EV-A71 induced IL-1β production in THP-1 macrophages is dependent on NLRP3, RIG-I, and TLR3

Enterovirus A71 (EV-A71) is an emerging enterovirus that can cause neurological complications. Enhanced serum IL-1β levels were observed in EV-A71 patients with severe neurological symptoms. However, the roles of sensors in enterovirus-induced IL-1β production are unclear. In this study, we identified that pattern recognition receptors, including RIG-I, TLR3, and TLR8, are implicated in EV-A71-triggered IL-1β release in human macrophages. EV-A71 infection results in caspase-1 and caspase-8, which act as regulators of EV-A71-induced NLRP3 and RIG-I inflammasome activation. Moreover, knockdown of the expression of TLR3 and TLR8 decreased the released IL-1β in an NLRP3-dependent manner. Since TLR3 and TLR8 ligands promote NLRP3 inflammasome activation via caspase-8, the alternative pathway may be involved. In summary, these results indicate that activation of the NLRP3 and RIG-I inflammasomes in EV-A71-infected macrophages is mediated by caspase-1 and caspase-8 and affected by TLRs, including TLR3 and TLR8.


Introduction
EV-A71 is a positive-sense RNA virus that belongs to the member of the family Picornaviridae. EV-A71 infection is associated with several diseases, including hand, foot, and mouth disease (HFMD) and herpangina. Moreover, EV-A71 may invade the central nervous system and cause severe neurological complications, causing signi cant concern in the Asia-Paci c region 1 . The severity of EV-A71 infection has been suggested to be correlated with the production of pro-in ammatory cytokines 2 . Previous studies suggested that the production of in ammatory cytokines such as IL-6 and IL-1β may be involved in central nervous system (CNS) damage caused by EV-A71 infection 3,4 .
As the major effector cells of the innate immune system, macrophages play essential roles in the recognition and destruction of invading microorganisms. When exposed to pathogens or in ammatory stimuli, they release cytokines and chemokines to induce enhanced vascular permeability and recruitment of immune cells 5 . However, many viruses are able to infect and replicate in macrophages to facilitate their dissemination 6 . Macrophage infection induces the expression of cytokines, which is associated with disease severity 7 . Moreover, emerging evidence suggests that virus infection can activate in ammasomes in macrophages to produce IL-1β and IL-18 8, 9 . Excessive IL-1β expression is associated with Theiler's murine encephalomyelitis virus (TMEV)-induced demyelinating disease by promoting the generation of Th17 cells 10 . Furthermore, a recent study indicated that the nucleotide-binding oligomerization domain 3 (NLRP3) in ammasome is involved in neuroinvasion and neuroin ammation during severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection 11 . In ammasomes are composed of multiple proteins, including NOD-like receptors (NLRs), the adaptor protein apoptosisassociated speck-like protein containing caspase recruitment domain (CARD) known as ASC, and the effector protein pro-caspase-1. Recently, several types of in ammasomes were characterized, and it was revealed that NLRP1, NLRP3, NLR family CARD domain-containing 4 (NLRC4) and pyrin in ammasome activation is mediated by NLR family members, while the absent in melanoma 2 (AIM2) in ammasome is activated by AIM2, which belongs to the I 202/IFI116 family 12,13 .

RNA viruses such as IAV, human immunode ciency virus (HIV), and encephalomyocarditis virus (EMCV)
can induce NLRP3 in ammasome activation [14][15][16] and the subsequent generation of IL-1β. The formation of NLRP3 in ammasomes has been demonstrated to be involved in viral pathogenesis 17−19 . The NLRP3 in ammasome can be activated by a wide range of stimuli, such as extracellular ATP, asbestos, silica, alum, amyloid β, single-stranded RNA (ssRNA), double-stranded (dsRNA) analogs, and pathogen products 20 . Accumulating evidence has demonstrated that picornaviruses, including EMCV, coxsackievirus B3 (CVB3), and poliovirus (PV), can produce IL-1β 16,21 . The induction of IL-1β is implicated in CVB3-induced myocarditis, and the transplantation of NLRP3-knockdown (KD) macrophages results in reduced IL-β secretion and milder symptoms 21 , indicating that IL-1β plays an essential role in pathogenesis. In human rhinovirus (HRV)-infected airway epithelial cells, in ammasomedependent IL-1β secretion is involved in the exacerbation of pulmonary symptoms 22 . Interestingly, the production of IL-1β can result in the pyroptosis of EV-A71-infected neural cells, which suggests that in ammasome activation may be involved in programmed cell death 3 . In contrast, the production of IL-1β has been reported to play bene cial roles during enteroviral infections. For example, Wang et al. showed that CVB3 infection resulted in more severe symptoms in NLRP3-knockout mice 23 , while the NLRP3 in ammasome was demonstrated to elicit bene cial effects on EV71-infected animals 24 . The con icting results of these studies suggest that in ammasome activation may exacerbate or ameliorate disease severity during enteroviral infections.
Previous research has shown that EV-A71 infection can activate IL-1β release mediated by the NLRP3 in ammasome activated via viral proteins 24,25 . However, the mechanisms associated with EV-A71induced IL-1β production in macrophages have not been completely characterized. In this study, we demonstrated that in addition to NLRP3, PRRs, including RIG-I and TLR-3, are involved in in ammasome activation in EV-A71-infected THP-1 macrophages. Moreover, both caspase-1 and caspase-8 are implicated in EV-A71-induced IL-1β production in THP-1 macrophages. Our results suggest that the release of IL-1β is modulated by multiple PRRs and caspases in THP-1 macrophages.

Isolation of PBMCs and differentiation of PBMCs toward macrophages
Human peripheral blood samples were collected after approval by IRB (Chang Gung Medical Foundation institutional review board, IRB2019050059). All the research process was performed in accordance with IRB guidelines and regulations. The informed consent was obtained from all subjects and/or their legal guardians. Mononuclear cells were harvested by Ficoll-Paque method. Brie y, peripheral blood was mixed with PBS (1:1). The diluted cell suspension was then layered on Ficoll-Paque (GE Healthcare Life Sciences, MA, USA) (volume 2:1) in a 50 mL canonical tube. After centrifugation at 400 x g for 20 minutes, the top layer was aspirated. The mononuclear cell layer in the interface was then transferred to a new tube. To differentiate the harvested PBMCs toward macrophages, RPMI medium supplemented with 20 ng/mL M-CSF (Peprotech, NJ, USA) and 1% human serum was added and cultured for 5 days.

RNA extraction and RT-qPCR
Trizol reagent (Life Technologies, Gaithersburg, MD) was applied to extract total RNA from cell samples.
The cells were lysis by Trizol reagent and then mixed with chloroform. After 5 min, the homogenate was centrifugated at 12,000 x g for 15 min at 4 o C. The aqueous phase was transferred to a new tube and added with an equal volume of isopropanol and then incubated for 10 min. The mixture was centrifuged at 12,000 x g for 8 min at 4 o C. After the centrifugation, the supernatant was removed. 500 µL 7 % ethanol was used to wash the RNA pellet then centrifugated at 12,000 x g for 5 min at 4 o C. Discard the supernatant and then air dried the RNA pellet. The RNA pellet was dissolved by sterile water and quanti ed the concertation of RNA by NonoDrop technology (Thermo-Fisher Scienti c, MA, USA). 1 µg of total RNA was used to synthesize cDNA. RevertAid First Strand cDNA Synthesis Kit (Thermo-Fisher Scienti c, MA, USA) was ap-plied according to the manufacturer's instructions. To detect the expression of target genes, speci c primers for TNF-α, IL-6, IL-1β, EV-A71 5'UTR were used (Table 1). qPCR assays were carried out on 384-well plates and analyzed by a Roche Lightcycler 480 instrument (Roche, Basel, SW). Triplicate for each sample in qPCR analysis and 18s rRNA was used as a reference gene. The relative expression level of each gene was analyzed by 2 −ΔΔCT method. Table 1 qPCR primer used in this study.

Statistical analysis
All experiments were repeated at least 3 times. Results were shown as the means ± SD. The data were analyzed by Student's unpaired T-test. The value of p < 0.05 was indicated statistical signi cance.

Data availability
The datasets used and analyzed during the current study available from the corresponding author on reasonable request.

Results
EV-A71 actively replicates in PMA-primed THP-1 cells without causing signi cant cell death. PMA-primed THP-1 cells were seeded and infected with EV-A71 at the M.O.I. of 2 for two days. Morphological changes were observed, and no obvious cytopathic effects were observed ( Figure 1A). Cell viability was examined by MTT assay, and no signi cant cell death appeared ( Figure 1B). The expression of EV-A71 viral protein 3D was examined by Western blot analysis, and the viral protein appeared at 8 hours post-infection (p.i.) and peaked at 12 hours p.i. (Figure 1C). Total lysates of infected cells were collected, and virus titers were determined using a plaque forming assay ( Figure 1D). The virus titers were increased during infection, indicating that the virus was actively replicating.
EV-A71 infection triggers the expression of proin ammatory cytokines. To determine whether EV-A71 infection can upregulate the expression of proin ammatory cytokines, total RNA extracted from EV-A71infected PMA-primed THP-1 cells was examined by RT-qPCR to detect the expression levels of TNF-α, IL-6 and IL-1β. Our results showed that the transcripts of these three cytokines were upregulated, while no signi cant increase was observed in CVA16-infected cells (Figure 2A). Furthermore, nuclear parts were isolated from EV-A71-infected THP-1 macrophage-like cells. The expression of phospho-P65 in the nucleus and cytosol was examined by immunoblotting. The expression levels of phospho-P65 in the nucleus were increased at 1 hour p.i. and declined at 12 hours p.i. (Figure 2B).
EV-A71 induces the production of IL-1β from PMA-primed THP-1 cells in a dose-dependent manner. To determine whether IL-1β can be secreted, THP-1 macrophages were infected with EV-A71 or treated with LPS. Cell lysates and supernatants were collected, and pro-IL-1β expression was measured by Western blot analysis ( Figure 3A). The amount of the secreted IL-1β p17 subunit was increased in a timedependent manner ( Figure 3B). Supernatants were harvested at different time points and subjected to ELISA. Our results revealed that EV-A71 infection and LPS stimulation resulted in the secretion of IL-1β from PMA-primed THP-1 cells ( Figure 3C). An increase in IL-1β levels was detected 12, 24, and 48 hours p.i.. To determine whether EV-A71 activates IL-1β in a dose-dependent manner, THP-1 macrophages were infected with EV-A71 at different M.O.I.s. The amount of secreted IL-1β was increased in the cells infected with EV-A71 in a dose-dependent manner ( Figure 3D). Different EV-A71 strains were used to infect cells, and the expression of mature IL-1β was measured to determine whether the increase in secreted IL-1β is strain-speci c. Our results revealed that all tested EV-A71 strains were able to cause the release of IL-1β ( Figure 3E). To test whether EV-A71 infection can induce the production of IL-1β in human macrophages, blood samples were collected from healthy volunteers, and peripheral blood mononuclear cells (PBMCs) were isolated. Isolated PBMCs were then forced to differentiate into human macrophages. These cells were then subjected to EV-A71 infection, and our results showed that the release of mature IL-1β was increased at 24 hr p.i. (Figure 3F). These ndings indicated that EV-A71 can activate the production of IL-1β in human macrocytic cells.
Knockdown of NLRP3, RIG-I, and TLR3 decreases the expression of IL-1β in EV-A71-infected cells. To elucidate which types of in ammasomes are involved in EV-A71-induced IL-1β production, siRNAs speci c for NLRP3 were transfected into THP-1 macrophages. The knockdown e ciency was con rmed by immunoblot analysis ( Figure 4A). Our results also showed that the expression of NLPR3 was increased in EV-A71-infected cells. The transfected cells were subsequently infected with EV-A71 at the M.O.I. of 2, and the release of mature IL-1β was assessed by ELISA. The production of IL-1β was drastically downregulated in NLRP3 KD cells, which suggests that NLRP3 is essential for EV-A71-induced IL-1β production ( Figure 4B). We noticed that NLRP3 KD was not able to diminish the synthesis of IL-1β. Thus, other RNA sensors may be implicated in EV-A71-induced in ammasome formation. Previous studies suggested that RIG-I and TLR3, as well as NLRP3, are involved in in ammasome activation in in uenza A virus-infected epithelial cells 14 . To test whether RIG-I and TLR3 were involved, siRNAs speci c for these two genes were applied for cell transfection. The expression levels of RIG-I and TLR3 in scrambled siRNA-and speci c siRNA-transfected THP-1 macrophages were examined by Western blot analysis ( Figure 4C and 4E). The production of IL-1β was determined by ELISA. Our results showed that knockdown of RIG-I and TLR3 resulted in approximately 50% downregulation of EV-A71-induced IL-1β release ( Figure 4D and 4F).
Caspase-1 and caspase-8 are involved in EV-A71-triggered IL-1β production. Caspase-1 has been demonstrated to be involved in IL-1β production by promoting pro-IL-1β processing 26 . This enzyme is activated by the formation of multiple in ammasomes 27 . The expression of the activated caspase-1 p20 subunit was detected in both LPS-treated and EV-A71-infected PMA-primed THP-1 cells ( Figure 5A). The addition of Ac-YVAD-cmk, a caspase-1 inhibitor, suppressed EV-A71-induced IL-1β secretion ( Figure 5B). To further characterize the role of IL-1β in macrophages upon EV-A71 infection, siRNA speci c for caspase-1 was transfected into THP-1 macrophages that were then infected with EV-A71. The knockdown e ciency was con rmed by Western blot analysis ( Figure 5C). Supernatants were harvested, and the amount of released IL-1β was analyzed by ELISA. The results showed that the amount of secreted IL-1β protein was drastically decreased in caspase-1-KD cells ( Figure 5D). In addition to caspase-1, caspase-8 has also been implicated in the regulation of the NLRP3 in ammasome in murine dendritic cells 28 . To examine the involvement of caspase-8, PMA-primed THP-1 cells were infected with EV-A71 to induce IL-1β synthesis and then treated with Z-IETD-FMK. The expression of cleaved caspase-8 was signi cantly decreased in infected cells treated with Z-IETD-FMK ( Figure 5E). Compared to the untreated cells, only 1/4 of the IL-1β protein was detected in the supernatant of Z-IETD-FMK-treated samples ( Figure  5F). These ndings suggested that both caspase-1 and caspase-8 are involved in IL-1β production in EV-A71-infected THP-1 macrophages.

Discussion
Macrophages have been shown to be the target cells for many viruses, such as dengue virus and in uenza virus 29,30 . The roles of macrophages in virus pathogenesis include altering the function of macrophages, enhancing the expression of proin ammatory cytokines, and supporting productive replication 30,31 . Prior studies showed that macrophages are productively infected by poliovirus, and these infected APCs can present the antigen 32 . Moreover, nonpolio enteroviruses, including CVB3 and rhinovirus, are also able to infect macrophages 33,34 . In this study, PMA-primed THP-1 cells, which are commonly used for studying the biological activities of macrophages, were used as an in vitro cell model.
PMA primed THP-1 cells 35 . Our data showed that EV-A71 infection did not cause morphological changes in infected cells, and no cells were observed. This could be attributed to the enhanced expression of IFN-b in EV71-infected THP-1 macrophages (data not shown). A similar phenomenon has been demonstrated before 35 .
Overproduction of proin ammatory cytokines is correlated with the severity of EV-A71 infection 2 . Prior studies have shown that the sustained overproduction of IL-6 is correlated with disease severity in neonatal mice 36 . However, contrasting results have been recently published by Wang et al., who demonstrated that the use of an anti-IL-6 antibody enhances the lethality of EV-A71-infected mice 37 . In contrast to IL-6, several clinical reports have shown the involvement of TNF-a in the severity of illness in EV-A71 patients [38][39][40] . In accordance with previous results, our data suggest that these cells could be responsible for cytokine production 41 . We also showed that the expression levels of in ammatory cytokines were higher in EV-A71-infected THP-1 macrophages than in CVA16-infected cells. The differential ability of EV-A71 and CVA16 to trigger cytokine expression has been reported before by using human enteroids as a cell model 42 . A recent study showed that EV-A71 genotype G, a clinical isolate from AFP patients, was able to induce higher expression of cytokine/chemokine genes than other isolates 43 .
Many studies have demonstrated that EV-A71 can lead to the production of IL-1β via activation of the NLPR3 in ammasome 16, 44,45 . Viral protein 2B in EMCV, PV, and EV-A71 is able to activate the NLRP3 in ammasome by inducing a ux of calcium ions 16 . Wang et al also demonstrated that EV-A71 induced in ammasome activation through the 3D protein 46 . However, the mechanism underlying in ammasome activation in EV-A71-infected THP macrophages is not clear. Our results suggest that the expression of TLR3, RIG-I, and NLRP3 is essential for EV-A71-triggered IL-1β release. This is the rst report demonstrating that RIG-I and TLR3 are implicated in EV-A71-induced IL-1β release. Cytosolic RIG-I has been shown to activate in ammasomes and drive subsequent IL-1β production in VSV-infected human peripheral blood-derived monocytes (PBMCs) 47 . Furthermore, in uenza A infection in human primary bronchial epithelial cells leads to IL-1β release through the interaction of RIG-I with ASC and caspase-1 14 .
It is well known that RIG-I is responsible for recognizing VSV and in uenza virus, while MDA5 is related to sensing picornaviruses 48 . However, RIG-I is implicated in EV-A71-induced in ammasome formation but not MDA5.
Treatment with RIG-I agonists is su cient for caspase-1 activation and the release of mature IL-1β 47 . Both RIG-I and NLRP3 in ammasomes can activate caspase-1 to produce mature IL-1β. However, Ac-YVAD-cmk, a caspase 1 inhibitor, was not able to totally abolish the production of IL-1β in EV-A71-infected THP-1 macrophages. Moreover, treatment with a caspase-8 inhibitor drastically downregulated the release of IL-1β. Therefore, both caspase-1 and caspase-8 are involved in virus-triggered in ammasome activation. Caspase-8 can serve as a positive regulator of the NLRP3 in ammasome to improve IL-1β maturation in mouse dendritic cells 28 . Furthermore, activated caspase 8 can act on pro-IL-1β directly to form mature IL-1β 49 . Many studies have demonstrated that caspase 8 is involved in the production of IL-1β in innate immune cells 50,51 . Our results suggest that more than one caspase is involved in EV-A71-induced IL-1β production in THP-1 macrophages. However, the underlying mechanisms are not clear. In summary, this study shows that EV-A71 can induce the upregulation of IL-1β with the involvement of TLR3, NLRP3, and RIG-I.