Imaging Applications in Neuroscience Summit

Generating adaptive behavioral responses to emotionally salient stimuli requires evaluation of complex associations between multiple sensations, the surrounding context, and current internal state. Neural circuits within the amygdala parse this emotional information, undergo synaptic plasticity to reflect learned associations, and evoke appropriate responses through their projections to the brain regions orchestrating these behaviors. Information flow within the amygdala is regulated by the intercalated cells (ITCs), which are densely packed clusters of GABAergic neurons that encircle the basolateral amygdala (BLA) and provide contextually relevant feedforward inhibition of amygdala nuclei. I will discuss the unique contribution of each ITC cluster and explore how different neuromodulators conveying internal state act via ITC gates to shape emotionally motivated behavior. I will focus on the confocal studies unveiling the role of a previously uncharacterized ITC cluster, the apical ITCs, as an integrator of thalamic and cortical information.

Functional calcium imaging in the brain using GCaMP allows us to observe the brain in action, linking cellular firing patterns to observable behavior. But without any further information aside from the calcium activity, we are left with few details about the neurons we have observed, lacking knowledge on the specific cell types or the brain regions with which they communicate. By utilizing the full range of spectral capabilities of a confocal microscope, we can increase the number of observable secondary markers by an order of magnitude. Spectral lambda imaging allows the discrimination of fluorophores with sufficiently different emission spectra. Combining this with the full range of available excitation lasers, we can further differentiate between fluorophores with similar emission spectra, but distinct excitation wavelengths. With this multiplexed lambda imaging approach, we were able to image nine additional colors together with GCaMP during in-vivo imaging. The additional markers were introduced using retrograde viruses injected into different regions of the brain, highlighting which of the active neurons was connected by long-range projections to those specific regions. This connectivity information allows the nuanced examination of circuit dynamics in cellular firing patterns observed during behavior. This technique allows us to not only overcome the optical limitations on spatial resolution and volumetric imaging, but to further allow the simultaneous recording of multiple neuronal subtypes at once, enabling the direct comparison of activity patterns during behavior.

General anesthetics are potent regulators of pain, but how they affect specific pain features, such as aversion, is not well understood. In this study, we investigated the effect of isoflurane, a general anesthetic, on the anterior cingulate cortex (ACC), a brain region implicated in generating pain's negative effect. Using in vivo imaging of neuronal calcium dynamics in mice, we found that spontaneous ACC activity persists at concentrations of isoflurane that induce loss of behavioral responsiveness (i.e., unconsciousness), and is only silenced at higher concentrations that induce deeper levels of anesthesia. We suggest that this persistent ACC activity underlies the phenomena of connected consciousness during general anesthesia, where otherwise anesthetized patients regain awareness intraoperatively and are capable of experiencing pain. As such, blocking the pain experience during general anesthesia likely occurs after isoflurane-induced silencing of the ACC, which we propose produces a similar functional outcome to ablating the ACC, a treatment known to provide pain relief.

In addition, we detail effective strategies for using the ZEISS LSM980 with Airyscan 2 to perform in vivo calcium imaging in awake and anesthetized mice. We also demonstrate how coupling Airyscan live imaging with chronically implanted gradient index (GRIN) lenses allows for identifying neural activity within distinct subpopulations of neurons. Lastly, we discuss methods for processing calcium imaging data to extract neural activity at the single neuron level.

Evaluation of in vivo distribution of biologics is often a critical step in drug development. Historically, tracking pharmacokinetics with radionuclides has been the most commonly used method because of its high sensitivity. Gamma emitting radionuclides also circumvent common issues with detection that can arise from tissue matrix interference or poor recovery when compared to ELISA and mass spectrometry based detection. While radiometric distribution will continue to be a primary mode of evaluation, it never approaches the resolution offered by microscopy. We have coupled higher resolution fluorescence and electron microscopy with radiometric distribution to answer key developmental questions for retinal and neurodegenerative diseases.

The human brain has many structural and functional features that are distinct from lower species traditionally used for medical research. To identify mechanisms of human brain development and find cures for human-specific neurological disorders such as autism, we ideally need a human brain model. However, experimentation with human brain tissue, particularly at fetal stages, is inherently challenging, curtailing long-term gene manipulation studies and environmental perturbations. Cerebral organoids generated from human pluripotent stem cells (hPSC) are thus emerging as a promising alternative system for studying human neocortical development and disease. We established reproducible and efficient methods for cortical organoid differentiation that faithfully recapitulate in vivo neocortical development. Neurons within the organoids are functional and exhibit network-like activities. We further demonstrate the utility of the organoid system for modeling neurodevelopmental disorders. Finally, we started employing tissue engineering technologies to improve the established telencephalic organoids. Together, these findings provide an essential foundation for the utilization human brain organoids to study human neural development and disease.

The presentation will review key findings regarding synaptic health in the context of cognitive aging, as well as recent efforts to develop rhesus monkey models of Alzheimer’s Disease (AD). We have identified multiple synaptic attributes in prefrontal cortex (PFC) of the rhesus monkey that are highly vulnerable to aging and their loss is strongly correlated with cognitive decline. Many of these attributes are protected by cyclical estradiol treatment (E2T) that also protects against age-related cognitive decline. We have also revealed several reflections of synaptic aging in rhesus monkey hippocampus that lead to age-related cognitive decline, and interestingly, the vulnerable synapses in hippocampus bear little resemblance to the patterns of vulnerability in PFC, demonstrating independent region-specific mechanisms of synaptic aging. Through these synaptic analyses we are developing a molecular and structural profile of Synaptic Health. Recently, we have developed and characterized both Ab oligomer (AβO) and tau based rhesus monkey models of AD. The AβO based model involves the chronic injection of AβOs into the lateral ventricle, and we view this model as representing the synaptic phase of AD, given the well-known synaptotoxic effects of AβOs. This model reveals AβO-induced synapse loss similar to that seen in aged monkeys with cognitive decline. The tau-based model is directed at the degenerative phase of AD and requires the injection of a double human mutant tau construct (AAV1-(P301L/S320F) directly into the entorhinal cortex (ERC). The exogenous human mutant tau coaptates with endogenous monkey tau, leading to extensive tau-based pathology and neurodegeneration across the ERC connectome. Both models induce an extensive neuroinflammatory response. Future experiments will reveal both behavioral and in vivo imaging consequences of the cellular and synaptic pathology reflective of the synaptic vs degenerative phases of AD.

The study of the dendritic and axonal structure of human neurons is an important area of research in neuroscience. Animal models have been traditionally used to investigate cellular and network properties in the central nervous system, but the translation of these results to humans is still being determined. To overcome this problem, ex vivo brain slices derived from neurosurgical-resected human tissue are being used to investigate the fundamental properties of human neurons. This technique enables the analysis of the morphology of neurons in correlation with their functional properties. Biocytin-filled neurons can be imaged using confocal microscopy to study the dendritic structure and spine morphology.

Human organotypic slice cultures further expand the possibilities for studying human neurons, including the ability to use AAV-mediated labeling and functional imaging of cells in the tissue. This technique allows live cell imaging experiments using GFP-labeled neurons can provide insights into the dynamic changes in dendritic and axonal structure over time. To increase the output and efficiency of studying human neurons, AI models are being trained on human datasets to automate spine detection. This will allow faster and more accurate analysis of large datasets, crucial for understanding complex neurological disorders such as neurodegeneration, tumor infiltration, or epileptogenesis. In conclusion, studying the dendritic and axonal structure of human neurons using ex vivo brain slices, live cell imaging, and AI models will continue to enhance our understanding of the human brain and its role in neurological diseases.

Every tissue clearing experiment has unique sample and labelling parameters. Because tissue type, thickness, pigmentation, molecular targets of interest and labelling modalities are different for every researcher, it’s naive to assume a previously published method will work reproducibly for everyone. Here, we describe how a modular approach to tissue clearing, in conjunction with quantitative structure-activity relationship (QSAR) modeling, can create customized protocols tailored to individual experiments. As a proof of principal, we use QSAR modelling in mouse tissue to optimize clearing and labelling conditions for multiple antibodies in cleared human brain tissue.

The combination of optical tissue clearing with light-sheet fluorescence imaging provides a powerful approach for understanding the brain basis of behavior and neurological disease. In this talk, I will describe tools we have developed to investigate recent neuronal and microglial activity in the mouse brain. Immediate early gene (IEG) products like Npas4 and cFos provide sensitive, cellular resolution snapshots of recent neuronal activity. When coupled with atlases suchl as the CCFv3, the number of active neurons can be quantified across hundreds of distinct regions in the mouse brain. Obtaining differential signals from optically cleared, intact brains poses unique challenges. Artifacts such as label diffusion and alignment accuracy can introduce significant error. Furthermore, the hierarchical structure of atlas regions imposes strong interdependence among regional signals. We have developed a statistical framework designed specifically to tackle these unique challenges. We apply this framework to identify both expected and surprising regional signatures of cFos and Npas4 activated when dark adapted mice are exposed to light. We have also applied these tools to quantify microglia and β-amyloid plaques in a mouse model of Alzheimer's Disease. We have developed two machine learning-enabled workflows, one that counts the microglia while the other measures microglial shape. We have also validated methods for labeling β-amyloid in cleared brains. Our automated methods can identify and count plaques and plaque-associated microglia throughout the brain. Together, we have combined these tools into BrainQuant3D, a scalable platform for whole-brain image segmentation and analysis that transforms differential signals into publication-ready figures.

Recent advances in tissue clearing and light sheet imaging have opened an exciting new avenue for brain-wide, cellular resolution immunostaining. At Translucence Biosystems, we have developed a services pipeline for clearing and staining, light sheet imaging, and cellular resolution machine learning-enabled quantification, providing access to the intricate anatomy of intact tissues. With help from BRAIN Initiative funding, we are developing hardware, software, and reagent kits to offer neuroscientists the tools they need for unbiased and complete views of brain anatomy and function. In this talk I will give an overview of our pipeline and highlight our Neuronal Activity Tissue Clearing Kit, Mesoscale Imaging System, and Stitchy software. Our tissue clearing kit is based on our modified iDISCO+ protocol and includes validated antibodies against Npas4 and cFos for brain-wide marking of recently active neurons. The Mesoscale Imaging System adapts the ZEISS Lightsheet Z.1 for imaging large intact tissues in high refractive index solutions. It is also sold by Zeiss as an add on to the new Lightsheet 7 to enable more rapid acquisition using a 2.5x objective. Finally, we developed our new Stitchy software to easily handle terabyte-scale image files and streamline our internal light sheet imaging and quantification pipeline. The recently released user-friendly commercial version of Stitchy reads native light sheet microscope files and produces stitched images in your desired output file format (e.g. Imaris, ome.tiff, NGFF). We created these tools for researchers as part of our mission to enable the dimensional shift from 2D to 3D histology.

Labeling neuronal samples with antibodies followed by diverse modes of imaging, plays a prominent role across neuroscience research. Such an approach can reveal the temporal and spatial localization of proteins critical for function in development, aging, injury, and disease. The value of the data arising from such approaches can be impacted by numerous experimental parameters including sample preparation, the efficacy and specificity of antibody labeling, and the quality of the imaging platform. I will discuss our efforts to systematically address these issues, with a particular focus on antibody development. I will emphasize our work to develop antibodies engineered to facilitate simultaneous multiplex labelling, expanding the utility of each precious sample with a greater ability to define cell types and spatial relationships.

While international brain-mapping initiatives remain focused on revealing the structure and working of the healthy brain, the need to map the unhealthy brain is compelling and urgent, yet unmet. We envision a set of digital maps that are accompanied by visual “search engines” that allow an exploratory analysis of the brain cytoarchitecture. There is an equally compelling need to map the complex and multi-scale brain cellular alterations associated with pathological conditions like traumatic brain injury, ischemia, binge alcohol, tumor growth, and experimental drug treatments. These alterations range from individual cells to multi-cellular functional units at multiple scales ranging from niches to the layered brain cytoarchitecture. These alterations represent a mixture of changes associated with the primary injury, secondary injuries, regenerative processes, inflammation, tissue remodeling, drug treatments, and drug side effects. Many of these alterations can be subtle and/or latent, only discernible by sensing changes in cell morphology, or the expression patterns of molecular markers. Unfortunately, immunohistochemical (IHC) methods reveal only a fraction of these alterations at a time, miss the many other alterations and side effects that are occurring concurrently, and do not provide quantitative readouts.
 
In this talk, we will discuss a practical approach to pathological brain tissue mapping with a focus on rational therapeutics development. The idea is to replace the many low information content assays with a single comprehensive assay based on imaging and analyzing highly multiplexed whole brain sections employing 10 – 100 molecular markers, sufficient to analyze all the major brain cell types and their functional states over large brain regions. We describe a combination of signal reconstruction, neural-network-based cell detection and phenotyping, and high-dimensional data analysis approaches to generate quantitative readouts of cellular alterations at multiple scales ranging from individual cells to multi-cellular units, large cellular ensembles (e.g., cortical layers), and atlas-mapped brain regions for comparative analysis. These data can be used to test hypotheses, screen individual drugs, and combination therapies, and initiate system-level studies.

Correlative light and electron microscopy (CLEM) is an effective tool to investigate the relationship between structure and function in the brain. We can apply advanced light microscopy (LM) techniques to record neuronal activity, then perform volume electron microscopy (EM) techniques to visualize the structure of synapses at high resolution. We established two CLEM workflows using different volume EM applications: serial block-face scanning EM (SBF-SEM) and serial section SEM array tomography (ssSEM). Using the first approach, we captured functional properties of neurons and their dendritic spines with 2-photon Ca2+ imaging in vivo, and targeted the same neuron, dendrites, and spines with SBF-SEM using inherent structural markers. This method allowed us to capture a relatively large volume of EM data with easy alignment, and we analyzed the morphological details of synapses and their surroundings. Volume EM data is extremely information-rich, but extracting features from images with high throughput remains a challenge. Examples of how we are working to overcome this challenge using deep learning will be discussed. A second approach involving CLEM can be applied when immuno-EM labeling methods for identification of specific cells or organelles is necessary. We induced structural plasticity in spines using 2-photon glutamate uncaging, applied an immuno-EM protocol to label particular neurons, and performed ssSEM to identify the exact spine stimulated by uncaging among other unstimulated spines. In this workflow, we can capture a region of interest multiple times at various magnifications. We found that the peri-synaptic membrane, directly adjacent to the synapse, expanded at an early time point of induced plasticity. We extended the workflow by applying a rapid cryo-fixation and ZEISS cryo-Airyscan imaging to capture cellular events at a high temporal resolution. There are several ways to optimize and apply a particular CLEM method for any given project, and several tips will be discussed.

Abstract coming soon.

Mapping neuronal networks underlying behavior has recently become a central focus in neuroscience. Serial section electron microscopy (ssEM) is one approach that can reveal the fine structure of neuronal networks (“connectomics”). However, this structural data doesn’t provide important molecular information which helps identify cell types or certain functional properties of brain cells. Volumetric correlated light and electron microscopy (vCLEM) combines ssEM and volumetric fluorescence microscopy to incorporate molecular information into ssEM datasets. We developed an approach using small fluorescent single-chain variable fragment (scFv) immuno-probes to do multiplexed detergent-free immuno-labeling and serial electron microscopy on the same samples. We generated eight such fluorescent scFvs that targeted useful markers for brain studies (GFP, GFAP, Calbindin, Parvalbumin, Kv 1.2, VGluT1, PSD-95, and NPY). To test the vCLEM approach, six different fluorescent probes were imaged in a sample of the cortex of a cerebellar lobule (Crus 1) using confocal microscopy with spectral unmixing and followed by ssEM imaging of the same sample. The results show both excellent ultrastructure and superimposition of many fluorescence channels. This work revealed a poorly described cell type in the cerebellum, two types of mossy fiber terminals, and the subcellular localization of ion channels. Because scFvs derive from the plentiful number of existing monoclonal antibodies, hundreds of such probes can be generated which may be useful for a range of connectomic studies.

Sexual dimorphism in brain activity has been observed in many species, including the nematode C. elegans. In C. elegans, males and hermaphrodites exhibit differences in their responses to environmental cues such as food and pheromones. These differences are believed to be mediated by structural and functional differences in the nervous system. The connectome of C. elegans has revealed substantial differences in neural wiring patterns between males and hermaphrodites. However, the extent to which functional connectivity and neuronal activity contribute to sexually dimorphic behavior is still not fully understood. Calcium imaging techniques offer a promising means to address this question by enabling simultaneous measurement of neuronal activity in large populations of neurons in vivo. In this study, we developed a novel system that records whole-brain neuronal activity in both male and hermaphrodite C. elegans while presenting them with a diverse set of external sensory cues. Our system uses a modified confocal microscope with single-cell resolution that can image the worm’s nervous system at up to 10 volumes/second. We developed microfluidic devices to accommodate both sexes, these devices enable us to simultaneously record the neuronal activity of almost all neurons in the nervous system, and to sequentially present 10 stimuli to the animal's nose. To assess the variability in state-dependent responses to repeated exposure of the same stimulus, we repeated the sequence 3 times and observed the animal's response for over 30 minutes. In our study, we investigated concentration-dependent attractive and repulsive responses to a diverse set of stimuli, including gustatory, olfactory, and nociceptive cues. Using our system, we discovered a substantial set of previously unknown dimorphisms. Our approach enabled us to evaluate stimulus-evoked responses across sensory, inter-, and motor neurons, revealing novel sexually dimorphic responses exhibiting significant differences in the shape, amplitude, and kinetics of male versus hermaphrodite neural activity. Nevertheless, on a global scale we found pairwise correlations between individual neurons to be largely similar between sexes, indicating the potential for a large degree of conserved functional connectivity.