Hubenak, Justin R. B.S.; Zhang, Qixu M.D., Ph.D.; Branch, Cynthia D. B.S.; Kronowitz, Steven J. M.D.
More than 1.6 million new patients in the United States were diagnosed with cancer in 2012, and almost two-thirds of these patients were treated with radiotherapy.1 The benefits of radiotherapy for cancer have been well documented for many years. These benefits, however, can be outweighed by radiation-induced damage to neighboring normal tissues as a result of either direct exposure to radiation or the so-called bystander effect, which refers to biological effects in nonirradiated cells caused by signals from irradiated cells.2,3
Radiotherapy is based on the concept that the DNA repair capacity of cells with sublethal damage from radiotherapy is generally greater in healthy cells than in cancerous cells. In other words, cancer cells are more susceptible to radiation than are normal cells. The mechanisms underlying radiotherapy-induced DNA damage and postradiotherapy DNA repair have been studied in detail; however, there still exist many gaps in knowledge on how these complex systems are entwined.
SELECTION OF ARTICLES FOR INCLUSION IN REVIEW
The PubMed and EMBASE databases were reviewed for articles on adverse effects of radiotherapy on normal tissue published from January of 2005 through May of 2012. Subsequently, abstracts of these articles were reviewed to identify articles with information relevant to the biological basis of radiotherapy-induced DNA damage and DNA repair. In addition, reference lists of the articles identified by the database search were reviewed, and referenced articles that seemed relevant were reviewed with no limitations on publication date. The database searches yielded 1751 publications. Of these, 1729 were eliminated because they did not address fundamental biology or were duplicates. A total of 22 articles were included.
TYPES OF RADIATION-INDUCED DNA DAMAGE
To address possible points of intervention to reduce normal tissue toxic effects, it is essential to first understand the methods by which ionizing radiation damages cells. Ionizing radiation is naturally encountered as cosmic rays; alpha, beta, and gamma rays; x-rays; and some portions of the ultraviolet spectrum. Visible light is not usually considered ionizing radiation but can cause upward of 1 × 105 DNA lesions per cell per day.4 On the basis of the rates at which energy is deposited into cells, the different forms of radiation can be classified as forms of low-linear-energy-transfer radiation, the most prevalent form used for cancer therapy, or high-linear-energy-transfer radiation, which is typically encountered as neutrons, heavy ions, and pions.
Low-linear-energy-transfer photon beams (photon radiation) from cesium 137, cobalt 60, iridium 192, and low-linear-energy-transfer electron beams from linear accelerators are the most common form of radiotherapy. Electron beams produced by linear accelerators are useful for treating superficial lesions because the dose is deposited near the surface. Proton beams have higher linear energy transfer than photon and electron beams and can be tuned to have low linear energy transfer through normal tissues with high linear energy transfer at the end of the beam. Neutron beams have high linear energy transfer and are used to treat some inoperable tumors, cancer of the head and neck, and prostate cancer. High-linear-energy-transfer radiation delivered using carbon ions (also called heavy ion radiation) can be effective against radioresistant cancers but is equally effective against normal tissues, raising risk.
Indirect DNA Damage
When the high-energy particles of ionizing radiation penetrate the body and liberate electrons from atoms and molecules, individual chemical bonds are destroyed, which results in highly reactive ions and ion pairs commonly referred to as reactive oxygen species. Reactive oxygen species have been proposed to account for the majority of radiation-induced cellular genetic damage.5
When water, present in high concentrations throughout the entire body, is split by ionizing radiation, free radicals such as the hydroxyl reactive oxygen species are produced. Hydroxyl radicals are highly reactive with DNA, proteins, and lipids. Hydroxyl radicals function in additional reactions across double bonds and are also able to detach hydrogen atoms from the methyl groups of thymine nucleic acids, creating mutations that must be repaired before cell division or transcription to maintain genomic stability.6 This type of damage is predominantly associated with the linear sequence of nucleotides in DNA (primary structure).
Direct DNA Damage
Ionizing radiation is also capable of directly damaging DNA and regulatory proteins. By splitting chemical bonds on the helical backbone, ionizing radiation is capable of creating single-strand breaks and double-strand breaks with any amount of exposure. Double-strand breaks are significantly more destabilizing for the genome than single-strand breaks and have a high likelihood of ultimately leading to cell death or to poor DNA repair, such as single-strand annealing, which is mutagenic.
Both low- and high-linear-energy-transfer radiation sources are capable of producing single-strand breaks and double-strand breaks; however, when the high-energy particles of low-linear-energy-transfer radiation pass through the cell, fewer ionization events are likely to occur, and most of those events are single-strand breaks. Double-strand breaks due to high- and low-linear-energy-transfer radiation are most likely to result from radiation exposure during phases of the cell cycle when chromatin is tightly condensed.
Cellular Susceptibility to DNA Damage by Cell-Cycle Phase
As radiation exposure increases, the amount of damage increases proportionally, until enough damage is accumulated via direct and indirect means that the repair mechanisms are overcome and the cell becomes senescent, enters apoptosis, or loses the ability to regulate its own growth and becomes tumorigenic.
Proliferating cells have complex, finely regulated systems in place to control each stage of the cell cycle. Cells in the resting phase (G0) must typically receive very specific signals before they begin to prepare for DNA replication, a checkpoint that makes cells in the G0 phase somewhat radioresistant. Cells in the first growth phase (G1), when the cell enlarges, are somewhat radioresistant because of noncondensed DNA and regulated checkpoints. Transition from the G1 phase into the DNA replication phase (S phase) is safeguarded by sequentially activated cyclin-dependent kinases tasked with recruitment and amplification of genes integral for genomic replication during S phase. The G1 to S phase checkpoint prevents cells with damaged DNA from replicating and is called the restriction checkpoint; critical proteins here include P53, P21, and P16, all of which are considered cyclin-dependent kinase inhibitors.7 Retinoblastoma, an important tumor suppressor protein in humans, binds transcription factors required for S-phase genes and prevents translocation of these transcription factors to the nucleus. Phosphorylation of retinoblastoma by cyclin-dependent kinases releases these transcription factors; however, DNA damage causes accumulation of P53, which in turn activates cyclin-dependent kinase inhibitors that inhibit phosphorylation of retinoblastoma, delaying the transition into S phase until repairs are performed.8
Cells in S phase are the most radioresistant, because the genes, enzymes, and proteins responsible for accurate genomic replication and repair are overexpressed in S phase compared with the other phases of the cell cycle. Once DNA replication has begun, cyclin-dependent kinase inhibitors remain prevalent to negatively regulate the progression of cells through the remainder of the cell cycle, and cyclin-dependent kinases can stop progression altogether if genomic instability is detected. Mismatch repair is also highly active during S phase and is capable of repairing DNA damage due to chemicals and radiation as well as natural replication errors.9
Cells in the second growth phase (G2) retain lingering repair enzymes from the S phase and as such are relatively radioresistant. Before mitosis (M phase) begins, the cell must pass through the G2 checkpoint, where CDC25 (cell division cycle 25) phosphatase activates M-phase promoting factor (a cyclin-dependent kinase complex) under good genomic conditions in cells without DNA damage. If DNA damage is present, ATM (ataxia telangiectasia mutated) kinase phosphorylates CDC25, which in normal cells halts progression to mitosis and can eventually lead to destruction of CDC25.10 ATM kinase is a critical enzyme for cyclin-dependent kinase–mediated checkpoints as well as the damage response in cells, interacting with other protein complexes such as P53 and H2AX in response to double-strand breaks11 (Fig. 1).
M-phase cells are especially sensitive to radiation. Upon entry into M phase, chromosomes condense, and the mitotic spindle begins to pull chromosomes apart. An ionizing event that occurs when DNA is in a condensed state has a higher probability of causing a double-strand break than does an event that occurs when DNA is not condensed. The only currently understood checkpoint in M-phase cells is based on tension in the spindle apparatus.
Susceptibility of Cancer Cells versus Normal Cells to DNA Damage
Cancer cells resemble undifferentiated stem cells more closely than they resemble the mature, differentiated cells around them. This is a result of genomic instability and failures in genetic checkpoints that allow cancer cells to escape the normal controls of the cell cycle, leading to an increase in the cells’ proliferative capacity. Failed checkpoint controls can allow damaged DNA to be passed on to daughter cells. Radiotherapy is effective in killing cancer cells because dividing cells (M-phase cells) are especially radiosensitive and cancer cells divide more often.
In addition, cancer cells generally have reduced capacity to repair genomic lesions because cancer mutations are often related to the DNA repair process, allowing for DNA damage from other stages of the cell cycle to accumulate and be passed on to daughter cells.12,13 In contrast, healthy cells subjected to low levels of ionizing radiation are generally capable of halting any ongoing replication and repairing the damage over time or, if the damage is irreparable, activating proper apoptosis pathways. Fractionated radiation capitalizes on this concept by spreading out treatments over time, allowing normal cells to regenerate before additional doses are given.
CELLULAR RESPONSE TO DNA DAMAGE
Most types of human cells handle DNA damage with a standard response, regardless of whether the damage is direct damage in the form of single-strand breaks and double-strand breaks or indirect damage caused by reactive oxygen species. Cells utilize a sensor-signal-effector mechanism driven by the initial perception of a DNA break or conformational change in the helical structure. The details of the mechanism differ according to the type of DNA lesion being processed, but the overall sensor-signal-effector pattern holds true in each case. The DNA repair pathways discussed in this section are illustrated in Figure 2.
Response to Double-Strand Breaks
When a double-strand break occurs in a normal human cell, it elicits a reaction from ATM kinase (the sensor). ATM kinase recruits the autophosphorylated form of ATM and DNA-dependent protein kinase.14 These kinases, in addition to other PI3 family kinases, phosphorylate histone H2AX on serine 139.15 The phosphorylated form is known as γ-H2AX and is an early reporter of double-strand breaks. When γ-H2AX is present, signaling a double-strand break, local DNA structure relaxes and residues become less condensed, allowing space for repair proteins to bind to the DNA and facilitate nonhomologous end joining or, if the cell is in S phase or G2 phase, homologous recombination. In homologous recombination, the strand is first resected from the 5′ end. This produces an overhang that the broken strand is matched to. Nonhomologous end joining is similar but without homologous pairs. The γ-H2AX signal protein can be visualized within seconds of initiation of a double-strand break and persists until repair is complete—for typical cells, approximately 24 to 48 hours. Other proteins (the effectors) are also present within minutes of detection of a double-strand break, including P53-binding protein 53BP1, and either guide the cell toward repair or ultimately force apoptosis by stimulating release of mitochondrial cytochrome c and subsequent activation of caspase-3.16 A double-strand break is the most difficult lesion for a cell to accurately repair and can lead to mutations, such as deletions and insertions.
Response to Single-Strand Breaks
The cellular response to single-strand breaks follows the same general pattern as the cellular response to double-strand breaks. The main pathways used by human and eukaryotic cells in response to single-strand damage are base excision repair, which is favored for minor DNA repair; nucleotide excision repair, which is favored when the helix is disrupted; and mismatch repair. Most minor DNA lesions—including most single-strand breaks, minor oxidative damage, and lesions resulting from light chemical mutagens and alkylation products—are processed with base excision repair. Damaged residues are sensed by a glycosylase (e.g., methylpurine DNA glycosylase or uracil DNA glycosylase). The damaged base is then removed, preserving the DNA backbone and creating an apurinic/apyrimidinic site. Endonucleases then excise a portion of the injured DNA strand, generating a single-strand break. Base excision repair machinery then binds to the damage site and finishes the repair by synthesizing new DNA.17 If the intermediate enzymes are disrupted and uncoupled from the repair process early, the damage site is treated as a single-strand break; otherwise, repair progresses to completion. Nucleotide excision repair is similar to base excision repair for damage such as bulky adducts and pyrimidine dimers. Mismatch repair is used primarily to address erroneous incorporation of bases during replication, but it also plays a role in repairing some forms of DNA damage.
In one mechanism of single-strand break repair, poly(ADP-ribose) polymerase acts as a damage sensor and signaling protein.18 Upon binding to a single-strand break, poly(ADP-ribose) polymerase begins synthesis of ADP-ribose polymer chains that signal DNA-repairing machinery to form a repair complex. Poly(ADP-ribose) polymerase can induce apoptosis in cells with excessive DNA damage by depleting the cell of NAD+ nucleosidase, which is required for every link in the growing ADP-ribose polymer chain.
Poly(ADP-ribose) polymerase can be involved in double-strand break repair pathways but is not required for homologous recombination.19 It is a popular target for inhibition by new radiosensitizing drugs (Table 1): when a cell’s ability to repair single-strand breaks through base excision repair is reduced, DNA damage accumulates, especially in cancer cells, many of which have degraded repair pathways to begin with. Ongoing research at The University of Texas M. D. Anderson Cancer Center has focused on poly(ADP-ribose) polymerase activation as a major pathway that can be damaged not only by pharmacologic agents but by radiation itself.
EFFECTS OF RADIATION ON BYSTANDER CELLS AND TISSUES
Bystander effects, defined as biological effects in nonirradiated cells as a result of signals from irradiated cells, have been increasingly studied as an important consequence of radiotherapy that may lead to normal tissue toxic effects.20,21 Damage in bystander tissues and cells can take years to manifest.22
Mechanisms of signaling from irradiated to nonirradiated cells (“bystander signaling”) are still mostly unknown; however, a number of mechanisms have been proposed as integral to such signaling, including signaling due to small soluble chemical factors, electrical currents, ion concentrations, and even pressure waves.2,3,20,23 In vitro studies looking at cell-cell communication pathways such as gap junctions have shed some light on the mechanisms of bystander signaling.24
In one study showing the bystander effect, P53, which accumulates in response to DNA damage,8 was analyzed in cultured fibroblasts after irradiation with high-linear-energy-transfer alpha particles.25 P53 was observed in a higher fraction of bystander cells than cells hit directly by alpha radiation. Gap junction inhibitors reduced this effect. When medium was transferred from irradiated cells to nonirradiated cells, P53 levels remained low in the nonirradiated cells, highlighting the significance of cell-cell interactions through gap junctions as the means of propagating stress responses in cells. However, other studies have found that medium transferred from irradiated cells to nonirradiated cells can have a detrimental effect on cell viability.26 Current research at M. D. Anderson Cancer Center is focused on determining whether bystander signaling is mediated by a chemical that can be found in the medium. Our current research has focused on coculture with inserts, which allows for medium exchange through a semipermeable membrane without cell-cell contact, eliminating the need to transfer the medium. This improved system has shown promise in determining the effects of bystander and stress signaling on apoptosis between isolated groups of cells.
Moving to living systems to look for bystander effects of radiation and their contribution to normal tissue toxic effects adds additional complexity. One study showed quantifiable DNA damage responses in shielded tissues in vivo when mice were administered targeted radiation.27 Shortly after exposure, γ-H2AX staining showed significant increases in double-strand break foci in tissues that had been shielded from direct exposure above levels of controls. This study is interesting because it showed similar results in two different strains of mice that conventionally react very differently to radiation. Results such as these are in line with results of other studies on the bystander effect in vivo.28 From a clinical standpoint, such results help to explain the wide patient-to-patient variability in normal tissue toxic effects by showing that radiation administration can elicit a far-reaching, invisible effect in patients regardless of genetic predisposition.29 Mechanisms underlying different organs reacting to the same isolated radiation event are unlikely to be based on cell-cell interactions, which suggests that soluble factors and reactive oxygen species are more likely than direct contact to have played roles in signaling to other cells in the mice-with-shielded-tissues study described above.
More research is necessary to clarify the implications of radiation downstream effects, such as reactive oxygen species in the context of genetic background and increases of double-strand breaks. While most healthy (noncancer) cells tolerate single-strand breaks very well, double-strand breaks in healthy cells have negative implications in terms of genomic stability and cell survival.30 Thus, better understanding of double-strand breaks in bystander cells and tissues is critical.
IMPACT OF LINEAR ENERGY TRANSFER ON RADIATION-INDUCED TISSUE INJURY
As radiation interacts with matter, it loses its energy through interactions with atoms in its direct path. In radiotherapy, linear energy transfer is defined as the average amount of energy lost per defined distance in tissue, as in the energy deposited into a handful of cells. Linear energy transfer occurs at different rates in different tissues, and quantification of it in cellular systems is an important component of determining correct dosage in radiology.
Brahme recently published a thorough review of mechanisms, equations, and models of cell death in relation to linear energy transfer, radiation type, double-strand break probability per dose, and more.31 The models proposed in this study illuminate differences in the impact of radiation type and dose on different cell types under different metabolic conditions. Oxygen is a great radiosensitizer through mechanisms including reactive oxygen species, so hypoxia, such as commonly exists in tumors, can drastically decrease response to radiation, whereas the normoxia prevailing in normal tissues makes them susceptible to normal tissue toxic effects.
Medium-linear-energy-transfer ions, such as proton and neutron beams, are most effective for causing double-strand breaks in radioresistant cancer cells while minimizing damage to normal cells. Common low-linear-energy-transfer radiation sources (photon and electron beams from cesium, cobalt, iridium, and linear accelerators) transfer too little energy for many deep or radioresistant cancers and require many fractions, while high-linear-energy-transfer radiation from carbon ions and other heavy ion radiation can produce a high frequency of double-strand breaks in normal cells, leading to normal tissue toxic effects.31 A proper radiation source should be chosen when available, and dose and fractionation should be determined on the basis of the tumor type, the tumor location, and the linear energy transfer of the radiation source.
The cellular response to DNA damage has been studied exhaustively; however, new discoveries are still being made, and connections between cellular signaling pathways involved in the DNA damage response are slowly being elucidated. Reactive oxygen species are a primary method of radiation-induced cellular damage, if not the primary method.24 They can serve as more than locally destructive agents by inducing damage response mechanisms in, and possibly even directly damaging, tissues far removed from the treatment zone.23 As most tumors are relatively hypoxic, reactive oxygen species–induced damage occurs primarily within normal tissues and is thus of great concern clinically.
Mechanisms of DNA damage repair are complex; some repair pathways, such as poly(ADP-ribose) polymerase signaling, have been highlighted in this article. While DNA damage repair pathways have some redundancy, a failure of any repair system increases the chances of normal tissue toxic effects and increases metabolic stress. DNA repair mechanisms exhibit feedback mechanisms balanced between damage repair and apoptotic pathways; by reducing the oxidative stress in nontarget cells, the likelihood of unintentional cell death and normal tissue toxic effects can be reduced.
Important questions remain regarding the long-term consequences of bystander signaling with various forms of radiation, as few forms of bystander signaling have been characterized. Research focusing on minor differences in repair mechanisms between different types of radiation will be an important step to increase understanding of the DNA repair process. Normal tissue toxic effects begin in the cell; thus, any advances in understanding processes at the cellular level can help spur innovations that improve patient outcomes.
The University of Texas M. D. Anderson Cancer Center is supported in part by the National Institutes of Health through Cancer Center Support Grant CA016672.
1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer J Clin. 2012;62:10–29
2. Mothersill C, Seymour C. Are epigenetic mechanisms involved in radiation-induced bystander effects? Front Genet. 2012;3:74
3. Morgan WF, Sowa MB. Non-targeted bystander effects induced by ionizing radiation. Mutat Res. 2007;616:159–164
4. Hoejmakers JHJ.. DNA damage, aging, and cancer. N Engl J Med. 2009;361:1475–1485
5. Zhao W, Diz DI, Robbins ME. Oxidative damage pathways in relation to normal tissue injury. Br J Radiol. 2007;80(Spec No 1):S23–S31
6. Lipinski B. Hydroxyl radical and its scavengers in health and disease. Oxid Med Cell Longev. 2011;2011:809696
7. Dotto GP. p21(WAF1/Cip1): More than a break to the cell cycle? Biochim Biophys Acta. 2000;1471:M43–M56
8. Brown KR, Rzucidlo E. Acute and chronic radiation injury. J Vasc Surg. 2011;53(1 Suppl):15S–21S
9. Crouse GF. Mismatch repair outside of replication. Cell Cycle. 2012;11:2773–2774
10. Strausfeld U, Labbé JC, Fesquet D, et al. Dephosphorylation and activation of a p34cdc2/cyclin B complex in vitro by human CDC25 protein. Nature. 1991;351:242–245
11. Huang X, Halicka HD, Darzynkiewicz Z. Detection of histone H2AX phosphorylation on Ser-139 as an indicator of DNA damage (DNA double-strand breaks). Curr Protoc Cytom. 2004;Chapter 7:Unit 7.27
12. Rassool FV, Tomkinson AE. Targeting abnormal DNA double strand break repair in cancer. Cell Mol Life Sci. 2010;67:3699–3710
13. Kuhfittig-Kulle S, Feldmann E, Odersky A, et al. The mutagenic potential of non-homologous end joining in the absence of the NHEJ core factors Ku70/80, DNA-PKcs and XRCC4-LigIV. Mutagenesis. 2007;22:217–233
14. Huen MS, Chen J. The DNA damage response pathways: At the crossroad of protein modifications. Cell Res. 2008;18:8–16
15. Kurose A, Tanaka T, Huang X, Traganos F, Dai W, Darzynkiewicz Z. Effects of hydroxyurea and aphidicolin on phosphorylation of ataxia telangiectasia mutated on Ser 1981 and histone H2AX on Ser 139 in relation to cell cycle phase and induction of apoptosis. Cytometry A. 2006;69:212–221
16. Coates PJ, Appleyard MV, Murray K, et al. Differential contextual responses of normal human breast epithelium to ionizing radiation in a mouse xenograft model. Cancer Res. 2010;70:9808–9815
17. Liu Y, Prasad R, Beard WA, et al. Coordination of steps in single-nucleotide base excision repair mediated by apurinic/apyrimidinic endonuclease 1 and DNA polymerase β. J Biol Chem. 2007;282:13432–13541
18. Ström CE, Johansson F, Uhlén M, Szigyarto CA, Erixon K, Helleday T. Poly (ADP-ribose) polymerase (PARP) is not involved in base excision repair but PARP inhibition traps a single-strand intermediate. Nucleic Acids Res. 2011;39:3166–3175
19. Burdak-Rothkamm S, Prise KM. New molecular targets in radiotherapy: DNA damage signalling and repair in targeted and non-targeted cells. Eur J Pharmacol. 2009;625:151–155
20. Little JB, Azzam EI, de Toledo SM, Nagasawa H. Bystander effects: Intercellular transmission of radiation damage signals. Radiat Prot Dosimetry. 2002;99:159–162
21. Olsson MG, Nilsson EJ, Rutardóttir S, Paczesny J, Pallon J, Akerström B. Bystander cell death and stress response is inhibited by the radical scavenger α(1)-microglobulin in irradiated cell cultures. Radiat Res. 2010;174:590–600
22. Kadhim MA, Macdonald DA, Goodhead DT, Lorimore SA, Marsden SJ, Wright EG. Transmission of chromosomal instability after plutonium alpha-particle irradiation. Nature. 1992;355:738–740
23. Mothersill C, Seymour C. Radiation-induced bystander and other non-targeted effects: Novel intervention points in cancer therapy? Curr Cancer Drug Targets. 2006;6:447–454
24. Burdak-Rothkamm S, Short SC, Folkard M, Rothkamm K, Prise KM. ATR-dependent radiation-induced gamma H2AX foci in bystander primary human astrocytes and glioma cells. Oncogene. 2007;26:993–1002
25. Iyer R, Lehnert BE, Svensson R. Factors underlying the cell growth-related bystander responses to alpha particles. Cancer Res. 2000;60:1290–1298
26. Belloni P, Latini P, Palitti F. Radiation-induced bystander effect in healthy G(o) human lymphocytes: Biological and clinical significance. Mutat Res. 2011;713:32–38
27. Koturbash I, Loree J, Kutanzi K, Koganow C, Pogribny I, Kovalchuk O. In vivo bystander effect: Cranial X-irradiation leads to elevated DNA damage, altered cellular proliferation and apoptosis, and increased p53 levels in shielded spleen. Int J Radiat Oncol Biol Phys. 2008;70:554–562
28. Rudqvist N, Parris TZ, Schüler E, Helou K, Forssell-Aronsson E. Transcriptional response of BALB/c mouse thyroids following in vivo astatine-211 exposure reveals distinct gene expression profiles. EJNMMI Res. 2012;2:32
29. Bentzen SM, Overgaard J. Patient-to-patient variability in the expression of radiation-induced normal tissue injury. Semin Radiat Oncol. 1994;4:68–80
30. Stewart RD. Two-lesion kinetic model of double-strand break rejoining and cell killing. Radiat Res. 2001;156:365–378
31. Brahme A. Accurate description of the cell survival and biological effect at low and high doses and LET’s. J Radiat Res. 2011;52:389–407