PROJECT SUMMARY
The human genome is constantly attacked from sources that include environmental pollutants, other
exogenous origins that include drug treatment, endogenous reactive oxygen species, and UV light. Among the
lesions/adducts are ones derived from polycyclic aromatic compounds, widespread byproducts of fossil fuel
combustion found at toxic waste dumps, superfund sites, in our air, food and water. The resulting DNA lesions
cause mutations that lead to cancer. However, not all DNA lesions are equally carcinogenic, as their mutagenic
propensities vary: a cascade of processes determines whether they are repaired, or survive for mutagenic or
error-free bypass by DNA polymerases. Human nucleotide excision repair (NER) is a key mechanism for
removal of many such DNA lesions. The vital importance of NER is demonstrated in the devastating human
disorder xeroderma pigmentosum, caused by mutations in NER genes. Notably, some lesions are rapidly
repaired, some slowly, and some are resistant and thus particularly genotoxic, a phenomenon that is poorly
understood. Likewise, there is a gap in our understanding of the mechanisms underlying DNA lesion bypass by
polymerases that can lead to a mutagenic or error-free outcome. The objective of this project is to provide
mechanistic insights into the puzzling variability of DNA lesion mutagenicity, focusing on the key steps of lesion
recognition for repair and mutagenic bypass, to yield integrated new molecular and dynamic understanding of
lesion mutagenic proclivity in unprecedented atomistic detail, using molecular dynamics simulations.
Our overall hypothesis is that the structure of the lesion and its base sequence context determine its
overall mutagenic propensity. In Aim 1, we will utilize a selected set of DNA lesions/adducts whose structures
differ greatly in size and shape, placed in differing sequence contexts, to determine structural, energetic and
dynamic characteristics of the lesion-containing DNAs as they bind to Rad4/XPC, the yeast homolog of the
human XPC lesion recognition protein. We will reveal how those that bind for productive recognition leading to
excision differ from those that fail to do so. In Aim 2 we will determine how the human XPD helicase in TFIIH,
that verifies the presence of lesions for NER by stalling, processes lesions of different sizes and shapes, and
how XPD mutations that cause human disease inhibit XPD’s function. In Aim 3 we will determine how differing
lesion structures in varying nucleosomal positions impose different distortions on the nucleosome and how
selected histone acetylations modulate these distortions, to promote or inhibit access for repair. In Aim 4 we
investigate endogenous and exogenous DNA peptide crosslink lesions, to determine how selected DNA
bypass polymerases process them error-free or mutagenically, in differing DNA sequence contexts.
Focusing on the most mutagenic lesions, our work will facilitate identification of appropriate biomarkers
for determining risk of developing cancer, advance design of chemotherapy drugs that are less repaired, and
yield a predictive tool to identify mutational hotspot sequences induced by different lesions in human tumors.