Possible role of Krüppel-like factor 5 in the remodeling of small airways and pulmonary vessels in chronic obstructive pulmonary disease

Abstract

Background: Small airway remodeling is an important cause of the airflow limitation in chronic obstructive pulmonary disease (COPD). A large population of patients with COPD also have pulmonary hypertension. Krüppel-like factor 5 (KLF5) is a zinc-finger transcription factor that contributes to tissue remodeling in cardiovascular diseases. Here, we evaluate the possible involvement of KLF5 in the remodeling of small airways and pulmonary vessels in COPD.

Methods: Lung tissues were obtained from 23 control never-smokers, 17 control ex-smokers and 24 ex-smokers with COPD. The expression of KLF5 in the lung tissues was investigated by immunohistochemistry. We investigated whether oxidative/nitrosative stress, which is a major cause of the pathogenesis in COPD, could augment the production of KLF5. We examined the role of KLF5 in the stress-mediated tissue remodeling responses. We also investigated the susceptibility of KLF5 expression to nitrosative stress using bronchial fibroblasts isolated from the lung tissues.

Results: The expression of KLF5 was up-regulated in the small airways and pulmonary vessels of the COPD patients and it was mainly expressed in bronchial fibroblasts and cells of the pulmonary vessels. The extent of the KLF5 expression in the small airway of the COPD group had a significant correlation with the severity of the airflow limitation. Oxidative/nitrosative stress augmented the production of KLF5 in lung fibroblasts as well as the translocation of KLF5 into the nuclei. Silencing of KLF5 suppressed the stress-augmented differentiation into myofibroblasts, the release of collagens and metalloproteinases. Bronchial fibroblasts from the patients with COPD highly expressed KLF5 compared to those from the control subjects under basal condition and were more susceptible to the induction of KLF5 expression by nitrosative stress compared to those from the control subjects.

Conclusion: We provide the first evidence that the expression of KLF5 is up-regulated in small airways and pulmonary vessels of patients with COPD and may be involved in the tissue remodeling of COPD.

Fig. 1

Semiquantitative measurement of Krüppel-like factor 5 (KLF5) production in the small airways by immunohistochemical staining. Lung tissues were obtained from control subjects that never or previously smoked and from former smokers with COPD. The localization of KLF5 in the lung tissues was investigated by immunostaining. Representative photographs of the immunoreactivity of KLF5 in the small airways are shown (original magnification × 200 for upper panels and left middle panel, × 400 for right middle panel; scale bars = 100 μm) (a). A negative control sample was provided to replace a non-specific polyclonal rabbit IgG antibody instead of the primary antibody on the lung tissue specimens of the patients with COPD. Arrow heads indicate KLF5-immunopositive cells. The immunopositive area of KLF5 in the small airways was quantified using Image J (b). **p < 0.01 versus control never smoker, ^†† p < 0.01 versus control ex-smoker

Fig. 2

Semiquantitative measurement of KLF5 production in the pulmonary vessels by immunohistochemical staining. The localization of KLF5 in the lung tissues was investigated by immunostaining. Representative photographs of the immunoreactivity of KLF5 in the pulmonary vessels are shown (original magnification × 200 for upper panels and left lower panel, × 400 for right lower panel; scale bars = 100 μm) (a). A negative control sample was provided to replace the non-specific polyclonal rabbit IgG antibody instead of the primary antibody for the lung tissue specimens of the patients with COPD. Arrow heads indicate KLF5-immunopositive cells. The immunopositive area of KLF5 in the pulmonary vessels was quantified using Image J (b). **p < 0.01 versus control never smoker, ^†† p < 0.01 versus control ex-smoker

Fig. 3

Immunoreactivity of KLF5 in the bronchial fibroblasts. To determine the cell types that expressed KLF5, double immunostaining for KLF5 and prolyl 4-hydroxylase β (P4HB) was carried out. P4HB is a marker of fibroblasts. Representative photographs of the immunoreactivity of KLF5 (green) and P4HB (red) are shown (original magnification × 200 for all panels; scale bars = 100 μm). A negative control sample was provided to replace the non-specific polyclonal rabbit IgG antibody instead of the primary antibody for the lung tissue specimens of the patients with COPD. DAPI: 4′,6-diamidino-2-phenylindole; HE: hematoxylin-eosin staining

Fig. 4

Relationship between the immunopositive area for KLF5 in the small airways and lung function. The relationship between the KLF5-positive area in the small airways and the values of FEV₁ %predicted (a), FVC %predicted (b), and D_LCO/V_A %predicted (c) was investigated in the COPD group. r is the correlation coefficient; the lines and p values correspond to the fitted regression equation. FEV₁: forced expiratory volume in 1 s; FVC: forced vital capacity; D_LCO: diffusing capacity of the lung for carbon monoxide; V_A: alveolar volume; N.S.: not significant

Fig. 5

Effect of hydrogen peroxide (H₂O₂) or peroxynitrite (ONOO⁻) on KLF5 production and translocation into the nucleus in human fetal lung fibroblasts (HFL-1). Production of KLF5 was analyzed by immunoblotting. HFL-1 cells were treated with various concentrations of H₂O₂ for 24 h (a) and with 10⁻⁶ M H₂O₂ for various durations (b) as well as ONOO⁻ for 24 h (c) and 10⁻⁷ M ONOO⁻ for various durations (d). Each band intensity of KLF5 was standardized by that of β-actin. Translocation of KLF5 into the nucleus was also analyzed by immunoblotting. The cells were treated with 10⁻⁶ M H₂O₂ (e) or 10⁻⁷ M ONOO⁻ (f) and the nuclear fraction was harvested at various time points. Each band intensity of KLF5 was standardized by that of lamin A/C. All values are expressed as mean ± SD for 4–7 separate experiments. *p < 0.05, **p < 0.01 versus control group

Fig. 6

Effects of KLF5 silencing by siRNA on the H₂O₂ or ONOO⁻-augmented the production of α-smooth muscle actin (α-SMA) and release of collagens. Effects of siRNA for KLF5 on the gene expression and the protein production were confirmed by quantitative reverse transcription-polymerase chain reaction (qRT-PCR) for 24 h after transfection (a) and immunoblotting for 72 h (b). Effects of KLF5 silencing on the H₂O₂ (c) or the ONOO⁻ (d)-augmented production of α-SMA were analyzed by immunoblotting. Each band intensity was analyzed by Quantity One. Effect of KLF5 silencing on the ONOO⁻-augmented release of collagens was analyzed (e). All values are expressed as mean ± SD for 4 separate experiments. **p < 0.01 versus vehicle-pretreated vehicle-exposed group. ^† p < 0.05, ^†† p < 0.01 versus vehicle-pretreated H₂O₂ or ONOO⁻-exposed group

Fig. 7

Effects of KLF5 silencing by siRNA on the ONOO⁻-augmented release of matrix metalloproteinases (MMPs). Effects of KLF5 silencing on the ONOO⁻-augmented release of MMP-9 and MMP-2 were investigated by gelatin zymography (a). After silencing, the infected cells were incubated with 10⁻⁷ M ONOO⁻ (filled bars) or vehicles (open bars) for 24 h. Latent form of MMP-9 (a and b), active form of MMP-9 (a and c), latent form of MMP-2 (a and d), and active form of MMP-2 were evaluated. Each band intensity was analyzed by Quantity One. All values are expressed as mean ± SD for 4 separate experiments. **p < 0.01 versus vehicle-pretreated vehicle-exposed group. ^† p < 0.05, ^†† p < 0.01 versus vehicle-pretreated ONOO⁻-exposed group

Fig. 8

Production of KLF5 in adult human bronchial fibroblasts and effects of ONOO⁻ on production of KLF5 in adult cells. Bronchial fibroblasts were isolated from the lungs of the subjects in each study group. Four different strains of bronchial fibroblasts in each group were obtained. The amounts of KLF5 were analyzed by immunoblotting (a). **p < 0.01 versus control never-smoker, ^†† p < 0.01 versus control ex-smoker. The fibroblasts were treated with 10⁻⁶ M ONOO⁻ for 24 h. The production of KLF5 was analyzed by immunoblotting (b). Each band intensity of KLF5 was standardized by that of β-actin. *p < 0.05, **p < 0.01. All values are expressed as mean ± SD. CNS = Control never-smoker; CES = Control ex-smoker, COPD = COPD ex-smoker