Evolutionary and Phenotypic Characterization of Two Spike Mutations in European Lineage 20E of SARS-CoV-2

Abstract

We have detected two mutations in the spike protein of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) at amino acid positions 1163 and 1167 that appeared independently in multiple transmission clusters and different genetic backgrounds. Furthermore, both mutations appeared together in a cluster of 1,627 sequences belonging to clade 20E. This cluster is characterized by 12 additional single nucleotide polymorphisms but no deletions. The available structural information on the S protein in the pre- and postfusion conformations predicts that both mutations confer rigidity, which could potentially decrease viral fitness. Accordingly, we observed reduced infectivity of this spike genotype relative to the ancestral 20E sequence in vitro, and the levels of viral RNA in nasopharyngeal swabs were not significantly higher. Furthermore, the mutations did not impact thermal stability or antibody neutralization by sera from vaccinated individuals but moderately reduce neutralization by convalescent-phase sera from the early stages of the pandemic. Despite multiple successful appearances of the two spike mutations during the first year of SARS-CoV-2 evolution, the genotype with both mutations was displaced upon the expansion of the 20I (Alpha) variant. The midterm fate of the genotype investigated was consistent with the lack of advantage observed in the clinical and experimental data. IMPORTANCE We observed repeated, independent emergence of mutations in the SARS-CoV-2 spike involving amino acids 1163 and 1167, within the HR2 functional motif. Conclusions derived from evolutionary and genomic diversity analysis suggest that the co-occurrence of both mutations might pose an advantage for the virus and therefore a threat to effective control of the epidemic. However, biological characterization, including in vitro experiments and analysis of clinical data, indicated no clear benefit in terms of stability or infectivity. In agreement with this, continuous epidemiological surveillance conducted months after the first observations revealed that both mutations did not successfully outcompete other variants and stopped circulating 9 months after their initial detection. Additionally, we evaluated the potential of both mutations to escape neutralizing antibodies, finding that the presence of these two mutations on their own is not likely to confer antibody escape. Our results provide an example of how newly emerged spike mutations can be assessed to better understand the risk posed by new variants and indicate that some spike mutations confer no clear advantage to the virus despite independently emerging multiple times and are eventually displaced by fitter variants.

Keywords: HR2; SARS-CoV-2; adaptive mutations; antibody escape; homoplasy; spike; variants.

FIG 1

Sequences mutated at positions 1163 and 1167 of the S protein. (a) The number of mutation events for amino acid replacement S:D1163Y (light orange) or another S:D1163 amino acid replacements (dark orange). (b) The number of mutation events for amino acid replacement S:G1167V (light turquoise) or another S:G1167 amino acid replacements (dark turquoise). Bars in magenta indicate the appearance of both the S:D1163Y and S:G1167V amino acid replacements in the same sequences. (c) Maximum-likelihood phylogeny of 10,450 SARS-CoV-2 genomes. The inner ring represents sequences with amino acid changes in position D1163 of the S protein. The outer ring represents sequences with amino acid changes in position G1167 of the S protein. Branches are colored in magenta for 1163.7, green for clade 20E, and orange for cluster 1163.654. The scale bar indicates the number of nucleotide substitutions per site. (d) Temporal distribution and frequency of sequences with variant 1163.7 colored by geographical origin.

FIG 2

Temporal distribution of sequences in GISAID per variant. Number of sequences classified as cluster 1163.7 (n = 2,106), 20E (n = 159,450), and 20I (n = 979,013) by date from June 2020 until the beginning of July 2021.

FIG 3

Structure of 1163 and 1167 in the pre- and postfusion states of S protein. (a) Schematic representation of the S protein. SP, signal peptide; NTD, N-terminal domain; RBD, receptor-binding domain; SD1 and SD2, subdomains 1 and 2; L-UH, linker-upstream helix; FP, fusion peptide; CR, connecting region; HR1, heptad repeat 1; CH-SD3, central helix subdomain 3; BH, β-hairpin; HR2, heptad repeat 2; TM, transmembrane; CD, cytoplasmic domain. Amino acid changes D1163Y and G1167V are indicated in purple, and other mutations described in the text are in green. (b) (Left) Cartoon representation of a structural model of prefusion membrane-bound trimeric S protein (77). In each subunit, the RBD, HR1, and HR2 domains are colored in different tones (light to dark) of blue, yellow, and green. The N-glycosylation of N1155 and N1176 is shown in stick representation and colored as the corresponding subunit. Functional and structural regions are marked. (Right) Close-up view of the N-terminal portion of HR2 where D1163Y and G1167V amino acid replacements are found. The side chains of mutated and hydrophobic residues in the HR2 region are shown in stick representation and colored as the corresponding subunit (mutated residues in a lighter tone). (c) Cartoon representation (left) of S2 subunit in postfusion conformation with HR1 and HR2 regions colored as in panel b and N-glycosylation around mutation position shown as sticks. (Right) Close-up view of the region encompassing the mutations (right), showing in stick representation the mutated and hydrophobic residues from the HR2 region shown in panel b. Dotted lines highlight HR2 disordered regions in the cryo-electron microscopy structure.

FIG 4

Comparison of the infectivity and stability of different S genotypes. (a and b) The infectivity of VSV particles pseudotyped with each S protein genotype in either Vero cells (a) or human A549 cells expressing ACE2 and TMPRSS2 (b). Means and standard deviations for three replicates are plotted. (c) Comparison of cycle threshold (C_T) values for the N gene from patients infected with viruses encoding different S protein variants. Data are derived from 2,534 sequences from the SeqCOVID consortium. The number of observations (N) analyzed for each genotype is indicated. (d) The thermal sensitivity of VSV pseudotyped with different S genotypes following incubation at 15 min. Data are standardized to the surviving fraction following incubation at 30°C, and the three-parameter log-logistic equation is plotted. FFU, focus forming units.

FIG 5

Antibody neutralization of 20E and 1163.7 variants. The reciprocal titer at which infection with the 20E S genotype (S:A222V and S:D614G) or 1163.7 S genotype (20E plus S:D1163Y and S:G1167V) is reduced by 80% (ID₈₀) by sera from individuals infected during the early stage of the pandemic (a) or during a later stage of the pandemic (b) and from donors vaccinated with the BNT162b2 vaccine (c). The means and standard errors for three replicates are plotted.