INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION APPENDIX REFERENCES

MUCINS ARE THE MAJOR MACROMOLECULAR components of mammalian mucus, the viscoelastic gel that coats and lubricates the epithelium of the respiratory tract and protects it against infectious and environmental agents (47). In the large-animal species, including humans, there are two main sources of airway mucins: the submucosal glands and the epithelial goblet cells. Although the secretions are likely increased to protect the airways, secretions in excess of airway clearance capacity are detrimental and contribute to the development of obstructive pulmonary diseases, such as asthma, chronic obstructive pulmonary disease, chronic bronchitis, and cystic fibrosis, by occluding the airways (45).

There is evidence that airway hypersecretion in some disorders is due in part to abnormalities in the nervous control of the airways (46). In the airways, cholinergic nerve fibers are closely associated with submucosal glands (33), which express muscarinic receptors (42, 65) and induce gland fluid and anion secretion (24). Cholinergic agonists such as acetylcholine and methacholine mimic the effects of parasympathetic nerve stimulation (8). It has been suggested that, in species with extensive submucosal glands, such as porcine and human (12, 27), the epithelial goblet cells receive no functional cholinergic innervation (43), whereas in species lacking extensive submucosal glands, such as mouse, hamster, and rabbit (27, 61), the epithelial goblet cells are under neural control (46). Also, it has been shown that the dense network of submucosal glands present in the proximal airways plays a critical role in airway maintenance (58). Therefore, to understand pulmonary diseases in which neurogenic mucous hypersecretion contributes to pathophysiology, it is important to study the regulation of gland fluid and ion secretion in an animal model with an extensive glandular network, such as the porcine airways. Moreover, the porcine tracheobronchial tree closely resembles that of humans, both morphologically and histologically (31).

In the present study, we used porcine tracheal epithelium with an optical hillocks technique (15) to measure the gland fluid secretion and an electrophysiological Ussing technique (55) to measure the transepithelial potential difference (PD) and short-circuit current (I_sc), which are bioelectric correlates of ion transport. A model for ion transport in the airway epithelium (60) predicts that activation of muscarinic receptors induces chloride (Cl) secretion from gland cells, leading to an increase in gland fluid secretion by osmosis. Because the Na⁺-K⁺-ATPase pump establishes the electrochemical gradients required for the Cl and associated gland fluid secretion, we measured the changes in cholinergically induced gland fluid and ion fluxes in the presence and in the absence of the sodium (Na⁺) channel inhibitor amiloride and the Cl channel inhibitors diphenylamine-2-carboxylic acid (DPC) or 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB). Our present results indicate that modulation of the cholinergically induced ion transport in the porcine glandular airway epithelium could provide a mechanism to control the neural component of gland fluid secretion.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION APPENDIX REFERENCES

Tracheas from 45 pigs (34-114 kg) were obtained from the local abattoir and were transported to the laboratory in ice-cold physiological saline solution. The tracheas were used within 2 h after removal from the pigs. Adventitious tissue was dissected from the external surface of the trachea, and the part of the trachea between the larynx and the lobar bronchus before the carina was cut into five or more tubes of ~2.5 cm in length. The tubes were then longitudinally cut through the anterior and posterior aspects, leaving a small piece of tissue as a tether to ensure a paired control tissue from the same location in the trachea, resulting in two tissues with exposed epithelium. Submucosal gland fluid flux (J_G) from randomly selected tissues with exposed epithelia was then measured by the hillocks technique (15, 37). The epithelium from one tube of each trachea was used to assess the viability of the trachea and to measure agent-induced changes in electrophysiological parameters via the Ussing technique (56).

Hillocks technique: submucosal J_G. The membrane preparation and subsequent J_G measurements were carried out as described in detail previously (39). Briefly, the pieces of each trachea were submerged for 1.5 h in a temperature-controlled (37°C), water-jacketed tissue bath (158400; Radnoti, Monrovia, CA) filled with 30 ml of Hanks' balance salt solution (HBSS) continuously bubbled with 95% O₂ and 5% CO₂. The tissues were then pretreated for 3 min or 1 h with either the Na⁺ channel inhibitor amiloride, or the Cl channel inhibitors DPC or NPPB, or the muscarinic antagonist 4-diphenylacetoxy-N-methylpiperidine methiodide (4-DAMP), or the vehicle DMSO. After pretreatment with one of these blockers, segments of trachea were removed from the bath, and the epithelial surface was blotted with a tissue wiper (Kimwipes; Kimberly-Clark, Roswell, GA) to remove any airway secretions present and then evenly coated with aerosolized tantalum (1- to 5-µm particle size tantalum; Atlantic Equipment Engineers, Bergenfield, NJ) from an aerosol generator (WDFII; BGI, Waltham, MA). The tissue was then placed in a bath containing 10 ml of warmed and oxygenated HBSS, with the cholinergic agonist methacholine (1,000 µM) and pretreatment agent bathing only the basolateral (cartilage) side and the tantalum-coated epithelium exposed to room air. A microscope equipped with a digital camera was used to capture ×25 images of 8 mm² of the epithelial surface area 3 min after the epithelium was coated with tantalum. Preliminary experiments showed that the initial J_G induced by methacholine in this preparation was completed in 3 min, consistent with previous observation in bovine (62) and ovine (29) airways. However, after this initial J_G, a low-sustained secretion rate was observed (data not shown), consistent with the observation of Joo et al. (29).

1

2

Ussing chamber: epithelial electrophysiology. The electrophysiological parameters of the epithelia mounted in Ussing chambers (55) were measured to determine the viability of the tissues and to monitor agent-induced changes in epithelial PD and I_sc (the current required to clamp the PD to 0 mV), indicators of transepithelial ion transport. One tube from each trachea was cut longitudinally through the anterior side to expose and dissect the posterior epithelium from the underlying cartilage. The epithelium was retained in a plastic slider that was mounted between the half-chambers of an Ussing chamber system (P2300; Physiologic Instruments, San Diego, CA). HBSS (5 ml) maintained at 37°C was circulated across the luminal and basolateral sides of the epithelium by gas (95% O₂ and 5% CO₂) lift oxygenators. Each half-chamber has two ports for the placement of salt bridges (pipette tip filled with 3% weight agar and 3 M KCl solution) containing Ag⁺-AgCl electrodes. A voltmeter and ammeter with current-voltage (I-V) clamp capabilities (VCC-MC2-HV; Physiologic Instruments) was connected to the I-V electrode pairs to measure the membrane's PD and I_sc, respectively. A computer with an analog-to-digital converter board (MF604; Humusoft) recorded the PD or I_sc at a 1-Hz interval from the analog outputs of the I-V clamp. The membrane's resistance (R_m) was calculated by using Ohm's law: R_m ( · cm²) · 10³ = PD (mV)/I_sc (µA/cm²). Before each experiment, the Ussing chamber was assembled in the absence of an epithelial membrane, filled with 10 ml of HBSS, and allowed to equilibrate at 37°C for 1.5 h. Any PD between the voltage-sensing electrodes was nulled, and the series resistance compensation circuitry was adjusted to compensate for the fluid resistance. Our studies were performed mostly under open-circuit conditions to simulate more closely the in vivo condition in which an electrical gradient (PD) is present across the epithelium. Each trachea was required to exhibit a baseline epithelial PD of >3 mV for inclusion of J_G or electrophysiological data in the present study.

Experimental protocol. Eight sets of experiments on porcine airway epithelia were conducted with the hillocks technique. 1) The baseline J_G (no pretreatment agent or methacholine challenge) was measured from at least one control tissue for each trachea. 2) Different tissues were challenged with different concentrations of methacholine at logarithmic intervals between 0.1 and 1,000 µM. 3) The effects of blocking cellular ion channels on the methacholine-induced J_G were also evaluated by pretreating the tissues for 3 min with either the epithelial Na⁺ channel blocker amiloride (10 µM) or 4) the Cl channel blocker DPC (1,000 µM) before challenge with methacholine (1,000 µM). 5) A longer period of exposure (1 h) with DPC and 6) another Cl channel blocker NPPB (300 µM) was also tested on methacholine (1,000 µM)-induced J_G. 7) To determine whether the methacholine-induced J_G was specifically a cholinergically induced secretion, the tissues were pretreated for 3 min with a muscarinic M₃ receptor antagonist, 4-DAMP (1 µM), before challenge with methacholine (100 µM). 8) Finally, to evaluate the role of the surface epithelium in methacholine-induced J_G, tissues were denuded of surface epithelium with a wooden dowel. The protocol was otherwise identical to that of the third set of experiments.

Data analysis. J_G, number of hillocks, PD, I_sc, and R_m were summarized as means ± SE, and n refers to the number of tissues tested. No more than three tissues from each trachea were used with the hillocks technique, and only one tissue per trachea was used with the Ussing technique. Paired, two-tailed Student's t-tests were performed to determine whether the changes in J_G, number of hillocks, or electrophysiological parameters were significantly different before and after different treatments (P < 0.05). Unpaired t-tests were used to determine whether the change in methacholine-induced J_G was significantly different between DPC- and NPPB-treated tissues and whether the change in methacholine-induced J_G in amiloride-treated tissues was significantly different between denuded and native tissues. The concentration-response curve in Fig. 2 was fit by the Levenberg-Marquardt (34) method for nonlinear least squares regression to the Hill equation J_G = (J * / { + , where [methacholine] is methacholine concentraton. The three-parameter curve fit estimated the maximum J_G , the effective concentration of methacholine that produces a gland half-maximal fluid flux (EC₅₀), and the Hill coefficient (n_H).

Solution composition and drugs. The HBSS contained 136.8 mM NaCl, 5.6 mM dextrose, 5.4 mM KCl, 4.2 mM NaHCO₃, 1.3 mM CaCl₂, 0.8 mM MgSO₄, 0.4 mM KH₂PO₄, 0.3 mM Na₂HPO₄, and phenol red. Amiloride and methacholine were purchased from ICN (Costa Mesa, CA). DPC, NPPB, 4-DAMP, and DMSO were purchased from Sigma Chemical (St. Louis, MO). All drugs were dissolved in DMSO, except methacholine, which was dissolved in water.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION APPENDIX REFERENCES

The average baseline submucosal J_G (n = 64 tissues from 45 porcine tracheas) was not significantly greater than zero, and no hillocks were apparent on 76% of the control tissues 3 min after the addition of tantalum powder (Fig. 1A). The areas of the smallest and largest hillocks measured in this study under the present magnification (×25) were 0.007 mm² under baseline conditions and 0.549 mm² after methacholine (1,000 µM) challenge in the presence of amiloride (10 µM). These hillocks had calculated volume secretions of 0.2 and 153.0 nl, respectively. The epithelia from 45 porcine tracheas had a baseline PD of 7.5 ± 0.5 mV (lumen negative) and generated an I_sc of 73 ± 4 µA/cm², resulting in a R_m of 105 ± 7  · cm². These values are in agreement with the PD of 9.7 mV and I_sc of 83 µA/cm² reported by Ballard et al. (3) in porcine tracheal epithelia.

View larger version (76K): [in this window] [in a new window]   Fig. 1.   Digitized images of porcine tracheal epithelium coated with tantalum powder. A: no hillocks are formed after 3-min baseline measurement period. Hillocks (indicated with arrows) are formed 3 min after addition of 10 µM methacholine [gland fluid flux (JG) = 0.22 µl · min1 · cm2; B] and 1,000 µM methacholine (JG = 0.47 µl · min1 · cm2; C). Scale is in micrometers.

The muscarinic agonist methacholine administered to the basolateral (cartilage) side caused concentration-dependent increases in J_G (Figs. 1 and 2). The number of hillocks per image also increased to 0.8 ± 0.5 (0.1 µM), 3.3 ± 0.5 (1 µM), 5.0 ± 0.6 (10 µM), 7.2 ± 0.6 (100 µM), and 7.2 ± 0.5 (1,000 µM). Methacholine (1,000 µM) induced a significant transient increase in baseline absolute PD (Fig. 3A) from 6.3 ± 1.2 to 7.2 ± 1.2 mV (n = 5) and I_sc (Fig. 3B) from 70 ± 14 to 82 ± 14 µA/cm² (n = 5), which caused a slight decrease in R_m of 6% from 104 ± 23 to 98 ± 19  · cm². The absolute PD and I_sc values reached their maximum in 80 ± 8 s and then decreased toward the baseline value (Fig. 3).

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Effect of methacholine on JG from porcine airway submucosal glands. The mean responses were fit to the Hill equation (solid line) with the maximum JG (0.41 µl · min1 · cm2), Hill coefficient (0.94), and EC50 (10.2 µM) determined by nonlinear least squares regression. Values are means ± SE. n = 6 Tissues from 2 tracheas at 0.1 µM and 4 tracheas at 1 µM; n = 13 tissues from 6 tracheas at 10 µM; n = 32 tissues from 11 tracheas at 100 µM; and n = 61 tissues from 24 tracheas at 1,000 µM.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Example of the transient increase in absolute magnitude of the epithelial potential difference (PD; A) and short-circuit current (Isc; B) after the addition of methacholine (1,000 µM, basolateral) in a porcine tracheal epithelium. Isc is measured in the intervals when the PD is clamped to 0.

Pretreatment with amiloride (10 µM) significantly increased the methacholine (1,000 µM)-induced J_G by 56% from 0.43 ± 0.04 to 0.67 ± 0.09 µl · min¹ · cm² (n = 21 tissues from 7 porcine tracheas, Fig. 4) and the number of hillocks from 7.1 ± 0.6 to 8.1 ± 0.7, although the latter was not significantly different. Amiloride alone had no significant effect on baseline J_G (n = 8 tissues from 4 porcine tracheas; data not shown). Amiloride (10 µM) significantly decreased the baseline absolute PD (Fig. 5A) from 8.2 ± 1.3 to 4.3 ± 0.7 mV (n = 5) and I_sc (Fig. 5B) from 72 ± 6 to 37 ± 4 µA/cm², causing a slight increase in R_m of 3% from 112 ± 13 to 115 ± 14  · cm². The absolute PD and I_sc stabilized at their lower values in 65 ± 9 s after the addition of amiloride (Fig. 5), consistent with previous studies in sheep and dog (1, 40). Subsequent addition of methacholine (1,000 µM) transiently and significantly increased the amiloride-inhibited absolute PD (Fig. 5A) to a peak value of 6.5 ± 0.8 mV in 57 ± 15 s (n = 5) and the amiloride-inhibited I_sc (Fig. 5B) to a peak value of 62 ± 4 µA/cm², causing a decrease in R_m of 10% to 104 ± 11  · cm².

View larger version (11K): [in this window] [in a new window]   Fig. 4.   Effect of the epithelial sodium channel blocker amiloride (10 µM, n = 21) on methacholine (1,000 µM)-induced submucosal JG. Values are means ± SE. * P < 0.05 compared with methacholine alone.

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Example of the rapid decrease in the absolute magnitude of the PD (A) and Isc (B) induced by amiloride (10 µM, luminal) and the transient increase in absolute magnitude of the epithelial PD (A) and Isc (B) after subsequent addition of methacholine (1,000 µM, basolateral) in a porcine tracheal epithelium. Isc is measured in the intervals when the PD is clamped to 0.

A 3-min pretreatment with DPC (1,000 µM) significantly decreased the methacholine (1,000 µM)-induced J_G by 50% from 0.40 ± 0.04 to 0.20 ± 0.06 µl · min¹ · cm² (n = 21 tissues from 7 porcine tracheas, Fig. 6), the number of hillocks from 7.4 ± 0.6 to 3.8 ± 0.5, the baseline absolute PD (Fig. 7A) from 8.9 ± 0.8 to 6.8 ± 1.1 mV (n = 5), and the I_sc (Fig. 7B) from 74 ± 10 to 51 ± 11 µA/cm², causing an increase in R_m of 12% from 128 ± 20 to 143 ± 21  · cm². The PD and I_sc stabilized at their lower values in 270 ± 95 s after the addition of DPC. Subsequent addition of methacholine (1,000 µM) significantly decreased the DPC-inhibited absolute PD (Fig. 7A) to 5.8 ± 1.2 mV (n = 5) and I_sc (Fig. 7B) to 48 ± 12 µA/cm², causing an increase in the R_m of 6% to 152 ± 34  · cm² after 80 s.

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Effects of the chloride channel blockers diphenylamine-2-carboxylic acid (DPC; 1,000 µM) and 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB; 300 µM) on methacholine (1,000 µM)-induced submucosal JG measured from tissues pretreated with DPC for 3 min (n = 21 tissues from 7 tracheas) and for 1 h (n = 10 tissues from 5 tracheas) and NPPB for 1 h (n = 10 tissues from 5 tracheas). Values are means ± SE. * P < 0.05 compared with flux induced by methacholine alone.

View larger version (20K): [in this window] [in a new window]   Fig. 7.   Example of the decrease in the absolute magnitude of the PD (A) and Isc (B) induced by DPC (1,000 µM, luminal) in a porcine tracheal epithelium. Note the absence of an increase in absolute magnitude of the epithelial PD (A) and Isc (B) after subsequent addition of methacholine (1,000 µM, basolateral). Isc is measured in the intervals when the PD is clamped to 0.

Because methacholine-induced J_G was not fully inhibited by the 3-min pretreatment with DPC, additional experiments on 10 tissues from five tracheas were performed with a longer 1-h DPC pretreatment period. Although a 1-h pretreatment with DPC (1,000 µM) significantly decreased the methacholine (1,000 µM)-induced J_G by 60% from 0.35 ± 0.07 to 0.14 ± 0.03 µl · min¹ · cm² (Fig. 6) and decreased the number of hillocks from 4.8 ± 0.3 to 3.8 ± 0.5 (P = 0.07), no significant difference in the methacholine-induced J_G or the number of hillocks was observed between tissues pretreated with DPC for 3 min or 1 h. A 1-h pretreatment with another Cl channel blocker NPPB (300 µM) significantly decreased the methacholine (1,000 µM)-induced J_G by 92% from 0.39 ± 0.08 to 0.03 ± 0.01 µl · min¹ · cm² (n = 10 tissues from 5 porcine tracheas, Fig. 6) and the number of hillocks from 7.3 ± 0.8 to 2.3 ± 0.5.

Also, the J_G and number of hillocks produced by 100 µM methacholine (0.27 ± 0.01 µl · min¹ · cm² and 6.9 ± 0.3, respectively, n = 9 tissues from 3 porcine tracheas) were completely inhibited by a 3-min pretreatment with the M₃-selective muscarinic antagonist 4-DAMP (1 µM). The J_G and number of hillocks for methacholine (1,000 µM)-induced gland fluid secretion from paired tissues pretreated with the antagonist vehicle DMSO (0.33 ± 0.07 µl · min¹ · cm² and 8.9 ± 0.9, respectively, n = 8 tissues from 3 porcine tracheas) were not significantly different from the J_G or number of hillocks of tissues not pretreated with DMSO (0.33 ± 0.06 µl · min¹ · cm² and 8.1 ± 1.3, respectively), indicating that DMSO at the concentration used in this study had no significant effect on gland fluid secretion. Finally, no difference was observed between tissues denuded of epithelium and those with intact epithelium in methacholine (1,000 µM)-induced J_G (0.30 ± 0.05 and 0.38 ± 0.05 µl · min¹ · cm², respectively; n = 8 tissues from 4 porcine tracheas). Also, the absence or the presence of epithelium did not affect the increase of methacholine (1,000 µM)-induced J_G in the presence of amiloride (0.51 ± 0.14 µl · min¹ · cm², n = 6 tissues from 3 porcine tracheas; and 0.62 ± 0.09 µl · min¹ · cm², n = 23 tissues from 8 porcine tracheas, respectively). Mechanical removal of the epithelium by rubbing the luminal surface of the porcine tracheal ring with a wooden dowel led to the complete obliteration of the epithelium as confirmed by histology (data not shown).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION APPENDIX REFERENCES

In the present study, we used the hillocks technique to measure gland fluid secretion and the Ussing technique to measure ion transport across isolated porcine tracheal epithelia. We showed that methacholine produces a concentration-dependent increase in J_G and an increase in the PD and I_sc. We also found that ion transport inhibitors, such as the Na⁺ channel blocker amiloride and the Cl channel blockers DPC or NPPB, can modulate the methacholine-induced J_G, PD, and I_sc across porcine tracheal epithelia. In agreement, the time course of the events measured by the hillocks and Ussing techniques suggests that the methacholine-induced increase in J_G and the hyperpolarization observed from unpretreated or amiloride-pretreated tissues occur concurrently, with both events decreasing toward premethacholine levels within 3 min.

With the development of a computer-assisted data analysis program calculating J_G from the microscopic hillock images (39), we can compare our data with the data obtained from the hillocks (16, 23, 32) and other techniques, such as the excised whole airway (5) and micropipette (13) techniques that measure gland flux or flow rate. The cholinergic agonist-induced J_G (Fig. 2) confirms the cholinergic agonist-induced secretions observed by other investigators from porcine tracheal gland cells (17, 64), glands (22, 52), and native epithelial membranes (5, 54, 65). The measured J_G induced by methacholine (10 µM) of 0.21 µl · min¹ · cm² in the present study is in agreement with the value reported for liquid secretion induced by another secretagogue, acetylcholine (10 µM), namely, 0.22 µl · min¹ · cm², from porcine bronchi (54). However, the methacholine (1,000 µM)-induced J_G of 0.43 µl · min¹ · cm² is substantially less than the J_G induced by mechanical stimulation from canine trachea (7.4 µl · min¹ · cm²) in vivo (23) or by electric field stimulation from ferret trachea (2.5 µl · min¹ · cm²) in vitro (32). This is expected, as these stimuli release endogenous acetylcholine and other gland secretagogues (44). The methacholine-induced J_G was completely inhibited by the M₃-selective muscarinic antagonist 4-DAMP, indicating that M₃ receptors were responsible for methacholine-induced gland fluid secretion, consistent with other studies in the pig (65) and ferret (42).

The Ussing chamber measures the electrophysiological parameters maintained by the active ion transport across the surface airway epithelium that is composed mainly of ciliated and goblet cells in pig (7) and human (10). Studies by Ballard et al. (2, 4) suggested that epithelial resistance (R_m) may be lower in airways with submucosal glands, and, recently, Wang et al. (58) confirmed that the glands affect the overall ion transport properties of the epithelium. Therefore, the Ussing chamber measures the concerted electrogenic ion transport of the surface epithelial cells as well as the submucosal glands. We measured an increase in the absolute PD (indicating epithelial hyperpolarization) and I_sc (indicating an increase in transepithelial active ion transport) after the addition of methacholine. These effects are likely created in part by methacholine-stimulated Cl secretion into the gland serous cell tubule lumen via an inositol triphosphate-dependent mechanism (48), as demonstrated in porcine (38, 65) and human gland cells (63).

The gland secretion composition is modified as the fluid progresses through the intricate structure (36) of the submucosal glands toward the airway surface. The gland serous tubules and acini secrete ions, water, glycoprotein, and antimicrobial proteins (19, 26, 36). The serous secretions collect additional glycoproteins as they pass through mucous tubules into a collecting duct. A gland collecting duct finally terminates into a ciliated gland duct opening at the airway epithelial surface with a density of approximately one duct opening per millimeter square of tracheal epithelium in pig (39) and human (41). The collecting duct cells express a relatively large number of Na⁺ channels (11, 20) and are mitochondria rich (31), suggesting a high-metabolic activity required for active Na⁺ and osmotically associated fluid absorption.

A simplified model of ion transport across airway epithelium (60) can be used to interpret the data collected by the hillocks and Ussing techniques. This model can be applied to the ciliated ducts and collecting ducts of the glands because Na⁺ transporters and Cl channels are present (11, 18, 20). We found that the methacholine-induced increase in epithelial absolute PD, I_sc, and J_G was greater when the tissues were pretreated with the Na⁺ channel blocker amiloride (Figs. 4 and 5). A previous study on canine tracheal explants also reported a slight increase in methacholine-induced glycoconjugate secretion when pretreated with amiloride (6). Pretreatment with amiloride depolarizes the surface and likely parts of the glandular epithelium by inhibition of Na⁺ absorption, creating a greater electrochemical gradient for the methacholine-induced Cl efflux. This suggests that part of the increased J_G from amiloride-pretreated (depolarized) surface and glandular tissues is osmotically associated with the hyperpolarization caused by the methacholine-induced Cl secretion and electrically silent NaCl secretion (9). Another part of the increased J_G observed from amiloride-pretreated tissues could be due to a decrease in fluid absorption in the gland tubule collecting ducts, which would be detected as an increase in J_G by the hillocks technique. Considering our data and the observation of Wu et al. (62) that amiloride inhibits Na⁺-associated fluid absorption in bovine airways, it is likely that a portion of this Na⁺ and associated fluid absorption blocked by amiloride takes place in the gland collecting ducts (28).

In our preparation, the epithelium had no effect on methacholine (1,000 µM)-induced J_G. We also found no significant difference in methacholine (1,000 µM)-induced J_G between epithelium-denuded and -intact tracheal tissues pretreated with amiloride (10 µM), indicating that amiloride affects the glands. Similarly, it has been shown that methacholine- and substance P-induced secretions from human (51) and porcine (53) epithelium-denuded bronchial tissues, respectively, did not significantly differ from those of control tissues. A mathematical model suggests that the effects of amiloride on the tracheal surface epithelia would only increase the volume of a hillock by 4% (see APPENDIX), whereas in our study we found an increase of 56%, implicating the importance of Na⁺ and associated fluid transport across some of the gland ducts.

We also found that the methacholine-induced J_G was reduced by pretreatment with DPC or NPPB (Fig. 6), with no increase in epithelial absolute PD or I_sc after the addition of methacholine to DPC-pretreated tissues (Fig. 7). Interestingly, a decrease in absolute PD (Fig. 7) was observed after the addition of methacholine (1,000 µM, basolateral) to tissues treated with DPC (1,000 µM, luminal). This could be due to methacholine-induced activation of basolaterally located Cl channels (57). In the framework of the simplified ion transport model, DPC or NPPB block the apical Cl channels that are active under homeostatic conditions and depolarize the surface and glandular epithelium by inhibition of basal transcellular Cl secretion. Unlike the amiloride-induced depolarization, DPC blocks the Cl channels that are opened by methacholine. Therefore, no increase in absolute PD or I_sc is observed after the addition of methacholine to the DPC-pretreated tissues (Fig. 7), unlike the unpretreated (Fig. 3) and amiloride-pretreated (Fig. 5) tissues. This suggests that part of the decrease in methacholine-induced J_G from DPC-pretreated tissues is osmotically associated with the inhibition of methacholine-induced Cl secretion, observed as a loss of methacholine-induced epithelial hyperpolarization (compare Figs. 3 and 7), from DPC-pretreated tissues.

Similar to the 92% inhibition of methacholine (1,000 µM)-induced tracheal J_G by NPPB (300 µM) in our study, Ballard et al. (5) reported a 91% inhibition in acetylcholine (10 µM)-induced porcine bronchial mucus by NPPB (300 µM). However, a difference was observed between the two studies when DPC was used (60% inhibition in the present study vs. 86%). This may be explained by different gland secretion measurement times (3 min in the present study vs. 2 h) or differences in the measurement techniques. The greater inhibition of methacholine-induced J_G in tissues treated with NPPB vs. DPC in our study is likely due to the greater potency of NPPB (59, 66). The methacholine-induced J_G not blocked by DPC or NPPB could be due to either contraction of the glandular myoepithelial cells (49), or increased mucin secretion and obligatory osmotic water transport, or a change in gland tubule permeability due to upregulation of aquaporin AQP5 (25, 50). Also, the decrease in the number of methacholine-induced hillocks on DPC- or NPPB-pretreated tissues is consistent with the finding that inhibition of anion transport causes occlusion of the glands by mucus (24).

In conclusion, we showed for the first time using the hillocks technique that the Na⁺ transport inhibitor amiloride increases and the Cl transport inhibitors DPC or NPPB decrease methacholine-induced porcine gland fluid secretion. The data suggest that ion transport in the glands is an important mechanism that modulates airway gland secretion. It is interesting to note that humans with genetically defective Na⁺ channels (pseudohypoaldosteronism) produce excessive airway fluid (30), and humans with defective cystic fibrosis transmembrane conductance regulator Cl channels (cystic fibrosis) may produce dehydrated airway secretions based on cultured cell study (35). The abnormal composition of airway fluid in other obstructive pulmonary diseases, such as asthma, chronic bronchitis, and chronic obstructive pulmonary disease, may be due in part to a neurogenic mechanism leading to abnormal gland fluid flow into the airway lumen (45). Also, amiloride-sensitive Na⁺ channels may be responsible for accumulation of fluid in the respiratory tract in influenza infection (21). Therefore, studies of gland fluid secretion and ion transport may provide a means for investigating agents to maintain airway patency in obstructive pulmonary diseases.