Clare Stirzaker

Identification and resolution of artifacts in bisulfite sequencing

Bisulfite sequencing has become the most widely used application to detect 5-methylcytosine (5-MeC) in DNA, and provides a reliable way of detecting any methylated cytosine at single-molecule resolution in any sequence context. The process of bisulfite treatment exploits the different sensitivity of cytosine and 5-MeC to deamination by bisulfite under acidic conditions, in which cytosine undergoes conversion to uracil while 5-MeC remains unreactive. In this article, we address the more commonly encountered experimental artifacts associated with bisulfite sequencing, and provide methods for the detection and elimination of these artifacts. In particular, we focus on conditions that inhibit complete bisulfite-mediated conversion of cytosines in a target sequence, and demonstrate the necessity of complete protein removal from DNA samples prior to bisulfite treatment. We also include a brief summary of the experimental protocol for bisulfite treatment and tips for designing polymerase chain reaction (PCR) primers to amplify from bisulfite-treated DNA.

Transcriptional Gene Silencing Promotes DNA Hypermethylation through a Sequential Change in Chromatin Modifications in Cancer Cells

It is well established that DNA hypermethylation of tumor suppressor and tumor-related genes can occur in cancer cells and that each cancer subtype has specific gene sets that are commonly susceptible to methylation and silencing. Glutathione S-transferase (GSTP1) is one example of a gene that is hypermethylated and inactivated in the majority of prostate cancers. We previously reported that hypermethylation of the GSTP1 CpG island promoter in prostate cancer cells is initiated by a combination of transcriptional gene silencing (by removal of the Sp1 sites) and seeds of methylation that, instead of being constantly removed because of demethylation associated with transcription, acts as a catalyst for the spread of methylation across the CpG island. In this study, we now demonstrate that the seeds of DNA methylation also play an important role in initiating chromatin modification. Our results address a number of central questions about the temporal relationship between gene expression, DNA hypermethylation, and chromatin modification in cancer cells. We find that for the GSTP1 gene, (a) histone acetylation is independent of gene expression, (b) histone deacetylation is triggered by seeds of DNA methylation, (c) the spread of DNA hypermethylation across the island is linked to MBD2 and not MeCP2 binding, and (d) histone methylation occurs after histone deacetylation and is associated with extensive DNA methylation of the CpG island. These findings have important implications for understanding the biochemical events underlying the mechanisms responsible for abnormal hypermethylation of CpG island-associated genes in cancer cells.

Hypermethylation trigger of the glutathione-S-transferase gene (GSTP1) in prostate cancer cells

Understanding what triggers hypermethylation of tumour suppressor genes in cancer cells is critical if we are to discern the role of methylation in the oncogenic process. CpG sites in CpG island promoters, that span most tumour suppressor genes, remain unmethylated in the normal cell, despite the fact that CpG sites are the prime target for de novo methylation by the DNA methyltransferases. The CpG island-associated with the GSTP1 gene is an intriguing example of a CpG rich region which is susceptible to hypermethylation in the majority of prostate tumours and yet is unmethylated in the normal prostate cell. In this study we evaluate a number of factors purported to be involved in hypermethylation to test their role in triggering hypermethylation of GSTP1 in prostate cancer DU145 and LNCaP cells. We find that hypermethylation is not associated with (1) elevated expression of the DNA methyltranferases, or (2) removal of Sp1 transcription factor binding sites in the CpG island or (3) removal of CpG island boundary elements or (4) prior gene silencing. Instead our results support a model that requires a combination of prior gene silencing and random ‘seeds’ of methylation to trigger hypermethylation of the GSTP1 gene in the prostate cancer cell. We propose that the GSTP1 gene is initially silenced in the prostate cancer and random sites of methylation accumulate that result in subsequent hypermethylation and chromatin remodelling.

Comparison of methyl-DNA immunoprecipitation (MeDIP) and methyl-CpG binding domain (MBD) protein capture for genome-wide DNA methylation analysis reveal CpG sequence coverage bias

DNA methylation primarily occurs at CpG dinucleotides in mammals and is a common epigenetic mark that plays a critical role in the regulation of gene expression. Profiling DNA methylation patterns across the genome is vital to understand DNA methylation changes that occur during development and in disease phenotype. In this study, we compared two commonly used approaches to enrich for methylated DNA regions of the genome, namely methyl-DNA immunoprecipitation (MeDIP) that is based on enrichment with antibodies specific for 5′-methylcytosine (5MeC), and capture of methylated DNA using a methyl-CpG binding domain-based (MBD) protein to discover differentially methylated regions (DMRs) in cancer. The enriched methylated DNA fractions were interrogated on Affymetrix promoter tiling arrays and differentially methylated regions were identified. A detailed validation study of 42 regions was performed using Sequenom MassCLEAVE technique. This detailed analysis revealed that both enrichment techniques are sensitive for detecting DMRs and preferentially identified different CpG rich regions of the prostate cancer genome, with MeDIP commonly enriching for methylated regions with a low CpG density, while MBD capture favors regions of higher CpG density and identifies the greatest proportion of CpG islands. This is the first detailed validation report comparing different methylated DNA enrichment techniques for identifying regions of differential DNA methylation. Our study highlights the importance of understanding the nuances of the methods used for DNA genome-wide methylation analyses so that accurate interpretation of the biology is not overlooked.

Bisulfite sequencing of chromatin immunoprecipitated DNA (BisChIP-seq) directly informs methylation status of histone-modified DNA

The complex relationship between DNA methylation, chromatin modification, and underlying DNA sequence is often difficult to unravel with existing technologies. Here, we describe a novel technique based on high-throughput sequencing of bisulfite-treated chromatin immunoprecipitated DNA (BisChIP-seq), which can directly interrogate genetic and epigenetic processes that occur in normal and diseased cells. Unlike most previous reports based on correlative techniques, we found using direct bisulfite sequencing of Polycomb H3K27me3-enriched DNA from normal and prostate cancer cells that DNA methylation and H3K27me3-marked histones are not always mutually exclusive, but can co-occur in a genomic region-dependent manner. Notably, in cancer, the co-dependency of marks is largely redistributed with an increase of the dual repressive marks at CpG islands and transcription start sites of silent genes. In contrast, there is a loss of DNA methylation in intergenic H3K27me3-marked regions. Allele-specific methylation status derived from the BisChIP-seq data clearly showed that both methylated and unmethylated alleles can simultaneously be associated with H3K27me3 histones, highlighting that DNA methylation status in these regions is not dependent on Polycomb chromatin status. BisChIP-seq is a novel approach that can be widely applied to directly interrogate the genomic relationship between allele-specific DNA methylation, histone modification, or other important epigenetic regulators.

Evaluation of affinity-based genome-wide DNA methylation data: Effects of CpG density, amplification bias, and copy number variation

DNA methylation is an essential epigenetic modification that plays a key role associated with the regulation of gene expression during differentiation, but in disease states such as cancer, the DNA methylation landscape is often deregulated. There are now numerous technologies available to interrogate the DNA methylation status of CpG sites in a targeted or genome-wide fashion, but each method, due to intrinsic biases, potentially interrogates different fractions of the genome. In this study, we compare the affinity-purification of methylated DNA between two popular genome-wide techniques, methylated DNA immunoprecipitation (MeDIP) and methyl-CpG binding domain-based capture (MBDCap), and show that each technique operates in a different domain of the CpG density landscape. We explored the effect of whole-genome amplification and illustrate that it can reduce sensitivity for detecting DNA methylation in GC-rich regions of the genome. By using MBDCap, we compare and contrast microarray- and sequencing-based readouts and highlight the impact that copy number variation (CNV) can make in differential comparisons of methylomes. These studies reveal that the analysis of DNA methylation data and genome coverage is highly dependent on the method employed, and consideration must be made in light of the GC content, the extent of DNA amplification, and the copy number.

Acetylation of H2A.Z is a key epigenetic modification associated with gene deregulation and epigenetic remodeling in cancer

Histone H2A.Z (H2A.Z) is an evolutionarily conserved H2A variant implicated in the regulation of gene expression; however, its role in transcriptional deregulation in cancer remains poorly understood. Using genome-wide studies, we investigated the role of promoter-associated H2A.Z and acetylated H2A.Z (acH2A.Z) in gene deregulation and its relationship with DNA methylation and H3K27me3 in prostate cancer. Our results reconcile the conflicting reports of positive and negative roles for histone H2A.Z and gene expression states. We find that H2A.Z is enriched in a bimodal distribution at nucleosomes, surrounding the transcription start sites (TSSs) of both active and poised gene promoters. In addition, H2A.Z spreads across the entire promoter of inactive genes in a deacetylated state. In contrast, acH2A.Z is only localized at the TSSs of active genes. Gene deregulation in cancer is also associated with a reorganization of acH2A.Z and H2A.Z nucleosome occupancy across the promoter region and TSS of genes. Notably, in cancer cells we find that a gain of acH2A.Z at the TSS occurs with an overall decrease of H2A.Z levels, in concert with oncogene activation. Furthermore, deacetylation of H2A.Z at TSSs is increased with silencing of tumor suppressor genes. We also demonstrate that acH2A.Z anti-correlates with promoter H3K27me3 and DNA methylation. We show for the first time, that acetylation of H2A.Z is a key modification associated with gene activity in normal cells and epigenetic gene deregulation in tumorigenesis.

Mining cancer methylomes: prospects and challenges.

There are over 28 million CpG sites in the human genome. Assessing the methylation status of each of these sites will be required to understand fully the role of DNA methylation in health and disease. Genome-wide analysis, using arrays and high-throughput sequencing, has enabled assessment of large fractions of the methylome, but each protocol comes with unique advantages and disadvantages. Notably, except for whole-genome bisulfite sequencing, most commonly used genome-wide methods detect <5% of all CpG sites. Here, we discuss approaches for methylome studies and compare genome coverage of promoters, genes, and intergenic regions, and capacity to quantitate individual CpG methylation states. Finally, we examine the extent of published cancer methylomes that have been generated using genome-wide approaches.

Methyl-CpG-binding domain proteins: readers of the epigenome

How DNA methylation is interpreted and influences genome regulation remains largely unknown. Proteins of the methyl-CpG-binding domain (MBD) family are primary candidates for the readout of DNA methylation as they recruit chromatin remodelers, histone deacetylases and methylases to methylated DNA associated with gene repression. MBD protein binding requires both functional MBD domains and methyl-CpGs; however, some MBD proteins also bind unmethylated DNA and active regulatory regions via alternative regulatory domains or interaction with the nucleosome remodeling deacetylase (NuRD/Mi-2) complex members. Mutations within MBD domains occur in many diseases, including neurological disorders and cancers, leading to loss of MBD binding specificity to methylated sites and gene deregulation. Here, we summarize the current state of knowledge about MBD proteins and their role as readers of the epigenome.

Epigenetic-induced repression of microRNA-205 is associated with MED1 activation and a poorer prognosis in localized prostate cancer

Deregulation of microRNA (miRNA) expression can have a critical role in carcinogenesis. Here we show in prostate cancer that miRNA-205 (miR-205) transcription is commonly repressed and the MIR-205 locus is hypermethylated. LOC642587, the MIR-205 host gene of unknown function, is also concordantly inactivated. We show that miR-205 targets mediator 1 (MED1, also called TRAP220 and PPARBP) for transcriptional silencing in normal prostate cells, leading to reduction in MED1 mRNA levels, and in total and active phospho-MED1 protein. Overexpression of miR-205 in prostate cancer cells negatively affects cell viability, consistent with a tumor suppressor function. We found that hypermethylation of the MIR-205 locus was strongly related with a decrease in miR-205 expression and an increase in MED1 expression in primary tumor samples (n=14), when compared with matched normal prostate (n=7). An expanded patient cohort (tumor n=149, matched normal n=30) also showed significant MIR-205 DNA methylation in tumors compared with normal, and MIR-205 hypermethylation is significantly associated with biochemical recurrence (hazard ratio=2.005, 95% confidence interval (1.109, 3.625), P=0.02), in patients with low preoperative prostate specific antigen. In summary, these results suggest that miR-205 is an epigenetically regulated tumor suppressor that targets MED1 and may provide a potential biomarker in prostate cancer management.