Targeting phosphoinositide-3-kinase-delta with theophylline reverses corticosteroid insensitivity in chronic obstructive pulmonary disease

Abstract

Rationale: Patients with chronic obstructive pulmonary disease (COPD) show a poor response to corticosteroids. This has been linked to a reduction of histone deacetylase-2 as a result of oxidative stress and is reversed by theophylline.

Objectives: To determine the role of phosphoinositide-3-kinase-delta (PI3K-δ) on the development of corticosteroid insensitivity in COPD and under oxidative stress, and as a target for theophylline.

Methods: Corticosteroid sensitivity was determined as the 50% inhibitory concentration of dexamethasone on tumor necrosis factor-α-induced interleukin-8 release in peripheral blood mononuclear cells from patients with COPD (n = 17) and compared with that of nonsmoking (n = 8) and smoking (n = 7) control subjects. The effect of theophylline and a selective PI3K-δ inhibitor (IC87114) on restoration of corticosteroid sensitivity was confirmed in cigarette smoke-exposed mice.

Measurements and main results: Peripheral blood mononuclear cells of COPD (50% inhibitory concentration of dexamethasone: 156.8 ± 32.6 nM) were less corticosteroid sensitive than those of nonsmoking (41.2 ± 10.5 nM; P = 0.018) and smoking control subjects (47.5 ± 19.6 nM; P = 0.031). Corticosteroid insensitivity and reduced histone deacetylase-2 activity after oxidative stress were reversed by a non-selective PI3K inhibitor (LY294002) and low concentrations of theophylline. Theophylline was a potent selective inhibitor of oxidant-activated PI3K-δ, which was up-regulated in peripheral lung tissue of patients with COPD. Furthermore, cells with knock-down of PI3K-δ failed to develop corticosteroid insensitivity with oxidative stress. Both theophylline and IC87114, combined with dexamethasone, inhibited corticosteroid-insensitive lung inflammation in cigarette-smoke-exposed mice in vivo.

Conclusions: Inhibition of oxidative stress dependent PI3K-δ activation by a selective inhibitor or theophylline provides a novel approach to reversing corticosteroid insensitivity in COPD.

Figure 1.

Phosphoinisitide-3-kinase (PI3K) inhibitor reversed corticosteroid insensitivity seen in peripheral blood mononuclear cells (PBMCs) obtained from chronic obstructive pulmonary disease (COPD). PBMCs were separated from 8 nonsmoking control subjects (NSC), 7 smoking control subjects (SC), and 17 patients with COPD. The cells were stimulated with tumor necrosis factor (TNF)-α (1 ng/ml) with or without serial concentrations of dexamethasone (Dex, 10⁻¹¹ –10⁻⁶M). Supernatants were taken after overnight incubation for interleukin(IL)-8 ELISA. (A) The average value of IL-8 with or without TNF-α in each group is shown; NS = not significant. (B) IC₅₀-Dex values were calculated and plotted individually. (C) The correlation between FEV₁ (% predicted) and IC₅₀-Dex in patients with COPD (n = 17) analyzed by Spearman correlation test. (D) IC₅₀-Dex values on TNF-α–induced IL-8 production were calculated individually in the presence or absence of LY-294002 (LY) in PBMCs obtained from 12 patients with COPD.

Figure 2.

Phosphoinisitide-3-kinase (PI3K) activation in chronic obstructive pulmonary disease (COPD). Whole tissue extracts were prepared from peripheral lung tissue of nonsmoking control subjects (NSCs) (n = 16), smoking control subjects (SCs) (n = 13), COPD stage 1 (mild; n = 10), COPD stage 2 (moderate; n = 18), COPD stage 3 (severe; n = 6), and COPD stage 4 (very severe; n = 10) in lysis buffer, and pAkt, total Akt and α-tubulin were determined by sodium dodecyl sulfate–polyacrylamide gel electrophoresis/Western blotting. (A) The ratio of pAkt and total Akt was calculated by measuring band density. (B) The mRNA of PI3K-δ in peripheral lung tissue was measured by the Taqman reverse transcriptase polymerase chain reaction system and transcripts were normalized to that of GNB2L1; P < 0.05 versus nonsmoker. (C) PI3K-δ was also immunoprecipitated from lung tissue of NSCs (n = 10), SCs (n = 10), COPD stage 1 (mild; n = 5), COPD stage 2 (moderate; n = 8), COPD stage 3 (severe; n = 5), and COPD stage 4 (very severe; n = 7) and its activity was measured using a PIP₃ assay mass kit. ** P < 0.01 or *** P < 0.001versus NSC, # P < 0.05 versus SC, $ P < 0.05 versus S1.

Figure 3.

Inhibition of PI3K-δ reverses corticosteroid insensitivity under oxidative stress. (A) Either PI3K-δ short interference (si) RNA, PI3K-γ siRNA, or scrambled oligonucleotides was transfected to U937 cells using Nucleofector (Amaxa Biosystems, Cologne, Germany). The cells were incubated for 48 hours and treated with 200 μM of H₂O₂ for 20 minutes. The cells were then stimulated with TNF-α (10 ng/ml) in the presence of dexamethasone and IC₅₀-Dex was calculated as the index of corticosteroid sensitivity. * P < 0.05. (B) HDAC2 was also immunoprecipitated from the PI3K knocked-down (KD) U937 cells after H₂O₂ treatment and the activity was measured. ##P < 0.01.

Figure 4.

Theophylline is able to reverse corticosteroid insensitivity by PI3Kδ inhibition. (A) Peripheral blood mononuclear cells (PBMCs) obtained from 16 patients with chronic obstructive pulmonary disease (COPD) were stimulated with tumor necrosis factor (TNF)-α (1 ng/ml) with or without serial concentrations of dexamethasone (Dex, 10⁻¹¹–10⁻⁶M) in the presence or absence of theophylline (Theo, 10⁻⁶M). IC₅₀-Dex values were calculated individually. (B) HDAC2 siRNA was transfected to PBMCs from healthy volunteers and cells were incubated for 24 hours. IC₅₀-Dex values on TNF-α–induced IL-8 production was calculated. (C) U937 cells were stimulated with H₂O₂ for 7 minutes in the presence or absence of theophylline (Theo, 10⁻⁶M) or LY294002 (LY, 10⁻⁶M) for 30 minutes before H₂O₂, and pAkt level was determined by Western blotting. (D) PI3K-δ was immunoprecipitated from the cells treated with 200 μM H₂O₂ for 7 minutes or nontreated cells, and PI3K-δ activity was measured by a PIP3 detection kit in the presence or absence of theophylline (10⁻⁹–10⁻³M).

Figure 5.

Reversal of corticosteroid insensitivity by theophylline and PI3K-δ inhibition in smoking mice. A/J mice were exposed to cigarette smoke (4%) for 30 minutes per day for 10 days. IC87114 (20 mg/kg orally: IC), dexamethasone (5 mg/kg orally: Dex), LY294002 (20 mg/kg orally: LY), a combination of these treatments, or PEG400 as control were administered twice daily for 3 days for five animals each following the last cigarette exposure. The number of alveolar macrophages (A) and neutrophils (B) in bronchoalveolar lavage fluid was calculated; * P < 0.05, ** P < 0.01 versus smoke control, ## P < 0.01 between air and smoke control. (C) Theophylline (10 mg/kg orally: Theo), dexamethasone (10 mg/kg orally: Dex), a combination of these treatments, or 0.5% carboxymethylcellulose as control were therapeutically administered once daily for 3 days following the last cigarette exposure. The number of neutrophils in bronchoalveolar lavage was shown. (D) HDAC2 was immunoprecipitated in nuclear-rich extract from lung tissue of the mice treated with indicated compounds, and its activity was measured by fluorescent-based HDAC activity assay kit. * P < 0.05, ** P < 0.01 versus smoke control.