Fig. 4: Soft substrates improve self-renewal in minimal conditions.

a Procedure for clonogenicity assays. Cells were seeded on different substrates (i) for 5 days or (ii) and (iii) passaged every 2–3 days. Passaging (psg.) was done (ii) by serial dilution from 1:3 to 1:4 or (iii) by keeping a constant plating density of 50,000 cells per substrate. Cells were then replated at clonal density (100 cells/cm²) on gelatin-coated TCP in 2i+LIF to allow self-renewing cells to form colonies. Finally, colonies were fixed and stained for alkaline phosphatase (AP). b Clonogenicity assay of mESCs after 5 days in serum following protocol (i). (Left) Brightfield images of mESCs after 5 days. Scale bar: 100 µm. (Centre) Snapshot of AP stainings. (Right) Quantification of the number of AP+ colonies in % of control conditions (serum + LIF on TCP) (n = 8 independent samples). c Clonogenicity assay after mESCs were cultured for three passages in N2B27 + CHIRON following protocol (iii). (Left) Snapshots of AP stainings. (Right) Quantification of the number of AP + colonies in % of plated cells (n = 5–8 independent samples). d Clonogenicity assay after mESCs were cultured for three passages in N2B27 + PD03 following protocol (iii). Cells were from a C57BL/6-Agouti background. (Left) Snapshots of AP stainings. (Right) Quantification of the number of AP+ colonies in % of plated cells (n = 8 independent samples). e First (left) and second (right) generation chimaeric mice obtained after the injection into blastocysts of GFP-transfected mESCs cultured on soft substrates in N2B27 + PD03 for three passages following protocol (ii). The bedding autofluorescence contributes to background in the bottom-left image, but clear GFP+ regions are visible on the mice. f Brightfield images of naïve hPSCs in control media “PDLGX” and in media without PD03, “LGX”. Scale bar: 100 µm. g Clonogenecity assay of naïve hPSCs in LGX. (Left) Snapshot of AP-stainings. (Right) Quantification of number of AP+ colonies in % of control conditions “PDLGX”. The replating efficiency was low with only 11% recovered colonies in PDLGX conditions (n = 8 independent samples). In all panels above, error bars show standard deviation, points show the values of independent repeats, and P-values were computed using a one-way ANOVA with Tukey–Kramer’s multiple comparison test. Source data are provided as a Source Data file.