Simon Powell: Clinical Translational Studies of DNA Repair in Human Cancers

Sporadic Breast Cancer Defects in Homologous Recombination

The BRCA1-BRCA2 pathway can be functionally inactivated in sporadic breast cancer, perhaps as frequently as in 20 percent of cases, suggesting that the pool of patients with defects in homologous recombinational repair is larger than initially expected. The functional assessment of defects in homologous recombination was assessed by ionizing-radiation- (ex-vivo-exposure-) induced Rad51 foci and confirmed by characteristic large block deletions of array comparative genomic hybridization. The inactivation of the BRCA1-BRCA2 pathway allows the use of specific therapies designed to exploit the defects in DNA repair.

Drug Screen for Homologous Recombination

We are screening drug libraries for agents that specifically target key double-strand break (DSB) repair pathways in tumor cells. We recently have designed and validated a cell-based, high-throughput screening platform to identify small molecules that disrupt the balance between homologous recombination and non-homologous end joining. Candidate drugs will be tested for their radiosensitizing potential using in vitro and in vivo secondary assays, as a means to further develop new therapeutics that may be combined with radiotherapy in order to enhance tumor cell kill.

A second screen is to look for drugs that are specifically toxic to BRCA1- or BRCA2-deficient cells, by utilizing the capacity to observe the simultaneous growth of both genotypes courtesy of different fluorescent markers. We have validated the screen for identifying known BRCA-sensitizing drugs such as mitomycin C and PARP inhibitors. The first screening dataset is currently being reviewed.

Why Do BRCA1 and BRCA2 Mutations Lead to Cancer in Specific Tissues?

The reasons why BRCA1 and BRCA2 mutation carriers develop breast and ovarian cancers are currently unknown. Breast and ovarian tissues are subject to oxidative stress as a result of hormonally driven proliferation during each menstrual cycle. BRCA1 and BRCA2 mutation carriers have almost identical clinical syndromes in spite of the significantly different molecular functions of the two proteins, suggesting that their common role in homologous recombination (HR) is protecting the epithelium from developing cancer.

BRCA1- or BRCA2-deficient cells have been found to be more sensitive to oxidative stress, modeled by treatment with hydrogen peroxide. The inhibition of short-tract or long-tract base excision repair (BER) by depletion of XRCC1 or FEN-1 respectively resulted in mild sensitization to hydrogen peroxide. However, no additional toxicity was found to be specific for BRCA-deficient cells, suggesting that BER is not involved, directly or indirectly, in the sensitivity of BRCA1- and BRCA2-deficient cells to hydrogen peroxide. BRCA-deficient cells accumulate in S-phase after exposure to hydrogen peroxide with associated DNA double-strand breaks (DSB), whereas BRCA-proficient cells develop DSB, but show no accumulation in S-phase. The consequence of developing DSB in S-phase in the absence of a functioning BRCA1-BRCA2 pathway of HR is the development of a large increase in chromatid-type aberrations and the observed large-scale gains and losses by whole genome analysis, characteristic of BRCA-deficient cells. Thus, BRCA1-BRCA2 mediated HR is important for the repair of DSB in S-phase, which is the cytotoxic lesion arising from oxidative stress.

Oxidative damage is formed after exposure to reactive oxygen species (ROS), such as hydrogen peroxide, superoxide anions, or hydroxyl radicals. Consequences of ROS include DNA damage, lipid peroxidation, and amino acid oxidation. ROS is able to cause oxidative damage to the DNA in the form of modified bases, abasic sites, or single-strand breaks. It has been shown that there is increased oxidative damage to DNA in epithelial tissues, potentially due to increased exposure to estrogen and its metabolites.

We have designed plasmids containing oxidative DNA lesions in close proximity. Using 2D gel electrophoresis followed by Southern blotting, we plan to identify whether an increased amount of oxidative damage causes replication stalling, and if so, which proteins are localized to the stalled replication fork using ChIP. The aim is to identify any differences in protein recruitment between wild type and BRCA1/2-deficient cells during oxidative damage processing.

The implied etiology of BRCA1- and BRCA2- associated breast cancer is that oxidative stress is the endogenous trigger of carcinogenesis arising from hormonally driven proliferation with each menstrual cycle. Thus, there is a potential role for antioxidants in breast cancer prevention.