Grass Carp Reovirus VP41 Targets Fish MITA To Abrogate the Interferon Response

ABSTRACT

Although fish possess an efficient interferon (IFN) system to defend against aquatic virus infection, grass carp reovirus (GCRV) still causes hemorrhagic disease in grass carp. To date, GCRV's strategy for evading the fish IFN response is still unknown. Here, we report that GCRV VP41 inhibits fish IFN production by suppressing the phosphorylation of mediator of IFN regulatory factor 3 (IRF3) activation (MITA). First, the activation of the IFN promoter (IFNpro) stimulated by mitochondrial antiviral signaling protein (MAVS) and MITA was decreased by the overexpression of VP41, whereas such activation induced by TANK-binding kinase 1 (TBK1) was not affected. Second, VP41 was colocalized in the cellular endoplasmic reticulum (ER) and associated with MITA. Furthermore, as a phosphorylation substrate of TBK1, VP41 significantly decreased the phosphorylation of MITA. Truncation assays indicated that the transmembrane (TM) region of VP41 was indispensable for the suppression of IFNpro activity. Finally, after infection with GCRV, VP41 blunted the transcription of host IFN and facilitated viral RNA synthesis. Taken together, our findings suggest that GCRV VP41 prevents the fish IFN response by attenuating the phosphorylation of MITA for viral evasion.

IMPORTANCE MITA is thought to act as an adaptor protein to facilitate the phosphorylation of IRF3 by TBK1 upon viral infection, and it plays a critical role in innate antiviral responses. Here, we report that GCRV VP41 colocalizes with MITA at the ER and reduces MITA phosphorylation by acting as a decoy substrate of TBK1, thus inhibiting IFN production. These findings reveal GCRV's strategy for evading the host IFN response for the first time.

INTRODUCTION

Grass carp reovirus (GCRV), a highly virulent pathogenic agent of fish, has caused severe epidemic outbreaks of hemorrhagic disease and resulted in tremendous mortality in grass carp (Ctenopharyngodon idella) (1). It is a double-stranded RNA (dsRNA) virus belonging to the genus Aquareovirus in the family Reoviridae (2). The genome consists of 11 segments (termed S1 to S11) encased in a multilayered icosahedral capsid shell (3, 4). Based on genomic and biological characteristics, the known GCRV strains can be clustered into three groups (group I to group III) (2). Moreover, a protein sequence comparison showed that the similarity among the three groups is less than 20%, so the functions of the encoded proteins are likely to be diverse (2). For instance, segment 8 of group I has been found to encode a clamping protein (VP6) that bridges the inner core with the outer shell (3). Segment 8 of group II GCRV has been predicted to encode a protein of approximately 41 kDa (VP41) with a hydrophobic α-helical transmembrane (TM) region at the N terminus (5). Amino acid sequence analysis of VP41 demonstrates that there are no homologous proteins in other aquareoviruses (6). Segment 8 of group III GCRV has been predicted to encode the core protein VP6 and may be involved in the formation of a continuous capsid shell via clamping to VP3 (7). During recent years, great progress has been made in understanding the pathogenesis of GCRV (8–10). For instance, in fish spleen and liver, infection with GCRV has been shown to significantly induce the transcription of interferon (IFN) and multiple IFN-stimulated genes (ISGs), which displayed powerful capacities to defend against the influence of GCRV (11, 12). Hence, for GCRV, the host cellular IFN response should be inhibited to facilitate viral proliferation.

For host cells, viral infection triggers the activation of signaling cascades to initiate antiviral immune responses. For example, the retinoic acid-inducible gene I (RIG-I)-like receptor (RLR) pathway is crucial for the activation of IFN expression (13). The RLR family is comprised of three members: RIG-I, melanoma differentiation-associated gene 5 (MDA5), and laboratory of genetics and physiology 2 (LGP2) (14). Upon binding with viral RNA, the N-terminal caspase recruiting domain (CARD) of RIG-I and MDA5 interacts with another CARD-containing protein, mitochondrial antiviral signaling protein (MAVS) (also known as IPS-1, VISA, and Cardif) (15–18). This activates the downstream mediator of IFN regulatory factor 3 (IRF3) activation (MITA) (also known as STING, ERIS, and MPYS) and TANK-binding kinase 1 (TBK1), leading to the phosphorylation of IRF3/7, which is translocated to the nucleus and initiates the transcription of IFN (19–21). Several studies demonstrated that fish also possess a functional RLR pathway. For example, fish RIG-I and MDA5 have been shown to intensively trigger IFN production (22–24); IRF3 and MITA can be phosphorylated by TBK1, and they display a powerful capacity to activate IFN (25–30).

MITA has been identified as a critical factor participating in the RLR signaling pathway (31–36). In response to viral infection, MITA interacts with MAVS and acts as a scaffold protein to facilitate the phosphorylation of IRF3/7 by TBK1, leading to the induction of IFN (37). Consistently, in antiviral assays, a deficiency in MITA expression impairs the host antiviral response and increases susceptibility to viruses and certain intracellular bacteria (38–40). In fish, multiple-sequence alignments have revealed that zebrafish MITA has a high level of conservation with mammalian MITA. Previous studies demonstrated that fish MITA is comprised of five putative TM domains within its N terminus and that it predominantly resides in the endoplasmic reticulum (ER), but the function of the TM domains in the process of cellular location is unclear (30). In addition, the dominant negative mutant of fish MITA, lacking the N-terminal TM domains, blocks IFN expression induced by RIG-I and MDA5, suggesting that fish MITA is involved in the RLR pathway and located downstream of RIG-I (30, 41). Regarding antiviral capability, the overexpression of MITA in fish cells leads to strong inductions of IFN and ISGs and confers to the host an antiviral state to combat spring viremia of carp virus (SVCV) and GCRV infection (41).

Since the IFN response enables the host to defend against viral infection, for viruses, it is necessary to abrogate or evade the IFN pathway. As MITA plays a pivotal role in IFN production, it is a popular target of viral antagonists in mammals. For example, the nonstructural protein NS4B of hepatitis C virus (HCV) blocks IFN signaling by a direct protein interaction with MITA (42, 43), the protease NS2B3 of dengue virus (DENV) inhibits IFN production through cleaving MITA (44, 45), and human coronavirus (HCoV) NL63 and severe acute respiratory syndrome coronavirus (SARS-CoV) papain-like protease (PLP) antagonize innate immune signaling by disrupting MITA dimerization to suppress IRF3 activation (46). As noted above, although mammalian MITA has been characterized as a key target for viruses, whether fish MITA is blunted by aquatic viruses is still unknown.

To date, there have been few studies regarding the evasion mechanisms used by GCRV to interfere with fish IFN production (8, 10). In this study, we show that GCRV VP41 colocalizes and associates with MITA at the ER. It acts as a competitive substrate of TBK1 to mediate the reduction of MITA phosphorylation, leading to the inhibition of IFN expression and the facilitation of viral RNA synthesis. These data reveal that the evasion mechanism of GCRV involves negatively regulating the function of MITA.

RESULTS

GCRV VP41 blocks IFN-φ1 and IFN-φ3 induction.

Although fish possess an efficient IFN system to defend against aquatic viruses, GCRV still causes high mortality rates in grass carp, with hemorrhagic symptoms (Fig. 1A), and the mechanisms that GCRV uses to escape the host IFN response are still unknown. Here, to understand the strategies used by GCRV to combat the host, several constructs of GCRV segments were employed for luciferase experiments in vitro, and the S8-encoded protein (VP41), containing a putative TM region in its N terminus (Fig. 1B), exhibited the potential to counteract host IFN activation (data not shown). Four type I IFNs (IFN-φ1 to IFN-φ4) have been identified in zebrafish, but only IFN-φ1 and IFN-φ3 can be significantly activated by poly(I·C), a mimic of viral RNA (30, 47–49), indicating that IFN-φ1 and IFN-φ3 participate in the host antiviral process. To investigate the effect of GCRV VP41 on IFN regulation, a luciferase assay was performed in the following study. In agreement with previously reported findings, poly(I·C) and SVCV induced the activation of the IFN-φ1 promoter (IFN-φ1pro) in epithelioma papulosum cyprini (EPC) cells; however, such an induction was significantly suppressed by the overexpression of VP41 (Fig. 1C and D). Moreover, VP41 also impeded the activation of IFN-φ3pro upon transfection with poly(I·C) (Fig. 1E). The IFN-stimulated regulatory element (ISRE) is considered a transcription factor binding motif in the promoter regions of IFN and ISGs, facilitating gene transcription (30). Consistently, VP41 decreased ISRE activity under stimulation (Fig. 1F). Together, these results demonstrate that GCRV VP41 serves as a negative regulator to interfere with host IFN production.

FIG 1.

GCRV VP41 blocks poly(I·C)-triggered IFN-φ1/3pro activation. (A) Young grass carp, with obvious skin hemorrhage and lesion, intraperitoneally injected with GCRV 106 (100 μl of the filtered virus-containing supernatant of frozen and thawed GCO cells). (B) Diagrammatic representation of VP41 used in this study. S8 encodes a 361-amino-acid protein, containing a TM region in its N terminus predicted by amino acid sequence analysis (TMHMM server v. 2.0 website). (C and E) Overexpression of VP41 inhibits poly(I·C)-induced IFN-φ1/3pro activation. EPC cells were seeded into 24-well plates and cotransfected the next day with 250 ng the empty vector or pcDNA3.1-VP41 and 25 ng pRL-TK plus 250 ng IFN-φ1pro–Luc (C) or IFN-φ3pro–Luc (E). At 24 h posttransfection, cells were untreated (null) or transfected with poly(I·C). The luciferase activities were monitored 24 h after stimulation. (D) VP41 blocks SVCV-induced activation of IFN-φ1pro. EPC cells were seeded into 24-well plates overnight and cotransfected with 250 ng IFN-φ1pro–Luc and 25 ng pRL-TK plus 250 ng the empty vector or pcDNA3.1-VP41. At 24 h posttransfection, cells were infected with SVCV at a multiplicity of infection (MOI) of 10. The luciferase activities were monitored 24 h after stimulation. (F) VP41 suppresses ISRE activation induced by poly(I·C). EPC cells were cotransfected with 250 ng ISRE-Luc and 25 ng pRL-TK plus 250 ng the empty vector or pcDNA3.1-VP41. At 24 h posttransfection, cells were treated with poly(I·C). Luciferase activities were analyzed 24 h after stimulation. The promoter activity is presented as relative light units normalized to Renilla luciferase activity. Error bars are the SDs obtained by measuring each sample in triplicate. Asterisks indicate significant differences from the control (*, P < 0.05).

Knockdown of VP41 potentiates GCRV-induced IFN transcription.

Next, we examined the effects of the knockdown of VP41 on the endogenous gene expression of ifn. We used two VP41-specific small interfering RNAs (siRNAs) in this study: one (siVP41#2) significantly inhibited the expression of VP41, while the other (siVP41#1) had no detectable effect (Fig. 2A). In quantitative real-time PCR (RT-qPCR) experiments, the knockdown of VP41 increased the GCRV-induced transcription of ifn in grass carp ovary (GCO) cells (Fig. 2B). In addition, upon infection with GCRV, the viral s1, s6, and s10 genes were downregulated in GCO cells by the knockdown of VP41 (Fig. 2C). Collectively, these data suggest that VP41 inhibits GCRV-triggered IFN expression.

FIG 2.

Effects of RNAi-mediated knockdown of VP41 on GCRV-induced IFN expression. (A) Effects of RNAi on expression of GCRV S8. GCO cells were seeded into 6-well plates overnight and transfected with 100 nM siVP41#1, siVP41#2, or siCon. At 6 h posttransfection, the cells were infected with GCRV (100 μl of the filtered virus-containing supernatant of frozen and thawed GCO cells, which was diluted 100 times in PBS). At 24 h postinfection, total RNAs were extracted to examine the transcriptional levels of s8. (B and C) Effects of RNAi on GCRV-induced ifn transcription of host cells and the viral s1, s6, and s10 transcripts of GCRV. GCO cells were seeded into 6-well plates and transfected with 100 nM siCon or siVP41#2. At 6 h posttransfection, cells were uninfected or infected with GCRV (100 μl of the filtered virus-containing supernatant of frozen and thawed GCO cells, which was diluted 100 times with PBS) for 42 h before RT-qPCR analysis was performed. The relative transcription levels were normalized to the transcription level of the β-actin gene and are represented as fold induction relative to the transcription level in control cells, which was set to 1. Error bars represent SDs obtained by measuring each sample in triplicate. Asterisks indicate significant differences from the control (*, P < 0.05).

VP41 inhibits IFN induction activated by the RLR signaling pathway.

Previous studies demonstrated that fish RLR signaling cascades also play vital roles in activating the expression of IFN (47, 50). Here, to determine whether the suppression of IFN mediated by VP41 occurred through the RLR pathway, zebrafish RLR constructs and IFN-φ1/3pro were employed in the following studies. As shown in Fig. 3A and B, IFN-φ1pro and IFN-φ3pro activities were induced by the RLR cascades, and these activations driven by the N terminus of RIG-I (RIG-I-Nter), MAVS, and MITA were all suppressed by VP41, while activation by TBK1 was not affected. Therefore, these results suggest that VP41 inhibits the RLR signaling pathway at the step of MITA.

FIG 3.

VP41 suppresses IFN-φ1/3 activation mediated by RIG-I, MAVS, or MITA. (A and B) EPC cells were seeded into 24-well plates overnight and cotransfected with a RIG-I-Nter-, MAVS-, MITA-, or TBK1-expressing plasmid; the empty vector or pcDNA3.1-VP41; plus IFN-φ1pro–Luc (A) or IFN-φ3pro–Luc (B) at a ratio of 1:1:1. pRL-TK was used as a control. At 48 h posttransfection, cells were lysed for luciferase activity detection. The promoter activity is presented as relative light units normalized to Renilla luciferase activity. Error bars are the SDs obtained by measuring each sample in triplicate. Asterisks indicate significant differences from the control (*, P < 0.05).

VP41 colocalizes with MITA at the ER.

Since the function of MITA on activating IFN was negatively regulated by VP41, the relationship between VP41 and MITA needs to be clarified. Fish MITA has been found to consist of five TM domains and to be located at the ER (30). Similarly, a putative TM region has been found in the N terminus of VP41. Thus, first, the subcellular localization of VP41 was investigated. After cotransfection with VP41-enhanced green fluorescent protein (EGFP) and ER-DsRed (ER marker) or the empty vector (DsRed), confocal microscopy analysis revealed that the green signals of VP41 mainly overlapped the red signals of the ER marker, suggesting that VP41 is located at the ER (Fig. 4A and B). As a positive control, cells transfected with MITA-EGFP and ER-DsRed also showed an overlapping image (Fig. 4C). Furthermore, the colocalization of VP41 with MITA was monitored. By cotransfection with VP41-EGFP and MITA-DsRed or TBK1-DsRed, the signals of VP41 were shown to almost overlap those of MITA and partly colocalized with those of TBK1 (Fig. 4D and E). These results indicate that VP41 predominantly colocalizes with MITA at the ER.

FIG 4.

Subcellular localization of VP41. (A and B) VP41 is localized at the ER. EPC cells seeded onto microscopy cover glass in 6-well plates were cotransfected with 2 μg VP41-EGFP and 2 μg the empty vector (A) or ER-DsRed (B). After 24 h, the cells were fixed and subjected to confocal microscopy analysis. The yellow staining in the merged image indicates that VP41 is localized at the ER. (C) Subcellular localization of MITA. EPC cells seeded onto microscopy cover glass in 6-well plates were cotransfected with 2 μg MITA-EGFP and 2 μg ER-DsRed. After 24 h, the cells were fixed and examined by using a confocal microscope. The yellow staining in the merged image indicates that MITA is localized at the ER. (D and E) VP41 colocalizes with MITA. EPC cells were plated onto coverslips in 6-well plates and cotransfected with 2 μg VP41-EGFP and 2 μg MITA-DsRed (D) or TBK1-DsRed (E). After 24 h, the cells were fixed and observed by confocal microscopy. Green signals represent overexpressed VP41. Red signals represent overexpressed MITA or TBK1, and blue staining indicates the nucleus region. The yellow staining in the merged image indicates the colocalization of VP41 and MITA (original magnification, ×63; oil immersion objective). Bar, 10 μm. All experiments were repeated at least three times, with similar results.

VP41 associates with MITA and TBK1.

To clarify the relevance of VP41 and MITA, the interaction pattern was examined by a coimmunoprecipitation (co-IP) assay. In HEK 293T cells cotransfected with hemagglutinin (HA)-VP41 and Myc-MAVS or with Myc-VP41 and Flag-TBK1 or Flag-MITA, the anti-Myc antibody (Ab)-immunoprecipitated protein complexes containing MAVS were not recognized by the anti-HA Ab. However, the anti-Flag Ab-immunoprecipitated protein complexes containing TBK1 and MITA were recognized by the anti-Myc Ab, suggesting that VP41 associates with MITA and TBK1 (Fig. 5A and B). Next, to identify the functional domain of MITA interacting with VP41, two truncated mutants of MITA were constructed, MITA-ΔN (lacking the N-terminal TM region) and MITA-ΔC (lacking the C terminus) (Fig. 5C). As shown in Fig. 5D, consistent with wild-type MITA, MITA-ΔC bound with VP41, while such an association was abrogated in the MITA-ΔN group. Furthermore, the role of the N terminus for MITA was also examined by subcellular localization. As shown in Fig. 5E, MITA lacking the N terminus led to complete cellular distribution and the disruption of colocalization with VP41. Collectively, these data suggest that VP41 directly or indirectly associates with MITA and TBK1 and that the TM region of MITA is indispensable.

FIG 5.

VP41 associates with MITA and TBK1. (A and B) VP41 associates with MITA and TBK1. HEK 293T cells seeded into 10-cm² dishes were transfected with the indicated plasmids (5 μg each). After 24 h, cell lysates were immunoprecipitated (IP) with anti-Myc/Flag affinity gel. The immunoprecipitates and cell lysates were then analyzed by immunoblotting (IB) with anti-HA, anti-Myc, and anti-Flag Abs, respectively. (C) Schematic representation of mutants of zebrafish MITA used in this study. The ER localization domain is indicated within its N-terminal TM region. There are two mutants of MITA: MITA-ΔN, containing the C-terminal 222 amino acids, and MITA-ΔC, containing the ER localization domain. (D) VP41 associates with MITA via its N terminus. The experiments were performed similarly as described above for panel A. (E) VP41 colocalizes with the N-terminal TM region of MITA. EPC cells were plated onto coverslips in 6-well plates and cotransfected with 2 μg VP41-DsRed and 2 μg MITA-EGFP, MITA-ΔN-EGFP, or MITA-ΔC-EGFP plasmids. After 24 h, the cells were fixed and subjected to confocal microscopy analysis (original magnification, ×63; oil immersion objective). Bar, 10 μm. All experiments were repeated three times, with similar results.

VP41 decreases TBK1-mediated MITA phosphorylation.

As a scaffold protein, mammalian MITA promotes TBK1-mediated IRF3 phosphorylation, triggering the robust expression of IFN. In this process, MITA is phosphorylated by TBK1 as well, which is indispensable for IFN signaling transduction (31, 37). In zebrafish, since both MITA and TBK1 associate with VP41, whether the activation of MITA is influenced by VP41 needs to be clarified. As shown in Fig. 6A, when Myc-MITA was cotransfected with Flag-TBK1, shifted bands with higher molecular weights were detected by the anti-Myc Ab. Furthermore, to confirm whether the shifted bands represented phosphorylated MITA, a dephosphorylation assay was performed in vitro, and the shifted bands partially disappeared after being treated with calf intestinal phosphatase (CIP), demonstrating that fish MITA can also be phosphorylated by TBK1. Given that the structural feature of the TM region and the cellular location of VP41 are similar to those of MITA and that VP41 can associate with both TBK1 and MITA, whether VP41 can be phosphorylated by TBK1 was assayed. As shown in Fig. 6B, when MITA or TBK1 was overexpressed individually, the molecular weight of VP41 remained unchanged compared with the control group. However, a band shift of VP41 was observed when MITA and TBK1 were cotransfected. To identify whether the shifted band represented phosphorylated VP41, the cell lysate was treated with CIP. As expected, the shifted band cooperatively induced by TBK1 and MITA disappeared, indicating that VP41 can also be phosphorylated by TBK1 in cooperation with MITA (Fig. 6C). Finally, since both MITA and VP41 can be phosphorylated by TBK1, the competition between VP41 and MITA to be phosphorylated by TBK1 was examined. As shown in Fig. 6D, the phosphorylation of MITA caused by TBK1 was reduced dose dependently with the overexpression of VP41. In addition, our previous study demonstrated that IRF3 can also be phosphorylated by TBK1 (29). Next, it was worth investigating whether VP41 affects the TBK1-induced phosphorylation of IRF3. As shown in Fig. 6E, cotransfection with Flag-TBK1 caused a shift of IRF3 to higher-molecular-weight bands. Furthermore, the in vitro dephosphorylation assay demonstrated that TBK1 was also a conserved phosphokinase for IRF3. Similarly, the level of phosphorylated IRF3 was gradually reduced by the exogenous expression of VP41 in a dose-dependent manner (Fig. 6F). Taken together, these data demonstrate that VP41 reduces the TBK1-triggered phosphorylation of MITA or IRF3 by being competitively phosphorylated by TBK1.

FIG 6.

VP41 decreases TBK1-mediated phosphorylation of MITA. (A) TBK1 mediates the phosphorylation of MITA. HEK 293T cells were seeded into 6-well plates overnight and transfected with the indicated plasmids (2 μg each) for 24 h. The cell lysates (100 μg) were treated with or without CIP (10 U) for 40 min at 37°C. The lysates were then detected by immunoblotting (IB) with anti-Myc, anti-Flag, and anti-β-actin Abs. (B) The MITA-TBK1 association induces the phosphorylation of VP41. HEK 293T cells were seeded into 6-well plates overnight and transfected with the indicated plasmids (1.5 μg each) for 24 h. The cell lysates were subjected to IB with anti-Myc, anti-HA, anti-Flag, and anti-β-actin Abs. (C) The level of TBK1-phosphorylated VP41 is reduced by CIP treatment. HEK 293T cells were seeded into 6-well plates overnight and transfected with the indicated plasmids (1.5 μg each) for 24 h. The cell lysates (100 μg) were treated with or without CIP (10 U) for 40 min at 37°C. The lysates were then detected by IB with anti-Myc, anti-HA, anti-Flag, and anti-β-actin Abs. (D) Overexpression of VP41 inhibits TBK1-mediated phosphorylation of MITA in a dose-dependent manner. HEK 293T cells were seeded into 6-well plates overnight and cotransfected with 1.5 μg Flag-TBK1 and 1.5 μg the empty vector or Myc-VP41 (1 and 2 μg, respectively), together with 1.5 μg HA-MITA for 24 h. The lysates were then subjected to IB with anti-HA, anti-Myc, anti-Flag, and anti-β-actin Abs. (E) TBK1 phosphorylates IRF3. HEK 293T cells were seeded into 6-well plates overnight and transfected with the indicated plasmids (2 μg each) for 24 h. The cell lysates (100 μg) were treated with or without CIP (10 U) for 40 min at 37°C. The lysates were then detected by IB with anti-Myc, anti-Flag, and anti-β-actin Abs. (F) VP41 decreases TBK1-mediated phosphorylation of IRF3. HEK 293T cells were seeded into 6-well plates overnight and cotransfected with 1.5 μg Flag-TBK1 and 1.5 μg the empty vector or Myc-VP41 (1 and 2 μg, respectively), together with 1.5 μg HA-IRF3 for 24 h. The lysates were then detected by IB with anti-HA, anti-Myc, anti-Flag, and anti-β-actin Abs. All experiments were repeated at least three times, with similar results.

The N-terminal TM region of VP41 is essential for its inhibitory activity.

Both VP41 and MITA contain a TM region in their N termini, and previous studies suggested that the TM region is essential for MITA to localize at the ER and to activate IFN production (32). Accordingly, to characterize the function of the TM region of VP41, the truncated VP41-ΔTM mutant lacking the TM region was generated (Fig. 7A). First, fluorescence microscopy suggested that wild-type VP41 was specifically localized at the ER, whereas VP41-ΔTM was distributed throughout the whole cell (Fig. 7B). The function of the TM region of VP41 in IFN regulation was then monitored by a luciferase assay. Compared with wild-type VP41, VP41-ΔTM could not inhibit the activation of IFN-φ1pro induced by poly(I·C) or SVCV (Fig. 7C and D). Not surprisingly, in the RLR axis, the induction of IFN-φ1pro by MAVS and MITA was not decreased with the overexpression of VP41-ΔTM (Fig. 7E). Finally, VP41-ΔTM did not affect the TBK1-mediated phosphorylation of MITA (Fig. 7F). Collectively, these data suggest that the N-terminal TM region is necessary for the subcellular localization and IFN inhibition of VP41.

FIG 7.

The N-terminal TM region of VP41 is essential for its cellular location and inhibitory activity. (A) Diagrammatic representation of mutants of VP41 used in this study. The putative TM is indicated within its N terminus. VP41-ΔTM contains the C-terminal 337 amino acids. (B) VP41 is localized at the ER via its N-terminal TM region. EPC cells were plated onto coverslips in 6-well plates and cotransfected with 2 μg ER-DsRed and 2 μg the VP41-EGFP or VP41-ΔTM–EGFP plasmid. After 24 h, the cells were fixed and subjected to confocal microscopy analysis (original magnification, ×63; oil immersion objective). Bar, 10 μm. (C and D) The VP41-ΔTM protein has no effect on poly(I·C)/SVCV-triggered IFN-φ1 induction. EPC cells were seeded into 24-well plates and cotransfected the next day with 250 ng the empty vector, pcDNA3.1-VP41, or pcDNA3.1-VP41-ΔTM; 25 ng pRL-TK; and 250 ng IFN-φ1pro–Luc. At 24 h posttransfection, cells were untreated (null) or treated with poly(I·C) (C) or SVCV (MOI = 10) (D). The luciferase activities were monitored 24 h after stimulation. The promoter activity is presented as relative light units normalized to Renilla luciferase activity. (E) VP41 inhibits IFN-φ1 activation mediated by MAVS or MITA, dependent on its N-terminal TM region. EPC cells were seeded into 24-well plates and cotransfected with MAVS-, MITA-, and TBK1-expressing plasmids; pcDNA3.1-VP41 or pcDNA3.1-VP41-ΔTM; and IFN-φ1pro–Luc at a ratio of 1:1:1. pRL-TK was used as a control. At 48 h posttransfection, cells were lysed for luciferase activity detection. The promoter activity is presented as relative light units normalized to Renilla luciferase activity. Error bars are the SDs obtained by measuring each sample in triplicate. Asterisks indicate significant differences from the control (*, P < 0.05). (F) VP41 inhibits TBK1-mediated phosphorylation of MITA, relying on its N-terminal TM region. HEK 293T cells were seeded into 6-well plates overnight and cotransfected with 1.5 μg Flag-TBK1 and 1.5 μg the empty vector, Myc-VP41, or Myc–VP41-ΔTM, together with 1.5 μg HA-MITA, for 24 h. The lysates were then detected by IB with anti-HA, anti-Myc, anti-Flag, and anti-β-actin Abs. All experiments were repeated at least three times, with similar results.

VP41 attenuates the IFN response and facilitates viral RNA synthesis.

To determine whether VP41 interferes with the cellular IFN response to facilitate viral RNA synthesis, GCO cells were transfected with VP41 and stimulated with GCRV. Total RNAs were extracted and monitored by RT-qPCR. As shown in Fig. 8A, the expression level of the ifn transcript in cells overexpressing VP41 was reduced compared to the levels in control cells. After infection with GCRV, the viral s1, s6, and s10 genes were upregulated in cells overexpressing VP41 (Fig. 8B). In addition, RT-qPCR analysis indicated that the overexpression of VP41 inhibited the SVCV-induced expression of ifn and several other ISGs, such as isg15-1 and rig-i (Fig. 8C). Meanwhile, upon stimulation with SVCV, the transcript levels of the viral g, m, n, and p genes were increased when VP41 was overexpressed (Fig. 8D). These data indicate that VP41 suppresses the cellular IFN response and enhances viral RNA synthesis.

FIG 8.

Overexpression of VP41 dampens the cellular IFN response and facilitates viral RNA synthesis. (A and B) GCO cells seeded into 6-well plates overnight and transfected with 2 μg Myc-VP41 or the empty vector, at 6 h posttransfection, were infected with GCRV (100 μl of the filtered virus-containing supernatant of frozen and thawed GCO cells, which was diluted 100 times with PBS) for 42 h. Total RNAs were extracted to examine the transcriptional levels of cellular ifn (A) and s1, s6, and s10 of GCRV (B) by RT-qPCR. (C and D) EPC cells seeded into 6-well plates overnight were transfected with 2 μg pcDNA3.1-VP41 or the empty vector and infected with SVCV (MOI = 10) at 24 h posttransfection. At 24 h postinfection, total RNAs were extracted to examine the transcriptional levels of cellular ifn, isg15-1, and rig-i (C) and the g, m, n, and p genes of SVCV (D) by RT-qPCR. The relative transcription levels were normalized to the transcription level of the β-actin gene and are represented as fold induction relative to the transcription level in control cells, which was set to 1. Error bars represent SDs obtained by measuring each sample in triplicate. Asterisks indicate significant differences from the control (*, P < 0.05).

DISCUSSION

As in mammals, the fish IFN system plays a crucial role in defense against viral infections. Meanwhile, aquatic viruses have evolved various tactics to antagonize the host IFN response for replication and proliferation. The aquatic virus GCRV has been reported to be a vital pathogen of grass carp, while the molecular mechanism underlying its subversion of the host IFN response is poorly understood. In this report, we have revealed that GCRV VP41 competes with MITA to be phosphorylated by TBK1, thereby blocking IFN production and promoting GCRV RNA synthesis. This finding enhances the current understanding of the immune evasion mechanisms of aquatic viruses.

The key adaptor molecule MITA plays critical roles in both DNA and RNA virus-triggered IFN induction (31, 32, 34, 35). Consequently, MITA is targeted by various viruses for immune evasion in mammals. Generally, three mechanisms are used to disable MITA signaling: disruption of the interaction, cleavage, and posttranslational modification (43, 45, 51). There are several reasons for viruses to choose MITA as a target; for example, MITA is involved in the regulation of cellular proliferation, differentiation, and programmed death (52). More crucially, MITA functions as a scaffold protein recruited by TBK1 to facilitate IRF3 phosphorylation and activate IFN expression. Thus, abolishing the MITA function is necessary and effective for viruses to escape from the host IFN response. The conserved RLR signaling of fish in response to virus infection has been demonstrated, and fish MITA predominantly resides at the ER and plays an important role in IFN activation, protecting fish cells from GCRV and SVCV infection (30, 41). Interestingly, in this study, GCRV VP41 was associated with the TM region of MITA, which is necessary for MITA to be located at the ER. Although both VP41 and MITA were located at the ER, and the signals of co-overexpressed VP41 and MITA overlapped, whether the ER location of MITA is influenced by VP41 remains unknown.

Actually, the specific ER localization of VP41 may be more crucial for viruses than we realize, because the ER is a vital organelle that enables both hosts and viruses to synthesize proteins. Previous studies demonstrated that the ER is related to the assembly of viral replication complexes. For example, the 2C protein of poliovirus reorganizes intracellular membranes, including the ER membrane, to support RNA replication (53). NS4B of HCV is an ER-localized protein that induces the convolution of the ER membrane and the formation of a membranous web that provides an important platform within the HCV replication complex (54). Plant alphaviruses use viral membrane-bound proteins as organizer proteins of the RNA replication complex (55). Thus, the localization of VP41 to the ER membrane may be indispensable for the viral particle assembly of GCRV. On the other hand, viruses are also directed to the ER membrane for the evasion of immune responses. For instance, the 3A protein of enteroviruses disrupts ER-to-Golgi traffic and inhibits host antiviral protein secretion (56). Future studies to uncover the specific functions of VP41 at the ER membrane should be conducted.

Although a hydrophobic α-helical TM region was predicted in the N terminus of VP41 (5), the functions of VP41 for the viral life cycle or immune evasion are still unclear. In this study, we demonstrated that VP41 was phosphorylated by TBK1 and facilitated the RNA synthesis of GCRV. Actually, several studies have revealed that the phosphorylated viral proteins are indispensable for viral genome transcription or particle assembling. For instance, the human immunodeficiency virus type 1 (HIV-1) Tat protein is phosphorylated by the cellular proteins cyclin-dependent kinase 2 (CDK2)/cyclin E and promotes viral transcription (57), and the HCV NS5A protein is phosphorylated by the host cellular protein CKII and regulates virion assembly (58). However, whether viral RNA synthesis was promoted by the inhibition of host IFN or by the phosphorylation of VP41 should be further investigated.

Viruses have evolved multiple strategies to interfere with the relevant signal transduction pathways at different steps to subvert the host IFN response. In this study, GCRV VP41 inhibited IFN production activated by the RLR signaling pathway. There are several other signaling pathways for IFN activation, such as the Toll-like receptor (TLR), MyD88, and JAK-STAT pathways. In some viruses, one viral protein can antagonize multiple host antiviral components to promote the efficiency of IFN inhibition. For example, the 3Cpro cysteine protease of coxsackievirus B3 (CVB3) blocks host immunity by cleaving two key adaptor molecules for innate immunity: MAVS and Toll–interleukin-1 receptor (TIR) domain-containing adaptor-inducing interferon beta (TRIF) (59). Meanwhile, the use of various proteins encoded by one virus to antagonize specific signaling pathways appears to be another strategy of the virus. For HCV, NS3/4A cleaves TRIF, and NS4B targets MITA, thereby subverting both the TLR3 and RLR pathways (42, 60). Actually, the protein encoded by S11 of group II GCRV also suppressed IFN expression in a luciferase assay (data not shown); hence, as GCRV is the leading cause of death in grass carp, the multifunctionality of its different proteins requires further study.

In conclusion, in this study, we showed that GCRV VP41 targets MITA to suppress IFN signaling, which sheds light on the novel manner of immune evasion utilized by this aquatic virus. Further study is needed to ascertain the other molecular events of GCRV, which will promote an in-depth understanding of GCRV pathogenesis and provide ideas for preventive strategies.

MATERIALS AND METHODS

Cells and viruses.

HEK 293T cells were grown at 37°C in 5% CO₂ in Dulbecco's modified Eagle's medium (DMEM; Invitrogen) supplemented with 10% fetal bovine serum (FBS; Invitrogen). GCO cells and EPC cells were maintained at 28°C in 5% CO₂ in medium 199 (Invitrogen) supplemented with 10% FBS. GCRV (strain 106, group II) was a gift from Lingbing Zeng (Yangtze River Fisheries Research Institute, Chinese Academy of Fishery Sciences). Because group II GVRV cannot cause a cytopathic effect (CPE) but can propagate in GCO cells, the cultured media with GCO cells infected with group II GCRV for 8 days were harvested and stored at −80°C until use. SVCV, a negative single-stranded RNA (ssRNA) virus, was propagated in EPC cells until CPE was observed, and the cultured media with cells were then harvested and stored at −80°C until use.

Plasmid construction and reagents.

The cDNA fragment encoding GCRV VP41 (GenBank accession no. KC201173.1) was amplified by RT-PCR from the total RNA of GCRV-infected cells and then cloned into the pcDNA3.1(+) (Invitrogen) or pCMV-Myc/pCMV-HA/pCMV-Tag 2C vectors (BD Clontech). The open reading frames (ORFs) of zebrafish MAVS (GenBank accession no. NM_001080584.2), TBK1 (accession no. NM_001044748.2), MITA (accession no. NM_001278837.1), and the truncated mutants of MITA were also subcloned into the pCMV-HA, pCMV-Myc, and pCMV-Tag 2C vectors. For subcellular localization, the ORF of GCRV VP41 was inserted into the pEGFP-N3 or pDsRed-N1 vector (BD Clontech). The ORFs of TBK1 and MITA were also inserted into the pDsRed-N1 vector. The whole ORF of MITA and the truncated mutants of MITA were also subcloned into the pEGFP-N3 vector. The cDNA fragment encoding truncated VP41 (VP41-ΔTM) was inserted into the pcDNA3.1(+), pCMV-Myc, or pEGFP-N3 vector. The pDsRed-ER plasmid was purchased from BD Clontech. The promoter sequences of zebrafish IFNs (IFN-φ1pro and IFN-φ3pro) were cloned and inserted into the pGL3-Basic luciferase reporter vector (Promega) to analyze promoter activity. The backbone of the pGL3-Basic luciferase reporter vector contains a modified coding region for firefly (Photinus pyralis) luciferase that has been optimized for monitoring transcriptional activity in transfected eukaryotic cells. The Renilla luciferase internal control vector (pRL-TK) was purchased from Promega. Plasmids containing the RIG-I-Nter, MAVS, MITA, TBK1, and ISRE luciferase reporter genes were described previously (61, 62). All constructs were confirmed by DNA sequencing. Primer sequences are available upon request. Poly(I·C) was purchased from Sigma-Aldrich and used at a final concentration of 1 μg/ml.

Luciferase activity assay.

EPC cells were seeded into 24-well plates overnight and cotransfected with various constructs at a ratio of 10:10:1 (RIG-I/MAVS/TBK1/MITA, IFN-φ1pro/IFN-φ3pro/ISRE-Luc, and pRL-TK expression vectors). The empty vector pcDNA3.1(+) was used to ensure that there were equivalent amounts of total DNA in each well. Transfection of poly(I·C) was performed 24 h before cell harvest. At 48 h posttransfection, luciferase activity was measured with the Dual-Luciferase Reporter Assay system (Promega) according to the manufacturer's instructions. Firefly luciferase activities were normalized on the basis of Renilla luciferase activity. The results are representative of data from more than three independent experiments, each performed in triplicate.

RNA interference (RNAi) experiment.

GCO cells were seeded into 6-well plates overnight and transfected with 100 nM siRNA of VP41 or the negative control (siCon) by using the X-tremeGENE HP DNA transfection reagent (Roche) according to the manufacturer's protocols. siRNA of VP41 and siCon were obtained from RiboBio Co., Ltd. (Guangzhou, China). The following sequences were targeted for GCRV VP41: siVP41#1 (GCCAAACGGACTCTACTTA) and siVP41#2 (TCTCCTCAAATGCCTGCAA).

RT-qPCR.

Total RNAs were extracted by using the TRIzol reagent (Invitrogen). cDNA was synthesized by using a GoScript reverse transcription system (Promega) according to the manufacturer's instructions. RT-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. The β-actin gene was used as an internal control. Primer sequences are available upon request. The relative fold changes were calculated by comparison to the corresponding controls using the 2^−ΔΔC_T method. Three independent experiments were conducted for statistical analysis.

Co-IP assay.

For transient-transfection and co-IP experiments, HEK 293T cells were used instead of EPC cells due to the superhigh transfection efficiency of HEK 293T cells. Cells seeded into 10-cm² dishes overnight were transfected with a total of 10 μg the indicated plasmids. At 24 h posttransfection, the medium was removed carefully, and the cell monolayer was washed twice with 10 ml ice-cold phosphate-buffered saline (PBS). The cells were then lysed in 1 ml radioimmunoprecipitation (RIPA) lysis buffer (1% NP-40, 50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 1 mM EDTA, 1 mM NaF, 1 mM sodium orthovanadate [Na₃VO₄], 1 mM phenylmethylsulfonyl fluoride [PMSF], 0.25% sodium deoxycholate) containing a protease inhibitor cocktail (Sigma-Aldrich) at 4°C for 1 h on a rocker platform. The cellular debris was removed by centrifugation at 12,000 × g for 15 min at 4°C. The supernatant was transferred to a fresh tube and incubated with 30 μl anti-HA-agarose beads or anti-Flag affinity gel (Sigma-Aldrich) overnight at 4°C with constant agitation. These samples were further analyzed by immunoblotting (IB). Immunoprecipitated proteins were collected by centrifugation at 5,000 × g for 1 min at 4°C, washed three times with lysis buffer, and resuspended in 50 μl 2× SDS sample buffer. The immunoprecipitates and whole-cell lysates were analyzed by IB with the indicated Abs.

Immunoblot analysis.

Immunoprecipitates or whole-cell extracts were separated by 10% SDS-PAGE and transferred onto a polyvinylidene difluoride (PVDF) membrane (Bio-Rad). The membranes were blocked for 1 h at room temperature in TBST buffer (25 mM Tris-HCl, 150 mM NaCl, 0.1% Tween 20 [pH 7.5]) containing 5% nonfat dry milk, probed with the indicated 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 the Immobilon Western chemiluminescent horseradish peroxidase (HRP) substrate (Millipore) and detected by using an ImageQuant LAS 4000 system (GE Healthcare). Abs were diluted as follows: anti-β-actin (Cell Signaling Technology) at 1:1,000, anti-Flag/HA (Sigma-Aldrich) at 1:3,000, anti-Myc (Santa Cruz Biotechnology) at 1:2,000, and HRP-conjugated anti-rabbit IgG or anti-mouse IgG (Thermo Scientific) at 1:5,000. Results are representative of data from three independent experiments.

In vitro protein dephosphorylation assay.

Transfected HEK 293T cells were lysed as described above, except that the phosphatase inhibitors (Na₃VO₄ and EDTA) were omitted from the lysis buffer. Protein dephosphorylation was carried out in 100-μl reaction mixtures consisting of 100 μg of cell protein and 10 U of CIP (Sigma-Aldrich). The reaction mixtures were incubated at 37°C for 40 min, followed by immunoblot analysis.

Fluorescence microscopy.

EPC cells were plated onto coverslips in 6-well plates and transfected with the indicated plasmids for 24 h. The cells were then washed twice with PBS and fixed with 4% paraformaldehyde (PFA) for 1 h. After being washed three times with PBS, the cells were stained with 4′,6-diamidino-2-phenylindole (DAPI) (1 μg/ml; Beyotime) for 15 min in the dark at room temperature. Finally, the coverslips were washed and observed with a confocal microscope under a ×63 oil immersion objective (SP8; Leica).

Statistics analysis.

Data are expressed as means ± standard deviations (SDs) of results from at least three independent experiments (n ≥ 3). The P values were calculated by one-way analysis of variance (ANOVA) with Dunnett's post hoc test (SPSS Statistics, version 19; IBM). A P value of <0.05 was considered statistically significant.