Persistently betanodavirus-infected barramundi (Lates calcarifer) exhibit resistances to red sea bream iridovirus infection

Developmental and Comparative Immunology 41 (2013) 666–674

Contents lists available at ScienceDirect

Developmental and Comparative Immunology journal homepage: www.elsevier.com/locate/dci

Persistently betanodavirus-infected barramundi (Lates calcarifer) exhibit resistances to red sea bream iridovirus infection Yu-Chi Wu a,1, Yu-Hsuan Kai b,1, Shau-Chi Chi a,b,⇑ a b

Department of Life Science, National Taiwan University, Taipei 10617, Taiwan Institute of Zoology, National Taiwan University, Taipei 10617, Taiwan

a r t i c l e

i n f o

Article history: Received 7 April 2013 Revised 13 June 2013 Accepted 17 June 2013 Available online 4 July 2013 Keywords: Betanodavirus Megalocytivirus Interferon Mx Barramundi

a b s t r a c t Nervous necrosis virus (NNV) and red sea bream iridovirus (RSIV) are two important pathogens that have caused acute, highly contagious, and widespread diseases among wild and cultured fish, especially at larval and juvenile stages. We discovered that the pathogenicity of NNV to the 80 days post-hatch (dph) barramundi is lower than that to the 14 dph barramundi. Following NNV challenge, no mortality occurred in the 80 dph barramundi, but NNV RNA2 and barramundi Mx (BMx) gene expression was detected in the brain and liver. The 80 dph barramundi pre-challenged with NNV became more resistant to the following RSIV challenge (mortality: 62%) compared to the NNV-free barramundi challenged with RSIV (mortality: 100%). A similar phenomenon was revealed in the cell culture system that RSIV proliferated less progeny in the barramundi brain (BB) cell line which exhibit persistent NNV infection than in NNV-free cured BB (cBB) cell line. The potential factors involved in the resistance of the persistently NNV-infected barramundi and BB cells to the secondary RSIV infection were examined in this study. We prove that barramundi anti-NNV polyclonal antibodies do not cross-neutralize RSIV, and NNV infection does not interfere with RSIV replication. However, the interferon (IFN) response and BMx gene expression in cBB cells suppresses the RSIV proliferation. Our study suggests that the NNV-induced IFN response and BMx expression are responsible for the resistance of barramundi to RSIV infection. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Piscine iridoviruses and nodaviruses are two important pathogens that caused destructive diseases among the larvae and juveniles of cultured marine fish in Taiwan (Chao and Pang, 1997; Chao et al., 2002; Chi et al., 1997, 2003; Chou et al., 1998; Huang et al., 2011; Wang et al., 2003, 2009). Piscine iridoviruses are large, cytoplasmic, and double-stranded DNA viruses with an icosahedral capsid of 120–240 nm in diameter that comprise three genera: Lymphocystivirus, Ranavirus, and Megalocytivirus (Williams et al., 2005). In Taiwan, the genetic analysis of iridovirus isolates in cultured fish from 2001 to 2009 showed that most isolates collected prior to 2005 belong to the Ranavirus genus and most isolates collected post 2005 belong to the Megalocytivirus genus (Huang et al., 2011). The red sea bream iridovirus (RSIV) is a member of the Megalocytiviruses and has caused epizootic diseases in Taiwan (Huang et al., 2011; Wang et al., 2003). Enlarged basophilic cells are consistently discovered in RSIV target organs, including the ⇑ Corresponding author. Address: Department of Life Science, National Taiwan University, No. 1, Sec. 4, Roosevelt Rd., Taipei 10617, Taiwan. Tel.: +886 2 33662505; fax: +886 2 23673852. E-mail address: [email protected] (S.-C. Chi). 1 These authors contributed equally to this study. 0145-305X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dci.2013.06.011

gills, kidneys, heart, liver, and spleen (Inouye et al., 1992; Wang et al., 2003). The nervous necrosis virus (NNV) belongs to Betanodavirus of Nodaviridae and is referred to as fish nodavirus. NNV is a small, icosahedral, and non-enveloped RNA virus with a 25–30 nm diameter, and comprises two single-stranded positive-sense RNA genomes without a poly-A tail. The RNA1 (3.1 kb) encodes viral RNA-dependent RNA polymerase (RdRp), and the RNA2 (1.4 kb) encodes the capsid protein (Chi et al., 2001; Mori et al., 1992; Munday et al., 2002). NNV mainly replicates in the central nervous system and causes vacuolation in the brain and retina (Chi et al., 1997). Fish surviving viral nervous necrosis (VNN) disease become NNV carriers (Johansen et al., 2003). The BB cell line was established from the brain tissue of a barramundi that survived VNN and was identified as exhibiting a persistent NNV infection (Chi et al., 2005). The persistence of NNV in BB cells involved the interferon (IFN) response (Wu and Chi, 2006). The IFN response is an important defense system against virus infection and induces numerous types of antiviral proteins, such as Mx protein – a marker of the IFN response (Samuel, 2001). After serial treatments of BB cells with NNV-specific polyclonal antibodies, an NNV-free cell line referred to as cured BB (cBB) cells was developed as a tool for studying the antiviral activity of IFN response and barramundi Mx

Y.-C. Wu et al. / Developmental and Comparative Immunology 41 (2013) 666–674

(BMx) protein (Wu and Chi, 2006). Our study shows that the BMx protein exhibits anti-NNV activity (Wu et al., 2010). The barramundi and grouper are two important cultured fish species in Taiwan and constantly suffer from fish nodavirus and iridovirus infections, separately or simultaneously. In our study regarding the pathogenicity of NNV and RSIV to these cultured fish, the barramundi or grouper that survived the NNV challenge normally became NNV carriers and acquired a resistance to the RSIV challenge. Discovering the mechanisms that promote the protection of NNV carriers against RSIV is important. Therefore, several potential mechanisms were speculated and verified, including whether a specific immune response induced by NNV antagonizes RSIV, whether NNV replication interferes with RSIV replication, and whether NNV-induced IFN response and BMx protein exhibits antiviral activity against RSIV.

2. Materials and methods 2.1. Fish, cell lines, viruses, and vaccine The 14 and 80 days post-hatch (dph) barramundi (Lates calcarifer) were obtained from a private barramundi hatchery farm in southern Taiwan. The average body weight and total body length of the 14 dph barramundi larvae were 0.009 g and 0.6 cm, and those of the 80 dph barramundi were 1.2 g and 4.3 cm. The barramundi were cultured in ozone-treated and re-circulated seawater, and fed with commercial dried pellets (Uni-President Enterprises Corp., Taiwan). The fish were confirmed to be NNV- and RSIV-free by real-time PCR that was described in Section 2.3. The BB and cBB cell lines were used in this study. The BB cell line exhibits a persistent NNV infection, and the cBB cell line is NNV-free (Chi et al., 2005; Wu and Chi, 2006). These two cell lines were cultured in Leibovitz’s L-15 medium (GIBCO) supplemented with 10% fetal bovine serum (FBS) (GIBCO) and incubated at 28 °C. Grouper fin cell line (GF-1) was cultured in L-15 medium with 5% FBS at 28 °C (Chi et al., 1999) and used for the propagation and titration of NNV. Sea bream kidney (SK) cell line was maintained in L-15 medium plus 10% FBS at 28 °C and used for the propagation of RSIV. A strain of fish iridovirus, named RSIV-GIG2004, was isolated from the head-kidney of a diseased giant grouper (Epinephelus lanceolatus) from the south of Taiwan, and identified as the Megalocytivirus via the gene sequence of the RSIV Pst I fragment which was amplified by the primer set IRD-1F/IRD-2R (Table 1) (Kurita et al., 1998). NNV was isolated from hump-back grouper (Chi et al., 2003). The preparation of inactivated NNV vaccine was based on the method described by Kai and Chi (2008).

667

2.2. Viral challenge test in barramundi The naive 80 dph barramundi were divided into seven groups (120 fish per group). Six groups were intramuscularly (IM) injected with NNV at doses of 104, 105, 106, 107, 108, and 109 TCID50 per fish, and one group was negative control injected with PBS. No mortality was recorded in each group 1 month post challenge. Fourteen days post NNV challenge, thirty fish pre-challenged with NNV (108 TCID50 per fish) were further intraperitoneally (IP) challenged with two doses of RSIV (104 and 106 TCID50 per fish). Concurrently, the negative control fish pre-challenged with PBS were challenged with the same doses of RSIV. The accumulated mortality of each group was recorded for 2 weeks. Naïve 80 dph barramundi from the same lot were divided into three groups (40 fish per group). Two groups were IM immunized with NNV vaccine at doses of 105 and 106 TCID50 per fish, and the third group (control group) was mock-immunized with PBS. Two months post NNV vaccination, all three groups were challenged with RSIV (104 TCID50 per fish). The accumulated mortality of each group was recorded for 2 weeks. The 14 dph barramundi larvae were divided into eight groups (120 fish per group). For bath challenge, four groups were respectively immersed in 5 l seawater supplemented with NNV (107, 108, and 109 TCID50) or with PBS as the negative control. For oral challenge, 50,000 of artemia were separately mixed with NNV at doses of 107, 108, and 109 TCID50 in 10 ml seawater for 30 min, and then three groups of the 14 dph barramundi larvae were fed with the NNV-supplemented artemia. One group of the 14 dph barramundi larvae fed with artemia pre-mixed with PBS was served as negative control. 2.3. Reverse transcription and real-time PCR Before challenge test, barramundi were identified as NNV/RSIVfree by real-time PCR. Five naïve 80 dph barramundi were randomly collected to detect the existence of NNV RNA2 in the brain and RSIV DNA in the liver. Because of the small body size of 14 dph barramundi larvae, entire fish was used for NNV and RSIV detection (five larvae for each virus detection). In NNV challenge test, five 80 dph barramundi were respectively sampled from the groups of 108 TCID50 NNV and PBS at 14 dpc. The brains and livers were removed to analyze the NNV RNA2 and BMx gene expression by real-time PCR. Besides, five 14 dph barramundi were respectively sampled from the groups of bath challenge (108 TCID50 NNV) and oral challenge (artemia pre-mixed with 108 TCID50 NNV) at 7 dpc. NNV RNA2 and BMx gene expression in the whole fish body of 14 dph barramundi was analyzed by real-time PCR. RNA extraction, reverse transcription (RT), real-time PCR for NNV RNA2 and BMx mRNA were performed following the method

Table 1 The sequences of primers and siRNAs. Name

Sequence

References

NNV RPCR-F NNV RPCR-R NNV R3 RSIV IRD-1F RSIV IRD-2R RSIV RY16-F RSIV RY16-R Mx150-F Mx150-R Actin-F Actin-R Mx-siRNA N-siRNA

50 -CAGTCCGACCTCAGTACAC-30 50 -AACACTCCAGCGACACAG-30 50 -CGAGTCAACACGGGTGAAGA-30 50 -CTCAAACACTCTGGCTCATC-30 50 -GCGTTAAAGTAGTGAGGGCA-30 50 -GGTATATGAGCAAGCGATGGAC-30 50 -TGGAGTGCCATACAGGATG-3 50 -TGAGGAGAAGGTGCGTCC-30 50 -GCGCCTCCAACACGGAGCTC-30 50 -CACTCAACCCCAAAGCCAACAGG-30 50 -AAAGTCCAGCGCCACGTAGCACAG-30 50 -r(GCAGAUAGAAGAGAAACUA)dTdT-30 50 -r(UUCUCCGAACGUGUCACGU)dTdT-30

Wu et al. (2010) Wu et al. (2010) Nishizawa et al. (1994) Kurita et al. (1998) Kurita et al. (1998) Chao et al. (2002) Chao et al. (2002) Trobridge and Leong (1995) Trobridge and Leong (1995) Larsen et al. (2004) Larsen et al. (2004)

668

Y.-C. Wu et al. / Developmental and Comparative Immunology 41 (2013) 666–674

described by Wu et al. (2010). The sequences of primers used in RT and real-time PCR are listed in Table 1. Briefly, the total RNA of brains, livers, or larvae was extracted by acid guanidinium thiocyanate–phenol–chloroform method (Chomczynski and Sacchi, 1987). The RNA was transcribed into cDNA by M-MLV reverse transcriptase (Promega) and primers, including NNV R3 and oligo dT20. Subsequently, the primer sets for BMx (Mx150-F and Mx150-R), NNV RNA2 (RPCR-F and RPCR-R) and actin (Actin-F and Actin-R) were used for real-time PCR. An aliquot (0.5 ll) of the cDNA was added into a real-time PCR mixture with a final volume of 20 ll containing 0.5 lM forward and reverse primers in 1 iQ SYBR Green Super-Mix (Bio-Rad). The amplification was carried out in MyiQ Real-time PCR Detection System (Bio-Rad) with an initial denaturing step of 95 °C for 3 min, followed by 40 cycles of 95 °C for 20 s, 60 °C for 20 s, 72 °C for 20 s, and fluorescence detection at 83 °C for 20 s. All samples were analyzed in triplicate. The expression level of BMx gene was normalized with internal control (actin). A pQE plasmid containing the NNV RNA2 ORF gene was used to establish the standard curve of NNV RNA2 copies in realtime PCR. The copies of NNV RNA2 in the brains, livers, or larvae were quantified. The total DNA of barramundi livers and larvae was extracted by Genomic DNA mini kit (Geneaid), and subjected to detection of RISV DNA by real-time PCR. The primer set for RSIV was RY16-F/ RY16-R, and the real-time PCR program was the same as that described above. A pGEM-T easy plasmid containing a RSIV gene fragment amplified by the primer set, RY16-F/RY16-R, was used to establish the standard curve of RSIV DNA copies in real-time PCR. The copies of RSIV DNA in the livers and larvae were quantified. 2.4. Neutralization assay Sera were collected from three 80 dph barramundi immunized with inactivated NNV vaccine, and the titers of neutralizing antibody against NNV and RSIV were examined. The sera were respectively 4-fold diluted in L-15 medium and filtrated through 0.22-lm membranes. Subsequently, the serum solutions were serially 2fold diluted in L-15 medium (1:4–1:2048). An equal volume of NNV or RSIV with a titer of 103 TCID50 ml1 was mixed with the diluted sera and shaken for 1 h at room temperature. Each mixture was inoculated into 4 wells of 96-well culture plate pre-seeded with GF-1 cells for NNV or cBB cells for RSIV (0.2 ml per well). After 1 h virus absorption, the mixtures were removed and replaced by fresh L-15 medium. The CPE were determined at 6 days post virus inoculation, and the neutralizing antibody titer against NNV and RSIV were calculated. Alpha neutralization test (constant antiserum plus serially diluted viral solution) was applied to measure the neutralization index (NI) of the rabbit anti-NNV serum against NNV and RSIV. Rabbit anti-NNV serum was first 100-fold diluted in L-15 medium and filtrated through a 0.22-lm membrane. NNV (1010 TCID50 ml1) and RSIV (106 TCID50 ml1) solutions were serially 10-fold diluted in L-15 medium, respectively, and mixed with equal volume of the prepared anti-NNV serum. After 1 h shaking at room temperature, each mixture was inoculated into 4 wells of 96-well culture plate pre-seeded with GF-1 cells for NNV titration or cBB cells for RSIV titration (0.2 ml per well). For the control, the serially 10-fold diluted viral solution was mixed with equal volume of L-15 medium, and each mixture was inoculated into 4 wells (0.2 ml per well) of 96-well cultured plate pre-seeded with cells. Viral titers were determined at 6 days post virus inoculation. The NI value of the anti-NNV serum against each virus was expressed as the viral titer from the viral solutions mixing with L-15 medium divided by the viral titer from the viral solutions mixing with anti-NNV serum. NI values over 50 (log NI = 1.7) are considered significant (Mahy and Kangro, 1996).

2.5. Infection of NNV and/or RSIV in BB and cBB cells To realize the influence of persistent NNV infection on RSIV proliferation in vitro, the BB and cBB cells respectively pre-seeded in the 4-well plate (4  104 cells per well) were infected with RSIV (MOI = 10). After 1 h virus adsorption, the virus which did not infect the cells was washed out by L-15 medium for 3 times, and then the cells were incubated with fresh L-15 medium. The titers of the progeny RSIV at 7 days post infection (dpi) were determined. To clarify whether NNV and RSIV interfere each other in the virus replication, the cBB cells pre-seeded in the 4-well plate (4  104 cells per well) were sole-infected with NNV or RSIV, or co-infected with these two viruses simultaneously (MOI = 10). After 1 h virus adsorption, the cells were washed and incubated with fresh L-15 medium. The progeny NNV and RSIV at 24 h post infection (hpi) were titrated. To realize whether the IFN response in cBB cells antagonizes RSIV replication, the cBB cells pre-seeded in the 4-well plate (4  104 cells per well) were transfected with poly I:C (1 lg ml1) by lipofectamine 2000 (Invitrogen) and incubated for 24 h. Following RSIV infection (MOI = 10) for 1 h, the cells were washed and incubated with fresh L-15 medium. The CPE was recorded at 24 hpi, and the titers of progeny RSIV at 7 dpi were determined. For the progeny virus titration, the cells and supernatants were harvested and subjected to 3 freeze/thaw cycles. The progeny NNV was titrated in GF-1 cells, and the progeny RSIV was titrated in cBB cells. GF-1 cells are sensitive to NNV only, and cBB cells are sensitive to both viruses. Because the progeny RSIV derived from NNV/ RSIV co-infected cBB cells and RSIV-infected BB cells contained both NNV and RSIV, the viral solutions were pre-treated with rabbit anti-NNV serum for 1 h before RSIV titration. The volume ratio of viral solution and rabbit anti-NNV serum was 100:1. Complete neutralization of NNV was confirmed by inoculating the viral solution into GF-1 cells, and no CPE occurred. 2.6. Western blot The cBB and BB cells pre-seeded in 30 mm dishes (2  105 cells per dish) were transfected with poly I:C (1 lg ml1) or infected with NVV or RSIV (MOI = 10). After 24 h, the cells were harvested to analyze the BMx protein expression by Western blot. The total protein extraction and Western blot was performed following the procedures described by Wu et al. (2010). Briefly, the cells were washed with PBS and lysed with 40 ll of lysis buffer containing 1% NP-40, 50 mM Tris (pH 8.0), 1 mM DTT and 1 protease inhibitor (Roche) for 30 min on ice. Following centrifugation at 12,000g at 4 °C for 15 min to discard debris, each cell lysate with 10 lg total protein was applied for 10% SDS–polyacrylamide gel electrophoresis (SDS–PAGE), and the proteins were electrophoretically transferred to a PVDF membrane (Millipore). After blocking with 5% skim milk in TBS buffer containing 0.5% Tween 20 for 1 h at room temperature, the membrane was reacted with rabbit anti-BMx antibodies and then with peroxidase-labeled goat anti-rabbit IgG (KPL). The signals were developed by Amersham ECL Plus Western Blotting Detection Reagent (GE) and visualized through Multifunction Gel Image System MultiGel-21 (TOPBIO). The internal control, actin, was reacted with anti-actin mAb (CHEMICON) and then with alkaline-phosphatase (AP)-conjugated goat anti-mouse IgG antibody (KPL). The protein bands were visualized using BCIP/NBT substrate Kit (KPL). 2.7. Down-regulation of BMx gene expression by siRNA The siRNAs were designed and purchased from GenePharma. The sense sequences of the BMx specific siRNA (Mx-siRNA) and the non-silencing control siRNA (N-siRNA) are listed in Table 1.

Y.-C. Wu et al. / Developmental and Comparative Immunology 41 (2013) 666–674

669

The cBB cells pre-seeded in the 4-well plate (4  104 cells per well) were transfected with Mx- or N-siRNA (100 nM). Four hours later, the cells were further transfected with poly I:C (25 ng ml1). After 14 h incubation, the cells were harvested to analyze the BMx gene expression by real-time PCR or infected with RSIV (MOI = 10). After 1 h virus adsorption, the cells were washed with L-15 medium for 3 times and incubated with fresh L-15 medium. The titers of progeny RSIV at 7 dpi were titrated in cBB cells. 3. Results 3.1. Pathogenicity of NNV and RSIV to barramundi Naïve 80 days post-hatch (dph) barramundi challenged with six doses of NNV (104–9 TCID50 per fish) showed no mortality within 1 month. The fish pre-challenged with NNV at dose of 108 TCID50 were collected at 14 days post challenge (dpc) and further challenged with RSIV (106 and 104 TCID50 per fish). For the 106 TCID50 RSIV challenge, the fish pre-challenged with PBS or NNV all died (Fig. 1A). For the 104 TCID50 RSIV challenge, the barramundi with PBS pre-challenge began dying 5 days post RSIV challenge, and the accumulated mortality was 100% (Fig. 1A). However, the NNV-RSIV-104 TCID50 dual-challenged fish began dying 7 days post RSIV challenge, and the accumulated mortality was 62% (Fig. 1A). These results indicate that the pre-challenge with NNV delayed the occurrence of death and reduced the mortality of the RSIVchallenged barramundi. Because no mortality occurred in the 80 dph barramundi challenged with NNV, the pathogenicity of the NNV strain was examined in the 14 dph barramundi larvae. The larvae were too small to be intramuscularly injected with NNV; thus, they were challenged with NNV by bath or feeding. The 14 dph barramundi were fed with artemia pre-mixed with NNV at doses of 109, 108, and 107 TCID50, and the accumulated mortality was 93%, 78%, and 79% for the high, medium, and low doses of NNV challenge (Fig. 1B), respectively. However, the bath challenge with NNV at doses of 109, 108, and 107 TCID50 did not induce mortality in the 14 dph barramundi. The accumulated mortality of the barramundi challenged with NNV or RSIV is summarized in Table 2. 3.2. NNV Existence and BMx gene expression in NNV-challenged barramundi All 80 dph barramundi challenged with 108 TCID50 NNV were alive, and 5 fish of them were collected at 14 dpc for analysis of NNV presence and BMx gene expression in the brain and liver using real-time PCR. Near 109 copies of NNV RNA2 were detected in the brain (1.5  109 copies) and liver (5.1  109 copies) of NNV-challenged fish, but no NNV RNA2 was detected in the same organs of the control fish challenged with PBS (Fig. 2A). Moreover, the BMx gene expression levels in the brain and liver of NNV-challenged barramundi were 26- and 202-fold higher than those of the control fish (Fig. 2B), respectively. Therefore, the 80 dph barramundi challenged with NNV obtain persistent NNV infections that induce BMx gene expression in the brain and liver. Besides, five fish of 14 dph barramundi challenged with NNV by bath (108 TCID50 NNV) or feeding (artemia pre-mixed with 108 TCID50 NNV) were sampled at 7 dpc for analyzing NNV presence and BMx gene expression in whole fish body. The results showed that no NNV RNA2 was detected in the bath-challenged barramundi, and no mortality occurred (Table 3). The BMx gene expression level of bath-challenged barramundi was the same as that of non-challenged fish, and the level was regarded as 1 (Table 3). In the oral challenge with NNV, near 1012 copies of

Fig. 1. The accumulated mortality of virus-infected barramundi. (A) The 80 days post hatch (dph) barramundi pre-challenged with NNV (108 TCID50 per fish) or PBS were challenged with two doses of RSIV (106 and 104 TCID50 per fish). (B) The 14 dph barramundi larvae were fed with artemia pre-mixed with three doses of NNV (109, 108 and 107 TCID50). ⁄P < 0.001 (Chi-square).

NNV RNA2 were detected in the whole fish body which was induced 213-fold BMx gene expression level compare to that of non-challenged fish, and then mortality occurred (Table 3). These results revealed that once NNV invaded into the barramundi and replicated inside the host, certain level of BMx gene expression could be induced. The BMx gene expression levels and NNV RNA2 copies in NNV-challenged barramundi were summarized in Table 3. 3.3. Cross-reaction of anti-NNV polyclonal antibodies to RSIV The barramundi were immunized with inactivated NNV (106 and 105 TCID50 per fish) for inducing NNV-specific antibodies to clarify whether NNV-specific antibodies cross-protect against RSIV challenge. The BMx gene did not express in the brain and liver of the barramundi at 2 months post immunization (data not shown), and the immunized fish were challenged with RSIV, resulting in 100% accumulated mortality (Fig. 3A). Furthermore, the sera of the immunized barramundi exhibited an average neutralizing antibody titer of 28 ND50 against NNV, but no neutralizing antibody titer (<22 ND50) against RSIV was detected (Fig. 3B). These results indicate that the antibody immune response induced by NNV does not protect barramundi against the RSIV challenge. Moreover, the cross-neutralizations of rabbit anti-NNV polyclonal antibodies against RSIV were analyzed. The results show that the neutralization index (NI) of a 100-fold diluted rabbit anti-NNV serum against NNV was greater than 109, but the same serum failed to neutralize RSIV (NI = 100–0.5).

670

Y.-C. Wu et al. / Developmental and Comparative Immunology 41 (2013) 666–674

Table 2 The accumulated mortality of barramundi challenged with NNV and RSIV. Days post-hatch Fish body length

14 dph 0.6 cm

80 dph 4.3 cm

Virus

NNV

NNV

Challenge mode

Bath

Oral

Challenge doses

109–107

109

108

107

0

93

78

79

Accumulated mortality (%) a

8

6

NNVa-RSIV

PBS-RSIV

IM

IM-IP

109–104

106

104

106

IM-IP 104

0

100

100

100

62

4

NNV-challenged fish (10 TCID50 per fish) were challenged with two doses of RSIV (10 and 10 TCID50 per fish).

Fig. 2. The NNV load and BMx gene expression of NNV-challenged barramundi at 14 dpc. (A) The copies of NNV RNA2 per organ (brain or liver) were measured by real-time PCR. N.D. indicated no NNV RNA2 could be detected. (B) The BMx gene expression in the brain and liver was analyzed by real-time PCR. Bars represent standard deviations. Column bars with different letters are significantly different at P < 0.05 (N = 5).

Table 3 BMx gene expression levels and NNV RNA2 copies in NNV-challenged barramundi. Days post-hatch Fish body length

14 dph 0.6 cm

80 dph 4.3 cm

Challenge mode

Bath

Oral

IM

Challenge doses

108

108

108

Accumulated mortality (%)

0

78

0

Copies of NNV RNA2 in brain (B), liver (L), or whole fish body (W)

W: ND

W: 9.00 ± 0.08  1012

BMx gene expression level in brain (B), liver (L), or whole fish body (W)

W: 1

W: 213.9 ± 14.7

Time point for real-time PCR analysis

7 dpc

7 dpc

B: 1.45 ± 0.77  109 L: 5.06 ± 0.76  109 B: 26.0 ± 3.7 L: 202.1 ± 99.3 14 dpc

ND: No NNV RNA2 could be detected.

3.4. Replication of RSIV in cBB and BB cells The cBB and BB cells are in vitro systems for mimicking NNVfree and persistently NNV-infected barramundi. Following RSIV infection with 10, 1, and 0.1 MOI, the average titers of progeny RSIV in cBB cells at 7 dpi were 106.3, 105.8, and 105.4 TCID50 ml1, respectively. Those in BB cells were 105.4, 105.2, and 104.2 TCID50 ml1 (Fig. 4), respectively. These results show that the efficiency of RSIV proliferation in BB cells is lower compared to that in cBB cells. 3.5. Influence of NNV and RSIV co-infection on respective viral proliferation To discover whether NNV infection interferes with RSIV replication, cBB cells were infected with either NNV or RSIV, or co-in-

fected with NNV and RSIV. The progeny viruses at 24 hpi were collected for viral titration. The titers of the progeny RSIV in RSIV sole-infected cBB cells were the same as that of NNV-RSIV co-infected cBB cells (104.9 TCID50 ml1) (Fig. 5A). However, the titer of the progeny NNV in co-infected cBB cells was 107.2 TCID50 ml1, which was lower than that in NNV sole-infected cBB cells (108.4 TCID50 ml1) (Fig. 5B). 3.6. Expression of BMx protein in cBB and BB cells after viral infection and poly I:C transfection The cBB and BB cells were separately infected with NNV and RSIV (MOI = 10), or transfected with poly I:C (1 lg ml1). The BMx protein expression in cBB cells was induced by both virus infection and poly I:C transfection (Fig. 6). However, the expression level of the BMx protein in RSIV-infected cBB cells was higher

Y.-C. Wu et al. / Developmental and Comparative Immunology 41 (2013) 666–674

671

Fig. 3. The cross-protection test of specific immune response induced by inactivated NNV vaccine against RSIV. (A) The accumulated mortality of barramundi pre-immunized with inactivated NNV vaccine and then challenged with RSIV. The naïve barramundi were first immunized with two doses of inactivated NNV (106 and 105 TCID50 per fish), and then challenged with RSIV (104 TCID50 per fish) at 2 months post immunization. (B) The sera of the barramundi were collected at 2 months post immunization, and the neutralizing antibody titers of the sera against NNV and RSIV were detected. Bars represent standard deviations (N = 3). N.D. indicated no neutralizing antibody titer could be detected.

Fig. 6. The expression of BMx protein in cBB and BB cells which were infected with virus or transfected with poly I:C.

3.7. Antiviral activity of IFN response and BMx against RSIV Fig. 4. The proliferation of RSIV in the cBB and BB cells. The cBB and BB cells were respectively infected with three doses of RSIV (MOI = 10, 1, 0.1), and the progeny RSIV at 7 dpi was titrated. Bars represent standard deviations. ⁄P < 0.05 (N = 3).

Fig. 5. The effect of co-infection on the titers of progeny NNV and RSIV. The cBB cells were sole- or co-infected with NNV and RSIV. The progeny RSIV (A) and NNV (B) were harvested at 24 hpi, and titrated on cBB and GF-1 cells, respectively. Bars represent standard deviations. ⁄⁄P < 0.01 (N = 3).

compared to that in the NNV-infected cBB cells. The BB cells with the persistent NNV infection always expressed the BMx protein. Although BB cells were further infected with NNV or RSIV, or transfected with poly I:C, the BMx protein expressed in the same manner as the mock-treated BB cells (Fig. 6).

The cBB cells were transfected with poly I:C to induce the IFN response and were subsequently infected with RSIV. The numbers of cells with CPE (90%) in mock-transfected cBB cells were higher than those (10%) in poly I:C-transfected cBB cells (Fig. 7A). The progeny RSIV titer at 7 dpi in mock-transfected cBB cells was 108.7 TCID50 ml1, but decreased to 107 TCID50 ml1 in poly I:Ctransfected cBB cells (Fig. 7B), indicating that the poly I:C-induced IFN response exhibited antiviral activity against RSIV replication. Furthermore, a BMx specific siRNA (Mx-siRNA) was designed to down-regulate BMx gene expression. The cBB cells were transfected with Mx-siRNA or non-silencing siRNA (N-siRNA) prior to poly I:C-transfection for the induction of the BMx gene expression. The expression level of the BMx gene in N-siRNA + poly I:C-transfected cBB cells was regarded as 100%, but approximately 82% of the BMx gene expression was down-regulated by the Mx-siRNA (Fig. 8A). Therefore, the cBB cells treated as those shown in Fig. 8A were infected with RSIV, and the progeny RSIV at 7 dpi was titrated. The progeny RSIV titer (106.3 TCID50 ml1) in the Mx-siRNA + poly I:Ctransfected cBB cells was higher than that in the N-siRNA + poly I:C-transfected cBB cells (105.5 TCID50 ml1) (Fig. 8B), indicating that BMx may suppress RSIV proliferation.

4. Discussion VNN disease frequently caused high mortality in larval and juvenile marine fish (Munday et al., 2002). The mortality of affected barramundi was high at 15–35 dph, but became moderate beyond 50 dph (Ransangan and Manin, 2010). In this study, high mortality was observed in the 14 dph barramundi larvae fed with

672

Y.-C. Wu et al. / Developmental and Comparative Immunology 41 (2013) 666–674

Fig. 7. Poly I:C-transfected cBB cells exhibited the antiviral activity against RSIV. The mock- and poly I:C-transfected cBB cells were infected with RSIV. (A) The cytopathic effects were observed at 24 hpi. The arrows indicate the CPE. Bar = 100 lm. (B) The progeny RSIV at 7 dpi was titrated in cBB cells. Bars represent standard deviations. ⁄⁄ P < 0.01 (N = 3).

Fig. 8. The antiviral activity of BMx against RSIV. (A) The cBB cells were transfected with Mx-siRNA or N-siRNA, and then further transfected with poly I:C. After 14 h incubation, BMx gene expression was analyzed by real-time PCR. The level of BMx gene expression in N-siRNA + poly I:C-transfected cBB cells was regarded as 100%. (B) The cBB cells with different levels of BMx gene expression, showed in Fig. 8(A), were infected with RSIV (MOI = 10). The progeny RSIV at 7 dpi was titrated. Bars represent standard deviations. Repeated experiments which have the same results lack the standard deviations. Column bars with different letters are significantly different at P < 0.05 (N = 3).

artemia pre-mixed with NNV, but no mortality was induced by the injection of NNV in 80 dph barramundi (Table 2), indicating that older barramundi were more resistant to NNV infection. Moreover, the oral challenge with NNV caused mortality in the 14 dph barramundi, but the bath challenge did not (Table 2). This finding was different with the groupers because their mortality was induced through the bath challenge with NNV (Kai and Chi, 2008). Furthermore, this demonstrated that oral transmission is the major route for NNV infection in 14 dph barramundi larvae. Conversely, the RSIV was a high pathogenic virus for the barramundi and caused

100% accumulated mortality (Table 2). The barramundi pre-challenged with NNV had the potential to survive the RSIV challenge (Table 2). Therefore, the key factor that contributed to the protection from RSIV was identified in this study. The barramundi challenged with NNV became NNV carriers, and the NNV replication that simultaneously induced BMx gene expression was detected in the brain and liver (Fig. 2). Both persistently NNV-infected barramundi and BB cells were more resistant to RSIV infection compared to NNV-free barramundi and cBB cells (Figs. 1 and 4). The resistance to RSIV infection was not related to

Y.-C. Wu et al. / Developmental and Comparative Immunology 41 (2013) 666–674

the NNV-induced specific immunity, because the barramundi immunized with inactivated NNV all expired after the RSIV challenge (Fig. 3A). Moreover, the barramundi anti-NNV sera did not neutralize the RSIV (Fig. 3B). Another speculation posits that NNV may interfere with RSIV replication. However, this possibility was excluded because the titer of progeny RSIV in sole-infected cBB cells was nearly equal to that of the NNV-RSIV co-infected cBB cells (Fig. 5A), indicating that NNV does not interfere with the RSIV replication. Conversely, RSIV replication can suppress NNV replication during co-infection (Fig. 5B). How RSIV interferes with NNV replication requires further examination. Is the IFN response related to the NNV-induced protection against RSIV? Our results showed that the 80 dph barramundi challenged with NNV expressed the BMx gene in the liver (Fig. 2B), one of the target organs for RSIV replication. When NNV-challenged fish were further challenged with RSIV, the accumulated mortality reduced to 62% (Fig. 1A). These results suggest that NNV-induced IFN response in the liver may be responsible for the resistance to RSIV. Moreover, persistently NNV-infected BB cells expressed the BMx protein (Fig. 6), and the titers of the progeny RSIV in BB cells were lower than those in the cBB cells after RSIV infection (Fig. 4). Furthermore, poly I:C-transfected cBB cells with BMx protein expression showed antiviral activity against RSIV (Figs. 6 and 7). When BMx gene expression was down-regulated by Mx-siRNA, RSIV proliferation increased (Fig. 8). These data suggest that the IFN response and BMx expression induced by NNV have a crucial role in antagonizing RSIV replication. Numerous studies indicated that fish with a primary infection of one virus can protect the fish from a secondary infection by a heterologous virus. For example, the rainbow trout pre-challenged with cutthroat trout virus (CTV), aquareovirus, or infectious pancreatic necrosis virus (IPNV) acquired resistance to infectious hematopoietic necrosis virus (IHNV) (Byrne et al., 2008; Hedrick et al., 1994; Kim et al., 2009; LaPatra et al., 1995). Protection of Atlantic salmon from infectious salmon anaemia virus (ISAV) is elicited from the pre-exposure of fish to IPNV (Johansen and Sommer, 2001). Moreover, a primary challenge of Japanese flounder and sevenband grouper by aquabirnavirus protected the fishes from viral hemorrhagic septicemia virus (VHSV) and NNV infection, respectively (Pakingking et al., 2003, 2005). The protection of the Japanese flounder against VHSV was the result of an IFN response induced by the aquabirnavirus (Pakingking et al., 2004). Virus infection may elicit the IFN response and Mx gene expression. NNV RNA2 was detected in numerous organs during the late stage of acute infection, including the brain, eyes, liver, pyloric gland, stomach, intestines, and blood cells (Chi et al., 2001). Moreover, Mx gene expression was detected in numerous organs of the NNV-infected grouper and sea bream, including the brain, liver, spleen, and kidneys (Chen et al., 2006; Poisa-Beiro et al., 2008). This indicates that the systemic infection of NNV stimulates an IFN response in not only the brain but also the target organs of RSIV. In this study, BMx gene expression occurred in the brain and liver of barramundi at 14 days post NNV challenge (Fig. 2B), which suggests that the NNV-induced systemic IFN response may restrict the RSIV replication. Both NNV and RSIV can induce BMx protein expression in cBB cells. Studies show that the BMx protein interacts with NNV RdRp and interferes with NNV replication (Wu et al., 2010). This study initially demonstrated the antagonism of BMx against RSIV, and the molecular mechanism of BMx against RSIV will be examined in the future. Acknowledgements We deeply thank Schweitzer Chemical Corporation for providing the SK cell line. This work was partially supported by National

673

Science Council of the Republic of China under the contract No. NSC 99-2313-B-002-024-MY3.

References Byrne, N., Castric, J., Lamour, F., Cabon, J., Quentel, C., 2008. Study of the viral interference between infectious pancreatic necrosis virus (IPNV) and infectious haematopoietic necrosis virus (IHNV) in rainbow trout (Oncorhynchus mykiss). Fish Shellfish Immunol. 24, 489–497. Chao, C.B., Pang, V.F., 1997. An outbreak of an iridovirus-like infection in cultured grouper (Epinephelus spp.) in Taiwan. J. Chin. Soc. Vet. Sci. 23, 411–422. Chao, C.B., Yang, S.C., Tsai, H.Y., Chen, C.Y., Lin, C.S., Huang, H.T., 2002. A nested PCR for the detection of grouper iridovirus in Taiwan (TGIV) in cultured hybrid grouper, giant seaperch, and largemouth bass. J. Aquat. Anim. Health 14, 104– 113. Chen, Y.M., Su, Y.L., Lin, J.H., Yang, H.L., Chen, T.Y., 2006. Cloning of an orangespotted grouper (Epinephelus coioides) Mx cDNA and characterisation of its expression in response to nodavirus. Fish Shellfish Immunol. 20, 58–71. Chi, S.C., Hu, W.W., Lo, B.J., 1999. Establishment and characterization of a continuous cell line (GF-1) derived from grouper, Epinephelus coioides (Hamilton): a cell line susceptible to grouper nervous necrosis virus (GNNV). J. Fish Dis. 22, 173–182. Chi, S.C., Lo, B.J., Lin, S.C., 2001. Characterization of grouper nervous necrosis virus (GNNV). J. Fish Dis. 24, 3–13. Chi, S.C., Lo, C.F., Kou, G.H., Chang, P.S., Peng, S.E., Chen, S.N., 1997. Mass mortalities associated viral nervous necrosis disease in two species of hatchery-reared grouper, Epinephelus fuscogutatus and Epinephelus akaara (temminck & Schlegel). J. Fish Dis. 20, 185–193. Chi, S.C., Shieh, J.R., Lin, S.J., 2003. Genetic and antigenic analysis of betanodaviruses isolated from aquatic organisms in Taiwan. Dis. Aquat. Org. 55, 221–228. Chi, S.C., Wu, Y.C., Cheng, T.M., 2005. Persistent infection of betanodavirus in a novel cell line derived from the brain tissue of barramundi Lates calcarifer. Dis. Aquat. Org. 65, 91–98. Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction. Anal. Biochem. 162, 156–159. Chou, H.Y., Hsu, C.C., Peng, T.Y., 1998. Isolation and characterization of a pathogenic iridovirus from cultured grouper (Epinephelus sp.) in Taiwan. Fish Pathol. 33, 201–206. Hedrick, R.P., LaPatra, S.E., Yun, S., Lauda, K.A., Jones, G.R., Congleton, J.L., de Kinkelin, P., 1994. Induction of protection from infectious hematopoietic necrosis virus in rainbow trout Oncorhynchus mykiss by pre-exposure to the avirulent cutthroat trout virus (CTV). Dis. Aquat. Org. 20, 111–118. Huang, S.M., Tu, C., Tseng, C.H., Huang, C.C., Chou, C.C., Kuo, H.C., Chang, S.K., 2011. Genetic analysis of fish iridoviruses isolated in Taiwan during 2001–2009. Arch. Virol. 156, 1505–1515. Inouye, K., Yamano, K., Maeno, Y., Nakajima, K., Matsuoko, M., Wada, Y., Sorimachi, M., 1992. Iridovirus infection of cultured red sea bream, Pagrus major. Fish Pathol. 27, 19–27. Johansen, L.H., Sommer, A.I., 2001. Infectious pancreatic necrosis virus infection in Atlantic salmon Salmo salar post-smolts affects the outcome of secondary infections with infectious salmon anaemia virus or Vibrio salmonicida. Dis. Aquat. Org. 47, 109–117. Johansen, R., Amundsen, M., Dannevig, B.H., Sommer, A.I., 2003. Acute and persistent experimental nodavirus infection in spotted wolffish Anarhichas minor. Dis. Aquat. Org. 57, 35–41. Kai, Y.H., Chi, S.C., 2008. Efficacies of inactivated vaccines against betanodavirus in grouper larvae (Epinephelus coioides) by bath immunization. Vaccine 26, 1450– 1457. Kim, H.J., Oseko, N., Nishizawa, T., Yoshimizu, M., 2009. Protection of rainbow trout from infectious hematopoietic necrosis (IHN) by injection of infectious pancreatic necrosis virus (IPNV) or poly(I:C). Dis. Aquat. Org. 83, 105–113. Kurita, J., Nakajima, K., Hirono, I., Aoki, T., 1998. Polymerase chain reaction (PCR) amplification of DNA of red sea bream iridovirus (RSIV). Fish Pathol. 33, 17–23. LaPatra, S.E., Lauda, K.A., Jones, G.R., 1995. Aquareovirus interference mediated resistance to infectious hematopoietic necrosis virus. Vet. Res. 26, 455–459. Larsen, R., Rokenes, T.P., Robertsen, B., 2004. Inhibition of infectious pancreatic necrosis virus replication by atlantic salmon Mx1 protein. J. Virol. 78, 7938– 7944. Mahy, B.W.J., Kangro, H.O., 1996. Virology Method Manual. Academic Press, London. Mori, K., Nakai, T., Muroga, K., Arimoto, M., Mushiake, K., Furusawa, I., 1992. Properties of a new virus belonging to nodaviridae found in larval striped jack (Pseudocaranx dentex) with nervous necrosis. Virology 187, 368–371. Munday, B.L., Kwang, J., Moody, N., 2002. Betanodavirus infection of teleost fish: a review. J. Fish Dis. 25, 127–142. Nishizawa, T., Mori, K., Nakai, T., Furusawa, I., Muroga, K., 1994. Polymerase chain reaction amplification of RNA of striped jack nervous necrosis virus (SJNNV). Dis. Aquat. Org. 18, 103–107. Pakingking, R.J., Mori, K., Sugaya, T., Oka, M., Okinaka, Y., Nakai, T., 2005. Aquabirnavirus-induced protection of marine fish against piscine nodavirus infection. Fish Pathol. 40, 125–131. Pakingking, R.J., Okinaka, Y., Mori, K., Arimoto, M., Muroga, K., Nakai, T., 2004. In vivo and in vitro analysis of the resistance against viral haemorrhagic

674

Y.-C. Wu et al. / Developmental and Comparative Immunology 41 (2013) 666–674

septicaemia virus in Japanese flounder (Paralichthys olivaceus) precedingly infected with aquabirnavirus. Fish Shellfish Immunol. 17, 1–11. Pakingking, R.J., Takano, R., Nishizawa, T., Mori, K., Iida, Y., Arimoto, M., Muroga, K., 2003. Experimental coinfection with aquabirnavirus and viral hemorrhagic septicemia virus (VHSV), Edwardsiella tarda or Streptococcus iniae in Japanese flounder Paralichthys olivaceus. Fish Pathol. 38, 15–21. Poisa-Beiro, L., Dios, S., Montes, A., Aranguren, R., Figueras, A., Novoa, B., 2008. Nodavirus increases the expression of Mx and inflammatory cytokines in fish brain. Mol. Immunol. 45, 218–225. Ransangan, J., Manin, B.O., 2010. Mass mortality of hatchery-produced larvae of Asian seabass, Lates calcarifer (Bloch), associated with viral nervous necrosis in Sabah, Malaysia. Vet. Microbiol. 145, 153–157. Samuel, C.E., 2001. Antiviral action of interferons. Clin. Microbiol. Rev. 14, 778–809. Trobridge, G.D., Leong, J.A., 1995. Characterization of a rainbow trout Mx gene. J. Interferon Cytokine Res. 15, 691–702.

Wang, C.S., Chao, S.Y., Ku, C.C., Wen, C.M., Shih, H.H., 2009. PCR amplification and sequence analysis of the major capsid protein gene of megalocytiviruses isolated in Taiwan. J. Fish Dis. 32, 543–550. Wang, C.S., Shih, H.H., Ku, C.C., Chen, S.N., 2003. Studies on epizootic iridovirus infection among red sea bream, Pagrus major (Temminck & Schlegel), cultured in Taiwan. J. Fish Dis. 26, 127–133. Williams, T., Barbosa-Solomieu, V., Chinchar, V.G., 2005. A decade of advances in iridovirus research. Adv. Virus Res. 65, 173–248. Wu, Y.C., Chi, S.C., 2006. Persistence of betanodavirus in Barramundi brain (BB) cell line involves the induction of Interferon response. Fish Shellfish Immunol. 21, 540–547. Wu, Y.C., Lu, Y.F., Chi, S.C., 2010. Anti-viral mechanism of barramundi Mx against betanodavirus involves the inhibition of viral RNA synthesis through the interference of RdRp. Fish Shellfish Immunol. 28, 467–475.