Growth, Pathogenesis, and Serological Characteristics of the Japanese Encephalitis Virus Genotype IV Recent Strain 19CxBa-83-Cv

Genotype IV Japanese encephalitis (JE) virus (GIV JEV) is the least common and most neglected genotype in JEV. We evaluated the growth and pathogenic potential of the GIV strain 19CxBa-83-Cv, which was isolated from a mosquito pool in Bali, Indonesia, in 2019, and serological analyses were also conducted. The growth ability of 19CxBa-83-Cv in Vero cells was intermediate between that of the genotype I (GI) strain Mie/41/2002 and the genotype V (GV) strain Muar, whereas 19CxBa-83-Cv and Mie/41/2002 grew faster than Muar in mouse neuroblastoma cells. The neuroinvasiveness of 19CxBa-83-Cv in mice was higher than that of Mie/41/2002 but lower than that of Muar; however, there were no significant differences in neurovirulence in mice among the three strains. The neutralizing titers of sera from 19CxBa-83-Cv- and Mie/41/2002-inoculated mice against 19CxBa-83-Cv and Mie/41/2002 were similar, whereas the titers against Muar were lower than those of the other two viruses. The neutralizing titers of JE vaccine-inoculated mouse pool serum against 19CxBa-83-Cv and Muar were significantly lower than those against Mie/41/2002. The neutralizing titers against the three viruses were similar in three out of the five serum samples from GI-infected JE patients, although the titers against Mie/41/2002 were higher than those against 19CxBa-83-Cv and Muar in the remaining two sera samples. In summary, we identified the basic characteristics of 19CxBa-83-Cv, but further studies are needed to better understand GIV JEV.


Introduction
Japanese encephalitis (JE) is a serious nervous disorder and is a significant public health problem in many Asian countries. Humans are infected with JEV by being bitten mainly by Culex mosquitoes. There are an estimated 68,000 cases of JE per year in East, South, and Southeast Asian countries, resulting in 15,000 fatalities, mostly in children [1][2][3]. There is no specific treatment available for JE other than preventive vaccination with JE vaccines.
JEV belongs to the genus Flavivirus in the family Flaviviridae and is amplified in bird and pig-mosquito transmission cycles [4]. Ardeid birds are the natural reservoir for JEV, whereas pigs act as the major amplifying hosts of JEV and are mainly involved in outbreaks in humans. Infected Culex mosquitoes transmit JEV to dead-end hosts, either humans or horses. JEV has a single-stranded positive-sense RNA genome that encodes three structural proteins (C, prM, and E) and seven non-structural proteins (NS1, NS2A,

Phylogenetic Analysis
The phylogenetic tree was constructed with the maximum likelihood method using 500 bootstrap replications (Molecular Evolutionary Genetics Analysis software; MEGA X) based on the complete nucleotide sequence (1500 nucleotides) of the JEV E gene [44].

Plaque Formation Assay for Titration of Infectious Viruses and Analysis of Growth Kinetics
Infectious viral titers for each sample were determined using plaque formation assays. Vero cells (approximately 5 × 10 5 /well) were plated in 12-well culture plates and inoculated with each virus for 1 h at 37 • C. Next, an MEM-based overlay medium containing 1% methylcellulose (FUJIFILM Wako Pure Chemical, Osaka, Japan) and 2% FBS was added to the wells, and the cells were incubated for 4-5 days at 36 • C, after which the cells were fixed using a 10% formalin-PBS solution and stained with methylene blue, as described previously [45]. The diameter of 20 plaques was measured and the mean plaque size in mm ±SEM was calculated. Differences in mean plaque sizes were analyzed using Welch's t-test. The ability of the JEV strains to grow in vitro was analyzed as previously described [46]. Briefly, cells were cultured in 6-well culture plates and infected with each JEV strain in 3 mL MEM supplemented with 2% FBS (2F/MEM) at a multiplicity of infection (MOI) of 0.02-0.1 plaque-forming units (PFU)/cell. Small aliquots (200 µL) of the media were collected at one-day intervals, and infectious viral titers were determined by a plaque formation assay in Vero cells, as described above.

Mouse Challenge Experiment
Female ddY mice (Japan SLC, Shizuoka, Japan) were used for the challenge tests. For neuroinvasive analysis, groups of mice (3 weeks old, n = 6) were inoculated intraperitoneally (i.p.) with 100 µL (1 × 10 3 , 1 × 10 4 , or 1 × 10 5 PFU) of virus solution diluted in 0.9% NaCl solution. The mice were observed, and their body weights were measured every day for 20 days after inoculation to assess survival rates. Survival curves were compared using Bell Curve for Excel (Social Survey Research Information, Tokyo, Japan) and the log-rank (Mantel-Cox) test. Statistical significance was set at p < 0.05. The surviving mice were sacrificed, and their sera were collected for serological neutralization tests, as described in Section 2.8. For analysis of neurovirulence, groups of mice (4 weeks old, n = 6) were inoculated intracerebrally (i.c.) with 30 µL (3 × 10 2 or 3 × 10 3 PFU) of virus solution, and the mice were observed for 14 days to determine survival rates, as described above.

Mouse Vaccination
Female ddY mice (4 weeks old) were vaccinated i.p. with 0.5 mL of Vero cell-derived inactivated JE Beijing-1 (GIII) vaccine (BIKEN: Lot no. 106VC-2009) and were immunized again one week after the initial immunization. The immunized mice were sacrificed 14 days after the initial immunization, and their sera were collected and mixed. The pooled serum solution (JSS-2020) was used for the neutralization test, as described in Section 2.8.

Sera of JE Patients
Sera from four JE patients in Japan between 2019 and 2021 were also used for the neutralization study. The patients were diagnosed with JE in our laboratory using JEV-specific IgM ELISA and neutralization tests, as described previously [47]. The JEV genotypes were determined by conventional RT-PCR amplification of the JEV E region, followed by nucleotide sequencing, as described previously [48].

Plaque Reduction Neutralization Test (PRNT)
Neutralizing antibodies against JEV were measured using the PRNT method. Each JEV strain was combined at a 1:1 ratio with 2-fold serial dilutions (1:10 to 1:10,240) of human and Viruses 2023, 15, 239 4 of 13 mouse sera, and then incubated at 37 • C for 90 min. Vero cell monolayers were inoculated with the mixtures in 12-well plates and incubated at 37 • C for 90 min. Subsequently, an overlay medium containing 1% methylcellulose was added, and cells were incubated at 36 • C for 4-5 days. The cells were fixed using a 10% formalin-PBS solution and stained with methylene blue. The PRNT titer (PRNT 50 ) was defined as the reciprocal of the highest dilution resulting in a 50% reduction relative to the mouse serum-free control.

Phylogenetic Analysis of the 19CxBa-83-Cv Strain
A phylogenetic tree of JEV was constructed using the nucleotide sequence of the E gene of the GI-GV JEV strains ( Figure 1). Strain 19CxBa-83-Cv was clustered in the GIV and showed higher homology to the Bali2019 strain isolated from an Australian JE patient infected in Bali in 2019, compared to the strains JEV/Aus/NT_Tiwi Islands/2021 (Hu_Tiwi 2021; OM867669) and sw-22-00722-11 (ON624132) identified in Australia in 2021 and 2022, respectively [30,31,38,40]. types were determined by conventional RT-PCR amplification of the JEV E region, followed by nucleotide sequencing, as described previously [48].

Plaque Reduction Neutralization Test (PRNT)
Neutralizing antibodies against JEV were measured using the PRNT method. Each JEV strain was combined at a 1:1 ratio with 2-fold serial dilutions (1:10 to 1:10,240) of human and mouse sera, and then incubated at 37 °C for 90 min. Vero cell monolayers were inoculated with the mixtures in 12-well plates and incubated at 37 °C for 90 min. Subsequently, an overlay medium containing 1% methylcellulose was added, and cells were incubated at 36 °C for 4-5 days. The cells were fixed using a 10% formalin-PBS solution and stained with methylene blue. The PRNT titer (PRNT50) was defined as the reciprocal of the highest dilution resulting in a 50% reduction relative to the mouse serum-free control.

Phylogenetic Analysis of the 19CxBa-83-Cv Strain
A phylogenetic tree of JEV was constructed using the nucleotide sequence of the E gene of the GI-GV JEV strains ( Figure 1). Strain 19CxBa-83-Cv was clustered in the GIV and showed higher homology to the Bali2019 strain isolated from an Australian JE patient infected in Bali in 2019, compared to the strains JEV/Aus/NT_Tiwi Islands/2021 (Hu_Tiwi 2021; OM867669) and sw-22-00722-11 (ON624132) identified in Australia in 2021 and 2022, respectively [30,31,38,40].

Growth Properties of the 19CxBa-83-Cv Strain In Vitro
To clarify the growth properties of the GIV strain 19CxBa-83-Cv in vitro, we first compared the plaques formed by infection with 19CxBa-83-Cv to those of the GI strain Mie/41/2002 and GV strain Muar in Vero cells ( Figure 2A). The plaque size induced by 19CxBa-83-Cv (mean diameter ±SEM: 0.994 ± 0.046) was smaller than that induced by Mie/41/2002 (1.313 ± 0.049; P < 0.001) but was larger than that induced by Muar (0.671 ± 0.020; P < 0.001). Analysis of the growth kinetics of the strains showed that  Figure 2C). We previously showed that the growth ability of the Muar strain was lower than that of Mie/41/2002 in murine neuroblastoma cells [27,29]. The growth rate of 19CxBa-83-Cv was quite close to that of Mie/41/2002 and strikingly higher than that of Muar in Neuro-2a cells ( Figure 2D).

Growth Properties of the 19CxBa-83-Cv Strain In Vitro
To clarify the growth properties of the GIV strain 19CxBa-83-Cv in vitro, we first compared the plaques formed by infection with 19CxBa-83-Cv to those of the GI strain Mie/41/2002 and GV strain Muar in Vero cells (Figure 2A). The plaque size induced by 19CxBa-83-Cv (mean diameter ±SEM: 0.994 ±0.046) was smaller than that induced by Mie/41/2002 (1.313 ±0.049; P < 0.001) but was larger than that induced by Muar (0.671 ±0.020; P < 0.001). Analysis of the growth kinetics of the strains showed that 19CxBa-83-Cv grew slower than Mie/41/2002, whereas 19CxBa-83-Cv grew faster than Muar in Vero cells ( Figure 2B). The growth rate of 19CxBa-83-Cv was similar to those of Mie/41/2002 and Muar in human neuroblastoma IMR-32 cells ( Figure 2C). We previously showed that the growth ability of the Muar strain was lower than that of Mie/41/2002 in murine neuroblastoma cells [27,29]. The growth rate of 19CxBa-38-Cv was quite close to that of Mie/41/2002 and strikingly higher than that of Muar in Neuro-2a cells ( Figure 2D).

Virulence of the 19CxBa-83-Cv Strain in Mice
Next, we examined the neuroinvasiveness of 19CxBa-83-Cv in mice. Mice were infected intraperitoneally with 19CxBa-83-Cv, Mie/41/2002, or Muar, and their body weights were recorded (Figure 3 and Figure S1). We previously reported that the neuroinvasiveness of Muar was significantly higher than that of Mie/41/2002 in mice, even though Muar grew slower than Mie/41/2002 in murine neuroblastoma cells, as shown in Figure 2D

Virulence of the 19CxBa-83-Cv Strain in Mice
Next, we examined the neuroinvasiveness of 19CxBa-83-Cv in mice. Mice were infected intraperitoneally with 19CxBa-83-Cv, Mie/41/2002, or Muar, and their body weights were recorded ( Figures  3 and S1). We previously reported that the neuroinvasiveness of Muar was significantly higher than that of Mie/41/2002 in mice, even though Muar grew slower than Mie/41/2002 in murine neuroblastoma cells, as shown in Figure 2D [27]. In the group infected with 10 5 PFU, all the mice infected with Muar died, and five out of six (83.3%) 19Cx-83-Cv-infected mice died by seven days after inoculation, although only two out of six (33.3%) Mie/41/2002-infected mice died by 10 days post-infection ( Figure 3A). The survival curve for the 19CxBa-83-Cv-infected group was similar to that of the Muar-infected group. In the group infected with 10 4 PFU, two (33.3%) and one (16.7%) out of six 19CxBa-83-Cv-infected mice and six Mie/41/2002-infected mice died within 10 days post-infection, respectively, whereas all Muar-infected mice died by seven days post-infection ( Figure 3B). In the group infected with 10 3 PFU, all mice infected with the 19CxBa-83-Cv or Mie/41/2002 survived, but one out of six (16.7%) Muar-infected mice died throughout the observation period ( Figure 3C). The survival curves of the 19CxBa-83-Cv-infected groups were similar to those of the Mie/41/2002-infected groups in mice infected with 10 4 PFU and 10 3 PFU.  To evaluate the neurovirulence of 19CxBa-83-Cv in mice, the mice were infected i.c. with 19CxBa-83-Cv, Mie/41/2002, or Muar, and their body weights were recorded (Figure 4 and Figure S2). In the groups infected with 3 × 10 3 PFU, all mice infected with 19CxBa-83-Cv, Mie/41/2002, or Muar died within six days after infection ( Figure 4A). In contrast, all mice infected with 3 × 10 2 PFU of 19CxBa-83-Cv or Mie/41/2002 survived, and one out of six (16.7%) mice infected with Muar died throughout the observation period ( Figure 4B). No significant differences were observed in the survival curves among the groups infected with 3 × 10 3 PFU and those infected with 3 × 10 2 PFU. (16.7%) mice infected with Muar died throughout the observation period ( Figure 4B). No significant differences were observed in the survival curves among the groups infected with 3 × 10 3 PFU and those infected with 3 × 10 2 PFU.

Neutralizing Ability of Sera from the 19CxBa-83-Cv-Infected Mice
Serum samples were collected from the surviving mice at the end of the observation period (20 days post-infection) in the neuroinvasive experiment described in Figure 3. The sera were subjected to neutralization analysis against 19CxBa-83-Cv, Mie/41/2002, and Muar (Table 1)  1 Sera were collected from the survived mice at 20 days after inoculation of the JEVs described in Figure 3.

Neutralizing Ability of Sera from the 19CxBa-83-Cv-Infected Mice
Serum samples were collected from the surviving mice at the end of the observation period (20 days post-infection) in the neuroinvasive experiment described in Figure 3. The sera were subjected to neutralization analysis against 19CxBa-83-Cv, Mie/41/2002, and Muar (Table 1). Sera from mice inoculated with 10 5 or 10 4 PFU 19CxBa-83-Cv showed PRNT 50   1 Sera were collected from the survived mice at 20 days after inoculation of the JEVs described in Figure 3.

Neutralizing Ability of Pooled Serum from Mice Vaccinated with Vero Cell-Derived Inactivated JE Vaccine against the 19CxBa-83-Cv
We wished to evaluate the ability of the Vero cell-derived inactivated JE vaccine, which was produced from the GIII JEV Beijing-1 strain in Japan, to induce neutralizing antibodies against 19CxBa-83-Cv. Pooled mouse serum JSS-2020, which was prepared from mice immunized twice with the vaccine, was used for PRNT against 19CxBa-83-Cv, Mie/41/2002, and Muar ( Table 2). The PRNT 50 titer of the serum against Mie/41/2002 was 1:160, which was higher than those against 19CxBa-83-Cv (1:20) and Muar (1:40).

Neutralizing Ability of Sera from JE Patients against 19CxBa-83-Cv
Five sera samples from four autochthonous JE patients were collected to assess the neutralizing ability against 19Cx-Ba-83-Cv, Mie/41/2002, and Muar ( Table 3). The nucleotide sequences of the E region of the JEV genome amplified from the samples of the patients indicated that all JE patients were infected with GI JEV. The PRNT 50 titers against the three viruses were similar in three of the five sera. In the remaining two sera, the titers against GI Mie/41/2002 were four-to sixteen-fold higher than those against GIV 19CxBa-83-Cv and GV Muar.  1  25  2560  2560  2560  2  7  160  80  80  16  5120  5120  5120  3  10  640  160  160  4 10 2560 320 160 1 Sera were collected from four domestic JE patients diagnosed by IgM ELISA, real-time RT-PCR, and PRNT 50 methods.

Discussion
The first case of JE in humans caused by GIV JEV infection was reported in 2019 [31]. The endemics of JE in Australia in 2021 and 2022 were also caused by GIV JEV [38][39][40]. However, only a few GIV JEV have been isolated before the epidemics, little attention has been paid to GIV JEV, and GIV JEV has not been well characterized. In this study, we characterized the recently isolated GIV JEV strain 19CxBa-83-Cv [30].
The phylogenetic tree in Figure 1 shows that the 2021-2022 Australian strains were located between the 2017-2019 Bali group, which contains 19CxBa-83-Cv, and the 1980s Indonesian group (JKT6468), suggesting that the Australian endemic strains may not originate from the recent Bali group. The E protein of 19CxBa-83-Cv differs by only one amino acid residue (0.2%) from that of the Bali2019 strain, which was isolated from an Australian JE patient returning from Bali in 2019, whereas the strain differs by eight residues (1.6%) from that of the 2021-2022 Australian strains ( Figure S3). The Australian strains also have some residues unique to the group in the E protein [40]. Further genomic analyses of more GIV strains will help us to understand the meaning of the variations in amino acid sequences among GIV strains.
The growth potential and plaque size of 19CxBa-83-Cv in Vero cells were intermediate between those of GI strain Mie/41/2002 and GV strain Muar. The three strains showed similar growth properties in human neuroblastoma-derived IMR-32 cells, whereas Muar grew notably slower than the other two strains, which showed similar growth patterns in mouse neuroblastoma Neuro-2a cells. We previously reported that Muar exhibits slower growth kinetics than Mie/41/2002 in mouse neuroblastoma cells and that an amino acid residue in the non-structural protein NS2A (NS2A 166 ) is crucial for the growth of GI and GV JEV in the cells [27,29]. The residue NS2A 166 in Muar is His, but in Mie/41/2002 and 19CxBa-83-Cv, it is Tyr, suggesting that the Tyr residue NS2A 166 in 19CxBa-83-Cv contributes to the efficient replication ability of 19CxBa-83-Cv in Neuro-2a.
The neuroinvasiveness of 19CxBa-83-Cv in mice was higher than that of Mie/41/2002 but lower than that of Muar. It is known that the E protein of JEV is the main factor determining its pathogenicity [49][50][51][52]. Our previous findings indicate that the amino acid at position 123 in the E protein (E 123 ) is involved in the neuroinvasiveness of JEV in mice. JEV with Ser at E123 shows lower neuroinvasiveness than JEV with Arg and His at this position [28,45]. His at E 123 is highly conserved in GV JEV, including Muar, whereas Mie/41/2002 and 19CxBa-83-Cv have Ser at this position ( Figure S4). In our results, 19CxBa-83-Cv exhibited higher neuroinvasiveness in mice than Mie/41/2002, indicating that E123 is not associated with 19CxBa-83-Cv in neuro-invasiveness in mice. Twentythree residues in the E protein differed between 19CxBa-83-Cv and Mie/41/2002, and these sites may include amino acid residues responsible for the difference in pathogenicity between the strains. Alternatively, it is also possible that viral proteins other than the E protein are involved in the variation in virulence. The establishment of reverse genetics for GIV JEV is needed to analyze the pathological mechanisms of GIV JEV. There were no significant differences in neurovirulence in mice among the three strains, implying that the direct ability of the strains to damage the brain is not involved in the differences in neuroinvasiveness among the viruses. The ability of the strains to proliferate in the peripheral tissues of mice may be important for determining the pathogenicity of the viruses.
The neutralizing titers of sera from the 19CxBa-83-Cv-and Mie/41/2002-inoculated mice against 19CxBa-83-Cv and Mie/41/2002 were similar, whereas the titers against Muar were four-fold lower than those against the other two viruses in most mice. These data suggest that 19CxBa-83-Cv is serologically closer to the GI strain Mie/41/2002 than to GV Muar. A comparison of the amino acid sequences of the E protein showed that the identity between 19CxBa-83-Cv and Mie/41/2002 was 95.4%, whereas that between 19CxBa-83-Cv and Muar was 90.6% ( Figure S4), supporting the idea that 19CxBa-83-Cv is closer to Mie/41/2002 than to Muar. GIII strains were widely and frequently isolated in most JE endemic areas until the 1990s; therefore, licensed JE vaccines are produced from GIII strains [53][54][55]. A previous report indicated that the chimeric JE vaccine induced protective levels of neutralizing antibodies against the classical GIV strain 9092 [56]. We investigated the neutralizing potency of the GIII Beijing-1-based inactivated JE vaccine against the three strains using mice and found that the titers of pooled sera from vaccinated mice against 19CxBa-83-Cv and Muar were lower than those against Mie/41/2002. In the E protein, there were only eight residues (1.6%) that differed between Beijing-1 and Mie/41/2002, but 26 residues (5.2%) and 42 residues (8.4%) differed between Beijing-1 and 19CxBa-83-Cv and between Beijing-1 and Muar, respectively ( Figure S4), implying that the differences in the amino acid sequences of the structural proteins may influence the neutralization potency of JE vaccines.
The major genotype of JEV isolated in Japan is GI [48,57,58]. In this study, we used five serum samples from four Japanese JE patients who were infected with GI JEV in Japan to evaluate the reactivity of the sera to the three strains. In three of the five serum samples, no clear differences in the neutralizing titers against the strain were observed. In contrast, in the other two serum samples, the titers against 19CxBa-83-Cv and Muar were four-to sixteen-fold lower than those against Mie/41/2002. Our previous reports showed that the ratio of the neutralizing titer against Muar to that against Mie/41/2002 was equivalent to or less than 1:2 in most JE patient sera in Japan and northern Vietnam, where only GI JEV has been identified in recent years [59,60]. These findings are partially inconsistent with the data from JEV-infected mouse samples described above. The reasons for this remain unknown.
In this study, we uncovered the basic characteristics of the recent GIV JEV strain 19CxBa-83-Cv by comparing the virus with well-characterized GI and GV JEV strains. However, there are certain differences in the amino acid sequences between the 2017-2019 Bali group and the 2021-2022 Australian group, which might critically affect the properties of the viruses. Therefore, it is essential to conduct similar analyses of other GIV isolates and accumulate information on the isolates in the future to understand the nature of GIV JEV.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/v15010239/s1, Figure S1: Body weight of mice inoculated with JEV strains as shown in Figure 3. Figure S2: Body weight of mice inoculated with JEV strains as shown in Figure 4. Figure S3: Comparison of the complete amino acid sequence of E protein (500 residues) in GIV JEV strains. Figure S4  Institutional Review Board Statement: The study was conducted in accordance with the Declaration of Helsinki and approved on 21 September 2022, by the Ethical Committee for Biomedical Science at the National Institute of Infectious Diseases (NIID) in Japan (no. 1449). Animal experiments were performed in accordance with the Guidelines for Animal Experiments Performed at the NIID, under approval on 14 July 2022 from the Animal Welfare and Animal Care Committee of the NIID, Japan (no. 122099). All efforts were made to minimize pain and distress. Mice infected with JEV were observed daily for adverse reactions and signs of disease. For the collection of organ samples, mice were euthanized using isoflurane.
Informed Consent Statement: Patient consent was waived due to the use of human specimens that were remnants of the laboratory diagnosis. This study did not involve identifiable private information or identifiable specimens.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.