DNA Demethylation Activity Research Articles

AtMBD9 modulates Arabidopsis development through the dual epigenetic pathways of DNA methylation and histone acetylation

Mutations within the Arabidopsis METHYL-CpG BINDING DOMAIN 9 gene (AtMBD9) cause pleotropic phenotypes including early flowering and multiple lateral branches. Early flowering was previously attributed to the repression of flowering locus C (FLC) due to a reduction in histone acetylation. However, the reasons for other phenotypic variations remained obscure. Recent studies suggest an important functional correlation between DNA methylation and histone modifications. By investigating this relationship, we found that the global genomic DNA of atmbd9 was over-methylated, including the FLC gene region. Recombinant AtMBD9 does not have detectable DNA demethylation activity in vitro, but instead has histone acetylation activity. Ectopic over-expression of AtMBD9 and transient DNA demethylation promotes flowering and causes partial recovery of the normal branching phenotype. Co-immunoprecipitation assays suggest that AtMBD9 interacts in vivo with some regions of the FLC gene and binds to histone 4 (H4). Gene expression profile analysis revealed earlier up-regulation of some flower-specific transcriptional factors and alteration of potential hormonal and signal transducer axillary branching regulatory genes. In accordance with this result, AtMBD9 itself was found to be localized in the nucleus and expressed in the flower and axillary buds. Together, these results suggest that AtMBD9 controls flowering time and axillary branching by modulating gene expression through DNA methylation and histone acetylation, and reveal another component of the epigenetic mechanism controlling gene expression.

DNA Demethylation in Zebrafish Involves the Coupling of a Deaminase, a Glycosylase, and Gadd45

Evidence for active DNA demethylation in vertebrates is accumulating, but the mechanisms and enzymes remain unclear. Using zebrafish embryos we provide evidence for 5-methylcytosine (5-meC) removal in vivo via the coupling of a 5-meC deaminase (AID, which converts 5-meC to thymine) and a G:T mismatch-specific thymine glycosylase (Mbd4). The injection of methylated DNA into embryos induced a potent DNA demethylation activity, which was attenuated by depletion of AID or the non enzymatic factor Gadd45. Remarkably, overexpression of the deaminase/glycosylase pair AID/Mbd4 in vivo caused demethylation of the bulk genome and injected methylated DNA fragments, likely involving a G:T intermediate. Furthermore, AID or Mbd4 knockdown caused the remethylation of a set of common genes. Finally, Gadd45 promoted demethylation and enhanced functional interactions between deaminase/glycosylase pairs. Our results provide evidence for a coupled mechanism of 5-meC demethylation, whereby AID deaminates 5-meC, followed by thymine base excision by Mbd4, promoted by Gadd45.

Modest reactivation of the mutant FMR1 gene by valproic acid is accompanied by histone modifications but not DNA demethylation

Fragile X syndrome (FXS), the leading cause of inherited mental retardation, is due to expansion and methylation of a CGG sequence in the FMR1 gene, which result in its silencing. We previously demonstrated a reactivation of FMR1 in FXS cells treated with the DNA demethylating drug 5-azadeoxycytidine, and, to a lesser extent, with the histone deacetylating drug butyrate. To identify other reactivating drugs, we now treated three FXS lymphoblastoid cell lines with valproic acid (VPA), a well-known antiepileptic drug, causing histone deacetylase inhibition and, possibly, DNA demethylation. After VPA treatment, FMR1-mRNA levels were low and FMRP protein was undetectable. The gene remained methylated, whereas histones were acetylated and a modest variation of histone methylation was observed. These results confirm the histone hyperacetylating effect of VPA but do not support its putative DNA demethylation activity. The primary role of DNA demethylation in the reactivation of the FMR1 gene was confirmed.

Epigenetic alterations due to diet and Helicobacter pylori infection in gastric carcinogenesis

Sporadic gastric cancer is considered to be the result of a progressive accumulation of genotypic changes due to an adverse environment (i.e., diet and Helicobacter pylori infection). The main molecular mechanism implicated in cancer-related molecular alterations is of epigenetic nature, which includes DNA methylation and histone modification. Diet may influence the methylation status supplying methyl groups S-adenosyl-methionine formation, modifying DNA methyltransferase activity and influencing DNA demethylation activity. H. pylori may affect DNA methyltransferase directly or through inflammatory mediators (e.g., reactive oxygen species or nitric oxide). In conclusion, gastric cancer is a multistep process due to an adverse environment over a long period of time. Dietary habit and H. pylori infection can induce epigenetic alterations that, in turn, trigger gastric carcinogenesis.

Modulation of gene methylation by genistein or lycopene in breast cancer cells

Dietary agents with chemopreventive potential, including soy-derived genistein and tomato-derived lycopene, have been shown to alter gene expression in ways that can either promote or potentially inhibit the carcinogenic processes. To begin to explore the mechanisms by which these agents may be acting we have examined the DNA methylation modulating capacity of genistein or lycopene for several genes relevant to breast cancer in the breast cancer cell lines MCF-7 and MDA-MB-468, as well as in immortalized but noncancer fibrocystic MCF10A breast cells. We find using methylation specific PCR (MSP) that a low, nontoxic concentration of genistein (3.125 microM, resupplemented every 48 hr for 1 week) or a single dose of lycopene (2 microM) partially demethylates the promoter of the GSTP1 tumor suppressor gene in MDA-MB-468 cells. RT-PCR studies confirm a lack of GSTP1 expression in untreated MDA-MB-468, with restoration of GSTP1 expression after genistein or lycopene treatment. The RARbeta2 gene however, was not demethylated by genistein or lycopene in either of these breast cancer cell lines. But, lycopene (2 microM, once per week for 2 weeks) did induce demethylation of RARbeta2 and the HIN-1 genes in the noncancer MCF10A fibrocystic breast cells. These data show for the first time that the tomato carotenoid lycopene has direct DNA demethylating activity. In summary, both genistein and lycopene, at very low, dietarily relevant concentrations can potentially mitigate tumorigenic processes via promoter methylation modulation of gene expression.

Epigenetic Lesions in Malignant Melanoma

Malignant melanoma arises through a series of genetic and epigenetic events. A more profound understanding of the biology of metastatic melanoma should greatly aid in the development of new and effective treatments. Currently, avenues being pursued to improve treatment of metastatic melanoma include dendritic cell vaccines and other vaccination strategies, tyrosine kinase inhibitors, adoptive transfer of ex vivo stimulated T cells, and, as reviewed here, epigenetic approaches. The "methylator phenotype", with inactivation by promoter hypermethylation of numerous genes in malignant melanoma cell lines and primary tumors (p16, PTEN, RASSF1, estrogen receptor, retinoic acid receptor beta, SOCS1 and -2, MGMT etc.) offers a strong rationale for treatment approaches based on the use of DNA demethylating agents. The clinical literature on treatment of metastasized malignant melanoma with either 5-azacytidine or 5-aza-2'-deoxycytidine (decitabine) is reviewed. Future trials in malignant melanoma with these compounds might profit from prolonged low-dose exposure, since they unfold their full effects not immediately but with a certain delay, which may be associated with their DNA demethylating activity. Combinations of DNA demethylation agents with either histone deacetylase inhibitors, interleukin-2, chemotherapy or tamoxifen have been embarked on both in in vitro models of melanoma and recent clinical trials. The in vitro synergism between inhibitors of DNA methylation and histone deacetylation strongly invites a systematic study of combinations of both groups of agents. Up-regulation of cancer testis antigens by epigenetic therapy in melanoma also offers a very strong rationale to place these drugs and schedules within a larger treatment concept of immunotherapy which may include also T cell activation e.g. by interleukin-2, and vaccination strategies. In conclusion, the epigenome of malignant melanoma, with a well-established in vitro reversal potential, holds promise as a novel molecular target.

110 COMPARATIVE ANALYSIS OF DNA AND HISTONE H3 METHYLATION REPROGRAMMING IN MOUSE, BOVINE, AND RABBIT ZYGOTES

Active demethylation of paternal DNA in zygotes has been documented for many mammalian species, including mouse (Mayer et al. 2000 Nature 403, 501–505), rat (Zaitseva et al. 2007 Mol. Reprod. Dev. 74, 1255–1261), pig (Fulka et al. 2006 Histochem. Cell. Biol. 126, 213–217), Homo sapiens (Fulka et al. 2004 Reproduction 128, 703–708), and cow (Beaujean et al. 2004 Curr. Biol. 14, 266–267). The generality of this concept has been challenged by reports stating that there is no detectable genome-wide paternal DNA demethylation in rabbit (Shi et al. 2004 Biol. Reprod. 71, 340–347) and ovine (Beaujean et al. 2004 Curr. Biol. 14, 266–267), whereas others report partial evidence for the opposite (Zhang et al. 2005 Mol. Reprod. Dev. 72, 530–533). In order to compare the fate of DNA and histone H3 methylation (H3/K4triMe and H3/K9diMe), we performed studies on mouse, bovine, and rabbit zygotes using specific antibodies in an indirect immunofluorescence approach. We observed a similar distribution of these epigenetic modifications in maternal and paternal pronuclei of the zygote, indicating that the overall epigenetic reprogramming activities — including the DNA demethylation of the paternal pronucleus — are conserved between mammalian species. For all 3 species we found a clear asymmetry in mouse monoclonal antimethylcytosine (MeC) antibody signal intensities between paternal and maternal pronuclei at advanced stages, which suggests the presence of paternal DNA demethylation. To obtain additional evidence for the presence of DNA demethylation activity in rabbit zygotes, we examined 1-cell embryos after somatic cell nuclear transfer (SCNT) using rabbit fetal fibroblasts as nuclear donors, and we found a strong signal reduction in SCNT embryos 4 h after activation. For the majority of SCNT 1-cell embryos, we estimated at least a 4- to 6-fold decrease of MeC signal. Along with the similarities of DNA demethylation kinetics in rabbit, mouse, and bovine zygotes, we found a conserved distribution of H3/K4triMe and H3/K9diMe signals on maternal and paternal chromatin. In all 3 species, H3/K4triMe, which predominantly demarcate open chromatin, is present in both pronuclear chromatin, whereas H3/K9diMe, a hallmark of condensed chromatin, is strongly enriched in maternal pronuclear chromatin and undetectable in paternal chromatin. Furthermore, we observed a similar asymmetric compartmentalization of paternal and maternal chromosomes in early 2-cell rabbit and bovine embryos as monitored by H3/K9diMe, which marks maternally derived chromosomes. In summary, our data suggest that mechanisms of epigenetic reprogramming are conserved in mammalian species both on the level of DNA and chromatin methylation.

Reversal of p15/ INK4b hypermethylation in AML1/ETO-positive and -negative myeloid leukemia cell lines

In vitro and in vivo, myeloid leukemic and preleukemic cells exhibit variable sensitivity to the antiproliferative and proapoptotic effects induced already at low concentrations of DNA methyltransferase (DNMT) inhibitors. The molecular mechanisms underlying this variable sensitivity of leukemic blasts to azanucleosides such as 5-azacytidine and 5-aza-2′-deoxycytidine (DAC) may involve modifier effects of specific fusion proteins such as AML1/ETO. The cyclin-dependent kinase inhibitor p15/ INK4b is one potential target of DNA demethylating activity in AML and MDS where it is frequently silenced by hypermethylation. To study sensitivity to DAC in myeloid leukemia cells, we chose the myeloid cell lines Kasumi-1 (expressing AML1/ETO), KG-1 and KG-1a (both AML1/ETO-negative) all of which a highly methylated p15/ INK4b gene. Treatment with DAC resulted in dose-dependent regional demethylation of p15/ INK4b in Kasumi-1 and KG-1, but only to a modest degree in KG-1a cells. Demethylation was associated with induction of p15/INK4b protein expression. Growth-inhibitory and proapoptotic activity of DAC was significantly higher in Kasumi-1 than in KG-1a cells, and sensitization of cells to a cooperating effect of All-trans retinoic acid and of the histone deacetylase (HDAC) inhibitor Trichostatin A was observed. DAC-induced growth inhibition and apoptosis were enhanced when AML1/ETO was conditionally expressed in AML1/ETO-negative U-937 cells. In conclusion, hypomethylation and reactivation of p15/ INK4b in myeloid cell lines are among the molecular events associated with DAC-induced growth arrest and apoptosis. Further studies of AML1/ETO as a modifier of the epigenotype and sensitivity of myeloid cells to inhibitors of DNMTs and HDACs appear warranted.

Hydralazine target: From blood vessels to the epigenome

Hydralazine was one of the first orally active antihypertensive drugs developed. Currently, it is used principally to treat pregnancy-associated hypertension. Hydralazine causes two types of side effects. The first type is an extension of the pharmacologic effect of the drug and includes headache, nausea, flushing, hypotension, palpitation, tachycardia, dizziness, and salt retention. The second type of side effects is caused by immunologic reactions, of which the drug-induced lupus-like syndrome is the most common, and provides clues to underscoring hydralazine's DNA demethylating property in connection with studies demonstrating the participation of DNA methylation disorders in immune diseases. Abnormalities in DNA methylation have long been associated with cancer. Despite the fact that malignant tumors show global DNA hypomethylation, regional hypermethylation as a means to silence tumor suppressor gene expression has attracted the greatest attention. Reversibility of methylation-induced gene silencing by pharmacologic means, which in turns leads to antitumor effects in experimental and clinical scenarios, has directed efforts toward developing clinically useful demethylating agents. Among these, the most widely used comprise the nucleosides 5-azacytidine and 2'deoxy-5-azacytidine; however, these agents, like current cytotoxic chemotherapy, causes myelosuppression among other side effects that could limit exploitation of their demethylating properties. Among non-nucleoside DNA demethylating drugs currently under development, the oral drug hydralazine possess the ability to reactivate tumor suppressor gene expression, which is silenced by promoter hypermethylation in vitro and in vivo. Decades of extensive hydralazine use for hypertensive disorders that demonstrated hydralazine's clinical safety and tolerability supported its testing in a phase I trial in patients with cancer, confirming its DNA demethylating activity. Hydralazine is currently being evaluated, along with histone deacetylase inhibitors either alone or as adjuncts to chemotherapy and radiation, for hematologic and solid tumors in phase II studies.

27 CYTOPLASM OF A GERMINAL VESICLE OOCYTE CAN PROMOTE SOMATIC NUCLEAR REPROGRAMMING IN MICE

A number of studies have shown that epigenetic reprogramming is severely deficient in cloned embryos, the majority of which exhibit histone hypermethylation. Xenopus geminal vesicle (GV) oocytes have been reported to have a DNA demethylating activity (Simonsson and Gurdon 2004 Nat. Cell Biol. 6, 984–990). In an attempt to develop a new method for erasing or reprogramming the epigenetic status of the donor cell prior to nuclear transfer, we examined whether the mammalian GV oocyte cytoplasm can demethylate H3-K9 of somatic cell nuclei and improve the quality of reconstructed embryos. In the first series of experiments, cumulus nuclei were injected into enucleated GV oocytes and cultured for various times, 0 h, 3 h, 7 h, 9 h, 11 h, and 16 h, before examining the chromosome morphology of somatic nuclei together with Me-H3-K9. In the second series of experiments, permeabilized cumulus cells were immersed with GV oocyte extract and injected into enucleated MII oocytes. These reconstructed oocytes were activated and cultured until the blastocyst stage. Preparation of oocyte extracts was as follows: 200 GV oocytes were collected and zonae pellucidae removed using Tyrode's solution. These zona-free oocytes were broken down in 5 �L of HEPES medium containing the ATP-generating system (1 mM ATP, 10 mM creatine phosphate, 25 �g mL-1 creatine kinase, 100 �M GTP). Permeabilized cumulus was incubated in oocyte extract for 45 min at 37�C. The demethylation of H3-K9 was analyzed in nuclear transfer embryos at the pronucleus, 2-cell, 8-cell, morula, and blastocyst stages. In 2 experiments, we examined a total of 234 enucleated GV oocytes and 358 enucleated MII oocytes. The results of first experiment showed that histone H3-K9 begins demethylation in the somatic chromosome 1 h after injection into GV oocytes (90%) and, importantly, this state was maintained until the MII-like stage. The donor somatic chromosomes in the enucleated GV oocyte can condense and undergo stages GVI (0–3 h), MI (7 h), and AI-TI (9–11 h), and the MII (16 h)-like stage. These results clearly show that enucleated GV oocytes can carry out spindle assembly and extrude the first polar body. However, some of these chromosomes are located not only on the metaphase plate of the spindle but also on spindle poles or dispersed on the spindle. In the second experiment, reconstructed embryos with GV extract-treated somatic cells showed a strong H3-K9 demethylation that was significantly different to nontreated somatic cells at the blastocyst stage. The demethylation was expressed at a rate comparable to that observed in ICSI embryos. Although there was no increase in the frequency of development at the blastocyst stage, we found an increase in cell number at the blastocyst stage. Our findings show that the GV oocyte cytoplasm has the ability to erase methylation of H3-K9 in somatic nuclei, which suggests that the incorporation of GV mammalian oocyte components may contribute to the reprogramming of somatic cell nuclei.

Continued Low-Dose Decitabine (DAC) Is an Active First-Line Treatment of Older AML Patients: First Results of a Multicenter Phase II Study.

Background: Few therapeutic options exist for older AML patients (pts) in whom induction chemotherapy is not feasible. Encouraged by the marked activity and low toxicity of the demethylating agent DAC in high-risk MDS, we initiated a phase II trial in untreated AML patients >60 years not eligible for induction.Methods: low-dose DAC was given as for MDS (i.e. 135 mg/m2 i.v. over 72 hrs), repeated q 6 weeks for up to 4 courses, with all-trans retinoic acid (ATRA, 45mg/m2/day for 28 days) given during course 2 in DAC-sensitive pts. Maintenance with 20 mg/m2 DAC i.v. over 1 hour on 3 days (total dose 60mg/m2, outpatient administration) q 8 weeks was offered to pts completing all 4 courses. Pts with a WBC of >20 000/ul received a short course of hydroxyurea (HU) prior to DAC. The primary endpoint was best response: complete (CR) or partial remission (PR) or an antileukemic effect (AE, >25% bone marrow blast reduction). Pts requiring HU beyond day 28 of course 1 and/or showing blast increase had progressive disease (PD). Secondary endpoints were: overall (OS) and progression-free survival, toxicity and hospitalization duration. Accompanying studies included quality-of-life and geriatric assessment, p15/INK4b methylation analyses, and sequential measurement of HbF (a potential marker of DNA demethylating activity).Results: 51 pts have been recruited, with a median age of 72 years (range 63–85). 33% of pts were > 75 years. Complex karyotype and/or preceding MDS were present in 65 and 51 %, respectively. Median WBC before treatment was 5300/μl (range, 600–241,000, 33% of pts >20 000/μl WBC). Median bone marrow blasts were 70%. The median number of DAC courses given was 2, and 6 pts received a total of 30 maintenance courses (median 4.5). In the 29 fully evaluable pts, the best response was CR in 4 pts (14%), and PR in 5 pts (17%), with a median of 13 weeks to best response. An AE occurred in 9 pts (31%), resulting in a 62% overall response rate. Stable disease (SD) was seen in 3 pts (10%), 7 had PD (24%), 1 pt (3%) died early (day 9 from start DAC). Two pts (1 SD, 1 AE) were taken off study after 2 courses because of prolonged neutropenia. Toxicities of inpatient DAC were very similar to those described for MDS (neutropenia, fever/infection, pancytopenia). No unexpected toxicities or ATRA syndrome were observed with the combination of DAC+ATRA. No hospitalizations occured during maintenance DAC. Median OS from start of treatment was 7.5 months (range, 0.3–21+), the 1-year survival 24%. In CR+PR pts, median OS was not reached, in pts with AE was 7.8 months, in nonresponders 1 month. Geriatric assessment revealed a high maintenance level of capabilities during the study period.Conclusion: low-dose DAC is very well tolerated by older AML pts ineligible for more aggressive treatment, with myelosuppression being the major toxicity. Objective responses occur in 31 % of pts. A frequent and often prolonged antileukemic effect, limited hospitalization times, the occurrence of late responses, and the good feasibility of outpatient maintenance also support continued DAC treatment.

Clinical Development of Decitabine as a Prototype for an Epigenetic Drug Program

This review highlights decitabine as a prototype epigenetic modifying drug to show how the clinical development of epigenetic agents differs from that of traditional cytotoxic chemotherapies. Decitabine, a cytosine analogue, is cytotoxic at high doses but has selective DNA demethylating activity at low doses. The focus of current decitabine investigations is twofold: to elucidate all of the mechanisms of action and to determine the optimal dose, schedule, and concomitant therapies. New phase I trials have identified a "biologically effective dose," which is 1 to 2 logs lower than the cytotoxic dose. A clinical development program with low-dose decitabine in malignant diseases is focused on myelodysplastic syndrome (MDS), acute myelogenous leukemia (AML), and chronic myelogenous leukemia (CML). A phase III trial in MDS showed objective responses (complete [CR] plus partial [PR] remission) and longer median time to progression to AML or death with decitabine than with supportive care alone. The optimal use of decitabine may be in combination with other agents that promote gene expression, namely, histone deacetylase (HDAC) inhibitors. Optimized decitabine doses and combinations with other epigenetic therapies that can be used at minimally toxic doses provide potentially safer therapeutic options and introduce novel combination therapies.

Demethylating agent, 5-azacytidine, reverses differentiation of embryonic stem cells

The de novo methylation activity is essential for embryonic development as well as embryonic stem (ES) cell differentiation, where the intensive and extensive DNA methylation was detected. In this study, we investigated the effects of a demethylating agent, 5-azacytidine (5-AzaC), on differentiated ES cells in order to study the possibility of reversing the differentiation process. We first induced differentiation of ES cells by forming embryoid bodies, and then the cells were treated with 5-AzaC. The cells showed some undifferentiated features such as stem cell-like morphology with unclear cell-to-cell boundary and proliferative responsiveness to LIF. Moreover, 5-AzaC increased the expressions of ES specific markers, SSEA-1, and alkaline phosphatase activity as well as ES specific genes, Oct4, Nanog, and Sox2. We also found that 5-AzaC demethylated the promoter region of H19 gene, a typical methylated gene during embryonic differentiation. These results indicate that 5-AzaC reverses differentiation state of ES cells through its DNA demethylating activity to differentiation related genes.

Diet and DNA methylation interactions in cancer prevention.

Epigenetic events constitute an important mechanism by which gene function is selectively activated or inactivated. Since epigenetic events are susceptible to change they offer potential explanations of how environmental factors, including diet, may modify cancer risk and tumor behavior. Abnormal methylation patterns are a nearly universal finding in cancer, as changes in DNA methylation have been observed in many cancer tissues (e.g., colon, stomach, uterine cervix, prostate, thyroid, and breast). Site-specific alterations in DNA methylation have also been observed in cancer and may play a significant role in gene regulation and cancer development. This review presents intriguing evidence that part of the anticancer properties attributed to several bioactive food components, encompassing both essential nutrients and non-essential components, may relate to DNA methylation patterns. Four sites where dietary factors may be interrelated with DNA methylation are discussed. First, dietary factors may influence the supply of methyl groups available for the formation of S-adenosylmethionine (SAM). Second, dietary factors may modify the utilization of methyl groups by processes including shifts in DNA methyltransferase (Dnmt) activity. A third plausible mechanism may relate to DNA demethylation activity. Finally, the DNA methylation patterns may influence the response to a bioactive food component.

Human DNA-demethylating activity: a glycosylase associated with RNA and PCNA.

We have partially purified and characterized the 5-methylcytosine removing activity (5-meC-DNA Glycosylase) from HeLa cells with 700-fold enrichment. This activity cleaves DNA specifically at fully methylated CpG sites. The mechanism of 5-meC removal is base excision from fully methylated CpG loci on DNA, producing abasic sites. Hemi-methylated DNA is not a substrate. A prominent 52 KDa protein is present in all partially purified fractions. This activity is tightly associated with other nuclear factors and proteins, which resulted in differential fractionation of this activity on ion exchange columns. One nuclear factor associated with this activity is identified as RNA. Another nuclear protein, proliferating cell nuclear antigen (PCNA) is also associated with this enzyme. Glycosylic removal of 5-meC from DNA by this activity could be involved in the regulation of transcription, replication, differentiation, and development through resultant hypomethylation of DNA.

Ras Induces a General DNA Demethylation Activity in Mouse Embryonal P19 Cells

We demonstrate that expression of v-Ha-ras in mouse embryonal P19 cells results in genome-wide demethylation. Analysis of the pattern of methylation of specific genes reveals that different types of genes are demethylated in the ras transfectants: skeletal muscle specific genes, a gene specifically expressed in the adrenal cortex (c21), ubiquitous genes, and exogenously introduced sequences. Transient transfection and in vitro demethylation assays reveal that the ras transfectants express high levels of a general DNA demethylation activity. This demonstrates that the general DNA demethylation activity in mouse embryonal cells is controlled by an important cellular signal transducer and that DNA demethylase is a potential downstream effector of Ras.

Presence of a DNA demethylating activity in the nucleus of murine erythroleukemic cells.

Nucleoplasm prepared from murine erythroleukemia cells was assayed from enzymatic activity which removes methyl groups from DNA methylated at the internal C of the sequence 5'-CCGG-3' (Hpa II sites). Under the conditions of assay, the release of methyl groups was linear with time for 1 h, and up to 2 pmol could be released per microgram of extract protein. The activity was proportional to the amount of added extract and was sensitive to heat ant to proteinase K exposure. The initial velocity was hyperbolic with increasing amounts of substrate, and the Km for methyl groups was estimated to be between 1 and 1.5 microM. These results show that DNA demethylation catalyzed by extracts from cell nuclei can occur in vitro where no DNA synthesis is occurring and suggest that such a process might occur in vivo to alter DNA methylation during differentiation.