A molecularly cloned, live-attenuated Japanese encephalitis vaccine SA14-14-2 virus: a conserved single amino acid in the ij hairpin of the viral E glycoprotein determines neurovirulence in mice.

PLoS Pathog. 2014 Jul 31;10(7):e1004290.

Yun SI1, Song BH1, Kim JK1, Yun GN2, Lee EY3, Li L4, Kuhn RJ4, Rossmann MG4, Morrey JD1, Lee YM1.

1 Department of Animal, Dairy, and Veterinary Sciences; Utah Science Technology and Research, College of Agriculture and Applied Sciences, Utah State University, Logan, Utah, United States of America, 2 Department of Microbiology, College of Medicine, Chungbuk National University, Cheongju, Republic of Korea, 3 Department of Anatomy, College of Medicine, Chungbuk National University, Cheongju, Republic of Korea, 4 Department of Biological Sciences, Purdue University, West Lafayette, Indiana, United States of America.



Japanese encephalitis virus (JEV), a mosquito-borne flavivirus that causes fatal neurological disease in humans, is one of the most important emerging pathogens of public health significance. JEV represents the JE serogroup, which also includes West Nile, Murray Valley encephalitis, and St. Louis encephalitis viruses. Within this serogroup, JEV is a vaccine-preventable pathogen, but the molecular basis of its neurovirulence remains unknown. Here, we constructed an infectious cDNA of the most widely used live-attenuated JE vaccine, SA14-14-2, and rescued from the cDNA a molecularly cloned virus, SA14-14-2MCV, which displayed in vitro growth properties and in vivo attenuation phenotypes identical to those of its parent, SA14-14-2. To elucidate the molecular mechanism of neurovirulence, we selected three independent, highly neurovirulent variants (LD50, <1.5 PFU) from SA14-14-2MCV (LD50, >1.5´105 PFU) by serial intracerebral passage in mice. Complete genome sequence comparison revealed a total of eight point mutations, with a common single G1708→A substitution replacing a Gly with Glu at position 244 of the viral E glycoprotein. Using our infectious SA14-14-2 cDNA technology, we showed that this single Gly-to-Glu change at E-244 is sufficient to confer lethal neurovirulence in mice, including rapid development of viral spread and tissue inflammation in the central nervous system. Comprehensive site-directed mutagenesis of E-244, coupled with homology-based structure modeling, demonstrated a novel essential regulatory role in JEV neurovirulence for E-244, within the ij hairpin of the E dimerization domain. In both mouse and human neuronal cells, we further showed that the E-244 mutation altered JEV infectivity in vitro, in direct correlation with the level of neurovirulence in vivo, but had no significant impact on viral RNA replication. Our results provide a crucial step toward developing novel therapeutic and preventive strategies against JEV and possibly other encephalitic flaviviruses.



The long-term goal of our research is to understand how JEV causes fatal neurological disease, a multifactorial and multidimensional process determined by an array of viral virulence factors that interact with their host targets at specific sites and times during infection in a susceptible host. Molecular genetic studies on the viral virulence factors to date have been limited because of the severe host cell toxicity and genetic instability of a complete cDNA copy of the viral genomic RNA during propagation in E. coli, which makes it extremely difficult to clone and maintain a functional cDNA for genetic manipulation of the viral genomic RNA (1-3).

In our previous study, we reported the successful construction of a genetically stable full-length infectious cDNA clone for a highly virulent JEV strain CNU/LP2 (4), by utilizing bacterial artificial chromosome (BAC) as a cloning vector (5). In our recently published work featured in this website, we used the same cloning strategy to generate a fully functional cDNA clone for SA14-14-2 (6, 7), a live JE vaccine virus that is highly attenuated in both neurovirulence and neuroinvasiveness (8). Using our newly developed infectious SA14-14-2 cDNA BAC, combined with a mouse model for JEV infection, we have performed a series of comprehensive and systematic experiments to investigate the genetic basis for the attenuation of SA14-14-2. The findings of our study are five-fold:

(i) We cloned a full-length SA14-14-2 cDNA into the BAC plasmid pBeloBAC11, which was shown to be able to accommodate a full-length JEV cDNA. More importantly, we synthesized infectious RNA transcripts in vitro from the full-length SA14-14-2 cDNA that were capable of producing a high titer of the molecularly cloned SA14-14-2 virus when transfected into JEV-permissive cells, such as BHK-21 (Figure 1). The infectious SA14-14-2 cDNA clone thus provides unique opportunities to study the viral genetic factors responsible for the attenuation of the live JE vaccine virus.


Figure 1 (Yun et al, 2015)

Figure 1. A schematic diagram showing the reverse genetics system based on a full-length infectious BAC of JEV SA14-14-2. The diagram has been modified from the original review article (6).

(ii) We generated three highly neurovirulent variants derived from the molecularly cloned SA14-14-2 virus after four serial passages in the brains of mice. We sequenced the complete genome of these three SA14-14-2 variants and compared them to that of the parental SA14-14-2 virus. This genome sequence analysis revealed one missense (G1708→A) and seven silent point mutations. Of these, the G1708→A substitution, causing a Gly→Glu change at position 244 of the viral E glycoprotein, was the only one found in all three variants. In mice, all three variants still lacked a detectable level of neuroinvasiveness but acquired a very high degree of neurovirulence.

(iii) We constructed eight derivatives of SA14-14-2, each containing one of the eight point mutations found in our three variants. After the intracerebral (IC) inoculation of mice, we found that one of the eight mutants with the G1708→A substitution had an IC LD50 of <1.5 PFU, but the remaining seven mutants all had IC LD50s of >1.5´105 PFU, as seen with the parent SA14-14-2. These results clearly showed that the single G1708→A substitution is responsible for the reversion of SA14-14-2 to lethal neurovirulence.

(iv) We determined by homology modeling that E-244 is located in the ij hairpin adjacent to the fusion loop of the viral E dimerization domain (Figure 2). We further demonstrated the biological importance of E-244 in neurovirulence by creating 14 E-244 mutants of three different classes: (a) neurovirulent, with an IC LD50 of £1.5 to 31 PFU, replacing G244 with E, D, T, S, Q, and P; (b) non-neurovirulent, with an IC LD50 of >1.5´104 or 105 PFU, exchanging G244 with R, K, F, W, N, and L; and (c) intermediate, with an IC LD50 of 1.2-5.8´103 PFU, substituting G244 with A and V.


Figure 2 (Yun et al, 2015)

Figure 2. A homology model illustrating an icosahedral asymmetric unit of the three E monomers on the viral membrane. Highlighted in the inset is the critical residue Glu at E-244 in the ij hairpin adjacent to the fusion loop of the viral E dimerization domain (DII). The model has been copied from the original figure (9).


(v) We discovered that E-244 plays a crucial role in establishing the productive infection of SA14-14-2 in neuronal but not non-neuronal cells, by performing multistep growth assays in two neuronal (mouse NSC-34 and human SH-SY5Y) and one non-neuronal (hamster BHK-21) cell line for comparison. We also found that E-244 is capable of altering JEV infectivity in a neuronal cell-specific manner, in agreement with the neurovirulence phenotype observed in mice.

Collectively, we report that E-244 in the ij hairpin of viral E glycoprotein acts as a key viral factor that promotes JEV infection of neurons in mice, likely through its role in one of three major steps involved in viral entry: binding, endocytosis, or membrane fusion (Figure 3). The results of this research open new avenues for the development of novel therapeutic interventions against JEV and other closely related neurotropic flaviviruses.


Figure 3 (Yun et al, 2015)

Figure 3. A simplified overview depicting the replication cycle of JEV. The overview has been modified from the original review article (6).



  1. Ruggli N, Rice CM. 1999. Functional cDNA clones of the Flaviviridae: strategies and applications. Adv Virus Res 53:183-207.
  2. Sumiyoshi H, Hoke CH, Trent DW. 1992. Infectious Japanese encephalitis virus RNA can be synthesized from in vitro-ligated cDNA templates. J Virol 66:5425-5431.
  3. Mishin VP, Cominelli F, Yamshchikov VF. 2001. A ‘minimal’ approach in design of flavivirus infectious DNA. Virus Res 81:113-123.
  4. Kim JM, Yun SI, Song BH, Hahn YS, Lee CH, Oh HW, Lee YM. 2008. A single N-linked glycosylation site in the Japanese encephalitis virus prM protein is critical for cell type-specific prM protein biogenesis, virus particle release, and pathogenicity in mice. J Virol 82:7846-7862.
  5. Yun SI, Kim SY, Rice CM, Lee YM. 2003. Development and application of a reverse genetics system for Japanese encephalitis virus. J Virol 77:6450-6465.
  6. Yun SI, Lee YM. 2014. Japanese encephalitis: the virus and vaccines. Hum Vaccin Immunother 10:263-279.
  7. Yu Y. 2013. Development of Japanese encephalitis attenuated live vaccine virus SA14-14-2 and its characteristics, p. 181-206. In Tkachev S (ed.), Encephalitis. InTech, Open Access.
  8. Song BH, Yun GN, Kim JK, Yun SI, Lee YM. 2012. Biological and genetic properties of SA14-14-2, a live-attenuated Japanese encephalitis vaccine that is currently available for humans. J Microbiol 50:698-706.
  9. Yun SI, Song BH, Kim JK, Yun GN, Lee EY, Li L, Kuhn RJ, Rossmann MG, Morrey JD, Lee YM. 2014. A molecularly cloned, live-attenuated Japanese encephalitis vaccine SA14-14-2 virus: a conserved single amino acid in the ij hairpin of the viral E glycoprotein determines neurovirulence in mice. PLoS Pathog 10:e1004290.


Multiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier SchönmannMultiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier Schönmann