Regulatory Link Between DNA Methylation and Active Demethylation in Arabidopsis

Regulatory link between DNA methylation and active demethylation in Arabidopsis

Mingguang Leia, Huiming Zhanga,b, Russell Juliana, Kai Tanga, Shaojun Xiea,b, and Jian-Kang Zhua,b,1 aDepartment of Horticulture and Landscape Architecture, Purdue University, West Lafayette, IN 47906; and bShanghai Center for Plant Stress Biology, Shanghai Institutes for Biological Science, Chinese Academy of Sciences, Shanghai 200032, China

Contributed by Jian-Kang Zhu, February 4, 2015 (sent for review December 20, 2014; reviewed by Jerzy Paszkowski and Yijun Qi) De novo DNA methylation through the RNA-directed DNA meth- dynamic transcriptional regulation (12). ROS1 counteracts the ylation (RdDM) pathway and active DNA demethylation play RdDM pathway to prevent DNA hypermethylation (5, 12–14). important roles in controlling genome-wide DNA methylation Interestingly, ROS1 gene expression is suppressed in mutants patterns in plants. Little is known about how cells manage the defective in RdDM (14–17) or MET1 (16), suggesting that DNA balance between DNA methylation and active demethylation methylation and active demethylation activities are coordinated. activities. Here, we report the identification of a unique RdDM However, the mechanism underlying such coordination remains target sequence, where DNA methylation is required for main- elusive. Here, we report the identification of a regulatory ele- taining proper active DNA demethylation of the Arabidopsis ge- ment at the ROS1 promoter that monitors DNA methylation nome. In a genetic screen for cellular antisilencing factors, we levels and accordingly modulates active DNA demethylation isolated several REPRESSOR OF SILENCING 1 (ros1) mutant alleles, of the Arabidopsis genome by controlling ROS1 expression. In as well as many RdDM mutants, which showed drastically reduced a genetic screen for Arabidopsis mutants that are defective in ROS1 gene expression and, consequently, transcriptional silencing transcriptional antisilencing of transgenic reporter genes, we of two reporter genes. A helitron transposon element (TE) in the isolated several ros1 mutant alleles and many RdDM mutants ROS1 gene promoter negatively controls ROS1 expression, whereas with drastically repressed ROS1 gene expression. We found that DNA methylation of an RdDM target sequence between ROS1 5′ UTR a helitron TE in the ROS1 gene promoter negatively regulates and the promoter TE region antagonizes this helitron TE in regulating ROS1 expression, and that such negative regulation is antago-

ROS1 PLANT BIOLOGY expression. This RdDM target sequence is also targeted by nized by RdDM activities. RdDM of a sequence between ROS1 ros1 ROS1, and defective DNA demethylation in loss-of-function mu- 5′ UTR and the promoter TE positively correlates with ROS1 tant alleles causes DNA hypermethylation of this sequence and con- gene expression. In addition, defective active DNA demethyla- ROS1 comitantly causes increased expression. Our results suggest tion resulted in hypermethylation of this sequence, concomitant ROS1 that this sequence in the promoter region serves as a DNA with enhanced ROS1 gene expression. ChIP analysis of ROS1 methylation monitoring sequence (MEMS) that senses DNA meth- occupancy at this region confirmed the existence of coregulation ylation and active DNA demethylation activities. Therefore, the ROS1 between DNA methylation and active demethylation. These promoter functions like a thermostat (i.e., methylstat) to results provide important insights into how cells sense the bal- sense DNA methylation levels and regulates DNA methylation by ROS1 ance between DNA methylation and demethylation, and fine- controlling expression. tune active DNA demethylation accordingly.

DNA demethylase | ROS1 | methylstat | sensor | rheostat Significance NA methylation is a conserved epigenetic mark important DNA methylation is critical for transposon silencing and gene for development and stress responses in plants and many D regulation. DNA methylation levels are determined by the animals (1–4). Genome-wide DNA methylation patterns are combined activities of DNA methyltransferases and demethy- dynamically regulated by establishment, maintenance, and re- lases. This study found a 39-bp DNA methylation monitoring moval activities (4, 5). In plants, de novo DNA methylation is sequence (MEMS) in the promoter of the DNA demethylase controlled by the RNA-directed DNA methylation (RdDM) REPRESSOR OF SILENCING 1 (ROS1) gene of Arabidopsis plants. pathway, in which complementary pairing between long non- DNA methylation of MEMS is responsive to both RNA-directed coding RNAs and siRNAs mediates cytosine methylation in DNA methylation and ROS1-dependent active demethylation. a sequence-specific manner (2, 4, 6, 7). Best characterized in Thus, MEMS can sense DNA methylation and demethylation Arabidopsis, RdDM involves a complex array of regulators and activities, and it regulates genomic DNA methylation by adjusting mainly targets heterochromatic regions that are enriched with ROS1 expression levels. Our results suggest that the ROS1 pro- transposon elements (TEs) and other DNA repeat sequences moter, with the MEMS and an adjacent helitron transposon ele- (8, 9). Once established, Arabidopsis DNA methylation is main- ment, functions as a “methylstat” that senses and regulates tained via different mechanisms depending on the cytosine contexts genomic DNA methylation levels. [i.e., CG, CHG, CHH (H represents A, T, or C)]. CG and CHG methylation is maintained by MET1 and CMT3, respectively (2), Author contributions: M.L. and J.-K.Z. designed research; M.L., H.Z., and R.J. performed whereas CHH methylation within pericentromeric long TEs can research; M.L., H.Z., K.T., S.X., and J.-K.Z. analyzed data; and M.L., H.Z., and J.-K.Z. wrote be catalyzed by CMT2 and CHH methylation at other loci is the paper. established de novo by DRM2 during every cell cycle (2, 10). In Reviewers: J.P., The Sainsbury Laboratory, University of Cambridge; and Y.Q., Tsinghua contrast to DNA methyltransferases that establish and/or maintain University. cytosine methylation, plant 5-methylcytosine DNA glycosylases The authors declare no conflict of interest. initiate a base excision repair pathway that erases DNA methyla- Data deposition: The high-throughput sequencing data generated in this study have been deposited in the Gene Expression Omnibus (GEO) database, www.ncbi.nlm.nih.gov/geo tion, thereby generating, together with the establishment and (accession nos. GSE53033, GSE64499, and GSE58790). An additional dataset used in this maintenance activities, a dynamic landscape of DNA meth- study is GEO accession no. GSE44209. ylation (11, 12). 1To whom correspondence should be addressed. Email: [email protected]. The Arabidopsis REPRESSOR OF SILENCING 1 (ROS1) is This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. a major DNA demethylase that prunes DNA methylation for 1073/pnas.1502279112/-/DCSupplemental. www.pnas.org/cgi/doi/10.1073/pnas.1502279112 PNAS | March 17, 2015 | vol. 112 | no. 11 | 3553–3557 Results and Discussion A Mutations in ROS1 and Its Regulator Increased DNA Methylation 1 Cause the Silencing of Cauliflower Mosaic Virus 35S Promoter:: Sucrose Transporter 2 and 2 × Cauliflower Mosaic Virus 35S 35S SUC2 2x35S HPT II Promoter::Hygromycin Phosphotransferase II Transgenes. We pre- viously reported an efficient transgene-based genetic screen for B cellular antisilencing factors in Arabidopsis (18, 19). In this system, a sucrose transporter 2 (SUC2) gene was driven by the Col-0 WT constitutive cauliflower mosaic virus 35S promoter (35S) and hygromycin phosphotransferase II (HPTII) gene was driven by the 2 × 35S promoter in Arabidopsis (Fig. 1A). When grown on sucrose-containing nutrient media, the transgenic plants (there- after referred as WT) absorb too much sucrose from the medium, resulting in root growth inhibition (20). In this genetic screen, we obtained seven new alleles of ROS1 (Fig. 1B and Fig. S1). Quantitative RT-PCR analysis showed that the transcript levels of SUC2 and HPTII were drastically reduced in ros1 mutant plants (Fig. 1 C and D). Bisulfite sequencing showed that the 35S promoter had higher levels of DNA methylation in ros1 than in the WT, especially in region A, although the promoter was already methylated in the WT (Fig. 1E). We also recovered eight C SUC2 D HPTII new alleles of increased DNA methylation 1 (IDM1) (Fig. 1B 1.2 1.2 and Fig. S1), a gene necessary for ROS1 function in active DNA 1 1 demethylation, by creating a favorable chromatin environment (21). The results show that active DNA demethylation is re- 0.8 0.8 quired for preventing transcriptional silencing of the reporter 0.6 0.6 genes. The large number of new ros1 and idm1 mutant alleles 0.4 0.4 provides valuable information regarding structure–function rela- 0.2 0.2 tionships for these two important epigenetic regulators. 0 0

35S::SUC2 2 × 35S::HPTII WT RdDM Mutations Cause the Silencing of and WT

Transgenes. Our genetic screen also recovered many RdDM Relative transcript level idm1-9 idm1-9 ros1-14 ros1-15 ros1-12 ros1-13 ros1-11 ros1-11 ros1-12 ros1-14 ros1-15 ros1-13 nrpd1-12 mutants, including six new alleles of nrpd1,2nrpe1,2nrpd2,5rdr2, nrpd1-12 5 dcl3,1dtf1,1ago4,1drm2,2drd1,4ktf1,2idn1,and3idn2 (Fig. E 35S promoter 2A and Figs. S1 and S2). In the RdDM pathway mutants that were tested, the SUC2 and HPTII transgenes were silenced (Fig. 2 B region A region B and C and Fig. S3A). Bisulfite sequencing showed that the 35S promoter is hypermethylated in the nrpd1-12 mutant compared WT with WT (Fig. 1E). Although the RdDM pathway is known to be required for epigenetic silencing of some transgenes and numerous ros1-13 endogenous loci, including many transposon repeats (22–25), our forward genetic screen here identified the RdDM as an anti- nrpd1-12 silencing pathway. The notion of the RdDM pathway also having an antisilencing role is consistent with a previous report (17). Fig. 1. Mutations in ROS1 resulted in transcriptional silencing of marker RdDM Maintains ROS1 Expression to Prevent the Silencing of 35S:: transgenes. (A) Schematic diagram of the construct harboring the 35S::SUC2 and × SUC2 and 2 × 35S::HPTII Transgenes. Because the ros1 and RdDM 2 35S::HPTII cassettes that were transformed in the parental line. Black rec- mutants were recovered from the same genetic screen, we hy- tangles represent exons. (B) Comparison of root growth of Col-0, WT, and seven ros1 mutants. Expression of SUC2 (C)andHPTII (D) transgenes in ros1 mutants. pothesized that RdDM may prevent the silencing of the 35S::SUC2 ± × Values are mean SD of three biological replicates and are normalized to and 2 35S::HPTII transgenes by maintaining ROS1 expression. transcript levels in WT. (E) DNA methylation status at the 35S promoter in WT, We examined ROS1 gene expression, and found that ROS1 was ros1-13,andnrpd1-12 mutants. The result from whole-genome bisulfite se- dramatically repressed in all of the tested RdDM mutants (Fig. 2D quencing is shown. Vertical bars on each track indicate DNA methylation levels. and Fig. S3B), which is consistent with previous reports (14, 15, 17). To test whether the SUC2 and HPTII silencing in the RdDM mutants is caused by down-regulation of ROS1, we tried to restore and consequently suppressed the silencing of SUC2 and HPTII ROS1 mRNA levels in nrpe1-17 plants by transgenic expression of (Fig. 3 A and B). The results confirmed that the antisilencing role ROS1 with a 2-kb native promoter. The attempt failed, as revealed of RdDM is indeed because it is required for ROS1 expression. The by the long root phenotype of the ROS1 promoter (PROS1)::ROS1/ results also suggest that the helitron TE negatively regulates ROS1 nrpe1-17 plants on sucrose-containing medium (Fig. 3A). Quanti- gene expression, and that RdDM-dependent methylation is re- tative RT-PCR results showed that the RNA levels of SUC2, quired for ROS1 expression only when this TE is present in ROS1 HPTII,andROS1 in these transgenic plants remained low (Fig. 3B). promoter. We hypothesized that an unknown RdDM target se- As a control, the PROS1::ROS1 transgene could rescue the root quence at the ROS1 locus functions to antagonize the TE to ensure phenotype of ros1-12 mutant plants (Fig. S4). These results show ROS1 expression. that like the endogenous ROS1,theROS1 transgene requires RdDM activity for expression. ROS1 gene promoter contains RdDM Regulates the DNA Methylation Level of DNA Methylation a helitron TE, which is located 155 bp upstream of the 5′ UTR of Monitoring Sequence in the ROS1 Promoter. To investigate how ROS1. Deletion of this helitron TE from the ROS1 promoter en- RdDM activity may antagonize the helitron TE in regulating abled transgenic expression of PROS1-ΔTE::ROS1 in RdDM mutants, ROS1 gene expression, we examined DNA methylation patterns

3554 | www.pnas.org/cgi/doi/10.1073/pnas.1502279112 Lei et al. A and active DNA demethylation through transcriptional regulation of ROS1. Accordingly, we referred to this sequence as the DNA Col-0 WT methylation monitoring sequence (MEMS) in subsequent analyses. DNA Methylation Level of MEMS Is also Regulated by Active DNA Demethylation. We found that DNA methylation at MEMS is also regulated by ROS1-dependent active demethylation, as shown by the increased DNA methylation levels when ROS1 is dysfunctional (Fig. 4A). In ros1-12 plants compared with WT plants, the levels of DNA methylation in CG, CHG, and CHH contexts were increased by 287%, 205%, and 196%, respectively (Fig. 4A). Similarly, DNA hypermethylation was observed in ros1-4 in comparison to its Col-0 WT control (Fig. S5A), further B C suggesting that MEMS is targeted by ROS1-dependent DNA 1.2 SUC2 1.2 HPTII demethylation. To confirm the regulation of MEMS by ROS1- 1.0 1.0 dependent DNA demethylation, chromatin occupancy of ROS1 0.8 0.8 at MEMS was examined by ChIP. We used ros1-4 plants com- plemented with transgenic ROS1 that was driven by its native 0.6 0.6 promoter and was epitope-tagged with 3× FLAG. As shown in 0.4 0.4 Fig. 4B, anti-FLAG antibody detected ROS1 signals at MEMS 0.2 0.2 and its neighboring regions, consistent with the local DNA 0 0 hypermethylation in ros1 mutants (Fig. 4A and Fig. S5A).

Relative transcript level Therefore, DNA methylation at MEMS is controlled by both RdDM and ROS1-dependent active demethylation. D We next asked whether DNA hypermethylation at MEMS might be accompanied by transcriptional elevation of ROS1.Fivepoint- ROS1 1.2 mutation alleles of ros1 mutants, including ros1-11 to ros1-15, were 1.0 examined. Quantitative RT-PCR results showed that ROS1 gene PLANT BIOLOGY 0.8 0.6 0.4 A 0.2 0 Relative transcript level

Col-0 WT Fig. 2. Dysfunctional RdDM caused transcriptional silencing of marker transgenes. (A) Comparison of root growth in Col-0, WT, and RdDM mutants, including nrpd1, nrpe1, nrpd2, rdr2, drd1,andktf1; ros1 and idm1 were included glucose as controls. Expression of SUC2 (B)andHPTII (C) transgenes and ROS1 (D)inthe RdDM mutants is shown. Values are mean ± SD of three biological replicates and are normalized to transcript levels in WT. ktf1, Kow domain-containing transcription factor 1.

sucrose in the ROS1 promoter. We performed individual bisulfite sequencing to compare DNA methylation levels in the region between the 5′ UTR and the helitron TE in WT, nrpd1-12, and B nrpe1-17. NRPD1 and NRPE1 encode the largest subunits of Pol SUC2 HPTII ROS1 IV and Pol V, respectively. Dysfunction of Pol IV and Pol V, 1.2 1.5 1.2 both of which are major components of the canonical RdDM 1 1.2 1 0.8 pathway (2, 6, 7, 26), substantially reduced DNA methylation 0.9 0.8 0.6 0.6 levels at the examined region, especially in CHG and CHH 0.6 contexts (Fig. 4A). Similar patterns were observed in nrpd1-3 and 0.4 0.4 0.2 0.3 0.2 nrpe1-11 plants compared with their WT control Columbia-0 0 0 0 (Col-0), as revealed in whole-genome bisulfite sequencing

(Fig. S5A). In the methylated ROS1 promoter region, Pol IV- Relative transcript level dependent and Pol V-dependent 24-nt siRNAs were detected by small RNA sequencing (27) (Fig. S5B). Consistently, Pol V could bind to this region, as shown by published NRPE1 ChIP sequencing data (28) (Fig. S5C). These results demonstrated that RdDM directly controls DNA methylation in this particular sequence of ROS1 promoter. Because the DNA hypomethylation Fig. 3. Antisilencing defect in RdDM mutants is rescued by ROS1 transgene driven by its promoter with TE deletion. (A) Root phenotype of nrpd1 and nrpe1 in this particular sequence in the RdDM mutants is concom- mutants was rescued by ROS1 gene driven by its promoter with deletion of the itant with ROS1 gene repression, we hypothesized that this TE (PROS1-ΔTE::ROS1). (B) Expression of SUC2 and HPTII transgenes in nrpd1 and sequence may serve as an indicator of general RdDM activities nrpe1 mutants transformed with PROS1-ΔTE::ROS1.Valuesaremean± SD of three to allow a dynamic coordination between DNA methylation biological replicates and are normalized to transcript levels in WT.

Lei et al. PNAS | March 17, 2015 | vol. 112 | no. 11 | 3555 TSS In the nrpd1 mutants, CG hypermethylation was observed at the A ATG ′ TE 3 endofthehelitronTEthatisadjacenttoMEMS(Fig. S5A). It is unlikely that ROS1 gene repression in nrpd1 is a consequence of this hypermethylation, because a greater level of CG hypermethylation CG CHG WT inthesameregionexistsinros1 mutants (Fig. S5A), which displayed WT CHH nrpd1-12 increased ROS1 gene expression. This conclusion is also supported 100 MEMS region nrpe1-17 80 ros1-13 by the observation that ROS1 expression is suppressed in met1 nrpd1-12 60 mutant plants (16), in which CG methylation is abolished. Recently, 40 it was shown that Pol V can be recruited by preexisting CG meth- nrpe1-17 20 ylation at RdDM target loci (31). Therefore, dysfunction of MET1 0 may impair RdDM at MEMS, and thereby decrease methylation- ros1-13 Methylation level (%) level Methylation CG CHG CHH total C dependent ROS1 gene expression. Indeed, DNA methylation at MEMS (-40 to -2) MEMS is lost in met1-3 mutant plants (32) (Fig. S7). B The DNA methylation at MEMS and its adjacent region in the TSS ATG ROS1 promoter tends to be variable (Figs. S5A and S7), although TE our results show that the methylation at MEMS decreases in nrpd1 1 2 and nrpe1 mutants and increases in ros1 mutants. The reason for Region 1 Region 2 0.2 0.2 this high variability is presently unclear but may be a reflection of the dynamic regulation of DNA methylation at the ROS1 promoter 0.16 0.16 in plants. It is also unclear how the helitron TE negatively regulates 0.12 0.12 ROS1 gene expression. One possibility is that the TE in ROS1 0.08 0.08

input (‰) promoter causes a chromatin conformation that is unfavorable for 0.04 0.04

Relative value to transcription initiation. DNA methylation at the adjacent MEMS 0 0 Col-0 line 1 line 2 Col-0 line 1 line 2 may help attract a transcriptional activator to overcome the transcriptional suppression by the TE. ROS1-3xFlag/ros1-4 C ROS1-3xFlag/ros1-4 Like a thermostat that senses temperature changes and maintains ROS1 7 a stable temperature, a methylstat may exist in plant cells to 6 monitor DNA methylation levels and accordingly adjust DNA 5 4 methylation and/or active demethylation activities to maintain 3 genomic DNA methylation patterns. Our results suggest that 2 the ROS1 promoter functions as a methylstat that senses both 1 0 DNA methylation and demethylation activities and accordingly fine-tunes ROS1 expression levels. In met1-3 plants, CG hypo- WT Relative transcript level

idm1-9 methylation at 5S ribosomal DNA is followed by progressive ros1-14 ros1-15 ros1-11 ros1-12 ros1-13 nrpd1-12 increases in CHH methylation levels in successive generations Fig. 4. DNA methylation at MEMS is coregulated by RdDM and active as a result of RdDM and lack of ROS1 expression, resulting in demethylation. (A) Analysis of DNA methylation levels at the 5′ region of re-establishment of transcriptional silencing (16). In humans, the ROS1 gene in nrpd1, nrpe1, and ros1 mutants by genomic bisulfite se- genome-wide DNA methylation landscape is dynamically fine- quencing. Methylation levels of boxed regions in each mutant were calcu- tuned, as shown by a strong correlation between biological age lated and are shown in the histograms. Black rectangles represent exons. The and DNA methylation increases at some loci and decreases at empty box represents the UTR. TSS, transcription start site. (B) ROS1 protein × other loci (33). It is possible that a similar methylstat may exist in enrichment at ROS1 promoter. ChIP against ROS1-3 FLAG was performed in mammals to regulate DNA methylation. Col-0 and two lines of ROS1-3×FLAG/ros1-4. All error bars indicate SD (n = 3). (C) Enhancement of ROS1 transcript levels in ros1 mutants. Materials and Methods Plant Materials. All mutant and transgenic Arabidopsis plants used in this expression was increased in four of the five ros1 mutant alleles, with study were in the Col-0 genetic background. Seeds were surface-sterilized in 20% the exception of ros1-12, which contains a C-to-T mutation, and thus (vol/vol) bleach, rinsed with sterilized water, and sown on half-strength Murashige and Skoog (MS) medium plates with 1% (wt/vol) agar. The root phenotype was a premature stop codon in the fourth exon (Fig. 4C and Fig. S1). observed when 2% (wt/vol) sucrose was added to the medium. When it was RNA decay can be triggered by mutations that cause premature stop necessary to avoid the side effects of sucrose accumulation and to monitor other codons (29) and may account for a lack of increased RNA levels of phenotypes, sucrose was replaced with 1% glucose. After 2 d of stratification at − − ROS1 in ros1-12. The increased ROS1 expression in all other ex- 4 °C, plates were moved to a growth chamber at 22 °C with 100 μmol·m 2·s 1 amined ros1 alleles indicates a positive correlation between MEMS fluorescent light (16 h of light and 8 h of dark). DNA hypermethylation and ROS1 gene up-regulation, further supporting a role of MEMS in sensing DNA methylation and Screening for Ethyl Methanesulfonate-Mutagenized Mutant Population. M2 demethylation activities and regulating ROS1 expression. seeds from ethyl methanesulfonate-mutagenized plants were grown vertically on half-strength MS medium with 2% (wt/vol) sucrose for 1 wk. In this con- DNA methylation in gene promoters is commonly considered as dition, root growth of WT plants was severely inhibited, whereas putative transcriptionally repressive. However, DNA methylation of MEMS mutants with long roots were selected and transferred to soil. To map the has a positive role in regulating ROS1 gene expression. Unlike mutated genes, mutants were crossed to Landsberg erecta ecotype WT promoter methylation, gene body methylation is preferentially as- plants. We selected a mapping population of 96 F2 seedlings that showed sociated with transcribed genes (8, 30). DNA methylation in the long roots when grown on half-strength MS medium with 2% (wt/vol) ROS1 gene body is unaffected in nrpd1-3 and nrpe1-11 mutants (Fig. sucrose. In this medium, hygromycin (20 μg/L) was added to exclude the S6), indicating that RdDM regulation of ROS1 gene expression is seedlings that lack the 35S::SUC2 transgene. After rough mapping with bulk segregant analysis, the mutant genome was sequenced to find the independent of DNA methylation in the ROS1 gene body. Mu- mutated genes. tation of ROS1 causes DNA hypermethylation in nearly 5,000 genomic regions, many of which are not RdDM targets (21). Plasmid Construction and Mutant Complementation. For complementation, Thus, ROS1 down-regulation by defective RdDM may cause ROS1 genomic DNA with an ∼2-kb promoter region was amplified from DNA hypermethylation in these non–RdDM-targeted regions. Col-0 genomic DNA and cloned into the pENTR/D-TOPO entry vector

3556 | www.pnas.org/cgi/doi/10.1073/pnas.1502279112 Lei et al. (Invitrogen). To delete the sequence corresponding to the helitron transposon, Real-Time RT-PCR. For real-time RT-PCR, total RNAs were extracted with the ROS1 entry clone was amplified using the primers listed in Table S1. After it was RNeasy Plant Kit (Qiagen). From 1 μg of total RNA, cDNAs were generated using sequenced, the genomic DNA was recombined with attR sites into the poly (T) primer and M-MuLV Reverse Transcriptase (New England Biolabs) in pGWB16 binary vector (34) using LR Clonase (Invitrogen). Agrobacterium a20-μL reaction mixture. The resulting cDNAs were used as templates for real- strain GV3101 carrying various constructs was used to transform the ros1-13, time PCR with iQ SYBR green supermix (Bio-RAD). PCR was performed with nrpd1-12,andnrpe1-17 mutants by the standard floral dip method (35). a CFX96 real-time PCR detection system (Bio-RAD). TUB8 wasusedasaninternal control. All of the primers used are listed in Table S1. Bisulfite Sequencing of Individual Loci. Two hundred nanograms of genomic DNA was treated with the BisulFlash DNA Modification Kit (Epigentek) fol- ChIP Assay. ChIP assays were performed as described by Wierzbicki et al. (28). lowing the manufacturer’s protocols. Two microliters of bisulfite-treated The antibodies used were anti-FLAG (no. F1804; Sigma). ChIP products were DNA was used for each PCR assay (20-μL final volume) using ExTaq (Takara) diluted with 80 μLofTEbuffer,and2μL was used for each quantitative PCR with primers specific to the corresponding regions (Table S1). PCR products reaction. At least two biological replicates were performed. were cloned into the pGEM-T easy vector (Promega) following the supplier’s instructions. For each region in each sample, at least 16 independent top- ACKNOWLEDGMENTS. This work was supported by NIH Grants R01GM070795 strand clones were sequenced and analyzed. and R01GM059138 (to J.-K.Z.) and the Chinese Academy of Sciences.

1. Reik W, Dean W, Walter J (2001) Epigenetic reprogramming in mammalian de- 19. Wang X, et al. (2013) RNA-binding protein regulates plant DNA methylation by velopment. Science 293(5532):1089–1093. controlling mRNA processing at the intronic heterochromatin-containing gene IBM1. 2. Law JA, Jacobsen SE (2010) Establishing, maintaining and modifying DNA methylation Proc Natl Acad Sci USA 110(38):15467–15472. patterns in plants and animals. Nat Rev Genet 11(3):204–220. 20. Lei M, et al. (2011) Genetic and genomic evidence that sucrose is a global regulator of 3. Wu SC, Zhang Y (2010) Active DNA demethylation: Many roads lead to Rome. Nat Rev plant responses to phosphate starvation in Arabidopsis. Plant Physiol 156(3):1116–1130. Mol Cell Biol 11(9):607–620. 21. Qian W, et al. (2012) A histone acetyltransferase regulates active DNA demethylation 4. Zhang H, Zhu JK (2012) Seeing the forest for the trees: A wide perspective on RNA- in Arabidopsis. Science 336(6087):1445–1448. directed DNA methylation. Genes Dev 26(16):1769–1773. 22. Eun C, et al. (2011) AGO6 functions in RNA-mediated transcriptional gene silencing in 5. Zhu JK (2009) Active DNA demethylation mediated by DNA glycosylases. Annu Rev shoot and root meristems in Arabidopsis thaliana. PLoS ONE 6(10):e25730. Genet 43:143–166. 23. Greenberg MV, et al. (2011) Identification of genes required for de novo DNA – 6. Pikaard CS, Haag JR, Pontes OM, Blevins T, Cocklin R (2012) A transcription fork model methylation in Arabidopsis. Epigenetics 6(3):344 354. for Pol IV and Pol V-dependent RNA-directed DNA methylation. Cold Spring Harb 24. Won SY, et al. (2012) Development of a luciferase-based reporter of transcriptional Symp Quant Biol 77:205–212. gene silencing that enables bidirectional mutant screening in Arabidopsis thaliana. 7. Matzke MA, Mosher RA (2014) RNA-directed DNA methylation: An epigenetic path- Silence 3(1):6. 25. Zheng Z, et al. (2010) An SGS3-like protein functions in RNA-directed DNA methyla- way of increasing complexity. Nat Rev Genet 15(6):394–408. – tion and transcriptional gene silencing in Arabidopsis. Plant J 62(1):92 99. PLANT BIOLOGY 8. Zhang X, et al. (2006) Genome-wide high-resolution mapping and functional analysis 26. Zhang H, Zhu JK (2011) RNA-directed DNA methylation. Curr Opin Plant Biol 14(2): of DNA methylation in arabidopsis. Cell 126(6):1189–1201. 142–147. 9. Henderson IR, Jacobsen SE (2007) Epigenetic inheritance in plants. Nature 447(7143): 27. Zhang H, et al. (2013) DTF1 is a core component of RNA-directed DNA methylation 418–424. and may assist in the recruitment of Pol IV. Proc Natl Acad Sci USA 110(20):8290–8295. 10. Zemach A, et al. (2013) The Arabidopsis nucleosome remodeler DDM1 allows DNA 28. Wierzbicki AT, et al. (2012) Spatial and functional relationships among Pol V-associ- methyltransferases to access H1-containing heterochromatin. Cell 153(1):193–205. ated loci, Pol IV-dependent siRNAs, and cytosine methylation in the Arabidopsis 11. Gehring M, et al. (2006) DEMETER DNA glycosylase establishes MEDEA polycomb epigenome. Genes Dev 26(16):1825–1836. gene self-imprinting by allele-specific demethylation. Cell 124(3):495–506. 29. Chang YF, Imam JS, Wilkinson MF (2007) The nonsense-mediated decay RNA sur- 12. Gong Z, et al. (2002) ROS1, a repressor of transcriptional gene silencing in Arabi- veillance pathway. Annu Rev Biochem 76:51–74. dopsis, encodes a DNA glycosylase/lyase. Cell 111(6):803–814. 30. Zilberman D, Gehring M, Tran RK, Ballinger T, Henikoff S (2007) Genome-wide 13. He XJ, et al. (2009) A conserved transcriptional regulator is required for RNA-directed analysis of Arabidopsis thaliana DNA methylation uncovers an interdependence be- – DNA methylation and plant development. Genes Dev 23(23):2717 2722. tween methylation and transcription. Nat Genet 39(1):61–69. 14. Gao Z, et al. (2010) An RNA polymerase II- and AGO4-associated protein acts in RNA- 31. Johnson LM, et al. (2014) SRA- and SET-domain-containing proteins link RNA poly- – directed DNA methylation. Nature 465(7294):106 109. merase V occupancy to DNA methylation. Nature 507(7490):124–128. 15. Huettel B, et al. (2006) Endogenous targets of RNA-directed DNA methylation and Pol 32. Stroud H, Greenberg MV, Feng S, Bernatavichute YV, Jacobsen SE (2013) Compre- – IV in Arabidopsis. EMBO J 25(12):2828 2836. hensive analysis of silencing mutants reveals complex regulation of the Arabidopsis 16. Mathieu O, Reinders J, Caikovski M, Smathajitt C, Paszkowski J (2007) Transgenera- methylome. Cell 152(1-2):352–364. tional stability of the Arabidopsis epigenome is coordinated by CG methylation. Cell 33. Hannum G, et al. (2013) Genome-wide methylation profiles reveal quantitative views 130(5):851–862. of human aging rates. Mol Cell 49(2):359–367. 17. Li X, et al. (2012) Antisilencing role of the RNA-directed DNA methylation pathway 34. Nakagawa T, et al. (2007) Development of series of gateway binary vectors, pGWBs, and a histone acetyltransferase in Arabidopsis. Proc Natl Acad Sci USA 109(28): for realizing efficient construction of fusion genes for plant transformation. J Biosci 11425–11430. Bioeng 104(1):34–41. 18. Lei M, et al. (2014) Arabidopsis EDM2 promotes IBM1 distal polyadenylation and 35. Clough SJ, Bent AF (1998) Floral dip: A simplified method for Agrobacterium- regulates genome DNA methylation patterns. Proc Natl Acad Sci USA 111(1):527–532. mediated transformation of Arabidopsis thaliana. Plant J 16(6):735–743.

Lei et al. PNAS | March 17, 2015 | vol. 112 | no. 11 | 3557 Supporting Information

Lei et al. 10.1073/pnas.1502279112

ROS1

IDM1

NRPD1

NRPE1

NRPD2

RDR2

DCL3

Fig. S1. Diagram of mutations identified in this study. The positions of mutations are counted from the first ATG in the genomic sequence. The changed amino acid is indicated in parentheses. The asterisk represents a stop codon.

Lei et al. www.pnas.org/cgi/content/short/1502279112 1of5 AGO4

DRM2

KTF1

DRD1

IDN1

IDN2

DTF1

Fig. S2. Diagram of mutations identified in this study. The positions of mutations are counted from the first ATG in the genomic sequence. The changed amino acid is indicated in parentheses. The asterisk represents a stop codon. AGO4, Argonaute 4; DTF1, DNA-binding transcription factor 1; KTF1, Kow domain- containing transcription factor 1.

Lei et al. www.pnas.org/cgi/content/short/1502279112 2of5 A 1.2 SUC2

1.0

0.8

0.6

0.4

0.2 Relative transcript level

0.0

B ROS1 1.2

1.0

0.8

0.6

0.4

0.2 Relative transcript level

0.0

Fig. S3. Dysfunctional RdDM caused transcriptional silencing of ROS1 and SUC2 transgenes. Expression of SUC2 (A) and ROS1 (B) in RdDM mutants is shown. Values are mean ± SD of three biological replicates, where the fold changes are normalized to transcript levels in WT. dcl3, dicer-like 3.

Col-0 WT ros1-13

glucose

sucrose

Fig. S4. Root phenotype of ros1-13 mutant was rescued by ROS1 gene driven by both its native promoter and the promoter with TE deletion.

Lei et al. www.pnas.org/cgi/content/short/1502279112 3of5 A TSS TE ROS1

Col-0_rep1

Col-0_rep2

nrpd1-3_rep1

nrpd1-3_rep2

nrpe1-11_rep1

nrpe1-11_rep2

ros1-4_rep1

ros1-4_rep2

B Col-0 nrpd1-3 nrpe1-11

C Col-0 NRPE1 ChIP-seq

nrpe1-11 NRPE1 ChIP-seq

Fig. S5. DNA methylation status and NRPE1 enrichment at ROS1 promoter. (A) DNA methylation status of ROS1 promoter in Col-0, nrpd1-3, nrpe1-11,and ros1-4 mutants. The result from whole-genome bisulfite sequencing is shown. Vertical bars on each track indicate DNA methylation levels. TSS, transcription start site. (B) Small RNA levels at ROS1 promoter in Col-0, nrpd1-3, and nrpe1-11 (1). (C) NRPE1 enrichment at ROS1 promoter. ChIP sequencing data (2) are from neomorph.salk.edu/pol_epigenomes/browser.html.

1. Zhang H, et al. (2013) DTF1 is a core component of RNA-directed DNA methylation and may assist in the recruitment of Pol IV. Proc Natl Acad Sci USA 110(20):8290–8295. 2. Wierzbicki AT, et al. (2012) Spatial and functional relationships among Pol V-associated loci, Pol IV-dependent siRNAs, and cytosine methylation in the Arabidopsis epigenome. Genes Dev 26(16):1825–1836.

Col-0_rep1

Col-0_rep2

nrpd1-3_rep1

nrpd1-3_rep2

nrpe1-11_rep1

nrpe1-11_rep2

Fig. S6. DNA methylation status at ROS1 gene in Col-0, nrpd1-3, and nrpe1-11 mutants. The result from whole-genome bisulfite sequencing is shown. Vertical bars on each track indicate DNA methylation levels.

Lei et al. www.pnas.org/cgi/content/short/1502279112 4of5 TSS

TE ROS1

Col-0 rep1

mCG Col-0 rep2

met1-3

Col-0 rep1

mCHG Col-0 rep2

met1-3

Col-0 rep1

mCHH Col-0 rep2

met1-3

Fig. S7. DNA methylation status at ROS1 promoter in Col-0 and met1-3 mutant. The result from whole-genome bisulfite sequencing (1) is shown. Vertical bars on each track indicate DNA methylation levels. TSS, transcription start site.

1. Johnson LM, et al. (2014) SRA- and SET-domain-containing proteins link RNA polymerase V occupancy to DNA methylation. Nature 507(7490):124–128.

Table S1. List of primers used in this study Primer name Sequences (5′→3′) Purpose

pEN-pROS1 CACCGCTTCACTTTGCTTGGCACA Cloning ROS1 genomic DNA into pENTR vector ROS1-r GGCGAGGTTAGCTTGTTGTC Cloning ROS1 genomic DNA into pENTR vector ROS1prodelF CATATCAGTGGGACGATGATGACGATGACGATA Deletion of TE in ROS1 promoter ROS1prodelR TCATCATCGTCCCACTGATATGGGAGCTAGTG Deletion of TE in ROS1 promoter ROS1proBiF TTTGATGATAAATTTATAAATAAAGTT Bisulfite sequencing of ROS1 promoter ROS1proBiR ATCACTAATACTTCATTTCTTCTCTT Bisulfite sequencing of ROS1 promoter qSUC2F TGCCTCAAGGCTCTCTTAACTT qPCR of SUC2 qSUC2R TGGAGAGGTATTTATAGGAAGAGAAGG qPCR of SUC2 qHPTII-f GGGATTCCCAATACGAGGTC qPCR of HPTII qHPTII-r GCTCCATACAAGCCAACCAC qPCR of HPTII qROS1F GGACAGAACACCGAGTTTACG qPCR of ROS1 qROS1R TCATCAGGTTCTCTCTTTTCCAA qPCR of ROS1 qTUB8F AGCTGAGAATTGTGACTGCTTG qPCR of TUB8 qTUB8R TCAACAACGTTCCCATTCC qPCR of TUB8 qGUS-f TCTACTTTACTGGCTTTGGTCG qPCR of GUS qGUS-r CGTAAGGGTAATGCGAGGTAC qPCR of GUS qACT2-r ATGACTCAGATCATGTTTGAGACC qPCR of ACT2 qACT2-r TCAGTAAGGTCACGACCAGCAA qPCR of ACT2 pROS1ChIPf1 CTGCATATGTAAACGAAAACGTGT ChIP-qPCR of ROS1 promoter pROS1ChIPr1 ACAAATAACATGTTATGTAATCCACGTT ChIP-qPCR of ROS1 promoter pROS1ChIPf2 TCTTTCCCATAGTCTAGGAGATTTTG ChIP-qPCR of ROS1 promoter pROS1ChIPr2 TTTGTTTATGTAGGGCGAAAGTTC ChIP-qPCR of ROS1 promoter

ACT2, Actin 2; f or F, forward; GUS, β-glucuronidase; qPCR, quantitative PCR; r or R, reverse; TUB8, Tublin 8.

Lei et al. www.pnas.org/cgi/content/short/1502279112 5of5