Back

Loss-of-function mutation in Omicron variants reduces spike protein expression and attenuates SARS-CoV-2 infection

Vu, M. N.; Morris, D. R.; Alvarado, R. E.; Lokugamage, K. G.; Zhou, Y.; Estes, L. K.; McLeland, A. M.; Schindewolf, C.; Plante, J. A.; Ahearn, Y. P.; Morgan, A. L.; Meyers, W. M.; Murray, J. T.; Weaver, S.; Walker, D. H.; Russell, W. K.; Routh, A. L.; Plante, K. S.; Menachery, V. D.

2023-04-18 microbiology
10.1101/2023.04.17.536926 bioRxiv
Show abstract

SARS-CoV-2 Omicron variants emerged in 2022 with >30 novel amino acid mutations in the spike protein alone. While most studies focus on receptor binding domain changes, mutations in the C-terminus of S1 (CTS1), adjacent to the furin cleavage site, have largely been ignored. In this study, we examined three Omicron mutations in CTS1: H655Y, N679K, and P681H. Generating a SARS-CoV-2 triple mutant (YKH), we found that the mutant increased spike processing, consistent with prior reports for H655Y and P681H individually. Next, we generated a single N679K mutant, finding reduced viral replication in vitro and less disease in vivo. Mechanistically, the N679K mutant had reduced spike protein in purified virions compared to wild-type; spike protein decreases were further exacerbated in infected cell lysates. Importantly, exogenous spike expression also revealed that N679K reduced overall spike protein yield independent of infection. Although a loss-of-function mutation, transmission competition demonstrated that N679K had a replication advantage in the upper airway over wild-type SARS-CoV-2 in hamsters, potentially impacting transmissibility. Together, the data show that N679K reduces overall spike protein levels during Omicron infection, which has important implications for infection, immunity, and transmission.

Published in Nature Communications (predicted rank #1) · training set

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.