variations larger than a preset cut-off level could occasionally require adjustments of the condenser lens and the light source position prior to the exposure time calculation. Reproducible and consistent stainings are essential for accurate quantifications. This is often emphasised when DIA is employed for quantification of intensity-dependent factors. The use of DIA enables measurement of a range of factors that are difficult to interpret by eye. Area, intensity and shapes are quantifiable parameters that can provide extra information about the stained target. These techniques are available for both bright field and fluorescence microscopy. Fluorescence offers a general advantage in providing easy separation of the stained targets, and superior contrast. A potential drawback can however be the previously mentioned need for calibration procedures. The papers of this thesis provide examples of how fluorescence microscopy and bright field microscopy can be employed in different DIA-based quantification procedures. Quantification of DSB-signalling foci using immunofluorescence are presented in Papers I, II and III. In these experiments, images of foci and DAPI nuclear stainings were acquired from epidermal regions of the stained sections at 40x magnification, resulting in image sets where number of foci per nuclei area was determined by DIA. Paper IV exemplifies how DIA can be used in combination with bright field microscopy. In this case, images of epidermal p21 stainings were collected by bright field microscopy at 40x magnification. Reproducible light 29

30 conditions were achieved simply by keeping all microscopy illumination settings fixed for all image acquisitions. All images were recorded using basic laboratory equipment; a bright field microscope equipped for fluorescence, connected to a CCD-camera and a computer Quantification of DSB foci by digital image analysis A semi-automatic quantification procedure, based on DIA, was employed to count number of foci in the tissue (Figure 6). Several later studies have described similar foci quantification procedures [Böcker et al. 2006, Costes et al. 2006, Wykes et al. 2006]. As described in Paper I, our quantification approach was based on a three-step procedure: feature extraction, segmentation and labelling. The fist step is the feature extraction. The purpose of this step was to enhance the appearance of targeted structures in the image; in this case the foci. In Paper I, a Top-hat filter was used for feature extraction. This filter creates images of the backgrounds that are subsequently subtracted from the corresponding original images, creating a more homogenous image with less background variations and equalised foci intensities. For Paper II and III, the Top-hat filter was replaced by a Laplace edge detecting filter. The Laplace filter was more sensitive to staining-correlated intensity variations but proved to be slightly advantageous in resolution. This replacement was possible due to improvements in the staining methods, mainly air-drying of sections prior to mounting, which reduced the overall intra-staining intensity variations. The feature extraction step simplifies the subsequent segmentation, making it possible to find a fixed threshold level that worked satisfactorily throughout the entire material. The segmentation results in a binary image, where pixels in the images are divided in two groups according to a selected threshold level. The threshold levels selecting the foci were kept constant throughout the experiments and were initially selected manually by eye. The number of foci within each image was determined by a labelling algorithm, where pixels are grouped according to their connection to other pixels. In this way it is possible to determine which pixels belong to the same foci and also determine the number of foci in each image. An additional analysis step was added to enable co-localisation studies, where foci locations of the DSB markers H2AX and 53BP1 could be compared in the same cells. This was achieved by a procedure where images of the two separate foci extractions were merged and re-analysed to count numbers of overlapping foci by an additional labelling step. All foci images were accompanied by an image of the corresponding DAPI nuclear staining. These DAPI images provided a reflection of the DNA content in the cell sections in which the foci were quantified. In combination with the foci numbers, the thresholded area of the cell nuclei pro- 30

31 vided a measure which correlates to the density of the DSB damage in the cells. The choice of this measure was based on three main considerations. Firstly, measuring number of foci, as opposed to measuring foci area or foci intensity directly, provided a measure that is less dependent on variations in staining intensities and optical focus. Secondly, measuring DNA area, as opposed to counting actual numbers of cells, was advantageous since only fractions of cells are present in the sections, containing varying amounts of DNA. Finally, the selected measure would also be independent of possible variations in DNA content due to variations in cell cycle phase distributions. Figure 6. Foci detection and DIA-quantification in epidermal cells. Immunofluorescence stainings allow for simultaneous detection of multiple targets in the same cells. a) H2AX b) 53BP1 c) DAPI nuclear staining d) Result of DIA. Red and green corresponds to detected H2AX and 53BP1 foci respectively. Areas of foci colocalisation are marked by yellow Quantification of p21 by digital image analysis The DIA-approach for p21 analysis described in Paper IV exemplifies how DIA can be employed to quantify DAB/HTX stainings and images acquired with bright field microscopy (Figure 7). The procedure was based on separate thresholdings of the red, green and blue (RGB) colour channels. The brown DAB staining, corresponding to the p21 expression was selected and measured in the blue channel of the image, using a fixed threshold level. The areas of the cell nuclei were determined in the red channel by an autothreshold algorithm. Areas not selected by these two thresholding procedures were considered as background. Brown DAB-area per total cell nuclei area was calculated for each stained section. 31

32 Figure 7. Quantification of p21 in epidermal basal keratinocytes by DIA. a) The dashed line outlines the manual selection of basal and suprabasal keratinocytes in the p21 stainings. b) Thresholding procedures provide a measure of p21 stained nuclear area and staining intensity within the selected region Manual quantifications of immunohistochemical stainings Distinct staining patterns were possible to quantify manually by eye (Figure 8). For comparative purposes, p21 was also evaluated by manual counting of the DAB/HTX stained sections in a bright field microscope. p21 positive cells per mm of the epidermal basal layer was determined by counting positive cells by eye, with a 100x objective, and measuring the length of the basal layer using an ocular scale (Figure 8c). Mitosis and BKD (Figure 8a, Figure 9) were evaluated using the same manual counting procedure as for p21. Apoptosis quantifications were performed in a similar way but in fluorescence, using the same stainings as for the H2AX foci quantifications, counting only intensely stained cells signalling apoptotic DNA fragmentation (Figure 8b). 53BP1 foci were also quantified manually in a bright field microscope for comparison with corresponding DIA-results (Figure 8d). This manual quantification by eye was possible due to the low background levels of the 53BP1 staining. The number of foci per cell was counted in 100 consecutive basal keratinocytes. The intensity and signal-to-noise level of this staining was superior compared to the other investigated DSB foci marker, H2AX Statistics Direct inspection of the plotted dose-response relationships and standard errors could in all cases be considered as sufficient evidence for low-dose hypersensitivity. A statistical approach was however set-up to further confirm the observed response. The approach was first described in Paper II where it was used to assess DSB foci dose response. In brief, the procedure is essentially a t-test for evaluation of dose-normalised dose-response relationships from individual patients. A linear regression is applied to each normalised and logarithmic transformed dose response providing a slope that 32

33 approximates the low-dose hypersensitivity for each patient. This test is therefore best suited for parameters where good linear fits are achieved. In our case, foci calculations providing large amounts of data provided better curve fits than other more infrequent and fluctuating parameters, such as mitosis and apoptosis. It should therefore be noted that the accuracy of the curve fits will not be directly reflected in the p-value. As well as reflecting low-dose hypersensitivity, the magnitude of the p-value will also be affected by the similarity in dose response between patients. Figure 8. Manually quantified immunostainings. a) Mitosis, detected by the HTA28 antibody, specific for a histone H3 phosphorylation. b) Apoptosis detected by H2AX. The intensely stained cell in the centre of the image displays apoptosis related DNA fragmentation. c) Intermediate, as well as intensely stained p21 nuclei can be observed in the basal layer of the epidermis. d) The distinct staining of DSBsignalling 53BP1 foci enables quantification by eye. 33

34 Figure 9. Reduction in basal keratinocyte density throughout the radiotherapy treatment course. Images illustrate how radiation exposed epidermis thins out during the treatment course. a) Epidermis during first treatment week. b) Epidermis during last treatment week. 34

35 4. Results and Discussion 4.1 Paper I: DNA double strand break quantification in skin biopsies A method for quantification of DSB-signalling H2AX foci in epidermal regions of paraffin embedded skin biopsies was set up and evaluated using skin biopsy material from patients undergoing radiotherapy. The quantification procedure was constructed with an effort to create a straightforward and robust method based on standard laboratory equipment and basic image analysis algorithms. This approach makes the method accessible but also introduces potential pitfalls as a result of inaccuracies in the equipment and irregularities in the material. An initial investigation revealed that HIER methods and mercury lamp variations were key factors that had to be monitored. A standardised immunohistochemistry protocol was set up with an aim to reduce both intra- and inter-staining variations. The observed intensity variations in the mercury lamp light source was monitored and compensated for by recurrent calibrations of the microscope. The design of the digital image analysis procedure could further compensate for remaining intensity variations. An initiating Top-hat background subtraction algorithm reduced remaining intensity variations. Segmentation and labelling steps were further included, where actual foci numbers are counted, rather than measuring foci intensity or area. This contributed to making the measure less influenced by intensity variations. The method was evaluated by foci quantification in skin punch biopsies taken from two prostate cancer patients undergoing radiotherapy. Biopsies taken from regions of the skin corresponding to a range of low doses ( Gy) provide a material where the relation between fraction dose size and H2AX foci density can be determined. The results revealed an overall linear increase in H2AX foci with dose, at 30 minutes following the delivery of the first fraction. Similar dose-response relationships were observed for additional biopsy sets, taken a week into treatment: 30 minutes after the 4 th or 5 th fractions. Furthermore, the two investigated patients provided very similar responses. Repeated staining and quantification procedures revealed convincing reproducibility. 35

36 4.2 Paper II: Low-dose hypersensitive H2AX response and infrequent apoptosis in epidermis from radiotherapy patients. Paper II expands the results from Paper I, employing the H2AX foci quantification procedure to investigate a more extensive skin biopsy material acquired from 5 prostate cancer patients. This allowed for a more detailed examination of the H2AX dose response during the first week of fractionated radiotherapy. All patients displayed a consistent shape of the H2AX dose response curves; an overall linear shape combined with a pronounced non-linearity for the low-dose region, suggesting a hypersensitive response to doses below ~0.3Gy. A re-examination of the dose-response relationships, observed for the two patients in Paper I, reveals that these patients also display low-dose hypersensitivity. The low-dose hypersensitive response was observed just 30 minutes following the first fraction of the treatment as well as one week into the treatment, 30 minutes after the 4 th or 5 th fractions. Assuming a linear induction of DSBs, this suggests that the non-linear response is a result of dose-dependent differences in DSB repair efficiency where the lower dose levels fail to trigger repair responses to the same extent as the higher doses (Figure 10). This would be in agreement with the observed threshold activation of ATM in vitro [Bakkenist et al. 2003]. The hypersensitive response was further illustrated by calculating the amounts of H2AX foci per dose unit. This provides a dose normalised measure which enables direct comparison of cellular response to the different doses. Statistical calculations on individually normalised dose-response relationships confirmed a low-dose hypersensitive response for all investigated time points (p<0.01). In an attempt to detect additional DSB repair at later time points, additional biopsies were collected 2 hours following delivered fractions. Unexpectedly, no significant decrease in foci levels could be detected from 30 minutes to 2 hours after damage induction. Considering the observed lowdose hypersensitive response, a possible explanation for the observed H2AX kinetics could be a DSB repair machinery with fast and slow repair components. Previous studies on radiotherapy related skin erythema have shown that short intra-fraction breaks of 15 minutes are sufficient to reduce these treatment side effects [Turesson et al. 1984, Turesson et al. 1989]. This supports the suggestion that a considerable amount of repair can occur in skin tissues within the first 30 minutes following DSB induction. As well as detecting DSB-signalling foci, the H2AX stainings also enabled quantification of apoptotic cells in the epidermis. A subset of intensely H2AX stained cells, possibly signalling DNA fragmentation, were present throughout the material and could easily be distinguished from the regular 36

37 foci patterns. These cells displayed several hallmarks of apoptosis, such as cellular rounding, shrinkage, apoptotic bodies and co-staining with cparp-1 [Steel (rev) 2001, Holubec et al. 2005]. The apoptotic frequencies in the epidermis were very low throughout the material and only 43 apoptotic cells in total could be detected among a total number of cells. The data was however sufficient to detect a dose dependency in the apoptotic frequency (p<0.01). Compared to the previous paper, the understanding and execution of the foci quantification methods were improved for this publication. For instance, air drying all sections prior to mounting reduced variations in staining intensities considerably. The reduced intra-staining variations allowed for improvements of the digital image analysis process. The feature extracting Top-hat filter was replaced with an edge detecting Laplace filter, providing an improvement in foci resolution. Figure 10. An interpretation of the low-dose hypersensitive dose response observed for DSB-signalling foci in epidermis. Assuming a linear DSB induction (solid line), the observed foci dose response suggests that a considerable amount of the initial damage is repaired as early as 30 minutes after damage induction (dashed lined). The shape of the dose response curve further suggests that doses below ~0.3Gy fail to activate repair mechanisms to the same extent as the higher doses. 37

38 4.3 Paper III: Low-dose hypersensitivity of initial and persistent DSB foci in epidermis throughout fractionated radiotherapy treatment Paper III is an extension of the study presented in Paper II. DSB foci dose response throughout the treatment course was evaluated in the epidermis using H2AX as well as 53BP1 as DSB-signalling markers. Digital image analysis was employed to quantify DSB foci dose response during the last week of radiotherapy. Low-dose hypersensitivity was observed 30 minutes following delivered fractions in skin biopsies collected from five patients. The dose response was further recorded 24 hours following delivered fractions, in biopsies from five additional patients. Since the patients receive daily fractions, this time point also corresponds to the epidermal state just prior to delivery of the subsequent fraction. The results reveal that foci levels are elevated prior to fraction delivery and that this persistent subset of foci also displays low-dose hypersensitivity. The DSB-signalling markers H2AX and 53BP1 was quantitatively and qualitatively compared in the described biopsy-sets taken during last treatment week. The quantitative comparison displayed overlapping foci doseresponse relationships for both 30 minute and 24 hour time points. A quantitative evaluation of co-localisation did however reveal that a considerable portion of the foci were unique to each individual marker. Further qualitative evaluations pointed out differences in background levels and signal-to-noise ratio. 53BP1 foci were generally more intense and distinct compared to H2AX foci. 53BP1 immunostainings also displayed lower background levels and an absence of apoptotic detection. The crisp and distinct staining pattern of 53BP1 enabled detection of DSB-signalling foci by conventional bright field microscopy using DAB/HTX stained biopsy sections. One advantage with this technique, compared to immunofluorescence, is that it is possible to preserve the slides post-analysis. Furthermore, problems with fading and light intensity variations are reduced. Biopsies taken throughout the treatment course were immunostained by 53BP1 and quantified by eye in a bright field microscope to determine foci dose response. The shape of the dose response curves agreed with previous observations where foci were quantified by immunofluorescence and digital image analysis confirming preserved low-dose hypersensitivity throughout the treatment course and persisting foci in between fractions. The low-dose hypersensitive shape of the DSB foci dose response, in combination with the observation of residual persistent foci in between fractions, suggests a possible link with the observations of low-dose hypersensitivity in basal keratinocyte reductions [Joiner et al. 2001, Harney et al. 2004#1, Paper IV]. More specifically, persistent foci could be involved in 38

39 mechanisms triggering a sustained growth arrest, which in turn contributes to the keratinocyte depression observed during the treatment course of fractionated radiotherapy. 4.4 Paper IV: A low-dose hypersensitive keratinocyte response to fractionated radiotherapy is mediated by growth arrest and apoptosis Paper II and III described a low-dose hypersensitive response in epidermal DSB foci during radiotherapy. The results of Paper IV reveal that the hypersensitive response is reflected in several other endpoints further downstream in the epidermal DDR pathways. The p21 kinase inhibitor is involved in several checkpoints that regulate cell cycle progression following DNA damage. Of particular importance is its involvement in the sensitive G 1 checkpoint. In this study, frequencies of p21 nuclear stained keratinocytes were determined in basal epidermal keratinocytes in skin biopsies taken during the first and last week of a 7 week treatment course. An evident increase in p21 expression was detected during treatment. Furthermore, a low-dose hypersensitive response in p21, to doses below ~0.3Gy, could be observed during the first treatment week, and was still evident at the end of the treatment. Manual counting of p21-positive cell nuclei was compared to quantifications by digital image analysis. The low-dose hypersensitive response could be observed independent of quantification method. The results suggest that p21-controlled checkpoints can be activated in epidermal basal keratinocytes after very small amounts of damage and that this response is amplified for low doses. The epidermal dose response was further evaluated by quantification of mitotic frequencies, using antibodies against phosphorylated histone H3, a distinct marker specific for mitosis. Decreased levels of mitosis could be observed throughout the treatment course for all dose levels. However, the reduction in mitotic frequency per dose unit was more pronounced for doses below ~0.3Gy, revealing a low-dose hypersensitive response for the mitosis endpoint as well. Apoptotic frequencies in the basal and suprabasal layer of epidermis were quantified in biopsies collected during the last treatment week in order to determine the dose response. Apoptotic events proved to be infrequent throughout the examined material. A general increase in apoptosis during treatment is however evident when comparing with apoptosis levels recorded after the first treatment week (see Paper II). Similar to previously described endpoints, apoptotic dose response during the last week of treatment display a low-dose hypersensitivity for doses below ~0.3Gy. 39

40 A decrease in basal cell density has been observed in normal skin from patients undergoing radiotherapy [Joiner et al. 2001, Harney et al. 2004#1]. In this paper, the dose response of keratinocytes, the major cell type in the basal layer, was monitored by quantifications of BKD levels during the treatment course. A low-dose hypersensitivity response in BKD reduction could be observed for doses below ~0.3Gy in samples collected throughout the treatment course. A general increase in the thickness of the cornified layer could be observed towards the end of the treatment, suggesting that the amount of cells entering differentiation have increased. This further suggests that a pathway where damaged basal cells are eliminated by differentiation could be an epithelial damage response of importance [von Wagenheim et al. (rev) 1995]. In conclusion, the results from Paper IV indicate that permanent growth arrest as well as apoptosis could explain the observed reductions in BKD throughout a radiotherapy treatment course. The observed low-dose hypersensitivity responses in p21, mitosis, apoptosis, and BKD possibly reflect a chain of events where the low-dose effects are preserved throughout the DDR pathways in agreement with our previous observations of low-dose hypersensitivity in DSB-signalling foci. 40

41 5. Conclusions Examination of the dose response in epithelial skin from prostate cancer patients undergoing fractionated radiotherapy treatments provides the following conclusions: DSB-signalling H2AX and 53BP1 foci were detectable and quantifiable in formalin fixed tissue material. A detailed dose response could be determined for clinically relevant dose levels ranging from 0.05 to 1.1Gy. Low-dose hypersensitivity in DSB-signalling foci was observed throughout the treatment course. DSB-signalling foci did not return to background levels in between fractions. The persistent foci expressed low-dose hypersensitivity. Assessment of checkpoint activation (p21), mitosis reduction (HTA28) and apoptosis induction ( H2AX) revealed low-dose hypersensitivity. The reduction of basal keratinocytes expressed low-dose hypersensitivity throughout the treatment course. Growth arrest will influence the dose-dependent reduction of keratinocytes observed during fractionated radiotherapy. Cell death mechanisms such as apoptosis could also contribute. 41

42 6. Acknowledgements The acknowledgements is probably the most read section in any thesis. I ve therefore tried to make the most of it by taking the opportunity to express my honest opinions about everyone in my surroundings. It is not often that you get the chance to put your thoughts into a printed book that will be stored in the Uppsala University archives for ever and ever and ever But don t worry; there are good things to say about all of you In fact some of the stuff is soppy, sentimental and pretentious, so if you re allergic to any of those things, stop reading now! My first thanks goes to my intelligent companion Martin. We have a collaboration completely free from competition and prestige where anything can be discussed or made fun of. I think that s why two stubborn minded, critical thinkers like us can work together as equals. I m proud of our collaboration and the fact that we have helped each other to cope with the difficulties. Thanks also for hosting smelly surströmming dinners for home sick exile northerners and for being a very good friend. Your lovely family, Anne and Fanny, are of course included in all my thanks. I d like to thank my highly motivated supervisor Ingela for her enthusiasm and for keeping me in touch with the clinical perspective of things. Your genuine interest in science is an inspiration. I also thank you and all the collaborators in Gothenburg (Jan, Ingegerd, Karl-Axel and Ragnhild) for the excellent skin biopsy material you have collected throughout the years. Special thanks to all prostate cancer patients that bravely donated significant amounts of themselves! Additional thanks to Ingegerd for all your meticulous cell countings and to Ragnhild for keeping track of endless amounts of data Our small group in Uppsala includes a few more people. Special thanks to our room mate Majlis for talks about important and unimportant stuff as well as all the tissue-slicing. Thanks to Ulf for supporting the fair and honest side of things when needed, and for your unique combination of sanity and insanity. Thanks also to Sunna for counting endless numbers of tissue sections with infinite patience. Special thanks to Didde for surfing the waves and keeping us afloat in the rough seas of paper work. On behalf of the oncology department; thanks to Gunilla Enblad and Jan Nyman for signing up as co-supervisors. Many thanks to Hans Garmo at ROC for very helpful discussions about statistics. 42

43 The best coffee in the building is served in our office. So, my sincere thanks goes to our coffee and coffee in general! Thanks also to two seasons of Flight of the Conchords for entertaining me when the thesis-writing got too tedious. Thanks to everyone in the oncology corridor who have made an effort to make fika -discussions interesting throughout the years. Thanks to Anders, Anita, Daniel, Mattias, Marie, Gustaf, Ingrid, Linda, Lena, Marie-Louise, Maryam, Nongnit, Norafiza, Per, Rose-Marie, Xuping and of course the genpat-gang during the early years. Special thanks to Ingrid for taking the lead and knowing when to do what in the thesis-writing-procedure. I would also like to thank the nice people in our neighbouring BMScorridor, in particular Jörgen, Bosse and Irina, that I ve had the pleasure of collaborating with in small projects separate from my main PhD project. Thanks to my mom Margareta for always being there. Also, for raising me without any pressure or demands. Thanks also for all discussions about small and big stuff. They ve helped prepare me for the upcoming challenges as well as for life in general. Most of all, thanks for all your concern and very warm heart. Thanks to moms husband Hans-Erik for teaching me essential skills from the old days and physics from a practical approach. How to cut down trees, move rocks and smoke perch. More importantly, thanks for being a cool and unique role model. Thanks to my dad Thomas for all your genuine concern and support. We re much more alike than you think. I ve made a list of all the good qualities that we share. Maybe we can discuss it during a relaxed road trip in Italy. I think we should give up our careers as para-gliding mounting climbers though, I m convinced that a more relaxed style suits us better. Thanks to mormor Irma for teaching me everything about fishing. Also for all the good food, fika and discussions. I think you would have been a brilliant scientist! Thanks to morfar Folke for having been the foundation of the family. Hope I inherited a bit of your strength and stamina. I know that you would have been proud of me and the fact that this acknowledgement section is very family-focused. Thanks to my aunt Monica and her husband Tommy for all the good times and for being so nice. Thanks especially for letting me come along on all those fishing adventures even though I was just a kid. Thanks to my cousin Sebastian for confirming that following-the-crowd just doesn t suit our genes. Thanks to my aunt Margot and her husband Calle for all work opportunities, mechanic advice etc, but more importantly for being very fun and authentic people that I definitely identify with. Thanks to my oldest sister Sara for taking care of me during my early days. You taught me how to play with dolls, style hair with bobbins, make mixed tapes and many more essential skills. Feel a bit guilty that I don t visit 43

44 you and your handsome boys (Mattias, Albin and Emil) so often, but I m gonna make up for it soon by cooking you all a nice dinner at your place! Thanks to my younger sister Hanna for thinking that I know stuff even when I don t. And for taking so good care of one of my favourite places, our summer house on Laxön. And also for always being such a great company. I include your very well-selected and matching husband David in all of those thanks. And the newest family-addition, Selma, who gets thanks just for being cute. Thanks to my youngest sister Linnea for your laid back and original style. I like to think that I m partly responsible for influencing you to be so cool and talented At the same time, you ll always be the baby sister that jumps around singing to herself with a svamphatt on her head Because of that, and many other things, I m also gonna put into writing that I think you are a true role model in many ways. Hope you and your equally cool boyfriend David, have time to hang out with your old brother this summer. Many thanks to Fionas family for taking very good care of weird Swedes when they re away from home. Just to mention a few things: Thanks to Patricia, for all the lovely food. Joe, for trying to teach me the essentials of rugby (I m still clueless). Eoin and Niall for philosophical birthday cards and advice on how to train rats. You re all brill. Thanks to Fiona for making me happy. A lot of things changed when I met you. I suddenly liked tea, long flights and the smell of Irish washing powder. Irish breakfasts as well, but I guess I would have fancied them anyway Silly love songs started to make complete sense and writing cheesy acknowledgements is not a problem anymore. Obvious and true things are very easy. 44

45 7. References Abramoff M, Magelhaes P, Ram S. Image Processing with ImageJ. Biophoton Int ;11(7): Alberts B, Johnson A, Lewis J, Raff M, Roberts K, Walter P. Chromosomal DNA and Its Packaging in the Chromatin Fiber. Ch 4 in: Molecular Biology of the Cell, 4th ed. Garland Science, New York, USA Andegeko Y, Moyal L, Mittelman L, Tsarfaty I, Shiloh Y, Rotman G. Nuclear retention of ATM at sites of DNA double strand breaks. J Biol Chem Oct 12;276(41): Anderson L, Henderson C, Adachi Y. Phosphorylation and rapid relocalization of 53BP1 to nuclear foci upon DNA damage. Mol Cell Biol Mar ;21(5): Bakkenist CJ, Kastan MB. DNA damage activates ATM through intermolecular autophosphorylation and dimer dissociation. Nature Jan 30;421(6922): Banin S, Moyal L, Shieh S, Taya Y, Anderson C, Chessa L, et al. Enhanced phosphorylation of p53 by ATM in response to DNA damage. Science ;281(5383): Bartkova J, Horejsi Z, Koed K, Kramer A, Tort F, Zieger K, et al. DNA damage response as a candidate anti-cancer barrier in early human tumorigenesis. Nature Apr 14;434(7035): Bassing CH, Chua KF, Sekiguchi J, Suh H, Whitlow SR, Fleming JC, et al. Increased ionizing radiation sensitivity and genomic instability in the absence of histone H2AX. Proc Natl Acad Sci U S A Jun 11;99(12): Bassing CH, Suh H, Ferguson DO, Chua KF, Manis J, Eckersdorff M, et al. Histone H2AX: a dosage-dependent suppressor of oncogenic translocations and tumors. Cell ;114(3): Blanpain C, Fuchs E. Epidermal homeostasis: a balancing act of stem cells in the skin. Nat Rev Mol Cell Biol Mar ;10(3): Bonner WM. Phenomena leading to cell survival values which deviate from linearquadratic models. Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis Dec 2;568(1): Bonner WM, Redon CE, Dickey JS, Nakamura AJ, Sedelnikova OA, Solier S, Pommier Y. GammaH2AX and cancer. Nat Rev Cancer Dec ;8(12): Botuyan MV, Lee J, Ward IM, Kim J, Thompson JR, Chen J, Mer G. Structural basis for the methylation state-specific recognition of histone H4-K20 by 53BP1 and Crb2 in DNA repair. Cell Dec 29;127(7): Bouquet F, Muller C, Salles B. The loss of gammah2ax signal is a marker of DNA double strand breaks repair only at low levels of DNA damage. Cell Cycle May ;5(10): Brenner DJ, Doll R, Goodhead DT, Hall EJ, Land CE, Little JB, Lubin JH, Preston DL, Preston R, Puskin JS, Ron E, Sachs RK, Samet JM, Setlow RB, Zaider M. 45

46 Cancer risks attributable to low doses of ionizing radiation: assessing what we really know. Proc Natl Acad Sci U S A Nov 25;100(24): Brenner DJ, Hall EJ. Computed tomography--an increasing source of radiation exposure. N Engl J Med Nov 29;357(22): Burma S, Chen B, Murphy M, Kurimasa A, Chen D. ATM phosphorylates histone H2AX in response to DNA double-strand breaks. J Biol Chem Nov 9;276(45): Buscemi G, Perego P, Carenini N, Nakanishi M, Chessa L, Chen J, et al. Activation of ATM and Chk2 kinases in relation to the amount of DNA strand breaks. Oncogene ;23(46): Böcker W, Iliakis G. Computational Methods for analysis of foci: validation for radiation-induced gamma-h2ax foci in human cells. Radiat Res Jan ;165(1): Canman C, Lim D, Cimprich K, Taya Y, Tamai K, Sakaguchi K, Appella E, Kastan M, Siliciano J. Activation of the ATM kinase by ionizing radiation and phosphorylation of p53. Science ;281(5383): Cardis E, Vrijheid M, Blettner M, Gilbert E, Hakama M, Hill C, 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 Apr ;167(4): Celeste A, Petersen S, Romanienko PJ, Fernandez-Capetillo O, Chen HT, Sedelnikova OA, et al. Genomic instability in mice lacking histone H2AX. Science May 3;296(5569): Celeste A, Difilippantonio S, Difilippantonio MJ, Fernandez-Capetillo O, Pilch DR, Sedelnikova OA, et al. H2AX haploinsufficiency modifies genomic stability and tumor susceptibility. Cell ;114(3): Celeste A, Fernandez-Capetillo O, Kruhlak MJ, Pilch DR, Staudt DW, Lee A, Bonner RF, Bonner WM, Nussenzweig A. Histone H2AX phosphorylation is dispensable for the initial recognition of DNA breaks. Nat Cell Biol Jul ;5(7): Chen H, Bhandoola A, Difilippantonio M, Zhu J, Brown M, Tai X, et al. Response to RAG-mediated VDJ cleavage by NBS1 and gamma-h2ax. Science Dec 8;290(5498): Chowdhury D, Keogh M, Ishii H, Peterson CL, Buratowski S, Lieberman J. gamma- H2AX dephosphorylation by protein phosphatase 2A facilitates DNA doublestrand break repair. Mol Cell Dec 9;20(5): Chronis F, Rogakou EP. Interplay Between H2AX and 53BP1 Pathways in DNA Double-Strand Break Repair Response. Ch 13 in: Apoptosis, Senescence, and Cancer, 2nd ed, Edited by Gewirtz DA, Holt SE and Grant S. Humana Press Inc, Totowa, NJ, USA p Costes SV, Boissiere A, Ravani S, Romano R, Parvin B, Barcellos-Hoff MH. Imaging features that discriminate between foci induced by high- and low-let radiation in human fibroblasts. Radiat Res May ;165(5): el-deiry W, Tokino T, Velculescu V, Levy D, Parsons R, Trent J, et al. WAF1, a potential mediator of p53 tumor suppression. Cell Nov 19;75(4): Enns L, Bogen KT, Wizniak J, Murtha AD, Weinfeld M. Low-dose radiation hypersensitivity is associated with p53-dependent apoptosis. Mol Cancer Res Oct ;2(10): Falck J, Coates J, Jackson SP. Conserved modes of recruitment of ATM, ATR and DNA-PKcs to sites of DNA damage. Nature Mar 31;434(7033): Fernandez-Capetillo O, Lee A, Nussenzweig M, Nussenzweig A. H2AX: the histone guardian of the genome. DNA Repair (Amst) ;3(8-9):

47 Gorgoulis VG, Vassiliou LF, Karakaidos P, Zacharatos P, Kotsinas A, Liloglou T, et al. Activation of the DNA damage checkpoint and genomic instability in human precancerous lesions. Nature Apr 14;434(7035): Gudkov AV, Komarova EA. The role of p53 in determining sensitivity to radiotherapy. Nat Rev Cancer Feb ;3(2): Halazonetis TD, Gorgoulis VG, Bartek J. An Oncogene-Induced DNA Damage Model for Cancer Development. Science Mar 7;319(5868): Harney J, Shah N, Short S, Daley F, Groom N, Wilson GD, et al. The evaluation of low dose hyper-radiosensitivity in normal human skin. Radiotherapy and Oncology Mar ;70(3): Harney J, Short SC, Shah N, Joiner M, Saunders MI. Low dose hyperradiosensitivity in metastatic tumors. International Journal of Radiation Oncology*Biology*Physics Jul 15;59(4): Harper J, Adami G, Wei N, Keyomarsi K, Elledge S. The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinases. Cell Nov 19;75(4): Helleday T, Petermann E, Lundin C, Hodgson B, Sharma RA. DNA repair pathways as targets for cancer therapy. Nat Rev Cancer Mar ;8(3): Hirata A, Inada K, Tsukamoto T, Sakai H, Mizoshita T, Yanai T, et al. Characterization of a Monoclonal Antibody, HTA28, Recognizing a Histone H3 Phosphorylation Site as a Useful Marker of M-phase Cells. J. Histochem. Cytochem ;52(11): Holubec H, Payne CM, Bernstein H, Dvorakova K, Bernstein C, Waltmire CN, et al. Assessment of apoptosis by immunohistochemical markers compared to cellular morphology in ex vivo-stressed colonic mucosa. J Histochem Cytochem Feb ;53(2): Honoré HB, Bentzen SM. A modelling study of the potential influence of low dose hypersensitivity on radiation treatment planning. Radiotherapy and Oncology Apr ;79(1): Huang L, Clarkin K, Wahl G. Sensitivity and selectivity of the DNA damage sensor responsible for activating p53-dependent G1 arrest. Proc Natl Acad Sci U S A May 14;93(10): Huen MS, Grant R, Manke I, Minn K, Yu X, Yaffe MB, Chen J. RNF8 transduces the DNA-damage signal via histone ubiquitylation and checkpoint protein assembly. Cell Nov 30;131(5): Huyen Y, Zgheib O, Ditullio RA, Gorgoulis VG, Zacharatos P, Petty TJ, Sheston EA, Mellert HS, Stavridi ES, Halazonetis TD. Methylated lysine 79 of histone H3 targets 53BP1 to DNA double-strand breaks. Nature Nov 18;432(7015): Joiner M, Denekamp J, Maughan R. The use of 'top-up' experiments to investigate the effect of very small doses per fraction in mouse skin. Int J Radiat Biol Relat Stud Phys Chem Med Apr ;49(4): Joiner M, Johns H. Renal damage in the mouse: the response to very small doses per fraction. Radiat Res May ;114(2): Joiner MC, Lambin P, Malaise EP, Robson T, Arrand JE, Skov KA, et al. Hypersensitivity to very-low single radiation doses: Its relationship to the adaptive response and induced radioresistance. Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis Nov 4;358(2): Joiner M, Marples B, Lambin P, Short S, Turesson I. Low-dose hypersensitivity: current status and possible mechanisms. Int J Radiat Oncol Biol Phys Feb 1;49(2):

48 Jones PH, Simons BD, Watt FM. Sic transit gloria: farewell to the epidermal transit amplifying cell? Cell Stem Cell Oct 11;1(4): Kao GD, McKenna W, Guenther MG, Muschel RJ, Lazar MA, Yen TJ. Histone deacetylase 4 interacts with 53BP1 to mediate the DNA damage response. J Cell Biol Mar 31;160(7): Kastan MB. DNA damage responses: mechanisms and roles in human disease: 2007 g.h.a. Clowes memorial award lecture. Mol Cancer Res Apr ;6(4): Kinner A, Wu W, Staudt C, Iliakis G. {gamma}-h2ax in recognition and signaling of DNA double-strand breaks in the context of chromatin. Nucl. Acids Res Oct 1;36(17): Kolas NK, Chapman J, Nakada S, Ylanko J, Chahwan R, Sweeney FD, et al. Orchestration of the DNA-damage response by the RNF8 ubiquitin ligase. Science Dec 7;318(5856): Krause M, Hessel F, Wohlfarth J, Zips D, Hoinkis C, Foest H, et al. Ultrafractionation in A7 human malignant glioma in nude mice. Int J Radiat Biol Jun ;79(6): Krause M, Prager J, Wohlfarth J, Hessel F, Dorner D, Haase M, et al. Ultrafractionation does not improve the results of radiotherapy in radioresistant murine DDL1 lymphoma. Strahlenther Onkol ;181(8): Krause M, Wohlfarth J, Georgi B, Pimentel N, Dorner D, Zips D, et al. Low-dose hyperradiosensitivity of human glioblastoma cell lines in vitro does not translate into improved outcome of ultrafractionated radiotherapy in vivo. Int J Radiat Biol Oct ;81(10): Krueger SA, Joiner MC, Weinfeld M, Piasentin E, Marples B. Role of apoptosis in low-dose hyper-radiosensitivity. Radiat Res Mar ;167(3): Krueger SA, Collis SJ, Joiner MC, Wilson GD, Marples B. Transition in Survival From Low-Dose Hyper-Radiosensitivity to Increased Radioresistance Is Independent of Activation of ATM SER1981 Activity. International Journal of Radiation Oncology*Biology*Physics Nov 15;69(4): Lee AC, Fernandez-Capetillo O, Pisupati V, Jackson SP, Nussenzweig A. Specific association of mouse MDC1/NFBD1 with NBS1 at sites of DNA-damage. Cell Cycle Jan ;4(1): Lin P, Wu A. Not all 2 gray radiation prescriptions are equivalent: Cytotoxic effect depends on delivery sequences of partial fractionated doses. International Journal of Radiation Oncology*Biology*Physics Oct 1;63(2): Linke S, Clarkin K, Wahl G. p53 mediates permanent arrest over multiple cell cycles in response to gamma-irradiation. Cancer Res Mar 15;57(6): Löbrich M, Jeggo PA. The two edges of the ATM sword: Co-operation between repair and checkpoint functions. Radiotherapy and Oncology Aug ;76(2): Löbrich M, Jeggo PA. The impact of a negligent G2/M checkpoint on genomic instability and cancer induction. Nat Rev Cancer Nov ;7(11): Mailand N, Bekker-Jensen S, Faustrup H, Melander F, Bartek J, Lukas C, Lukas J. RNF8 ubiquitylates histones at DNA double-strand breaks and promotes assembly of repair proteins. Cell Nov 30;131(5): Markova E, Schultz N, Belyaev I. Kinetics and dose-response of residual 53BP1/gamma-H2AX foci: co-localization, relationship with DSB repair and clonogenic survival. Int J Radiat Biol May ;83(5):

49 Marples B, Joiner M. The response of Chinese hamster V79 cells to low radiation doses: evidence of enhanced sensitivity of the whole cell population. Radiat Res Jan ;133(1): Marples B, Wouters BG, Collis SJ, Chalmers AJ, Joiner MC. Low-dose hyperradiosensitivity: a consequence of ineffective cell cycle arrest of radiationdamaged G2-phase cells. Radiat Res Mar ;161(3): Marples B, Krueger SA, Collis SJ, Joiner MC. Low Dose Hyper-Radiosensitivity: A Historical Perspective. Ch 19 in: Targeted Radionuclide Tumor Therapy. Edited by Stigbrand T, Carlsson J and Adams GP. Springer, Netherlands p Marples B, Collis SJ. Low-dose hyper-radiosensitivity: past, present, and future. Int J Radiat Oncol Biol Phys Apr 1;70(5): Mukherjee B, Kessinger C, Kobayashi J, Chen BP, Chen DJ, Chatterjee A, et al. DNA-PK phosphorylates histone H2AX during apoptotic DNA fragmentation in mammalian cells. DNA Repair May 10;5(5): Mundt AJ, Roeske JC, Chung TD, Weichselbaum RR, Priciples of Radiation Oncology, Biologic Basis of Radiation Therapy. Sect 9:39 in: Cancer Medicine 6. Edited by Kufe DW, Pollock RE, Weichselbaum RR et al. BC Decker Inc, London, UK Nakamura AJ, Chiang Y, Hathcock KS, Horikawa I, Sedelnikova OA, Hodes RJ, Bonner WM. Both telomeric and non-telomeric DNA damage are determinants of mammalian cellular senescence. Epigenetics Chromatin ;1(1):6. Nowak E, Etienne O, Millet P, Lages CS, Mathieu C, Mouthon M, et al. Radiationinduced H2AX phosphorylation and neural precursor apoptosis in the developing brain of mice. Radiat Res Feb ;165(2): Nyman J, Turesson I. Basal cell density in human skin for various fractionation schedules in radiotherapy. Radiother Oncol Nov ;33(2): Paull T, Rogakou E, Yamazaki V, Kirchgessner C, Gellert M, Bonner W. A critical role for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage. Curr Biol Jul 27;10(15): Preston D, Ron E, Tokuoka S, Funamoto S, Nishi N, Soda M, Mabuchi K, Kodama K. Solid cancer incidence in atomic bomb survivors: Radiat Res Jul ;168(1):1-64. Rappold I, Iwabuchi K, Date T, Chen J. Tumor suppressor p53 binding protein 1 (53BP1) is involved in DNA damage-signaling pathways. J Cell Biol Apr 30;153(3): Rogakou EP, Pilch DR, Orr AH, Ivanova VS, Bonner WM. DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. J Biol Chem Mar 6;273(10): Rogakou EP, Boon C, Redon C, Bonner WM. Megabase chromatin domains involved in DNA double-strand breaks in vivo. J Cell Biol Sep 6;146(5): Rogakou EP, Nieves-Neira W, Boon C, Pommier Y, Bonner WM. Initiation of DNA Fragmentation during Apoptosis Induces Phosphorylation of H2AX Histone at Serine 139. J. Biol. Chem Mar 24;275(13): Rothkamm K, Löbrich M. Evidence for a lack of DNA double-strand break repair in human cells exposed to very low x-ray doses. Proc Natl Acad Sci U S A Apr 29;100(9): Sanders SL, Portoso M, Mata J, Bahler J, Allshire RC, Kouzarides T. Methylation of histone H4 lysine 20 controls recruitment of Crb2 to sites of DNA damage. Cell Nov 24;119(5):

50 Savitsky K, Bar-Shira A, Gilad S, Rotman G, Ziv Y, Vanagaite L, et al. A single ataxia telangiectasia gene with a product similar to PI-3 kinase. Science Jun 23;268(5218): Schultz LB, Chehab NH, Malikzay A, Halazonetis TD. p53 Binding Protein 1 (53BP1) Is an Early Participant in the Cellular Response to DNA Double-Strand Breaks. J. Cell Biol Dec 18;151(7): Sedelnikova OA, Rogakou EP, Panyutin IG, Bonner WM. Quantitative detection of (125)IdU-induced DNA double-strand breaks with gamma-h2ax antibody. Radiat Res Oct ;158(4): Steel GG. Introduction: The significance of radiobiology for radiotherapy. Ch 1 in: Basic Clinical Radiobiology, 3rd ed. Edited by Steel GG. Arnold, London, UK p1-7. Steel GG. The Case Against Apoptosis. Acta Oncologica ;40(8): Stiff T, O'Driscoll M, Rief N, Iwabuchi K, Lobrich M, Jeggo PA. ATM and DNA- PK function redundantly to phosphorylate H2AX after exposure to ionizing radiation. Cancer Res Apr 1;64(7): Turesson I. The progression rate of late radiation effects in normal tissue and its impact on dose-response relationships. Radiother Oncol Jul ;15(3): Turesson I, Notter G, Wickström I, Johansson KA, Eklund S. The influence of irradiation time per treatment session on acute and late skin reactions: a study on human skin. Radiother Oncol Oct ;2(3): Turesson I, Bernefors R, Book M, Flogegård M, Hermansson I, Johansson KA, et al. Normal tissue response to low doses of radiotherapy assessed by molecular markers--a study of skin in patients treated for prostate cancer. Acta Oncol ;40(8): Turesson I, Thames HD. Repair capacity and kinetics of human skin during fractionated radiotherapy: erythema, desquamation, and telangiectasia after 3 and 5 year's follow-up. Radiother Oncol Jun ;15(2): Van der Kogel AJ. Radiation response and tolerance of normal tissues. Ch 5 in: Basic Clinical Radiobiology, 2nd ed. Edited by Steel GG. Arnold, London, UK p von Wangenheim K, Peterson H, Schwenke K. Review: a major component of radiation action: interference with intracellular control of differentiation. Int J Radiat Biol Okt ;68(4): Wang H, Wang M, Wang H, Bocker W, Iliakis G. Complex H2AX phosphorylation patterns by multiple kinases including ATM and DNA-PK in human cells exposed to ionizing radiation and treated with kinase inhibitors. J Cell Physiol Feb ;202(2): Ward I, Chen J. Histone H2AX is phosphorylated in an ATR-dependent manner in response to replicational stress. J Biol Chem Dec 21;276(51): Ward IM, Minn K, Jorda KG, Chen J. Accumulation of Checkpoint Protein 53BP1 at DNA Breaks Involves Its Binding to Phosphorylated Histone H2AX. J. Biol. Chem May 23;278(22): Watt FM, Lo Celso C, Silva-Vargas V. Epidermal stem cells: an update. Curr Opin Genet Dev Oct ;16(5): Weinberg WC, Denning MF. P21Waf1 control of epithelial cell cycle and cell fate. Crit Rev Oral Biol Med ;13(6): Wouters B, Skarsgard L. The response of a human tumor cell line to low radiation doses: evidence of enhanced sensitivity. Radiat Res Apr ;138(1 Suppl):

51 Wouters B, Sy A, Skarsgard L. Low-dose hypersensitivity and increased radioresistance in a panel of human tumor cell lines with different radiosensitivity. Radiat Res Oct ;146(4): Wouters B, Skarsgard L. Low-dose radiation sensitivity and induced radioresistance to cell killing in HT-29 cells is distinct from the "adaptive response" and cannot be explained by a subpopulation of sensitive cells. Radiat Res Nov ;148(5): Wykes SM, Piasentin E, Joiner MC, Wilson GD, Marples B. Low-dose hyperradiosensitivity is not caused by a failure to recognize DNA double-strand breaks. Radiat Res May ;165(5): Zgheib O, Pataky K, Brugger J, Halazonetis TD. An oligomerized 53BP1 tudor domain suffices for recognition of DNA double-strand breaks. Mol Cell Biol Feb ;29(4):

52