Portal de Periódicos da CAPES

A Unique Developmental Pattern of Oct-3/4DNA Methylation Is Controlled by a cis-demodification Element

2002; Elsevier BV; Volume: 277; Issue: 37 Linguagem: Inglês

10.1074/jbc.m203338200

1083-351X

Sharon Gidekel, Yehudit Bergman,

RNA modifications and cancer

Oct-3/4 is the earliest expressed transcription factor that is known to be crucial in murine pre-implantation development. In this report we asked whether methylation participates in controlling changes in Oct-3/4 expression and thus may play an important role in controlling normal embryogenesis. We show that the Oct-3/4 gene is unmethylated from the blastula stage but undergoes de novomethylation at 6.5 days post-coitum and remains modified in all adult somatic tissues analyzed. Oct-3/4 remains unmethylated in 6.25 days post-coitum epiblast cells when other genes, such as apoAI, undergo de novo methylation. We show that methylation of the Oct-3/4 promoter sequence strongly compromises its ability to direct efficient transcription. Moreover, DNA methylation inhibits basal transcription of the endogenous Oct-3/4 gene in vivo. We found that the Oct-3/4 gene harbors a cis-specific demodification element that includes the proximal enhancer sequence. This element leads to demethylation in embryonal carcinoma cells when the sequence is initially methylated and protects the local region from de novo methylation in post-implantation embryos. These results indicate that in the embryo protection from de novo methylation is not a unique feature of imprinted or housekeeping genes that carry a CpG island, but is also applicable to tissue-specific genes expressed during early stages of embryogenesis. Methylation of Oct-3/4 may be analogous to methylation of CpG islands on the inactive X chromosome that also occurs at later stages of development. Oct-3/4 is the earliest expressed transcription factor that is known to be crucial in murine pre-implantation development. In this report we asked whether methylation participates in controlling changes in Oct-3/4 expression and thus may play an important role in controlling normal embryogenesis. We show that the Oct-3/4 gene is unmethylated from the blastula stage but undergoes de novomethylation at 6.5 days post-coitum and remains modified in all adult somatic tissues analyzed. Oct-3/4 remains unmethylated in 6.25 days post-coitum epiblast cells when other genes, such as apoAI, undergo de novo methylation. We show that methylation of the Oct-3/4 promoter sequence strongly compromises its ability to direct efficient transcription. Moreover, DNA methylation inhibits basal transcription of the endogenous Oct-3/4 gene in vivo. We found that the Oct-3/4 gene harbors a cis-specific demodification element that includes the proximal enhancer sequence. This element leads to demethylation in embryonal carcinoma cells when the sequence is initially methylated and protects the local region from de novo methylation in post-implantation embryos. These results indicate that in the embryo protection from de novo methylation is not a unique feature of imprinted or housekeeping genes that carry a CpG island, but is also applicable to tissue-specific genes expressed during early stages of embryogenesis. Methylation of Oct-3/4 may be analogous to methylation of CpG islands on the inactive X chromosome that also occurs at later stages of development. embryonal stem promoter proximal enhancer embryonal carcinoma retinoic acid RA response element days post-coitum wild type whole cell extracts trichostatin A Development is a multistep process involving interactions of a large number of trans-acting factors with specificcis-regulatory elements leading to the activation and repression of many genes. Oct-3/4 is the earliest expressed transcription factor that is known to be crucial in murine pre-implantation development (1Scholer H.R. Trends Genet. 1991; 7: 323-329Abstract Full Text PDF PubMed Scopus (328) Google Scholar). The Oct-3/4gene is a member of the POU family of transcription factors; it is expressed in embryonal stem (ES)1 and in embryonal carcinoma (EC) cells (1Scholer H.R. Trends Genet. 1991; 7: 323-329Abstract Full Text PDF PubMed Scopus (328) Google Scholar, 2Herr W. Cleary M.A. Genes Dev. 1995; 9: 1679-1693Crossref PubMed Scopus (345) Google Scholar). Oct-3/4 expression is down-regulated in these cells upon induction to differentiate with retinoic acid (RA) (3Minucci S. Botquin V. Yeom Y.I. Dey A. Sylvester I. Zand D.J. Ohbo K. Ozato K. Scholer H.R. EMBO J. 1996; 15: 888-899Crossref PubMed Scopus (85) Google Scholar, 4Okamoto K. Okazawa H. Okuda A. Sakai M. Muramatsu M. Hamada H. Cell. 1990; 60: 461-472Abstract Full Text PDF PubMed Scopus (607) Google Scholar, 5Okazawa H. Okamoto K. Ishino F. Ishino Kaneko T. Takeda S. Toyoda Y. Muramatsu M. Hamada H. EMBO J. 1991; 10: 2997-3005Crossref PubMed Scopus (154) Google Scholar, 6Rosner M.H. Vigano M.A. Ozato K. Timmons P.M. Poirier F. Rigby P.W. Staudt L.M. Nature. 1990; 345: 686-692Crossref PubMed Scopus (754) Google Scholar, 7Scholer H.R. Dressler G.R. Balling R. Rohdewohld H. Gruss P. EMBO J. 1990; 9: 2185-2195Crossref PubMed Scopus (491) Google Scholar, 8Pikarsky E. Sharir H. Ben Shushan E. Bergman Y. Mol. Cell. Biol. 1994; 14: 1026-1038Crossref PubMed Scopus (105) Google Scholar). The Oct-3/4 gene is expressed throughout the pre-implantation embryo (6Rosner M.H. Vigano M.A. Ozato K. Timmons P.M. Poirier F. Rigby P.W. Staudt L.M. Nature. 1990; 345: 686-692Crossref PubMed Scopus (754) Google Scholar, 9Palmieri S.L. Peter W. Hess H. Scholer H.R. Dev. Biol. 1994; 166: 259-267Crossref PubMed Scopus (520) Google Scholar). Oct-3/4 protein is present in unfertilized oocytes, and the zygotic expression is activated prior to the 8-cell stage (9Palmieri S.L. Peter W. Hess H. Scholer H.R. Dev. Biol. 1994; 166: 259-267Crossref PubMed Scopus (520) Google Scholar, 10Pesce M. Wang X. Wolgemuth D.J. Scholer H. Mech. Dev. 1998; 71: 89-98Crossref PubMed Scopus (406) Google Scholar, 11Yeom Y.I., Ha, H.S. Balling R. Scholer H.R. Artzt K. Mech. Dev. 1991; 35: 171-179Crossref PubMed Scopus (92) Google Scholar). In the embryo Oct-3/4 is abundantly and uniformly expressed in all cells through the morula stage (11Yeom Y.I., Ha, H.S. Balling R. Scholer H.R. Artzt K. Mech. Dev. 1991; 35: 171-179Crossref PubMed Scopus (92) Google Scholar). Its expression is down-regulated in trophectoderm cells and becomes restricted to cells of the inner cell mass in the blastocyst (9Palmieri S.L. Peter W. Hess H. Scholer H.R. Dev. Biol. 1994; 166: 259-267Crossref PubMed Scopus (520) Google Scholar). Oct-3/4 expression is high and persists through day 7.5 in the unsegmented presomitic mesoderm, decreasing anteriorly to posteriorly as the somatic lineages form (7Scholer H.R. Dressler G.R. Balling R. Rohdewohld H. Gruss P. EMBO J. 1990; 9: 2185-2195Crossref PubMed Scopus (491) Google Scholar). From day 8.5, expression cannot be detected in any somatic tissue but is restricted to the premigratory progenitor germ cells (9Palmieri S.L. Peter W. Hess H. Scholer H.R. Dev. Biol. 1994; 166: 259-267Crossref PubMed Scopus (520) Google Scholar, 12Yeom Y.I. Fuhrmann G. Ovitt C.E. Brehm A. Ohbo K. Gross M. Hubner K. Scholer H.R. Development. 1996; 122: 881-894Crossref PubMed Google Scholar). Recently, using conventional gene targeting technology, it was found that the activity of Oct-3/4 is essential for the identity of the pluripotential founder cell population in the inner cell mass (13Nichols J. Zevnik B. Anastassiadis K. Niwa H. Klewe Nebenius D. Chambers I. Scholer H. Smith A. Cell. 1998; 95: 379-391Abstract Full Text Full Text PDF PubMed Scopus (2627) Google Scholar). Furthermore, it was shown that a critical amount of Oct-3/4 is required to sustain stem cell self-renewal, and any up- or down-regulation induces divergent developmental programs (14Niwa H. Miyazaki J. Smith A.G. Nat. Genet. 2000; 24: 372-376Crossref PubMed Scopus (2857) Google Scholar). A very modest modulation of Oct-3/4 levels in ES cells results in a dramatic readout. Thus, it is now recognized that the mere presence of Oct-3/4 protein does not define pluripotency because it is the level of expression that is the key to regulation. One of the molecular mechanisms that regulate gene expression is methylation (15Bird A. Genes Dev. 2002; 16: 6-21Crossref PubMed Scopus (5254) Google Scholar). DNA methylation plays a crucial role in normal embryogenesis by acting in cis to modulate protein-DNA interactions. Several studies have demonstrated that DNA methylation patterns go through dynamic changes during embryogenesis. A global loss of DNA methylation occurs in the mammalian preimplantation embryo prior to the 16-cell morula stage, and the DNA remains unmethylated throughout blastulation (16Kafri T. Ariel M. Brandeis M. Shemer R. Urven L. McCarrey J. Cedar H. Razin A. Genes Dev. 1992; 6: 705-714Crossref PubMed Scopus (586) Google Scholar, 17Monk M. Boubelik M. Lehnert S. Development. 1987; 99: 371-382Crossref PubMed Google Scholar). Following implantation, an extensive wave of de novo methylation occurs, and only genes containing CpG islands, such as housekeeping genes, escape this process (16Kafri T. Ariel M. Brandeis M. Shemer R. Urven L. McCarrey J. Cedar H. Razin A. Genes Dev. 1992; 6: 705-714Crossref PubMed Scopus (586) Google Scholar, 18Brandeis M. Frank D. Keshet I. Siegfried Z. Mendelsohn M. Nemes A. Temper V. Razin A. Cedar H. Nature. 1994; 371: 435-438Crossref PubMed Scopus (620) Google Scholar, 19Macleod D. Charlton J. Mullins J. Bird A.P. Genes Dev. 1994; 8: 2282-2292Crossref PubMed Scopus (514) Google Scholar). At a later stage of development, tissue-specific genes undergo selective demethylation (20Lichtenstein M. Keini G. Cedar H. Bergman Y. Cell. 1994; 76: 913-923Abstract Full Text PDF PubMed Scopus (173) Google Scholar, 21Mostoslavsky R. Bergman Y. Biochim. Biophys. Acta. 1997; 1333: F29-F50PubMed Google Scholar). Because Oct-3/4 plays a crucial role in regulating initial differentiation decisions in early development and has a unique expression pattern, we studied the dynamics of the generation of its methylation pattern and the role of methylation in controllingOct-3/4 expression. We have shown previously (22Ben Shushan E. Pikarsky E. Klar A. Bergman Y. Mol. Cell. Biol. 1993; 13: 891-901Crossref PubMed Scopus (68) Google Scholar) that treatment of EC cells with RA reducesOct-3/4 expression that is associated with increased methylation and changes in chromatin structure in the immediate upstream regulatory region that includes the promoter (P) and proximal enhancer (PE) elements. In this study we show that in vivo the Oct-3/4 gene is unmethylated from the blastula stage and starts to undergo de novo methylation at 6.5 dpc. The Oct-3/4 gene remains modified in all adult somatic organs analyzed. We show that during embryogenesisOct-3/4 remains unmethylated at a time when genome wide methylation occurs; thus it was interesting to find out whether it carries a cis-demethylation element that protects the gene from global modification mechanisms. By using a series of stable transfection assays into EC cells, we have found that the Oct-3/4 gene does harbor acis-specific demodification element that includes theOct-3/4 PE sequence. Furthermore, mutagenesis at the protein-binding sites of this sequence clearly affected its demethylation activity. By using a mini-transgene containing the PE sequences only, we have shown that in the embryo the PE protects the DNA from de novo methylation in 6.25 dpc epiblast cells, and when it is mutated the sequence becomes methylated. In vitromethylation of the Oct-3/4 upstream sequence inhibits its ability to induce transcription. Moreover, we have shown that DNA methylation inhibits the endogenousOct-3/4 basal expression in vivo. Murine P19 (23McBurney M.W. Rogers B.J. Dev. Biol. 1982; 89: 503-508Crossref PubMed Scopus (348) Google Scholar) EC cells and L8 cells were maintained as described previously (8Pikarsky E. Sharir H. Ben Shushan E. Bergman Y. Mol. Cell. Biol. 1994; 14: 1026-1038Crossref PubMed Scopus (105) Google Scholar). p53−/− and p53−/− Dnmt1−/− fibroblasts were grown from 9 dpc embryos obtained by mating p53−/−Dnmt1−/+ mice. These cells (grown in Dulbecco's modified Eagle's medium containing 15% fetal calf serum) appear to be immortal, still not having undergone senescence even after 100 passages. Nearest neighbor analysis (24Naveh-Many T. Cedar H. Proc. Natl. Acad. Sci. U. S. A. 1981; 78: 4246-4250Crossref PubMed Scopus (100) Google Scholar) revealed that less than 5% of CpG residues in these cells are methylated, as compared with 73% for normal fibroblasts. These cells were treated with 50 ng/ml TSA for 24 h in order to study Oct-3/4 transcription. The constructs 342 and pOct-luc have been described previously (see Refs. 25Siegfried Z. Eden S. Mendelsohn M. Feng X. Tsuberi B.Z. Cedar H. Nat. Genet. 1999; 22: 203-206Crossref PubMed Scopus (279) Google Scholar and 26Sylvester I. Scholer H.R. Nucleic Acids Res. 1994; 22: 901-911Crossref PubMed Scopus (73) Google Scholar, respectively). pOctPE-luc was constructed by cloning the PCR amplification product spanning positions −1215 to +155 of theOct-3/4 upstream region containing the PE and P sequences into the PstI site of pOct-luc. The pre-existing P was deleted by digestion with HindIII- and the plasmid was religated. The primers used were Nsi 2 and Nsi 3. The following plasmids were constructed by digesting the different constructs with NsiI and subcloning them into thePstI site in plasmid 342. pPEwt was constructed by cloning the PCR amplification product spanning positions −1215 to −900 of the Oct-3/4 PE region. The primers used were Nsi 2 and Nsi 1. pPEPwt was constructed by cloning the PCR amplification product spanning the −1215 to +155 positions of theOct-3/4 PE and P region. The primers used were Nsi 2 and Nsi 3. pPE1A* construct is identical to the above described pPEwt plasmid except for a 16-bp mutation inserted in the 1A site. Two distinct PCR amplification reactions were done on pBluescript KS(−) containing the −1215 to −541 BamHI-SauIIIA upstream region of the Oct-3/4 gene. The primers used were RARE1A and Primer 2 and RARE1Arev and Primer 1. Products were mixed, amplified with Primers 1 and 2, digested withBamHI/ScaI, and subcloned intoBamHI/EcoRV sites in pBluescript KS(−) to create pmRARE1A. pPE1A* was constructed by cloning the PCR amplification product of pmRARE1A using the primers Nsi 1 and Nsi 2 into thePstI site in 342. pPE1B* was constructed by cloning the PCR amplification product spanning positions −1215 to −942 of the Oct-3/4 PE region, harboring a deletion of the 1B site. The primers used were Nsi 2 and Nsi 4. pPE1A*1B* construct is identical to the above described pPE1B* plasmid except for a 16-bp mutation inserted in the 1A site. pPE1A*1B* was constructed by cloning the PCR amplification product of pmRARE1A using the primers Nsi 2 and Nsi 4. pPEP1A*1B* construct is identical to the above described pPEPwt plasmid except for a 16-bp mutation inserted in the 1A site and a 28-bp deletion of the 1B site. pmRARE1A was subjected to PCR amplification with Primer 2 and Primer RARE1Brev; the product was digested withBamHI/ScaI and ligated toScaI/BamHI fragment from pPEPwt. The ligation fragment was subjected to PCR amplification using the primers Nsi 2 and Nsi 3 and subcloned into the PstI site in 342 to create pPEP1A*1B*. Primers and PCR amplification conditions are described in Table I.Table IPrimers for plasmid constructionsNameSequenceNsi 15′-TGCATGCATAAACAAGTACTCAACCC-3′Nsi 25′-TGCATGCATGGATCCTCAGACTGGG-3′Nsi 35′-TGCATGCATGGATCCACCCAGCCC-3′Nsi 45′-TGCATGCATGGGGACAACTTCCTGCTCCC-3′RARE1A5′-CCGTGCAGATCTCGAGCTTCCTGTGCTGGCGG-3′RARE1Arev5′-GCTCGAGATCTGCACGGACGAGGATGAACACCGG-3′Primer 15′-CCAGGGTTTTCCCAGTCACG-3′Primer 25′-GCCAAGCTCGGAATTAACCC-3′RARE1Brev5′-AAAAGTACTCCCCTGGGGACAACTTCC-3′All PCR amplification reactions were carried out as follows: 2 min at 94 °C; 24 cycles of 1 min at 94 °C, 30 s at 60 °C, and 30 s at 72 °C, and finally 3 min at 72 °C. All amplified fragments were checked by sequencing. Open table in a new tab All PCR amplification reactions were carried out as follows: 2 min at 94 °C; 24 cycles of 1 min at 94 °C, 30 s at 60 °C, and 30 s at 72 °C, and finally 3 min at 72 °C. All amplified fragments were checked by sequencing. Plasmid DNAs were methylated in vitro as described previously (27Kirillov A. Kistler B. Mostoslavsky R. Cedar H. Wirth T. Bergman Y. Nat. Genet. 1996; 13: 435-441Crossref PubMed Scopus (208) Google Scholar). P19 and L8 cells were transfected by the calcium phosphate precipitation method (28Wigler M. Pellicer A. Silverstein S. Axel R. Urlaub G. Chasin L. Proc. Natl. Acad. Sci. U. S. A. 1979; 76: 1373-1376Crossref PubMed Scopus (834) Google Scholar). Cells (5 × 105) were plated 24 h before transfection and cotransfected with 15 μg of methylated plasmid together with 1.5 μg of a plasmid carrying the neoselection gene. Stably transfected cells were selected by G418 (250 mg per ml, Invitrogen), and clonally propagated. Pre-implantation embryos were obtained as described previously (29Birger Y. Shemer R. Perk J. Razin A. Nature. 1999; 397: 84-88Crossref PubMed Scopus (107) Google Scholar). The post-implantation embryos were dissected from the uterus of the foster mothers at dpc 6.25, 6.5, 7.5, 8.5, 12.5, or 14.5. Genomic DNAs were extracted from the following groups: 1) a pool of 10–20 preimplantation blastocysts; 2) a pool of epiblast cells dissected from 7 embryos of 6.25 dpc; 3) pools of post-implantation embryos at stages 6.5, 7.5, 8.5, 12.5, and 14.5 dpc. DNA samples were divided into 3 aliquots that were subjected to digestion with PvuII alone and PvuII together with HpaII or HhaI. The PCRs were carried out with 5 ng of DNA using the following primers: for analyzingHpaII sites 1–4 primers Hp1a and Hp1b, Hp2a and Hp2b, Hp3a and Hp3b, Hp4a and Hp4b were used, respectively. In order to analyze the HhaI site, primers Hp3a and Hp4b were used. To assay the Jκ region that lacks HpaII or HhaI sites, we used 5GL-2 and Jκ1110 primers. Primers and PCR amplification conditions are described in Table II.Table IIPrimers for methylation analysis and transgenic mice identificationNameSequenceHp1a5′-CCTCCTAATCCCGTCTCC-3′Hp1b5′-CCAGCTCTCCACCTCTCC-3′Hp2a5′-CCCAGTATTTCAGCCCATGTCC-3′Hp2b5′-GTTAGAGCTGCCCCTCTG-3′Hp3a5′-GGCACACGAACATTCAATGG-3′Hp3b5′-GGAGAAACTGAGGCGAGCGC-3′Hp4a5′-GGGATTGGGGAGGGAGAGG-3′Hp4b5′-GGTGGGGGTGAGAAGGCG-3′5GL-25′-CTTTCGCCTACCCACTGCTC-3′Jκ11105′-CCCTCCGAACGTGTACACAC-3′Dra1–3′5′-CCTTGATTTCAGAATCTTGCC-3′All PCR amplification reactions were carried out as follows: 2 min at 95 °C; 29 cycles of 45 s at 95 °C, 30 s at 56 °C, and 45 s at 72 °C, and finally 5 min at 72 °C. The PCR products were electrophoresed on 1–3% agarose gels, stained with ethidium bromide, and photographed. Open table in a new tab All PCR amplification reactions were carried out as follows: 2 min at 95 °C; 29 cycles of 45 s at 95 °C, 30 s at 56 °C, and 45 s at 72 °C, and finally 5 min at 72 °C. The PCR products were electrophoresed on 1–3% agarose gels, stained with ethidium bromide, and photographed. HpaII and HhaI sites in the first intron of the apoAI gene were studied using "fragment I" primers as described previously (16Kafri T. Ariel M. Brandeis M. Shemer R. Urven L. McCarrey J. Cedar H. Razin A. Genes Dev. 1992; 6: 705-714Crossref PubMed Scopus (586) Google Scholar). WCEs and electrophoretic mobility shift assays (EMSAs) were done as described previously (30Ben Shushan E. Thompson J.R. Gudas L.J. Bergman Y. Mol. Cell. Biol. 1998; 18: 1866-1878Crossref PubMed Scopus (215) Google Scholar). 1Awt and mut1A oligonucleotides (described in TableIII) were used for EMSAs.Table IIIOligonucleotides for EMSAsNameDescriptionSequence1Awtwt Oct-3/4 PE 1AGCACAGGAATGGGGGAGGGGTGGGTGACGAGGmut1A16 nucleotides mutations in the Oct-3/4 PE 1AGCACAGGAAgctcgagatctgcacgGACGAGG Open table in a new tab The differentially methylated pOct-luc or pOctPE-luc constructs were transiently cotransfected with the pCMV-Renilla luciferase construct into P19 EC cells using the calcium phosphate precipitation method (28Wigler M. Pellicer A. Silverstein S. Axel R. Urlaub G. Chasin L. Proc. Natl. Acad. Sci. U. S. A. 1979; 76: 1373-1376Crossref PubMed Scopus (834) Google Scholar). Luciferase activity was determined 48 h post-transfection (Dual-luciferase Reporter Assay system kit, Promega) and was corrected for transfection efficiency by measuring the Renilla luciferase activity. Total RNA was collected from the cells with Trizol (Sigma) following the manufacturer's instructions. Reverse transcription was carried out following the manufacturer's protocol (Promega). RT-PCR analysis of Oct-3/4 transcription in p53−/−, p53−/−/Dnmt1−/−, and blastocyst cells were carried out using the primers 251L and Oct4-a. PCR was performed using 200 mm of the four dNTPs, 100 ng of each primer, and 0.1 μl of [32P]dCTP. A second half-nested PCR was carried out in the same reaction conditions using the following primers: Oct4–5′ and Oct4-a. One-third of the reaction products was subjected to electrophoresis on a 7.2% polyacrylamide gel. Primers and PCR amplification conditions are described in TableIV.Table IVPrimers for RT-PCR analysisNameSequence251L5′-ATGGCATACTGTGGACCTCA-3′Oct4-a5′-CCTGGGAAAGGTGTCCTGTA-3′Oct4–5′5′-CAGGCAGGAGCACGAGTGGAA-3′Amplification was performed as follows: 4 min at 94 °C, 40 cycles of 30 s at 94 °C, 30 s at 62 °C, and 1 min at 72 °C, and, finally, 5 min at 72 °C. Open table in a new tab Amplification was performed as follows: 4 min at 94 °C, 40 cycles of 30 s at 94 °C, 30 s at 62 °C, and 1 min at 72 °C, and, finally, 5 min at 72 °C. The 1141- and 1099-bp DraI fragments, harboring the wt or double mutated Oct-3/4 PE sequences, were isolated from pPEwt and pPE1A*1B* plasmids, respectively. Fragments were purified from low melting point agarose gels by a gel extraction kit (Qiagen). DNA samples were microinjected into the pronucleus of (C57BL/6 × BALB/c)F1 fertilized mouse eggs and transferred into pseudo-pregnant CB6/F1 foster mother. Transgenic mice were identified by PCR amplification using primersDraI-3′ and Hp1b (Table II). Epiblast cells were isolated from 6 wt PEwt and 13 double mutated PE1A*1B* 6.25 dpc transgenic embryos as described (31Sturm K. Tam P.P. Methods Enzymol. 1993; 225: 164-190Crossref PubMed Scopus (94) Google Scholar). DNA from the wt and mutated pools of embryos was extracted, and 10 ng were subjected to digestion with PvuII in the presence or absence of HpaII. Products were analyzed by PCR amplification using primers DraI-3′ and Hp1b (Table II) flankingHpaII site 1 in the Oct-3/4 PE. As a control, we assayed the Jκ region that lacks HpaII sites using 5GL-2 and Jκ1110 primers. Primers and PCR amplification conditions are described in Table II. Four HpaII and oneHhaI sites were analyzed in theOct-3/4 1370-bp upstream region (Fig.1A). To study the methylation status of the Oct-3/4 gene in early mouse embryos, genomic DNA was isolated from mouse embryos at various stages of development. We adopted a PCR-based assay in which genomic DNA was subjected to digestion with or without HpaII orHhaI methyl-sensitive restriction enzymes. Digestion conditions were carefully titrated. After digestion, oligonucleotide primers flanking each site were used. A PCR product is observed only when the site is methylated and refractory to digestion. It was shown previously (16Kafri T. Ariel M. Brandeis M. Shemer R. Urven L. McCarrey J. Cedar H. Razin A. Genes Dev. 1992; 6: 705-714Crossref PubMed Scopus (586) Google Scholar) that when properly calibrated, this assay is linear over a wide range of DNA concentrations and can be used to measure accurately the degree of DNA methylation at specific sites. By using this PCR assay, we show that the HhaI and the fourHpaII sites tested are clearly unmethylated in blastocysts and in epiblast cells isolated from 6.25 dpc embryos. These sites become methylated in 6.5 dpc embryos almost to the full extent (Fig.1B). Control experiments were included to validate our results. Tail DNA and DNA from P19 cells served as controls for methylated and unmethylated DNA, respectively. Moreover, as an internal control, a primer pair that amplifies a Jκ sequence that does not encompass either HpaII or HhaI sites was included in each reaction. As expected, the fragment was amplified from all DNA samples to a similar level (Fig. 1B). We have quantified the methylation status of each site by PhosphorImager analysis and found that different sites undergo modification to various levels, and for most of these sites the level of methylation is sustained throughout embryogenesis (Fig. 1D). Similar results were obtained from quantifications of Southern blots containing DNA isolated from 6.5, 8.5, 12.5, and 14.5 dpc embryos, P19 cells, and murine tails (data not shown). In order to investigate whether this pattern of modification serves as the prototype for the basic stable methylation pattern in somatic cells, we analyzed the methylation status of theOct-3/4 upstream regulatory region in adult kidney and tail DNA. As can be seen in Fig. 1D, the cleavage patterns of all sites analyzed of both DNA samples were roughly similar to those found in 8.5-day embryos. Results from PhosphorImager analysis of the corresponding Southern blots were graphed (Fig. 1E), and it is interesting to note that the HhaI site located in the P region and the juxtaposed HpaII site 3 are almost fully methylated, serving as a core methylation center, whereas sites flanking this center (HpaII sites 1, 2, and 4) are partially methylated. These experiments clearly show that the methylation pattern established during development faithfully propagates itself to adulthood. The apoA1 gene, similar to other tissue-specific genes analyzed, undergoes de novo methylation due to the onset of global methylation that occurs at post-implantation embryos (16Kafri T. Ariel M. Brandeis M. Shemer R. Urven L. McCarrey J. Cedar H. Razin A. Genes Dev. 1992; 6: 705-714Crossref PubMed Scopus (586) Google Scholar, 17Monk M. Boubelik M. Lehnert S. Development. 1987; 99: 371-382Crossref PubMed Google Scholar). As can be seen in Fig. 1C, the apoA1 sequences are methylated in the 6.25 dpc epiblast cells, at the time whenOct-3/4 is still unmethylated. Thus, theOct-3/4 gene remains unmethylated at a time when the apoA1 gene undergoes de novo methylation. One possible mechanism to explain the ability of the Oct-3/4gene to escape the first de novo methylation wave is the involvement of a cis-element that either actively demethylates the newly methylated Oct-3/4 or protects it from de novo methylation. At this stage of our research we cannot differentiate between these two alternatives, and for simplicity reasons this element will be designated a protection/demethylation or demodification element. In the first series of experiments we wished to determine whether a long (1370 bp) fragment from the upstream regulatory element of theOct-3/4 gene could bring about demethylation of an in vitro methylated substrate when integrated into the genome of P19 EC cells. This element contains the P, intervening, and PE sequences. We inserted this fragment into a vector (25Siegfried Z. Eden S. Mendelsohn M. Feng X. Tsuberi B.Z. Cedar H. Nat. Genet. 1999; 22: 203-206Crossref PubMed Scopus (279) Google Scholar) that was engineered to harbor a non-CpG island sequence as well as recognition sites for methylation-sensitive restriction endonucleases (designated pPEPwt, Figs. 1A and2A). We generated stably transfected cell lines containing the above described in vitro methylated (by HpaII and HhaI methylases) construct and examined the methylation status of the insert as well as the vector sequences. To minimize the effect of particular integration sites and copy number, we analyzed more than 10 independent clones for each of the constructs tested and chose those with a similar copy number. The sites that were analyzed were the same HpaII and HhaI sites that were analyzed in the endogenous Oct-3/4 gene (Fig. 1) and an extra HhaI site in the vector sequence (Fig.2A). All sites analyzed were undermethylated (Fig. 2B, lanes 2 and 3), indicating that this fragment harbors a demodification element that is able to demethylate sequences inside and outside the fragment. Next we analyzed the ability of the P alone, the proximal enhancer (PE) alone, and the P and PE elements without the intervening sequences to induce demethylation. It was clear that both the P and the intervening sequences are dispensable for the demethylation process (data not shown). In striking contrast, we found that the PE fragment (pPEwt) plays a key role in directing the demethylation reaction (Fig.2B, compare lanes 4 and 5). This PE was previously found to be a critical element for P19 EC cell-specific expression and for RA-mediated down-regulation of the Oct-3/4 protein (1Scholer H.R. Trends Genet. 1991; 7: 323-329Abstract Full Text PDF PubMed Scopus (328) Google Scholar, 2Herr W. Cleary M.A. Genes Dev. 1995; 9: 1679-1693Crossref PubMed Scopus (345) Google Scholar). We wished to determine whether the Oct-3/4demodification element is specific for EC cells or alternatively functions in other cells as well. We transfected in vitromethylated pPEwt into L8 fibroblast cells and performed Southern blot analysis. This analysis showed that the integrated construct was resistant to HpaII and HhaI digestion and therefore maintained methylation at these sites (Fig. 2C, compare lane 1to lanes 3and 4). Taken together, these experiments clearly indicate that theOct-3/4 PE is required for directing demodification and for mediating demethylation of theOct-3/4 gene in a cell-specific manner. This cell specificity encouraged us to delineate the demodification element responsible for Oct-3/4 demethylation in EC cells. It was shown previously (3Minucci S. Botquin V. Yeom Y.