Activation of the cystic fibrosis transmembrane conductance regulator by the flavonoid quercetin: potential use as a biomarker of ΔF508 cystic fibrosis transmembrane conductance regulator rescue

Abstract

Therapies to correct the ΔF508 cystic fibrosis transmembrane conductance regulator (CFTR) folding defect require sensitive methods to detect channel activity in vivo. The β₂ adrenergic receptor agonists, which provide the CFTR stimuli commonly used in nasal potential difference assays, may not overcome the channel gating defects seen in ΔF508 CFTR after plasma membrane localization. In this study, we identify an agent, quercetin, that enhances the detection of surface ΔF508 CFTR, and is suitable for nasal perfusion. A screen of flavonoids in CFBE41o⁻ cells stably transduced with ΔF508 CFTR, corrected to the cell surface with low temperature growth, revealed that quercetin stimulated an increase in the short-circuit current. This increase was dose-dependent in both Fisher rat thyroid and CFBE41o⁻ cells. High concentrations inhibited Cl⁻ conductance. In CFBE41o⁻ airway cells, quercetin (20 μg/ml) activated ΔF508 CFTR, whereas the β₂ adrenergic receptor agonist isoproterenol did not. Quercetin had limited effects on cAMP levels, but did not produce detectable phosphorylation of the isolated CFTR R-domain, suggesting an activation independent of channel phosphorylation. When perfused in the nares of Cftr(+) mice, quercetin (20 μg/ml) produced a hyperpolarization of the potential difference that was absent in Cftr(-/-) mice. Finally, quercetin-induced, dose-dependent hyperpolarization of the nasal potential difference was also seen in normal human subjects. Quercetin activates CFTR-mediated anion transport in respiratory epithelia in vitro and in vivo, and may be useful in studies intended to detect the rescue of ΔF508 CFTR by nasal potential difference.

Figure 1.

Flavonoid activation of ΔF508 CFTR-dependent Cl⁻ transport. Screen of flavonoid compounds for activation of temperature-corrected ΔF508 CFTR in CFBE41o⁻ monolayers. Dose–response experiments in modified Ussing chambers were performed with flavonoid compounds. In each case, the concentration of compound eliciting maximal activation is indicated. Concentrations tested include 1, 10, 50–100 μM. *P < 0.05 versus control, n = 4–6 filters in each condition, ± SEM. 100 μM quercetin = 33 μg/ml.

Figure 2.

Quercetin activates CFTR through a mechanism independent of PKA and R-domain phosphorylation. (A) Concentrations of cAMP after forskolin, genistein, or quercetin exposure. CFBE41o⁻ cells were exposed to quercetin (50 μg/ml), genistein (50 μM), forskolin (20 μM), or vehicle (DMSO, 1:1,000 dilution) for 10 minutes before the assay. Quercetin induced a small but statistically significant increase in cAMP compared with genistein and vehicle controls, whereas forskolin-stimulated cAMP was statistically increased above all tested conditions. *P < 0.05, **P < 0.001 compared with control, n = 4 per condition ± SEM. (B) Quercetin does not induce R-domain phosphorylation. Recombinant CFTR R-domain was expressed in stably transfected NIH-3T3 cells and detected by Western blotting against an HA tag (solid arrowhead). A 2- to 4-kD mobility shift is seen, indicative of R-domain phosphorylation (open arrowhead) after a 5-minute treatment with forskolin (20 μM, positive control). Quercetin had no detectable effect on R-domain phosphorylation at any concentration tested. Results are representative of three experiments.

Figure 3.

Dose–response effects of quercetin on I_sc in ΔF508 CFTR-transduced FRT and CFBE41o⁻ monolayers. Cells were studied in modified Ussing chambers, and stimulated I_sc was determined after treatment with a low chloride gradient + amiloride (100 μM), followed by the sequential addition of increasing concentrations of quercetin (1 μM, 10 μM, and 100 μM) or DMSO vehicle at the concentrations shown. Forskolin (20 μM), genistein (50 μM), and CFTR_Inh-172 (10 μM) were included at the end of the experiments to confirm the ion transport response of the monolayer. The data reflect stimulated I_sc over the baseline current (before the first application of quercetin or vehicle). (A) Representative I_sc tracing in ΔF508 CFTR-transduced FRT cells grown at 27°C. (B) Quercetin stimulates a dose-dependent increase in I_sc in ΔF508 CFTR-transduced FRT cells grown at 27°C (right), with smaller effects seen in cells grown at 37°C (left). High doses of quercetin (100 μg/ml) were inhibitory in temperature-corrected cells. *P < 0.05 versus vehicle **P < 0.005 versus vehicle and P < 0.05 versus quercetin-treated cells without temperature correction (37°C growth), n = 4–8 filters per condition, ± SEM. (C) Dose–response studies demonstrating that peak stimulation in CFTR-dependent I_sc occurs at 10 and 20 μg/ml of quercetin in wild-type CFTR FRT cells. *P < 0.05 versus vehicle, n = 4, ± SEM. (D) Complementary dose–response experiments in ΔF508 CFTR-transduced CFBE41o⁻ cells grown at 27°C to correct ΔF508 CFTR misprocessing, using the same protocol as in A. No stimulation of I_sc was seen in cells without temperature correction of ΔF508 CFTR processing (0 ± 0 μA/cm², data not shown). *P < 0.005 versus vehicle, n = 4/condition, ± SEM. (E) In CFBE41o⁻ cells transduced with wild-type CFTR, quercetin exhibited similar effects on ion transport. **P = 0.10, n = 4/condition, ± SEM.

Figure 4.

Quercetin stimulates CFTR-dependent I_sc in conditions simulating human NPD protocols. A representative short-circuit current tracing in temperature-corrected ΔF508 CFTR CFBE41o⁻ monolayers shows a poor response to isoproterenol perfusion, followed by a robust response to quercetin (A) compared with DMSO vehicle (B). After the establishment of a basolateral to mucosal Cl⁻ gradient and the addition of amiloride (100 μM) to the apical compartment (LoCl⁻, Amil), cell monolayers were stimulated in the apical membrane by isoproterenol (10 μM), followed by quercetin (20 μg/ml) or vehicle. Glybenclamide (Glyb, 200 μM, apical addition) was used to confirm the CFTR dependence of the stimulated current. (C) Summary data. The ΔF508 CFTR-transduced CFBE41o⁻ monolayers grown at 27°C × 48 hours exhibited activation of I_sc after quercetin (20 μg/ml). *P < 0.001, n = 6 filters per condition, ± SEM. (D) Quercetin pretreatment does not affect ion transport response. The ΔF508 CFTR transduced CFBE41o⁻ monolayers were exposed to either quercetin (20 μg/ml) or vehicle for 30 minutes daily for 2 consecutive days before Ussing chamber analysis. Cells were grown at 27°C × 48 hours. Stimulation with DMSO vehicle is included as a negative control (gray). *P < 0.05 versus vehicle, n = 5–6/condition, ± SEM. (E) Representative short-circuit current tracing of fully differentiated primary human airway cells treated with quercetin (20 μg/ml) or vehicle control. Cells were harvested from a single normal donor, expanded in culture, grown at the air–liquid interface until terminally differentiated, and then loaded into modified Ussing chambers under voltage clamp conditions. Cells were sequentially treated with amiloride (100 μM), low Cl⁻ secretory gradient, quercetin or vehicle control, forskolin (20 μM; to confirm that CFTR-dependent response was intact), and finally CFTR_Inh-172. (F) Summary data of experiments shown in E. In these cells, the response to forskolin (20 μM) under the exact same conditions is approximately 25 μA/cm². *P < 0.05, n = 4 filters/condition, ± SEM.

Figure 5.

Quercetin activates CFTR-dependent ion transport across the murine nasal mucosa in vivo. Mice underwent a standardized NPD protocol with the addition of quercetin (or vehicle control) in the final perfusion solution, that is, 20 μg/ml in Cl⁻-free gluconate solution and amiloride (amil) (100 μM). Representative tracings are from Cftr⁺ mice stimulated with quercetin (A) or DMSO vehicle (B) in the final perfusate. (C) Summary data of studies described above. Quercetin perfusion resulted in a −2.9-mV mean PD polarization that was significantly different from that seen in Cftr⁺ mice receiving vehicle alone, or Cftr^−/− mice exposed to quercetin. *P < 0.005 versus vehicle or Cftr^−/− mice, n = 6–8 per condition, ± SEM. wt, wild-type. (D) The ΔF508 homozygous mice demonstrated no hyperpolarization after perfusion with isoproterenol (Iso, 10 μM) or quercetin with isoproterenol in the final perfusate. n = 6–8 per condition, ± SEM.

Figure 6.

Quercetin activates CFTR in humans by nasal potential difference. (A) Representative NPD tracing after sequential perfusion of Ringer solution, Ringer solution with amiloride (Amil), Cl⁻-free solution (Zero Cl⁻), Cl⁻-free solution + quercetin (, 10, and 20 μg/ml), and finally quercetin + isoproterenol (Iso, 10 μM). (B) Summary data of NPD studies on 12 normal human subjects (mean of each two nostril tracings for each subject). *P < 0.05 for within tracing change, ± SEM.