INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

MOVEMENT OF SOLUTES AND WATER between the air space and capillary compartments in the distal air spaces of the lung involves transport across epithelial, interstitial, and endothelial barriers. The tight epithelium of the alveolus and distal airways provides the principal permeability barrier and carries out the active absorption of excess fluid from the air spaces (10, 15, 29). The alveolar epithelium, which occupies the majority of the distal air space surface, contains type I and type II epithelial cells. Cell culture experiments indicate that type II cells are able to transport Na⁺ and Cl utilizing luminal epithelial Na⁺ channels (ENaC) and possibly CFTR Cl channels and basolateral membrane Na⁺-K⁺ pumps (16, 20, 21, 27). Less information is available about the physiology of type I cells because suitable cell culture models are not available; however, measurements in freshly isolated immunopurified type I cells indicate that they are highly water permeable (9) and may contain ENaC (22). Distal airways are moderately water permeable (12), and, although little information is available, they are thought to carry out rapid fluid transport as do epithelial cells in the larger proximal airways (1, 3, 5, 23, 28).

Intact lung preparations have been used to measure salt and water transport across the distal air space barrier (13, 29). The air space is generally filled with fluid containing a radiolabeled volume marker such as ¹²⁵I-albumin, and fluid movement into or out of the air space compartment is deduced from changes in volume marker concentration in air space fluid samples. There is a considerable body of data showing that fluid is absorbed isotonically from the air space compartment by an amiloride-sensitive, active fluid transport process that is stimulated by cAMP agonists (10, 29). Rapid, osmotically driven water transport across the air space-capillary barrier was originally demonstrated in perfused sheep lung from the rate of osmotic equilibration after instillation of hyperosmolar fluid into the air spaces (13). Ion transport in intact lung has been measured from the disappearance of radioactive ²²Na⁺ and ³⁶Cl from instilled air space fluid (11, 32). Utilizing volume marker and tracer uptake methods, it was reported recently that cAMP-stimulated fluid absorption and Cl transport out of the air space fluid in mouse lung were inhibited by glibenclamide and CFTR gene deletion, suggesting the involvement of CFTR in distal lung Cl transport (11). However, a concern in the use of volume markers and radioactive labels is that invasive air space fluid sampling is required, which limits the number of samples that can be obtained in each experiment (generally 1-2 samples in mouse lung) and introduces uncertainties due to contamination of samples by fluid in proximal airways.

Our laboratory previously introduced a pleural surface fluorescence method to measure osmotic water transport continuously in perfused lungs (6). The air space was filled with a fluorescent volume marker whose concentration (proportional to air space fluid osmolality) was deduced from the fluorescence signal at the pleural surface. Osmotic water permeability was measured from the time course of pleural surface fluorescence in response to changes in osmolality of the pulmonary artery perfusate. The pleural surface fluorescence method was used to quantify changes in lung water permeability near the time of birth (8) and to determine the role of aquaporin water channels AQP1, AQP4, and AQP5 in lung water permeability utilizing knockout mice (2, 24, 35). An interesting observation was that reduction of lung water permeability by >30-fold by aquaporin deletion had no effect on active fluid absorption from the distal air spaces (24).

The purpose of this study was to develop and apply pleural surface fluorescence methods to measure the transport of Cl and Na⁺ into the air space compartment in perfused mouse lung. Membrane-impermeant fluorescent indicators of Cl and Na⁺ were introduced into the fluid-filled air space compartment to measure Cl ([Cl]) and Na⁺ concentrations ([Na⁺]) continuously in response to changes in perfusate ionic composition, addition of transporter agonists or inhibitors, or gene knockout. Fluorescent indicators were selected with bright, long-wavelength fluorescence and with high sensitivities to Cl and Na⁺. The new methods were validated by analysis of air space fluid samples and applied to characterize the mechanisms of Cl and Na⁺ transport across the distal air space barrier in mouse lung. Measurements were also made in CFTR null (cystic fibrosis) mice to analyze the role of CFTR in distal lung salt transport, and in aquaporin null mice to test whether AQP1 or AQP5 deletion in lung induces compensatory changes in functional salt transporters.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Mice. Mice in a CD1 genetic background (age 6-8 wk, 22-35 g) were bred for these studies. CFTR null (and matched wild-type) mice in a B6D2/129 genetic background were bred at the University of North Carolina (34). AQP1 and AQP5 null mice were generated by targeted gene disruption (25, 26). All animal procedures were approved by the University of California San Francisco Committee on Animal Research.

Isolated lung perfusion. Mice were euthanized using intraperitoneal ketamine (40 mg/kg) and xylazine (8 mg/kg). The trachea was cannulated with polyethylene PE-90 tubing, and the pulmonary artery was cannulated with PE-20 tubing. The left atrium was transected to permit fluid exit. The pulmonary artery was gravity perfused at constant pressure (8-10 cmH₂O) at room temperature, giving flow of ~0.5 ml/min. The air space was filled with 0.5 ml of buffered isosmolar solutions containing the fluorescent indicators lucigenin or sodium green. The compositions of air space instillate and pulmonary artery perfusate solutions are given in Table 1.

Pleural surface fluorescence measurements. Air space fluid fluorescence was measured by a pleural surface fluorescence method, as modified from our laboratory's previous method for measurement of lung water transport (6, 7). The heart and lungs were positioned in a perfusion chamber for observation by epifluorescence microscopy. Lucigenin (50 µM, Molecular Probes) or Sodium Green (20 µM, Molecular Probes) was added to the air space fluid as indicators of Cl or Na⁺, respectively. The fluorescence from a 3- to 5-mm-diameter spot on the lung pleural surface was monitored continuously with an inverted epifluorescence microscope using a ×10 air objective (Leitz, numerical aperture of 0.25) and filter set containing 470 ± 15-nm (for lucigenin) or 490 ± 15-nm (for Sodium Green) excitation filter, 510-nm dichroic mirror, and >530-nm cut-on filter. Signals were detected by a photomultiplier, amplified, digitized, and recorded at a rate of 1 Hz.

Cl transport measurements. The pulmonary artery was first perfused with Cl free solution (solutions A or E, see Table 1) until the lungs became white. Air (0.5 ml) was injected into the air space with the lungs submersed under water to confirm that the lung was intact. After deflation, 0.5 ml of isosmolar fluid containing 50 µM lucigenin were infused into the air space, and the trachea was sutured closed. The total time for lung preparation from chest incision to fluorescence measurements was 10-15 min. The heart-lung block was positioned on the microscope stage, and a fluorescence baseline was recorded. The perfusate was then switched to a Cl-containing solution (solution C). In some experiments, the perfusate and instillate contained amiloride (100 µM), glibenclamide (100 µM), or bumetanide (100 µM), or the perfusate contained IBMX (200 µM) and forskolin (20 µM). Some experiments were done using Na⁺-free solutions (solutions E and F), or K⁺-free solutions in which K⁺ was replaced by Na⁺.

Na+ transport measurements. After lung preparation as described above, 0.5 ml of an isosmolar Na⁺-free solution (solution B) containing 20 µM Sodium Green were instilled into the air spaces. The pulmonary artery was perfused with the same Na⁺-free solution for 5 min and then switched to a Na⁺-containing solution (solution D). In some experiments, the instillate and/or perfusate contained forskolin/IBMX, amiloride, or glibenclamide at the concentrations given above. Some experiments were done using Cl-free solutions in the instillate and perfusate (solutions G and H).

Computation of [Cl] and [Na⁺] and fluxes from fluorescence data. To determine influx rates of Cl and Na⁺ (in mM/min), the time course of air space fluid [Cl] and [Na⁺] was computed from the measured fluorescence time course, F(t). The data in Fig. 1B, left, for lucigenin quenching by Cl define a Stern-Volmer relation, F_o/F = 1 + K_Cl [Cl], where F_o is fluorescence in the absence of Cl, F is fluorescence in the presence of Cl, and K_Cl is the Stern-Volmer quenching constant (0.395 mM¹). For measurements in intact lung, small corrections are needed for background signal (F_b) and photobleaching/dye leakage (r_b, defined below). If F_c(t) is the background and photobleaching-corrected time course of pleural surface fluorescence, the Stern-Volmer equation becomes

(1)

where [Cl(t)] is the time course of air space fluid [Cl]. If r_b is the linear rate of photobleaching/dye leakage as defined by F(t)  F_b = [(F_o  F_b)(1  r_bt)], then

(2)

where F(t) is measured (uncorrected) fluorescence. [Cl(t)] thus becomes

(3)

In each experiment, F_b is measured from the lung surface before airway fluid instillation (F_b is generally <5% of F_o), and r_b is determined from a linear regression of F(t) before Cl addition (r_b is generally <0.003 min¹). The initial rate of Cl influx (d[Cl]/dt, in mM/min) is computed from the derivative of [Cl(t)] evaluated at t = 0 

(4)

where dF(0)/dt is computed from measured F(t) using a second-order polynomial regression.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Characteristics of Cl and Na+ indicators for measurement of distal lung ion transport. A: fluorescence excitation and emission spectra of lucigenin (50 µM in solution A, see Table 1; left) and Sodium Green (20 µM in solution B; right). B: sensitivity of lucigenin fluorescence to Cl concentration ([Cl]; left) and of Sodium Green fluorescence to Na+ concentration ([Na+]; right). Values are means ± SE. Fo, fluorescence in the absent of [Cl] or [Na+]; F, fluorescence in the presence of [Cl] or [Na+].

(5)

(6)

Assay of Na+ and Cl in air space fluid samples. To validate fluorescence measurements, in some experiments air space fluid samples were obtained and assayed for [Cl] or [Na⁺]. Lungs were isolated, the pulmonary artery was perfused, and the air space was filled with 0.5 ml of fluid as described above. Fluid samples (20 µl) were withdrawn at specified times using PE-90 tubing attached to a 1-ml syringe. For each fluid sample, the initial 20 µl withdrawn were discarded. [Cl] was assayed using a dual-wavelength fluorescence method as described previously (36), and [Na⁺] was measured by flame photometry.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

For measurement of Cl and Na⁺ transport into the fluid-filled air space compartment, ion-selective indicators were chosen that were membrane impermeant and had bright, long-wavelength fluorescence, pH insensitivity, and sensitivity to Cl and Na⁺ in a concentration range appropriate for transport measurements in perfused lung preparations. Because of the relatively slow rates of ion transport in lung, estimated from ³⁶Cl and ²²Na measurements in mouse lung to be ~0.7%/min equilibration in response to air space-capillary gradients of [Cl] or [Na⁺] (11), experiments were designed to measure ion influx using highly [Cl]- and [Na⁺]-sensitive indicators in the air space fluid. Figure 1A shows the fluorescence spectra of lucigenin (left) and Sodium Green (right), and Fig. 1B shows their sensitivities to [Cl] and [Na⁺], respectively. Both indicators are highly polar and have bright, long-wavelength, pH-insensitive fluorescence. Lucigenin fluorescence is also insensitive to cation composition and to the anions sulfate, phosphate, and NO (4), as well as HCO in the range 0-30 mM. Sodium Green fluorescence is insensitive to anion composition and to the cations choline⁺ and N-methylglucamine⁺ (NMG⁺) and has a Na⁺-to-K⁺ selectivity ratio of ~40 (39).

Cl transport into the fluid-filled air space compartment was measured from the time course of air space fluid lucigenin fluorescence, measured at the pleural surface, in response to an increase in pulmonary artery perfusate [Cl]. The pulmonary artery was initially perfused with a Cl-free isosmolar solution (NO replacing Cl), and the air space was instilled with the same solution but contained lucigenin. Figure 2A shows that Cl addition to the perfusate produced a slow decrease in lucigenin fluorescence whose initial slope depended on perfusate [Cl]. In some experiments, there was a slow decrease in fluorescence (0.8-1.5%/min), even in the absence of Cl, which weakly depended on illumination intensity and probably corresponds to a combination of photobleaching and slow dye leakage out of the air space compartment. Computed rates of Cl influx (after correction for dye photobleaching/leakage) were approximately linear with [Cl]: 0.8, 1.8, and 3.0 mM/min for perfusate [Cl] of 30, 60, and 120 mM, respectively.

View larger version (43K): [in this window] [in a new window]   Fig. 2.   Pleural surface measurement of Cl transport across the distal lung barrier. A: representative time course of pleural surface lucigenin fluorescence in response to addition of indicated [Cl] to the pulmonary artery perfusate. The Cl free perfusate (solution A, Table 1) was switched to Cl containing perfusates (NO replacing Cl). Curves are displaced for clarity. B: confocal fluorescence micrograph of pleural surface of lung prepared as in A. Scale bar: 200 µm. C: time course of air space fluid [Cl] (measured in air space fluid samples) for increase in perfusate [Cl] from 0 to 30 mM. Values are means ± SE; n = 3 lungs.

Figure 2B shows a confocal micrograph of the pleural surface of a lung containing lucigenin in the air space compartment, prepared for functional studies as in Fig. 2A. Fluorescence was seen in the fluid-filled air spaces with dark septa separating alveoli. Selective air space fluid staining was seen for >1 h in the perfused lung preparation. Occasionally seen microvessels were nonfluorescent. To validate the accuracy of [Cl] measurements, in some studies air space fluid samples were obtained at 0, 1, 2, and 5 min and assayed for [Cl]. Figure 2C shows an approximately linear increase in perfusate [Cl] over 5 min, with an influx rate (slope) of 0.9 mM/min, in good agreement with the influx rate of 1.0 ± 0.1 mM/min determined from the lucigenin pleural surface fluorescence measurements (computed using Eq. 4).

The pleural surface fluorescence method was used to study Cl transporting mechanisms. Figure 3A, first curve, shows that, after addition of 30 mM Cl to the perfusate, cAMP stimulation by forskolin and IBMX produced a significantly increased negative slope of the fluorescence vs. time curve, indicating more rapid Cl influx. Cl influx increased from 1.0 ± 0.1 mM/min before to 2.1 ± 0.3 mM/min after forskolin addition (averaged data summarized in Fig. 3B). The increase in slope was blocked by inclusion of 100 µM glibenclamide in the instillate and perfusates throughout the experiment (second curve). Experiments were done to deduce whether Cl transport into the air space compartment was accompanied by cation entry and/or NO exit (to maintain electroneutrality). There was no significant effect of inhibition of ENaC-mediated transcellular Na⁺ transport by amiloride (third curve), or Na⁺-K⁺-Cl cotransporter-mediated transport by bumetanide (not shown), providing evidence for Cl/NO exchange as the primary mechanism for Cl entry into the air space fluid in these experiments. This conclusion was supported by the finding that Na⁺ replacement by the relatively impermeant cation NMG⁺ (fourth curve) in the instillate and perfusates had little effect on the rates of Cl entry before or after forskolin addition. Also, Cl entry was not affected by reduction of perfusate [K⁺] to zero (K⁺ replaced by NMG⁺) (influx rate 0.77 ± 0.14 mM/min), providing evidence against K⁺-Cl cotransport.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Mechanistic analysis of distal lung Cl transport. Experiments were done as in Fig. 2A with increase in perfusate [Cl] from 0 to 30 mM. A, first curve: increased rate of Cl influx after cAMP stimulation by addition of forskolin (20 µM) and IBMX (200 µM) in buffer A; second curve: same experiment but with inclusion of glibenclamide (100 µM) in the instillate and perfusates throughout the experiment; third curve: same but with amiloride (100 µM) in instillate and perfusates; fourth curve: same as in first curve, but with Na+ replaced by N-methylglucamine+ (solutions E and F) in instillate and perfusates. B: normalized slopes (left ordinate) and deduced rates of Cl influx (d[Cl]/dt; right ordinate) for the maneuvers in A. Values are means ± SE; n = 4-10 lungs. * P < 0.01 for cAMP-stimulated vs. basal Cl influx (Student's t-test).

Na⁺ transport into the fluid-filled air space compartment was measured by a similar approach in which the pulmonary artery was initially perfused with Na⁺-free isosmolar solution, and the air space was instilled with the same solution but containing Sodium Green. Figure 4A shows that the addition of Na⁺ to the perfusate resulted in an increase in fluorescence. As done for Cl, the accuracy of [Na⁺] measurements was validated by [Na⁺] assays on air space fluid samples. The rate of Na⁺ influx measured by fluorescence in response to a 30 mM Na⁺ gradient, 1.2 ± 0.3 (SE) mM/min (4 lungs, computed from Eq. 7), agreed with Na⁺ influx measured in fluid samples, 0.8 mM/min (3 lungs).

View larger version (31K): [in this window] [in a new window]   Fig. 4.   Pleural surface measurement of Na+ transport across the distal lung barrier. A: time course of pleural surface Sodium Green fluorescence in response to Na+ addition to the pulmonary artery perfusate. The Na+-free perfusate (solution B) was switched to Na+-containing perfusates (N-methylglucamine+ replacing Na+). Curves were displaced for clarity. B: mechanistic analysis of distal lung Na+ transport. Experiments were done as in A with increase in perfusate [Na+] from 0 to 30 mM. First curve: increased rate of Na+ influx after cAMP stimulation by addition of forskolin (20 µM) and IBMX (200 µM); second curve: same experiment but with inclusion of amiloride (100 µM) in the instillate and perfusates throughout the experiment; third curve: same but with glibenclamide (100 µM) in instillate and perfusates; fourth curve: same as in first curve but with Cl replaced by gluconate (solutions G and H). C: normalized slopes (left ordinate) and deduced rates of Na+ influx (d[Na+]/dt; right ordinate) for the maneuvers in B. Values are means ± SE; n = 4 lungs. Slopes were computed just before (forskolin/IBMX) vs. after forskolin/IBMX addition (+forskolin/IBMX). * P < 0.05 compared with forskolin/IBMX-stimulated Na+ transport under control conditions (ANOVA).

Na⁺ transporting mechanisms were studied by using transport agonists, inhibitors, and ion substitution. Figure 4B (first curve) shows that cAMP stimulation by forskolin and IBMX produced an increase in slope corresponding to an increased rate of Na⁺ influx from 1.2 ± 0.3 mM/min (just before cAMP agonist addition) to 5.9 ± 0.7 mM/min (averaged data summarized in Fig. 4C). Inclusion of amiloride in the instillate and perfusates throughout the experiment (second curve) had no significant effect on basal Na⁺ influx but significantly inhibited cAMP-stimulated Na⁺ influx. In these experiments, Na⁺ was substituted by NMG⁺, a relatively membrane-impermeant cation. Substitution of Na⁺ by choline⁺, a commonly used membrane-impermeant cation, resulted in progressive dye leakage and increased pulmonary vascular resistance, suggesting toxicity to the lung. We note that, unlike the situation for Cl/NO substitution, there is not suitable membrane-permeant cation for Na⁺ replacement. To determine whether Na⁺ entry into the air space fluid was accompanied by transcellular transport of Cl to maintain electroneutrality, effects of glibenclamide and Cl replacement by gluconate were studied. Inclusion of glibenclamide (Fig. 4B, third curve) partially inhibited cAMP-stimulated Na⁺ influx, consistent with the glibenclamide inhibition of Cl influx (see Fig. 3) and the need for Cl entry to maintain electroneutrality. Replacement of Cl by the relatively impermeant anion gluconate (fourth curve) produced stronger inhibition of cAMP-stimulated Na⁺ influx.

Measurements of Na⁺ and Cl transport were carried out in lungs of AQP1 and AQP5 null mice to test the hypothesis that aquaporin deletion causes an upregulation of ion transporters to compensate for reduced air space-capillary water permeability. Our laboratory found previously that deletion of these aquaporins did not affect active fluid absorption from the distal air spaces, even when fluid absorption was maximally stimulated by cAMP agonists and type II cell upregulation by keratinocyte growth factor (24). Figure 5A, left, shows that air space-capillary osmotic water permeability, measured by a pleural surface fluorescence method using FITC-dextran as a volume indicator (6), was reduced ~10-fold by AQP1 or AQP5 deletion. Figure 5A, right, is a control study showing that air space fluid volume, assessed by pleural surface FITC-dextran fluorescence, was not changed significantly under the ion substitution conditions used for measurement of Cl and Na⁺ influx. Figure 5B shows representative data for Cl and Na⁺ influx into the air space fluid using the experimental protocols established in Figs. 3 and 4. Influx rates were qualitatively similar, with preservation of cAMP agonist stimulation. The averaged data summarized in Fig. 5C show no effect of AQP1 or AQP5 deletion on basal or simulated Cl or Na⁺ transport across the distal air space barrier in mouse lung.

View larger version (30K): [in this window] [in a new window]   Fig. 5.   Salt and water transport in lungs of aquaporin null mice. A, left: osmotically induced water transport across the air space-capillary barrier of AQP1 and AQP5 null mice. Time course of air space fluid FITC-dextran (volume marker) fluorescence was measured in response to indicated changes in pulmonary artery perfusate osmolality (solution I without or with 200 mM sucrose). Right: time course of FITC-dextran fluorescence in response to Cl and Na+ addition (as in Figs. 3A and 4B), showing little effects of these maneuvers on air space fluid volume. B: representative time courses of Cl transport (left) and Na+ transport (right) in lungs from mice of indicated genotype. Experimental protocol is as in Figs. 3A (first curve) and 4B (first curve). C: averaged Cl (left) and Na+ influx rates (right) for experiments as in B. Values are means ± SE; n = 4-10 mice. Differences are not significant.

Inhibition of cAMP-stimulated Cl transport by glibenclamide (Fig. 3) suggested the involvement of CFTR in this process, although glibenclamide may also affect other Cl transporters as well as K⁺ transporters at concentrations needed to inhibit CFTR (33). Measurements of Cl transport in cystic fibrosis (CFTR null) mice were done to investigate the role of CFTR in basal and cAMP-stimulated Cl entry into the air space fluid compartment. Cl entry studies (as in Fig. 3B, first curve) were done in a series of matched wild-type (Fig. 6A, left) and CFTR null (right) mice. In nearly all wild-type mice, cAMP elevation produced a marked increase in the rate of Cl entry, as seen from the increased slope after forskolin/IBMX addition. In many, but not all, CFTR null mice, there was little or no cAMP-stimulated Cl entry. Figure 6B summarizes rates of Cl entry before and after cAMP stimulation. CFTR deletion in mice caused on average a small increase in basal Cl influx and a decrease in cAMP-stimulated Cl transport (Fig. 6B, left), producing a significant approximately twofold reduction in cAMP-dependent Cl transport (right, P < 0.001).

View larger version (25K): [in this window] [in a new window]   Fig. 6.   Role of CFTR in Cl transport into the air space compartment. A: Cl transport measurements done as in Fig. 2A (first curve) in wild-type (left) and CFTR null (right) mice. Each lucigenin trace was obtained from a different mouse. Fitted slopes (thin lines) are shown of the fluorescence time course after forskolin/IBMX addition. B: Cl influx rates before and after forskolin addition (d[Cl]/dt; left), shown with cAMP-stimulated Cl influx as the difference between forskolin/IBMX-stimulated and basal Cl influx (d[Cl]/dt; right). Values are means ± SE. * P < 0.05 (Student's t-test).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study establishes a pleural surface fluorescence method to measure unidirectional Cl and Na⁺ transport into the air space compartment in a perfused mouse lung preparation. Membrane-impermeant ion-selective indicators were chosen with appropriate optical properties and sensitivities to measure Cl and Na⁺ transport. The ion flux rates deduced by measurement of pleural surface fluorescence were validated by analysis of air space fluid samples. The pleural surface methodology was applied to characterize unidirectional Cl and Na⁺ transport across the distal air space barrier in intact lung and to investigate two questions: Does ion transporter upregulation occurs in transgenic mouse models of aquaporin deficiency, and does distal lung Cl transport require CFTR? Compared with cell culture models, measurements in intact lung preparations are more closely related to in vivo lung physiology because the complex tissue anatomy and cellular heterogeneity are preserved. However, a limitation of intact lung measurements is that it is generally not possible to define the details of apical and basolateral membrane Cl and Na⁺ transporting mechanisms because of the complex mixture of cell types participating in the response in intact lung and because of uncertainties in driving forces. In addition, because of the [Cl] and [Na⁺] sensitivities of available ion indicators and the fairly slow rates of lung ion movement, the experiments required initial perfusion with solutions containing zero [Cl] or [Na⁺] and measurement of Cl and Na⁺ entry (rather than exit) into the fluid-filled air space compartment.

Cl entry into the air space compartment was measured from the time course of air space fluid lucigenin fluorescence after an increase in perfusate [Cl]. The rate of Cl entry increased linearly with perfusate [Cl]. Elevation of cellular cAMP by forskolin and IBMX increased the rate of Cl influx by 2.1-fold. These results are consistent with the study of Nielsen et al. (30), showing an approximately twofold forskolin-induced stimulation of Cl secretion into rabbit lungs containing a zero Cl solution. The cAMP-dependent Cl influx, but not the basal Cl influx, was inhibited by glibenclamide. Neither basal nor cAMP-stimulated Cl influx was significantly inhibited by amiloride or bumetanide, or Na⁺ and K⁺ replacement by NMG⁺, indicating that, under our experimental conditions, Cl influx were accompanied by NO efflux rather than Na⁺ or K⁺ influx. These results indicate that Cl transport across the distal air space barrier is cAMP regulated, which may be responsible, in part, for the cAMP-dependent elevation in alveolar fluid clearance measured in lungs of mice and other mammals.

Na⁺ entry into the air space compartment was measured from the time course of air space fluid Sodium Green fluorescence after a change in perfusate [Na⁺]. However, a limitation in the Na⁺ transport studies was that Na⁺ entry was accompanied in large part by Cl entry to maintain electroneutrality, as evidenced by partial inhibition of Na⁺ entry by glibenclamide and Cl substitution. The incomplete inhibition of Na⁺ entry by Cl substitution may be due to the nonzero gluconate permeability, incomplete interstitial Cl depletion, and/or movement of other ions (such as K⁺) to maintain electroneutrality. The robust cAMP-stimulated Na⁺ entry suggests direct activation of the Na⁺ pathway by cAMP. The persistent cAMP stimulation of Na⁺ entry in the absence of Cl, albeit lesser in magnitude than in the presence of Cl, suggests that cAMP-stimulated Na⁺ entry is not produced exclusively by cAMP-stimulated Cl influx and electrogenic Na⁺-Cl coupling. An interesting observation was the weak inhibition of Na⁺ entry by amiloride (by ~30%), compared with the strong inhibition (by up to 90%) of isosmolar fluid absorption in mouse lung (11). Active fluid absorption is a transcellular process in which the basolateral membrane Na⁺-K⁺ pump generates the electrochemical potential to drive Na⁺ (and Cl) transport. Thus active fluid transport involves transcellular Na⁺ transport through the same cell that generates the driving force. The strong amiloride inhibition of active fluid absorption in mouse lung suggests that ENaC is the principal Na⁺ channel involved in this process. In contrast, passive unidirectional Na⁺ transport occurs through all pathways, including different epithelial cell types and the paracellular pathway. The lesser inhibition of amiloride on unidirectional Na⁺ flux than active fluid absorption suggests that the major pathways for unidirectional Na⁺ flux are not amiloride sensitive. Although paracellular permeability is likely to be a major contributor to unidirectional Na⁺ flux in this model, the studies here do not quantify relative contributions from type I alveolar epithelial cells and various cells lining the airways.

The pleural surface fluorescence approach was applied to determine whether aquaporin deletion in mice causes upregulation of Na⁺ and Cl transporters to compensate for reduced osmotic water permeability. Salt transporter upregulation is a possible explanation of our observations that AQP1 and AQP5 deletion in lung does not impair active fluid absorption from the distal air spaces, even after maximizing the rate of fluid absorption by cAMP agonists and type II cell upregulation by keratinocyte growth factor (24). We reasoned that functional measurement of salt transport is superior to genomic or proteomic analysis because of the complex posttranslational processing and trafficking of the major lung salt transporters such as ENaC and CFTR and because of the incomplete knowledge of the transporters responsible for alveolar fluid clearance. The data here show that passive Na⁺ and Cl transport were unaffected by deletion of AQP1 or AQP5, despite an ~10-fold reduction in osmotically induced water permeability across the air space-capillary osmotic barrier. These results provide evidence against functionally significant upregulation of salt transport in distal lung in aquaporin knockout mice. The data support the conjecture that high lung water permeability and aquaporins are not needed for active fluid absorption because the rate of fluid absorption in lung is substantially lower than that in epithelia (25, 36, 37) where aquaporins were shown to be required.

Mutations in CFTR Cl channels cause the most common lethal genetic disease, cystic fibrosis. The mechanisms by which nonfunctional CFTR causes lung disease are not clear (31). Proposed consequences of CFTR mutation on airway physiology include abnormally low airway surface liquid (ASL) salt concentration, low ASL volume because of increased ENaC function, and low ASL pH because of defective HCO transport (5, 38). Our laboratory reported, using fluorescent indicators, that ASL salt concentration, pH, and osmolality are not different in cystic fibrosis (18, 19) but that submucosal gland viscosity is increased (17). Fang et al. (11) recently proposed a novel function of CFTR in distal air space function: impaired cAMP-dependent active fluid absorption. They found that cAMP-stimulated alveolar fluid clearance and air space fluid ³⁶Cl exit in wild-type mice were inhibited by glibenclamide. cAMP-stimulated alveolar fluid clearance and air space fluid ³⁶Cl exit were also reduced in cystic fibrosis (F508 CFTR mutant) mice. The data here support the conclusion that cAMP-stimulated, but not basal Cl, transport across the distal air space barrier is blocked by glibenclamide but suggest that the interpretation of Cl transport in cystic fibrosis mice is more complex. Although there was a significant approximately twofold reduction in cAMP-stimulated Cl influx in CFTR null mice compared with wild-type mice, in agreement with the involvement of CFTR as proposed by Fang et al. (11), we found cAMP-stimulated Cl transport in ~50% of the CFTR null mice. In addition to CFTR, our data suggest the presence of other cAMP-regulated Cl transporting pathway(s) in distal lung. The heterogeneity in Cl transport from mouse to mouse in CFTR null mice may represent effects of modifier genes and environmental or other factors on the expression of alternative cAMP-regulated Cl transporters. The challenge will be to identify the transporter(s) responsible for the cAMP-dependent Cl transport process defined functionally in our experiments.

In summary, the results here establish a pleural surface fluorescence method to measure continuously the transport of Cl and Na⁺ into the air space compartment in perfused mouse lungs. Compared with prior fluid sampling methods, our approach allows continuous noninvasive monitoring of distal air space fluid ionic content during perfusate solution changes. This methodology is technically simple and should be readily applicable to study lung salt transport in other transgenic mouse models and mammalian species and to study salt secretion in the perinatal lung.