Interleukin-17A induces bicarbonate secretion in normal human bronchial epithelial cells

Abstract

The innate immune functions of human airways include mucociliary clearance and antimicrobial peptide activity. Both functions may be affected by changes in epithelial ion transport. Interleukin-17A (IL-17A), which has a receptor at the basolateral membrane of airway epithelia, is a T cell cytokine that has been shown to increase mucus secretion and antimicrobial peptide production by human bronchial epithelial (HBE) cells. Furthermore, IL-17A levels are increased in sputum from patients during pulmonary exacerbations of cystic fibrosis. Therefore, we investigated the effects of IL-17A on basal, amiloride-sensitive, and forskolin-stimulated ion transport in mature, well-differentiated HBE cells. Exposure of HBE monolayers to IL-17A for 48 h induced a novel forskolin-stimulated bicarbonate secretion in addition to forskolin-stimulated chloride secretion and resulted in alkalinization of liquid on the mucosal surface of polarized cells. IL-17A-induced bicarbonate secretion was cystic fibrosis transmembrane conductance regulator (CFTR)-dependent, mucosal chloride-dependent, partially Na⁺-dependent, and sensitive to serosal, but not mucosal, stilbene inhibition. These data suggest that IL-17A modulates epithelial bicarbonate secretion and implicate a mechanism by which airway surface liquid pH changes may be abnormal in cystic fibrosis.

the innate defense of the lung against inhaled pathogens and toxins incorporates both physical and immunological mechanisms. A primary physical innate defense is mucociliary clearance, the trapping of inhaled particles in the airway surface liquid, and propulsion of trapped particles toward the oropharynx by a combination of ciliary beating and cough clearance. Immunological defense mechanisms of the airway epithelium in response to both innate and adaptive immune recognition include the production and secretion of antimicrobial peptides and antimicrobial chemicals such as hydrogen peroxide. Both mucociliary clearance and antimicrobial activity are regulated in part by epithelial ion transport, which may be altered through the impact of cytokines on human bronchial epithelial (HBE) cells (3, 8–10).

Interleukin (IL)-17A is a homodimeric, disulfide-linked cytokine produced predominantly by CD4-positive T cells of the Th17 lineage (14, 25), although it may be produced by other T cell subsets (21). It is a member of a unique cytokine family comprising six members, IL-17A through IL-17F. IL-17A binds preferentially to the IL-17RA, which is expressed in the basolateral membranes of bronchial epithelial cells (23). In the lung, IL-17A plays a critical role in the pulmonary inflammatory response to extracellular, gram-negative organisms such as Klebsiella pneumonia and Pseudomonas aeruginosa (5, 13). In vitro studies have demonstrated that binding of IL-17A to its receptor on HBE cells results in the production of proinflammatory cytokines and chemokines (23), increased mucin production (2), and increased production of antimicrobial peptides (17). These data suggest that IL-17A is a potent mediator of epithelial host defense in the airways. Furthermore, IL-17A levels are elevated in sputum from patients during pulmonary exacerbations of CF and subsequently decrease with intravenous antibiotic therapy (23), suggesting that IL-17A may have a role in the pathophysiology of pulmonary exacerbations of CF.

Because mucociliary clearance, antimicrobial peptide activity, and antimicrobial systems such as reactive oxygen species production are host defense mechanisms that depend on ion transport, we investigated the effects of IL-17A on the basal, amiloride-sensitive, and forskolin-stimulated ion transport properties of mature, well-differentiated HBE cells. Understanding how IL-17A affects epithelial ion transport will provide insights into the physiology of the airways during periods of inflammation and may lead to novel therapies for diseases of chronic airways inflammation such as cystic fibrosis (CF).

MATERIALS AND METHODS

Cell culture.

HBE cells, purchased from Lonza (Walkersville, MD), were cultured according to Gray and colleagues (11). Primary human CF airway cells were obtained from excess pathological tissue remaining after lung transplantation of ΔF508 homozygous CF patients, as previously described (24) under a protocol approved by the University of Pittsburgh Institutional Review Board. P2 HBEs were seeded in plastic T-75 flasks (Costar) and grown in bronchial epithelial cell growth medium (BEGM) (Biowhittaker) supplemented with bovine pituitary extract (52 μg/ml), hydrocortisone (0.5 μg/ml), human recombinant epidermal growth factor (0.5 ng/ml), epinephrine (0.5 μg/ml), transferrin (10 μg/ml), insulin (5 μg/ml), retinoic acid (0.1 μg/ml), triiodothyronine (6.5 μg/ml), gentamicin (50 μg/ml), and amphotericin-B (50 ng/ml). Media were changed every 48 h until cells were 90% confluent. Cells were then passaged and seeded on Transwell permeable inserts (Costar) in differentiation media containing 50% DMEM in BEGM with the same supplements as above but without triiodothyronine and with a final retinoic acid concentration of 50 nM (all-trans retinoic acid; Sigma). Cells were seeded at a density of 3 × 10⁴ per 0.33 cm². HBEs were maintained submerged for the first 7 days in culture after which time they were exposed to an apical air interface for the remainder of the culture period. Medium was refreshed two to three times each week. At all stages of culture, cells were maintained at 37°C in 5% CO₂ in an air incubator. Under these conditions, HBEs formed a well-differentiated mucociliary phenotype with the classical ion transport phenotype associated with this tissue. A total of five tissue donors were used to complete these studies.

Solutions.

The normal bath solution contained (in mM): 120 NaCl, 25 NaHCO₃, 3.3 KH₂PO₄, 0.8 K₂HPO₄, 1.2 MgCl₂, 1.2 CaCl₂, and 10 glucose. The pH of this solution was 7.3–7.4 when gassed with a mixture of 95% O₂-5% CO₂ at 37°C. For Cl⁻-free solutions, Cl⁻ was replaced with equimolar gluconate. For Na⁺-free solutions, Na⁺ was replaced with equimolar N-methyl-d-glucamine. HCO₃⁻-free solutions contained (in mM): 145 NaCl, 3.3 KH₂PO₄, 0.8 K₂HPO₄, 1.2 MgCl₂, 1.2 CaCl₂, 10 HEPES, and 10 glucose; pH 7.4. This solution was bubbled with air rather than O₂/CO₂. Cl⁻ and HCO₃⁻-free solutions contained (in mM): 145 sodium gluconate, 3.3 KH₂PO₄, 0.8 K₂HPO₄, 1.2 magnesium gluconate, 1.2 calcium gluconate, 10 HEPES, and 10 glucose, pH 7.4, and were bubbled with air.

Two different solutions were used for calibration of pH with fluorescence emissions. For calibration of SNARF-free acid (Invitrogen, Carlsbad, CA) the solution contained (in mM): 120 NaCl, 25 sodium gluconate, 5 KCl, 1.2 MgCl₂, 1.2 CaCl₂, and 20 HEPES at five different pH (6.8, 7.2, 7.6, 8.0, and 8.4). For calibration of SNARF-1 dextran (Invitrogen) and intracellular pH, solutions contained (in mM): 120 KCl, 20 NaCl, 1 CaCl₂, 1 MgCl₂, 20 HEPES, plus 10 μM nigericin, 10 μM valinomycin, and 10 μM forskolin. pH was adjusted to 6.5, 7, and 8 with NaOH.

Short-circuit current measurements.

Transwell inserts were mounted in a modified, vertical Ussing chamber (Costar), and the monolayers were voltage-clamped continuously to 0 mV after fluid resistance compensation using an automatic voltage clamp (VCC 600; Physiologic Instruments). Short-circuit current (I_SC) was digitized at 1 Hz, and data were stored on a computer hard drive using Acquire and Analyze software build 2.2.0 (Physiologic Instruments). Transepithelial resistance (R_T) was determined by stepping the clamp potential from 0 to ± 2 mV for 4 s every minute, recording the change in I_SC, and calculating R_T from Ohm's law (R_T = ΔV/ΔI, where V is voltage and I is current).

I_SC was allowed to stabilize at the beginning of each experiment and after each drug addition. By convention, an upward deflection in the I_SC tracing represents net serosal-to-mucosal anion movement or mucosal-to-serosal cation movement. Mean data are presented as change in I_SC from the stable I_SC recorded before and after each drug addition. For example, in Fig. 1A, the change in I_SC from points A to B represents the change in I_SC in response to amiloride. Unless otherwise stated, X-axis labels on bar graphs represent the manipulation responsible for corresponding changes in I_SC and are provided in chronological order from left to right.

Fig. 1.

Dose-dependent increase in forskolin-stimulated short-circuit current (I_SC) with interleukin (IL)-17A treatment. A: representative I_SC traces demonstrating changes in I_SC after sequential additions of amiloride, forskolin, and bumetanide. Points A and B indicate the times used to measure the change in I_SC in response to amiloride. Note that there is enhanced forskolin-stimulated I_SC in IL-17A-treated cells and that the additional I_SC is not sensitive to bumetanide or barium. B: mean changes in I_SC in response to amiloride, forskolin, and bumetanide after incubation of cells with IL-17A in the basolateral medium for 48 h. IL-17A induced a dose-dependent additional forskolin-stimulated I_SC (n = 7–8 filters/group, 2 donors). C: IL-17A-induced forskolin-stimulated I_SC is sensitive to ouabain (n = 6 filters/group, 1 donor). *P < 0.05 compared with 0 ng/ml. ***P < 0.001 compared with vehicle.

Low-light level, live-cell fluorescence microscopy.

Mature, well-differentiated HBE cells on Transwell supports were placed in a specially designed chamber (Bioptechs, Butler, PA) that allowed for perfusion of the serosal side of the cells. The chamber was mounted on the stage of an inverted fluorescence microscope (Nikon Eclipse TE2000-U) and imaged with a ×20 air objective (numeric aperture 0.75; Nikon S-Fluor). The serosal side of the cells was perfused continuously with normal bath solution that was held in an overhead, heated glass reservoir. Solution in the reservoir was gassed with 95% O₂-5% CO₂ and fed by gravity through glass tubing to glass heating coils close to the chamber inlet. Basolateral solution changes were made using an automated pinch-valve system (Warner Instruments, Hamden, CT). The entire chamber holding the Transwell insert was covered with a small box with an inlet for gas injection. SNARF dye was excited with light from a xenon arc lamp filtered with a 480/20 nm band-pass filter, and the ratio of fluorescence emissions at 640 and 580 nm (F₆₄₀/F₅₈₀) were captured every 10 s using a Hammamatsu ORCA-ER camera and Image Suite software (Perkin Elmer, Waltham, MA). Data were stored on a computer hard drive and subsequently analyzed using Origin 7.5 software (OriginLab, Northampton, MA).

HBE surface pH was measured using two separate protocols. The first protocol was used to measure resting mucosal surface pH based on the protocol of Jayaraman et al. (16). Briefly, SNARF-1 dextran (10,000 mol wt; Invitrogen) was dispersed by sonication in perfluorocarbon (FC-77; Sigma) at 1 mg/ml. SNARF-1 dextran in FC-77 (20 μl) was applied to the mucosal surface of mature, well-differentiated HBE cells. After a 5- to 10-min incubation at 37°C and 5% CO₂, filters were mounted on the microscope stage and perfused on the serosal side with HCO₃⁻-buffered solution. The F₆₄₀-to-F₅₈₀ (F₆₄₀/F₅₈₀) ratio was captured every minute. For calibration, cells were incubated in HEPES-buffered, high-K⁺ solution (pH 6.5, 7.0, or 7.5) for 2 h at 37°C. Mucosal high-K⁺ buffer was removed and replaced with 20 μl of 1 mg/ml SNARF-1 dextran in buffer of the same pH. F₆₄₀/F₅₈₀ was captured, and its relationship to pH was calculated by linear regression.

The second protocol was designed to measure HCO₃⁻ secretion in a small volume of HCO₃⁻-free fluid added to the mucosal surface of polarized cells. The mucosal surface of the cells was submerged in solution containing (in mM): 120 NaCl, 25 sodium gluconate, 5 KCl, 1.2 MgCl₂, 1.2 CaCl₂, and 1 HEPES, pH 7.37. Amiloride hydrochloride (10 μM) and SNARF-5-free acid (1 μg/ml) (Invitrogen) were added to the solution immediately before its application to the mucosal surface of the cells. After mounting the chamber containing the cells on the microscope stage, the chamber was covered with a small box that was continuously infused with humidified 100% O₂. F₆₄₀/F₅₈₀ was captured every min for 5 min after which the serosal solution was changed to one containing 2 μM forskolin. Data are reported as means ± SE of the rate of change in pH per min for each group.

To measure intracellular pH, mature HBE cells were loaded with acetoxymethyl ester SNARF-5 (SNARF-5-AM; Invitrogen) for 15–20 min at room temperature in the presence of 1 mg/ml water-soluble probenecid (Invitrogen). Cells were mounted on the microscope stage as described for extracellular pH measurements except the mucosal surface was submerged in 100 μl of normal, HCO₃⁻-buffered solution with 10 μM amiloride. After mounting the chamber containing the cells on the microscope stage, the chamber was covered with a small box that was continuously infused with humidified 95% O₂-5% CO₂. F₆₄₀/F₅₈₀ was captured every minute as described above.

Chemicals.

Recombinant human IL-17A (R&D Systems) was dissolved in PBS and used at the indicated concentrations. Amiloride (Sigma, St. Louis, MO) and BaCl₂ (Sigma) were dissolved in distilled, deionized water as 1,000× stocks. Bumetanide (Sigma) was dissolved in ethanol as a 1,000× stock. Forskolin (EMD; Calbiochem, San Diego, CA) was dissolved in dimethyl sulfoxide (DMSO) as a 5,000× stock. CFTRinh-172 (Calbiochem) was dissolved in DMSO as a 1,000× stock. DNDS (Invitrogen, Carlsbad, CA) was dissolved in normal buffer as a 100 mM stock and used at a final concentration of 3 mM. 4,4′-diisothiocyanostilbene-2,2′-disulfonic acid (Invitrogen) was dissolved in DMSO as a 1,000× stock.

Statistical analysis.

Unless otherwise stated, values are means ± SE. Statistical comparisons were made with Prism 4 (GraphPad Software, San Diego, CA). Comparisons between groups were made with either Student's t-test or with two-way ANOVA followed by Bonferroni posttest analysis for significance between groups. Significance was defined as a P value ≤0.05.

RESULTS

A total of 213 HBE filters representing five nondisease donors were used for these studies. Each experiment was performed on cells from one to three donors. At baseline, the mean R_T and I_SC of untreated filters studied in normal bathing solution were 954 ± 555 Ω·cm² and 8.6 ± 6.4 μA/cm², respectively. For filters treated with 50 ng/ml IL-17A, the mean baseline R_T and I_SC were 755 ± 390 Ω·cm² and 9.7 ± 5.8 μA/cm², not different from the untreated control cells. CF cells were obtained from three different tissue samples.

Dose-response studies.

In an initial set of experiments, we examined the effects of increasing doses of IL-17A on basal, amiloride-sensitive, and forskolin-stimulated ion transport in HBE cells. HBE cells were incubated in the presence of 0, 10, or 50 ng/ml IL-17A in the basolateral medium for 48 h. We chose this time based on previous studies that showed effects of cytokines on ion transport in epithelial cells at 48 h (3, 31). Following IL-17A pretreatment, we studied I_SC changes in response to sequential addition of amiloride, forskolin, bumetanide, and Ba²⁺ (Fig. 1A). Compared with vehicle, IL-17A pretreatment did not change baseline or amiloride-sensitive I_SC. However, it augmented the magnitude of forskolin-stimulated I_SC in a dose-dependent fashion (Fig. 1B), with a maximum effect at 50 ng/ml (20.3 ± 1.9 μA/cm² for IL-17A vs. 10.5 ± 1.7 μA/cm² for vehicle, P < 0.01). In contrast, the magnitude of the I_SC responses to subsequent bumetanide and Ba²⁺ additions were not different between IL-17A-treated cells and controls (Fig. 1A). Consequently, the I_SC in the IL-17A-treated cells was higher than that in control cells after forskolin stimulation and subsequent exposure to blockers of Cl⁻ secretion. These data suggest that the additional, IL-17A-dependent, cAMP-activated current was not the result of Cl⁻ secretion, because Cl⁻ secretion depends on bumetanide-sensitive NKCC-1 mediated Cl⁻ entry across the basolateral membrane, as well as Ba²⁺-sensitive basolateral K⁺ conductance for hyperpolarization of the cell and maintenance of a driving force for Cl⁻ exit (12). Nonetheless, the additional forskolin-stimulated I_SC in IL-17A-treated cells was completely inhibited by ouabain (Fig. 1C), suggesting that it depended on the maintenance of transmembrane ion gradients.

Time course studies.

To investigate the time dependence of IL-17A-induced changes in ion transport, IL-17A either was added to the serosal bath in the Ussing chamber 0.25 h before addition of amiloride or was added to cells in differentiation media in the tissue culture incubator for 4, 24, or 48 h. No effects of IL-17A were observed at 0.25 or 4 h (Fig. 2, A and B); whereas additional forskolin-stimulated I_SC was observed at 24 h and continued to increase during the subsequent 24 h (P < 0.05; Fig. 2, C and D). The appearance of an effect at 24 h, but not at 4 h, suggests that new protein production likely is required to produce the additional IL-17A-induced, forskolin-stimulated I_SC. Because the most robust response to IL-17A was at 48 h at 50 ng/ml, we used these conditions for the remainder of the studies.

Fig. 2.

Time dependence of IL-17A-induced changes in ion transport. Human bronchial epithelial (HBE) cells were incubated with IL-17A (50 ng/ml) in the basolateral medium for 0.25 (A), 4 (B), 24 (C), or 48 (D) h (n = 4–6 filters/group, 2 donors). A significant response to IL-17A treatment was not seen until 24 h, which increased further by 48 h. *P < 0.05 compared with 0.25 h (*) and compared with 24 h (#).

HCO₃⁻ inhibition studies.

A previous demonstration of cAMP-activated, bumetanide-insensitive HCO₃⁻ secretion across human airways (15) suggested that the IL-17A-dependent, bumetanide-insensitive, forskolin-stimulated I_SC could be associated with HCO₃⁻ secretion. To test this hypothesis, we sequentially exposed amiloride-inhibited, forskolin-stimulated HBE cells to acetazolamide, an inhibitor of carbonic anhydrase, and DNDS, an inhibitor of Na⁺/HCO₃⁻ cotransporters and Cl⁻/HCO₃⁻ exchangers (Fig. 3A). Acetazolamide increased I_SC in control cells (P < 0.05 compared with no change by one-way t-test; Fig. 3B) but decreased I_SC in IL-17A-treated cells (P < 0.05 when compared with no change or with the increase seen in controls). Serosal addition of DNDS decreased I_SC in both controls and IL-17A-treated cells (Fig. 3B, DNDS), but the decrease in IL-17A-treated cells was larger (−1.1 ± 0.3 μA/cm² for vehicle vs. −3.9 ± 0.9 μA/cm² for IL-17A, P < 0.05, Mann-Whitney test). Therefore, the total inhibition of I_SC by acetazolamide and DNDS was greater in the IL-17A-treated group than in controls (P < 0.05, 2-way ANOVA with Bonferroni posttests; Fig. 3C). Notably, the difference in mean total inhibition of I_SC by acetazolamide and DNDS (8 μA/cm²) was close in magnitude to the increase in forskolin-stimulated I_SC observed in IL-17A-treated HBE cells (6.5 μA/cm²) (Fig. 3C). These data indicate that the additional forskolin-stimulated I_SC in IL-17A-treated cells was mediated by an acetazolamide- and DNDS-sensitive mechanism, consistent with HCO₃⁻ secretion. Of note, the IL-17A-dependent I_SC was not inhibited by mucosal addition of DIDS, a nonspecific inhibitor of Cl⁻/HCO₃⁻ exchangers (22) (data not shown).

Fig. 3.

IL-17A-induced I_SC is eliminated by inhibitors of HCO₃⁻ secretion. A: I_SC trace demonstrating that IL-17A-induced I_SC is inhibited by sequential addition of acetazolamide and DNDS. B: mean changes in I_SC with sequential addition of amiloride, forskolin, acetazolamide, and DNDS (n = 7 filters/group, 2 donors). C: mean changes in I_SC with acetazolamide and DNDS inhibition summed to demonstrate total inhibition of HCO₃⁻ secretion in control cells vs. IL-17A-treated cells. *P < 0.05 and **P < 0.001 compared with vehicle.

Ion substitution studies.

To verify that the additional forskolin-stimulated I_SC in IL-17A-treated cells was primarily the result of HCO₃⁻ secretion, we performed I_SC measurements in HCO₃⁻-free solutions buffered to pH 7.4 with HEPES/NaOH and gassed with air. In these nominally HCO₃⁻-free conditions, there was no difference in forskolin-stimulated I_SC between vehicle and IL-17A-treated cells [16.9 ± 2.2 μA/cm² vs. 14.6 ± 0.3 μA/cm², respectively, P = not significant (Fig. 4A)]. These results support the hypothesis that IL-17A promotes HCO₃⁻ secretion in HBE cells under voltage-clamped conditions. To investigate the Na⁺ dependence of IL-17A-induced HCO₃⁻ secretion, I_SC was measured in HBE cells bathed in Na⁺-free solutions. Under these conditions, both vehicle and IL-17A-treated cells had reduced forskolin-activated I_SC. However, IL-17A treated cells had greater I_SC than vehicle-treated cells (2.3 ± 0.3 μA/cm² for vehicle vs. 4.6 ± 0.5 μA/cm² for IL-17A, P < 0.01, 2-way ANOVA with Bonferroni posttests) (Fig. 4B), suggesting that the IL-17A-induced I_SC was composed of both Na⁺-dependent and Na⁺-independent mechanisms.

Fig. 4.

IL-17A-induced I_SC is dependent on the presence of both HCO₃⁻ and Cl⁻. A: HBE filters were mounted in HCO₃⁻-free solutions buffered with HEPES and gassed with air. In the absence of soluble HCO₃⁻, there was no difference in forskolin-stimulated I_SC between non-IL-17A-treated and IL-17A-treated cells (n = 6 filters/group, 1 donor). B: HBE filters were mounted in symmetric Na⁺-free solutions and treated sequentially with amiloride and forskolin. Data demonstrate mean forskolin-stimulated I_SC ± SE, which is significantly greater in IL-17A-treated cells compared with control. Note the different y-axis scale because the total amount of forskolin-stimulated I_SC is significantly reduced in Na⁺-free solutions when compared with normal conditions. C: HBE filters were mounted in symmetric Cl⁻-free solutions and treated sequentially with amiloride and forskolin. Data represent mean forskolin-stimulated I_SC under Cl⁻-free conditions, which is not different between IL-17A-treated cells and controls (n = 6 filters/group, 2 donors). D: HBE cells were mounted in asymmetric conditions with Cl⁻-free solution on the serosal side and normal Cl⁻- and HCO₃⁻-containing solution on the mucosal side (n = 4 filters/group, 2 donors). Removal of serosal Cl⁻ does not inhibit the effect of IL-17A on forskolin-stimulated I_SC. *P < 0.05 by unpaired t-test. **P < 0.01 by Mann-Whitney test.

To investigate the Cl⁻ dependence of IL-17A-induced HCO₃⁻ secretion, I_SC measurements were performed in HBE cells bathed in Cl⁻-free, HCO₃⁻-containing solutions. Forskolin-stimulated I_SC in symmetrical Cl⁻-free conditions was the same for vehicle and IL-17A-treated cells (3.8 ± 0.4 vs. 3.7 ± 0.4 μA/cm², respectively) (Fig. 4C). However, when I_SC measurements were performed in the absence of serosal Cl⁻ and presence of mucosal Cl⁻, IL-17A-treated cells had higher forskolin-stimulated I_SC (P < 0.05, unpaired t-test with Welch's correction; Fig. 4C). Taken together, these findings suggest that the I_SC induced by IL-17A requires the presence of HCO₃⁻ and Cl⁻ in the mucosal bath solution.

Chloride addition studies.

To test the hypothesis that IL-17A promoted Cl⁻-dependent HCO₃⁻ secretion, we performed experiments in which Cl⁻ was added to the mucosal bath as a transport stimulus. In the first set of experiments, HBE cells were mounted in Ussing chambers in symmetrical Cl⁻-free solutions. After addition of amiloride and forskolin, 30 mM NaCl was added to the mucosal solution, and 30 mM sodium gluconate was added to the serosal solution. Under these conditions, there is a driving force for movement of Cl⁻ across the epithelium from the mucosal to the serosal solution. The predicted negative I_SC resulting from mucosal-to-serosal movement of Cl⁻ down its concentration gradient was observed in control cells (−0.15 μA/cm²; Fig. 5A). In IL-17A-treated cells, however, addition of Cl⁻ to the mucosal bath caused a sustained increase in I_SC of 4.0 ± 0.4 μA/cm² (Fig. 5A), different from the change seen in control cells (P < 0.01; Fig. 5B). To test the hypothesis that the I_SC observed in IL-17A-treated cells under these conditions was the result of Cl⁻-dependent HCO₃⁻ secretion, we performed the same NaCl addition experiment in the absence of HCO₃⁻. In the absence of HCO₃⁻, both untreated and IL-17A-treated cells responded to the addition Cl⁻ to the mucosal bath with a decrease in I_SC (Fig. 5C). Therefore, in the absence of HCO₃⁻, the polarity of the observed I_SC changes is consistent with the direction of the imposed Cl⁻ gradient. Taken together, these results strongly support a model of a novel mucosal Cl⁻-dependent HCO₃⁻ secretion in IL-17A-treated HBE cells.

Fig. 5.

IL-17A induces Cl⁻-dependent HCO₃⁻ secretion. HBE cells were mounted in Ussing chambers with Cl⁻-free solutions on both the serosal and mucosal sides (n = 3 - 6 filters/group, 3 donors). After addition of amiloride and forskolin, 30 mM NaCl was added to the mucosal bath, and 30 mM sodium gluconate was added to the serosal bath. A: representative I_SC traces demonstrating the effect of addition of Cl⁻ to the mucosal bath. In the presence of HCO₃⁻, IL-17A-treated cells respond to mucosal Cl⁻ addition with an increase in I_SC (solid line). This I_SC change in is opposite to the expected I_SC change for movement of Cl⁻ down its concentration gradient. B: mean I_SC changes for control and IL-17A-treated cells with mucosal Cl⁻ addition. C: mean I_SC changes for control and IL-17A-treated cells with mucosal Cl⁻ addition in the absence of HCO₃⁻. In the absence of HCO₃⁻, addition of mucosal Cl⁻ results in a decrease in I_SC in both control and IL-17A-treated cells. This I_SC change is consistent with the expected I_SC generated by movement of Cl⁻ down its concentration gradient. **P < 0.01 by 2-way ANOVA.

Dependence of HCO₃⁻ secretion on cystic fibrosis transmembrane conductance regulator.

Because the IL-17A-enhanced I_SC was dependent on cAMP, we assessed the possible role of cystic fibrosis transmembrane conductance regulator (CFTR) in the IL-17A-induced, forskolin-stimulated I_SC. Primary bronchial epithelial cells from CF patients with ΔF508 homozygous genotype were treated with IL-17A and studied using the same drug addition protocols as above. As anticipated, the magnitude of the forskolin-stimulated I_SC was much smaller in CF cells than in non-CF cells. Notably, IL-17A was without effect on forskolin-stimulated I_SC in CF cells (Fig. 6A). This result demonstrates a requirement for functional CFTR expression in IL-17A-induced Cl⁻-dependent HCO₃⁻ secretion. To test this further, normal HBE cells were treated with the CFTR inhibitor CFTRinh-172 (30 μM for 90 min) before study in the Ussing chamber. CFTRinh-172 pretreatment inhibited forskolin-stimulated I_SC in control cells by 55%. Importantly, it eliminated the difference in forskolin-stimulated I_SC between controls and IL-17A-treated cells (Fig. 6B). These data further support the idea that the I_SC response to IL-17A pretreatment depends on CFTR function.

Fig. 6.

Cystic fibrosis transmembrane conductance regulator (CFTR) dependence of IL-17A-induced HCO₃⁻ secretion. A: HBE cells from cystic fibrosis (CF) patients were stimulated with IL-17A at 0 or 50 ng/ml in the basolateral medium for 48 h before study (n = 6 filters/group, 3 donors). We observed very little forskolin-stimulated I_SC in either group, and there was no difference between IL-17A-treated cells and controls (P = 0.09). B: normal HBE cells were incubated with vehicle (dimethyl sulfoxide) or CFinh-172 (30 μM) in normal bath solution applied to the apical surface for 90 min before study (n = 5 filters/group, 2 donors). Pretreatment with CFTRinh-172 eliminated the ability of IL-17A to augment forskolin-stimulated I_SC.

Low-light level, live-cell fluorescence microscopy.

To confirm that the additional forskolin-stimulated I_SC seen in IL-17A-treated HBE cells reflected HCO₃⁻ secretion, we measured pH of the apical solution bathing control and IL-17A-treated cells using low-light level fluorescence microscopy of the pH indicator dye SNARF-5. The relationship of the F₆₄₀/F₅₈₀ emission ratio of SNARF-5-free acid to mucosal fluid pH was found to be linear over the pH range examined (Fig. 7A). To monitor changes in mucosal solution pH over time, we placed 100 μl of a HCO₃⁻-free solution with low buffering capacity on the mucosal surface of the cells while normal buffer was perfused across the serosal surface at a rate of ∼1–2 ml/min. Under these conditions, HBE cells acidified the mucosal solution at a constant rate over a 10-min observation period (Fig. 7B, left). The rate of acidification was attenuated when the mucosal solution was replaced by one with stronger buffering capacity (10 mM HEPES, pH 7.4) (Fig. 7B, right), suggesting that the technique specifically measured H⁺ equivalent secretion in the mucosal solution.

Fig. 7.

IL-17A augments forskolin-stimulated HCO₃⁻ secretion by HBE cells. A: calibration of mucosal SNARF-5-free acid fluorescence ratio [ratio of fluorescence emissions at 640 and 580 nm (F₆₄₀/F₅₈₀)] to pH is linear from pH 6.8 to pH 8.4. B: mucosal fluid pH acidifies over time (left), and the rate of acidification can be attenuated by increasing the buffering capacity of the mucosal fluid by increasing HEPES concentration to 10 mM (right). C: representative traces of mucosal fluid pH vs. time in controls and IL-17A-treated cells. D: mean change in rate of mucosal fluid pH change at baseline and after forskolin stimulation (P < 0.01 by Mann-Whitney).

To examine the effects of IL-17A, we compared the rates of mucosal solution pH changes in vehicle-treated and IL-17A-treated cells. For these experiments, we measured a basal rate of acidification during an initial 5-min period before adding 2 μM forskolin to the serosal solution. After 2–3 min to allow for forskolin-mediated activation of cAMP signaling, the rate of pH change was measured over the subsequent 5 min. Representative traces are shown in Fig. 7C. The baseline rate of mucosal acidification was not different between vehicle and IL-17A-treated cells. Both vehicle- and IL-17A-treated cells responded to forskolin stimulation by decreasing the rate of mucosal fluid acidification (P < 0.01 for both vehicle and IL-17A-treated cells). However, whereas vehicle-treated cells continued to acidify the mucosal fluid after forskolin stimulation, IL-17A-treated cells alkalinized the mucosal fluid after forskolin stimulation (P < 0.01 compared with vehicle by Mann-Whitney, n = 6 filters total representing 3 donors; Fig. 7D).

We also examined the effect of IL-17A on resting airway surface pH in vehicle and IL-17A-treated cells using SNARF-labeled dextran dispersed in PFC-77. The application of 20 μl of SNARF-dextran in PFC labeled the mucosal surface of the cells in the absence of additional buffer. The F₆₄₀-to-F₅₈₀ ratio of SNARF-1 dextran is linear between pH 6.5 and 7.5 (Fig. 8A). The surface pH of IL-17A-treated cells (7.2 ± 0.2) was more alkaline than that of controls [6.6 ± 0.1, P < 0.05 (Fig. 8B)].

Fig. 8.

IL-17A alkalinizes surface pH in HBE cells. SNARF-1 dextran (1 mg/ml in perfluorocarbon) was added to the mucosal surface of mature HBE cells, and the F₆₄₀-to-F₅₈₀ ratio was measured every min. A: calibration of SNARF-1 dextran in aqueous solutions of pH 6.5, 7, and 7.5. B: surface pH measurements from 8 vehicle (open circles) and IL-17A-treated (filled circles) filters representing 3 donors. Note that resting surface pH is more alkaline in IL-17A-treated cells than controls. Horizontal bar represents group mean. *P < 0.05 by Mann-Whitney.

To examine whether IL-17A affected intracellular pH, we loaded cells with SNARF-5-AM. For these experiments, we followed a procedure similar to that described for measurement of mucosal fluid pH except that the mucosal surface of the cells was submerged in 100 μl of HCO₃⁻-buffered solution in the presence of 95% O₂-5% CO₂, and all solutions contained 1 mM probenicid. The F₆₄₀-to-F₅₈₀ ratio of intracellular SNARF-5 is linear between pH 6.5 and 7.5 (Fig. 9A). Both control and IL-17A-treated cells displayed a stable baseline 5–7 min after initiating perfusion, with an intracellular pH of 7.33 ± 0.02 and 7.34 ± 0.02, respectively. After forskolin stimulation, intracellular pH in control and IL-17A-treated cells was 7.36 ± 0.02 and 7.38 ± 0.02, respectively. These results reveal no difference in intracellular pH between control and IL-17A-treated cells.

Fig. 9.

Intracellular pH is not different in IL-17A-treated cells vs. controls. Mature HBE cells were loaded with SNARF-5-AM before mounting on the microscope stage. Cells were overlayed with 100 μl of HCO₃⁻-buffered solution and kept in the presence of 95% O₂-5% CO₂. Cells were perfused on the serosal side with CO₂-gassed HCO₃⁻-buffered solution. F₆₄₀/F₅₈₀ was measured every min until stability after which the serosal solution was changed to one containing 2 μM forskolin. A: calibration of intracellular pH using SNARF-5-AM after equilibration in solutions of pH 6.5, 7, and 7.5. B: mean intracellular pH ± SE in controls (open circles) vs. IL-17A-treated cells (filled circles) (n = 8 filters from 3 donors). There is no difference between IL-17A-treated cells and controls in either baseline pH or change in pH with forskolin.

DISCUSSION

Our data demonstrate that IL-17A treatment of mature, well-differentiated HBE cells specifically augments forskolin-stimulated I_SC that likely represents a novel HCO₃⁻ secretion. The conclusion that the IL-17A-enhanced I_SC is the result of HCO₃⁻ secretion is supported by the sensitivity of the additional current to inhibitors of HCO₃⁻ production and transport, acetazolamide and DNDS (Fig. 3, B and C), and by the absence of the IL-17A-induced current when HCO₃⁻/CO₂ were removed from the mucosal and serosal solutions (Fig. 4A). This conclusion also is supported by direct measurements of mucosal fluid pH (Fig. 7D) that demonstrated enhanced forskolin-stimulated secretion of base equivalents in IL-17A-treated cells. By pharmacological and ion substitution experiments, we conclude that this IL-17A-induced HCO₃⁻ secretion mechanism is CFTR-dependent, mucosal Cl⁻-dependent, partially Na⁺-dependent, and inhibitable by serosal DNDS but not by mucosal DIDS. Accordingly, the HCO₃⁻ secretion pathway likely involves HCO₃⁻ entry across the basolateral membrane on a stilbene-sensitive Na+-HCO₃⁻ cotransporter and possibly an electrogenic Cl⁻/HCO₃⁻ exchanger. HCO₃⁻ exit across the apical membrane takes place via a CFTR-dependent mechanism that may represent direct conductance of HCO₃⁻ through CFTR or exit of HCO₃⁻ on a Cl⁻/HCO₃⁻ exchanger that is dependent on CFTR.

HCO₃⁻ can enter epithelial cells across the basolateral membrane via Na⁺-dependent or Na⁺-independent mechanisms, or be generated intracellularly through metabolic production via carbonic anhydrase activity (6). HCO₃⁻ secretion by HBE cells requires a sufficient electrochemical driving force and a HCO₃⁻ exit pathway across the apical membrane. Under the ionic conditions used in the present study, the apical membrane equilibrium potentials for Cl⁻ and HCO₃⁻ are approximately −30 and −15 mV, respectively, based on previous studies in Calu-3 and HBE cells (4, 20). Blocking Na⁺ conductance at the apical membrane with amiloride hyperpolarizes the apical membrane, creating an enhanced driving force for anion exit across the apical membrane (29). With the application of forskolin, anion conductance is activated, and the apical membrane potential is expected to approach the Cl⁻ equilibrium potential (20), remaining sufficiently hyperpolarized to provide a driving force for HCO₃⁻ secretion. CFTR may provide an exit pathway for HCO₃⁻ in HBE cells, and, therefore, HBE cells can secrete HCO₃⁻ in response to forskolin (27).

In IL-17A-treated HBE cells, simultaneous Cl⁻ and HCO₃⁻ secretion may occur by permeation of both anions through CFTR. In support, absence of CFTR anion conductance, either in ΔF508 homozygous bronchial epithelial cells (Fig. 6A) or by pharmacological inhibition of CFTR in normal HBE cells (Fig. 6B), abolished the effect of IL-17A on I_SC. Alternatively, it is possible that HCO₃⁻ is secreted via a non-CFTR pathway in IL-17A-treated HBE cells, but the driving force for HCO₃⁻ secretion across the apical membrane is unfavorable in the absence of CFTR. However, this explanation seems unlikely since the presence of CFTR in the apical membrane would be expected to be a depolarizing force on apical membrane potential. Finally, CFTR dependence could reflect IL-17A-induced activation of an anion exchanger that is dependent on CFTR for its activity or expression (19, 28).

In support of the idea that IL-17A activated an anion exchange mechanism, the IL-17A-induced change in I_SC was dependent on the presence of both Cl⁻ and HCO₃⁻ (Fig. 4). Also, under initially Cl⁻-free conditions, addition of Cl⁻ to the mucosal bath evoked a HCO₃⁻-dependent increase in I_SC in IL-17A-treated HBE cells (Fig. 5). This result is consistent with a model in which Cl⁻ moving down its electrochemical gradient is exchanged at either the apical or basolateral membrane for HCO₃⁻ in a stoichiometry of >1 HCO₃⁻:1 Cl⁻. Together these data suggest the involvement of Cl⁻/HCO₃⁻ exchange as one possible mechanism responsible for IL-17A-induced HCO₃⁻ secretion across HBE cells.

IL-17A-enhanced forskolin-stimulated I_SC was observed only after >4 h of exposure to IL-17A, suggesting that new protein expression was required. Accordingly, we hypothesize that IL-17A increases the expression, trafficking, or activation of a HCO₃⁻ transport pathway. This hypothesis is supported by our measurements of a more alkaline surface pH in IL-17A-treated cells compared with controls (Fig. 8). Moreover, the mucosal surface pH values that we measured are consistent with previous in vitro and in vivo measurements of airway surface pH in normal and inflamed states (reviewed in Ref. 6). Despite enhanced HCO₃⁻ secretion in IL-17A-treated cells, intracellular pH was not different in IL-17A-treated cells and controls (Fig. 9B). This result suggests that HCO₃⁻ transport at the apical and basolateral membrane is coordinated such that intracellular HCO₃⁻ concentration remains constant during enhanced secretion.

Our results do not address the molecular identity of the transport pathway responsible for the effect of IL-17A on I_SC. Members of the SLC26 sulfate transporter family function as Cl⁻/HCO₃⁻ exchangers that interact with CFTR (19). Of note, previous studies performed under noninflammatory conditions demonstrated a role for SLC26a3 in HBE cells (28), whereas stimulation of HBE cells with IL-4 was associated with upregulation of SLC26a4 activity (26). Together, these data suggest that airway epithelial cells are capable of responding to external stimuli, including inflammatory cytokines, with alterations in the expression of HCO₃⁻ transporters, including Cl⁻/HCO₃⁻ exchangers.

Inflammatory cytokines from both the Th1 and Th2 families can affect the ion transport phenotype of HBE cells. For example, IL-1β increases CFTR activity, expression, and HCO₃⁻ concentration in samples of mucosal fluid from HBE cell cultures (10). Interferon-γ, another Th1 cytokine, decreases Na⁺ absorption and increases UTP-stimulated anion secretion by HBE cells (7). Both IL-4 and IL-13, Th2 cytokines, decrease Na⁺ absorption and increase anion secretion in HBE cells in response to cAMP and purinergic agonists (3, 9). Here, we demonstrated that IL-17A, a Th17 cytokine, increased forskolin-stimulated HCO₃⁻ secretion by HBE cells. Therefore, there may be a common signaling pathway through which these cytokines act in HBE cells to augment both physical and immunological innate defense mechanisms through altered ion transport.

Many physiological processes of airway epithelia are upregulated by IL-17A, including mucin production (2), chemokine elaboration (18, 23), and neutrophil recruitment (30). Because mucin secretion and neutrophil degranulation would be expected to acidify the airway surface, epithelial cells may also increase HCO₃⁻ secretion in response to IL-17A to maintain pH homeostasis (10). Therefore, in diseases in which CFTR conductance is impaired, such as CF and possibly chronic obstrucive pulmonary disease (1), the innate defense mechanisms of the airways may be altered by inadequate epithelial responses to inflammatory stimuli.

GRANTS

This work was supported by National Institutes of Health Grants K08 HL-081080 (J. L. Kreindler), R01 HL-079142, (J. K. Kolls), P50 HL-084932 (J. K. Kolls, R. A. Frizzell), P30 DK-72506, and RO1 DK-68196 (R. A. Frizzell) and CFF RDP to the University of Pittsburgh. R. J. Lee was supported by predoctoral fellowships from the National Science Foundation and the Cystic Fibrosis Foundation.