Clade 2.3.4.4b but not historical clade 1 HA replicating RNA vaccine protects against bovine H5N1 challenge in mice

Abstract

The ongoing circulation of influenza A H5N1 in the United States has raised concerns of a pandemic caused by highly pathogenic avian influenza. Although the United States has stockpiled and is prepared to produce millions of vaccine doses to address an H5N1 pandemic, currently circulating H5N1 viruses contain multiple mutations within the immunodominant head domain of hemagglutinin (HA) compared to the antigens used in stockpiled vaccines. It is unclear if these stockpiled vaccines will need to be updated to match the contemporary H5N1 strains. Here we show that a replicating RNA vaccine expressing the HA of an H5N1 isolated from a US dairy cow confers complete protection against homologous lethal challenge in mice. A repRNA encoding the HA of a clade 1 H5 from 2004 (A/Vietnam/1203/2004) as utilized by some stockpiled vaccines, confers only partial protection. Our data highlight the utility of nucleic acid vaccines to be rapidly updated to match emergent viruses of concern while demonstrating that contemporary bovine H5N1 viruses can evade immunity elicited by historical HA antigens.

Fig. 1. Mutations within the A/bovine HA may impact neutralizing antibodies and vaccine efficacy.

The amino acid sequence homology between A/Vietnam and A/bovine was mapped onto the HA trimer structure (PDB 6CFG). Two of the protomers are displayed in the surface representation, and 1 protomer is displayed in ribbon representation. Cyan indicates regions with differences in sequence and areas colored white indicate conserved sequence homology. Fabs from well characterized cross-neutralizing antibodies (nAbs) that target three sites of vulnerability on the HA1 domain of the HA trimer are shown (PDB 4MHJ, 3GBN, and 4HG4 are aligned). nAb 2G1 targets the region including the receptor binding site (RBS) and is colored in purple, nAb H5M9 binds the VE domain (gray), and nAb FLUA-20 binds an occluded epitope on prefusion HA along the HA1 interface that can be exposed during reversible conformational changes (yellow).

Fig. 2. RepHA-Bovine is immunogenic and protective against lethal A/bovine challenge in mice.

a Mice (n = 4 per group) were immunized with indicated repRNAs and HA-binding antibodies measured against indicated recombinant HAs (a) or NP (b). c Viral neutralization (VN) titers of pooled sera (n = 4 per group) were measured against infectious A/Vietnam or A/bovine virus and data shown as technical replicates. Splenocytes were evaluated for HA-specific T cells by IFNγ ELISpot. d Splenocytes were stimulated with overlapping peptides spanning the A/Vietnam HA in five pools and data shown as the cumulative spot count from all five peptide pools. e, f Mice (n = 6 per group) were challenged with 100,000 TCID₅₀ of A/bovine via the intranasal route and weighed daily (d) and euthanized when they reached >20% weight loss or exhibited signs such as severe lethargy or neurological signs. Box plots indicate mean plus standard deviation. a, b P values calculated with an ordinary one-way ANOVA with Tukey’s multiple comparisons test. d P values calculated with an ordinary one-way ANOVA with Dunnett’s multiple comparisons test to sham-vaccinated mice. e P values calculated with log-rank test with Bonferroni correction for multiple comparisons. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Fig. 3. RepHA-Vietnam and repHA-Bovine elicit antibodies with distinct homosubtypic reactivity.

Luminex microspheres were coated with the indicated recombinant HAs. H5 clade is indicated in parentheses. Antibody binding to recombinant HAs was measured in serum from sham or repHA-Vietnam or repHA-Bovine vaccinated mice (n = 9) (a, b). Data in (b) is duplicated from (a) but shown in heatmap format. N = 9 per group. As comparison, pooled human serum from humans receiving an experimental vaccine based on Influenza A/Indonesia/05/2005 (H5N1) was also evaluated. Data shown as median fluorescence intensity (MFI). Microspheres coated with protein G served as an internal positive control while microspheres coated with human serum albumin (HSA) served as an internal negative control. c, d Binding antibodies from serum from mice that survived challenge with indicated viruses was evaluated as in (a) and data in (d) is duplicated from (c) but shown in heatmap format. As comparison, sera from a convalescent human infected with H1N1 in 2009 was included. a, c P values calculated with a two-way ANOVA with Tukey’s multiple comparisons test. a Only comparisons between repHA-Vietnam and repHA-Bovine mouse sera with P values < 0.05 are shown for clarity. All comparisons between sham and repHA-Vietnam or repHA-Bovine vaccinated mouse sera was significant, P < 0.01 except for HSA and Hong Kong 2007 which were non-significant. c Only comparisons between mice and with P values of <0.05 are shown, all other comparisons were P > 0.05. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.