Deep mutational scans of XBB.1.5 and BQ.1.1 reveal ongoing epistatic drift during SARS-CoV-2 evolution

Abstract

Substitutions that fix between SARS-CoV-2 variants can transform the mutational landscape of future evolution via epistasis. For example, large epistatic shifts in mutational effects caused by N501Y underlied the original emergence of Omicron variants, but whether such large epistatic saltations continue to define ongoing SARS-CoV-2 evolution remains unclear. We conducted deep mutational scans to measure the impacts of all single amino acid mutations and single-codon deletions in the spike receptor-binding domain (RBD) on ACE2-binding affinity and protein expression in the recent Omicron BQ.1.1 and XBB.1.5 variants, and we compared mutational patterns to earlier viral strains that we have previously profiled. As with previous RBD deep mutational scans, we find many mutations that are tolerated or even enhance binding to ACE2 receptor. The tolerance of sites to single-codon deletion largely conforms with tolerance to amino acid mutation. Though deletions in the RBD have not yet been seen in dominant lineages, we observe many tolerated deletions including at positions that exhibit indel variation across broader sarbecovirus evolution and in emerging SARS-CoV-2 variants of interest, most notably the well-tolerated Δ483 deletion in BA.2.86. The substitutions that distinguish recent viral variants have not induced as dramatic of epistatic perturbations as N501Y, but we identify ongoing epistatic drift in SARS-CoV-2 variants, including interaction between R493Q reversions and mutations at positions 453, 455, and 456, including mutations like F456L that define the newly emerging EG.5 lineage. Our results highlight ongoing drift in the effects of mutations due to epistasis, which may continue to direct SARS-CoV-2 evolution into new regions of sequence space.

A deep learning framework for predicting the neutralizing activity of COVID-19 therapeutics and vaccines against evolving SARS-CoV-2 variants

“…On further analysis of mutations linked to reduced responsiveness, our model identifies the spike mutation F456L, present in EG.5 and FL.1.5.1, as a primary contributor to reduced neutralization by selected COVID-19 antibodies and vaccines. This observation aligns remarkably with recent studies using pseudovirus neutralization assays, 28,30 and deep mutational scanning, 31 which highlighted the mutation’s role in ACE2 receptor binding. The F456L mutation emerged independently in XBB descendant strains such as EG.5, FL.1.5.1, and XBB.1.16.6 and was found in 40% of newly uploaded SARS-CoV-2 sequences by August 2023.…”

“…Additionally, consistent with recent studies, 27,33,34 our model predicted that monoclonal antibodies including Sotrovimab, Cilgavimab, and Bebtelovimab that previously retained some activity against parental BA.2 variants have reduced neutralization activity against BA.2.86. Regarding mutations linked to reduced effectiveness, our findings align with a recent deep mutational scanning study, 31 which suggests that BA.2.86 mutations are generally well-tolerated and individual mutations do not significantly impact neutralization activities.…”

“…The F456L mutation emerged independently in XBB descendant strains such as EG.5, FL.1.5.1, and XBB.1.16.6 and was found in 40% of newly uploaded SARS-CoV-2 sequences by August 2023. 28 Continuous shifts at this position and its vicinity, highlighted by deep mutational scanning, 31 suggest the virus is evolving to dodge antibody responses while effectively binding to its target receptors. Remarkably, our deep learning framework successfully discerned the robust immune escape properties of this mutation, despite not seeing it within training sequences for the uncertainty-based neural network, although the mutation was encountered in the VAE training sequences.…”

Delineating the functional activity of antibodies with cross-reactivity to SARS-CoV-2, SARS-CoV-1 and related sarbecoviruses

“…Another explanation for this differential effect in neutralization despite sequence conservation across sarbecoviruses may be due to epistatic shifts as the accumulation of mutations outside the antibody binding site may lead to shifts in the effect of residues on antibody function, which has been observed in other viral contexts (89,90) as well as in SARS-CoV-2 variants (65,69). While we performed our DMS experiments in the genetic background of the SARS-CoV-2 WH-1 or Omicron BA.2, the effects of mutations may differ in different viral contexts (15,65). SARS-CoV-2 evolution is likely shaped by both mutation and positive selection to evade immunity, yet the capacity to escape antibody recognition can be limited by selection to maintain functional capabilities (41).…”

“…Recently, one of these clade 3 sarbecovirus has been found to utilize human ACE2 to infect cells, and human ACE2-independent clade 3 sarbecoviruses can acquire mutations that permit binding to human ACE2 with just two substitutions thus potentially broadening their host cell range to include humans as well (11,13). The COVID-19 pandemic has demonstrated the ability of the SARS-CoV-2 RBD to tolerate mutations (14,15), and evidence of recombinants (16) highlights the diversity of RBD that can be generated within a single sarbecovirus lineage that has circulated in the human population in just a few years. The large number of sarbecoviruses in animal reservoirs that can use the human ACE2 receptor for entry, coupled with the pathways for diversity and mutational tolerance among these viruses, underscores the urgent need to understand the immune responses that may be useful for eliciting in future spillover events.…”