A highly potent agonist to protease-activated receptor-2 reveals apical activation of the airway epithelium resulting in Ca²⁺-regulated ion conductance

Abstract

The airway epithelium provides a barrier that separates inhaled air and its various particulates from the underlying tissues. It provides key physiological functions in both sensing the environment and initiating appropriate innate immune defenses to protect the lung. Protease-activated receptor-2 (PAR₂) is expressed both apically and basolaterally throughout the airway epithelium. One consequence of basolateral PAR₂ activation is the rapid, Ca²⁺-dependent ion flux that favors secretion in the normally absorptive airway epithelium. However, roles for apically expressed PAR₂ activation have not been demonstrated, in part due to the lack of specific, high-potency PAR₂ ligands. In the present study, we used the newly developed PAR₂ ligand 2at-LIGRLO(PEG₃-Pam)-NH₂ in combination with well-differentiated, primary cultured airway epithelial cells from wild-type and PAR₂^−/− mice to examine the physiological role of PAR₂ in the conducting airway after apical activation. Using digital imaging microscopy of intracellular Ca²⁺ concentration changes, we verified ligand potency on PAR₂ in primary cultured airway cells. Examination of airway epithelial tissue in an Ussing chamber showed that apical activation of PAR₂ by 2at-LIGRLO(PEG₃-Pam)-NH₂ resulted in a transient decrease in transepithelial resistance that was due to increased apical ion efflux. We determined pharmacologically that this increase in ion conductance was through Ca²⁺-activated Cl⁻ and large-conductance K⁺ channels that were blocked with a Ca²⁺-activated Cl⁻ channel inhibitor and clotrimazole, respectively. Stimulation of Cl⁻ efflux via PAR₂ activation at the airway epithelial surface can increase airway surface liquid that would aid in clearing the airway of noxious inhaled agents.

the airway epithelium is the first site of cellular contact for inhaled allergens and other insults. It provides a barrier to protect underlying submucosa from inhalation threats and orchestrates critical aspects of airway innate immunity, such as inflammation, airway remodeling, and mucociliary clearance (16, 34). A subset of inhaled allergens (e.g., house dust mites, cockroaches, and various fungi) has been associated with the onset and pathogenesis of allergic asthma (36, 40). Allergic asthma is a chronic disease associated with airway remodeling that includes inflammation, secretory alterations, and airway hyperreactivity culminating in airway obstruction (20). Although allergens associated with asthma (i.e., asthmagens) are quite diverse, many of them contain proteases that can interact with the airway epithelium through a variety of mechanisms that include the activation of protease-activated receptor-2 (PAR₂) (22).

PAR₂ is a G protein-coupled receptor (GPCR) belonging to a family of GPCRs composed of four members (2). Unlike typical GPCRs, PAR ligand activation sequences are contained within the extracellular portion of the GPCR. Proteolytic cleavage of the NH₂-terminus reveals the tethered ligand [SLIGKV for humans and SLIGRL for rodents (30)], which binds an activation pocket, leading to cellular signaling cascades. Exogenous asthmagen-associated proteases that have been demonstrated to activate PAR₂ include those from the house dust mite, cockroaches, and fungi (22, 23, 32). Individuals with asthma are often sensitized to these protease-containing allergens; thus, understanding roles for PAR₂ in normal and pathological airway physiology have been an intense research focus (2, 17, 21, 33, 41).

The full impact of PAR₂ in allergic asthma is quite complex. PAR₂ activation has been linked to long-term, detrimental airway inflammation and hypersecretion as well as beneficial bronchodilation (7, 8, 10, 12, 19, 38, 42). It has been suggested that these diametrically opposed outcomes may be separable by the activation of separate signaling pathways downstream of PAR₂ (43). In studies examining more rapid responses in the airway epithelium, PAR₂ has been linked to changes in airway epithelial ion conductance (5, 9, 25, 31, 35, 37). Interestingly, studies that have uncovered roles for PAR₂ in ion conductance changes across the airway epithelium have focused on the basolateral activation of PAR₂ (1, 24, 25, 35). This has occurred despite immunocytochemical evidence for apical expression of PAR₂ in mouse and human tissues (6, 8, 25), where the receptor would most likely interact with inhaled proteases in healthy individuals. Further complicating the role of PAR₂ in airway epithelial ion movement is the finding that, in some mouse strains, PAR₂ ligands (i.e., SLIGRL-NH₂) have been shown to activate ion conductance changes that appear independent of PAR₂ (1).

In the present study, we evaluated the potency and specificity of a recently developed, lipidated PAR₂ agonist, 2at-LIGRLO(PEG₃-Pam)-NH₂ (14), for use in well-differentiated, primary cultured airway epithelial cells from wild-type and PAR₂^−/− mice. We first examined PAR₂ signaling with digital imaging microscopy of intracellular Ca²⁺ concentration ([Ca²⁺]_i) changes. We next examined apical activation of PAR₂ in Ussing chamber experiments designed to evaluate PAR₂ effects on transepithelial resistance (TER) and alterations in ion conductance. We demonstrated that novel apical stimulation of PAR₂ in the primary airway epithelium leads to the activation of Ca²⁺-regulated ion channels consistent with a subsequent apical efflux of Cl⁻. The availability of potent and specific PAR₂ agonists suggests that apical PAR₂ activation can contribute to mucociliary clearance and thus aid in the rapid clearing of asthmagens from healthy airways.

MATERIALS AND METHODS

Materials.

Cellgro DMEM:F-12 was from Mediatech (Manassas, VA). Lechner and LaVeck basal media, HBSS, glutamax, penicillin, and streptomycin were from InVitrogen (Carlsbad, CA). Fibronectin, collagen type I, and Nu-Serum were from Becton-Dickinson (Franklin Lakes, NJ). FBS and clotrimazole were from Sigma-Aldrich (St. Louis, MO). CaCCinh-A01 was from Tocris Bioscience (Bristol, UK). Semipermeable filters were Corning Costar 6.5-mm Transwell filters with a 0.4-μm Pore Polyester Membrane Insert (Lowell, MA). All other chemicals were from Sigma-Aldrich or Fisher Scientific (Pittsburgh, PA).

Tissue culture methods.

Animal protocols were approved by the Institutional Animal Care and Use Committee of The University of Arizona. Primary mouse tracheal epithelial (MTE) cells were cultured as previously described (39). C57Bl/6 wild-type and PAR₂^−/− mice on a C57Bl/6 background were used for these cultures. Isolated primary MTE cell cultures for digital imaging microscopy were seeded onto glass coverslips coated with rat tail collagen (11), cultured at 37°C with 5% CO₂, and fed every other day in full culture medium. Although these cells establish cell junctions, they do not become well differentiated. MTE cells used in Ussing chamber experiments were seeded onto 6.5-mm semipermeable filters coated with a collagen-fibronectin-BSA matrix and cultured at 37°C with 5% CO₂. MTE cells were grown in full culture medium to allow for confluent monolayers. After cell monolayers reached a TER of >500 Ω·cm² (∼5 days), apical media were removed to establish an air interface, and media were changed to a 1:1 mixture of DMEM and Ham's F-12, 1% penicillin-streptomycin, and 2% Nu-Serum. These cells become a mixed population of well-differentiated cells with clearly established cilia.

[Ca²⁺]_i measurements.

Cell monolayers were washed with modified HBSS (1.3 mM CaCl₂, 5.0 mM KCl, 0.3 mM KH₂PO₄, 0.5 mM MgCl₂, 0.4 mM MgSO₄, 137.9 mM NaCl, 0.3 mM Na₂PO₄, and 1% glucose additionally buffered with 25 mM HEPES, pH 7.4) and loaded for 45–60 min with 5 μM fura 2-AM in HBSS. Cells were removed from fura 2-AM loading solution and placed back in matching HBSS for at least 20 min before Ca²⁺ imaging. Fura-2 fluorescence was observed on an Olympus IX70 microscope with a ×40 oil-immersion objective after alternating excitation between 340 and 380 nm by a 75-W Xenon lamp linked to a Delta Ram V illuminator (PTI, London, ON, Canada) and a gel optic line. Images of emitted fluorescence above 505 nm were recorded by an ICCD camera (PTI) or an EVO-512 camera (Roper Scientific, Tucson, AZ) and simultaneously displayed on a color monitor. The imaging system was under software control (ImageMaster, PTI) and collected a ratio approximately every 0.6 s. [Ca²⁺]_i was calculated by ratiometric analysis of fura-2 fluorescence using equations originally published in Ref. 18. A typical field of view contained 30–40 cells at a resting [Ca²⁺]_i estimated to be ∼75 nM. A change in [Ca²⁺]_i was considered positive if the cell increased its [Ca²⁺]_i twofold or more.

Ussing chamber experiments.

MTE cells on 6.5-mm Transwell filters were mounted in an EasyMount Ussing chamber system (Physiologic Instruments, San Diego, CA) and bathed on both sides with Krebs-Ringer buffer containing (in mM) 115 NaCl, 25 NaHCO₃, 0.4 KH₂PO₄, 2.4 K₂HPO₄, 1.2 CaCl₂, and 1.2 MgCl₂ with 10 glucose. Bath solutions were continuously circulated with a gas lift by bubbling with 95% air and 5% CO₂ at 37°C (pH 7.4). MTE cell monolayers were either current clamped and observed for changes in TER or voltage clamped and monitored for changes in short circuit current (I_sc) and TER with a multichannel voltage/current-clamp VCC MC8 (Physiologic Instruments). For current-clamp experiments, MTE monolayers were clamped to 0 μA, and a 5-μA pulse of 200-ms duration was imposed every 10 s to monitor TER. Changes in I_sc from those experiments were calculated using Ohm's law. Changes in TER (ΔTER) were calculated from the difference between the TER measurement at baseline and the peak measurement change after the addition of agonist. In voltage-clamp experiments, MTE cell monolayers were clamped to 0 mV, and a 5-mV pulse of 200-ms duration was imposed every 10 s. Changes in I_sc (ΔI_sc) were calculated from the difference between the initial I_sc measurement at baseline and the peak measurement change after the addition of agonist. Data were analyzed using Acquire and Analyze software (version 2.3, Physiologic Instruments).

Statistics.

An unpaired Student's t-test was used for comparisons between two data sets. One-way ANOVA with Tukey's multiple-comparison test was used for data sets with more than two comparisons. P values of <0.05 were used to establish significant differences between data sets. Values are expressed as means ± SE unless otherwise noted.

RESULTS

High-potency lipidated ligand specifically activates PAR₂ in primary cultured airway cells.

We initially tested the specificity and concentration-response relationship of the PAR₂ synthetic tethered ligand 2at-LIGRLO(PEG₃-Pam)-NH₂ (14) in the primary MTE cell model using digital imaging microscopy. In each experiment, MTE cells from either wild type or PAR₂^−/− mice were exposed to varying concentrations of 2at-LIGRLO(PEG₃-Pam)-NH₂, and [Ca²⁺]_i was monitored over time. In a typical experiment with wild-type MTE cells, 2at-LIGRLO(PEG₃-Pam)-NH₂ elicited a rise in [Ca²⁺]_i with a population average peak of ∼200 nM that recovered to baseline within 90 s (Fig. 1, A and D). MTE cells from PAR₂^−/− mice, however, showed no measureable change in [Ca²⁺]_i after the addition of 2at-LIGRLO(PEG₃-Pam)-NH₂ (Fig. 1B). To confirm that the loss of [Ca²⁺]_i signaling was due to lack of PAR₂ in MTE cells, 5 μM ATP was applied 3 min after the addition of 2at-LIGRLO(PEG₃-Pam)-NH₂. The ATP response was similar in both wild-type and PAR₂^−/− MTE cells (Fig. 1, A and B, respectively). A concentration-response relationship was established for the percentage of cells that responded with a change in [Ca²⁺]_i to 2at-LIGRLO(PEG₃-Pam)-NH₂ in wild-type MTE cells (EC₅₀: 835 nM, 95% confidence interval: 261 nM–2.68 μM; Fig. 1C).

Fig. 1.

2at-LIGRLO(PEG₃-Pam)-NH₂ activates protease-activated receptor-2 (PAR₂) in mouse tracheal epithelial (MTE) cells. Average intracellular Ca²⁺ concentration ([Ca²⁺]_i) was plotted over time. A: wild-type MTE cells increased [Ca²⁺]_i after 2at-LIGRLO(PEG₃-Pam)-NH₂ (1 μM) and responded to ATP (5 μM) with a second rise in [Ca²⁺]_i. B: 2at-LIGRLO(PEG₃-Pam)-NH₂ (1 μM) evoked no change in [Ca²⁺]_i in PAR₂^−/− MTE cells; however, their response to ATP (5 μM) was similar to wild-type cells. C: concentration-response curve for 2at-LIGRLO(PEG₃-Pam)-NH₂ in wild-type MTE cells [EC₅₀: 835 nM, 95% confidence interval (CI): 261 nM–2.68 μM]. D: color panels of [Ca²⁺]_i from wild-type MTE cells after 2at-LIGRLO(PEG₃-Pam)-NH₂. White lines depict cell borders. The time after application is noted. A color bar indicating approximate [Ca²⁺]_i is given in the bottom left of D. Traces in A and B represent average [Ca²⁺]_i of >35 cells in a single experiment and are means ± SE; n ≥ 3 for each experiment protocol.

Apically applied lipidated PAR₂ agonist alters TER and I_sc in polarized primary airway epithelial cells.

In healthy individuals, asthmagen exposure in the airway lumen first contacts the apical membrane of the airway epithelium. Thus, we evaluated physiological responses of well-differentiated primary MTE cells after apical exposure of PAR₂-specific agonists in Ussing chamber experiments. To better mimic endogenous airway epithelial physiology, MTE cell cultures were first current clamped and measured for TER changes after exposure to 10 μM 2at-LIGRLO(PEG₃-Pam)-NH₂; I_sc was subsequently calculated. In primary MTE cell cultures from wild-type mice, a transient decrease in TER was observed (141.2 ± 42.15 Ω·cm², n = 6; Fig. 2, A and B). However, in MTE cell cultures from PAR^−/− mice, changes in TER in response to 2at-LIGRLO(PEG₃-Pam)-NH₂ were negligible (13.5 ± 12.84 Ω·cm², n = 4; Fig. 2, A and B). The observed loss in TER after the addition of 2at-LIGRLO(PEG₃-Pam)-NH₂ could not be explained by a substantial breakdown in the epithelial barrier as fluorescently labeled permeability markers as small as 600 Da did not significantly penetrate the epithelial barrier of wild-type cells (data not shown). The loss in TER, however, was concurrent with a transient increase in calculated I_sc in wild-type epithelia (6.8 ± 1.5 μA/cm², n = 6; Fig. 2C and D). In contrast, the calculated change in I_sc in response to 2at-LIGRLO(PEG₃-Pam)-NH₂ from PAR₂^−/− epithelia was negligible (1.0 ± 0.18 μA/cm², n = 4; Fig. 2, C and D).

Fig. 2.

2at-LIGRLO(PEG₃-Pam)-NH₂ activation of PAR₂ alters transepithelial resistance (TER) and short circuit current (I_sc) in MTE cells. Ussing chamber experiments were conducted under current-clamp conditions and apical application of 2at-LIGRLO(PEG₃-Pam)-NH₂ (10 μM). A: representative TER traces in wild-type MTE cells and PAR₂^−/− MTE cells. B: quantified peak changes in TER in wild-type MTE cells (−141.2 ± 42.15 Ω_·cm², n = 6) compared with PAR₂^−/− MTE cells (13.5 ± 12.84 Ω_·cm², n = 4). C: representative traces of I_sc in wild-type MTE cells and PAR₂^−/− cells. D: quantification of peak changes in I_sc from wild-type MTE cells (6.8 ± 1.5 μA/cm², n = 6) and PAR₂^−/− MTE cells (1.0 ± 0.18 μA/cm², n = 4). *Significant difference (P < 0.05 by Student's t-test).

Since our initial experiments demonstrated apical PAR₂-specific activation, we next compared the sensitivity of wild-type MTE cell cultures to an activating peptide [SLIGRL-NH₂ (3)], a more potent nonlipidated peptidomimetic PAR₂ agonist [2at-LIGRL-NH₂ (15)], and the most potent, lipidated PAR₂ agonist [2at-LIGRLO(PEG₃-Pam)-NH₂] in Ussing chamber experiments. Under voltage-clamp conditions, 10 μM apical addition of SLIGRL-NH₂ resulted in a minimal increase in I_sc and a correspondingly minimal decrease in TER (ΔI_sc: 2.91 ± 0.285 μA/cm² and ΔTER: −25.27 ± 2.85 Ω·cm², n = 3; Fig. 3, A–C). A slightly larger but not significantly different response from SLIGRL-NH₂ was observed after the apical addition of 2at-LIGRL-NH₂ (ΔI_sc: 5.27 ± 0.394 μA/cm² and ΔTER: −42.00 ± 14.0 Ω·cm², n = 3; Fig. 3, A–C). However, the lipidated PAR₂ ligand 2at-LIGRLO(PEG₃-Pam)-NH₂ evoked a change in I_sc and TER (ΔI_sc: 17.09 ± 1.88 μA/cm² and ΔTER: −270.3 ± 51.1 Ω·cm², n = 6; Fig. 3, A–C) that was significantly different from SLIGRL-NH₂ and 2at-LIGRL-NH₂ changes. In contrast to the results observed after apical addition of PAR₂ agonists, basolateral application of SLIGRL-NH₂, 2at-LIGRL-NH₂, or 2at-LIGRLO(PEG₃-Pam)-NH₂ to wild-type MTE cells all yielded full PAR₂ activation [SLIGRL-NH₂: ΔI_sc 25.05 ± 2.31 μA/cm² and ΔTER −428.3 ± 54.7 Ω·cm², n = 3; 2at-LIGRL-NH₂: ΔI_sc 26.37 ± 2.86 μA/cm² and ΔTER −517.5 ± 19.6 Ω·cm², n = 3; and 2at-LIGRLO(PEG₃-Pam)-NH₂: ΔI_sc 17.27 ± 0.43 μA/cm² and ΔTER −326.4 ± 35 Ω·cm², n = 3; Fig. 4, A–C].

Fig. 3.

2at-LIGRLO(PEG₃-Pam)-NH₂ activates apical PAR₂ in MTE cells. All traces are from Ussing chamber experiments conducted under voltage-clamp conditions. A: representative relative I_sc traces of apical addition of 10 μM PAR₂ agonists demonstrating that only 2at-LIGRLO(PEG₃-Pam)-NH₂ was sufficiently potent to evoke a robust PAR₂ response. B: representative relative TER traces of apical addition of PAR₂ agonists demonstrating that only 2at-LIGRLO(PEG₃-Pam)-NH₂ was sufficiently potent to evoke a robust PAR₂ response. C: quantification of apical addition of PAR₂ agonists showing a minimal response from SLIGRL-NH₂ (ΔI_sc: 2.91 ± 0.285 μA/cm² and ΔTER: −25.27 ± 2.85 Ω_·cm², n = 3) and a slightly higher but still minimal response from 2at-LIGRL-NH₂ (ΔI_sc: 5.27 ± 0.394 μA/cm² and ΔTER: −42.00 ± 14.0 Ω_·cm², n = 3) in contrast to 2at-LIGRLO(PEG₃-Pam)-NH₂, which showed a robust response (ΔI_sc: 17.09 ± 1.88 μA/cm² and ΔTER −270.3 ± 51.1 Ω_·cm², n = 6). *Significant difference (P < 0.05 by one-way ANOVA).

Fig. 4.

Basolateral activation of PAR₂ in MTE cells is similar among PAR₂ agonists. Ussing chamber experiments were conducted under voltage clamp. A: representative relative I_sc traces showing that PAR₂ agonists (10 μM) all evoke a similar and robust increase in I_sc when applied basolaterally. B: representative relative traces of TER after basolateral addition of PAR₂ agonists showing a similar transient reduction in TER. C: quantification of responses to PAR₂ agonists after basolateral application was similar [SLIGRL-NH₂: ΔI_sc 25.05 ± 2.31 μA/cm² and ΔTER −428.3 ± 54.7 Ω_·cm², n = 3; 2at-LIGRL-NH₂: ΔI_sc 26.37 ± 2.86 μA/cm² and ΔTER −517.5 ± 19.6 Ω·cm², n = 3; and 2at-LIGRLO(PEG₃-Pam)-NH₂: ΔI_sc 17.27 ± 0.43 μA/cm² and ΔTER −326.4 ± 35 Ω·cm², n = 3]. ^Significant difference from 2at-LIGRL-NH₂ (P < 0.05 by one-way ANOVA).

Apical activation of PAR₂ alters primary airway epithelial cell ion conductance via Ca²⁺-regulated ion channels.

Because PAR₂ activation in airway epithelia leads to an increase in [Ca²⁺]_i, we examined the potential contribution of the rise in [Ca²⁺]_i to observed PAR₂ changes in ion conductance. MTE cell cultures from wild-type mice were voltage clamped and measured for changes in I_sc and TER after the apical addition of 2at-LIGRLO(PEG₃-Pam)-NH₂ with or without a 30-min preincubation with the Ca²⁺ chelator BAPTA-AM (Fig. 5, A–C). Under normal conditions, apical addition of 2at-LIGRLO(PEG₃-Pam)-NH₂ resulted in ΔI_sc of 17.09 ± 1.88 μA/cm² and ΔTER of −270.3 ± 51.1 Ω·cm² (n = 6). However, after intracellular Ca²⁺ chelation with BAPTA, apical application of 2at-LIGRLO(PEG₃-Pam)-NH₂ resulted in an attenuated ΔI_sc of 2.54 ± 0.346 μA/cm² and ΔTER of −84.07 ± 3.44 Ω·cm² (n = 3).

Fig. 5.

PAR₂ alterations in airway epithelial ion conductance are largely mediated by [Ca²⁺]_i increases. Ussing chamber experiments were conducted under voltage-clamp conditions. A: representative I_sc traces showing that apical addition of 2at-LIGRLO(PEG₃-Pam)-NH₂ (10 μM) alone evoked a robust PAR₂ response that was attenuated with BAPTA (50 μM) preincubation. B: MTE cells pretreated with BAPTA showed a diminished loss of TER after the apical application of 2at-LIGRLO(PEG₃-Pam)-NH₂ (10 μM) compared with cells treated with 2at-LIGRLO(PEG₃-Pam)-NH₂ (10 μM) alone. C: quantification of I_sc and TER demonstrating a significant reduction in the PAR₂ response to 2at-LIGRLO(PEG₃-Pam)-NH₂ (10 μM) in MTE cells pretreated with BAPTA (50 μM) (BAPTA treatment: ΔI_sc 2.54 ± 0.346 μA/cm² and ΔTER −84.07 ± 3.44 Ω·cm², n = 3; BAPTA free: ΔI_sc 17.09 ± 1.88 μA/cm² and ΔTER −270.3 ± 51.1 Ω·cm², n = 6). *Significant difference (P < 0.05 by unpaired Student's t-test).

To elucidate the mechanism(s) by which a PAR₂-evoked rise in [Ca²⁺]_i increased airway epithelial ion conductance, we examined the role of Ca²⁺-regulated ion channels in the apical membrane of primary MTE cells. Primary MTE cell cultures were voltage clamped and pretreated apically with 10 μM Ca²⁺-activated Cl⁻ channel (CaCC) inhibitor for 3 min followed by the apical addition of 10 μM 2at-LIGRLO(PEG₃-Pam)-NH₂ (Fig. 6, A and B). CaCC inhibitor significantly reduced 2at-LIGRLO(PEG₃-Pam)-NH₂-evoked changes in I_sc and TER (ΔI_sc: 3.76 ± 0.677 μA/cm² and ΔTER: −32.25 ± 2.09 Ω·cm², n = 4; Fig. 6G).

Fig. 6.

PAR₂ activates Ca²⁺-regulated ion channels in MTE cells. Ussing chamber experiments were conducted under voltage clamp-conditions. A and B: representative traces of I_sc (A) and TER (B) showing that the apical addition of 2at-LIGRLO(PEG₃-Pam)-NH₂ (10 μM) alone evoked a robust PAR₂ response that was largely attenuated by Ca²⁺-activated Cl⁻ channel inhibitor (CaCC-inh; 10 μM). C and D: representative traces of I_sc (C) and TER (D) showing that clotrimazole (30 μM) reduced the apical 2at-LIGRLO(PEG₃-Pam)-NH₂-evoked PAR₂ response compared with agonist alone. E and F: representative traces of I_sc (E) and TER (F) showing that combined apical treatment of CaCC-inh (10 μM), clotrimazole (30 μM), and 2at-LIGRLO(PEG₃-Pam)-NH₂ (10 μM) completely diminished the PAR₂ activation seen with 2at-LIGRLO(PEG₃-Pam)-NH₂ (10 μM) alone. G: quantification of changes in I_sc (left) and TER (right) showing that compared with 2at-LIGRLO(PEG₃-Pam)-NH₂ alone (ΔI_sc: 17.09 ± 1.88 μA/cm² and ΔTER: −270.3 ± 51.1 Ω·cm², n = 6), inhibition with CaCC-inh (ΔI_sc: 3.76 ± 0.677 μA/cm² and ΔTER: −32.25 ± 2.09 Ω·cm², n = 4) or clotrimzole (ΔI_sc: 3.21 ± 0.73 μA/cm² and ΔTER: −69.63 ± 8.149 Ω·cm², n = 3) significantly reduced the PAR₂ response. The two inhibitors combined completely abolished the apical PAR₂ response to 2at-LIGRLO(PEG₃-Pam)-NH₂ (ΔI_sc: 0.017 ± 0.017 μA/cm² and ΔTER: −0.533 ± 0.393 Ω·cm², n = 3). *Significant difference (P < 0.05 by one-way ANOVA).

We next investigated the potential involvement of large-conductance Ca²⁺-activated K⁺ channels in PAR₂-dependent alterations in ion conductance. Voltage-clamped MTE cell cultures were pretreated for 3 min with 30 μM clotrimazole followed by the apical addition of 10 μM 2at-LIGRLO(PEG₃-Pam)-NH₂ (Fig. 6, C and D). Pretreatment with clotrimazole significantly reduced 2at-LIGRLO(PEG₃-Pam)-NH₂-evoked changes in I_sc and TER (ΔI_sc: 3.21 ± 0.73 μA/cm² and ΔTER: −69.63 ± 8.149 Ω·cm², n = 3; Fig. 6G).

To evaluate if the two channel types above accounted for the observed PAR₂-dependent transcellular ion conductances, primary MTE cell cultures were voltage clamped and pretreated apically with a combination of 10 μM CaCC inhibitor and 30 μM clotrimazole for 3 min followed by the apical addition of 10 μM 2at-LIGRLO(PEG₃-Pam)-NH₂ (Fig. 6, E and F). In these experiments, the response to 2at-LIGRLO(PEG₃-Pam)-NH₂ was almost completely abolished (ΔI_sc: 0.017 ± 0.017 μA/cm² and ΔTER: −0.533 ± 0.393 Ω·cm², n = 3; Fig. 6G).

DISCUSSION

In the present study, we show, for the first time in primary airway epithelial cells, the robust apical activation of PAR₂ leading to a transient decrease in TER accompanied by an increase in Ca²⁺-regulated Cl⁻ and K⁺ conductances. These changes could only be observed with the recently developed, high-potency, lipidated PAR₂ ligand 2at-LIGRLO(PEG₃-Pam)-NH₂ (14). Apically positioned PAR₂ in the conducting airway epithelium is ideally located to detect endogenous proteases that can accompany allergen inhalation. We postulated that the transient changes in ion conductance could provide a temporary increase in airway surface liquid. This would be complemented by Ca²⁺-induced increases in ciliary beat (13). Thus, initial activation of PAR₂ in the lumenal airway could aid innate immune function via rapid, local upregulation of mucociliary clearance.

PAR₂ is expressed throughout the airway epithelium. Its activation after asthmagen exposure has been associated with both protective affects via bronchorelaxation (7, 8, 10, 19) and detrimental effects (e.g., cytokine release and subsequent chronic inflammation) in animal models of asthma (12, 38, 42). It has come to light in recent work that this dichotomy in PAR₂ activation effects may be due to dual signaling pathways downstream of PAR₂, namely, G_q-dependent Ca²⁺ signaling and G protein-independent β-arrestin signaling (29). Similar signaling has been reported for another GPCR targeted in asthma treatment, β-adrenergic receptor (βAR) signaling (43). Because of the diametrically opposed responses, continued development of novel, potent, efficacious, and (potentially) biased signaling drugs that affect PAR₂ or βAR are highly desirable. Recent progress in improving PAR₂ ligand potency and efficacy has occurred through the lipidation of the known PAR₂ agonists (4, 14). Such lipidated compounds, which include 2at-LIGRLO(PEG₃-Pam)-NH₂ used in this study, have orders of magnitude increased potency with improved selectivity for PAR₂ over traditionally used peptides (e.g., SLIGRL-NH₂) or peptidomimetics (e.g., 2at-LIGRL-NH₂ or 2-furoyl-LIGRLO-NH₂).

In the present study, we characterized the use of lipidated PAR₂ ligands in the activation of primary cultured MTE cells. Interestingly, apical application of the lipidated PAR₂ agonist to cultured MTE cells initiated a Ca²⁺ flux with a markedly higher EC₅₀ (835 nM) and wider 95% confidence interval (261 nM–2.68 μM) than those observed in an immortalized human bronchial airway cell line (16HBE14o- cells; EC₅₀: 3.95 nM, 95% confidence interval: 3.43–4.55 nM). The need for higher concentrations of lipidated PAR₂ ligands in primary cultured cells was not unexpected as it is not uncommon to observe differences in immortalized and primary cultured cells. However, this change in activity emphasizes the need for potent and efficacious ligands to evaluate roles for apically expressed PAR₂ in the airway. As a measure of specificity, PAR₂^−/− cells did not respond to the lipidated ligand with a transient change in [Ca²⁺]_i or ion flux.

Roles for PAR₂ in airway ion conductance in the mouse or human airway epithelium have traditionally been evaluated using basolateral application, in part due to the lack of response to activating peptides and/or peptidomimetics when applied apically (9, 25, 28, 31, 35, 37, 45). Our results examining basolaterally activated PAR₂ were similar to previous findings in that all of the PAR₂ ligands tested [SLIGRL-NH₂, 2at-LIGRL-NH₂, and 2at-LIGRLO(PEG₃-Pam)-NH₂ at 10 μM] evoked rapid, transient changes in TER and I_sc. It was somewhat surprising to us that a more robust response with the lipidated compound compared with traditional activating peptide and/or peptidomimetic was not observed after basolateral activation. We also noted a slightly different shape in the representative traces from those experiments with 2at-LIGRLO(PEG₃-Pam)-NH₂, which required more time to develop a full response in Ussing chamber experiments. It is likely these changes represent the lipid sequence interacting with the tissue culture filter and/or collagen during application and thus hindering presentation of the agonist to the receptor on the basolateral membrane.

Proteases in asthmagens have been shown to alter the barrier function of healthy and asthmatic airways (22, 27, 44). Furthermore, remodeled airway epithelium in asthmatics includes a reduction in barrier function (20). Thus, basolateral activation of PAR₂ could be relevant after extended asthmagen exposure or in diseased or compromised airways. However, basolateral activation of PAR₂ by inhaled asthmagens would not necessarily be expected in a healthy individual. A further caveat of interpreting basolateral PAR₂ activation is the demonstration in model epithelial cells of distinct pools of PAR₂ that display differential signaling when activated at apical or basolateral membranes (26).

As mentioned above, there exists limited data using traditional activating peptides (e.g., SLIGRL-NH₂) and peptidomimetics (e.g., 2at-LIGRL-NH₂) applied to the apical airway epithelial membrane. Although Cl⁻ current was observed after high concentrations (30 μM) of SLIGRL-NH₂ to the mouse (BALB/c) trachea, these changes were attributed to nonspecific neurokinin-1 receptor activation (1). Using the potent and selective lipidated compound 2at-LIGRLO(PEG₃-Pam)-NH₂, we were able to demonstrate, for the first time, apical PAR₂ activation in primary mouse tracheal epithelia that resulted in transient changes in TER and I_sc. To rule out the possibility that the lipidated agonist disrupted the epithelium and gained access to the basolateral side for activation, we repeated our recordings with 10 μM concentrations of SLIGRL-NH₂ and O(PEG₃-Pam)-NH₂. As with 10 μM SLIGRL-NH₂ alone, no epithelial response was observed.

Apical activation of PAR₂-dependent Ca²⁺-regulated Cl⁻ and K⁺ currents that could drive secretion in our well-differentiated MTE cells is consistent with the basolateral activation of the whole mouse trachea previously reported (25). Basolateral activation of primary human cells from absorptive airway also displayed Ca²⁺-regulated Cl⁻ and K⁺ currents that were supplemented with cAMP-dependent ion flow (9, 24, 25). These differences can be attributed to known differences in ion channel and transporter expression in the mouse and human airway (25). We conclude from these data that although PAR₂ is differentially sensitive in the apical and basolateral absorptive airway epithelium, activation results in similar, transient Ca²⁺ dependent Cl⁻ and K⁺ currents that temporarily provide a secretory phenotype (i.e., net Cl⁻ efflux). It would be interesting to use the high-potency lipidated PAR₂ agonist to determine if apical activation of secretory airway epithelial model cells [e.g., Calu-3 cells (31, 37, 45)] would result in similar signaling as we observed above or differential signaling, as observed in human colon epithelial models (26).

In summary, we have shown apical activation of PAR₂ on primary airway epithelial cells using the highly potent and specific lipidated peptidomimetic 2at-LIGRLO(PEG₃-Pam)-NH₂. To our knowledge, this is the first time activation of PAR₂ has been demonstrated at the apical surface of airway epithelial cells. Using highly potent and specific agonists to PAR₂ we can isolate receptor-specific contributions to airway epithelial function not previously possible. A plausible physiological implication for PAR₂ activation at the apical airway surface that results in Cl⁻ efflux is an increase in airway surface liquid. Increases in airway surface liquid would serve to aid in flushing out noxious stimuli such as inhaled asthmagens in a healthy individual.

GRANTS

This work was primarily supported by National Institutes of Health (NIH) Grant R01-NS-073664 (multi-principal investigator; to S. Boitano, T. J. Price, and J. Vagner), NIH Grant R01-AI-083403 (to M. O. Daines), and NIH Training Grant ES-007091 (to C. L. Sherwood). This work was additionally supported in part by NIH Superfund Research Grant ES-04940 (to S. Boitano).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: C.L.S., M.O.D., T.J.P., J.V., and S.B. conception and design of research; C.L.S. and S.B. performed experiments; C.L.S. and S.B. analyzed data; C.L.S., T.J.P., J.V., and S.B. interpreted results of experiments; C.L.S. and S.B. prepared figures; C.L.S. and S.B. drafted manuscript; C.L.S., M.O.D., T.J.P., J.V., and S.B. edited and revised manuscript; C.L.S., M.O.D., T.J.P., J.V., and S.B. approved final version of manuscript.