SARS-CoV and SARS-CoV-2 display limited neuronal infection and lack the ability to transmit within synaptically connected axons in stem cell-derived human neurons

Abstract

Sarbecoviruses such as SARS and SARS-CoV-2 have been responsible for two major outbreaks in humans, the latter resulting in a global pandemic. While sarbecoviruses primarily cause an acute respiratory infection, they have been shown to infect the nervous system. However, mechanisms of sarbecovirus neuroinvasion and neuropathogenesis remain unclear. In this study, we examined the infectivity and trans-synaptic transmission potential of the sarbecoviruses SARS and SARS-CoV-2 in human stem cell-derived neural model systems. We demonstrated limited ability of sarbecoviruses to infect and replicate in human stem cell-derived neurons. Furthermore, we demonstrated an inability of sarbecoviruses to transmit between synaptically connected human stem cell-derived neurons. Finally, we determined an absence of SARS-CoV-2 infection in olfactory neurons in experimentally infected ferrets. Collectively, this study indicates that sarbecoviruses exhibit low potential to infect human stem cell-derived neurons, lack an ability to infect ferret olfactory neurons, and lack an inbuilt molecular mechanism to utilise retrograde axonal trafficking and trans-synaptic transmission to spread within the human nervous system.

Keywords: Axonal trafficking; Neurotropism; SARS; SARS-CoV-2; Sarbecovirus; Stem cell–derived neurons; Trans-synaptic transmission.

Fig. 1

Infection of iPSC-derived human neural cultures with sarbecoviruses. A Representative confocal images of stem cell–derived human neural cultures at differentiation day 21–24 infected with different strains of SARS and SARS-CoV-2 at MOI 1 for 24 h. Neurons are identified by staining with TUJ1 (red) and viral antigen identified by staining with anti-SARS S1 antibody (green) and horse antiserum recognising SARS coronavirus (magenta). B Quantification of sarbecovirus-infected day 21–24 differentiated neurons at 24 hpi. C Quantification of sarbecovirus-infected undifferentiated NPCs and day 50 differentiated neurons at 24 hpi. 100–400 neural cells in a single image, totalling at least 1000 neurons, were counted using ImageJ from one to three independent experiments (n = 1, Delta; n = 3, SARS-HK, VIC01, VIC02). Error bars represent mean ± SEM. *p < 0.05, **p < 0.001

Fig. 2

Sarbecovirus infection of ex vivo model of interconnected human neural network to assess trans-synaptic transmission. A Stem cell–derived NPCs were seeded in both panels of microfluidic device separated by microchannels which allows only the neurites to pass through. After seeding, NPCs were differentiated for 20 (RABV) to 21 days (sarbecoviruses) to generate neurons in both panels interconnected by neurites through the microchannels. The neuronal network is identified by TUJ1 staining (red) and F-actin stain to visualise the neural architecture (magenta). The ex vivo models were infected MOI 1 with RABV for 48 h, which served as a positive control for trans-synaptic transmission (B) or different strains of sarbecoviruses (C, D, E) at for 24 h. To prevent passive diffusion of virus from the infected panel into the non-infected panel, a higher volume of media is maintained in the non-infected panel. After 24 h of infection, neural cultures were fixed, stained, and imaged in the microfluidic device. Tile images were taken with 20 × objective with Z-stacks and stitched together. Image B shows trans-synaptic spread of RABV from the infected to non-infected panel identified by staining with a rabies anti-nucleoprotein antibody (green). Images C, D, and E show SARS and SARS-CoV-2 infections stained with anti-SARS2-S1 antibody (green) in a subset of neurons in the infected panel (boxed area) but viral antigen was not observed in the non-infected panel

Fig. 3

Detection of SARS-CoV-2 antigen in ferret nasal neuroepithelium. High-resolution Airyscan confocal imaging of ferret nasal olfactory epithelium infected with SARS-CoV-2 (VIC01 strain). Ferrets were inoculated with 4.64 × 10⁴ TCID₅₀ VIC01 via the intranasal route and ferret tissue collected 7 dpi (Au et al. 2022). Ferret tissue sections analysed herein were stained with anti-SARS-CoV-2 nucleocapsid antibody (red) (A–C), mature olfactory neuronal marker (OMP) (A), immature olfactory neuronal marker (TUJ1) (B), or epithelial (sustentacular) cell marker (cytoK-18) (C). The top two panels of images A–C are images taken with 20 × objective, while the bottom three panels are images taken with 63 × objective using a high-resolution Airyscan module. All images were taken as z-stacks and the maximum intensity projection of z-stacks is shown. Images show that the viral antigen (green) is more prominently co-localised with the sustentacular cells (cytoK-18, red), indicated by white arrows. While virus antigen is observed close to the OMP or TUJ1 positive neurons in images A and B, a clear localisation within the neurons is not observed compared to sustentacular cells (C)