Foot-and-Mouth Disease Virus Induces Porcine Gasdermin E-Mediated Pyroptosis through the Protease Activity of 3C^pro

ABSTRACT

Foot-and-mouth disease (FMD) is an acute, highly contagious disease of cloven-hoofed animals caused by FMD virus (FMDV). Currently, the molecular pathogenesis of FMDV infection remains poorly understood. Here, we demonstrated that FMDV infection induced gasdermin E (GSDME)-mediated pyroptosis independent of caspase-3 activity. Further studies showed that FMDV 3C^pro cleaved porcine GSDME (pGSDME) at the Q271-G272 junction adjacent to the cleavage site (D268-A269) of porcine caspase-3 (pCASP3). The inhibition of enzyme activity of 3C^pro failed to cleave pGSDME and induce pyroptosis. Furthermore, overexpression of pCASP3 or 3C^pro-mediated cleavage fragment pGSDME-NT was sufficient to induce pyroptosis. Moreover, the knockdown of GSDME attenuated the pyroptosis caused by FMDV infection. Our study reveals a novel mechanism of pyroptosis induced by FMDV infection and might provide new insights into the pathogenesis of FMDV and the design of antiviral drugs.

IMPORTANCE Although FMDV is an important virulent infectious disease virus, few reports have addressed its relationship with pyroptosis or pyroptosis factors, and most studies focus on the immune escape mechanism of FMDV. GSDME (DFNA5) was initially identified as being associated with deafness disorders. Accumulating evidence indicates that GSDME is a key executioner for pyroptosis. Here, we first demonstrate that pGSDME is a novel cleavage substrate of FMDV 3C^pro and can induce pyroptosis. Thus, this study reveals a previously unrecognized novel mechanism of pyroptosis induced by FMDV infection and might provide new insights into the design of anti-FMDV therapies and the mechanisms of pyroptosis induced by other picornavirus infections.

INTRODUCTION

Foot-and-mouth disease (FMD) is an acute, febrile, highly contagious infectious disease caused by foot-and-mouth disease virus (FMDV), which mainly affects artiodactyls (pigs, cattle, and sheep), occasionally in humans and other animals (1). Animals suffering from FMD develop blisters and ulcers on the oral mucosa, hooves, and breast skin, which seriously affects the productivity and quality of livestock and causes huge economic losses to the breeding industry (2). FMD is mainly divided into seven serotypes, including A, O, and C, South African Territories type 1 (SAT1), SAT2, SAT3, and Asian 1, of which O-type FMD is widely prevalent in the world (3). FMDV belongs to the genus Aphthovirus in the Picornaviridae family. The genome consists of a single positive-strand RNA of about 8.5 kb in length, encoding a large open reading frame (ORF), which is translated into a polyprotein precursor and then cleaved by viral proteases and processed into four structural proteins (VP4, VP2, VP3, and VP1) and eight nonstructural proteins (L^pro, 2A, 2B, 2C, 3A, 3B, 3C^pro, and 3D) (4).

FMDV 3C protease (3C^pro) belongs to the chymotrypsin-like cysteine protease family and is crucial in processing viral polyprotein precursors and viral replication. In addition, FMDV 3C^pro also plays an important role in the regulation of host innate immunity (5). Accumulating evidence has shown that 3C^pro inhibited multiple innate immune response signaling pathways, thereby escaping the host antiviral response. FMDV 3C^pro inhibited host protein synthesis by cleaving the translation initiation factors eIF4AI and eIF4G (6). Furthermore, 3C^pro promoted viral translation and replication by antagonizing the internal ribosomal entry site transacting factors (ITAFs) DDX23 and hnRNP K (7, 8). 3C^pro also escaped the innate immune response by degrading various RIG-I, MDA5, and NF-κB signaling molecules. Moreover, the proteolytic activity of 3C^pro is particularly important for inducing the degradation of these molecules, showing its strong potential as an antiviral target (6, 9, 10).

Programmed cell death is an important innate immune mechanism during the infection of host cells by pathogenic microorganisms, while excessive responses are also the main cause of its pathogenicity. Pyroptosis is a novel inflammation-associated programmed cell death characterized by the formation of membrane pores in the plasma membrane, cell swelling, and rupture of the plasma membrane, followed by the release of its cytoplasmic contents (including inflammatory cytokines) to the extracellular environment (11). The canonical pyroptosis pathway is mainly mediated by caspase-1. The inflammasome is assembled in response to danger signals, and caspase-1 is activated. Activated caspase-1 cleaves the pore-forming protein gasdermin D (GSDMD), eventually releasing GSDMD N-terminal fragments to bind membrane lipids and form membrane pores, resulting in cell swelling, dissolution, and death. Activated caspase-1 also cleaves pro-IL-1β and pro-IL-18 into mature interleukin-1β (IL-1β) and IL-18, which are then secreted to the extracellular environment and promote inflammatory responses (12–14).

Furthermore, it has been observed that both human caspase-4/5 and mouse caspase-11 exhibit binding affinity toward lipopolysaccharide (LPS), which subsequently triggers their activation and cleavage of GSDMD, ultimately resulting in the formation of membrane pores and induction of pyroptosis via the noncanonical pyroptosis pathway (15–17). Recent studies have shown that other members of the gasdermin family can also act as effector proteins to trigger pyroptosis (18–24). Caspase-8 activates GSDMC and converts apoptosis into pyroptosis (23). Similarly, caspase-3 activation drives pyroptosis by cleavage of GSDME and provides new insights into cancer chemotherapy (18). Furthermore, granzymes A and B derived from cytotoxic lymphocytes cleave GSDMB (19) and GSDME (20) to induce pyroptosis, enhancing antitumor immunity. Moreover, pyroptosis mediated by GSDME plays a crucial role in the pathogenesis of viral infection (21, 22).

Pyroptosis is a double-edged sword. On the one hand, it can effectively defend against pathogen infection. On the other hand, excessive inflammation will damage the immune system, which is conducive to the replication of viruses (25). Previous studies have shown that picornavirus infection can induce pyroptosis to facilitate viral replication and disrupt the host immune response (26, 27). However, the mechanism of pyroptosis induced by FMDV infection remains unclear.

This study demonstrated that FMDV infection induces GSDME-mediated pyroptosis independent of caspase-3 activity. In addition, FMDV 3C^pro cleaves pGSDME close to the cleavage site of pCASP3, and the protease activity of 3C^pro is required for pGSDME cleavage. Furthermore, we found that overexpression of pCASP3 or the 3C^pro-mediated cleavage fragment pGSDME-NT is sufficient to induce pyroptosis. Our study reveals a novel mechanism of pyroptosis induced by FMDV infection and might provide new insights into the pathogenesis of FMDV and the design of antiviral drugs.

RESULTS

FMDV infection induces GSDME-mediated pyroptosis in SK6 cells.

Programmed cell death caused by FMDV infection is an important defense mechanism against viral infection (11, 28, 29). Previous studies have reported that FMDV infection induces apoptosis in host cells (9). To investigate whether FMDV infection causes pyroptosis, the novel inflammation-associated programmed cell death, SK6 cells were infected with FMDV O/HN/CHA/93 strain at a multiplicity of infection (MOI) of 1 at the indicated time (Fig. 1). Morphological observation showed that SK6 cells infected with FMDV exhibited typical pyroptosis features. The cell morphology was similar to that of SK6 cells infected with the Seneca Valley virus (SVV) (26), characterized by swelling and rounding of the cells. The shape of fried eggs in the middle of the bulge (Fig. 1A). In addition, cell viability decreased significantly as the virus replicated (Fig. 1B). Lactate dehydrogenase (LDH), as a stable cytoplasmic enzyme, is released during cell lysis. To this end, we further measured the LDH release after FMDV infection in different cells. As shown in Fig. 1C, LDH release was dramatically increased at 4 and 8 h postinfection (hpi) in PK-15 and SK6 cells (porcine cells) and was significantly higher than that in BHK-21 (mouse-derived cell line) and HEK-293T cells (human cell lines), while there was no significant difference between PK-15 and SK6 cells. Propidium iodide (PI), a nucleic acid red fluorescent dye, can only stain necrotic cells with loss of cell membrane integrity and binds to nucleic acids to emit bright red fluorescence. To further characterize pyroptosis caused by FMDV infection in SK6 cells, the FMDV- or mock-infected cells were stained with PI. The results showed that a large amount of red fluorescence was observed in FMDV-infected cells, while no fluorescent signal was observed in mock-infected cells (Fig. 1D). Since pyroptosis is a kind of inflammatory cell death (11), we further detected the expression of inflammasome-associated genes. The results showed that the mRNA levels of IL-1β and IL-18 (Fig. 1E and F) and NLRP2 and NLRP3 (see Fig. S1 in the supplemental material) increased with virus infection, while the mRNA levels of ASC, GSDMD, and NLRP1 decreased significantly (see Fig. S1). Interestingly, the mRNA levels of caspase-1 slightly increased in the early stage of the virus infection and decreased in the late stage, although there was no significant difference throughout the infection (see Fig. S1).

FIG 1.

FMDV infection induces pyroptosis in SK6 cells. SK6 cells were infected with FMDV O/HN/CHA/93 strain at an MOI of 1 at the indicated times. (A) Morphological examination of SK6 cells with FMDV or mock infection. SVV infection was used as a positive control. The cell morphology was observed using an Olympus IX71 microscope (white arrows, pyroptotic cells). (B) FMDV inhibited SK6 cell viability. The cell viability of SK6 was determined by using a Cell Counting Kit-8. (C) SK6, PK-15, BHK-21, and HEK-293T cells were infected with FMDV O/HN/CHA/93 strain at an MOI of 1 at the indicated times. The release of LDH from FMDV-infected cells was measured with an LDH cytotoxicity assay kit. (D) PI staining of SK6 cells with FMDV or mock infection. (E and F) The relative expression levels of IL-1β and IL-18 were analyzed by quantitative RT-PCR (qRT-PCR) in SK6 cells, and the housekeeping gene GAPDH was used as the control. (G) FMDV infection induces GSDME-mediated pyroptosis in SK6 cells. Proteolytic cleavage of GSDMD or GSDME in SK6 cells with FMDV infection was determined by Western blotting. The abundance of caspase-3 (CASP3), IL-1β, MLKL, phospho-MLKL, FMDV VP1 protein, and α-tubulin as an internal control was also determined by Western blotting. Two lanes represent two biological replicates. (H and I) The release of IL-1β and IL-18 from FMDV-infected cells was measured with porcine IL-1β and IL-18 using an enzyme-linked immunosorbent assay (ELISA) kit. (J and K) SK6 cells were transfected with plasmids encoding pGSDME for 24 h and then infected with FMDV O/HN/CHA/93 strain at an MOI of 1 at the indicated times. (L and M) SK6 cells were infected with FMDV at various MOIs as indicated for 24 h. Proteolytic cleavage of GSDME in SK6 cells with FMDV infection was determined by Western blotting. The release of LDH from FMDV-infected cells was measured with LDH Cytotoxicity assay kit. The data in panels B, C, E, F, H, I, K, and M are presented as means ± the SD. Experiments were performed independently with at least two biological replicates. Student t test or one/two-way ANOVA was used for analysis. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Previous studies have shown that picornavirus infection-induced pyroptosis is executed through the caspase-1-GSDMD axis (11, 12, 28–31). However, our results showed that caspase-1 was not activated after FMDV infection. Therefore, FMDV infection might cause pyroptosis through other pathways. Western blotting revealed that consistent with mRNA levels, caspase-1 was not activated after FMDV infection, and GSDMD was degraded rather than cleaved (Fig. 1G). Excitingly, GSDME, another pyroptotic executive protein in the gasdermin family (18), was activated, and an obvious cleavage product appeared. Also, activation of caspase-3 and maturation of IL-1β was observed (Fig. 1G), suggesting that FMDV infection-induced pyroptosis might be executed through the caspase-3-GSDME axis. Furthermore, we examined whether MLKL (mixed-lineage kinase domain-like) protein, the executor in the necroptosis pathway, was phosphorylated (29, 32–35). The results showed that MLKL was phosphorylated after FMDV infection, indicating that FMDV infection causes multiple forms of cell death (pyroptosis, necroptosis, and apoptosis). In addition, the levels of proinflammatory cytokines IL-1β and IL-18 in the supernatant of SK6 cells infected with FMDV were significantly higher than those in the control group (Fig. 1H and I).

To verify the cleavage of pGSDME, SK6 cells were transfected with Flag-pGSDME, followed by FMDV infection at the indicated time. Western blotting found an obvious cleavage product of transfected pGSDME in FMDV-infected SK6 cells at 12 hpi (Fig. 1J). The release of LDH from FMDV-infected cells was dramatically increased compared to the mock-infected cells (Fig. 1K). In addition, similar results were observed when pGSDME-transfected cells were infected with different doses of FMDV (Fig. 1L and M). These data indicated that FMDV infection induces GSDME-mediated pyroptosis in SK6 cells.

FMDV infection induces pyroptosis independent of caspase-3 activity.

As the upstream initiator or executor of the programmed cell death pathway, caspases play a crucial role in pyroptosis (11). Since our results showed that FMDV infection might induce pyroptosis through the caspase-3/GSDME pathway, we investigated whether the caspase-3 activity is required for pyroptosis induced by FMDV infection. SK6 cells were infected with FMDV O/HN/CHA/93 strain at an MOI of 1 for 12 h in the presence or absence of the caspase-3-specific inhibitor Ac-DEVD-CHO (20 μM). As shown in Fig. 2A, the cells treated with Ac-DEVD-CHO still exhibited typical pyroptosis features. LDH release from Ac-DEVD-CHO-treated cells was not significantly different from that of DMSO-treated cells (Fig. 2B). PI staining also showed amounts of red fluorescence in Ac-DEVD-CHO-treated cells (Fig. 2C). Western blotting revealed that FMDV infection still caused pyroptosis in the presence of Ac-DEVD-CHO, which was manifested as the cleavage of pGSDME, activation of caspase-3, and maturation of IL-1β (Fig. 2D). As expected, there were no significant differences of the IL-1β and IL-18 levels in the FMDV-infected SK6 cell supernatant between the AC-DEVD-CHO- and DMSO-treated groups (Fig. 2E and F).

FIG 2.

FMDV infection induces pyroptosis in SK6 cells independent of caspase-3 activity. SK6 cells were infected with FMDV O/HN/CHA/93 strain at an MOI of 1 for 12 h in the presence or absence of Ac-DEVD-CHO (20 μM). (A) Morphological examination of SK6 cells with FMDV or mock infection. The cell morphology was observed using an Olympus IX71 microscope (white arrows, pyroptotic cells). (B) The release of LDH from FMDV-infected cells was measured with an LDH cytotoxicity assay kit. (C) PI staining of SK6 cells with FMDV or mock infection. (D) Proteolytic cleavage of GSDME in SK6 cells with FMDV infection was determined by Western blotting. The abundance of caspase-3 (CASP3), IL-1β, FMDV VP1 protein, and α-tubulin as an internal control was also determined by Western blotting. (E and F) The release of IL-1β and IL-18 from FMDV-infected cells was measured with porcine IL-1β and IL-18 using an ELISA kit. The data in panels B, E, and F are presented as means ± the SD. Experiments were performed independently with at least three biological replicates. Two-way ANOVA was used for analysis. ns, not significant.

To further exclude the role of caspase-3 in pyroptosis induced by FMDV infection, caspase-3 knockdown was performed using small interfering RNAs (siRNAs). Similar to the results of AC-DEVD-CHO treatment, caspase-3 knockdown had no significant effect on pyroptosis caused by FMDV infection (see Fig. S2). These results indicated that FMDV infection induces pyroptosis independent of caspase-3 activity.

FMDV 3C^pro induces pyroptosis in SK6 cells.

Since FMDV infection induces pyroptosis in SK6 cells independent of caspase-3 activity, we speculated that viral proteins might be involved. SK6 cells were cotransfected with plasmids encoding pGSDME and viral proteins to identify the key proteins for 24 h. The results showed that only 3C^pro had a cleavage effect on pGSDME (Fig. 3A), suggesting that 3C^pro might be involved in the pyroptosis induced by FMDV infection. Furthermore, typical pyroptotic morphology was found in cells overexpressing 3C^pro (Fig. 3B). Moreover, significantly increased LDH release was observed in 3C^pro-transfected cells (Fig. 3C). To verify the cleavage effect of 3C^pro on pGSDME, SK6 cells were cotransfected with plasmids encoding 3C^pro and pGSDME for 24 h. As indicated in Fig. 3D, a distinct cleavage product was observed in the cells cotransfected with 3C^pro and pGSDME. These results indicated that FMDV 3C^pro induces pyroptosis in SK6 cells.

FIG 3.

FMDV 3C^pro induces pyroptosis in SK6 cells. (A) Proteolytic cleavage of pGSDME in SK6 cells by viral proteins. SK6 cells were cotransfected with plasmids encoding pGSDME and viral proteins for 24 h. Proteolytic cleavage of pGSDME in SK6 cells by viral proteins was determined by Western blotting. The red asterisk indicates the pGSDME N-terminal cleavage product. (B) SK6 cells were transfected with recombinant plasmids encoding 3C^pro. The cell morphology was observed using an Olympus IX71 microscope (white arrows, pyroptotic cells). (C) The release of LDH from 3C^pro-transfected cells was measured using an LDH cytotoxicity assay kit. (D) Proteolytic cleavage of pGSDME in SK6 cells by 3C^pro. SK6 cells were cotransfected with plasmids encoding 3C^pro and pGSDME for 24 h. Proteolytic cleavage of pGSDME in SK6 cells by 3C^pro was determined by Western blotting. Two lanes represent two biological replicates. The data in panel C are presented as means ± the SD. Experiments were performed independently with at least three biological replicates. A Student t test was used for analysis. *, P < 0.05.

Catalytic residues in the active sites of 3C^pro are essential for the pGSDME cleavage induced by 3C^pro.

The catalytic triad of H46, D84, and C163 in FMDV 3C^pro has been identified as critical sites that play important roles in its enzyme activity (9, 10, 36). To determine whether the enzyme activity of 3C^pro is necessary for pGSDME cleavage, three 3C^pro mutants (HA-3C^pro-H46Y, HA-3C^pro-D84N, and HA-3C^pro-C163G) that eliminated the protease activity of 3C^pro were constructed. HEK-293T cells were cotransfected with plasmids encoding pGSDME and 3C^pro or its mutants for 12 h. The results showed that the elimination of the enzyme activity of 3C^pro abolished its cleavage effect on pGSDME, indicating that the enzyme activity of 3C^pro was necessary for the cleavage of pGSDME (Fig. 4A). LDH release from 3C^pro mutant-transfected cells was suppressed compared to wild-type HA-3C^pro (HA-3C^pro-WT) (Fig. 4B). To determine whether 3C^pro interacts with pGSDME for cleavage, we performed coimmunoprecipitation analysis. The results showed that 3C^pro interacts with and cleaves pGSDME (Fig. 4C). Due to the strong degradation effect of 3C^pro-WT on pGSDME, we did not detect pGSDME in the immunoprecipitation samples. As expected, we detected the cleavage product of pGSDME in the input samples (Fig. 4C). In addition, with the elimination of the enzyme activity of 3C^pro, the number of dead cells decreased significantly (Fig. 4D). Consistent with the results of coimmunoprecipitation, 3C^pro mutants and pGSDME had good intracellular colocalization. In contrast, only a small amount of colocalization was observed in the WT group due to the degradation of pGSDME by 3C^pro-WT (Fig. 4E). Collectively, these results suggested that catalytic residues in the active sites of 3C^pro are essential for the pGSDME cleavage induced by 3C^pro.

FIG 4.

Catalytic residues in the active sites of 3C^pro are essential for the pGSDME cleavage induced by 3C^pro. (A) The effects of catalytic residues of the 3C^pro active sites on 3C^pro-induced pGSDME cleavage. HEK-293T cells were cotransfected with plasmids encoding pGSDME and 3C^pro or its mutants for 12 h. Proteolytic cleavage of pGSDME in HEK-293T cells by 3C^pro or its mutants was determined by Western blotting. (B) The release of LDH from pGSDMD and 3C^pro-transfected cells was measured with an LDH cytotoxicity assay kit. (C) FMDV 3C^pro interacts with and cleaves pGSDME. HEK-293T cells were cotransfected with plasmids encoding pGSDME and 3C^pro or its mutants for 12 h. The interaction between 3C^pro and pGSDME was analyzed by using coimmunoprecipitation. (D) PI staining of the transfected cells. (E) The interaction between 3C^pro and pGSDME was analyzed by using confocal immunofluorescence. The data in panel B are presented as means ± the SD. Experiments were performed independently with at least three biological replicates. One-way ANOVA was used for analysis. **, P < 0.01.

Rupintrivir inhibits FMDV-induced pyroptosis in SK6 cells.

Rupintrivir (AG7088) (Fig. 5A) is a potent inhibitor of picornavirus 3C^pro and has been applied in many studies against picornaviruses (37–39). Since the enzyme activity of 3C^pro is critical for pGSDME cleavage, we used the 3C^pro inhibitor rupintrivir to further investigate its important role in pyroptosis induced by FMDV infection. First, we performed molecular docking analysis to evaluate the affinity of rupintrivir for FMDV 3C^pro. Results showed that rupintrivir could effectively bind to 3C^pro. In addition, the hydrophobic pocket containing the catalytic residues (H46 and D84) of 3C^pro was occupied successfully by rupintrivir (Fig. 5B). Next, we assessed the inhibitory effect of rupintrivir on the enzyme activity of 3C^pro. HEK-293T cells were cotransfected with plasmids encoding 3C^pro and pGSDME for 12 h in the presence or absence of rupintrivir. We found that rupintrivir inhibits the enzyme activity of 3C^pro in a dose-dependent manner (Fig. 5C). Then, SK6 cells were infected with FMDV at an MOI of 1 for 12 h in the presence or absence of rupintrivir (8 μM). PI staining and LDH release assays revealed significant inhibition of FMDV proliferation in rupintrivir-treated cells (Fig. 5D and E). As expected, the rupintrivir treatment inhibited: (i) the cleavage of pGSDME, (ii) the activation of caspase-3, and (iii) the maturation of IL-1β. (Fig. 5F). Moreover, rupintrivir significantly inhibited the secretion of proinflammatory cytokines IL-1β and IL-18 in FMDV-infected SK6 cells (Fig. 5G and H).

FIG 5.

Rupintrivir inhibits FMDV-induced pyroptosis in SK6 cells. SK6 cells were infected with FMDV O/HN/CHA/93 strain at an MOI of 1 for 12 h in the presence or absence of rupintrivir (8 μM). (A) Chemical formula of rupintrivir (AG7088). (B) Binding mode of rupintrivir to FMDV 3C^pro. Molecular docking studies were performed by Autodock Vina 1.2.2 (http://autodock.scripps.edu/) and visualized using the PyMOL 2.5 software (DeLano Scientific LLC), with PDB ID 5hm2 (FMDV 3C^pro) and 4ght (EV71 3C^pro-AG7088 complex) used as the templates. The yellow structure represents rupintrivir, and the pink residues represent His46 (H46) and Asp84 (D84). (C) Rupintrivir inhibits the activity of 3C^pro in a dose-dependent manner. HEK-293T cells were cotransfected with plasmids encoding 3C^pro and pGSDME for 12 h in the presence or absence of rupintrivir. Proteolytic cleavage of pGSDME in HEK-293T cells by 3C^pro was determined by Western blotting. (D) PI staining of SK6 cells with FMDV or mock infection. (E) The release of LDH from FMDV-infected cells was measured with an LDH cytotoxicity assay kit. (F) Proteolytic cleavage of GSDME in SK6 cells with FMDV infection was determined by Western blotting. The abundance of caspase-3 (CASP3), IL-1β, FMDV VP1 protein, and α-tubulin as an internal control was also determined by Western blotting. (G and H) The release of IL-1β and IL-18 from FMDV-infected cells was measured with porcine IL-1β and IL-18 using an ELISA kit. The data in panels E, G, and H are presented as means ± the SD. Experiments were performed independently with at least three biological replicates. Two-way ANOVA was used for analysis. ****, P < 0.0001.

FMDV 3C^pro cleaves pGSDME adjacent to the cleavage site of pCASP3.

Previous studies have shown that human GSDME (hGSDME) has a cleavage motif (267DMPD270) that can be recognized by caspase-3 (18, 21). However, in porcine GSDME (pGSDME), the cleavage site by porcine caspase-3 (pCASP3) remains ambiguous. To address this question, we first aligned the amino acid sequences of pGSDME, hGSDME, and mouse GSDME (mGSDME) and revealed that there might be a similar cleavage motif (265DMTD268) in pGSDME to those in hGSDME (Fig. 6A and B). We performed a cleavage experiment using the D268A mutant to verify this site. HEK-293T cells were cotransfected with plasmids encoding pCASP3 and pGSDME or its mutants for 24 h. Meanwhile, tumor necrosis factor alpha (TNF-α; 50 ng/mL) was added to activate pCASP3 (18). As shown in Fig. 6C, the D268A mutation appears to remove the cleavage effect of pCASP3 on pGSDME, suggesting that pCASP3 might cleave pGSDME at D268-A269. Intriguingly, there is a clear decrease in pCASP3 cleavage of pGSDME-Q271A mutant, indicating that this mutation also interferes with pCASP3 processing of pGSDME. Since 3C^pro can also cleave pGSDME independently, we further explored the cleavage site of 3C^pro on pGSDME. Based on the cleavage preference of picornavirus 3C^pro (Q-G or E-Q) (26, 27) and the cleavage fragment size of pGSDME, we constructed a panel of potential pGSDME mutants. We found that the pGSDME-Q271A mutant was resistant to 3C^pro-induced cleavage (Fig. 6D; see also Fig. S3 in the supplemental material), indicating that 3C^pro probably cleaves pGSDME at Q271-G272. Collectively, FMDV 3C^pro cleaves pGSDME adjacent to the cleavage site of pCASP3.

FIG 6.

FMDV 3C^pro cleaves pGSDME adjacent to the cleavage site of pCASP3. (A) Sequence alignment of GSDME in pigs, humans, and mice. (B) Logo analysis of the cleavage sites predicted from the polyprotein cleavage of pCaspase-3 (pCASP3) or 3C^pro. (C) Cleavage sites of pGSDME induced by pCASP3. HEK-293T cells were cotransfected with plasmids encoding pCASP3 and pGSDME or its mutants for 24 h in the presence or absence of TNF-α (50 ng/mL). (D) Cleavage sites of pGSDME induced by 3C^pro. HEK-293T cells were cotransfected with plasmids encoding 3C^pro and pGSDME or its mutants for 12 h. Proteolytic cleavage of pGSDME or its mutants in HEK-293T cells by pCASP3 or 3C^pro was determined by Western blotting.

pCASP3 or 3C^pro-mediated cleavage fragment pGSDME-NT localizes to the cell membrane and triggers pyroptosis.

The gasdermin-N domain oligomerizes and targets the cell membrane to form membrane pores, thus triggering pyroptosis (11, 26, 27, 31, 40–42), which provoked us to wonder whether overexpression of pCASP3 or 3C^pro-mediated cleavage fragment pGSDME-NT was sufficient to induce pyroptosis. HEK-293T cells were transfected with recombinant plasmids encoding pGSDME or its cleavage fragments for 48 h (Fig. 7A and B). Confocal immunofluorescence analysis revealed that the N-terminal fragments of pGSDME cleaved by pCASP3 and 3C^pro was localized to the cell membrane, whereas the full-length and C-terminal fragments of pGSDME were found throughout the cytoplasm (Fig. 7C). Furthermore, the LDH release of pGSDME-NT1-268- and pGSDME-NT1-271-transfected cells was significantly higher than that of other transfected cells (Fig. 7D). The PI uptake in N-terminal transfected cells increased dramatically (Fig. 7E). As expected, the N-terminal transfected cells exhibited typical pyroptotic morphology (Fig. 7F). Taken together, these results indicated that pCASP3- or 3C^pro-mediated cleavage fragment pGSDME-NT localizes to the cell membrane and triggers pyroptosis.

FIG 7.

pCASP3 or 3C^pro -mediated cleavage fragment pGSDME-NT localizes to the cell membrane and triggers pyroptosis. (A) Schematic diagram of pGSDME and its cleavage fragments induced by pCaspase-3 (pCASP3) or 3C^pro. HEK-293T cells were transfected with recombinant plasmids encoding pGSDME or its cleavage fragments for 48 h. (B) The expression of pGSDME and its cleavage fragments was determined by Western blotting. (C) The localization of pGSDME or its cleavage fragments was analyzed by confocal immunofluorescence. (D) The release of LDH from the transfected cells was measured with an LDH cytotoxicity assay kit. (E) PI staining of the transfected cells. (F) The cell morphology was observed using an Olympus IX71 microscope (white arrows, pyroptotic cells). The data in panel D are presented as means ± the SD. Experiments were performed independently with at least three biological replicates. One-way ANOVA was used for analysis. ****, P < 0.0001.

pGSDME knockdown attenuates FMDV-induced pyroptosis in SK6 cells.

To further confirm the crucial role of GSDME in FMDV infection-induced pyroptosis, pGSDME knockdown was performed using short hairpin RNAs (shRNAs). First, we verified the knockdown efficiency of the three shRNAs, and the results showed that shRNA3 had the highest efficiency (Fig. 8A). Therefore, we chose it for subsequent experiments. We then determined that the knockdown of pGSDME did not affect FMDV replication (see Fig. S4). SK6 cells were transduced with shRNA lentivirus targeting pGSDME (sh-pGSDME) or negative control (sh-NC) for 24 h, and then the cells were infected with FMDV at an MOI of 1 for 12 h. Morphological examination showed that knockdown of pGSDME reduced cell death caused by FMDV infection (Fig. 8B). LDH release assays and PI staining also revealed that pGSDME knockdown attenuated cell death induced by FMDV infection (Fig. 8C and D). In addition, pGSDME knockdown suppressed the cleavage of pGSDME and maturation of IL-1β (Fig. 8E). Moreover, pGSDME knockdown significantly diminished the secretion of proinflammatory cytokines IL-1β and IL-18 in FMDV-infected SK6 cells (Fig. 8F and G). These results indicated that pGSDME knockdown attenuates FMDV-induced pyroptosis in SK6 cells.

FIG 8.

pGSDME knockdown attenuates FMDV-induced pyroptosis in SK6 cells. SK6 cells were transduced with lentivirus targeting pGSDME (sh-pGSDME) or negative control (sh-NC) for 24 h, and then the cells were infected with FMDV O/HN/CHA/93 strain at an MOI of 1 for 12 h. (A) The knockdown efficiency of pGSDME was determined by Western blotting. (B) Morphological examination of SK6 cells with FMDV or mock infection. The cell morphology was observed using an Olympus IX71 microscope (white arrows, pyroptotic cells). (C) The release of LDH from FMDV-infected cells was measured with an LDH cytotoxicity assay kit. (D) PI staining of SK6 cells with FMDV or mock infection. (E) Proteolytic cleavage of GSDME in SK6 cells with FMDV infection was determined by Western blotting. The abundance of caspase-3, IL-1β, FMDV VP1 protein, and α-tubulin as an internal control was also determined by Western blotting. (F and G) The release of IL-1β and IL-18 from FMDV-infected cells was measured with porcine IL-1β and IL-18 using an ELISA kit. The data in panels C, F, and G are presented as means ± the SD. Experiments were performed independently with at least three biological replicates. Two-way ANOVA was used for analysis. *, P < 0.05; ***, P < 0.001.

DISCUSSION

FMD is listed as a class A infectious disease by the World Organization for Animal Health (OIE) due to its wide prevalence, multiple transmission routes, and fast spread speed. In addition, it always threatens the safety of livestock and the trade of livestock products in countries and regions without FMD (4). Thus, detailed research is required on its pathogenic mechanisms to provide better guidance for preventing and controlling FMD. In the present study, we revealed that FMDV infection induces GSDME-mediated pyroptosis independent of caspase-3 activity. In addition, FMDV 3C^pro induces pyroptosis by cleaving pGSDME adjacent to the cleavage site of pCaspase-3, and the protease activity of 3C^pro is required for pGSDME cleavage (Fig. 9).

FIG 9.

Schematic diagram showing the proposed mechanism of FMDV-induced GSDME-mediated pyroptosis. FMDV infection induces GSDME-mediated pyroptosis independent of caspase-3 activity. In addition, FMDV 3C^pro induces pyroptosis by cleaving pGSDME adjacent to the cleavage site of pCaspase-3, and the protease activity of 3C^pro is required for pGSDME cleavage.

Several pore-forming proteins in the gasdermin family are involved in various cellular processes, such as inflammation and cell death (43). Previous studies have shown that GSDMD, a pore-forming protein, is critical in pyroptosis induced by pathogen infection (11, 26, 30, 31, 40, 44–47). Gram-negative bacterial infection activates mouse caspase-11 through LPS and induces GSDMD-mediated noncanonical pyroptosis (46). SARS-CoV-2 NSP6 triggers NLRP3-dependent pyroptosis by targeting ATP6AP1 (47). SVV 3C^pro induces pyroptosis in picornaviruses by directly cleaving porcine GSDMD (26). Recent studies have shown that other members of the gasdermin family also play key roles in pyroptosis. GSDME was initially identified as being associated with deafness disorders (11). Accumulating evidence indicates that GSDME switches caspase-3-mediated apoptosis to pyroptosis (18, 20). Furthermore, GSDME is required to induce pyroptosis and severe disease during EV71 infection (21). Moreover, ZIKV causes placental pyroptosis and associated adverse fetal outcomes through the activation of GSDME (22). In this study, initial morphological observations indicated that FMDV infection of SK6 cells could induce pyroptosis (swelling and rounding of the cells and the shape of fried eggs in the middle of the bulge) (Fig. 1A). To further confirm the pyroptosis caused by FMDV infection, we detected the key proteins in the pyroptosis pathway (activation of CASP1/CASP3 and cleavage of GSDMD/GSDME) by Western blotting, and the results showed that the activation of CASP3 and the cleavage of GSDME were detected (Fig. 1G). In addition, we also detected some other common features of cell death, such as cell viability, LDH and inflammatory cytokines (IL-1β and IL-18) levels in the cell culture supernatant, and PI staining of the cell nucleus. In summary, these results demonstrated that GSDME mediates FMDV infection-induced pyroptosis. Therefore, our study might reveal a previously unrecognized novel mechanism of pyroptosis induced by FMDV infection.

FMDV 3C^pro affects antiviral innate immune responses by degrading or cleaving various host proteins (6–10). Here, we revealed for the first time that pGSDME is a novel cleavage substrate of FMDV 3C^pro and can induce pyroptosis. In addition, we found that FMDV 3C^pro cleaved pGSDME at a putative cleavage site (Q271-G272), which is similar to other picornaviruses (26, 27). However, FMDV 3C^pro had only one cleavage site at the N-terminal of pGSDME, which was close to the cleavage site of pCaspase-3, while in other picornaviruses, 3C^pro had two cleavage sites at the N-terminal of pGSDMD and inhibited pyroptosis by cleavage of the pGSDMD-NT fragment (26, 27). Our results suggest that 3C^pro has different cleavage modes for pyroptotic executive proteins. The protease activity of 3C^pro is crucial for its function, and mutation of the active sites of 3C^pro will abolish the cleavage function (10, 38). Hence, 3C^pro is a potential FMDV antiviral therapy target (37–39). In this study, 3C^pro lost the ability to cleave GSDME after mutating the catalytic triad. In addition, we further demonstrated the importance of the protease activity of 3C^pro by using the 3C^pro inhibitor rupintrivir (AG7088). Molecular docking analysis showed that rupintrivir inhibited protease activity, possibly by binding to the enzyme active site of 3C^pro. Small molecule anti-FMDV drugs with rupintrivir analogs will be of interest.

Pyroptosis is a double-edged sword. On the one hand, pyroptosis is beneficial to the elimination of pathogens. On the other hand, pyroptosis activates massive inflammatory factors to induce the inflammatory response, which may be the main cause of clinical symptoms such as blisters and ulcers in pigs caused by FMDV infection. Previous studies have shown that GSDME deficiency inhibits pyroptosis and alleviates pathological symptoms caused by EV71 infection in mice (21). In addition, adverse fetal outcomes associated with placental pyroptosis caused by ZIKV infection were effectively attenuated in GSDME-deficient mice (22). In our study, the knockdown of GSDME attenuated the pyroptosis caused by FMDV infection, but it was insufficient to abolish the cell death. The possible reason is that there might be multiple forms of cell death (pyroptosis, necroptosis, apoptosis, or ferroptosis) caused by FMDV infection (34, 35, 48). Given the current difficulties and limitations in developing and applying FMD vaccines, the design and comprehensive application of antiviral drugs targeting 3C^pro might be a more feasible strategy for preventing and controlling the disease.

MATERIALS AND METHODS

Materials and reagents.

jetPRIME in vitro DNA and siRNA transfection reagent was purchased from Polyplus-Transfection Inc. (New York, NY). A Pierce BCA protein assay kit and Alexa Fluor 555-goat anti-mouse antibody were purchased from Thermo Fisher Scientific (Waltham, MA). Dulbecco modified Eagle medium (DMEM), fetal bovine serum (FBS), and Pen-Strep antibiotics (10,000 U/mL penicillin and 10,000 µg/mL streptomycin) were purchased from Gen-View Scientific, Inc. (Jacksonville, FL). Mouse anti-His monoclonal antibody (MAb), anti-Flag MAb, anti-HA MAb, rabbit anti-HA MAb, and horseradish peroxidase-conjugated goat anti-mouse or rabbit IgG were obtained from Medical & Biological Laboratories Co., Ltd. (Nagoya, Japan). GSDMD rabbit MAb, GSDME rabbit polyclonal antibody (pAb), IL-1β rabbit pAb, IL-18 rabbit pAb, MLKL rabbit pAb, Phospho-MLKL-T357, S358, S360 rabbit pAb, and caspase-1 rabbit pAb were purchased from ABclonal Technology Co., Ltd. (Wuhan, China). Caspase-3 rabbit pAb was purchased from Proteintech (Wuhan, China). An enhanced chemiluminescence (ECL) system was purchased from UElandy, Inc. (Suzhou, China). A Cell Counting Kit-8 and rupintrivir (AG7088) were obtained from MedChemExpress (Shanghai, China). TRIpure reagent was obtained from Aidlab Biotechnologies Co., Ltd. (Beijing, China). Hifair first-strand cDNA synthesis SuperMix and Hieff UNICON Universal Blue qPCR SYBR green Master Mix were purchased from Yeasen Biotechnology Co., Ltd. (Shanghai, China). An LDH cytotoxicity assay kit, PI, caspase inhibitor Z-VAD-FMK, caspase-3 inhibitor Ac-DEVD-CHO, dimethyl sulfoxide (DMSO), puromycin, protein A+G agarose (beads), and DAPI (4′,6′-diamidino-2-phenylindole) were purchased from Beyotime Biotechnology (Shanghai, China). FMDV VP1 rabbit polyclonal antibody was prepared and maintained in our laboratory.

Swine Kidney-6 (SK6) cells, Porcine Kidney-15 (PK-15) cells, Baby Hamster Syrian Kidney (BHK-21) cells, and human embryonic kidney (HEK-293T) cells were cultured in DMEM supplemented with 10% FBS and 1% Pen-Strep antibiotics at 37°C with 5% CO₂.

FMDV O/HN/CHA/93 strain was maintained in the National Foot-and-Mouth Disease Reference Laboratory, Lanzhou Veterinary Research Institute, Chinese Academy of Agricultural Sciences. All virus experiments were conducted at the Biosafety Level 3 Laboratory, Lanzhou Veterinary Research Institute, Chinese Academy of Agricultural Sciences.

Plasmid construction and transfection.

All the eukaryotic expression plasmids of FMDV viral proteins used in this study were constructed and cloned into pCAGGS-HA, pEBG-GST, or pEGFP-C1 vector using the primers listed in Table S1 in the supplemental material. Three 3C^pro mutants (HA-3C^pro-H46Y, HA-3C^pro-D84N, and HA-3C^pro-C163G) that eliminated the protease activity of 3C^pro were constructed and cloned into the pCAGGS-HA vector. The porcine GSDME (pGSDME) gene (GenBank accession no. XM_003134855.4) was amplified from SK6 cells and cloned into pcDNA3.1(+)3×Flag vector, and the mutants and cleavage fragments of pGSDME were also cloned into pcDNA3.1(+)3×Flag vector. Porcine caspase-3 (GenBank accession no. NM_214131.1) was synthesized by Sangon Biotech (Shanghai) Co., Ltd., and cloned into pcDNA3.1(+)3×Flag vector, and the 3×Flag tag was substituted with 6×His tag. Cell transfection was conducted using the jetPRIME in vitro DNA and siRNA transfection reagent according to the manufacturer’s instructions.

Drug treatment.

In caspase-3 activity inhibition experiments, caspase-3 inhibitor Ac-DEVD-CHO or DMSO was added to the medium 2 h before virus infection. For 3C^pro enzyme activity inhibition tests, rupintrivir (AG7088) was added to the medium for 4 h after plasmid transfection or virus infection.

Knockdown of pCASP3 or pGSDME.

pCASP3 knockdown was performed using small interfering RNAs (siRNAs). SK6 cells were transfected with siRNAs targeting pCASP3 for 48 h. The sequences (5′–3′) of siRNAs were as follows: siRNA-1 (CAGCGCUGAAACAGUAUGU and ACAUACUGUUUCAGCGCUG), siRNA-2 (CUGCUGUAGAACUCUAACU and AGUUAGAGUUCUACAGCAG), and siRNA-3 (CCGAAAGGUAGCAGUAGAA and UUCUACUGCUACCUUUCGG). Three shRNAs targeting pGSDME were cloned into the lentiviral vector pLKO.1. For lentivirus generation, 0.7 μg of psPAX2, 0.3 μg of pMD2.G, and 1 μg of shRNA constructs were cotransfected into HEK-293T cells. After 48 h of transfection, the lentivirus was harvested and further transduced into SK6 cells. The shRNA-expressing cells were selected with 1.5 μg/mL of puromycin and then plated into 12-well plates for further experiments. The sequences (5′–3′) of shRNAs were as follows: shRNA-1, GGCTCTTGAAGAGGAACATCC; shRNA-2, GCTGACATTAACCTGGAAATG; and shRNA-3, GGAGAACAATCGATCTGAAGA.

Western blotting.

Cell samples were harvested, lysed, and quantified by using a Pierce BCA protein assay kit. Then, the lysate samples were mixed with 5× loading buffer, boiled at 95°C for 10 min, subjected to 12% SDS-PAGE, and transferred onto polyvinylidene difluoride membranes. The membranes were blocked, washed, and incubated with the indicated primary and secondary antibodies. After intensive washing, protein bands were detected using an ECL system and analyzed with Image Lab software 4.0.1 (Bio-Rad, Hercules, CA).

Cytotoxicity assay.

Cell death was assessed by measuring cell viability or LDH release with a Cell Counting Kit-8 and an LDH cytotoxicity assay kit, respectively.

Propidium iodide staining.

Cell membrane integrity was assessed by observing the cells’ PI uptake. The infected or transfected cells were stained with PI at the indicated times. The fluorescence signal was observed by using an inverted fluorescence microscope (Ti-U-Nikon, Tokyo, Japan).

Quantitative reverse transcription-PCR.

Cell samples were collected, and total RNA was extracted with TRIpure Reagent, followed by reverse transcription with Hifair first-strand cDNA synthesis SuperMix. All qPCR assays were performed with Hieff UNICON Universal Blue qPCR SYBR green Master Mix on a ViiA 7 real-time PCR system (Applied Biosystems, Grand Island, NY) according to the manufacturer’s instructions.

Determination of TCID₅₀.

BHK-21 cells were cultured in DMEM supplemented with 10% FBS and 1% Pen-Strep antibiotics at 37°C with 5% CO₂ and seeded into 96-well plates. Then, the diluted virus was incubated with the cells for 3 days. The cytopathic effect was observed daily, and the virus titers were calculated according to the Reed-Muench method (49).

Coimmunoprecipitation assay.

HEK-293T cells were cultured in DMEM supplemented with 10% FBS and 1% Pen-Strep antibiotics at 37°C with 5% CO₂ and seeded into 12-well plates. The cells were transfected with the indicated plasmids for 12 h. Then, the cell samples were harvested, lysed, and subjected to Western blotting and immunoprecipitation analysis. Briefly, the lysates were centrifuged at 4°C for 10 min, and the supernatants were incubated with anti-HA binding beads at 4°C for 4 h. Subsequently, the binding beads were washed with PBST six times and then denatured in 1× SDS-PAGE loading buffer for 10 min. Finally, the supernatants were analyzed by Western blotting.

Confocal immunofluorescence assay.

HEK-293T cells were seeded on coverslips in a 24-well plate and transfected with the indicated plasmids. After 12 h of transfection, cells were washed with phosphate-buffered saline (PBS) three times, fixed with 4% paraformaldehyde for 30 min at room temperature, permeabilized with 0.5% Triton X-100 for 20 min at –20°C, and further blocked with 2% bovine serum albumin for 2 h at room temperature. Subsequently, the blocked cells were incubated with the primary antibodies (rabbit anti-HA MAb or mouse anti-Flag MAb) for 2 h at 37°C. The cells were then washed with PBST three times and incubated with the secondary antibodies (Alexa Fluor 488-goat anti-rabbit antibody or Alexa Fluor 555-goat anti-mouse antibody) for 1 h at 37°C. Nuclei were stained with DAPI. After intensive washing, the fluorescence signal was analyzed using a laser scanning confocal microscope (LSM 510; Zeiss, USA).

Molecular docking analysis.

To analyze the binding affinities and modes of interaction between rupintrivir and FMDV 3C^pro, AutodockVina 1.2.2, a silico protein-ligand docking software, was employed (50, 51). The molecular structure of rupintrivir (AG7088) was retrieved from the EV71 3C^pro-AG7088 complex (PDB ID 4ght). The three-dimensional coordinate of FMDV 3C^pro (PDB ID 5hm2) was downloaded from the PDB (http://www.rcsb.org/pdb/home/home.do). For docking analysis, all protein and molecular files were converted into PDBQT format with all water molecules excluded and polar hydrogen atoms added. The grid box covered each protein’s domain and accommodated free molecular movement. Molecular docking studies were performed by Autodock Vina 1.2.2 (http://autodock.scripps.edu/) and visualized using the PyMOL 2.5 software (DeLano Scientific LLC).

Sequence alignment and logo analysis.

Porcine GSDME (GenBank accession no. XM_003134855.4), human GSDME (GenBank accession no. NM_004403.3), and mouse GSDME (GenBank accession no. NM_018769.4) amino acid sequences were obtained from GenBank (https://www.ncbi.nlm.nih.gov/nuccore/). Sequence alignment was performed by CLUSTALW (https://www.genome.jp/tools-bin/clustalw) and visualized with BioEdit v7.2.5 (http://www.mbio.ncsu.edu/BioEdit/bioedit.html). Logo analysis of the cleavage sites predicted from the polyprotein cleavage of pCaspase-3 or 3C^pro was generated by WebLogo (https://weblogo.threeplusone.com/).

Statistical analysis.

All data were analyzed using Prism 8.0 software (GraphPad Software, Inc., La Jolla, CA) and are presented as means ± the standard deviations (SD) of three independent experiments. Comparisons were performed through one- or two-way analysis of variance (ANOVA), followed by Tukey’s test. A P value of <0.05 was considered statistically significant.