SUMMARY The central hypothesis of this Program Project is that defects in base excision repair (BER) drive human carcinogenesis and affect responses to cancer treatments. To test this hypothesis, we are using our strengths in fundamental biochemistry, molecular biology, structural biology and biophysics to examine human genetic variation in the BER enzymes. Our program is informed and driven by the identification of germline and tumor- associated enzyme variants that may alter the DNA repair capacity of human BER enzymes. The single- molecule approaches proposed in Project 4 will provide insights into how the human DNA glycosylases and the downstream enzymes in BER search for their targets in a sea of undamaged DNA as well as in a chromatin milieu. We hypothesize that, as we have shown for their bacterial homologs, the human DNA glycosylases, which remove oxidized bases, scan the DNA in a rotational manner employing an amino acid wedge to search for base damage. We further postulate that the downstream enzymes search for the product- bound prior enzyme in the pathway and that DNA compacted into nucleosomes may reduce BER enzyme diffusion rates. These hypotheses will be tested by: Elucidating the real-time DNA damage search behavior of the human DNA glycosylases and their variants (Aim 1); elucidating the spatial relationships and interactions between enzymes in the BER pathway relative to the site of DNA damage (Aim 2); and determining the diffusive behavior of the BER enzymes in the context of chromatin (Aim 3). To do this, we will examine the diffusive properties of the BER enzymes on both undamaged and site-specifically damaged DNA and on chromatin as well as on DNA or chromatin bound by the prior enzyme in the pathway. We also will determine if the diffusion of glycosylases is affected by mutations in putative reading head amino acids or by germline and tumor-associated variants that potentially affect the BER search process.