Effect of anion secretion inhibitors on mucin content of airway submucosal gland ducts

Sarah K. Inglis, Michel R. Corboz, Stephen T. Ballard


In porcine bronchi, inhibition of both Cl and HCO3 transport is required to block the anion secretion response to ACh and to cause mucus accumulation within ACh-treated submucosal gland ducts [S. K. Inglis, M. R. Corboz, A. E. Taylor, and S. T. Ballard.Am. J. Physiol. 272 (Lung Cell. Mol. Physiol. 16): L372–L377, 1997]. In this previous study, a combination of three potential HCO3 transport inhibitors [1 mM acetazolamide, 1 mM DIDS, and 0.1 mM dimethylamiloride (DMA)] was used to block carbonic anhydrase, Cl/ HCO3 exchange, and Na+/H+exchange, respectively. The aim of the present study was to obtain a better understanding of the mechanism of ACh-induced HCO3 secretion in airway glands by determining which of the three inhibitors, in combination with bumetanide, is required to block anion secretion and so cause ductal mucin accumulation. Gland duct mucin content was measured in distal bronchi isolated from domestic pigs. Addition of either bumetanide alone, bumetanide plus acetazolamide, or bumetanide plus DIDS had no significant effect on ACh-induced mean gland duct mucin content. In contrast, glands treated with bumetanide plus DMA as well as glands treated with all four anion transport blockers were almost completely occluded with mucin after the addition of ACh. These data suggest that mucin is cleared from the ducts of bronchial submucosal glands by liquid generated from Cl- and DMA-sensitive HCO3 transport.

  • acetylcholine
  • airway epithelium
  • bicarbonate secretion
  • bronchi
  • cystic fibrosis
  • mucus

the volume and composition of airway surface liquid is thought to be regulated by active ion transport across airway epithelia. Although Na+ absorption and Cl secretion are the two predominant transepithelial transport processes, a variety of airway epithelia also secretes HCO3 [nasal glands (6), trachea (18), bronchi (7), and Clara cells (20)]. It is likely that HCO3 is secreted into the airway lumen through apical Cl channels such as the cystic fibrosis (CF) transmembrane conductance regulator (CFTR) (18). In the genetic disorder CF, CFTR is mutated and channel function is disrupted (21). As a consequence, active secretion of Cl (2), HCO3 (18), and liquid (10) are probably impaired in CF airways.

A number of observations indicate that the submucosal glands may be principally involved in the secretion of airway surface liquid. First, submucosal glands secrete liquid at rest and in response to adrenergic and cholinergic stimulation (8, 11, 15). Second, basal rates of active Cl secretion are higher in glandular airways (bronchi) than in aglandular airways (bronchioles) (1). Third, pulmonary expression of the CFTR is highest in submucosal glands (5). Fourth, one of the first signs of pulmonary pathology in CF newborns is hypertrophy and mucus occlusion of the submucosal glands (17).

Inglis et al. (7) recently showed that ACh, a stimulator of gland mucus and liquid secretion, induces electrogenic anion secretion in excised porcine bronchi. After blockade of both Cl and HCO3 secretion, ACh-induced anion secretion is inhibited, and the submucosal gland ducts become occluded with mucin (9). These studies suggest that both Cl and HCO3 secretion drive glandular liquid secretion and that glandular liquid secretion is required to flush secreted mucins out of the gland ducts. Because CFTR-mediated anion secretion is defective in CF, it is likely that CF submucosal gland ducts become occluded with mucus as a result of defective glandular anion and liquid secretion.

Because we initially considered that carbonic anhydrase, Cl/ HCO3 exchange, and Na+/H+exchange could potentially be involved in airway epithelial HCO3 transport, the combination of acetazolamide, DIDS, and dimethylamiloride (DMA) was used in previous studies by Inglis et al. (7, 9) to block HCO3 secretion. The aim of the present study was to obtain a more precise understanding of the mechanism of HCO3 secretion from airway submucosal glands by determining which of the three HCO3 secretion inhibitors is required, in combination with bumetanide, to block ACh-induced anion secretion as assessed by the mucin content of gland ducts in excised bronchi.


Airway excision. Young domestic pigs (8–18 kg) were obtained from a local vendor. The animals were sedated with intramuscular injections of ketamine (8 mg/kg) and xylazine (0.4 mg/kg) and killed with an intravenous overdose of pentobarbital sodium. Portions of the left and right lungs were rapidly excised and placed in Krebs-Ringer bicarbonate (KRB) solution at room temperature (25°C). Distal bronchi (external diameter 3–4 mm) were carefully dissected from the surrounding parenchyma, transferred to a tissue bath containing KRB solution, and slowly warmed (0.1–0.2°C/min) from room temperature to 37°C.

Histological analysis of glandular mucin content. Isolated, warmed distal bronchi were cut into 1-cm lengths, and side branches were severed flush with the main airway trunk. The bronchi were then incubated for 30 min in KRB solution containing blockers of Cland HCO3 secretion. Bumetanide (10 μM), a blocker of Na+-K+-2Clcotransport, was used to inhibit Cl secretion. Blockers of HCO3 secretion were chosen based on the model for airway epithelial HCO3 secretion proposed by Smith and Welsh (18). According to this model, HCO3 and H+ are produced from CO2 and H2O by intracellular carbonic anhydrase. Protons are extruded from the cell interior by basolateral Na+/H+exchange, and HCO3 exits across the apical membrane through anion channels. The three HCO3 secretion blockers used in the present study were acetazolamide (1 mM) to block carbonic anhydrase, DMA (100 μM) to block Na+/H+exchange, and DIDS (1 mM) to block possible basolateral entry of HCO3 by Cl/ HCO3 exchange. Figure 1 shows a composite cell model of anion secretion and summarizes the intended targets for these anion secretion inhibitors. To identify which HCO3 blockers were needed to inhibit HCO3 secretion, bronchi were treated with either 1) bumetanide alone,2) bumetanide and DMA,3) bumetanide and acetazolamide,4) bumetanide and DIDS, or5) bumetanide and all three HCO3 secretion blockers. After pretreatment with the anion transport blockers, 10 μM ACh was added, and the tissues were incubated for 30 min. The bronchi were then removed and placed immediately into Millonig’s buffer containing 1% paraformaldehyde and 1% glutaraldehyde at 4°C. Later, the tissues were embedded in paraffin, cut into 5- to 6-μm-thick sections, and stained for mucins with Alcian blue and pyronin Y at pH 2.5 as previously described (9). This procedure stains sulfated glycoproteins red-purple and sialylated glycoproteins blue but does not stain neutral glycoproteins. We determined in parallel studies, however, that the distribution of neutral glycoproteins in porcine bronchi closely mirrors that of sulfated and sialylated glycoproteins (Inglis, unpublished observations).

Fig. 1.

Model of potential anion secretion pathways in airway epithelial cells. Model of anion secretion is modified from Smith and Welsh (18) to account for possible existence of basolateral Cl/ Formula exchange. Cl secretion: Cl enters cell by basolateral Na+-K+-2Clcotransport and exits through an apical anion channel. Formula secretion: Formula and H+ are formed inside cell by action of carbonic anhydrase (ca). H+ passes across basolateral membrane by Na+/H+exchange, and Formula exits cell through an apical anion channel. Alternatively, Formula could enter cell by basolateral Cl/ Formula exchange and exit through an apical anion channel. Dotted lines, intended site of inhibitor action.

A morphometric analysis was used to estimate the quantity of mucin within the submucosal gland ducts. All gland ducts within a single slide section were examined for each bronchus. The mucin content of each gland duct was scored in a blind fashion on a subjective linear scale from zero (no mucin in duct) to five (duct completely occluded with mucin). Each numerical increment between zero and five represents an ∼20% increase in the ductal lumen area occupied by mucin.

Solution composition and drugs. The KRB solution contained (in mM) 112 NaCl, 4.7 KCl, 2.5 CaCl2, 1.2 MgSO4, 1.2 KH2PO4, 25 NaHCO3, and 11.6 glucose. Solution pH was maintained at 7.4 by frequent gassing with 5% CO2 in O2. All drugs and chemicals were purchased from Sigma. DMSO was used as the vehicle for bumetanide, acetazolamide, DIDS, and DMA.

Statistical analysis. Within an experimental group, each airway was obtained from a different animal. All data are reported as means ± SE. ANOVA with Tukey’s post hoc test was used to compare mean values.P < 0.05 was considered the level of significance.


Histological analysis of glandular mucin content. To determine which anion transporters are required to keep submucosal gland ducts clear of mucin, bronchi were pretreated with various combinations of the anion transport blockers before the addition of ACh. The gland duct mucin contents for each of the five different treatments are shown in Figs.2 and 3. Ductal mucin content was scored on a linear scale from zero (duct contains no mucin) to five (duct occluded with mucin; seemethods). Figure 2 shows the frequency distribution of mucin scores, and Fig. 3 shows the average mucin content of the gland ducts. In bronchi treated only with ACh, >50% of the ducts contained no measurable mucin (Fig. 2) and mean ductal mucin content was 0.66 ± 0.07 (Fig. 3). In contrast, gland ducts from bronchi that were pretreated with all four anion transport blockers before ACh treatment contained significantly greater quantities of mucin (3.60 ± 0.09; Fig. 3). Forty percent of the ducts in these bronchi were completely occluded with mucin (Fig. 2). Mucin content of gland ducts pretreated with either bumetanide alone (1.33 ± 0.46), bumetanide and acetazolamide (1.95 ± 0.45), or bumetanide and DIDS (1.32 ± 0.45) was not significantly different from gland ducts treated only with ACh (Fig. 3). In contrast, the mucin content of gland ducts that were pretreated with bumetanide and DMA (3.61 ± 0.21) was comparable to the mucin content of gland ducts pretreated with all four anion transport blockers (Fig. 3) and was significantly greater than tissues that were pretreated with either no inhibitor, bumetanide alone, bumetanide and acetazolamide, or bumetanide and DIDS. Forty-one percent of the ducts that were pretreated with bumetanide and DMA were completely occluded with mucin (Fig. 2). These results confirm previous findings that glands pretreated with Cl and HCO3 secretion blockers become occluded with mucin when ACh is added (9). Photomicrographs of tissues pretreated with either bumetanide and DMA or the DMSO vehicle before the addition of ACh illustrate the extent of gland duct occlusion induced by these agents (Fig. 4).

Fig. 2.

Effect of anion secretion blockers on frequency distribution of ductal mucin content. Abscissa, amount of mucin present in submucosal gland of bronchi expressed on a scale from 0 (no mucin) to 5 (duct completely occluded with mucin). Ordinate, frequency of mucin scores. ACh (10 μM) was added to bronchi in 6 different pretreatment groups: drug vehicle (A); 10 μM bumetanide (B); bumetanide and 100 μM dimethylamiloride (DMA; C); bumetanide and 1 mM acetazolamide (D); bumetanide and 1 mM DIDS (E); bumetanide, acetazolamide, DMA, and DIDS (F). Values are means ± SE; n = 6 tissues.

Fig. 3.

Effect of anion secretion blockers on mean mucin content of gland ducts on addition of ACh. Scale is 0 (no mucin) to 5 (duct occluded with mucin). +, Presence of blocker. Values are replotted as means ± SE from data in Fig. 2; n = 6 tissues. * Significant difference from all other groups.

Fig. 4.

Effect of anion secretion inhibitors on submucosal gland duct mucin content. Isolated bronchi were pretreated with either DMSO vehicle (A) or 10 μM bumetanide and 100 μM DMA (B) before addition of 10 μM ACh. Tissue sections were stained with Alcian blue-pyronin Y at pH 2.5 to identify mucin glycoproteins (see text). Note that gland ducts from tissues pretreated with anion transport inhibitors are filled with mucin, whereas ducts in untreated tissues are clear. Bar, 50 μm.


A previous study (9) of distal bronchi provided indirect evidence that1) Cl and HCO3 secretion drives the glandular liquid secretion response to ACh and2) ACh-induced glandular liquid secretion can be uncoupled from mucus secretion by pretreatment with blockers of both Cl(bumetanide) and HCO3 (acetazolamide, DMA, and DIDS) transport. Unfortunately, this previous study provided little insight into the mechanism of HCO3 secretion. In the present study, measurements of gland duct mucin content were used to determine which of the three HCO3 secretion blockers is specifically required to inhibit ACh-induced, HCO3 - dependent liquid secretion. Pretreatment with either acetazolamide or DIDS, in combination with bumetanide, did not result in mucin accumulation after the addition of ACh. In contrast, DMA plus bumetanide was as effective as the cocktail of all three HCO3 blockers plus bumetanide at causing gland duct mucin accumulation after addition of ACh. These results suggest that DMA-sensitive Na+/H+exchange is critical for glandular HCO3 and liquid secretion.

The finding that glands pretreated with DMA become occluded with mucin after addition of ACh is consistent with a major role for Na+/H+exchange in airway gland HCO3 secretion. Similar models of HCO3 secretion that require Na+/H+exchange have been described in a variety of tissues including pancreatic ducts, ileum, and salivary acini (3, 13, 14). Muscarinic agonists are known to stimulate Na+/H+exchange in acinar cells of the pancreas and salivary, lacrimal, and nasal glands (4, 6, 16, 22). The mechanism for ACh-induced stimulation of the Na+/H+exchanger has not been fully determined, but elevation of intracellular calcium concentration is likely to play a role (22). An alternative explanation for the DMA-induced mucin occlusion of submucosal gland ducts could be that DMA stimulates gland mucin secretion. However, in a previous study, Inglis et al. (9) demonstrated that the combination of DMA, acetazolamide, and DIDS had no effect on gland duct mucin content in the absence of bumetanide. If DMA indeed stimulated mucin secretion, we would have expected an increase in ductal mucin content under those conditions. This previous study also demonstrated that the effect of the HCO3 blockers on gland mucin secretion is reversible, indicating that the blockers are unlikely to be toxic to the tissues.

The inability of DIDS to affect gland duct mucin content suggests that Cl/ HCO3 exchange activity is low or absent from airway glandular epithelium. Taken together, the effects of DMA and DIDS on ductal mucin accumulation suggest that the majority of secreted HCO3 is generated within the cell cytoplasm and that little or no HCO3 enters the cells across the basolateral membrane (Fig. 1). It is surprising, therefore, that acetazolamide did not induce glandular mucin accumulation. It is possible that carbonic anhydrase was incompletely blocked despite the presence of acetazolamide or that the uncatalyzed generation of intracellular HCO3 supported sufficient liquid secretion to flush mucin from the gland ducts. Ductal mucin content is unlikely to be a truly quantitative measure of gland liquid secretion. A threshold level of liquid secretion is probably required to flush mucin from the gland ducts. If gland liquid secretion is only partially inhibited, mucin may still be adequately cleared.

Evidence for HCO3 secretion has been found in nasal glands (6) as well as in tracheal (18), bronchial (7), and bronchiolar epithelia (20). Although its physiological role has not yet been determined, airway epithelial HCO3 secretion may play an important role in the regulation of airway surface liquid pH. Because both the physical properties of mucus (19) and the activity of cilia (12) are sensitive to changes in H+concentration, mucociliary clearance could be strongly influenced by the regulation of airway surface liquid pH.

In summary, this study suggests that ACh-induced HCO3 secretion in airway glandular epithelium is critically dependent on basolateral Na+/H+exchange. These findings reinforce the previous suggestion by Inglis et al. (9) that ACh-induced liquid secretion from airway submucosal glands is driven by both Cl and HCO3 secretion.


We thank Dr. Walter Wilborn and Barbara Hyde of the Structural Research Center (Mobile, AL) for performing the histology.


  • Address for reprint requests: S. T. Ballard, Dept. of Physiology, MSB 3024, Univ. of South Alabama, Mobile, AL 36688.

  • This work was supported by National Heart, Lung, and Blood Institute Grant HL-48622.


View Abstract