A self-amplifying RNA vaccine against COVID-19 with long-term room-temperature stability

Abstract

mRNA vaccines were the first to be authorized for use against SARS-CoV-2 and have since demonstrated high efficacy against serious illness and death. However, limitations in these vaccines have been recognized due to their requirement for cold storage, short durability of protection, and lack of access in low-resource regions. We have developed an easily-manufactured, potent self-amplifying RNA (saRNA) vaccine against SARS-CoV-2 that is stable at room temperature. This saRNA vaccine is formulated with a nanostructured lipid carrier (NLC), providing stability, ease of manufacturing, and protection against degradation. In preclinical studies, this saRNA/NLC vaccine induced strong humoral immunity, as demonstrated by high pseudovirus neutralization titers to the Alpha, Beta, and Delta variants of concern and induction of bone marrow-resident antibody-secreting cells. Robust Th1-biased T-cell responses were also observed after prime or homologous prime-boost in mice. Notably, the saRNA/NLC platform demonstrated thermostability when stored lyophilized at room temperature for at least 6 months and at refrigerated temperatures for at least 10 months. Taken together, this saRNA delivered by NLC represents a potential improvement in RNA technology that could allow wider access to RNA vaccines for the current COVID-19 and future pandemics.

Fig. 1. Design and characterization of SARS-CoV-2 D614G saRNA/NLC vaccine.

a SARS-CoV-2 saRNA vaccine schematic. b RNA/nanostructured lipid carrier (NLC) vaccine formulation. Design by Cassandra Baden. c Size distribution of NLC formulation particles alone (black) and saRNA/NLC complex (pink) showing mean and standard deviation of n = 3 replicate measurements for each sample. d SARS-CoV-2 saRNA complexed to the outside of the NLC particles is full-length intact saRNA that is protected by the NLC complexation from RNase degradation. Lanes were derived from the same gel and re-arranged (see Supplementary Fig. 7). e Western blot showing SARS-CoV-2 spike (S) protein expression in saRNA/NLC vaccine-transfected HEK-293 cells. Blot was derived from the same experiment and processed in parallel. See Supplementary Fig. 7 for the original unprocessed blot image.

Fig. 2. Immunogenicity of SARS-CoV-2 D614G saRNA/NLC vaccine after prime or prime-boost immunization of C57BL/6J mice.

a Serum SARS-CoV-2 spike protein-binding IgG. Horizontal lines show geometric mean. b Serum SARS-CoV-2 neutralizing antibodies against the original Wuhan strain. Results show geometric mean and 95% confidence interval. c Serum SARS-CoV-2 spike protein-binding IgG1 versus IgG2a indicates a strong Th1-biased response. Horizontal lines show median. The vector control represents mice injected with 10 μg of NLC-complexed saRNA expressing the non-immunogenic secreted embryonic alkaline phosphatase gene. n = 10 mice per group.

Fig. 3. Immunogenicity of SARS-CoV-2 D614G, D614G-2P, and D614G-2P-3Q saRNA/NLC vaccines after 10 µg prime or prime-boost immunization of C57BL/6J mice.

a Serum SARS-CoV-2 spike protein-binding IgG. Horizontal lines show geometric mean. b–d Serum SARS-CoV-2 neutralizing antibody titers against the original Wuhan, Alpha (B.1.1.7), and Beta (B.1.351) pseudovirus variants. The vector control represents mice injected with 10 μg of NLC-complexed saRNA expressing the non-immunogenic secreted embryonic alkaline phosphatase gene. Results show geometric mean and geometric standard deviation. n = 10 mice per group.

Fig. 4. Humoral immunogenicity profiles of optimized SARS-CoV-2 D614G-2P-3Q saRNA/NLC vaccine (AAHI-SC2) after prime or prime-boost immunization of C57BL/6 J mice.

a Serum SARS-CoV-2 spike protein-binding IgG. Horizontal lines show geometric mean. Serum SARS-CoV-2 neutralizing antibody titers post-prime (b) and post-boost (c). Horizontal lines show geometric mean. Data were log-transformed and evaluated by mixed effects analysis with multiple comparisons. Induction of bone marrow (BM)-resident IgA- (d) and IgG-secreting (e) cells by ELISpot. Results show geometric mean and geometric standard deviation. Analyzed with one-way ANOVA with multiple comparisons. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. The vector control represents mice injected with 10 μg of NLC-complexed saRNA expressing the non-immunogenic secreted embryonic alkaline phosphatase (SEAP) gene. n = 6 mice for SEAP control group and n = 8 for dosing groups.

Fig. 5. Detailed cellular immunogenicity profiles of optimized SARS-CoV-2 D614G-2P-3Q saRNA/NLC vaccine (AAHI-SC2) after prime or prime-boost immunization of C57BL/6 J mice.

a SARS-CoV-2 spike-reactive spleen-resident IFNγ⁺ T cells by ELISpot. SFU = spot forming units. Results show mean and standard deviation (SD). b SARS-CoV-2 spike-reactive spleen-resident IL-5⁺ T cells by ELISpot. Results shows mean and SD. c SARS-CoV-2 spike-reactive spleen-resident IL-17-A⁺ T cells by ELISpot. Results shows mean and SD. CD4 (d) and CD8 (e) T cells responding with any Th1 (IFNγ, IL-2, or TNFα) or Th2 (IL-5 or IL-10) cytokines post-prime and post-boost were plotted representing the total number of responding cells, for CD4 or CD8 cells per mouse. Horizontal lines show median. f, g Quality of responding CD4 and CD8 T cells. Total Th1 or Th2 responses were subdivided by cells responding with one, two, or three cytokine(s) to show magnitude and quality of Th1 and Th2 responses, with bar graphs showing the average percentage across each group of mice. The vector control represents mice injected with 10 μg of NLC-complexed saRNA expressing the non-immunogenic secreted embryonic alkaline phosphatase (SEAP) gene. Results show mean and SD. n = 6 mice for SEAP control group and n = 8 for dosing groups.

Fig. 6. Stability of the lyophilized SARS-CoV-2 saRNA/NLC vaccine (AAHI-SC2) after 10 months of storage at different temperatures.

a Mean hydrodynamic (Z-average) diameter of the vaccine complex with error bars indicating the standard deviation (SD) of n = 3 replicate measurements for each condition at each timepoint. b Integrity of vaccine RNA at 0 and 10 months of storage at the indicated temperatures. “Uncomplexed” refers to saRNA alone that had been stored at −80 °C for the indicated length of time. Lanes were derived from the same gel and re-arranged for the 0 months of storage image. See Supplementary Fig. 8 for original unprocessed gel images. c Serum SARS-CoV-2 spike protein-binding IgG induced in female C57BL/6J mice by the vaccine after storage under the indicated conditions for the indicated times. n = 5 mice per condition per timepoint. Antibody plots show geometric mean and geometric SD. Statistical analysis conducted on log-transformed data by Welch’s ANOVA test with Dunnett’s T3 multiple comparison test at each timepoint, comparing each stored vaccine preparation to freshly complexed vaccine. *p < 0.05, **p < 0.01, ****p < 0.0001. d Pseudovirus neutralization titers induced in female C57BL/6 J mice by the vaccine after storage under the indicated conditions for the indicated times. LOD = limit of detection. n = 5 mice per condition per timepoint. Results show geometric mean and geometric SD. Statistical analysis conducted on log-transformed data by Welch’s ANOVA test with Dunnett’s T3 multiple comparison test at each timepoint, comparing each stored vaccine preparation to freshly complexed vaccine.