Extended Data Fig. 1 Exemplary production of hiPSC-derived cardiomyocytes from vitronectin and (GMP-conforming) CTS vitronectin pre-culture.

a and b , Appearance of cells on CTS vitronectin ( a ) and vitronectin ( b ) on day −2 of the production protocol in monolayer (scale bar, 500 µm). c , Expression of markers of an undifferentiated state at CP 0 before induction of differentiation in samples from spinner flasks pre-cultured on vitronectin or CTS vitronectin. d , Cell yield at CP I of respective spinner-flask runs. e , Aggregate diameter of representative samples taken at CP I from respective spinners. Shown are individual values and mean ± s.d. in red. f , Percentage of cells positive for cardiac markers in representative samples from the respective spinner flasks at CP I.

Extended Data Fig. 2 Setup of the spinner flask.

In a sterile environment, unpack from plastic and install the lid with the stirrer. Remove one of the lids for the side ports and install a sampling device (optional).

Extended Data Fig. 3 Aggregate diameter size distribution for one differentiation run at CP 0, CP I and CP II.

Shown are individual aggregate diameters (gray) and mean values ± s.d. (red). The hiPSC line used is Amber.

Extended Data Fig. 4 Total aggregate count per milliliter of spinner flask experiments at CP 0 ( n = 3) and CP I ( n = 2).

Aggregate numbers were determined through dilution of a culture sample and manual counting through microscopy.

Extended Data Fig. 5 Confocal microscopy of aggregates at CP 0 and CP I to analyze cell number per aggregate.

a–c , Light microscope pictures of a representative cell sample at CP 0 (scale bar, 500 µm) ( a ) and confocal images of the central Z-stack of aggregates of the same batch (scale bar, 50 µm) ( b and c ). d–f , Light microscope pictures of a representative cell sample at CP I (scale bar, 500 µm) ( d ) and confocal images in a differentiated state as CM aggregates at CP I (scale bar, 50 µm) ( e and f ). Nuclei were stained with Sytox red after fixation; afterwards, aggregates were dehydrated with ethanol and cleared with methyl salicylate/benzyl benzoate (MSBB). Notable are pronounced cavities at CP 0 as well as at CP I. g , Aggregates treated for confocal imaging do not differ in diameter compared to untreated, equivalent samples measured according to QC3 at CP 0 or CP I. h , Nuclei identified at CP 0 and CP I through automatic counting of confocal z-stacks imaging whole aggregates. This nuclei count should closely resemble the total cell count per aggregate. i and j , Identified nuclei per aggregate in comparison to the individual aggregate diameter for aggregates at CP 0 ( i ) and CP I ( j ). Linear regression was added to depict the goodness of fit (CP 0 R 2 = 0.73; CP I R 2 = 0.55). For examples, see supplementary file z-stacks for CP 0 and CP I (open with the hyperstack function of FIJI/ImageJ). PFA, paraformaldehyde. The hiPSC line Phoenix was used in all experiments.

Supplementary Information

Supplementary Tables 1 and 2

Supplementary Video 1

CM harvest for downstream application

Supplementary Video 2

hPSC splitting

Supplementary Video 3

Spinner medium exchange

Supplementary Data 1

Confocal z-stack pluripotent aggregates CP 0

Supplementary Data 2

Confocal z-stack CM aggregates CP I

Supplementary Code 1

ImageJ macro for aggregate size determination

Supplementary Code 2

ImageJ macro for viability score determination