Extended Data Fig. 8: Blastoderm cell density does not correlate with presence or absence of the CF, and nuclear double-layering is not due to delamination or pseudostratification.
From: Divergent evolutionary strategies pre-empt tissue collision in gastrulation
a–f, Representative images of fixed embryos stained with DRAQ5 (see Methods) and segmented nuclei (gray-scale; red dots mark individual nuclei). White boundaries enclose regions used to calculate nuclear density (red dots/area). Scale bars, 20 µm. g, Nuclear density distributions for six species, dashed vertical line separates species with and without the CF. Each dot represents an embryo; bold lines, median; whiskers, 95% CI. n = 9, 11, 25, 22, 15, 12 for D. melanogaster, M. abdita, H. illucens, C. fuscipes, C. riparius, C. albipunctata, respectively. Statistical significance determined using one-way ANOVA (non-parametric Kruskal-Wallis test with Dunn’s uncorrected test for multiple comparisons), non-significant differences are indicated in gray. h, Time-lapse imaging of the early C. riparius head region shows cells retain columnar shapes until mitosis, without evidence for cone/frustum morphologies. This argues against pseudostratification or cell extrusion underlying nuclear double-layering (n = 5 embryos, 16 cells tracked). Images: time frames (columns), z-depths (rows), each panel 50 × 50 µm. Colored dots track cell positions; estimated positions are indicated when not visible due to embryo curvature (asterisk) or mitotic rounding (open circle).
