Anionic pulmonary surfactant phospholipids inhibit inflammatory responses from alveolar macrophages and U937 cells by binding the lipopolysaccharide-interacting proteins CD14 and MD-2

Abstract

Lipopolysaccharide (LPS), derived from Gram-negative bacteria, is a major cause of acute lung injury and respiratory distress syndrome. Pulmonary surfactant is secreted as a complex mixture of lipids and proteins onto the alveolar surface of the lung. Surfactant phospholipids are essential in reducing surface tension at the air-liquid interface and preventing alveolar collapse at the end of the respiratory cycle. In the present study, we determined that palmitoyl-oleoyl-phosphatidylglycerol and phosphatidylinositol, which are minor components of pulmonary surfactant, and synthetic dimyristoylphosphatidylglycerol regulated the inflammatory response of alveolar macrophages. The anionic lipids significantly inhibited LPS-induced nitric oxide and tumor necrosis factor-alpha production from rat and human alveolar macrophages and a U937 cell line by reducing the LPS-elicited phosphorylation of multiple intracellular protein kinases. The anionic lipids were also effective at attenuating inflammation when administered intratracheally to mice challenged with LPS. Binding studies revealed high affinity interactions between the palmitoyl-oleoyl-phosphatidylglycerol and the Toll-like receptor 4-interacting proteins CD14 and MD-2. Our data clearly identify important anti-inflammatory properties of the minor surfactant phospholipids at the environmental interface of the lung.

FIGURE 1.

Anionic phospholipids inhibit inflammatory mediator production induced by LPS. Liposomes composed of SM, PE, DPPC, PS, POPG, and PI were formed by bath sonication for 30 min at room temperature. LPS (10 ng/ml) and different concentrations of phospholipids were added to monolayer cultures of differentiated U937 cells (A) or primary rat alveolar macrophages (B). At 6 h after stimulation, media were collected, and secreted TNF-α levels were determined in U937 cultures. NO production was determined 24 h after stimulating rat alveolar macrophages. LPS stimulation in the absence of phospholipid was set as 100%. The data shown are the means ± S.E. from three separate experiments with duplicate samples in each experiment. The average TNF-α production upon LPS stimulation was (8.0 ± 0.54 ng/ml). The average NO production upon LPS stimulation was 12.17 ± 0.27 μm.

FIGURE 2.

The inhibitory effect of phosphatidylglycerols on LPS-induced inflammatory mediator production is molecular species-specific. PG liposomes were formed by bath sonication for 30 min at room temperature. LPS (10 ng/ml) and different concentrations of PG were added to monolayer cultures of differentiated U937 cells (A) or rat alveolar macrophages (B). Media TNF-α measurements were performed 6 h after stimulation. Media NO measurements were performed 24 h after stimulation. LPS stimulation without PG was set at 100%. The molecular species of PG shown on the graph are: 8:0, dioctanoyl-phosphatidylglycerol; 12:0, dilauroyl-phosphatidylglycerol (DLPG); 14:0, DMPG; 16:0, DPPG; 18:0, distearoyl-phosphatidylglycerol; and 16:0/18:1, POPG. The data shown are the means ± S.E. from three separate experiments with duplicate samples in each experiment. The average TNF-α production upon LPS stimulation was 11.3 ± 0.7 ng/ml. The average NO production upon LPS stimulation was 10.1 ± 0.6 μm.

FIGURE 3.

Homotypic PG containing liposomes are most effective in antagonizing LPS action in the presence of surfactant lipids. Surfactant lipid (SL) and POPG were dried under nitrogen and hydrated at 37 °C for 1 h. A, surfactant lipid and POPG were mixed in organic solvents prior to drying and hydrating, and subsequently, liposomes were produced. B, surfactant lipid and POPG were made as independent populations of liposomes that were subsequently mixed prior to macrophage treatment. 10 ng/ml LPS and different concentrations of liposome mixtures were added to monolayer cultures of differentiated U937 cells. 6 h after stimulation, media were collected, and TNF-α production was determined. LPS stimulation without phospholipid was set as 100%. The data shown are the means ± S.E. from three separate experiments with duplicate samples in each experiment. The average TNF-α production upon LPS stimulation was 7.0 ± 0.2 ng/ml.

FIGURE 4.

POPG inhibits LPS-induced MAPK and IκBα phosphorylation and MKP-1 expression. A, POPG liposomes (200 μg/ml) were added to monolayer cultures of differentiated U937 cells that received either no treatment or 10 ng/ml LPS. After incubating for the indicated times, cells were lysed using buffer containing detergent, protease inhibitors, and phosphatase inhibitors. Aliquots with 15 μg of protein from lysates were separated by SDS-PAGE and transferred onto nitrocellulose membranes. The amount of phosphorylation was detected using phospho-specific antibodies to p38 MAPK, p42/p44 ERK, p46-p54 JNK, and phosphorylated IκBα. To determine equal loading of proteins between samples, the membranes were probed with rabbit polyclonal p46 JNK, p42/p44 ERK, p38 MAPK, and IκBα antibodies. The expression of MKP-1 was detected with a polyclonal MKP-1 antibody. B, POPC and DPPG liposomes (200 μg/ml) were compared with POPG liposomes (200 μg/ml) as antagonists of LPS activation of cells using the same conditions as described for panel A. The time of analysis following LPS exposure for p38, ERK, and JNK was 30 min, and that for IκBα and MKP-1 was 60 min.

FIGURE 5.

Quantification of POPG inhibition of LPS-induced signaling. Western blot analysis as described in the legend for Fig. 4 was performed three or four times on separate samples, and the intensity of phospho-p38 (P-p38), phospho-IκBα (P-IκBα), phospho-ERK (P-ERK), phospho-JNK (P-JNK), total IκBα (T-IκBα), and MKP-1 was calculated using NIH Image J1.34 software. Significance is as follows: *, p < 0.05, **, p < 0.01, ***, p < 0.001 when compared between LPS treatment and LPS plus POPG treatment. No PLs, no phospholipids.

FIGURE 6.

POPG, DMPG, and PI antagonize the effects of LPS on primary human alveolar macrophages. Human alveolar macrophages were isolated from healthy volunteer BALF and plated onto a 96-well plate. 2 days after plating, 10 ng/ml LPS and 20 μg/ml phospholipids were added to monolayer cultures of human alveolar macrophages. 6 h after stimulation, media were collected, and TNF-α production was determined by ELISA. LPS stimulation without phospholipid was set at 100%. The data shown are the means ± S.E. from three separate experiments with duplicate samples in each experiment. The average TNF-α secretion after LPS stimulation was 30.7 ± 15.1 ng/ml. Significance is as follows: *, p < 0.05, **, p < 0.01, ***, p < 0.001 when compared with LPS stimulation in the absence of POPG.

FIGURE 7.

Anionic phospholipids (PLs) modulate lung inflammation induced by intratracheally administered LPS. A mixture of LPS (1 μg) and phospholipids (30 μg) in 20 μl of PBS was sprayed into murine trachea using a MicroSprayer^TM aerosolizer. At 18 h after stimulation, lungs were lavaged via the trachea. TNF-α production (A) was determined by ELISA. The number of leukocytes was counted, and differential cell counts (B) were determined from at least 300 cells on cytocentrifuged preparations. Mouse KC (C) and MIP-2 (D) secretion were determined using Quantikine kits (R&D Systems). The data shown are the means ± S.E. from six to eight mice. Significance is as follows: *, p < 0.05, **, p < 0.01, when compared between LPS and LPS plus POPG.

FIGURE 8.

Anionic phospholipids (PLs) modulate lung inflammation induced by intravenously administered LPS. Phospholipids were dried under nitrogen and hydrated, and liposomes were formed using a Liposofast^TM. The phospholipids (50 μg) in 20 μl of PBS were sprayed into murine trachea using a MicroSprayer^TM aerosolizer. At the same time, LPS (50 μg) in 200 μl of PBS was intravenously administered to mice. 3 h after stimulation, lungs were lavaged via the trachea. TNF-α production (A) was determined by ELISA. The number of leukocytes was counted, and differential cell counts (B) were determined from at least 300 cells on cytocentrifuged preparations. Mouse KC (C) and MIP-2 (D) levels were determined using Quantikine kits (R&D System). The data shown are the means ± S.E. for six to eight mice. Significance is as follows: *, p < 0.05, **, p < 0.01, when compared between LPS and LPS plus POPG.

FIGURE 9.

CD14 binds to solid phase lipids. A, affinity-purified preparations (2 μg each) of the extracellular domains of CD14 (sCD14) and TLR4 (sTLR4) and the full-length MD-2 were analyzed by gel electrophoresis under reducing and denaturing conditions. The electrophoretic gels were stained with Coomassie Blue. mol. mass st., molecular mass standards. B, phospholipids (1.25 nmol) in 20 μl of ethanol were placed onto microtiter wells, and the solvent was evaporated. Nonspecific binding was blocked with 20 mm Tris buffer (pH 7.4) containing 0.15 m NaCl, 5 mm CaCl₂ (in the upper panel), or 2 mm EGTA (in the lower panel) and 5% (w/v) bovine serum albumin (buffer A). Varying concentrations of human CD14 in buffer A were added and incubated at 37 °C for 1 h. The binding of CD14 to phospholipids was detected using anti-CD14 monoclonal antibody as described under “Experimental Procedures.” The data shown are the means ± S.E. from three separate experiments with duplicate samples in each experiment.

FIGURE 10.

PG Inhibits CD14 binding to solid phase LPS. A, various types of PG were coated onto microtiter plates and incubated with CD14 (1 μg/ml) at 37 °C for 1 h. The binding of CD14 to PG was detected using anti-CD14 monoclonal antibody, and the ELISA-based absorbance of CD14 bound to POPG was defined as 100%. Types of PG shown on the graph are: dilauroylphosphatidylglycerol (DLPG), DMPG, DPPG, and 16:0/18:1 POPG. B, LPS (2 μg) in 20 μl of ethanol was placed onto microtiter wells, and the solvent was evaporated. After blocking the nonspecific binding with buffer A, the mixture of CD14 (1 μg/ml) and phospholipid liposomes (20 μg/ml) in buffer A, which was preincubated at 37 °C for 1 h, was added and incubated at 37 °C for 1 h. The binding of CD14 to LPS was detected using anti-CD14 monoclonal antibody. The ELISA-based absorbance of CD14 bound to LPS was defined as 100%. The data shown are the means ± S.E. from three separate experiments with duplicate samples in each experiment. *, p < 0.05, **, p < 0.01, when compared with LPS-CD14 binding in the absence of phospholipids.

FIGURE 11.

Monoclonal antibodies specific for the LPS binding site inhibit CD14 interaction with POPG and PI. POPG (A) or PI (B) were coated onto microtiter plates. After blocking the nonspecific binding with buffer A, the mixture of CD14 (1 μg/ml) and monoclonal antibodies or isotype control IgG (50 μg/ml) in buffer A, which was preincubated at 37 °C for 1 h, was added and incubated at 37 °C for 1 h. The binding of CD14 to phospholipids was detected using sheep anti-CD14 polyclonal antibody, and the ELISA-based absorbance of CD14 bound to phospholipid was defined as 100%. The data shown are the means ± S.E. from three separate experiments with duplicate samples in each experiment. *, p < 0.05, when compared with CD14 binding in the absence of monoclonal antibody. mIG, mouse Ig. C, CD14 (2 μg) was coated onto microtiter plates, and nonspecific binding was blocked with buffer A. Monoclonal antibodies or isotype control IgG (50 μg/ml) in buffer A were added and incubated at 37 °C for 1 h. The CD14 was detected using sheep anti-CD14 polyclonal antibody, and the ELISA-based absorbance of solid phase CD14 alone was defined as 100%.

FIGURE 12.

MD-2 preferentially binds POPG. A, POPG (1.25 nmol) was placed onto microtiter wells, and the solvent was evaporated. After blocking nonspecific binding with buffer A, MD-2, sTLR4, and PstB2 (1 μg/ml) in buffer A were added and incubated at 37 °C for 1 h. The binding of recombinant proteins to POPG was detected using anti-His antibody. B, phospholipids (1.25 nmol) were placed onto microtiter wells, and the solvent was evaporated. After blocking, MD-2 (1 μg/ml) in buffer A was added and incubated at 37 °C for 1 h. The binding of MD-2 to phospholipids was detected using anti-His antibody. The ELISA-based absorbance of MD-2 bound to POPG was defined as 100%. The data shown are the means ± S.E. from three separate experiments each with duplicate determinations.

FIGURE 13.

POPG disrupts MD-2 interaction with TLR4. sTLR4 (100 ng) was adsorbed onto microtiter wells. After blocking nonspecific binding with buffer A, the mixture of MD-2 (1 μg/ml) and phospholipid liposomes (20 μg/ml) (A) or different concentrations of phospholipids (B) in buffer A, which was preincubated at 37 °C for 1 h, was added and incubated at 37 °C for 2 h. The binding of MD-2 to sTLR4 was detected using horseradish peroxidase-conjugated anti-V5 monoclonal antibody. The ELISA-based absorbance of MD-2 binding without phospholipids was defined as 100%. The data shown are the means ± S.E. from three separate experiments each with duplicate determinations. *, p < 0.05, **, p < 0.01, when compared with MD-2-sTLR4 binding in the absence of phospholipids.