Fig 2: Colony spreading regulates NPC rosette formation. (A) Quantitation of the growth of PODXL-GFP colonies over time. Imaging initiated 2 h after plating. Red dotted line at the 6-hr time point (8 h after plating) indicates the start of rosette morphogenesis. (B) Live imaging of 1196a hiPSC expressing Lifeact-GFP 2 h after roller cutting and re-plating. Imaging was started 5 h after re-plating. Top panels show images of whole colonies. (i) and (ii) panels show magnified images at the colony edge (i) or at the center (ii). Scales indicate 150 µm for whole colony, 50 µm for magnified images in i and ii. (C) Confocal images of 8 h NPC rosette colony edge cells stained with indicated markers. (D) Quantitation of basal cell area (based on N-CADHERIN membrane staining) of roller-dissociated 1196a hiPSC-derived NPC colonies 1 h after plating at the center (1 h (center)) and 6 h after plating in cells that are inside (Rosette) and outside (Non-Rosette) of rosettes. (E) Quantitation of basal cell area of H9 hESC-derived day 10 NPC monolayer (no roller-cut) in cells that are inside (Rosette) and outside (Non-Rosette) of rosettes. (F) Confocal images of NPC rosettes treated with DMSO (control) or ROCK-i, stained with indicated markers. (G) Quantitation of colony size from samples in panel (F). (H) Immunolocalization analysis of PODXL+ domain in ROCK-i treated samples: the X–Z plane is shown to reveal the internalized PODXL+ apical domain. (I–K) Quantitation of number of rosettes per colony (I), individual lumen area (J) and nuclear aspect ratio (K) from samples in panel (F). Scales as indicated. Student’s t-test was used for statistical analyses: *p = 0.05; **p = 0.01; ***p = 0.001; ***p < 0.0001.

Fig 3: Live-tracking of NPC rosette morphogenesis. (A–C) Live imaging of 1196a hiPSC expressing Lifeact-GFP (A) or PODXL-GFP (C) for 12 h after colonies were allowed to reattach for 2 h. (B) Confocal images of PODXL-GFP NPC rosettes stained with anti-PODXL. GFP localization mirrors endogenous PODXL localization. Note that, while a uniform prominent GFP signal is initially seen throughout the apical surface of the recently plated colony, at later time points, GFP-enriched regions are concentrated at the constricted centers of the rosettes (lumen). It is likely that, as the colony expands rapidly, the apical surface area of non-rosette cells dramatically increases, resulting in reduced GFP signal intensity per unit area. Scales as indicated.

Fig 4: Effects of colony confinement on NPC rosette morphogenesis. (A) Confocal images of NPC rosettes, 12 h after culturing on un-patterned PDMS, or on small (0.008 mm2), medium (0.049 mm2) or large (0.196 mm2) circular patterns coated with 1% Geltrex in the absence (control, DMSO) or presence of ROCK-i, stained with indicated markers. (B,C) Quantitation of number of rosettes per colony (left), individual lumen area (middle) and nuclear aspect ratio (right) from DMSO (B) and ROCK-i (C) treated samples in panel (A). Graphs containing all conditions are shown in Supplementary Figure 5. (D) Confocal images of control and RhoA-WT expressing NPC rosettes stained with apical markers, PODXL (green) and aPKC? (red). (E–G) Quantitation of colony sizes (E), individual lumen area (F), and rosettes per colony (G) in control and RhoA-WT colonies. Scales as indicated. Student’s t-test was used for statistical analysis: ns = p > 0.05; *p = 0.05; **p = 0.01.

Fig 5: Self-organisation of hiPSCs in response to extracellular matrix signaling.hiPSCs were plated in a 3D matrix of matrigel and analysed at the indicated time points. a-b, hiPSCs stained for aPKC, OCT-4 and F-actin. c, hiPSCs stained for PAR-6, GM130 and F-actin. d, hiPSCs cultured in the presence of the ROCK inhibitor Y-27632 and stained for aPKC and F-actin. e-f, hiPSCs cultured in the presence (e) or absence (f) of ROCK inhibitor, and stained for PODXL and F-actin g, hiPSCs were treated with the caspase 3 inhibitor Z-DEVD FMK and lumen formation was analysed 24 and 48 hours after plating by staining for aPKC and F-actin. h, Quantification of lumen formation in the presence of Z-DEVD FMK. Data is shown as a contingency table. Fisher’s exact test. ns: not significant (n=12 hiPSC organoids per condition -24h time point-; n=10 hiPSC organoids per condition -48h time point-). Data presented in this figure involved the assessment of a minimum of 10 hESC organoids per panel and per condition across a minimum of 2 independent experiments per panel.