Fig 2: CLEC5A and TLR2 mediate DV-activated and DV-activated platelet-induced NET formation. a Human neutrophils were pre-treated with isotype control (300 µg/ml) or anti-CLEC5A mAb, followed by incubation with platelets and DV for 3 h at 37 °C. b Alternatively, neutrophils were pre-incubated with anti-CLEC5A mAb and anti-TLR2 mAb, followed by incubation with DV-activated platelets (MOI = 5) or aggretin-activated platelets at 37 °C for 3 h. c Neutrophils from WT (clec5a+/+tlr2+/+), clec5a-/-tlr2+/+, clec5a+/+tlr2-/-, and clec5a-/-tlr2-/- mice were incubated with WT platelets in the presence of DV (NGC-N) for 3 h at 37 °C. NETs were measured by immunofluorescence staining (µm2). Data are mean ± SEM from at least three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001 (Student’s t-test). Source data are provided as Source Data file

Fig 3: DV-induced NETs contribute to changes in vascular permeability. a and b Neutrophils were incubated with DV, EVs (from 8 × 106 platelets, including both MVs and EXOs), or DV-EVs (from 8 × 106 DV-activated platelets) and overlaid on HMEC-1 monolayers in presence or absence of DNase I (100 U/ml) at 37 °C for 3 h (a). Alternatively, neutrophils were pretreated with isotype control, anti-CLEC5A mAb, and anti-TLR2 mAb at 37 °C for 1 h, followed by incubation with DV or EVs (b). These assays used 1.5 × 108 EVs per experiment (a and b). Endothelial monolayer permeability was measured by determining HRP activity in media from the lower chamber. c Stat1-/- mice were challenged by intravenous administration of DV, in the presence or absence of DNase I, followed by injecting Evans blue (i.v.) at 1 h before sacrifice on day 5 post-infection (n = 4 in saline (vehicle) group, n = 3 in DNase I group). d Stat1-/- (n = 6 in saline and isotype control group, n = 4 in anti-TLR2 mAb group) and stat1-/-clec5a-/- (n = 5 in saline group, n = 4 in isotype control and anti-TLR2 mAb group) mice were pre-treated with anti-TLR2 mAb or isotype control, followed by inoculation with DV (NGC-N) via intraperitoneal route. Evans blue was injected intravenously at day 5 post-infection, 1 h before mice were sacrificed. Plasma leakage of visceral organs was determined by measuring absorbancies at 610 nm for each organ. Data are mean ± SEM from at least three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001 (Student’s t-test). Source data are provided as Source Data file

Fig 4: EVs play critical roles in the pathogenesis of viral infectons. DV activates macrophages and neutrophils via CLEC5A. DV also stimulate CLEC2 on platelets to promote release of EVs (DV-EXOs and DV-MVs), which further enhance the secretion of proinflammatory cytokines and NET formation in neutrophils via CLEC5A and TLR2

Fig 5: DV stimulates platelets to release EVs that induce NET formation via CLEC2. a Human platelets were incubated with DV (PL046) at 37 °C for 1 h in the presence or absence of anti-CLEC2 mAb, followed by harvesting of EVs by centrifugation for flow cytometry analysis. b Human platelets were incubated with DV (MOI = 0.5) or aggretin (3 nM) at 37 °C for 1 h, followed by harvesting EVs by ultracentrifugation. Both activated platelets and PLT-EVs were incubated with neutrophils and NET formation was observed. c Human platelets (8 × 106) were preincubated with isotype control or anti-CLEC2 mAb for 15 min before DV (MOI = 0.5) stimulation, followed by harvesting EVs to stimulate human neutrophils and observation of NET formation. d Neutrophils (8 × 105) were pre-pretreated with isotype control, anti-CLEC5A mAb, or anti-TLR2 mAb at 37 °C for 1 h, followed by incubation with EVs isolated from DV-PLTs (8 × 106) and aggretin-PLTs at 37 °C for 3 h. NETs were measured by immunofluorescence staining (µm2). Data are mean ± SEM from at least three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001 (Student’s t-test). Source data are provided as Source Data file