Dense 4D nanoscale reconstruction of living brain tissue

Abstract

Three-dimensional (3D) reconstruction of living brain tissue down to an individual synapse level would create opportunities for decoding the dynamics and structure–function relationships of the brain’s complex and dense information processing network; however, this has been hindered by insufficient 3D resolution, inadequate signal-to-noise ratio and prohibitive light burden in optical imaging, whereas electron microscopy is inherently static. Here we solved these challenges by developing an integrated optical/machine-learning technology, LIONESS (live information-optimized nanoscopy enabling saturated segmentation). This leverages optical modifications to stimulated emission depletion microscopy in comprehensively, extracellularly labeled tissue and previous information on sample structure via machine learning to simultaneously achieve isotropic super-resolution, high signal-to-noise ratio and compatibility with living tissue. This allows dense deep-learning-based instance segmentation and 3D reconstruction at a synapse level, incorporating molecular, activity and morphodynamic information. LIONESS opens up avenues for studying the dynamic functional (nano-)architecture of living brain tissue.

Light-microscopy based dense connectomic reconstruction of mammalian brain tissue

“…LICONN has been used to acquire volumes up to roughly one million cubic microns in size ( Fig. 1A ), which is sufficiently large for biological analysis of mammalian neural circuits ( 1 , 10 , 66 , 67 ) and achieved reasonable acquisition rates (6.5h acquisition time for 0.39 Teravoxel, effective voxel rate of 17 MHz including overhead due to tile overlap and sample stage positioning) even without dedicated optimizations for imaging speed ( 25 , 68 , 69 ). However, the largest EM datasets span a cubic millimeter, and a major long-term ambition in the field is mapping the hundreds of cubic millimeters encompassing an entire mouse brain ( 70 ).…”

Fluorescence microscopy shadow imaging for neuroscience

“…Organoids offer exciting opportunities for delving into human organ development, and brain organoids hold great promise for investigating the complexities of human brain development in health and disease ( Kim et al, 2020 ; Eichmüller and Knoblich, 2022 ). SUSHI has been applied to reconstruct the volume of a living human brain organoid grown in vitro ( Figure 3E ), demonstrating the potential of this technique ( Velicky et al, 2023 ).…”

“…The tradeoff for the speed of the light-sheet approach is a more complex dual microscope objective setup and slightly lower contrast images than confocal microscopy-based shadow imaging, though the light-sheet images still reveal a wealth of neuropil and ECS structure. Another recent report combined SUSHI of neurons with confocal microscopy of genetically expressed synaptic proteins, including synaptophysin-1 and PSD-95 ( Velicky et al, 2023 ).…”

“…(E) Live human cerebral organoid imaged by SUSHI, including the off-line work flow used for image processing and tissue reconstruction. From Velicky et al (2023) , with permission.…”

“…This approach allows analysis of individual cells within their parenchymal context, and we previously used this to stoichiometrically analyze unlabeled presynaptic boutons with respect to postsynaptic dendritic spines on positively YFP-labeled neurons ( Tønnesen et al, 2018 ), which would not be feasible using positive labeling ( Figures 4A–C ). Further, as mentioned above, SUSHI and confocal shadow imaging have been combined in a sample with genetic expression of synaptic markers to allow two- or three-color combined STED and confocal imaging to reveal the neuropil context around the synaptic markers ( Velicky et al, 2023 ). Two-color shadow imaging was recently used to visualize GFP-labeled tumor cells and unlabeled cortical neurons by in vivo two-photon microscopy ( Dembitskaya et al, 2023 ), providing proof of concept for visualizing the invasion and displacement of healthy parenchyma by cancerous cells ( Figure 4D ).…”