lable at ScienceDirect

Plant Physiology and Biochemistry 87 (2015) 102e108

Contents lists avai

Plant Physiology and Biochemistry

journal homepage: www.elsevier .com/locate/plaphy

Research article

High-temperature effect on genes engaged in DNA methylation andaffected by DNA methylation in Arabidopsis

Mladen Naydenov a, Vesselin Baev a, Elena Apostolova a, Nadezhda Gospodinova a,Gaurav Sablok b, Mariyana Gozmanova a, Galina Yahubyan a, *

a Department of Plant Physiology and Molecular Biology, University of Plovdiv, 24 Tsar Assen St, 4000 Plovdiv, Bulgariab Department of Biodiversity and Molecular Ecology, Fondazione Edmund Mach, IASMA, San Michele 38010, Italy

a r t i c l e i n f o

Article history:Received 2 October 2014Accepted 31 December 2014Available online 1 January 2015

Keywords:Heat stressDNA methylationGene expressionPolIVPolV

* Corresponding author.E-mail address: galina.yahubyan@gmail.com (G. Y

http://dx.doi.org/10.1016/j.plaphy.2014.12.0220981-9428/© 2015 Elsevier Masson SAS. All rights re

a b s t r a c t

Along with its essential role in the maintenance of genome integrity, DNA methylation takes part inregulation of genes which are important for plant development and stress response. In plants, DNAmethylation process can be directed by small RNAs in process known as RNA-directed DNA methylation(RdDM) involving two plant-specific RNA polymerases e PolIV and PolV. The aim of the present studywas to investigate the effect of heat stress on the expression of genes encoding key players in DNAmethylation e DNA methyltransferase (MET1, CMT3, and DRM2), the largest subunits of PoIIV and PolV(NRPD1 and NRPE1 respectively) and the DNA demethylase ROS1. We also examined the high-temperature effect on two protein-coding genes e At3g50770 and At5g43260 whose promoterscontain transposon insertions and are affected by DNA-methylation, as well as on the AtSN1, a SINE-likeretrotransposon. To assess the involvement of PolIV and PolV in heat stress response, the promotermethylation status and transcript levels of these genes were compared between wild type and doublemutant lacking NRPD1 and NRPE1. The results demonstrate coordinated up-regulation of the DRM2,NRPD1 and NRPE1 in response to high temperature and suggest that PolIV and/or PolV might be requiredfor the induction of DRM2 expression under heat stress. The ROS1 expression was confirmed to besuppressed in the mutant lacking active PolIV and PolV that might be a consequence of abolished DNAmethylation. The increased expression of At3g50770 in response to elevated temperature correlated withreduced promoter DNA methylation, while the stress response of At5g43260 did not show inversecorrelation between promoter methylation and gene expression. Our results also imply that PolIV and/orPolV could regulate gene expression under stress conditions not only through RdDM but also by acting inother regulatory processes.

© 2015 Elsevier Masson SAS. All rights reserved.

1. Introduction

Methylation of cytosine to methylcytosine is a common DNAmodification for plants, animals and fungi. DNA methylation cangenerally be found in transposable elements (TEs) and repetitiveelements as well as in gene bodies in Arabidopsis (Cokus et al.,2008; Lister et al., 2008). DNA methylation of TEs and repeats in-activates their transcription and is an evolutionary mechanism ofdefense against selfish DNA. Gene-body methylation was found tocorrelate with high expression levels (Tran et al., 2005; Zemachet al., 2010; Zhang et al., 2006) or, alternatively, DNA methylationcan define exons boundaries or regulate alternative splicing,

ahubyan).

served.

because it has been observed that exons are more highly methyl-ated than introns (Feng et al., 2010; Laurent et al., 2010).

DNA methylation occurs in three different sequence contexts:CG, CNG (where N is any base) and asymmetric CHH (where H ¼ A,T, or C), and is catalyzed by DNA methyltransferase enzymes(DNMT). They can either methylate DNA, establishing a newmethylation state (de novo methylation), or maintain preexistingmethylation (Feng et al., 2010). The maintenance of CG methylationrelies on site symmetry and requires the highly conservativemammalian DNMT1 homologue e Methylase 1 (MET1) in Arabi-dopsis (Bestor and Verdine, 1994; Cao and Jacobsen, 2002b;Finnegan et al., 1996; Kishimoto et al., 2001). De novo methylationoccurs in CNG and CHH context. Chromomethylase 3 (CMT3) is achromodomain containing plant specific enzyme (Henikoff andComai, 1998) that is responsible for the CNG methylation and

M. Naydenov et al. / Plant Physiology and Biochemistry 87 (2015) 102e108 103

reveals some locus-specific effects on asymmetric methylation(Bartee and Bender, 2001; Cao and Jacobsen, 2002a; Lindroth et al.,2001; McCallum et al., 2000). The de novo methytransferases e

Domains Rearranged Methyltransferase DRM1 and DRM2, are ho-mologues of the mammalians DNMT3 and are responsible forasymmetric methylation In Arabidopsis (Cao et al., 2000). Similarlyto CMT3, DRMmethyltransferases have locus-specific effects on theCNG methylation (Cao and Jacobsen, 2002a). In nondividing cells,CG methylation can be removed through active demethylationresulting from the action of the DNA glycosylase/lyases, RepressorOf Silencing 1 (ROS1) and Demeter (DME) (Choi et al., 2002;Kinoshita et al., 2004).

RNA-directed DNA methylation (RdDM) is a plant-specific DNAmethylation pathway, in which short interfering RNA (siRNA)molecules guide cytosine de novo methylation (Law and Jacobsen,2010). RNA-dependent RNA polymerase 2 (RDR2), Dicer-like 3(DCL3), Argonaute 4 (AGO4) and two types of DNA-dependent RNApolymerases e Polymerase IV (PolIV) and Polymerase V (PolV) aremain components in this pathway. RDR2 makes PolIV transcriptsdouble-stranded and the endonuclease DCL3 cleaves double-stranded RNAs to 24 nt siRNAs (Xie et al., 2004). In turn, AGO4, asa component of the RNA-induced silencing complex, binds siRNAsand targets cognate DNA sequence for methylation (Chan et al.,2005; Huettel et al., 2007). The plant specific PolIV and PolV aredistinguished by their unique largest subunits, NRPD1 and NRPE1,and act at different steps and loci of the RdDM pathway. WhilePolIV is needed to produce or amplify the siRNA trigger (Smithet al., 2007), PolV generally acts downstream of this step to facili-tate de novo methylation at the siRNA-targeted site (Kanno et al.,2005; Pontier et al., 2005). RdDM may be passively lost individing cells if siRNA signal is withdrawn (Chan et al., 2005).

Along with its essential role in the maintenance of genomeintegrity through silencing transposons and harmful DNA, DNAmethylation takes also part in regulation of genes that are impor-tant for plant physiology and development (Bartee and Bender,2001; McCallum et al., 2000). Environmental stress can inducechromatin structure changes like histone modifications and DNAmethylation/demethylation. In maize roots, cold stress leads to areduction of methylation in the DNA of the nucleosome's coreswhich correlates with increased expression of ZmMI1 which con-tains part of the coding region of a putative protein and part of aretrotransposon-like sequence (Steward et al., 2000). Changes inthe ambient conditions may also activate transposons which arenormally inactivated by epigenetic silencing through methylationof their promoters (Grandbastien, 2004). Heat stress mobilizesONSEN retrotransposons, making them transcriptionally active, sothey synthesize extrachromosomal DNA copies in Arabidopsisseedlings (Ito et al., 2011). Cold stress results in hypomethylationand transposition of Tam-3 transposon in Antirrhinum majus(Hashida et al., 2006). Drought and salt stress induced CNGhypermethylation of the satellite DNA in the facultative halophyteMesembryanthemum crystallinum that in turn leads to a metabolicswitch from C3 to CAM photosynthesis mode (Dyachenko et al.,2006).

It has been shown that differential expression of endogenes andtransgenes in response to stress can be regulated by RdDM of TEswhich reside in promoter regions (Grandbastien, 2004; Kashkushet al., 2003; Steward et al., 2002). Recently, it was reported thatread-through transcription from transposon remnants result bothin up- and down-regulation of adjacent protein-coding genese thepentatricopeptide repeat (PPR)-containing protein gene andseveral auxin-responsive genes (Popova et al., 2013). In our previ-ous study, we determined that the promoter regions of 81 abioticstress-responsive genes contain TE insertions, which are potentialtargets for 24 nt siRNAs and DNA methylation in Arabidopsis (Baev

et al., 2010). In the present study, we examined the high-temperature effect on expression of key DNA-methylation relatedgenes, and on promoter methylation status and transcript levels oftwo protein-coding genes, At3g50770 and At5g43260, which areunder control of promoter regions containing TE insertions (Baevet al., 2010). The AtSN1 was chosen in order to test more generaleffect of elevated temperature on DNA methylation and expressionof repeated endogenous DNA. To assess the involvement of theplant-specific RNA polymerases in stress response, the transcriptlevels of these genes and the methylation status of their promoterregions were compared between the wild type (Col-0) and thenrpd1a-1 nrpd1b-1 double mutant that lacks the largest subunits ofPolIV (NRPD1) and Pol V (NRPE1).

2. Materials and methods

2.1. Plant material and stress treatment

Seeds of Arabidopsis thaliana (Col-0), wild type and the nrpd1a-1nrpd1b-1 double mutant provided by T. Lagrange (Pontier et al.,2005) were planted on soil and stratified at 4 �C for 4 days. Seed-lings were grown in a Percival chamber at 21 �C, 16 h light (100umol m�2 s�1)/8 h dark for 5 weeks. For high-temperature treat-ment, plants were transferred to 36 �C. Plant material (rosetteleaves) of four plants was taken at different time points e 0, 6, 24and 48 h after the beginning of stress experiment. The experimentwas performed three times with nearly identical results.

2.2. Total RNA and DNA extraction

Total RNA was extracted using RNeasy Plant Mini Kit (Qiagen)and treated with DNase (Qiagen) according to the manufacturer'sinstructions. RNA samples from all time points were used for cDNAsynthesis and qRT-PCR expression analysis. Total DNA was extrac-ted using DNeasy Plant Mini Kit (Qiagen) according to the manu-facturer's instructions. DNA samples from all time points weredigested with McrBC and used for methylation-sensitive qPCR.

2.3. qRT-PCR

cDNA synthesis: 1 mg of RNA was reverse transcribed withRevertAid reverse Transcriptase (Thermo Scientific) underfollowing conditions: 65 �C for 5 min, 4 �C for 5 min; 42 �C for60 min; final inactivation at 70 �C for 10 min.

Amplification: PCR amplification was performed using a stan-dard SYBR Green protocol (Fermentas) on 7500 Real-time PCRmachine (Applied Biosystems). All reactions were carried out in atotal volume of 25 ml and contained 5 ml of diluted cDNA, 1.5 ml ofprimer mix to final concentration 0.6 mM, 4.5 ml of Nuclease-freewater and 12.5 ml of SYBR Green mix with ROX. PCR conditionswere as follows: 50 �C for 2 min, 95 �C for 10 min, followed by 40amplification cycles of 95 �C for 15 s and 60 �C for 1 min. All re-actions were performed in triplicates. The housekeeping gene -EF1a was used as endogenous control for normalization(Czechowski et al., 2005), and the untreated sample (0 h of treat-ment) was accepted as reference (RQ j 1). Ct values were calculatedusing 7500 software v.2.0.1 (ABI). Relative quantitation of geneexpression (RQ) was determined with the equation: E to the powerof eDCt of gene of interest/E to the power of �DCt of housekeepinggene, where E is the primer efficiency measured by standard curveexperiment with serial dilutions and DCt is the difference betweenthe Ct values of target gene for each sample and for the referencesample. All primer sequences are given in Supplemental file 1.

Fig. 2. Response of the retrotransposon AtSN1 to high-temperature stress in wild-typeand nrpd1a-1 nrpd1b-1 mutant. (A) Promoter analysis: relative DNA methylation wasassayed by qPCR of undigested and McrBC-digested DNA. (B) Transcript analysis:relative expression levels were assayed by qRT-PCR. Expression levels were normalizedto non-stressed controls. Representative experiment from 3 independent replicates isshown. Error bars indicate SD of replicates.

M. Naydenov et al. / Plant Physiology and Biochemistry 87 (2015) 102e108104

2.4. DNA digestion and methylation-sensitive qPCR

DNA digestion: 40 ng of genomic DNAwere digested with McrBC(NEB), which recognizes and cleaves DNA containing at least two 5-methylcytosines preceded by a purine and located in one of CG,CNG, and CNN sequence contexts. McrBC digestion was performedat 37��C for 16 h, according to the manufacturer's conditions. Inmock sample, the incubation was performed with 1 ml of 50%Glycerol instead of enzyme.

Amplification: PCR amplification was performed using a stan-dard SYBR Green protocol (Fermentas) on StepOnePlus Real-timePCR machine (Applied Biosystems). All reactions were performedin a total volume of 15 ml and contained 50 pg of DNA (digested orundigested), 0.9 ml of primermix (to final concentration 0.6 mM), 4.5of Nuclease-free water and 7.5 ml of SYBR Green with ROX. PCRconditions were as follows: 50 �C for 2 min, 95 �C for 10 min, fol-lowed by 40 amplification cycles of 95 �C for 15 s and 60 �C for1 min. All reactions were performed in triplicates. The undigested(mock) sample for each time point was used as reference sample.Relative quantitation of methylation (RQ) was calculated using theStepOnePlus (ABI) Ct Values and the equation: E to the power ofeDCt of gene of interest/E to the power of �DCt of endogenouscontrol, where E is the primer efficiency measured by standardcurve experiment with serial dilutions and DCt is the differencebetween the Ct values for mock sample and for digested sample.

Primer pairs were designed to flank the TE of At3g50770 pro-moter and each of the two TEs of the At5g43260 promoter (Figs. 3Aand 4A, Supplemental File 1), and the primers for AtSN1 methyl-ation were according (Hamilton et al., 2002).

2.5. Software and web based analysis tools

Masking of TEs in analyzed promoter regions was performedusing the software starPRO DB v1.0 http://bioinfo1.uni-plovdiv.bg/cgi-bin/starpro/(Baev et al., 2010) and Arabidopsis epigenomemap. Sequence data were retrieved from The Arabidopsis Infor-mation Resource (TAIR, www.arabidopsis.org). Primers weredesigned using the Geneious.Pro.v4.8.5 software and www.ncbi.com based tool Primer-BLAST.

Fig. 1. High-temperature response of DNA-methylation related genes. Expression patterntransferase 2 (DRM2), the largest subunit of Polymerase IV (NRPD1) and of Polymerase Vrepresented as fold change of expression relative to non-stressed controls. Representative e

3. Results

3.1. High-temperature response of key genes engaged in DNAmethylation

To explore the effect of high temperature on key genes engagedin DNA methylation, we analyzed the expression levels of severalgenes closely related to this process: the DNA methyltransferasegenes e MET1, CMT3, and DRM2, the genes encoding the largestsubunits of Polymerase IV and V (NRPD1 and NRPE1) and the geneencoding ROS1 DNA demethylase. mRNA steady-state abundancewas measured during high-temperature treatment (0 h, 6 h, 24 hand 48 h) by RT-qPCR (Fig. 1). Expression analysis revealed that theDNAmethylation-related genes responded to elevated temperaturespecifically in thewild type: (1) the expression of DRM2,NRPD1 andNRPE1, and to a lower extent that of ROS1, was stably up-regulatedin the course of high-temperature treatment; (2) the transcriptlevels of MET1 and CMT3 were increased during the first 6 h oftreatment but the longer treatment resulted in decreased transcriptlevels compared to untreated sample.

In the nrpd1a-1 nrpd1b-1 double mutant, high-temperature

of Methylase1 (MET1), (Chromomethylase 3) CMT3, Domains Rearranged Methyl-(NRPE1) and Repressor Of Silencing 1 (ROS1) was analyzed by qRT-PCR. Expression isxperiment from 3 independent replicates is shown. Error bars indicate SD of replicates.

Fig. 3. Response of the Calmodulin-like 41(CML41) encoding gene (At3g50770) to high-temperature stress in wild-type and nrpd1a-1 nrpd1b-1 mutant. (A) Gene structure e datawere obtained from the Arabidopsis epigenome map and RepeatMasker Pre-masked Genomes. Arrows indicate the position of the primers used for the following qPCR and qRT-PCR. (B) Promoter analysis: relative DNA methylation was assayed by qPCR of undigested and McrBC-digested DNA. (C) Transcript analysis: relative expression levels were assayedby qRT-PCR. Expression levels were normalized to non-stressed controls. Representative experiment from 3 independent replicates is shown. Error bars indicate SD of replicates.

Fig. 4. Response of At5g43260 to high-temperature stress in wild-type and nrpd1a-1 nrpd1b-1 mutant. (A) Gene structure e data were obtained from the Arabidopsis epigenomemap and RepeatMasker Pre-masked Genomes. Arrows indicate the position of the primers used for the following qPCR and qRT-PCR. (B) Promoter analysis: relative DNAmethylation was assayed by qPCR of undigested and McrBC-digested DNA. (C) Transcript analysis: relative expression levels were assayed by qRT-PCR. Expression levels werenormalized to non-stressed controls. Representative experiment from 3 independent replicates is shown. Error bars indicate SD of replicates.

M. Naydenov et al. / Plant Physiology and Biochemistry 87 (2015) 102e108 105

treatment resulted in expression profiles, which differed fromthose of wild type (Fig. 1): (1) the strong stimulating effect of high

temperature on DRM2 and MET1 expression that was observed inwild-type was strongly suppressed in the mutant; (2) the ROS1

M. Naydenov et al. / Plant Physiology and Biochemistry 87 (2015) 102e108106

expression was almost completely abolished in the mutant.

3.2. High-temperature response of genes affected by DNA-methylation

3.2.1. Response of the retrotransposon AtSN1Previous findings have demonstrated that temperature stress

changes the activity of LINE and copia-like retrotransposons inA. thaliana (Cavrak et al., 2014; Lang-Mladek et al., 2010; Matsunagaet al., 2012; Tittel-Elmer et al., 2010). Here, we tested the effect ofhigh temperature on methylation and transcript levels of the SINE-like retrotransposon, AtSN1. Methylation sensitive-qPCR showedthat temperature elevation resulted in enhanced DNA methylationof AtSN1 with a peak at 24 h of treatment (Fig. 2A). The AtSN1transcript levels were down-regulated throughout the entireperiod of heat treatment (48 h). The AtSN1 methylation wasstrongly reduced in the double mutant compared to wild type dueto the lack of active PolIV and PolV in the mutant. The AtSN1transcript levels were down-regulated in the mutant compared towild type before and up to 24 h of heat application, but highaccumulation of AtSN1 transcripts (over 4 times higher than that ofuntreated sample) was observed in the double mutant at 48 h oftreatment (Fig. 2B).

3.2.2. Response of the protein-coding genes At3g50770 andAt5g43260

At3g50770 encodes Calmodulin-like 41 (CML41) protein thatwas identified as being induced by oligogalacturonides, oligosac-charides derived from the plant cell wall, and by the bacterialflagellin peptide Flg22, general elicitors of the basal defenseresponse in plants. (Denoux et al., 2008). The At3g50770 promotercontains a DNA type TE insertion (At3TE76530), which lies veryclosely to the transcriptional start site (Fig. 3A). We have previouslyreported that the DNA methylation of the TE region was RdDMdependent (Baev et al., 2010). Here, the exposure of Arabidopsis tohigh temperature resulted in a strong increase of At3g50770 tran-script levels with a peak at the 24 h of treatment (over 5.5 timeshigher than those of untreated sample) (Fig. 3C). The high-temperature profiles of gene expression and promoter DNAmethylation revealed that the enhanced expression was accom-panied by reduced promoter methylation levels (Fig. 3B). The lackof PolIV and PolV resulted in highly reduced DNA methylation ofthe At3g50770 promoter region in the double mutant compared towild type (Fig. 3B).

At5g43260 encodes chaperon-like protein, there are two TEinsertions in its promoter region which are located approximatelyat 1200 bp and 1900 bp upstream of the Transcriptional Start Site(TSS) and flank a region massively targeted by siRNAs. The two TEsbelong to the DNA TE families ATREP6 and ATREP8 and are RdDMdependent (Fig. 4A). High temperature up-regulated the At5g43260transcript levels, which reached their maximum at 24 h of treat-ment (over 2.5 times higher than those of untreated sample)(Fig. 4C). Themethylation levels of the twoTE-derived regions werehigher compared to untreated sample during high-temperaturetreatment (Fig. 4B, C). These findings did not show the expectedinverse correlation between the levels of promoter methylationand transcript levels of At5g43260. The comparison of thewild typeand double mutant showed that the methylation of the two TEinsertions was highly reduced in the double mutant, while thetranscript levels were unchanged when plants were grown atnormal temperature (Fig. 4C). So, it is difficult to see any directrelationship between methylation status of the two TE regions ofthe promoter and gene expression, and that could be due to theremote position of TE insertions from the TSS.

4. Discussion

A number of studies have been devoted to the elucidation ofimpact of temperature stress on protein-coding gene expression, TEactivity and heterochromatic silencing (Ito et al., 2011; Tittel-Elmeret al., 2010; Zeller et al., 2009), yet there is a limited documentationon the involvement of the genes encoding important componentsof DNA-methylation machinery in heat stress response.

4.1. Heat stress stimulates DRM2, NRPD1 and NRPE1 expression

One of the main findings of the present study was the heat-induced expression of DRM2, NRPD1 and NRPE1, suggesting thatthe RdDM pathwaywasmobilized by high temperature to reinforcestabilization of DNA methylation. The simultaneous increase ofROS1 and DRM2 expression, which was observed in response toelevated temperature, could result in target-specific deposition andremoval of DNA methylation. There are increasing evidences forDNA-methylation changes and altered expression of DNA-methylation related enzymes in response to abiotic and bioticstress (Bilichak et al., 2012; Boyko et al., 2010; Dowen et al., 2012;Ou et al., 2012). For example the rice orthologous genes of DRM2,DRM2-1 and DRM2-2, were reported to be up-regulated by heavymetal stress (Ou et al., 2012). Stress-induced DNA methylation wasshown to be impaired in dcl3 deficiencymutant (Boyko et al., 2010).Thus, the RdDM pathway, which can operate in all sequencecontext methylation and accounts for approximately 30% of all DNAmethylation in Arabidopsis (Cokus et al., 2008; Lister et al., 2008),seems to play an indispensable role in plant response to stressfactors.

4.2. PolIV and/or PolV might be required for the induction of DRM2expression under heat stress, and for ROS1 expression under normaltemperature

The coordinated expression of DRM2, NRPD1 and NRPE1 inresponse to high-temperature stress was clearly seen not only inwild type but also in the double mutant, in which the deficiency ofthe largest subunits of PolIV and PolV repressed the stimulatingeffect of high temperature on DRM2 expression. Previous studyshowed that some proteins of the RdDM pathway were mis-localized in PolIV-null mutants (Pontes et al., 2006) that indicatedthe importance of PolIV for the pathway stability (Haag andPikaard, 2011). In addition to that, our observation suggests thatPolIV and/or PolV might be required for the induction of DRM2expression under heat stress. The observed almost completelyabolished ROS1 expression in the double mutant confirms theprevious report for down-regulated ROS1 transcript levels in thesingle mutants nrpd1, nrpe1 and nrpd2 (Huettel et al., 2006). Takentogether, our mutant analysis comes in line with the assumption ofHuettel and coworkers that plant specific RNA polymerases cancooperate with either DNA methyltransferases to catalyze cytosinemethylation, or DNA glycosylases to catalyze cytosinedemethylation.

4.3. RdDM machinery could interact with other factors to regulateAtSN1 expression at elevated temperature

High temperature caused increased DNA methylation of AtSN1that resulted in its transcriptional repression throughout the entireperiod of treatment (48 h). The increased AtSN1 methylationcorrelated with the up-regulated transcript levels of DRM2, NRPD1and NRPE1 (Figs. 1 and 2). Taken together, these results reveal theimportance of RdDM for AtSN1 silencing in plants exposed to heatstress. Notably, expression analysis revealed increased AtSN1

M. Naydenov et al. / Plant Physiology and Biochemistry 87 (2015) 102e108 107

transcript levels in the double mutant after longer heat exposurethat cannot be linked to DNA methylation (Fig. 2). AtSN1 dere-pression has been previously observed in several flowering mu-tants grown under normal temperature and, together with otherTEs, was inked to RNA-mediated chromatin silencing (Baurle andDean, 2008). The AtSN1 response reminds of the heat response ofanother TEe ONSEN, whose activationwas linked to general loss ofepigenetic control and required heat shock transcriptional factors(Cavrak et al., 2014). Likely, the RdDM machinery interacts withother factors (most likely such as chromatin remodelers or/andtranscriptional factors) to regulate AtSN1 expression at elevatedtemperature.

4.4. The increased expression of At3g50770 in response to elevatedtemperature correlates with reduced promoter DNA methylation,and suggests that PolIV and/or PolV may regulate gene expressionnot only through RdDM but also by acting in other regulatoryprocesses

The observed results indicate a negative correlation betweenthe promoter methylation and transcript levels of At3g50770, andreveal a role of DNA methylation for transcriptional regulation ofCML41 in response to elevated temperature. Recently, using mu-tants lacking CG or non-CG methylation, it was demonstrated thatDNA methylation modulated expression of some pathogen-responsive genes (Dowen et al., 2012). The lack of PolIV and PolVresulted in highly reduced DNA methylation of the At3g50770promoter region in the double mutant compared to wild type(Fig. 3B). We expected for the double mutant that the highlyreduced promoter methylation would result in At3g50770 reac-tivation under normal conditions and that effect would persist atelevated temperature. In fact, the At3g50770 transcript levels didnot differ significantly between the wild type and double mutantwhen plants were grown under normal temperature conditions(Fig. 3C). Interestingly, unlike the observed strong stimulating ef-fect of high temperature on At3g50770 expression in wild type, thetranscript levels were slightly altered by high temperature in thedouble mutant (Fig. 3C) suggesting a role for PolIV and/or PolV inregulatory mechanism different than DNA-methylation. TheAt3g50770 promoter reminds that of the FWA gene, which hastandem repeats that are targeted by siRNAs and RdDM, but incontrast to At3g50770, the FWA transgene silencing was released ofnrpd1 and nrpe1 mutants (Pontier et al., 2005). Our findings indi-cate that the high-temperature dependent At3g50770 up-regulation requires PolIV and/or PolV. It seems that these pro-teins take part not only in RdDM of the gene promoter but also in aregulatory process that could not directly be linked to promoterDNA methylation.

4.5. The At5g43260 response to elevated temperature does notshow inverse correlation between promoter methylation and geneexpression but yet requires PolIV and/or PolV

Notably, the At5g43260 expression profiles differed significantlybetween the wild type and double mutant treated with high tem-perature (Fig. 4C) e the stimulating effect of high temperature onAt5g43260 expression in the wild typewas completely abolished inthe double mutant. This observation is similar to that we describedabove for At3g50770 expression and supports the involvement ofPolIV and/or PolV in high-temperature stress response ofAt5g43260 through a regulatory mechanism distinct from pro-moter methylation. Since it has been shown that body methylationhas only minor but positive effects on levels of gene expression(Zilberman et al., 2007), one can speculate that high temperaturecould alter methylation within coding regions (or body

methylation) of these genes via PolIV or PolV-mediated mecha-nism. Besides, several studies of nrpe1 mutants show that loss ofPolV might have roles in heterochromatin organization and chro-matin remodeling pathways that are independent of RdDMpathway (Douet et al., 2009; Pontes et al., 2009). We found anumber of cis-regulatory elements in the TE-related insertions ofthe promoters of the two genes. It could be possible that high-temperature induced changes of RdDM (and of related siRNAs)and chromatin structure alter the interactions between those ele-ments and specific regulatory factors.

5. Conclusions

The current study demonstrates that key genes of plant DNAmethylation/demethylation pathway respond specifically toelevated temperature that may result in target-specific depositionand removal of DNA methylation. It provides supportive evidencesfor the involvement of plant specific RNA polymerases in plantstress response. The comparison of methylation and transcriptlevels of analyzed genes in the wild type and double mutant in-dicates that PolIV and/or PolV could regulate stress-responsivegene expression through RdDM and thus enlarges the list ofgenes affected by RdDM upon heat stress (Popova et al., 2013). Inaddition, our analysis suggests that the plant specific RNA poly-merases might act under heat stress in regulatory mechanismsdifferent from RdDM. It will be of great importance to analyzenrpd1 and nrpe1 single mutants in order to distinguish the partic-ular role of each of the two RNA polymerases in stress response andto go further in elucidating other possible mechanisms of theiraction.

Contributions

M.N., G.Y. designed research, M.N., E.A., M.G, N.G. carried outexperiments, V.B performed bioinformatics analyses, G.Y. wrote thepaper.

Acknowledgment

We thank Thierry Lagrange and lab colleagues from the LGDP,University of Perpignan, France, for providing mutant seed stocks.This work was partially supported by Bulgarian Ministry of Edu-cation and Sciences Grant DO02-279, DKOF7RP02/20 and EC FP7REGPOT Project BioSupport.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.plaphy.2014.12.022.

References

Baev, V., Naydenov, M., Apostolova, E., Ivanova, D., Doncheva, S., Minkov, I.,Yahubyan, G., 2010. Identification of RNA-dependent DNA-methylation regu-lated promoters in Arabidopsis. Plant Physiol. Biochem. 48, 393e400.

Bartee, L., Bender, J., 2001. Two Arabidopsis methylation-deficiency mutationsconfer only partial effects on a methylated endogenous gene family. NucleicAcids Res. 29, 2127e2134.

Baurle, I., Dean, C., 2008. Differential interactions of the autonomous pathway RRMproteins and chromatin regulators in the silencing of Arabidopsis targets. PLoSOne 3, e2733.

Bestor, T.H., Verdine, G.L., 1994. DNA methyltransferases. Curr. Opin. Cell. Biol. 6,380e389.

Bilichak, A., Ilnystkyy, Y., Hollunder, J., Kovalchuk, I., 2012. The progeny of Arabi-dopsis thaliana plants exposed to salt exhibit changes in DNA methylation,histone modifications and gene expression. PLoS One 7, e30515.

Boyko, A., Blevins, T., Yao, Y., Golubov, A., Bilichak, A., Ilnytskyy, Y., Hollunder, J.,Meins Jr., F., Kovalchuk, I., 2010. Transgenerational adaptation of Arabidopsis tostress requires DNA methylation and the function of Dicer-like proteins. PLoS

M. Naydenov et al. / Plant Physiology and Biochemistry 87 (2015) 102e108108

One 5, e9514.Cao, X., Jacobsen, S.E., 2002a. Locus-specific control of asymmetric and CpNpG

methylation by the DRM and CMT3 methyltransferase genes. Proc. Natl. Acad.Sci. U. S. A. 99 (Suppl. 4), 16491e16498.

Cao, X., Jacobsen, S.E., 2002b. Role of the arabidopsis DRM methyltransferases in denovo DNA methylation and gene silencing. Curr. Biol. 12, 1138e1144.

Cao, X., Springer, N.M., Muszynski, M.G., Phillips, R.L., Kaeppler, S., Jacobsen, S.E.,2000. Conserved plant genes with similarity to mammalian de novo DNAmethyltransferases. Proc. Natl. Acad. Sci. U. S. A. 97, 4979e4984.

Cavrak, V.V., Lettner, N., Jamge, S., Kosarewicz, A., Bayer, L.M., Mittelsten Scheid, O.,2014. How a retrotransposon exploits the plant's heat stress response for itsactivation. PLoS Genet. 10, e1004115.

Chan, S.W., Henderson, I.R., Jacobsen, S.E., 2005. Gardening the genome: DNAmethylation in Arabidopsis thaliana. Nat. Rev. Genet. 6, 351e360.

Choi, Y., Gehring, M., Johnson, L., Hannon, M., Harada, J.J., Goldberg, R.B.,Jacobsen, S.E., Fischer, R.L., 2002. DEMETER, a DNA glycosylase domain protein,is required for endosperm gene imprinting and seed viability in arabidopsis.Cell 110, 33e42.

Cokus, S.J., Feng, S., Zhang, X., Chen, Z., Merriman, B., Haudenschild, C.D.,Pradhan, S., Nelson, S.F., Pellegrini, M., Jacobsen, S.E., 2008. Shotgun bisulphitesequencing of the Arabidopsis genome reveals DNA methylation patterning.Nature 452, 215e219.

Czechowski, T., Stitt, M., Altmann, T., Udvardi, M.K., Scheible, W.R., 2005. Genome-wide identification and testing of superior reference genes for transcriptnormalization in Arabidopsis. Plant Physiol. 139, 5e17.

Denoux, C., Galletti, R., Mammarella, N., Gopalan, S., Werck, D., De Lorenzo, G.,Ferrari, S., Ausubel, F.M., Dewdney, J., 2008. Activation of defense responsepathways by OGs and Flg22 elicitors in Arabidopsis seedlings. Mol. Plant 1,423e445.

Douet, J., Tutois, S., Tourmente, S., 2009. A Pol V-mediated silencing, independent ofRNA-directed DNA methylation, applies to 5S rDNA. PLoS Genet. 5, e1000690.

Dowen, R.H., Pelizzola, M., Schmitz, R.J., Lister, R., Dowen, J.M., Nery, J.R., Dixon, J.E.,Ecker, J.R., 2012. Widespread dynamic DNA methylation in response to bioticstress. Proc. Natl. Acad. Sci. U. S. A. 109, E2183eE2191.

Dyachenko, O.V., Zakharchenko, N.S., Shevchuk, T.V., Bohnert, H.J., Cushman, J.C.,Buryanov, Y.I., 2006. Effect of hypermethylation of CCWGG sequences in DNA ofMesembryanthemum crystallinum plants on their adaptation to salt stress.Biochem. (Mosc) 71, 461e465.

Feng, S., Cokus, S.J., Zhang, X., Chen, P.Y., Bostick, M., Goll, M.G., Hetzel, J., Jain, J.,Strauss, S.H., Halpern, M.E., Ukomadu, C., Sadler, K.C., Pradhan, S., Pellegrini, M.,Jacobsen, S.E., 2010. Conservation and divergence of methylation patterning inplants and animals. Proc. Natl. Acad. Sci. U. S. A. 107, 8689e8694.

Finnegan, E.J., Peacock, W.J., Dennis, E.S., 1996. Reduced DNA methylation in Ara-bidopsis thaliana results in abnormal plant development. Proc. Natl. Acad. Sci.U. S. A. 93, 8449e8454.

Grandbastien, M.A., 2004. Stress activation and genomic impact of plant retro-transposons. J. Soc. Biol. 198, 425e432.

Haag, J.R., Pikaard, C.S., 2011. Multisubunit RNA polymerases IV and V: purveyors ofnon-coding RNA for plant gene silencing. Nat. Rev. Mol. Cell. Biol. 12, 483e492.

Hamilton, A., Voinnet, O., Chappell, L., Baulcombe, D., 2002. Two classes of shortinterfering RNA in RNA silencing. EMBO J. 21, 4671e4679.

Hashida, S.N., Uchiyama, T., Martin, C., Kishima, Y., Sano, Y., Mikami, T., 2006. Thetemperature-dependent change in methylation of the Antirrhinum transposonTam3 is controlled by the activity of its transposase. Plant Cell. 18, 104e118.

Henikoff, S., Comai, L., 1998. A DNA methyltransferase homolog with a chromo-domain exists in multiple polymorphic forms in Arabidopsis. Genetics 149,307e318.

Huettel, B., Kanno, T., Daxinger, L., Aufsatz, W., Matzke, A.J., Matzke, M., 2006.Endogenous targets of RNA-directed DNA methylation and Pol IV in Arabi-dopsis. EMBO J. 25, 2828e2836.

Huettel, B., Kanno, T., Daxinger, L., Bucher, E., van der Winden, J., Matzke, A.J.,Matzke, M., 2007. RNA-directed DNA methylation mediated by DRD1 and PolIVb: a versatile pathway for transcriptional gene silencing in plants. Biochim.Biophys. Acta 1769, 358e374.

Ito, H., Gaubert, H., Bucher, E., Mirouze, M., Vaillant, I., Paszkowski, J., 2011. An siRNApathway prevents transgenerational retrotransposition in plants subjected tostress. Nature 472, 115e119.

Kanno, T., Huettel, B., Mette, M.F., Aufsatz, W., Jaligot, E., Daxinger, L., Kreil, D.P.,Matzke, M., Matzke, A.J., 2005. Atypical RNA polymerase subunits required forRNA-directed DNA methylation. Nat. Genet. 37, 761e765.

Kashkush, K., Feldman, M., Levy, A.A., 2003. Transcriptional activation of retro-transposons alters the expression of adjacent genes in wheat. Nat. Genet. 33,102e106.

Kinoshita, T., Miura, A., Choi, Y., Kinoshita, Y., Cao, X., Jacobsen, S.E., Fischer, R.L.,Kakutani, T., 2004. One-way control of FWA imprinting in Arabidopsis

endosperm by DNA methylation. Science 303, 521e523.Kishimoto, N., Sakai, H., Jackson, J., Jacobsen, S.E., Meyerowitz, E.M., Dennis, E.S.,

Finnegan, E.J., 2001. Site specificity of the Arabidopsis METI DNA methyl-transferase demonstrated through hypermethylation of the superman locus.Plant Mol. Biol. 46, 171e183.

Lang-Mladek, C., Popova, O., Kiok, K., Berlinger, M., Rakic, B., Aufsatz, W., Jonak, C.,Hauser, M.T., Luschnig, C., 2010. Transgenerational inheritance and resetting ofstress-induced loss of epigenetic gene silencing in Arabidopsis. Mol. Plant 3,594e602.

Laurent, L., Wong, E., Li, G., Huynh, T., Tsirigos, A., Ong, C.T., Low, H.M., KinSung, K.W., Rigoutsos, I., Loring, J., Wei, C.L., 2010. Dynamic changes in thehuman methylome during differentiation. Genome Res. 20, 320e331.

Law, J.A., Jacobsen, S.E., 2010. Establishing, maintaining and modifying DNAmethylation patterns in plants and animals. Nat. Rev. Genet. 11, 204e220.

Lindroth, A.M., Cao, X., Jackson, J.P., Zilberman, D., McCallum, C.M., Henikoff, S.,Jacobsen, S.E., 2001. Requirement of CHROMOMETHYLASE3 for maintenance ofCpXpG methylation. Science 292, 2077e2080.

Lister, R., O'Malley, R.C., Tonti-Filippini, J., Gregory, B.D., Berry, C.C., Millar, A.H.,Ecker, J.R., 2008. Highly integrated single-base resolution maps of the epi-genome in Arabidopsis. Cell 133, 523e536.

Matsunaga, W., Kobayashi, A., Kato, A., Ito, H., 2012. The effects of heat inductionand the siRNA biogenesis pathway on the transgenerational transposition ofONSEN, a copia-like retrotransposon in Arabidopsis thaliana. Plant Cell. Physiol.53, 824e833.

McCallum, C.M., Comai, L., Greene, E.A., Henikoff, S., 2000. Targeted screening forinduced mutations. Nat. Biotechnol. 18, 455e457.

Ou, X., Zhang, Y., Xu, C., Lin, X., Zang, Q., Zhuang, T., Jiang, L., von Wettstein, D.,Liu, B., 2012. Transgenerational inheritance of modified DNA methylation pat-terns and enhanced tolerance induced by heavy metal stress in rice (Oryzasativa L.). PLoS One 7, e41143.

Pontes, O., Costa-Nunes, P., Vithayathil, P., Pikaard, C.S., 2009. RNA polymerase Vfunctions in Arabidopsis interphase heterochromatin organization indepen-dently of the 24-nt siRNA-directed DNA methylation pathway. Mol. Plant 2,700e710.

Pontes, O., Li, C.F., Costa Nunes, P., Haag, J., Ream, T., Vitins, A., Jacobsen, S.E.,Pikaard, C.S., 2006. The Arabidopsis chromatin-modifying nuclear siRNApathway involves a nucleolar RNA processing center. Cell 126, 79e92.

Pontier, D., Yahubyan, G., Vega, D., Bulski, A., Saez-Vasquez, J., Hakimi, M.A., Lerbs-Mache, S., Colot, V., Lagrange, T., 2005. Reinforcement of silencing at trans-posons and highly repeated sequences requires the concerted action of twodistinct RNA polymerases IV in Arabidopsis. Genes. Dev. 19, 2030e2040.

Popova, O., Dinh, H., Aufsatz, W., Jonak, C., 2013. The RdDM pathway is required forbasal heat tolerance in Arabidopsis. Mol. Plant 6, 396e410.

Smith, L.M., Pontes, O., Searle, I., Yelina, N., Yousafzai, F.K., Herr, A.J., Pikaard, C.S.,Baulcombe, D.C., 2007. An SNF2 protein associated with nuclear RNA silencingand the spread of a silencing signal between cells in Arabidopsis. Plant Cell. 19,1507e1521.

Steward, N., Ito, M., Yamaguchi, Y., Koizumi, N., Sano, H., 2002. Periodic DNAmethylation in maize nucleosomes and demethylation by environmental stress.J. Biol. Chem. 277, 37741e37746.

Steward, N., Kusano, T., Sano, H., 2000. Expression of ZmMET1, a gene encoding aDNA methyltransferase from maize, is associated not only with DNA replicationin actively proliferating cells, but also with altered DNA methylation status incold-stressed quiescent cells. Nucleic Acids Res. 28, 3250e3259.

Tittel-Elmer, M., Bucher, E., Broger, L., Mathieu, O., Paszkowski, J., Vaillant, I., 2010.Stress-induced activation of heterochromatic transcription. PLoS Genet. 6,e1001175.

Tran, R.K., Henikoff, J.G., Zilberman, D., Ditt, R.F., Jacobsen, S.E., Henikoff, S., 2005.DNA methylation profiling identifies CG methylation clusters in Arabidopsisgenes. Curr. Biol. 15, 154e159.

Xie, Z., Johansen, L.K., Gustafson, A.M., Kasschau, K.D., Lellis, A.D., Zilberman, D.,Jacobsen, S.E., Carrington, J.C., 2004. Genetic and functional diversification ofsmall RNA pathways in plants. PLoS Biol. 2, E104.

Zeller, G., Henz, S.R., Widmer, C.K., Sachsenberg, T., Ratsch, G., Weigel, D.,Laubinger, S., 2009. Stress-induced changes in the Arabidopsis thaliana tran-scriptome analyzed using whole-genome tiling arrays. Plant J. 58, 1068e1082.

Zemach, A., McDaniel, I.E., Silva, P., Zilberman, D., 2010. Genome-wide evolutionaryanalysis of eukaryotic DNA methylation. Science 328, 916e919.

Zhang, X., Yazaki, J., Sundaresan, A., Cokus, S., Chan, S.W., Chen, H., Henderson, I.R.,Shinn, P., Pellegrini, M., Jacobsen, S.E., Ecker, J.R., 2006. Genome-wide high-resolution mapping and functional analysis of DNA methylation in arabi-dopsis. Cell 126, 1189e1201.

Zilberman, D., Gehring, M., Tran, R.K., Ballinger, T., Henikoff, S., 2007. Genome-wideanalysis of Arabidopsis thaliana DNA methylation uncovers an interdependencebetween methylation and transcription. Nat. Genet. 39, 61e69.