Molecular Life Sciences (2014)
DNA is not a pristine collection of the canonical four bases in Watson-Crick geometry. Modification of the bases and sugars in DNA is an ongoing process, with endogenous damage as well as exogenous (Lawley 1984). Oxygen and light are major causes of DNA damage. Although the frequency of total modifications can be considered low (i.e., one in 105 to 106 bases), the number is more than one per average gene. The modifications not only block DNA polymerases but also cause miscoding and the introduction of mutations into the genome. Paradoxically, drugs that modify DNA are used to treat cancers, although there is a finite risk of introducing DNA damage that could lead to a new cancer. More the 100 human genes are involved in DNA repair, indicating the importance of the damage, and deficiencies in these genes have been shown to have major phenotypic consequences.
DNA is the genetic material of organisms, and its integrity is essential for the maintenance of life. However, DNA is under constant attack from chemical and physical agents from inside and outside cells. Thus, DNA is not a pristine mixture of the four nucleobases (six including 5-methylcytosine and 5-hydroxymethylcytosine) but contains traces of a myriad of damaged products.
Chemistry and biology of DNA damage
The sources of DNA damage are many, including both exogenous and endogenous stressors. The endogenous stressors include physical agents such as ultraviolet (UV) light and other radiation, plus a plethora of chemicals such as pollutants, carcinogens in food, and even chemotherapeutic agents. Although one can make conscious lifestyle decisions that may reduce exposure to exogenous chemicals and physical agents that damage DNA, one can never completely avoid these agents, e.g., UV light. Aside from the exogenous stressors, there are many chemical modifiers of DNA within cells, and an individual has less control over these processes. These endogenous factors include alkylating agents (e.g., S-adenosylmethionine), oxidants (reactive oxygen and nitrogen species), and electrophilic products generated from oxidative and other intracellular reactions (e.g., Michael acceptors derived from lipids).
Before discussing the types of modification in more detail, a review of the significance of DNA modification is in order. The mutation theory of cancer goes back at least to Bauer in 1928 (Bauer 1928), predating the recognition of DNA in the genetic code. One of the first connections between
DNA modifications and mutations was mustard gas, which was shown to modify nucleic acid bases and also be mutagenic. The question of how modifications cause miscoding was already considered in the classic paper of Watson and Crick in 1953. The relationship between carcinogens and mutagens was also unclear for many years, but in vitro activation systems were very helpful in this regard (Ames et al. 1973) (although the caveat should be included that not all carcinogens are genotoxins). The study of mutagenesis is of relevance not only regarding cancer but also teratology (birth defects), atherosclerosis, and a number of other diseases (Ramos and Moorthy 2005).
The relationship between DNA modification and mutation was unambiguously established using an approach termed site-specific mutagenesis by Essigmann and his colleagues (Basu and Essigmann 1988). The approach involves the preparation of a vector containing a modified DNA base (in the first example, O6-methylguanine (Green et al. 1984)), its introduction into cells, and the analysis of the resulting mutations. Further, many molecular details of how modified DNA bases interact with DNA polymerases have now been established through combinations of structural and kinetic studies.
It is now generally agreed that DNA modifications can lead to a variety of potentially detrimental genotoxic effects, including base pair and frameshift mutations, strand breaks, and also complex events, e.g., deletions and recombination events. The incidence of DNA damage is inherently low, but a single problem in a gene could produce a dramatic biological effect. The problem is not so much that the process would lead to an inactive protein, in that the other allele is not damaged in the same way, but a mistake leading to a protein with abnormal function/regulation is more serious. An example is the oncogenes, in which modification at particular sites leads to loss of control and ultimately aberrant signaling in a cell. The tumor suppressor gene p53 has a number of functions, and loss of one allele can be quite detrimental (and heterozygotic mice have high tumor incidence). Thus, one can ask the question of how many adducts are a problem. Indeed, this question is very relevant in consideration of practical toxicology and setting exposure limits to potential carcinogens, which has considerable economic as well as health issues. The answer is that in principle even a single DNA adduct could conceivably result in cancer. However, the frequency of detrimental biological events associated with each DNA modification is low and, in addition, >100 genes are present to repair damage. In settings with experimental animals, a level of DNA modification of 1/105 bases is considered high and 1/104 would be dramatic. Many individual adducts are present at levels of 1/107 bases, some in apparently healthy individuals. What has been shown in experimental animal models is that there is often a correlation of the levels of DNA adducts with tumor incidence, e.g., with afiatoxin B1. Moreover, in people a higher incidence of DNA adducts (with afiatoxin bound) is seen in areas where people have exposure to afiatoxin and there is a high incidence of liver cancer (Wogan 1992).
A number of drugs that have been used to treat cancer act by modification of DNA, i.e., alkylating agents. The concept is that these drugs modify DNA and block replication; DNA is rather quiescent in most tissues but being replicated rapidly in tumors. Unfortunately the alkylation of DNA is not totally specific for the tumors and DNA in other tissues is modified. Related to this phenomenon is an increased risk of future tumors related to cancer chemotherapy (Kaldor et al. 1990).
The general mechanism of DNA modification follows several routes. Some chemicals are inherently reactive, e.g., certain direct alkylating agents used as drugs. Many other chemicals are rather inert but are activated by oxidation, reduction, or other processes to electrophilic agents or radicals that modify DNA. The list of chemicals thus activated includes oxygen. These activated chemicals vary in their stability, but in general they have half-lives on the order of at least seconds and can migrate to the nucleus to react with DNA. Some of the bulkier chemical entities have affinity for DNA due to intercalation and other physical forces, e.g., epoxides derived from afiatoxins and polycyclic aromatic hydrocarbons (Johnson and Guengerich 1997). Reaction then occurs with DNA, yielding a covalent product. The chemistry of these modifications generally fits into two categories: radical reactions (involving unshared electrons) and the reactions (2-electron reactions) of electrophiles with the nucleophilic sites of DNA.
Some of the DNA adducts thus formed are inherently stable and rather persistent, if they are not removed through DNA repair. In other cases the chemistry is unstable and hydrolytic, or rearrangement processes convert the initial adduct into another, with a possible increase or decrease in the potential for miscoding.
A plethora of genes and enzymes for handling DNA damage exist, not only to repair the damage but also to signal cells to arrest and quiesce. Information about these processes is contained in other parts of this book. One of the most striking pieces of evidence to support the premise that DNA damage is important in humans is the strong predilection of individuals with defects in some of the genes to cancer and neurological problems, e.g., xeroderma pigmentosum.
Ames BN, Durston WE, Yamasaki E et al (1973) Carcinogens are mutagens: a simple test system combining liver homogenates for activation and bacteria for detection. Proc Natl Acad Sci U S A 70:2281–2285
Basu AK, Essigmann JM (1988) Site-specifically modified oligodeoxynucleotides as probes for the structural and biological effects of DNA-damaging agents. Chem Res Toxicol 1:1–18
Bauer KH (1928) Mutationstheorie der Geschwulstenstehung. Springer, Berlin
Green CL, Loechler EL, Fowler KW et al (1984) Construction and characterization of extrachro- mosomal probes for mutagenesis by carcinogens: site-specific incorporation of O6- methylguanine into viral and plasmid genomes. Proc Natl Acad Sci U S A 81:13–17
Johnson WW, Guengerich FP (1997) Reaction of afiatoxin B1 exo-8,9-epoxide with DNA: kinetic analysis of covalent binding and DNA-induced hydrolysis. Proc Natl Acad Sci U S A 94:6121–6125
Kaldor JM, Day NE, Pettersson F et al (1990) Leukemia following chemotherapy for ovarian cancer. N Engl J Med 322:1–6
Lawley PD (1984) Carcinogenesis by alkylating agents. In: Searle CE (ed) Chemical carcinogens, vol 1, 2nd edn. American Chemical Society, Washington, DC, pp 325–484
Ramos KS, Moorthy B (2005) Bioactivation of polycyclic aromatic hydrocarbon carcinogens within the vascular wall: implications for human atherogenesis. Drug Metab Rev 37:595–610
Wogan GN (1992) Afiatoxins as risk factors for hepatocellular carcinoma in humans. Cancer Res Suppl 52:2114–2118