Differential regulation of pulmonary endothelial monolayer integrity by varying degrees of cyclic stretch

Abstract

Ventilator-induced lung injury is a life-threatening complication of mechanical ventilation at high-tidal volumes. Besides activation of proinflammatory cytokine production, excessive lung distension directly affects blood-gas barrier and lung vascular permeability. To investigate whether restoration of pulmonary endothelial cell (EC) monolayer integrity after agonist challenge is dependent on the magnitude of applied cyclic stretch (CS) and how these effects are linked to differential activation of small GTPases Rac and Rho, pulmonary ECs were subjected to physiologically (5% elongation) or pathologically (18% elongation) relevant levels of CS. Pathological CS enhanced thrombin-induced gap formation and delayed monolayer recovery, whereas physiological CS induced nearly complete EC recovery accompanied by peripheral redistribution of focal adhesions and cortactin after 50 minutes of thrombin. Consistent with differential effects on monolayer integrity, 18% CS enhanced thrombin-induced Rho activation, whereas 5% CS promoted Rac activation during the EC recovery phase. Rac inhibition dramatically attenuated restoration of monolayer integrity after thrombin challenge. Physiological CS preconditioning (5% CS, 24 hours) enhanced EC paracellular gap resolution after step-wise increase to 18% CS (30 minutes) and thrombin challenge. These results suggest a critical role for the CS amplitude and the balance between Rac and Rho in mechanochemical regulation of lung EC barrier.

Figure 1

Effects of CS preconditioning at 5% and 18% linear distension on thrombin-induced disruption and recovery of pulmonary EC monolayer integrity. A: Human pulmonary ECs were exposed to CS for 2 hours followed by 5-minute and 50-minute thrombin stimulation (50 nmol/L) with continuous CS. Actin cytoskeletal remodeling was examined by immunofluorescent staining with Texas Red-conjugated phalloidin. Intercellular gaps induced by thrombin stimulation are marked by arrows. B: Quantitative analysis of thrombin-induced gap formation in pulmonary ECs exposed to low- and high-magnitude CS was performed as described in Materials and Methods. Maximal gap formation observed in EC monolayers treated with 50 nmol/L of thrombin for 5 minutes was taken as 100%. Shown are representative results of five independent experiments. *P < 0.05.

Figure 2

Focal adhesion remodeling in CS-preconditioned ECs during endothelial monolayer recovery after thrombin stimulation. A: ECs were exposed to 5% or 18% CS for 2 hours followed by stimulation with 50 nmol/L of thrombin for 5 minutes or 50 minutes. Immunofluorescence analysis of focal adhesion remodeling after thrombin stimulation in static cultures (left) and ECs exposed to 5% (middle) or 18% CS (right) was performed using paxillin staining as described in Materials and Methods. B: Higher magnification images of CS-exposed HPAECs after 50 minutes of thrombin stimulation. White arrows show peripheral accumulation of paxillin in ECs preconditioned at 5% CS. Random distribution of paxillin-containing focal adhesions in ECs preconditioned at 18% CS after 50 minutes of thrombin stimulation is marked by filled arrows. Shown are representative results of five independent experiments.

Figure 3

Cortactin distribution in CS-preconditioned ECs during monolayer recovery after thrombin stimulation. Human pulmonary ECs were exposed to 5% CS or 18% CS for 2 hours followed by stimulation with 50 nmol/L of thrombin for 50 minutes. Immunofluorescent detection of cortactin translocation after thrombin stimulation in static cultures and ECs exposed to CS was performed using anti-cortactin antibody. Areas of cortactin accumulation are marked by arrows. Shown are representative results of three independent experiments.

Figure 4

Differential activation of Rac and Rho by 5% and 18% CS. Human pulmonary ECs were exposed to CS at 5% elongation for the indicated periods of time, and Rac (A) and Rho (B) activation was measured as described in Materials and Methods. The graphs below depict results of quantitative analysis of Rac and Rho activation by scanning densitometry of the autoradiography films, normalized to the total Rho or Rac content in the cell lysates, and expressed in relative density units (RDU). Sphingosine 1-phosphate (0.5 μmol/L) and thrombin (50 nmol/L) served as positive controls for Rac and Rho activation, respectively. C: Bar graphs represent comparative analysis of Rac and Rho activation in response to 5% CS and 18% CS stimulation for 30 minutes. Shown are representative results of three independent experiments. *P < 0.05.

Figure 5

Effects of 5% and 18% CS preconditioning on thrombin-induced Rho and Rac activation. A: Human pulmonary ECs were exposed to 5% CS or 18% CS for 2 hours followed by thrombin stimulation (50 nmol/L, 5 minutes) and measurements of Rho activation. B: Static human pulmonary ECs were exposed to thrombin (50 nmol/L) for 5 minutes, 30 minutes, 50 minutes, or left untreated (0 minutes) followed by measurements of Rac activity. C: Human pulmonary ECs were exposed to 5% CS or 18% CS for 2 hours, stimulated with thrombin (50 nmol/L, 50 minutes) and Rac activities were measured during EC recovery phase after thrombin challenge (50 minutes). Rho and Rac activation was assessed by in vitro pulldown assays, quantified by scanning densitometry of the autoradiography films, normalized to the total Rho or Rac content in the cell lysates, and expressed in RDU. Shown are representative results of three independent experiments. *P < 0.05.

Figure 6

Effects of Rac protein depletion on the monolayer integrity in HPAECs exposed to CS. A: Cells grown in D35 plastic dishes were incubated with siRNA to Rac1 or treated with nonspecific RNA oligonucleotide duplexes for 48 hours, and Rac protein depletion was examined by immunoblotting with corresponding antibody. Control blots represent β-actin expression in ECs treated with siRNA. B: HPAECs grown on BioFlex plates were incubated with siRNA to Rac1 (bottom row) or treated with nonspecific RNA duplexes (top row) for 48 hours as described in Materials and Methods. Cells were next left static (left) or were exposed to 2-hour 5% CS (middle) or 18% CS (right). Actin cytoskeletal remodeling was evaluated by immunofluorescent staining with Texas Red-conjugated phalloidin. Paracellular gaps are shown by arrows. Enlarged images of cell interface areas (insets) illustrate significant paracellular gaps formed in CS-stimulated EC monolayers with depleted Rac1 (shown by arrow), as compared to cells treated with nonspecific RNA duplexes. Shown are representative results of three independent experiments.

Figure 7

Effects of Rac inhibition on monolayer recovery in HPAECs exposed to CS and thrombin. A: HPAECs were grown on BioFlex plates. Cells exposed to 5% CS were preincubated with Rac inhibitor NSC-23766 (200 μmol/L, 1 hour) or vehicle. After a total of 2 hours of CS exposure (with or without inhibitor), ECs were stimulated with thrombin (50 nmol/L) for 5 minutes, and Rac activation was assessed by in vitro pulldown assays, as described in Materials and Methods. Shown are representative results of three independent experiments. B: Cells exposed to 5% CS or 18% CS were preincubated with Rac inhibitor NSC-23766 (200 μmol/L, 1 hour) or vehicle. After a total of 2 hours of CS exposure (with or without inhibitor), HPAECs were stimulated with thrombin (50 nmol/L) for 5 minutes (left row) and 50 minutes (right row). Actin rearrangement was analyzed by immunofluorescent staining with Texas Red-conjugated phalloidin. Paracellular gaps are shown by arrows. Shown are representative results of three independent experiments.

Figure 8

Effects of long-term physiological CS preconditioning on EC monolayer disruption and recovery induced by 18% CS and thrombin. A: Static EC cultures, or CS-preconditioned cells (24 hours, 5% distension) were either immediately treated with thrombin (50 nmol/L) or exposed to 30 minutes of 18% CS before thrombin stimulation. F-actin was visualized after 5 minutes (middle row) and 50 minutes (bottom row) of thrombin treatment by immunofluorescent staining with Texas Red phalloidin. Intercellular gaps are marked by arrows. B: Quantitative analysis of thrombin-induced gap formation in static and CS-preconditioned pulmonary ECs on thrombin stimulation was performed as described in Materials and Methods. Shown are representative results of three independent experiments. *P < 0.05.

Figure 9

Effects of chronic CS preconditioning and thrombin on Rho and Rac activation. After 24-hour preconditioning at 5% CS (A) or 18% CS (B), human pulmonary ECs were challenged with 50 nmol/L of thrombin for 5 minutes or 50 minutes without interruption of CS stimulation, and Rho activation assays were performed as described in Materials and Methods. HPAEC cultures grown on BioFlex plates without CS and stimulated with thrombin for 5 minutes or 50 minutes served as static controls. C: Human pulmonary ECs preconditioned at 5% CS or 18% CS for 24 hours were challenged with vehicle or thrombin (50 nmol/L) for 50 minutes without interruption of CS stimulation, and Rac activation assays were performed. The graphs depict results of quantitative analysis of Rho and Rac activation by scanning densitometry of the autoradiography films, normalized to the total Rho or Rac content in the cell lysates, and expressed in RDU. Shown are representative results of three independent experiments. *P < 0.05.

Figure 10

Hypothetical mechanism of magnitude-dependent regulation of agonist-induced EC permeability by CS. Physiological CS stimulates Rac by activating putative Rac-specific guanine nucleotide exchange factor (GEF). Rac activation triggers focal adhesion peripheral redistribution, induces cortactin peripheral translocation, activation of cortical actin polymerization, and thus accelerates re-establishment of EC monolayer integrity after thrombin challenge. In turn, high-magnitude CS suppresses Rac activities and enhances Rho-mediated signaling, which leads to increased EC barrier disruption and delays barrier recovery induced by thrombin via Rho-dependent activation of stress fiber formation, actomyosin contraction, and random focal adhesion redistribution in thrombin-challenged lung EC monolayers.