Vero cell was purchased from American Type Culture Center (ATCC) and maintained in RPMI-1640 medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. HEK293T cell was obtained from ATCC and cultured using MEM medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. Both JEV (strain ZJ14-52, GenBank Accession No. MK558811) and ZIKV (strain Zhejiang04, GenBank Accession No. KX117076.1) were propagated in Vero cells and the viral stocks were tittered using plaque-forming assay (PFA) as described below. The viral titer was expressed as plaque-forming unit (PFU)/mL. HRP-conjugated goat anti-mouse IgG and FITC-conjugated goat anti-mouse IgG were purchased from Proteintech (Chicago, America). 3,3′,5,5′-Tetramethylbenzidine (TMB) was purchased from Macklin (Shanghai, China). Methyl cellulose M450 was purchased from Ourchem (Beijing, China). Isopropyl-β-D-thiogalactoside (IPTG) was purchased from Solarbio (Beijing, China). Polyetherimide (PEI) transfection reagent was purchased from GlpBio (California, United States).

Six-week-old male and female wild-type (WT) C57BL/6 mice were purchased from the Animal Model Research Center at Nanjing University, China, and were maintained and bred in the Animal Center of Ningbo University, China. The experiments were performed in strict accordance with the animal welfare act and public health service policy on humane care and use of laboratory animals and were approved by the ethical committee of Ningbo University (No. 2020-55). Sample sizes were estimated on the basis of our previous studies. Animal experiments were randomized but not blinded.

Three previously identified H-2^b-restricted CD8⁺ T cell epitopes in ZIKV (ZIKV-E_4–12, ZIKV-E_57–65, and ZIKV-E_420–428) and their variants in JEV (JEV-E_4–12, JEV-E_57–65, and JEV-E_416–424) (Chen et al., 2020) were synthesized by ChinaPeptides with a purity >95% (Supplementary Table S1).

The genes (optimized for E. coli codon usage) encoding the extracellular domain of E protein of JEV (E₄₀₆, aa 1-406) and ZIKV (E₄₁₀, aa 1-410) were synthesized by GeneWiz (Suzhou, China) and cloned into the pET21a expression vectors, each featuring a 6 × His tag at the 3′ terminal of gene sequence. pET21a-JEV-E₄₀₆ and pET21a-ZIKV-E₄₁₀ were, respectively transformed into E. coli BL21 strain, and protein expression was induced using IPTG at a final concentration of 1 mM. The recombinant JEV-E₄₀₆ and ZIKV-E₄₁₀ proteins were purified using a Ni-NTA (TransGen Biotech, Beijing, China), followed by further concentrated through a 10 kDa ultrafiltration tube (Millipore, Darmstadt, Germany). The protein purity was assessed by SDS-PAGE and the protein concentration was quantified using a BCA kit (TransGen Biotech).

The genes (optimized for Homo sapiens codon usage) encoding the complete amino acid sequences of wild-type E protein (E^WT, 500 aa) or E protein mutant (E^mut, 500 aa, with W101A and F108A amino acid substitutions in the FL sequence) of JEV were synthesized by GeneWiz and cloned into the pCAGGS plasmid, resulting in recombinant plasmids pCAGGS-JEV-E^WT and pCAGGS-JEV-E^mut. After dual enzyme digestion of the recombinant plasmid and pShuttle-CMV plasmid, the purified gene fragment was ligated with the empty pShuttle-CMV plasmid and transformed into competent bacteria Stbl3 strain to produce recombinant pShuttle-CMV plasmids (pShuttle-CMV-JEV-E^WT or pShuttle-CMV-JEV-E^mut). The linearized plasmid pShuttle-CMV-JEV-E^WT or pShuttle-CMV-JEV-E^mut was electroporated into competent E. coli BJ5183 strain containing a human adenovirus type 5 (Ad5) genomic backbone plasmid (pAdEasy-1) for homologous recombination, and the recombinant Ad5 plasmids (Ad5-JEV-E^WT and Ad5-JEV-E^mut) were screened out. The recombinant Ad5 plasmids were transfected into HEK293T cell to produce recombinant Ad5 which was then tittered by checking the GFP-positive cell using confocal microscope on the basis of Vero cell. The titer of recombinant Ad5 was expressed as infectious unit (IFU) per ml. Recombinant adenoviruses were used to infect Vero cells and the expression of JEV-E protein in the cell supernatant was evaluated using WB. JEV-E₄₀₆-immune mouse sera were used as the primary antibody and HRP-conjugated goat anti-mouse IgG mAb worked as the secondary antibody.

To prepare a mRNA sequence encoding JEV-E protein in vitro, we designed a DNA sequence consisting of T7 promoter, a 5′ UTR sequence, a Kozak sequence, a signal peptide sequence, gene sequence encoding JEV-E^WT or JEV-E^mut (W101A and F108A substitutions in the FL sequence), a 3′ UTR sequence, and a poly A tail from 5′ terminal to 3′ terminal, respectively (Supplementary Table S2). The DNA sequence was engineered into an optimized sequence with as much as more codons containing A or G (Supplementary Table S2) and cloned into the eukaryotic plasmid pBluescript II KS (+) to produce recombinant plasmids pBluescript II KS (+)-JEV-E^WT and pBluescript II KS (+)-JEV-E^mut. Using the linearized pBluescript II KS (+)-JEV-E^WT or pBluescript II KS (+)-JEV-E^mut as template, RNA was produced using an in vitro transcription kit after supplementing with N1-methyl-pseudouridine and other components. Finally, the 5′ terminal of RNA was capped using capping reagent kit (Novoprotein, Suzhou, China) to generate functional JEV-E^WT mRNA and JEV-E^mut mRNA.

293T cells were seeded to cell culture plate (10 cm²) and incubated for 24 h in a 37°C/5%CO₂ incubator. Next day, the cell supernatant was discarded and MEM medium without FBS was added to the cells. Four hours later, 10 μg mRNA was diluted in PBS to reach 0.5 mL. 0.5 mL PEI (20 μg) was added to diluted mRNA and incubated for 20 min at room temperature. The PEI/mRNA mixture was added to cells and transfected for 4 h. After discarding the transfection reagent, complete MEM medium (with 10% FBS) was added to the cells and incubated for 2 days in the incubator. The cell supernatant was harvested for detecting JEV-E protein using WB. JEV-E₄₀₆-immune mouse sera were used as the primary antibody and HRP-conjugated goat anti-mouse IgG mAb worked as the secondary antibody.

The purified mRNA was diluted into citrate buffer (pH4, 50 mM) to form the aqueous phase with a final concentration of 170 μg/mL. Briefly, 4-(N,N-dimethylamino) butyric acid (dilinoleic) methyl ester (Dlin-MC3-DMA; cationic lipid), Diethyl phosphatidylcholine (DSPC; non-cationic lipid), Cholesterol (sterol), and 1, 2-dimyristoyl-rac-glycero-3-methoxypolyethylene glycol-2000 (DMG-PEG2000; PEG-modified lipid) were dissolved into ethanol to make the organic phase, reaching the final concentration of 6.25 mM, 1.25 mM, 4.8125 mM, 0.1875 mM, respectively. The aqueous phase and the organic phase were mixed at a volume ratio of 3:1 and passed through a Microflow S microfluidic nanomedicine fabrication instrument at a flow rate of 12 mL/min to produce JEV-E^WT mRNA-LNP or JEV-E^mut mRNA-LNP. The zeta potential and polydispersity of mRNA-LNP were determined using particle size analyzer. The particle size of mRNA-LNP was measured using transmission electron microscope. The mRNA-LNP was cleaved using 2% Triton X-100 and the amount of released mRNA was measured for calculating the mRNA encapsulation rate.

Recombinant adenovirus and JEV-E mRNA-LNP were used to immunize C57BL/6 mice. As for recombinant adenovirus immunization, 6-week-old male mice were subcutaneously (s.c.) injected with Ad5-JEV-E^WT or Ad5-JEV-E^mut (1 × 10⁸ IFU/mouse). The mice receiving PBS were used as control. Two weeks later, the mice were boosted with the same recombinant adenovirus. Two weeks after the second immunization, the mice were sacrificed and the mouse sera were collected for ELISA, plaque-reduction neutralization test (PRNT), and in vitro ADE assay. The mRNA-LNP was used to immunize adult (6-week-old) or young (1-week-old) mice. In one protocol, 6-week-old male or female mice were injected s.c. with JEV-E^WT mRNA-LNP or JEV-E^mut mRNA-LNP (5 or 50 μg mRNA/mouse) twice at a two-week interval. The mice receiving PBS were used as control. Two weeks after the second immunization, the male mice were sacrificed and the mouse sera were used for ELISA, PRNT, and passive transfer experiments. The splenocytes were used for ELISpot assay. The immunized female mice were mated with 10-week-old naïve male mice. Approximate 3 weeks later, the 1-day-old neonates born to the mRNA-LNP-immunized mice were injected s.c. with 1,000 PFU of ZIKV, and the mouse weight and mortality were recorded daily for 28 days. In another protocol, 1-week-old neonates were injected s.c. with mRNA-LNP (0.1 or 1 μg mRNA/mouse) and boosted with the same mRNA-LNP 2 weeks later. Two weeks after the second immunization, some immunized mice were sacrificed and the mouse sera were used for ELISA and PRNT; the remaining mice were injected s.c. with 1 × 10⁵ PFU of JEV. Three days later, the mice were sacrificed and the mouse sera were harvested for viral load detection using PFA.

The antigen-reactive IgG in mouse sera (recombinant adenovirus-infected mice; mRNA-LNP-immunized adult mice; mRNA-LNP-immunized young mice) was detected using indirect ELISA. In brief, the high-affinity ELISA plates were coated with recombinant JEV-E₄₀₆ or ZIKV-E₄₁₀ proteins (20 μg/mL, 100 μL/well) and incubated for 1.5 h at 37°C. The plates were then blocked for 1.5 h at 37°C using 5% non-fat milk/PBS solution. Serially diluted mouse sera were added to the plates and incubated for 1.5 h at 37°C. HRP-conjugated Goat anti-mouse IgG (1:2,000 dilution) was added to the plate and incubated for 1.5 h at 37°C. The plate was developed with fresh TMB solution for 10 min at room temperature. The OD₄₅₀ value of each well was measured using a microplate reader.

The ability of recombinant adenovirus-immune or mRNA-LNP-immune mouse sera to neutralize JEV in vitro was evaluated on Vero cells using PRNT. In brief, Vero cells were seeded in each well (1 × 10⁵ cells/well) of 24-well cell culture plate and incubated for 24 h in a 37°C/5% CO₂ incubator. Mouse sera (recombinant adenovirus-infected mice; mRNA-LNP-immunized adult mice; mRNA-LNP-immunized young mice) were 3-fold serially diluted in the well (240 μL/well) of 96-well culture plate, followed by addition of 50 μL of JEV solution (containing 50 PFU), and incubated for 1 h at 37°C. After discarding the supernatant of each well, the mixture of serum/virus was added to Vero cell and infected for 1 h at 37°C. After discarding the viral solution, the cells were overlaid with 1% methyl cellulose/medium and incubated for 3 days in a 37°C/5% CO₂ incubator. The cells were fixed with 4% paraformaldehyde for 1 h at room temperature and stained with 0.5% crystal violet solution for 10 min at room temperature. The plaques of each well were counted and the NT₅₀ (the highest serum dilution that can reduce viral plaques by 50%) titers were calculated.

The potential of mouse sera to enhance the infection of ZIKV was assessed on the basis of K562 cells. K562 cells were plated to each well (2 × 10⁴ cells/well) of U-bottom 96-well cell culture plates and incubated for 24 h in the 37°C/5% CO₂ incubator. Mouse sera (Ad5-JEV-E^WT-immune, Ad5-JEV-E^mut-immune) were serially diluted in the well (50 μL/well) of 96-well plate and 50 μL of ZIKV (containing 1,000 PFU) were added to each well, and incubated at 37°C for 1 h. After centrifuged for 10 min at 2,000 rpm, the supernatant of K562 cells was discarded, followed by addition of the serum/virus mixture. After 1-h infection at 37°C, the viral solution was discarded and fresh medium was added to cells. After 24-h incubation, the cells were collected and fixed with 4% paraformaldehyde for 40 min at room temperature. The cells were then permeabilized with 0.5% Triton X-100/PBS for 40 min at temperature. The cells were incubated with viral antigen-specific antibody (ZIKV-E₄₁₀-immune mouse serum, 1:200 dilution) for 1 h at room temperature, followed by incubation with FITC-labeled goat anti-mouse IgG mAb (1:1000 dilution) for 1 h at room temperature. The percentage of FITC-positive cells among each sample was determined using flow cytometry.

Mouse splenocytes were plated to each well (1 × 10⁶/100 μL/well) of 96-well precoated ELISpot plate (DAKEWE, coated with anti-mouse IFNγ mAb), followed by addition of individual peptide (2 μg/well). Wells with cells but without peptide were served as negative control. After 24 h incubation in a 37°C/CO₂ incubator, the plates were incubated sequentially with biotinylated anti-mouse IFN-γ mAb, HRP-conjugated streptavidin, and developed with 3-amino-9-ethylcarbazole color development solution. The colored spots of each well were counted and the results were expressed as the number of spot-forming cell (SFC) per 1 × 10⁶ mouse splenocytes. The presence of >20 SFCs/well with a stimulation index of >2 (index = ratio of test SFCs to control SFCs) was considered a positive response.

The effect of JEV-E mRNA-LNP-immune mouse sera on JEV or ZIKV infection in vivo was determined using passive transfer experiments. Different volume (3 or 15 μL) of either JEV-E^WT mRNA-LNP-immune mouse sera or JEV-E^mut mRNA-LNP-immune mouse sera were injected s.c. into 1-day-old mice. Two hours later, the mice were injected s.c. with 10 PFU of JEV or 1,000 PFU of ZIKV. The mouse weights and mortality were recorded daily for 14 days and 28 days, respectively.

The titers of viral stock and the viral load in mouse sera were determined using PFA. Briefly, Vero cells are seeded to each well (1 × 10⁵ cells/well) of 24-well cell culture plate and incubated for 24 h in a 37°C/5% CO₂ incubator. After discarding the cell supernatant, 240 μL of serially diluted viral stock or mouse serum was added to cells and the plate was incubated for 1 h at 37°C. After discarding the samples, the cells were overlaid with 1% Methyl cellulose/medium and incubated for 3–5 days in a 37°C/5% CO₂ incubator. The cells were fixed with 4% paraformaldehyde for 1 h at room temperature and stained with 0.5% crystal violet solution for 10 min at room temperature. The numbers of plaque in each well were counted and the viral load in mouse serum was expressed as PFU/mL.

All data were analyzed using Prism 8 software (GraphPad Software, La Jolla, CA, United States) and expressed as the means ± SEM. The comparison between group means and survival data were analyzed using the Mann–Whitney test and a log-rank test, respectively. p < 0.05 was considered statistically significant.

To explore whether optimizing the FL sequence of the JEV-E protein affects the production of ZIKV-cross-reactive Abs, we initially designed the amino acid sequence of JEV-E protein mutant (JEV-E^mut) by incorporating two amino acid substitutions to the wild-type E protein (JEV-E^WT). With the assistance of shuttle plasmid, the genes encoding the JEV-E^WT or JEV-E^mut were successfully transferred into the human Ad5 genomic backbone plasmid to produce a recombinant adenovirus plasmid carrying the JEV-E^WT gene or JEV-E^mut gene. After transfecting the linearized recombinant adenovirus plasmid into 293T cells, recombinant adenovirus particles (having an average diameter of around 100 nm) were identified in the cytoplasm of cells using transmission electron microscopy (Figures 1A,B). We collected the cell supernatants and purified the recombinant adenoviruses Ad5-JEV-E^WT and Ad5-JEV-E^mut. Two recombinant adenoviruses were sent to company GeneWiz for sequencing, and the inserted nucleotide sequences showed no nucleotide loss, insertion, or mutation as compared to our original sequences (data not shown). Both Ad5-JEV-E^WT and Ad5-JEV-E^mut were used to infect Vero cells and the presence of JEV-E protein in the cell supernatant was determined using WB (Figure 1C). Since the recombinant adenovirus carries the gene encoding GFP, the expression of GFP in the Vero cells infected with Ad5-JEV-E^WT or Ad5-JEV-E^mut was observed using confocal microscopy and the titer of recombinant adenovirus was determined by counting GFP-positive cells (Figures 1D,E).

We further designed DNA sequence for producing mRNA in vitro and cloned it into pBluscript II KS (+) plasmid, resulting into pBluscript II KS (+)-JEV-E^WT and pBluscript II KS (+)-JEV-E^mut (Figure 2A). On the basis of linearized recombinant plasmid, we proceeded to generate mRNA encoding JEV-E^WT or JEV-E^mut with a length of 1875 bp (Figure 2B; Supplementary Figure S1). Following transfection of Vero cells with either JEV-E^WT mRNA or JEV-E^mut mRNA, the expression levels of the E protein were assessed using WB, displaying a similar expression proficiency (Figure 2C). We next encapsulated the mRNA within four components (cationic lipid, non-cationic lipid, cholesterol, and PEG-modified lipid) and produced the mRNA-LNP. The transmission electron micrograph showed that the mRNA-LNPs have an average diameter of around 100 nm (Figures 2D,E).

We next immunized 6-week-old mice with recombinant adenoviruses Ad5-JEV-E^WT or Ad5-JEV-E^mut, and found that each recombinant adenovirus immunization induced the production of JEV-E-reactive IgG (Supplementary Figures S2A,B), which could cross-recognize ZIKV-E protein. Interestingly, the ability of Ad5-JEV-E^mut-immune mouse serum Abs to recognize ZIKV-E protein was significantly weaker than that of Ad5-JEV-E^WT-immune mouse serum Abs (Supplementary Figures S2C,D). Next, 6-week-old mice were immunized with different doses of JEV-E^WT mRNA or JEV-E^mut mRNA in the form of LNP twice at a two-week internal, and the levels of JEV-E protein-reactive IgG and cross-reactive Abs against ZIKV-E protein were quantified using an ELISA. The results demonstrated that whether it is 5 μg or 50 μg, JEV-E^WT mRNA-LNP and JEV-E^mut mRNA-LNP induced a similar level of JEV-E-reactive antibody response (Figures 3A,B). However, 50 μg of JEV-E^mut mRNA induced a significantly lower level of ZIKV-E-cross-reactive antibody response compared to JEV-E^WT mRNA (Figures 3C,D). Considering 10-week-old mice resist JEV infection and have no obvious viremia post JEV infection whereas 5-week-old mice are susceptible to JEV infection and can maintain viremia for several days post JEV infection (data not shown), we then immunized 1-week-old young mice with different doses of JEV-E^WT mRNA-LNP or JEV-E^mut mRNA-LNP twice at a two-week interval and determined the JEV-E protein-reactive IgG in the mouse sera at 2 weeks post the second immunization. Similar Ab response against JEV-E protein and ZIKV-E protein were observed in mouse sera (Figures 3E–H). These data indicate that the substitution of amino acid residues among the FL of JEV can significantly attenuate the induction of ZIKV-E-cross-reactive Abs, which is not affected by mouse age. We next evaluate the cellular immune response induced by JEV-E mRNA-LNP immunization by detecting the frequencies of peptide-specific T cells in the splenocytes of JEV-E mRNA-LNP-immunized mice. The results show that 5 μg JEV-E^WT mRNA induced a similar T cell immune response with 5 μg JEV-E^mut mRNA while 50 μg JEV-E^WT mRNA induced a stronger T cell immune response than 50 μg JEV-E^mut mRNA (Figures 4A,B). But, 50 μg mRNA did not significantly increase T cell response than 5 μg mRNA (Figures 4C,D). These results indicate that substitution of amino acid residues in the FL sequence of JEV can significantly reduce the level of ZIKV-cross-reactive antibody response, but does not affect the humoral and cellular immune responses against JEV antigens.

Due to the resistance of K562 cells to ZIKV infection and its inability to be directly infected, the ZIKV-binding Abs can mediate the virus entry to cells by binding to the Fc receptors on the surface of K562 cells, making it a commonly used cell line for performing in vitro ADE assay (Chen et al., 2020). Therefore, we compared the differences in enhancing ZIKV infection in vitro between Ad5-JEV-E^WT-immune mouse sera and Ad5-JEV-E^mut-immune mouse sera based on K562 mice. As shown in Figures 5A,B, Ad5-JEV-E^WT-immune mouse serum has a strong enhancing effect on ZIKV infection, and enhanced infection fold decreases continuously with serum diluted. When the serum is diluted at 1:10, the infection of ZIKV is increased by about 16 folds. Although Ad5-JEV-E^mut-immune mouse serum still has an enhancing effect on ZIKV infection, there is a significant reduction compared to Ad5-JEV-E^WT-immune mouse serum in total. When the serum is diluted at 1:10, it enhances ZIKV infection by about 8 folds.

In light of the aforementioned substitution of two amino acids within the FL region of the E protein were shown to attenuate the induction of ZIKV-cross-reactive Abs, we were then propelled to investigate the potential impact of such amino acid substitutions on the production of neutralizing Abs against JEV using PRNT. The results demonstrate that sera from mice immunized with Ad5-JEV-E^mut effectively neutralized JEV in vitro, with a NT₅₀ titer of around 1:199 which is significantly higher than that of Ad5-JEV-E^WT-immune mouse sera (NT₅₀ titer is about 1:70, Figures 6A,B). Interestingly, sera from 6-week-old or 1-week-old mice immunized with JEV-E^WT mRNA-LNP or JEV-E^mut mRNA-LNP exhibit high level of JEV-neutralizing antibody response. The NT₅₀ titers are 1:199 versus 1:933, 1:468 versus 1:588, respectively (Figures 6C–F). These results suggest that substitution of amino acid residues in the FL sequence of JEV does not affect the level of JEV-neutralizing antibody response induced by JEV-E^mut.

To evaluate whether two types of JEV-E mRNA-LNP-immune mouse sera can protect recipient mice from JEV attack, we injected 15 μL of JEV-E^WT mRNA-LNP or JEV-E^mut mRNA-LNP-immune mouse sera into 1-day-old neonatal mice, and then attacked them with JEV, followed by mouse survival study. As shown in Figures 7A,B, the percent survival of recipient mice transferred with 15 μL of JEV-E^WT mRNA-LNP-immune mouse sera or JEV-E^mut mRNA-LNP-immune mouse sera was significantly higher than that of the control group (55.5% versus 0 and 37.5% versus 0, respectively).

Given the efficacy observed in passive serum transfer assays, whereas mouse-immune sera conferred protection against lethal JEV challenges, this next experiment aimed to ascertain whether our developed mRNA vaccine candidates could directly protect mice from JEV. To this end, one-week-old mice were immunized with 1 μg of JEV-E^WT mRNA or JEV-E^mut mRNA for twice and then challenged with JEV 2 weeks after the second immunization. Three days post JEV attack, the mouse sera were collected for viral load analysis. The data demonstrated that both mRNA-LNP immunization significantly shielded the mice from JEV infection. PFA results revealed that the levels of viral load in the sera of mice immunized with either JEV-E^WT mRNA-LNP or JEV-E^mut mRNA-LNP were substantially lower as compared to those in naïve mice (43 PFU/mL versus 501 PFU/mL, p = 0.0148; 26 PFU/mL versus 501 PFU/mL, p = 0.0093, respectively), confirming the protective efficacy of the JEV-E mRNA-LNP against JEV infection (Figure 8). These results indicate that mRNA constructed based on E protein mutant with optimized FL sequences showed no significant difference in protecting mice against ZIKV attacks compared to mRNA encoding wild-type E protein of JEV.

Considering a significant decrease in the level of ZIKV-cross-reactive antibody induced by JEV-E^mut mRNA-LNP, we next to explore whether the reduction of ZIKV-cross-reactive antibody response can potentially affect the occurrence of ADE of ZIKV infection in vivo. Specifically, we injected 3 or 15 μL of pooled JEV-E^WT mRNA-LNP- or JEV-E^mut mRNA-LNP-immune mouse sera into 1-day-old neonates and performed mouse survival study post ZIKV challenge. As a result, 1-day-old mice receiving 3 μL of pooled naive mouse sera all died within 22 days after ZIKV infection (Figures 9A,B). However, 1-day-old mice given with 3 μL of JEV-E^WT mRNA-LNP-immune mouse sera showed an accelerated death (all die within 14 days after ZIKV challenge) (Figures 9A,B). But, the 1-day-old mice transferred with 3 μL of JEV-E^mut mRNA-LNP-immune mouse sera all died at around 19 days after ZIKV infection and exhibited a significant reduction of weight loss compared with mice receiving naive mouse sera at multiple timepoints post ZIKV infection (Figures 9A,B).

Next, 1-day-old pups injected with 15 μL of naive mouse serum also died within 20 days after ZIKV infection. However, 1-day-old pups given with 15 μL of JEV-E^WT mRNA-LNP-immune mouse serum exhibited a significant acceleration of death (all died within 11 days after ZIKV infection). However, 1-day-old pups receiving 15 μL of JEV-E^mut mRNA-LNP-immune mouse serum died around 23 days after ZIKV infection. Compared to JEV-E^mut mRNA-LNP-immune mouse serum, JEV-E^WT mRNA-LNP-immune mouse serum significantly accelerated the death of 1-day-old mice (Figures 9C,D). These results indicate that the Abs induced by mRNA encoding wild-type E protein of JEV can exacerbate ZIKV infection in mice at the specific doses, leading to earlier death of mouse, representing speculated ADE of ZIKV infection. The Abs induced by mRNA encoding the E protein mutant of JEV did not have an obvious enhancing effect on ZIKV infection, suggesting optimized JEV mRNA vaccine is safer than unmodified mRNA vaccine.

Although our passive transfer experiment results showed that Abs against the JEV-E^WT protein can accelerate the death of ZIKV-infected mice at certain doses, but, do anti JEV-E^WT Abs have ADE against ZIKV infection in the real world? If so, whether the Abs induced by optimized mRNA vaccine will not mediate ADE of ZIKV infection? Considering that IgG in the mother’s body can across the placenta and enter the body of fetus or newborn, as well as that many women who have been infected with JEV or vaccinated with JEV vaccine will have children who may subsequently be infected with ZIKV, we then evaluate the impact of anti-JEV Abs obtained by the pups from the female mice vaccinated with either JEV-E^WT mRNA-LNP or JEV-E^mut mRNA-LNP on subsequent ZIKV infection. The results showed that the 1-day-old mice born to JEV-E^mut mRNA-LNP-immunized female mice loss less weight compared to 1-day-old mice born to naive females post ZIKV challenge (Figure 10A). Approximately 6% of 1-day-old mice born to naive female mice survived ZIKV infection, while all 1-day-old mice born to JEV-E^WT mRNA-LNP-immunized female mice died within 17 days post ZIKV challenge. In comparison, around 14% of 1-day-old mice born to JEV-E^mut mRNA-LNP-immunized female mice survived ZIKV infection (Figure 10). These results suggest that maternally acquired Abs from JEV-E^WT mRNA-LNP-immunized female mice can accelerate the death of 1-day-old mice during ZIKV infection, while the maternally acquired Abs against JEV-E^mut protein do not exacerbate the death of 1-day-old mice infected with ZIKV, but rather slightly improve the percent survival of mice, suggesting that Abs induced by JEV-E^mut mRNA immunization may not make the host develop severe diseases after ZIKV infection.