Baby hamster kidney BHK-21 cells (BHK-21; ATCC, CCL-10), human embryonic kidney HEK293T cells (HEK293T; ATCC, CRL-11268), and porcine kidney (PK-15; ATCC, CCL-33) cells were grown in Dulbecco’s modified essential medium (DMEM; Invitrogen, Carlsbad, CA, USA) containing 10% fetal bovine serum (Gibco, Grand Island, NY, USA), 100 U/ml penicillin (Glenview, USA), and 10 μg/ml streptomycin sulfate (Glenview, USA) at 37°C in a humidified 5% CO₂ incubator. PSV and VSV were used in this study, and virus titers were determined using plaque-forming unit assays in BHK-21 cells. Sev was amplified in chicken eggs, and viral titer was determined using the hemagglutination (HA) test. IFN-β (Sino Biological, Shanghai, China) was used for IFN stimulation. The proteasome inhibitor MG132 (S1748-1mg), the caspase inhibitor zVAD-FMK (C1202), dimethyl sulfoxide (DMSO, ST038), and NH4Cl were obtained from Beyotime Biotechnology.

The commercial antibodies used in this study including anti-FLAG monoclonal antibody and anti-HA monoclonal antibody were obtained from Medical and Biological Laboratories (Nagoya, Japan); anti-IRF3 polyclonal antibody (11312-1-AP) and anti-TANK polyclonal antibody (27065-1-AP) were obtained from Proteintech (Wuhan, China); anti-p-IRF3-386 (AP0995) monoclonal antibody, anti-TBK1 monoclonal antibody (A14641), anti-GAPDH (AC002) monoclonal antibodies, and anti-β-tubulin monoclonal antibodies (AC012) were obtained from ABclonal (Woburn, MA, USA); anti-MAVS monoclonal antibody (SC-166583) was obtained from Santa Cruz Biotechnology (Dallas, TX, USA); anti-MDA5 polyclonal antibody (M00263) was obtained from Boster (Pleasanton, CA, USA); anti-PARP1 polyclonal antibody (GB111500) was obtained from Servicebio (Wuhan, China); and anti-PSV VP1 polyclonal antibody was prepared in our laboratory.

The full-length CDs of individual PSV viral proteins were amplified from PSV (MT080999.1) cDNA. VP1, VP2, VP3, 3C^pro, and 3D were cloned into vector pCAGGS-HA. L, 2A, 2C, and 3A were cloned into pEBG-GST. VP4 was cloned into vector pEGFP-C1, and 2B was cloned into vector pCDNA3.1-Flag. Flag-TANK was a kind gift from Changjiang Weng (Harbin Veterinary Research Institute of Chinese Academy of Agricultural Sciences). Flag-MAVS and Flag-TRIF were kindly provided by Mei lin Jin (Hua Zhong Agricultural University). Mutagenesis of PSV 3C^pro, MAVS, and TANK constructs was performed using overlap PCR (C214-01/02, Vazyme, Nanjing, China). All constructs were identified through DNA sequencing. Primer pairs used in this study are shown in Supporting Table 1. Mutants of human MAVS (D429A, D490A, and D498A) and PSV 3C^pro (H40A, C146A, and H40A/C146A) were constructed by overlap extension using Pfu DNA polymerase (Stratagene, La Jolla, CA). The cDNAs encoding deletion mutants of MAVS, including N429 (1–429 amino acids) and C430 (430–540 amino acids), were cloned into the pFlag vector.

Total cellular proteins were prepared using lysis buffer (1.19% HEPES, 0.88% NaCl, 0.04% EDTA, 1% NP-40, and a protease inhibitor) with occasional vortexing. Lysates were then collected by centrifugation at 12,000 rpm for 10 min at 4°C, and protein concentrations were determined by the bicinchoninic acid protein assay kit (Thermo Scientific, Waltham, MA, USA). The equal amounts of protein for each sample were loaded and separated by 8% to 12% SDS-PAGE and transferred onto polyvinylidene fluoride (PVDF) membranes (Roche, Welwyn Garden City, UK). Membranes were blocked with 5% skim milk in PBST with 5% Tween 20 (DGBio, Beijing, China) for 2 h at room temperature. The membranes were then washed two times with PBST and incubated with primary antibodies at room temperature for 2 h. Afterward, the membranes were washed five times with PBST and incubated with anti-rabbit or anti-mouse IgG antibodies conjugated to horseradish peroxidase (HRP) at room temperature for 1 h. The membranes were developed using enhanced chemiluminescence detection reagents (Thermo Scientific, USA).

HEK293T cells were seeded in 24-well plates, and the monolayer cells were co-transfected with 100 ng/well of pGL3-IFNβ-Luc plasmid, 5 ng/well of pRL-TK plasmid (as an internal control), and the indicated expression plasmids or an empty control plasmid. The cells were infected PSV or Sev 20 h after the initial co-transfection. The cells were then collected and lysed, and firefly luciferase and Renilla luciferase activities were determined using the dual-luciferase reporter assay system (Beyotime, Shanghai, China) according to the manufacturer’s protocol. Three independent experiments were performed in duplicate. Data are presented as means ± standard deviations(SD).

Cells were fixed in 4% paraformaldehyde at 4°C for 10 min and permeabilized with 0.1% Triton X-100 at room temperature for 10 min; the cells were incubated with 2% BSA in PBS for 1 h at room temperature. The cells were then washed three times with PBS and incubated with anti-IRF3 for 2 h. After four washes with PBS, the cells were incubated with Alexa Fluor 488-conjugated goat anti-rabbit IgG antibodies at room temperature for 1 h. The cells were washed again four times with PBS and then incubated with DAPI (Sigma, St. Louis, MO, USA) for 10 min, followed by five washes. Fluorescent images were examined using inverted fluorescence.

Total RNA was extracted from the targeted cells using TRIzol reagent (Invitrogen) in accordance with the manufacturer’s instructions. 1 µg total RNA was reverse transcribed to cDNA using HiScript III 1st Strand cDNA Synthesis Kit (Vazyme). The relative mRNA level of targeted genes was measured with qPCR using SYBR Green real-time PCR master mix (Applied Biological Materials Inc) with specific prime. All reactions were performed in triplicate. The mRNA level of housekeeping gene GAPDH was used as an internal control. The relative gene fold was determined with the comparative cycle threshold (2^-ΔΔCT) method.

Statistical analysis was conducted using GraphPad Prism software, version 5. All results were determined by at least three times independent experiments. The various treatments were compared using an unpaired, two-tailed Student t-test with an assumption of unequal variance. P value < 0.05 was considered statistically significant.

Previous studies have reported that IFN-β can inhibit several picornavirus replication, such as FMDV, EMCV, and SVV (Chinsangaram et al., 2001; Wang et al., 2011; Qian et al., 2017). To evaluate whether IFN-β could inhibit PSV proliferation, HEK293T cells were pretreated with different doses of IFN-β for 12 h and then infected with PSV (0.01 MOI) for 24 h. We found that IFN-β inhibited PSV-induced cytopathic effects (CPE) and PSV proliferation in a dose-dependent manner ( Figure 1A ). Taken together, our results demonstrate that PSV is an interferon-sensitive virus. The impact of PSV infection on the host innate immune response is unknown. The effects of PSV infection on IFN-β promoter activity were analyzed by dual-luciferase assays. We found that IFN-β promoter activity was hardly activated in PSV-infected cells ( Figure 1B ). HEK293T cells were infected with PSV (0.1 MOI) or Sev (2⁵ HA) for the indicated time, and the expression level of IFN-β mRNA was detected through quantitative real-time RT-PCR assay. Interestingly, the mRNA level of IFN-β hardly increased in PSV-infected cells as the infected time increased ( Figure 1C ). Similarly, we observed that the IFN-β promoter activity and the mRNA level of IFN-β hardly increased in PSV-infected PK15 cells ( Figures 1D, E ). We also measured the phosphorylation level of IRF3 via Western blotting. HEK293T cells were infected with PSV (0.1 MOI) or Sev (2⁵ HA), and then the cells were collected at indicated times. The phosphorylation of IRF3 could not be detected ( Figure 1F ). In addition, we also analyzed whether PSV could influence Sev-induced IFN-β production. The quantitative real-time PCR results showed that PSV infection inhibited the expression level of Sev-induced IFN-β mRNA ( Figure 1G ). Collectively, these results indicate that PSV infection cannot induce type I interferon production.

Based on above findings, we evaluated which viral proteins would likely be involved in the inhibition of type I production. HEK293T cells were transfected with plasmids expressing L, VP1, VP2, VP3, VP4, 2A, 2B, 2C, 3A, 3C^pro, or 3D in combination with IFN-β-Luc and pRL-TK. The results indicated that viral proteins L, 3C^pro, and 3D could significantly inhibit IFN-β promoter activity ( Figure 2A ). Previous studies revealed that picornaviruses evolved various strategies to inhibit IFN-β production by certain viral proteins, such as L^pro, VP2, VP3, 2A^pro, 2B, 2C, 3A, 3B, and 3C^pro (Dang et al., 2011; Mukherjee et al., 2011; Bei et al., 2013; Dang et al., 2014; Zixiang et al., 2016; Dan et al., 2016; Dan et al., 2016; Wen et al., 2019; Xiangle et al., 2020). Our data showed that overexpression of L, 3C^pro, and 3D reduced Sev-induced IFN-β production. Moreover, the results of quantitative real-time PCR and Western blotting demonstrated that 3C^pro exhibited an extreme inhibitory effect on IFN-β production ( Figures 2B, C ). This implies that 3C^pro is an antagonist protein to inhibit IFN-β production. HEK293T cells were transfected with an increasing dose of HA-3C for 24 h. As shown in Figure 2D , PSV 3C^pro could significantly inhibit Sev-induced IFN-β promoter activity. In addition, PSV 3C^pro decreased Sev-induced endogenous transcription of IFN-β in a dose-dependent manner ( Figure 2E ). Furthermore, we also found that the phosphorylation of IRF3 gradually reduced when the protein level of PSV 3C^pro was increased ( Figure 2F ). Taken together, these data confirm that PSV 3C^pro inhibits type I IFN production in a dose-dependent manner.

Similar to other picornaviruses 3C^pro, PSV 3C^pro contains the conserved catalytic box with histidine (His) and cysteine (Cys) residues ( Figure 3A ). Therefore, we constructed a series of PSV 3C^pro mutation expression plasmids, including single-site mutations H40A and C146A and double-site mutation H40A-C146A (3CDM) ( Figure 3B ). The result showed that wild-type 3C^pro (WT) suppressed Sev-induced phosphorylation of IRF3. In contrast, all 3C^pro mutants [3C-H40A, 3C-C146A, and H40A-C146A (3C-DM)] lost the ability to inhibit the phosphorylation of IRF3 ( Figure 3C ). This result suggested that 3C^pro inhibited Sev-induced type I interferon production via its protease activity.

Previous studies demonstrated that several picornavirus 3C proteases cleave or degrade innate immune adaptors to evade host innate immune responses. SVV 3C^pro cleaves MAVS, TRIF, and TANK to weaken type I IFN production (Qian et al., 2017). CVB3 and EV71 cleave RIG-I via viral 3C^pro (Feng et al., 2014). SVV 3C^pro can target RIG-I, IRF3, and IRF7 for degradation to downregulate IFN-β production (Qiao et al., 2018; Wen et al., 2019). To investigate which innate immune adaptors are cleaved or degraded by PSV 3C^pro, HEK293T cells were transfected with plasmids expressing FLAG-MAVS, -RIG-I, -MDA5, -IRF3, -TBK1, or -IKKξ in combination with an empty vector or plasmids expressing HA-3C. As shown in Figure 4A , overexpression of PSV 3C^pro could cleave MAVS, and we observed that TBK1 and MDA5 were degraded by PSV 3C^pro. To confirm whether PSV 3C^pro degraded TBK1 and MDA5 in a dose-dependent manner, HEK293T cells were co-transfected with increasing amounts of PSV 3C^pro and MDA5 or TBK1 for 24 h. Western blotting results showed that the abundance of MDA5 and TBK1 was gradually reduced when the protein level of PSV 3C^pro was increased ( Figure 4B ). Meanwhile, HEK293T cells were co-transfected with FLAG-MAVS and an increasing dose of PSV 3C^pro. We observed that the cleavage products of MAVS were increased by PSV 3C^pro in a dose-dependent manner ( Figure 4C ). In addition, we also assessed the effects of PSV 3C^pro on endogenous MDA5 and TBK1 protein expression and endogenous MAVS cleavage. HEK293T cells were transfected with increasing amounts of HA-3C for 24 h. The result showed that endogenous MDA5 and TBK1 were degraded by 3C^pro and 3C^pro cleaved endogenous MAVS ( Figure 4D ). We also analyzed the expression of MDA5 and TBK1 and the cleavage of MAVS during PSV infection. HEK293T cells were infected with 0.5 MOI PSV for the indicated time. We found that the expression of MDA5 and TBK1 was gradually reduced during PSV infection, and PSV infection induced the cleavage of MAVS ( Figure 4E ). Collectively, our data indicated that PSV 3C^pro targets MAVS, MDA5, and TBK1 to impair type I IFN production.

A previous study demonstrated that SVV 3C^pro degraded RIG-I through the caspase pathway (Wen et al., 2019). To investigate whether PSV 3C^pro-mediated MDA5 and TBK1 is dependent on apoptosis-related caspase activity, proteasome, or lysosome signaling, HEK293T cells expressing MDA5 or TBK1 and PSV 3C^pro were treated with DMSO, proteasome inhibitor MG132 (10 μM), lysosomal inhibitor NH4Cl (10 mM), or pan-caspase inhibitor Z-VAD-FMK (50 μM). As shown in Figures 5A, B , we found that none of these inhibitors rescued MDA5 or TBK1 expression. We next examined whether 3C^pro-induced reduction of MDA5 and TBK1 depended on its protease activity; HEK293T cells were co-transfected with FLAG-MDA5 or -TBK1 and all 3C^pro mutants (H40A, C146A, and H40A-C146A (3C-DM)) for 24 h. Western blot results showed that 3C^pro degraded MDA5 and TBK1 through its protease activity ( Figures 5C, D ). Next, we tested whether PSV 3C^pro downregulated the level of TBK1 and MDA5 mRNA. The result of qRT-PCR suggested that PSV 3C^pro could significantly reduce the mRNA level of TBK1 and MDA5 ( Figures 5E, F ).

Previous reports have clearly shown that MAVS could be cleaved by virus proteases (NS3/4A and 3ABC) or caspases (Yang et al., 2007; Yu et al., 2010; Anggakusuma et al., 2016). To determine whether cellular proteases participate in 3C^pro-mediated cleavage of MAVS, the effect of proteasome inhibitor MG132, lysosomal inhibitor NH4Cl, and pan-caspase Z-VAD-FMK on the cleavage of MAVS were detected by Western blot. Flag-MAVS and PSV 3C^pro were co-transfected into HEK293T cells for 12 h, and then the cells were treated with DMSO, MG132 (10 μM), NH4Cl (10 mM), and Z-VAD-FMK (50 μM) for another 12 h. The results showed that PSV 3C^pro lost the ability to mediate the cleavage of MAVS in the presence of Z-VAD-FMK, indicating that caspases were involved in 3C^pro-mediated MAVS cleavage ( Figure 6A ). To investigate whether 3C protease activity also participated in 3C^pro-mediated cleavage of MAVS, FLAG-MAVS and 3C mutants [3C-H40A, 3C-C146A, and 3C-H40A-C146A (3C-DM)] were co-transfected into HEK293T cells for 24 h. Compared to the wild-type PSV 3C^pro, all 3C^pro mutants could not cleave MAVS ( Figure 6B ). To confirm whether PSV 3C^pro could induce cell apoptosis, HEK293T cells were transfected with an increased amount of HA-3C. As shown in Figure 6C , PSV 3C^pro cleaved PARP1 in a dose-dependent manner. This result revealed that 3C^pro could induce apoptosis. Interestingly, we found that PSV could also induce cell apoptosis ( Figure 6D ). To determine whether 3C^pro induced apoptosis through its protease activity, HEK293T cells were transfected with the HA-3C or 3C^pro mutant. PSV 3C^pro lost the ability to cleave PARP1 in the presence of Z-VAD-FMK, and PSV 3C^pro without protease activity could not cleave PARP1 ( Figure 6E ). These data showed that the protease activity was essential for PSV 3C^pro-mediated apoptosis.

To investigate the 3C^pro-mediated MAVS cleavage site, we constructed three MAVS mutants in which aspartic (D) was replaced with alanine (A) ( Figure 7A ). HEK293T cells were transfected with MAVS or MAVS mutants and HA-3C. As shown in Figure 7B , we observed that the cleavage fragment entirely disappeared in MAVS-D429A. Taken together, these results demonstrated that PSV 3C^pro cleaved MAVS at D429. MAVS is the key adaptor molecule in RLR signaling pathways. To investigate whether PSV 3C^pro could disrupt MAVS-mediated type I interferon production, a dual-luciferase activity assay was performed to detect the IFN-β promoter activity. The result showed that MAVS-mediated IFN-β promoter activity was significantly inhibited by 3C-WT, but not 3C-DM ( Figure 7C ). In addition, MAVS-D429A-mediated IFN-β promoter activity was not affected by 3C-WT or 3C-DM, suggesting that MAVS-D429A could resist 3C^pro-mediated cleavage and retain the ability of IFN-β promoter activity ( Figure 7C ). To confirm the results, HEK293T cells were transfected with MAVS, MAVS-D429A, and 3C-WT or 3C-DM for 24 h, and the cells were collected for qRT-PCR analysis. As expected, MAVS-D429A could induce IFN-β production and 3C-WT did not inhibit MAVS-D429A-mediated IFN-β production ( Figure 7D ). We next assessed whether MAVS cleavage fragments maintain the activity of MAVS-mediated IFN-β production; the truncated plasmids expressing the N-terminal (residues 1 to 429; N429) and C-terminal (residues 430 to 540; C430) fragments of MAVS were constructed, and the effect was evaluated by dual-luciferase activity assay and qRT-PCR. The cleavage products cannot activate IFN-β promotor activity and increase the IFN-β mRNA level, indicating that N429 and C430 lost the ability to induce IFN-β production ( Figures 7E, F ). Taken together, PSV 3C^pro-mediated MAVS cleavage products lose their ability to induce type I interferon production.