Low Protective Efficacy of the Current Japanese Encephalitis Vaccine against the Emerging Genotype 5 Japanese Encephalitis Virus

Background The current Japanese encephalitis (JE) vaccine derived from G3 JE virus (JEV) can induce protective immunity against G1–G4 JEV genotypes. However, protective efficacy against the emerging G5 genotype has not been reported. Methods/Principal Findings Using in vitro and in vivo tests, biological phenotype and cross-immunoreactions were compared between G3 JEV and G5 JEV (wild strains). The PRNT90 method was used to detect neutralizing antibodies against different genotypes of JEV in JE vaccine-immunized subjects and JE patients. In JE vaccine-immunized mice, the lethal challenge protection rates against G3 and G5 JEV wild strains were 100% and 50%, respectively. The seroconversion rates (SCRs) of virus antibodies against G3 and G5 JEV among vaccinated healthy subjects were 100% and 35%, respectively. All clinically identified JE patients showed high levels of G3 JEV neutralizing antibodies (≥1:10–1280) with positive serum geometric mean titers (GMTs) of 43.2, while for G5 JEV, neutralizing antibody conversion rates were only 64% with positive serum GMTs of 11.14. Moreover, the positive rate of JEV neutralizing antibodies against G5 JEV in pediatric patients was lower than in adults. Conclusions/Significance Low levels of neutralizing/protective antibodies induced by the current JE vaccine, based on the G3 genotype, were observed against the emerging G5 JEV genotype. Our results demonstrate the need for more detailed studies to reevaluate whether or not the apparent emergence of G5 JEV can be attributed to failure of the current vaccine to induce appropriate immune protectivity against this genotype of JEV.


Introduction
Japanese encephalitis (JE), probably the world's most frequently occurring viral encephalitis, is a neurological infectious disease caused by Japanese encephalitis virus (JEV), transmitted via mosquito bite [1]. The mortality rate of JE patients is 30-35% and 50% of JE survivors live with neurological sequelae. Therefore, JE is considered a disease with significant public health and economic burdens [2,3]. JE, prevalent in 24 Asia-Pacific countries, is a mosquito-borne zoonotic natural focal disease with an estimated 67,900 cases each year. Approximately, 30 million people live in JE-endemic areas [4]. In addition, with increased international travel and business to these areas, more people are at risk of JE infection [5,6], presenting a potentially serious international public health problem.
Since there is no effective therapeutic antiviral treatment, JE vaccination is the most effective prevention and control measure [7]. Studies have shown that the current vaccine derived from the G3 JEV can protect against G1-G4 JEV infections [8]. However, G5 JEV, first discovered in 1951, has not been epidemic over the last 60 years but reemerged in 2009 and simultaneously occurred on the Asian continent in East Asia (Korea) and southern Asia (Tibet) [9,10,11], presenting new challenges for JE prevention and control. Importantly, G5 JEV and G1-G4 JEV differ significantly in their molecular biological characteristics [12,13,14]. Nevertheless, a recent report demonstrated protective effects of the current JE vaccine against G5 JEV in animals [14]. However, whether or not humans immunized with the current G3 JEV-based vaccine can produce protective antibodies against the emerging G5 JEV is currently unknown. Moreover, it is not known if JE-infected patients have protective antibodies against G5 JEV.
Here, using wild-type strains of G3 and G5 JEV genotypes, we report comparative studies of neutralizing antibody levels against G5 JEV in both JE-vaccinated subjects and clinically diagnosed JE patients.

Virus
All JEV strains used in this study were isolated from field samples using cell culture and stored at low passage level in our laboratory: Department of Viral Encephalitis and Arbovirus, Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention (IVDC, China CDC). G1 JEV was GZ56 strain (GenBank: HM366552) [15], G3 JEV was P3 strain (GenBank: JEU47032) [16], and G5 JEV was XZ0934 strain (GenBank: JF915894.1) [9].

Measurement of virus multiplication (plaque assay)
BHK-21 cells were inoculated at a multiplicity of infection (MOI) of 0.01 and the cytopathic effects (CPE) of the cells were observed. Simultaneously, aliquots of the infected media were collected every 12 h and measured using plaque assays to generate a viral reproduction curve [14]. Aliquots of 10-fold diluted virus suspension (from 10 −1 to 10 −6 ) were added to BHK-21 cells in 6-well plates (0.1 mL/well). After 1 h of adsorption at 37°C, the cells were overlaid with 1.3% methylcellulose-MEM medium containing 2% FBS (5 mL/well). After culturing for 3~5 days, the cells were stained with crystal violet and virus titres expressed as plaque-forming units (pfu) were calculated. For each virus, the diameter of 8 plaques was measured and the mean plaque size in mm and standard error (SEM) was calculated.

Serum samples
Serum samples from JE-vaccinated children (3 mL of blood) were collected from two-year-old children 28 days before and after vaccination SA14-14-2 JE live attenuated vaccine (LAV), (Wuhan Institute of Biological Products Co. Ltd Liability, Wuhan, China). Serum samples (5 mL of blood) were collected from patients diagnosed with JE. All serum samples were obtained from whole blood after clotting at room temperature and centrifuged at 5,000 rpm for 5 min. Aliquots of serum were stored at -80°C until use.

Ethics statement
The implementation agencies of the present study were the National Reference Laboratory of Japanese Encephalitis and the World Health Organization JE Regional Reference Laboratory, China (WHO-JE-RRL, China). Therefore, all the human serum samples used in this study were from those collected and maintained during the JE epidemiological monitoring period by these JE reference laboratories. All adult subjects provided written informed consent, and a parent or guardian of any child participant provided informed consent on their behalf. This research project was approved by the Ethics Committee of the IVDC, China CDC (No.2014112509).

Detection of IgM antibodies against JEV
Two kits were used to duplicate test serum samples collected from JE patients: the JE-Dengue IgM Combo ELISA (PanBio, Brisbane, Australia), and the JEV IgM-Capture ELISA kit (Shanghai B & C Enterprise Development Co. Ltd, Shanghai, People's Republic of China). Methods and result interpretation were performed according to manufacturers' instructions [17]. Both detection kits were Capture ELISA.

Animal experiments
BALB/c mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. (quality qualifier SYXK Beijing 2012-0022) and were female except for the neonatal mice. All mice were housed under pathogen-free conditions at the China CDC animal facility. All animal experiments were conducted in strict compliance with the regulations set by the Animal Ethics Committee of China CDC (No.2014112509).
Lethality of viruses to mice. BALB/c neonatal mice (1-2 days old, 12 mice/group) were intracerebrally (i.c.) inoculated with 1×10 5 pfu/mL JEV (0.01 mL/each); 5~6-week-old BALB/c (5 mice/group) were intraperitoneally (i.p.) inoculated with 10-fold diluted JEV (from 10 −1~1 0 −6 , 0.03 mL/each), followed by a sham i.c. injection. The incidence and mortality of mice were observed and recorded every day and the Reed-Muench method was used to calculate the 50% lethal dose (LD 50 ) of JEV to adult mice. Viral immunology. G3 and G5 JEV (2000 pfu/300 μL each) were thermally inactivated at 56°C for 30 min and then 4-week-old BALB/c mice were separately immunized (7 mice/group) by multiple subcutaneous injections (300 μL/each) with an equal volume of mixed Freund's complete adjuvant at day 0 followed by direct i.p. injection (300 μL/each) every two weeks for a total of three times. Two weeks after the third immunization, sera were collected to test the cross-immunoreaction between G3 and G5 JEV.

Neutralization tests
Serum neutralizing antibody to JEV was detected using the plaque reduction neutralization test (PRNT) method. The inactivated serum was diluted two-fold (from 1:5 to 1:1280) and mixed with an equal volume of 200 pfu JEV, incubated in 37°C for 1 h. The mixture was then added to 6-well BHK-21 plates for 1 h followed by overlaying with 1.3% methylcellulose-MEM medium containing 2% FBS for 3~5 days. The plaques were stained with crystal violet and counted. Neutralizing antibody titers were calculated as the reciprocal of the highest dilution resulting in 90% plaque reduction (PRNT 90 ) compared to an unvaccinated control serum [19].

Statistical analyses
In this study, the positive cut-off value of neutralizing antibody titer was defined as PRNT 90 ≧1:10. Seropositive subjects were defined as those with a titer above or equal to the cut-off value and when negative, the result was given an arbitrary value of 5 for calculating the geometric mean titer (GMT). All data were processed and drawn using GraphPad Prism 5.0 software (GraphPad, La Jolla, CA, USA), Student's t-tests were used for all analyses and a P-value <0.05 was considered statistically significant.

Results
General phenotypic characteristics of G5 and G3 JEV CPE due to virus infection showed that G5 JEV (XZ0934 strain) and G3 JEV (P3 strain) can cause BHK cells to shrink and shed (Fig 1A). The plaque diameter of G5 JEV was smaller than that of G3 JEV (Fig 1B). Virus infectivity generated from BHK-21 cells inoculated with MOI of 0.01 showed that G3 JEV and G5 JEV proliferated and reached their highest titers in 48 h and 60 h, respectively (Table 1 and Fig 1C), indicating that the two viruses had substantially similar multiplication capacities in BHK-21 cells.
The 1~2-day-old BALB/c neonatal mice i.c. inoculated with the two JEV genotypes (10 3 pfu/mouse; 12/group) all died in the first 5 days. The 5~6-week-old BALB/c adult mice were i. p. inoculated with the two JEV genotypes (5/group) and the LD 50 of each virus was calculated to assess the virulence of JEV. The LD 50 values of the G5 and G3 virus titers (logarithmic) were 3.5 and 4.1, respectively, demonstrating no significant difference in virulence (Table 1).

Cross-immunoreaction between G3 and G5 JEV
To compare the relative immunogenicities between G3 and G5 JEV, we measured the degree of cross-immunoreaction between the two JEV genotypes using PRNT. The results showed that the PRNT 90 titer of G3 JEV immune serum against G3 JEV was higher than 1:320 and only 1:10 against G5 JEV. The PRNT 90 titer of G5 JEV immune serum against G5 JEV was 1:40, and 1:10 against G3 JEV (Table 1). Both JEV genotypes showed strong immunoreactions against Protective efficacy of the current JE vaccine against G5 JEV challenge Mice were i.p. injected with the JE vaccine at high, medium or low dilution LAV/SA14-14-2 at 2510 pfu, 251 pfu and 25 pfu, IPV/P3 at 1:5, 1:25 and 1:125, respectively. After 14 days from initial immunization, they were then challenged i.p. with 500 LD 50 of G5 or G3 JEV. The results showed that high and medium doses of LAV completely protected against G3 JEV challenge, but high, medium and low doses of LAV provided only a 50% protection rate against G5 JEV challenge. All IPV doses completely protected against challenge with G3 JEV, while protection rates against challenge with G5 JEV were 80% for high and medium doses in the IPV group (1:5 and 1:25 dilution, respectively) and 60% in the low dose group (1:125 dilution; Table 1 and Fig 2).
In serum samples after immunization with the current JE vaccine, neutralizing antibody titers estimated using PRNT 90 (expressed as GMTs) against G3 JEV and SCRs were analyzed  and the results showed that when neutralizing antibodies against G3 JEV serum were 1:40, SCRs against G1 JEV were as high as 100%, but none were seropositive against G5 JEV (0% SCRs). When neutralizing antibodies against G3 JEV serum were 1:160-1:320, SCRs against G5 JEV were only 50%. Protective antibodies were produced against G5 JEV (SCRs up to 100%) only when the dilution of G3 JEV serum was higher than 1:320. The results are shown in Table 2.

Detection of neutralizing antibody against G5 JEV in serum samples collected during acute phase from clinical JE patients
This study further examined levels of neutralizing antibody against G5 JEV in serum samples from clinically diagnosed JE patients. All 45 serum samples were confirmed IgM-positive when using two JEV IgM antibody detection kits. Neutralizing antibodies against G1, G3 and G5 JEV were detected in all 45 serum samples. The results showed that neutralizing antibodies ( 1:10-1280) against G3 JEV were detected in all serum samples from JE patients with serum GMT of 43.2. Neutralizing antibodies ( 1:10-80) against G1 JEV were detected in 87% (39/ 45) of serum samples with serum GMT of 16.37. Additionally, neutralizing antibodies against G5 JEV were detected in only 64% (29/45) of serum samples with serum GMT of 11.14. Evaluation of JE patient's age and neutralizing antibodies against G5 virus showed that 41% (9/22) of JE patients 1-15 years of age and 87% (20/23) of patients over 16 years of age had positive neutralizing antibodies against G5 JEV (Table 3).

Discussion
Poor cross-immunogenicity G5 JEV and G3 JEV wild-type isolates were used in this study. Comparative studies of wild strains showed that the biological characteristics of G5 JEV, such as CPE, virus titers and virus lethality to animals, were similar to those of G3 JEV. However, cross-neutralization studies showed that antigenic cross-reactivity between G3 and G5 JEV was low (Table 1) and this was also reflected in the low cross-protectivity studies. A study of a G5 JEV infectious molecular clone (the infectious molecular clone was prepared from the full-length cDNA of G5 JEV XZ0934 strain) showed similar results [20] as did the use of G5 JEV wild strains isolated from viral JE patients in 1951 [14]. Further animal experiments also showed that the protection capacity of the current JE vaccine against G5 JEV challenge on animals was only 50% (Table 1).
These results indicate that G3 and G5 JEV were significantly different in terms of their capacity to stimulate cross-protective immune responses to the heterologous virus. Our present study about molecular characterization of full-length genome of G5 JEV (SA14-14-2 strain) shows that G5 JEV is significantly different from that of the known G3 JEV (SA14-14-1) [12]. The open reading frame (ORF) of G5 JEV(10302 nucleotides (nt)) was 3 nt longer than that of G3 JEV(10299 nt), which encoding an additional amino acid (aa) Ser residue in the NS4A gene of G5 JEV. What is more, They share low identity of 90.7% in nt and 79.3% in aa. The envelope protein (E protein) encoded by the E gene plays an important role in eliciting protective neutralizing antibodies and is crucial for the neurovirulence of JEV. Except the similar 8 key aa residues in E protein for virulence of JEV, the aa identity of E protein between G5 and G3 JEV is only 89.6%. These heterologous differences may involve in the induction of genotype-specific neutralizing antibodies and led to poor cross-immunogenicity between heterologous G3 and G5 JEV.
Low protective efficacy of the current JE vaccine against G5 JEV To date, regardless of whether attenuated or inactivated JEV vaccines are used, they are all currently based on JEV strains representing the single genotype 3 (G3) [21]. Previous studies showed these vaccines can protect against G1-G4 JE virus infection [8,18,19]. In this study, PRNT 90 assays were used to detect neutralizing antibodies against JE viruses in specimens collected from JE-vaccinated healthy children. The results showed that after receiving JE vaccines, SCRs of neutralizing antibodies against G3, G1 and G5 JEV in the serum samples were 100%, 96% and 35%, respectively (Fig 3). SCRs of neutralizing antibodies against G3 and G1 JEV in this study were similar to those observed in previous research; G3 was 100% and G1 slightly lower at 97% (45/47) [8,18,19]. However, in this study, neutralizing antibodies against G5 JEV were detected in only 35% of the healthy children after they received the JE vaccination and 100% SCRs against G5 were detected only when the titer of neutralizing antibodies against G3 JEV was as high as 1: 320 (Table 2). This study is the first to report the low SCR of G5 JEV after administration of the current JE vaccine. Although a larger number of post-vaccination serum samples needs to be evaluated, the results of this study indicate that the protective efficacy of the current JE vaccine is very limited against G5 JE virus infection, regardless of the SCRs or protective neutralizing antibody titers.

Patients with natural infectious JE have a risk of re-infection with G5 virus
In this study, JEV IgM-positive serum samples (n = 45) collected from clinically diagnosed JE patients were used to evaluate them for the presence of G5 JEV-specific neutralizing antibodies.
The positive rates of neutralizing antibodies against G3 and G1 JEV in the serum were 100% (45/45) and 89% (39/45), respectively. However, the positive rate of neutralizing antibodies against G5 JEV was only 64% (29/45). Further analysis showed that the positive rates of neutralizing antibodies against G5 JEV in JE patients 1-15 years of age were 41% (9/22) and 87% (20/23) in JE patients over 15 years of age, indicating that JE patients of different ages showed different levels of G5 JEV protective antibodies. This possibly reflects the greater number of exposures to JEV in the older age-group. Additionally, in this study, the serum sample of a 9-year-old child collected on the 26th day after onset of JE (S2 Table, the 1st case) and serum samples of two children collected on the 11th day after onset of JE (S2 Table, the 5th and 6th cases), showed detectable seroconversion antibodies against G3 JEV (PRNT 90 1:10). However, antibodies against G5 were all negative (PRNT 90 <1:10). These results indicate that within 11-26 days after being infected with JEV, patients failed to produce protective antibodies against G5 JEV. Therefore, pediatric patients with natural JEV infection still had significant potential risk of G5 virus re-infection. Based on our results, whether patients infected with JEV can produce protective antibodies against G5 JEV in the recovery period was inconclusive. The results did, however, imply that JE pediatric patients may not produce protective antibodies against G5 JEV after infection with non-G5 strains of JEV.

Significance to public health
JE is a vaccine-preventable disease and widespread and long-standing vaccination programmes have significantly reduced JE cases in traditional endemic areas such as Japan and Korea. However, since the discovery of the G5 virus in South Korea in 2009 [11], JE cases in Korea have been reported yearly (26 in 2010, 3 in 2011, 20 in 2012, 14 in 2013 and 26 in 2014) [22]. Moreover, six G5 JEVs have been recently detected in local Korean orientalis and Culex pipiens mosquitoes, suggesting that, in addition to the dominant Culex tritaeniorhynchus, other species of mosquito can spread G5 JEV in Korea [23]. The Chinese government included the JE vaccine in their EPI management programme in 2008, allowing school-age children to receive the JE vaccination for free. However, thousands of JE cases still occurred yearly after 2008 [24,25]. Therefore, research on whether JE patients in Korea or China were related to the emerging G5 JEV infection needs to be carried out. To evaluate in detail the protective efficacy of the current JE vaccine against G5 JEV infection, increasing the surveillance of G5 JEV in vectors in endemic areas, detecting levels of G5 JEV protective antibodies across different age groups in JE-endemic areas, and finding clinical JE cases due to G5 JEV infection should be implemented.
Supporting Information S1