Electronic Thesis and Dissertation Repository


Doctor of Philosophy




Dr. Hong Ling


Y family DNA polymerases are specialized enzymes for replication through sites of DNA damage in the genome. Although the DNA damage bypass activity of these enzymes is important for genome maintenance and integrity, it is also responsible for DNA mutagenesis due to the error-prone nature of the Y family. Understanding how these enzymes select incoming nucleotides during DNA replication will give insight into their role in cancer formation, aging, and evolution. This work attempts to mechanistically explain, primarily through X-ray crystallography and enzymatic activity assays, how Y family polymerases select incoming nucleotides in various DNA replication contexts. Initially, we sought to determine how the model Y family polymerase Dpo4 differentiates between ribo and deoxyribonucleotides. Crystal structures were solved of a mutant Dpo4 enzyme (Y12A) deficient in ribonucleotide discrimination, incorporating either deoxy-adenine (dA) or ribo-adenine (rA) nucleotides opposite template thymine DNA. It was revealed that the Dpo4 Y12A mutant allowed rA incorporation by accommodating the 2’-OH group of the ribose sugar. Thus Y family polymerases block ribonucleotide incorporation during DNA replication by clashing with the 2’-OH group of ribose sugars. Next, we examined how human DNA polymerase iota (polɩ) prefers to incorporate mis-matched nucleotides opposite an undamaged thymine base. Crystal structures of polɩ in complex with template thymine DNA incorporating either correct adenine (A) or mis-matched thymine (T) or guanine (G) revealed the structural basis of error-prone replication. Correct A was destabilized by a narrow polɩ active site and mis-matched G was preferred by hydrogen bonding with glutamine59 from the finger domain. Domain swapping experiments confirmed the role of the polɩ finger domain in nucleotide selection opposite T. We then investigated how polɩ selects the correct cytosine (C) nucleotide opposite the mutagenic oxidative lesion 8-oxo-guanine. Crystal structures of polɩ in complex with 8-oxo-guanine DNA incorporating correct C or mis-matched A, T, or G revealed the structural basis of error-free replication. The narrow polɩ active site destabilizes A and G purine bases while selecting correct C due to the greatest hydrogen bonding potential with the 8-oxo-guanine Hoogsteen edge. We also show how Glu59 from the finger domain is involved in nucleotide selection and bypass activity through site-directed mutagenesis. Lastly, we examined how polɩ replicates opposite a bulky lesion produced form environmental pollution: N-[deoxyguanosin-8-yl]-1-amino-pyrene (APG). Crystal structures of polɩ in complex with APG DNA incorporating correct C or mis-matched A reveal the structural mechanism of APG replication. Correct C is preferred opposite the lesion due to Watson-Crick base pairing while mis-matched A is incorporated by base stacking above the lesion. We also demonstrate using the model Y family polymerase Dpo4, that the hydrophobic lesion interacts with protein side chains from the little finger domain, which inhibits DNA replication past the lesion site. Taken together, these results further our understanding of how Y family polymerases select incoming nucleotides and how this selection can result in error-free or error-prone replication depending on the chemical nature of the template base.