Fig 2: NMI is released by activated macrophages. a NMI in the supernatant (Sup) and cytoplasmic fraction of human THP1 cells (ATCC TIB-202™) were analyzed by immunoblotting post S. typhimurium stimulation, using high-mobility group box protein 1 (HMGB1), l-lactate dehydrogenase B 1 (LDHB1), and N-myc as control. ß-actin in the cell and bovine serum albumin (BSA) in the supernatant were used to show equal loading. BSA was shown using commassie blue staining. b NMI in the supernatant (Sup) and cytoplasmic fraction of THP1 cells were analyzed by immunoblotting. The cells were pretreated with (+) or without (-) Brefeldin A for 8 h and stimulated with (+) or without (-) S .typhimurium for 4 h. c NMI released by mouse RAW264.7 cells (ATCC TIB-71™) (2 × 106 cells ml-1) was analyzed by ELISA at different time points post LPS (1 µg ml-1) stimulation. Error bars indicate ± s.e.m. from three biological replicates. d Mice (n = 5 for each group) were intraperitoneally injected with 1 × 104 CFU live S. typhimurium per mouse and the concentration of the NMI in the serum were determined by ELISA. Data are presented as the mean ± s.e.m. **P < 0.01. e, f NMI in the sera of mice (n = 5 for each group) was determined by ELISA 3, 6 or 24 h after intraperitoneal injection of LPS (10 mg kg-1 or 20 mg kg-1) or acetaminophen (APAP) (300 mg kg-1). Data are presented as the mean ± s.e.m. **P < 0.01. g ELISA analysis of NMI in the serum of 24 patients and 8 healthy individuals. Patients succumbed to severe inflammation were labeled with solid cycles. Significance in d and e were tested by unpaired Student’s t-test. Significance in g was tested by Mann–Whitney U-test. **P < 0.01

Fig 3: Recombinant NMI protein induces TNF release and NF-?B activation. a, b Tumor necrosis factor (TNF) and interleukin 6 (IL-6) in the sera of mice (n = 5) were determined by ELISA, 6 or 24 h after intravenous injection of recombinant mNMI (10 mg kg-1) or PBS. c, d TNF and IL-6 released by mouse RAW264.7 cells (ATCC TIB-71™), 4 h post-stimulation with increasing concentrations of mNMI. e, f TNF and IL-6 released by RAW264.7 cells stimulated with 1 µg ml-1 mNMI. g TNF released by RAW264.7 cells stimulated with mNMI (5 µg ml-1) or LPS (100 ng ml-1) for 8 h, with (+) or without (-) pre-treatment of polymyxin B (PMB) (25 µg ml-1), heat (80 °C for 30 min) and trypsin (5 µg ml-1, 37 °C overnight). h I?B, p-I?B, NF-?B (p50 and p65), phosphorated p65 at Ser536 (p-p65 (S536)) in the cytoplasm and NF-?B (p50 and p65) protein in nucleus extracts of RAW264.7 cells were assessed by immunoblotting, after the cells were treated with 5 µg ml-1 recombinant mNMI protein. ß-actin in the cytoplasm and LaminB in the nucleus were used as control. LaminB in the cytoplasm was detected to ensure that there was no contamination during the fractionation. Error bars in c–g indicate ± s.e.m. from three biological replicates. Significance was tested by unpaired Student’s t-test test. *P < 0.05, **P < 0.01

Fig 4: Nanoemulsion uptake and impact on macrophage viability in vitro. PBS-based PFC-NE uptake when exposed to RAW 264.7 (ATCC TIB 71) macrophage cell line. Microscopy images were taken on EVOS Imager at 40X Magnification. White bar on the lower right corner represents 100 µm A Image of macrophage morphology when exposed to PFC-NE under the White Light channel. B Image of macrophages’ nuclei when exposed to PFC-NE, and treated with Hoechst stain under the DAPI channel. Cell nucleus can be observed in blue. C Image of macrophages’ cytoplasm when exposed to NE under the Red Fluorescent Channel. D Macrophage cell viability upon 24 h exposure of NE with 10 µM and 25 µM of Near infrared (NIRF) dye DiR concentration. Cellular viability was assessed using CellTiter-Glo® 2.0. The error bar represents the standard deviation from six independent measurements

Fig 5: IFP35 serves as a DAMP and stimulates the inflammatory responses. a, b Immunoblotting a and ELISA analysis b of IFP35 in the supernatant and cytoplasmic fraction of mouse RAW264.7 cells (ATCC TIB-71™), treated with LPS (1 µg ml-1). c, d Mice (n = 5 for each group) were intraperitoneally injected with LPS from E. coli (100 ng ml-1, c) or live S. typhimurium (1 × 104 CFU per mouse, d). The serum concentrations of IFP35 in mice were determined by ELISA. Data are presented as the mean ± s.e.m. Significance was tested by unpaired Student’s t-test (**P < 0.01). e TNF level in the supernatants of human THP1 cells (ATCC TIB-202™) was analyzed by ELISA 8 h post incubation with LPS (100 ng ml-1) or different purified proteins (5 µg ml-1 respectively). Data are presented as the mean ± s.e.m of three independent experiments. **P < 0.01. f The luciferase activity of HEK293T cells (ATCC CRL-11268™), transiently transfected with the genes of hTLR4–MD2–CD14 and NF-?B promoter with luciferase, and stimulated with 5 µg ml-1 hIFP35?N for 4 h (in the presence or absence of 25 µg ml-1 polymyxin B (PMB)). 100 ng ml-1 LPS was administrated as positive control. Data are presented as the mean ± s.e.m. of three independent transfections and are given as the ratio of luciferase activity relative to unstimulated cells. Significance in d and e was tested by one-way ANOVA followed by Student–Newman–Keuls test. **P < 0.01. g The survival rate of wild type (n = 14) and Ifp35 -/- mice (n = 14) intraperitoneally injected with LPS (50 µg kg-1) and d-gal (1 g kg-1). Significance was tested by log-rank test.*P < 0.05