Analysis and interpretation of epigenomic patterns in colonic epithelia
Barnicle, Alan
Barnicle, Alan
Loading...
Repository DOI
Publication Date
2015-12-18
Type
Thesis
Downloads
Citation
Abstract
The methylation of cytosine nucleotides at the 5' position of DNA is a crucial epigenetic mechanism for the control of gene expression. Epigenome-wide associated studies have demonstrated that specific methylome patterns exist in certain disease states, that methylome signatures can predict cancer susceptibility and that methylation patterns are capable of characterizing epigenomic events that stimulate the survival of cancer cells. Little is known about the effect of chronic inflammation on DNA methylation. In this work, we analysed the effect of chronic inflammation on DNA methylation patterns in ulcerative colitis (UC), a condition that predisposes to colon cancer. Due to the cell type specific nature of DNA methylation patterns, this PhD initially focussed on designing an epithelial cell isolation method that was capable of enriching a purified epithelial population both in diseased and non-diseased states. Secondly, we wished to test the hypothesis that distinct colonic regions have specific methylation signatures, while also comparing DNA methylation patterns in isolated epithelial cells and in whole colonic mucosal biopsies. Finally, using the chronic inflammatory condition of ulcerative colitis (UC) as a pathogenic phenotype, this PhD aimed to identify potential epigenomic and genomic dysregulation that occurs in intestinal epithelial cells in inflamed areas of the colon. Isolation of epithelial cells from mucosal biopsies resulted in purified populations of crypts consisting of ~90% pure epithelium. Using these fractions isolated at 4°C to minimize degradation of nucleic acids, it was demonstrated that stable and integral mRNA and gDNA could successfully be extracted in both diseased and non-diseased states. Using genome-wide technology, specific DNA methylation signatures in the proximal and distal colon in samples from healthy colon were identified. Computational deconvolution was also used to characterize accurate epithelial cell estimates from whole colonic biopsies, while also highlighting the increased DNA methylation variability caused by samples comprised of a mixture of cell lineages. Results from this study identified region-specific epigenomic patterns of HOX genes (specifically HOXB and HOXC families). These patterns represent a valuable tool for interpreting experimental data on diseases that exhibit region-specific expression in the colon such as inflammatory bowel disease and colorectal cancer. By performing genome-wide DNA methylation and transcriptome profiling of purified intestinal epithelial cells, it was demonstrated that significant DNA methylation and gene expression variation occurs in UC. Genes showing inverse correlation between DNA methylation and gene expression levels were also identified, most notably promoter hypermethylation and down regulation of RARB. This gene was previously identified as a tumour suppressor in colorectal adenocarcinoma as well as in breast, lung and prostate cancer. However this is the first finding of RARB potentially playing a functional role in UC. This integrative epigenomic dataset will enhance our understanding of UC pathophysiology, potentially adding to our knowledge of the links between chronic inflammation and carcinogenesis.
Funder
Publisher
Publisher DOI
Rights
Attribution-NonCommercial-NoDerivs 3.0 Ireland