Neurotransmitter classification from electron microscopy images at synaptic sites in Drosophila melanogaster.

High-resolution electron microscopy of nervous systems has enabled the reconstruction of synaptic connectomes. However, we do not know the synaptic sign for each connection (i.e., whether a connection is excitatory or inhibitory), which is implied by the released transmitter. We demonstrate that artificial neural networks can predict transmitter types for presynapses from electron micrographs: a network trained to predict six transmitters (acetylcholine, glutamate, GABA, serotonin, dopamine, octopamine) achieves an accuracy of 87% for individual synapses, 94% for neurons, and 91% for known cell types across a D. melanogaster whole brain. We visualize the ultrastructural features used for prediction, discovering subtle but significant differences between transmitter phenotypes. We also analyze transmitter distributions across the brain and find that neurons that develop together largely express only one fast-acting transmitter (acetylcholine, glutamate, or GABA). We hope that our publicly available predictions act as an accelerant for neuroscientific hypothesis generation for the fly.

A petavoxel fragment of human cerebral cortex reconstructed at nanoscale resolution.

To fully understand how the human brain works, knowledge of its structure at high resolution is needed. Presented here is a computationally intensive reconstruction of the ultrastructure of a cubic millimeter of human temporal cortex that was surgically removed to gain access to an underlying epileptic focus. It contains about 57,000 cells, about 230 millimeters of blood vessels, and about 150 million synapses and comprises 1.4 petabytes. Our analysis showed that glia outnumber neurons 2:1, oligodendrocytes were the most common cell, deep layer excitatory neurons could be classified on the basis of dendritic orientation, and among thousands of weak connections to each neuron, there exist rare powerful axonal inputs of up to 50 synapses. Further studies using this resource may bring valuable insights into the mysteries of the human brain.

Array tomography of in vivo labeled synaptic receptors.

Array tomography (AT) allows one to localize sub-cellular components within the structural context of cells in 3D through the imaging of serial sections. Using this technique, the z-resolution can be improved physically by cutting ultra-thin sections. Nevertheless, conventional immunofluorescence staining of those sections is time consuming and requires relatively large amounts of costly antibody solutions. Moreover, epitopes are only readily accessible at the section's surface, leaving the volume of the serial sections unlabeled. Localization of receptors at neuronal synapses in 3D in their native cellular ultrastructural context is important for understanding signaling processes. Here, we present in vivo labeling of receptors via fluorophore-coupled tags in combination with super-resolution AT. We present two workflows where we label receptors at the plasma membrane: first, in vivo labeling via microinjection with a setup consisting of readily available components and self-manufactured microscope table equipment and second, live receptor labeling by using a cell-permeable tag. To take advantage of a near-to-native preservation of tissues for subsequent scanning electron microscopy (SEM), we also apply high-pressure freezing and freeze substitution. The advantages and disadvantages of our workflows are discussed.

Targeting of membrane proteins with fluoronanogold probes for high-resolution correlative microscopy.

Correlative light and electron microscopy (CLEM) can provide valuable information about a biological sample by giving information on the specific localization of a molecule of interest within an ultrastructural context. In this work, we describe a simple CLEM method to obtain high-resolution images of neurotransmitter receptor distribution in synapses by electron microscopy (EM). We use hippocampal organotypic slices from a previously reported mouse model expressing a modified AMPA receptor (AMPAR) subunit that binds biotin at the surface (Getz et al., 2022). This tag can be recognized by StreptAvidin-Fluoronanogold™ conjugates (SA-FNG), which reach receptors at synapses (synaptic cleft is 50-100nm thick). By using pre-embedding labeling, we found that SA-FNG reliably bind synaptic receptors and penetrate around 10-15µm in depth in live tissue. However, the silver enhancement was only reaching the surface of the slices. We show that permeabilization with triton is highly effective at increasing the in depth-gold amplification and that the membrane integrity is well preserved. Finally, we also apply high-resolution electron tomography, thus providing important information about the 3D organization of surface AMPA receptors in synapses at the nanoscale.

RoboEM: automated 3D flight tracing for synaptic-resolution connectomics.

Mapping neuronal networks from three-dimensional electron microscopy (3D-EM) data still poses substantial reconstruction challenges, in particular for thin axons. Currently available automated image segmentation methods require manual proofreading for many types of connectomic analysis. Here we introduce RoboEM, an artificial intelligence-based self-steering 3D 'flight' system trained to navigate along neurites using only 3D-EM data as input. Applied to 3D-EM data from mouse and human cortex, RoboEM substantially improves automated state-of-the-art segmentations and can replace manual proofreading for more complex connectomic analysis problems, yielding computational annotation cost for cortical connectomes about 400-fold lower than the cost of manual error correction.

Synaptic wiring motifs in posterior parietal cortex support decision-making.

The posterior parietal cortex exhibits choice-selective activity during perceptual decision-making tasks1-10. However, it is not known how this selective activity arises from the underlying synaptic connectivity. Here we combined virtual-reality behaviour, two-photon calcium imaging, high-throughput electron microscopy and circuit modelling to analyse how synaptic connectivity between neurons in the posterior parietal cortex relates to their selective activity. We found that excitatory pyramidal neurons preferentially target inhibitory interneurons with the same selectivity. In turn, inhibitory interneurons preferentially target pyramidal neurons with opposite selectivity, forming an opponent inhibition motif. This motif was present even between neurons with activity peaks in different task epochs. We developed neural-circuit models of the computations performed by these motifs, and found that opponent inhibition between neural populations with opposite selectivity amplifies selective inputs, thereby improving the encoding of trial-type information. The models also predict that opponent inhibition between neurons with activity peaks in different task epochs contributes to creating choice-specific sequential activity. These results provide evidence for how synaptic connectivity in cortical circuits supports a learned decision-making task.

Microanatomy and ultrastructure of the nervous system of adult Renicola parvicaudatus (Digenea: Renicolidae).

The digenean complex life cycle includes various morphological forms with different locomotory and behavioral activities, and the functional specialization of their nervous system is of importance for the transmission of these parasites. Adult digeneans acquire many adaptive features associated with the final settlement in a vertebrate host. Our study describes the general morphology and ultrastructure of the nervous system of the adult renicolid digenean Renicola parvicaudatus parasitizing the renal tubules of herring gulls. Using immunocytochemical and electron microscopic methods, we identified the distinctive characteristics of ganglia and synapses in the studied species. A comparative analysis of the organization of the nervous system of adult individuals and their continuously-swimming stylet cercariae revealed a number of stage-related differences in the composition of ganglia, the distribution of serotonin- and FMRFamide-immunoreactive neurons, the cytomorphology of neuron somata and free sensory endings. Thus, in adults, the presence of FMRFamide-positive neuron somata, accessory muscle bundles in the ganglionic cortex, and eight types of neuronal vesicles was detected, but no glia-like elements were identified. Their neurons are characterized by a larger volume of cytoplasm and also show greater ultrastructural diversity. Although the sensory papillae of adults do not vary in their external morphology as much as those of larvae, their sensory bulbs are more diverse in cytomorphology. Following our previous data on the "support" cell processes related to various tissues of the larvae and considered as glia-like structures, we also briefly present the identified features of the parenchyma, attachment organs and excretory system of adult individuals. The excretory system of adult R. parvicaudatus is characterized by the presence of unique terminal cells with several flame tufts, which are not typical either for the larvae of this species or for other digeneans studied so far. We also used molecular phylogenetic analysis to clarify species identification.

Light and electron microscopic observations on retinal neurons of red-tail shark (Epalzeorhynchos bicolor H. M. Smith, 1931).

The structure of photoreceptors (PR) and the arrangement of neurons in the retina of red-tail shark were investigated using light and electron microscopy. The PR showed a mosaic arrangement and included double cones, single cones (SC), and single rods. Most cones occur as SC. The ratio between the number of cones and rods was 3:1.39 (±0.29). The rods were tall that reached the pigmented epithelium. The outer plexiform layer (OPL) showed a complex synaptic connection between the horizontal and photoreceptor terminals that were surrounded by Müller cell processes. Electron microscopy showed that the OPL possessed both cone pedicles and rod spherules. Each rod spherule consisted of a single synaptic ribbon within the invaginating terminal endings of the horizontal cell (hc) processes. In contrast, the cone pedicles possessed many synaptic ribbons within their junctional complexes. The inner nuclear layer consisted of bipolar, amacrine, Müller cells, and hc. Müller cells possessed intermediate filaments and cell processes that can reach the outer limiting membrane and form connections with each other by desmosomes. The ganglion cells were large multipolar cells with a spherical nucleus and Nissl' bodies in their cytoplasm. The presence of different types of cones arranged in a mosaic pattern in the retina of this species favors the spatial resolution of visual objects. RESEARCH HIGHLIGHTS: This is the first study demonstrating the structure and arrangement of retinal neurons of red-tail shark using light and electron microscopy. The current study showed the presence of different types of cones arranged in a mosaic pattern that may favor the spatial resolution of visual objects in this species. The bipolar, amacrine, Müller, and horizontal cells could be demonstrated.

Ultrastructural analysis of nigrostriatal dopaminergic terminals in a knockin mouse model of DYT1 dystonia.

DYT1 dystonia is associated with decreased striatal dopamine release. In this study, we examined the possibility that ultrastructural changes of nigrostriatal dopamine terminals could contribute to this neurochemical imbalance using a serial block face/scanning electron microscope (SBF/SEM) and three-dimensional reconstruction to analyse striatal tyrosine hydroxylase-immunoreactive (TH-IR) terminals and their synapses in a DYT1(ΔE) knockin (DYT1-KI) mouse model of DYT1 dystonia. Furthermore, to study possible changes in vesicle packaging capacity of dopamine, we used transmission electron microscopy to assess the synaptic vesicle size in striatal dopamine terminals. Quantitative comparative analysis of 80 fully reconstructed TH-IR terminals in the WT and DYT1-KI mice indicate (1) no significant difference in the volume of TH-IR terminals; (2) no major change in the proportion of axo-spinous versus axo-dendritic synapses; (3) no significant change in the post-synaptic density (PSD) area of axo-dendritic synapses, while the PSDs of axo-spinous synapses were significantly smaller in DYT1-KI mice; (4) no significant change in the contact area between TH-IR terminals and dendritic shafts or spines, while the ratio of PSD area/contact area decreased significantly for both axo-dendritic and axo-spinous synapses in DYT1-KI mice; (5) no significant difference in the mitochondria volume; and (6) no significant difference in the synaptic vesicle area between the two groups. Altogether, these findings suggest that abnormal morphometric changes of nigrostriatal dopamine terminals and their post-synaptic targets are unlikely to be a major source of reduced striatal dopamine release in DYT1 dystonia.

Ultrastructure of noise-induced cochlear synaptopathy.

Acoustic overexposure can eliminate synapses between inner hair cells (IHCs) and auditory nerve fibers (ANFs), even if hair-cell function recovers. This synaptopathy has been extensively studied by confocal microscopy, however, understanding the nature and sequence of damage requires ultrastructural analysis. Here, we used focused ion-beam scanning electron microscopy to mill, image, segment and reconstruct ANF terminals in mice, 1 day and 1 week after synaptopathic exposure (8-16 kHz, 98 dB SPL). At both survivals, ANF terminals were normal in number, but 62% and 53%, respectively, lacked normal synaptic specializations. Most non-synapsing fibers (57% and 48% at 1 day and 1 week) remained in contact with an IHC and contained healthy-looking organelles. ANFs showed a transient increase in mitochondrial content (51%) and efferent innervation (34%) at 1 day. Fibers maintaining synaptic connections showed hypertrophy of pre-synaptic ribbons at both 1 day and 1 week. Non-synaptic fibers were lower in mitochondrial content and typically on the modiolar side of the IHC, where ANFs with high-thresholds and low spontaneous rates are normally found. Even 1 week post-exposure, many ANF terminals remained in IHC contact despite loss of synaptic specializations, thus, regeneration efforts at early post-exposure times should concentrate on synaptogenesis rather than neurite extension.

GRF2 Is Crucial for Cone Photoreceptor Viability and Ribbon Synapse Formation in the Mouse Retina.

Using constitutive GRF1/2 knockout mice, we showed previously that GRF2 is a key regulator of nuclear migration in retinal cone photoreceptors. To evaluate the functional relevance of that cellular process for two putative targets of the GEF activity of GRF2 (RAC1 and CDC42), here we compared the structural and functional retinal phenotypes resulting from conditional targeting of RAC1 or CDC42 in the cone photoreceptors of constitutive GRF2KO and GRF2WT mice. We observed that single RAC1 disruption did not cause any obvious morphological or physiological changes in the retinas of GRF2WT mice, and did not modify either the phenotypic alterations previously described in the retinal photoreceptor layer of GRF2KO mice. In contrast, the single ablation of CDC42 in the cone photoreceptors of GRF2WT mice resulted in clear alterations of nuclear movement that, unlike those of the GRF2KO retinas, were not accompanied by electrophysiological defects or slow, progressive cone cell degeneration. On the other hand, the concomitant disruption of GRF2 and CDC42 in the cone photoreceptors resulted, somewhat surprisingly, in a normalized pattern of nuclear positioning/movement, similar to that physiologically observed in GRF2WT mice, along with worsened patterns of electrophysiological responses and faster rates of cell death/disappearance than those previously recorded in single GRF2KO cone cells. Interestingly, the increased rates of cone cell apoptosis/death observed in single GRF2KO and double-knockout GRF2KO/CDC42KO retinas correlated with the electron microscopic detection of significant ultrastructural alterations (flattening) of their retinal ribbon synapses that were not otherwise observed at all in single-knockout CDC42KO retinas. Our observations identify GRF2 and CDC42 (but not RAC1) as key regulators of retinal processes controlling cone photoreceptor nuclear positioning and survival, and support the notion of GRF2 loss-of-function mutations as potential drivers of cone retinal dystrophies.

Ultrastructural Characteristics and Synaptic Connectivity of Photoreceptors in the Simplex Retina of Little Skate (Leucoraja erinacea).

The retinas of the vast majority of vertebrate species are termed "duplex," that is, they contain both rod and cone photoreceptor neurons in different ratios. The retina of little skate (Leucoraja erinacea) is a rarity among vertebrates because it contains only a single photoreceptor cell type and is thus "simplex." This unique retina provides us with an important comparative model and an exciting opportunity to study retinal circuitry within the context of a visual system with a single photoreceptor cell type. What is perhaps even more intriguing is the fact that the Leucoraja retina is able use that single photoreceptor cell type to function under both scotopic and photopic ranges of illumination. Although some ultrastructural characteristics of skate photoreceptors have been examined previously, leading to a general description of them as "rods" largely based on outer segment (OS) morphology and rhodopsin expression, a detailed study of the fine anatomy of the entire cell and its synaptic connectivity is still lacking. To address this gap in knowledge, we performed serial block-face electron microscopy imaging and examined the structure of skate photoreceptors and their postsynaptic partners. We find that skate photoreceptors exhibit unusual ultrastructural characteristics that are either common to rods or cones in other vertebrates (e.g., outer segment architecture, synaptic ribbon number, terminal extensions), or are somewhere in between those of a typical vertebrate rod or cone (e.g., number of invaginating contacts, clustering of multiple ribbons over a single synaptic invagination). We suggest that some of the ultrastructural characteristics we observe may play a role in the ability of the skate retina to function across scotopic and photopic ranges of illumination. Our findings have the potential to reveal as yet undescribed principles of vertebrate retinal design.

Studying zebrafish nervous system structure and function in health and disease with electron microscopy.

Zebrafish (Danio rerio) is a well-established model for studying the nervous system. Findings in zebrafish often inform studies on human diseases of the nervous system and provide crucial insight into disease mechanisms. The functions of the nervous system often rely on communication between neurons. Signal transduction is achieved via release of signaling molecules in the form of neuropeptides or neurotransmitters at synapses. Snapshots of membrane dynamics of these processes are imaged by electron microscopy. Electron microscopy can reveal ultrastructure and thus synaptic processes. This is crucial both for mapping synaptic connections and for investigating synaptic functions. In addition, via volumetric electron microscopy, the overall architecture of the nervous system becomes accessible, where structure can inform function. Electron microscopy is thus of particular value for studying the nervous system. However, today a plethora of electron microscopy techniques and protocols exist. Which technique is most suitable highly depends on the research question and scope as well as on the type of tissue that is examined. This review gives an overview of the electron microcopy techniques used on the zebrafish nervous system. It aims to give researchers a guide on which techniques are suitable for their specific questions and capabilities as well as an overview of the capabilities of electron microscopy in neurobiological research in the zebrafish model.

Differential effects of aging on hippocampal ultrastructure in male vs. female rats.

Age-related decline in physical and cognitive functions are facts of life that do not affect everyone to the same extent. We had reported earlier that such cognitive decline is both sex- and context-dependent. Moreover, age-associated ultrastructural changes were observed in the hippocampus of male rats. In this study, we sought to determine potential differences in ultrastructural changes between male and female rats at various stages of life. We performed quantitative electron microscopic evaluation of hippocampal CA1 region, an area intimately involved in cognitive behavior, in both male and female adolescent, adult and old Wistar rats. Specifically, we measured the number of docking synaptic vesicles in axo-dendritic synapses, the length of active zone as well as the total number of synaptic vesicles. Distinct age- and sex-dependent effects were observed in several parameters. Thus, adult female rats had the lowest synaptic active zone compared to both adolescent and old female rats. Moreover, the same parameter was significantly lower in adult and old female rats compared to their male counterparts. On the other hand, old male rats had significantly lower number of total synaptic vesicles compared to both adolescent and adult male rats as well as compared to their female counterparts. Taken together, it may be suggested that age- and sex-dependent ultrastructural changes in the hippocampus may underlie at least some of the differences in cognitive functions among these groups.

Comprehensive analysis of sex differences in the function and ultrastructure of hippocampal presynaptic terminals.

Sex differences in the brain, encompassing variations in specific brain structures, size, cognitive function, and synaptic connections, have been identified across numerous species. While previous research has explored sex differences in postsynaptic structures, synaptic plasticity, and hippocampus-dependent functions, the hippocampal presynaptic terminals remain largely uninvestigated. The hippocampus is a critical structure responsible for multiple brain functions. This study examined presynaptic differences in cultured hippocampal neurons derived from male and female mice using a combination of biochemical assays, functional analyses measuring exocytosis and endocytosis of synaptic vesicle proteins, ultrastructural analyses via electron microscopy, and presynaptic Ca2+-specific optical probes. Our findings revealed that female neurons exhibited a higher number of synaptic vesicles at presynaptic terminals compared to male neurons. However, no significant differences were observed in presynaptic protein expression, presynaptic terminal ultrastructure, synaptic vesicle exocytosis and endocytosis, or presynaptic Ca2+ alterations between male and female neurons.

Glycogen in retinal horizontal cells of the African mud catfish Clarias gariepinus (Burchell, 1822) and its physiological significance.

This paper reports on glycogen store in the retinal horizontal cells (HC) of the African mud catfish Clarias gariepinus, as seen by histochemical reaction with periodic acid Schiff (PAS) and transmission electron microscopy in light- as well as dark-adapted state. Glycogen is abundant in the large somata and less in their axons, characterised ultrastructurally by many microtubules and extensive gap junctions interconnecting them. There was no apparent difference in glycogen content in HC somata between light- and dark adaptation, but the axons clearly showed absence of glycogen in dark condition. The HC somata (presynaptic) make synapses with dendrites in the outer plexiform layer. Müller cell inner processes, which contain more densely packed glycogen, invest the HC. Other cells of the inner nuclear layer do not show any appreciable content of glycogen. Rods, but not cones, contain abundant glycogen in their inner segments and synaptic terminals. It is likely that glycogen is used as energy substrate in hypoxia for this species that dwell muddy aquatic environment with low oxygen content. They appear to have high energy demand, and a high glycogen content in HC could act as a ready source to fulfil physiological processes, like microtubule-based transport of cargo from the large somata to axons and the maintenance of electrical activities across the gap junctions between the axonal processes. It is also likely that they can supplement glucose to the neighbouring inner nuclear layer neurons, which are clearly devoid of glycogen.

Expression of α-Synuclein in the mouse retina is confined to inhibitory presynaptic elements.

α-Synuclein (α-Syn) is enriched in presynaptic terminals of the central nervous system including the retina and plays a role in the synaptic vesicle cycle and synaptic transmission. Abnormal aggregation of α-Syn is considered to be the main component of the Lewy bodies that are the pathological hallmarks of Parkinson's disease. Although expression pattern of α-Syn has been described in the retinas, its precise cellular and subcellular locations are poorly understood. We investigated the precise expression of α-Syn using light microscopy (LM) and electron microscopy (EM) with antibodies against α-Syn in the mouse retina. We found that the majority of α-Syn immunoreactivity (IR) is located in GABAergic, glycinergic, and dopaminergic amacrine cells, and their processes often make a direct synapse to other labeled or unlabeled amacrine profiles, bipolar cell terminals, or ganglion cell dendrites. Further, our LM and immuno-EM results confirm the absence of α-Syn in excitatory photoreceptors, bipolar cell bodies, and their ribbon synapses, providing evidence, for the first time, that ribbon synapses do not express α-Syn. Additionally, α-Syn IR is located in the ganglion cells, some of which are intrinsically photosensitive retinal ganglion cells. These results reveal a previously unappreciated inhibitory synapse-specific expression pattern of α-Syn in the retina, suggesting that α-Syn may play a distinct role in the modulation and integration of inhibitory synaptic transmission in the retina.

Three-Dimensional Structure of Inner Ear Hair Cell Ribbon Synapses in a Zebrafish Model of Usher Syndrome Type 1B.

Our understanding of inner ear hair cell ultrastructure has heretofore relied upon two-dimensional imaging; however, serial block-face scanning electron microscopy (SBFSEM) changes this paradigm allowing for three-dimensional evaluation. We compared inner ear hair cells of the apical cristae in myo7aa-/- null zebrafish, a model of human Usher Syndrome type 1B, to hair cells in wild-type zebrafish by SBFSEM to investigate possible ribbon synapse ultrastructural differences. Previously, it has been shown that compared to wild type, myo7aa-/- zebrafish neuromast hair cells have fewer ribbon synapses yet similar ribbon areas. We expect the recapitulation of these results within the inner ear apical crista hair cells furthering the knowledge of three-dimensional ribbon synapse structure while resolving the feasibility of therapeutically targeting myo7aa-/- mutant ribbons. In this report, we evaluated ribbon synapse number, volume, surface area, and sphericity. Localization of ribbons and their distance from the nearest innervation were also evaluated. We determined that myo7aa-/- mutant ribbon synapses are smaller in volume and surface area; however, all other measurements were not significantly different from wild-type zebrafish. Because the ribbon synapses are nearly indistinguishable between the myo7aa-/- mutant and wild type, it suggests that the ribbons are structurally receptive, supporting that therapeutic intervention may be feasible.

Rod bipolar cells receive cone photoreceptor inputs through both invaginating synapses and flat contacts in the mouse and guinea pig retinas.

The light pathways are segregated into rod and cone pathways in which rods synapse with rod bipolar cells (RBCs), while cones contact cone bipolar cells (CBCs). However, previous studies found that cones can make synapse with RBCs (cone-RBC synapses) and rods can contact OFF CBC in primate and rabbit retinas. Recently, such cone-RBC synapses have been reported physiologically and morphologically in the mouse retina. Nevertheless, the precise subcellular evidence to determine whether it is the invaginating synapse or the flat contact remains absent. This is due to a lack of immunochemically verified ultrastructural data. Here, we investigated the precise expression of protein kinase C alpha (PKCα) using pre-embedding immunoelectron microscopy (immuno-EM) with a monoclonal antibody against PKCα, a biomarker for the RBCs. We determined the nanoscale localization of PKCα in the outer plexiform layer of the mouse and guinea pig retinas. Our results demonstrate the existence of both the direct invaginating synapse and the basal/flat contact of the cone-RBCs, providing for the first time immunochemically verified ultrastructural evidence for the cone-RBC synapse in the mouse and guinea pig retinas. These results suggest that the cross talk between cone and rod pathways is much more extensive than previously assumed.

The connectome of an insect brain.

Brains contain networks of interconnected neurons and so knowing the network architecture is essential for understanding brain function. We therefore mapped the synaptic-resolution connectome of an entire insect brain (Drosophila larva) with rich behavior, including learning, value computation, and action selection, comprising 3016 neurons and 548,000 synapses. We characterized neuron types, hubs, feedforward and feedback pathways, as well as cross-hemisphere and brain-nerve cord interactions. We found pervasive multisensory and interhemispheric integration, highly recurrent architecture, abundant feedback from descending neurons, and multiple novel circuit motifs. The brain's most recurrent circuits comprised the input and output neurons of the learning center. Some structural features, including multilayer shortcuts and nested recurrent loops, resembled state-of-the-art deep learning architectures. The identified brain architecture provides a basis for future experimental and theoretical studies of neural circuits.