UC Davis

The DNA nucleobases are highly susceptible to modification from a wide variety of oxidizing and alkylating agents. Alterations to the nucleobase can threaten genomic integrity by disrupting normal cellular processes including replication and transcription or by introducing mutations. Damage to the DNA base is identified and removed by DNA glycosylases that initiate the base excision repair pathway (BER). The diversity of DNA damage, the location of damage, and how that DNA damage is identified has given valuable insights into disease related to DNA repair. The NEIL family of DNA glycosylases can excise lesions arising from all four nucleobases in a variety of DNA contexts, including duplex, single stranded, and G quadruplex (G4). The loss or dysfunction of NEIL enzymes is associated with a diverse set of disease phenotypes, such as immune deficiencies, anxiety, impaired memory retention, and cancer. Yet, it remains unclear how NEIL dysfunction is related to these phenotypes. In this work, I evaluate the molecular and structural features involved in recognition and excision of a diverse set of lesions across multiple DNA contexts by the NEIL family of glycosylases. First, I examine the features of the damaged nucleobase that influence differences in excision between the two isoforms of NEIL1. RNA editing of the NEIL1 pre-mRNA by the Adenosine Deaminase Acting on RNA (ADAR1) creates two isoforms of NEIL1 via a recoding event that converts a lysine to arginine in the lesion recognition loop of NEIL1. Notably, previous studies have demonstrated that the two isoforms display different enzymatic properties on the pyrimidine lesion, thymine glycol (Tg), where the unedited (UE, K242) isoform showed a significantly faster rate of excision compared to edited NEIL1 (Ed, R242). To further the understanding of NEIL1 activity, I continued this analysis on lesion processing by the two NEIL1 isoforms on a large number of substrates to evaluate the impact of the ADAR1-mediated recoding event. Notably, UE NEIL1 demonstrates better excision of U/T pyrimidines, such as Tg, uracil glycol (Ug), 5-hydroxyuracil (5-OHU), and 5-hydroxymethyluracil (5-hmU), than the edited isoform. However, the relative difference in the rate of excision between Ed and UE NEIL1 is not consistent between lesions and decreases markedly in the series with Tg > Ug > 5-OHU > 5 hmU. Calculations performed in the gas phase examine tautomer stability (2-OH vs 4-OH) and N3 proton affinity of each lesion, and the relative rates of base excision track with the N3 proton affinity of the most stable tautomer. These data suggest that enzyme-promoted tautomerization affects cleavage of the glycosidic bond and enhances excision observed with the unedited enzyme. As a result, the differences in activity between the two isoforms of NEIL1 imply a unique regulatory mechanism for DNA repair by RNA editing. Additionally, the ability of NEIL enzymes to excise DNA damage from G4 structures was evaluated. The G-rich nature of G4 sequences makes them highly susceptible to oxidative damage, and DNA damage in promoter containing G4 sequences and interaction of BER enzymes have been implicated in gene regulation. An oxidation product of guanine, guanidinohydantoin (Gh), was positioned at varying locations in G4 sequences from VEGF, KRAS, and RAD17 promoters, and excision by NEIL1 and mouse NEIL3 (mNeil3), was monitored using in vitro glycosylase assays. The production curves are biphasic providing two rates, indicating that a fraction of Gh within the G4 is excised rapidly, while another fraction is processed more slowly. Also, the percent base removed does not reach 100% despite excess enzyme. The G4 sequence was the largest contributing factor to the differences in rates, associated amplitudes, and overall Gh excision by the NEIL enzymes. These observations suggest that the G4 structure and stability impact accessibility and ability of NEIL1 and mNeil3 to position Gh for cleavage. Thus, reduced repair by NEIL may increase the persistence of mutations or alter gene regulation. In the last chapter, I evaluated the repair of Tg, Gh, and 2,6-diamino-4-hydroxy-5-formamidopyrimidine (FapyG) in human cell lines, as repair can vary between them, using a green fluorescent protein (GFP) reporter that had been previously used to evaluate the repair of Gh and Tg initiated by Neil1 in MEFs. Here, I evaluated the repair of Tg, Gh, and FapyG in HEK293FT cells and the cancer cell lines, HeLa and U87. The extent of repair was the greatest in HEK 293FT cells, while reduced repair of all lesions was observed in HeLa and U87 cells. In comparing the lesions, FapyG was repaired to the greatest extent followed by Gh and Tg across all cell lines. These lesions have never been explored in a cell-based assay, and lesion containing plasmid based cellular assays are useful tool to examine DNA damage and DNA repair capacity as they reflect the influence of factors such as protein expression and stability, which can be altered in disease states. My dissertation evaluates the molecular basis of NEIL initiated repair to provide insights between NEIL activity and its diverse disease phenotypes. These results demonstrate that the type of damage and the location of damage can influence excision by the NEIL DNA glycosylases and altered NEIL activity can impact cellular function and progression of disease. NEIL1 specificity can be modified by RNA editing, and aberrant RNA editing and high levels Ed NEIL1 is associated with cancer. Also, NEIL enzymes interact with G4 structures and reduced excision may lead to increase mutation or participate in regulatory mechanisms. Each chapter examines unique repair properties of the NEIL enzymes and its DNA damage substrates and provides further opportunities to explore repair of these diverse lesions.