GSBS Dissertations and Theses

Publication Date


Document Type

Doctoral Dissertation

Academic Program

Interdisciplinary Graduate Program


Surgery, Cell Biology, Physiology

First Thesis Advisor

Dr. Shuk-Mei Ho


Gene Expression Regulation, Neoplastic, Gene Silencing, DNA Methylation, Prostatic Neoplasms, Gene Expression Profiling, Neoplasms, Hormone-Dependent


One important objective of prostate cancer (PCa) research is to understand the molecular basis underlying the progression of these cancers from an androgen dependent to an androgen independent state. Hypermethylation of the promoter CpG islands is associated with the transcriptional silencing of specific gene sets in each tumor type and subtype. Transcriptional silencing of antitumor genes via CpG island hypermethylation could be a mechanism mediating PCa progression from an androgen-dependent to an androgen-independent state.

Hypermethylation associated gene silencing has been reported for a great number of genes in PCa with the exception of the genes that undergo methylation associated silencing specifically during cancer development to androgen independence. The first aim of this thesis is to identify novel glenes which undergo DNA hypermethylation associated gene silencing during the cancer progression. The androgen-dependent (AD, as defined as the inability of celill to proliferate in the absence of androgen) PCa cell line LNCaP gives rise to the androgen-independent (AI) subline LNCaPcs generated by maintaining LNCaP in medium with charcoal-stripped (CS) serum for over 30 passages. This LNCaP cell model was used to identify differentially methylated sequences between the two genomes using the Methylation-Sensitive Restriction Fingerprinting (MSRF) technique. One sequence identified is located in a 5' CpG island, which encompasses part of the promoter, exon 1, and part of intron 1, of the Peroxisomal Membrane Protein 24 KD (PMP24) gene. PMP24 is silenced in concert with the hypermethylation of its CpG island in AI LNCaPcsand PC-3 cell lines. The silencing is reactivated by the treatment with a DNA methyltransferase inhibitor, 5-aza-2'-deoxycytidine (5AZAdC). PMP24 is specifically silenced in PCa cancer cell lines and shows potential antitumor properties. These results demonstrate the utility of MSRF in the identification of novel, differentially methylated DNA sequences in the genome and suggest that hypermethylation-mediated silencing of PMP24 is an epigenetic event involved in PCa progression to androgen independence.

The next study investigated the molecular mechanism for DNA methylation associated gene silencing of PMP24 in AI LNCaPcs and PC-3 cell lines. We demonstrated that PMP24 transcription is repressed by the disruption of transcription factor binding to a critical cis-element by hypermethylation of its promoter CpG island. We found a CpG containing activator protein 2 (AP-2) cis-element in the intron 1 of PMP24 whose first CpG dinucleotidle is essential for the sequence-specific protein binding and the promoter activity of the gene. We presented first in cellulo evidence that the methylation of AP-2 cis-element alone but not the whole CpG island, using a newly developed methylated oligonucleotides treatment, is sufficient for the downregulation of PMP24. Our study is the first to report that the silencing mechanism for PMP24 in AI LNCaPcs and PC-3 is mediated by the complete methylation of a single GpG site of AP-2 cis-element in the intron 1 part of the CpG island, which interferes with transcription factor binding. Most interestingly, the promoter CpG island of PMP24 is hypermethylated in AD LNCaP cells with the incomplete methylation specifically at the AP-2 cis-element. The silencing of PMP24 in AD LNCaP cells was reactivated not by the 5AZAdC treatment but by the treatment with Trichostatin A (TSA), a histone deacetylase inhibitor. An alternative silencing mechanism for PMP24 other than the interference with transcription factor binding by methylation is therefore likely involved at this androgen-dependent stage. During the androgen ablation process, this mechanism is either evolved by the spread of methylation in the promoter CpG island or selected against, leading to the methylation-dominant silencing mechanism in the AI cells as seen in LNCaPcsand PC-3 cells.

Taken together, this thesis emphasized the important role of DNA methylation in the progression of PCa into androgen independence. Particular respect should be paid to the specific CpG dinucleotides in cis-elements critical for the promoter activity, whose complete methylation could dominate the silencing mechanism which is independent of androgen. This thesis also pointed to the importance of monitoring the effects of cell culture on the methylation status of genes. Most importantly, this thesis raised the possibility that the silencing mechanisms for PMP24 could be different in AD LNCaP cells as compared to AI LNCaPcs and PC-3 cells. Either the evolution of such mechanism or the selectivity against it during the androgen ablation process would result in a methylation-dominant silencing mechanism of the genes such as PMP24 in AI cells and may contribute to the overall androgen independence of the cells.


Some images did not scan well. Please consult original document.



Rights and Permissions

Copyright is held by the author, with all rights reserved.