Developmentally non-redundant SET domain proteins SUVH2 and SUVH9 are required for transcriptional gene silencing in Arabidopsis thaliana
- 1.7k Downloads
In plants, RNA-directed DNA methylation (RdDM) and related transcriptional gene silencing (TGS) involve members of the suppressor of variegation 3-9-homologous (SUVH) group of putative histone methyltransferases. Utilizing a reverse genetic approach in Arabidopsis thaliana, we demonstrate that two closely related SUVH members, SUVH2 and SUVH9, act partially non-redundant in RdDM. DNA methylation, transcript accumulation and association with histone modifications were analyzed at the endogenous RdDM target AtSN1 (a SINE-like retroelement) in suvh2 and suvh9 single as well as suvh2 suvh9 double mutants. SUVH2 was found to be required for full DNA methylation at AtSN1 in early seed development and was also higher expressed in seeds than at later developmental stages. SUVH9 had its impact on RdDM later during vegetative development of the plant and was also higher expressed during that stage than at earlier developmental stages. The strongest reduction of RdDM at AtSN1 was found in suvh2 suvh9 double mutant plants. Histone 3-lysine 9-dimethylation (H3K9me2) associated with AtSN1 was reduced only in the simultaneous absence of functional SUVH2 and SUVH9. Thus, SUVH2 and SUVH9 functions in RdDM and TGS are overlapping in spite of some developmental specialization. Pol V specific transcripts were reduced in suvh2 suvh9 plants. This might indicate a role of these SUVH proteins in Pol V complex recruitment.
KeywordsSET domain protein RNA-directed DNA methylation AtSN1 SUVH2 SUVH9
SET domain containing proteins can be divided into four different classes typified by the corresponding proteins of Drosophila E(Z), TRX, ASH1 and SU(VAR)3-9 (Baumbusch et al. 2001; Thorstensen et al. 2011). In A. thaliana, 15 genes similar to the SU(VAR)3-9 were identified, of which 10 belong to the group of suppressor of variegation [SU(VAR)3-9] homologous genes (SUVH) defined by the presence of an internal YDG/SRA-domain and C-terminal pre-SET and SET domains. Among those, SUVH2 (encoded by At2g33290) and SUVH9 (encoded by At4g13460), show considerable sequence similarity (Naumann et al. 2005) and form a subgroup within the SUVH-family that is characterized by the absence of the catalytically important post-SET domain (Rea et al. 2000). SUVH2 has been reported to participate in gene silencing via heterochromatization (Naumann et al. 2005; Ay et al. 2009) and both, SUVH2 and SUVH9, are involved in DRM2-mediated de novo DNA methylation of particular target regions in A. thaliana (Johnson et al. 2008). The SET domain is essential for histone methyltransferase activity (Rea et al. 2000). Nevertheless, in vitro histone methylation tests were successful so far only for a subset of SUVH proteins, possibly due to rather specific substrate requirements of some of the family members. SUVH4, SUVH5 and SUVH6 readily methylate bovine H3 (Jackson et al. 2004; Ebbs and Bender 2006). SUVH5 additionally showed some activity on bovine H2A (Ebbs and Bender 2006). SUVH2 has been reported to methylate H4 in one study using reconstituted recombinant nucleosomes (Naumann et al. 2005), but was found inactive in a second study using bovine histones or A. thaliana nucleosomes (Johnson et al. 2008).
The second characterized domain in the SUVH proteins is the SET and RING associated (SRA) domain (Johnson et al. 2007, also termed YDG domain according to Baumbusch et al. 2001). This domain was shown to be responsible for the binding of proteins to methylated DNA in vitro (Rajakumara et al. 2011). The SRA domains of histone methyltransferases SUVH4 and SUVH6, which are linked to MET1- and CMT3-mediated maintenance of DNA methylation, have preferential affinity to methylated cytosines in CHG and CHH (H indicating A,C or T) context (Johnson et al. 2007). SUVH2 and SUVH9 are involved in de novo RNA-directed DNA methylation (RdDM) (Naumann et al. 2005; Johnson et al. 2008). The SRA domain of SUVH2 shows high affinity to methylated cytosines in CG motifs, while the SRA domain SUVH9 has the strongest affinity to methylated cytosines in CHH context (Johnson et al. 2008). Based on these results, Johnson et al. (2008) postulated that SUVH2 and SUVH9 may act in RdDM rather via binding to methylated cytosines in particular sequence contexts, than by actually methylating histones.
AtSN1 (A. thaliana SINE like element 1) is a putative transposable retroelement present in 71 copies in the A. thaliana Col-0 reference genome (Myouga et al. 2001). A single copy of AtSN1 located on chromosome 3 that is flanked by genic sequences has served as a model for the genetic analysis of RdDM of genomic sequences of A. thaliana in many studies (Zilberman et al. 2003; Xie et al. 2004; He et al. 2009a, b; Gao et al. 2010). Its internal region between the long terminal repeats (LTR) contains in total 44 cytosine sites; 11 in symmetric (4 CG sites, 7 CHG sites) and 33 in asymmetric (CHH) context. ~40 % of these cytosines are methylated in wild type plants. In symmetric context, cytosine methylation of AtSN1 and other plant sequences, is primarily mediated by the DNA methyltransferase MET1 (Finnegan et al. 1993; Lindroth et al. 2001; Bartee et al. 2001) while methylation in asymmetric context is mainly mediated by DRM2 (RdDM pathway; Cao and Jacobsen 2002). In addition to the de novo DNA methyltransferase DRM2, the DNA dependent RNA polymerases Pol VI and Pol V contribute to RdDM and transcriptional suppression of AtSN1 (Wierzbicki et al. 2008) by synthesis of siRNA homologous to the methylated region (Pontier et al. 2005). Further factors involved in RdDM are RDR2, DRD1 and DDM1 (Wierzbicki et al. 2009). Most likely, the RdDM pathway involves a self amplifying feedback loop of siRNA dependent DNA methylation and DNA methylation dependent Pol IV-RDR2 mediated formation of siRNAs (Law and Jacobsen 2010).
In wild type plants with functional RdDM, no AtSN1 sense transcripts are found (Wierzbicki et al. 2009; Haag et al. 2009), although the internal LTR-flanking region of AtSN1 shares structural similarity with promoters of tRNA (Box A and Box B) and of genes typically transcribed by DNA-dependent RNA Pol III (Myouga et al. 2001). Formation of AtSN1 sense transcripts is seen in mutants with disturbed RdDM such as nrpe1, ago4 and dms3. As the transcriptional activation was not associated with acetylation of H3 at lysines 9 and 14 or Pol II association at AtSN1, it is assumed that AtSN1 transcription is Pol III-dependent (Wierzbicki et al. 2009). AtSN1 silencing in wild type plants is associated with complementary and adjacent antisense transcripts, which are most likely products of Pol V (Wierzbicki et al. 2008) which are missing in nrpe1 mutant plants (NRPE1 is the largest subunit of the Pol V complex) and reduced in rdm3 mutants (He et al. 2009a, b, c). Interaction of NRPE1 with the AtSN1 region was shown by chromatin immunoprecipitation (ChIP) (Rowley et al. 2011).
In the silenced state, the AtSN1 region is associated with high levels of repressive histone marks such as H3K9me2 and H3K27me1, while acetylation of histone 3 is mainly absent (Wierzbicki et al. 2008). Addionally, association of AtSN1 with H3K4me2, a mark associated with active chromatin was described, defining this region as “intermediate heterochromatin” (Habu et al. 2006). In plants without functional Pol V, the level of H3K9me2 remained unaltered in the internal AtSN1 region (Wierzbicki et al. 2008).
Here, we show that the two putative histone methyltransferases SUVH2 and SUVH9 are differentially expressed during plant development and contribute, at least in part, non-redundantly to RdDM. SUVH2 plays a major role at the seedling stage and SUVH9 during vegetative growth. Nevertheless, both are important for transcriptional silencing and persistence of repressive histone marks.
Materials and methods
Plant material and growth conditions
The A. thaliana accession Col-0 was used in all experiments. The suvh2 T-DNA insertion lines were ordered from NASC (Nottingham Arabidopsis Stock Centre, T-DNA insertion line Gabi_kat_516A07, Suppl. Figure 2). The suvh9 T-DNA insertion line was ordered from NASC (T-DNA insertion line SALK_048033). Double mutant suvh2 suvh9 line was generated by crossing Gabi_kat_516A07 and SALK_048033 and genotyping of the F3 generation. SUVH2 over-expression plants were obtained from the laboratory of G. Reuter (Martin-Luther University Halle-Wittenberg, Germany). These plants contain a strong constitutive expressed myc-tagged SUVH2 gene controlled by the CaMV35 promoter. Plants with enhanced level of SUVH2 were selected by their dwarf phenotype, curled leaves and delayed senescence.
Plants were cultivated on soil at 21 °C under a 16 h light/8 h dark (long day) regime for propagation and seed production or under an 8 h light/16 h dark (short day) regime to generate material for RNA and DNA analysis. All seeds were stratified before germination for 3 days at 4 °C in the dark. Kanamycin and phosphinotricin (BASTA®) resistance of the plants was tested on germination medium (1/2 MS salts; 10 g/l sucrose) plates with 40 mg/l kanamycin or 20 mg/l Glufosinate-ammonium under long day conditions.
Primers used for genotyping: Supplementary Table.
Total RNA samples for transcript measurements were extracted from seedlings grown for 1 week on GM medium under long day conditions and from rosette leaves of plants grown for 6–8 weeks under short day conditions. For analysis of RNA in imbibed seeds, seeds were soaked in water over night at RT. RNA was prepared in triplicates from 100 mg of seedlings grown on agarplates. Total RNA and small RNA were extracted from 1 g tissue with the mirVana™ miRNA Isolation kit (Ambion) according to manufacturers protocol E. and F. II. cDNA was prepared using RevertAid™ H Minus M-MuLVRT (Fermentas) according to manufacturers protocol. mRNA and AtSN1-A analysis was performed on oligo dT transcribed cDNA, AtSN1-B (and C) on cDNA using specific primers. cDNAs were quantified by the RT-qPCR method using the iCycler (Bio-rad) and the iQ™ SyBR® Green Supermix (Bio-rad). Program: 1: 5′ 95 °C; 2: 15″ 95 °C; 3: 30″ 65 °C; 4: 30″ 72 °C; 5: goto2 40×; 6: Melting curve 65 °C 10″ +0.5 °C 60 repeats; 7: 4 °C.
Primers used for qPCR are defined in the Supplementary Table.
ACTIN2 (ACT2; At3g18780) and PHOSPHOFRUCTOKINASE (PFK; At4g04040; transcripts only present in seedlings and leaf tissue) were used as reference genes (Correlation Coefficient and PCR efficiency: Suppl. Figure 8).
Northern blot analysis of siRNA/miRNA
Small RNA were detected by Northern blot analysis of a RNA fraction (10 μg per lane for miRNA 157a and 50 μg per lane for AtSN1 siRNA) enriched for small molecules. Hybridization was performed as described by Mette et al. (2005) using a 32P-labeled AtSN1 antisense transcript.
DNA methylation analysis (Bisulfite sequencing)
Genomic DNA for methylation analysis was extracted with a DNeasy Plant Maxi Kit (Qiagen, Hilden, Germany) from rosette leaves of individually genotyped plants grown for 6–8 weeks under short day conditions and from seedlings grown on GM medium for 1 week under long day conditions.
Bisulfite-mediated chemical conversion of DNA was done using a Qiagen EpiTect Bisulfite Kit (Qiagen, Hilden, Germany) following the manufacturers instructions. Amplicons were transformed with the help of the StrataClone™ PCR Cloning Kit. Sequencing of single clonal colonies was done with the Cyclereader Auto Sequencing Kit (Fermentas, St.Leon-Rot, Germany) and Licor 4300 DNA Analyzer.
Statistical analysis was performed using Chi-square test comparing the number of methylation events out of all possible cytosine sites detected by bisulfite sequencing. Standard deviation is calculated by QUICKBASIC Program for Exact and Mid-P Confidence Intervals for a Binomial Proportion (see Suppl. Fig 4a–h).
Primers used for amplification of converted target regions are defined in the Supplementary Table.
Chromatin immuno precipitation (ChIP)
The ChIP was performed according Jovtchev et al. (2011), Gendrel et al. (2005) and Lawrence and Pikaard (2004) with modifications. Four g of 1 week old seedlings were used for crosslinking by 1 % w/v formaldehyde solution under vacuum for 15 min (seedlings) or 10 min (leaves). Crosslinking was quenched with 2.5 ml 2 M glycine under vacuum for five min. The plant material was homogenized in liquid nitrogen and suspended in 30 ml extraction buffer. The extract was filtered through 2 layers of Miracloth and centrifuged at 1,800g for 20 min at 4 °C. The pellet was resuspended in 1 ml of extraction buffer 2 and centrifuged at 16,000g for 10 min. The pellet was resuspended in 300 μl extraction buffer 3 and overlayed on 300 μl of extraction buffer 3. The nuclei are collected by centrifugation at 16,000g for 1 h. The pellet was resuspended in 300 μl of cold nuclei lysis buffer. The sonication of the chromatin was performed using a Diagenode Disruptor for 6 cycles with 30 s of high energy sonication and 30 s break at 4 °C. Nuclear debris was removed by centrifugation at 16,000g for 10 min. One aliquot of the chromatin extract was used for the input and the others were diluted 1:10 in ChIP dilution buffer. The immunoprecipitation was performed with Dynabead Protein A (Dynal) coupled antibodies. H3: #39163 Histone H3 C-terminal rabbit pAB, Activemotif; H3K4me3: #39159 Histone H3 trimethyl Lys4 Rabbit pAB, Activemotif; H3K9me2: mouse monoclonal [mAbcam1220), Abcam; H3K27me3:; H4K20me1: ab9051, Rabbit polyclonal Histone H4 (monomethyl K20), Abcam were used.
After overnight incubation at 4 °C, the coupled beads were washed 2 times 5 min each with low salt buffer :150 mM NaCL, 0.1 % SDS, 1 % Triton X100, 2 mM EDTA, 20 mM Tris–HCl (pH 8.1); high salt buffer: 500 mM NaCL, 0.1 % SDS, 1 % Triton X-100, 2 mM EDTA, 20 mM Tris–HCl (pH 8.1); LiCl wash buffer: 0.25 LiCl, 1 % NP40, 1 % Na deoxycholate, 1 mM EDTA, 10 mM Tris–HCl (pH 8.1) and TE buffer: 10 mM Tris–HCl (pH 8.), 1 mM EDTA. The DNA was eluted twice by incubation with 250 μl elution buffer at 65 °C for 15 min.
Primers used for quantitative amplification of genomic regions after ChIP are defined in the supplementary table.
SUVH2 and SUVH9 are differentially expressed during A. thaliana development
Transcripts of SET-domain protein family genes SUVH2 (At2g33290) and SUVH9 (At4g13460) are detectable throughout the lifecycle of A. thaliana, but show different expression levels when material of different stages of plant development is compared (Suppl. Figure 1). In seeds imbibed in water for 1 day, SUVH2 expression was highest and decreased during seedling development and in mature leaves of 6 week old plants grown under short day conditions (Fig. 1). SUVH9 showed an expression pattern inverse to that of SUVH2, i.e., a low SUVH9 transcript abundance in seeds and a higher one in seedlings and leaves of mature plants. The observed changes in SUVH2 and SUVH9 expression could be due to differential gene regulation over developmental time or, provided that SUVH2 and SUVH9 are preferentially expressed in different cell types, might reflect the different cell type-composition of the analysed tissues.
Single suvh2 and suvh9 mutations affect RdDM to different extent at different developmental stages
To test the significance of this differential expression of SUVH2 and SUVH9, we applied a reverse genetic approach. T-DNA segregation analysis, sequencing of the PCR-amplified T-DNA integration locus (Suppl. Figure 2) and transcript quantification by RT-qPCR (Suppl. Figure 3) in T-DNA insertion lines identified the loss-of-function alleles suvh2 (GABI-Kat 516A07; Rosso et al. 2003) and suvh9 (SALK_048033; Alonso et al. 2003; Johnson et al. 2008) that both contained a single T-DNA insertion locus and showed robust specific reduction of SUVH2 and SUVH9 transcripts, respectively. A suvh2 suvh9 double mutant line was obtained by crossing of the single mutant lines followed by self-pollination of F1 and F2 plants. Homozygous suvh2 suvh9 plants were identified in the F3 generation using PCR-based markers selective for T-DNA insertion and wild type alleles. For all following analyses, material from homozygous F4 and F5 plants was used.
AtSN1 methylation was also determined in leaves from a line hemizygous for a Pro35S-mycSUVH2 over-expression construct (Naumann et al. 2005; Ay et al. 2009). Despite their dwarf habitus typical for increased SUVH2 transcript levels, Pro35S-mycSUVH2 plants did not show an obvious increase in AtSN1 methylation (Suppl. Figure 5). Thus, SUVH2 is required for RdDM of AtSN1, but is not the rate limiting factor if functional SUVH2 and SUVH9 alleles are present.
The GC-rich retroelement AtCOPIA4 (At4g16870) is hypermethylated in the A. thaliana genome, mainly due to maintenance methylation involving MET1 and SUVH4 (Johnson et al. 2007). AtCOPIA4 methylation is hardly affected in mutants of RdDM components. We performed bisulfite sequencing of an AtCOPIA4 element containing 83 cytosines (14 in CG, 6 in CHG and 63 in CHH context) (Fig. 3b, Suppl. Figure 4 I to P) in order to test for effects suvh2 and suvh9 mutations on its DNA methylation. No obvious changes of DNA methylation were found for all genotypes in seedlings and leaves from mature plants indicating that COPIA4 methylation is apparently not affected by mutants of the RdDM pathway.
Only the suvh2 suvh9 double mutant shows released silencing of AtSN1 sense transcription
To investigate the involvement of SUVH2 in silencing of Pol II dependent transcription, we used of a well characterized two component transgene system mediating RNA-directed transcriptional gene silencing (TGS; Aufsatz et al. 2002; Fischer et al. 2008). Although impaired reporter gene silencing was seen in the suvh2 mutant (Suppl. Figure 7), this observation turned out to be inconclusive due to the structure of the used suvh2 T-DNA insertion allele. Line GABI-Kat 516A07 contains two T-DNA copies in a tail-to-tail arrangement (Suppl. Figure 2). The resulting inverted repeat structure of CaMV Pro35S sequences leads to TGS of the silencer Pro35S-NOSpro-proNOS construct. Such cross-silencing effects were already reported for other T-DNA insertion lines (Daxinger et al. 2008).
Release of AtSN1 sense transcription in suvh2 suvh9 is correlated with altered histone modification
During plant reproduction, processes of genome-wide epigenetic reprogramming are taking place (Slotkin et al. 2009). This can include changes of histone modifications, DNA methylation and siRNA levels. Nevertheless, potentially harmful elements such as AtSN1 stay transcriptional silenced throughout development. Both SET domain proteins SUVH2 and SUVH9 are required for complete transcriptional silencing of AtSN1, acting in a seemingly redundant way. However, with regard to RNA-directed DNA methylation, SUVH2 and SUVH9 activities are at least in part developmentally non-redundant. This might be related to the different binding properties of the two proteins for methylated DNA (Johnson et al. 2008). Taking into account the reported in vitro binding preferences of the SUVH2 (methylated cytosines in CG context) and SUVH9 (methylated cytosines in CHH context) and the ratio of potentially methylated cytosines in these contexts in AtSN1 (4 sites in CG vs. 33 CHH context), one would predict that loss of SUVH9 should have a stronger impact than loss of SUVH2, irrespective of the developmental stage of sampled plants. The effects of the single suvh2 and suvh9 mutations on AtSN1 methylation levels observed in our study do not meet this expectation. Rather, from the observed reduction of DNA methylation in suvh2 seedlings and suvh9 mature leaves, we would conclude that the differential expression patterns of SUVH2 and SUVH9 can explain their non-redundant involvement in RdDM over development. The high expression of the SUVH2 in imbibed seeds indicates a possible role in the epigenetic reprogramming during early development. Stage-specific regulation of SUVH2 might explain the delayed senescence phenotype observed by Ay et al. (2009) due to ectopic SUVH2 expression during late development and the resulting mis-regulation of senescence controlling genes.
In A. thaliana interphase nuclei, most heterochromatin resides in structures called chromocenters, which are characterized by highly methylated DNA and specific marks such as histone H3K9me2 on the molecular level (Soppe et al. 2002). A clear reduction of histone H3K9me2 at chromocenters has been reported for example for a suvh4 suvh5 suvh6 triple mutant (Johnson et al. 2008; Jovtchev et al. 2011). For a suvh2 single mutant a reduction of H3K9me2 at chromocenters of young leaves has been reported (Naumann et al. 2005), while for the same allele in a suvh2 suvh9 double mutant no immunologically detectable change of H3K9me2 in chromocenters was reported in another study (Johnson et al. 2008). This might be explained by different availability of SUVH2 or SUVH9 at different developmental stages as well as by possible intensity changes of the histone marks themselves during development. While for example chromocenters show highly condensed H3K9me2 and H3K27me2 signals in nuclei of mature leaves, these signals become successively decondensed in nuclei of senescent leaves (Ay et al. 2009).
We thank Beate Kamm, Christa Fricke and Inge Glaser for excellent technical assistance, Armin Meister and Rhonda Meyer for support with statistical data analysis and Ingo Schubert for critically reading the manuscript. We are grateful to Gunter Reuter for contribution of transgenic A. thaliana line containing the Pro35S-mycSUVH2 construct and critical reading of the manuscript. This work was supported by German research foundation (DFG) collaborative research centre (SFB) 648 “Molecular Mechanisms of Information Processing in Plants”.
This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.
- Alonso JM, Stepanova AN, Leisse TJ, Kim CJ, Chen H, Shinn P, Stevenson DK, Zimmerman J, Barajas P, Cheuk R, Gadrinab C, Heller C, Jeske A, Koesema E, Meyers CC, Parker H, Prednis L, Ansari Y, Choy N, Deen H, Geralt M, Hazari N, Hom E, Karnes M, Mulholland C, Ndubaku R, Schmidt I, Guzman P, Guilar-Henonin L, Schmid M, Weigel D, Carter DE, Marchand T, Risseeuw E, Brogden D, Zeko A, Crosby WL, Berry CC, Ecker JR (2003) Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science 301:653–657PubMedCrossRefGoogle Scholar
- Aufsatz W, Mette MF, van der WJ, Matzke AJ, Matzke M (2002) RNA-directed DNA methylation in Arabidopsis. Proc Natl Acad Sci USA 99(Suppl 4):16499–16506Google Scholar
- Baumbusch LO, Thorstensen T, Krauss V, Fischer A, Naumann K, Assalkhou R, Schulz I, Reuter G, Aalen RB (2001) The Arabidopsis thaliana genome contains at least 29 active genes encoding SET domain proteins that can be assigned to four evolutionarily conserved classes. Nucleic Acids Res 29:4319–4333PubMedCrossRefGoogle Scholar
- Gao Z, Liu HL, Daxinger L, Pontes O, He X, Qian W, Lin H, Xie M, Lorkovic ZJ, Zhang S, Miki D, Zhan X, Pontier D, Lagrange T, Jin H, Matzke AJ, Matzke M, Pikaard CS, Zhu JK (2010) An RNA polymerase II- and AGO4-associated protein acts in RNA-directed DNA methylation. Nature 465:106–109PubMedCrossRefGoogle Scholar
- Johnson L, Cao X, Jacobsen S (2002) Interplay between two epigenetic marks. DNA methylation and histone H3 lysine 9 methylation. Curr Biol 12:1360–1367Google Scholar
- Jovtchev G, Borisova BE, Kuhlmann M, Fuchs J, Watanabe K, Schubert I, Mette MF (2011) Pairing of lacO tandem repeats in Arabidopsis thaliana nuclei requires the presence of hypermethylated, large arrays at two chromosomal positions, but does not depend on H3-lysine-9-dimethylation. Chromosoma 120:609–619PubMedCrossRefGoogle Scholar
- Law JA, Jacobsen SE (2010) Establishing, maintaining and modifying DNA methylation patterns in plants and animals. Nat Rev Genet 11:204–220Google Scholar
- Pontier D, Yahubyan G, Vega D, Bulski A, Saez-Vasquez J, Hakimi MA, Lerbs-Mache S, Colot V, Lagrange T (2005) Reinforcement of silencing at transposons and highly repeated sequences requires the concerted action of two distinct RNA polymerases IV in Arabidopsis. Genes Dev 19:2030–2040Google Scholar
- Soppe WJ, Jasencakova Z, Houben A, Kakutani T, Meister A, Huang MS, Jacobsen SE, Schubert I, Fransz PF (2002) DNA methylation controls histone H3 lysine 9 methylation and heterochromatin assembly in Arabidopsis. EMBO J 21:6549–6559Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 2.0 International License (https://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.