Regulation of DNA transposition by CpG methylation and chromatin structure in human cells
© Jursch et al.; licensee BioMed Central Ltd. 2013
Received: 18 December 2012
Accepted: 19 April 2013
Published: 15 May 2013
Skip to main content
© Jursch et al.; licensee BioMed Central Ltd. 2013
Received: 18 December 2012
Accepted: 19 April 2013
Published: 15 May 2013
The activity of transposable elements can be regulated by different means. DNA CpG methylation is known to decrease or inhibit transpositional activity of diverse transposons. However, very surprisingly, it was previously shown that CpG methylation of the Sleeping Beauty (SB) transposon significantly enhanced transposition in mouse embryonic stem cells.
In order to investigate the unexpected response of SB transposition to CpG methylation, related transposons from the Tc1/mariner superfamily, that is, Tc1, Himar1, Hsmar1, Frog Prince (FP) and Minos were tested to see how transposition was affected by CpG methylation. A significant increase of >20-fold in transposition of SB, FP and Minos was seen, whereas Tc1, Himar1 and Hsmar1 showed no difference in transposition upon CpG-methylation. The terminal inverted repeats (TIRs) of the SB, FP and Minos elements share a common structure, in which each TIR contains two functionally important binding sites for the transposase (termed the IR/DR structure). The group of IR/DR elements showed increased excision after CpG methylation compared to untreated transposon donor plasmids. We found that de novo CpG methylation is not required for transposition. A mutated FP donor plasmid with depleted CpG sites in both TIRs was as efficient in transposition as the wild-type transposon, indicating that CpG sites inside the TIRs are not responsible for altered binding of factors potentially modulating transposition. By using an in vivo one-hybrid DNA-binding assay in cultured human cells we found that CpG methylation had no appreciable effect on the affinity of SB transposase to its binding sites. However, chromatin immunoprecipitation indicated that CpG-methylated transposon donor plasmids are associated with a condensed chromatin structure characterized by trimethylated histone H3K9. Finally, DNA compaction by protamine was found to enhance SB transposition.
We have shown that DNA CpG methylation upregulates transposition of IR/DR elements in the Tc1/mariner superfamily. CpG methylation provokes the formation of a tight chromatin structure at the transposon DNA, likely aiding the formation of a catalytically active complex by facilitating synapsis of sites bound by the transposase.
Co-evolution of transposable elements (TEs) with their host species gave rise to several mechanisms that regulate the transposition reaction , for example, cell or tissue type [2, 3], cell-cycle timing , transcriptional regulation [5–7], posttranscriptional regulation by small RNAs [8–10], regulation of the transposase protein , interactions with DNA repair , and target site selectivity. The activity of TEs can be regulated by chromatin at different stages of the transpositional reaction. For example, the formation of a catalytically active synaptic complex requires expression of the transposase and a DNA topology that makes the element accessible for the protein machinery required for catalysis. The eukaryotic genome is typically organized into either of two types of chromatin: euchromatin, a relatively relaxed chromatin structure, in which the DNA is packed less tightly and heterochromatin, a more inaccessible and highly condensed fraction of the genome. Heterochromatic regions carry characteristic features, which distinguish them from euchromatic DNA, such as dense cytosine-methylation (5-Me-C) of CpG sites, hypo-acetylation of lysine residues in the N-terminal tails of histones H3 and H4 and methylation of specific lysine residues such as lysine 9 in histone H3. In contrast to euchromatin, which is largely composed of unique (protein coding) sequences, the DNA sequence of heterochromatin is usually repetitive and gene poor . One class of repetitive sequences found in heterochromatic regions of different genomes are TEs, and therefore it is believed that the accumulation of transposable DNA sequences in heterochromatic regions provides a safe place, where the deleterious potential of these elements can be kept on a leash . Indeed, there is a strong correlation between chromatin structure and the activity of TEs. For example, recruiting transposable DNAs into heterochromatic regions may provide efficient silencing of transcription of element-encoded proteins, and thus provides genome stability. In addition to its repressive function on transcription, heterochromatin also exerts a repressive influence on recombination [15, 16]; hence, the containment of repeated sequences in heterochromatic regions may prevent irregular recombination and genome instability.
CpG methylation is known to decrease or inhibit transpositional activity of diverse transposons. However, very surprisingly, Yusa et al. showed that CpG methylation of the Sleeping Beauty (SB) transposon and to a smaller extent the Tc3 element of Caenorhabditis elegans produced elevated transpositional activity in mouse embryonic stem (ES) cells . Chromatin immunoprecipitation experiments revealed that hyperactive genomic donor sites have the characteristics of a heterochromatic structure. The SB transposase was found to co-localize with heterochromatin protein 1 (HP1), a well-established marker for heterochromatin, suggesting that the transposase preferentially associates with heterochromatic DNA . Based on these results, it was postulated that heterochromatin formation at the transposon donor site can upregulate SB transposition .
Transiently transfected plasmids can become CpG-methylated in the cell de novo. CpG methylation in mammals is catalyzed by three DNA-methyltransferases (Dnmts): Dnmt1 is responsible for the maintenance of CpG methylation patterns, and Dnmt3a and Dnmt3b perform de novo CpG methylation. We tested the excision activity of untreated and CpG-methylated SB transposon donors in HCT116 knockout/knockdown cell lines lacking either Dnmt1 or Dnmt3b or both [28, 29] by PCR-based excision assay (Figure 5B). All cell lines supported transposon excision. However, in these cells Dnmt3a is still active and could provide de novo CpG methylation of transfected donor plasmids. Thus, to inhibit the activity of Dnmt3a, we used the chemical aza-deoxycytidine, which is a known inhibitor of cellular de novo CpG methylation. Aza-deoxycytidine was added to cultured, HCT116-derived Dnmt3b knockout cells in a range of 1 μM to 1 mM two days prior to co-transfection of the transposon plasmids. The PCR-based excision assay (Figure 5C) showed that transposon excision is still detectable at high concentrations of aza-deoxycytidine. Even though PCR band intensity became weaker at increasing aza-deoxycytidine concentrations, this is likely due to the cytotoxic effects of aza-deoxycytidine, which is accompanied by reduced cell survival and cell growth. The enhancing effect of in vitro CpG methylation was not affected by aza-deoxycytidine treatment. We conclude that host-mediated de novo CpG methylation is not required for transposition.
As quantified by bacterial transformation, anti-acetylated histone H3 (anti-AcH3) antibodies precipitated threefold less CpG-methylated transposon plasmids than non-methylated transposon plasmids, suggesting an enrichment of non-methylated plasmids in open chromatin (Figure 6A). Conversely, immunoprecipitation with anti-trimethylated histone H3 lysine 9 (anti-H3triMeK9) antibodies resulted in reduced recovery of non-methylated plasmids compared to CpG-methylated plasmids, implying an enrichment of condensed chromatin status for CpG-methylated plasmids. PCR analyses of the immunoprecipitated DNA samples showed, in accordance with the results obtained by bacterial transformation, a relative enrichment of non-methylated plasmids in the euchromatic fraction (precipitation with anti-AcH3 antibody) and a relative enrichment of methylated plasmids in the heterochromatic fraction (precipitation with anti-H3triMeK9 antibody) (Figure 6B). We conclude that CpG-methylated plasmids are associated with condensed chromatin.
To test if DNA condensation is an underlying structural determinant of enhanced transposition in our experiments, we sought to examine the effect of condensation introduced into transposon DNA by means other than CpG methylation. We opted to package and condense transposon donor plasmids using protamine. Protamines are small, basic, arginine-rich peptides that largely replace histones in sperm chromatin, and package DNA into the most condensed eukaryotic DNA known .
The original finding that CpG methylation enhances Sleeping Beauty transposition  was very surprising, as CpG methylation has been known to play a crucial role in the cellular defense against TEs. We now extend these observations to other members of the Tc1/mariner superfamily, and show that this phenomenon is not restricted to SB, but seems to be an intrinsic feature associated with the characteristic IR/DR structure of the SB, Frog Prince and Minos elements (Figure 2).
We tested several hypotheses that provided possible explanations for the enhancing effect of CpG methylation on transposition. It is formally possible that CpG methylation is a prerequisite for transposition of SB, FP and Minos. In this case donor plasmids that are CpG-methylated in vitro would have a head start against non-methylated donors, which would need to be methylated by cellular factors following transfection. We tested transposition of non-methylated donor plasmids in Dnmt1 and/or Dnmt3b knockout/knockdown cell lines and in the presence of aza-deoxycytidine, which blocks de novo CpG methylation. Transposition was as efficient under these conditions as in wild-type or untreated cells (Figure 5), thus we conclude that CpG methylation is not required for transposition.
An alternative hypothesis predicts that one (or several) particular CpG sites inside the TIRs might affect CpG methylation. An enhancer of transposition might possibly be attracted by methylated CpG sites or, alternatively, a potential repressor might be distracted. However, a mutated FP donor plasmid with depleted CpG sites in both TIRs was as efficient in transposition as the wild-type transposon (Figure 4). Thus, CpG sites inside the TIRs do not seem to be responsible for the binding of any transposition-enhancing factor or the blocking of any transposition repressor.
Binding of the SB, FP and Minos transposases might directly be supported by CpG sites inside the TBSs, the TIRs or close to them. Alternatively, CpG methylation could increase transposition indirectly; for example by the formation of heterochromatin following binding of methylated CpG sites by the HP1 or MeCP2 (a protein that binds specifically to methylated DNA) proteins [31, 32]. Yusa et al. tested SB transposase binding to TBSs in vitro and did not detect any influence from CpG methylation. In line with these previous observations, the expression of luciferase as a reporter for TBS binding in an in vivo one-hybrid assay was affected by CpG methylation to the same extent in all background controls, binding experiments and under competition conditions (Figure 3). We thus conclude that CpG methylation had neither a direct nor an indirect effect on SB transposase binding.
CpG methylation/heterochromatin is one of the regulatory mechanisms that silence and inhibit TE activity. The potential of TEs to escape a regulatory mechanism imposed by the host is a strong evolutionary advantage (at least for the transposon). Assuming that the transposase source is provided by a transcriptionally active element located in euchromatin, host-cell induced CpG methylation/heterochromatin-based silencing of TEs can be offset by higher transposition efficiency out of condensed chromatin, thereby constituting a potential mechanism for SB and other, similar-structured transposons to escape CpG methylation-mediated silencing.
A 200 bp-minivariant of Himar1 (GenBank #U11644) was amplified out of pMM2611  via PCR using the primer Himar1/IR (TAACAGGTTGGCTGATAAGTCCCC). A 1.6 kb fragment of Tc1 (GenBank #X01005) was amplified out of pTc1_Ex/01 with primer Tc1/IR (TACAGTGCTGGCCAAAAAGATATCC). The PCR products were purified, phosphorylated and ligated into a SmaI digested, dephosphorylated vector pUC19. After transformation in E. coli DH10B cells, plasmids were isolated, test digested and sequenced. pUC19-Himar1 was SmaI digested, pUC19-Tc1 was StyI digested; both were dephosphorylated and ligated with the Klenow-treated gene trap cassette (GTC) (HindIII-NotI-XmnI digest of plasmid Δ170_CMV_zeo#1) . After transformation in E. coli DH10B, plasmids were prepared and control digested to verify the expected plasmid layout. For Minos (GenBank #X61695) the plasmid pMiLRneo was HindIII-NotI digested and ligated with the formerly described HindIII-NotI fragment of the GTC. The TIRs harboring the GTC were amplified out using PCR primer Minos/IR (TACGAGCCCCAACCACTATTAATTC). The PCR product was purified and ligated into SmaI digested pUC19. The product was checked by test digestion and sequencing. pSB-GTC (= GT/neo_CMV/zeo #2), derived from pT/neo , pFP-GTC (= pFP/GT-neo)  and pHsMar1-GTC and pHsmar1-GTCrev  were provided by C Miskey. The plasmids containing the transposases have been previously published: Tc1: pCMV/Tc1 , Himar1: pCMV/Himar3x , pSB10: , Minos: pJGD/ILMi , FP: pFV-FP , Hsmar1: pHsmar1. pFPΔCpG-GTC: FP TIRs on plasmid pFP-GTC were mutated using the Stratagene QuikChange Multi Site Directed Mutagenesis Kit following the manufacturer’s instructions with phosphorylated primers FP/IR_CpG_delete_syn (P-TGTTTGTCACACTTAA-GTGTTTCA GAACATCAAA-CCAATTTAAACAATAG) and FP/IR_CpG_delete_anti (P-CTATTGTTTAAATTGGTTTGATGTTCTG AAACA-CTTAAGTGTGACAAACA). The mutated plasmid pFPΔCpG-GTC was transformed in QuikChange XL1-Blue Supercompetent Cells (Stratagene), purified and sequenced.
Donor plasmids were CpG-methylated by SssI CpG methylase (NEB), and purified with the Qiagen PCR purification kit. Complete methylation was tested by control digests using methylation-sensitive restriction enzymes NotI (pMiLRgeo) or SalI (all others), and compared to a control digest of the respective untreated donor plasmids.
HeLa cells were cultured in DMEM with 4.5 g/L glucose and 110 mg/L pyruvate supplemented with 10% fetal calf serum (FCS) and an antibiotic/antimycotic cocktail (DMEM+/+) at 37°C. Cells were passaged via trypsinization with 1:5 dilution of 0.5% trypsin in 5.3 mM ethylenediaminetetraacetic acid (EDTA). Transfections were done at 60% to 80% cell confluency in DMEM without antibiotics (DMEM+/−) with the Fugene6 transfection reagent (Roche). For transposition assays, 6 × 105 cells were transfected with CpG-methylated or untreated donor plasmids (500 ng) together with transposase expression plasmids (50 ng). Complexing with protamine sulfate was done as described previously  by pre-incubating 500 ng plasmid DNA with 1 μg protamine sulfate (Sigma-Aldrich) for 15 min at room temperature, followed by the addition of Fugene6 as described above.
Cells (6 × 105) were transfected with equal amounts of CpG-methylated or untreated donor plasmids plus equal amounts of transposase-carrying helper or control plasmids. Typically 250 ng or 500 ng of donor plasmid was used and 50 ng of helper plasmid. Two days after transfection, the cells were harvested, plasmid DNA was prepared with the Qiagen Miniprep kit in a volume of 50 μL. A 1:10 dilution of 1 μL of the extracted DNA served as template for input normalization PCR utilizing primers Amp-For (TGCACGAGTGGGTTACATCGAACT) and Amp-Rev (TTGTTGCCATTGCTACAGGCATCG). Normalized DNA amounts served as templates for nested PCR utilizing pUC2 (GCGAAAGGGGGATGTGCTGCAAGG) and pUC5 (TCTTTCCTGCGTTATCCCCTGATTC) in the first, and pUC19-3F (GTTTTCCCAGTCACGACGTT) and pUC19-3R (TGTGGAATTGTGAGCGGATA) in the second PCR. PCR protocol: 95°C 3 min, 30 cycles of 95°C 30 s, 58°C 20 s, 64°C 10 s, and 72°C 2 min. The PCR products were subjected to gel electrophoresis.
Two days prior to transfection cells were treated with different amounts of 5-aza-2′-deoxycytidine (aza-dC), ranging from 1 μM to 1 mM. Aza-dC addition was repeated daily after previous medium exchange. Then, 6 h before transfection, aza-dC was removed from the cells and added again 6 h post-transfection.
HeLa cells (6 × 105) were transfected with 300 ng CpG-methylated or untreated p5SB-luc reporter plasmid, 90 ng pc-SB(123)-AD activator plasmid  and 1.1 μg pCMV-SB10 competitor plasmid (or filled with adequate control plasmids). Two days later, transfected cells were washed, treated with CCLR buffer (5× buffer: 125 mM Tris*H3PO4 pH 7.8, 10 mM dithiothreitol (DTT), 10 mM 1,2 cyclohexane-diaminetetra-acetic acid (CDTA), 50% glycerol, 5% Triton X-100), incubated for 15 min on ice and vortexed for 15 s. Cell debris was pelleted for 2 min with 12,000g at 4°C. 30 μL of the supernatant was combined with 100 μL luciferase buffer (20 mM tricine-NaOH pH7.8, 1.07 mM Mg(CO3)4 Mg(OH)2*5H2O, 2.67 mM MgSO4, 0.1 mM EDTA, 470 μM luciferin, 530 μM ATP, 33.3 mM DTT), vortexed briefly and measured in the luminometer.
Cells (4×106) were transfected in either four 5-cm or one 10-cm culture dish with approximately 3 μg CpG-methylated or non-methylated pFP-GTC donor plasmid. 24 h after incubation, 37% formaldehyde was added to the medium to an end concentration of 1%. The culture dishes were put into a plastic bag, sealed and incubated for 10 min at 37°C. After removal of the medium and two washing steps with ice-cold PBS containing protease inhibitors (1 mM PMSF, 1 μg/mL aprotinin, 1 μg/mL pepstatin A), the cells were scraped, transferred into 1.5 mL Eppendorf tubes and centrifuged for 4 min at 4°C at 2,000g. The pellet was resuspended in 400 μL lysis buffer [1% sodium dodecyl sulfate (SDS), 10 mM EDTA, 50 mM Tris–HCl pH 8.1, protease inhibitors]. Each sample was split into 2× 200 μL samples, incubated for 10 min on ice and centrifuged for 10 min at 4°C at 13,000 rpm. Supernatants were transferred to fresh 2 ml Eppendorf tubes and 1,800 μL ChIP dilution buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris–HCl pH 8.1, 167 mM NaCl, protease inhibitors) and 80 μL salmon sperm DNA/Protein A agarose-50% slurry (0.2 mg/mL sonicated salmon sperm DNA, 0.5 mg/mL BSA, approximately 1.5 mg/mL recombinant Protein A, 0.05% sodium azide) were added. The Eppendorf tubes were incubated for 30 min on a rotating plate at 4°C. The agarose was pelleted using brief centrifugation and the supernatants were transferred to new tubes. Polyclonal antibodies anti-acetylated histone H3 (anti-AcH3) (provided with the ChIP Kit, Upstate #06-599, Lot 27610) or anti-trimethylated histone H3 lysine 9 (anti-H3triMeK9) (Abcam #ab8898) were added in varying amounts (4 μL to 10 μL) and incubated overnight at 4°C on a rotating plate. The next day, 60 μL of salmon sperm DNA/Protein A agarose-50% slurry was added and incubated for 1 h at 4°C with rotation. The agarose was pelleted at 900 rpm for 1 min at 4°C and washed with low salt immune complex wash buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris–HCl pH 8.1, 150 mM NaCl), high salt immune complex wash buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris–HCl pH 8.1, 500 mM NaCl), LiCl immune complex wash buffer [0.25 M LiCl, 1% Nonidet P-40 (NP40), 1% deoxycholate, 1 mM EDTA, pH 8.0] and twice with TE buffer (10 mM Tris–HCl, 1 mM EDTA, pH 8.0). The chromatin was eluted by the repeated addition of 250 μL elution buffer (1% SDS, 0.1 M NaHCO3) to the pellet, vortexing and 15 min incubation followed by pelleting the agarose and saving the supernatant in a fresh 1.5 mL tube. To the combined (500 μL) eluates, 20 μL 5 M NaCl was added and heated for 4 h at 65°C to reverse cross-linking (samples were stored at this point overnight at −20°C). Using 10 μL 0.5 M EDTA, 20 μL 1 M Tris–HCl pH 6.5 and 2 μL proteinase K, histones and other DNA-bound proteins were digested for 1 h at 45°C. The remaining residues were extracted twice with phenol/chloroform and the DNA was precipitated with isopropanol (standard protocol). The DNA was taken up in H2O and re-purified by dialysis. For the quantification of plasmids, the extracted DNA was either transformed into electrocompetent E. coli DH10B or used in a semi-quantitative PCR (primer Amp-For and Amp-Rev). Transformed cells were plated on ampicillin/zeocin/LB (Luria-Bertani) agar and incubated overnight at 37°C. Colonies were counted the next morning.
Anti-acetylated histone H3
Anti-trimethylated histone H3 lysine 9
Bovine serum albumin
1,2 cyclohexane-diaminetetra-acetic acid
Dulbecco’s modified Eagle’s medium
Fetal calf serum
Gene trap cassette
Heterochromatin protein 1
Polymerase chain reaction
Sodium dodecyl sulfate
Transposase binding site
Terminal inverted repeat.
We thank B Savakis for kindly providing Minos transposon vectors. The luciferase-based Sleeping Beauty reporter and activator plasmids were obtained from S Yant. TJ was funded by the International MDC/HU PhD Program.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.