A

DNA DAMAGE AND REPAIR

DNA is the master molecule and serves as the blueprint for the formation of all proteins and enzymes in every organism. The proteins then generate all the other substances in our cells. Thus, it is essential for reproduction, growth, and maintenance, and for sustaining normal living, that the DNA remains intact so that the genetic code can be read correctly. The stability and intactness of the DNA is a prerequisite for normal cellular functions, and there is good evidence that damage to the DNA can lead to cellular dysfunction, cancer and other diseases, or cell death. A major theory of aging holds that much of the aging phenotype (changes that can be observed) is caused by the gradual accumulation of DNA damage over a life span. DNA damage occurs at a high frequency due to metabolic processes and environmental factors such as various types of exposures and the intake of food and drugs. The prevention or repair of DNA damage is thus a major concern in biology and medicine.

DNA damage

The long, thin DNA molecules contain three components: nitrogen-rich bases, sugar groups, and phosphate groups. The composition of the four types of bases—adenine, guanine, cytosine, and thymidine—makes up the genetic code. Damage to DNA can occur to any of its three components. The bases are the most reactive, and there is far more knowledge about changes to them than to the sugar or phosphate components of DNA. Also, many chemicals or carcinogens form adducts (lesions) with new chemical groups being attached to existing DNA bases.

Base modifications in DNA after exposures

Living organisms are constantly exposed to stress from environmental agents and from endogenous metabolic processes. An important factor is exposure to oxidative reagents or oxidative stress. The resulting reactive oxygen species (ROS) attack proteins, lipids and DNA. Since proteins and lipids are readily degraded and resynthesized, the most significant consequence of oxidative stress is thought to be DNA modifications, which can become permanent via the formation of mutations and other types of genomic instability.

Many different DNA base changes have been observed following oxidative stress, and these lesions are widely considered to instigate the development of cancer, aging, and neurological degradation. The attack on DNA by ROS generates a low steady-state level of DNA adducts that have been detected in the DNA from human cells. More than one hundred oxidative base modifications in DNA have been detected in human cells (Dizdaroglu), and a few of these are shown in Figure 1. The best-known and most widely studied oxidative DNA base adduct is 8-hydroxyguanosine (8-oxoG).

Oxidative DNA damage is thought to contribute to carcinogenesis, and studies have shown that it accumulates in cancerous tissue. Furthermore, the cumulative risk of cancer increases dramatically with age in humans (Ames), and cancer can in general terms be regarded as a degenerative disease of old age. There is evidence for the accumulation of oxidative DNA damage with age, based on studies mainly measuring the increase in 8-oxoG, the best-known and most widely studied oxidative DNA base lesion.

DNA base damage also can occur after direct attack by external sources. Irradiation from various sources can directly damage bases in DNA. For example, ultraviolet irradiation from exposure to sunlight creates certain DNA lesions. Irradiation from γ-ray sources, such as X-rays, leads to many different kinds of lesions in DNA, including base modifications, sites with a loss of base, and breaks in the DNA strand. Since DNA contains two strands running in parallel but opposite directions, breaks can be either single-stranded or double-stranded. A large number of components of food intake can directly damage DNA. These include carcinogens and chemicals that cause DNA damage, either by direct reactions or via metabolic modification. For example, aromatic amines are found in variety of foods and are known to cause DNA damage and to be highly mutagenic. A number of poisons work by attacking the DNA and damaging it. An example of this is the poisonous gas nitrogen mustard, which causes modification of DNA bases and also can link DNA bases on opposite DNA strands. These cause serious havoc in the cell by completely blocking the progression of polymerases.

Detection of DNA damage

DNA can be extracted from human cells or tissues and then analyzed chemically for its components. There are various assays to detect DNA modifications. Some of these techniques are very sensitive and can detect rare changes in DNA. They include a number of chemical analyses and chromatographic measurements of DNA. A hotly debated issue is the choice of method to purify DNA from cells or tissues. Many of the available extraction procedures introduce new DNA damage in the process of purification. A number of enzymatic methods have been used in which specific enzymes or antibodies detect certain kinds of DNA base modifications and/or adducts formed in DNA. These enzymes become molecular "probes" for the damage, an approach that can be very sensitive. Radioactive labeling procedures can be a simple and easy way to measure several modifications in DNA.

The free radical theory of aging was first put forward by Denham Harman in 1956. He proposed that free radicals would be produced in the utilization of molecular oxygen by animal cells, and that as a consequence of free radical reactions with nucleic acids and other cellular components, the animal would develop mutations and cancer. He also suggested that damage by endogenous free radicals was the fundamental cause of aging. A second theory, proposed in 1959 by Leo Szilard, postulated specifically that time-dependent changes in somatic DNA, rather than other cellular constituents, were the primary cause of senescence. Both authors based their theories in large part on the belief, common at the time, that radiation accelerated aging independently of its effects on carcinogenesis.

Consequences of DNA damage

Figure 2 shows some of the consequences of DNA damage. As mentioned above, DNA damage can be induced by external or internal sources. Ultraviolet (UV) irradiation and ionizing irradiation are examples of exogenous sources of stress. Reactive oxygen species generated by the oxidative phosphorylation that occurs in mitochondria, and thus via cellular metabolism, is an example of an endogenous type of stress. Mutations in DNA can occur via replication of the damaged DNA whereby they become "fixed." Lesion bypass or replication errors can give rise to other forms of genomic instability. A lesion in DNA can block transcription (conversion of DNA to RNA) completely, it may truncate the transcript, or it may cause errors in the transcription. Alternatively, the DNA damage may induce new transcripts, and a number of genes have been shown to be inducible by various forms of cellular stress. These changes in transcription patterns that are caused by DNA damage may be part of the origin of the malignant phenotype; many changes in transcription have been reported in cancers. They are also likely to be a cause of some of the changes seen in aging, where reductions, or in some cases increases, in transcriptional activity are well established (Bohr and Anson). Lesions in DNA also can lead to cell cycle arrest, or they can cause strand breaks in DNA.

It is estimated that there are several thousand DNA alterations in each cell in the human organism per day (Lindahl), caused by both endogenous and exogenous stresses. Were it not for an efficient DNA repair process, genetic material would be destroyed by these processes over a normal human lifetime.

DNA repair

Mammalian cells can make use of a variety of DNA repair pathways. An overview of these is presented in Figure 3. Most studies over the years have been based on the assumption that the DNA repair pathways listed were confined to the removal of specific lesions within certain categories. For example, nucleotide excision repair (NER) was the system that removed bulky lesions in DNA that dramatically changed the DNA structure. In contrast, base excision repair (BER) was the process responsible for the removal of simple lesions in DNA, which are considered to cause only small structural changes in DNA and may not represent major blocks to transcription and replication. This concept has changed somewhat in recent years as it has become evident that many of the DNA repair pathways listed in Figure 3 are overlapping and share components. Thus, although it is useful to think of repair pathways as confined to the removal of different types of DNA lesions, that distinction is of limited validity. It would be much too ambitious to review all the pathways listed in Figure 3. Two of the most predominant pathways are NER and BER, and these will be discussed in more detail. The mismatch repair pathway is of particular interest in relation to cancer.

In general terms, the DNA repair process consists of a number of steps that act in concert to accomplish the complete repair of DNA damage. The first step is the recognition of the DNA lesion, and this is accomplished by proteins that constantly survey the DNA for any unwanted modifications. In the next step, incision, enzymes are cut into the DNA to remove the damaged DNA base. This step is complex and involves many proteins. When the damaged base has been removed, there is a step of resynthesis, in which new DNA is made to replace that which was removed in the incision step. This is accomplished by proteins called DNA polymerases, and there are several kinds of these in each mammalian cell. Since the DNA damage was in one strand of DNA, the other strand has the information required to copy itself. After the DNA synthesis, there is a ligation process in which the gaps in the DNA are sealed and a new, intact double helix is formed.

Base excision repair. Base excision repair (BER) of oxidative DNA damage is initiated by DNA glycosylases, a class of enzymes that recognize and remove damaged bases from DNA by hydrolytic cleavage of the base-sugar bond, leaving an abasic site (AP site). There are at least two pathways for further processing of the AP site. One of these is catalyzed by AP endonuclease, and results in a single-nucleotide gap that is then filled and sealed.

An alternative pathway, long patch BER, has been reported. In addition to a DNA glycosylase and AP endonuclease, it also involves a single-strand flap structure that is recognized and excised, and then the DNA is ligated (Klugland and Lindahl). These repair events result in a repair patch two to seven nucleotides long. There has been much research activity in the BER area (see Krokan et al.; Friedberg et al.).

Nucleotide excision repair. Most of the understanding of the nucleotide excision repair (NER) pathway has come through the study of the human disorder xeroderma pigmentosum (XP). There are seven genetically different types of this disease, designated A–G. XP proteins are designated after the cell line in which they are mutated (e.g., the XPA protein is the one mutated in XPA cells). The individuals afflicted with this condition suffer from high incidences of skin and internal cancers, hyperpigmentation, and premature aging (Friedburg et al.). Cells from XP patients cannot incise their DNA at a site of a UV-induced lesion, and thus are in general deficient in the incision process of NER. The enzymatic steps involved in NER are recognition of the lesion, incision of the DNA, excision of the damaged DNA template, resynthesis of new DNA based on the intact template, and ligation of the newly formed DNA repair patch into the reconstructed double helix. A number of reviews have discussed NER in much more detail (see, e.g., Friedberg).

Genomic heterogeneity of NER. A major advance in the study of DNA repair has been the insight that NER and possibly other repair pathways operate with considerable heterogeneity over the mammalian genome. The NER pathway can thus be subdivided into pathways relating to the functional and structural organization of the genome. Only about 1 percent of the genome is transcriptionally active, and a repair pathway called transcription coupled repair (TCR) operates here. The remainder of the genome, the 99 percent that is inactive, has a separate pathway, general genome repair (GGR). Much work has been dedicated to the further delineation and clarification of these pathways. For a more thorough review and discussion of the repair in genes, see Balajee and Bohr, which also discusses the human premature aging syndrome, Cockayne syndrome (CS), where TCR is deficient. CS is a rare and severe clinical condition in which patients appear to be much older than their chronological age. It is a premature aging disorder because many of the clinical signs and symptoms are those seen in the aging process in normal individuals.

Mitochondrial DNA repair in mammalian cells

Mitochondria are the energy stations inside the cells. Here, oxidative phosphorylation occurs, generating adenosine triphosphate (ATP). In this process, reactive oxygen species are formed at high frequencies, and the mitochondrial DNA (mtDNA) is directly exposed. The mitochondrial DNA does not have a recognized chromatin structure and thus is particularly exposed to formation of oxidative DNA base lesions.

There are about one thousand mitochondria per mammalian cell. Each mitochondrion has four to five DNA plasmids. This means that about 2 percent of total human DNA is in the mitochondria. All mtDNA is transcribed, whereas only about 1 percent of the nuclear DNA is transcribed. Thus the mtDNA makes up a large fraction of the total transcribed DNA in a mammalian cell.

MtDNA does not code for any DNA damage-processing enzymes. Thus, all repair enzymes functioning in the mitochondria need to be transported into them. It has been shown that mitochondria do not repair UV-induced lesions; this observation provided the basis for the notion that there is no DNA repair capacity in mtDNA. There have been many observations indicating the presence of BER in mitochondria. BER enzymes have been identified and characterized, and studies have shown that a number of oxidative DNA base lesions are efficiently removed from mtDNA. Other repair processes have also been detected (see Croteau et al. and other articles in the same issue of Mutation Research). An important question remaining is whether mitochondria possess any capabilities to repair bulky lesions via the NER pathway.

Knowledge about mtDNA repair is limited because it has been very difficult to study. Experimental techniques and methods have not been nearly as well developed as those for the study of nuclear DNA repair. Whereas in vitro repair studies have been performed with great success on nuclear or whole cell extracts from cells, this kind of biochemical approach has only very recently become available for mtDNA. Recent advances suggest that mtDNA repair can now be studied using more sophisticated biochemical analysis (Stierum et al.), and this should provide great advances in the near future.

Oxidative phosphorylation in mitochondria (which produces ATP) results in the production of reactive oxygen species (ROS). Other processes that contribute significantly to the pool of ROS include heat, ultraviolet light, drugs such as those used in the treatment of HIV, and ionizing radiation. Hydrogen peroxide, singlet oxygen, and hydroxyl radicals are among the ROS produced. The interactions between ROS and mtDNA result in oxidation of specific mtDNA bases, and such base modifications have been detected in human cells. Thus, insufficient mtDNA repair may result in mitochondrial dysfunction and thereby cause degenerative diseases, loss of energy formation, and pathophysiological processes leading to aging and cancer.

Identification of BER enzymes in mitochondria. Early indications for a BER mechanism in mitochondria came with the isolation of a mammalian mitochondrial endonuclease which specifically recognizes AP sites and cleaves the DNA strand. Later, it was demonstrated that a combination of enzymes purified from Xenopus laevis (a frog) mitochondria efficiently repair abasic sites in DNA.

The isolation of mitochondrial glycosylases has provided further evidence for a BER mechanism in mitochondria. Endonucleases specific for oxidative damage (mtODE), and for thymine glycols have been purified from rat mitochondria (Croteau et al., 1999).

The molecular mechanisms that lead to aging in multicellular organisms are still unclear. Many theories have arisen to explain the aging process, and among them the mitochondrial theory of aging, described earlier, has received much attention.

Age-related changes in DNA repair

There has been a great deal of interest in the question of whether DNA repair declines with age. Investigators have looked at many conditions of premature and normal aging, and many different types of assays have been used. In general, there seems to be a growing consensus that DNA repair declines slightly with age (see Bohr and Anson). One study reported a 1 percent decline in DNA repair capacity per year with advancing age in individuals (Wei et al.). Attempts to correlate DNA repair capacity of different organisms with maximum life span have been made. Hart and Setlow (1974) demonstrated a linear correlation between the logarithm of life span and the DNA repair capacity in cells from different mammalian species, suggesting that higher DNA repair activity is associated with longer life span. Although these studies document a connection between DNA repair capacity and age, there are also a large number of studies in which no connection between these two parameters were found.

Changes in mitochondrial function with age have been observed in several organisms. Experimental data from many laboratories suggest that the mitochondrial genome indeed accumulates DNA damage with age (Bohr and Anson, 1995).

Since oxidative DNA damage accumulates in mitochondria, changes in mtDNA repair with age have been studied. Initially, mtODE activity in mitochondrial extracts obtained from livers and hearts of rats six, twelve, and twenty-three months old was compared. In contrast to the common notion that DNA repair decreases with age, an increase in mtODE activity with increasing age was found. In both organs, activities at twelve and twenty-three months were significantly higher than at six months (p<0.01) (Souza-Pinto et al.). These results suggest that the changes observed in mtODE activity reflect a specific upregulation of the oxidative DNA damage repair mechanisms. Similar results were obtained when investigating mtODE activity in extracts from mouse liver mitochondria. The activity increased from six to fourteen months of age.

Perspectives

DNA repair is a complex and fascinating process, and it is very interactive with other DNA metabolic processes. It is tightly linked to the transcription process and also to DNA replication and a number of signal transduction pathways. An age-associated decline in DNA repair could explain why older individuals suffer from age-associated diseases and become highly susceptible to cancer. This understanding could also lead to therapeutic interventions in the future where DNA repair activities could be enhanced.

VILHELM BOHR

See also BIOLOGY OF AGING; CANCER, BIOLOGY; CELLULAR AGING: BASIC PHENOMENA; CELLULAR AGING: DNA POLYMORPHISMS; MOLECULAR BIOLOGY OF AGING; MUTATION; THEORIES OF BIOLOGICAL AGING: DNA DAMAGE.

BIBLIOGRAPHY

AMES, B. N. "Endogenous Oxidative DNA Damage, Aging and Cancer." Free Radical Research Communications 7 (1998): 121–128.

BALAJEE, A. S., and BOHR, V. A. "Genomic Heterogeneity of Nucleotide Excision Repair." Gene 250 (2000): 15–30.

BOHR, V. A., and ANSON, R. M. "DNA Damage, Mutation and Fine Structure DNA Repair in Aging." Mutation Research 338 (1995): 25–34.

CROTEAU, D. L.; STIERUM, R. H.; and BOHR, V. A. "Mitochondrial DNA Repair Pathways." Mutation Research 434 (1999): 137–148.

DIZDAROGLU, M. "Chemical Determination of Free Radical-Induced Damage to DNA." Free Radical Biology and Medicine 10 (1991): 225–242.

FRIEDBERG, E. C.; WALKER, G. C.; and SIEDE, W. DNA Repair and Mutagenesis. New York: ASM Press, 1995.

HARMAN, D. "Aging: A Theory Based on Free Radical and Radiation Chemistry." Journal of Gerontology 11 (1956): 298–300.

HART, R. W., and SETLOW, R. B. "Correlation Between Deoxyribonucleic Acid Excision Repair and Life Span in a Number of Mammalian Species." Proceedings of the National Academy of Sciences of the United States of America 71 (1994): 2169–2173.

KLUNGLAND, A., and LINDAHL, T. "Secondary Pathway for Completion of Human DNA Base Excision-Repair: Reconstitution with Purified Proteins and Requirement for DNase IV (FEN1)." EMBO Journal 16 (1997): 3341–3348.

KROKAN, H. E.; STANDAL, R.; and SLUPPHAUG, G. "DNA Glycosylases in the Base Excision Repair of DNA." Biochemical Journal 325 (1997): 1–16.

LINDAHL, T. "Instability and Decay of the Primary Structure of DNA." Nature 362 (1993): 709–715.

SOUZA-PINTO, N. C.; CROTEAU, D. L.; HUDSON, E. K.; HANSFORD, R. G.; and BOHR, V. A. "Age-Associated Increase in 8-Oxo-Deoxyguanosine Glycosylase/AP Lysase Activity in Rat Mitochondria." Nucleic Acids Research 27 (1999): 1935–1942.

STIERUM, R. H.; DIANOV, G. L.; and BOHR, V. A. "Single-Nucleotide Patch Base Excision Repair of Uracil in DNA by Mitochondrial Protein Extracts." Nucleic Acids Research 27 (1999): 3712–3719.

WEI, Q.; MATANOSKI, G. M.; FARMER, E. R.; HEDAYAT, M. A.; and GROSSMAN, L. "DNA Repair and Aging in Basil Cell Carcinoma: A Molecular Epidemiology Study." Proceedings of the National Academy of Sciences of the United States of America 90 (1993): 1614–1618.