The establishment of separated pulmonary and systemic circulation in vertebrates, via cardiac outflow tract (OFT) septation, is a sensitive developmental process accounting for 10% of all congenital anomalies. Neural Crest Cells (NCC) colonising the heart condensate along the primitive endocardial tube and force its scission into two tubes. Here, we show that NCC aggregation progressively decreases along the OFT distal-proximal axis following a BMP signalling gradient. Dullard, a nuclear phosphatase, tunes the BMP gradient amplitude and prevents NCC premature condensation. Dullard maintains transcriptional programs providing NCC with mesenchymal traits. It attenuates the expression of the aggregation factor Sema3c and conversely promotes that of the epithelial-mesenchymal transition driver Twist1. Altogether, Dullard-mediated fine-tuning of BMP signalling ensures the timed and progressive zipper-like closure of the OFT by the NCC and prevents the formation of a heart carrying the congenital abnormalities defining the tetralogy of Fallot.

Generating comprehensive image maps, while preserving spatial 3D context, is essential to quantitatively assess and locate specific cellular features and cell-cell interactions during organ development. Despite the recent advances in 3D imaging approaches, our current knowledge of the spatial organization of distinct cell types in the embryonic pancreatic tissue is still largely based on 2D histological sections. Here, we present a light-sheet fluorescence microscopy approach to image the pancreas in 3D and map tissue interactions at key time points in the mouse embryo. We demonstrate the utility of the approach by providing volumetric data, 3D distribution of three main cellular components (epithelial, mesenchymal, endothelial) within the developing pancreas, and quantification of their relative cellular abundance within the tissue. Interestingly, our 3D images show that endocrine cells are constantly and increasingly in contact with endothelial cells forming small vessels, while the interactions with mesenchymal cells decrease over time. These findings suggest distinct cell-cell interaction requirement for early endocrine cell specification and late differentiation. Lastly, we combine our image data in an open-source online repository (referred to as Pancreas Embryonic Cell Atlas).

During organ formation, progenitor cells need to acquire the diversity of cell identities found in the organ as well as organize themselves into distinct structural units. How these processes are coordinated, and how tissue architecture(s) are preserved despite the dramatic cell rearrangements occurring in developing organs remain unclear. Here, we identified cellular rearrangements between acinar and ductal progenitors as a mechanism to drive branching morphogenesis in the pancreas while preserving the integrity of the acinar-ductal functional unit. Using ex vivo and in vivo mouse models, we found that pancreatic ductal cells form clefts by protruding and pulling on the acinar basement membrane, which lead to acini splitting. Newly formed acini remain connected to bifurcated branches generated by ductal cell rearrangement. IGF/PI3K pathway regulates this process by controlling ductal cell fluidity. If components of the pathway are genetically or chemically dysregulated, ductal cell fluidity prevents branching and affects pancreatic cell fates. Hence, our results explain how acinar multiplication and branch bifurcation are synchronized during pancreas organogenesis.

No abstract

Generating comprehensive image maps, while preserving spatial 3D context, is essential to quantitatively assess and locate specific cellular features and cell-cell interactions during organ development. Despite the recent advances in 3D imaging approaches, our current knowledge of the spatial organization of distinct cell types in the embryonic pancreatic tissue is still largely based on 2D histological sections. Here, we present a light-sheet fluorescence microscopy approach to image the pancreas in 3D and map tissue interactions at key development time points in the mouse embryo. We used transgenic mouse models and antibodies to visualize the three main cellular components within the developing pancreas, including epithelial, mesenchymal and endothelial cell populations. We demonstrated the utility of the approach by providing volumetric data, 3D distribution of distinct progenitor populations and quantification of relative cellular abundance within the tissue. Lastly, our image data were combined in an open source online repository (referred to as Pancreas Embryonic Cell Atlas). This image dataset will serve the scientific community by enabling further investigation on pancreas organogenesis but also for devising strategies for the in vitro generation of transplantable pancreatic tissue for regenerative therapies.