NLRP3 inflammasome activation by Foot-and-mouth disease virus infection mainly induced by viral RNA and non-structural protein 2B

ABSTRACT

Foot-and-mouth disease virus (FMDV) is a positive-strand RNA virus of the family Picornaviridae. Early studies show that some viruses of Picornaviridae, such as EMCV and EV71, induce NLRP3 inflammasome activation. Our current study demonstrates that FMDV induces the secretion of caspase-1 and interleukin 1 beta (IL-1β), as well as activates the NLRP3 inflammasome in a dose- and time-dependent manner. Meanwhile, NLRP3 inflammasome can suppress FMDV replication during virus infection. Both FMDV RNA and viroporin 2B stimulate NLRP3 inflammasome activation. FMDV RNA triggers NLRP3 inflammasome through p-NF-κB/p65 pathway not dependent on RIG-I inflammasome. FMDV 2B activates NLRP3 inflammasome through elevation of intracellular ion, but not dependent on mitochondrial reactive oxygen species (ROS) and lysosomal cathepsin B. It further demonstrates that 2B viroporin activates NLRP3 inflammasome and induces IL-1β in mice, which enhances the specific immune response against FMDV as an ideal self-adjuvant for FMD VLPs vaccine in guinea pigs. The results reveal a series of regulations between NLRP3 inflammasome complex and FMDV. Amino acids 140–145 of 2B is essential for forming an ion channel. By mutating the amino acid and changing the hydrophobic properties, the helical transmembrane region of the viroporin 2B is altered, so that the 2B is insufficient to trigger the activation of NLRP3 inflammasome. This study demonstrates the functions of FMDV RNA and 2B viroporin activate NLRP3 inflammasome and provides some useful information for the development of FMD vaccine self-adjuvant, which is also helpful for the establishment of effective prevention strategies by targeting NLRP3 inflammasome.

1. Introduction

Foot-and-mouth disease (FMD) is an acute, febrile and contact infectious disease caused by Foot-and-mouth disease virus (FMDV) infection in cloven-hoofed animals. Therefore, after each outbreak, infected animals will be slaughtered and burned, which the production of livestock will decline markedly. In addition, there are individual variations of FMDV that can be transmitted to humans. FMD is known as the ‘number one killer’ of animal husbandry. However, the molecular mechanisms by which FMDV stimulates inflammation are not defined. Macrophages play a pivotal role in triggering inflammation during FMDV infection [1].

The innate immune system is a universal form of first-line defence of the host against virus invasions [2–5]. Its recognition relies on germline-encoded pattern recognition receptors (PRRs), including Toll-like receptors (TLRs), retinoic acid-inducible gene (RIG-I)-like receptors, NOD-like receptors (NLRs), and C-type lectin receptors (CLRs) [6–14]. NLRs, as one of PRRs, recognize PAMPs (pathogen-associated molecular patterns) or DAMPs (damage-associated molecular patterns) and trigger assembly of ‘inflammasome’, which can lead to the activation of caspase-1 and the secretion of pro-inflammatory cytokines, such as IL-1β and IL-18, to further regulate the innate immune response [15]. The NLRP3 inflammasome, including NLRP3, apoptosis-associated speck-like protein (ASC), and pro-Caspase-1, are the most fully identified. Once NLRP3 recognizes some signals of PAMPs or DAMPs, it oligomerizes through the oligomerization domain and directly recruits the ASC CARD domain through the N-terminal PYRIN domain of NLRP3. Afterwards, the CARD domain of ASC directly recruits pro-caspase-1 which will cleave to mature caspase-1 and finally make pro-IL-1β to IL-1β, inducing the inflammation.

HCV is a single-stranded positive-sense RNA virus, induces chronic inflammation and mediates liver damage. HCV induces IL-1β/IL-18 mRNA expression through NF-κB activation and produces IL-1β and IL-18 from a monocyte macrophage [16]. HCV RNA can activate NLRP3 inflammasome and induce IL-1β secretion with transfection into monocytes or macrophages. HCV RNA activates signal 2 and triggers ASC oligomerization and caspase-1 cleavage not dependent on RIG-I. HCV RNA induces activation of the NLRP3 inflammasome dependent on ROS production [17]. HCV RNA triggers NLRP3 inflammasome activation through MyD88-mediated TLR7 signalling to induce IL-1β mRNA expression and drive IL-1β secretion [18]. Meanwhile, FMDV RNA is a single-stranded positive-sense RNA, which may be similar to HCV RNA.

Many kinds of viruses can induce inflammatory responses and produce pro-inflammatory cytokines to stimulate NLRP3 inflammasome and prevent pathogenic invasions [19,20]. FMDV has not been reported whether it can also activate NLRP3 inflammasome. At present, some studies have reported FMDV can elicit an innate immune response in macrophage cells [1]. Especially, Protein VP1, L^pro, 3C^pro, 2C, and 2B of FMDV can inhibit the expression of IFN-α/β, in which the cellular proteins, eIF4G, NEMO, KPNA1, and RIG-I are involved [21–28]. Previous studies confirm that FMDV 2B protein is a viroporin-like protein and forms hydrophilic pore at the cellular membranes [29–31]. Recently, some reports have shown that viroporins play an important role in the activation of the NLRP3 inflammasome by regulating the antiviral innate immune responses [32]. These viroporins mainly include the influenza virus M2 protein [33–35], encephalomyocarditis virus (EMCV) 2B protein [36], human respiratory syncytial virus SH [37], human rhinovirus 2B protein [38], and hepatitis C virus (HCV) p7 protein [16,18]. Hence, it will be very interesting to find a new function of FMDV 2B in activation of the NLRP3 inflammasome.

To assess the contribution of NLRP3 inflammasome and IL-1β secretion in FMDV pathogenesis, we firstly investigated the inflammasome response to FMDV infection and then focused on the FMDV RNA and 2B viroporin function of inflammasome activation. Our study reveals that FMDV can activate the NLRP3 inflammasome. NLRP3 inflammasome plays a protective role against FMDV infection. Furthermore, we find that FMDV RNA can activate p-NF-κB/p65 pathway and viroporin 2B protein can induce ion efflux and trigger NLRP3 inflammasome activation. 2B protein activates NLRP3 inflammasome in mice and enhances the specific immune response against FMDV as an ideal self-adjuvant for FMD VLPs vaccine in guinea pigs. Helical transmembrane region of the viroporin 2B is important to induce ion channel to activate NLRP3 inflammasome. This study could provide some useful information to scientists for the development of antiviral strategy targeting NLRP3 inflammation pathway.

2. Materials and methods

2.1. Mice

8–10-week old C57BL/6 mice were used. NLRP3^−/- and ASC^−/- mice were given by Dr. Zhengfan Jiang in Peking University of China.

2.2. Cell and virus

Mouse monocyte macrophage cell line (RAW264.7), baby hamster Syrian kidney (BHK-21), porcine kidney epithelial cell (PK-15), and swine testis cell (ST) were purchased from American Type Culture Collection (Manassas, VA, USA). Mouse bone marrow-derived macrophage cell line (iBMM) was a gift from Dr. Zhengfan Jiang in Peking University of China. All cell lines above were cultured in Dulbecco’s modified Eagle’s medium (Gibco) supplemented with 10% foetal bovine serum (FBS) (Gibco) at 37°C with 5% CO₂ atmosphere. Porcine myelomonocytic cell lines (3D4/2), which were given by Dr. Zengjun Lu in Lanzhou Veterinary Research Institute of China, were cultured in RPMI 1640 modified medium (Gibco) supplemented with 10% FBS at 37°C with 5% CO₂ atmosphere.

Mouse bone marrow-derived macrophage (BMM) and Mouse bone marrow-derived dendritic cell (BMDC) were isolated and cultured from NLRP3^−/-, ASC^−/- and WT mice as described previously [36].

The serotype O of FMDV (GenBank accession number: F149009) was passaged in BHK-21 cells at 1 MOI. The viral titres were calculated using the Reed–Muench method as 50% tissue culture infective dose. The virus stock was clarified and stored at −80°C. For inactivation of the virus, ultraviolet light was used at 9500 w for 15 min with spectrolinker UV crosslinker.

2.3. Microarray data deposition

RAW264.7 macrophages were infected with FMDV or mock at 9 h post-infection (hpi) to analysis total RNA using an Agilent Mouse Gene Expression Microarray 8 × 60 K (G4852A). Data were submitted to the gene expression omnibus (GEO). The GEO accession number is GSE96588.

2.4. Antibodies, siRNA, and chemicals

Monoclonal mouse anti-NLRP3 antibody (AG-20B-0014), polyclonal rabbit anti-ASC antibody (AG-25B-0006), and monoclonal mouse anti-Caspase-1 antibody (AG-20B-0042) were purchased from AdipoGen (San Diego, USA). Polyclonal rabbit anti-IL-1β antibody (ab9722), polyclonal rabbit anti-NLRP3 antibody (ab214185), polyclonal rabbit anti-DDX58 (ab45428), polyclonal rabbit anti-NF-κB/p65 (phospho S536) (p-p65) antibody (ab86299) and rabbit anti-HA tag antibody (ab9110) were purchased from Abcam (Cambridge, UK). Monoclonal mouse anti-flag (sc-166355), monoclonal mouse anti-HA (sc-7392), monoclonal mouse anti-NF-κB/p65 antibody (sc-8008) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Polyclonal rabbit anti-flag (SAB4301135), monoclonal mouse anti-β-actin (A5314), and all of the secondary antibodies conjugated with HRP, FITC, and TRITC were purchased from Sigma–Aldrich (Louis, MO, USA). Polyclonal rabbit anti-TRAF3 antibody (#4729), monoclonal rabbit anti-IRF-3 antibody (#4302) and monoclonal rabbit anti-phospho-IRF-3 antibody (#4947) were purchased from Cell Signal Technology (CST, Beverly, MA, USA). Polyclonal pig anti-FMDV antibody was provided by the OIE Reference Laboratory of China (Lanzhou, Gansu, China). NLRP3 siRNA (sc-45470), ASC siRNA (sc-37282), Caspase-1 siRNA (sc-29922), RIG-I siRNA (sc-61418) and control siRNA (sc-36869) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Lipopolysaccharide (LPS) (L2280), ATP (A7699), and Mito-TEMPO (SML0737) were purchased from Sigma–Aldrich (St. Louis, MO, USA). Z-YVAD-FMK (ab141388), MSU (ab144305), and BAPTA-AM (ab120503) were purchased from Abcam (Cambridge, UK). Ca-074-Me was purchased from ApexBio (Houston, USA). Fluo-3-AM calcium indicator (F1242), MitoSOX^TM red mitochondrial superoxide indicator (M36008), TRIzol reagent (15596018), Lipofectamine 2000 (11668019), and Lipofectamine RNAiMAX (13778-075) were purchased from Invitrogen (Carlsbad, California, USA). Fluo-3-AM calcium indicator (S1056) and DAPI (C1002) were purchased from Beyotime (Shanghai, China). PrimeScript RT Reagent Kit (RR036A) and SYBR Premix Ex Taq II (Tli RNaseH Plus) (RR820B) were purchased from Takara (Dalian, China). BD FACS^TM Lysing Solution (349202) was purchased from BD (Lake Franklin, New Jersey, USA). Montandie ISA 206 was purchased from Seppic (Paris, France).

2.5. Plasmid construction

The plasmids, expressing FMDV proteins 2B, 3C, 3A, 3B, and 3C, as well as 3D, VP0, VP1, and VP3, were named as p2B-HA, p2B-flag, p3C-flag, p3A-flag, p3B-flag, p3C-flag, p3D-flag, pL-flag, pVP0-flag, pVP1-flag, and pVP3-flag stored in our lab. The plasmids of pNLRP3-flag, pASC-flag, and pCaspase-1-flag were given from Dr. Changjiang Weng of the Haerbin Veterinary Research Institute of China. The mutants of p2B-HA (M1, M2 and M3) were constructed and named as p2B-M1-HA, p2B-M2-HA and p2B-M3-HA.

2.6. Infection

All types of cells were infected with FMDV at different MOIs for 1 h at 37°C. Then, the supernatants were discarded. After washing with PBS, DMEM containing 2% FBS was added for further culture. Cells were treated with the supernatants of BHK-21 cell culture lysate, as mock-infected controls (Mock).

2.7. Transfection

PK-15, RAW264.7, and iBMM were seeded on 12-well plates at approximately 1 × 10⁵/well. The cells were transfected with the indicated plasmids using Lipofectamine 2000 according to the manufacturer’s instructions. After 18–24 hpi, the cell-free supernatants were collected for ELISA, and cells were harvested and subjected to IFA, Western blot analysis, or qRT-PCR.

2.8. RNA interference

RAW264.7 and iBMM were seeded on 12-well plates with approximately 1 × 10⁵/well. The cells were transfected with the indicated plasmids using Lipofectamine RNAiMAX (Invitrogen) according to the manufacturer’s instructions. Upon reaching 16 h, cells were transfected with siRNA or FMDV RNA. The cell-free supernatants were collected for ELISA, and cell pellets were harvested and subjected to Western blot analysis or qRT-PCR.

2.9. Quantitative real-time PCR (qRT-PCR)

For qRT-PCR, all samples were collected for RNA extraction using TRIzol reagent. The concentration of FMDV RNA was determined using NanoDrop 2000 (Thermo Scientific, USA). The cDNA was obtained by reverse transcription with a PrimeScript RT Reagent Kit, and then qRT-PCR was performed using SYBR Premix Ex Taq II (Tli RNaseH Plus) to quantify the level of mRNA transcription on the ABI 7500 PCR system (Applied Biosystems, USA). The level of glyceraldehyde-3-phosphate dehydrogenase mRNA was used as an internal control. Differences in mRNA expression were represented using the 2^−ΔΔCt method. All primers were synthesized by Genewiz (Suzhou, China), and the sequences are listed in Tables S1 and S2.

2.10. Western blot analysis

All samples were loaded and ran in 10% acrylamide gel electrophoresis and then transferred to polyvinylidene difluoride membranes (Millipore, USA) using a mini-trans blot system (Bio-Rad, USA). The membranes were blocked in 5% non-fat milk/TBST for 1 h and incubated with the primary antibody at 4°C overnight. After five rounds of washing with TBST, the membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibody for 1 h at room temperature. Signals of protein were revealed using the ECL Western Blotting Analysis System.

2.11. TCID₅₀ assay

The supernatants of RAW264.7, iBMM, and PK-15 cells were collected and centrifuged, and 10-fold diluted. One hundred micro litres of supernatants and virus stock were added into BHK-21 cells which were seeded in 96-well plate at 90% confluence, at 37°C for about 72 h. Cytopathic effect (CPE) of each of well was counted. TCID₅₀/100μl values were calculated by the Reed-Muench method. Each test was performed at least three times.

2.12. ELISA

The cell supernatant was collected and analysed for the presence of Caspase-1 (AdipoGen, USA), IL-1β (mouse: eBiosciences, USA; porcine: R&D, USA) by ELISA according to the manufacturer’s instructions.

2.13. Flow cytometry

Cells were incubated with Fluo-3 AM for 1 h at 37°C. Using the minimum concentration of dye required was desirable to yield fluorescence signals with adequate signal-to-noise ratios. Before measuring fluorescence, cells were washed for five times and then incubated for 20–30 min to ensure that the Fluo-3 AM in the cells was completely transformed into Fluo-3. Cells were then washed five times and resuspended in PBS containing 2% FBS, ready for detecting. For staining of MitoSOX, cells were normally incubated with 5 µM MitoSOX for 10 min at 37°C, away from light. The cells were washed gently three times with warm PBS buffer and resuspended in PBS containing 2% FBS. All untreated cells were recognized as a negative control. Flow cytometric analysis was performed on a Guava easyCyte instrument.

2.14. Confocal microscopy

RAW264.7 was seeded on glass slides in 24-well plates and then transfected with plasmids. At 24 h post-transfection, the cells were washed with PBS, added in 4% paraformaldehyde for 15 min, and then permeabilized with 0.1% Triton-100 for 15 min. After washing, cells were blocked with PBS containing 5% FBS for 1 h and then incubated with primary and secondary antibodies for 2 h, respectively. After washing three times, the cells were incubated with DAPI for 15 min and finally analysed using a confocal scanning microscope (Leica SP8, Germany).

2.15. Animal experiments

Thirty-two SPF female BALB/c mice (20–25 g) and twenty-one SPF female guinea pigs (300–350 g) were used in this research from Experimental Animal Centre of Lanzhou Veterinary Research Institute. Animal experiment was carried out in National Foot and Mouth Disease Reference Laboratory, Lanzhou Veterinary Research Institute, under the guidelines of Animal Research Ethics Board of Lanzhou Veterinary Research Institute, Chinese Academy of Agricultural Science (CAAS), China.

BALB/c mice were randomly divided into four groups. Group 1, pH 7.4 PBS; Group 2, 50 µg FMDV serotype O VLPs (made in our lab) and 50 µg FMDV 2B (expressed and purified in our lab) emulsified by Montandie ISA 206 (Seppic, Paris, France); Group 3, 50 µg FMDV serotype O VLPs emulsified by Montandie ISA 206; Group 4, 50 µg FMDV 2B emulsified by Montandie ISA206. The BALB/c mouse were injected with 0.2 mL vaccine in tibialis cranialis muscle of both rear legs. Blood was collected at the eye socket, and sera were separated before immunization (0 day) and after immunization at 1, 3, 5 and 7 days. The spleens and the leg bones were sterilely removed. Cells of spleens and the bones of tibia and femur were washed with PBS. Erythrocyte was lysed using BD FACS^TM Lysing Solution. Total RNA of splenic lymphocytes and bone marrow cells were extracted using TRIzol reagent and determined by NanoDrop 2000. The levels of mRNA were detected by qRT-PCR. The guinea pigs were randomly divided into three groups. Group 1, pH 7.4 PBS; Group 2, 50 µg FMDV serotype O VLPs emulsified by Montandie ISA 206; Group 3, 50 µg FMDV serotype O VLPs and 50 µg FMDV 2B emulsified by Montandie ISA 206. The guinea pigs were injected with 0.2 mL vaccine in tibialis cranialis muscle of both rear legs. The serum samples of guinea pigs were collected from the heart at every 7 days. Guinea pigs were given booster dose at 28 days after the first immunization. The challenge was performed at 28 days after secondary immunizations. All guinea pigs were challenged with 0.2 mL of guinea pig infectious dose 100 ID₅₀ per guinea pig in the back sole of rear foot. After the challenge, the animals were examined daily for 7 days.

2.16. Micro-neutralization assay

Sera taken from all guinea pigs were analysed for neutralizing antibody titres by using a micro-neutralization assay with a monolayer of BHK-21 cells. Double dilutions of sera were reacted with 100 TCID₅₀ of FMDV at 37°C for 1 h. Cells were then added as indicators of residual infectivity and microplates were incubated at 37°C for 3 days. The endpoint titres were calculated as the reciprocal of the last serum dilution to neutralize 100 TCID₅₀ homologous FMDV in 50% of the wells.

2.17. Cell viability assay

Cell viability was determined by MTS Assay CellTiter 96 Aqueous One Solution Cell Proliferation Assay (Promega G3582) according to the manufacturer’s instructions.

2.18. Statistical analysis

The data presented in this article were expressed as mean ±standard (SD) at least three times with two-tailed Student’s t-test using GraphPad PRISM. ns: not significant, *P < 0.05 was considered significant, ** P < 0.01 was considered more significant, and *** P < 0.001 was considered highly significant.

3. Results

3.1. FMDV infection activates NLRP3 inflammasome in macrophage

To determine the mechanism underlying activation of inflammation after FMDV infection in macrophages, we performed FMDV infection in mouse macrophage RAW264.7 cells and found that FMDV serotype O infection triggers the NLRP3 inflammasome in macrophages. These highly enriched, differently expressed genes were mainly responsible for the regulation of inflammatory responses using an Agilent Mouse Gene Expression Microarray. NLRP3, IL-1β and NF-κB were up-regulated after FMDV infection (Fig. 1(A)). The NLRP3 inflammasome is activated by a wide range of stress signals that are derived from pathogens and environmental stresses. LPS leads to activation of NF-κB, which triggers ‘signal 1’ of NLRP3 inflammasome [39,40]. Thus, we determined the LPS primed- or unprimed-FMDV on the secretion of IL-1β. LPS primed- or unprimed-RAW264.7 macrophages were infected with FMDV or treated with positive control (LPS+ATP) and negative control (mock). The production of NLRP3 and pro-IL-1β in lysis, and secretion of cleaved IL-1β and Caspase-1 in supernatants were induced by LPS primed- or unprimed-FMDV infection and LPS+ATP treatment in a time-dependent manner (Fig. 1(B,C)). Hence, to determine whether FMDV activates the NLRP3 inflammasome, mouse macrophage RAW264.7, iBMM, and porcine macrophage 3D4/2 cells were infected with FMDV at 2 multiplicity of infection (MOI) for 9 h. As shown in Fig. 1(D), the expression of NLRP3 and pro-IL-1β protein in lysis and the production of IL-1β and caspase-1 in supernatant significantly increased in the FMDV-infected RAW264.7 and iBMM. Accordingly, the expression of FMDV structural protein VP0, 1, 2 and 3 was detected in the lysates, indicating that FMDV replicated well in mouse macrophages. Especially, the IL-1β and Caspase-1 increased significantly in the supernatants of mouse macrophage induced by FMDV (Fig. 1(E)). The secretion of IL-1β was detected in FMDV-infected 3D4/2, indicating that FMDV causes IL-1β secretion in porcine macrophage (Fig. 1(F)). Similarly, the mRNA of NLRP3, pro-IL-1β and FMDV were detected in FMDV-infected RAW264.7, iBMM and 3D4/2 (Figure S1(a)). To further investigate whether the activation of NLRP3 inflammation is positively related to the FMDV infection, RAW264.7 and iBMM macrophages were used to infect the FMDV with different MOIs and were detected at 9 hpi. The protein expression in the cell lysates (Fig. 1(G)) and the secretions of IL-1β and Caspase-1 in the cell supernatants increased in a dose-dependent manner (Fig. 1(H)). The mRNA results of PK-15 and ST cells were similar to those of RAW264.7 and iBMM (Fig. S1(B)). These results suggested that FMDV infection can induce the NLRP3 inflammation activation with dose- and time-dependent manner.

Figure 1.

NLRP3 inflammasome was activated by FMDV. (A) Cluster data showing differential gene expression from mock- and FMDV-infected RAW 264.7 cells at 9 hpi using an Agilent Mouse Gene Expression Microarray. (B,C) 100 ng/mL LPS primed- or unprimed-RAW264.7 macrophages were infected with 2 MOIs FMDV, treated with mock or LPS (100 ng/mL, 3 h)+ATP (5 mM, 30 min) and collected at different time points as indicated. (B) Western blot analysis was used to determine the proteins in the cell lysates (cell) and supernatants (Sup). (C) Supernatants were analysed for mouse IL-1β by ELISA. (D,F) RAW264.7, iBMM and 3D 4/2 cells were infected with 2 MOIs FMDV, mock or LPS (100 ng/mL, 3 h)+ATP (5 mM, 30 min) for 9 h. Mock was used as a negative control and LPS+ATP was used as a positive control. (D) Western blot analysis was used to determine the proteins in the cell lysates (cell) and supernatants (Sup) of RAW264.7 and iBMM.(E) Supernatants were analysed for the mature IL-1β Cand Caspase-1 (p20) of RAW264.7 and iBMM by ELISA (F) Supernatants were analysed for porcine IL-1β (p17) of 3D 4/2 by ELISA. (G,H) RAW264.7 and iBMM were infected with 0.2, 2, and 20 MOIs of FMDV for 9 h. (G) Western blot analysis in the cell lysates (cell) and supernatants (Sup). (H) Supernatants were analysed for IL-1β and Caspase-1 of RAW264.7 and iBMM by ELISA. Data are represented as the mean ± SD of triplicate measurement in three independent experiments. * P < 0.05, ** P < 0.01, *** P < 0.001 vs mock (Student’s t-test).

3.2. Roles of inflammasome complex components on IL-1β secretion following FMDV infection

Previous study has determined that FMDV infection induces IL-1β, activating NLRP3 inflammasome. The NLRP3 inflammasome is a complex, containing NLRP3, ASC and pro-Caspase-1. BMM and BMDC were isolated and cultured from NLRP3^−/-, ASC^−/- and WT mice. A significant increase in IL-1β was detected in the supernatant from wild-type BMM, in addition, the IL-1β secretion decreased in the supernatant from NLRP3^−/- and ASC^−/- BMM (Fig. 2(A)). The mRNA expression of NLRP3 and ASC of BMDC and BMM were significantly down-regulated after the gene of NLRP3^−/- and ASC^−/- knocked-out, respectively (Fig. S2(A,B)). Moreover, FMDV mRNA expression increased in NLRP3^−/- and ASC^−/- BMDC and BMM (Fig. 2(B)). To directly assess the relationship between FMDV and components in inflammasome complex, siRNAs were transfected into RAW264.7 and iBMM macrophages to knockdown the expression levels of NLRP3 (siNLRP3), ASC (siASC), and Caspase-1 (siCapase-1). The activated IL-1β were significantly down-regulated after RNA interference (Fig. 2(C)). Western blot analysis showed that the expression levels of NLRP3, ASC, and Caspase-1 significantly decreased (Fig. 2(D)). FMDV protein and virus titres increased after NLPR3, ASC and Caspase-1 siRNA transfected into RAW264.7, iBMM and PK-15 cells (Fig.2(D–F)). Moreover, FMDV titres decreased after overexpression of NLPR3, ASC and Caspase-1 in PK-15 cells (Fig. 2(F)). These results above showed that FMDV infection activated NLRP3 inflammasome depending on all the components of the NLRP3 inflammasome. When knockdown of NLRP3 inflammasome components, it promoted replication of FMDV. Whereas overexpressing of the NLRP3 inflammasome components, it inhibited replication of FMDV.

Figure 2.

Role of components in NLRP3 inflammasome complex after FMDV infection. (A,B) BMM and BMDC were isolated and cultured from NLRP3^−/-, ASC^−/- and WT mice, and then treated with mock or 2 MOIs of FMDV for 9 h. (A) ELISA analysis for the mature IL-1β of BMM. (B) The mRNA levels of FMDV were quantified by qRT-PCR of BMM and BMDC. (C,E) Sicontrol, siNLRP3, siASC and siCaspase-1 were transfected into RAW264.7 and iBMM for 16 h and then treated with mock or 2 MOIs of FMDV for 9 h. (C) ELISA analysis of the mature IL-1β in the supernatants. (D) Western blot analysis. (E) TCID₅₀ analysis of FMDV titres in the supernatants. (F) Sicontrol, siNLRP3, siCaspase-1, and siASC were transfected into PK-15 for 16 h and then treated with 1 MOI of FMDV for 9 h. Empty vector (EV), pNLRP3, pCaspase-1, and pASC were transfected into PK-15 for 20 h and then treated with 1 MOI of FMDV for 9 h. TCID₅₀ analysis of FMDV titres in the supernatants. (G,I) RAW264.7 cells were incubated with yVAD (5 or 50 µM) for 1 h and then infected with 2 MOIs FMDV for 9 h. (G) Western blot analysis. (H) ELISA analysis of the mature IL-1β and Caspase-1 in the supernatants. (I) TCID₅₀ analysis of FMDV titres in the supernatants. Data are presented as the mean ± SD of triplicate measurement in three independent experiments. ns: not significant, * P < 0.05, ** P < 0.01, *** P < 0.001 vs sicontrol-FMDV or p-FMDV (Student’s t-test).

Z-YVAD-FMK (yVAD) is a specific peptide inhibitor of Caspase-1 that can strongly decrease IL-1β secretion [41]. RAW264.7 cells were incubated with yVAD (5 or 50 µM) for 1 h and then infected with 2 MOIs of FMDV for 9 h. The secretion of Caspase-1 and IL-1β were significantly decreased (Fig. 2(H)). However, the protein expression and mRNA level of pro-Caspase-1 were not reduced (Figs. 2(G) and S2(C)), suggesting that yVAD can effectively inhibit the secretion of Caspase-1 but has no obvious effect on the precursor of Caspase-1. Meanwhile, FMDV protein, virus titres and mRNA level of FMDV were increased significantly, when the activity of caspase-1 is inhibited (Figs. 2(G,I) and S2(D)).

Hence, when the three components of NLRP3 inflammasome were silenced or the Caspase-1 activity was inhibited, the secretion of Caspase-1 and IL-1β decreased significantly. Meanwhile, the NLRP3 inflammasome can suppress FMDV replication and plays a protective role against FMDV infection.

3.3. FMDV RNA activates the NLRP3 inflammasome

To determine which component of the FMDV virion can activate the NLRP3 inflammasome, including the RNA genome or viral proteins, is necessary. The NLRP3 in cell lysates and IL-1β in cell supernatants were detected after transfection with FMDV-infected RNA (V-RNA) (10 µg) (Figs. 3(A) and S3(A)), which was consistent with the results of the mRNA levels of NLRP3 and pro-IL-1β (Fig. S3(B)). These results confirmed that FMDV RNA can trigger IL-1β expression in mouse macrophage. To further demonstrate the pathway in which FMDV RNA triggers NLRP3 inflammasome, we found that p-NF-κB/p65 were significantly activated (Fig. 3(B)). As reported, the activated NF-κB up-regulates the transcription of NLRP3 and pro-IL-1β, which further promotes the synthesis of NLRP3 inflammasome [42]. Hence, FMDV RNA activated the ‘signal 1’ of NLRP3 inflammasome (Fig. 3(C)).

Figure 3.

FMDV RNA activates the NLRP3 inflammasome. (A) RAW264.7 and iBMM cells were treated with mock (control) or infected with 2 MOIs FMDV for 9 h. Total RNA of control-RNA (C-RNA) and FMDV-infected-RNA (V-RNA) were extracted and transfected into RAW264.7 and iBMM cells at various amounts (2.5, 5, and 10 µg) for 24 h. ELISA analysis of IL-1β. (B) RAW264.7 were treated with mock or infected with 2 MOIs FMDV for 9 h. 10 µg FMDV-infected-RNA (FMDV RNA) were extracted and transfected into RAW264.7 for 24 h. Western blot analysis of the proteins of RIG-I and NF-kB pathway. (C) Model of RIG-I and NF-kB signalling pathways. (D,E) Sicontrol, siRIG-I were transfected into RAW264.7 for 16 h and then transfected with control-RNA (Mock) and FMDV-infected-RNA (FMDV RNA) for 24 h into RAW264.7. (D) Supernatants were analysed for IL-1β by ELISA. (E) Supernatants were analysed for IFN-β by ELISA. (F) RAW264.7 and iBMM cells were infected with mock, two MOIs FMDV or 2, 5 MOIs UV–inactivated FMDV (UV-FMDV) for 5 and 9 h. Supernatants were analysed for IL-1β by ELISA. (G) RAW264.7 and iBMM cells were infected with mock, 2 MOIs FMDV or UV-inactivated FMDV (UV-FMDV) for 9 h. Supernatants were analysed by ELISA for Caspase-1. Data are presented as mean ± SD of triplicate measurement in three independent experiments. ns: not significant, * P < 0.05, ** P < 0.01, *** P < 0.001 vs mock (Student’s t-test).

RIG-I, which belongs to the RIG-I like receptor (RLR) family, containing two N-terminal CARDs and pro-caspase-1. On activation, RIG-I inflammasome induces the activated caspase-1 and the mature IL-1β [10,43]. As shown in Fig. 3(B,C), both FMDV and FMDV RNA can be recognized by RIG-I to activate IRF-3 (IFN regulatory factor 3), we tested whether RIG-I was involved in inflammasome activation. Therefore, transfected siRNA of RIG-I into RAW264.7 cells, we confirmed that the knock-down efficiency was significant (Fig. S3(C)). However, when FMDV RNA was transfected into RAW264.7 macrophages, IL-1β secretion was not reduced in the absence of RIG-I (Fig. 3(D)), while the secretion of IFN-β was clearly decreased (Fig. 3(E)). These results indicated that FMDV RNA activated NF-κB signal pathway, up-regulated the levels of mRNA of NLRP3 and pro-IL-1β and induce the IL-1β secretion not dependent on RIG-I inflammasome.

To further confirm this result and simultaneously verify whether the FMDV replication can induce the secretion of IL-1β, UV-inactivated FMDV was used to treat the macrophages RAW264.7 and iBMM, and then the cellular lysates and supernatants were collected at 5 and 9 h. Results demonstrated that UV-inactivated FMDV did not induce FMDV replication (Fig. S3(D)), but triggered robust secretions of IL-1β and caspase-1 in a time- and dose-dependent manner, making the mRNA up-regulated at 9 h (Figs. 3(F,G) and S3(E)).

Hence, the results suggested that NLRP3 inflammasome activation is independent of FMDV viral replication, which can be induced by the FMDV RNA genome. Furthermore, IL-1β secretion was activated by NLRP3 inflammasome, not dependent on RIG-I inflammasome.

3.4. FMDV nonstructural 2B protein activates the NLRP3 inflammasome

To confirm whether viral structural or nonstructural proteins, except for the FMDV RNA, can induce the activation of the NLRP3 inflammasome pathway, the different plasmids were transfected into RAW264.7 cells, the secretion of IL-1β significantly increased in the supernatants of cells only transfected with p2B-flag compared with those in the empty vector control (Fig. 4(A)). This finding suggested that FMDV 2B protein alone can trigger IL-1β expression in mouse macrophage RAW264.7 cells. ASC specks are the indirect indicators for inflammasome activation [44], we detected ASC specks in LPS+ATP, FMDV-infected and 2B-transfected mouse macrophage. Immunofluorescence revealed that endogenous ASC formed specks in the cytoplasm or extracellular space the after activation of NLRP3 inflammasome in RAW264.7 cells (Fig. 4(B)). The p2B-flag was transfected into RAW264.7 or iBMM, induced IL-1β and Caspase-1 secretion levels in the cell supernatant (Figs. 4(C) and S4(A)). The cell viability experiment indicated that FMDV infection and 2B transfection did not cause cell death to RAW264.7 cells. The cell viability rates were close to 100% (Fig. S4(B)). FMDV and 2B can trigger IL-1β release under conditions that do not trigger cell death in RAW264.7. The results indicated that FMDV and 2B activated NLRP3 inflammasome independent of producing pyroptosis, as reported previously [45].

Figure 4.

FMDV nonstructural protein 2B induces the secretion of IL-1β. (A) RAW264.7 cells were transfected for 24 h with different Flag-tagged plasmids encoding FMDV nonstructural, structural protein and empty vector (EV) as indicated. Supernatant was analysed for IL-1β by ELISA. (B) RAW264.7 macrophages were infected with 2 MOIs FMDV, treated with mock or LPS (100 ng/mL, 3 h)+ATP (5 mM, 30 min) for 9 h. RAW264.7 were transfected with 1 µg FMDV p2B-flag for 24 h. Cells were stained with anti-flag (green) or FMDV (green) or ASC (red) and analysed by confocal microscopy. Nuclei were stained with DAPI (blue). (C) RAW264.7 and iBMM were transfected with 1 µg of empty vector (EV) and FMDV p2B-flag for 24 h. ELISA analysis of the mature IL-1β and Caspase-1 in the supernatants. (D,E) BALB/c mice (20–25 g) were randomly divided into four groups. Group 1, pH 7.4 PBS; Group 2, 50 µg FMDV serotype O Virus-like proteins (VLPs) (made in our lab) and 50 µg FMDV 2B emulsified by Montandie ISA 206; Group 3, 50 µg FMDV serotype O VLPs emulsified by Montandie ISA 206; Group 4, 50 µg FMDV 2B emulsified by Montandie ISA 206. The BALB/c mice were injected with 0.2 mL vaccine in tibialis cranialis muscle of both rear legs. (D) In the mouse splenic lymphocytes and bone marrow cells, the mRNA expression level of NLRP3 and pro-IL-1β increased after vaccination with 2B or 2B together with FMD VLPs at 3, 5 and 7 days. (E) ELISA analysis of the mature IL-1β in the sera of the mice at 0, 1, 3, 5 and 7 days. (F) The guinea pigs (300–350 g) were randomly divided into three groups. Group 1, pH 7.4 PBS; Group 2, 50 µg FMDV serotype O VLPs emulsified by Montanide ISA 206; Group 3, 50 µg purified FMDV 2B and VLPs emulsified by Montanide ISA 206. The guinea pigs were injected with 0.2 mL vaccine in the tibialis cranialis muscle of both rear legs. The serum samples were collected from the heart at every 7 days. Guinea pigs were given booster dose at 28 days after the first immunization. All the sera were collected. Specific neutralizing antibodies of guinea pigs were detected. Data are presented as mean ± SD of triplicate measurement in three independent experiments. ns: not significant, * P < 0.05, ** P < 0.01, *** P < 0.001 vs EV or PBS (Student’s t-test).

Aluminium adjuvants can enhance immunity of vaccine by activating the NLRP3 inflammasome [46–51]. Hence, to further testify whether the 2B protein can irritate the NLRP3 inflammasome and exert adjuvant functions as aluminium salt, purified 2B proteins [29] were injected into mice together with FMDV virus-like particles (VLPs) vaccine. In the mouse splenic lymphocytes and bone marrow cells, the mRNA expression level of NLRP3 and pro-IL-1β increased after vaccination with 2B or 2B together with FMD VLPs at 5 and 7 days (Fig. 4(D)). The production of IL-1β in the mouse sera increased and peaked at 5 days (Fig. 4(E)). In addition, FMDV 2B can sufficiently induce the secretion of IL-1β in the mice. The Guinea pig is an ideal animal model for FMD vaccine, which can appear specific symptoms of FMDV infection, such as vesicle on the sole of the foot. Hence, 2B protein together with FMD VLPs further immunized guinea pigs to detect whether the 2B protein can improve the protection against FMDV infection. It showed that 2B protein significantly enhanced the neutralization antibody titre from 14 to 28 days in the first vaccination (Fig. 4(F)). This finding suggested that the 2B protein improved the immunity of FMD VLPs and increase the neutralization antibody response in the early stage of immunization. Moreover, the protection potency of the VLPs or VLPs with 2B reached 75% in guinea pigs challenge experiment (Table 1). Hence, the animal experiment further demonstrated that the FMDV 2B protein enhanced the immunity of the VLPs vaccine by activating NLRP3 inflammasome, especially, in the early stage of vaccination. It showed that 2B not only activated NLRP3 inflammasome and secreted IL-1β, but also acted as an adjuvant without any cytotoxicity to induce immunity protection, producing cytokines and neutralizing antibodies, and established the foundation to the development of self-adjuvants for VLPs vaccines.

Table 1.

Guinea pigs immunization and challenge experiments vaccination and challenge. Animal experimentation was carried out in National Foot and Mouth Disease Reference Laboratory, Lanzhou Veterinary Research Institute, under the guidelines of the Animal Research Ethics Board of Lanzhou Veterinary Research Institute, Chinese Academy of Agricultural Sciences (CAAS), China. The guinea pigs (300–350 g) were randomly divided into three groups. Group 1, pH 7.4 PBS; Group 2, 50 µg FMDV serotype O Virus-like proteins (VLPs) (made in our lab) emulsified by Montanide ISA 206 (Seppic, Paris, France); Group 3, 50 µg FMDV 2B (expressed and purified) +50 µg FMDV serotype O Virus-like particles (VLPs). The guinea pigs were injected with 0.2 mL vaccine in the tibialis cranialis muscle of both rear legs. The serum samples were collected from the heart at every 7 days. Guinea pigs were given booster dose at 28 days after the first immunization. The challenge was performed at 28 days after secondary immunizations. All guinea pigs were challenged with 0.2 mL of guinea pig infectious dose 100 ID₅₀ per guinea pig in back sole of rear foot. After the challenge, the animals were examined daily for 7 days.

Number of guinea pigs	Group 1 (PBS)	Group 2 (VLPs)	Group 3 (VLPs+2B)
Protection
1	None	Partial	Partial
2	None	Total	Total
3	None	Total	Total
4	Partial	Total	Total
Severity of symptoms
1	Severe	Mild	Mild
2	Severe	None	None
3	Severe	None	None
4	Mild	None	None
Rate of protection (%)	0% (0/4)	75% (3/4)	75% (3/4)

Total protection was defined as no lesions on the footpad. Partial protection was defined as only one lesion on the injected foot only. No protection was defined two or more lesions on feet of the challenged animal.

3.5. NLRP3 inflammasome activation by FMDV is dependent on ion efflux

Even though 2B proteins can induce the secretion of IL-1β, the relationship between 2B and NLRP3 inflammasome remains unclear. It is reported that ROS plays an important role in the activation of NLRP3 inflammasome [52]. The levels of ROS peaked at 7 hpi in RAW264.7 and iBMM with 2 MOIs FMDV infection (Fig. 5(A)), suggesting that FMDV can trigger ROS production. Monosodium urate (MSU) and ATP are NLRP3 activators [53,54], and Mito-TEMPO is an antioxidant, inhibiting ROS [33,55–57], as an NLRP3 inhibitor. Similarly, no effect was observed on the secretion of IL-1β in FMDV-infected macrophages treated with Mito-TEMPO but significantly inhibited IL-1β secretion in response to MSU or ATP (Fig. 5(B)). Cathepsin B is a lysosomal protease, activating of NLRP3 inflammasome [58]. Hence, treated with Ca-074-Me, a specific inhibitor of cathepsin B [59], decreased the IL-1β secretion of MSU-induced macrophages, and no effects were observed in the FMDV-infected macrophages (Fig. 5(C)). Above all, the activation of FMDV-induced NLRP3 inflammasome was independent of the ROS and lysosome models.

Figure 5.

Activation of NLRP3 inflammasome independent of mitochondrial ROS and lysosomal cathepsin B, dependent on ion efflux. (A) RAW264.7 and iBMM cells were infected with Mock or 2 MOIs FMDV for 1, 3, 5, 7, 9 h. Then, the cells were incubated with 5 µM MitoSOX for 10 min at 37°C. Results were analysed by flow cytometric analysis with FlowJo software. (B) RAW264.7 and iBMM cells were treated with mock, 2 MOIs FMDV, MSU (2.5 mM) and LPS (100 ng/mL, 3 h)+ATP (5 mM, 30 min) in the presence or absence of Mito-TEMPO (500 µM). The supernatants of cells were collected and analysed by ELISA for IL-1β. (C) RAW264.7 and iBMM cells were treated with mock, 2 MOIs FMDV and MSU (2.5 mM) in presence or absence of Ca-074-Me (10 µM). The supernatants of cells were collected and analysed by ELISA for IL-1β. (D,E) RAW264.7 and iBMM cells were infected with mock or 2 MOIs FMDV for 9 h. Meanwhile, RAW264.7 and iBMM cells were transfected with 1 µg of empty vector (EV) and p2B-flag. Samples were collected at 12, 16, and 20 h. Cells were incubated with Fluo-3 AM for 1 h at 37°C avoiding light. Intracellular Ca²⁺ concentration of RAW264.7 and iBMM cells were analysed by flow cytometry with FlowJo software. (D) FMDV infection. (E) p2B-flag transfection. (F) RAW264.7 and iBMM cells were infected with 2 MOIs FMDV or transfected with empty vector (EV) or 1 µg of p2B-flag. Cells were, respectively, treated with BAPTA-AM at doses of 5, 10, 20, and 40 µM. The supernatants of FMDV-infected or EV or 2B-transfected cells were collected at 9 and 16 h, respectively, then analysed by ELISA assay for IL-1β. (G) RAW264.7 were treated with different doses of KCl or CaCl₂ and infected with 2 MOIs FMDV for 9 h, then analysed by ELISA assay for IL-1β. Data are represented as the mean ± SD of triplicate measurement in three independent experiments. ns: not significant, * P < 0.05, ** P < 0.01, *** P < 0.001 vs EV or Mock or FMDV (Student’s t-test).

To examine whether the ion efflux induced by the FMDV or 2B proteins as viroporin plays a role in NLRP3 inflammasome activation, the level of intracellular Ca²⁺ was first detected. Cells were infected with FMDV or transfected with p2B-flag followed staining with fluorescent dye Fluo-3 AM, as a calcium-dependent fluorescent probe [60–62]. It showed that the levels of intracellular Ca²⁺ concentration significantly increased after FMDV infection and 2B expression (Fig. 5(D,E)). Treated with different doses of BAPTA-AM, a cell-permeant chelator of Ca²⁺ [63], the IL-1β secretion significantly decreased (Fig. 5(F)). The results demonstrated that FMDV and FMDV 2B protein activated NLRP3 inflammasome and released IL-1β through Ca²⁺ efflux. Furthermore, potassium efflux is required for the activation of the NLRP3 inflammasome [64]. The results showed that incubation of FMDV-infected macrophages in media containing gradient concentrations of KCl, to activate NLRP3 inflammasome with low concentrations of K⁺, to suppress NLRP3 inflammasome with high concentrations of K⁺. IL-1β secretion was inhibited when K⁺ efflux was blocked with increased concentrations of extracellular K⁺ (Fig. 5(G)). As shown in Fig. 5(G), incubation with gradient concentrations of CaCl₂, the IL-1β secretion increased with dose-dependent of Ca²⁺. In conclusion, NLRP3 inflammasome was activated by FMDV infection through ions efflux, but independent of mitochondrial ROS and lysosomal cathepsin B.

3.6. FMDV 2B protein activates the NLRP3 inflammasome is dependent on the structure of viroporin

Nonstructural protein 2B of FMDV is viroporin-containing 154 amino acids, which is a pore-forming protein. There are two helical transmembrane regions between residues 83 and 104 and between residues 119 and 137, containing two distinct pore domains [29]. Three possible hydrophobic regions could be part of a transmembrane domain corresponding to amino acids 60 to 78, 84 to 104, and 121 to 141 [31]. To confirm the hydrophobic domains of 2B that were important for forming ion channel, we constructed a series of mutants of FMDV 2B which hydrophobic amino acids form 119 to 145 were changed into hydrophilic amino acid (Fig. 6(A,B)). RAW264.7 cells were transfected with EV (empty vector), p2B-HA and mutants of p2B-HA (M1, M2 and M3). It showed that the levels of intracellular Ca²⁺ concentration and the production of IL-1β significantly decreased after p2B-M3-HA (Fig. 6(C,D)). By mutating the amino acid and changing the hydrophobic properties, the helical transmembrane region of the viroporin 2B was altered, so that the 2B did not form a channel inserted into the cell membrane. Ions did not flow from the channel, moreover the intracellular calcium ion was significantly reduced, and the production of IL-1β decreased significantly. The mutant p2B-M3-HA could not trigger the activation of NLRP3 inflammasome. As viroporin, 2B-activated NLRP3 inflammasome depending on the structure of transmembrane regions.

Figure 6.

FMDV 2B protein activates the NLRP3 inflammasome is dependent on the transmembrane regions of viroporin. (A) The mutants of hydrophobic amino acids were designed. (B) A series of FMDV 2B mutants were constructed and analysis with DNASTAR software. (C,D) RAW264.7 were transfected with 1 µg of empty vector (EV), p2B-flag and mutants of 2B-flag (M1, M2 and M3). Samples were collected at 20 h following treatment with 2.5 µM Fluo-3 AM for 1 h at 37°C. (C) Intracellular Ca²⁺ concentration of RAW264.7 were analysed by flow cytometry with FlowJo software. (D) The supernatants were collected and analysed by ELISA assay for IL-1β. Data are represented as the mean ± SD of triplicate measurement in three independent experiments. ns: not significant, * P < 0.05, ** P < 0.01, *** P < 0.001 vs p2B-HA (Student’s t-test).

4. Discussion

NLRP3 inflammasome is critical for innate immune responses against pathogenic infections. However, the activation of inflammasome by FMDV has not been completely elucidated. In this study, we have demonstrated that FMDV activates NLRP3 inflammasome and induces secretion of IL-1β (Fig. 7). We find that FMDV can stimulate the production of Caspase-1 and IL-1β in mouse and porcine macrophages, mouse BMM and BMDC with a dose- and time-dependent manner. The components of NLRP3 inflammasome complex play a very important role in NLRP3 inflammasome activation with FMDV infection. FMDV RNA induces NF-κB activation, upregulates the transcription of NLRP3 and pro-IL-1β, triggers NLRP3 inflammasome and produces IL-1β expression in macrophage, independent of RIG-I inflammasome. FMDV 2B activates NLRP3 inflammasome through elevation of intracellular ion, independent of ROS and lysosomal cathepsin B. FMDV 2B viroporin activates NLRP3 inflammasome and induces IL-1β dependent of the helical transmembrane region.

Figure 7.

Schematic of FMDV-activated NLRP3 inflammasome. FMDV activates NLRP3 inflammasome and induces secretion of IL-1β. During FMDV infection, FMDV RNA recognizes RIG-I and activates NF-κB, which causes the mRNA of NLRP3 or pro-IL-1β synthesis and initial inflammasome activation at the transcriptional level. FMDV RNA triggers NLRP3 inflammasome and produces IL-1β expression in macrophage. The activation of NLRP3 inflammasome by FMDV and 2B does not pass through the ROS and lysosome models. NLRP3 inflammasome is activated by FMDV 2B protein, though Ca²⁺ and K⁺ flux. FMDV viroporin 2B protein triggers the activation of NLRP3 inflammasome by ion channel.

Our study reveals that FMDV activates NLRP3 inflammasome as triggered by FMDV RNA and 2B protein. The results provide insights into the innate immune mechanism of FMDV infection. More interestingly, we found that knockdown or inhibited the protein expression of NLRP3 inflammasome complex, FMDV protein and virus titres were increased. Moreover, FMDV titres decreased after proteins overexpression, inhibited replication of FMDV. Thus, we reveal that the NLRP3 inflammasome plays a protective role against FMDV infection. Therefore, NLRP3 inflammasome activity may be useful to control FMDV infection, based on the current finding that the NLRP3 inflammasome is protective against EV71 infection [19] and NOD2 inhibits FMDV replication in the infected cells [65].

As reported, two signals that sequentially induce the NLRP3 inflammasome to exist. The signal 1 is priming, corresponding to the activation of NF-κB which is a transcription factor [1]. RIG-I recognizes viral RNA via caspase recruitment domain (CARD)–CARD interactions to activate RIG-I inflammasome pathways [10,43,66]. FMDV RNA is sensed by PRRs like the NLRP3 and RIG-I to induce the production of IN-β and IL-1β. With the RIG-I knockdown, IL-1β secretion was not reduced after FMDV RNA transfection, meaning that FMDV RNA activated NLRP3 inflammasome, induced IL-1β secretion not dependent on RIG-I inflammasome. FMDV RNA activated NF-κB p65 and triggered the activation of NLRP3 inflammasome as the signal 1. The signal 2 is activation. There are three activation models of NLRP3 inflammasome activation. The first one is ROS produced by NADPH oxidase (NOX) in mitochondria is crucial [52]. In our study, using Mito-TEMPO, FMDV infection did not affect the secretion of IL-1β. The second model is the lysosomal rupture involved in NLRP3 inflammasome activation. Hence, a single lysosomal protease, cathepsin B, is found to play an important role in activating NLRP3 inflammasome [59]. Our results showed that FMDV had no effect on IL-1β secretion using Ca-074-Me. Thus, FMDV triggered NLRP3 inflammasome activation independent of ROS and cathepsin B. The third model is ion flux. Some K⁺ efflux, extracellular Ca²⁺ influx, and ER Ca²⁺ release cause mitochondrial dysfunction and triggered NLRP3 activation [59,67]. In our study, FMDV and FMDV 2B-activated Ca²⁺ efflux. In our previous research, we confirmed that FMDV 2B protein is a viroporin that is mainly localized in the ER, and it modifies membrane permeability and disturbs the Ca²⁺ balance in host cells [29]. VSV or EMCV induces activation of the NLRP3 inflammasome-depending K⁺ efflux [68]. Our results showed that NLRP3 inflammasome was activated with low concentrations of K⁺, and suppressed with high concentrations of K⁺. IL-1β secretion was inhibited when K⁺ efflux was blocked with increased concentrations of extracellular K⁺. Subsequently, many reports have demonstrated that viroporins of different viruses play important roles in the regulation of NLRP3 inflammasome activation [33–35]. Consistently, FMDV viroporin 2B protein triggers ‘signal 2ʹ of NLRP3 inflammasome through ion efflux, independent of ROS and lysosome models. Moreover, FMDV and 2B trigger IL-1β release and NLRP3 inflammasome activation not inducing cell death.

FMDV 2B, as a viroporin, forms pores on the cellular membrane to induce ion efflux, which finally induced NLRP3 inflammasome activation as the second signal. The animal experiments confirmed that 2B protein can be used as ‘autologous’ adjuvants to enhance the immune response of FMD VLPs vaccine or other subunit protein vaccines without boost vaccination. 2B can activate NLRP3 inflammasome and act as an ideal self-adjuvant for FMD VLPs vaccine in guinea pigs. Thus, our study broadens understanding of the mechanism of the 2B-activated NLRP3 inflammasome and enhances the host’s ability to eliminate viruses through the innate immune response pathway, providing new ideas for the preparation of novel vaccines.

There are two helical transmembrane regions between residues 83 and 104 and between residues 119 and 137, containing two distinct pore domains [29]. Three possible hydrophobic regions could be part of a transmembrane domain corresponding to amino acids 60 to 78, 84 to 104, and 121 to 141 [31]. Helical transmembrane regions were mutanted to change hydrophobic amino acids, to confirm the hydrophobic domains of 2B that were important for forming ion channel. The regions of 140–145 amino acids of 2B (p2B-M3-HA) were changed and transfected into macrophage. The intracellular calcium ion was significantly reduced, and the production of IL-1β decreased significantly. The regions of 140–145 amino acids of 2B are very important to form a channel inserted into the cell membrane. As viroporin, 2B induces NLRP3 inflammasome activation depending on 140–145 amino acids of transmembrane regions.

In conclusion, we have demonstrated that FMDV triggers NLRP3 inflammasome activation. FMDV RNA activates the signal 1 of NLRP3 inflammasome depending on NF-κB activation, and FMDV 2B viroporin activates the signal 2, which is dependent on the ion channel. NLRP3 inflammasome can suppress FMDV replication and play a protective role against FMDV infection. Further, 2B protein activates NLRP3 inflammasome in mouse, also improves specific immune response against FMD VLPs vaccine, which provides clues for effective interventions and new therapeutic approaches.

Funding Statement

This work was supported by grants from the National Natural Science Foundation of China [31672592, 31811540395], The National Key Research and Development Program [2017YFD0500900, 2017YFD0501100, and 2016YFE0204100], Central Public-interest Scientific Institution Basal Research Fund [1610312018003 and 1610312016002].

Acknowledgments

We thank the Dr. Mingzhou Chen from Wuhan University and Dr. Xin Cao from Jilin Agricultural University for advice and critical reading of the manuscript.

Authors’ contributions

GH devised the project and designed the experiments. ZX carried out the experimental work. ZY, SS and ZZ carried out some Western blot and all of ImageJ analysis. DH and LX purified protein and performed the animal experiment. WY performed Real-time PCR and ELISA. LZ, DY, JZ, WC and SH supplied some experimental materials. ZX, GH and WR prepared the manuscript. All authors contributed to and approved the final manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplementary data for this article can be accessed here.