Mouse tripartite synapse imaged with ZEISS Volutume | University of Turin

User Story

Exploring Astrocyte Morphology: How Volume Electron Microscopy is Advancing Brain Research

byZEISS Microscopy 17 June 2025 5 min read

Introduction

Exploring Astrocyte Morphology: How Volume Electron Microscopy is Advancing Brain Research

Astrocytes, star-shaped glial cells, are the most abundant glial cell type in the central nervous system (CNS). They play a crucial role in maintaining brain homeostasis, regulating synaptic activity, supporting neuronal metabolism, and responding to injury. Their diverse roles highlight their importance in both healthy brain function and various neurological disorders.

Astrocytes have branches extending from their soma, which divide into smaller structures called branchlets, lamelliform perisynaptic astrocytic processes (PAPs), and endfeet. Due to the nanometric size of these structures, electron microscopy (EM) is the only technique capable of fully imaging astrocytes, revealing their spatial distribution, heterogeneity, and synaptic coverage. Understanding astrocyte ultrastructure is vital for advancing knowledge of the CNS in health and disease.

Vanessa Chappiani & Prof. Corrado Calì | University of Turin — Vanessa Chiappini and Prof. Corrado Calì from the University of Turin

Corrado Calì, an Associate Professor of Human Anatomy at the University of Turin, has made significant contributions to this field. Calì and his team have focused on investigating the neuroanatomy of the brain at the nanoscopic level. Using 3D electron microscopy (3DEM, volume EM) and 3D reconstruction, they emphasize astrocytes and their morphological coupling with neurons and vasculature.

In addition to his academic roles, Calì is the President and Founder of IntraVides, a company developing smart assistant surgeon solutions through augmented reality. In his former team, he pioneered the use of Virtual Reality in neuroscience, developing tools to explore the intricate relationships between these cell types.

Studying Astrocyte Ultrastructure over Time

The recently published research delves into the scientific literature on the three-dimensional ultrastructure of astrocytes in the CNS, spanning from the 1960s to the present. While volume electron microscopy (vEM) gained momentum in this area in the 2000s – due to advancements in automated sectioning and backscattered electron sensors – serial section EM and 3D reconstructions were already present in the 1950s. The study provides a comprehensive overview of astrocyte ultrastructure across brain regions, species, and imaging techniques – bridging knowledge gaps and supporting researchers in structural biology and computational neuroscience.

Read the paper: Ultrastructure of astrocytes using volume electron microscopy: A scoping review

Mouse cortex imaged with ZEISS Volutome | University of Turin — Electron micrograph of a mouse brain cortex acquired with ZEISS Volutome.
Imaging parameters: 2 kV, 70 pA, dwell time 0.8 µs, pixel size: 6 nm, cutting thickness: 50 nm.

Advantages of Volume Electron Microscopy (vEM)

vEM surpasses traditional light microscopy in studying astrocyte morphology. Glial fibrillary acidic protein immunostaining, for instance, highlights only 15% of the astrocytic volume, failing to capture the entire astrocytic arborization. Additionally, finer lamelliform processes, which interface with synapses, have cross-sections as small as tens of nanometers – beyond the resolution of fluorescence microscopy.

Electron microscopes achieve nanometric resolution, enabling the detailed visualization of astrocytic processes. Scientists can study and image these processes in unprecedented detail by integrating technologies such as a focused ion beam or an ultramicrotome into a scanning electron microscope.

Corrado Calì University of Turin

Insights into the Complex Architecture of the Brain

Imaged with ZEISS Volutome

3D reconstruction from mouse hippocampus showing a neuron (blue), astrocytic process (light blue), and blood vessel (red) superimposed with the original electron micrograph.

3D reconstruction of a tripartite synapse: presynaptic bouton (red), dendritic spine (green), neuron (blue) and astrocytic process (light blue, transparent).

Computational and Imaging Innovations

Deep learning algorithms for segmentation and analysis of astrocytic networks show promise. Emerging tools like the Segment Anything Model can segment complex morphologies with minimal manual input, addressing structural variability and reducing human intervention. Further refinement is needed to reduce error rates and handle the complexity of astrocytic processes. Innovations in data management, such as cloud-based storage and precomputed multi-scale data pyramids, could also handle the large datasets generated by vEM studies.

Addressing Challenges in vEM Sample Preparation

The most common vEM sample preparation method is chemical fixation, which can cause extracellular shrinkage and astrocyte swelling. Although high-pressure freezing enables better tissue preservation, it struggles to vitrify samples that are hundreds of micrometers thick. These larger samples are needed to study human astrocytes in a comprehensive view. Refining high-pressure freezing protocols could improve astrocyte ultrastructure preservation and subsequent analysis.

Advancements in volume correlative light-electron microscopy could combine the molecular specificity of light microscopy with the spatial resolution of electron microscopy, enabling the identification of functional molecules within astrocytic processes.

Corrado Calì University of Turin

Astrocytes are crucial for maintaining brain homeostasis, regulating synaptic activity, supporting neuronal metabolism, and responding to injury. Their diverse functions highlight their importance in both healthy brain function and various neurological disorders.
Volume electron microscopy (vEM) provides nanometric resolution, allowing for detailed visualization of astrocytic processes that traditional light microscopy cannot achieve. For example, glial fibrillary acidic protein immunostaining only captures about 15% of the astrocytic volume, while vEM can image the entire astrocytic arborization, including finer lamelliform processes that are critical for synaptic interactions.
Deep Learning Algorithms: Emerging deep learning tools are being utilized for the segmentation and analysis of astrocytic networks. These algorithms can efficiently segment complex morphologies with minimal manual input, addressing the structural variability of astrocytes and reducing the need for human intervention. This automation enhances the accuracy and speed of data analysis, allowing researchers to focus on interpreting results and advancing their understanding of astrocyte functions.
Cloud-Based Data Management: Innovations in data management, including cloud-based storage and precomputed multi-scale data pyramids, are being developed to handle the large datasets generated by vEM studies. These advancements facilitate better organization, accessibility, and analysis of data for researchers. Cloud-based solutions enable the efficient storage and retrieval of vast amounts of imaging information, while multi-scale data pyramids allow for scalable and detailed examination of astrocyte structures. Together, these technologies streamline the workflow and enhance the overall efficiency of vEM studies.