Cigarette smoke suppresses Bik to cause epithelial cell hyperplasia and mucous cell metaplasia

Abstract

Rationale: Aberrant regulation of airway epithelial cell numbers in airways leads to increased mucous secretions in chronic lung diseases such as chronic bronchitis. Because the Bcl-2 family of proteins is crucial for airway epithelial homeostasis, identifying the players that reduce cigarette smoke (CS)-induced mucous cell metaplasia can help to develop effective therapies.

Objectives: To identify the Bcl-2 family of proteins that play a role in reducing CS-induced mucous cell metaplasia.

Methods: We screened for dysregulated expression of the Bcl-2 family members.

Measurements and main results: We identified Bik to be significantly reduced in bronchial brushings of patients with chronic epithelial cell hyperplasia compared with nondiseased control subjects. Reduced Bik but increased MUC5AC mRNA levels were also detected when normal human airway epithelial cells (HAECs) were exposed to CS or when autopsy tissues from former smokers with and without chronic bronchitis were compared. Similarly, exposure of C57Bl/6 mice to CS resulted in increased numbers of epithelial and mucous cells per millimeter of basal lamina, along with reduced Bik but increased Muc5ac expression, and this change was sustained even when mice were allowed to recover in filtered air for 8 weeks. Restoring Bik expression significantly suppressed CS-induced mucous cell metaplasia in differentiated primary HAEC cultures and in airways of mice in vivo. Bik blocked nuclear translocation of phospho-ERK1/2 to induce apoptosis of HAECs. The conserved Leu61 within Bik and ERK1/2 activation were essential to induce cell death in hyperplastic mucous cells.

Conclusions: These studies show that CS suppresses Bik expression to block airway epithelia cell death and thereby increases epithelial cell hyperplasia in chronic bronchitis.

Figure 1.

Exposure to cigarette smoke (CS) decreases Bik expression. (A) Bronchial brushings were obtained from 11 patients with chronic bronchitis (CB) and 9 subjects with no lung diseases as control (CTR) and subjected to quantitative real-time polymerase chain reaction (qRT-PCR) for mRNAs of Bcl-2 family members. Box plots represent the median value and 25th and 75th quartile. Whiskers are the 5th and 95th percentile. Open circles are outlier values (> 1.5 × interquartile range). Results for Bnip32, Bok, and Bid showed no significant differences among groups. (B) Bik and MUC5AC mRNAs quantified by qRT-PCR in lung autopsy tissues from healthy never-smokers or current smokers with chronic bronchitis. (C) Bik and MUC5AC mRNAs levels in the lungs of C57BL/6 mice exposed to 250 mg/m³ CS or filtered air for 6 h/d, 5 d/wk for 10 weeks. (D) Bik mRNA and (E) protein levels in human airway epithelial cells (HAECs) treated with nothing as control or 1,000 ng/ml cigarette smoke extract (CSE) for 24 hours and with 50 ng/ml IFN-γ for the following 24 hours. Bar = means ± SEM (n = 4 mice/group or n = 3 different treatments/group). *Denotes statistically significant difference (P < 0.05). CurS-CB = current smokers with CB; NS-NCB = never smokers without CB.

Figure 2.

Exposure to cigarette smoke (CS) leads to increased epithelial cell hyperplasia (ECH). Airway epithelial cell numbers in bik^+/+ and bik^−/− mice exposed to 250 mg/m³ CS or filtered air (FA) for 6 h/d, 5 d/wk for (A) 10 weeks or (B) 3 weeks followed by 8 weeks of recovery in filtered air. The left lungs were fixed under constant pressure perfusion, cut into 4-mm slices from distal to caudal, and slices embedded in paraffin. Tissue sections (5 μm) were stained with hematoxylin and eosin and the total epithelial cell number in the airways quantified. (C) C57BL/6 mice were exposed to 250 mg/m³ CS or filtered air for 6 h/d, 5 d/wk for 3 weeks followed by 8 weeks of recovery in air and Bik and Muc5ac mRNAs quantified by quantitative real-time polymerase chain reaction (qRT-PCR). (D) Bik and MUC5AC mRNAs in lung tissues from former smokers with and without chronic bronchitis (CB) quantified by qRT-PCR. (E) Primary human airway epithelial cells (HAECs) placed in culture in an air–liquid interface (ALI) culture were left untreated or treated with 1,000 ng/ml CSE for 24 hours and harvested 3 days later. Bik and MUC5AC mRNA levels were quantified from CS-treated and nontreated control subjects by qRT-PCR. (F) Primary HAECs placed in culture in an ALI culture were left untreated or treated with 1,000 ng/ml CS for 24 hours and 3 days later membranes were embedded in paraffin and stained with Alcian blue and periodic acid Schiff (AB/PAS) followed by hematoxylin and eosin staining. The number of mucous cells was significantly increased in CS-treated culture compared with untreated controls. Representative photomicrographs of AB/PAS-stained culture showing increased number of mucous cells in primary differentiated HAECs treated with 1,000 ng/ml CSE for 24 hours and harvested 3 days later, compared with untreated groups. Bars = group means ± SEM (n = 4 mice/group or n = 3 different treatments/group). *Denotes significant differences among groups (P < 0.05). Ctr = control; FS-CB = former smokers with CB; FS-NCB = former smokers without CB.

Figure 3.

Restoration of Bik expression reduces cigarette smoke (CS)-induced mucous cell metaplasia (MCM) and epithelial cell hyperplasia (ECH). (A) Differentiated human airway epithelial cells (HAECs) were treated with 1,000 ng/ml CSE for 24 hours and harvested 3 days later. Cultures were infected with nothing, Ad-Bik^L61G, or Ad-Bik for 24 hours before harvest. Western blot showing expression of Bik in Ad-Bik^L61G– or Ad-Bik–infected HAECs. Bars = group means ± SEM (n = 3/group). (B, C) C57BL/6 mice were exposed to 250 mg/m³ CS or filtered air for 6 h/d, 5 d/wk for 3 weeks and instilled with phosphate-buffered saline (PBS), Ad-GFP, or Ad-Bik in 50 μL PBS on 2 consecutive days and lungs harvested the following day. (B) Western blot analysis showing expression of HA-tagged Bik in three representative lungs of Ad-Bik– (lanes 1, 2, and 3) but not in Ad-GFP–instilled (lanes 4, 5, and 6) mice. (C) ECH was significantly reduced in the airways of mice instilled with Ad-Bik compared with those instilled with Ad-GFP or PBS. (D) The percentage of Muc5ac-positive cells was significantly reduced in the airways of mice instilled with Ad-Bik compared with those instilled with Ad-GFP or PBS. Bars = group means ± SEM (n = 6 mice/group). *Denotes significantly different among groups (P < 0.05).

Figure 4.

Bik expression reduces nuclear localization of cigarette smoke (CS)-activated ERK1/2 and phopho-ERK1/2 enhances Bik-induced cell death. (A) Human airway epithelial cells (HAECs) were treated with 1,000 ng/ml CSE, harvested over a period of 5 to 45 minutes and protein lysates analyzed for activated ERK1/2 by Western blotting. ERK1/2 was activated within 5 minutes of treatment, and this activation was sustained for 15 minutes of CSE treatment. (B) Nuclear and cytosolic fractions from HAECs infected with 100 MOI Ad-Bik or Ad-Bik^L61G 15 minutes after CSE treatment analyzed by Western blotting for phospho-ERK1/2, total ERK1/2, Bik, lamin, and actin. The figure is representative of three independent experiments. (C) CS enhances Ad-Bik–induced cell death. HAECs infected with Ad-Bik or Ad-Bik^L61G at 100 MOI were either treated with CSE or left untreated and counted 24 hours later. (D) HAECs were treated with different concentrations of insulin-like growth factor 1 (IGF-1) for 0 to 30 minutes and protein lysates analyzed for p-ERK1/2, ERK1/2, and actin by Western blotting. (E) HAECs were infected with 100 MOI Ad-Bik or Ad-Bik^L61G and treated with different concentrations of IGF-1, and cell viability was quantified. (F) Inhibition of ERK1/2 activation by U0126 reduced in CS-exposed cells. Western blot analysis of extracts from HAECs treated with U0126 and/or CS to reduce CS-induced ERK1/2 activation. (G) HAECs were treated with 1 μM U0126 and infected with 100 MOI Ad-Bik followed by 1,000 ng/ml CSE treatment 24 hours later. Cell death induced by Ad-Bik in the presence of CS was diminished significantly when ERK1/2 activation was inhibited by 1 uM U0126. Bars = group means ± SEM (n = 3 different treatments/group). *Denotes statistically significant difference (P < 0.05).