The entire preparation process of extracellular TLR4 domain and MD-2, including construction of expression vectors of sTLR4 and MD-2, removal of endotoxin, preparation and purification of sTLR4/sMD2 complex according to the molar ratio of 1:1, was carried out as described previously [21]. The molar ratio of the sTLR4/sMD-2 complex (1:1) in treated mice was the same as that in treated cells.

Macrophage-like cell line THP-1 was obtained from China Cell Line Bank (Shanghai, China). The cells were cultured in six-well plates (BD Biosciences, Franklin Lakes, NJ, USA) at a density of 5×10⁶ cells/ml. To induce THP-1 cells to the mature macrophages-like state, cells were cultured in RPMI 1640 containing 10% FBS with 100 ng/ml PMA for 24 h and further incubated in the culture medium without PMA for another 24 h. Then, cells were incubated with medium alone (Con), LPS (1.0 µg/ml), LPS + sTLR4 (LPS 1.0 µg/ml +sTLR4 5.0 µg/ml), LPS + sMD-2 (LPS 1.0 µg/ml + sMD-2 1.25 µg/ml), LPS + sTLR4/sMD-2 (LPS 1.0 µg/ml +sTLR4/sMD-2 6.25 µg/ml). After 24 h, cell supernatant was collected and total RNA was extracted.

Total RNAs were extracted from cells using RNAiso Plus reagent according to the manufacturer's instructions. Primers used in our study were listed in Table I. The 25 µl PCR reactions included Premix TaqTM (Ex Taq v.2.0 plus dye; Takara Biotechnology Co., Ltd., Dalian, China), 20 µM primers, cDNA and dd H₂O. The PCR reaction was carried out by the manufacturer's instructions. The electrophoresis image of PCR products were analyzed by the AlphaImager gel analysis system (ProteinSimple, San Jose, CA, USA). Each analysis was repeated three times. The semiquantitative value was expressed as the ration of the integrated optical density, as previously reported (21).

Health male BALB/c mice with the body weight of 22–28 g were obtained from the Experimental Animal Center of Guangxi Medical University. All of the animals were kept in sterile conditions. Mice were randomly divided into five groups (n=5/group), i.e. phosphate buffered saline-treated group (PBS), LPS-treated group (LPS), sTLR4+LPS-treated group (LPS+sTLR4), sMD-2+LPS-treated group (LPS+sMD-2) and sTLR4/sMD-2+LPS-treated group (LPS+sTLR4/sMD-2). Before surgery, all animals were anaesthetized by i.p. with 60 mg/kg sodium pentobarbital (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany). After successful endotracheal intubation, LPS (400 µg/kg in 50 µl of PBS) was instilled intratracheally with sTLR4 (650 µg/kg), sMD-2 (175 µg/kg) or sTLR4/sMD-2 (825 µg/kg). Control mice received intratracheal instillation of sterile PBS alone. The severity of lung injury was evaluated by the wet lung-to-body weight ratio, histopathology changes of the lung tissues and cellular profiles in the bronchoalveolar lavage fluid (BALF) after LPS administration for 16 h. Experimental mice were used strictly in accordance with the national animal experimental requirements, and the study was approved by the Ethical Committee for Animal Experiments from the Fourth Affiliated Hospital of Guangxi Medical University (Liuzhou, China).

After LPS infection for 16 h, mice were euthanized by anesthetic overdose with sodium pentobarbital (120 mg/kg, i.p.). BALF was collected by washing the lungs four times with 0.5 ml of ice-cold saline. Then the cells were centrifuged at 400 g for 10 min at 4°C. The supernatant was frozen at −80°C for further analysis. Subsequently, the sedimentary cells were resuspended in PBS and the total counts of cells were counted using a hemocytometer, and the May-Giemsa staining (Thermo Fisher Scientific, Inc., Waltham, MA, USA) method was used for manual classification count. We only counted the number of inflammatory cells, including neutrophils, monocytes, macrophages and lymphocytes for the total cells in BALF. A professional technician performed cell staining and counting manually. At least 300 cells/slides were counted no less than twice and the mean was calculated in each experiment.

The lung tissue samples were excised and extraneous tissues were cleared away. Then the lung tissue were weighed and the wet lung-to-body weight ratio was calculated to assess the severity of lung inflammation.

To detect the TNF-α, IL-8 and IL-6 protein levels, the cell supernatant and/or BALF of mice were determined by ELISA kits from RayBiotech, Inc., (Norcross, GA, USA). Quantitation of secreted CXCL1 protein levels was analyzed by ELISA kits (R&D Systems, Inc., Minneapolis, MN, USA).

The right inferior lung lobe were fully fixed in 10% neutral buffered formalin for one week, followed by paraffin-embedding in line with the standard procedures. All paraffin-embedded tissues were sectioned to 4 µm thick, and stained with hematoxylin and eosin for examination under Nikon Eclipse E800 microscope (×200). The lung injury score was evaluated by one pathologist in a blinded fashion as previously described (22). The extent of the pathological lesions was scaled from 0 to 3 according to the criteria shown in Table II.

All paraffin-embedded tissues were cut in 4 µm thick for immunohistochemical analysis. 0.3% hydrogen peroxide in methanol was used to block endogenous peroxidase activity. Then sections were treated with 0.1 M citrate buffer (pH 6.0) as an antigen retrieval system. Anti-MPO antibody was used for immunostaining (Santa Cruz Biotechnology, Inc., Dallas, TX, USA). The avidin biotin peroxidase complex method was used to perform immunostaining and the immunohistochemical analyses were performed with the chromogen diaminobenzidine.

Data were presented as mean ± SD. The results were statistically evaluated by one-way analysis of variance followed by the LSD post hoc test. All data were analyzed using with SPSS (v.17.0) software. P<0.05 was considered to indicate a statistically significant difference.

Our laboratory has successfully constructed extracellular TLR4 domain and sMD-2 [21] and we have reported the binding assay of LPS to THP-1 cells in previous paper by Zou et al (21). The fluorescence intensity was significantly decreased in sTLR4/sMD-2 complex treated group instead of sTLR4 or MD-2 alone. These results suggested that recombinant sTLR4/sMD-2 complex competes with TLR4/MD-2 complex on cell surface in binding LPS.

Cytokine microarray analysis of the changes in transcription levels caused by LPS-treated THP-1 cells showed an up-regulation of TNF-α, IL-8, MIP-1α and MIP-1β (23). We examined the expression of TNF-α, IL-8 and CXCL1 in LPS stimulated THP-1 cells after being treated with sTLR4, sMD-2 and sTLR4/sMD-2 complex. After 24 h, there was a significant increase of TNF-α, IL-8 and CXCL1 on mRNA and protein levels in LPS treated group, whereas the mRNA and protein levels of TNF-α, IL-8 and CXCL1 were significantly inhibited in sTLR4/sMD-2 complex treated group (Fig. 1). However, administration of sTLR4 or sMD-2 alone caused inconsistent mRNA expression of TNF-α, IL-8, and CXCL1 (Fig. 1A-C). Furthermore, administration of sTLR4 or sMD-2 alone gave a trend for decreasing the expression of TNF-α, IL-8 and CXCL1 protein induced by LPS (Fig. 1D-F). Of note, the anti-inflammatory effect of sTLR4/sMD-2 complex treatment group was better than that of sTLR4 or sMD-2 alone group. Together, these findings suggested that sTLR4/sMD-2 complex could suppress LPS-induced pro-inflammatory mediators and chemokines production in THP-1 cells.

To evaluate the potential role of sTLR4/sMD-2 complex in the pathological changes of lung in LPS-treated ALI mice, lung histological changes were evaluated after administration of LPS with sTLR4, sMD-2 or sTLR4/sMD-2 complex. The results showed marked inflammatory infiltrate, thick alveolar septum and severe interstitial edema in LPS-induced ALI mice. When sTLR4 or sMD-2 was separately treated, no significant decrease in LPS-induced pulmonary inflammation was observed. In contrast, administration of sTLR4/sMD-2 complex effectively reduced the airway inflammation (Fig. 2A). Furthermore, the semiquantitative histopathology score system was used to evaluate the severity of lung injury as shown in Table II (22). We found that administration of sTLR4/sMD-2 complex could significantly reduce lung injury score (Fig. 2B). The pulmonary index was also significantly decreased in sTLR4/sMD-2 complex treated mice. In contrast, lung injury score and pulmonary index ratio in the sTLR4 or sMD-2 groups did not decrease significantly compared to that in the LPS group (Fig. 2B and C).

Then we evaluated the potential role of sTLR4/sMD-2 complex in the changes of total cells and neutrophil count in BALF of LPS-treated mice. LPS administration resulted in dramatically increased total inflammatory cell count (P<0.05), most of which were neutrophils (Fig. 2D). Notably, we found that sTLR4/sMD-2 complex could significantly decrease the total inflammatory cell and neutrophil count in BALF, whereas sTLR4 or sMD-2 alone did not have the same effect (Fig. 2D).

The proinflammatory cytokines TNF-α, IL-6 and chemokine CXCL1 in BALF were quantified using method of ELISA. We found that the increased expression of TNF-α, IL-6 and CXCL1 in BALF cells were significantly suppressed by sTLR4/sMD-2 complex treatment (Fig. 3A). Administration of sTLR4 or sMD-2 alone decreased the expression of CXCL1 in BALF cells, but the effects were worse than that of sTLR4/sMD-2 complex. However, administration of sMD-2 alone significantly increased the expression of TNF-α in BALF cells, which was not consistent with results in THP-1 cells. We believed that sMD-2 can combine with LPS directly and induce homodimerization of the TLR4/MD-2 complex to trigger a proinflammatory signaling cascade when treated with a high concentration (400 µg/kg in 50 µl of PBS) of LPS in vivo. And TNF-α is an early inflammatory factor, so the expression of TNF-α increases first. However, when we used a low concentration (1 µg/ml) of LPS in vitro, no extra LPS was combined with sMD-2, so the addition of sMD-2 have no effects. Immunohistochemical staining also revealed a large amount of inflammatory cells infiltrated in lung tissue. The production of MPO, which is an indicator of neutrophil infiltration, was also detected in lung tissue. As shown in Fig. 3B and C, the number of MPO-positive expression in the LPS group was greater than that in PBS group (P<0.01). We found a significant reduction of MPO-positive cells in the sMD-2 group compared with the LPS group (P<0.05), whereas sTLR4/sMD-2 complex markedly decreased MPO-positive cells (P<0.01). In summary, sTLR4/sMD-2 complex could effectively protect against ALI induced by LPS.