Deep learning-based aberration compensation improves contrast and resolution in fluorescence microscopy.

Optical aberrations hinder fluorescence microscopy of thick samples, reducing image signal, contrast, and resolution. Here we introduce a deep learning-based strategy for aberration compensation, improving image quality without slowing image acquisition, applying additional dose, or introducing more optics into the imaging path. Our method (i) introduces synthetic aberrations to images acquired on the shallow side of image stacks, making them resemble those acquired deeper into the volume and (ii) trains neural networks to reverse the effect of these aberrations. We use simulations to show that applying the trained 'de-aberration' networks outperforms alternative methods, and subsequently apply the networks to diverse datasets captured with confocal, light-sheet, multi-photon, and super-resolution microscopy. In all cases, the improved quality of the restored data facilitates qualitative image inspection and improves downstream image quantitation, including orientational analysis of blood vessels in mouse tissue and improved membrane and nuclear segmentation in C. elegans embryos.

Incorporating the image formation process into deep learning improves network performance.

We present Richardson-Lucy network (RLN), a fast and lightweight deep learning method for three-dimensional fluorescence microscopy deconvolution. RLN combines the traditional Richardson-Lucy iteration with a fully convolutional network structure, establishing a connection to the image formation process and thereby improving network performance. Containing only roughly 16,000 parameters, RLN enables four- to 50-fold faster processing than purely data-driven networks with many more parameters. By visual and quantitative analysis, we show that RLN provides better deconvolution, better generalizability and fewer artifacts than other networks, especially along the axial dimension. RLN outperforms classic Richardson-Lucy deconvolution on volumes contaminated with severe out of focus fluorescence or noise and provides four- to sixfold faster reconstructions of large, cleared-tissue datasets than classic multi-view pipelines. We demonstrate RLN's performance on cells, tissues and embryos imaged with widefield-, light-sheet-, confocal- and super-resolution microscopy.

Differential adhesion regulates neurite placement via a retrograde zippering mechanism.

During development, neurites and synapses segregate into specific neighborhoods or layers within nerve bundles. The developmental programs guiding placement of neurites in specific layers, and hence their incorporation into specific circuits, are not well understood. We implement novel imaging methods and quantitative models to document the embryonic development of the C. elegans brain neuropil, and discover that differential adhesion mechanisms control precise placement of single neurites onto specific layers. Differential adhesion is orchestrated via developmentally regulated expression of the IgCAM SYG-1, and its partner ligand SYG-2. Changes in SYG-1 expression across neuropil layers result in changes in adhesive forces, which sort SYG-2-expressing neurons. Sorting to layers occurs, not via outgrowth from the neurite tip, but via an alternate mechanism of retrograde zippering, involving interactions between neurite shafts. Our study indicates that biophysical principles from differential adhesion govern neurite placement and synaptic specificity in vivo in developing neuropil bundles.

Structural and developmental principles of neuropil assembly in C. elegans.

Neuropil is a fundamental form of tissue organization within the brain1, in which densely packed neurons synaptically interconnect into precise circuit architecture2,3. However, the structural and developmental principles that govern this nanoscale precision remain largely unknown4,5. Here we use an iterative data coarse-graining algorithm termed 'diffusion condensation'6 to identify nested circuit structures within the Caenorhabditis elegans neuropil, which is known as the nerve ring. We show that the nerve ring neuropil is largely organized into four strata that are composed of related behavioural circuits. The stratified architecture of the neuropil is a geometrical representation of the functional segregation of sensory information and motor outputs, with specific sensory organs and muscle quadrants mapping onto particular neuropil strata. We identify groups of neurons with unique morphologies that integrate information across strata and that create neural structures that cage the strata within the nerve ring. We use high resolution light-sheet microscopy7,8 coupled with lineage-tracing and cell-tracking algorithms9,10 to resolve the developmental sequence and reveal principles of cell position, migration and outgrowth that guide stratified neuropil organization. Our results uncover conserved structural design principles that underlie the architecture and function of the nerve ring neuropil, and reveal a temporal progression of outgrowth-based on pioneer neurons-that guides the hierarchical development of the layered neuropil. Our findings provide a systematic blueprint for using structural and developmental approaches to understand neuropil organization within the brain.

A polymer index-matched to water enables diverse applications in fluorescence microscopy.

We demonstrate diffraction-limited and super-resolution imaging through thick layers (tens-hundreds of microns) of BIO-133, a biocompatible, UV-curable, commercially available polymer with a refractive index (RI) matched to water. We show that cells can be directly grown on BIO-133 substrates without the need for surface passivation and use this capability to perform extended time-lapse volumetric imaging of cellular dynamics 1) at isotropic resolution using dual-view light-sheet microscopy, and 2) at super-resolution using instant structured illumination microscopy. BIO-133 also enables immobilization of 1) Drosophila tissue, allowing us to track membrane puncta in pioneer neurons, and 2) Caenorhabditis elegans, which allows us to image and inspect fine neural structure and to track pan-neuronal calcium activity over hundreds of volumes. Finally, BIO-133 is compatible with other microfluidic materials, enabling optical and chemical perturbation of immobilized samples, as we demonstrate by performing drug and optogenetic stimulation on cells and C. elegans.

A journey to 'tame a small metazoan organism', ‡ seen through the artistic eyes of C. elegans researchers.

In the following pages, we share a collection of photos, drawings, and mixed-media creations, most of them especially made for this JoN issue, manifesting C. elegans researchers' affection for their model organism and the founders of the field. This is a celebration of our community's growth, flourish, spread, and bright future. Descriptions provided by the contributors, edited for space. 1.

Cadherin preserves cohesion across involuting tissues during C. elegans neurulation.

The internalization of the central nervous system, termed neurulation in vertebrates, is a critical step in embryogenesis. Open questions remain regarding how force propels coordinated tissue movement during the process, and little is known as to how internalization happens in invertebrates. We show that in C. elegans morphogenesis, apical constriction in the retracting pharynx drives involution of the adjacent neuroectoderm. HMR-1/cadherin mediates this process via inter-tissue attachment, as well as cohesion within the neuroectoderm. Our results demonstrate that HMR-1 is capable of mediating embryo-wide reorganization driven by a centrally located force generator, and indicate a non-canonical use of cadherin on the basal side of an epithelium that may apply to vertebrate neurulation. Additionally, we highlight shared morphology and gene expression in tissues driving involution, which suggests that neuroectoderm involution in C. elegans is potentially homologous with vertebrate neurulation and thus may help elucidate the evolutionary origin of the brain.

Rapid image deconvolution and multiview fusion for optical microscopy.

The contrast and resolution of images obtained with optical microscopes can be improved by deconvolution and computational fusion of multiple views of the same sample, but these methods are computationally expensive for large datasets. Here we describe theoretical and practical advances in algorithm and software design that result in image processing times that are tenfold to several thousand fold faster than with previous methods. First, we show that an 'unmatched back projector' accelerates deconvolution relative to the classic Richardson-Lucy algorithm by at least tenfold. Second, three-dimensional image-based registration with a graphics processing unit enhances processing speed 10- to 100-fold over CPU processing. Third, deep learning can provide further acceleration, particularly for deconvolution with spatially varying point spread functions. We illustrate our methods from the subcellular to millimeter spatial scale on diverse samples, including single cells, embryos and cleared tissue. Finally, we show performance enhancement on recently developed microscopes that have improved spatial resolution, including dual-view cleared-tissue light-sheet microscopes and reflective lattice light-sheet microscopes.

The G2-to-M Transition Is Ensured by a Dual Mechanism that Protects Cyclin B from Degradation by Cdc20-Activated APC/C.

In the eukaryotic cell cycle, a threshold level of cyclin B accumulation triggers the G2-to-M transition, and subsequent cyclin B destruction triggers mitotic exit. The anaphase-promoting complex/cyclosome (APC/C) is the E3 ubiquitin ligase that, together with its co-activator Cdc20, targets cyclin B for destruction during mitotic exit. Here, we show that two pathways act in concert to protect cyclin B from Cdc20-activated APC/C in G2, in order to enable cyclin B accumulation and the G2-to-M transition. The first pathway involves the Mad1-Mad2 spindle checkpoint complex, acting in a distinct manner from checkpoint signaling after mitotic entry but employing a common molecular mechanism-the promotion of Mad2-Cdc20 complex formation. The second pathway involves cyclin-dependent kinase phosphorylation of Cdc20, which is known to reduce Cdc20's affinity for the APC/C. Cooperation of these two mechanisms, which target distinct APC/C binding interfaces of Cdc20, enables cyclin B accumulation and the G2-to-M transition.

Isotropic Light-Sheet Microscopy and Automated Cell Lineage Analyses to Catalogue Caenorhabditis elegans Embryogenesis with Subcellular Resolution.

Caenorhabditis elegans (C. elegans) stands out as the only organism in which the challenge of understanding the cellular origins of an entire nervous system can be observed, with single cell resolution, in vivo. Here, we present an integrated protocol for the examination of neurodevelopment in C. elegans embryos. Our protocol combines imaging, lineaging and neuroanatomical tracing of single cells in developing embryos. We achieve long-term, four-dimensional (4D) imaging of living C. elegans embryos with nearly isotropic spatial resolution through the use of Dual-view Inverted Selective Plane Illumination Microscopy (diSPIM). Nuclei and neuronal structures in the nematode embryos are imaged and isotropically fused to yield images with resolution of ~330 nm in all three dimensions. These minute-by-minute high-resolution 4D data sets are then analyzed to correlate definitive cell-lineage identities with gene expression and morphological dynamics at single-cell and subcellular levels of detail. Our protocol is structured to enable modular implementation of each of the described steps and enhance studies on embryogenesis, gene expression, or neurodevelopment.