Structure-Based Design of Nipah Virus Vaccines: A Generalizable Approach to Paramyxovirus Immunogen Development

Abstract

Licensed vaccines or therapeutics are rarely available for pathogens with epidemic or pandemic potential. Developing interventions for specific pathogens and defining generalizable approaches for related pathogens is a global priority and inherent to the UN Sustainable Development Goals. Nipah virus (NiV) poses a significant epidemic threat, and zoonotic transmission from bats-to-humans with high fatality rates occurs almost annually. Human-to-human transmission of NiV has been documented in recent outbreaks leading public health officials and government agencies to declare an urgent need for effective vaccines and therapeutics. Here, we evaluate NiV vaccine antigen design options including the fusion glycoprotein (F) and the major attachment glycoprotein (G). A stabilized prefusion F (pre-F), multimeric G constructs, and chimeric proteins containing both pre-F and G were developed as protein subunit candidate vaccines. The proteins were evaluated for antigenicity and structural integrity using kinetic binding assays, electron microscopy, and other biophysical properties. Immunogenicity of the vaccine antigens was evaluated in mice. The stabilized pre-F trimer and hexameric G immunogens both induced serum neutralizing activity in mice, while the post-F trimer immunogen did not elicit neutralizing activity. The pre-F trimer covalently linked to three G monomers (pre-F/G) induced potent neutralizing antibody activity, elicited responses to the greatest diversity of antigenic sites, and is the lead candidate for clinical development. The specific stabilizing mutations and immunogen designs utilized for NiV were successfully applied to other henipaviruses, supporting the concept of identifying generalizable solutions for prototype pathogens as an approach to pandemic preparedness.

Keywords: G attachment protein; Nipah virus; pandemic preparedness; pre-F/G chimeric immunogen; stabilized prefusion F; structure-based vaccine design.

FIGURE 1

Structure-based design of a stabilized prefusion NiV F trimer. (A) Structure of prefusion NiV F glycoprotein trimer (PDB ID 5EVM) in green, orange and sand, and with residues that undergo >5 Å conformational change to transition to the postfusion conformation shown in black. GCN4 trimerization (TD) motif is shown in magenta, linked to NiV F residue 488. Zoom insets highlight mutated residues (red) to stabilize the prefusion F structure including L104C-I114C disulfide bond, a S191P helix-breaking proline substitution and a L172F cavity-filling mutation. (B) Two-dimensional class averages of the prefusion F trimer (left), postfusion F trimer (middle) and prefusion F trimer bound to h5B3 Fab (right) obtained by negative-stain electron microscopy (EM). (C) Binding kinetics were measured using a fortéBio Octet Red384 instrument. Table summarizing binding affinities of F designs to monoclonal antibody h5B3. (D) Non-reduced and reduced SDS-PAGE analysis of prefusion and postfusion NiV F glycoproteins.

FIGURE 2

Structure-based design of NiV G immunogens and NiV F/G chimeric immunogens. (A) Two-dimensional class averages of NiV G head domain multimer designs (left: trimeric G, center: hexameric G, right: ferritin-G) obtained by negative-stain EM. (B) Negative-stain EM analysis of NiV F/G chimeras, showing pre-F/G (left), G/pre-F (center), and post-F/G (right). TD, trimerization domain.

FIGURE 3

Thermodynamic and colloidal stability assessment of Pre-F, Post-F, Hex G, and Pre-F/G constructs. (A) Differential scanning calorimetry; based on the T_m (transition midpoint), the thermodynamic stability of the individual domains can be ranked Post-F>> Hex G > Pre-F, and Pre-F/G. (B) Dynamic light scattering indicates a similar degree of colloidal stability for Hex G, Pre-F and Pre-F/G, while Post-F is extremely colloidally stable, as indicated by lack of change in size within the temperature range of the experiment. (C) Table summarizing the conformational transitions (A) and colloidal stability (B) of Pre-F, Post-F, Hex G, and Pre-F/G.

FIGURE 4

Immunogenicity of NiV Pre-F stabilized immunogens, multimeric forms of G and Pre-F/G chimeric immunogens. (A,B) Recognition of pre-F (A) or post-F (B) NiV F proteins by sera from mice immunized twice with NiV F designs or unimmunized. (C,D) Recognition of pre-F NiV F (C) or mono G (D) protein by sera from mice immunized twice with NiV F, NiV G multimers or NiV F/G chimeric designs or unimmunized. (E) Recognition of pre-F NiV F or mono G proteins by sera from mice immunized twice with NiV Hex G or NiV Stalk G. Binding kinetics were measured using a fortéBio Octet Red384 instrument. Line represents mean of all animals in each group ± standard deviation (using GraphPad Prism).

FIGURE 5

Neutralization of NiVF/G VSVΔG-luciferase pseudovirus by sera from mice immunized with NiV stabilized pre-F, multimeric forms of G and F/G chimeric immunogens. VSVΔG-luciferase pseudovirus (expresses both NiV F_WT and NiV G on surface) neutralization assays were performed on individual mouse sera collected at week 5. The log₁₀ reciprocal IC₈₀ titer for each sample was calculated by curve fitting and non-linear regression using GraphPad Prism. P-values were calculated using one-way ANOVA with Tukey’s multiple comparisons test (***p < 0.001, ****p < 0.0001). Line represents mean of log₁₀ reciprocal IC₈₀ titer ± standard deviation (using GraphPad Prism).

FIGURE 6

Application of NiV antigen designs to phylogenetically related Henipaviruses. (A,B) Two-dimensional negative-stain EM class averages of NiV designs applied to Hendra (A) and Cedar (B) viruses. (C) Binding kinetics of HeV F pre-F, post-F, and NiV pre-F/HeV G chimeric immunogen binding to h5B3, NiV pre-F-specific antibody. (D) Binding kinetics of multimeric forms of NiV G, HeV G, CedPV G and NiV pre-F/HeV G chimeric immunogen binding to m102.4, HeV G-specific antibody. (E,F) Recognition of HeV pre-F (E) or HeV Mono G (F) proteins by sera from mice immunized twice with NiV F, G or F/G immunogens. Binding kinetics were measured using a fortéBio Octet Red384 instrument. Line represents mean of all animals in each group ± standard deviation (using GraphPad Prism).