Omicron variant spike protein shows much weaker membrane fusion activity than other variants

The Omicron variant for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) showed significantly higher transmission rates than the other variants, despite lower rates of severe disease. Researchers from Harvard Medical School are studying the membrane fusion activity and spike protein structure of this variant in order to better understand these differences.

Stady: Structural and functional influence by SARS-CoV-2 Omicron mutants. Image Credit: MattLphotography / Shutterstock

Researchers study can be found at bioRxiv* Server prepress, while article is subject to peer review.


The researchers transfected HEK cells with the Omicron spike protein expression construct and compared the membrane fusion activity with combinations of variants such as Alpha, Beta and delta in order to characterize the full-length spike protein. All spike proteins were expressed at similar levels, although Omicron proteins cleaved less frequently between the two subunits at 24 h after transfection, which may indicate that the two mutations near the furin cleavage site do not enhance spike protein processing. Cells producing these proteins successfully fused with ACE2 cells, although the fusion activity of Omicron spike protein was again lower than that of the other variants.

The scientists then performed a time course experiment with a cell-cell fusion assay, using both spike protein and ACE2 at saturating levels, in order to better test whether Omicron spike protein could stimulate membrane fusion more efficiently than other variables. Again, there were no significant differences in fusion activity other than the Omicron spike protein showing slightly lower activity. When using cells expressing a minimal level of endogenous ACE2, all other variants showed significant fusion activities at later time points, but the Omicron spike protein was mostly inactive.

The researchers assumed that Omicron struggled to infect cells with low levels of ACE2, so they tried using HEK cells transfected with different amounts of ACE2. Again, the Omicron spike protein lagged behind the other variants, requiring a 10-fold increase in ACE2 in order to reach similar fusion activity. This was observed again when spike-producing cells were transfected with the furin expression construct, and target cells were co-transfected with TMPRSS2.

The full-length Omicron spike protein was produced without any modifications using a C-terminal bacterial-tagged construct for expression. The purified protein was then examined using gel filtration chromatography. The wild-type Wuhan-Hu-1 spike protein is resolved into three peaks, which correspond to the infusion trimmer, the post-fusion separated S1 monomer, and the separated S1 monomer. Another variant, G614, shows a spike protein with a single peak of trimer processing.

The Omicron protein similarly presents one main peak, but there is a large amount of aggregate on the anterior side, and a scapula on the reverse side. This indicates that the Omicron spike protein is significantly less stable than the G614 protein. Further analysis using SDS-PAGE reveals a significant portion of the unbroken protein at 84 h after transfection. This supports the theory that Omicron struggles to transport cells with low levels of ACE2 due to reduced cleavage of the spiny protein.

Next, the researchers attempted to determine the cutter’s structure based on cryo-EM images. 3D classification revealed three different classes of spike protein, which represent a closed mode with all three receptor-binding domains (RBDs) facing ‘down’, forming a ‘single RBD-up’ and the RBD intermediate conformation seen in the G614 trimmer. They were then further refined and modeled to reveal no significant differences in the structure of the full-length Omicron spike protein and the G614 spike protein. For both variants, comprehensive closed confirmation showed that the N-terminal domain, RBD, C-terminal domain 1 (CTD1) and C-terminal domain 2 (CTD2) of S1 wrap around the S2 trimer. The upward RBD shape retained the helical core structure of S2 despite the eversion of the RBD upward, displacing the two adjacent N-terminal domains away and dramatically opening the trimer.

The researchers suggest that additional residues near the furin cleavage site that are significantly more organized due to the N679K mutation could reduce the flexibility at the furin cleavage site, slowing docking at the active furin site.


The authors demonstrate that the Omicron spike protein requires much higher levels of ACE2 for effective membrane fusion, lagging behind other variants. They also provided strong evidence that the Omicron spike protein is significantly less prone to cleavage than other variants. With further investigation of the structure of the spike protein, they identified a mutation and resulting structural change that could be responsible for this change. This information could be key to drug developers, and could help locate furin cleavage as a target for treatment, which could alleviate disease.

Important note:

bioRxiv publishes preliminary scientific reports that are not subject to peer review, and therefore should not be considered conclusive, guide clinical practice/health-related behavior, or be treated as established information.


Leave a Comment