in approximately 10 mL cold XF HA media per 150 cm2 culture surface area. Cells were counted using a ViCell automated cell counter, then seeded into ultra-low adhesion 6-well trays at 16106 cells/mL in XF HA media and 5.5 mL per well. The 6-well trays were incubated at 37uC on orbital rotators set at 95 rpm. During overnight culture, single cells aggregated to form spherical clusters approximately 100200 mm in diameter and roughly 5,000 aggregates per well. The differentiation media and rotation conditions used for each day are detailed in Methods S1. To initiate differentiation, aggregates were pooled into conical tube and allowed to settle by gravity, or in some experiments by centrifugation at 500 rpm for 90 seconds, followed by a wash using Stage-1 media without growth factors. The aggregates were resettled, then resuspended in day 1 media and evenly redistributed to the same number of wells using new 6-well trays in 5.5 mL total volume per well. After 2024 hrs of differentiation some cell death and a wispy mass, or clump, of 
DNA was visible in each well. On this day, clumps were disrupted by triturating with a P200 pipettor and returned to the rotator for 30 seconds to allow aggregates to collect in the center of the well. Approximately 4.5 mL of media was aspirated from each well and wells were fed with 4.5 mL of day 2 media. Aspiration of 4.5 mL of media followed by feeding with 4.5 mL of the subsequent day’s media was performed daily thereafter. In some experiments, at the end of Stage-2 the aggregates were pooled and redistributed to half the original number of wells in 5.5 mL of day 5 media. Aggregates were sampled at the indicated time points for histological, immunofluorescence, and gene expression analyses. Materials and Methods AGI-6780 web Ethics Statement Animal experiments were performed at Absorption Systems, San Diego, CA and were approved by their Institutional Animal Care and Use Committee, protocol #VC-02-09-148. Undifferentiated hESC Culture The CyT49 hESC line was used in these studies. Xeno-free growth media consisted of DMEM/F12 containing GlutaMAX supplemented with 10% of Xeno-free KnockOut Serum Replacement, 1% non-essential amino acids, 0.1 mM 2-mercaptoethanol, 1% penicillin/streptomycin, 10 ng/mL heregulin-1b and 10 ng/mL activin A. Upon thaw, or at regular passaging, dissociated hESC were plated at 50,000 or 33,000 cells/cm2 for three and four day growth cycles, respectively. On the day of plating only, cell attachment was facilitated by including 10% of non-heat inactivated human AB serum. A standardized plating volume of 0.2 mL/cm2 was used for different tissue culture plates, T-flasks and cell factories. The volume of growth media used was increased for each additional day of feeding according to Methods S1, and did not include human serum. On the day of passaging, cultures were fed with fresh growth media and cultured 48 hrs before dissociation. Cultures were washed with PBS and dissociated for 6 mins at 37uC using pre-warmed Accutase. Dissociated cells were gently collected using 36 volume of cold hESC PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/22188834 media, counted using a ViCell automated cell counter, or a hemocytometer, centrifuged for 5 mins at 2006 g and the pellet resuspended in fresh growth media at 1106106 cells/mL for subsequent plating. Where indicated, the StemPro hESC SFM medium supplemented with 10 ng/mL heregulin-1b, 10 ng/mL activin A, 10 ng/mL FGF2 and 200 ng/mL LR3-IGF1, or the same media without FGF2 were used for hESC culture or suspension c
