K+ channels play a crucial role in epithelia by repolarizing cells and maintaining electrochemical gradient for Na+ absorption and Cl- secretion. In the airway epithelium, the most frequently studied K+ channels are KvLQT1 and KCa. A functional role for KATP channels has been also suggested in the lung, where KATP channel openers activate alveolar clearance and attenuate ischemia-reperfusion injury. However, the molecular identity of this channel is unknown in airway and alveolar epithelial cells (AEC). We adopted an RT-PCR strategy to identify, in AEC, cDNA transcripts for Kir channels (Kir6.1 or 6.2) and sulfonylurea receptors (SUR1, 2A, or 2B) forming KATP channels. Only Kir6.1 and SUR2B were detected in freshly isolated and cultured alveolar cells. To determine the physiological role of K+ channels in the transepithelial transport of alveolar monolayers, we studied the effect, on total short-circuit currents (Isc), of basolateral application of glibenclamide, an inhibitor of KATP channels, as well as clofilium, charybdotoxin, clotrimazole, and iberiotoxin, inhibitors of KvLQT1 and KCa channels, respectively. Interestingly, activity of the three types of K+ channels was detected, since all tested inhibitors decreased Isc. Furthermore, these K+ channel inhibitors reduced amiloride-sensitive Na+ currents (mediated by ENaC) and completely abolished stimulation of Cl- currents by forskolin. Conversely, pinacidil, an activator of KATP channels, increased Na+ and Cl- transepithelial transport by 33–35%. These results suggest the presence, in AEC, of a KATP channel, formed from Kir6.1 and SUR2B subunits, which plays a physiological role, with KvLQT1 and KCa channels, in Na+ and Cl- transepithelial transport.
- ATP-sensitive K+ channel
- Na+ and Cl- transepithelial transport
in absorptive and secretory epithelia, K+ channels play a major role in maintaining an electrochemical gradient necessary for Na+ and Cl- transepithelial transport. Na+ reabsorption and Cl- secretion are important processes in airway epithelial cells. Although the molecular identities of the ion channels and transporters involved in Na+ and Cl- transport [like the epithelial sodium channel (ENaC) and the cystic fibrosis transmembrane regulator (CFTR), NaK2Cl cotransporter, and Na+-K+-ATPase] are relatively well defined, very little is known about the molecular nature and physiological role of the system involved in K+ transport.
The four principal classes of K+ channels are represented in nasal and lung epithelial cells (for review, see Refs. 12 and 46). Indeed, 1) multiple voltage-dependent K+ channels (including Kv1.1, 1.3, 1.4, 4.1–4.3, Kv9.3, and KvlQT1); 2) calcium-activated K+ channels (KCa channels SK4 or Slo); 3) inwardly rectifying K+ channels [Kir2.1 (33), Kir6.1 (KATP channel), Kir7 (19)]; and 4) four transmembrane domain K+ channels (Twik, Trek, Task) have been found to be expressed. Among all these K+ channels expressed in the lung, we know nothing about the functional role of several of them. In addition, the physiological relevance of the presence of so many K+ channels is still unclear. Nevertheless, a potential importance of KvLQT1, KCa, and ATP-sensitive K+ (KATP) channels in transepithelial ion transport has been suggested.
KvLQT1 (1, 35, 53) is expressed at high levels in lung and nasal epithelial cells (15, 25, 43). This small-conductance K+ channel (<3 pS, Ref. 54), activated by cAMP and inhibited by clofilium, could play a role in Cl- secretion in nasal and bronchial cells (13, 41–43), as well as in tracheal Na+ absorption (25). Two types of KCa channels seem to be present in the airways, one with intermediate conductance (IKCa or SK4 subtype) detected in trachea and bronchial cells (13, 44, 55), the other with high conductance (200 pS, maxi-KCa or Slo1 subtype) reported in nasal cells and the alveolar A549 cell line (36, 37, 47). Both of these KCa channels could be important in Cl- secretion (16, 42, 50).
KATP channels are formed from two different types of subunits: inwardly rectifying pore-forming subunits (Kir6.1 or 6.2) and sulfonylurea receptors (SUR1, 2A, or 2B). Kir6.1, cloned for the first time from a lung library (29), was found to be expressed at a moderate level in the rat lung (30), whereas Kir6.2 and SUR1, highly expressed in pancreatic islets, seem absent in the lung (28). In addition, SUR2A is predominantly expressed in heart and skeletal muscle, whereas SUR2B is ubiquitously expressed, including in the lung (31). Such distribution indicates that the lung epithelial KATP channel could be formed from Kir6.1 and SUR2B subunits. However, the precise molecular identity and cellular localization of KATP in the lung are unknown. In fact, although such a composition was predicted in the kidney, we have recently shown that the proximal tubule KATP channel could be formed from the Kir6.1 plus SUR2A and/or SUR2B subunits (10). KATP channels are sensitive to ATP, inhibited by glibenclamide, and activated by diazoxide or pinacidil.
KATP channels might have an important role in lung physiology and pathophysiology. Indeed, YM934, a KATP channel opener, increases alveolar liquid clearance in the human lung (48). This result indicates that KATP activation might stimulate Na+ transport in the lung, since it is the main mechanism involved in alveolar liquid clearance (4). KATP channel openers, with a well-known protective effect against ischemia-reperfusion injury in the heart, have also been shown to exert a beneficial action in the lungs (24, 34, 56).
Because KATP channels might have an important physiological impact on alveolar epithelial cell functions and in view of the dearth of knowledge on these channels in the lung, we decided to characterize their molecular identity and determine their physiological role in transepithelial transport in alveolar epithelial cells. We found that Kir6.1 and SUR2B are subunits that form the KATP channel in alveolar cells, and its modulation has an impact on Na+ and Cl- transepithelial transport.
MATERIALS AND METHODS
Alveolar Epithelial Cell Isolation and Primary Culture
Alveolar epithelial type II cells (ATII) were isolated from male Sprague-Dawley rats according to a well-established protocol (23) and according to a procedure approved by the animal care and use committee of the institution. In brief, after anesthesia, tracheotomy, and cannulation of the pulmonary artery, the lungs were perfused to remove blood cells and simultaneously inflated with air. After retrieval from the thoracic cavity, they were washed 10 times to remove alveolar macrophages and treated subsequently with elastase. They were then minced, and the resulting suspensions were filtered. Alveolar cells were collected and purified by a differential adherence technique that enhances the purity of the ATII cell pool (17). Briefly, “pre-IgG” cell suspensions were incubated for 45–60 min on bacteriological plastic plates coated with rat IgG at 37°C in a 5% CO2 incubator. Most macrophages were bound to IgG, whereas nonadherent “post-IgG” cells were collected and recovered by centrifugation. Alkaline phosphatase staining, to identify ATII cells (21), showed that the proportion of ATII cells of 69% in the pre-IgG mix increased to 86% in the post-IgG mix. This freshly isolated cell suspension was used directly for RNA extraction (day 0) or plated at 1 × 106 cells/cm2 on Transwell Costar permeant filters (4 cm2) in MEM containing 10% FBS, 0.08 mg/l gentamicin, septra (3 μg/ml trimethoprime + 17 μg/ml sulfamethoxazole), 0.2% NaHCO3, 10 mM HEPES, and 2 mM l-glutamine and then cultured at 37°C with 5% CO2 in a humidified incubator. The medium was replaced after 3 days by the same MEM without septra.
RNA purification. Total RNA from ATII cells (freshly isolated or cultured for 1–4 days on permeant filters) was purified with TRIzol reagent according to the manufacturer's instructions (Invitrogen, Burlington, Ontario, Canada).
PCR amplification of full-length KATP subunits. cDNAs were obtained with the Smart PCR cDNA synthesis kit (Clontech, Palo Alto, CA) from total RNAs purified from freshly isolated ATII cells. The multiple cDNA synthesis cycles of this protocol allow significant enrichment of full-length cDNA, a very appropriate tool, especially for long cDNAs such as SUR. Primers designed from the sequences of cloned rat Kir6.1 (Ref. 30, GenBank NM017099), Kir6.2 (Ref. 31, GenBank D86039), SUR1 (Ref. 11, GenBank AB052294), SUR2A (Ref. 27, GenBank D83598), and SUR2B (Ref. 11, GenBank AB045281) were used to amplify PCR products from ATII cell cDNAs with the Long Expand Long Template PCR System (Roche, Laval, Québec, Canada). The Kir6.1 primers (sense: 5′-ccgccatgctggccaggaagagcatcatcc3′, antisense: 5′ccctcatgattctgatgggcactggtttcc-3′) were designed to amplify a 1,283-bp product corresponding to full-length Kir6.1, whereas the Kir6.2 primers (sense: 5′-ccgccatgctgtcccgaaaaggcattatcc-3′, antisense: 5′-gcaactcaggacaaggaatccggagagatgc-3′) would amplify a 1,181-bp product corresponding to full-length Kir6.2. For full-length SUR1 (4,760-bp), the following primers were deployed: sense: 5′-ccaccatgcctttggccttctgcggcaccg-3′, antisense: 5′-ggctggtcatttgtccgcgcggacaaagg-3′. The SUR2 primer pair (sense: 5′-ccgccatgagcctttccttctgtggtaac-3′, antisense: 5′-cctgtcacatgtccgcacgaacgaacgagg-3′) amplifies a 4.6-bp product corresponding to full-length SUR2A or SUR2B. Indeed, the 5′-ends of these two subunits are identical, and the selected sequence of the 3′-primer, corresponding to the 3′-end of SUR2B, is also present in the 3′-untranslated region of SUR2A. In fact, the nucleotide sequences of SUR2A and SUR2B are identical, apart from an insertion in the 3′-end of the coding region of SUR2A, which generates a COOH terminus different from SUR2B. Therefore, a different strategy was adopted to discriminate between the two isoforms (see below).
Semiquantitative PCR. Five micrograms of total RNA, purified from freshly isolated or cultured ATII cells, were reverse-transcribed to cDNA with Moloney murine leukemia virus reverse transcriptase (RT; Invitrogen, Burlington, Ontario, Canada) in the presence of oligo-dT primers. cDNAs were amplified with Taq polymerase (Invitrogen) using specific primers designed from sequences of cloned rat Kir6.1 and SUR2. The Kir6.1b primers (sense: 5′-cgcccacggggacatctatgc-3′, antisense: 5′-agggggctacgcttatcaat-3′, 1 μM final concentration of each) were designed to amplify a 544-bp product. To discriminate between SUR2A and SUR2B, we employed a new pair of primers (sense: 5′-gcggatcgcacggttgtaaccatagctc-3′, antisense: 5′-cctgtcacatgtccgcacgaacgaacgagg-3′, 1 μM final concentration of each) to encompass the 3′-insert region of SUR2A and, therefore, amplify products of different size for SUR2A (343 bp) or SUR2B (169 bp). The β-actin primer pair (sense: 5′-ctaaggccaaccgtgaaaag-3′ and antisense: gccatctcttgctcgaagtc, 0.25 μM final concentration of each) amplified a 311-bp product. Semiquantitative RT-PCR amplification was undertaken according to a well-established laboratory protocol (9, 14). Briefly, Kir6.1 and SUR2 products were amplified for 30 cycles, whereas β-actin amplification was stopped after 20 cycles to remain in the linear phase (Fig. 2C). Because β-actin amplification, even in the linear phase of amplification, remains stable under all test conditions (see agarose gels in Fig. 2, A and B), Kir6.1 and SUR2B PCR products were normalized with the β-actin signal for each cDNA sample. RT-PCR products were finally separated on agarose gels, stained with ethidium bromide, and analyzed with the Typhoon Gel Imager, as described previously (9, 14).
Sequencing. The 544-bp and 169-bp bands, obtained respectively with Kir6.1 and SUR2 primer pairs, were extracted from agarose gel with QIAEXII (Qiagen, Mississauga, Ontario, Canada) according to the manufacturer's instructions. The products were purified by filtration on Microcon-PCR filters (Amicon, Beverly, MA) and ligated in pGEMT-Easy vector (Promega, Madison, WI). The nucleotide sequence of the isolated clones was confirmed by sequencing at the Centre hospitalier de l'Université de Montréal (CHUM) sequencing facilities with an ABI Prism 3100 Genetic Analyzer (Applied Biosystems, Foster City, CA).
Short-circuit current measurements in an Ussing chamber. Rat ATII cell monolayers (day 3 or 4) with high resistance (>1,200 Ω·cm2) and high Na+ transport rates were mounted in a heated (37°C) Ussing chamber and perfused on the apical and basolateral sides with warm physiological solution (containing in mM: 140 NaCl, 5 KCl, 10 TES, 1 MgCl2, 1 CaCl2, and 10 glucose). The transepithelial potential difference was clamped to zero by an external voltage clamp amplifier (apical and basolateral sides of the monolayer connected via KCl agar-calomel half-cells and Ag-AgCl electrodes), and the resulting short-circuit current (Isc) was recorded continuously on a computer. Resistance was determined from the current needed to clamp voltage from 0 to 1 mV for 1 s every 10 s.
Statistics. Average values are given as means ± SE, and n represents the number of experiments that were performed on at least four different animals.
Molecular Identity and Expression of KATP Channels in Alveolar Epithelial Cells
The molecular identity of the lung KATP channel subunits was investigated in alveolar cells. Different primer pairs, designed to specifically amplify rat Kir6.1, Kir6.2, SUR1, and SUR2, were used to generate PCR products from cDNAs of freshly isolated ATII cells (Fig. 1). The Kir6.1 primer pair (see materials and methods) amplified a ∼1,200- to 1,300-bp product, which could correspond to full-length Kir6.1 (Fig. 1A). In addition, the SUR2 primer pair amplified a PCR product that corresponded to the size expected (∼4.6 kb) for full-length SUR2A or SUR2B (Fig. 1D). However, this primer pair could not discriminate between SUR2A and SUR2B (see materials and methods). No product could be detected with primers designed from Kir6.2 or SUR1 (Fig. 1, B and C). These results suggested that the alveolar KATP channel could be formed from the Kir6.1 and SUR2 subunits.
It is well established that ATII cells progressively lose their phenotype in culture. For this reason, we decided to follow the expression of the Kir6.1 and SUR2 subunits from freshly isolated ATII cells (day 0) to ATII monolayers cultured for 1–4 days (time when electrophysiological experiments were performed). For this purpose, new primers were designed to amplify smaller PCR products for Kir6.1 and SUR2 to produce a more reliable semiquantitative protocol. For Kir6.1, the new primers (see materials and methods) amplified, as expected, a 544-bp fragment. This 544-bp band was extracted, purified, and subcloned into pGEMT-Easy vector. Sequencing the insert confirmed 100% identity with cloned rat Kir6.1 (Ref. 30, GenBank NM017099). As represented in Fig. 2A, Kir6.1 expression decreased progressively from freshly isolated cells (day 0) to cells cultured for 2 days (78 ± 2% decline between days 0 and 2, P < 0.001, n = 4). Expression increased afterwards, on days 3 and 4 (44 ± 10% increment between days 2 and 4, P < 0.025, n = 4).
As detailed (in materials and methods), the nucleotide sequences of SUR2A and SUR2B being identical, apart from an insertion in the 3′-end of the coding region of SUR2A, the primer pair used in Fig. 1D could therefore amplify the two SUR2 subunits. A new pair of PCR primers was designed to frame the 3′-end insert of SUR2A and amplify products of different size from either SUR2A or SUR2B (343 and 169 bp, respectively; see materials and methods). As observed in Fig. 2B, only a 169-bp product could be amplified. Sequencing this PCR product confirmed 100% identity to the rat SUR2B subunit (Ref. 11, GenBank AB045281). The expression of this SUR2B subunit was then followed in primary cultured cells. The observed expression profile of SUR2B was similar to Kir6.1. Indeed, there was an initial 88 ± 2% decrease (P < 0.001, n = 6) between freshly isolated ATII cells (day 0) and cultured cells until day 2, followed by a 21 ± 6% rise between days 2 and 4 (P < 0.025, n = 6).
Evidence of KATP, KvLQT1, and KCa Channel Activities in Alveolar Epithelial Cells
ATII cells in primary culture formed tight epithelia with a high Na+ transport rate. Mean total Isc and transepithelial resistance, measured in the Ussing chamber, of ATII monolayers cultured for 3 or 4 days on permeable filters were 6.7 ± 0.2 μA/cm2 and 1,372 ± 84 Ω·cm2, respectively (n = 67).
As KATP subunits were detected in alveolar epithelial cells, we decided to explore the presence of KATP activity in alveolar monolayers. It was observed that basolateral application of glibenclamide (100 μM), an inhibitor of KATP channels, reduced the Isc by 50 ± 5% (n = 7, P < 0.001, Table 1).
The activities of the two main K+ channels (KvLQT1 and KCa), described in nasal and lung epithelial cells, were also explored. The KvLQT1 channel was inhibited by clofilium. Basolateral application of clofilium (100 μM) elicited 79 ± 3% inhibition of total Isc (n = 7, P < 0.001, Table 1). Charybdotoxin (200 nM), an inhibitor of KCa channels (IKCa or maxi-KCa), and iberiotoxin (100 nM), an inhibitor of maxi-KCa channels, induced a smaller decrease of Isc (15 ± 2%, P < 0.001, n = 6, and 16 ± 3%, P < 0.01, n = 5, with charybdotoxin or iberiotoxin, respectively; Table 1). Clotrimazole (20 μM, basolateral side), an inhibitor of the IKCa channel sub-class, reduced total Isc by 41 ± 5% (n = 6, P < 0.001, Table 1). Higher doses of clofilium, clotrimazole, and glibenclamide failed to increase the inhibitory effects (results not shown).
The decrease of Isc observed after basolateral application of these K+ channel inhibitors was relatively slow, with a maximum impact measured after 40 min of treatment (see glibenclamide outcome in Fig. 3B). This suggests that the 15–79% Isc reduction induced by K+ channel inhibitors could be due, in part, to K+ channel suppression and subsequent depression of other pathways, such as Na+ or Cl- channels. We therefore decided to test the impact of K+ activity on Na+ and Cl- transport.
Role of the KATP Channel in Na+ and Cl- Transport in Alveolar Cells
Effect of K+ channel inhibitors on Na+ transport. In alveolar epithelial monolayers, total Isc measured in the Ussing chamber (6.3 ± 0.8 μA/cm2) was for the most part achieved by the apical ENaC channel (Fig. 3). Indeed, application of the ENaC blocker amiloride (1 μM, apical side) resulted in a sharp Isc decrease (i.e., 68 ± 2% inhibition of total Isc, P < 0.001, n = 27), reaching a stable value after 5 min. Amiloride also induced a small increase of transepithelial resistance (528 ± 146 Ω·cm2 increment).
The 5.61 ± 0.25 μA/cm2 decline in Isc (Fig. 3A) corresponded to amiloride-sensitive current (Iamil). This response was stable over a 70-min period, since a second amiloride treatment induced a similar effect (5.83 ± 1.01 μA/cm2; Fig. 3, A and D). To test the effect of K+ channel activity on Na+ transport, we then compared the magnitude of Iamil before and after treatment with K+ channel inhibitors. As reported previously (Table 1), glibenclamide (100 μM, basolateral side) progressively reduced total Isc (Fig. 3B). Subsequent addition of amiloride (1 μM, apical side) elicited additional Isc inhibition. The magnitude of Iamil measured after KATP inhibition by glibenclamide treatment was reduced by 54 ± 6% (P < 0.001, n = 7; Fig. 3, B and D). To verify that glibenclamide specifically affected the KATP channel, we tested apical glibenclamide treatment (100 μM), which could potentially block apical CFTR channels. After a small decrease of total Isc (from 6.22 ± 0.8 to 5.26 ± 0.7, corresponding to a 13 ± 4% decline, n = 5), we verified that this treatment failed to reduce Iamil (i.e., Iamil of 3.69 ± 0.5 and 3.42 ± 0.4 before and after apical glibenclamide treatment, 5.4 ± 3.6% diminution, not significant). The effects of KvLQT1 and KCa inhibitors on Iamil were also studied. A 40-min treatment with 100 μM clofilium, the KvLQT1 channel blocker, reduced Iamil by 74 ± 3% (P < 0.001, n = 7, Fig. 3D). Charybdotoxin (200 nM), an inhibitor of IKCa and maxi-KCa channels, as well as iberiotoxin, the maxi-KCa inhibitor, induced a small but significant decrease of Iamil (i.e., 13 ± 4% inhibition, P < 0.025, n = 6, and 15 ± 3%, P < 0.01, n = 5, respectively), whereas IKCa suppression with 20 μM clotrimazole evoked a 37 ± 6% reduction (P < 0.005, n = 6, Fig. 3D).
To determine the effect of concomitant blockage of the three classes of K+ channels (KATP, KvLQT1, and KCa) on Na+ transport, glibenclamide, clofilium, and clotrimazole were also applied in combination. Comparison of Fig. 3B (glibenclamide alone) and 3C (combination of the three inhibitors) shows that the kinetics of inhibition of total Isc were faster in the presence of the three inhibitors. In addition, 20-min treatment with these inhibitors allowed a 86 ± 1% reduction in Iamil (n = 5, P < 0.001), whereas 40 min were necessary to deplete Iamil by 54, 74, or 37% with glibenclamide, clofilium, or clotrimazole alone, respectively (Fig. 3D).
Effect of K+ channel inhibitors on Cl- transport. Various studies have shown that K+ channels (KvLQT1 and KCa) play a role in Cl- secretion in nasal and lung epithelial cells (for example, see 13, 16, 41, 42, 43, 50). We, therefore, tested whether the KATP channel could play a similar role in alveolar epithelial cells.
ATII monolayers were first treated with 1 μM amiloride (apical side) to block ENaC. Forskolin (10 μM), which increases intracellular cAMP (5), was then added to the basolateral side, since cAMP is a known activator of Cl- channels, including CFTR or Ca-activated Cl- channels (3). As observed in Fig. 4, after a small, transient decrease, 55 min of forskolin application induced a progressive Isc rise from 2.36 ± 0.10 to 3.31 ± 0.19 μA/cm2. This 0.95 ± 0.14 μA/cm2 increase corresponded to 40 ± 6% stimulation (P < 0.001, n = 8). In addition, 5-nitro-2-(3-phenylpropylamino)benzoate (NPPB)-sensitive Cl- current (0.72 ± 0.27 μA/cm2, in the control condition, Fig. 4B) rose to 1.89 ± 0.21 μA/cm2 (Fig. 4A) after forskolin application (163% increment, Fig. 4C). Finally, pretreatment with NPPB (200 μM, apical side, Fig. 4B) totally prevented the forskolin-activated Isc increase, suggesting that amiloride-insensitive, forskolin-stimulated current is a Cl- current.
The impact of K+ channel inhibitors was then tested on stimulation of Cl- current by forskolin. Inhibition of the KATP channel by basolateral glibenclamide application was investigated first. This treatment completely abolished forskolin-activated Cl- current (Fig. 5A), as did inhibition of KvLQT1 by clofilium (Fig. 5B) or of KCa by clotrimazole (Fig. 5C).
Effect of the KATP channel activator, pinacidil, on Na+ and Cl- transport. Iamil was then measured after stimulation with the KATP channel activator pinacidil. It was observed that this treatment increased Iamil from 4.09 ± 0.17 μA/cm2 (in the control condition) to 5.47 ± 0.37 μA/cm2 (after 10-min treatment with 100 μM pinacidil, i.e., 33 ± 4% increment, P < 0.001, n = 4, Fig. 6A).
The amplitude of NPPB-sensitive Cl- current measured after forskolin stimulation was also compared in the presence and absence of 100 μM pinacidil. A 10-min pretreatment with pinacidil increased forskolin-stimulated Cl- current by 35 ± 8% (P < 0.01, n = 6, Fig. 6B).
The results presented in this study have allowed us to determine the molecular identity of KATP channels (Kir6.1 plus SUR2B subunits) present in alveolar epithelial cells and to demonstrate their role in Na+ and Cl- transport in these cells. Indeed, KATP channel inhibition by glibenclamide reduced 1) total Isc and 2) amiloride-sensitive Na+ current and 3) abolished Cl- current stimulation by forskolin in cultured ATII monolayers. Conversely, KATP channel activation by pinacidil increased Na+ and Cl- transepithelial transport.
Molecular Identity and Expression of the Alveolar KATP Channels
The use of PCR primers designed from cloned rat Kir6.1 and Kir 6.2, SUR1, 2A, and 2B subunits allowed us to amplify, from freshly isolated epithelial alveolar cells, PCR products for Kir6.1 and SUR2B, suggesting that the alveolar KATP channel could be formed from these two subunits. Alkaline phosphatase staining showed that this final freshly isolated cell mix contained 86% of ATII cells. Most alkaline phosphatase-negative cells seemed to be macrophages. However, it is unlikely that the amplified signal came from these cells since KATP channels appeared to be absent in macrophage (26). In addition, the lungs were perfused by the pulmonary artery to remove most blood cells. However, minor contamination in the freshly isolated cell mix could not be excluded. Indeed, KATP channels have been observed in lamprey red blood cells (38). In leukocytes, Kir6.1 mRNA but not SUR mRNAs have been detected (57). However, in culture, these blood cells are unlikely to be present in mRNA extracts of alveolar epithelial monolayers.
It has been demonstrated that ATII cells progressively change their phenotype in primary culture. In particular, they rapidly lose some specific characteristics, such as phosphatase alkaline activity (21), surfactant synthesis, and the presence of lamellar bodies. The cells also lose their cuboidal shape (49), to progressively gain a more flattened morphology and the expression of ATI markers, such as RTI40 (18). In a previous study, we reported that CFTR expression decreased progressively in primary culture (Brochiero E, Dagenais A, Berthiaume Y, and Grygorczyk R, unpublished observations; Ref. 9). In fact, channel expression seemed related to culture conditions (Brochiero E, Dagenais A, Berthiaume Y, and Grygorczyk R, unpublished observations; Refs. 9, 32). We indeed observed that, on permeable filters, the level of CFTR and surfactant protein A (SP-A) expression remained detectable, even after 4 days of culture (Brochiero E, Dagenais A, Berthiaume Y, and Grygorczyk R, unpublished observations; Ref. 9). Because conditions for Isc measurements (with high currents and resistance) were optimal on alveolar cell monolayers cultured on filter for 3 or 4 days, we decided to verify if Kir6.1 and SUR2B mRNA, identified on freshly isolated cells, were still detectable in cultured monolayers. Conversely to the continuous decrease in SP-A and CFTR expression that we reported previously (9), we found here that Kir6.1 and SUR2B expression first decreased between days 0 and 2 and then rose surprisingly on days 3 and 4. Nevertheless, the estimated level of expression on day 4 remained lower than in freshly isolated cells. The similarity between the expression profile of Kir6.1 and SUR2B mRNA suggests that expression of the two different types of subunits forming the KATP channel is modulated in parallel as a function of culture duration. In a recent study, Lee et al. (40) reported that Kir6.1, Kir6.2, SUR1, or SUR2 mRNA could not be detected in cultured rat alveolar epithelial cells. Differences in culture (different culture media) and RT-PCR (primers and annealing temperatures) conditions could explain the discrepancy. Most importantly, as we observed, KATP expression varied with time of culture. In the study of Lee et al. (40), RT-PCR experiments were performed on rat alveolar cells maintained in culture for 5–7 days, whereas Kir6.1 and SUR2B expression was followed in our work from freshly isolated alveolar cells to monolayers cultured for a maximum of 4 days. We intentionally chose the same culture period for the PCR and electrophysiological experiments, as days 3 and 4 are optimal (high current and resistance) for Isc measurements.
The Kir6.1 and SUR2B mRNA detection that we have reported here is consistent with the known organ distribution of Kir6 and SUR subunits. Indeed, Kir6.1 and SUR2B mRNAs are ubiquitously expressed, including in the lung, whereas Kir6.2 and SUR1 are predominantly expressed in pancreatic cells, and SUR2A in heart and skeletal muscle (28, 30, 31). Although Kir6.1 and SUR2B mRNAs have been detected in whole lung extracts, our results are the first to demonstrate that the two subunits are present in alveolar epithelial cells.
Evidence of KvLQT1, KCa, and KATP Channel Activities in Epithelial Alveolar Cells
RT-PCR experiments suggested the presence of a KATP channel in alveolar epithelial cells. However, the contribution of this channel to ion membrane transport in alveolar monolayers needed to be explored. In fact, a single study by Sakuma et al. (48) reported KATP channel activity in the human lung. They established that YM934, a KATP channel opener, increased potassium influx into the alveolar spaces. In this study, we now report that KATP channel inhibition by 40-min basolateral treatment with glibenclamide reduced total Isc by 50%. The basolateral effect of glibenclamide indicates that the KATP channel could be located at the basolateral membrane of alveolar epithelial cells. Western blot and immunohistochemistry experiments are, however, required to confirm, at the protein level, the presence of the channel at the basolateral membrane.
The sensitivity of total Isc to basolateral glibenclamide suggests KATP channel activity in epithelial alveolar monolayers under resting conditions. Such KATP channel activity has been detected in various native cells with mM intracellular ATP concentrations. However, the sensitivity of KATP channels to ATP measured in excised-patch experiments is in the micromolar range. Various hypotheses have been postulated to explain this complex paradox observed in several tissues expressing KATP channels. For example, it has been observed that Mg-ADP reduced sensitivity to ATP, suggesting that the ATP/ADP ratio could be more important than the ATP level to control KATP activity. More recently, it has been shown that membrane phospholipids might reduce the channel's sensitivity to ATP into the physiological range (for review, see Ref. 2). It has also been suggested that the microenvironment in the vicinity of the membrane KATP channel could influence its regulation (6, 51). The spatial arrangement of proteins like adenylate kinases (AK), which convert ATP plus AMP into ADP, could also provide an effective phosphotransfer system to modulate the ATP/ADP ratio in the microenvironment of KATP channels (20). We previously identified AK3 protein from proximal tubules, which activated KATP currents (7).
Because KvLQT1 and KCa channels are K+ channels generally described in nasal and lung epithelial cells, their activity was also investigated. The effect of KATP, KvLQT1, and KCa inhibitors on transepithelial transport was then compared.
We observed that KvLQT1 channel inhibition by clofilium considerably reduced total Isc (79% inhibition after 40 min). KvLQT1 channels have been reported in many airway epithelial cells, including human (43) and murine (41) nasal cells, murine trachea (25), mouse ciliated epithelial cells of terminal bronchioles (15), and Calu-3 cells (13). In a previous study, Demolombe et al. (15) noted that a KvLQT1 antisense probe failed to stain alveoli in the mouse lung. Even if the level of KvLQT1 expression could be too low for in situ hybridization, we nevertheless detected its activity in alveolar monolayers.
Inhibition of total Isc by charybdotoxin (IKCa and maxi-KCa inhibitor, 15% inhibition), clotrimazole (IKCa inhibitor, 41% inhibition), and iberiotoxin (maxi-KCa inhibitor, 16% inhibition) revealed the presence of a third class of K+ channels (KCa channels) in cultured alveolar epithelial cells. The magnitude of inhibition of total Isc was smaller with these inhibitors compared with that observed in the presence of glibenclamide or clofilium. In fact, KCa channels could be only partially active in the resting condition with low intracellular calcium concentrations. Patch-clamp or PCR experiments are necessary to clearly define the identity (IKCa-SK or maxi-KCa-Slo1) of alveolar KCa channels. The small-intermediate KCa channel (IKCa or SK4 channel), sensitive to clotrimazole and charybdotoxin, has been reported in the trachea and bronchi (13, 44, 55). Nasal cells and the alveolar A549 cell line developed high KCa conductance sensitive to charybdotoxin (maxi-K or Slo1 channel) (36, 37, 47). Conversely, in a recent review, O'Grady and Lee (46) recorded as unpublished data that Slo1 mRNA could not be detected in adult rat alveolar epithelial cells. However, their culture conditions and duration were not described. In fact, the clotrimazole and iberiotoxin sensitivity observed in our study suggests the presence of both IKCa and maxi-KCa activity.
The observed kinetics of Isc inhibition by any K+ channel inhibitors are slow (see Fig. 3). Our hypothesis was that the progressive Isc decline could be due to sequential events. Indeed, K+ current inhibition could be followed by changes in electrochemical gradients that will secondarily affect other ion transport, such as the Na+ or Cl- conductance. To test this hypothesis, we decided to investigate the effect of K+ channel inhibitors on Na+ and Cl- transport.
Role of the KATP Channel in Na+ Transport
We noted that inhibition of KATP, KvLQT1, and KCa channels with glibenclamide, clofilium, and clotrimazole reduced amiloride-sensitive Na+ currents by 54, 74, and 37%, respectively. In addition, pinacidil, a KATP channel activator, increased Iamil significantly (33%).
A previous study (48) has shown that the KATP channel opener YM934 augmented K+ transport and alveolar clearance in the human resected lung. In addition, it has been found that glibenclamide and amiloride blocked the increase of alveolar clearance stimulated by YM934, suggesting that KATP and ENaC channels mediated this effect. In our study, we confirmed that KATP channel activity could be related to Na+ transepithelial transport, since KATP channel inhibitors and activators could respectively down- and upregulate amiloride-sensitive Na+ current. The absence of effect with apical glibenclamide treatment on the Na+ transport confirmed that the reduction of Iamil with basolateral glibenclamide was really secondary to KATP inhibition and not to a nonspecific effect through CFTR channels.
Although several investigations have established a role for KvLQT1 channels in Cl- secretion in nasal and bronchial epithelial cells (13, 41, 42, 43), a single study (25) recently demonstrated that KvLQT1 could play a role in Na+ absorption in the trachea. Our results underscore here that KvQT1 activity seems also related to Na+ transport in alveolar cells.
It has been shown that both intermediate- and high-conductance KCa channels are related to Cl- secretion (16, 42, 50); however, to the best of our knowledge, there is no evidence that these KCa channels could play a role in Na+ transport in lung epithelial cells. Our study reveals, for the first time, that KCa inhibition could affect Na+ transport in alveolar cells.
Nature of Forskolin-Stimulated Cl- Current
Forskolin raises cAMP levels by directly activating adenylate cyclase in a variety of tissues, including alveolar epithelial cells (5). We observed that this cAMP increase progressively stimulated a Cl- current sensitive to NPPB in alveolar epithelial monolayers. The precise nature of Cl- channel(s) contributing to this current is still undetermined. In fact, in fetal alveolar cells, the CFTR channel has been frequently detected, and its role in Cl- secretion is well documented (39, 45, 52). On the other hand, in the adult alveolar epithelium, a tissue involved principally in Na+ reabsorption, the presence and role of CFTR are still controversial (for review, see Ref. 46). Yet we have recently shown the presence of functional CFTR channels in freshly isolated adult ATII cells. Even reduced compared with freshly isolated cells, CFTR expression was still detectable in primary cultured ATII monolayers (9), suggesting that the observed forskolin-stimulated Cl- current could be mediated by CFTR. However, the kinetics of Cl- current activation (Fig. 4) were slower than those generally reported for CFTR-mediated currents (39). The reduced number of CFTR channels in cultured monolayers could explain these slow kinetics of activation. Alternatively, forskolin may also indirectly stimulate other Cl- channels, like calcium-activated Cl- channels (3). Indeed, we observed that both apical glibenclamide (to block CFTR channels) and DIDS (a general inhibitor of Cl- channels without effect on CFTR) partially inhibit NPPB-sensitive, forskolin-stimulated Cl- current (unpublished data), indicating that CFTR as well as additional Cl- channels could be involved in this current.
Interestingly, this Cl- conductance could play a physiological role in ion transport. More precisely, some studies have possibly implicated Cl- conductance in stimulation of Na+ transport by cAMP analogs or β-agonists as stated in a recent review (46). In particular, recent work by Fang and coworkers (22) supports that hypothesis, since isoproterenol stimulation of fluid clearance was found to be abolished in CFTR knockout mice (Delta F508 mice).
Role of the KATP Channel in cAMP-stimulated Cl- Transport
An interesting feature of the observed cAMP-stimulated Cl- current is its modulation by K+ inhibitors and activators. Indeed, we saw that KATP, KvLQT1, or KCa inhibition by basolateral glibenclamide, clofilium, or clotrimazole abolished forskolin-stimulated Cl- current, whereas KATP activation significantly increased it. KvLQT1 has been previously shown to play a crucial role in Cl- secretion in bronchial and nasal cells (13, 41, 43). Moreover, previous studies have related KCa channels to Cl- secretion in the murine tracheal epithelium (16) and human nasal epithelium (50). In this report, we note that a third class of K+ channels, KATP, could be necessary for Cl- transport.
Concerted Role of K+ Channels
The slow inhibition kinetics of Isc observed in the presence of any of the three classes of K+ channel blockers (basolateral glibenclamide, clofilium, charybdotoxin, clotrimazole, or iberiotoxin) suggested a progressive effect, first on K+ current, which could secondarily affect other ion transports, such as Na+ and Cl- transepithelial transport. Indeed, a change in the electrochemical gradient following K+ channel inhibition could explain this effect on Na+ and Cl- currents. However, the coupling mechanism between basolateral K+ channels and apical Na+ or Cl- channels is still unknown, and various mechanisms are possible.
The kinetics of inhibition of total Isc and Iamil were faster when these inhibitors were used in combination. This observation indicated that concomitant inhibition of KATP, KvLQT1, and KCa channels could probably modify the electrophysiological characteristics of the cell more rapidly, followed by faster secondary inhibition of Na+ transport. If the kinetics were modified in the presence of the three inhibitors, it should be noted that the maximal inhibition (86% in 20 min) was quite similar to that observed with one inhibitor (74% in 40 min for example, with clofilium). This observation suggests that inhibition of one type of K+ channel was sufficient to alter, albeit slowly, Na+ transport.
These data suggest that the three different classes of K+ channels, KATP, KvLQT1, and KCa, could participate in the modulation of transepithelial transport in epithelial alveolar cells. Although this hypothesis is somewhat surprising, it is possible that a variety of K+ channels is needed to respond to the different physiological stimuli leading to changes in ion transport, since these channels have different electrophysiological characteristics (conductance and open probability) and regulatory mechanisms (ATP, cAMP, calcium).
In summary, our results identified the presence of a KATP channel, formed from Kir6.1 and SUR2B subunits, in alveolar epithelial cells. This channel could play a significant role in epithelial lung physiology, since KATP activity, in association with KvLQT1 and KCa activities, seems related to Na+ Cl- ion transport. and
The authors acknowledge the editorial assistance of Ovid Da Silva, Editor, Research Support Office, CHUM Research Center. E. Brochiero was the recipient of a postdoctoral fellowship from the Canadian Institutes of Health Research in partnership with the Canadian Cystic Fibrosis Foundation, followed by a scholarship (Bourse de la relève) from the Faculty of Medicine of the Université de Montreal, then a scholarship from Fonds de la recherche en santé du Québec [Fonds de la Recherche en Santé du Québec (FRSQ) Junior 1]. Y. Berthiaume is a national scholar from the FRSQ.
This work was supported by the Canadian Institutes of Health Research (Y. Berthiaume), the Canadian Cystic Fibrosis Foundation (Y. Berthiaume), and the CHUM Research Center (E. Brochiero).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society