SummaryDeep learning technology is rapidly advancing and is now used to solve complex problems. Here, we used deep learning in convolutional neural networks to establish an automated method to identify endothelial cells derived from induced pluripotent stem cells (iPSCs), without the need for immunostaining or lineage tracing. Networks were trained to predict whether phase-contrast images contain endothelial cells based on morphology only. Predictions were validated by comparison to immunofluorescence staining for CD31, a marker of endothelial cells. Method parameters were then automatically and iteratively optimized to increase prediction accuracy. We found that prediction accuracy was correlated with network depth and pixel size of images to be analyzed. Finally, K-fold cross-validation confirmed that optimized convolutional neural networks can identify endothelial cells with high performance, based only on morphology.

Alteration of the nuclear Ca2+ transient is an early event in cardiac remodeling. Regulation of the nuclear Ca2+ transient is partly independent of the cytosolic Ca2+ transient in cardiomyocytes. One nuclear membrane protein, emerin, is encoded by EMD, and an EMD mutation causes Emery-Dreifuss muscular dystrophy (EDMD). It remains unclear whether emerin is involved in nuclear Ca2+ homeostasis. The aim of this study is to elucidate the role of emerin in rat cardiomyocytes by means of hypertrophic stimuli and in EDMD induced pluripotent stem (iPS) cell-derived cardiomyocytes in terms of nuclear structure and the Ca2+ transient. The cardiac hypertrophic stimuli increased the nuclear area, decreased nuclear invagination, and increased the half-decay time of the nuclear Ca2+ transient in cardiomyocytes. Emd knockdown cardiomyocytes showed similar properties after hypertrophic stimuli. The EDMD-iPS cell-derived cardiomyocytes showed increased nuclear area, decreased nuclear invagination, and increased half-decay time of the nuclear Ca2+ transient. An autopsied heart from a patient with EDMD also showed increased nuclear area and decreased nuclear invagination. These data suggest that Emerin plays a crucial role in nuclear structure and in the nuclear Ca2+ transient. Thus, emerin and the nuclear Ca2+ transient are possible therapeutic targets in heart failure and EDMD.

Cardiogenesis relies on the precise spatiotemporal coordination of multiple progenitor populations. Understanding the specification and differentiation of these distinct progenitor pools during human embryonic development is crucial for advancing our knowledge of congenital cardiac malformations and designing new regenerative therapies. By combining genetic labelling, single-cell transcriptomics, and ex vivo human-mouse embryonic chimeras we uncovered that modulation of retinoic acid signaling instructs human pluripotent stem cells to form heart field-specific progenitors with distinct fate potentials. In addition to the classical first and second heart fields, we observed the appearance of juxta-cardiac field progenitors giving rise to both myocardial and epicardial cells. Applying these findings to stem-cell based disease modelling we identified specific transcriptional dysregulation in first and second heart field progenitors derived from stem cells of patients with hypoplastic left heart syndrome. This highlights the suitability of our in vitro differentiation platform for studying human cardiac development and disease.