Encephalomyocarditis Virus Abrogates the Interferon Beta Signaling Pathway via Its Structural Protein VP2

Encephalomyocarditis virus is an important pathogen that can cause encephalitis, myocarditis, neurological diseases, and reproductive disorders. It also causes huge economic losses for the swine industry worldwide.

ABSTRACT

Type I interferon (IFN)-mediated antiviral responses are critical for modulating host-virus responses, and indeed, viruses have evolved strategies to antagonize this pathway. Encephalomyocarditis virus (EMCV) is an important zoonotic pathogen, which causes myocarditis, encephalitis, neurological disease, reproductive disorders, and diabetes in pigs. This study aims to understand how EMCV interacts with the IFN pathway. EMCV circumvents the type I IFN response by expressing proteins that antagonize cellular innate immunity. Here, we show that EMCV VP2 is a negative regulator of the IFN-β pathway. This occurs via the degradation of the MDA5-mediated cytoplasmic double-stranded RNA (dsRNA) antiviral sensing RIG-I-like receptor (RLR) pathway. We show that structural protein VP2 of EMCV interacts with MDA5, MAVS, and TBK1 through its C terminus. In addition, we found that EMCV VP2 could significantly degrade RLRs by the proteasomal and lysosomal pathways. For the first time, EMCV VP2 was shown to play an important role in EMCV evasion of the type I IFN signaling pathway. This study expands our understanding that EMCV utilizes its capsid protein VP2 to evade the host antiviral response.

IMPORTANCE Encephalomyocarditis virus is an important pathogen that can cause encephalitis, myocarditis, neurological diseases, and reproductive disorders. It also causes huge economic losses for the swine industry worldwide. Innate immunity plays an important role in defending the host from pathogen infection. Understanding pathogen microorganisms evading the host immune system is of great importance. Currently, whether EMCV evades cytosolic RNA sensing and signaling is still poorly understood. In the present study, we found that viral protein VP2 antagonized the RLR signaling pathway by degrading MDA5, MAVS, and TBK1 protein expression to facilitate viral replication in HEK293 cells. The findings in this study identify a new mechanism for EMCV evading the host’s innate immune response, which provide new insights into the virus-host interaction and help develop new antiviral approaches against EMCV.

INTRODUCTION

Encephalomyocarditis virus (EMCV) is a nonenveloped single-stranded RNA virus belonging to the Cardiovirus genus of the Picornaviridae family. Its genome resembles closely that of foot-and-mouth disease virus (FMDV) (1). EMCV is an important zoonotic pathogen that can cause myocarditis and encephalitis in a variety of mammals, resulting in neurological diseases, reproductive disorders, and diabetes (2, 3). Recently, an epidemic of cerebral myocarditis in swine was reported in several countries, including China, inflicting serious economic losses on the swine industry worldwide (4, 5). Etiological and serological investigations have shown EMCV to be widely distributed in China (6), concurring with previous speculations noting that most of these human cases are probably asymptomatic and/or unrecognized (7).

EMCV viral RNA consists of a 5′ untranslated region (UTR), a single open reading frame (ORF), and a 3′ UTR. The single ORF encodes four structural (VP1, VP2, VP3, and VP4) and eight nonstructural (L^pro, 2A, 2B, 2C, 3A, 3B, 3C, and 3D) proteins (8, 9). EMCV structural proteins play essential roles in virus entry, uncoating, and assembly, whereas nonstructural proteins form the main components of the viral replication complex. Notably, EMCV L^pro, 2C, and 3C play important roles in blocking interferon (IFN) expression and downstream antiviral signaling (10–15).

The host innate immune system is the first line of defense against viral infections (16). During a viral infection, host pattern recognition receptors (PRRs) recognize viral components and trigger antiviral responses by producing type I IFNs (17). Viral RNAs are primarily recognized by two classes of PRRs: Toll-like receptors (TLRs) and retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs). RIG-I and melanoma differentiation-associated gene 5 (MDA5) are important viral RLR sensors (18–20). When cytoplasmic viral RNA binds the RNA helicases RIG-I and/or MDA5, it interacts with the downstream adaptor protein mitochondrial antiviral signaling protein (MAVS) to facilitate its oligomerization. The oligomerization of MAVS is important for the assembly of the signalosome complex containing the TRAF3- and IκB-related kinases TBK1 and IKKε, which in turn directly phosphorylate IFN-regulatory factor 3 (IRF3) and IRF7, resulting in their translocation from the cytoplasm to the nucleus, which activates the transcription of antiviral genes, including IFN and IFN-stimulated genes (ISGs) (20–26).

Work to date studying the interplay of EMCV and the host immune system have referenced only studies related to viral proteins derived from poliovirus (PV), Tyler mouse encephalomyelitis virus (TMEV), and FMDV (27). Furthermore, these studies have focused solely on nonstructural proteins 2B, 2C, 3C, and L^pro. However, the role of EMCV structural proteins in evading host immunity has not been explored. Here, we show that EMCV structural protein VP2 is a strong RLR pathway antagonist and acts via directly degrading critical components of the RLR pathway.

RESULTS

EMCV structural protein VP2 promotes viral proliferation in vitro.

To study the effect of EMCV structural proteins on viral replication, EMCV VP1, VP2, and VP3 structural proteins were transfected into HEK293 cells (Fig. 1A) and then infected with EMCV for 36 h. All tested structural proteins significantly increased viral copy numbers compared to the control (Fig. 1B). Notably, the effect of VP2 expression on increased EMCV copy numbers was not due to higher VP2 expression (Fig. 1A and C). Thus, we focused our efforts on VP2 protein from here on. As VP2 is not toxic in HEK293 cells (Fig. 1D), we studied the effect of VP2 expression on EMCV proliferation over time. Indeed, VP2 overexpression significantly increased EMCV copy numbers (Fig. 1E) and titers (Fig. 1F) at 36 h and 48 h postinfection (hpi), suggesting that VP2 can promote EMCV proliferation in host cells.

FIG 1.

EMCV structural protein VP2 promotes viral proliferation in vitro. HEK293 cells were transfected with 1 μg of EMCV viral protein-containing plasmids (pCMV-HA-VP1, pCMV-HA-VP2, and pCMV-Myc-VP3) or an empty vector (EV) plasmid for 24 h before infecting cells with 0.001 MOI of EMCV for 36 h. (A) VP1, VP2, and VP3 protein expressions in HEK293 cells at 24 h were confirmed by immunoblotting. (B) Cell culture supernatants that were infected with EMCV were collected at 36 hpi to measure viral copy numbers by RT-PCR. (C) The empty vector or the pCMV-HA-VP1, pCMV-HA-VP2, or pCMV-Myc-VP3 plasmid (1 μg) was transfected into HEK293 cells for 24 h before measuring VP1, VP2, and VP3 expression by RT-PCR. Data are represented as means ± SD from three independent experiments and were measured in technical duplicates. (D) The effects of increasing doses of EMCV VP2 (0.3 μg, 0.6 μg, 0.9 μg, and 1.2 μg) on the viability of HEK293 cells were measured using CCK-8. (E and F) HEK293 cells were transfected with either pCMV-HA-VP2 (VP2) (1 μg) or an empty plasmid (EV) (1 μg) into HEK293 cells for 24 h and then infected with 0.001 MOI of EMCV before measuring viral copy numbers and titers at different times postinfection (6 h, 12 h, 24 h, 36 h, and 48 h). Quantitative PCR (qPCR) was used to determine viral copy numbers, and a 50% tissue culture infective dose (TCID₅₀) assay was performed for viral titer detection (Reed-Muench method). Data are represented as means ± SD from three independent experiments and were measured in technical duplicate. **, P < 0.01; ***, P < 0.001.

VP2 antagonizes critical components of the type I IFN signaling pathway.

We speculate that VP2-mediated enhancement of EMCV proliferation could occur through the blockade of type I IFN production. Indeed, VP1 and 2B proteins from FMDV have been shown to promote virus proliferation by inhibiting type I IFN signaling (28, 29). As EMCV could induce sufficient IFN-β expression as late as 36 h postinfection (Fig. 2A), we tested whether the overexpression of VP2 could block EMCV-triggered IFN-β responses. To test this, we overexpressed VP2 protein in HEK293 cells before infecting them with EMCV. As predicted, VP2 overexpression significantly inhibited EMCV-triggered IFN-β expression (Fig. 2B and C). Indeed, VP2 overexpression in EMCV-infected cells significantly reduced the expression of a number of important ISGs and components of the RLR and JAK/STAT pathways, including IFN-β (Fig. 2D). Given that EMCV is used as a model organism to study double-stranded RNA (dsRNA) responses (27), we hypothesized that VP2 is blocking key components of the dsRNA-sensing RLR pathway. VP2 overexpression decreased the expression of endogenous proteins MDA5, MAVS, TBK1, IRF3, and p-TBK1 in a dose-dependent manner (Fig. 2E and F). EMCV infection alone decreased MDA5, MAVS, and TBK1 expression at an early time point (12 h) but increased expression later (24 h) (Fig. 2G), consistent with the increased IFN-β expression seen at 24 h but not at 12 h (Fig. 2A). Importantly, EMCV infection in VP2-overexpressing cells significantly reduced the expression of MDA5, MAVS, and TBK1 at 24 h and 36 h (Fig. 2H), suggesting that VP2 can inhibit components of the RLR/MDA5 signaling pathway. As the EMCV nonstructural protein 3C is a known type I IFN pathway antagonist (13), we compared the abilities of the 3C protease and VP2 to inhibit components of the RLR/MDA5 signaling pathway. Indeed, EMCV-induced MAVS, TBK1, and IRF3 expression was dampened in 3C- and VP2-overexpressing cells (Fig. 2I and J). Consistent with this observation, VP2 and 3C promote better EMCV infectivity in HEK293 cells (Fig. 2K). Overall, the VP2-mediated blockade of components of the RLR/MDA5 signaling pathway is comparable to that seen with EMCV 3C protease.

FIG 2.

VP2 antagonizes critical components of the type I IFN signaling pathway. (A) HEK293 cells were either mock infected (mock) or infected (EMCV) with 0.001 MOI of EMCV, and IFN-β mRNA was detected by RT-PCR at various times postinfection (12 h, 24 h, and 36 h). (B) HEK293 cells were transfected with pCMV-HA-VP2 (1 μg) for 24 h before infecting cells with EMCV (0.001 MOI) for a further 36 h, and the IFN-β concentration in the cell culture supernatant was determined by an ELISA. (C) HEK293 cells were transfected with pCMV-HA-VP2 (1 μg) for 24 h before infecting cells with EMCV (0.001 MOI) for a further 36 h, and IFN-β mRNA expression was detected by RT-PCR. (D) HEK293 cells were transfected with either pCMV-HA-VP2 (VP2) (1 μg) or an empty vector plasmid (EV) (1 μg) for 24 h before infecting cells with 0.001 MOI of EMCV for a further 36 h. STAT1, STAT2, STAT3, IRF9, ISG15, ISG56, MX1, OAS1, IFITM1, IFITM2, and IFITM3 mRNA expression levels were measured by RT-PCR. (E) Increasing doses (0.3 μg, 0.6 μg, 0.9 μg, and 1.2 μg) of the pCMV-HA-VP2 plasmid were transfected into HEK293 cells for 36 h, and the expressions of MDA5, MAVS, TBK1, and IRF3 were detected by immunoblotting. Increasing doses (0.5 μg, 1.0 μg, and 1.5 μg) of the pCMV-HA-VP2 plasmid and 0.25 μg/ml poly(I·C) were transfected into HEK293 cells for 36 h, and the expression of p-TBK1 was detected by immunoblotting. β-Actin was used as a loading control. (F) HEK293T cells were cotransfected with pCMV-HA-VP2 (1.2 μg) and poly(I·C) (0.25 μg/ml) for 36 h before cell extracts were subjected to immunoblotting for MDA5, MAVS, TBK1, and IRF3 proteins. β-Actin served as a loading control. (G) The expressions of MDA5, MAVS, TBK1, and IRF3 in cells infected with EMCV (0.001 MOI) at 12 h, 24 h, and 36 h were detected by immunoblotting. (H) The effects of VP2 on the expression of MDA5, MAVS, TBK1, and IRF3 in cells infected with EMCV at 12 h, 24 h, and 36 h were detected by immunoblotting. β-Actin was used as a loading control, whereas VP1 was used as a marker for EMCV infection. (I) HEK293 cells were transfected with either the 3C or VP2 plasmid (1 μg) for 24 h before performing RT-PCR to screen for transgene expression. Data are represented as means ± SD from three independent experiments and were measured in technical duplicates. (J) HEK293 cells overexpressed 3C (pCMV-HA-3C) (1 μg) or VP2 (1 μg) for 24 h and, following infection with EMCV (0.001 MOI), for another 36 h. Cells were collected and immunoblotted for MAVS, TBK1, and IRF3 expression. β-Actin was used as a loading control, whereas VP1 was used as a marker for EMCV infection. (K) HEK293 cells were transfected with either the 3C or VP2 plasmid (1 μg) for 24 h and then infected with EMCV at an MOI of 0.001 for 36 h. Virus was collected for RNA, and RT-PCR was performed to measure EMCV copy numbers. Data are represented as means ± SD from three independent experiments and were measured in technical duplicates. *, P < 0.05; ***, P < 0.001.

VP2 antagonizes poly(I·C)-induced IFN-β activation.

We then wanted to show whether dsRNA-mediated IFN induction by poly(I·C), a dsRNA synthetic mimic that is known to trigger the RLR/MDA5 signaling pathway (30), could be blocked by VP2. VP2 significantly inhibited poly(I·C)-induced IFN-β activation in a dose-dependent manner (Fig. 3A). Increasing amounts of VP2 also inhibited the transcriptional levels of poly(I·C)-induced MDA5, MAVS, TBK1, and IRF3 expression (Fig. 3B to E).

FIG 3.

VP2 suppresses poly(I·C)-induced IFN-β activation. (A) HEK293 cells were cotransfected with increasing doses of pCMV-HA-VP2 (0.3 μg, 0.6 μg, 0.9 μg, and 1.2 μg) and poly(I·C) (0.25 μg/ml) for 36 h before measuring IFN-β secretion by an ELISA. (B to E) HEK293 cells were cotransfected with increasing doses of pCMV-HA-VP2 (0.3 μg, 0.6 μg, 0.9 μg, and 1.2 μg) and poly(I·C) (0.25 μg/ml) for 36 h, and RT-PCR was used to detect RLR pathway gene MDA5, MAVS, TBK1, and IRF3 mRNA expression. Data are represented as means ± SD from three independent experiments and were measured in technical duplicate. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

VP2 directly interacts with MDA5, MAVS, and TBK1.

To further investigate whether the VP2-mediated blockade of IFN-β responses is due to VP2 directly associating with components of the RLR signaling pathway, MDA5-, MAVS-, TBK1-, and IRF3-expressing plasmids were cotransfected with increasing amounts of the VP2-expressing plasmid in HEK293 cells. We found that VP2 significantly decreased MDA5 mRNA expression in a dose-dependent manner (Fig. 4A). In addition, VP2 also significantly inhibited IFN-β induced by MDA5 at both the mRNA (Fig. 4B) and protein (Fig. 4C) levels. By immunoblotting, we showed that VP2 can inhibit MDA5, MAVS, and IRF3 expression induced by MDA5 (Fig. 4D). Similar findings were also seen in cells cotransfected with increasing doses of VP2 and MAVS (Fig. 4E to H)-, TBK1 (Fig. 4I to L)-, and IRF3 (Fig. 4M to P)-expressing plasmids. The proteasome pathway is used by host cells to degrade proteins quickly and efficiently with the help of ubiquitin (Ub) molecules and related enzymes (31). To determine whether this pathway is involved in the VP2-mediated loss of MDA5, MAVS, TBK1, and IRF3, the proteasome inhibitor MG132 was used. MG132 treatment restored MDA5, MAVS, TBK1, and IRF3 levels in VP2-overexpressing cells activated with either poly(I·C) (Fig. 5A) or exogenous IRF3 (Fig. 5B). The lysosome-dependent pathway is another pathway for intracellular protein degradation, so chloroquine (CQ) (lysosome pathway inhibitor) was added to HEK293 cells cotransfected with VP2 and poly(I·C). In addition to MDA5 and IRF3, CQ treatment attenuates the inhibition of MAVS and TBK1 expression by VP2 (Fig. 5C). Collectively, these results suggest that VP2 can inhibit the expression of key RLR proteins and the activation of the RLR signaling pathway induced by MDA5, MAVS, TBK1, and IRF3, consistent with that seen with poly(I·C)-induced RLR signaling (Fig. 3). Furthermore, we showed that the VP2-mediated loss of the RLR signaling pathway components occurs through the proteasomal and lysosomal degradation pathways. Next, we wanted to confirm whether VP2 physically interacts with components of the RLR signaling pathway, MDA5, MAVS, and TBK1. Coimmunoprecipitation (co-IP) assays and confocal microscopy reveal that MDA5 (Fig. 6A to C), MAVS (Fig. 6D to F), and TBK1 (Fig. 6G to I) can all directly bind to VP2. Although colocalization of IRF3 and VP2 was seen by confocal microscopy (Fig. 6N), no direct interaction between VP2 and either exogenous IRF3 (Fig. 6J and K) or endogenous IRF3 (Fig. 6L and M) was observed. It is possible that VP2 and IRF3 interactions may require the participation of other host proteins. Collectively, the data indicated that EMCV VP2 can specifically interact with host MDA5, MAVS, and TBK1 proteins.

FIG 4.

VP2 inhibits the activation of the RLR pathway induced by MDA5, MAVS, TBK1, and IRF3. HEK293 cells were cotransfected with increasing doses of pCMV-HA-VP2 (0.3 μg, 0.6 μg, 0.9 μg, and 1.2 μg) with 0.5 μg of pCMV-FLAG-MDA5 for 36 h. (A) MDA5 mRNA expression was detected by RT-PCR. (B and C) IFN-β mRNA expression and protein concentrations were measured by RT-PCR and ELISAs, respectively. (D) Protein extracts were also made to measure the effect of increasing VP2 doses (0.6 μg, 0.9 μg, and 1.2 μg) with MDA5 on the expression of RLR pathway factors by immunoblotting. HEK293 cells were cotransfected with increasing doses of pCMV-HA-VP2 (0.6 μg and 1.2 μg) with 0.5 μg of pCMV-FLAG-MAVS for 36 h. (E) MAVS mRNA expression was detected by RT-PCR. (F and G) IFN-β mRNA expression and protein concentrations were measured by RT-PCR and ELISAs, respectively. (H) Protein extracts were also made to measure the effect of increasing VP2 doses (0.6 μg, 0.9 μg, and 1.2 μg) with MAVS on the expression of RLR pathway components by immunoblotting. HEK293 cells were cotransfected with increasing doses of pCMV-HA-VP2 (0.6 μg and 1.2 μg) with 0.5 μg of pCMV-FLAG-TBK1 for 36 h. (I) TBK1 mRNA expression was detected by RT-PCR. (J and K) IFN-β mRNA expression and protein concentrations were measured by RT-PCR and ELISAs, respectively. (L) Protein extracts were also made to measure the effect of increasing VP2 doses (0.6 μg, 0.9 μg, and 1.2 μg) with TBK1 on the expression of RLR pathway components by immunoblotting. HEK293 cells were cotransfected with increasing doses of pCMV-HA-VP2 (0.6 μg and 1.2 μg) with 0.5 μg of pCMV-FLAG-IRF3(5D) for 36 h. (M) IRF3 mRNA expression was detected by RT-PCR. (N and O) IFN-β mRNA expression and protein concentrations were measured by RT-PCR and ELISAs, respectively. (P) Protein extracts were also made to measure the effect of increasing VP2 doses (0.6 μg, 0.9 μg, and 1.2 μg) with IRF3(5D) on the expression of RLR pathway components by immunoblotting. β-Actin was used as a loading control. Data are represented as means ± SD from three independent experiments and were measured in technical duplicate. **, P < 0.01; ***, P < 0.001.

FIG 5.

VP2 degrades key proteins of the RLR signaling pathway through the proteasomal degradation pathway. (A and B) Poly(I·C) (0.25 μg/ml) and pCMV-HA-VP2 (0.5 μg) (A) or pCMV-FLAG-IRF3(5D) (0.5 μg) and pCMV-HA-VP2 (0.5 μg) (B) were cotransfected into HEK293 cells for 24 h and treated with either the proteasomal inhibitor MG132 or dimethyl sulfoxide (DMSO) (vehicle control) for a further 6 h before immunoblotting for MDA5, MAVS, TBK1, and IRF3. (C) Poly(I·C) (0.25 μg/ml) and pCMV-HA-VP2 (0.5 μg) were cotransfected into HEK293 cells for 24 h and treated with either the lysosomal inhibitor CQ or DMSO (vehicle control) for a further 12 h before immunoblotting for MDA5, MAVS, TBK1, and IRF3 proteins. β-Actin was used as the loading control.

FIG 6.

VP2 protein directly interacts with MDA5, MAVS, and TBK1. (A, B, D, E, G, H, J, and K) HEK293 cells were transfected with pCMV-HA-VP2 (0.5 μg) or pCMV-HA (0.5 μg) together with either pCMV-FLAG-MDA5 (0.5 μg) (A and B), pCMV-FLAG-MAVS (0.5 μg) (D and E), pCMV-FLAG-TBK1 (0.5 μg) (G and H), or pCMV-FLAG-IRF3 (0.5 μg) (J and K) for 12 h before performing coimmunoprecipitations and immunoblotting (IB) with either anti-FLAG or anti-HA antibodies. WB, Western blotting. (C, F, and I) The pCMV-HA-VP2 (1 μg) or pCMV-HA (mock) (1 μg) plasmid was transfected into HEK293 cells for 12 h before fixing cells for immunofluorescence imaging under a confocal microscope for VP2 (green), nuclei (blue), and MDA5, MAVS, and TBK1 proteins (red). DAPI, 4′,6-diamidino-2-phenylindole. (L and M) The pCMV-HA-VP2 (1 μg) plasmid was transfected into HEK293 cells for 12 h before performing coimmunoprecipitations and immunoblotting with either anti-IRF3 or anti-HA antibody. (N) pCMV-HA-VP2 (0.5 μg) and pCMV-FLAG-IRF3 (0.5 μg) were cotransfected into HEK293 cells for 12 h before fixing cells for immunofluorescence imaging under a confocal microscope for VP2 (red), nuclei (blue), and IRF3 (green). Bar = 5 μm. (O) HEK293 cells were cotransfected with HA-VP2, FLAG-MDA5, and HA-Ub at 30 h posttransfection. The MDA5-ubiquitin complexes were immunoprecipitated using anti-FLAG antibody and immunoblotted with anti-HA antibody to detect ubiquitinated proteins.

Proteins degraded via the proteasome pathway must be ubiquitinated first. As the above-described data demonstrated that VP2 interacted with MDA5, we next tested whether VP2 affects MDA5 ubiquitination. The ubiquitination assay showed that VP2 could increase MDA5 polyubiquitination (Fig. 6O).

The C-terminal domain of EMCV VP2 interacts with MDA5, MAVS, and TBK1 and inhibits the virus-triggered IFN-β signaling pathway.

To further determine the structural domains of EMCV VP2 that directly interact with MDA5, MAVS, and TBK1, we generated a series of truncated mutants, named HA-VP2-1–140aa, HA-VP2-90–215aa, and HA-VP2-170–256aa. C-terminal amino acids (aa) 170 to 256 of VP2 interacted with MDA5, MAVS, and TBK1 (Fig. 7A and B). To functionally confirm this, we showed that only full-length VP2 and C-terminal aa 170 to 256 of VP2 inhibited poly(I·C)-triggered IFN-β activation (Fig. 7C). Collectively, these results suggest that the C-terminal domain of EMCV VP2 interacts with MDA5, MAVS, and TBK1 and inhibits the virus-triggered IFN-β signaling pathway.

FIG 7.

C-terminal aa 170 to 256 of VP2 interact with MDA5, MAVS, and TBK1. (A and B) HEK293 cells were transfected with VP2 truncated mutants (1 μg) or pCMV-HA (1 μg) together with either pCMV-FLAG-MDA5 (0.5 μg), pCMV-FLAG-MAVS, or pCMV-FLAG-TBK1 (0.5 μg) for 30 h before performing coimmunoprecipitations and immunoblotting with either anti-FLAG or anti-HA antibodies. (C) HEK293 cells were transfected with VP2 truncated mutants (1 μg) or pCMV-HA (1 μg) for 24 h following infection with 0.001 MOI of EMCV for 36 h. Culture supernatants were collected for an ELISA to detect secreted IFN-β concentrations. Data are represented as means ± SD from three independent experiments and were measured in technical duplicates. ***, P < 0.001; NS, not significant.

DISCUSSION

Innate immunity is the first line of host defense against pathogenic infections. However, throughout evolution, viruses have acquired numerous strategies to evade the host immune system (32), notably the type I IFN antiviral response. Previous studies have shown the ability of EMCV to evade host immunity via interfering with IFN signaling (10–15). To our knowledge, we are the first to show the ability of the EMCV structural protein VP2 to antagonize the components of the RLR/MDA5 signaling pathway. Although we showed that VP2 directly interacts with MDA5, MAVS, and TBK1, we have yet to prove whether this is directly responsible for RLR/MDA5 signaling antagonism. A closely related structural viral protein, VP3, encoded by FMDV, has been shown to inhibit Sendai virus-triggered activation of IRF3 and RIG-I/MDA5 expression (33, 34). Those researchers also showed that VP3 interacts with MAVS via the C terminus (aa 111 to 220) of VP3, and our research indicated that the C-terminal aa 170 to 256 of the EMCV VP2 protein are responsible for MDA5, MAVS, and TBK1 binding and that VP2 abrogates the IFN-β signaling pathway by degrading MDA5, MAVS, TBK1, and IRF3. This degrading effect was associated with the proteasome and lysosome pathways.

We and others have used HEK293 cells as a model cell line for EMCV infection studies (10, 11, 35, 36). We are aware of the limitations that using this cell line poses, as it is inherently virally transformed and expresses the E1A oncogene from adenovirus (Adv), and also of the artificial nature of using a cell line model. Although it is known that E1A could impact type I IFN signaling, this effect occurs at an earlier time and is transient in nature (37–39). In addition, inhibition of Adv replication by IFN-α and IFN-γ is the consequence of repressing E1A immediate early gene product transcription. Both IFN-α and IFN-γ impede the association of the transactivator GABP (the cellular transcription factor GA-binding protein) with the E1A enhancer region during the early phases of infection (40). This could explain the observed late kinetics of IFN-β induction (36 h postinfection) with EMCV infection in our experiments (Fig. 2A). Importantly, VP2-mediated blockade of EMCV-triggered IFN activation (Fig. 2D) and EMCV replication (Fig. 1E and F) was notable only at this and later time points. Here, we further show that EMCV VP2 can inhibit the RLR/MDA5 and JAK/STAT signaling pathways (Fig. 2E).

Several viruses have been reported to antagonize the RLR pathway. Previous studies have shown that FMDV VP3 inhibits the expression of the key adaptor molecule in the RLR pathway, MAVS, as a strategy for evading innate antiviral immunity during the signal transduction phase (33, 34). Indeed, FMDV 3A was identified as a negative regulator of the virus-triggered IFN-β signaling pathway by inhibiting the expression of RIG-I, MDA5, and VISA (41). FMDV 3C protease is known to cleave nuclear transcription factor kappa B (NF-κB) essential modulator (NEMO), a bridging adaptor protein essential for the activation of both the NF-κB and interferon-regulatory factor signaling pathways, to impair innate immune signaling (42). On the other hand, FMDV L protease targets laboratory of genetics and physiology 2 (LGP2), a nonsignaling member of the RLRs, for cleavage, resulting in lower levels of IFN-β and antiviral activity (43). The NS1 protein of West Nile virus (WNV) was also reported to be an antagonist of IFN induction via interacting with and degrading RIG-I and MDA5 (44).

In conclusion, this study provides a new perspective on the role of EMCV VP2 in regulating the RLR/MDA5 signaling pathway. We reveal a new mechanism of EMCV VP2 evading the host’s natural immune response by interacting with MDA5, MAVS, and TBK1 to lead to its degradation (Fig. 8). These findings shed light on how EMCV evades host immunity by interrupting type I IFN signaling pathways.

FIG 8.

Model of how EMCV VP2 protein inhibits the RLR/MDA5 signaling pathway. EMCV VP2 protein degrades MDA5, MAVS, TBK1, and IRF3 through proteasomal and lysosomal pathways and inhibits the activation of the IFN-β signaling pathway through interaction with MDA5, MAVS, and TBK1. UPP, ubiquitin proteasome pathway.

MATERIALS AND METHODS

Cells, viruses, plasmids, and reagents.

Human embryonic kidney HEK293 cells and baby hamster kidney BHK-21 cells were obtained from the ATCC and cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% new bovine serum (NBS) at 37°C in a 5% CO₂ incubator. SYBR green real-time supermix was purchased from Bio-Rad (Hercules, CA, USA). EMCV strain PV21 was purchased from the ATCC (GenBank accession no. X74312) and propagated in BHK-21 cells. Infected cell supernatants were collected and stored at −80°C. Anti-MDA5 (catalog no. 66770-1-Ig), anti-MAVS (catalog no. 66911-1-Ig), anti-hemagglutinin (HA) tag, anti-IRF3, and anti-p-IRF3 at Ser396 (catalog no. 4947s) antibodies (Abs) were purchased from Proteintech (China); anti-FLAG tag Ab was purchased from Sigma (St. Louis, MO, USA) (catalog no. F1804); TBK1 antibody (catalog no. 3013S) and phospho-TBK1/NAK (Ser172) (D52C2) rabbit monoclonal antibody (mAb) (catalog no. 5483S) were bought from Cell Signaling Technology (CST); anti-β-actin Ab was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA) (catalog no. sc-81178); and Cy3-conjugated goat anti-rabbit IgG (red) and fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG (green) were purchased from Boster (China). The proteasome inhibitor MG132 was purchased from Beyotime (China) (catalog no. S1748). The lysosomal pathway inhibitor chloroquine (CQ) was purchased from InvivoGen. Anti-VP1 antibody was kindly provided by Juan Bai of Nanjing Agricultural University.

To construct plasmids encoding HA-tagged (VP1, VP2, and 3C) or Myc-tagged (VP3) EMCV proteins, genes were generated from EMCV PV21 strain-derived cDNAs and cloned into the pCMV-HA or pCMV-Myc vector. Truncated VP2 plasmids HA-VP2-1–140aa, HA-VP2-90–215aa, and HA-VP2-170–256aa were constructed in-house. The Ub gene was generated from HEK293 cell cDNA, and plasmid HA-Ub was constructed and verified by sequencing. The primer sequences used in this study are available upon request. All plasmids were verified by sequencing. IRF3 constitutively active mutant IRF3(5D)-FLAG, pCMV-MDA5-FLAG, pCMV-MAVS-FLAG, pCMV-TBK1-FLAG, and pCMV-IRF3-FLAG expression plasmids were all constructed in-house.

Transfections, real-time PCR assays, and enzyme-linked immunosorbent assays.

HEK293 cells in 6-well plates were transfected with various plasmids using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s protocol and infected with EMCV at a multiplicity of infection (MOI) of 0.001 for 36 h. mRNA expression levels of MDA5, RIG-I, MAVS, IFN-β, ISG15, ISG56, OAS1, and MX1 were quantitated by real-time PCR (RT-PCR). Total RNA was extracted from HEK293 cells using RNAiso Plus (TaKaRa, Dalian, China) according to the manufacturer’s protocol. RT-PCR was performed with SYBR green real-time PCR master mixes (Applied Biosystems, USA) on the ABI 7500 real-time PCR system (Applied Biosystems). β-Actin was used as the reference gene, and all data were determined by the threshold cycle (2^Δ−CT) method, with the results expressed as relative fold changes. Primers for RT-PCR are available upon request. Concentrations of IFN-β in cell culture supernatants were determined by using a human IFN-β enzyme-linked immunosorbent assay (ELISA) kit (Enzyme-Linked Biotechnology, Shanghai, China), according to the manufacturer’s instructions.

Coimmunoprecipitation and immunoblotting.

HEK293 cells were lysed in l ml of lysis buffer (20 mM Tris-HCl [pH 7.4 to 7.5], 150 mM NaCl, 1 mM EDTA, 1% NP-40, 10 mg/ml aprotinin, 10 mg/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride). For coimmunoprecipitation (co-IP) experiments, 500 μl of the cell lysate was incubated with 0.5 mg of the appropriate antibody and 1.5 mg of Dynabeads protein G (catalog no. 10004D; Invitrogen) at 4°C for 2 h. After beads were washed in phosphate-buffered saline (PBS)–Tween 20, precipitates were fractionated using SDS-PAGE, and immunoblotting was performed. Primary antibodies were incubated at 37°C for 2 h with the appropriate primary antibodies (mouse anti-FLAG at 1:3,000, mouse anti-HA at 1:3,000, or mouse anti-β-actin at 1:5,000) before incubation with horseradish peroxidase (HRP)-labeled goat anti-mouse antibody at 37°C for 45 min. Membranes were treated with ECL (Thermo Fisher Scientific), and bound proteins were visualized with the Amersham (Boston, MA, USA) Imager 600 imaging system.

Immunofluorescence microscopy.

HEK293 cells were grown on 35-mm glass-bottomed plates and cotransfected with the indicated plasmids. At 12 h posttransfection, cells were washed in PBS, fixed in 4% paraformaldehyde, and permeabilized with 0.2% Triton X-100 for 15 min at room temperature (RT) before blocking overnight in PBS containing 5% bovine serum albumin at 4°C. Primary antibodies were incubated at a dilution of 1:100 for 2 h at RT. Secondary antibodies used were Cy3-conjugated goat anti-rabbit IgG and/or FITC-conjugated goat anti-mouse IgG at a dilution of 1:100 for 1 h at RT. The cells were washed with PBS and observed on a confocal microscope (LSM900; Zeiss, Oberkochen, Germany).

MDA5 polyubiquitination assay.

HEK293 cells were cotransfected with HA-VP2, FLAG-MDA5, and HA-Ub expression vectors at a 1:1:1 ratio using the Lipofectamine 2000 method. Protein was extracted at 30 h posttransfection. The MDA5-ubiquitin complexes were immunoprecipitated using anti-FLAG antibody M2-conjugated beads (catalog no. A2220; Sigma) and immunoblotted with anti-HA antibody to detect ubiquitinated proteins.

EMCV infection and cell viability assays.

HEK293 cells were transfected with the EMCV protein expression plasmids using Lipofectamine 2000 (Thermo Scientific) for 24 h and subsequently infected with EMCV at an MOI of 0.001. The viral inoculum was removed 2 h later, and cells were washed twice in PBS (pH 7.4) before replacement with DMEM containing 3% NBS. IFN-β mRNA and EMCV genome copies in HEK293 cells were assessed by RT-PCR. Cell viability was measured using cell counting kit 8 (CCK-8) (Beyotime, China) according to the manufacturer’s protocol.

Statistical analysis.

All data were analyzed with GraphPad Prism 5.0 using Student’s t tests. Data are presented as the means ± standard deviations (SD) from three independent experiments. P values are indicated using asterisks (*, P < 0.05; **, P < 0.01; ***, P < 0.001).