Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)

Graduate Group

Biochemistry & Molecular Biophysics

First Advisor

Rahul M. Kohli

Second Advisor

Benjamin A. Garcia


A tetrad of epigenetic cytosine modifications imbues the DNA code with complex, dynamic meaning. DNA methyltransferase enzymes deposit methyl marks on the 5-carbon of cytosine, forming 5-methylcytosine (mC), which generally mediates long-term, locus-specific transcriptional repression during development and reprogramming. Ten-eleven translocation (TET) family enzymes oxidize the methyl group in three steps, forming predominantly 5-hydroxymethylcytosine (hmC) but also low levels of 5-formylcytosine (fC) and 5-carboxylcytosine (caC). These additional bases likely provide pathways for erasing methylation, but they may also harbor epigenetic functions in their own right. Questions regarding how each base forms and functions drive at the fundamental biology of the epigenome. In this thesis, I chronicle our lab’s efforts to probe the epigenome at its source—by deciphering and manipulating TET enzyme mechanisms. In particular, we aim to understand how and why TET enzymes generate rare fC and caC bases rather than hmC alone. Following a review of the field, I describe in Chapter 2 the methods that we developed to study rare cytosine modifications with high sensitivity. In Chapter 3, we applied these techniques to a rigorous kinetic study of how mouse Tet2 establishes and maintains oxidized cytosines. We found that Tet2 is capable of iterative oxidation, staying on a DNA strand to catalyze multiple rounds of oxidation and thereby enabling efficient generation of fC and caC under certain circumstances. In Chapter 4, we asked what structure-function determinants could allow for the generation of fC and caC. We discovered a conserved active site scaffold in human TET2 that specifically supports the formation of all three oxidized bases, not just hmC. By mutating the active site, we could alter the interactions between key residues to achieve stalling of oxidation at hmC. These mutants have now paved the way for applications in model systems to examine the function of hmC independently of fC and caC, which will allow us to dissect whether the rare, highly oxidized bases are truly critical for epigenetic processes. I describe our progress to date in Chapter 5, along with further mechanistic explorations of the dynamic epigenome.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."