Chromatin and siRNA pathways cooperate to maintain DNA methylation of small transposable elements in Arabidopsis

Abstract

Background: DNA methylation occurs at preferred sites in eukaryotes. In Arabidopsis, DNA cytosine methylation is maintained by three subfamilies of methyltransferases with distinct substrate specificities and different modes of action. Targeting of cytosine methylation at selected loci has been found to sometimes involve histone H3 methylation and small interfering (si)RNAs. However, the relationship between different cytosine methylation pathways and their preferred targets is not known.

Results: We used a microarray-based profiling method to explore the involvement of Arabidopsis CMT3 and DRM DNA methyltransferases, a histone H3 lysine-9 methyltransferase (KYP) and an Argonaute-related siRNA silencing component (AGO4) in methylating target loci. We found that KYP targets are also CMT3 targets, suggesting that histone methylation maintains CNG methylation genome-wide. CMT3 and KYP targets show similar proximal distributions that correspond to the overall distribution of transposable elements of all types, whereas DRM targets are distributed more distally along the chromosome. We find an inverse relationship between element size and loss of methylation in ago4 and drm mutants.

Conclusion: We conclude that the targets of both DNA methylation and histone H3K9 methylation pathways are transposable elements genome-wide, irrespective of element type and position. Our findings also suggest that RNA-directed DNA methylation is required to silence isolated elements that may be too small to be maintained in a silent state by a chromatin-based mechanism alone. Thus, parallel pathways would be needed to maintain silencing of transposable elements.

Figure 1

Raw data plots for the gene-oligo array. For each genotype pair, the average log₂(exp/ref) ratio is plotted versus the corresponding average log₂ fluorescent intensity. Each plot contains the results of six array measurements, that is dye-reversed measurements on three biological replicates. All data were lowess normalized as described in the Materials and methods section. Red dots represent statistically significant target loci, where those with positive log ratios indicate hypomethylation and those with negative log ratios indicate hypermethylation. Blue dots represent the rest of the loci.

Figure 2

CNG methylation targets of epigenetic silencing components. (a) Venn diagram summaries of positive loci using random-PCR arrays in cmt3, kyp, drm1/2 and ago4 mutant backgrounds. Loci were scored as positive if methylation was significantly changed in the indicated mutant relative to the Ler wild-type background. (b) Venn diagram summaries of positive loci using gene-oligo arrays, where cmt3, kyp, drm1/2 and ago4 were in a Ler (clk-st) and crm3 drm1/2 was in a Ws wild-type background. Gene-oligo and random-PCR datasets of targets are available with a graphical interface for browsing and for downloading [25]. (c) Table showing the number of positives and overlaps for each mutant class. Mutants are color coded for clarity in the Venn diagrams.

Figure 3

Location of transposable elements, CMT3, KYP and DRM1/2 targets along chromosome arms. (a) Transposable elements and CMT3 targets. (b) Comparison of KYP, DRM1/2 and CMT3 DRM1/2 targets to CMT3 targets. (c) Comparison of all CMT3 targets to the subset of CMT3 targets that are not also KYP targets, and comparison of CMT3 targets to the subset that are not also CMT3 DRM1/2 targets. To map repeats relative to the centromere, Repbase library sequences were searched using BLASTN with default Repbase parameters against TIGR Release 5 of the Arabidopsis genome sequence. All CMT3 targets, single-copy CMT3 targets and DRM1/2 targets from the gene-oligo array (Figure 2b) were also mapped on the same scale. The fraction of the total number of hits within each 1 Mb bin is shown. To compensate for differences in oligo abundance on the array, bins were normalized by dividing each raw fraction by the fraction of oligos in the bin.

Figure 4

Methylation occupancies of selected target loci determined by bisulfite sequencing. Elements are: (a) Mu-PCR (locus 4:1813417-1814107 on random-PCR array hypomethylated in cmt3, kyp and cmt3 drm1/2); (b) 229-R1 (left side of A000229 on gene-oligo array hypomethylated in cmt3 and drm1/2); (c) 229-R2 (right side of A000229); (d) Mu-4802 (A004802 hypomethylated in cmt3, kyp and cmt3 drm1/2); (e) TA11-4217 (A004217 hypomethylated in cmt3); (f) SINE3-5300 (A005300); (g) SINE3-11193 (A011193 hypomethylated in drm1/2). Wild-type lines are Ler, clk-st (parental line of cmt3, kyp, and ago4 derived from Ler) and Ws (parental line of drm1/2). See Table 3 for details.

Figure 5

Methylation by DRM1/2 and AGO4 is associated with the size of their targets. (a) The loss of methylation for each locus is calculated from the reduction seen in drm1/2 and ago4 when measured by bisulfite sequencing (drm1/2: correlation coefficient r = 0.82, p < 0.003; ago4: r = 0.90, p = 0.0002). The fraction methylated is the ratio of mutant to wild-type percentages listed in Table 3. Regression lines are shown for clarity. (b) A similar comparison of CMT3 and KYP reveals no significant associations (cmt3: r = -0.48, p = 0.2; kyp: r = -0.32, p = 0.5), so no regression lines are shown. The comparisons include data reported in this study supplemented with previously published data for other loci [11,12].