IFN-β promoter stimulator-1 (IPS-1)– and stimulator of IFN genes (STING)-mediated type I IFNs play a critical role in antiviral responses. Myxovirus resistance (Mx) proteins are pivotal components of the antiviral effectors induced by IFNs in many species. An unprecedented expansion of Mx genes has occurred in fish. However, the functions and mechanisms of Mx family members remain largely unknown in fish. In this study, we found that grass carp (Ctenopharyngodon idella) MxG, a teleost-specific Mx protein, is induced by IFNs and viruses, and it negatively regulates both IPS-1- and STING-mediated antiviral responses to facilitate grass carp reovirus, spring viremia of carp virus, and cyprinid herpesvirus-2 replication. MxG binds and degrades IPS-1 via the proteasomal pathway and STING through the lysosomal pathway, thereby negatively regulating IFN1 antiviral responses and NF-κB proinflammatory cytokines. MxG also suppresses the phosphorylation of STING IFN regulatory factor 3/7, and it subsequently downregulates IFN1 and NF-κB1 at the promoter, transcription, and protein levels. GTPase and GTPase effector domains of MxG contribute to the negative regulatory function. On the contrary, MxG knockdown weakens virus replication and cytopathic effect. Therefore, MxG can be an ISG molecule induced by IFNs and viruses, and degrade IPS-1 and STING proteins in a negative feedback manner to maintain homeostasis and avoid excessive immune responses after virus infection. To our knowledge, this is the first identification of a negative regulator in the Mx family, and our findings clarify a novel mechanism by which the IFN response is regulated.

The innate immune system depends on germline-encoded pattern recognition receptors to sense microbe-associated molecular patterns and initiate immune responses (1). Fish share most of the key antiviral pathways with humans and mice, which rely on a number of pattern recognition receptors, such as TLRs (2), retinoic acid–inducible gene I (RIG-I)–like receptors (RLRs) (3), and cytosolic DNA sensors (CDSs) (4). RLRs consist of RIG-I, melanoma differentiation–associated protein 5 (MDA5), and laboratory of genetics and physiology 2 (LGP2), which recognize viral dsRNA and activate IFN-β promoter stimulator-1 (IPS-1, also known as mitochondrial antiviral signaling protein [MAVS]) for downstream signal transmission (3). cGMP-AMP synthase is a cytosolic DNA sensor that activates innate immune responses through the production of the second messenger cGMP-AMP, which activates the stimulator of IFN genes (STING, also known as mediator of IFN regulatory factor [IRF]3 activation [MITA]) (5). IPS-1 and STING are key factors in the cellular antiviral signaling pathways (6). The interaction between viral RNA and RIG-I/MDA5 induces a conformational change that enables them to bind to the adaptor IPS-1, which causes the activation of NF-κB and IRFs, thereby leading to the induction of proinflammatory cytokines, chemokines, and antiviral responses, among others (7, 8). IFN binds to the cell surface receptors and activates JAK-STAT signal transduction cascades. The effector molecules suppress the replication of various pathogens (9).

To avoid host damage due to excessive responses, antiviral signaling pathways require tight regulation. Many studies have reported the critical roles of posttranslational modifications in regulation of the virus-triggered IFN pathway. In terms of the regulation of IPS-1, previous studies identified several ubiquitination-related factors that negatively regulate IPS-1 activity. MARCH5, Smad ubiquitin regulatory factor (Smurf2), and poly(rC)-binding protein 2 (PCBP2) degrade IPS-1 by promoting the polyubiquitin chains linked through lysine at position 48 of ubiquitin (K48) for ubiquitination of IPS-1 (1012). STING primarily mediates the innate immune signaling pathway in response to DNA viruses. Several studies demonstrate that STING also mediates the RNA virus-triggered signaling pathway (13, 14). In addition to defense against viral infection, STING is also involved in autoimmunity in humans and mice (15, 16). Thus, the activity of STING needs to be tightly controlled to inhibit excessive autoimmunity and aberrant inflammation in defense against viral infections. Zebrafish IRF10 inhibits STING-mediated IFN production and binds to the IFN-stimulated response element motif of the IFN gene to avoid excessive IFN expression (17).

The Mx family was originally identified and named for its resistance to Myxoviridae (18, 19), and it exists in nearly all vertebrate genomes, from fish to primates. Mx proteins are a family of the GTPase dynamin superfamily, with a molecular mass of 70–80 kDa (20). The primary structure of Mx proteins consists of three domains: the GTPase domain, the central interactive domain, and the GTPase effector domain (GED) with leucine zipper motifs (21). Two Mx genes have been reported in humans and mice (22), whereas three have been described in rats (23). Interestingly, there is high interspecies variability in the number of Mx genes in teleosts, such as seven in zebrafish (24), nine in Oncorhynchus mykiss (25), and three in Salmo salar (26). It is thought that selective pressure against virus infections has expanded the Mx family, resulting in greatly varied gene numbers among fish species. In the 50 years since the discovery of the mouse Mx1 gene, researchers have learned the molecular details of many antiviral mechanisms mediated by Mx proteins, including the unique steps of Mx in the virus life cycle to inhibit various viruses, and the proposed mode of Mx actions (27). Mx-sensitive viruses include bunyaviruses, orthomyxoviruses, paramyxoviruses, rhabdoviruses, togaviruses, picornaviruses, reoviruses, and hepatitis B virus (28). Human MxA and MxB can inhibit the two most famous viruses (influenza A virus and HIV) (29, 30), which has greatly promoted research on animal-derived Mx proteins. The subcellular localizations of Mx proteins contribute to their antiviral specificity (31). Normally, human MxA locates in the cytoplasm and inhibits viral transcription in the cytoplasm (32). Nuclear Mx proteins (e.g., mouse Mx1) defend against viruses that replicate in the nucleus (33, 34). Antiviral activity has also been reported in fish Mx proteins. For example, S. salar encodes three Mx proteins, all of which can suppress infectious salmon anemia virus (26). In our previous study, we also reported three antiviral Mx genes in grass carp (35). However, previous studies mainly focused on the study of Mx structure and its inhibitory effect on viruses (36, 37), but as an ISG molecule, its effect on the immune pathway is still poorly understood.

In this study, a new member of grass carp Mx family, MxG, was found to be highly transcribed in cells and individuals susceptible to grass carp reovirus (GCRV) as compared with those resistant to GCRV (38, 39). Online basic local alignment search tool (BLAST) and phylogenetic tree analyses indicate that MxG is a teleost-specific gene. In this study, we reported that grass carp MxG is a negative regulator in antiviral immunity, degrading IPS-1 through the proteasome pathway, and degrading STING through the lysosome pathway, thereby negatively regulating type I IFN (IFN-I) and NF-κB. The results indicate the negative roles of MxG in IFN production, and that MxG facilitates virus replication, which provide novel insights into the functions of the Mx family.

The resistant/susceptible Ctenopharyngodon idella kidney (CIK) cell lines to GCRV infection were isolated and identified in our previous study (39). The common CIK cells were used in the studies except where described. All of the CIK cells were cultured in DMEM supplemented with 10% FBS (Life Technologies), 100 U/ml penicillin (Sigma), and 100 μg/ml streptomycin (Sigma). Cells were incubated at 28°C with a 5% CO2 humidified atmosphere. Fathead minnow (FHM) cells were cultured at 28°C in 5% CO2 in M199 (Life Technologies) supplemented with 10% FBS (Life Technologies). epithelioma papulosum cyprini (EPC) cells were maintained at 28°C in 5% CO2 in MEM (HyClone) supplemented with 10% FBS (Life Technologies). Human embryonic kidney (HEK) 293T cells were grown at 37°C in 5% CO2 in DMEM (HyClone) supplemented with 10% FBS (Life Technologies). GCRV-GZ1208, a GCRV type I strain, spring viremia of carp virus (SVCV), and cyprinid herpesvirus-2 (CyHV-2) were used for cell infection at a multiplicity of infection of 0.1.

The open reading frame of grass carp MxG and grass carp Mx2 were amplified by PCR and then cloned into pcDNA4.0 (specific primers in Supplemental Table I). Overexpression plasmids IPS-1-hemagglutinin (HA), STING-FLAG, MDA5-HA, RIG-I-HA, TANK-binding kinase 1 (TBK1)-HA, IRF3-Myc, IRF7-Myc, IFN1-enhanced GFP (eGFP), and IFN3-eGFP were constructed as described previously (40). For subcellular localization, the open reading frame of grass carp MxG, IPS-1, and STING were introduced into pCMV-eGFP. MxG, voltage-dependent anion channel 1 (VDAC1), glucose-regulated protein 78 (GRP78), and lysosome-associated membrane protein 2 (LAMP2) were ligated into pDsRed1-C1, and red fluorescent protein (RFP)-GRP78 and RFP-LAMP2 were saved in our laboratory (41, 42). For dual-luciferase reporter assays, the valid promoters MxG and Mx2 were cloned into the pGL3-basic luciferase reporter vector (Promega), and IFN1 promoter-Luc, IFN3 promoter-Luc, NF-κB1 promoter-Luc, NF-κB2 promoter-Luc, IRF3 promoter-Luc, and IRF7 promoter-Luc were saved in our laboratory (43). The site-directed mutagenesis PCR product was ligated into pCMV-eGFP. HA-Ub-K48O plasmid was provided by Prof. Hongbing Shu (Wuhan University, Wuhan, China). FuGENE 6 transfection reagent was purchased from Promega. Transfected small interfering RNA (siRNA) with GP-siRNA-Mate Plus reagents were purchased from GenePharma (Jiangsu, China). MG132 and 3-MA were purchased from Selleck Chemicals and Sigma, respectively. NH4Cl was formulated with NH4Cl powder.

For IB analysis, protein extracts were separated by 8–12% SDS-PAGE gels and transferred onto nitrocellulose membranes (Millipore). The membranes were blocked in fresh 2% BSA dissolved in TBST buffer at 4°C overnight, then incubated with appropriate indicated primary Ab for 2 h at room temperature. They were then washed twice with TBST buffer and incubated with secondary Ab for 45 min at room temperature. After washing three times with TBST buffer, the nitrocellulose membrane was scanned and imaged by an Odyssey CLx imaging system (LI-COR Biosciences) or an ImageQuant (GE Healthcare). The results were obtained from three independent experiments.

For co-IP analysis, CIK cells in 10-cm2 dishes were cotransfected with the indicated plasmids for 48 h. The cells were lysed in IP lysis buffer (20 mM Tris [pH 7.4], 150 mM NaCl, 1% Triton X-100, 1 mM EDTA, 1 mM Na3VO4, 0.5 mg/ml leupeptin, 2.5 mM sodium pyrophosphate) and 1 mM PMSF (Beyotime) for 30 min on ice, and the cellular debris was removed by centrifugation at 12,000 × g for 30 min at 4°C. The supernatant was transferred to a fresh tube and incubated with 1 μg of Ab with gentle shaking overnight at 4°C. Protein A+G-Sepharose beads (Beyotime) (50 μl) were added to the mixture and incubated for 6 h at 4°C. After centrifugation at 3000 × g for 5 min, the beads were collected and washed three times with lysis buffer. Subsequently, the beads were suspended in 20 μl of 2× SDS loading buffer and denatured at 95°C for 10 min, followed by IB detection.

Mouse polyclonal Abs of grass carp IFN1, IFN3, and NF-κB1 were prepared and conserved in our laboratory. The anti-IRF7 rabbit polyclonal antiserum was previously prepared in our laboratory (40, 44). Anti-IRF3 rabbit polyclonal antiserum was previously prepared and presented by Prof. Yibing Zhang (Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan, China). Anti-HA tag (ab18181) mouse mAb, anti-FLAG tag (ab125243) mouse mAb, anti–β-tubulin rabbit polyclonal Ab (ab6046), and anti-Myc tag mouse mAb (ab32) were purchased from Abcam. Anti-GFP mouse mAb (AE012) was purchased from ABclonal. IRDye 800CW donkey anti–rabbit-IgG (926-32213) and anti–mouse-IgG (H+L) (926-32212) secondary Abs were purchased from LI-COR Bioscienes. Goat–anti-mouse Ig-HRP conjugate secondary Ab (A0216) was purchased from Beyotime.

Recombinant grass carp IFN1 protein was prepared in our laboratory as previously described (45).

Samples (1 × 106 cells per well) were infected with virus (multiplicity of infection of 0.1). Supernatants were collected at 24 h, sequentially diluted 10-fold (10−1 to 10−8), and incubated with CIK cells in a flat 96-well plate at 28°C for 7 d to determine the 50% tissue culture–infective dose. On day 7, the plate was examined for the presence of viral cytopathic effect (CPE) under a microscope.

Cells were seeded in a 24-well plate for 24 h. Cotransfection was performed with corresponding overexpression plasmid, target promoter-luciferase plasmid, and internal control reporter vector (pRL-TK). After a 24-h transfection, cells were infected with virus or PBS. Twenty-four hours postchallenge, cells were washed with PBS, then lysed with passive lysis buffer (Promega) for 30 min. Luciferase activities were detected by a Dual-Luciferase reporter assay system (Promega). The luciferase reading was normalized against that in the pRL-TK level, and the relative light unit intensity was presented as the ratio of luciferase of firefly to Renilla.

The samples were homogenized in TRIzol LS reagent (Invitrogen). Total RNA was isolated according to the manufacturer’s instructions and incubated with RNase-free DNase I (Roche) to remove contaminated genomic DNA. qRT-PCR was performed in a Roche LightCycler 480 system, and EF1α was used as an internal control gene for cDNA normalization. The qRT-PCR amplification was carried out in a total volume of 15 μl, containing 7.5 μl of BioEasy Master Mix (SYBR Green) (Hangzhou Bioer Technology), 3.1 μl of nuclease-free water, 4 μl of diluted cDNA (200 ng), and 0.2 μl of each gene-specific primer (10 μM) (Supplemental Table I). The data were analyzed as previously described (46).

CIK cells stably transfecting MxG or empty vector were screened by G418 as previously reported (47). RNA isolation, cDNA library construction, and sequencing were performed by Majorbio Biotech (Shanghai, China). Data analysis was achieved using i-Sanger (https://www.i-sanger.com/). BLAST search and annotation were carried out by the Kyoto Encyclopedia of Genes and Genomes databases, Gene Ontology, Clusters of Orthologous Groups, the non-redundant protein database, Swiss-Prot, and Pfam. Gene expression according to transcriptome was validated by quantitative real-time PCR with specific primers (Supplemental Table I).

Knockdown of MxG in CIK cells was achieved by transfection of siRNA targeting MxG mRNA. siRNA sequences targeting different regions of MxG mRNA were designed by three online siRNA designing software programs (https://www.invivogen.com/sirnawizard/design.php; http://biodev.extra.cea.fr/DSIR/DSIR.html; http://sidirect2.rnai.jp). Sequence similarity was searched against grass carp genome and transcriptomes. Nine siRNA sequences were selected and synthesized by GenePharma (Supplemental Table I). CIK cells were transfected with siRNA using GP-siRNA-Mate Plus reagents (GenePharma) for 36 h. The gene-silencing efficiencies of the siRNA candidates were first evaluated by qRT-PCR, comparing with that in the blank control CIK cells, and control siRNA (siCon, provided by the supplier) was used as a negative control. Then, CIK cells stably transfected MxG-GFP were plated in six-well plates and transfected with siRNAs for protein level tests. The cells were lysed, and proteins were extracted for IB. Furthermore, the optimal three siRNAs were used to investigate the effects on IFN1, similar sequence genes, and VP4 by qRT-PCR. In addition, mRNA and protein levels of MxG were examined in RNA interference cells with the representative optimal siRNA after GCRV infection. Finally, a CPE assay was used to investigate the general effect of MxG knockdown.

For the subcellular localization and colocalization of MxG, CIK cells were cotransfected with the indicated plasmids and plated onto coverslips in 12-well plates for 24 h, then the cells were washed, fixed, and stained as reported previously (43). Finally, images were taken with an UltraVIEW VoX 3D live cell imaging system (PerkinElmer).

Statistical analysis and presentation graphics were carried out using SPSS 16.0 and GraphPad Prism 8.0 software. Results are presented as mean ± SD for at least three independent experiments. All data were subjected to one-way ANOVA, followed by an unpaired, two-tailed t test. A p value <0.05 was considered to be statistically significant (*p < 0.05, **p < 0.01, and ***p < 0.001).

There are more isoforms of Mx in teleosts in contrast to mammals. At present, the function of the Mx family mainly involves antiviral reactions. However, we found that the expression trend of grass carp MxG was opposite to other members of the Mx family in the transcriptome data of grass carp (both individual and cell levels) that are resistant/susceptible to GCRV (38, 39). MxG is highly expressed in susceptible CIK cells, and its expression level is 5-fold higher than that in resistant CIK cells, as confirmed by qRT-PCR (Fig. 1A), so we speculated that it has a particular function. First, MxG is a teleost-specific gene by online BLAST and phylogenetic tree analyses (Supplemental Fig. 1). Then, we used G418 to screen CIK cells stably overexpressing MxG or vector only. The screening effect was detected by qRT-PCR and IB (Fig. 1B). The results showed that the stably transfected cell line expresses MxG more than 40-fold higher than the control group, and this proved that we successfully obtained stably transfected cells. After that, we used stably transfected cell lines to perform GCRV infection experiments, and, as shown in (Fig. 1C and (1D, a stronger CPE was observed in the overexpressed MxG group. Finally, we examined the effect of MxG overexpression on GCRV titer in 96-well plates, as shown in (Fig. 1E. The viral titer results are shown in (Fig. 1F, including the finding that the 50% tissue culture infectious dose in the MxG group was ∼4 orders of magnitude higher than that in the control group. Mx2 possesses substantial antiviral activity and was used as a control in this experiment. CIK cells overexpressing Mx2 were infected with GCRV, and a weaker CPE appeared (Fig. 1G). Then the viral titers in the supernatants of the infected cells were examined by a plaque assay, which showed that the titer in Mx2-overexpressing cells was ∼1 order of magnitude lower than that in control cells (Fig. 1H). The results indicated that MxG promotes GCRV replication and CPEs.

FIGURE 1.

MxG promotes GCRV replication. (A) Expression of MxG in susceptible/resistant cells. MxG mRNA expression in susceptible cells is 5-fold higher than that in resistant cells. (B) The stably transfected CIK cells with MxG-GFP and GFP were obtained by G418 screening. mRNA expressions of MxG in stably transfected cells were detected by qRT-PCR (upper panel) with EF1α gene as an internal control and IB (lower panel) with the β-tubulin gene as the reference. (C and D) CPEs in MxG overexpression cells after viral infection. CIK cells stably expressing MxG-GFP or GFP were infected with GCRV. The cells were observed under an inverted microscope at 48 h postinfection (C), and the supernatants were collected for plaque assays. The surviving cells were fixed with 4% paraformaldehyde and stained with 0.05% (w/v) crystal violet (D). (E and F) Effect of MxG on the titer of GCRV. CIK cells stably expressing MxG-GFP or GFP were inoculated into 6-well plates for 24 h and infected with GCRV. The cell culture supernatants were collected at 36 h postinfection, diluted at the indicated ratio, and the virus titer was measured in a 96-well plate. The cells were fixed with 4% paraformaldehyde and stained with 0.05% (w/v) crystal violet (E). The virus titer results are shown (F). (G and H) Effect of Mx2 on the titer of GCRV. CIK cells seeded in 6-well plates overnight were transiently transfected with 2 μg of Mx2-Myc and empty plasmid, respectively. The transfected cells were infected with GCRV at 24 h posttransfection. The supernatants were collected at 36 h postinfection for titer assays. Other performances were the same as above. All experiments were performed in triplicate. **p < 0.01.

FIGURE 1.

MxG promotes GCRV replication. (A) Expression of MxG in susceptible/resistant cells. MxG mRNA expression in susceptible cells is 5-fold higher than that in resistant cells. (B) The stably transfected CIK cells with MxG-GFP and GFP were obtained by G418 screening. mRNA expressions of MxG in stably transfected cells were detected by qRT-PCR (upper panel) with EF1α gene as an internal control and IB (lower panel) with the β-tubulin gene as the reference. (C and D) CPEs in MxG overexpression cells after viral infection. CIK cells stably expressing MxG-GFP or GFP were infected with GCRV. The cells were observed under an inverted microscope at 48 h postinfection (C), and the supernatants were collected for plaque assays. The surviving cells were fixed with 4% paraformaldehyde and stained with 0.05% (w/v) crystal violet (D). (E and F) Effect of MxG on the titer of GCRV. CIK cells stably expressing MxG-GFP or GFP were inoculated into 6-well plates for 24 h and infected with GCRV. The cell culture supernatants were collected at 36 h postinfection, diluted at the indicated ratio, and the virus titer was measured in a 96-well plate. The cells were fixed with 4% paraformaldehyde and stained with 0.05% (w/v) crystal violet (E). The virus titer results are shown (F). (G and H) Effect of Mx2 on the titer of GCRV. CIK cells seeded in 6-well plates overnight were transiently transfected with 2 μg of Mx2-Myc and empty plasmid, respectively. The transfected cells were infected with GCRV at 24 h posttransfection. The supernatants were collected at 36 h postinfection for titer assays. Other performances were the same as above. All experiments were performed in triplicate. **p < 0.01.

Close modal

Since MxG associates with host antiviral responses, IFNs are critical factors in antiviral immunity, and IFN1 and IFN3 play the major role in grass carp (48); accordingly, the current study mainly investigated these two IFNs. To further explore the potential role of MxG in regulating antiviral responses, we elucidated the effect of MxG on IFN1 and IFN3 in the promoter, transcription, and protein levels to identify it as a positive or negative regulator in IFN-I signaling. After GCRV challenge, MxG significantly inhibited the transcriptional level of IFN1 at all time points (0, 12, 24, and 48 h post-challenge) (Fig. 2A), but it did not inhibit expression of IFN3 (Fig. 2B). Then, we examined IFN-I response using a luciferase reporter. IFN1pro and IFN3pro were transfected into CIK cells. After treatment with GCRV or poly(I:C), the activities of IFN1pro and IFN3pro were significantly upregulated, whereas IFN1pro activity in the MxG group was significantly downregulated compared with the control group (Fig. 2C), but IFN3pro activity was unaffected (Fig. 2D). mRNA expression of MxG was significantly upregulated by recombinant IFN1 protein stimulation (Fig. 2E). The expression in protein level was consistent with the above experimental results. After GCRV infection, the protein expression of endogenous IFN1 was inhibited by MxG (Fig. 2F), whereas IFN3 was unaffected (Fig. 2G). These experiments indicated that MxG negatively regulates IFN1 expression.

FIGURE 2.

MxG inhibits GCRV/poly(I:C)-induced IFN1 signaling. (A and B) MxG overexpression inhibits the transcription of IFN1, but it does not inhibit expression of IFN3. CIK cells stably expressing MxG-GFP or GFP were inoculated into 6-well plates for 12 h and infected with GCRV. mRNA expression levels of IFN1 and IFN3 were examined by qRT-PCR at 0, 6, 12, 24, and 48 h postinfection. (C and D) MxG overexpression suppresses GCRV/poly(I:C)-induced IFN1 promoter activity, but it does not suppress IFN3 promoter activity. CIK cells were seeded in 24-well plates and transfected on the next day with 380 ng of IFN1pro-Luc (C) or IFN3pro-Luc (D) and 38 ng of pRL-TK, plus 380 ng of pcDNA4.0-MxG or pcDNA4.0 (control vector). Cells were untreated (null) or treated with GCRV or poly(I:C) at 24 h posttransfection. Luciferase activities were monitored at 24 h postchallenge. (E) mRNA expression of MxG was upregulated after recombinant grass carp IFN1 stimulation (final concentration 1 μg/ml). (F and G) MxG inhibits the expression of endogenous IFN1 but not IFN3 after GCRV infection. CIK cells were inoculated in 6-well plates for 12 h, transfected with pcDNA4.0-MxG or pcDNA4.0 for 24 h, and infected with GCRV, and cell lysates were used for IB using IFN1 (E) or IFN3 (F) polyclonal antiserum at 36 h after GCRV infection. **p < 0.01, ***p < 0.001.

FIGURE 2.

MxG inhibits GCRV/poly(I:C)-induced IFN1 signaling. (A and B) MxG overexpression inhibits the transcription of IFN1, but it does not inhibit expression of IFN3. CIK cells stably expressing MxG-GFP or GFP were inoculated into 6-well plates for 12 h and infected with GCRV. mRNA expression levels of IFN1 and IFN3 were examined by qRT-PCR at 0, 6, 12, 24, and 48 h postinfection. (C and D) MxG overexpression suppresses GCRV/poly(I:C)-induced IFN1 promoter activity, but it does not suppress IFN3 promoter activity. CIK cells were seeded in 24-well plates and transfected on the next day with 380 ng of IFN1pro-Luc (C) or IFN3pro-Luc (D) and 38 ng of pRL-TK, plus 380 ng of pcDNA4.0-MxG or pcDNA4.0 (control vector). Cells were untreated (null) or treated with GCRV or poly(I:C) at 24 h posttransfection. Luciferase activities were monitored at 24 h postchallenge. (E) mRNA expression of MxG was upregulated after recombinant grass carp IFN1 stimulation (final concentration 1 μg/ml). (F and G) MxG inhibits the expression of endogenous IFN1 but not IFN3 after GCRV infection. CIK cells were inoculated in 6-well plates for 12 h, transfected with pcDNA4.0-MxG or pcDNA4.0 for 24 h, and infected with GCRV, and cell lysates were used for IB using IFN1 (E) or IFN3 (F) polyclonal antiserum at 36 h after GCRV infection. **p < 0.01, ***p < 0.001.

Close modal

Next, the subcellular localization of MxG was investigated. To do so, we constructed eGFP-fused MxG and RFP-fused GRP78/VDAC1 plasmids. The green signal representing MxG completely overlapped with the red signal representing GRP78 (endoplasmic reticulum [ER]) and partially overlapped with the red signal representing VDAC1 (mitochondrion) (Fig. 3A). IPS-1 locates on the mitochondrion and STING localizes on the ER, and thus MxG may be related to IPS-1 and STING in space. IPS-1 and STING are the essential adaptor proteins in the RLR pathway, so we characterized the function of MxG in RLR activation of IFN signaling. CIK cells were cotransfected with vectors overexpressing MxG and RLR molecules of grass carp. We examined the activity of IFN1-Luc by overexpression of the RLR signaling pathway genes, including MDA5, RIG-I, IPS-1, STING, TBK1, IRF3, and IRF7 together with empty vector or MxG-Myc. As shown in (Fig. 3B, IFN1pro activity was induced by the RLRs, and these activations triggered by MDA5, RIG-I, IPS-1, and STING were suppressed by MxG, whereas activity induced by TBK1, IRF3, or IRF7 was unaffected. Taken together, these results suggested that MxG inhibits the RLR signaling pathway at the point of IPS-1 or STING. Thus, MxG downregulates the activities of IPS-1 or STING, leading to overall suppression of the antiviral responses.

FIGURE 3.

MxG plays broadly negative roles in antiviral immune responses. (A) MxG localizes on ERs and mitochondria. CIK cells were cotransfected with MxG-GFP and GRP78-RFP, an ER protein marker, or MxG-GFP and VDAC1-RFP, a mitochondrial outer membrane protein marker, respectively, and seeded on observation dishes for confocal microscopy. The cells were fixed with 4% (v/v) paraformaldehyde and stained with Hoechst 33342 at 48 h. All samples were imaged under a confocal microscope. Green signals represent overexpressed MxG, and red signals stand for overexpressed GRP78 or VDAC1. The blue staining indicates the nucleus. The yellow signals in the merged images indicate the colocalization between MxG and organelles (original magnification, ×63; oil immersion objective). All experiments were repeated at least three times. (B) MxG suppresses IFN1 promoter activity mediated by MDA5, RIG-I, IPS-1, and STING, respectively. CIK cells were seeded in 24-well plates and cotransfected on the next day with 260 ng of IFN1pro-Luc, 26 ng of pRL-TK, and 260 ng of MDA5-, RIG-I-, IPS-1-, STING-, TBK1-, IRF3-, or IRF7-overexpressing plasmid plus 260 ng of pcDNA4.0-MxG or pcDNA4.0 (control vector). The cells were lysed for luciferase activity detection at 36 h posttransfection. Other captions are the same as in (Fig. 2C. (C and D) Transcriptomic data of key genes in NF-κB and JAK-STAT pathways were verified by qRT-PCR. MxG downregulates mRNA expressions of these selected genes, which is consent with the transcriptomic data. (EH) CPE in MxG overexpression cells after viral infection. CIK cells stably expressing MxG-GFP or GFP were inoculated in 6-well plates for 24 h and infected with SVCV and CyHV-2, respectively. The cells were observed at 48 h postinfection under an inverted microscope (E and G), and then the supernatants were collected. CIK cells were plated in 24-well plates for 12 h and treated with the supernatants at different dilution rates. The viable cells were fixed with 4% paraformaldehyde and stained with 0.05% (w/v) crystal violet (F and H) at 36 h. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 3.

MxG plays broadly negative roles in antiviral immune responses. (A) MxG localizes on ERs and mitochondria. CIK cells were cotransfected with MxG-GFP and GRP78-RFP, an ER protein marker, or MxG-GFP and VDAC1-RFP, a mitochondrial outer membrane protein marker, respectively, and seeded on observation dishes for confocal microscopy. The cells were fixed with 4% (v/v) paraformaldehyde and stained with Hoechst 33342 at 48 h. All samples were imaged under a confocal microscope. Green signals represent overexpressed MxG, and red signals stand for overexpressed GRP78 or VDAC1. The blue staining indicates the nucleus. The yellow signals in the merged images indicate the colocalization between MxG and organelles (original magnification, ×63; oil immersion objective). All experiments were repeated at least three times. (B) MxG suppresses IFN1 promoter activity mediated by MDA5, RIG-I, IPS-1, and STING, respectively. CIK cells were seeded in 24-well plates and cotransfected on the next day with 260 ng of IFN1pro-Luc, 26 ng of pRL-TK, and 260 ng of MDA5-, RIG-I-, IPS-1-, STING-, TBK1-, IRF3-, or IRF7-overexpressing plasmid plus 260 ng of pcDNA4.0-MxG or pcDNA4.0 (control vector). The cells were lysed for luciferase activity detection at 36 h posttransfection. Other captions are the same as in (Fig. 2C. (C and D) Transcriptomic data of key genes in NF-κB and JAK-STAT pathways were verified by qRT-PCR. MxG downregulates mRNA expressions of these selected genes, which is consent with the transcriptomic data. (EH) CPE in MxG overexpression cells after viral infection. CIK cells stably expressing MxG-GFP or GFP were inoculated in 6-well plates for 24 h and infected with SVCV and CyHV-2, respectively. The cells were observed at 48 h postinfection under an inverted microscope (E and G), and then the supernatants were collected. CIK cells were plated in 24-well plates for 12 h and treated with the supernatants at different dilution rates. The viable cells were fixed with 4% paraformaldehyde and stained with 0.05% (w/v) crystal violet (F and H) at 36 h. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

CIK cells stably overexpressing MxG or vector only were used for infection experiments. GCRV-infected and uninfected samples were collected, and RNA was extracted for RNA sequencing. All reads have been submitted to the Sequence Read Archive at NCBI (accession number: PRJNA594117). According to the enrichment analysis of the Kyoto Encyclopedia of Genes and Genomes pathway, we found that MxG also associates with the NF-κB and JAK-STAT pathways (Supplemental Fig. 2) and selected some genes from transcriptome data for qRT-PCR verification (Fig. 3C, 3D). The results were basically consistent with the transcriptome sequencing, namely, MxG also inhibits NF-κB and JAK-STAT pathways.

We wondered whether MxG facilitates immune escape of other types of viruses in addition to the dsRNA virus GCRV. We used SVCV, a ssRNA virus (Fig. 3E, 3F), and CyHV-2, a dsDNA virus (Fig. 3G, 3H), to infect CIK cells stably overexpressing MxG or vector only and found that the MxG group exhibited a stronger CPE than did the control. All of the results indicated that MxG can inhibit antiviral immune responses and promote immune escape of multiple viruses.

Because MxG promotes GCRV replication, we speculated that MxG knockdown inhibits GCRV replication. First, we screened out the optimal siRNAs at mRNA and protein levels (Fig. 4A, 4B). The results indicated that the interference efficiencies of Si2, Si4, and Si5 (final concentration 100 nM) are the optimal. Then, we investigated the effects of Si2, Si4, and Si5 on IFN the pathway, including off-target and antiviral activity. Si2, Si4, and Si5 share the highest nucleotide identities with IL-17N (12/21), INSRb (12/21), and SCF5a1 (13/21), respectively. All three optimal siRNAs neither activate the IFN response (Fig. 4C) nor show an off-target effect (Fig. 4D). All of them significantly suppress viral replication (Fig. 4E). Si2 was used as a representative optimal siRNA for subsequent experiments. After that, we monitored the interference effects of MxG at different time points after GCRV infection to ensure the accuracy of the follow-up experiments (Fig. 4F, 4G). There is still effective interference at 48 h after the virus infection. Finally, the supernatants at 48 h after GCRV infection were collected and subjected to plaque assays. The results are consistent with the above, that is, CPEs weaken after MxG knockdown by microscopic observation (Fig. 4H) and crystal violet staining (Fig. 4I).

FIGURE 4.

MxG knockdown enhances antiviral responses. (A and B) Screening the optimal siRNA for MxG knockdown. (A) Nine specific siRNAs of MxG, along with siCon, were transfected into CIK cells, respectively. The cells were harvested for qRT-PCR at 36 h. The transcripts of MxG were normalized to the blank control CIK cells. (B) CIK cells stably expressing MxG-GFP were inoculated into 6-well plates for 12 h and transfected with siRNAs as above, and the cells were harvested for IB at 36 h (left). The relative intensity of IPS-1 to β-tubulin was quantified by ImageJ software (right). (C and D) Si2, Si4, Si5 and siCon were transfected into CIK cells for 36 h. mRNA expressions of IFN1 (C) and similar sequence genes (IL-17N, INSRb, and SCF5a1) (D) were compared with the corresponding genes in the blank control CIK cells. (E) Si2, Si4, Si5, and siCon were transfected into CIK cells for 36 h, then infected with GCRV for 48. mRNA expression levels of VP4 (viral gene) were normalized to the CIK cells infected with GCRV. (F and G) mRNA (F) and protein (G) levels of MxG in RNA interference cells with Si2 for 36 h at different time points after GCRV infection. (H and I) CPE after GCRV infection in MxG knockdown cells with si2. MxG was interfered in CIK cells for 36 h, then infected with GCRV for 48 h, and the supernatants were collected. Other captions are the same as in (Fig. 1D. EF1α gene and β-tubulin protein were used as internal controls for qRT-PCR and IB, respectively. **p < 0.01, ***p < 0.001.

FIGURE 4.

MxG knockdown enhances antiviral responses. (A and B) Screening the optimal siRNA for MxG knockdown. (A) Nine specific siRNAs of MxG, along with siCon, were transfected into CIK cells, respectively. The cells were harvested for qRT-PCR at 36 h. The transcripts of MxG were normalized to the blank control CIK cells. (B) CIK cells stably expressing MxG-GFP were inoculated into 6-well plates for 12 h and transfected with siRNAs as above, and the cells were harvested for IB at 36 h (left). The relative intensity of IPS-1 to β-tubulin was quantified by ImageJ software (right). (C and D) Si2, Si4, Si5 and siCon were transfected into CIK cells for 36 h. mRNA expressions of IFN1 (C) and similar sequence genes (IL-17N, INSRb, and SCF5a1) (D) were compared with the corresponding genes in the blank control CIK cells. (E) Si2, Si4, Si5, and siCon were transfected into CIK cells for 36 h, then infected with GCRV for 48. mRNA expression levels of VP4 (viral gene) were normalized to the CIK cells infected with GCRV. (F and G) mRNA (F) and protein (G) levels of MxG in RNA interference cells with Si2 for 36 h at different time points after GCRV infection. (H and I) CPE after GCRV infection in MxG knockdown cells with si2. MxG was interfered in CIK cells for 36 h, then infected with GCRV for 48 h, and the supernatants were collected. Other captions are the same as in (Fig. 1D. EF1α gene and β-tubulin protein were used as internal controls for qRT-PCR and IB, respectively. **p < 0.01, ***p < 0.001.

Close modal

The function of IPS-1 to activate IFN was negatively regulated by MxG. Therefore, it was necessary to elucidate the relationship between MxG and IPS-1. We speculated that MxG may interact with IPS-1 to exert its inhibitory effect. First, we performed a colocalization assay. RFP-MxG and IPS-1-GFP were cotransfected to CIK cells, and their colocalization was observed by confocal microscopy (Fig. 5A). Subsequently, we conducted co-IP assays to evaluate the potential interaction between MxG and IPS-1. EPC cells were cotransfected with MxG-Myc and IPS-1-HA. As shown in (Fig. 5B (left), MxG-Myc could be pulled down by IPS-1-HA using anti-HA agarose beads. Similarly, IPS-1-HA could also be pulled down by MxG-Myc using anti-Myc agarose beads (Fig. 5B, right). MxG interacts with IPS-1. How does MxG affect IPS-1? First, we examined whether MxG affects IPS-1 expression at the transcriptional level. As shown in (Fig. 5C, MxG did not inhibit but promoted the transcription of IPS-1. Next, we tested whether MxG affects IPS-1 expression at the protein level. With the overexpression of MxG, IPS-1 protein was decreased in a dose-dependent manner (Fig. 5D). MG132, a potent proteasome inhibitor, was used to treat cells for 8 h to further investigate the degradation pattern of IPS-1. Interestingly, the expression of IPS-1 protein was rescued with MG132 (Fig. 5E), indicating that IPS-1 may be degraded in a ubiquitin-proteasome manner. Furthermore, IPS-1-GFP, HA-Ub-K48O, and MxG-Myc or not, and were cotransfected in the presence of MG132. Following IP of IPS-1-GFP, IB analysis revealed that the MxG promotes the ubiquitination (K48) of IPS-1 (Fig. 5F). Taken together, MxG inhibits IPS-1 at the protein level, not the transcription level. MxG interacts with IPS-1 to promote its degradation by the ubiquitin-proteasome pathway.

FIGURE 5.

MxG degrades IPS-1 protein through the ubiquitin-proteasomal pathway. (A) MxG colocalizes with IPS-1. CIK cells were cotransfected with RFP-MxG and IPS-1-GFP, and seeded on observation dishes for confocal microscopy. Other captions are the same as in (Fig. 3A. (B) MxG interacts with IPS-1. Left, FHM cells were cotransfected with MxG-Myc/vector and IPS-1-HA for 36 h. Co-IP was performed with anti-HA mAb and mouse IgG (control), and IB with the corresponding Abs. Right, FHM cells were cotransfected with IPS-1-HA/vector and MxG-Myc for 36 h. Co-IP was performed with anti-Myc mAb and mouse IgG (control), and IB was performed with the corresponding Abs. (C) MxG facilitates IPS-1 at the transcription level. CIK cells were transfected with MxG or empty vector and infected with GCRV, and mRNA expression levels of IPS-1 were detected by qRT-PCR. (D) MxG overexpression degrades IPS-1 protein in a dose-dependent manner. HEK 293T cells were seeded in 6-well plates overnight and cotransfected with IPS-1-GFP (2 μg), MxG-GFP (0, 1, 2 μg), and empty vector (2, 1, 0 μg) for 24 h. The cell lysates were subjected to IB with anti-GFP and anti–β-tubulin Abs. The relative intensity of IPS-1 to β-tubulin was quantified by ImageJ software. (E) IPS-1 protein degradation induced by MxG was rescued in the presence of MG132. HEK 293T cells were cotransfected with 2 μg of IPS-1-GFP and 2 μg of MxG-Myc or empty vector for 24 h, then treated with DMSO or MG132 for 6 h. Subsequently, the cells were harvested for IB with anti-GFP and anti–β-tubulin Abs, respectively. The relative intensity of IPS-1 to β-tubulin was quantified by ImageJ software. (F) MxG promotes the ubiquitination (K48) of IPS-1 protein. EPC cells were transfected with 4 μg of IPS-1-GFP, 3 μg of MxG-Myc or empty vector, and 1 μg of HA-Ub-K48O for 24 h, then treated with MG132 for 6 h. Cell lysates were carried out with IP with the anti–GFP-agarose conjugate and IB with anti-HA, anti-GFP, and anti-Myc Abs, respectively. *p < 0.05, **p < 0.01.

FIGURE 5.

MxG degrades IPS-1 protein through the ubiquitin-proteasomal pathway. (A) MxG colocalizes with IPS-1. CIK cells were cotransfected with RFP-MxG and IPS-1-GFP, and seeded on observation dishes for confocal microscopy. Other captions are the same as in (Fig. 3A. (B) MxG interacts with IPS-1. Left, FHM cells were cotransfected with MxG-Myc/vector and IPS-1-HA for 36 h. Co-IP was performed with anti-HA mAb and mouse IgG (control), and IB with the corresponding Abs. Right, FHM cells were cotransfected with IPS-1-HA/vector and MxG-Myc for 36 h. Co-IP was performed with anti-Myc mAb and mouse IgG (control), and IB was performed with the corresponding Abs. (C) MxG facilitates IPS-1 at the transcription level. CIK cells were transfected with MxG or empty vector and infected with GCRV, and mRNA expression levels of IPS-1 were detected by qRT-PCR. (D) MxG overexpression degrades IPS-1 protein in a dose-dependent manner. HEK 293T cells were seeded in 6-well plates overnight and cotransfected with IPS-1-GFP (2 μg), MxG-GFP (0, 1, 2 μg), and empty vector (2, 1, 0 μg) for 24 h. The cell lysates were subjected to IB with anti-GFP and anti–β-tubulin Abs. The relative intensity of IPS-1 to β-tubulin was quantified by ImageJ software. (E) IPS-1 protein degradation induced by MxG was rescued in the presence of MG132. HEK 293T cells were cotransfected with 2 μg of IPS-1-GFP and 2 μg of MxG-Myc or empty vector for 24 h, then treated with DMSO or MG132 for 6 h. Subsequently, the cells were harvested for IB with anti-GFP and anti–β-tubulin Abs, respectively. The relative intensity of IPS-1 to β-tubulin was quantified by ImageJ software. (F) MxG promotes the ubiquitination (K48) of IPS-1 protein. EPC cells were transfected with 4 μg of IPS-1-GFP, 3 μg of MxG-Myc or empty vector, and 1 μg of HA-Ub-K48O for 24 h, then treated with MG132 for 6 h. Cell lysates were carried out with IP with the anti–GFP-agarose conjugate and IB with anti-HA, anti-GFP, and anti-Myc Abs, respectively. *p < 0.05, **p < 0.01.

Close modal

RFP-MxG and STING-GFP were cotransfected into CIK cells, and their colocalization was observed by confocal microscopy. As shown in (Fig. 6A, MxG collocated with STING. Then, co-IP assays between MxG and STING were carried out. MxG-Myc could be pulled down by STING-FLAG using anti-FLAG agarose beads (Fig. 6B, left), and vice versa (Fig. 6B, right). MxG interacts with STING. We examined the effect of MxG on the STING transcription level. The result was similar to IPS-1, that is, MxG promotes STING transcription (Fig. 6C). Next, we tested whether MxG has any effect on STING at the protein level. With the MxG increase, the degradation of STING was more evident (Fig. 6D). First, MG132 was used to rescue the degradation of STING but failed to do so (Fig. 6E). Then, the cells were treated with other indicated inhibitors. The MxG-mediated degradation of STING was completely inhibited by the lysosomal inhibitor NH4Cl, but it was not inhibited by 3-MA (the inhibitor for the autophagosome pathway) (Fig. 6E). STING is a critical mediator of innate antiviral responses. However, many viruses have mechanisms to inhibit STING. In this experiment, the expression of STING protein was inhibited or degraded at the protein level after GCRV infection (Fig. 6F), and MxG enhanced this phenomenon (Fig. 6G). Finally, the phosphorylation of STING mediated by TBK1 is an important step in IFN signal transduction. As shown in (Fig. 6H, phosphorylation of STING mediated by TBK1 was inhibited by MxG. In summary, MxG enhances GCRV immune escape by targeting STING and inducing lysosome-dependent degradation.

FIGURE 6.

MxG mediates STING protein degradation through the lysosomal pathway. (A) MxG colocalizes with STING. CIK cells were cotransfected with RFP-MxG and STING-GFP and seeded on observation dishes for confocal microscopy. Other captions are the same as in (Fig. 3A. (B) MxG interacts with STING. Left, FHM cells were cotransfected with MxG-Myc/vector and STING-FLAG for 36 h. Co-IP was performed with anti-FLAG mAb and mouse IgG (control), and IB with the corresponding Abs. Right, FHM cells were cotransfected with STING-FLAG/vector and MxG-Myc for 36 h. Co-IP was performed with anti-Myc mAb and mouse IgG (control), and IB with the corresponding Abs. (C) MxG facilitates STING at the transcription level. CIK cells were transfected with MxG or empty vector, and infected with GCRV. mRNA expression levels of STING were detected by qRT-PCR. (D) MxG overexpression degrades STING protein in a dose-dependent manner. EPC cells were seeded in 6-well plates overnight and cotransfected with STING-GFP (2 μg), MxG-GFP (0, 1, 2 μg), and empty vector (2, 1, 0 μg) for 24 h. The cell lysates were subjected to IB with anti-GFP and anti–β-tubulin Abs, respectively. The relative intensity of STING to β-tubulin was quantified by ImageJ software. (E) Effects of inhibitors on MxG-mediated degradation of STING protein. EPC cells were seeded in 6-well plates overnight and cotransfected with the indicated plasmids (2 μg each) for 24 h. The cells were treated with the indicated inhibitors for 6 h, then carried out with IB assays. The relative intensity of STING to β-tubulin was quantified by ImageJ software. The working concentrations of MG132, NH4Cl, and 3-MA are 25 μM, 15 mM, and 5 mM, respectively. (F) GCRV inhibits the protein expression of STING. EPC cells were seeded in 6-well plates overnight and transfected with 2 μg of STING-GFP for 24 h, infected or not infected with GCRV for 24 h, and then the STING protein level was investigated. (G) MxG promotes the degradation of STING by GCRV infection. EPC cells were cotransfected with 2 μg of STING-GFP and 2 μg of empty vector or MxG-GFP for 24 h, infected with GCRV for another 24 h, and then STING protein expression levels were examined. (H) MxG overexpression inhibits STING phosphorylation mediated by TBK1. HEK 293T cells were seeded in 6-well plates overnight and cotransfected with 1.5 μg of TBK1-HA, 1.5 μg of STING-GFP, and 1.5 μg of the empty vector or MxG-Myc for 24 h. The lysates were then subjected to IB with anti-HA, anti-Myc, anti-GFP, and anti–β-tubulin Abs, respectively. *p < 0.05, **p < 0.01.

FIGURE 6.

MxG mediates STING protein degradation through the lysosomal pathway. (A) MxG colocalizes with STING. CIK cells were cotransfected with RFP-MxG and STING-GFP and seeded on observation dishes for confocal microscopy. Other captions are the same as in (Fig. 3A. (B) MxG interacts with STING. Left, FHM cells were cotransfected with MxG-Myc/vector and STING-FLAG for 36 h. Co-IP was performed with anti-FLAG mAb and mouse IgG (control), and IB with the corresponding Abs. Right, FHM cells were cotransfected with STING-FLAG/vector and MxG-Myc for 36 h. Co-IP was performed with anti-Myc mAb and mouse IgG (control), and IB with the corresponding Abs. (C) MxG facilitates STING at the transcription level. CIK cells were transfected with MxG or empty vector, and infected with GCRV. mRNA expression levels of STING were detected by qRT-PCR. (D) MxG overexpression degrades STING protein in a dose-dependent manner. EPC cells were seeded in 6-well plates overnight and cotransfected with STING-GFP (2 μg), MxG-GFP (0, 1, 2 μg), and empty vector (2, 1, 0 μg) for 24 h. The cell lysates were subjected to IB with anti-GFP and anti–β-tubulin Abs, respectively. The relative intensity of STING to β-tubulin was quantified by ImageJ software. (E) Effects of inhibitors on MxG-mediated degradation of STING protein. EPC cells were seeded in 6-well plates overnight and cotransfected with the indicated plasmids (2 μg each) for 24 h. The cells were treated with the indicated inhibitors for 6 h, then carried out with IB assays. The relative intensity of STING to β-tubulin was quantified by ImageJ software. The working concentrations of MG132, NH4Cl, and 3-MA are 25 μM, 15 mM, and 5 mM, respectively. (F) GCRV inhibits the protein expression of STING. EPC cells were seeded in 6-well plates overnight and transfected with 2 μg of STING-GFP for 24 h, infected or not infected with GCRV for 24 h, and then the STING protein level was investigated. (G) MxG promotes the degradation of STING by GCRV infection. EPC cells were cotransfected with 2 μg of STING-GFP and 2 μg of empty vector or MxG-GFP for 24 h, infected with GCRV for another 24 h, and then STING protein expression levels were examined. (H) MxG overexpression inhibits STING phosphorylation mediated by TBK1. HEK 293T cells were seeded in 6-well plates overnight and cotransfected with 1.5 μg of TBK1-HA, 1.5 μg of STING-GFP, and 1.5 μg of the empty vector or MxG-Myc for 24 h. The lysates were then subjected to IB with anti-HA, anti-Myc, anti-GFP, and anti–β-tubulin Abs, respectively. *p < 0.05, **p < 0.01.

Close modal

The interaction of viral RNA with RIG-I and MDA5 induces a conformational change that enables interaction with IPS-1 and STING, facilitates downstream signaling molecules (such as IRF3, IRF7, and NF-κB), and produces effectors for host defense. We wondered whether MxG suppresses activation of signaling factors IRF3/7 or NF-κB. After GCRV infection, IRF3pro and IRF7pro promoter activities were significantly induced in the control group, whereas they were impaired significantly in the MxG group (Fig. 7A, 7B). IFN activation by phosphorylated IRF3 and IRF7 is a pivotal process during host antiviral infection (49). The effects of MxG on phosphorylation of IRF3 and IRF7 were detected. As shown in (Fig. 7C–F, the phosphorylation of IRF3 and IRF7 increased at 12 h after GCRV infection and decreased slightly at 36 h postinfection in control group. Phosphorylation levels of IRF3 and IRF7 in the MxG group were consistently lower than those in controls. In addition, MxG inhibits NF-κB at the promoter and protein levels (Fig. 7G–I). Taken together, these results suggested that MxG negatively regulates activation of IRF3/7 and NF-κB.

FIGURE 7.

MxG inhibits IRF3/7 and NF-κB. (A and B) MxG overexpression suppresses IRF3 and IRF7 promoter activities. CIK cells were inoculated in 24-well plates for 12 h, and transfected with 380 ng of IRF3/7pro-Luc, and 38 ng of pRL-TK, plus 380 ng of pcDNA4.0-MxG or pcDNA4.0 for 24 h. Cells were untreated (null) or treated with GCRV for 24 h. Luciferase activities were monitored. (CF) MxG reduced the IRF3 and IRF7 phosphorylation. CIK cells were inoculated in 6-well plates for 12 h, transfected with MxG or vector for 24 h, and infected with GCRV. The cell lysates were subjected to IB with anti-IRF7 polyclonal antiserum, anti-GFP, and anti–β-tubulin Abs (C) at 0, 12, and 36 h postinfection. FHM cells were seeded into 6-well plates overnight and cotransfected with 1.5 μg of IRF3-Myc and 1.5 μg of (MxG-GFP or vector) for 24 h, and infected with GCRV (0, 12, and 36 h). The lysates were then detected by IB with anti-Myc, anti-GFP, and anti–β-tubulin Abs (E). The relative intensity of phosphorylated IRF7 or phosphorylated IRF3 to β-tubulin was quantified by ImageJ software (D and F). (G and H) MxG overexpression suppresses NF-κB1 and NF-κB2 promoter activities. The captions are same as above. (I) MxG inhibits the expression of endogenous NF-κB1 at the protein level. The captions are the same as in (Fig. 2F. **p < 0.01, ***p < 0.001.

FIGURE 7.

MxG inhibits IRF3/7 and NF-κB. (A and B) MxG overexpression suppresses IRF3 and IRF7 promoter activities. CIK cells were inoculated in 24-well plates for 12 h, and transfected with 380 ng of IRF3/7pro-Luc, and 38 ng of pRL-TK, plus 380 ng of pcDNA4.0-MxG or pcDNA4.0 for 24 h. Cells were untreated (null) or treated with GCRV for 24 h. Luciferase activities were monitored. (CF) MxG reduced the IRF3 and IRF7 phosphorylation. CIK cells were inoculated in 6-well plates for 12 h, transfected with MxG or vector for 24 h, and infected with GCRV. The cell lysates were subjected to IB with anti-IRF7 polyclonal antiserum, anti-GFP, and anti–β-tubulin Abs (C) at 0, 12, and 36 h postinfection. FHM cells were seeded into 6-well plates overnight and cotransfected with 1.5 μg of IRF3-Myc and 1.5 μg of (MxG-GFP or vector) for 24 h, and infected with GCRV (0, 12, and 36 h). The lysates were then detected by IB with anti-Myc, anti-GFP, and anti–β-tubulin Abs (E). The relative intensity of phosphorylated IRF7 or phosphorylated IRF3 to β-tubulin was quantified by ImageJ software (D and F). (G and H) MxG overexpression suppresses NF-κB1 and NF-κB2 promoter activities. The captions are same as above. (I) MxG inhibits the expression of endogenous NF-κB1 at the protein level. The captions are the same as in (Fig. 2F. **p < 0.01, ***p < 0.001.

Close modal

Given that MxG plays a critical role in the RLR pathway, we attempted to elucidate the regulatory mechanism. First, we investigated the promoter activity and mRNA expression profiles of MxG after virus (GCRV, SVCV, CyHV-2) and IFN (IFN1 and IFN3) stimulation. As shown in (Fig. 8A and (8B, the promoter activity and mRNA expression of MxG were significantly increased after virus or IFN challenge. Then, we analyzed the transcription factor binding sites in the 5'-flanking sequence of MxG with ALGGEN PROMO software, and the results are shown in Supplemental Fig. 3, which contains one IFN-stimulated response element and one NF-κB1, two IRF7, three IRF3, one IRF2, and one STAT6 binding sites. In order to verify whether these predicted transcription factors really regulate the expression of MxG, we constructed the corresponding overexpression vectors and performed dual-luciferase reporter experiments. The results showed that IRF3 and IRF7 activate the promoter activity of MxG, whereas IRF2 inhibits MxG promoter activity (Fig. 8C). Overexpression of MxG promotes viral gene transcription (Fig. 8D–H), whereas Mx2 overexpression inhibits viral gene transcription (Fig. 8I). MxG knockdown reduces mRNA expression of viral genes (Fig. 8J, 8K). Collectively, viruses and IFNs enhance MxG expression, and the newly expressed MxG negatively modulates antiviral immune responses, suppressing IFN and facilitating viral replication.

FIGURE 8.

Viruses, IFNs, and IRF3/7 induce MxG expression and MxG facilitates viral replication. (A and B) Viruses and IFNs activate MxG. The promoter activity and transcriptional level of MxG were upregulated under GCRV, SVCV, and CyHV-2 challenge and IFN1/IFN3 overexpression. CIK cells were seeded in 6-well plates overnight and infected with GCRV, SVCV, CyHV-2, or stimulated with IFN1 and IFN3 overexpression plasmids, respectively. qRT-PCR assays were performed at 0, 12, 24, and 48 h poststimulation, and dual-luciferase reporter tests were carried out at 48 h poststimulation. (C) Transcriptional regulation of MxG. CIK cells were seeded into 24-well plates overnight and cotransfected with 380 ng of MxGpro-Luc and 38 ng of pRL-TK, plus 380 ng of pcDNA4.0-NF-κB1, pcDNA4.0-STAT6, pcDNA4.0-IRF2, pcDNA4.0-IRF3, pcDNA4.0-IRF7, or pcDNA4.0 (control vector), respectively. Luciferase activities were monitored at 36 h posttransfection. (DH) MxG promotes viral proliferation. CIK cells stably expressing MxG-GFP or GFP were inoculated in 6-well plates for 12 h and infected with GCRV. mRNA expression levels of VP4, VP5, VP6, VP7, NS38 were examined at 6, 12, 24, and 48 h postinfection. (I) Mx2 inhibits viral proliferation. CIK cells were seeded in 6-well plates overnight, transfected with 2 μg of Mx2-Myc or empty plasmid for 24 h, and infected with GCRV for 24 h. qRT-PCR assays were performed. (J and K) MxG knockout leads to the downregulation of viral genes. CIK cells were seeded in 6-well plates overnight, transfected with Si2 or siCon for 36 h, and infected with GCRV. mRNA expression levels of VP4, VP5, VP6, VP7, and NS38 were examined at 12 and 48 h postinfection. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 8.

Viruses, IFNs, and IRF3/7 induce MxG expression and MxG facilitates viral replication. (A and B) Viruses and IFNs activate MxG. The promoter activity and transcriptional level of MxG were upregulated under GCRV, SVCV, and CyHV-2 challenge and IFN1/IFN3 overexpression. CIK cells were seeded in 6-well plates overnight and infected with GCRV, SVCV, CyHV-2, or stimulated with IFN1 and IFN3 overexpression plasmids, respectively. qRT-PCR assays were performed at 0, 12, 24, and 48 h poststimulation, and dual-luciferase reporter tests were carried out at 48 h poststimulation. (C) Transcriptional regulation of MxG. CIK cells were seeded into 24-well plates overnight and cotransfected with 380 ng of MxGpro-Luc and 38 ng of pRL-TK, plus 380 ng of pcDNA4.0-NF-κB1, pcDNA4.0-STAT6, pcDNA4.0-IRF2, pcDNA4.0-IRF3, pcDNA4.0-IRF7, or pcDNA4.0 (control vector), respectively. Luciferase activities were monitored at 36 h posttransfection. (DH) MxG promotes viral proliferation. CIK cells stably expressing MxG-GFP or GFP were inoculated in 6-well plates for 12 h and infected with GCRV. mRNA expression levels of VP4, VP5, VP6, VP7, NS38 were examined at 6, 12, 24, and 48 h postinfection. (I) Mx2 inhibits viral proliferation. CIK cells were seeded in 6-well plates overnight, transfected with 2 μg of Mx2-Myc or empty plasmid for 24 h, and infected with GCRV for 24 h. qRT-PCR assays were performed. (J and K) MxG knockout leads to the downregulation of viral genes. CIK cells were seeded in 6-well plates overnight, transfected with Si2 or siCon for 36 h, and infected with GCRV. mRNA expression levels of VP4, VP5, VP6, VP7, and NS38 were examined at 12 and 48 h postinfection. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

Which domains or sites contribute to the negative regulatory function? The plasmids overexpressing full-length MxG and its domains were constructed (Fig. 9A) and transfected into FHM cells. GCRV infected the overexpression cells, after which the supernatants were collected for CPE (Fig. 9B). All of the constructs promote viral replication, but constructs ΔGTPase-MxG and ΔGED-MxG have relatively weakly promoting effects. The results indicated that GTPase and GEDs contribute to the negative regulatory function. The amino acid 645E in human MxA contributes greatly to its antiviral effect (50). The R614E mutation of mouse Mx1 protein strengthens the antiviral ability against swine fever virus (51). The amino acid 625A (in GED) in grass carp MxG was different from the corresponding known amino acid 645E in human MxA through sequence alignment (Fig. 9C). Therefore, the mutant A625E of MxG was constructed to test the effect on virus replication (Fig. 9D). The results showed that the mutant A625E of MxG weakens the promotion effect of full-length MxG on virus replication.

FIGURE 9.

The domains and site of MxG for negative regulatory function. (A) Schematic presentation of full-length MxG and its domain constructs. MxG consists of the GTPase domain, central interactive domain (CID), and GED. (B) The plasmids of full-length MxG and its constructs were transfected into FHM cells for 24 h, then the cells were infected with GCRV for 24 h. The cell supernatants were collected for CPE. (C) The protein sequences of human MxA, mouse Mx1, and grass carp MxG were aligned to find out the corresponding site in grass carp MxG to human MxA 645E and mouse Mx1 614R. (D) The plasmids of full-length MxG and A625E-MxG mutant were transfected into FHM cells for 24 h, and the cells were challenged with GCRV for 24 h. The supernatants were gathered for CPE.

FIGURE 9.

The domains and site of MxG for negative regulatory function. (A) Schematic presentation of full-length MxG and its domain constructs. MxG consists of the GTPase domain, central interactive domain (CID), and GED. (B) The plasmids of full-length MxG and its constructs were transfected into FHM cells for 24 h, then the cells were infected with GCRV for 24 h. The cell supernatants were collected for CPE. (C) The protein sequences of human MxA, mouse Mx1, and grass carp MxG were aligned to find out the corresponding site in grass carp MxG to human MxA 645E and mouse Mx1 614R. (D) The plasmids of full-length MxG and A625E-MxG mutant were transfected into FHM cells for 24 h, and the cells were challenged with GCRV for 24 h. The supernatants were gathered for CPE.

Close modal

To our knowledge, our findings provide the first evidence of the negative regulatory role of Mx family members in antiviral signaling and help us understand the balancing mechanism of IFN-I. The Mx family has extensive and strong antiviral effects. Human MxA and MxB can inhibit influenza A virus and HIV (29, 30), respectively. We found the opposite functional members of the Mx family. MxG promotes GCRV proliferation (Fig. 1) by inhibiting IFN1 (Fig. 2), and MxG knockdown enhances antiviral responses and inhibits GCRV replication (Fig. 4). Multiple viruses can use MxG for immune escape (Fig. 3E–H), which suggested that MxG plays an important role in the maintenance of immune homeostasis. In this study, we demonstrated that MxG negatively regulates the antiviral functions of IPS-1 and STING (Figs. 5, 6), suppressing IFN1 and NF-κB responses and facilitating viral replication (Figs. 1, 2, 3E–H, 7G–I). IPS-1 and STING are critical for host defense against viruses, and their activation must be under a tight regulation, otherwise aberrant immune responses may occur, leading to autoimmune or chronic inflammatory diseases (52, 53).

ISG56, optineurin, receptor for globular head domain of complement component C1q (gC1qR), LGP2 and sterile α, TIR motif containing 1 (SARM1), nucleotide-binding leucine-rich repeat containing X1 (NLRX1), and dihydroxyacetone kinase negatively regulate the antiviral responses by interfering with interactions between signaling components (5460), whereas ISG15, A20, ring-finger protein 125 (RNF125), Triad3A, PCBP2, and Ro52 induce degradation of signaling factors via the ubiquitin-proteasome system (12, 6165). Protein degradation is the main strategy involved in modulating protein functions in biological processes, and there are three main systems for protein degradation, that is, the ubiquitin-proteasomal, lysosomal, and autophagosomal pathways (66). In this study, MxG degrades IPS-1 through the ubiquitin-proteasomal pathway (Fig. 5E, 5F). Ubiquitination is a critical posttranslational modification that modulates innate immune signals. E3 ubiquitin-protein ligases (E3s) recognize and bind the target protein and catalyze the transfer of ubiquitin to lysine residue on the target protein (67). There are three types of E3s (homologous to the E6-AP C terminus [HECT] E3s, really interesting new gene [RING] E3s, and U-box E3s) (68, 69). MxG has no classical domain of the existing E3s via SMART software. Hence, whether MxG acts as a new unknown E3 or recruits another E3 to catalyze the ubiquitination (K48) of IPS-1 deserves further research. In this study, MxG degrades STING by the lysosomal pathway (Fig. 3E). The lysosome-dependent degradation is another important degradation system of cellular proteins (70). The lysosomal machinery interfaces with most cellular stress-response pathways, including those involved in controlling immune responses and inflammation. Thus, lysosome proteins may function as a central fulcrum that balances the beneficial and harmful effects of the host response to infection and other immunological stimuli (71).

According to the transcriptome data, MxG links to the NF-κB and JAK-STAT pathways. We performed preliminary validation by qRT-PCR. MxG downregulates some important molecules in the NF-κB and JAK-STAT pathways (Fig. 3C, 3D). Furthermore, MxG also promotes the replication of ssRNA virus (SVCV) and dsDNA virus CyHV-2 (Fig. 3E–H). These findings indicated that MxG has broad negative roles in antiviral immunity and anti-inflammatory responses. The GTPase and GEDs contribute to the negative regulatory function (Fig. 9A, 9B). The amino acid 645E in human MxA is very important in the antiviral activity (50). When the corresponding site 614R in mouse Mx1 is mutated to E, the mutant mouse Mx1 enhances the antiviral activity against classical swine fever virus (51). Therefore, we aligned the sequences of grass carp MxG, human MxA, and mouse Mx1 (Fig. 9C) and mutated A625E in grass carp MxG. The mutation attenuates the ability of facilitating virus replication (Fig. 9D). The site 625A in grass carp MxG might contribute to its negative regulatory role.

MxG expression is induced by viruses or IFNs (Fig. 8A, 8B), then inhibits IFN1 and promotes viral replication (Figs. 1–(3). IRF3 and IRF7 activate the promoter activity of MxG, whereas IRF2 inhibits it (Fig. 8C). As an ISG molecule, MxG negative feedback regulates the antiviral response of IFN to avoid the damage caused by excessive immune reaction. Based on our results and the literature, we proposed the working model to illustrate how MxG negatively regulates IPS-1- and STING-mediated IFN1 signaling (Fig. 10).

FIGURE 10.

A schematic working model illustrates how MxG negatively regulates antiviral responses. Upon DNA or RNA viruses infect, CDSs-STING and RLR signaling pathways respond, respectively. RLRs recognize viral RNA and interact with the adaptor protein IPS-1. CDSs bind viral DNA and interact with the adaptor protein STING. These interactions trigger downstream signaling, resulting in the production of IFNs and cytokines, which ultimately induce the MxG through the JAK-STAT pathway (72). IRF2 suppresses the transcriptional activation of IFN (73), which leads to the downregulation of MxG expression. Meanwhile, IRF2 also inhibits the promoter activity of MxG. The newly synthesized MxG is then recruited to the ER and mitochondria. MxG interacts with STING at the ERs and mediates its degradation by the lysosomal pathway. MxG interacts with IPS-1 at the mitochondria and mediates the degradation of the IPS-1 via the ubiquitin-proteasome pathway. Thus, MxG inhibits the signal transmission via IPS-1 or STING to TBK1, thereby inhibiting IRF3/7 phosphorylation and reducing IFN production. MxG also inhibits the signal from IPS-1 to NF-κB, leading to reduction in cytokines. In addition, MxG downregulates the expression of STAT1. In summary, MxG acts as a negative feedback regulator in the antiviral immune responses.

FIGURE 10.

A schematic working model illustrates how MxG negatively regulates antiviral responses. Upon DNA or RNA viruses infect, CDSs-STING and RLR signaling pathways respond, respectively. RLRs recognize viral RNA and interact with the adaptor protein IPS-1. CDSs bind viral DNA and interact with the adaptor protein STING. These interactions trigger downstream signaling, resulting in the production of IFNs and cytokines, which ultimately induce the MxG through the JAK-STAT pathway (72). IRF2 suppresses the transcriptional activation of IFN (73), which leads to the downregulation of MxG expression. Meanwhile, IRF2 also inhibits the promoter activity of MxG. The newly synthesized MxG is then recruited to the ER and mitochondria. MxG interacts with STING at the ERs and mediates its degradation by the lysosomal pathway. MxG interacts with IPS-1 at the mitochondria and mediates the degradation of the IPS-1 via the ubiquitin-proteasome pathway. Thus, MxG inhibits the signal transmission via IPS-1 or STING to TBK1, thereby inhibiting IRF3/7 phosphorylation and reducing IFN production. MxG also inhibits the signal from IPS-1 to NF-κB, leading to reduction in cytokines. In addition, MxG downregulates the expression of STAT1. In summary, MxG acts as a negative feedback regulator in the antiviral immune responses.

Close modal

In summary, MxG interacts with IPS-1 and STING and degrades IPS-1 and STING through the ubiquitin-proteasomal and lysosomal pathways, respectively. MxG also inhibits the phosphorylation of STING and IRF3/7, and it ultimately blunts the IFN and NF-κB expressions, reducing the resistance to viral infection. This finding provides a clue for screening highly effective MxG-specific inhibitors, which may facilitate the development of antiviral therapy to combat piscine viruses. In addition, to our knowledge, this is the first discovery of a negative regulatory factor in the Mx family, which will provide an important reference for other species to explore new members or new functions of the Mx family.

We thank Prof. Hong-bing Shu (Wuhan University, Wuhan, China) for providing HA-Ub-K48O plasmid and Prof. Yibing Zhang (Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan, China) for the IRF3 antiserum. We thank Prof. Qing Wang (Pearl River Fisheries Research Institute, Chinese Academy of Fishery Sciences, Guangzhou, China) for kindly providing the GCRV-GZ1208 strain, Prof. Xueqin Liu (Huazhong Agricultural University, Wuhan, China) for the SVCV virus, and Dr. Junfa Yuan (Huazhong Agricultural University, Wuhan, China) for the CyHV-2 virus. We also thank Dr. Xiaoling Liu, Dr. Gailing Yuan, Dr. Youliang Rao, Dr. Quanyuan Wan, Jianfei Ji, and Chuang Xu for discussions and help in the experiments.

This work was supported by the National Natural Science Foundation of China (31930114 and 31873044).

The sequence presented in this article has been submitted to GenBank (https://www.ncbi.nlm.nih.gov/nuccore/MN807245) under accession number MN807245, and transcriptomic data have been submitted to the National Center for Biotechnology Information’s Sequence Read Archive (https://www.ncbi.nlm.nih.gov/sra/?term=PRJNA594117) under BioProject accession number PRJNA594117.

The online version of this article contains supplemental material.

Abbreviations used in this article

CDS

cytosolic DNA sensor

CIK

Ctenopharyngodon idella kidney

co-IP

coimmunoprecipitation

CPE

cytopathic effect

CyHV-2

cyprinid herpesvirus-2

E3

E3 ubiquitin-protein ligase

EPC

epithelioma papulosum cyprini

FHM

fathead minnow

GCRV

grass carp reovirus

GED

GTPase effector domain

GRP78

glucose-regulated protein 78

HEK

human embryonic kidney

IB

immunoblotting

IFN-I

type I IFN

IPS-1

IFN-β promoter stimulator-1

IRF

IFN regulatory factor

MDA5

melanoma differentiation–associated protein 5

RFP

red fluorescent protein

RIG-I

retinoic acid–inducible gene I

RLR

RIG-1–like receptor

siCon

control siRNA

siRNA

small interfering RNA

STING

stimulator of IFN genes

SVCV

spring viremia of carp virus

TBK1

TANK-binding kinase 1

VDAC1

voltage-dependent anion channel 1

1.
Akira
S.
,
S.
Uematsu
,
O.
Takeuchi
.
2006
.
Pathogen recognition and innate immunity.
Cell
124
:
783
801
.
2.
Honda
K.
,
T.
Taniguchi
.
2006
.
IRFs: master regulators of signalling by Toll-like receptors and cytosolic pattern-recognition receptors.
Nat. Rev. Immunol.
6
:
644
658
.
3.
Chen
S. N.
,
P. F.
Zou
,
P.
Nie
.
2017
.
Retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs) in fish: current knowledge and future perspectives.
Immunology
151
:
16
25
.
4.
Keating
S. E.
,
M.
Baran
,
A. G.
Bowie
.
2011
.
Cytosolic DNA sensors regulating type I interferon induction.
Trends Immunol.
32
:
574
581
.
5.
Zhou
R.
,
X.
Xie
,
X.
Li
,
Z.
Qin
,
C.
Wei
,
J.
Liu
,
Y.
Luo
.
2020
.
The triggers of the cGAS-STING pathway and the connection with inflammatory and autoimmune diseases.
Infect. Genet. Evol.
77
:
104094
.
6.
Deng
L.
,
Q.
Zeng
,
M.
Wang
,
A.
Cheng
,
R.
Jia
,
S.
Chen
,
D.
Zhu
,
M.
Liu
,
Q.
Yang
,
Y.
Wu
, et al
2018
.
Suppression of NF-κB activity: a viral immune evasion mechanism.
Viruses
10
:
409
.
7.
Seth
R. B.
,
L.
Sun
,
C. K.
Ea
,
Z. J.
Chen
.
2005
.
Identification and characterization of MAVS, a mitochondrial antiviral signaling protein that activates NF-κB and IRF3.
Cell
122
:
669
682
.
8.
Liu
S.
,
X.
Cai
,
J.
Wu
,
Q.
Cong
,
X.
Chen
,
T.
Li
,
F.
Du
,
J.
Ren
,
Y. T.
Wu
,
N. V.
Grishin
,
Z. J.
Chen
.
2015
.
Phosphorylation of innate immune adaptor proteins MAVS, STING, and TRIF induces IRF3 activation.
Science
347
:
aaa2630
.
9.
Shuai
K.
,
B.
Liu
.
2003
.
Regulation of JAK-STAT signalling in the immune system.
Nat. Rev. Immunol.
3
:
900
911
.
10.
Yoo
Y. S.
,
Y. Y.
Park
,
J. H.
Kim
,
H.
Cho
,
S. H.
Kim
,
H. S.
Lee
,
T. H.
Kim
,
Y.
Sun Kim
,
Y.
Lee
,
C. J.
Kim
, et al
2015
.
The mitochondrial ubiquitin ligase MARCH5 resolves MAVS aggregates during antiviral signalling.
Nat. Commun.
6
:
7910
.
11.
Pan
Y.
,
R.
Li
,
J. L.
Meng
,
H. T.
Mao
,
Y.
Zhang
,
J.
Zhang
.
2014
.
Smurf2 negatively modulates RIG-I-dependent antiviral response by targeting VISA/MAVS for ubiquitination and degradation.
J. Immunol.
192
:
4758
4764
.
12.
You
F.
,
H.
Sun
,
X.
Zhou
,
W.
Sun
,
S.
Liang
,
Z.
Zhai
,
Z.
Jiang
.
2009
.
PCBP2 mediates degradation of the adaptor MAVS via the HECT ubiquitin ligase AIP4.
Nat. Immunol.
10
:
1300
1308
.
13.
Cai
X.
,
Y. H.
Chiu
,
Z. J.
Chen
.
2014
.
The cGAS-cGAMP-STING pathway of cytosolic DNA sensing and signaling.
Mol. Cell
54
:
289
296
.
14.
Holm
C. K.
,
S. H.
Rahbek
,
H. H.
Gad
,
R. O.
Bak
,
M. R.
Jakobsen
,
Z.
Jiang
,
A. L.
Hansen
,
S. K.
Jensen
,
C.
Sun
,
M. K.
Thomsen
, et al
2016
.
Influenza A virus targets a cGAS-independent STING pathway that controls enveloped RNA viruses.
Nat. Commun.
7
:
10680
.
15.
Ahn
J.
,
D.
Gutman
,
S.
Saijo
,
G. N.
Barber
.
2012
.
STING manifests self DNA-dependent inflammatory disease.
Proc. Natl. Acad. Sci. USA
109
:
19386
19391
.
16.
Liu
Y.
,
A. A.
Jesus
,
B.
Marrero
,
D.
Yang
,
S. E.
Ramsey
,
G. A.
Montealegre Sanchez
,
K.
Tenbrock
,
H.
Wittkowski
,
O. Y.
Jones
,
H. S.
Kuehn
, et al
2014
.
Activated STING in a vascular and pulmonary syndrome.
N. Engl. J. Med.
371
:
507
518
.
17.
Li
S.
,
L. F.
Lu
,
H.
Feng
,
N.
Wu
,
D. D.
Chen
,
Y. B.
Zhang
,
J. F.
Gui
,
P.
Nie
,
Y. A.
Zhang
.
2014
.
IFN regulatory factor 10 is a negative regulator of the IFN responses in fish.
J. Immunol.
193
:
1100
1109
.
18.
Haller
O.
,
H.
Arnheiter
,
I.
Gresser
,
J.
Lindenmann
.
1979
.
Genetically determined, interferon-dependent resistance to influenza virus in mice.
J. Exp. Med.
149
:
601
612
.
19.
Lindenmann
J.
1964
.
Inheritance of resistance to influenza virus in mice.
Proc. Soc. Exp. Biol. Med.
116
:
506
509
.
20.
Haller
O.
,
P.
Staeheli
,
G.
Kochs
.
2007
.
Interferon-induced Mx proteins in antiviral host defense.
Biochimie
89
:
812
818
.
21.
Gao
S.
,
A.
von der Malsburg
,
S.
Paeschke
,
J.
Behlke
,
O.
Haller
,
G.
Kochs
,
O.
Daumke
.
2010
.
Structural basis of oligomerization in the stalk region of dynamin-like MxA.
Nature
465
:
502
506
.
22.
Aebi
M.
,
J.
Fäh
,
N.
Hurt
,
C. E.
Samuel
,
D.
Thomis
,
L.
Bazzigher
,
J.
Pavlovic
,
O.
Haller
,
P.
Staeheli
.
1989
.
cDNA structures and regulation of two interferon-induced human Mx proteins.
Mol. Cell Biol.
9
:
5062
5072
.
23.
Sandrock
M.
,
M.
Frese
,
O.
Haller
,
G.
Kochs
.
2001
.
Interferon-induced rat Mx proteins confer resistance to Rift Valley fever virus and other arthropod-borne viruses.
J. Interferon Cytokine Res.
21
:
663
668
.
24.
Li
G.
,
J.
Zhang
,
Y.
Sun
,
H.
Wang
,
Y.
Wang
.
2009
.
The evolutionarily dynamic IFN-inducible GTPase proteins play conserved immune functions in vertebrates and cephalochordates.
Mol. Biol. Evol.
26
:
1619
1630
.
25.
Wang
T.
,
F.
Liu
,
G.
Tian
,
C. J.
Secombes
,
T.
Wang
.
2019
.
Lineage/species-specific expansion of the Mx gene family in teleosts: differential expression and modulation of nine Mx genes in vitro and in vivo in rainbow trout Oncorhynchus mykiss.
Fish Shellfish Immunol.
91
:
447
448
.
26.
Jensen
I.
,
B.
Robertsen
.
2002
.
Effect of double-stranded RNA and interferon on the antiviral activity of Atlantic salmon cells against infectious salmon anemia virus and infectious pancreatic necrosis virus.
Fish Shellfish Immunol.
13
:
221
241
.
27.
Haller
O.
,
P.
Staeheli
,
M.
Schwemmle
,
G.
Kochs
.
2015
.
Mx GTPases: dynamin-like antiviral machines of innate immunity.
Trends Microbiol.
23
:
154
163
.
28.
Haller
O.
,
S.
Stertz
,
G.
Kochs
.
2007
.
The Mx GTPase family of interferon-induced antiviral proteins.
Microbes Infect.
9
:
1636
1643
.
29.
Kochs
G.
,
O.
Haller
.
1999
.
Interferon-induced human MxA GTPase blocks nuclear import of Thogoto virus nucleocapsids.
Proc. Natl. Acad. Sci. USA
96
:
2082
2086
.
30.
Goujon
C.
,
O.
Moncorgé
,
H.
Bauby
,
T.
Doyle
,
C. C.
Ward
,
T.
Schaller
,
S.
Hué
,
W. S.
Barclay
,
R.
Schulz
,
M. H.
Malim
.
2013
.
Human MX2 is an interferon-induced post-entry inhibitor of HIV-1 infection.
Nature
502
:
559
562
.
31.
Haller
O.
,
M.
Frese
,
D.
Rost
,
P. A.
Nuttall
,
G.
Kochs
.
1995
.
Tick-borne thogoto virus infection in mice is inhibited by the orthomyxovirus resistance gene product Mx1.
J. Virol.
69
:
2596
2601
.
32.
Haller
O.
,
G.
Kochs
.
2011
.
Human MxA protein: an interferon-induced dynamin-like GTPase with broad antiviral activity.
J. Interferon Cytokine Res.
31
:
79
87
.
33.
Grimm
D.
,
P.
Staeheli
,
M.
Hufbauer
,
I.
Koerner
,
L.
Martínez-Sobrido
,
A.
Solórzano
,
A.
García-Sastre
,
O.
Haller
,
G.
Kochs
.
2007
.
Replication fitness determines high virulence of influenza A virus in mice carrying functional Mx1 resistance gene.
Proc. Natl. Acad. Sci. USA
104
:
6806
6811
.
34.
Zimmermann
P.
,
B.
Mänz
,
O.
Haller
,
M.
Schwemmle
,
G.
Kochs
.
2011
.
The viral nucleoprotein determines Mx sensitivity of influenza A viruses.
J. Virol.
85
:
8133
8140
.
35.
Peng
L.
,
C.
Yang
,
J.
Su
.
2012
.
Protective roles of grass carp Ctenopharyngodon idella Mx isoforms against grass carp reovirus.
PLoS One
7
:
e52142
.
36.
Gao
S.
,
A.
von der Malsburg
,
A.
Dick
,
K.
Faelber
,
G. F.
Schröder
,
O.
Haller
,
G.
Kochs
,
O.
Daumke
.
2011
.
Structure of myxovirus resistance protein a reveals intra- and intermolecular domain interactions required for the antiviral function.
Immunity
35
:
514
525
.
37.
Verhelst
J.
,
P.
Hulpiau
,
X.
Saelens
.
2013
.
Mx proteins: antiviral gatekeepers that restrain the uninvited.
Microbiol. Mol. Biol. Rev.
77
:
551
566
.
38.
Wan
Q.
,
J.
Su
.
2015
.
Transcriptome analysis provides insights into the regulatory function of alternative splicing in antiviral immunity in grass carp (Ctenopharyngodon idella).
Sci. Rep.
5
:
12946
.
39.
Shang
X.
,
C.
Yang
,
Q.
Wan
,
Y.
Rao
,
J.
Su
.
2017
.
The destiny of the resistance/susceptibility against GCRV is controlled by epigenetic mechanisms in CIK cells.
Sci. Rep.
7
:
4551
.
40.
Rao
Y.
,
Q.
Wan
,
C.
Yang
,
J.
Su
.
2017
.
Grass carp laboratory of genetics and physiology 2 serves as a negative regulator in retinoic acid-inducible gene I- and melanoma differentiation-associated gene 5-mediated antiviral signaling in resting state and early stage of grass carp reovirus infection.
Front. Immunol.
8
:
352
.
41.
Rao
Y.
,
Q.
Wan
,
H.
Su
,
X.
Xiao
,
Z.
Liao
,
J.
Ji
,
C.
Yang
,
L.
Lin
,
J.
Su
.
2018
.
ROS-induced HSP70 promotes cytoplasmic translocation of high-mobility group box 1b and stimulates antiviral autophagy in grass carp kidney cells.
J. Biol. Chem.
293
:
17387
17401
.
42.
Su
H.
,
C.
Fan
,
Z.
Liao
,
C.
Yang
,
J. L.
Clarke
,
Y.
Zhang
,
J.
Su
.
2020
.
Grass carp reovirus major outer capsid protein VP4 interacts with RNA sensor RIG-I to suppress interferon response.
Biomolecules
10
:
560
.
43.
Ji
J.
,
Y.
Rao
,
Q.
Wan
,
Z.
Liao
,
J.
Su
.
2018
.
Teleost-specific TLR19 localizes to endosome, recognizes dsRNA, recruits TRIF, triggers both IFN and NF-κB pathways, and protects cells from grass carp reovirus infection.
J. Immunol.
200
:
573
585
.
44.
Wan
Q.
,
C.
Yang
,
Y.
Rao
,
Z.
Liao
,
J.
Su
.
2017
.
MDA5 induces a stronger interferon response than RIG-I to GCRV infection through a mechanism involving the phosphorylation and dimerization of IRF3 and IRF7 in CIK cells.
Front. Immunol.
8
:
189
.
45.
Xiao
X.
,
W.
Zhu
,
Y.
Zhang
,
Z.
Liao
,
C.
Wu
,
C.
Yang
,
Y.
Zhang
,
S.
Xiao
,
J.
Su
.
2021
.
Broad-spectrum robust direct bactericidal activity of fish IFNφ1 reveals an antimicrobial peptide-like function for type I IFNs in vertebrates.
J. Immunol.
206
:
1337
1347
.
46.
Su
J.
,
R.
Zhang
,
J.
Dong
,
C.
Yang
.
2011
.
Evaluation of internal control genes for qRT-PCR normalization in tissues and cell culture for antiviral studies of grass carp (Ctenopharyngodon idella).
Fish Shellfish Immunol.
30
:
830
835
.
47.
Rao
Y.
,
J.
Su
,
C.
Yang
,
N.
Yan
,
X.
Chen
,
X.
Feng
.
2015
.
Dynamic localization and the associated translocation mechanism of HMGBs in response to GCRV challenge in CIK cells.
Cell. Mol. Immunol.
12
:
342
353
.
48.
Liao
Z.
,
Q.
Wan
,
J.
Su
.
2016
.
Bioinformatics analysis of organizational and expressional characterizations of the IFNs, IRFs and CRFBs in grass carp Ctenopharyngodon idella.
Dev. Comp. Immunol.
61
:
97
106
.
49.
Paz
S.
,
Q.
Sun
,
P.
Nakhaei
,
R.
Romieu-Mourez
,
D.
Goubau
,
I.
Julkunen
,
R.
Lin
,
J.
Hiscott
.
2006
.
Induction of IRF-3 and IRF-7 phosphorylation following activation of the RIG-I pathway.
Cell. Mol. Biol.
52
:
17
28
.
50.
Zürcher
T.
,
J.
Pavlovic
,
P.
Staeheli
.
1992
.
Mechanism of human MxA protein action: variants with changed antiviral properties.
EMBO J.
11
:
1657
1661
.
51.
Chen
J.
,
Y.
Wu
,
X. D.
Wu
,
J.
Zhou
,
X. D.
Liang
,
A. S.
Baloch
,
Y. F.
Qiu
,
S.
Gao
,
B.
Zhou
.
2020
.
The R614E mutation of mouse Mx1 protein contributes to the novel antiviral activity against classical swine fever virus.
Vet. Microbiol.
243
:
108621
.
52.
Biacchesi
S.
,
E.
Mérour
,
A.
Lamoureux
,
J.
Bernard
,
M.
Brémont
.
2012
.
Both STING and MAVS fish orthologs contribute to the induction of interferon mediated by RIG-I.
PLoS One
7
:
e47737
.
53.
Kato
H.
,
T.
Fujita
.
2015
.
RIG-I-like receptors and autoimmune diseases.
Curr. Opin. Immunol.
37
:
40
45
.
54.
Li
Y.
,
C.
Li
,
P.
Xue
,
B.
Zhong
,
A. P.
Mao
,
Y.
Ran
,
H.
Chen
,
Y. Y.
Wang
,
F.
Yang
,
H. B.
Shu
.
2009
.
ISG56 is a negative-feedback regulator of virus-triggered signaling and cellular antiviral response.
Proc. Natl. Acad. Sci. USA
106
:
7945
7950
.
55.
Mankouri
J.
,
R.
Fragkoudis
,
K. H.
Richards
,
L. F.
Wetherill
,
M.
Harris
,
A.
Kohl
,
R. M.
Elliott
,
A.
Macdonald
.
2010
.
Optineurin negatively regulates the induction of IFNβ in response to RNA virus infection.
PLoS Pathog.
6
:
e1000778
.
56.
Rothenfusser
S.
,
N.
Goutagny
,
G.
DiPerna
,
M.
Gong
,
B. G.
Monks
,
A.
Schoenemeyer
,
M.
Yamamoto
,
S.
Akira
,
K. A.
Fitzgerald
.
2005
.
The RNA helicase Lgp2 inhibits TLR-independent sensing of viral replication by retinoic acid-inducible gene-I.
J. Immunol.
175
:
5260
5268
.
57.
Xu
L.
,
N.
Xiao
,
F.
Liu
,
H.
Ren
,
J.
Gu
.
2009
.
Inhibition of RIG-I and MDA5-dependent antiviral response by gC1qR at mitochondria.
Proc. Natl. Acad. Sci. USA
106
:
1530
1535
.
58.
Carty
M.
,
R.
Goodbody
,
M.
Schröder
,
J.
Stack
,
P. N.
Moynagh
,
A. G.
Bowie
.
2006
.
The human adaptor SARM negatively regulates adaptor protein TRIF-dependent Toll-like receptor signaling.
Nat. Immunol.
7
:
1074
1081
.
59.
Diao
F.
,
S.
Li
,
Y.
Tian
,
M.
Zhang
,
L. G.
Xu
,
Y.
Zhang
,
R. P.
Wang
,
D.
Chen
,
Z.
Zhai
,
B.
Zhong
, et al
2007
.
Negative regulation of MDA5- but not RIG-I-mediated innate antiviral signaling by the dihydroxyacetone kinase.
Proc. Natl. Acad. Sci. USA
104
:
11706
11711
.
60.
Moore
C. B.
,
D. T.
Bergstralh
,
J. A.
Duncan
,
Y.
Lei
,
T. E.
Morrison
,
A. G.
Zimmermann
,
M. A.
Accavitti-Loper
,
V. J.
Madden
,
L.
Sun
,
Z.
Ye
, et al
2008
.
NLRX1 is a regulator of mitochondrial antiviral immunity.
Nature
451
:
573
577
.
61.
Kim
M. J.
,
S. Y.
Hwang
,
T.
Imaizumi
,
J. Y.
Yoo
.
2008
.
Negative feedback regulation of RIG-I-mediated antiviral signaling by interferon-induced ISG15 conjugation.
J. Virol.
82
:
1474
1483
.
62.
Wertz
I. E.
,
K. M.
O’Rourke
,
H.
Zhou
,
M.
Eby
,
L.
Aravind
,
S.
Seshagiri
,
P.
Wu
,
C.
Wiesmann
,
R.
Baker
,
D. L.
Boone
, et al
2004
.
De-ubiquitination and ubiquitin ligase domains of A20 downregulate NF-κB signalling.
Nature
430
:
694
699
.
63.
Arimoto
K.
,
H.
Takahashi
,
T.
Hishiki
,
H.
Konishi
,
T.
Fujita
,
K.
Shimotohno
.
2007
.
Negative regulation of the RIG-I signaling by the ubiquitin ligase RNF125.
Proc. Natl. Acad. Sci. USA
104
:
7500
7505
.
64.
Higgs
R.
,
J.
Ní Gabhann
,
N.
Ben Larbi
,
E. P.
Breen
,
K. A.
Fitzgerald
,
C. A.
Jefferies
.
2008
.
The E3 ubiquitin ligase Ro52 negatively regulates IFN-β production post-pathogen recognition by polyubiquitin-mediated degradation of IRF3.
J. Immunol.
181
:
1780
1786
.
65.
Nakhaei
P.
,
T.
Mesplede
,
M.
Solis
,
Q.
Sun
,
T.
Zhao
,
L.
Yang
,
T. H.
Chuang
,
C. F.
Ware
,
R.
Lin
,
J.
Hiscott
.
2009
.
The E3 ubiquitin ligase Triad3A negatively regulates the RIG-I/MAVS signaling pathway by targeting TRAF3 for degradation.
PLoS Pathog.
5
:
e1000650
.
66.
Korolchuk
V. I.
,
A.
Mansilla
,
F. M.
Menzies
,
D. C.
Rubinsztein
.
2009
.
Autophagy inhibition compromises degradation of ubiquitin-proteasome pathway substrates.
Mol. Cell
33
:
517
527
.
67.
Ardley
H. C.
,
P. A.
Robinson
.
2005
.
E3 ubiquitin ligases.
Essays Biochem.
41
:
15
30
.
68.
Metzger
M. B.
,
V. A.
Hristova
,
A. M.
Weissman
.
2012
.
HECT and RING finger families of E3 ubiquitin ligases at a glance.
J. Cell Sci.
125
:
531
537
.
69.
Hatakeyama
S.
,
K. I.
Nakayama
.
2003
.
U-box proteins as a new family of ubiquitin ligases.
Biochem. Biophys. Res. Commun.
302
:
635
645
.
70.
Lee
I. H.
2019
.
Mechanisms and disease implications of sirtuin-mediated autophagic regulation.
Exp. Mol. Med.
51
:
1
11
.
71.
Levine
B.
,
N.
Mizushima
,
H. W.
Virgin
.
2011
.
Autophagy in immunity and inflammation.
Nature
469
:
323
335
.
72.
Cheon
H.
,
E. G.
Holvey-Bates
,
J. W.
Schoggins
,
S.
Forster
,
P.
Hertzog
,
N.
Imanaka
,
C. M.
Rice
,
M. W.
Jackson
,
D. J.
Junk
,
G. R.
Stark
.
2013
.
IFNβ-dependent increases in STAT1, STAT2, and IRF9 mediate resistance to viruses and DNA damage.
EMBO J.
32
:
2751
2763
.
73.
Zhang
X. J.
,
P.
Zhang
,
H.
Li
.
2015
.
Interferon regulatory factor signalings in cardiometabolic diseases.
Hypertension
66
:
222
247
.

The authors have no financial conflicts of interest.

Supplementary data