Figure 1.
Alterations of gene-expression following cyclic mechanical stretch exposure of pulmonary epithelial cells—An in vitro model for ALI.
Confluent Calu-3 or A549 cells were plated on collagen-coated BioFlex plates and underwent cyclic mechanical stretch for 24 h at 30% maximum, 0.7% stretch minimum, and sine wave 5 s on, 5 s off. (A) Pathway analysis. Microarray technology was used to assess transcriptional changes in stretch versus control conditions. (B) Calu-3 pulmonary epithelia exposed to different time periods of 30% stretch. HIF1A protein levels determined by Western blot. One representative blot of three is displayed. (C) HIF reporter plasmid consisting of four tandem HIF1A enhancer sequences from the 3′-region of the erythropoietin gene transfected into pulmonary epithelia (A549) and exposed to 30% stretch (n = 3). (D) Calu-3 pulmonary epithelia exposed to different degrees of cyclic mechanical stretch. HIF1A protein levels determined by Western blot. One representative blot of three is displayed. (E) Human alveolar epithelial cells (HPAEpiC) were exposed to different time-periods of stretch. Representative Western blots are shown following 30% stretch. (F) Transcript level of the glycolytic enzyme LDHA from HIF1A KD Calu-3 pulmonary epithelia (lentivirus mediated HIF1A knockdown, see Figure S2) or Calu-3 epithelia transduced with a control virus (lentiviral scrambled siRNA, Scr) after 24 h of stretch were determined by real-time RT-PCR relative to housekeeping gene beta-actin (mean ± s.d., n = 3). (G) Lactate levels in supernatants obtained from Calu-3 with a lentiviral-mediated HIF1A KD or controls (Scr) after 24 h of 30% stretch exposure. Controls consisted of Calu-3 cells treated with lentiviral scrambled siRNA exposed to the same experimental conditions (n = 4). (H) Calu-3 controls (Scr Calu-3; transfected with lentiviral scrambled siRNA) or Calu-3 HIF1KD were exposed to 24 h of 30% stretch with media containing the 13C-glucose isotope. For nuclear magnetic resonance (NMR) analysis of metabolites, cells were shock frozen immediately after stretch. Rate of glycolysis is given as ratio of incorporated 13C intermediates of glycolysis compared to 13C glucose levels (n = 4).
Figure 2.
Mechanisms of cyclic mechanical stretch-mediated HIF1A stabilization.
(A) Confluent Calu-3 cells underwent cyclic mechanical stretch for 24 h. Partial oxygen pressures (pO2) from supernatants were determined using I-STAT analyzer and compared to un-stretched controls (mean ± s.d., n = 3). (B) Following 8 or 24 h of cyclic mechanical stretch exposure, mitochondrial fractions of Calu-3 cells were obtained and analyzed for succinate dehydrogenase (SDH, mitochondrial Complex II) activity using ELISA. Activity is given as OD (optical density) change over time (mean ± s.d., n = 3). (C) siRNA knockdown of succinate-CoA ligase (SUCLG). RT PCR for SUCLG from A549 cells after siRNA treatment compared with control siRNA treatment (CsiR; beta-actin was used as house-keeping gene; n = 3). Note: siRNA knockdown revealed 82% reduction of SUCLG transcript. (D) A549 cells following siRNA repression of succinate-CoA ligase (siSUCLG) or treatment with nonspecific control siRNA (CsiR) were stretched for 0 or 8 h, lysed, and blotted for HIF1A. One representative blot of three is displayed. (E) Confluent Calu-3 cells underwent cyclic mechanical stretch for 8 or 24 h with and without α-ketoglutarate (aKG) treatment. Cells were lysed and blotted for HIF1A. One representative blot of three is displayed. (F–H) Confluent Calu-3 cells underwent cyclic mechanical stretch for 8 h with or without inhibitors of JNK, ERK, or p38. Mitochondrial fractions of Calu-3 cells were obtained and analyzed for succinate dehydrogenase (SDH, mitochondrial Complex II) activity using ELISA. (I) Confluent Calu-3 cells underwent cyclic mechanical stretch for 8 or 24 h with and without neutralizing antibodies for TNFA or IL-6. Cells were lysed and blotted for HIF1A. One representative blot of three is displayed.
Figure 3.
Functional consequences of cyclic mechanical stretch mediated HIF1A stabilization.
(A) Calu-3 controls (Scr Calu-3; treated with lentiviral scrambled siRNA) or Calu-3 HIF1KD were exposed to 24 h of stretch with media containing the 13C-glucose isotope. For nuclear magnetic resonance (NMR) analysis of metabolites, cells were shock frozen immediately after stretch. Rate of tricarboxylic acid cycle (TCA) flux is given as ratio of incorporated 13C intermediates of the TCA compared to 13C glucose levels. (B) After 24 h of cyclic mechanical stretch, mitochondrial fractions of Calu-3 cells were obtained and analyzed for mitochondrial Complex IV (COX4) activity using ELISA. Activity is given as OD (optical density) change over time (mean ± s.d., n = 3). (C) ATP levels from Calu-3 controls (Scr Calu-3) or Calu-3 HIF1KD exposed to 24 of stretch. (D) Pyruvate dehydrogenase (PDH) activity levels from Calu-3 controls (Scr Calu-3) or Calu-3 HIF1KD exposed to 8 or 24 h of stretch. (E) Hydrogen peroxide levels of supernatants from Calu-3 controls or Calu-3 cells after 24 h at 30% stretch and 20% O2. The H2O2 data are expressed as the mean fluorescence levels from two independent experiments normalized by protein concentration. Error bars represents s.d. (n = 3). (F–G) IL-6 and TNF-α levels were evaluated in supernatants from stretched Calu-3 controls or Calu-3 HIF1KD exposed to 24 h of stretch using a human enzyme linked immunosorbent assay (ELISA). Results are presented as mean ± s.d. (n = 3).
Figure 4.
Normoxic HIF1A stabilization during ALI.
(A) Wild-type mice (BL6C57) were exposed to ventilator induced lung injury (VILI; pressure-controlled mechanical ventilation at an inspiratory pressure of 45 mbar with an inspired oxygen concentration of 100%, exposure time 180 minutes). To examine the influence of mechanical ventilation on pulmonary HIF1A expression patterns, lungs were stained with antibodies for Hif1a. IgG controls were used at identical concentrations and staining conditions as the target primary antibodies (magnification ×400; n = 4). (B) Frozen lung tissue was lysed and proteins resolved by SDS-PAGE. Resultant Western blots were probed with anti-Hif1a antibody. A representative blot of three is displayed. (C) Quantitative analysis of Western blot in (B) assessed by densitometry. (D) Imaging HIF1A using previously described HIF reporter mice expressing luciferase upon HIF stabilization (“ODD-Luc mice”) during ALI. Prior to imaging, mice were injected with i.p. luciferin (50 mg/kg) and mice were euthanized. Left column (control): Lungs excised without mechanical ventilation. To induce ALI, mice were ventilated with pressure controlled ventilation (45 mbar) at 21% oxygen concentration over 3 h. Middle column, 3 h at 15 mbar; right column, 3 h at 35 mbar. Color bar indicates photons/(cm2·s·steradian) with minimum and maximum threshold values (n = 4). (E) “ODD-Luc mice” were exposed to VILI (45 mbar, 3 h) using 20% or 100% inspired oxygen. Lungs were harvested, homogenized, and analyzed for luciferase gene expression using the Dual-Luciferase Reporter Assay System from Promega. (F) After 3 h of ventilation at 45 mbar, mitochondrial fractions obtained from lung tissue were analyzed for succinate dehydrogenase (SDH, mitochondrial Complex II) activity using ELISA. Activity is given as OD (optical density) change over time (mean ± s.d., n = 3).
Figure 5.
Functional consequences of HIF1A activation during ALI.
(A–E) HIF1A activator dimethyl-oxaloylglycine (DMOG) during ALI: BL6C57 mice were treated with 1 mg DMOG or vehicle control 4 h prior to the experimental procedure. (A) Mechanical ventilation was instituted and mice were ventilated for 0 or 180 min using pressure-controlled settings (inspiratory pressure of 45 mbar, 100% inspired oxygen concentration). Albumin concentration in the bronchoalveolar fluid (BAL) was determined by enzyme linked immunosorbent assay (n = 6). (B and C) Pulmonary neutrophil sequestration was quantified using a myeloperoxidase (MPO) assay. MPO activity was assessed using a murine ELISA from lung tissue (B) or BAL (C) (mean ± s.d., n = 6). (D) To assess pulmonary gas exchange, blood gas analyses were performed by obtaining arterial blood via cardiac puncture. Analysis was performed immediately and the ratio of the arterial partial pressure of oxygen (PaO2) to the fraction of inspired oxygen (FiO2) was determined (mean ± s.d., n = 6). (E) Mechanical ventilation was instituted and mice were ventilated using pressure-controlled settings (inspiratory pressure of 35 mbar, 100% inspired oxygen concentration) until a cardiac standstill was observed in the surface electrocardiogram (n = 6).
Figure 6.
Functional consequences of HIF1A inhibition during ALI.
(A–E) HIF1A inhibition by Echinomycin during VILI: BL6C57 mice were treated with 30 µg Echinomycin i.p. or vehicle control 1 h prior to the experimental procedure. (A) Mechanical ventilation was instituted and mice were ventilated for 0 or 180 min using pressure-controlled settings (pressure-controlled mechanical ventilation at an inspiratory pressure of 45 mbar with an inspired oxygen concentration of 100%, exposure time 180 min). Albumin concentration in the bronchoalveolar fluid (BAL) was determined by enzyme linked immunosorbent assay (n = 6). (B and C) Pulmonary neutrophil sequestration was quantified using a myeloperoxidase (MPO) assay. MPO activity was assessed using a murine ELISA from lung tissue (B) or BAL (C) (n = 6). Results are presented as mean ± s.d. and derived from six animals in each condition. (D) To assess pulmonary gas exchange, blood gas analyses were performed by obtaining arterial blood via cardiac puncture. Analysis was performed immediately and the ratio of the arterial partial pressure of oxygen (PaO2) to the fraction of inspired oxygen (FiO2) was determined (mean ± s.d., n = 6). (E) Mechanical ventilation was instituted and mice were ventilated using pressure-controlled settings (inspiratory pressure of 35 mbar, 100% inspired oxygen concentration) until a cardiac standstill was observed in the surface electrocardiogram (n = 6).
Figure 7.
Tissue-specific function of HIF1A during ALI.
Hif1af/f Actin-Cre+ mice (A), Hif1af/f CadherinCre+ (B), Hif1af/f LysozymCre+ (C), Hif1af/f ClaraCellCre+ (D), or Hif1af/f SurfactantCre+ (E) or corresponding age, weight, and gender-matched littermate controls (Cre expressing mice) were exposed to ventilator-induced lung injury (VILI; pressure-controlled mechanical ventilation at an inspiratory pressure of 35 mbar with an inspired oxygen concentration of 100%). Survival time was determined by measuring the time until a cardiac standstill was observed in the surface electrocardiogram. Note significantly attenuated survival of Hif1af/f Actin-Cre+ mice or Hif1af/f SurfactantCre+ (A, E, mean ± s.d., p<0.05, n = 6). Albumin concentration in the bronchoalveolar fluid was determined by enzyme-linked immunosorbent assay (ELISA; pressure-controlled mechanical ventilation at an inspiratory pressure of 45 mbar with an inspired oxygen concentration of 100%, exposure time 180 min). Note significantly increased albumin concentration in the bronchoalveolar fluid of Hif1af/f Actin-Cre+ mice or Hif1af/f SurfactantCre+ (A, E, mean ± s.d., p<0.05, n = 6).
Figure 8.
Functional consequences of alveolar HIF1A deletion during ALI.
(A–J) Hif1af/f SurfactantCre+ mice or age, gender, and weight-matched littermate controls (SurfactantCre+) were exposed to ventilator-induced lung injury (VILI; pressure-controlled mechanical ventilation at an inspiratory pressure of 45 mbar with an inspired oxygen concentration of 100%, exposure time 180 min). (A) Isolation of alveolar epithelial cells from Hif1af/f SurfactantCre+ or control mice after VILI exposure. Frozen cells were lysed and proteins resolved by SDS-PAGE. Resultant Western blots were probed with anti-Hif1a antibody. A representative blot of three is displayed. (B) Pulmonary neutrophil accumulation was quantified using myeloperoxidase (MPO) ELISA. (C–E) IL-6, KC, or TNF-α, levels were evaluated in lung tissue homogenates using a mouse-enzyme-linked immunosorbent assay (ELISA). Results are presented as mean ± s.d. (n = 6). (F) For quantification of histological tissue damage by ventilator-induced lung injury following 180 min ventilation, VILI scores were assessed in Hif1af/f SurfactantCre+ or corresponding littermate control mice (SurfactantCre+). Results are displayed as median and range (n = 4). (G) One of four representative photomicrographs (×200) stained with hematoxylin/eosin is displayed. (H–J) Hif1af/f SurfactantCre+ or matched controls (Surfactant Cre+) were treated with the pharmacologic HIF activator dimethyl-oxaloylglycine DMOG (1 mg/mouse i.p. 4 h period to VILI) and subsequently exposed to VILI. (H) Albumin concentration in the bronchoalveolar fluid by enzyme-linked immunosorbent assay (ELISA), (I) pulmonary gas exchange by the ratio of the arterial partial pressure of oxygen (PaO2) to the fraction of inspired oxygen (FiO2), and (J) MPO activity by using a murine ELISA from lung tissue (mean ± s.d.; n = 6).
Figure 9.
Functional role of alveolar epithelial HIF1A on glycolysis during ALI.
(A–D) Hif1af/f SurfactantCre+ mice or littermate controls (SurfactantCre+) matched in age, weight, and gender were exposed to ventilator-induced lung injury (VILI; pressure-controlled mechanical ventilation at an inspiratory pressure of 45 mbar with an inspired oxygen concentration of 100%, exposure time 120 min). 13C glucose was administered i.p. 30 min prior to the experimental procedure. Determination of 13C glucose and 13C carbohydrates during VILI was performed using liquid chromatography–tandem mass spectrometry (LC-MS). Glut 1 transcript level was determined by real-time RT-PCR relative to house-keeping gene beta-actin and expressed as fold induction relative to sham-operated controls (mean ± SD, n = 3). (A) 13C glucose. (B) Glut-1 transcript levels. (C) 13C fructose 1,6 bisphosphate. (D) 13C lactate. (E–H) To examine the functional role of glycolysis, wild-type mice treated with the glycolysis inhibitor 2-deoxy-D-glucose (200 mg/kg dose, i.p.) 30 min prior to the experimental procedure, and subsequently exposed to VILI (see above). Albumin concentration in the bronchoalveolar fluid (E) by enzyme-linked immunosorbent assay (ELISA), pulmonary gas exchange (F) by the ratio of the arterial partial pressure of oxygen (PaO2) to the fraction of inspired oxygen (FiO2), and MPO activity by using a murine ELISA from lung tissue (G) or IL-6 levels (H) in lung tissue homogenates using a mouse enzyme-linked immunosorbent assay (ELISA). Results are presented as mean ± s.d. (n = 6). (I and J) Mechanical ventilation was instituted and SurfactantCre+ controls or Hif1af/f SurfactantCre+ mice with and without 2-DG treatment were exposed to VILI (pressure-controlled mechanical ventilation at an inspiratory pressure of 35 mbar with an inspired oxygen concentration of 100%) until a cardiac standstill was observed in the surface electrocardiogram (p<0.05, n = 6).
Figure 10.
Consequences of alveolar-epithelial Hif1a deletion on mitochondrial function during ALI.
(A–G) Hif1af/f SurfactantCre+ mice or littermate controls (SurfactantCre+) matched in age, weight, and gender were exposed to ventilator-induced lung injury (VILI; pressure-controlled mechanical ventilation at an inspiratory pressure of 45 mbar with an inspired oxygen concentration of 100%, exposure time 120 min). 13C glucose was administered i.p. 30 min prior to the experimental procedure. Determination of 13C carbohydrates during VILI was performed using liquid chromatography–tandem mass spectrometry (LC-MS). (A) 13C malate. (B) Tricarboxylic acid (TCA) cycle flux rates were determined by measuring the ratio of 13C2 glutamate to total creatine (n = 4 for all experiments). (C–F) Hif1af/f SurfactantCre+ mice or age-, gender-, and weight-matched littermate controls (SurfactantCre+) were exposed to VILI (see above). (C) After 3 h of VILI exposure, mitochondrial fractions were obtained from lung tissue and analyzed for Complex IV activity using ELISA. Activity is given as OD (optical density) change over time (mean ± s.d., n = 3). (D) Frozen lung tissue was lysed and mitochondrial proteins resolved by SDS-PAGE. Resultant Western blots were probed with anti-COX4-2 antibody. A representative blot of three is shown. (E) Pulmonary ATP content in Hif1af/f SurfactantCre+ mice or littermate controls (SurfactantCre+) were exposed to VILI (inspired oxygen concentration 100%, exposure time 120 min, pressure-controlled ventilation with an inspiratory pressure of 45 mbar). (F) Hydrogen peroxide lung tissue levels in Hif1af/f SurfactantCre+ mice or littermate controls (SurfactantCre+) exposed to VILI. Data are expressed as the mean fluorescence levels from three independent experiments normalized by protein concentration. Error bars represent s.d. (n = 3). (G and H) BL6C57 mice were treated with Rotenone—an inhibitor of mitochondrial respiration—or vehicle control 1 h prior to the beginning of the experimental procedure. IL-6 levels in lung tissue homogenates (G) or Albumin levels in BAL (H) were evaluated using a mouse enzyme-linked immunosorbent assay (ELISA) following exposure to VILI (see above). Results are presented as mean ± s.d. (n = 6).
Figure 11.
Proposed model of ALI-associated HIF1A stabilization.
(1) During ALI, stretch of pulmonary epithelial cells results in the inhibition of succinate dehydrogenase (SDH), leading to normoxic stabilization of alveolar epithelial HIF1A. (2) Alveolar epithelial HIF1A results in increased glycolytic capacity, TCA flux, and optimized mitochondrial respiration via induction of Complex IV. (3) HIF1A-dependent prevention of mitochondrial dysfunction during ALI is associated with increased alveolar epithelial capacity to produce ATP, while concomitantly preventing ROS accumulation and attenuating lung inflammation.