DNA Interstrand Crosslinks, Cancer and Aging

DESCRIPTION (provided by applicant): DNA is continuously damaged and repaired. The consequences of unrepaired damage are revealed by genetic diseases in which DNA repair pathways are disrupted. Repair deficiency in humans can lead to a profound risk of cancer or dramatic premature aging. Prevention of cancer and age-related morbidity are therefore dependent upon identifying the sources of DNA damage and mechanisms of damage repair. DNA interstrand crosslinks (ICLs) present a unique challenge to cells because both DNA strands are covalently modified and inseparable, precluding transcription, replication and simple error-free excision repair of the DNA. The mechanism by which ICLs are repaired in mammalian cells is poorly understood, as are levels of endogenous crosslink damage and the physiological impact of these lesions. The long term objectives of this research are to understand the biological impact of unrepaired DNA ICLs and the mechanism by which cells repair ICL damage. Ercc1-Xpf is an endonuclease required for nucleotide excision repair of bulky lesions on one strand of DNA and the repair of bivalent ICLs. Genetic disruption of Ercc1 in the mouse causes accelerated aging, which cannot be attributed to defective nucleotide excision repair. We hypothesize that the phenotype of the Ercc1-deficient mice is due to their inability to repair ICLs, and therefore the consequence of endogenous ICLs. Experiments proposed will utilize Ercc1 mouse models to test this hypothesis and to reveal the consequence of unrepaired ICLs. In addition, cells derived from these animals will be used to delineate the steps of lCL repair. The specific aims of this project are: Aim 1 is to test the hypothesis that unrepaired DNA ICLs contribute to aging and carcinogenesis. We engineered two novel mutations in the mouse Ercc1 genomic locus that result in hypomorphic mice. Preliminary data indicates that these mice have a similar, but more extensive, progeroid phenotype compared to the Ercc1-deficient mice. However, the disease-free period and longevity of the mice are extended proportional to Ercc1-Xpf protein levels. To test our hypothesis, we will treat these mice with a drug that causes ICLs and compare aging parameters to untreated animals. We predict that the phenotype of the Ercc1-depleted mice will be exacerbated by drug treatment indicating a causal role for ICLs. In addition, we will measure aging parameters including tumor incidence in the longest-lived of the Ercc1 mice to determine the contribution of spontaneous ICLs to cancer. Aim 2 is to determine if lipid peroxidation (LPO) promotes aging in mice with defective ICL repair. LPO is caused by oxygen radical damage to membranes and can yield products able to crosslink DNA. We hypothesize that LPO is a source of spontaneous ICLs that contribute to the phenotype of the Ercc1 mice. LPO will be induced in Ercc1 hypomorphic mice chemically and through nutritional intervention. We predict that both will exacerbate the progeria of the mice. Results from these experiments will indicate if LPO can affect the rate of aging and if so whether this effect can be controlled through diet. Aim 3 is to develop a method to locally induce DNA ICLs and to use this technique to determine the sequential steps of lCL repair. The study of ICL repair is hampered by the fact that ICLs are formed inefficiently and drugs that cause ICLs induce a host of other DNA damages. Thus we propose to develop a technique to introduce ICLs only in a well-defined region of cell nuclei using near infrared multiphoton photoactivation of the chemotherapeutic agent psoralen. This method will enable us to discriminate ICL-specific events from other repair events within a single cell. The technique will be applied to wild type and ICL repair-defective Ercc1-/- cells to determine what proteins assemble specifically at sites of lCL damage and in what chronological order.