Figure 3. TEs are mis-expressed in the absence of Stella.

(A) Top panel shows boxplots of the ratio of reads mapped to TEs to total reads mapped in WT and KO oocytes, 1-cell and 2-cell embryos. The number of reads mapped to TEs are based on uniquely and multi-mapped counts (Figure 3—source data 2). (B) A heatmap of the relative expression of LTR, LINE and SINE in oocyte (green), 1-cell (yellow) and 2-cell embryos (red). Blue indicates higher expression and white indicates lower expression. Samples are clustered by row. (C) A score plot of the first three principal components for 66 cells using uniquely mapped TE counts (Figure 3—source data 1). The lower panels represent the score plots of the first two principal components using cells belonging to a specific developmental time point. The developmental time points are indicated by colour and the genotypes of Stella are indicated by shape. (D) A bar chart of the odds ratio of TEs up and downregulated in KO 2-cell embryos compared to WT, intersected with ‘maternal TEs’ and ‘zygotic TEs’. For maternal TEs enriched in TEs upregulated in KO 2-cell, ***p=5.35 × 10⁻⁶; for zygotic TEs enriched in TEs downregulated in KO 2-cell, ***p=8.13 × 10⁻⁷. (E) Bar plots showing the relative expression of TE families (LTR-ERVL, LINE1-L1 and SINE-Alu) in WT (white) and KO (light blue) oocytes, 1-cell and 2-cell embryos, data analysed from single-cell / embryo RNA sequencing. Two-sided Wilcoxon rank sum test performed between WT and KO samples and statistically significant p-values stated. (F) Top is an illustration of the structure of the full-length MuERV-L element flanked by 5’ and 3’ LTRs. Bottom is boxplots of the relative expression (counts per million) of MuERV-L Int and MT2_Mm transcript in WT (white) and KO (light-blue) oocytes, 1-cell and 2-cell embryos (p<0.05 corresponds to * and p<0.01 corresponds to **). (G) Immunofluorescence staining against MuERV-L Gag antibody in 2-cell embryos. (Left) The top panel shows bright-field, middle panel is MuERV-L Gag (green) and bottom panel are merged images of MuERV-L and DAPI, which counterstains DNA. Representative projections of z-stacks are shown for WT and KO embryos. (Right) A boxplot of the z-stack quantifications of the relative fluorescence intensity of MuERV-L Gag protein between WT and KO 2-cell embryos. Also see Figure 3—figure supplements 1–3 and Figure 3—source data 1–2.

DOI: http://dx.doi.org/10.7554/eLife.22345.014

Figure 3—figure supplement 1. TEs are mis-expressed in the absence of Stella.

(A) Heatmaps of the relative expression of individual families of LTR, LINE and SINE elements in oocyte (green), 1-cell (yellow) and 2-cell (red) embryos. Samples are clustered by row. Blue indicated relatively high expression while white indicates relatively low expression of TEs. (B) Bar charts of relative expression of LTR elements, LINE and SINE in WT (white) and KO (light blue) oocytes, 1-cell and 2-cell embryos. Two-sided Wilcoxon rank sum test performed between WT and KO samples and statistically significant p-values are stated. (C) Validation of MuERV-L Gag antibody. (Left) IF staining of an ESC reporter line for MuERV-L – 2C::tdTomato (Macfarlan et al., 2012), illustrating the MuERV-L Gag (GFP) staining overlaps well with tdTomato⁺ ESC. (Right) IF staining showing robust detection of MuERV-L Gag (GFP) in 2-cell embryos, while it is virtually undetectable in metaphase II (MII) oocytes. DNA is counterstained with DAPI.

DOI: http://dx.doi.org/10.7554/eLife.22345.017

Figure 3—figure supplement 2. Dppa3 does not affect MuERV-L expression in mESCs.

(A) Left shows histogram plots of tdTomato expression in 2C::tdTomato mESCs 48 hr and 96 hr after addition of DOX to induce Dppa3 expression. Red represents tdTomato frequency with Dppa3 over-expression (OE), while blue represents empty vector transfection control. The horizontal bars depict the expression threshold set for dtTomato and the proportion of tdTomato⁺ cells in each condition is stated. Right shows a bar plot of Dppa3 expression 48 hr and 96 hr after addition of DOX. Dppa3 expression is normalized to a housekeeping gene (GAPDH) and relative to empty vector sample. (B) Left shows histogram plots of tdTomato expression in 2C::tdTomato mESCs 48 hr and 96 hr after siRNA transfection. Red represents tdTomato frequency with Dppa3 siRNA, while blue represents scramble siRNA control. The bar depicts the expression threshold set for dtTomato and the proportion of tdTomato⁺ cells in each condition is stated. Right shows a bar plot of Dppa3 expression 48 hr and 96 hr after siRNA transfection. Dppa3 expression is normalized to a housekeeping gene (GAPDH) and relative to scramble siRNA sample.

DOI: http://dx.doi.org/10.7554/eLife.22345.018

Figure 3—figure supplement 3. Quality control of TE reads mapping and normalization.

(A) A bar plot of the proportion of multi-mapped TE reads (Figure 3—source data 2) which can be unambiguously assigned at the level of class (repClass), family (repFamily) and element (repName) based on the RepeatMasker annotation provided by UCSC table browser. (B) A heatmap of the expression of LTR, LINE and SINE in oocyte (green), 1-cell (yellow) and 2-cell (red) embryos based on between-sample normalization method (See Materials and methods). Samples are clustered by row.

DOI: http://dx.doi.org/10.7554/eLife.22345.019