IFI16 Inhibits Porcine Reproductive and Respiratory Syndrome Virus 2 Replication in a MAVS-Dependent Manner in MARC-145 Cells

Porcine reproductive and respiratory syndrome virus (PRRSV) is a single-stranded positive-sense RNA virus, and the current strategies for controlling PRRSV are limited. Interferon gamma-inducible protein 16 (IFI16) has been reported to have a broader role in the regulation of the type I interferons (IFNs) response to RNA and DNA viruses. However, the function of IFI16 in PRRSV infection is unclear. Here, we revealed that IFI16 acts as a novel antiviral protein against PRRSV-2. IFI16 could be induced by interferon-beta (IFN-β). Overexpression of IFI16 could significantly suppress PRRSV-2 replication, and silencing the expression of endogenous IFI16 by small interfering RNAs led to the promotion of PRRSV-2 replication in MARC-145 cells. Additionally, IFI16 could promote mitochondrial antiviral signaling protein (MAVS)-mediated production of type I interferon and interact with MAVS. More importantly, IFI16 exerted anti-PRRSV effects in a MAVS-dependent manner. In conclusion, our data demonstrated that IFI16 has an inhibitory effect on PRRSV-2, and these findings contribute to understanding the role of cellular proteins in regulating PRRSV replication and may have implications for the future antiviral strategies.


Introduction
Porcine reproductive and respiratory syndrome (PRRS), caused by PRRS virus (PRRSV), is an economically important viral disease all over the word [1,2]. PRRSV is a member of the Arteriviridae family in the order Nidovirales and consists of an enveloped 15 kb positive-strand RNA genome containing at least ten open reading frames (ORFs) [3][4][5][6]. PRRSV is divided into two genotypes: the European genotype (type 1) and the North American genotype (type 2). There is considerable sequence variability within both groups, and only about 50-60% nucleotide sequence identity between the two subtypes [7,8]. Recently, based on a new proposed classification scheme, type 1 and type 2 PRRSV

Antibodies and Reagents
Mouse anti-Flag M2 monoclonal antibody, mouse anti-c-Myc monoclonal antibody, anti-Flag affinity gel beads, anti-c-Myc affinity gel beads, and mouse IgG agarose were purchased from Sigma-Aldrich. PRRSV N protein antibody was purchased from GeneTex. Mouse anti-β-actin monoclonal antibody was purchased from GenScript. Rabbit anti-MAVS monoclonal antibody was purchased from Cell Signaling Technology (CST). The secondary antibodies conjugated to HRP were purchased from Jackson Immuno Research. Polyinosinic-polycytidylic acid (polyI:C) was purchased from Sigma-Aldrich. Recombinant human IFN-β was purchased from PEPROTECH. Lipofectamine2000 transfection reagent and Lipofectamine RNAiMAX transfection reagent were purchased from Invitrogen (Carlsbad, CA, USA).

Plasmids
To generate IFI16-Myc, the cDNA of IFI16 from MARC-145 cells was amplified and cloned into pCMV-Myc vector (Beyotime, Shanghai, China). The genes of RIG-I, MDA5, MAVS, TBK1, and IRF3 from MARC-145 cells were cloned into pCMV-Flag (Beyotime), creating RIG-I-Flag, MDA5-Flag, MAVS-Flag, TBK1-Flag, and IRF3-Flag respectively. The expression plasmids for 3×Flag-tagged MAVS and MAVS truncations were constructed by standard molecular cloning method from cDNA templates. All constructs were confirmed by DNA sequencing. The IFN-β promoter reporter plasmid (p-284) has been constructed as previous described [31]. All primers used are listed in Table 1. Table 1. Primers used for expression plasmid construction.

Quantitative Real-Time PCR Analysis
Total RNA was extracted from cells using RNeasy Mini Kit (Qiagen), reverse transcription was accomplished with PrimeScript™ RT Master Mix (Perfect Real Time) (Takara, Kyoto, Japan) according to the manufacturer's protocol. And then the samples were subjected to real-time PCR analysis with specific primers by using a Fast Start Universal SYBR Green Master Kit (Roche, Basel, Switzerland) on a 7500 fast RT-PCR system (Applied Biosystems, Foster City, CA, USA). Relative analysis of gene expression was evaluated using the 2 −∆∆CT method, in which target gene expression was normalized by glyceraldehyde 3-phosphate dehydrogenase (GAPDH) expression. Primers are shown in Table 2.

Dual Luciferase Reporter Assay
MARC-145 cells in 24-well plates were transfected with indicated expression plasmids and IFN-β luciferase reporter plasmids by using Lipofectamine 2000 transfection reagent (Invitrogen). At 36 hpt, the cells were transfected with polyI:C (10 µg/mL) (Sigma-Aldrich, St. Louis, MO, USA) for 12 h, and then the cells were harvested and the luciferase activity was measured by using dual-luciferase reporter assay system (Promega, Madison, WI, USA).

Co-Immunoprecipitation
HEK-293T cells were transfected with IFI16-Myc and MAVS-3×Flag, MAVS truncations plasmids or control vector. At 48 hpt, the cells were lysed in IP-lysis buffer (Sigma-Aldrich) containing protease inhibitor cocktail (Roche) for 1 h at 4 • C. The lysates were centrifuged at 12,000× g for 15 min at 4 • C, and the supernatants were pre-cleared with mouse IgG-Agarose (Sigma-Aldrich) at 4 • C for 2 h. Then, the pre-cleared supernatants were incubated with anti-c-Myc affinity gel beads or anti-Flag affinity gel beads (Sigma-Aldrich) for 4 h or overnight at 4 • C. The precipitates were washed five times with TBS buffer and detected by western blotting.

Virus Titration
Virus titers were determined according to a previous report [32]. Briefly, MARC-145 cells, grown in 96-well plates, were infected with ten-fold serial dilution of samples. After 1 h incubation at 37 • C, the supernatants were replaced with fresh DMEM containing 2% FBS. Five days post infection, the cytopathic effect (CPE) characterized by clumping and shrinkage of cells was obviously visible in MARC-145 cells and the viral titers, expressed as 50% tissue culture infective dose (TCID 50 ), calculated according to the method of Reed-Muench [33].

Statistical Analysis
Statistical graphs were created with GraphPad Prism software, and all data were analyzed using Student's t tests as the mean values ± the standard deviations (SD) of at least three independent experiments. The asterisks in the figures indicate significant differences (*, p < 0.05; **, p < 0.01).

IFI16 Inhibits PRRSV-2 Replication
Since type I interferon and interferon-induced genes could efficiently inhibit PRRSV replication in MARC-145 cells [21,23], and it has been reported that IFI16 could be induced by type I interferon [34][35][36], we firstly confirmed whether IFI16 could be induced by type I interferon in MARC-145 cells, and then explored whether IFI16 could inhibit PRRSV replication. MARC-145 cells were treated with IFN-β, and then the expression of IFI16 was detected. Consistent with the results of previous reports, IFN-β could also efficiently induce the expression of IFI16 ( Figure 1A), and the expression of IFI16 was enhanced in an IFN-β-dose-dependent manner and peaked at 24 h in MARC-145 cells ( Figure 1B). In addition, the transcription level of IFI16 was increased in the cells infected with PRRSV-2 ( Figure 1C,D). MARC-145 cells, and then explored whether IFI16 could inhibit PRRSV replication. MARC-145 cells were treated with IFN-β, and then the expression of IFI16 was detected. Consistent with the results of previous reports, IFN-β could also efficiently induce the expression of IFI16 ( Figure 1A), and the expression of IFI16 was enhanced in an IFN-β-dose-dependent manner and peaked at 24 h in MARC-145 cells ( Figure 1B). In addition, the transcription level of IFI16 was increased in the cells infected with PRRSV-2 ( Figure 1C, D). Next, to explore whether IFI16 could inhibit PRRSV-2 replication, MARC-145 cells were transfected with IFI16-Myc and control vector for 24 h, and then the cells were infected with PRRSV-2 at a MOI of 0.1. The results in Figure 2 show that overexpression of IFI16 reduced the RNA levels of PRRSV (Figure 2A), the PRRSV titers ( Figure 2B), and the expression levels of the N protein of PRRSV ( Figure 2C,D). To further examine the antiviral effects of IFI16 on PRRSV-2, we used specific siRNA to down-regulate the expression of endogenous IFI16, which led to lower expression of IFI16 than that in cells transfected with control siRNA ( Figure 3A). Compared with the nontargeting control siRNA (siNC)-transfected control cells, the RNA levels of PRRSV ( Figure  3B), viral titers ( Figure 3C), and the expression of N protein ( Figure 3D,E) were significantly increased in MARC-145 cells transfected with siIFI16. Together, these results indicated that IFI16 acts as an antiviral protein against PRRSV-2. To further examine the antiviral effects of IFI16 on PRRSV-2, we used specific siRNA to down-regulate the expression of endogenous IFI16, which led to lower expression of IFI16 than that in cells transfected with control siRNA ( Figure 3A). Compared with the nontargeting control siRNA (siNC)-transfected control cells, the RNA levels of PRRSV ( Figure 3B), viral titers ( Figure 3C), and the expression of N protein ( Figure 3D,E) were significantly increased in MARC-145 cells transfected with siIFI16. Together, these results indicated that IFI16 acts as an antiviral protein against PRRSV-2.

IFI16 Enhances the MAVS-mediated Type I IFN Signaling
IFI16 plays an important role in Sendai virus (SeV)-mediated production of type I IFN [30,37], and in light of the fact that PRRSV could slightly induce the production of type I IFN [31,38], we wonder whether IFI16 was useful to the transcription of type I IFN during PRRSV infection. MARC-145 cells were transfected with IFI16-Myc or control vector, and 24 h later, the cells were infected with PRRSV-2 BJ-4 at a MOI of 1. The results showed that IFI16 could enhance PRRSV-induced production of IFN-β (Figure 4).
Viruses 2019, 11, x FOR PEER REVIEW 7 of 14 (E) The relative intensity ratios of N protein were analyzed by Image J software. All experiments were repeated at least three times with similar results. * p < 0.05, ** p < 0.01.

IFI16 enhances the MAVS-mediated type I IFN signaling
IFI16 plays an important role in Sendai virus (SeV)-mediated production of type I IFN [30,37], and in light of the fact that PRRSV could slightly induce the production of type I IFN [31,38], we wonder whether IFI16 was useful to the transcription of type I IFN during PRRSV infection. MARC-145 cells were transfected with IFI16-Myc or control vector, and 24 h later, the cells were infected with PRRSV-2 BJ-4 at a MOI of 1. The results showed that IFI16 could enhance PRRSV-induced production of IFN-β (Figure 4). Given the fact that IFI16 plays a key role in the signaling through RIG-I, and among the different RNA sensors, DExD/H-box RNA helicases of the RLR family have been identified as essential sensors of RNA viruses [15,30,39,40], we explored the role of IFI16 in the RIG-I-mediated signaling pathway. Firstly, we found that IFI16 could enhance polyI:C-induced IFN-β promoter activity and the transcriptional levels of IFN-β ( Figure 5A,B). Next, to further investigate the mechanism of IFI16 in enhancing type I IFN signaling, MARC-145 cells were co-transfected with plasmids encoding IFI16, components of RIG-I pathway, and IFN-β reporter plasmid. The results showed that IFI16 greatly enhanced the IFN-β promoter activity, the mRNA transcriptional levels of IFN-β, ISG15, and ISG56 induced by MAVS ( Figure 5C-F). Subsequently, cells were co-transfected with MAVS, IFN-β reporter plasmid, and different concentrations of IFI16, and the results showed that IFI16 strongly increased MAVS-mediated IFN-β promoter activity and the mRNA transcription levels of IFN-β, ISG15, and ISG56 in a dose-dependent manner ( Figure 5G-J).
Since IFI16 could enhance MAVS-mediated type I IFN signaling, to identify whether MAVS is indispensable for IFI16 to regulate the production of type I IFN, MARC-145 cells were transfected with siMAVS, IFI16-Myc, and IFN-β reporter plasmid, and 36 h later, the cells were stimulated with polyI:C. And the results showed that IFI16 could not promote polyI:C-induced type I IFN production upon silencing of MAVS ( Figure 5K-L). Collectively, these data indicated that IFI16 positively regulates MAVS-mediated type I IFN signaling. Given the fact that IFI16 plays a key role in the signaling through RIG-I, and among the different RNA sensors, DExD/H-box RNA helicases of the RLR family have been identified as essential sensors of RNA viruses [15,30,39,40], we explored the role of IFI16 in the RIG-I-mediated signaling pathway. Firstly, we found that IFI16 could enhance polyI:C-induced IFN-β promoter activity and the transcriptional levels of IFN-β ( Figure 5A,B). Next, to further investigate the mechanism of IFI16 in enhancing type I IFN signaling, MARC-145 cells were co-transfected with plasmids encoding IFI16, components of RIG-I pathway, and IFN-β reporter plasmid. The results showed that IFI16 greatly enhanced the IFN-β promoter activity, the mRNA transcriptional levels of IFN-β, ISG15, and ISG56 induced by MAVS ( Figure 5C-F). Subsequently, cells were co-transfected with MAVS, IFN-β reporter plasmid, and different concentrations of IFI16, and the results showed that IFI16 strongly increased MAVS-mediated IFN-β promoter activity and the mRNA transcription levels of IFN-β, ISG15, and ISG56 in a dose-dependent manner ( Figure 5G-J).
Since IFI16 could enhance MAVS-mediated type I IFN signaling, to identify whether MAVS is indispensable for IFI16 to regulate the production of type I IFN, MARC-145 cells were transfected with siMAVS, IFI16-Myc, and IFN-β reporter plasmid, and 36 h later, the cells were stimulated with polyI:C. And the results showed that IFI16 could not promote polyI:C-induced type I IFN production upon silencing of MAVS ( Figure 5K,L). Collectively, these data indicated that IFI16 positively regulates MAVS-mediated type I IFN signaling.

IFI16 Interacts with MAVS
IFI16 is mainly localized in nucleus and partially localized in cytoplasmic and mitochondria [29], while MAVS could localize in cytoplasmic and mitochondria [41]. Besides, IFI16 could facilitate the MAVS-mediated type I IFN signaling, so to determine whether IFI16 could interact with MAVS, the co-immunoprecipitation assays (co-IP) were performed in 293T cells, and the results showed that IFI16 specifically interacted with MAVS ( Figure 6A,B).
As a central adaptor of IFN signaling, MAVS contains an N-terminal CARD domain, a proline-rich domain, and a C-terminal transmembrane domain (TM) [42,43]. So, to determine the functions of MAVS domains in the binding with IFI16, a series of MAVS truncations were generated ( Figure 6C). In the assays of co-IP, IFI16 only interacted with full-length MAVS, but not with truncations of MAVS ( Figure 6D). These results indicated that the CARD domain and TM domain of MAVS are sufficient and efficient for the binding with IFI16.

IFI16 Interacts with MAVS
IFI16 is mainly localized in nucleus and partially localized in cytoplasmic and mitochondria [29], while MAVS could localize in cytoplasmic and mitochondria [41]. Besides, IFI16 could facilitate the MAVS-mediated type I IFN signaling, so to determine whether IFI16 could interact with MAVS, the co-immunoprecipitation assays (co-IP) were performed in 293T cells, and the results showed that IFI16 specifically interacted with MAVS ( Figure 6A,B).
As a central adaptor of IFN signaling, MAVS contains an N-terminal CARD domain, a proline-rich domain, and a C-terminal transmembrane domain (TM) [42,43]. So, to determine the functions of MAVS domains in the binding with IFI16, a series of MAVS truncations were generated ( Figure 6C). In the assays of co-IP, IFI16 only interacted with full-length MAVS, but not with truncations of MAVS ( Figure 6D). These results indicated that the CARD domain and TM domain of MAVS are sufficient and efficient for the binding with IFI16.

Antiviral Activity of IFI16 is Dependent on MAVS
Since IFI16 possesses significant antiviral activity and enhances MAVS-mediated type I IFN signaling, to investigate whether the antiviral activity of IFI16 is dependent on MAVS, we silenced MAVS using siRNAs, and as shown in Figure 7, marked reduction of MAVS expression was observed in the MAVS silenced cells (Figure 7A,B). And, IFI16 could not decrease the PRRSV RNA levels ( Figure 7C

Antiviral Activity of IFI16 is Dependent on MAVS
Since IFI16 possesses significant antiviral activity and enhances MAVS-mediated type I IFN signaling, to investigate whether the antiviral activity of IFI16 is dependent on MAVS, we silenced MAVS using siRNAs, and as shown in Figure 7, marked reduction of MAVS expression was observed in the MAVS silenced cells (Figure 7A,B). And, IFI16 could not decrease the PRRSV RNA levels ( Figure 7C

Discussion
In this study, we revealed a novel mechanism by which IFI16 inhibits PRRSV-2 replication. Initially, we showed that IFI16 could efficiently repress PRRSV-2 replication. Subsequently, we demonstrated that IFI16 markedly enhances MAVS-mediated type I IFN signaling and binds directly with MAVS. Finally, the ability of IFI16 to antagonize the replication of PRRSV-2 is dependent on MAVS. Taken together, these findings suggest that IFI16 plays an important role in response to PRRSV-2 in MARC-145 cells.
IFI16 is an intracellular DNA sensor and could mediate the induction of IFN-β by stimulating with single-stranded and double-stranded DNA sequences or DNA virus infection [29]. And IFI16 has also been described as an antiviral restriction factor against DNA viruses since it could directly interact with DNA virus components and block viral replication [44,45]. Besides, IFI16 plays a broader role in the regulation of the type I IFN response to DNA viruses in antiviral immunity [30]. However, the functions of IFI16 in regulating RNA virus including PRRSV replication are largely unknown. In this study, we provide the first evidence that IFI16 could inhibit PRRSV-2 replication in MARC-145 cells. Additionally, IFI16 is widely expressed in endothelial and epithelial cells [46], and although the roles of IFI16 in porcine alveolar macrophages (PAMs) currently could not be investigated, this study may have implications for other RNA viruses.
It has been reported that IFI16 was essential for SeV-mediated production of type I IFN [30,37], and SeV-mediated induction of type I IFN comes primarily from RIG-I activation [39,40,47], which indicated that IFI16 may be involved in regulating the RIG-I signal pathway. Here, we have shown that IFI16 could interact with MAVS and promote MAVS-mediated type I IFN signaling, and MAVS was essential for IFI16 to regulate type I IFN signaling. MAVS is a central adaptor protein for RIG-I signaling pathway, which indicted that IFI16 may positively regulate RIG-I signaling pathway. Given that IFI16 transcriptionally regulates ISGs to enhance IFN responses to DNA

Discussion
In this study, we revealed a novel mechanism by which IFI16 inhibits PRRSV-2 replication. Initially, we showed that IFI16 could efficiently repress PRRSV-2 replication. Subsequently, we demonstrated that IFI16 markedly enhances MAVS-mediated type I IFN signaling and binds directly with MAVS. Finally, the ability of IFI16 to antagonize the replication of PRRSV-2 is dependent on MAVS. Taken together, these findings suggest that IFI16 plays an important role in response to PRRSV-2 in MARC-145 cells.
IFI16 is an intracellular DNA sensor and could mediate the induction of IFN-β by stimulating with single-stranded and double-stranded DNA sequences or DNA virus infection [29]. And IFI16 has also been described as an antiviral restriction factor against DNA viruses since it could directly interact with DNA virus components and block viral replication [44,45]. Besides, IFI16 plays a broader role in the regulation of the type I IFN response to DNA viruses in antiviral immunity [30]. However, the functions of IFI16 in regulating RNA virus including PRRSV replication are largely unknown. In this study, we provide the first evidence that IFI16 could inhibit PRRSV-2 replication in MARC-145 cells. Additionally, IFI16 is widely expressed in endothelial and epithelial cells [46], and although the roles of IFI16 in porcine alveolar macrophages (PAMs) currently could not be investigated, this study may have implications for other RNA viruses.
It has been reported that IFI16 was essential for SeV-mediated production of type I IFN [30,37], and SeV-mediated induction of type I IFN comes primarily from RIG-I activation [39,40,47], which indicated that IFI16 may be involved in regulating the RIG-I signal pathway. Here, we have shown that IFI16 could interact with MAVS and promote MAVS-mediated type I IFN signaling, and MAVS was essential for IFI16 to regulate type I IFN signaling. MAVS is a central adaptor protein for RIG-I signaling pathway, which indicted that IFI16 may positively regulate RIG-I signaling pathway. Given that IFI16 transcriptionally regulates ISGs to enhance IFN responses to DNA viruses [29,30], this may be a common mechanism in response to DNA or RNA viruses, which needs further demonstration in other PRRSV strains, and even other RNA viruses.
Since PRRSV could be recognized by RIG-I and induce the production of type I IFN [31,38], and IFI16 could positively regulate the RIG-I signaling pathway, we explored the mechanism by which IFI16 inhibits PRRSV-2 replication. We found that IFI16 could facilitate the PRRSV-mediated induction of IFN-β, interact with MAVS, and the anti-PRRSV activity of IFI16 is dependent on MAVS, which indicted that the ability of IFI16 to inhibit PRRSV-2 replication may be relevant to its role in the type I IFN signaling. Interestingly, IFI16 expression was increased after PRRSV infection, while possibly inhibiting virus replication. However, the mechanisms underlying need to be explored further.
In fact, in contrast to the results that IFI16 plays an important role in SeV-mediated production of type I IFN [30], several other groups have shown that both knockout and knockdown of IFI16 could not influence the ployI:C and SeV-mediated production of type I interferon [26,29,48]. The reason of these different results may be due to different cell lines they used. Here, we showed another piece of evidence that IFI16 may take part in the RIG-I signaling pathway. Previous studies have shown that STING could interact with RIG-I and MAVS, and there was also a marked reduction in the induction of IFN-β by RIG-I and MAVS in the absence of STING [49,50], while Brunette et al. have shown that the IFN response to poly(dA:dT) was reduced by >99% in STING −/− , MAVS −/− macrophages, and DCs, but not in phagocytes deficient in STING alone [51], which indicated that IFI16 may be a crosstalk between RIG-I-MAVS-type I interferon and STING-DNA-sensing pathways [52]. It is well known that IFI16 recruits STING to induce IFN-β transcription, so the roles of IFI16 in the crosstalk between RIG-I-MAVS-RNA-sensing pathways and IFI16-STING-DNA-sensing pathways need further studies.
In conclusion, our data show for the first time that IFI16 could inhibit PRRSV-2 replication in a MAVS-dependent manner, which may have implications for other RNA viruses and contribute to understand the antiviral mechanism, as well as virus-host interactions.