Fast extraction of neuron morphologies from large-scale SBFSEM image stacks

Rapid structural remodeling of peripheral taste neurons is independent of taste cell turnover

Taste bud cells are constantly replaced in taste buds as old cells die and new cells migrate into the bud. The perception of taste relies on new taste bud cells integrating with existing neural circuitry, yet how these new cells connect with a taste ganglion neuron is unknown. Do taste ganglion neurons remodel to accommodate taste bud cell renewal? If so, how much of the structure of taste axons is fixed and how much remodels? Here, we measured the motility and branching of individual taste arbors (the portion of the axon innervating taste buds) in mice over time with two-photon in vivo microscopy. Terminal branches of taste arbors continuously and rapidly remodel within the taste bud. This remodeling is faster than predicted by taste bud cell renewal, with terminal branches added and lost concurrently. Surprisingly, blocking entry of new taste bud cells with chemotherapeutic agents revealed that remodeling of the terminal branches on taste arbors does not rely on the renewal of taste bud cells. Although terminal branch remodeling was fast and intrinsically controlled, no new arbors were added to taste buds, and few were lost over 100 days. Taste ganglion neurons maintain a stable number of arbors that are each capable of high-speed remodeling. We propose that terminal branch plasticity permits arbors to locate new taste bud cells, while stability of arbor number supports constancy in the degree of connectivity and function for each neuron over time.

A Segmentation Scheme for Complex Neuronal Arbors and Application to Vibration Sensitive Neurons in the Honeybee Brain

The morphology of a neuron is strongly related to its physiological properties, application of logical product and thus to information processing functions. Optical microscope images are widely used for extracting the structure of neurons. Although several approaches have been proposed to trace and extract complex neuronal structures from microscopy images, available methods remain prone to errors. In this study, we present a practical scheme for processing confocal microscope images and reconstructing neuronal structures. We evaluated this scheme using image data samples and associated “gold standard” reconstructions from the BigNeuron Project. In these samples, dendritic arbors belonging to multiple projection branches of the same neuron overlapped in space, making it difficult to automatically and accurately trace their structural connectivity. Our proposed scheme, which combines several software tools for image masking and filtering with an existing tool for dendritic segmentation and tracing, outperformed state-of-the-art automatic methods in reconstructing such neuron structures. For evaluating our scheme, we applied it to a honeybee local interneuron, DL-Int-1, which has complex arbors and is considered to be a critical neuron for encoding the distance information indicated in the waggle dance of the honeybee.

Hierarchical integration of porosity in shales

Pore characterization in shales is challenging owing to the wide range of pore sizes and types present. Haynesville-Bossier shale (USA) was sampled as a typical clay-bearing siliceous, organic-rich, gas-mature shale and characterized over pore diameters ranging 2 nm to 3000 nm. Three advanced imaging techniques were utilized correlatively, including the application of Xe⁺ plasma focused ion beam scanning electron microscopy (plasma FIB or PFIB), complemented by the Ga⁺ FIB method which is now frequently used to characterise porosity and organic/inorganic phases, together with transmission electron microscope tomography of the nano-scale pores (voxel size 0.6 nm; resolution 1-2 nm). The three pore-size scales each contribute differently to the pore network. Those <10 nm (greatest number), 10 nm to 100 nm (best-connected hence controls transport properties), and >100 nm (greatest total volume hence determines fluid storativity). Four distinct pore types were found: intra-organic, organic-mineral interface, inter-mineral and intra-mineral pores were recognized, with characteristic geometries. The whole pore network comprises a globally-connected system between phyllosilicate mineral grains (diameter: 6-50 nm), and locally-clustered connected pores within porous organic matter (diameter: 200-800 nm). Integrated predictions of pore geometry, connectivity, and roles in controlling petrophysical properties were verified through experimental permeability measurements.

Generation of dense statistical connectomes from sparse morphological data

Sensory-evoked signal flow, at cellular and network levels, is primarily determined by the synaptic wiring of the underlying neuronal circuitry. Measurements of synaptic innervation, connection probabilities and subcellular organization of synaptic inputs are thus among the most active fields of research in contemporary neuroscience. Methods to measure these quantities range from electrophysiological recordings over reconstructions of dendrite-axon overlap at light-microscopic levels to dense circuit reconstructions of small volumes at electron-microscopic resolution. However, quantitative and complete measurements at subcellular resolution and mesoscopic scales to obtain all local and long-range synaptic in/outputs for any neuron within an entire brain region are beyond present methodological limits. Here, we present a novel concept, implemented within an interactive software environment called NeuroNet, which allows (i) integration of sparsely sampled (sub)cellular morphological data into an accurate anatomical reference frame of the brain region(s) of interest, (ii) up-scaling to generate an average dense model of the neuronal circuitry within the respective brain region(s) and (iii) statistical measurements of synaptic innervation between all neurons within the model. We illustrate our approach by generating a dense average model of the entire rat vibrissal cortex, providing the required anatomical data, and illustrate how to measure synaptic innervation statistically. Comparing our results with data from paired recordings in vitro and in vivo, as well as with reconstructions of synaptic contact sites at light- and electron-microscopic levels, we find that our in silico measurements are in line with previous results.

Serial block-face imaging and its potential for reconstructing diminutive cell systems: a case study from arthropods

Until recently, three-dimensional reconstruction on an ultrastructural level was only possible using serial section transmission electron microscopy (ssTEM). However, ssTEM is highly challenging and prone to artifacts as, e.g., section loss and image distortions. New methods, such as serial block-face scanning electron microscopy (SBFSEM) overcome these limitations and promise a high lateral resolution. However, little is known about the usability of SBFSEM in diminutive, but highly complex cellular systems. We used spider sperm (~3 µm in diameter), which fulfills these conditions, to analyze the potential of SBFSEM compared with ssTEM. Our data suggest that the resolution obtained by SBFSEM allows depicting structures on a cellular level and is sufficient to discriminate subcellular components, but is highly dependent on previous staining procedures and electron density of the target structures.

Beyond counts and shapes: studying pathology of dendritic spines in the context of the surrounding neuropil through serial section electron microscopy

Because dendritic spines are the sites of excitatory synapses, pathological changes in spine morphology should be considered as part of pathological changes in neuronal circuitry in the forms of synaptic connections and connectivity strength. In the past, spine pathology has usually been measured by changes in their number or shape. A more complete understanding of spine pathology requires visualization at the nanometer level to analyze how the changes in number and size affect their presynaptic partners and associated astrocytic processes, as well as organelles and other intracellular structures. Currently, serial section electron microscopy (ssEM) offers the best approach to address this issue because of its ability to image the volume of brain tissue at the nanometer resolution. Renewed interest in ssEM has led to recent technological advances in imaging techniques and improvements in computational tools indispensable for three-dimensional analyses of brain tissue volumes. Here we consider the small but growing literature that has used ssEM analysis to unravel ultrastructural changes in neuropil including dendritic spines. These findings have implications in altered synaptic connectivity and cell biological processes involved in neuropathology, and serve as anatomical substrates for understanding changes in network activity that may underlie clinical symptoms.

Computational tissue volume reconstruction of a peripheral nerve using high-resolution light-microscopy and reconstruct

The development of neural cuff-electrodes requires several in vivo studies and revisions of the electrode design before the electrode is completely adapted to its target nerve. It is therefore favorable to simulate many of the steps involved in this process to reduce costs and animal testing. As the restoration of motor function is one of the most interesting applications of cuff-electrodes, the position and trajectories of myelinated fibers in the simulated nerve are important. In this paper, we investigate a method for building a precise neuroanatomical model of myelinated fibers in a peripheral nerve based on images obtained using high-resolution light microscopy. This anatomical model describes the first aim of our "Virtual workbench" project to establish a method for creating realistic neural simulation models based on image datasets. The imaging, processing, segmentation and technical limitations are described, and the steps involved in the transition into a simulation model are presented. The results showed that the position and trajectories of the myelinated axons were traced and virtualized using our technique, and small nerves could be reliably modeled based on of light microscopy images using low-cost OpenSource software and standard hardware. The anatomical model will be released to the scientific community.

Large-volume reconstruction of brain tissue from high-resolution serial section images acquired by SEM-based scanning transmission electron microscopy

With recent improvements in instrumentation and computational tools, serial section electron microscopy has become increasingly straightforward. A new method for imaging ultrathin serial sections is developed based on a field emission scanning electron microscope fitted with a transmitted electron detector. This method is capable of automatically acquiring high-resolution serial images with a large field size and very little optical and physical distortions. In this chapter, we describe the procedures leading to the generation and analyses of a large-volume stack of high-resolution images (64 μm × 64 μm × 10 μm, or larger, at 2 nm pixel size), including how to obtain large-area serial sections of uniform thickness from well-preserved brain tissue that is rapidly perfusion-fixed with mixed aldehydes, processed with a microwave-enhanced method, and embedded into epoxy resin.

Marching cubes technique for volumetric visualization accelerated with graphics processing units

Volume visualization has numerous applications that benefit different knowledge domains, such as biology, medicine, meteorology, oceanography, geology, among others. With the continuous advances of technology, it has been possible to achieve considerable rendering rates and a high degree of realism. Visualization tools have currently assisted users with the visual analysis of complex and large datasets. Marching cubes is one of the most widely used real-time volume rendering methods. This paper describes a methodology for speeding up the marching cubes algorithm on a graphics processing unit and discusses a number of ways to improve its performance by means of auxiliary spatial data structures. Experiments conducted with use of several volumetric datasets demonstrate the effectiveness of the developed method.

A multiscale parallel computing architecture for automated segmentation of the brain connectome

Several groups in neurobiology have embarked into deciphering the brain circuitry using large-scale imaging of a mouse brain and manual tracing of the connections between neurons. Creating a graph of the brain circuitry, also called a connectome, could have a huge impact on the understanding of neurodegenerative diseases such as Alzheimer's disease. Although considerably smaller than a human brain, a mouse brain already exhibits one billion connections and manually tracing the connectome of a mouse brain can only be achieved partially. This paper proposes to scale up the tracing by using automated image segmentation and a parallel computing approach designed for domain experts. We explain the design decisions behind our parallel approach and we present our results for the segmentation of the vasculature and the cell nuclei, which have been obtained without any manual intervention.