The ongoing improvements in cancer therapy and health care have led to a population of long-term cancer survivors that continues to grow: 62% of adult and 77% of pediatric cancer patients survive beyond 5 years. Consequently, for most patients, cancer can be considered a chronic disease. The total radiation dose that can be administered safely to cancer patients is limited by the risk of complications arising in those normal tissues unavoidably included within the treatment volume. Of particular concern are the late effects that can arise several months to years postirradiation. While improvements in radiation oncology such as intensity-modulated radiation therapy (IMRT) have led to a reduction in the volume of normal tissue irradiated, late effects remain a significant risk. The National Cancer Institute has identified long-term survival from cancer as one of the new areas of public health emphasis, particularly studying adverse long-term or late effects of cancer and its treatment (National Cancer Institute's Plans and Priorities for Cancer Research). Given the increasing population of long-term survivors, the need to mitigate or treat late effects has emerged as a primary area of radiation biology research (1,2).

This is of particular relevance to patients receiving brain irradiation. The total dose of radiation therapy that can be administered safely to the brain of patients presenting with primary or metastatic brain tumors is limited by the risk of normal brain morbidity. The need to both understand and minimize the side effects of brain irradiation is exacerbated by the ever-increasing number of patients with secondary brain metastases that require treatment with large-field partial or whole brain irradiation (WBI). Around 20-40% of the 1,399,970 new cancer patients diagnosed in 2006 (3) will develop brain metastases (4) making this the second most common site of metastatic cancer, the most common neurological manifestation of cancer, and a cancer problem more common in incidence than newly diagnosed lung, breast or prostate cancer combined. Approximately 250,000 of these individuals will be treated ultimately with large-field partial or WBI for brain metastases. Over 125,000 of these patients will survive long enough to develop radiation-induced brain injury, including cognitive impairment. At present, there are no successful long-term treatments or effective preventive strategies for radiation-induced brain injury (5).

Oxidative Stress and Neurodegenerative Disorders Edited by G. Ali Qureshi and S. Hassan Parvez

© 2007 Elsevier B.V. All rights reserved.

The classic model of specific target cell clonogenic death hypothesized that radiation-induced late effects were solely a consequence of clonogenic cell loss (6,7). These late effects were inevitable, progressive and untreatable, leading to chronic and progressive reductions in organ function. In the last decade, our views on the pathogenesis of radiation-induced late normal tissue injury have undergone a paradigm shift. Rather than simply reflecting a loss of normal cellular components, radiation-induced late effects are viewed now as a combination of not only cell loss and loss of normal cellular function, but also an orchestrated, albeit limited, response to injury that involves interactions between multiple cell types within a particular target organ (8-11). Some of the important lesions include fibrosis, necrosis, atrophy and vascular damage.

In general, irradiating late-responding normal tissues leads to an acute activation of stress-sensitive kinases, transcription factors (12) and increased production of inflammatory cytokines. This is followed by an aberrant chronic inflammatory/wound healing response in which vascular and parenchymal cell dysfunction and cell loss, associated with chronic overproduction of particular cytokines and growth factors, results in fibrosis and/or necrosis, depending on the particular organ involved (11).

This new paradigm offers an exciting and new approach to radiation-induced late effects, namely the possibility that radiation-induced injury can be modulated by therapies directed at mitigating the cascade of events resulting from normal tissue injury. Indeed, recent findings support the hypothesis that radiation-induced injury can be modulated by therapies directed at mitigating the cascade of events resulting from normal tissue injury (13-15). Since these events are unlikely to occur in tumors, where direct clonogenic cell kill predominates, such treatments should not negatively impact radiation-induced tumor cell kill. However, the mechanisms responsible for the clinical expression and progression of late, radiation-induced normal tissue injury, remain poorly understood.

A growing body of evidence suggests that the development and progression of radiation-induced late effects are driven, in part, by an acute and chronic oxidative stress and/or inflammation (15). This chapter will (i) review radiation-induced late effects in the brain, and briefly the kidney, lung and skin; (ii) introduce reactive oxygen/nitrogen oxide species; (iii) discuss radiation-induced oxidative stress and its role in radiation-induced late effects and (iv) provide a rationale for anti-inflammatory-based interventional approaches directed at the treatment of late normal tissue injury, particularly in the brain.


Based on the time of clinical expression, radiation-induced brain injury is described in terms of acute, early delayed and late delayed reactions (16). Acute injury (acute radiation encephalopathy), expressed in days to weeks after irradiation, is fairly rare under current radiotherapy regimens. Early delayed injury occurs from 1 to 6 months postirradiation and can involve transient demyelination with somnolence. While both these early injuries can result in severe reactions, they are normally reversible and resolve spontaneously. In marked contrast, late delayed effects, characterized by demyelination, vascular abnormalities and ultimate white matter necrosis (9), are observed > 6 months postirradiation and are viewed as irreversible and progressive. In addition to these histopathologic endpoints, there is a growing awareness of intellectual deterioration in patients receiving brain irradiation (17). Cognitive dysfunction, including dementia, induced by large-field partial or WBI, is reported to occur in 20-50% of brain tumor patients who are long-term survivors (> 12 months postirradiation) (17-19).

Vascular abnormalities and demyelination are the predominant histological changes seen in radiation-induced brain injury. Classically, late delayed injury was viewed as due solely to a reduction in the number of surviving clonogens of either parenchymal, i.e. oligodendrocyte (20) or vascular, i.e. endothelial (21) target cell populations leading to white matter necrosis.

Vascular hypothesis

Proponents of the vascular hypothesis argue that vascular damage leads to ischemia with secondary white matter necrosis. In support of this hypothesis is the large amount of data describing radiation-induced vascular changes including vessel wall thickening, vessel dilation and endothelial cell nuclear enlargement (9,22). Quantitative studies in the irradiated rat brain have noted time- and dose-related reductions in the number of endothelial cell nuclei and blood vessels prior to the development of necrosis (22). Further, recent boron neutron capture studies in which radiation was delivered essentially to the vas-culature alone, still led to the development of white matter necrosis (23). A potential limitation of these studies is that they have used relatively large single doses or fractionated large doses of ionizing radiation given over a short period (23,24). The results from these studies may not be an accurate indication of events that occur after the more protracted low-dose-per-fraction regimens used clinically (25). More recent studies (26) have utilized a more clinically relevant fractionated regimen of WBI in which adult male rats received a total dose of 40 Gy delivered as eight fractions of 5 Gy administered twice per week for 4 weeks. This dose (Biologically Equivalent Dose (BED) = 106.7 Gy) is expected to be biologically similar to that used clinically (60 Gy in 30 fractions over 6 weeks) in the treatment of primary gliomas (BED = 100.2 Gy) (27). Brain capillary and arteriole pathology were studied using a novel alkaline phosphatase enzyme histo-chemistry methodology (28); vessel density and length were quantified using a stereology method with computerized image processing and analysis. Vessel density and length were unchanged 24 h after the last dose, but at 10 weeks postirradiation, both were substantially decreased. After 20 weeks, the rate of decline in the vessel density and length in irradiated rats was similar to that in unirradiated age-matched controls. No gross gliosis or demyelination was observed 12 months postirradiation using conventional histopathol-ogy techniques. These findings suggest that the early (10 weeks) and persistent vascular damage that occurs after a prolonged WBI fractionation scheme may play an important role in the development of late delayed radiation-induced brain injury.

In contrast, radiation-induced necrosis has been reported in the absence of vascular changes (9). Moreover, while the vascular hypothesis argues that ischemia is responsible for white matter necrosis, the most sensitive component of the central nervous system (CNS) to oxygen deprivation, the neuron, is located in the gray matter, a relatively radioresistant region. Thus, it seems unlikely that radiation injury is due solely to damage to the vasculature alone.

Parenchymal hypothesis

The parenchymal hypothesis for radiation-induced CNS injury focuses on the oligodendrocyte, required for the formation of the myelin sheath. The key cell for the generation of mature oligodendrocytes is the oligodendrocyte type 2 astrocyte (O-2A) progenitor cell (29). Irradiation results in the loss of reproductive capacity of the O-2A progenitor cells in the rat CNS (30,31). It is hypothesized that radiation induces loss of O-2A progenitor cells, leading to a failure to replace oligodendrocytes and demyelination. However, a mechanistic link between loss of oligodendrocytes and demyelination has yet to be established. Further, while the kinetics of oligodendrocytes is consistent with the early transient demyelination seen in the early delayed reactions, it is inconsistent with the late onset of white matter necrosis (32). Thus, it is unlikely that loss of O-2A progenitor cells and oligodendrocytes alone can lead to late radiation injury in the brain.

As noted above, recent findings suggest that the classic model of parenchymal or vascular target cells is over-simplistic. Pathophysiological data from a variety of late-responding tissues, including the CNS, indicate that the expression of radiation-induced normal tissue injury involves complex and dynamic interactions between several cell types within a particular organ (9,13,33). In the brain, these include not only the oligodendrocytes and endothelial cells, but also the astrocytes, microglia and neurons.


Once viewed as playing a mere supportive role in the CNS, astrocytes are now recognized as a heterogeneous class of cells with many important and diverse functions in the normal CNS (34). Astrocytes secrete a variety of cytokines, proteases and growth factors that regulate the response of the vasculature, neurons and oligodendrocyte lineage in the normal CNS (35,36). Recent data suggest that hippocampal astrocytes are capable of regulating neurogenesis by instructing the stem cells to adopt a neuronal fate (36). In addition, astrocytes assume a critical role in the reaction of the CNS to various forms of injury, including radiation, and are vital for the protection of endothelial cells, oligodendrocytes and neurons from oxidative stress (37,38). In response to injury, astrocytes exhibit two common reactions, a relatively acute cellular swelling and a more chronic hypertrophy-hyperplasia. Of note, time- and dose-dependent increases in astrocyte number have been observed in the irradiated rat and mouse brains (21,22,39). In addition to increased cell number, an increase in GFAP staining intensity indicative of reactive astrocytes has been observed (39).


Microglia are the immunocompetent cells of the CNS, and are present in the brain in substantial numbers. They are derived from hematopoietic precursors early in embryonic development (40) and share characteristics of peripheral immune cells (41). In the normal brain, microglia have a ramified morphology (42). Although often referred to as quiescent, resting microglia actively survey the CNS environment (43). Microglia respond to virtually any, even minor pathological event in the brain, by becoming activated. During this process, they change their morphology from the resting ramified state to a reactive amoeboid appearance (42) and produce and release a variety of factors including cytokines, prostanoids and proteases as well as 'NO and O^- (44). The association between activated microglia and subsequent induction of pro-inflammatory mediators makes the presence of activated microglia a good marker of inflammation. Irradiation of the brain has been shown to result in increased numbers of microglia 2-18 months postirradiation (45-47), and can occur during the latent period before the clinical expression of injury (24,39). An integral part of the inflammatory response is the production of various cytokines that enable the microglia to communicate in an autocrine and paracrine manner.


In view of the classic model of radiation-induced normal tissue injury, where DNA damage and loss of slowly turning over stem cell populations led to late effects, the non-proliferating neuron was thought to be radioresistant and a non-participant in radiation-induced brain injury. Recent data documenting chronic and progressive cognitive dysfunction in both children (48-50) and adults (17,51,52) following large-field partial or WBI have suggested that neurons are indeed sensitive to radiation. Moreover, in vivo and in vitro experimental studies have shown radiation-induced changes in hippocampal cellular activity, synaptic efficiency and spike generation (53,54), and in neuronal gene expression (55). Thus, it seems likely that radiation-induced alterations in neuron function play a role in the development and progression of radiation-induced brain injury. An additional and important component of radiation injury is the relatively recent observation that irradiation can inhibit hippocampal neurogenesis.

Neural stem cells/neurogenesis

The hippocampus is central to short-term declarative memory and spatial information processing. It consists of the dentate gyrus, CA3 and CA1 regions. The dentate gyrus represents a highly dynamic structure and a major site of postnatal/adult neurogenesis. Resident in the hippocampus are neural stem cells, self-renewing cells capable of generating neurons, astrocytes and oligodendrocytes (56,57). Neurogenesis depends on the presence of a specific neurogenic microenvironment; both endothelial cells and astrocytes can promote/regulate neurogenesis (36,58). Experimental studies have indicated that brain irradiation results in increased apoptosis (59), decreased cell proliferation and a decreased stem/precursor cell differentiation into neurons within the neurogenic region of the hippocampus (45,60,61). Rats irradiated with a single dose of 10 Gy produce only 3% of the new hippocampal neurons formed in control animals (45). Of note, these changes were observed after doses of radiation that failed to produce demyelination and/or white matter necrosis of the rat brain. The radiation-induced chronic inhibition of neurogene-sis has been associated with delayed cognitive impairment using hippocampal-dependent behavioral tasks in adult rodents (62). Radiation does not simply ablate the progenitor cell population. Progenitor cells can be successfully cultured from irradiated hippocampi and these cells retain their neurogenic capacity in vitro (45). Of interest, evidence demonstrating the importance of the microenvironment for successful neurogenesis comes from studies showing that non-irradiated stem cells transplanted into the irradiated hippocampus failed to generate neurons; this may reflect a pronounced microglial inflammatory response, since neuroinflammation is a strong inhibitor of neurogenesis (63). In contrast to the reduction in neurogenesis, gliogenesis appears to be enhanced after irradiation; microglia and immature oligodendrocytes increase in total and relative number in both in vitro and in vivo conditions (45). These results suggest that brain irradiation does not eradicate hippocampal progenitor cells or even alter their intrinsic capability to produce new neurons, but radiation induces currently undefined signals that regulate the proliferation, differentiation and survival of these cells. These likely include pro-inflammatory cytokines.

Pro-inflammatory cytokines

The prototypical pro-inflammatory cytokines include interleukin-one beta (IL-ip) and tumor necrosis factor alpha (TNF-a). Brain irradiation has been shown to increase gene expression of both IL-ip and TNF-a within 24 h; TNF-a gene expression remains elevated up to 6 months postirradiation (64,65). Moreover, radiation-induced increases in IL-ip, TNF-a and IL-6 protein have been observed weeks to months after experimental irradiation (64,66,67). IL-6 and TNF-a suppress hippocampal progenitor proliferation and induce progenitor apoptosis in vitro; this might help explain radiation- (63) and LPS-induced inflammation (68) suppressing both basal and inflammation-induced increases in neurogenesis in vivo. TNF-a and IL-1 can also regulate astrocyte proliferation in response to various forms of CNS injury (69,70), implicating them in gliosis. TNF-a is also cytotoxic to oligodendrocytes in vitro (71) and is expressed in multiple sclerosis and experimental autoimmune encephalitis (EAE) (72,73), suggesting a role for this cytokine in demyelina-tion. Given the radiation-induced changes in TNF-a expression following brain irradiation, Daigle et al. (66) examined the role of TNF-a signaling in the response to brain irradiation using TNFRp55- or TNFRp75-deficient mice compared with control mice. Contrary to the hypothesized protection the lack of TNF-a signaling pathways might provide in the irradiated brain, mice lacking TNFRp75 exhibited increased acute radiation-induced apoptosis in putative stem regions of the mouse brain. At 1 month after single doses of 20-45 Gy, the TNFRp75 mice showed reduced proliferative responses in the same regions, and by 3 months, they were exhibiting dose-dependent seizures and additional severe neurological abnormalities that were not seen in the TNFRp55-deficient or control mice. The seizure activity was correlated with the onset of extensive demyelination, and by 6 months levels of myelin basic protein in the irradiated TNFRp75-deficient mice were approximately 40% of those seen in the other two strains. At this stage the animals were moribund and were euthanized. These exciting observations point out the need to view pro-inflammatory cytokines as not merely "pro-injury" mediators, but rather as part of a normal orchestrated response to injury. Indeed, not only has TNF-a been shown to be protective in other demyelinating diseases, such as EAE (74), but signaling through the TNFRp75 has been shown to be neuroprotective in EAE, hypoxia and P-amyloid toxicity (75,76).

Kidney, lung and skin

As described for the brain, irradiation of other late-responding tissues, including the kidney, lung and skin can lead to the development of late effects, primarily evidenced as fibrosis. Thus, kidney irradiation can lead to the development of radiation nephropathy, associated with a chronic and progressive reduction in renal function associated with glomerulosclerosis and tubulointerstitial fibrosis (77). In the case of the lung, clinically significant radiation-induced fibrosis is usually described as chronic progressive dyspnea associated with scarring that is seen months to years after irradiation (78). Radiation overexposure frequently induces late damage in the skin characterized by dermal atrophy, telangiectasia, late ulceration and ultimately fibrosis (79,80). In each of these normal tissues, the classic model of loss of vascular or parenchymal target cell clono-gens has been replaced by a more dynamic multicellular model in which the radiation response reflects ongoing interactions between several cell types within each organ. While the specific pathogenic mechanism(s) involved in the development and progression of radiation-induced late effects remains unclear, there is increasing appreciation of the putative role of acute and chronic oxidative stress in this process (15). Before reviewing these findings, we will discuss reactive oxygen/nitrogen oxide species and their regulation.


All aerobic organisms produce ROS, partially reduced metabolites of molecular oxygen (dioxygen; O2) that have higher activities relative to molecular O2 (81,82). These include superoxide anion (O2-) and hydrogen peroxide (H2O2), formed by one- and two-electron reductions of O2, respectively, and hydroxyl radical (*OH). Superoxide is a free radical, defined as an atom or group of atoms possessing one or more unpaired electrons (83). In spite of being a free radical O2- is not highly reactive; i.e. O2- cannot penetrate lipid membranes and is therefore restricted to the intracellular compartment where it is generated. O2- is primarily generated in the mitochondria because of leakage of electrons from the electron transport chain. In addition, O2- is produced endogenously by flavoenzymes such as xanthine oxidase (84), lipoxygenase (85), cyclooxygenase (86) and plasma membrane-associated oxidases, e.g. the phagocytic NAPDH oxidase (87). O2- is rapidly dismutated to H2O2 by the antioxidant enzyme superoxide dismutase (SOD). There are three known isoforms of SOD in eukaryotes; manganese SOD (MnSOD), a homotetrameric protein with a molecular weight of 88 kD located within the mitochondrial matrix (88), copper-zinc SOD (CuZnSOD), a 32-kD homodimer located in the cytoplasm, nucleus and lysosomes, and extracellular SOD (EC-SOD), a 135-kD homotetramer released from cells into the extracellular space (89).

H2O2 is not a free radical and is a weaker oxidizing agent than O2-. Its importance lies in its ability to cross biological membranes. In addition to its generation through dismutation of O2-, H2O2 can be formed by direct two-electron reduction of O2 catalyzed by a variety of flavoprotein oxidases (90). H2O2 serves as an intermediate in the generation of more reactive ROS, such as hypochlorous acid via the action of myeloperoxidase, present in the phagosomes of neutrophils (91). In the presence of transition metals, H2O2 can give rise to the most reactive and toxic ROS, *OH by the Fenton reaction (92). At high concentrations, H2O2 is converted to water and O2 by catalase, which in mammalian cells is localized predominantly in the peroxisomes. Catalase is composed of four identical subunits, each having a molecular weight of approximately 55-60 kD and containing a protoporphyrin group (Fe III) at its active site (93). Each subunit has one molecule of NADPH bound to it, which is thought to help stabilize the enzyme and aid in peroxidation of molecules other than H2O2 (94).

At low concentrations, H2O2 is converted to water by the selenium-containing glutathione peroxidase (GPx). In mammals, the GPx family is comprised of five family members. The cytosolic and mitochondrial GPx (GPxl) is ubiquitously expressed in the cytosol of most tissues, particularly erythrocytes, kidney and liver (95). Gastrointestinal GPx (GPx2) and plasma GPx (GPx3) are mainly expressed in the gastrointestinal tract and kidney, respectively (96,97), while phospholipid hydroperoxide GPx (GPx4), a cytosol and a membrane-bound protein, is highly expressed in renal epithelial cells and testes (98). GPx5 is a recently identified selenium-independent GPx localized to the mouse epi-didymis (99). GPx reduces hydroperoxides including H2O2 to the respective alcohols and water with glutathione as the electron donor (100).

More recently, an additional family of antioxidant enzymes, the 2-Cys peroxiredoxins (2-Cys Prxs), have been recognized as important regulators of peroxide-mediated signaling cascades (101). In mammals, six distinct Prx family members have been identified and located primarily in the cytosol. Prx-3 is unique, being localized specifically to the mitochondria, while Prx-5 is found in both the mitochondria and peroxisomes. All members contain conserved reactive cysteine residues in the active site(s) that are essential for the enzymatic oxidation-reduction reaction. 2-Cys-Prxs are hypothesized to serve as "floodgates," keeping constitutive levels of H2O2 low, while permitting higher levels during signal transduction (102).

There is a growing appreciation of the important roles that the diatomic free radical nitric oxide ('NO) and reactive nitrogen oxide species (RNOS), formed from the reaction of 'NO with molecular oxygen or O2-, play in physiological and pathophysiological mechanisms (103,104). 'NO is synthesized enzymatically from L-arginine by NO synthase (NOS) via electron transfer from NADPH. Three distinct isoforms of NOS have been identified (for reviews see (105,106)). nNOS (also called Type 1, NOS-I and NOS-1) is predominantly localized in neuronal tissue. iNOS (also known as Type II, NOS-II and NOS-2) is the inducible or calcium-independent isoform found in a wide range of cells and tissues. eNOS (also known as Type III, NOS-III and NOS-3) which was first identified in vascular endothelial cells is, like nNOS, a calcium-dependent, constitutively expressed isoform.

At physiological concentrations 'NO functions as an intracellular messenger; 'NO can cross cell membranes and transmit signals to other cells (107). 'NO can also act as an excellent antioxidant; e.g. iron-catalyzed oxidation reactions are inhibited by 'NO (108). In pathophysiological situations where iNOS is upregulated, the most common RNOS generated in vivo are dinitrogen trioxide (N2O3) and peroxynitrite (O=NOO-), both of which can induce nitrosative and oxidative stresses (109). In normal cells, ROS/RNOS are believed to play an important role in intracellular signaling (110,111), gene expression (112,113) and physiological function (114). Under normal conditions, ROS/RNOS generation is approximately in balance with the cell's antioxidant defenses (antioxidants/antioxidant enzymes). Any imbalance between ROS/RNOS generation and destruction in favor of ROS/RNOS generation can create oxidative stress.


In vitro

Irradiating biological material leads to a rapid burst of ROS generated primarily as a result of the ionization of water molecules and direct ionization of target molecules (115). The radicals and ROS generated include e"q, H*, *OH, O2" and H2O2. Due to their instability/reactivity e" and *OH will react with target molecules within 10"9 s of their generation. In contrast, O2" and H2O2 are relatively stable in water and can persist for 10 and >102 s, respectively, in water (115). However, the yields of O2" and H2O2 generated as a consequence of a primary ionization event are considerably lower than those produced by normal cellular metabolism (116). Recent data indicate that, in addition to the rapid burst of radicals and ROS observed immediately following irradiation, cells can exhibit more persistent and prolonged increases in ROS/RNOS over time periods ranging from several minutes to several days postirradiation (117-121). Mitochondria appear to be the primary site of increased ROS/RNOS generation (119-122). A growing body of literature supports a role for RNOS, particularly *NO, in early radiation-induced signaling mechanisms (121-124).

In vivo

A radiation-induced increase in ROS generation and/or an oxidative stress has also been observed in vivo. Due to the transient nature of the ROS species generated, direct measurements are extremely difficult. Thus, the evidence has been primarily derived from studies showing increases in the formation of oxidized products. Total body irradiation (TBI) has been shown to lead to increased markers of lipid peroxidation, including thiobarbituric acid reaction products (TBARs), 4-hydroxynonenal (4-HNE) and hexane in animal models and in patients (125-129). This radiation-induced oxidative stress appears to result not from radicals and ROS generated at the time of irradiation, but from the propagation of radicals and ROS occurring from 2 to 10 days postirradiation. The mechanisms responsible for this chronic oxidative stress remain ill-defined; putative mechanisms include a reduction in the antioxidant vitamins C and E (125,128), an increase in radical generation resulting from changes in the xanthine oxidoreductase system (130) and altered arachidonic acid metabolism (131,132). More recently, a role for metabolic oxidative stress, initiated by radiation-induced damage to critical biomolecules regulating the metabolic production of prooxidant species, has been proposed (133).

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook

Post a comment