5. DNA DAMAGE AND REPAIR
Living cells are constantly exposed to environmental agents and endogenous processes that inflict damage to DNA. Several complex enzymatic mechanisms have evolved to repair DNA lesions, and lately there is tremendous progress towards greater understanding of the mechanisms involved. There are several major pathways of DNA repair, and the particular pathway used depends in part upon the type of DNA lesion that is being removed. Ultraviolet (UV) exposure generates two major adducts in DNA: the pyrimidine dimer and the 6-4 photoproduct.
Both of these adducts, as well as other bulky lesions, are removed by the nucleotide excision repair (NER) pathway. Endogenous metabolic processes generate oxidative DNA damage that is removed from the DNA mainly by the base excision repair (BER) pathway. A large number of DNA base modifications caused by oxidative stress have been detected in various mammalian cells, and have been found at higher levels in cancer cells (1). It is not clear which of these many modifications are most biologically relevant, but the one that is most widely studied is 8-hydroxydeoxyguanine (8-oxoG) (Fig. 5.1). Several investigators have reported accumulation of 8-oxoG with age (2).
Fig. 5.1. Comparison of the DNA structures after 500ps of molecular dynamics; from the left: native DNA and DNA with 8-oxoG. Deformed B-DNA structure at around the lesion is result of collapsing hydrogen bond network. Full circles represent atoms of 8-oxoG and guanine of native DNA that is replaced with 8-oxoG.
The DNA repair process varies significantly in efficiency between different regions of the genome. Specifically, DNA repair occurs preferentially in genes and particularly those that are actively transcribed (3). A component of this repair is directly linked to the basal transcription process, and is termed transcription coupled DNA repair (TCR) or "strand specific" DNA repair, since the transcribed strand of the active genes is preferentially repaired (4). Both CS and WS cells show deficiencies in TCR, and this may contribute to the molecular phenotypes of the diseases.
5.1. SEQUENCE-SPECIFIC DNA DAMAGE BY IRON-MEDIATED FENTON REACTIONS
The cytotoxicity of hydrogen peroxide is mediated by highly reactive oxidants generated by the reaction of reduced transition metals with H2O2 via Fenton reactions (5). H2O2
Fe2+ + H2O2 + H+ —> Fe2+ + •OH + H2O
DNA is an important target in H2O2-mediated cell killing through the reaction with Fe2+ (6, 7), for which the oxidizing species is probably not a freely diffusible hydroxyl radical (oOH) but possibly a localized oOH or a related iron-oxo species (8-10). The properties of strand-breaking oxidants are apparently governed by the chelation state of the iron upon which they are generated (11). DNA damage in vivo and in vitro mediated by Fe2+ and H2O2 has a peculiar dose response to H2O2 concentration (6, 7, 11). The rates of cell killing and DNA damage induction are maximal near 0.5-2.5 mM H2O2 then drop off and become independent of H2O2 concentration between approximately 5 and 50 mM. This response is the result of two types of oxidants. Type I oxidants are scavenged by H2O2 to form a less reactive species,
exemplified by the reaction of -OH with H2O2:
OH + H2O2 —> HO2• + H2O
Hence, these oxidants appear to damage DNA only at lower H2O2 concentrations; they are also scavenged by alcohols. The constant rate of damage observed with H2O2 concentrations between 5 and 50 mM was rationalized by proposing that this DNA damage might be induced by the reaction of H2O2 with Fe2+ that is intimately bound to DNA so that the nascent Type II oxidants would react with the DNA before encountering exogenous scavengers such as H2O2 or ethanol (11). Alternatively, one could propose that Type II oxidants do not react with H2O2 or ethanol. In either case, these oxidants would be formed on Fe2+ ions associated differently within the DNA helix than those which give rise to Type I oxidants. DNA strand breaks arise from reactions of the oxidants with deoxyribose residues (7, 11).
Because of the high reactivity of oOH and related oxidants, damage to DNA in vivo or in the presence of scavengers occurs predominantly with such radicals only when they are generated by reaction of H2O2 with metal ions bound to the DNA (12-14). The binding of metals to preferred sequences within DNA and the relationship of such associations to DNA damage by the Fenton reaction have been the subject of recent investigations (13, 15-17), and it has been proposed that transition metals in the presence of H2O2 might cause sequence-specific damage to DNA (12, 18).
Nucleoside damage following exposure of DNA to Fe2+ and H2O2 was not equally distributed among the nucleosides. For example, with 2.0 mM H2O2, the damage frequencies were dA:dG:dC:dT 1:1.3:1.5:2.0 (19). It has been proposed that DNA damage in 0.5 mM H2O2 was caused by Fe2+ ions which associated differently to DNA than those which cause damage at 50 mM H2O2, it was natural to ask whether strand breaks varied in sequence location, following the two conditions of reaction. Indeed we have very recently demonstrated that preferential cleavage sites due to iron/ H2O2 differ for Type I (0.5 mM H2O2) and Type II (50 mM H2O2) oxidants, and that these sites can be grouped into consensus sequences.
In some studies it was examined the location of cleavages after treating 144- and 191-base pair PM2 DNA restriction fragments with Fe2+ and either 0.5 mM or 50 mM. When 50 mM H2O2 was present, preferential cleavage occurred at the nucleosides 5' to dG moieties in the consensus sequence, RGGG. With 0.5 mM H2O2, however, small amounts of cleavage occurred at RGGG as would be expected if Type II radicals were to form to a limited extent in 0.5 mM H2O2, but preferential cutting occurred mainly within the consensus sequences RTGR, and also within the sequences (T)ATTY, CTTR, and YAAGT. (10 mM ethanol inhibited all of these preferential cleavages (except those at RGGG), as expected for cleavages by Type I radicals).
5.1.1. FENTON REACTIONS, DNA DAMAGE AND AGING
If we define "development" as the programmed and ordered establishment of chromosomal regulatory and structural patterns, then we might distinguish "aging" as the unprogrammed and random (stochastic) disestablishment of chromosomal regulatory and structural patterns. Should we apply this definition, it becomes relevant to aging that the iron-mediated Fenton reactions would appear to lead to damage of regulatory elements containing RTGR as well as structural elements (telomers that contain RGGG and centromers that contain another susceptible sequence, TATTY). How could these damages result in aging? Perhaps the repair efficacy of these damages drops with age, or the rate of DNA (chromosome) damage increases with age alternatively. DNA repair might be constant, but slightly less effective than is necessary for the frequency of DNA damage or certain types of DNA damage might not be subject to DNA repair and thus accumulate.
Finally, certain types of damage might not be repairable in certain types of cells (e.g. non-mitotic cells). We might also consider damage to mitochondrial DNA. Whereas mitochondrial genomes do not contain telomeres, they do contain the sequences that are sensitive to iron-mediated DNA damage. How are such damages processed? How do mitochondria selectively remove damaged mitochondrial genomes from a mitochondrion with several hundred genomes within a cell with several hundred mitochondria? Do mitochondrial DNA damages accumulate with age, and does this accumulation contribute to an aging error catastrophe? Each of these questions is actively being explored in many laboratories at present and we eagerly await their answers and conclusions.