Dr Yuri Miura PhD, Research Team for Functional Genomics, Tokyo Metropolitan Institute of Gerontology, 35-2 Sakaecho, Itabashi-ku, Tokyo 173-0015, Japan. Email: email@example.com
Oxidative stress is recognized as an important environmental factor in aging; however, because reactive oxygen species (ROS) and related free radicals are normally produced both intra- and extracellularly, air-living organisms cannot avoid the risk of oxidative stress. Consequently, these organisms have evolved various anti-oxidant systems to prevent ROS, scavenge free radicals, repair damaged components and adaptive responses. This review will focus on the repair and adaptive response to oxidative stress, and summarize the changes of these systems as a result aging and their relationship to premature aging. Geriatr Gerontol Int 2010; 10 (Suppl. 1): S1–S9.
Oxidative stress is a serious cause of cell damage associated with the initiation and progression of many diseases; therefore, air-living organisms have developed various anti-oxidant protective systems, such as enzymes, small molecules, metal chelation and various repair systems. In order to use these protective and repair systems effectively, organisms recognize the redox imbalance caused by reactive oxygen species (ROS) sensitively, followed by the initiation of biological response signaling. Oxidative stress elicits death signals or survival signals in cells (Fig. 1). Death signals proceed to apoptosis and necrosis, whereas survival signals induce three kinds of biological responses as follows: (i) repair response, which excises and repairs damaged components in cells; (ii) adaptive response, which enhances defense and repair systems; and (iii) hormesis, which yields beneficial stimulant effects.
As a result of non-specific and extensive damage caused by ROS, the subjects of repair response are ubiquitous, such as proteins, lipids and DNA. Oxidized proteins are reduced by enzymes; for example, methionine sulphoxide reductase1 and sestrin,2 or decomposed by protein-degradation systems, such as proteasome and various proteases. Fatty acid peroxides in phospholipids contained in membrane bilayers are excised by phospholipase A2, and reduced to alcohol by glutathione peroxidase (GPx); however, when peroxides react with metal complexes, several aldehydes, such as malondialdehyde and 4-hydroxy-nonenal, and hydrocarbon gases, such as ethane and pentane, are generated as final products of lipid peroxidation. Because the aldehydes can react with –SH or –NH2 group on proteins, the generation of aldehydes gives rise to further protein damage. DNA lesions, including strand breaks and base modifications, are repaired through nonhomologous end-joining (NHEJ), homologous recombination (HR), base excision repair (BER), or nucleotide excision repair (NER), as described later. It has been reported that the DNA repair response plays a key role in genome stability, the determination of apoptosis or survival, and premature aging; therefore, in the present review, we would like to focus attention on molecules involved in the initial response of DNA repair responses.
In general, low-dose stresses induce hormesis and adaptive response. Although hormesis and adaptive response are often confused, hormesis is defined as a beneficial stimulant effect induced by “chronic” low-dose stress, whereas adaptive response is defined as a defensive response to a “single” low-dose stress. From an experimental point of view, adaptive response is distinct from hormesis, because it requires an optimum experimental condition, which consists of low-dose stress and subsequent high-dose stress with an appropriate interval. Namely, adaptive response is a biological defensive response induced by a single low-dose stress, which is evaluated by the degree of damage as a result of subsequent high-dose stress. In contrast, hormesis is generally estimated epidemiologically or statistically by several indices, such as growth, longevity and immune response. Although it is often utilized in risk assessment and public health policies for environmental toxins, hormesis is extremely difficult to assess.3 In frequent cases, scientific experimental verification is impossible, for example, radiation hormesis caused by high background radiation. Thus, the details of hormesis are not described here further and we will focus on the repair response and adaptive response among survival responses to oxidative stress (Fig. 1).
Aging might be defined as progressive functional decline and increasing mortality over time. Although the mechanisms underlying aging remain obscure, it has recently been accepted that aging is not caused by a single factor or process, but is rather a multi-factorial process modulated by interplay among genetic and environmental factors. Because oxidative stress is an important environmental factor, here we review the correlation between aging and biological responses to oxidative stress, focusing attention on the following three points: (i) the phenotypes resembling premature aging due to gene mutations; (ii) the decline of survival responses with aging; and (iii) the adaptive process to the oxygen environment during aging.
Repair response to oxidative stress and aging
The accurate maintenance of nuclear DNA is critical to cellular function, and therefore numerous DNA repair systems have been developed corresponding to many different types of DNA lesions. Double-strand breaks (DSB) are repaired through NHEJ and HR; whereas single-strand lesions (SSL), which consist of oxidation products, such as 8-oxoguanine and thymine glycol, and some alkylation products, are repaired through BER and NER.4 The initial step in DNA repair is the detection of DNA lesions by sensor molecules, second is the activation of signaling proteins as transducer molecules and third is the recruitment of repair proteins as effector molecules. ROS oxidize DNA to 8-oxoguanine-DNA and produce DSB as a result of ionizing radiation. Here we will describe ataxia telangiectasia mutated (ATM), an important molecule in DSB sensing, and 8-oxoguanin-DNA glycosylase (Ogg 1), 8-oxoguanine excision enzyme, as a sensor molecule of BER. Furthermore, in mice and humans, mutations in certain DNA repair genes lead to phenotypes of premature aging;5–11 thus, we will also describe premature aging caused by the mutation of Atm genes.
Ataxia telangiectasia mutated (ATM)
ATM is a guard at the gate of genome stability. This multifunctional protein kinase organizes intricate cellular responses to DSB. The absence or dysfunction of ATM leads to a pleiotropic genetic disorder, ataxia-telangiectasia (AT), whose pathologies are neuronal degeneration, immunodeficiency, genomic instability, premature aging, and cancer susceptibility.12 ATM is activated by the formation of DSB, and phosphorylates various substrates involved in repair responses, apoptosis and cell cycle arrest; therefore, ATM is a key molecule in genome stability, including DNA repair and checkpoint regulation. Furthermore, it has recently been reported that oxidative stress facilitates ATM deficiency, suggesting that ATM is also involved in anti-oxidant systems. Here, we will describe two functions of ATM, genome stability and oxidative stress regulation, and their connection.
DSB activate ATM mediated by intramolecular autophosphorylation on Ser-1981 and dimer dissociation.13 Although the precise mechanisms of ATM activation remain uncertain, it is reported that the Mre11/Rad50/Nbs1 complex is involved in the autophosphorylation of ATM.14 Activated ATM is a kinase characterized by the signature motifs of phosphatidylinositol 3-kinase in a carboxy-terminal region and phosphorylates numerous proteins;15 for example, histone H2AX, one of the nucleosome core histone H2A, p53 protein, which plays an important role in the cellular response to numerous genotoxic insults, and so on (Fig. 2). Phosphorylated H2AX on Ser-139 by ATM kinase has been named γH2AX. γH2AX forms distinct nuclear γH2AX foci at DSB sites, followed by several signaling and repair proteins being recruited to γH2AX foci; therefore, γH2AX is considered to be a specific indicator of DSB.16 Phosphorylation of p53 on Ser-15 by ATM is a major mechanism leading to the activation of p53. Activated p53 stimulates the transcription of many target genes and coordinates checkpoint, senescence and apoptosis pathways in response to DSB (Fig. 2).17,18
Another function of ATM is reported to regulate intracellular ROS levels.19,20 In Atm-deficient cells or tissues, the following four pieces of evidence were observed: (i) abnormal response to agents that induce oxidative stress;21,22 (ii) abnormalities in the levels and/or function of anti-oxidant systems;23 (iii) evidence of increased oxidative stress;24 and (iv) observation of constitutive stress responses.25 Persistently elevated ROS levels might cause chronic damage to DNA and other cellular macromolecules, leading to various pathologies of a human disease, AT; however, there is a discrepant observation in Atm-deficient mice, which display increased levels of oxidative stress and damage, and are utilized as a cancer-prone model of AT. Erker et al. examined the effects of oxidative stress on the survival and tumorigenesis of Atm-deficient mice using oxidative stress-enhanced and stress-depressed mice, which had Sod genetically removed or reduced, or were treated with α-tocopherol. As a result, these treatments of Atm-deficient mice had no effects on survival or tumorigenesis, suggesting that oxidative stress and damage should be a secondary consequence of tumorigenesis in Atm-deficient mice and not the initial cause of cancer.26 Further investigations are necessary to clarify the connection between DNA genome instability and oxidative stress in Atm deficiency.
Mutants of various molecules involved in genome stability are reported to represent phenotypes of premature aging (Table 1). In mice, dysfunction of Atm, DNA-PKcs, Ku86, p53, Wrn and XpdTTD/XPA, which participate in the DNA repair response, cause premature aging phenotypes in various tissues and organs, suggesting that molecules involved in DNA repair play an important role in the normal aging process. In humans, the pathology of AT includes accelerated aging. Genome instability and DNA repair defects as a result of Atm deficiency could contribute to the premature features of this disorder. In addition, ATM plays an important role in telomere maintenance.27,28 AT cells show shortened telomeres and an increased incidence of telomeric fusions. Human AT patients might show premature features, at least in part, as a consequence of telomeric dysfunction.
DNA lesions have been implicated in the etiology of many diseases and in aging. The repair of oxidized bases in all organisms occurs primarily through the BER pathway. The critical step in the BER pathway is the recognition and excision of damaged bases by DNA glycosylase. Two mammalian DNA glycosylases, OGG1 and endonuclease III homolog (NTH1), were previously characterized, which excise the majority of oxidized base lesions.29 These DNA glycosylases have broad substrate specificities, but with preference for either pyrimidine or purine derivatives. NTH1 recognizes a wide range of oxidized pyrimidine derivatives, such as thymine glycol, 5-hydroxycytosine, dihydrouracil and at least six other oxidized pyrimidines. OGG 1 is primarily responsible for the repair of oxidation products of guanine, such as 7,8-dihydro-8-oxo guanine (8-oxoG).
The capacity of BER decreases with aging, accompanied with the decline in the activity of Ogg 1. Consequently, 8-oxoguanine lesions accumulate with aging.30 Furthermore, it was reported that in Huntington's disease, the somatic mutation associated with the onset and progression of disease, CAG-expansion, occurs in the process of removing these oxidized base lesions.31 Aging-dependent accumulation of 8-oxoguanine-lesions should elicit the increase of CAG expansion mediated by the repair reaction, resulting in the age-dependent onset of Huntington's disease.
Adaptive response to oxidative stress and aging
Adaptive responses are popular phenomena in various organisms. Among them, the radiation adaptive response is the most widely studied for social demands, such as energy, medical care and space development. Here we will describe the radiation adaptive response, and discuss its mechanisms and the effects of aging.
Over several hundred million years, aerobic organisms from prokaryotes to eukaryotes have developed adaptive mechanisms to regulate oxygen homeostasis.32,33 Thus, the longevity of a species likely depends on how well they accommodate the oxidative environment. Furthermore, in an individual organism, the process of aging is considered as the history of adaptation to the oxidative environment; therefore, an aged organism has various means of accommodation in the oxidative environment, different from a young organism. In the present study, adaptation to a chronic oxidative environment will be discussed, especially glutathione (GSH) and heat-shock proteins (HSP).
Radiation adaptive response
Ionizing radiation is the most widely studied for its adaptive response ranging from prokaryotes to eukaryotes (Table 2).34–49 As shown in Figure 3, the radiation adaptive response proceeds through three steps as follows: (i) recognition of redox imbalance as a result of low-dose irradiation; (ii) activation of signaling molecules from sensors to effectors, such as transcription factors; and (iii) the biosynthesis of functional proteins resulting in the acquirement of radioresistance. Because time is required for these processes, an interval between low-dose irradiation and subsequent high-dose irradiation is necessary to achieve radioresistance. Many factors contribute to the adaptive response in each process, as summarized in Table 3.40,43,50–65
Table 2. Radiation adaptive response in various biological systems
As a DNA damage sensor, ATM and DNA-PK are reported to be involved in the radiation adaptive response in some biological systems.43,50,51
Signaling molecules involved in the cell response to low-dose irradiation are PKC, p38, p53, ERK1/2, or NF-κB,40,43,52–56 etc. The factors localized to the cytoplasm, such as PKC, p38 and ERK1/2, are involved in the cell response to low-dose irradiation, suggesting that sensors of redox imbalance might probably exist in not only the nucleus but also the cytoplasm and membrane.
Bravard et al. examined the activities of anti-oxidant enzymes after low-dose and/or subsequent high-dose irradiation in AHH-1 lymphoblasts.59 They found that 1 or 3 h after high-dose irradiation, the activities of SOD2 (Mn-SOD), glutathione S-transferase, glutathione peroxidase and catalase were more increased in adapted cells than in non-adapted cells. The increased activity of some anti-oxidant enzymes results in the rapid scavenging of ROS and consequently less cell damage; however, it was not the same as in rat astrocytes.66 We examined the activities of catalase, glutathione peroxidase, glutathione reductase and glutathione content after low-dose and subsequent high-dose X-irradiation in rat astrocytes. Even though the experimental condition induced a radiation adaptive response in rat astrocytes, the activities of catalase, glutathione peroxidase and glutathione reductase, and the content of GSH were not significantly increased; therefore, anti-oxidant defenses only partly contribute to the radiation adaptive mechanism in astrocytes. The discrepancy regarding the induction of anti-oxidant systems in the radiation adaptive response seemed to depend on the biological system, such as the type of cells and the experimental conditions, including irradiation doses and their interval.
It was also reported that redox alteration through the activation of NF-κB and the induction of MnSOD might play a role in the adaptive response rather than in anti-oxidant activity.55 Guo et al. proposed that the induction of MnSOD after irradiation caused redox alterations, resulting in the upregulation of stress-responsive genes and the radiation adaptive response.
Organisms generally undergo qualitative changes with aging and their biological functions gradually degenerate. For the radiation adaptive response, aging represses the extent of adaptation. Venkat et al. examined the effect of low-dose irradiation on the frequency of micronuclei in human lymphocytes in younger (age 25–30 years), middle-aged (age 31–40 years) and older (age 41–57 years) people.67 They reported that the adaptive response depends on the age of the donor and decreases with age. Gadhia also found that aging abolishes the adaptive response by examining chromatid and isochromatid breaks in human peripheral blood lymphocytes.68
We have reported the radiation adaptive response in the growth of astrocytes.43 The radiation adaptive response in cultured astrocytes decreased with age and was not observed in 24-month-old rat cells. Furthermore, in this system, when the interval between low-dose and subsequent high-dose irradiation was 3 h, the radiation adaptive response was induced, whereas it did not occur with a 24 h-interval, even in young rat astrocytes. This suggests that the proteins responsible for the adaptive response should be induced 3 h after low-dose irradiation in astrocytes from young rats, and not induced 24 h after pre-irradiation or in astrocytes from aged rats. In order to explore the proteins responsible for the radiation adaptive response in astrocytes, we examined the effects of low-dose irradiation on protein expression and phosphorylation in young rat astrocytes using proteomics of 2-D PAGE.69,70 As a result, low-dose irradiation altered the expression of the elongation factor (EF)-2 fragment (spot A), phospho-β-actin (spots B and C), and phospho-EF-1β (spot D) (Fig. 4). The EF-2 fragment increased 3 h and reduced 24 h after 0.1 Gy-irradiation, and aging decreased temporal upregulation of the EF-2 fragment as well as induction of the radiation adaptive response (Fig. 5a). Alteration of the EF-2 fragment was not observed in 5 Gy-irradiated cells, showing that the response of the EF-2 fragment was specific to low-dose irradiation. In contrast, the alterations of phospho-β-actin and phospho-EF-1β were not significantly affected by aging (Fig. 5b–d). These results suggested that phospho-β-actin and phospho-EF-1β were not involved in the adaptive response depressed with age. In order to identify the EF-2 fragment as the protein responsible for the radiation adaptive response, further investigation is needed in relation to the function of the EF-2 fragment in the radiation adaptive response in cultured astrocytes. The EF-2 fragment is an attractive candidate for the factor responsible for adaptive response in rat astrocytes.
Adaptive mechanisms to chronic oxidative stress
Among anti-oxidant defense systems in aerobic organisms, glutathione (GSH) is one of the most ubiquitous anti-oxidants and the major regulator of the intracellular redox state. It is a tripeptide containing cysteine, which is a major reducing agent for maintaining intracellular redox balance. It was reported that GSH, GSH peroxidase and GSH reductase are induced in the lungs of rats exposed to ozone or treated with hyperoxia for several days.71,72 Furthermore, there is an interesting report that in long-lived Ames dwarf mice, the GSH level was significantly higher than in their wild-type siblings. Brown-Borg and Rakoczy suggested that the enhancement of resistance to oxidative stress is the mechanism underlying the extended longevity of these dwarf mice.73,74
In mammalian cells, heat-shock protein (HSP) synthesis is induced not only after hyperthermia, but also after exposure to oxidative stress. Induction of HSP synthesis can result in stress tolerance and cytoprotection against stress-induced molecular damage. In mammalian cells, induction of the heat-shock response requires activation and translocation to the nucleus of one or more heat-shock transcription factors that control the expression of a specific set of genes encoding cytoprotective HSP.75 Complementary DNA microarray analysis has provided evidence that the basal expression of HSP actually increases with aging.76,77 In the aged rat brain or human fibroblasts, it also results in an increase in the basal level of HSP70.76–79 This elevated expression should occur as a response to the age-associated accumulation of protein damage by oxidation. Age-associated elevations in the expression of heat-shock proteins have also been observed in Drosophila.80
Biological responses to oxidative stress have been investigated from various aspects, such as repair response and adaptive response. From these studies, the final functional molecules, which are involved in the protection and repair of oxidative damage, have been clarified, such as anti-oxidant molecules, DNA repair molecules, chaperone molecules and so on. Furthermore, it has been reported that these molecules also participate in the functional decline as a result aging and premature aging. Biological responses to oxidative stress play an important role in aging mechanisms as well as anti-oxidant capacity.
We thank Dr Shozo Suzuki for his guidance and warm encouragement throughout this study, Dr Tosifusa Toda (TMIG) for his thoughtful advice about proteomics and Ms Mayumi Kano for her technical assistance. This work was partly supported by a Grant-in Aid for Scientific Research (C) from the Japan Society for the Promotion of Science (No. 20590082, to Y.M.), a Grant from the “Ground-based Research Announcement for Space Utilization” promoted by the Japan Space Forum and a Grant from the Cosmetology Research Foundation.