Stellenbosch University
Welcome to Stellenbosch University
Insights achieved with volumetric electron microscopy
Author: Dr Jurgen Kriel
Published: 25/08/2022

​Electron microscopy (EM) has long been an essential tool for researchers to assess the ultrastructural detail of cells and tissues at nanoscale resolution. Although impactful, con­ven­tional EM approaches only relay information in a two-dimensional space, which limits the ability to assess the true structural organisation of biological material. From the way that connections amongst individual neurons are formed to how bacteria infiltrate plant roots, certain insights can only be gained once material is assessed in a three-dimensional space. Serial block-face (SBF) scanning electron microscopy (SEM) is a powerful volumetric EM modality that allows for such three-dimensional acquisition of information and was made available to researchers in Africa for the very first time at the end of 2020 by CAF.


The ThermoFisher Apreo Volumescope is one of the most recent and advanced volumetric SEMs available today. Housed at the Tygerberg Medical Campus, the Apreo Volumescope is the only such microscope on the African continent and opens the door to many new avenues of research in addition to the enhancement of current ongoing research. 

Here, we provide the progress made in volumetric EM within the South African research landscape to showcase the novel insights that can be achieved through SBF-SEM.

In stark contrast to the brain tissue for which SBF-SEM was originally designed, the first South African study to make use of the Apreo Volumescope focused on characterising melanin-containing cells in lizard skin in order to understand how they are capable of changing colour under different weather conditions. The question posed by Dr Fredericco Masetti and Prof Susana Clusella-Trullas was whether the pigment-containing structures, referred to as melanophores, were shared amongst individual skin cells or whether these protrusions remained contained within the boundaries of single cells. The volumetric data revealed that the black pigmented structures (Figure 22) remained contained within the boundaries of distinct areas within the tissue sample, with melanophores protruding ‘in between’ other cells rather than being shared amongst them (displayed as purple rendered structures in Figure 22). 
Moving even further across disciplines, the Volumescope has also been utilised to understand pathogen interaction with plant material. As part of PhD candidate Alno Carstens’ research (under supervision of Prof Lydia Joubert and Prof Gideon Wolfaardt), initial SEM analysis revealed bacterial association on the outer membrane of certain plant roots (Figure 23 A). However, it is not known whether these bacteria are able to penetrate the plant cell wall and if so, to what extent. SBF-SEM allowed for the visualisation of both plant material and bacterial association. Of note is that a single cross-section through the sample (Figure 23 B) would not accurately relay the extent of bacterial association. However, after subsequent image analysis and size exclusion-based thresholding (Figure 23 C, D), the spread of bacterial association was made visible, with only a small number of bacteria able to enter the plant. 

Although the main aim of SBF-SEM is to render out structures of interest, it can also be used as an explorative tool. Three-dimensional cell culture models are greatly beneficial to cancer researchers as they mimic tumour growth in the human micro-environment more closely than single cells in a petri dish. However, a limiting factor in the ultrastructural analysis of cancer spheroids is being able to identify regions at different depths and to make accurate comparisons between the outer proliferating layer and the inner necrotic compartment. As part of MSc student Kim Frederick’s thesis (under supervision of Prof Ben Loos), regional comparisons in the cytosolic ultrastructure could be made in brain cancer spheroids (Figure 24 A) at both the inner (Figure 24 B) and outer (Figure 24 C) layer in consecutive sections. This allowed for the identification and comparison in the volume of necrotic/dead regions between control and treated samples (Figure 24 D) to study the potential therapeutic outcomes of novel compounds. 

Connectomics is the field of study dedicated to mapping neuronal circuitry in order to identify specific pathways associated with behaviour or to establish how neuronal interactions become damaged under pathological conditions. Although an area of great advancement in the rest of the world, such research has never before been possible in South Africa. Since the acquisition of the Apreo Volumescope, Dr Kriel has worked closely with Prof Carine Smith to make connectomics research possible in South Africa. Using zebrafish as a model organism, they have been able to segment connected neurons in a subset of zebrafish brain tissue. Currently, efforts are being made to conduct this segmentation automatically using deep learning-based approaches. Figure 25 shows the mapping of a connected subset of neurons and quantification of the network connections in a single circuit. 

The Central Analytical Facilities (CAF) Microscopy Unit remains the only microscopy unit in South Africa dedicated to the advancement of multimodal imaging. Although 2D correlative light and electron microscopy (CLEM) is now a streamlined service, 3D CLEM has long been out of reach for South African researchers. CAF has now successfully set up a 3D CLEM workflow to combine confocal and volumetric EM datasets. This approach is currently aiding infectious disease researchers Dr Caroline Beltran and Prof Gerhard Walzl to enhance the understanding of Mycobacterium tuberculosis (mTB) pathophysiology, specifically in the formation of granulomas. Comprised of a variety of immune cells, assessing the distribution of mTB-infected cells in granulomas is made possible by genetic modification of mTB to produce a fluorescent protein. Image B in Figure 26 shows a confocal microscopy image render of granuloma in mouse lung tissue, with the mTB in the red fluorescence channel. Although this allows for the combined visualisation of multiple immune cell markers, the individual bacteria within each infected cell cannot be adequately studied through light microscopy alone. Through a series of careful sample preparation steps, SBF-SEM was conducted on a single granuloma after it had been identified through confocal microscopy (Figure 26 A – E). Overlaying these two datasets gave a clear indication of where the bacteria were located (Figure 26 E – G) and allowed for the segmentation of individual bacteria within infected macrophages. Furthermore, the multinucleated nature of these cells could also be more accurately studied, and preliminary data suggest that these infected cells form an organised network (Figure 26 G). 

Figure 26 A) Confocal tile scan of mouse lung vibratome section. B) Render of confocal image stack (60 x magnification). C) Flat resin embedding and fine cutting of region of interest (ROI). D) SBF-SEM overview image of entire ROI during acquisition. E) Overlay of confocal and SBF-SEM images, with high-resolution ROI outlined in red. F) Region positively identified to contain bacteria. G) Render of infected cells. H) Combined surface render of a single macrophage (green) contained multiple nuclei (grey) and mTB (red). 

The data presented here is a remarkable display of how researchers from various disciplines can gain valuable insights through volumetric EM. Though much progress has been made, the application of SBF-SEM is still in its infancy in South Africa. It is hoped that with further expansion of the skills and expertise necessary to acquire and process SBF-SEM datasets, novel discoveries can be made to pave the way for advances in critical sectors ranging from healthcare to agriculture. 

Images referred to can be seen in the Annual Report

Reference list

Ramón y Cajal, S. Histologie du système nerveux de l’homme & des vertébrés. Vol. 1 (Maloine, 1909).
Denk, W. & Horstmann, H. Serial Block-Face Scanning Electron Microscopy to Reconstruct Three-Dimensional Tissue Nanostructure. PLOS Biol. 2, e329 (2004).