* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Epigenetic Inactivation of Chalcone Synthase-A
Oncogenomics wikipedia , lookup
History of RNA biology wikipedia , lookup
Genomic library wikipedia , lookup
DNA vaccination wikipedia , lookup
Nucleic acid double helix wikipedia , lookup
SNP genotyping wikipedia , lookup
Molecular cloning wikipedia , lookup
Metagenomics wikipedia , lookup
No-SCAR (Scarless Cas9 Assisted Recombineering) Genome Editing wikipedia , lookup
Genetic engineering wikipedia , lookup
Cre-Lox recombination wikipedia , lookup
Extrachromosomal DNA wikipedia , lookup
Transcription factor wikipedia , lookup
DNA supercoil wikipedia , lookup
RNA interference wikipedia , lookup
Site-specific recombinase technology wikipedia , lookup
Polycomb Group Proteins and Cancer wikipedia , lookup
Non-coding RNA wikipedia , lookup
Point mutation wikipedia , lookup
Long non-coding RNA wikipedia , lookup
Designer baby wikipedia , lookup
Nucleic acid analogue wikipedia , lookup
Transgenerational epigenetic inheritance wikipedia , lookup
Epigenetics of neurodegenerative diseases wikipedia , lookup
Epigenetics of depression wikipedia , lookup
Epigenetic clock wikipedia , lookup
Genome editing wikipedia , lookup
Cell-free fetal DNA wikipedia , lookup
Non-coding DNA wikipedia , lookup
DNA methylation wikipedia , lookup
Epitranscriptome wikipedia , lookup
Vectors in gene therapy wikipedia , lookup
Epigenetics of human development wikipedia , lookup
Microevolution wikipedia , lookup
RNA silencing wikipedia , lookup
Behavioral epigenetics wikipedia , lookup
Epigenetics wikipedia , lookup
Deoxyribozyme wikipedia , lookup
Helitron (biology) wikipedia , lookup
Cancer epigenetics wikipedia , lookup
Epigenetics of diabetes Type 2 wikipedia , lookup
Epigenetics in stem-cell differentiation wikipedia , lookup
Artificial gene synthesis wikipedia , lookup
History of genetic engineering wikipedia , lookup
Epigenomics wikipedia , lookup
Epigenetics in learning and memory wikipedia , lookup
Therapeutic gene modulation wikipedia , lookup
Primary transcript wikipedia , lookup
Plant Cell Physiol. 48(4): 638–647 (2007) doi:10.1093/pcp/pcm028, available online at www.pcp.oxfordjournals.org ß The Author 2007. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: [email protected] Short Communication Epigenetic Inactivation of Chalcone Synthase-A Transgene Transcription in Petunia Leads to a Reversion of the Post-Transcriptional Gene Silencing Phenotype Akira Kanazawa 1, *, Michael O’Dell 2 and Roger P. Hellens 3 1 Graduate School of Agriculture, Hokkaido University, Sapporo, 060-8589 Japan Department of Genetics, John Innes Centre, Norwich Research Park, Colney, Norwich NR4 7UH, UK 3 Gene Discovery and Function Department, HortResearch, Private Bag 92 169, Auckland, New Zealand 2 of DNA methylation in the affected promoters, as well as changes in histone modification. The siRNAs may also direct DNA methylation in the nucleus (for a review, see Matzke et al. 2004). Like siRNAs, small RNAs called micro RNAs also negatively regulate the expression of endogenous genes through either RNA cleavage or arrest of translation, which is another pathway of RNA silencing (reviewed by Baulcombe 2004). Overexpression of the chalcone synthase-A (CHS-A) gene under the control of the cauliflower mosaic virus (CaMV) 35S promoter and the nopaline synthase (NOS) terminator causes the production of white sectors or completely white flowers in transformed petunia (Petunia hybrida) plants, and was one of the first examples of PTGS (also termed co-suppression; Napoli et al. 1990, Van der Krol et al. 1990). The production of wild-type pigment is inhibited because CHS performs an essential step in the biosynthesis of anthocyanins. Various silencing patterns have been observed in the petunia CHS-A silencing system (Jorgensen 1995, Jorgensen et al. 1996), and patterns can be transmitted both somatically and germinally (Jorgensen 1995). Using a transgene-induced silencing system that exploits genes involved in anthocyanin biosynthsis, Sijen et al. (2001) demonstrated that a transgene that expresses dsRNA can induce PTGS when the coding sequence is transcribed, and can induce TGS when the promoter sequence is transcribed. A flower color pattern in nontransgenic petunia plants has also been attributed to sequence-specific degradation of the CHS-A RNA (Koseki et al. 2005), suggesting that similar mechanisms are involved in the induction of CHS-A silencing in transgenic and non-transgenic petunia plants. We have maintained silenced lines of petunia including the line C001, which produces completely silenced white flowers (for details of plant materials, see Materials and Methods). A C001 plant is male-sterile and has been maintained through backcrossing as a dominant heterozygous trait for410 generations. The purple-flowered Petunia plants that exhibit a white-flowering phenotype as a consequence of chalcone synthase transgene-induced silencing occasionally give rise to revertant branches that produce flowers with wild-type pigmentation. Transcription run-on assays confirmed that the production of white flowers is caused by post-transcriptional gene silencing (PTGS), and indicated that transgene transcription is repressed in the revertant plants, providing evidence that induction of PTGS depends on the transcription rate. Transcriptional repression of the transgene was associated with cytosine methylation at CpG, CpNpG and CpNpN sites, and the expression was restored by treatment with either 5-azacytidine or trichostatin A. These results demonstrate that epigenetic changes occurred in the PTGS line, and these changes interfere with the initiation of transgene transcription, leading to a reversion of the PTGS phenotype. Keywords: Bisulfite sequencing analysis — Cauliflower mosaic virus 35S promoter — Chalcone synthase — Cytosine methylation — Petunia hybrida — Posttranscriptional gene silencing. Abbreviations: CaMV, cauliflower mosaic virus; CHS, chalcone synthase; dsRNA, double-stranded RNA; Npt, neomycin phosphotransferase; NOS, nopaline synthase; OCS, octopine synthase; PTGS, post-transcriptional gene silencing; RdDM, RNA-directed DNA methylation; RT–PCR, reverse transcription– PCR; siRNA, short interfering RNA; TGS, transcriptional gene silencing. The silencing of genes induced by the presence of homologous double-stranded RNA (dsRNA) sequences is termed RNA silencing. RNA silencing in plants includes post-transcriptional gene silencing (PTGS) and transcriptional gene silencing (TGS; for a review, see Baulcombe 2004). PTGS involves sequence-specific degradation of RNA in the cytoplasm and is mediated by short interfering RNAs (siRNAs). TGS is the repression or inactivation of transcription and is associated with an increased level *Corresponding author: E-mail, [email protected]; Fax, þ81-11-706-4933. 638 Epigenetic changes suppressing PTGS C002 line arose from a C001 plant as a spontaneous revertant; the mechanism of the reversion is unknown. The CHS-A mRNA levels differ greatly in the two lines, as demonstrated by RNA gel-blot analysis and reverse transcription–PCR (RT–PCR) (Metzlaff et al. 1997). The levels of CHS-A mRNA from both the CHS-A transgene and the endogenous CHS-A gene (CHS-A endogene) are very low in C001 flowers and leaves, as expected from its white flower color. In C002, transcripts from both the CHS-A transgene and the CHS-A endogene are high in flowers, as predicted from the purple flower color, whereas leaves contain low levels of the CHS-A endogene transcript and no observed CHS-A transgene transcript (Metzlaff et al. 1997, Metzlaff et al. 2000). In this study, we investigated the mechanisms underlying the reversion of the PTGS phenotype (white flowers) to the normal phenotype (purple flowers) by comparing the C001 and C002 lines. Transcription run-on assays were performed to determine the activity of both the CHS-A transgene and the CHS-A endogene in leaf and flower tissues of the C001 (silenced; white flowers) and C002 (revertant; purple flowers) lines (Fig. 1). We were able to distinguish the transcriptional activity of the two genes by hybridization of probes specific to an intron sequence (endogene specific) or the NOS terminator sequence (transgene specific) relative to a positive control (Kanazawa et al. 2000; for probes, see Fig. 1B). The hybridization to CHS-A exon 1 (common to both the endogene and the transgene) reflects the transcription from both genes. The pBluescript plasmid and the Antirrhinum majus ubiquitin gene (EMBL: AMUBIMRP X67957) were used as negative and positive controls, respectively. Significant levels of CHS-A transgene transcription were detected in both flowers and leaves of C001 (Fig. 1A). Northern blot analysis showed that little to no CHS-A mRNA accumulated in the flowers or leaves of this line, and the steady-state mRNA levels of both the CHS-A endogene and the CHS-A transgene have also been shown to be very low by RT–PCR (Metzlaff et al. 1997). The run-on data demonstrate that the mechanism of the silencing in C001 is post-transcriptional, as was previously shown in other transgenic petunia lines (Van Blokland et al. 1994, Stam et al. 1998). There was no significant difference in the transcription rate of the CHS-A endogene in C001 and C002 flower tissues. However, differences were observed in the transcription rate of the CHS-A transgene in the two lines: only a weak signal hybridizing to the NOS terminator sequence was observed in C002 flower tissues, and no signal was detected in C002 leaf tissues (Fig. 1A). This result shows that transcription of the transgene was repressed during the conversion from C001 to C002. This observation was confirmed by the intensity of hybridization to the CHS-A A V26 639 C001 C002 J-type (silenced) (revertant) (silenced) pBS CHS exon 1 CHS intron Flower NOS ter Ubiquitin pBS CHS exon 1 Leaf CHS intron NOS ter Ubiquitin B CHS-A endogene Exon 1 Exon 2 Intron CHS exon 1 probe CHS intron probe 0.1 kb CHS-A transgene CaMV 35S pro NOS T CHS exon 1 probe NOS terminator probe Fig. 1 The production of white flowers in the transgenic petunia plant is caused by PTGS, and transgene transcription is repressed in the revertant. (A) Run-on transcription assay using isolated nuclei from flower and leaf tissues of V26, C001, C002 and junction-type (J-type) plants. Nuclei were isolated from flowers and leaves of these plants, and used to synthesize radiolabeled run-on transcripts. The labeled RNA was hybridized to DNAs blotted on the filter that contain pBluescript plasmid (negative control), exon 1 of the CHS-A gene (hybridizes to transcripts from both the CHS-A transgene and the CHS-A endogene), the intron of the CHS-A gene (hybridizes to transcripts from the CHS-A endogene), the NOS terminator (hybridizes to transcripts from the CHS-A transgene) and the ubiquitin gene (positive control). For each material, the run-on assay was repeated three times, and the results of one of the experiments are shown. (B) Structure of the CHS-A endogene and the CHS-A transgene and positions of probes for the run-on assay. Exons of the CHS-A gene are indicated by hatched boxes. The positions of probes are indicated by lines. A dotted line indicates a portion of the CHS-A exon 1 probe that is hybridized with the intron of the CHS-A endogene transcript but not with the CHS-A transgene transcript. exon 1 sequence: in C002, very similar hybridization intensities to those observed for wild-type V26 flowers and leaves were observed, indicating that the transgene did not contribute to the labeled RNA run-on transcription pool. Quantification of the radioactivity of hybridization signals (CHS-A exon 1 vs. the ubiquitin gene) indicated that the total transcription activity of the CHS-A genes in flower tissues of C002 was reduced to 52.6 5.3% of the activity in 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 B − + CHS-A transgene/tubulin A 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 5-azacytidin C 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 D − + Trichostatin A − + 5-azacytidin CHS-A transgene/tubulin flower tissues of C001. These data explain the results of Northern blot and RT–PCR analyses (Metzlaff et al. 1997, Metzlaff et al. 2000) including the lack of CHS-A transgene transcript in C002 leaf tissues where no transcriptional activity of the gene was detected (see Fig.1A). A transcription run-on assay was also performed using nuclei isolated from a transgenic plant line that produces purple flowers with white centers of various sizes, which is called a ‘junction’-patterned flower phenotype (see Jorgensen 1995). This line contains a single-copy transgene, which may possibly account for the less stable phenotype of PTGS than in C001 that produces completely silenced flowers (discussed below). The transcription rate of the CHS-A transgene in the junction-type (J-type) plant was slightly lower than the transcription rate in C001, due possibly to the lower copy number of the gene, and was apparently higher than that in C002 (Fig. 1A). This result also indicates that the transgene transcription is repressed in C002. We suspected that the reversion to a purple-flowering phenotype was epigenetic, since regenerated calli from C002 leaves produced some white-flowering individuals (R. P. Hellens et al. unpublished data). This result suggests that rather than resulting from a mutation at the DNA level, the change from C001 to C002 is reversible, most probably involving a heritable epigenetic modification. We therefore examined whether the level of mRNA from the CHS-A transgene was affected by treatments of plants with the demethylating agent 5-azacytidine, or with trichostatin A, a specific inhibitor of histone deacetylase. Trichostatin A is also known to cause demethylation in Neurospora (Selker 1998) and plants (Lawrence et al. 2004), as well as down-regulation of DNA methyltransferase in human cells (Januchowski et al. 2006). Surface-sterilized seeds of C002 were sown on a solid medium containing 20 mM 5-azacytidine or 2 mM trichostatin A. Twenty plants were grown on the medium in a plate, and RNA was extracted from whole seedlings of plants grown on a plate. This experiment was repeated three times. Changes in the levels of mRNAs from the CHS-A endogene and the CHS-A transgene were analyzed by real-time RT–PCR. The levels of these mRNAs were quantified relative to the level of a-tubulin mRNA. The level of the mRNA from the CHS-A transgene was found to increase prominently in C002 plants after treatment with 5-azacytidine, whereas the mRNA level of the CHS-A endogene was unchanged (Fig. 2A, B). Similar results were obtained when plants were treated with trichostatin A (Fig. 2C, D). No PTGS was induced in seedlings by these treatments, although the transcription of the transgene appeared to be restored. This may be due to the intrinsically low level of transcription of the CHS-A endogene in plants of this developmental stage, and/or CHS-A endogene/tubulin Epigenetic changes suppressing PTGS CHS-A endogene/tubulin 640 5.0 4.0 3.0 2.0 1.0 0.0 − + Trichostatin A Fig. 2 Effects of 5-azacytidine and trichostatin A on the mRNA levels of the CHS-A endogene and the CHS-A transgene in young plants of C002. Real-time RT–PCR was conducted to analyze the mRNA levels of the CHS-A endogene (A and C) and the CHS-A transgene (B and D) in plants treated with 5-azacytidine (A and B) and trichostatin A (C and D). The mRNA levels were quantified relative to the mRNA level of the a-tubulin, and the relative values of plants treated with 5-azacytidine or trichostatin A and untreated control plants were compared. The value of untreated plants was set at 1. The data represent the mean and standard errors obtained from three replicates of the analysis. partial restoration of transgene transcription, which may not be sufficient to induce the RNA degradation reactions of PTGS. In any case, the results indicate that the mechanisms of transcriptional repression of the CHS-A transgene in C002 involve epigenetic changes including methylation of genomic DNA. We next examined the methylation status of the transgene promoter in C001 and C002 by bisulfite sequencing analysis. For each material, 20–22 clones obtained from four independent PCR amplifications from bisulfitetreated DNA templates were sequenced. The data were compiled in Fig. 3 and Table 1. The results indicated that cytosine residues in the 298 to 47 region of the CaMV 35S promoter were more methylated in C002 (Fig. 3B) than in C001 (Fig. 3A) at CpG, CpNpG and CpNpN sites. Cytosine methylation was detected in 10/10, 7/7 and 38/51 instances of CpG, CpNpG (symmetrical positions) and CpNpN (non-symmetrical positions) sequences, respectively, in the PCR products from C002. The frequency of cytosine methylation was 73.6, 60.4 and 23.5% at CpG, CpNpG and CpNpN sites, respectively, in C002. On the other hand, the frequency was 8.0, 0 and 10.5% at CpG, CpNpG and CpNpN sites, respectively, in C001 (Table 1). The presence of more methylcytosine residues overall in the Epigenetic changes suppressing PTGS A C001 641 CpG 1 CpNpG CpNpN 0.75 0.5 0.25 0 −298 −248 −198 −148 C002 B −98 CpG 1 As-1 CpNpG −48 CpNpN 0.75 0.5 0.25 0 −298 C −248 −198 −148 C002 (+ 5-azacytidine) −98 CpG 1 As-1 CpNpG −48 CpNpN 0.75 0.5 0.25 0 −298 D −248 −198 −148 C002 (+ trichostatin A) −98 CpG 1 As-1 CpNpG −48 CpNpN 0.75 0.5 0.25 0 −298 −248 −198 −148 −98 As-1 −48 Fig. 3 DNA methylation status of the CaMV 35S promoter. Sequencing data of PCR products amplified from bisulfite-treated DNA have been compiled. (A) C001 plants, (B) C002 plants, (C) C002 plants treated with 5-azacytidine, (D) C002 plants treated with trichostatin A. The height of the vertical lines shows the frequency of methylcytosines at respective positions per total PCR clones sequenced. Red, green and blue lines indicate frequencies of methycytosine at CpG, CpNpG and CpNpN sites, respectively. For C001, C002, 5-azacytidinetreated C002 and trichostatin A-treated C002 plants, 20, 22, 21 and 22 clones, were sequenced, respectively. Clones of each material contain products of four independent PCR amplifications from bisulfite-treated DNA templates. Numbers below the line indicate nucleotide positions relative to the transcription start site of the CaMV 35S promoter. The analyzed sequences cover the 298 to 47 region of the promoter. The position of two CpG sites in the as-1 element is indicated by arrows. 642 Table 1 Epigenetic changes suppressing PTGS A summary of bisulfite sequencing analysis of the CaMV 35S promoter in C001 and C002 plants Plant lines and treatments Frequency of cytosine methylation No. of sites at which cytosine methylation was detected Frequency of methylation at two CpG sites in the as-1 element (%) CpG CpNpG CpNpN Total CpG CpNpG CpNpN Total C001 8.0% (16/200) 0% (0/140) 10.5% (107/1020) 9.0% (123/1360) 3/10 0/7 21/51 24/68 25.0 C002 73.6% (162/220) 60.4% (93/154) 23.5% (264/1122) 34.7% (519/1496) 10/10 7/7 38/51 55/68 81.8 C002 (þ 5-azacytidine) 62.9% (132/210) 55.8% (82/147) 15.5% (166/1071) 26.6% (380/1428) 10/10 6/7 23/51 39/68 64.3 C002 (þ trichostatin A) 44.1% (97/220) 46.1% (71/154) 13.0% (146/1122) 21.0% (314/1496) 10/10 7/7 29/51 46/68 50.0 Data obtained from 20, 22, 21 and 22 clones of C001, C002, 5-azacytidine-treated C002 and trichostatin A-treated C002 plants, respectively, were compiled. Clones of each material contain products of four independent PCR amplifications from bisulfite-treated DNA templates. promoter region in C002 is consistent with the notion that cytosine methylation in the promoter interferes with transcription directly or indirectly. The frequency of cytosine methylation decreased when C002 plants were treated with 5-azacytidine (Fig. 3C) or trichostatin A (Fig. 3D). The frequencies of cytosine methylation at CpG/CpNpG/CpNpN sites were reduced to 62.9%/55.8%/15.5% and 44.1%/46.1%/13.0% by treatments with 5-azacytidine and trichostatin A, respectively. The CaMV 35S promoter contains a 21 bp element (83 to 63) designated as the activation sequence-1 (as-1) that binds basic leucine zipper-type transcription factors (Benfey and Chua 1990). The binding of nuclear factors to the as-1 element is influenced by cytosine methylation in vitro (Kanazawa et al. 2007). The frequency of CpG methylation at 79 and 67 in the as-1 element was higher in C002 (81.8%) than in C001 (25.0%), and was also reduced to 64.3 and 50.0% by treatments with 5-azacytidine and trichostatin A, respectively. These results are consistent with the restoration of the mRNA levels of the CHS-A transgene by these agents (see Fig. 2), and support the notion that methylation of cytosines in the promoter is involved in the transcriptional repression of the CHS-A transgene in C002. It appeared that 5-azacytidine was slightly more effective than trichostatin A in the restoration of the mRNA level, but was less effective than trichostatin A in demethylation. This implies that the transcriptional repression may also be mediated by chromatin modification rather than DNA methylation alone, and the mechanisms of the transcriptional restoration by these agents may not be identical. To examine whether the structure of the transgene locus has a specific feature that may account for PTGS of the CHS-A genes and/or transcriptional repression of the CHS-A transgene, we analyzed the genomic DNA of the transgene locus (Fig. 4). Using a Southern blot strategy previously reported (Cluster et al. 1996), two sequences, the neomycin phosphotransferase II (NptII) gene and the CaMV 35S promoter, were used as probes for the analysis. This analysis distinguishes structural organizations of the Agrobacterium-transferred DNA in the genome including three possible orientations of the DNA copies as repeated sequences (Fig. 4C–E) as well as those dispersed in the genome (Fig. 4B). Hybridization signals of 6.6 and 7.2 kb were detected with the NptII probe when C001 DNA was digested with HindIII and XbaI, respectively. In addition, two signals were detected with the CaMV 35S promoter probe for both HindIII- and XbaI-digested DNA (Fig. 4F). These results indicate that the Agrobacterium-transferred DNA (Fig. 4A) is integrated in the genome as an inverted repeat centered on the left border (Fig. 4D). We also looked for the presence or absence of siRNA, a hallmark of the occurrence of RNA silencing. A Northern blot analysis of the small RNA fraction revealed that siRNAs corresponding to the CHS-A coding sequence accumulated in the C001 line, but not the C002 line (Fig. 4G). In contrast, no siRNA corresponding to the CaMV 35S promoter sequence was detected in either C001 or C002 lines (data not shown). The results of the transcription run-on assays confirmed previous observations that showed that the white phenotype of C001 flowers is caused by PTGS Epigenetic changes suppressing PTGS A F NOS pro 1 kb OCS ter NOS ter NptII CHS-A LB LB aI Xb II I nd Hi aI Xb 35S pro HindIII XbaI B III nd Hi 643 kb 19.3 RB 7.7 6.2 RB 4.2 XbaI 3.5 HindIII C LB RB XbaI 2.7 LB 5.2 kb XbaI HindIII 5.8 kb HindIII D RB LB Probe: NptII G E XbaI 7.2 kb XbaI HindIII 6.6 kb HindIII LB RB LB XbaI HindIII 6.2 kb Probe: 35S pro RB 6 V2 C0 01 002 C siRNA RB XbaI 20 mer tRNA + 5S rRNA 6.2 kb HindIII Fig. 4 Structural organization of Agrobacterium-transferred DNA and accumulation of CHS-A siRNA in the transgenic petunia plants. (A) Structure of transgenes used for transformation. Restriction sites of HindIII and XbaI are indicated below the map. Directions of transcription are indicated by arrows. NOS pro, nopaline synthase promoter; OCS ter, octopine synthase terminator; LB, left border; RB, right border. (B–E) Possible organization of Agrobacterium-transferred DNA in the petunia genome and expected sizes of HindIII- or XbaI-digested DNA fragments. (B) A dispersed copy; (C) a right border-centered inverted repeat; (D) a left border-centered inverted repeat; (E) a direct repeat. White and black arrows indicate the CHS-A transgene and the NptII gene, respectively. Positions of sequences hybridizing with the probes for the CaMV 35S promoter and the NptII gene are indicated by open and filled boxes, respectively, below the arrows. The expected sizes of HindIII- or XbaI-digested DNA fragments are shown. The sizes of other fragments depend on the distance to the nearest flanking genomic restriction site. (F) Southern blot analysis of C001 DNA. HindIII- or XbaI-digested DNA was hybridized with probes specific for the NptII gene or the CaMV 35S promoter sequences. Note that the observed hybridization profiles were consistent with the left border-centered inverted repeat structure (D): the NptII probe hybridized with a 6.6 kb HindIII-digested fragment and a 7.2-kb XbaIdigested fragment, and the CaMV 35S promoter probe hybridized with two HindIII-digested fragments and two XbaI-digested fragments. (G) Northern blot analysis of low molecular weight RNA of flower tissues from V26, C001 or C002, probed for the CHS-A gene. A DNA oligonucleotide (20-mer) was also loaded as a size control; its position is indicated on the left. Ethidium bromide-stained tRNA and 5S rRNA bands are shown as a loading control. (Metzlaff et al. 1997, Metzlaff et al. 2000). Despite the extensive understanding of proteins involved in PTGS, primary factor(s) that are responsible for initiation of transgene-mediated silencing have not been fully understood. In particular, PTGS caused by transcripts from inverted repeat DNA remains one of the least understood RNA silencing processes in plants (reviewed by Brodersen and Voinnet 2006). The different transcription rates in C001 and C002 flowers (see Fig. 1) appear to be responsible for determining whether PTGS is triggered. This resembles an observation that the transgene transcription rate was reduced in revertants of co-suppressed nitrate reductase genes in tobacco (Vaucheret et al. 1997). Although it is unknown why a partial reduction in transgene transcription leads to a complete loss of PTGS, this phenomenon is consistent with the RNA threshold model (Smith et al. 1994), in which mRNA degradation is triggered when the mRNA accumulates to a certain level. This notion is similar to that in a report in which the frequency of PTGS occurrence in petunia is correlated with the strength of the promoter of a transgene (Que et al. 1997). On the contrary, lack of correlation between the transcription rate of a transgene and induction of PTGS has also been reported based on comparisons between independently transformed 644 Epigenetic changes suppressing PTGS plant lines (Van Blokland et al. 1994, English et al. 1996). In the present study, we compared C001 and C002 plants whose genetic backgrounds (e.g. locus and copy number of transgene) are the same. This eliminated possible genetic differences between materials, and the results consequently provided strong support for the threshold model. Neither swapping the CaMV 35S promoter with a weaker promoter nor swapping the NOS terminator with a less stable terminator caused CHS-A PTGS in petunia (R. P. Hellens et al. unpublished data). In light of these results, the present study suggests that the triggering of PTGS depends on the amount of CHS-A transgene transcript that has accumulated and/or the transcription rate of the CHS-A transgene at a certain stage of development. An inverted repeat structure of a gene has often been implicated in efficient silencing of the gene (reviewed by Muskens et al. 2000). It is possible that the repeated structure of the CHS-A transgene may account for the stable induction of PTGS that causes production of completely white flowers in C001 if read-through transcription occurs over the repeated transgene copies and, consequently, produces dsRNA. In this scenario, de novo methylation of the corresponding coding sequence via an RNA intermediate (reviewed by Wassenegger 2000, Matzke et al. 2004) is conceivable. A secondary and more efficient long-term response may be to spread this methylation (see Wassenegger 2000, Fojtova et al. 2003, Mishiba et al. 2005) to promoter regions, which results in the activation of TGS. RNA-directed DNA methylation (RdDM) of the transgene promoter may also occur directly if unexpected transcription from upstream of the promoter occurs, together with the read-through transcription. These transcription events may cause the production of dsRNA corresponding to the CaMV 35S promoter sequence. DsRNA corresponding to even a portion of the promoter might induce TGS: we have demonstrated that a 120 bp portion of the CaMV 35S promoter cloned in an RNA virus vector can induce both RdDM and TGS of a reporter gene driven by the promoter (Otagaki et al. 2006). Although we have not detected siRNA corresponding to the CaMV 35S promoter sequence, it is possible that extremely small amounts of siRNA corresponding to the CaMV 35S promoter sequence are present. RdDM mediated by siRNAs whose level is below the limit of detection of a standard gel-blot method has also been postulated in the silencing system of PAI genes in Arabidopsis thaliana (for a review, see Mathieu and Bender 2004). In addition to these RNA-mediated mechanisms, de novo methylation in the transgene promoter sequences may also be caused by DNA–DNA interaction between inverted repeats of the transgene (see Muskens et al. 2000). Bisulfite sequencing analysis revealed an elevated frequency of cytosine methylation in the transgene promoter in C002 (Fig. 3). Unlike the methylation status of most of the cytosine residues analyzed in the present study, a slightly higher level of methylation was detected in C001 at a restriction site located far upstream (1.2 kb) of the promoter (O’Dell et al. 1999). One plausible explanation for the hypermethylation at the site may be that RNA-directed de novo methylation preferentially occurred at limited cytosines as a consequence of transgene transcription in C001. It is also possible that the methylated state at the site was established and maintained in C001, whereas the state was destabilized in C002. Although whether the methylation status at the site reflects methylation frequency in its surrounding region remains to be examined, these results suggest that cytosine methylation in the distal region is not directly involved in transcriptional repression. In contrast, a higher frequency of cytosine methylation overall within 0.3 kb upstream of the transcription start site of the promoter has been detected in various transformed plants that exhibit TGS (see, for example, Meyer et al. 1994) as observed in the present study. We presume that the cytosine methylation that primarily affects promoter activity may be limited to the proximal region of the promoter in C001 and C002. The observed changes in DNA methylation status may be closely associated with changes in chromatin structure. A model of TGS suggests that a methylcytosine-binding protein (such as MeCP2) recruits Sin3 and histone deacetylases to the chromatin containing methylated DNA to which it is bound, subsequently leading to an alteration in the chromatin structure (reviewed by Kass et al. 1997, Ashraf and Ip 1998). A recent report also suggested that heterochromatin formation marked by site-specific histone modifications may in turn lead to DNA methylation by DNA methyltransferase recruited by heterochromatin protein 1 (HP1; Jackson et al. 2002). It is known that heterochromatin-mediated silencing can spread (Noma et al. 2004), although a direct relationship between spreading of DNA methylation and heterochromatin formation has not yet been understood. Our findings have shown that epigenetic change(s) occurred during the process that resulted in the generation of the C002 line from the C001 line, as evidenced by the difference in the methylation statuses of the transgene promoter as well as the restoration of transgene expression with either 5-azacytidine or trichostatin A in C002 plants. The results demonstrate that at least three events were involved in the change: (i) epigenetic changes including cytosine methylation of the transgene promoter; (ii) repression of transgene transcription; and (iii) suppression of the specific degradation of the CHS-A mRNA. If the epigenetic changes occur, such as in C002, there is a very low rate of transcription of the transgene, and PTGS can no longer initiate (Fig. 5). Somatic conversion from PTGS Epigenetic changes suppressing PTGS CHS-A transgene Epigenetic changes C ··· C m5 m5 C ··· C High rate of transcription PTGS Low rate of transcription (co-suppression) no PTGS White flower Purple flower Fig. 5 Schematic diagram of the generation of the C002 revertant line from the C001 PTGS line. In C001 plants, the CHS-A transgene is transcribed at a high rate that causes PTGS (co-suppression). Epigenetic changes that involve cytosine methylation in the CaMV 35S promoter occurred during the generation of C002, in which the CHS-A transgene is transcribed at a very low rate that does not cause PTGS. to TGS may explain at least some of the various phenotypes related to co-suppression and the changeable gene silencing phenotypes sometimes observed during development (see Jorgensen 1995). A similar change in a transgene conferring drug resistance was observed in a long-term tobacco callus culture (Fojtova et al. 2003). Our present results indicate that conversion from PTGS to TGS can occur during the growth of a plant as a consequence of the presence of a transgene homologous to an endogenous gene in the plant genome. Materials and Methods Transgenic petunia lines C001 and C002 (Metzlaff et al. 1997, O’Dell et al. 1999, Kanazawa et al. 2007) were used. These plants were obtained by the transformation of the wild-type plant V26 with the CHS-A transgene controlled by the CaMV 35S promoter and the NOS terminator (Napoli et al. 1990). C001 and C002 refer to pedigree numbers of a transformant CHS38 (Jorgensen et al. 1996), which was originally designated 218.38 (Napoli et al. 1990), and have also been designated CHS38W and CHS38P, respectively (R. A. Jorgensen, personal communication). C001 shows a silenced phenotype with white flowers, while C002 is a spontaneous revertant plant from C001 with purple flowers. A junction-type plant is a descendant of the CHS223 line (Cluster et al. 1996, Jorgensen et al. 1996) and contains a single-copy CHS-A transgene. Flowers (35–45 mm long) and young leaves of these plants were used for run-on transcription assays. The isolation of nuclei and transcription run-on assay were performed using the 645 mini-scale method described by Kanazawa et al. (2000). For each material, the experiment was repeated three times. For quantitative assays, membranes were exposed to imaging plates (Fuji Film, Tokyo, Japan) and radioactivity was quantified with a Bio-imaging analyzer (Fujix BAS 2000; Fuji Film). For treatment of plants with 5-azacytidine or trichostatin A, surface-sterilized seeds of C002 were sown on plates containing half concentration standard MS medium (Murashige and Skoog 1962) that contained 10 g l1 sucrose. The pH of the medium was adjusted to 5.7 before autoclaving. Media were solidified with 0.8% (w/v) agar. 5-Azacytidine or trichostatin A was added to the medium just before plates were poured. 5-Azacytidine (dissolved in water) was added to a final concentration of 20 mM. Trichostatin A (dissolved in methanol) was added to a final concentration of 2 mM. After sowing seeds, plates were incubated at a photoperiod of 16 h light/8 h dark, 248C for 1 month. We found that these treatments slightly affected the development of plants (e.g. delay in germination). Total RNA was isolated from plants as described by Napoli et al. (1990), except that we removed genomic DNA from the RNA fraction using DNase I (TAKARA BIO INC., Otsu, Japan). A 1 mg aliquot of the total RNA was used as the template for cDNA synthesis. The cDNA synthesis reaction mixture was prepared by mixing 4 ml of 5 reaction buffer [250 mM Tris–HCl (pH 8.3), 375 mM KCl, 15 mM MgCl2], 2 ml of 0.1 M dithiothreitol (DTT), 1 ml of 100 mM oligo(dT)20 primer, 4 ml of 2.5 mM dNTPs, the total RNA solution, and water to a final volume of 19 ml. The mixture was heated at 658C for 5 min and rapidly cooled on ice. After the addition of 1 ml of reverse transcriptase (M-MLV, Invitrogen, Carlsbad, CA, USA), the cDNA synthesis was performed at 428C for 1 h. The reverse transcriptase was inactivated by heating the sample at 998C for 1 min. Real-time RT–PCR was carried out using a 1 ml aliquot of the reaction mixture and SYBR Premix Ex Taq kit (TAKARA BIO INC.) with a DNA Engine Opticon 2 System (MJ Research, Waltham, MA, USA). The PCR cycle was 958C for 30 s, 588C for 30 s, 728C for 30 s and 788C for 2 s. This cycle was repeated 40 times. Fluorescence quantification was carried out before and after the incubation at 788C to monitor for the formation of primer dimers. A reaction mixture without reverse transcriptase was used as a control to confirm that no amplification occurred from genomic DNA contaminates in the RNA sample. Primers for RT–PCR were as follows: CHS-A endogene, 50 -GATACTTACACTTGTCACGTA-30 (‘995’) and 50 -GTGCTT TGATCAACACAGTTTG-30 (‘2350’); CHS-A transgene, 50 -ACA CGCTCGAGCTCATTTC-30 (‘996’) and primer ‘2350’; a-tubulin, 50 -GCCACCATCAAGACCAAGC-30 (‘tubulin F’) and 50 -ACCT CAGCAACACTGGTTGA-30 (‘tubulin R’). Primers ‘995’ and ‘996’ anneal the 50 -untranslated sequences of the CHS-A endogene and the CHS-A transgene, respectively, and thereby allow amplification of transcripts specific to the CHS-A endogene and the CHS-A transgene, respectively. For analysis of DNA methylation by bisulfite sequencing, DNA was isolated from young seedlings of C001, C002, 5-azacytidine-treated C002 and trichostatin A-treated C002 plants using a Nucleon PhytoPure DNA extraction kit (Amersham Biosciences, Piscataway, NJ, USA). The protocol of bisulfite treatment in this study is based on the methods of Frommer et al. (1992) and Paulin et al. (1998). DNA was cleaved with the restriction enzyme HindIII, extracted with phenol/ chloroform, and precipitated by ethanol. The cleaved DNA was alkali denatured in 0.3 M NaOH at 378C for 20 min. Denatured DNA was incubated in a total volume of 600 ml with freshly prepared 6.4 M urea/3.1 M sodium bisulfite/0.5 mM hydroquinone, 646 Epigenetic changes suppressing PTGS pH 5.0, at 608C for 30 h under mineral oil. DNA was then recovered by a Qiaquick PCR purification kit (Qiagen GmbH, Hilden, Germany). NaOH was added to the DNA solution to a concentration of 0.3 M and then incubated at 378C for 20 min. Glycogen and ammonium acetate were added to the solution to final concentrations of 0.16 mg ml1 and 2.64 M, respectively. DNA was then precipitated with ethanol and dissolved in 20 ml of TE (pH 8.0) buffer. Two rounds of PCR were carried out for amplification. The first round of the PCR was carried out in a total volume of 50 ml using 1 ml of bisulfite-treated DNAs as a template. The second round of the PCR was carried out using a 1 ml aliquot of the reaction mixture from the first round of PCR. To amplify target sequences, primers ‘35S –346F bisulfite T’ (50 -TATTGAGATTTTTTAATAAAGGGTAA-30 ) and ‘35S þ 1R bisulfite A’ (50 -TCCTCTCCAAATAAAATAAACTTC-30 ) were used for the first round PCR, and primers ‘35S –323F bisulfite T’ (50 -TAATATTTGGAAATTTTTTTGGATT-30 ) and ‘35S –21R bisulfite A’ (50 -TTCCTTATATAAAAAAAAAATCTTAC-30 ) were used for the second round of PCR. The PCR cycling conditions were: 948C for 30 s, 528C for 30 s and 728C for 1 min. This cycle was repeated 40 times, and the reaction mixture was then further incubated at 728C for 10 min. The PCR products were cloned into the pGEM-T Easy vector (Promega, Madison, WI, USA) and were subjected to sequence analysis. As a control to ensure that bisulfite treatment was complete, DNA isolated from Arabidopsis thaliana leaves was simultaneously treated. A region of the A. thaliana ASA1 gene that is not methylated was amplified as previously reported (see Kusaba et al. 2003). All five cloned sequences of PCR products showed complete conversion of cytosines to thymidines. The structural organization of transgenes in the petunia genome was analyzed by DNA gel-blot analysis. Total DNA was isolated from leaves as described by Kanazawa and Tsutsumi (1992). DNA was digested with restriction enzymes and fractionated by electrophoresis on a 1% (w/v) agarose gel. After electrophoresis, the DNA was transferred to a nylon membrane (Hybond Nþ, Amersham Biosciences) and allowed to hybridize with labeled probes. Labeling of probes, hybridization, washing of membranes and detection of signals were carried out using the Alkphos Direct nucleic acid labeling and detection system (Amersham Biosciences). The CaMV 35S promoter sequence amplified by PCR using primers ‘35S –345F’ and ‘35S þ 1R’ (Kanazawa et al. 2007) and the NptII gene sequence amplified by PCR using primers 50 -GAATGAACTCCAAGACGAGG-30 and 50 -AAGAACTCGTCAAGAAGGCG-30 were labeled for use as hybridization probes. The CHS-A siRNAs were detected by Northern blot analysis of the small RNA fraction using a digoxigenin-labeled probe as described by Goto et al. (2003). Acknowledgments We would like to express our appreciation to Cathie Martin for giving us helpful advice and constructive comments throughout this research and on the manuscript. We are grateful to Richard B. Flavell for valuable suggestions and financial support, and Michael Metzlaff for helpful discussions. We are also grateful to Peter Shaw, Alison Beven and Nadia S. Al-Kaff for technical advice on transcription run-on assay, Tomoki Matsuyama, Mineo Senda and Makoto Kusaba for technical advice on bisulfite sequencing analysis, Chikara Masuta and Kazunori Goto for valuable information on virus-induced gene silencing, Andrea Davies, Akiko Takahashi, Maiko Koseki and Shungo Otagaki for technical assistance, Sayuri Tsukahara for her help in constructing a bisulfite sequencing system, and Neal Gutterson and Richard Jorgensen for plant materials. This work was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan. References Ashraf, S.I. and Ip, Y.T. (1998) Transcription control: repression by local chromatin modification. Curr. Biol. 8: R683–R686. Baulcombe, D. (2004) RNA silencing in plants. Nature 431: 356–363. Benfey, P.N. and Chua, N.-H. (1990) The cauliflower mosaic virus 35S promoter: combinatorial regulation of transcription in plants. Science 250: 959–966. Brodersen, P. and Voinnet, O. (2006) The diversity of RNA silencing pathways in plants. Trends Genet. 22: 268–280. Cluster, P.D., O’Dell, M., Metzlaff, M. and Flavell, R.B. (1996) Details of T-DNA structural organization from a transgenic Petunia population exhibiting co-suppression. Plant Mol. Biol. 32: 1197–1203. English, J.J., Mueller, E. and Baulcombe, D.C. (1996) Suppression of virus accumulation in transgenic plants exhibiting silencing of nuclear genes. Plant Cell 8: 179–188. Fojtova, M., van Houdt, H., Depicker, A. and Kovarik, A. (2003) Epigenetic switch from posttranscriptional to transcriptional silencing is correlated with promoter hypermethylation. Plant Physiol. 133: 1240–1250. Frommer, M., McDonald, L.E., Millar, D.S., Collis, C.M., Watt, F., Grigg, G.W., Molloy, P.L. and Paul, C.L. (1992) A genomic sequencing protocol that yields a positive display of 5-methylcytosine residues in individual DNA strands. Proc. Natl Acad. Sci. USA 89: 1827–1831. Goto, K., Kanazawa, A., Kusaba, M. and Masuta, C. (2003) A simple and rapid method to detect plant siRNAs using nonradioactive probes. Plant Mol. Biol. Rep. 21: 51–58. Jackson, J.P., Lindroth, A.M., Cao, X. and Jacobsen, S.E. (2002) Control of CpNpG DNA methylation by the KRYPTONITE histone H3 methyltransferase. Nature 416: 556–560. Januchowski, R., Da browski, M., Ofori, H. and Jagodzinski, P.P. (2006) Trichostatin A down-regulates DNA methyltransferase 1 in Jurkat T cells. Cancer Lett. 246: 313–317. Jorgensen, R.A. (1995) Cosuppression, flower color patterns and metastable gene expression states. Science 268: 686–691. Jorgensen, R.A., Cluster, P.D., English, J., Que, Q. and Napoli, C.A. (1996) Chalcone synthase cosuppression phenotypes in petunia flowers: comparison of sense vs. antisense constructs and single-copy vs. complex T-DNA sequences. Plant Mol. Biol. 31: 957–973. Kanazawa, A., O’Dell, M. and Hellens, R.P. (2007) The binding of nuclear factors to the as-1 element in the CaMV 35S promoter is affected by cytosine methylation in vitro. Plant Biol. in press. Kanazawa, A., O’Dell, M., Hellens, R.P., Hitchin, E. and Metzlaff, M. (2000) Mini-scale method for nuclear run-on transcription assay in plants. Plant Mol. Biol. Rep. 18: 377–383. Kanazawa, A. and Tsutsumi, N. (1992) Extraction of restrictable DNA from plants of the genus Nelumbo. Plant Mol. Biol. Rep. 10: 316–318. Kass, S.U., Pruss, D. and Wolffe, A.P. (1997) How does DNA methylation repress transcription?. Trends Genet 13: 444–449. Koseki, M., Goto, K., Masuta, C. and Kanazawa, A. (2005) The star-type color pattern in Petunia hybrida ‘Red Star’ flowers is induced by the sequence-specific degradation of the chalcone synthase RNA. Plant Cell Physiol 46: 1879–1883. Kusaba, M., Miyahara, K., Iida, S., Fukuoka, H., Takano, T., Sassa, H., Nishimura, M. and Nishio, T. (2003) Low glutelin content1: a dominant mutation that suppresses the Glutelin multigene family via RNA silencing in rice. Plant Cell 15: 1455–1467. Lawrence, R.J., Earley, K., Pontes, O., Silva, M., Chen, J., Neves, N., Viegas, W. and Pikaard, C.S. (2004) A concerted DNA methylation/ histone methylation switch regulates rRNA gene dosage control and nucleolar dominance. Mol. Cell 13: 599–609. Epigenetic changes suppressing PTGS Mathieu, O. and Bender, J. (2004) RNA-directed DNA methylation. J. Cell Sci. 117: 4881–4888. Matzke, M., Aufsatz, W., Kanno, T., Daxinger, L., Papp, I., Mette, M.F. and Matzke, A.J.M. (2004) Genetic analysis of RNA-mediated transcriptional gene silencing. Biochim. Biophys. Acta 1677: 129–141. Metzlaff, M., O’Dell, M., Cluster, P.D. and Fravell, R.B. (1997) RNA-mediated degradation and chalcone synthase A silencing in petunia. Cell 88: 845–854. Metzlaff, M., O’Dell, M., Hellens, R. and Flavell, R.B. (2000) Developmentally and transgene regulated nuclear processing of primary transcripts of chalcone synthase A in petunia. Plant J. 23: 63–72. Meyer, P., Niedenhof, I. and ten Lohuis, M. (1994) Evidence for cytosine methylation of non-symmetrical sequences in transgenic Petunia hybrida. EMBO J. 13: 2084–2088. Mishiba, K., Nishihara, M., Nakatsuka, T., Abe, Y., Hirano, H., Yokoi, T., Kikuchi, A. and Yamamura, S. (2005) Consistent transcriptional silencing of 35S-driven transgenes in gentian. Plant J. 44: 541–556. Murashige, T. and Skoog, F. (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant. 15: 473–497. Muskens, M.W.M., Vissers, A.P.A., Mol, J.N.M. and Kooter, J.M. (2000) Role of inverted DNA repeats in transcriptional and post-transcriptional gene silencing. Plant Mol. Biol. 43: 243–260. Napoli, C., Lemieux, C. and Jorgensen, R. (1990) Introduction of a chimeric chalcone synthase gene into petunia results in reversible co-suppression of homologous genes in trans. Plant Cell 2: 279–289. Noma, K., Sugiyama, T., Cam, H., Verdel, A., Zofall, M., Jia, S., Moazed, D. and Grewal, S.I.S. (2004) RITS acts in cis to promote RNA interference-mediated transcriptional and post-transcriptional silencing. Nat. Genet. 36: 1174–1180. O’Dell, M., Metzlaff, M. and Flavell, R.B. (1999) Post-transcriptional gene silencing of chalcone synthase in transgenic petunias, cytosine methylation and epigenetic variation. Plant J. 18: 33–42. Otagaki, S., Arai, M., Takahashi, A., Goto, K., Hong, J.-S., Masuta, C. and Kanazawa, A. (2006) Rapid induction of transcriptional and posttranscriptional gene silencing using a novel Cucumber mosaic virus vector. Plant Biotechnol. 23: 259–265. 647 Paulin, R., Grigg, G.W., Davey, M.W. and Piper, A.A. (1998) Urea improves efficiency of bisulphite-mediated sequencing of 50 -methylacytosine in genomic DNA. Nucleic Acids Res. 26: 5009–5010. Que, Q., Wang, H.-Y., English, J.J. and Jorgensen, R.A. (1997) The frequency and degree of cosuppression by sense chalcone synthase transgenes are dependent on transgene promoter strength and are reduced by premature nonsense codons in the transgene coding sequence. Plant Cell 9: 1357–1368. Selker, E.U. (1998) Trichostatin A causes selective loss of DNA methylation in Neurospora. Proc. Natl Acad. Sci. USA 95: 9430–9435. Sijen, T., Vijn, I., Rebocho, A., van Blokland, R., Roelofs, D., Mol, J.N.M. and Kooter, J.M. (2001) Transcriptional and posttranscriptional gene silencing are mechanistically related. Curr. Biol. 11: 436–440. Smith, H.A., Swaney, S.L., Parks, T.D., Wernsman, E.A. and Dougherty, W.G. (1994) Transgenic plant virus resistance mediated by untranslatable sense RNAs: expression, regulation and fate of nonessential RNAs. Plant Cell 6: 1441–1453. Stam, M., Viterbo, A., Mol, J.N.M. and Kooter, J.M. (1998) Position-dependent methylation and transcriptional silencing of transgenes in inverted repeats: implications for posttranscriptional silencing of homologous host genes in plants. Mol. Cell Biol. 18: 6165–6177. Van Blokland, R., Van der Geest, N., Mol, J.N.M. and Kooter, J.M. (1994) Transgene-mediated suppression of chalcone synthase expression in Petunia hybrida results from an increase in RNA turnover. Plant J. 6: 861–877. Van der Krol, A.R., Mur, L.A., Beld, M., Mol, J.N.M. and Stuitje, A.R. (1990) Flavonoid genes in petunia: addition of a limited number of gene copies may lead to a suppression of gene expression. Plant Cell 2: 291–299. Vaucheret, H., Nussaume, L., Palauqui, J.-C., Quillere, I. and Elmayan, T. (1997) A transcriptionally active state is required for post-transcriptional silencing (cosuppression) of nitrate reductase host genes and transgenes. Plant Cell 9: 1495–1504. Wassenegger, M. (2000) RNA-directed DNA methylation. Plant Mol. Biol. 43: 203–220. (Received October 30, 2006; Accepted February 17, 2007)