Light-microscopy based dense connectomic reconstruction of mammalian brain tissue

Abstract

The information-processing capability of the brain’s cellular network depends on the physical wiring pattern between neurons and their molecular and functional characteristics. Mapping neurons and resolving their individual synaptic connections can be achieved by volumetric imaging at nanoscale resolution with dense cellular labeling. Light microscopy is uniquely positioned to visualize specific molecules but dense, synapse-level circuit reconstruction by light microscopy has been out of reach due to limitations in resolution, contrast, and volumetric imaging capability. Here we developed light-microscopy based connectomics (LICONN). We integrated specifically engineered hydrogel embedding and expansion with comprehensive deep-learning based segmentation and analysis of connectivity, thus directly incorporating molecular information in synapse-level brain tissue reconstructions. LICONN will allow synapse-level brain tissue phenotyping in biological experiments in a readily adoptable manner.One-Sentence SummaryHydrogel expansion enables molecularly informed reconstruction of brain tissue at synaptic resolution with light microscopy.

Human hippocampal CA3 uses specific functional connectivity rules for efficient associative memory

“…6C ). To confirm that thorny dendrites were indeed the target of mossy inputs in human tissue, we applied LICONN (Tavakoli et al, 2024), an iterative hydrogel-expansion technology, enabling synapse-level tissue reconstruction. An effective spatial resolution of <20 nm laterally (∼16-fold increased over confocal imaging) with high-fidelity tissue preservation and comprehensive structural labeling (pan-protein), allowed us to reconstruct the mossy fiber connectome of PN dendrites, which displayed extensive large boutons contacting thorny excrescences in both mouse and human ( Fig.…”

“…Mossy fiber connections in human and mouse samples were reconstructed using high-resolution expansion with LICONN technology (Tavakoli et al, 2024), with slight adaptations for the human specimens as detailed below, and manual cell shape annotation.…”

“…Human acute tissue slices (350 µm) were fixed immediately after slicing in RT fixative solution before quenching in 100 mM glycine-containing PBS for 30 min. Further handling was according to (Tavakoli et al, 2024) with adaptations for human tissue. Images were acquired using an ANDOR Dragonfly spinning disk microscope with a 40x objective lens as described above.…”

“…Slices were expanded by incubation in MilliQ water before imaging on an ANDOR Dragonfly spinning disk microscope as above. Expansion factors were calculated by comparison of pre and post expansion images and were measured to be 3.9 ± 0.1, 3.9 ± 0.1, and 3.8 ± 0.1 for mouse, rat and human respectively (custom code, (Tavakoli et al, 2024)). Spines were either detected using a machine learning algorithm in ilastik (Berg et al, 2019) and manually corrected using Imaris 9 (Bitplane), or manually segmented using Neutube.…”