Iterative structure-based improvement of a fusion-glycoprotein vaccine against RSV

Abstract

Structure-based design of vaccines, particularly the iterative optimization used so successfully in the structure-based design of drugs, has been a long-sought goal. We previously developed a first-generation vaccine antigen called DS-Cav1, comprising a prefusion-stabilized form of the fusion (F) glycoprotein, which elicits high-titer protective responses against respiratory syncytial virus (RSV) in mice and macaques. Here we report the improvement of DS-Cav1 through iterative cycles of structure-based design that significantly increased the titer of RSV-protective responses. The resultant second-generation 'DS2'-stabilized immunogens have their F subunits genetically linked, their fusion peptides deleted and their interprotomer movements stabilized by an additional disulfide bond. These DS2 immunogens are promising vaccine candidates with superior attributes, such as their lack of a requirement for furin cleavage and their increased antigenic stability against heat inactivation. The iterative structure-based improvement described here may have utility in the optimization of other vaccine antigens.

Figure 1

Design, structure, immunogenicity, and informatics of pre-fusion stabilized single-chain RSV F glycoproteins from design cycle 1. (a) RSV F DS-Cav1 structure in ribbon with F₂ domain in dark gray and F₁ domain in light gray. Close-up of the cleavage site of a single protomer with the fusion peptide colored orange. (b) Single-chain RSV F constructs were designed by replacing both of the furin cleavage sites (R109/E110 and R136/F137) and the fusion peptide with a variable-length linker replacing either residues 97–151 or residues 105–145. DS-Cav1 mutations (DS: disulfide between residues 155 and 290; Cav1: Ser to Phe mutation at residue 190 and Val to Leu at residue 207) were incorporated into all designs with other mutations labeled and indicated by vertical black lines. (c) RSV F sc9 DS-Cav1 structure in ribbon and overlaid with the DS-Cav1 structure (gray). Residues removed in the single chain design or that differed by rmsd > 5 Å highlighted in red. (d) Details of the inter-subunit linker between F₂ and F₁ (electron density (2F_o-F_c) is shown at 1σ in blue). (e) Serum neutralization titers from mice immunized with RSV F DS-Cav1 single chain variants with optimized linker (blue symbols) and Postfusion F and RSV F DS-Cav1 as controls (gray symbols). Titers from each mouse are shown, with geometric means indicated by horizontal red lines (values shown at the bottom). Each group contained 10 mice except the DS-Cav1 group which included 29 mice. P values <0.0001 (****), <0.01 (**) or <0.05 (*) were determined by two-tailed Mann-Whitney tests. (f) Statistical analysis of design cycle 1. Left: spearman correlation of antigenic and physical properties with neutralization titers for design cycle 1 immunogens (including DS-Cav1). Right: correlation of elution volume with neutralization titer. Pearson correlation for single-chain constructs denoted in green.

Figure 2

Design, properties, structure, immunogenicity, and informatics of single-chain RSV F with altered F₂-F₁ linkers from design cycle 2. (a) Design of altered F₂-F₁ linkers. (b)Representative dynamic light scattering profiles and gel filtration chromatograms of single-chain RSV F glycoprotein variants with altered linkers. A representative gel filtration profile for each protein is shown as assessed on a Superdex-200 16/60 column (120 ml column volume). sc9-10 DS-Cav1 exhibited similar elution profiles characteristic as DS-Cav1. (c) Structures of single-chain RSV F glycoproteins with optimized F₂-F₁ linkers. (d) Structural details of the F₂-F₁ linker with the linker region in red, mutations in black, and electron density in blue. (e) Serum neutralization titers from mice immunized with RSV F DS-Cav1 single chain variants with optimized linker (blue symbols) and Postfusion F and RSV F DS-Cav1 as controls (gray symbols). Immunization titers from sc9 DS-Cav1 (green symbols) are shown for reference. Titers from each mouse are shown, with geometric means indicated by horizontal red lines (values shown at the bottom). Each group contained 10 mice except the sc9-10 DS-Cav1 (20 mice), sc9-19 DS-Cav1 (19 mice), sc9-24 DS-Cav1 (20 mice) and DS-Cav1 (29 mice) groups. P values <0.0001 (****) or <0.001 (***) were determined by two-tailed Mann-Whitney tests. (f)Statistical analysis of design cycles 1–2. Left: spearman correlation of antigenic and physical properties with neutralization titers for variants from design cycles 1 and 2 (including DS-Cav1). Correlations with adjusted P value (Bonferroni correction) of less than 0.05 were marked with “*”. Right: correlation of quaternary antibody antigenicity with neutralization titer. “10,000” was used for antigenicity with “N.B.” values.

Figure 3

Design, structure, immunogenicity, and informatics of single-chain RSV F glycoproteins with interprotomer disulfides (DS2) from design cycle 3. (a) Crystal structure of RSV sc9-10 DS-Cav1 trimer in ribbon. Insets show close-ups of Cycle 3 designs for residues mutated to Cys to form interprotomer disulfides (mutations labeled and in stick representation). (b) Engineered single-chain RSV F glycoproteins with interprotomer disulfides by SDS-PAGE (uncropped form is shown in Supplementary Fig. 1c). (c) Negative-stain EM class-averaged images of RSV F DS2 variants. (d) Structure of single-chain RSV F glycoproteins sc9-10 with interprotomer disulfide A149C Y458C shown in close-up with electron density. (e) Serum neutralization titers of 10 mice immunized with sc9-10 DS-Cav1 A149C Y458C and 15 mice immunized with sc9-10 DS-Cav1 N183GC N428C. Titers from each mouse are shown as individual purple symbols, with geometric means indicated by horizontal red lines (values shown at the bottom). Sera titers from 20 mice immunized with sc9-10 DS-Cav1 (blue symbols) and 39 mice immunized with RSV F DS-Cav1 (gray symbols) are shown for reference. P values <0.0001 (****) or <0.001 (***) were determined by two-tailed Mann-Whitney tests. (f)Statistical analysis of design cycles 1–3. Left: spearman correlation of each physical property with neutralization titers for design cycle 1, 2, and 3 variants (including DS-Cav1). Correlations with adjusted P value (Bonferroni correction) of less than 0.05 were marked with “*”. Right: correlation of quaternary antibody antigenicity and physical stability with neutralization titer. Constructs containing a mutation at residue 183 were not included in the correlation for quaternary antibody antigenicity as N183 is part of the AM14 epitope. “10,000” was used for antigenicity with “N.B.” values.

Figure 4

Design, structure, immunogenicity, and informatics from design cycle 4 comprising combinations of interprotomer disulfides (DS2) and other mutations. (a) Design cycle 4 involving interprotomer disulfides combined with other variants highlighted here in orange and detailed in Table S1. (b) Phylogenetic tree of strains tested. (c) Structures of single-chain RSV F glycoproteins sc9-10 DS-Cav1 A149C Y458C S46G E92D S215P K465Q with details of surface mutations shown in insets. (d) Serum neutralization titers from mice immunized with RSV F DS-Cav1 single-chain variants (n=10). Titers from each mouse are shown as individual orange symbols, with geometric means indicated by horizontal red lines. Titers from mice immunized with designs based on sc9-10 DS-Cav1 A149C Y458C are shown with differently formatted circle symbols. Designs based on sc-9-10 DS-Cav1 N183GC N428C are shown with orange square symbols, and designs based on sc9-10 DS-Cav1 Q98C Q361C with orange triangle symbols. These latter two designs were based on the RSV CH18537 and Long strains. Sera titers from 39 mice immunized with RSV F DS-Cav1 (gray symbols) are shown for reference. P values <0.0001 (****), <0.01 (**) or <0.05 (*) were determined by two-tailed Mann-Whitney tests. (e) Statistical analysis of design cycles 1–4. Left: spearman correlation of antigenic and physical properties with neutralization titers for variants from design cycles 1–4 (including DS-Cav1). Correlations with adjusted P value (Bonferroni correction) of less than 0.05 are marked with “*”. Right: correlation of quaternary antibody antigenicity and physical stability with neutralization titer. Constructs containing mutations at residue 183 were not included in the correlation for quaternary antibody antigenicity as N183 is part of the AM14 epitope. “10,000” was used for antigenicity with “N.B.” values.