Pseudotyped Bat Coronavirus RaTG13 is efficiently neutralised by convalescent sera from SARS-CoV-2 infected patients

Abstract

RaTG13 is a close relative of SARS-CoV-2, the virus responsible for the COVID-19 pandemic, sharing 96% sequence similarity at the genome-wide level. The spike receptor binding domain (RBD) of RaTG13 contains a number of amino acid substitutions when compared to SARS-CoV-2, likely impacting affinity for the ACE2 receptor. Antigenic differences between the viruses are less well understood, especially whether RaTG13 spike can be efficiently neutralised by antibodies generated from infection with, or vaccination against, SARS-CoV-2. Using RaTG13 and SARS-CoV-2 pseudotypes we compared neutralisation using convalescent sera from previously infected patients or vaccinated healthcare workers. Surprisingly, our results revealed that RaTG13 was more efficiently neutralised than SARS-CoV-2. In addition, neutralisation assays using spike mutants harbouring single and combinatorial amino acid substitutions within the RBD demonstrated that both spike proteins can tolerate multiple changes without dramatically reducing neutralisation. Moreover, introducing the 484 K mutation into RaTG13 resulted in increased neutralisation, in contrast to the same mutation in SARS-CoV-2 (E484K). This is despite E484K having a well-documented role in immune evasion in variants of concern (VOC) such as B.1.351 (Beta). These results indicate that the future spill-over of RaTG13 and/or related sarbecoviruses could be mitigated using current SARS-CoV-2-based vaccination strategies.

Fig. 1. Differences in neutralisation titres between SARS-CoV-2 and RaTG13 by pMN Assay.

a Comparison of neutralisation titres between SARS-CoV-2 and RaTG13 using the WHO International Reference Panel for anti-SARS-CoV-2 immunoglobulin. Three of the four sera showed increased neutralisation titres against RaTG13. b Comparison of neutralisation titres between SARS-CoV-2 (n = 25), RaTG13 (n = 25, p = <0.0001), B.1.351 (Beta) (n = 25, p = <0.0001) variant of concern, SARS-CoV-1 (n = 25, p = <0.0001) and WIV16 (n = 25, p = <0.0001) using convalescent sera derived from patients and healthcare workers. c Comparison of neutralisation titres between SARS-CoV-2 and RaTG13 against sera from single-dose vaccinated healthcare workers (n = 21, p = 0.0016). d Differences in neutralisation titre from single dose vaccinated healthcare workers split by ‘no prior infection’ (n = 11, p = 0.001) or ‘prior infection’ (n = 10, p = 0.084) with SARS-CoV-2. Full circles denote healthcare workers vaccinated with BNT162b2 (n = 12), whereas open circles denote vaccination with AZD1222 (n = 9). Numbers in brackets denote fold changes relative to SARS-CoV-2. Wilcoxon matched pairs signed rank tests were used in panels (b), (c) and (d). Dotted lines in graphs denote the assay’s lower limit of detection. IC50 was calculated by fitting a non-linear regression curve using Graphpad Prism 8 software. All n values constitute of biologically independent samples.

Fig. 2. Key amino acid residues affect antibody neutralisation against SARS-CoV-2 and RaTG13.

a Structures of SARS-CoV-2 (PDB: 6X2A) and RaTG13 (PDB: 7CN4). The highlighted (red) amino acids denote the ACE2 contact residues that were substituted to generate the Multi RBD plasmids for subsequent pMN assays. b Simplified schematic highlighting the functional domains within the SARS-CoV-2 Spike protein. The amino acid substitutions found in RaTG13 have been labelled across the diagram, with red text denoting the substitutions displayed in panel (a), which were built into chimeric Spikes. (FCS, furin cleavage site; NTD, N-terminal domain; RBD, receptor binding domain; RBM, receptor binding motif; FP, fusion peptide; HR, heptad repeat; TM, transmembrane domain). (c) Repeated experiments showing the neutralisation titres, by pMN assay, of SARS-CoV-2 (n = 25), RaTG13 (n = 25, p = <0.0001) and both Sars-CoV-2 Multi RBD (n = 25, p = 0.0005) and RaTG13 Multi RBD (n = 25, p = 0.0043) mutants with the same set of convalescent sera used in Fig. 1. We did not observe a significant difference between the Multi RBD results (n = 25, p = 0.9007). d, e Using a set of four convalescent sera derived from patients, each pseudotype mutant was assayed by pMN and IC50s were then converted to fold changes against their original RaTG13 (d) (K439N; n = 4, p = 0.0194) or SARS-CoV-2 (e) (Q439Y; n = 4, p = 0.0088, N501D; n = 4, p = 0.0069) background. *RaTG13 Y498Q was not performed due to low PV titre (Supplementary Fig 3). f Lysine substitution at position 484 in SARS-CoV-2 Spike significantly reduced neutralisation (n = 20, p = <0.0001). g Lysine substitution at position 484 in RaTG13 however showed a subtle, non-significant increase in neutralisation titres (n = 20, p = 0.1054), whereas an arginine substitution at position 403 significantly reduced the neutralisation titres compared to RaTG13 (n = 20, p < 0.0001) and RaTG13 T484K (n = 20, p < 0.0001). Comparisons of neutralisation in panel (f) and (g) were made using sera from doubly vaccinated HCWs (n = 20). Full circles denote healthcare workers vaccinated with BNT162b2 (n = 11), whereas open circles denote vaccination with AZD1222 (n = 9). Numbers in brackets denote fold changes. Wilcoxon matched pairs signed rank tests was used in panel (c), (f) and (g). Student’s test was used in panels (d) and (e). IC50 was calculated by fitting a non-linear regression curve using Graphpad Prism 8 software. All n values constitute biologically independent samples.