Cooperativity between DNA methyltransferases in the maintenance methylation of repetitive elements

A population-epigenetic model to infer site-specific methylation rates from double-stranded DNA methylation patterns

Cytosine methylation is an epigenetic mechanism in eukaryotes that is often associated with stable transcriptional silencing, such as in X-chromosome inactivation and genomic imprinting. Aberrant methylation patterns occur in several inherited human diseases and in many cancers. To understand how methylated and unmethylated states of cytosine residues are transmitted during DNA replication, we develop a population-epigenetic model of DNA methylation dynamics. The model is informed by our observation that de novo methylation can occur on the daughter strand while leaving the opposing cytosine unmethylated, as revealed by the patterns of methylation on the two complementary strands of individual DNA molecules. Under our model, we can infer site-specific rates of both maintenance and de novo methylation, values that determine the fidelity of methylation inheritance, from double-stranded methylation data. This approach can be used for populations of cells obtained from individuals without the need for cell culture. We use our method to infer cytosine methylation rates at several sites within the promoter of the human gene FMR1.

Homology-dependent methylation in primate repetitive DNA

In mammals, several studies have suggested that levels of methylation are higher in repetitive DNA than in nonrepetitive DNA, possibly reflecting a genome-wide defense mechanism against deleterious effects associated with transposable elements (TEs). To analyze the determinants of methylation patterns in primate repetitive DNA, we took advantage of the fact that the methylation rate in the germ line is reflected by the transition rate at CpG sites. We assessed the variability of CpG substitution rates in nonrepetitive DNA and in various TE and retropseudogene families. We show that, unlike other substitution rates, the rate of transition at CpG sites is significantly (37%) higher in repetitive DNA than in nonrepetitive DNA. Moreover, this rate of CpG transition varies according to the number of repeats, their length, and their level of divergence from the ancestral sequence (up to 2.7 times higher in long, lowly divergent TEs compared with unique sequences). This observation strongly suggests the existence of a homology-dependent methylation (HDM) mechanism in mammalian genomes. We propose that HDM is a direct consequence of interfering RNA-induced transcriptional gene silencing.

DNA methylation profiles of CpG islands for cellular differentiation and development in mammals

DNA methylation has been implicated in mammalian development. Transcription units contain CpG islands, but expression of CpG island associated genes in normal tissues was not believed to be controlled by DNA methylation. There are, however, numerous CpG islands containing tissue-dependent and differentially methylated regions (T-DMR), which are potential methylation sites in normal cells and tissues. Genomic scanning which focused on T-DMRs in CpG islands revealed that the DNA methylation profile of each cell/tissue is more complicated than previously considered. Differentiation of cells is associated with both methylation and demethylation, which occur at multiple loci. The epigenetic system characterized by DNA methylation requires cells to memorize gene expression patterns, thus, standardizing cellular phenotypes.

The de novo methylation activity of Dnmt3a is distinctly different than that of Dnmt1

Background

Though Dnmt1 is considered the primary maintenance methyltransferase and Dnmt3a and Dnmt3b are considered de novo methyltransferases in mammals, these three enzymes may work together in maintaining as well as establishing DNA methylation patterns. It has been proposed that Dnmt1 may carry out de novo methylation at sites in the genome with transient single-stranded regions, such as replication origins, and then spread methylation from these nucleation sites in vivo, even though such activity has not been reported.

Results

In this study, we show that Dnmt3a does not act on single-stranded substrates in vitro, indicating that Dnmt3a is not likely to initiate DNA methylation at such proposed nucleation sites. Dnmt3a shows similar methylation activity on unmethylated and hemimethylated duplex DNA, though with some substrate preference. Unlike Dnmt1, pre-existing cytosine methylation at CpG sites or non-CpG sites does not stimulate Dnmt3a activity in vitro and in vivo.

Conclusion

The fact that Dnmt3a does not act on single stranded DNA and is not stimulated by pre-existing cytosine methylation indicates that the de novo methylation activity of Dnmt3a is quite different from that of Dnmt1. These findings are consistent with a model in which Dnmt3a initiates methylation on one of the DNA strands of duplex DNA, and these hemimethylated sites then stimulate Dnmt1 activity for further methylation.

Epigenetics and cancer

Both genetics and epigenetics regulate gene expression in cancer. Regulation by genetics involves a change in the DNA sequence, whereas epigenetic regulation involves alteration in chromatin structure and methylation of the promoter region. During the initiation, development, and progression of cancer, a number of genes undergo epigenetic changes. Some of these changes can be used as biomarkers for early detection of cancer as well as to follow treatment. A panel of epigenetic biomarkers is preferred to a single biomarker in clinical assays. Changes in gene expression due to epigenetic regulation can be reversed by chemicals, and this approach opens up a novel approach in cancer prevention and treatment.

Therapeutic implications of DNA methylation

Cancer growth and metastasis requires reprogramming of the expression of multiple genes. The epigenome, which is comprised of chromatin and the patterns of DNA methylation, sets up and maintains gene expression programs. As expected from the broad changes in gene expression in cancer, which are characterized by both silencing and activation of multiple genes, the epigenome of cancer cells is distinguished by aberration of DNA methylation patterns, which include both hypo- and hypermethylation and aberrant regulation of DNA methylation enzymes. In contrast to genetic alterations, which are fixed and are not amenable to therapeutic intervention, pharmacological agents could alter DNA methylation patterns. This raises the prospect that DNA methylation-targeted drugs will reverse cancer growth and metastasis. One of the main challenges however, is to understand the relative role of hypo- and hypermethylation in order to achieve a balance of epigenetic therapeutic agents with positive outcome and reduced adverse effects.

Dynamic expression of de novo DNA methyltransferases Dnmt3a and Dnmt3b in the central nervous system

To explore the role of DNA methylation in the brain, we examined the expression pattern of de novo DNA methyltransferases Dnmt3a and Dnmt3b in the mouse central nervous system (CNS). By comparing the levels of Dnmt3a and Dnmt3b mRNAs and proteins in the CNS, we showed that Dnmt3b is detected within a narrow window during early neurogenesis, whereas Dnmt3a is present in both embryonic and postnatal CNS tissues. To determine the precise pattern of Dnmt3a and Dnmt3b gene expression, we carried out X-gal histochemistry in transgenic mice in which the lacZ marker gene is knocked into the endogenous Dnmt3a or Dnmt3b gene locus (Okano et al. [1999] Cell 99:247-257). In Dnmt3b-lacZ transgenic mice, X-gal-positive cells are dispersed across the ventricular zone of the CNS between embryonic days (E) 10.5 and 13.5 but become virtually undetectable in the CNS after E15.5. In Dnmt3a-lacZ mice, X-gal signal is initially observed primarily in neural precursor cells within the ventricular and subventricular zones between E10.5 and E17.5. However, from the newborn stage to adulthood, Dnmt3a X-gal signal was detected predominantly in postmitotic CNS neurons across all the regions examined, including olfactory bulb, cortex, hippocampus, striatum, and cerebellum. Furthermore, Dnmt3a signals in CNS neurons increase during the first 3 weeks of postnatal development and then decline to a relatively low level in adulthood, suggesting that Dnmt3a may be of critical importance for CNS maturation. Immunocytochemistry experiments confirmed that Dnmt3a protein is strongly expressed in neural precursor cells, postmitotic CNS neurons, and oligodendrocytes. In contrast, glial fibrillary acidic protein-positive astrocytes exhibit relatively weak or no Dnmt3a immunoreactivity in vitro and in vivo. Our data suggest that whereas Dnmt3b may be important for the early phase of neurogenesis, Dnmt3a likely plays a dual role in regulating neurogenesis prenatally and CNS maturation and function postnatally.

Epigenetic regulation of mammalian imprinted genes: from primary to functional imprints

Parental genomic imprinting was discovered in mammals some 20 years ago. This phenomenon, crucial for normal development, rapidly became a key to understanding epigenetic regulation of mammalian gene expression. In this chapter we present a general overview of the field and describe in detail the 'imprinting cycle'. We provide selected examples that recapitulate our current knowledge of epigenetic regulation at imprinted loci. These epigenetic mechanisms lead to the stable repression of imprinted genes on one parental allele by interfering with 'formatting' for gene expression that usually occurs on expressed alleles. From this perspective, genomic imprinting remarkably illustrates the complexity of the epigenetic mechanisms involved in the control of gene expression in mammals.

Conserved methylation patterns of human papillomavirus type 16 DNA in asymptomatic infection and cervical neoplasia

DNA methylation contributes to the chromatin conformation that represses transcription of human papillomavirus type16 (HPV-16), which is prevalent in the etiology of cervical carcinoma. In an effort to clarify the role of this phenomenon in the regulation and carcinogenicity of HPV-16, 115 clinical samples were studied to establish the methylation patterns of the 19 CpG dinucleotides within the long control region and part of the L1 gene by bisulfite modification, PCR amplification, DNA cloning, and sequencing. We observed major heterogeneities between clones from different samples as well as between clones from individual samples. The methylation frequency of CpGs was measured at 14.5%. In addition, 0.21 and 0.23%, respectively, of the CpA and CpT sites, indicators of de novo methylation, were methylated. Methylation frequencies exceeded 30% in the CpGs overlapping with the L1 gene and were about 10% for most other positions. A CpG site located in the linker between two nucleosomes positioned over the enhancer and promoter of HPV-16 had minimal methylation. This region forms part of the HPV replication origin and is close to binding sites of master-regulators of transcription during epithelial differentiation. Methylation of most sites was highest in carcinomas, possibly due to tandem repetition and chromosomal integration of HPV-16 DNA. Methylation was lowest in dysplasia, likely reflecting the transcriptional activity in these infections. Our data document the efficient targeting of HPV genomes by the epithelial methylation machinery, possibly as a cellular defense mechanism, and suggest involvement of methylation in HPV oncogene expression and the early-late switch.

Sex- and tissue-specific expression of maintenance and de novo DNA methyltransferases upon low dose X-irradiation in mice

DNA methylation is crucial for normal development, proliferation, and proper maintenance of genome stability for a given organism. A variety of DNA damaging agents that are known to affect genome stability were also shown to alter DNA methylation patterns. We have recently pioneered the studies in the area of the radiation effects on DNA methylation, and found that radiation exposure led to substantial dose-dependent and tissue-specific DNA hypomethylation, which was much more pronounced in spleen and liver of female animals. The exact mechanisms of radiation-induced DNA hypomethylation are still to be uncovered. We have previously shown that one of those mechanisms may potentially be DNA repair related. Another possible mechanism may be linked to changes in the expression of DNA methyltransferases (DNMTs). In the current study, we examined the radiation-induced changes in expression of maintenance DNMT1, and de novo methyltransferases DNMT3a and DNMT3b in spleen and liver of irradiated animals. This was paralleled by the studies of acute and chronic IR-induced methylation changes in spleen and liver of intact animals, as well as in animals with altered sex hormone status. Here we report that radiation-induced DNA methylation changes correlated with radiation-induced alterations in expression of DNA methyltransferases. We present the data on tissue-specificity in radiation-induced expression of DNA methyltransferases, and prove that changes in the expression of de novo methyltransferases DNMT3a and DNMT3b are the most important in radiation-induced DNA methylation alterations. We also discuss the role of sex hormones, especially estrogen, in the generation of the sex-specific radiation-induced methylation changes.

Molecular dissection of the events leading to inactivation of the FMR1 gene

The analysis of a lymphoblastoid cell line (5106), derived from a rare individual of normal intelligence with an unmethylated full mutation of the FMR1 gene, allowed us to reconstruct the chain of molecular events leading to the FMR1 inactivation and to fragile X syndrome. We found that lack of DNA methylation of the entire promoter region, including the expanded CGG repeat, correlates with methylation of lysine 4 residue on the N-tail of histone H3 (H3-K4), as in normal controls. Normal levels of FMR1 mRNA were detected by real-time fluorescent RT-PCR (0.8-1.4 times compared with a control sample), but mRNA translation was less efficient (-40%), as judged by polysome profiling, resulting in reduced levels of FMRP protein (approximately 30% of a normal control). These results underline once more that CGG repeat amplification per se does not prevent FMR1 transcription and FMRP production in the absence of DNA methylation. Surprisingly, we found by chromatin immunoprecipitation that cell line 5106 has deacetylated histones H3 and H4 as well as methylated lysine 9 on histone H3 (H3-K9), like fragile X cell lines, in both the promoter and exon 1. This indicates that these two epigenetic marks (i.e. histone deacetylation and H3-K9 methylation) can be established in the absence of DNA methylation and do not interfere with active gene transcription, contrary to expectation. Our results also suggest that the molecular pathways regulating DNA and H3-K4 methylation are independent from those regulating histone acetylation and H3-K9 methylation.

The PWWP domain of Dnmt3a and Dnmt3b is required for directing DNA methylation to the major satellite repeats at pericentric heterochromatin

Dnmt3a and Dnmt3b are responsible for the establishment of DNA methylation patterns during development. These proteins contain, in addition to a C-terminal catalytic domain, a unique N-terminal regulatory region that harbors conserved domains, including a PWWP domain. The PWWP domain, characterized by the presence of a highly conserved proline-tryptophan-tryptophan-proline motif, is a module of 100 to 150 amino acids found in many chromatin-associated proteins. However, the function of the PWWP domain remains largely unknown. In this study, we provide evidence that the PWWP domains of Dnmt3a and Dnmt3b are involved in functional specialization of these enzymes. We show that both endogenous and green fluorescent protein-tagged Dnmt3a and Dnmt3b are particularly concentrated in pericentric heterochromatin. Mutagenesis analysis indicates that their PWWP domains are required for their association with pericentric heterochromatin. Disruption of the PWWP domain abolishes the ability of Dnmt3a and Dnmt3b to methylate the major satellite repeats at pericentric heterochromatin. Furthermore, we demonstrate that the Dnmt3a PWWP domain has little DNA-binding ability, in contrast to the Dnmt3b PWWP domain, which binds DNA nonspecifically. Collectively, our results suggest that the PWWP domains of Dnmt3a and Dnmt3b are essential for targeting these enzymes to pericentric heterochromatin, probably via a mechanism other than protein-DNA interactions.

DNA methylation and breast cancer

DNA methylation and chromatin structure patterns are tightly linked components of the epigenome, which regulate gene expression programming. Two contradictory changes in DNA methylation patterns are observed in breast cancer; regional hypermethylation of specific genes and global hypomethylation. It is proposed here that independent mechanisms are responsible for these alterations in DNA methylation patterns and that these alterations deregulate two different processes in breast cancer. Regional hypermethylation is brought about by specific regional changes in chromatin structure, whereas global demethylation is caused by a general increase in demethylation activity. Hypermethylation silences growth regulatory genes resulting in uncontrolled growth whereas hypomethylation leads to activation of genes required for metastasis. DNA methylation inhibitors activate silenced tumor suppressor genes resulting in arrest of tumor growth and are now being tested as candidate anticancer drugs. Demethylation inhibitors are proposed here to be potential novel candidate antimetastatic agents, which would bring about methylation and silencing of metastatic genes. Future therapeutic application of either methylation or demethylation inhibitors in cancer therapy would require understanding of the relative role of these processes in the evolution of cancer.

DNA demethylation and cancer: therapeutic implications

The epigenome, which is comprised of chromatin and its associated proteins and the patterns of covalent modification of DNA by methylation, sets up and maintains gene expression programs. A hallmark of cancer is a paradoxical aberration of DNA methylation patterns, a global loss of DNA methylation, that coexists with regional hypermethylation of certain genes. The hypermethylation of tumor-suppressor genes has attracted significant attention recently and DNA methylation inhibitors are being tested as potential anticancer agents. However, emerging data suggests that hypomethylation plays a role in activating genes required for metastasis and invasion. It is proposed here that hypermethylation and hypomethylation in cancer are independent processes, which target different programs at different stages in tumorigenesis. Understanding the relative roles of hypomethylation and hypermethylation in cancer has clear implications on the therapeutic use of agents targeting the DNA methylation machinery, which are discussed in this review.

Preference of DNA methyltransferases for CpG islands in mouse embryonic stem cells

Many CpG islands have tissue-dependent and differentially methylated regions (T-DMRs) in normal cells and tissues. To elucidate how DNA methyltransferases (Dnmts) participate in methylation of the genomic components, we investigated the genome-wide DNA methylation pattern of the T-DMRs with Dnmt1-, Dnmt3a-, and/or Dnmt3b-deficient ES cells by restriction landmark genomic scanning (RLGS). Approximately 1300 spots were detected in wild-type ES cells. In Dnmt1(-/-) ES cells, additional 236 spots emerged, indicating that the corresponding loci are methylated by Dnmt1 in wild-type ES cells. Intriguingly, in Dnmt3a(-/-)Dnmt3b(-/-) ES cells, the same 236 spots also emerged, and no additional spots appeared differentially. Therefore, Dnmt1 and Dnmt3a/3b share targets in CpG islands. Cloning and virtual image RLGS revealed that 81% of the RLGS spots were associated with genes, and 62% of the loci were in CpG islands. By contrast to the previous reports that demethylation at repeated sequences was severe in Dnmt1(-/-) cells compared with Dnmt3a(-/-)Dnmt3b(-/-) cells, a complete loss of methylation was observed at RLGS loci in Dnmt3a(-/-)Dnmt3b(-/-) cells, whereas methylation levels only decreased to 16% to 48% in the Dnmt1(-/-) cells. We concluded that there are CpG islands with T-DMR as targets shared by Dnmt1 and Dnmt3a/3b and that each Dnmt has target preferences depending on the genomic components.

Human DNA methyltransferase 1 is required for maintenance of the histone H3 modification pattern

DNA methyltransferase 1 (DNMT1) plays an essential role in murine development and is thought to be the enzyme primarily responsible for maintenance of the global methylation status of genomic DNA. However, loss of DNMT1 in human cancer cells affects only the methylation status of a limited number of pericentromeric sequences. Here we show that human cancer cells lacking DNMT1 display at least two important differences with respect to wild type cells: a profound disorganization of nuclear architecture, and an altered pattern of histone H3 modification that results in an increase in the acetylation and a decrease in the dimethylation and trimethylation of lysine 9. Additionally, this phenotype is associated with a loss of interaction of histone deacetylases (HDACs) and HP1 (heterochromatin protein 1) with histone H3 and pericentromeric repetitive sequences (satellite 2). Our data indicate that DNMT1 activity, via maintenance of the appropriate histone H3 modifications, contributes to the preservation of the correct organization of large heterochromatic regions.

Genomic hypomethylation is specific for preneoplastic liver in folate/methyl deficient rats and does not occur in non-target tissues

Chronic dietary insufficiency of the lipotropic nutrients choline and methionine is hepatocarcinogenic in male rats and certain mouse strains. Despite the fact that DNA hypomethylation is a hallmark of most cancer genomes, the tissue-specific consequences of this alternation with respect to tumorigenesis remain to be determined. In the present study, the folate/methyl deficient model of multistage hepatocarcinogenesis was used to evaluate in vivo alterations in DNA methylation in the liver, the carcinogenesis target tissue, and in non-target tissues, including pancreas, spleen, kidney, and thymus, of male F344 rats. By utilizing the HpaII/MspI-based cytosine extension assay, we demonstrated that the percent of CpG sites that lost methyl groups on both strands progressively increased in liver tissue after 9, 18, and 36 weeks of folate/methyl deficiency. The endogenous activity of DNA methyltransferase in liver of rats fed with folate/methyl deficient diet for the 36-week period gradually increased with time. In contrast, non-target tissues displayed no changes in DNA methylation level or activity of DNA methyltransferase. The failure of DNA methyltransferase to restore and maintain DNA methylation patterns in preneoplastic liver tissue may lead to the establishment of tumor-specific DNA methylation and DNA methyltransferase profiles that are not expressed in normal liver. These results provide additional information about alterations in DNA methylation during early preneoplastic stages of carcinogenesis. They also demonstrate that DNA hypomethylation is localized to tissue that undergoes carcinogenesis, and is not altered in non-target tissues.

Genetic analysis of RNA-mediated transcriptional gene silencing

The 'nuclear side' of RNA interference (RNAi) is increasingly recognized as an integral part of RNA-mediated gene silencing networks. Current data are consistent with the idea that epigenetic changes, such as DNA (cytosine-5) methylation and histone modifications, can be targeted to identical DNA sequences by short RNAs derived via Dicer cleavage of double-stranded RNA (dsRNA). To determine the relationships among RNA signals, DNA methylation and chromatin structure, we are carrying out a genetic analysis of RNA-mediated transcriptional gene silencing (TGS) in Arabidopsis. Results obtained so far indicate that in response to RNA signals, different site-specific DNA methyltransferases (DMTases) cooperate with each other and eventually with histone-modifying enzymes to establish and maintain a transcriptionally inactive state at a homologous target promoter. Processing of dsRNA in Arabidopsis occurs in the nucleus and in the cytoplasm, where distinct Dicer-like (DCL) activities are thought to generate functionally distinct classes of short RNAs. RNA silencing pathways thus operate throughout the cell to defend against invasive nucleic acids and to regulate genome structure and function.

Distinct localization of histone H3 acetylation and H3-K4 methylation to the transcription start sites in the human genome

Almost 1-2% of the human genome is located within 500 bp of either side of a transcription initiation site, whereas a far larger proportion (approximately 25%) is potentially transcribable by elongating RNA polymerases. This observation raises the question of how the genome is packaged into chromatin to allow start sites to be recognized by the regulatory machinery at the same time as transcription initiation, but not elongation, is blocked in the 25% of intragenic DNA. We developed a chromatin scanning technique called ChAP, coupling the chromatin immunoprecipitation assay with arbitrarily primed PCR, which allows for the rapid and unbiased comparison of histone modification patterns within the eukaryotic nucleus. Methylated lysine 4 (K4) and acetylated K9/14 of histone H3 were both highly localized to the 5' regions of transcriptionally active human genes but were greatly decreased downstream of the start sites. Our results suggest that the large transcribed regions of human genes are maintained in a deacetylated conformation in regions read by elongating polymerase. Common models depicting widespread histone acetylation and K4 methylation throughout the transcribed unit do not therefore apply to the majority of human genes.

The role of MET1 in RNA-directed de novo and maintenance methylation of CG dinucleotides

A genetic screen for mutants defective in RNA-directed DNA methylation and transcriptional silencing of the constitutive nopaline synthase (NOS) promoter in Arabidopsis identified two independent mutations in the gene encoding the DNA methyltransferase MET1. Both mutant alleles are disrupted structurally in the MET1 catalytic domain, suggesting that they are complete loss of function alleles. Experiments designed to test the effect of a met1 mutation on both RNA-directed de novo and maintenance methylation of the target NOS promoter revealed in each case approximately wild type levels of non-CG methylation together with significant reductions of CG methylation. These results confirm a requirement for MET1 to maintain CG methylation induced by RNA. In addition, the failure to establish full CG methylation in met1 mutants, despite normal RNA-directed de novo methylation of Cs in other sequence contexts, indicates that MET1 is required for full de novo methylation of CG dinucleotides. We discuss MET1 as a site-specific DNA methyltransferase that is able to maintain CG methylation during DNA replication and contribute to CG de novo methylation in response to RNA signals.

Continuous zebularine treatment effectively sustains demethylation in human bladder cancer cells

During tumorigenesis, tumor suppressor and cancer-related genes are commonly silenced by aberrant DNA methylation in their promoter regions. Recently, we reported that zebularine [1-(beta-D-ribofuranosyl)-1,2-dihydropyrimidin-2-one] acts as an inhibitor of DNA methylation and exhibits chemical stability and minimal cytotoxicity both in vitro and in vivo. Here we show that continuous application of zebularine to T24 cells induces and maintains p16 gene expression and sustains demethylation of the 5' region for over 40 days, preventing remethylation. In addition, continuous zebularine treatment effectively and globally demethylated various hypermethylated regions, especially CpG-poor regions. The drug caused a complete depletion of extractable DNA methyltransferase 1 (DNMT1) and partial depletion of DNMT3a and DNMT3b3. Last, sequential treatment with 5-aza-2'-deoxycytidine followed by zebularine hindered the remethylation of the p16 5' region and gene resilencing, suggesting the possible combination use of both drugs as a potential anticancer regimen.

Overall DNA methylation and chromatin structure of normal and abnormal X chromosomes

DNA methylation patterns were studied at the chromosome level in normal and abnormal X chromosomes using an anti-5-methylcytosine antibody. In man, except for the late-replicating X of female cells, the labeled chromosome structures correspond to R- and T-bands and heterochromatin. Depending on the cell type, the species, and cell culture conditions, the late-replicating X in female cells appears to be more or less undermethylated. Under normal conditions, the only structures that remain methylated on the X chromosomes correspond to pseudoautosomal regions, which harbor active genes. Thus, active genes are usually hypomethylated but are located in methylated chromatin. Structural rearrangements of the X chromosome, such as t(X;X)(pter;pter), induce a Turner syndrome-like phenotype that is inconsistent with the resulting triple-X constitution. This suggests a position effect controlling gene inactivation. The derivative chromosomes are always late replicating, and their duplicated short arms, which harbor pseudoautosomal regions, replicate later than the normal late-replicating X chromosomes. The compaction or condensation of this segment is unusual, with a halo of chromatin surrounding a hypocondensed chromosome core. The chromosome core is hypomethylated, but the surrounding chromatin is slightly labeled. Thus, unusual DNA methylation and chromatin condensation are associated with the observed position effect. This strengthens the hypothesis that DNA methylation at the chromosome level is associated with both chromatin structure and gene expression.

Hairpin-bisulfite PCR: assessing epigenetic methylation patterns on complementary strands of individual DNA molecules

Epigenetic inheritance, the transmission of gene expression states from parent to daughter cells, often involves methylation of DNA. In eukaryotes, cytosine methylation is a frequent component of epigenetic mechanisms. Failure to transmit faithfully a methylated or an unmethylated state of cytosine can lead to altered phenotypes in plants and animals. A central unresolved question in epigenetics concerns the mechanisms by which a locus maintains, or changes, its state of cytosine methylation. We developed "hairpin-bisulfite PCR" to analyze these mechanisms. This method reveals the extent of methylation symmetry between the complementary strands of individual DNA molecules. Using hairpin-bisulfite PCR, we determined the fidelity of methylation transmission in the CpG island of the FMR1 gene in human lymphocytes. For the hypermethylated CpG island of this gene, characteristic of inactive-X alleles, we estimate a maintenance methylation efficiency of approximately 0.96 per site per cell division. For de novo methylation efficiency (E(d)), remarkably different estimates were obtained for the hypermethylated CpG island (E(d) = 0.17), compared with the hypomethylated island on the active-X chromosome (E(d) < 0.01). These results clarify the mechanisms by which the alternative hypomethylated and hypermethylated states of CpG islands are stably maintained through many cell divisions. We also analyzed a region of human L1 transposable elements. These L1 data provide accurate methylation patterns for the complementary strand of each repeat sequence analyzed. Hairpin-bisulfite PCR will be a powerful tool in studying other processes for which genetic or epigenetic information differs on the two complementary strands of DNA.

Role of the DNA Methyltransferase Variant DNMT3b3 in DNA Methylation

Several alternatively spliced variants of DNA methyltransferase (DNMT) 3b have been described. Here, we identified new murine Dnmt3b mRNA isoforms and found that mouse embryonic stem (ES) cells expressed only Dnmt3b transcripts that contained exons 10 and 11, whereas the Dnmt3b transcripts in somatic cells lacked these exons, suggesting that this region is important for embryonic development. DNMT3b2 and 3b3 were the major isoforms expressed in human cell lines and the mRNA levels of these isoforms closely correlated with their protein levels. Although DNMT3b3 may be catalytically inactive, it still may be biologically important because D4Z4 and satellites 2 and 3 repeat sequences, all known DNMT3b target sequences, were methylated in cells that predominantly expressed DNMT3b3. Treatment of cells with the mechanism-based inhibitor 5-aza-2′-deoxycytidine (5-Aza-CdR) caused a complete depletion of DNMT1, 3a, 3b1, and 3b2 proteins. Human DNMT3b3 and the murine Dnmt3b3-like isoform, Dnmt3b6, were also depleted although less efficiently, suggesting that DNMT3b3 also may be capable of DNA binding. Moreover, de novo methylation of D4Z4 in T24 cancer cells after 5-Aza-CdR treatment only occurred when DNMT3b3 was expressed, reinforcing its role as a contributing factor of DNA methylation. The expression of either DNMT3b2 or 3b3, however, was not sufficient to explain the abnormal methylation of DNMT3b target sequences in human cancers, which may therefore be dependent on factors that affect DNMT3b targeting. Methylation analyses of immunodeficiency, chromosomal instabilities, and facial abnormalities cells revealed that an Alu repeat sequence was highly methylated, suggesting that Alu sequences are not DNMT3b targets.

Structure and function of eukaryotic DNA methyltransferases

DNA methylation is a common epigenetic modification found in eukaryotic organisms ranging from fungi to mammals. Over the past 15 years, a number of eukaryotic DNA methyltransferases have been identified from various model organisms. These enzymes exhibit distinct biochemical properties and biological functions, partly due to their structural differences. The highly variable N-terminal extensions of these enzymes harbor various evolutionarily conserved domains and motifs, some of which have been shown to be involved in functional specializations. DNA methylation has divergent functions in different organisms, consistent with the notion that it is a dynamically evolving mechanism that can be adapted to fulfill various functions. Genetic studies using model organisms have provided evidence suggesting the progressive integration of DNA methylation into eukaryotic developmental programs during evolution.

Mammalian DNA (cytosine-5) methyltransferases and their expression

Two classes of functional DNA (cytosine-5) methyltransferases have been discovered in mammals to date. One class methylates the unmodified DNA and is designated as the de novo enzyme, whereas the other maintains the methylation status of the daughter strand during DNA replication and thus is referred to as a maintenance DNA methyltransferase. Each enzyme catalyzes methyl group transfer from S-adenosyl-L-methionine to cytosine bases in DNA. During methylation the enzyme flips its target base out of the DNA duplex into a typically concave catalytic pocket. This flipped cytosine base is then a substrate for the enzyme-catalyzed reaction. The newly formed 5-methylcytosine confers epigenetic information on the parental genome without altering nucleotide sequences. This epigenetic information is inherited during DNA replication and cell division. In mammals, DNA methylation participates in gene expression, protection of the genome against selfish DNA, parental imprinting, mammalian X chromosome inactivation, developmental regulation, T cell development, and various diseases.

DNA methylation and cancer therapy

Vertebrate DNA is modified by methyl moieties at the 5'-position of cytosine rings residing in the di-nucleotide sequence CpG. Approximately 80% of CpG dinucleotide sequences are methylated. The pattern of distribution of methylated CGs is cell-type specific and correlates with gene expression programming and chromatin structure. Three kinds of seemingly contradictory aberrations in DNA methylation are observed in cancer, global hypomethylation, and regional hypermethylation and deregulated level of expression of DNA methyltransferases. It was previously proposed that the DNA methylation machinery is a candidate target for anticancer therapy. Inhibition of hypermethylation was the first therapeutic target. However, recent data suggests that inhibition of DNA methylation might have untoward effects such as induction of genes involved in metastasis. This review discusses the relative role of the three levels of alteration in the DNA methylation in cancer, proposes a unified hypothesis on the relative roles of increased DNA methyltransferase as well as the coexistence of hypo -and hyper- methylation in cancer and its possible implications on anticancer therapy.

Maintenance DNA methylation of nucleosome core particles

The enzyme responsible for maintenance methylation of CpG dinucleotides in vertebrates is DNMT1. The presence of DNMT1 in DNA replication foci raises the issue of whether this enzyme needs to gain access to nascent DNA before its packaging into nucleosomes, which occurs very rapidly behind the replication fork. Using nucleosomes positioned along the 5 S rRNA gene, we find that DNMT1 is able to methylate a number of CpG sites even when the DNA major groove is oriented toward the histone surface. However, we also find that the ability of DNMT1 to methylate nucleosomal sites is highly dependent on the nature of the DNA substrate. Although nucleosomes containing the Air promoter are refractory to methylation irrespective of target cytosine location, nucleosomes reconstituted onto the H19 imprinting control region are more accessible. These results argue that although DNMT1 is intrinsically capable of methylating some DNA sequences even after their packaging into nucleosomes, this is not the case for at least a fraction of DNA sequences whose function is regulated by DNA methylation.

Reactivation of silenced genes and transcriptional therapy

The purpose of this review is to discuss the potential role of "transcriptional therapy" to modulate the expression of target genes in order to treat monogenic as well as multifactorial disorders. In vitro and in vivo experiments with DNA demethylating and histone hyperacetylating drugs are currently performed in several laboratories on a variety of genes. In attempting to place these results into perspective, we divided the target genes into four major categories: (1) single genes with a hypermethylated CpG island; (2) single genes without a CpG island; (3) groups of genes silenced by aberrant DNA methylation; and (4) groups of genes silenced by lack of histone acetylation. We discuss the latest advances in the field of chromatin regulation and, in particular, the role of histone methylation and that of RNA interference in gene silencing. We can expect that in the future regulation of transcription will become an effective treatment for several genetic conditions.

Age-related epigenetic changes and the immune system

The role of DNA methylation in immune function is discussed extensively in other papers in this issue. Many of these discussions assume that DNA methylation, a major mediator of epigenetic information, is fairly immutable and uniform in adult cells and tissues. There is, however, growing evidence that DNA methylation changes subtly with age. Normal aging cells and tissues show a progressive loss of 5-methylcytosine content, primarily within DNA repeated sequences, but also in potential gene regulatory areas. In parallel, selected genes show progressive age-related increases in promoter methylation, which, once a critical methylation density is reached, have the potential to permanently silence gene expression. These changes are highly mosaic within a given tissue and introduce a high degree of epigenetic variability in aging cells. Such epigenetic phenomena could impact immune response through masking/unmasking potential tissue antigens as well as by modulating the differentiation and response of immune effector cells. The contribution of epigenetic changes to the altered immune function observed in aging humans deserves careful investigation.

Methylation and demethylation in the regulation of genes, cells, and responses in the immune system

DNA methylation is a focus of epigenetic research in the immune system. This overview begins with a synopsis of the players and processes involved in DNA methylation, demethylation, methyl-CpG-recognition, histone modification, and chromatin remodeling. The role of these mechanisms in immune responses, with a focus on T lymphocytes, is then reviewed. There is evidence for epigenetic regulation of several key immune processes including thymocyte development, antigen presentation, differentiation, cytokine expression, effector function, and memory. DNA methylation contributes, along with other epigenetic mechanisms, to the establishment of transcriptional thresholds that vary between genes and T cell types. The immune system is a fertile field for studies of epigenetic regulation of cell fate and function.

The human Dnmt2 has residual DNA-(cytosine-C5) methyltransferase activity

The human Dnmt2 protein is one member of a protein family conserved from Schizosaccharomyces pombe and Drosophila melanogaster to Mus musculus and Homo sapiens. It contains all of the amino acid motifs characteristic for DNA-(Cytosine-C5) methyltransferases, and its structure is very similar to prokaryotic DNA methyltransferases. Nevertheless, so far all attempts to detect catalytic activity of this protein have failed. We show here by two independent assay systems that the purified Dnmt2 protein has weak DNA methyltransferase activity. Methylation was observed at CG sites in a loose ttnCGga(g/a) consensus sequence, suggesting that Dnmt2 has a more specialized role than other mammalian DNA methyltransferases.

Establishment and maintenance of genomic methylation patterns in mouse embryonic stem cells by Dnmt3a and Dnmt3b

We have previously shown that the DNA methyltransferases Dnmt3a and Dnmt3b carry out de novo methylation of the mouse genome during early postimplantation development and of maternally imprinted genes in the oocyte. In the present study, we demonstrate that Dnmt3a and Dnmt3b are also essential for the stable inheritance, or "maintenance," of DNA methylation patterns. Inactivation of both Dnmt3a and Dnmt3b in embryonic stem (ES) cells results in progressive loss of methylation in various repeats and single-copy genes. Interestingly, introduction of the Dnmt3a, Dnmt3a2, and Dnmt3b1 isoforms back into highly demethylated mutant ES cells restores genomic methylation patterns; these isoforms appear to have both common and distinct DNA targets, but they all fail to restore the maternal methylation imprints. In contrast, overexpression of Dnmt1 and Dnmt3b3 failed to restore DNA methylation patterns due to their inability to catalyze de novo methylation in vivo. We also show that hypermethylation of genomic DNA by Dnmt3a and Dnmt3b is necessary for ES cells to form teratomas in nude mice. These results indicate that genomic methylation patterns are determined partly through differential expression of different Dnmt3a and Dnmt3b isoforms.

Increased protein expression of DNA methyltransferase (DNMT) 1 is significantly correlated with the malignant potential and poor prognosis of human hepatocellular carcinomas

Alteration of DNA methylation is one of the most consistent epigenetic changes in human cancers. DNA methyltransferase (DNMT) 1 is a major enzyme involved in establishing genomic methylation patterns. Most of the studies concerning DNMT1 expression in human cancers have been performed only at the mRNA level. To directly examine DNMT1 protein expression levels during human hepatocarcinogenesis, 16 histologically normal liver tissues, 51 noncancerous liver tissues exhibiting chronic hepatitis or cirrhosis, which are considered to be precancerous conditions, and 53 hepatocellular carcinomas (HCCs) were subjected to immunohistochemic examination. If more than 20% of the cells exhibited nuclear DNMT1 staining, the tissue sample was considered to be DNMT1-positive. DNMT1 immunoreactivity was observed in 23 (43%) of the HCCs, but in none (0%) of the histologically normal liver or noncancerous liver tissues exhibiting chronic hepatitis or cirrhosis. The incidence of increased DNMT1 protein expression in HCCs correlated significantly with poor tumor differentiation (p = 0.0006) and portal vein involvement (p = 0.0002). Moreover, the recurrence-free (p = 0.0001) and overall (p < 0.0001) survival rates of patients with HCCs exhibiting increased DNMT1 protein expression were significantly lower than those of patients with HCCs that did not exhibit increased expression. Increased DNMT1 protein expression may play a critical role in the malignant progression of HCCs and be a biologic predictor of both HCC recurrence and a poor prognosis in HCC patients.

Targeting DNA methylation in cancer

There is overwhelming evidence that DNA methylation patterns are altered in cancer. Methylation of CG-rich islands in regulatory regions of genes marks them for transcriptional silencing. Multiple genes, which confer selective advantage upon cancer cells such as tumor suppressors, adhesion molecules, inhibitors of angiogenesis and repair enzymes are silenced. In parallel, tumor cell genomes are globally less methylated than their normal counterparts. In contrast to regional hypermethylation, this loss of methylation in cancer cells occurs in sparsely distributed CG sequences. We now understand that DNA methylation machineries might include a number of DNA methyltransferases, proteins that direct DNA methyltransferases to specific promoters, chromatin modifying enzymes as well as demethylases. There is also data to suggest that pharmacological down regulation of some members of the DNA methylation machinery could inhibit cancer in vitro, in vivo and in clinical trials. Understanding which functions of DNA methylation machinery are critical for cancer is essential for the design of inhibitors of the DNA methylation machinery as anticancer agents. This review discusses the possible role of DNA methyltranferases and demethylases in tumorigenesis and the possible pharmacological and therapeutic implications of the DNA methylation machinery.

A role for poly(ADP-ribosyl)ation in DNA methylation

The aberrant DNA methylation of promoter regions of housekeeping genes leads to gene silencing. Additional epigenetic events, such as histone methylation and acetylation, also play a very important role in the definitive repression of gene expression by DNA methylation. If the aberrant DNA methylation of promoter regions is the starting or the secondary event leading to the gene silencing is still debated. Mechanisms controlling DNA methylation patterns do exist although they have not been ultimately proven. Our data suggest that poly(ADP-ribosyl)ation might be part of this control mechanism. Thus an additional epigenetic modification seems to be involved in maintaining tissue and cell-type methylation patterns that when formed during embryo development, have to be rigorously conserved in adult organisms.

Decrease of DNA methyltransferase 1 expression relative to cell proliferation in transitional cell carcinoma

In many common cancers such as transitional cell carcinoma (TCC), specific genes are hypermethylated, whereas overall DNA methylation is diminished. Genome-wide DNA hypomethylation mostly affects repetitive sequences such as LINE-1 retrotransposons. Methylation of these sequences depends on adequate expression of DNA methyltransferase I (DNMT1) during DNA replication. Therefore, DNMT1 expression relative to proliferation was investigated in TCC cell lines and tissue as well as in renal carcinoma (RCC) cell lines, which also display hypomethylation, as indicated by decreased LINE-1 methylation. Cultured normal uroepithelial cells or normal bladder tissue served as controls. In all tumor cell lines, DNMT1 mRNA as well as protein was decreased relative to the DNA replication factor PCNA, and DNA hypomethylation was present. However, the extents of hypomethylation and DNMT1 downregulation did not correlate. Reporter gene assays showed that the differences in DNMT1 expression between normal and tumor cells were not established at the level of DNMT1 promoter regulation. Diminished DNMT1:PCNA mRNA ratios were also found in 28/45 TCC tissues but did not correlate with the extent of DNA hypomethylation. In addition, expression of the presumed de novo methyltransferases DNMT3A and DNMT3B mRNAs was investigated. DNMT3B overexpression was observed in about half of all high-stage TCC (DNMT3B vs. tumor stage, chi(2): p = 0.03), whereas overexpression of DNMT3A was rarer and less pronounced. Expression of DNMT3A and DNMT3B in most RCC lines was higher than in TCC lines. Our data indicate that DNMT1 expression does not increase adequately with cell proliferation in bladder cancer. This relative downregulation probably contributes to hypomethylation of repetitive DNA but does not determine its extent alone.

Biochemical fractionation reveals association of DNA methyltransferase (Dnmt) 3b with Dnmt1 and that of Dnmt 3a with a histone H3 methyltransferase and Hdac1

De novo DNA methyltransferases, Dnmt3a and 3b, were purified by fractionation of S-100 extract from mouse lymphosarcoma cells through several chromatographic matrices followed by glycerol density gradient centrifugation. Dnmt3a was separated from Dnmt3b and Dnmt1 in the first column, Q-Sepharose whereas Dnmt3b co-purified with Dnmt1 after further fractionation through Mono-S and Mono-Q columns and glycerol density gradient centrifugation. Following purification, the majority of de novo DNA methyltransfearse activity was associated with Dnmt3b/Dnmt1 fractions. By contrast, the fractions containing Dnmt3a alone exhibited markedly reduced activity, which correlated with diminished expression of this isoform in these cells. Histone deacetylase 1(Hdac1) cofractionated with Dnmt3a throughout purification whereas Hdac1 was separated from Dnmt3b/Dnmt1 following chromatography on Mono-Q column. Dnmt3a purified through glycerol gradient centrifugation was also associated with a histone H3 methyltransferase (HMTase) activity whereas purified Dnmt3b/Dnmt1 was devoid of any HMTase activity. The activity of this HMTase was abolished when lysine 9 of N-terminal histone H3 peptide was replaced by leucine whereas mutation of lysine 4 to leucine inhibited this activity only partially. This is the first report on the identification of a few key co-repressors associated with endogenous Dnmt3a and of a complex containing Dnmt3b and a minor form of Dnmt1 following extensive biochemical fractionation.

The power and the promise of DNA methylation markers

The past few years have seen an explosion of interest in the epigenetics of cancer. This has been a consequence of both the exciting coalescence of the chromatin and DNA methylation fields, and the realization that DNA methylation changes are involved in human malignancies. The ubiquity of DNA methylation changes has opened the way to a host of innovative diagnostic and therapeutic strategies. Recent advances attest to the great promise of DNA methylation markers as powerful future tools in the clinic.

Mutation of the DNA methyltransferase (DNMT) 1 gene in human colorectal cancers

Alteration of DNA methylation is one of the most consistent epigenetic changes in human cancers. DNA methyltransferase (DNMT) 1 is a major enzyme that determines genomic methylation patterns. In order to understand the significance of mutations of the DNMT1 gene during human carcinogenesis, we performed polymerase chain reaction-single strand conformation polymorphism analysis using 46 oligonucleotide primer sets for all 40 coding exons and the 5'-flanking region (450 bp) of the DNMT1 gene in 29 colorectal cancers, 32 stomach cancers, 40 hepatocellular carcinomas (HCCs) and a corresponding sample of non-cancerous tissue from each case. Mutations in coding exons of the DNMT1 gene were detected in two (7%) of the colorectal cancers: they consisted of one-base deletion resulting in deletion of the whole catalytic domain and a point mutation resulting in a single amino acid substitution. No stomach cancers or HCCs showed mutations in the coding exons of the DNMT1 gene. No mutation in the 5'-flanking region of the DNMT1 gene was detected in any of the colorectal and stomach cancers or HCCs. These data suggest that mutational inactivation of the DNMT1 gene that potentially causes a genome-wide alteration of DNA methylation status may be a rare event during human carcinogenesis.

Inhibition of DNA methylation and reactivation of silenced genes by zebularine

BACKGROUND

Gene silencing by abnormal methylation of promoter regions of regulatory genes is commonly associated with cancer. Silenced tumor suppressor genes are obvious targets for reactivation by methylation inhibitors such as 5-azacytidine (5-Aza-CR) and 5-aza-2'-deoxycytidine (5-Aza-CdR). However, both compounds are chemically unstable and toxic and neither can be given orally. We characterized a new demethylating agent, zebularine [1-(beta-D-ribofuranosyl)-1,2-dihydropyrimidin-2-one], which is a chemically stable cytidine analog.

METHODS

We tested the ability of zebularine to reactivate a silenced Neurospora crassa gene using a hygromycin gene reactivation assay. We then analyzed the ability of zebularine to inhibit DNA methylation in C3H 10T1/2 Cl8 (10T1/2) mouse embryo cells as assayed by induction of a myogenic phenotype and in T24 human bladder carcinoma cells, using the methylation-sensitive single nucleotide primer extension (Ms-SNuPE) assay. We also evaluated the effects of zebularine (administered orally or intraperitoneally) on growth of EJ6 human bladder carcinoma cells grown in BALB/c nu/nu mice (five mice per group) and the in vivo reactivation of a methylated p16 gene in these cells. All statistical tests were two-sided.

RESULTS

In N. crassa, zebularine inhibited DNA methylation and reactivated a gene previously silenced by methylation. Zebularine induced the myogenic phenotype in 10T1/2 cells, which is a phenomenon unique to DNA methylation inhibitors. Zebularine reactivated a silenced p16 gene and demethylated its promoter region in T24 bladder carcinoma cells in vitro and in tumors grown in mice. Zebularine was only slightly cytotoxic to T24 cells in vitro (1 mM zebularine for 48 hours decreased plating efficiency by 17% [95% confidence interval (CI) = 12.8% to 21.2%]) and to tumor-bearing mice (average maximal weight change in mice treated with 1000 mg/kg zebularine = 11% [95% CI = 4% to 19%]). Compared with those in control mice, tumor volumes were statistically significantly reduced in mice treated with high-dose zebularine administered by intraperitoneal injection (P<.001) or by oral gavage (P<.001).

CONCLUSIONS

Zebularine is a stable DNA demethylating agent and the first drug in its class able to reactivate an epigenetically silenced gene by oral administration.

Epigenetics in carcinogenesis and cancer prevention

Carcinogenesis is a stepwise process of accumulation of genetic and epigenetic abnormalities that can lead to cellular dysfunction. It has become clear that epigenetic changes are equally important for this multistep process to produce its results. This article describes the different roles that epigenetic modulation may play during carcinogenesis and how an early detection and chemopreventive intervention strategy that takes both sides of the equation into account would be advantageous.

Epidemiologic considerations to assess altered DNA methylation from environmental exposures in cancer

Epidemiologic studies in human populations have identified a broad spectrum of risk factors for cancer. Gene-damaging agents have been a primary focus of cancer epidemiology; however, all xenobiotics do not interact with DNA directly. Some exogenous agents induce epigenetic changes. In view of this, markers that measure changes to the epigenome must also be incorporated into molecular epidemiologic studies. We review the current understanding of the impact of exogenous agents including: micronutrients, chemotherapeutic agents, metals, and others, on DNA methylation. Two categories of genes are described: (1) genes that can alter susceptibility to aberrant DNA methylation and (2) genes that increase susceptibility to cancer when they are silenced through DNA methylation. Methods for incorporating markers of DNA methylation status into etiologic investigations of the impact of environmental exposures on disease (e.g., cancer) are discussed.

Resistance of IAPs to methylation reprogramming may provide a mechanism for epigenetic inheritance in the mouse

Genome-wide epigenetic reprogramming by demethylation occurs in early mouse embryos and primordial germ cells. In early embryos many single-copy sequences become demethylated both by active and passive demethylation, whereas imprinted gene methylation remains unaffected. In primordial germ cells single-copy and imprinted sequences are demethylated, presumably by active demethylation. Here we investigated systematically by bisulphite sequencing the methylation profiles of IAP and Line1 repeated sequence families during preimplantation and primordial germ cell development. Whereas Line1 elements were substantially demethylated during both developmental periods, IAP elements were largely resistant to demethylation, particularly during preimplantation development. This may be desirable in order to prevent IAP retrotransposition, which could cause mutations. In turn, this can result in the transgenerational inheritance of epigenetic states of IAPs, which could lead to heritable epimutations of neighbouring genes through influencing their transcriptional states.

DNA methylation in animal development

Nuclear transfer experiments have demonstrated that epigenetic mechanisms operate to limit gene expression during animal development. In somatic cells, silenced genes are associated with defined chromatin states which are characterised by hypermethylation of DNA, hypoacetylation of histones and specific patterns of methylation at distinct residues of the N-terminal tails of histone H3 and H4. This review describes the role of the DNA methylation-mediated repression system (Dnmt1's, MeCPs and MBDs and associated chromatin remodelling activities) in animal development. DNA methylation is essential for normal vertebrate development but has distinct regulatory roles in non-mammalian and mammalian vertebrates. In mammals, DNA methylation has an additional role in regulating imprinting. This suggests that epigenetic regulation is plastic in its application and should be considered in a developmental context that may be species specific.

DNMT1 is required to maintain CpG methylation and aberrant gene silencing in human cancer cells

Transcriptional silencing by CpG island methylation is a prevalent mechanism of tumor-suppressor gene suppression in cancers. Genetic experiments have defined the importance of the DNA methyltransferase Dnmt1 for the maintenance of methylation in mouse cells and its role in neoplasia. In human bladder cancer cells, selective depletion of DNMT1 with antisense inhibitors has been shown to induce demethylation and reactivation of the silenced tumor-suppressor gene CDKN2A. In contrast, targeted disruption of DNMT1 alleles in HCT116 human colon cancer cells produced clones that retained CpG island methylation and associated tumor-suppressor gene silencing, whereas HCT116 clones with inactivation of both DNMT1 and DNMT3B showed much lower levels of DNA methylation, suggesting that the two enzymes are highly cooperative. We used a combination of genetic (antisense and siRNA) and pharmacologic (5-aza-2'-deoxycytidine) inhibitors of DNA methyl transferases to study the contribution of the DNMT isotypes to cancer-cell methylation. Selective depletion of DNMT1 using either antisense or siRNA resulted in lower cellular maintenance methyltransferase activity, global and gene-specific demethylation and re-expression of tumor-suppressor genes in human cancer cells. Specific depletion of DNMT1 but not DNMT3A or DNMT3B markedly potentiated the ability of 5-aza-2'-deoxycytidine to reactivate silenced tumor-suppressor genes, indicating that inhibition of DNMT1 function is the principal means by which 5-aza-2'-deoxycytidine reactivates genes. These results indicate that DNMT1 is necessary and sufficient to maintain global methylation and aberrant CpG island methylation in human cancer cells.

DNA methylation density influences the stability of an epigenetic imprint and Dnmt3a/b-independent de novo methylation

DNA methylation plays an important role in transcriptional repression. To gain insight into the dynamics of demethylation and de novo methylation, we introduced a proviral reporter, premethylated at different densities, into a defined chromosomal site in murine erythroleukemia cells and monitored the stability of the introduced methylation and reporter gene expression. A high density of methylation was faithfully propagated in vivo. In contrast, a low level of methylation was not stable, with complete demethylation and associated transcriptional activation or maintenance-coupled de novo methylation and associated silencing occurring with equal probability. Deletion of the proviral enhancer increased the probability of maintenance-coupled de novo methylation, suggesting that this enhancer functions in part to antagonize such methylation. The DNA methyltransferases (MTases) Dnmt3a and Dnmt3b are thought to be the sole de novo MTases in the mammalian genome. To determine whether these enzymes are responsible for maintenance-coupled de novo methylation, the unmethylated or premethylated proviral reporter was introduced into DNA MTase-deficient embryonic stem cells. These studies revealed the presence of a Dnmt3a/Dnmt3b-independent de novo methyltransferase activity that is stimulated by the presence of preexisting methylation.

Dnmt3a and Dnmt1 functionally cooperate during de novo methylation of DNA

Dnmt3a is a de novo DNA methyltransferase that modifies unmethylated DNA. In contrast Dnmt1 shows high preference for hemimethylated DNA. However, Dnmt1 can be activated for the methylation of unmodified DNA. We show here that the Dnmt3a and Dnmt1 DNA methyltransferases functionally cooperate in de novo methylation of DNA, because a fivefold stimulation of methylation activity is observed if both enzymes are present. Stimulation is observed if Dnmt3a is used before Dnmt1, but not if incubation with Dnmt1 precedes Dnmt3a, demonstrating that methylation of the DNA by Dnmt3a stimulates Dnmt1 and that no physical interaction of Dnmt1 and Dnmt3a is required. If Dnmt1 and Dnmt3a were incubated together a slightly increased stimulation is observed that could be due to a direct interaction of these enzymes. In addition, we show that Dnmt1 is stimulated for methylation of unmodified DNA if the DNA already carries some methyl groups. We conclude that after initiation of de novo methylation of DNA by Dnmt3a, Dnmt1 becomes activated by the pre-existing methyl groups and further methylates the DNA. Our data suggest that Dnmt1 also has a role in de novo methylation of DNA. This model agrees with the biochemical properties of these enzymes and provides a mechanistic basis for the functional cooperation of different DNA MTases in de novo methylation of DNA that has also been observed in vivo.

Mechanisms underlying epigenetically mediated gene silencing in cancer

It has become apparent that epigenetically mediated alterations, which establish heritable abnormalities in gene expression, are a fundamental feature of human cancer. The best studied of these changes are aberrant gene silencing events which involve transcriptional inactivation associated with abnormally methylated promoter region CpG islands. A most important aspect of understanding this change, which can cause loss of key gene function, concerns dissection of the molecular mechanisms that mediate the transcriptional repression and those responsible for establishing the abnormal methylation and associated chromatin events. This chapter reviews the progress in these arenas.

Identification and characterization of alternatively spliced variants of DNA methyltransferase 3a in mammalian cells

CpG methylation is mediated by the functions of at least three active DNA methyltransferases (DNMTs). While DNMT1 is thought to perform maintenance methylation, the more recently discovered DNMT3a and DNMT3b enzymes are thought to facilitate de novo methylation. Murine Dnmt3a and 3b are developmentally regulated and a new Dnmt3a isoform, Dnmt3a2, has been recently shown to be expressed preferentially in mouse embryonic stem (ES) cells. Here we have characterized four alternatively spliced variants of human and mouse DNMT3a. These transcripts included a novel exon 1 (1beta) that was spliced into the same exon 2 acceptor splice site used by the original exon 1 (1alpha). Cloning and sequencing of the 5' region of the human DNMT3a gene revealed that exon 1beta was situated upstream of exon 1alpha and that the entire region was contained within a CpG island. We also identified other alternatively spliced species containing intron 4 inclusions that were associated with either exon 1alpha or 1beta. These were expressed at low levels in mouse and human cells. All transcripts were highly conserved between human and mouse. The levels of Dnmt3a mRNA containing exon 1beta were 3-25-fold greater in mouse ES cells than in various somatic cells as determined by semiquantitative reverse transcription-polymerase chain reaction analysis, while the levels of exon 1alpha-containing transcripts were slightly higher in human and mouse somatic cells. The preferential expression of the beta transcript in ES cells suggests that this transcript, in addition to Dnmt3a2, may also be important for de novo methylation during development.