Spring Viremia of Carp Virus N Protein Suppresses Fish IFNφ1 Production by Targeting the Mitochondrial Antiviral Signaling Protein

This article is featured in In This Issue, p.3497

The IFN system is the first line of defense of the host against viral invasion. Once host cells are infected by virus, the retinoic acid–inducible gene I (RIG-I)–like receptors (RLRs) sense the viral RNAs and activate IFN expression (1–4). In this process, a member of the RLR family recognizes viral RNAs and activates a downstream adaptor molecule, the mitochondrial antiviral signaling protein (MAVS, also known as VISA/IPS-1/Cardif), which initiates TANK-binding kinase 1 (TBK1) and canonical IKK complex (IKKα/β/γ) to activate IFN regulatory factor 3/7 (IRF3/7) and NF-κB, which translocate into the nucleus and initiate the expression of IFN (5–10). This sets up the antiviral status of the cells.

However, viruses have evolved a number of elaborate strategies to avoid being eliminated. One of those mechanisms is to block the production of IFN (11). In mammals, MAVS seems to be an important target for virus. Generally, two mechanisms are used to inhibit the expression of IFN by targeting MAVS, including cleavage and degradation. As examples, MAVS can be cleaved by the NS3/4A serine protease of hepatitis C virus and the cysteine protease 3C^pro of coxsackievirus B3 (12, 13), or degraded by the hepatitis B virus X and open reading frame (ORF)-9b of coronaviruses in an ubiquitin (Ub)–proteasome manner (14, 15).

In aquaculture, a number of viruses have been identified as the etiological agents of aquatic animal diseases (16), however, the mechanism(s) used by aquatic viruses to escape the host IFN system are still unclear. Spring viremia of carp virus (SVCV) is the causative agent of SVC and causes significant mortality in common carp (Cyprinus carpio). It is a negative-sense ssRNA virus that belongs to the genus Vesiculovirus of the family Rhabdoviridae (17). The genome of SVCV is ∼11 kb and encodes five proteins including the nucleoprotein (N), phosphoprotein (P), matrix protein (M), glycoprotein (G), and viral RNA–dependent RNA polymerase (L) (18). The availability of sequence information for the entire SVCV genome and homology analysis with mammalian rhabdoviruses allowed more accurate deduction of the functions of the proteins of SVCV. The N protein interacts with the viral RNA to form the nucleocapsid. The P and L proteins associate with the nucleocapsid, which are required for the transcription and replication of SVCV. The M protein takes part in the assembly and budding of SVCV, and the G protein binds to cellular receptors and mediates viral endocytosis (17, 19).

There is some information available concerning the host response to an SVCV infection (20, 21). For example, by using high-throughput methods such as pathway-targeted microarrays and transcriptomic and proteomic analyses, many immune- and autophagy-related genes have been identified to be involved in an SVCV infection (22–25). The expression of IFN could be elicited by an SVCV infection, and overexpression of IFN could significantly increase the survival of zebrafish embryos infected with SVCV (26–28). Similar to mammals, it has been shown that zebrafish MAVS plays an important role in IFN activation, which can significantly decrease the probability of infection with SVCV (29). In the RLR pathway, the truncated splicing variant of MAVS blocked the IFN induction by RIG-I, indicating the position of MAVS in the signaling pathway is conserved (30, 31). These results demonstrate that the defense mechanisms of fish against SVCV infection are similar to that of mammals, especially regarding IFN production being activated by MAVS.

Until now, there have been few studies regarding the mechanisms used by aquatic viruses to interfere with fish IFN production. In study reported herein we show that the N protein of SVCV could inhibit host IFN expression by degrading MAVS in a Ub–proteasome manner. These results suggest that the immune evasion by an aquatic virus by interfering with the host IFN response also exists in lower vertebrates.

Epithelioma papulosum cyprini (EPC) cells were maintained in medium 199 (Invitrogen) supplemented with 10% FBS (Invitrogen). Zebrafish liver (ZFL) cells were cultured in 50% L-15 (Invitrogen), 35% DMEM-HG (Invitrogen), and 15% Ham’s F12 medium (Invitrogen) supplemented with 0.15 g/l sodium bicarbonate (Sigma-Aldrich), 15 mM HEPES (Sigma-Aldrich), and 10% FBS. Zebrafish embryo fibroblast-like (ZF4) cells were grown in a 1:1 mixture of DMEM and Ham’s F12 medium supplemented with 10% FBS. All cells were maintained at 28°C and 5% CO₂. SVCV was propagated in EPC cells until cytopathic effect (CPE) was complete, and the culture medium was harvested and stored at −80°C until needed.

The plasmids encoding RIG-I-Nter, TBK1, and IRF7 in the pcDNA3.1(+)vector (Invitrogen) and the plasmids containing IFNφ1pro-Luc, IFNφ2pro-Luc, IFNφ3pro-Luc, IFNφ4pro-Luc, and NF-κB-Luc in the pGL3-Basic luciferase reporter vector (Promega) were constructed as described previously (31–33). The IFN-stimulated response element (ISRE) luciferase reporter construct (ISRE-Luc) containing five ISRE motifs in series was purchased from Stratagene. The ORF of MAVS was amplified by PCR from ZF4 cells and cloned into the pcDNA3.1(+) and pHAGE-CMV-MCS-PGK (BD Clontech) vectors. The pcDNA3.1-HA-IRF7 plasmid was constructed by subcloning the DNA fragment encoding HA-IRF7 into the pcDNA3.1(+) vector. To generate the MAVS-mCherry, TBK1-mCherry, and IRF7-mCherry expression plasmids, the cDNA fragments encoding MAVS, TBK1, and IRF7 were cloned into the pCS2-mCherry vector (BD Clontech). The MAVS_tv2-Flag plasmid was generated by inserting MAVS_tv2 cDNA into the pHAGE-CMV-MCS-PGK vector. The cDNA fragments of N, P, and G genes were amplified by RT-PCR from the RNA of SVCV infected cells and then cloned into the pcDNA3.1(+) vector. C-terminally GFP-tagged N (N-GFP) was generated by inserting the ORF of the N protein of SVCV into the pEGFP-N3 vector (BD Clontech). The pcDNA3.1-HA-N, pcDNA3.1-N-Flag and pcDNA3.1-HA-P plasmids were constructed by inserting the DNA fragments encoding HA-N, N-Flag and HA-P (P protein of SVCV) into the pcDNA3.1(+) vector. The cDNA fragment of N was also subcloned into the pCMV-Myc vector (BD Clontech). All constructs were confirmed by DNA sequencing. The primers including the restriction enzyme cutting sites used for plasmid construction are listed in Supplemental Table I. MG132 and polyinosinic:polycytidylic acid [poly(I:C)] were purchased from Sigma-Aldrich and used at a final concentration of 20 μM/ml and 1 μg/ml, respectively.

EPC cells were seeded in 10-cm² dishes or 24-well plates and transfected with various plasmids by using X-tremeGENE HP DNA Transfection Reagent (Roche), according to the manufacturer’s instructions. For the antiviral assay, EPC cells seeded in 24-well plates were transfected with 0.5 μg pcDNA3.1-N or the empty vector. At 24 h posttransfection, cells were infected with SVCV at a multiplicity of infection (MOI) of 1, 10, 100, or 1000. After 2 or 3 d, aliquots of the supernatant were harvested for determination of virus titers, and cell monolayers were stained with 1% crystal violet for visualizing CPE. For virus titration, 200 μl culture medium was collected at 48 h postinfection and used for plaque assay. The supernatants were subjected to 3-fold serial dilutions and then added (100 μl) onto a monolayer of EPC cells cultured in a 96-well plate. After 48 or 72 h, the medium was removed, and the monolayers were washed with PBS, fixed by 4% paraformaldehyde (PFA), and stained with 1% crystal violet. The virus titer was expressed as 50% tissue culture infectious dose (TCID₅₀/ml).

EPC cells were seeded in 24-well plates and 24 h later cotransfected with 250 ng of the luciferase reporter plasmid (IFNφ1pro-Luc, ISRE-Luc or NF-κB-Luc) and 25 ng of the pRL-TK vector (Promega). The Renilla luciferase internal control was used to normalize the expression levels of the transefected plasmids. The empty vector pcDNA3.1(+) was used to maintain equivalent amounts of DNA in each well. The transfection of poly(I:C) was performed at 24 h before the cells were harvested. At 48 h posttransfection, the cells were washed in PBS and lysed for measuring luciferase activity by Dual-Luciferase Reporter Assay System according to the manufacturer’s instructions (Promega). The results were obtained from more than three independent experiments, each performed in triplicate.

Total RNAs were extracted using the TRIzol reagent (Invitrogen). cDNA was synthesized using a GoScript reverse transcription system (Promega), according to the manufacturer’s instructions. Quantitative real-time PCR (qPCR) was performed with Fast SYBR Green PCR Master Mix (Bio-Rad) on the CFX96 Real-Time System (Bio-Rad). PCR conditions were as follows: 95°C for 5 min and then 40 cycles of 95°C for 20 s, 60°C for 20 s, and 72°C for 20 s. All primers used for qPCRs are shown in Supplemental Table I, and β-actin was used as an internal control. The relative fold changes were calculated by comparison with the corresponding controls using the 2^−ΔΔCt method. Three independent experiments were conducted for statistical analysis purposes.

Whole-cell extracts were separated by 10% SDS-PAGE and transferred to polyvinylidene difluoride membranes (Bio-Rad). The membranes were blocked for 1 h at room temperature in TBST buffer (25 mM Tris-HCl, 150 mM NaCl, and 0.1% Tween 20 [pH 7.5]) containing 5% nonfat dry milk and probed with various primary Abs at an appropriate dilution overnight at 4°C, washed three times with TBST, and then incubated with secondary Abs for 1 h at room temperature. After three additional washes with TBST, the membranes were stained with Immobilon TM Western Chemiluminescent HRP Substrate (Millipore) and detected using an ImageQuant LAS 4000 system (GE Healthcare). Rabbit polyclonal anti-MAVS and anti-IRF7 antisera were made by immunization of rabbits with prokaryotic expressed zebrafish MAVS_tv2 and IRF7-DBD as described in a previous report (31). The antiserum against MAVS_tv2 or IRF7 was diluted 1:2000. The other Abs were also diluted including anti–β-actin (Cell Signaling Technology) at 1:1000, anti-Flag/HA (Sigma-Aldrich) at 1:2000, anti-myc (Santa Cruz Biotechnology) at 1:2000, and HRP-conjugated anti-rabbit IgG or anti-mouse IgG (Thermo Scientific) at 1:5000. The results were the representative of three independent experiments. ImageJ was used for quantifying the protein levels based on the band density obtained from Western blot analysis.

EPC cells were transiently transfected with 5 μg MAVS-Flag, 4 μg Myc-N, and 1 μg of the hemagglutinin (HA)-Ub, HA-Ub-K48O, or HA-Ub-K63O expression plasmids. At 18 h posttransfection, the cells were treated with 20 μM MG132. Samples were harvested at 24 h posttransfection, lysed using a radioimmunoprecipitation assay lysis buffer (1% Nonidet P-40, 50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 1 mM EDTA, 1 mM NaF, 1 mM Na₃VO₄, 1 mM PMSF, and 0.25% sodium deoxycholate) that contained the protease inhibitor mixture (Sigma-Aldrich). The samples were denatured by heating for 10 min in 1% SDS and diluted with lysis buffer until the concentration of SDS was decreased to 0.1%. The diluted supernatants were then immunoprecipitated overnight at 4°C with constant agitation with 30 μl anti-Flag agarose conjugate (Sigma-Aldrich). Immunoprecipitated protein was washed three times with lysis buffer and resuspended in 2× SDS sample buffer. The immunoprecipitates and whole-cell lysates were analyzed by immunblotting with various Abs that are indicated.

EPC cells were seeded in a 6-well plate and transfected with various plasmids for 24 h. Then, the cells were washed twice with PBS and fixed with 4% PFA for 1 h. Images were obtained using a fluorescent microscope (Nikon) through a ×10 objective lens.

Data are expressed as mean ± SD of at least three independent experiments (n ≥ 3). The p values were calculated by one-way ANOVA with Dunnett’s posthoc test (version 19 SPSS Statistics; IBM). A p value < 0.05 was considered statistically significant.

Previous studies have shown that many strategies are used by viruses to counteract host IFN production. Whether the expression of fish IFN could be inhibited by SVCV was explored by the luciferase reporter gene assay. Because poly(I:C) mimics viral RNA, it was chosen for inducing the expression of the four type I IFNs (IFNφ1 to IFNφ4) in zebrafish. As shown in Fig. 1A, the activation of IFNφ1 and IFNφ3 promoters were significantly upregulated by transfection with poly(I:C); however, the IFNφ2 and IFNφ4 promoters were unaffected. IFNφ1 and IFNφ3 had similar regulation patterns and IFNφ1 but not IFNφ3 could be upregulated by IRF3 (32, 33). In agreement with the results above, the IFNφ1 promoter (IFNφ1pro) activity was again induced by poly(I:C); however, the induction was remarkably suppressed by infection with SVCV in a dose-dependent manner (Fig. 1B). These results demonstrated that the upregulation of IFNφ1 by poly(I:C) could be blocked by SVCV.

Similar to mammals, fish RLR signaling cascades have also been reported to activate IFN expression (29). To determine whether the zebrafish RLR pathway could be disrupted by an SVCV infection, the N-terminal domain of RIG-I (RIG-I-Nter), MAVS, TBK1, IRF3, and IRF7 were cotransfected with IFNφ1pro following infection with different titers of SVCV. As shown in Fig. 2A, a low level of expression of IFNφ1 could be induced with a high concentration of SVCV. In addition, overexpression of the RLR cascades led to a higher induction of IFNφ1pro activity. However, the activation of IFNφ1 driven by RIG-I-Nter and MAVS were blocked by SVCV in a dose-dependent manner (Fig. 2B, 2C), whereas the induction of IFNφ1 by TBK1 and IRF7 were not affected (Fig. 2D, 2E). Interestingly, the induction of IFNφ1 by IRF3 was significantly blocked by a high concentration of SVCV, which might be due to SVCV developing multiple strategies to evade host immune surveillance (Fig. 2F). Collectively, because MAVS is a unique adaptor protein downstream of RIG-I and upstream of TBK1, these data indicated that SVCV disrupts the activation of IFNφ1 by targeting MAVS.

To understand how SVCV blocks MAVS signal transduction, the expression pattern of MAVS was analyzed in SVCV-infected cells. Remarkably, immunoblotting revealed that the abundance of MAVS was substantially decreased postinfection with SVCV for 24 h. However, the IRF7 concentration did not appear to change (Fig. 3A). Similarly, a reduction in MAVS-Flag and no effect on HA-IRF7 were observed in cells overexpressing exogenous MAVS-Flag or HA-IRF7 postinfection with SVCV (Fig. 3B, 3C). MG132, a potent inhibitor of the proteasome, was used to treat cells for 8 h to further identify the degradation pattern of MAVS. Interestingly, the expression of MAVS was rescued gradually with the increasing concentration of MG132, indicating that MAVS was degraded in an Ub-proteasome manner (Fig. 3D). The expression of mavs at the mRNA level was also analyzed. As shown in Fig. 3E, compared with ifnφ1, there was no significant change in the transcription of mavs after challenge with SVCV from 0 to 24 h. These results suggested that SVCV reduces the expression of MAVS by an Ub-proteasome system at the protein level but not at the mRNA level.

To further elucidate the mechanism of the blocking of IFNφ1 signaling by SVCV and based on the results of preliminary experiments (data not shown), the N protein of SVCV was chosen for further study. As shown in Fig. 4A and 4B, the activation of IFNφ1pro induced by poly(I:C) was suppressed by overexpression of the N protein, whereas IFNφ1pro activity did not change in the G protein group. In addition, the IFNφ1pro activity was suppressed gradually with increasing amounts of the N protein, which occurred in a dose-dependent manner (Fig. 4C). Because many viruses have developed diverse strategies to interfere with the NF-κB signaling to inhibit a host antiviral response (34), the NF-κB promoter was also monitored in the presence of the N protein. As shown in Fig. 4D, similar to IFNφ1pro, the activation of the NF-κB promoter induced by poly(I:C) also decreased with the overexpression of the N protein. These results indicated that the N protein is a potential factor of SVCV that interferes with the IFNφ1 production.

Given the fact that MAVS plays a paramount role in IFNφ1 activation, it was speculated that MAVS was the target of the N protein. As shown in Fig. 5A, IFNφ1pro activity was significantly upregulated by MAVS; however, this induction could be suppressed by the overexpression of the N protein. Similarly, the ISRE activity upregulated by MAVS was also decreased with the overexpression of the N protein (Fig. 5B). In the RLR signaling pathway, MAVS is a downstream adaptor of RIG-I and upstream activator of TBK1. When the N protein was overexpressed, the induction of IFNφ1 by RIG-I-Nter and MAVS was remarkably inhibited; however, no apparent effects were observed on IFNφ1 activation induced by TBK1 (Fig. 5C). These results suggested that the N protein might inhibit the transcription of IFNφ1 induced by the RLR axis through the negative regulation of MAVS.

To investigate the mechanism of regulation of the N protein on MAVS, the modulation pattern of MAVS at the protein level during the overexpression of the N protein was analyzed. MAVS-, TBK1-, and IRF7-mCherry expression vectors were cotransfected with N-EGFP or the empty vector (EGFP). As shown in Fig. 6A, a striking reduction in the abundance of MAVS-mCherry was observed, however, the fluorescence signals of TBK1- and IRF7-mCherry did not appear to be affected. The fluorescence signals of N-EGFP and EGFP in the MAVS-, TBK1-, and IRF7-mCherry groups were similar to the control group (data not shown). In addition, the exogenous MAVS was substantially reduced again with the overexpression of the N protein in a dose-dependent manner (Fig. 6B, 6C). As a control, the P protein of SVCV had little effect on MAVS (Fig. 6D). In contrast, the content of exogenous IRF7 did not change with the overexpression of the N protein (Fig. 6E). Our previous study showed that a truncated MAVS (MAVS_tv2) in zebrafish, lacking the C-terminal transmembrane (TM) domain, could negatively regulate IRF7-mediated IFNφ1 production (31). To further characterize the degradation region of MAVS, MAVS_tv2 was used. As shown in Fig. 6F, the abundance of MAVS_tv2 also decreased when cotransfected with the N protein, suggesting that the TM domain of MAVS is not the target of the N protein. Consistent with the result that the transcription of mavs was not suppressed by SVCV, it was also not decreased by the overexpression of the N protein (Fig. 6G). These results indicated that overexpression of the N protein of SVCV specifically promotes the degradation of MAVS at the protein level and the target region is beyond the TM domain.

The results above suggested that MAVS was degraded by SVCV in a proteasome-dependent manner. To delineate the mechanisms responsible for the degradation of MAVS by the N protein of SVCV, cells were cotransfected with MAVS-Flag and HA-N in the presence of MG132. As shown in Fig.7A, the band of MAVS in the cells coexpressing the N protein was weaker than that in the control cells; however, the reduction of MAVS was significantly rescued when treated with MG132. In addition, the repression of MAVS was rescued with the increasing concentration of MG132, which also displayed a dose-dependent manner (Fig. 7B). To further clarify whether MAVS was degraded by the Ub-proteasome pathway, MAVS-Flag, Myc-N, and HA-Ub were cotransfected in the presence or absence of MG132. Following immunoprecipitation of MAVS-Flag, immunoblot analysis revealed that the N protein promoted the ubiquitination of MAVS (Fig. 7C). K48 and K63, the lysines at positions 48 and 63 of ubiquitin linked with polyubiquitin chains, are two canonical polyubiquitin chain linkages. A series of cases have reported that the target proteins were degraded by K48-linked polyubiquitin chains in a proteasome-dependent manner, whereas their functions were stabilized and enhanced by K63-linked polyubiquitin chains (35–37). To dissect the pattern of ubiquitination of MAVS through K48- or K63-linked ubiquitination, cells were cotransfected with MAVS-Flag, Myc-N, HA-Ub-K48O, or HA-Ub-K63O. As shown in Fig. 7D, the N protein promoted K48-linked ubiquitination of MAVS but not K63-linked ubiquitination. These results indicated that the K48-linked Ub-proteasomal degradation of MAVS was triggered by the N protein of SVCV.

To determine whether the replication of SVCV is enhanced by the N protein, cells were transfected with pcDNA3.1-N or the empty vector and then infected with different titers of SVCV. As shown in Fig. 8A, more CPE was observed in the N protein group at 2 d postinfection. This was confirmed by the titer of SVCV that significantly increased (32,000-fold) in the N protein overexpressing cells compared with the control cells (Fig. 8B). To identify the effect of the N protein on the replication of SVCV, the transcripts of the genes of the G, M, and P proteins of SVCV in EPC cells were monitored by qPCR. Upon infection with SVCV, overexpression of the N protein could facilitate the transcriptions of the G, M, and P of SVCV (Fig. 8C). These results demonstrated that the N protein enhances the capacity of SVCV to replicate.

Similar to that in mammals, fish IFNs play important roles in the host innate immune system and trigger the expression of massive IFN-stimulated genes that can interfere with viral transcription, translation, assembly and release (38). For example, overexpression of zebrafish IFN in embryos induced remarkable upregulation of viperin and exhibited strong antiviral activity against infectious hematopoietic necrosis virus (39). In addition, products of two typical IFN-stimulated genes, PKZ and PKR, were able to inhibit the replication of grass carp reovirus by phosphorylating the α subunit of eukaryotic initiation factor 2 (40, 41). However, many aquatic viruses still cause significant losses in the cultured fish industries. The immune evasion mechanisms involved in the pathogenesis of aquatic viruses remain poorly understood. During the early stage of viral infection it is crucial for viruses to replicate and to be able to evade the attack of the hosts’ immune system. Many mammalian viruses have developed numerous strategies to evade the hosts’ IFN system, including inhibition of IFN production, suppression of IFN-mediated signaling pathways, and blocking the functions of IFN-induced antiviral proteins. For example, the 3C^pro cysteine protease of hepatitis A virus blocked IFN induction by cleaving MAVS (42); The NS5 protein of dengue virus inhibited IFN-dependent signaling by targeting STAT2 for proteasome-mediated degradation (43); The NS1 protein of influenza A and B viruses inhibited the activation of antiviral effector protein PKR by directly binding to the N-terminal (44). Blocking the production of IFN is crucial for viruses to escape this host defense mechanism, but until now, few mechanisms used by aquatic viruses had been described. The current study revealed that the N protein of SVCV inhibited zebrafish IFNφ1 expression by degrading MAVS in a Ub-proteasome manner. Because fish are a lower vertebrate, these findings indicate that the fight between a virus and the hosts’ IFN response has existed for a long time during evolution of vertebrates.

Recent studies have identified that MAVS is an adaptor molecule of RIG-I, which binds viral RNA directly and transmits signaling to induce IFN production (29). Interestingly, many viruses possess different strategies to suppress MAVS signaling, including cleaving MAVS from the mitochondria, assembling a degradasome to disable MAVS and disrupting the function of MAVS-containing complexes (14, 15, 45). Compared with other molecules, choosing MAVS as the target seems to be preferred by viruses. First, MAVS is an upstream factor in the RLR signaling pathway, can transduce signaling, and activate the expression of many genes not only IFN, which can provide a more powerful antiviral response (46, 47). Second, the transcription, translation, and assembly of SVCV is carried out in cytoplasm. Therefore, selecting a host cytoplasmic protein such as MAVS as the target will avoid encoding a nuclear localization signal and transferring this into the nucleus. Third, MAVS functions not only on the IFN system but also on inducing host cell apoptosis. It was reported that MAVS could be cleaved by dengue virus to retard host cell apoptosis and to maximize the use of the intracellular resource (48). Finally, MAVS is localized to the outer membrane of mitochondria that are crucial organelles responsible for ATP production (49). Degradation of MAVS might hinder the function of mitochondria and make host cells unable to provide energy for the antiviral system that could also decrease potency.

One of the host immune molecules could be the target of several viruses or one virus could also possess a number of different strategies to elude host defense mechanisms. Previous studies demonstrated that SVCV induced the autophagy of host cells and upregulated the expression of related genes (23). In the current study, the N protein inhibited IFN expression by degrading MAVS, which provided another strategy for SVCV to resist the host defense system. Among the other proteins of SVCV, the M protein has been reported to interfere with the posttranscriptional machinery of the host, such as blocking the nuclear transport of spliced mRNAs and small nuclear RNAs and to slow the nuclear transport of many other molecules (50). In addition, our initial study showed that the P protein could suppress the activation of IFNφ1 induced by poly(I:C) (data not shown). RNA viruses appear to use one protein to antagonize the IFN response by different strategies because of their relatively limited genome capacity (51–54). As an RNA virus causing significant mortality of fish, the functions of the different proteins of SVCV should be studied further.

In mammals, several viruses regulate the components of the IFN signal transduction pathway through ubiquitination or phosphorylation (55, 56). Ubiquitination is a reversible covalent modification that regulates the stability, activity, and localization of target proteins. The polyubiquitin chains linked through lysine at position 48 of ubiquitin (K48) target proteins for proteasomal degradation. For example, AIP4 catalyzed the K48-linked ubiquitination of MAVS, in cooperation with PCBP2, leading to the degradation of MAVS (57), whereas K63-linked ubiquitination usually carries out signaling functions independent of proteolysis. For example, TRIM25 targeted RIG-I for K63-linked ubiquitination and enhanced the binding of RIG-I to MAVS (58). In the current study, the N protein degraded MAVS that underwent K48-linked ubiquitination. However, the coimmunoprecipitation failed to show direct interaction between the N protein and MAVS (data not shown). Thus, our working hypothesis is that the N protein would recruit an E3 ligase or enhance its activity and then trigger the ubiquitination of MAVS. However, recent studies revealed that several E3 ligases were involved in SVCV infection (24); future studies should find out the exact E3 ligase(s) related with the N protein and the K48-linked ubiquitination residues of MAVS.

In conclusion, the current study revealed a potential vital mechanism used by the N protein of SVCV to negatively regulate MAVS through the Ub-proteasome degradation pathway (Fig. 9). These data provide new insights into the strategy of an aquatic virus to evade host innate immunity in a lower vertebrate. Further studies should focus on the functions of other proteins of SVCV in immune evasion.

We thank Dr. Hong-Bing Shu (Wuhan University, Wuhan, China) for providing plasmids (HA-Ub, HA-Ub-K48O, and HA-Ub-K63O) and Dr. Xing Liu (Institute of Hydrobiology) for providing vector (pCMV-Myc).

The authors have no financial conflicts of interest.