Engineering antigens to direct immune responses is a promising methodology that has the potential to address shortcomings in vaccinology. Cellular and humoral immune responses tend to bias towards specific antigens and particular epitopes, in a phenomenon known as immunodominance. During the SARS-CoV-2 pandemic, the immunodominance of key viral antigens (such as the spike protein) were shown to play a role in the emergence of highly virulent variants that drove significant increases in morbidity and mortality.
The receptor binding domain (RBD), is an immunodominant region within the spike protein that mediates contact with hACE2 receptors, thus making it vulnerable to antibody neutralisation. The epitopes of the RBD are organised into classes, ranging from I – V. Class I & II epitopes are designated contact points for ACE2 and are mutational hotspots. Emergent strains of SARS-CoV-2 have demonstrated significant mutability within these regions, helping to facilitate escape from prior established immunity. Therefore, directing immune responses away from highly mutable epitopes towards more conserved regions of the spike and RBD, was a promising avenue to developing more variant-resistant vaccines.
Skewing of immune responses was achieved by masking immunodominant regions (Class I/II epitopes) through glycan obstruction. Glycans elicit immunosuppressive effects on regions of protein they are conjugated to and cover, a strategy utilised by several pathogenic viruses such as SARS-CoV-2 & HIV.
Membrane-bound RBD constructs were profiled for N-linked glycosylation motifs and a selection of single and double glycan motif knock-ins were designed (n=15). Designed RBD constructs were transfected into HEK-293 cells in vitro and then analysed through flow cytometry against a panel of class I, III and V binding antibodies. RBD constructs that demonstrated significant knockdowns in class I binding, but no effect on class III or V binding were determined to have effective glycan shielding of class I/II sites and viable protein expression.
Successful glycan RBD variants (n=2) were then transcribed into mRNA for encapsulation in Lipid Nanoparticles (LNP) formulation for in vivo immunisation. BALB/c mice (n=5) were immunised with 4ug of either WT-RBD or Glyc-RBD mRNA-LNP formulations via IM injections on days 0 and 21, with serum responses measured at day 35.
Immunised mice were sacrificed and spleens were harvested for lymphocyte immunophenotyping and antigen-positive B-cell quantification via flow cytometry. Antigen-positive lymphocytes will be stained with labelled ACE2-Fc to ascertain the level of class I skewing achieved by glycan blockade.