Subacute TGFβ Exposure Drives Airway Hyperresponsiveness in Cystic Fibrosis Mice through the PI3K Pathway

Abstract

Cystic fibrosis (CF) is a lethal genetic disease characterized by progressive lung damage and airway obstruction. The majority of patients demonstrate airway hyperresponsiveness (AHR), which is associated with more rapid lung function decline. Recent studies in the neonatal CF pig demonstrated airway smooth muscle (ASM) dysfunction. These findings, combined with observed CF transmembrane conductance regulator (CFTR) expression in ASM, suggest that a fundamental defect in ASM function contributes to lung function decline in CF. One established driver of AHR and ASM dysfunction is transforming growth factor (TGF) β1, a genetic modifier of CF lung disease. Prior studies demonstrated that TGFβ exposure in CF mice drives features of CF lung disease, including goblet cell hyperplasia and abnormal lung mechanics. CF mice displayed aberrant responses to pulmonary TGFβ, with elevated PI3K signaling and greater increases in lung resistance compared with controls. Here, we show that TGFβ drives abnormalities in CF ASM structure and function through PI3K signaling that is enhanced in CFTR-deficient lungs. CF and non-CF mice were exposed intratracheally to an adenoviral vector containing the TGFβ1 cDNA, empty vector, or PBS only. We assessed methacholine-induced AHR, bronchodilator response, and ASM area in control and CF mice. Notably, CF mice demonstrated enhanced AHR and bronchodilator response with greater ASM area increases compared with non-CF mice. Furthermore, therapeutic inhibition of PI3K signaling mitigated the TGFβ-induced AHR and goblet cell hyperplasia in CF mice. These results highlight a latent AHR phenotype in CFTR deficiency that is enhanced through TGFβ-induced PI3K signaling.

Keywords: CFTR; airway hyperresponsiveness; airway smooth muscle; cystic fibrosis; transforming growth factor β.

Figure 1.

(A) Adenoviral (Ad)-transforming growth factor (TGF) β–exposed cystic fibrosis (CF) mice have greater increases in lung resistance at methacholine doses up to 100 mg/ml compared with PBS-exposed CF mice. *P < 0.05 versus CF mice treated with PBS by two-way ANOVA with Tukey’s post hoc test. (B) Ad-TGFβ–exposed CF mice demonstrate a delayed methacholine (MTCH)-induced peak change in resistance and delayed resolution of resistance alterations compared with non-CF mice. Resistance was measured during each of 12 single-frequency forced oscillation technique perturbations, taken approximately every 12 seconds after each nebulized treatment. *P < 0.05 by two-way ANOVA. (C) The areas under the curve of the resistance changes induced by 25 and 100 mg/ml of methacholine are higher in CF than non-CF TGFβ-exposed mice. *P < 0.05 by two-tailed t test. Data are presented as mean ± SD.

Figure 2.

(A) Albuterol treatment does not reduce the heightened TGFβ-dependent baseline resistance in either CF or non-CF mice (ns = not significant by two-tailed t test, P > 0.05). (B) Nebulized albuterol (Alb) treatment immediately before methacholine challenge blunted the methacholine-induced resistance increases in Ad-TGFβ–exposed CF mice. Control mice were treated with PBS before methacholine challenge. (C) Albuterol pretreatment significantly reduced the slope of methacholine-induced resistance in CF mice, but not in non-CF mice. *P < 0.05 by one-way ANOVA with Tukey’s post hoc analysis. Data are presented as mean ± SD.

Figure 3.

(A) Immunohistochemistry for ACTA2-smooth muscle actin (αSMA) demonstrated airway smooth muscle (ASM) (brown) around airways in PBS control and Ad-TGFβ–treated CF and non-CF mice. Scale bar: 100 μm. (B) Quantification of αSMA area, corrected to basement membrane (BM) perimeter squared, demonstrated that TGFβ exposure significantly increased ASM area at Day 7 in CF mice only. *P < 0.05 by one-way ANOVA with Tukey’s post hoc analysis. Data are presented as mean ± SD.

Figure 4.

After treatment with the pan-PI3K inhibitor LY294002 (LY), PI3K signaling, but not mitogen-activated protein kinase (MAPK) or canonical Smad signaling, is inhibited in the lungs of mice treated with intratracheal Ad-TGFβ. Western blot analysis was performed on whole-lung homogenates from mice treated with intraperitoneal LY before Ad-TGFβ exposure. (A) PI3K signaling, as measured by phosphorylation of S6, is inhibited in both CF and non-CF mouse lungs at Day 1 after Ad-TGFβ treatment as compared with vehicle (veh)-treated mice. At 3 days after TGFβ exposure, CF mice, but not non-CF mice, have overcome PI3K inhibition. (B and C) Neither Ad-TGFβ–induced MAPK (as measured by phosphorylated extracellular signal-regulated protein kinase [ERK] 1/2) nor canonical Smad (as measured by phosphorylated Smad2) signaling is affected by LY treatment as compared with vehicle treatment. Values are corrected to vehicle-treated mice of the same genotype and time point. *P < 0.05 by two-tailed t test. ns indicates P > 0.05 by two-tailed t test. Data are presented as mean ± SD.

Figure 5.

(A and B) Treatment with a PI3K inhibitor, LY, did not reduce total cell count (A) or alter differential cell count (B) in the BAL of Ad-TGFβ–treated CF or non-CF mice. E = eosinophils; L = lymphocytes; M = macrophages; N = neutrophils. (C) Similarly, LY treatment did not alter TGFβ levels, either total or active, in BAL of treated mice. (D) PI3K inhibition with LY did not reduce ASM area in TGFβ-treated CF or non-CF mice. (E) Despite similar levels of pulmonary inflammation, TGFβ, and ASM burden, LY treatment significantly reduced airway hyperresponsiveness (AHR) in Ad-TGFβ–exposed CF mice, but not Ad-TGFβ–exposed non-CF mice. *P < 0.05 versus CF LY-treated mice by two-way ANOVA with Sidak’s post hoc analysis. (F) Slope of max resistances was significantly decreased by LY in CF mice, indicating reduced AHR after PI3K inhibition in the setting of CF transmembrane conductance regulator deficiency only. *P < 0.05 by one-way ANOVA with Tukey’s post hoc analysis. Data are presented as mean ± SD.

Figure 6.

PI3K inhibition with LY decreased goblet cell hyperplasia in Ad-TGFβ–exposed CF, but not non-CF, mice. (A) Periodic acid–Schiff–stained, mucin-containing goblet cells (arrows) were more numerous in Ad-TGFβ–exposed CF mice treated with vehicle. Scale bar: 50 μm. (B) The percentage of goblet cells around airways was significantly increased in vehicle-treated CF mice after Ad-TGFβ exposure, but this goblet cell hyperplasia was decreased by LY treatment. Percent goblet cells was not changed by PI3K inhibition in Ad-TGFβ–exposed non-CF mice. *P < 0.05 by one-way ANOVA with Tukey’s post hoc analysis. Data are presented as mean ± SD.