key: cord-0963464-y2voi0n5
authors: Kim, Hyejin; Kim, Ah-Young; Choi, Jieun; Park, Sun Young; Park, Sang Hyun; Kim, Jae-Seok; Lee, Sim-In; Park, Jong-Hyeon; Park, Choi-Kyu; Ko, Young-Joon
title: Foot-and-Mouth Disease Virus Evades Innate Immune Response by 3C-Targeting of MDA5
date: 2021-01-29
journal: Cells
DOI: 10.3390/cells10020271
sha: 862f529547604ad708161a689c63b59774a875f0
doc_id: 963464
cord_uid: y2voi0n5

Foot-and-mouth disease (FMD) is a highly contagious disease caused by FMD virus (FMDV) in cloven-hoofed animals. Retinoic acid-inducible gene I (RIG-I) and melanoma differentiation-associated gene 5 (MDA5) are representative receptors in the cytoplasm for the detection of viral RNA and trigger antiviral responses, leading to the production of type I interferon. Although MDA5 is a crucial receptor for sensing picornavirus RNA, the interplay between MDA5 and FMDV is relatively unknown compared to the interplay between RIG-I and FMDV. Here, we observed that the FMDV infection inhibits MDA5 protein expression. Of the non-structural proteins, the Lb and 3C proteinases (Lb(pro) and 3C(pro)) were identified to be primarily responsible for this inhibition. However, the inhibition by 3C(pro) was independent of proteasome, lysosome and caspase-dependent pathway and was by 3C protease activity. A direct interaction between 3C(pro) and MDA5 protein was observed. In conclusion, this is the first report that 3C(pro) inhibits MDA5 protein expression as a mechanism to evade the innate immune response during FMDV infection. These results elucidate the pathogenesis of FMDV and provide fundamental insights for the development of a novel vaccine or therapeutic agent.

The Foot-and-Mouth Disease virus (FMDV) is a single-stranded, positive-sense RNA virus that belongs to the genus Apthovirus of the family Picornaviridae [1] . The FMDV causes foot-and-mouth disease (FMD), which is highly contagious in various clovenhoofed animals, such as pigs, cows and goats [2, 3] . The FMDV genome is 8.5 kb and consists of one open reading frame (ORF), which encodes a single polyprotein that is subsequently proteolyzed into four structural proteins (SPs: VP1, VP2, VP3 and VP4) and eight nonstructural proteins (NSPs: L pro , 2A, 2B, 2C, 3A, 3B, 3C pro and 3D) [2, 4, 5] . The innate immune system plays an important role in defending against invading pathogens. Pathogen recognition receptors (PRRs) recognize viral pathogen-associated molecular patterns (PAMPs) in host cells and activate signaling cascades, leading to the expression of the host immune response, type 1 interferons (IFNs; alpha/beta interferon [IFN-α/β]) genes and pro-inflammatory cytokines [6] . PRRs involved in RNA viral genome detection include Toll-like receptor (TLR) and retinoic acid-inducible gene I (RIG-I)-like receptor (RLR) [7] . TLRs are expressed on the surface or endosomal compartments of macrophages, dendritic cells and other immune cells [8] . The RLR family mainly encompasses two members: RIG-I and melanoma differentiation-associated gene 5 (MDA5),

Porcine Kidney-15 (PK-15; ATCC CCL-33, Manassas, VA, USA) cells, human embryonic kidney 239T (HEK 293T; ATCC, Manassas, VA, USA) cells and baby hamster kidney-21 (BHK-21; ATCC C-13, Manassas, VA, USA) cells were grown in Dulbecco Modified Eagle's medium (DMEM; Corning Inc., Corning, NY, USA) containing 10% fetal bovine serum (FBS; Gibco, Grand Island, NY, USA) and 1% antibiotic-antimycotic (Anti-Anti; Gibco) at 37 • C with 5% CO 2 . The BHK-21 cell is commonly used for FMDV propagation and titration. The FMDV serotype O strain (Boeun/SKR/2017) was propagated in BHK-21 cells. The FMDV type O strain was inoculated at a multiplicity of infection (MOI) of 0.5 on PK-15 cells of 70-80% confluency and was incubated at 37 • C with 5% CO 2 . Viral titers were calculated by endpoint titration on BHK-21 cells using the Spearman-Kärber method and the results were expressed as log 10 TCID 50 /0.1 mL [26, 27] .

The commercial antibodies for target proteins used in this study were as follows: anti-HA-tag polyclonal antibody (MBL, Nagoya, Japan), anti-HA-tag monoclonal antibody (Cell Signaling Technology, Danvers, MA, USA), anti-MDA5 polyclonal antibody (Abcam, Cambridge, UK), anti-MAVS polyclonal antibody (Abcam), anti-IRF3 polyclonal antibody (Cell Signaling Technology), phospho-IRF3 monoclonal antibody (Cell Signaling Technology), OctA (FLAG)-Probe monoclonal antibody (Santa Cruz Technology, Dallas, TX, USA), anti-mouse IgG (Cell Signaling Technology), anti-rabbit IgG (H+L) secondary antibody (Invitrogen, Meridian Rd., Rockford, IL, USA), anti-mouse IgG (H+L) secondary antibody (Invitrogen), anti-mouse IgG-Fc Fragment antibody (Bethyl Laboratories, Montgomery, Texas, USA) and anti-β-actin polyclonal antibody (Cell Signaling Technology). The chemical reagents used for the inhibitor assay used in this study include: the proteasome inhibitor MG132 (Millipore, Billerica, MA, USA), the lysosome inhibitor, chloroquine diphosphate (CQ; Sigma, St. Louis, MO, USA) and the caspase inhibitor, benzyloxy-carbonyl-Val-Ala-Asp (OMe) fluoromethylketone (Z-VAD-FMK; Merck, Darmstadt, Germany).

To generate expression vectors for FMDV Lb (GenBank accession No. AJ539139), 2B (GenBank accession No. AY312587) and 2C (GenBank accession No. AY312587), 3A (GenBank accession No. AY312587) and 3B (GenBank accession No. AY312587) and 3C (GenBank accession No. AJ539139) or 3D (GenBank accession No. AJ539139), each NSP cDNA was cloned into the pCMV-HA vector to construct plasmids expressing HA-tagged protein (Enzynomics, Daejeon, Korea). Similarly, the cDNA of porcine MDA5 (GenBank accession No. EU006039) was cloned into the pCMV-HA vector or p3xFLAG-CMV vector (Enzynomics). Specific mutations of 3C (H46Y, D84N and C163G) were introduced to pCMV-HA-3C (Enzynomics).

HEK293T cells, approximately 70-80% confluent in 6-well plates on the day of transfection, were washed once with pre-warmed phosphate-buffered saline (PBS; Corning Inc.) and were completely replaced by Opti-MEM I (1×) Reduced Serum Medium (Invitrogen, Grand Island, NY, USA) per plate, followed by 30 min incubation at 37 • C. According to the manufacturer's instructions for FuGENE HD Transfection Reagent (Promega, Madison, WI, USA), the transfection reagent and plasmid DNA were mixed gently at a ratio of 3:1. The transfection reagent/DNA mixture was incubated at room temperature (RT) for 15 min and added to the plates. To assay proteasome, lysosome and caspase inhibition, media of transfected cells were supplemented with the proteasome inhibitor MG132, the lysosome inhibitor CQ and the caspase inhibitor Z-VAD-FMK, respectively. After 4 h posttransfection (hpt), the medium was replaced with 2% FBS and the plates were incubated in an incubator at 37 • C with 5% CO 2 .

Total RNA was extracted from FMDV-infected PK-15 cells or plasmid-transfected HEK293 cells using the RNeasy mini kit (Qiagen, Valencia, CA, USA) and the cDNA was synthesized by the Super Script III First-Strand Synthesis System (Invitrogen, Carlsbad, CA, USA) from the extracted RNA, according to the manufacturer's protocol. The level of mRNA for FMDV was detected using an AccuPower FMDV Real-time RT-PCR Kit (Bioneer, Daejeon, Korea). To quantify the abundance of mRNA, the quantitative real-time RT-PCR experiment was performed with iQ SYBR Green Super mix (Bio-Rad, Hercules, CA, USA) on a CFX96 Real-Time System (Bio-Rad, Munich, Germany) under the following conditions: 95 • C for 5 min, followed by 40 cycles of 95 • C for 10 s and 60 • C for 30 s. The Glyceraldehyde-3-Phosphate dehydrogenase (GAPDH) gene was used as an internal control. Relative expression of mRNA expression was determined by the threshold cycle (2 −∆∆CT ) method. The primer sequences for target gene expression are listed in Table 1 [28] .

The FMDV-infected PK-15 and plasmid-transfected HEK293T cells were washed three times with pre-warmed PBS (Corning Inc.) and lysed in chilled Pierce radioimmunoprecipitation assay buffer (RIPA buffer; Thermo Fisher Scientific, Rockford, IL, USA) or Pierce IP lysis buffer (Thermo Fisher Scientific) containing protease inhibitor cocktail (Complete Mini, Roche Diagnostics, Germany) and phosphatase inhibitor (Thermo Fisher Scientific) using a cell scraper on ice. To remove the cell debris, the lysates were centrifuged at 13,000× g for 20 min at 4 • C and the supernatant was transferred to a new tube for further analysis. Total protein concentrations were determined using the Pierce™ BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's instructions.

For immunoprecipitation (IP), 500 µL of whole cell lysate was incubated with 2 µg of OctA (FLAG)-Probe monoclonal antibody or mouse IgG antibody overnight at 4 • C with shaking. The next day, the lysates were incubated with 100 µL of protein A/G agarose resin (Thermo Fisher Scientific) at 4 • C for 2 h and centrifuged 3000× g for 1 min to pellet resin. The agarose resin was washed five times with 500 µL of IP lysis buffer containing protease inhibitor cocktail and the antigen-antibody complex is eluted from the resin by heating in 100 µL of SDS loading buffer without Sample Reducing agent (Invitrogen) for 10 min at 50 • C and centrifuged 3000× g for 1 min.

For Western blot analysis, the samples were mixed with 4× lithium dodecyl sulfate sample buffer (LDS; Invitrogen) containing Sample Reducing agent (Invitrogen) and were then heated for 10 min at 95 • C. Target proteins were separated on 4-12% Bis-Tris gels (Invitrogen) and transferred onto a PVDF membrane (Invitrogen) using the iBlot gel transfer device (Invitrogen). The membranes were blocked with 2% skim milk in PBS containing 0.1% Tween 20 (PBS-T) for an hour at RT with shaking, washed three times with PBS-T for 10 min and then incubated with appropriate primary antibodies overnight at 4 • C with shaking. The next day, the membranes were washed three times with PBS-T and incubated with suitable secondary antibodies for an hour at RT with shaking. The antibodyantigen complexes were visualized with ECL Western blotting substrate (Amersham, Buckinghamshire, UK) using the Azure C600 device (Azure Biosystem, Dublin, CA, USA). Band intensities were quantified using Image J software (Wayne Rasband, NIH, Bethesda, MD, USA). 

Primer

The statistical significance was evaluated with one-way ANOVA followed by Tukey's post hoc test using GraphPad Prism Version 5 (GraphPad Software, San Diego, AD, USA). All data are representative of three independent experiments and values are represented as the mean ± standard error of the mean (SEM). P-values < 0.05 were considered statistically significant.

To determine whether FMDV infection induced of IFN-β production through RLRmediated antiviral pathways, PK-15 cells were infected with FMDV at a multiplicity of infection (MOI) of 0.5 for 0, 4, 8, 12, 16 and 24 h ( Figure 1 ). The mRNA and protein levels were examined by real-time RT-PCR and western blot analysis. FMDV mRNA levels and viral titers were highest at 12 h post-infection (hpi) and decreased from 16 hpi (Figure 1a ). The mRNA levels of MDA5, RIG-I and IFN-β gradually increased and were highest at 16 hpi ( Figure 1b) . The levels of the three mRNAs decreased after 16 hpi. The protein expression levels of MDA5, RIG-I and MAVS gradually decreased in a timedependent manner as FMDV replicated in the PK-15 cells. In particular, the expression of the MDA5 protein was significantly reduced 16 hpi compared to that of the RIG-I protein ( Figure 1c ). While the amount of IRF3 was not affected by FMDV infection in PK-15 cells, the phosphorylated form of IRF3 (pIRF3), which induces IFN-β transcription, increased until 16 hpi and decreased at 24 hpi ( Figure 1d ). These results suggest that MDA5 plays a more important role than RIG-I in innate immune evasion of FMDV. All experiments were performed in triplicate and presented as mean ± SEM (*, p < 0.05; **, p < 0.01; ***, p < 0.001 and n.s., not significant). The abundance of proteins were calculated by Image J software to quantify the band intensity, normalized to β-actin and respectively compared with the 0 hpi. All experiments were performed in triplicate and presented as mean ± SEM (*, p < 0.05; **, p < 0.01; ***, p < 0.001 and n.s., not significant). The abundance of proteins were calculated by Image J software to quantify the band intensity, normalized to β-actin and respectively compared with the 0 hpi.

FMDV infection contributed to the decline in MDA5 protein expression. Therefore, to identify the FMDV NSP that affect the expression of endogenous MDA5 and RIG-I proteins, HEK293T cells were transfected with the plasmid encoding each viral NSP. After 16 h, endogenous MDA5 and RIG-I were detected by western blotting in the transfected cells ( Figure 2a ) and quantified using ImageJ software (Figure 2b ). The Lb pro and 3C pro of FMDV NSPs significantly reduced endogenous MDA5 expression compared to HAvector-transfected cells. Meanwhile, RIG-I expression was not affected by any type of FMDV NSP. To identify whether the Lb pro and 3C pro were responsible for the decline in exogenous MDA5 protein, plasmids encoding HA-tagged MDA5 and NSPs were cotransfected into HEK293T cells for 16 h (Figure 3a) . MDA5 was not detected in the cells transfected with Lb pro and 3C pro by western blot analysis. MDA5 protein expression was reduced in a dose-dependent manner by Lb pro and 3C pro (Figure 3b ). MDA5 expression was also reduced by 2B, 2C and 3A but not by 3B and 3D. The inhibition by 2B, 2C and 3A was dose-dependent for exogenous MDA5 (Figure 3c ). In particular, the Lb pro and 3C pro treatments decreased MDA5 mRNA levels ( Figure 3b ). These results indicate that both endogenous and exogenous MDA5 protein levels were prominently reduced by Lb pro and 3C pro .

To determine whether the proteasome, lysosome and caspase-dependent pathways play roles in the FMDV Lb pro and 3C pro -induced reduction of MDA5, the effects of proteasome inhibitor (MG132), lysosome inhibitor (CQ) and caspase inhibitor (Z-VAD-FMK) were evaluated (Figure 4 ). MDA5-expressing plasmids and either HA-Lb or HA-3C plasmids were co-transfected into HEK293 cells and the cells were cultured in the presence or absence of the inhibitors. The protein levels of MDA5, Lb pro and 3C pro were detected by western blotting at 12 hpt. MG132, CQ or Z-VAD-FMK did not affect the L-or 3C-induced reduction of MDA5 ( Figure 4 ). These results suggest that FMDV Lb pro and 3C pro -induced reduction of MDA5 occurs via a process independent of the proteasome, lysosome and caspase pathways. 

FMDV infection contributed to the decline in MDA5 protein expression. Therefore, to identify the FMDV NSP that affect the expression of endogenous MDA5 and RIG-I proteins, HEK293T cells were transfected with the plasmid encoding each viral NSP. After 16 h, endogenous MDA5 and RIG-I were detected by western blotting in the transfected cells ( Figure 2a ) and quantified using ImageJ software (Figure 2b ). The Lb pro and 3C pro of FMDV NSPs significantly reduced endogenous MDA5 expression compared to HA-vector-transfected cells. Meanwhile, RIG-I expression was not affected by any type of FMDV NSP. To identify whether the Lb pro and 3C pro were responsible for the decline in exogenous MDA5 protein, plasmids encoding HA-tagged MDA5 and NSPs were co-transfected into HEK293T cells for 16 h (Figure 3a) . MDA5 was not detected in the cells transfected with Lb pro and 3C pro by western blot analysis. MDA5 protein expression was reduced in a dose-dependent manner by Lb pro and 3C pro (Figure 3b ). MDA5 expression was also reduced by 2B, 2C and 3A but not by 3B and 3D. The inhibition by 2B, 2C and 3A was dose-dependent for exogenous MDA5 (Figure 3c ). In particular, the Lb pro and 3C pro treatments decreased MDA5 mRNA levels (Figure 3b ). These results indicate that both endogenous and exogenous MDA5 protein levels were prominently reduced by Lb pro and 3C pro . 

To determine whether the proteasome, lysosome and caspase-dependent pathways play roles in the FMDV Lb pro and 3C pro -induced reduction of MDA5, the effects of proteasome inhibitor (MG132), lysosome inhibitor (CQ) and caspase inhibitor (Z-VAD-FMK) were evaluated (Figure 4 ). MDA5-expressing plasmids and either HA-Lb or HA-3C plasmids were co-transfected into HEK293 cells and the cells were cultured in the presence or absence of the inhibitors. The protein levels of MDA5, Lb pro and 3C pro were detected by western blotting at 12 hpt. MG132, CQ or Z-VAD-FMK did not affect the L-or 3C-induced reduction of MDA5 (Figure 4) . These results suggest that FMDV Lb pro and 3C pro -induced reduction of MDA5 occurs via a process independent of the proteasome, lysosome and caspase pathways. 

The catalytic triad of H46, D84 and C163 in FMDV 3C pro have been determined as crucial sites that play essential roles in its enzyme activity [29, 30] . To determine whether the enzyme activity of 3C pro was involved in MDA5 reduction, we constructed three mutant plasmids that expressed the 3C pro mutant (HA-3C-H46Y, HA-3C-D84N and HA-3C-C163G), which eliminated the protease activity of 3C pro ( Figure 5 ). The pCMV-HA-vector plasmids, HA-3C plasmids or mutant plasmids were co-transfected with Flag-MDA5 plasmids into HEK293T cells and the expression levels of Flag-MDA5 were detected at 12 hpt (Figure 5a) . The wild-type HA-3C pro (WT) strongly suppressed MDA5 protein expression, compared to the pCMV-HA-Vector (Figure 5b ). In contrast, the H46Y, D84N and C163G mutants had no suppressive effect on MDA5 expression, demonstrating that the protease activity of 3C pro inhibited MDA5 protein expression (Figure 5b ). An immunoprecipitation was then performed to identify whether the interaction between 3C pro and MDA5 was associated with the 3C protease activity of ( Figure  6 ). The wild-type and the mutants of HA-3C pro interacted with FLAG-MDA5 and the wild-type of 3C pro inhibited MDA5 protein expression. Meanwhile, the D84N and C163G mutants did not reduce MDA5 protein expression. These results suggest that 3C pro directly interacts with MDA5 and induces the reduction of MDA5 by its protease activity. 

The catalytic triad of H46, D84 and C163 in FMDV 3C pro have been determined as crucial sites that play essential roles in its enzyme activity [29, 30] . To determine whether the enzyme activity of 3C pro was involved in MDA5 reduction, we constructed three mutant plasmids that expressed the 3C pro mutant (HA-3C-H46Y, HA-3C-D84N and HA-3C-C163G), which eliminated the protease activity of 3C pro ( Figure 5 ). The pCMV-HA-vector plasmids, HA-3C plasmids or mutant plasmids were co-transfected with Flag-MDA5 plasmids into HEK293T cells and the expression levels of Flag-MDA5 were detected at 12 hpt (Figure 5a) . The wild-type HA-3C pro (WT) strongly suppressed MDA5 protein expression, compared to the pCMV-HA-Vector (Figure 5b ). In contrast, the H46Y, D84N and C163G mutants had no suppressive effect on MDA5 expression, demonstrating that the protease activity of 3C pro inhibited MDA5 protein expression (Figure 5b ). An immunoprecipitation was then performed to identify whether the interaction between 3C pro and MDA5 was associated with the 3C protease activity of ( Figure 6 ). The wild-type and the mutants of HA-3C pro interacted with FLAG-MDA5 and the wild-type of 3C pro inhibited MDA5 protein expression. Meanwhile, the D84N and C163G mutants did not reduce MDA5 protein expression. These results suggest that 3C pro directly interacts with MDA5 and induces the reduction of MDA5 by its protease activity. HEK293T cells (1.5 × 10 6 cells) were co-transfected with HA-tagged, MDA5-expressing plasmid (2 µg) and various plasmids expressing 3C pro or individual mutants of 3C pro (2 µg) for 12 h. The abundance of MDA5 proteins was calculated by Image J software to quantify the band intensity, normalized to β-actin and respectively compared with the group of HA-Vec. or HA-3C pro . The result from one of the triplicate experiments are shown and presented as mean ± SEM (**, p < 0.01; ***, p < 0.001 and n.s., not significant).

Interaction between 3C pro and MDA5. HEK293T cells grown in 10-cm dishes were co-transfected with FLAG-MDA5 plasmid (8 µL) and the indicated plasmids (8 µL). The cells were lysed at 10 h and immunoprecipitated with mouse anti-FLAG antibody or mouse normal IgG antibody and then subjected to western blotting with anti-FLAG or anti-HA antibodies. Whole cell lysates (WCL) were analyzed by immunoblotting (IB) using anti-FLAG or anti-HA antibodies. All the data were repeated in three independent experiments and data from one of the triplicate experiments are shown.

Upon infection of the host, the virus will face an attack from the host immune response. In the battle with the host immune response, the virus has evolved a series of immune escape mechanisms to overcome the antiviral responses induced by the host immune system [31] . Although MDA5 is known to be responsible for picornavirus recognition [32, 33] , little is known about the interplay between MDA5 and FMDV. This study aimed to elucidate the mechanism of FMDV in evading the immune response of the host and to define specific NSPs that play a key role in the interplay with MDA5.

The evidence that FMDV interferes with interferon signaling cascade pathways is supported by two results in this study. Firstly, FMDV infection degraded MDA5 and MAVS. Secondly, IRF3 phosphorylation, which is a hallmark of IRF3 activation, in- 

Upon infection of the host, the virus will face an attack from the host immune response. In the battle with the host immune response, the virus has evolved a series of immune escape mechanisms to overcome the antiviral responses induced by the host immune system [31] . Although MDA5 is known to be responsible for picornavirus recognition [32, 33] , little is known about the interplay between MDA5 and FMDV. This study aimed to elucidate the mechanism of FMDV in evading the immune response of the host and to define specific NSPs that play a key role in the interplay with MDA5.

The evidence that FMDV interferes with interferon signaling cascade pathways is supported by two results in this study. Firstly, FMDV infection degraded MDA5 and MAVS. Secondly, IRF3 phosphorylation, which is a hallmark of IRF3 activation, in- 

Upon infection of the host, the virus will face an attack from the host immune response. In the battle with the host immune response, the virus has evolved a series of immune escape mechanisms to overcome the antiviral responses induced by the host immune system [31] . Although MDA5 is known to be responsible for picornavirus recognition [32, 33] , little is known about the interplay between MDA5 and FMDV. This study aimed to elucidate the mechanism of FMDV in evading the immune response of the host and to define specific NSPs that play a key role in the interplay with MDA5.

The evidence that FMDV interferes with interferon signaling cascade pathways is supported by two results in this study. Firstly, FMDV infection degraded MDA5 and MAVS.

Secondly, IRF3 phosphorylation, which is a hallmark of IRF3 activation, increased until 16 h post-FMDV infection and then decreased that is similar pattern to the mRNAs for interferon beta, MDA5 and RIG-I. In this study, the PK-15 cells were employed to see phenomena related with interferon signal transduction after FMDV infection because the PK-15 was susceptible to FMDV and have interferon signaling pathway [34] [35] [36] . Endogenous MDA5 in PK-15 decreased significantly on FMDV infection, resulting in no protein band by western blot analysis (Figure 1) . Previous studies have shown that MDA5 is inhibited by 3A or 2B. They, however, were carried out using Sendai virus [11] or a plasmid expressing MDA5 [37] . This study is the first to report that endogenous MDA5 expression is inhibited by FMDV infection. The result that MDA5 mRNA increased and MDA5 protein decreased until 16 hpi indicates that protein level rather than transcription was interfered with by FMDV infection. Once FMDV enter the host's cell, interferon signaling pathway initiates after MDA5 recognizes FMDV RNA, leading to increasing MDA5 mRNA. Meantime, NSP that is accumulated as FMDV infection progress in the cells interferes with interferon signaling pathway by degrading MDA5 protein translated from mRNA. We suggest that that is why the amount of MDA5 between mRNA and protein level was different. In previous studies, the similar result was reported by other research group that FMDV infection can induce RIG-I mRNA expression, while RIG-I protein was gradually downregulated as an infection progress [24] .

This study showed that MDA5 was more significantly reduced than RIG-I by FMDV infection in PK-15 cells. Previous study reported that FMDV was related with MDA5 but not by TLR3 in porcine epithelial cells using lentivirus-driven RNA interference. Although a previous study showed that MDA5 overexpression showed more significant antiviral activity than RIG-I overexpression against FMDV [24] , it dealt with the interplay between 2B and RIG-I only. Another study reported that L pro cleaved exogenous MDA5 but it did not show any evidence of degradation of endogenous MDA5 [38] .

Since it was found that FMDV evaded interferon induction pathways mainly via MDA5 inhibition in this study, we investigated which NSP could play a key role in the evasion mechanism. A total of seven NSPs (excluding 2A, which directs co-translational "ribosome skipping" at its own C-terminus [39] ), were evaluated for the interplay with MDA5 by overexpressing each protein. Of these, L pro and 3C pro significantly inhibited endogenous MDA5 expression in contrast to RIG-I, which was not reduced by any of them. The slight decrease in RIG-I caused by FMDV infection (Figure 1c ) might be attributed to the structural proteins such as VP1 or VP3 that were reported previously with regard to the interferon signaling pathway [28, 40] .

In addition, exogenous MDA5 expression was evaluated with the same NSPs using HA-tagged plasmids. Consistent with endogenous MDA5, L pro and 3C pro inhibited the most significant exogenous MDA5 expression. Other NSPs, such as 2B, 2C and 3A, also inhibited exogenous MDA5 compared to 3B and 3D. The contradictory results of 2B, 2C and 3A for endogenous and exogenous MDA5 might be due to the fact that endogenous MDA5 was derived from humans (HEK293T) and the exogenous MDA5 plasmid was constructed based on the porcine sequence. In this study, HEK293T cells were used due to its much higher transfection efficiency than that of PK-15 cells. Although this study focused on L pro and 3C pro , the interplay of 2B, 2C and 3A with MDA5 still has to be investigated.

In response to FMDV infection, the proteasome, lysosome and caspase-dependent pathways were not involved in the reduction of MDA5 by 3C pro . Similar results were previously reported with RIG-I, which was degraded during FMDV, EMCV, poliovirus and rhinovirus infection [24, 41] . Since we assumed that the protease activity of 3C would cleave MDA5, 3C mutants with point mutations in the catalytic triad of H46, D84 and C163 were constructed. As anticipated, all three inactive 3C pro mutants (H46Y, D84N and C163G) did not cleave the MDA5 at all. Therefore, we performed immunoprecipitation assay using HA-3C and Flag-MDA5 to observe the direct interaction between 3C pro and MDA5. The 3C pro interacted with MDA5 and that its protease activity degraded MDA5 protein expression. In this study, the degradation product of 3C pro -induced MDA5 was not detected. Previous studies reported that Lb pro degraded MDA5 without degradation products [35] and that 2B degraded RIG-I without intermediate products [24] . In further studies, we are planning to research regarding its degradation mechanism because the MDA5 was degraded independent of the proteasome, lysosome and caspase pathways. FMDV L pro is expressed in two forms, Lab and Lb. Lb is more abundant than Lab in infected cells [42] and is associated with translocation to the nucleus and cleavage of NF-kB and IRF3/7 proteins [43] [44] [45] . Since the direct interaction between the FMDV Lb pro and MDA5 has recently been reported [38] , we focused on the interplay between FMDV 3C pro and MDA5, which has not been elucidated at all.

FMDV 3C pro specializes in cleaving FMDV polyprotein into viral proteins with biological functions and induces cleavage of eIF4A and histone H3 to block the host translation system [46, 47] . To antagonize the interferon signaling pathway, the protease activity of FMDV 3C pro blocks STAT1/STAT2 nuclear translocation [36] . These biological functions of FMDV 3C pro appear to ruin the IFN system in the broad spectrum [48] . However, there is no report that FMDV 3C pro could exert a role to inhibit innate immunity by targeting MDA5.

FMDV infection initially triggers interferon production via IRF3 phosphorylation but eventually inhibits the expression of the MDA5 protein, thereby disrupting p-IRF3 activity. In this regard, we identified the antagonistic role of FMDV Lb pro and 3C pro on MDA5-mediated immune responses. In particular, the protease activity of 3C pro inhibited MDA5 expression independent of other protein degradation pathways. To our knowledge, this is the first report that FMDV 3C pro targets MDA5 in the mechanism to evade the immune response of the host. These results will improve our understanding of FMDV pathogenesis in the host and provide fundamental insights for the development of novel vaccines or therapeutic agents.

Comparative evaluation of non-structural protein-antibody detecting ELISAs for foot-and-mouth disease sero-surveillance under intensive vaccination

Foot-and-mouth disease

Molecular basis of pathogenesis of FMDV

Chemiluminescence immunoassay for the detection of antibodies against the 2C and 3ABC nonstructural proteins induced by infecting pigs with foot-and-mouth disease virus

A simple and rapid assay to evaluate purity of foot-and-mouth disease vaccine before animal experimentation

Enterovirus 2Apro targets MDA5 and MAVS in infected cells

Antiviral RNA recognition and assembly by RLR family innate immune sensors

Cytosolic sensing of viruses

Nucleic acid immunity

Molecular mechanisms of foot-and-mouth disease virus targeting the host antiviral response

Foot-and-mouth disease virus non-structural protein 3A inhibits the interferon-beta signaling pathway

RIG-I-like receptors: Their regulation and roles in RNA sensing

Immune signaling by RIG-I-like receptors

MDA-5 is cleaved in poliovirus-infected cells

Preference of RIG-I for short viral RNA molecules in infected cells revealed by next-generation sequencing

Flavivirus induces interferon-beta gene expression through a pathway involving RIG-I-dependent IRF-3 and PI3K-dependent NF-kappaB activation

Cytosolic 5 -triphosphate ended viral leader transcript of measles virus as activator of the RIG I-mediated interferon response

RNase L releases a small RNA from HCV RNA that refolds into a potent PAMP

MDA5 detects the double-stranded RNA replicative form in picornavirus-infected cells

MDA-5 recognition of a murine norovirus

Murine coronavirus mouse hepatitis virus is recognized by MDA5 and induces type I interferon in brain macrophages/microglia

Foot-and-mouth disease virus 3B protein interacts with pattern recognition receptor RIG-I to block RIG-I-mediated immune signaling and inhibit host antiviral response

Innate immune sensor LGP2 is cleaved by the leader protease of foot-and-mouth disease virus

Foot-and-mouth disease virus viroporin 2B antagonizes RIG-I-mediated antiviral effects by inhibition of its protein expression

Foot-and-mouth disease virus infection inhibits LGP2 protein expression to exaggerate inflammatory response and promote viral replication

Beitrag zur kollektiven Behandlung pharmakologischer Reihenversuche. Naunyn-Schmiedebergs Arch

The method of right and wrong cases (constant stimuli) without Gauss's formulae

The VP3 structural protein of foot-and-mouth disease virus inhibits the IFN-beta signaling pathway

Crystal structure of foot-and-mouth disease virus 3C protease. New insights into catalytic mechanism and cleavage specificity

Identification of the active-site residues of the 3C proteinase of foot-and-mouth disease virus

Molecular mechanisms of immune escape for foot-and-mouth disease virus

Differential roles of MDA5 and RIG-I helicases in the recognition of RNA viruses

Identification of the role of RIG-I, MDA-5 and TLR3 in sensing RNA viruses in porcine epithelial cells using lentivirus-driven RNA interference

CRISPR-Cas9 Mediated RNase L knockout regulates cellular function of PK-15 cells and increases PRV replication

In vitro coinfection and replication of classical swine fever virus and porcine circovirus type 2 in PK15 cells

3Cpro of foot-and-mouth disease virus antagonizes the interferon signaling pathway by blocking STAT1/STAT2 nuclear translocation

Foot-and-mouth disease virus non-structural protein 2B negatively regulates the RLR-mediated IFN-beta induction

MDA5 cleavage by the Leader protease of foot-and-mouth disease virus reveals its pleiotropic effect against the host antiviral response

Cleavage of foot-and-mouth disease virus polyprotein is mediated by residues located within a 19 amino acid sequence

Foot-and-mouth disease virus VP1 target the MAVS to inhibit type-I interferon signaling and VP1 E83K mutation results in virus attenuation

RIG-I is cleaved during picornavirus infection

Functional analysis of the two alternative translation initiation sites of foot-andmouth disease virus

Differential gene expression in bovine cells infected with wild type and leaderless foot-and-mouth disease virus

Degradation of nuclear factor kappa B during foot-and-mouth disease virus infection

Foot-and-mouth disease virus leader proteinase inhibits dsRNA-induced type I interferon transcription by decreasing interferon regulatory factor 3/7 in protein levels

Foot-and-mouth disease virus protease 3C induces specific proteolytic cleavage of host cell histone H3

Cleavage of translation initiation factor 4AI (eIF4AI) but not eIF4AII by foot-and-mouth disease virus 3C protease: Identification of the eIF4AI cleavage site

Type I interferon induced and antagonized by foot-and-mouth disease virus

The authors declare no conflict of interest.