DNA repair causes damage

A DNA repair mechanism may come at a cost -- cancer-causing mutations, according to a study published this week in Science.

DNA ligase repairs chromosomal damage
Image: Wikimedia commons, Tom
Ellenberger Washington
University School of Medicine

A supposedly accurate DNA repair mechanism employed by cells to fix double-strand breaks can surprisingly increase the nearby mutation rate by up to 1400 times, providing a possible explanation for the accumulation of tumor-causing mutations in cancerous tissues. Previously, "we saw that DNA breaks are present in precancerous legions," said molecular oncologist linkurl:Thanos Halazonetis;http://www.molbio.unige.ch/halazonetis/ of the University of Geneva, who did not participate in the research. "These breaks, based on this finding now, could perhaps explain [some of the] mutations in oncogenes that drive cancer progression." Cancer cells are the subject of unusually high numbers of mutations, but for many cancers, the cause of those mutations is unknown. Chromosomes are commonly rearranged in precancerous tissues, a result of double-strand breaks (DSBs) in the DNA. But how the breaks lead to these mutations has been "a longstanding mystery," said geneticist linkurl:Dmitry Gordenin;http://www.niehs.nih.gov/research/atniehs/labs/lmg/cs/staff.cfm of the Laboratory of Molecular Genetics at the National Institute of Environmental Health Sciences, who was also not involved in the study. To see how these two events were related, linkurl:James Haber;http://www.bio.brandeis.edu/faculty/haber.html of Brandeis University in Massachusetts and his colleagues induced DSBs in a specific location in the yeast genome, and measured the resulting mutation rate surrounding that locus. They found that the mutation rate indeed increased -- up to 1400 times more than the normal spontaneous mutation rate would predict -- and it was the cells' attempts to repair the damage that was responsible. "Even though the repair process is designed to be very accurate," Haber said, "it turns out that it comes at the cost -- a high rate of mutation associated with the event." But what was most surprising, researchers say, was the repair mechanism that was messing everything up -- the usually accurate process of homologous recombination, in which DNA polymerase uses a sister chromatid or homologous chromosome as a template for re-synthesizing the missing DNA. While the DNA polymerases commonly involved in mismatch repair are known to cause relatively frequent errors, the DNA polymerases involved in recombination are the same as those that are responsible for normal DNA replication, which were "assumed to be very accurate," said molecular oncologist linkurl:Lawrence Loeb;http://depts.washington.edu/mcb/faculty_profile.php?fcid=96 of the University of Washington, who was not involved in the research. "People usually thought that the process of recombination would be error-free, and here [they have] shown it can be highly error-prone." In addition to point mutations, where an incorrect nucleotide is inserted into the DNA sequence, the researchers also observed more dramatic alterations called frameshift mutations in the area of the DSB. These mutations occur when the DNA polymerase that is re-synthesizing the missing code slips off the homologous sequence it is reading and reattaches somewhere else, either on the same chromosome or a different one entirely. "These [types of mutations] involve incredible instability of the copying process," Haber said. "[The polymerase] is not holding on to the template; it gets partway across and falls off." When this happens, "the other bit of homology doesn't fully align, [and] the switch itself causes error," added Loeb. Once detached, the polymerases could even reattach to the strands of DNA they had just constructed, folding back on themselves to create a little "hairpin." These hairpin mutations are particularly interesting with regard to their implications for cancer, Haber noted, as such mutations can be found in the human p53 mutation database. "These odd hairpin-like mutations are not just a yeast thing; they're apparently frequently found in human cancers," he said. Of course, before this mechanism can be assumed to apply to human cancers, "one would have to validate or repeat this study with human cells," Halazonetis said. "But because homologous recombination is conserved in yeast and human cells, I would think that this probably would be true in human cells." Consequently, "these breaks may explain a fraction of the mutations in human cancers." W.M. Hicks, et al., "Increased mutagenesis and unique mutation signature associated wit mitotic gene conversion," Science, 329:82-5, 2010.
**__Related stories:__***linkurl:Ancient microbes repair DNA;http://www.the-scientist.com/news/display/53546/
[27th August 2007]*linkurl:DNA repair mapped, systems-wide;http://www.the-scientist.com/news/display/23466/
[19th May 2006]*linkurl:DNA Damage Responses: Cancer and Beyond;http://www.the-scientist.com/article/display/15766/
[10th October 2005]