Diane H. Cho ^{1, 2}, Cortlandt P. Thienes ¹, Sarah E. Mahoney ¹, Erwin Analau ¹, Galina N. Filippova ¹, and Stephen J. Tapscott ^{1, @},

¹ Division of Human Biology, Fred Hutchinson Cancer Research Center, Seattle, Washington 98109
² Molecular and Cellular Biology Program, University of Washington, Seattle, Washington 98105

^@ Corresponding Author Contact Information:    Phone: 206-667-4499; Fax: 206-667-6524
e-mail:   stapscot@fhcrc.org

Prior studies of the DM1 locus have shown that the CTG repeats are a component of a CTCF-dependent insulator element and that repeat expansion results in conversion of the region to heterochromatin. We now show that the DM1 insulator is maintained in a local heterochromatin context: an antisense transcript emanating from the adjacent SIX5 regulatory region extends into the insulator element and is converted into 21 nucleotide (nt) fragments with associated regional histone H3 lysine 9 (H3-K9) methylation and HP1g recruitment that is embedded within a region of euchromatin-associated H3 lysine 4 (H3-K4) methylation. CTCF restricts the extent of the antisense RNA at the wild-type (wt) DM1 locus and constrains the H3-K9 methylation to the nucleosome associated with the CTG repeat, whereas the expanded allele in congenital DM1 is associated with loss of CTCF binding, spread of heterochromatin, and regional CpG methylation.

Myotonic dystrophy (DM1) is an autosomal-dominant, multisystem disorder caused by an expanded CTG repeat in the 3' UTR of a protein kinase gene, DMPK. The general population has 5–38 CTG repeats, whereas individuals with DM1 have greater than 100 repeats, and frequently thousands of repeats, with disease severity correlated with the number of repeats. Research efforts thus far strongly support the hypothesis that the myotonia associated with DM1 results from the interaction between the CUG expansion in the DMPK RNA and the RNA binding proteins of the muscleblind family (Kanadia et al., 2003). The CTG expansion also alters the regional chromatin structure and gene expression at the DM1 locus. At the wt locus, the region is in a phased nucleosome array with a nucleosome positioned over the repeats (Filippova et al., 2001), whereas on the expanded allele, the region surrounding the CTG repeats becomes nuclease resistant and the expression of the adjacent SIX5 gene is suppressed (Klesert et al., 1997, Thornton et al., 1997 and Otten and Tapscott, 1995). The induction of heterochromatin and gene silencing by expanded CTG repeats was also demonstrated in transgenic mice and shown to be sensitive to the dose of heterochromatin protein 1 (HP1) (Saveliev et al., 2003). The clear distinctions in chromatin structure between the normal and expanded CTG repeats at the DM1 locus provide an opportunity to study the mechanisms of chromatin modifications associated with unstable repeats in the human genome.

Results:

A model for heterochromatin formation at repetitive elements is emerging from studies in lower eukaryotes (reviewed in Bernstein and Allis [2005]). In this model (Figure 1A), bidirectional RNA transcripts are converted to siRNA, which recruits histone methyltransferases, HP1, and DNA methyltransferases with associated conversion of the region to heterochromatin. Antisense RNA transcripts and siRNAs induce similar local chromatin modifications and gene silencing in mouse and human cells (Kawasaki and Taira, 2004, Morris et al., 2004 and Tufarelli et al., 2003), indicating that the mechanism of siRNA-directed chromatin modifications is conserved in mammalian cells. We hypothesized that the human DM1 locus containing unstable CTG repeats might activate this conserved RNAi pathway to alter chromatin structure. To identifiy potential siRNAs, we performed RNase protection assays with probes to the region in both wt- and DM1-cultured human fibroblasts. Probe B, an RNA probe that would protect an antisense transcript across the CTG repeats and CTCF site I (Figure 1B), protected 21 nt fragments in fibroblasts from adult onset DM (DM1) and congenital DM (CDM1) cells, but surprisingly, roughly equal abundance fragments were also detected in the wt cells (Figure 1C). The RNA fragments were not restricted to the repeats because a ³²P-ATP-labeled probe, in which the CUG repeat would remain unlabeled, also protected 21 nt RNA fragments. The small RNAs did appear to be localized to the CTCF site 1 and CTG repeat region, however, because Probes A and C covering regions adjacent to Probe B (Figure 1B) failed to detect any small RNA fragments (Figure 1D).

Figure 1. Twenty-One Nucleotide-Protected RNA Fragments Remain after RNase Digestion of a Probe Spanning the CTG Repeat Domain.

Figure 1. Twenty-One Nucleotide-Protected RNA Fragments Remain after RNase Digestion of a Probe Spanning the CTG Repeat Domain.

(A) Model for siRNA heterochromatin formation. dsRNA arising from repetitive DNA sequences is diced to siRNA fragments. Histone methyltransferases (HMTase) are then recruited to methylate lysine 9 residues of histone H3, providing a binding site for the heterochromatin protein HP1. HP1 binding further propagates heterochromatin formation, which can be additionally modified by DNA methyltransferases (DNMT) to methylate CpG residues as indicated by a star (*).

(B) RNase protection probes were generated to target either the sense or antisense strand for the regions shown. Probes A, B, and C protect antisense transcripts, and Probe B' protects a sense strand to the indicated regions.

(C) Probe B or B' has either a radiolabeled CTP (*CUG/GA*C) or ATP (CUG/G*AC) in order to determine if the 21 nt RNA fragments are limited to the CTG repeat region. Note that the hybridization conditions used were optimized for small RNA and, therefore, did not hybridize to the 3 kb sense transcript.

(D) Regions flanking Probe B do not exhibit 21 nt RNA fragments from the antisense transcript. RNA is from normal human fibroblast cell line (wt), congenital DM1 cell line (CDM1), and adult-onset DM1 cell line (DM1).

The presence of a DMPK “sense” transcript through this region has been well documented (Krahe et al., 1995), whereas antisense transcripts have not been characterized. Strand-specific RT-PCR mapped the origin of an antisense transcript in wt cells to a 124 bp region previously characterized as a positive regulator of the sense transcript for the SIX5 gene (Klesert et al., 1997), termed the hypersensitive-site enhancer (HSE) (Figures 2A and 2B). This region was cloned in either orientation upstream of the luciferase gene in the pGL3 vector, and higher luciferase activity was expressed when the regulatory region was in the antisense orientation (Figure 2C). A deletion series mapped the antisense promoter to a 229 bp region of the HSE (Figure 2D). This region contains conserved E boxes, and conversion of the fibroblasts to skeletal muscle by cotransfection with a MyoD expression vector resulted in a substantial increase in the antisense transcripts from this regulatory region (Figure 2D), suggesting that the level of antisense transcripts might be regulated during cell differentiation. The antisense transcript contains several short open reading frames but none encoding a polyglutamine repeat (data not shown).

Figure 2. Start of Antisense Transcription Maps to the HSE Region.

Figure 2. Start of Antisense Transcription Maps to the HSE Region.

(A) To distinguish between self-priming and strand-specific priming, a linker sequence (LK) is attached to the DM1-specific primer for cDNA synthesis and is followed by nested PCR.

(B) Nested PCR product results from template derived from primer A, N(A), and not primer B, N(B), indicating that the antisense transcription begins in the 124 base interval between location of primer A and B. Primers A and B both amplified genomic DNA when paired with a DM1-specific primer (A' and B').

(C) Map of the region cloned into pGL3 basic to test for promoter activity in “sense” (1659–2518 For) and “antisense” (1659–2518 Rev) orientations is shown. pGL3 with indicated insert was transfected into 3T3 cells to determine promoter activity and normalized to cotransfected b-gal expression vector. DM regions in constructs are numbered relative to position 37547 from AC011530 GenBank sequence. Both sense and antisense luciferase reporter activity is present; however, the region has higher antisense transcription than sense transcription.

(D) Deletion constructs map the predominant antisense promoter activity to 1871–2110. Cotransfection with MyoD expression vector (CS2-MyoD) or control (CS2) shows that both sense and antisense reporter activity is enhanced by MyoD activation in standard low-serum differentiation medium (note that the two panels have different scales to accommodate the higher induction of 1871–2110 by MyoD).

Error bars represent the SD of the mean.

We next used chromatin immunoprecipitation (ChIP) to determine whether the 21 nt RNA fragments at the DM1 locus were associated with histone H3-K9 methylation and recruitment of HP1. In order to distinguish between the wt and expanded alleles, we took advantage of the high degree of DNA methylation on the expanded allele, but not the wt allele, in congenital DM1 cells (Steinbach et al., 1998), which was confirmed by bisulfite sequencing (data not shown). Digestion of the precipitated DNA with methylation-sensitive endonucleases allowed for the direct assessment of histone modifications and HP1 recruitment on the expanded allele in the CDM1 cells and inferential determination of binding to the wt allele, whereas PCR amplification across the CTG repeats selectively assessed the wt allele because the CTG expansion is ~5.5 kb. Figure 3A shows enrichment for H3-K9 trimethylation on the expanded allele and opposing H3-K4 trimethylation on the wt allele, consistent with our prior studies showing that the expansion converts the surrounding region to heterochromatin (Otten and Tapscott, 1995). Mammals have three HP1 isoforms, a, b, and g, and all three isoforms are associated with heterochromatin (Maison and Almouzni, 2004). HP1g is also found in some euchromatic regions; however, expression of either mammalian HP1a or HP1g in Drosophila enhances heterochromatin-mediated gene silencing (Ma et al., 2001), indicating similar roles in mediating heterochromatin formation. ChIP for HP1a and HP1b did not show enrichment at the DM1 region (data not shown); however, interestingly, HP1g was associated with both the wt and expanded alleles in CDM1 cells (Figure 3B). Further investigation into HP1g binding in wt cells revealed an enrichment for HP1g in the region of the wt CTG repeats, but not in the regions 1.4 kb upstream in intron 11 (Figure 3C) or 13 kb upstream in exon 1 of DMPK (data not shown).

Figure 3. H3 Modifications, HP1g and CTCF Enrichment on Wt, and Expanded Alleles at the DM1 Locus.

Figure 3. H3 Modifications, HP1g and CTCF Enrichment on Wt, and Expanded Alleles at the DM1 Locus.

(A) Representative ChIP for H3-K4 trimethylation (K4 tri) and H3-K9 trimethylation (K9 tri) in congenital DM1 (CDM1). Wt row represents PCR primers spanning wt CTG repeats (amplicon B in diagram). WT/EXP row (amplicon A in diagram) amplifies both wt and expanded (EXP) alleles upstream of the CTG repeats without methylation-sensitive HpaII digestion (HpaII-) and expanded allele only with HpaII digestion (HpaII+). There is enrichment for K4 tri on the wt allele in HpaII- lane in the wt row, but not on the expanded allele in HpaII+ lane in the WT/EXP row. K9 trienrichment is present on the expanded allele in the WT/EXP row, HpaII+ lane, and not on the wt allele in the wt row, HpaII- lane. Panel below shows relative K4 tri- and K9 trienrichment values normalized to control myogenin (MGN).

(B) Summary of three ChIP experiments for H3 modifications and HP1g recruitment in the CDM1 cell line using methylation-sensitive HpaII or HhaI digestion and indicated PCR amplicons A or B. All relative ChIP enrichments with a p value < 0.05 based on a t test comparing the input amplicon:control ratios to the IP amplicon:control ratios (see Experimental Procedures) are marked by an asterisk (*).

(C) Wt cells show localized enrichment for H3-K4 trimethylation and HP1g recruitment near the CTG repeat domain (amplicons B, C, and D), but not 1.4 kb upstream of CTCF site 1 with PCR amplicon A.

(D) The first panel shows DNA from wt and CDM1 cell lines digested with methylation-sensitive HpaII (H) or its methylation-insensitive isoschizomer, MspI (M), before (INPUT) and after a-CTCF IP from fixed chromatin and amplified with amplicon A primers. HpaII digestion eliminates the enriched fragment after CTCF ChIP, indicating that CTCF binds the unmethylated wt allele, but not the methylated expanded allele. Similarly, the second panel shows CDM1 DNA digested with methylation-sensitive HhaI and assayed for CTCF enrichment relative to internal control MGN. Amplicon B, WT/EXP primers amplify a region upstream of the repeats and show enrichment for CTCF in the HhaI- lane, but not in the HpaI+ lane. Fold enrichments for CTCF are indicated below each IP lane relative to actin or myogenin (MGN) internal controls. Fold enrichments for CTCF in the amplicon C:WT IP lanes (HhaI-/+) are a control for HhaI digestion.

Error bars represent the SD of the mean.

Previously, we demonstrated that the CTCF insulator factor binds to sites flanking the CTG repeats at the wt DM1 locus and forms an insulator element at this site (Filippova et al., 2001). To determine whether CTCF protein is binding at the expanded and methylated allele in the CDM1 cells, we performed allele-specific analysis of CTCF binding by ChIP. Figure 3D shows that there is abundant enrichment for CTCF at the wt allele in both wt and CDM1 cells but substantially reduced or absent CTCF binding at the expanded and methylated allele in the CDM1 cells. The DNA sequence of the CTCF binding region was identical in the wt and expanded allele (data not shown).

CTCF is associated with chromatin domain boundaries (Filippova et al., 2005). To determine whether CTCF sites flanking the DM1 repeats can influence regional transcription, we used recombination-mediated cassette exchange (RMCE) (Feng et al., 1999) to stably integrate a portion of the human DM1 locus, with wt or mutant CTCF sites, in the same orientation at a random genomic site in the mouse 3T3 fibroblast cell line. Strand-specific real-time RT-PCR showed that wt CTCF sites significantly decreased the extension of antisense transcription beyond the CTG repeats and CTCF site 1 relative to mutated CTCF sites (Figure 4A). Sense transcripts did not demonstrate an analogous change across the CTCF sites (data not shown). ChIP experiments confirmed that CTCF was bound to the wt locus and not to the locus with the mutant CTCF binding sites (Figure 4B), and ChIP for RNA polymerase II (pol II) binding showed an increased amount of pol II and phosphorylated pol II at the locus with mutated CTCF sites compared to the wt locus (Figure 4B), consistent with the RT-PCR studies that showed increased transcription of this region on the locus with the mutated CTCF sites (see Figure 4A).

Figure 4. CTCF Impedes Antisense Transcription at the Integrated Human DM1 Sequence in 3T3 Recombinant Lines, and the Repeat Region Is Associated with RNA Polymerase II, H3-K9 Dimethylation, and HP1g Enrichment.

Figure 4. CTCF Impedes Antisense Transcription at the Integrated Human DM1 Sequence in 3T3 Recombinant Lines, and the Repeat Region Is Associated with RNA Polymerase II, H3-K9 Dimethylation, and HP1g Enrichment.

(A) Shown is the 1.8 kb human DM1 sequence stably integrated into 3T3 cell lines by RMCE containing CTG₁₉ repeats with wt CTCF sites (wt) or mutated CTCF sites (MUT). Quantitative real-time PCR analysis on first round PCR products for amplicons A, B, and C shows that CTCF sites block the progression of antisense transcription compared to mutated CTCF sites. Independent wt and mutant clones showed similar results. Distances between the 5' ends of taqman probes and the HSE are shown.

(B) Representative CTCF, pol II, or phosphorylated pol II ChIPs are shown for wt and MUT CTCF cell lines with amplicon A primers. Summary of wt pol II and phospho pol II enrichments relative to MUT from two (pol II) or three (phospho pol II) experiments is shown for amplicon B primers. ChIP enrichments with a p value < 0.05 are marked by an asterisk, indicating a significant enrichment of the IP amplicon:control ratio compared to the input/control ratio.

(C) Representative ChIP for histone H3 methylation and HP1g recruitment in the 3T3 recombinant cell lines was assessed by real-time PCR and normalized to mouse GAPDH.

(D) ChIP from fixed chromatin digested by micrococcal nuclease to mononucleosomes shows enrichment for H3-K9 dimethylation localized mainly to amplicon C repeat domain as determined by real-time PCR and normalized to mouse ACTA1. There is H3-K4 trimethyl enrichment across amplicons A, B, C, and D.

Error bars represent the SD of the mean.

Similar to the endogenous DM1 alleles in human cells, these recombinant 3T3 cell lines also demonstrated enrichment for HP1g at the integrated human DNA sequences; however, neither the wt nor mutant locus was associated with H3-K9 trimethylation (Figure 4C). The recruitment of HP1g to this locus in the absence of H3-K9 trimethylation, both at the RMCE human loci integrated in the 3T3 cells and at the endogenous human wt locus (see Figures 3B and 3C), led us to assess H3-K9 dimethylation, which is also associated with HP1 recruitment and heterochromatin. Indeed, the region of the CTG repeats, both with the wt or the mutant CTCF sites, was associated with H3-K9 dimethylation (Figure 4C). In addition, both loci were enriched for the H3-K4 trimethylation characteristic of active loci, indicating that the area of the wt CTG repeats has features of heterochromatin embedded within a region of euchromatin.

A recent publication shows that H3-K9 methylation and HP1g are associated with transcription elongation (Vakoc et al., 2005) and suggests that these features will be present at any transcribed region. To determine whether the H3-K9 methylation was restricted to the region of the CTG repeat and dsRNA, we prepared mononucleosomes for ChIP by using micrococcal nuclease. At the human minilocus in 3T3 cells, H3-K9 dimethylation was restricted to the single nucleosome covering the CTG repeats, whereas H3-K4 trimethylation was spread throughout the region (Figure 4D). It is interesting to note that the allele with mutations in the CTCF binding sites appears to have a less restricted distribution of H3-K9 dimethylation, suggesting that CTCF might limit the spread of H3-K9 methylation to adjacent nucleosomes. In human fibroblasts, a similar mononucleosome preparation also showed H3-K9 dimethylation restricted to the nucleosome associated with the CTG repeat (data not shown). Therefore, H3-K9 methylation appears to be highly restricted to the region of the CTG repeats and siRNA-sized fragments.

Discussion:

Taken together, these findings suggest an interesting model of chromatin dynamics at the DM1 locus. The wt allele is associated with bidirectional transcription, 21 nt RNA fragments, H3-K9 dimethylation, and HP1g recruitment in the region of the CTG repeats. The presence of heterochromatin-associated H3-K9 dimethylation and HP1g on the wt allele surrounded by euchromatin-associated H3-K4 trimethylation represents a very local chromatin modification that might be initiated by the presence of bidirectional transcripts through an siRNA mechanism; however, we have not directly shown that the 21 nt RNA fragments from this region are generated by Dicer, and it is possible that the CTG repeat contributes to establishing the local chromatin modifications independent of the siRNA pathway. The restriction of the bidirectional transcripts to a small region appears to be accomplished, at least in part, by the flanking CTCF sites, and the spread of H3-K9 dimethylation to a broader region when the CTCF sites are mutated suggests CTCF might prevent the propagation of heterochromatin to the surrounding regions, which is consistent with its location at chromatin boundaries on the X chromosome (Filippova et al., 2005). When the CTG repeat is expanded in DM1 cells, the region of heterochromatin spreads, as previously demonstrated by our nuclease access studies (Otten and Tapscott, 1995) and shown in this study by the replacement of H3-K4 trimethylation with H3-K9 trimethylation on the expanded allele. In this regard, our study demonstrates that the wt allele contains a constrained island of chromatin modifications and that CTG expansion is sufficient to nucleate the spreading of heterochromatin to surrounding regions.

CTCF might also have a role in preventing DNA methylation. It is interesting that the recruitment of HP1g does not result in DNA methylation either at the wt locus or at the expanded locus associated with DM1. Binding of CTCF has been shown to prevent DNA CpG methylation at the H19 locus (Pant et al., 2004 and Schoenherr et al., 2003), and it is possible that CTCF binding protects the DM1 CTG region from methylation. Consistent with this, the expanded allele in congenital DM1 is associated with loss of CTCF binding, spread of heterochromatin, and regional CpG methylation.

How these changes might contribute to the phenotype of congenital myotonic dystrophy remains unknown, but it is reasonable to speculate that loss of insulator function due to impaired CTCF binding between DMPK and SIX5 might result in higher DMPK expression late in embryogenesis when the SIX5 enhancer is most active (Klesert et al., 2000). This might contribute to the earlier disease phenotype in congenital DM1, because the major clinical features of DM1 are caused by expression of a CUG repeat containing RNA (Mankodi et al., 2000) through its interaction with RNA binding proteins of the muscleblind-like (MBNL) family (Jiang et al., 2004 and Kanadia et al., 2003). Finally, our prior demonstration that CTCF sites flank CAG/CTG repeats at other disease causing loci (Filippova et al., 2001) suggests that the features described here at the DM1 locus might be present at other repeat associated loci in the genome.

Experimental Procedures:

Cell Cultures

Primary fibroblast cell cultures from an individual with wt CTG repeats (wt: CTG₅/CTG₁₁) and DM1 were established from skin biopsies and previously characterized (Otten and Tapscott, 1995). Cell cultures from a CDM1 fetus with a 5.5 kb expansion as well as a DM1 fetus with a 0.6 kb expansion were previously described (Steinbach et al., 1998).

RNase Protection Assays

We hybridized 5–10 g of total RNA to [³²P]CTP or [³²P]ATP-labeled RNA probes at 50°C by using reagents from Ambion's mirVana kit (#1552). We generated RNA probes from PCR products with the following T7-containing primers: DM955, 5'-CCTGCCAGTTCACAACCGCTCCGAG-3'; DM1180, 5'-AAGAAATGGTCTGTGATCCCCCCAG-3'; DM1191, 5'-AGGCTGAGGCCCTGACGTGGATGG-3'; and DM1384, 5'-GCAAAAGCAAATTTCCCGAGTAAGCAG-3'. We used a 15% polyacrylamide/7% UREA gel to visualize the RNase-protected fragments.

Strand-Specific RT-PCR

RNA was harvested with TRIZOL and extracted twice with acidic phenol/chloroform to remove contaminating DNA. For strand-specific RT-PCR, we attached a linker sequence, LK 5'-CGACTGGAGCACGAGGACACTGA-3', to the 5' end of a primer specific for the antisense strand of DMPK: DM765, 5'-TTCGCCGTTGTTCTGTCTCGTGC-3'; DM1078, 5'-GGGTCCTTGTAGCCGGGAATGCTG-3'; or DM1192, 5'-GGCTGAGGCCCTGACGTGGATGGGCAAAC-3' and used 2–5 mg of RNA and Superscript III (Invitrogen) at 50°C. The first round of PCR contained the LK sequence alone as a primer and a DMPK-specific primer (20 cycles of 94°C 30 s, 62°C 30 s, and 72°C 1 min). The second round of PCR used the LK primer and nested DMPK-specific primers for 30 additional cycles. For real-time PCR analysis, we used 2 ml of 1:100 diluted sample of first round PCR products with primers specific for the LK sequence paired with primers for the DMPK sequence (available upon request). Mouse GAPDH was used to normalize the real-time PCR reactions as it self-primed efficiently in the cDNA reactions without the addition of random or specific primers. We calculated values from duplicate reactions for each sample from standards constructed from the first round PCR products specific to each LK/DMPK primer pair.

ChIP

ChIPs were performed as previously described (Filippova et al., 2001) except as follows. We incubated 750 mg of protein overnight with 5–10 ml of antibodies to H3-K4 trimethyl (Abcam 8580), H3-K9 trimethyl (Abcam 8898), H3-K9 dimethyl (Upstate 07-521), HP1g (Upstate 07-332), HP1a (Upstate 05-689), HP1b (Upstate 07-333), pol II (Santa Cruz 899), or phospho pol II (Santa Cruz 13583). For methylation-sensitive ChIP, we digested 20–50 ng of recovered DNA before (INPUT) or after IP with HpaII or HhaI in 10 ml volume and used 1–2 ml of the digested DNA in PCR reactions. We carried out duplex PCR reactions with internal controls (human myogenin or b-actin; mouse amylase or GAPDH) and determined enrichment values by calculating IP DM-amplicon/control ratios over INPUT DM-amplicon:control ratios. We used real-time PCR analysis for HP1g and H3 methyl ChIPs in 3T3 recombinant lines and calculated relative IP/INPUT values normalized to GAPDH, unless stated otherwise, from standards using INPUT DNA with either DMPK-specific primers or mouse GAPDH primers. For micrococcal nuclease ChIPs, we digested fixed chromatin with 30 U of micrococcal nuclease for 10 min at 37°C. We followed the same protocol as above for the IP and DNA recovery. P values were calculated by comparing the IP DM-amplicon:control ratios to the INPUT DM-amplicon:control ratios by using a one-tailed, two-sample equal variance t test.

Reporter Assays

We used PCR-amplified products with attached BglII or KpnI sites for forward or reverse orientation cloning into pGL3 basic (Promega). DMPK-specific primer sequences were as follows: DM1659, 5'-ACAGTGTCCGCGGTTTGCGTTGT-3'; DM1871, 5'-CCCAGCTGTGCCACCGAGCGTCGAG-3'; DM2100, 5'-TGTGGGGGCCTGCAGAGGAC-3'; DM2110, 5'-AGGCCCCCACATTCCCCATCT-3'; and DM2518, 5'-GTTTTGCAACTTTGGGAAGTTCCTCCC-3'. 3T3 cells were transiently transfected with 2 mg of DMPK constructs and 0.5 mg of CMV-bgal as an internal control. Luciferase assays and b-galactosidase assays were performed as described elsewhere (Bergstrom and Tapscott, 2001).

RMCE

DM1 minilocus constructs were inserted into the same chromosomal position in the fibroblast cell line 3T3 with RMCE essentially by the method of Feng et al. (1999). The b-galactosidase-coding region of the targeting vector L1-PGK-HyTk-1L was first deleted by removing a 267 bp NdeI-XhoI fragment. Stable clones of the targeting construct L1-PGK-HyTk-1L were generated in 3T3 cells with Fugene (Promega) and analyzed for single copy integration by Southern blot. A clone with a single copy integration at a random genomic locus was chosen to exchange the targeting cassette with the human DM1 minilocus (positions 735–2518) with either wt or mutant CTCF sites. This 3T3 targeting line was transfected in 6 cm tissue culture dishes by Fugene with 3 mg of the minilocus constructs and 2 mg of CMV-Cre and selected after 4 days with 2 mg/ml gancyclovir and 560 mg/ml G418. RMCE clones were screened by PCR and Southern blot to identify clones with single copy integrants in the same orientation.

DM1 Sequence

All DM1 primers and positions are numbered relative to nucleotide 37547 from AC011530 GenBank sequence.

Acknowledgments:

This work was supported by National Institutes of Health (NIH) R01 AR4203 (S.J.T.) and NIH U54 AR050741 to the Paul D. Wellstone Muscular Dystrophy Cooperative Research Center. C.P.T. was supported by the Chromosome Metabolism and Cancer Training Grant T32 CA09657 and S.E.M. by a Poncin Foundation Fellowship.

References

Bergstrom and Tapscott, 2001
D.A. Bergstrom and S.J. Tapscott, Molecular distinction between specification and differentiation in the myogenic basic helix-loop-helix transcription factor family, Mol. Cell. Biol. 21 (2001), pp. 2404–2412.

Bernstein and Allis, 2005
E. Bernstein and C.D. Allis, RNA meets chromatin, Genes Dev. 19 (2005), pp. 1635–1655.

Feng et al., 1999
Y.Q. Feng, J. Seibler, R. Alami, A. Eisen, K.A. Westerman, P. Leboulch, S. Fiering and E.E. Bouhassira, Site-specific chromosomal integration in mammalian cells: highly efficient CRE recombinase-mediated cassette exchange, J. Mol. Biol. 292 (1999), pp. 779–785.

Filippova et al., 2001
G.N. Filippova, C.P. Thienes, B.H. Penn, D.H. Cho, Y.J. Hu, J.M. Moore, T.R. Klesert, V.V. Lobanenkov and S.J. Tapscott, CTCF-binding sites flank CTG/CAG repeats and form a methylation-sensitive insulator at the DM1 locus, Nat. Genet. 28 (2001), pp. 335–343.

Filippova et al., 2005
G.N. Filippova, M.K. Cheng, J.M. Moore, J.P. Truong, Y.J. Hu, D.K. Nguyen, K.D. Tsuchiya and C.M. Disteche, Boundaries between chromosomal domains of X inactivation and escape bind CTCF and lack CpG methylation during early development, Dev. Cell 8 (2005), pp. 31–42.

Jiang et al., 2004
H. Jiang, A. Mankodi, M.S. Swanson, R.T. Moxley and C.A. Thornton, Myotonic dystrophy type 1 is associated with nuclear foci of mutant RNA, sequestration of muscleblind proteins and deregulated alternative splicing in neurons, Hum. Mol. Genet. 13 (2004), pp. 3079–3088.

Kanadia et al., 2003
R.N. Kanadia, K.A. Johnstone, A. Mankodi, C. Lungu, C.A. Thornton, D. Esson, A.M. Timmers, W.W. Hauswirth and M.S. Swanson, A muscleblind knockout model for myotonic dystrophy, Science 302 (2003), pp. 1978–1980.

Kawasaki and Taira, 2004
H. Kawasaki and K. Taira, Induction of DNA methylation and gene silencing by short interfering RNAs in human cells, Nature 431 (2004), pp. 211–217.

Klesert et al., 1997
T.R. Klesert, A.D. Otten, T.D. Bird and S.J. Tapscott, Trinucleotide repeat expansion at the myotonic dystrophy locus reduces expression of DMAHP, Nat. Genet. 16 (1997), pp. 402–406.

Klesert et al., 2000
T.R. Klesert, D.H. Cho, J.I. Clark, J. Maylie, J. Adelman, L. Snider, E.C. Yuen, P. Soriano and S.J. Tapscott, Mice deficient in Six5 develop cataracts: implications for myotonic dystrophy, Nat. Genet. 25 (2000), pp. 105–109.

Krahe et al., 1995
R. Krahe, T. Ashizawa, C. Abbruzzese, E. Roeder, P. Carango, M. Giacanelli, V.L. Funanage and M.J. Siciliano, Effect of myotonic dystrophy trinucleotide repeat expansion on DMPK transcription and processing, Genomics 28 (1995), pp. 1–14.

Ma et al., 2001
J. Ma, K.K. Hwang, H.J. Worman, J.C. Courvalin and J.C. Eissenberg, Expression and functional analysis of three isoforms of human heterochromatin-associated protein HP1 in Drosophila, Chromosoma 109 (2001), pp. 536–544.

Maison and Almouzni, 2004
C. Maison and G. Almouzni, HP1 and the dynamics of heterochromatin maintenance, Nat. Rev. Mol. Cell Biol. 5 (2004), pp. 296–304.

Mankodi et al., 2000
A. Mankodi, E. Logigian, L. Callahan, C. McClain, R. White, D. Henderson, M. Krym and C.A. Thornton, Myotonic dystrophy in transgenic mice expressing an expanded CUG repeat, Science 289 (2000), pp. 1769–1773.

Morris et al., 2004
K.V. Morris, S.W. Chan, S.E. Jacobsen and D.J. Looney, Small interfering RNA-induced transcriptional gene silencing in human cells, Science 305 (2004), pp. 1289–1292.

Otten and Tapscott, 1995
A.D. Otten and S.J. Tapscott, Triplet repeat expansion in myotonic dystrophy alters the adjacent chromatin structure, Proc. Natl. Acad. Sci. USA 92 (1995), pp. 5465–5469.

Pant et al., 2004
V. Pant, S. Kurukuti, E. Pugacheva, S. Shamsuddin, P. Mariano, R. Renkawitz, E. Klenova, V. Lobanenkov and R. Ohlsson, Mutation of a single CTCF target site within the H19 imprinting control region leads to loss of Igf2 imprinting and complex patterns of de novo methylation upon maternal inheritance, Mol. Cell. Biol. 24 (2004), pp. 3497–3504.

Saveliev et al., 2003
A. Saveliev, C. Everett, T. Sharpe, Z. Webster and R. Festenstein, DNA triplet repeats mediate heterochromatin-protein-1-sensitive variegated gene silencing, Nature 422 (2003), pp. 909–913.

Schoenherr et al., 2003
C.J. Schoenherr, J.M. Levorse and S.M. Tilghman, CTCF maintains differential methylation at the Igf2/H19 locus, Nat. Genet. 33 (2003), pp. 66–69.

Steinbach et al., 1998
P. Steinbach, D. Glaser, W. Vogel, M. Wolf and S. Schwemmle, The DMPK gene of severely affected myotonic dystrophy patients is hypermethylated proximal to the largely expanded CTG repeat, Am. J. Hum. Genet. 62 (1998), pp. 278–285.

Thornton et al., 1997
C.A. Thornton, J.P. Wymer, Z. Simmons, C. McClain and R.T. Moxley 3rd, Expansion of the myotonic dystrophy CTG repeat reduces expression of the flanking DMAHP gene, Nat. Genet. 16 (1997), pp. 407–409.

Tufarelli et al., 2003
C. Tufarelli, J.A. Stanley, D. Garrick, J.A. Sharpe, H. Ayyub, W.G. Wood and D.R. Higgs, Transcription of antisense RNA leading to gene silencing and methylation as a novel cause of human genetic disease, Nat. Genet. 34 (2003), pp. 157–165.

Vakoc et al., 2005
C.R. Vakoc, S.A. Mandat, B.A. Olenchock and G.A. Blobel, Histone H3 lysine 9 methylation and HP1g are associated with transcription elongation through mammalian chromatin, Mol. Cell 19 (2005), pp. 381–391.

1. Song X, Sun Y, and Garen A, "Roles of PSF protein and VL30 RNA in reversible gene regulation".

2. Hovsepian JA, and Frenster JH, "Sense and Antisense during RNA Initiation of the DNA Transcription Bubble".

3. Frenster JH, and Hovsepian JA, "Ultrastructure  of Closed Loops within Euchromatin of Isolated Lymphocyte Nuclei".

4. Ling J, Ainol L, Zhang L, Yu X, Pi W, and Tuan D, "HS2 Enhancer Function Is Blocked by a Transcriptional Terminator Inserted between the Enhancer and the Promoter".

5. Jiang M, Ma N, Vassylyev DG, and McAllister WT, "RNA Displacement and Resolution of the Transcription Bubble during Transcription by T7 RNA Polymerase".

6. Buskirk AR, Landrigan A, and Liu DR, "Engineering a Ligand-Dependent RNA Transcriptional Activator".

7. Chambeyron S,  and Bickmore WA, "Chromatin decondensation and nuclear reorganization of the HoxB locus upon induction of transcription".

8. Frenster JH, and Hovsepian JA, "Activator RNA Exchange during Interphase Chromatin Reprogramming".

Links to RNA and Biological Causality:

Links to Euchromatin Activator RNA Reviews:
Links to Euchromatin Activator RNA Research:
Links to Ultrastructural Probes of DNase I-Sensitive Sites:
Links to RNA as a Therapeutic Agent:
Links to Hodgkin Lymphoma Immuno-Pathology:
Links to Activated T-Lymphocyte Immunotherapy:
Links to Medical Systems Biology:
Links to Selective Gene Transcription:
Links to RNA-Induced Epigenetics:
Links to RNA-Induced Embryogenesis:
Links to RNA and Biological Causality:
Links to Reprogramming and Neoplasia:

A Brief History of Activator RNA:

"Ultrastructural Probes of Active DNA Sites, and the RNA Activators of DNA". (PowerPoint Presentation).