Cigarette smoke disrupts monolayer integrity by altering epithelial cell-cell adhesion and cortical tension

Abstract

Chronic obstructive pulmonary disease (COPD) is a major cause of morbidity and mortality. Cigarette smoke (CS) drives disease development and progression. The epithelial barrier is damaged by CS with increased monolayer permeability. However, the molecular changes that cause this barrier disruption and the interaction between adhesion proteins and the cytoskeleton are not well defined. We hypothesized that CS alters monolayer integrity by increasing cell contractility and decreasing cell adhesion in epithelia. Normal human airway epithelial cells and primary COPD epithelial cells were exposed to air or CS, and changes measured in protein levels. We measured the cortical tension of individual cells and the stiffness of cells in a monolayer. We confirmed that the changes in acute and subacute in vitro smoke exposure reflect protein changes seen in cell monolayers and tissue sections from COPD patients. Epithelial cells exposed to repetitive CS and those derived from COPD patients have increased monolayer permeability. E-cadherin and β-catenin were reduced in smoke exposed cells as well as in lung tissue sections from patients with COPD. Moreover, repetitive CS caused increased tension in individual cells and cells in a monolayer, which corresponded with increased polymerized actin without changes in myosin IIA and IIB total abundance. Repetitive CS exposure impacts the adhesive intercellular junctions and the tension of epithelial cells by increased actin polymer levels, to further destabilize cell adhesion. Similar changes are seen in epithelial cells from COPD patients indicating that these findings likely contribute to COPD pathology.

Keywords: COPD; E cadherin; cigarette smoke; cortical tension; epithelial barrier.

Fig. 1.

Cigarette smoke (CS) increases epithelial monolayer permeability. A: in normal human airway epithelial cells, one CS exposure did not increase epithelial monolayer permeability as measured by FITC-dextran assay, but 2 CS exposures (2 cigarettes, 2 h rest followed by 2 additional cigarettes) did cause a statistically significant increase in epithelial monolayer permeability (*P = 0.008, n = 4 patients, with 3 technical replicates per patient). B: epithelial cells derived from COPD patients have a more permeable monolayer compared with normal human airway epithelial cells (*P = 0.02) and have a further increase in monolayer permeability following one in vitro exposure to CS (*P = 0.008) (n = 4 patients, with 3 technical replicates per patient).

Fig. 2.

Cigarette smoke alters adherens junction proteins. A, left: representative Western analysis showing a reduction in total E-cadherin protein in NHBE cells following 3 exposures to whole CS in the Vitrocell smoke chamber. A, right: densitometry measurement of E-cadherin levels, *P < 0.05, n = 10. B: Western analysis showing a reduction in total β-catenin in NHBE cells following 3 exposures to whole CS in the Vitrocell smoke chamber. n = 3 patients. C, left: representative Western analysis showing reduction in total E-cadherin in HBEC-3KT cells following 3 exposures to 10% CSE. C, right: densitometry measurement of E-cadherin levels, *P < 0.05, n = 10. D: representative Western analysis showing no change in total β-catenin in HBEC-3KT cells after 3 exposures to 10% CSE. n = 3. E: immunofluorescence of NHBE cells shows that in normal cells E-cadherin is enriched along the basolateral membrane (arrow); representative image is shown, n = 8. Following 2 exposures to whole CS, there is a reduction of E-cadherin along this surface (arrow). In addition, there is evidence of actin stress fibers across the apical surface of the epithelium after CS exposure (arrow). Scale bar, 20 µm.

Fig. 3.

E-cadherin is reduced in COPD patients. A, left: representative Western analysis showing a reduction in total E-cadherin protein in a COPD patient compared with a patient with no history of COPD. Cell pellets were obtained from bronchial brushings. A, right: densitometry measurement of E-cadherin levels, n = 4 patient replicates. *P < 0.05. B: immunohistochemistry of lung tissue sections demonstrating that in normal individuals E-cadherin is enriched along the lateral surface. In tissue sections from patients with COPD, E-cadherin staining is reduced along this surface. C: in tissue sections from patients with COPD, β-catenin staining is less prominent in the epithelium of COPD patients compared with lung sections from normal subjects. n = 8 patients/condition were analyzed for B and C. Scale bar, 50 µm.

Fig. 4.

Repetitive CS exposure increases the tension of individual cells and cells in a monolayer. A: stiffness of HBEC-3KT cells in a monolayer is determined using MTC in which ferromagnetic beads are applied to the monolayer surface and displacement of beads is measured after a sinusoidal twisting force is applied to the beads. Following short-term exposure to CSE, cells in an epithelial monolayer have a statistically significant decrease in stiffness compared with cells in untreated monolayers. In contrast, following prolonged CSE exposure, cells within the epithelial monolayer show a statistically significant increase in stiffness compared with untreated controls (*P < 0.005). B: MPA was used to measure the cortical tension of individual cells. Using a micropipette of a set radius (inset), force is applied to the surface of the cell causing a portion of the cell membrane to pull into the pipette. Scale bar, 10 µm. Following short-term exposure to CSE, cortical tension decreases ~2-fold in smoke-exposed cells compared with unexposed cells (*P < 0.005). In contrast, following prolonged exposure to CSE, cortical tension increases ~2-fold in smoke exposed cells compared with unexposed cells (**P < 0.005). C: the cortical tension of cells increases in a near linear fashion with cell passage. Cortical tension measurements were normalized to the calculated value at each passage number based upon the linear fit to the entire data set.

Fig. 5.

Cigarette smoke extract does not alter total amount of myosin II. A, left: Western analysis showing no change in total myosin IIA following one exposure to CSE (n = 3). A, right: Western analysis showing no change in total myosin IIA following 3 exposures to CSE (n = 3). B, left: Western analysis showing no change in total myosin IIB following one exposure to CSE (n = 3). B, right: Western analysis showing no change in total myosin IIB following 3 exposures to CSE (n = 3).

Fig. 6.

Cigarette smoke causes cytoskeletal rearrangement. A: immunofluorescence of NHBE cells shows that myosin IIB is enriched along the basolateral membrane and apical stress fibers in control cells and following 2 CS exposures (arrows). B: following short-term exposure to CSE, the actin polymer fraction was reduced slightly, but this change was not statistically significant (P = 0.13, n = 14 controls, 18 CSE, representative Western to the right showing soluble and insoluble fractions of actin from a given sample. Total actin is calculated as the summation of soluble and insoluble fraction). C: in contrast, following exposure to prolonged CSE, the actin polymer fraction increased ~50% increase (*P = 0.01, n = 15 controls, 17 CSE; representative Western to the right showing soluble and insoluble fractions of actin from a given sample. Total actin is calculated as the summation of soluble and insoluble fraction). D: immunofluorescence imaging of lung sections from patients with COPD showing an overall reduction in myosin IIB and myosin IIC compared with tissue sections from patients without COPD. Despite the overall decrease, myosin IIB appears to be relatively well preserved in the apical subcompartment. In contrast, myosin IIA does not change in its amount or distribution (representative images depicted, n = 8 patients/condition). Scale bar, 20 µm.

Fig. 7.

E-cadherin and cigarette smoke have opposing effects on the cell’s cortical tension. A: knockdown of E-cadherin in HBE3KT cells (left) and NHBE cells (right) using adenoviral transduction of either a scrambled shRNA control virus or shRNA directed against E-cadherin (CHD1) causes over 50% knockdown of E-cadherin protein at 72 h postincubation. B: knockdown of E-cadherin does not increase paracellular permeability of decrease transepithelial resistance. C: E-cadherin knockdown (KD) results in a twofold reduction in cortical tension of cells (*P < 0.005). Prolonged CSE exposure (2 exposures) results in a twofold increase in cortical tension for both control (**P < 0.005) and E-cadherin knockdown, relative to the untreated comparators for each cell type (each data point shows measurement of an individual cell, averaged over 2 or 3 measurements. Data show aggregated measurements over 3–5 different experiments). ***P < 0.005.