Volume Electron Microscopy of Cortical Organoids: Methods for Region Identification, Connectome Reconstruction, and Organelle Segmentation

Abstract

Abstract Volume electron microscopy (vEM) has become a powerful tool for 3D ultrastructural analysis of neural circuits, yet its application to human brain organoids remains limited, particularly for connectomic studies. Here, we established a comprehensive and scalable workflow for applying vEM to human cortical organoids, integrating correlative light and electron microscopy, large-area SEM (scanning electron microscopy) mosaic imaging, focused ion beam–SEM (FIB-SEM), and transmission electron microscopy (TEM) validation. By systematically comparing two embedding protocols in use, we demonstrated that the DeFelipe and Fairén (1993); Cano-Astorga et al. (2024a) method provides optimal compatibility with toluidine blue–stained semithin sectioning and enables reliable synapse segmentation and neurite tracing. In contrast, the Deerinck’s protocol (2010) offers enhanced membrane contrast but limits postsynaptic density visualization. Using FIB-SEM imaging of peripheral, neuropil-like regions of cortical organoids, we achieved accurate 3D reconstruction of synapses, neurites and intracellular organelles, enabling quantitative assessment of synaptic apposition surfaces, neurite trajectories, and organelle distribution across defined cellular compartments. Together, our results demonstrate for the first time the feasibility of micro-connectomic reconstruction in human cortical organoids at nanometer resolution. This methodological framework expands the applicability of vEM to organoid systems and provides a robust foundation for future studies of human brain development, disease modeling, and therapeutic evaluation at the synaptic and subcellular level. Graphical Abstract Workflow for vEM–based micro-connectomic analysis in human cortical organoids. Cortical organoids underwent sample preparation, resin embedding, and semithin/ultrathin sectioning. Large-area SEM mosaic imaging was used to identify regions of interest for targeted FIB-SEM acquisition. Ultrastructural observations were independently validated by TEM, establishing a scalable pipeline for connectome reconstruction and subcellular analysis in human brain organoids. This image was created with BioRender