Large hypomethylated domains serve as strong repressive machinery for key developmental genes in vertebrates.

DNA methylation is a fundamental epigenetic modification in vertebrate genomes and a small fraction of genomic regions is hypomethylated. Previous studies have implicated hypomethylated regions in gene regulation, but their functions in vertebrate development remain elusive. To address this issue, we generated epigenomic profiles that include base-resolution DNA methylomes and histone modification maps from both pluripotent cells and mature organs of medaka fish and compared the profiles with those of human ES cells. We found that a subset of hypomethylated domains harbor H3K27me3 (K27HMDs) and their size positively correlates with the accumulation of H3K27me3. Large K27HMDs are conserved between medaka and human pluripotent cells and predominantly contain promoters of developmental transcription factor genes. These key genes were found to be under strong transcriptional repression, when compared with other developmental genes with smaller K27HMDs. Furthermore, human-specific K27HMDs show an enrichment of neuronal activity-related genes, which suggests a distinct regulation of these genes in medaka and human. In mature organs, some of the large HMDs become shortened by elevated DNA methylation and associate with sustained gene expression. This study highlights the significance of domain size in epigenetic gene regulation. We propose that large K27HMDs play a crucial role in pluripotent cells by strictly repressing key developmental genes, whereas their shortening consolidates long-term gene expression in adult differentiated cells.

Coordinated changes in DNA methylation in antigen-specific memory CD4 T cells.

Memory CD4(+) T cells are central regulators of both humoral and cellular immune responses. T cell differentiation results in specific changes in chromatin structure and DNA methylation of cytokine genes. Although the methylation status of a limited number of gene loci in T cells has been examined, the genome-wide DNA methylation status of memory CD4(+) T cells remains unexplored. To further elucidate the molecular signature of memory T cells, we conducted methylome and transcriptome analyses of memory CD4(+) T cells generated using T cells from TCR-transgenic mice. The resulting genome-wide DNA methylation profile revealed 1144 differentially methylated regions (DMRs) across the murine genome during the process of T cell differentiation, 552 of which were associated with gene loci. Interestingly, the majority of these DMRs were located in introns. These DMRs included genes such as CXCR6, Tbox21, Chsy1, and Cish, which are associated with cytokine production, homing to bone marrow, and immune responses. Methylation changes in memory T cells exposed to specific Ag appeared to regulate enhancer activity rather than promoter activity of immunologically relevant genes. In addition, methylation profiles differed between memory T cell subsets, demonstrating a link between T cell methylation status and T cell differentiation. By comparing DMRs between naive and Ag-specific memory T cells, this study provides new insights into the functional status of memory T cells.

Genome-wide genetic variations are highly correlated with proximal DNA methylation patterns.

5-methyl-cytosines at CpG sites frequently mutate into thymines, accounting for a large proportion of spontaneous point mutations. The repair system would leave substantial numbers of errors in neighboring regions if the synthesis of erased gaps around deaminated 5-methyl-cytosines is error-prone. Indeed, we identified an unexpected genome-wide role of the CpG methylation state as a major determinant of proximal natural genetic variation. Specifically, 507 Mbp (∼18%) of the human genome was within 10 bp of a CpG site; in these regions, the single nucleotide polymorphism (SNP) rate significantly increased by ∼50% (P < 10(-566) by a two-proportion z-test) if the neighboring CpG sites are methylated. To reconfirm this finding in another vertebrate, we compared six single-base resolution methylomes in two inbred medaka (Oryzias latipes) strains with sufficient genetic divergence (3.4%). We found that the SNP rate also increased by ∼50% (P < 10(-2170)), and the substitution rates in all dinucleotides increased simultaneously (P < 10(-441)) around methylated CpG sites. In the hypomethylated regions, the "CGCG" motif was significantly enriched (P < 10(-680)) and evolutionarily conserved (P = ∼ 0.203%), and slow CpG deamination rather than fast CpG gain was seen, indicating a possible role of CGCG as a candidate cis-element for the hypomethylation state. In regions that were hypermethylated in germline-like tissues but were hypomethylated in somatic liver cells, the SNP rate was significantly smaller than that in hypomethylated regions in both tissues, suggesting a positive selective pressure during DNA methylation reprogramming. This is the first report of findings showing that the CpG methylation state is significantly correlated with the characteristics of evolutionary change in neighboring DNA.

Genome-wide profiling of DNA methylation in human cancer cells.

Global changes in DNA methylation correlate with altered gene expression and genomic instability in cancer. We have developed a methylation-specific digital sequencing (MSDS) method that can assess DNA methylation on a genomic scale. MSDS is a simple, low-cost method that combines the use of methylation-sensitive restriction enzymes with second generation sequencing technology. DNA methylation in two colon cancer cell lines, HT29 and HCT116, was measured using MSDS. When methylation levels were compared between the two cell lines, many differentially methylated regions (DMRs) were identified in CpG island shore regions (located within 2kb of a CpG island), gene body regions and intergenic regions. The number of DMRs in the vicinity of gene transcription start sites correlated with the level of expression of TACC1, CLDN1, and PLEKHC1 (FERMT2) genes, which have been linked to carcinogenesis. The MSDS method has the potential to provide novel insight into the functional complexity of the human genome.

High-resolution analysis of the 5'-end transcriptome using a next generation DNA sequencer.

Massively parallel, tag-based sequencing systems, such as the SOLiD system, hold the promise of revolutionizing the study of whole genome gene expression due to the number of data points that can be generated in a simple and cost-effective manner. We describe the development of a 5'-end transcriptome workflow for the SOLiD system and demonstrate the advantages in sensitivity and dynamic range offered by this tag-based application over traditional approaches for the study of whole genome gene expression. 5'-end transcriptome analysis was used to study whole genome gene expression within a colon cancer cell line, HT-29, treated with the DNA methyltransferase inhibitor, 5-aza-2'-deoxycytidine (5Aza). More than 20 million 25-base 5'-end tags were obtained from untreated and 5Aza-treated cells and matched to sequences within the human genome. Seventy three percent of the mapped unique tags were associated with RefSeq cDNA sequences, corresponding to approximately 14,000 different protein-coding genes in this single cell type. The level of expression of these genes ranged from 0.02 to 4,704 transcripts per cell. The sensitivity of a single sequence run of the SOLiD platform was 100-1,000 fold greater than that observed from 5'end SAGE data generated from the analysis of 70,000 tags obtained by Sanger sequencing. The high-resolution 5'end gene expression profiling presented in this study will not only provide novel insight into the transcriptional machinery but should also serve as a basis for a better understanding of cell biology.