Myosin cables control the timing of tissue internalization in the Drosophila embryo.

Compartment boundaries prevent cell mixing during animal development. In the early Drosophila embryo, the mesectoderm is a group of glial precursors that separate ectoderm and mesoderm, forming the ventral midline. Mesectoderm cells undergo one round of oriented divisions during axis elongation and are eventually internalized 6 h later. Using spinning disk confocal microscopy and image analysis, we found that after dividing, mesectoderm cells reversed their planar polarity. The polarity factor Bazooka was redistributed to mesectoderm-mesectoderm cell interfaces, and the molecular motor non-muscle Myosin II and its upstream activator Rho-kinase (Rok) accumulated at mesectoderm-ectoderm (ME) interfaces, forming supracellular cables flanking the mesectoderm on either side of the tissue. Laser ablation revealed the presence of increased tension at ME cables, where Myosin was stabilized, as shown by fluorescence recovery after photobleaching. We used laser nanosurgery to reduce tension at the ME boundary, and we found that Myosin fluorescence decreased rapidly, suggesting a role for tension in ME boundary maintenance. Mathematical modelling predicted that increased tension at the ME boundary was necessary to prevent the premature establishment of contacts between the two ectodermal sheets on opposite sides of the mesectoderm, thus controlling the timing of mesectoderm internalization. We validated the model in vivo: Myosin inhibition disrupted the linearity of the ME boundary and resulted in early internalization of the mesectoderm. Our results suggest that the redistribution of Rok polarizes Myosin and Bazooka within the mesectoderm to establish tissue boundaries, and that ME boundaries control the timely internalization of the mesectoderm as embryos develop.

REEP5 depletion causes sarco-endoplasmic reticulum vacuolization and cardiac functional defects.

The sarco-endoplasmic reticulum (SR/ER) plays an important role in the development and progression of many heart diseases. However, many aspects of its structural organization remain largely unknown, particularly in cells with a highly differentiated SR/ER network. Here, we report a cardiac enriched, SR/ER membrane protein, REEP5 that is centrally involved in regulating SR/ER organization and cellular stress responses in cardiac myocytes. In vitro REEP5 depletion in mouse cardiac myocytes results in SR/ER membrane destabilization and luminal vacuolization along with decreased myocyte contractility and disrupted Ca2+ cycling. Further, in vivo CRISPR/Cas9-mediated REEP5 loss-of-function zebrafish mutants show sensitized cardiac dysfunction upon short-term verapamil treatment. Additionally, in vivo adeno-associated viral (AAV9)-induced REEP5 depletion in the mouse demonstrates cardiac dysfunction. These results demonstrate the critical role of REEP5 in SR/ER organization and function as well as normal heart function and development.

Role of α-Catenin and its mechanosensing properties in regulating Hippo/YAP-dependent tissue growth.

α-catenin is a key protein of adherens junctions (AJs) with mechanosensory properties. It also acts as a tumor suppressor that limits tissue growth. Here we analyzed the function of Drosophila α-Catenin (α-Cat) in growth regulation of the wing epithelium. We found that different α-Cat levels led to a differential activation of Hippo/Yorkie or JNK signaling causing tissue overgrowth or degeneration, respectively. α-Cat can modulate Yorkie-dependent tissue growth through recruitment of Ajuba, a negative regulator of Hippo signaling to AJs but also through a mechanism independent of Ajuba recruitment to AJs. Both mechanosensory regions of α-Cat, the M region and the actin-binding domain (ABD), contribute to growth regulation. Whereas M is dispensable for α-Cat function in the wing, individual M domains (M1, M2, M3) have opposing effects on growth regulation. In particular, M1 limits Ajuba recruitment. Loss of M1 causes Ajuba hyper-recruitment to AJs, promoting tissue-tension independent overgrowth. Although M1 binds Vinculin, Vinculin is not responsible for this effect. Moreover, disruption of mechanosensing of the α-Cat ABD affects tissue growth, with enhanced actin interactions stabilizing junctions and leading to tissue overgrowth. Together, our findings indicate that α-Cat acts through multiple mechanisms to control tissue growth, including regulation of AJ stability, mechanosensitive Ajuba recruitment, and dynamic direct F-actin interactions.

Quantitative modelling of epithelial morphogenesis: integrating cell mechanics and molecular dynamics.

Epithelial tissues form and repair in complex processes influenced by molecular and physical factors. Recent years have witnessed the development of new microscopy modalities that push the limits of spatial resolution, and enable long-term monitoring of developing animals. Increasingly, methods from the physical sciences are used to investigate the role of mechanical forces in living organisms. The application of these new technologies to developmental biology has led to ever-expanding volumes of data that must be interpreted and integrated. For these reasons, computer models are being applied to investigate tissue morphogenesis. Here, we discuss the use of vertex models to study the morphogenesis of epithelial tissues. We motivate the use of computational models and consider their advantages and limitations. We provide an introduction to the theoretical foundation of vertex models and describe how they can integrate mechanical and biochemical dynamics. Finally, we review recent advances in the application of vertex models to investigate dorsal closure, a morphogenetic process in the Drosophila embryo with parallels to both embryonic development and wound repair in vertebrate organisms.