Journal of Applied Physiology
Vol. 83, No. 4, pp. 1348-1359, October 1997
GAS EXCHANGE, MECHANICS, AND AIRWAYS

Interaction between ion transporters and the mucociliary transport system in dog and baboon

Scot L. Winters and Donovan B. Yeates

Pulmonary Biophysics and Bioengineering Research Laboratory, Departments of Medicine and Chemical Engineering, University of Illinois at Chicago, Chicago 60680; and Veterans Affairs Chicago Health Care System, Chicago, Illinois 60612

ABSTRACT

INTRODUCTION

METHODS

RESULTS

DISCUSSION

ACKNOWLEDGEMENTS

FOOTNOTES

REFERENCES

ABSTRACT

Winters, Scot L., and Donovan B. Yeates. Interaction between ion transporters and the mucociliary transport system in dog and baboon. J. Appl. Physiol. 83(4): 1348-1359, 1997.To gain insight into the role of epithelial ion channels, pumps, and cotransporters in regulating airway water and mucociliary transport, we administered inhibitors of the Na⁺ channel (amiloride), 3Na-2K-adenosinetriphosphatase (acetylstrophanthidin), and Na-K-2Cl cotransporter (furosemide) to anesthetized dogs and/or baboons. Tracheal ciliary beat frequency was measured by using heterodyne laser light scattering. Tracheal mucus velocity (TMV) and bronchial mucociliary clearance (BMC) or lung mucociliary clearance were measured by using radioaerosols and nuclear imaging. Respiratory tract fluid output was collected by using a secretion-collecting endotracheal tube. In six dogs, amiloride aerosol [lung deposition, 96 ± 11 µg (means ± SE)] had minimal effect, whereas acetylstrophanthidin aerosol (lung deposition, 71 ± 9 µg) increased BMC, and furosemide (40 mg iv) markedly increased TMV. In five baboons, TMV increased after iv furosemide administration (2 mg/kg) as well as by aerosol (lung deposition, 20 ± 3 mg), coincident with increases in ciliary-mucus coupling from 11.5 ± 0.1 to 29.5 ± 0.4 and 46.5 ± 0.7 µm/beat, respectively. Furosemide also increased lung mucociliary clearance in baboons. In dogs, respiratory tract fluid output increased after intravenous furosemide from 2.2 ± 0.5 to 6.8 ± 1.7 mg/min. When combined with dry-air inhalation, furosemide failed to stimulate TMV and reversed the inhibition of BMC by dry air. Thus pharmacological manipulation of the Na-K-2Cl cotransporter and the 3Na-2K-adenosinetriphosphatase pump may provide increases of clinical relevance in airway hydration and mucociliary transport.

fluid transport; ciliary beat; ion transport; water balance; mucus transport

INTRODUCTION

THE DIRECTION OF NET WATER FLUX across many epithelia, including that of the tracheobronchial airways, has been suggested to result from transepithelial ion transport with secondary osmotic water movement. The effective transport of airway mucus atop cilia extending 5-7 µm from the subjacent epithelia is considered critically dependent on the depth of the periciliary fluid. Therefore, acute changes in hydration of the airway lining fluid must induce compensatory mechanisms within the epithelium, possibly through altered concentrations of Na⁺ and Cl, the predominant osmolytes in unperturbed airway lining fluid. The Na-K-2Cl cotransporter and the 3Na-2K-adenosinetriphosphatase (ATPase) pump, in conjunction with Na⁺ and Cl channels, are central to the maintenance of Na⁺ and Cl homeostasis in the epithelium. Although extensive studies on these transporters have been aimed at understanding their function at the molecular level as well as in epithelial tissues, the relationship of these transporters in the homeostatic regulation of the airway lining fluid, i.e., the periciliary fluid and thus mucociliary transport in vivo, is still largely a matter of speculation. In vitro studies have suggested that water absorption can be suppressed by inhibition of Na⁺ channels (14) or 3Na-2K-ATPase pumps (22) while secretion by tracheal epithelium, presumably associated with an induction of Cl flux (29), is inducible by various agonists as well as by evaporation (26).

We hypothesized that the unperturbed tracheobronchial airways favor absorption of water into the mucosa, leading to a low basal level of mucociliary transport, which would be further reduced during acute inspiration of dry air. On the basis of in vitro studies, inhibition of Na⁺ channels or 3Na-2K-ATPase pumps can be predicted to inhibit flux of ions and water from the lumen into the mucosa, increasing airway hydration and leading to an increase in mucociliary transport. In the unperturbed airways, Cl secretion is low (35). 1) If this low Cl flux is associated with water flux from the mucosa into the lumen, then inhibition of the Na-K-2Cl cotransporter may slightly reduce the already low basal level of mucociliary transport. 2) However, if administration of the Na-K-2Cl cotransport inhibitor furosemide induces a decrease in intracellular chloride and cellular volume associated with an increase in epithelial permeability to water, an increase in airway fluid depth and mucociliary transport will result. If, as in 1, inhibition of the cotransporter inhibits water flux into the airways, administration of furosemide before inspiration of dry air will further reduce the dry-air-induced decrease in mucociliary transport. On the other hand, if, as in 2, furosemide causes an increase in water flux from the mucosa into the lumen, the decrease in mucus transport due to dry air will be negated. We addressed the above hypotheses and questions regarding the joint regulation of airway ion/water flux and mucociliary transport by investigating responses of the canine (and in specific cases the baboon) mucociliary transport system in vivo to amiloride (an inhibitor of mucosal Na⁺ channels), acetylstrophanthidin (an inhibitor of the basolateral 3Na-2K-ATPase pump), furosemide (an inhibitor of the basolateral Na-K-2Cl cotransporter), inhalation of dry air, and the combination of furosemide and inhalation of dry air.

The bronchi of both dogs and humans absorb Na⁺ (2, 15), and the tracheae of dogs and humans can both secrete Cl and absorb Na⁺ in vitro (35, 36). However, the balance between these differs under short-circuit conditions, with the canine trachea being predominantly chloride secreting and the primate trachea being predominantly sodium absorbing. Thus canine and primate airways may be considered to have different basal "set points," suggesting that basal levels of ion and water fluxes differ. We hypothesized that furosemide would produce similar effects on the mucociliary transport system in both canine and primate airways, assuming both airways possess similar mechanisms to respond to perturbation. Furthermore, although intravenous (iv) administration of furosemide may inhibit the basolateral Na-K-2Cl cotransporter in the airway epithelium, mucociliary transport responses might be attributed to systemic effects of this agent. On the basis of the expectation that mucociliary responses to furosemide would be due to its action on the airway epithelia, we postulated that aerosolized and iv administration of furosemide would produce similar effects on mucociliary transport. To investigate the above hypotheses, we also studied the baboon tracheobronchial mucociliary transport system and its response to iv and aerosolized furosemide.

Parameters monitored individually or in combination included tracheal ciliary beat frequency (CBF), tracheal mucus velocity (TMV), mucociliary clearance of the bronchi (BMC) or entire lung (LMC), and volumetric respiratory tract fluid output at the larynx (RTFO) in complementary experiments.

METHODS

The experimental procedures and protocols for these studies on dogs and baboons were approved by the Animal Care Committee of the Biological Resources Laboratory at the University of Illinois at Chicago. The Biological Resources Laboratory is sanctioned by the American Association of Accreditation for Laboratory Animal Care.

1

1

RESULTS

The responses of the canine mucociliary transport system, as indicated by CBF, TMV, and BMC, to lung deposition of 96 ± 11 µg of amiloride or 71 ± 9 µg of acetylstrophanthidin and to iv administration of 40 mg of furosemide are shown in Fig. 1. For the study in which amiloride was administered, CBF in the 15 min before the perturbation was started, was significantly higher than in the control study. The CBF from the first 15 min after the sham control perturbation was started, was significantly elevated with respect to subsequent time periods in the control study (Fig. 1A). This is possibly caused by mechanical disturbance of the system inherent in moving equipment and the coupling of ventilatory circuits. Considering CBF at all times throughout the study, CBF was increased compared with the control by each of the perturbations. However, the increases that reached statistical significance (many CBF observations) are limited in magnitude and perhaps importance, i.e., particularly those after administration of amiloride. After furosemide injection, TMV increased from the control result of 10.9 ± 0.7 to 14.4 ± 0.5 mm/min (Fig. 1B). The R24% for the control study was 36.9 ± 3.0%, compared with 33.9 ± 11.7, 28.5 ± 9.5, and 42.5 ± 6.6% for the study in which amiloride, acetylstrophanthidin, or furosemide was administered, respectively. This indicator of radioaerosol deposition pattern did not vary significantly either between test perturbations or between individual dogs. However, the study in which acetylstrophanthidin was administered had the lowest R24% value. BMC was markedly increased compared with the control study by administration of aerosolized acetylstrophanthidin (17.2 ± 1.3 vs. 9.0 ± 1.3%, BMC_15,30 vs. control, respectively; Fig. 1C). The increase in BMC caused by administration of furosemide iv (14.2 ± 1.5 vs. 9.0 ± 1.3%, BMC_15,30 vs. control, respectively) did not reach statistical significance.

The response in baboons of CBF, TMV, and LMC to administration of 2 mg/kg iv furosemide or lung deposition of 20 ± 3 mg of furosemide is shown in Fig. 2. In the study in which aerosolized furosemide was inhaled, CBF was significantly lower than the comparable CBF in the control study during the 15 min before the inhalation was started (Fig. 2A). CBF during the 60 min after inhalation of the aerosolized furosemide was remarkably still lower (7.9 ± 0.3 vs. 10.1 ± 0.3 Hz, CBF_0,60 vs. control, respectively). TMV markedly increased after furosemide iv (TMV_0,90, 6.6 ± 0.4 mm/min; Fig. 2B) compared with the control study (TMV_0,90, 2.1 ± 0.1 mm/min). An even more dramatic increase was observed in the 45 min after furosemide aerosol (11.2 ± 1.4 vs. 2.2 ± 0.1 mm/min, TMV_0,45). The increase in LMC caused by administration of furosemide iv (8.9 ± 1.3 vs. 4.9 ± 0.4%, LMC_15,30 vs. control, respectively; Fig. 2C) did not reach statistical significance. Although mucociliary clearance was slightly faster than in the comparable control study before the inhalation of aerosolized furosemide, the increase in LMC after inhalation of aerosolized furosemide was statistically significant within 15 min of the aerosol delivery (LMC_0,15, 2.4 ± 0.4 to 7.1 ± 0.6%) and continued to increase throughout the study (LMC_45,60, 8.4 ± 0.4 to 23.7 ± 0.9%).

The responses of CBF, TMV, and BMC in dogs to inhalation of dry air or dry air plus furosemide, compared with the control study and the study where only furosemide was administered, are shown in Fig. 3. Dry-air inspiration decreased CBF compared with the control study during the first 15 min of exposure, but, surprisingly, CBF recovered during subsequent dry-air exposure to approach or exceed control values (Fig. 3A). When TMV from the dry air inhalation was considered as one sample throughout the 90-min delivery period, no difference from the control was evident (Fig. 3B). However, when the dry-air exposure was examined in three periods of 30 min each, a reduction was suggested during the first 30 min (TMV_0,30, 8.6 ± 1.0 vs. 10.9 ± 0.7 mm/min), followed by a return to near baseline during the second 30 min (TMV_30,60, 11.3 ± 1.1 mm/min), and a suggested TMV increase between 60 and 90 min of exposure (TMV_60,90, 15.1 ± 1.7 mm/min). TMV (Fig. 3B) also showed responses that suggested counterbalancing effects of dry air and furosemide. Dry air eliminated the substantial TMV increase after furosemide administration, and furosemide administration eliminated the suggested TMV increase during prolonged dry-air exposure. In the radioactivity retention studies, the R24% values for the studies in which dry air and dry air plus furosemide were administered were 44.9 ± 3.9 and 47.5 ± 4.0%, respectively, and were not significantly different from the control or from the study in which furosemide only was administered. Mucociliary clearance in the 15 min before the start of both dry-air and dry-air plus furosemide perturbations was slightly slower than the control study (Fig. 3C). Beyond 30 min of dry-air exposure, BMC was significantly reduced compared with the control (9.3 ± 0.9 vs. 14.0 ± 1.4%, BMC_30,45 vs. control, respectively). BMC for the combined perturbation of dry air and iv furosemide (Fig. 3C) was not significantly different from that of the control study and was approximately midway between BMC curves for dry air and furosemide alone.

RTFO for the control study and for the study in which 40 mg of furosemide were administered iv is compared for each of the sampling periods in Fig. 4. The control RTFO may be artificially large in the first collection period because fluid accumulates at the endotracheal tube cuff from the time of intubation, not from the start of collection. Furthermore, secretion may be transiently stimulated by intubation. On the basis of the latter two sampling periods, RTFO collection rate was significantly larger than the control study rate after administration of furosemide iv (6.8 ± 1.7 vs. 2.2 ± 0.5 mg/min, furosemide vs. control, respectively). On the basis of postmortem measurement of one dog and radiographic image scaling of another dog, the mean diameter of the trachea for the dogs used in these studies was estimated to be 19 mm. By eliminating the time dimension from RTFO and TMV, and assuming a uniform circular cylinder of transport, the estimated airway lining fluid transport depth in the control study was 4.5 ± 0.9 µm. This value was increased to 8.3 ± 2.3 µm by the administration of iv furosemide.

In the dog, coordination of regional tracheobronchial transport, i.e., between BMC and TMV, was maintained similarly to the control study in the studies in which aerosolized acetylstrophanthidin or both dry air and iv furosemide were used (Fig. 5). Bronchial clearance decreased relative to tracheal transport in the studies after perturbation with aerosolized amiloride, iv furosemide, or dry air. In the baboon, lung clearance of radioactivity also decreased relative to tracheal transport compared with the control study during study of both iv furosemide and aerosolized furosemide.

The interaction between tracheal CBF and TMV, as well as its variation in response to perturbation of airway lining fluid and ion transport, is shown in Fig. 6. It is notable that none of the perturbations in the present study significantly decreased the index of micrometers per beat. These values are presented for the dogs (Fig. 6A) and the baboons (Fig. 6B), considering all results from the start of the perturbation to the end of the study ~60 min later. The conversion of ciliary beat into mucus transport for the control study was notably several times less in the baboon (13.6 ± 0.3 µm/beat) relative to the dog (57.8 ± 0.7 µm/beat). In the dog, the beat-transport translation was not changed by either aerosolized acetylstrophanthidin or dry-air exposure. A minimal increase above the control study was observed for the study in which amiloride was delivered (69.8 ± 0.7 vs. 57.8 ± 0.7 µm/beat, amiloride vs. control, respectively). The interaction was consistently improved in studies where iv furosemide was administered (90.0 ± 0.9 µm/beat). Furosemide also dramatically increased beat-transport translation interaction in the baboon animal model (Fig. 6B) with respect to the control study (11.5 ± 0.1 µm/beat), whether delivered iv or by aerosol (29.5 ± 0.4 vs. 46.5 ± 0.7 µm/beat, respectively). It can be seen in Fig. 7 that aerosol administration resulted not only in a larger increase in cilia-mucus interaction but also in an earlier maximal response than did iv administration.

DISCUSSION

One of the major findings of the present study is the considerable evidence that furosemide causes an increase in the net water flux toward the airway lumen, facilitating an increase of mucus transport. 1) Furosemide increased mucociliary transport when administered iv to both dogs and baboons and when administered by aerosol to baboons (Figs. 1 and 2). 2) Furosemide reversed the decrease in bronchial clearance caused by inhalation of dry air (Fig. 3). 3) Dry-air inhalation reversed the furosemide-induced increase in mucociliary transport (Fig. 3). 4) RTFO and tracheal fluid depth increased after iv administration in dogs (Fig. 4). 5) The increase in mucus transport by furosemide was effected by an improved coupling of the viscoelastic mucus to the cilia (Figs. 6 and 7). The finding that enhanced mucociliary transport was facilitated by increased airway hydration is consistent with a model of mucociliary transport operating at a low basal level under resting conditions.

The minor transient responses of the mucociliary transport system to amiloride observed herein (Fig. 1) are consistent with other reports on the effects of amiloride in mammals. Bronchial clearance in patients with cystic fibrosis has been found to be not statistically different from control >10 min after delivery of aerosolized amiloride (1). This transient response is consistent with a half-life of 10.5 min for amiloride in the airways (20). Although inhibition of the apical Na⁺ channel by amiloride may increase the mucosal Na⁺ concentration, it does not necessarily result in an increased hydration of the airway lining (33). The nature of the responses of the epithelium to amiloride may be related to restoration of intracellular Na⁺ concentration by the Na-K-2Cl cotransporter and/or restoration of electrochemical membrane potential by Cl efflux.

Acetylstrophanthidin was chosen instead of ouabain to inhibit the 3Na-2K-ATPase due to a more rapid onset, shorter duration of action, and higher lipid solubility. Aerosol delivery was chosen to minimize the systemic effects typical of cardiac glycosides while attempting to deliver an effective dose to inhibit basolateral epithelial 3Na⁺/2K⁺ exchange. Although less effective than the same dose applied basolaterally, apical application of ouabain can inhibit 3Na-2K-ATPase (38), and the more lipid-soluble acetylstrophanthidin was expected to be even more effective. The increase in CBF observed after administration of acetylstrophanthidin (Fig. 1) is consistent with the increased CBF in Stentor polymorphus (a sessile protozoan) induced by digitoxin (27), another 3Na-2K-ATPase inhibitor. Such inhibitors also dramatically decrease intracellular Cl concentrations (L. B. Wong, private communication). Whereas TMV did not increase after administration of acetylstrophanthidin in the present study, TMV in dogs was enhanced after administration of an extremely high (0.025 mg/kg) iv dose of acetylouabain (17), also a 3Na-2K-ATPase inhibitor. The large increase in BMC, without complementary increase in TMV, may in part be the result of a more proximal iron oxide aerosol-deposition pattern, as suggested but not substantiated in R24% analysis.

The effects of iv furosemide in the baboon (Fig. 2) correlated well with the results in the dog (Fig. 1), indicating that basal electrolyte-transport differences between dog and primate airways determined in vitro (35) are not indicative of different physiological responses of the mucociliary transport system in vivo. That furosemide was more effective when delivered by aerosol (Fig. 2) should not be surprising, because furosemide, which has some luminal action on the Na-K-2Cl cotransporter (36, 37), likely achieved a high concentration in the lining fluid, and, because of its polar nature, was unlikely to diffuse readily into the blood.

In an apparent contradiction of our findings, inhaled furosemide has been reported to have no effect on mucociliary clearance in humans (12). In that study, only 10% of the 40 mg aerosolized was likely deposited within the lungs, compared with the 20-mg deposition confirmed within the lungs in the present study. In addition, the mass of furosemide required to cause the same effect in the standard 70-kg human subjects vs. 15- to 40-kg baboons could be substantially larger. Given the differences in deposition and dosing, the study by Hasani et al. (12) suggests only that furosemide deposition at <10% of that used in our experiments may be ineffective.

The transient suppression of BMC, TMV, and CBF during dry-air inspiration (Fig. 3) is consistent with a previous report of TMV reduction in dogs by dry air (13) as well as with our conjecture that the hydration of airway lining fluid in the unstimulated airway is at a low basal level. The proportionate decrease in CBF and TMV during initial dry-air inspiration may be an external consequence of increased viscous resistance from a dehydrated mucus, opposing ciliary motion. However, there was minimal change in the translation of CBF into TMV (Fig. 6), suggesting no important changes occurred in viscous resistance. The decrease of CBF in response to airway dehydration may result from an inhibitory neural suppression of the mucociliary transport system.

The subsequent reversal of dry-air-induced mucociliary suppression, with proportionate increases in TMV and CBF (Fig. 3), is consistent with studies that showed a stimulation of BMC in patients with asthma after 6- to 8-min hyperventilation with dry air (5). The absence of recovery of BMC in our results may derive from the lower bronchial hydration stress of spontaneous dry-air inspiration compared with hyperventilation, as both water transport and heat transport occur deeper in the lung with moderate hyperventilation, or it may derive from the prolonged bronchial hydration stress of continuous dry-air inspiration compared with that after cessation of a short (6-8 min) hyperventilation. The recovery of TMV and CBF implies the existence of a compensatory mechanism to increase water transport to the airway in response to dehydration. The time course of the inhibition and recovery of TMV and CBF is similar to the time course of decrease and subsequent increase in expired humidity during dry-air inhalation (30), as well as the delay suggested to be required for insertion of Na-K-2Cl cotransporters (10) or aquaporins into the epithelial membrane (16). The finding that RTFO and the estimated depth of the tracheal fluid increased after furosemide administration (Fig. 4) is suggestive that net water flux to the mucosa increased.

Since informal presentation of our findings, the effects of furosemide combined with dry air have been studied by others. Specifically, Daviskas et al. (4) combined aerosolized furosemide with dry-air hyperventilation in healthy and asthmatic subjects. In healthy subjects, inhalation of the vehicle control appears to have inhibited the increase in mucociliary transport, previously reported to be induced by dry-air hyperventilation (5). Under such conditions, the ability to observe effects due to the administration of furosemide is compromised. The inhibition of a hyperventilation-induced increase in mucociliary transport by furosemide in persons with asthma is consistent with data in the present study and with our initial findings in dogs (42).

The coordination of regional tracheobronchial transport (Fig. 5) appears to be easily disturbed. However, it is quite reasonable that, when altered, bronchial transport should be reduced relative to tracheal transport (or that tracheal transport should be increased relative to bronchial transport) to avoid the accumulation of secretions at the tracheobronchial junction.

Consistent with the lack of change in rabbit tracheal CBF in vitro on exposure to 10⁴ M furosemide (31), CBF in the present study was largely unaffected by iv furosemide in both dogs and baboons (Figs. 1 and 2). Aerosolized furosemide decreased CBF while producing the largest increases in mucociliary transport. Therefore, the increases in mucociliary transport induced by furosemide were due to an increase in the distance mucus was transported per ciliary beat rather than to an increase in beat frequency (Figs. 6 and 7).

In the dog control study, the coordination of mucus movement per ciliary beat (Figs. 6 and 7) indicates a mean transport of mucus of 58 µm per ciliary beat. This distance is several times the length of the cilia and suggests that the mucus layer is disengaged and moving above cilia in the recovery phase of the stroke and/or within the metachronal field. The increase of this transport distance to 90 µm with perturbation implies that the mucus layer is advancing toward an optimal interaction with the tips of the cilia, with mucus velocity approximating ciliary tip velocity. Ciliary tip velocities of 1,000 µm/s at a beat rate of 10 Hz (i.e., 100 µm/beat) have been reported (28). The transport of mucus per ciliary beat in the baboon control study (14 µm) is closer to the length of the cilia and thus suggests that the mucus layer is normally more fully engaged with the cilia in the baboon, consistent with the differing flux "set point" in the trachea between the dog (primarily Cl secreting) and baboon (primarily Na⁺ absorbing). It is notable that none of these perturbations decreased the effectiveness of the translation of ciliary beat to mucus transport, indicating the robustness of the mucociliary transport system and its ability to adapt to perturbation.

This calculated index of ciliary beat effectiveness, although informative, must be interpreted cautiously because other components of ciliary motion, such as metachronal wave period of the cilia bed (41), amplitude of the ciliary beat, or cilia stroke velocity were not measured and may be regulated independently of the regulation of CBF. Furthermore, this index requires the implicit assumption that the CBF, determined from backscattered photons of a laser beam with a focal diameter of 7 µm on the epithelium in the midtrachea, was representative of ciliary motion throughout the trachea and that measurement of each component of mucociliary transport on separate days of experimentation was representative of the systems response to each perturbation.

In addition to its primary action on the cotransporter, there are other potential mechanisms whereby furosemide may have elicited the observed responses. Furosemide appears to relax smooth muscle constriction by inhibiting prostaglandin synthesis. However, administration of furosemide did not change prostaglandin synthesis in the nasal mucosa (21) or inhibit its production in airway epithelium (18). Diuresis from furosemide might also induce the release of vasopressin, angiotensin, or other mediators that may stimulate mucociliary transport. However, inhalation of furosemide (1 mg/kg) did not cause diuresis either in the present study or in the study of Rastogi et al. (25). Also, furosemide increased mucociliary transport in the dog studies in which urine output was matched with iv volume replacement, as well as in baboon studies where there was no attempt at volume replacement. Furosemide is also reported to inhibit cholinergic and excitatory nonadrenergic, noncholinergic neurotransmission in the airways (6). However, for this effect to cause the observed result, furosemide would have to block inhibitory sensory nerves that regulate secretion so that their inhibition thereby stimulates mucociliary transport. A furosemide-induced increase in microvascular leakage (8) is consistent with our observations.

As stated above, the data strongly imply that furosemide increases the net water balance in the airway lumen. We suggest that it is more likely that furosemide causes this water balance by inhibition of the basolateral Na-K-2Cl cotransporter rather than by other documented or undocumented actions of furosemide. On cursory review, this may seem an antithesis to the generally held paradigm that water flux toward the airway lumen is associated with Cl transport across the apical epithelial membrane. The data need not be so considered. Rather, the data suggest that an increase in net water content in the airway lumen is not always associated with an increase in net Cl transport into the airway lumen.

In the case of severe airway perturbation, such as hyperventilation with dry air (4, 5), released mediators may cause a receptor-induced increase of intracellular inositol triphosphate or adenosine 3,5-cyclic monophosphate, with the subsequent activation of apical Cl channels and consequent efflux of Cl from the cell, followed by an obligatory cell shrinkage. Such shrinkage could promote an increase in net water flux from the submucosa to the lumen. Cell shrinkage is also associated with compensatory increased activity of the basolateral Na-K-2Cl cotransporter (10), likely through phosphorylation, in an effort to maintain cellular volume homeostasis. Thus this sequence of events results in an augmentation of basolateral-to-luminal transport of both Cl and water. Teleologically, such a response would protect a dehydrated epithelial lining.

During mild dry-air perturbation or under homeostatic conditions in which apical Cl channels are not likely to be receptor activated, augmentation of Cl transport into the airway lumen is not expected. However, in both mild dry-air and furosemide perturbations (as well as the extreme perturbation above), a predicted decrease occurs in the intracellular Cl content and cell volume (7, 8, 36). It is possible these decreases act as second messengers (32) to initiate a net water flux from the submucosa to the airway lumen. Such water flux may result from decreased cell volume by increased paracellular permeability to water flux into the airway, analogous to responses in the ophthalmic trabecular network (23). This is consistent with an osmotically induced increase in mannitol permeability in native tracheal epithelia (43), although it is inconsistent with the osmotically induced decrease in mannitol permeability in cultured nasopharyngeal epithelial cells (39). A decrease in intracellular Cl has also been linked to increased transmembrane permeability to water flux through the increased expression of aquaporins within epithelial membranes (9). Aquaporins permit water flux down an electrochemical gradient (16) that, although it may be minimal during homeostatic perturbations, may still allow ample water movement if the permeability is substantially increased. Finally, active water transport has recently been proposed to rationalize occurrences related to water transport across biological membranes that are not explained by simple flux down an electrochemical gradient (44). Such active water transport may have a role in producing water flux without Cl transport. Therefore tracheobronchial secretion may not be strictly associated with Cl flux into the airway but rather with increased net flux of Cl out of the epithelial cells, resulting from either increased Cl efflux or decreased Cl influx.

In summary, the use of aerosolized amiloride to enhance the clearance of secretions from the lungs in normal airways is not supported by the minimal, transient responses observed in mucociliary transport with this Na⁺-channel inhibitor. The potential stimulation of the mucociliary transport system (BMC and CBF) by aerosolized acetylstrophanthidin-induced inhibition of 3Na-2K-ATPase should be confirmed. The increases in airway fluid transport, mucociliary transport, and mucus transport-ciliary beat when furosemide was administered either to dogs or to baboons, either intravenously or by aerosol, as well as the counterbalancing mucociliary transport responses of dry-air inhalation and furosemide, all suggest an important role of the Na-K-2Cl cotransporter in the regulation of airway hydration and mucociliary transport.

Thus pharmacological manipulation of the periciliary layer may provide an effective mechanism to treat persons with dry, inspissated secretions and to augment an impaired mucociliary clearance system. Aerosolized loop diuretics and cardiac glycosides may be clinically effective in improving the removal of secretions from the airways.

ACKNOWLEDGEMENTS

We extend special thanks to A. Daza for assistance in data collection and to Dr. L. Wong for advice and comments.

FOOTNOTES

Address for reprint requests: D. B. Yeates, Dept. of Medicine (M/C 788), Univ. of Illinois at Chicago, 1940 West Taylor St., Rm. 212, Chicago, IL 60612 (E-mail: YEATES-D{at}UIC.EDU).

Received 7 March 1996; accepted in final form 10 June 1997.

REFERENCES

1.	Bennett, W. D., K. N. Oliver, K. L. Zeman, K. W. Hohneker, R. C. Boucher, and M. R. Knowles. Effect of uridine 5-triphosphate plus amiloride on mucociliary clearance in adult cystic fibrosis. Am. J. Respir. Crit. Care Med. 153: 1796-1801, 1996[Abstract].
2.	Boucher, R. C., M. J. Stutts, and J. T. Gatzy. Regional differences in bioelectric properties and ion flow in excised canine airways. J. Appl. Physiol. 51: 706-714, 1981[Abstract/Free Full Text].
3.	Chandra, T., D. B. Yeates, I. F. Miller, and L. B. Wong. Stationary and nonstationary correlation-frequency analysis of heterodyne mode laser light scattering: magnitude and periodicity of canine tracheal ciliary beat frequency in vivo. Biophys. J. 66: 878-890, 1994.
4.	Daviskas, E., S. D. Anderson, I. Gonda, D. Bailey, G. Bautovich, and J. P. Seale. Mucociliary clearance during and after isocapnic hyperventilation with dry air in the presence of furosemide. Eur. Respir. J. 9: 716-724, 1996[Abstract].
5.	Daviskas, E., S. D. Anderson, I. Gonda, S. Eberl, S. Meikle, J. P. Seale, and G. Bautovich. Changes in mucociliary clearance during and after isocapnic hyperventilation with dry air asthmatic and healthy subjects. Eur. Respir. J. 8: 742-751, 1995[Abstract].
6.	Elwood, W., J. O. Lötvall, P. J. Barnes, and K. F. Chung. Loop diuretics inhibit cholinergic and noncholinergic nerves in guinea pig airways. Am. Rev. Respir. Dis. 143: 1340-1344, 1991[Medline].
7.	Foskett, J. K. [Ca2+] modulation of Cl content controls cell volume in single salivary acinar cells during fluid secretion. Am. J. Physiol. 259 (Cell Physiol. 28): C998-C1004, 1990[Abstract/Free Full Text].
8.	Freed, A., V. Tashar, B. Schofield, and C. Omari. Effect of furosemide on hypernea-induced airway obstruction, injury and microvascular leakage. J. Appl. Physiol. 81: 2461-2467, 1996[Abstract/Free Full Text].
9.	Grosso, A., P. Jaquet, P. Brawand, and R. C. Desousa. Anion-induced dynamic behavior of apical water channels in vasopressin-sensitive epithelia exposed to mercury. Am. J. Physiol. 266 (Cell Physiol. 35): C1577-C1585, 1994[Abstract/Free Full Text].
10.	Haas, M., L. G. Johnson, and R. C. Boucher. Regulation of Na-K-Cl cotransport in cultured canine epithelia: a [3H]bumetanide binding study. Am. J. Physiol. 259 (Cell Physiol. 28): C557-C569, 1990[Abstract/Free Full Text].
11.	Hager, J., K. H. Gober, J. P. Löhr, and M. Dürr. Measurement of particle and mass distribution of pentamidine aerosol by ultrasonic and air jet nebulizers. J. Aerosol Med. 5: 65-79, 1992. [Medline]
12.	Hasani, A., D. Pavia, M. A. Spiteri, C. T. Yeo, J. E. Agnew, S. W. Clarke, and F. K. Chung. Inhaled frusemide does not affect lung mucociliary clearance in healthy and asthmatic subjects. Eur. Respir. J. 7: 1497-1500, 1994[Abstract].
13.	Hirsch, J. A., J. L. Tokayer, M. J. Robinson, and M. A. Sackner. Effects of dry air and subsequent humidification on tracheal mucous velocity in dogs. J. Appl. Physiol. 39: 242-246, 1975[Abstract/Free Full Text].
14.	Jiang, C., W. E. Finkbeiner, J. H. Widdicombe, P. B. McCray, Jr., and S. S. Miller. Altered fluid transport across airway epithelium in cystic fibrosis. Science 262: 424-427, 1993[Abstract/Free Full Text].
15.	Knowles, M., G. Murray, J. Shallal, F. Askin, V. Ranga, J. Gatzy, and R. Boucher. Bioelectric properties and ion flow across excised human bronchi. J. Appl. Physiol. 56: 868-877, 1984[Abstract/Free Full Text].
16.	Kuwahara, M., and A. S. Verkman. Direct fluorescence measurement of diffusional water permeability in the vasopressin-sensitive kidney collecting tubule. Biophys. J. 54: 587-593, 1988[Abstract/Free Full Text].
17.	Laurenzi, G. A., and S. Yin. Studies of mucus flow in the mammalian respiratory tract. I. The beneficial effects of acetyl ouabain on respiratory tract mucus flow. Am. Rev. Respir. Dis. 103: 800-807, 1971[Medline].
18.	Levasseur-Acker, G. M., M. Molimard, J. Regnard, E. Naline, C. Freche, and A. Lockhart. Effect of furosemide on prostaglandin synthesis by human nasal and bronchial epithelial cells. Am. J. Respir. Cell Mol. Biol. 10: 378-383, 1994[Abstract].
19.	McMurry, P. H., N. Whitby, and E. Whitby. Droplet Distributions from Two Ultrasonic Medical Nebulizers. Denver, CO: Mountain Medical Equipment, 1990. (Consulting report)
20.	Mentz, W. M., J. B. Brown, M. Fredman, J. Stutts, J. T. Gatzy, and R. C. Boucher. Deposition, clearance, and effects of aerosolized amiloride in sheep airways. Am. Rev. Respir. Dis. 134: 938-943, 1986[Medline].
21.	Mullol, J., I. Ramis, J. Prat, J. Roselló-Catafau, A. Xaubert, C. Piera, E. Gelpi, and C. Picado. Failure of frusemide to increase production of prostaglandin E2 in human nasal mucosa in vivo. Thorax 48: 260-263, 1993[Abstract].
22.	Nathanson, I., J. H. Widdicombe, and J. A. Nadel. Effects of amphotericin B on ion and fluid movement across dog tracheal epithelium. J. Appl. Physiol. 55: 1257-1261, 1983[Abstract/Free Full Text].
23.	O'Donnell, M. E., J. D. Brandt, and F. R. E. Curry. Na-K-Cl cotransport regulates intracellular volume and monolayer permeability of trabecular meshwork cells. Am. J. Physiol. 268 (Cell Physiol. 37): C1067-C1074, 1995[Abstract/Free Full Text].
24.	Pillai, R. S., D. B. Yeates, M. Eljamal, I. F. Miller, and A. J. Hickey. Generation of concentrated aerosols for inhalation studies. J. Aerosol Sci. 25: 187-197, 1994.
25.	Rastogi, A., M. Luayon, O. A. Ajayi, and R. S. Pildes. Nebulized furosemide in infants with bronchopulmonary dysplasia. J. Pediatr. 125: 976-979, 1994[Medline].
26.	Shephard, K. L., and H. Rohmoune. Evaporation-induced changes in airway surface liquid on an isolated guinea pig trachea. J. Appl. Physiol. 76: 1156-1165, 1994[Abstract/Free Full Text].
27.	Sleigh, M. A. Metachronism and frequency of beat in the peristomial cilia of "Stentor." J. Exp. Biol. 33: 15-28, 1956. [Abstract]
28.	Sleigh, M. A., J. R. Blake, and N. Liron. The propulsion of mucus by cilia. Am. Rev. Respir. Dis. 137: 726-741, 1988[Medline].
29.	Smith, J. J., and M. J. Welsh. Fluid and electrolyte transport by cultured human airway epithelia. J. Clin. Invest. 91: 1590-1597, 1993.
30.	Tabka, Z., J. Ben Jebria, and H. Guénard. Effect of breathing dry warm air on respiratory water loss at rest and during exercise. Respir. Physiol. 67: 115-125, 1987[Medline].
31.	Tamaoki, J., M. Kondo, and T. Takizawa. Effect of cAMP on ciliary function in rabbit tracheal epithelial cells. J. Appl. Physiol. 66: 1035-1039, 1989[Abstract/Free Full Text].
32.	Tohda, H., J. K. Foskett, H. O'Brodovich, and Y. Marunaka. Cl regulation of Ca2+-activated non-selective cation channel in -agonist-treated fetal distal lung epithelium. Am. J. Physiol. 266 (Cell Physiol. 35): C104-C109, 1994[Abstract/Free Full Text].
33.	Tomkiewicz, R. P., E. M. App, J. G. Zayaz, O. Ramirez, N. Church, R. C. Boucher, M. Knowles, and M. King. Amiloride inhalation therapy in cystic fibrosis. Am. Rev. Respir. Dis. 148: 1002-1007, 1993[Medline].
34.	Wales, K., G. Petrow, and D. B. Yeates. Production of Tc-99m labelled iron oxide aerosols for human lung deposition and clearance studies. Int. J. Appl. Radiat. Isotopes 31: 689-694, 1980. [Medline]
35.	Welsh, M. J. Electrolyte transport by airway epithelia. Physiol. Rev. 67: 1143-1184, 1987[Free Full Text].
36.	Welsh, M. J. Inhibition of chloride secretion by furosemide in canine tracheal epithelium. J. Membr. Biol. 71: 219-226, 1983[Medline].
37.	Widdicombe, J. H., I. T. Nathanson, and E. Highland. Effects of "loop" diuretics on ion transport by dog tracheal epithelium. Am. J. Physiol. 245 (Cell Physiol. 14): C388-C396, 1983[Abstract/Free Full Text].
38.	Widdicombe, J. H., I. F. Ueki, I. Bruderman, and J. Nadel. The effects of sodium substitution and ouabain on ion transport by dog tracheal epithelium. Am. Rev. Respir. Dis. 120: 385-392, 1979[Medline].
39.	Willumsen, N. J., C. W. Davis, and R. C. Boucher. Selective response of human airway epithelia to luminal but not serosal solution hypertonicity. J. Clin. Invest. 94: 779-787, 1994.
40.	Winters, S. L., and D. B. Yeates. Roles of hydration, sodium, and chloride in regulation of canine mucociliary transport system. J. Appl. Physiol. 83: 1360-1369, 1997[Abstract/Free Full Text].
41.	Wong, L. B., I. F. Miller, and D. B. Yeates. Nature of the mammalian ciliary metachronal wave. J. Appl. Physiol. 75: 458-467, 1993[Abstract/Free Full Text].
42.	Wong, L. B., and D. B. Yeates. Dynamics of the regulation of ciliated epithelial function. Comments Theor. Biol. 4: 183-208, 1997.
43.	Yankaskas, J. R., J. T. Gatzy, and R. C. Boucher. Effects of raised osmolarity on canine tracheal epithelial ion transport function. J. Appl. Physiol. 62: 2241-2245, 1987[Abstract/Free Full Text].
43a.	Yeates, D. B., and M. Mayhugh. A phoswich multidetector probe for measuring tracheal mucus velocity. Clin. Phys. Physiol. Meas. 5: 313-320, 1984[Medline].
44.	Zeuthen, T., and W. D. Stein. Cotransport of salt and water in membrane proteins: membrane proteins as osmotic engines. J. Membr. Biol. 137: 179-195, 1994[Medline].