help button home button
AJRCMB
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Published ahead of print on May 30, 2003, doi:10.1165/rcmb.2003-0109OC
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2003-0109OCv1
29/6/710    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Szkotak, A. J.
Right arrow Articles by Duszyk, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Szkotak, A. J.
Right arrow Articles by Duszyk, M.
American Journal of Respiratory Cell and Molecular Biology. Vol. 29, pp. 710-720, 2003
© 2003 American Thoracic Society
DOI: 10.1165/rcmb.2003-0109OC

The Role of the Basolateral Outwardly Rectifying Chloride Channel in Human Airway Epithelial Anion Secretion

Artur J. Szkotak, S. F. Paul Man and Marek Duszyk

Departments of Physiology and Medicine, University of Alberta, Edmonton, Alberta, Canada

Address correspondence to: Dr. Marek Duszyk, Department of Physiology, University of Alberta, 7-46 Medical Sciences Bldg., Edmonton, Alberta, T6G 2H7 Canada. E-mail: marek.duszyk{at}ualberta.ca


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The purpose of this study was to characterize basolateral anion channels in Calu-3 and normal human bronchial epithelial cells, and their role in anion secretion. Patch clamp studies identified an outwardly rectifying Cl- channel (ORCC), which could be activated by the adenosine receptor agonist 5'-(N-ethylcarboxamido)adenosine (NECA). Short-circuit current measurements revealed that NECA activates a basolateral, but not an apical, anion conductance sensitive to 4,4'-diisothiocyanatostilbene-2, 2'-disulfonic acid, and to 9-anthracenecarboxylic acid, but not to 4,4'-dinitrostilbene-2,2'-disulfonic acid. Apical membrane permeabilization studies confirmed the presence of basolateral anion channels, established their halide permeability sequence (Cl- >= Br- >> I-), and demonstrated their outwardly rectifying nature. Experiments using H-89, forskolin, and Ht31 demonstrated that adenosine receptor dependent activation of basolateral ORCC was cAMP- and potentially A-kinase anchoring protein–dependent. Neither BAPTA-AM treatment nor basolateral Ca2+ removal had any effect on the activation of these channels. Anion replacement and 36Cl- flux studies show that Calu-3 cells primarily secrete HCO3- when stimulated with NECA, and that Cl- secretion can be stimulated by blocking basolateral ORCC, whereas normal human bronchial epithelial cells exclusively secrete Cl- under all conditions studied. We propose a novel model of anion secretion in which ORCC recycles Cl- across the basolateral membrane, allowing preferential HCO3- secretion.

Abbreviations: A-kinase anchoring protein, AKAP • A1, A2A, A2B, A3, adenosine receptor subtypes • bronchial epithelial growth media, BEGM • intracellular Ca2+ concentration, [Ca2+]i • cystic fibrosis transmembrane conductance regulator, CFTR • glyceraldehyde-3-phosphate dehydrogenase, GAPDH • short-circuit current, Isc • 36Cl- apical to basolateral flux, JClAB • 36Cl- basolateral to apical flux, JClBA • 36Cl- net flux, JClnet • calculated net HCO3- flux, JHCO3net • Ca2+-dependent K+ channels, KCa • Krebs-Henseleit solution, KHS • normal human bronchial epithelia, NHBE • Na+-K+-2Cl-, NKCC • outwardly rectifying chloride channel, ORCC • closed probability, PC • protein kinase A, PKA • open probability, PO • transepithelial resistance, RT • reverse transcription polymerase chain reaction, RT-PCR • negative pipette potential, -VPIP


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Anion secretion by airway epithelial cells acts as a driving force for hydration of the airway surface liquid and is therefore a major determinant of mucociliary clearance. The transport pathways comprising the Cl- and HCO3- secretory mechanism have been extensively studied (1–3). Cl- enters the cell across the basolateral membrane via a Na+-K+-2Cl- (NKCC) cotransporter, which represents a secondary active transport process driven by electrochemical gradients established by the basolateral Na+/K+-ATPase. K+ ions are recycled via basolateral K+ channels and serve to adjust membrane potential, whereas vectorial secretion of Cl- is achieved by its exit through apically localized anion channels. HCO3- ions are thought to enter the cell via basolaterally localized Na+-HCO3- cotransporters, and to exit the cell through apical anion channels, which serve as a final common pathway for both Cl- and HCO3- transport. It is well established that the cystic fibrosis transmembrane conductance regulator (CFTR), the channel mutated in individuals with cystic fibrosis (CF), is the primary apical Cl- conductance. However, it has only recently been shown that this channel is also responsible for conducting HCO3- across the apical membrane of airway epithelia. Illek and coworkers (4) have shown that basolaterally permeabilized Calu-3 cells, a human model of serous submucosal airway epithelium, mounted in a HCO3- gradient will conduct this anion via CFTR. Similarly, Paradiso and colleagues (5) used pH-sensitive dyes to show that human nasal epithelia will acidify/alkalinize across their apical membranes via CFTR. This is further supported by the finding that the apical membranes of these cells do not display anion exchange activity, which could account for apical HCO3- translocation (5, 6).

The finding that CFTR conducts Cl- and HCO3- poses new questions. In particular, it is unclear how cells control which anion is secreted at any given time. It has been suggested that opening of basolateral K+ channels, which causes cell hyperpolarization, elicits a switch from HCO3- to Cl- secretion (3, 7). Hyperpolarization inhibits basolateral Na+-HCO3- cotransporters, and therefore HCO3- secretion, while promoting Cl- secretion by increasing the driving force for apical exit of this anion. Thus, opening of basolateral K+ channels could explain a switch from HCO3- to Cl- secretion. However, the mechanism by which switching in the opposite direction occurs, from Cl- to HCO3- secretion, remains unknown. Presumably, reversing the logic and closing K+ channels would not suffice in switching secretion from Cl- to HCO3- because this would depolarize cells and therefore inhibit secretion all together. Furthermore, the mechanism by which HCO3- secretion can dominate Cl- secretion at any given moment is unknown, particularly considering that Cl- is both more abundant than HCO3- and 4 to 10 times more permeable through CFTR channels (8, 9).

Before the cloning of CFTR, the outwardly rectifying Cl- channel (ORCC) was suspected of being the primary defect in CF when it was shown that this channel could be activated by cAMP-dependent mechanisms in normal, but not in CF, airway epithelia (10, 11). After the cloning of CFTR, it was shown that CFTR regulates ORCC such that in CF ORCC is unresponsive to cAMP-dependent stimuli (12). Schwiebert and coworkers (13) proposed a mechanism for this regulation, suggesting that CFTR mediates the autocrine release of ATP, which stimulates purinergic receptors and leads to activation of the ORCC. More recently, Xia and colleagues (14) used single-channel patch clamp studies of Calu-3 cells to demonstrate that ORCCs are rarely found in the apical membranes of confluent layers, but are abundant in nonpolarized dispersed cells, thus suggesting that these channels could be localized to the basolateral membrane.

Despite the ongoing interest in the ORCC, few anion transport models have proposed a clear physiologic role for this channel in airway epithelial cells. Willumsen and associates (1) suggested that these channels contribute to Cl- secretion by providing an additional pathway for Cl- entry across the basolateral membrane, thereby supplementing the electrically silent cotransport system. However, this does not explain how vectorial Cl- transport could be achieved, under short-circuit conditions, if both the apical and basolateral membrane transport processes are passive. Other studies suggest that the presence of a basolateral ORCC may facilitate cAMP-dependent Cl- absorption across airway epithelium, under open-circuit conditions (15).

In previous studies we have shown that adenosine A2 receptor stimulation leads to the activation of CFTR Cl- channels in Calu-3 cells (16). During the course of those experiments we observed that anion channels other than CFTR were frequently activated by adenosine receptor stimulation. Therefore, the purpose of the present study was to characterize the role of these channels in transepithelial anion secretion in Calu-3 cells, and in primary cultures of normal human bronchial epithelium (NHBE). The results of our study show that adenosine not only activates CFTR, but also basolateral ORCCs, via a process that is cAMP-dependent and may involve A kinase anchoring proteins (AKAPs). The fact that inhibition of these channels increases Cl- secretion indicates that at least some of the Cl- ions that enter the cell via NKCC cotransport are recycled across the basolateral membrane. Furthermore, we propose a physiologic role for basolateral Cl- channels as a molecular switch that controls the type of anion secreted. Opening of basolateral Cl- channels allows for Cl- recycling across the basolateral membrane and therefore preferential HCO3- secretion, whereas closing of these channels favors Cl- secretion.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell Culture
Calu-3 cells were obtained from the American Type Culture Collection (Rockville, MD), and grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 5 µg/ml gentamycin sulfate, 6 µg/ml penicillin-G, and 10 µg/ml streptomycin. The cells were maintained in T75 tissue-culture flasks (Costar, Cambridge, MA), and typically required 6–8 d to reach ~ 85% confluence. At this time the cells were passaged using saline solution containing 0.05% trypsin and 0.02% EDTA. Cells were seeded at a density of 3.5 x 105 cells/cm2 onto Costar Snapwell inserts (0.45-µm pore size, 1 cm2 surface area) for short-circuit current (Isc) measurements, and maintained in medium which differed only in that it contained 20% fetal bovine serum. For the first 6 d, cells were grown submerged in culture medium. Subsequently, air interface culturing was used, in which the medium was added only to the basolateral side of the inserts. Inserts were used for experiments 10–16 d after the establishment of an air interface. For patch clamp studies, 1 x 105 cells were seeded onto 15-mm coverslips (Fisherbrand, Pittsburgh, PA) 24 h before experiments. All cells used for experiments were at passage 24–27.

NHBE cells were obtained from BioWhittaker (San Diego, CA) as frozen passage 1 stocks, containing >= 500,000 cells, and were cultured as described previously (17). The cells were thawed according to the instructions provided and seeded into T75 tissue culture flasks. They were grown in bronchial epithelial growth media (BEGM; BioWhittaker) supplemented with bovine pituitary extract (52 µg/ml), hydrocortisone (0.5 µg/ml), human recombinant epidermal growth factor (0.5 ng/ml), epinephrine (0.5 µg/ml), transferrin (10 µg/ml), insulin (5 µg/ml), triiodothyronine (6.5 ng/ml), gentamicin (50 µg/ml), amphotericin-B (50 ng/ml), penicillin G (6.2 µg/ml), streptomycin (10 µg/ml), and retinoic acid (330 nM). Retinoic acid (all-trans) was obtained from Sigma (St. Louis, MO), prepared in aliquots that were stored at -70°C, and was added to aliquots of media just before use. Care was taken to shield cells and media from light, particularly after supplementing with retinoic acid. When the cells reached 80% confluence they were passaged with trypsin/EDTA (BioWhittaker). A portion of the harvested cells was seeded into a new T75 flask at 3,500 cells/cm2 at passage 2, whereas the rest were frozen at >= 500,000 cells/cryovial in 10% dimethyl sulfoxide (DMSO), for use later. Once 80% confluent the cells were passaged again, but now onto Costar Snapwell inserts (0.45-µM pore size, 1 cm2 surface area) at 1 x 105 cells/insert, at passage 3. Cells plated onto inserts were maintained in media containing a 1:1 (vol/vol) mixture of BEGM:Dulbecco's modified Eagle's medium supplemented as above except that amphotericin-B and triiodothyronine were excluded, whereas retinoic acid was used at a reduced concentration (50 nM). The cells were grown submerged for the first 7 d, followed by air interface culturing for an additional 14–20 d. All NHBE cells used for experiments were at passage 3.

All cultures of both Calu-3 and NHBE cells were incubated at 37°C in a humidified atmosphere of 5% CO2 in air. Media was changed the day after seeding and subsequently every 2–3 d, until the desired confluence was reached.

Cell-Attached Patch Clamp
Cells, cultured on 15-mm round coverslips, were rinsed three times in bath solution, containing (in mM): 160 Tris Cl, 30 sucrose (pH 7.0). They were then mounted into an open bath chamber (Warner Instruments Inc., Hamden, CT), maintained at 37°C and fixed to the stage of an Olympus IMT-2 Inverted Research Microscope (Lake Success, NY). Pipette electrodes were made from standard borosilicate glass (Sutter Instrument Co., Novato, CA) using a two-stage vertical puller (Narishige, Japan), fire polished to a final resistance of 8–20 M{Omega}, and back-filled with bath solution. After pipette immersion in bath solution, offset potentials were compensated and a G{Omega} seal was formed. Currents were recorded, using an Axopatch 200A amplifier and Clampex 8.0 software, both from Axon Instruments (Foster City, CA). Continuous recordings of channel activity, sampled at 5 kHz, were made when voltages were clamped from -80 mV to 80 mV in 20-mV increments. All currents were reported with reference to zero in the bath and data were analyzed by pClamp 8.0 (Axon Instruments) and Microcal Origin 6.0 (Northampton, MA) software. Recordings were filtered at 100 Hz, using an 8-pole Bessel filter; current-amplitude histograms were made and fit with Gaussian functions. The closed probability (PC) was calculated as the proportion of the area under the curve that corresponded to the state in which all channels were closed. The open probability (PO) was then calculated, for patches that contained one or more non-CFTR channels, as the total area under the curve minus the closed probability (1 - PC). Voltage is expressed as the negative pipette potential (-VPIP).

Transepithelial Measurements
Standard techniques were used in Ussing chamber studies. Cells grown on inserts were bathed on apical and basolateral sides with 10 ml of Krebs-Henseleit Solution (KHS), or a modified version thereof (Table 1). Solutions were warmed to 37°C and continually circulated with a gas lift using either 95% O2–5% CO2 if the solution was HCO3- buffered, or 100% O2 if the solution was HEPES buffered. Chemicals were added from concentrated stock solutions, and both chambers were continuously and separately perfused to ensure proper oxygenation and stirring. The transepithelial potential difference was clamped using a DVC 1,000 voltage/current amplifier (WPI, Sarasota, FL), and the resulting current was recorded through Ag-AgCl electrodes and 3 M KCl agar bridges. In most cases the voltage was clamped to zero and the resultant Isc was sampled at 10 Hz using a PowerLab 8SP series data acquisition converter and Chart software, both from ADInstruments (Castle Hill, NSW, Australia). Brief (1 s) pulses to 0.5 mV were applied every 90 s, in order to calculate resistances. All values are expressed as an average {Delta}Isc, which was calculated as the mean change in current in the first 300 s after drug addition, unless otherwise noted. During all experiments, the Isc was allowed to stabilize for 20 min before treatments, and all experiments were performed in the presence of 10 µM apical amiloride.


View this table:
[in this window]
[in a new window]
  		TABLE 1 Composition of solutions used in transepithelial measurements (mM)

 
When clamping at variable potentials was necessary, experiments were performed in the same way as described, with a few exceptions. The data were acquired at 100 Hz using the PowerLab 8SP series converter in conjunction with Scope software (ADInstruments); cells were clamped from -80 to 80 mV in 20-mV increments with each step lasting 1.5 s, and with brief (0.5 s) periods between each step at 0 mV. The 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid (DIDS)-sensitive current was obtained by subtracting post-DIDS traces, from pre-DIDS traces. Positive currents were defined as anion secretion or movement from the basolateral to the apical side, and current–voltage relationships were obtained from the peak currents.

Radioisotopic Flux
36Cl- flux assays were carried out on Calu-3 inserts being used for simultaneous Isc experiments in KHS. Inserts were short-circuited for 20 min before beforethe addition of the radioisotope. At time zero (T0), samples for determination of background radioactivity were taken, followed by the addition of 3 µCi of 36Cl- (Amersham Pharmacia Biotech, Little Chalfont, England) to the basolateral compartment, and another 20 min allowed for the establishment of equilibrium. At this time (T20) 0.5 ml samples were taken from the apical side and replaced with fresh KHS; this was repeated at every 10-min interval thereafter. Three samples were taken (T20–T40) before the addition of bilateral 5'-(N-ethylcarboxamido)adenosine (NECA) (10 µM); this was followed by two more samples taken (T50 and T60) before the addition of basolateral DIDS (50 µM), and a further two more samples taken (T70 and T80) after DIDS treatment. In addition, two samples were taken from the basolateral side, immediately before treatment with NECA, in order to calculate the specific activity. Samples were counted for radioactivity using a model 1219 Rackbeta (LKB, Turku, Finland) liquid scintillation counter. The unidirectional flux (JClBA) was calculated according to standard equations (18). 36Cl- fluxes in the apical to basolateral (JClAB) direction were measured in exactly the same fashion, except that the radioisotope was added to the apical bathing solution. Net 36Cl- flux (JClnet) was calculated as JClnet = JClBA - JClAB.

Isc data represent the secretion of Cl- and HCO3-, assuming that Na+ absorption is negligible in the presence of amiloride (apical, 10 µM), and K+ secretion is insignificant in Calu-3 cells (3). Therefore, net HCO3- flux (JHCO3net) can be calculated as JHCO3net = Isc - JClnet, where Isc is the average short-circuit current expressed in flux units (µEq/cm2·h) over the 10-min period during which flux was assayed.

Chemicals
Stock solutions were prepared in H2O for DIDS (5 mM), 4,4'-dinitrostilbene-2,2'-disulfonic acid (DNDS, 100 mM), 2-p-(2-carboxyethyl)phenethylamino-5'-(N-ethylcarboxamido) adenosine (CGS-21680, 1 mM), carbachol (100 mM), and amiloride (10 mM). Stock solutions of 9-anthracenecarboxylic acid (9-AC) were in 0.1 N NaOH (10 mM), NECA in 0.1 N HCl (10 mM), furosemide in 0.1 N NaOH (100 mM), forskolin in 95% ethanol (10 mM), N-(2-[p-bromocinnamylamino]ethyl)-5-isoquinolinesulfonamide (H-89) in 50% ethanol (5 mM), ionomycin in 95% ethanol (10 mM), nystatin in DMSO (90 mg/ml), 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid tetrakis acetoxymethyl ester (BAPTA-AM) in DMSO (30 mM), and ouabain in KHS (10 mM), or a modified version thereof (Table 1). DIDS and DNDS were purchased from Molecular Probes Inc. (Eugene, OR), forskolin from Calbiochem (San Diego, CA), and all other chemicals were from Sigma. InCELLect stearated-Ht31 (St-Ht31) peptide and its analog St-Ht31P, a modified version of St-Ht31 in which two isoleucine residues are replaced with prolines, was obtained as a 10 mM stock in 50 mM Tris HCl (pH = 7.5) from Promega (Madison, WI).

Data Analysis
Data are presented as means ± SEM, unless otherwise indicated; n refers to the number of experiments. The paired Student's t test was used to compare the means of two groups. Statistically significant differences among the means of multiple groups were determined by one-way analysis of variance (ANOVA) with the Tukey-Kramer post-test using Graphpad Instat 3.05 software (San Diego, CA). A value of P < 0.05 was considered statistically significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Adenosine Receptor Stimulation Activates ORCC
Figure 1A shows representative traces of cell-attached patch clamp current recordings obtained before and after adenosine receptor agonist treatment (NECA, 10 µM). In cell-attached patches the channels were usually closed, but could be activated by NECA in 40% of patches (n = 10). Figure 1B shows the current–voltage relationship obtained at voltages ranging from -80 to 80 mV. Activity was greatest at depolarizing potentials where the open probability was PO = 0.3 ± 0.1 and conductances were between 115 and 140 pS (n = 4). These channels have been characterized previously in Calu-3 cells (14, 19) and are known to be DIDS-sensitive (20), but the finding that they are activated by adenosine receptor stimulation is novel.



View larger version (20K):
[in this window]
[in a new window]
  		Figure 1. Adenosine-receptor mediated activation of an ORCC in Calu-3 cells. (A) Typical traces, obtained in 160 mM Tris Cl, 30 mM sucrose solution, showing ORCC activity in a cell-attached patch before and after NECA (10 µM) treatment. The recordings are representative of 10 experiments, 4 of which displayed channel activity, and the channel closed (C) and open (O) states are shown. (B) The current–voltage relationship corresponds to the experiment shown in A after NECA treatment. Values are expressed as means ± SD.

 
Basolaterally Localized Anion Channels
Further characterization of ORCC and their regulation by adenosine receptors was performed in transepithelial Isc studies. In total, we evaluated 134 inserts with Calu-3 cells and 65 with NHBE cells. The basal Isc and RT in Calu-3 cells in normal KHS was 23.1 ± 1.1 µA/cm2 and 210 ± 15 {Omega}cm2 (n = 59), respectively. Similarly, the basal Isc and RT in NHBE cells was 10.2 ± 0.8 µA/cm2 and 811 ± 58 {Omega}cm2 (n = 28), respectively. In all experiments the cells were pretreated with apical amiloride (10 µM), which had no effect on Calu-3 cells, but reduced baseline Isc in NHBE cells to 2.9 ± 0.2 µA/cm2 (n = 28).

Adenosine receptor stimulation with bilateral NECA (10 µM) induced a biphasic Isc response, with an initial peak followed by a plateau (Figure 2). Overall, NECA increased Isc by 40.5 ± 2.7 µA/cm2 (n = 10) in Calu-3 cells and by 20.9 ± 0.9 µA/cm2 (n = 3) in NHBE cells. The NECA response was abolished in the presence of a CFTR Cl- channel blocker, DPC (1 mM, apical), in both cell types (n = 4, P<0.005). Apical DIDS (50 µM) had no effect on either baseline or NECA-stimulated Isc. However, basolateral DIDS caused a small but significant increase in baseline Isc in Calu-3 cells (1.4 ± 0.1 µA/cm2, n = 3, P < 0.05, Figure 2B) and a much larger increase in cells pre-treated with NECA (15.6 ± 2.0 µA/cm2, n = 8, Figure 2A). Similar results were obtained in NHBE cells, where DIDS only had an effect when applied basolaterally after NECA pretreatment (4.2 ± 0.4 µA/cm2, n = 3). Interestingly, in the presence of basolateral DIDS, the NECA response was potentiated to 57.5 ± 2 µA/cm2 (n = 3, P < 0.01, Figure 2B) and 30.1 ± 2.9 µA/cm2 (n = 4, P < 0.05) in Calu-3 and NHBE cells, respectively, bearing in mind that effects were quantified as the average Isc in the first 5 min of the drug response. This suggests that basolateral DIDS-sensitive anion channels are stimulated through adenosine signaling in both cell types. However, it is difficult to gauge the magnitude of basolateral anion channel inhibition on Isc, because DIDS is also known to block electrogenic Na+-HCO3- cotransporters, and electroneutral Cl-/HCO3- exchangers.



View larger version (13K):
[in this window]
[in a new window]
  		Figure 2. The effect of NECA and basolateral anion channel blockers on Isc in Calu-3 cells. Transepithelial Isc measurements were performed in KHS in the presence of apical amiloride (10 µM). (A) Representative recording (n = 8) showing the activation of Isc by bilateral NECA (10 µM), and subsequent basolateral DIDS (50 µM). (B) The effect of basolateral DIDS on baseline Isc, and potentiation of the subsequent NECA response (n = 3). (C) Basolateral 9-AC (1 mM) application stimulates Isc to a similar extent as basolateral DIDS, in the presence of NECA (n = 3). (D) Basolateral DNDS (3 mM) inhibits Isc, when applied after NECA; the subsequent basolateral DIDS response is potentiated (n = 3).

 
In an attempt to isolate the effect of basolateral anion channel inhibition, we used the alternative anion channel blockers 9-AC and DNDS. In Calu-3 cells pretreated with NECA, the application of basolateral 9-AC (1 mM) stimulated Isc by 18.0 ± 0.9 µA/cm2 (n = 3, Figure 2C). This effect was similar to that produced by DIDS (P > 0.05). However, other studies indicate that this high concentration of 9-AC is below the IC50 for ORCC inhibition, and thus likely does not represent complete channel blockage (11). The application of basolateral DNDS (3 mM), in the presence of NECA, inhibited Isc by 12.8 ± 1.6 µA/cm2 (n = 3, Figure 2D), indicating that this drug does not block basolateral anion channels. Because both DNDS and DIDS inhibit Na+-HCO3- cotransporters (and Cl-/HCO3- exchangers), but only one appears to inhibit basolateral anion channels, the sequential application of these drugs allows us to isolate the contributions of each of these factors to the Isc. In other words, when DIDS is applied after DNDS, the Isc response reflects the inhibition of basolateral anion channels in isolation. Accordingly, the observed increase in Isc induced by DIDS, in the presence of DNDS, is nearly doubled to 29.7 ± 1.1 µA/cm2 (n = 3, P < 0.005, Figure 2D), when compared to the application of DIDS alone.

The Basolateral Membrane Contains ORCC
Apical membrane permeabilization experiments were performed to further characterize the basolateral anion channels. These experiments were done in the presence of basolateral ouabain (1 mM), to inhibit the electrogenic Na+/K+-ATPase. Basolateral Na+-K+-2Cl- and Na+-HCO3- cotransporters were also inhibited, by furosemide (1 mM) and HCO3- removal, respectively. An apical (HCO3- Free KHS) to basolateral (HCO3- Free, Low Cl- solution) Cl- gradient (129.1:17.1 mM) was established before permeabilization (Table 1). Under these conditions, apical nystatin (90 µg/ml) increased Isc, in Calu-3 cells, by -102 ± 13 µA/cm2 (n = 4, Figure 3A) and -163 ± 20 µA/cm2 (n = 3, Figure 3B) in the absence and presence of NECA, respectively (P < 0.05). Furthermore, DIDS inhibited Isc by 30.9 ± 1.0 µA/cm2 (n = 4, Figure 3A) and 54.6 ± 10.9 µA/cm2 (n = 3, Figure 3B) in the absence and presence of NECA, respectively (P < 0.05). Similar results were obtained in NHBE cells where nystatin stimulated Isc was larger in NECA pretreated cells (-62.1 ± 2.6 µA/cm2, n = 4), than in nontreated cells (-45.5 ± 0.2 µA/cm2, n = 4, P < 0.01). DIDS reduced Isc by 14.1 ± 0.7 µA/cm2 and 7.6 ± 0.3 µA/cm2 in the presence and absence of NECA, respectively (n = 4, P < 0.001).



View larger version (13K):
[in this window]
[in a new window]
  		Figure 3. Permeabilization of the apical membrane with nystatin in Calu-3 cells reveals a basolateral Cl- conductance that can be stimulated by NECA and is inhibited by DIDS. Isc measurements were performed in the presence of an apical (HCO3--free KHS, Table 1) to basolateral (HCO3--free, Low Cl- solution, Table 1) Cl- gradient (129.1 mM:17.1 mM). Cells were pretreated with basolateral ouabain (1 mM), to inhibit the Na+/K+-ATPase, and basolateral furosemide (1 mM) to inhibit Na+-K+-2Cl- cotransport. Representative recordings are shown from control (A, n = 4) and NECA-treated (B, bilateral, 10 µM, n = 3) monolayers exposed to apical nystatin (90 µg/ml) and subsequent basolateral DIDS (50 µM).

 
Nystatin permeabilization experiments were also used to determine the relative halide permeabilities of the basolateral ORCC in Calu-3 cells. Under conditions in which an equimolar Br- gradient replaced the Cl- gradient (Table 1), the {Delta}Isc responses to nystatin and DIDS were reduced, though not significantly (P > 0.05), to -97.8 ± 3.8 µA/cm2 and 28.1 ± 1.1 µA/cm2, respectively (n = 4). When Cl- ions were replaced by I- ions, the {Delta}Isc response to nystatin was again insignificantly reduced (-84.8 ± 7.4 µA/cm2, n = 4, P > 0.05), but the subsequent response to DIDS was reduced to 3.3 ± 0.3 µA/cm2 (n = 4, P > 0.0001). Based on the DIDS-sensitive inhibition of the basolateral anion channels in Calu-3 cells, their halide permeability sequence is: Cl- (1.00) >= Br- (0.90) >> I- (0.11).

Apical membrane permeabilization experiments were further used to investigate the current-voltage relationship of Cl- permeation across the basolateral membrane of NHBE cells. Monolayers were bathed in symmetrical HCO3--free choline solution (Table 1), and were clamped from -80 to 80 mV in 20-mV increments. The recordings, in Figure 4, show DIDS-sensitive current under control (Figure 4A) and NECA-treated conditions (n = 3, Figure 4B). The current–voltage relationships, in Figure 4C, show that the DIDS-sensitive Cl- current displays strong outward rectification, and is significantly stimulated by NECA (P < 0.05). Similar experiments in Calu-3 cells were not possible as transepithelial resistances proved to be insufficient.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 4. Permeabilization of the apical membrane with nystatin in NHBE cells reveals the presence of DIDS-sensitive ORCC in the basolateral membrane. Current measurements were performed in symmetrical HCO3--free choline Cl- solution (Table 1), in cells permeabilized with apical nystatin (90 µg/ml), and clamped from -80 to 80 mV in 20-mV increments. Once currents had stabilized after nystatin permeabilization, basolateral DIDS (50 µM) was applied and further recordings were made. Representative DIDS-sensitive current recordings from control (A) and NECA (bilateral, 10 µM, B)-treated monolayers (n = 3) are shown. (C) Current–voltage relationships for traces shown in A.

 
Adenosine Signaling to Basolateral Anion Channels
Having established the localization and characteristics of the DIDS-sensitive anion channels activated by adenosine receptor stimulation, we investigated the signaling mechanisms involved. Our previous studies had demonstrated adenosine receptor expression and localization in Calu-3 cells (16), so the present study has only supplemented that information with data from NHBE cells. RT-PCR experiments revealed that NHBE cells express mRNA for A2A and A2B but not A1 or A3 receptors (Figure E1 in the online supplement). Adenosine receptor function and localization to either apical or basolateral membranes was characterized in transepithelial Isc measurements in KHS (Figure E2 in the online supplement). These results show that adenosine-dependent Isc can be nearly maximally stimulated by basolateral A2B receptor stimulation, approximately half maximally stimulated by apical A2B receptor stimulation and only slightly stimulated by basolateral A2A receptor stimulation. Overall, this pattern is similar to that seen in Calu-3 cells.

Previous studies have shown that adenosine signaling is cAMP-dependent in human airway epithelium (21). We have therefore investigated if cAMP-dependent signaling is responsible for ORCC activation in KHS. When Calu-3 cells are pretreated with the PKA inhibitor H-89 (bilateral, 10 µM) and NECA (bilateral, 10 µM), the Isc response to DIDS (basolateral, 50 µM) is reduced by 80% to 2.7 ± 0.2 µA/cm2 (n = 3, P < 0.0005; Figure 5A), when compared with experiments in which there is no H-89 pretreatment. Furthermore, forskolin (10 µM, bilateral), an adenylyl cyclase activator, stimulated a basolaterally DIDS-sensitive Isc that was similar to that activated by NECA (n = 4, data not shown).



View larger version (22K):
[in this window]
[in a new window]
  		Figure 5. Adenosine receptor–mediated activation of basolateral anion channels is PKA- and AKAP-dependent, but Ca2+-independent in Calu-3 cells. Experiments were performed on monolayers pre-treated with apical amiloride (10 µM). (A) In cells bathed in bilateral KHS and pretreated with the PKA inhibitor, H-89 (bilateral, 10 µM), the bilateral NECA (10 µM), and basolateral DIDS (50 µM) responses were greatly reduced (n = 3). (B) Calu-3 monolayers, bathed in bilateral KHS, were pretreated with either the AKAP-neutralizing peptide St-Ht31 (bilateral, 10 µM), or an equimolar amount of AKAP-inactive St-Ht31P as a negative control. Isc experiments were then performed by adding bilateral NECA (10 µM, not shown), followed by basolateral DIDS (50 µM). The averages of three of each experiment are shown. The values expressed are means ± SEM. (C) In cells bathed with bilateral KHS and pretreated with bilateral NECA (10 µM) and basolateral DIDS (50 µM), there is a significant response to subsequent application of basolateral ionomycin (1 µM). (D) In cells bathed with apical KHS and basolateral Ca2+-free KHS, the responses to bilateral NECA (10 µM) and basolateral DIDS (50 µM) are unchanged, but application of basolateral ionomycin (1 µM) has no effect; subsequent application of CaCl2 (2.5 mM, basolateral) stimulates Isc.

 
It has previously been shown that cAMP-dependent CFTR activation by adenosine receptors is AKAP-dependent (22). To investigate whether the same is true of basolateral Cl- channel activation, we used the stearated cell permeable form of the AKAP-inhibiting peptide St-Ht31 and its AKAP-inactive analog St-Ht31P, as a control. Calu-3 cells, in KHS, were pretreated with either St-Ht31 (bilateral, 10 µM) or equimolar St-Ht31P, followed by NECA (bilateral, 10 µM). Under these conditions, the response to DIDS was reduced by 25% (P < 0.02) in cells pretreated with St-Ht31 (n = 3) when compared with those pretreated with St-Ht31P (n = 3; Figure 5B). Therefore, these experiments further support the involvement of cAMP in the activation of basolateral Cl- channels and indicate that this activation may be partially AKAP-dependent, though it is possible that disruption of AKAPs inhibits the DIDS response by subverting other channel activities.

Previous studies suggested that basolateral DIDS activates Isc by stimulating an increase in [Ca2+]i (23, 24). We have therefore performed numerous experiments to test this hypothesis. First, we used the intracellular Ca2+ chelator BAPTA-AM in Isc measurements. We found that application of BAPTA-AM (bilateral, 30 µM) to Calu-3 cells bathed in KHS induced a small decrease in Isc of 2.4 ± 0.1 µA/cm2 (n = 6), which recovered over the next 30 min. There were no significant changes in the subsequent Isc responses to either NECA (41.6 ± 1.4 µA/cm2, n = 3) or DIDS (18.7 ± 0.2 µA/cm2, n = 3) treatment. Second, we performed Isc measurements on Calu-3 monolayers bathed in apical KHS, and basolateral Ca2+-free KHS (Table 1). It was previously shown that removal of basolateral Ca2+ does not disturb tight junctions, and allows for investigation of Ca2+-independent responses (23–25). Indeed, in Calu-3 cells tight junctions remained intact, as measured by RT, and allowed for measurement of Isc (Figure 5D). In control experiments performed in bilateral KHS, the application of basolateral ionomycin (1 µM) induced a marked increase in Isc (24.9 ± 5.7 µA/cm2, n = 4), when applied after NECA and DIDS (Figure 5C). In Calu-3 cells bathed with apical KHS and basolateral Ca2+ free KHS, the responses to NECA (31.8 ± 7.3 µA/cm2) and DIDS (14.2 ± 3.2 µA/cm2) were unchanged (P > 0.05), whereas the basolateral ionomycin response was completely absent until basolateral CaCl2 was applied (2.5 mM, 21.1 ± 4.8 µA/cm2, n = 3; Figure 5D). The carbachol response was also greatly inhibited by basolateral Ca2+ removal, from 151.3 ± 8.4 µA/cm2 (n = 6) to 16.6 ± 3.2 µA/cm2 (n = 3, P < 0.0001). Experiments performed in bilateral Ca2+-free KHS resulted in greatly reduced RT and therefore unreliable Isc, though even then small but consistent and proportional responses to NECA and DIDS were observed (data not shown). Thus, our experiments show that the NECA and DIDS responses are Ca2+-independent.

Basolateral ORCC and Anion Secretion
Figure 6 shows typical Isc responses to NECA and DIDS in HCO3--free and Cl--free KHS in Calu-3 cells (Table 1). When HCO3- is removed from KHS, Calu-3 baseline Isc decreases by 60% to 9.6 ± 1.0 µA/cm2 (n = 8, P < 0.01), whereas in NHBE cells it does not change. Under these conditions the response to NECA in Calu-3 cells is not sustained. A robust initial peak in Isc is followed by a return to baseline (Figure 6A), resulting in an average increase of only 3.4 ± 0.8 µA/cm2 (n = 4). However, subsequent application of basolateral DIDS still induces a marked increase in Isc of 9.4 ± 0.7 µA/cm2 (n = 4; Figure 6A). When the order of drug addition was reversed (Figure 6B), basolateral DIDS increased Isc by 2.0 ± 0.1 µA/cm2 (n = 4) and subsequent NECA application resulted in Isc activation by 6.3 ± 0.8 µA/cm2 (n = 4), an effect that, despite the lack of an initial peak, was significantly greater then the response to NECA in the absence of DIDS (P < 0.05). Lack of HCO3- ions in KHS did not affect Isc responses in NHBE cells. NECA stimulated Isc by 23.2 ± 1.4 µA/cm2 (n = 4) and subsequent DIDS application stimulated Isc by 3.9 ± 0.6 µA/cm2 (n = 4), whereas DIDS did not affect baseline Isc, and in its presence NECA induced an increase of 32.0 ± 2.9 µA/cm2 (n = 4).



View larger version (14K):
[in this window]
[in a new window]
  		Figure 6. The effect of NECA and DIDS on Isc in Calu-3 cells in HCO3-- and Cl--free solutions. All monolayers were pretreated with apical amiloride (10 µM). Representative recordings are shown from experiments performed by either applying bilateral NECA (10 µM) followed by basolateral DIDS (50 µM) (A, C, n = 4) or by applying basolateral DIDS followed by bilateral NECA (B, D, n = 3).

 
When Cl- is removed from KHS, baseline Isc in Calu-3 cells is 16.7 ± 0.8 µA/cm2 (n = 3) while NECA stimulates the current by 27.4 ± 1.8 µA/cm2 (n = 3; Figure 6C). Both of these values account for ~ 70% of the Isc in normal KHS. This result is consistent with HCO3--free KHS experiments, which indicate that these cells primarily secrete HCO3-. Subsequent application of basolateral DIDS inhibits the Isc by 10.4 ± 0.5 µA/cm2 (n = 3), likely due to its effects on basolateral Na+-HCO3- cotransport. If DIDS is applied first in Calu-3 cells, it induces a small but significant rise in Isc of 1.3 ± 0.1 µA/cm2 (n = 3, Figure 6D). Subsequent application of NECA results in a reduced Isc increase of 14.4 ± 0.7 µA/cm2 (n = 3), consistent with the notion that DIDS inhibits basolateral Na+-HCO3- cotransport. In NHBE cells, Cl- removal almost completely eliminates responses to adenosine receptor stimulation; NECA induces meager Isc increases of 2.1 ± 0.7 µA/cm2 (n = 3) and 2.0 ± 0.5 µA/cm2 (n = 4), in the absence and presence of basolateral DIDS, respectively.

To further distinguish between Cl- and HCO3- secretion, we have performed simultaneous Isc and radioisotopic flux measurements on Calu-3 cells in KHS. Figure 7A shows the averaged Isc traces (n = 8; the thickness of the line represents the SEM) and the sample intervals used for counting (T40–T80). Figure 7B compares the net 36Cl- flux rate (JClnet) and the calculated net HCO3- flux rate (JHCO3net) during each sample interval. This information is supplemented with measured unilateral 36Cl- flux data in Table 2; fluxes were measured n = 4 times in each direction. These data confirm that the majority of baseline Isc is HCO3- secretion (T40), though the contribution of Cl- to baseline varies greatly. Consistent with the findings from HCO3--free experiments (Figure 6A), the application of bilateral NECA (10 µM) stimulates transient Cl- secretion (Figure 7, T50) followed by sustained HCO3- secretion (Figure 7, T60). Unilateral flux data (Table 2) shows that the switch between Cl- and HCO3- secretion occurs due to an initial stimulation of JClBA followed by an eventual "catch-up" by JClAB. Subsequent application of basolateral DIDS (Figure 7, T70 and T80) stimulates JClnet, by decreasing JClAB, and reduces calculated HCO3- flux, likely by inhibiting Na+-HCO3- co-transport.



View larger version (27K):
[in this window]
[in a new window]
  		Figure 7. Simultaneous 36Cl- flux and Isc measurements demonstrate the relative contributions of HCO3- and Cl- to Calu-3 anion secretion. Experiments were performed in KHS on monolayers pretreated with apical amiloride (10 µM). (A) The averaged transepithelial Isc from the n = 8 experiments during which radioisotopic flux measurements were performed. The data are expressed as a mean ± SEM, where the thickness of the trace represents the SEM. The sampling periods for radioisotopic flux measurements, designated TX (where X is the time in minutes), are shown in relation to the points at which bilateral NECA (10 µM) and basolateral DIDS (50 µM) were added. (B) Measured 36Cl- net flux (JClnet = JClBA - JClAB) and calculated HCO3- flux (JHCO3net = Isc µ JClnet) are shown for each sampling period. Unilateral fluxes and other relevant data are summarized in Table 2. All values are expressed as means ± SEM.

 

View this table:
[in this window]
[in a new window]
  		TABLE 2 Unidirectional and net ion fluxes across short-circuited Calu-3 monolayers

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Our study shows that human airway epithelial cells possess basolateral Cl- channels that are activated by A2 adenosine receptors in a cAMP-dependent manner. Cell-attached patch-clamp experiments identified an ORCC that was activated by adenosine receptor stimulation in Calu-3 cells. The channel properties, particularly its current–voltage relationship, were similar to those described previously (10, 11, 14). Similarly, Zhang and coworkers (26) have previously reported the presence of a cAMP-dependent outwardly rectifying DIDS-sensitive Cl- conductance in patch-clamp studies of NHBE cells. Our transepithelial Isc measurements failed to detect any DIDS-sensitive anion channel activity in the apical membranes of either Calu-3 or NHBE cells, under control or adenosine receptor stimulated conditions. However, similar experiments demonstrated the presence of DIDS-sensitive adenosine receptor activated anion channel activity in the basolateral membranes of both cell types. These channels were 9-AC–sensitive, but DNDS-insensitive, and this finding was corroborated with apical membrane nystatin permeabilization studies, which also showed that DIDS-sensitive basolateral Cl- current is outwardly rectifying in NHBE cells, and that this current can be stimulated by adenosine receptor activation. Similarly, Hwang and colleagues (20) have previously shown that the basolateral Cl- conductance, in primary cultures of rat airway epithelium, is outwardly rectifying and cAMP-dependent. Thus there is strong evidence for the presence of basolateral ORCC in airway epithelial cells.

The halide permeability of the basolateral ORCC in Calu-3 cells (Cl- >= Br- >> I-) is characteristic of Eisenman selectivity sequence IV (27). It has previously been shown that CFTR channel halide permeability is most consistent with Eisenman sequence III (28, 29). However, it has been suggested that I- may actually be more permeable than Cl-, resulting in an Eisenman sequence I pattern of permeability (I- > Br- > Cl-), but because I- also blocks the pore it gives the apparent halide permeability sequence initially described (28). According to the Eisenman theory, the Cl-:HCO3- permeability ratio increases with the sequence number (27). Thus, the Cl-:HCO3- permeability ratio for CFTR (sequence I or III) is smaller than for ORCC (IV). In other words, the relative permeability of HCO3- is greater for CFTR than for ORCC, whereas the relative permeability of Cl- is greater for ORCC than for CFTR.

Cobb and associates (21) have shown that adenosine receptors act through cAMP-dependent pathways that do not involve a change in [Ca2+]i in Calu-3 cells. We have confirmed and extended this finding by showing that neither the adenosine nor the subsequent DIDS response is Ca2+-dependent. These findings are significant as other studies have suggested that DIDS may stimulate anion secretion by raising [Ca2+]i in T84 (23) and Calu-3 (24) cells. In contrast to our findings, Brayden and coworkers used T84 cells to show that when basolateral Ca2+ is depleted, the basolateral DIDS-induced increase in Isc is absent (23). These findings may represent differences between intestinal and airway epithelia. Studies by Ito and colleagues (24) have suggested that basolateral DIDS increases Isc partially by increasing [Ca2+]i in Calu-3 cells. Our results may differ because they studied the DIDS effect on baseline Isc, whereas we studied its effect on adenosine stimulated Isc. Neither these studies nor ours could address the Ca2+ issue directly because DIDS cannot be used in fluorescence studies (23). Perhaps the best argument in favor of our interpretation is the simplest: DIDS is a well known inhibitor of anion channels and our results can be entirely explained by this mechanism of action; there is no need to ascribe new roles for extracellular DIDS in the form of effects on intracellular Ca2+. Indeed, the data obtained by Ito and colleagues (24) can be reinterpreted in light of our findings. They showed that Ca2+ depletion and basolateral KCa blockers inhibit the basolateral DIDS response to a similar extent. They interpret this finding to mean that basolateral DIDS activates KCa by raising intracellular Ca2+. However, it is more likely that in the absence of KCa activity, either through pharmacologic block or by depletion of [Ca2+]i, the activity of any Isc stimulating agent will be reduced because the driving force for anion secretion is reduced. In our experiments we prestimulated cells via a cAMP-dependent pathway, which is known to activate basolateral cAMP-dependent K+ channels in Calu-3 cells (30). These cAMP-dependent K+ channels likely compensated for the lack of KCa activity in the absence of [Ca]i, and thus the basolateral DIDS response was conserved. This demonstrates that the DIDS response is neither Ca2+- nor KCa channel–dependent in Calu-3 cells. We further support our argument by showing that another anion channel blocker structurally unrelated to DIDS, 9-AC, stimulates anion secretion when applied basolaterally. A similar explanation may be applicable to T84 cells, to account for the apparent Ca2+ dependence of the basolateral DIDS effect (23), but this will require further experimentation and may be aided by the use of 9-AC. Thus, it is unlikely that basolateral DIDS alters [Ca2+]i; instead changes in [Ca2+]i may alter the Isc response to basolateral DIDS, under some conditions.

It has been shown previously that adenosine receptor signaling to CFTR in Calu-3 cells involves AKAPs (22). The results of our study show that AKAPs may also be involved in the activation of basolateral ORCC by adenosine receptors. Further experimentation, particularly in patch-clamp studies, will be necessary to prove this. However, binding to AKAPs may be an important clue in the determination of the molecular identity of this channel.

NHBE cells appear to secrete little or no HCO3-, under any conditions tested, and Isc generated by these cells is due to Cl- transport. In contrast, Calu-3 cells can secrete either Cl- or HCO3-. HCO3- removal inhibits the majority of baseline Isc and results in an adenosine response that displays an initial peak in Isc, but no plateau. 36Cl- flux studies show that baseline Isc is due to HCO3- secretion, the adenosine-dependent peak is due to Cl- secretion, and the subsequent plateau is due to HCO3- secretion. To explain these results we propose the following model. CFTR channels are rapidly activated by adenosine resulting in an initial rise in JClBA (Table 2, T50), whereas basolateral ORCC are activated after a delay, such that JClAB "catches up" shortly thereafter (Table 2, T60). Opening of apical CFTR channels, which are more permeable to Cl- than to HCO3-, results in Cl- secretion. Subsequently, basolateral ORCC are activated, effectively redirecting Cl- away from the apical membrane and causing it to be recycled across the basolateral membrane. At the same time, CFTR, which is more permeable to HCO3- than the basolateral ORCC, is now available to conduct HCO3-. This model explains our results, as well as those of others (3), that show that prolonged cAMP elevation in Calu-3 cells raises unidirectional Cl- flux in both directions, resulting in no net secretion. However, although we have shown that this is likely due to activation of cAMP-dependent basolateral Cl- channels, others suggested that this may be due to activation of cAMP-dependent electroneutral Cl-:Cl- exchange via basolateral anion exchangers (3). Either explanation would be possible, but our studies also show that this effect is electrogenic indicating that it is mediated at least in part by ORCC.

Our model can also explain why Calu-3 HCO3- secretion dominates Cl- secretion. Tamada and coworkers (7) showed that forskolin can activate apical membrane conductance so dramatically that HCO3- secretion can be easily sustained, despite the reduced permeability of CFTR to this anion. These authors postulated that preferential HCO3- secretion could be achieved if the basolateral NKCC cotransporters were inactive, in agreement with the fact that bumetanide has little effect on Isc. Our data show that preferential HCO3- secretion can be achieved, even in the presence of NKCC activity, because Cl- can be recycled across the basolateral membrane via ORCC.

Several other epithelia have been shown to secrete HCO3- preferentially over Cl- (4, 31), but the mechanisms by which this occurs are largely unknown. One of systems that has been well characterized is the pancreatic duct epithelium, which secretes Cl- through apical CFTR channels, and then exchanges Cl- for HCO3- through apical anion exchangers (31). On first inspection this model of HCO3- secretion as well as the one presented in the current study seems unnecessarily complicated. It would be far simpler for epithelia to express apical anion channels that are more permeable to HCO3- than Cl-, and thus circumvent the need for apical anion exchangers or basolateral anion channels. However, currently there are no known anion channels that are more permeable to HCO3- than to Cl-, and their existence is unlikely. An examination of anion isotherms (27) indicates that in order for an anion channel to be more permeable to HCO3- than to Cl-, the pore would have to be very large, so large in fact that the channel may cease to display any selectivity. Therefore, epithelia may have been forced to evolve mechanisms, such as the one described in the present study, to circumvent structural limitations of ion channels.

In our studies, when Cl- is removed from the bathing solution basolateral DIDS inhibits adenosine-dependent Isc in Calu-3 cells. This effect is likely due to inhibition of Na+-HCO3- cotransporters, a known target for DIDS. Therefore, in KHS, the Isc stimulated by DIDS in the presence of adenosine reflects simultaneous basolateral Cl- channel inhibition, which tends to raise Isc, and basolateral Na+-HCO3- cotransport inhibition, which tends to lower Isc. In support of this argument, the inhibitory effect of basolateral DIDS in Cl--free KHS (10.4 ± 0.5 µA/cm2, Figure 6C) is not statistically different from the inhibitory effect of basolateral DNDS in KHS (12.8 ± 1.6 µA/cm2, n = 3, Figure 2D). Thus the DIDS stimulated Isc increase (15.6 ± 2.0 µA/cm2, Figure 2A) measured in KHS after NECA addition underestimates the contribution of basolateral Cl- channels. A better measure of their contribution to Isc is derived from experiments in which cells are pretreated with NECA and DNDS before DIDS application (29.7 ± 1.1 µA/cm2, Figure 2D). This value agrees very closely with the directly measured JClnet of 1.20 ± 0.13 µEq/cm2h (32.2 ± 3.5 µA/cm2, Figure 7B, Table 2, T80) stimulated by DIDS, in KHS, and in the presence of NECA. Thus, these results suggest that in cells in HCO3--free KHS and pretreated with NECA, DIDS should stimulate Isc by ~ 30 µA/cm2. However, Isc measurements in HCO3--free KHS demonstrated increases of only 9.4 ± 0.7 µA/cm2 (n = 4). This indicates that Cl- secretion in Calu-3 cells is HCO3--dependent, which can occur for a number of reasons: (i) HCO3- removal may depolarize the cell and therefore reduce the driving force for Cl- secretion; (ii) basolateral anion exchangers are involved; and (iii) soluble adenylyl cyclase, which is HCO3--dependent and is present in Calu-3 cells (unpublished data), is involved. In particular the possibility of anion exchanger involvement is interesting because it is also known to be DIDS-sensitive. The fact that there is a small, but statistically significant (P < 0.0001) Cl- absorption (T60, Figure 7B) in KHS occurring under short-circuit conditions after adenosine receptor stimulation, is consistent with the involvement of such an anion exchanger. A finding supported in previous studies of Calu-3 cells that demonstrated the existence of a basolateral anion exchange process, likely mediated by Anion Exchanger 2, that contributed modestly to Isc (6). This anion exchanger is a target for both DNDS and DIDS and thus its effects, like those of the Na+-HCO3- cotransporter, can be factored out by sequential application of these drugs.

Although our data clearly show that basolateral anion channels are important in determining the magnitude of Cl- secretion, we also propose that they may be important in determining HCO3- secretion. Similar to the role of basolateral K+ channels, the opening of basolateral anion channels may affect membrane potential. Specifically, the opening of basolateral anion channels is predicted to depolarize the cell thereby stimulating Na+-HCO3- cotransporters and HCO3- secretion (7). However, because of limited pharmacologic tools, and the lack of information regarding the molecular identity of ORCC, we could not directly address this issue.

In summary, this study not only expands our knowledge of epithelial cell function, but may also have important clinical implications. Our conclusions suggest that the ORCC may be an important modifier of the CF phenotype and that modulation of this channel may be a useful adjunct therapy for this or other illnesses characterized by abnormal airway epithelial anion secretion.


   Acknowledgments
 
A.S. is an M.D./Ph.D. student, and M.D. is a Senior Scholar, both of the Alberta Heritage Foundation for Medical Research. This work was supported by the Canadian Cystic Fibrosis Foundation and the Canadian Institutes of Health Research.

Received in original form March 25, 2003

Received in final form May 15, 2003


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Willumsen, N. J., C. W. Davis, and R. C. Boucher. 1989. Intracellular Cl- activity and cellular Cl- pathways in cultured human airway epithelium. Am. J. Physiol. Cell Physiol. 256:C1033–C1044.[Abstract/Free Full Text]
  2. Smith, J. J., and M. J. Welsh. 1992. cAMP stimulates bicarbonate secretion across normal, but not cystic fibrosis airway epithelia. J. Clin. Invest. 89:1148–1153.
  3. Devor, D. C., A. K. Singh, L. C. Lambert, A. DeLuca, R. A. Frizzell, and R. J. Bridges. 1999. Bicarbonate and Chloride Secretion in Calu-3 Human Airway Epithelial Cells. J. Gen. Physiol. 113:743–760.[Abstract/Free Full Text]
  4. Illek, B., J. R. Yankaskas, and T. E. Machen. 1997. cAMP and genistein stimulate HCO3- conductance through CFTR in human airway epithelia. Am. J. Physiol. Lung Cell. Mol. Physiol. 272:L752–L761.[Abstract/Free Full Text]
  5. Paradiso, A. M., R. D. Coakley, and R. C. Boucher. 2003. Polarized distribution of HCO3- transport in human normal and cystic fibrosis nasal epithelia. J. Physiol.(Lond.) 548:203–218.[Abstract/Free Full Text]
  6. Loffing, J., B. D. Moyer, D. Reynolds, B. E. Shmukler, S. L. Alper, and B. A. Stanton. 2000. Functional and molecular characterization of an anion exchanger in airway serous epithelial cells. Am. J. Physiol. Cell Physiol. 279:C1016–C1023.[Abstract/Free Full Text]
  7. Tamada, T., M. J. Hug, R. A. Frizzell, and R. J. Bridges. 2001. Microelectrode and impedance analysis of anion secretion in Calu-3 cells. JOP 2:219–228.[Medline]
  8. Gray, M. A., C. E. Pollard, A. Harris, L. Coleman, J. R. Greenwell, and B. E. Argent. 1990. Anion selectivity and block of the small-conductance chloride channel on pancreatic duct cells. Am. J. Physiol. Cell Physiol. 259:C752–C761.[Abstract/Free Full Text]
  9. Linsdell, P., J. A. Tabcharani, and J. W. Hanrahan. 1997. Multi-ion mechanism for ion permeation and block in the cystic fibrosis transmembrane conductance regulator chloride channel. J. Gen. Physiol. 110:365–377.[Abstract/Free Full Text]
  10. Frizzell, R. A., G. Rechkemmer, and R. L. Shoemaker. 1986. Altered regulation of airway epithelial cell chloride channels in cystic fibrosis. Science 233:558–560.[Abstract/Free Full Text]
  11. Welsh, M. J. 1986. An apical-membrane chloride channel in human tracheal epithelium. Science 232:1648–1650.[Abstract/Free Full Text]
  12. Gabriel, S. E., L. L. Clarke, R. C. Boucher, and M. J. Stutts. 1993. CFTR and outward rectifying chloride channels are distinct proteins with a regulatory relationship. Nature 363:263–268.[CrossRef][Medline]
  13. Schwiebert, E. M., M. E. Egan, T. H. Hwang, S. B. Fulmer, S. S. Allen, G. R. Cutting, and W. B. Guggino. 1995. CFTR regulates outwardly rectifying chloride channels through an autocrine mechanism involving ATP. Cell 81:1063–1073.[CrossRef][Medline]
  14. Xia, Y., C. M. Haws, and J. J. Wine. 1997. Disruption of monolayer integrity enables activation of a cystic fibrosis "bypass" channel in human airway epithelia. Nat. Med. 3:802–805.[CrossRef][Medline]
  15. Uyekubo, S. N., H. Fischer, A. Maminishkis, B. Illek, S. S. Miller, and J. H. Widdicombe. 1998. cAMP-dependent absorption of chloride across airway epithelium. Am. J. Physiol. Lung Cell. Mol. Physiol. 275:L1219–L1227.[Abstract/Free Full Text]
  16. Szkotak, A. J., A. M. L. Ng, S. F. P. Man, S. A. Baldwin, C. E. Cass, J. D. Young, and M. Duszyk. 2003. Coupling of CFTR-mediated anion secretion to nucleoside transporters and adenosine homeostasis in Calu-3 cells. J. Membr. Biol. (In press)
  17. Danahay, H., H. Atherton, G. Jones, R. J. Bridges, and C. T. Poll. 2002. Interleukin-13 induces a hypersecretory ion transport phenotype in human bronchial epithelial cells. Am. J. Physiol. Lung Cell. Mol. Physiol. 282:L226–L236.[Abstract/Free Full Text]
  18. Schultz, S. G., and R. Zalusky. 1964. Ion transport in isolated rabbit ileum. J. Gen. Physiol. 47:567–584.[Abstract/Free Full Text]
  19. Haws, C., W. E. Finkbeiner, J. H. Widdicombe, and J. J. Wine. 1994. CFTR in Calu-3 human airway cells: Channel properties and role in cAMP-activated Cl- conductance. Am. J. Physiol. Lung Cell. Mol. Physiol. 266:L502–L512.[Abstract/Free Full Text]
  20. Hwang, T., H. Lee, N. Lee, and Y. C. Choi. 2000. Evidence that basolateral but not apical membrane localization of outwardly rectifying depolarization-induced Cl- channel in airway epithelia. J. Membr. Biol. 176:217–221.[CrossRef][Medline]
  21. Cobb, B. R., F. Ruiz, C. M. King, J. Fortenberry, H. Greer, T. Kovacs, E. J. Sorscher, and J. P. Clancy. 2002. A2 adenosine receptors regulate CFTR through PKA and PLA2. Am. J. Physiol. Lung Cell. Mol. Physiol. 282:L12–L25.[Abstract/Free Full Text]
  22. Huang, P., K. Trotter, R. C. Boucher, S. L. Milgram, and M. J. Stutts. 2000. PKA holoenzyme is functionally coupled to CFTR by AKAPs. Am. J. Physiol. Cell Physiol. 278:C417–C422.[Abstract/Free Full Text]
  23. Brayden, D. J., M. E. Krouse, T. Law, and J. J. Wine. 1993. Stilbenes stimulate T84 Cl- secretion by elevating Ca2+. Am. J. Physiol. Gastrointest. Liver Physiol. 264:G325–G333.[Abstract/Free Full Text]
  24. Ito, Y., M. Son, H. Kume, and K. Yamaki. 2001. Novel effects of minocycline on Ca2+-dependent Cl- secretion in human airway epithelial Calu-3 cells. Toxicol. Appl. Pharmacol. 176:101–109.[CrossRef][Medline]
  25. Kerstan, D., J. Thomas, R. Nitschke, and J. Leipziger. 1999. Basolateral store-operated Ca2+-entry in polarized human bronchial and colonic epithelial cells. Cell Calcium 26:253–260.[CrossRef][Medline]
  26. Zhang, Z. H., F. Jow, R. Numann, and J. Hinson. 1998. The airway-epithelium: a novel site of action by guanylin. Biochem. Biophys. Res. Commun. 244:50–56.[CrossRef][Medline]
  27. Wright, E. M., and J. M. Diamond. 1977. Anion selectivity in biological systems. Physiol. Rev. 57:109–156.[Abstract/Free Full Text]
  28. Tabcharani, J. A., P. Linsdell, and J. W. Hanrahan. 1997. Halide permeation in wild-type and mutant cystic fibrosis transmembrane conductance regulator chloride channels. J. Gen. Physiol. 110:341–354.[Abstract/Free Full Text]
  29. Illek, B., A. W. Tam, H. Fischer, and T. E. Machen. 1999. Anion selectivity of apical membrane conductance of Calu-3 human airway epithelium. Pflugers Arch. 437:812–822.[CrossRef][Medline]
  30. Cowley, E. A., and P. Linsdell. 2002. Characterization of basolateral K+ channels underlying anion secretion in the human airway cell line Calu-3. J. Physiol.(Lond.) 538:747–757.[Abstract/Free Full Text]
  31. Sohma, Y., M. A. Gray, Y. Imai, and B. E. Argent. 2001. 150 mM HCO3- -how does the pancreas do it? Clues from computer modelling of the duct cell. JOP 2:198–202.[Medline]



This article has been cited by other articles:


Home page
 Am. J. Physiol. Lung Cell. Mol. Physiol.Home page
O. A. Itani, F. S. Lamb, J. E. Melvin, and M. J. Welsh
Basolateral chloride current in human airway epithelia
Am J Physiol Lung Cell Mol Physiol, October 1, 2007; 293(4): L991 - L999.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Lung Cell. Mol. Physiol.Home page
H. Fischer, B. Illek, W. E. Finkbeiner, and J. H. Widdicombe
Basolateral Cl channels in primary airway epithelial cultures
Am J Physiol Lung Cell Mol Physiol, June 1, 2007; 292(6): L1432 - L1443.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2003-0109OCv1
29/6/710    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Szkotak, A. J.
Right arrow Articles by Duszyk, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Szkotak, A. J.
Right arrow Articles by Duszyk, M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Proc. Am. Thorac. Soc. Am. J. Respir. Crit. Care Med.
Copyright © 2003 American Thoracic Society.