Post by Nadica (She/Her) on Dec 4, 2024 3:27:00 GMT
Enhanced immune evasion of SARS-CoV-2 variants KP.3.1.1 and XEC through N-terminal domain mutations - Published Nov 22, 2024
KP.3, a subvariant of JN.1, has rapidly emerged as the dominant strain of SARS-CoV-2 in several countries, and has been designated as a Variant Under Monitoring. Previous studies indicate that the unique Q493E substitution in KP.3 Spike glycoprotein enhances its receptor ACE2-binding affinity and immune evasion, enabling it to outcompete KP.2.1–6 Notably, KP.3.1.1, which only carries one additional S31 deletion compared to KP.3, has surpassed KP.3 to become the new dominant strain globally (figure A; appendix p 4).7 Meanwhile, XEC, a recombinant variant of KS.1.1 and KP.3.3, shows strong potential to become the next dominant strain, rapidly expanding across Europe and North America. Compared with KP.3, XEC has only two additional spike mutations, F59S and T22N. Both S31del and T22N introduce potential glycosylation on the N-terminal domain. Consequently, there is an imperative need to characterise the antigenicity and infectivity of KP.3.1.1 and XEC.
Here we first used surface plasmon resonance to assess the binding affinity between human ACE2 and the Spike of JN.1, KP.3, KP.3.1.1, and XEC. We found KP.3, KP.3.1.1, and XEC showed a significant increase in ACE2-Spike binding affinity compared with JN.1; however, we did not observe significant changes in the receptor binding of KP.3.1.1 and XEC relative to KP.3, despite the N-terminal domain glycosylation mutations (figure B). We then evaluated the infectivity of KP.3.1.1 and XEC using vesicular stomatitis virus-based pseudoviruses in Vero cells. We found that although the pseudovirus infectivity of KP.3, KP.3.1.1, and XEC increased compared with JN.1, no significant differences were observed among KP.3, KP.3.1.1, and XEC (figure C), which is consistent with another recent study.8
With the vesicular stomatitis virus-based pseudovirus system, in our study we assessed the neutralisation of convalescent plasma against KP.3.1.1 and XEC, involving plasma samples from two cohorts. Participants in both cohorts received two or three doses of inactivated SARS-CoV-2 vaccines and had breakthrough infections with BA.5 or BF.7. One cohort (n=29) was reinfected by JN.1, and the other (n=21) by JN.1 or XDV with F456L (appendix p11). The specific variant causing reinfections was inferred from local prevalence (>90%) at the sampling timepoint (appendix p6). Of note, XDV shares the same Spike sequence as JN.1. We found that KP.3.1.1 and XEC showed significantly enhanced immune evasion capabilities compared with KP.3 (figure D). Compared with KP.3, the 50% neutralisation titre values against KP.3.1.1 decreased by 1·2-fold in the JN.1 group and 1·5-fold in the JN.1 or XDV with F456L group, whereas against XEC values decreased by 1·8-fold in the JN.1 group and 2·1-fold in the JN.1 or XDV with F456L group. Notably, the JN.1 or XDV with F456L group showed significant plasma neutralisation improvement against KP.2 only, with no differences observed against KP.3, KP.3.1.1, and XEC (appendix p7).
Since KP.3.1.1 and XEC do not have additional receptor-binding domain mutations compared with KP.3, and most neutralising antibodies should target the receptor-binding domain instead of the N-terminal domain, we hypothesised the unexpected strong evasion mediated by the N-terminal domain glycosylation mutations could be attributed to potential effects on the neutralising activity of receptor-binding domain-targeting neutralising antibodies through allosteric effects. Indeed, compared with KP.3, KP.3.1.1 and XEC showed enhanced escape capabilities against neutralising antibodies targeting various epitopes on the receptor-binding domain in Vero cells, especially those competing with ACE2 (figure E; appendix p 8). The inhibition efficiency of soluble ACE2 against KP.3.1.1 and XEC also showed a slight reduction. This enhanced antibody escape capability indicates that the glycosylation mutations on the N-terminal domain of KP.3.1.1 and XEC, located away from the epitopes on the receptor-binding domain, potentially reduce receptor-binding domain-targeting neutralising antibody activity via allostery.
In summary, we found that KP.3.1.1 and especially XEC showed enhanced humoral immune evasion and receptor-binding domain-targeting antibody escape capabilities, supporting the foreseeable dominance of the XEC variant of SARS-CoV-2. However, the mechanisms by which T22N and S31del enhance antibody escape from receptor-binding domain-targeting antibodies remain unclear and require further exploration. The potential mechanisms could involve kinetic changes in receptor-binding domain conformational dynamics induced by additional N-terminal domain glycosylation or effect on membrane fusion efficiency. Further studies are necessary to elucidate the mechanism.
KP.3, a subvariant of JN.1, has rapidly emerged as the dominant strain of SARS-CoV-2 in several countries, and has been designated as a Variant Under Monitoring. Previous studies indicate that the unique Q493E substitution in KP.3 Spike glycoprotein enhances its receptor ACE2-binding affinity and immune evasion, enabling it to outcompete KP.2.1–6 Notably, KP.3.1.1, which only carries one additional S31 deletion compared to KP.3, has surpassed KP.3 to become the new dominant strain globally (figure A; appendix p 4).7 Meanwhile, XEC, a recombinant variant of KS.1.1 and KP.3.3, shows strong potential to become the next dominant strain, rapidly expanding across Europe and North America. Compared with KP.3, XEC has only two additional spike mutations, F59S and T22N. Both S31del and T22N introduce potential glycosylation on the N-terminal domain. Consequently, there is an imperative need to characterise the antigenicity and infectivity of KP.3.1.1 and XEC.
Here we first used surface plasmon resonance to assess the binding affinity between human ACE2 and the Spike of JN.1, KP.3, KP.3.1.1, and XEC. We found KP.3, KP.3.1.1, and XEC showed a significant increase in ACE2-Spike binding affinity compared with JN.1; however, we did not observe significant changes in the receptor binding of KP.3.1.1 and XEC relative to KP.3, despite the N-terminal domain glycosylation mutations (figure B). We then evaluated the infectivity of KP.3.1.1 and XEC using vesicular stomatitis virus-based pseudoviruses in Vero cells. We found that although the pseudovirus infectivity of KP.3, KP.3.1.1, and XEC increased compared with JN.1, no significant differences were observed among KP.3, KP.3.1.1, and XEC (figure C), which is consistent with another recent study.8
With the vesicular stomatitis virus-based pseudovirus system, in our study we assessed the neutralisation of convalescent plasma against KP.3.1.1 and XEC, involving plasma samples from two cohorts. Participants in both cohorts received two or three doses of inactivated SARS-CoV-2 vaccines and had breakthrough infections with BA.5 or BF.7. One cohort (n=29) was reinfected by JN.1, and the other (n=21) by JN.1 or XDV with F456L (appendix p11). The specific variant causing reinfections was inferred from local prevalence (>90%) at the sampling timepoint (appendix p6). Of note, XDV shares the same Spike sequence as JN.1. We found that KP.3.1.1 and XEC showed significantly enhanced immune evasion capabilities compared with KP.3 (figure D). Compared with KP.3, the 50% neutralisation titre values against KP.3.1.1 decreased by 1·2-fold in the JN.1 group and 1·5-fold in the JN.1 or XDV with F456L group, whereas against XEC values decreased by 1·8-fold in the JN.1 group and 2·1-fold in the JN.1 or XDV with F456L group. Notably, the JN.1 or XDV with F456L group showed significant plasma neutralisation improvement against KP.2 only, with no differences observed against KP.3, KP.3.1.1, and XEC (appendix p7).
Since KP.3.1.1 and XEC do not have additional receptor-binding domain mutations compared with KP.3, and most neutralising antibodies should target the receptor-binding domain instead of the N-terminal domain, we hypothesised the unexpected strong evasion mediated by the N-terminal domain glycosylation mutations could be attributed to potential effects on the neutralising activity of receptor-binding domain-targeting neutralising antibodies through allosteric effects. Indeed, compared with KP.3, KP.3.1.1 and XEC showed enhanced escape capabilities against neutralising antibodies targeting various epitopes on the receptor-binding domain in Vero cells, especially those competing with ACE2 (figure E; appendix p 8). The inhibition efficiency of soluble ACE2 against KP.3.1.1 and XEC also showed a slight reduction. This enhanced antibody escape capability indicates that the glycosylation mutations on the N-terminal domain of KP.3.1.1 and XEC, located away from the epitopes on the receptor-binding domain, potentially reduce receptor-binding domain-targeting neutralising antibody activity via allostery.
In summary, we found that KP.3.1.1 and especially XEC showed enhanced humoral immune evasion and receptor-binding domain-targeting antibody escape capabilities, supporting the foreseeable dominance of the XEC variant of SARS-CoV-2. However, the mechanisms by which T22N and S31del enhance antibody escape from receptor-binding domain-targeting antibodies remain unclear and require further exploration. The potential mechanisms could involve kinetic changes in receptor-binding domain conformational dynamics induced by additional N-terminal domain glycosylation or effect on membrane fusion efficiency. Further studies are necessary to elucidate the mechanism.