Supplemental Information. Antagonistic Activities of Sox2. and Brachyury Control the Fate Choice. of Neuro-Mesodermal Progenitors

Transcription

1 Developmental Cell, Volume 42 Supplemental Information Antagonistic Activities of Sox2 and Brachyury Control the Fate Choice of Neuro-Mesodermal Progenitors Frederic Koch, Manuela Scholze, Lars Wittler, Dennis Schifferl, Smita Sudheer, Phillip Grote, Bernd Timmermann, Karol Macura, and Bernhard G. Herrmann

2 A Koch_Figure S1 1.4% / 0.2% 2.5% B / Group I Wnt3a Fgf8 Rspo3 Group II Group III Group IV Group V + - / Lef1 Foxc1 Foxf1 Osr1 Sp8 Nkx1-2 Sox1 Sox3 C 2 Pseudotime Component Component 1

3 Figure S1. Related to Figure 1. (A) FACS plot of cells from hearts of / embryos, used to set-up the gating in Figure 1B. (B) Genome browser snapshots showing normalized read densities across examples of differentially expressed genes clustered into groups I-V. (C) Same PCA plot showing the cell trajectory as in Figure 1F, but colored according to pseudotime.

4 Koch_Figure S2 A B d2 d3 / / 2% 83% 98% 17% C / JM8 JM8 T Sox2 Tbx6 * H3 * 70kDa 55kDa 55kDa 40kDa 55kDa 40kDa 25kDa 15kDa D Group I Group II Group III Group IV Group V Sox2 T Wnt3a Lef1 Tbx3 Sp8 Sox3 E peaks 228 genes 542 peaks 319 genes Frequency Distance from T to Closest Sox2 Peak Maximum (bp)

5 Figure S2. Related to Figure 2. (A) Fluorescence microscopy of cells containing and reporters, differentiated in vitro from ES cells to NMPs. The majority of reporter-positive cells are /. Scale bars, 50µm. (B) FACS analysis of cells containing and reporters, differentiated in vitro from ES cells to NMPs either at day 2 (d2, left) or day 3 (d3, right) of differentiation. (C) Western blotting confirming that both T and Sox2 protein are co-expressed at day 3 (d3) of differentiation marking the NMP stage, in wild-type JM8 cells () (right) as well as in cells containing and reporters (left). The expression of Tbx6 at d3 and d4 indicates that the NMP stage is transient, followed by differentiation to (presomitic) mesoderm. Controls are undifferentiated ES-cells grown in 2i + LIF medium (d0); histone H3 was used as a loading control; * indicates background bands. (D) Genome browser screenshots showing T and Sox2 -Seq read densities across exemplary genes from groups I-V and their surrounding genomic regions. (E) To investigate whether T and Sox2 co-bind the same potential regulatory elements at the 574 shared putative target genes (see Figure 2C), the distances from the maxima of T peaks (n=1990) to the closest Sox2 peak associated with the same gene were retrieved. In total, 324 and 542 (314 and 524 when correcting for peak associations to two genes due to intergenic binding) are found within 1kb and 5kb respectively, corresponding to 228 and 319 of the 574 putative target genes.

6 Koch_Figure S3 A Brightfield merge B Neg. (head) CE TS % 1.3% 66.0% 31.1% T 2J/2J T 2J/2J 99.2% 0.6% 47.4% 49.4% C n = 4, ,541 2,372 D Sox2 Promoter Intergenic Genic n = 2, e T bits 1 0 motif frequency distance from peak summit (bp) β-catenin Rspo3

7 Figure S3. Related to Figure 3. (A) Fluorescence microscopy pictures of embryos obtained by tetraploid complementation using either wild-type () or mutant T 2J/2J ES cells containing the reporter, verifying the correct reporter expression in the posterior mesoderm. Scale bars, 500µm. (B) FACS profiles of positive cells sorted from caudal ends (CE) of wild-type () or T 2J/2 embryos of stage TS13, used for differential RNA-Seq analysis. Forebrains (heads) were used as a negative control and the relative amounts of cells in each subpopulation in and T 2J/2J mutants are indicated in percentages. (C) Analysis of β-catenin -Seq data showing the distribution of binding sites, the de novo motif analysis and the frequency distribution of that motif around the peak maxima. (D) Exemplary genome browser screenshot showing Sox2, T and β-catenin binding near the Rspo3 locus.

8 A B Koch_Figure S4 Neg. (heart) CE TS12 CE TS13 T 2J/2J T 2J/2J 0% 0.1% / 0% 1.3% / 0% 51.1% 31.1% / 26.0% 9.7% / 26.3% 53.7% / 17.6% 10.7% 26.3% / 16.4% 0.9% 11.3% 14.4% C TS12 TS13 T 2J/2J normalized frequency (x10,000) normalized frequency (x10,000) / / bp 200bp 400bp 600bp 0bp 200bp 400bp 600bp 0bp 200bp 400bp 600bp 0bp 200bp 400bp 600bp 0bp 200bp 400bp 600bp 0bp 200bp 400bp 600bp bp 200bp 400bp 600bp 0bp 200bp 400bp 600bp 0bp 200bp 400bp 600bp 0bp 200bp 400bp 600bp 0bp 200bp 400bp 600bp / / bp 200bp 400bp 600bp D Before Normalization 1600 After Normalization Mass Mass

9 Figure S4. Related to Figure 4. (A) Fluorescence microscopy pictures of embryos obtained by tetraploid complementation using either wild-type () or T 2J/2J ES cells containing the and reporters, verifying the correct expression patterns. Scale bars, 500µm. (B) FACS profiles of and/or reporter-positive cells isolated from caudal ends of wild-type (, upper) or T 2J/2J embryos (lower panel) of stages TS12 (left) and TS13 (right), used for in vivo ATAC-Seq analysis. Samples from hearts (left) were used as negative controls; the relative amounts of cells in each subpopulation are indicated in percentages. (C) Frequency distribution plots of DNA fragment sizes sequenced in ATAC- Seq experiments of sorted cell populations. (D) Box plots of genomic coverage distributions of all samples across different chromosomes shown before (left) and after (right) normalization.

10 A Koch_Figure S5 T 2J/2J / B T 2J/2J C chr5 chr6 chr10 chr11 T 2J/2J / T 2J/2J chr4 chr3 chr1 chr7 chr9 chr12 chr8 chr13 chr15 chr16 chr14 chr2 chr17 chrx chr18 chr /

11 Figure S5. Related to Figure 4. (A) Genome-browser snapshot of chromosome 2 showing ATAC-Seq data profiles derived from FACS purified cells of wild-type () or T 2J/2J TS12 embryos relative to stage-matched wild-type NMP cells. (B) Genome-wide correlation matrix of all ATAC-Seq datasets derived from indicated cell types at and genotypes at TS12 relative to their respective stage-matched wild-type NMP dataset. (C) Heatmap of the genome-wide Hilbert curve of global ATAC-Seq data (relative to the NMP dataset) derived from FACSpurified, / or cells from TS12 wild-type () or mutant (T 2J/2J ) embryos. The map of the underlying Hilbert curves shows the location of each chromosome.

12 Koch_Figure S6 A Brightfield mcherry B Neg. (head) CE TS14 Tbx6 -/- Tbx6 -/- 99.0% 99.4% Tbx6 mch + Tbx6 mch + 0.9% 39.2% 59.5% Tbx6::mCherry Tbx6::mCherry Tbx6 mch + Tbx6 mch + 0.5% 33.5% 65.8% Tbx6::mCherry Tbx6::mCherry C KO F T D E Tbx6 mch + Tbx6 mch + KO KO KO KO Msgn1 Wnt3a Foxf1 Fgf8 G H Tbx6 T Tbx6 T Fgf8 Msgn1 Tbx6 mch + KO Tbx6 Sox1 Sox3 Sox3

13 Figure S6. Related to Figure 5. (A) Fluorescence microscopy pictures of embryos obtained by tetraploid complementation using wild-type () or Tbx6 -/- ES cells containing the Tbx6::mCherry reporter, verifying the correct expression pattern at TS14. In the mcherry signal is mainly observed in paraxial mesoderm, however the signal persists in several somite pairs due to mcherry stability. For FACS sorting, only caudal end material posterior to presomite -2 in wild-type, representing mostly presomitic mesoderm tissue was isolated by microdissection. (B) FACS profiles of Tbx6::mCherry cells sorted from (upper panel) or Tbx6 -/- embryos (lower panel) of stage TS14 used for transcriptome or ATAC-Seq analysis. Samples from heads were used as negative controls; the relative amounts of cells in each subpopulation are indicated in percentages. (C-E) Exemplary genome-browser views of RNA-seq read densities from dysregulated genes behaving similarly in T 2J/2J and Tbx6 -/- knock-outs (C), genes downregulated in T 2J/2J and upregulated in Tbx6 -/- knock-outs (D), and genes upregulated in T 2J/2J and unchanged in Tbx6 -/- knock-outs (E). (F-H) Exemplary genome-browser views of T and Tbx6 -seq read densities from dysregulated genes behaving similarly in T 2J/2J and Tbx6 -/- knock-outs (F), downregulated in T 2J/2J and upregulated in Tbx6 -/- knock-outs (G), and upregulated in T 2J/2J and unchanged in Tbx6 -/- knock-outs (H).