Date on Master's Thesis/Doctoral Dissertation


Document Type

Doctoral Dissertation

Degree Name

Ph. D.


Microbiology and Immunology

Degree Program

Microbiology and Immunology, PhD

Committee Chair

Ratajczak, Marius

Committee Co-Chair (if applicable)

Kosiewicz, Michele

Committee Member

Kosiewicz, Michele

Committee Member

Kakar, Sham

Committee Member

Zhang, Huang-Ge

Committee Member

Li, Bing

Author's Keywords

stem cells; cancer; embryonal carcinoma; VSELs; epigenetics; genomic imprinting


Primordial germ cells (PGCs) are hypothesized to deposit hematopoietic stem cells (HSCs) along their migration route through the embryo during the early stages of embryogenesis. PGCs also undergo global chromatin remodeling, including the erasure and reestablishment of genomic imprints, during this migration. While PGCs do not spontaneously form teratomas, their malignant development into germ cell tumors (GCTs) in vivo is often accompanied by the retention of hypomethylation at the IGF2-H19 imprinting control differentially methylated region (DMR). Previous studies in bimaternal embryos determined that proper genomic imprinting at two paternally imprinted loci was necessary for their growth and development: Igf2-H19 and Dlk1-Meg3. Hypomethylation at DMRs within these two loci confers a tumor-suppressing phenotype, thus provoking the question of whether changes in genomic imprinting at these loci may be important for the development of GCTs. Similarly, these loci were recently implicated in the quiescence and maintenance of HSCs, and there is evidence to suggest that both loci are involved in leukemogenesis. Here, I investigated the DLK1-MEG3 locus in acute myeloid leukemia (AML) patient samples, and discovered significant associations between patient survival and the methylation and expression patterns from this locus. In addition, I investigated the methylation of DMRs within the IGF2-H19 and DLK1-MEG3 loci in the human embryonal carcinoma (EC) cell line NTera2 and found that, while the IGF2-H19 control DMR was hypomethylated, the DLK1-MEG3 control DMR and secondary MEG3 DMR were hypermethylated in these cells. The expression ratio of imprinted genes from both loci also agreed with proposed imprinting mechanisms for these phenotypes, and changes in these expression ratios accompanied a decrease in the proliferation rate of these cells during treatment with the DNA methyltransferase inhibitor 5-aza-2’-deoxycytidine. While NTera2 cells functionally responded to exogenous insulin-like growth factors, including IGF2, these cells exhibited strong nuclear staining for DLK1, and shRNA-mediated knockdown of DLK1 revealed a requirement for this gene for the in vitro and in vivo malignant properties of these cells. Furthermore, isolation of potential cancer stem cells (CSCs) from the NTera2 cell line based on CD133 and SSEA4 surface expression produced subpopulations of cells with unique gene expression signatures and migratory characteristics. However, little difference in the DLK1 or OCT4 expression was found among these subpopulations, and the emergence of CD133+SSEA4+ cells from in vitroexpanded CD133-, SSEA4-, and CD133-SSEA4- singly-sorted cells indicated that, while the overall stemness of these cells was fixed, the phenotype of this established cell line is actually in flux. In conclusion, DLK1 is a potential target to treat AML and EC, meriting future investigations into the development of DLK1-targeting therapies, including the use of specific antibodies, aptamers, and vaccination strategies. EC cell growth and metastasis could also be inhibited by employing DNA methyltransferase inhibitors, and investigations into the effect of these drugs on the expression of genes from the DLK1-MEG3 locus in AML could provide valuable information for the development of patient-specific treatments for this disease.