Dengue Virus Envelope Protein Domain III Induces Nlrp3 Inflammasome-Dependent NETosis-Mediated Inflammation in Mice

Abnormal immune responses and cytokine storm are involved in the development of severe dengue, a life-threatening disease with high mortality. Dengue virus-induced neutrophil NETosis response is associated with cytokine storm; while the role of viral factors on the elicitation of excessive inflammation mains unclear. Here we found that treatments of dengue virus envelope protein domain III (EIII), cellular binding moiety of virion, is sufficient to induce neutrophil NETosis processes in vitro and in vivo. Challenges of EIII in inflammasome Nlrp3−/− and Casp1−/− mutant mice resulted in less inflammation and NETosis responses, as compared to the wild type controls. Blockages of EIII-neutrophil interaction using cell-binding competitive inhibitor or selective Nlrp3 inflammasome inhibitors OLT1177 and Z-WHED-FMK can suppress EIII-induced NETosis response. These results collectively suggest that Nlrp3 inflammsome is a molecular target for treating dengue-elicited inflammatory pathogenesis.


INTRODUCTION
Dengue is one the most important mosquito borne diseases in the tropical and subtropical areas of the world (1, 2), while specific treatments and effective vaccines are currently unavailable (3)(4)(5)(6)(7)(8). Infections with dengue viruses (DENV) can lead to a wide range of clinical manifestations and disease severity. Severe dengue (also known as dengue hemorrhage fever, DHF) is characterized by plasma leakage and abnormal bleeding that can lead to shock and high mortality. Because DHF typically occurs during secondary infections with DENVs, abnormal adaptive immune responses are considered as part of the pathophysiology. For example, reports have suggested that antibody-dependent enhancement (9), original antigenic sin (10), autoantibody production (11) may be involved. However, detrimental innate immune responses such as excessive inflammation and cytokine storm are likely the critical pathological changes that lead to exacerbated disease, tissue injuries and ultimate death in DHF (9,10,(12)(13)(14)(15).
The mechanism underlying dengue-induced unregulated inflammation remains elusive (9,10,12,13). In the innate immune system, neutrophils are first line of defense against infection through engulfment of microbes, secretion of antimicrobials and induction of neutrophil extracellular traps (NETs)-releasing cell death process termed NETosis (16,17). NETs are extracellular DNA-protein complexed networks, which bind pathogens and modulate inflammation (16,18). Pathogenic roles of NETosis have been found in non-infectious diseases, such as autoimmunity, coagulation, acute injuries and cancer (19). In addition, NETosis has been reported associating with cytokine storm in various infectious diseases, including dengue (14,20,21). Interleukin (IL)-1β, a potent proinflammatory cytokine released by DENV-infected leukocytes, has been considered as a critical component in cytokine storm (22)(23)(24). Inflammasomes, cellular sensors for pathogen associated molecular patterns (PAMPs) and damage associated molecular patterns (DAMPs), are critical for IL-1 activation (14,25), and the cell death process pyroptosis (26). Reports revealed that the elevated levels of circulating IL-1β and gene expression in DHF patients suggesting the involvement of IL-1β in the disease severity (27,28). IL-1β enhances the vascular permeability, particularly in association with other proinflammatory cytokines such as tumor necrosis factor (TNF)-α and interferon in clinical profiles of DF and DHF (14,(29)(30)(31).
The DENV viral factor that contributes to NETosis remains unclear. Plasma EIII levels could be detected in acute DENV infection (34). Evidences have shown that EIII treatments induced inflammasome activation and inflammation of macrophages (35). Our previous study revealed that challenges with the DHF-viral-load-equivalent levels of EIII can suppress megakaryocyte, and endothelial cell function through initiating cell death (33,36). Accordingly, EIII may be a cytotoxic virulence factor of DENV to cause NETosis and initiate downstream inflammation. As a result, in this present study, we would like to investigate whether DENV and EIII can directly initiate NETosis, and whether Nlrp3 inflammasome is involved. In addition, whether we can ameliorate EIII-mediated inflammation through suppression of Nlrp3 inflammasome and NETosis pathways is also addressed. Relevant implications and applications are discussed.

Ethics Statement
The animal experiments in this report were conducted in agreement with National (Taiwan Animal Protection Act, 2008) directive for protection of laboratory animals. All experimental protocols for examining the experimental animals were approved by the Animal Care and Use Committee of Tzu-Chi University, Hualien, Taiwan (approval ID: 101019).

Statistical Analyses
In this report, the means, standard deviations, and statistics of the quantifiable data were calculated using the SigmaPlot 10 and SPSS 17 software packages. Significance of the data was examined using one-way ANOVA, followed by the post hoc Bonferroni-corrected t-test. The probability of type-1 error α = 0.05 was recognized as the threshold of statistical significance.

DENV and rEIII Induce Multiple Regulated Cell Death Pathways of Neutrophils
NETosis is a type of regulated cell death (RCD) (58). Previous reports suggested that neutrophil RCD exacerbate pathogenesis in infectious diseases (59,60). As a result, we would like to investigate whether various RCD pathways are also involved in DENV-and rEIII-induced neutrophil cell death.
We first found that DENV and rEIII induced neutrophil cell death in a dose dependent manner ( Figure 3A). Overall, various cell death inducers, including doxorubicin (apoptosis) (61,62), rapamycin (autophagy) (63), erastin (ferroptosis) (64), TNF-α (necroptosis) (65, 66), nigericin (pyroptosis) (67), served as positive control agents to induce respective RCD pathways of the tested neutrophils (Figures 3B,C, dead cell population adjusted to 100%; Supplementary Figure 2, flow cytometry gating and calculation). Notably, when compared with cell death agonists, DENV and rEIII treatment induced considerable pyroptosis, necroptosis, autophagy and NETosis responses in the neutrophils, while only minor or no ferroptosis and apoptosis levels ( Figure 3B, % of total cells; Figure 3C, % of total dead cell). In addition, the cell type specific RCD patterns/profiles (CTS-RCDPs) (33) of neutrophil in the DENV-, and rEIII-treated groups were somewhat similar, with pyroptosis exhibiting the highest levels in both groups among all tested RCD pathways ( Figure 3B; ∼40%), suggesting that DENV-induced CTS-RCDP in the neutrophils is likely mediated through EIII on the DENV virion. In case one dead cell may display multiple RCDs, here we defined CTS-RCD as a detection ratio of RCDs in 1 cell type at a specific condition.  An unexpected finding is that the DENV and rEIII induced NETosis only displayed approximately 20% of total RCDs ( Figure 3B); and a classical NETosis inducer TPA (a phorbol ester) also induced NETosis about only 40% of total RCDs (Figure 3B, TPA groups). This let us wondered whether DENV and EIII-mediated induction of such a low percentage of NETosis in total RCD, could sufficiently lead to neutrophil dysfunction. In addition, we would like to investigate whether Nlrp3 inflammasome is involved in rEIII-induced neutrophil death. Accordingly, Nlrp3 inhibitor OLT1177 and inflammasome/caspase1 inhibitor Z-WHED-FMK were used to further characterizations of whether Nlrp3 inflammasome is involved in respective RCD responses.
We found that treatments with inflammasome inhibitors OLT1177 and Z-WHED-FMK both suppressed EIII-induced neutrophil cell death (Figures 4A,B; Supplementary Figure 3, percentage pie charts of OLT1177 and Z-WHED-FMK treatments; equivalent to some data in Figures 4A,B,I).

Nlrp3 Inflammasome Deficiency and Inhibitor Treatments Rescue DENV-and rEIII-Exacerbated Mitochondria Metabolic Burden and Inflammation of Neutrophils
Because inflammasome-mediated pyroptosis is a major RCD involving in rEIII-induced neutrophil defect, here we would like to further investigate whether suppression of neutrophil Nlrp3 inflammasome through inhibitor treatments is sufficient to ameliorate DENV rEIII-induced neutrophil defects. Here we found that treatments of rEIII-increased mitochondria mass (Figure 5A), membrane potential ( Figure 5C) and superoxide ( Figure 5E) levels in a dose dependent manner, while treatments of Nlrp3 inflammasome inhibitors OLT1177 and Z-WHED-FMK ameliorated such metabolic burden of neutrophil mitochondria (Figures 5B,D,F). Levels of caspase-1 activation in the neutrophils were analyzed and confirmed in parallel using colorimetric assay (Supplementary Figure 5). In agreement with this, mouse experiments further revealed that, treatments of Nlrp3 inflammasome inhibitors OLT1177 markedly ameliorated rEIII-induced elevation of circulating soluble CitH3 (Figure 6A), IL-1β (Figure 6B), and TNF-α ( Figure 6C) levels in mice. Consistently, compared to wild type mice, neutrophils from Nlrp3 −/− and Casp1 −/− mutant mice displayed markedly reduced levels of total mitochondria ROS (Figure 7A), hydrogen peroxide (Figure 7B), superoxide ( Figure 7C) after in vitro treatments of rEIII. Similarly, rEIII treatments markedly induced circulating CitH3 in wild type mice, but not in Nlrp3 −/− and Casp1 −/− mutant mice ( Figure 7D). These results collective suggest that EIII is a virulence factor to induce neutrophil defects, and Nlrp3 inflammasome is a critical target for DENV and EIII to induce neutrophil dysfunction, NETosis, and inflammation.
predominantly displayed in serum samples of DHF patients (75). Despite of these findings, the mechanism underlining DENVinduced NETosis formation is not fully understood; particularly, the viral factor leading to NETosis remains elusive. Here we found that, the DENV EIII could be a potential virulence fact that elicits abnormal neutrophil responses and NETosis. Because DENV and EIII induced similar RCD patterns of neutrophils ( Figure 3B, DENV and rEIII groups), suggesting that, during the speak viremia stage, high levels of DENV virion conduct the cytotoxicity to enhance NETosis through EIII moiety. This is partly consistent with the findings that NET formation is markedly increased in neutrophils isolated from dengue patients during the acute phase of the infection (75). Additionally, our data suggests that such cell-type-specific RCD patterns may be a useful analysis method on the functional characterization of biologically drugs and hazardous materials at the cellular levels.
Inflammasomes regulate and interact with various RCDs (72,(76)(77)(78). The inhibitors of either Nlrp3 inflammasome/pyroptosis (inhibitor OLT1177) or NETosis (inhibitor GSK484) can suppress to each other (Figure 4), suggests there could exist a cross-talk between these 2 pathways. First, such crosstalk could be regulated at an intracellular signaling level; for example, GSDMD, an effector protein of pyroptosis, plays a critical role in the generation of NET (79). Consistently, here we found that treatments of GSDMD inhibitor DMF drastically reduced EIII-induced pyroptosis and NETosis in neutrophils (Supplementary Figure 4). Accordingly, if Nlrp3 inflammasome-activated GSDMD can further enhance NET formation, it is reasonable to observe an inhibition of NETosis using inhibitors from a pyroptosis pathway. Second, the crosstalk may also explain by a model of intercellular regulations between macrophage and neutrophil. For example, the NET is able to prime macrophages to produce IL-1 and IL-18 through the Nlrp3 inflammasome, thus amplifying the inflammatory response (80). At the same time, IL-18 effectively stimulated NET release and caspase-1 activation in primed macrophages compared to IL-18 alone (80). This suggests a feed-forward loop that NET increase the production of IL-1β and IL-18 in macrophages, which in turn can stimulate NET formation in neutrophils (80). These evidences may explain our observations that treatments of Nlrp3 inhibitor OLT1177 can markedly suppressed rEIII-induced neutrophil pyroptosis and NETosis in vitro (Figure 4), and markedly suppressed rEIII-induced NETosis and inflammation in mice (Figure 6). However, as these . n = 6, # P < 0.05, ## P < 0.01, vs. respective wild type (WT) groups; *P < 0.05, **P < 0.01, vs. respective vehicle groups.
regulations still not been clearly demonstrated in dengue, more detailed mechanism is worth of further investigations.
As a glycosaminoglycan (GAG) binding lectin (81), EIII could have multiple neutrophil surface targets. Recent evidences have suggested that lectin DC-SIGN mediated DENV infection in dendritic cells (82); lectins CLEC2 and CLEC5A plays critical roles on DENV-induced inflammation (24,74,83,84). CLEC5A and DC-SIGN are expressed by leukocyte subpopulations (82,84); while their respective isoforms CLEC2 and DC-SIGNR preferentially expressed by the other cell types (84,85). Here we used recombinant soluble CLEC2, CLEC5A, DC-SIGN, DC-SIGNR, to perform EIII competition experiments. Analysis results revealed that CLEC5A-, and DC-SIGN-, but not CLEC2-, and DC-SIGNR-coated beads can bind to rEIII (Supplementary Figure 6). Consistently, CLEC5A and DC-SIGN, but not CLEC2 and DC-SIGNR can block rEIII-neutrophil binding (Supplementary Figure 6). In addition, CLEC5A and DC-SIGN can reduce EIII-induced increased ROS and NETosis levels in neutrophils (Supplementary Figure 6). These evidences collectively suggested that CLEC5A and DC-SIGN are cellular targets of EIII on neutrophils. As a critical inflammatory regulator, the role of CLEC5A and DC-SIGN in EIII-mediated pathogenesis is worthy of further investigations.
Cell population is heterogeneous even in one cell line. This could be a reason that reports have revealed treatments of pathogens and cytotoxic agents leading to multiple types of RCDs simultaneously (86)(87)(88)(89)(90)(91)(92). For example, when cellular stress can activate both receptor-induced lysosomal-dependent, and mitochondrial-mediated cell death pathways, which will lead to both programmed necrosis and apoptosis (88). Similarly, as DENV and EIII been reported to have multiple cellular targets, it is reasonable to detect multiple RCDs after DENV and EIII challenges. Here, we found that DENV and EIII but not the other RCD inducers, induced a similar RCD pattern in neutrophils (Figures 3B,C). Previous reports have suggested that, upon stimulation by PAMP, neutrophil activation results in necrotic RCDs (93,94); by contrast, neutrophil apoptosis contributes to the resolution of inflammation (93)(94)(95)(96). Consistently, here we found that DENV and EIII (pathogen, PAMP) preferentially induced necrotic RCDs with almost no detectable apoptosis levels in neutrophils (Figure 4). Although further investigations are needed, our data suggested that the neutrophil CTS-RCDPs are useful on the characterization of cellular tendency on the induction of RCDs in particular conditions.
In addition to EIII, DENV membrane protein (M) and nonstructural proteins NS2A, NS2B were also shown to induce inflammasome activation (14,97,98), and NS1 was demonstrated to enhance inflammation through a toll-like receptor 4 (99, 100). NS2A, NS2B have shown to serve as viroporins to increase cellmembrane permeability and activate inflammasome (14,101,102). Viroporins primarily affect virus-infected cells (103), and clinical course of DHF occurs specifically when the viremia markedly decreased (104, 105); these evidences suggest that NS2A, NS2B-mediated inflammasome activations may be not timely associated with the clinical course of DHF. DENV virus particle-expressed EIII and soluble NS1 are detected at high levels prior to the acute phase of DHF (104, 105), and may be considered as two virulence factors for severe dengue. Here we found that both EIII and NS1 treatments can induce increased NETosis levels in neutrophils, and EIII has a relative higher activity ( Supplementary Figure 7; DENV envelop domain I, domain II protein fragment, served as a control protein). Because the induction of virion-expressed EIII and soluble NS1 are induced in a similar, but not a same time course (105), the respective pathogenic role of EIII and NS1 on the elicitation of DHF-related pathogenesis remains to be further studied. However, data obtained in the study suggested that virionassociated EIII is a candidate virulence factor that contributes to dengue-elicited NET formation.
In this present study, we found that treatments of both DENV and EIII, a cell-surface-binding and cytotoxic protein, can induce multiple cell death pathways in neutrophils with a similar cell-type-specific RCD pattern ( Figure 3B). NETosis inhibitor GSK484 treatments sufficiently suppressed EIII-induced cell death (Figure 4A), NETosis (Figure 4G), and mitochondria metabolic burden (Figure 5) of neutrophils. This suggested that, despite the NETosis seems to be a relatively minor response that counts approximately 20% of total neutrophil RCDs, the EIII-stimulated NETosis is sufficiently leading to an abnormal neutrophil activation and cell death. In addition, our data revealed that blockage of Nlrp3 inflammasome, through treatments of inhibitor OLT1177 or gene deficiencies in Nlrp3 and caspase 1 expression, protects neutrophils from rEIIIinduced NETosis (Figures 4, 7) and proinflammatory cytokines TNF and IL-1 secretion in mice (Figure 6). These results collectively suggest that Nlrp3 inflammasome is a promising target for treating DENV-induced inflammation.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

ETHICS STATEMENT
The animal study was reviewed and approved by Animal Care and Use Committee of Tzu-Chi University, Hualien, Taiwan (approval ID: 101019).

AUTHOR CONTRIBUTIONS
H-HC conceptualized and supervised this project and wrote this manuscript. T-SL, D-SS, S-CH, and W-SW performed experiments and analyzed the data. All authors contributed to the article and approved the submitted version.