Molecular Determinants of Radiation Response

Current Cancer Research Series Editor Wafik El-Deiry For other titles published in this series, go to www.springer.co...

Author: Theodore L. DeWeese | Marikki Laiho

44 downloads 1801 Views 7MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Current Cancer Research

Series Editor Wafik El-Deiry

For other titles published in this series, go to www.springer.com/series/7892

Theodore L. DeWeese Marikki Laiho ●

Editors

Molecular Determinants of Radiation Response

Editors Theodore L. DeWeese Department of Radiation Oncology and Molecular Radiation Sciences Johns Hopkins University School of Medicine, 401 North Broadway Suite 1440 Baltimore, MD 21231 USA [email protected]

Marikki Laiho Department of Radiation Oncology and Molecular Radiation Sciences Johns Hopkins University School of Medicine Baltimore, MD 21231 USA [email protected]

ISBN 978-1-4419-8043-4 e-ISBN 978-1-4419-8044-1 DOI 10.1007/978-1-4419-8044-1 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2011922499 © Springer Science+Business Media, LLC 2011 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)

Preface

Every day, mammalian cells accumulate an estimated 100,000 lesions in their DNA resulting from exposure to reactive oxygen species, chemical deterioration of their bases, and exposure to exogenous agents such as ultraviolet and ionizing radiation. Cells have evolved complex response mechanisms to recognize and repair this injury in order to maintain genomic integrity in the face of this unrelenting assault. The study of DNA damage response has a long and storied history beginning in the 1940s by scientists like Albert Kelner and Renato Dulbecco. Their work revealed the existence of enzymatic photoreactivation and ultimately laid the foundation for the idea that cells respond to DNA damage and that DNA damage repair does exist. Since that time and with the development of sophisticated molecular techniques, an evolving story regarding the cellular response to DNA injury and its importance has emerged. The spectrum of diseases that have benefited from this contemporary research effort is broad and includes virtually all fields where genotoxic stress from oxidative injury, radiation insult, and chemical exposure plays a role in disease initiation, evolution, and treatment. One of the most important areas where these modern studies have had an impact is in the area of cancer biology. Through these studies, we now have a firmer grasp on the molecular response of the cell to DNA injury and how these responses influence mutagenesis, cell cycle progression, DNA damage signaling, neoplastic transformation, and cancer therapy. This book reviews a number of these important topics and will serve to provide a review of key basic and translational aspects of each. We have divided the book into two general categories for the purpose of organization: (1) Molecular Basis of DNA Damage Responses, and (2) Modulation of Radiation Responses – Opportunities for Therapeutic Exploitation. In the first section, Dr. Redon and colleagues provide a review of the histone variant, H2AX, and the role of this protein in the early response of the cell to DNA injury. Dr. Bunz discusses key proteins involved in ATM-dependent signaling of DNA injury and the importance of this protein in controlling DNA damage repair. Similarly, Dr. Lobrich, and colleagues discuss the role of signaling molecules and regulators of a complex signaling system that modulates cellular progression through the cell cycle following radiation injury and the link to repair of DNA damage. Drs. Shen and Falbo provide an overview on the critical role of chromatin structure and

v

vi

Preface

proteins associated with chromatin remodeling in the response of the cell to DNA damage and in the maintenance of DNA fidelity. Finally, one of the most versatile model systems that has served to elucidate specific DNA damage response pathways at the organismal level is Caenorhabditis elegans. Data derived from this model have been key to our understanding of data generated in higher organisms, including mammalian cells. In their chapter, Drs. Bailly and Gartner review important data on DNA damage signaling, repair, and cell fate generated using this model system. The book then transitions into a discussion of DNA damage response topics that have therapeutic relevance. First, Dr. Hammond and her colleagues discuss cellular and tumor hypoxia, a critically important microenvironmental condition of many human tumors. They review the molecular underpinnings of the hypoxic state, and how these factors alter radiation-induced DNA injury and repair. Collectively, the chapters by Drs. Vischioni, et al., Freytag, et al., Yazlovitskaya and Hallahan, and Dunn, et al., provide in-depth analyses of molecularly based therapies directed toward a variety of cellular targets, all of which have potential to modulate the response of the cell to radiation-induced DNA damage for therapeutic benefit. This includes discussions on (1) inhibitors of specific proteins known to be involved in DNA strand break repair; (2) development of viral and nonviral gene therapy systems that enhance radiation injury through a variety of DNA-directed mechanisms; (3) identification of pro-survival proteins, induced by radiation in tumor vasculature that can serve as molecular targets for radiation-modifying drugs; and (4) the potential for anti-EGFR agents in combination with radiation to substantially improve the therapeutic benefit of radiation therapy. Next, Dr. Roti Roti, et al., provide a comprehensive review of heat effects on signaling proteins, nuclear matrix-associated proteins and chromatin remodeling, and demonstrate how these heat-induced effects result in substantial alterations in cellular response to radiation. Finally, Dr. Drake reviews the evolving knowledge on the immunological effects of radiation, an area of investigation that has great potential to change the future of cancer care. Together, these chapters are a collection of contemporary works on DNA injury and the cellular response associated with it. While not every topic in the DNA damage response domain could be reviewed in a monograph of this size, we do believe the authors have done an outstanding job in providing timely and relevant discussions on their respective subjects, allowing the reader to become more familiar with the field and where the future lies within it. We firmly believe the information contained in this book underscores the significance of DNA damage response in cancer research and the need for continued investigation in this area in order to make substantive progress toward eliminating the suffering associated with cancer. We would fully concur with the opinion expressed by Dr. Bruce Alberts when he said: “If I were the czar of cancer research, I would give a higher priority to recruiting more of our best young scientists to decipher the detailed mechanisms of both apoptosis and DNA repair, and I would give them the resources to do so” (ref: Alberts, B, Science, 320:19, 2008). We hope that his sentiment and this book will

Preface

vii

provide inspiration to those young scientists seeking to work in an area with great potential and importance for our collective future. The editors wish to acknowledge Ms. Debbie Stankowski for her tireless efforts in collecting and organizing all of the manuscripts from our illustrious contributors. Baltimore, MD

Theodore L. DeWeese Marikki Laiho

Contents

Part I Molecular Basis of the DNA Damage Responses 1 H2AX in DNA Damage Response........................................................... Christophe E. Redon, Jennifer S. Dickey, Asako J. Nakamura, Olga A. Martin, and William M. Bonner

3

2 DNA Damage Signaling Downstream of ATM...................................... Fred Bunz

35

3 Checkpoint Control Following Radiation Exposure............................. Markus Lobrich, Aaron A. Goodarzi, Tom Stiff, and Penny A. Jeggo

53

4 Chromatin Responses to DNA Damage................................................. Karina Falbo and Xuetong Shen

79

5 Caenorhabditis elegans Radiation Responses........................................ 101 Aymeric Bailly and Anton Gartner Part II Modulation of Radiation Responses: Opportunities for Therapeutic Exploitation 6 Hypoxia and Modulation of Cellular Radiation Response.................. 127 Ester M. Hammond, Monica Olcina, and Amato J. Giaccia 7 Inhibitors of DNA Repair and Response to Ionising Radiation................................................................................................... 143 Barbara Vischioni, Nils H. Nicolay, Ricky A. Sharma, and Thomas Helleday 8 Gene Therapy and Radiation.................................................................. 173 Svend O. Freytag, Kenneth N. Barton, Farzan Siddiqui, Mohamed Elshaikh, Hans Stricker, and Benjamin Movsas ix

x

Contents

9 Molecular Targeted Drug Delivery Radiotherapy................................ 187 Eugenia M. Yazlovitskaya and Dennis E. Hallahan 10 EGFR Signaling and Radiation.............................................................. 201 Emily F. Dunn, Shyhmin Huang, and Paul M. Harari 11 Thermal Modulation of Radiation-Induced DNA Damage Responses................................................................................... 227 Joseph L. Roti Roti, Robert P. VanderWaal, and Andrei Laszlo 12 Radiation-Induced Immune Modulation............................................... 251 Charles G. Drake Index.................................................................................................................. 265

Contributors

Aymeric Bailly Wellcome Trust Centre for Gene Regulation and Expression, University of Dundee, Dow Street, Dundee DD1 5EH, UK Kenneth N. Barton Department of Radiation Oncology, Henry Ford Hospital, 2799 West Grand Boulevard, Detroit, MI 48202, USA William M. Bonner Laboratory of Molecular Pharmacology, Center for Cancer Research, National Cancer Institute, 9000 Rockville Pike, Bethesda, MD 20892, USA Fred Bunz Associate Professor, Department of Radiation Oncology and Molecular Radiation Sciences, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, David H. Koch Cancer Research Building (CRB2), Room 453, 1550 Orleans Street, CRB II, Room 462, Baltimore, MD 21231, USA Jennifer S. Dickey Laboratory of Molecular Pharmacology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USA Charles G. Drake Johns Hopkins Kimmel Cancer Center, 1650 Orleans Street, Baltimore, MD 21231, USA Emily F. Dunn Department of Human Oncology, University of Wisconsin Comprehensive Cancer Center, K4/332, 600 Highland Avenue, Madison, WI 53792, USA Mohamed Elshaikh Department of Radiation Oncology, Henry Ford Hospital, 2799 West Grand Boulevard, Detroit, MI 48202, USA xi

xii

Contributors

Karina Falbo Department of Carcinogenesis, The University of Texas M.D. Anderson Cancer Center, MDA SP/RD, 1808 Park Road 1-C, Box 389 Smithville, TX 78957, USA Svend O. Freytag Department of Radiation Oncology, Henry Ford Health System, One Ford Place, 5D, Detroit, MI 48202, USA Anton Gartner Wellcome Trust Centre for Gene Regulation and Expression, University of Dundee, Dow Street, Dundee DD1 5EH, UK Amato J. Giaccia Department of Radiation Oncology, Center for Clinical Sciences Research, Stanford University, Stanford, CA 94303-5152, USA Aaron A. Goodarzi Genome Damage and Stability Centre, University of Sussex, Brighton, East Sussex BN1 9RQ, UK Dennis E. Hallahan Department of Radiation Oncology, Washington University School of Medicine, 4511 Forest Park Avenue, Suite 200, St. Louis, MO 63130, USA Ester M. Hammond The Cancer Research UK/MRC Gray Institute for Radiation Oncology and Biology, University of Oxford, Old Road Campus Research Building, Churchill Hospital, Oxford OX3 7DQ, UK Paul M. Harari Department of Human Oncology, University of Wisconsin, 600 Highland Avenue, K4/336, Madison, WI 53792, USA Thomas Helleday Cancer Research UK-Medical Research Council, Gray Institute for Radiation Oncology and Biology, University of Oxford, Oxford, OX3 7DQ, UK and Department of Genetics Microbiology and Toxicology Stockholm University, Stockholm S-106 91, Sweden Shyhmin Huang Department of Human Oncology, University of Wisconsin, 600 Highland Avenue, K4/336, Madison, WI 53792, USA Penny A. Jeggo Genome Damage and Stability Centre, University of Sussex, Brighton, East Sussex BN1 9RQ, UK

Contributors

Andrei Laszlo Radiation and Cancer Biology Division, Department of Radiation Oncology, Washington University School of Medicine, St. Louis, MO 63108, USA Markus Lobrich Darmstadt University of Technology, Radiation Biology and DNA Repair, Darmstadt 64287, Hesse, Germany Olga A. Martin Laboratory of Molecular Pharmacology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USA Benjamin Movsas Department of Radiation Oncology, Henry Ford Hospital, 2799 West Grand Boulevard, Detroit, MI 48202, USA Asako J. Nakamura Laboratory of Molecular Pharmacology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USA Nils H. Nicolay Gray Institute for Radiation Oncology & Biology, University of Oxford, Old Road Campus Research Building, Roosevelt Drive, Oxford OX3 7DQ, UK Monica Olcina The Cancer Research UK/MRC Gray Institute for Radiation Oncology and Biology, University of Oxford, Old Road Campus Research Building, Churchill Hospital, Oxford OX3 7DQ, UK Christophe E. Redon Laboratory of Molecular Pharmacology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USA Joseph L. Roti Roti Radiation and Cancer Biology Division, Department of Radiation Oncology, Washington University School of Medicine, St. Louis, MO, USA Ricky A. Sharma Gray Institute for Radiation Oncology & Biology, University of Oxford, Old Road Campus Research Building, Roosevelt Drive, Oxford OX3 7DQ, UK Xuetong Shen Department of Carcinogenesis, The University of Texas, MD Anderson Cancer Center, MDA SP/RD, Unit 116, 1808 Park Road 1-C, Box 389, Smithville, TX 78957, USA

xiii

xiv

Contributors

Farzan Siddiqui Department of Radiation Oncology, Henry Ford Hospital, 2799 West Grand Boulevard, Detroit, MI 48202, USA Hans Stricker Department of Radiation Oncology, Henry Ford Hospital, 2799 West Grand Boulevard, Detroit, MI 48202, USA Tom Stiff Genome Damage and Stability Centre, University of Sussex, Brighton, East Sussex BN1 9RQ, UK Robert P. VanderWaal Radiation and Cancer Biology Division, Department of Radiation Oncology, Washington University School of Medicine, St. Louis, MO 63108, USA Barbara Vischioni Division of Radiation Oncology, European Institute of Oncology and University of Milan, Milan, Italy Eugenia M. Yazlovitskaya Department of Radiation Oncology, Vanderbilt University School of Medicine, B-902 Vanderbilt Clinic, 1301 22nd Ave South, Nashville, TN 37232-5671, USA

Part I Molecular Basis of the DNA Damage Responses

Chapter 1

H2AX in DNA Damage Response Christophe E. Redon, Jennifer S. Dickey, Asako J. Nakamura, Olga A. Martin, and William M. Bonner

Abstract Histone H2AX is a tumor suppressor, helping to preserve genome integrity. It does this by becoming massively and quickly phosphorylated at the sites of nascent DNA double-strand breaks (DSBs) in chromatin. In this chapter, we discuss the state of current knowledge concerning the various aspects of H2AX metabolism, including DSB formation, H2AX phosphorylation to form foci, foci size, and foci stoichiometry with DSBs. While H2AX is essential for efficient DSB repair, it is not essential for basic DSB repair. Mice lacking H2AX are viable under sterile conditions. However, H2AX plays an essential role in two processes necessary for long-term animal survival: immune system development and male fertility. We discuss the phosphorylation of H2AX during telomere dysfunction, interrupted replication/transcription, virus infection, and apoptosis. Phosphorylated H2AX foci serve as platforms for the recruitment of DNA repair and chromatin remodeling factors as well as factors involved in the cell-cycle checkpoint. Phosphorylated H2AX foci are also useful to detect DSBs related to cancer, senescence, and the radiation-induced bystander effect. We conclude by discussing clinical applications which are also coming to the forefront. These include applications in radiation biodosimetry, individual radiosensitivity, efficiency of chemotherapeutic agents to damage cancer and normal cells, and the evaluation of environmental toxins. Keywords g-H2AX foci • Double-strand break repair • Ionizing radiation • Cancer • Clinical applications • Environmental toxins

W.M. Bonner (*) Laboratory of Molecular Pharmacology, Center for Cancer Research, National Cancer Institute, 9000 Rockville Pike, Bethesda, MD 20892, USA e-mail: [email protected] T.L. DeWeese and M. Laiho (eds.), Molecular Determinants of Radiation Response, Current Cancer Research, DOI 10.1007/978-1-4419-8044-1_1, © Springer Science+Business Media, LLC 2011

3

4

C.E. Redon et al.

1.1 Introduction Damage to cellular constituents including DNA is a routine risk of cellular existence, and cells maintain an impressive array of pathways dedicated to repairing DNA lesions and hopefully restoring the damaged DNA to its original sequence. While most single-stranded lesions are easily repaired, DNA doublestrand breaks (DSBs) present special challenges. DSBs compromise the redundancy built into the DNA molecule so that restoring the DNA to its original sequence or even to accurately rejoin the broken fragment is not certain. Also, the broken chromatin ends may rejoin with fragments other than their original partners, leading to chromosomal rearrangements. In addition, if DSBs persist into mitosis, fragments of chromatids may be lost during anaphase, leading to an unbalanced genome. The DNA DSB, therefore, is considered one of the most serious types of DNA damage. One mechanism that has evolved in eukaryotes involves the formation of foci at DSB sites, which serve to concentrate many molecules of various types of proteins involved in DNA repair, chromatin remodeling, and cell-cycle control (McKinnon and Caldecott 2007; Bonner et al. 2008). The foundation of these foci involves the phosphorylation of a histone H2A isoform, H2AX, on a conserved C-terminal sequence centered on an omega-4-serine residue (Rogakou et al. 1998). Upon formation of a DSB, many H2AX molecules, hundreds to thousands in mammals and dozens in budding yeast, become phosphorylated within minutes in the chromatin flanking the DSB site. H2AX phosphorylated on the omega-4-serine is named g-H2AX (Rogakou et al. 1998, 1999; Bonner et al. 2008). In this chapter, we discuss the various aspects of g-H2AX involvement in DNA damage repair and possible applications of the response for human health.

1.2 DSB Formation and H2AX Phosphorylation The genetic information of eukaryotic cells is maintained on DNA fibers, which are organized and compacted into nucleosomes, the fundamental chromatin subunit (Campos and Reinberg 2009) (Fig. 1.1a left panel). In humans, this compaction enables ~2 m of DNA fibers to fit inside a 0.01 mm diameter nucleus. The nucleosome comprises 145 bp DNA and a core histone octamer containing two molecules of each of the four core histone species, H2A, H2B, H3, and H4. Each nucleosome is separated by a minimum of 20 bp of DNA (linker DNA) that is associated with histone H1 (Redon et al. 2002; Campos and Reinberg 2009). H2AX accounts for about 10% of the total H2A protein in normal human fibroblasts (Rogakou et al. 1998). Besides phosphorylation of H2AX on omega-4 serine, it contains two more known phosphorylation sites, an N-terminal serine and a C-terminal tyrosine (Pantazis and Bonner 1981; Cook et al. 2009; Krishnan et al. 2009). The designation g-H2AX is independent of the phosphorylation state of these other two sites.

1 H2AX in DNA Damage Response

5

As opposed to most core histones that are synthesized during S phase, H2AX synthesis is both DNA replication-dependent and -independent (Wu et al. 1982). There is a single H2AX gene in mammals, which is structured like replication-linked histone genes. The H2AX gene lacks introns and contains a stem–loop motif just downstream of the translation termination site, which serves as a transcription termination site; however, there is also a considerable amount of read-through transcription to a polyA site about 1 Kb further downstream. The result is a protein synthesized at a basal level throughout the cell cycle and in quiescent cells with a peak during S-phase (Wu et al. 1983). The cell-cycle-independent synthesis of H2AX may serve its function in DNA repair as DNA lesions can happen in any stage of the cell cycle. Thus, newly synthesized H2AX would be available as part of the cellular DNA damage response.

1.2.1 Characteristics of g -H2AX Focal Formation and Removal H2AX, in common with the other histone species, is modified posttranslationally by acetylation, biotinylation, methylation, phosphorylation, and ubiquitylation (Pantazis and Bonner 1981; Redon et al. 2002; Chew et al. 2006; Ikura et al. 2007; Pinto and Flaus 2010). Most of these modifications happen on both N-terminal and C-terminal tails and have specific functions in the cell life such as DNA metabolism, DNA condensation, or DNA damage repair. While H2AX and other H2A species share highly conserved core structure, H2AX also contains a unique C-terminal omega-4-serine residue, also highly conserved through evolution (Redon et al. 2002; Martin et al. 2003; Dickey et al. 2009a) (Fig. 1.1a right panel). However, there is a linker sequence connecting these two conserved regions, which does not appear to be conserved either in sequence or length. The variability of the linker length may be related to the variations in internucleosomal lengths of organisms through evolution (Pinto and Flaus 2010). Three related PIK-kinases, ATM, ATR, and DNA-PK, are primarily responsible for H2AX phosphorylation on the omega-4-serine (Bonner et al. 2008). An anti-g-H2AX antibody raised to the H2AX phosphorylated C-terminal peptide CKATQAS(PO4)QEY in humans revealed that the g-H2AX molecules are localized at DNA DSB sites (Rogakou et al. 1999).

1.2.1.1 Size of g -H2AX Foci When the maximal extent of H2AX phosphorylation is measured by 2D gel electrophoresis on cell types with different fractions of H2AX relative to total H2A, it is the fraction of phosphorylated H2AX relative to total H2AX that remains relatively constant regardless of the relative H2AX content, from 2% in lymphocytes and HeLa to 20% in SF268 (Rogakou et al. 1998). This fraction is ~0.03% g-H2AX per DSB (Redon et al. 2003). If H2AX is randomly distributed in the chromatin, 0.03%


7

corresponds to ~2 Mbp in mammals. In haploid yeast, which contain 1/400th the amount of the mammalian DNA complement, and in which the H2AX homolog comprises ~95% of the total H2A, stoichiometric measurements also result in a value ~0.03% of the H2A(X) phosphorylated per DSB. In contrast to these stoichiometric determinations, immunocytochemistry suggests that g-H2AX molecules cover much more than 2 Mbp DNA (Fig. 1.1b). Measurements in muntjac and human mitotic cells have revealed that g-H2AX foci formed postirradiation cover ~20 Mbp or more of chromatin (Rogakou et al. 1999). The two different estimates of focus size from stoichiometry vs. immunocytochemistry, 2 vs. 20 Mbp, are consistent with a model in which only ~10% of the H2AX is phosphorylated at any

Fig. 1.1 H2AX and g-H2AX foci. (a) Left panel: H2AX is an element of the nucleosome, the fundamental repeating unit of chromatin. The nucleosome core particles consist of approximately 147-bp DNA wrapped around a histone octamer containing two of each core histones H2A, H2B, H3, and H4 and connected by a “linker DNA” of 20 nucleotides minimum length. Linker histone (H1 and its isoforms), binding the “linker DNA,” sits at the top of the nucleosome at the DNA entry and exit points (to simplify, H1 tails are not represented). Because H2AX represents 2–20% of H2A isoforms, a nucleosome could in principle contain either H2A or H2AX or both H2A and H2AX polypeptides. Both H2A and H2AX are composed of a N-terminal tail and a central globular domain containing the lysine 119, which is ubiquitylated during the processes of DNA repair and histone exchange. H2AX contains an exclusive C-terminal tail, consisting of two evolutionarily conserved motifs (light gray and black) separated by a linker varying in sequence and length between animal and/or plant species (dark gray) (right panel). H2AX C-terminal conserved motif (in black) contains the omega-4 serine that is phosphorylated upon DNA DSB formation (asterisk) to form g-H2AX. (b) Model for g-H2AX focus in mammal and yeast. Upper panel: Stoichiometric studies show that ~2,000 H2AX molecules are phosphorylated at a DSB site in human and ~50 in yeast (red bars). However, microscopy and chromatin immunoprecipitation have shown that the g-H2AX molecules are distributed over a substantially larger region than expected. These observations suggest that ~10% of H2AX in the focus are phosphorylated at any time (green bars). Lower panel: g-H2AX formation in yeast and human lymphocytes 30 min after 0.4 and 200 Gy, respectively. Yeasts insets were approximately scaled to lymphocytes to show the relative g-H2AX focus sizes between species (left image). The right image is a enlargement of the inset from the left image. (c) Schematic representation of the g-H2AX foci distribution being dependent on the nature of irradiation. While g-H2AX foci are spread in cell nuclei irradiated with low-LET (X-rays, g-rays), g-H2AX foci are spaced closely together along the particle track in nuclei after high-LET irradiation. The track is a direct visualization of the ionizing radiation track led by a particle traveling through a cell nucleus. (d) Heterochromatin is refractory for g-H2AX foci formation. Lymphocytes were irradiated with 6 Gy of g-irradiation and harvested 5 min later. The left image shows a representative single optical slice of g-H2AX foci distribution in lymphocyte after g-irradiation, while the g-H2AX signal was subtracted in the middle image. Heterochromatin masses are represented by strong PI staining. To better visualize the position of g-H2AX foci in the nucleus, g-H2AX were replaced with dotted line (left image). Note that g-H2AX foci preferentially localize in the low PI staining regions (euchromatin) and/or at the border of heterochromatin masses. (e) Are all DSBs labeled by g-H2AX? g-H2AX detection in blood white cells 30 min after 0.6 Gy g-irradiation of blood sample. In opposite to lymphocytes (Ly.) where cells show a robust g-H2AX foci formation, most neutrophils (Ne.) are deprived of g-H2AX staining. This observation could lead to consider that in some cell types, DSB formation does not lead to g-H2AX focus formation. (f) g-H2AX focus spreading on Muntiacus muntjak mitotic chromosomes at the indicated time after 0.6 Gy-g-IR. (Part F modified with permission from Rogakou et al. (1999), The Rockfeller University Press). g-H2AX: green; DNA: red

8

C.E. Redon et al.

one time in the focal region flanking the DSB site. A similar situation exists in budding yeast (Shroff et al. 2004). Thus, each g-H2AX focus appears to subsume very similar fractions of the genome in different species (Fig. 1.1b). 1.2.1.2 DSB/g -H2AX Stoichiometry One fundamental question about using g-H2AX as a marker for DSB induction is whether all DSBs formed by irradiation result in g-H2AX focal formation. Specifically, how well do DSB numbers calculated from enumerating g-H2AX foci correlate with those from pulsed-field gel electrophoresis measurements (PFGE). Estimates of DSB numbers from PFGE have varied in the range 3.6–9.7 DSBs per 1,000 Mbp per Gy (Whitaker et al. 1995), which translates to 23–62 per G1 human cell (6,400 Mbp) per Gy, but most values clustered in the 30–40 DSBs per G1 human cell (6,400 Mbp) per Gy (Lobrich et al. 1994, 1995; Cedervall et al. 1995). Variations in this number can result from the cell type, the oxygen content of the cell and tissue, and the type of radiation (low LET vs. high LET). Rothkamm and Lobrich (2003) investigated the induction of DSBs in primary human fibroblasts in the G1 phase of the cell cycle by examining both g-H2AX formation and by PFGE. An average of 36 g-H2AX foci per Gy per cell was visible after 3 min irradiation using doses of 0.2 or 2 Gy, while PFGE studies performed in the dose range between 10 and 80 Gy gave 39 DSBs per Gy per cell. Thus, their study presented the evidence that the numbers of g-H2AX foci are almost the same as the number of DSBs. 1.2.1.3 Spatial Distribution of g -H2AX Foci The pattern of g-H2AX foci observed in nuclei after radiation exposure is not uniform. First, it is dependent on the type of radiation. Several studies have shown that exposure of cells to high-LET particles is more damaging than to low-LET radiation (photons, X-rays/g-rays) (Goodhead and Nikjoo 1989; Cucinotta et al. 2000, 2003). Immunofluorescent detection of g-H2AX reveals that the foci appear to be homogeneously distributed throughout the nuclei following low-LET irradiation (Redon et al. 2009). In contrast, high-LET irradiation produces streaks of g-H2AX foci along the tracks of individual particle traversals (Fig. 1.1c) (Desai et al. 2005). The repair of DSBs is much slower when induced by high-LET than by low-LET (Jenner et al. 1993), reflecting the complexity of DSBs induced by such radiation. Immunofluorescence of g-H2AX confirmed observations made previously with pulsed-field electrophoresis by showing a slower disappearance of g-H2AX foci after high-LET than after low-LET irradiation (Leatherbarrow et al. 2006). Second, g-H2AX focal formation in nuclei is dependent on the level of chromatin compaction, regardless of the radiation type. Several studies in both yeast and mammalian cells have shown that H2AX is refractory to phosphorylation in compacted chromatin


9

(heterochromatin), occurring preferentially in open chromatin (euchromatin) (Cowell et al. 2007; Kim et al. 2007; Goodarzi et al. 2009). Thus g-H2AX focal formation is observed more often in euchromatin and at the euchromatin/heterochromatin border than in heterochromatin masses (Fig. 1.1d). The repair of these DSBs linked to heterochromatin regions (10–25% of total DSBs) requires ATM/ KAP-1 signaling for repair (Goodarzi et al. 2008). Third, not all cell types respond similarly with g-H2AX focal formation. Although comparison of irradiation-induced H2AX foci in different animal tissues, cell cultures, and primary cells from different individuals indicate a high degree of the assay reproducibility and it’s main dependence on the dose rather than cell type (Martin et al. 2004; Rube et al. 2008; Redon et al. 2009), there are some exceptions. Cellular DNA content needs to be taken into account. In normal human lymphocytes, an average of 15 g-H2AX foci per Gy are formed (Redon et al. 2009), which is low compared to estimates usually reported for cultured cells (Yasui 2004; Kato et al. 2006; Leatherbarrow et al. 2006). However, G0 lymphocytes also possess less DNA per cell on average than do proliferating cells. Kato et al. (2008) showed that twice as many g-H2AX foci are observed in metaphase cells relative to G1 cells at any given dose. In contrast to lymphocytes, neutrophils exhibit a poor g-H2AX response, showing few or no foci in response to irradiation (Fig. 1.1e). This observation correlates with the findings that unlike lymphocytes, mature neutrophils are unable to repair DNA breaks induced by g-irradiation, perhaps due to their lack of proteins involved in DNA repair (Terai et al. 1991; Ajmani et al. 1995; Kurosawa et al. 2003). Since the postmitotic life span of neutrophils is ~2 days, DNA repair and genomic integrity may not be a priority. However, together, these results indicate that if a cell is capable of responding to DNA damage, then most if not all DSBs induce g-H2AX focus formation. 1.2.1.4 Temporal Distribution of g -H2AX Foci Mass staining and immunocytochemistry both show an increase in g-H2AX amounts relative to radiation dose, and both show a rapid increase in g-H2AX levels for 30 min after exposure followed by a slower decline (Rogakou et al. 1999) (Fig. 1.1f). Short-term analysis of fixed cells reveals the presence of radiation-induced g-H2AX foci in cells 1–2 min after irradiation, and 10 min postexposure, the foci are usually well formed and contain DSB repair factors such as 53BP1, MRE11, and RAD50, indicating that active repair is in progress (Paull et al. 2000; Martin et al. 2008). Numbers of irradiation-induced foci decrease as DSBs are rejoined. The majority of foci disappear by 8 h postirradiation (Rogakou et al. 1999; Martin et al. 2002, 2004; Cucinotta et al. 2008; Redon et al. 2009). However, a substantial number of g-H2AX foci are detectable from 48 h to 7 days after radiation exposures greater than 1 Gy (Redon et al. 2009; Bhogal et al. 2010). It has been suggested that these residual g-H2AX foci may be a critical factor determining cell survival (Banath et al. 2010). While their exact properties are obscure, these residual irradiation-induced foci can be attributed to incomplete or stalled repair of more complex DSB lesion,

10

C.E. Redon et al.

faulty rejoining of DSBs, lethal DNA lesions, persistent chromatin alterations, apoptosis, activity of several kinases and phosphatases, and checkpoint signaling (Goodhead 1994; Nikjoo et al. 2001; Banath et al. 2010).

1.2.1.5 Removal of g -H2AX Foci The mechanism of g-H2AX removal after DSB repair is still unclear. Several protein species including the PP2A and the PP4 family of phosphatases and WIP1 are involved in g-H2AX dephosphorylation (Chowdhury et al. 2005, 2008; Keogh et al. 2006; Macurek et al. 2010). Chromatin remodeling complexes such as TIP60 in human cells and INO80-C and SWR-C in yeast cells also contribute to g-H2AX removal from chromatin by histone exchange (Thiriet and Hayes 2005). The association between TIP60 and UBC13 is enhanced after DNA damage and correlates with H2AX acetylation on lysine 5, which is necessary for H2AX ubiquitylation by UBC13 and release from chromatin (Kusch et al. 2004; Ikura et al. 2007). It is suggested that both direct dephosphorylation of g-H2AX and its removal by histone exchange occurs simultaneously distant from the DSB site and near the DSB site, respectively (Chowdhury et al. 2005). Histone removal during repair would allow some specific factors to access and to rejoin DNA directly at the DSB site. This observation is correlated with the finding that g-H2AX loss is significant at the break site where DNA becomes single stranded by resection (Shroff et al. 2004). Nucleosome-displaced g-H2AX could also be dephosphorylated by phosphatase(s). Thus, in budding yeast Pph3 is thought to dephosphorylate g-H2AX only after its removal from chromatin (Keogh et al. 2006).

1.2.1.6 Changes in g -H2AX Kinetics DSB repair deficiencies during senescence and aging and indirect cellular responses such as bystander effects can compromise typical g-H2AX phosphorylation/dephosphorylation kinetics. For example, radiation-induced g-H2AX foci and DSB repair complexes form substantially more slowly in cells from older healthy donors and even more slowly in cells from Werner syndrome (premature aging) patients. Cells from younger Werner syndrome patients recruit DNA repair proteins to g-H2AX foci at rates similar to older healthy donors (Martin et al. 2008). This suggests that an ageassociated decline in the integrity of the genome may be due to the decreased speed with which aging cells can assemble DNA repair machinery at a damage site. One possible consequence of slower focal formation is incorrect repairs, producing cells more prone to cancer. DNA repair and maintenance protein deficiencies also lead to altered g-H2AX responses including delayed repair kinetics and/or higher g-H2AX focal formation (Smilenov et al. 2005; Lavin 2008; Rube et al. 2008; Hamilton 2009). These altered focal kinetics may form the basis for clinical or environmental applications of the g-H2AX assay and are discussed below.


11

1.2.2 g -H2AX Formation as an Essential Step in Biological Processes H2AX is phosphorylated not only in response to DNA damage induction by external agents but also in response to programmed biological process. In fact, H2AX-null mice live normally in a germ-free environment, and while they are hypersensitive to irradiation, there are two other defects preventing them from surviving in the wild (Celeste et al. 2002).

1.2.2.1 Immune System Development H2AX-null mice are incapable of performing class switch recombination, leading to compromised immunity. In addition, these mice exhibit lower numbers of lymphocytes that have undergone V(D)J recombination (Celeste et al. 2002). Since these breaks are programmed in immune system development, their presence would not be expected to be stressful, but they do make the cell more susceptible to chromosome translocations and B-cell malignancies when incorrectly performed.

1.2.2.2 Male Fertility Male H2AX-null mice are sterile (Celeste et al. 2002). Meiosis is characterized by the formation of programmed DSBs catalyzed by SPO11, which triggers the initiation of homologous recombination (Li and Ma 2006). During homologous recombination in meiosis I, g-H2AX is present in the chromatin and disappears as the homologous chromosomes pair in late pachytene (Celeste et al. 2002). It is important to note that H2AX-null male mice show the normal appearance of SPO11-dependent RAD51 foci. In contrast, female H2AX-null mice remain fertile. This indicates that the deficiency is male-specific and that H2AX is not essential for homologous recombination during meiosis I. SPO11-dependent g-H2AX foci disappear as the homologous chromosomes pair in late pachytene (Bellani et al. 2005). However, the X-Y chromosome pair retains g-H2AX staining even after they have condensed to form the XY body (also known as the sex body). In contrast to SPO11-dependent g-H2AX formation by ATM, XY bodyinduced phosphorylation of H2AX is regulated by ATR (Bellani et al. 2005). In H2AX-null male mice, there is a higher incidence of the X and Y chromosomes failing to pair during meiosis I, and often these two chromosomes are found distant from each other in pachytene spermatocytes (Celeste et al. 2002). The unique feature of the X and Y chromosomes is that they share a rather short pseudoautosomal region between them, which is used for pairing. Apparently, H2AX makes the pairing process more efficient, which is critical only for the X-Y pairing. Thus, it is not known whether g-H2AX is being induced by a DSB somewhere in the XY body or whether in this case, g-H2AX may be responding to a different stimulus.

12

C.E. Redon et al.

1.2.3 g -H2AX Formation in Other Biological Processes 1.2.3.1 Telomere Dysfunction Telomeres, the repetitive DNA sequence motifs and specialized proteins that cap the ends of linear chromosomes, are essential for chromosomal stability. However, since the DNA polymerase cannot duplicate the DNA to the ends of the double helix during replication, telomeric DNA consequently becomes shorter with each cell division in human somatic cells (de Lange 2004). After many cell divisions, the telomeric DNA becomes too short to bind the basic capping structure and becomes an eroded telomere. The eroded telomere structure is a type of double-stranded DNA damage, which also activates g-H2AX focus formation (Takai et al. 2003). In somatic cells, which lack telomerase, the enzyme that adds telomeric DNA repeats, eroded telomeres persistently induce formation of g-H2AX foci and may be highly susceptible to chromosomal rearrangements, similar to more canonical DSBs.

1.2.3.2 Replication/Transcription Accidental DSBs can be generated from DNA single-stranded breaks (SSBs) during replication and/or transcription. DNA SSBs are produced by a variety of exogenous and endogenous sources including radiation, ROS, chemicals, defective DNA repair processes, and inhibited enzymes such as DNA topoisomerase I (top1) or topoisomerase II (top2) (Bonner et al. 2008). Top1 is a ubiquitous enzyme that removes DNA supercoiling by producing top1 cleavage complexes. The cleavage complexes are generally widely distributed throughout the genome and are transient. However, when the rapid resealing of the TOP1 cleavage complex is inhibited by common DNA base alterations (abasic sites, mismatches, oxidized bases, carcinogenic adducts, and UV lesions) or by the top1 inhibitors, DNA polymerase (or RNA polymerase II during transcription) is stalled at the site of the lesion (Pommier et al. 2006). This results in the formation of a DNA DSB and the appearance of a g-H2AX focus. Likewise, the stabilization of top2-DNA covalent complexes by a top2 poison such as etoposide or doxorubicin also may generate DNA DSBs (Banath and Olive 2003).

1.2.3.3 Virus Infection g-H2AX formation by virus have been reported and is due to both viral DNA integration as well as the hijacking of cellular processes involved in DNA metabolism by virus oncoproteins (Duensing and Munger 2002; Daniel et al. 2004; Xie and Scully 2007). Viral infection can create DNA lesions in the host cell, which activate the DNA damage response (DDR) pathway. H2AX is phosphorylated at the sites of retroviral infection, but g-H2AX is not required for the repair itself (resealing)


13

of these integration sites, supporting the notion that H2AX would be involved in anchoring broken DNA together (Daniel et al. 2004). The murine g-herpesvirus 68 (gHV68) actively induces g-H2AX through expression of the viral kinase, orf36, which plays a vital role for viral replication in infected animals (Tarakanova et al. 2007). The role of H2AX for g-HV68 infection is shown by the fact that, 1 week postinfection, gHV68 amounts in H2AX−/− macrophages were almost 100-fold lower than in H2AX+/+ hosts. 1.2.3.4 Apoptosis Apoptosis is a tightly regulated multistep pathway that is responsible for cell death during development as well as in response to cellular stress. Apoptosis leads to a variety of morphological changes in a cell, including loss of membrane asymmetry and attachment, cell shrinkage, nuclear fragmentation, chromatin condensation, and DNA fragmentation. H2AX phosphorylation occurs during apoptosis concurrently with the initial appearance of high molecular weight DNA fragments but before the appearance of internucleosomal DNA fragments (Rogakou et al. 2000). g-H2AX formation is primarily due to the action of DNA-PK and JNK-1 kinases (Lu et al. 2006; Mukherjee et al. 2006; Sluss and Davis 2006). During apoptosis, g-H2AX phosphorylation is usually robust, showing a pan-nuclear staining or an “apoptotic g-H2AX ring” staining (Mukherjee et al. 2006; Solier and Pommier 2009). In addition to phosphorylation at S139, the phosphorylation of another H2AX C-terminal residue, tyrosine 142, has a role in apoptosis (Cook et al. 2009; Stucki 2009) (Fig. 1.2a). This phosphorylation, generated by the WSTF kinase, is constitutive, and dephosphorylation by the EYA1/3 phosphatases occurs after DNA damage. Tyrosine 142 dephosphorylation is required for proper DNA repair by allowing g-H2AX formation and its binding to MDC1, both mechanisms indispensable for a correct DNA damage response. If for some reason after DNA damage, tyrosine 142 dephosphorylation is aborted, the cell response will change to apoptosis.

1.3 g -H2AX in DNA DSB Repair H2AX is a multirole player in the DNA damage response. First, H2AX phosphorylation creates a zone around the DSB site that facilitates the concentration of DNA repair and signaling proteins that will participate in DNA resection and cell-cycle checkpoint (Celeste et al. 2002; Bonner et al. 2008). Second, g-H2AX recruits cohesins to promote sister chromatid-dependent recombinational repair (Unal et al. 2004; Xie et al. 2004). Third, g-H2AX participates directly, or by recruiting other factors, to chromatin remodeling to support DSB repair (van Attikum and Gasser 2005). Finally, preventing the dissociation of broken ends is thought to be another task of g-H2AX (Bassing and Alt 2004).

14

C.E. Redon et al.

Fig. 1.2 g-H2AX formation and role in recruitments of DNA repair proteins. (a) H2AX in the crossroad between cell survival and cell death in response to DNA damage. In addition to the phosphorylation occurring at serine 139 in response to DNA damage, the phosphorylation of another H2AX C-terminal residue, tyrosine 142, modulates apoptosis. WSTF phosphorylates Tyr 142 constitutively, while dephosphorylation by EYA1/3 occurs after DNA damage. During the DNA repair response, MDC1 binds to phosphoserine 139, binding that requires dephosphorylation of phosphotyrosine 142. In the case of the programmed cell death, the phosphotyrosine is not dephosphorylated and results in both the inhibition of repair factors binding to g-H2AX and the recruitment of proapoptotic factors such as the JNK-1 kinase. (b) Several kinases are involved in g-H2AX formation. If they have a redundant role in H2AX modification, their primary role in H2AX phosphorylation is dependent on the origin of the DNA damage. ATM is the primary kinase for H2AX phosphorylation after radiationinduced DNA damage (left) while ATR role is preponderant after replication stress-induced DSBs (middle). Finally both DNA-PK and JNK-1 kinases are involved in H2AX phosphorylation in response to DNA fragmentation in apoptosis (left). (c) g-H2AX plays a role in the


15

Fig. 1.2 (continued) recruitment of 53BP1 and BRCA1. (1) Following DSB formation, the initial response consists in the binding of the MRN complex (Mre11-Rad50-Nbs1) to the DSB ends and the ATM enrollment and activation. The outcome of ATM activation results in the subsequent phosphorylation of several DNA repair factors including H2AX phosphorylation on serine 139 (g-H2AX). MDC1 is then recruited at the DNA breaks by binding to g-H2AX via its C-Terminal BRCT domains. The MRN-ATM complex is also recruited at the DSB sites via the binding of NBS1 to the CK2-dependent constitutive phosphorylation of MDC1. The increased ATM enrollment results in a feedback loop with accrued phosphorylation of several DNA repair factors including MDC1, NBS1, and H2AX. (2) The next step in the response to DSB formation consists in chromatin remodeling. ATM phosphorylation of MDC1 allows the recruitment of the RNF8-UBC13 E3 ubiquitin ligase complex and consecutively H2A and H2AX ubiquitylation. H2A and H2AX modification, in turn, mobilize RNF168-UBC13, the other E3 ubiquitin ligase complex, amplifying the ubiquitin–histone conjugates at the DSB sites. (3) Subsequently, the ubiquitylation of both H2A and H2AX provides docking sites for RAP80, a component of the BRCA1-A complex, resulting in BRCA1 recruitment via the adaptor protein ABRA1. De-ubiquitylation of histones can occur by the action of the BRCA1-A complex-associated deubiquitylating enzyme BRCC36. It is not known if at this point BRCA1 itself participates in additional H2A or H2AX ubiquitylation. Finally, histone modifications, including histone ubiquitylation, also cause an alteration in chromatin remodeling, allowing exposure of previously hidden methylated H3 and H4 histone tails. 53BP1 TUDOR domains, then, bind unmasked methylated H3 and H4, allocating 53BP1 recruitment to DSB sites

16

C.E. Redon et al.

Following DSB induction, ATM, ATR, and DNA-PK, three members of the phosphatidylinositol-3 kinase (PIKKs) family with specialized and overlapping functions, are activated and phosphorylate proteins involved in DNA repair including H2AX (Bonner et al. 2008) (Fig. 1.2b). If the three PIKK kinases can potentially phosphorylate H2AX, the kinase responsible for H2AX phosphorylation at a given time is dependent on the manner of damage induction (Bonner et al. 2008). DSBs induced by irradiation primarily recruit ATM, while replication-induced DSBs activate primarily ATR via replication protein A (RPA). DNA-PK, which was shown to be responsible for H2AX phosphorylation during apoptosis, is also responsible for H2AX modification along with both ATM and ATR but less frequently.

1.3.1 g -H2AX and the Recruitment of DNA Repair and Chromatin Remodeling Factors In the case of irradiation-induced DSBs, the ends are recognized by the Mre11Rad50-NBS1 (MRN) complex, leading to ATM dimer recruitment. Autophosphory lation and activation results in ATM monomerization (Durocher and Jackson 2001; Shiloh 2001; Yang et al. 2003). Upon activation, ATM phosphorylates hundreds of H2AX molecules that are necessary for the accumulation and the retention at the DNA break sites of numerous proteins involved in DNA repair and signaling (Shiloh 2003). The g-H2AX formation is the starting point for a hierarchical and timely regulated recruitment cascade leading to the accumulation of mediator proteins, including BRCA1 and 53BP1, at the DSB sites (Fig. 1.2c). BRCA1 and 53BP1 play key roles in genome integrity by modulating checkpoints and DNA repair (Fernandez-Capetillo et al. 2002; FitzGerald et al. 2009; Huen et al. 2010). Though H2AX plays an important role in BRCA1 mobilization in response to DNA damage, the H2AX null mutant exhibits less severe phenotype than the BRCA1 mutant (Celeste et al. 2002). This can be explained by the fact that a portion of BRCA1 is also recruited to the DSB sites by another mechanism, independent of H2AX phosphorylation, through its interaction with MRN (Greenberg 2008). The fraction of BRCA1 accumulating at the DSB site in a g-H2AX dependent manner is found in the BRCA1-A complex (BRCA1/BARD1/ABRA1/BRCC36/RAP80 and recently discovered MERIT40/BRE/NBA1) (Wang et al. 2009; Solyom et al. 2010). Its recruitment at the DNA break happens in a succession of protein modification and protein–protein recognition steps described below.

1.3.1.1 BRCA1-A Complex The early response to DSB formation consists of the binding of MDC1, another mediator, to g-H2AX through its BRCT domains (Stucki et al. 2005). MDC1 contributes to the propagation of the DNA damage repair signal by recruiting more MRN


17

complex (via NBS1) and ATM (Fig. 1.2c panel 1). The MDC1-MRN complex is dependent on a serine phosphorylated in a cluster of conserved repeat motifs by casein kinase 2 (Lukas et al. 2004; Chapman and Jackson 2008). Enhanced recruitment of ATM leads in turn to escalating H2AX phosphorylation and MDC1 recruitment to several megabases around the DSB site. This signal amplification loop also promotes MDC1 phosphorylation by ATM. Phosphorylated MDC1 then interacts with RNF8 (Huen et al. 2007), which, along with the E2 enzyme UBC13, catalyzes the addition of Lys63-linked polyubiquitin chains to both H2A and H2AX (Mailand et al. 2007). Histone ubiquitylation by RNF8/UBC13 triggers the recruitment of the RNF168/UBC13 complex via the C-terminal MIU domain of RNF168 (Doil et al. 2009; Stewart et al. 2009). Mutations in the E3 ubiquitin ligase RNF168 have been recently linked to the DNA repair deficiency disorder RIDDLE syndrome (Stewart 2009). RIDDLE (Radiosensitivity, Immunodeficiency, Dysmorphic features, and learning difficulties) syndrome is a human immunodeficiency disorder associated with defective DSB repair (Stewart et al. 2007). The RNF168/UBC13 complex is then involved in the ubiquitylation amplification of substrates such as H2A and H2AX (Fig. 1.2c panel 2). Ubiquitinated H2A and H2AX provide docking sites for the UIM domains of ubiquitin ligase RAP80, which in turn, recruits ABRA1, a mediator protein that also binds simultaneously BRCA1/BARD1 and BRCC36 (Huen et al. 2007; Kolas et al. 2007; Mailand et al. 2007).

1.3.1.2 53BP1 53BP1 accumulation at DSB sites is also dependent on H2AX phosphorylation. The role of 53BP1 in the cell-cycle checkpoint may be explained by its ability to interact with p53 via its BRCT motifs, a phenomenon that would allow its phosphorylation/ activation by ATM at the DSB site (FitzGerald et al. 2009). 53BP1 also contains two tandem Tudor domains that bind methylated histones H3 and H4 with high affinity for H4K20Me20 and that are necessary for its accumulation at DSB sites (FitzGerald et al. 2009). Like BRCA1, 53BP1 accumulation at DSB sites is dependent on RNF8 (Mailand et al. 2007). However, in contrast to BRCA1, the relationship between RNF8 and 53BP1 is not well understood. It is thought that in response to DSB formation, the accumulation of DNA damage response factors and the subsequent histone modifications result in chromatin and nucleosome remodeling, which expose H4 N-terminal tails, revealing H4K20me2 and allowing 53BP1 accumulation (Mailand et al. 2007) (Fig. 1.2c panel 3).

1.3.1.3 Chromatin Remodeling Complexes H2AX phosphorylation is also implicated in stimulating the binding to chromatin or enhancing the activity of chromatin remodeling complexes such as the human histone acetyltransferase, TIP60 complex, and its yeast counterpart the NuA4 complex

18

C.E. Redon et al.

(Thiriet and Hayes 2005; van Attikum and Gasser 2005). These complexes are involved in the DNA damage response by allowing both histone modifications and chromatin relaxation, two mechanisms necessary for further DNA repair factor recruitment and/or histone exchange/removal (Thiriet and Hayes 2005). NuA4 accumulates in the vicinity of the DSB, and genetic evidences indicate that both of INO80-C and SWR-C complexes are involved in DNA damage repair (Thiriet and Hayes 2005). Chromatin remodeling complexes also play a role after DNA is repaired by restoring the chromatin environment to its predamaged condition (i.e., removal of g-H2AX from chromatin after repair, for example) (Thiriet and Hayes 2005). Studies in yeast demonstrated that Arp4, a protein found in the NuA4, INO80-C and SWR-C (Thiriet and Hayes 2005; van Attikum and Gasser 2005) chromatin remodeling complexes, binds to the yeast g-H2AX homolog in vitro (Downs et al. 2004). Morrison et al. (2004) showed that INO80-C can also be recruited to DSB sites by interaction between another one of its components, Nhp10, and the g-H2AX yeast homolog (Morrison et al. 2004). The Arp4 homolog in mammals, Baf53, is found in several chromatin remodeling complexes such as Tip60 and SWI-SNF (Lee et al. 2007). In humans, the SWI/SNF chromatin remodeling complex is targeted to DNA lesions by its interaction with BRIT1/MCPH1 (Peng and Lin 2009a, b; Peng et al. 2009), a protein linked to a disease gene called microcephalin and identified in numerous cancers (Bartek 2006). In addition to its role in chromatin condensation and tumor suppression, BRIT1/MCPH1 participates in the accumulation of numerous proteins at the DSB sites and has been shown to control the G2–M checkpoint, through its ability to regulate Chk1 and BRCA1 transcriptional levels (Lin et al. 2005; Wood et al. 2007). As it was suggested that the C-terminal BRCT domains of BRIT1/MCPH1 interact with g-H2AX (Wood et al. 2007), the recruitment of SWI/SNF at the DSB sites could be dependent on H2AX phosphorylation.

1.3.1.4 Cohesins In mammalian cells, the formation of g-H2AX is important for homologous recombination between sister chromatids (Xie et al. 2004). Decreased intersister chromatid distances after DSB induction suggest that cohesin loading at DSBs elevates sister-chromatid cohesion (Dodson and Morrison 2009) and could facilitate accurate homologous recombination. Cohesins are responsible for chromatid cohesion, holding sister chromatids together from the time they are replicated to the metaphase–anaphase transition (Unal et al. 2004). Following DNA damage formation, cohesins are loaded to the DSB sites in a g-H2AX-dependent manner (Unal et al. 2004).

1.3.2 g -H2AX and the Cell-Cycle Checkpoint Cell-cycle checkpoints halt progression of cells through the replication cycle if damage is present. The G2/M checkpoint prevents cells with DNA damage from passing through mitosis, since chromatid fragments that lack a centromere may not attach to


19

the spindle and not be pulled toward a pole (Fig. 1.1f). If DNA damage cannot be fixed, cells undergo programmed cell death in which H2AX also plays a role (Lu et al. 2006; Mukherjee et al. 2006). In mammalian cells, H2AX was shown to be essential for proper functioning of the G2/M checkpoint after low amounts of DNA damage. This observation can be related to the finding that g-H2AX is involved in BRCA1 and 53BP1 recruitment to the break site (as discussed in the previous paragraph). For example, Fernandez-Capetillo et al. (2002) showed that 1 h after cells are exposed to various amounts of ionizing radiation and mitotic cells are isolated, MEFs from control mice exhibit greater than 75% inhibition in passing through mitosis at doses greater than 0.5 Gy. However, H2AX-null MEFs exhibit little if any inhibition at this dose. Both ATM, which phosphorylates H2AX, and 53BP1, whose recruitment is dependent on g-H2AX, are also essential for proper checkpoints after 0.5 Gy. Studies with yeast show that the g-H2AX homolog is important for prolonged G2/M checkpoint arrest in response to radiation (Nakamura et al. 2004). The role of the yeast g-H2AX homolog in the cell-cycle checkpoint is linked to the fact that g-H2AX is necessary for both Crb2 (the yeast 53BP1 homolog) phosphorylation and foci formation.

1.3.3 Complexity of Repair Foci Containing g -H2AX and Other Factors Following DSB and subsequent g-H2AX formation, more than 50 DDR proteins are recruited and/or modified at the break sites. It is unknown if more DDR proteins located in the g-H2AX foci remain to be discovered, and the spatial organization of these proteins in the repair foci is still largely unknown. However, a few studies in both yeast and mammalian cells have examined the structure of g-H2AX foci, and their observations confirm a time-dependent and hierarchical distribution of DDR proteins (Nakamura et al. 2010). Both g-H2AX and MDC1 were found to span large domains flanking DSBs sites (with MDC1 localizing as doublets in foci), while 53BP1 and the components of the MRN complexes were restricted to single regions, shown to correspond to DSB sites in yeast (Shroff et al. 2004). In yeast, deposition of g-H2AX and cohesins following DSB formation was similarly found to span large domains flanking DSB breaks (Unal et al. 2004). DDR protein distribution in g-H2AX foci seems to be related to their function in DNA repair. Thus, MDC1 and perhaps cohesins, which play primary roles in homologous recombination and/or sister chromatid recombination, are distributed over several areas, while 53BP1 and the MRN complex, which have a more implicit role in nonhomologous end joining, are distributed over smaller areas perhaps more central to the break. In addition to their spatial complexity, g-H2AX foci also show a temporal complexity with DDR proteins recruitment exhibiting both dependent and independent behavior during the cell cycle. Thus, g-H2AX foci change their composition through the cell cycle, with 53BP1, NBS1, and MRE11 dissociating from the foci at G2/mitosis to return at early G1, while MDC1 remains at g-H2AX foci through the entire cell cycle (Nelson et al. 2009; Nakamura et al. 2010).

20

C.E. Redon et al.

1.4 g -H2AX as a Marker g-H2AX focal formation in response to DSB induction is a highly amplified process and, as discussed earlier, is capable of detecting individual DSBs. This sensitivity enables researchers to study DNA DSBs at physiological levels of damage. As such, g-H2AX focal formation has the potential to become a useful in vivo biodosimeter for processes that result in DSB formation. This includes not only processes that result in frank DSBs but also those that may indirectly result in DSBs. Some of these are discussed below. In addition, there are several areas in which clinical applications of g-H2AX focal formation are being developed.

1.4.1 Biological Processes 1.4.1.1 Cancer Elevated levels of g-H2AX are present in different human cancer model systems, including cervical cancer cells, melanoma cells, colon carcinomas, fibrosarcoma, osteosarcoma, glioma, and neuroblastoma cells, as well as in tumor biopsies compared to normal primary cells or normal tissues adjacent to tumors (Banath et al. 2004; Warters et al. 2005; Martin and Bonner 2006; Yu et al. 2006). These results suggest that an increased level of DNA damage is a general characteristic of cancer development (Bartkova et al. 2005; Gorgoulis et al. 2005). Moreover, an increase in g-H2AX content has been found in colonocytes from patients with ulcerative colitis that predisposes them to colorectal cancer (Risques et al. 2008). Damaged telomeres make up the majority of endogenous DNA double-strand damage in tumor cells, and the numbers of telomere-associated g-H2AX foci correlate inversely with the cells’ telomerase activity (Nakamura et al. 2009). The origin of nontelomeric g-H2AX foci in these cells remains an open question. These foci may be caused by specific attributes of cancer cells such as excess reactive oxygen production and accumulation of DNA repair and cell-cycle checkpoint defects (Lavin 2008; Hamilton 2009; Martin et al. 2010).

1.4.1.2 Senescence Animals, including humans, age and primary somatic cell lines derived from them undergo only a limited number of divisions. This exhaustion of proliferative potential, called cellular senescence, may protect organisms from cell immortalization and tumorigenesis, thus providing a tumor suppressor function (Prieur and Peeper 2008). Factors thought to play a role in aging at the cellular level include telomere shortening, oxidative stress, accumulating DNA damage, compromised DNA repair


21

machinery, and inappropriate gene amplification (Ju et al. 2006; Ozgenc and Loeb 2006; Martin et al. 2008; Zhao et al. 2009; Reddel 2010). It has been shown that g-H2AX foci increase during cellular senescence in culture, whether from proliferation, DNA-damaging agents, or oxidative stress (Nakamura et al. 2008). Likewise, the incidence of foci increases as humans and mice age (Martin et al. 2004, 2008; Hesse et al. 2009). 1.4.1.3 Radiation-Induced Bystander Effect The radiation-induced bystander effect (RIBE) was first described in the early 1990s by Nagasawa and Little when it was noted that when 1% of the cells in a dish were traversed with alpha particles, 30% of the cells displayed increased sister chromatid exchanges (Nagasawa and Little 1992). These results indicated for the first time that cellular damage may be experienced by naïve “bystander” cells when they are in close contact with damaged cells. Since that time, the RIBE has been studied in a number of cellular models as well as in vivo (Mothersill and Seymour 2004; Hei et al. 2008; Prise and O’Sullivan 2009). The RIBE generally manifests itself in the form of DNA damage that leads to various downstream effects. The RIBE has been shown to increase the incidence of apoptosis, point mutations, sister-chromatid exchanges, global genome demethylation, and senescence (Mothersill and Seymour 2004; Hei et al. 2008; Prise and O’Sullivan 2009). As DNA damage, and specifically DNA DSBs, seems to be a precipitating event in many of these manifestations of the RIBE, g-H2AX has been often used as a marker of bystander effect induction and propagation (Zhou et al. 2002; Sokolov et al. 2005, 2007; Martin et al. 2007). Based on these studies, it was found that the DNA damage induced as a result of bystander signaling is distinct in several respects from the DNA damage that might be induced as a result of direct cellular injury (Sokolov et al. 2005; Martin et al. 2007). For instance, bystander-effectinduced DNA damage arises more slowly than that from direct cellular irradiation. When cells are irradiated directly, using either X-rays or alpha particles, DNA DSBs are formed almost immediately. The g-H2AX foci are visible less than 1 min postirradiation and the focal size and incidence peaks by 30 min postirradiation (Bonner et al. 2008). In contrast, it takes several hours for g-H2AX foci to form in bystander cell populations, and focal incidence may peak as late as 8 h postirradiation (Kashino et al. 2004; Sokolov et al. 2005; Martin et al. 2007) (Fig. 1.3a). These kinetic characteristics indicate that a slowly diffusing substance may be contributing to bystander DNA DSB production and that breaks may actually be made indirectly (Kashino et al. 2004; Hei et al. 2008). Likewise, while the g-H2AX foci in directly irradiated cell populations begin to disappear shortly after irradiation as the cellular machinery repairs the breaks, bystander DNA damage can persist for several days (Morgan 2003; Kashino et al. 2004; Sokolov et al. 2005; Martin et al. 2007) (Fig. 1.3a). In some cellular systems, increased DNA damage can be seen in bystander cell populations for as long as

22

C.E. Redon et al.

Fig. 1.3 (a) Upper diagram: Brief description of the bystander effect. When cells are irradiated directly, DSBs are formed immediately and g-H2AX foci appear within minutes. Irradiated cells release factors and ROS that diffuse to unirradiated cells and initiate a delayed response leading to g-H2AX foci formation. g-H2AX foci are represented by green dots. Lower graph: Typical g-H2AX foci kinetics for direct (open circle) and bystander (closed circle) responses to irradiation are shown. Both intensity and timing of the bystander response differ from the direct response. (b) The use of g-H2AX for radiation biodosimetry. Left panel: g-H2AX formation in lymphocytes 30 min after exposure to 0, 0.1, 0.2, 0.6, 1, 1.5 and 3 Gy g-irradiation. Right graph: Linear relationship between the number of g-H2AX foci per cell and the irradiation dose. Red: DNA; Green: g-H2AX

7 days (Martin et al. 2007). These findings indicate either that the DNA damage created by bystander signaling is more difficult to repair than damage induced directly by irradiation or that a feedback mechanism is continually creating new DNA damage long after the original precipitating event is finished. Some of the signaling molecules that have been proposed to function as primary bystander effect mediators, including TGF-b and nitric oxide (NO), induce the transcription of stress response elements that then themselves create increased DNA DSBs (Choy et al. 2007; Shao et al. 2008; Dickey et al. 2009b). For instance, TGF-b has been shown to induce the transcription of nitric oxide synthase (NOS), which leads


23

to increased production of NO, and increased DNA DSBs through lipid peroxidation as well as direct DNA damage (Hussain et al. 2003; Zablocka and Janusz 2008). This feedback loop then serves to create DNA DSBs over a longer time than what would be caused by directly damaging the genome with radiation. Other observations show that the bystander effect is a global response that is not restricted to stress-induced irradiation. Ultraviolet exposure, photolytic DNA damage, and exposure to tumorigenic and senescent cells can also cause bystander responses including DNA damage in neighboring cells without the need for any outside damaging stimulus (Dickey et al. 2009b; Pawelec et al. 2009). Identifying the cell populations that are vulnerable to bystander effect signaling, either from radiation or from other forms of stress, remains an active area of investigation. Some research has indicated that actively replicating cells, particularly cells in S-phase, are the most susceptible cells to form g-H2AX foci (BurdakRothkamm et al. 2007; Burdak-Rothkamm et al. 2008; Dickey et al. 2009a, b).

1.4.2 Clinical Applications 1.4.2.1 Biodosimetry and Individual Radiosensitivity Assays predictive of individual radiosensitivity have been the “Holy Grail” of experimental radiation oncologists for decades, but prospects have been limited by the insensitivity of assays, and also the need to use activated and/or transformed lymphocytes. Other issues impeding progress are the cost and labor intensity of clonogenic survival assays, and the delay (8–10 days) in obtaining results to subsequently plan individualized therapy. The literature is replete with failures of predictive assays for both tumors and normal tissues. For example, no correlation has been observed between the radiosensitivity of transformed lymphocytes and primary fibroblasts in radiotherapy patients (Geara et al. 1992). The g-H2AX assay presents a novel approach. Considered to be the most sensitive modern assay for DSB detection (Bonner et al. 2008), it can detect both a single DSB in cell nuclei and the cell responses to radiation doses as low as 1 mGy (Rothkamm and Lobrich 2003). Because of this sensitivity, g-H2AX has been recently identified as a useful biomarker with clinical implications (see below). It is possible to detect irradiation-induced DNA damage in situ in nonactivated human lymphocytes, as well as in tissue biopsies (Qvarnstrom et al. 2004; Lobrich et al. 2005; Redon et al. 2009), hair follicles (unpublished results), or oral cells (Yoon et al. 2009), although in these cells the sensitivity of the assay is generally lower, possibly due to high endogenous g-H2AX level and large number of dying cells. Additionally, radiation dose-dependent responses and persistence of foci make g-H2AX detection a good potential biodosimeter in the event of a major radiation emergency as well as clinical examination or radiotherapy (Olive and Banath 2004; Lobrich et al. 2005; Rothkamm et al. 2007; Geisel et al. 2008; Kuefner et al. 2009; Redon et al. 2009; Bhogal et al. 2010).

24

C.E. Redon et al.

The g-H2AX assay in lymphocytes irradiated in vivo (for CT scan examinations, angiography, and radiotherapy) has been used, showing a good correlation between the g-H2AX focal numbers and the radiation doses received by patients (Qvarnstrom et al. 2004; Lobrich et al. 2005; Sak et al. 2007; Geisel et al. 2008; Kuefner et al. 2009) (Fig. 1.3b). Importantly, the g-H2AX kinetics is similar in human lymphocytes and fibroblasts after ex vivo and in vivo irradiation (Lobrich et al. 2005). The rapid response (maximal number of foci at 30 min postexposure and repair within several hours) gives the base for 1-day assessment of individual sensitivity in a minimally invasive test of primary lymphocytes exposed to ex vivo irradiation, identifying individuals with defective DNA repair. A severely radiosensitive patient with a DSB repair deficiency was identified in a tomography examination study by the g-H2AX focus formation assay and PFGE. That patient had previously shown exceptionally severe side effects after radiotherapy (Lobrich et al. 2005). Thus, feedback on a patient radiosensitivity would allow clinicians to better adjust treatments to individuals, avoiding serious radiotherapy accidents (Abbaszadeh et al. 2009). For this purpose, g-H2AX has been used to identify individuals with ataxia telangiectasia (A-T), a disease predisposing patients to cancer, as well as to identify ataxia telangiectasia heterozygotes, which represent 0.36–1% of the general population (Porcedda et al. 2006, 2008, 2009). The ataxia telangiectasia gene codes for ATM, the kinase responsible for phosphorylation of a broad range of DNA repair factor including H2AX (Shiloh 2003). Preclinical mouse models of radiosensitivity testing allowed researchers to verify different DSB repair deficiencies after whole-body irradiation (DNA-PK, SCID, AT). Even slight impairments caused changes in DSB repair that were detected in both lymphocytes and tissues using the g-H2AX focus formation assay. Thus, future examinations of patient blood ex vivo could give precious information about individual repair capacity (Rube et al. 2008). Finally, the use of g-H2AX for radiation biodosimetry was also valuable to measure daily space radiation exposure in the International Space Station. The study showed that measured doses were similar between biological (g-H2AX) and physical (track and luminescent detectors) dosimetries (0.7 mSv/day vs. 0.5 mSv, respectively) (Ohnishi et al. 2009). 1.4.2.2 Chemotherapy In addition to its growing role in drug development in research laboratories, g-H2AX is increasingly used in clinical trials. Most cancer therapeutic treatments cause DSBs directly or indirectly (Bonner et al. 2008). Since g-H2AX is a robust DSB biomarker, measuring g-H2AX in cells may provide clinicians valuable information about individual patient responses. Numerous clinical trials have measured g-H2AX signals in patients (Bonner et al. 2008). Because collecting blood is relatively noninvasive, this tissue has been the most often used for g-H2AX detection (Lobrich et al. 2005; Rothkamm et al. 2007). However, in contrast to radiotherapy where DSBs are induced regardless of cell


25

cycle, analysis of g-H2AX signals in the nondividing lymphocytes during chemotherapy could be challenging as most of the cancer drugs target dividing cells. The use of skin biopsies and plucked hairs, both containing dividing cells, can be used in addition to blood (Qvarnstrom et al. 2004; Fong et al. 2009). However, because of their genetic makeup and/or differences in microenvironments such as vascularization, normal tissues may respond more differently than the targeted tumors. Therefore, g-H2AX detection in tumor biopsies will still be required. In recent years, phase 1 clinical studies have started utilizing the g-H2AX assay. A study using clofarabine (a ribonucleotide reductase and DNA polymerases inhibitor) and cyclophosphamide (an alkylating agent effectively used for treatment of leukemias) for adults with refractory cancers showed increased g-H2AX levels in circulating leukemia cells of 12 of the 13 treated patients (Karp et al. 2007). Another trial confirmed the inhibition of the poly(adenosine diphosphate [ADP]-ribose) polymerase (PARP) by the drug olaparib by analyzing g-H2AX in plucked-eyebrow hairs (Fong et al. 2009). Still another study demonstrated the genotoxic activity of the minor groove binding agent SJG-136 by increased g-H2AX levels in both patient lymphocytes and tumor biopsies (Hochhauser et al. 2009). Finally, a trial testing the combination of tipifamid (a farnesyltransferase inhibitor) and etoposide (a topoisomerase II inhibitor) as orally bioavailable regimen for elderly adults diagnosed with myelogenous leukemia used a g-H2AX assay in AML marrow blasts (Karp et al. 2009). This study showed good patient responses to the drug combination demonstrated by H2AX phosphorylation correlating with increased subdiploid DNA content. Quantifying H2AX phosphorylation has also been used for clinical diagnostics. Increased levels of g-H2AX and shorter telomeres were found in colonocytes of patients with ulcerative colitis, a chronic inflammation disease linked to colorectal cancer (Risques et al. 2008). Potentially, g-H2AX assays could be used to prescreen individuals, determining the ones at risk for cancer that would result in preemptive treatments (Martin and Bonner 2006). Other diagnostic tools using g-H2AX revealed its usefulness in the diagnosis of metastatic renal cell carcinoma (Wasco and Pu 2008) and predicting the recurrence of bladder urothelial carcinoma (Cheung et al. 2009). 1.4.2.3 Environmental Toxins The formation of DSBs can be used as a sensor to determine the genotoxicity of compounds/chemicals, occupational activities, and other human behaviors. For example, g-H2AX has been used to determine the DNA-damaging effects of cigarette smoking (Albino et al. 2004; Toyooka and Ibuki 2009), nonthermal microwaves of cell phones (Markova et al. 2005; Belyaev et al. 2009), electromagnetic fields (Luo et al. 2006), dental resin composites (Urcan et al. 2009), crude oil (Ibuki et al. 2007), chromium VI (Ha et al. 2004), actinomycin D (Mischo et al. 2005), norethindrone (Gallmeier et al. 2005), NO-aspirin (Tanaka et al. 2006), carbon

26

C.E. Redon et al.

nanotubes (Zhu et al. 2007), isocyanates (Mishra et al. 2008), and the herbicide paraquat (Kim et al. 2004). Thus, the future use of g-H2AX in routine screens could be valuable to examine the genotoxicity of doubtful agents and/or human activities that, if detected in time, would help reduce preventable human diseases.

1.5 Conclusions H2AX is phosphorylated in response to DNA DSB formation, irrespective of whether those breaks are induced directly by environmental agents or indirectly as a result of programmed and nonprogrammed biological processes. In addition, the amplification of the process makes it applicable for use at physiological levels of DSBs, leading to possible clinical assays. While the phosphorylation of H2AX is understood at the level of the individual protein, how it contributes to foci formation is still unclear. Also, evidence is accumulating that foci have considerable complexity and undergo changes with cell cycle and possibly other states. Understanding the dynamics of foci formation as opposed to repair protein recruitment will help further elucidate how cells deal with DSB damage. Acknowledgements This work was supported by the Intramural Research Program of the National Cancer Institute, NIH.

References Abbaszadeh F, Clingen PH, Arlett CF et al (2009) A novel splice variant of the DNA-PKcs gene is associated with clinical and cellular radiosensitivity in a xeroderma pigmentosum patient. J Med Genet doi: 10.1136/jmg.2009.068866 Ajmani AK, Satoh M, Reap E et al (1995) Absence of autoantigen Ku in mature human neutrophils and human promyelocytic leukemia line (HL-60) cells and lymphocytes undergoing apoptosis. J Exp Med 181:2049–2058 Albino AP, Huang X, Jorgensen E et al (2004) Induction of H2AX phosphorylation in pulmonary cells by tobacco smoke: a new assay for carcinogens. Cell Cycle 3:1062–1068 Banath JP, Klokov D, Macphail SH et al (2010) Residual gammaH2AX foci as an indication of lethal DNA lesions. BMC Cancer 10:4 doi: 10.1186/1471-2407-10-4 Banath JP, Macphail SH, Olive PL (2004) Radiation sensitivity, H2AX phosphorylation, and kinetics of repair of DNA strand breaks in irradiated cervical cancer cell lines. Cancer Res 64:7144–7149 Banath JP, Olive PL (2003) Expression of phosphorylated histone H2AX as a surrogate of cell killing by drugs that create DNA double-strand breaks. Cancer Res 63:4347–4350 Bartek J (2006) Microcephalin guards against small brains, genetic instability, and cancer. Cancer Cell 10:91–93 Bartkova J, Horejsi Z, Koed K et al (2005) DNA damage response as a candidate anti-cancer barrier in early human tumorigenesis. Nature 434:864–870 Bassing CH, Alt FW (2004) H2AX may function as an anchor to hold broken chromosomal DNA ends in close proximity. Cell Cycle 3:149–153 Bellani MA, Romanienko PJ, Cairatti DA et al (2005) SPO11 is required for sex-body formation, and Spo11 heterozygosity rescues the prophase arrest of Atm−/− spermatocytes. J Cell Sci 118:3233–3245


27

Belyaev IY, Markova E, Hillert L et al (2009) Microwaves from UMTS/GSM mobile phones induce long-lasting inhibition of 53BP1/gamma-H2AX DNA repair foci in human lymphocytes. Bioelectromagnetics 30:129–141 Bhogal N, Kaspler P, Jalali F et al (2010) Late residual gamma-H2AX foci in murine skin are dose responsive and predict radiosensitivity in vivo. Radiat Res 173:1–9 Bonner WM, Redon CE, Dickey JS et al (2008) GammaH2AX and cancer. Nat Rev Cancer 8:957–967 Burdak-Rothkamm S, Rothkamm K, Prise KM (2008) ATM acts downstream of ATR in the DNA damage response signaling of bystander cells. Cancer Res 68:7059–7065 Burdak-Rothkamm S, Short SC, Folkard M et al (2007) ATR-dependent radiation-induced gamma H2AX foci in bystander primary human astrocytes and glioma cells. Oncogene 26:993–1002 Campos EI, Reinberg D (2009) Histones: annotating chromatin. Annu Rev Genet 43: 559–599 Cedervall B, Wong R, Albright N et al (1995) Methods for the quantification of DNA doublestrand breaks determined from the distribution of DNA fragment sizes measured by pulsedfield gel electrophoresis. Radiat Res 143:8–16 Celeste A, Petersen S, Romanienko PJ et al (2002) Genomic instability in mice lacking histone H2AX. Science 296:922–927 Chapman JR, Jackson SP (2008) Phospho-dependent interactions between NBS1 and MDC1 mediate chromatin retention of the MRN complex at sites of DNA damage. EMBO Rep 9:795–801 Cheung WL, Albadine R, Chan T et al (2009) Phosphorylated H2AX in noninvasive low grade urothelial carcinoma of the bladder: correlation with tumor recurrence. J Urol 181:1387–1392 Chew YC, Camporeale G, Kothapalli N et al (2006) Lysine residues in N-terminal and C-terminal regions of human histone H2A are targets for biotinylation by biotinidase. J Nutr Biochem 17:225–233 Chowdhury D, Keogh MC, Ishii H et al (2005) Gamma-H2AX dephosphorylation by protein phosphatase 2A facilitates DNA double-strand break repair. Mol Cell 20:801–809 Chowdhury D, Xu X, Zhong X et al (2008) A PP4-phosphatase complex dephosphorylates gamma-H2AX generated during DNA replication. Mol Cell 31:33–46 Choy JC, Wang Y, Tellides G et al (2007) Induction of inducible NO synthase in bystander human T cells increases allogeneic responses in the vasculature. Proc Natl Acad Sci USA 104:1313–1318 Cook PJ, Ju BG, Telese F et al (2009) Tyrosine dephosphorylation of H2AX modulates apoptosis and survival decisions. Nature 458:591–596 Cowell IG, Sunter NJ, Singh PB et al (2007) gammaH2AX foci form preferentially in euchromatin after ionising-radiation. PLoS One 2:e1057 Cucinotta FA, Nikjoo H, Goodhead DT (2000) Model for radial dependence of frequency distributions for energy imparted in nanometer volumes from HZE particles. Radiat Res 153:459–468 Cucinotta FA, Pluth JM, Anderson JA et al (2008) Biochemical kinetics model of DSB repair and induction of gamma-H2AX foci by non-homologous end joining. Radiat Res 169:214–222 Cucinotta FA, Wu H, Shavers MR et al (2003) Radiation dosimetry and biophysical models of space radiation effects. Gravit Space Biol Bull 16:11–18 Daniel R, Ramcharan J, Rogakou E et al (2004) Histone H2AX is phosphorylated at sites of retroviral DNA integration but is dispensable for postintegration repair. J Biol Chem 279:45810–45814 de Lange T (2004) T-loops and the origin of telomeres. Nat Rev Mol Cell Biol 5:323–329 Desai N, Durante M, Lin ZW et al (2005) High LET-induced H2AX phosphorylation around the Bragg curve. Adv Space Res 35:236–242 Dickey JS, Redon CE, Nakamura AJ et al (2009a) H2AX: functional roles and potential applications. Chromosoma 118:683–692 Dickey JS, Baird BJ, Redon CE et al (2009b) Intercellular communication of cellular stress monitored by gamma-H2AX induction. Carcinogenesis 30:1686–1695 Dodson H, Morrison CG (2009) Increased sister chromatid cohesion and DNA damage response factor localization at an enzyme-induced DNA double-strand break in vertebrate cells. Nucleic Acids Res 37:6054–6063 Doil C, Mailand N, Bekker-Jensen S et al (2009) RNF168 binds and amplifies ubiquitin conjugates on damaged chromosomes to allow accumulation of repair proteins. Cell 136:435–446 Downs JA, Allard S, Jobin-Robitaille O et al (2004) Binding of chromatin-modifying activities to phosphorylated histone H2A at DNA damage sites. Mol Cell 16:979–990

28

C.E. Redon et al.

Duensing S, Munger K (2002) The human papillomavirus type 16 E6 and E7 oncoproteins independently induce numerical and structural chromosome instability. Cancer Res 62:7075–7082 Durocher D, Jackson SP (2001) DNA-PK, ATM and ATR as sensors of DNA damage: variations on a theme? Curr Opin Cell Biol 13:225–231 Fernandez-Capetillo O, Chen HT, Celeste A et al (2002) DNA damage-induced G2-M checkpoint activation by histone H2AX and 53BP1. Nat Cell Biol 4:993–997 FitzGerald JE, Grenon M, Lowndes NF (2009) 53BP1: function and mechanisms of focal recruitment. Biochem Soc Trans 37:897–904 Fong PC, Boss DS, Yap TA et al (2009) Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers. N Engl J Med 361:123–134 Gallmeier E, Winter JM, Cunningham SC et al (2005) Novel genotoxicity assays identify norethindrone to activate p53 and phosphorylate H2AX. Carcinogenesis 26:1811–1820 Geara FB, Peters LJ, Ang KK et al (1992) Intrinsic radiosensitivity of normal human fibroblasts and lymphocytes after high- and low-dose-rate irradiation. Cancer Res 52:6348–6352 Geisel D, Heverhagen JT, Kalinowski M et al (2008) DNA double-strand breaks after percutaneous transluminal angioplasty. Radiology 248:852–859 Goodarzi AA, Noon AT, Deckbar D et al (2008) ATM signaling facilitates repair of DNA doublestrand breaks associated with heterochromatin. Mol Cell 31:167–177 Goodarzi AA, Noon AT, Jeggo PA (2009) The impact of heterochromatin on DSB repair. Biochem Soc Trans 37:569–576 Goodhead DT (1994) Initial events in the cellular effects of ionizing radiations: clustered damage in DNA. Int J Radiat Biol 65:7–17 Goodhead DT, Nikjoo H (1989) Track structure analysis of ultrasoft X-rays compared to high- and low-LET radiations. Int J Radiat Biol 55:513–529 Gorgoulis VG, Vassiliou LV, Karakaidos P et al (2005) Activation of the DNA damage checkpoint and genomic instability in human precancerous lesions. Nature 434:907–913 Greenberg RA (2008) Recognition of DNA double strand breaks by the BRCA1 tumor suppressor network. Chromosoma 117:305–317 Ha L, Ceryak S, Patierno SR (2004) Generation of S phase-dependent DNA double-strand breaks by Cr(VI) exposure: involvement of ATM in Cr(VI) induction of gamma-H2AX. Carcinogenesis 25:2265–2274 Hamilton R (2009) Genetics: breast cancer as an exemplar. Nurs Clin North Am 44: 327–338 Hei TK, Zhou H, Ivanov VN et al (2008) Mechanism of radiation-induced bystander effects: a unifying model. J Pharm Pharmacol 60:943–950 Hesse JE, Faulkner MF, Durdik JM (2009) Increase in double-stranded DNA break-related foci in early-stage thymocytes of aged mice. Exp Gerontol 44:676–684 Hochhauser D, Meyer T, Spanswick VJ et al (2009) Phase I study of sequence-selective minor groove DNA binding agent SJG-136 in patients with advanced solid tumors. Clin Cancer Res 15:2140–2147 Huen MS, Grant R, Manke I et al (2007) RNF8 transduces the DNA-damage signal via histone ubiquitylation and checkpoint protein assembly. Cell 131:901–914 Huen MS, Sy SM, Chen J (2010) BRCA1 and its toolbox for the maintenance of genome integrity. Nat Rev Mol Cell Biol 11:138–148 Hussain SP, Hofseth LJ, Harris CC (2003) Radical causes of cancer. Nat Rev Cancer 3:276–285 Ibuki Y, Toyooka T, Shirahata J et al (2007) Water soluble fraction of solar-simulated light-exposed crude oil generates phosphorylation of histone H2AX in human skin cells under UVA exposure. Environ Mol Mutagen 48:430–439 Ikura T, Tashiro S, Kakino A et al (2007) DNA damage-dependent acetylation and ubiquitination of H2AX enhances chromatin dynamics. Mol Cell Biol 27:7028–7040 Jenner TJ, deLara CM, O’Neill P et al (1993) Induction and rejoining of DNA double-strand breaks in V79-4 mammalian cells following gamma- and alpha-irradiation. Int J Radiat Biol 64:265–273


29

Ju YJ, Lee KH, Park JE et al (2006) Decreased expression of DNA repair proteins Ku70 and Mre11 is associated with aging and may contribute to the cellular senescence. Exp Mol Med 38:686–693 Karp JE, Flatten K, Feldman EJ et al (2009) Active oral regimen for elderly adults with newly diagnosed acute myelogenous leukemia: a preclinical and phase 1 trial of the farnesyltransferase inhibitor tipifarnib (R115777, Zarnestra) combined with etoposide. Blood 113:4841–4852 Karp JE, Ricklis RM, Balakrishnan K et al (2007) A phase 1 clinical-laboratory study of clofarabine followed by cyclophosphamide for adults with refractory acute leukemias. Blood 110:1762–1769 Kashino G, Prise KM, Schettino G et al (2004) Evidence for induction of DNA double strand breaks in the bystander response to targeted soft X-rays in CHO cells. Mutat Res 556:209–215 Kato TA, Nagasawa H, Weil MM et al (2006) Levels of gamma-H2AX Foci after low-dose-rate irradiation reveal a DNA DSB rejoining defect in cells from human ATM heterozygotes in two at families and in another apparently normal individual. Radiat Res 166:443–453 Kato TA, Okayasu R, Bedford JS (2008) Comparison of the induction and disappearance of DNA double strand breaks and gamma-H2AX foci after irradiation of chromosomes in G1-phase or in condensed metaphase cells. Mutat Res 639:108–112 Keogh MC, Kim JA, Downey M et al (2006) A phosphatase complex that dephosphorylates gammaH2AX regulates DNA damage checkpoint recovery. Nature 439:497–501 Kim JA, Kruhlak M, Dotiwala F et al (2007) Heterochromatin is refractory to gamma-H2AX modification in yeast and mammals. J Cell Biol 178:209–218 Kim SJ, Kim JE, Moon IS (2004) Paraquat induces apoptosis of cultured rat cortical cells. Mol Cells 17:102–107 Kolas NK, Chapman JR, Nakada S et al (2007) Orchestration of the DNA-damage response by the RNF8 ubiquitin ligase. Science 318:1637–1640 Krishnan N, Jeong DG, Jung SK et al (2009) Dephosphorylation of the C-terminal tyrosyl residue of the DNA damage-related histone H2A.X is mediated by the protein phosphatase eyes absent. J Biol Chem 284:16066–16070 Kuefner MA, Grudzenski S, Schwab SA et al (2009) DNA double-strand breaks and their repair in blood lymphocytes of patients undergoing angiographic procedures. Invest Radiol 44:440–446 Kurosawa A, Shinohara K, Watanabe F et al (2003) Human neutrophils isolated from peripheral blood contain Ku protein but not DNA-dependent protein kinase. Int J Biochem Cell Biol 35:86–94 Kusch T, Florens L, Macdonald WH et al (2004) Acetylation by Tip60 is required for selective histone variant exchange at DNA lesions. Science 306:2084–2087 Lavin MF (2008) Ataxia-telangiectasia: from a rare disorder to a paradigm for cell signalling and cancer. Nat Rev Mol Cell Biol 9:759–769 Leatherbarrow EL, Harper JV, Cucinotta FA et al (2006) Induction and quantification of gammaH2AX foci following low and high LET-irradiation. Int J Radiat Biol 82:111–118 Lee K, Kang MJ, Kwon SJ et al (2007) Expansion of chromosome territories with chromatin decompaction in BAF53-depleted interphase cells. Mol Biol Cell 18:4013–4023 Li W, Ma H (2006) Double-stranded DNA breaks and gene functions in recombination and meiosis. Cell Res 16:402–412 Lin SY, Rai R, Li K et al (2005) BRIT1/MCPH1 is a DNA damage responsive protein that regulates the Brca1-Chk1 pathway, implicating checkpoint dysfunction in microcephaly. Proc Natl Acad Sci USA 102:15105–15109 Lobrich M, Ikpeme S, Kiefer J (1994) Measurement of DNA double-strand breaks in mammalian cells by pulsed-field gel electrophoresis: a new approach using rarely cutting restriction enzymes. Radiat Res 138:186–192 Lobrich M, Rief N, Kuhne M et al (2005) In vivo formation and repair of DNA double-strand breaks after computed tomography examinations. Proc Natl Acad Sci USA 102:8984–8989 Lobrich M, Rydberg B, Cooper PK (1995) Repair of x-ray-induced DNA double-strand breaks in specific Not I restriction fragments in human fibroblasts: joining of correct and incorrect ends. Proc Natl Acad Sci USA 92:12050–12054

30

C.E. Redon et al.

Lu C, Zhu F, Cho YY et al (2006) Cell apoptosis: requirement of H2AX in DNA ladder formation, but not for the activation of caspase-3. Mol Cell 23:121–132 Lukas C, Melander F, Stucki M et al (2004) Mdc1 couples DNA double-strand break recognition by Nbs1 with its H2AX-dependent chromatin retention. EMBO J 23:2674–2683 Luo Q, Yang J, Zeng QL et al (2006) 50-Hertz electromagnetic fields induce gammaH2AX foci formation in mouse preimplantation embryos in vitro. Biol Reprod 75:673–680 Macurek L, Lindqvist A, Voets O et al (2010) Wip1 phosphatase is associated with chromatin and dephosphorylates gammaH2AX to promote checkpoint inhibition. Oncogene doi: 10.1038/ onc.2009.501 Mailand N, Bekker-Jensen S, Faustrup H et al (2007) RNF8 ubiquitylates histones at DNA doublestrand breaks and promotes assembly of repair proteins. Cell 131:887–900 Markova E, Hillert L, Malmgren L et al (2005) Microwaves from GSM mobile telephones affect 53BP1 and gamma-H2AX foci in human lymphocytes from hypersensitive and healthy persons. Environ Health Perspect 113:1172–1177 Martin OA, Bonner WM (2006) GammaH2AX in cancer cells: a potential biomarker for cancer diagnostics, prediction and recurrence. Cell Cycle 5:2909–2913 Martin OA, Horikawa I, Redon C et al (2008) Delayed kinetics of DNA double-strand break processing in normal and pathological aging. Aging Cell 7:89–100 Martin OA, Horikawa I, Zimonjic DB et al (2004) Senescing human cells and ageing mice accumulate DNA lesions with unrepairable double-strand breaks. Nat Cell Biol 6:168–170 Martin OA, Nakamura A, Kovalchuk O et al (2007) DNA double-strand breaks form in bystander cells after microbeam irradiation of three-dimensional human tissue models. Cancer Res 67:4295–4302 Martin OA, Pilch DR, Redon C et al (2003) Histone H2AX in DNA damage and repair. Cancer Biol Ther 2:233–235 Martin OA, Redon CE, Dickey JS et al (2010) Role of oxidatively induced DNA lesions in human pathogenesis. Mutat Res doi:10.1016/j.mrrev.2009.12.005 Martin OA, Rogakou EP, Panyutin IG et al (2002) Quantitative detection of (125) IdU-induced DNA double-strand breaks with gamma-H2AX antibody. Radiat Res 158:486–492 McKinnon PJ, Caldecott KW (2007) DNA strand break repair and human genetic disease. Annu Rev Genomics Hum Genet 8:37–55 Mischo HE, Hemmerich P, Grosse F et al (2005) Actinomycin D induces histone gamma-H2AX foci and complex formation of gamma-H2AX with Ku70 and nuclear DNA helicase II. J Biol Chem 280:9586–9594 Mishra PK, Panwar H, Bhargava A et al (2008) Isocyanates induces DNA damage, apoptosis, oxidative stress, and inflammation in cultured human lymphocytes. J Biochem Mol Toxicol 22 429–440 Morgan WF (2003) Is there a common mechanism underlying genomic instability, bystander effects and other nontargeted effects of exposure to ionizing radiation? Oncogene 22:7094–7099 Morrison AJ, Highland J, Krogan NJ et al (2004) INO80 and gamma-H2AX interaction links ATP-dependent chromatin remodeling to DNA damage repair. Cell 119:767–775 Mothersill C, Seymour CB (2004) Radiation-induced bystander effects – implications for cancer. Nat Rev Cancer 4:158–164 Mukherjee B, Kessinger C, Kobayashi J et al (2006) DNA-PK phosphorylates histone H2AX during apoptotic DNA fragmentation in mammalian cells. DNA Repair (Amst) 5:575–590 Nagasawa H, Little JB (1992) Induction of sister chromatid exchanges by extremely low doses of alpha-particles. Cancer Res 52:6394–6396 Nakamura AJ, Chiang YJ, Hathcock KS et al (2008) Both telomeric and non-telomeric DNA damage are determinants of mammalian cellular senescence. Epigenetics Chromatin 1:6 doi: 10.1186/1756-8935-1-6 Nakamura AJ, Redon CE, Martin OA (2009) Where did they come from? The origin of the endogenous gamma-H2AX foci in tumor cells. Cell Cycle 8:2324 Nakamura AJ, Rao VA, Pommier Y et al (2010) The complexity of phosphorylated H2AX foci formation and DNA repair assembly at DNA double-strand breaks. Cell Cycle 9:389–397 Nakamura TM, Du LL, Redon C et al (2004) Histone H2A phosphorylation controls Crb2 recruitment at DNA breaks, maintains checkpoint arrest, and influences DNA repair in fission yeast. Mol Cell Biol 24:6215–6230


31

Nelson G, Buhmann M, von Zglinicki T (2009) DNA damage foci in mitosis are devoid of 53BP1. Cell Cycle 8:3379–3383 Nikjoo H, O’Neill P, Wilson WE et al (2001) Computational approach for determining the spectrum of DNA damage induced by ionizing radiation. Radiat Res 156:577–583 Ohnishi T, Takahashi A, Nagamatsu A et al (2009) Detection of space radiation-induced double strand breaks as a track in cell nucleus. Biochem Biophys Res Commun 390:485–488 Olive PL, Banath JP (2004) Phosphorylation of histone H2AX as a measure of radiosensitivity. Int J Rad Onc Biol Phys 58:331–335 Ozgenc A, Loeb LA (2006) Werner Syndrome, aging and cancer. Genome Dyn 1:206–217 Pantazis P, Bonner WM (1981) Quantitative determination of histone modification. H2A acetylation and phosphorylation. J Biol Chem 256:4669–4675 Paull TT, Rogakou EP, Yamazaki V et al (2000) A critical role for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage. Curr Biol 10:886–895 Pawelec G, Bohr V, Campisi J (2009) Special issue on cancer and ageing. Mech Ageing Dev 130:1–2 Peng G, Lin SY (2009a) BRIT1/MCPH1 is a multifunctional DNA damage responsive protein mediating DNA repair-associated chromatin remodeling. Cell Cycle 8:3071–3072 Peng G, Lin SY (2009b) The linkage of chromatin remodeling to genome maintenance: contribution from a human disease gene BRIT1/MCPH1. Epigenetics 4:457–461 Peng G, Yim EK, Dai H et al (2009c) BRIT1/MCPH1 links chromatin remodelling to DNA damage response. Nat Cell Biol 11:865–872 Pinto DM, Flaus A (2010) Structure and Function of Histone H2AX. Subcell Biochem 50: 55–78 Pommier Y, Barcelo JM, Rao VA et al (2006) Repair of topoisomerase I-mediated DNA damage. Prog Nucleic Acid Res Mol Biol 81:179–229 Porcedda P, Turinetto V, Brusco A et al (2008) A rapid flow cytometry test based on histone H2AX phosphorylation for the sensitive and specific diagnosis of ataxia telangiectasia. Cytometry A 73:508–516 Porcedda P, Turinetto V, Lantelme E et al (2006) Impaired elimination of DNA double-strand break-containing lymphocytes in ataxia telangiectasia and Nijmegen breakage syndrome. DNA Repair (Amst) 5:904–913 Porcedda P, Turinetto V, Orlando L et al (2009) Two-tier analysis of histone H2AX phosphorylation allows the identification of Ataxia Telangiectasia heterozygotes. Radiother Oncol 92:133–137 Prieur A, Peeper DS (2008) Cellular senescence in vivo: a barrier to tumorigenesis. Curr Opin Cell Biol 20:150–155 Prise KM, O’Sullivan JM (2009) Radiation-induced bystander signalling in cancer therapy. Nat Rev Cancer 9:351–360 Qvarnstrom OF, Simonsson M, Johansson KA et al (2004) DNA double strand break quantification in skin biopsies. Radiother Oncol 72:311–317 Reddel RR (2010) Senescence: an antiviral defense that is tumor suppressive? Carcinogenesis 31:19–26 Redon C, Pilch D, Rogakou E et al (2002) Histone H2A variants H2AX and H2AZ. Curr Opin Genet Dev 12:162–169 Redon C, Pilch DR, Rogakou EP et al (2003) Yeast histone 2A serine 129 is essential for the efficient repair of checkpoint-blind DNA damage. EMBO Rep 4:678–684 Redon CE, Dickey JS, Bonner WM et al (2009) gamma-H2AX as a biomarker of DNA damage induced by ionizing radiation in human peripheral blood lymphocytes and artificial skin. Adv Space Res 43:1171–1178 Risques RA, Lai LA, Brentnall TA et al (2008) Ulcerative colitis is a disease of accelerated colon aging: evidence from telomere attrition and DNA damage. Gastroenterology 135:410–8 Rogakou EP, Boon C, Redon C et al (1999) Megabase chromatin domains involved in DNA double-strand breaks in vivo. J Cell Biol 146:905–916 Rogakou EP, Nieves-Neira W, Boon C et al (2000) Initiation of DNA fragmentation during apoptosis induces phosphorylation of H2AX histone at serine 139. J Biol Chem 275:9390–9395

32

C.E. Redon et al.

Rogakou EP, Pilch DR, Orr AH et al (1998) DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. J Biol Chem 273:5858–5868 Rothkamm K, Balroop S, Shekhdar J et al (2007) Leukocyte DNA damage after multi-detector row CT: a quantitative biomarker of low-level radiation exposure. Radiology 242:244–251 Rothkamm K, Lobrich M (2003) Evidence for a lack of DNA double-strand break repair in human cells exposed to very low X-ray doses. Proc Natl Acad Sci USA 100:5057–5062 Rube CE, Grudzenski S, Kuhne M et al (2008) DNA double-strand break repair of blood lymphocytes and normal tissues analysed in a preclinical mouse model: implications for radiosensitivity testing. Clin Cancer Res 14:6546–6555 Sak A, Grehl S, Erichsen P et al (2007) gamma-H2AX foci formation in peripheral blood lymphocytes of tumor patients after local radiotherapy to different sites of the body: dependence on the dose-distribution, irradiated site and time from start of treatment. Int J Radiat Biol 83:639–652 Shao C, Folkard M, Prise KM (2008) Role of TGF-beta1 and nitric oxide in the bystander response of irradiated glioma cells. Oncogene 27:434–440 Shiloh Y (2001) ATM and ATR: networking cellular responses to DNA damage. Curr Opin Genet Dev 11:71–77 Shiloh Y (2003) ATM and related protein kinases: safeguarding genome integrity. Nat Rev Cancer 3:155–168 Shroff R, Arbel-Eden A, Pilch D et al (2004) Distribution and dynamics of chromatin modification induced by a defined DNA double-strand break. Curr Biol 14:1703–1711 Sluss HK, Davis RJ (2006) H2AX is a target of the JNK signaling pathway that is required for apoptotic DNA fragmentation. Mol Cell 23:152–153 Smilenov LB, Lieberman HB, Mitchell SA et al (2005) Combined haploinsufficiency for ATM and RAD9 as a factor in cell transformation, apoptosis, and DNA lesion repair dynamics. Cancer Res 65:933–938 Sokolov MV, Dickey JS, Bonner WM et al (2007) gamma-H2AX in bystander cells: not just a radiation-triggered event, a cellular response to stress mediated by intercellular communication. Cell Cycle 6:2210–2212 Sokolov MV, Smilenov LB, Hall EJ et al (2005) Ionizing radiation induces DNA double-strand breaks in bystander primary human fibroblasts. Oncogene 24:7257–7265 Solier S, Pommier Y (2009) The apoptotic ring: a novel entity with phosphorylated histones H2AX and H2B and activated DNA damage response kinases. Cell Cycle 8:1853–1859 Solyom S, Patterson-Fortin J, Pylkas K et al (2010) Mutation screening of the MERIT40 gene encoding a novel BRCA1 and RAP80 interacting protein in breast cancer families. Breast Cancer Res Treat 120:165–168 Stewart GS (2009) Solving the RIDDLE of 53BP1 recruitment to sites of damage. Cell Cycle 8:1532–1538 Stewart GS, Panier S, Townsend K et al (2009) The RIDDLE syndrome protein mediates a ubiquitin-dependent signaling cascade at sites of DNA damage. Cell 136:420–434 Stewart GS, Stankovic T, Byrd PJ et al (2007) RIDDLE immunodeficiency syndrome is linked to defects in 53BP1-mediated DNA damage signaling. Proc Natl Acad Sci USA 104:16910–16915 Stucki M (2009) Histone H2A.X Tyr142 phosphorylation: a novel sWItCH for apoptosis? DNA Repair (Amst) 8:873–876 Stucki M, Clapperton JA, Mohammad D et al (2005) MDC1 directly binds phosphorylated histone H2AX to regulate cellular responses to DNA double-strand breaks. Cell 123:1213–1226 Takai H, Smogorzewska A, de Lange T (2003) DNA damage foci at dysfunctional telomeres. Curr Biol 13:1549–1556 Tanaka T, Kurose A, Halicka HD et al (2006) Nitrogen oxide-releasing aspirin induces histone H2AX phosphorylation, ATM activation and apoptosis preferentially in S-phase cells: involvement of reactive oxygen species. Cell Cycle 5:1669–1674 Tarakanova VL, Leung-Pineda V, Hwang S et al (2007) Gamma-herpesvirus kinase actively initiates a DNA damage response by inducing phosphorylation of H2AX to foster viral replication. Cell Host Microbe 1:275–286


33

Terai C, Wasson DB, Carrera CJ et al (1991) Dependence of cell survival on DNA repair in human mononuclear phagocytes. J Immunol 147:4302–4306 Thiriet C, Hayes JJ (2005) Chromatin in need of a fix: phosphorylation of H2AX connects chromatin to DNA repair. Mol Cell 18:617–622. Toyooka T, Ibuki Y (2009) Cigarette sidestream smoke induces phosphorylated histone H2AX. Mutat Res 676:34–40 Unal E, Arbel-Eden A, Sattler U et al (2004) DNA damage response pathway uses histone modification to assemble a double-strand break-specific cohesin domain. Mol Cell 16:991–1002 Urcan E, Scherthan H, Styllou M et al (2009) Induction of DNA double-strand breaks in primary gingival fibroblasts by exposure to dental resin composites. Biomaterials 31:2010–4 van Attikum H, Gasser SM (2005) ATP-dependent chromatin remodeling and DNA double-strand break repair. Cell Cycle 4:1011–1014. Wang B, Hurov K, Hofmann K et al (2009) NBA1, a new player in the Brca1 A complex, is required for DNA damage resistance and checkpoint control. Genes Dev 23:729–739 Warters RL, Adamson PJ, Pond CD et al (2005) Melanoma cells express elevated levels of phosphorylated histone H2AX foci. J Invest Dermatol 124:807–817 Wasco MJ, Pu RT (2008) Utility of antiphosphorylated H2AX antibody (gamma-H2AX) in diagnosing metastatic renal cell carcinoma. Appl Immunohistochem Mol Morphol 16:349–356 Whitaker SJ, Ung YC, McMillan TJ (1995) DNA double-strand break induction and rejoining as determinants of human tumour cell radiosensitivity. A pulsed-field gel electrophoresis study. Int J Radiat Biol 67:7–18 Wood JL, Singh N, Mer G et al (2007) MCPH1 functions in an H2AX-dependent but MDC1independent pathway in response to DNA damage. J Biol Chem 282:35416–35423 Wu RS, Tsai S, Bonner WM (1982) Patterns of histone variant synthesis can distinguish G0 from G1 cells. Cell 31:367–374 Wu RS, Tsai S, Bonner WM (1983) Changes in histone H3 composition and synthesis pattern during lymphocyte activation. Biochemistry 22:3868–3873 Xie A, Puget N, Shim I et al (2004) Control of sister chromatid recombination by histone H2AX. Mol Cell 16:1017–1025 Xie A, Scully R (2007) Hijacking the DNA damage response to enhance viral replication: gammaherpesvirus 68 orf36 phosphorylates histone H2AX. Mol Cell 27:178–179 Yang J, Yu Y, Hamrick HE et al (2003) ATM, ATR and DNA-PK: initiators of the cellular genotoxic stress responses. Carcinogenesis 24:1571–1580 Yasui LS (2004) GammaH2AX foci induced by gamma rays and 125idU decay. Int J Radiat Biol 80:895–903 Yoon AJ, Shen J, Wu HC et al (2009) Expression of activated checkpoint kinase 2 and histone 2AX in exfoliative oral cells after exposure to ionizing radiation. Radiat Res 171:771–775 Yu T, MacPhail SH, Banath JP et al (2006) Endogenous expression of phosphorylated histone H2AX in tumors in relation to DNA double-strand breaks and genomic instability. DNA Repair (Amst) 5:935–946 Zablocka A, Janusz M (2008) The two faces of reactive oxygen species. Postepy Hig Med Dosw (Online) 62:118–124 Zhao B, Benson EK, Qiao R et al (2009) Cellular senescence and organismal ageing in the absence of p21(CIP1/WAF1) in ku80(−/−) mice. EMBO Rep 10:71–78 Zhou H, Randers-Pehrson G, Suzuki M et al (2002) Genotoxic damage in non-irradiated cells: contribution from the bystander effect. Radiat Prot Dosimetry 99:227–232 Zhu L, Chang DW, Dai L et al (2007) DNA damage induced by multiwalled carbon nanotubes in mouse embryonic stem cells. Nano Lett 7:3592–3597

Chapter 2

DNA Damage Signaling Downstream of ATM Fred Bunz

Abstract ATM is the apical signaling molecule that triggers diverse cellular responses to double-strand DNA breaks. Directly and indirectly, ATM initiates a two-tiered cascade of protein kinase activation, composed of upstream phosphatidylinositol 3-kinase-like kinases, mediator proteins, and checkpoint kinases. Together, these proteins signal a broad network of downstream effectors that modulate virtually every aspect of cell growth and death. This review will focus on the signaling molecules required for the diverse ATM-dependent responses to DNA damage, with an emphasis on the extensively characterized pathways that suppress proliferation and promote DNA repair. Keywords DNA damage • ATM • ATR • Checkpoints • Signaling network

2.1 Introduction Damage to genomic DNA stimulates a profound and functionally diverse cellular response that affects fundamental cellular processes. As described in previous chapters, ATM plays a central role in initiating the collective responses to a particularly lethal form of DNA damage, the double-strand DNA break (DSB). Elegant biochemical experiments have demonstrated the fundamental mechanisms by which ATM kinase activity is activated at DSB sites (Bakkenist and Kastan 2004). The transduction of ATM signals arising from focal sites on damaged chromosomes to the many cellular compartments that mount responses to DSBs will be the focus of this review. F. Bunz (*) Associate Professor, Department of Radiation Oncology and Molecular Radiation Sciences, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, David H. Koch Cancer Research Building (CRB2), Room 453, 1550 Orleans Street, CRB II, Room 462, Baltimore, MD 21231, USA e-mail: [email protected] T.L. DeWeese and M. Laiho (eds.), Molecular Determinants of Radiation Response, Current Cancer Research, DOI 10.1007/978-1-4419-8044-1_2, © Springer Science+Business Media, LLC 2011

35

36

F. Bunz Budding yeast (S. Cerevisiae)

Fission yeast (S. Pombe)

DNA Damage

Replication Stress

DNA Damage

Tel1

Mec1 / Ddc2

Tel1

Replication Stress

Rad3 / Rad26

Mammalian cells

DNA Damage

Replication Stress

ATR / ATRIP

ATM

PIKK Rad9

BRCA1 53BP1 MDC1/NFBD1 MCPH1/BRIT1

Crb2

Mediator

Checkpoint Kinase

Effector

Rda53

Cdc25

Chk1

Cds1

Chk1

Cdc25

Chk2

Chk1

Cdc25A, B, C p53

Fig. 2.1 Conserved DNA damage signaling pathways. The general organization of the two-tiered DNA damage signaling cascade was defined by genetic studies in budding and fission yeasts. Double-strand DNA breaks activate orthologs of ATM, while DNA replication intermediates trigger the activation of orthologs of ATR and its binding partner ATRIP. Mediator proteins facilitate the activation of the checkpoint kinases by the upstream PIKKs. In mammalian cells, a family of BRCT domain-containing proteins appears to provide the functions of the single mediator proteins in the yeasts. Downstream signals converge on cell cycle regulatory proteins including Cdc25 protein phosphatases. Mammalian cells contain additional regulators of cell cycle checkpoints, including p53. Both the PIKKs and the checkpoint kinases have many additional downstream substrates. See text for details

Studies of the DNA damage pathways in evolutionarily divergent yeasts have revealed a two-tiered kinase cascade that is conserved in human cells (Fig. 2.1). DNA lesions directly activate apical serine/threonine protein kinases that are structurally related to the phosphatidylinositol 3-kinase and are accordingly known as the (PI3-kinase-like kinase) PIKK family. Identified as the gene mutated in ataxiatelangiectasia patients in 1995, ATM is one of six human PIKKs (Lavin et al. 2005). A second PIKK known to be functionally conserved in human cells is the ataxia-telangiectasia and Rad3-related (ATR) kinase. In both humans and yeasts, upstream PIKK proteins activated by DNA damage phosphorylate downstream serine/threonine protein kinases known as checkpoint kinases (Bartek and Lukas 2003; McGowan and Russell 2004; Reinhardt and Yaffe 2009). Chk1 and Chk2, the human orthologs of the yeast checkpoint kinases, transduce DNA damage signals from lesions and stalled replication forks to spatially and functionally distinct compartments of the cell (Stracker et al. 2009). An emerging class of phosphoprotein-interacting proteins cumulatively play the role of yeast Rad9, the first identified checkpoint protein, and mediate the activation of the checkpoint kinases by the PIKKs (Mohammad and Yaffe 2009). By illuminating the relationship between PIKKs, mediator proteins and the checkpoint kinases, genetic analyses of yeast have provided the theoretical framework for understanding the functional organization of the DNA damage response signaling pathways in human cells.

2 DNA Damage Signaling Downstream of ATM

37

2.2 Ataxia-Telangiectasia and Rad3-Related ATR is a large serine/threonine protein kinase with significant homology to ATM and the other members of the PIKK family (Abraham 2001; Cimprich and Cortez 2008). Both ATM and ATR preferentially phosphorylate serine residues that are followed by glutamine (SQ sites), and therefore target overlapping sets of substrates. However, ATM and ATR respond to distinct stimuli and thus have nonredundant functions. ATM is primarily activated by DSBs. ATR responds to DSBs but is additionally activated by a wide range of DNA lesions and DNA structures caused by environmental and therapeutic agents that inhibit or impede DNA replication, including ultraviolet (UV) radiation, DNA cross-linking agents and antimetabolites that interfere with nucleotide metabolism (Abraham 2001; Osborn et al. 2002). From a genetic perspective, the ATM and ATR signaling pathways exhibit important differences that suggest distinct roles in the cell. Both ATM and Chk2 are encoded by tumor suppressor genes that confer cancer predisposition – most prominently an increased risk of breast cancer – when inactivated by germline mutations. The complete loss of ATM signaling in patients with ataxia-telangiectasia or in ATM-knockout mice largely eliminates the response to DSBs and leads to extreme sensitivity to ionizing radiation (Shiloh 2006) but does not obviously impair unperturbed cell growth. In contrast, both ATR and Chk1 are essential genes that are required for cell viability and proliferation (Liu et al. 2000; Brown and Baltimore 2003). Only rare hypomorphic ATR alleles are tolerated in the germline; it is unclear if these predispose carriers to cancer (O’Driscoll et al. 2004). While ATR and CHK1 mutations have been found in small number of mismatch repair-deficient cancers, these genes do not have the requisite characteristics of established tumor suppressors. Inhibition of ATR activity in the presence of low levels of DNA replication stress causes DNA breaks at defined loci known as fragile sites (Casper et al. 2002). Detected at the cytogenetic scale, fragile sites are thought to represent structurally distinct regions in the genome where DNA replication forks tend to stall. The requirement of ATR for normal cell growth, its activation by DNA replication inhibitors and its role as a suppressor of fragile site expression all support a critical role for ATR in the stabilization of DNA replication forks (Fig. 2.2). Recent studies have determined the fundamental mechanism by which ATR is activated at stalled replication forks (Zou and Elledge 2003; Kumagai and Dunphy 2006; Cimprich and Cortez 2008; Mordes et al. 2008). The molecular machines that replicate genomic DNA include multiprotein DNA polymerase complexes that function to synthesize nascent strands and DNA helicase complexes that unwind the DNA template ahead of the advancing fork. Critical to the activation of ATR is the decoupling of polymerase and helicase activities. When a polymerase complex runs out of nucleotide substrate or if it encounters a DNA lesion, such as a strand break or an adduct, DNA synthesis can stall (Zegerman and Diffley 2009). In this case, the helicase complex can continue unwinding DNA, exposing a stretch of single-stranded DNA that would otherwise be rapidly replicated. The single-stranded DNA at the lagging strand of the stalled fork

38

F. Bunz

a

Pol

Helicase

b Stall

Helicase

Pol

RPA

c ATR ATRIP

TopBP1 TopBP1

9-1-1

Helicase

d ATR P

ATRIP

P TopBP1 P

Stable P Helicase

Fig. 2.2 Activation of ATR. (a) Advancing DNA replication forks are driven by the coordinated activities of multiprotein DNA polymerase (Pol) and DNA helicase complexes. (b) DNA polymerases that synthesize the leading and lagging strands may run short of nucleotides, or encounter a DNA lesion that impairs DNA synthesis. In such cases, the polymerase complexes will pause and become uncoupled from the helicase complexes, which continue unwinding DNA ahead of the stalled fork. Regions of single-stranded DNA are rapidly coated with the trimeric replication protein A complex (RPA). (c) ATR is recruited to the RPA-coated single-stranded DNA via association with ATRIP. The Rad9–Rad1–Hus1 (9-1-1) complex is loaded at the single strand–double strand DNA junction by the ATP-dependent Rad17 clamp loader. The TopBP1 protein associates with both the 9-1-1 complex and the ATR–ATRIP complex, which are all required for ATR activation. (d) Activated ATR phosphorylates numerous proteins assembled at the stalled fork (phosphates shown in yellow), which apparently serves to increase fork stability and eventually facilitate resumption of DNA synthesis. Thus activated, ATR also phosphorylates Chk1 and many other downstream substrates


39

becomes rapidly coated with the heterotrimeric single-strand DNA-binding protein complex known as replication protein A (RPA) (Fanning et al. 2006). The junction between RPA-coated single-stranded DNA and adjacent double-stranded DNA creates a structure that is recognized by the protein complexes that function to restart stalled forks. The RPA-coated single-strand DNA attracts binding by the ATR-interacting protein (ATRIP), the binding partner of ATR (Cortez et al. 2001). A heterotrimeric ring-like protein complex composed of Rad9-Rad1-Hus, known as the 9-1-1 sliding clamp, is loaded onto the single-strand DNA/double-strand DNA junction by an ATP-dependent clamp loader derived from the replicative replication factor C (RFC) complex (Parrilla-Castellar et al. 2004). The 9-1-1 complex and ATR function together to recruit the topoisomerase-interacting protein TopBP1 (Lee et al. 2007). The binding of ATR–ATRIP to RPA-coated single-strand DNA and the recruitment of TopBP1 are all required for activation of ATR kinase activity at stalled, unwound replication forks (Kumagai and Dunphy 2006; Mordes et al. 2008). ATR is known to phosphorylate several proteins involved in DNA replication, including subunits of RFC and RPA, proteins in the minichromosome maintenance (MCM) complex required for replication initiation and fork progression, and several DNA polymerases (Matsuoka et al. 2007; Cimprich and Cortez 2008). The consequences of individual ATR-dependent phosphorylation events remain poorly understood, but the overall effect of ATR on the replication fork appears to be to stabilize the protein complexes of the replisome and thereby maintain fork integrity and velocity (Cimprich and Cortez 2008; Wilsker et al. 2008). The role of ATR– ATRIP as both a sensor of stalled replication forks and as a stabilizer of the multiprotein complexes at the fork would appear to represent, in essence, a feedback circuit that ensures the efficient replication of the genome.

2.3 The Checkpoint Kinases The human homologs of the yeast checkpoint kinases serve as transducers of the signals generated by upstream PIKKs (Bartek and Lukas 2003). PIKKs are large kinases that, once activated, appear to remain closely located with DNA breaks and stalled replication forks. The smaller, more mobile checkpoint kinases are the mechanism by which signals originating at chromatin are transmitted throughout the cell to spatially and functionally diverse effectors (Smits 2006; Stracker et al. 2009). The checkpoint kinases preferentially phosphorylate R-X-X-S/T sites flanked by hydrophobic residues, and like the PIKKs have overlapping sets of substrates. In addition to the well-described ATM–Chk2 and ATR–Chk1 pathways, an additional complex consisting of the stress-responsive kinases p38MAPK and MK2 have been found to share substrate preferences with Chk1 and Chk2, and to function in the activation of checkpoints (Reinhardt and Yaffe 2009). The crossspecificity between Chk1, Chk2, and MK2 substrates has made it difficult to evaluate the respective roles of these kinases in downstream pathways. Recent studies

40

F. Bunz

suggest that Chk1 plays a primary role in cell cycle arrest pathways (Jallepalli et al. 2003; Wilsker et al. 2008) and is the most relevant checkpoint kinase in terms of therapeutic responses (Xiao et al. 2006; Wilsker et al. 2008).

2.3.1 Chk1 The most extensively characterized of ATR’s many substrates is the checkpoint kinase Chk1. The phosphorylation and activation of Chk1 requires a binding partner called Claspin that mediates the ATR–Chk1 physical interaction (Kumagai and Dunphy 2000; Chini and Chen 2004; Liu et al. 2006). Associated with the clamp loading complex at active replicons, Claspin is phosphorylated when replication forks stall (Wang et al. 2006). The association of Claspin with Chk1 is mediated by repeated phosphopeptide motifs that are phosphorylated by Chk1 and perhaps by other kinases as well (Kumagai and Dunphy 2003; Chini and Chen 2006). Thus, the Chk1–Claspin interaction is promoted by stalled replication forks and by the basal activity of Chk1 itself. Brought into proximity of the Chk1–Claspin complex, ATR phosphorylates Chk1 at two serine residues (S317 and S345) in a regulatory domain located in the c-terminus (Zhao and Piwnica-Worms 2001). While the Chk1 regulatory domain is evolutionarily divergent, the serine residues required for human Chk1 activation are within short regions of homology that are conserved in yeast. Detailed analyses of the functions of the individual Chk1 phosphorylation sites have provided insights into the functions of Chk1 and the distinct roles played by Chk1 in unperturbed and damaged cells. Genetic model systems in mouse and human cells have demonstrated Chk1 to be essential for cellular viability, as is ATR (Liu et al. 2000). Phosphorylation of Chk1 on its S345 site appears to be intrinsic to the essential function of Chk1; CHK1 alleles with mutations (S to A) that disrupt the S345 site do not support cellular viability (Niida et al. 2007; Wilsker et al. 2008). In contrast, mutations of the S317 are tolerated. However, cells harboring S317 mutant Chk1 proteins lose the ability to mediate the G2/M checkpoint in response to DNA damage and exhibit defects in DNA replication, including decreased replication fork velocity and increased fork stalling in unperturbed cells (Wilsker et al. 2008). Studies of Chk1 mutants have therefore revealed that Chk1 plays an essential role during normal cell growth and also has a nonessential role in the DNA damage response. These distinct roles are genetically separable by mutation of a single ATR phosphorylation site at S317 (Niida et al. 2007; Wilsker et al. 2008). Phosphorylation of Chk1 plays a prominent role in its localization. In unstressed cells, a pool of Chk1 protein remains stably bound to chromatin. Phosphorylation of Chk1 on S317 and S345 by ATR results in the rapid dissociation of this bound fraction (Smits et al. 2006). Following DNA damage or replicative stress, the majority of Chk1 in the cell becomes rapidly phosphorylated, suggesting that both chromatin-bound and unbound fractions of Chk1 are ultimately modified by ATR. In the prevailing model, unphosphorylated Chk1 in effect circulates to sites of


41

activated ATR (Smits 2006). Once phosphorylated by ATR at regions of RPA-coated single-stranded DNA, Chk1 is released from chromatin to phosphorylate its downstream substrates. Chk1 localization is also controlled by cellular pathways generally unrelated to DNA damage signaling. The AKT kinase, which is negatively regulated by the PTEN tumor suppressor, phosphorylates Chk1 on S280 (Puc et al. 2005). The phosphorylation of Chk1 at this site promotes its monoubiqutination and promotes its sequestration in the cytoplasm. It is believed that the enhanced phosphorylation of Chk1 by AKT contributes to the checkpoint deficiencies observed in PTENdeficient cancers (Puc and Parsons 2005; Puc et al. 2005). Chk1 controls the intra-S and the G2/M checkpoints, DNA damage-responsive pathways by which damaged cells transiently halt DNA replication and are prevented from entering mitosis, respectively (Stracker et al. 2009). Key components of the checkpoint pathways are the Cdc25 family of phosphatases. The three Cdc25 proteins Cdc25A, Cdc25B, and Cdc25C remove inhibitory phosphate moieties from the cyclin-dependent kinases (CDKs), and thereby promote cell cycle transitions (Boutros et al. 2006; Karlsson-Rosenthal and Millar 2006). The effect of Chk1 on the Cdc25 proteins is to inhibit their activity by triggering their degradation (in the case of Cdc25A) or causing their sequestration in the cytoplasm (Cdc25B and Cdc25C). By inhibiting the Cdc25 phosphatases required for CDK activity, activated Chk1 is required for the checkpoint pathways that block cell cycle transitions. Other functions of Chk1 include the activation of DNA repair by the Fanconi Anemia pathway. Required for the repair of interstrand DNA cross-links, the Fanconi Anemia genes encode a multisubunit protein complex that is initially activated by ATR (Gurtan and D’Andrea 2006). Chk1 phosphorylates the FancE subunit on two residues (T346 and S374) that promote its degradation (Wang et al. 2007). The Chk1 sites on FancE are required for cell survival after treatment with DNA cross-linking agents.

2.3.2 Chk2 As Chk1 is directly phosphorylated by ATR in response to RPA-coated singlestranded DNA, so is Chk2 is phosphorylated by ATM in response to DSBs (Bartek et al. 2001; Ahn et al. 2004; Stracker et al. 2009). The kinase domains of Chk2 and Chk1 are highly related, and pharmacological inhibitors of these kinases typically exhibit cross-specificity. Beyond their kinase domains, Chk2 and Chk1 are structurally and functionally distinct (McGowan 2002). In striking contrast with the ATR–Chk1 pathway, the ATM–Chk2 pathway is nonessential. The Chk2 gene can be homozygously disrupted in human and mouse cells (Hirao et al. 2000; Jallepalli et al. 2003), and therefore does not play a critical role in cell proliferation. Chk2 mutations occur at a significant frequency, but the role of Chk2 in cancer has been controversial. The original identification of Chk2

42

F. Bunz

mutations in patients with a variant of Li Fraumeni syndrome known as Li Fraumeni-like syndrome suggested that Chk2 mutations might phenocopy highly penetrant p53 mutations (Bell et al. 1999). In accordance with this hypothesis, studies of Chk2 activity in vitro demonstrated that Chk2 can phosphorylate p53 on a site (S20) involved in its activation (Hirao et al. 2000; Shieh et al. 2000). However, the most prevalent Chk2 alteration is a truncating mutation (designated 1100delC) that is found in some populations at a frequency as high as 1% (Sodha et al. 2000). The discordance between the high frequency of the CHK21100delC allele and the low incidence of Li Fraumeni-like syndrome suggests that Chk2 mutations are not functionally equivalent to p53 mutations (Sodha et al. 2002). Studies in human cells demonstrate that p53 can be activated in the absence of Chk2 (Jallepalli et al. 2003), while analyses of knockout mice suggest that Chk2 is required for some p53dependent functions in some cells types (Jack et al. 2002). It thus would appear that Chk2 can promote p53 activation in some tissues, perhaps in response to tissuespecific stimuli. Large-scale population-based studies have demonstrated that carriers of the CHK21100delC allele have three- to fivefold increased risk of developing breast cancer (Meijers-Heijboer et al. 2002; Weischer et al. 2008). CHK2 is therefore a tumor suppressor gene with incomplete penetrance, similar to ATM. Chk2 has an N-terminal cluster of ATM/ATR recognition sites known as the SCD, a Forkhead-associated domain (FHA) that is involved in phosphorylationdependent protein–protein interactions and a kinase catalytic domain toward the c-terminus (Ahn et al. 2004; Stracker et al. 2009). More than 25 distinct Chk2 phosphorylation sites have been identified. Activated ATM phosphorylates Chk2 on many of the recognition sites in the c-terminal SCD, including residue T68. Once phosphorylated, the SCD becomes a docking site for a second Chk2 molecule. The multimerization of Chk2 brings the activation domains of the kinase loops into close proximity, thus promoting the autophosphorylation of multiple residues in the catalytic site. The autophosphorylation of the N-terminal sites increases enzymatic activity. This two-stage activation mechanism is common to the homologs of Chk2 in yeast (Oliver et al. 2007). The substrate specificity of Chk2 closely resembles that of Chk1 (Stracker et al. 2009). Accordingly, Chk2 has been shown to phosphorylate key cell cycle regulators such as p53 and the Cdc25 protein phosphatases. The phenotypes of Chk2 knockout human cells and mice are generally mild and do not phenocopy mutations in ATM. Chk2 knockout cells retain their ability to upregulate p53 and activate checkpoints following DNA damage, and apoptotic pathways appear to be defective in only a subset of tissues (Jack et al. 2002; Jallepalli et al. 2003).

2.3.3 p38MAPK/MK2 The mitogen-activated protein kinase-activated protein kinase-2 (MK2) is potently stimulated by the various activators of the p38 family, which response to diverse forms of cell stress (Roux and Blenis 2004). Loss of function of MK2 in knockout


43

mice leads most apparently to deficiencies in immune responses. A role for the p38MAPK/MK2 in checkpoint control was suggested by the finding that MK2 can function downstream of ATR and phosphorylate the same sites on Cdc25 proteins as Chk1 and Chk2, following exposure to UV radiation (Manke et al. 2005). Cdc25B and Cdc25C are sequestered in the cytoplasm following MK2-dependent phosphorylation.

2.4 Cooperation Between ATM and ATR ATM and ATR are both activated by DSBs (Abraham 2001; Osborn et al. 2002). The activation of ATM occurs within minutes of a DSB, in cells that are in any phase of the cell cycle. In contrast, activation of ATR by DSBs is delayed and restricted to cells that are in S and G2 (Jazayeri et al. 2006). Cells from ataxiatelangiectasia patients deficient in ATM function exhibit markedly impaired phosphorylation of downstream substrates and diminished DSB responses, despite the presence of wild-type ATR. Such observations underscore the primary importance of ATM in the DSB response. Recent studies have demonstrated that ATM is in fact required for activation of ATR by DSBs (Adams et al. 2006; Cuadrado et al. 2006; Jazayeri et al. 2006; Myers and Cortez 2006). Unlike other downstream components of the DSB response, ATR is not known to be an ATM substrate. Rather, ATR is activated indirectly as a result of ATM-initiated processing of DSBs that involves the components of the MRN complex. The DNA end processing by the MRN complex in effect causes DSBs that efficiently activate ATM to be converted to regions of RPA-coated single-stranded DNA that efficiently activate ATR. This DSB conversion is CDK-dependent, which causes ATR activation to be cell cycle phase-specific (Jazayeri et al. 2006). In the absence of ATM, the conversion of DSBs to RPA-coated single-stranded DNA is significantly slower and less robust, and the activation of ATR is similarly decreased. Following its indirect activation by ATM, ATR plays a critical role in the overall DSB response. Most notably, cells engineered to be completely or even partially deficient in ATR function exhibit clear defects in the G2/M checkpoint triggered by ionizing radiation (Hurley et al. 2007). Additionally, ATR-deficient cells exposed to ionizing radiation fail to enter S-phase and complete DNA replication, suggesting a failure to stabilize early replication complexes in the presence of DSBs (Hurley et al. 2007). ATM has been observed to be detectably activated by agents that primarily inhibit DNA replication (Dodson and Tibbetts 2006; Stiff et al. 2006); this crossactivation has been attributed to phosphorylation by ATR of the ATM S1981 autophosphorylation site (Stiff et al. 2006). The ATR-dependent phosphorylation of this site is independent of the MRN complex, suggesting that ATM activation by this pathway is mechanistically distinct from its autoactivation after ionizing radiation. Cumulatively, these data show that the cooperation between ATM and ATR appears

44

F. Bunz

to work in both directions. While ATM and ATR control parallel pathways, these apical sensor kinases clearly work together to respond to structurally diverse DNA lesions (Hurley and Bunz 2007). Is there crosstalk between the ATM and ATR pathways? The shared substrate specificity of ATM and ATR presents the possibility that downstream substrates could be activated by either kinase. Activation of the ATM substrate Chk2 has been shown to result from treatment with DNA replication stressors known to activate ATR (Stiff et al. 2006). Chk1, known to be strongly phosphorylated by ATR, is phosphorylated in an ATM-dependent manner after exposure to ionizing radiation (Gatei et al. 2003). Each of these observations could be interpreted as evidence of crosstalk. However, recent insights into the cooperation between ATM and ATR have cast these observations in a new light. It is important to consider that DNA lesions are not static structures, but rather are rapidly processed and metabolized (Jazayeri et al. 2006). DSBs can be converted to regions of single-stranded DNA by the processing of broken ends that is ATM-dependent and MRN-mediated. DSB can alternatively arise from stalled replication forks that expose single-stranded DNA to the effects of nucleases. It now appears that relationships that on the surface appear to be crosstalk between upstream kinases are in fact the result of interconversion between DSBs and RPA-coated single-strand DNA. It is now generally believed that the respective activation of Chk1 and Chk2 by ATR and ATM is highly specific.

2.5 Mediators of the DNA Damage Response: BRCT-Containing Proteins Downstream of ATM and ATR are a complex family of proteins that share a common, highly conserved motif first identified in the breast cancer susceptibility gene 1 (BRCA1). The BRCA1 c-terminal (BRCT) motif is a phosphoprotein-binding domain involved in protein–protein interactions and oligomerization (Yu et al. 2003). Proteins containing tandem BRCT domains at their C-termini function to facilitate the interaction between signaling (PIKK) and transduction (checkpoint kinase) molecules, and thus act as mediators of the signaling cascade activated by DNA damage or DNA replication stress (Mohammad and Yaffe 2009). Among the BRCT-containing proteins recently found to function in DNA damage signaling are BRCA1, 53BP1, MDC1/NFBD1, and MCPH1/BRIT1. Together, these proteins represent the functional homologs of the first checkpoint protein ever described, budding yeast Rad9 (which is unrelated to the human Rad9 component of the 9-1-1 complex). Importantly, while yeast Rad9 functions to unidirectionally mediate signals from the PIKKs to the checkpoint kinases, several of the human homologs appear to additionally affect the upstream activation of ATM (Mochan et al. 2003; Aglipay et al. 2006; Wilson and Stern 2008). Both the diversity of mediators and the bidirectionality of the signaling pathways suggest a higher order of complexity in mammalian cells as compared with yeasts.


45

2.5.1 BRCA1 Many of the genes involved in the DNA damage repair pathways function as tumor suppressors. Prominent among these is BRCA1, which mutated in about one-half of inherited breast and ovarian cancers. Many of the cancer predisposing mutations in BRCA1 affect the BRCT domain, suggesting that the role of BRCA1 in the DNA damage response is intrinsic to its role in tumor suppression. BRCA1 is phosphorylated after DNA damage on multiple sites by ATM (Cortez et al. 1999) and Chk2 (Zhang et al. 2004) and relocates rapidly to sites of damage and stalled DNA replication forks (Okada and Ouchi 2003; Ouchi 2006). Cells containing BRCA1 mutant proteins display defective intra-S and G2/M checkpoints and hypersensitivity to ionizing radiation (Deng 2006). The individual phosphorylation sites are differentially required for the G2/M and the intra-S checkpoints, demonstrating the independent regulation of these pathways and the overall complexity of the relationship between BRCA1 and checkpoints (Xu et al. 2001). The phosphorylation of BRCA1 by Chk2, also encoded by a tumor suppressor gene involved in breast cancer, is required for efficient repair of DSBs (Lee et al. 2000; Zhang et al. 2004).

2.5.2 53BP1 53BP1 is a component of the DNA damage response that was originally identified by virtue of its physical interaction with the tumor suppressor protein p53 (Iwabuchi et al. 1994). Genetic analyses have demonstrated the importance of 53BP1 for the stabilization of p53, the phosphorylation of Chk2 and the activation of intra-S and G2/M checkpoints after DNA damage (DiTullio et al. 2002; Wang et al. 2002; Ward et al. 2003). After exposure to ionizing radiation, 53BP1 relocalizes rapidly to nuclear foci that also contain ATM, which correspond to sites of DSBs. Cells deficient in ATM exhibit defects in the relocalization of 53BP1, indicating that 53BP1 functions in ATM-dependent pathways (DiTullio et al. 2002). 53BP1-knockout mice are cancer prone, radiosensitive, and develop similar cancers as ATM-knockout mice (Ward et al. 2004). Interestingly, loss of one or both copies of 53BP1 in p53-null mice significantly accelerated the rate of cancer development, demonstrating that 53BP1 and p53 proteins function together to suppress tumors (Ward et al. 2005). While a significant body of data suggests that 53BP1 mediates the phosphorylation of ATM substrates, the exact function of 53BP1 remains unclear. 53BP1 contains tandem BRCT motifs, but these are not required for its oligomerization or for downstream DNA repair functions (Ward et al. 2006). Localization of 53BP1 to damage sites has been shown to require a tandem Tudor domain, a methyl-proteinbinding motif required for the initial recruitment of 53BP1 to chromatin (Huyen et al. 2004). 53BP1 contains 15 identified ATM/ATR phosphorylation sites (Matsuoka et al. 2007). These sites are phosphorylated after ionizing radiation as well as UV radiation, suggesting that ATR as well as ATM can play an upstream role in its activation (Jowsey et al. 2007).

46

F. Bunz

2.5.3 MDC1/NFBD1 Mediator of DNA damage checkpoint protein 1 (MDC1/NFBD1) is a BRCT-domain-containing protein that is phosphorylated in response to DNA damage (Stucki and Jackson 2004). After its phosphorylation, MDC1/NFBD1 relocalizes to sites of DNA lesions and functions to facilitate the recruitment and accumulation of cell cycle checkpoint and DNA repair factors to sites of DNA damage (Lukas et al. 2004; Xu et al. 2008), a characteristic shared with the other BRCTcontaining proteins. Originally identified in a random screen of large cDNAs, MDC1/NFBD1 was predicted to function in the DNA damage response on the basis of its domain structure (Stucki and Jackson 2004). Depletion of MDC1/NFBD1 causes defects in checkpoint activation and apoptosis and sensitizes cells to DNAdamaging agents (Goldberg et al. 2003; Stewart et al. 2003).

2.5.4 MCPH1/BRIT1 MCPH1/BRIT1 was identified independently as a disease gene that encodes microcephalin and as a BRCT domain-containing inhibitor of the catalytic component of telomerase (hTERT). Like the more extensively characterized BRCT-domain protein 53BP1, MCPH1/BRIT1 regulates the intra-S and G2/M checkpoints, and localizes to gH2AX foci after irradiation (Xu et al. 2004; Lin et al. 2005; Alderton et al. 2006). Depletion of MCPH1/BRIT1 leads to decreased expression of BRCA1 and Chk1, and causes radiosensitivity. The clinical similarity of primary microcephaly, caused by MCPH1/BRIT1 mutations, and ATR–Seckel syndrome, a recessive disease caused by hypomorphic mutations in ATR (O’Driscoll et al. 2004), suggest the importance of an ATR–BRCA1–Chk1 signaling pathway in brain development (Lin et al. 2005; Alderton et al. 2006). MCPH1/BRIT1 contains three BRCT domains that facilitate its interactions with the chromatin-remodeling complex SWI/SNF. By recruiting this complex to the sites of DNA lesions, MCPH1/BRIT1 promotes chromatin relaxation which is thought to facilitate the access to the lesions by DNA repair proteins (Peng et al. 2009).

2.6 Activation of p53 by Upstream Kinases Among the first human proteins found to contribute to the DNA damage response was the tumor suppressor p53. The p53 gene is among the most widely mutated of cancer genes, and contributes to the development of approximately one-half of all cancers. Upon its activation by upstream signaling pathways, p53 contributes significantly to numerous growth inhibitory pathways, including cell cycle checkpoints, apoptosis and senescence (Vogelstein et al. 2000). The p53 protein is modified by a broad array of posttranslational modifications, many of which are believed to


47

contribute to its activation (Horn and Vousden 2007; Kruse and Gu 2008). The best characterized among these are the phosphorylation moieties that are added by the kinases of the DNA damage signaling network (Appella and Anderson 2001). P53 normally has a very short half-life, due to its association with Hdm2, a ubiquitin E3-ligase that targets p53 for proteasomal degradation (Kubbutat et al. 1997). In unperturbed cells, p53 is therefore maintained at low levels. After DNA damage, p53 is phosphorylated on multiple sites. These phosphorylation events are concurrent with the dissociation of p53 from Hdm2 and its stabilization. Thus activated by DNA damage, p53 transactivates the transcription of a large number of genes that contribute to the diverse outcomes of the DNA damage response (Vogelstein et al. 2000). While p53 is known to be strongly activated by DNA damage and replication stress, the exact mechanism of its activation remains unclear (Kruse and Gu 2009). ATM-null cells from patients with ataxia-telangiectasia exhibit a markedly diminished p53 response to ionizing radiation, that is both less robust and temporally delayed compared with cells with wild-type ATM (Canman et al. 1994; Kastan and Lim 2000). ATM, ATR, Chk1, and Chk2 have all been reported to directly phosphorylate p53 on several N-terminal sites (Canman et al. 1998; Tibbetts et al. 1999; Chehab et al. 2000; Hirao et al. 2000), most prominently S15 and S20, that appear to play a role in protein stabilization in vitro. It has been unclear whether the direct phosphorylation of p53 by any one of the upstream kinases is primarily important, or to what extent the p53 response is due to indirect effects of DNA damage signaling. In recent years, studies of knock-in mouse models have called the role of the N-terminal p53 phosphorylation sites into question. Mutations of the mouse equivalents of human S15 and S20 (S18 and S23 in mouse) notably fail to eliminate the responsiveness of p53 levels to DNA damage and other stimuli (Wu et al. 2002; Saito et al. 2003; MacPherson et al. 2004; Sluss et al. 2004), although double mutants in both of these residues do exhibit significant apoptotic defects in some tissues (Chao et al. 2006). These results suggest that p53 stabilization may be a combination of direct and indirect effects of the DNA damage signaling pathways, including the stimulation of other types of posttranslational modifications (Kruse and Gu 2009).

2.7 Diverse Substrates of the Human PIKKs The two-tiered structure of the yeast DNA damage and replication stress pathways (Fig. 2.1) has guided the identification of homologous proteins that function in the human DNA damage response. The checkpoint and mediator proteins that play critical roles in the growth arrest and survival following DNA damage remain the most highly characterized substrates of ATM and ATR. Recently, unbiased proteomic approaches have significantly broadened our view of the extent and scope of the DNA damage responses. Over 700 proteins are robustly phosphorylated at more than 900 sites by the combined activation of ATM and ATR in response to the

48

F. Bunz

DSBs caused by ionizing radiation (Matsuoka et al. 2007). Similarly, UV radiation, a potent inhibitor of DNA replication, was found to induce the phosphorylation of nearly 500 proteins (Stokes et al. 2007; Stokes and Comb 2008). Within these two unexpectedly large sets of proteins are many that are commonly activated by both stimuli and therefore probably represent the combined effects of ATM and ATR activation. A significant number of substrates appear to be exclusively dependent on either ATM or ATR (Stokes and Comb 2008). A common theme among the downstream targets of ATM/ATR is the phosphorylation of multiple targets within individual pathways, suggesting many potential nodes of regulation (Matsuoka et al. 2007). For example, many substrates participate in the regulation of the successive stages of DNA replication. Prevalent among the ATM/ATR target proteins are those that are involved in each step of DNA replication, including origin recognition (ORC proteins), replication complex assembly (MCM proteins), clamp loading (RFC1 and RFC3), and DNA synthesis (DNA polymerase epsilon, GINS). Identification of these substrates promises mechanistic insights into the genetically observed roles of ATM and ATR in controlling DNA replication after DNA damage. Other functional modules impacted by the combined functions of ATM and ATR are the Fanconi anemia pathway and the nucleotide excision repair pathway, and combined pathways required for DNA repair by homologous recombination (Matsuoka et al. 2007). The functional diversity of the ATM/ATR phosphoproteome is striking. Many of the recently identified ATM/ATR substrates implicate pathways with no prior relationship to DNA damage signaling. For example, multiple proteins in the PTEN/ AKT pathway, which responds to growth stimuli such as insulin, were found to be robust ATM/ATR substrates, including AKT, its adaptors, regulators, and downstream effectors (Matsuoka et al. 2007). The elucidation of these pathways, and the ways they are functionally integrated will provide experimental challenges for years to come.

References Abraham RT (2001) Cell cycle checkpoint signaling through the ATM and ATR kinases. Genes Dev 15:2177–96 Adams KE, Medhurst AL, Dart DA et al (2006) Recruitment of ATR to sites of ionising radiationinduced DNA damage requires ATM and components of the MRN protein complex. Oncogene 25:3894–904 Aglipay JA, Martin SA, Tawara H et al (2006) ATM activation by ionizing radiation requires BRCA1-associated BAAT1. J Biol Chem 281:9710–8 Ahn J, Urist M, Prives C (2004) The Chk2 protein kinase. DNA Repair (Amst) 3:1039–47 Alderton GK, Galbiati L, Griffith E et al (2006) Regulation of mitotic entry by microcephalin and its overlap with ATR signalling. Nat Cell Biol 8:725–33 Appella E, Anderson CW (2001) Post-translational modifications and activation of p53 by genotoxic stresses. Eur J Biochem 268:2764–72. Bakkenist CJ, Kastan MB (2004) Initiating cellular stress responses. Cell 118:9–17 Bartek J, Lukas J (2003) Chk1 and Chk2 kinases in checkpoint control and cancer. Cancer Cell 3:421–9


49

Bartek J, Falck J, Lukas J (2001) CHK2 kinase – a busy messenger. Nat Rev Mol Cell Biol 2:877–86. Bell DW, Varley JM, Szydlo TE et al (1999) Heterozygous germ line hCHK2 mutations in Li–Fraumeni syndrome. Science 286:2528–31 Boutros R, Dozier C, Ducommun B (2006) The when and wheres of CDC25 phosphatases. Curr Opin Cell Biol 18:185–91 Brown EJ, Baltimore D (2003) Essential and dispensable roles of ATR in cell cycle arrest and genome maintenance. Genes Dev 17:615–28 Canman CE, Wolff AC, Chen CY et al (1994) The p53-dependent G1 cell cycle checkpoint pathway and ataxia- telangiectasia. Cancer Res 54:5054–8 Canman CE, Lim DS, Cimprich KA et al (1998) Activation of the ATM kinase by ionizing radiation and phosphorylation of p53. Science 281:1677–9 Casper AM, Nghiem P, Arlt MF et al (2002) ATR regulates fragile site stability. Cell 111:779–89 Chao C, Herr D, Chun J et al (2006) Ser18 and 23 phosphorylation is required for p53-dependent apoptosis and tumor suppression. EMBO J 25:2615–22 Chehab NH, Malikzay A, Appel M et al (2000) Chk2/hCds1 functions as a DNA damage checkpoint in G(1) by stabilizing p53. Genes Dev 14:278–88 Chini CC, Chen J (2004) Claspin, a regulator of Chk1 in DNA replication stress pathway. DNA Repair (Amst) 3:1033–7 Chini CC, Chen J (2006) Repeated phosphopeptide motifs in human Claspin are phosphorylated by Chk1 and mediate Claspin function. J Biol Chem 281:33276–82 Cimprich KA, Cortez D (2008) ATR: an essential regulator of genome integrity. Nat Rev Mol Cell Biol 9:616–27 Cortez D, Wang Y, Qin J et al (1999) Requirement of ATM-dependent phosphorylation of brca1 in the DNA damage response to double-strand breaks. Science 286:1162–6 Cortez D, Guntuku S, Qin J et al (2001) ATR and ATRIP: Partners in Checkpoint Signaling. Science 294:1713–6 Cuadrado M, Martinez-Pastor B, Murga M et al (2006) ATM regulates ATR chromatin loading in response to DNA double-strand breaks. J Exp Med 203:297–303 Deng CX (2006) BRCA1: cell cycle checkpoint, genetic instability, DNA damage response and cancer evolution. Nucleic Acids Res 34:1416–26 DiTullio RA Jr, Mochan TA, Venere M et al (2002) 53BP1 functions in an ATM-dependent checkpoint pathway that is constitutively activated in human cancer. Nat Cell Biol 4:998–1002 Dodson GE, Tibbetts RS (2006) DNA replication stress-induced phosphorylation of cyclic AMP response element-binding protein mediated by ATM. J Biol Chem 281:1692–7 Fanning E, Klimovich V, Nager AR (2006) A dynamic model for replication protein A (RPA) function in DNA processing pathways. Nucleic Acids Res 34:4126–37 Gatei M, Sloper K, Sorensen C et al (2003) Ataxia-telangiectasia-mutated (ATM) and NBS1dependent phosphorylation of Chk1 on Ser-317 in response to ionizing radiation. J Biol Chem 278:14806–11 Goldberg M, Stucki M, Falck J et al (2003) MDC1 is required for the intra-S-phase DNA damage checkpoint. Nature 421:952–6 Gurtan AM, D’Andrea AD (2006) Dedicated to the core: understanding the Fanconi anemia complex. DNA Repair (Amst) 5:1119–25 Hirao A, Kong YY, Matsuoka S et al (2000) DNA damage-induced activation of p53 by the checkpoint kinase Chk2. Science 287:1824–7 Horn HF, Vousden KH (2007) Coping with stress: multiple ways to activate p53. Oncogene 26:1306–16 Hurley PJ, Bunz F (2007) ATM and ATR: Components of an integrated circuit. Cell Cycle 6:414–7 Hurley PJ, Wilsker D, Bunz F (2007) Human cancer cells require ATR for cell cycle progression following exposure to ionizing radiation. Oncogene 26:2535–42 Huyen Y, Zgheib O, Ditullio RA Jr et al (2004) Methylated lysine 79 of histone H3 targets 53BP1 to DNA double-strand breaks. Nature 432:406–11

50

F. Bunz

Iwabuchi K, Bartel PL, Li B et al (1994) Two cellular proteins that bind to wild-type but not mutant p53. Proc Natl Acad Sci USA 91:6098-102 Jack MT, Woo RA, Hirao A et al (2002) Chk2 is dispensable for p53-mediated G1 arrest but is required for a latent p53-mediated apoptotic response. Proc Natl Acad Sci USA 3:3 Jallepalli PV, Lengauer C, Vogelstein B et al (2003) The Chk2 tumor suppressor is not required for p53 responses in human cancer cells. J Biol Chem 278:20475–9 Jazayeri A, Falck J, Lukas C et al (2006) ATM- and cell cycle-dependent regulation of ATR in response to DNA double-strand breaks. Nat Cell Biol 8:37–45 Jowsey P, Morrice NA, Hastie CJ et al (2007) Characterisation of the sites of DNA damageinduced 53BP1 phosphorylation catalysed by ATM and ATR. DNA Repair (Amst) 6:1536–44 Karlsson-Rosenthal C, Millar JB (2006) Cdc25: mechanisms of checkpoint inhibition and recovery. Trends Cell Biol 16:285–92 Kastan MB, Lim DS (2000) The many substrates and functions of ATM. Nat Rev Mol Cell Biol 1:179–86 Kruse JP, Gu W (2008) SnapShot: p53 posttranslational modifications. Cell 133:930–30.e1 Kruse JP, Gu W (2009) Modes of p53 regulation. Cell 137:609–22 Kubbutat MH, Jones SN, Vousden KH (1997) Regulation of p53 stability by Mdm2. Nature 387:299–303 Kumagai A, Dunphy WG (2000) Claspin, a novel protein required for the activation of Chk1 during a DNA replication checkpoint response in Xenopus egg extracts. Mol Cell 6:839–49 Kumagai A, Dunphy WG (2003) Repeated phosphopeptide motifs in Claspin mediate the regulated binding of Chk1. Nat Cell Biol 5:161–5 Kumagai A, Dunphy WG (2006) How cells activate ATR. Cell Cycle 5:1265–8 Lavin MF, Birrell G, Chen P et al (2005) ATM signaling and genomic stability in response to DNA damage. Mutat Res 569:123–32 Lee JS, Collins KM, Brown AL et al (2000) hCds1-mediated phosphorylation of BRCA1 regulates the DNA damage response. Nature 404:201–4 Lee J, Kumagai A, Dunphy WG (2007) The Rad9-Hus1-Rad1 checkpoint clamp regulates interaction of TopBP1 with ATR. J Biol Chem 282:28036–44 Lin SY, Rai R, Li K et al (2005) BRIT1/MCPH1 is a DNA damage responsive protein that regulates the Brca1-Chk1 pathway, implicating checkpoint dysfunction in microcephaly. Proc Natl Acad Sci USA 102:15105–9 Liu Q, Guntuku S, Cui XS et al (2000) Chk1 is an essential kinase that is regulated by Atr and required for the G(2)/M DNA damage checkpoint. Genes Dev 14:1448–59 Liu S, Bekker-Jensen S, Mailand N et al (2006) Claspin operates downstream of TopBP1 to direct ATR signaling towards Chk1 activation. Mol Cell Biol 26:6056–64 Lukas C, Melander F, Stucki M et al (2004) Mdc1 couples DNA double-strand break recognition by Nbs1 with its H2AX-dependent chromatin retention. EMBO J 23:2674–83 MacPherson D, Kim J, Kim T et al (2004) Defective apoptosis and B-cell lymphomas in mice with p53 point mutation at Ser 23. EMBO J 23:3689–99 Manke IA, Nguyen A, Lim D et al (2005) MAPKAP kinase-2 is a cell cycle checkpoint kinase that regulates the G2/M transition and S phase progression in response to UV irradiation. Mol Cell 17:37–48 Matsuoka S, Ballif BA, Smogorzewska A et al (2007) ATM and ATR substrate analysis reveals extensive protein networks responsive to DNA damage. Science 316:1160–6 McGowan CH (2002) Checking in on Cds1 (Chk2): A checkpoint kinase and tumor suppressor. Bioessays 24:502–11 McGowan CH and Russell P (2004) The DNA damage response: sensing and signaling. Curr Opin Cell Biol 16:629-633 Meijers-Heijboer H, van den Ouweland A, Klijn J et al (2002) Low-penetrance susceptibility to breast cancer due to CHEK2(*)1100delC in noncarriers of BRCA1 or BRCA2 mutations. Nat Genet 31:55–9 Mochan TA, Venere M, DiTullio RA Jr et al (2003) 53BP1 and NFBD1/MDC1-Nbs1 function in parallel interacting pathways activating ataxia-telangiectasia mutated (ATM) in response to DNA damage. Cancer Res 63:8586–91


51

Mohammad DH, Yaffe MB (2009) 14-3-3 proteins, FHA domains and BRCT domains in the DNA damage response. DNA Repair (Amst) 8:1009–17 Mordes DA, Glick GG, Zhao R et al (2008) TopBP1 activates ATR through ATRIP and a PIKK regulatory domain. Genes Dev 22:1478–89 Myers JS, Cortez D (2006) Rapid activation of ATR by ionizing radiation requires ATM and Mre11. J Biol Chem 281:9346–50 Niida H, Katsuno Y, Banerjee B et al (2007) Specific role of Chk1 phosphorylations in cell survival and checkpoint activation. Mol Cell Biol 27:2572–81 O’Driscoll M, Gennery AR, Seidel J et al (2004) An overview of three new disorders associated with genetic instability: LIG4 syndrome, RS-SCID and ATR–Seckel syndrome. DNA Repair (Amst) 3:1227–235 Okada S, Ouchi T (2003) Cell cycle differences in DNA damage-induced BRCA1 phosphorylation affect its subcellular localization. J Biol Chem 278:2015–20 Oliver AW, Knapp S, Pearl LH (2007) Activation segment exchange: a common mechanism of kinase autophosphorylation? Trends Biochem Sci 32:351–6 Osborn AJ, Elledge SJ, Zou L (2002) Checking on the fork: the DNA-replication stress-response pathway. Trends Cell Biol 12:509–16 Ouchi T (2006) BRCA1 phosphorylation: biological consequences. Cancer Biol Ther 5:470–5 Parrilla-Castellar ER, Arlander SJ, Karnitz L (2004) Dial 9-1-1 for DNA damage: the Rad9-Hus1Rad1 (9-1-1) clamp complex. DNA Repair (Amst) 3:1009–14 Peng G, Yim EK, Dai H et al (2009) BRIT1/MCPH1 links chromatin remodelling to DNA damage response. Nat Cell Biol 11:865–72 Puc J, Parsons R (2005) PTEN loss inhibits CHK1 to cause double stranded-DNA breaks in cells. Cell Cycle 4:927–9 Puc J, Keniry M, Li HS et al (2005) Lack of PTEN sequesters CHK1 and initiates genetic instability. Cancer Cell 7:193–204 Reinhardt HC, Yaffe MB (2009) Kinases that control the cell cycle in response to DNA damage: Chk1, Chk2, and MK2. Curr Opin Cell Biol 21:245–5 Roux PP, Blenis J (2004) ERK and p38 MAPK-activated protein kinases: a family of protein kinases with diverse biological functions. Microbiol Mol Biol Rev 68:320–44 Saito S, Yamaguchi H, Higashimoto Y et al (2003) Phosphorylation site interdependence of human p53 post-translational modifications in response to stress. J Biol Chem 278:37536–44 Shieh SY, Ahn J, Tamai K et al (2000) The human homologs of checkpoint kinases Chk1 and Cds1 (Chk2) phosphorylate p53 at multiple DNA damage-inducible sites. Genes Dev 14:289–300 Shiloh Y (2006) The ATM-mediated DNA-damage response: taking shape. Trends Biochem Sci 31:402–10 Sluss HK, Armata H, Gallant J et al (2004) Phosphorylation of serine 18 regulates distinct p53 functions in mice. Mol Cell Biol 24:976–84 Smits VA (2006) Spreading the signal: dissociation of Chk1 from chromatin. Cell Cycle 5:1039–43 Smits VA, Reaper PM, Jackson SP (2006) Rapid PIKK-dependent release of Chk1 from chromatin promotes the DNA-damage checkpoint response. Curr Biol 16:150–9 Sodha N, Williams R, Mangion J et al (2000) Screening hCHK2 for mutations. Science 289:359. Sodha N, Houlston RS, Bullock S et al (2002) Increasing evidence that germline mutations in CHEK2 do not cause Li–Fraumeni syndrome. Hum Mutat 20:460–2 Stewart GS, Wang B, Bignell CR et al (2003) MDC1 is a mediator of the mammalian DNA damage checkpoint. Nature 421:961–6 Stiff T, Walker SA, Cerosaletti K et al (2006) ATR-dependent phosphorylation and activation of ATM in response to UV treatment or replication fork stalling. EMBO J 25:5775–82 Stokes MP, Comb MJ (2008) A wide-ranging cellular response to UV damage of DNA. Cell Cycle 7:2097–9 Stokes MP, Rush J, Macneill J et al (2007) Profiling of UV-induced ATM/ATR signaling pathways. Proc Natl Acad Sci USA 104:19855–60 Stracker TH, Usui T, Petrini JH (2009) Taking the time to make important decisions: the checkpoint effector kinases Chk1 and Chk2 and the DNA damage response. DNA Repair (Amst) 8:1047–54

52

F. Bunz

Stucki M, Jackson SP (2004) MDC1/NFBD1: a key regulator of the DNA damage response in higher eukaryotes. DNA Repair (Amst) 3:953–7 Tibbetts RS, Brumbaugh KM, Williams JM et al (1999) A role for ATR in the DNA damageinduced phosphorylation of p53. Genes Dev 13:152–7 Vogelstein B, Lane D, Levine AJ (2000) Surfing the p53 network. Nature 408:307–10 Wang B, Matsuoka S, Carpenter PB et al (2002) 53BP1, a mediator of the DNA damage checkpoint. Science 298:1435–8 Wang X, Zou L, Lu T et al (2006) Rad17 phosphorylation is required for claspin recruitment and Chk1 activation in response to replication stress. Mol Cell 23:331–41 Wang X, Kennedy RD, Ray K et al (2007) Chk1-mediated phosphorylation of FANCE is required for the Fanconi anemia/BRCA pathway. Mol Cell Biol 27:3098–108 Ward IM, Minn K, van Deursen J et al (2003) p53 Binding protein 53BP1 is required for DNA damage responses and tumor suppression in mice. Mol Cell Biol 23:2556–63 Ward IM, Minn K, Chen J (2004) UV-induced ataxia-telangiectasia-mutated and Rad3-related (ATR) activation requires replication stress. J Biol Chem 279:9677–80 Ward IM, Difilippantonio S, Minn K et al (2005) 53BP1 cooperates with p53 and functions as a haploinsufficient tumor suppressor in mice. Mol Cell Biol 25:10079–86 Ward I, Kim JE, Minn K et al (2006) The tandem BRCT domain of 53BP1 is not required for its repair function. J Biol Chem 281:38472–7 Weischer M, Bojesen SE, Ellervik C et al (2008) CHEK2*1100delC genotyping for clinical assessment of breast cancer risk: meta-analyses of 26,000 patient cases and 27,000 controls. J Clin Oncol 26:542–8 Wilsker D, Petermann E, Helleday T et al (2008) Essential function of Chk1 can be uncoupled from DNA damage checkpoint and replication control. Proc Natl Acad Sci USA 105:20752–7 Wilson KA, Stern DF (2008) NFBD1/MDC1, 53BP1 and BRCA1 have both redundant and unique roles in the ATM pathway. Cell Cycle 7:3584–94 Wu Z, Earle J, Saito S et al (2002) Mutation of mouse p53 Ser23 and the response to DNA damage. Mol Cell Biol 22:2441–9 Xiao Z, Xue J, Sowin TJ et al (2006) Differential roles of checkpoint kinase 1, checkpoint kinase 2, and mitogen-activated protein kinase-activated protein kinase 2 in mediating DNA damageinduced cell cycle arrest: implications for cancer therapy. Mol Cancer Ther 5:1935–43 Xu B, Kim S, Kastan MB (2001) Involvement of Brca1 in S-phase and G(2)-phase checkpoints after ionizing irradiation. Mol Cell Biol 21:3445–50 Xu X, Lee J, Stern DF (2004) Microcephalin is a DNA damage response protein involved in regulation of CHK1 and BRCA1. J Biol Chem 279:34091–4 Xu C, Wu L, Cui G et al (2008) Structure of a second BRCT domain identified in the nijmegen breakage syndrome protein Nbs1 and its function in an MDC1-dependent localization of Nbs1 to DNA damage sites. J Mol Biol 381:361–72 Yu X, Chini CC, He M et al (2003) The BRCT domain is a phospho-protein binding domain. Science 302:639–42 Zegerman P, Diffley JF (2009) DNA replication as a target of the DNA damage checkpoint. DNA Repair (Amst) 8:1077–88 Zhang J, Willers H, Feng Z et al (2004) Chk2 phosphorylation of BRCA1 regulates DNA doublestrand break repair. Mol Cell Biol 24:708–18 Zhao H, Piwnica-Worms H (2001) ATR-mediated checkpoint pathways regulate phosphorylation and activation of human Chk1. Mol Cell Biol 21:4129–39 Zou L, Elledge SJ (2003) Sensing DNA damage through ATRIP recognition of RPA–ssDNA complexes. Science 300:1542–8

Chapter 3

Checkpoint Control Following Radiation Exposure Markus Lobrich, Aaron A. Goodarzi, Tom Stiff, and Penny A. Jeggo

Abstract The DNA damage response involves processes of DNA repair and signal transduction pathways. DNA double-strand breaks (DSBs) are the most biologically significant lesion induced by ionizing radiation (IR). DNA nonhomologous end-joining and homologous recombination represent the major DSB repair pathways and ataxia telangiectasia mutated (ATM) lies at the heart of the DSB signaling response. However, ATM and Rad3 related (ATR) can also be activated in S and G2 phase following IR exposure. A major end point of damage response signaling is the activation of cell cycle checkpoint arrest. In addition to the initiation of checkpoint arrest, recent studies have demonstrated that the signaling pathway monitors the progress of DSB repair to ensure timely checkpoint release, a process called the maintenance of checkpoint arrest. In this chapter, we overview DNA damage-induced cell cycle checkpoint arrest following exposure to IR. We discuss how cell cycle stage impacts upon the roles of ATM and ATR, how they influence the initiation and maintenance of checkpoint signaling, and the interface between checkpoint arrest and DSB repair. We evaluate current insight into the sensitivity of the processes and the impact of higher order chromatin structure on damage response signaling. Keywords DNA double strand breaks • Cell cycle checkpoint arrest • Radiationinduced damage responses • G1/S checkpoint arrest • G2/M checkpoint arrest

P.A. Jeggo (*) Genome Damage and Stability Centre, University of Sussex, Brighton, East Sussex BN1 9RQ, UK e-mail: [email protected] T.L. DeWeese and M. Laiho (eds.), Molecular Determinants of Radiation Response, Current Cancer Research, DOI 10.1007/978-1-4419-8044-1_3, © Springer Science+Business Media, LLC 2011

53

54

M. Lobrich et al.

3.1 Introduction The DNA damage response (DDR) encompasses processes of DNA repair and two overlapping signal transduction pathways. A range of distinct DNA repair pathways recognize different classes of DNA damage, some of which have inbuilt additional specificity, such as the distinct glycolyases that function in base excision repair. Collectively, these processes allow the recognition and repair of a broad range of DNA lesions. In contrast, the signal transduction response is activated by one of two broad classes of lesion, a DNA double-strand break (DSB) or a single-stranded (ss) region of DNA. Two phosphoinositol 3-kinase-like kinases (PIKKs) lie at the center of the signaling response; ataxia telangiectasia mutated (ATM) regulates the response to DSBs and the related ATM and Rad3-related protein (ATR) is activated by ss DNA regions. The major DNA repair processes [e.g., nucleotide excision repair (NER), single-strand break (SSB), and DSB repair] function largely independently of the signal transduction pathways. However, there is increasing evidence of interplay between DNA repair pathways and ATM/ATR signaling. An additional and arguably more significant aspect of the signaling response is the regulation of cell cycle checkpoint arrest and, in higher organisms, the process of apoptosis. In this review, we focus on the processes of cell cycle checkpoint arrest and their regulation by damage response signaling. Our major focus lies on the response to DNA DSBs. Since DSBs are a biologically significant lesion induced by ionizing radiation (IR), our analysis will primarily consider IR-induced checkpoint arrest. Three damage response checkpoints have been described: G1/S, G2/M, and intra-S-phase arrest. We will review the current knowledge of the processes underlying these checkpoints and specifically consider their efficacy in enhancing survival and limiting genomic instability.

3.2 PIKK Activation and Signaling The first step in signaling DNA damage to the cell cycle checkpoint machinery is the activation of one or both of the PIKKs. IR induces base damage, SSBs, DSBs and a lower number of DNA inter- and intrastrand cross-links. Base damage can be processed into SSBs, and the ratio of SSB:DSB damage induced by IR is 20:1 (Nikjoo et al. 2002). A critical aspect in considering the mechanisms underlying cell cycle checkpoint arrest is the distinct structures that lead to PIKK activation. First, we will overview the current knowledge of the steps activating ATM and ATR and then consider PIKK activation following IR exposure.

3.2.1 ATM Activation The mechanism leading to ATM activation is, perhaps, better understood than that resulting in ATR activation. Several excellent recent reviews have discussed the

3 Checkpoint Control Following Radiation Exposure

55

steps involved in ATM activation and the assembly of ionizing radiation-induced foci (IRIF) and only an outline of the process will be provided here (Lavin 2008; Panier and Durocher 2009; van Attikum and Gasser 2009). Increasing evidence suggests that the Mre11-Rad50-NBS1 (MRN) complex represents the initial DNA DSB sensor (Carson et al. 2003; Uziel et al. 2003). Importantly, the extreme C terminus of NBS1 encompasses an evolutionary conserved motif that promotes interaction with ATM, providing a route by which ATM is initially recruited to DSBs (Falck et al. 2005). In undamaged cells, ATM exists as an inactive homodimer and its DSB recruitment results in autophosphorylation and monomerisation, a process leading to sustained activation (Bakkenist and Kastan 2003). An early step following ATM activation is phosphorylation of the C terminus of H2AX, the variant form of the histone H2A, creating phosphorylated H2AX (phospho-S139 H2AX), termed gH2AX (Rogakou et al. 1998). Although H2AX phosphorylation initiates in the region directly flanking the DSB, phosphorylation occurs at increasing distances from the break site with time so that the region containing phosphorylated H2AX can cover mega DNA base pairs at persisting DSBs (Rogakou et al. 1999). Thus, an extended region of H2AX flanking the DSB becomes phosphorylated, a critical factor leading to visible IRIF. gH2AX binds to the tandem BRCT domains of the mediator protein, MDC1 (Mediator of DNA damage checkpoint 1), providing a critical first step in mediator protein recruitment (Stucki and Jackson 2004). The N terminus of MDC1 encompasses a recently identified SDT (Ser–Asp–Thr) repeat motif in close proximity to the tandem BRCT domain. The SDT motifs are phosphorylated by CK2 both endogenously and following IR exposure and mediate an interaction with the FHA domain of NBS1, providing a route to tether MRN and ATM at the DSB site (Chapman and Jackson 2008; Melander et al. 2008; Spycher et al. 2008). A highly significant finding is that MDC1 facilitates the recruitment of two ubiquitin E3-ligases, RNF8 (RING finger 8) and RNF168 (RING finger 168), to the DSB site, a pivotal result exposing the importance of ubiquitylation as a post-translational modification in the DDR (Huen et al. 2007; Kolas et al. 2007; Mailand et al. 2007; Doil et al. 2009; Stewart et al. 2009). Ubiquitylation requires a ubiquitin-activating enzyme (E1), a ubiquitin-conjugating enzyme (E2) and a ubiquitin ligase (E3). The FHA domain of RNF8 interacts with an ATM phosphorylation motif in the N terminus of MDC1. RNF8 also interacts with UBC13, an E2 ubiquitin-conjugating enzyme and catalyzes K63-linked ubiquitylation (Plans et al. 2006). The second E3 ligase, RNF168, was identified as the gene defective in a human disorder termed RIDDLE (radiosensitivity, immunodeficiency, dysmorphic features and learning difficulties) Syndrome (Stewart et al. 2007, 2009; Doil et al. 2009). RNF168 recruitment requires the RNF8 RING finger motif suggesting that its recruitment itself requires ubiquitylation at DSBs. In very recent work, HERC2 has been identified as a further protein that is required for ubiquitylation at DSBs (Bekker-Jensen et al. 2010). HERC2 encompasses HECT domains and is an ATM substrate. ATM-dependent phosphorylation of HERC2 mediates its interaction with the FHA domain of RNF8, which facilitates assembly of UBC13 with RNF8, thereby promoting K63-linked ubiquitylation. The main target of these ubiquitin ligases is H2A/H2AX. An important end point of H2A

56

M. Lobrich et al.

ubiquitylation is recruitment of another mediator protein, 53BP1 (Huen et al. 2007; Kolas et al. 2007; Mailand et al. 2007). The recruitment of BRCA1 to the DSB also depends on the RNF8-RNF168 ubiquitin ligases although the precise mechanism and function of this loop of the signaling process is unclear. BRCA1 is itself an ubiquitin ligase and it has been suggested that it further enhances the concentration of ubiquitin moieties at the DSB. Recent studies have provided evidence that 53BP1 interacts with Rad50 via 53BP1’s BRCT domain, providing an additional step promoting MRN hyper-accumulation at the DSB (Lee et al. 2009). This step also appears to be critical to promote ATM chromatin retention and the formation of p-1981-ATM foci (Noon et al. 2010). Importantly, although the phosphorylation of H2AX and the recruitment of the mediator proteins are important steps leading to p-1981-ATM foci formation, they are not essential for ATM activation and have only a modest impact on ATM substrate phosphorylation, although the precise impact is substrate specific. Collectively, these findings demonstrate that ATM regulates an orchestrated and sophisticated assembly of proteins at the DSB, which results in modifications of histones in the DSB vicinity. Increasing evidence suggests that changes to the higher order chromatin superstructure, including factors influencing heterochromatinisation, promotes DSB repair within regions of heterochromatin (Goodarzi et al. 2008; Noon et al. 2010).

3.2.2 ATR Activation ATR activation and signaling has also previously been reviewed in detail and will only be briefly discussed here (Cimprich and Cortez 2008). In contrast to ATM, ATR is essential since it is activated during every S phase. ATR is activated by ss regions of DNA, following their coating with the ss DNA-binding protein, RPA (Zou and Elledge 2003). Localization of ATR to ss DNA also requires ATR interacting protein (ATRIP), in part because ATRIP directly binds to RPA (Ball et al. 2007). However, additionally ATR and ATRIP closely interact and co-regulate each other’s stability (Cortez et al. 2001). Although ATR localization to RPA-coated ss DNA may require only ATRIP, sustained ATR activation also requires the colocalization of Rad17 and the so-called 9-1-1 complex, involving Rad9-Rad1-Hus1. The 9-1-1 complex represents a heterotrimeric ring-shaped molecule akin to PCNA. Loading of the 9-1-1 complex requires the recruitment of the damage-specific clamp loader, Rad17, to the RPA-coated ss DNA (Zou et al. 2003). Critically, the recruitment of the 9-1-1 complex promotes the recruitment of TOPBP1, a multiBRCT domain protein, which appears to be a critical step in ATR activation (Furuya et al. 2004; Kumagai et al. 2006; Delacroix et al. 2007; Lee et al. 2007). Interestingly, the recruitment and assembly of the Rad17/9-1-1 complex to RPA-coated ss DNA occurs largely independently of ATR-ATRIP recruitment, a feature which is distinct to the more step-wise orchestrated recruitment of proteins at a DSB. A critical issue in evaluating ATR signaling is the length of ss DNA required for ATR activation. Consistent with the findings discussed above that RPA binding to


57

ss DNA regions is required for ATR activation, neither base damage nor ss nicks appear to activate ATR. ss DNA of sufficient length to bind RPA occurs predominantly at stalled or collapsed replication forks, and indeed this is a situation where ATR signaling plays its major role. ss DNA regions can also occur during the processing of certain DNA lesions, such as a pyrimidine dimer, although it is unclear whether ATR signaling is activated by the ss regions generated during NER or whether it requires exonucleolytic processing to generate longer and perhaps more accessible ss DNA regions (Nakada et al. 2004; Marini et al. 2006). ss DNA of sufficient length can also be generated in G2 phase at a resected DSB, which will be discussed in further detail below.

3.2.3 Activation of ATM vs. ATR Following IR Exposure Although base damage and SSBs are abundant lesions induced by IR, they are rapidly repaired and, importantly, do not directly activate either ATM or ATR signaling. In G0/G1 phase, SSBs do not undergo extensive resection and the available evidence suggests that ATR is not activated even at prolonged times after IR exposure (Stiff et al. 2004). Thus, following IR exposure, the signaling response and G1/S checkpoint arrest is predominantly ATM dependent in G0/G1 phase cells (Kastan et al. 1992; Lavin et al. 1994). SSBs are also rapidly repaired in G2 phase and it appears that, as in G1 phase, they do not activate damage signaling response (Shibata, unpublished findings). However, in G2 phase, resection can occur at least at a subset of DSBs resulting in ATR activation (Jazayeri et al. 2006). Importantly, resection at DSBs in G2 phase is ATM dependent (Jazayeri et al. 2006). Thus, in G2 and in G1 phase cell cycle checkpoint arrest after IR is ATM dependent although ATR may additionally contribute to checkpoint signaling in G2 phase as resection ensues (see below for further discussion of the interplay between ATM and ATR in G2/M checkpoint arrest) (Lavin et al. 1994; Shibata et al. 2010). In S phase, the situation is different since ATR can be activated as a consequence of replication fork stalling. Given the 20:1 ratio of SSBs:DSBs, replication fork stalling at SSBs can be significant, although the relevance of SSBs for damage response signaling is diminished by their rapid rate of repair. The prevailing evidence suggests that DSBs generated in S phase can directly activate ATM, as they do in G1 or G2 phase, and that, additionally, signaling can be complemented by ATR activation at stalled or collapsed replication forks (Byun et al. 2005). In addition to ATM-dependent activation of ATR following resection in S or G2 phase, there is also evidence that the generation of one-ended DSBs at collapsed replication forks can lead to ATM activation (Mirzoeva and Petrini 2003; Helleday et al. 2007). Further it has been proposed that ATR can phosphorylate ATM leading to ATR-dependent ATM activation (Stiff et al. 2006). Thus, the interplay between ATM and ATR signaling is complex and critically depends upon cell cycle phase. In summary, in G1 and G2 phase, almost all damage response signaling after IR exposure arises from DSBs and is ATM dependent at early times (Fig. 3.1). In S phase, ATR can additionally contribute due to its direct

58

M. Lobrich et al.

Fig. 3.1 Overview of roles of ATM and ATR in cell cycle checkpoint arrest processes. (1) DSBs in G1 phase directly activate ATM leading to Chk2/p53 activation and G1/S checkpoint arrest. This is a sensitive process. (2) In S phase, replication fork collision with SSBs or DSBs can lead to ATR and Chk1 activation and inhibition of origin firing. ATM-Chk2 can also be directly activated by DSBs in S phase. (3) In G2 phase, DSBs directly activate ATM-Chk2 causing G2/M phase arrest. Additionally, ATM activation promotes DSB resection leading to ATR-Chk1 activation. G2/M phase arrest has a threshold sensitivity of 10–20 DSBs. S phase arrest slows progression through S phase but cells can eventually enter G2 phase with damage. ATR also contributes to G2/M phase arrest of irradiated G1 or S phase cells. (a) Describes the progression through the cell cycle and pathways activated and (b) Describes the sensitivity of the checkpoint response

activation at stalled/collapsed replication forks. ATM-dependent resection can lead to ATR activation in S and G2 phase cells.

3.2.4 Signaling from ATM/ATR to the Transducer Kinases ATM and ATR signal to the checkpoint machinery via one of two transducer kinases, Chk1 and Chk2 (Matsuoka et al. 1998, 2000; see Bartek and Lukas 2003,


59

for a review). A major function of these kinases is to relay the presence of DNA damage or the lack of genome integrity to the proteins that control cell cycle progression, namely the cyclin-dependent kinases (Cdks), and is achieved via inhibition/activation of phosphatases or kinases that regulate Cdk phosphorylation and activity. More recently, however, there is evidence that the transducer kinases may also regulate aspects of DNA repair (Sorensen et al. 2005). Perhaps surprisingly, although ATM and ATR share many substrates, their signaling to the transducer kinases appears to be specific, with ATM primarily phosphorylating and activating Chk2 while ATR signals in a similar way via Chk1 (Chaturvedi et al. 1999; Liu et al. 2000; Jazayeri et al. 2006). Chk2 is constitutively expressed but becomes activated uniquely in the presence of DNA damage (Lukas et al. 2001). In contrast, Chk1 is labile and its expression is increased in S/G2 phase (Lukas et al. 2001). Like ATR, Chk1 is activated during normal S phase progression. The activation of both Chk1 and Chk2 is regulated by PIKK-dependent phosphorylation. Chk2 is phosphorylated at a unique site in its regulatory N-terminal region, T68, predominantly by ATM. This triggers dimerisation and autophosphorylation in the C-terminal kinase domain (Matsuoka et al. 2000; Ahn et al. 2002; Xu et al. 2002b). Although p-T68-Chk2 can be observed in foci after DNA damage, live cell imaging analysis has shown that p-Chk2 is rapidly released from the damage site consistent with the notion that it functions as a signal transducer (Bartek and Lukas 2003). Chk1 is phosphorylated by ATR at several SQ sites in its C terminus including S317 and S345. Further details of the pathways leading from transducer kinase activation to cell cycle arrest will be discussed for the individual checkpoints below.

3.3 Cell Cycle Checkpoint Activation, Maintenance, and Adaptation The process of cell cycle checkpoint arrest can be divided into a number of stages which, at least to some extent, may be genetically and mechanistically distinct. The exploitation of IR is particularly useful for considering these distinct stages since the damage is rapidly induced allowing the processes of checkpoint initiation and maintenance to be distinguished. The presence of phosphorylated Chk1 or Chk2 (p-Chk1 or p-Chk2) can be observed at very early times (5–30 min) following IR treatment, although full checkpoint arrest takes more time to become evident. Nonetheless, in most cell lines, G2/M checkpoint arrest assessed by a dramatic reduction in mitotic cells is evident by 1 h post-IR. This process represents checkpoint activation. Subsequently, as DSB repair ensues a point is reached when the molecular signals relaying checkpoint arrest are no longer maintained and entry into mitosis recommences. Following exposure to higher IR doses, the process of initial checkpoint activation can be distinguished from the maintenance of checkpoint arrest. At lower doses, this is difficult to distinguish since the duration of checkpoint arrest may be short. It is also difficult to verify whether checkpoint initiation occurs at very low doses if the arrest is of short duration. From our own

60

M. Lobrich et al.

work, we have found that the presence of pChk1 or pChk2 provides a good correlation with checkpoint arrest and that addition of a Chk1/2 inhibitor (e.g., 600 nM UCN-01, which inactivates both checkpoint kinases) abolishes both the initiation and maintenance of checkpoint arrest (examined by adding UCN-01 after initiation has been established) (Shibata et al. manuscript submitted). Published findings have shown that the duration of checkpoint arrest depends upon dose and DSB repair capacity (Deckbar et al. 2007). Collectively, these findings provide strong evidence that the checkpoint machinery monitors the status of DSB repair to ensure timely checkpoint release. This process will be discussed further in the section considering G2/M checkpoint arrest. An additional factor in considering the maintenance of checkpoint arrest is the progression of cells through the cell cycle, which will also be considered below. In evaluating the distinct steps leading to checkpoint initiation and maintenance, the phenomenon of checkpoint escape should be clarified. Here, we define this to represent cells, which most likely due to their position in the cell cycle, fail to undergo checkpoint arrest – that is, they escape arrest. We distinguish this phenomenon from cells which initiate arrest but are subsequently released earlier than anticipated, which we describe as premature release. Cells that “escape” checkpoint arrest may be significant in the analysis of mitotic breakage when the time of sample collection is within the period of checkpoint arrest. Thus, although the majority of cells remain checkpoint arrested, a few “escape” arrest and enter mitosis. Since the results of most cytogenetic analysis are expressed as chromosome breaks per mitotic cell, the analysis can be highly dependent upon the small number of mitotic cells which have “escaped” checkpoint arrest rather than the majority of the population that has undergone checkpoint arrest (Lobrich and Jeggo 2007). Finally, checkpoint adaptation is a distinct process which was originally described in yeast and, more recently, in mammalian cells (Toczyski et al. 1997; Pellicioli et al. 2001; Yoo et al. 2004; Syljuasen et al. 2006). This represents a regulated process, whereby the checkpoint is released in cells that have endured prolonged arrest despite multiple persisting DSBs. Rather than remain arrested, cells appear to make a last ditch attempt to progress through the cell cycle albeit with the potential for increased chromosome breakage.

3.4 Mechanism Underlying DNA Damage-Induced G2/M Checkpoint Arrest The G2/M checkpoint has, arguably, been the best studied of the checkpoints, in part because the process is relatively well conserved between yeast and mammalian cells. Further, unlike the G1/S checkpoint, the process can be readily studied in immortalized or tumor cell lines. The G2/M checkpoint prevents the mitotic progression of cells which incurred DNA damage in G2 phase or which progressed into G2 phase with DNA damage. Progression from G2 into mitosis is regulated primarily by Cyclin B1/Cdk1 kinase. Cdk1 activity is itself regulated by inhibitory Tyr15 and Thr14 phosphorylation by the Wee1 and Myt1 kinases, respectively (Booher


61

et al. 1997). Activation of Cdk1 requires dephosphorylation of Thr14 and Tyr15 by the Cdc25 phosphatases. Three Cdc25 phosphatases, A, B, and C, have been described in mammalian cells and their interplay is still poorly understood. Cdc25A and C are important as mitotic regulators with Cdc25B triggering the initial process. A critical step in damage-induced checkpoint arrest is the inactivation of the Cdc25 phosphatases, which is regulated by the Chk1/Chk2 kinases (Matsuoka et al. 1998). Cdc25A and C are phosphorylation targets of the Chk1/2 kinases. Cdc25A is phosphorylated by the transducer kinases at multiple sites and the regulation of its activity by phosphorylation is complex. Phosphorylation can directly inhibit Cdc25A activity and its proteasome-dependent degradation (Mailand et al. 2000; Molinari et al. 2000; Falck et al. 2001). However, Cdc25A is also phosphorylated by cyclin B-Cdk1 in mitosis, which results in its stabilization rather than degradation, a step which appears to consolidate mitotic entry (Mailand et al. 2002).

3.4.1 The Initiation of G2/M Checkpoint Arrest Early checkpoint arrest, representing the sharp reduction in mitotic entry within 1–2 h of IR exposure, must arise in irradiated G2 phase cells since primary and transformed cells take ~5 h to progress from S into G2 phase. As discussed above, IR exposure of G2 phase cells initially only activates ATM, since IR directly induces DSBs and SSBs but not regions of ss DNA, the lesion activating ATR. In G2 phase, DSB repair can occur by DNA nonhomologous end-joining (NHEJ) or by homologous recombination (HR). Although a widely accepted dogma is that HR represents the major DSB repair pathway in G2 phase while NHEJ predominates in G1 phase, recent studies have shown that ~80% of DSBs are repaired with fast kinetics by NHEJ in G2 phase while HR functions to repair those DSBs (approximately 20% of IR-induced DSBs) which are repaired with slow kinetics (Beucher et al. 2009). Making the assumption that the DSBs rejoined by NHEJ do not undergo resection but activate the ATM-Chk2 arm of the signaling response, the finding argues that ATM-Chk2 signaling must play a major role in the initiation of checkpoint arrest. Resection at the slowly repaired DSBs in G2 phase is ATM dependent but leads to ATR activation at the ss DNA regions generated (Jazayeri et al. 2006). Thus, signaling at resected DSBs occurs via ATM-ATRChk1. Recent in vitro studies have suggested that as resection ensues at DSBs, there is a switch from ATM to ATR signaling since ATM activity becomes diminished as the length of the ss DNA tail increases (Rhind 2009; Shiotani and Zou 2009). It is important to appreciate, however, that although the completion of DSB repair by HR appears to represent a slow process, resection occurs relatively rapidly after IR exposure. Indeed, using RPA foci formation to assess resection, we have observed that maximal numbers of RPA (and Rad51) foci can be observed by 30 min post-IR (earlier times are difficult to evaluate using this method of assessing resection) (Shibata, manuscript submitted). Consistent with this notion, pChk1 levels in G2 phase cells are also at a maximum level at 30–60 min post-IR

62

M. Lobrich et al.

(Shibata et al. 2010). Thus, for cells in G2 at the time of irradiation, we have observed that both ATM-Chk2 and ATM-ATR-Chk1 contribute to the initiation of checkpoint arrest (Shibata et al. 2010). In summary, early checkpoint arrest in G2 phase cells is ATM dependent, but Chk1/Chk2 have overlapping, redundant functions (Fig. 3.2). Indeed, the lack of dependency of G2/M checkpoint initiation upon Chk2 has been reported previously (Hirao et al. 2002). It might be expected, however, that ATM-Chk2 have the most significant impact since the majority of DSBs are repaired without resection by NHEJ. Such a dependency might be detectable at lower doses when the number of DSBs undergoing resection becomes small (and potentially less than one per cell), if Chk1 is only activated at 20% of IR-induced DSBs. Additionally, the significance of Chk1 vs. Chk2 in the initiation of G2/M checkpoint arrest may vary between cell types and may be distinct in tumor cells. One factor influencing this may be the rapidity with which DSBs undergo resection. Interestingly, checkpoint arrest in DT40 chicken cells is Chk1 dependent after 4 Gy, a finding consistent with the suggestion that more DSBs undergo resection and repair by HR in G2 in DT40 compared to mammalian cells (Sonoda et al. 2006; Rainey et al. 2008).

Fig. 3.2 Initiation and maintenance of G2/M checkpoint arrest in irradiated G2 phase cells. Irradiated G2 phase cells initially activate ATM and Chk2 from directly induced DSBs. However, ATM-dependent resection at DSBS can occur at early times after IR exposure in G2 cells leading to ATR-Chk1 activation. Both processes can contribute to the initiation of checkpoint arrest. Recent studies have shown that homologous recombination repairs the DSBs that are located at heterochromatin and are repaired with slow kinetics. Thus, resected DSBs represent a greater fraction of unrepaired DSBs at later times post-IR and ATR-Chk1 signaling makes a greater contribution to the maintenance of checkpoint arrest


63

3.4.2 The Maintenance of G2/M Checkpoint Arrest Studies have demonstrated that the duration of checkpoint arrest is dependent upon dose and DSB repair capacity with DSB repair defective cells undergoing prolonged arrest compared to repair proficient cells (Wang et al. 2002b; Sturgeon et al. 2006; Deckbar et al. 2007). This demonstrates that G2/M checkpoint arrest is not activated for a defined period of time but instead supports a model by which the checkpoint machinery responds to the status of DSB repair. Thus, it is likely that DSBs are continuously sensed and Chk1/2 continuously activated. The maintenance of checkpoint arrest involves two distinct classes of cells. First, checkpoint arrest must be maintained in irradiated G2 phase cells. Since, the prevention of mitotic entry can be sustained for prolonged times after higher doses or in cells lacking DSB repair capacity, the checkpoint signal must be capable of being maintained in G2 phase cells for prolonged times. Additionally, at longer times after IR exposure (>5 h), irradiated and potentially damaged S phase cells progress into G2 phase. Such cells can also undergo G2/M arrest for prolonged periods (Krempler et al. 2007; Fernet et al. 2009). Interestingly, these findings suggest that the intraS-phase checkpoint arrest process does not efficiently prevent irradiated S phase cells progressing into G2 phase and that the G2/M checkpoint plays a significant role in preventing progression of such cells. However, the efficacy of the S phase checkpoint has not been carefully evaluated and will not be discussed in detail in this review. As discussed above, ATR can be activated in an ATM-independent manner in S phase cells after IR exposure as a consequence of replication fork arrest. Consistent with this notion, G2/M checkpoint arrest at later times after IR occurs by a molecularly distinct process to the early G2/M arrest being ATR-Chk1 dependent and ATM independent (Beamish et al. 1994; Xu et al. 2002a; Wang et al. 2003; Fernet et al. 2009). In recent studies, we have employed aphidicolin, an inhibitor of the replicative polymerase, to prevent S phase cells progressing into G2 phase postirradiation to allow an analysis of the maintenance of checkpoint arrest and DSB repair in irradiated G2 phase cells (Beucher et al. 2009; Shibata, manuscript submitted). Since HR repairs the slow DSB repair component, it is likely that, at later times postirradiation, the persisting DSBs will be those that have undergone resection, and hence ATR-Chk1 activation will be important for the maintenance of checkpoint arrest. Consistent with this notion, we observed that loss of either ATR or Chk1 results in premature mitotic entry of irradiated G2 phase cells (Shibata et al. 2010). However, additionally, ATM-Chk2 signaling contributes to the maintenance of checkpoint arrest, although less significantly compared with the role of Chk1 (Fig. 3.2). Thus, it appears likely that as DSB repair ensues and the checkpoint signal nears a threshold level required to maintain checkpoint arrest, both Chk1 and Chk2 provide some contribution. This starkly contrasts to the redundant and overlapping functions in initiating checkpoint arrest likely due to either activity alone being sufficient to reach a threshold signal (after 3 Gy IR).

64

M. Lobrich et al.

3.4.3 Sensitivity of the G2/M Checkpoint An important question in evaluating the efficacy of the G2/M checkpoint is its ability to respond to low levels of DNA damage. A simple and possibly the anticipated prediction would be that G2/M checkpoint arrest would be triggered by a single DSB. This is a reasonable expectation given that the checkpoint arrest has evolved to prevent genomic instability arising from the progression of cells with DSBs through mitosis. However, several studies have now provided evidence that the G2/M checkpoint is not sensitive to a single DSB but rather has a defined threshold of 10–20 DSBs (Deckbar et al. 2007; Fernet et al. 2009). In one approach, we evaluated the duration of checkpoint arrest with the status of DSB repair, using gH2AX to monitor the progress of DSB repair. We observed that cells enter mitosis after 3 Gy when approximately 15–20 gH2AX foci remain. Although the duration of arrest changes in repair proficient vs. deficient cells, the number of gH2AX foci remaining at the point of mitotic entry was similar, providing suggestive evidence that there is a defined sensitivity threshold (Deckbar et al. 2007). One limitation of this approach is that there may be a lag between completion of DSB repair and loss of gH2AX foci. Indeed, this is a possibility suggested by the finding that although the kinetics of DSB repair monitored by pulsed field gel electrophoresis (PFGE) is similar to that monitored by gH2AX foci enumeration, the repair lags 1–2 h when assessed by gH2AX foci enumeration compared to the use of PFGE. However, even 2 h after commencing mitotic entry, gH2AX foci remain detectable in the remaining G2 phase cells. Further supporting the notion that mitotic entry occurs prior to the completion of DSB repair, we also observed that mitotic cells derived from G2 phase cells that were released from G2/M arrest harbor 1–2 chromosome breaks (Deckbar et al. 2007). We also assessed chromosome breakage in G2 phase cells using calyculin A to induce premature chromosome condensation (PCC) of G2 phase cells. At the time of mitotic entry, we observed ~3 PCC chromosome breaks per G2 phase cell. Thus, mitotic entry occurs when ~15 gH2AX foci and 3 PCC breaks remain in G2 phase cells and gives rise to 1–2 chromosome breaks in the cells that enter mitosis after checkpoint arrest. Given that gH2AX foci detect nearly all DSBs, while a smaller subset of DSBs may be manifest as PCC breaks and an even smaller subset are visualized as chromosome breaks, this ratio appears reasonable. Importantly, since the majority of the G2 phase population undergoes G2/M arrest, our findings show that the major contribution to chromosome breakage arises in cells that undergo checkpoint arrest (Fig. 3.3). This contribution is more significant than the higher number of chromosome breaks observed in cells that “escape” checkpoint arrest, since the latter represents just a tiny percentage of the irradiated G2 population (Lobrich and Jeggo 2007). Collectively, these results provide strong evidence that the G2/M checkpoint has a defined threshold that allows for a low level of chromosome breakage in cells that are released from arrest. The ATMdependent G2/M checkpoint is therefore inefficient and lacks sensitivity. Interestingly, a recent study has suggested that cells that traverse S phase after irradiation have a lower threshold sensitivity compared to cells that were in G2 at the time of irradiation (Fernet et al. 2009).


65

Fig. 3.3 Estimation of the kinetics for total chromosome breakage considering the level of chromosome breaks per mitotic cell and the number of cells in mitosis. The number of mitotic cells assessed by phosphoH3 analysis at various times post 1 Gy IR (upper left panel) was multiplied with the number of chromosome breaks per mitotic cell (upper right panel) providing an estimation of the total number of mitotic breaks (total chromosome breakage). Artemis cells harbor more chromosome breaks per mitotic cell than wt cells at any time point. Strikingly, however, both Artemis and wt cells harbor a similar level of chromosome breaks per mitotic cell at times when the checkpoint is released (which is delayed in Artemis cells). Hence, the total number of chromosome breaks is only marginally higher in Artemis than in wt cells although they arise with delayed kinetics in Artemis cells. This demonstrates the efficiency of checkpoint arrest in limiting the impact of a repair defect, i.e., the co-operative interplay between checkpoint and repair function. Because of the lack of checkpoint arrest, A-T cells display elevated numbers of chromosome breaks early after irradiation which then decreases because of repair and depletion of the G2 population. The total number of chromosome breaks in a population of irradiated G2 A-T cells is several-fold higher than in normal and Artemis cells (see Deckbar et al. 2007 for details of this analysis)

A consideration of these issues raises the interesting question of how signaling is maintained as cells initiate the process of HR. The available evidence suggests that NHEJ and ATM signaling can function together at DSBs – thus NHEJ is largely unaffected by the absence of ATM and similarly, ATM signaling is activated

66

M. Lobrich et al.

normally at DSBs in the absence of NHEJ proteins (although arrest is maintained for a prolonged period due to persisting unrepaired DSBs). This suggests that ATM can be recruited and remain active at non-resected DSBs despite ongoing NHEJ. Following resection at a DSB, the ss DNA generated is rapidly bound by RPA, initiating the recruitment and activation of ATR and Chk1 as described above. However, this process is rapidly followed by the replacement of RPA-coated ss DNA with Rad51. This raises the important question of how signaling to Chk1 is maintained at DSBs undergoing HR. Interestingly, studies in yeast, where most DSB repair is carried out by HR, show that Crb2, the yeast homologue of 53BP1, is required for the maintenance of checkpoint arrest raising the possibility that the mediator proteins help maintain damage response signaling as the process of HR proceeds (Nakamura et al. 2005). Additionally, HR proteins themselves have also been suggested to help sustain checkpoint arrest (Badie et al. 2009). A further possibility is that the lack of efficient ATM and ATR signaling to DSBs during later stages of HR contributes to the insensitivity of G2/M checkpoint maintenance. The discussion above reflects the sensitivity of the G2/M checkpoint in maintaining arrest once initiated. A related, and arguably more important, question is the number of DSBs required to activate the G2/M checkpoint. Substantial evidence has now shown that for most primary cell lines, the G2/M checkpoint arrest is not efficiently activated following exposure to doses below 0.5 Gy (Marples et al. 2004; Deckbar et al. 2007; Fernet et al. 2009). Although it is difficult to distinguish transient checkpoint activation from the complete lack of checkpoint activation, we have not observed any significant perturbation in primary human fibroblasts exposed to 0.25 Gy X or g-rays. Assuming that 1 Gy X-rays induce ~25 DSBs in G1 phase (i.e., 50 DSBs in G2 phase cells), this suggests that the introduction of 12–15 DSBs does not activate G2/M phase checkpoint arrest – a number of DSBs remarkably similar to the threshold at which checkpoint arrest is not maintained (Kegel et al. 2007). Interestingly, the phenomenon of low dose radiation hypersensitivity has also been attributed to a failure of low dose radiation exposure to activate G2/M checkpoint arrest (Marples et al. 2004; Fernet et al. 2009).

3.4.4 Role of Damage Response Mediator Proteins in G2/M Checkpoint Arrest Perhaps surprisingly, although the DNA damage mediator proteins have frequently been described as “checkpoint” proteins, and, indeed MDC1 derives its name from this phenotype, loss of mediator protein function confers only a modest defect in checkpoint arrest, only being observed following exposure to low radiation doses (Fernandez-Capetillo et al. 2002; Wang et al. 2002a; Stewart et al. 2003; Lou et al. 2006). It has, therefore, been argued that the mediator proteins, H2AX, MDC1, and 53BP1, function primarily to amplify ATM signaling with their role only being detectable at low doses when amplification is required to generate a sufficiently strong signal to activate the checkpoint arrest machinery (Fernandez-Capetillo et al.


67

2002). A possible model based on current understanding of DDR signaling is that this is achieved by their ability to enhance the retention of ATM at the DSB (Noon et al. 2010). In this context, it is also significant that recent in vitro studies suggest that 53BP1 enhances ATM activation when the MRN concentration is low (Lee et al. 2009). Strikingly, our recent studies suggest that the mediator proteins are important for maintaining checkpoint arrest following exposure to high radiation doses (Shibata et al. 2010). Significantly, this demonstrates a potential value of their “amplification” role after exposure to high doses of IR, a role which appears to be significant for the maintenance of genomic stability (see below). Interestingly, ATM and the mediator proteins have recently been shown to be required for the slow component of DSB repair, which represents the repair of DSBs located at regions of heterochromatin (Goodarzi et al. 2008; Noon et al. 2010). Further, their role involves the concentrated, localized phosphorylation of KAP-1, a heterochromatic building factor (Noon et al. 2010). Thus it is possible that KAP-1 phosphorylation is also important for signaling from DSBs located at regions of heterochromatin and that this represents the “amplification” step. Interestingly, previous studies have suggested that chromatin compaction limits the strength of the DNA damage signaling response and that an enhanced response is observed when histone H1 levels are diminished (Murga et al. 2007). Indirect evidence suggests that heterochromatin may also create a barrier to ATM signal expansion, which is, at least partially, relieved by ATM-dependent phosphorylation of Kap1 (Jeggo, unpublished findings). This may be important in maintaining checkpoint arrest since heterochromatic DSBs are repaired slowly.

3.5 G1/S Arrest In mammalian cells, there is strong evidence for a process of G1/S phase arrest that is dependent upon p53 (Kastan et al. 1992). This process, which does not appear to be present in yeast, appears to be particularly important in multicellular organisms in preventing the proliferation of damaged cells and, thereby, represents an important step in cancer avoidance (Kastan and Bartek 2004). Additionally, there is evidence for another process that prevents G1/S progression that functions via ATM-Chk2. We will first describe p53-dependent G1/S arrest, then discuss evidence for a second process more akin to the mechanism underlying G2/M checkpoint arrest.

3.5.1 p53 Dependent G1/S Arrest p53 is subject to complex regulation, which occurs at both the transcriptional and post-translational level. p53 stability is regulated predominantly by the ubiquitin ligase, MDM2, which binds to the N-terminal region of p53, ubiquitylates the C terminus and targets it for proteasome-dependent degradation, thereby ensuring low levels of p53 in undamaged cells (Khosravi et al. 1999). ATM, ATR, and the transducer kinase, Chk2, can phosphorylate both MDM2 and p53 (Kastan et al. 1992;

68

M. Lobrich et al.

Hirao et al. 2000; Bartek and Lukas 2001b). Phosphorylation of p53 or MDM2 diminishes MDM2 binding to p53, leading to decreased p53 ubiquitylation and consequently increased p53 stability. These processes additionally impact upon p53 transcriptional activity. p53 is a transcriptional regulator of proteins that harbor p53 binding motifs. The Cdk inhibitor, p21waf1/Cip1, is the most significant p53 regulated gene in the context of G1/S checkpoint arrest (Harper et al. 1993; Sherr and Roberts 1999; Bartek and Lukas 2001b). However, p53 also regulates MDM2, providing a feedback, autoregulatory mechanism. Thus, following DNA damage, p53 is rapidly activated by post-translational modifications, leading to enhanced levels of p21, which occurs more slowly since it requires transcriptional activation. Entry into S phase is regulated by phosphorylation of Rb protein, which in turn is controlled by cyclindependent kinases (Sherr 2000). p21, a Cdk2 inhibitor, prevents Cdk2 phosphory lation of Rb and S phase entry (Sherr and Roberts 1999; Bartek and Lukas 2001b). Previous studies have suggested that the G1/S checkpoint is highly sensitive and is activated by a single DSB (Di Leonardo et al. 1994; Huang et al. 1996; Linke et al. 1997). In a recent study, we examined the sensitivity of the G1/S checkpoint and found that it was transiently activated by doses less than 100 mGy, suggesting that it is, indeed, highly sensitive (Deckbar et al. 2010). We observed that irradiated G2 cells that were released from G2/M checkpoint arrest were subsequently arrested at the G1/S checkpoint, suggesting that the G1/S checkpoint provides a barrier to counteract the inefficiency of the G2/M checkpoint. However, we revealed a significant limitation of the G1/S checkpoint, namely that full arrest did not occur for >4 h post-IR. We attribute this feature to the time required to fully activate p21 since transcriptional activation is a slower process than post-translational modification. It should also be mentioned that although p53 is a substrate for both ATM and ATR, after IR exposure G1/S checkpoint arrest is ATM-dependent, most likely because ATR is not activated in G0/G1 cells as discussed above (Lavin 2008).

3.5.2 A Second Process Inhibiting S Phase Entry After Radiation Exposure Previous studies have shown that, in a similar manner to the G2/M checkpoint, the Cdc25 phosphatases can regulate CyclinE-Cdk2 activity in G1 phase and that DNA damage results in the rapid destruction of Cdc25A via ubiquitylation and proteasome mediated degradation (Mailand et al. 2000). These findings, therefore, provide strong evidence for the existence of another process regulating G1/S entry (Bartek and Lukas 2001a). This process has been proposed to represent an early response, thereby providing a dual wave of G1/S regulation; an early Cdc25A-dependent process and a more slowly activated p53-dependent process (Bartek and Lukas 2001a). Our analysis of G1/S entry provided findings consistent with this (Fig. 3.4). Following IR exposure, we did not observe complete blockage to S phase entry until >4 h post-IR treatment but rather a slowing in the rate of S phase entry. Such slowing was ATM-Chk2 dependent and Chk1 independent demonstrating that it represents a regulated DNA damage response. Since p53 enhances S phase entry even in undamaged cells, it was unclear


69

Fig. 3.4 Two processes delay entry from G1 into S phase following IR exposure. DSBs activated in G1 phase lead to ATM activation. ATM phosphorylation of Chk2 leads to Cdc25 activation and causes a delay but not complete arrest of entry into S phase. ATM also phosphorylates p53, which activates p21. This process causes a complete block in S phase entry. However, this process requires transcriptional activation and thus leads to a block in S phase entry that is only observed >4 h postirradiation. Thus, at early times post-IR, a delay in S phase entry is observed following by a defined blockage at slightly later times

whether this process is p53 dependent. Taken together with the previous analysis of the impact of IR exposure on Cdc25 stability, our findings suggest that this represents an initial process that regulates G1/S entry but does not confer complete blockage (Mailand et al. 2000). At >4 h after IR exposure, ATM activates p53-dependent G1/S checkpoint arrest via p21 activation. Although the p53-dependent process has a low threshold of sensitivity, i.e., it is activated by 1–3 DSBs, it is activated slowly allowing damage cells to enter S phase at early times after IR exposure.

3.5.3 Maintenance of G1/S Checkpoint Arrest Previous studies have suggested that p53-dependent checkpoint arrest permanently eliminates damaged cells rather than providing a transient delay to enhance the time for DSB repair (Di Leonardo et al. 1994; Wahl et al. 1997). However, in our recent study, we observed that cells can reenter S phase following full G1/S checkpoint activation (Deckbar et al. 2010). Indeed, given the fact that primary fibroblasts are predominantly in G0/G1 phase and that G1/S phase is activated in all cells even following low dose exposure, the ability of these cells to survive IR demonstrates that

70

M. Lobrich et al.

they must be released in a timely manner following checkpoint activation. Indeed, we observed that following exposure to low IR doses (4 Gy, we observed that a low number of cells (~5%) can enter S phase at later times even though DSB repair is not yet completed. The rate of S phase entry is slower than observed in untreated cells, suggesting that it does not represent release of the entire G1 population but rather the failure of a few cells to remain arrested. Analysis of DNA DSB levels assessed by gH2AX foci enumeration and by PCC formation provided strong evidence that DSB repair is not completed when the cells enter S phase. Thus, we suggest that the G1/S checkpoint even if fully activated also has limitations that can affect its maintenance and hence its efficacy for preventing genomic instability (Fig. 3.5).

Fig. 3.5 The limitations of G1/S and G2/M checkpoint arrest. At early times post-IR, G1/S checkpoint arrest slows but does not abolish S phase entry allowing cells to enter S phase. By 6 h postirradiation, full arrest of S phase entry is observed following exposure to doses >1 Gy. However, G1/S arrest is not efficiently maintained and slow release of cells with DSBs can occur. Such cells can progress into G2 phase with unrepaired DSBs. The G2/M checkpoint arrest is initiated within 1 h of IR exposure. However, the G2/M checkpoint has a defined threshold of sensitivity and cells are released into mitosis when they harbor 10–20 DSBs. This can generate mitotic cells with DSBs. The hatched regions represent periods when cells can enter S phase or mitosis with DSBs


71

3.6 Intra-S-Phase Checkpoint Arrest Studies in yeast and mammalian cells have also demonstrated that there is a process(es) that monitors replication fork integrity in S phase and functions to stabilize replication forks following their arrest (Branzei and Foiani 2009; Segurado and Tercero 2009). Additionally, DNA damage in S phase can lead to transient arrest of origin firing (Caspari and Carr 1999). These processes will only be discussed briefly. An extremely early characterized feature of cell lines derived from ataxia telangiectasia patients, a human disorder caused by mutations in ATM, was a phenomenon called radioresistant DNA synthesis, now known to be due to a failure to elicit S phase checkpoint arrest (Painter and Young 1980). Thus, while control cells show a rapid and pronounced decrease in DNA replication, normally monitored by thymidine incorporation, at 1–4 h postirradiation, A-T cells continue to traverse S phase at a rate similar to that observed in untreated cells. This is now known to represent ATM-dependent inhibition of Cdc25A phosphorylation that likely occurs via two independent pathways that involve NBS1 and Chk2 (Falck et al. 2001; Falck et al. 2002). This process, at least in part, involves inhibition of Cdk2-dependent loading of Cdc45 onto replication origins (Falck et al. 2002). At later times (>4 h) post-IR, S phase arrest is ATM-independent but ATR-Chk1 dependent, a process considered to represent the inhibition of fork elongation and/ or origin firing via an ATR-dependent mechanism most likely from ss regions of DNA generated at stalled replication forks (Zhou et al. 2002). Although not studied in detail, it is likely that the intra-S-phase checkpoint also has limitations. This is perhaps most convincingly demonstrated by the observation that S phase cells can progress to G2 phase following high dose exposure (e.g. 6 Gy), even in cells with impaired DSB repair (Krempler et al. 2007).

3.7 Significance of Cell Cycle Checkpoint Arrest The DDR functions to achieve two distinct goals, enhancing survival and maintaining genomic stability. Although these two end points are related they are distinct. Indeed, the maintenance of genomic stability may necessitate decreasing survival to achieve the elimination of damaged cells and enhancing survival by increasing genomic instability is an important step in the etiology of carcinogenesis. Checkpoint arrest likely enhances survival by allowing more time for DSB repair. However, although repair can progress in the subsequent cell cycle stage, the accuracy of repair will be comprised. Thus, for example, progression through mitosis and cytokinesis in the presence of DSBs may result in loss of acentric fragments diminishing the opportunity to rejoin the correct DNA ends. Further, replication past a DSB enhances the opportunity for translocation events. The ability to repair DSBs makes the greatest impact on survival post-IR since unrepaired DSBs are particularly lethal lesions, and cells lacking DSB repair mechanisms, e.g., NHEJdeficient cell lines, are dramatically radiation sensitive. Although there are multiple

72

M. Lobrich et al.

routes for repairing DSBs including telomere fusion, end capping, and translocation, most of them utilize NHEJ proteins. Thus, mice and patients deficient in NHEJ proteins show dramatic radiation sensitivity but only modestly increased carcinogenesis. This is perhaps best demonstrated by SCID mice which show dramatic radiation sensitivity but little elevated radiation-induced tumorigenesis (Bosma and Carroll 1991). In short, diminished DSB repair dramatically enhances radiation sensitivity with only modestly enhanced instability. In contrast, checkpoint arrest most likely exerts a significant impact on genomic instability compared to survival. The phenomenon of low dose radiation hypersensitivity, however, represents the enhanced sensitivity to low radiation doses. Recent evidence has consolidated the model that this phenomenon arises as a consequence of a failure to activate G2/M checkpoint arrest following low doses strongly suggesting that checkpoint arrest can enhance survival (Marples et al. 2004; Marples and Collis 2008; Fernet et al. 2009). The overlapping roles of Chk1 and Chk2 in checkpoint arrest and the fact that Chk1 is essential makes it difficult to assess the impact of a checkpoint defect. However, loss of p53, which abolishes the late G1/S checkpoint results in enhanced survival and genomic instability, consistent with the notion that checkpoint arrest might exert a greater impact on the maintenance of genomic stability. However, p53 also regulates apoptosis limiting defined conclusions. The loss of ATM confers loss of checkpoint arrest coupled with comprised DSB repair (Lobrich and Jeggo 2005). Artemis defective cells harbor the same DSB repair defect as ATM defective cells but are checkpoint proficient. We previously compared chromosome breakage arising from irradiated G2 phase cells in A-T vs. Artemis null cells (Deckbar et al. 2007). Strikingly, we observed that although Artemis null cells showed only modestly enhanced chromosome breakage in mitosis, demonstrating the efficacy of the G2/M checkpoint in preventing the progression of damaged cells, loss of ATM conferred a high level of chromosome breakage. Use of UCN01, a checkpoint inhibitor, to abolish checkpoint arrest, conferred a small defect similar to that observed in Artemis-defective cells. Thus, we suggest that loss of either checkpoint arrest or a DSB repair defect enhances chromosome breakage but loss of both is more than additive (Deckbar et al. 2007; Lobrich and Jeggo 2007). This likely explains the dramatically increased cancer predisposition and genomic instability observed in A-T.

3.8 Conclusions The activation of cell cycle checkpoint arrest is a critical end point of DNA damage response signaling that plays an important role in limiting genomic instability and also contributes to enhancing survival following radiation exposure. Two distinct processes that arrest (or slow) entry from G1 into S phase have been described (Fig. 3.4). Intra-S-phase arrest slows progression through S phase and G2/M arrest prevents entry into mitosis. Radiation exposure induces DSBs, SSBs, and base damage, but only DSBs directly activate damage response signaling via ATM.


73

However, ATR can be activated following replication fork stalling or collapse in S phase or following resection in S/G2 phase. This provides a complex signaling response that differs with cell cycle phase (Fig. 3.1). Cell cycle checkpoint arrest is a sensitive process yet has defined limitations (Fig. 3.5). An evaluation of these limitations is important for considering the impact of radiation on genomic stability and survival. Further understanding of the processes can enable them to be exploited to enhance the therapeutic potential of radiotherapy.

References Ahn JY, Li X, Davis HL et al (2002) Phosphorylation of threonine 68 promotes oligomerization and autophosphorylation of the Chk2 protein kinase via the forkhead-associated domain. J Biol Chem 277:19389–19395 Badie S, Liao C, Thanasoula M et al (2009) RAD51C facilitates checkpoint signaling by promoting CHK2 phosphorylation. J Cell Biol 185:587–600 Bakkenist CJ, Kastan MB (2003) DNA damage activates ATM through intermolecular autophosphorylation and dimer dissociation. Nature 421:499–506 Ball HL, Ehrhardt MR, Mordes DA et al (2007) Function of a conserved checkpoint recruitment domain in ATRIP proteins. Mol Cell Biol 27:3367–3377 Bartek J, Lukas J (2001a) Mammalian G1- and S-phase checkpoints in response to DNA damage. Curr Opin Cell Biol 13:738–747 Bartek J, Lukas J (2001b) Pathways governing G1/S transition and their response to DNA damage. FEBS Lett 490:117–122 Bartek J, Lukas J (2003) Chk1 and Chk2 kinases in checkpoint control and cancer. Cancer Cell 3:421–429 Beamish H, Khanna KK, Lavin MF (1994) Ionizing radiation and cell cycle progression in ataxia telangiectasia. Radiat Res 138:S130–133 Bekker-Jensen S, Danielsen JR, Fugger K et al (2010) HERC2 coordinates ubiquitin-dependent assembly of DNA repair factors on damaged chromosomes. Nat Cell Biol 12:80–86 Beucher A, Birraux J, Tchouandong L et al (2009) ATM and Artemis promote homologous recombination of radiation-induced DNA double-strand breaks in G2. EMBO J 28:3413–3427 Booher RN, Holman PS, Fattaey A (1997) Human Myt1 is a cell cycle-regulated kinase that inhibits Cdc2 but not Cdk2 activity. J Biol Chem 272:22300–22306 Bosma MJ, Carroll AM (1991) The SCID mouse mutant: definition, characterization, and potential uses. Annu Rev Immunol 9:323–350 Branzei D, Foiani M (2009) The checkpoint response to replication stress. DNA Repair (Amst) 8:1038–1046 Byun TS, Pacek M, Yee MC et al (2005) Functional uncoupling of MCM helicase and DNA polymerase activities activates the ATR-dependent checkpoint. Genes Dev 19:1040–1052 Carson CT, Schwartz RA, Stracker TH et al (2003) The Mre11 complex is required for ATM activation and the G2/M checkpoint. EMBO J 22:6610–6620 Caspari T, Carr AM (1999) DNA structure checkpoint pathways in Schizosaccharomyces pombe. Biochimie 81:173–181 Chapman JR, Jackson SP (2008) Phospho-dependent interactions between NBS1 and MDC1 mediate chromatin retention of the MRN complex at sites of DNA damage. EMBO Rep 9:795–801 Chaturvedi P, Eng WK, Zhu Y et al (1999) Mammalian Chk2 is a downstream effector of the ATM-dependent DNA damage checkpoint pathway. Oncogene 18:4047–4054

74

M. Lobrich et al.

Cimprich KA, Cortez D (2008) ATR: an essential regulator of genome integrity. Nat Rev Mol Cell Biol 9:616–627 Cortez D, Guntuku S, Qin J et al (2001) ATR and ATRIP: partners in checkpoint signaling. Science 294:1713–1716 Deckbar D, Birraux J, Krempler A et al (2007) Chromosome breakage after G2 checkpoint release. J Cell Biol 176:748–755 Deckbar D, Stiff T, Koch B et al (2010) The limitations of the G1-S checkpoint. Cancer Res 70: 4412–4421 Delacroix S, Wagner JM, Kobayashi M et al (2007) The Rad9-Hus1-Rad1 (9-1-1) clamp activates checkpoint signaling via TopBP1. Genes Dev 21:1472–1477 Di Leonardo A, Linke SP, Clarkin K et al (1994) DNA damage triggers a prolonged p53-dependent G1 arrest and long-term induction of Cip1 in normal human fibroblasts. Genes Dev 8:2450–2551 Doil C, Mailand N, Bekker-Jensen S et al (2009) RNF168 binds and amplifies ubiquitin conjugates on damaged chromosomes to allow accumulation of repair proteins. Cell 136:435–446 Falck J, Coates J, Jackson SP (2005) Conserved modes of recruitment of ATM, ATR and DNAPKcs to sites of DNA damage. Nature 434:605–611 Falck J, Mailand N, Syljuasen RG et al (2001) The ATM-Chk2-Cdc25A checkpoint pathway guards against radioresistant DNA synthesis. Nature 410:842–847 Falck J, Petrini JH, Williams BR et al (2002) The DNA damage-dependent intra-S phase checkpoint is regulated by parallel pathways. Nat Genet 30:290–294 Fernandez-Capetillo O, Chen HT, Celeste A et al (2002) DNA damage-induced G2-M checkpoint activation by histone H2AX and 53BP1. Nat Cell Biol 4:993–997 Fernet M, Megnin-Chanet F, Hall J et al (2009) Control of the G2/M checkpoints after exposure to low doses of ionising radiation: implications for hyper-radiosensitivity. DNA Repair (Amst) 9:48–57 Furuya F, Poitelea M, Guo L et al (2004) Chk1 activation requires Rad9 S/TQ-site phosphorylation to promote association with C-terminal BRCT domains of Rad4TOPBP1. Genes Dev 18:1154–1164 Goodarzi AA, Noon AT, Deckbar D et al (2008) ATM signaling facilitates repair of DNA doublestrand breaks associated with heterochromatin. Mol Cell 31:167–177 Harper JW, Adami GR, Wei N et al (1993) The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin- dependent kinases. Cell 75:805–816 Helleday T, Lo J, van Gent DC et al (2007) DNA double-strand break repair: from mechanistic understanding to cancer treatment. DNA Repair (Amst) 6:923–935 Hirao A, Cheung A, Duncan G et al (2002) Chk2 is a tumour suppressor that regulates apoptosis in both an ataxia telangiectasia mutad (ATM)-dependent and an ATM-independent manner. Mol Cell Biol 22:6521–6532 Hirao A, Kong YY, Matsuoka S et al (2000) DNA damage-induced activation of p53 by the checkpoint kinase Chk2. Science 287:1824–1827 Huang LC, Clarkin KC, Wahl GM (1996) Sensitivity and selectivity of the DNA damage sensor responsible for activating p53-dependent G1 arrest. Proc Natl Acad Sci USA 93:4827–4832 Huen MS, Grant R, Manke I et al (2007) RNF8 transduces the DNA-damage signal via histone ubiquitylation and checkpoint protein assembly. Cell 131:901–914 Jazayeri A, Falck J, Lukas C et al (2006) ATM- and cell cycle-dependent regulation of ATR in response to DNA double-strand breaks. Nat Cell Biol 8:37–45 Kastan MB, Bartek J (2004) Cell-cycle checkpoints and cancer. Nature 432:316–323 Kastan MB, Zhan Q, El-Deiry WS et al (1992) A mammalian cell cycle checkpoint pathway utilizing p53 and GADD45 is defective in ataxia-telangiectasia. Cell 71:587–597 Kegel P, Riballo E, Kuhne M et al (2007) X-irradiation of cells on glass slides has a dose doubling impact. DNA Repair (Amst) 6:1692–1697 Khosravi R, Maya R, Gottlieb T et al (1999) Rapid ATM-dependent phosphorylation of MDM2 precedes p53 accumulation in response to DNA damage. Proc Natl Acad Sci USA 96:14973–14977


75

Kolas NK, Chapman JR, Nakada S et al (2007) Orchestration of the DNA-damage response by the RNF8 ubiquitin ligase. Science 318:1637–1640 Krempler A, Deckbar D, Jeggo PA et al (2007) An imperfect G(2)/M checkpoint contributes to chromosome instability following irradiation of S and G(2) phase cells. Cell Cycle 6:1682–1686 Kumagai A, Lee J, Yoo HY et al (2006) TopBP1 activates the ATR–ATRIP complex. Cell 124:943–955 Lavin MF (2008) Ataxia-telangiectasia: from a rare disorder to a paradigm for cell signalling and cancer. Nat Rev Mol Cell Biol 9:759–769 Lavin MF, Khanna KK, Beamish H et al (1994) Defect in radiation signal transduction in ataxiatelangiectasia. Int J Radiat Biol 66:S151–S156 Lee JH, Goodarzi AA, Jeggo PA et al (2009) 53BP1 promotes ATM activity through direct interactions with the MRN complex. EMBO J 29:574–585 Lee J, Kumagai A, Dunphy WG (2007) The Rad9-Hus1-Rad1 checkpoint clamp regulates interaction of TopBP1 with ATR. J Biol Chem 282:28036–28044 Linke SP, Clarkin KC, Wahl GM (1997) p53 mediates permanent arrest over multiple cell cycles in response to gamma-irradiation. Cancer Res 57:1171–1179 Liu Q, Guntuku S, Cui XS et al (2000) Chk1 is an essential kinase that is regulated by Atr and required for the G(2)/M DNA damage checkpoint. Genes Dev 14:1448–1459 Lobrich M, Jeggo PA (2005) Harmonising the response to DSBs: a new string in the ATM bow. DNA Repair (Amst) 4:749–759 Lobrich M, Jeggo PA (2007) The impact of a negligent G2/M checkpoint on genomic instability and cancer induction. Nat Rev Cancer 7:861–869 Lou Z, Minter-Dykhouse K, Franco S et al (2006) MDC1 maintains genomic stability by participating in the amplification of ATM-dependent DNA damage signals. Mol Cell 21:187–200 Lukas C, Bartkova J, Latella L et al (2001) DNA damage-activated kinase Chk2 is independent of proliferation or differentiation yet correlates with tissue biology. Cancer Res 61: 4990–4993 Mailand N, Bekker-Jensen S, Faustrup H et al (2007) RNF8 ubiquitylates histones at DNA doublestrand breaks and promotes assembly of repair proteins. Cell 131:887–900 Mailand N, Falck J, Lukas C et al (2000) Rapid destruction of human Cdc25A in response to DNA damage. Science 288:1425–1429 Mailand N, Podtelejnikov AV, Groth A et al (2002) Regulation of G(2)/M events by Cdc25A through phosphorylation-dependent modulation of its stability. EMBO J 21:5911–5920 Marini F, Nardo T, Giannattasio M et al (2006) DNA nucleotide excision repair-dependent signaling to checkpoint activation. Proc Natl Acad Sci USA 103:17325–17330 Marples B, Collis SJ (2008) Low-dose hyper-radiosensitivity: past, present, and future. Int J Radiat Oncol Biol Phys 70:1310–1318 Marples B, Wouters BG, Collis SJ et al (2004) Low-dose hyper-radiosensitivity: a consequence of ineffective cell cycle arrest of radiation-damaged G2-phase cells. Radiat Res 161:247–255 Matsuoka S, Huang M, Elledge SJ (1998) Linkage of ATM to cell cycle regulation by the Chk2 protein kinase. Science 282:1893–1897 Matsuoka S, Rotman G, Ogawa A et al (2000) Ataxia telangiectasia-mutated phosphorylates Chk2 in vivo and in vitro. Proc Natl Acad Sci USA 97:10389–10394 Melander F, Bekker-Jensen S, Falck J et al (2008) Phosphorylation of SDT repeats in the MDC1 N terminus triggers retention of NBS1 at the DNA damage-modified chromatin. J Cell Biol 181:213–226 Mirzoeva OK, Petrini JH (2003) DNA replication-dependent nuclear dynamics of the Mre11 complex. Mol Cancer Res 1:207–218 Molinari M, Mercurio C, Dominguez J et al (2000) Human Cdc25 A inactivation in response to S phase inhibition and its role in preventing premature mitosis. EMBO Rep 1:71–79 Murga M, Jaco I, Fan Y et al (2007) Global chromatin compaction limits the strength of the DNA damage response. J Cell Biol 178:1101–1108 Nakada D, Hirano Y, Sugimoto K (2004) Requirement of the Mre11 complex and exonuclease 1 for activation of the Mec1 signaling pathway. Mol Biol Cell 22:10016–10025

76

M. Lobrich et al.

Nakamura TM, Moser BA, Du LL et al (2005) Cooperative control of Crb2 by ATM family and Cdc2 kinases is essential for the DNA damage checkpoint in fission yeast. Mol Cell Biol 25:10721–10730 Nikjoo H, Bolton CE, Watanabe R et al (2002) Modelling of DNA damage induced by energetic electrons (100 eV to 100 keV). Radiat Prot Dosimetry 99:77–80 Noon AT, Shibata A, Rief N et al (2010) 53BP1-dependent robust, localized KAP-1 phosphorylation is essential for heterochromatic DNA double strand break repair. Nat Cell Biol 12:177–84 Painter RB, Young BR (1980) Radiosensitivity in ataxia-telangiectasia: a new explanation. Proc Natl Acad Sci USA 77:7315–7317 Panier S, Durocher D (2009) Regulatory ubiquitylation in response to DNA double-strand breaks. DNA Repair (Amst) 8:436–443 Pellicioli A, Lee SE, Lucca C et al (2001) Regulation of Saccharomyces Rad53 checkpoint kinase during adaptation from DNA damage-induced G2/M arrest. Mol Cell 7:293–300 Plans V, Scheper J, Soler M et al (2006) The RING finger protein RNF8 recruits UBC13 for lysine 63-based self polyubiquitylation. J Cell Biochem 97:572–582 Rainey MD, Black EJ, Zachos G et al (2008) Chk2 is required for optimal mitotic delay in response to irradiation-induced DNA damage incurred in G2 phase. Oncogene 27:896–906 Rhind N (2009) Changing of the guard: how ATM hands off DNA double-strand break signaling to ATR. Mol Cell 33:672–674 Rogakou EP, Boon C, Redon C et al (1999) Megabase chromatin domains involved in DNA double-strand breaks in vivo. J Cell Biol 146:905–916 Rogakou EP, Pilch DR, Orr AH et al (1998) DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. J Biol Chem 273:5858–5868 Segurado M, Tercero JA (2009) The S-phase checkpoint: targeting the replication fork. Biol Cell 101:617–627 Sherr CJ (2000) The Pezcoller lecture: cancer cell cycles revisited. Cancer Res 60:3689–3695 Sherr CJ, Roberts JM (1999) CDK inhibitors: positive and negative regulators of G1-phase progression. Genes Dev 13:1501–1512 Shibata A, Barton O, Noon AT (2010) Role of ATM and the damage response mediator proteins 53BP1 and MDC1 in the maintenance of G(2)/M checkpoint arrest. Mol Cell Biol 30: 3371–3383 Shiotani B, Zou L (2009) Single-stranded DNA orchestrates an ATM-to-ATR switch at DNA breaks. Mol Cell 33:547–558 Sonoda E, Hochegger H, Saberi A et al (2006) Differential usage of non-homologous end-joining and homologous recombination in double strand break repair. DNA Repair (Amst) 5:1021–1029 Sorensen CS, Hansen LT, Dziegielewski J et al (2005) The cell-cycle checkpoint kinase Chk1 is required for mammalian homologous recombination repair. Nat Cell Biol 7:195–201 Spycher C, Miller ES, Townsend K et al (2008) Constitutive phosphorylation of MDC1 physically links the MRE11-RAD50-NBS1 complex to damaged chromatin. J Cell Biol 181:227–240 Stewart GS, Panier S, Townsend K et al (2009) The RIDDLE syndrome protein mediates a ubiquitin-dependent signaling cascade at sites of DNA damage. Cell 136:420–434 Stewart GS, Stankovic T, Byrd PJ et al (2007) RIDDLE immunodeficiency syndrome is linked to defects in 53BP1-mediated DNA damage signaling. Proc Natl Acad Sci USA 104:16910–16915 Stewart GS, Wang B, Bignell CR et al (2003) MDC1 is a mediator of the mammalian DNA damage checkpoint. Nature 421:961–966 Stiff T, O’Driscoll M, Rief N et al (2004) ATM and DNA-PK function redundantly to phosphorylate H2AX following exposure to ioninsing radiation. Cancer Res 64:2390–2396 Stiff T, Walker SA, Cerosaletti K et al (2006) ATR-dependent phosphorylation and activation of ATM in response to UV treatment or replication fork stalling. EMBO J 25:5775–5782 Stucki M, Jackson SP (2004) MDC1/NFBD1: a key regulator of the DNA damage response in higher eukaryotes. DNA Repair (Amst) 3:953–957


77

Sturgeon CM, Knight ZA, Shokat KM et al (2006) Effect of combined DNA repair inhibition and G2 checkpoint inhibition on cell cycle progression after DNA damage. Mol Cancer Ther 5:885–892 Syljuasen RG, Jensen S, Bartek J et al (2006) Adaptation to the ionizing radiation-induced G2 checkpoint occurs in human cells and depends on checkpoint kinase 1 and Polo-like kinase 1 kinases. Cancer Res 66:10253–10257 Toczyski DP, Galgoczy DJ, Hartwell LH (1997) CDC5 and CKII control adaptation to the yeast DNA damage checkpoint. Cell 90:1097–1106 Uziel T, Lerenthal Y, Moyal L et al (2003) Requirement of the MRN complex for ATM activation by DNA damage. EMBO J 22:5612–5621 van Attikum H, Gasser SM (2009) Crosstalk between histone modifications during the DNA damage response. Trends Cell Biol 19:207–217 Wahl GM, Linke SP, Paulson TG et al (1997) Maintaining genetic stability through TP53 mediated checkpoint control. Cancer Surv 29:183–219 Wang B, Matsuoka S, Carpenter PB et al (2002a) 53BP1, a mediator of the DNA damage checkpoint. Science 298:1435–1438 Wang X, Khadpe J, Hu B et al (2003) An overactivated ATR/CHK1 pathway is responsible for the prolonged G2 accumulation in irradiated AT cells. J Biol Chem 278:30869–30874 Wang X, Li GC, Iliakis G et al (2002b) Ku affects the CHK1-dependent G(2) checkpoint after ionizing radiation. Cancer Res 62:6031–6034 Xu B, Kim ST, Lim DS et al (2002a) Two molecularly distinct G(2)/M checkpoints are induced by ionizing irradiation. Mol Cell Biol 22:1049–1059 Xu X, Tsvetkov LM, Stern DF (2002b) Chk2 activation and phosphorylation-dependent oligomerization. Mol Cell Biol 22:4419–4432 Yoo HY, Kumagai A, Shevchenko A et al (2004) Adaptation of a DNA replication checkpoint response depends upon inactivation of Claspin by the Polo-like kinase. Cell 117:575–588 Zhou XY, Wang X, Hu B et al (2002) An ATM-independent S-phase checkpoint response involves CHK1 pathway. Cancer Res 62:1598–1603 Zou L, Elledge SJ (2003) Sensing DNA damage through ATRIP recognition of RPA-ssDNA complexes. Science 300:1542–1548 Zou L, Liu D, Elledge SJ (2003) Replication protein A-mediated recruitment and activation of Rad17 complexes. Proc Natl Acad Sci USA 100:13827–13832

Chapter 4

Chromatin Responses to DNA Damage Karina Falbo and Xuetong Shen

Abstract The preservation of genome stability depends upon tightly regulated mechanisms that continuously search the genome for irregularities that, if left ignored, could be dangerous to both the genomic integrity and cell survival. These mechanisms are mediated by a multiplicity of proteins and factors that physically approach the DNA to either search for the damage or to fix it. Therefore, in eukaryotic cells, where the DNA molecule is tightly packed into nucleosomes forming a highly compacted structure, the chromatin, access of these factors to the DNA molecule represents an additional step, as well as an opportunity for extra regulatory mechanisms. Although a plethora of information has been accumulated over the past years on the factors and mechanisms involved in the response to DNA damage, very little is known about the role of the chromatin structure itself on the DNA damage response (DDR). Interestingly, several remodeling complexes have recently been described to be involved in the DNA damage response. In particular, the INO80 remodeling complex seems to be involved in several aspects of the DDR response. Thus, this chapter will describe the novel roles of the INO80 remodeling complex in DNA damage tolerance, double strand break repair (DSB) and telomere maintenance. Keywords INO80 • Chromatin remodeling

4.1 Chromatin Remodeling is an Integral Component of the DNA Damage Response The therapeutic activity of ionizing radiation on tumors is primarily based on the cell cytotoxicity derived from the inhibitory effects of radiation on vital cellular processes such as DNA replication and transcription. Radiation also induces the activation of

X. Shen (*) Department of Carcinogenesis, The University of Texas, MD Anderson Cancer Center, MDA SP/RD, Unit 116, 1808 Park Road 1-C, Box 389, Smithville, TX 78957, USA e-mail: [email protected] T.L. DeWeese and M. Laiho (eds.), Molecular Determinants of Radiation Response, Current Cancer Research, DOI 10.1007/978-1-4419-8044-1_4, © Springer Science+Business Media, LLC 2011

79

80

K. Falbo and X. Shen

the so-called DNA damage response, a surveillance mechanism responsible for the preservation of genetic stability. Interestingly, therapeutic inactivation of this response seems to represent a window of opportunity to enhance the effects of radiation therapy (Ljungman 2009). As part of the DDR, several proteins have been identified and studied in the past few years leading to a clear understanding of the different steps involved in these repair processes (Powell and Bindra 2009). However, since the DNA is not naked but wrapped around histone proteins forming a highly compacted structure, the chromatin, it became evident that remodeling of the chromatin might be an essential component of the DDR (Morrison and Shen 2006, 2009). Furthermore, recent investigations have shed light on the importance of chromatin structure during the DDR opening new questions on whether chromatin remodeling complexes could represent a new player in the field of radiation oncology.

4.2 The Chromatin Environment In order to fit the long DNA molecule into the nucleus, eukaryotic cells evolved to create a highly compacted structure, the chromatin. In the chromatin, the DNA molecule is wrapped around proteins called histones forming the basic chromatin unit, the nucleosome. Nucleosomes interact with each other to further fold the DNA molecule to create a higher level of compaction, the so-called 30 nm fiber (Staynov 2008). Thus, as a consequence of this complexity, the chromatin creates impediments and constrains to nuclear processes that require access to DNA, such as DNA replication or the DDR. Histones, a group of highly evolutionarily conserved proteins, are the main mediators of DNA folding into the chromatin structure, since they can interact with each other, with the DNA and with other chromosomal proteins. Each nucleosome core is formed by two molecules of each of four different histone proteins: H2A, H2B, H3, and H4, with histones H2A and H2B forming a stable dimer (H2A/H2B) and histones H3 and H4 forming a stable tetramer (H3/H4)2 (Vernet et al. 1990). The H3/H4 histone tetramer has a pivotal role in nucleosome organization, while the H2A/H2B dimer is mainly responsible for the interaction between the DNA and the nucleosome core. Finally, the histone fold domains of the core histones are responsible for organizing the DNA around the histone core, primarily through electrostatic interactions (Luger et al. 1997). Histones are actively involved in the regulation of the chromatin environment, through their ability to be evicted, exchanged or displaced along the DNA (Huertas et al. 2009). In addition, histones are posttranslationally modified by phosphorylation, ubiquitylation, methylation, acetylation, etc., at both the N-terminal tail as well as at the globular domain, modifications that are essential for many nuclear processes. Interestingly, while the role of histone mobilization, exchange, or eviction in the DDR is essentially an unexplored area, the involvement of posttranslational modifications is, instead, in active investigation (Altaf et al. 2007). Specifically, in the yeast Saccharomyces cerevisiae, histone H2AX is phosphorylated at Ser129 in

4 Chromatin Responses to DNA Damage

81

the region around a double strand break (Rogakou et al. 1998). The phosphorylated form of H2AX, gH2AX, occurs in cis on the DNA and extends in just minutes to around 50–100 kb from the DSB in both directions after irradiation. Moreover, the recruitment of factors involved in the early steps of DSB repair depends on H2AX phosphorylation, supporting a pivotal role for this posttranslational modification in the chromatin structure and in the response to DNA damage. Finally, although very little is known about the role of chromatin structure in the DNA damage response, several lines of evidence indicate that nucleosome mobilization, exchange, and eviction mediated by chromatin remodeling complexes constitute a central component of this response. Therefore, this chapter will focus on the rapidly expanding research related to the role of chromatin structure and chromatin remodeling complexes in DNA damage-related nuclear processes.

4.2.1 Chromatin Remodeling Complexes and the Dynamic Nature of Chromatin The dynamic nature of the chromatin is possible thanks to the concerted activities of several chromatin modifier factors. In fact, these factors are responsible for the regulation of essential nuclear processes such as: covalent modification of histone tails, histone exchange, DNA methylation and nucleosome mobilization. Here, we will focus on a subgroup of chromatin modifiers that utilize the energy provided by ATP to mobilize nucleosomes or exchange histones, the ATP-dependent remodeling complexes. More specifically, we will focus our discussion on the INO80 remodeling complex, since the last few years have witnessed an accelerated rate on the discovery of novel functions for this complex, including the DNA damage tolerance pathways during replication and DSB repair.

4.2.2 The INO80 Remodeling Complex The INO80 remodeling complex is an evolutionarily conserved complex that bears both an ATPase activity and a 3¢–5¢ helicase activity. Even though the mechanism is still unclear, INO80 can mobilize nucleosomes using the energy provided by ATP (Shen et al. 2000). In fact, the first function attributed to INO80 was during transcription, where INO80 binds to the promoter regions of certain activated genes and mobilizes the surrounding nucleosomes, using energy provided by ATP, to allow access to the DNA by transcription factors (Shen et al. 2003b). Furthermore, in the last few years, several novel transcriptional-independent functions were discovered, including regulation of DSB repair, DNA damage tolerance response during DNA replication, and telomere maintenance (Fig. 4.1). Thus, the main discussion topics of this section will be the INO80 structural characteristics that relate to these novel research areas.

82


a

b

DSB repair

Other Arp8 subunits

ATP

Ino80

DNA replication

Other Arp8 subunits ATP

Ino80

Nhp10 P

PCNA

H2AX Nucleosome eviction Strand invasion Homologous Recombination

c

Telomere maintenance Other subunits

ATP

Ino80

Ies3 Est1

Rad18 Rad51

Template switch Damage avoidance

Promotion of recombinational telomere maintenance?

Fig. 4.1 INO80 is involved in several nuclear functions. (a) INO80 activities at DSBs. After MMS treatment, INO80 binds to double strands breaks and mediates homologous recombination by interaction with the phosphorylated form of histone H2AX. This interaction depends on H2AX phosphorylation and the INO80’s subunit Nhp10. At the DSBs site INO80 evicts or displaces nucleosomes facilitating the proper recruitment of HR factors, such as Rad51. (b) INO80 activities during DNA replication. When replication forks face an obstruction, such as an MMS adduct, INO80 binds to the DNA to help process stalled replication forks. INO80 allows proper recruitment of DNA damage tolerance factors leading to fork resolution though template switching and the completion of DNA replication without DNA damage generation. (c) INO80 activities at telomeres. INO80 is involved in the regulation of telomere structure and function. Ies3 subunit interacts with the Est1 telomeric protein, suggesting a direct role for INO80 at telomeres

Structurally, INO80 is composed of 15 subunits: Ino80, Arp8, Rvb1, Nhp10, Rvb2, Arp4, Arp5, Actin, Ies1, Ies2, Ies4, Anc1/Taf14, Ies5, and Ies6. Interestingly, while some of these subunits are well characterized, some others still remain to be investigated. However, due to the extensive list of subunits and space limitations, we will consider only the subunits that are relevant to our discussion (Bao and Shen 2007). Probably the most important subunit is Ino80 that confers the ATPase activity. Ino80 contains the GXGKT motif at the ATPase domain, a highly conserved nucleotide-binding motif that also contains a lysine residue known to interact with an ATP phosphate, and whose alteration renders the INO80 complex nonfunctional. The INO80 complex also contains several Arp subunits with diverse functions. Arp5 and Arp8 are INO80 exclusive subunits, and both Arp4 and actin incorporate into the complex depend on Arp8. In addition, Arp5 and Arp8 are important for the chromatin remodeling activity of the complex, while Arp4 is the only subunit that binds ATP (Shen et al. 2003a). Furthermore, Arp8 and Arp4 are known to interact in vitro and in vivo with histones H3 and H4, possibly to perform chaperone activities between histones and the DNA. Another important subunit is Nhp10, an HMG-1 like protein that could potentially bind to structured DNA or nucleosomes. Nhp10 is not important for INO80-mediated nucleosomes mobilization, but it is relevant for the recruitment of Ies3 into the complex (Bao and Shen 2007). More importantly, we will discuss later in this chapter, a key interaction between Nhp10 and gH2AX during DSB mediated by INO80. Finally, the molecular functions of Ies1, 3, 4, and 5 are still unclear; however, we will discuss below recent findings that provide interesting clues about the putative roles of some of these subunits in important nuclear functions. Therefore, the


83

diversity and complexity of activities performed by the different INO80 subunits support the view that the high degree of conservation reflects a pivotal, essential, and versatile role of this complex in vital nuclear processes.

4.3 INO80 is Directly Involved in the DNA Damage Response The most important effect of ionizing radiation is the generation of DNA double strand breaks, lesions that are repaired by either homologous recombination (HR) or nonhomologous end joining (NHEJ). Thus, cells that are sensitized by inhibition of DSB repair pathways could represent an excellent target for selective therapeutic radiation (Friedberg et al. 2006; Shrivastav et al. 2008). As such, numerous inhibitors have been developed to target DNA-dependent protein kinase (DNA-PK), a key player in the NHEJ pathway. However, due to the highly compacted nature of the chromatin, providing DDR proteins access to the DNA requires additional factors, such as chromatin remodeling complexes. Surprisingly, very little is known about the role of chromatin structure in the DDR response. Indeed, three chromatin remodeling complexes have recently been shown to be involved in DDR in the yeast S. cerevisiae: SWI/SNF, RSC, and INO80 (Osley and Shen 2006), opening a novel research area for the exploration of new therapeutic options. The INO80 remodeling complex was originally suspected to be involved in the DDR response due to the sensitivity of the ino80, arp8, and arp5 mutants to agents that cause DSB, such as ionizing radiation (IR) and methyl methanesulfonate (MMS) (Shen et al. 2000, 2003a; Morrison et al. 2004; van Attikum et al. 2004; Tsukuda et al. 2005). Also, it was well established that INO80 binds to the promoter regions of a multiplicity of genes to direct nucleosome mobilization. As such, one possibility was that the sensitivity of the mutants to DNA-damaging agents could be a consequence of INO80 remodeling activity at promoter regions of genes involved in DDR. Nonetheless, as it was later discovered, it was also possible that INO80 activity at DDR was direct instead of transcription-related.

4.3.1 The INO80 Remodeling Complex Binds to Double Strand Breaks After DSB formation, DNA integrity can be restored by two major mechanisms: an error prone process, where the DNA broken ends are directly ligated through a nonhomologous end-joining mechanism, or an homologous recombination process that involves copying information from a sister chromatid or homolog. Due to the highly compacted nature of the chromatin, both processes require chromatin changes to pave the way for the DDR machinery to access the DSB site. Interesting insights into the putative direct role of the INO80 remodeling complex during the DDR response came from investigations by Morrison et al. and van Attikum et al. with the discovery that the INO80 remodeling complex is directly involved in the DDR response.

84


This original discovery was performed using a well-studied genetic system developed by J. Harber. In the yeast S. cerevisiae, a single DSB can be created at the MAT locus by induction of a specific HO endonuclease. This single DSB cannot be efficiently repaired due to the lack of recombination donors, thus, allowing the characterization of recruited DNA repair factors to the DSB using chromatin immunoprecipation (Moore and Haber 1996; Haber 2000). In this model system, chromatin immunoprecipitation (ChIP) analysis of flag-tagged versions of Ino80, Arp8, and Arp5 shows that these subunits are recruited to a DNA region that extends up to 9 kb from the HO-induced DSB break. In addition, the kinetics of recruitment to DSB sites is similar to that of several proteins involved in DSB repair, suggesting a coordinated function of INO80 with other repair factors (Morrison et al. 2004; van Attikum et al. 2004).

4.3.2 INO80 Binding to Double Strand Breaks Depends on Histone H2AX Phosphorylation One of the first responses to DNA damage in yeast is the recruitment of the phosphorylated form of histone H2AX (gH2AX) to DSBs (Downs et al. 2000; Redon et al. 2003). gH2AX is detected in the region surrounding the DSB within 15 min after induction of an HO cut (Shroff et al. 2004). Interestingly, the observation that INO80 associates with histones under physiological conditions, together with the finding that INO80 binds to DSBs, led Morrison and Van Attikum to envision a mechanism in which INO80 recruitment to DSBs is, somehow, mediated by phosphorylation of H2AX. In fact, the phosphorylated form of H2AX is enriched with the INO80 complex in cell cultures treated with MMS to induce DSBs. Using an antibody that specifically recognizes gH2AX, INO80 can be detected in association with gH2AX when the complex is purified under mild salt conditions. Moreover, INO80 recruitment to DSBs is significantly reduced in a strain bearing a mutation at the H2AX serine residue that is phosphorylated, indicating that INO80 recruitment to DSBs is mediated by H2AX phosphorylation. Therefore, after DSB induction, INO80 binds preferentially to the phosphorylated form of H2AX present at the DSB and this interaction depends on H2AX phosphorylation (Fig. 4.1). Interestingly, several INO80 subunits are dispensable for INO80 recruitment to DSBs, including the Arps. However, only Nhp10, an exclusive INO80 subunit, is necessary for the INO80–gH2AX interaction. Thus, since Nhp10 deletion does not significantly affect INO80 remodeling activity (Shen et al. 2003b), it can be reasoned that the Nhp10 subunit confers upon INO80 the exclusive capability to bind to gH2AX independently of its remodeling activity. Finally, the direct role of INO80 at DSBs is strongly supported by data collected by comparing the global transcription profile of the wild-type and the ino80 strains. As such, a survey of the global transcription profile of the ino80 mutant reflects no major changes in the expression profile of genes associated with DNA damagerelated pathways, including cell cycle arrest, checkpoint activation, homologous


85

recombination and nonhomologous end joining (Mizuguchi et al. 2004), reinforcing the idea that INO80 activities at HO sites are not related to the complex ability to remodel the chromatin at promoters (Shen et al. 2000), but rather to a direct activity of INO80 at the HO cut.

4.3.3 INO80 is Involved in Homologous Recombination-Mediated DSB Repair INO80’s presence at DSBs and its interaction with gH2AX led to the early speculation that the complex could be necessary for NHEJ and/or HR at DSBs. In yeast, the Mre11–Rad50–Xrs2 complex is recruited to the DNA immediately after DSB formation to mediate DNA resection that leads to ssDNA formation (Lee et al. 1998; Nakada et al. 2004), an early step in HR. Interestingly, this original speculation was supported by the observation that in S. cerevisiae, 5¢–3¢ resection of HO-induced DNA ends is reduced in the arp8 mutant, when assessed by a quantitative PCR-based amplification assay (Morrison et al. 2004). However, there are currently two opposing views based on the investigations of two different labs. In one study, both the arp8 and nhp10 mutants were shown to be defective in strand resection when assessed by an RT-PCR method called QAOS (van Attikum et al. 2004). Using this method, ssDNA can be detected within 1 hr at a site located 1.6 kb of the HO cut in a wild-type strain, but the signal is significantly reduced in the arp8 and nhp10 mutants, which leads to the conclusion that chromatin remodeling driven by INO80 favors DNA end processing. Moreover, recruitment of Mre11, a factor involved in DNA resection, is significantly reduced in the arp8 mutant at HO sites, indicating that INO80 remodeling activity at HO sites facilitates Mre11 recruitment that leads to proper DNA resection. In a second independent study, using Southern blot to detect strand resection at the MAT locus, and a recruitment assay to measure the association and spreading of RPA with single-stranded MAT DNA, Tsukuda et al. described that end-strand resection occurs normally in the arp8 mutant even though the recruitment of Rad51 and Rad52, two proteins involved in stand invasion, is delayed in the same mutant (Tsukuda et al. 2005). Therefore, this data is in agreement with a theory in which the INO80 complex binds to HO sites and allows the recruitment of early HR factors, possibly by regulating DNA resection at the DSB. Further research is required to clarify these disparate views.

4.3.4 INO80 is Involved in the Early and Late Steps of Homologous Recombination INO80 has been described to be indispensable for HR repair in several organisms. In Arabidopsis, the frequency of HR is reduced by 15% in a mutant that does not

86


express INO80. Interestingly, reports in yeast seem to indicate that INO80 has no significant effect on the frequency of HR-mediated DSB repair when assessed in a haploid HO MAT system that lacks the donor sequences (van Attikum et al. 2004; Tsukuda et al. 2005). Thus, the disparity in the data described above, as to whether or not INO80 affects the outcome of HR-mediated DSB repair, prompted the use of a different genetic system to further analyze DSB repair in a step-by-step approach to address this issue. Using a diploid HO MAT system where the donor sequence is present, Tsukuda et al. showed that the total allelic HR frequency is reduced approximately fivefold with all HR classes reduced to a similar extent in mutants that lack INO80. In this allelic system, homologous recombination between chromosomes is stimulated by a single HO DSB generated in an ura3 allele, allowing the analysis of conversion tract lengths, G2 crossovers, DSB-dependent chromosome loss, break-induced replication (BIR), and DSB-induced cell death. Remarkably, this study showed that Arp8 is necessary during early and late steps of HR (Tsukuda et al. 2009). In fact, during early HR, Arp8 accelerates the strand invasion step, despite the fact that it does not affect DNA resection. Indeed, in an HO MAT system where the donor sequences are available, strand invasion is dramatically impaired in the arp8 mutant when assessed using a single-end invasion assay (SEI), a PCR-based assay that allows quantification of the initial strand invasion/repair synthesis reaction after a DBS. Furthermore, dissection of the late HR steps using the ura3 system reveals important differences. First, the frequency of conversion tracts, a hallmark of defective HR, is increased in the arp8 mutants. Second, the frequency of marker conversion as a function of distance from the DSB, another parameter related to tract length, is altered in the arp8 mutants. In wild-type cells, markers 5¢ of the DSB convert more frequently than equidistant 3¢ markers, but in the arp8 mutant several 3¢ markers convert at significantly higher frequencies that in wild type. Thus, since the Arp8 subunit is indispensable for INO80 ATPase activity, it can be inferred from the analysis of HR in diploids that chromatin remodeling mediated by the INO80 complex regulates both early stand strand invasion and the later steps that control gene conversion, tract length, and continuity (Tsukuda et al. 2009).

4.3.5 INO80 Can Evict Histones at DSBs An important question on the role of chromatin remodeling during DSBs refers to the putative mechanism involved, and in this regard, several lines of evidence indicate that ATP-driven remodeling activity is involved in INO80 activities at DSBs. As described above, arp8 mutants that lack the INO80 ATPase activity are defective in several HR steps. Also, a point mutation that specifically abolishes the INO80 ATPase remodeling activity is sensitive to chemicals that induce DSBs. Nonetheless, robust data addressing whether or not the chromatin remodeling activity is important during DSB repair comes from the analysis of nucleosome and histone occupancy at DSBs showing that the INO80 complex is involved in nucleosome eviction at DSBs.


87

The HM loci is tightly packaged into heterochromatin, thus, during MATa ating-type switching, the normally silenced HO site in the HMLa that is occluded m by a strategically positioned nucleosome is cleaved, indicating chromatin remodeling takes place at this locus. Similarly, during MATa switching, histones located at the HMRa region are displaced, indicating that the search for homology of the invading strand requires chromatin remodeling at the donor sequence. Accordingly, analysis of histone H2B occupancy by microccocal nuclease digestion in both MATa and HMRa shows that while H2A is rapidly evicted from Mata in both wild-type and arp8 strains, H2A presence at the HMRa is retained only in the arp8 mutant. In addition, transfer of a HR mediator, Rad51, from the recipient to the donor strand that is a key step in homology search, is less efficient in the arp8 mutant, indicating INO80 plays an important role in chromatin remodeling at the donor sequence, possibly by mediating histone eviction that facilitates strand invasion by the recipient template (Tsukuda et al. 2009). Similarly, ChIP analysis of histones using the haploid HO MAT system shows that INO80 is involved in general nucleosome eviction at an HO cut. Indeed, after 1 hr of HO induction gH2AX accumulates and spreads from the DSB at MAT, peaking at +9.6 kb from the DSB in wild-type as well as in arp8 and nhp10 mutant strains. However, after 2–4 hr of HO induction, gH2AX decreases dramatically in the wild-type but not in the arp8 and nhp10 mutants, indicating that gH2AX is removed from the chromatin at an HO DSB in an INO80-dependent manner. Moreover, ChIP analysis of core histones shows similar results, a marked reduction in the presence of core histones that correlates with the reduction in H2AX occupancy in the arp8 and nhp10, leading to the conclusion that INO80 is required for the general eviction of nucleosomes at HO sites after DSB generation (Tsukuda et al. 2009).

4.3.6 INO80 is Involved in the Recruitment of DNA Repair Factors to DSBs INO80-mediated nucleosome eviction, and consequent chromatin remodeling, allows the recruitment of several proteins involved in different steps of HR. In fact, the recruitment of DSB repair factors, such as Rad51 and Rad52 and Mre11 to DSB is significantly reduced in the arp8 mutant, when assessed in a MAT strain that lacks donor loci (Tsukuda et al. 2009). Although it is still not clear whether or not Rad51 recruitment affects DNA resection (see disparate results described before), the evidence supports the fact that Rad51 recruitment impacts the efficiency of the strand invasion step during HR-mediated repair at DSBs. Accordingly, when HM donor loci are present, Rad51 is transferred from recipient MAT ssDNA to the donor template to effect DNA strand exchange. Furthermore, ChIP analysis using a MAT system in which the donor locus is present shows that while Rad51 recruitment to the Mata locus is similar in wild-type and arp8 mutants, in the MATa locus Rad51 accumulates to higher levels in the arp8 mutant, indicating INO80-dependent

88


chromatin remodeling is required for efficient Rad51 transfer from the invading strand to the donor HMRa strand (Tsukuda et al. 2009). In summary, a mechanism can be envisioned in which, by evicting or displacing nucleosomes at DSBs, INO80 allows proper recruitment of proteins involved in DSB repair. Specifically, strand invasion depends on the interaction of a ssDNARad51 nucleoprotein with a donor locus, interaction that is promoted by the INO80 chromatin remodeling activity. Therefore, INO80’s remodeling activity induces nucleosome eviction specifically at the donor sequences allowing or facilitating the transfer of Rad51 from the invading to the donor strand. Interestingly, since INO80 was never reported to have histone transfer activity (Shen et al. 2000), it is possible that INO80 originally disrupts the nucleosome structure to facilitate the subsequent nucleosome displacement by a chaperone, possibly Asf1, a global chaperon known to be involved in DSB repair (Tyler 2002; Adkins and Tyler 2004; Prado et al. 2004).

4.4 INO80 is Phosphorylated at the Ies4 Subunit An important question on the role of INO80 in DNA repair concerns the regulatory mechanisms that direct INO80 activities during the DDR. Insights into this issue came from the investigations published by Morrison et al., describing the presence of a posttranslational modification in the INO80 remodeling complex. Specifically, Morrison et al. found that the Ies4 subunit of the INO80 complex is phosphorylated after MMS treatment. Indeed, 2D gel analysis of MMS-treated wild-type cultures reveals five phosphorylated Ies4 spots that map to the Ies4 N-terminal peptide that contains two serines in Mec1/Tel1 kinase (S/T)Q consensus sites. In addition, Ies4 phosphorylation after MMS treatment is impaired in strains that lack either the Mec1 or Tel kinases, early effectors in the DNA damage-related checkpoint response. After DNA damage, Mec1/Tel1 phosphorylates several checkpoint proteins, many of which have a Mec1/Tel1 kinase (S/T)Q consensus site, leading to checkpoint activation (Fig. 4.2). Remarkably, purified INO80 can be phosphorylated at the Ies4 subunit in an in vitro kinase, assay, in a mix that contains precipitated Mec1 or Tel1, indicating that INO80 is, in fact, phosphorylated by Mec1 and Tel1 and thus involved in the checkpoint pathway (Morrison et al. 2007). Interestingly, while ies4 mutants that cannot be phosphorylated by Mec1/Tel1 are not sensitive to DNA-damaging agents such as MMS, a serine to glutamic acid mutation that mimics phosphorylation results in high sensitivity to MMS, suggesting Ies4 phosphorylation is relevant during DSB repair. However, analysis of several steps of HR using the HO MAT system that repair the DSB via an ectopic donor or single strand annealing shows that none of these phosphorylation mutants are defective in any of the steps analyzed, including strand invasion, ligation of repair ends, product formation, and cell survival, indicating that Ies4 phosphorylation by Mec1/Tel1 seems to be dispensable for HR-mediated DSB repair (Morrison et al. 2007).


Mec1

Tel1

P

89

Mec1

Tel1

H2AX

Nhp10

Other subunits Arp8 Ino80 ATP

DNA repair

P Ies4

Ino80 Other subunits

Checkpoint modulation

Fig. 4.2 INO80 activities at DSBs are influenced by Mec1/Tel1. When cells are treated with DNA-damaging agents, Mec1/Tel1 mediates either the phosphorylation of H2AX that leads to the recruitment of INO80, or the phosphorylation the Ies4. Following INO80 binding to a DSB, some subunits such as Nhp10, Arp8, and Arp5 are involved in repair mechanisms while Ies4 is involved in checkpoint activation

Therefore, since Mec1/Tel1 activation is not only involved in HR repair but also in the early steps of checkpoint activation, it is possible that Ies4 has an active role during checkpoint pathways. After DNA damage generation in yeast, ssDNA formation induces recruitment of RPA, a single strand DNA-binding complex that consequently recruits the Ddc2–Mec1 complex, leading to Rad53 phosphorylation and checkpoint activation (Harrison and Haber 2006). Interestingly, Rad53 phosphorylation and its activity are enhanced in the ies4 mutant cells that mimic a persistent phosphorylation, which leads to the consequent enhanced checkpoint response and pronounced cell cycle arrest. Surprisingly in the arp8 and nhp10 mutants, Rad53 phosphorylation and activity are significantly reduced after induction of an HO site, as compared with a wild-type strain. Moreover, this defect could be attributed to an impaired ability of the arp8 mutant to recruit Mec1, a Rad53 downstream effector. In fact, ChIP analysis using a HO MAT system lacking a donor locus shows that while Mec1 is present at HO sites within 1 h after DSB induction, this recruitment is significantly reduced in the arp8 and nhp10 mutants (Morrison et al. 2007). In summary, after DNA damage, the INO80 remodeling complex is posttranslationally phosphorylated by the Mec1/Tel1. Thus, while the lack of the Arp8 subunit impairs Mec1/Tel1 recruitment to DSB and the consequent Rad53 phosphorylation, failure to regulate Ies4 phosphorylation induces Rad53 hyperphosphorylation and prolonged cell cycle arrest, supporting the idea that different INO80 subunits could be responsible for different aspects of the DDR response. Finally, a future extension of this research could reveal a more detailed mechanism for INO80 posttranslational modifications and its role in checkpoint activation.

90


4.5 INO80 is Important for Telomere Maintenance Telomeres, the end terminal structure of chromosomes, consist of an array of repetitive sequences that, due to their resemblance to DSBs, are important for the maintenance of genomic stability. In fact, to avoid recognition by the NHEJ machinery, telomeres are arranged in a higher order nucleoprotein structure that attaches to the nuclear matrix (Misri et al. 2008). Nonetheless, despite the extensive research accumulated on the protein component of telomeres, very little is known about the role of the chromatin structure at telomeres. The chromatin structure at telomeres is unique; it is known that in eukaryotes telomere DNA is organized in tightly packed nucleosomes separated by 10–20 bp of linker DNA (Pisano et al. 2008) and that the telomeric DNA is associated with several proteins important for telomere structure and maintenance. Interestingly, several lines of evidence indicate that most DNA damage response proteins are, somehow, involved in telomere maintenance. Indeed, in yeast, the MRN complex that plays a critical role in DNA DSB repair in eukaryotes, also binds to telomeres (Lombard and Guarente 2000; Nakamura et al. 2002; Takata et al. 2005). In human cells, the MRN complex, together with ATM (Tel1), is involved in the activation of the ATR kinase (Mec1) after IR exposure (Jazayeri et al. 2006). Moreover, a similar situation is observed in yeast, where Mre11, a MRN complex component, is implicated in telomere resection and Mec1 loading during late S phase (Larrivee et al. 2004), indicating that the telomeric function of Mre11 might reflect the role of Mre11 in the recognition and processing of DNA DSBs. Therefore, the involvement of chromatin remodeling complexes in DSBs leads to the possibility that a similar process is present at telomeres. In this regard, the discovery by Yu et al. of a physical interaction between a telomeric protein, Est1, and an INO80 subunit, Ies3, have shed light into the role of chromatin remodeling at telomeres (Yu et al. 2007). Using a two hybrid screening system Yu et al. found that the Ies3 subunit of the INO80 remodeling complex interacts directly with Est1, a highly conserved polypeptide essential for telomerase maintenance and protection in vivo. Est1 is involved in telomerase recruitment to chromosome ends as well as in activation functions after recruitment that are still not well understood (Virta-Pearlman et al. 1996; Pennock et al. 2001; Seto et al. 2002; Taggart et al. 2002). In addition, Yu et al. observed that some INO80 subunits could influence the length as well as the structure of telomeres (Yu et al. 2007). As such, the moderate but significant telomere elongation characteristic of the ies3, arp8, and nhp10 mutants, observed by analysis of telomere restriction fragments, together with the significant reduction of telomere position effect (TPE) in the ies3 mutant, suggest that the INO80 remodeling complex is involved in telomere structure and function as described below. In yeast that had lost telomere proteins, such as Est1, telomeres experience attrition that leads to growth defects. Interestingly, when these mutants are subcultured, after several passages, a subpopulation of survivor cells arises with growth rates that are similar to wild-type cell rates. Moreover, it is believed that these survivors use a recombination-mediated mechanism for telomere maintenance, since loss of several recombination proteins impairs the ability of this strain to form survivors (Lundblad


91

and Szostak 1989; Lundblad and Blackburn 1993; Teng and Zakian 1999; ass-Eisler and Greider 2000). Interestingly, the est1 ies3 double mutant presents K this type of growth retardation. Moreover, est1 mutant subcultured cells recover after 3 days, similar to wild type, while the est1 ies3 double mutant fails to recover until day 7, suggesting that Ies3 is important for HR at telomeres. In addition, both strains accumulate long and heterogeneous terminal restriction fragments that are typical of telomere-related mutants. However, 2D gel analysis shows that only the double mutants accumulate unusual telomeric DNA structures, such as circular DNA, as well as high levels of G strand overhangs (Yu et al. 2007). Furthermore, altered structures at telomeres can also be repaired by nonhomologous end joining, whose main products are T–T fusions (van Gent et al. 2001). Interestingly, although PCR analysis of the ino80 and est1 mutants shows no significant presence of these aberrant fusions, a high level of T–T fusions is present in the double est1 ies3 mutant, suggesting a role for the ies3 protein in NHEJ at telomeres. Similar to the situation at DBSs, the question remains as to whether or not INO80 remodeling activity is necessary for INO80 telomere-related activities. Interestingly, the ino80 mutant that lacks the ATP-dependent remodeling activity (Yu et al. 2007) does not present any major telomere defects, suggesting that the ies3 mutant effect on telomeres could be independent of any remodeling activity. However, in the arp8 mutant that is also required for the chromatin remodeling activity of INO80, telomere length, as well as formation of survivor clones, are affected, suggesting that at least some INO80 activities at telomeres are remodeling-dependent. In support of this observation, chromatin immunoprecipitation of Ino80, Ies3, and Nhp10 indicates that these subunits bind to telomeric regions, supporting the view that INO80 could be remodeling the chromatin at telomeres. Furthermore, INO80 binds to telomeres independently of Est1, suggesting that Est1 direct interaction with Ies3 may not have a structural role, such as helping INO80 bind to telomeres, but instead a functional one, such as the activation of INO80 after recruitment (Yu et al. 2007). In summary, several INO80 subunits are involved in the regulation of telomere structure and function. Specifically, the Ies3 subunit directly interacts with Est1, a telomere protein, interaction that does not seem to be necessary for INO80 recruitment to telomeres (Yu et al. 2007). Moreover, the preferential localization of some INO80 subunits to telomeres suggests that INO80’s activities at telomeres might be direct, instead of transcription-related. Nonetheless, further investigations are required to shed light on several unaddressed issues, especially since current data seems to indicate that different subunits could have different roles at telomeres.

4.6 INO80 is Involved in the DNA Damage Tolerance Pathways 4.6.1 The DNA Damage Response During Replication To achieve self-perpetuation, the eukaryotic cell has to make an accurate copy of the entire genome. Thus, replication represents a main challenge to the cell, since the DNA molecule could be easily broken or altered during the process. Moreover,

92


since the DNA is highly compacted into the chromatin, the DNA replication machinery depends on several chromatin remodeling factors to obtain access to the DNA molecule. As such, DNA replication has developed into a highly coordinated and regulated mechanism to ensure replication fidelity as well as the avoidance of replication-related damage (Falbo and Shen 2006). In fact, an important aspect of the DNA damage response is linked to the process of replicating the DNA, provided that replication forks might be able to process any dangerous DNA alterations that could compromise replication fork stability, and the correct completion of DNA replication. Fork instability leads to fork collapse and the generation of broken chromosomes, genome aberrations, and mutations, the most common hallmarks of genome instability. Thus, when replication forks encounter an obstruction, the S phase checkpoint is activated. S phase checkpoint activation halts DNA replication and induces the recruitment of several factors to stabilize the fork and to allow proper processing of the obstruction (Branzei and Foiani 2009). Interestingly, a stalled replication fork has two possible outcomes. On one hand, failure to stabilize obstructed replication forks leads to fork collapse and the generation of DSB that can be repaired by gene conversion or break-induced replication, two DNA damage repair processes characteristic of late S phase or G2. On the other hand, stalled replication forks can resume DNA replication by different mechanisms all of which are mediated by PCNA, a key molecule during unperturbed DNA replication. These so-called DNA damage tolerance mechanisms confer cell the ability to avoid DNA damage generation as a consequence of DNA replication itself. In fact, during DNA damage tolerance lesions can be bypassed by using specialized polymerases. Replication can also be restarted downstream of the lesion leaving a gap that is filled by translesion synthesis (TLS) polymerase, or it can undergo template switching, a mechanism that involves homologous recombination, where the newly synthesized DNA strand is utilized as a template for DNA synthesis across the gap (Branzei and Foiani 2009).

4.6.2 Chromatin Remodeling at the Onset of DNA Replication During DNA replication, the chromatin structure is radically affected, resulting in the rupture and establishment of histone–DNA interactions that are mostly mediated by chromatin remodeling complexes. Surprisingly, although the role of chromatin in transcription is well known from studies in the past two decades, the role of chromatin in DNA replication has not been widely investigated. In fact, from the limited research in the area we currently know that several chromatin remodeling complexes are involved in different steps during unperturbed DNA replication (Falbo and Shen 2006). However very little is known about the role of these complexes in the DNA damage tolerance response during DNA replication. Thus, this section will discuss the recent discoveries linking the INO80 remodeling complex to DNA replication and more specifically to the DNA damage avoidance pathways.


93

4.6.3 A Direct Role for INO80 During DNA Replication Originally, accumulated evidence argued in favor of an indirect, transcriptionalmediated, role of chromatin remodeling during DNA replication. However, the observation that hydroxyurea (HU), a compound that blocks DNA replication, dramatically induces Ino80 expression during S phase, strongly supported the idea of a direct role of INO80 during DNA replication. Moreover, the ino80 mutant shows a prolonged cell cycle and is hypersensitive to DNA replication blocking agents, such as MMS and HU (Falbo et al. 2009). Therefore, these early observations suggested a defective response to stalled replication forks in mutants that lack INO80.

4.6.3.1 INO80 Binds to Origins of Replication During S Phase An important indicator of the direct role of INO80 during replication is the fact that INO80 binds to ARS during S phase. In fact, in the yeast S. cerevisiae, whole genome ChIP–chip analysis of cells tagged at the INO80 locus shows that INO80 binds to 45% of ARS in cells arrested in S phase with HU but only to 5% of cells arrested in G2 with nocodazole, indicating that INO80 binding to ARS is S phase specific. In addition, INO80 does not affect the transcription of genes known to be implicated in DNA replication, when assessed by transcription microarray analysis, supporting the idea that INO80’s role in S phase is direct instead of transcriptionrelated. Finally, INO80 does not seem to discriminate between early and late ARS, a characteristic shared by many factors involved in the S phase checkpoint. Instead, INO80 binds to early, as well as late ARS almost with the same probability genome wide (Falbo et al. 2009). Therefore, since HU induces the checkpoint response, and many checkpoint factors are recruited only to early ARS, INO80 binding distribution to ARS does not correspond with a checkpoint-related function, as it will be explain in the next section.

4.6.3.2 INO80 and the S Phase Checkpoint After HU treatment, replication forks stall, leading to S phase checkpoint activation that is mainly mediated by Rad53 phosphorylation. Checkpoint activation induces cell cycle delay, fork stabilization, and prevention of late firing origins activation. Thus, in the absence of Rad53 early replication forks collapse and late firing origins repression is lost (Branzei and Foiani 2007). Interestingly, the involvement of INO80 in the S phase checkpoint seems to be controversial, since analysis of Rad53 phosphorylation in the ino80 mutant by different research groups has shown dissimilar results. On one hand, Shimada et al. described a delayed Rad53 phosphorylation in the ino80 mutant after HU treatment. Moreover, this group also described that Rad53 inactivation is delayed in the

94


ino80 mutant after release and recovery from HU arrest (Shimada et al. 2008). In addition, using 2D gel analysis to follow replication fork movement in cells released from HU treatment Shimada et al. observed a small, yet, significant delay in replication progression that could be due to the delayed effect on Rad53 phosphorylation. On the other hand, Falbo et al. found no significant differences in the Rad53 phosphorylation pattern between wild-type and ino80 mutants. More importantly, 2D gel analysis of replication intermediates that are generated only after fork’s collapse, a well-established method to test for replication fork collapse at specific ARS, indicates that unlike the rad53 mutant, the ino80 mutant shows no sign of fork collapse, such as the presence of a cone signal or the failure to repress late firing origins (Falbo et al. 2009). Therefore, available evidence indicates that INO80 does not seem to affect replication fork stabilization after HU treatment. As such, the hypersensitivity of the ino80 mutant to HU on plates, which is not connected to the defect in replication fork stabilization shown by 2D gel electrophoresis, is known to be a characteristic of mutants of genes that mediate DNA damage tolerance during replication. Therefore, as it will be discussed in the next section, INO80 has a pivotal role in the DNA damage tolerance pathways during replication. 4.6.3.3 INO80 and the DNA Damage Tolerance Pathways In S. cerevisiae, replication forks that encounter an MMS-induced adduct during S phase stall, and the synthesis of the leading and lagging strands becomes uncoupled. Thus, if the obstruction is in the template for the leading strand, a gap is formed, and the subsequent excision of this region creates a DSB by destruction of the replication fork leading to H2AX phosphorylation. Moreover, in yeast, bypass of this type of lesions is performed by the RAD18/RAD6-mediated DNA damage tolerance pathway and, similar to the ino80 mutant, mutants of genes involved in this pathway are hypersensitive to MMS (Hoege et al. 2002; Papouli et al. 2005; Watts 2006; Ulrich 2007). In support of a role for INO80 in DNA damage avoidance during replication, H2AX phosphorylation, a DSB marker, is significantly increased in the ino80 mutants when MMS treated, G1 synchronized ino80 mutant cells are released into the S phase (Falbo et al. 2009). Moreover, this phosphorylation depends on cells’ progression through the S phase, which strongly suggests that INO80 is necessary to avoid replication-related damage. Furthermore, analysis of S phase synchronized cells treated with MMS by two different techniques confirms these results. One of these techniques, pulse field gel electrophoresis (PFGE) allows differentiation of replicating DNA vs. non-replicating DNA, since replicating DNA is unable to enter the agarose gel, remaining as a single band on the top. Thus, PFGE analysis of the ino80 and arp8 mutants shows a significant delay in the mutants’ ability to reconstitute their chromosomes, indicating an impaired ability of the mutants to complete DNA replication after MMS treatment. The other technique, DNA combing analysis, that allows direct visualization and quantitation


95

of replicating chromosomes, indicates that the length of the BrdU tracks, indicative of replication fork movement, is significantly reduced in the arp8 mutant after release from MMS (Falbo et al. 2009). Therefore, the altered chromosome mobility of arp8 cells in PFGE after MMS treatment is due to the persistence of unreplicated gaps that represents an impaired ability of the mutants to complete DNA replication. 4.6.3.4 INO80 Chromatin Remodeling Activity is Required for Efficient PCNA Ubiquitylation Mutants of genes involved in the RAD6/RAD18 pathway and the ino80 mutant have similar phenotypes, such as MMS hypersensitivity and an impaired ability to resume DNA replication after MMS treatment. In the budding yeast, the RAD6/ RAD18 damage tolerance pathway is dependent upon PCNA. After MMS treatment, PCNA is monoubiquitylated at its K164 residue by Rad6-Rad18, and subsequently, K164 is polyubiquitylated by the Mms2-Ubc13-Rad5 enzyme complex. Thus, PCNA mono and/or polyubiquitylation are the main cellular mechanisms to resolve replication fork obstructions that could potentially lead to DNA damage (Ulrich 2007). Remarkably, INO80’s impaired ability to resume replication fork movement after MMS treatment seems to be related to a defective PCNA ubiquitylation. In fact, pull down from trichloroacetic acid extracted proteins followed by western blot analysis using an anti-Ub antibody in the ino80 mutant, shows that in S phase synchronized MMS-treated cells, the ubiquitylated forms of PCNA are significantly lower in abundance as compared to the wild-type strain. Moreover, the same effect is observed in a point mutant (K737A) that specifically abolishes the ATPase activity of INO80, indicating that the INO80 ATPase activity is required for PCNA ubiquitylation. Furthermore, transcriptional microarray analysis of S phase synchronized MMS-treated ino80 mutants shows no significant regulation of genes involved in the DNA damage tolerance pathways, which supports a direct role of INO8O during this process (Falbo et al. 2009). 4.6.3.5 INO80 is Required for the Formation of Rad51-Dependent Recombination Intermediates Induced by MMS Treatment After MMS treatment, Rad51 is recruited to stalled forks to perform recombinationmediated activities conducing to fork resolution and completion of DNA replication. Rad51 activity at forks generates DNA hemicatenate-like structures called X-shaped structures. Moreover, Rad51-mediated accumulation of these structures depends on PCNA polyubiquitylation and Rad18 recruitment to the fork, indicating that Rad18 and Rad51 work in conjunction to promote lesion bypass by a template switching mechanism that uses the information in the newly synthesized chromatid (Branzei et al. 2008).

96


Concordantly with the reduced PCNA ubiquitylation in the ino80 mutant, chromatin immunoprecipitation analysis of several ARS in S phase synchronized, MMS-treated cultures shows that Rad18 recruitment is significantly reduced in the ino80 mutant (Falbo et al. 2009). Thus, it is possible that INO80 binds to replication forks, and remodels the chromatin to allow proper recruitment of DNA damage tolerance factors like Rad18 and Rad51. Moreover, Rad51 recruitment to ARS is, in fact, significantly reduced in the ino80 mutant, when assessed by chromatin immunoprecipitation analysis. Thus, as explained previously, improper recruitment of Rad51 leads to impaired recombination and the consequent reduction in the generation of recombination intermediates, as it can be concluded by comparison of the arp8 mutant and wild-type cells using 2D gel electrophoresis analysis (Falbo et al. 2009). Therefore, INO80 is necessary to recruit Rad18 and Rad51 to origins of replication when forks face MMSinduced adducts. In conclusion, INO80 is a novel regulator of DNA damage tolerance during replication, through its ability to influence the recruitment of factors in both the RAD6 and RAD51 pathways. In S. cerevisiae, INO80 binds to origins of replication and allows the proper recruitment of proteins involved in replication fork resolution, probably through its ATPase-dependent remodeling activity. Thus, INO80 allows recruitment of Rad18 that leads to PCNA ubiquitylation, which is necessary for Rad51 recruitment. Rad51 recruitment leads to activation of recombinationmediated processes that possibly, as part of a template switching mechanism, will lead to fork resolution and DNA replication completion without the generation of DSBs, genomic aberrations, and genomic instability (Fig. 4.3).

Rad18 DNA adduct

PCNA

INO80

Chromatin Remodeling?

INO80

PCNA

PCNA

Rad18 Rad51

Ub Progressing replication fork

Rad51

Template switching Replication resumption

Fig. 4.3 INO80 is involved in the DNA damage tolerance pathway during DNA replication. When replication forks encounter obstructions caused by DNA damage the INO80 chromatin remodeling complex is recruited to blocked replication forks. Then, INO80 remodels the chromatin environment to facilitate the recruitment of factors from both the RAD6 and RAD51 pathways, such as Rad18 and Rad51. These initiating factors activate subsequent pathways to avoid DNA damage that could arise as a consequence of improper handling of stalled replication forks


97

4.7 Conclusions and Perspectives The past few years have witnessed an accelerated phase in the discovery of chromatin regulation in eukaryotic organisms. From epigenetic modifications to nucleosome mobilization, it is now clear that chromatin structure and its respective modifiers are essential players in a multitude of DNA-related activities. Specifically, chromatin remodeling complexes, through their ability to remodel the structure of chromatin, act as DNA gatekeepers, granting or denying access to the molecule and, thus, regulating many nuclear functions. In particular, one remodeling complex, INO80, has been implicated in several unrelated novel nuclear activities, constituting a good example of the extreme versatility these complexes have achieved through evolution. INO80 is involved not only in transcription, but also in DNA damage-related activities, telomere maintenance and DNA replication, and an important clue to this multitasking skill is INO80’s structure and composition. Yet, it is known that each different subunit has its own specific function, but it is also clear that they work together to perform different tasks. A relevant example of INO80’s multitasking skills is the role of INO80 in DNA damage-related activities. INO80 is involved in both DSB repair and DNA damage avoidance during DNA replication, two processes that, despite their sharing of some mechanisms, are radically different in nature. In brief, the Nhp10 subunit is required during conventional DSB due to its ability to bind gH2AX, but it is not required for INO80’s role in DNA damage tolerance, since, for example, lack of Nhp10 does not affect fork recovery after MMS treatment as measured by PFGE. In addition, the presence of posttranslational modifications in some subunits adds a different level of complexity to the picture, related to the possible regulatory mechanisms that could be directing INO80 activities. Indeed, posttranslational modifications might increase INO80’s opportunities to participate in multiple and different processes, since these marks could constitute a code that, for example, would shift INO80 focus from one activity to another. Clearly, further research would certainly advance our knowledge of this complex as well as the intricate relations established among its subunits. Interestingly, the INO80 remodeling complex is highly conserved from yeast to humans, and in fact, an INO80 homolog was recently described in humans, leading to the question of whether the INO80 complex might have similar functions during DDR in humans (Cai et al. 2007). In support of this idea, it was described that deletion of the human homolog of INO80 increases a cell’s sensitivity to DNA-damaging agents. In addition, YYI, a transcription factor closely associated with INO80, has the ability to bind recombination intermediate structures (Wu et al. 2007). Therefore, further exploration of this new research area could bring innovative and invaluable tools to improve the knowledge and treatment of diseases such as cancer, where accurate DNA replication and proper activation of DNA damage avoidance pathways have an important impact on the disease progression.

98


References Adkins MW, Tyler JK (2004) The histone chaperone Asf1p mediates global chromatin disassembly in vivo. J Biol Chem 279:52069–52074 Altaf M, Saksouk N, Cote J (2007) Histone modifications in response to DNA damage. Mutat Res 618:81–90 Bao Y, Shen X (2007) INO80 subfamily of chromatin remodeling complexes. Mutat Res 618:18–29 Branzei D, Foiani M (2007) Interplay of replication checkpoints and repair proteins at stalled replication forks. DNA Repair (Amst) 6:994–1003 Branzei D, Foiani M (2009) The checkpoint response to replication stress. DNA Repair (Amst) 8:1038–1046 Branzei D, Vanoli F, Foiani M (2008) SUMOylation regulates Rad18-mediated template switch. Nature 456:915–920 Cai Y, Jin J, Yao T et al (2007) YY1 functions with INO80 to activate transcription. Nat Struct Mol Biol 14:872–874 Downs JA, Lowndes NF, Jackson SP (2000) A role for Saccharomyces cerevisiae histone H2A in DNA repair. Nature 408:1001–1004 Falbo KB, Alabert C, Katou Y et al (2009) Involvement of a chromatin remodeling complex in damage tolerance during DNA replication. Nat Struct Mol Biol 16:1167–1172 Falbo KB, Shen X (2006) Chromatin remodeling in DNA replication. J Cell Biochem 97:684–689 Friedberg EC, Aguilera A, Gellert M et al (2006) DNA repair: from molecular mechanism to human disease. DNA Repair (Amst) 5:986–996 Haber JE (2000) Lucky breaks: analysis of recombination in Saccharomyces. Mutat Res 451:53–69 Harrison JC, Haber JE (2006) Surviving the breakup: the DNA damage checkpoint. Annu Rev Genet 40:209–235 Hoege C, Pfander B, Moldovan GL et al (2002) RAD6-dependent DNA repair is linked to modification of PCNA by ubiquitin and SUMO. Nature 419:135–141 Huertas D, Sendra R, Munoz P (2009) Chromatin dynamics coupled to DNA repair. Epigenetics 4:31–42 Jazayeri A, Falck J, Lukas C et al (2006) ATM- and cell cycle-dependent regulation of ATR in response to DNA double-strand breaks. Nat Cell Biol 8:37–45 Kass-Eisler A, Greider CW (2000) Recombination in telomere-length maintenance. Trends Biochem Sci 25:200–204 Larrivee M, LeBel C, Wellinger RJ (2004) The generation of proper constitutive G-tails on yeast telomeres is dependent on the MRX complex. Genes Dev 18:1391–1396 Lee SE, Moore JK, Holmes A et al (1998) Saccharomyces Ku70, mre11/rad50 and RPA proteins regulate adaptation to G2/M arrest after DNA damage. Cell 94:399–409 Ljungman M (2009) Targeting the DNA damage response in cancer. Chem Rev 109:2929–2950 Lombard DB, Guarente L (2000) Nijmegen breakage syndrome disease protein and MRE11 at PML nuclear bodies and meiotic telomeres. Cancer Res 60:2331–2334 Luger K, Mader AW, Richmond RK et al (1997) Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature 389:251–260 Lundblad V, Blackburn EH (1993) An alternative pathway for yeast telomere maintenance rescues est1- senescence. Cell 73:347–360 Lundblad V, Szostak JW (1989) A mutant with a defect in telomere elongation leads to senescence in yeast. Cell 57:633–643 Misri S, Pandita S, Kumar R et al (2008) Telomeres, histone code, and DNA damage response. Cytogenet Genome Res 122:297–307 Mizuguchi G, Shen X, Landry J et al (2004) ATP-driven exchange of histone H2AZ variant catalyzed by SWR1 chromatin remodeling complex. Science 303:343–348


99

Moore JK, Haber JE (1996) Cell cycle and genetic requirements of two pathways of nonhomologous end-joining repair of double-strand breaks in Saccharomyces cerevisiae. Mol Cell Biol 16:2164–2173 Morrison AJ, Highland J, Krogan NJ et al (2004) INO80 and gamma-H2AX interaction links ATP-dependent chromatin remodeling to DNA damage repair. Cell 119:767–775 Morrison AJ, Kim JA, Person MD et al (2007) Mec1/Tel1 phosphorylation of the INO80 chromatin remodeling complex influences DNA damage checkpoint responses. Cell 130:499–511 Morrison AJ, Shen X (2006) Chromatin modifications in DNA repair. Results Probl Cell Differ 41:109–125 Morrison AJ, Shen X (2009) Chromatin remodelling beyond transcription: the INO80 and SWR1 complexes. Nat Rev Mol Cell Biol 10:373–384 Nakada D, Hirano Y, Sugimoto K (2004) Requirement of the Mre11 complex and exonuclease 1 for activation of the Mec1 signaling pathway. Mol Cell Biol 24:10016–10025 Nakamura TM, Moser BA, Russell P (2002) Telomere binding of checkpoint sensor and DNA repair proteins contributes to maintenance of functional fission yeast telomeres. Genetics 161:1437–1452 Osley MA, Shen X (2006) Altering nucleosomes during DNA double-strand break repair in yeast. Trends Genet 22:671–677 Papouli E, Chen S, Davies AA et al (2005) Crosstalk between SUMO and ubiquitin on PCNA is mediated by recruitment of the helicase Srs2p. Mol Cell 19:123–133 Pennock E, Buckley K, Lundblad V (2001) Cdc13 delivers separate complexes to the telomere for end protection and replication. Cell 104:387–396 Pisano S, Galati A, Cacchione S (2008) Telomeric nucleosomes: forgotten players at chromosome ends. Cell Mol Life Sci 65:3553–3563 Powell SN, Bindra RS (2009) Targeting the DNA damage response for cancer therapy. DNA Repair (Amst) 8:1153–1165 Prado F, Cortes-Ledesma F, Aguilera A (2004) The absence of the yeast chromatin assembly factor Asf1 increases genomic instability and sister chromatid exchange. EMBO Rep 5:497–502 Redon C, Pilch DR, Rogakou EP et al (2003) Yeast histone 2A serine 129 is essential for the efficient repair of checkpoint-blind DNA damage. EMBO Rep 4:678–684 Rogakou EP, Pilch DR, Orr AH et al (1998) DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. J Biol Chem 273:5858–5868 Seto AG, Livengood AJ, Tzfati Y et al (2002) A bulged stem tethers Est1p to telomerase RNA in budding yeast. Genes Dev 16:2800–2812 Shen X, Mizuguchi G, Hamiche A et al (2000) A chromatin remodelling complex involved in transcription and DNA processing. Nature 406:541–544 Shen X, Ranallo R, Choi E et al (2003a) Involvement of actin-related proteins in ATP-dependent chromatin remodeling. Mol Cell 12:147–155 Shen X, Xiao H, Ranallo R et al (2003b) Modulation of ATP-dependent chromatin-remodeling complexes by inositol polyphosphates. Science 299:112–114 Shimada K, Oma Y, Schleker T et al (2008) Ino80 chromatin remodeling complex promotes recovery of stalled replication forks. Curr Biol 18:566–575 Shrivastav M, De Haro LP, Nickoloff JA (2008) Regulation of DNA double-strand break repair pathway choice. Cell Res 18:134–147 Shroff R, Arbel-Eden A, Pilch D et al (2004) Distribution and dynamics of chromatin modification induced by a defined DNA double-strand break. Curr Biol 14:1703–1711 Staynov DZ (2008) The controversial 30 nm chromatin fibre. Bioessays 30:1003–1009 Taggart AK, Teng SC, Zakian VA (2002) Est1p as a cell cycle-regulated activator of telomerebound telomerase. Science 297:1023–1026 Takata H, Tanaka Y, Matsuura A (2005) Late S phase-specific recruitment of Mre11 complex triggers hierarchical assembly of telomere replication proteins in Saccharomyces cerevisiae. Mol Cell 17:573–583 Teng SC, Zakian VA (1999) Telomere-telomere recombination is an efficient bypass pathway for telomere maintenance in Saccharomyces cerevisiae. Mol Cell Biol 19:8083–8093

100


Tsukuda T, Fleming AB, Nickoloff JA et al (2005) Chromatin remodelling at a DNA doublestrand break site in Saccharomyces cerevisiae. Nature 438:379–383 Tsukuda T, Lo YC, Krishna S et al (2009) INO80-dependent chromatin remodeling regulates early and late stages of mitotic homologous recombination. DNA Repair (Amst) 8:360–369 Tyler JK (2002) Chromatin assembly. Cooperation between histone chaperones and ATPdependent nucleosome remodeling machines. Eur J Biochem 269:2268–2274 Ulrich HD (2007) Conservation of DNA damage tolerance pathways from yeast to humans. Biochem Soc Trans 35:1334–1337 van Attikum H, Fritsch O, Hohn B et al (2004) Recruitment of the INO80 complex by H2A phosphorylation links ATP-dependent chromatin remodeling with DNA double-strand break repair. Cell 119:777–788 van Gent DC, Hoeijmakers JH, Kanaar R (2001) Chromosomal stability and the DNA doublestranded break connection. Nat Rev Genet 2:196–206 Vernet G, Sala-Rovira M, Maeder M et al (1990) Basic nuclear proteins of the histone-less eukaryote Crypthecodinium cohnii (Pyrrhophyta): two-dimensional electrophoresis and DNA-binding properties. Biochim Biophys Acta 1048:281–289 Virta-Pearlman V, Morris DK, Lundblad V (1996) Est1 has the properties of a single-stranded telomere end-binding protein. Genes Dev 10:3094–3104 Watts FZ (2006) Sumoylation of PCNA: wrestling with recombination at stalled replication forks. DNA Repair (Amst) 5:399–403 Wu S, Shi Y, Mulligan P et al (2007) A YY1-INO80 complex regulates genomic stability through homologous recombination-based repair. Nat Struct Mol Biol 14:1165–1172 Yu EY, Steinberg-Neifach O, Dandjinou AT et al (2007) Regulation of telomere structure and functions by subunits of the INO80 chromatin remodeling complex. Mol Cell Biol 27:5639–5649

Chapter 5

Caenorhabditis elegans Radiation Responses Aymeric Bailly and Anton Gartner

Abstract Over the past 10 years a number of laboratories have started to focus on Caenorhabditis elegans radiation responses, taking advantage of a multi-cellular experimental model system that enables studying DNA damage responses at the organismal level. Here we provide a comprehensive review of C. elegans DNA damage responses, largely focusing on recombinational repair, DNA damage signalling and DNA damage-induced apoptosis in response to ionizing radiation. To better explain C. elegans DNA damage response phenotypes and DNA damage response pathway, we also provide an introduction to the C. elegans life cycle and indicate key experimental procedures. Keywords Germ cell • Nematode

5.1 Introduction The use of a model organism to dissect fundamental biological processes has been instrumental in the establishment of various fields in biology. The ionizing radiation (IR) response is no exception, and early studies focused on the effect of DNA damage on single cell organisms such as yeasts. Using the budding yeast Saccharomyces cerevisiae as a model system has been a powerful method to identify genes required for DNA damage repair. Most of the key factors in this process have been isolated by performing forward genetic screens for mutants sensitive to DNA-damaging agents (Weinert and Hartwell 1993). The majority of those genes are conserved in higher eukaryotic organisms, and some alleles lead to heritable cancer predisposition syndromes in the human population (Bartek et al. 2007). However, yeast cells lack an important aspect of the DNA damage response present in higher eukaryotic organisms – the removal of genetically compromised cells by programmed cell A. Gartner (*) Wellcome Trust Centre for Gene Regulation and Expression, University of Dundee, Dow Street, Dundee DD1 5EH, UK e-mail: [email protected] T.L. DeWeese and M. Laiho (eds.), Molecular Determinants of Radiation Response, Current Cancer Research, DOI 10.1007/978-1-4419-8044-1_5, © Springer Science+Business Media, LLC 2011

101

102

A. Bailly and A. Gartner

death (apoptosis). Furthermore, DNA damage response and repair pathways are more complex in higher organisms, and diverse cell types differentially respond to DNA damage. It is still technically challenging to perform forward genetic screens with vertebrate cell lines, and large-scale RNAi screening procedures do not permit the generation of stable mutations or hypomorphic alleles in essential genes. Therefore, the use of a model system that recapitulates the integral DNA damage response of mammalian cells but concurrently provides the ease of maintenance and manipulation of a small and simple organism is of great interest to the scientific community. Over the past years an increasing number of laboratories have engaged in establishing and exploiting the nematode worm Caenorhabditis elegans as an experimental system to study DNA damage response genes. Before going into the details of C. elegans DNA damage response pathways, it is important to summarize the basic biological features of this nematode, as well as to explain experimental approaches unique to the nematode system. A basic knowledge of worm physiology and life cycle is also necessary to comprehend the various strategies used by somatic and germ cell tissues to respond to DNA damage, as well as to understand the basic assays that are used to assess defects in DNA damage response pathways.

5.2 The C. elegans Life Cycle and Implications for Radiation Responses C. elegans is a relatively small animal and adults are approximately 1 mm long (Fig. 5.1a). Studies mostly focus on worm development and are aided by the invariant somatic cell lineage. All individual animals display the same cell division patterns to produce identical groups of differentiated cells, and the complete process of embryogenesis can be monitored by light microscopy. The pioneering work of Sulston and co-workers established the complete map of the C. elegans embryonic cell lineage (Sulston et al. 1983). Knowledge of the invariant lineage enables using the worm to ask basic biological questions, such as the molecular details of inductive developmental processes at the level of individual cells. Upon fertilization, the first cleavage divisions produce five somatic founder cells and one germline precursor cell (the P cell) (Sulston et al. 1983), and different adult tissues are produced according to a highly complex lineage pattern. During embryogenesis, which takes less than 12 h, 556 somatic cells are generated and two cells act as germ precursor cells (Sulston et al. 1983). Embryonic cell divisions occur very rapidly, especially during early development where some cell divisions occur in less than 15 min. Embryogenesis is followed by four larval stages, termed the L1 to L4 stages, before worms develop into fertile adults, 50–70 h after fertilization. During larval development, further cell divisions occur in somatic tissues, especially in the ventral cord and during development of the vulva. However, most cell proliferation happens in the C. elegans germline. The germline consists of only two germ cells at the L1 stage, which expand into two gonad arms each comprised of ~500 cells each in the adult worm (Sulston et al. 1983) (Fig. 5.1a). In the adult all somatic cells are post-mitotic and the germline is the only proliferative tissue. This dichotomy in proliferative capacity is important when

Fig. 5.1 (a) Adult Caenorhabditis elegans hermaphrodite (upper panel). Schematic representation of a C. elegans germline (lower panel). The distal end of the worm germline is comprised of mitotic cells. Germ cells enter meiosis at the “transition zone.” After passing though the transition zone most germ cells are in early and late pachytene; late pachytene germ cells can undergo apoptosis. (b) Apoptotic germ cells, highlighted by arrows appear as bottom-like structures under Nomarski optics. (c) CED-1::GFP accumulation around dying cells during the corpse engulfment highlights earlystage apoptotic cells. (d) Cell cycle arrest phenotype. Nuclei (arrow) and the surrounding cytoplasm are enlarged in irradiated cells due to continued cellular growth in the absence of cell division

104


considering radiation responses in adult worms. It appears that germ cells and somatic cells are subject to divergent selective pressures. Somatic cells are optimized to contribute to the fitness for only one generation, whereas germ cells are optimized to maintain proliferative potential and to ensure the accurate transmission of the genetic material. It is thus not surprising that somatic tissues tolerate higher levels of DNA damage and use other DNA repair strategies compared to germ cells. Germlinespecific checkpoints and accurate recombinational repair are used to guard germ cells from acquiring deleterious mutations that could be passed on to the next generation. Germ cell proliferation occurs in a stem cell compartment, and germ cell divisions are regulated by the somatic distal tip cell (DTC) located at the tip of the mitotic region, which forms a stem cell niche that constantly restocks the germ cell popula tion (Kimble and White 1981) (see Fig. 5.1a). Within this mitotic germ cell compartment cell divisions occur much more slowly compared to embryonic divisions, and estimates for cell cycle timing range from 12 to 20 h, contrasting with cell division times of less than 15 min during early embryogenesis (Crittenden et al. 2006; Crittenden and Kimble 2008). Germ cell nuclei are actually not fully cellularized and are thus part of a syncytium, although for convenience they are referred to as cells both in this review and generally in the C. elegans germ cell field (Hirsh et al. 1976). The mitotic stem cell compartment (commonly referred to as the mitotic zone) is followed by the so-called transition zone where events related to the entry into meiotic prophase occur, such as the formation of DNA double-strand breaks (DSBs) and the initiation of meiotic chromosome pairing (Dernburg et al. 1998) (Fig. 5.1a). Proximal to the transition zone most cells are in meiotic pachytene; homologous chromosomes are tightly aligned to each other forming the synaptonemal complex. Germ cells subsequently complete meiosis and concomitantly undergo cellularization. C. elegans is a hermaphrodite, and thus both sperm cells and oocytes must be produced in the same animal. Worms achieve this by temporally switching the sex of the germ line (Hodgkin and Brenner 1977). The germline acquires a male fate during the L4 larval stage and all germ cells differentiate into amoeboid sperm cells, stored in an organ termed the spermatheca. Oocytes are produced when the germline switches to its female fate towards the end of the L4 larval stage. In adult worms, fertilization occurs when oocytes pass through the spermatheca, allowing oogenesis to occur at the proximal end of the gonad (Fig. 5.1a). Failure to undergo meiotic recombination does not lead to overt changes in germ cell identity or morphology, and meiotic chromosome pairing is not affected. Mutants defective in recombinational repair of DNA DSBs often also show defects in meiotic recombination, leading to defects in meiotic crossover formation and meiotic chromosome segregation, and resulting in embryonic lethality in the following generation.

5.3 The C. elegans as an Experimental System C. elegans is a unique and versatile experimental system, which is easy to propagate and amenable to long-term storage. Nematodes can be routinely propagated on bacteria on petri dishes, or grown in liquid culture to obtain large quantities of

5 Caenorhabditis elegans Radiation Responses

105

worms amenable for biochemical analysis, and are thus an inexpensive and scalable experimental system. Further, worms are tractable to long-term storage as frozen stocks can maintain their viability, which facilitates the development of worm strain collections. Shorter-term storage is also possible due to the ability of worms to differentiate into starvation-resistant dauer larvae, which enables them to survive up to several months under normal culture conditions. As C. elegans worms are predominantly hermaphrodites, self-fertilization results in the easy maintenance of isogenic populations. Genetic crosses in worms employ males, which arise spontaneously from hermaphrodite populations by sex chromosome non-disjunction. Furthermore, C. elegans was the first multi-cellular organism to be fully sequenced (1998) and also made headlines for uncovering the basic apoptosis pathway (Sulston and Horvitz 1977; Ellis and Horvitz 1986; Hengartner et al. 1992; Conradt and Horvitz 1998), for the discovery of RNAi (Fire et al. 1998) and for being the first animal where ectopic green fluorescent protein (GFP) fusion proteins were expressed (Chalfie et al. 1994). Over the past four decades since C. elegans was first introduced as a model system (Brenner 1974), the worm research community has grown and produced invaluable resources, such as a centralized strain depository, gene knock-out mutant collections, and genome-wide RNAi libraries. Indeed, C. elegans is unique as RNAi can be applied to the entire organism. This is conveniently done by expressing double-stranded RNA corresponding to both strands of a single worm gene in Escherichia coli, ingestion of which leads to specific gene inactivation (Kamath et al. 2001). Such RNAi feeding procedures can be performed in liquid cultures using a 96-well microtiter plate format, allowing such screens to be performed on a genome-wide scale (Kamath et al. 2003; Lee et al. 2003). Candidate genes identified using this procedure can then be validated by analyzing the corresponding genetic mutations. In addition to the revolution brought about by the advent of genome-wide RNAi screens, it is predicted that the C. elegans DNA damage response field will benefit tremendously from new technology driving the revival of classic genetic mutagenesis and screening approaches. While in the past the community was hampered by the bottleneck associated with tedious and time-consuming positional cloning approaches, next-generation sequencing, which already allows for re-sequencing multiple C. elegans strains in a single sequencing reaction, will revolutionize this approach (Hillier et al. 2008; Sarin et al. 2008; Shendure and Ji 2008).

5.4 Basic Phenotypes Associated with DSB Repair and DNA Damage-Signalling Defects Pioneering work conducted by Hartmann (Hartman and Herman 1982) first aimed to use C. elegans as a model system to isolate genes implicated in DNA damage. Nine radiation-sensitive genes, rad-1 to rad-9 were identified by their ability to protect worms against UV light during embryogenesis. Loss-of-function mutations in these genes had detrimental effects on worm viability under UV exposure. Some of these mutants were also sensitive to IR. Following these initial studies, it was

106


almost two decades before work on those mutants was resumed and the first rad genes were mapped and positionally cloned (Ahmed et al. 2001) (see below). Even earlier, mutants were isolated that showed defects in meiotic chromosome segregation, a phenotype arising as a consequence of defects in meiotic chromosome pairing and recombination. Defective meiotic chromosomes segregation leads to embryonic lethality and an enhanced incidence of males due to sex chromosome non-disjunction. We now know that many of these “him” (high incidence of males) mutants are defective in meiotic recombination, and show an enhanced sensitivity to IR. The first recombinational repair gene isolated by mapping and positional cloning of a him mutant was him-6, the C. elegans orthologue of the mammalian Blooms syndrome gene (Wicky et al. 2004). C. elegans genetics has been instrumental in working out the conserved core apoptotic pathway. The early studies described above took advantage of the invariant development of the worm and the discovery that 131 cells die during the development of the worm by apoptosis (Sulston et al. 1983). Further studies on the germline revealed that an average of one to three corpses can be detected in each gonad arm at any given time. Taking the rapid engulfment of those corpses into account, it became apparent that every second germ cell is eliminated by so-called physiological germ cell apoptosis (Gumienny et al. 1999). However, preliminary attempts to study DNA damage-induced apoptosis failed, as IR-induced apoptosis could not be detected in either embryos or the somatic tissues of adult worms. Given that the germline is the only proliferative tissue of the adult worm and may therefore be relatively sensitive to damage, it was then asked whether DNA damage (caused by DNA DSB-inducing agents such as IR) was able to induce germ cell apoptosis. This was indeed found to be the case. A key study focusing on C. elegans DNA damage response phenotypes established that DNA damage activates apoptotic pathways in the germline, and revealed that germline apoptosis is induced in a dose-dependent manner (Gartner et al. 2000). Using this system, up to thirty dying cells can be counted 24 h after a dose of 120 Gy IR. These studies have exploited the unique opportunity provided by C. elegans to directly observe apoptotic cells in a living multi-cellular organism (Gumienny et al. 1999). Irradiated worms can be transferred individually onto a slide covered with agarose, anesthetized with levamisole, and analyzed using a microscope enabled for Nomarski optics (Gartner et al. 2004). Under these conditions, germline apoptotic cells appear as button-like structures (Fig. 5.1b). Alternatively, apoptotic cells can be visualized using live cell dyes such as acridine-orange, which visualizes the reduced intracellular pH associated with apoptotic dying cells. Finally, apoptotic cells can be visualized by the characteristic pattern of ectoptic CED-1::GFP localization around dying cells during the engulfment process (Fig. 5.1c) (Schumacher et al. 2005a). CED-1 is a transmembrane receptor related to the human SREC scavenger receptor and is expressed at the surface of engulfing cells that face the apoptotic corpses (Zhou et al. 2001). It is now evident that radiation-induced germ cell apoptosis relies on the same core apoptotic genes that are required for developmental and physiological germ cell apoptosis (Gartner et al. 2000). The CED-3 caspase and the CED-4 Apaf1-like


107

molecules are both required to induce apoptosis, while the CED-9 Bcl-2 homologue prevents cells from undergoing apoptosis (Ellis and Horvitz 1986; Hengartner et al. 1992). In addition, the transcriptional induction of the C. elegans BH3 domain protein EGL-1, whose cell type-specific transcriptional induction is necessary for C. elegans somatic developmental apoptosis, is also required for DNA damageinduced germ cell apoptosis (Conradt and Horvitz 1998; Gartner et al. 2000) (Fig. 5.2). EGL-1 is not required for physiological germ cell apoptosis, which is not dependent on DNA damage (Gumienny et al. 1999). IR-induced EGL-1 transcription is mediated by C. elegans CEP-1, the sole worm primordial p53-like transcription factor (Derry et al. 2001; Schumacher et al. 2001). CEP-1 is the subject of extensive current studies and will be described in the last chapter of this review. Interestingly, germ cell apoptosis is restricted to a subset of germ cells, namely cells in the late pachytene stage, a stage of meiotic chromosome pairing where recombination has already occurred, and homologous chromosomes are tightly associated with each other within the synaptonymal complex. DNA damageinduced apoptosis does not occur in any other germ cell types such as the mitotic stem cell compartment, or during embryogenesis. Germ cell apoptosis is also induced by meiotic recombination failure, and either mutations or RNAi-mediated depletion of a number of recombination genes such as rad-51, and the worm homologues of the BRCA1 and BRCA2 genes (Gartner et al. 2000; Alpi et al. 2003; Martin et al. 2005).

ic ce lls

ati

us cle m nu plas o SIR-2.1 t y c

on

?

?

?

CED-9

lls

c slo an

egl-1, ced-13 (BH3)

?

KRI-1

?

Tr

ING-3

ce

SIR-2.1

m

CEP-1

er G

RAD-9,HUS-1,MRT- 2 HRP-17 ATL-1

Checkpoint signaling

RB1

at

m

So

DNA damage

inactive CED-3

?

CED-4

CED-3 ?

GERM CELL APOPTOSIS

active

[ceramide]

Fig. 5.2 Diagram depicting Caenorhabditis elegans germ cell apoptosis pathways in response to ionizing irradiation

108


Thus, it is likely that the primary function of DNA damage-induced apoptosis is to provide a fail-safe mechanism to ensure the elimination of germ cells with recombination defects. Indeed, germ cell apoptosis appears to have a more general role in monitoring apoptosis progression, given that an independent checkpoint leading to the elimination of meiotic germ cells with meiotic chromosome pairing defects has been defined (Bhalla and Dernburg 2005). The importance of germ cell apoptosis induction in cells with meiotic chromosome pairing and a recombination defect becomes evident when partially pairing and recombination-defective mutants are combined with apoptosis-defective mutants. Under those conditions, an enhanced level of lethality is observed when analyzing the viability of embryos that result from such defective germ cells (Rinaldo et al. 2002; Bhalla and Dernburg 2005; Adamo et al. 2008). A second damage-induced germline phenotype is a transient cell cycle arrest that occurs in mitotic germ cells following treatment with DNA-damaging agents such as IR (Gartner et al. 2000; Chin and Villeneuve 2001). Cell cycle arrest induced by DNA damage can be simply observed in living anesthetized worms under Nomarski optics without any treatment, extraction, or fixation steps. Arrested mitotic germ cells can be easily recognized since they appear larger than their unirradiated counterparts (Fig. 5.1d), although quantification of the size increase is best quantified by DAPI staining of dissected germlines. The cellular enlargement occurs because cells stop proliferation due to checkpoint activation but continue to grow. Radiation-induced mitotic cell cycle arrest predominately occurs in the G2 stage, as demonstrated by staining with an antibody that recognizes a conserved epitope corresponding to tyrosine-15 of mammalian CDK1 (Hachet et al. 2007; Moser et al. 2009). Phosphorylation of this tyrosine residue is required to keep CDK-1 inactive before S-phase entry, and is antagonized by the CDC25 phosphatase which activates the CDK-1 kinase, allowing progression into mitosis (Russell and Nurse 1986). Elucidation of a DNA damage checkpoint-dependent phenotype initiated a search for mutants defective in either germline DNA damage-induced apoptosis or cell cycle arrest. The first such screen for apoptotic mutants was undertaken by merely scanning through radiation-sensitive mutants previously identified by the Herman laboratory. The sensitivity of these mutants was first assayed by the L4 survival assay, which was designed to measure the radiation sensitivity of meiotic pachytene cells. For this assay, worms in the fourth larvae stage (L4) are irradiated, allowed to mature into adults, and the survival rate of embryos laid by the adult worms on the following day is assessed after approximately 48 h. As cells progress from the pachytene stage to fertilized embryos in approximately 1 day, this assay measures the extent of repair of meiotic cells residing in the pachytene stage. Typically, 50% of the progeny of wild-type worms will die while the other half will develop into fertile adults for 60 Gy of irradiation. DSB-defective repair mutants, however, fail to develop into adults under those conditions. A slightly modified assay measures the radiation sensitivity of proliferative mitotic germ cells. In this assay, worms at the L1 larval stage are irradiated. These can be conveniently isolated from a developmentally asynchronous liquid culture by filtration through an


109

11-mm filter. Early L1 stage worms only have two germ cells. These need to proliferate in the following four larval stages to generate the pool of approximately 500 cells contained in each of the two arms of the adult gonad. Thus, L1 larval stage germ cells are very sensitive to IR, which at a high dose leads to sterility and the lack of viable progeny in the next generation.

5.5 C. elegans DNA Damage Response Signalling The first mutations associated with defects in radiation-induced apoptosis turned out to impair all DNA damage responses, and were later mapped to the mrt-2, clk-2, and hus-1 loci (Ahmed and Hodgkin 2000; Ahmed et al. 2001; Hofmann et al. 2002). mrt-2 and hus-1 also function in telomere replication (see below), and both have homologues in mammals and yeasts. The proteins corresponding to mrt-2 and hus-1 form part of the conserved 9-1-1 PCNA-like DNA-sliding complex and MRT-2 is related to fission yeast and human RAD1. The DNA damage-specific clamp-loader, comprising RAD17 and the four smallest RFC subunits, recruits this complex to the dsDNA–ssDNA transition of resected DSBs (Kostrub et al. 1998; Green et al. 2000; Venclovas and Thelen 2000; Bermudez et al. 2003; Majka and Burgers 2003). The worm homologues of the mammalian ATM and ATR kinases, which act as upstream DNA damage sensors and are mutated in human genome instability syndromes, have also been implicated in DNA damage checkpoint signalling. Worm ATR (referred to as ATL-1) and CHK1 are needed for checkpoint responses to IR, UV and replication stress. Both are also essential for germ cell proliferation, possibly by being required for faithful DNA replication (Kalogeropoulos et al. 2004; Garcia-Muse and Boulton 2005). As expected, ATL-1 accumulates in DNA damage foci upon irradiation or DNA replication stress. ATM appears to play a more minor role in the C. elegans DNA damage checkpoint response, as atm-1 mutants are only partially defective in responding to IR, and progeny survival is reduced only upon high doses of IR (Stergiou et al. 2007). Mutants of the C. elegans Werner Syndrome helicase wrn-1 share the partial cell cycle arrest phenotype of atm-1 mutants in response to IR (Lee et al. 2010). It is possible that WRN-1 acts upstream of ATM-1 because ATM-1 nuclear accumulation and RPA-1 focus formation in response to IR are defective in wrn-1 mutants (Lee et al. 2010). The first conserved DNA damage checkpoint gene uncovered by C. elegans genetics was clk-2. The mn159 allele was found in an initial screen for mutations conferring increased lethality to the progeny of irradiated germ cells and was originally named rad-5 (mn159), while clk-2 (qm37) was found in a screen for mutations with slow development and an extended life span (Lakowski and Hekimi 1996; Ahmed et al. 2001). However, clk-2 (qm37) lifespan extension is weak, and may derive indirectly from its slow growth phenotype. Both clk-2 mutations are temperaturesensitive, resulting in embryonic lethality associated with checkpoint defects in response to IR, UV and DNA replication stress at the restrictive temperature (Moser et al. 2009). CLK-2 is the orthologue of S. cerevisiae Tel2, which has been

110


implicated in regulation of yeast telomere length, and which binds telomeric DNA in vitro (Lim et al. 2001). In contrast, C. elegans clk-2 does not seem to affect telomere length. Several studies indicated that CLK2/TEL2 is functionally conserved between yeasts and humans. The fission yeast homologue has been implicated in S-phase regulation and DNA damage checkpoint responses (Shikata et al. 2007). Human CLK2/TEL2 is required for both the replication checkpoint and DNA cross-link repair pathways, where it affects the DNA damage-induced mono-ubiquitination of the FancD2 repair protein. CLK2/TEL2 appears to promote S-phase checkpoint activation by preventing CHK1 degradation (Collis et al. 2007). Intriguingly, human and fission yeast CLK2/TEL2 was recently shown to bind to all PI3 kinases, which include TOR2, ATM and ATR (Takai et al. 2007). It was argued that CLK2/ TEL2 might function by stabilizing all PI3 kinases, as their protein levels were dramatically reduced upon CLK2/TEL2 depletion. However, the notion that CLK2/ TEL2 might act in the DNA damage response pathway via the regulation of PI3 kinases was contended in recent studies using the budding yeast model (Anderson and Blackburn 2008; Anderson et al. 2008). Furthermore, ATR/ATL-1 and CLK-2 depletion lead to opposite phenotypes during C. elegans embryogenesis (Moser et al. 2009). In addition, while both C. elegans CLK-2 and ATR/ATL-1 are required for germ cell proliferation, germ cell cycle arrest phenotypes are distinct. clk-2 mutant germ cells arrest in a G2-like stage with condensed chromosomes while atl-1 mutation results in excessive genome instability. CLK2/TEL2 does not contain an obvious catalytic domain, but it is now clear that CLK2/TEL2 belongs to the ARM repeat superfamily of structurally related proteins. Tandem ARM repeats form a super-helical fold capable of forming a surface for protein–protein interactions. ARM repeat proteins are structurally related to proteins containing tandem HEAT motifs. The demonstrated interactions between TEL2/CLK2 and the HEAT repeat containing PI3 kinases suggest that TEL2/CLK2 might act as an adaptor protein that may impinge on multiple signalling pathways (Takai et al. 2007). Another conserved gene that was first implicated in DNA damage checkpoint signalling using C. elegans is gen-1 (Bailly et al. 2010). GEN1 belongs to the XPG family of endonucleases (Clarkson 2003) and is the worm orthologue of the recently purified GEN1 Holliday junction-resolving enzyme (Ip et al. 2008). Holliday junction-resolving enzymes are required for the resolution of Holliday junctions, four-way DNA structures that are crucial intermediates of homologous recombination (HR) and which have to be resolved for separation of the two DNA double strands, in order to complete recombination (Lilley and White 2001). A gen-1 mutant was isolated in a screen for radiation-sensitive mutants also defective in both checkpoint-mediated apoptosis and cell cycle arrest in response to IR. Positional cloning revealed that the mutant affected the worm homologue of the human GEN1-resolving enzyme. C. elegans GEN-1 facilitates repair of DNA DSBs but surprisingly is not essential for meiotic recombination. The role of the meiotic resolving enzyme is apparently taken over by the SLX-1/SLX-4 nuclease complex in worms (Saito et al. 2009). The mammalian homologues of this nuclease complex have been shown to exhibit Holliday junction resolvase activity in vitro (Fekairi et al. 2009;


111

Munoz et al. 2009; Svendsen et al. 2009), and the phenotypes of fly and worm SLX-4 mutants are consistent with a defect in meiotic Holliday junction resolution (Yildiz et al. 2002; Andersen et al. 2009). Mutational analysis of C. elegans GEN-1 reveals that the DNA damage-signalling function of GEN-1 is separable from its role in DNA repair, and epistasis analysis indicates that GEN-1 defines a DNA damagesignalling pathway acting in parallel to the canonical pathway mediated by CHK-1 phosphorylation and induction of CEP-1/p53-mediated apoptosis. Furthermore, GEN-1 acts redundantly with the aforementioned 9-1-1 complex to ensure genome stability in response to endogenous replication stress. It is intriguing to speculate that GEN-1 might coordinate DNA damage-signalling and DNA DSB repair. It is possible that GEN-1 acts as a dual-function protein to resolve Holliday junctions, but also to signal and maintain cell cycle arrest until every Holliday junction is resolved (Bailly et al. 2010). It will be interesting to see if either a DNA repair function or a DNA damage-signalling function will be found for human GEN1.

5.6 C. elegans DSB Repair The major repair pathway used in the C. elegans germline to repair IR-induced DSB is the HR pathway. Many of the components of this pathway are used for DSB repair as well as for meiotic recombination. Indeed, many genes involved in recombinational repair are preferentially expressed in the germline. In contrast, nonhomologous end joining (NHEJ) appears to be the predominant repair pathway in somatic tissues (Clejan et al. 2006). Mutations in C. elegans NHEJ genes such as homologues of lig-4 or cku-70 exhibit developmental defects when embryos are irradiated, as well as fused somatic nuclei, indicative of a repair defect (Clejan et al. 2006). Nevertheless, reporter constructs that reconstitute functional LacZ by recombinational repair indicate that recombinational DSB repair contributes to DSB repair in somatic tissues during development (Pontier and Tijsterman 2009). The same study also indicates that the single-strand annealing pathway as well and alternative NHEJ pathway also contribute to DSB repair in somatic cells. HR is essential for meiotic recombination as well as for DSB repair in response to DSB-generating agents. During meiosis DSBs are generated endogenously by the activity of the conserved nuclease SPO-11 (Dernburg et al. 1998). C. elegans provides an excellent model to study meiotic chromosome pairing and recombination. As part of this review we, however, will focus on recombinational repair in response to IR. A crucial step in the initiation of HR is the processing of the DSB and the generation of a single-stranded DNA tail, which is the molecular substrate of the universal RAD-51 recombinase protein. This nucleolytic event, often referred to as resection, results presumably from the activity of nucleases and/or helicases, both of which might operate in concert (Mimitou and Symington 2009). In C. elegans the MRE-11 nuclease appears to be required for resection, as expected from previous work in budding yeast (Nairz and Klein 1997; Chin and Villeneuve 2001). Similarly, the CtIP nuclease was identified independently in human cells and

112


worms (Penkner et al. 2007; Sartori et al. 2007). CtIP promotes DNA resection in DSB repair following IR exposure in mitotic human cell lines and is required for RAD51 loading (Sartori et al. 2007). Interestingly, the nematode orthologue of CtIP, called com-1, is required for DNA resection during meiotic DSB repair, a phenotype that is similar to the phenotype associated with mutations of the budding yeast homologue sae-2 (Penkner et al. 2007). As expected for a resection-defective phenotype, com-1 mutant worms are defective in RAD-51 loading, although the rate of DSBs generated by SPO-11 appears to be normal. Recombinational DSB repair requires the bona fide DNA strand exchange protein RAD-51, a member of the RecA family (Ogawa et al. 1993; Benson et al. 1994). RAD-51 oligomerizes on the resected ssDNA and catalyzes strand invasion into an intact DNA double-strand template. RAD-51 oligomerization can be detected by immunofluorescence as discrete foci, which are the most important markers for DNA DSBs in C. elegans (Alpi et al. 2003). Similar to the human situation, RAD-51 specifically interacts with BRC-2 (Martin et al. 2005), the orthologue of the human protein BRAC2 which has been implicated as a mediator of RAD51 loading (Galkin et al. 2005). Cytological evidence suggests that BRC-2, like its mammalian counterpart, is required for RAD-51 foci accumulation following exposure to IR (Martin et al. 2005). BRC-2, like RAD-51, is required for the repair of meiotic DSBs generated by SPO-11. BRCA1 was originally identified in the human population as being linked to a heritable form of breast and ovarian cancer, similar to BRCA2 (Futreal et al. 1994; Miki et al. 1994; Wooster et al. 1995). BRCA1-deficient cells are defective for HR, rendering them extremely sensitive to genotoxic treatment (Jasin 2002). In cells subjected to IR treatment, BRCA1 rapidly relocates to sites of DSB repair and forms discrete foci similar to other repair factors, supporting the idea of its direct role in repair (Scully et al. 1997a, b). BRCA1 is part of a heterodimeric complex with BARD1 (Wu et al. 1996). BRCA1 displays an E3 ubiquitin ligase activity in vitro when associated with the E2 enzyme UBC5 (Lorick et al. 1999). However, no substrate of BRCA1/BARD1 has been identified so far and its function in the repair process remains elusive. Nematode orthologues of BRCA1 and BARD1 have been identified (Boulton et al. 2004), and worms carrying either brc-1 or brd-1 mutations share the same phenotype, consistent with a joint function in vivo. These mutants display enhanced p53-dependent apoptosis, compromised survival and chromosome fragmentations following IR treatment, strongly suggesting a DNA repair defect (Boulton et al. 2004). BRC-1 and BRD-1 proteins form discrete structures called foci at sites of DNA damage and co-localize with RAD-51 (Polanowska et al. 2006). Interestingly, it appears that brc-1 might be specifically required for interchromatid repair as opposed to interchromosomal repair in meiotic germ cells (Adamo et al. 2008). Recent publications have provided important insights into novel mechanisms that prevent excessive recombination. The gene encoding for RTEL1 has been identified on the basis of the synthetic lethality associated with its deletion in conjunction with a loss-of-function mutation in the C. elegans Blooms helicase (Barber et al. 2008). Such double mutants confer an increased rate of meiotic recombination


113

and elevated levels of RAD-51 foci. These genetic observations were supported by biochemical experiments, which directly demonstrated that RTEL disrupts RAD-51/ssDNA filaments. Indeed, worms carrying an rtel-1 deletion are hypersensitive to inter-strand cross-linking (ICL) agents such as nitrogen mustard. Interestingly, mutants of the human orthologue RTEL1 display a similar sensitivity towards ICL agents (Barber et al. 2008). The efficiency of using C. elegans to screen for synthetic lethal interactions has been recently highlighted by the identification of the HELQ-1 helicase as a factor required for the disassembly of RAD-51 filaments during meiotic DSB repair (Ward et al. 2010). Prior to this, the critical step of filament disassembly had remained poorly defined at the molecular level. helq-1 was isolated based on a synthetic lethal interaction with the only C. elegans rad-51 paralogue, rfs-1. RFS-1 has been shown to be required for ICL repair (Ward et al. 2007). Worms mutant for both rfs-1 and helq-1 show an accumulation of SPO-11-dependent RAD-51 foci (Ward et al. 2010). Purified RFS-1 and HELQ-1 both interact independently with RAD-51 in vitro and promote the RAD-51 disassembly from duplex, but not singlestranded DNA (Ward et al. 2010). These genetic and biochemical data strongly suggest that both RFS-1 and HELQ-1 function in concert to remove and process RAD-51 filaments during meiosis I. C. elegans genetics can be used to assess whether checkpoint-defective mutants lead to an increased mutation rate in the absence of DNA-damaging agents. For example, using a mutation in the unc-58 locus that leads to an “uncoordinated” phenotype (Hodgkin et al. 1979), it has been shown that the number of revertants in a growing worm population increases 8–15-fold when DSB repair is compromised by mutated checkpoint genes (Harris et al. 2006). Importantly, many of these reversions were small deletions, a finding that could be confirmed by mapping spontaneously arising mutations.

5.7 Divergence Between Vertebrates and Nematodes Although DSB repair pathways are highly conserved throughout eukaryotes, some critical genes present in vertebrates have not been identified in nematodes. This is notably the case for one of the first factors phosphorylated in response to DSB formation, the histone variant H2AX, also present in the budding yeast genome (Downs et al. 2000; Paull et al. 2000). In fact, none of the H2A variant genes present in the C. elegans genome encodes the conserved C-terminal tail phosphoserine that is targeted by the ATM kinase in response to DSB in vertebrates and yeasts (unpublished). Another critical factor of the vertebrate response to DSB is MDC1, which is also phosphorylated by ATM, and is absent in the worm genome. In vertebrates, MDC1 directly interacts with phosphorylated H2AX through its BRCT domains (Goldberg et al. 2003; Stewart et al. 2003) and subsequently recruits the RNF8 ubiquitin ligase (Mailand et al. 2007). This process leads to the polyubiquitination of histone H2A and the recruitment of further ubiquitination ligases to

114


the vicinity of the DSB, and is believed to help maintain the structure of repair centres. It is intriguing to speculate why none of these factors (RNF8, H2AX, MDC1) have been found in the C. elegans genome. However, it is unlikely to be due to the low level of homology of these factors with putative C. elegans orthologues. We rather speculate that they have been lost during nematode evolution. Thus, the above repair factors might function to optimize recombination repair and DNA damage signalling without being integral parts of this process. Other DNA repair genes that could not be identified in C. elegans include a homologue of the NBS1 gene, which exists in a complex with the MRE11 nuclease and RAD50 in mammalian cells (Alani et al. 1989; Usui et al. 1998; Hopfner et al. 2002), and is implicated in recombinational repair, NHEJ and DNA damage signalling. Another noticeable absence is ATRIP, a protein needed to recruit ATR to sites of DNA damage. Both NBS1 and ATRIP are evolutionary very divergent and their homology to their respective budding yeast counterparts is barely detectable (Edwards et al. 1999; Paciotti et al. 2000; Rouse and Jackson 2000). It is thus likely that their orthologues have not been identified in nematodes due to a very low level of sequence identity.

5.8 Telomere Replication and Mortal Germline Mutations Studies in C. elegans have addressed the connection between DNA damage checkpoint pathways, germ cell apoptosis and DNA repair factors, and the maintenance of immortal germlines. Unbiased genetic screens led to the identification of mutants that are defective in germ cell immortality. These mrt (mortal germline) mutants proliferate normally for several generations, but produce reduced progeny numbers in later generations, before succumbing to sterility caused by defects in germ cell proliferation (Ahmed and Hodgkin 2000). Several such mutants have been linked to telomere replication defects (Ahmed and Hodgkin 2000; Boerckel et al. 2007). A telomere replication defect in C. elegans leads to a progressive shortening of chromosome ends and ultimately to sterility after 15–20 generations, due to anaphase bridges and mitotic catastrophe. In addition to the telomerase catalytic subunit TRT-1 (Meier et al. 2006), another group of genes, namely hus-1, mrt-2 and hpr-17 that encode for the aforementioned PCNA-like 9-1-1 complex (the RAD-17 clamp-loader), have been shown to be essential for telomere replication. These studies guided investigations in mammalian tissue culture cells that showed that the 9-1-1 complex is indeed required for telomerase loading (Francia et al. 2006). This finding and double mutant analysis with C. elegans telomerase and 9-1-1 mutants lead to the notion that all these genes act in the same telomere replication pathway as the worm telomerase. It was also argued that there might be a common process required for both recombinational DSB repair and telomere replication. This notion was recently challenged by the description of the MRT-1 nuclease, a C. elegans hybrid protein composed of a conserved SMC1 nuclease domain fused to an OB-DNA binding fold related to the second such fold of the POT1 telomere-binding


115

protein (Meier et al. 2009). MRT-1 mutants, similar to mutants in the 9-1-1 complex, are DNA cross-link repair defective, but not IR-sensitive. In addition, a TRT-1 mutation leads to neither DSB repair nor DNA cross-link defects. However, given that MRT-1, the 9-1-1 complex and TRT-1 all act in the same telomere replication pathway, nucleolytic DNA processing akin to DNA cross-link repair is likely to be required for telomerease activity in vivo. In yeast and mammalian cells telomere end-protection proteins are known (Garvik et al. 1995; O’Sullivan and Karlseder 2010). These protect telomeres from being recognized and processed as DNA DSBs, and hence depletion of such genes leads to immediate checkpoint activation resulting in senescence and apoptotic cell death (Verdun and Karlseder 2006, 2007). However, no such genes have been detected in C. elegans. Only late-stage telomere replication mutants show chromosome fusions and evidence for activation of the DNA damage checkpoint pathway (data not shown). Several further DNA repair mutants are known to have a mortal germline defect, although none of these mutants have the progressive telomere shortening phenotype of mrt mutants. Nevertheless, RFS-1, the sole worm homologue of the mammalian RAD51 paralogue family, seems to be required for intact telomeres. rfs-1 mutants display fluctuating telomere length, and progression into germline mortality is much less uniform compared to telomerase mutants (Ward et al. 2007; Yanowitz 2008). Indeed, rfs-1 mutants were shown to be able to occasionally partially rescue (or delay) the onset of sterility of telomerase mutants, implying that RFS-1 might have a role in the recombination of subtelomeric sequences. In addition, mutants in DNA mismatch repair genes, as well as in the C. elegans him-6 helicase, can lead to a mortal germline phenotype (Degtyareva et al. 2002; Ahmed 2006). It is intriguing to speculate that the mortal germline phenotype of these DNA repair factors is due to the accumulation of deleterious mutations and/or chromosome fusions. Such a phenotype could resemble the phenotype of hypermorphic DNA double-strand repair mutants in mouse models, where actively proliferating tissues become progressively depleted (Morales et al. 2005).

5.9 The Regulation of DNA Damage-Induced Germ Cell Apoptosis Germ cells damaged by an exposure to genotoxic agents such as IR, UV light or N-ethyl-N-nitrosourea (ENU) are removed by programmed cell death mediated by the C. elegans p53 homologue, CEP-1 (C. elegans p53-like 1). Homology to the mammalian p53 family was initially found in the DNA-binding domain (Derry et al. 2001; Schumacher et al. 2001) and subsequent biochemical and bioinformatic analysis has also identified a SAM protein–protein interaction domain in CEP-1 (Ou et al. 2007). A domain structure that includes a SAM domain is reminiscent of the mammalian p63 and p73 superfamily members, which are indeed slightly more related to the CEP-1 DNA-binding domain than is p53. Surprisingly, both p63 and p73 have predominately developmental roles, in skin development and neurogenesis

116


respectively, while cep-1 mutants do not have any overt developmental phenotype. Nevertheless, the exact phylogenetic relationship of CEP-1 to any of the three mammalian p53 members is not clear, given that gene radiation into the three p53 families occurred during vertebrate development. The recent discovery that an isoform of mammalian p63 is specifically required for the elimination of female meiotic germ cells indicates that germ cell apoptosis might be a primordial function of the p53 family, and that p63 might be evolutionarily ancient (Suh et al. 2006). CEP-1 appears to act as a transcription factor and is required for the transcriptional induction of the egl-1 (Hofmann et al. 2002) and ced-13 genes (Schumacher et al. 2005b), that encode BH3-only proteins (Fig. 5.2). These BH3-only proteins induce apoptosis in worms by binding to the Bcl-2 homologue, CED-9, which normally holds the Apaf-1 homologue CED-4 in an inactive state. EGL-1 binding to CED-9 releases CED-4, which can then activate the caspase CED-3. Loss of CEP-1 function through RNAi knockdown, gene deletion, or expression of a dominant-negative protein lacking the DNA-binding domain, results in viable worms without any overt developmental defect (Derry et al. 2001; Schumacher et al. 2001) (Fig. 5.2). In addition, CEP-1 does not affect IR-induced germ cell cycle arrest, and cep-1 mutants are only weakly radiosensitive in radiation survival assays. Besides egl-1 and ced-13, there are very few genes that are transcriptionally induced by IR, and the induction of most of these genes does not depend on CEP-1. Indeed, the transcriptional induction of genes in response to IR only seems to play a rather minor role in the damage response. None of the IR-induced genes encodes a DNA repair or a DNA damage checkpoint signalling protein. IR-induced genes rather seem to be involved in responding to oxidative stress, and overlap with the set of genes induced in response to infection with bacterial pathogens, oxidative stress and ageing (Greiss et al. 2008b). Several genes have been shown to negatively regulate CEP-1-dependent apop tosis. The ABL-1 and AKT kinases antagonize CEP-1-induced apoptosis following irradiation, and lead to elevated levels of CEP-1, but their mode of action is unknown (Deng et al. 2004; Quevedo et al. 2007). Loss of components of the SCFFSN-1 ubiquitin ligase and of the neddylation pathway also result in increased CEP-1 phosphorylation and protein levels, along with increased egl-1 transcription (Gao et al. 2008). However, it is not clear whether SCFFSN-1 regulates CEP-1 directly, as no direct interaction between FSN-1 and CEP-1 has been detected. The best-studied negative regulator of CEP-1 is GLD-1. GLD-1 is an mRNA-binding protein that inhibits the translation of multiple target mRNAs involved in germ cell differentiation, mainly through interaction with target sequences in their 3¢ UTRs. A hypomorphic allele of gld-1 was identified that, in contrast to gld-1 null alleles, shows no overt defects in germ cell differentiation but specifically leads to the hyper-activation of CEP-1-dependent apoptosis (Schumacher et al. 2005a). The corresponding mutant protein fails to recognize the cep-1 3¢ UTR target, which leads to increased CEP-1 expression, while other developmental targets of GLD-1 are regulated normally. In the absence of GLD-1 more CEP-1 is translated, resulting in increased apoptosis. It will be interesting to see if similar regulatory mechanisms are important for the regulation of the mammalian p53 family. In addition, the sole


117

C. elegans ASSP proteins iASSP was shown to bind CEP-1, similar to mammalian p53 (Bergamaschi et al. 2003). Worm iASSP bound both CEP-1 and p53 in vitro, indicating that it might negatively regulate CEP-1 by direct interaction. Finally, the protein arginine methyltransferase PRMT-5 can bind to both CEP-1 and the coactivator CBP-1 (p300/CBP homologue) and methylates the latter (Yang et al. 2009). When PRMT-5 is absent, cep-1-induced apoptosis and egl-1 mRNA levels are increased following DNA damage, indicating that PRMT-5 normally acts to modulate CEP-1-dependent transcription of egl-1. Mutations in several other genes have been shown to specifically result in reduced apoptosis induction in response to IR, reminiscent of cep-1 mutants (Fig. 5.2). RNAi-mediated depletion of lag-3 the worm homologue of the mammalian MAML Notch transcriptional coactivator resulted in decreased apoptosis and decreased egl-1 and ced-13 induction following irradiation. LAG-3 may thus act as a CEP-1 coactivator similar to the function of MAML in p53 activation (Zhao et al. 2007), but these studies, however, still have to be confirmed by the analysis of a lag-3 deletion mutant. Unexpectedly, egl-1 and ced-13 are still normally induced in several further mutants that show defects in DNA damage-induced apoptosis. This phenotype indicates that the corresponding genes impinge on apoptosis induction in one or more genetic pathways that act either downstream or in parallel of the cep-1-dependent transcriptional induction of egl-1. The first such factors described are components of the C. elegans retinoblastoma complex (RB1) (Schertel and Conradt 2007) which also affect basal, DNA damage independentgerm cell apoptosis (Fig. 5.2). The worm homologue of the human SIRT1 histone deacetlylase SIR-2, which has been implicated in regulating transcriptional repression and ageing also affects IR-induced apoptosis (Greiss et al. 2008a). Interestingly, it was found that SIR-2 translocates from the nucleus to the cytoplasm in earlystage apoptotic germ cells, and evidence was provided that such translocation may be functionally linked to apoptosis induction (Fig. 5.2). Other such factors, including the ING-3 transcription factor (Luo et al. 2009) and components of the ceramide synthesis pathway, have been shown to be required for DNA damage-induced apoptosis (Deng et al. 2008) (Fig. 5.2). Interestingly, the apoptotic defect associated with these mutants is restored upon microinjection of long-chain natural ceramide, and ceramide concentrations in mitochondria are increased upon IR. The involvement of ceramide in mammalian apoptosis has been controversial, and this study indeed confirms that ceramide has a role in inducing apoptosis (Fig. 5.2). Finally, KRI-1, an orthologue of KRIT1/CCM1, which is mutated in the human neurovascular disease cerebral cavernous malformation, is required to activate DNA damagedependent cell death independently of cep-1/p53 (Ito et al. 2010). While it has been hypothesized that the major regulatory events controlling cell death occur by cellautonomous mechanisms, kri-1 and at least in part the retinoblastoma complex components appear to function in C. elegans somatic tissues to affect apoptosis induction in the C. elegans germline (Fig. 5.2). In summary, C. elegans encodes a primordial p53 pathway needed for DNA damage-induced apoptosis. Intriguingly, several other genes are also required in what appears to be parallel pathways to promote germ cell apoptosis in response to IR. At the moment, these factors cannot

118


be placed into a simple genetic or biochemical pathway, and future studies will be required to mechanistically link these genes to apoptosis regulation. Furthermore, it is not clear whether those factors acting in parallel to CEP-1 are regulated by the ATM/ATR kinases. It will be interesting to see if the same genes have a role in apoptosis induction in vertebrate cells. Acknowledgements A CRUK career development award to Anton Gartner funded work in the Gartner lab. Aymeric Bailly was supported by a Wellcome Trust VIP award. We are grateful to Ashley Craig and Ulrike Gartner for proofreading.

References Adamo A, Montemauri P, Silva N et al (2008) BRCA-1 acts in the inter-sister pathway of meiotic double-strand break repair. EMBO Rep 9:287–292 Ahmed S (2006) Uncoupling of pathways that promote postmitotic life span and apoptosis from replicative immortality of Caenorhabditis elegans germ cells. Aging Cell 5:559–563 Ahmed S, Hodgkin J (2000) MRT-2 checkpoint protein is required for germline immortality and telomere replication in C. elegans. Nature 403:159–164 Ahmed S, Alpi A, Hengartner MO et al (2001) C. elegans RAD-5/CLK-2 defines a new DNA damage checkpoint protein. Curr Biol 11:1934–1944 Alani E, Subbiah S, Kleckner N (1989) The yeast RAD50 gene encodes a predicted 153-kD protein containing a purine nucleotide-binding domain and two large heptad-repeat regions. Genetics 122:47–57 Alpi A, Pasierbek P, Gartner A et al (2003) Genetic and cytological characterization of the recombination protein RAD-51 in Caenorhabditis elegans. Chromosoma 112:6–16 Andersen SL, Bergstral DT, Kohl KP et al (2009) Drosophila MUS312 and the vertebrate ortholog BTBD12 interact with DNA structure-specific endonucleases in DNA repair and recombination. Mol Cell 35:128–135 Anderson CM, Blackburn EH (2008) Mec1 function in the DNA damage response does not require its interaction with Tel2. Cell Cycle 7:3695–3698 Anderson CM, Korkin D, Smith DL et al (2008) Tel2 mediates activation and localization of ATM/Tel1 kinase to a double-strand break. Genes Dev 22:854–859 Bailly AP, Freeman A, Hall J et al (2010) The Caenorhabditis elegans homolog of Gen1/Yen1 resolvases links DNA damage signaling to DNA double-strand break repair. PLoS Genetics. Jul 15;6(7):e1001025 Barber LJ, Youds JL, Ward JD et al (2008) RTEL1 maintains genomic stability by suppressing homologous recombination. Cell 135:261–271 Bartek J, Bartkova J, Lukas J (2007) DNA damage signalling guards against activated oncogenes and tumour progression. Oncogene 26:7773–7779 Benson FE, Stasiak A, West SC (1994) Purification and characterization of the human Rad51 protein, an analogue of E. coli RecA. EMBO J 13:5764–5771 Bergamaschi D, Samuels Y, O’Neil NJ et al (2003) iASPP oncoprotein is a key inhibitor of p53 conserved from worm to human. Nat Genet 33:162–167 Bermudez VP, Lindsey-Boltz LA, Cesare AJ et al (2003) Loading of the human 9-1-1 checkpoint complex onto DNA by the checkpoint clamp loader hRad17-replication factor C complex in vitro. Proc Natl Acad Sci USA 100:1633–1638 Bhalla N, Dernburg AF (2005) A conserved checkpoint monitors meiotic chromosome synapsis in Caenorhabditis elegans. Science 310:1683–1686 Boerckel J, Walker D, Ahmed S (2007) The Caenorhabditis elegans Rad17 homolog HPR-17 is required for telomere replication. Genetics 176:703–709


119

Boulton SJ, Martin JS, Polanowska J et al (2004) BRCA1/BARD1 orthologs required for DNA repair in Caenorhabditis elegans. Curr Biol 14:33–39 Brenner S (1974) The genetics of Caenorhabditis elegans. Genetics 77:71–94 C. elegans Sequencing Consortium (1998) Genome sequence of the nematode C. elegans: a platform for investigating biology. Science 282:2012–2018 Chalfie M, Tu Y, Euskirchen G et al (1994) Green fluorescent protein as a marker for gene expression. Science 263:802–805 Chin GM, Villeneuve AM (2001) C. elegans mre-11 is required for meiotic recombination and DNA repair but is dispensable for the meiotic G(2) DNA damage checkpoint. Genes Dev 15:522–534 Clarkson SG (2003) The XPG story. Biochimie 85:1113–1121 Clejan I, Boerckel J, Ahmed S (2006) Developmental modulation of nonhomologous end joining in Caenorhabditis elegans. Genetics 173:1301–1317 Collis SJ, Barber LJ, Clark AJ et al (2007) HCLK2 is essential for the mammalian S-phase checkpoint and impacts on Chk1 stability. Nat Cell Biol 9:391–401 Conradt B, Horvitz HR (1998) The C. elegans protein EGL-1 is required for programmed cell death and interacts with the Bcl-2-like protein CED-9. Cell 93:519–529 Crittenden SL, Kimble J (2008) Analysis of the C. elegans germline stem cell region. Methods Mol Biol 450:27–44 Crittenden SL, Leonhard KA, Byrd DT et al (2006) Cellular analyses of the mitotic region in the Caenorhabditis elegans adult germ line. Mol Biol Cell 17:3051–3061 Degtyareva NP, Greenwell P, Hofmann ER et al (2002) Caenorhabditis elegans DNA mismatch repair gene msh-2 is required for microsatellite stability and maintenance of genome integrity. Proc Natl Acad Sci USA 99:2158–2163 Deng X, Hofmann ER, Villanueva A et al (2004) Caenorhabditis elegans ABL-1 antagonizes p53-mediated germline apoptosis after ionizing irradiation. Nat Genet 36:906–912 Deng X, Yin X, Allan R et al (2008) Ceramide biogenesis is required for radiation-induced apoptosis in the germ line of C. elegans. Science 322:110–115 Dernburg AF, McDonald K, Moulder G et al (1998) Meiotic recombination in C. elegans initiates by a conserved mechanism and is dispensable for homologous chromosome synapsis. Cell 94:387–398 Derry WB, Putzke AP, Rothman JH (2001) Caenorhabditis elegans p53: role in apoptosis, meiosis, and stress resistance. Science 294:591–595 Downs JA, Lowndes NF, Jackson SP (2000) A role for Saccharomyces cerevisiae histone H2A in DNA repair. Nature 408:1001–1004 Edwards RJ, Bentley NJ, Carr AM (1999) A Rad3–Rad26 complex responds to DNA damage independently of other checkpoint proteins. Nat Cell Biol 1:393–398 Ellis HM, Horvitz HR (1986) Genetic control of programmed cell death in the nematode C. elegans. Cell 44:817–829 Fekairi S, Scaglione S, Chahwan C et al (2009) Human SLX4 is a Holliday junction resolvase subunit that binds multiple DNA repair/recombination endonucleases. Cell 138:78–89 Fire A, Xu S, Montgomery MK et al (1998) Potent and specific genetic interference by doublestranded RNA in Caenorhabditis elegans. Nature 391:806–811 Francia S, Weiss RS, Hande MP et al (2006) Telomere and telomerase modulation by the mammalian Rad9/Rad1/Hus1 DNA-damage-checkpoint complex. Curr Biol 16:1551–1558 Futreal PA, Liu Q, Shattuck-Eidens D et al (1994) BRCA1 mutations in primary breast and ovarian carcinomas. Science 266:120–122 Galkin VE, Esashi F, Yu X et al (2005) BRCA2 BRC motifs bind RAD51-DNA filaments. Proc Natl Acad Sci USA 102:8537–8542 Gao MX, Liao EH, Yu B et al (2008) The SCF FSN-1 ubiquitin ligase controls germline apoptosis through CEP-1/p53 in C. elegans. Cell Death Differ 15:1054–1062 Garcia-Muse T, Boulton SJ (2005) Distinct modes of ATR activation after replication stress and DNA double-strand breaks in Caenorhabditis elegans. EMBO J 24:4345–4355 Gartner A, Milstein S, Ahmed S et al (2000) A conserved checkpoint pathway mediates DNA damage-induced apoptosis and cell cycle arrest in C. elegans. Mol Cell 5:435–443

120


Gartner A, MacQueen AJ, Villeneuve AM (2004) Methods for analyzing checkpoint responses in Caenorhabditis elegans. Methods Mol Biol 280:257–274 Garvik B, Carson M, Hartwell L (1995) Single-stranded DNA arising at telomeres in cdc13 mutants may constitute a specific signal for the RAD9 checkpoint. Mol Cell Biol 15:6128–6138 Goldberg M, Stucki M, Falck J et al (2003) MDC1 is required for the intra-S-phase DNA damage checkpoint. Nature 421:952–956 Green CM, Erdjument-Bromage H, Tempst P et al (2000) A novel Rad24 checkpoint protein complex closely related to replication factor C. Curr Biol 10:39–42 Greiss S, Hall J, Ahmed S et al (2008a) C. elegans SIR-2.1 translocation is linked to a proapoptotic pathway parallel to cep-1/p53 during DNA damage-induced apoptosis. Genes Dev 22:2831–2842 Greiss S, Schumacher B, Grandien K et al (2008b) Transcriptional profiling in C. elegans suggests DNA damage dependent apoptosis as an ancient function of the p53 family. BMC Genomics 9:334 Gumienny TL, Lambie E, Hartwieg E et al (1999) Genetic control of programmed cell death in the Caenorhabditis elegans hermaphrodite germline. Development 126:1011–1022 Hachet V, Canard C, Gonczy P (2007) Centrosomes promote timely mitotic entry in C. elegans embryos. Dev Cell 12:531–541 Harris J, Lowden M, Clejan I et al (2006) Mutator phenotype of Caenorhabditis elegans DNA damage checkpoint mutants. Genetics 174:601–616 Hartman PS, Herman RK (1982) Radiation-sensitive mutants of Caenorhabditis elegans. Genetics 102:159–178 Hengartner MO, Ellis RE, Horvitz HR (1992) Caenorhabditis elegans gene ced-9 protects cells from programmed cell death. Nature 356:494–499 Hillier LW, Marth GT, Quinlan AR et al (2008) Whole-genome sequencing and variant discovery in C. elegans. Nat Methods 5:183–188 Hirsh D, Oppenheim D, Klass M (1976) Development of the reproductive system of Caenorhabditis elegans. Dev Biol 49:200–219 Hodgkin JA, Brenner S (1977) Mutations causing transformation of sexual phenotype in the nematode Caenorhabditis elegans. Genetics 86:275–287 Hodgkin J, Horvitz HR, Brenner S (1979) Nondisjunction mutants of the nematode Caenorhabditis elegans. Genetics 91:67–94 Hofmann ER, Milstein S, Boulton SJ et al (2002) Caenorhabditis elegans HUS-1 is a DNA damage checkpoint protein required for genome stability and EGL-1-mediated apoptosis. Curr Biol 12:1908–1918 Hopfner KP, Craig L, Moncalian G et al (2002) The Rad50 zinc-hook is a structure joining Mre11 complexes in DNA recombination and repair. Nature 418:562–566 Ip SC, Rass U, Blanco MG et al (2008) Identification of Holliday junction resolvases from humans and yeast. Nature 456:357–361 Ito S, Greiss S, Gartner A et al (2010) Cell-nonautonomous regulation of C. elegans germ cell death by kri-1. Curr Biol 20:333–338 Jasin M (2002) Homologous repair of DNA damage and tumorigenesis: the BRCA connection. Oncogene 21:8981–8993 Kalogeropoulos N, Christoforou C, Green AJ et al (2004) chk-1 is an essential gene and is required for an S-M checkpoint during early embryogenesis. Cell Cycle 3:1196–1200 Kamath RS, Martinez-Campos M, Zipperlen P et al (2001) Effectiveness of specific RNAmediated interference through ingested double-stranded RNA in Caenorhabditis elegans. Genome Biol 2:RESEARCH0002 Kamath RS, Fraser AG, Dong Y et al (2003) Systematic functional analysis of the Caenorhabditis elegans genome using RNAi. Nature 421:231–237 Kimble JE, White JG (1981) On the control of germ cell development in Caenorhabditis elegans. Dev Biol 81:208–219


121

Kostrub CF, Knudsen K, Subramani S et al (1998) Hus1p, a conserved fission yeast checkpoint protein, interacts with Rad1p and is phosphorylated in response to DNA damage. EMBO J 17:2055–2066 Lakowski B, Hekimi S (1996) Determination of life-span in Caenorhabditis elegans by four clock genes. Science 272:1010–1013 Lee SS, Lee RY, Fraser AG et al (2003) A systematic RNAi screen identifies a critical role for mitochondria in C. elegans longevity. Nat Genet 33:40–48 Lee SJ, Gartner A, Hyun M et al (2010) The Caenorhabditis elegans Werner syndrome protein functions upstream of ATR and ATM in response to DNA replication inhibition and doublestrand DNA breaks. PLoS Genet 6:e1000801 Lilley DM, White MF (2001) The junction-resolving enzymes. Nat Rev Mol Cell Biol 2:433–443 Lim CS, Mian IS, Dernburg AF et al (2001) C. elegans clk-2, a gene that limits life span, encodes a telomere length regulator similar to yeast telomere binding protein Tel2p. Curr Biol 11:1706–1710 Lorick KL, Jensen JP, Fang S et al (1999) RING fingers mediate ubiquitin-conjugating enzyme (E2)-dependent ubiquitination. Proc Natl Acad Sci USA 96:11364–11369 Luo J, Shah S, Riabowol K et al (2009) The Caenorhabditis elegans ing-3 gene regulates ionizing radiation-induced germ-cell apoptosis in a p53-associated pathway. Genetics 181:473–482 Mailand N, Bekker-Jensen S, Faustrup H et al (2007) RNF8 ubiquitylates histones at DNA doublestrand breaks and promotes assembly of repair proteins. Cell 131:887–900 Majka J, Burgers PM (2003) Yeast Rad17/Mec3/Ddc1: a sliding clamp for the DNA damage checkpoint. Proc Natl Acad Sci USA 100:2249–2254 Martin JS, Winkelmann N, Petalcorin MI et al (2005) RAD-51-dependent and -independent roles of a Caenorhabditis elegans BRCA2-related protein during DNA double-strand break repair. Mol Cell Biol 25:3127–3139 Meier B, Clejan I, Liu Y et al (2006) trt-1 is the Caenorhabditis elegans catalytic subunit of telomerase. PLoS Genet 2:e18 Meier B, Barber LJ, Liu Y et al (2009) The MRT-1 nuclease is required for DNA crosslink repair and telomerase activity in vivo in Caenorhabditis elegans. EMBO J 28:3549–3563 Miki Y, Swensen J, Shattuck-Eidens D et al (1994) A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science 266:66–71 Mimitou EP, Symington LS (2009) DNA end resection: many nucleases make light work. DNA Repair 8:983–995 Morales M, Theunissen JW, Kim CF et al (2005) The Rad50S allele promotes ATM-dependent DNA damage responses and suppresses ATM deficiency: implications for the Mre11 complex as a DNA damage sensor. Genes Dev 19:3043–3054 Moser SC, von Elsner S, Bussing I et al (2009) Functional dissection of Caenorhabditis elegans CLK-2/TEL2 cell cycle defects during embryogenesis and germline development. PLoS Genet 5:e1000451 Munoz IM, Hain K, Declais AC et al (2009) Coordination of structure-specific nucleases by human SLX4/BTBD12 is required for DNA repair. Mol Cell 35:116–127 Nairz K, Klein F (1997) mre11S – a yeast mutation that blocks double-strand-break processing and permits nonhomologous synapsis in meiosis. Genes Dev 11:2272–2290 Ogawa T, Yu X, Shinohara A et al (1993) Similarity of the yeast RAD51 filament to the bacterial RecA filament. Science 259:1896–1899 O’Sullivan RJ, Karlseder J (2010) Telomeres: protecting chromosomes against genome instability. Nat Rev Mol Cell Biol 11:171–181 Ou HD, Lohr F, Vogel V et al (2007) Structural evolution of C-terminal domains in the p53 family. EMBO J 26:3463–3473 Paciotti V, Clerici M, Lucchini G et al (2000) The checkpoint protein Ddc2, functionally related to S. pombe Rad26, interacts with Mec1 and is regulated by Mec1-dependent phosphorylation in budding yeast. Genes Dev 14:2046–2059

122


Paull TT, Rogakou EP, Yamazaki V et al (2000) A critical role for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage. Curr Biol 10:886–895 Penkner A, Portik-Dobos Z, Tang L et al (2007) A conserved function for a Caenorhabditis elegans Com1/Sae2/CtIP protein homolog in meiotic recombination. EMBO J 26:5071–5082 Polanowska J, Martin JS, Garcia-Muse T et al (2006) A conserved pathway to activate BRCA1dependent ubiquitylation at DNA damage sites. EMBO J 25:2178–2188 Pontier DB, Tijsterman M (2009) A robust network of double-strand break repair pathways governs genome integrity during C. elegans development. Curr Biol 19:1384–1388 Quevedo C, Kaplan DR, Derry WB (2007) AKT-1 regulates DNA-damage-induced germline apoptosis in C. elegans. Curr Biol 17:286–292 Rinaldo C, Bazzicalupo P, Ederle S et al (2002) Roles for Caenorhabditis elegans rad-51 in meiosis and in resistance to ionizing radiation during development. Genetics 160:471–479 Rouse J, Jackson SP (2000) LCD1: an essential gene involved in checkpoint control and regulation of the MEC1 signalling pathway in Saccharomyces cerevisiae. EMBO J 19:5801–5812 Russell P, Nurse P (1986) cdc25+ functions as an inducer in the mitotic control of fission yeast. Cell 45:145–153 Saito TT, Youds JL, Boulton SJ et al (2009) Caenorhabditis elegans HIM-18/SLX-4 interacts with SLX-1 and XPF-1 and maintains genomic integrity in the germline by processing recombination intermediates. PLoS Genet 5:e1000735 Sarin S, Prabhu S, O’Meara MM et al (2008) Caenorhabditis elegans mutant allele identification by whole-genome sequencing. Nature Meth 5:865–867 Sartori AA, Lukas C, Coates J et al (2007) Human CtIP promotes DNA end resection. Nature 450:509–514 Schertel C, Conradt B (2007) C. elegans orthologs of components of the RB tumor suppressor complex have distinct pro-apoptotic functions. Development 134:3691–3701 Schumacher B, Hofmann K, Boulton S et al (2001) The C. elegans homolog of the p53 tumor suppressor is required for DNA damage-induced apoptosis. Curr Biol 11:1722–1727 Schumacher B, Hanazawa M, Lee MH et al (2005a) Translational repression of C. elegans p53 by GLD-1 regulates DNA damage-induced apoptosis. Cell 120:357–368 Schumacher B, Schertel C, Wittenburg N et al (2005b) C. elegans ced-13 can promote apoptosis and is induced in response to DNA damage. Cell Death Differ 12:153–161 Scully R, Chen J, Ochs RL et al (1997a) Dynamic changes of BRCA1 subnuclear location and phosphorylation state are initiated by DNA damage. Cell 90:425–435 Scully R, Chen J, Plug A et al (1997b) Association of BRCA1 with Rad51 in mitotic and meiotic cells. Cell 88:265–275 Shendure J, Ji H (2008) Next-generation DNA sequencing. Nat Biotechnol 26:1135–1145 Shikata M, Ishikawa F, Kanoh J (2007) Tel2 is required for activation of the Mrc1-mediated replication checkpoint. J Biol Chem 282:5346–5355 Stergiou L, Doukoumetzidis K, Sendoel A et al (2007) The nucleotide excision repair pathway is required for UV-C-induced apoptosis in Caenorhabditis elegans. Cell Death Differ 14:1129–1138 Stewart GS, Wang B, Bignell CR et al (2003) MDC1 is a mediator of the mammalian DNA damage checkpoint. Nature 421:961–966 Suh EK, Yang A, Kettenbach A et al (2006) p63 protects the female germ line during meiotic arrest. Nature 444:624–628 Sulston JE, Horvitz HR (1977) Post-embryonic cell lineages of the nematode, Caenorhabditis elegans. Dev Biol 56:110–156 Sulston JE, Schierenberg E, White JG et al (1983) The embryonic cell lineage of the nematode Caenorhabditis elegans. Dev Biol 100:64–119 Svendsen JM, Smogorzewska A, Sowa ME et al (2009) Mammalian BTBD12/SLX4 assembles a Holliday junction resolvase and is required for DNA repair. Cell 138:63–77 Takai H, Wang RC, Takai KK et al (2007) Tel2 regulates the stability of PI3K-related protein kinases. Cell 131:1248–1259


123

Usui T, Ohta T, Oshiumi H et al (1998) Complex formation and functional versatility of Mre11 of budding yeast in recombination. Cell 95:705–716 Venclovas C, Thelen MP (2000) Structure-based predictions of Rad1, Rad9, Hus1 and Rad17 participation in sliding clamp and clamp-loading complexes. Nucleic Acids Res 28:2481–2493 Verdun R, Karlseder J (2006) The DNA damage machinery and homologous recombination pathway act consecutively to protect human telomeres. Cell 127:709–720 Verdun R, Karlseder J (2007) Replication and protection of telomeres. Nature 447:924–931 Ward JD, Barber LJ, Petalcorin M et al (2007) Replication blocking lesions present a unique substrate for homologous recombination. EMBO J 26:3384–3396 Ward JD, Muzzini DM, Petalcorin MI et al (2010) Overlapping mechanisms promote postsynaptic RAD-51 filament disassembly during meiotic double-strand break repair. Mol Cell 37:259–272 Weinert TA, Hartwell LH (1993) Cell cycle arrest of cdc mutants and specificity of the RAD9 checkpoint. Genetics 134:63–80 Wicky C, Alpi A, Passannante M et al (2004) Multiple genetic pathways involving the Caenorhabditis elegans Bloom’s syndrome genes him-6, rad-51, and top-3 are needed to maintain genome stability in the germ line. Mol Cell Biol 24:5016–5027 Wooster R, Bignell G, Lancaster J et al (1995) Identification of the breast cancer susceptibility gene BRCA2. Nature 378:789–792 Wu LC, Wang ZW, Tsan JT et al (1996) Identification of a RING protein that can interact in vivo with the BRCA1 gene product. Nat Genet 14:430–440 Yang M, Sun J, Sun X et al (2009) Caenorhabditis elegans protein arginine methyltransferase PRMT-5 negatively regulates DNA damage-induced apoptosis. PLoS Genet 5:e1000514 Yanowitz JL (2008) Genome integrity is regulated by the Caenorhabditis elegans Rad51D homolog rfs-1. Genetics 179:249–262 Yildiz O, Majumder S, Kramer B et al (2002) Drosophila MUS312 interacts with the nucleotide excision repair endonuclease MEI-9 to generate meiotic crossovers. Mol Cell 10:1503–1509 Zhao Y, Katzman RB, Delmolino LM et al (2007) The notch regulator MAML1 interacts with p53 and functions as a coactivator. J Biol Chem 282:11969–11981 Zhou Z, Hartwieg E, Horvitz HR (2001) CED-1 is a transmembrane receptor that mediates cell corpse engulfment in C. elegans. Cell 104:43–56

Part II Modulation of Radiation Responses: Opportunities for Therapeutic Exploitation

Chapter 6

Hypoxia and Modulation of Cellular Radiation Response Ester M. Hammond, Monica Olcina, and Amato J. Giaccia

Abstract Tumors possess many features that differentiate them from normal tissues which can potentially be targeted for therapy. One such difference is the presence of an inefficient vascular network, which is often unable to meet the tumor’s needs. The interaction of this faulty vascular system with the cells within the tumor leads to the development of a unique tumor microenvironment not found in normal tissues. This particular environment has a number of distinct features such as low oxygen concentrations, high interstitial fluid pressures and low pH. Previous studies have found that the presence of these microenvironmental conditions potentiates the development of certain adaptive responses by the tumor. These adaptive responses can have a number of adverse consequences, including increased resistance to chemo and radiotherapy. Nevertheless, since these microenvironmental characteristics are found in most solid tumor types, they have a profound effect on the tumor’s behavior and represent attractive therapeutic targets. Keywords Hypoxia • DNA damage response • DNA repair • Reoxygenation

6.1 Characteristics of the Tumor Microenvironment Normal tissues are perfused by a well-structured vasculature which ensures that the needs of the tissue are met. This vasculature comprises a network of blood vessels, which are capable of delivering oxygen and nutrients, as well as a comprehensive lymphatic system, that functions to eliminate waste products and excess fluids. Partial pressures of oxygen in normal tissues lies in the range of 10–80 mmHg (Brown and Giaccia 1998). Within tumors, however, the situation is very different. Blood vessels are chaotic, having a distorted structure that prevents the proper A.J. Giaccia (*) Department of Radiation Oncology, Center for Clinical Sciences Research, Stanford University, Stanford, CA 94303-5152, USA e-mail: [email protected] T.L. DeWeese and M. Laiho (eds.), Molecular Determinants of Radiation Response, Current Cancer Research, DOI 10.1007/978-1-4419-8044-1_6, © Springer Science+Business Media, LLC 2011

127

128

E.M. Hammond et al.

delivery of oxygen. Furthermore, tumor vessels usually have a number of structural abnormalities that make them more permeable than normal vessels. In addition, the organization of such vessels is often disturbed and chaotic, and many vessels are functionally incapable to adequately supply the tumor with oxygen and nutrients. The consequences of a malformed vasculature are areas of chronic hypoxia, where viable tumor cells exist in a state of low oxygen (typically 100–200 mm from the lumens of a blood vessel). Moreover, acute hypoxic areas may also occur where blood vessels temporarily collapse due to, for instance, a high interstitial fluid pressure (IFP) or hemodynamic changes. The majority of hypoxia research is carried out between 2% oxygen, which gives robust hypoxia inducible factor-1 (HIF-1) activation and 0.1% oxygen which induces a block in DNA synthesis because within a tumor, a hypoxic gradient exists, with tumor cells at physiologic oxygen levels near the vessels and regions of anoxia surrounding the necrotic core. Tumor cells can adapt to hypoxia in a number of ways (Vaupel and Harrison 2004). For example, hypoxia may lead to cell cycle arrest particularly at the G1/S phase border through p21 and p27 activation by HIF-1 (Green et al. 2001; Gardner et al. 2003). In addition, hypoxia may promote apoptosis by both a p53-dependent and a p53-independent manner (Alarcon et al. 1999, 2001; Soengas et al. 1999; Koumenis et al. 2001). Necrosis may also be induced in response to hypoxia if cells do not receive oxygen (Vaupel and Harrison 2004). In addition, hypoxia induces a number of proteomic changes, which will enable the tumor to survive and proliferate in a low oxygen environment. Interestingly, many of these same changes involved in adaptation to hypoxia also promote tumor progression and provide the tumor with an increased ability to invade and metastasize to distant sites. Many of these changes in gene expression are mediated by the HIF family of transcription factors. The HIF transcription factor family are heterodimers composed of an a-subunit (HIF-1a, HIF-2a, HIF-3a) and a b-subunit. In the presence of oxygen, HIF-1a undergoes a prolyl hydroxylase (PHD-mediated hydroxylation) and is degraded by the von Hippel Lindau (VHL) protein. As oxygen decreases, HIF-1a is stabilized due to a decrease in hydroxylation by the PHDs (Maxwell et al. 1999; Jaakkola et al. 2001; Kaelin and Ratcliffe 2008). HIF-1 acts as a transcriptional regulator for genes involved in angiogenesis, glycolysis, glucose transport, and increased oxygen delivery (Poon et al. 2009). Examples of critical HIF-regulated genes include vascular endothelial growth factor (VEGF, angiogenesis), GLUT1 (glycolysis) and carbonic anhydrase IX (CAIX, a regulator of pH). These and others have all been used as markers of hypoxia. Oncogene activation and tumor suppressor loss can also have an effect on the tumor’s response to hypoxia. Furthermore, clonal selection of those cells that have developed mutations in certain tumor suppressor gene pathways, which increase survival in the hypoxic environment, will also lead to increased tumor aggressiveness (Graeber et al. 1996). Thus, hypoxia will inevitably be an adverse prognostic factor that predicts for increased metastasis and treatment resistance (Brizel et al. 1996). Increased IFP is another unique characteristic of the tumor microenvironment. A number of factors lead to a high IFP. First, the gaps in the tumor vasculature contribute to the leakiness of the tumor, allowing fluids and proteins to escape.

6 Hypoxia and Modulation of Cellular Radiation Response

129

This creates a high pressure in the interstitium that cannot be corrected, since tumors possess an inefficient lymphatic system and reduced blood flow. This problem is exacerbated further since the extracellular matrix (ECM) is influenced by the action of platelet-derived growth factor (PDGF) and transforming growth factor (TGF) b receptors as well as b1 integrin and collagen (Cairns et al. 2006). As mentioned above, the hypoxic environment found in tumors perturbs normal gene and protein expression at a number of levels. Some of the targets induced by HIF-1 include enzymes that promote a switch to glycolysis. This switch will increase the amount of lactate present. However, while it was originally thought that the low pH found in tumors was solely due to the increased lactic acid produced as a result of this metabolic switch, it has now been shown that other factors may also play a role. The overall acidic pH in tumors seems to be influenced by both extracellular and intracellular pH. Gradients of pH exist within cells which require the action of ion channels such as the monocarboxylate H+ co-transporter (MCT), the vacuolar H+ATPase, the Na+/H+ exchanger and the Na+-dependent Cl/ bicarbonate exchanger. The expression of these membrane transporters is thought to be increased as part of the adaptive response to hypoxia, resulting in a decreased extracellular pH. Many elegant studies have demonstrated that the degree of hypoxia in human tumors negatively correlates with prognosis (Hockel et al. 1996; Nordsmark et al. 1996a, b). This has been attributed to the finding that the hypoxic regions of tumors are more resistant to killing by both radio- and chemotherapy. Chemotherapeutic drugs are delivered to tumors through the blood stream and are therefore reliant on the tumor vasculature for optimum delivery. As discussed above, the tumor vasculature is malformed and inefficient, therefore impeding the delivery of chemotherapeutic agents and antibodies to hypoxic regions. This is also compounded by the finding that some chemotherapies are less efficient in conditions of low oxygen. The reasons behind this include both reduced drug activity in hypoxia (e.g., bleomycin and etoposide) and reduced killing of cells which are not proliferating as found in severely hypoxic areas and decreased pH regions (e.g., alkylating agents) (Shannon et al. 2003; Minchinton and Tannock 2006). Many studies have described the effect of oxygen on radiosensitivity (Brown and Wilson 2004). Oxygen enhances radiosensitivity and conversely a lack of oxygen or hypoxia increases radiosensitivity. Hypoxic cells are two to three times more resistant to killing by ionizing radiation than cells at normal oxygen concentrations. This ratio is termed the oxygen enhancement ratio (OER) and is irrespective of cell type and inherent radiosensitivity but is dependent on radiation quality (O’Neill and Wardman 2009). The toxic effects induced by radiation are mediated through the generation of reactive oxygen species which, if in proximity to the DNA will cause damage. In the presence of oxygen this damage becomes “fixed” or permanent. This is, of course, a simplification of what actually occurs. It has been proposed that the presence of oxygen radicals on DNA bases leads to double-strand breaks (DSBs) as well as damaged bases and that this radical site can then be transferred to the sugar and result in an additional single-strand break. This is referred to as damage transfer by oxygen and results in an amplification of the initial damage.

130

E.M. Hammond et al.

In order for this to occur, it has been estimated that oxygen must be present within a thousandth of second after the irradiation (Wardman 2009). This is also true for chemical radiosensitizers such as nitroimidazoles (Wardman 2007). The most toxic type of DNA damage is an unrepaired DSBs, which is lethal to a cell. More recently, hypoxia has been shown to negatively impact tumorigenesis by driving what is known as a mutator phenotype (Loeb 2001). Several studies recently reviewed by Bristow and Hill (2008) have demonstrated that mutation frequencies are increased both in vitro and in vivo in response to hypoxia (Reynolds et al. 1996; Sandhu et al. 2000; Li et al. 2001). More and more evidence is emerging that hypoxia has a significant impact on the DNA repair capabilities of a cell and that this directly impacts genomic instability (Bertrand et al. 2003).

6.2 Hypoxic Induction of a Unique DNA Damage Response In response to genotoxic agents, which induce DNA damage, a well-characterized signaling cascade is initiated by members of the PI3 Kinase family (Jackson and Bartek 2009). This includes; ataxia telangiectasia mutated (ATM), ATM- and Rad3-related (ATR) and DNA protein kinase (DNA-PK) (Cimprich and Cortez 2008; Lavin 2008). Despite their accepted presence at the top of this signaling pathway, these kinases must first be activated or relocalized to the site of DNA damage. In order to achieve this, first the cell must register that it has received a damaging insult. In response to DSBs, the MRN complex, consisting of MRE11, Rad50, and NBS1, localizes to DNA ends and recruits ATM which is then activated through autophosphorylation (Carney et al. 1998; Costanzo et al. 2001; D’Amours and Jackson 2002; Lee and Paull 2004). Mutation of the ATM, MRE11, and NBS1 genes have all been demonstrated to result in radiosensitivity syndromes (Stewart et al. 1999; Lavin 2008). However, it has been shown experimentally that DNA damage signaling does occur in the absence of these sensor proteins, although with delayed kinetics (Uziel et al. 2003). In some cases, genotoxic stress results in a replication arrest. For example, an induced bulky lesion can block the DNA polymerases leading to stalled replication forks which are sensed by single-stranded DNA-binding proteins (Richard et al. 2008). The most characterized of these singlestrand binding proteins is replication protein A (RPA). Once bound to regions of ssDNA, RPA then recruits ATR and its binding partner ATRIP to sites of stalled replication, therefore bringing the kinase into proximity of the damaged DNA (Zou and Elledge 2003). Once localized accordingly and/or activated, the PI3 kinases phosphorylate numerous downstream targets to bring about cell cycle arrest, apoptosis, DNA repair, and senescence. More recently, both ATM and ATR have been described as being active in hypoxic conditions (Hammond et al. 2002; Gibson et al. 2005; Freiberg et al. 2006; Bencokova et al. 2009; Economopoulou et al. 2009). The finding that exposure of minimally transformed cells to hypoxia induces p53-dependent apoptosis provided one of the first pieces of evidence that hypoxia induces a DNA damage response

6 Hypoxia and Modulation of Cellular Radiation Response

131

(Graeber et al. 1996). Additional studies demonstrated that p53 was both p hosphorylated and active in hypoxic conditions but only under low oxygen tensions ( 20 ng/mL) features (Freytag et al. 2007b). Based on these results, a randomized, controlled phase 2/3 has been opened in newly diagnosed, intermediate-risk prostate cancer. In the second study, a possible treatment effect (100% adenocarcinoma free) was observed across all prognostic risk groups (Teh et al. 2004). These latter results should be interpreted cautiously as only two biopsy cores, rather than the standard 6–14, were taken in posttreatment biopsies. This is not standard practice and likely overestimates the true activity of the investigational therapy. In primary glioblastoma, HSV-1 TK suicide gene therapy has been combined with radiotherapy in two randomized trials (Rainov 2000; Immonen et al. 2004). In this setting, patients on the investigational therapy arm received a local injection of a replication-defective retrovirus or replication-defective adenovirus expressing HSV-1 TK following surgical resection of the primary tumor. They then received 2 weeks of GCV along with a standard course of radiotherapy. Patients on the control arm received standard therapy only (surgery and radiotherapy). Although the combined therapy was well tolerated, neither strategy produced convincing evidence of efficacy. In the trial that utilized the replication-defective retrovirus (248 patients), median survival on the investigational arm was 365 days vs. 354 days on the control arm. In the smaller trial that utilized the replication-defective adenovirus (36 patients), median survival on the investigational therapy arm was 62.4 weeks vs. 37.7 weeks on the control arm, raising the possibility that the gene therapy could in fact improve survival. However, these preliminary results were not confirmed in a larger, four-arm study of 251 patients. Here, patients were randomized to receive surgery followed by radiation ± temozolomide (control arms), or surgery followed by HSV-1 TK gene therapy (Cerepro) and radiation ± temozolomide (investigational arms). Surgery followed by chemoradiation (with temozolomide) is the standard of care for primary glioblastoma. The primary end point was survival, which was defined as death or re-intervention, the latter of which opens up the possibility of investigator bias and would likely drive the survival data. When comparing the two arms that received chemoradiation, median survival was essentially identical (~600 days). As a result, the European Medicines Agency rejected marketing authorization for Cerepro.

8.2.2 p53 Gene Therapy The p53 tumor suppressor gene codes for a sequence-specific DNA-binding protein that has pleotropic functions including the regulation of cell cycle arrest and apoptosis and is mutated in at least 60% of human cancers. Exposure of normal cells to DNA damaging agents, such as radiation, induces p53 expression and cell cycle delays in G1 and G2 permitting the repair of damaged DNA before cells enter S phase or mitosis. Following insult, p53 can induce cells down the apoptotic pathway and may be a major determinant of the sensitivity of cells to chemotherapeutic agents and radiation (Lowe et al. 1993, 1994; Hamada et al. 1996). Because p53 can induce apoptosis following irradiation, several groups proposed using p53 gene

8 Gene Therapy and Radiation

177

therapy as a tumor cell radiosensitizer (Gallardo et al. 1996; Spitz et al. 1996; Badie et al. 1998; Li et al. 1999; Sah et al. 2003). Adenovirus-mediated transfer of the wild-type p53 gene was found to potentiate the effects of radiation in a variety of human cancer cell lines in vitro. The magnitude of the radiosensitization effect (i.e., SERs) was rather modest (1.1–1.3) and lower than what can be achieved with suicide gene therapy. As expected, the increased radiosensitivity correlated with the induction of p53-dependent apoptosis. Based on these preclinical results, a number of p53 gene therapy trials were conducted targeting cancers of the head and neck, lung, ovary, urinary bladder, and brain (Clayman et al. 1998, 1999; Swisher et al. 1999, 2003; Weill et al. 2000; Nemunaitis et al. 2000; Lang et al. 2003; Pagliaro et al. 2003; Wolf et al. 2004; Yoo et al. 2009). These trials used adenovirus-mediated p53 gene therapy as a single agent or in combination with chemotherapy and/or radiotherapy. Overall, p53 gene therapy was well tolerated. Gene therapy-related side effects included local inflammation and mild flu-like symptoms. Preliminary signs of efficacy were observed in a phase 2 trial of non-small lung cancer (NSLC) (Swisher et al. 2003). Nineteen patients received 60 Gy of radiation over a 6-week period along with three intratumoral injections of a p53-expressing replication-defective adenovirus (INGN 201, Advexin). Following the gene therapy, 12 (63%) biopsies showed no viable tumor, which is better than expected (17–20%) for patients receiving radiotherapy only. Seventeen patients were evaluable by computed tomography (CT) and bronchoscopy at 3 months. The overall (p53-injected and noninjected) tumor response was complete response in 1 (5%) patient, partial response in 5 (26%) patients, stable disease in 1 (5%) patient, and progressive disease in 11 (58%) patients. With a median follow-up of 37 months, there was 47% overall survival at 1 year and 26% at 3 years. Five patients were alive without evidence of disease at 34–48 months.

8.2.3 Tumor Necrosis Factor Alpha (TNFa) Gene Therapy TNFa is a multifunctional cytokine that demonstrated antitumor activity over 30 years ago (Carswell et al. 1975; Creasey et al. 1986; Asher et al. 1987). However, the development of recombinant TNFa as a cancer therapeutic was hampered by its severe toxicity when delivered systemically (Chapman et al. 1987; Feinberg et al. 1988; Spriggs et al. 1988). Weischelbaum et al. were the first to propose using TNFa gene therapy as a tumor radiosensitizer (Weichselbaum et al. 1994; Hallahan et al. 1995; Chung et al. 1998). Their approach was somewhat unique in that they placed the TNFa gene under the control of the radiation-inducible Egr-1 promoter. Up to tenfold greater TNFa levels were observed in tumors following irradiation. Combining adenovirus-mediated TNFa gene therapy and radiation resulted in significant tumor control under conditions that produced only modest effects with either single modality. Although it was initially thought that TNFa might actually increase the intrinsic radiosensitivity of tumor cells, it is now believed that most of TNFa antitumor activity is mediated through its effects on the tumor vasculature (Staba et al. 1998).

178

S.O. Freytag et al.

Adenovirus-mediated TNFa gene therapy (TNFerade) is another gene therapy strategy that has progressed to phase 3. Early-stage trials of solid cancers demonstrated that multiple injections of TNFerade were generally well tolerated up to a dose of 4 × 1011 particle units (pu)/injection (Senzer et al. 2004; Mundt et al. 2004; McLoughlin et al. 2005). The most common gene therapy-related side effects were mild to moderate injection-site pain and flu-like symptoms (~20%). Objective tumor responses were observed in a significant fraction of patients (up to 50%). Some patients had synchronous lesions that allowed for a comparison of the tumor response following the gene therapy plus radiotherapy vs. radiotherapy alone. In one study, four of five (80%) patients showed a greater response in the index lesion that received the same radiation dose as the control lesion. Based on these phase 1 results, a randomized, controlled phase 2/3 trial was opened in locally advanced pancreatic cancer comparing TNFerade plus chemoradiation vs. chemoradiation alone. The study called for 330 patients that were enrolled using a 2:1 (investigational arm:control arm) randomization scheme. Patients on the investigational therapy arm received five weekly intratumoral injections of TNFerade concomitant with a standard 5½ week course of 5-FU-based chemoradiation, whereas patients on the control arm received only standard chemoradiation. The primary end point was survival. At the first interim analysis (92 of expected 276 deaths), median survival on both the investigational therapy and control arms was 9.9 months (Chang et al. 2009). There was preliminary evidence of a “late effect” with 30.5% of patients being alive on the investigational therapy arm at 18 months vs. 11.3% on the control arm. Final results showed no difference in survival between the two arms and further development of TNFerade for pancreatic cancer was terminated.

8.3 Gene Therapy Strategies That Have Been Combined with Radiation in Preclinical Models 8.3.1 Replication-Competent Oncolytic Adenoviruses There are several possible mechanisms by which replication-competent adenoviruses may enhance the therapeutic effects of radiation. E1A expression is required for viral replication and it is well-established that E1A can sensitize cells to both chemotherapeutic agents and radiation (Lowe et al. 1993, 1994; Shao et al. 1997; 2001). Another hypothesis is that expression of the adenovirus E4-34K protein, which is known to disrupt V(D)J recombination through its interaction with DNAdependent protein kinase (DNA-PK), may block the repair of double-strand breaks (DSB) following irradiation (Collis et al. 2003a, b). DSB correlate closely with radiation lethality and cells defective in DSB repair and V(D)J recombination exhibit a radiation-sensitive phenotype (Rooney et al. 2002). A variety of replication-competent, oncolytic adenoviruses have been combined with radiation in preclinical models including Ad5-CD/TKrep (Freytag et al. 1998,


179

2002; Rogulski et al. 2000b), Ad5-yCD/mutTKSR39rep-ADP (Freytag et al. 2007a), ONYX-015 virus (Rogulski et al. 2000a), CV706 (Chen et al. 2001), Ad5-D24RGD (Lamfers et al. 2002), KD1, KD3, and VRX-007 (Toth et al. 2003). Ad5-CD/TKrep and Ad5-yCD/mutTKSR39rep-ADP are armed with therapeutic “suicide” genes and have progressed into late-stage clinical trials, which has been discussed previously, whereas the other oncolytic adenoviruses lack a therapeutic gene. All exhibited additive or greater antitumor activity when combined with radiation in preclinical models.

8.3.2 Targeting Signal Transduction and Apoptotic Pathways One of the hallmark characteristics of the malignant cell is the dysregulation of growth- and survival-related signal transduction pathways that confer resistance to conventional cancer treatments such as chemotherapy and radiation therapy. Not surprisingly, components of these pathways including the epidermal growth factor receptor (EGFR) (Lammering et al. 2001, 2003), Raf-1 kinase (Soldatenkov et al. 1997; Kasid and Dritschilo 2003), NF-kB (Jung et al. 1995; Pajonk et al. 1999; Mukogawa et al. 2003), Bcl-2 (Arafat et al. 2000, 2003), mda-7/IL-24 (Kawabe et al. 2002; Su et al. 2003; Yacoub et al. 2003), and tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) (Chinnaiyan et al. 2000; Belka et al. 2001), have been the targets of many gene therapy strategies. Ablation of signal transduction activity using a dominant negative inhibitors (EGFR-CD533 with EGFR, IkB mutants with NF-kB, Bax with Bcl-2) or antisense oligonucleotides (Raf-1) demonstrated SERs of 1.3–2.4 in vitro and increased tumor radiocurability in vivo (Jung et al. 1995; Soldatenkov et al. 1997; Pajonk et al. 1999; Lammering et al. 2001, 2003; Kasid and Dritschilo 2003; Mukogawa et al. 2003). Given that EGFR, Raf-1, and NF-kB signaling are dysregulated in a wide spectrum of human cancers, these targeted gene therapies have the potential to have widespread applicability in the clinic. However, if these gene therapy strategies do not exhibit a bystander effect, they will be limited by the efficiency of gene delivery in vivo. The safety of administering a replication-defective adenovirus expressing mda-7/IL-24 (INGN241) as a single agent has been evaluated in advanced cancers (Tong et al. 2005). Multiple (twice weekly × 3) injections proved to be safe up to a dose of 2 × 1012 vp/injection. Significant tumor apoptosis was observed along with immuneactivating events. Clinical trials combining adenovirus-mediated mda-7/IL-24 gene therapy with radiation therapy are pending.

8.3.3 Targeting DNA Repair Pathways Genes involved in the repair of radiation-induced DNA damage are ideal targets for cancer therapies designed to increase tumor cell radiosensitivity. Double-strand DNA breaks are particularly lethal and are repaired by homologous recombination,

180

S.O. Freytag et al.

single-strand annealing and nonhomologous end joining (NHEJ), the latter of which predominates in human cells. A number of genes involved in these DNA repair pathways have been identified including Ku (Ku70 and Ku80), DNA-PK, XRCC4, DNA ligase IV, RAD51, RAD 52, RAD 54, BCRA1, BRCA2, and others. These repair pathways are intimately linked to cell cycle checkpoints that involve the sensors of radiation-induced DNA damage, namely, ATM, ATR, DNA-PK, and their downstream effector, p53. Hence, several have been the targets of gene therapy strategies. Ablation of Ku, ATM, ATR, and DNA-PK expression by antisense strategies resulted in marked radiosensitization in vitro and improved tumor radiocurability (Nussenzweig et al. 1997; Marangoni et al. 2000; Li et al. 2003; Collis et al. 2003b). Further investigations of these gene therapy strategies are warranted.

8.3.4 Other Strategies Other gene therapy strategies that have been combined with radiation include nitric synthase (iNOS; Wang et al. 2004), phosphatase and tensin homolog (PTEN; Tomioka et al. 2008), and interference of genes that play a role in autophagy (Apel et al. 2008). Nitric oxide (NO) has many activities and is known to sensitize oxic tumor cells to irradiation. Adenovirus-mediated iNOS gene transfer into human colorectal cells resulted in respectable radiosensitization with SERs of ~1.6. An advantage of iNOS gene therapy is that it exhibits a bystander effect. PTEN is a tumor suppressor that negatively regulates intracellular levels of phosphatidylinositol-3,4,5-trisphosphate in cells and functions as a tumor suppressor by negatively regulating Akt/PKB signaling pathway. Delivery of PTEN by microcapsules was shown to improve the radiocurability of prostate xenografts. Authophagy or “selfeating” is frequently activated in tumor cells treated with chemotherapy or irradiation. Inhibition of genes involved in autophagy using small interfering RNAs (siRNAs) potentiated the effects of irradiation in vitro. All of these approaches warrant further investigation.

8.4 Conclusion With the exception of perhaps ex vivo vaccine-based strategies, gene therapy has not demonstrated significant activity against metastatic disease in the clinic. This is not surprising given the low efficiency of gene transfer in vivo when delivered systemically. Until this limitation is overcome, it makes more sense to apply gene therapy as an adjuvant (meaning auxiliary) to loco-regional cancer treatments such as radiation therapy. As discussed above, a variety of gene therapy strategies designed to improve the effectiveness of radiation therapy have been developed to date. Of those translated into the clinic, all have been well tolerated and have not exacerbated the side effects of radiation therapy. Several have generated provocative


181

antitumor activity in early-stage trials. These preliminary results await confirmation in larger randomized trials, which are in progress. In the future, we should focus our efforts on improving the efficiency of gene transfer (both locally and systemically) as well as targeting pathways that play a role in the radiation response and tumor radioresistance. Much has been learned about these pathways in recent years (discussed elsewhere) and this knowledge will undoubtedly lead to the development of new gene therapy strategies designed to make the tumor cell more susceptible to the therapeutic effects of radiation. However, as with all cancer therapies, it will be important that these new strategies specifically target the malignant cell while sparing normal tissue. Although much work lies ahead, we believe combining gene therapy with radiotherapy will someday earn a place in the management of human cancer. Acknowledgment This work was supported by a grant (P01 CA097012) from the National Institutes of Health to SOF.

References Aghi M, Kramm C, Chou T et al (1998) Synergistic anticancer effects of ganciclovir/thymidine kinase and 5-fluorocytosine/cytosine deaminase gene therapies. J Natl Cancer Inst 90:370–380 Apel A, Herr I, Schwarz H et al (2008) Blocked autophagy sensitizes resistant carcinoma cells to radiation therapy. Cancer Res 68:1485–1494 Arafat W, Buchsbaum D, Gomez-Navarro J et al (2003) An adenovirus encoding proapoptotic Bax synergistically radiosensitizes malignant glioma. Int J Radiat Oncol Biol Phys 55:1037–1050 Arafat W, Gomez-Navarro J, Xiang J et al (2000) An adenovirus encoding proapoptotic Bax induces apoptosis and enhances the radiation effect in human ovarian cancer. Mol Ther 1:545–554 Asher A, Mule J, Reichert C et al (1987) Studies on the anti-tumor efficacy of systemically administered recombinant tumor necrosis factor against several murine tumors in vivo. J Immunol 138:963–974 Badie B, Kramer M, Lau R et al (1998) Adenovirus-mediated p53 gene delivery potentiates the radiation-induced growth inhibition of experimental brain tumors. J Neurooncol 37:217–222 Barton K, Stricker H, Brown S et al (2008) Phase I study of noninvasive imaging of adenovirusmediated gene expression in the human prostate. Mol Ther 16:1761–1769 Belka C, Schmid B, Marini P et al (2001) Sensitization of resistant lymphoma cells to irradiationinduced apoptosis by the death ligand TRAIL. Oncogene 20:2190–2196 Black M, Kokoris M, Sabo P (2001) Herpes simplex virus-1 thymidine kinase mutants created by semi-random sequence mutagenesis improve prodrug-mediated tumor cell killing. Cancer Res 61:3022–3026 Black M, Newcomb T, Wilson H et al (1996) Creation of drug-specific herpes simplex virus type 1 thymidine kinase mutants for gene therapy. Proc Natl Acad Sci USA 93:3525–3529 Boucher P, Im M, Freytag S et al (2006) A novel mechanism of synergistic cytotoxicity with 5-FC and GCV in double suicide gene therapy. Cancer Res 66:3230–3237 Carswell E, Olds L, Kassel R et al (1975) An endotoxin-induced serum factor that causes necrosis of tumors. Proc Natl Acad Sci USA 72:366–3670 Chang K, Fisher W, Kenady D et al (2009) Multi-center randomized controlled phase 3 clinical trial using TNFerade with chemoradiation in patients with locally advanced pancreatic cancer: interim analysis of overall survival. ASCO, Abstract 4605

182

S.O. Freytag et al.

Chapman PB, Lester TJ, Casper ES et al (1987) Clinical pharmacology of human tumor necrosis factor in patients with advanced cancer. J Clin Oncol 5:1942–1951 Chen Y, DeWeese T, Dilley J et al (2001) CV706, a prostate cancer-specific adenovirus variant, in combination with radiotherapy produces synergistic antitumor efficacy without increasing toxicity. Cancer Res 61:5453–5460 Chhikara M, Huang H, Vlachaki M et al (2001) Radio-gene therapy: enhanced therapeutic effect of HSV-tk-GCV gene therapy and ionizing radiation for prostate cancer. Mol Ther 3:536–542 Chinnaiyan A, Prasad U, Shankar S et al (2000) Combined effect of tumor necrosis factor-related apoptosis-inducing ligand and ionizing radiation in breast cancer therapy. Proc Natl Acad Sci USA 97:1754–1759 Chung T, Mauceri H, Hallahan D et al (1998) Tumor necrosis factor-based gene therapy enhances radiation cytotoxicity in human prostate cancer. Cancer Gene Ther 5:344–349 Clayman G, el-Naggar A, Lippman S et al (1998) Adenovirus-mediated p53 gene transfer in patients with advanced recurrent head and neck squamous cell carcinoma. J Clin Oncol 16:2221–2232 Clayman G, Frank D, Bruso P et al (1999) Adenovirus mediated wild-type p53 gene transfer as a surgical adjuvant in advanced head and neck cancers. Clin Cancer Res 5:1715–1722 Collis S, Ketner G, Hicks J et al (2003a) Expression of the DNA-PK binding protein E4-34K fails to confer radiation sensitivity to mammalian cells. Int J Radiat Biol 79:53–60 Collis S, Swartz M, Nelson W et al (2003b) Enhanced radiation and chemotherapy-mediated cell killing of human cancer cells by small inhibitory RNA silencing of DNA repair factors. Cancer Res 63:1550–1554 Creasey A, Reynolds M, Laird W (1986) Cures and partial regression of murine and human tumors by recombinant human tumor necrosis factor. Cancer Res 46:5687–5690 Culver K, Ram Z, Wallbridge S et al (1992) In vivo gene transfer with retroviral vector-producer cells for treatment of experimental brain tumors. Science 256:1550–1552 Feinberg B, Kurzrock R, Talpaz M et al (1988) A phase I trial of intravenously administered recombinant tumor necrosis factor-alpha in cancer patients. J Clin Oncol 6:1328–1334 Freytag S, Barton K, Zhang Y et al (2007a) Replication-competent adenovirus-mediated suicide gene therapy with radiation in a preclinical model of pancreatic cancer. Mol Ther 15:1600–1606 Freytag S, Movsas B, Aref I et al (2007b) Phase I trial of replication-competent adenovirus-mediated suicide gene therapy with IMRT for prostate cancer. Mol Ther 15:1016–1023 Freytag S, Paielli D, Wing M et al (2002) Efficacy and toxicity of replication-competent adenovirusmediated double suicide gene therapy in combination with radiation therapy in an orthotopic mouse prostate cancer model. Int J Radiat Oncol Biol Phys 54:873–886 Freytag S, Rogulski K, Paielli D et al (1998) A novel three-pronged approach to selectively kill cancer cells: concomitant viral, double suicide gene, and radiotherapy. Hum Gene Ther 9:1323–1333 Freytag S, Stricker H, Pegg J et al (2003) Phase I study of replication-competent adenovirus-mediated double suicide gene therapy in combination with conventional dose three-dimensional conformal radiation therapy for the treatment of newly diagnosed, intermediate- to high-risk prostate cancer. Cancer Res 63:7497–7506 Gable M, Kim J, Kolozsvary A et al (1997) Selective in vivo radiosensitization by 5-fluorocytosine of human colorectal carcinoma cells transduced with the E. coli cytosine deaminase (CD) gene. Int J Radiat Oncol Biol Phys 41:883–887 Gallardo D, Drazan K, McBride W (1996) Adenovirus-based transfer of wild-type p53 gene increase ovarian tumor radiosensitivity. Cancer Res 56:4891–4893 Hallahan D, Mauceri J, Seung L et al (1995) Spatial and temporal control of gene therapy using ionizing radiation. Nat Med 1:786–791 Hamada M, Fujiwara T, Hizuta A et al (1996) The p53 gene is a potent determinant of chemosensitivity and radiosensitivity in gastric and colorectal cancers. J Cancer Res Clin Oncol 122:360–365


183

Hamstra D, Rice D, Fahmy S et al (1999) Enzyme/prodrug therapy for head and neck cancer using a catalytically superior cytosine deaminase. Hum Gene Ther 10:1993–2003 Hanna N, Mauceri H, Wayne J et al (1997) Virally directed cytosine deaminase/5-fluorocytosine gene therapy enhances radiation response in human cancer xenografts. Cancer Res 57:4205–4209 Huber B, Austin E, Richards C et al (1994) Metabolism of 5-fluorocytosine to 5-fluorouracil in human colorectal tumor cells transduced with the cytosine deaminase gene: significant antitumor effects when only a small percentage of tumor cells express cytosine deaminase. Proc Natl Acad Sci USA 91:8302–8306 Hwang H, Davis T, Houghton J et al (2000) Radiosensitivity of thymidylate synthase-deficient human tumor cells is affected by progression through the G1 restriction point into S-phase: implications for fluoropyrimidine radiosensitization. Cancer Res 60:92–100 Immonen A, Vapalahti M, Tyynelä K et al (2004) AdvHSV-tk gene therapy with intravenous ganciclovir improves survival in human malignant glioma: a randomised, controlled study. Mol Ther 10:967–972 Jung M, Zhang Y, Lee S et al (1995) Correction of radiation sensitivity in ataxia telangiectasia cells by a truncated I kappa B-alpha. Science 268:1619–1621 Kasid U, Dritschilo A (2003) RAF antisense oligonucleotide as a tumor radiosensitizer. Oncogene 22:5876–5884 Kawabe S, Nishikawa T, Munshi A et al (2002) Adenovirus-mediated mda-7 gene expression radiosensitizes non-small cell lung cancer cells via TP53-independent mechanisms. Mol Ther 6:637–644 Khil M, Kim J, Mullen C et al (1995) Radiosensitization by 5-fluorocytosine of human colorectal carcinoma cells in culture transduced with cytosine deaminase gene. Clin Cancer Res 2:53–57 Kievit E, Bershad E, Ng E et al (1999) Superiority of yeast over bacterial cytosine deaminase for enzyme/prodrug gene therapy in colon cancer xenografts. Cancer Res 59:1417–1421 Kievit E, Nyati M, Ng E et al (2000) Yeast cytosine deaminase improves radiosensitization and bystander effect by 5-fluorocytosine of human colorectal cancer xenografts. Cancer Res 60:6649–6655 Kim J, Kim S, Brown S et al (1994) Selective enhancement by an antiviral agent of the radiationinduced cell killing of human glioma cells transduced with HSV-tk gene. Cancer Res 54:6053–6056 Kim J, Kim S, Kolozsvary A et al (1995) Selective enhancement of radiation response of herpes simplex virus thymidine kinase transduced 9L gliosarcoma cells in vitro and in vivo by antiviral agents. Int J Radiat Oncol Biol Phys 33:861–868 Kim J, Kolozsvary A, Rogulski K et al (1998) Double suicide fusion gene: a selective radiosensitization of 9L glioma in rat brain. Cancer J Sci Am 4:364–369 Kim S, Kim J, Kolozsvary A et al (1997) Preferential radiosensitization of 9L glioma cells transduced with HSV-TK gene by acyclovir. J Neurooncol 33:189–194 Lamfers M, Grill J, Dirven C et al (2002) Potential of the conditionally replicative adenovirus Ad5-Delta24RGD in the treatment of malignant gliomas and its enhanced effect with radiotherapy. Cancer Res 62:5736–5742 Lammering G, Hewit T, Valerie K et al (2003) Anti-erbB receptor strategy as a gene therapeutic intervention to improve radiotherapy in malignant human tumours. Int J Radiat Biol 79:561–568 Lammering G, Lin P, Contessa J et al (2001) Adenovirus-mediated overexpression of dominant negative epidermal growth factor receptor-CD533 as a gene therapeutic approach radiosensitizes human carcinoma and malignant glioma cells. Int J Radiat Oncol Biol Phys 51:775–784 Lang F, Bruner J, Fuller G et al (2003) Phase I trial of adenovirus-mediated p53 gene therapy for recurrent glioma: biological and clinical results. J Clin Oncol 21:2508–2518 Li G, He F, Shao X et al (2003) Adenovirus-mediated heat-activated antisense Ku70 expression radiosensitizes tumor cells in vitro and in vivo. Cancer Res 63:3268–3274

184

S.O. Freytag et al.

Li J, Lax S, Kim J et al (1999) The effects of combining ionizing radiation and adenoviral p53 gene therapy in nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 43:607–616 Lowe S, Bodis S, McClatchey A et al (1994) p53 status and the efficacy of cancer therapy in vivo. Science 266:807–810 Lowe S, Ruley H, Jacks T et al (1993) p53-dependent apoptosis modulates the cytotoxicity of anticancer agents. Cell 74:957–967 Marangoni E, Le Romancer M, Foray N et al (2000) Transfer of Ku86 RNA antisense decreases the radioresistance of human fibroblasts. Cancer Gene Ther 7:339–346 McGinn C, Shewach D, Lawrence T (1996) Radiosensitizing nucleosides. J Natl Cancer Inst 88:1193–1203 McLoughlin J, McCarty T, Cunningham C et al (2005) TNFerade, an adenovector carrying the transgene for human tumor necrosis factor alpha, for patients with advanced tumors: surgical experience and long-term follow-up. Ann Surg Oncol 12:825–830 Moolten F, Wells J (1990) Curability of tumors bearing herpes thymidine kinase genes transferred by retroviral vectors. J Natl Cancer Inst 82:297–300 Mukogawa T, Koyama F, Tachibana M et al (2003) Adenovirus-mediated gene transduction of truncated I kappa B alpha enhances radiosensitivity in human colon cancer cells. Cancer Sci 94:745–750 Mundt A, Vijayakumar S, Nemunaitis J et al (2004) A phase I trial of TNFearde biologic in patients with soft tissue sarcoma in the extremities. Clin Cancer Res 10:5747–5753 Nemunaitis J, Swisher S, Timmons T et al (2000) Adenovirus-mediated p53 gene transfer in sequence with cisplatin to tumors of patients with non-small-cell lung cancer. J Clin Oncol 18:609–622 Nussenzweig A, Sokol K, Burgman P et al (1997) Hypersensitivity of Ku80-deficient cell lines and mice to DNA damage: the effects of ionizing radiation on growth, survival, and development. Proc Natl Acad Sci USA 94:13588–13593 Pagliaro L, Keyhani A, Williams D et al (2003) Repeated intravesical instillations of an adenoviral vector in patients with locally advanced bladder cancer: a phase I study of p53 gene therapy. J Clin Oncol 21:2247–2253 Pajonk F, Pajonk K, McBride W (1999) Inhibition of NF-kappa B, clonogenicity, and radiosensitivity of human cancer cells. J Natl Cancer Inst 91:1956–1960 Qiao J, Black M, Caruso M (2000) Enhanced ganciclovir killing and bystander effect of human tumor cells transduced with a retroviral vector carrying a herpes simplex virus thymidine kinase gene mutant. Hum Gene Ther 11:1569–1576 Rainov N (2000) A phase III clinical evaluation of herpes simplex virus type 1 thymidine kinase and ganciclovir gene therapy as an adjuvant to surgical resection and radiation in adults with previously untreated glioblastoma multiforme. Hum Gene Ther 11:2389–2401 Rogulski K, Freytag S, Zhang K et al (2000b) Anti-tumor activity of ONYX-015 is influenced by p53 status and is augmented by radiotherapy. Cancer Res 60:1193–1196 Rogulski K, Kim J, Kim S et al (1997a) Glioma cells transduced with an E. coli CD/HSV-1 TK fusion gene exhibit enhanced metabolic suicide and radiosensitivity. Hum Gene Ther 8:73–85 Rogulski K, Wing M, Paielli D et al (2000a) Double suicide gene therapy augments the therapeutic efficacy of an oncolytic adenovirus through enhanced cytotoxicity and radiosensitization. Hum Gene Ther 11:67–76 Rogulski K, Zhang K, Kolozsvary A et al (1997b) Pronounced anti-tumor effects and tumor radiosensitization of double suicide gene therapy. Clin Cancer Res 3:2081–2088 Rooney S, Sekiguchi J, Zhu C et al (2002) Leaky SCID phenotype associated with defective V(D) J coding end processing in Artemis-deficient mice. Mol Cell 10:1379–1390 Sah N, Munshi A, Nishikawa T et al (2003) Adenovirus-mediated wild-type p53 radiosensitizes human tumor cells by suppressing DNA repair capacity. Mol Cancer Ther 2:1123–1131 Senzer N, Mani S, Rosemurgy A, Nemunaitis J et al (2004) TNFerade biologic, an adenovector with a radiation-inducible promoter, carrying the human tumor necrosis factor alpha gene: a phase I study in patients with solid tumors. J Clin Oncol 22:592–601


185

Shao R, Karunagaran D, Zhou B et al (1997) Inhibition of nuclear factor-kappa B activity is involved in E1A-mediated sensitization of radiation-induced apoptosis. J Biol Chem 272:32739–32742 Shao R, Tsai E, Wei K et al (2001) E1A inhibition of radiation-induced NF-kappa B activity through suppression of IKK activity and I-kappa B degradation, independent of Akt activation. Cancer Res 61:7413–7416 Soldatenkov V, Dritschilo A, Wang F et al (1997) Inhibition of Raf-1 protein kinase by antisense phosphorothioate oligodeoxyribonucleotide is associated with sensitization of human laryngeal squamous carcinoma cells to gamma radiation. Cancer J Sci Am 3:13–20 Spitz F, Nguyen D, Skibber J et al (1996) Adenoviral-mediated wild-type p53 gene expression sensitizes colorectal cancer cells to ionizing radiation. Clin Cancer Res 2:1665–1671 Spriggs DR, Sherman ML, Michie H et al (1988) Recombinant tumor necrosis factor administered as a 24-hour intravenous infusion. A phase I and pharmacologic study. J Natl Cancer Inst 80:1039–1044 Staba M, Mauceri H, Kufe D et al (1998) Adenoviral TNF-a gene therapy and radiation damage tumor vasculature in a human malignant glioma xenograft. Gene Ther 5:293–300 Stackhouse M, Pederson L, Grizzle W et al (2000) Fractionated radiation therapy in combination with adenoviral delivery of the cytosine deaminase gene and 5-fluorocytosine enhances cytotoxic and antitumor effects in human colorectal and cholangiocarcinoma models. Gene Ther 7:1019–1026 Su Z, Lebedeva I, Sarkar D et al (2003) Melanoma differentiation associated gene-7, mda-7/IL-24, selectively induces growth suppression, apoptosis and radiosensitization in malignant gliomas in a p53-independent manner. Oncogene 22:1164–1180 Swisher S, Roth J, Komaki R et al (2003) Induction of p53-regulated genes and tumor regression in lung cancer patients after intratumoral delivery of adenoviral p53 (INGN 201) and radiation therapy. Clin Cancer Res 9:93–101 Swisher S, Roth J, Nemunaitis J et al (1999) Adenovirus-mediated p53 gene transfer in advanced non-small cell lung cancer. J Natl Cancer Inst 91:763–771 Teh B, Aguilar-Cordova E, Kernen K et al (2001) Phase I/II trial evaluating combined radiotherapy and in situ gene therapy with or without hormonal therapy in the treatment of prostate cancer – a preliminary report. Int J Rad Oncol Biol Phys 51:605–613 Teh BS, Ayala G, Aguilar L et al (2004) Phase I–II trial evaluating combined intensity-modulated radiotherapy and in situ gene therapy with or without hormonal therapy in treatment of prostate cancer-interim report on PSA response and biopsy data. Int J Radiat Oncol Biol Phys 58:1520–1529 Tomioka A, Tanaka M, De Velasco M et al (2008) Delivery of PTEN via a novel gene microcapsule sensitizes prostate cancer cells to irradiation. Mol Cancer Ther 7:1864–1870 Tong A, Nemunaitis J, Su D et al (2005) Intratumoral injection of INGN 241, a nonreplicating adenovector expressing the melanoma-differentiation associated gene-7 (mda-7/IL24): biologic outcome in advanced cancer patients. Mol Ther 11:160–172 Toth K, Tarakanova V, Doronin K et al (2003) Radiation increases the activity of oncolytic adenovirus cancer gene therapy vectors that overexpress the ADP (E3-11.6K) protein. Cancer Gene Ther 10:193–200 Uckert W, Kammertons T, Haack K et al (1998) Double suicide gene (cytosine deaminase and herpes simplex virus thymidine kinase) but not single gene transfer allows reliable elimination of tumor cells in vivo. Hum Gene Ther 9:855–865 Vlachaki M, Chhikara M, Aguilar L et al (2001) Enhanced therapeutic effect of multiple injections of HSV-TK + GCV gene therapy in combination with ionizing radiation in a mouse mammary tumor model. Int J Radiat Oncol Biol Phys 52:1008–1017 Wang Z, Cook T, Alber S et al (2004) Adenoviral gene transfer of the human inducible nitric oxide synthase gene enhances the radiation response of human colorectal cancer associated with alterations in tumor vascularity. Cancer Res 64:1386–1395 Weichselbaum R, Hallahan D, Beckett M et al (1994) Gene therapy targeted by radiation preferentially radiosensitizes tumor cells. Cancer Res 54:4266–4269

186

S.O. Freytag et al.

Weill D, Mack M, Roth J et al (2000) Adenoviral-mediated p53 gene transfer to non-small cell lung cancer through endobronchial injection. Chest 118:966–970 Wolf J, Bodurka D, Gano J et al (2004) A phase 1 study of Adp53 (INGN201; ADVEXIN) for patients with platinum and paciltaxel-resistant epithelial ovarian cancer. Gynecol Oncol 94:442–448 Xie Y, Gilbert J, Kim J et al (1999) Efficacy of adenovirus-mediated CD/5-FC and HSV-1 TK/ GCV suicide gene therapies concomitant with p53 gene therapy. Clin Cancer Res 5:4224–4232 Yacoub A, Mitchell C, Lister A et al (2003) Melanoma differentiation-associated 7 (interleukin 24) inhibits growth and enhances radiosensitivity of glioma cells in vitro and in vivo. Clin Cancer Res 9:3272–3281 Yoo G, Moon J, LeBlanc M et al (2009) A phase 2 trial of surgery with perioperative INGN 201 (Ad5CMV-p53) gene therapy followed by chemoradiotherapy for advanced, resectable squamous cell carcinoma of the oral cavity, oropharynx, hypopharynx, and larynx. Arch Otolaryngol Head Neck Surg 135:869–874

Chapter 9

Molecular Targeted Drug Delivery Radiotherapy Eugenia M. Yazlovitskaya and Dennis E. Hallahan

Abstract This chapter discusses prosurvival signal transduction pathways, such as PI3K/Akt and MAPK/ERK signaling, induced by ionizing radiation in the tumor microvasculature. Molecular targeting of these radiation-induced signaling pathways provides a means to enhance tumor control. Preclinical studies show that this approach improves the efficacy of radiotherapy in mouse models of cancer. Keywords Signal transduction pathways • Molecular targeting • Radiotherapy

9.1 Introduction Roughly 50% of cancer patients are treated with ionizing radiation (IR) therapy (Owen et al. 1992). Despite this wide use and effectiveness, radiotherapy has limitations, such as restricted tolerance of normal tissues and tumor radioresistance. One approach to enhance the therapeutic ratio of ionizing radiation is to combine it with chemotherapy (Lawrence et al. 2003; Wilson et al. 2006). In general, radiotherapy and chemotherapy target deregulated cancer cells, and the initial rationale for the combination of radiation and chemotherapeutic agents was based on the “spatial cooperation” concept (Hall and Giaccia 2006). More recently, a new paradigm had emerged from the findings that the host component of cancer (e.g., microvasculature, stroma, and immune system) is an important target for the effects of radiation and cytotoxic drugs (Lu et al. 2005; Kim et al. 2006; Heath and Bicknell 2009). For example, the tumor vasculature is particularly interesting because of its resistance to clinically relevant low IR doses of 2–5 Gy (Geng et al. 2001; Edwards et al. 2002; Tan et al. 2006; Linkous et al. 2009).

D.E. Hallahan (*) Department of Radiation Oncology, Washington University School of Medicine, 4511 Forest Park, Suite 200, St. Louis, MO 63130, USA e-mail: [email protected]

T.L. DeWeese and M. Laiho (eds.), Molecular Determinants of Radiation Response, Current Cancer Research, DOI 10.1007/978-1-4419-8044-1_9, © Springer Science+Business Media, LLC 2011

187

188

E.M. Yazlovitskaya and D.E. Hallahan

9.2 Role of Vasculature in Development and Treatment of Solid Tumors Recent preclinical studies suggest that the combination of radiotherapy with antiangiogenic or vascular targeting agents enhances the therapeutic effect of ionizing radiation (Siemann and Horsman 2004; Wachsberger et al. 2004; Kobayashi and Lin 2006). Using tumor vascular endothelium for molecular targeted therapy has specific advantages. Formation of new blood vessels is very limited in developed tissues, thus, antiangiogenic drugs targeting proliferating tumor vascular endothelium causes little normal tissue toxicity. Most importantly, each tumor capillary supplies hundreds of tumor cells leading to potentiation of therapeutic effect (Wachsberger et al. 2003). The response of tumor microvasculature to radiation is dependent upon the dose and time interval after treatment. Tumor blood flow decreases when tumors are treated with single doses in the range of 20–45 Gy (Song et al. 1972). One of the underlying reasons for this response is that radiation doses in the range of 15–20 Gy given as a single treatment induce apoptosis in tumor vascular endothelial cells (Schwentker et al. 1998; Garcia-Barros et al. 2003, 2004). This apoptosis was shown to be dependent upon acid sphingomyelinase-mediated production of ceramide (Paris et al. 2001; Garcia-Barros et al. 2003, 2004), FGF signaling pathway (Paris et al. 2001), and/or Bax (Garcia-Barros et al. 2003, 2004). Tumor blood vessels show less response to radiation doses in the range of 2–3 Gy, which is typically used to treat cancer (Geng et al. 2001; Edwards et al. 2002; Tan et al. 2006; Linkous et al. 2009). Endothelial cells grown in culture are also resistant to the low doses of IR (Edwards et al. 2002; Tan et al. 2006; Yazlovitskaya et al. 2008). Molecular mechanisms of this radioresistance are not known; however, some recent studies suggest the involvement of prosurvival signaling pathways, such as activation of the receptor tyrosine kinases (RTK) (Geng et al. 2001; Kim et al. 2004; Yazlovitskaya et al. 2008), Src pathway (Cuneo et al. 2006), PI3K/Akt (Edwards et al. 2002; Tan and Hallahan 2003; Tan et al. 2006; Yazlovitskaya et al. 2008; Linkous et al. 2009), and ERK ( Maclachlan et al. 2005; Yazlovitskaya et al. 2008; Linkous et al. 2009). Therefore, blocking of radiation-induced activation of these pathways could potentially abrogate radioresistance of the tumor vasculature and potentiate tumor response to radiotherapy.

9.3 Radiation-Induced Prosurvival Signal Transduction Pathways Exposure of cells to ionizing radiation induces multiple cellular and biological effects by either direct DNA damage or through the formation of free radical species (Fig. 9.1a). Accumulating evidence suggests that these initial events activate multiple signaling pathways (Valerie et al. 2007). A balance exists in the

9 Molecular Targeted Drug Delivery Radiotherapy

189

activation of prosurvival and cell killing pathways which determine cellular response to radiation, ultimately controlling cell metabolic, proliferative, or death processes (Fig. 9.1a) (Schmidt-Ullrich et al. 2000; Dent et al. 2003a; Valerie et al. 2007). The main signal transduction pathways activated by ionizing radiation in a

Ionizing radiation

DNA damage

Free radical species

Receptor activation

Death receptors

Pro-survival receptors

Cell death kinase cascades

Pro-survival kinase cascades

Cell death

Cell survival

b PC

cPLA2 RTK

LPC

LPLD/ATX

GPCR

LPA

cPLA2

PA

PLD

PC

GPCR

c

AA, LPC

Fig. 9.1 Radiation-induced signal transduction pathways. (a) Schematic representation of r adiation-induced signal transduction pathways, including prodeath and prosurvival paths, and their interaction. (b) Schematic representation of metabolic transformations of cellular phospholipids (PC, LPC, PA, LPA) by lipases (cPLA2, LPLD/ATX, PLD) and subsequent activation of receptors (GPCR, RTK). (c) (Adapted from Hirabayashi et al. 2004.) Proposed model for cPLA2a activation. Ca2+ binding to the C2 domain promotes cPLA2a translocation from the cytosol to the membrane containing phosphatidylcholine (PC). Phosphorylation on serine residues and/or binding to anionic phospholipids stabilize the association of cPLA2a with the membrane and increase its catalytic activity

190


vascular endothelium are phosphatidylinositol 3-kinase/Akt (PI3K/Akt) and mitogen-activated protein kinase/extracellular signal-regulated kinase (MAPK/ ERK) pathways. Radiation-induced activation of these pathways occurs very early, at 1–15 min after exposure (Hallahan et al. 1992; Edwards et al. 2002; Dent et al. 2003b; Maclachlan et al. 2005; Tan et al. 2006; Yazlovitskaya et al. 2008; Linkous et al. 2009).

9.3.1 PI3K/Akt Pathway The PI3K/Akt signal transduction pathway has been implicated in survival signaling in various cell types (Brazil and Hemmings 2001; Vivanco and Sawyers 2002). In vascular endothelial cells, PI3K/Akt can be activated through a family of receptor tyrosine kinases (RTKs) that are activated by growth factors (Burgering and Coffer 1995; Mazure et al. 1997). For example, Akt signaling participates in angiogenesis following VEGF stimulation of endothelial cells and regulates capillarylike tubule formation (Gerber et al. 1998). In addition, Akt can be activated independently of growth factors or PI3K signaling (Bianco et al. 2003). Recently, it has been shown that radiation induces phosphorylation of Akt within endothelial cells (Tan and Hallahan 2003). Low dose of irradiation (2–3 Gy) activates the PI3K/ Akt-mediated cell viability pathway, which also involves inhibitory phosphorylation of the proapoptotic kinase GSK-3b (Tan et al. 2006). Taken together, growing evidence suggests that radiation-induced PI3K/Akt pathway is a vital target for enhancement of the radiation response in tumor vascular endothelium. Inhibition of PI3K/Akt pathway at the various steps of activation prior to IR leads to the potentiation of radiation-induced cell death in vascular endothelial cells. Inhibition of RTKs or inhibition of their ligands was studied (Kim et al. 2004, 2006). Much of the effort has been directed at compounds that specifically inhibit VEGF, such as neutralizing antibody (Gorski et al. 1999), or VEGFR, such as DC101 and SU5416 (Geng et al. 2001). The broad-spectrum RTK inhibitors have the advantage of targeting multiple RTKs that may be present in the tumor, its microenvironment, or both. One such small molecule is PTK787 that targets VEGFR2, PDGF, and c-Met (Wood et al. 2000). Other examples are SU11248 and SU6668. SU11248 is a selective inhibitor of VEGFR2 and PDGFR, as well as Kit and VEGFR1 (Mendel et al. 2003). SU6668 was designed to inhibit kinase activity of FGF, PDGF, and VEGF receptors and have demonstrated significant radiation sensitization effects in preclinical models (Lu et al. 2004). Inhibition of PI3K, Akt, and their down-stream targets was evaluated. Inhibition of p110 subunit of PI3K (Geng et al. 2004) either using specific chemical inhibitors (wortmannin, LY294002, IC486068) or overexpression of the mutant p85 component of PI3K-enhanced radiation-induced apoptosis and minimized capillary tubule formation (Edwards et al. 2002; Tan and Hallahan 2003; Geng et al. 2004; Tan et al. 2006). Chemical inhibition of Akt with ALX-349 as well as overexpression of dominant-negative mutants Akt or its down-stream target GSK-3b also lead to


191

radiosensitization of vascular endothelium (Tan et al. 2006). Another down-stream target of PI3K/Akt signaling, mTOR, was shown to be a promising molecule for radiosensitization. Inhibitors of mTOR RAD001 and rapamycin were potent radiosensitizers of vascular endothelial cells in vitro and led to improved tumor-growth delay in vivo (Shinohara et al. 2005; Albert et al. 2006).

9.3.2 MAPK/ERK Pathway Activation of “classical” MAP kinases, ERK1 and 2, by IR is dependent on the cell type, the expression of multiple growth factor receptors, and genetic alterations (Dent et al. 2003b; Lammering et al. 2004). In some cases, Inhibition of MEK1/2, upstream kinases regulating ERK1/2 activity, enhances cell killing by radiation due to increased G2/M arrest and apoptosis (Abbott and Holt 1999); while in some cell types, activation of ERK pathway following irradiation has been shown to promote radiosensitivity by abrogating G2/M checkpoint (McKinstry et al. 2002). Radiationinduced ERK activation was linked to increased expression of radioprotective transcription factors and DNA repair proteins (Amorino et al. 2003; Yacoub et al. 2003). More importantly, ERK activation in endothelial cells plays a radioprotective role resulting in increased survival of irradiated cells (Maclachlan et al. 2005; Yazlovitskaya et al. 2008; Linkous et al. 2009). The cytotoxic effect of radiation can be enhanced by the inhibition of radiation-induced activity of both Akt and ERK (Dent et al. 2003b; Taira et al. 2006).

9.3.3 Phospholipids and Cytosolic Phospholipase A2 (cPLA2) Mechanisms of radiation-induced activation of PI3K/Akt and MAPK/ERK pathways in vascular endothelium involve biologically active lipids and proteins, such as phospholipases, lipid kinases, and phosphatases, which regulate production and activity of these lipids (Cabral 2005; Farooqui and Horrocks 2006). One group of cellular lipids involved in signal transduction comprises the phospholipids. Major phospholipids found in membranes of mammalian cells are phosphatidylcholine (PC) and phosphatidylinositol (Ivanova et al. 2004). Phospholipid metabolism is regulated through G-protein- and tyrosine kinase-mediated enzymes, including phospholipases A, C, and D (PLA, PLC, PLD) as well as cyclooxygenases 1 and 2, lipid kinases and phosphatases (Ivanova et al. 2004). These enzymes convert lipids into bioactive signaling molecules, such as arachidonic acid (AA), phosphatidic acid (PA), diacylglycerol, phosphorylated phosphatidylinositols, lysophosphatidic acid (LPA), and lysophosphatidylcholine (LPC). Phospholipases are enzymes which hydrolyze phospholipids. Phospholipase A2 (PLA2) hydrolyzes phospholipids at the sn-2-acyl ester bond, generating free fatty acids and lysophospholipids (Chakraborti 2003) (Fig. 9.1b). The PLA2 superfamily can be divided into ten groups by gene sequences. Based on

192


biological properties, the classification of PLA2s can be simplified into three main types: the cytosolic (cPLA2), the secretory (sPLA2) and the intracellular Ca2+independent (iPLA2) (Chakraborti 2003). The cPLA2, or group IV PLA2, includes at least four paralogs in mammals: PLA2a, -b, -g, and -d (Chakraborti 2003; Hirabayashi et al. 2004). cPLA2a is the most ubiquitously expressed enzyme in the group. The protein contains N-terminal regulatory C2 domain, which is involved in the Ca2+-dependent phospholipid binding, and two catalytic domains A and B (Nalefski et al. 1994) (Fig. 9.1c, adapted from Hirabayashi et al. 2004). Upon membrane binding, conformational changes remove the lid from the active site and allow the fatty acid chain of a substrate molecule (PC) to enter the active site (Fig. 9.1c, adapted from Hirabayashi et al. 2004). Regulatory mechanisms of cPLA2a activity are complex and include subcellular localization, intracellular Ca2+ content, phosphorylation, protein–protein interaction, and cleavage (Chakraborti 2003; Hirabayashi et al. 2004). The specific structural changes and regulatory mechanisms involved in activation of cPLA2 serve as a foundation for the design of chemical inhibitors of cPLA2 (Farooqui et al. 2006). We have recently determined that activation of cPLA2 is an initial event (within 2 min) required for radiation-induced activation of Akt and ERK1/2 in vascular endothelial cells leading to the increased cell viability (Fig. 9.2a) (Yazlovitskaya et al. 2008). We studied cPLA2 inhibitors methyl arachidonyl fluorophosphonate (MAFP) and arachidonyl trifluoromethyl ketone (AACOCF3). AACOCF3, a cellpermeable trifluoromethyl ketone analog of arachidonic acid, is potent cPLA2 inhibitor with IC50 = 1.5 mM (Farooqui et al. 2006). NMR studies have demonstrated

a

b

c

Vascular endothelial cell

Vascular endothelial or tumor cell

Vascular endothelial or tumor cell

LPLD ATX

LPC Ionizing radiation

PC cPLA2

RTK (Flk-1)

GPCR (GPR4)

LPA GPCR (LPA1-7)

LPC Pro-survival pathways Survival, inflammation

Pro-survival pathways Survival

Fig. 9.2 Schematic representation of interaction of radiation-induced prosurvival responses in vascular endothelial and tumor cells. Radiation-induced signaling pathways involve activation of cPLA2 followed by the increased production of LPC (a). LPC in turn activates RTK or GPR4 (b). LPC can be converted to LPA by LPLD resulting in activation of LPA1-7 receptors (c)


193

that the carbon chain of AACOCF3 binds in a hydrophobic pocket of cPLA2 and the carbonyl group of AACOCF3 forms a covalent bond with serine 228 in the enzyme active site (Street et al. 1993; Trimble et al. 1993; Farooqui et al. 2006). AACOCF3 has been used extensively to study the role of cPLA2 in platelet aggregation and inflammation-associated apoptosis (Fabisiak et al. 1998; Calzada et al. 2001; Duan et al. 2001; Kirschnek and Gulbins 2006). MAFP is a potent irreversible inhibitor of cPLA2 with IC50 = 0.5 mM (Farooqui et al. 2006). Since at concentration of 5 mM it also completely inhibits iPLA2, MAFP is considered less selective inhibitor of cPLA2 than AACOCF3. In our study, inhibition of cPLA2 with both inhibitors significantly enhanced radiation-induced cytotoxicity of human umbilical vein endothelial cells (HUVEC) due to mitotic catastrophe and delayed programmed cell death (Yazlovitskaya et al. 2008). This effect was confirmed using HUVEC transfected with shRNA for cPLA2a and primary cell cultures from cPLA2a−/− mice. Endothelial functions were also affected by inhibition of cPLA2 during irradiation resulting in attenuated cell migration and tubule formation (Yazlovitskaya et al. 2008). When combined with radiation, the inhibition of cPLA2 with AACOCF3 disrupts the biological functions of the tumor vasculature, enhances destruction of tumor blood vessels and suppresses tumor growth (Fig. 9.3, reproduced from Linkous et al. 2009). This confirmed that cPLA2 contributes to vascular endothelial cell radioresistance and presents a potential molecular target for tumor sensitization to radiotherapy.

9.3.4 Signaling by the cPLA2 Products LPC and LPA Since the most represented phospholipid in mammalian membranes is PC, the main lysophospholipid produced by activated cPLA2a is lysophosphatidylcholine (LPC) (Figs. 9.1b and 9.2a). Alternatively, LPC could be generated in the reaction mediated by lecithin-cholesterol acyltransferase that transfers fatty acid residue from PC to cholesterol (Prokazova et al. 1998). This biologically active lipid works as the second messenger in the signal transduction pathways, regulating a number of cellular responses (Prokazova et al. 1998; Chakraborti 2003). It is noteworthy that cellular response to exogenously added LPC critically depends on LPC concentration. Doses less than 25 mM usually trigger cell survival and proliferation (Prokazova et al. 1998; Fujita et al. 2006; Gwak et al. 2006). LPC signaling could be mediated by receptor activation or be receptor independent. LPC may bind to the plasma membrane in receptor-independent manner or directly enter the lipid bilayer (Prokazova et al. 1998). Receptor-dependent LPC signaling could involve LPC-specific G-protein-coupled receptors (GPCRs), G2A and GPR4 (Figs. 9.1b and 9.2b) (Xu 2002). LPC stimulates proliferation in HUVEC by transactivating the VEGFR 2 and activating Akt and ERK1/2 (Figs. 9.1b and 9.2b) (Fujita et al. 2006). Increased levels of LPC has been linked directly to cytokine and chemokine production in endothelial cells by activating MAPK and PI3K/Akt pathways, thus regulating the chemotaxis of leukocyte subpopulations

194


Control

3 Gy

AACOCF3

b

20

Vascular index (%)

a

AACOCF3+ 3 Gy

15

10

∗ ∗

5

AA y CO CO CF CF 3+ 3 3 Gy

G

AA

3

co

nt

ro

l

0

3

5

7

9 11 13 15 17 19 21

Time after treatment (days)

0 R

1

+I

0

500

A

200

∗

IR

400

1000

A

600

1500

A

800

A

1000

2000

ol

1200

tr

1400

H460 2500

on

Control IR AACOCF3 AACOCF3 + IR

d

C

LLC 1600

Tumor volume (mm3)

Tumor volume (mm3)

c

Fig. 9.3 Inhibition of cPLA2 with AACOCF3 attenuates tumor vascularity and decreases tumor size in irradiated mouse models. (Reproduced from Linkous et al. 2009.) Using heterotopic tumor models of Lewis lung carcinoma (LLC) (a–c) or H460 large cell carcinoma of the lung (d), mice were treated intraperitoneally with vehicle DMSO (control) or 10 mg/kg AACOCF3 and tumors were irradiated 30 min later with 3 Gy. (a, b) Treatment was repeated for 5 consecutive days. Twenty-four hours after the final treatment, tumor blood flow was analyzed by three-dimensional Power Doppler sonography. Shown are representative images of tumor blood flow (a). Shown is a bar graph of the average percent vascular index with SEM from group of three to five animals; *p

Molecular Determinants of Radiation Response

Medical Response to Effects of Ionizing Radiation

Determinants of Financial Development

Determinants of Financial Development

Social Determinants of Health

Electromagnetic Radiation: Atomic, Molecular, and Optical Physics: Atomic, Molecular, And Optical Physics: Electromagnetic Radiation

Response

Macrosocial Determinants of Population Health

Fundamental determinants of exchange rates

Fundamental Determinants of Entrepreneurial Behaviour

Fundamental Determinants of Exchange Rates

Sound and Structural Vibration, Second Edition: Radiation, Transmission and Response

Macrosocial Determinants of Population Health

The Entropy of Radiation

Fundamentals of Radiation Chemistry

Atoms, Radiation and Radiation Protection

Atoms, Radiation, and Radiation Protection

Radiation Preservation of Foods

Fundamentals of Radiation Chemistry

Fundamentals of Atmospheric Radiation

Traces and Determinants of Linear Operators

Determinants of Animal Behaviour (Routledge Modular Psychology)

Common Determinants of Liquidity and Trading

Determinants of Tropical Savannas (IUBS Monograph Series)

Common determinants of liquidity and trading

A Treatise on the Theory of Determinants

Radiation Hydrodynamics

ATOMIC AND MOLECULAR DATA FOR RADIOTHERAPY AND RADIATION RESEARCH

Sudden Response

The Physics of Radiation Therapy

Molecular Determinants of Radiation Response

Medical Response to Effects of Ionizing Radiation

Determinants of Financial Development

Determinants of Financial Development

Social Determinants of Health

Electromagnetic Radiation: Atomic, Molecular, and Optical Physics: Atomic, Molecular, And Optical Physics: Electromagnetic Radiation

Response

Macrosocial Determinants of Population Health

Fundamental determinants of exchange rates

Fundamental Determinants of Entrepreneurial Behaviour

Fundamental Determinants of Exchange Rates

Sound and Structural Vibration, Second Edition: Radiation, Transmission and Response

Macrosocial Determinants of Population Health

The Entropy of Radiation

Fundamentals of Radiation Chemistry

Atoms, Radiation and Radiation Protection

Atoms, Radiation, and Radiation Protection

Radiation Preservation of Foods

Fundamentals of Radiation Chemistry

Fundamentals of Atmospheric Radiation

Traces and Determinants of Linear Operators

Determinants of Animal Behaviour (Routledge Modular Psychology)

Common Determinants of Liquidity and Trading

Determinants of Tropical Savannas (IUBS Monograph Series)

Common determinants of liquidity and trading

A Treatise on the Theory of Determinants

Radiation Hydrodynamics

ATOMIC AND MOLECULAR DATA FOR RADIOTHERAPY AND RADIATION RESEARCH

Sudden Response

The Physics of Radiation Therapy

Recommend Documents