Sex-specific expression and function of TRIM28 during mouse primordial germ differentiation

Figures

• 

  Figure 1. TE accessibility changes during PGC differentiation (A) Alluvial plots showing the chromatin accessibility state of TEs (ERV, LINE, and SINE) overlapping with ATAC peaks in XY (left) and XX (right). (B) Motif enrichment of TEs transitioning from OC (left) or CO (right) in either sex. p values from HOMER. (C) Chromatin accessibility of OC (top) or CO (bottom) TEs, enrichment of H3K27ac in in vitro PGCs. Percentages represent peak overlap. (D) Percentage of TEs within families for which the largest proportion of elements are represented. Top, TEs open to close; bottom, TEs close to open. (E) Heatmap showing differential gene expression between E11.5 and E13.5 testicular PGCs (left) and ovarian PGCs (right). (F) Venn diagram showing sex-specific and shared KRAB-ZFPs. See also Figure S1; Table S5.

• 

  Figure 2. TRIM28 repression of TEs is sex specific (A) Immunofluorescence showing TRIM28 at E12.5, E13.5, and E16.5 in XY (left) and XX (right) samples. Germ cells are marked with VASA (green) and TRIM28 in magenta. Scale bars represent 20 μm. N ≥ 3 for all conditions. (B) Quantification of TRIM28 abundance in PGCs relative to gonadal soma. Each dot represents a biological replicate. Pairwise comparisons are t tests, and three-way comparisons are ANOVA tests. Error bars in SEM. (C) Heatmaps showing ERV accessibility in control and TRIM28 knockout PGCs at E10.5 and E14.5. (D) Profile plots of ChIP-seq data of TRIM28 at E12.5 (left) or H3K9me3 at E13.5 (right) at ERVs accessible in control PGCs (top) or TCKO PGCs (bottom). Graphs show input-subtracted normalized to read count. (E) H3K9me3 enrichment at LTRs identified at E10.5 (left), E14.5 (right), and E13.5. Significance tested using unpaired 2-sided t test. Boxplot shows median with hinges equal to 25th and 75th percentiles, whiskers are 1.5 interquartile range. (F) Proportional representation of TEs, which are upregulated (false discovery rate [FDR] < 0.005, Log2FC > 2), downregulated (FDR < 0.005, Log2FC < 0.25), or unchanged (log2fold…

• 

  Figure 3. TRIM28 suppresses emergence of a 2C-like transcriptome (A) Expression of select genes in E11.5, E12.5, and E13.5 control and TCKO PGCs in XY (top) and XX (bottom). Significance by DESeq2 (log2 fold change ≥ 2 or ≤ −2 and padj. < 0.05). (B) Normalized RNA-seq of MERVL-int and MT2_Mm at E11.5, E12.5, and E13.5 in control and TCKO PGCs. p value is Wilcoxon t test. (C) Correlation of Log2 fold change 2C transcriptome genes between mESCs/2CLCs (x axis) and control/TCKO PGCs (y axis) at E12.5. Colored dots are significant in TRIM28 DEG analysis. (D) Browser track plots of ATAC-seq at E14.5 in control and TCKO XY (top) and XX (bottom) PGCs at the Dux locus (left) and Zscan4c (right). RPKM normalized. (E) Boxplot of Oct4-eGFP+ PGCs isolated at indicated time point. Significance testing by t test. (F) Representative images of Ki67 and TRIM28 staining in testicular (XY) and ovarian (XX) PGCs. VASA marks germ cells. Scale bars represent 20 μm. (G) Quantification of background-corrected Ki67 intensity in E12.5 (top) and E13.5 (bottom) testicular and ovarian PGCs. Average ratio of each biological replicate. Significance testing by Welch’s t test. Error bars in SEM. At least, n = 3 biological…

• 

  Figure 4. TRIM28 is required for timely acquisition of competency (A) Representative images at E12.5 and E13.5 showing PGCs (VASA) and DAZL. Scale bars represent 20 μm. (B) Quantification of percentage of PGCs positive for DAZL and VASA or only VASA in testicular (XY) and ovarian (XX) PGCs at E12.5 (top) (XX control n = 4, XX TCKO n = 3, XY control n = 3, XY TCKO n = 3) and E13.5 (bottom) (XX control n = 4, XX TCKO n = 4, XY control n = 3, XY TCKO n = 3). Significance testing by Welch’s t test. Error bars in SEM. (C) Normalized expression of Dazl, Ddx4, and Prdm1. Error bars in SD. (D) Gene tracks showing abundance of TRIM28 over the Dazl locus (left) and Ddx4 locus (right) at E12.5. (E) Representative images of embryonic testes and ovaries with TFAP2C and VASA. Scale bars represent 20 μm. (F) Quantification of TFAP2C intensity over background in PGCs, each dot represents the mean of one biological replicate. Significance by Welch’s t test. Error bars in SEM. (G) Normalized expression of Tfap2c, Nanog, and Sox2 in E13.5 PGCs. Error bars in SD. For t tests, ns = p > 0.05, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗p < 0.0001. See also Figure S4.

• 

  Figure 5. TRIM28 is necessary for successful gametogenesis (A) Representative image of embryonic testes showing MILI (left) and MIWI2 (right) in pro-spermatogonia (VASA). Scale bars represent 20 μm. (B) Quantification of percentage of MILI+ (left) and MIWI2+ (right). Each point is one biological replicate. Significance by Welch’s t test. Error bars show SEM. (C) Volcano plot showing changes in expression in E13.5 testicular PGCs. Select genes are labeled. (D) Representative image of embryonic ovaries showing expression of SCP3 (left) and yH2Ax (right) in meiotic germ cells (VASA). Scale bars represent 20 μm. (E) Quantification of percent SCP3+ (left) and yH2Ax+ (right) in E16.5 ovaries. Each point is one biological replicate. Significance by Welch’s t test. Error bars show SEM. (F) Volcano plot showing changes in expression in E13.5 testicular PGCs. Select genes labeled. (G) Immunofluorescence images of week 7 (day 48) ovaries from 1 control female and 1 TCKO female. Ovarian soma (NR2F2), granulosa cells (FOXL2), and oocyte (DPPA3) are shown. Each TCKO ovary is shown. Scale bars represent 100 μm. (H) Measurement of ovary area from 2D image in cm2. Each dot is a single ovary from one…