Induction of a Protective Immune Response ... - Journal of Virology

7 downloads 0 Views 115KB Size Report
May 29, 2006 - striped jack nervous necrosis virus (SJNNV), barfin ffounder nervous necrosis .... g) for 45 min in an SW50.1 rotor (Beckman). After the run, the ...
JOURNAL OF VIROLOGY, Oct. 2006, p. 10201–10207 0022-538X/06/$08.00⫹0 doi:10.1128/JVI.01098-06 Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Vol. 80, No. 20

Induction of a Protective Immune Response against Viral Nervous Necrosis in the European Sea Bass Dicentrarchus labrax by Using Betanodavirus Virus-Like Particles R. Thie´ry,1* J. Cozien,1 J. Cabon,1 F. Lamour,1 M. Baud,1 and A. Schneemann2 French Food Safety Agency, BP 70, F-29280 Plouzane´, France,1 and Department of Molecular Biology, The Scripps Research Institute, La Jolla, California 920372 Received 29 May 2006/Accepted 26 July 2006

Betanodaviruses are causative agents of viral nervous necrosis (VNN), a devastating disease of cultured marine fish worldwide. Virus particles contain a single type of coat protein that spontaneously assembles into virus-like particles (VLPs) when expressed in a baculovirus expression system. In the present study, the immunogenicity of betanodavirus VLPs and the protection they confer against VNN in the European sea bass Dicentrarchus labrax were investigated. Enzyme-linked immunosorbent assay and seroneutralization tests performed on plasma from fish vaccinated intramuscularly with doses as low as 0.1 ␮g of VLPs indicated that the VLPs elicited the synthesis of specific antibetanodavirus antibodies with neutralizing activity. Moreover, fish vaccinated with VLPs were protected from challenge with live virus. Both the immune response and the protective effect against viral challenge were dose dependent. Reverse transcription-PCR data indicated that higher doses of vaccine also reduced the number of fish containing detectable quantities of betanodavirus RNA on day 30 after challenge. Taken together these data strongly support the hypothesis that VLPs obtained in the baculovirus expression system may represent an effective vaccine against VNN. The family Nodaviridae comprises the recently established genera Alphanodavirus and Betanodavirus, members of which predominantly infect insects and fish, respectively (25). Viruses belonging to the betanodavirus genus are the causative agents of viral encephalopathy and retinopathy, also known as viral nervous necrosis (VNN), a devastating disease of many species of marine fish cultured worldwide (3, 19). Infected fish commonly display neurological disorders, which are often associated with strong vacuolization of the central nervous system and the retina. Betanodaviruses are small, spherical, nonenveloped viruses with a genome that is composed of two single-stranded, positive-sense RNA molecules. The larger genomic segment, RNA1 (3.1 kb), encodes the RNA-dependent RNA polymerase (8, 20, 31), while the smaller genomic segment, RNA 2 (1.4 kb), encodes the coat protein (11, 22). Phylogenetic analysis of the coat protein sequences of several betanodavirus isolates prompted their classification into four genotypes, designated striped jack nervous necrosis virus (SJNNV), barfin flounder nervous necrosis virus (BFNNV), tiger puffer nervous necrosis virus, and red-spotted grouper nervous necrosis virus (RGNNV) (10, 21, 26, 34). At present, there is neither a treatment nor a vaccine available to prevent VNN in fish. The control of the disease is based upon virus detection in infected animals. This relies on several diagnostic methods, including isolation of the causative agent and/or detection of viral components such as antigens or genome fragments (23). Identification of putative virus-free breeders can be accomplished by specific antibetanodavirus

* Corresponding author. Present address: French Food Safety Agency, BP111, F-06902 Sophia Antipolis, France. Phone: 33 (0) 4 92 94 37 07. Fax: 33 (0) 4 92 94 37 01. E-mail: [email protected].

antibody screening of the brood stock (2, 6). In addition, strict disinfection procedures using various physical or chemical agents capable of inactivating VNN viruses are recommended on farms where infection has occurred, but the procedures are difficult to apply in practice (12). It is widely accepted that a vaccine capable of preventing VNN in fish would be a major improvement that could lead to effective control of the disease and to a reduction of significant economical losses in the fish industry. Partial protective immunity has been obtained in several fish species by using recombinant betanodavirus coat proteins expressed in Escherichia coli (14, 32, 35), whereas synthetic peptides derived from the coat protein of sea bass nodavirus were poorly protective (9). DNA-based vaccination with constructs encoding betanodavirus coat protein has been met with limited success in Atlantic halibut (27), turbot (30), and sea bass (Kerbart-Boscher and Thie´ry, unpublished observations). On the other hand, a plasmid carrying the gene for the glycoprotein of viral hemorrhagic septicemia virus was recently reported to induce an early protection against betanodavirus challenge in turbot, suggesting a role of nonspecific defense mechanisms (29). Virus-like particles (VLPs) of the betanodavirus malabaricus grouper nervous necrosis virus (MGNNV) were previously generated by expression of the coat protein in Sf21 cells, using a recombinant baculovirus vector (16). The morphology of the recombinant particles is similar, if not identical, to that of native virions, but they do not contain the viral genome (33). Rather, VLPs package random cellular RNA and are therefore not infectious. In the present study, the potential of such VLPs as a vaccine against VNN in sea bass was investigated in vivo. A strong protective immune response against experimental infection with native virus was obtained in sea bass vaccinated with purified MGNNV VLPs or VLPs obtained after expression of the coat protein of SB2, a previously characterized

10201

10202

´ RY ET AL. THIE

betanodavirus isolate from clinically affected sea bass (34). Both the immune response and the protection were found to correlate with the administered dose. To our knowledge, this is the first report demonstrating the use of VLPs to protect fish against viral infection. MATERIALS AND METHODS Construction of recombinant baculoviruses expressing the coat protein of betanodavirus isolates. Construction of a recombinant baculovirus expressing the coat protein of MGNNV, a betanodavirus isolated from the brain of infected Epinephelus malarabicus grouper, was previously described in detail (16). Basically, the same procedure was applied to obtain a recombinant baculovirus containing the coat protein gene of a betanodavirus obtained from diseased sea bass Dicentrarchus labrax, isolate V26 (24), with slight modifications. Both viruses belong to the RGNNV genotype (34). The coat protein gene of the V26 isolate was amplified by using reverse transcription (RT)-PCR and cloned into plasmid pcDNAI (Invitrogen) as described previously (34). It is here referred to as pSB2. The plasmid pSB2 was used as a template to reamplify the entire V26 coat gene by using PCR with PfuTurbo polymerase (Stratagene) and specific N-terminal and C-terminal primers, including a BglII site or a NotI site, respectively, to facilitate subsequent cloning steps. The amplified product was gel purified, using the QIAEX II Gel extraction kit (QIAGEN), digested with BglII and NotI restriction enzymes, and cloned into plasmid pBacPAK9 (Clontech), which was linearized with the same restriction enzymes. After screening of the bacterial colonies by PCR, a recombinant shuttle plasmid vector, here referred to as pBacPAK9/SB2, was selected, purified, and then cotransfected with BacPAK6 viral DNA (Clontech) into Sf21 cells according to the manufacturer’s protocols. Five plaque-purified baculovirus recombinants were used for amplification and diagnostic tests regarding expression of SB2 coat protein and assembly into VLPs. To this end, 2.5 ⫻ 106 Sf21 cells were infected with a single plaquepurified recombinant. After a 6-day incubation at 27°C, the infected cells were pelleted, resuspended in 1 ml phosphate-buffered saline (PBS), and lysed with 0.5% (vol/vol) NP-40. Cell debris was removed by centrifugation in a microcentrifuge. The supernatant was transferred to SW50.1 ultracentrifuge tubes (Beckman) and underlain with 0.5 ml of 20% (wt/wt) sucrose in 10 mM Tris, pH 8. The tubes were filled to the top with PBS and centrifuged at 45,000 rpm (243,000 ⫻ g) for 45 min in an SW50.1 rotor (Beckman). After the run, the tubes were drained and the pellets, containing putative VLPs, were resuspended in 50 ␮l PBS. Five microliters was analyzed on a 12% Laemmli sodium dodecyl sulfate polyacrylamide gel to determine whether SB2 coat protein was present. A recombinant baculovirus judged positive by protein gel electrophoresis was selected for further experiments. This virus stock was designated BV-SB2. Large-scale production and purification of MGNNV VLPs and SB2 VLPs. Tricoplusia ni cells were propagated in serum-free ExCell405 medium (JRH Biosciences). A 1-liter T. ni cell culture at a density of approximately 2 ⫻ 106 cells/ml was infected with 30 ml of each recombinant baculovirus stock (BV-B9M or BV-SB2) and incubated at 27°C for 3 to 5 days. Cells were then pelleted by low-speed centrifugation, resuspended in 200 ml of buffer (10 mM Tris, pH 8 [MGNNV VLPs], or 50 mM HEPES–10 mM EDTA, pH 7.4 [SB2 VLPs]), and lysed with 0.5% NP-40 (vol/vol) at 4°C for 10 min. The cell debris was pelleted by centrifugation at 15,000 ⫻ g for 15 min at 4°C. The supernatant was subjected to ultracentrifugation (245,000 ⫻ g at 11°C for 2.5 h) in 20% (MGNNV VLPs) or 30% (SB2 VLPs) (wt/wt) sucrose in 10 mM Tris, pH 8 (MGNNV VLPs), or 50 mM HEPES–10 mM EDTA, pH 7.4 (SB2 VLPs). The VLP pellet was resuspended in 0.5 ml of 10 mM Tris, pH 8 (MGNNV VLPs), or 50 mM HEPES–10 mM EDTA, pH 7.4 (SB2 VLPs). For the MGNNV VLPs, a step of RNase A digestion (5 ␮g/ml for 10 min at room temperature) followed by low-speed centrifugation was then performed. The samples were subsequently layered on 10 to 40% (wt/wt) sucrose gradients in 10 mM Tris, pH 8 (MGNNV VLPs), or 50 mM HEPES–10 mM EDTA, pH 7.4 (SB2 VLPs), and centrifuged at 141,000 ⫻ g for 3 h at 11°C. The gradients were fractionated, and 2 to 3 ␮l of each fraction was analyzed on a protein gel. Fractions containing VLPs were pooled and dialyzed against 10 mM Tris, pH 8–10 mM NaCl (MGNNV VLPs) or 50 mM HEPES, pH 7.4 (SB2 VLPs). MGNNV VLPs were further purified by ultracentrifugation at 112,000 ⫻ g in 32% CsCl (wt/wt) overnight at 11°C. The gradient was fractionated, and fractions containing the VLPs were pooled. The pooled sample was dialyzed against 10 mM Tris (pH 8) and concentrated. The integrity of the VLPs was confirmed by electron microscopy of negatively stained samples, and the number of VLPs in each preparation was estimated by protein gel electrophoresis and comparison with a known amount of the insect nodavirus flock house virus. Purified VLPs were stored at ⫺80°C until use.

J. VIROL. Vaccination of sea bass. Two independent vaccination trials were done, using MGNNV VLPs (trial 1) and SB2 VLPs (trial 2), respectively. In both trials, sea bass Dicentrarchus labrax originating from a French commercial hatchery with no history of VNN were used. Fish weighing 66 g (trial 1) or 22 g (trial 2) on average were transferred into 40-liter tanks supplied with running, UV-treated, aerated seawater regulated at 25°C ⫾ 1°C (25 fish per tank). Fish were fed with a commercial dried pellet and acclimatized for 2 weeks (trial 1) or 1 week (trial 2) before being vaccinated. Prior to vaccination and sampling, the fish were starved for 24 h and anesthetized, using 0.2 ‰ (vol/vol) phenoxyethanol, just before delivery of the vaccine. The vaccine preparation consisted of thawed VLPs that were diluted in PBS. The fish were vaccinated individually with a single intramuscular injection of 100 ␮l of vaccine containing different amounts of VLPs. In trial 1, each fish received 20 ␮g or 100 ␮g of MGNNV VLPs. In trial 2, six doses of vaccine were tested: 20 ␮g, 10 ␮g, 5 ␮g, 1 ␮g, 0.5 ␮g, or 0.1 ␮g of SB2 VLPs. Each dose of vaccine was tested in triplicate on fish held in separate tanks (25 fish/tank). Unvaccinated control fish were treated in the same way but received 100 ␮l PBS each (trial 1, two tanks of 25 fish; trial 2, three tanks of 25 fish). Fish behavior and mortality were monitored during the immunization period (about 4 weeks) in both experiments. Serological assays, ELISA, and seroneutralization. Blood samples were taken from 3 to 5 individual fish in each tank at 27 days (trial 1) or 29 days (trial 2) postimmunization. The fish were starved for 24 h before sampling, sacrificed, and bled by puncture of the dorsal aorta or the caudal vein, using vacutainers coated with lithium-heparin (Becton Dickinson). The tubes were kept on ice and centrifuged at 1,200 ⫻ g for 10 min at 4°C. Each individual plasma sample was aliquoted and kept at ⫺80°C until use. The titer of specific antibetanodavirus antibodies from each individual plasma sample was determined by enzymelinked immunosorbent assay (ELISA) as described previously (6), with modifications as follows. A biotinylated rabbit polyclonal antibody to sea bass immunoglobulins was used instead of a monoclonal antibody for signal development. Twofold serial dilutions of each tested plasma sample were assayed in parallel to the same dilutions of a positive standard plasma sample obtained from a sea bass that was experimentally infected with betanodavirus strain V26. A negative standard plasma sample from naive fish was also included on each plate. Colorimetric detection using extravidin-peroxydase and orthophenylenediamine was performed by reading the optical density at 492 nm. The betanodavirus antibody titer was expressed as the optical density at 492 nm (OD492) for a constant dilution (1:8,192). Statistical differences between titers from the different groups were assessed by analysis of variance (ANOVA) and paired Student’s t tests. Titration of neutralizing antibodies was performed on individual plasma samples heat treated at 45°C for 30 min to inactivate the complement. Twofold serial dilutions (1:40 to 1:5,120) were prepared in 96-well plates by mixing 50 ␮l of plasma with 50 ␮l of L15 medium. Then 50 ␮l of a suspension of V26 isolate at 10 50% tissue culture infective doses (TCID50)/␮l was added to each well. Following incubation of the plates at 24°C overnight, 100 ␮l from each well was transferred to 96-well plates containing SNN-1 cells grown for 24 to 48 h. The plates were incubated at 24°C for 5 to 6 days, after which they were processed for immunofluorescence staining, using a rabbit polyclonal antibody to betanodavirus as described previously (7). Neutralizing antibody titer was expressed as the reciprocal of the plasma dilution for which the number of fluorescent cells was approximately equal to half of that observed for a negative control plasma. Nodavirus challenge in sea bass and sampling. To investigate if vaccinated fish were protected against VNN, nodavirus strain W80 (RGNNV genotype) (34) was used in both trials. Strain W80 was propagated in the SSN-1 cell line, and the titer was determined as described previously (7). On day 27 (trial 1) or day 30 (trial 2) postvaccination, all remaining fish (20 per tank) from the vaccinated or nonvaccinated groups were anesthetized and infected by intramuscular injection as described previously (24), using 100 ␮l of culture supernatant with a virus concentration of 103 TCID50/␮l. The tanks were supplied with aerated seawater at 25°C ⫾ 1°C, as during the immunization phase. Fish behavior, clinical signs of disease, and mortality were recorded for 1 month after challenge. Differences in cumulative mortality among fish receiving different treatments were tested for statistical significance by ANOVA and by Student’s t tests. Relative percent survival (RPS) values were calculated for each treatment (15). In both trials, the mortality was considered stable after about 1 month postchallenge. At this point, varying numbers of fish were sacrificed and frozen at ⫺80°C for subsequent detection of the nodavirus genome by RT-PCR. In trial 1, all surviving fish were stored frozen at ⫺80°C. In trial 2, five apparently healthy fish per tank, as well as all the remaining fish that displayed signs of disease, such as floating belly up and swimming in an uncoordinated manner, were stored frozen. Fish that died during trial 2 were also kept at ⫺80°C, except for those whose eyes and brains had been eaten due to cannibalism.

VOL. 80, 2006

VLP VACCINE AGAINST VIRAL NERVOUS NECROSIS IN SEA BASS TABLE 1. Immunization dataa

Trial

1 2

Treatment or VLP dose

No. of fish seropositive/no. of fish examined per tank

Range of neutralizing titerb

20␮g · fish⫺1 100␮g · fish⫺1 Unvaccinated 0.1␮g · fish⫺1 0.5␮g · fish⫺1 1␮g · fish⫺1 5␮g · fish⫺1 10␮g · fish⫺1 20␮g · fish⫺1 Unvaccinated

5/5, 5/5, 5/5 5/5, 5/5, 5/5 0/5, 0/5 4/4, 5/5, 5/5 4/4, 5/5, 5/5 5/5, 4/4, 5/5 5/5, 5/5, 5/5 4/4, 5/5, 5/5 5/5, 5/5, 5/5 0/4, 0/3, 0/3

0–⬎5120 1280–⬎5120 0 0–320 0–1,280 40–1,280 1,280–2,560 1,280–5,120 1280–⬎5,120 0

a Fish were immunized with MGNNV VLPs (trial 1) or SB2 VLPSs (trial 2). Total levels of specific antibetanodavirus antibodies were measured by ELISA in plasma sampled from 3 to 5 fish per tank, 27 days (trial 1) or 29 days (trial 2) after vaccination. Plasma samples with OD values of ⬎0.2 at a dilution of 1:8,192 were considered positive. The neutralizing titer is expressed as the reciprocal value of the plasma dilution that gave at least 50% reduction of the titer of nodavirus strain W80 grown on the SSN-1 cell line. b 0, plasma dilution of ⬍ 40.

Detection of nodavirus by RT-PCR. Betanodavirus RNA detection was performed by RT-PCR, using total RNAs extracted from the eyes and the brain of individual fish that had been stored at ⫺80°C. The RT-PCR assay was performed with primers derived from the coat protein gene of strain W80 and Ready-to-Go RT-PCR beads (Amersham Biosciences) as previously described (34). After amplification, the PCR products were analyzed by electrophoresis on a 2% agarose gel stained with ethidium bromide. Differences in the number of asymptomatic fish scoring positive for the presence of RNA2 between the different treatment groups were statistically assessed by chi-square analysis at a 95% confidence level.

RESULTS Fish behavior following immunization. Two independent vaccination trials were conducted in the sea bass Dicentrarchus labrax, using purified MGNNV VLPs (trial 1) and SB2 VLPs (trial 2). In both trials, a total volume of 100 ␮l was administrated intramuscularly by injection into the dorsal muscle of individual fish. In trial 1, fish were vaccinated with either 20 ␮g or 100 ␮g of MGNNV VLPs in PBS, whereas in trial 2, fish were vaccinated with doses ranging from 0.1 ␮g to 20 ␮g of SB2 VLPs in PBS. No adverse effects of these vaccine preparations could be observed during the immunization phase. Specifically, the behavior of the fish (swimming, appetite, excitability) was normal in all groups. In total, only three fish died during this phase (trial 1, one fish vaccinated with 100 ␮g of VLPs; trial 2, two fish vaccinated with 5 and 10 ␮g of VLPs, respectively). The mortality was unexplained, as no clinical signs of disease were observed before these fish died. No further mortality or morbidity was observed before challenge with live virus. VLP vaccine immunogenicity in sea bass. Plasma recovered from 3 to 5 fish per tank prior to challenge was assayed for the presence of specific antibetanodavirus antibodies by ELISA. In both trials, samples from all vaccinated fish were positive (Table 1), whereas samples from nonimmunized control fish were seronegative. In trial 1, the mean antibody titers at a dilution of 1:8,192 were 1.562 ⫾ 0.062 and 2.032 ⫾ 0.122 for fish vaccinated with 20 ␮g and 100 ␮g MGNNV VLPs, respectively. The difference between the two treatments was found to be statistically significant using the Student’s t test

10203

(P ⫽ 0.0083). In trial 2, the following values were obtained: 0.215 ⫾ 0.058 (0.1 ␮g), 0.468 ⫾ 0.086 (0.5 ␮g), 0.572 ⫾ 0.097 (1 ␮g), 1.105 ⫾ 0.163 (5 ␮g), 1.227 ⫾ 0.162 (10 ␮g), and 1.520 ⫾ (20 ␮g) (Fig. 1). ANOVA with one factor revealed statistically significant differences among treatments. This was confirmed by using paired Student’s t tests between all treatments. Strong statistical support was observed by comparing all treatments (P ⬍ 0.0001), except between fish groups vaccinated with 0.5 ␮g or 1 ␮g and with 5 ␮g or 10 ␮g, respectively (P ⬍ 0.05). Nevertheless, the mean antibetanodavirus antibody titer clearly increased with increasing VLP dose (Fig. 1). The same plasma samples were analyzed for betanodavirus neutralizing activity in cell culture. As shown in Table 1, plasma from all vaccinated groups displayed significant neutralizing titers. In general, the titers increased with the amount of VLPs in the inoculum. However, the neutralizing titers were found to be quite variable between individual fish, and several plasma samples did not exhibit any neutralizing activity at all. Surprisingly, in trial 1, plasma from 4 of 15 fish, each vaccinated with 20 ␮g of VLPs, did not contain detectable neutralizing activity, whereas all fish vaccinated with an equivalent dose responded in trial 2. Protection against virus challenge. On day 27 (trial 1) or day 30 (trial 2) postvaccination, all remaining fish from the vaccinated and nonvaccinated control groups were challenged with 105 TCID50 of betanodavirus strain W80 (RGNNV genotype). In trial 1, mortality and conspicuous clinical signs of VNN appeared between 5 and 6 days postinfection in the unvaccinated group. Cumulative mortality in unvaccinated fish, held in two separate tanks, reached a plateau after 9 days postinfection (90%) and 21 days postinfection (70%), respectively. At the end of the trial, several of the survivors displayed noticeable signs of VNN, such as hyperinflation of the swimming bladder and uncoordinated swimming. Mortality in the vaccinated group of fish was drastically lower than that for unvaccinated controls (Table 2 and Fig. 2). Specifically, fish vaccinated with 100 ␮g of VLPs displayed very low mortality (mean, 8.5%) compared to the control group (80.5%), and no mortality at all was observed in one of the

FIG. 1. Bar graph showing results of ELISA. Fish were vaccinated with SB2 VLPs (approximately 0.1 ␮g to 20 ␮g per fish) by intramuscular injection or received 100 ␮l of PBS. Twofold serial dilutions of individual plasma samples from 3 to 5 fish per tank were tested for the presence of antibetanodavirus antibodies by using a sandwich ELISA method. Detection was performed by colorimetric reading at OD492. The average OD readings at plasma dilution 1:8,192 for treatment is indicated.

10204

´ RY ET AL. THIE

J. VIROL. TABLE 2. Mortality and morbidity after betanodavirus challengea

Trial

Vaccine

1

MGNNV VLPs

2

SB2 VLPs

VLP dose (␮g · fish⫺1)

No. of fish dead/ no. of fish challenged per replicate

No. of moribund fish per tank

Mean RPS

20 100 Unvaccinated 0.1 0.5 1 5 10 20 Unvaccinated

4/19, 2/19, 7/19 0/19, 2/20, 3/20 18/20, 15/21 2/20, 4/20, 9/20 8/19, 7/20, 5/20 4/20, 3/20, 3/20 1/19, 2/20, 1/20 2/20, 2/20, 0/20 1/20, 1/19, 1/19 11/20, 7/20, 10/20

0, 0, 0 0, 0, 0 0, 3 1, 1, 1 0, 0, 1 1, 0, 1 0, 0, 0 0, 0, 0 1, 0, 0 2, 5, 2

71.7 89.4 46.5 27.4 64.2 85.6 85.6 88.9

a Dead fish were counted daily, and the cumulative mortality rate was calculated at 30 days postinfection (trial 1) or 28 days postinfection (trial 2). The number of moribund fish with conspicuous clinical signs of nodavirosis at the end of the experiments is also indicated. The RPS for each treatment was calculated as follows: (C ⫺ V)/C ⫻ 100, where V and C are the percentages of cumulative mortality in vaccinated fish and unvaccinated control fish, respectively.

tanks. Additionally, time to death was delayed by 3 days relative to that of unvaccinated controls. The data from trial 2 confirmed the importance of the VLP dose upon protection against challenge. Fish from treatment groups that received 0.1, 0.5, or 1 ␮g of SB2 VLPs appeared partially protected against W80 challenge, based on the calculated RPS values (Table 2), but the mean cumulative mortalities were not significantly different from that of the unvaccinated controls (P ⬎ 0.05). A high level of protection was observed when fish received at least 5 ␮g of SB2 VLPs (RPS values of ⬎80%), with strong statistical support (P ⬍ 0.001) (Table 2 and Fig. 3). This level of protection was similar to that observed in trial 1. Increasing the vaccine dose also delayed the onset of mortality (not shown). Detection of betanodavirus RNA2 in challenged fish. In order to verify that fish were actually infected after challenge, RT-PCR detection of betanodaviral RNA2 was performed on total RNA samples extracted from individual fish. The fish were classified into three categories: (i) fish that had died during the course of the experiment, (ii) apparently asymptom-

atic fish alive at the end of the experiment, and (iii) fish displaying conspicuous clinical signs of disease at the end of the experiment (Table 3). All dead fish that were analyzed were found positive by RT-PCR (trial 1, n ⫽ 2; trial 2, n ⫽ 69). As expected, betanodavirus RNA2 was also detected in all RNA samples from fish displaying clinical signs of VNN in both trials (trial 1, n ⫽ 3; trial 2, n ⫽ 16). On the other hand, many of the asymptomatic fish appeared to be free of virus. Specifically, the number of fish scoring positive for RNA2 decreased with increasing vaccine dose. In trial 1, only 4% of asymptomatic fish vaccinated with 100 ␮g of VLPs were found positive, compared to 27% of asymptomatic fish vaccinated with 20 ␮g of VLPs. The observed difference was found to be statistically significant using Fisher’s exact test (P ⫽ 0.0011) and chi-square analysis with Yates correction. In trial 2, vaccination with 20 ␮g of VLPs per fish also significantly reduced the average proportion of infected asymptomatic fish compared to that of the unvaccinated controls, but not when lower doses of the vaccine were used.

FIG. 2. Average cumulative mortality after betanodavirus challenge (trial 1). Fish were vaccinated with MGNNV VLPs (approximately 20 ␮g or 100 ␮g per fish) by intramuscular injection. Twenty-seven (27) days postvaccination, the fish were challenged with strain W80 (105 TCID50/ fish). Fish mortality was recorded daily. The percentage of average cumulative mortality is plotted against the number of days after challenge. Unvaccinated control fish received 100 ␮l of PBS.

FIG. 3. Average cumulative mortality rate after betanodavirus challenge (trial 2). Fish were vaccinated with different doses of SB2 VLPs (0.1 ␮g to 20 ␮g per fish) by intramuscular injection. Twentynine days postvaccination, the fish were challenged with betanodavirus strain W80 (105 TCID50/fish). Fish mortality was recorded daily. The percentage of average cumulative mortality for each group of fish on day 29 postbetanodaviral challenge is indicated. Groups showing a statistical difference compared to the unvaccinated control fish are marked with an asterisk.

VOL. 80, 2006

VLP VACCINE AGAINST VIRAL NERVOUS NECROSIS IN SEA BASS

10205

TABLE 3. Detection of betanodavirus RNA2 in infected fisha Treatment or VLP dose (␮g · fish⫺1)

Trial

Vaccine

1

MGNNV VLPs

2

SB2 VLPs

a b

20 100 Unvaccinated 0.1 0.5 1 5 10 20 Unvaccinated

No. of fish with VNN clinical signs positive by RT-PCR/no. of fish examined

3/3 3/3 1/1 2/2 1/1 9/9

No. of dead fish positive by RT-PCR/no. of fish examined

No. of asymptomatic fish positive by RT-PCR/no. of fish examined

NDb 1/1 1/4 14/14 18/18 9/9 4/4 4/4 3/3 17/17

12/44 2/54 1/5 14/15 15/15 14/15 14/15 11/15 6/15 17/17

RT-PCR analysis was performed using RNA extracted from the eyes and brains of challenged fish as described in Materials and Methods. ND, not determined.

DISCUSSION Vaccination is widely used in the salmon industry for effective control of a number of economically important diseases. To date, most available fish vaccines against viral diseases are traditional inactivated vaccines (28). One noticeable exception is a multivalent vaccine that includes a recombinant VP2 from infectious pancreatic necrosis virus, which can be used to protect salmonids against infectious pancreatic necrosis (13). However, the immunological basis of its efficacy is unknown. Although vaccines against bacterial diseases are available for the European sea bass, at present there is no available vaccine against VNN, one of the most significant viral diseases that can affect this fish species and other species of economical interest. In the present study, betanodavirus VLPs obtained from recombinant baculoviruses proved to be efficient against VNN in vaccine efficacy trials using the European sea bass as a model. Recombinant VLPs were not infectious and did not induce any noticeable side effects, as demonstrated by the insignificant level of mortality recorded in vaccinated fish. In fact, in the present study, only 3 out of 900 vaccinated fish died during the immunization phase. This low level of mortality is frequently observed in sea bass held in captivity. There were no obvious clinical signs of disease or gross pathological changes associated with these mortalities, and the vast majority of the fish displayed normal behavior, including appetite. Given that the VLPs do not contain replication-competent viral RNA, this result was expected. Unlike live attenuated viruses, there is no risk of residual infectivity or reversion to virulence, which makes VLPs excellent vaccine candidates. Because of the multivalent structure of the VLPs, which mimics that of native virions, it is likely that the stimulation of the fish immune system closely resembles that induced during a natural infection and thus does not necessitate the use of an adjuvant (4). In our study, the specific humoral immune response to betanodavirus VLPs was investigated, and a strong response was observed both by ELISA and by seroneutralization tests. The ELISA antibody titers were high and strongly correlated with the amount of VLPs given to the fish. High doses (20 and 100 ␮g · fish⫺1) induced a response of the same order of magnitude as observed for a positive standard plasma from sea bass that were experimentally infected with a live strain (data not shown). Compared to results of vaccination of

adult sea bass with an inactivated virus preparation (6), higher betanodavirus antibody titers were obtained in the present study. Heat inactivation at high temperature (100°C) of the virus used in the previous study may have destroyed the multimeric structure of the virions, thus explaining the reduction in antibetanodavirus antibody titer. Alternatively, differences in the sensitivity of the ELISA method, the age of the fish, or the water temperature could have accounted for the differences in results between the two studies. Injection of synthetic peptides derived from the betanodavirus coat protein sequence into juvenile sea bass triggered the synthesis of antibetanodavirus antibodies in a low proportion of the fish (9), but the antibody titers were not determined in that study. In contrast, in the present study 100% of VLP-vaccinated fish were found seropositive 4 weeks postimmunization. Recombinant coat proteins expressed in E. coli have been used to immunize fish against VNN in several species. An oil-emulgated recombinant partial coat protein (rT2) from SJNNV conferred partial protection to juvenile turbot Scophthalmus maximus challenged with SJNNV, although mortality in the unvaccinated control fish was low. This vaccine preparation also induced the synthesis of rT2-specific antibodies in adult turbot and Atlantic halibut Hippoglossus hippoglossus (14, 30). The kinetics of antibody induction were slower than those observed in the present study, which is likely due to differences in the immune systems of the studied fish species. Turbot vaccinated with 50 ␮g of recombinant coat protein demonstrated a threefold-lower survival rate than fish that received only 10 ␮g (30), in contrast to the present study where the higher doses of vaccines induced the higher protection. The sevenband grouper Epinephelus septemfasciatus was also protected against an experimental challenge when the fish received two intramuscular injections of 60 ␮g of recombinant coat protein with a 10-day interval between injections (32). However, the RPS value was lower than that obtained by using VLPs for an equivalent challenge dose (ca. 105 TCID50 · fish⫺1), which is probably due to the lower neutralizing antibody titers present in groupers at the time of challenge. Groupers Cromileptes altivelis were also partially protected against an experimental challenge (20 to 40% mortality versus 100% for the unvaccinated control) when fish received two intramuscular injections of 70 ␮g of a mixture of three recombinant coat

10206

´ RY ET AL. THIE

proteins (RGNNV genotype) with a 10-day interval between injections (35). Although naked DNA vaccination has been shown to be a promising approach to stimulate the fish immune system of salmonids in several models (1, 5, 17), disappointing results were obtained with a plasmid carrying the coat protein gene of the Atlantic halibut nodavirus, as it failed to protect fish against a viral challenge (30). Negative data were also obtained by using several plasmid constructs carrying the coat protein gene of nodaviruses isolated from sea bass in challenge experiments performed in sea bass (Kerbart-Boscher and Thie´ry, unpublished). Although these negative results are not fully understood, it is likely that the recombinant coat protein expressed in fish cells in vivo is not properly presented to the fish immune system or does not fold correctly. This does not occur with baculovirus-expressed betanodavirus coat protein, where the expressed protein itself assembles into VLPs. Recent work showed that betanodavirus strains fall into three major serotypes named A, B, and C (18). This serogrouping was in part consistent with their genotypes, i.e., serotype A for the SJNNV genotype, serotype B for the tiger puffer nervous necrosis virus genotype, and serotype C for both the RGNNV and BFNNV genotypes. In the present study, vaccine/challenge experiments were performed by using betanodavirus VLPs and a homologous wild-type strain for challenge (i.e., all belonging to the RGNNV serotype). This could explain the high level of protection that was observed in our experiments. It is likely that RGNNV-genotype VLPs would also protect fish against a heterologous challenge with BFNNV, as predicted from their serological properties, but not against strains belonging to serotypes A or B. However, more work is needed to substantiate this assumption. To conclude, we have demonstrated that a candidate vaccine comprised of betanodavirus VLPs produced in the baculovirus expression system could induce protective immunization against a homologous challenge in the European sea bass. Due to their inherent nature, such VLP-based vaccines could provide an efficient, safe, and economically viable strategy to control viral nervous necrosis in sea bass and possibly in other fish species. ACKNOWLEDGMENTS This work was supported in part by grant GM53491 (A.S.) from the National Institutes of Health and by the French Food Safety Agency (R.T.). REFERENCES 1. Anderson, E. D., D. V. Mourich, S. C. Fahrenkrug, S. LaPatra, J. Shepherd, and J. A. Leong. 1996. Genetic immunization of rainbow trout (Oncorhynchus mykiss) against infectious hematopoietic necrosis virus. Mol. Mar. Biol. Biotechnol. 5:114–122. 2. Arimoto, M., K. Muschiake, Y. Mizuta, T. Nakai, K. Muroga, and I. Furusawa. 1992. Detection of striped jack nervous necrosis virus (SJNNV) by enzyme linked immunosorbent assay (ELISA). Fish Pathol. 27:191–195. 3. Azad, I. S., M. S. Shekhar, A. R. Thirunavukkarasu, M. Poornima, M. Kailasam, J. J. Rajan, S. A. Ali, M. Abraham, and P. Ravichandran. 2005. Nodavirus infection causes mortalities in hatchery produced larvae of Lates calcarifer: first report from India. Dis. Aquat. Organ. 63:113–118. 4. Bachmann, M., and R. Zinkernagel. 1997. Neutralizing antiviral B cell responses. Annu. Rev. Immunol. 15:235–270. 5. Boudinot, P., M. Blanco, P. de Kinkelin, and A. Benmansour. 1998. Combined DNA immunization with the glycoprotein gene of viral hemorrhagic septicemia virus and infectious hematopoietic necrosis virus induces doublespecific protective immunity and nonspecific response in rainbow trout. Virology 249:297–306.

J. VIROL. 6. Breuil, G., and B. Romestand. 1999. A rapid ELISA method for detecting specific antibody level against nodavirus in the serum of the sea bass, Dicentrarchus labrax (L.): application to the screening of spawners in a sea bass hatchery. J. Fish Dis. 22:45–52. 7. Castric, J., R. Thie´ry, J. Jeffroy, P. de Kinkelin, and J. C. Raymond. 2001. Sea bream Sparus aurata, an asymptomatic contagious fish host for nodavirus. Dis. Aquat. Organ. 47:33–38. 8. Chi, S.-C., B.-J. Lo, and S.-C. Lin. 2001. Characterization of grouper nervous necrosis virus (GNNV). J. Fish Dis. 24:3–14. 9. Coeurdacier, J. L., F. Laporte, and J. F. Pepin. 2003. Preliminary approach to find synthetic peptides from nodavirus capsid potentially protective against sea bass viral encephalopathy and retinopathy. Fish Shellfish Immunol. 14:435–447. 10. Dalla Valle, L., E. Negrisolo, L. Patarnello, L. Zanella, C. Maltese, G. Bovo, and L. Colombo. 2001. Sequence comparison and phylogenetic analysis of fish nodaviruses based on the coat protein gene. Arch. Virol. 146:1125–1137. 11. Delsert, C., N. Morin, and M. Comps. 1997. A fish encephalitis virus that differs from other nodaviruses by its capsid protein processing. Arch. Virol. 142:2359–2371. 12. Frerichs, G. N., A. Tweedie, W. G. Starkey, and R. H. Richards. 2000. Temperature, pH and electrolyte sensitivity, and heat, UV disinfectant inactivation of sea bass (Dicentrarchus labrax) neuropathy nodavirus. Aquaculture 185:13–24. 13. Frost, P., and A. Ness. 1997. Vaccination of Atlantic salmon with recombinant VP2 of infectious pancreatic necrotic virus (IPNV), added to a mutivalent vaccine, suppresses viral replication following IPNV challenge. Fish Shellfish Immunol. 7:609–619. 14. Husgard, S., S. Grotmol, B. K. Hjeltnes, O. M. Rodseth, and E. Biering. 2001. Immune response to a recombinant capsid protein of striped jack nervous necrosis virus (SJNNV) in turbot Scophthalmus maximus and Atlantic halibut Hippoglossus hippoglossus, and evaluation of a vaccine against SJNNV. Dis. Aquat. Organ. 45:33–44. 15. Jarp, J., and A. Tverdal. 1997. Statistical aspects of fish vaccination, p. 311–320. In L. A. Gudding, R. Midtlyng, and P. J. Brown (ed.), Fish vaccinology. Karger, Basel, Switzerland. 16. Lin, C. S., M. W. Lu, L. Tang, W. Liu, C. B. Chao, C. J. Lin, N. K. Krishna, J. E. Johnson, and A. Schneemann. 2001. Characterization of virus-like particles assembled in a recombinant baculovirus system expressing the capsid protein of a fish nodavirus. Virology 290:50–58. 17. Lorenzen, N., E. Lorenzen, K. Einer-Jensen, J. Heppell, and H. L. Davis. 1999. Genetic vaccination of rainbow trout against viral haemorrhagic septicaemia virus: small amounts of plasmid DNA protect against a heterologous serotype. Virus Res. 63:19–25. 18. Mori, K., T. Mangyoku, T. Iwamoto, M. Arimoto, S. Tanaka, and T. Nakai. 2003. Serological relationships among genotypic variants of betanodavirus. Dis. Aquat. Organ. 57:19–26. 19. Munday, B., J. Kwang, and N. Moody. 2002. Betanodavirus infection of teleost fish: a review. J. Fish Dis. 25:127–142. 20. Nagai, T., and T. Nishizawa. 1999. Sequence of the non-structural protein gene encoded by RNA1 of striped jack nervous necrosis virus. J. Gen. Virol. 80:3019–3022. 21. Nishizawa, T., M. Furuhashi, T. Nagai, T. Nakai, and K. Muroga. 1997. Genomic classification of fish nodaviruses by molecular phylogenetic analysis of the coat protein gene. Appl. Environ. Microbiol. 63:1633–1636. 22. Nishizawa, T., K. I. Mori, T. Nakai, I. Furusawa, and K. Muroga. 1994. Polymerase chain reaction (PCR) amplification of RNA of striped jack nervous necrosis virus (SJNNV). Dis. Aquat. Org. 18:103–107. ´ pizooties (OIE). 2000. Viral encephalopathy and 23. Office International des E retinopathy, p. 69–73. In Diagnostic manual for aquatic animal diseases, 3rd ed. Office international des ´epizooties (OIE), Paris, France. 24. Peducasse, S., J. Castric, R. Thie´ry, J. Jeffroy, A. Le Ven, and F. Baudin Laurencin. 1999. Comparative study of viral encephalopathy and retinopathy in juvenile sea bass Dicentrarchus labrax infected in different ways. Dis. Aquat. Org. 36:11–20. 25. Schneemann, A., L. A. Ball, C. Delsert, J. E. Jonhson, and T. Nishizawa. 2004. Nodaviridae, p. 863–870. In C. M. Fauquet, M. A. Mayo, J. Maniloff, U. Dusselberger, and L. A. Ball (ed.), Virus taxonomy: eighth report of the International Committee on Taxonomy of Viruses. Academic Press, London, United Kingdom. 26. Skliris, G. P., J. V. Krondiris, D. C. Sideris, A. P. Shinn, W. G. Starkey, and R. H. Richards. 2001. Phylogenetic and antigenic characterization of new fish nodavirus isolates from Europe and Asia. Virus Res. 75:59–67. 27. Sommerset, I., S. Husgaard, and A. H. Nerland. 2001. Development of vaccines against nodavirus affecting Atlantic halibut (Hippoglossus hippoglossus). In 10th International Conference of the European Association of Fish Pathologists, Dublin, Ireland. 28. Sommerset, I., B. Krossoy, E. Biering, and P. Frost. 2005. Vaccines for fish in aquaculture. Expert Rev. Vaccines 4:89–101. 29. Sommerset, I., E. Lorenzen, N. Lorenzen, H. Bleie, and A. H. Nerland. 2003. A DNA vaccine directed against a rainbow trout rhabdovirus induces early protection against a nodavirus challenge in turbot. Vaccine 21:4661–4667. 30. Sommerset, I., R. Skern, E. Biering, H. Bleie, I. Uglenes Fiksdal, S. Grove,

VOL. 80, 2006

VLP VACCINE AGAINST VIRAL NERVOUS NECROSIS IN SEA BASS

and A. H. Nerland. 2005. Protection against Atlantic halibut nodavirus in turbot is induced by recombinant capsid protein vaccination but not following DNA vaccination. Fish Shellfish Immunol. 18:13–29. 31. Tan, C., B. Huang, S. F. Chang, G. H. Ngoh, B. Munday, S. C. Chen, and J. Kwang. 2001. Determination of the complete nucleotide sequences of RNA1 and RNA2 from greasy grouper (Epinephelus tauvina) nervous necrosis virus, Singapore strain. J. Gen. Virol. 82:647–653. 32. Tanaka, S., K. Mori, M. Arimoto, T. Iwamoto, and T. Nakai. 2001. Protective immunity of sevenband grouper, Epinephelus septemfasciatus Thunberg, against experimental viral nervous necrosis. J. Fish Dis. 24:15–22. 33. Tang, L., C. S. Lin, N. K. Krishna, M. Yeager, A. Schneemann, and J. E.

10207

Johnson. 2002. Virus-like particles of a fish nodavirus display a capsid subunit domain organization different from that of insect nodaviruses. J. Virol. 76:6370–6375. 34. Thie´ry, R., J. Cozien, C. de Boisse´son, S. Kerbart-Boscher, and L. Ne´varez. 2004. Genomic classification of new betanodavirus isolates by phylogenetic analysis of the coat protein gene suggests a low host fish species specificity. J. Gen. Virol. 85:3079–3087. 35. Yuasa, K., I. Koeshayani, K. Roza, K. Mori, M. Katata, and T. Nakai. 2002. Immune response of humpback grouper, Cromileptes altivelis (Valenciennes) injected with the recombinant coat protein of betanodavirus. J. Fish Dis. 25: 53–56.