Receptor for advanced glycation end-products is a marker of type I cell injury in acute lung injury

Abstract

Rationale: Receptor for advanced glycation end-products (RAGE) is one of the alveolar type I cell-associated proteins in the lung.

Objectives: To test the hypothesis that RAGE is a marker of alveolar epithelial type I cell injury.

Methods: Rats were instilled intratracheally with 10 mg/kg lipopolysaccharide or hydrochloric acid. RAGE levels were measured in the bronchoalveolar lavage (BAL) and serum in the rats and in the pulmonary edema fluid and plasma from patients with acute lung injury (ALI; n = 22) and hydrostatic pulmonary edema (n = 11).

Main results: In the rat lung injury studies, RAGE was released into the BAL and serum as a single soluble isoform sized approximately 48 kD. The elevated levels of RAGE in the BAL correlated well with the severity of experimentally induced lung injury. In the human studies, the RAGE level in the pulmonary edema fluid was significantly higher than the plasma level (p < 0.0001). The median edema fluid/plasma ratio of RAGE levels was 105 (interquartile range, 55-243). The RAGE levels in the pulmonary edema fluid from patients with ALI were higher than the levels from patients with hydrostatic pulmonary edema (p < 0.05), and the plasma RAGE level in patients with ALI were significantly higher than the healthy volunteers (p < 0.001) or patients with hydrostatic pulmonary edema (p < 0.05).

Conclusion: RAGE is a marker of type I alveolar epithelial cell injury based on experimental studies in rats and in patients with ALI.

Figure 1.

RAGE protein levels in the bronchoalveolar lavage (BAL) and serum in the rat hydrochloric acid (HCl) instillation lung injury model. Rats were divided into three groups: (1) an HCl group, in which rats were intratracheally instilled with 4 ml/kg of 0.1 N HCl and mechanically ventilated for 2 h (n = 7); (2) a saline group, in which rats were instilled with 4 ml/kg of saline intratracheally and mechanically ventilated for 2 h (n = 3); and (3) a control group that was mechanically ventilated without fluid instillation (n = 3). (A) Protein recovery in the BAL. (B) Wet-to-dry ratio of the lung. (C) Immunoblot assay using anti-RAGE polyclonal rabbit antibody. After 0.1 N HCl, RAGE was detected in the BAL (10 μl) and in serum (10 μg protein) as an approximately 48 kD band. (D) Quantification of RAGE level in the BAL by dot blot analysis. Results are expressed in arbitrary units of RAGE antigen concentration expressed as μg/ml of recombinant RAGE-Fc chimera protein equal to that in the sample. Instillation of 0.1 N HCl caused significantly higher RAGE level in the BAL as well as protein recovery in the BAL (A) and lung wet-to-dry ratio (B). *p < 0.05.

Figure 2.

The relationship between the severity of lung injury and abundance of RAGE in the BAL and serum in the rat HCl lung injury model. The volume of instilled HCl (0.1 N) was increased from 1 to 4 ml/kg (n = 3 in each setting except 4 ml/kg [n = 7]), and the severity of lung injury and the RAGE levels were evaluated. (A) BAL protein recovery. (B) Wet–dry ratio of the lung. (C) RAGE abundance in the BAL (10 μl) and in serum (10 μg protein) demonstrated by representative Western blot. (D) Quantification of RAGE level in the BAL by the dot blot analysis. Results are expressed in arbitrary units of RAGE antigen concentration expressed as μg/ml of recombinant RAGE-Fc chimera protein equal to that in the sample. Instillation of 4 ml/kg HCl resulted in significantly higher level of RAGE than the control level. *p < 0.05 versus control.

Figure 3.

The relationship between the severity of lung injury and abundance of RAGE in the BAL and serum in the rat HCl lung injury model. The concentration of instilled HCl (4 ml/kg) was increased from 0 N (saline without HCl) to 0.1 N (n = 3 in each setting except 0.1 N [n = 7]), and the severity of lung injury and the RAGE levels were evaluated. (A) BAL protein recovery. (B) Wet-to-dry ratio of the lung. (C) RAGE abundance in the BAL (10 μl) demonstrated by representative Western blot. (D) Quantification of RAGE level in the BAL by the dot blot analysis. Results are expressed in arbitrary units of RAGE antigen concentration expressed as μg/ml of recombinant RAGE-Fc chimera protein equal to that in the sample. Instillation of 0.1 N HCl resulted in a significantly higher level of RAGE than the control level. *p < 0.05 versus saline.

Figure 4.

RAGE expression in a rat LPS-induced lung injury model. Rats were divided into an LPS group (n = 3) and a control group (n = 3). In the LPS group, LPS (10 mg/kg) in 250 μl of saline was instilled intratracheally. The same volume of saline was instilled in control animals. Instillation of LPS into the distal airspaces increased protein recovery in the BAL (10 μl) (A), the lung wet-to-dry weight ratio (B), and the level of RAGE in the BAL (C). *p < 0.05 versus saline.

Figure 5.

RAGE expression in the normal rat lung and in the LPS-induced lung injury model. RAGE-positive cells (red) were observed on the alveolar septa, but they did not colocalize with p180 lamellar body protein–positive cells (alveolar type II epithelial cells; green). Intratracheal instillation of LPS (10 mg/kg) resulted in edematous alveolar septa, but the pattern of RAGE positivity did not change compared with control conditions. White scale bar indicates 10 μm.

Figure 6.

RAGE expression in the normal rat lung and in the LPS-induced lung injury model. RAGE-positive cells (red) were observed on the alveolar septa, but they did not colocalize with CD68-positive cells (alveolar macrophages; green). Intratracheal instillation of LPS (10 mg/kg) resulted in edematous alveolar septa, but the pattern of RAGE positivity was not changed compared with control conditions. White scale bar indicates 10 μm.

Figure 7.

(A) RAGE expression in the BAL and lung homogenate of rat. The immunoblot assay using anti-RAGE polyclonal rabbit antibody demonstrates lung homogenate and its membrane fraction contained three bands sized approximately 48, 50, and 55 kD, whereas the soluble fraction of lung homogenate and BAL demonstrated a single band sized approximately 48 kD. (B) RAGE expression in the pulmonary edema fluid and lung homogenate of human lung. The immunoblot assay using anti-RAGE monoclonal antibody demonstrates that lung homogenate and its membrane fraction contained three bands sized approximately 48, 50, and 55 kD, whereas the soluble fraction of lung homogenate and the pulmonary edema fluid demonstrated a single band sized approximately 48 kD. Homo, homogenate; Mem, membrane fraction; Sol, soluble fraction.

Figure 8.

RAGE levels in pulmonary edema fluid and plasma. (A) In the pulmonary edema fluid, RAGE levels in the patients with acute lung injury/acute respiratory distress syndrome (ALI/ARDS) were significantly higher than the RAGE levels in the patients with hydrostatic pulmonary edema (HYDRO; p < 0.05). (B) In plasma, RAGE levels in the ALI/ARDS group were higher than in the healthy volunteers (p < 0.001). The median levels of plasma RAGE levels in ALI/ARDS were higher than hydrostatic pulmonary edema group (p < 0.05). Data expressed as median (horizontal line), with 25th to 75th confidence intervals in the boxes and the 10th to 90th percentiles in the whiskers.