Deep mutational scanning for SARS-CoV-2 Omicron BA.1 and BA.2 subvariants

In a recent study posted to the bioRxiv* preprint server, researchers performed deep mutational scanning (DMS) for the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Omicron variant of concern (VOC) sub-VOC BA.1 and BA.2 receptor-binding domains (RBDs).

Background

The continual evolution of SARS-CoV-2 by acquiring mutations (mut) in the spike (S) protein RBD may impact angiotensin-converting enzyme 2 (ACE2) binding, folding stability, and antibody (Ab) recognition. DMS can characterize mutational impacts on such biochemical properties and inform SARS-CoV-2 surveillance efforts. However, the mutational impact may vary with SARS-CoV-2 evolution due to epistasis, warranting the need for updated DMS analysis.

The Omicron VOC has shown the most prominent evolutionary changes in the SARS-CoV-2 RBD with Omicron BA.1 and Omicron BA.2 sub-VOCs comprising 15 RBD mut and 16 RBD mut, respectively. However, the reshaping of SARS-CoV-2 evolutionary characteristics by Omicron is not well-characterized. The authors of the present study previously showed that core mut such as Y365W, F492W, and I358F considerably enhance the stability and yield of the Wuhan-Hu-1 strain (ancestral) RBD and RBD-based nanoparticle vaccines with no alterations in antigenicity.

About the study

In the present study, researchers extended their previous analysis by performing DMS to evaluate the Omicron BA.1 and Omicron BA.2 RBD mutational impacts on ACE2 receptor binding, RBD folding, and bebtelovimab monoclonal Ab (mAb) recognition. In addition, the team compared obtained data to previously published data of pre-Omicron VOCs to determine potential epistatic shifts.

Protein stability was assessed in terms of RBD expression. Duplicate libraries were prepared by site-saturation mutagenesis in the background of Omicron BA.1 and Omicron BA.2 RBD, and the mutant VOC libraries were cloned in yeast-display platforms in the Wuhan-Hu-1 background. The impact of RBD mut on ACE2 receptor-binding affinity was evaluated based on the surface expression by pooled yeast-display VOC platforms incubated with varying hACE2 (human ACE2) concentrations.

Fluorescence-activated cell sorting-sequencing (FACS-seq) analysis was performed to quantify the RBD expression and ACE2 binding strength of every mutant library at every ACE2 concentration. Epistatic shifts in the Omicron VOC’s mutational landscape were assessed in relation to the ancestral strain by computing an epistatic shift metric to identify positions in Omicron BA.1 and BA.2 RBDs with considerable mutational alterations in relation to the ancestral strain. Further, biolayer interferometry (BLI) analysis was performed to assess Omicron BA.1 and BA.2 RBD ACE2-binding kinetics. Further, the team assessed pathways of bebtelovimab escape facilitated by RBD mut.

Results

The impact of a few mut in Omicron RBDs differed from those in the ancestral strain RBD, with epistatic shifts resembling those observed for the Beta VOC due to the convergent and epistatically altering N501Y mutation. However, Omicron VOCs showed additional sublineage-specific epistatic shifts, including epistatic entrenchment that enhanced the favourability of N501Y and Q498R Omicron mut for ACE2 binding in comparison to pre-Omicron VOCs.

Contrastingly, the Omicron Q493R mutation showed no entrenchment, with the R493 mutation being as unfavorable for ACE2 receptor binding in Omicron RBDs as in the ancestral strain RBD. The R493Q reversion must have occurred in Omicron sub-VOCs such as BA.4/5 and BA.2.75 and potentiated concomitant antigenic alterations. Omicron sub-VOCs BA.1 and BA.2 showed lower RBD expression and Ab escape site widening in comparison to those of the ancestral strain.

Similar to previously documented DMS findings for the Alpha VOC, Beta VOC, Delta VOC, and Eta strain RBDs, the two Omicron sub-VOC RBDs showed high mutational tolerance. For Omicron, a few affinity-increasing mut were secondary alterations or reversions at R493 or N417 residue sites that developed mut when Omicron emerged. Omicron stabilizing mut were identified at positions 358, 363, 365, and 392 in the RBD core and positions 369, 374, and 376 in the peripheral loop region.

Of note, the ‘rpk9’ stabilizing mut combination in the ancestral strain RBD (V395I, F392W, and Y365F) enhanced Omicron BA.1 surface expression. Sites 439, 453, and 455 were unaltered between Wuhan-Hu-1 and Omicron; however, the positions showed epistatic shifts that altered the affinity-enhancing mut availability. E.g., N439K enhanced ancestral strain ACE2-binding affinity and convergently occurred in initially circulating SARS-CoV-2 VOCs reducing the Omicron RBDs binding affinity for ACE2. Conversely, L455W reduced Wuhan-Hu-1 affinity but increased Omicron sub-VOC affinity for ACE2.

Y455 and W455 mut in SARS-CoV-1 and RsSHC014 (SARS-CoV-2 numbering), respectively, and epistatic shifts that enhance ACE2 receptor binding in Omicron indicated that the site 455 was potentially relevant for the future evolution of SARS-CoV-2. The N501Y ancestral strain mut enhanced ACE2-binding affinity by 12-fold but Y501N reduced the affinity by 288-fold and 1096-fold for Omicron BA.1 and Omicron BA.2, respectively. However, the entrenchment patterns indicated that N501Y reversion was unlikely to occur in the future Omicron sub-VOCs. The BLI analysis showed similar findings.

The R493Q reversion in Omicron RBD enhanced ACE2 receptor-binding affinity, enabling additional ACE2-binding affinity-reducing mut such as F486V in the VOCs, and facilitated immune evasion. Bebtelovimab escape in Wuhan-Hu-1 was attributed to V445, K444 and P499 residue mut, and was seen in Omicron sub-VOC RBDs, with more mut and escape sites, including those at site 466.

Conclusion

Overall, the study findings highlighted epistatic shifts and mutational profiles of Omicron BA.1 and Omicron BA.2 by DMS, that could inform continued efforts in SARS-CoV-2 surveillance and forecasting.

*Important notice

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