Fig. 5: Deletion of the ETV genes affects germ-layer formation and pancreatic in vitro differentiation.

A Scheme illustrating the micropatterning approach to in vitro generation of the organized embryo-like structures (gastruloids). hPSCs (light blue) were plated on laminin 521-coated micropatterns and exposed for 46 h to BMP4 to induce self-organization and formation of gastruloids, containing extraembryonic-like cells (red), ectodermal cells (green) and the mesodermal and endodermal cells (dark blue). B Representative immunofluorescence staining of ISL1 (red), SOX2 (green), and BRA (blue) in gastruloids generated from ESI (hPSC control cell line), HUES8 (hPSC control cell line), iCas9 (WT hPSCs, parental control), KO and tKO hPSCs. Note, the disrupted germ layers organization in KO hPSCs (SOX2+ and BRA+), the overgrowth of extraembryonic-like cells (ISL1+), and the absence of SOX2+ cells in tKO, compared to control hPSC lines (ESI, HUES8 and iCas9). Scale bars = 200 µm. C Distribution from the center to the gastruloid edge of ISL1, SOX2, and BRA positive signals based on immunofluorescence staining on ESI (blue), HUES8 (red), iCas9 (green), tKO (black), and KO (pink) gastruloids. Data are presented as the mean ± SEM. N = 3 biological replicates. D WT, KO, and tKO differentiation to definitive endoderm using a standard protocol with full doses of Activin A and CHIR99021 (a WNT activator). N = 3 biological replicates. KO vs. WT, p = ns, tKO vs. WT, p = 0.0232, tKO vs. KO, p = 0.0002. E WT, KO, and tKO differentiation to definitive endoderm using a suboptimal protocol with decreased concertation of Activin A, with or without CHIR99021. N = 3 biological replicates. KO vs. WT and tKO vs. KO, p < 0.0001, tKO vs. WT, p = ns. F Scheme illustrating hPSC spontaneous differentiation into embryoid bodies (EBs). G Gene expression changes in KO EBs reveal altered differentiation potential. Dot plot shows log2FC in gene expression (qPCR) between KO and WT EBs at days 8 and 10 of spontaneous differentiation. The increased expression of endodermal and mesodermal markers and decreased expression of ectodermal markers were observed in KO at both timepoints. N = 3 biological replicates. The p-values were determined by two-sided student’s t-test. H Representative immunofluorescence images of human fetal pancreas at weeks 10.6 and 13 of embryogenesis stained with PDX1 (red), CHGA (gray), and ETV1 (green). N = 3 tissue samples. Scale bar = 100 µm. I Representative immunofluorescence staining of ETV1 (green) in WT hPSC-derived definitive endoderm (DE, upper panel), and pancreatic progenitors (PP, bottom panel). SOX17 (red) marks DE, and PDX1 (green) and NKX6-1 (red) mark PPs. N = 3 biological replicates. Scale bar = 200 µm. J Scheme illustrating hPSC differentiation into pancreatic progenitors (PP). During differentiation, hPSCs pass through the following stages: definitive endoderm (DE) marked by FOXA2 and SOX17 expression, PP marked by PDX1 and NKX6-1 expression. At the PP stage (day 12) single-cell RNA sequencing and ATAC sequencing were performed on WT and KO samples. K Representative immunofluorescence staining of PDX1 (green), NKX6-1 (red), and CHGA (gray) proteins in WT and KO PPs. N = 3 biological replicates. Scale bar = 200 µm. L Representative flow cytometry analysis showing the absence of NKX6-1+KO PPs and decreased PDX1 protein levels in KO compared to WT PPs. Two distinct PDX1+ populations are observed in KO cells: low PDX1+ (~50% of cells) and high PDX1+ (~50% of cells). In contrast, WT PPs exhibit predominantly high PDX1 expression. For plots (D, E), a one-way ANOVA with Turkey’s correction for multiple comparisons was used to determine the p-values shown on the graph. The data are presented as means ± SDs.