AOBPreview originally published online on March 2, 2007
Annals of Botany 2007 99(5):845-856; doi:10.1093/aob/mcm021
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Properties of a Tobacco DNA Methyltransferase, NtMET1 and Its Involvement in Chromatin Movement during Cell Division
Research and Education Center for Genetic Information, Nara Institute of Science and Technology, Nara 630-0192, Japan
* For correspondence. E-mail sano{at}gtc.naist.jp
Received: 6 November 2006 Returned for revision: 5 December 2006 Accepted: 9 January 2007 Published electronically: 2 March 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Background and Aims: Plants possess three types of DNA methyltransferase, among which methyltransferase type 1 (MET1) is considered to play a major role by maintaining the CpG methylation patterns. However, little information is available as to its enzymatic activity, interacting proteins and spatial and temporal behaviours during DNA replication. In the present study, one example, NtMET1 from tobacco plants, was selected and an analysis was made of its biochemical properties and cellular localization.
Methods: NtMET1 was expressed in Sf9 insect cells, and a purified sample was subjected to a standard in vitro methylation assay. Intramolecular interaction was examined by the yeast two-hybrid and pull-down assays. Transgenic tobacco plants (Nicotiana tabacum) over-expressing NtMET1 were constructed via Agrobacterium-mediated transformation. Cellular localization was examined by fluorescence protein fusion, which was expressed in tobacco bright yellow 2 cells.
Key Results In vitro: assays showed no detectable methylation activity when both hemimethylated and unmethylated DNA samples were used as the substrate. In planta assays with over-expressing transgenic lines showed no hypermethylation but rather hypomethylation of genomc DNA. The inability of methylation was conceivably due to a tight intramolecular interaction between the N- and C-terminal regions with the catalytic domain residing on the C-terminus being completely masked. Cellular localization analyses indicated that NtMET1 localized to the nucleus in the resting stage and migrates to the cytoplasm during mitosis, particularly at metaphase. The pattern observed resembled that of Ran GTPase, and in vitro pull-down assays showed a clear interaction between NtMET1 and AtRAN3, an Arabidopsis orthologue of tobacco Ran GTPase, NtRan-A1.
Conclusions: The results suggest that enzymatic activity of NtMET1 is well adjusted by its own intra/intermolecular interaction and perhaps by interactions with other proteins, one of which was found to be Ran GTPase. Results also revealed that NtMET1 becomes localized to the vicinity of chromatin with the aid of Ran GTPase during cell division, and may play an important role in progress through mitosis independently of methylation activity.
Key words: DNA methyltransferase, intra/inter molecular interaction, 5-methylcytosine, mitosis, Ran GTPase, Nicotiana tabacum
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Methylation of cytosine residues is commonly observed in the DNA of most eukaryotes, this often being called DNA methylation, and has been considered to play certain roles in controlling gene expression (Yoder et al., 1997). Cytosine methylation is enzymatically catalysed by DNA methyltransferases, which transfer a methyl group from S-adenosylmethionine to the 5-position of cytosines in DNA. In plants, three types of DNA methyltransferases have been identified based on sequence analyses: DNA methyltransferase type I (MET1), chromomethylase and domains rearranged methylase (DRM), among which MET1 is believed to predominantly catalyse methylation of hemimethylated symmetrical CpG, thereby maintaining methylation patterns after DNA replication (Finnegan and Dennis, 1993). In addition, the green alga, Chlamydomonas reinhardtii was shown to possess a DNA methyltransferase, whose structure resembles the MET1, but exhibits catalytic activities similar to DRM (Nishiyama et al., 2004).
MET1 consists of the N-terminal regulatory domain and the C-terminal catalytic domain (Fig. 1A). The N-terminal domain contains several putative signal motifs, including nuclear localization, bromo-adjacent homology domains, zinc fingers and serine-rich regions (Fig. 1A). Similar signal motifs are also found in mammalian DNA methyltransferase 1 (Dnmt1), the orthologue of plant MET1 proteins (Fig. 1A). However, despite the similar molecular size, few common motifs with plant MET1 were found in the N-terminal of the Chlamydomonas enzyme (Fig. 1A), suggesting a complicated function of the N-terminal regions. In contrast, the C-terminal domain contains highly conserved protein motifs, which are common among DNA methyltransferases of different types from different organisms (Fig. 1B).
|
Curiously, however, biochemical studies on catalytic properties of MET1 enzymes have been hitherto limited. As far as is known, only two preliminary reports are available; one is a study on maize MET1 (ZmMET1, AF229183 [GenBank] ), of which N-terminal-truncated protein was reported to exhibit a low methylation activity in vitro (Steward et al., 2000). The other case is pea MET (C-5 MTase, AF034419 [GenBank] ), which showed methylation activity towards Cp(A/T)pG and poly(dI-dC) synthetic oligomers (Pradhan et al., 1998). However, lack of enough knowledge on catalytic properties has made it difficult to fully understand the biological functions of MET1 proteins. Indeed, most functional studies, including Arabidopsis enzymes, have been carried out on genetic bases without substantial activity assays (Finnegan and Kovac, 2000; Goll and Bestor, 2005).
In addition to methylation activity, DNA methyltransferase proteins have been proposed to directly participate in regulatory function (Rountree et al., 2000). The N-terminus of mammalian Dnmt1 was shown to possess transcriptional repressing activity by directly interacting with histone deacetylase (Rountree et al., 2000), histone methyltransferase (Fuks et al., 2000) and retinoblastoma tumour suppressor proteins (Robertson et al., 2000). These observations suggest that Dnmt1 plays an important role in the gene regulation network independent of DNA methylation (Milutinovic et al., 2004). In accordance with this idea, intracellular localization analyses revealed that mouse Dnmt1 moved between nuclei and cytoplasm during embryo development (Mertineit et al., 1998). It was speculated that Dnmt1 family proteins might be involved in sister chromatid segregation during the mitotic phase (Hung et al., 1999). As to plant MET1 proteins, no such information has so far been available.
In order to understand the general features of MET1 function, a tobacco example, NtMET1, was selected and was found apparently not to exhibit DNA methylation actitivty in vitro, conceivably due to tight intra/intermolecular interactions. It is also reported here that NtMET1 interacts with Ran GTPase, thereby reversibly changing cellular localization during mitosis in tobacco cells.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Plant materials
Wild-type and transgenic tobacco plants (Nicotiana tabacum ‘Xanthi nc’) were grown in a growth cabinet at 23 °C under a 14/10 h light/dark cycle. Tobacco-cultured bright yellow 2 (BY2) cells, from both wild-type and transgenic lines, were grown in Murashige and Skoog medium on a rotary shaker (115 rpm, 25 °C) in the dark.
Isolation of NtMET1 cDNA
Since a full-length cDNA of previously reported NtMET1 (accession no. AB030726
[GenBank]
) was partly deduced from genomic sequence (Nakano et al., 2000), the entire coding region spanning over 4·5 kb was amplified by PCR using tobacco (N. tabacum) cDNA and a set of primers containing attB sites and specific nucleotide sequences of NtMET1 (forward, 5'-AAAAAGCAGGCTTAATGGGTTCCCTGGCGGGGTTG-3'; reverse, 5'-AGAAAGCTGGGTCCTAAGTGGACCTCTTCTTGCT-3'; NtMET1-specific sequences are in underlined) (registered under accession no. AB280788
[GenBank]
). PCR products were subjected to a second PCR with another set of primers (forward, 5'-GGGGACAAGTTTCTACAAAAAAGCAGGC-3'; reverse, 5'-GGGGACCACTTTGTACAAGAAAGCTGGGT-3'), and the resulting 4·5-kb products were cloned into the pDONR201 (Invitrogen, Carlsbad, CA, USA) vector plasmid, and used for further experiments including protein expression, GFP-fusion and transformation.
Fusion proteins
Full-length NtMET1 was fused to the gene for glutathione S-transferase (GST) at the N-terminus and cloned into the pDEST20 vector plasmid, which was transformed into Escherichia coli DH10BacTM cells (Invitrogen) as described (Wada et al., 2003). Insect Sf9 cells were maintained in Grace's insect medium (Invitrogen) supplemented with 10 % foetal bovine serum (Invitrogen) and 500 µg mL–1 gentamycin. Approximately 6 x 109 cells per dish were infected with the recombinant baculovirus stock (500 µL) at suitable titre using Cellfectin Reagent (Invitrogen) and incubated at 27 °C for 3 d. Infected cells were suspended in lysis buffer (20 mM Tris–HCl, pH 7·5, 5 mM EDTA, 1 % Nonidet P-40, 1 mM dithiothreitol, 1 mM phenylmethylsulfonylfluoride and 100 µg mL–1 aprotinin) and sonicated for 10 s twice. The resulting solution was used as the crude enzyme solution, or subjected to purification through a glutathione–Sepharose column (Amersham Biosciences, Piscataway, NJ, USA) as described by Wada et al. (2003). GST-fused NtH4 was constructed using pGEX4T-2 (Pharmacia Biotechnology, Piscataway, NJ, USA) according to the method described by Seo et al. (1995). His-tagged AtRAN3 was prepared as described by Yano et al. (2006). Protein concentrations were estimated using the Bradford assay with bovine serum albumin as standard.
DNA methyltransferase assay
DNA methyltransferase activity was assayed by measuring 3H-labelled methyl group transfer from S-adenosylmethionine (AdoMet) to substrates as described by Wada et al. (2003) with modification. For the initial assay, 2 µg of poly(dI-dC)/poly(dI-dC) (Sigma, St Louis, MI, USA) was used as the substrate. For sequence specificity assays, the synthetic oligonucleotides 5'-ACGATCGTACGATCGTACGATCGT-3' (for CpG), 5'-ACTGCAGTACTGCAGTACTGCAGT-3' (for CpNpG, where N is A or T) and 5'-AGCATGCTAGCATGCTAGCATGCT-3' (for CpNpN) were prepared. All these sequences are palidromic, which form duplexes. For hemi-methylation analysis, synthetic 28-mer oligonucleotides containing five 5-methylcytosines (m5C) and their complementary strands without m5C were independently prepared, and annealed to form duplexes. The substrates for CpG were 5'-ATTCGATCGAATCGTATACGTACGTATT-3' and 3'-TAAGCTAGCTTAGCATATGCATGCATAA-5' (m5C is underlined), and those for CpNpG were 5'-ATTCAGTCAGATCTGATCAGTACTGATT-3' and 3'-TAAGTCAGTCTAGACTAGTCATGACTAA-5'. The reaction mixture contained 20 mM MOPS–NaOH (pH 7·0), 5 mM EDTA, 200 µg mL–1 BSA, 25 % (v/v) glycerol, 1 mM dithiothreitol, 100 µg mL–1 RNase A, 2 µmole of AdoMet (methyl-3H, specific activity 307·1 GBq mmol–1) (PerkinElmer, Wellesley, MA, USA), appropriate amounts of substrate DNA and purified protein fractions. After incubation for 90 min at 37 °C, the reactions were stopped by adding 500 µL of proteinase K–SDS buffer (1 mg mL–1 proteinase K, 1 % SDS, 2 mM EDTA, 125 mM NaCl and 0·5 mg mL–1 salmon sperm DNA) with further incubation at 50 °C for 1 h. The substrate DNA was extracted with phenol–chloroform, recovered by ethanol precipitation and spotted on DEAE paper, dried and washed with 0·5 M sodium phosphate followed by 70 % ethanol. Papers were placed in scintillation mixture, and DNA methylation activity was determined by scintillation counting of radioactivity. The amounts of the methyl group transferred were calculated based on the specific activity of [3H]AdoMet (7·7 x 104 cpm pmol–1).
Transformation
Transformation of tobacco leaf discs was performed as described previously by Nakano et al. (2000). Cultured BY2 cells were transformed via the Agrobacterium-mediated method as described by Yamaguchi et al. (2003).
High performance liquid chromatography (HPLC)
The amount of 5-methylcytosine (m5C) in total genomic DNA was measured by HPLC according to the protocols described previously by Wada et al. (2003). A 15-µg aliquot of DNA was digested with 2 units of nuclease P1 (Sigma) in 100 µL of buffer containing 3 mM sodium acetate (pH 5·4) and 0·5 mM ZnSO4 at 37 °C for 16 h. Nucleotides were dephosphorylated with 20 units of calf intestine alkaline phosphatase (Takara, Kyoto, Japan) and filtered through a membrane with a pore size of 0·2 µm. Samples (10 µL) were injected into a Supelcosil LC-18-S column (Supelco, Bellafonte, PA, USA) and separated with a 2·5–20 % methanol gradient in the presence of 50 mM KH2PO4 (pH 4·3).
Fluorescence microscopy
For green fluorescence (GFP)-fusion, the attB-PCR product of NtMET1 was cloned into the pGWB6 (CaMV35S–sGFP–NOS3') vector to fuse to GFP using the GATEWAY cloning system (Invitrogen). A cDNA for tobacco histone H4 (NtH4) (accession no. AB280787
[GenBank]
) was obtained by PCR using specific primers containing SalI and NcoI sites. Cyan fluorescence protein (CFP)-fused histone H4 (NtH4) was constructed by inserting the cDNA into the CaMV35S–CFP–NOS3' vector as previously described by Ito et al. (2003). Yellow fluorescence protein (YFP)–AtRan3 and CFP-RCC1 vectors were prepared as described by Yano et al. (2006). All constructs were transformed into A. tumefeciens strain EHA105 and the resulting strains were used to transform tobacco BY2 cells (Wada et al., 2003). Transgenic BY2 cells were observed with an AX70 fluorescence microscope (Olympus, Tokyo, Japan), with the U-MGFPHQ cube for GFP, YFP and CFP individually, and captured with a cooled charge-coupled device camera (CoolSNAPHQ; Photometrics, Tucson, AZ, USA). DNA staining was performed with an aliquot of LS medium containing 1 mg mL–1 4'6-diamidino-2-phenylindole (DAPI) as described by Yano et al. (2006).
DNA blot and RT-PCR analyses
Genomic DNA was extracted from fresh leaves by the cetyl-trimethyl ammonium bromide method (Murray and Thompson, 1980). A 20-µg aliquot was digested with either HpaII or MspI, and fractionated by 1 % agarose gel electrophoresis, transferred onto nylon membrane (Hybond N + ; Amersham Pharmacia Biotech, Piscataway, NJ, USA), and subjected to hybridization with a 32P-labelled cDNA fragment of tobacco retrotransposon 1 (Tto1; accession no. D83003
[GenBank]
) (Hirochika, 1993) or BamHI tandem repeat element (HRS60·1; accession no. X12489
[GenBank]
). For RT-PCR, total RNA was isolated from young leaf tissues by the AGPC method (Verwoerd et al., 1989). A 1-µg aliquot of total RNA was used for reverse transcription using an RNA PCR kit (Takara). The resulting cDNA was amplified with sets of specific primers under the following conditions: 28 cycles at 94 °C for 30 s, 57 °C for 30 s, 72 °C for 1 min with a final extension at 72 °C for 7 min (GeneAmp 2400, PerkinElmer). The products were fractionated on 1·5 % agarose gel.
Immuno-blot analysis
Total protein extract (20 µg) was prepared from 0·5 g young leaf tissues, and suspended in 1 mL cold lysis buffer containing 50 mM Tris–HCl, pH 7·5, 0·1 % SDS, 10 % glycerol, 100 mM DTT and 1 mM PMSF. Proteins were fractionated by 12 % SDS–PAGE, and afterwards blotted to polyvinylidene difluoride membranes (Millipore, Billerica, MA, USA), NtMET1 protein was detected by immunoblotting using rabbit anti-ZmMET1 antibodies raised against specific peptide in catalytic regions (Steward et al., 2000) and peroxidase-conjugated goat anti-rabbit secondary antibodies (BioRad, Hercules, CA, USA).
Construction of Ran GTPase mutants
A specific point mutant was introduced into AtRAN3 cDNA by PCR site-directed mutagenesis. The active form was constructed by substituting Gly at position 22 with valine (G22V), and the inactive form with Thr at position 27 with Asn (T27N) (Dascher and Balch, 1994; Haizel et al., 1997). Briefly, the 5'-oligonucleotide primer, 5'-CACCATGGCTCTACCTAACCAGCAAAC-3' was used to facilitate directional cloning into the pBAD102/D-TOPO expression vectors (Invitrogen). The anti-sense mutagenic oligonucleotides, 5'-GGTCTTCCCTGTGCCTACATC-3' for G22V (mutation point is underlined) and 5'-ACA AATGTGTTCTTCCCTGTG-3' for T27N, were prepared to serve as primers for amplification of the 5'-fragments. The 3'-fragments were generated using oligonucleotides complementary to the above mutagenic primers in combination with the 3'-end anti-sense oligonucleotide primer 5'-CTCGAAGGTGTCATCATCGTCAT-3'. After the second overlap extension, the recombinant PCR products were cloned into the pBAD102/D-TOPO vector and expressed in E. coli strain LMG194 according to the manufacturer's instructions (Invitrogen).
Pull-down binding assay
The pull-down assay was performed as described by Yap et al. (2005). Briefly, approx. 5 µg of purified GST, GST-fused NtMET1 or GST-fused NtH4 proteins were immobilized on glutathione (GSH)-Sepharose (Amersham Biosciences) at 4 °C for 2 h. After blocking, beads were incubated with 25 µg of His-tagged fusion proteins at 4 °C for 16 h. Beads were collected and washed, and proteins were eluted with a buffer containing 10 mM reduced glutathione in 50 mM Tris–HCl at pH 8. Eluted proteins were fractionated on 12 % SDS–polyacrylamide gels and subjected to immuno-blot staining using mouse anti-His-tag monoclonal antibodies and horse radish peroxidase-conjugated anti-mouse IgG antibodies (MBL, Nagoya, Japan). The antibody–antigen complex was detected using the ECL system (Amersham Biosciences). For far-western assays, a 20-µL aliquot of antibodies raised against the C-terminal region-specific peptide of ZmMET1 (Steward et al., 2000) was bound to protein A–Sepharose (Pharmacia Biotechnology), and washed three times with 500 µL of 50 mM Tris–Cl buffer (pH 7·0). A 25-µg 
NtMET1/N protein was pre-incubated with 25-µg full-length NtMET1 or NtMET1/C at 4 °C for 2 h on a rotator. The antibody–protein A complex was then mixed with appropriate amounts of proteins, and incubated at 4 °C for 16 h on a rotator. After being washed with the buffer, the samples were fractionated on SDS–PAGE, transferred to a membrane and subjected to immuno-staining by anti-GST antibodies.
Yeast two-hybrid assay
Yeast two-hybrid assays were performed as described previously by Yap et al. (2005). The GAL4 DNA-binding domain (BD) was fused to full-length NtMET1 (NtMET1) or C-terminal-truncated NtMET1 (NtMET1/N), and the activation domain (AD) was fused to N-terminal-truncated NtMET1 (NtMET1/C) using the MATCHMAKER GAL4 system (Clonetech, Palo Alto, CA, USA) Yeast AH109 cells were co-transformed with the combinations indicated. The co-transformants for interacting constructs were screened on Quadruple Dropout Medium (SD/-Ade/-His/-Leu/-Trip) and subjected to the ß-galactosidase colony-lift filter and activity assays as described in the yeast protocols handbook (Clonetech).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Methylation activity
Full-length or N-terminus-truncated NtMET1 as glutathione S-transferase (GST) fusion protein was expressed in a baculovirus-mediated expression system in Spodoptera frugiperda (Sf9) insect cell line, which lacks endogenous DNA methyltransferase activity. After 72 h post-infection, cell extracts were sampled, and protein was purified through a glutathione–Sepharose column, and immunoblotted with anti-GST antibodies (Fig. 2A). It is evident that full-length (200 kDa, NtMET1) and N-terminus-truncated (117 kDa,
NtMET1/C) proteins were successfully synthesized (Fig. 2A). DNA methyltransferase activity was then estimated using purified protein samples. As positive and negative controls, tobacco DRM1, a de novo DNA methyltransferase (Wada et al., 2003), and bacterial ß-glucuronidase (GUS), respectively, were similarly expressed in insect cells and subjected to the same assay. Initially, de novo DNA methyltransferase activity was estimated by measuring 3H-labelled methyl group transfer from AdoMet into synthetic poly(dI-dC), which provides a large number of potential dinucleotide sites for methylation. Results indicated that, while GST–NtDRM1 was clearly active, both GST–NtMET1 and GST-NtMET1/C were totally inactive, showing the same level as the control GUS protein (Fig. 2B). The maintenance methyl transfer activity was then analysed. Substrates were constructed using two types of synthetic 28-mer oligonucleotide substrates containing either CpG or CpNpG (N is A or T). In one sample, all cytosines were substituted with m5C and, in the other, all cytosines remained intact. By mixing and annealing these two types, hemimethylated substrate was created. Unmethylated substrate was made from unmethylated oligomers. Results showed that GST–NtMET1 was inactive towards both unmethylated and hemimethylated substrates in the CpG and CpNpG sequence context (Fig. 2C). Time-course analyses also exhibited no detectable activity up to 2 h of incubation (data not shown). That the reaction system was appropriate was confirmed by clear activity of GST–NtDRM1, which preferentially methylates both unmethylated and hemimethylated CpNpG and CpNpN (Wada et al., 2003) (Fig. 2C).
|
Transgenic tobacco
The inability of NtMET1 to methylate DNA in vitro was puzzling, since its critical domains necessary for activity were highly conserved among DNA methyltransferases, of which enzymatic activities have been confirmed in vitro (Fig. 1B). Expecting in vivo activity, transgenic tobacco plants were constructed, in which NtMET1 was overexpressed under the control of cauliflower mosaic virus (CaMV) 35S promoter. More than ten lines were initially selected and grown on soil. At maturity, their phenotype was apparently normal, except that the growth rate was slow, resulting in short plants compared with the control wild-type plants (representative two lines are shown in Fig. 3A). Four lines were finally selected and high NtMET1 expression was confirmed by RT-PCR (Fig. 3B) and NtMET1 proteins produced by immuno-blot staining (Fig. 3C). Total DNA was isolated, hydrolysed to mononucleotides and m5C contents were estimated by HPLC. In wild-type plants, up to 12 % of cytosines were m5C, while in transgenic lines, m5Cs were only 8 % (line #1), 7·3 % (line #2) and 6·3 % (line #4) of total cytosines (Fig. 3D). Thus global methylation in these transgenic lines was clearly reduced up to nearly 50%. The methylation status at a particular loci was then analysed by DNA blot hybridization using a pair of methylation-sensitive restriction endonucleases (Fig. 3E and F). When probed with cDNA of Tto1, a multicopy retrotransposon of tobacco, genomic DNA from wild-type plants showed a clear resistance to cleavage by HpaII, which is sensitive to Cm5CGG, whereas it was susceptible to MspI, which is insensitive to Cm5CGG (Fig. 3E). In contrast, genomic DNA from all transgenic lines was efficiently cleaved by HpaII, showing similar patterns to those by MspI (Fig. 3E). This indicated that the second cytosines in the CCGG sequence in Tto1 loci were highly methylated in the wild-type control but demethylated in transgenic lines. Similar results were obtained with a tandem repeat element, HRS60·1, which comprises 2·3 % of the total tobacco genome (Kovarik et al., 1994). In wild-type plants, this sequence was sensitive to cleavage by MspI but highly resistant to HpaII, indicating a high frequency of the second cytosine methylation in the CCGG sequence (Fig. 3F). In transgenic tobacco lines #1, #2 and #4, the locus was apparently demethylated, showing a cleavage pattern by HpaII similar to that by MspI (Fig. 3F). These results indicated that, in spite of efficient expression of NtMET1 protein in planta, genomic DNA was not over-methylated but rather demethylated, perhaps due to excess NtMET1 protein interfering with the normal methylation process.
|
Intramolecular interaction
A lack of enzymatic activity both in vitro and in planta suggested that NtMET1 activity is highly negatively controlled at protein level. One such controlling factor could be higher conformation of protein itself, since the N-terminal region of mammalian Dnmt1 was reported to efficiently bind to the C-terminus (Margot et al., 2003). Subsequently, the possibility of intramolecular interaction within the NtMET1 molecule was examined. Using the N-terminus with 895 amino acids (
NtMET1/N) and the C-terminus with 798 amino acids (NtMET1/C) (Fig. 4A), the in vitro far-western assay was performed by specific antibodies that recognized the C-terminal but not the N-terminal polypeptides. Results showed that the N-terminus (NtMET1/N) was efficiently precipitated only in the presence of the C-terminus (NtMET1/C) (Fig. 4B). This interaction was abolished when the full-length NtMET1 was used instead of the C-terminus (Fig. 4B). The inability of co-precipitation might be due to the C-terminus being already masked by the intact N-terminus, being thus prevented from further interaction with exogenous N-terminal fragments. Yeast two-hybrid assays confirmed these results, showing an effective interaction between NtMET1/N and NtMET1/C, but no interaction between full-length NtMET1 and NtMET1/C (Fig. 4C). These experiments proved a specific intra-molecular interaction between the N- and C-terminal regions in NTMET1 protein.
|
Intracellular localization
The cellular localization of NtMET1 protein was examined with GFP-fused constructs. Plasmid containing the cauliflower mosaic virus (CaMV) 35S promoter NtMET1-GFP was constructed and introduced into tobacco-cultured BY2 cells by the Agrobacterium-mediated transformation method. When cells during the resting stage were examined, NtMET1–GFP constructs clearly gave fluorescent signals in nuclei (Fig. 5, middle panel), while the control GFP fluorescence was observed throughout the cell (Fig. 5, upper panel). Upon merging with DAPI-stained images, NtMET1 protein was apparently localized to regions where DNA is condensed. This was distinct when an image of the nucleus was enlarged, showing well-matching DAPI and GFP images (Fig. 5, lower panel). Localization of NtMET1 was then examined in cells undergoing mitosis. At inter- and prophases, NtMET1–GFP was only localized in the nucleus (Fig. 6A, top and second panels). At early metaphase, however, NtMET1–GFP was mostly found in cytoplasm (Fig. 6A, third panel), and in both cytoplasm and chromatins at middle and late metaphases (Fig. 6A, fourth panel). At anaphase, the protein began to align on chromatin (Fig. 6A, fifth panel) and, at telophase, NtMET1 was again localized only in the nucleus (Fig. 6A, bottom panel). These localization patterns were inherent in NtMET1 protein, since during all phases of mitosis, histone H4 protein distinctly remained with the chromatin (Fig. 6C), and the control GFP protein was ubiquitously observed throughout the cell (Fig. 6D).
|
|
Localization patterns of NtMET1 during mitosis appeared to resemble that of Ran GTPase, AtRAN3, which is an orthologue of tobacco NtRan-A1 showing a 96 % homology (Yano et al., 2006). Subsequently, localization of the YFP–AtRAN3 fusion protein was examined in parallel with NtMET1. At inter- and prophases, AtRAN3 was mainly localized to the nucleus with weak dispersion into the cytoplasm (Fig. 6B, top and second panels). At metaphase, AtRAN3 was found at the periphery of the nucleus, particularly at the position of the spindle microtubules (Fig. 6B, third panel). At anaphase, AtRAN3 was localized in and between chromosomes (Fig. 6B, fourth panel), and at telophase, it was localized not only into the nucleus, but also into the region of the growing cell plate (Fig. 6B, bottom panel). When comparing the localization patterns of NtMET1 and AtRAN3, it is evident that they behave concertedly at pro- and metaphases, but differently at ana- and telophases. These results indicated that NtMET1 changes its localization during mitosis, and that such changes partly coincide with that of AtRAN3. Since Ran GTPases are considered to play critical roles in spindle formation (Kahana and Cleveland, 1999), co-localization of NtMET1 with AtRAN3 proteins may suggest that both proteins are positively involved in chromatin movement during cell division.
Physical interaction
Physical interaction between NtMET1 and AtRAN3 was assessed by an in vitro pull-down assay, using GST-tagged NtMET1 (GST–NtMET1) and His-tagged AtRAN3 (His-AtRAN3) proteins. His-AtRAN3 was applied to a glutathione–Sepharose column containing GST–NtMET1 or GST proteins, eluted with a buffer containing reduced GST, separated by SDS–PAGE and subjected to immuno-blot assays with anti-His-tag antibodies. The results showed that AtRAN3 was only detectable upon incubation with GST–NtMET1 but not with GST protein alone, indicating specific binding of AtRAN3 to NtMET1 (Fig. 7A). Ran GTPase often changes its location and targeting of interacting proteins depending on whether it is in the GTP-bound active or GDP-bound inactive forms (Zheng, 2004). In order to determine which preferentially binds to GST–MET1, pull-down assays were performed with active and inactive AtRAN3 mutants, which were constructed by single amino acid substitution. Binding tests showed that GST–MET1 interacted equally with both forms of AtRAN3 (Fig. 7A).
|
Ran GTPase was previously shown to bind directly to histone H4 in the Xenopus cell (Bilbao-Cortes et al., 2002). To test this possibility in the plant cell, tobacco histone H4 fused to GST was bacterially expressed and subjected to a pull-down assay with His-AtRAN3 in a similar manner to GST–NtMET1 assays. Results showed that interaction indeed takes place between AtRAN3 and NtH4 (Fig. 7B), and that this interaction was equally effective with mutant proteins (Fig. 7B). The possibility of NtMET1 interacting with NtH4 was finally examined, but the results were apparently negative, showing no direct binding on the pull-down assay (Fig. 7C).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Despite using intensively purified recombinant protein expressed in insect cells, a tobacco type 1 DNA methyltransferase, NtMET1 did not exhibit any detectable enzymatic activity in vitro towards either unmethylated or hemi-methylated DNA substrates. The inability was not due to experimental procedures such as amino acid mutation, since the total cDNA sequence was confirmed to be correct after the genomic sequence (Kim et al., unpubl. res.). Also all amino acid motifs necessary for activity were well conserved (see Fig. 1A). Transgenic tobacco producing excess NtMET1 protein exhibited hypomethylation of genomic DNA, indicating enzymatic inability in planta. Several causes are conceivable: (1) NtMET1 is inherently deficient in enzymatic activity; (2) it needs modification to be activated, including glucosylation, phosphorylation and/or acetylation/methylation; (3) it needs counterpart factors such as protein(s) for activation; and (4) the activity is blocked by protein structure.
The first argument that MET1 is defective in enzyme activity may not be exclusive. However, such a probability appears to be low, since the catalytic domain of NtMET1 highly resembles mammalian Dnmt1 and Chlamydomonas CrMET1, both of which have been confirmed to possess high activity (Okano et al., 1999; Nishiyama et al., 2004). Genetic analyses also point to a clear involvement of MET1 in genomic CpG methylation (Ronemus et al., 1996), although the possibility remains that the effects are indirect. The second idea that post-translational modification is prerequisite for conferring activity is also possible. In silico analysis indicated that NtMET1 can potentially be phosphorylated at 84 Ser/Thr sites throughout the molecule. In mouse DNA methyltransferase, phosphorylation of serine at position 514 was shown to be critical for bringing the enzyme to the replication foci (Glickman et al., 1997). It was suggested that phosphorylation may affect subcelluar localization and attenuate substrate inhibition due to the allosteric effect of the enzyme (Glickman et al., 1997). A similar modification could occur in NtMET1 thereby modulating the activity. The third point suggesting that NtMET1 requires a protein factor for in vivo activity is highly probable. To date, mammalian Dnmt1 has been shown to form stable complexes with histone deacetylase (HDAC) 1 and 2, tumour suppressor gene product (Rb), histone methyltransferase (SUV39H1) and DNA methyltransferase-associated protein (DMAP1) (Robertson et al., 2000; Rountree et al., 2000; Fuks et al., 2003). These proteins were supposed to modulate Dnmt1 function by, for example, mutually affecting methylation activity of DNA and histones (Fuks et al., 2003), or recruiting it to a transcription complex (Rountree et al., 2000). For plant MET1, however, no protein factors have so far been identified that directly regulate its activity. The fourth idea suggesting that intramolecular interaction might interfere with the activity was partly supported by the present finding that shows a tight interaction between the N- and C-terminal regions. A similar but opposite case was reported for mouse Dnmt1, in which full activity required N-terminus interaction with the C-terminus (Margot et al., 2003). It was concluded that a physical interaction between the N- and C-terminal domains was prerequisite for activation of the catalytic domain. Hence intramolecular interaction might be critical for regulation of Dnmt1 and NtMET1 activities. However, structural relaxation alone appears not to be sufficient for activation of NtMET1, as judged from the present observation. For example, the N-terminus-truncated construct, which does not interact with the C-terminus, did not exhibit detectable methylation activity (Fig. 1B), and treatments of purified enzyme with mild detergents such as Tween-20 and Nonidet P-40 did not recover the activity (Kim et al., unpubl. res.). It can thus be safely concluded that NtMET1 activity is finely regulated by complex formation with various proteins, which may be important to relax and/or tighten its higher conformation, and to assist in activity expression.
A question then arises as to how and which factors activate the enzyme in vivo. The best clue might be obtained through identification of interacting protein(s) by the yeast two-hybrid method. Subsequent intensive screening of a cDNA library containing 1·2 x 109 clones by the NtMET1 bait yielded encoding clones such as ubiquitin carrier and DnaJ heat-shock proteins, together with several unknown proteins, but it was decided not to characterize them further (Kim et al., unpubl. res.). Another clue could be obtained through NtMET1 behaviour during cell division, since this enzyme is supposed to function during DNA replication to maintain the CpG methylation (Goll and Bestor, 2005). Mammalian Dnmt1 was shown to be localized in the cytoplasm during interphase, and translocated to the nucleus at S-phase to associate with replication foci (Leonhardt et al., 1992; Goll and Bestor, 2005), so that the enzyme specifically acts on newly replicated DNA strands. In contrast, NtMET1 appeared to stay within the nucleus throughout the cell cycle, except during M-phase, when it diffused into the cytoplasm. This apparently opposite localization between mammalian and plant enzymes may be correlated with their activity. Mammalian Dnmt1 was shown to be constitutively active when the N-terminus interacts with the C-terminus catalytic domain (Margot et al., 2003). NtMET1 could be constitutively inactive when the N-terminus interacts with the C-terminus. Thus, in mammalian cells, methylation can be performed simply by bringing the active enzyme to the replication foci while, in plants, reactivation of inactive enzyme is necessary. This can be accomplished by other protein factor(s), and such an example will be identified in a future study.
Translocation of NtMET1 into the cytoplasm during metaphase is notable. Since no DNA replication takes place during M-phase, such a translocation is probably not involved in maintenance of DNA methylation. In the case of Drosophila Dnmt1-like protein, DmMTR1, a cell cycle-specific switch of its localization in cytoplasm and nucleus was noted (Hung et al., 1999). During interphase, DmMTR1 was located outside the nucleus, and rapidly translocated into the nucleus during mitosis. Since genomic DNA of Drosophila is not methylated, this protein was suggested to play an essential function in the cell cycle-regulated condensation of chromosomes (Hung et al., 1999). Although the translocation phase differs between DmMTR1 and NtMET1, these observations suggested that, independently of DNA methylation, both proteins may participate in the progress of mitosis together with many other proteins. One such protein was found to be a Ran GTPase for NtMET1. Ran GTPase was shown to promote microtubule nucleation during mitotic spindle assembly and nuclear envelop assembly (Clarke and Zhang, 2001). It takes two alternative GTP- and GDP-bound forms, among which the former has been considered to be biologically active due to its localization on the chromatin surface (Kahana and Cleveland, 1999). Recent studies, however, revealed that Ran directly binds to nucleosomes and to histones H3 and H4 independently of GTP/GDP forms (Bilbao-Cortes et al., 2002). It was found here that both forms of Ran GTPase equally bind to NtMET1, and also to histone H4, consistent with the observation above. This suggests that NtMET1 is one of the chromatin proteins, attaching to H4 via Ran GTPase, since NtMET1 did not directly bind to H4. Previously it had been shown that the active form of Ran GTPase efficiently interacts with a methyl CpG-binding protein (MBD), which migrates around chromatin during mitosis (Yano et al., 2006). Since the behaviour of NtMET1 resembles this pattern, it is tempting to speculate that one of the protein complexes surrounding the chromatin structure specifies the recognition of, and interaction with, methylated DNA and that the dynamic movement of such a complex during mitosis is driven by Ran GTPase.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
This work was partly supported by a grant for the Research for the Future Program from the Japan Society for the Promotion of Science. H.-J.K. is a recipient of a National Scholarship for Foreign Students from the Ministry of Education, Culture, Sports, Science and Technology, Japan.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
-
Bilbao-Cortes D, Hetzer M, Langst G, Becker PB, Mattaj IW. Ran binds to chromatin by two distinct mechanisms. Current Biology (2002) 12:1151–1156.[CrossRef][Web of Science][Medline]
Clarke PR, Zhang C. Ran GTPase: a master regulator of nuclear structure and function during the eukaryoitic cell division cycle? Trends in Cell Biology (2001) 11:366–371.[CrossRef][Web of Science][Medline]
Dascher C, Balch WE. Dominant inhibitory mutants of ARF1 block endoplasmic reticulum to Golgi transport and trigger disassembly of the Golgi apparatus. Journal of Biological Chemistry (1994) 269:1437–1448.
Finnegan EJ, Dennis ES. Isolation and identification by sequence homology of a putative cytosine methyltransferase from Arabidopsis thaliana. Nucleic Acids Research (1993) 21:2383–2388.
Finnegan EJ, Kovac KA. Plant DNA methytransferase. Plant Molecular Biology (2000) 43:189–201.[CrossRef][Web of Science][Medline]
Fuks F, Burgers WA, Brehm A, Hughes-Davies L, Kouzarides T. DNA methyltransferase Dnmt1 associates with histone deacetylase activity. Nature Genetics (2000) 24:88–91.[CrossRef][Web of Science][Medline]
Glickman JF, Pavlovich JG, Reich NO. Peptide mapping of the murine DNA methyltransferase reveals a major phosphorylation site and the start of translation. Journal of Biological Chemistry (1997) 272:17851–17857.
Goll MG, Bestor TH. Eukaryotic cytosine methyltransferases. Annual Review of Biochemistry (2005) 74:481–514.[CrossRef][Web of Science][Medline]
Haizel T, Merkle T, Pay A, Fejes E, Nagy F. Characterization of proteins that interact with the GTP-bound form of the regulatory GTPase Ran in Arabidopsis. The Plant Journal (1997) 11:93–103.[CrossRef][Web of Science][Medline]
Hirochika H. Activation of tobacco retrotransposons during tissue culture. EMBO Journal (1993) 12:2521–2528.[Web of Science][Medline]
Hung MS, Karthikeyan N, Huang B, Koo HC, Kiger J, Shen CKJ. Drosophila proteins related to vertebrate DNA (5-cytosine) methyltransferases. Proceedings of the National Academy of Science of the USA (1999) 96:11940–11945.[CrossRef]
Ito M, Koike A, Koizumi N, Sano H. Methylated DNA-binding proteins from Arabidopsis. Plant Physiology (2003) 133:1747–1754.
Kahana JA, Cleveland DW. Beyond nuclear transport: Ran-GTP as a determinant of spindle assembly. Journal of Cell Biology (1999) 146:1205–1209.
Kovarik A, Koukalova B, Holy A, Bezdek M. Sequence-specific hypomethylation of the tobacco genome induced with dihydroxypropyladenine, ethinine and 5-azacytidine. FEBS Letters (1994) 353:309–311.[CrossRef][Web of Science][Medline]
Leonhardt H, Page AW, Weier HU, Bestor TH. A targeting sequence directs DNA methyltransferase to sites of DNA replication in mammalian nuclei. Cell (1992) 71:865–873.[CrossRef][Web of Science][Medline]
Margot JB, Ehrenhofer-Murray AE, Leonhardt H. Interactions within the mammalian DNA methyltransferase family. BMC Molecular Biology (2003) 4:1–9.[CrossRef][Medline]
Mertinart C, Yoder JA, Taketo T, Laird DW, Trasler JM, Bestor TH. Sex-specific exons control DNA methyltransferase in mammalian germ cells. Development (1998) 125:889–897.[Abstract]
Milutinovic S, Brown SE, Zhuang Q, Szyf M. DNA methyltransferase 1 knock down induces gene expression by a mechanism independent of DNA methylation and histone deacetylation. Journal of Biological Chemistry (2004) 279:27915–27927.
Murray MG, Thompson WF. Rapid isolation of high molecular weight plant DNA. Nucleic Acids Research (1980) 8:4321–4325.
Nakano Y, Steward N, Sekine M, Kusano T, Sano H. A tobacco NtMET1 cDNA encoding a DNA methyltransferase: molecular characterization and abnormal phenotypes of transgenic tobacco plants. Plant and Cell Physiology (2000) 41:448–457.[Web of Science][Medline]
Nishiyama R, Wada Y, Mibu M, Yamaguchi Y, Shimogawara K, Sano H. Role of a nonselective de novo DNA methyltransferase in maternal inheritance of chloroplast genes in the green alga, Chlamydomonas reinhardtii. Genetics (2004) 168:809–916.
Okano M, Bell DW, Haber DA, Li E. DNA methyltransferases Dnmt3b are essential for de novo methylation and mammalian development. Cell (1999) 99:247–257.[CrossRef][Web of Science][Medline]
Pradhan S, Cummings M, Roberts RJ, Adams RLP. Isolation, characterization and baculovirus-mediated expression of the cDNA encoding cytosine DNA methyltransferase from Pisum sativum. Nucleic Acids Research (1998) 26:1214–1222.
Robertson KD, Ait SA, Yokochi T, Wade PA, Jones PL, Wolffe AP. DNMT1 forms a complex with Rb, E2F1 and HDAC1 and represses transcription from E2F-responsive promoters. Nature Genetics (2000) 25:338–342.[CrossRef][Web of Science][Medline]
Ronemus MJ, Galbiati M, Ticker C, Chen J, Dellaporta SL. Demethylation-induced developmental pleiotropy in Arabidopsis. Science (1996) 273:654–657.[Abstract]
Rountree MR, Bachman KE, Baylin SB. DNMT1 binds HDAC2 and a new co-repressor, DMAP1, to form a complex at replication foci. Nature Genetics (2000) 5:269–277.
Seo S, Okamoto M, Seto H, Ishizuka K, Sano H, Ohashi Y. Tobacco MAP kinase: a possible mediator in wound signal transduction pathways. Science (1995) 270:1988–1992.
Steward N, Kusano T, Sano H. Expression of ZmMET1, a gene encoding a DNA methyltransferase from maize, is associated not only with DNA replication in actively proliferating cells, but also with altered DNA methylation status in cold-stressed quiescent cells. Nucleic Acids Research (2000) 28:3250–3259.
Verwoerd TC, Dekker BM, Hoekema A. A small-scale procedure for the rapid isolation of plant RNAs. Nucleic Acids Research (1989) 17:2362.
Wada Y, Ohya H, Yamaguchi Y, Koizumi N, Sano H. Preferential de novo methylation of cytosine residues in non-CpG sequences by a domains rearranged DNA methyltransferase from tobacco plants. Journal of Biological Chemistry (2003) 278:42386–42393.
Yamaguchi Y, Yoda H, Akiyama K, Sano H. Induction of transcripts encoding a novel transmembrane protein during the hypersensitive response upon tobacco mosaic virus infection in tobacco plants. Planta (2003) 218:249–253.[CrossRef][Web of Science][Medline]
Yano A, Kodama Y, Koike A, Shinya T, Kim HJ, Matsumoto M, et al. Interaction between methylated-DNA binding protein and Ran GTPase during cell division in tobacco cultured cells. Annals of Botany (2006) 98:1179–1187.
Yap YK, Kodama Y, Waller F, Chung KM, Ueda H, Nakamura K, et al. Activation of a novel transcription factor through phosphorylation by WIPK, a wound-induced mitogen-activated protein kinase in tobacco plants. Plant Physiology (2005) 139:127–137.
Yoder JA, Walsh CP, Bestor TH. Cytosine methylation and the ecology of intragenomic parasites. Trends in Genetics (1997) 13:335–340.[CrossRef][Web of Science][Medline]
Zheng Y. G protein control of microtubule assembly. Annual Review of Cell and Developmental Biology (2004) 20:867–894.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||