Vaccine 29 (2011) 6352–6357

Contents lists available at ScienceDirect

Vaccine
journal homepage: www.elsevier.com/locate/vaccine

A DNA vaccine encoding the E protein of West Nile Virus is protective and
can be boosted by recombinant domain DIII
Anne Schneeweiss a , Stefan Chabierski a , Mathias Salomo a , Nicolas Delaroque a , Samiya Al-Robaiy a ,
Thomas Grunwald b , Kurt Bürki c , Uwe G. Liebert d , Sebastian Ulbert a,∗
a
Vaccine Technologies Unit, Fraunhofer Institute for Cell Therapy and Immunology, Perlickstrasse 1, D-04103 Leipzig, Germany
b
Department of Molecular and Medical Virology, Ruhr-University, Universitätsstrasse 150, 44801 Bochum, Germany
c
Institute of Animal Science, University of Zürich, Winterthurerstrasse 190, CH-8057 Zürich, Switzerland
d
Institute of Virology, Leipzig University, Johannisallee 30, 04103 Leipzig, Germany

a r t i c l e i n f o a b s t r a c t

Article history: West Nile Virus (WNV) is an emerging pathogenic flavivirus with increasing distribution worldwide. Birds
Received 8 December 2010 are the natural host of the virus, but also mammals, including humans, can be infected. In some cases,
Received in revised form 18 April 2011 a WNV infection can be associated with severe neurological symptoms. All currently available WNV
Accepted 29 April 2011
vaccines are in the veterinary sector, and there is a need to develop safe and effective immunization
Available online 17 May 2011
technologies, which can also be used in humans. An alternative to current vaccination methods is DNA
immunization. Most current DNA vaccine candidates against flaviviruses simultaneously express the
Keywords:
viral envelope (E) and membrane (prM) proteins, which leads to the formation of virus-like particles.
DNA vaccine
West Nile Virus
Here we generated a DNA plasmid, which expresses only the E-protein ectodomain. Vaccination of mice
Heterologous prime boost stimulated anti-WNV T-cell responses and neutralizing antibodies that were higher than those obtained
after immunizing with a recombinant protein previously shown to be a protective WNV vaccine. A single
dose of the plasmid was sufficient to protect animals from a lethal challenge with the virus. Moreover,
immunogenicity could be boosted when DNA injection was followed by immunization with recombinant
domain DIII of the E-protein. This resulted in significantly enhanced neutralizing antibody titers and a
more prominent cellular immune response. The results suggest that the WNV E-protein is sufficient
as a protective antigen in DNA vaccines and that protection can be significantly improved by adding a
recombinant protein boost to the DNA prime.
© 2011 Elsevier Ltd. All rights reserved.

1. Introduction affecting European countries such as Hungary, Italy and

West Nile Virus (WNV) belongs to the genus Flavivirus, which
includes several medically important pathogens such as Japanese
Encephalitis Virus (JEV), Dengue Virus (DENV), Yellow Fever Virus
(YFV) or the Tick-Borne Encephalitis Virus (TBEV). WNV is a
zoonotic virus, which is transmitted by mosquitoes and mainly
infects birds. However, mosquitoes can also transmit WNV to
mammals such as humans or horses. Particularly older or immune-
suppressed individuals are at risk for severe neurological symptoms
caused by WNV, such as encephalitis, meningitis or flaccid paraly-
sis, which might even be fatal [1].
WNV is one of the most widespread flaviviruses, and
over the last years outbreaks have frequently been observed
in Africa, Asia, Australia and, since 1999, America [2,3]. In
Europe, severe WNV cases have increased in recent years,
∗ Corresponding author. Tel.: +49 341 35536 2106; fax: +49 341 35536 82106.
E-mail address: Sebastian.ulbert@izi.fraunhofer.de (S. Ulbert).
Romania [4–6].
In mammals, only about 20% of the infected individuals develop
disease symptoms upon an infection with WNV, indicating effective
immune responses against the virus in most cases.
Studies performed in mouse models have demonstrated that the
mammalian innate and adaptive immune systems are able to con-
trol a WNV infection. Inhibiting the class 1 interferon system, as
well as antibody- or T-cell-responses leads to increased suscepti-
bility to WNV (recently reviewed in [7,8]).
Several veterinary WNV vaccines have been commercialized
to date, including inactivated virus or viral vector systems [9,10].
In contrast, no vaccine is yet available to protect humans. WNV
affects older and immune-compromised people most severely;
therefore live vaccines are not optimal due to potential side effects.
Innovative techniques for inactive subunit vaccines are under
development, amongst them are recombinant proteins or DNA vac-
cines [11–13].
DNA vaccination has become a well-established technique over
the last two decades and is an alternative to existing immunization

0264-410X/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.vaccine.2011.04.116
 A. Schneeweiss et al. / Vaccine 29 (2011) 6352–6357
approaches, especially due to the development of novel delivery
techniques such as in vivo electroporation [14]. A DNA vaccine
against WNV for horses was licensed in the USA. This vaccine, sim-
ilar to most WNV DNA vaccines reported in the literature, is based
on the co-expression of the viral envelope (E) and membrane (prM)
proteins [15–17]. The E protein has functions in several aspects of
the viral life cycle, and prM stabilizes the conformation of E during
virion assembly. Both proteins are targets for neutralizing anti-
bodies (NA) [18,19]. The E protein consists of a large ectodomain
and a small hydrophobic C-terminal transmembrane region. Within
the ectodomain, domains DI, DII and DIII function in virion assem-
bly, membrane fusion and receptor binding, respectively. NAs are
directed against all three domains, but DIII elicits the strongest neu-
tralizing responses [19,20]. The co-expression of flaviviral prM/E
leads to the formation of virus-like particles (VLPs). It is believed
that these structures stabilize the proteins and are critical for opti-
mal presentation of the antigens to the immune system [15,21].
This is supported by results showing that DNA vaccines exclusively
encoding the E protein, or its domain DIII, have until now conferred
only partial protection to mice [21–23]. On the other hand, several
studies have demonstrated that in recombinant form the E protein
(in the absence of prM) or DIII can be used successfully to protect
mammals against WNV [12,13,24]. Therefore, it remains to be clar-
ified for DNA vaccines whether the WNV E protein can be used as
a protective antigen without prM.
In this study, a DNA plasmid coding for the ectodomain of the
WNV E protein was generated. Mice vaccinated with a single dose
of this DNA are protected against a lethal challenge with the virus.
The plasmid elicits neutralizing antibodies, which can be enhanced
by a recombinant protein boost.
2. Materials and methods
2.1. Cells, viruses and materials
HeLa cells were grown in DMEM supplemented with 10% heat-
inactivated fetal calf serum (FCS) and 1% antibiotics. Vero V76 (from
vervet monkey) cells were grown in DMEM supplemented with 5%
FCS and 1% antibiotics. Media und supplements were from Invitro-
gen (Karlsruhe, Germany).
WNV strains NY 2000, which is WNV 2000-crow3356 (Bern-
hard Nocht Institute, Hamburg, Germany), and IS-98-ST1 (Pasteur
Institute, Paris, France) were propagated on VeroV76 cells. Cell
culture supernatant was harvested and virus was purified by ultra-
centrifugation. Virus pellets were diluted in phosphate-buffered
saline (PBS) and titrated on Vero V76 cells in 96 well culture plates.
Titers are expressed as tissue culture infective dose 50 (TCID50 ).
2.2. Construction of a WNV DNA vaccine candidate
The pVAX1 vector (Invitrogen) was used as backbone. To
increase gene expression levels, two SV40 enhancer elements
(72 bp each) [25] were integrated into a unique MluI restriction
site upstream of the CMV promoter. The genomic sequence
for the WNV NY E -protein (accession number AF346318) was
amplified by RT-PCR on genomic RNA from WNV and cloned via
BamHI and ApaI restriction sites into the backbone vector, yielding
pWNV-FL. In addition, the sequence coding for the C-terminal
amino acids G401-A501 was deleted, yielding only the ectodomain
of the protein (pWNV-E). A secretory signal sequence from the
murine tissue plasminogen activator (TPA) was added by PCR
in-frame at the N-terminus of the full length E-Protein and E
ectodomain, using the primers TPA-sense: CGGGATCCACCAT-
GAAGAGAGAGCTGCTGTGTGTACTGCTGCTTTGTGGACTGGCTTTCC-
CATTGCCTGACCAGGGAATACATGGGAGGTTCGCATTCAACTGCCTT
6353
GGAATGAG (TPA sequence is underlined) and 3 fullE:
GCGGGCCCTAGTCAGCGTGCACGTTCACGG or 3 ectoE: GCCT-
GCAGCTAAGACTTGTGCCAATGGTGATTG. The resulting PCR
fragments were inserted into the backbone vector via BamHI
and PstI restriction sites, resulting in pT-WNV-FL (full length E
with TPA) and pT-WNV-E (E ectodomain with TPA). pCBWN was
kindly provided by Michael Diamond (Washington University,
USA).
DNA was purified by endotoxin-free GigaPrep Kit
(Sigma–Aldrich, Hamburg, Germany) according to the man-
ufacturer’s instructions. DNA concentration and purity were
determined by UV spectroscopy. HeLa cells were transiently trans-
fected using Lipofectamine 2000 (Invitrogen, Karlsruhe, Germany)
according to the manufacturer’s instructions. Cell culture super-
natants were collected and cells were lysed with 0.1% SDS Tris
24 h post transfection. The cell lysates and supernatants were
analyzed by SDS PAGE and western blotting using a polyclonal
rabbit antiserum raised against the peptide GEAHNDKRADPAFVC
from the WNV E protein.
2.3. Expression and purification of recombinant WNV proteins
The sequence encoding the domain DIII of the WNV E pro-
tein was PCR amplified from pT-WNV-E and cloned via BamHI and
XhoI restriction sites into the bacterial expression vector pMAL-
c2X (New England Biolabs, Ipswich, USA), which was modified to
include a C-terminal 6× His tag. The resulting coding sequences
contained an N-terminal maltose binding protein (MBP) and a C-
terminal His tag. The fusion-protein was expressed in the E. coli
Rosetta DE3 strain (Merck, Darmstadt, Germany) by induction
with 1 mM isopropyl-␤-d-thiogalactopyranoside. Bacteria were
harvested and lysed in buffer I (20 mM Tris pH 7.4, 200 mM NaCl,
1 mM EDTA) by repeated sonification and freezing in liquid nitro-
gen. Insoluble cell debris containing the WNV fusion-protein was
collected by centrifugation, resuspended in lysis buffer II (20 mM
Tris pH 7.4, 200 mM NaCl, 1 mM EDTA, 8 M Urea) and sonicated.
After centrifugation the recombinant WNV fusion-protein was
purified from the supernatant on HisTrap HP columns (GE Health-
care Bio-Sciences, Munich, Germany) by liquid chromatography.
Thereafter, the MBP-fusion protein was purified on an amylose
resin column (New England Biolabs). Subsequently the MBP-tag
was removed by cleaving the fusion-protein with Factor Xa (New
England Biolabs) and the resulting His-tagged protein was isolated
by a second His-tag purification as above and dialyzed against PBS.
2.4. Mouse immunizations and WNV challenge studies
Vaccination studies were carried out in groups of five female
Balb/c mice (obtained from Sociéte Janvier, Le Genest-St-Isle,
France), 6–8 weeks old. Experiments were performed after
approval of local authorities and according to the guidelines
of the Ethics Committees for Animal Studies of Germany and
Switzerland. For DNA vaccination mice were narcotised with
Ketamin/Xylazin (5 mg/kg Xylazin and100 mg/kg Ketamin; Bayer,
Leverkusen, Germany). 25 ␮l of DNA (at 1 mg/ml) in 0.9% sodium
chloride (w/v) were injected intramuscularly (i.m.) into the quadri-
ceps muscle (29-gauge needle). Injections were immediately
followed by electroporation with the electric pulse generator Clin-
ivet (IGEA, Modena, Italy). A four-electrode array was placed at
the injection site and two electric pulses with 450 V (50 ␮s) fol-
lowed by eight low volt pulses (110 V, 20 ms, 20 ms interval) were
applied. Alternatively, one group of mice received pT-WNV-E sub-
cutaneously (s.c.) without electroporation.
For protein immunizations 20 ␮g of recombinant protein were
injected intraperitoneally (i.p.) together with 20 ␮g ODN 1826
(Eurofins, Ebersberg, Germany), a CpG containing oligonucleotide,
6354 A. Schneeweiss et al. / Vaccine 29 (2011) 6352–6357

as adjuvant. In a BSL3 animal facility, mice were challenged intra-

venously 6 weeks after the first immunization with a lethal dose
(2 × 107 TCID50 ) of the WNV strain IS-98-ST1 (E-protein sequence
100% identical in amino acids to T-WNV-E) and monitored for sur-
vival, body weight and clinical symptoms for 15 days.

2.5. Murine IFN- ELISpot

The ELISpot test for detection of murine IFN-␥ was carried out
according to the manufacturer’s instructions (Mabtech, Hamburg,
Germany). Two weeks after the last immunizations, splenocytes
were isolated from the mice and stimulated with 1 ␮g of recom-
binant ectodomain of WNV E (insect cell derived and kindly
provided by Michael Diamond). IFN-␥ activity was measured using
an ELISpot Reader (AID Diagnostika, Straßberg, Germany).

2.6. Virus neutralisation test and antibody analysis

Vero V76 cells were seeded into 96-well plates at a density of

7000 cells in 100 ␮l per well 1 day before start of the test. For
the virus neutralisation test, heat inactivated (30 min at 56 ◦ C) sera
from immunized mice were serially diluted in culture medium. Two
hundred TCID50 of WNV NY were added to each serum dilution, the
mixtures were incubated for 1 h at 37 ◦ C and added to the cells (in
triplicates). After 60 min culture fluids were replaced by fresh Vero
V76 medium (5% FCS) and cells were propagated for additional 5
days. Titers of neutralising antibodies were determined by monitor-
ing of the cytopathic effect (CPE). Neutralising activity was reported
until two out of three wells of infected cells showed no CPE [26].
To investigate IgG production upon vaccination, sera from
immunized mice were pooled, diluted 1:100 and analyzed in an
enzyme-linked immunosorbent assay (ELISA) using rDIII as antigen
(4 ng per well of a microtiter plate), following standard protocols.
Bound IgG were detected with a rabbit anti-mouse antibody conju-
gated with horseradish peroxidase diluted 1:1000 (Dako, Hamburg,
Germany). Mean values of the optical density at 450 nm (OD450) of
triplicate measurements were determined. Correct conformation of
rDIII was analyzed in an ELISA with the monoclonal antibody E16
(kindly provided by Michael Diamond) in a 1:1000 dilution. Fig. 1. (A) HeLa cells were transfected with plasmids coding for the proteins men-
tioned below and analyzed for expression of the WNV E-protein 24 h later. The
3. Results top panel shows cell lysates, and the bottom panel displays cell culture super-
natants. Lanes: 1, full length E-protein (pWNV-FL); 2, full length E-protein carrying a
N-terminal TPA signal sequence (pT-WNV-FL, specific band marked with an arrow-
3.1. Development of a DNA vaccine head); 3, Ecto-domain of the E-protein (pWNV-E, specific band marked with an
asterisk); 4, Ecto-domain of the E-protein carrying a N-terminal TPA signal sequence
Expression of the full length WNV E protein from plasmid (pT-WNV-E); 5, non-transfected cells. (B) SDS-gel electrophoresis with recombinant
domain DIII. The protein was purified from bacteria using nickel and amylose based
pWNV-FL was tested via in vitro transfection studies in HeLa cells.
affinity chromatography (lane 1); the fusion protein was released from the MBP tag
However, as can be seen in Fig. 1A (lane 1), no clear signal for the via proteolytic cleavage by factor X (lane 2); a second purification of the His contain-
E protein was detectable in a western blot, indicating poor pro- ing proteins separated the MBP (lane 3) from the WNV antigen (lane 4). The correct
duction of the protein. Therefore, the plasmid was modified by size of rDIII is indicated on the right. (C) ELISA using the monoclonal antibody E16 on
deleting the hydrophobic C-terminal domain of E, which might be recombinant DIII protein and recombinant MBP as a negative control. The proteins
were used in the amounts indicated, rDIII after the removal of the MBP tag.
responsible for the weak expression level. The resulting protein is
a truncated version of E which resembles the ectodomain (pWNV-
E). Now, the protein was detectable in the western blot, although
at very low level (lane 3). In addition, a secretory signal sequence,
derived from the murine tissue plasminogen activator (TPA), was However, as the TPA tagged version of the full-length E protein
added to the N-terminus. The resulting plasmid, termed pT-WNV- was absent in the supernatant, these results indicate that the TPA
E, showed a very pronounced increase in expression of the protein, signal sequence mediated the secretion of the E-ectodomain.
as compared to the wild type version (Fig. 1A, lane 4). Also the For control immunization experiments recombinant DIII (rDIII)
full length version of the E-protein was expressed more efficiently was used, as this domain is efficiently expressed in bacteria
when the TPA sequence was added to the N-terminus (pT-WNV-FL; (Fig. 1B), and it was shown previously to be a protective vaccine
lane 2). antigen [12,24]. To test for the correct conformation of the recom-
The TPA-modified version of the E-ectodomain was found in the binant domain DIII after expression, the protein was analyzed with
cell culture supernatant (Fig. 1A, lower panel), visible as a triple the monoclonal antibody E16, which recognizes a structural epi-
band. In contrast, the proteins without the TPA sequences were not tope along the lateral ridge of DIII [27]. There was a clear signal
detectable in the supernatant. It might be possible that the signals detectable in the ELISA (Fig. 1C), indicating the correct conforma-
were under the detection limit due to the low expression levels. tion of the protein.
 A. Schneeweiss et al. / Vaccine 29 (2011) 6352–6357 6355

Table 1 with a WNV antigen (E-protein) or a control protein (MBP). Spleen

Vaccination schemes used in this study (groups of 5 mice).
cells derived from mice immunized once with pT-WNV-E showed
Group Nr. Vaccine Day 0 Day 14 Day 28 elevated IFN-␥ activity upon stimulation with the WNV antigen
1 Empty plasmid DNA (Fig. 2). I.m. injection with electroporation seems to be essential
2 pT-WNV-E DNA for this T-cell response, as s.c. injection did not lead to detectable
3 pT-WNV-E 2× DNA DNA IFN-␥. Booster injections with rDIII increased the cellular response
4 pT-WNV-E + 1× rDIII DNA Protein to pT-WNV-E, regardless whether the adjuvant was used or not.
5 pT-WNV-E + 2× rDIII DNA Protein Protein
Immunizing twice with pT-WNV-E led to the highest IFN-␥ produc-
6 pT-WNV-E + 2× rDIII no adj. DNA Protein Protein
7 pT-WNV-E s.c. DNA tion. Also the cellular response to pCBWN increased by subsequent
8 pWNV-E DNA immunizations with rDIII (Fig. 2). These results indicate that boost-
9 pCBWN DNA ing with recombinant protein can enhance the cellular immune
10 pCBWN + 2× rDIII DNA Protein Protein
response to these DNA immunizations. However, boosting with a
11 3× rDIII Protein Protein Protein
second application of pT-WNV-E is most powerful.

3.2. Immunizations 3.4. Antibody production

Table 1 shows the different immunization groups. Balb/c mice
were immunized with the plasmids pT-WNV-E and pWNV-E
(groups 2–7 and 8, respectively). In addition, for comparison with
a construct encoding the prM/E expression cassette plus a viral sig-
nal sequence, the plasmid pCBWN [15] was used (groups 9 and 10).
Combinations of DNA and rDIII were applied as heterologous DNA-
prime and protein-boost strategies (groups 4–6, 9, 10). In group 6,
the adjuvant was omitted in the protein boost injections. Three rDIII
immunizations (without DNA prime) served as a positive control
(group 11).
DNA was injected i.m. followed by in vivo electroporation. To
compare the results with an alternative delivery route, group 7
received pT-WNV-E s.c. without electroporation (Table 1). Con-
trol animals (group 1) received one dose of the empty backbone
plasmid. Injections were performed with 2-weeks intervals, and
immune responses were analyzed six weeks after the first applica-
tion.
3.3. IFN- production
Induction of cellular immunity was determined with an ELISpot
assay, which measured the IFN-␥ production upon stimulation
The humoral immune response following the immunizations
was evaluated by analysing the mouse sera for the presence of
WNV-specific IgG in an ELISA. Specific antibodies were detected
with all three immunization strategies (Table 2). Antibody lev-
els of animals immunized once with pT-WNV-E were comparably
low and increased after a second application. Boosting with rDIII
increased the IgG values, both for priming with pT-WNV-E and
pCBWN (Table 2, groups 2, 3 and 10). The plasmid pWNV-E (group
8) and the empty DNA backbone vector led to no detectable anti-
WNV IgG (group 1), Similar to the T-cell response, electroporation
of pT-WNV-E seems to be crucial for antibody production, as no sig-
nals were detected in animals that received the DNA s.c. without
electroporation (group 7).
To investigate if the WNV-specific immune response elicited by
the different immunization protocols exhibited WNV-neutralising
activity, the sera were studied in a virus neutralisation assay. Titers
of NA are shown in Table 2. Overall, the results were similar to
the total IgG responses. The highest neutralizing activities were
detected after immunization with the heterologous DNA-prime
protein-boost, when either pT-WNV-E or pCBWN was used as DNA.
The addition of the ODN1826 adjuvant did not seem to influence
the outcome of protein boost, as both total IgG and neutralizing

Fig. 2. WNV-specific IFN-␥ levels after immunization. Splenocytes from mice immunized with the antigens indicated were stimulated with cell culture medium (white
columns), recombinant WNV E ectodomain (grey columns) or recombinant MBP (black columns). An ELISpot assay was performed to determine IFN-␥ activity upon
stimulation. Error bars represent the standard deviation.
6356 A. Schneeweiss et al. / Vaccine 29 (2011) 6352–6357
Table 2
Anti-WNV IgG production and neutralizing titers upon immunization with the indicated plasmids and recombinant proteins (mean values of groups of 5 mice). Samples
were taken six weeks after the first immunizations.
Group Nr. Vaccine
1 Empty plasmid
2 pT-WNV-E
3 pT-WNV-E 2×
4 pT-WNV-E + 1× rDIII
5 pT-WNV-E + 2× rDIII
6 pT-WNV-E + 2× rDIII no adj.
7 pT-WNV-E s.c.
8 pWNV-E
9 pCBWN
10 pCBWN + 2× rDIII
11 3× r DIII
activity in group 6 were comparable to the groups receiving the
adjuvant (Table 2, group 5). Mice which obtained the empty DNA
vector showed no neutralising activity. Vaccination of rDIII alone
did not lead to neutralizing antibodies above the detection limit of
1:16, although high IgG levels were measured in this group.
3.5. In vivo protection studies
To test the effectiveness of pT-WNV-E as a protective vaccine,
viral challenge experiments were performed. Mice were immu-
nized either with one dose of pT-WNV-E (Fig. 3, group 1), with one
dose of pT-WNV-E plus two doses of rDIII (group 2), or with three
doses of rDIII (group 3). Negative control mice received one dose
of the empty backbone plasmid (group 4). Six weeks after the first
immunization, animals were injected intravenously with a lethal
dose of WNV and their survival was monitored. Mice that received
the negative control DNA all developed severe symptoms between
5 and 7 days post infection (Fig. 3) and were euthanized according
to the termination criteria set. In contrast, all animals immunized
with pT-WNV-E, with rDIII or with the heterologous prime-boost
survived without displaying any measurable signs of disease. These
data demonstrate that a single vaccination with pT-WNV-E is suf-
ficient to protect mice from a lethal WNV infection.
Fig. 3. In vivo protection study. Mice were vaccinated and subsequently challenged
with a lethal dose of WNV. Survival rates were determined for 15 days post infection
(p.i). Immunizations were performed with one dose of pT-WNV-E (group 1), one
dose of pT-WNV-E plus two doses of rDIII (group 2), three doses of rDIII (group 3)
or the empty DNA backbone plasmid (group 4). Control mice remained uninfected
(group 5).
Anti WNV IgG (OD) (±STDV) WNV neutralizing titer
0.006 (±0.001) <16
0.306 (±0.01) 16
0.768 (±0.03) 128
0.874 (±0.12) 32
1.799 (±0.03) 256
1.804 (±0.06) 512
0.003 (±0.005) <16
0.007 (±0.005) <16
0.167 (±0.02) 32
1.729 (±0.02) 256
1.646 (±0.03) <16
4. Discussion
In this study the immunization with a DNA vaccine plasmid
(pT-WNV-E) coding for the ectodomain of the WNV E protein is
reported. Fusion of the TPA leader sequence to the ectodomain
leads to a significant increase in expression level compared to
the wildtype version and to the generation of protective immune
responses. In contrast, the plasmid without the TPA sequence
did not induce measurable anti-WNV IgG or neutralizing activity,
although a minor cellular immune response was observed.
Splenocytes from mice vaccinated with pT-WNV-E produced
IFN-␥ upon stimulation with WNV antigen, and the level was
strongly increased after a second injection of the plasmid. IFN-␥
values were also markedly increased after two booster injections
with rDIII, but were still lower as compared to the DNA boost. This
could be explained by the fact that the ectodomain of E contains
more T cell epitopes than its domain DIII [28]. In addition, DNA
vaccines are known potent inducers of cellular immune responses,
especially after delivery via electroporation [29].
Booster injections with rDIII increased antibody titers in serum
to a higher extend than a second application of pT-WNV-E.
Together, this indicates that, after a DNA prime, the DNA boost
especially augments cellular immunity, whereas the rDIII boost
strengthens antibody responses. In contrast, NA titers in animals
immunized with rDIII were borderline, although the total IgG level
against the antigen was high. This result differs from two previ-
ous studies, which showed that bacterially expressed rDIII leads
to efficient NA production [12,24]. The rDIII protein used here was
of the correct conformation, as it was bound by the monoclonal
antibody E16, which recognizes a structural epitope. Therefore,
technical differences in study design (e.g. injection route, mouse
strain, neutralization assays) might contribute to this discrepancy.
In addition, low-level NA titers after rDIII immunization have been
reported previously [23]. Nevertheless, consistent with the studies
described in [12,24], immunization with rDIII led to protection of
mice from WNV in the work described here.
When challenged with WNV, mice immunized with a single
dose of pT-WNV-E (with or without rDIII boost) did not show
signs of disease. Protection from a lethal WNV-challenge was also
achieved with DNA vaccines containing the prM/E expression cas-
sette, such as the plasmid pCBWN [15], which was included in
the present study. Expression of pCBWN leads to the formation
of VLPs, which forms the basis of most current DNA immunization
approaches against flaviviruses [15–17,30]. The pT-WNV-E plas-
mid does not contain the prM or viral signal sequences present
in the prM/E constructs. Moreover, it does not lead to formation
of VLPs (data not shown). However, humoral or cellular responses
were not significantly different when comparing immunizations
with pT-WNV-E and pCBWN. This indicates that co-expression of
prM and the formation of VLPs is not a prerequisite for induction
of an efficient anti-WNV immune response upon DNA vaccina-
 A. Schneeweiss et al. / Vaccine 29 (2011) 6352–6357
tion. Therefore, a plasmid, which encodes only the ectodomain of
WNV-E, fused to the TPA sequence seems to be a valid alternative
to constructs expressing the proteins prM and E together with a
viral signal sequence. This is supported by recent studies on DIII
based DNA vaccines which were optimized by the addition of DNA
encoded protein adjuvants [23,31]. In addition, it was also shown
that the ectodomain of E from JEV is more efficiently expressed
after fusion to a TPA signal sequence in the context of vaccination
[22].
The results presented here indicate synergistic effects when a
single immunization with pT-WNV-E is followed by two booster
immunizations with rDIII. Mice were equally well protected in the
challenge experiment, but the protein boost strongly enhanced the
immunogenicity of a single pT-WNV-E application.
The combination of DNA prime and protein boost is reportedly
associated with an improvement of cellular and humoral immune
responses, although the exact reasons for this synergy still have
to be determined (reviewed in [32]). An inactivated WNV vaccine
was used to optimize a DNA vaccine based on the pr/M expression
cassette [17]. However, the use of rDIII as booster component has
the advantages that the protein is relatively easy to express and
that the vaccine manufacturing does not involve culturing of WNV.
For vaccine strategies encoding single antigens, such as the plas-
mids described here or recombinant proteins [13], care must be
taken to monitor emerging viral strains with potential sequence
variations in order to guarantee protection. On the other hand,
especially DNA vaccines can rapidly be adapted to novel sequences
by adapting the nucleotide sequences and by co-injection of several
sequence variants.
In summary, a DNA plasmid encoding a part of the E protein of
WNV was generated, which protects animals from WNV. Neutral-
izing antibody and cellular immune responses to a single injection
of this plasmid were strongly increased by subsequent injections
of recombinant DIII protein. This strategy was clearly superior to
vaccination with DNA or protein alone. As a vaccine to be used in
humans against WNV has to be both, effective and long-lasting,
further studies are warranted using the approach presented.
Acknowledgements
We thank Diana Riethmüller for excellent technical assistance,
Drs. Michael Szardenings and Gustavo Makert dos Santos for crit-
ically reading of the manuscript and Dr. Matthias Giese (Institute
for Molecular Vaccines, Heidelberg) for acquisition of funding from
BLE. We also thank Dr. Philippe Despres (Pasteur Institute, Paris)
for supplying the WNV strain IS-98-ST1 and Dr. Michael Diamond
(Washington University, St. Louis) for providing the E16 antibody,
pCBWN and recombinant ectoE. This work was supported by the
German Federal Agency for Agriculture and Food (BLE) and by the
European Commission under FP7, Project 261426 (WINGS-West
Nile Integrated Shield Project).
References
[1] Hayes EB, Sejvar JJ, Zaki SR, Lanciotti RS, Bode AV, Campbell GL. Virology, pathol-
ogy, and clinical manifestations of West Nile virus disease. Emerg Infect Dis
2005;11:1174–9.
[2] Calistri P, Giovannini A, Hubalek Z, Ionescu A, Monaco F, Savini G, et al. Epi-
demiology of west Nile in Europe and in the Mediterranean basin. Open Virol
J 2010;4:29–37.
[3] Murray KO, Mertens E, Desprès P. West Nile virus and its emergence in the
United States of America. Vet Res 2010;41:67.
[4] Krisztalovics K, Ferenczi E, Molnar Z, Csohan A, Ban E, Zoldi V, et al. West Nile
virus infections in Hungary. August–September 2008. Euro Surveill 2008;13,
pii:19030.
[5] Barzon L, Franchin E, Squarzon L, Lavezzo E, Toppo S, Martello T, et al. Genome
sequence analysis of the first human West Nile virus isolated in Italy in 2009.
Euro Surveill 2009;14, pii:19384.
6357
[6] Sirbu A, Ceianu CS, Panculescu-Gatej RI, Vazquez A, Tenorio A, Rebreanu R, et al.
Outbreak of West Nile virus infection in humans, Romania, July to October 2010.
Euro Surveill 2011;16(January (2)).
[7] Diamond MS, Mehlhop E, Oliphant T, Samuel MA. The host immunologic
response to West Nile encephalitis virus. Front Biosci 2009;14:3024–34.
[8] Ulbert S, West Nile Virus—the complex biology of an emerging pathogen. Inter-
virology, in press.
[9] Ng T, Hathaway D, Jennings N, Champ D, Chiang YW, Chu HJ. Equine vaccine
for West Nile virus. Dev Biol (Basel) 2003;114:221–7.
[10] El Garch H, Minke JM, Rehder J, Richard S, Edlund Toulemonde C, Dinic S, et al.
A West Nile virus (WNV) recombinant canarypox virus vaccine elicits WNV-
specific neutralizing antibodies and cell-mediated immune responses in the
horse. Vet Immunol Immunopathol 2008;123:230–9.
[11] Martin JE, Pierson TC, Hubka S, Rucker S, Gordon IJ, Enama ME, et al. A West
Nile virus DNA vaccine induces neutralizing antibody in healthy adults during
a phase 1 clinical trial. J Infect Dis 2007;196:1732–40.
[12] Martina BE, Koraka P, van den Doel P, van Amerongen G, Rimmelzwaan GF,
Osterhaus AD. Immunization with West Nile virus envelope domain III protects
mice against lethal infection with homologous and heterologous virus. Vaccine
2008;26:153–7.
[13] Lieberman MM, Nerurkar VR, Luo H, Cropp B, Carrion Jr R, de la Garza M, et al.
Immunogenicity and protective efficacy of a recombinant subunit West Nile
virus vaccine in rhesus monkeys. Clin Vaccine Immunol 2009;16:1332–7.
[14] Kutzler MA, Weiner DB. DNA vaccines: ready for prime time? Nat Rev Genet
2008;9:776–88.
[15] Davis BS, Chang GJ, Cropp B, Roehrig JT, Martin DA, Mitchell CJ, et al. West Nile
virus recombinant DNA vaccine protects mouse and horse from virus challenge
and expresses in vitro a noninfectious recombinant antigen that can be used in
enzyme-linked immunosorbent assays. J Virol 2001;75:4040–7.
[16] Anwar A, Chandrasekaran A, Ng ML, Marques E, August JT. West Nile
premembrane-envelope genetic vaccine encoded as a chimera containing the
transmembrane and cytoplasmic domains of a lysosome-associated membrane
protein: increased cellular concentration of the transgene product, targeting
to the MHC II compartment, and enhanced neutralizing antibody response.
Virology 2005;332:66–77.
[17] Ishikawa T, Takasaki T, Kurane I, Nukuzuma S, Kondo T, Konishi E. Co-
immunization with West Nile DNA and inactivated vaccines provides
synergistic increases in their immunogenicities in mice. Microbes Infect
2007;9:1089–95.
[18] Vázquez S, Guzmán MG, Guillen G, Chinea G, Pérez AB, Pupo M, et al.
Immune response to synthetic peptides of dengue prM protein. Vaccine
2002;20:1823–30.
[19] Beasley DW, Barrett AD. Identification of neutralizing epitopes within
structural domain III of the West Nile virus envelope protein. J Virol
2002;76:13097–100.
[20] Oliphant T, Nybakken GE, Austin SK, Xu Q, Bramson J, Loeb M, et al. Induc-
tion of epitope-specific neutralizing antibodies against West Nile virus. J Virol
2007;81:11828–39.
[21] Aberle JH, Aberle SW, Allison SL, Stiasny K, Ecker M, Mandl CW, et al. A
DNA immunization model study with constructs expressing the tick-borne
encephalitis virus envelope protein E in different physical forms. J Immunol
1999;163:6756–61.
[22] Ashok MS, Rangarajan PN. Protective efficacy of a plasmid DNA encoding
Japanese encephalitis virus envelope protein fused to tissue plasminogen acti-
vator signal sequences: studies in a murine intracerebral virus challenge model.
Vaccine 2002;20:1563–70.
[23] Dunn MD, Rossi SL, Carter DM, Vogt MR, Mehlhop E, Diamond MS, et al.
Enhancement of anti-DIII antibodies by the C3d derivative P28 results in lower
viral titers and augments protection in mice. Virol J 2010;7:95.
[24] Chu JH, Chiang CC, Ng ML. Immunization of flavivirus West Nile recombinant
envelope domain III protein induced specific immune response and protection
against West Nile virus infection. J Immunol 2007;178:2699–705.
[25] Li HS, Liu Y, Li DF, Zhang RR, Tang HL, Zhang YW, et al. Enhancement of DNA
vaccine-induced immune responses by a 72-bp element from SV40 enhancer.
Chin Med J (Engl) 2007;120:496–502.
[26] Reed LJ, Muench H. A simple method of estimating fifty per cent endpoints. Am
J Hygiene 1938;27:493–7.
[27] Nybakken GE, Oliphant T, Johnson S, Burke S, Diamond MS, Fremont DH. Struc-
tural basis of West Nile virus neutralization by a therapeutic antibody. Nature
2005;437:764–9.
[28] Parsons R, Lelic A, Hayes L, Carter A, Marshall L, Evelegh C, et al. The memory
T cell response to West Nile virus in symptomatic humans following natural
infection is not influenced by age and is dominated by a restricted set of CD8+
T cell epitopes. J Immunol 2008;181:1563–72.
[29] Otten GR, Schaefer M, Doe B, Liu H, Megede JZ, Donnelly J, et al. Potent
immunogenicity of an HIV-1 gag-pol fusion DNA vaccine delivered by in vivo
electroporation. Vaccine 2006;24:4503–9.
[30] Chang GJ, Hunt AR, Davis B. A single intramuscular injection of recombinant
plasmid DNA induces protective immunity and prevents Japanese encephalitis
in mice. J Virol 2000;74:4244–52.
[31] Ramanathan MP, Kutzler MA, Kuo YC, Yan J, Liu H, Shah V, et al. Coimmuniza-
tion with an optimized IL15 plasmid adjuvant enhances humoral immunity via
stimulating B cells induced by genetically engineered DNA vaccines expressing
consensus JEV and WNV E DIII. Vaccine 2009;27:4370–80.
[32] Lu S. Heterologous prime-boost vaccination. Curr Opin Immunol
2009;21:346–51.