- Research article
- Open Access
Complementation between inactive fragments of SssI DNA methyltransferase
© Slaska-Kiss et al.; licensee BioMed Central Ltd. 2012
Received: 2 March 2012
Accepted: 30 May 2012
Published: 30 May 2012
Silencing mammalian genes by targeted DNA (cytosine-5) methylation of selected CG sites in the genome would be a powerful technique to analyze epigenomic information and to study the roles of DNA methylation in physiological and pathological states. A promising approach of targeted DNA methylation is based on the ability of split fragments of a monomeric DNA methyltransferase (C5-MTase) to associate and form active enzyme. A few C5-MTases of different specificities have been shown to possess the ability of fragment complementation, but a demonstration of this phenomenon for a C5-MTase, which has CG specificity and thus can be targeted to methylate any CG site, has been lacking. The purpose of this study was to test whether the CG-specific prokaryotic C5-MTase M.SssI shows the phenomenon of fragment complementation.
We show that truncated inactive N-terminal fragments of M.SssI can assemble with truncated inactive C-terminal fragments to form active enzyme in vivo when produced in the same E. coli cell. Overlapping and non-overlapping fragments as well as fragments containing short appended foreign sequences had complementation capacity. In optimal combinations C-terminal fragments started between conserved motif VIII and the predicted target recognizing domain of M.SssI. DNA methyltransferase activity in crude extracts of cells with the best complementing fragment pairs was ~ 4 per cent of the activity of cells producing the full length enzyme. Fusions of two N-terminal and two C-terminal fragments to 21.6 kDa zinc finger domains only slightly reduced complementation ability of the fragments.
The CG-specific DNA methyltransferase M.SssI shows the phenomenon of fragment complementation in vivo in E. coli. Fusion of the split fragments to six unit zinc finger domains does not substantially interfere with the formation of active enzyme. These observations and the large number of complementing fragment combinations representing a wide range of MTase activity offer the possibility to develop M.SssI into a programmable DNA methyltransferase of high specificity.
Cytosine C5-methylation plays important roles in several biological phenomena, such as restriction-modification in prokaryotes, genomic imprinting and carcinogenesis in eukaryotes . This reaction is catalyzed by DNA (cytosine-5) methyltransferases (C5-MTase), which transfer a methyl group from S-adenosyl-methionine (AdoMet) to carbon 5 of cytosine in specific DNA sequences . Prokaryotic C5-MTases contain ten conserved sequence motifs and a so-called variable region located between conserved motifs VIII and IX . The conserved motifs are responsible for the general steps of the methyl transfer reaction [4–9], whereas specific sequence recognition is mediated mainly by the variable region . Our understanding of the three-dimensional structure of C5-MTases and of their interaction with substrate DNA is mainly based on the X-ray structures of two enzymes: M.HhaI and M.HaeIII. They revealed that both MTases fold in two domains, the large domain encompassing most of the conserved motifs and the small domain containing the variable region and conserved motif IX. The two domains form a cleft where the DNA substrate fits with the major groove facing the small domain and the minor groove facing the large domain. All specific DNA-protein interactions are at the small domain – major groove interface [12, 13]. Eukaryotic C5-MTases are larger proteins but the sequence homology they share with prokaryotic C5-MTases and the available biochemical data suggest that they have the same catalytic mechanism [14, 15].
Although the vast majority of characterized C5-MTases function as monomers, there are exceptions: M.AquI (C YCGRG) and M.EcoHK31I (YGGC CR) consist of two polypeptides. The larger subunit of M.AquI contains conserved motifs I – VIII and part of the variable region, whereas the smaller subunit contains the distal half of the variable region and conserved motifs IX - X [16, 17]. In M.EcoHK31I, the larger subunit encompasses conserved motifs I – VIII, X as well as the predicted target recognition domain (TRD), and only motif IX is located in the smaller polypeptide . The structural plasticity of C5-MTases is also supported by the phenomenon of protein fragment complementation observed with three enzymes: N- and C-terminal inactive fragments of three naturally monomeric C5-MTases (M.BspRI, M.BsuRI and M.HhaI) can assemble to form active MTase if expressed in the same E. coli cell, [19, 20].
The goal of this work was to test whether M.SssI, which has the same specificity (CG) as the eukaryotic DNA MTases  and therefore has special importance as an experimental tool in the study of eukaryotic DNA methylation, is capable of fragment complementation. In higher eukaryotes, DNA methylation occurs at CG dinucleotides (CpG sites) and is associated with gene silencing . Targeted DNA methylation, i.e. selective methylation of predetermined CpG sites in the genome is emerging as a promising technique for selective gene silencing [23–27]. The applicability of targeted methylation as a research tool or as a potential therapeutic approach critically depends on the specificity of targeting, i. e. on the difference of methylation between targeted and non-targeted sites. One approach to increase targeting specificity capitalized on the phenomenon of functional complementation between inactive fragments of the MTase. In the first implementation of this technique complementing N- and C-terminal fragments of the HhaI MTase were genetically fused to zinc finger proteins (ZFP) engineered to recognize different nine bp sequences. When the MTase fragment-ZFP fusion proteins were expressed in the same E. coli cell, the targeted M.HhaI recognition site, which was flanked by the two closely spaced ZFP binding sites, became methylated, whereas the other M.HhaI recognition sites on the same plasmid stayed unmethylated . Although this strategy is likely to require improvement to suppress the non-targeted background methylation mainly deriving from the reconstitution of the MTase in unbound state , it probably remains the most promising approach for achieving the specificity required for using targeted methylation as a reliable research tool [30, 31]. However, of the C5-MTases shown to possess the capacity of fragment complementation, only M.HhaI can be used to target CpG sites, and even M.HhaI can methylate only a small subset of CpG sites (1 in 16, those in GC GC context). To be able to target any CpG site, one needs a C5-MTase with CG specificity such as M.SssI.
Here, we report that M.SssI shows the phenomenon of fragment complementation, thus it is, in principle, suitable for the split fragment approach of targeted DNA methylation.
Construction of plasmids expressing truncated fragments of M.SssI
Plasmids encoding N- or C-terminal fragments of M.SssI
6-ZFP-B zinc finger domain
6-ZFP-B zinc finger domain
6-ZFP-A zinc finger domain
6-ZFP-A zinc finger domain
6-ZFP-A zinc finger domain
Functional complementation between inactive fragments of M.SssI in vivo
Since the truncated genes of the complementing N- and C-terminal polypeptides contained overlapping segments, it was important to exclude the possibility that MTase activity arose by reconstitution of the intact MTase gene via homologous recombination. To test this, the araBAD promoter and part of the araC gene was deleted from pOB-Sss[191–386] and pOB-Sss[241–386] leaving the [191–386] and [241–386] coding sequence intact. The resulting plasmids (pSss[191–386]del and pSss[241–386]del) did not complement pBNH-Sss[1–304] (not shown) indicating that for restoration of MTase activity synthesis of the C-terminal peptide was required, and a recombination mechanism can be excluded.
In a reciprocal approach, nested deletions were introduced from the 3’-end of the M.SssI gene. Plasmids expressing the following polypeptides were isolated: [1–176], [1–222], [1–239], [1–248], [1–257], [1–317], [1–318], [1–324] and [1–356] (Table 1 and Figure 1). Because in these plasmids the truncated gene was fused to vector sequence, translation termination is determined by the next in-frame stop codon downstream of the PaeI site (Figure 2), and the fragments carry short C-terminal appendages of foreign sequence (Table 1). These plasmids were tested in combination with pOB-Sss[241–386] and pOB-Sss[242–386], two of the plasmids that most efficiently complemented the N-terminal polypeptide [1–304]. The two shortest fragments ([1–176] and [1–222]) did not complement [241–386] (Figure 4B). Complementation by fragment [1–239] was rather efficient, whereas only poor methylation activity was observed with fragments [1–248] and [1–257]. The low complementation capacity of the latter two fragments is probably due to the relatively long foreign sequence at their C-terminus (Table 1). All longer fragments were active in the complementation test and resulted in comparable levels of methylation (Figure 4B). Similar results were obtained when [242–386] was used as C-terminal fragment although in this combination fragments [1-248/257] appeared more active than with [241–386] (Figure 4C vs. 4B). The complementation observed with the [1–239] + [241/242-386] pairs is noteworthy because in these cases there is no overlap between the fragments.
It has been observed in several systems that sequence overlap between the fragment pairs was beneficial or even essential for complementation [20, 37, 38]. To test this requirement for M.SssI in more detail, a series of C-terminal fragments were tested in combination with [1–239]. Fragments [240/241/242-386] were highly active in pair with [1–239], whereas the few residues shorter [246–386] fragment was already ineffective (Figure 4D). It was shown before that the [1–189] + [191–386] and the [1–304] + [304–386] combinations were inactive (see above). Taking into account that fragments [1–304] and [191–386] were active in combination with fragment(s), which they overlapped, these observations show that the requirement for overlap depends on the location of the break. If the bisection point is in the proximal half of the variable region, no overlap is required, and even the loss of a few amino acids can be compatible with MTase activity. In contrast, if the bisection point is in regions less optimal for fitting the two polypeptides together, a short overlap between the fragment pairs appears to be necessary for the formation of the active complex.
To compare M.SssI activities in vivo between representatives of the best complementing fragment pairs and intact M.SssI, extracts were prepared from arabinose induced cells and MTase activity was determined using radioactively labeled AdoMet as described in Methods. The following MTase activities (derived from three independent experiments and expressed in cpm/μg extract protein) were obtained (with the enzyme in parentheses): 1016 ± 369 (full-length), 16 ± 0 ([1–304]), 53 ± 14 ([1–304] + [242–386]) and 65 ± 8 ([1–304] + [246–386]). Assuming similar levels of expression, these values suggest that the activities of even the best complementing pairs is below 5 per cent of that of the intact enzyme.
In this study we identified N- and C-terminal fragments of the SssI MTase, which can assemble with a counterpart fragment to form active MTase when produced in the same E. coli cell. All available evidence supports the conclusion that the observed phenomenon is indeed peptide complementation and not the result of DNA recombination restoring the intact MTase gene. Firstly, deletion of the promoter of two C-terminal fragments abolished complementation (see above). Secondly, plasmid pairs with longer overlaps between the 5’- and 3’ gene segments were not necessarily more efficient in complementation than fragments with shorter overlaps (Figure 4A). Thirdly, in the experiment in which the [191–386] fragment was shortened from the N-terminus by exonuclease III digestion, many of the randomly picked clones contained M.SssI gene segments with substantial overlap with the gene of the [1–304] fragment but fused to the ATG initiation codon in incorrect reading frame. When these plasmids were co-transformed with pBNH-Sss[1–304], methylation was not detectable, confirming that M.SssI activity required synthesis of the C-terminal peptide (not shown). Finally, complementation observed between some plasmids containing non-overlapping gene segments such as [1–239] + [241/242-386] (Figures 4D) also argues against the recombination model.
Although a large number of fragment pairs resulted in detectable M.SssI activity, the most efficient methylation was observed with fragment combinations involving C-terminal fragments starting between conserved motif VIII and the assumed TRD, with a sharp optimum between residues 241 and 246 (Figures 1 and 5). In lack of an X-ray structure it is difficult to rationalize these results but the location of split sites in the region connecting the predicted large and small domains  would be consistent with the general principle that fragment complementation is most efficient between folded units . The flexible random coil conformation predicted for the segment Glu241 - Ile262  could also be an advantage for fitting the two subunits together. It is noteworthy that the beginning (K210) of the C-terminal M.HhaI fragment producing complementation was in the same region, between motif VIII and the TRD .
Evaluation of the phenotypes of the N-terminal fragments generated by exonuclease III digestion is less straightforward because the extraneous C-terminal peptides appended by the cloning procedure (Table 1) are likely to interfere with assembly of the subunits. In contrast to the C-terminal fragments, the complementation ability of the N-terminal fragments did not show a clear optimum. All overlapping fragments, which did not have longer extraneous C-terminal peptide could complement fragments [241–386] and [242–386] and the complementation capacity of these fragments appeared to be similar (Figures 4 B and C).
This work was motivated by the desire to develop M.SssI into a programmable DNA MTase suitable to methylate unique CpG sites in the genome. Here, as a first step toward this goal, the ability of M.SssI for fragment complementation has been demonstrated and fragment combinations displaying different levels of MTase activity in vivo have been identified. These results open the possibility for developing a system for targetable CpG methylation using the split fragment approach, which will involve construction of fusions between fragments of M.SssI and appropriately designed ZFP targeting domains. The split fragment approach is in many ways reminiscent of the technique using ZFP-FokI chimeric nucleases for targeted DNA cleavage . Low activity of the heterodimers can be an advantage for increasing targeting specificity especially if it derives from the low affinity between the MTase halves . Ideally, the split MTase fragments should not be able to assemble in free state at cellular concentrations but should form active heterodimers due to increased local concentration when co-localized to the target site via the ZFP domains. The best strategy to create a site specific MTase could be to start with fragments, which cannot associate into active MTase as has been suggested . The level of targeting specificity required for methylation of unique sites in the genome will probably need extensive optimization of different parameters governing targetable methylation specificity, such as affinity to target and non-target DNA, affinity between the MTase subunits, distance between the ZFP binding site and the targeted CG, etc.. The large number of complementing fragment combinations representing a wide range of MTase activity will be a rich source of starting material to engineer ZFP-M.SssI fusions for programmable DNA methylation.
The CG-specific M.SssI, similarly to three other naturally monomeric C5-MTases of different specificity, shows the phenomenon of fragment complementation in vivo in E. coli. Fusion of the split fragments to six unit zinc finger domains only slightly decreases the efficiency of complementation. These observations offer the possibility to develop M.SssI into a programmable DNA methyltransferase of high specificity that can be useful in the study of DNA methylation in higher eukaryotes.
Strains and growth conditions
The Escherichia coli strain DH10B F−endA1 recA1 galE15 galK16 nupG rpsL ΔlacX74 80dlacZ ΔM15 araD139 Δ(ara leu)7697 mcrA Δ(mrr-hsdRMS-mcrBC) relA1 spoT1 λ− was used as general cloning host and CJ236 FΔ(HindIII)::cat (Tra+ Pil+ CamR)/ung-1 relA1 dut-1 thi-1 spoT1 mcrA as host for site-directed mutagenesis.
Cells were grown in LB medium at 30 or 37 °C. Ampicillin (Ap) and kanamycin (Kn) were used at 100 and 50 μg/ml, respectively. L-arabinose (Sigma) was used at 0.5-0.005 % as indicated at the experiments.
pBAD24 is an expression vector plasmid, in which transcription of the target gene is tightly controlled by the araBAD promoter and the AraC protein . pOK-BAD, a Kn-resistant derivative of pBAD24, has p15A replicon, thus is compatible with ColE1-based plasmids such as pBAD24 .
All variants of the sssIM gene described in this paper are derived from the allele in the plasmid pCAL7 obtained from New England Biolabs. The sssIM allele in pCAL7, considered as the WT allele for this work, carries mutations that change four TGA stop codons, which normally specify Trp in the native host Spiroplasma sp., to TGG (W. Jack, personal communication). The pBAD24-based plasmids pB-MSssI and pBNH-MSssI encode the WT and an N-terminally His-tagged M.SssI variant, respectively (Figure 2).
Plasmids encoding N-terminal M.SssI fragments are based on the vector pBAD24, and have names starting with pB, whereas plasmids that encode C-terminal M.SssI fragments, are based on the vector pOK-BAD and carry names starting with pOB- (Table 1).
Plasmid pB-Sss[1–189] was constructed by deleting the Ecl136II – SmiI fragment of pB-MSssI (Figure 2). Ecl136II cuts after the codons of the conserved E186NV motif and SmiI after the stop codon. This deletion created a reading frame ending two triplets downstream of the Gly189 codon. Thus, the MTase fragment encoded by pB-Sss[1–189] carries a two amino acid C-terminal appendage ENVGEI (underlined).
Deoxyoligonucleotides used in this work
M.SssI gene sense strand, positions 566-586. Starts with a SacI site
Forward primer for [190-304]
M.SssI gene anti-sense strand, positions 912-895, stop codon and PstI site added
Reverse primer for [1-304]
M.SssI gene sense strand, positions 835-855, NcoI site added
Forward primer for [279-386]
M.SssI gene anti-sense strand, stop codon and PstI site added
Reverse primer for His6-Cys-tagged C-terminal fragments
M.SssI gene sense strand, positions 910-930, NcoI site added
Forward primer for [304-386]
M.SssI gene sense strand, positions 721-741, NcoI site added
Forward primer for [241-386]
6-ZFP-A gene antisense strand
Mutagenic primer for elimination of a HindIII site
Forward primer for amplification of the 6-ZFP-A gene
Reverse primer for amplification of the 6-ZFP-A gene
Plasmids pBNH-Sss[1–176], pBNH-Sss[1–222], pBNH-Sss[1–239], pBNH-Sss[1–248], pBNH-Sss[1–257], pBNH-Sss[1–317], pBNH-Sss[1–318], pBNH-Sss[1–324] and pBNH-Sss[1–356] are deletion derivatives of pBNH-MSssI and were generated by exonuclease III digestion. pBNH-MSssI was digested with Bpu1102I and PaeI, which have single sites in the plasmid downstream of the M.SssI gene (Figure 2). Unidirectional deletions starting from the Bpu1102I end (5’-overhang) were made by exonuclease III . The 3’-overhang of the PaeI end protected the downstream sequences. The M.SssI fragments expressed from the plasmids obtained by exonuclease digestion carry short C-terminal appendages (Table 1) determined by the fused vector sequence downstream of the PaeI site.
Plasmids pOB-Sss[191–386], pOB-Sss[241–386], pOB-Sss[279–386] and pOB-Sss[304–386] are based on the expression vector pOK-BAD and encode C-terminal M.SssI fragments. They were constructed either by direct fragment transfer or PCR amplification from a plasmid expressing a C-terminally His-tagged M.SssI variant. This variant and the derived C-terminal polypeptides used in this work carry a cysteine at the end of the His-tag: KIGGSHHHHHHC. (The C-terminal cysteine, not relevant to the present study, was added for future applications.) Plasmid pOB-Sss[191–386] was constructed by cloning the Ecl136II-PstI fragment encoding the His-tagged [191–386] polypeptide between the filled-in Acc65I site and the PstI site in the pOK-BAD polylinker. The variant encoded by pOB-Sss[191–386] has the following N-terminal sequence MVP L191LHKK. The underlined three amino acid extension is encoded by the pOK-BAD polylinker. To construct plasmids expressing the [241–386], [279–386] and [304–386] polypeptides, the respective gene segments were PCR-amplified using AK189, AK185 or AK188 as forward primer and AK186 as reverse primer (Table 2). In pOK-BAD the NcoI site overlaps the translational start codon. To facilitate cloning of the synthesized fragments in pOK-BAD, the forward primers were designed to carry a 5’-extension with an NcoI site. The amplified fragments were inserted between the NcoI and PstI sites of pOK-BAD.
Plasmids pSss[191–386]del and pSss[241–386]del were made by deleting the BssHII–NcoI fragment of pOB-Sss[191–386] and pOB-Sss[241–386], respectively. BssHII cuts in the middle of the araC gene and the NcoI site contains the ATG start codon of the truncated MTase genes (Figure 2). Thus, this deletion removes part of the araC gene along with the araBAD promoter and translational initiation signals for the [191–386] and [241–386] polypeptides.
Plasmids pOB-Sss[231–386], pOB-Sss[240–386], pOB-Sss[242–386], pOB-Sss[246–386], pOB-Sss[250–386], pOB-Sss[254–386], pOB-Sss[261–386], pOB-Sss[316–386] and pOB-Sss[321–386] are deletion derivatives of pOB-Sss[191–386] and were obtained by exonuclease III digestion. To have appropriate restriction sites for unidirectional exonuclease III treatment, pUC18 linearized with KpnI was inserted into the unique KpnI site of pOB-Sss[191–386] located in the gene section coding for the N-terminal MVP appendage (see above) i.e. between the start codon and the gene segment encoding the [191–386] fragment. A plasmid, in which the orientation of pUC18 was such that the unique SacI site was upstream (towards the translational start codon), whereas the unique SalI site was downstream, was chosen for further work. Part of the inserted pUC18, which was not needed for controlling the exonuclease III digestion, was deleted by ApaLI digestion and circularization of the large fragment to yield pOB-Sss[191–386]SS (Figure 2). To generate random 5’-deletions in the gene of the [191–386] fragment, pOB-Sss[191–386]SS was cut with SalI (5’-overhang) and SacI (3’-overhang) and treated with exonuclease III as described above. The SalI end allowed exonuclease III digestion to start whereas the SacI end protected upstream sequences, such as the PBAD promoter and the initiation codon. As byproducts of the cloning steps, some of the M.SssI fragments encoded by these plasmids carry short N-terminal extensions (Table 1).
The plasmids pcDNA3.1mnhk-up1 and pcDNA3.1mnhk-up2 encode the zinc finger proteins 6-ZFP-A and 6-ZFP-B, respectively , and were kindly provided by M. Rots. To construct fusions between the N-terminal M.SssI fragments [1–304], [1–239] and 6-ZFP-B, the 6-ZFP-B coding region was transferred, on an NcoI-XhoI fragment, from pcDNA3.1mnhk-up2, into pBNH-Sss[1–304] and pBNH-Sss[1–239] to yield pB6ZB-Sss[1–304] and pB6ZB-Sss[1–239], respectively. For fusions to C-terminal M.SssI fragments, the HindIII site in the 6-ZFP-A zinc finger gene was eliminated by site directed mutagenesis using the mutagenic primer AK258. The resulting 6-ZFP-A gene carrying the silent mutation was PCR-amplified using primers AK259 and AK260 containing PvuII and PstI sites as 5’-extensions, respectively (Table 2). The PCR product was T/A-cloned in pTZ57R/T, then excised by PvuII - PstI double-digestion and cloned between the HpaI and PstI sites of pBS-CAL75, a plasmid containing the 3’-half of the sssIM gene with an engineered HpaI site overlapping the M.SssI stop codon (Figure 2). Ligation of the PvuII end of the PCR fragment to the HpaI end of pBS-CAL75 creates in-frame fusion between the MTase and the 6-ZFP-A zinc finger domain (Figure 2, plasmid pBS-Sss6ZA). C-terminal fusions between the M.SssI fragments [241–386]/[261–386] and 6-ZFP-A were constructed by inserting the HindIII fragment of pBS-Sss6ZA containing the 3’-end of the M.SssI gene and the coding sequence of 6-ZFP-A (Figure 2) into the unique HindIII site located in the 3’ half of the M.SssI coding sequence (Figure 2) in pOB-Sss[241–386] and pOB-Sss[261–386] to yield pOB-Sss[241–386]-6ZA and pOB-Sss[261–386]-6ZA, respectively.
Nucleotide sequences of the plasmids are available upon request.
Restriction digestion, polymerase chain reaction, agarose gel electrophoresis and cloning in E. coli plasmid vectors was carried out using standard procedures . Oligonucleotides were purchased from IDT or were synthesized in this institute and are listed in Table 2. Restriction endonucleases, DNA polymerase large (Klenow) fragment, Taq DNA polymerase, exonuclease III, S1 nuclease and T4 DNA ligase were purchased from Fermentas or New England Biolabs. Site-directed mutagenesis was performed by the Kunkel method . Unidirectional deletions were generated by the combined use of exonuclease III, S1 nuclease, Klenow DNA polymerase and T4 DNA ligase . DNA sequence was determined by an automated sequencer (ABI).
Methyltransferase assay in crude extracts
E. coli cells with plasmids encoding M.SssI fragment(s) or the full-length enzyme were grown at 37 °C to OD550 ~ 0.5, then 0.1 % arabinose was added and culturing was continued for 3 hrs at 30 °C. Cells sedimented from 200 ml culture by centrifugation were resuspended in 0.2 ml 50 mM Tris–HCl pH 8.0, 10 mM 2-mercaptoethanol and disrupted by sonication. After removing cell debris by centrifugation, M.SssI activity was determined in 30 μl reactions containing 0.9 μg lambda phage DNA (dam-dcm-), 50 mM Tris–HCl pH 8.5, 50 mM NaCl, 10 mM EDTA, 5 mM DTT, 5 μM [methyl-3 H]AdoMet and 2 μl cell extract. The [methyl-3 H]AdoMet used in the reaction was prepared by diluting the 370 GBq/mmol preparation (NET155, New England Nuclear) to 116 GBq/mmol with unlabeled AdoMet (New England Biolabs). After incubation at 30 °C for 20 min, reactions were stopped by 1 % SDS, then the incorporated radioactivity was determined as described . Protein concentration was determined by the Bradford reaction (Sigma). MTase activity was calculated by dividing the cpm values by the amount of total protein in 2 μl cell extract (in μg) used in the assay.
We thank William Jack (New England Biolabs) for the plasmid pCAL7, Marianne Rots (University Medical Center Groningen) for plasmids encoding zinc finger domains, Tamás Raskó for pBS-CAL75 and for contribution to construct pBNH-MSssI. We also thank Ibolya Anton for the technical assistance. This work was supported by the K 83983 grant from the Hungarian Scientific Research Fund (OTKA) and by the COOP-CT-2005-017984 grant from the European Commission’s Sixth Framework Programme.
- Jeltsch A: Beyond Watson and Crick: DNA methylation and molecular enzymology of DNA methyltransferases. ChemBioChem. 2002, 3: 274-293. 10.1002/1439-7633(20020402)3:4<274::AID-CBIC274>3.0.CO;2-SView ArticlePubMedGoogle Scholar
- Kumar S, Cheng X, Klimasauskas S, Mi S, Pósfai J, Roberts RJ, Wilson GG: The DNA (cytosine-5) methyltransferases. Nucleic Acids Res. 1994, 22: 1-10. 10.1093/nar/22.1.1View ArticlePubMedPubMed CentralGoogle Scholar
- Pósfai J, Bhagwat AS, Pósfai G, Roberts RJ: Predictive motifs derived from cytosine methyltransferases. Nucleic Acids Res. 1989, 17: 2421-2435. 10.1093/nar/17.7.2421View ArticlePubMedPubMed CentralGoogle Scholar
- Wu JC, Santi DV: Kinetic and catalytic mechanism of HhaI methyltransferase. J Biol Chem. 1987, 262: 4778-4786.PubMedGoogle Scholar
- Chen L, MacMillan AM, Chang W, Ezaz-Nikpay K, Lane WS, Verdine GL: Direct identification of the active-site nucleophile in a DNA (cytosine-5)-methyltransferase. Biochemistry. 1991, 30: 11018-11025. 10.1021/bi00110a002View ArticlePubMedGoogle Scholar
- Estabrook RA, Lipson R, Hopkins B, Reich N: The coupling of tight DNA binding and base flipping: identification of a conserved structural motif in base flipping enzymes. J Biol Chem. 2004, 279: 31419-31428. 10.1074/jbc.M402950200View ArticlePubMedGoogle Scholar
- Shieh FK, Reich NO: AdoMet-dependent methyl-transfer: Glu119 is essential for DNA C5-cytosine methyltransferase M.HhaI. J Mol Biol. 2007, 373: 1157-1168. 10.1016/j.jmb.2007.08.009View ArticlePubMedGoogle Scholar
- Darii MV, Cherepanova NA, Subach OM, Kirsanova OV, Rasko T, Slaska-Kiss K, Kiss A, Deville-Bonne D, Reboud-Ravaux M, Gromova ES: Mutational analysis of the CG recognizing DNA methyltransferase SssI: Insight into enzyme-DNA interactions. Biochim Biophys Acta. 2009, 1794: 1654-1662.View ArticlePubMedGoogle Scholar
- Sankpal UT, Rao DN: Mutational analysis of conserved residues in HhaI DNA methyltransferase. Nucleic Acids Res. 2002, 30: 2628-2638. 10.1093/nar/gkf380View ArticlePubMedPubMed CentralGoogle Scholar
- Balganesh TS, Reiners L, Lauster R, Noyer-Weidner M, Wilke K, Trautner TA: Construction and use of chimeric SPR/phi3T DNA methyltransferases in the definition of sequence recognizing enzyme regions. EMBO J. 1987, 6: 3543-3549.PubMedPubMed CentralGoogle Scholar
- Klimasauskas S, Nelson JL, Roberts RJ: The sequence specificity domain of cytosine-C5 methylases. Nucleic Acids Res. 1991, 19: 6183-6190. 10.1093/nar/19.22.6183View ArticlePubMedPubMed CentralGoogle Scholar
- Klimasauskas S, Kumar S, Roberts RJ, Cheng X: HhaI methyltransferase flips its target base out of the DNA helix. Cell. 1994, 76: 357-369. 10.1016/0092-8674(94)90342-5View ArticlePubMedGoogle Scholar
- Reinisch KM, Chen L, Verdine GL, Lipscomb WN: The crystal structure of Haelll methyltransferase covalently complexed to DNA: An extrahelical cytosine and rearranged base pairing. Cell. 1995, 82: 143-153. 10.1016/0092-8674(95)90060-8View ArticlePubMedGoogle Scholar
- Jurkowska RZ, Jurkowski TP, Jeltsch A: Structure and function of mammalian DNA methyltransferases. ChemBioChem. 2011, 12: 206-222. 10.1002/cbic.201000195View ArticlePubMedGoogle Scholar
- Song J, Rechkoblit O, Bestor TH, Patel DJ: Structure of DNMT1-DNA complex reveals a role for autoinhibition in maintenance DNA methylation. Science. 2011, 331: 1036-1040. 10.1126/science.1195380View ArticlePubMedPubMed CentralGoogle Scholar
- Karreman C, de Waard A: Agmenellum quadruplicatum M.AquI, a novel modification methylase. J Bacteriol. 1990, 172: 266-272.PubMedPubMed CentralGoogle Scholar
- Pinarbasi H, Pinarbasi E, Hornby D: Recombinant alpha and beta subunits of M.AquI constitute an active DNA methyltransferase. J Biochem Mol Biol. 2002, 35: 348-351. 10.5483/BMBRep.2002.35.3.348View ArticlePubMedGoogle Scholar
- Lee KF, Kam KM, Shaw PC: A bacterial methyltransferase M.EcoHK311 requires two proteins for in vitro methylation. Nucleic Acids Res. 1995, 23: 103-108. 10.1093/nar/23.1.103View ArticlePubMedPubMed CentralGoogle Scholar
- Pósfai G, Kim SC, Szilák L, Kovács A, Venetianer P: Complementation by detached parts of GGCC-specific DNA methyltransferases. Nucleic Acids Res. 1991, 19: 4843-4847. 10.1093/nar/19.18.4843View ArticlePubMedPubMed CentralGoogle Scholar
- Choe W, Chandrasegaran S, Ostermeier M: Protein fragment complementation in M.HhaI DNA methyltransferase. Biochem Biophys Res Commun. 2005, 334: 1233-1240. 10.1016/j.bbrc.2005.07.017View ArticlePubMedGoogle Scholar
- Renbaum P, Abrahamove D, Fainsod A, Wilson GG, Rottem S, Razin A: Cloning, characterization, and expression in Escherichia coli of the gene coding for the CpG DNA methylase from Spiroplasma sp. strain MQ1(M.SssI). Nucleic Acids Res. 1990, 18: 1145-1152. 10.1093/nar/18.5.1145View ArticlePubMedPubMed CentralGoogle Scholar
- Klose RJ, Bird AP: Genomic DNA methylation: the mark and its mediators. Trends Biochem Sci. 2006, 31: 89-97. 10.1016/j.tibs.2005.12.008View ArticlePubMedGoogle Scholar
- Xu GL, Bestor TH: Cytosine methylation targetted to pre-determined sequences. Nat Genet. 1997, 17: 376-378. 10.1038/ng1297-376View ArticlePubMedGoogle Scholar
- Li F, Papworth M, Minczuk M, Rohde C, Zhang Y, Ragozin S, Jeltsch A: Chimeric DNA methyltransferases target DNA methylation to specific DNA sequences and repress expression of target genes. Nucleic Acids Res. 2007, 35: 100-112. 10.1093/nar/gkm575View ArticlePubMedPubMed CentralGoogle Scholar
- Smith AE, Hurd PJ, Bannister AJ, Kouzarides T, Ford KG: Heritable gene repression through the action of a directed DNA methyltransferase at a chromosomal locus. J Biol Chem. 2008, 283: 9878-9885. 10.1074/jbc.M710393200View ArticlePubMedGoogle Scholar
- Jurkowska RZ, Jeltsch A: Silencing of gene expression by targeted DNA methylation: concepts and approaches. Methods Mol Biol. 2010, 649: 149-161. 10.1007/978-1-60761-753-2_9View ArticlePubMedGoogle Scholar
- Rivenbark AG, Stolzenburg S, Beltran AS, Yuan X, Rots MG, Strahl BD, Blancafort P: Epigenetic reprogramming of cancer cells via targeted DNA methylation. Epigenetics. 2012, 7: 350-360. 10.4161/epi.19507View ArticlePubMedPubMed CentralGoogle Scholar
- Nomura W, Barbas CF: In vivo site-specific DNA methylation with a designed sequence-enabled DNA methylase. J Am Chem Soc. 2007, 129: 8676-8677. 10.1021/ja0705588View ArticlePubMedGoogle Scholar
- Meister GE, Chandrasegaran S, Ostermeier M: An engineered split M.HhaI-zinc finger fusion lacks the intended methyltransferase specificity. Biochem Biophys Res Commun. 2008, 377: 226-230. 10.1016/j.bbrc.2008.09.099View ArticlePubMedPubMed CentralGoogle Scholar
- Kiss A, Weinhold E: Functional reassembly of split enzymes on-site: a novel approach for highly sequence-specific targeted DNA methylation. ChemBioChem. 2008, 9: 351-353. 10.1002/cbic.200700662View ArticlePubMedGoogle Scholar
- Meister GE, Chandrasegaran S, Ostermeier M: Heterodimeric DNA methyltransferases as a platform for creating designer zinc finger methyltransferases for targeted DNA methylation in cells. Nucleic Acids Res. 2010, 38: 1749-1759. 10.1093/nar/gkp1126View ArticlePubMedPubMed CentralGoogle Scholar
- Koudan EV, Bujnicki JM, Gromova ES: Homology modeling of the CG-specific DNA methyltransferase SssI and its complexes with DNA and AdoHcy. J Biomol Struct Dyn. 2004, 22: 339-345. 10.1080/07391102.2004.10507005View ArticlePubMedGoogle Scholar
- Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, Valentin F, Wallace IM, Wilm A, Lopez R, et al.: Clustal W and Clustal X version 2.0. Nucleic Acids Res. 2007, 23: 2947-2948.Google Scholar
- Guzman LM, Belin D, Carson MJ, Beckwith J: Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J Bacteriol. 1995, 177: 4121-4130.PubMedPubMed CentralGoogle Scholar
- Tímár E, Venetianer P, Kiss A: In vivo DNA protection by relaxed-specificity SinI DNA methyltransferase variants. J Bacteriol. 2008, 190: 8003-8008. 10.1128/JB.00754-08View ArticlePubMedPubMed CentralGoogle Scholar
- van der Gun BTF, Maluszynska-Hoffman M, Kiss A, Arendzen AJ, Ruiters MHJ, McLaughlin PMJ, Weinhold E, Rots MG: Targeted DNA methylation by a DNA methyltransferase coupled to a triple helix forming oligonucleotide to down-regulate the epithelial cell adhesion molecule. Bioconjugate Chem. 2010, 21: 1239-1245. 10.1021/bc1000388. 10.1021/bc1000388View ArticleGoogle Scholar
- Chen Y, Li S, Chen T, Hua H, Lin Z: Random dissection to select for protein split sites and its application in protein fragment complementation. Protein Sci. 2009, 18: 399-409. 10.1002/pro.42View ArticlePubMedPubMed CentralGoogle Scholar
- Paschon DE, Patel ZS, Ostermeier M: Enhanced catalytic efficiency of aminoglycoside phosphotransferase (3')-IIa achieved through protein fragmentation and reassembly. J Mol Biol. 2005, 353: 26-37. 10.1016/j.jmb.2005.08.026View ArticlePubMedGoogle Scholar
- Tan S, Guschin D, Davalos A, Lee YL, Snowden AW, Jouvenot Y, Zhang HS, Howes K, McNamara AR, Lai A, et al.: Zinc-finger protein-targeted gene regulation: genomewide single-gene specificity. Proc Natl Acad Sci U S A. 2003, 100: 11997-12002. 10.1073/pnas.2035056100View ArticlePubMedPubMed CentralGoogle Scholar
- Gommans WM, McLaughlin PM, Lindhout BI, Segal DJ, Wiegman DJ, Haisma HJ, van der Zaal BJ, Rots MG: Engineering zinc finger protein transcription factors to downregulate the epithelial glycoprotein-2 promoter as a novel anti-cancer treatment. Mol Carcinog. 2007, 46: 391-401. 10.1002/mc.20289View ArticlePubMedGoogle Scholar
- Durai S, Mani M, Kandavelou K, Wu J, Porteus MH, Chandrasegaran S: Zinc finger nucleases: custom-designed molecular scissors for genome engineering of plant and mammalian cells. Nucl Acids Res. 2005, 33: 5978-5990. 10.1093/nar/gki912View ArticlePubMedPubMed CentralGoogle Scholar
- Durfee T, Nelson R, Baldwin S, Plunkett G, Burland V, Mau B, Petrosino JF, Qin X, Muzny DM, Ayele M, et al.: The complete genome sequence of Escherichia coli DH10B: insights into the biology of a laboratory workhorse. J Bacteriol. 2008, 190: 2597-2606. 10.1128/JB.01695-07View ArticlePubMedPubMed CentralGoogle Scholar
- Sambrook J, Russell DW: The Condensed Protocols. From Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. 2006Google Scholar
- Kunkel TA, Roberts JD, Zakour RA: Rapid and efficient site-specific mutagenesis without phenotypic selection. Methods Enzymol. 1987, 154: 367-382.View ArticlePubMedGoogle Scholar
- Greene PH, Poonian MS, Nussbaum AL, Tobias L, Garfin DE, Boyer HW, Goodman HM: Restriction and modification of a self-complementary octanucleotide containing the EcoRI substrate. J Mol Biol. 1975, 99: 237-261. 10.1016/S0022-2836(75)80143-4View ArticlePubMedGoogle Scholar
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.