AICAR decreases acute lung injury by phosphorylating AMPK and upregulating heme oxygenase-1

Abstract

Aim: We investigated the mechanisms by which N1-(β-d-ribofuranosyl)-5-aminoimidazole-4-carboxamide ribonucleotide (AICAR), an activator of AMP-activated protein kinase (AMPK), decreases lung injury and mortality when administered to mice post exposure to bromine gas (Br₂).

Methods: We exposed male C57BL/6 mice and heme oxygenase-1 (HO-1)-deficient (HO-1^-/-) and corresponding wild-type (WT) littermate mice to Br₂ (600 ppm for 45 or 30 min, respectively) in environmental chambers and returned them to room air. AICAR was administered 6 h post exposure (10 mg·kg^-1, intraperitoneal). We assessed survival, indices of lung injury, high mobility group box 1 (HMGB1) in the plasma, HO-1 levels in lung tissues and phosphorylation of AMPK and its upstream liver kinase B1 (LKB1). Rat alveolar type II epithelial (L2) cells and human club-like epithelial (H441) cells were also exposed to Br₂ (100 ppm for 10 min). After 24 h we measured apoptosis and necrosis, AMPK and LKB1 phosphorylation, and HO-1 expression.

Results: There was a marked downregulation of phosphorylated AMPK and LKB1 in lung tissues and in L2 and H441 cells post exposure. AICAR increased survival in C57BL/6 but not in HO-1^-/- mice. In WT mice, AICAR decreased lung injury and restored phosphorylated AMPK and phosphorylated LKB1 to control levels and increased HO-1 levels in both lung tissues and cells exposed to Br₂. Treatment of L2 and H441 cells with small interfering RNAs against nuclear factor erythroid 2-related factor 2 or HO-1 abrogated the protective effects of AICAR.

Conclusions: Our data indicate that the primary mechanism for the protective action of AICAR in toxic gas injury is the upregulation of lung HO-1 levels.

FIGURE 1

Phosphorylated AMP-activated protein kinase (pAMPK) is depleted in Br₂-injured mice lung and rat alveolar type II epithelial (L2) cells. a–d) Representative Western blot images of pAMPK and total AMPK (tAMPK) levels and quantitative pAMPK to tAMPK ratio in lung tissue of mice exposed to Br₂ (600 ppm for 45 min) and returned to room air for the indicated periods of time (a, b) and of rat L2 cells (100 ppm, 10 min). pAMPK level was significantly reduced in mice lung tissue and in L2 cells by 24 h post Br₂ exposure (c, d). e, f) In the groups receiving AICAR 6 h post injury, pAMPK levels were significantly elevated compared to their untreated groups and approached normal (air control) values. Administration of AICAR in the absence of injury in animals and cells produced no significant difference to the air control group. Data are presented as mean±sem, n=5–7 animals per group. Significance was determined by one-way ANOVA followed by Tukey’s post hoc test. Fold air in b, d, f indicates the change relative to the air control.

FIGURE 2

AMP-activated protein kinase (AMPK) activation attenuates lung oedema and inflammation after Br₂ exposure. a) Representative images of haematoxylin and eosin (H&E) staining and myeloperoxidase (MPO) staining of peripheral lung tissue demonstrated proteinaceous and inflammatory cell infiltration into air spaces, and thickening of alveolar septae at 24 h post Br₂ exposure. Lungs from Br₂+AICAR-treated mice did not exhibit altered morphology or inflammatory cell infiltration and were similar in appearance to the control (air-exposed) lungs. Asterisks denote the region of lung magnified in the insert. Scale bars: 100 μm; 50 μm (inset). b) The lung injury score was significantly elevated at 24 h in the Br₂+vehicle group compared to the control and AICAR-treated groups. c) Protein levels were significantly elevated in bronchoalveolar lavage fluid (BALF) at 24 h in the Br₂+vehicle group compared to the control and AICAR-treated groups. d–f) Total cell, neutrophils and macrophage counts were significantly elevated in the Br₂+vehicle group compared to the control and AICAR-treated groups. g) Lung oedema was evaluated by the wet/dry (W/D) ratio, which was significantly elevated at 24 h post Br₂ exposure compared to the control groups, and was alleviated by AICAR treatment. Administration of AICAR in the absence of injury in animals and cells produced no significant difference to the control group. Data are presented as mean±sem, n=5–9 animals per group. Significance was determined by one-way ANOVA followed by Tukey’s post hoc test.

FIGURE 3

AICAR administration improves respiratory function in Br₂-exposed mice. Mice were anaesthetised at 24 h post Br₂ exposure (600 ppm for 45 min) and arterial blood was collected from the abdominal aorta. AICAR administration improved blood gas post Br₂, including a) arterial CO₂ tension (P_aCO2), c) arterial O₂ saturation (S_aO2) and d) arterial O₂ tension (P_aO2). b) AICAR did not improve blood H₊ significantly. e) Br₂ did not cause HCO₃⁻ changes at 24 h. Data are presented as mean±sem, n=5–9 animals per group. Significance was determined by one-way ANOVA followed by Tukey’s post hoc test. f) Male and female C57BL/6 mice were exposed to Br₂ (600 ppm, 45 min) and mortality was assessed over 7 days. AICAR-treated mice showed a significant improvement in survival compared to the vehicle-treated mice (n=8–20 animals per group). Administration of AICAR in the absence of injury in animals and cells produced no significant difference to the air control group. Significant changes in survival were determined by Kaplan–Meier Mantel–Cox log-rank test.

FIGURE 4

Phosphorylated liver kinase B1 (pLKB1) is decreased in lungs and high mobility group box 1 (HMGB1) is increased in plasma in Br₂-exposed mice. Effect of exogenous HMGB1 on LKB1 and AMP-activated protein kinase (AMPK) signalling. a) Representative Western blot images of pLKB1 and total LKB1 levels and b, c) summarised pLKB1 to total LKB1 ratio demonstrated that pLKB1 levels (b) and total LKB1 levels (c) were significantly decreased in the lungs of mice at 24 h post Br₂ exposure. d) ELISA showed that Br₂-exposed mice had significantly higher levels of HMGB1. e) Representative Western blots and f) quantitative analysis from immunoprecipitation, using an antibody to HMGB1 and then pull down protein immunoblotted for LKB1, showed that LKB1 interacts with HMGB1 and forms a complex in mice lung tissue. Data are presented as mean±sem, n=5–7. Significance was determined by unpaired two-tailed t-test. g–j) The ratio of pLKB1 to total LKB1 (g, h) and phosphorylated AMPK (pAMPK) to total AMPK (i, j) in rat alveolar type II epithelial (L2) cells showed that the levels of pLKB1 and pAMPK were significantly decreased after incubation with recombinant HMGB1 (rHMGB1) protein at 10 μg·mL⁻¹ for 24 h. Data are presented as mean±sem, n=6 independent experiments. Significance was determined by unpaired two-tailed t-test. Fold air/saline in b, c, f, h, j indicates the change relative to the air/saline control.

FIGURE 5

AICAR induces heme oxygenase-1 (HO-1) in mice lung tissue and rat alveolar type II epithelial (L2) cells. a) Representative Western blot images and b) quantitative graph of HO-1 in mice lung tissue showing significant upregulation of HO-1 by AICAR. c) Representative Western blot images and d) quantitative graph of HO-1 from L2 cells exposed to air or Br₂ (100 ppm for 10 min) and then returned to room air 24 h later, showing significant upregulation of HO-1 by AICAR. Administration of AICAR in the absence of injury in animals and cells produced no significant difference to the air control group. Data are presented as mean±sem, n=5–6. Significance was determined by one-way ANOVA followed by Tukey’s post hoc test. Fold air in b, d indicates the change relative to the air control.

FIGURE 6

AMP-activated protein kinase (AMPK) activation does not protect HO-1^−/− mice after exposure to Br₂. a) Male and female HO-1^−/− mice and their wild-type (WT) littermates were exposed to Br₂ (600 ppm for 30 min) and mortality assessed over 10 days. AICAR treatment showed no significant protection in HO-1^−/− mice. Significance was determined by Kaplan-Meier Mantel-Cox log-rank test (n=4–11 animals per group). b, c) Representative Western blot of HO-1^−/− mice lung tissue showing a significant increase in AMPK and phosphorylated AMPK (pAMPK) post AICAR administration. Fold air indicates the change relative to the air control. d–f) AICAR did not reduce the levels of bronchoalveolar lavage fluid (BALF) protein (d), total cell count (e) or neutrophil count (f) in HO-1^−/− mice 24 h post Br₂ gas exposure. Data are presented as mean±sem, n=5. Significance was determined by one-way ANOVA followed by Tukey’s post hoc test.

FIGURE 7

AICAR reduces apoptosis and necrosis following exposure of rat alveolar type II epithelial (L2) cells to Br₂ but not in those treated with small interfering RNA (siRNA) against heme oxygenase-1 (HO-1). L2 cells were treated with scrambled RNA (scRNA) or siRNA against HO-1 and then divided into three groups: Air, Br₂ and Br₂+AICAR. a) Representative flow cytometry analysis of Annexin V FITC-stained cells collected 24 h post Br₂ exposure showed a significant decrease in live cells (quadrant (Q) 3), an increase in late apoptotic/necrotic (Q2) cells and an increase in necrotic (Q1) cells. Q4 showed early apoptotic cells. b, c) Administration of a siRNA against HO-1 significantly reduced the levels of HO-1 protein 24 h post Br₂ and in non-exposed L2 cells. AICAR did not increase the protein levels of HO-1 in siRNA-treated cells. The control scRNA did not affect the levels of HO-1. d–g) Quantification of flow cytometry data. In AICAR-treated L2 cells that received the control scRNA, all indicators of cell injury were reduced compared to the vehicle-treated Br₂ groups. Administration of the siRNA against HO-1 did not increase the injury after Br₂ exposure in vehicle-treated cells, but prevented the protective effects of AICAR seen in the scRNA groups. Data are presented as individual data points and mean±sem. Experiments were conducted in triplicate. Significance was determined by one-way ANOVA followed by Tukey’s post hoc test.

FIGURE 8

AICAR reduces apoptosis and necrosis following Br₂ injury in vitro through heme oxygenase-1 (HO-1) in rat alevolar type II epithelial (L2) cells and human club-like epithelial (H441) cells but not in cells treated with small interfering RNA (siRNA) against nuclear factor erythroid 2-related factor 2 (Nrf2). a–d) Quantification of flow cytometry data from L2 cells treated with scrambled RNA (scRNA) or siRNA against Nrf2. In AICAR-treated cells that received scRNA, all indicators of cell injury were reduced compared to the vehicle-treated Br₂ groups. Administration of the siRNA against Nrf2 did not increase the injury after Br₂ exposure in vehicle-treated cells, but prevented the protective effects of AICAR seen in the scRNA groups. Data are presented as individual points and mean±sem. Experiments were replicated in triplicate. Significance was determined by one-way ANOVA followed by Tukey’s post hoc test. e–h) Quantification of flow cytometry data from H441 cells treated with scRNA or siRNA against HO-1 and then divided further into three groups: Air, Br₂ and Br₂+AICAR. In AICAR-treated cells that received the control scRNA, all indicators of cell injury were reduced compared to the vehicle-treated Br₂ groups that received the scRNA. Administration of the siRNA against HO-1 did not increase the injury after Br₂ exposure in vehicle-treated cells, but prevented the protective effects of AICAR seen in the scRNA groups. Data are presented as individual data points and mean±sem. Experiments repeated in triplicate. Significance was determined by one-way ANOVA followed by Tukey’s post hoc test. i, j) Representative Western blot against Nrf2 and HO-1 after Nrf2 siRNA administration and its quantification in L2 cells. Data are presented as mean±sem, n=4. Significance determined by one-way ANOVA followed by Tukey’s post hoc test. k, l) Representative Western blot against HO-1 after HO-1 siRNA administration and its quantification in H441 cells. Data are presented as mean±sem, n=4. Significance determined by one-way ANOVA followed by Tukey’s post hoc test.