The H240R Protein of African Swine Fever Virus Inhibits Interleukin 1β Production by Inhibiting NEMO Expression and NLRP3 Oligomerization

ABSTRACT The H240R protein (pH240R), encoded by the H240R gene of African swine fever virus (ASFV), is a 241-amino-acid capsid protein. We previously showed that the deletion of H240R from the ASFV genome, creating ASFV-ΔH240R, resulted in an approximately 2-log decrease in infectious virus production compared with the wild-type ASFV strain (ASFV-WT), and ASFV-ΔH240R induced higher interleukin 1β (IL-1β) production in porcine alveolar macrophages (PAMs) than did ASFV-WT, but the underlying mechanism remains to be elucidated. Here, we demonstrate that the activation of the NF-κB signaling and NLRP3 inflammasome was markedly induced in PAMs upon ASFV-ΔH240R infection compared with ASFV-WT. Moreover, pH240R inhibited NF-κB activation by interacting with NEMO and promoting the autophagy-mediated lysosomal degradation of NEMO, resulting in reduced pro-IL-1β transcription. Strikingly, NLRP3 deficiency in PAMs inhibited the ASFV-ΔH240R-induced IL-1β secretion and caspase 1 activation, indicating an essential role of NLRP3 inflammasome activation during ASFV-ΔH240R replication. Mechanistically, pH240R interacted with NLRP3 to inhibit its oligomerization, leading to decreased IL-1β production. Furthermore, the inhibition of the NF-κB signaling and NLRP3 inflammasome activation promoted ASFV-ΔH240R replication in PAMs. Taken together, the results of this study reveal an antagonistic mechanism by which pH240R suppresses the host immune response by manipulating activation of the NF-κB signaling and NLRP3 inflammasome, which might guide the rational design of live attenuated vaccines or therapeutic strategies against ASF in the future. IMPORTANCE African swine fever (ASF), a lethal hemorrhagic disease, is caused by African swine fever virus (ASFV). There are no commercially available vaccines or antivirals for the disease. Here, we showed that ASFV with a deletion of the H240R gene exhibits high-level expression of interleukin 1β (IL-1β), a proinflammatory cytokine, in porcine alveolar macrophages and that the H240R protein (pH240R) exhibits robust inhibitory effects on IL-1β transcription and production. More specifically, pH240R inhibited NF-κB activation via the autophagy-mediated lysosomal degradation of NEMO, leading to the decrease of pro-IL-1β transcription. In addition, pH240R interacted with NLRP3 to inhibit its oligomerization, leading to decreased IL-1β production. Our results indicate that pH240R is involved in the evasion of host innate immunity and provide a novel target for the development of a live attenuated vaccine against ASF.

A frican swine fever (ASF), a lethal hemorrhagic disease, is caused by African swine fever virus (ASFV); there are no commercially available vaccines or antiviral strategies for this disease (1)(2)(3). ASFV is a member of the family Asfarviridae and the only known DNA arbovirus. ASFV particles are 260 to 300 nm in diameter and composed of complex multilayered structures with an overall icosahedral morphology (4,5). ASFV is composed of five distinct structures, namely, an outer icosahedral protein capsid containing an inner lipid envelope, a core shell, a double-stranded DNA genome, and an external lipid envelope. The viral genome is a double-stranded DNA that is 170 to 194 kb in length and encodes more than 150 viral proteins, which are involved in viral replication, virus-host interactions, and modulation of the host innate immune responses (6). A previous study identified 68 different structural proteins in the virion by mass spectrometry analysis (7), and these proteins included multiple components associated with the virus structure as well as a set of enzymes involved in viral mRNA transcription or modification and the maintenance of viral genome integrity. Although the functions of several ASFV genes have been elucidated, large gaps remain in the knowledge about the functions of viral structural proteins, which hinders vaccine and antiviral strategy development.
H240R encodes a structural protein that is relatively abundant in the virion (7). Our previous study showed that the H240R protein (pH240R) is a capsid protein, and it is not absolutely essential for virus replication in vitro. However, the infectious-virus titers from porcine alveolar macrophages (PAMs) infected with a recombinant ASFV with H240R deleted (ASFV-DH240R) are approximately 100-fold lower than those from the cells with the wild-type virus (ASFV-WT) (8). Unexpectedly, higher transcription and production of the proinflammatory cytokine interleukin 1b (IL-1b) were observed in PAMs infected with ASFV-DH240R than ASFV-WT (8).
IL-1b, which is primarily produced by macrophages, is a key mediator of innate immune responses to various invading pathogens (9,10), such as adenovirus (11), Zika virus (12) and respiratory syncytial virus (13). Importantly, the production of biologically active IL-1b is mediated by two distinct innate immune signaling cascades. The first signaling cascade involves transcriptional activation of nuclear factor k B (NF-k B), leading to the synthesis of mRNAs that encode pattern recognition receptors (PRRs) (14)(15)(16). The second signaling cascade includes PRR-induced assembly of a multiprotein complex called the inflammasome and the maturation of caspase 1, which leads to the cleavage of pro-IL-1b into the mature IL-1b (17). Secretion of IL-1b from cells infected with pathogens is closely regulated by inflammasomes. An inflammasome typically consists of a sensor protein, the adaptor molecule apoptosis-associated speck-like molecule containing a caspase recruitment domain (ASC), and the zymogen procaspase 1 (18,19). Assembly of the inflammasome is triggered by the recognition of a variety of stimuli by sensor proteins. The activated sensor protein recruits procaspase 1 by ASC, leading to the autocleavage of caspase 1 to generate an active caspase 1-p10/p20 tetramer. Then, the activated caspase 1 cleaves inactive pro-IL-1b, resulting in IL-1b production. Ultimately, the cytokines processed by the inflammasome contribute to the antiviral response by establishing a proinflammatory state in tissues, which in turn facilitates the recruitment and activation of immune cells (20,21), the establishment of a cellular antiviral state (22), and the induction of pyroptotic cell death (23).
IL-1b plays an important role in the control of infection by various pathogens, as described above. However, pathogens antagonize the host antiviral responses to replicate and exert pathogenic effects in vivo (9,10). These observations suggest that pathogens have evolved several strategies to evade the protective innate immune responses regulated by proinflammatory cytokines. Previous studies have shown that several viruses counteract NF-k B and inflammasome activation (12,(24)(25)(26)(27)(28)(29)(30). However, the correlation between the NFk B signaling and/or inflammasome activation and viral replication remains unclear in vitro.
ASFV infection can induce low-level IL-1b and tumor necrosis factor alpha (TNF-a) in pigs as early as 7 days postinfection (31). The ASFV L83L protein specifically interacts with IL-1b, and the ASFV mutant with L83L deleted is not attenuated in pigs (32). pA238L inhibits the expression of IL-1b by inhibiting the activation of NF-k B, and the mutant ASFV with A238L deleted remains virulent in pigs (33). The ASFV pMGF505-7R specifically interacts with the IKK complex to inhibit pro-IL-1b transcription, and a mutant ASFV with MGF505-7R deleted exhibits attenuated phenotype in pigs and induces high-level IL-1b in vitro and in vivo (34). We have recently shown that pH240R is involved in the ASFV replication cycle, and the H240R-deleted ASFV induced greater IL-1b response than did ASFV-WT in PAMs. Moreover, it has not been elucidated that the relationship between the phenotype with reduced ASFV-DH240R replication and IL-1b production in ASFV-DH240R infection. We speculate that pH240R may inhibit the NF-k B signaling and/or inflammasome activation to regulate downstream inflammatory responses. Therefore, we investigated the effects of pH240R on IL-1b production in PAMs during viral infection.
Here, we demonstrated that the NF-k B signaling was activated in the ASFV-DH240R-infected PAMs by measuring the transcription of pro-IL-1b and the phosphorylation of Ik Ba and p65. Mechanistically, pH240R interacted with the adaptor protein NEMO of the IKK complex. Importantly, pH240R inhibited NF-k B activation by degrading NEMO, leading to the decrease of pro-IL-1b transcription. Intriguingly, ASFV-DH240R infection induced NLRP3 inflammasome activation to promote IL-1b secretion from PAMs. Moreover, pH240R interacted with NLRP3 to inhibit its oligomerization, leading to decreased IL-1b production. Altogether, our study reveals that pH240R counteracts the first and second signaling pathways of IL-1b production, highlighting the importance of IL-1b in the host antiviral immune responses against ASFV infection.

RESULTS
The H240R gene-deleted ASFV activates the NF-jB signaling pathway in PAMs. Ik Ba acts as a regulator of the NF-k B signaling pathway, and the phosphorylation of Ik Ba and p65 is the indicator of NF-k B activation (35). To determine whether the NF-k B signaling pathway is activated in the ASFV-DH240R-infected PAMs, the transcription of pro-IL-1b was analyzed by using reverse transcription-quantitative PCR (RT-qPCR), and the phosphorylation levels of Ik Ba and p65 were measured by using Western blotting at 12 and 24 h postinfection (hpi). The results showed that the transcription levels of pro-IL-1b were increased by approximately 100-and 600-fold at 12 and 24 hpi, respectively, in the PAMs infected with ASFV-DH240R compared with those with ASFV-WT (Fig. 1A). The phosphorylation levels of Ik Ba and p65 in ASFV-DH240R-infected PAMs were increased by 1.88-and 1.45-fold at 12 h (Fig. 1B, left) and by 3.42-and 2.20-fold at 24 h (Fig. 1B, right) compared with those in the ASFV-WT-infected cells. These results indicate that the NF-k B signaling is activated upon ASFV-DH240R infection.
To determine if the NF-k B signaling pathway is specifically activated by ASFV-DH240R infection, BAY11-7082, an inhibitor of the NF-k B signaling pathway, was used to treat PAMs, and the transcription levels of pro-IL-1b and the phosphorylation levels of Ik Ba and p65 were analyzed at 12 and 24 hpi as described above. The results showed that the transcription levels of pro-IL-1b in the BAY11-7082-treated PAMs were decreased by 2.75-and 2.65-fold at 12 and 24 h, respectively, compared with those in the medium-treated PAMs upon ASFV-DH240R infection (Fig. 1C). The phosphorylation levels of Ik Ba and p65 in the inhibitor-treated PAMs were decreased by 1.78-and 1.67fold at 12 h and by 2.66-and 2.48-fold at 24 h, compared with those in the mediumtreated PAMs with ASFV-DH240R infection (Fig. 1D). However, there were no significant differences in the transcription of pro-IL-1b and the phosphorylation levels of Ik Ba and p65 between the inhibitor-and the medium-treated PAMs infected with ASFV-WT. The data suggest that the activation of NF-k B induced by ASFV-DH240R infection is suppressed by the inhibitor of the NF-k B pathway.
Taken together, our results demonstrate that ASFV-DH240R infection specifically activates the NF-k B signaling to facilitate pro-IL-1b transcription.
pH240R inhibits the activation of the NF-jB signaling pathway. The transcription and expression of pro-IL-1b are regulated by the NF-k B signaling (36). To investigate the effects of pH240R on the NF-k B promoter activation, HEK293T cells were cotransfected with the NF-k B reporter and different amounts of the H240R-expressing plasmid, in combination with cGAS-and STING-expressing plasmids or stimulation with FIG 1 ASFV-DH240R infection activates the NF-k B signaling to promote pro-IL-1b transcription. (A) pro-IL-1b transcription was upregulated in PAMs infected with ASFV-DH240R compared with ASFV-WT. PAMs were infected with equal genome copies of ASFV-DH240R and ASFV-WT for 12 h or 24 h, and pro-IL-1b transcription was examined by RT-qPCR using primers against pro-IL-1b and GAPDH. (B) The phosphorylation levels of Ik Ba and p65 were increased in the PAMs infected with ASFV-DH240R compared with ASFV-WT. PAMs were infected with equal genome copies of ASFV-DH240R and ASFV-WT for 12 or 24 h, and the phosphorylation levels of Ik Ba and p65 were analyzed by Western blotting. The protein band intensities were analyzed by Image Studio software. (C) pro-IL-1b transcription was decreased in the inhibitor-treated PAMs induced by ASFV-DH240R infection. BAY11-7082, an inhibitor of the NF-k B signaling pathway, was added to the PAMs infected with ASFV-DH240R or ASFV-WT, and the transcription of pro-IL-1b was analyzed by RT-qPCR at 12 or 24 hpi. (D) The phosphorylation levels of Ik Ba and p65 were decreased in the inhibitortreated PAMs induced by ASFV-DH240R infection. BAY11-7082 was added to the PAMs infected with ASFV-DH240R or ASFV-WT, and the phosphorylation levels of Ik Ba and p65 were analyzed by Western blotting at 12 or 24 hpi. Data are from three independent experiments. The error bars denote standard errors of the means. The results of densitometry analysis to quantify the ratio of p-p65 or p-Ik Ba to GAPDH are shown at the bottom. The significance of differences between groups (n = 3) was determined using Student's t test (***, P , 0.001; ns, not significant). lipopolysaccharides (LPS), and then the cells were lysed to examine the luciferase signals. As shown in Fig. 2A and B, ectopic expression of pH240R inhibited the cGAS/ STING-induced activation of the NF-k B reporter but not the IRF3 reporter in a dosedependent manner. In addition, pH240R inhibited the activation of the NF-k B reporter HEK293T cells were cotransfected with pFlag-cGAS, pFlag-STING, pNF-k B-FLuc, or pIRF3-FLuc (0.1 mg) and pTK-RLuc (0.01 mg) in combination with pCAGGS-Flag or different amounts of pFlag-H240R, and the luciferase activity was measured at 24 hpt. The expression of each protein was measured by Western blotting. (C) pH240R inhibited the NF-k B promoter activation induced by LPS stimulation. HEK293T cells were cotransfected with pCAGGS-Flag and different amounts of pFlag-H240R for 24 h, followed by treatment with or without LPS (100 mM) for another 12 h, and the luciferase activity was determined. The expression of pH240R and GAPDH were measured by Western blotting. Data are from three independent experiments. The error bars denote standard errors of the means. The significance of the difference between groups (n = 3) was determined using Student's t test (**, P , 0.01; ***, P , 0.001; ns, not significant). in the LPS-stimulated HEK293T cells compared with the mock-treated HEK293T cells (Fig.  2C). These results suggest that pH240R efficiently inhibits the activation of the NF-k B signaling.
pH240R inhibits proinflammatory cytokine expression. TNF-aor LPS-induced NF-k B activation facilitates the expression of various proinflammatory cytokines, including IL-1b, IL-6, and TNF-a (37). We have shown that pH240R inhibited the activation of the NF-k B signaling; thus, we examined whether pH240R inhibited the expression of these proinflammatory cytokines. The effects of pH240R on the TNF-aand LPS-induced transcription of IL-1b, IL-6, and TNF-a were evaluated in HEK293T cells and PAMs. Stimulation with TNF-a efficiently induced the transcription of IL-1b, IL-6, and TNF-a, while overexpression of pH240R significantly inhibited the TNF-a-induced FIG 3 The H240R protein (pH240R) inhibits TNF-a-or LPS-induced proinflammatory cytokine expression. (A and B) pH240R inhibited TNF-a-or LPS-induced proinflammatory cytokine expression. HEK293T cells were transfected with pCAGGS-Flag or pFlag-H240R for 24 h, followed by treatment with or without TNF-a (100 mg/mL) (A) or LPS (100 mM) (B) for another 10 h, and the transcription levels of IL-1b, IL-6, and TNF-a were measured by RT-qPCR. (C and D) pH240R inhibited TNF-a-or LPS-induced proinflammatory cytokine expression. PAMs were transfected with pCAGGS-Flag or pFlag-H240R for 24 h, followed by treatment with or without TNF-a (100 mg/mL) (A) or LPS (100 mM) (B) for another 10 h, and the transcription levels of IL-1b, IL-6, and TNF-a were measured by RT-qPCR. Data are from three independent experiments. The error bars denote standard errors of the means. The significance of the difference between groups (n = 3) was determined using Student's t test (*, P , 0.05; **, P , 0.01; ***, P , 0.001). expression of these proinflammatory cytokines (Fig. 3A). As expected, LPS stimulation efficiently induced the expression of IL-1b, IL-6, and TNF-a, and overexpression of pH240R significantly inhibited the LPS-induced expression of these proinflammatory cytokines (Fig. 3B). Similarly, pH240R significantly suppressed the expression of proinflammatory cytokines in PAMs ( Fig. 3C and D). These data suggest that pH240R significantly inhibits the expression of proinflammatory cytokines.
pH240R interacts with the NEMO of the IKK complex. To further investigate the mechanisms underlying the pH240R-mediated inhibition of NF-k B activation, we first determined whether pH240R is associated with various adaptor proteins (STING, TBK1, IKKa, IKKb, or NEMO) that are involved in the activation of the NF-k B promoter. The IKK complex is composed of IKKa, IKKb, and the regulatory ubiquitin-binding protein NEMO, which regulates the activity of the kinase subunits IKKa and IKKb. The data showed that pH240R markedly inhibited the activation of the NF-k B promoter induced by STING, TBK1, IKKa, IKKb, or NEMO ( Fig. 4A to E) but not by p65 (Fig. 4F), suggesting that pH240R might negatively regulate NF-k B activation by targeting the IKK complex.
To identify the adaptor proteins that interact with pH240R to facilitate its inhibition of NF-k B activation, cGAS, STING, TBK1, IKKa, IKKb, and NEMO were selected, and the interactions of these proteins with pH240R were validated by coimmunoprecipitation (co-IP) assay. The results showed that the Flag-tagged cGAS, STING, or TBK1 was not coprecipitated with the Myc-tagged pH240R (data not shown), while the Myc-tagged NEMO, but not the Myc-tagged IKKa or IKKb, was coprecipitated with the Flag-tagged pH240R (Fig. 4G). The Flag-tagged pH240R was coprecipitated with the Myc-tagged NEMO upon incubation with anti-Flag monoclonal antibody (MAb) and protein G-agarose (Fig. 4H). To determine whether pH240R interacts with the NEMO in the context of ASFV infection, virus-infected cell lysates were immunoprecipitated with rabbit anti-pH240R polyclonal antibody (PAb) and probed for the presence of NEMO. As a result, NEMO could be detected in the ASFVinfected cells (Fig. 4I). We also investigated whether pH240R colocalizes with NEMO in HEK293T cells cotransfected with the H240R-and NEMO-expressing plasmids. Consistent with the above results, the colocalization of pH240R with NEMO, but not with IKKa or IKKb, was observed in the cytoplasm of HEK293T cells with a Pearson's correlation coefficient of 0.93 by laser confocal microscopy (Fig. 4J). These data suggest that pH240R inhibits NF-k B activation by interacting with NEMO.
pH240R antagonizes NF-jB activation by promoting the autophagy-mediated lysosomal degradation of NEMO. To investigate the effects of pH240R on the expression of NEMO, which is a polyubiquitin-binding protein (38), pMyc-IKKa, pMyc-IKKb, or pMyc-NEMO and increasing amounts of pFlag-H240R were cotransfected into HEK293T cells. The results showed that pH240R did not affect the transcription of IKKa, IKKb, or NEMO ( Fig. 5A to C). However, pH240R reduced the expression of NEMO, but not that of IKKa or IKKb, in a dose-dependent manner ( Fig. 5D to F). The expression of NEMO was lower in the ASFV-infected PAMs than that in the ASFV-DH240R-infected PAMs (Fig. 5G). There are two main pathways for protein degradation, i.e., the proteasome pathway and lysosomal proteolysis. To determine the pathways involved in the NEMO degradation by pH240R, the proteasome inhibitor MG132 and the lysosome inhibitors chloroquine and bafilomycin A1 (BafA1) were used to treat the cells. The results showed that the pH240R-induced NEMO degradation was rescued by chloroquine or BafA1 but not by MG132 (Fig. 5H), indicating that pH240R downregulates NEMO levels via autophagy-mediated lysosomal degradation. Ik Ba is a regulator of NF-k B activation and binds tightly to NF-k B dimers (39). Phosphorylation by the IKK complex triggers Ik Ba ubiquitination and degradation. To investigate whether pH240R functions as an inhibitor of the degradation of Ik Ba, the increasing amounts of pFlag-H240R were transfected into HEK293T cells for 24 h, and then stimulated with TNF-a to analyze the transcription levels of endogenous Ik Ba. The results showed that pH240R reduced the transcription of endogenous Ik Ba in a dose-dependent manner (Fig. 5I). pEGFP-Ik Ba and increasing amounts of the H240R-expressing plasmid were cotransfected into HEK293T cells, and the expression of Ik Ba was analyzed. The results showed that pH240R stabilized the expression of Ik Ba in a dose-dependent manner (Fig. 5J). Taken together, our results demonstrate that pH240R inhibits NF-k B activation by degrading NEMO, leading to the decrease of pro-IL-1b transcription.
ASFV-DH240R activates the NLRP3 inflammasome to induce IL-1b secretion in PAMs. The inflammasome plays a key role in host innate immune responses by promoting procaspase 1 cleavage to generate the activated subunits p20 and p10, ultimately leading to the maturation and secretion of IL-1b. To determine whether ASFV-DH240R infection activates the inflammasome, the protein levels of IL-1b and caspase 1 in the PAMs infected with ASFV-DH240R or ASFV-WT were measured. As shown by enzymelinked immunosorbent assay (ELISA), the concentrations of IL-1b in the cell supernatants and the levels of caspase 1 in the cell lysates of the ASFV-DH240R-infected PAMs were increased by 3.0-fold (Fig. 6A) and 10.0-fold (Fig. 6B), respectively, compared with those of the ASFV-WT-infected PAMs. Collectively, ASFV-DH240R infection induced the activation of caspase 1 and the release of IL-1b from the ASFV-DH240R-infected PAMs, suggesting that the inflammasome is activated upon ASFV-DH240R infection.
NLRP3 and AIM2 are known to play important roles in virus-induced inflammasome activation (40), but the AIM2 sensor and AIM2-like proteins are absent in porcine cells. We hypothesize that the ASFV-H240R-induced inflammasome activation may depend on NLRP3. Then, NLRP3 expression in PAMs was knocked down with three different small interfering RNAs (siRNAs) targeting the NLRP3 gene (siNLRP3), and the NLRP3 mRNA levels were determined by RT-qPCR. As shown in Fig. 6C, the knockdown efficiency of siNLRP3-1 was the most significant. Transfection of the three siRNAs at the tested concentrations resulted in no notable cytotoxicity at various time points (Fig. 6D). To further confirm whether the inflammasome activation induced by ASFV-DH240R infection depends on NLRP3, the levels of IL-1b and caspase 1 in wild-type PAMs or the PAMs with NLRP3 knockdown infected with ASFV-DH240R or ASFV-WT were measured. As shown in Fig. 6E and F, the concentrations of IL-1b in the cell supernatants and the protein levels of caspase 1 in the cell lysates of the ASFV-DH240R-infected PAMs with NLRP3 knockdown were decreased by 2.5-and 3.1-fold, respectively, compared with those of the ASFV-DH240R-infected PAMs with nontargeting siRNA (siNC) transfection. However, there were no significant differences in the protein levels of IL-1b and caspase 1 between the siNC-and siNLRP3-treated PAMs infected with ASFV-WT. The data show that IL-1b maturation as well as caspase 1 cleavage is abrogated in the NLRP3 knockdown PAMs, suggesting that inflammasome activation induced by ASFV-H240R infection depends on NLRP3.
Taken together, these results show that ASFV-DH240R infection specifically activates the NLRP3 inflammasome to promote the secretion of IL-1b.
pH240R inhibits IL-1b maturation. To evaluate whether pH240R inhibits NLRP3-dependent inflammasome activation, the effects of pH240R on NLRP3 inflammasome activation in HEK293T cells with a Gaussia luciferase (GLuc)-based inflammasome reconstruction assay were investigated, and the luciferase activities were measured. As shown in Fig. 7A, transfection with the H240R-expressing plasmid significantly inhibited the NLRP3-dependent GLuc activation in a dose-dependent manner. Ectopic expression of pH240R significantly inhibited IL-1b maturation, but not NLRP3, ASC, or procaspase 1 expression, as demonstrated by NLRP3 inflammasome reconstitution assay (Fig. 7B). These results indicate that pH240R antagonizes the NLRP3-dependent inflammasome activation to inhibit IL-1b maturation. HEK293T cells were transfected with pCMV-Myc or pFlag-H240R alone and cotransfected with pMyc-IKKa, pMyc-IKKb, or pMyc-NEMO in combination with pFlag-H240R. IKKa, IKKb, NEMO, and pH240R and were then analyzed by laser confocal microscopy. Bars, 10 mm. Data are from three independent experiments. The error bars denote standard errors of the means. The significance of the difference between groups (n = 3) was determined using Student's t test (**, P , 0.01; ***, P , 0.001; ns, not significant). ASC forms speck-like structures, which are known as pyroptosomes, upon inflammasome activation (41). It has been reported that ASC forms specks when it is ectopically expressed in cells treated with nigericin and LPS, and ASC speck formation correlates with IL-1b processing (42). To investigate whether pH240R inhibits ASC speck formation, HEK293T cells were cotransfected with the ASC-expressing plasmid together with H240R or control, and then the cells were treated with nigericin and LPS and analyzed by laser confocal The production of IL-1b or caspase 1 was higher in PAMs infected with ASFV-DH240R than ASFV-WT. PAMs were infected with equal genome copies of ASFV-DH240R and ASFV-WT for 24 h. The level of IL-1b in cell supernatants or caspase 1 in cell lysates was measured by ELISA. (C) Validation of knockdown efficiency of siNLRP3 in PAMs. PAMs were transfected with three different siRNAs targeting NLRP3 (siNLRP3-1, siNLRP3-2, and siNLRP3-3), and the mRNA levels of NLRP3 were measured by RT-qPCR. (D) Cell viability determined with a CCK-8 counting kit. (E and F) The production of IL-1b or caspase 1 in PAMs with NLRP3 knockdown was decreased compared with that in wild-type PAMs upon ASFV-DH240R infection. PAMs were transfected with siNLRP3-1 for 24 h and infected with equal genome copies of ASFV-DH240R and ASFV-WT for another 24 h. The level of IL-1b in cell supernatants or caspase 1 in cell lysates was measured by ELISA. Data are from three independent experiments. The error bars denote standard errors of the means. The significance of the differences between groups (n = 3) was determined using Student's t test (*, P , 0.05; **, P , 0.01; ***, P , 0.001; ns, not significant). microscopy. ASC expression in the presence of nigericin and LPS treatment induced speck formation in most ASC-expressing cells; in contrast, pH240R significantly impaired speck formation (Fig. 7C). The results indicate that pH240R inhibits the NLRP3-dependent inflammasome activation prior to or during ASC speck formation.
pH240R interacts with NLRP3 and inhibits its oligomerization. To elucidate the mechanism by which pH240R inhibits the NLRP3-dependent inflammasome activation, we clarified the protein of inflammasome component that interacting with pH240R by glutathione S-transferase (GST) pulldown assay. As shown in Fig. 8A, the GST-tagged pH240R (GST-pH240R) protein pulled down the Flag-tagged NLRP3 (Flag-NLRP3), but not Flag-ASC or Flag-procaspase 1. Consistent with the results, colocalization of pH240R and NLRP3 was clearly observed in the cytoplasm with a Pearson's correlation coefficient of 0.89 (Fig. 8B). Together, these results suggest that pH240R interacts with the NLRP3 protein. Moreover, to investigate whether pH240R affects the transcription or expression of NLRP3, pFlag-NLRP3 and increasing amounts of pMyc-H240R were cotransfected into HEK293T cells. The results showed that neither the transcription nor the translation of NLRP3 was affected by  (Continued on next page) ASFV H240R Protein Inhibits IL-1b Production Journal of Virology pH240R ( Fig. 8C and D). The expression levels of endogenous NLRP3 were not affected by pH240R during ASFV or ASFV-DH240R infection ( Fig. 8E and F). We have shown that pH240R inhibits the NLRP3-dependent inflammasome activation prior to or during ASC speck formation; thus, we hypothesized that pH240R may inhibit NLRP3 oligomerization and/or ASC recruitment to NLRP3 oligomers. To determine whether pH240R inhibits the oligomerization of NLRP3, HEK293T cells were cotransfected with the plasmids pFlag-NLRP3, pHA-ASC, and pMyc-H240R to analyze oligomerization as described previously (34). As shown in Fig. 8G, the oligomerization levels of NLRP3 were inhibited by pH240R. Next, HEK293T cells were cotransfected with the plasmids harboring four truncated mutants of NLRP3 (Fig. 8H), and the H240R-expressing plasmid to analyze the NLRP3 domains responsible for the interaction with pH240R by co-IP assay. The results showed that pH240R was coprecipitated with the Myc-tagged NLRP3-D1(743-1036) (Fig. 8I). As shown in Fig. 8J, most of the transfected cells displayed oligomers in the HEK293T cells cotransfected with the NLRP3-D1(743-1036)-expressing plasmid in combination with an empty expression plasmid, while the ratio of the oligomer forms was decreased approximately 40% in the HEK293T cells cotransfected with the NLRP3-D1(743-1036)-expressing plasmid combined with the H240R-expressing plasmid. However, no oligomerization occurred in the cells expressing NLRP3-D2, NLRP3-D3, or NLRP3-D4, and these proteins were dispersed throughout the cytoplasm of the transfected cells.
Taken together, these results indicate that pH240R interacts with NLRP3 and inhibits its oligomerization, thereby inhibiting NLRP3 inflammasome activation.
Inhibition of NF-jB signaling and NLRP3 inflammasome activation promotes ASFV-DH240R replication. In response to viral infection, macrophages secrete large amounts of cytokines, including IL-1b, to inhibit the replication of viruses. However, the production of many proinflammatory cytokines in response to viral infection is regulated principally at the transcriptional or translational level. Our results described above showed that the viral growth of ASFV-DH240R was significantly decreased in PAMs and that NF-k B signaling and NLRP3 inflammasome activation were induced upon ASFV-DH240R infection ( Fig. 1 and 6).
To investigate the effects of NF-k B activation on the replication of ASFV-DH240R in PAMs, BAY11-7082, an inhibitor of the NF-k B signaling pathway, was added to PAMs, and the viral genome copies and viral titers in the PAMs infected with ASFV-DH240R or ASFV-WT were determined. As shown in Fig. 9A and B, the viral genome copies and the viral titers of ASFV-DH240R in the inhibitor-treated PAMs were increased by 25.1and 9.1-fold, respectively, compared with those in the mock-treated PAMs (P , 0.01), but there were no significant differences between the medium-and inhibitor-treated PAMs upon ASFV-WT infection, indicating that inhibition of the NF-k B signaling significantly promotes the replication of ASFV-DH240R in PAMs.
To further investigate the effects of NLRP3 knockdown on the replication of ASFV-DH240R in PAMs, the viral genome copies and viral titers were measured in the siNLRP3-or siNC-transfected PAMs infected with ASFV-DH240R or ASFV-WT. The results showed that the viral genome copies and the viral titers of ASFV-DH240R in NLRP3 knockdown PAMs were significantly increased by 10.9-and 4.7-fold, respectively, compared with those in the siNC-treated PAMs, but there were no significant differences between the siNC-and siNLRP3-transfected PAMs ( Fig. 9C and D), indicating that inhibition of NLRP3 inflammasome activation significantly promotes the replication of ASFV-DH240R in PAMs.
Collectively, the data show that the activated NF-k B signaling and NLRP3 inflammasome induced by ASFV-H240R infection exhibit the antiviral effect.

DISCUSSION
In the coevolution of hosts and pathogens, natural selection results in a delicate balance between a host's ability to defend against pathogens and a pathogen's ability to evade host defenses (43). Previous studies have shown that the secretion of IL-1b, which is regulated by inflammasome activation, is a critical step in the host innate immune responses to infections with various pathogens (9,10). Upregulated expression of adhesion molecules of endothelial cells targets the activated immune cells to sites of infection via proinflammatory cytokines (9). Moreover, proinflammatory cytokines also activate the NF-k B, JNK, and p38 pathways to upregulate the expression of multiple genes that contribute to the innate immune responses against various pathogens at the sites of infections (44). Components of various pathogens can be recognized by inflammasome sensor proteins, which induce inflammasome activation and proinflammatory cytokine secretion (45,46). In addition, it has been shown that administration or depletion of proinflammatory cytokines can increase or decrease the load and/or pathogenicity of  (10 5 ) seeded in 48-well plates were cultured for 12 h, treated with 5 nM BAY11-7082 (an inhibitor of the NK-k B signaling pathway) in RPMI 1640 medium for another 2 h, and then infected with 10 5 genome copies of ASFV-DH240R or ASFV-WT. At 1.5 hpi, the medium was replaced with fresh medium. At 72 hpi, the viral genome copies (A) and viral titers (B) in PAMs infected with ASFV-DH240R or ASFV-WT were determined. (C and D) Inhibition of the NLRP3 inflammasome activation promotes ASFV-DH240R replication in PAMs. PAMs (10 5 ) were transfected with siNLRP3-1 for 24 h, followed by infection with 10 5 genome copies of ASFV-DH240R or ASFV-WT for 72 h, and the viral genome copies (C) and viral titers (D) in the PAMs were quantitated. Data represent three independent experiments. The error bars denote standard errors of the means. The significance of the difference between the groups (n = 3) was measured using Student's t test (**, P , 0.01; ***, P , 0.001; ns, not significant). various pathogens in vivo, respectively (44). The production of IL-1b is regulated by two pathways: the transcription of pro-IL-1b mRNA is regulated by NF-k B, which also regulates the synthesis of mRNAs encoding caspase 1 and pro-IL-1b (14)(15)(16), and the processing of IL-1b is modulated by the PRR-induced assembly of the inflammasome and the maturation of caspase 1, which leads to the cleavage of pro-IL-1b into the mature IL-1b (17). In this study, a novel mechanism by which ASFV pH240R suppresses NF-k B signaling and NLRP3 inflammasome activation to inhibit IL-1b transcription and maturation was revealed.
The low-level production of IL-1b induced by ASFV infection was identified by us and others (8,34), indicating that ASFV proteins have evolved multiple strategies to escape from the host immune responses and further suppress IL-1b production. pA238L, which is encoded by the ASFV A238L gene, is an Ik B homolog protein that significantly inhibits NFk B activation, thereby inhibiting the expression of IL-1b, but the virulence of ASFV with A238L deleted is not attenuated in vivo (33). pL83L, which is encoded by the ASFV L83L gene, interacts directly with IL-1b, but the virulence of ASFV with L83L deleted is not attenuated in vivo (32). However, pDP96R, which is encoded by the ASFV DP96R gene, inhibits the NF-k B promoter activation by interacting with IKKb, and the virulence of the DP96R deletion mutant is attenuated in vivo (47,48). pMGF505-7R, encoded by the ASFV MGF505-7R gene, binds to the IKK complex to inhibit pro-IL-1b transcription and interacts with NLRP3 to inhibit IL-1b secretion, and the virulence of ASFV with MGF505-7R deleted is attenuated in vivo (34).
Our previous study showed that pH240R is a capsid protein that interacts with p72 and is associated with ASFV morphogenesis during the ASFV life cycle (8). We found that ASFV-DH240R failed to efficiently assemble after entry into the cells and generated 98% aberrant tubular and bilobulate structures, resulting in major growth defects. In addition, ASFV-DH240R infection induced higher expression of proinflammatory cytokines, including IL-1b, IL-6, TNF-a, and C-X-C motif chemokine ligand 8, in PAMs than did ASFV-WT infection (8). In this study, we confirmed that the phosphorylation of Ik Ba and p65 and the transcription of pro-IL-1b were more highly induced by ASFV-DH240R infection than ASFV-WT ( Fig. 1 and 6), indicating that NF-k B signaling and inflammasome activation were induced in the ASFV-DH240R-infected PAMs. To further investigate the molecular mechanisms underlying the pH240R-mediated inhibition of IL-1b production, an NF-k B luciferase reporter and NLRP3 inflammasome reconstruction system were constructed based on the NF-k B and GLuc reporter systems to evaluate the activation of the NF-k B promoter and the production of IL-1b. We further confirmed that pH240R significantly inhibited NF-k B activation (Fig. 2) and IL-1b maturation (Fig. 7).
The inhibition of the IL-1b-mediated inflammatory response associated with various mechanisms, including the first signaling (transcription of pro-IL-1b) and the second signaling (maturation and release of IL-1b) (14)(15)(16)(17). Our results showed that pH240R inhibited both the transcription of pro-IL-1b and maturation of IL-1b, indicating that pH240R functions as a pleiotropic regulator of antiviral innate immune responses, including the NF-k B and NLRP3 inflammasome signaling cascades. Once NF-k B signaling is activated, the NF-k B dimer p50/p65 is translocated to the nucleus and induces the expression of proinflammatory cytokines. Viruses have evolved several strategies to inhibit the activation of NF-k B signaling, mainly regulating receptors, the IKK complex, Ik Ba, or p50/p65. It has been shown that herpes simplex virus 1 ubiquitin-specific protease (UL36USP) inhibits NF-k B activation by deubiquitinating Ik Ba and restricting its degradation (49). Kaposi's sarcoma-associated herpesvirus latency-associated nuclear antigen 1 inhibits NF-k B activation by degrading p65 through the ubiquitin-proteasome pathway (50). The human cytomegalovirus UL26 protein inhibits the phosphorylation of IKKb to antagonize the NF-k B activation (51). ORFV073 inhibits the activation of the IKK complex by interacting with the NEMO protein (52). Herpes simplex virus 1 ICP0 inhibits NF-k B activation by binding to the p65/p50 complex and degrading p50 through the ubiquitin-proteasome pathway (53). In this study, we found that pH240R inhibited NEMO expression by interacting with NEMO to inhibit NF-k B activation, thereby inhibiting the transcription of pro-IL-1b. However, effects of pH240R on the functions of other molecules cannot be ruled out.
When the NLRP3 inflammasome is activated, virus-encoded proteins inhibit IL-1b secretion by disrupting the NLRP3 oligomerization, ASC oligomerization, inflammasome complex formation, caspase 1 activity, and IL-1b maturation. It has been shown that the NS1 protein from influenza A virus inhibits IL-1b maturation by inhibiting caspase 1 activation (54). Overexpression of the influenza A virus NS1 protein in THP-1 cells inhibits NLRP3 inflammasome activation (55). The Sendai virus V protein inhibits ASC oligomerization and NLRP3 self-oligomerization by interacting with NLRP3, thereby inhibiting the secretion of IL-1b (29). In this study, we demonstrated that pH240R interacts with NLRP3 to inhibit its oligomerization, thereby inhibiting IL-1b maturation. Previous studies have shown that several proteins of ASFV, including F317L, MGF505-7R, MGF360-12L, D345L, and I215L, antagonize NF-k B signaling via multiple adaptors of signaling (34,(56)(57)(58)(59). We showed here that ASFV-DH240R infection induced a strong mature IL-1b response. The possible explanation is that pH240R blocks both the first and second signaling pathways of IL-1b maturation (Fig. 10).
ASFV targets the mononuclear-phagocytic system, and the infected tissues suffer extensive damage following infection with highly virulent isolates (60). ASFV isolates with low virulence also infect these cells, but the degree of tissue involvement and the resulting tissue damage are much less severe (61,62). ASFV is able to replicate efficiently and induce apoptosis in macrophages in vivo, which is likely to be critical to ASFV virulence (60). Although the pathogenicity of ASFV-DH240R has not been studied in vivo, the biological characteristics of ASFV-DH240R infection in vitro indicate the  (34), indicating that cytokine production in pigs is also one of the mechanisms underlying virulence attenuation. However, the virulence of ASFV-DH240R in pigs remains to be further investigated.
In summary, we demonstrate for the first time that pH240R inhibits IL-1b production by inhibiting NEMO expression and NLRP3 oligomerization. Our findings provide a novel target for the development of vaccines and drugs for the prevention and control of ASF.
RNA extraction and RT-qPCR. Total RNA was extracted from HEK293T cells or PAMs using RNAiso Plus (catalog no. 9109; TaKaRa) according to the manufacturer's protocols. The RNA was treated with DNase (catalog no. 143582; Roche) to remove genomic DNA. The cDNA was processed in a 20-mL volume with avian myeloblastosis virus (AMV) reverse transcriptase XL (catalog no. 2621; TaKaRa) and used as a template for RT-qPCR. The pro-IL-1b, IL-6, TNF-a, IKKa, IKKb, NEMO, Ik Ba, and NLRP3 cDNA levels were quantified by RT-qPCR in a reaction mixture containing 10 mM pro-IL-1b, IL-6, TNF-a, IKKa, IKKb, NEMO, Ik Ba, and NLRP3 forward and reverse primers (Table 1) and 2 mL of the cDNA template under the condition of 30 s at 95°C, followed by 40 cycles of 95°C for 15 s, 60°C for 30 s, and 72°C for 30 s. GAPDH was used as the internal control in this experiment.
Dual-luciferase reporter assay. HEK293T cells were cotransfected with reporter plasmids (pNF-k B-FLuc or pIRF3-FLuc [0.1 mg]) and pTK-RLuc (0.01 mg) with or without the indicated expression plasmids using X-tremeGENE HP (catalog no. 6366546001; Roche). At 24 h posttransfection (hpt), the cells were washed with phosphate-buffered saline (PBS) three times, 80 mL of passive lysis buffer was added in each well followed by incubation on a shaker for 20 min at 4°C, and then the luciferase activities were measured using a dual-luciferase reporter assay system (catalog no. E1910; Promega) according to the manufacturer's instructions. The results are presented as the ratios of firefly luciferase (FLuc) to Renilla luciferase (RLuc).
Co-IP and Western blotting assays. HEK293T cells were cotransfected with the pMyc-H240R and pFlag-cGAS, pFlag-STING, or pFlag-TBK1 (2 mg each), or with pFlag-H240R and pMyc-IKKa, pMyc-IKKb, or pMyc-NEMO (2 mg each) for the co-IP assay. At 48 hpi, the cells were lysed with NP-40 for 20 min and centrifuged for 12 min at 4°C; the supernatants in loading buffer were boiled for 8 min at 100°C. The supernatants were incubated with the indicated primary antibodies in the presence of protein G-agarose (catalog no. IP10-10ml; Merck-Millipore) at 4°C overnight. The agarose was washed five times with PBS, and the protein expression was analyzed by Western blotting as described previously (62).
Measurement of inflammatory-cytokine levels. A total of 4 Â 10 6 PAMs in 6-well plates were infected with 10 7 genome copies of ASFV-DH240R or ASFV-WT for 1.5 h at 37°C and then washed twice with PBS. The PAMs were then cultured in fresh medium for 24 h. The cell supernatants were collected to measure the concentrations of cytokines using ELISA kits (Ray Biotech, Norcross, GA). The cells were collected for RT-qPCR and Western blotting, and uninfected cells were used as a negative control. All the samples were analyzed in triplicate.
RNA interference assay. siRNAs targeting the porcine NLRP3 genes were synthesized by Gemma. The siRNA sequences targeting NLRP3 were GCU GCA AGC UGG CUC GUU ATT (siNLRP3-1), GCC UGU GCA CAC GGU AGU ATT (siNLRP3-2), and CCU GUU CAC CAU GUG CUU UTT (siNLRP3-3). The siNC sequence was UUC UCC GAA CGU GUC ACG UTT. A total of 5 Â 10 5 PAMs were seeded in 24-well plates and cultured for 12 h. The PAMs were transfected with 100 nM siNLRP3 or siNC using the XtremeGENE siRNA transfection reagent (catalog no. 4476093001; Roche) according to the manufacturer's instructions. At 24 hpt, the transfected PAMs were infected with ASFV-DH240R for 1.5 h, and the supernatants were washed with PBS and replaced with fresh medium. At 72 hpi, the supernatants and cell lysates were harvested to measure the viral titers and genomic copies.
Cell viability assay. A cell viability assay was performed with the Cell Counting Kit-8 (CCK-8) (catalog no. CK04; Dojindo) according to the manufacturer's instructions.
Reconstitution of the NLRP3 inflammasome. An NLRP3 inflammasome reconstitution assay was performed as described previously (64). Briefly, HEK293T cells were seeded in a 24-well plate and cotransfected