Fig. 6: Heterogeneous properties of VE cells.

a ImSAnE extracted 3 μm-thick VE from an E5.5 Cdh1-GFP (green) plotted back to a 3D projected embryo. CERL protein (gray) positive cells are migrating DVE. b ImSAnE surface projection showing VE segmented based on subtypes defined by the location of the Cerl-positive VE and ExE. ExVE, DVE, anterior EmVE (aEmVE), and posterior EmVE (pEmVE). c Quantification of mean shape index between EmVE and ExVE of E5.5 embryos. EmVE showed a higher shape index than the ExVE within the same embryo. n = 6 embryos. p-value = 0.02, two-sided paired Student’s t test. *P < 0.05. d Quantification of mean shape index of various VE subtypes in E5.5 embryos: DVE, aEmVE, pEmVE, and the anterior VE (aEmVE and DVE). n = 7 embryos. Error bars: standard deviation. e Quantification of mean shape index between aEmVE and pEmVE of E5.5 embryos. aEmVE showed a higher shape index than the pEmVE within the same embryo. n = 11 embryos. p-value = 0.0037(**); two-sided paired Student’s t test. f Confocal cross-section of a representative E5.25 Cdh1-GFP embryo, with brightfield (BF) and immunofluorescence. CERL (gray) was undetected at this stage. Scale bars: 20 μm. g ImSAnE surface projection showing VE of the embryo in (f) segmented and the shape index extracted, showing a 3D projected embryo with color-coded shape indexes. h Comparison of the mean shape index between two halves of EmVE in E5.0 embryos. n = 6 embryos. p-value = 0.02(*); two-sided paired Student’s t test. i Comparison of mean shape index between two halves of EmVE in E5.25 embryos. n = 18 embryos. p-value = 0.0005(***); two-sided paired Student’s t test. j Comparison of the mean shape index between two halves of EmVE in basement membrane-depleted (BM depleted) E5.25 embryos. n = 4 and 5 embryos. p-value = 0.04(*); two-sided unpaired Student’s t test. k Diagram showing an approximated E5.5 embryo with its proximal-distal axis projected on a polar projection, with the distal tip in the center of the circle for physical modeling. l Cells are represented by dots, with color indicating simulation time. Shade of green indicates the degree of perforation of the basement membrane. Black arrows indicate a distal to proximal migratory cue. If the strength of the migratory force depends on the degree of basement membrane perforations, a radial migration cue, and cell–cell adhesion led to directed DVE migration. Note that the direction of cell migration force is always azimuthally symmetric, only its magnitude is modified by basement membrane heterogeneity. The mechanism described above depends on cell–cell adhesion in the DVE, predicting that directed migration should fail if it is disrupted. m Basement membrane perforations must modulate active force generation by DVE cells. If only the resistance DVE cells experience is dependent on the basement membrane cue, DVE migration becomes disoriented. Passive resistance is modeled by surface friction. n When the basement membrane is uniformly depleted, directed migration fails and the DVE ruptures into multiple patches. o DVE rupture occurs in approximately 50% of simulations where the basement membrane is uniformly depleted. n = 20 simulations. p, q Maximum projection of time-lapse confocal stacks of E5.5 Cerl-GFP embryos in control (p) and 10 μM Batimastat-treated condition (q). Note that the length of protrusions (magenta arrowhead) in reduced in Batimastat-treated embryo. Scale bars: 20μm. r Quantifications of the length of DVE protrusions in control and in Batimastat-treated embryos. n = 13 and 12 embryos. p-value < 00001(****); two-sided unpaired Student’s t test.