Glycan masking of a non-neutralising epitope enhances neutralising antibodies targeting the RBD of SARS-CoV-2 and its variants

Abstract

The accelerated development of the first generation COVID-19 vaccines has saved millions of lives, and potentially more from the long-term sequelae of SARS-CoV-2 infection. The most successful vaccine candidates have used the full-length SARS-CoV-2 spike protein as an immunogen. As expected of RNA viruses, new variants have evolved and quickly replaced the original wild-type SARS-CoV-2, leading to escape from natural infection or vaccine induced immunity provided by the original SARS-CoV-2 spike sequence. Next generation vaccines that confer specific and targeted immunity to broadly neutralising epitopes on the SARS-CoV-2 spike protein against different variants of concern (VOC) offer an advance on current booster shots of previously used vaccines. Here, we present a targeted approach to elicit antibodies that neutralise both the ancestral SARS-CoV-2, and the VOCs, by introducing a specific glycosylation site on a non-neutralising epitope of the RBD. The addition of a specific glycosylation site in the RBD based vaccine candidate focused the immune response towards other broadly neutralising epitopes on the RBD. We further observed enhanced cross-neutralisation and cross-binding using a DNA-MVA CR19 prime-boost regime, thus demonstrating the superiority of the glycan engineered RBD vaccine candidate across two platforms and a promising candidate as a broad variant booster vaccine.

Keywords: SARS-CoV-2 antibody; glycan masking; neutralising antibodies; pseudotype neutralisation; receptor binding domain (RBD).

Figure 1

Rational immunogen design of glycan engineered SARS CoV-2 RBD mutants. (A) Mesh surface representation of SARS CoV-2 RBD protein. The epitope recognised by B38 is represented in red-brown solid surface, CR3022 in yellow solid surface, and S309 in grey solid surface. (B) The glycan sites in wild-type SARS-CoV-2 RBD, M7 and M8 designs are highlighted by the stick representation of sugar molecule. (C) Surface representation of wild-type SARS-CoV-2 and glycan engineered M7 and M8 with glycans shown as green spheres. (D) Multiple sequence alignment of RBD WT with M7, M8, and VOCs.

Figure 2

SARS CoV-2 RBD DNA-based vaccine candidates induce humoral immune response in Balb/c mice. (A) Expression analysis of DNA based vaccine candidates encoding glycan engineered SARS CoV-2 RBD mutants by Western blot of transfected HEK293T cells harvested after 48 h. Size in kilodaltons (kDa) and size of the molecular weight marker are indicated. The “DNA empty” lane refers to cells transfected with the same vector but missing the gene insert, the “Cells” lane represents untransfected cells.(B) Immunization schedule of BALB/c mice (n=6) vaccinated four times with DNA-based vaccines encoding SARS CoV-2 RBD WT, M7 and M8, respectively. Bleeds were taken at two-week intervals (light right symbol) and bleeds at week 2 and 8 (dark red symbol) were anlaysed for humoral repsonse. Binding antibody titers (C) represented as area under the curve (AUC) and neutralizing antibodies (D) shown as logIC50 values against SARS CoV-2 two weeks after the first DNA immuniztation. (E) Ratio of log IC50 and log (AUC) was calculated to capture the proportion of neutralising antibodies to binding antibodies 2 weeks after the first DNA immunization. Binding (F) and neutralizing antibodies (G) induced after 4 immunisations with the respective SARS CoV-2 RBD DNA vaccine candidate. (H) Calculated ratio of log IC50 and log(AUC) at week 8 after 4 DNA immunizations. The Mann-Whitney statistical test was applied (*p < 0.05; **p < 0.005 as asteriks or ns for non-significant).

Figure 3

Construction and biochemical characterisation of recombinant MVAs encoding for SARS CoV-2 RBD WT and SARS CoV-2 RBD M7 antigens. (A) Schematic representation of the MVA genome and design of the recombinant SARS CoV-2 RBD WT and SARS CoV-2 RBD M7 MVAs. The J2R region or TK locus was used to insert the antigens for SARS CoV-2 RBD WT and SARS CoV-2 RBD M7 via homologous recombination. (B) Western blot analysis of recombinant MVAs encoding SARS CoV-2 RBD WT and SARS CoV-2 M7 RBD using HEK293T cell lysates infected with a MOI of 2.0 and harvested after 24 h. Size in kilodaltons (kDa) and corresponding bands of the protein standard are indicated. The “MVA CR19 empty” lane represents cells infected with the same MVA construct, but lacking the antigen gene in the backbone, the “Cells” lane represents uninfected cells.

Figure 4

DNA/MVA superior to DNA/DNA regimen in induction of binding and neutralising antibodies against VOCs. (A) Immunization schedule of BALB/c mice vaccinated with either DNA/DNA or DNA/MVA regimen. For the DNA immunization, mice (n=6) received 50 µg DNA vaccine subcutaneuosly, whereas for the MVA boost mice were immunized either with SARS CoV-2 RBD WT or SARS CoV-2 RBD M7 MVA at a dose of 2x10⁷ pfu. Bleeds were taken at two-week intervals (light red symbol) and the terminal bleed at week 11 (dark red symbol) was analysed. (B) Binding antibody titers against VOCs represented as AUC values measured in mice sera from week 11. (C) Neutralization titers shown as logIC50 against VOCs in mice sera collected at week 11.

Figure 5

Challenge in human ACE2 transduced mice with SARS-CoV-2 wildtype virus. (A) Immunization schedule of BALB/c mice vaccinated (n=12) using different DNA prime/MVA boost regimen followed by a challenge with SARS-CoV-2 live virus. For challenge, BALB/c mice were transduced with 1x10⁷ pfu of the ad5-huACE2 vector five days before infection with 1x10⁴ pfu Australia/VIC01/2020 SARS-CoV-2 virus. Bleeds were taken in two-week intervals (light red symbol) until week 8 and then again at week 16, 18/19 and 6 days post challenge (6 d.p.c). Bleeds at week 4, 8 and 6 d.p.c were analysed (dark red symbol). Binding antibody titers (AUC) (B) and neutralising antibodies (logIC50) (C) against SARS-CoV-2 were analyzed 4 weeks after prime and boost respectively and 6 days post challenge (d.p.c). (D) Neutralisation titers (logIC50) against all circulating VOCs to date analyzed in mouse sera obtained after 6 d.p.c. (E) SARS-CoV-2 genome copies from the lungs of infected mice at day 3 (D3) and day 6 (D6) post infection shown as log10 copies/gram of lung. (F–H) Correlation of binding (AUC) and neutralising (logIC50) antibody titer versus lung titre.