DNA methylation-dependent modulator of Gsg2/Haspin gene expression.

The Gsg2 (Haspin) gene encodes a serine/threonine protein kinase and is predominantly expressed in haploid germ cells. In proliferating somatic cells, Gsg2 is shown to be expressed weakly but plays an essential role in mitosis. Although the Gsg2 minimal promoter recognized by the spermatogenic cell-specific nuclear factor(s) has been found, to date, the molecular mechanism that differentially controls Gsg2 expression levels in germ and somatic cells remains to be sufficiently clarified. In this study, we analyzed the DNA methylation status of the upstream region containing the Gsg2 promoter. We found a tissue-dependent and differentially methylated region (T-DMR) upstream (-641 to -517) of the authentic promoter that is hypomethylated in germ cells but hypermethylated in other somatic tissues. Profiling of Gsg2 expression and DNA methylation status at the T-DMR in spermatogenic cells indicated that the hypomethylation of the T-DMR is maintained during spermatogenesis. Using the reporter assay, we also demonstrated that DNA methylation at the T-DMR of Gsg2 reduced the promoter activity by 60-80%, but did not fully suppress it. Therefore, the T-DMR functions as a modulator in a DNA methylation-dependent manner. In conclusion, Gsg2 is under epigenetic control.

Establishment of trophoblast stem cell lines from somatic cell nuclear-transferred embryos.

Placental abnormalities occur frequently in cloned animals. Here, we attempted to isolate trophoblast stem (TS) cells from mouse blastocysts produced by somatic cell nuclear transfer (NT) at the blastocyst stage (NT blastocysts). Despite the predicted deficiency of the trophoblast cell lineage, we succeeded in isolating cell colonies with typical morphology of TS cells and cell lines from the NT blastocysts (ntTS cell lines) with efficiency as high as that from native blastocysts. The established 10 ntTS cell lines could be maintained in the undifferentiated state and induced to differentiate into several trophoblast subtypes in vitro. A comprehensive analysis of the transcriptional and epigenetic traits demonstrated that ntTS cells were indistinguishable from control TS cells. In addition, ntTS cells contributed exclusively to the placenta and survived until term in chimeras, indicating that ntTS cells have developmental potential as stem cells. Taken together, our data show that NT blastocysts contain cells that can produce TS cells in culture, suggesting that proper commitment to the trophoblast cell lineage in NT embryos occurs by the blastocyst stage.

Aberrant DNA methylation status in human uterine leiomyoma.

Aberrant DNA methylation has been implicated in tumorigenesis. This study was undertaken to establish the genome-wide DNA methylation profile in uterine leiomyomas and to investigate whether DNA methylation status is altered in uterine leiomyomas. For this purpose, restriction landmark genomic scanning (RLGS) was performed on a paired sample of leiomyoma and adjacent normal myometrium. The RLGS profile revealed 29 aberrant methylation spots (10 methylated and 19 demethylated) in leiomyoma in comparison with myometrium. One of the differently methylated genomic loci was newly identified as GS20656 from the human genome sequence database. In 9 of the 10 paired samples, the DNA methylation levels of the first exon of GS20656 were significantly lower in leiomyoma than in myometrium, suggesting the existence of a genomic locus under epigenetic regulation in uterine leiomyomas. Unexpectedly, DNA methyltransferase 1 (DNMT1) and DNMT3a mRNA expression levels were higher in leiomyoma than in myometrium. These facts suggest that other epigenetic factors besides DNMT are involved in local changes of DNA methylation at genome loci. The present study indicates not only aberrant genome-wide DNA methylation status in uterine leiomyomas but also the existence of a genomic locus that is differently methylated between normal myometrium and uterine leiomyoma.

DNA methylation profile of tissue-dependent and differentially methylated regions (T-DMRs) in mouse promoter regions demonstrating tissue-specific gene expression.

DNA methylation constitutes an important epigenetic regulation mechanism in many eukaryotes, although the extent of DNA methylation in the regulation of gene expression in the mammalian genome is poorly understood. We developed D-REAM, a genome-wide DNA methylation analysis method for tissue-dependent and differentially methylated region (T-DMR) profiling with restriction tag-mediated amplification in mouse tissues and cells. Using a mouse promoter tiling array covering a region from -6 to 2.5 kb ( approximately 30,000 transcription start sites), we found that over 3000 T-DMRs are hypomethylated in liver compared to cerebrum. The DNA methylation profile of liver was distinct from that of kidney and spleen. This hypomethylation profile marked genes that are specifically expressed in liver, including key transcription factors such as Hnf1a and Hnf4a. Genes with T-DMRs, especially those lacking CpG islands and those with HNF-1A binding motifis in their promoters, showed good correlation between their tissue-specific expression and liver hypomethylation status. T-DMRs located downstream from their transcription start sites also showed tissue-specific gene expression. These data indicate that multilayered regulation of tissue-specific gene function could be elucidated by DNA methylation tissue profiling.

Potential link between estrogen receptor-alpha gene hypomethylation and uterine fibroid formation.

Uterine leiomyomas are the most common uterine tumors in women. Estrogen receptor-alpha (ER-alpha) is more highly expressed in uterine leiomyomas than in normal myometrium, suggesting a link between uterine leiomyomas and ER-alpha expression. DNA methylation is an epigenetic mechanism of gene regulation and plays important roles in normal embryonic development and in disease progression including cancers. Here, we investigated the DNA methylation status of the ER-alpha promoter region (-1188 to +229 bp) in myometrium and leiomyoma. By sodium bisulfite sequencing, 49 CpG sites in the proximal promoter region of ER-alpha gene were shown to be unmethylated in both leiomyoma and normal myometrium. At seven CpG sites in the distal promoter region of the ER-alpha gene, there was a variation in DNA methylation status in myometrium and leiomyoma. Further analysis of the DNA methylation status by bisulfite restriction mapping among 11 paired samples of myometrium and leiomyoma indicated that CpG sites in the distal region of ER-alpha promoter are hypomethylated in leiomyomas of nine patients. In those patients, ER-alpha mRNA levels tended to be higher in the leiomyoma than in the myometrium. In conclusion, there was an aberrant DNA methylation status in the promoter region of ER-alpha gene in uterine leiomyoma, which may be associated with high ER-alpha mRNA expression.

Epigenetics: the study of embryonic stem cells by restriction landmark genomic scanning.

During mammalian development, it is essential that the proper epigenetic state is established across the entire genome in each differentiated cell. To date, little is known about the mechanism for establishing epigenetic modifications of individual genes during the course of cellular differentiation. Genome-wide DNA methylation analysis of embryonic stem cells by restriction landmark genomic scanning provides information about cell type- and tissue-specific DNA methylation profiles at tissue-specific methylated regions associated with developmental processes. It also sheds light on DNA methylation alterations following fetal exposure to chemical agents. In addition, analysis of embryonic stem cells deficient in epigenetic regulators will contribute to revealing the mechanism for establishing DNA methylation profiles and the interplay between DNA methylation and other epigenetic modifications.

Sequential changes in genome-wide DNA methylation status during adipocyte differentiation.

DNA methylation is an epigenetic mark on the mammalian genome. There are numerous tissue-dependent and differentially methylated regions (T-DMRs) in the unique sequences distributed throughout the genome. To determine the epigenetic changes during adipocyte differentiation, we investigated the sequential changes in DNA methylation status of 3T3-L1 cells at the growing, confluent, postconfluent and mature adipocyte cell stages. Treatment of 3T3-L1 cells with 5-aza-2'-deoxycytidine inhibited differentiation in a stage-dependent manner, supporting the idea that formation of accurate DNA methylation profile, consisting of methylated and unmethylated T-DMRs, may be involved in differentiation. Analysis by methylation-sensitive quantitative real-time PCR of the 65 known T-DMRs which contain NotI sites detected 8 methylations that changed during differentiation, and the changes in the patterns of these methylations were diverse, confirming that the differentiation process involves epigenetic alteration at the T-DMRs. Intriguingly, the dynamics of the methylation change vary depending on the T-DMRs and differentiation stages. Restriction landmark genomic scanning detected 32 novel T-DMRs, demonstrating that differentiation of 3T3-L1 cells involves genome-wide epigenetic changes by temporal methylation/demethylation, in addition to maintenance of a static methylated/demethylated state, and both depend on differentiation stage.

A new class of tissue-specifically methylated regions involving entire CpG islands in the mouse.

CpG islands, which have higher GC content and CpG frequencies compared to the genome as a whole, are generally believed to be unmethylated in tissues except at promoters of genes undergoing X chromosome inactivation or genomic imprinting. Recent studies, however, have shown that CpG islands at promoters of a number of genes contain tissue-dependent, differentially methylated regions (T-DMRs). In general, the tissue-specific methylation is restricted to a part of the promoter CpG island, with hypomethylation of the remaining sequence. In the current study, using comparison between Restriction Landmark Genomic Scanning (RLGS) and in silico RLGS, we identified ten sperm-specific unmethylated NotI sites, T-DMRs located in CpG islands that were hypomethylated in sperm but near-completely methylated in the kidney and brain. Unusually, these T-DMRs involve the whole CpG island at each of these loci. We characterized one of these genes, adenine nucleotide translocator 4 (Ant4), which is expressed in germ cells. Using a promoter assay, we demonstrated that expression of Ant4 gene is controlled by DNA methylation at the CpG island sequences within the promoter region. Ant4 and other sperm-specific hypomethylated loci represent a new class of CpG islands that become completely methylated in different cell lineages. T-DMRs at CpG islands are functionally important gene regulatory elements that may now be categorized into two classes: T-DMRs involving a subregion of the CpG island and those that occupy the whole CpG island.

DNA methylation errors in cloned mice disappear with advancement of aging.

Cloned animals have various health problems. Aberrant DNA methylation is a possible cause of the problems. Restriction landmark genomic scanning (RLGS) that enabled us to analyze more than 1,000 CpG islands simultaneously demonstrated that all cloned newborns had aberrant DNA methylation. To study whether this aberration persists throughout the life of cloned individuals, we examined genome-wide DNA methylation status of newborn (19.5 dpc, n=2), adult (8-11 months old, n=3), and aged (23-27 months old, n=4) cloned mice using kidney cells as representatives. In the adult and aged groups, cloning was repeated using cumulus cells of the adult founder clone of each group as nucleus donor. Two newborn clones had three with aberrantly methylated loci, which is consistent with previous reports that all cloned newborns had DNA methylation aberrations. Interestingly, we could detect only one aberrantly methylated locus in two of the three adult clones in mid-age and none of four senescent clones, indicating that errors in DNA methylation disappear with advancement of animals' aging.

Cell type-specific methylation profiles occurring disproportionately in CpG-less regions that delineate developmental similarity.

Our previous studies using restriction landmark genomic scanning (RLGS) defined tissue- or cell-specific DNA methylation profiles. It remains to be determined whether the DNA sequence compositions in the genomic contexts of the NotI loci tested by RLGS influence their tendency to change with differentiation. We carried out 3834 methylation measurements consisting of 213 NotI loci in the mouse genome in 18 different tissues and cell types, using quantitative real-time PCR based on a Virtual image rlgs database. Loci were categorized as CpG islands or other, and as unique or repetitive sequences, each category being associated with a variety of methylation categories. Strikingly, the tissue-dependently and differentially methylated regions (T-DMRs) were disproportionately distributed in the non-CpG island loci. These loci were located not only in 5'-upstream regions of genes but also in intronic and non-genic regions. Hierarchical clustering of the methylation profiles could be used to define developmental similarity and cellular phenotypes. The results show that distinctive tissue- and cell type-specific methylation profiles by RLGS occur mostly at NotI sites located at non-CpG island sequences, which delineate developmental similarity of different cell types. The finding indicates the power of NotI methylation profiles in evaluating the relatedness of different cell types.

Regulation of tissue-specific expression of the human and mouse urate transporter 1 gene by hepatocyte nuclear factor 1 alpha/beta and DNA methylation.

Expression of Urate transporter 1 (URAT1/SLC22A12) is restricted to the proximal tubules in the kidney, where it is responsible for the tubular reabsorption of urate. To elucidate the mechanism underlying its tissue-specific expression, the transcriptional regulation of the hURAT1 and mUrat1 genes was investigated. Hepatocyte nuclear factor 1 alpha (HNF1alpha) and HNF1beta positively regulate minimal promoter activity of the URAT1 gene as shown by reporter gene assays. Electrophoretic mobility shift assays revealed binding of HNF1alpha and/or HNF1beta to the HNF1 motif in the hURAT1 promoter. Furthermore, the mRNA expression of Urat1 is reduced in the kidneys of Hnf1alpha-null mice compared with wild-type mice, confirming the indispensable role of HNF1alpha in the constitutive expression of URAT1 genes. It was also shown that the proximal promoter region of mUrat1 was hypermethylated in the liver and kidney medulla, whereas this region was relatively hypomethylated in the kidney cortex. These methylation profiles are in a good agreement with the proximal tubule-restricted expression of mUrat1 in the kidney cortex. Taken together, these results strongly suggest that tissue-specific expression of the URAT1 genes is coordinately regulated by the transcriptional activation by HNF1alpha/HNF1beta heterodimer and repression by DNA methylation.

Epigenetic regulation of Nanog gene in embryonic stem and trophoblast stem cells.

The Nanog and Oct-4 genes are essential for maintaining pluripotency of embryonic stem (ES) cells and early embryos. We previously reported that DNA methylation and chromatin remodeling underlie the cell type-specific mechanism of Oct-4 gene expression. In the present study, we found that there is a tissue-dependent and differentially methylated region (T-DMR) in the Nanog up-stream region. The T-DMR is hypomethylated in ES cells, but is heavily methylated in trophoblast stem (TS) cells and NIH/3T3 cells, in which the Nanog gene is repressed. Furthermore, in vitro methylation of T-DMR suppressed Nanog promoter activity in reporter assay. Chromatin immunoprecipitation assay revealed that histone H3 and H4 are highly acetylated, and H3 lysine (K) 4 is hypermethylated at the Nanog locus in ES cells. Conversely, histone deacetylation and H3-K4 demethylation occurred in TS cells. Importantly, in TS cells, hypermethylation of H3-K9 and -K27 is found only at the Nanog locus, not the Oct-4 locus, indicating that the combination of histone modifications associated with the Nanog gene is distinct from that of the Oct-4 gene. In conclusion, the Nanog gene is regulated by epigenetic mechanisms involving DNA methylation and histone modifications.

Genome-wide and locus-specific DNA hypomethylation in G9a deficient mouse embryonic stem cells.

In the mammalian genome, numerous CpG-rich loci define tissue-dependent and differentially methylated regions (T-DMRs). Euchromatin from different cell types differs in terms of its tissue-specific DNA methylation profile as defined by these T-DMRs. G9a is a euchromatin-localized histone methyltransferase (HMT) and catalyzes methylation of histone H3 at lysines 9 and 27 (H3-K9 and -K27). To test whether HMT activity influences euchromatic cytosine methylation, we analyzed the DNA methylation status of approximately 2000 CpG-rich loci, which are predicted in silico, in G9a(-/-) embryonic stem cells by restriction landmark genomic scanning (RLGS). While the RLGS profile of wild-type cells contained about 1300 spots, 32 new spots indicating DNA demethylation were seen in the profile of G9a(-/-) cells. Virtual-image RLGS (Vi-RLGS) allowed us to identify the genomic source of ten of these spots. These were confirmed to be cytosine demethylated, not just at the Not I site detected by the RLGS but extending over several kilobase pairs in cis. Chromatin immunoprecipitation (ChIP) confirmed these loci to be targets of G9a, with decreased H3-K9 and/or -K27 dimethylation in the G9a(-/-) cells. These data indicate that G9a site-selectively contributes to DNA methylation.

Methylation in embryonic stem cells in vitro.

Stem cells raise the possibility of regenerating failing body parts with new tissue. Before stem cells can safely fulfill their promise, many technical problems, including understanding the stem cell phenotype, must be overcome. DNA methylation, which is responsible for gene silencing and is associated with chromatin remodeling, is an epigenetic system that determines the specific characteristic of a variety of cells, including stem cells. Each cell type has a unique DNA methylation profile produced by varied loci-specific methylation. Investigation of such DNA methylation profiles provides a way of identifying pluripotent stem cells. Further, it is likely that analysis of the epigenetic status of stem cells may provide novel information regarding "sternness" within these populations.

Dimethyl sulfoxide has an impact on epigenetic profile in mouse embryoid body.

Dimethyl sulfoxide (DMSO), an amphipathic molecule, is widely used not only as a solvent for water-insoluble substances but also as a cryopreservant for various types of cells. Exposure to DMSO sometimes causes unexpected changes in cell fates. Because mammalian development and cellular differentiation are controlled epigenetically by DNA methylation and histone modifications, DMSO likely affects the epigenetic system. The effects of DMSO on transcription of three major DNA methyltransferases (Dnmts) and five well-studied histone modification enzymes were examined in mouse embryonic stem cells and embryoid bodies (EBs) by reverse transcription-polymerase chain reaction. Addition of DMSO (0.02%-1.0%) to EBs in culture induced an increase in Dnmt3a mRNA levels with increasing dosage. Increased expression of two subtypes of Dnmt3a in protein levels was confirmed by Western blotting. Southern blot analysis revealed that DMSO caused hypermethylation of two kinds of repetitive sequences in EBs. Furthermore, restriction landmark genomic scanning, by which DNA methylation status can be analyzed on thousands of loci in genic regions, revealed that DMSO affected DNA methylation status at multiple loci, inducing hypomethylation as well as hypermethylation depending on the genomic loci. In conclusion, DMSO has an impact on the epigenetic profile: upregulation of Dnmt3a expression and alteration of genome-wide DNA methylation profiles with phenotypic changes in EBs.

Regulation of the expression of human organic anion transporter 3 by hepatocyte nuclear factor 1alpha/beta and DNA methylation.

Human organic anion transporter 3 (hOAT3/SLC22A8) is predominantly expressed in the proximal tubules of the kidney and plays a major role in the urinary excretion of a variety of organic anions. The promoter region of hOAT3 was characterized to elucidate the mechanism underlying the tissue-specific expression of hOAT3. The minimal promoter of hOAT3 was identified to be located approximately 300 base pairs upstream of the transcriptional start site, where there are canonical TATA and hepatocyte nuclear factor (HNF1) binding motifs, which are conserved in the rodent Oat3 genes. Transactivation assays revealed that HNF1alpha and HNF1beta markedly increased hOAT3 promoter activity, where the transactivation potency of HNF1beta was lower than that of HNF1alpha. Mutations in the HNF1 binding motif prevented the transactivation. Electrophoretic mobility shift assays demonstrated binding of the HNF1alpha/HNF1alpha homodimer or HNF1alpha/HNF1beta heterodimer to the hOAT3 promoter. It was also demonstrated that the promoter activity of hOAT3 is repressed by DNA methylation. Moreover, the expression of hOAT3 was activated de novo by forced expression of HNF1alpha alone or both HNF1alpha and HNF1beta together with the concomitant DNA demethylation in human embryonic kidney 293 cells that lack expression of endogenous HNF1alpha and HNF1beta, whereas forced expression of HNF1beta alone could not activate the expression of hOAT3. This suggests a synergistic action of the HNF1alpha/HNF1alpha homodimer or HNF1alpha/HNF1beta heterodimer and DNA demethylation for the constitutive expression of hOAT3. These results indicate that the tissue-specific expression of hOAT3 might be regulated by the concerted effect of genetic (HNF1alpha and HNF1beta) and epigenetic (DNA methylation) factors.

Proliferation related acidic leucine-rich protein PAL31 functions as a caspase-3 inhibitor.

Proliferation related acidic leucine-rich protein PAL31 (PAL31) is expressed in proliferating cells and consists of 272 amino acids with a tandem structure of leucine-rich repeats in the N-terminus and a highly acidic region with a putative nuclear localization signal in the C-terminus. We previously reported that PAL31 is required for cell cycle progression. In the present study, we found that the antisense oligonucleotide of PAL31 induced apoptosis to the transfected Nb2 cells. Stable transfectants, in which PAL31 was regulated by an inducible promoter, were generated to gain further insight into the signaling role of PAL31 in the regulation of apoptosis. Expression of PAL31 resulted in the marked rescue of Rat1 cells from etoposide and UV radiation-induced apoptosis and the cytoprotection was correlated with the levels of PAL31 protein. Thus, cytoprotection from apoptosis is a physiological function of PAL31. PAL31 can suppress caspase-3 activity but not cytochrome c release in vitro, indicating that PAL31 is a direct caspase-3 inhibitor. In conclusion, PAL31 is a multifunctional protein working as a cell cycle progression factor as well as a cell survival factor.

Analysis of tissue-specific DNA methylation during development.

The establishment of a cell identification method more accurate than the conventional morphological and cell-type-specific marker analyses has been desired. DNA methylation is related to the gene activity including gene-silencing and is a key mechanism of epigenetics underling cellular differentiaion and development in mammals. Recent studies indicated that there exist unique genomic DNA methylation profiles specific to the cell type. DNA methylation profiles a mechanism for memorizing the set of genes inherent in individual type of cells. In this chapter, we present the methods to analyze DNA methylation status for identifying cells or tissues.

Skewed X-inactivation in cloned mice.

In female mammals, dosage compensation for X-linked genes is accomplished by inactivation of one of two X chromosomes. The X-inactivation ratio (a percentage of the cells with inactivated maternal X chromosomes in the whole cells) is skewed as a consequence of various genetic mutations, and has been observed in a number of X-linked disorders. We previously reported that phenotypically normal full-term cloned mouse fetuses had loci with inappropriate DNA methylation. Thus, cloned mice are excellent models to study abnormal epigenetic events in mammalian development. In the present study, we analyzed X-inactivation ratios in adult female cloned mice (B6C3F1). Kidneys of eight naturally produced controls and 11 cloned mice were analyzed. Although variations in X-inactivation ratio among the mice were observed in both groups, the distributions were significantly different (Ansary-Bradley test, P<0.01). In particular, 2 of 11 cloned mice showed skewed X-inactivation ratios (19.2% and 86.8%). Similarly, in intestine, 1 of 10 cloned mice had a skewed ratio (75.7%). Skewed X-inactivation was observed to various degrees in different tissues of different individuals, suggesting that skewed X-inactivation in cloned mice is the result of secondary cell selection in combination with stochastic distortion of primary choice. The present study is the first demonstration that skewed X-inactivation occurs in cloned animals. This finding is important for understanding both nuclear transfer technology and etiology of X-linked disorders.

Non-coding RNA directed DNA demethylation of Sphk1 CpG island.

The formation of DNA methylation patterns is one of the epigenetic events that underlie mammalian development. The Sphk1 CpG island is a target for tissue-dependent DNA methylation as well as a template for generating multiple subtypes. The number of mammalian non-coding RNA genes is rapidly expanding. In this study, we found endogenous antisense transcripts, Khps1 subtypes with different sizes (600-20,000nt). A subtype, Khps1a, was a 1290-bp, non-coding, 5'-capped and 3'-polyadenylated RNA that originated from the CpG island and overlapped with a tissue-dependent differentially methylated region (T-DMR) of Sphk1. Intriguingly, overexpression of two fragments of Khps1 caused demethylation of CG sites in the T-DMR. Furthermore, this RNA-directed demethylation was associated with DNA methylation at three CC(A/T)GG sites in the T-DMR. The link between the RNA-directed CG demethylation and non-CG methylation provides a novel mechanism of epigenetic regulation and potential tool for epigenetic manipulation of mammalian cells.