All animal experiments were approved by the Institutional Animal Care and Use Committee of the Shanghai Veterinary Research Institute, China (IACUC No: SV-20231229-G03), and were conducted in accordance with the Guidelines on the Humane Treatment of Laboratory Animals (Ministry of Science and Technology of the People’s Republic of China, Policy No. 2006398).

The JEV genotype I strains SH2201 (GenBank No. PQ488563) and SD12 (GenBank No. MH753127.1), along with the genotype III strain N28 (GenBank No. GQ918133.2) were provided by our laboratory at the Shanghai Veterinary Research Institute. A live-attenuated genotype I strain, SD12-F120 (GenBank No. MN544779), was generated through serial passage of its virulent SD12 strain on BHK-21 cells, combined with multiple plaque purification and virulence selection in mice. The SD12-F120 vaccine strain (GenBank No. MN544780.1) is a vero cell-adapted JEV strain (30). The SA14-14-2 vaccine strain (GenBank No. AF315119.1) was sourced from Wuhan Keqian Biology, Wuhan, China. All JEV strains were propagated and titrated on newborn hamster kidney cells (BHK-21), which were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM; Thermo Fisher Scientific, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS) at 37°C in a 5% CO₂ atmosphere. The 50% lethal dose (LD₅₀) of each JEV strain was determined using three-week-old female C57BL/6 mice (Shanghai SLAC Laboratory Animal Co. LTD) through intraperitoneal inoculation of serially diluted virus ( Supplementary Table S1 ).

A total of 307 sheep serum samples were collected from July to September 2022 from sheep with neurological symptoms (primarily characterized by limb spasms, and ataxia). These samples were provided by the Shanghai Branch of the China Animal Health and Epidemiology Center. Neutralizing antibodies against the JEV SH2201 strain were tested in all 307 serum samples, resulting in the identification of 64 positive serum samples. Subsequently, neutralizing antibodies against the JEV genotype I (SD12) and genotype III (N28) strains were assessed in these 64 positive serum samples.

To perform multiple sequence alignment of the E and NS1 protein amino acid sequence of JEV isolates from sheep, classical JEV strains were retrieved from GenBank using DNASTAR Lasergene 7.1 (MegAlign).

Three-week-old female C57BL/6 mice, purchased from Shanghai SLAC Laboratory Animal Co., LTD, were randomly divided into vaccinated and control groups (n = 10 mice/group). The vaccinated groups received intraperitoneally injected of the SA14-14-2 or SD12-F120 vaccine at a dose of 10⁵ PFU per animal. Serum samples were collected 14 days post-vaccination to assess neutralizing antibody titers.

In another experiment, 3-week-old female C57BL/6 mice from Shanghai SLAC Laboratory Animal Co., LTD were also randomly assigned to vaccinated and control groups (n = 15 mice/group). In order to maintain animal welfare and reduce the use of mice, we simultaneously conducted cross protection experiments with the SA14-14-2 and SD12-F120 vaccine strain against the SH2201 strain, and shared data from the JEV strain SH2201, SD12, and N28 strain challenge control groups. Mice were intraperitoneally injected with the SA14-14-2 or SD12-F120 vaccine at doses of 10,000, 1,000 or 100 PFU (plaque-forming unit) per animal. And then were challenged intraperitoneally with a dose of 20 LD₅₀ of JEV SH2201 at 14 days post-vaccine boost. The challenged mice were monitored daily for 21 days and the mortality rates were calculated accordingly. Mice blood (n = 5/group) was collected for viraemia detection from day 0 to day 3 post challenge and euthanized on day 7 post challenge and tissue was collected for viral load testing. Histopathologic analysis of brain tissue.

The viral load of each sample was assessed by RT-qPCR using JEV NS1 gene-specific primers and the viral titer was determined using the TCID₅₀ method, as previously described (37). For histopathological analysis, the brain of mice were sectioned and subjected to hematoxylin and eosin staining (Wuhan Servicebio Technology Co., Ltd, Wuhan, China). The severity of histopathological lesions was assessed by determining the integrated optical density in three fields of view using Case Viewer software (Wuhan Servicebio Technology Co., LTD) (38).

Neutralizing antibodies in serum samples collected from both mice and sheep were detected by 50% plaque reduction neutralization test (PRNT₅₀), as previously described (39). Briefly, serum samples were inactivated in a water bath at 56˚C for 30 minutes, then serially diluted two-fold, starting from a 1:5 dilution up to 1: 320. The diluted serum samples were mixed with an equal volume of JEV at a concentration of 200 PFU/0.1 ml and incubated at 37˚C for 1 hour. The mixture was then dispensed onto BHK-21 cells grown in 6-well plates and incubated at 37˚C for 2 hours. Following the adsorption period, the cells were overlaid with 1.2% methylcellulose (Thermo Fisher Scientific) in DMEM containing 2% fetal bovine serum (FBS) and incubated at 37˚C for 3 to 5 days until significant viral plaques were observed. The plaques were stained with 0.5% crystal violet, and the neutralizing antibody titers were calculated (40).

In this study, the positive cut-off value for neutralizing antibody titers was defined as PRNT₅₀ ≥1:10. For PRNT₅₀ titers below this positive threshold, the geometric mean titer (GMT) was calculated using a minimum value of 5. All data were processed and analyzed using GraphPad Prism 8.0 software (GraphPad, La Jolla, CA, USA). Student’s t-tests were used for statistical analyses. A p value < 0.05 was considered statistically significant.

We compared the amino acid differences in the E protein between the JEV SH2201 strain, isolated from sheep with neurological signs and genotype I and genotype III vaccine and virulent strains ( Table 1 ). The SH2201 strain differs at E138(R→E) from the genotype I live-attenuated vaccine strain SD12-F120. It also exhibits variations at E107 (F→L), E129 (T→M), E138 (K→E), E176 (V→I), E222 (A→S), E244 (G→E), E264 (H→Q), E327 (S→T) and E366 (A→S) compared to the genotype III live-attenuated vaccine strain SD14-14-2 ( Table 1 ). Domain III of the E protein is a critical antigenic site antigenic site recognized by the host’s immune system, playing a vital role in generating antibodies against the virus (23). Mutations in Domain III can impact the virus’s virulence, tropism, and antigenicity. Particularly, mutations at E327 (S→T) and E366 (A→S) within Domain III may influence the protective efficacy of the SA14-14-2 vaccine against GI strain infections (39).

Additionally, NS1 is a multifunctional protein that plays a role in JEV replication, pathogenesis, and the immune response to infection (41, 42). The SH2201 strain differs from the genotype III live attenuated vaccine strain SA14-14-2 and the genotype I live attenuated vaccine strain SD12-F120 at amino acid position 73 (R→G) and 178 (T→W). The SH2201 strain differs from the genotype I live attenuated vaccine strain SD12-F120 at 219 (G→R) ( Supplementary Figure S1 ).

We assessed the neutralizing antibody titers induced by the SA14-14-2 and SD12-F120 vaccines against the sheep-derived SH2201 (GI) strain. Serum samples were collected from mice (n = 10) vaccinated with the SA14-14-2 and SD12-F120 vaccines, and the neutralizing antibody titers (PRNT₅₀) specific to the sheep-derived SH2201 (GI), homologous genotype SD12 (GI), and heterologous N28 (GIII) strains were measured. Before vaccination, the neutralizing antibody titers in serum samples were all below 10 ( Figure 1A ). Following vaccination with the SA14-14-2 vaccine, the PRNT₅₀ titer against the homologous genotype N28 (GIII) strain was 26.5, higher than the titer (12) for the sheep-derived SH2201 (GI) strain, which had a specificity of 12. The PRNT₅₀ titer against the heterologous genotype SD12 (GI) strain was 12, also higher than that for the SH2201 (GI) strain ( Figure 1B ). The seropositivity rate for the homologous N28 strain was 90.0% (9/10), while the seropositivity rates for both the sheep-derived SH2201 (GI) and the heterologous genotype SD12 (GI) strains were 40.0% (4/10) ( Figure 1C ). After vaccination with the SD12-F120 vaccine, the PRNT₅₀ titer against the sheep-derived SH2201 (GI) strain was 29, exceeding which was higher than the PRNT₅₀ titer of 10 for the heterologous genotype N28 (GIII) strain, but lower than the PRNT₅₀ titer of 33.5 for the homologous genotype SD12 (GI) strain ( Figure 1D ). The seropositivity rate for the sheep-derived SH2201 (GI) strain was 80.0% (8/10), while the seropositivity rate for the homologous genotype SD12 (GI) strain was 90.0% (9/10), and the rate for the heterologous genotype N28 (GIII) strain was 40.0% (4/10) ( Figure 1E ). The results above indicate that mice vaccinated with the SA14-14-2 vaccine have a low neutralizing antibody titer against the heterologous genotype SH2201 (GI) strain, with a weaker immune response.

Considering that mice vaccinated with the SA14-14-2 vaccine exhibited relatively low titers of specific neutralizing antibodies against the sheep-derived SH2201 (GI), we assessed the protective efficacy of the SA14-14-2 vaccine against challenge with this strain. Mice susceptible to JEV infection were vaccinated with high (10,000 PFU), medium (1,000 PFU), or low (100 PFU) doses of the SA14-14-2 vaccine, and then were challenged intraperitoneally with a dose of 20 LD₅₀ of the sheep-derived SH2201 (GI) strain or the homologous genotype N28 (GIII) strain. The results demonstrated that mice vaccinated with high and medium doses of the SA14-14-2 vaccine were completely protected from the homologous genotype N28 (GIII) strain. However, those receiving a low dose showed only partial protection, achieving a survival rate of 60.0% ( Figure 2C ). Furthermore, none of the dosage groups provided complete protection against the sheep-derived SH2201 (GI) strain, with protection rates of 80%, 60%, and 40% observed for the high, medium, and low dose groups, respectively ( Figures 2A-C ).

For additional validation of the immunoprotective effect of the SA14-14-2 live-attenuated vaccine, blood was collected from mice on day 0 to day 3 post challenge to measure the viremia. The results showed that the viremia in the N28 and SH2201 strain control group mice peaked on day 2 post-challenge, and viremia up to 2.1×10⁴ RNA copies/ml to 4.8×10⁴ RNA copies/ml ( Figures 2D–F ). No viremia was detected in the high and medium-dose SA14-14-2 vaccine groups following challenge with the homologous N28 strain ( Figures 2D, E ). In the low-dose group, viremia was detected in 2 out of the mice challenged with the homologous N28 strain ( Figures 2F ). For the challenge with the heterologous SH2201 strain, viremia was detected in 2 mice in the medium-dose vaccine group and in 3 mice in the low-dose group ( Figures 2E, F ).

Additionally, we also measured the viral load and titer in various tissues of the control and vaccine-immunized mice. The results showed that the highest viral loads were observed in the brains of N28 and SH2201 control mice, with titers ranging from 7.6×10¹ to 7.6×10³ TCID₅₀/ml. Other tissues (heart, liver, spleen, lung and kidney) were also detected at different levels of viral load ( Figures 2G–L ). JEV was not detected in the tissues of mice challenged with homozygous N28 in the high-dose and medium-dose SA14-14-2 vaccine immunization groups ( Figures 2G, H ). In the low-dose group, JEV was detected in the brain, heart, liver, spleen, and lung tissues of two mice challenged with the homologous N28 strain, with the highest viral titer in the brain, ranging from 5.1×10¹ to 1.9×10² TCID₅₀/ml ( Figures 2I, L ). In the medium and low-dose vaccine groups, 2-3 mice had JEV detected in the brain, heart, liver, spleen, and lung tissues after challenge with the heterologous SH2201 strain ( Figures 2H, I ), with the highest viral titer in the brain tissue of 3.2×10² TCID₅₀/ml ( Figures 2K, L ). A significant difference in the viral titer in the brain tissue was observed between the high and medium-dose vaccine groups and the control group. These data indicate that the SA14-14-2 vaccine offers partial protection against the sheep-derived SH2201 (GI) strain.

The development of genotype I (GI) vaccine has been proposed to control GI encephalitis virus infections, as the genotype III (GIII) encephalitis vaccine has been shown to lack complete cross-protection against GI strains (43). Given that the JEV strain SH2201, isolated from neurologically symptomatic sheep, belongs to genotype I, we evaluated the protective efficacy of the SD12-F120 (GI) vaccine against challenge with the SH2201 (GI) virus. Mice were vaccinated with high (10,000 PFU), medium (1,000 PFU), or low (100 PFU) doses of the SD12-F120 (GI) vaccine and then challenged with both homologous SD12 (GI) and sheep-derived SH2201 (GI) viruses. The high and medium doses of the SD12-F120 vaccine provided complete protection against both the homologous SD12 (GI) and the sheep-derived SH2201 (GI) strains ( Figures 3A, B ). However, the low dose did not fully protect the mice, yielding protection rates of 90.0% against SD12 (GI) and 80.0% against SH2201 (GI) ( Figure 3C ).

For additional validation of the immunoprotective effect of the SD12-F120 live-attenuated vaccine, blood was collected from mice on day 0 to day 3 post challenge to measure the viremia. The results showed that the viremia in the SH2201 and SD12 strains control group mice peaked on day 2 post-challenge, and viremia up to 4.8×10⁴ RNA copies/mL to 5.6×10⁴ RNA copies/ml ( Figures 3D–F ). No viremia was detected in the high and medium-dose SD12-F120 vaccine groups following challenge with the homologous SD12 and SH2201 strain ( Figures 3D, E ). In the low-dose group, viremia was detected in one out of the mice challenged with the homologous SD12 strain and two out of the mice challenged with the homologous SH2201 strain ( Figures 3F ).

Additionally, we also measured the viral load and titer in various tissues of the control and vaccine-immunized mice. The results showed that the highest viral loads were observed in the brains of SD12 and SH2201 control mice, with titers ranging from 1.3×10² to 7.6×10³ TCID₅₀/ml. Other tissues (heart, liver, spleen, lung and kidney) were also detected at different levels of viral load ( Figures 3G–L ). JEV was not detected in the tissues of mice challenged with homozygous SD12 and SH2201 in the high-dose and medium-dose SD12-F120 vaccine immunization groups ( Figures 3G, H ). In the low-dose group, JEV was detected in the brain, heart, and lung tissues of one mouse challenged with the homologous SD12 strain, with the highest viral titer was 2.5×10¹ TCID₅₀/ml in the brain. And in the brain, heart, liver, and lung tissues of one mouse challenged with the homologous SH2201 strain, with the highest viral titer was 1.3×10² TCID₅₀/ml in the brain ( Figures 3I, L ). A significant difference in the viral titer in the brain tissue was observed between vaccine immunization groups and the control group. These results demonstrate that the SD12-F120 vaccine offers substantial protective efficacy against the sheep-derived SH2201 (GI) strain, achieving over 80.0% protection even at low doses.

To further explore the cross-protection, brain samples were collected from the vaccinated and control groups mice at 7 days post challenge for analysis of histopathological lesions. The histopathological lesions of encephalitis were not observed in blank control mice, whereas the control mice challenged with the virulent N28 (GIII), SH2201 (GI), or SD12 (GI) strain developed encephalitis histopathologically characterized by the multifocal lymphohistiocytic ( Figures 4A–D ). The histopathological lesions of encephalitis were not observed in the brain tissue of mice from the SA14-14-2 vaccine immunization group challenged with the homologous N28 (GIII) strain. However, histopathological lesions of encephalitis were observed of mice challenged with the heterologous SH2201 (GI) strain ( Figures 4E, F ). The histopathological lesions of encephalitis were not observed in the brain tissue of mice from the SD12-F120 vaccine immunization group challenged with either the homologous SD12 (GI) or the SH2201 (GI) strain ( Figures 4G, H ).

Neutralizing antibody titers (expressed as GMT) were assessed in 64 serum samples from convalescent sheep with Japanese encephalitis. The results indicated that when the neutralizing antibody titer against the SH2201 strain ranged from 1:10 to 1:20, there was protective efficacy against the SD12 (GI) strain (GMT ≥ 10). When the neutralizing antibody titer against SH2201 reached ≥ 1:40 (GMT ≥ 10), protective efficacy was observed against both the SD12 (GI) and N28 (GIII) strains, with serum positivity rates of 100% for both strains ( Table 2 ). These findings confirm the presence of a certain degree of JEV infection and prevalence among sheep in clinical settings.