Text

theses

Fry is an evolutionarily-conserved gene involved in cellular differentiation, division, morphology, polarity, and adhesion. Decreased expression of the FRY protein in breast cancer cell lines reduced the expression of gene networks associated with epithelial differentiation, morphology, polarity and adhesion. Ectopic expression of the wild-type rat Fry gene in human breast cancer cells restored a gene expression profile associated with differentiation, suppression of epithelial-mesenchymal transition. Breast cancer cells expressing wild-type Fry at physiological levels reacquired a non-transformed morphology in vitro and a non-tumorigenic in vivo phenotype in the nude mouse xenograft model. We further determined that expression of FRY mRNA and protein correlated with breast cancer phenotypes such as Elston tumor grade, estrogen receptor, progesterone receptor, and Her2 receptor status. Despite these finding, significant gaps in knowledge remain in our understanding of how FRY expression is decreased during tumor progression. Based on preliminary data, we hypothesized modulation of FRY expression occurs through epigenetic mechanisms which include DNA methylation, histone modification and chromatin remodeling, and the activity/interaction of non-coding RNAs. To understand how DNA methylation regulates FRY transcription in breast tumors, we examined publicly available RNA-seq and DNA methylation data from The Cancer Genome Atlas (TCGA). Analysis of FRY promoter methylation revealed no significant changes in DNA methylation in normal vs. breast tumors. However, when we subdivided the tumors by hormone receptor status, we observed an increase in FRY promoter methylation in ER- and PR- breast tumors. Treatment of various breast cancer cell lines with the DNMT inhibitor, 5-Aza-2-deoxycytidine, resulted in upregulation of FRY mRNA and protein expression. These results indicate that loss of FRY expression may in part be due to increased methylation. To elucidate the role of chromatin in the role of FRY expression we searched publicly available chromatin immunoprecipitation sequencing data from the Encyclopedia of DNA elements (ENCODE) to identify key chromatin markers and transcription factors that bind to the FRY promoter. Using this data, we demonstrated that FRY has a canonical gene promoter characterized by low levels of H3K4me1 and high levels H3K4me2/me3. Furthermore, using data from the ENCODE, we identified 28 transcription factors predicted to binds to the FRY gene promoter, the most significant of which were estrogen receptor 1 (ESR1) and progesterone receptor (PR). To determine if histone deacteylases (HDACs) impacted FRY expression, we dosed various breast cancer cell lines with commonly used the HDAC inhibitors, Trichostatin-A and Panobinostat (LBH589). Through treatment of our cell lines with these HDAC inhibitors, we were able to modulate FRY expression, indicating that its expression is in part regulated by protein acetylation. Next, we explored whether transcriptional repressor, enhancer of Zeste Homolog 2 (EZH2) can bind to the FRY promoter to down regulate its expression. Using Chromosome Immunoprecipitation DNA sequencing (ChIP-seq) data from the ENCODE database, we identified that EZH2 binds to the FRY promoter in both normal human mammary epithelial cells (HMECs) and in the breast cancer derived MCF-7 cell line. Furthermore, when we treated cells with EZH2 inhibitors, GSK343 and DZNep, we induce the expression of FRY mRNA and protein in the MDA-MB231, MDA-MB-468, MCF-7, T47-D and HCC1954 cell lines. Our results suggest that elevated expression of EZH2 contributes to decreased FRY expression in breast cancer cells. One unique feature of the FRY gene promoter is that it also encodes an antisense long coding RNA (lncRNA), FRY-AS1. Many lncRNAs have been shown to be biologically active and have been implicated various cellular processes such as transcriptional interference, the induction of chromatin remodeling, and modulation of protein activity. To date, no studies have focused on the functional characterization of FRY-AS1. In this present study, we analyzed publicly available RNA-Seq data. We found that FRY-AS1 is differentially expressed in a variety of human tissue, and that FRY mRNA expression is strongly correlated with FRY-AS1 expression. Next, using RNA-seq data from TCGA BRCA dataset, we found that FRY-AS1 expression was significantly decreased in basal-like tumors, and in estrogen negative, and progesterone negative tumors. In vitro studies indicated that FRY-AS1 was decreased in 4 of 5 cell lines examined relative to the levels non-tumorigenic mammary epithelial cell line. Cellular fractionation studies revealed that FRY-AS1 is primarily localized to the nucleus. Ectopic over-expression of FRY-AS1 in the MDA-MB-231 induced differentiation to a more differentiated epithelial phenotype in vitro and induced the formation the formation acinar structures in 3D-Matrigel cultures. Significantly, overexpression of FRY-AS1 also increased endogenous FRY expression, indicating that the lncRNA regulates FRY expression in trans. Lastly, we treated the MDA-MB-231, MDA-MB-468, MCF-7, HCC1954, and T47-D cell lines with HDAC, DNMT, or EZH2 inhibitors. In these studies, we were able to induce FRY-AS1 expression in all cancer cell lines examined, which indicates that the expression of this lncRNA is regulated in part through multiple epigenetic mechanisms. Taken together, our results demonstrate that FRY-AS1 is a bona-fide lncRNA with functional significance in regulation of the FRY carcinoma susceptibility gene, and hence may be a target therapeutic intervention.

ETD_8697

Ph.D.

Includes bibliographical references

by Brian Estrella

doi:10.7282/T3KK9G7K

ETD doctoral

The author owns the copyright to this work.