Development of a novel antigen for hemagglutination inhibition of Japanese encephalitis virus genotype 5
Article information
Abstract
Japanese encephalitis virus (JEV) is a zoonotic arbovirus transmitted by mosquitoes and poses a significant public health concern. JEV is classified into 5 genotypes (G1–G5). Since 2010, JEV genotype 5 (G5) has been predominant in the Republic of Korea; however, a specific antigen for hemagglutination (HA) inhibition (HI) assays has been lacking. This study aimed to develop a novel antigen to enable genotype-specific HI antibody detection for JEV G5. The Sangju strain of JEV G5 was serially passaged 30 times in Vero and C6/36 cells, resulting in Sangju-30, which exhibited 2 envelope protein mutations and an HA titer of 8, increasing to 512 following polyethylene glycol 8000 concentration. The antigen demonstrated a strong correlation (R = 0.95, p < 0.05, 95% confidence interval, 6.3–31.9) with HI antibody levels in pig sera. These findings suggest that the Sangju-30 antigen is a viable alternative for JEV G5 serological monitoring, enhancing diagnostic accuracy and offering a replacement for traditional murine-derived antigen production.
Introduction
Japanese encephalitis virus (JEV), a mosquito-borne flavivirus in the Flaviviridae family, poses significant veterinary and public health concerns. JEV infects a variety of hosts, including pigs, horses, chickens, and wild birds, with its transmission cycle primarily involving Culex mosquitoes. Pigs serve as amplifying hosts, supporting viral replication following mosquito bites and leading to viremia. Subsequent feeding by mosquitoes on infected pigs facilitates viral transmission to other animals and humans, perpetuating the spread of the disease [1]. Pigs, particularly pregnant sows, may suffer reproductive disorders such as abortions and stillbirths upon JEV infection. While most horses experience asymptomatic infections, approximately 5% develop encephalitis, which can prove fatal [1]. Given the critical role of pigs in viral transmission, vaccination of swine populations has been implemented as a preventive strategy. The live attenuated JEV vaccine used in animals in the Republic of Korea (ROK), developed in the 1980s, was derived from a pig-isolated JEV strain that underwent 300 serial passages, resulting in the Anyang 300 strain [2]. Korean veterinary authorities have mandated the annual vaccination of sows and fattening pigs prior to mosquito season, which has effectively prevented JEV cases in pigs since 2008. Despite these efforts, human infections continue to be reported annually [3].
JEV is genetically classified into 5 genotypes (G1–G5) based on its envelope (E) gene or full genome sequences [4]. Until 1990, JEV G3 was the predominant strain in the ROK. However, in the post-1990, JEV G1 replaced G3, and since 2010, JEV G5 has been the primary circulating strain in the ROK [5,6]. Notably, experimental studies in mice indicate that JEV G5 exhibits higher pathogenicity than JEV G1 and G3 [7]. Current vaccines and serological assays primarily target JEV G1 and G3, raising concerns regarding potential diagnostic and surveillance gaps for JEV G5 infections [5]. Despite the increasing prevalence of JEV G5 in the ROK, no commercial serological assays are currently available for detecting JEV G5-specific antibodies in humans or animals.
Several serological techniques, including the virus neutralization test, enzyme-linked immunosorbent assay (ELISA), and hemagglutination (HA) inhibition (HI) assay, are employed to detect JEV antibodies in serum samples [8–10]. Among these, the HI assay is a cost-effective method frequently used for detecting antibodies in thoracic fluid from aborted pig fetuses [11]. Traditional HI antigen preparation for JEV G3 involves inoculating suckling mice with JEV and harvesting brain homogenates post-mortem [12]. However, this method poses several limitations, including biosafety risks, animal welfare issues, and environmental contamination due to acetone usage.
Given the dominance of JEV G5 in the ROK since 2010, the absence of a standardized HI assay for detecting JEV G5 antibodies represents a critical gap in disease surveillance and control. Therefore, the development of a JEV G5-specific antigen is essential for strengthening veterinary and public health measures. In this study, we introduce a novel antigen for HI antibody detection that specifically targets JEV G5.
Materials and Methods
Ethical approval
All animal experiments were conducted with ethical approval from the Animal and Plant Quarantine Agency of Korea (approval number: 2024–756).
Virus strains and cell culture
The JEV G5 strain, designated as the Sangju strain (NCCP-490200), was provided by the Korea Disease Control and Prevention Agency. This strain was originally isolated from the Culex orientalis mosquito in the ROK in 2020. The JEV G1 strain, KV1899 (GenBank: AY316157), was isolated from the blood of Korean pigs in 1999. Vero cells (African green monkey kidney cell line [ATCC]: CCL-81), BHK-21 cells (ATCC: C-13), LFBK cells (CVCL-RX27), and NG108-15 cells (ATCC: HB-12317) were cultured in Dulbecco’s modified Eagle medium supplemented with 10% fetal bovine serum and 1% antibiotics. The cells were maintained at 37°C in a humidified atmosphere containing 5% CO₂. C6/36 cells (Aedes albopictus larval cell line, ATCC: CRL-1660) were cultured under identical conditions, except at an incubation temperature of 28°C. Sf-9 cells (Spodoptera frugiperda ovarian tissue, ATCC: CRL-1711) were cultured in Grace’s medium (Gibco BRL, USA) at 28°C in a non-CO₂ environment.
Passaging and characterization of the Sangju-30 strain
The Sangju strain was subjected to 30 sequential passages using a limiting dilution method, alternating between Vero and C6/36 cells. The resulting strain, designated Sangju-30, was deposited in the Korean Veterinary Culture Collection (KVCC; VR2500011). To evaluate viral propagation efficiency, the virus titer of Sangju-30 was measured and compared with that of the original Sangju strain in Vero cells. Sangju-30 was inoculated into Vero, BHK-21, LFBK, NG108-15, C6/36, and Sf-9 cells, and HA titers were evaluated. The whole-genome nucleotide sequence of Sangju-30 was determined using next-generation sequencing (NGS; Sanigen, Korea) and compared with the genome of the parental Sangju strain.
Hemagglutination inhibition test
Inactivated pig sera were used in the HI test. The KV1899 (G1 strain) and Sangju-30P (G5 strain) of JEV were used as the antigens for the HI test. An HI test was performed in 96-well microplates, using slightly modified standard methods, to estimate the JEV antibody prevalence in the pig sera [13,14]. Using a sucrose/acetone extraction method, JEV G1 antigen was prepared from the brains of suckling mice infected with the KV1899 strain. JEV G5 antigen was prepared by concentrating on the virus obtained from Vero cells infected with Sangju-30 strain. As an HI test procedure, 100 µL of serum were mixed with 900 µL of 14.5% kaolin (Sigma, USA) to remove non-specific inhibitors and incubated for 30 minutes. After pipetting, the kaolin was removed by centrifugation at 3,600 rpm for 15 minutes. The resultant clear supernatant was mixed with 20 µL of packed goose erythrocytes to remove any natural agglutinins. After incubation at 37°C for 1 hour, the treated sera were separated from the goose erythrocytes by centrifugation. The treated sera (25 µL) were diluted 2-fold from 1:20 to 1:10,240 in round-bottom 96-well microplates and reacted with 8 HA units of JEV antigen. After incubation at 37°C for 1 hour, 50 µL of 0.33% goose erythrocytes was added, and the microplates were incubated for 30 min at 37°C. To confirm test reliability, positive and negative JEV infection pig control sera were used in HI test. HI titer was expressed as the reciprocal of the highest dilution of serum showing complete HI.
Propagation and inactivation of the Sangju-30 strain
The Sangju-30 strain, used for HI testing, was propagated in Vero cells under standard culture conditions. When approximately 90% cytopathic effects were observed, infected cell cultures in 175 cm² flasks were subjected to 3 freeze–thaw cycles. The harvested supernatant was inactivated at 37°C for 10 h using either 2 mM binary ethylenimine (BEI) or 0.1% formaldehyde, in accordance with established protocols [15,16]. To confirm complete virus inactivation, the treated supernatant was dialyzed and subsequently inoculated into Vero cells, with no viral replication detected. The inactivated Sangju-30 antigen was then concentrated 50-fold or 100-fold using polyethylene glycol 8000 (PEG 8000; Sigma-Aldrich) for further evaluation. The HA titers of the unconcentrated and concentrated antigens were measured.
Preparation of inactivated JEV G1 and G5 vaccines
To evaluate the immunogenicity of Sangju-30, the KV1899 and Sangju-30 strains were propagated in Vero cells and inactivated using 2 mM BEI to prepare 2 types of test vaccines. Aluminum hydroxide gel (Rehydragel LV; Chemtrade Logistics, Canada) was used as an adjuvant and mixed with the JEV antigens at a 9:1 ratio.
Preparation of pig sera specific for JEV G1 and G5
A total of 14 eight-week-old, JEV antibody-negative pigs were purchased and divided into 3 experimental groups: group 1 (n = 6), vaccinated twice with inactivated JEV G1 vaccine at two-week intervals; group 2 (n = 6), vaccinated twice with inactivated JEV G5 vaccine at two-week intervals; and group 3 (n = 2), a non-vaccinated control group. Blood samples were collected biweekly from all animals. Serum samples were isolated and subjected to HI antibody titration using the KV1899 and Sangju-30 antigens.
Statistical analysis
Least-squares linear regression analysis was conducted to determine the correlation coefficients (R values) between the HI titers of the 2 JEV genotypes. The p-value and the 95% confidence interval (CI) between HI titers were calculated using the Wilcoxon matched pair’s test. All statistical calculations were performed using Excel ver. 2010 (Microsoft Corp., USA).
Results
Genomic characterization of Sangju-30
The complete genome sequence of the Sangju-30 strain was confirmed and compared with that of the parental Sangju strain. The Sangju-30 genome contained 11,900 nucleotides, with no detectable structural defects. Comparative sequence analysis identified 5 amino acid substitutions located in the E, NS3, NS4B, and NS5 genes, which may contribute to phenotypic differences between the strains (Table 1).
Amino acid mutations identified in viral proteins of the Japanese encephalitis virus Sangju-30 strain
Growth efficiency and HA activity of Sangju-30
The replication efficiency of the Sangju-30 strain was evaluated by measuring the 50% tissue culture infective dose (TCID50/mL). The strain demonstrated a viral titer of 107.0 TCID50/mL, which was 17.8 times higher than the 105.6 TCID50/mL observed for the parental Sangju strain (Fig. 1). Evaluation of viral proliferation in 6 different cell lines revealed that Vero cells supported the highest level of Sangju-30 replication. HA activity against goose erythrocytes was evaluated for Sangju-30 propagated in each of the 6 cell types. Only Sangju-30 propagated in Vero cells exhibited HA activity, with a titer of 8, while no HA was observed in BHK-21, LFBK, NG108-15, C6/36, or Sf-9 cells (Fig. 2).
Comparison of viral titers between the parental Sangju strain and the passed Sangju-30 strain in Vero cells. P indicates the number of passages. TCID50, 50% tissue culture infective dose.
Virus and hemagglutination (HA) titers of the Sangju-30 strain propagated in 6 different cell lines. TCID, tissue culture infective dose.
Effect of inactivation agents on HA activity
To evaluate the impact of chemical inactivation on HA activity, 200 mL of Sangju-30 virus propagated in Vero cells was treated with either 2 mM BEI or 0.1% formaldehyde. The BEI-inactivated Sangju-30 antigen retained an HA titer of 8, whereas the formaldehyde-inactivated antigen exhibited no detectable HA activity (Table 2).
Hemagglutination (HA) titer of the Sangju-30 strain based on the inactivation agent
Enhancement of HA activity through virus concentration
Following the selection of BEI as the preferred inactivation agent, Sangju-30 was concentrated 50-fold and 100-fold using PEG 8000 to enhance HA activity. The concentrated virus demonstrated HA titers of 256 and 512, respectively, compared with the unconcentrated virus and control (Fig. 3).
Hemagglutination (HA) titers of the Sangju-30 strain propagated in Vero cells and concentrated 50-fold and 100-fold using polyethylene glycol 8000. The HA titers of the unconcentrated and 50-fold and 100-fold concentrated Sangju-30 samples were 8, 256, and 512, respectively.
Serological evaluation using Sangju-30 antigen
The HI titers of JEV G1 and G5 antigens were analyzed in serum samples obtained from pigs vaccinated with inactivated JEV G1 and G5 test vaccines, as well as in control group samples. In serum collected from JEV G1-vaccinated pigs, the HI titers of JEV G1 and G5 antigens exhibited a correlation coefficient R = 0.90 (p < 0.5, 95% CI, 11.2–35.7) Similarly, in serum from JEV G5-vaccinated pigs, R = 0.95 (p < 0.05, 95% CI, 6.3–31.9) was observed between the HI titers of JEV G1 and G5 antigens (Fig. 4).
Correlation between hemagglutination inhibition (HI) antibody titers against Japanese encephalitis virus (JEV) genotype (G) 1 and 5 antigens in sera collected from pigs vaccinated with JEV G1 (A) and JEV G5 (B) vaccines. The correlation coefficient (r) between HI titers of JEV G1 and G5 antigens in (A) and (B) was 0.90 and 0.95, respectively.
Discussion
JEV remains endemic across Asia, with genotype prevalence shifting over time owing to various ecological and evolutionary pressures. In the ROK, the dominant JEV genotypes have transitioned from G3 to G1 and, more recently, to G5 since the virus was first identified in 1947 [17]. These genotype shifts are influenced by factors such as changes in mosquito vector populations, host immune dynamics, and environmental conditions. The replacement of G3 by G1 has been attributed to enhanced adaptability in avian hosts, allowing for greater transmission efficiency [18,19]. Similarly, the emergence of JEV G4 in Australia during 2021 and 2022 was likely driven by ecological changes, including alterations in mosquito populations and climatic conditions [20]. This emergence was further supported by viral adaptation to new reservoir hosts, such as feral pigs and ardeid birds. The genotype shift from G1 to G5 in the ROK after 2010 is also believed to be associated with changes in the vector population and viral adaptability [21]. Given the current predominance of JEV G5 in the ROK, the development of a JEV G5-specific antigen is critical for improving serological diagnostics in veterinary settings.
Repeated passaging of JEV in cell culture facilitates viral adaptation, primarily by accumulating mutations, particularly in the E protein, a key determinant of neurovirulence. Previous studies have identified residue E138 as a critical site influencing neurovirulence, with acidic amino acid residues enhancing virulence and basic residues contributing to attenuation [22]. In addition, passage-induced mutations can enhance viral replication by improving host cell membrane interactions and altering replication organelle formation [23]. In this study, serial passaging of the Sangju strain in Vero and C6/36 cells resulted in 2 amino acid substitutions in the E protein, likely contributing to a 17.8-fold increase in viral replication efficiency. The enhanced replication capacity may also explain the increased HA activity of the Sangju-30 strain in goose erythrocytes. Notably, Sangju-30 propagated in Vero cells exhibited significantly higher viral titers compared to other cell types, suggesting that host cell adaptation plays a pivotal role in replication dynamics. These findings provide critical insights into how passage-induced mutations in the E protein influence JEV replication efficiency and may inform future research on viral adaptation and virulence.
The method of JEV inactivation plays a crucial role in preserving antigenic integrity. BEI inactivation maintains the native structural conformation of the E protein, thereby preserving HA activity in goose erythrocytes. In contrast, formalin inactivation induces extensive cross-linking of viral proteins, which can disrupt the E protein structure and abolish its HA activity. Our previous study reported that BEI-inactivated JEV retained high HA titers (≥ 2,048), whereas formalin-inactivated JEV G1 showed no detectable HA activity [12]. In this study, propagation and concentration of the Sangju-30 strain in Vero cells confirmed robust HA activity, supporting the utility of Vero cell-derived antigen as a viable alternative to traditional sucrose-acetone extraction methods [24]. However, additional studies are required to evaluate the consistency and efficacy of this approach across the 5 JEV genotypes.
Despite genomic diversity among JEV G1–G5, JEV is considered a single serotype owing to conserved immunodominant epitopes, strong cross-neutralization among genotypes, and minimal antigenic variation affecting immune recognition [19]. In this study, the strong correlation (r ≥ 0.9) observed between HI antibody responses to JEV G1 and G5 antigens indicates substantial serological cross-reactivity. This finding aligns with those of previous studies demonstrating broad immunogenicity of JEV, wherein antibodies elicited by one genotype can effectively neutralize heterologous strains [25]. For instance, JEV G3-based vaccines have been shown to induce cross-protective immunity against multiple genotypes, reinforcing the significance of cross-reactive antibody responses. The HI assays revealed that the antibodies elicited by the JEV G5 antigen exhibited a higher correlation (R = 0.95, p < 0.05) with the homologous antigen than with those induced by the JEV G1 antigen, suggesting that JEV G5 may possess antigenic features conducive to broader cross-protection. Given its predominance in the ROK, JEV antibody detection assays should incorporate the JEV G5 strain to enhance the sensitivity and relevance of serological surveillance.
In conclusion, we successfully developed a novel Sangju-30 antigen through serial passaging of the JEV G5 Sangju strain in Vero and C6/36 cells, achieving an HA titer of 512 following PEG8000 concentration. Genomic analysis revealed 2 key amino acid substitutions in the E protein, which may contribute to its immunogenic properties. The Sangju-30 antigen demonstrated a strong correlation (R = 0.95) with HI antibody responses in pig sera. These findings suggest that the antigen represents a robust and precise alternative to traditional murine-derived antigens for efficient serological monitoring of JEV. Future studies should focus on developing a high-throughput ELISA method to facilitate large-scale surveillance of JEV G5-specific antibodies.
Notes
The authors declare no conflict of interest.
Author’s Contributions
Conceptualization: Yang DK: Data curation: Yang DK; Formal analysis: Yang DK; Funding acquisition: Cho YS, Yang DK; Investigation: Yang DK, Methodology: Lee JY, Lee HJ, Kwon GH, Park GN; Project administration: Yang DK, Software: Yang DK, Lee HJ; Validation: Cho YS; Writing–original draft: Yang DK; Writing–review & editing: Yang DK, Lee JY, Lee HJ, Cho YS.
Funding
This study was supported financially by a grant from Animal and Plant Quarantine Agency, Ministry of Agriculture, Food and Rural Affairs (MAFRA), Republic of Korea (N-1549085-2017-36-01).