Taste Receptor Activation in Tracheal Brush Cells by Denatonium Modulates ENaC Channels via Ca2+, cAMP and ACh

Abstract

Mucociliary clearance is a primary defence mechanism of the airways consisting of two components, ciliary beating and transepithelial ion transport (I_SC). Specialised chemosensory cholinergic epithelial cells, named brush cells (BC), are involved in regulating various physiological and immunological processes. However, it remains unclear if BC influence I_SC. In murine tracheae, denatonium, a taste receptor agonist, reduced basal I_SC in a concentration-dependent manner (EC₅₀ 397 µM). The inhibition of bitter taste signalling components with gallein (G_βγ subunits), U73122 (phospholipase C), 2-APB (IP3-receptors) or with TPPO (Trpm5, transient receptor potential-melastatin 5 channel) reduced the denatonium effect. Supportively, the I_SC was also diminished in Trpm5^-/- mice. Mecamylamine (nicotinic acetylcholine receptor, nAChR, inhibitor) and amiloride (epithelial sodium channel, ENaC, antagonist) decreased the denatonium effect. Additionally, the inhibition of G_α subunits (pertussis toxin) reduced the denatonium effect, while an inhibition of phosphodiesterase (IBMX) increased and of adenylate cyclase (forskolin) reversed the denatonium effect. The cystic fibrosis transmembrane conductance regulator (CFTR) inhibitor CFTR_inh172 and the KCNQ1 potassium channel antagonist chromanol 293B both reduced the denatonium effect. Thus, denatonium reduces I_SC via the canonical bitter taste signalling cascade leading to the Trpm5-dependent nAChR-mediated inhibition of ENaC as well as G_α signalling leading to a reduction in cAMP-dependent I_SC. Therefore, BC activation contributes to the regulation of fluid homeostasis.

Keywords: ACh; CFTR; ENaC; Trpm5; Ussing chamber; airway epithelium; brush cell; ion transport; mucociliary clearance.

Figure 1

Effect of the bitter taste receptor agonist denatonium on transepithelial ion transport of mouse tracheal epithelium. (A) Apical application of denatonium (1 mM) resulted in a decrease in the net short circuit current (I_SC). Representative current trace. (B) The denatonium-induced effect was dose-dependent with an EC₅₀ of 397 µM in wildtype mouse tracheae (n = 5 for each concentration). (C) Repeated application of denatonium (1 mM, apical) on the same trachea showed similar denatonium-modulated current changes (ΔI_SC) upon the second application (n = 5, ns: not significant). (D) The current changes (ΔI_SC) upon basolateral application of denatonium (1 mM, n = 5) were smaller than upon apical application (* p < 0.05).

Figure 2

Denatonium acts on G protein-coupled taste receptors. (A) Schematic drawing of the action of the G_βγ inhibitor gallein. (B) The G_βγ subunit antagonist gallein (50 µM, n = 5, apical) reduced the denatonium-induced current (∆I_SC, n = 5, * p < 0.05).

Figure 3

Denatonium activates Tas2R108 and Tas2R105. (A) Application of denatonium (1 mM) increased [Ca²⁺]_i in HEK293 cells transfected with Tas2R108 (red) or Tas2R105 (green) but not in untransfected HEK293 cells. The ATP (10 µM) control stimulus increased [Ca²⁺]_i in all cell lines. Representative curves for the Ca²⁺ response. (B) The denatonium-induced [Ca²⁺]_i was increased in the HEK293 cells transfected with Tas2R108 (n = 99) or Tas2R105 (n = 160) compared to untransfected HEK293 cells (n = 227) and the [Ca²⁺]_i response to denatonium was higher in HEK293 Tas2R105 cells than in HEK293 Tas2R108 cells (* p < 0.05).

Figure 4

Involvement of components of the canonical bitter taste signalling cascade in the denatonium-induced effect. (A) Schematic drawing of the action of the PLC_β2 inhibitor U-73122. (B) In the presence of the PLC_β2 inhibitor U-73122 (10 µM, apical), the denatonium-induced current was reduced compared to control conditions (∆I_SC, n = 5, * p < 0.05). (C) Schematic drawing of the influence of the 1,4,5-trisphosphate (IP3) receptor inhibitor 2-APB, the gap junction inhibitor carbenoxolone and the SERCA inhibitor thapsigargin. (D) The IP3 receptor inhibitor 2-aminoethoxydiphenyl borate (2-APB, 100 µM, apical) reduced the denatonium-induced effect (ΔI_SC, n = 5, * p < 0.05). (E) Depletion of endoplasmic reticulum Ca²⁺ stores with 1µM thapsigargin (thap, apical) in the presence of extracellular Ca²⁺ did not impact the denatonium-induced effect (ΔI_SC, n = 5, * p < 0.05). Control: denatonium-effect without thapsigargin, thap + basal: denatonium-effect in the presence of thapsigargin, basal: denatonium-effect in the presence of thapsigargin without the thapsigargin-induced increase of basal I_SC.

Figure 5

Role of the Trpm5 channel in the denatonium-induced current changes. (A) Schematic drawing of the bitter taste signalling cascade activated by denatonium in murine tracheal brush cells. (B) In Trpm5^−/− mice, the response to denatonium was concentration-dependent with an EC₅₀ of 40 µM (n = 5 for each concentration, black curve). The corresponding curve from wildtype mice is depicted in green (data from Figure 1B). (C) The denatonium-induced current was reduced in the presence of the Trpm5 antagonist TPPO (triphenylphosphine oxide; 100 µM) when TPPO was applied apically or basolaterally in wildtype mice (∆I_SC, n = 5, * p < 0.05). (D) Comparison of the apical application of denatonium (1 mM) in the tracheae of wildtype (wt) and Trpm5^−/− mice. The denatonium-induced effect (∆I_SC) was reduced in Trpm5^−/− compared to wt mice (n = 17, * p < 0.05). (E) In the presence of the gap junction antagonist carbenoxolone (carb, 100 µM, apical and basolateral), the denatonium-induced effect was increased (ΔI_SC, n = 5, * p < 0.05).

Figure 6

Cholinergic signalling is involved in the denatonium-induced effect. (A) Schematic drawing of the influence of the muscarinic ACh receptor inhibitor atropine, the nicotinic ACh receptor inhibitor mecamylamine, the FLAP antagonist MK-886, and the TRPV1 agonist resiniferatoxin. (B) In the presence of the nAChR antagonist mecamylamine (MEC, 25 µM, apical and basolateral), the denatonium-induced effect was significantly reduced in wildtype but not in Trpm5^−/− mice (∆I_SC, n = 4–5, * p < 0.05, ns: not significant). (C) Application of MEC (25 µM, apical and basolateral) in the presence of denatonium (1 mM, apical) increased I_SC, but to a lesser extent than the wash out of denatonium. (D) The mAChR antagonist atropine (50 µM, apical and basolateral) did not influence the denatonium-induced current (∆I_SC, n = 5). (E) The FLAP inhibitor MK-886 (1 µM, apical and basolateral) did not alter the denatonium-induced effect (∆I_SC, n = 6). (F) Immunostaining of DRG sections from an untreated and an RTX-treated wildtype mouse stained for Trpv1 (arrows: Trpv1⁺ neurons). (G) Immunostaining of a tracheal whole mount preparation of an untreated trachea from a Trpm5-tauGFP mouse and a trachea from a Trpm5-tauGFP mouse treated with resiniferatoxin (RTX) stained for brush cells (GFP, green, arrows) and CGRP (yellow) and the pan-neuronal marker PGP9.5 (red) (arrowheads: nerves, stars: neuroendocrine cells). (H) The denatonium-induced effect was not changed in RTX-treated mice compared to the control mice (∆I_SC, n = 6–7).

Figure 7

Denatonium influences cAMP-dependent ion currents. (A) Schematic drawing of the effect of pertussis toxin (PTX). (B) The inhibitor of the G_αi subunit pertussis toxin (500 ng/mL, apical) reduced the denatonium-induced current (∆I_SC, n = 5, * p < 0.05). (C) In the presence of the phosphodiesterase inhibitor IBMX (100 µM apical), the denatonium-induced effect was increased (∆I_SC, n = 5, * p < 0.05). (D) Apical application of 1 mM denatonium decreased I_SC, which was reversed by forskolin (50 µM, apical). Representative current trace. (E) Forskolin increased I_SC in the presence of denatonium by approximately the same extent as it was decreased before with denatonium (∆I_SC, n = 6).

Figure 8

Involvement of ion channels in the denatonium-induced effect. (A) Schematic drawing of the influence of the epithelial sodium channel (ENaC) inhibitor amiloride, the Cl⁻-channel antagonist NPPB, the cystic fibrosis transmembrane conductance regulator (CFTR) blocker CFTR_inh172, and the K⁺-channel inhibitor BaCl₂ and the KCNQ1 channel antagonist chromanol 293B as well as the inhibitor of the Na-K-2Cl cotransporter 1 (NKCC1) bumetanide. (B) The ENaC inhibitor amiloride (10 µM, apical) reduced the denatonium-induced current in wildtype but not in Trpm5^−/− mice (ΔI_SC, n = 5–6, * p < 0.05, ns: not significant). (C) The effect of amiloride (10 µM, apical) was significantly reduced in the presence of denatonium (1 mM apical) (ΔI_SC, n = 4–5, * p < 0.05). (D) The Cl⁻-channel inhibitor NPPB (100 µM, apical) reduced the denatonium-induced effect in wildtype but not in Trpm5^−/− mice (ΔI_SC, n = 5–6, * p < 0.05, ns: not significant). (E) The CFTR antagonist CFTR_inh172 (10 µM, apical) reduced the denatonium-induced current (∆I_SC, n = 4, * p < 0.05, con: control). (F) The non-selective K⁺-channel inhibitor BaCl₂ (5 mM, apical) did not change the denatonium-induced current (∆I_SC, n = 5, ns: not significant) compared to the control effect. (G) In the presence of 5 mM BaCl₂ on the basolateral side of the epithelium, the denatonium-induced current was reduced (∆I_SC, n = 4, * p < 0.05, con: control). (H) The KCNQ1 antagonist chromanol 293B (100 µM, basolateral, chr 293B) reduced the denatonium-induced current (∆I_SC, n = 5, * p < 0.05, con: control). (I) The NKCC inhibitor bumetanide (200 µM, basolateral) reduced the denatonium-induced current (∆I_SC, n = 5, * p < 0.05, con: control).

Figure 9

Proposed signalling pathway of the influence of denatonium on transepithelial ion transport. Denatonium binding to Tas2R in brush cells leads to a G_βγ-dependent activation of PLC_β2, which results in an activation of IP3 receptors. The increase in [Ca²⁺]_i leads to an activation of Trpm5 followed by a release of ACh. The released ACh then binds to nAChR on neighbouring cells, resulting in an inhibition of ENaC and chloride channels. Simultaneously, the G_α-dependent reduction of [cAMP]_i in brush cells might lead to an additional decrease in ENaC, CFTR, and KCNQ1 as well as NKCC1 activity.