Self-Amplifying RNA Vaccines Give Equivalent Protection against Influenza to mRNA Vaccines but at Much Lower Doses

Abstract

New vaccine platforms are needed to address the time gap between pathogen emergence and vaccine licensure. RNA-based vaccines are an attractive candidate for this role: they are safe, are produced cell free, and can be rapidly generated in response to pathogen emergence. Two RNA vaccine platforms are available: synthetic mRNA molecules encoding only the antigen of interest and self-amplifying RNA (sa-RNA). sa-RNA is virally derived and encodes both the antigen of interest and proteins enabling RNA vaccine replication. Both platforms have been shown to induce an immune response, but it is not clear which approach is optimal. In the current studies, we compared synthetic mRNA and sa-RNA expressing influenza virus hemagglutinin. Both platforms were protective, but equivalent levels of protection were achieved using 1.25 μg sa-RNA compared to 80 μg mRNA (64-fold less material). Having determined that sa-RNA was more effective than mRNA, we tested hemagglutinin from three strains of influenza H1N1, H3N2 (X31), and B (Massachusetts) as sa-RNA vaccines, and all protected against challenge infection. When sa-RNA was combined in a trivalent formulation, it protected against sequential H1N1 and H3N2 challenges. From this we conclude that sa-RNA is a promising platform for vaccines against viral diseases.

Keywords: DNA; H1N1; RNA; alphavirus; influenza; replicon; trivalent; vaccine.

Graphical abstract

Figure 1

Different mRNA Vaccine Platforms Are Both Protective against Influenza A Disease in a Prime-Boost Regime, but IVT-mRNA Requires More Material BALB/c mice were immunized i.m. with 120, 80, or 20 μg H1N1/PR8-HA coding mRNA, with 5 μg of inactivated virus (def-Virus) or ringer-lactate solution only (buffer), followed by a homologous boost 3 weeks later. H1N1-specific antibody was measured by HAI (A) and VNT (B) 8 weeks after the first vaccine was administered. Animals were infected i.n. with 10-fold MLD₅₀ of H1N1/PR8. Survival (C) and weight change (D) were monitored daily. BALB/c mice were immunized i.m. with 1.25, 0.25, or 0.05 μg H1N1/PR8-HA coding sa-RNA followed by a homologous boost 3 weeks later. H1N1-specific antibody was measured by HAI (E) and VNT (F) 8 weeks after the first vaccine was administered. Thereafter, animals were infected i.n. with 10-fold MLD₅₀ of H1N1/PR8. Survival (G) and weight change (H) were monitored daily. Lines and points represent means and SEM of n = 5 mice. *p < 0.05 and **p < 0.001 indicate significance measured by one-way ANOVA.

Figure 2

In Vivo Imaging of Luciferase Encoded by mRNA and Self-Amplifying RNA BALB/c mice were intramuscularly injected with 4 μg sa-RNA (2 μg per leg) or synthetic mRNA encoding luciferase genes in PBS. At various time points after inoculation, expression was visualized using an IVIS spectrum in vivo imaging system after intraperitoneal injection of D-luciferin. One representative image is shown per time point (A). Luciferase levels from n = 6 animals were quantified as relative light units (B). Points represent means ± SEM.

Figure 3

Formulating sa-RNA with PEI Significantly Increases the Antibody Response BALB/c mice were immunized twice, on days 0 and 21, with 1.25 μg PEI-formulated sa-RNA encoding HA or sa-RNA encoding HA alone. Sera was collected at days 19 (A) and 54 (B) and analyzed for influenza virus neutralization. Responses were compared to animals immunized with buffer alone. Points represent individual animals, and lines represent mean of n = 8 animals.

Figure 4

Self-Amplifying RNA Vaccines Are Protective against Seasonal H1N1 and B Influenza Disease and Reduce Viral Load in a Prime-Boost Regime BALB/c mice were i.m. immunized intramuscularly in a prime boost regime with a 3-week interval (indicated by arrows) with 1.5 or 0.5 μg Cal’09 H1N1 HA sa-RNA (A–F), Flu B-Mass (G–I), or X31 H3N2 (J–L). Responses were compared to 1.5 μg HIV gp140 sa-RNA (negative control) or 1.8 μg licensed protein flu vaccine (A–I) or naive animals (J–L). (A) H1N1-specific IgG was measured after vaccination. At 7 weeks, mice were infected intranasally with Cal’09 H1N1 influenza. Weight change was monitored daily (B), and influenza M gene copy number was measured in the lung (C). H1N1-specific total IgG (D) and the ratio of specific IgG2a:IgG1 was measured in serum 4 days after infection (E). (F) H1-specific CD8⁺ T cells were measured in lung tissue on day 7 of infection. (G) For Flu B-Mass-immunized animals, specific IgG was measured by ELISA. (H) At 7 weeks, mice were infected i.n. with B/Florida/06 influenza, and weight change was monitored daily. (I) Influenza B NS gene copy number was measured in the lung. (J) For H3N2-immunized animals, specific IgG was measured by ELISA. (K) At 7 weeks, mice were infected i.n. with X31 H3N2 influenza, and weight change was monitored daily. (L) Influenza A M gene copy number was measured in the lung. Lines and points represent mean of n ≥ 4 mice. *p < 0.05, **p < 0.01, ***p < 0.001 between 1.5 μg flu RNA and negative control; +p < 0.05, ++p < 0.01, +++p < 0.001 between 0.5 μg flu RNA and negative control; and ##p < 0.01, ###p < 0.001 between protein vaccine and negative control.

Figure 5

Self-Amplifying RNA Vaccines Are Immunogenic and Protective against H1N1 in Trivalent Combination BALB/c mice were primed i.m. with 1.5 μg each of Cal’09 H1N1, B-Mass, X31 H3N2 HA sa-RNA, 1.5 μg Cal’09 H1N1 sa-RNA alone, or 1.8 μg licensed protein flu vaccine, followed by a homologous boost 3 weeks later. H1N1 (A), H3N2 (B), or Flu B (C) specific antibody was measured by ELISA in sera 7 days after infection. (D) At 7 weeks, mice were infected i.n. with Cal’09 H1N1 influenza, and weight change was monitored daily. (E) 7 days later, the Cal’09 RNA and trivalent RNA groups from the same study were challenged with X31 H3N2 influenza, and responses were compared to new naive controls. (A)–(C) points represent individual animals and lines represent mean. (D) and (E) points represent the mean of n = 5 animals ± SEM. ***p < 0.001 between trivalent sa-RNA and naive; ###p < 0.001 between monovalent sa-RNA and naive; and xxx between monovalent and trivalent RNA (E).

Figure 6

A Single Dose of Self-Amplifying RNA Vaccine Gives Equivalent Protection to Electroporated DNA and Greater Protection than mRNA Encoding the Same Gene (A) BALB/c mice were primed i.m. with 1.5 μg Cal’09 H1N1 as DNA or self-amplifying RNA. RNA was delivered as a formulation; DNA was delivered as a formulation or naked with electroporation. 4 weeks later, mice were infected i.n. with Cal’09 H1N1 influenza, and weight change was monitored daily. (B) M gene copy number was measured in lungs 7 days after infection. 4 days post-infection, H1N1-specific total IgG was measured in serum (C), and the ratio of specific IgG2a:IgG1 determined (D). (E) 7 days post-infection, proportions of flu-specific CD8⁺ T cells were measured in lung tissue by pentamer staining. (A) points represent the mean of n = 5 animals ± SEM. (B)–(E) points represent individual animals, and lines represent mean. *p < 0.05; **p < 0.001, and ***p < 0.001 indicate significance measured by one-way ANOVA.