Herberg S et al. (2015), Histone H3 lysine 9 trimethylation is required ...

XB-ART-51302

Sci Rep 2015 Sep 21;5:14236. doi: 10.1038/srep14236.

Show Gene links Show Anatomy links

Histone H3 lysine 9 trimethylation is required for suppressing the expression of an embryonically activated retrotransposon in Xenopus laevis.

Herberg S , Simeone A , Oikawa M , Jullien J , Bradshaw CR , Teperek M , Gurdon J , Miyamoto K .

???displayArticle.abstract???
Transposable elements in the genome are generally silenced in differentiated somatic cells. However, increasing evidence indicates that some of them are actively transcribed in early embryos and the proper regulation of retrotransposon expression is essential for normal development. Although their developmentally regulated expression has been shown, the mechanisms controlling retrotransposon expression in early embryos are still not well understood. Here, we observe a dynamic expression pattern of retrotransposons with three out of ten examined retrotransposons (1a11, λ-olt 2-1 and xretpos(L)) being transcribed solely during early embryonic development. We also identified a transcript that contains the long terminal repeat (LTR) of λ-olt 2-1 and shows a similar expression pattern to λ-olt 2-1 in early Xenopus embryos. All three retrotransposons are transcribed by RNA polymerase II. Although their expression levels decline during development, the LTRs are marked by histone H3 lysine 4 trimethylation. Furthermore, retrotransposons, especially λ-olt 2-1, are enriched with histone H3 lysine 9 trimethylation (H3K9me3) when their expression is repressed. Overexpression of lysine-specific demethylase 4d removes H3K9me3 marks from Xenopus embryos and inhibits the repression of λ-olt 2-1 after gastrulation. Thus, our study shows that H3K9me3 is important for silencing the developmentally regulated retrotransposon in Xenopus laevis.

???displayArticle.pubmedLink??? 26387861
???displayArticle.pmcLink??? PMC4585706
???displayArticle.link??? Sci Rep
???displayArticle.grants??? [+]

Species referenced: Xenopus laevis
Genes referenced: 10a1.1 1a11 cadm1 mars2 myc pwp1 sox17a

???attribute.lit??? ???displayArticles.show???

	Figure 1. Developmentally regulated expression of retrotransposons 1a11, λ-olt 2-1 and xretpos(L) during Xenopus laevis embryonic development.(a) Schematic diagram of expression analysis during embryonic development. Embryos were collected at different stages. After RNA extraction and reverse transcription the transcript level was determined by qPCR. The container was drawn by S.H. (b) Relative changes of the transcript levels of 1a11, λ-olt 2-1 and xretpos(L) during embryogenesis in comparison to their transcript levels at the one-cell stage (St1 = 1) (n = 3–14). pwp1 is a constantly expressed gene that was used as control (n = 1–4). All values were normalized to the overall RNA concentration of the respective sample. Sample RT- is a negative control. Error bars represent SEM. *P = 0.005, P = 0.021. (c) Schematic diagram of the accumulated transcript levels of 1a11, λ-olt 2-1 and xretpos(L) during embryogenesis, explaining that their expression is upregulated after midblastula transition and shows the highest level at the late gastrula stage. (d) 1a11, λ-olt 2-1 and xretpos(L) are mainly detected as sense transcripts although antisense transcription can also be found. Forward primers were used for reverse transcription to analyze the expression of antisense transcripts from retrotransposons (blue bars), while reverse primers were used as a control (red bars). Upper graphs summarize antisense transcription at stages 1 and 12.5. Lower graphs represent the comparison between sense and antisense transcription although antisense transcription is almost invisible due to its much weaker expression than sense transcription. Relative transcript levels were compared to the transcript levels at the one-cell stage (St1 = 1) (n = 4). (e) Normalization by the cell number also indicates the upregulated expression of retrotransposons after midblastula transition and the downregulated expression after the gastrula stage. Relative changes of the transcript levels of 1a11, λ-olt 2-1, xretpos(L) and pwp1 per cell during embryogenesis in comparison to their transcript levels at the one-cell stage (St1 = 1) (n = 3–13). Error bars represent SEM. Only values from stage 7 onwards are shown, since the increase of the cell number is less constant during the first six stages (Supplementary Fig. S3).
	Figure 2. The differential expression pattern between mars2-like and conventional mars2 during embryonic development.(a) The structure of λ-olt 2-1, mars2-like and conventional mars2. Grey boxes represent exons and the white box represents the LTR of λ-olt 2-1. The first exon of mars2-like and its surrounding region show 97% sequence identity to the LTR of λ-olt 2-1; the second exon shows 87% sequence identity to the third exon of conventional Xenopus mars2; the third exon shows no sequence identity to any reported gene. The full nucleotide sequence of mars2-like is shown in Supplementary Figure S5. Arrows represent binding sites of primer pairs used for qPCR. The forward primer of primer pair 1 binds to the splicing junction between exon 1 and exon 2 of mars2-like. Primer 2 binds to exon1 of conventional mars2. (b) Changes in the transcript levels of mars2-like and conventional mars2 during embryogenesis in comparison to their transcript levels at stage 28/29 (St28/29 = 1). Stage 28/29 was used as a reference here instead of stage 1, since the transcript level of conventional mars2 is already very high at stage 1. The figure at the bottom shows a merge of mars2-like (primer 1) and conventional mars2 (primer 2). All values were normalized to the overall RNA concentration of the respective sample. Sample RT- is a negative control. All error bars represent SEM. n = 3–7.
	Figure 3. 1a11, λ-olt 2-1, xretpos(L) and mars2-like are transcribed by RNA polymerase II.(a) Schematic diagram of the transcriptional inhibition experiment. 0.1 ng or 2 ng of α-amanitin were injected into one-cell stage embryos. Embryos were collected at the blastula stage (stage 9) and RT-qPCR was performed. The container was drawn by S.H. (b) and (c) The effect of α-amanitin injection on the transcript levels of sox17, pwp1, 1a11, λ-olt 2-1, xretpos(L), and mars2-like. Transcript level of non-injected embryos were used as a reference (non = 1). All values were normalized to the RNA concentration of the respective sample. Sample RT- is a negative control. All error bars represent SEM. n = 3. P < 0.05, *P < 0.005.
	Figure 4. KDM4d overexpression removes H3K9me3 in embryos and upregulates λ-olt 2-1 at the neurula stage.(a) ChIP analysis indicates enrichment of H3K4me3 and H3K9me3 on LTRs of retrotransposons at different stages of embryogenesis. The precipitated DNA/input DNA of 1a11, λ-olt 2-1 and xretpos(L) was determined by qPCR. (H3K4me3 and H3K9me3: n = 3–4, IgG: n = 2). All values of modified histone and control IgG were normalized to H3 (Supplementary Fig. 6b). All error bars represent SEM. *P = 0.00008, P < 0.05. Blue = H3K4me3, red = H3K9me3, and green = IgG. (b) KDM4d mRNA was injected into fertilized embryos to remove H3K9me3 during embryogenesis. At the gastrula stage (stage 10) and at the neurula stage (stage 17) embryos were collected and prepared for western blotting or qPCR. Myc-GFP mRNA was used as a control. The container was drawn by S.H. (c) Removal of H3K9me3 by KDM4d mRNA injection was verified by western blot. The same samples were used to detect the expression of HP1β in Xenopus embryos. Histone 4 (H4) was used as a loading control. (d) The effect of KDM4d overexpression on retrotransposons expression and on the expression of mars2-like and conventional mars2 at the gastrula (St10) and the neurula stages (St17). Transcription was measured by RT-qPCR. The relative change of the transcript level in comparison to the transcript level of control myc-GFP mRNA-injected embryos is shown. The transcript level of myc-GFP mRNA-injected embryos at stage 10 was set 1. All values were normalized to the RNA concentration of the sample. All error bars represent SEM. n = 3–5. *P < 0.05. Blue bars represent myc-GFP mRNA-injected embryos while red bars are KDM4d mRNA-injected embryos.

References [+] :

Akkers, A hierarchy of H3K4me3 and H3K27me3 acquisition in spatial gene regulation in Xenopus embryos. 2009, Pubmed, Xenbase

	Figure 1. Developmentally regulated expression of retrotransposons 1a11, λ-olt 2-1 and xretpos(L) during Xenopus laevis embryonic development.(a) Schematic diagram of expression analysis during embryonic development. Embryos were collected at different stages. After RNA extraction and reverse transcription the transcript level was determined by qPCR. The container was drawn by S.H. (b) Relative changes of the transcript levels of 1a11, λ-olt 2-1 and xretpos(L) during embryogenesis in comparison to their transcript levels at the one-cell stage (St1 = 1) (n = 3–14). pwp1 is a constantly expressed gene that was used as control (n = 1–4). All values were normalized to the overall RNA concentration of the respective sample. Sample RT- is a negative control. Error bars represent SEM. *P = 0.005, P = 0.021. (c) Schematic diagram of the accumulated transcript levels of 1a11, λ-olt 2-1 and xretpos(L) during embryogenesis, explaining that their expression is upregulated after midblastula transition and shows the highest level at the late gastrula stage. (d) 1a11, λ-olt 2-1 and xretpos(L) are mainly detected as sense transcripts although antisense transcription can also be found. Forward primers were used for reverse transcription to analyze the expression of antisense transcripts from retrotransposons (blue bars), while reverse primers were used as a control (red bars). Upper graphs summarize antisense transcription at stages 1 and 12.5. Lower graphs represent the comparison between sense and antisense transcription although antisense transcription is almost invisible due to its much weaker expression than sense transcription. Relative transcript levels were compared to the transcript levels at the one-cell stage (St1 = 1) (n = 4). (e) Normalization by the cell number also indicates the upregulated expression of retrotransposons after midblastula transition and the downregulated expression after the gastrula stage. Relative changes of the transcript levels of 1a11, λ-olt 2-1, xretpos(L) and pwp1 per cell during embryogenesis in comparison to their transcript levels at the one-cell stage (St1 = 1) (n = 3–13). Error bars represent SEM. Only values from stage 7 onwards are shown, since the increase of the cell number is less constant during the first six stages (Supplementary Fig. S3).
	Figure 2. The differential expression pattern between mars2-like and conventional mars2 during embryonic development.(a) The structure of λ-olt 2-1, mars2-like and conventional mars2. Grey boxes represent exons and the white box represents the LTR of λ-olt 2-1. The first exon of mars2-like and its surrounding region show 97% sequence identity to the LTR of λ-olt 2-1; the second exon shows 87% sequence identity to the third exon of conventional Xenopus mars2; the third exon shows no sequence identity to any reported gene. The full nucleotide sequence of mars2-like is shown in Supplementary Figure S5. Arrows represent binding sites of primer pairs used for qPCR. The forward primer of primer pair 1 binds to the splicing junction between exon 1 and exon 2 of mars2-like. Primer 2 binds to exon1 of conventional mars2. (b) Changes in the transcript levels of mars2-like and conventional mars2 during embryogenesis in comparison to their transcript levels at stage 28/29 (St28/29 = 1). Stage 28/29 was used as a reference here instead of stage 1, since the transcript level of conventional mars2 is already very high at stage 1. The figure at the bottom shows a merge of mars2-like (primer 1) and conventional mars2 (primer 2). All values were normalized to the overall RNA concentration of the respective sample. Sample RT- is a negative control. All error bars represent SEM. n = 3–7.
	Figure 3. 1a11, λ-olt 2-1, xretpos(L) and mars2-like are transcribed by RNA polymerase II.(a) Schematic diagram of the transcriptional inhibition experiment. 0.1 ng or 2 ng of α-amanitin were injected into one-cell stage embryos. Embryos were collected at the blastula stage (stage 9) and RT-qPCR was performed. The container was drawn by S.H. (b) and (c) The effect of α-amanitin injection on the transcript levels of sox17, pwp1, 1a11, λ-olt 2-1, xretpos(L), and mars2-like. Transcript level of non-injected embryos were used as a reference (non = 1). All values were normalized to the RNA concentration of the respective sample. Sample RT- is a negative control. All error bars represent SEM. n = 3. P < 0.05, *P < 0.005.
	Figure 4. KDM4d overexpression removes H3K9me3 in embryos and upregulates λ-olt 2-1 at the neurula stage.(a) ChIP analysis indicates enrichment of H3K4me3 and H3K9me3 on LTRs of retrotransposons at different stages of embryogenesis. The precipitated DNA/input DNA of 1a11, λ-olt 2-1 and xretpos(L) was determined by qPCR. (H3K4me3 and H3K9me3: n = 3–4, IgG: n = 2). All values of modified histone and control IgG were normalized to H3 (Supplementary Fig. 6b). All error bars represent SEM. *P = 0.00008, P < 0.05. Blue = H3K4me3, red = H3K9me3, and green = IgG. (b) KDM4d mRNA was injected into fertilized embryos to remove H3K9me3 during embryogenesis. At the gastrula stage (stage 10) and at the neurula stage (stage 17) embryos were collected and prepared for western blotting or qPCR. Myc-GFP mRNA was used as a control. The container was drawn by S.H. (c) Removal of H3K9me3 by KDM4d mRNA injection was verified by western blot. The same samples were used to detect the expression of HP1β in Xenopus embryos. Histone 4 (H4) was used as a loading control. (d) The effect of KDM4d overexpression on retrotransposons expression and on the expression of mars2-like and conventional mars2 at the gastrula (St10) and the neurula stages (St17). Transcription was measured by RT-qPCR. The relative change of the transcript level in comparison to the transcript level of control myc-GFP mRNA-injected embryos is shown. The transcript level of myc-GFP mRNA-injected embryos at stage 10 was set 1. All values were normalized to the RNA concentration of the sample. All error bars represent SEM. n = 3–5. *P < 0.05. Blue bars represent myc-GFP mRNA-injected embryos while red bars are KDM4d mRNA-injected embryos.