Unknown,Transcriptomics,Genomics,Proteomics

Dataset Information

Identification and functional analysis of epigenetically silenced microRNAs in colorectal cancer cells

ABSTRACT: Abnormal miRNA expression has been linked to the development and progression of human cancers, and such dysregulation can result from aberrant DNA methylation. We combined the analysis of miRNA expression data deposited with empirical DNA methylation data in HCT116 and DKO colon cancer cells (SRA accession# SRP001414) to identify novel DNA methylation regulated miRNAs. ABSTRACT: Abnormal microRNA (miRNA) expression has been linked to the development and progression of several human cancers, and such dysregulation can result from aberrant DNA methylation. While a small number of miRNAs is known to be regulated by DNA methylation, we postulated that such epigenetic regulation is more prevalent. By combining MBD-isolated Genome Sequencing (MiGS) to evaluate genome-wide DNA methylation patterns and microarray analysis to determine miRNA expression levels, we systematically searched for candidate miRNAs regulated by DNA methylation in colorectal cancer cell lines. We found 64 miRNAs to be robustly methylated in HCT116 cells and/or DNMT-1 and 3B doubleknock cells (DKO); eighteen of them were located in imprinting regions or already reported to be regulated by DNA methylation. In the remaining 46 miRNAs, expression levels of 18 were consistent with their DNA methylation status. Finally, 10 miRNAs were up-regulated by 5-aza-2'-deoxycytidine treatment and identified to be novel miRNAs regulated by DNA methylation. Moreover, we demonstrated the functional relevance of these epigenetically silenced miRNAs by ectopically expressing select candidates, which resulted in inhibition of growth and migration of cancer cells. Our study also provides a reliable strategy to identify DNA methylation-regulated miRNAs by combining DNA methylation profiles and expression data. [miRNA expression]: Total RNA was extracted from 3 biological replicate sets of HCT116 and DMNT-1 and 3B double knock out HCT116 (DKO) colorectal cancer cells.

ORGANISM(S): Homo sapiens

SUBMITTER: Angela Ting

PROVIDER: E-GEOD-26819 | biostudies-arrayexpress |

REPOSITORIES: biostudies-arrayexpress

ACCESS DATA

Similar Datasets

Project description:Epigenetic methyl-CpG silencing of the ribosomal RNA (rRNA) genes is thought to down-regulate rRNA synthesis in mammals. In contrast, we now show that CpG methylation in fact positively influences rRNA synthesis and processing. Human HCT116 cells, inactivated for DNMT1 and DNMT3b or treated with aza-dC, lack CpG methylation and reactivate a large fraction of normally silent rRNA genes. Unexpectedly, these cells display reduced rRNA synthesis and processing, and accumulate unprocessed 45S rRNA. Reactivation of the rRNA genes is associated with their cryptic transcription by RNA polymerase II. Ectopic expression of cryptic rRNA gene transcripts recapitulates the defects associated with loss of CpG methylation. The data demonstrate that rRNA gene silencing prevents cryptic RNA polymerase II transcription of these genes. Lack of silencing leads to the partial disruption of rRNA synthesis and rRNA processing, providing an unexpected explanation for the cytotoxic effects of loss of CpG methylation. Agilent custom microarray analyses of transcripts from sense and antisense strands of the human rRNA genes present in total nuclear RNA from HCT116-DNMT1-/-;DNMT3b-/- (DKO) and HCT116 (PS) cells. The DKO/PS RNA ratio was normalized within the 5â region of the 45S ribosomal RNA, since this sense coding region was found by quantitative Northern and S1-mapping analyses to equally present in the RNA pools from both cells (as compared to total RNA levels and 18S and 28S levels). The distribution for sense RNAs from the coding region is strongly influenced by the predominant RPI 45S rRNA transcript and hence peaks at 0, the DKO and PS 45S rRNA signals having been normalized. 60-mer probes were designed to tile Human rDNA (GenBank accession U13369) on both strands using ArrayOligoSelector (Wardle et al., 2006). A total of 1,146 different non-overlapping probes (573 for each DNA strand) were designed to cover the 43 kb of human rDNA and each printed in triplicate on the array. The experiment was performed using a dye-swap duplicate analysis. The microarrays were scanned on a G2565CA DNA microarray scanner (Agilent Technologies) and the quantification performed using the Feature Extraction software (Agilent technologies). The data were analyzed using the LIMMA software package in R (Smyth and Speed, 2003).

Project description:This is one of expressional parts of the study. These data were correlated to epigenetic changes and CG density of genes in analyzed cells. The whole study has a following summary: To elucidate possible roles of DNA methylation and chromatin marks in transcription, we performed epigenetic profiling of chromosome 19 in human bronchial epithelial cells (HBEC) and in the colorectal cancer cell line HCT116 as well as its counterpart with double knockout of DNMT1 and DNMT3B (HCT116-DKO). We found that H3K9me3 forms intragenic chromatin blocks along genes with low CpG density in the gene body. Analysis of H3K36me3 profiles indicated that this mark associates either with active genes with low CpG density and H3K9me3 in the gene body or with active genes with high CpG density and DNA hypermethylation in the gene body. In HCT116 cells with double knockout of DNMT1 and DNMT3B, transcription of genes with low CpG density in the gene body was highly elevated and associated with promoter DNA demethylation and rearrangement of H3K9me3 and H3K36me3 occupation. Our finding suggests that similar to DNA methylation, H3K9me3 may play a role in intragenic gene regulation. Further, we observed that a combination of low CpG density in gene bodies together with H3K9me3 and H3K36me3 marking is a specific epigenetic feature of zinc finger (ZNF) genes, which comprise 90% of all genes carrying both histone marks on chromosome 19. For high CpG density genes, transcription and H3K36me3 occupancy were not changed in condition of partial or intensive loss of DNA methylation in gene bodies in the HCT116-DKO cell line. siRNA experiments with SETD2 knockdown in both HBEC and HCT116-DKO cell lines failed to reduce DNA methylation in gene bodies under conditions of H3K36me3 depletion. Our study suggests that the H3K36me3 and DNA methylation marks in gene bodies are established independently from each other and points to similar functional roles of intragenic DNA methylation and intragenic H3K9me3 for CpG-rich and CpG-poor genes, respectively. To elucidate how a loss of DNA methylation affects a gene expression and epigenome we studied colon cancer cell line HCT116 and its derivative, HCT116 DNMT1-/- DNMT3b -/- (HCT116-DKO). According to previous studies HCT116-DKO is characterized by 95% of loss of DNA methylation. In HCT116 and HCT116-DKO cells, we profiled diverse epigenetic marks and gene expression. Further a crosstalk between epigenetic changes and transcriptional changes was analyzed. This is the expressional part of study.

Dataset Information

Identification and functional analysis of epigenetically silenced microRNAs in colorectal cancer cells

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets