Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Nov 2;21(5):694-703.e7.
doi: 10.1016/j.stem.2017.10.004.

An endosiRNA-Based Repression Mechanism Counteracts Transposon Activation During Global DNA Demethylation in Embryonic Stem Cells

Affiliations
Free PMC article

An endosiRNA-Based Repression Mechanism Counteracts Transposon Activation During Global DNA Demethylation in Embryonic Stem Cells

Rebecca V Berrens et al. Cell Stem Cell. .
Free PMC article

Abstract

Erasure of DNA methylation and repressive chromatin marks in the mammalian germline leads to risk of transcriptional activation of transposable elements (TEs). Here, we used mouse embryonic stem cells (ESCs) to identify an endosiRNA-based mechanism involved in suppression of TE transcription. In ESCs with DNA demethylation induced by acute deletion of Dnmt1, we saw an increase in sense transcription at TEs, resulting in an abundance of sense/antisense transcripts leading to high levels of ARGONAUTE2 (AGO2)-bound small RNAs. Inhibition of Dicer or Ago2 expression revealed that small RNAs are involved in an immediate response to demethylation-induced transposon activation, while the deposition of repressive histone marks follows as a chronic response. In vivo, we also found TE-specific endosiRNAs present during primordial germ cell development. Our results suggest that antisense TE transcription is a "trap" that elicits an endosiRNA response to restrain acute transposon activity during epigenetic reprogramming in the mammalian germline.

Keywords: DNMT1; IAP elements; RNAi; endogenous retroviruses; germ line; primordial germ cell; repeats; small RNAs; transposable element.

Figures

None
Figure 1
Figure 1
Transcriptional Upregulation of Specific TE Classes upon Acute Dnmt1 Deletion (A) Left: schematic overview of epigenetic reprogramming during preimplantation and male (blue) and female (red) germline development. Right: schematic of Dnmt1 cKO as an in vitro system for mechanistic study of TE regulation during epigenetic reprogramming. (B) Violin plots showing the distribution of CpG methylation levels measured by WGBS-seq of WT (gray) and conditional Dnmt1 cKO ESCs induced for days depicted in the figure. The percentage of methylated cytosines was quantified in consecutive 50 CpG windows genome-wide. CGI, CpG island. For significance analysis, Wilcoxon rank-sum test with Bonferroni correction testing with a p value threshold of <0.05. (C) Heatmap of unbiased hierarchical clustering of all TEs responsive to Dnmt1 cKO across the time course of KO induction. The relative expression (Z score) of TEs upon Dnmt1 cKO is shown; n = 2. (D) Bar graph showing the percentage of genic antisense transcription upon Dnmt1 deletion in KO relative to WT samples; n = 2. (E) Chromosome view of TE inserted antisense to gene. Position of TE is denoted (top panel) along with sense strand-specific RNA-seq reads (lower panels; sense transcription shown in blue; antisense transcription shown in red). Each read is depicted. Arrows indicate directionality of reads. (F) Expression of TEs in conditional Dnmt1 cKO ESC. Shown are normalized RNA-seq read counts overlapping different TE classes in sense (filled bars) or antisense (hatched bars) orientation. The figure shows mean of n = 2. See also Figures S1 and S4I and Data S1.
Figure 2
Figure 2
Generation of TE-Derived Small RNAs following Global Demethylation (A) Schematic displaying the hypothesis of pervasive transcription overlapping TEs acting as a “trap” of transcriptional activation of TEs. This could work through the production of dsRNAs from sense and antisense transcripts that feed into the RNAi pathway, which subsequently silences the TEs. (B) Small RNA-seq reads mapped to different classes of TEs from WT (gray) and conditional Dnmt1 cKO ESCs. p < 0.05, ∗∗p < 0.005, two-tailed Student’s t test. Bars represent mean ± SD, n = 3. All reads of a size between 20 and 24 nt were mapped to TE consensus sequences. (C) Small RNA-seq reads mapped to the consensus sequence of IAPEZ. All reads of a size between 20 and 36 nt were mapped to the IAPEZ consensus sequence. (D) Schematic displaying AGO2 IP of small RNAs. (E) Size distribution of AGO2-bound small RNAs after AGO2 IP of sense (black) and antisense (gray) small RNAs mapping to repeatmasker consensus sequences using the piPipes small RNA-seq pipeline (Han et al., 2015). (F) Small RNA-seq of AGO2-bound small RNAs mapped to TE classes of WT (gray) and conditional Dnmt1 cKO ESCs induced for 9 days (light blue). p < 0.05, ∗∗p < 0.005, two-tailed Student’s t test. Bars represent mean ± SD, n = 4. See also Figures S2 and S4I and Data S1.
Figure 3
Figure 3
TEs Are Repressed by a DICER Mechanism (A) Knockdown (KD) of RNAi players. Left: schematic of siRNA KD in Dnmt1 cKO ESCs. The genome gets demethylated (5mC, orange) and IAPs get transcriptionally activated and resilenced (red) if small RNAs are present (gray); however, KD of the RNAi pathway will deplete small RNAs. Lower right: quantitative real-time PCR analysis showing KD efficiencies of Dicer, Ago2, and Dgcr8 upon treatment with siRNAs after Dnmt1 deletion. Upper right: expression of IAPs upon Dicer, Ago2, Dgcr8, or non-targeting siRNA transfection. The data are normalized to non-targeting control. Bars represent mean ± SD, n = 3. p < 0.05, ∗∗p < 0.005, two-tailed Student’s t test. (B) Small RNA-seq of Dicer/Dnmt1 cDKO and Dnmt1 cKO ESCs. Sense (orange) and antisense (blue) small RNAs are separated by size and were mapped to all TEs. Reads were normalized to non-induced WT (Dicerfl/fl/Dnmt1fl/fl) ESCs. (C) Quantitative real-time PCR analysis of TE classes in ESCs following conditional Dnmt1 cKO or Dnmt1/Dicer cDKO by treatment with 4OHT or Dicer KO. Bars represent mean of two biological replicates with two technical replicates. Values were normalized to Atp5b and Hspcb, and major satellites were normalized to U1. p < 0.05, ∗∗p < 0.005, two-tailed Student’s t test. (D) Quantitative real-time PCR analysis of IAPEz in the indicated ESC lines. Conditional deletions were induced by treatment with 4OHT for the indicated days. Values were normalized to Atp5b and Hspcb and are relative to the respective WT sample for each KO line, indicated by dashed line. Error bars represent mean ± SD; n = 3 for Dnmt1 cKO, Dicer KO/Dnmt1 cKO, and Ago2 KO/Dnmt1 cKO; n = 2 for Dicer/Dnmt1 cDKO and Ago2/Dnmt1 cDKO. Ago2 KO/Dnmt1 cKO time points days 9 and 11 were not collected. (E) Heatmap of unbiased hierarchical clustering of all TE classes responsive to Dicer KO. Heatmap depicts relative expression (Z score) of TEs upon Dicer KO. See also Figures S3 and S4I and Tables S2 and S3.
Figure 4
Figure 4
Repressive Histone Modifications Enriched at TEs upon Global Demethylation (A) Heatmap showing relative enrichment (Z score) of repressive histone marks (H3K9me3, H3K27me3, and H3K9me2) at TE classes differentially regulated upon both Dicer KO (Figure 3A) and Dnmt1 cKO (Figure 1C) and normalized to enrichment in WT ESCs upon acute deletion of Dnmt1. (B) H3K27me3, H3K9me3, and H3K9me2 enrichment over TEs dependent on Dicer and Dnmt1. Heatmap depicts ChIP-seq data of H3K27me3 mapped to TE families at depicted days after Dnmt1 cKO, Dicer KO, and Dnmt1/Dicer cDKO in comparison to WT ESCs. (C) Schematic of the two levels of TE control upon global demethylation. Upon deletion of Dnmt1, DNA methylation (5mC; orange)-mediated repression is lost, and transposon expression increases (as an example, IAP expression is shown in green). Subsequently, small RNAs (red; “immediate” response) and repressive histone marks (chromatin, blue; “chronic” response) establish a new repressive environment. Also indicated are the time points at which the different experimental manipulations interfere with the system. See also Figure S4 and Data S1.

Similar articles

See all similar articles

Cited by 12 articles

See all "Cited by" articles

References

    1. Aravin A.A., Sachidanandam R., Bourc’his D., Schaefer C., Pezic D., Toth K.F., Bestor T., Hannon G.J. A piRNA pathway primed by individual transposons is linked to de novo DNA methylation in mice. Mol. Cell. 2008;31:785–799. - PMC - PubMed
    1. Babiarz J.E., Ruby J.G., Wang Y., Bartel D.P., Blelloch R. Mouse ES cells express endogenous shRNAs, siRNAs, and other Microprocessor-independent, Dicer-dependent small RNAs. Genes Dev. 2008;22:2773–2785. - PMC - PubMed
    1. Bodak M., Cirera-Salinas D., Yu J., Ngondo R.P., Ciaudo C. Dicer, a new regulator of pluripotency exit and LINE-1 elements in mouse embryonic stem cells. FEBS Open Bio. 2017;7:204–220. - PMC - PubMed
    1. Chen C.Y., Morris Q., Mitchell J.A. Enhancer identification in mouse embryonic stem cells using integrative modeling of chromatin and genomic features. BMC Genomics. 2012;13:152. - PMC - PubMed
    1. Creyghton M.P., Cheng A.W., Welstead G.G., Kooistra T., Carey B.W., Steine E.J., Hanna J., Lodato M.A., Frampton G.M., Sharp P.A. Histone H3K27ac separates active from poised enhancers and predicts developmental state. Proc. Natl. Acad. Sci. USA. 2010;107:21931–21936. - PMC - PubMed

LinkOut - more resources

Feedback