In vitro analysis of integrated global high-resolution DNA methylation profiling with genomic imbalance and gene expression in osteosarcoma - PubMed (original) (raw)
In vitro analysis of integrated global high-resolution DNA methylation profiling with genomic imbalance and gene expression in osteosarcoma
Bekim Sadikovic et al. PLoS One. 2008.
Abstract
Genetic and epigenetic changes contribute to deregulation of gene expression and development of human cancer. Changes in DNA methylation are key epigenetic factors regulating gene expression and genomic stability. Recent progress in microarray technologies resulted in developments of high resolution platforms for profiling of genetic, epigenetic and gene expression changes. OS is a pediatric bone tumor with characteristically high level of numerical and structural chromosomal changes. Furthermore, little is known about DNA methylation changes in OS. Our objective was to develop an integrative approach for analysis of high-resolution epigenomic, genomic, and gene expression profiles in order to identify functional epi/genomic differences between OS cell lines and normal human osteoblasts. A combination of Affymetrix Promoter Tilling Arrays for DNA methylation, Agilent array-CGH platform for genomic imbalance and Affymetrix Gene 1.0 platform for gene expression analysis was used. As a result, an integrative high-resolution approach for interrogation of genome-wide tumour-specific changes in DNA methylation was developed. This approach was used to provide the first genomic DNA methylation maps, and to identify and validate genes with aberrant DNA methylation in OS cell lines. This first integrative analysis of global cancer-related changes in DNA methylation, genomic imbalance, and gene expression has provided comprehensive evidence of the cumulative roles of epigenetic and genetic mechanisms in deregulation of gene expression networks.
Conflict of interest statement
Competing Interests: The authors have declared that no competing interests exist.
Figures
Figure 1. Integrative epigenetic, genetic, and expression profiling.
(A) Schematic workflow of microarray data analysis and integration. Individual microarrays in replicates (grey boxes), are imported, background corrected, and significantly enriched or depleted regions are detected and assigned to specific genes for DNA methylation, genomic imbalance, and gene expression. All data were analyzed, and integrated using Partek Genomic Suite (PGS) software, and network analysis was performed using Ingenuity Pathway Analysis (IPA). IN – input DNA, IP – immunoprecipitated DNA Cy3 and Cy5 – replicate experiment dye flip for a-CGH labeling, 1 and 2 – individual expression replicates, cor. – background normalized/corrected arrays, HMM – Hidden Markov Model algorithm, gen. seg. – genomic segmentation algorithm, ANOVA – analysis of variance algorithm. (B) Chromosome view of epigenetic, genetic, and expression changes at chromosome 7 in U2OS cells. A PGS generated visualization of the significant regions of genes with significant changes in expression (lane 1 profile), DNA methylation (lane 2 profile), genomic segmentation algorithm results (lane 3 heat map), and the corresponding a-CGH profile (lane 4 profile). The genomic segmentation scale and a-CGH profile values are in log2, while DNA methylation and gene expression y-axis scale represents average fold change to osteoblast levels, and size of each bar is proportional to the fold change (green – decrease, red – increase).
Figure 2. Visualization and validation of Me-DIP-chip detected regions.
Top panel is the PGS-generated region view of the enriched/hypermethylated genomic region (grey box) in the WT1 gene promoter, featuring the colour-coded profile of the signal from each cell type (in log2), and the corresponding heat-map of the replicate array experiments bellow (in log2). Note the reproducibility of the signal, as well as the dendrogram of the hierarchical clustering for the region to the left of the heat maps. Middle panel shows the PGS-generated .wig file of this region imported into UCSC Genome Browser mapped to the WT1 gene promoter, and the corresponding CpG island, allowing identification of genomic features of interest at 35 nucleotide resolution, and design of the primers for the validation experiments including the real-time PCR, and bisulfite EpiTYPER quantitation of DNA methylation. Bottom panel represents the MethPrimer-generated view of the region and the corresponding CpG dinucleotides (red bars) whose methylation levels are quantitated using EpiTYPER.
Figure 3. Gene-specific real-time PCR validation of the Me-DIP-chip data.
Me-DIP-chip detected genes in Table 1 were subject to real-time PCR quantitative analysis of enrichment. The y-axis represents fold enrichment values generated by calculating the ration of U2OS or MG63 IP (Ct)/IN(Ct) over osteoblast IP(Ct)/IN(Ct). Each real-time PCR reaction was performed in triplicate and average values were used for enrichment calculation.
Figure 4. Validation of Me-DIP-chip data using quantitative DNA methylation analysis.
Six genes from Table 1 were subject to EpiTYPER quantitative analysis of DNA methylation in CpG dinucleotides across 400 nucleotide regions detected as significantly enriched/depleted in Me-DIP-chip experiment. On left, the bar charts show levels of methylation (0–1–0–100%), on y-axis and individual CpG dinucleotides on the x-axis, and the corresponding error bars based on triplicate experiment. On right, the PGS-generated region views of the corresponding significantly enriched/depleted genes are labelled as in Figure 3.
Figure 5. VENN analysis of gene-specific epigenetic, genetic, and gene expression changes between U2OS and MG63 cells.
The lists of PGS generated genes with significant changes in DNA methylation, genomic imbalance, and gene expression, in relation to normal osteoblasts are compared using VENN analysis between the U2OS and MG63 cells.
Figure 6. VENN analysis of gene-specific epigenetic, genetic, and gene expression changes in U2OS and MG63.
(A) The lists of PGS generated genes with significant changes in DNA methylation, genomic imbalance, and gene expression, in relation to normal osteoblasts are compared using VENN analysis in U2OS and MG63 cells. (B) Analysis of the two-way intersects for the gain/loss form Figure 6A between DNA methylation and gene expression (left), genomic imbalance and gene expression (middle), and DNA methylation and genomic imbalance (right). y-axis represents the number of genes. (C) Pie chart of the 3-way intersect for the gain/loss changes from Figure 6A for DNA methylation, genomic imbalance, and gene expression.
Figure 7. MYC network-related changes in gene expression, DNA methylation, and genomic imbalance in MG63 cells.
IPA analysis of gene expression, DNA methylation , and genomic imbalance changes in MYC oncogene related pathways. Red denotes gain, and green loss of the corresponding variable.
Figure 8. Validation of array-CGH abnormality calls by metaphase FISH.
A: the copy number abnormality calls identified by array-CGH analysis are shown on the left side of the chromosome 8 ideogram (850-band resolution) for each cell lines. The log2 ratios of a-CGH enrichment detected by genomic segmentation algorithm are represented by a spectrum from green (−3) to red (+3). Arrows indicate the chromosomal localization of the FISH probes. Metaphases from MG-63 (B), and U-2 OS (C), were co-hybridized with the following probes: chromosome 8 centromere (pale blue), RP11-440N18 (8q24.21) (red) and RP11-349C2 (8q24.3) (green).
References
- Hanahan D, Weinberg RA. The hallmarks of cancer. Cell. 2000;100:57–70. - PubMed
- Esteller M. Cancer epigenomics: DNA methylomes and histone-modification maps. Nat Rev Genet. 2007;8:286–298. - PubMed
- Momparler RL, Bovenzi V. DNA methylation and cancer. J Cell Physiol. 2000;183:145–154. - PubMed
- Das PM, Singal R. DNA methylation and cancer. J Clin Oncol. 2004;22:4632–4642. - PubMed
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Medical
Research Materials