High-throughput widefield fluorescence imaging of 3D samples using deep learning for 2D projection image restoration.

Fluorescence microscopy is a core method for visualizing and quantifying the spatial and temporal dynamics of complex biological processes. While many fluorescent microscopy techniques exist, due to its cost-effectiveness and accessibility, widefield fluorescent imaging remains one of the most widely used. To accomplish imaging of 3D samples, conventional widefield fluorescence imaging entails acquiring a sequence of 2D images spaced along the z-dimension, typically called a z-stack. Oftentimes, the first step in an analysis pipeline is to project that 3D volume into a single 2D image because 3D image data can be cumbersome to manage and challenging to analyze and interpret. Furthermore, z-stack acquisition is often time-consuming, which consequently may induce photodamage to the biological sample; these are major barriers for workflows that require high-throughput, such as drug screening. As an alternative to z-stacks, axial sweep acquisition schemes have been proposed to circumvent these drawbacks and offer potential of 100-fold faster image acquisition for 3D-samples compared to z-stack acquisition. Unfortunately, these acquisition techniques generate low-quality 2D z-projected images that require restoration with unwieldy, computationally heavy algorithms before the images can be interrogated. We propose a novel workflow to combine axial z-sweep acquisition with deep learning-based image restoration, ultimately enabling high-throughput and high-quality imaging of complex 3D-samples using 2D projection images. To demonstrate the capabilities of our proposed workflow, we apply it to live-cell imaging of large 3D tumor spheroid cultures and find we can produce high-fidelity images appropriate for quantitative analysis. Therefore, we conclude that combining axial z-sweep image acquisition with deep learning-based image restoration enables high-throughput and high-quality fluorescence imaging of complex 3D biological samples.

LIVECell-A large-scale dataset for label-free live cell segmentation.

Light microscopy combined with well-established protocols of two-dimensional cell culture facilitates high-throughput quantitative imaging to study biological phenomena. Accurate segmentation of individual cells in images enables exploration of complex biological questions, but can require sophisticated imaging processing pipelines in cases of low contrast and high object density. Deep learning-based methods are considered state-of-the-art for image segmentation but typically require vast amounts of annotated data, for which there is no suitable resource available in the field of label-free cellular imaging. Here, we present LIVECell, a large, high-quality, manually annotated and expert-validated dataset of phase-contrast images, consisting of over 1.6 million cells from a diverse set of cell morphologies and culture densities. To further demonstrate its use, we train convolutional neural network-based models using LIVECell and evaluate model segmentation accuracy with a proposed a suite of benchmarks.

Tissue mechanics drives regeneration of a mucociliated epidermis on the surface of Xenopus embryonic aggregates.

Injury, surgery, and disease often disrupt tissues and it is the process of regeneration that aids the restoration of architecture and function. Regeneration can occur through multiple strategies including stem cell expansion, transdifferentiation, or proliferation of differentiated cells. We have identified a case of regeneration in Xenopus embryonic aggregates that restores a mucociliated epithelium from mesenchymal cells. Following disruption of embryonic tissue architecture and assembly of a compact mesenchymal aggregate, regeneration first restores an epithelium, transitioning from mesenchymal cells at the surface of the aggregate. Cells establish apico-basal polarity within 5 hours and a mucociliated epithelium within 24 hours. Regeneration coincides with nuclear translocation of the putative mechanotransducer YAP1 and a sharp increase in aggregate stiffness, and regeneration can be controlled by altering stiffness. We propose that regeneration of a mucociliated epithelium occurs in response to biophysical cues sensed by newly exposed cells on the surface of a disrupted mesenchymal tissue.

Spatiotemporally Controlled Mechanical Cues Drive Progenitor Mesenchymal-to-Epithelial Transition Enabling Proper Heart Formation and Function.

During early cardiogenesis, bilateral fields of mesenchymal heart progenitor cells (HPCs) move from the anterior lateral plate mesoderm to the ventral midline, undergoing a mesenchymal-to-epithelial transition (MET) en route to forming a single epithelial sheet. Through tracking of tissue-level deformations in the heart-forming region (HFR) as well as movement trajectories and traction generation of individual HPCs, we find that the onset of MET correlates with a peak in mechanical stress within the HFR and changes in HPC migratory behaviors. Small-molecule inhibitors targeting actomyosin contractility reveal a temporally specific requirement of bulk tissue compliance to regulate heart development and MET. Targeting mutant constructs to modulate contractility and compliance in the underlying endoderm, we find that MET in HPCs can be accelerated in response to microenvironmental stiffening and can be inhibited by softening. To test whether MET in HPCs was responsive to purely physical mechanical cues, we mimicked a high-stress state by injecting an inert oil droplet to generate high strain in the HFR, demonstrating that exogenously applied stress was sufficient to drive MET. MET-induced defects in anatomy result in defined functional lesions in the larval heart, implicating mechanical signaling and MET in the etiology of congenital heart defects. From this integrated analysis of HPC polarity and mechanics, we propose that normal heart development requires bilateral HPCs to undergo a critical behavioral and phenotypic transition on their way to the ventral midline, and that this transition is driven in response to the changing mechanical properties of their endoderm substrate.

On the role of mechanics in driving mesenchymal-to-epithelial transitions.

The mesenchymal-to-epithelial transition (MET) is an intrinsically mechanical process describing a multi-step progression where autonomous mesenchymal cells gradually become tightly linked, polarized epithelial cells. METs are fundamental to a wide range of biological processes, including the evolution of multicellular organisms, generation of primary and secondary epithelia during development and organogenesis, and the progression of diseases including cancer. In these cases, there is an interplay between the establishment of cell polarity and the mechanics of neighboring cells and microenvironment. In this review, we highlight a spectrum of METs found in normal development as well as in pathological lesions, and provide insight into the critical role mechanics play at each step. We define MET as an independent process, distinct from a reverse-EMT, and propose questions to further explore the cellular and physical mechanisms of MET.

Mechanochemical actuators of embryonic epithelial contractility.

Spatiotemporal regulation of cell contractility coordinates cell shape change to construct tissue architecture and ultimately directs the morphology and function of the organism. Here we show that contractility responses to spatially and temporally controlled chemical stimuli depend much more strongly on intercellular mechanical connections than on biochemical cues in both stimulated tissues and adjacent cells. We investigate how the cell contractility is triggered within an embryonic epithelial sheet by local ligand stimulation and coordinates a long-range contraction response. Our custom microfluidic control system allows spatiotemporally controlled stimulation with extracellular ATP, which results in locally distinct contractility followed by mechanical strain pattern formation. The stimulation-response circuit exposed here provides a better understanding of how morphogenetic processes integrate responses to stimulation and how intercellular responses are transmitted across multiple cells. These findings may enable one to create a biological actuator that actively drives morphogenesis.