Inhibition of Glycogen Synthase Kinase 3β Blocks Mesomesenchymal Transition and Attenuates Streptococcus pneumonia-Mediated Pleural Injury in Mice

Abstract

Pleural loculation affects about 30,000 patients annually in the United States and in severe cases can resolve with restrictive lung disease and pleural fibrosis. Pleural mesothelial cells contribute to pleural rind formation by undergoing mesothelial mesenchymal transition (MesoMT), whereby they acquire a profibrotic phenotype characterized by increased expression of α-smooth muscle actin and collagen 1. Components of the fibrinolytic pathway (urokinase plasminogen activator and plasmin) are elaborated in pleural injury and strongly induce MesoMT in vitro. These same stimuli enhance glycogen synthase kinase (GSK)-3β activity through increased phosphorylation of Tyr-216 in pleural mesothelial cells and GSK-3β mobilization from the cytoplasm to the nucleus. GSK-3β down-regulation blocked induction of MesoMT. Likewise, GSK-3β inhibitor 9ING41 blocked induction of MesoMT and reversed established MesoMT. Similar results were demonstrated in a mouse model of Streptococcus pneumoniae-induced empyema. Intraperitoneal administration of 9ING41, after the induction of pleural injury, attenuated injury progression and improved lung function (lung volume and compliance; P < 0.05 compared with untreated and vehicle controls). MesoMT marker α-smooth muscle actin was reduced in 9ING41-treated mice. Pleural thickening was also notably reduced in 9ING41-treated mice (P < 0.05). Collectively, these studies identify GSK-3β as a newly identified target for amelioration of empyema-related pleural fibrosis and provide a strong rationale for further investigation of GSK-3β signaling in the control of MesoMT and pleural injury.

Copyright © 2017 American Society for Investigative Pathology. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1

Glycogen synthase kinase (GSK)-3β expression is enhanced in nonspecific pleuritis. Lung tissue sections from patients diagnosed with nonspecific pleuritis were immunofluorescently labeled. A: GSK-3β expression (red) increases in the pleural mesothelial cells (calretinin; green) of the nonspecific pleuritis tissue. GSK-3β and calretinin colocalization increases in nonspecific pleuritis lung sections (merged; yellow). B: The pleural mesothelium of patients diagnosed with nonspecific pleuritis demonstrates increased GSK-3β (green) and α-smooth muscle actin (α-SMA; red) compared with normal lung sections. GSK-3β and α-SMA colocalization (merged; orange) increases in nonspecific pleuritis lung sections. Arrows indicate the mesothelium. n = 30 fields per slide; n = 3 to 4 patients per group. Scale bars = 50 μm. Original magnification, ×40.

Figure 2

Glycogen synthase kinase (GSK)-3β mobilizes to the nucleus of human primary mesenchymal cells (HPMCs) undergoing mesothelial mesenchymal transition (MesoMT). Serum-starved HPMCs were treated with phosphate-buffered saline (PBS), 5 ng/mL transforming growth factor (TGF)-β, 7 nmol/L plasmin (PLN), or 20 nmol/L urokinase plasminogen activator (uPA) (for 24 and 48 hours). A: PBS-, TGF-β–, and uPA-treated cells (24 hours) were immunofluorescently labeled for GSK-3β (green) and nuclei (red). Colocalization of GSK-3β and nuclei appear orange. Arrows indicate the location of the nucleus. B: PBS-, TGF-β–, plasmin-, and uPA-treated HPMCs were lyzed and resolved by SDS-PAGE. Lysates were probed for α-smooth muscle actin (α-SMA), total GSK-3β, and Tyr-216–phosphorylated GSK-3β (pGSKT) by Western blot analysis. Akt was used as loading control. n = 30 fields per slide (A); n = 3 slides per treatment (A); n = 2 independent experiments (B).

Figure 3

Glycogen synthase kinase (GSK)-3β down-regulation attenuates the induction of mesothelial mesenchymal transition (MesoMT). Untransfected, control siRNA, and GSK-3β siRNA–transfected cells were serum-starved for 24 hours. Cells were then treated with phosphate-buffered saline (PBS) or 5 ng/mL transforming growth factor (TGF)-β for 48 hours. A: Conditioned media and cell lysates were resolved by SDS-PAGE and immunoblotted for collagen (Col)-1, plasminogen activator inhibitor (PAI)-1, α-smooth muscle actin (α-SMA), and GSK-3β by Western blot analysis. β-Actin was used as loading control. B: Total RNA was isolated from untransfected, control siRNA–, and GSK-3β siRNA–transfected cells that had been treated with TGF-β for 24 hours. Changes in α-SMA, Col-1, and GSK-3β mRNA levels were determined by real-time quantitative PCR (qPCR) analyses. GAPDH was used as the reference gene. C: Total RNA was isolated from untransfected, control siRNA– and GSK-3α siRNA–transfected cells that had been treated with TGF-β for 24 hours. Changes in α-SMA, Col-1, and GSK-3α mRNA levels were determined by qPCR analyses. GAPDH was used as the reference gene. Data are expressed as means ± SEM. n = 3 independent experiments. ∗P < 0.05; †P < 0.05 versus PBS control; ‡P < 0.05 versus TGF-β controls. Cont, control.

Figure 4

Glycogen synthase kinase (GSK)-3β inhibition with 9ING41 blocks mesothelial mesenchymal transition (MesoMT) induction. A–C: Human primary mesenchymal cells (HPMCs) were treated with varying doses of 9ING41 (10 to 0.05 μmol/L) in serum-free media for 24 hours before the addition of 5 ng/mL transforming growth factor (TGF)-β (A), 7 nmol/L plasmin (PLN) (B), or 20 nmol/L urokinase plasminogen activator (uPA) (C). Cells were then allowed to incubate for 48 hours. Conditioned media and lysates were then resolved by SDS-PAGE and immunoblotted for collagen (Col)-1 and α-smooth muscle actin (α-SMA) by Western blot analysis. β-Actin was used as loading control. D: HPMCs were treated with varying doses of 4-benzyl-2-methyl-1,2,4-thiadiazolidine-3,5-dione (TDZD-8; 20 to 1 μmol/L) in serum-free media for 24 hours before the addition of 5 ng/mL TGF-β. Conditioned media and lysates were then resolved by SDS-PAGE and immunoblotted for Col-1 and α-SMA by Western blot analysis. Akt was used as loading control. n = 2 to 3 independent experiments. PBS, phosphate-buffered saline.

Figure 5

The glycogen synthase kinase (GSK)-3β inhibitor 9ING41 reverses established mesothelial mesenchymal transition (MesoMT). Serum-starved human primary mesenchymal cells (HPMCs) were treated with 5 ng/mL transforming growth factor (TGF)-β for 24 hours. A: Varying doses of 9ING41 (10 to 0.5 μmol/L) were then added to the TGF-β–treated cells and allowed to incubate for 48 hours. Conditioned media and lysates were resolved by SDS-PAGE and immunoblotted for collagen (Col)-1 and α-smooth muscle actin (α-SMA). β-Actin was used as loading control. B: For real-time quantitative PCR (qPCR) analyses varying doses of 9ING41 (10 to 0.5 μg/mL) were added to TGF-β–treated cells and then allowed to incubate for 24 hours. Total RNA was then isolated and transcribed into cDNA. α-SMA and Col-1 expression was determined by qPCR analyses. GAPDH served as the reference gene. Data are expressed as means ± SEM. n = 2 to 3 independent experiments (A); n = 3 independent experiments (B). ∗P < 0.05 versus TGF-β treatment. PBS, phosphate-buffered saline.

Figure 6

Glycogen synthase kinase (GSK)-3β inhibition by 9ING41 reduces Tyr-216 phosphorylation of GSK-3β and phosphorylation of NF-κB and Smad2. A: Serum-starved human primary mesenchymal cells (HPMCs) were treated with phosphate-buffered saline (PBS), 7 nmol/L plasmin (PLN), and 20 nmol/L urokinase plasminogen activator (uPA) in the presence or absence of 10 μmol/L 9ING41. Cell lysates were then resolved by SDS-PAGE and immunoblotted for Tyr-216 phosphorylated GSK-3β (p-GSKT), total GSK-3β, and phosphorylated p65 (p-p65). β-Actin was used as loading control. B: Serum-starved HPMCs were treated with PBS and 5 ng/mL transforming growth factor (TGF)-β in the presence and absence of 9ING41 (10 to 1 μmol/L) for 48 hours. Conditioned media and cell lysates were then resolved by SDS-PAGE and immunoblotted for collagen (Col)-1, phosphorylated Smad2 (p-Smad2), total Smad2, and α-smooth muscle actin (α-SMA). β-Actin was used as loading control. n = 2 independent experiments.

Figure 7

Streptococcus pneumoniae–mediated pleural injury is attenuated by treatment with 9ING41 A: Serum-starved murine primary mesenchymal cells (MPMCs) were treated with 5 ng/mL transforming growth factor (TGF)-β in the presence or absence of 10 μmol/L 9ING41. Cell lysates were then resolved by SDS-PAGE and immunoblotted for α-smooth muscle actin (α-SMA). Akt was used as loading control. B: C57Bl/6J mice were intrapleurally injected with saline or S. pneumoniae (1.8 × 108 cfu). After 24 hours, mice were either left untreated or treated with dimethyl sulfoxide (vehicle) or 30 mg/kg 9ING41 for 6 days. At the completion of the 7-day course, lung compliance was determined using the Scireq flexivent. Lung renditions were then collected by computed tomographic scan to determine lung volumes. C: Lung tissue sections from vehicle- and 9ING41-treated mice were Trichrome stained and images were taken. Pleural thicknesses were then measured and compared. Arrows indicate the mesothelium. D and E: Pleural sections from vehicle- and 9ING41-treated mice were immunostained for the mesothelial cell marker calretinin (green) and α-SMA (red; D) or Tyr-216–phosphorylated glycogen synthase kinase (GSK)-3β (Tyr-216-P) (E) and imaged by confocal microscopy. Arrows indicate the mesothelium. Colocalization of α-SMA and calretinin is orange. Data are expressed as means ± SEM. n = 2 (A); n = 6 to 10 mice per treatment (B); n = 30 fields per slide per mouse (C); n = 6 to 10 animals per treatment (C); n = 30 fields per mouse (D and E); n = 3 mice per treatment (D and E). ∗P < 0.05 versus saline; †P < 0.05 versus no treatment; ‡P < 0.05 versus vehicle. Scale bar = 50 μm (C). Original magnification: ×20 (C); ×40 (D and E). Trx, treatment.