| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
B Ligand (RANKL) Gene Promoter Modulates Cell- and Tissue-Specific RANKL Expression and Osteoclastogenesis
Division of Molecular Pathology, Kobe University Graduate School of Medicine, Kobe 650-0017, Japan
Address all correspondence and requests for reprints to: Sohei Kitazawa, M.D., Ph.D., Division of Molecular Pathology, Kobe University Graduate School of Medicine, 7-5-1 Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan. E-mail: kitazawa{at}med.kobe-u.ac.jp.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
B ligand (RANKL) expression is tissue specific and limited to certain subsets of T-lymphocytes and stromal/osteoblastic cells. Even among osteoblasts, RANKL is expressed on about 20% of osteoblasts of the normal mouse. To clarify the mechanism of population-specific RANKL expression, we analyzed the effect of CpG methylation on its transcription, mRNA and protein expression as well as on osteoclastogenesis. Subpopulations of ST2 cells were used: P9, which expresses RANKL and supports osteoclastogenesis, and P16, which does not. By sodium bisulfite mapping, the rate of CpG methylation of the –65/+350 region, especially of CpG locus no. 1 three bases upstream of the TATA-box, was higher in P16 than in P9 ST2 cells. ChIP and gel shift assay showed that methylated CpG locus no. 1 was a target of MeCP2 binding that, in turn, blocked the binding of the TATA-box binding protein to the TATA-box. In vitro methylation by SssI of the promoter construct reduced its transcriptional activity at the steady state and its response to 1
,25(OH)2 vitamin D3. Conversely, treatment with DNA methylase inhibitor, 5-aza-2'-deoxycytidine, significantly restored RANKL expression and osteoclastogenesis in P16 cells. Except for primary cultured osteoblasts, CpG locus no. 1 was frequently methylated in various normal mouse tissues. We propose that the methylation status of the CpG locus three bases upstream of the TATA-box modulates the control of cell- and tissue-specific expression of RANKL gene and osteoclastogenesis. The heterogeneity of stromal/ osteoblastic cells in response to bone-resorbing stimuli may be attributed, in part, to the methylation status of the RANKL gene promoter. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
B ligand (RANKL) expressed on stromal cells and osteoblasts (7, 8, 9), and its cognate receptor, RANK (9, 10), expressed at high levels on osteoclast precursors. These observations have provided a solid molecular understanding of the coupling between osteoblastic bone formation and osteoclastic bone resorption. RANKL gene expression is, on the other hand, tissue-specific and limited to certain subsets of T-lymphocytes and stromal/ osteoblastic cells (11). Even among stromal/osteoblastic cells, RANKL-expressing cells are heterogeneously distributed, comprising about 20% of otherwise morphologically indistinguishable stromal/ osteoblastic cells lining the bone surface in normal mouse bone tissue, as revealed by our previous in situ hybridization studies (12, 13). Probably reflecting such heterogeneity and selected populations of RANKL-expressing cells, bone formation and resorption, during remodeling, takes place not as a generalized event but as a localized packet of bone, called bone-remodeling unit (14), whose distribution density and pattern ultimately define the response to mechanical loading, bone-seeking steroids and other cytokines (2). To explain cell- and tissue-specific expression of the RANKL gene, the contribution of tissue-specific transcription factors such as Runx2 (15), Sp1 and 3 (16), and activating transcription factor (ATF)-4 (17) has been postulated. Additionally, a higher chromatin structure including an epigenetic conformation and other far upstream elements are also assumed to be required for full induction and tissue-specific RANKL gene expression (18, 19).
In this study, because epigenetic regulation by CpG methylation at the promoter region is often associated with both tissue-specific and diverse or heterogeneous expression of genes (20, 21), we analyzed the contribution of CpG methylation of the promoter and that of MeCP2, a major methyl-CpG binding protein, in mouse RANKL gene expression.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
,25(OH)2 vitamin D3 (vitamin D3) decreased significantly in the subpopulations after passage 12; tartrate-resistant acid phosphatase (TRACP)-positive multinucleated cells were hardly generated after passage 14 (Fig. 1A
). RANKL and osteoprotegerin (OPG) mRNA expression in representative subpopulations, P9 and P16, was then assessed by quantitative real-time RT-PCR. Mirroring the results of osteoclastogenesis in the coculture, vitamin D3 caused a 17-fold increase of RANKL mRNA in P9 but not in P16: it reduced OPG mRNA expression equally in both P9 and P16 ST2 cells (Fig. 1B). Reduction in the osteoclastogenesis of P16 correlated with the lack of responsiveness of the RANKL gene to vitamin D3
|
). Because mainly VDR-retinoid X receptor (RXR) ß heterodimer binds to vitamin D-responsive element (VDRE) (–937/–922) of the mouse RANKL gene promoter (22), the binding of VDR-RXRß and the oligonucleotide spanning VDRE in P9 and P16 ST2 cells was assessed by EMSA. In both P9 and P16, protein DNA binding, supershift by anti-VDR antibody and blockshift by anti-RXRß antibody were also equally demonstrated (Fig. 2B).
|
). The transiently introduced transcriptional activity of promoter constructs was assessed in P9 and P16 ST2 cells with or without vitamin D3 treatment. Each of the promoter constructs showed almost the same level of activity at the steady state. The pGL3-1005 construct containing VDRE demonstrated increased luciferase activity in response to vitamin D3, but pGL3–723 lacking VDRE did not. Thus, in vitro, the exogenously transfected RANKL gene promoter constructs showed similar transcriptional activity in both P9 and P16.
Methylation Status of CpG Loci and Chromatin Acetylation of the Mouse RANKL Gene Promoter in P9 and P16 ST2 Cells
Because quantitative or functional deterioration of VDR was not observed in P16 ST2 cells, the cis-regulating mechanism of the endogenous RANKL gene promoter was examined. In the 5'-flanking region of the mouse RANKL gene, clusters of CpG loci were identified in region A (9 CpG loci downstream of VDRE) and in region B (39 CpG loci around the transcription and the translation start sites). Bisulfite mapping using genomic DNA extracted from P9 and P16 ST2 cells showed that the CpG loci in region A were totally methylated in both P9 and P16 (Fig. 3A); the CpG loci in region B were rarely methylated in P9, whereas CpG, especially cytosines no. 1 (three bases upstream of the TATA-box) and nos. 3–7 (around the transcription start site), nos. 17–21 (around the translation start site) and nos. 28–37 were frequently methylated in P16 (Fig. 3B). Regardless of the total methylation of the CpG loci located in region A, H4 histone acetylation in response to vitamin D3 was observed in P9, suggesting that CpG methylation in region A does not interfere with vitamin D3-induced chromatin relaxing in either –950/–680 (Fig. 3C) or –250/+10 (Fig. 3D). CpG methylation in region B that accumulated in P16 ST2 cells, on the other hand, affected vitamin D3-induced histone acetylation in both –950/–680 and –250/+10.
|
). When transfected into ST2 cells, the promoter activity of the methylated construct decreased to less than 30% of the original construct at the steady state and failed to increase promoter activity in response to vitamin D3 in both P9 and P16 (Fig. 4B), suggesting that methylation of the RANKL gene promoter suppresses its transcriptional activity and responsiveness to vitamin D3.
|
), suggesting that demethylation increases transcription and vitamin D3 responsiveness of the RANKL gene. Western blotting showed that, 5-aza-dC restored RANKL protein expression in P16 ST2 cells in the presence of vitamin D3 (Fig. 5B). As assessed by immunohistochemistry, 5-aza-dC-treated P16 cells showed focal expression of RANKL protein in response to vitamin D3 (Fig. 5C). To examine whether demethylation restored osteoclastogenesis as well as RANKL expression in P16 ST2 cells, 5-aza-dC-treated ST2 cells were cocultured with bone marrow macrophages with or without vitamin D3. P16 ST2 cells treated with 5-aza-dC supported a significant number of osteoclasts in response to vitamin D3 in the coculture, although the increase in the number of osteoclasts was less than that in P9 ST2 cells (Fig. 5D).
|
). Bisulfite mapping of osteoblastic cells, however, revealed that the CpG loci at –100/+350 except CpG no. 9 were almost nonmethylated in both 8-wk- and 12-month-old mice (Fig. 6B). On the other hand, organs such as the heart, lung, liver, small and large intestine, cerebellum and testis had a high rate of methylation at CpG locus no. 1 (Fig. 6, C and D), and CpG loci nos. 10–39 showed less than 10% methylation (data not shown).
|
). A reactive PCR product reflecting endogenous MeCP2 binding within –96/+12 was observed in nuclear precipitate from methylation-prone P16 cells. Conversely, a PCR product reflecting TBP binding was found only in P9 and those reflecting H4 and H3 histone acetylation were detected predominantly in P9 (Fig. 7A). The effect of introducing single-CpG methylation adjacent to the TATA-box on the TATA binding protein and MeCP2 was further analyzed by EMSA. A distinct difference between methylated and unmethylated probes was observed. The unmethylated oligonucleotide containing an inverted TATA-box (–42/–13) showed a clear protein DNA binding complex (arrow in Fig. 7B, lane 2). The specificity of the DNA-TBP binding was verified by the supershift with an anti-TBP antibody (lane 3) and by the competition assay with an excess amount of the unlabeled oligonucleotide containing a consensus TATA-box sequence (lanes 5 and 6). The oligonucleotide with a methyl-cytosine introduced at CpG locus no. 1 showed a strong protein DNA binding complex at higher position (arrowhead in Fig. 7B, lane 8), which was selectively and completely blockshifted by an anti-MeCP2 antibody (lane 10). The methylated probe also showed a protein DNA binding complex at the same position as of the unmethylated probe, which was washed out by the unlabeled consensus TATA oligonucleotide (lanes 11 and 12), and conversely, which increased by the addition of an anti-MeCP2 antibody (lane 10).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). Both VDR expression and its in vitro DNA binding capacity were of almost the same level in P9 and P16 ST2 cells (Fig. 2, A and B). Furthermore, the transient transfection study showed that the external RANKL gene promoter containing VDRE had almost the same level of activity in P9 and P16 ST2 cells. These data suggested that the passage number-dependent lack of responsiveness of the RANKL gene to vitamin D3 was due not to the deterioration of VDR expression and its binding capacity to VDRE, but probably to the epigenetic alteration of the RANKL gene. Consequently, we focused on the CpG methylation status of the intrinsic RANKL gene promoter in P9 and P16 ST2 cells.
By sodium bisulfite mapping (Fig. 3B), P16 exhibited accumulation of methylation at the CpG loci in –10/+350 of the RANKL gene promoter, especially around no. 1 (three bases upstream of TATA-box), nos. 3–7 (downstream of the transcription start site) and nos. 17–21 (around ATG codon). A CpG at no. 1 located three nucleotides upstream of the TATA-box was highly methylated in P16 ST2 cells that did not express RANKL. In light of a recent observation that DNA-enriched fragments containing A/T-rich bases (more than four continuous A/T sequences) adjacent to methyl-CpG are required for MeCP2, a most abundant methyl-binding domain protein, to selectively bind to the target DNA (23), we noticed that methylation at the CpG locus no. 1 located just 3 nucleotides upstream of the inverted TATA-box (CGaggTTTATAA) was a typical MeCP2 target region and could, therefore, be a critical site for silencing the RANKL gene. In agreement with this hypothesis, both ChIP assay and EMSA showed that CpG locus no. 1 was a target of MeCP2 binding and that the recruitment of MeCP2 at the TATA-box interferes with the binding of the TBP to the TATA-box (Fig. 7A). Moreover, the methylated oligonucleotide (–42/–13) showed a TBP binding at the TATA-box sequence, which was increased by blocking the DNA-MeCP2 binding by the addition of an interfering anti-MeCP2 antibody (Fig. 7B, lane 10). From these ChIP and gel shift assay studies, we speculated that the TATA-box of the mouse RANKL gene promoter cannot be co-occupied by TBP and MeCP2 and that the occupancy of the TATA-box by MeCP2 may shut down RANKL gene transcription. This hypothesis would explain the sudden decrease in osteoclastogenesis and RANKL gene expression starting at P12 and P13 ST2 cells (Fig. 1A). In fact, in drug-resistant human breast cancer cells, CpG methylation close to the TATA-box is also associated with the silencing of the WTH3 gene (24). Besides such site-specific methylation critical for TBP binding, generalized accumulation of methylation at a CpG island in the mouse RANKL gene promoter may result in the recruitment of histone deacetylase (25) and chromatin condensation leading to the stable epigenetic silencing of the RANKL gene. Indeed, mirroring the accumulation of methylation in P16, acetylation of histone H4 was not observed in P16 even after vitamin D3 treatment (Fig. 3, C and D). The in vitro observation that SssI-methylated RANKL gene promoter showed reduced transcriptional activity and poor response to vitamin D3 (Fig. 4B) and that DNA methylase inhibitor, 5-aza-dC, treatment partially restored RANKL mRNA (Fig. 5A), protein expression (Fig. 5, B and C) and osteoclastogenesis in methylation-prone P16 ST2 cells (Fig. 5D) also supported our hypothesis. Because the effects of vitamin D3 on gene transcription are mediated through the recruitment of coregulators containing histone acetylase activity, the converse (26), DNA methylation is interpreted by a family of methyl-CpG binding domain proteins that repress transcription through the recruitment of corepressors with histone deacetylase activity (27); thus, the inducible effect of vitamin D3 may partly be counteracted by CpG methylation at the histone modification level.
Next, to address the issue of whether or not methylation of the RANKL gene promoter is related to the tissue-specific expression of the gene, we analyzed the in vivo status of CpG methylation by Southern blotting and bisulfite mapping. In all genomic DNA from mouse tissues examined, methylation-protected bands were observed even in whole bone tissue (Fig. 6A). By sodium bisulfite mapping, stromal/ osteoblastic cells isolated from 8-wk- and 12-month-old mice (both of which support osteoclastogenesis and express RANKL in the presence of vitamin D3; data not shown) showed no methylation at CpG locus no. 1 of the RANKL gene promoter (Fig. 6B), whereas other organs without RANKL expression presented a high rate of methylation at CpG locus no. 1 (Fig. 6, C and D). In view of the fact that RANKL expression is limited among stromal/ osteoblastic cells and a small subset of T-lymphocytes (9, 28), we speculate that methylation of the RANKL gene promoter region is associated with the tissue-specific expression of the gene. Alterations of the methylation status, on the other hand, have been implicated as a consequence of aging. Whereas age-related global hypomethylation probably attributable to decreased methyltransferase activity has been reported in genes lacking a CpG island (29), accumulation or increased methylation in CpG islands of certain genes is observed in an age-dependent manner (the so-called type A methylation) (30). In comparing primary cultured stromal/osteoblastic cells isolated from 8-wk- and 12-month-old mice, no difference of methylation in the RANKL gene promoter was noted (Fig. 6B). We speculate that methylation of RANKL gene promoter does not belong to type A methylation but is related to differentiation and tissue-specific RANKL gene expression.
In conclusion, our data suggest that CpG methylation of the RANKL gene promoter reversibly suppresses RANKL gene activation by vitamin D, and that this epigenetic mechanism is one of the factors defining the heterogeneity and diversity of stromal/ osteoblastic cells in response to bone-resorbing stimuli as well as to the tissue-specific expression of the RANKL gene.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
MEM (Sigma, St. Louis, MO) supplemented with 10% fetal bovine serum (Sigma). ST2 cells were passed every 7 d and, after being treated with 0.25% trypsin/ EDTA (Sigma), were subcultured at a dilution of 1:10 (104/cm2). The cells cultured in phenol red-free MEM supplemented with 2% charcoal-stripped fetal bovine serum were treated with 10 nM of 1,25(OH)2 vitamin D3 (vitamin D3) for 12–24 h and subjected to quantitative RT-PCR and Western blotting. ST2 cells were seeded at a density of 5 x 105 cells/ 60 mm-diameter dish for 24 h then treated with 1 µM of 5-aza-dC (Sigma). After 48 h, the cells were subjected to quantitative RT-PCR and Western blotting. Mouse bone marrow mononuclear cells (2 x 105/cm2), prepared as previously described (5), were cocultured with ST2 cells (2 x 104/cm2) sequentially from passage 8 (P8) or passage 17 (P17) for 7 d in the presence of 10 nM vitamin D3 in multiwell plates (Becton Dickinson, Lincoln Park, NJ); at the end of the culture, the cells were stained for TRACP with a commercial kit (Sigma), and the number of osteoclast-like TRACP-positive multinuclear (>3) cells was counted. The coculture experiment was repeated three times and the representative results are expressed as the means ± SD of the number of osteoclast-like cells obtained from four culture wells.
Quantitative RT-PCR
Total RNA was extracted from ST2 cells treated with or without 10 nM of vitamin D3 and/or 1 µM of 5-aza-dC by standard methods using an RNeasy Protect Mini kit (Qiagen KK, Tokyo, Japan). To assess the amount of mRNA of RANKL, OPG and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 1 µg of total RNA was reverse-transcribed to produce cDNA, which was then amplified and quantified by the ABI PRISM 7300 Real-Time PCR system (Applied Biosystems, Foster City, CA) using sets of primers and probes (assay ID of RANKL: OPG: Mm00441908, Mm00435452 and Taq Man Rodent GAPDH no. 4308313). The amount of RANKL and OPG mRNA was quantified relative to that of GAPDH in each reaction according to the manufacturer’s protocol (Applied Biosystems).
Western Blotting
P9 and P16 ST2 cells with or without 10 nM vitamin D3 and/or 1 nM of 5-aza-dC treatment were lysed with 1.0 ml of ice-cold lysis buffer [50 mM Tris-HCl (pH 8.0), 100 mM NaCl, 0.2% Nonidet P-40, 0.2% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS), and 50 µm/ ml aprotinin]. After sonication for 1 min, the lysates were centrifuged at 15,000 rpm for 20 min. The supernatant, equalized by protein concentrations, was separated by 10% SDS-PAGE, transferred to the nylon membrane, Hybond-P (Amersham Biosciences, Piscataway, NJ) and immunoblotted with polyclonal antibodies against ß-actin (BT-560; Biomedical Technologies, Stoughton, MA), RANKL (sc-7627) and VDR (sc-1008; Santa Cruz Biotechnology, Santa Cruz, CA). Immunocomplexes were visualized by the enhanced chemiluminescence (ECL) Western blotting system (Amersham) according to the manufacturer’s instructions.
Immunohistochemistry
P9 and P16 ST2 cells (105/cm2) were seeded onto Lab-Tek Chamber Slides (Nalge Nunc International, Rochester, NY) before treatment with a combination of 2 nM of 5-aza-dC and/or 10 nM vitamin D3 for 48 h. After fixation with 4% paraformaldehyde and blocking endogenous peroxidase with 3% H2O2, the cells were incubated with an anti-RANKL polyclonal antibody (sc-7627; Santa Cruz) at a final concentration of 2 µg/ml for 1 h at a room temperature. Nonimmunized goat serum was used for preparing negative controls. A rabbit antigoat Ig as the second antibody (Dako, Carpinteria, CA) and a DAKO LSAB kit (Dako) of the avidin-biotin-peroxidase complex method were used. Final development was carried out with 3,3'-diaminobenzidine containing 0.03% H2O2.
In Situ Hybridization
Tibiae and visceral organs such as the heart, lung, liver, small and large intestine, cerebellum, and testis were excised from 8-wk-old mice, fixed with 4% paraformaldehyde in 0.1 M phosphate buffer, decalcified with 20% EDTA in 0.1 M of phosphate buffer, and embedded in paraffin. Serial sections 5 µm thick were used for hematoxylin and eosin staining and in situ hybridization. The cDNA fragments from the coding region of mouse RANKL (RANKL-1: 364–613 and RANKL-2: 1096–1422) were amplified and used as templates for the digoxigenin (DIG)-labeling procedure in the presence of DIG-deoxyuridine triphosphate (DIG DNA labeling mixture; Roche, Basel, Switzerland). DIG-labeled single-stranded antisense DNA probes for mouse RANKL were generated with antisense primer-primed unidirectional PCR, as previously described (12). For negative controls, DIG-labeled sense probes were generated with sense primer-primed unidirectional PCR. The mouse tissue sections were pretreated and incubated with the hybridization medium containing 1 µg/ml of antisense DNA probes (RANKL-1 and -2) at 50 C for 12 h, as previously described (12). After hybridization, the slides were incubated with alkaline phosphatase-conjugated anti-DIG antibody (Roche) and subjected to the colorimetric reaction with a solution of nitroblue tetrazolium salt and bromo-4-chloro-3-indol phosphatase (Roche).
Mouse RANKL Gene Promoter Construct and Transient Transfection Study
The 5'-flanking region of the mouse RANKL gene was cloned, and nested deletion mutants were subcloned into the pGL3-Basic vector, as –1005-luc and –723-luc by PCR as previously described (15, 22); –1005-luc with AGGTCAGCCTGGTTCA (VDRE: –937/–922), and –723-luc lacking VDRE. Each of the plasmid constructs and the pRL-TK vector (Promega, Madison, WI) were cotransfected into the ST2 cells by the liposome-mediated technique, Lipofectamine (Invitrogen, Carlsbad, CA). Transfected ST2 cells cultured in phenol red-free MEM supplemented with 2% charcoal-stripped fetal bovine serum were treated with 10 nM vitamin D3 or the vehicle (0.1% ethanol) 24 h before harvesting, then luciferase activity of firefly and Renilla from cell lysates was determined with a luminometer (ATP-3010; Advantec, Tokyo, Japan). The transfection study was repeated three times; the representative data are expressed as the means ± SD of relative luciferase activity standardized by thymidine kinase (TK) promoter activity obtained from four culture wells.
SssI Methylase Treatment
Promoter construct pGL3-1005 was treated with SssI methylase (New England BioLabs, Beverly, MA) in the presence of S-adenomethionine. Part of the DNA purified by ethanol precipitation was digested with HpaII and electrophoresed to monitor the SssI treatment. Methylated pGL3-1005 was transfected into P9 and P16 cells, treated with vitamin D3 or vehicle for 24 h and then subjected to luciferase assay, as described above.
EMSA
Nuclear protein extracts were prepared from P9 and P16 ST2 cells cultured with or without 10 nM vitamin D3, as previously described (22, 32). For EMSA, the double-stranded oligonucleotide spanning VDRE of the mouse RANKL gene promoter (–942/–917), 5'-GTTTGAGGTCAGCCTGGTTCATATAG-3', was 5'-end labeled with 
32P-ATP (3000 Ci/ mmol) by T4 polynucleotide kinase (Promega). The binding reaction was carried out by preincubating the labeled nucleotide with nuclear extract protein (5 µg) in 20 mM HEPES (pH 7.9), 50 mM of NaCl, 5% glycerol, 0.1 mM of dithiothreitol, and 1 µg poly (deoxyinosine-deoxycytosine) at room temperature for 15 min. Antibodies against VDR, sc-1008, and RXRß, sc-831, (Santa Cruz) were added to the binding reaction for 45 min at 4 C. Samples were electrophoresed on a 5% polyacrylamide gel at room temperature for 3 h at 100 V, and then the gel was dried and analyzed with image analyzer BAS-EWS 4075 (FUJIX, Tokyo, Japan). Nuclear protein extracts from ST2 cells were subjected to the binding assay using the double-stranded oligonucleotide spanning the inverted TATA-box (–42/–13) with one or without methylated-CpG dinucleotide (Operon Biotechnologies, Huntsville, AL), 5'-GCTCTCTCCACGAGGTTTATAAGAGTTAGG-3' (–42/–13), or 5'-GCTCTCTCCA[5Me-dC]GAGGTTTATAAGAGTTAGG-3'. An anti-TBP antibody (ab818; Abcam, Cambridge, MA) and an anti-MeCP2 antibody (ab2828) were used for the binding reaction. Unlabeled oligonucleotide containing the consensus TATA-box, 5'-GGGCACAGCCCAGAGGGTATAAAACAGTGCTGGA-3' (adapted from the human osteocalcin gene promoter) and that containing the non-TATA sequence (shuffled sequence of the consensus TATA-box), 5'-GGGCAGACACGCACAGTGATGAGATAGCACTGGA-3', were used for the competition assay.
ChIP Assay
According to the instructions of the Acetyl-Histone H3/ H4 Immunoprecipitation Assay Kit (Upstate Biotechnology, Lake Placid, NY), soluble chromatin was prepared from 1 x 106 ST2 cells fixed with formaldehyde, precipitated with salmon sperm DNA/ protein A agarose-50% slurry, then 2 ml of the supernatant solution was incubated with an antiacetyl-Histone H3 or H4 antibody overnight at 4 C with rotation. The immunoprecipitate was collected after treatment with salmon sperm DNA/ protein A agarose slurry for 1 h at 4 C with rotation, and then washed. From the eluted immunoprecipitate, DNA was recovered by phenol/ chloroform extraction and ethanol precipitation, and resuspended with 1x Tris-EDTA buffer. PCR with the above DNA solution as a template was conducted with the following sets of primers covering regions 1 (–950/–680) and 2 (–250/+10):
5'-GTTTGAGGTCAGCCTGGTTCATATAG-3':RANKL-C1 (sense)
5'-AGCCTCACTGCTTAAGAAATCCTTATG-3':RANKL-C1 (antisense)
5'-GGACCCAACCCACAGCCTCCA-3':RANKL-C2 (sense)
5'-GAGATGGCAGGGTACCCCAGGCAG-3':RANKL-C2 (antisense)
The soluble chromatin was also incubated with an anti-MeCP2 antibody (ab2828) or an anti-TBP antibody (ab818). DNA was recovered from the eluted immunoprecipitate and subjected to PCR to amplify the 108-bp region (–96/+12) containing the inverted TATA-box with the following primers,
5'-AGATGTGGGAGTGAAAGAGGCAC-3' (sense)
5'-AGATGGCAGGGTACCCCAGGCAG-3' (antisense)
Southern Blotting
Genomic DNA was extracted from the heart, lung, liver, spleen, kidney, small and large intestine, cerebrum, cerebellum, thyroid, pancreas, testis, and femoral bone of 8-wk-old male BALB/c mice. In total, 10 µg of each DNA were digested for 24 h at 37 C with 3 µl of either MspI or HpaII, electrophoresed and transferred onto high bond n + nylon membranes (Amersham). A 580-bp DNA fragment downstream of the transcription start site of the mouse RANKL gene was used as a probe. The membrane was hybridized at 65 C for 12 h, washed twice in 2x SSPE containing 0.1% SDS, twice in 1x SSPE containing 0.1% SDS, at 65 C, and then analyzed with a BAS-2000 image analyzer (FUJIX).
Mapping of Methylation Site by Sodium Bisulfite Modification
Genomic DNA was extracted from P9 and P16 ST2 cells, the primary cultured bone-explanting cells from minced tibiae of either 8-wk- or 12-month-old male BALB/c mice and visceral organs such as the heart, lung, liver, small and large intestine, cerebellum and testis from 8-wk-old mice. The bisulfite reaction was carried out as previously described (33). A quantity of 1 µg of DNA in a volume of 50 µl of Tris-EDTA was denatured by NaOH (final concentration, 0.2 M) for 10 min at 37 C. Freshly prepared 30 µl of 10 mM hydroquinone and 520 µl of 3 M sodium bisulfite at pH 5 were added to the samples. Each sample was incubated under mineral oil at 50 C for 16 h. Modified DNA was purified with Wizard DNA purification resin according to the manufacturer’s recommended protocol (Promega) and eluted into 50 µl of H2O. Modification was completed by NaOH (final concentration, 0.3 M) treatment for 5 min at room temperature, then by ethanol precipitation. DNA was resuspended in 20 µl of H2O and used immediately or stored at –20 C. Bisulfite-modified DNA (100 ng) was amplified with nested PCR using the following sets of converted primers covering the CpG loci (A, –920/–721), and (B, –66/+357),
5'-GAGGTTAGTTTGGTTTATATAGTAAGTTTTAG-3': A-sense
5'-CAAAACTAATATCAAAATTAAACCTCACTACT-3': A-antisense
5'-ATAGTAAGTTTTAGGTTAGTTTAGTTTAT-3': A-nested-sense
5'-CAACTAATAAAAACACTTACAACTTATCTTA-3': A-nested-antisense
5'-AAGGAGGGTAGATGTGGGAGTGAAAGAGGTATT-3': B-sense
5'-AATACAAAAACAAAACAATACAAACCACCT-3': B-antisense
5'-GAGGTTGATTGGTTTTGGAGGTTAGT-3': B-nested-sense
5'-CAATACTACAAACCACCTTTCCCAATC-3': B-nested-antisense
The PCR condition was as follows: 30 cycles of 94 C for 30 sec, 60 C for 30 sec, 72 C for 30 sec, and the final elongation step for 5 min at 72 C. The PCR mixture contained 1x buffer (Takara, Tokyo, Japan) with 1.5 mM of MgCl2, 20 pmol of each primer, 0.2 mM of deoxynucleotide triphosphates, and bisulfite-modified DNA (50 ng) in a final volume of 50 µl. Each PCR product was loaded onto a 3% agarose gel, and the purified DNA from the gel was cloned into the pCR 2.1 plasmid vector (Invitrogen, Groningen, The Netherlands). After transformation and culturing of the competent bacteria (INVF’; Invitrogen) overnight on an LB/agar/ampicillin plate, colonies (at least 12) were randomly selected, and the recombinant plasmid was recovered for DNA sequencing with an M13F or M13R primer. The sequencing reactions for the cloned PCR products were carried out with a DNA sequencing kit (Applied Biosystems) and the reaction products were analyzed on a 310 Genetic Analyzer (Applied Biosystems). In all molecules examined, all cytosine residues not preceding guanine residues were converted by bisulfite treatment. Bisulfite treatment, PCR, cloning and sequencing analysis of DNA were repeated independently.
Statistical Analysis
The significance of the data was analyzed by Student’s t test (two tailed).
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Disclosure Statement: The authors have nothing to disclose.
First Published Online September 28, 2006
Abbreviations: ChIP, Chromatin immunoprecipitation; DIG, digoxigenin; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; OPG, osteoprotegerin; P, passage; RANKL, receptor activator of nuclear factor-B ligand; RXR, retinoid X receptor; SDS, sodium dodecyl sulfate; TBP, TATA-box binding protein; TRACP, tartrate-resistant acid phosphatase; VDR, vitamin D receptor; VDRE, vitamin D-responsive element.
Received for publication May 12, 2006. Accepted for publication September 20, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
B ligand) mRNA and protein in skeletal and extraskeletal tissues. Bone 25:525–534[Medline]B ligand gene in osteoblasts and bone marrow stromal cells. J Cell Biochem 96:716–727[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  J. Gimenez, C. Montgiraud, G. Oriol, J.-P. Pichon, K. Ruel, V. Tsatsaris, P. Gerbaud, J.-L. Frendo, D. Evain-Brion, and F. Mallet Comparative Methylation of ERVWE1/Syncytin-1 and Other Human Endogenous Retrovirus LTRs in Placenta Tissues DNA Res, August 1, 2009; 16(4): 195 - 211. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Miyamoto, T. Tsukiyama, Y. Yang, N. Li, N. Minami, M. Yamada, and H. Imai Cell-Free Extracts from Mammalian Oocytes Partially Induce Nuclear Reprogramming in Somatic Cells Biol Reprod, May 1, 2009; 80(5): 935 - 943. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. L. Foley, J. M. Craig, R. Morley, C. J. Olsson, T. Dwyer, K. Smith, and R. Saffery Prospects for Epigenetic Epidemiology Am. J. Epidemiol., February 15, 2009; 169(4): 389 - 400. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
