Author ORCID Identifier


Date of Graduation


Document Type

Dissertation (PhD)

Program Affiliation

Epigenetics and Molecular Carcinogenesis

Degree Name

Doctor of Philosophy (PhD)

Advisor/Committee Chair

Taiping Chen, Ph.D.

Committee Member

Mark Bedford, Ph.D.

Committee Member

Rick Finch, Ph.D. DABT

Committee Member

Xiaobing Shi, Ph.D.

Committee Member

Blaine Bartholomew, Ph.D.


DNA methylation (5-methylcytosine, 5mC) is essential for the regulation of gene expression and integrity of the mammalian genome. It occurs predominantly in the context of CpG dinucleotides to form a symmetrical pattern on both DNA strands, which allows DNA methylation patterns to be semi-conservatively maintained during DNA replication. There are two classes of DNA methyltransferases (DNMTs): DNMT3A and DNMT3B function primarily as de novo methyltransferases that establish DNA methylation patterns, whereas DNMT1 is the major enzyme responsible for maintaining DNA methylation patterns by converting hemi-methylated CpGs to fully methylated CpGs during DNA replication. Two accessory factors also play critical regulatory roles: DNMT3L (DNMT3-like) facilitates de novo methylation by DNMT3A/3B, and UHRF1, a 5mC reader, targets DNMT1 to hemi-methylated CpG sites during DNA replication to facilitate maintenance methylation.

The finding that 5mC can be converted to 5-hydroxymethylcytosine (5hmC) and other oxidized derivatives by the ten-eleven translocation (TET) family of dioxygenases represents a breakthrough in the field. Since the discovery, great progress has been made in elucidating the DNA demethylation pathways. In addition to being an intermediate in DNA demethylation, 5hmC may also serve as an epigenetic mark. The discovery of 5hmC raises a fundamental question: how are methylation (5mC) marks maintained at hemi-hydroxymethylated CpG sites during DNA replication? Another important question in the field is how de novo methylation is regulated during embryonic development and in somatic cells, as DNMT3L, an essential co-factor of DNMT3A in germ cells, is quickly silenced during embryonic development and is not expressed in somatic cells. My dissertation aims to address these questions.

In one project, I showed that DNMT3B3, a catalytically inactive DNMT3B isoform due to alternative splicing, preferentially facilitates DNMT3B-mediated DNA methylation. This study was conducted in collaboration with Dr. Xiaodong Cheng’s lab. The Cheng lab provided biochemical and biophysical data showing that DNMT3B3, like DNMT3L, significantly stimulates DNMT3A/3B activity in vitro. I performed rescue experiments in Dnmt3a/3b/3l triple knockout (TKO) mouse embryonic stem cells (mESCs) and demonstrated that DNMT3B3 strongly enhances DNMT3B activity and only weakly enhances DNMT3A activity. These results, together with previous findings that DNMT3B is mainly responsible for the establishment of DNA methylation patterns during embryogenesis and that DNMT3B3 is ubiquitously expressed in differentiating and somatic cells, led us to propose that DNMT3B3 is a critical accessory factor of DNMT3B for de novo methylation in somatic tissues, similar to the role of DNMT3L in assisting DNMT3A in germ cells.

In another project, I showed that DNMT3A and DNMT3B, in addition to carrying out de novo methylation, are also required for faithful ‘maintenance’ of DNA methylation by antagonizing TET-mediated 5mC erosion. Specifically, I provided genetic evidence that the progressive loss of methylation exhibited by Dnmt3a/3b double KO (DKO) mESCs, as reported previously, can be prevented by deletion of the Tet genes (Tet1/2/3). Based on the results, I hypothesized that one or more 5hmC readers target DNMT3A/3B to hemi-hydroxymethylated CpG sites to prevent 5mC erosion. The hypothesis would predict that disruption of the 5hmC reader(s) in mESCs would exhibit progressive hypomethylation, like Dnmt3a/3b DKO mESCs. Candidate approaches by disrupting the putative 5hmC-binding proteins that have been reported, unfortunately, failed to identify such a gene. The project is now being pursued by other personnel in Dr. Taiping Chen’s lab.

In summary, my dissertation projects advance our understanding of the fundamental mechanisms that regulate de novo methylation during embryonic development and in somatic cells and ensure stable and faithful maintenance of DNA methylation patterns in normal developmental and cellular processes.


DNA cytosine methylation, Dnmt3a, Dnmt3b, Dnmt3b3, de novo methylation, DNA methylation maintenance



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.