Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Jul 13;37(14):e98482.
doi: 10.15252/embj.201798482. Epub 2018 Jun 5.

Transcriptional Control and Exploitation of an Immune-Responsive Family of Plant Retrotransposons

Affiliations
Free PMC article

Transcriptional Control and Exploitation of an Immune-Responsive Family of Plant Retrotransposons

Jérôme Zervudacki et al. EMBO J. .
Free PMC article

Abstract

Mobilization of transposable elements (TEs) in plants has been recognized as a driving force of evolution and adaptation, in particular by providing genes with regulatory modules that impact their transcription. In this study, we employed an ATCOPIA93 long-terminal repeat (LTR) promoter-GUS fusion to show that this retrotransposon behaves like an immune-responsive gene during pathogen defense in Arabidopsis We also showed that the endogenous ATCOPIA93 copy "EVD", which is activated in the presence of bacterial stress, is negatively regulated by both DNA methylation and polycomb-mediated silencing, a mode of repression typically found at protein-coding and microRNA genes. Interestingly, an ATCOPIA93-derived soloLTR is located upstream of the disease resistance gene RPP4 and is devoid of DNA methylation and H3K27m3 marks. Through loss-of-function experiments, we demonstrate that this soloLTR is required for the proper expression of RPP4 during plant defense, thus linking the responsiveness of ATCOPIA93 to biotic stress and the co-option of its LTR for plant immunity.

Keywords: Arabidopsis; DNA methylation; innate immunity; polycomb silencing; transposable element.

Figures

Figure 1
Figure 1. ATCOPIA93 LTR::GUS transcriptional fusion behaves like a canonical immune‐responsive gene

Cytosine methylation analyzed by bisulfite‐sequencing at the LTR::GUS transgene. Genomic DNA of a pool of LTR::GUS transgenic plants (four rosette‐stage plants, line T3#12) was treated with sodium bisulfite, amplified with primers specific for the LTR contained in the LTR::GUS construct and cloned for sequencing (19 clones). The endogenous LTR of ATCOPIA93 EVD was sequenced as a positive control (17 clones). The analysis of another T3 line led to the same results (Fig EV1B). The percentage of methylated cytosines is indicated by vertical bars. The number of CG, CHG, and CHH sites is indicated on the right. This result was also reproduced in various T3 and T1 lines by a Sau96I methylation‐sensitive assay analyzing the first CG site (black asterisk; Fig EV1C).

Accumulation of GUS protein detected in response to bacterial elicitors of basal immunity. Upper panel: Representative pictures of leaves infiltrated with water (mock), Pto DC3000 deleted of 28 effectors (PtoΔ28E) at 2 × 108 colony‐forming unit per ml (cfu/ml) or 1 μM of flg22, and incubated with GUS substrate 24 h post‐infiltration (24 hpi). The number of leaves showing this representative phenotype is indicated in brackets. The T3 line LTR::GUS #12 (used for the remainder of the study) is shown here, but two additional homozygous T3 lines displayed the same phenotype and are presented in Fig EV1D. Lower panel: Western blot analysis over a 24‐h time course; RbC: Rubisco. Three to four plants (two leaves per plant) were infiltrated for each condition and time point, and leaves pooled by condition and time point before extracting the proteins. Samples derived from the same experiment, and gels and blots were processed in parallel. This experiment was repeated twice with similar results.

Time‐course analysis of GUS mRNA (plain lines) and PTI‐marker WRKY29 mRNA (dashed lines) by RT–qPCR. Leaves were infiltrated with water (mock), or PtoΔ28E bacteria at 2 × 108 cfu/ml; two similar leaves of three to four plants were pooled by condition and by time point after infiltration (as in B) before extracting the RNA subjected to RT–qPCR. Values are relative to the expression of the UBIQUITIN gene (At2g36060). This experiment was repeated twice independently, and another independent experiment is shown in Fig EV1E.

Accumulation of GUS protein detected in response to virulent Pto DC3000 versus PtoΔ28E. Upper panel: Representative pictures of leaves infiltrated with water (mock), effectorless (PtoΔ28E), and virulent (Pto) bacteria Pto DC3000, both at 1 × 107 cfu/ml, and incubated with GUS substrate at 24 hpi. The number of leaves showing this representative phenotype is shown in brackets. Lower panel: Western blot analysis of the GUS protein accumulated at 9 hpi; RbC: Rubisco. Two similar leaves of three to four plants were pooled by condition and time point before extracting the proteins. This experiment was repeated twice with similar results.

Activation of the GUS expression upon PtoΔ28E elicitation in LTR::GUS plants with mutated W‐boxes. Experiments were performed on 28, 22, and 19 primary transformants for the LTR::GUS WT, m1, and m2 constructs, respectively; the point mutations introduced are depicted on the right (W‐boxes are the sequences in bold). Mock and PtoΔ28E (at 2 × 108 cfu/ml) treatments were performed on four similar leaves of each individual transformant, and GUS staining performed 24 h later on two leaves. One representative picture (in brackets is the number of plants showing this phenotype) for one primary transformant is shown for each construct with each treatment. Plants were classified into three categories: normal GUS induction, loss of GUS induction, constitutive GUS expression and percentages of plants belonging to each category over the total number of plants tested were calculated.

Source data are available online for this figure.
Figure EV1
Figure EV1. ATCOPIA93 LTR::GUS transcriptional fusion behaves like a canonical immune‐responsive gene

Alignment between the LTRs of ATCOPIA93‐EVD and ATCOPIA93‐ATR showing that they are identical in sequence. The two W‐boxes tested in Fig 1 are highlighted in blue; the beginning and end of the LTRs of the ATCOPIA93 family are highlighted in green and orange, respectively; the GGGCC sequence in red is the site recognized by Sau96I methylation‐sensitive restriction enzyme and underlined is the CG site analyzed by Chop‐qPCR in Figs EV1C and 3D. The five CG sites of the LTR are highlighted with pink boxes.

Cytosine methylation analyzed by bisulfite‐sequencing at the LTR::GUS transgene. Genomic DNA of a pool of LTR::GUS transgenic plants (four rosette‐stage plants of the line T3#12) was treated with sodium bisulfite, amplified with primers specific for the LTR contained in the LTR::GUS construct and cloned for sequencing (21 to 11 clones—toward the end of the sequence—were analyzed for T3#6, and 12 clones were analyzed for T4#12) as in Fig 1A. The percentage of methylated cytosines is indicated by vertical bars.

Methylation status of the DNA (CG site) at the LTR in LTR::GUS transgenic plants by Sau96I Chop‐qPCR. DNA from a pool of leaves of single primary transformants and pools of T3 homozygous plants was digested with the methylation‐sensitive restriction endonuclease Sau96I which recognizes GGNCC sites—here GGGCCG—and is sensitive to methylation of the second C. Digested DNA was quantified by using qPCR with primers spanning a Sau96I restriction site in the LTR. On the left, primers were specific for the transgenic LTR (one primer in the vector): Lack of amplification shows unmethylation of the CG site in the transgenic LTR. On the right, as a control, the same DNA was analyzed with primers specific for the endogenous EVD‐LTR (one primer upstream of EVD): Amplification shows methylation of the CG site in the endogenous 5′ LTR sequence. The signal was normalized to an undigested control. The assay principle is schematized under the graph.

Analysis of additional, independent LTR::GUS transgenic lines: Representative pictures of leaves infiltrated with water (mock), Pto DC3000 deleted of 28 effectors (PtoΔ28E) at 2 × 108 colony‐forming unit per ml (cfu/ml) with GUS substrate 24 h post‐infiltration (24 hpi). The number of leaves showing this representative phenotype is indicated in brackets.

Additional independent experiment for the time‐course analysis of GUS mRNA (plain lines) and PTI‐marker WRKY29 mRNA (dashed lines) by RT–qPCR. The experiment was performed as in Fig 1C.

Source data are available online for this figure.
Figure 2
Figure 2. ATCOPIA93 reactivation is negatively controlled by DNA methylation

Time‐course analysis of ATCOPIA93 mRNA by RT–qPCR. Leaves were infiltrated with water (mock) or PtoΔ28E bacteria at 2 × 108 cfu/ml; two similar leaves of three to four plants were pooled by condition and by time point after infiltration before extracting total RNAs. Values were determined by RT–qPCR and are relative to the expression of the UBIQUITIN gene (At2g36060). This experiment was repeated twice with similar results, and another independent experiment is shown in Fig EV2.

ATCOPIA93 mRNA analysis in two methylation‐defective mutants, ddm1 and met1, at 6 h post‐treatment with either water or PtoΔ28E at 2 × 108 cfu/ml. Material and data were generated as in (A). The data points for four independent experiments are plotted. Two‐tailed P‐values were calculated by paired t‐test to take into account inter‐experiment variability. The variability observed between biological replicates is inherent to the developmental stage (adult leaves) analyzed, which often shows differences in the timing and extent of PTI from one experiment to the other (see Figs 1C, EV1E, 2A, and EV2).

Source data are available online for this figure.
Figure EV2
Figure EV2. ATCOPIA93 reactivation is negatively controlled by DNA methylation
Biological replicate (independent experiment) for the time‐course analysis of ATCOPIA93 mRNA by RT–qPCR. The experiment was performed as in Fig 2A.Source data are available online for this figure.
Figure 3
Figure 3. H3K27m3 and DNA methylation co‐exist at ATCOPIA93

IGB (integrative genome browser) views showing H3K9m2 levels and H3K27m3 levels in WT and met1 rosette leaves, at ATCOPIA93 EVD and ATR (ChIP‐chip public data, Deleris et al, 2012). Orange horizontal bars: protein‐coding genes; horizontal green bars: transposable elements. The LTRs are delineated by pink bars. Vertical blue bars: H3K9m2 signal relative to H3 (two top lanes); vertical purple bars: H3K27m3 signal relative to H3 for each probe.

Analysis of H3K27m3 marks at ATCOPIA93 EVD and ATR by ChIP on rosette leaves, followed by qPCR, in wild‐type plants and in clf plants mutated for the H3K27 methyltransferase CURLY LEAF. Data were normalized to the input DNA. ATCOPIA93 CDS is a region in ATCOPIA93 GAG common to EVD and ATR. AT5g17120 is a region in the protein‐coding gene located upstream of EVD. FLC is a region located in the first intron of FLOWERING LOCUS C which shows high levels of H3K27m3 in vegetative tissues and serves as a positive control. TA3 is a transposon and serves as a negative control. Because of technical variability in the ChIP efficiency, one ChIP experiment is presented here and two other independent experiments are presented in Fig EV3B. ChIPs were performed on a pool of rosette leaves from eight to 10 plants/genotype.

Genomic distribution of H3K27m3 marks between EVD and ATR loci by ChIP‐PCR pyrosequencing. Upper panel: Depiction of the pyrosequenced region (in yellow) within the GAG biotinylated qPCR amplicon obtained after H3K27m3 ChIP‐qPCR and purification with streptavidin beads. The position interrogated corresponds to the discriminating SNP between EVD (C/G) and ATR (A/T). Lower panel: The % indicated represents the % of G (EVD, dark blue bar) or T (ATR, light blue bar) at that position. The PCR and sequencing primers were designed so that other ATCOPIA93‐derived sequences (divergent and presumably nonfunctional) such as AT4G04410 and AT1G43775 cannot be amplified and so that the allelic ratio between the two active ATCOPIA93 copies EVD and ATR only can be evaluated. To verify this, the qPCR GAG product is also amplified from the Input gDNA as a control where a 50–50% ratio is expected. For clarity, an average of two experiments performed on two independent Input and ChIPs samples is shown (error bars represent standard error (SE) of the mean) and individual datasets presented in Fig EV3C.

Methylation status of the DNA captured with H3K27m3 by Sau96I Chop‐qPCR. H3K27m3 ChIP‐DNA from two independent ChIPs was digested with the methylation‐sensitive restriction endonuclease Sau96I which is sensitive to the methylation of the second C at the GGGCCG site in the LTR (as in Fig EV1). The values plotted correspond to the ratio between the amount of amplified DNA in the Sau96I digestion and the amount of amplified DNA in the undigested control, as calculated by the formula 2−(Ct.digestedDNA–Ct.undigestedDNA) and using primers specific for a region of EVD‐LTR spanning this Sau96I restriction site. Dark and light symbols are used for the first and second experiments, respectively. Results show that the WT ChIP‐DNA had significantly less digestion compared with the ddm1 control; thus, there was more methylation.

Source data are available online for this figure.
Figure EV3
Figure EV3. H3K27m3 and DNA methylation co‐exist at ATCOPIA93

IGV (integrative genome viewer) showing H3K27m3 levels at ATCOPIA93 EVD in wild‐type Col and clf mutant. Publically available ChIP‐seq data (Wang et al, 2016) were processed by mapping all reads of the H3K27m3‐ChIP libraries including non‐unique multi‐mappers such as the ones coming from ATCOPIA93 LTR and CDS. The bottom panel is a close‐up view of the top panel.

Additional biological replicates for H3K27m3 analysis at ATCOPIA93 by ChIP‐qPCR. The ChIPs were performed with different amounts of starting material for each batch (bio.rep.), which contributes to explain the differences in ChIP efficiency. Left panel: Biological replicate for loss of H3K27m3 marks in clf. Right panel: Additional biological replicate that was used for pyrosequencing of H3K27m3‐immunoprecipitated DNA in wild‐type plants.

Detail of pyrosequencing replicates on H3K27m3 ChIP‐DNA at ATCOPIA93 CDS.

Analysis of H3K9m2 marks at ATCOPIA93 EVD and ATR by ChIP in rosette leaves, followed by qPCR. Data were normalized to the Input DNA. Loci tested are as in Fig 3B. Because of variability in the ChIP efficiency (different amounts of starting material), the three biological replicates are shown separately.

Absence of H3K27m3 marks at the transgenic LTR sequence (“trLTR”) in LTR::GUS plants (rosette leaves). qPCRs were performed using primers specific for the transgenic LTR; UBQ (UBIQUITIN), and TA3 transposon sequences are used as a negative control for the ChIP while FLC (FLOWERING LOCUS C) is used as a positive control. Because of variability in the ChIP efficiency (different amounts of starting material), the two biological replicates are shown separately.

Source data are available online for this figure.
Figure 4
Figure 4. PcG‐mediated silencing and DNA methylation exert a dual and differential negative control on the induction of EVD and ATR during PAMP‐triggered immunity

ATCOPIA93 mRNA levels in pools of rosette leaves of DNA methylation mutant ddm1 and PRC2 mutant clf (three to four plants per condition), 6 h post‐infiltration with either water or PtoΔ28E bacteria at 2 × 108 cfu/ml. Values were determined by RT–qPCR and are relative to the expression of the UBIQUITIN (At2g36060) gene. Five independent experiments were performed, and the five corresponding biological replicates are shown and represented by black, blue, green, brown, and purple symbols, respectively. Two‐tailed P‐values were calculated by paired t‐test to take into account inter‐experiment variability. The variability observed between biological replicates is inherent to the developmental stage (adult leaves) analyzed which often shows differences from one experiment to the other, in the timing and extent of PTI (see Figs 1C, EV1E, 2A, and EV2).

Left: Qualitative analysis by pyrosequencing of the RT–qPCR products quantified in (A). The pyrosequenced region and SNP interrogated are the same as in Fig 3. For clarity, the average of three experiments on three of the biological replicates (A) is shown (error bars represent SE of the mean); the independent replicates are shown individually in Fig EV4B with the corresponding color code. Right: Determination of EVD and ATR transcripts levels 6 hpi with PtoΔ28E by integrating ATCOPIA93 total transcript levels (A) with pyrosequencing data. Calculations were made by applying the average respective ratios of EVD and ATR (left panel) to the average RNA values (relative to UBIQUITIN and shown in A) of the three pyrosequenced biological replicates. Pyrosequencing could not be performed in wild type because of too low amount of ATCOPIA93 transcript; thus, the EVD/ATR ratio could not be determined and an intermediate gray color is used.

ATCOPIA93 mRNA levels in pools of rosette leaves of ddm1‐clf double mutants (three to four plants per condition), 6 h post‐infiltration with either water or PtoΔ28E bacteria at 2 × 108 cfu/ml. Values were determined as in (A). Three independent experiments were performed, and the corresponding biological replicates are shown and represented by purple, green, and brown symbols, respectively. Due to high values obtained in the double mutants, the scale is different from the scale in (A). Two‐tailed P‐values were calculated by paired t‐test.

Left: Qualitative analysis by pyrosequencing of the RT–qPCR products quantified in (C) and as in (B). The average of the values obtained from the three experiments above (C) is shown (error bars represent SE of the mean); the independent replicates are shown individually in Fig EV4E with the corresponding color code. Right: Determination of EVD and ATR transcripts levels 6 hpi with PtoΔ28E by integrating ATCOPIA93 total transcript levels (C) with pyrosequencing data. Calculations were made as in (B).

Top: ATCOPIA93 linear extrachromosomal DNA (ecDNA) was detected by adaptor‐ligation PCR as previously described (Takeda et al, 2001; Mirouze et al, 2009; see scheme) on DNA from various genotypes. Pools of rosette leaves (three to four plants per condition), 24 h post‐infiltration with either water or PtoΔ28E bacteria at 2 × 108 cfu/ml, were used. UBIQUITIN (UBQ) was used to control genomic DNA amounts; epiRIL “454” (Marí‐Ordóñez et al, 2013) was used as a positive control for ATCOPIA93 ecDNA accumulation and two F3 ddm1‐clf lines were tested. A technical replicate (independent ligation experiment) along with negative controls lacking ligation is shown in Fig EV4F. Bottom: qPCR analysis of the ecDNA using the same primers as above in ddm1‐clf mutants. Three biological replicates are shown.

Source data are available online for this figure.
Figure EV4
Figure EV4. PcG‐mediated silencing and DNA methylation exert a dual and differential negative control on the induction of EVD and ATR during PAMP‐triggered immunity

ATCOPIA93 mRNA levels in seedlings of ddm1 and clf mutants. About thirty to forty of 3‐week‐old seedlings (grown in plates then transferred to liquid medium) were vacuum‐infiltrated with either water or a suspension of PtoΔ28E bacteria at 2 × 108 cfu/ml and collected 2 h later (this time point was determined on the basis of LTR::GUS expression in a pilot experiment). Less variability is observed at this stage as shown by the two biological replicates (black and gray symbols).

Detail of the pyrosequencing biological replicates for ATCOPIA93 cDNA analysis. Each color corresponds to each biological replicate shown in Fig 4A.

Analysis of H3K9m2 and H3K27m3 marks in clf and ddm1 at ATCOPIA93 EVD and ATR by ChIP in rosette leaves, followed by qPCR. Data were normalized to the Input DNA. Two biological replicates are presented. ChIPs were performed in parallel in WT, ddm1, and clf samples. The analysis of the WT ChIP‐DNA is already presented in Figs 3B and EV3C (bio.rep.1 and 2) for comparisons between LTR and CDS, and between EVD and ATR. Here, the ddm1 and clf data were included to show that H3K9m2 marks persist in clf mutant (and, as expected, are almost absent in the ddm1 negative control) and H3K27m3 marks persist in ddm1 (and are strongly reduced in clf as already shown in Fig 3B).

Representative pictures of ddm1‐clf mutants alongside WT plants and ddm1 and clf single mutants (4.5‐week‐old plants grown in short days). The red arrow indicates that the clf mutants have started to bolt.

Detail of the pyrosequencing biological replicates for ATCOPIA93 cDNA analysis. Each color corresponds to each biological replicate shown in Fig 4D. In addition, values obtained for mock‐treated ddm1‐clf double mutants are shown in the right panel.

Technical replicate (independent ligation) for ecDNA detection on the same samples as Fig 4E. The corresponding negative controls lacking ligation (“ligation”) were included.

Source data are available online for this figure.
Figure EV5
Figure EV5. Cis‐regulation of the RPP4 immune resistance gene by a ATCOPIA93‐derived, unmethylated soloLTR

Summary of the characteristics of the ATCOPIA93‐derived soloLTRs in Col‐0 wild‐type plants. chr: chromosome; (+): plus strand; (−): minus strand; mC: methylated cytosines. Methylation status at cytosines and transcription was inferred from inspection of public data at unique reads (http://neomorph.salk.edu/arabidopsis_methylomes/stressed_ath_methylomes.html); trimethylation status of H3K27 was inferred from inspection of both ChIP‐chip (Deleris et al, 2012) and ChIP‐seq data for unique reads (Wang et al, 2016). Transcription downstream of the soloLTRs was inspected in plants treated with various bacteria (http://neomorph.salk.edu/arabidopsis_methylomes/stressed_ath_methylomes.html; Dowen et al, 2012); in addition, RPP4 induction was observed in PtoΔ28E‐treated plants (B) and in response to various bacterial and oomycete elicitors (https://bar.utoronto.ca/eplant).

RPP4 mRNA levels in wild‐type plants, at 6 hpi with either water or PtoΔ28E bacteria. Two similar rosette leaves of three plants were used per condition. Values were determined by RT–qPCR and are relative to the expression of the UBIQUITIN gene. Three independent biological replicates are shown.

Alignment between the soloLTR‐5 (upstream of RPP4) and the EVD/ATR soloLTR. The W‐box 1 (GGTCAA), which is conserved in both, is depicted in blue. The blue asterisk highlights a single nucleotide polymorphism (SNP) between the two sequences.

Source data are available online for this figure.
Figure 5
Figure 5. Cis‐regulation of the RPP4 immune resistance gene by a ATCOPIA93‐derived, unmethylated soloLTR

Absence of H3K27m3 and H3K9m2 marks at the soloLTR‐5 [two primers sets (1) and (2)], upstream of RPP4. qPCRs were performed on ChIP‐DNA previously analyzed in Figs 3B and EV3C (bio.rep.1) to further validate the epigenetic status inferred from both H3K27m3 and H3K9m2 ChIP‐chip (Deleris et al, 2012) and H3K27m3 ChIP‐seq data for unique reads (Wang et al, 2016). UBQ: negative control for both H3K27m3 and H3K9m2 ChIPs; FLC and TA3: positive controls for H3K27m3 and H3K9m2 ChIP, respectively.

Depiction of the constructs used to transform the rpp4 null mutant to assess the impact of LTR mutations on RPP4 expression. Blue large bars: exons, blue medium bars: transcribed and untranslated regions (UTRs), purple bar: soloLTR‐5.

Box plot representing the mRNA levels of RPP4 in the presence/absence of the soloLTR‐5. 18 primary transformants were analyzed for the “pRPP4::RPP4 + mock” and “pRPP4::RPP4 + PtoΔ28E” datasets, and 18 primary transformants were analyzed for “pRPP4ΔLTR::RPP4 + mock” and “pRPP4ΔLTR::RPP4 + PtoΔ28E” datasets. Mock and PtoΔ28E (2 × 108 cfu/ml) infiltrations were performed on two leaves of each individual transformant that were collected at 6 h post‐infiltration (hpi). RNA was extracted for each transformant individually, for each treatment, and analyzed by RT–qPCR to determine the RPP4 mRNA levels relative to UBIQUITIN (At2g36060) expression. The values obtained for each primary transformant were plotted. The horizontal line in the box represents the median; the edges of the box represent the 25th and 75th percentiles, the whiskers stretch out to the 10–90 percentile above and below the edges of the box; the symbols (dots) represent the outliers. Two‐tailed P‐values were calculated by unpaired t‐test with Welch's correction. In addition, a nonparametric test (Mann–Whitney) was used and showed the same statistical differences (see Source data for P‐value summary).

Box plot representing the mRNA levels of RPP4 in the presence of the W‐box1 or the mutated W‐box1 (according to Fig 1E) in the soloLTR‐5. Twenty‐three primary transformants were analyzed for the “pRPP4::RPP4 + mock” and “pRPP4::RPP4 + PtoΔ28E” datasets, and 24 primary transformants were analyzed for “pRPP4w1::RPP4 + mock” and “pRPP4w1::RPP4 + PtoΔ28E” datasets. Analyses were as in (C) (t‐test and nonparametric test showing the same statistical differences, see Source data for P‐value summary).

The oomycete PAMP NLP20 induces the same molecular responses as PtoΔ28E. Top: Representative pictures of leaves infiltrated with water (mock), 1 μM of NLP20 or effectorless bacteria PtoΔ28 at 2 × 108 cfu/ml as a positive control and incubated with GUS substrate 24 hpi (two leaves from three plants per treatment). This result was repeated three times. Bottom: GUS mRNA levels at 3 hpi with NLP20 or PtoΔ28E. Analyses were performed as in Fig 1.

Box plot representing the mRNA levels of RPP4 in the presence/absence of the soloLTR‐5 and presence/absence of the W‐box1, in response to 1 μM of NLP20. Sixteen primary transformants for each construct were analyzed as in (C and D) (t‐test and nonparametric test show the same statistical differences, see Source data for P‐value summary).

Source data are available online for this figure.

Similar articles

See all similar articles

Cited by 4 articles

References

    1. Asai S, Shirasu K (2015) Plant cells under siege: plant immune system versus pathogen effectors. Curr Opin Plant Biol 28: 1–8 - PubMed
    1. Basenko EY, Sasaki T, Ji L, Prybol CJ, Burckhardt RM, Schmitz RJ, Lewis ZA (2015) Genome‐wide redistribution of H3K27me3 is linked to genotoxic stress and defective growth. Proc Natl Acad Sci USA 112: E6339–E6348 - PMC - PubMed
    1. Beguiristain T, Grandbastien M‐A, Puigdomènech P, Casacuberta JM (2001) Three Tnt1 subfamilies show different stress‐associated patterns of expression in tobacco. Consequences for retrotransposon control and evolution in plants. Plant Physiol 127: 212–221 - PMC - PubMed
    1. Bernatavichute YV, Zhang X, Cokus S, Pellegrini M, Jacobsen SE (2008) Genome‐wide association of histone H3 lysine nine methylation with CHG DNA methylation in Arabidopsis thaliana . PLoS One 3: e3156 - PMC - PubMed
    1. Böhm H, Albert I, Oome S, Raaymakers TM, Van den Ackerveken G, Nürnberger T (2014) A conserved peptide pattern from a widespread microbial virulence factor triggers pattern‐induced immunity in Arabidopsis . PLoS Pathog 10: e1004491 - PMC - PubMed

Publication types

LinkOut - more resources

Feedback