Development Of A Novel DNA Vaccine Against Japanese Encephalitis Virus

Japanese encephalitis (JE) is a serious mosquito-borne disease in far eastern and southeastern Asia, with an estimated 45,000 cases and 10,000 deaths annually. The etiologic agent, Japanese encephalitis virus (JEV), belongs to the Flaviviridae family. In humans, JEV infection can cause severe central nervous system disease, and vaccination remains the only promising way of controlling the present JEV outbreaks in endemic areas. Several JEV vaccines have been used in Asian countries with measurable success. One is the formalin-inactivated JEV vaccine purified from infected mouse brain (BIKEN or JE-VAX). It is the only WHO-recommended vaccine and is currently used worldwide. Another inactivated vaccine and a live-attenuated SA 14-14-2 vaccine, both of which are prepared from infected primary hamster kidney cells, are mainly used in China due to the regulatory issues surrounding international standards. Moreover, the recently developed Vero cell-derived inactivated vaccine containing the purified, inactivated JEV strain SA 14-14-2 has been approved (IXIARO). All vaccines have effectively decreased the morbidity of JE irrespective of the drawbacks. The disadvantages of inactivated vaccine include the high cost of preparation, the lack of long-term immunity, and the risk of allergic reactions. Meanwhile, the risk of virulence reversion for the attenuated vaccine is also considered, despite its high efficacy when used in China. Moreover, there have recently been recurring outbreaks of JEV in India. Therefore, it is imperative that a safer, more effective and less costly vaccine be developed for protection against JEV infection. Recently, there has been a significant improvement aimed at employing plasmid DNA-based vaccination, in order to overcome the shortcomings of the traditional JEV vaccines.The flavivirus genome is a single-stranded, positive-sense RNA of approximately 11 kb. The genomic RNA that contains a single open reading frame (ORF) encodes three structural proteins: the capsid protein C, the precursor of membrane protein M (prM), and the envelope protein E, followed by seven non-structural proteins (NS1 to NS5). Among these viral proteins, the glycosylated E protein plays a dominant role in generating the protective neutralizing antibody (NAb). PrM protein is cleaved to M protein during virion maturation. However, in certain instances, the prM cleavage may not be complete, thus allowing the prM protein to be an additional target on virions for NAb, and the correct fold of the E protein requires co-synthesis with the prM protein. The NS1 protein can evoke a protection response through an antibody-dependent complement-mediated pathway. Several investigations have reported the inoculation of plasmids containing a JEV prM-E/ E or NS1 gene to elicit specific protective immune responses in mice. Moreover, protection of mice, from a lethal virus challenge, has also been shown through the passive transfer of monoclonal antibodies against the prM, E and NS1 proteins. Taken together, for the further enhancement of the immune response against JEV, a plasmid expressing these three protective proteins (prM-E-NS1) seems likely to be a better candidate, compared with previous studies that focus mainly on DNA vaccines that express the JEV prM-E protein.Because the immune responses that are induced by DNA vaccines are still suboptimal against pathogens, several methods were established in order to improve the efficiency of DNA vaccines. The use of cytokines has been actively investigated as an adjuvant to modulate immune responses in DNA immunization. Additionally, the granulocyte-macrophage colony-stimulating factor (GM-CSF) has been evaluated extensively with numerous viral, bacterial, parasitic and tumorous vaccines, based on its ability to recruit antigen-presenting cells (APCs) to the site of antigen synthesis, as well as its ability to stimulate the differentiation and maturation of dendritic cells (DCs). Meanwhile, a vector pCAGGSP7 that has a strong eukaryotic promoter derived from the chickenβ-actin was chosen to enhance the expression of JEV proteins.With the strategies described above, a bicistronic plasmid that co-expresses the JEV prM-E-NS1 protein and GM-CSF was constructed as a candidate vaccine. Plasmids expressing prM-E-NS1, or GM-CSF were also constructed. Thus, the purpose of the present study was to determine if co-expression of the JEV antigen and GM-CSF on the same plasmid could enhance the generation of anti-virus immunity after DNA vaccination, when compared to co-administration of separate plasmids that express the viral antigen or GM-CSF. Interestingly, immunization with the bicistronic plasmid coexpressing viral prM-E-NS1 and GM-CSF resulted in the highest IgG response and sufficient protection against virus challenged in BALB/c mice; in contrast, co-injection of the GM-CSF plasmid dramatically suppressed the antibody response and resulted in decreased protective immunity against virus challenge. Our results suggest that a specific immunization response against JEV prM-E-NS1 was optimally induced by bicistronic plasmid DNA coexpressing the viral proteins and GM-CSF.Results and Conclusions1. Plasmid constructionThe internal ribosome entry site (IRES) sequence from the Encephalomyocarditis virus (EMCV) has been used to co-express heterologous gene. Briefly, genomic RNA of JEV was extracted from the JEV Bejing-1 strain-infected C6/36 cells, while GM-CSF RNA was isolated from the spleen of BALB/c mice using the Trizol reagent. JEV prM-E-NS1 and GM-CSF fragments that were obtained by RT-PCR were inserted into the multiple cloning sites A and B of the pIRES vector, respectively. Then, the total prM-E-NS1-IRES-GM-CSF fragment was digested with the Xho I and Not I restriction enzymes, and subcloned into the efficient plasmid, pCAGGSP7. The plasmid co-expressing JEV prM-E-NS1 and GM-CSF was named pCAG-JEGM. Meanwhile, plasmid pCAG-JE (that expresses JEV prM-E-NS1 alone) and pCAG-GM (that expresses GM-CSF alone) was constructed using the same pCAGGSP7 plasmid backbone. DNA sequencing was used to verify the plasmids, and expression of these plasmids was further confirmed by indirect immunofluorescence in mammalian cells. For immunization, the plasmids were extracted and purified, with an endotoxin-free plasmid extraction kit (Omega), from transformed Escherichia coli JM09. Then, the purified plasmids were dissolved in sterile saline and adjusted to 1.0 mg/ml prior to use.2. Designment of mice experimentsFor DNA immunization, six-week-old female BALB/c mice were immunized intramuscularly with 100μg of plasmid DNA, three times, at three-week intervals. The mice were inoculated with the plasmid pCAG-JEGM, or pCAG-JE, or a mixture of pCAG-JE and pCAG-GM, or a mixture of pCAG-JE and pCAGGSP7. Mice immunized with pCAGGSP7 served as negative controls. Three weeks after the final immunization, some mice were euthanized for evaluation of the pre-challenge serum neutralizing antibody (NAb titers). For the protection test, three weeks after the last immunization, the mice were challenged intraperitoneally with a lethal dose (50 LD50) of JEV Beijing-1 strain, followed by a sham intracerebral injection. These mice were observed for mortality for 21 days, and surviving mice were bled for evaluation of the post-challenge serum NAb titers.3. Determination of antibody titers after immunizationGroups of six-week-old BALB/c mice were immunized with the plasmids, three times at intervals of three weeks (0, 3 and 6), and the dynamic anti-JEV response was analyzed at various time points by ELISA. Sera samples were obtained before prime or booster immunization. As expected, the values of optical density (OD) in the control group (pCAGGSP7) always maintained a low level during the observation period, while the group inoculated with pCAG-JEGM, or pCAG-JE, or a mixture with pCAG-JE and pCAGGSP7, showed a significant antibody response three weeks after the first immunization. Moreover, levels of these antibodies were further augmented after the double booster immunization. Surprisingly, mice co-injected with pCAG-JE and pCAG-GM produced relatively lower levels of anti-JEV IgG compared with other vaccinated groups (P <0.00001), and the antibody levels were not apparently enhanced, even with the double booster immunization.Serum samples from before and after the challenge were also analyzed for end-point titers by anti-JEV IgG ELISA. At pre-challenge, co-expression of GM-CSF enhanced the IgG level, and the titer in mice immunized with pCAG-JEGM (1:2900) was 2.2-fold higher than that obtained from the pCAG-JE immunized mice, or was 1.8-fold higher than that obtained from the mice immunized with a mixture of pCAG-JE and pCAGGSP7. Mice co-immunized with plasmids pCAG-JE and pCAGGSP7 (inoculated with 50μg of antigen plasmid) showed similar antibody titers as that of mice immunized with plasmid pCAG-JE alone (100μg). However, the mice co-immunized with pCAG-JE and pCAG-GM showed a very low antibody titer (1:130). Not surprisingly, the post-challenge end-point anti-JEV titers were improved in all of the groups. Among them, group of pCAG-JEGM showed the highest anti-JEV level, with end-point titers about 1:25,600, whereas the group that received pCAG-JE and pCAG-GM had the lowest titer, only up to 1:6,400.Meanwhile, the amount of JEV-specific IgG1 and IgG2a in the sera was also determined by ELISA. The IgG1 isotype is generated by the Th2 immune response, while the IgG2a isotype is generated by Th1 immune response. It has been reported that intramuscular injection of plasmid DNA produced mostly IgG2a antibody. However, in this study, intramuscular immunization induced similar levels of anti-JEV IgG1 and IgG2a response in the pre-challenge sera. In the post-challenge sera, the level of IgG1 subtype was apparently augmented, showing an increased IgG1/IgG2a ratio in the anti-JEV antibody response. A bias toward a Th2 phenotype response was induced following the challenge of virus.4. Measurement of NAb titers by plaque reduction neutralization test (PRNT)Sera samples that were obtained from mice three weeks after the final immunization were assayed for JEV NAb titers. Following the last immunization, mice were challenged with 50 LD50 JEV; sera were also collected from surviving mice at the end of the three-week observation period. The result indicated that the pre-challenge NAb titers were very low, and ranged from 1:10 to 1:20 in the different groups, even though the DNA vaccination induced high anti-JEV titers in the ELISA and high protection rates in the challenge experiment. NAb titers of mice immunized with the empty vector did not exceed 1:10. Unexpectedly, co-expression of GM-CSF with prM-E-NS1 had little effect on generation of the NAb response. On the contrary, the NAb titers in the post-challenge sera were improved significantly; the highest post-challenge NAb titers were seen in sera from the pCAG-JEGM immunized mice, which was as high as 1:390, while NAb titers with 1:260 and 1:290 were seen in group pCAG-JE and the group co-immunized with pCAG-JE and pCAGGSP7, respectively. Additionally, the NAb titers in the sera from mice co-immunized with pCAG-JE and pCAG-GM reached to 1:140, which was the lowest value among these groups.5. Antibody-dependent complement mediated cytotoxicity in infected cellsIn the present study, mice were vaccinated with plasmids expressing the JEV prM-E-NS1 protein. Although specific NAb against E-protein played an important role in protection, it is of interest to know whether the NS1-specific antibodies can also block JEV replication through a complement-dependent manner. Therefore, sera from immunized mice were further analyzed for their ability to lyse JEV-infected L929 cells in the presence of complement, by measuring the release of LDH. Meanwhile, sera from mice immunized with plasmid pCAG-ME (that expresses the JEV prM-E protein, without the NS1 protein) were also analyzed in this experiment. The results showed that sera from mice immunized with pCAG-JEGM, pCAG-JE, or a mixture of pCAG-JE and pCAGGSP7, were able to lyse the JEV-infected cells, and had over 50% specific lysis ability; on the contrary, sera from mice co-immunized with pCAG-JE and pCAG-GM showed a low level of cytolytic ability, which indicated a low level of NS1 antibody and is consistent with the result of a low antibody level in ELISA. The control sera from normal mice or sera from pCAG-ME immunized mice only showed low and non-specific lysis activity.6. Cell-mediated immune response detected by lymphocyte proliferation and CTL assayTo evaluate the antigen-specific immunological memory in lymphocytes, the splenocytes derived from vaccinated mice were stimulated with JEV antigen in vitro, and a cell proliferation assay was performed using CCK-8. The splenocytes from pCAGSSP7-immunized mice were used as a negative control, and no distinct proliferation was observed (Simulation index, SI=1.09±0.14) in this group. As expected, significant proliferation was shown in each of the other four vaccinated groups (SI ranged from 2.5 to 2.8, P<0.001,) when compared with the negative control, demonstrating that the lymphocyte proliferation was in an antigen-specific manner.Moreover, the splenocytes collected from the different immunized groups were also stimulated with JEV in vitro and were examined for CTL activity to lyse the JEV-infected L929 cells. Immunization with pCAG-JE or pCAG-JE+pCAGGSP7 only produced very low JEV-specific CTL activity, even at high E:T ratios (200:1). Co-expession or co-immunization with GM-CSF failed to enhance CTL activity, even though we performed the CTL experiments several times, only inferior amounts of CTL activity could be detected with or without GM-CSF.7. Mouse challenge experimentsThe plasmids were administered three times to BALB/c mice by an intramuscular route, and then the mice were challenged at three weeks after the final immunization with 50 LD50 of JEV Beijing-1 strain. pCAGGSP7-immunized mice served as the negative controls, and none of the mice survived the JEV challenge in this group. In contrast, mice that received pCAG-JEGM, pCAG-JE and a mixture of pCAG-JE and pCAGGSP7 were significantly protected from the virus challenge, with a 100% survival rate (seven of seven, P < 0.0004); on the other hand, mice that received a mixture of pCAG-JE and pCAG-GM were not fully protected, with 71% survival rate (five of seven, P < 0.002), probably due to the low antibody level detected by ELISA.8. Adoptive transfer experimentsAdoptive transfer experiments were preformed with the sera that were collected from mice immunized with plasmid pCAG-JEGM or pCAG-JE before the challenge, and the corresponding sera from mice that survived the virus challenge. Interestingly, a 100% survival rate was seen in mice that received the post-challenge sera, whereas there were no mice that survived when inoculated with the pre-challenge sera of pCAG-JEGM. Only one mouse survived in the group inoculated with the pre-challenge sera of pCAG-JE. The results of adoptive transfer experiments strongly suggested that the JEV-specific antibody, especially the post-challenge sera, which have high IgG antibody titers and NAb titers, could mediate viral clearance and protect mice from the lethal viral challenge.In conclusion, our DNA vaccine expressing JEV prM-E-NS1 protein could generate high protection efficiency in mice challenge model, and the anamnestic antibody response or the antibody-mediated mechanism plays dominant role in the infection course. Coexpression with GM-CSF induces better immune response than co-inoculation with plasmid of GM-CSF. The possible advantage of using GM-CSF as an adjuvant for vaccines is still a matter of debate and the mechanism of how the GM-CSF play its adjuvant role in vivo need to be further investigated. Thus, GM-CSF is still very useful as an adjuvant with some caution.