Label-free imaging of 3D pluripotent stem cell differentiation dynamics on chip.

Massive, parallelized 3D stem cell cultures for engineering in vitro human cell types require imaging methods with high time and spatial resolution to fully exploit technological advances in cell culture technologies. Here, we introduce a large-scale integrated microfluidic chip platform for automated 3D stem cell differentiation. To fully enable dynamic high-content imaging on the chip platform, we developed a label-free deep learning method called Bright2Nuc to predict in silico nuclear staining in 3D from confocal microscopy bright-field images. Bright2Nuc was trained and applied to hundreds of 3D human induced pluripotent stem cell cultures differentiating toward definitive endoderm on a microfluidic platform. Combined with existing image analysis tools, Bright2Nuc segmented individual nuclei from bright-field images, quantified their morphological properties, predicted stem cell differentiation state, and tracked the cells over time. Our methods are available in an open-source pipeline, enabling researchers to upscale image acquisition and phenotyping of 3D cell culture.

Single-cell characterization of neovascularization using hiPSC-derived endothelial cells in a 3D microenvironment.

The formation of vascular structures is fundamental for in vitro tissue engineering. Vascularization can enable the nutrient supply within larger structures and increase transplantation efficiency. We differentiated human induced pluripotent stem cells toward endothelial cells in 3D suspension culture. To investigate in vitro neovascularization and various 3D microenvironmental approaches, we designed a comprehensive single-cell transcriptomic study. Time-resolved single-cell transcriptomics of the endothelial and co-evolving mural cells gave insights into cell type development, stability, and plasticity. Transfer to a 3D hydrogel microenvironment induced neovascularization and facilitated tracing of migrating, coalescing, and tubulogenic endothelial cell states. During maturation, we monitored two pericyte subtypes evolving mural cells. Profiling cell-cell interactions between pericytes and endothelial cells revealed angiogenic signals during tubulogenesis. In silico discovered ligands were tested for their capability to attract endothelial cells. Our data, analyses, and results provide an in vitro roadmap to guide vascularization in future tissue engineering.

Single-cell profiling of GP2-enriched pancreatic progenitors to simultaneously create acinar, ductal, and endocrine organoids.

Rationale: Pancreatic lineage specification follows the formation of tripotent pancreatic progenitors (PPs). Current protocols rebuilding PPs in vitro have an endocrine lineage bias and are mostly based on PDX1/NKX6-1 coexpression neglecting other markers decisive for PP heterogeneity and lineage potential. However, true tripotent PPs are of utmost interest to study also exocrine disorders such as pancreatic cancer and to simultaneously generate all three pancreatic lineages from the same ancestor. Methods: Here, we performed a comprehensive compound testing to advance the generation of multipotent progenitors, which were further characterized for their trilineage potential in vitro and in vivo. The heterogeneity and cell-cell communication across the PP subpopulations were analyzed via single-cell transcriptomics. Results: We introduce a novel PP differentiation platform based on a comprehensive compound screening with an advanced design of experiments computing tool to reduce impurities and to increase Glycoprotein-2 expression and subsequent trilineage potential. Superior PP tripotency was proven in vitro by the generation of acinar, endocrine, and ductal cells as well as in vivo upon orthotopic transplantation revealing all three lineages at fetal maturation level. GP2 expression levels at PP stage ascribed varying pancreatic lineage potential. Intermediate and high GP2 levels were superior in generating endocrine and duct-like organoids (PDLO). FACS-based purification of the GP2high PPs allowed the generation of pancreatic acinar-like organoids (PALO) with proper morphology and expression of digestive enzymes. scRNA-seq confirmed multipotent identity, positioned the GP2/PDX1/NKX6-1high population next to human fetal tip and trunk progenitors and identified novel ligand-receptor (LR) interactions in distinct PP subpopulations. LR validation experiments licensed midkine and VEGF signaling to increase markers labelling the single cell clusters with high GP2 expression. Conclusion: In this study, we guide human pluripotent stem cells into multipotent pancreatic progenitors. This common precursor population, which has the ability to mature into acinar, ductal and functional ß-cells, serves as a basis for studying developmental processes and deciphering early cancer formation in a cell type-specific context. Using single-cell RNA sequencing and subsequent validation studies, we were able to dissect PP heterogeneity and specific cell-cell communication signals.

Single-cell-resolved differentiation of human induced pluripotent stem cells into pancreatic duct-like organoids on a microwell chip.

Creating in vitro models of diseases of the pancreatic ductal compartment requires a comprehensive understanding of the developmental trajectories of pancreas-specific cell types. Here we report the single-cell characterization of the differentiation of pancreatic duct-like organoids (PDLOs) from human induced pluripotent stem cells (hiPSCs) on a microwell chip that facilitates the uniform aggregation and chemical induction of hiPSC-derived pancreatic progenitors. Using time-resolved single-cell transcriptional profiling and immunofluorescence imaging of the forming PDLOs, we identified differentiation routes from pancreatic progenitors through ductal intermediates to two types of mature duct-like cells and a few non-ductal cell types. PDLO subpopulations expressed either mucins or the cystic fibrosis transmembrane conductance regulator, and resembled human adult duct cells. We also used the chip to uncover ductal markers relevant to pancreatic carcinogenesis, and to establish PDLO co-cultures with stellate cells, which allowed for the study of epithelial-mesenchymal signalling. The PDLO microsystem could be used to establish patient-specific pancreatic duct models.

Modeling plasticity and dysplasia of pancreatic ductal organoids derived from human pluripotent stem cells.

Personalized in vitro models for dysplasia and carcinogenesis in the pancreas have been constrained by insufficient differentiation of human pluripotent stem cells (hPSCs) into the exocrine pancreatic lineage. Here, we differentiate hPSCs into pancreatic duct-like organoids (PDLOs) with morphological, transcriptional, proteomic, and functional characteristics of human pancreatic ducts, further maturing upon transplantation into mice. PDLOs are generated from hPSCs inducibly expressing oncogenic GNAS, KRAS, or KRAS with genetic covariance of lost CDKN2A and from induced hPSCs derived from a McCune-Albright patient. Each oncogene causes a specific growth, structural, and molecular phenotype in vitro. While transplanted PDLOs with oncogenic KRAS alone form heterogenous dysplastic lesions or cancer, KRAS with CDKN2A loss develop dedifferentiated pancreatic ductal adenocarcinomas. In contrast, transplanted PDLOs with mutant GNAS lead to intraductal papillary mucinous neoplasia-like structures. Conclusively, PDLOs enable in vitro and in vivo studies of pancreatic plasticity, dysplasia, and cancer formation from a genetically defined background.