JGV Papers in Press. Published June 5, 2015 as doi:10.1099/vir.0.000213
Journal of General Virology In vitro Growth, Pathogenicity, and Serological Characteristics of the Japanese Encephalitis Virus Genotype V Muar strain. --Manuscript Draft-Manuscript Number:
VIR-D-15-00245R1
Full Title:
In vitro Growth, Pathogenicity, and Serological Characteristics of the Japanese Encephalitis Virus Genotype V Muar strain.
Short Title:
Characterization of the Japanese Encephalitis Virus Genotype V Muar strain.
Article Type:
Standard
Section/Category:
Animal - Positive-strand RNA Viruses
Corresponding Author:
Shigeru Tajima, Ph.D. National Institute of Infectious Diseases Shinjuku, Tokyo JAPAN
First Author:
Shigeru Tajima, Ph.D.
Order of Authors:
Shigeru Tajima, Ph.D. Kazumi Yagasaki Akira Kotaki Takumi Tomikawa Eri Nakayama Meng Ling Moi Chang-Kweng Lim Masayuki Saijo Ichiro Kurane Tomohiko Takasaki
Abstract:
The characteristics of the genotype V Japanese encephalitis virus (GV JEV) remain poorly understood since only two strains have been isolated, thus far. In this study, we examined the effects of the GV JEV Muar strain on in vitro growth and pathogenicity in mice; we also evaluated the efficacy of inactivated Japanese encephalitis (JE) vaccines against the Muar strain. Although the growth of the Muar strain in mouse neuroblastoma N18 cells was clearly worse than those of the GIII Beijing-1 and GI Mie/41/2002 strains, neuroinvasiveness of the Muar strain was similar to that of the Beijing-1 strain and significantly higher than that of the Mie/41/2002 strain. The results of a plaque reduction neutralization test suggest that the neutralization ability of the JE vaccines against the Muar strain is reduced compared with GI and GIII strains. However, the protection potency of the JE vaccine against the Muar strain was similar to that for the Beijing-1 strain in mice. Our data indicate that GV JEV has unique growth, virulence, and antigenicity features.
Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
Manuscript Including References (Word document) Click here to download Manuscript Including References (Word document): Text(PRNT)150603JGV.docx
1
In vitro Growth, Pathogenicity, and Serological Characteristics of the Japanese
2
Encephalitis Virus Genotype V Muar strain.
3 4
Shigeru Tajima*, Kazumi Yagasaki, Akira Kotaki, Takumi Tomikawa, Eri Nakayama, Meng
5
Ling Moi, Chang-Kweng Lim, Masayuki Saijo, Ichiro Kurane and Tomohiko Takasaki.
6 7
Department of Virology I, National Institute of Infectious Diseases, 1-23-1 Toyama, Shinjuku,
8
Tokyo 162-8640, Japan
9 10
Key words:
11
Japanese encephalitis virus / Genotype / Pathogenicity / Neutralization / Vaccine
12 13
Correspondence: Shigeru Tajima, Ph.D.
14
Department of Virology 1, National Institute of Infectious Diseases
15
1-23-1 Toyama, Shinjuku, Tokyo 162-8640, Japan
16
E-mail: [email protected]
17
Phone/Fax: +81-3-5285-1188
18 19 20
Footnote
21
The GenBank accession numbers for the JEV E gene sequence we determined in this study are as
22
follow: Nakayama strain, AB920348; Beijing-1 strain, AB920347; Kumamoto/65/2005 strain,
23
AB920445; Kumamoto/81/2006 strain, AB920446; Kumamoto/104/2006 strain, AB920447;
24
Chiba/150/2007 strain, AB920448; Chiba/103/2008, AB920449.
25
Summary
26
The characteristics of the genotype V Japanese encephalitis virus (GV JEV) remain poorly
27
understood since only two strains have been isolated, thus far. In this study, we examined the
28
effects of the GV JEV Muar strain on in vitro growth and pathogenicity in mice; we also
29
evaluated the efficacy of inactivated Japanese encephalitis (JE) vaccines against the Muar strain.
30
Although the growth of the Muar strain in mouse neuroblastoma N18 cells was clearly worse
31
than those of the GIII Beijing-1 and GI Mie/41/2002 strains, neuroinvasiveness of the Muar
32
strain was similar to that of the Beijing-1 strain and significantly higher than that of the
33
Mie/41/2002 strain. The results of a plaque reduction neutralization test suggest that the
34
neutralization ability of the JE vaccines against the Muar strain is reduced compared with GI and
35
GIII strains. However, the protection potency of the JE vaccine against the Muar strain was
36
similar to that for the Beijing-1 strain in mice. Our data indicate that GV JEV has unique growth,
37
virulence, and antigenicity features.
38 39
40
Introduction
41
Japanese encephalitis (JE), resulting from an infection of the Japanese encephalitis virus
42
(JEV), is a significant public health problem in many Asian countries. JEV causes serious
43
nervous disorders such as meningitis and encephalitis. There are an estimated 68,000 cases of JE
44
per year, occurring mainly in China, India, and Southeast Asian countries, which result in 15,000
45
fatalities, mostly in children (Campbell et al., 2011; Erlanger et al., 2009; Tsai, 2000). There is
46
no specific treatment available for JE, other than preventive vaccination with JE vaccines. In
47
Japan, the number of reported JE cases has dramatically decreased since the late 1960s, and less
48
than 10 JE cases have occurred in Japan over the past two decades (Arai et al., 2008). This
49
decrease is thought to be due to active vaccination, a decrease in the number of mosquitoes that
50
can transmit JEV, and urbanization. However, a study on the prevalence of JEV that used
51
domestic pigs indicated that, in most regions of Japan, naive pigs are still seroconverted to JEV
52
every year, which suggests that JEV is still a serious threat (Arai et al., 2008).
53
JEV belongs to the genus Flavivirus in the family Flaviviridae and is amplified in a bird
54
and pig-mosquito transmission cycle (Pierson, 2013). The infected mosquitoes, mainly Culex
55
tritaeniorhynchus, transmit JEV to dead-end hosts: either humans or horses. JEV has a single-
56
stranded positive-sense RNA genome, which encodes three structural proteins (C, prM, and E)
57
and seven non-structural proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5). JEV is
58
classified into five genotypes (GI - GV) based on its genomic sequence (Solomon et al., 2003),
59
of which the GIII strain was the most widely distributed and most frequently isolated in JE
60
endemic areas until the 1990s. Consequently, all licensed inactivated and live-attenuated JE
61
vaccines are derived from GIII JEV. However, the major genotype of JEV isolated in Japan
62
changed from GIII to GI strain in the early 1990s (Ma et al., 2003; Yoshida et al., 2005). In
63
recent years, a similar genotype shift has been observed in South Korea, northern Vietnam,
64
China, Taiwan, and Thailand (Chen et al., 2011; Nam et al., 1996; Nga et al., 2004; Nitatpattana
65
et al., 2008; Pan et al., 2011; Wang et al., 2007; Yang et al., 2004; Yun et al., 2010; Zhang et al.,
66
2011; Zhang et al., 2009). GI JEV has also been found in India since 2009 (Fulmali et al., 2011).
67
Thus, the dominant genotype in JEV endemic areas is now GI, although the reasons for the broad
68
shift from GIII to GI remain unclear. The Indonesian/Malaysian region is thought to be the origin
69
of all genotypes of JEV (Solomon et al., 2003), and in several Southeast Asian countries, GI was
70
isolated as a minor JEV strain before the 1990s. Comprehensive phylogenetic analysis of the GI
71
JEV strains demonstrates that the old and the new GI strains form distinct lineages (Gao et al.,
72
2013; Schuh et al., 2014), suggesting that genetic differences in the new GI genome might be
73
responsible for the spread of GI viruses. Recent findings indicate that GI JEV exhibits a higher
74
replicative ability in mosquitoes than does GIII JEV (Schuh et al., 2014).
75
A GV strain of JEV was first isolated from an encephalitis patient in Malaysia in 1952
76
(also referred to as the Muar strain), and it remained the only known instance of this genotype for
77
over 50 years (Hasegawa et al., 1994; Solomon et al., 2003). Recently, the GV JEV genome has
78
been identified in Culex mosquitoes in China in 2009 and in South Korea in 2010 and 2012 (Kim
79
et al., 2015; Li et al., 2011; Takhampunya et al., 2011). A group in China succeeded in isolating
80
a new GV strain (referred to as the XZ0934 strain) from C. tritaeniorhynchus (Li et al., 2011). In
81
South Korea, genomes of GV JEV were detected in Culex bitaeniorhynchus, Culex orientalis,
82
and Culex pipiens, but none were identified in the major JEV vector C. tritaeniorhynchus (Kim
83
et al., 2015; Takhampunya et al., 2011). These findings raise the possibility that the other species
84
of Culex mosquitoes may be involved in the natural transmission cycle of the GV strain of JEV.
85
Thus, GV JEV may also be emerging in other endemic areas, and more attention needs to be
86
focused on investigating the dynamics of circulating JEV strains in these areas.
87
Since only two GV strains, Muar and XZ0934, have been isolated thus far and biological
88
studies on the features of GV JEV have been very limited, the growth and pathogenic properties
89
of GV remain poorly understood. The spread of JEV strains, other than GIII, into endemic areas
90
makes it imperative that systematic evaluation is conducted on the protective efficacy of current
91
GIII-derived JE vaccines against other JEV genotypes. Previous reports call attention to the
92
reduced neutralizing potency of inactivated and live-attenuated GIII JE vaccines against
93
circulating GI JEV (Fan et al., 2013; Fan et al., 2012). On the other hand, other researchers have
94
shown that inactivated and live-attenuated JE vaccines are effective against both GIII and GI
95
JEV strains that are prevalent in China (Liu et al., 2011), and inactivated vaccines used for
96
European travelers induce protective levels of neutralizing antibodies against GI - GIV JEV
97
(Erra et al., 2013a; b). Thus, the efficacy of the current JE vaccines against heterologous
98
genotypes remains unresolved. Furthermore, these studies did not assess the Beijing-1 strain-
99
derived vaccine that is currently available in Japan, the Nakayama strain-derived vaccine, or a
100
GV JEV strain as a target virus.
101
To uncover the characteristics of GV JEV, we investigated in vitro growth and
102
pathogenicity in mice for the GV JEV Muar strain. We also evaluated the protective efficacy of
103
the inactivated JE vaccines against the Muar strain as well as several GI and GIII JEV strains.
104 105 106 107 108
109
Results
110
In vitro growth properties of the Muar strain
111
To characterize the growth of the GV Muar strain in vitro, we first compared the plaques
112
induced by the Muar strain to those of GI Mie/41/2002 strain and GIII Beijing-1 strains in Vero
113
cells (Fig. 1A). The plaque size induced by Muar (average diameter and SEM: 0.98±0.03 mm)
114
was less than that of Mie/41/2002 (1.89±0.05 mm; P < 0.001) but was similar to that of Beijing-
115
1 (1.02±0.03 mm; P = 0.37). Similar results were obtained using a replication kinetics assay for
116
these viruses in Vero cells and BHK-21 cells (Fig. 1B and C). These results suggest that the
117
growth of Muar is similar to that of Beijing-1 in interferon response-deficient cells. Our previous
118
report showed that, in contrast to the results in Vero cells, the steady state number of infectious
119
particles for Beijing-1 was apparently higher than that of Mie/41/2002 in mouse neuroblastoma-
120
derived N18 cells (Tajima et al., 2010). We next examined the growth properties of Muar in N18
121
cells (Fig. 1D). However, the growth rate of Muar in N18 cells was slower than the other strains,
122
and the infectious particles of Muar at steady state were 100-fold and 10-fold less than those of
123
Beijing-1 and Mie/41/2002, respectively. These data indicate that Muar might exhibit a reduced
124
amplification in N18 cells, and the characteristics of the Muar strain are clearly different from
125
those of the Beijing-1 and Mie/41/2002 strains in vitro. Using real-time RT-PCR, we also
126
confirmed in all three cultured cells that the quantity of viral genomes in the supernatant and in
127
the cells infected the viruses corresponded to those of the virus titers (data not shown), indicating
128
that the reduced growth of Muar in cultured cells does not reflect defective particle assembly and
129
secretion.
130 131
Virulence of the Muar strain in mice
132
Next, we examined the pathogenicity of the Muar strain in mice. Mice were infected
133
intraperitoneally (i.p.) with Muar, Beijing-1, or Mie/41/2002 and observed for 3 weeks (Table 1).
134
Our previous report using the Beijing-1 and Mie/41/2002 strains demonstrated that the growth
135
characteristics of some JEV strains in N18 cells were positively correlated with
136
neuroinvasiveness in mice (Tajima et al., 2010); therefore, we speculated that the Muar strain
137
might exhibit a lower virulence in mice compared with the Beijing-1 and Mie/41/2002 strains. In
138
the 103, 104, and 105 plaque forming units (p.f.u.)-infected groups, more than 90% of the mice
139
infected with Muar or Beijing-1 died; however, less than 30% of the mice infected with
140
Mie/41/2002 died. The survival curves for the Muar-infected groups were similar to those of the
141
Beijing-1-infceted groups. These results indicate that, in contrast to the growth kinetics in N18
142
cells, neuroinvasiveness of Muar is equivalent to that of Beijing-1 and significantly higher than
143
that of Mie/41/2002 in mice.
144 145
The neutralizing ability of inactivated JE vaccines against Muar and recombinant
146
intertypic JEVs.
147
We next evaluated the ability of inactivated Nakayama and Beijing-1 vaccines, which
148
were produced in Japan, to induce neutralizing antibodies against Muar. We also used three GIII
149
and 11 GI viruses with differing amino acid sequences in the E protein as target viruses
150
(Supplementary table 1). The E protein is the major structural protein that constitutes the surface
151
structure of the virus particles, and a majority of neutralizing antibodies bind to the E protein and
152
can inhibit viral infection (Wu et al., 2003). Serum samples collected from mice at 14 days after
153
initial immunization with 2-fold-diluted and 8-fold-diluted JE vaccines were used for the plaque
154
reduction neutralization test (PRNT), and the value of 50% reduction relative to the non-serum
155
control (PRNT50) for each virus was determined (Table 2). In the sera of mice that were
156
immunized with 2-fold dilutions, the PRNT titer against GI and GIII viruses was between 1:80
157
and 1:320. In those with 8-fold dilutions, it was between 1:40 and 1:160. We statistically
158
compared the PRNT titers of sera against GI viruses to those against GIII viruses using a
159
parametric Welch’s t-test and a non-parametric Mann-Whitney U-test. No significant differences
160
(P > 0.05) were observed between the GI group and GIII group in all cases, indicating that the
161
neutralizing efficacy of inactivated Nakayama and Beijing-1 vaccines against the GI strains is
162
similar to that for the GIII strains. In contrast to the GI and GIII strains, the PRNT titers in the
163
sera of mice immunized against GV Muar were 4-fold less than those against the GI and GIII
164
strains. These results suggest that the neutralizing capacity of the JE vaccines against GV JEV
165
was considerably lower than that against GI and GIII JEV. We also examined the neutralizing
166
potency (PRNT) of sera obtained from mice at 2 and 4 months after the initial immunization for
167
the 2-fold-diluted and 8-fold-diluted Nakayama and Beijing-1 vaccines (Table 3). Two vaccine
168
strains, two GI strains, three GIII strains, and one GV strain were used as target viruses. The
169
PRNT titers for GV Muar were less than those for the GI and GIII JEV, indicating that the
170
neutralization capacity of the inactivated JE vaccines is relatively weak against the GV Muar
171
strain.
172
A comparison of the E protein amino acid sequences amount the viral genotypes showed a
173
higher homology between GI and GIII than that between GI and GV or that between GIII and
174
GV (data not shown). There were 40 residues (8.0%) that differed between Beijing-1 and Muar,
175
and 44 residues (8.8%) were different between Nakayama and Muar. In contrast, less than 12
176
residues (2.4%) were different between the vaccine strains and the GI strains, and less than 8
177
residues (1.6%) were different between the vaccine strains and the GIII strains (Supplementary
178
table 2 and 3). Other GV strains, XZ0934 and 10-1827, also showed similar homology to the
179
vaccine strains, and only two residues (0.4%) were different between the XZ0934 and 10-1827
180
strains. These findings raise the possibility that the neutralization potency of the inactivated
181
vaccines against the other GV strains is also relatively weak. To examine the efficacy of the
182
vaccines against GV strains other than Muar, we generated recombinant intertypic JEVs rJEV-
183
EMuar-M41 and rJEV-EXZ0934-M41, which contain the majority of the E region of Muar and
184
XZ0934, respectively, in the backbone of the Mie/41/2002 genome. We also generated rJEV-
185
ENakayama-M41 and rJEV-EBeijing-1-M41 as control viruses for use in the neutralization assay
186
(Tajima et al., 2010). The PRNT titers in the sera from mice immunized with the vaccines
187
against rJEV-EMuar-M41 and rJEV-EXZ0934-M41 were clearly less than those against the parent
188
virus Mie/41/2002, rJEV-ENakayama-M41, and rJEV-EBeijing-1-M41 (Table 4). These results
189
indicate that the neutralization ability of the vaccines may be weak against both Muar and other
190
GV JEV strains.
191 192
Protection efficacy of the JE vaccines against Muar infection in mice
193
To examine whether immunization of mice with the inactivated JE vaccines inhibits the
194
onset of Muar-induced disease, mice that were vaccinated with the inactivated Beijing-1 vaccine
195
were inoculated i.p. with 104 PFU of Muar or Beijing-1 strains (Table 5). Mice vaccinated with
196
2-fold diluted vaccine had a 100% survival rate after inoculation with Beijing-1 or Muar.
197
Fourteen of 16 mice (87.5%) and 12 of 16 mice (75%) that were vaccinated with 32-fold diluted
198
vaccine survived after infection with Beijing-1 or Muar, respectively. There was no significant
199
difference in the survival rates between Beijing-1-infected and Muar-infected groups. These
200
results suggest that the protection efficacy of the Beijing-1 vaccine against infection with Muar is
201
similar to that against infection with Beijing-1 in vivo.
202 203
The neutralizing ability of the sera from Muar-infected mice against various JEVs
204
Finally, we assessed the ability of GV Muar-infected mouse sera to block the infection of
205
GI and GIII JEVs in vitro using the neutralization assay (Table 6). The PRNT titer against Muar
206
was 2- or 4-fold higher than those against two GI, three GIII, and Beijing-1 strains, though the
207
titer against Nakayama was equivalent to that against Muar. Thus, we confirmed that sera from
208
Muar-infected mice were able to neutralize the infection of GI and GIII strains; however, the
209
neutralization potency of these sera against GI and GIII viruses was weaker than that against
210
Muar.
211 212
213
Discussion
214
To date, only two GV JEV strains have been isolated and amplified in vitro. Therefore, the
215
characteristics of GV JEV remains poorly understood. In this study, we examined the in vitro
216
growth properties and in vivo pathogenicity of the GV strain Muar. Furthermore, we investigated
217
the potential of inactivated JE vaccines that are currently available in Japan to induce a
218
protective immune response against GV JEV using mouse models. Our results indicate that GV
219
JEV has different growth, pathogenic, and antigenic characteristics compared with GI and GIII
220
JEV.
221
Our previous report suggested that there is a correlation between the growth characteristics
222
of JEV in mouse neuroblastoma N18 cells and virulence in mice (Tajima et al., 2010). The
223
growth rate of Muar was less than that of the highly virulent GIII Beijing-1 strain and the
224
moderately virulent GI Mie/41/2002 strain in N18 cells; however, the neuroinvasiveness of Muar
225
was similar to that of Beijing-1. These results suggest that the growth properties of JEV in N18
226
cells do not necessarily reflect pathogenicity in mice. There is little information regarding the
227
regions of the Muar genome that are associated with virulence in mice. We previously
228
demonstrated that the E protein contributes to the difference in virulence between Mie/41/2002
229
and Beijing-1 (Tajima et al., 2010). Other reports have also highlighted the importance of the E
230
protein as a determinant of pathogenicity (Monath et al., 2002; Sumiyoshi et al., 1995; Zhao et
231
al., 2005). On the other hand, our group determined that an amino acid substitution in the JEV
232
NS4A protein alters virulence in mice (Yamaguchi et al., 2011). To clarify the regions
233
responsible for increased pathogenicity in the Muar genome, various GI-GV intertypic and point
234
mutant viruses may be required. Recently, Ishikawa et al. identified the infectious molecular
235
clone and reporter replicon of the Muar strain, which is also beneficial for generally
236
understanding the characteristics of GV JEV (Ishikawa et al., 2015). In addition, de Wispelaere
237
et al. constructed a molecular clone of the GV XZ0934 JEV strain and demonstrated that this
238
recombinant GV virus exhibited a greater pathogenicity in BALB/c mice than that of GIII JEV
239
(de Wispelaere et al., 2015).
240
The neutralization tests using the vaccinated mouse sera indicated that both the Nakayama
241
and Beijing-1 vaccines induce sufficient neutralizing antibodies against GI JEV. However, these
242
vaccines induce fewer neutralizing antibodies against GV Muar or the recombinant intertypic
243
JEVs, which contain the E protein from the GV Muar or XZ0934 strains. Furthermore, the sera
244
from Muar-infected mice had a lower neutralization titer against GI and GIII viruses compared
245
with the Muar strain. These data indicate that GV JEV may be distinct from GI and GIII JEV in
246
antigenicity. The E protein sequences between GI and GIII JEV differ by approximately 3%
247
(Solomon et al., 2003). However, this number increases to 8.8 - 9.2% between GV strains and
248
the Nakayama strain, and 8.2 - 8.6% between GV strains and the Beijing-1 strain
249
(Supplementary table 2 and 3). The decrease in homology between the vaccine strains and the
250
GV strains might be responsible for the observed weak efficacy of the vaccines against the Muar
251
strain and recombinant JEVs that contain GV E proteins. The amino acid differences between the
252
vaccine strains and the GI strains did not exceed 2.4%, indicating that the difference may not
253
influence the efficacy of the vaccines. Previous studies have identified individual residues within
254
the E protein that are related to the recognition of JEV by neutralizing antibodies (Crill & Chang,
255
2004; Goncalvez et al., 2008; Kobayashi et al., 1985; Morita et al., 2001; Vogt et al., 2009; Wu
256
et al., 2003). Although the sequence variation between the vaccine strains and GV strains do not
257
correspond to epitope residues, only a limited number of residues could be responsible for the
258
decreased neutralization ability of the vaccines against GV strains. All inactivated and live-
259
attenuated JE vaccines that are currently available are derived from GIII strains. Our results
260
indicate that live-attenuated GIII-derived vaccines may have a reduced potency in inducing
261
neutralizing antibodies against GV JEV. In this report, we used vaccinated mouse sera for all
262
neutralization analyses. It is generally accepted that neutralizing antibodies with a PRNT of 1:10
263
or greater are sufficient to protect from the onset of JE (Kitano, 1996; Kurane & Takasaki, 2000).
264
Further studies using human sera from JE vaccine-immunized healthy and JE case individuals
265
will be needed in order to validate our findings in humans.
266
In contrast to the results of the neutralization tests, the challenge analysis in mice
267
immunized with the inactivated Beijing-1 vaccine showed no clear difference in mortality
268
between Beijing-1-infected and Muar-infected groups for either vaccine concentration. This
269
indicates that the ability of the Beijng-1 vaccine to block the onset of encephalitis by Muar might
270
be comparable to that by Beijing-1 in vivo. Our results indicated that the PRNT titers of 2-fold
271
and 8-fold diluted Beijing-1 vaccine against Muar were 1:20 and 1:10, respectively, which were
272
16-fold less than those against Beijing-1 (1:320 and 1:160, respectively) (Table 2). These
273
findings suggest that although the neutralization potency of the Beijing-1 vaccine against GV
274
JEV is weaker than those against GI and GIII JEV, the potency of the vaccine is not critical to
275
inhibit disease caused by GV JEV, as long as the titer exceeds a threshold of 1:10. We inoculated
276
104 PFU of Muar and Beijing-1 per 5-week-old mouse in the protection assay as this virus titer
277
was sufficient to kill all 3-week-old mice, as shown in Table 1. However, only 50% of the mice
278
were deceased in the non-vaccinated Beijing-1 inoculated group, while over 90% of mice were
279
deceased in the non-vaccinated Muar inoculated group. It is possible that Muar may exhibit a
280
modestly increased virulence compared with Beijing-1 when 5-week-old mice are used for the
281
evaluation of neuroinvasiveness.
282
For more than 50 years, the Muar strain was the only example of GV. Recently, several
283
GV genomes were detected in Culex mosquitoes in China and South Korea (Kim et al., 2015; Li
284
et al., 2011; Takhampunya et al., 2011), but there has been no GV JEV detected in Japan, thus
285
far. However, in both Japan and South Korea, a GIII-to-GI shift occurred in the early 1990s,
286
raising the possibility that GV JEV has already invaded Japan. Recently, Kim et al. suggested
287
that two new vector mosquito species might play a role in the transmission of GV JEV (Kim et
288
al., 2015). Continuous and careful monitoring of the spread of GV viruses must be undertaken.
289
In South Korea, 26 JE patients were reported in 2010, representing a 12-fold increase compared
290
with the mean number that was reported over the previous 26 years (Seo et al., 2013). The
291
relationship between the increase in the number of JE patients in South Korea in 2010 and the
292
emergence of GV JEV remains unknown. Continuous isolation of JEV from JE patients,
293
mosquitoes, and other amplifying hosts is important to understand the effects of the GV
294
emergence.
295 296 297
298
Methods
299
Cells, virus strains, and genetic analysis
300
Vero cells (strain 9013) and BHK-21 cells were maintained at 37 C with 5% CO2 in
301
Eagle’s Minimal Essential Medium supplemented with 10% heat-inactivated fetal bovine serum
302
and 100 U Penicillin-Streptomycin mL-1. JEV strains used in this study are listed in
303
Supplementary table S1. Briefly, we used two JE vaccine strains (Nakayama and Beijing-1),
304
eleven GI strains, three GIII strains, and one GV strain. All of the GI and GIII strains were
305
isolated in Japan, while the GV Muar strain was isolated from a patient with encephalitis in
306
Malaysia in 1952. Working virus stocks were prepared in Vero cells. The nucleotide sequence of
307
the E region containing 1,500 bp were determined for all strains used in this study, including the
308
two vaccine strains (Nakayama and Beijing-1). These nucleotide sequences and their
309
corresponding amino acid sequences were subjected to sequence alignments using GENETYX
310
Ver. 11 genetic software (Genetyx Corporation, Japan).
311 312
Analysis of growth kinetics and plaque morphology
313
The growth properties of the JEV strains were analyzed as described previously (Tajima et
314
al., 2010). Briefly, cells were plated into 6-well culture plates and infected with the JEV strains
315
at a multiplicity of infection of 0.01 p.f.u./cell. Aliquots of the media were removed periodically
316
and their viral titers were determined using a plaque assay on Vero cells. To evaluate the plaque
317
size, Vero cells were plated in 12-well plates and inoculated with the viruses. Four days after
318
inoculation, cells were fixed using a 10% formalin-PBS solution and subsequently stained with
319
methylene blue. The diameter of 20 plaques was measured and the mean plaque size in mm and
320
standard error (SEM) was calculated.
321 322
Production of recombinant JEV
323
Four recombinant molecular clones, rJEV-EBeijing-1-M41/pMW119, rJEV-ENakayama-
324
M41/pMW119, rJEV-EMuar-M41/pMW119, and rJEV-EXZ0934-M41/pMW119, were constructed
325
in the GI JEV strain Mie/41/2002 backbone as described previously (Tajima et al., 2010). Briefly,
326
the E region, from position 927 to 2410, of rJEV(Mie/41/2002)/pMW119 was replaced with the
327
corresponding region of the Beijing-1, Nakayama, Muar, or XZ0934 strain. The recombinant
328
viruses were recovered by transfection of in vitro-transcribed recombinant viral RNA into Vero
329
cells, as previously described (Tajima et al., 2010).
330 331
Mouse vaccination
332
Mouse experiments were performed in accordance with the “Guidelines for animal
333
experiments performed at National Institute of Infectious Diseases” published by the Animal
334
Welfare and Animal Care Committee of the National Institute of Infectious Diseases, Japan.
335
Mouse brain-derived inactivated JE Nakayama vaccine (Biken: lot no. 101-2004) and Vero cell-
336
derived inactivated JE Beijing-1 vaccine (Biken: lot no. 106VC-2009) were used for
337
immunization of the mice. Both vaccines were diluted 1:2 and 1:8 with PBS. Groups of inbred
338
ddY mice (4 weeks old, n=30; Japan SLC) were vaccinated i.p. with 0.5 mL of the diluted
339
vaccines and were immunized again one week after the initial immunization. The immunized
340
mice were sacrificed at 14 days, 2 months, or 4 months after the initial immunization, and their
341
sera were pooled independently (10 mice/pool) for each period. These sera were used to measure
342
neutralization antibody titers as described below.
343
344
Mouse challenge test
345
For evaluation of neuroinvasiveness, groups of ddY mice (3 weeks old, n=10) were
346
inoculated i.p. with 100 l of serially-diluted recombinant virus in 0.9% NaCl solution. The mice
347
were observed for 3 weeks after inoculation to determine survival rates. Survival curve
348
comparisons were performed using Prism software (GraphPad software) statistical analysis with
349
a log-rank (Mantel-Cox) test. To evaluate the protection efficacy of the vaccines, groups of ddY
350
mice (3 weeks old, n=14 or 15) were vaccinated i.p. with 0.5 mL of the inactivated Beijing-1
351
vaccine diluted 1:2 or 1:32 and were immunized again one week after the initial immunization.
352
The immunized mice were inoculated i.p. with 100 l of 104 p.f.u. Beijing-1 and Muar strains.
353
The mice were observed for 3 weeks after inoculation to determine survival rates, and survival
354
curve comparisons were performed as described above. To obtain anti-Muar antisera, ddY mice
355
(3 weeks old, n=4) were inoculated i.p. with 100 l of 10 p.f.u. Muar and were inoculated
356
subsequently with 100 l of 102 p.f.u. Muar at 11 days after the initial inoculation. The Muar-
357
infected mice were sacrificed 22 days after the first inoculation to collect pooled antisera, and the
358
sera were used for neutralization tests.
359 360
Plaque reduction neutralization test (PRNT)
361
Neutralizing antibodies to JEV were measured using the PRNT method. Each JEV strain
362
was combined at a 1:1 ratio with 2-fold serial dilutions (1:10 to 1:640) of the sera from mice
363
immunized with the Nakayama vaccine or the Beijing-1 vaccine and then incubated at 35 C for
364
90 min. Vero cell monolayers were inoculated with these mixtures in 12-well plates and
365
incubated at 35 C for 90 min. Subsequently, overlay medium containing 1% methylcellulose
366
was added and cells were incubated at 35 C for 4 to 6 days. The cells were fixed using a 10%
367
formalin-PBS solution and stained using methylene blue. The PRNT titer was defined as the
368
reciprocal of the highest dilution that resulted in 50% (PRNT50) reduction relative to the non-
369
serum control.
370 371
Acknowledgements
372
This work was supported by a Health and Labour Sciences Research Grant for Research on
373
Emerging and Re-emerging Infectious Diseases by the Ministry of Health, Labor, and Welfare in
374
Japan (H24-Shinkou-007), a Grant-in-Aid for Scientific Research (C) from the JSPS (25460577),
375
and a grant from the Japan Foundation for Pediatric Research (13-008).
376 377
The authors declare that they have no conflict of interest.
378 379
References
380
Arai, S., Matsunaga, Y., Takasaki, T., Tanaka-Taya, K., Taniguchi, K., Okabe, N., Kurane,
381
I. & Vaccine Preventable Diseases Surveillance Program of, J. (2008). Japanese
382
encephalitis: surveillance and elimination effort in Japan from 1982 to 2004. Jpn J Infect
383
Dis 61, 333-338.
384
Campbell, G. L., Hills, S. L., Fischer, M., Jacobson, J. A., Hoke, C. H., Hombach, J. M.,
385
Marfin, A. A., Solomon, T., Tsai, T. F., Tsu, V. D. & Ginsburg, A. S. (2011).
386
Estimated global incidence of Japanese encephalitis: a systematic review. Bull World
387
Health Organ 89, 766-774, 774A-774E.
388
Chen, Y. Y., Fan, Y. C., Tu, W. C., Chang, R. Y., Shih, C. C., Lu, I. H., Chien, M. S., Lee,
389
W. C., Chen, T. H., Chang, G. J. & Chiou, S. S. (2011). Japanese encephalitis virus
390
genotype replacement, Taiwan, 2009-2010. Emerg Infect Dis 17, 2354-2356.
391 392
Crill, W. D. & Chang, G. J. (2004). Localization and characterization of flavivirus envelope glycoprotein cross-reactive epitopes. Journal of virology 78, 13975-13986.
393
de Wispelaere, M., Frenkiel, M. P. & Despres, P. (2015). A Japanese Encephalitis virus
394
genotype 5 molecular clone is highly neuropathogenic in a mouse model: implication of
395
the structural proteins region in virulence. Journal of virology.
396 397 398
Erlanger, T. E., Weiss, S., Keiser, J., Utzinger, J. & Wiedenmayer, K. (2009). Past, present, and future of Japanese encephalitis. Emerg Infect Dis 15, 1-7. Erra, E. O., Askling, H. H., Yoksan, S., Rombo, L., Riutta, J., Vene, S., Lindquist, L.,
399
Vapalahti, O. & Kantele, A. (2013a). Cross-protection elicited by primary and booster
400
vaccinations against Japanese encephalitis: a two-year follow-up study. Vaccine 32, 119-
401
123.
402
Erra, E. O., Askling, H. H., Yoksan, S., Rombo, L., Riutta, J., Vene, S., Lindquist, L.,
403
Vapalahti, O. & Kantele, A. (2013b). Cross-protective capacity of Japanese encephalitis
404
(JE) vaccines against circulating heterologous JE virus genotypes. Clin Infect Dis 56,
405
267-270.
406
Fan, Y. C., Chen, J. M., Chen, Y. Y., Lin, J. W. & Chiou, S. S. (2013). Reduced neutralizing
407
antibody titer against genotype I virus in swine immunized with a live-attenuated
408
genotype III Japanese encephalitis virus vaccine. Vet Microbiol 163, 248-256.
409
Fan, Y. C., Chen, J. M., Chiu, H. C., Chen, Y. Y., Lin, J. W., Shih, C. C., Chen, C. M.,
410
Chang, C. C., Chang, G. J. & Chiou, S. S. (2012). Partially neutralizing potency
411
against emerging genotype I virus among children received formalin-inactivated Japanese
412
encephalitis virus vaccine. PLoS Negl Trop Dis 6, e1834.
413
Fulmali, P. V., Sapkal, G. N., Athawale, S., Gore, M. M., Mishra, A. C. & Bondre, V. P.
414
(2011). Introduction of Japanese encephalitis virus genotype I, India. Emerg Infect Dis 17,
415
319-321.
416
Gao, X., Liu, H., Wang, H., Fu, S., Guo, Z. & Liang, G. (2013). Southernmost Asia is the
417
source of Japanese encephalitis virus (genotype 1) diversity from which the viruses
418
disperse and evolve throughout Asia. PLoS Negl Trop Dis 7, e2459.
419
Goncalvez, A. P., Chien, C. H., Tubthong, K., Gorshkova, I., Roll, C., Donau, O., Schuck,
420
P., Yoksan, S., Wang, S. D., Purcell, R. H. & Lai, C. J. (2008). Humanized
421
monoclonal antibodies derived from chimpanzee Fabs protect against Japanese
422
encephalitis virus in vitro and in vivo. Journal of virology 82, 7009-7021.
423
Hasegawa, H., Yoshida, M., Fujita, S. & Kobayashi, Y. (1994). Comparison of structural
424
proteins among antigenically different Japanese encephalitis virus strains. Vaccine 12,
425
841-844.
426
Ishikawa, T., Abe, M. & Masuda, M. (2015). Construction of an infectious molecular clone of
427
Japanese encephalitis virus genotype V and its derivative subgenomic replicon capable of
428
expressing a foreign gene. Virus research 195, 153-161.
429
Kim, H., Cha, G. W., Jeong, Y. E., Lee, W. G., Chang, K. S., Roh, J. Y., Yang, S. C., Park,
430
M. Y., Park, C. & Shin, E. H. (2015). Detection of Japanese encephalitis virus genotype
431
V in Culex orientalis and Culex pipiens (Diptera: Culicidae) in Korea. PloS one 10,
432
e0116547.
433
Kitano, T. O., A. (1996). Japanese encephalitis vaccine. In Vaccine handbook, pp. 103-113.
434
Edited by Y. A. Arai, T.; Chino, T.; Katow, S.; Miyamura, T.; Nakamura, R.; Oya, A.;
435
Sato, H. Tokyo: Maruzen.
436
Kobayashi, Y., Hasegawa, H. & Yamauchi, T. (1985). Studies on the antigenic structure of
437
Japanese encephalitis virus using monoclonal antibodies. Microbiol Immunol 29, 1069-
438
1082.
439
Kurane, I. & Takasaki, T. (2000). Immunogenicity and protective efficacy of the current
440
inactivated Japanese encephalitis vaccine against different Japanese encephalitis virus
441
strains. Vaccine 18 Suppl 2, 33-35.
442
Li, M. H., Fu, S. H., Chen, W. X., Wang, H. Y., Guo, Y. H., Liu, Q. Y., Li, Y. X., Luo, H. M.,
443
Da, W., Duo Ji, D. Z., Ye, X. M. & Liang, G. D. (2011). Genotype v Japanese
444
encephalitis virus is emerging. PLoS Negl Trop Dis 5, e1231.
445
Liu, X., Yu, Y., Li, M., Liang, G., Wang, H., Jia, L. & Dong, G. (2011). Study on the
446
protective efficacy of SA14-14-2 attenuated Japanese encephalitis against different JE
447
virus isolates circulating in China. Vaccine 29, 2127-2130.
448
Ma, S. P., Yoshida, Y., Makino, Y., Tadano, M., Ono, T. & Ogawa, M. (2003). Short report:
449
a major genotype of Japanese encephalitis virus currently circulating in Japan. Am J Trop
450
Med Hyg 69, 151-154.
451
Monath, T. P., Arroyo, J., Levenbook, I., Zhang, Z. X., Catalan, J., Draper, K. &
452
Guirakhoo, F. (2002). Single mutation in the flavivirus envelope protein hinge region
453
increases neurovirulence for mice and monkeys but decreases viscerotropism for
454
monkeys: relevance to development and safety testing of live, attenuated vaccines.
455
Journal of virology 76, 1932-1943.
456
Morita, K., Tadano, M., Nakaji, S., Kosai, K., Mathenge, E. G., Pandey, B. D., Hasebe, F.,
457
Inoue, S. & Igarashi, A. (2001). Locus of a virus neutralization epitope on the Japanese
458
encephalitis virus envelope protein determined by use of long PCR-based region-specific
459
random mutagenesis. Virology 287, 417-426.
460
Nam, J. H., Chung, Y. J., Ban, S. J., Kim, E. J., Park, Y. K. & Cho, H. W. (1996). Envelope
461
gene sequence variation among Japanese encephalitis viruses isolated in Korea. Acta
462
Virol 40, 303-309.
463
Nga, P. T., del Carmen Parquet, M., Cuong, V. D., Ma, S. P., Hasebe, F., Inoue, S., Makino,
464
Y., Takagi, M., Nam, V. S. & Morita, K. (2004). Shift in Japanese encephalitis virus
465
(JEV) genotype circulating in northern Vietnam: implications for frequent introductions
466
of JEV from Southeast Asia to East Asia. The Journal of general virology 85, 1625-1631.
467
Nitatpattana, N., Dubot-Peres, A., Gouilh, M. A., Souris, M., Barbazan, P., Yoksan, S., de
468
Lamballerie, X. & Gonzalez, J. P. (2008). Change in Japanese encephalitis virus
469
distribution, Thailand. Emerg Infect Dis 14, 1762-1765.
470
Pan, X. L., Liu, H., Wang, H. Y., Fu, S. H., Liu, H. Z., Zhang, H. L., Li, M. H., Gao, X. Y.,
471
Wang, J. L., Sun, X. H., Lu, X. J., Zhai, Y. G., Meng, W. S., He, Y., Wang, H. Q.,
472
Han, N., Wei, B., Wu, Y. G., Feng, Y., Yang, D. J., Wang, L. H., Tang, Q., Xia, G.,
473
Kurane, I., Rayner, S. & Liang, G. D. (2011). Emergence of genotype I of Japanese
474
encephalitis virus as the dominant genotype in Asia. Journal of virology 85, 9847-9853.
475
Pierson, T. C. D., M. S. (2013). Flaviviruses. In Fields Virology, 6 edn, pp. 747-794. Edited by
476 477
D. M. H. Knipe, P. M. Philadelphia: Lippincott Williams & Wilkins. Schuh, A. J., Ward, M. J., Leigh Brown, A. J. & Barrett, A. D. (2014). Dynamics of the
478
Emergence and Establishment of a Newly Dominant Genotype of Japanese Encephalitis
479
Virus throughout Asia. Journal of virology 88, 4522-4532.
480
Seo, H. J., Kim, H. C., Klein, T. A., Ramey, A. M., Lee, J. H., Kyung, S. G., Park, J. Y., Cho,
481
Y. S., Cho, I. S. & Yeh, J. Y. (2013). Molecular detection and genotyping of Japanese
482
encephalitis virus in mosquitoes during a 2010 outbreak in the Republic of Korea. PloS
483
one 8, e55165.
484
Solomon, T., Ni, H., Beasley, D. W., Ekkelenkamp, M., Cardosa, M. J. & Barrett, A. D.
485
(2003). Origin and evolution of Japanese encephalitis virus in southeast Asia. Journal of
486
virology 77, 3091-3098.
487
Sumiyoshi, H., Tignor, G. H. & Shope, R. E. (1995). Characterization of a highly attenuated
488
Japanese encephalitis virus generated from molecularly cloned cDNA. The Journal of
489
infectious diseases 171, 1144-1151.
490
Tajima, S., Nerome, R., Nukui, Y., Kato, F., Takasaki, T. & Kurane, I. (2010). A single
491
mutation in the Japanese encephalitis virus E protein (S123R) increases its growth rate in
492
mouse neuroblastoma cells and its pathogenicity in mice. Virology 396, 298-304.
493
Takhampunya, R., Kim, H. C., Tippayachai, B., Kengluecha, A., Klein, T. A., Lee, W. J.,
494
Grieco, J. & Evans, B. P. (2011). Emergence of Japanese encephalitis virus genotype V
495
in the Republic of Korea. Virol J 8, 449.
496
Tsai, T. F. (2000). New initiatives for the control of Japanese encephalitis by vaccination:
497
minutes of a WHO/CVI meeting, Bangkok, Thailand, 13-15 October 1998. Vaccine 18
498
Suppl 2, 1-25.
499
Vogt, M. R., Moesker, B., Goudsmit, J., Jongeneelen, M., Austin, S. K., Oliphant, T.,
500
Nelson, S., Pierson, T. C., Wilschut, J., Throsby, M. & Diamond, M. S. (2009).
501
Human monoclonal antibodies against West Nile virus induced by natural infection
502
neutralize at a postattachment step. Journal of virology 83, 6494-6507.
503
Wang, H. Y., Takasaki, T., Fu, S. H., Sun, X. H., Zhang, H. L., Wang, Z. X., Hao, Z. Y.,
504
Zhang, J. K., Tang, Q., Kotaki, A., Tajima, S., Liang, X. F., Yang, W. Z., Kurane, I.
505
& Liang, G. D. (2007). Molecular epidemiological analysis of Japanese encephalitis
506
virus in China. The Journal of general virology 88, 885-894.
507
Wu, K. P., Wu, C. W., Tsao, Y. P., Kuo, T. W., Lou, Y. C., Lin, C. W., Wu, S. C. & Cheng,
508
J. W. (2003). Structural basis of a flavivirus recognized by its neutralizing antibody:
509
solution structure of the domain III of the Japanese encephalitis virus envelope protein. J
510
Biol Chem 278, 46007-46013.
511
Yamaguchi, Y., Nukui, Y., Tajima, S., Nerome, R., Kato, F., Watanabe, H., Takasaki, T. &
512
Kurane, I. (2011). An amino acid substitution (V3I) in the Japanese encephalitis virus
513
NS4A protein increases its virulence in mice, but not its growth rate in vitro. The Journal
514
of general virology 92, 1601-1606.
515
Yang, D. K., Kim, B. H., Kweon, C. H., Kwon, J. H., Lim, S. I. & Han, H. R. (2004).
516
Molecular characterization of full-length genome of Japanese encephalitis virus
517
(KV1899) isolated from pigs in Korea. J Vet Sci 5, 197-205.
518
Yoshida, Y., Tabei, Y., Hasegawa, M., Nagashima, M. & Morozumi, S. (2005). Genotypic
519
analysis of Japanese encephalitis virus strains isolated from swine in Tokyo, Japan. Jpn J
520
Infect Dis 58, 259-261.
521
Yun, S. M., Cho, J. E., Ju, Y. R., Kim, S. Y., Ryou, J., Han, M. G., Choi, W. Y. & Jeong, Y.
522
E. (2010). Molecular epidemiology of Japanese encephalitis virus circulating in South
523
Korea, 1983-2005. Virol J 7, 127.
524
Zhang, J. S., Zhao, Q. M., Guo, X. F., Zuo, S. Q., Cheng, J. X., Jia, N., Wu, C., Dai, P. F. &
525
Zhao, J. Y. (2011). Isolation and genetic characteristics of human genotype 1 Japanese
526
encephalitis virus, China, 2009. PloS one 6, e16418.
527 528 529
Zhang, J. S., Zhao, Q. M., Zhang, P. H., Jia, N. & Cao, W. C. (2009). Genomic sequence of a Japanese encephalitis virus isolate from southern China. Arch Virol 154, 1177-1180. Zhao, Z., Date, T., Li, Y., Kato, T., Miyamoto, M., Yasui, K. & Wakita, T. (2005).
530
Characterization of the E-138 (Glu/Lys) mutation in Japanese encephalitis virus by using
531
a stable, full-length, infectious cDNA clone. The Journal of general virology 86, 2209-
532
2220.
533 534 535
536
Table 1. Mouse neuroinvasiveness of the GI, GIII, and GV JEV strains.
Virus Survival† Muar/1952 9/10 Beijing-1 8/10 Mie/41/2002 10/10 103 Muar/1952 1/10 Beijing-1 0/10 Mie/41/2002 8/10 4 10 Muar/1952 0/10 Beijing-1 0/10 Mie/41/2002 7/10 105 Muar/1952 0/10 Beijing-1 0/10 Mie/41/2002 7/10 † No. of mice surviving/no. mice inoculated. Dose 102
537 538 539 540
‡
Average time to death (day) 9 8 7.8 7.2 9 6.6 7.2 10.7 7.7 8.7 11.3
P value‡ (vs. Mie/41/2002)
P value‡ (vs. Muar/1952)
0.32 0.15
Journal of General Virology In vitro Growth, Pathogenicity, and Serological Characteristics of the Japanese Encephalitis Virus Genotype V Muar strain. --Manuscript Draft-Manuscript Number:
VIR-D-15-00245R1
Full Title:
In vitro Growth, Pathogenicity, and Serological Characteristics of the Japanese Encephalitis Virus Genotype V Muar strain.
Short Title:
Characterization of the Japanese Encephalitis Virus Genotype V Muar strain.
Article Type:
Standard
Section/Category:
Animal - Positive-strand RNA Viruses
Corresponding Author:
Shigeru Tajima, Ph.D. National Institute of Infectious Diseases Shinjuku, Tokyo JAPAN
First Author:
Shigeru Tajima, Ph.D.
Order of Authors:
Shigeru Tajima, Ph.D. Kazumi Yagasaki Akira Kotaki Takumi Tomikawa Eri Nakayama Meng Ling Moi Chang-Kweng Lim Masayuki Saijo Ichiro Kurane Tomohiko Takasaki
Abstract:
The characteristics of the genotype V Japanese encephalitis virus (GV JEV) remain poorly understood since only two strains have been isolated, thus far. In this study, we examined the effects of the GV JEV Muar strain on in vitro growth and pathogenicity in mice; we also evaluated the efficacy of inactivated Japanese encephalitis (JE) vaccines against the Muar strain. Although the growth of the Muar strain in mouse neuroblastoma N18 cells was clearly worse than those of the GIII Beijing-1 and GI Mie/41/2002 strains, neuroinvasiveness of the Muar strain was similar to that of the Beijing-1 strain and significantly higher than that of the Mie/41/2002 strain. The results of a plaque reduction neutralization test suggest that the neutralization ability of the JE vaccines against the Muar strain is reduced compared with GI and GIII strains. However, the protection potency of the JE vaccine against the Muar strain was similar to that for the Beijing-1 strain in mice. Our data indicate that GV JEV has unique growth, virulence, and antigenicity features.
Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
Manuscript Including References (Word document) Click here to download Manuscript Including References (Word document): Text(PRNT)150603JGV.docx
1
In vitro Growth, Pathogenicity, and Serological Characteristics of the Japanese
2
Encephalitis Virus Genotype V Muar strain.
3 4
Shigeru Tajima*, Kazumi Yagasaki, Akira Kotaki, Takumi Tomikawa, Eri Nakayama, Meng
5
Ling Moi, Chang-Kweng Lim, Masayuki Saijo, Ichiro Kurane and Tomohiko Takasaki.
6 7
Department of Virology I, National Institute of Infectious Diseases, 1-23-1 Toyama, Shinjuku,
8
Tokyo 162-8640, Japan
9 10
Key words:
11
Japanese encephalitis virus / Genotype / Pathogenicity / Neutralization / Vaccine
12 13
Correspondence: Shigeru Tajima, Ph.D.
14
Department of Virology 1, National Institute of Infectious Diseases
15
1-23-1 Toyama, Shinjuku, Tokyo 162-8640, Japan
16
E-mail: [email protected]
17
Phone/Fax: +81-3-5285-1188
18 19 20
Footnote
21
The GenBank accession numbers for the JEV E gene sequence we determined in this study are as
22
follow: Nakayama strain, AB920348; Beijing-1 strain, AB920347; Kumamoto/65/2005 strain,
23
AB920445; Kumamoto/81/2006 strain, AB920446; Kumamoto/104/2006 strain, AB920447;
24
Chiba/150/2007 strain, AB920448; Chiba/103/2008, AB920449.
25
Summary
26
The characteristics of the genotype V Japanese encephalitis virus (GV JEV) remain poorly
27
understood since only two strains have been isolated, thus far. In this study, we examined the
28
effects of the GV JEV Muar strain on in vitro growth and pathogenicity in mice; we also
29
evaluated the efficacy of inactivated Japanese encephalitis (JE) vaccines against the Muar strain.
30
Although the growth of the Muar strain in mouse neuroblastoma N18 cells was clearly worse
31
than those of the GIII Beijing-1 and GI Mie/41/2002 strains, neuroinvasiveness of the Muar
32
strain was similar to that of the Beijing-1 strain and significantly higher than that of the
33
Mie/41/2002 strain. The results of a plaque reduction neutralization test suggest that the
34
neutralization ability of the JE vaccines against the Muar strain is reduced compared with GI and
35
GIII strains. However, the protection potency of the JE vaccine against the Muar strain was
36
similar to that for the Beijing-1 strain in mice. Our data indicate that GV JEV has unique growth,
37
virulence, and antigenicity features.
38 39
40
Introduction
41
Japanese encephalitis (JE), resulting from an infection of the Japanese encephalitis virus
42
(JEV), is a significant public health problem in many Asian countries. JEV causes serious
43
nervous disorders such as meningitis and encephalitis. There are an estimated 68,000 cases of JE
44
per year, occurring mainly in China, India, and Southeast Asian countries, which result in 15,000
45
fatalities, mostly in children (Campbell et al., 2011; Erlanger et al., 2009; Tsai, 2000). There is
46
no specific treatment available for JE, other than preventive vaccination with JE vaccines. In
47
Japan, the number of reported JE cases has dramatically decreased since the late 1960s, and less
48
than 10 JE cases have occurred in Japan over the past two decades (Arai et al., 2008). This
49
decrease is thought to be due to active vaccination, a decrease in the number of mosquitoes that
50
can transmit JEV, and urbanization. However, a study on the prevalence of JEV that used
51
domestic pigs indicated that, in most regions of Japan, naive pigs are still seroconverted to JEV
52
every year, which suggests that JEV is still a serious threat (Arai et al., 2008).
53
JEV belongs to the genus Flavivirus in the family Flaviviridae and is amplified in a bird
54
and pig-mosquito transmission cycle (Pierson, 2013). The infected mosquitoes, mainly Culex
55
tritaeniorhynchus, transmit JEV to dead-end hosts: either humans or horses. JEV has a single-
56
stranded positive-sense RNA genome, which encodes three structural proteins (C, prM, and E)
57
and seven non-structural proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5). JEV is
58
classified into five genotypes (GI - GV) based on its genomic sequence (Solomon et al., 2003),
59
of which the GIII strain was the most widely distributed and most frequently isolated in JE
60
endemic areas until the 1990s. Consequently, all licensed inactivated and live-attenuated JE
61
vaccines are derived from GIII JEV. However, the major genotype of JEV isolated in Japan
62
changed from GIII to GI strain in the early 1990s (Ma et al., 2003; Yoshida et al., 2005). In
63
recent years, a similar genotype shift has been observed in South Korea, northern Vietnam,
64
China, Taiwan, and Thailand (Chen et al., 2011; Nam et al., 1996; Nga et al., 2004; Nitatpattana
65
et al., 2008; Pan et al., 2011; Wang et al., 2007; Yang et al., 2004; Yun et al., 2010; Zhang et al.,
66
2011; Zhang et al., 2009). GI JEV has also been found in India since 2009 (Fulmali et al., 2011).
67
Thus, the dominant genotype in JEV endemic areas is now GI, although the reasons for the broad
68
shift from GIII to GI remain unclear. The Indonesian/Malaysian region is thought to be the origin
69
of all genotypes of JEV (Solomon et al., 2003), and in several Southeast Asian countries, GI was
70
isolated as a minor JEV strain before the 1990s. Comprehensive phylogenetic analysis of the GI
71
JEV strains demonstrates that the old and the new GI strains form distinct lineages (Gao et al.,
72
2013; Schuh et al., 2014), suggesting that genetic differences in the new GI genome might be
73
responsible for the spread of GI viruses. Recent findings indicate that GI JEV exhibits a higher
74
replicative ability in mosquitoes than does GIII JEV (Schuh et al., 2014).
75
A GV strain of JEV was first isolated from an encephalitis patient in Malaysia in 1952
76
(also referred to as the Muar strain), and it remained the only known instance of this genotype for
77
over 50 years (Hasegawa et al., 1994; Solomon et al., 2003). Recently, the GV JEV genome has
78
been identified in Culex mosquitoes in China in 2009 and in South Korea in 2010 and 2012 (Kim
79
et al., 2015; Li et al., 2011; Takhampunya et al., 2011). A group in China succeeded in isolating
80
a new GV strain (referred to as the XZ0934 strain) from C. tritaeniorhynchus (Li et al., 2011). In
81
South Korea, genomes of GV JEV were detected in Culex bitaeniorhynchus, Culex orientalis,
82
and Culex pipiens, but none were identified in the major JEV vector C. tritaeniorhynchus (Kim
83
et al., 2015; Takhampunya et al., 2011). These findings raise the possibility that the other species
84
of Culex mosquitoes may be involved in the natural transmission cycle of the GV strain of JEV.
85
Thus, GV JEV may also be emerging in other endemic areas, and more attention needs to be
86
focused on investigating the dynamics of circulating JEV strains in these areas.
87
Since only two GV strains, Muar and XZ0934, have been isolated thus far and biological
88
studies on the features of GV JEV have been very limited, the growth and pathogenic properties
89
of GV remain poorly understood. The spread of JEV strains, other than GIII, into endemic areas
90
makes it imperative that systematic evaluation is conducted on the protective efficacy of current
91
GIII-derived JE vaccines against other JEV genotypes. Previous reports call attention to the
92
reduced neutralizing potency of inactivated and live-attenuated GIII JE vaccines against
93
circulating GI JEV (Fan et al., 2013; Fan et al., 2012). On the other hand, other researchers have
94
shown that inactivated and live-attenuated JE vaccines are effective against both GIII and GI
95
JEV strains that are prevalent in China (Liu et al., 2011), and inactivated vaccines used for
96
European travelers induce protective levels of neutralizing antibodies against GI - GIV JEV
97
(Erra et al., 2013a; b). Thus, the efficacy of the current JE vaccines against heterologous
98
genotypes remains unresolved. Furthermore, these studies did not assess the Beijing-1 strain-
99
derived vaccine that is currently available in Japan, the Nakayama strain-derived vaccine, or a
100
GV JEV strain as a target virus.
101
To uncover the characteristics of GV JEV, we investigated in vitro growth and
102
pathogenicity in mice for the GV JEV Muar strain. We also evaluated the protective efficacy of
103
the inactivated JE vaccines against the Muar strain as well as several GI and GIII JEV strains.
104 105 106 107 108
109
Results
110
In vitro growth properties of the Muar strain
111
To characterize the growth of the GV Muar strain in vitro, we first compared the plaques
112
induced by the Muar strain to those of GI Mie/41/2002 strain and GIII Beijing-1 strains in Vero
113
cells (Fig. 1A). The plaque size induced by Muar (average diameter and SEM: 0.98±0.03 mm)
114
was less than that of Mie/41/2002 (1.89±0.05 mm; P < 0.001) but was similar to that of Beijing-
115
1 (1.02±0.03 mm; P = 0.37). Similar results were obtained using a replication kinetics assay for
116
these viruses in Vero cells and BHK-21 cells (Fig. 1B and C). These results suggest that the
117
growth of Muar is similar to that of Beijing-1 in interferon response-deficient cells. Our previous
118
report showed that, in contrast to the results in Vero cells, the steady state number of infectious
119
particles for Beijing-1 was apparently higher than that of Mie/41/2002 in mouse neuroblastoma-
120
derived N18 cells (Tajima et al., 2010). We next examined the growth properties of Muar in N18
121
cells (Fig. 1D). However, the growth rate of Muar in N18 cells was slower than the other strains,
122
and the infectious particles of Muar at steady state were 100-fold and 10-fold less than those of
123
Beijing-1 and Mie/41/2002, respectively. These data indicate that Muar might exhibit a reduced
124
amplification in N18 cells, and the characteristics of the Muar strain are clearly different from
125
those of the Beijing-1 and Mie/41/2002 strains in vitro. Using real-time RT-PCR, we also
126
confirmed in all three cultured cells that the quantity of viral genomes in the supernatant and in
127
the cells infected the viruses corresponded to those of the virus titers (data not shown), indicating
128
that the reduced growth of Muar in cultured cells does not reflect defective particle assembly and
129
secretion.
130 131
Virulence of the Muar strain in mice
132
Next, we examined the pathogenicity of the Muar strain in mice. Mice were infected
133
intraperitoneally (i.p.) with Muar, Beijing-1, or Mie/41/2002 and observed for 3 weeks (Table 1).
134
Our previous report using the Beijing-1 and Mie/41/2002 strains demonstrated that the growth
135
characteristics of some JEV strains in N18 cells were positively correlated with
136
neuroinvasiveness in mice (Tajima et al., 2010); therefore, we speculated that the Muar strain
137
might exhibit a lower virulence in mice compared with the Beijing-1 and Mie/41/2002 strains. In
138
the 103, 104, and 105 plaque forming units (p.f.u.)-infected groups, more than 90% of the mice
139
infected with Muar or Beijing-1 died; however, less than 30% of the mice infected with
140
Mie/41/2002 died. The survival curves for the Muar-infected groups were similar to those of the
141
Beijing-1-infceted groups. These results indicate that, in contrast to the growth kinetics in N18
142
cells, neuroinvasiveness of Muar is equivalent to that of Beijing-1 and significantly higher than
143
that of Mie/41/2002 in mice.
144 145
The neutralizing ability of inactivated JE vaccines against Muar and recombinant
146
intertypic JEVs.
147
We next evaluated the ability of inactivated Nakayama and Beijing-1 vaccines, which
148
were produced in Japan, to induce neutralizing antibodies against Muar. We also used three GIII
149
and 11 GI viruses with differing amino acid sequences in the E protein as target viruses
150
(Supplementary table 1). The E protein is the major structural protein that constitutes the surface
151
structure of the virus particles, and a majority of neutralizing antibodies bind to the E protein and
152
can inhibit viral infection (Wu et al., 2003). Serum samples collected from mice at 14 days after
153
initial immunization with 2-fold-diluted and 8-fold-diluted JE vaccines were used for the plaque
154
reduction neutralization test (PRNT), and the value of 50% reduction relative to the non-serum
155
control (PRNT50) for each virus was determined (Table 2). In the sera of mice that were
156
immunized with 2-fold dilutions, the PRNT titer against GI and GIII viruses was between 1:80
157
and 1:320. In those with 8-fold dilutions, it was between 1:40 and 1:160. We statistically
158
compared the PRNT titers of sera against GI viruses to those against GIII viruses using a
159
parametric Welch’s t-test and a non-parametric Mann-Whitney U-test. No significant differences
160
(P > 0.05) were observed between the GI group and GIII group in all cases, indicating that the
161
neutralizing efficacy of inactivated Nakayama and Beijing-1 vaccines against the GI strains is
162
similar to that for the GIII strains. In contrast to the GI and GIII strains, the PRNT titers in the
163
sera of mice immunized against GV Muar were 4-fold less than those against the GI and GIII
164
strains. These results suggest that the neutralizing capacity of the JE vaccines against GV JEV
165
was considerably lower than that against GI and GIII JEV. We also examined the neutralizing
166
potency (PRNT) of sera obtained from mice at 2 and 4 months after the initial immunization for
167
the 2-fold-diluted and 8-fold-diluted Nakayama and Beijing-1 vaccines (Table 3). Two vaccine
168
strains, two GI strains, three GIII strains, and one GV strain were used as target viruses. The
169
PRNT titers for GV Muar were less than those for the GI and GIII JEV, indicating that the
170
neutralization capacity of the inactivated JE vaccines is relatively weak against the GV Muar
171
strain.
172
A comparison of the E protein amino acid sequences amount the viral genotypes showed a
173
higher homology between GI and GIII than that between GI and GV or that between GIII and
174
GV (data not shown). There were 40 residues (8.0%) that differed between Beijing-1 and Muar,
175
and 44 residues (8.8%) were different between Nakayama and Muar. In contrast, less than 12
176
residues (2.4%) were different between the vaccine strains and the GI strains, and less than 8
177
residues (1.6%) were different between the vaccine strains and the GIII strains (Supplementary
178
table 2 and 3). Other GV strains, XZ0934 and 10-1827, also showed similar homology to the
179
vaccine strains, and only two residues (0.4%) were different between the XZ0934 and 10-1827
180
strains. These findings raise the possibility that the neutralization potency of the inactivated
181
vaccines against the other GV strains is also relatively weak. To examine the efficacy of the
182
vaccines against GV strains other than Muar, we generated recombinant intertypic JEVs rJEV-
183
EMuar-M41 and rJEV-EXZ0934-M41, which contain the majority of the E region of Muar and
184
XZ0934, respectively, in the backbone of the Mie/41/2002 genome. We also generated rJEV-
185
ENakayama-M41 and rJEV-EBeijing-1-M41 as control viruses for use in the neutralization assay
186
(Tajima et al., 2010). The PRNT titers in the sera from mice immunized with the vaccines
187
against rJEV-EMuar-M41 and rJEV-EXZ0934-M41 were clearly less than those against the parent
188
virus Mie/41/2002, rJEV-ENakayama-M41, and rJEV-EBeijing-1-M41 (Table 4). These results
189
indicate that the neutralization ability of the vaccines may be weak against both Muar and other
190
GV JEV strains.
191 192
Protection efficacy of the JE vaccines against Muar infection in mice
193
To examine whether immunization of mice with the inactivated JE vaccines inhibits the
194
onset of Muar-induced disease, mice that were vaccinated with the inactivated Beijing-1 vaccine
195
were inoculated i.p. with 104 PFU of Muar or Beijing-1 strains (Table 5). Mice vaccinated with
196
2-fold diluted vaccine had a 100% survival rate after inoculation with Beijing-1 or Muar.
197
Fourteen of 16 mice (87.5%) and 12 of 16 mice (75%) that were vaccinated with 32-fold diluted
198
vaccine survived after infection with Beijing-1 or Muar, respectively. There was no significant
199
difference in the survival rates between Beijing-1-infected and Muar-infected groups. These
200
results suggest that the protection efficacy of the Beijing-1 vaccine against infection with Muar is
201
similar to that against infection with Beijing-1 in vivo.
202 203
The neutralizing ability of the sera from Muar-infected mice against various JEVs
204
Finally, we assessed the ability of GV Muar-infected mouse sera to block the infection of
205
GI and GIII JEVs in vitro using the neutralization assay (Table 6). The PRNT titer against Muar
206
was 2- or 4-fold higher than those against two GI, three GIII, and Beijing-1 strains, though the
207
titer against Nakayama was equivalent to that against Muar. Thus, we confirmed that sera from
208
Muar-infected mice were able to neutralize the infection of GI and GIII strains; however, the
209
neutralization potency of these sera against GI and GIII viruses was weaker than that against
210
Muar.
211 212
213
Discussion
214
To date, only two GV JEV strains have been isolated and amplified in vitro. Therefore, the
215
characteristics of GV JEV remains poorly understood. In this study, we examined the in vitro
216
growth properties and in vivo pathogenicity of the GV strain Muar. Furthermore, we investigated
217
the potential of inactivated JE vaccines that are currently available in Japan to induce a
218
protective immune response against GV JEV using mouse models. Our results indicate that GV
219
JEV has different growth, pathogenic, and antigenic characteristics compared with GI and GIII
220
JEV.
221
Our previous report suggested that there is a correlation between the growth characteristics
222
of JEV in mouse neuroblastoma N18 cells and virulence in mice (Tajima et al., 2010). The
223
growth rate of Muar was less than that of the highly virulent GIII Beijing-1 strain and the
224
moderately virulent GI Mie/41/2002 strain in N18 cells; however, the neuroinvasiveness of Muar
225
was similar to that of Beijing-1. These results suggest that the growth properties of JEV in N18
226
cells do not necessarily reflect pathogenicity in mice. There is little information regarding the
227
regions of the Muar genome that are associated with virulence in mice. We previously
228
demonstrated that the E protein contributes to the difference in virulence between Mie/41/2002
229
and Beijing-1 (Tajima et al., 2010). Other reports have also highlighted the importance of the E
230
protein as a determinant of pathogenicity (Monath et al., 2002; Sumiyoshi et al., 1995; Zhao et
231
al., 2005). On the other hand, our group determined that an amino acid substitution in the JEV
232
NS4A protein alters virulence in mice (Yamaguchi et al., 2011). To clarify the regions
233
responsible for increased pathogenicity in the Muar genome, various GI-GV intertypic and point
234
mutant viruses may be required. Recently, Ishikawa et al. identified the infectious molecular
235
clone and reporter replicon of the Muar strain, which is also beneficial for generally
236
understanding the characteristics of GV JEV (Ishikawa et al., 2015). In addition, de Wispelaere
237
et al. constructed a molecular clone of the GV XZ0934 JEV strain and demonstrated that this
238
recombinant GV virus exhibited a greater pathogenicity in BALB/c mice than that of GIII JEV
239
(de Wispelaere et al., 2015).
240
The neutralization tests using the vaccinated mouse sera indicated that both the Nakayama
241
and Beijing-1 vaccines induce sufficient neutralizing antibodies against GI JEV. However, these
242
vaccines induce fewer neutralizing antibodies against GV Muar or the recombinant intertypic
243
JEVs, which contain the E protein from the GV Muar or XZ0934 strains. Furthermore, the sera
244
from Muar-infected mice had a lower neutralization titer against GI and GIII viruses compared
245
with the Muar strain. These data indicate that GV JEV may be distinct from GI and GIII JEV in
246
antigenicity. The E protein sequences between GI and GIII JEV differ by approximately 3%
247
(Solomon et al., 2003). However, this number increases to 8.8 - 9.2% between GV strains and
248
the Nakayama strain, and 8.2 - 8.6% between GV strains and the Beijing-1 strain
249
(Supplementary table 2 and 3). The decrease in homology between the vaccine strains and the
250
GV strains might be responsible for the observed weak efficacy of the vaccines against the Muar
251
strain and recombinant JEVs that contain GV E proteins. The amino acid differences between the
252
vaccine strains and the GI strains did not exceed 2.4%, indicating that the difference may not
253
influence the efficacy of the vaccines. Previous studies have identified individual residues within
254
the E protein that are related to the recognition of JEV by neutralizing antibodies (Crill & Chang,
255
2004; Goncalvez et al., 2008; Kobayashi et al., 1985; Morita et al., 2001; Vogt et al., 2009; Wu
256
et al., 2003). Although the sequence variation between the vaccine strains and GV strains do not
257
correspond to epitope residues, only a limited number of residues could be responsible for the
258
decreased neutralization ability of the vaccines against GV strains. All inactivated and live-
259
attenuated JE vaccines that are currently available are derived from GIII strains. Our results
260
indicate that live-attenuated GIII-derived vaccines may have a reduced potency in inducing
261
neutralizing antibodies against GV JEV. In this report, we used vaccinated mouse sera for all
262
neutralization analyses. It is generally accepted that neutralizing antibodies with a PRNT of 1:10
263
or greater are sufficient to protect from the onset of JE (Kitano, 1996; Kurane & Takasaki, 2000).
264
Further studies using human sera from JE vaccine-immunized healthy and JE case individuals
265
will be needed in order to validate our findings in humans.
266
In contrast to the results of the neutralization tests, the challenge analysis in mice
267
immunized with the inactivated Beijing-1 vaccine showed no clear difference in mortality
268
between Beijing-1-infected and Muar-infected groups for either vaccine concentration. This
269
indicates that the ability of the Beijng-1 vaccine to block the onset of encephalitis by Muar might
270
be comparable to that by Beijing-1 in vivo. Our results indicated that the PRNT titers of 2-fold
271
and 8-fold diluted Beijing-1 vaccine against Muar were 1:20 and 1:10, respectively, which were
272
16-fold less than those against Beijing-1 (1:320 and 1:160, respectively) (Table 2). These
273
findings suggest that although the neutralization potency of the Beijing-1 vaccine against GV
274
JEV is weaker than those against GI and GIII JEV, the potency of the vaccine is not critical to
275
inhibit disease caused by GV JEV, as long as the titer exceeds a threshold of 1:10. We inoculated
276
104 PFU of Muar and Beijing-1 per 5-week-old mouse in the protection assay as this virus titer
277
was sufficient to kill all 3-week-old mice, as shown in Table 1. However, only 50% of the mice
278
were deceased in the non-vaccinated Beijing-1 inoculated group, while over 90% of mice were
279
deceased in the non-vaccinated Muar inoculated group. It is possible that Muar may exhibit a
280
modestly increased virulence compared with Beijing-1 when 5-week-old mice are used for the
281
evaluation of neuroinvasiveness.
282
For more than 50 years, the Muar strain was the only example of GV. Recently, several
283
GV genomes were detected in Culex mosquitoes in China and South Korea (Kim et al., 2015; Li
284
et al., 2011; Takhampunya et al., 2011), but there has been no GV JEV detected in Japan, thus
285
far. However, in both Japan and South Korea, a GIII-to-GI shift occurred in the early 1990s,
286
raising the possibility that GV JEV has already invaded Japan. Recently, Kim et al. suggested
287
that two new vector mosquito species might play a role in the transmission of GV JEV (Kim et
288
al., 2015). Continuous and careful monitoring of the spread of GV viruses must be undertaken.
289
In South Korea, 26 JE patients were reported in 2010, representing a 12-fold increase compared
290
with the mean number that was reported over the previous 26 years (Seo et al., 2013). The
291
relationship between the increase in the number of JE patients in South Korea in 2010 and the
292
emergence of GV JEV remains unknown. Continuous isolation of JEV from JE patients,
293
mosquitoes, and other amplifying hosts is important to understand the effects of the GV
294
emergence.
295 296 297
298
Methods
299
Cells, virus strains, and genetic analysis
300
Vero cells (strain 9013) and BHK-21 cells were maintained at 37 C with 5% CO2 in
301
Eagle’s Minimal Essential Medium supplemented with 10% heat-inactivated fetal bovine serum
302
and 100 U Penicillin-Streptomycin mL-1. JEV strains used in this study are listed in
303
Supplementary table S1. Briefly, we used two JE vaccine strains (Nakayama and Beijing-1),
304
eleven GI strains, three GIII strains, and one GV strain. All of the GI and GIII strains were
305
isolated in Japan, while the GV Muar strain was isolated from a patient with encephalitis in
306
Malaysia in 1952. Working virus stocks were prepared in Vero cells. The nucleotide sequence of
307
the E region containing 1,500 bp were determined for all strains used in this study, including the
308
two vaccine strains (Nakayama and Beijing-1). These nucleotide sequences and their
309
corresponding amino acid sequences were subjected to sequence alignments using GENETYX
310
Ver. 11 genetic software (Genetyx Corporation, Japan).
311 312
Analysis of growth kinetics and plaque morphology
313
The growth properties of the JEV strains were analyzed as described previously (Tajima et
314
al., 2010). Briefly, cells were plated into 6-well culture plates and infected with the JEV strains
315
at a multiplicity of infection of 0.01 p.f.u./cell. Aliquots of the media were removed periodically
316
and their viral titers were determined using a plaque assay on Vero cells. To evaluate the plaque
317
size, Vero cells were plated in 12-well plates and inoculated with the viruses. Four days after
318
inoculation, cells were fixed using a 10% formalin-PBS solution and subsequently stained with
319
methylene blue. The diameter of 20 plaques was measured and the mean plaque size in mm and
320
standard error (SEM) was calculated.
321 322
Production of recombinant JEV
323
Four recombinant molecular clones, rJEV-EBeijing-1-M41/pMW119, rJEV-ENakayama-
324
M41/pMW119, rJEV-EMuar-M41/pMW119, and rJEV-EXZ0934-M41/pMW119, were constructed
325
in the GI JEV strain Mie/41/2002 backbone as described previously (Tajima et al., 2010). Briefly,
326
the E region, from position 927 to 2410, of rJEV(Mie/41/2002)/pMW119 was replaced with the
327
corresponding region of the Beijing-1, Nakayama, Muar, or XZ0934 strain. The recombinant
328
viruses were recovered by transfection of in vitro-transcribed recombinant viral RNA into Vero
329
cells, as previously described (Tajima et al., 2010).
330 331
Mouse vaccination
332
Mouse experiments were performed in accordance with the “Guidelines for animal
333
experiments performed at National Institute of Infectious Diseases” published by the Animal
334
Welfare and Animal Care Committee of the National Institute of Infectious Diseases, Japan.
335
Mouse brain-derived inactivated JE Nakayama vaccine (Biken: lot no. 101-2004) and Vero cell-
336
derived inactivated JE Beijing-1 vaccine (Biken: lot no. 106VC-2009) were used for
337
immunization of the mice. Both vaccines were diluted 1:2 and 1:8 with PBS. Groups of inbred
338
ddY mice (4 weeks old, n=30; Japan SLC) were vaccinated i.p. with 0.5 mL of the diluted
339
vaccines and were immunized again one week after the initial immunization. The immunized
340
mice were sacrificed at 14 days, 2 months, or 4 months after the initial immunization, and their
341
sera were pooled independently (10 mice/pool) for each period. These sera were used to measure
342
neutralization antibody titers as described below.
343
344
Mouse challenge test
345
For evaluation of neuroinvasiveness, groups of ddY mice (3 weeks old, n=10) were
346
inoculated i.p. with 100 l of serially-diluted recombinant virus in 0.9% NaCl solution. The mice
347
were observed for 3 weeks after inoculation to determine survival rates. Survival curve
348
comparisons were performed using Prism software (GraphPad software) statistical analysis with
349
a log-rank (Mantel-Cox) test. To evaluate the protection efficacy of the vaccines, groups of ddY
350
mice (3 weeks old, n=14 or 15) were vaccinated i.p. with 0.5 mL of the inactivated Beijing-1
351
vaccine diluted 1:2 or 1:32 and were immunized again one week after the initial immunization.
352
The immunized mice were inoculated i.p. with 100 l of 104 p.f.u. Beijing-1 and Muar strains.
353
The mice were observed for 3 weeks after inoculation to determine survival rates, and survival
354
curve comparisons were performed as described above. To obtain anti-Muar antisera, ddY mice
355
(3 weeks old, n=4) were inoculated i.p. with 100 l of 10 p.f.u. Muar and were inoculated
356
subsequently with 100 l of 102 p.f.u. Muar at 11 days after the initial inoculation. The Muar-
357
infected mice were sacrificed 22 days after the first inoculation to collect pooled antisera, and the
358
sera were used for neutralization tests.
359 360
Plaque reduction neutralization test (PRNT)
361
Neutralizing antibodies to JEV were measured using the PRNT method. Each JEV strain
362
was combined at a 1:1 ratio with 2-fold serial dilutions (1:10 to 1:640) of the sera from mice
363
immunized with the Nakayama vaccine or the Beijing-1 vaccine and then incubated at 35 C for
364
90 min. Vero cell monolayers were inoculated with these mixtures in 12-well plates and
365
incubated at 35 C for 90 min. Subsequently, overlay medium containing 1% methylcellulose
366
was added and cells were incubated at 35 C for 4 to 6 days. The cells were fixed using a 10%
367
formalin-PBS solution and stained using methylene blue. The PRNT titer was defined as the
368
reciprocal of the highest dilution that resulted in 50% (PRNT50) reduction relative to the non-
369
serum control.
370 371
Acknowledgements
372
This work was supported by a Health and Labour Sciences Research Grant for Research on
373
Emerging and Re-emerging Infectious Diseases by the Ministry of Health, Labor, and Welfare in
374
Japan (H24-Shinkou-007), a Grant-in-Aid for Scientific Research (C) from the JSPS (25460577),
375
and a grant from the Japan Foundation for Pediatric Research (13-008).
376 377
The authors declare that they have no conflict of interest.
378 379
References
380
Arai, S., Matsunaga, Y., Takasaki, T., Tanaka-Taya, K., Taniguchi, K., Okabe, N., Kurane,
381
I. & Vaccine Preventable Diseases Surveillance Program of, J. (2008). Japanese
382
encephalitis: surveillance and elimination effort in Japan from 1982 to 2004. Jpn J Infect
383
Dis 61, 333-338.
384
Campbell, G. L., Hills, S. L., Fischer, M., Jacobson, J. A., Hoke, C. H., Hombach, J. M.,
385
Marfin, A. A., Solomon, T., Tsai, T. F., Tsu, V. D. & Ginsburg, A. S. (2011).
386
Estimated global incidence of Japanese encephalitis: a systematic review. Bull World
387
Health Organ 89, 766-774, 774A-774E.
388
Chen, Y. Y., Fan, Y. C., Tu, W. C., Chang, R. Y., Shih, C. C., Lu, I. H., Chien, M. S., Lee,
389
W. C., Chen, T. H., Chang, G. J. & Chiou, S. S. (2011). Japanese encephalitis virus
390
genotype replacement, Taiwan, 2009-2010. Emerg Infect Dis 17, 2354-2356.
391 392
Crill, W. D. & Chang, G. J. (2004). Localization and characterization of flavivirus envelope glycoprotein cross-reactive epitopes. Journal of virology 78, 13975-13986.
393
de Wispelaere, M., Frenkiel, M. P. & Despres, P. (2015). A Japanese Encephalitis virus
394
genotype 5 molecular clone is highly neuropathogenic in a mouse model: implication of
395
the structural proteins region in virulence. Journal of virology.
396 397 398
Erlanger, T. E., Weiss, S., Keiser, J., Utzinger, J. & Wiedenmayer, K. (2009). Past, present, and future of Japanese encephalitis. Emerg Infect Dis 15, 1-7. Erra, E. O., Askling, H. H., Yoksan, S., Rombo, L., Riutta, J., Vene, S., Lindquist, L.,
399
Vapalahti, O. & Kantele, A. (2013a). Cross-protection elicited by primary and booster
400
vaccinations against Japanese encephalitis: a two-year follow-up study. Vaccine 32, 119-
401
123.
402
Erra, E. O., Askling, H. H., Yoksan, S., Rombo, L., Riutta, J., Vene, S., Lindquist, L.,
403
Vapalahti, O. & Kantele, A. (2013b). Cross-protective capacity of Japanese encephalitis
404
(JE) vaccines against circulating heterologous JE virus genotypes. Clin Infect Dis 56,
405
267-270.
406
Fan, Y. C., Chen, J. M., Chen, Y. Y., Lin, J. W. & Chiou, S. S. (2013). Reduced neutralizing
407
antibody titer against genotype I virus in swine immunized with a live-attenuated
408
genotype III Japanese encephalitis virus vaccine. Vet Microbiol 163, 248-256.
409
Fan, Y. C., Chen, J. M., Chiu, H. C., Chen, Y. Y., Lin, J. W., Shih, C. C., Chen, C. M.,
410
Chang, C. C., Chang, G. J. & Chiou, S. S. (2012). Partially neutralizing potency
411
against emerging genotype I virus among children received formalin-inactivated Japanese
412
encephalitis virus vaccine. PLoS Negl Trop Dis 6, e1834.
413
Fulmali, P. V., Sapkal, G. N., Athawale, S., Gore, M. M., Mishra, A. C. & Bondre, V. P.
414
(2011). Introduction of Japanese encephalitis virus genotype I, India. Emerg Infect Dis 17,
415
319-321.
416
Gao, X., Liu, H., Wang, H., Fu, S., Guo, Z. & Liang, G. (2013). Southernmost Asia is the
417
source of Japanese encephalitis virus (genotype 1) diversity from which the viruses
418
disperse and evolve throughout Asia. PLoS Negl Trop Dis 7, e2459.
419
Goncalvez, A. P., Chien, C. H., Tubthong, K., Gorshkova, I., Roll, C., Donau, O., Schuck,
420
P., Yoksan, S., Wang, S. D., Purcell, R. H. & Lai, C. J. (2008). Humanized
421
monoclonal antibodies derived from chimpanzee Fabs protect against Japanese
422
encephalitis virus in vitro and in vivo. Journal of virology 82, 7009-7021.
423
Hasegawa, H., Yoshida, M., Fujita, S. & Kobayashi, Y. (1994). Comparison of structural
424
proteins among antigenically different Japanese encephalitis virus strains. Vaccine 12,
425
841-844.
426
Ishikawa, T., Abe, M. & Masuda, M. (2015). Construction of an infectious molecular clone of
427
Japanese encephalitis virus genotype V and its derivative subgenomic replicon capable of
428
expressing a foreign gene. Virus research 195, 153-161.
429
Kim, H., Cha, G. W., Jeong, Y. E., Lee, W. G., Chang, K. S., Roh, J. Y., Yang, S. C., Park,
430
M. Y., Park, C. & Shin, E. H. (2015). Detection of Japanese encephalitis virus genotype
431
V in Culex orientalis and Culex pipiens (Diptera: Culicidae) in Korea. PloS one 10,
432
e0116547.
433
Kitano, T. O., A. (1996). Japanese encephalitis vaccine. In Vaccine handbook, pp. 103-113.
434
Edited by Y. A. Arai, T.; Chino, T.; Katow, S.; Miyamura, T.; Nakamura, R.; Oya, A.;
435
Sato, H. Tokyo: Maruzen.
436
Kobayashi, Y., Hasegawa, H. & Yamauchi, T. (1985). Studies on the antigenic structure of
437
Japanese encephalitis virus using monoclonal antibodies. Microbiol Immunol 29, 1069-
438
1082.
439
Kurane, I. & Takasaki, T. (2000). Immunogenicity and protective efficacy of the current
440
inactivated Japanese encephalitis vaccine against different Japanese encephalitis virus
441
strains. Vaccine 18 Suppl 2, 33-35.
442
Li, M. H., Fu, S. H., Chen, W. X., Wang, H. Y., Guo, Y. H., Liu, Q. Y., Li, Y. X., Luo, H. M.,
443
Da, W., Duo Ji, D. Z., Ye, X. M. & Liang, G. D. (2011). Genotype v Japanese
444
encephalitis virus is emerging. PLoS Negl Trop Dis 5, e1231.
445
Liu, X., Yu, Y., Li, M., Liang, G., Wang, H., Jia, L. & Dong, G. (2011). Study on the
446
protective efficacy of SA14-14-2 attenuated Japanese encephalitis against different JE
447
virus isolates circulating in China. Vaccine 29, 2127-2130.
448
Ma, S. P., Yoshida, Y., Makino, Y., Tadano, M., Ono, T. & Ogawa, M. (2003). Short report:
449
a major genotype of Japanese encephalitis virus currently circulating in Japan. Am J Trop
450
Med Hyg 69, 151-154.
451
Monath, T. P., Arroyo, J., Levenbook, I., Zhang, Z. X., Catalan, J., Draper, K. &
452
Guirakhoo, F. (2002). Single mutation in the flavivirus envelope protein hinge region
453
increases neurovirulence for mice and monkeys but decreases viscerotropism for
454
monkeys: relevance to development and safety testing of live, attenuated vaccines.
455
Journal of virology 76, 1932-1943.
456
Morita, K., Tadano, M., Nakaji, S., Kosai, K., Mathenge, E. G., Pandey, B. D., Hasebe, F.,
457
Inoue, S. & Igarashi, A. (2001). Locus of a virus neutralization epitope on the Japanese
458
encephalitis virus envelope protein determined by use of long PCR-based region-specific
459
random mutagenesis. Virology 287, 417-426.
460
Nam, J. H., Chung, Y. J., Ban, S. J., Kim, E. J., Park, Y. K. & Cho, H. W. (1996). Envelope
461
gene sequence variation among Japanese encephalitis viruses isolated in Korea. Acta
462
Virol 40, 303-309.
463
Nga, P. T., del Carmen Parquet, M., Cuong, V. D., Ma, S. P., Hasebe, F., Inoue, S., Makino,
464
Y., Takagi, M., Nam, V. S. & Morita, K. (2004). Shift in Japanese encephalitis virus
465
(JEV) genotype circulating in northern Vietnam: implications for frequent introductions
466
of JEV from Southeast Asia to East Asia. The Journal of general virology 85, 1625-1631.
467
Nitatpattana, N., Dubot-Peres, A., Gouilh, M. A., Souris, M., Barbazan, P., Yoksan, S., de
468
Lamballerie, X. & Gonzalez, J. P. (2008). Change in Japanese encephalitis virus
469
distribution, Thailand. Emerg Infect Dis 14, 1762-1765.
470
Pan, X. L., Liu, H., Wang, H. Y., Fu, S. H., Liu, H. Z., Zhang, H. L., Li, M. H., Gao, X. Y.,
471
Wang, J. L., Sun, X. H., Lu, X. J., Zhai, Y. G., Meng, W. S., He, Y., Wang, H. Q.,
472
Han, N., Wei, B., Wu, Y. G., Feng, Y., Yang, D. J., Wang, L. H., Tang, Q., Xia, G.,
473
Kurane, I., Rayner, S. & Liang, G. D. (2011). Emergence of genotype I of Japanese
474
encephalitis virus as the dominant genotype in Asia. Journal of virology 85, 9847-9853.
475
Pierson, T. C. D., M. S. (2013). Flaviviruses. In Fields Virology, 6 edn, pp. 747-794. Edited by
476 477
D. M. H. Knipe, P. M. Philadelphia: Lippincott Williams & Wilkins. Schuh, A. J., Ward, M. J., Leigh Brown, A. J. & Barrett, A. D. (2014). Dynamics of the
478
Emergence and Establishment of a Newly Dominant Genotype of Japanese Encephalitis
479
Virus throughout Asia. Journal of virology 88, 4522-4532.
480
Seo, H. J., Kim, H. C., Klein, T. A., Ramey, A. M., Lee, J. H., Kyung, S. G., Park, J. Y., Cho,
481
Y. S., Cho, I. S. & Yeh, J. Y. (2013). Molecular detection and genotyping of Japanese
482
encephalitis virus in mosquitoes during a 2010 outbreak in the Republic of Korea. PloS
483
one 8, e55165.
484
Solomon, T., Ni, H., Beasley, D. W., Ekkelenkamp, M., Cardosa, M. J. & Barrett, A. D.
485
(2003). Origin and evolution of Japanese encephalitis virus in southeast Asia. Journal of
486
virology 77, 3091-3098.
487
Sumiyoshi, H., Tignor, G. H. & Shope, R. E. (1995). Characterization of a highly attenuated
488
Japanese encephalitis virus generated from molecularly cloned cDNA. The Journal of
489
infectious diseases 171, 1144-1151.
490
Tajima, S., Nerome, R., Nukui, Y., Kato, F., Takasaki, T. & Kurane, I. (2010). A single
491
mutation in the Japanese encephalitis virus E protein (S123R) increases its growth rate in
492
mouse neuroblastoma cells and its pathogenicity in mice. Virology 396, 298-304.
493
Takhampunya, R., Kim, H. C., Tippayachai, B., Kengluecha, A., Klein, T. A., Lee, W. J.,
494
Grieco, J. & Evans, B. P. (2011). Emergence of Japanese encephalitis virus genotype V
495
in the Republic of Korea. Virol J 8, 449.
496
Tsai, T. F. (2000). New initiatives for the control of Japanese encephalitis by vaccination:
497
minutes of a WHO/CVI meeting, Bangkok, Thailand, 13-15 October 1998. Vaccine 18
498
Suppl 2, 1-25.
499
Vogt, M. R., Moesker, B., Goudsmit, J., Jongeneelen, M., Austin, S. K., Oliphant, T.,
500
Nelson, S., Pierson, T. C., Wilschut, J., Throsby, M. & Diamond, M. S. (2009).
501
Human monoclonal antibodies against West Nile virus induced by natural infection
502
neutralize at a postattachment step. Journal of virology 83, 6494-6507.
503
Wang, H. Y., Takasaki, T., Fu, S. H., Sun, X. H., Zhang, H. L., Wang, Z. X., Hao, Z. Y.,
504
Zhang, J. K., Tang, Q., Kotaki, A., Tajima, S., Liang, X. F., Yang, W. Z., Kurane, I.
505
& Liang, G. D. (2007). Molecular epidemiological analysis of Japanese encephalitis
506
virus in China. The Journal of general virology 88, 885-894.
507
Wu, K. P., Wu, C. W., Tsao, Y. P., Kuo, T. W., Lou, Y. C., Lin, C. W., Wu, S. C. & Cheng,
508
J. W. (2003). Structural basis of a flavivirus recognized by its neutralizing antibody:
509
solution structure of the domain III of the Japanese encephalitis virus envelope protein. J
510
Biol Chem 278, 46007-46013.
511
Yamaguchi, Y., Nukui, Y., Tajima, S., Nerome, R., Kato, F., Watanabe, H., Takasaki, T. &
512
Kurane, I. (2011). An amino acid substitution (V3I) in the Japanese encephalitis virus
513
NS4A protein increases its virulence in mice, but not its growth rate in vitro. The Journal
514
of general virology 92, 1601-1606.
515
Yang, D. K., Kim, B. H., Kweon, C. H., Kwon, J. H., Lim, S. I. & Han, H. R. (2004).
516
Molecular characterization of full-length genome of Japanese encephalitis virus
517
(KV1899) isolated from pigs in Korea. J Vet Sci 5, 197-205.
518
Yoshida, Y., Tabei, Y., Hasegawa, M., Nagashima, M. & Morozumi, S. (2005). Genotypic
519
analysis of Japanese encephalitis virus strains isolated from swine in Tokyo, Japan. Jpn J
520
Infect Dis 58, 259-261.
521
Yun, S. M., Cho, J. E., Ju, Y. R., Kim, S. Y., Ryou, J., Han, M. G., Choi, W. Y. & Jeong, Y.
522
E. (2010). Molecular epidemiology of Japanese encephalitis virus circulating in South
523
Korea, 1983-2005. Virol J 7, 127.
524
Zhang, J. S., Zhao, Q. M., Guo, X. F., Zuo, S. Q., Cheng, J. X., Jia, N., Wu, C., Dai, P. F. &
525
Zhao, J. Y. (2011). Isolation and genetic characteristics of human genotype 1 Japanese
526
encephalitis virus, China, 2009. PloS one 6, e16418.
527 528 529
Zhang, J. S., Zhao, Q. M., Zhang, P. H., Jia, N. & Cao, W. C. (2009). Genomic sequence of a Japanese encephalitis virus isolate from southern China. Arch Virol 154, 1177-1180. Zhao, Z., Date, T., Li, Y., Kato, T., Miyamoto, M., Yasui, K. & Wakita, T. (2005).
530
Characterization of the E-138 (Glu/Lys) mutation in Japanese encephalitis virus by using
531
a stable, full-length, infectious cDNA clone. The Journal of general virology 86, 2209-
532
2220.
533 534 535
536
Table 1. Mouse neuroinvasiveness of the GI, GIII, and GV JEV strains.
Virus Survival† Muar/1952 9/10 Beijing-1 8/10 Mie/41/2002 10/10 103 Muar/1952 1/10 Beijing-1 0/10 Mie/41/2002 8/10 4 10 Muar/1952 0/10 Beijing-1 0/10 Mie/41/2002 7/10 105 Muar/1952 0/10 Beijing-1 0/10 Mie/41/2002 7/10 † No. of mice surviving/no. mice inoculated. Dose 102
537 538 539 540
‡
Average time to death (day) 9 8 7.8 7.2 9 6.6 7.2 10.7 7.7 8.7 11.3
P value‡ (vs. Mie/41/2002)
P value‡ (vs. Muar/1952)
0.32 0.15
