Circular RNA vaccines with long-term lymph node-targeting delivery stability after lyophilization induce potent and persistent immune responses

ABSTRACT Several recent attempts to improve the stability and immunogenicity of messenger RNA (mRNA) vaccines include the use of circular RNA (circRNA), targeted delivery, and lyophilization. However, these research directions have often been pursued independently, ignoring the impact of the modification of lipid nanoparticle-encapsulated mRNA vaccines on their targeted delivery after lyophilization and on their subsequent immunogenicity. Here, we develop a circRNA vaccine targeting lymph nodes that expresses rabies virus glycoprotein (G), termed circRNA-G. Mannose modification is introduced directly into the process of synthesizing PEG lipids, and the resulting PEG-mannose can be used in the preparation of mannose-LNPs (mLNPs) that target dendritic cells, thereby promoting the specific distribution of circRNA-G to lymph nodes (mLNP-circRNA-G). We demonstrated that mLNP-circRNA-G has continuous antigen availability that promotes the generation of T follicular helper cells, germinal center B cells, long-lived plasma cells, and memory B cells in mice. Importantly, the vaccine with this targeting modification remained stable for at least 24 weeks of storage at 4℃ after lyophilization, and its immunogenicity was also maintained. Notably, this strategy also enhances the antibody production of the SARS-CoV-2 trimeric receptor-binding domain circRNA vaccine and the stability of immunogenicity after lyophilization. In summary, this study provides a general platform for the design of lyophilized vaccines with targeted stability, demonstrating the potential of lymph node-targeting circRNAs as next-generation vaccines. IMPORTANCE messenger RNA (mRNA) vaccines are a key technology in combating existing and emerging infectious diseases. However, the inherent instability of mRNA and the nonspecificity of lipid nanoparticle-encapsulated (LNP) delivery systems result in the need for cold storage and a relatively short-duration immune response to mRNA vaccines. Herein, we develop a novel vaccine in the form of circRNAs encapsulated in LNPs, and the circular structure of the circRNAs enhances their stability. Lyophilization is considered the most effective method for the long-term preservation of RNA vaccines. However, this process may result in irreversible damage to the nanoparticles, particularly the potential disruption of targeting modifications on LNPs. During the selection of lymph node-targeting ligands, we found that LNPs modified with mannose maintained their physical properties almost unchanged after lyophilization. Additionally, the targeting specificity and immunogenicity remained unaffected. In contrast, even with the addition of cryoprotectants such as sucrose, the physical properties of LNPs were impaired, leading to an obvious decrease in immunogenicity. This may be attributed to the protective role of mannose on the surface of LNPs during lyophilization. Freshly prepared and lyophilized mLNP-circRNA vaccines elicited comparable immune responses in both the rabies virus model and the SARS-CoV-2 model. Our data demonstrated that mLNP-circRNA vaccines elicit robust immune responses while improving stability after lyophilization, with no compromise in tissue targeting specificity. Therefore, mannose-modified LNP-circRNA vaccines represent a promising vaccine design strategy.

Supplemental Table Table S1

Figure S2 .
Figure S2.CircRNA-G vaccines are more stable than mRNA-G vaccines.(A-C) The degradation rate and antigen expression of circRNA-G and mRNA-G were stored at different temperatures.

Figure S7 .
Figure S7.CircRNA-G and LN-targeting delivery immunization induces higher Ab titers.(A-B) The ratio of IgG and nAb titers induced by the LNP-circRNA-G vaccine to that induced by the LNP-mRNA-G vaccine.(C-D) The ratio of IgG and nAb titers induced by the mLNP-circRNA-G vaccine to that induced by the LNP-circRNA-G vaccine.

Figure
Figure S8.A balanced Th1/Th2 immune response was induced by mLNP-circRNA-G.ICR mice were immunized with a single injection of 2 μg of mLNP-circRNA-G or LNP-circRNA-G.Sera were collected week 4 post immunization and assessed by ELISA for RABV G-specific IgG1, IgG2a and IgG2b titers.Titer ratios of IgG2a to IgG1 and IgG2b to IgG1 were calculated.Data are means ± SEMs (n = 10).

Figure
Figure S14.A balanced Th1/Th2 immune response is induced by mLNP-circRNA-RBD.ICR mice were immunized with a single injection of 5 μg of mLNP-circRNA-RBD or LNP-circRNA-RBD.Sera were collected at week 4 after primary and booster immunization and assessed by ELISA for RABV Gspecific IgG1 and IgG2a titers.Titer ratios of IgG2a to IgG1 were calculated.Data are means ± SEMs (n = 5).

Figure S15 .
Figure S15.Safety evaluation of mLNP-circRNA-G in vivo and in vitro.(A) Cytotoxicity analysis of LPP-mRNA-G in HEK-293T cells by MTT assays.One-way ANOVA was used to evaluate intergroup differences.ns, no significant difference; Error bars represent SEM (n = 3).(B) ICR mice were injected intramuscularly with 30 μg of mLNPs or mLNP-circRNA-G for 24 h to evaluate the safety of mLNP-circRNA-G.Body weight changes in the mice were continuously monitored.Error bars represent SEM (n = 3).(C) Biochemical indicators of mLNPs and mLNP-circRNA-G-vaccinated mice were detected by an automatic hematological biochemical analyzer.(D) H&E analysis of major organs from mLNPs and mLNP-circRNA-G-vaccinated mice on day 2. Scale bars, 50 μm.