DEV021519 Supplementary Material

Files in this Data Supplement:

• Supplemental Tables 1-4 - Adobe PDF

• Supplemental Figure S1 -

  Fig. S1. Detection of Pdgfrα protein and transcripts in early embryos. (A) Endogenous Pdgfrα protein is undetectable in embryos before the 32-cell stage, although weak GFP expression is detected. (B) In embryos of >32 cells, Pdgfrα protein first appears in a subset of GFP-positive cells. (C) In >64-cell stage embryos, Pdgfrα and GFP are colocalised. Red, Pdgfrα; green, GFP; blue, Hoechst. Scale bar: 20 µm. (D) Pdgfrα transcripts are present during early mouse development. Pdgfrα is initially expressed at low levels in 8-cell and morula stage embryos. Increased levels are observed in E3.5 and E4.5 embryos.

• Supplemental Figure S2 -

  Fig. S2. Total cell number during early development in relation to age post-coitum and expression of lineage markers. (A) Total cell number versus age post-coitum for all embryos analysed. (B) Total cell numbers versus age post-coitum, where each time-point is represented by one selected litter to demonstrate variation in cell number within litters. (C) Mean total cell number of embryos collected at progressive 3-hourly time-points. Error bars indicate s.d. The supporting table indicates numbers of embryos and litters for each time-point. (D) Frequency histogram of embryo stages classified by total cell number. Cohort sizes are adjusted to linearise an exponential scale of increasing cell number. Major peaks in 15 to 18- and 30 to 36-cell cohorts, and a putative minor peak in the 58 to 72-cell cohort, are interpreted as reflecting synchronicity in the cell cycle. Cells were counted by printing on paper multiple confocal sections of nuclear staining. Nuclei, which were always present in a least two sections, were then individually identified and numbered to determine the total number of cells.

• Supplemental Figure S3 -

  Fig. S3. Nanog and Cdx2 are co-expressed at the early blastocyst stage. (A,B) At the early blastocyst stage (∼32-cells), Nanog can be expressed in Cdx2-positive cells, sometimes more weakly (A) and sometimes more strongly (B) than in a typical ICM cell. In rare cases, Cdx2 can be detected in ICM cells (arrowhead, B). (C) In a later (60-cell) expanded blastocyst, Cdx2 is expressed exclusively in the TE and Nanog exclusively in the ICM. White, Nanog; red, Cdx2; blue, Hoechst. Scale bar: 20 µm.

• Supplemental Figure S4 -

  Fig. S4. Dynamic expression of GFP in Pdgfrα^H2B-GFP/+ embryos of more than 128-cells visualized by 3D time-lapse microscopy. (A-H) Dynamic GFP expression is shown in a single blastocyst over time. Single optical sections of GFP fluorescence are merged with bright-field images. (A′-H′) Higher magnifications of A-H showing the GFP channel only. Some cells that have not formed part of the nascent PrE layer express GFP weakly (arrowheads). One cell (red arrowhead) upregulates GFP briefly (B′) before undergoing DNA fragmentation (two red arrowheads, D′). Another cell (yellow arrowhead) downregulates GFP to a level that is undetectable (H′). Note the upregulation of GFP in cells lining the cavity. Scale bar: 20 µm.

• Supplemental Figure S5 -

  Fig. S5. After 17 hours of time-lapse imaging, embryos remain viable and express PrE- and EPI-specific markers as per freshly collected embryos. (A) Embryos exposed to 17 hours of 3D time-lapse confocal microscopy starting at E3.5, which were shown to form PrE (which we assessed by observation of GFP-positive cell distribution), were fixed and stained for the presence of PrE- and EPI- specific markers. Gata4 staining in these embryos was restricted to the PrE layer and was mutually exclusive with Nanog staining in the EPI (n=26), as we observed in wild-type embryos. Single optical sections of the same Pdgfrα^H2B-GFP/+ embryo fixed after 17 hours of time-lapse imaging. Green, GFP; red, Gata4; white, Nanog; blue, Hoechst. Scale bar: 20 µm. (B) Pdgfrα^H2B-GFP/+ embryos exposed to 17 hours of 3D time-lapse confocal microscopy starting at E3.5 were transferred to recipient females. Seven out of 20 developed to term. Newborn animals were morphologically indistinguishable from animals derived from genetically distinct CAG::mRFP1 non-imaged embryos that were co-transferred. Furthermore, they were viable and exhibited normal fertility in adult life. The discrepancy in the number of newborn animals is likely to result from the postimplantation embryonic lethality of Pdgfrα^{H2B-GFP/H2B-GFP} embryos, which were used in some of our experiments. (C) It should be noted that Pdgfrα^H2B-GFP/+ embryos were indistinguishable from Pdgfrα^{H2B-GFP/H2B-GFP} embryos at preimplantation stages examined, except for a stronger GFP signal.

• Supplemental Figure S6 -

  Fig. S6. Incidence of apoptosis increases after the 64-cell stage. (A,B) In embryos comprising >64-cells, apoptotic cells were detectable by both TUNEL assay (A) and cleaved caspase staining (B). Red: TUNEL (A), cleaved caspase (B); green, GFP; blue, Hoechst. Scale bar: 20 µm. (C) Percentage incidence of apoptosis in relation to total cell number. Apoptosis is infrequent (only one apoptotic cell recorded) before the 64-cell stage but increases progressively thereafter.

• Movie 1 -

  Movie 1. Pdgfrα and Gata4 are colocalised in the PrE layer of E4.0 blastocysts. Sequential x-y images of a z-stack taken through the E4.0 blastocyst depicted in Fig. 1A. Green, Pdgfrα; red, Gata4. Merged with bright field.

• Movie 2 -

  Movie 2. GFP expression is restricted to the ICM of about 50% of Pdgfrα^H2B-GFP blastocysts. Sequential x-y images of a z-stack taken through a 33-cell blastocyst. Green, GFP; red, FM4-64.

• Movie 3 -

  Movie 3. Intensity of Gata6 and Nanog staining is independent of cell position within the morula. Sequential x-y images of a z-stack taken through a late morula. White, Nanog; red, Gata6; blue, Hoechst.

• Movie 4 -

  Movie 4. Gata4 and Nanog expression patterns do not overlap. Sequential x-y images of a z-stack taken through a mid-blastocyst. White, nanog; red, Gata4; blue, Hoechst.

• Movie 5 -

  Movie 5. GFP-expressing cells are randomly distributed in Pdgfrα^H2B-GFP morulae and early blastocysts. 3D time-lapse movie (15 minutes per frame) of early Pdgfrα^H2B-GFP embryos.

• Movie 6 -

  Movie 6. After a phase of heterogeneous distribution, GFP-positive cells form a layer lining the cavity. 3D time-lapse movie (15 minutes per frame) of 3D reconstructed images of Pdgfrα^H2B-GFP blastocysts.

• Movie 7 -

  Movie 7. Transition from a random distribution of GFP-positive cells to a nascent PrE layer can occur within only 15 minutes. 3D time-lapse movie (15 minutes per frame) of 3D reconstructed images of a Pdgfrα^H2B-GFP blastocyst.

• Movie 8 -

  Movie 8. In Pdgfrα^H2B-GFP embryos beyond the 128-cell stage, GFP signal intensity increases in cells lining the cavity, while cells away from the cavity downregulate GFP or undergo cell death. 3D time-lapse movie (15 minutes per frame) of 3D reconstructed images of Pdgfrα^H2B-GFP blastocyst.

• Movie 9 -

  Movie 9. Diverse behaviour of GFP-positive cells contributing to the PrE layer in Pdgfrα^H2B-GFP/+ embryos. 3D time-lapse movie (15 minutes per frame) of 3D reconstructed images of the Pdgfrα^H2B-GFP blastocyst depicted in Fig. 6A. The dark blue dot indicates a cell staying in contact with the cavity throughout the movie. The yellow dot indicates a weakly GFP-positive cell migrating away from the cavity. The pale blue dot indicates a cell intercalating into the PrE due to cavity expansion and upregulating GFP expression. The magenta and red dots indicate GFP-positive cells intercalating into the PrE layer.