Supplementary Figure 7: Effect of cell density on the patterning of hESC induced by a localized BMP4 source.

(a) Computational simulation showing the local percentage of phospho-SMAD1 positive cells in a colony with an initially high cell density exposed to a localized BMP4 source of 50ng/ml. (b) Representative pictures (DAPI) of HES3 MIXL1^GFP/+ cell suspensions at different concentrations 30 hours after the loading in the cell chambers. (c) Boxplots of the measured cell density for the different cell suspension concentrations (n=8 independent samples per condition). For simplicity the conditions are named “Low”, “Intermediate” and “High” density. ANOVA and subsequent multiple two-tailed t-test Benjamini-Hochberg corrected were performed. (d) Boxplots of the distribution of HES3 MIXL1^GFP/+ cells positive for the reported markers along the axis of the cell chamber after 48 hours of exposure to a localized 50ng/ml BMP4 source starting with high cell density (n=8 independent samples). (e) Boxplots of the distribution of HES3 MIXL1^GFP/+ cells positive for the reported markers along the axis of the chamber after 48 hours of exposure to a localized 250ng/ml BMP4 source starting with high cell density (n=6 independent samples). For all the boxplots the center line represents the median, the box limits represent the first and third quartiles, the whiskers represent the upper and lower extremes of the distribution, while white circles represent outliers. Single data points are also shown in red. (f) Time points from a computational simulation showing the local percentage of phospho-SMAD1 positive cells in a colony with an initially high cell density exposed to a localized BMP4 source of 250ng/ml. (g) Representative fluorescence pictures for indicated markers in HES3 MIXL1^GFP/+ cells exposed 48 hours to a localized 50ng/ml source of BMP4 in the microfluidic device starting with low cell density. This experiment was repeated two times on a total of 8 samples and showed similar results. Scale bar 200μm.