All living cells accumulate DNA damage during the course of cellular reproduction and also as a result of exposure to radiation, oxidative agents, and radiomimetic chemicals in the environment. Many forms of DNA damage are caused by these processes, the most severe being double-strand breaks in the DNA helix. This form of damage is lethal in most organisms if unrepaired and can also lead to large-scale rearrangements of the genome in the form of chromosomal translocations, deletions, and loss of chromosome fragments. These types of rearrangements are common in cancer cells, which often have deficiencies in recognizing and repairing double-strand breaks in DNA. Genomic instability can result from misrepaired breaks and DNA rearrangements, which can ultimately lead to unregulated cell growth. Our long-term objectives are to understand at a mechanistic level how the recognition and repair of DNA double-strand breaks occur and also to characterize the molecular pathways that are misregulated during tumorigenesis.
The basic pathways of DNA double-strand break repair are conserved in all organisms, and many of the protein complexes responsible for the biochemical steps of these pathways are also conserved. One of these is the Mre11/Rad50 complex, which is found in archaea, prokaryotes, and eukaryotes, and is recognized as one of the first complexes in cells that senses and responds to DNA breaks. Together, Mre11 and Rad50 form the catalytic core of the complex and are conserved in all organisms, from archaebacteria to humans. The Nbs1(Xrs2) component is specific to eukaryotes and regulates the catalytic activities of Mre11/Rad50; it also performs essential roles in the signaling of DNA double-strand breaks. Much of our current understanding of the MRN(X) [Mre11/Rad50/Nbs1(Xrs2)] complex and its biological functions comes from Saccharomyces cerevisiae, in which mutants in each of the components were isolated on the basis of phenotypes of DNA damage sensitivity or failure in sporulation.
DNA End Processing
DNA breaks in eukaryotic cells are repaired by two major pathways, nonhomologous end joining and homologous recombination. In budding yeast, the MRN(X) complex is important for both of these pathways, although the nature of its role in homologous recombination has been elusive. Genetic evidence from S. cerevisiae and from multicellular eukaryotes suggests that the MRN(X) complex acts at the initiating stage of double-strand break repair, which includes recognition and processing of the DNA end in preparation for homologous recombination.
DNA end processing in eukaryotic cells involves extensive 5' to 3' resection to generate 3' single-strand overhangs that are bound by the Rad51 DNA recombinase. The Rad51-DNA filament catalyzes DNA strand exchange with a homologous template, which initiates DNA synthesis across the break site, followed by resolution of the intermediates into recombined products. One of the major questions still unanswered in the DNA repair field is the nature of the processing enzymes that create the 3' single-strand overhangs. The RecBCD enzyme performs this role in Escherichia coli but is not conserved beyond prokaryotes. The MRN(X) complex is widely believed to be responsible for this processing event, because of the rapid association of MRX with DNA ends in vivo, the presence of nuclease domains in the Mre11 protein, and observations of unprocessed DNA ends in hypomorphic mutants.
Our work using the human and yeast MRN(X) complexes in vitro has characterized the DNA-binding and DNA-unwinding activities of this enzyme, showing that MRN(X) recognizes and partially unwinds the ends of linear DNA. Our current work is aimed at understanding how MRN(X) complexes initiate the 5'-processing event in collaboration with other cellular enzymes and at investigating how this processing is regulated during cell cycle progression.
In S. cerevisiae, the Sae2 protein has been shown to act in the same pathway as MRN(X) in DNA double-strand break processing, but the role of this protein in DNA repair has been unclear. We have shown that recombinant Sae2 exhibits endonuclease activity and acts cooperatively with MRN(X) on model DNA substrates in vitro. We are continuing to characterize Sae2 catalytic activities and also to investigate the functions of human CtIP, a Brca1-interacting protein that appears to be the functional homolog of Sae2 in vertebrate cells. A mechanistic understanding of the factors that perform and regulate DNA double-strand break processing may also facilitate future efforts to manipulate the efficiency of homologous recombination in human cells.
DNA Damage Signaling
In eukaryotic cells, DNA repair is accompanied by a rapid signaling cascade mediated by protein phosphorylation. The response to DNA double-strand breaks is primarily mediated by the ataxia-telangiectasia mutated (ATM) protein kinase, which phosphorylates many substrates that initiate cell cycle arrest, DNA repair, and in some cases apoptosis. The phosphorylation of these substrates by the ATM protein kinase is critical for their activity after DNA damage. Some of these substrates are known tumor suppressors in humans, including p53, Chk2, and Brca1. Loss of the ATM protein in humans results in the genomic instability disorder ataxia-telangiectasia (A-T), characterized by sensitivity to DNA-damaging agents, immunodeficiency, and cerebellar degeneration.
The ATM protein has been refractory to biochemical analysis because of its large size and incompatibility with standard recombinant expression systems. We established an expression system for the protein and purified it in its physiological dimeric form. Using this protein, we reconstituted DNA damage–induced signaling in vitro with purified components, which showed that dimeric ATM depends on the MRN complex for multiple steps in the activation process. Specifically, the Mre11/Rad50 complex binds to and unwinds DNA ends, recruits ATM to these DNA sites, converts ATM dimers into active monomers, and facilitates stable binding of ATM to its substrates.
We are investigating the specific roles of the MRN components in the transition of ATM from its inactive to active state and exploring the roles of several tumor suppressors in regulating this process. ATM has been shown to be activated in early cancerous lesions and in aging tissues in humans. A mechanistic understanding of the activation process by DNA damage and other forms of cellular stress will help us to elucidate how ATM functions in these diverse cellular contexts.
Oxidative Stress Signaling through ATM
Cells from patients with A-T show signs of chronic oxidative stress and high levels of reactive oxygen species (ROS). Many of the deficiencies associated with ATM loss in mammals have also been linked to the effects of ROS, including loss of cerebellar neurons and hematopoietic stem cells, and higher rates of lymphoid cancers. We have identified a pathway of ATM activation that occurs through direct oxidation of ATM and does not require the MRN complex or DNA breaks. This pathway is required for apoptosis induced by ROS and also appears to regulate antioxidant functions in human cells. Our current work addresses the targets of ATM responsible for these effects and how this ROS-induced pathway functions in parallel with the DNA damage–dependent pathway of ATM activation in mammals.
This work was also supported by grants from the National Institutes of Health and the Cancer Research and Prevention Institute of Texas.
As of March 08, 2013