'/~°ur~ta/. ~/~.~r~ral..V!r°l°blY..! ! ? 96).'..7 ~'..21.59-2 !..63'....P.[!~!e.d.!p..G[eat B~i.!a!n................................................ S H O R T C O M M U N I C A T I O N
Expression of muscovy duck parvovirus capsid proteins (VP2
and VP3) in a baculovirus expression system and
demonstration of immunity induced by the recombinant
proteins
Ghislaine Le GalI-Recul4,1 V4ronique Jestin, 1 Patrice Chagnaud, 2 Philippe Blanchard z
and Andr4 Jestin z
1.2 Unite de Pathologie Aviaire1 and Laboratoire de Biologie Moleculaire2, CNEVA, Laboratoire Central de Recherches Avicole et Porcine,
BP 53, 22440 Ploufragan, France
The gene encoding the muscovy duck parvovirus
(DPV)-strain 8 9 3 8 4 capsid proteins VP2 and VP3
was cloned in a baculovirus expression system and
expressed in insect cells. The recombinant proteins
were found to react with specific anti-DPV serum by
Western blotting and to be located in the nucleus of
insect cells (Sf9) as shown by immunofluorescence.
Empty virus-like particles (VLPs) identical in size
and appearance to DPV virions were observed by
electron microscopy. The antigenicity and immunogenicity of the recombinant proteins were
evaluated by ELISA and seroneutralization.
Immunization of 3-week-old muscovy ducklings
induced anti-DPV antibodies; neutralizing antibody
titres were consistent with those observed in
ducklings inoculated with a commercial inactivated
vaccine. The way to develop these promising results
is discussed.
The muscovy duck parvovirus (DPV) is the causative agent
(Jestin et al., 1991a; Foumier & Gaudry, 1992) of a disease
which appeared in France in 1989 (Jestin, 1990; Drouin et al.,
1991) that is characterized by high mortality in very young
muscovy ducklings, stunting, markedly reduced body weights,
and delay in the process of feathering in older ones (Jestin,
1990). In addition, nervous signs and associated microscopic
lesions have been reported (Dalibart et al., 1993). In the field,
only attenuated live and inactivated vaccines are available
(Foumier & Gaudry, 1992; Jestin et al., 1993). They are
Authorfor correspondence:GhislaineLe GalI-Recul4.
Fax +33 96 76 O1 23. e-mail [email protected]
The nucleotidesequencedata reported in this paper have been
deposited with the EMBL, DDBJand GenBankdatabasesunder
accession number Z68272.
0001-4066 © 1996 SGM
intended (i) to give high antibody levels in breeders in order to
obtain good maternal immunity (ii) to actively immunize
ducklings that become vulnerable after the drop in maternal
antibodies. For that purpose, the vaccination schedule is
expensive and time consuming since several injections are
required. Moreover, the quality of protection of 2- to 3-weekold ducklings is questionable (Jestin et al., I993). This can
account for chronic cases of the disease that are still observed
later on and are responsible for significant economic losses.
Therefore, there is the need for further improvement of the
vaccination strategy.
DPV has a linear single-stranded DNA genome of about
5300 bases with terminal palindromic hairpins (Le Gall-Recul~
& Jestin, 1994). The right open reading flame (ORF) of the
DPV genome encodes three capsid proteins VP1, VP2 and
VP3 (G. Le Gall-Recul6 & V. Jestin, unpublished results). The
corresponding nucleic sequence was established for the 89384DPV strain (accession number Z68272 in nucleotide sequence
databases), and comparison of its predicted amino acid
sequence with those of mammalian parvoviruses showed a
higher homology level with the defective parvovirus, the
adeno-associated virus type 2 (AAV2) (G. Le Gall-Recul6 & V.
Jestin, unpublished results). By analogy with the AAV2 capsid
proteins gene organization (Becerra et al., 1988; Cassinotti et
al., 1988; Trempe & Carter, 1988), and by reference to the
observed molecular masses of DPV VP2 and VP3 proteins (Le
Gall-Recul6 & JesUn, I994), we determined the most favourable
initiation codon positions of the two genes.
The baculovirus expression system has been used to
efficiently produce large amounts of viral antigens (Luckow &
Summers, 1988). When expressed using such a system,
parvovirus capsid proteins self-assemble and form empty
virus-like particles (VLPs). In most cases, the recombinant
particles were shown to be antigenically and immunologically
identical to native parvovirus virions (Kajigaya ef al., 1991;
Lopez de Turiso et al., 1992; Martinez et al., 1992; Saliki et al.,
1992; Christensen et al., 1994). The purpose of this paper is (i)
to describe the cloning and the expression in baculovirus of the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:01:10
_~15 c.
(a)
(b)
1
2
3
1
kDa
2
kDa
119-0
97.4 - 66.2
-
-
- - 73-0
84.0
59.0
64.0
-
-
-
-
VP1 (87.0kDa)
m VP2 (74-0kDa)
- - VP3 (58.0kDa)
45-0
-
-
31.0 - 48.0
Fi 9. 1. Western-blot analysis of muscovy duck parvovirus (DPV) recombinant proteins using a specific DPV duck antiserum. (a)
total lysates of insect Sf9 cells infected with wild-type baculovirus (lane 2) and with recombinant baculovirus AcDPV1 14 (lane
3) ; low molecular mass markers (lane 1 ). (b) CsCI-gradient-purified DPV virions (lane 2) ; low molecular mass markers (lane
1).
DPV gene encoding VP2 and VPd, and (ii) to demonstrate the
antigenicity and the immunogenicity of the expressed proteins.
The 89384-DPV strain was isolated from liver, heart and
spleen of a muscovy duck showing symptoms of the disease
(Jestin et al., 1991a). DPV virions were purified by banding in
CsCI and the DPV DNA was extracted as previously described
(Le Gall-Recul6 & Jestin, 1994, 1995). We cloned inserts
containing the entire right ORF of the DPV genome and the
recombinant plasmid pDPV2 was selected for the recombinant
transfer vector construction. We chose to express both DPV
VP2 and VP3 by construction of a recombinant baculovirus
containing the genomic segment coding for the two genes.
Indeed, Ruffing et a]. (1992) showed that empty capsid-like
structures were observed only when AAV2 VP2 was
coexpressed with any other capsid protein. Moreover, some
authors succeeded in expressing both Aleutian mink disease
parvovirus VP1 and VP2 structural proteins, using a recombinant baculovirus generated from a single cDNA containing
the start ATGs for the two polypeptides (Christensen et al.,
I993). The gene segment coding for VP2 and VP3 was
amplified from purified pDPV2 by using the polymerase chain
reaction (PCR) technique. The two primers DPV-U
5' CCCCATGGATGGCTCCTGCTAA 3' for the upper one
and DPV-L3 5' CCCCATGGAAAATCAAAAGAA 3' for the
lower one, correspond in their DPV-specific sequence to the
nucleotides 670-683 and 2455-2443 respectively, according
to the DPV sequence in the nucleotide sequence databases.
This procedure permitted us to mutate the VP2 ACG codon to
an ATG one (bold T in the upper primer) corresponding to a
more favourable context for the initiation of translation, and to
introduce a NcoI restriction endonuclease site allowing cloning
:16(
into the NcoI site of the pGmAc3MR baculovirus transfer
vector. This vector contains essentiaI parts of the poIyhedrin
gene of Autographa californica nuclear polyhedrosis virus
(AcNPV) which were cloned into the pUC8 vector. KpnI and
NcoI sites constitute the multiple cloning site region. Since the
DPV capsid protein gene contains a KpnI site (Le Gall-Recul6
& Jestin, 1994), we had to clone into the NcoI one. The
amplification was performed as previously detailed (lestin et al.,
1995) except that the denaturation step at 95 °C for 5 min was
followed by 5 cycles of 20 s at 95 °C, 30 s at 50 °C, 30 s at
75 °C, then 30 further cycles of 20 s at 95 °C, 30 s at 55"5 °C
and 30 s at 75 °C. The amplification was terminated with an
elongation step at 75 °C for 6 min. The purified PCR product
was digested by Ncol and successfully cloned into the transfer
vector. Recombinant plasmids containing a single insert in the
correct orientation were selected by restriction enzyme
analysis. One recombinant plasmid (p114) was amplified and
the insert sequence was checked by DNA sequencing. Five
micrograms of p114 plasmid DNA was cotransfected with
500 ng of purified wild-type AcNPV DNA into Spodoptera
fr~giperda (Sf9) insect cells by using DOTAP transfection
reagent (Boehringer), according to the supplier's instructions.
Recombinant baculoviruses were purified and one (AcDPV114)
was used to express DPV recombinant proteins.
Sf9 cells were infected at an m.o.i, of 10 p.f.u./cell either
with wild-type or AcDPV114 recombinant baculoviruses.
Cells were harvested 48 h post-infection and total proteins
from 2 x 106 pelleted cells were separated by 10% SDS-PAGE
before being transferred onto a nitrocellulose membrane. The
proteins were identified with a specific DPV cluck antiserum
and a peroxidase-labelled rabbit anti-duck IgG (H + L) con-
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:01:10
(a)
(b)
Fig. 2. Electron micrograph of recombinant VLPs (a) and CsCI gradientpurified DPV virions (b). Scale bar represents 50 nm.
jugate (Nordic). The antiserum was prepared by inoculating
DPV virions into specific pathogen free (SPF) muscovy ducks.
Western blot analysis showed two bands of approximately
equal intensity that corresponded to two expressed proteins of
approximately 73 and 59 kDa (Fig. ia). Their relative molecular masses were similar to those of the DPV VP2 and VP3
proteins (74 and 58 kDa, respectively) established by Western
blotting in the same conditions (Fig. lb). No bands of these
sizes were detected in wild-type baculovirus-infected Sf9 cells
(Fig. l a). To localize the subcellular distribution of the
expressed proteins in Sf9 cells, we performed an indirect
immunofluorescence analysis. Infected Sf9 cells were prepared
as previously described and 2 x 104 cells were dispensed per
well. After being fixed, the slides were incubated with a specific
DPV duck antiserum or a SPF duck serum and stained with
an FITC-conjugated rabbit anti-duck immunoglobulin serum
(Nordic). Observation of the slides containing recombinant
baculovirus-infected Sf9 cells incubated with DPV duck serum,
confirmed the expression of DPV capsid proteins and showed
that they were localized in the nucleus of the cells (data not
shown). No reaction was observed either with these cells
following incubation with the SPF duck serum, or with wildtype baculovirus-infected Sf9 cells incubated with the DPV
antiserum. Nuclear location was described previously for
baculovirus expression of both VP1 and VP2 structural
proteins of Aleutian mink disease parvovirus (Christensen et
a]., 1993) and of recombinant AAV2 proteins, when VP1 or
VP2 with VP3 were coexpressed (Ruffing eta]., 1992).
We examined whether the recombinant proteins could selfassemble into VLPs. For this purpose, Sf9 cells infected by
recombinant baculoviruses as described above were pelleted
by centrifugation, washed with cold PBS, and resuspended in
a buffer containing 1% deoxycholate, 0"1% SDS, 0"1 mMEDTA, 10 mM-Tris-HC1 pH 8"0. Cells were incubated on ice
for 30 rain and centrifuged (12000g, 10 min, 4 °C). The
supernatant was layered onto a double cushion of 50 and 30 %
(w/v) sucrose and ultracentrifuged at l O 0 0 0 0 g for 2 h
(4 °C). The interface was diluted in PBS, centrifuged at
200 000 g for 2 h (4 °C) and the pellet was resuspended in PBS.
A fraction was examined by electron microscopy. For morphology comparison with wild-type DPV, virions were CsC1
gradient-purified as previously described (Le Gall-Recul6 &
Jestin, 1994) and examined at the same magnification. Electron
micrographs showed some aggregates of virus-like particles
with a morphology very similar to that of native DPV virions
(Fig. 2). The diameter of the particles was estimated to be about
22 nm, which is consistent with the size of DPV virions (Le
Gall-Recul6 & Jestin, 1994). However, few particles only were
detected in our purified preparations although we used a
combination of 0"I % SDS and 1% deoxycholate in case they
were present as hardly soluble aggregates in Sf9 cells. These
results revealed that recombinant capsid proteins VP2 and VP3
T a b l e 1. DPV antibodies in muscovy ducks inoculated w i t h recombinant baculovirus expressing VP2 and VP3 capsid
proteins
Before the third
inoculation
2 weeks after the third
inoculation
3 weeks after the third
inoculation
Duck no.
ELISA*
log z SNt-
ELISA
log z SN
ELISA
log z SN
1
2
3
4
5
6
7
8q:
Mean ratio
of the group
8.8 (o.s)
3-7 (0.6)
5-7 (0-6)
8"4 (1'1)
5"3 (0"6)
7"6 (1"8)
8"6 (0"6)
5"0 (0-0)
6'6
7-0
6.3
8"3
9"6
7"0
8"3
8"3
7"6
7"8
9-0 (0-0)
3.0 (0.0)
6"0 (0'0)
7"0 (0"0)
6'3 (0'6)
3"7 (0'6)
8"6 (0"6)
10-6 (0-6)
6"8
9.0
6.3
8"3
9'6
8"3
7"0
8"3
12-3
8"6
7-3 (1.2)
0-7 (0.6)
3"7 (0"6)
3"0 (0"0)
3"3 (0"6)
0"7 (0"6)
7"0 (0'0)
8"4 (0-6)
4"3
9-6
6-3
7-6
8-3
8"3
6"3
8-3
10"3
8"1
* Mean of at least three repetitions/serum (standard deviation).
t- Statistical mean of three repetitions.
:1: Duck no. 8 received 0'2 dose, 0"25 dose and 0"75 dose at first, second and third inoculation.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:01:10
_~16
iiiiii iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii !i i!ii!ii !iiii iii ii i ii iiii iiiiiiiiiiiiiiiiiiii
¸
were able to self-assemble into DPV-like empty capsids,
demonstrating that the presence of DPV VP1 was not
necessary for capsid formation. Similar observations have been
reported for the formation of VLPs of AAV2 (Ruffing et al.,
1992) and some autonomous parvoviruses (Kajigaya et al.,
1991; Martinez et al., 1992; Saliki et al., 1992; Christensen et
al., 1993, 1994).
For evaluating the immunity induced by the expressed
DPV capsid proteins, St9 cells were infected either with
recombinant baculovirus (RB) or with wild-type baculovirus
(WB) at an m.o.i, of 10 to 20 and harvested 48--60 h postinfection. Nineteen 3-week-old SPF muscovy ducks were
separated into three groups. Eight and five ducks received
subcutaneously three injections of respectively 107 infected
cells plus adjuvant in a final volume of I ml, at 2 week intervals.
The adjuvant was identical to the one included in the Parvol
vaccine (Rh6ne-M6rieux, France) against muscovy duck parvovirosis. Six more ducks were put in close contact with the latter
ones. Thereafter, the absence of specific antibodies having been
checked before the first injection, sera were collected from all
birds before the third injection, then 2 and 3 weeks later. For
titration of DPV antibodies, the G M ELISA test was used as
described (Jestin et al., 1992). This test employs the DPV G M
strain that had been shown to be very closely related to strain
89384 (Jestin et al., 1991 b). Each serum sample was at least
tested three times. Neutralizing antibodies were titrated in
decomplemented sera as previously described (Jestin et al.,
1991a) with the exception that Eagle's medium modified
McPherson-Stocker BHK21 (Eurobio) was used. Irrespective
of the time of serum sampling, no antibodies were detected in
ducks from group WB or contact ducks. O n the contrary, all
the RB ducks exhibited high to medium antibody ELISA titres
and medium serum neutralizing antibody (SN) titres (Table 1).
Two weeks after the third inoculation, no booster effect was
noticed except for the duck no. 8 that received lower doses
at each inoculation. One week later, antibody ELISA titres
were decreasing whereas SN titres were more stable. Thus, we
demonstrated that the expressed parvovirus capsid proteins
induced specific duck parvovirus antibodies. Our results are
very consistent with those obtained by testing the French
commercial bivalent inactivated oil vaccine (designed to
stimulate maternal immunity) on 2-week-old muscovy
ducklings. In fact, 2 and 3 weeks after the second injection
mean ELISA titres reached 3"4 then 4"3, respectively, and mean
seroneutralization titres (log2) 6"3 then 7"6 (V. Jestin, unpublished results). Thus, for breeder immunization, purified
VLPs plus adjuvant could be used.
In the present study, we have described for the first time the
expression of VP2 and VP3 capsid proteins of the muscovy
duck parvovirus in insect cells infected by recombinant
baculoviruses and we have demonstrated that immunization
with VLPs induced neutralizing antibodies against DPV in
ducks. A recombinant vaccine would be highly desirable. For
this purpose further improvements are required, notably the
_>16;
optimization of VLP expression levels, and the construction of
new recombinant baculoviruses is currently under investigation.
We are grateful to Dr Gdrard Devauchelle and Dr Martine Cdrutti
(INRA, CNRS UA 1184, St-Christol-les-A16s, France) for providing the
wild-type baculovirus, the St9 cells and the pGmAc3MR transfer vector.
We thank Claire Arnauld for DNA sequencing. This work was supported
by a grant from the General Council of Brittany Region (BRITTA
convention).
References
Becerra, S. P., Koczot, F., Fabisch, P. & Rose, J. A. (1988). Synthesis
of adeno-associated virus structural proteins requires both alternative
mRNA splicing and alternative initiations from a single transcript. Journal
of Virology 62, 2745-2754.
Cassinotti, P., Weitz, M. & Tratschin, J.-D. (1988). Organization of the
adeno-associated virus (AAV) capsid gene: mapping of minor spliced
mRNA coding for virus capsid protein 1. Virology 167, 176--184.
Christensen, J., Storgaard, T., Bloch, B., Alexandersen, S. & Aasted, B.
(1993). Expression of Aleutian mink disease parvovirus proteins in a
baculovirus vector system. Journal of Virology 67, 229-238.
Christensen, J., Alexandersen, S.~ Bloch, B., Aasted, B. & Uttenthal, A.
(1994). Production of mink enteritis parvovirus empty capsids by
expression in a baculovirus vector system: a recombinant vaccine for
mink enteritis parvovirus in mink. Journal of General Virology 75, 149-155.
Dalibart, V., Plassiart, G., Cherel, Y., Hurtrel, M. & Wyers~ M. (1993).
Caract6risation des 16sions histologiques de la Parvovirose spontan6e du
canard de Barbarie (Cairina moschata). Diagnostic histopathologique
diff6rentiel avec la R6ovirose. Recueil de Mddecine Vdtdrinaire 169,
763-772.
Drouin, P., Toux, J. Y., L'Hospitalier, R., Guittet, M. & Bennejean, G.
(1991). D'apr~s les donn6es du r6seau national d'observations
6pid6miologiques en aviculture, Evolution de l'incidence de la maladie de
Derzsy en 1989 et 1990. L'Aviculteur 521, 36--40.
Fournier, D. & Gaudry, D. (1992). Muscovy duck parvovirus (MDP) in
France: field vaccination trials. Proceedings of the 9th International
symposium on Waterfowl, Pisa, Italy, 16--18 September, 36-41.
Jestin, V. (1990). Le point sur l'6pizootie actuelle de maladie de Derzsy
chez le canard de Barbarie. L'Aviculteur 516, 77-78.
Jestin, V., Le Bras, M. 0., Cherbonnel, M., Le Gall, G. & Bennejean, G.
(1991 a). Isolement de virus de la maladie de Derzsy tr6s pathog~nes
chez le canard de Barbarie. Recueil de Mddecine Vdtdrinaire 167, 849-857.
Jestin, V., Cherbonnel, M., Le Bras, M. 0., Morin, Y., Petit, J., Le Coq,
L., Le Moal, L., Jarnet, G., Ogier, R. & Quitin, E. (1991 b). Contr61e de
l'immunit4 contre la maladie de Derzsy. L'Aviculteur 521, 59--60.
]estin, V., Le Bras, M. O. & Cherbonnel, bl. (1992). Control of muscovy
duck parvovirosis by vaccination. European Commission Proceedings: New
and Evolving Virus Diseases of Poultry. Brussels, Belgium, 15-16 December,
167-181.
Jestin, V., Le Bras, M. O. & Cherbonnel, M. (1993). Contr61e de la
parvovirose du canard par la vaccination. Proceeding I~res )ourn~es de la
recherche sur Ies palmipides ~ foie gras, INTRA/ITAVI, Bordeaux, France,
27-28 avril, 119-137.
Jestin, V., Le GalI-Recul4, G., Arnauld, C., Cherbonnel, M. & Jestin, A.
(1995). Detection of muscovy duck parvovirus DNA by the polymerase
chain reaction, lmmunobiotogy of viral infections. Proceedings of the 3rd
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:01:10
Congress of the European Society for Veterinary Virology, Interlaken 1994,
528-532. Edited by M. Schwyzer.
Kajigaya, S., Fujii, H., Field, A. M., Rosenfeld, S., Anderson, L.l.,
Shimada, T. & Young, N.S. (1991). Self-assembled B19 parvovirus
capsids, produced in a baculovirus system, are antigenically and
immunogenically similar to native virions. Proceedings of the National
Academy of Sciences, USA 88, 4646-4650.
Le GalI-Recul6, G. & lestin, V. (1994). Biochemical and genomic
characterization of muscovy duck parvovirus. Archives of Virology 139,
121-I31.
Le GalI-Recul4, G. & lestin, V. (1995). Production of digoxigeninlabelled DNA probe for detection of muscovy duck parvovirus. Molecular
and Cellular Probes 9, 39-44.
Lopez de Turiso, J. A., Cortes, E., Martinez, C., Ruiz de Ybanez, R.,
Simarro, I., Vela, C. & Casal, I. (1992). Recombinant vaccine for canine
parvovirus in dogs. Journal of Virology 66, 2748-2753.
Luckow, V. A. & Summers, M. D. (1988). Trends in the development of
baculovirus expression vectors. Biotechnology 6, 47-55.
Martinez, C., Dalsgaard, K., Lopez de Turiso, J. A., Cortes, E., Vela, C.
& Casal, J. I. (1992). Production of porcine parvovirus empty capsids
with high immunogenic activity. Vaccine 10, 684---690.
Ruffing, M., Zentgraf, H. & Kleinschmidt, J. A. (1992). Assembly of
viruslike particles by recombinant structural proteins of adeno-associated
virus type 2 in insect ceils. Journal of Virology 66, 6922--6930.
Saliki, J.T., Mizak, B., Flore, H.P., Gettig, R.R., Burand, J.P.,
Carmichael, L E., Wood, H.A. & Parrish, C.R. (1992). Canine
parvovirus empty capsids produced by expression in a baculovirus
vector: use in analysis of viral properties and immunization of dogs.
]ournal of General Virology 73, 369-374.
Trempe, I.P. & Carter, B.J. (1988). Altemate mRNA splicing is
required for synthesis of adeno-associated virus VP1 capsid protein.
Journal of Virology 62, 3356-3363.
Received 4 April 1996; Accepted 23 April 1996
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:01:10
~_16!