The Department of Infectious Diseases, Virology, of Heidelberg University uses Amira Software to characterize the replication cycle of SARS-CoV-2 inside host cells

Viruses extensively remodel host cells during infection to induce their replication and production of viral progeny. Characterizing the replication cycle inside the host cell is key to understanding virulence and discovering targets restricting virus replication. In light of the current COVID-19 pandemic, we applied in situ cryo-electron tomography on cryo-focused ion beam-milled lamella of cells infected with SARS-CoV-2 to visualize hubs of virus replication and assembly. We characterized double-membrane vesicles, compartments associated with viral replication, and directly visualized enclosed viral RNA filaments. We observed virion assembly sites at cisternae enriched in assembled viral spike proteins that laterally reorganized on the envelope during budding.

SARS-CoV-2 virion budding and assembly at the ERGIC membrane. A 3D volume rendering is shown with cellular and viral membranes in green and magenta respectively. Viral spike proteins are shown in yellow, and viral ribonucleoproteins in cyan.

The first aim was to enhance the contrast of viral and cellular membranes. We applied a combination of 3D non-local means filter and the Membrane Enhancement Filter implemented in Amira Software. An initial automated segmentation was achieved using the Top-hat segmentation tool. This initial segmentation was further manually refined utilizing a Wacom graphics tablet. Based on the refined membrane segmentation, a 3D surface was calculated using unconstrained smoothing, and the surface was visualized as 3D model.

In the next step, subtomogram averages of the viral proteins S and vRNP were imported to the 3D rendering and visualized using isosurface rendering. Using the advanced scripting capabilities of Amira Software, the known positions of all S and vRNPs present in the tomogram were imported, and individual isosurfaces were positioned accordingly.

Finally, the complete 3D rendering showing cellular and viral membranes together with subtomogram averages of viral S proteins and vRNPs was exported as high-quality figures and movies used for publication.

Spatial distribution and length of RNA filaments in a double-membrane vesicle. Inner and outer membrane are represented in light and dark green, respectively. Individual segmented filaments are colored according to their length.

We particularly like the general idea of how Amira Software is organized in the project view. This helps to set up complex data processing workflows. The great variety of specialized image processing modules makes Amira Software particularly useful for EM data. The segmentation modules are very flexible, and it is easy to combine different tools enabling the segmentation of complex structures.

The authors acknowledge microscopy support from the Infectious Diseases Imaging Platform (IDIP) at the Center for Integrative Infectious Disease Research Heidelberg. The authors acknowledge access to the infrastructure and support provided by the Cryo-EM Network at the Heidelberg University (HD-cryoNet). This work was supported by a research grant from the Chica and Heinz Schaller Foundation (Schaller Research Group Leader Program) and by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation).

This project would not be possible without the highly collaborative environment of the Center for Integrative Infectious Diseases Research (CIID). Work with SARS-CoV-2 was performed in the BSL3 laboratories, and all infectious samples were inactivated before undergoing plunge freezing and cryo-ET performed in the Thermo Scientific™ Krios™ Cryo-TEM (cryo-transmission electron microscope). The “Infection Diseases Imaging Platform (IDIP)” provided Leica advanced equipment for cryo-EM sample preparation and light microscopy. The Cryo-EM Network at the Heidelberg University (HD-cryoNet) provides state-of-the-art cryo-EM infrastructure.