High Individual Heterogeneity of Neutralizing Activities against the Original Strain and Nine Different Variants of SARS-CoV-2

Figures

• 

  Figure 1. Figure showing the neutralising response against the original virus (Original/B.1.1) and the 9 SARS-CoV-2 variants in convalescent patients and in vaccinated participants. The Y-axis represents the neutralising antibody titres obtained by MNT, and the X-axis represents the corresponding group of tested human sera. Data are shown as the mean and standard error of the mean (SEM). Solid lines represent the geometric mean titre, and the whiskers show the 95% confidence interval. Each scatter represents one serum. Scatter symbols and colours are attributed on the basis of the sera group. Serology results are shown for the convalescent patients and for the vaccinated patients. For seroneutralisation, the 10 quadrants correspond to one different tested strain each. The last quadrant shows the IgG titres in AU/mL as obtained by chemiluminescent immunoassay (CLIA) for each serum group. IgG titres > 400 are represented in the graph above the maximum threshold of detection (400 AU/mL). The same statistical significance was obtained by excluding these nonquantitative values (>400) from the ANOVA test. In all graphs, significance is represented by an asterisk for p ≤ 0.05. The absence of an…

• 

  Figure 2. Figure showing the neutralising response of each serum group against the original virus (Original/B.1.1) and the 9 SARS-CoV-2 variants in convalescent patients and in vaccinated participants. The Y-axis represents the neutralising antibody titres obtained by MNT, and the X-axis consists of the different tested SARS-CoV-2 strains. Data are shown as the mean and standard error of the mean (SEM). Solid lines represent the geometric mean titre, and the whiskers show the 95% confidence interval. Each scatter represents one serum. Scatter’s symbols and colours are strain-specific. Each quadrant in the figure corresponds to one different sera group. In all graphs, significance is represented by an asterisk for p ≤ 0.05. Significance is represented as an asterisk for significant p-values. The absence of an asterisk means no significant variance was detected between groups.

• 

  Figure 3. Figure showing a scheme of the neutralising activity results for the monoclonal antibody bamlanivimab. Table showing the neutralising activity of LY-CoV555 against the 10 tested strains. Neutralising concentrations are represented by grey colour gradient cases and are displayed in μg/mL units. Darker grey colours reflect higher neutralisation activity.

• 

  Figure 4. Scheme showing the molecular mechanisms of nAb escape in the RBD of SARS-CoV-2 variants. (A) The epitope recognised by the LY-CoV555 nAb (pdb file #7KMG) consists of several amino acid (coloured in green) residues distributed on the surface of the RBD. The anionic carboxylic group of E484 interacts with the cationic charge of R50 (heavy chain of LY-CoV555 nAb) through an electrostatic bridge. The aromatic ring of Y490 interacts with a methyl group of I52 (heavy chain of LY-CoV555 nAb) by a CH–π interaction, which is reinforced by vicinal apolar amino acid residues (I54 and I55). (B) In the Marseille-484K.V1/R.1 variant, E484 (in red in the left panel) is mutated in E484K (in blue in the right panel). The consequence of this mutation is a shift of the side chain of E484K whose cationic group (which replaces the negative charge of E484) now forms a cation–π bond with the aromatic ring of F490. In this new context, neither E484K nor F490 can still interact with the LY-CoV555 nAb. Indeed, R50, I52, I54, and L55 of the heavy chain of the antibody are no longer involved in RBD recognition. (C) Mutational pattern of the Marseille-501/A.27 variant (L452R/N501Y). (D) Mutational pattern of…

• 

  Figure 5. Figure showing the molecular mechanism of nAb escape by the RBD of the Marseille-501/A.27 (Mrs-501) variant. (A) In the Original/B.1.1 RBD (PDB file #7KMG), the heavy chain of the LY-CoV 555 nAb interacts with the side chains of L452 (Van der Waals network) and Y449 (NH–π). (B) In the Marseille-501/A.27 variant, the mutant R452 is displaced out of the Van der Waals network, which reorients Y449 so that the NH–π interaction with N-31 is no longer possible. The aromatic ring of Y501 comes closer to Y449, which definitely prevents any contact with the heavy chain of the nAb.

• 

  Figure 6. Figure showing the molecular mechanism of nAb escape by the RBD of the Delta/B.1.617.2 variant. (A) A π–π aromatic cluster is involved in the recognition of the RBD (Original/B.1.1 strain) by the LY-CoV555 nAb (pdb file #7KMG). Y32 and Y92 of the light chain of the antibody clamp the aromatic ring of RBD residue F486. This cluster is stabilised by a CH–π interaction between the methyl groups of T478 and F486. It should be noted that this interaction is important to functionally orient and wedge the side chain of F486 between Y32 and Y92. (B) In the Delta/B.1.617.2 variant, the mutation T478K prevents the formation of this network by forming a cation–π interaction between the cationic group of this residue and the aromatic ring of F486. This new bond reorients the side chain of F486 towards the RBD surface, thereby preventing any possibility of association with the antibody. The models are shown in sphere (upper panels) or stick (lower panels) representations.

• 

  Figure 7. Figure showing the variability of the main neutralising epitope in the NTD among virus strains. (A) Molecular mechanism of NTD recognition (Original/B.1.1 strain) by the 4A8 nAb. The NTD–nAb complex (pdb file #7C2 L) has a global affinity of −225 kJ/mol−1. The antibody clamps two distal zones of the NTD (the N3 loop with amino acid residues K147, K150, and W152) and the N5 loop (R246, Y248, and L249), which together form the main neutralising epitope of the NTD. (B) The NTD of the Alpha variant retains this crescent-shaped structure, which displays a slightly higher affinity (+ 7% compared with the Original/B.1.1 strain) for the 4A8 nAb due to the repositioning of the amino acids of the N3 loop. (C) In the case of the Betavariant, only the N3 loop part of the epitope is conserved, and therefore that the affinity for the 4A8 nAb is decreased by 50%. Such a truncated epitope may elicit a poor antibody response, consistent with seroneutralisation data.

• 

  Figure 8. Figure showing the intralineage variability of the main neutralising epitope in the NTD: the case of India variants. (A) Delta/B.1.617.2 variant. (B) Delta/AY.37 variant. The left panels in (A) and (B) show the positions of residues K147 and R246 that face the antibody. The superposition of the secondary structure shows that the conformational changes between these variants chiefly involve the amino acid side chain orientation. In the case of Delta/B.1.617.2, both K147 and R246 interact with the antibody, whereas in the case of Delta/AY.37, R246 moves away from the antibody, while K147 gets closer, resulting in decreased binding to avoid steric clash. The middle panels show the NTD surface as “seen” by the antibody. The right panels show the subtle rearrangements of the NTD structure of both India variants. The main change concerns the orientation of K147.