also be repaired as showed by others in cultured rat cells. They measured the frequency of single-strand breaks and oxidative base damage in mtDNA light and heavy strands by ligation-mediated PCR and a quantitative Southern blot technique coupled with digestion by the enzymes endonuclease III and formamidopyrimidine DNA glycosylase. After treatment with 5-^M alloxan for 1 h they found a rapid repair, being complete prior to 4 h with no evidence that damage removal occurred in a strand-specific manner (61).
While mitochondria lack NER due to damage induced by UV radiation, transcription-coupled repair pathways that do not involve NER could theoretically be present. This possibility was investigated in 1998. The authors used photoactivated methylene blue to damage purified nuclear DNA and mtDNA of human fibroblasts in culture (64). The primary product of this reaction, 8-oxo-dG, was quantified using E. coli FPG DNA glyco-sylase in a gene-specific damage and repair assay under conditions that would produce an average of three oxidative lesions per double-stranded mitochondrial genome. Using this approach, they presented the first results on efficient removal of singlet oxygen-induced base damage from human mtDNA. Within 9 h, 47% of the induced damage had been removed by the cells. This was neither due to replication or cell loss nor due to degradation of damaged genomes - a so far believed hypothesis - but due to a specific mtDNA repair pathway. The mtDNA repair, however, was not transcription coupled as it usually is in nuclear DNA. In mitochondria, the heavy and the light strands are transcribed in different frequencies. One would expect this to result in differences in the rate of repair. However, no differences were observed, suggesting that most or all of the regions are repaired independent of transcription.
Mitochondria show base excision repair (BER) activities
Certain lesions that are known to be removed by BER from the nuclear DNA, are efficiently removed in mitochondria, whereas some bulky lesions, e.g. complex alkylation damage produced by nitrogen mustard, that are usually removed by NER, are not repaired in these organelles (51). These findings suggest efficient BER activities in mitochondria. Further support for BER in mitochondria comes from the fact that enzymes involved in BER have been purified from mitochondria. Pinz and Bogenhagen isolated a combination of enzymes purified from Xenopus laevis mitochondria and demonstrated efficient repair of abasic sites in DNA being the first to report a complete reconstitution of BER by using mitochondrial enzymes (65). They could also prove that the mtDNA polymerase y is not only a replicative polymerase but has the ability to participate in an efficient short patch repair process.
BER is an important mechanism that cells use for the removal of oxidative damage. It is initiated by DNA glycosylases, a class of enzymes that recognize a specific set of modified bases such as 8-oxo-dG or TG: a damage-specific glycosylase recognizes a damaged base and then cleaves the N-glycosylic bond between the sugar and the base, generating an AP site (apurinic/apyrimidinic). Some glycosylases have an associated AP sylase function that cleaves the DNA phosphate backbone, while others rely on AP endonucleases for strand cleavage. Then, a phosphodiesterase excises the 3'-terminal unsaturated sugar derivate, and a DNA polymerase resynthesizes the resulting one-nucleotide gap; the ends are sealed by a DNA ligase (54,66).
Biochemical changes may be caused by the threshold effect
The level of any one mtDNA mutation, be it a point mutation or deletion will be insufficient to result in the gross physiological and pathological changes associated with the decline of bioenergy capacity observed in aged tissues. Thus, it has been reported that the occurrence of the common 4977-bp deletion in skeletal muscle accumulates with age but only up to 0.1% at age 80-90 years. This can support the idea of the threshold effect, by which a phenotypic lesion becomes evident only when over 80% of the mtDNA in a cell is mutated by different alterations, and of mitotic segregation, by which the proportion of mutant mitochondrial genomes may shift in daughter cells during cell division. It is suggested that 5-10% intact mtDNA molecules per cell may be sufficient to maintain viable cellular bioenergetic functions (67).
Porteous et al. encouraged this theory by showing that the bioenergetic function of cybrids containing less than 50-55% of 4977-bp deleted mtDNA was equivalent to those cybrids with only intact mtDNA molecules. But more than 55% deleted mtDNA led to a decrease of mitochondrial membrane potential, rate of ATP synthesis, and the cellular ATP:ADP ratio (68). Other studies on different cell lines confirmed these findings. EtBr, for example, is an intercalating drug that, in isolated HeLa cell mitochondria, inhibits preferentially rRNA synthesis over mRNA synthesis. Total mtDNA decreased when treating NT2 cells with 25-^M and 50-^M EtBr, converting to 9.85 and 19.7 ^g/ml. Concentrations over 6 ^g/ml led in a mouse pancreatic P-cell line to a complete cessation of cell growth (69) which approximately matched our results (70). These cells finally died (or nearly died) after 16 and 22 days, respectively when they had only about 10% of their mtDNA left. This might be a threshold, which is important for cell survival and differs between cell types depending on the energy demand of the particular tissue (71). This same threshold could be found in HeLa cells treated with 50 ng/ml EtBr. The mtDNA was depleted to about 10% of normal after 6 days of growth; longer treatment resulted in extensive cell death (72).
Complex lesions such as cisplatin ICLs are removed via homologous recombination in yeast and prokaryotes (reviewed in (73,74)). Since mitochondria from hamster cells also show repair of these alterations (51), a recombination repair pathway in mammalia seems possible. In 1996, this hypothesis was supported by the work of Thyagarajan et al. (75). They demonstrated recombination between plasmids using human mitochondrial protein extracts from normal and immortalized mammalian somatic cells. By showing that pretreatment of these protein extracts with affinity-purified anti-recA antibody (recA is a protein involved in recombination in bacteria) reduced homologous recombination activities by approximately 90%, further evidence for this hypothesis was gained (75).
Another indication for recombination is the removal of 4NQO damage (76). This kind of damage is generally thought to be removed via a NER pathway in nuclear DNA. In mitochondria, which lack the NER pathway, as mentioned above, recombination activities might be an alternative method to repair DNA alterations.
Was this article helpful?