Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Adenine nucleotides such as adenosine 5'-triphosphate (ATP) play fundamental roles in cellular energy metabolism, and are recognized as extracellular mediators in a wide variety of cells and organ systems (1, 2). They exert their biologic effects through specific cell-surface receptors, and regulate cellular responses including mast cell degranulation, visceral and vascular smooth-muscle contractility, and synaptic neurotransmission (1, 2). In airway epithelium, ATP is known to activate ciliary motility, mucus secretion, and Cl transport (3). Because extracellular ATP stimulates Cl channels other than cystic fibrosis (CF) transmembrane conductance regulator (CFTR) (5, 6), ATP may be a therapeutic potential for CF (7). Further, recent evidence suggests that CFTR possesses an ability not only to secrete Cl but also to induce ATP release from the epithelial cells themselves (8, 9).

The ATP-specific receptors, termed P2-prinoceptors, are structurally and functionally distinct from the P1-prinoceptors that are activated by extracellular adenosine, a metabolite of ATP (1, 10), and are further classified into two families: intrinsic ion channels termed P2X prinoceptors, and G protein-coupled receptors termed P2Y prinoceptors on the basis of pharmacologic properties coupled with the cloning of prinoceptors (10). Among them, P2Y₂ receptor (formerly P_2U), which is stimulated by both ATP and uridine 5'-triphosphate (UTP) (2), is present on airway epithelium, and the stimulation of the receptor produces cytosolic Ca²⁺ mobilization via activation of phospholipase (PL) C (4, 5) It is thus likely that ATP can stimulate Ca²⁺-activated Cl channel and, hence, Cl transport by airway epithelium. However, there are several reports indicating that activation of Cl channel by ATP occurs without an increase in cystolic Ca²⁺. That is, extracellular ATP directly activates outwardly rectifying Cl channel (ORCC) via prinoceptors on epithelial apical surface (9, 14). The ORCC stimulation with ATP is supposed to be mediated by P2Y₂ receptor (9) but not P1 or P2X (14). Together with ATP released via CFTR, the regulation mode of apical membrane Cl conductance by extracellular ATP appears to be more complicated than expected. Therefore, the purpose of the present study was to clarify the ATP-specific receptors involved and to elucidate signal transduction which leads to the activation of Cl conductance. We demonstrate here pharmacologic evidence that ATP but not UTP is capable of stimulating P1-like prinoceptor in addition to P2Y₂, thereby activating protein kinase (PK) A-dependent Cl secretion across primary cultured airway epithelium from bovine trachea.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture

Bovine tracheas were obtained from a slaughterhouse, and tracheal epithelial cells were isolated by protease as previously described (17). Briefly, strips of epithelium were pulled off the submucosa, washed four times with phosphate-buffered saline (PBS) containing 5 mM dithiothreitol (DTT), and rinsed twice with PBS. Epithelial tissues were digested with PBS containing 0.05% protease (type XIV; Sigma Chemical Co., St. Louis, MO) at 4°C overnight. After neutralization of the protease with 5% fetal calf serum (FCS), cells were pelleted (200 × g, 10 min) and suspended in 50% Dulbecco's modified Eagle's medium (DMEM) and 50% Ham's nutrient F-12 that contained 5% FCS, non-essential amino acids, penicillin (100 U/ml), streptomycin (100 µg/ml), and gentamicin (50 µg/ml). The cells were then seeded at a density of 2.5 × 10⁵ cells/cm² onto collagen-coated polycarbonate inserts (porous filters of 24-mm diameter, 10-µm thickness, and 0.4-µm pore size; Costar Transwell, Cambridge, MA). The medium was changed every other day, and the cells were cultured under air-liquid interface condition at 37°C in a CO₂ incubator (95% air/5% CO₂) for 10 to 12 d (18).

Measurement of Bioelectric Properties

The short-circuit technique for measuring electrical properties of cultured airway epithelium has been described previously (19). Briefly, the porous filter on which tracheal epithelial cells were grown was mounted between Ussing chambers (0.5-cm² surface area) and bathed with Krebs Ringer bicarbonate solution of the following composition (in mM): 115 NaCl, 25 NaHCO₃, 2.4 K₂HPO₄, 1.2 CaCl₂, 1.2 MgCl₂, and 10 glucose, equilibrated with 95% O₂/5% CO₂ warmed to 37°C. Transepthelial potential difference (PD) was measured with two polyethylene bridges containing 3% agar in saline, positioned within 1 mm from each side of the epithelial surface and connected to calomel electrodes (model 2080A-06T; Horiba Ltd, Tokyo, Japan) and a high- impedance voltmeter (model CEZ-9100; Nihon Kohden, Tokyo, Japan). Another pair of polyethylene bridges (3% agar in saline), positioned 10 mm from the orifice and connected to Ag/AgCl wires, was used to pass sufficient current through both the chamber and cells to bring the PD to zero. This short-circuit current (Isc) was automatically corrected for solution resistance between the PD-detecting bridges, and recorded continuously on a pen recorder (model SR6335; Graphtec, Tokyo, Japan). The cells were allowed to equilibrate for 20 min, and amiloride (100 µM) was added to the mucosal side of the solution to eliminate a component of Na⁺ movement in the Isc (20). When the response of Isc to amiloride became a stability that did not vary by more than 0.2 µA/cm² in any 5-min interval thereafter, drugs studied were added to the chamber.

Measurement of Intracellular Ca²⁺ Concentration

Bovine tracheal epithelial cells were cultured on coverslips (15-mm diameter; Matsunami Ltd, Tokyo, Japan) coated with human placental collagen. After confluence was achieved, the coverslip was washed with Hanks' balanced salt solution (HBSS) that contained 10 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (Hepes) at pH 7.4 and loaded with 2 µM fura 2-acetoxymethyl ester (AM) for 20 min at 37°C. The coverslip was then washed again and held with a rigid holder in a continuously stirred cuvette containing Hepes-buffered HBSS maintained at 37°C, and the fluorescence intensity was measured with a spectrophotometer (CAF-110; Japan Spectroscopic Co., Tokyo, Japan) (21). For excitation of fura 2 fluorescence, ultraviolet lights of 340- and 380-nm wavelength were automatically exchanged at a rate of 50 Hz, emitted light from cells (F340 and F380) was detected with a photomultiplier tube through a 510 ± 10-nm band-pass filter, and the fluorescence intensity ratio F340/F380 was automatically calculated. Maximal and minimal values for the ratio were determined in the presence of ionomycin (10 µM) and 5 mM ethyleneglycol-bis-(-aminoethyl ether)-N,N,N',N',-tetraacetic acid (EGTA), respectively. Intracellular Ca²⁺ concentration ([Ca²⁺]_i) was calculated using the formula described by Grynkiewicz and associates (22).

Measurement of PKA Activity

Confluent cells plated on collagen-coated 6-cm culture dishes were washed with Hepes-buffered HBSS and were rendered quiescent for 30 min at 37°C. After stimulation with suitable drugs for 1 min, assays were stopped by rapid aspiration of media followed by rinsing with ice-cold PBS. The cells were scraped; homogenized in lysis buffer (10 mM potassium phosphate [pH 6.8], 1 mM EGTA, 1 mM DTT, 0.1 mM sodium vanadate, 1.2 mM MgCl₂, 1 mM 3-isobutyl-1-methylxanthine, and 0.02% Triton X-100) containing 10 µg/ml each of aprotinin, leupeptin, and phenylmethylsulfonyl fluoride; and centrifuged (12,000 × g, 20 min) at 4°C. The supernatants were then removed and assayed for PKA activity.

PKA was assayed by measuring phosphorylation of the peptide substrate Kemptide (Leu-Arg-Arg-Ala-Ser-Leu-Gly) according to the Pierce Colorimetric PKA assay kit (Pierce, Rockford, IL). However, PKA determined by this assay system reflects the total amount of PKA in each sample because of the use of 100 mM cyclic adenosine 3',5'-monophosphate (cAMP) as an activator. We therefore modified the system and used deionized water instead of the activator. Standard curves were generated with increasing concentrations of the catalytic subunit of PKA (0 to 0.2 U/µl, bovine heart). The background (substrate phosphorylation in the presence of the PKA inhibitor peptide PKI, 2 µM) was subtracted from each sample, and PKA activity was expressed as units per milligram of cellular protein.

Drugs

DMEM, Ham's F-12, and non-essential amino acids were purchased from GIBCO BRL (Tokyo, Japan). Fura 2-AM was obtained from Dojindo Lab (Kumamoto, Japan). PKI was a product of Calbiochem (San Diego, CA). 2-Methylthio-ATP, ,-methylene-ATP, and 8-p-sulfophenyltheophylline (8-SPT) were obtained from Research Biochemicals Ins. (Natick, MA). Diphenylamine-2-carboxylate (DPC) was obtained from Nakarai, Inc. (Tokyo, Japan). Amiloride, 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid (DIDS), glybenclamaide, and all other chemicals were obtained from Sigma.

Statistics

Data are expressed as means ± standard error (SE). Comparisons between two groups were made by two-tailed paired or unpaired Student's t test. Statistical analyses of multiple comparisons were performed by one-way analysis of variance using Fisher's PLSD-test, and a P value of less than 0.05 was considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Differences between ATP- and UTP-Induced Increases in Isc

The effects of exogenous ATP and UTP on cultured sheets of bovine tracheal epithelium were investigated using bioelectric measurements of ion transport. After the cell layer cultured under air-liquid interface condition was confirmed to possess sufficient transepithelial resistance of more than 200 ·cm², Isc was measured in the presence of mucosal 100 µM amiloride. The basal Isc was 3.3 ± 0.5 µA/cm² (n = 24). Addition of 100 µM ATP to the mucosal side rapidly increased Isc (Figure 1A). This Isc change consisted of biphasic responses: an initial transient spike peaked within 20 s (first peak) and a second rise peaked within 100 s (second peak). Increases in Isc from the baseline to the first and second peak values were 19.3 ± 2.0 and 15.9 ± 1.8 µA/cm², respectively (n = 12). Addition of 100 µM UTP to the mucosal solution likewise elicited a rapid increase in Isc, as did ATP (Figure 1B). However, the UTP-induced increase in Isc lacked a large second peak. The changes of Isc in response to UTP from the baseline to the first and second peak values were 20.1 ± 2.2 and 7.9 ± 1.3 µA/cm², respectively (n = 12), and the second peak was significantly smaller than that induced by ATP (P < 0.01). As shown in Figures 1C and 1D, these responses to ATP and UTP were concentration-dependent: the nucleotide concentrations required to produce a half-maximal effect were 10 and 7.1 µM, respectively, for the first peak, and 29.6 and 14.4 µM, respectively, for the second peak. When the bathing medium was replaced with Cl free solution, in which Cl was substituted with iodide that cannot be transported across the airway epithelium, the responses of Isc to ATP and UTP were abolished (data not shown), implying that the effects are dependent entirely on Cl secretion.

View larger version (13K): [in this window] [in a new window]   Figure 1.   Differential effects between ATP and UTP on Isc in bovine tracheal epithelial cells cultured under air-liquid interface condition. (A) Mucosal ATP (100 µM) induced an initial transient increase (first peak) in Isc, followed by a biphasic response with a peak (second peak) and a sustained increase. (B) Second peak induced by UTP (100 µM) was smaller than that induced by ATP. (C and D) Concentration-dependent effects of ATP or UTP on first (open circles) and second (filled circles) peaks. Data are means ± SE; n = 4-12 for each concentration.

[Ca²⁺]i Measurement

The basal [Ca²⁺]i in the bovine tracheal epithelium was 116 ± 5 nM (n = 37). As shown in Figure 2, the cells were stimulated by several agonists for prinoceptors. Exposure to 100 µM ATP produced a rapid increase in F340/F380. This [Ca²⁺]i response was biphasic, consisting of an initial transient rise that peaked within 15 s followed by a sustained response. The latter gradually decreased but remained elevated from the baseline for more than 10 min. The values of [Ca²⁺]i in transient peak and sustained response determined at 2 min after the addition of ATP were 509 ± 27 and 191 ± 8 nM, respectively (n = 12). The quantity of 100 µM UTP caused a similar increase in [Ca²⁺]i, and the peak and the sustained values were 524 ± 32 and 197 ± 11 nM, respectively (n = 12). There were no significant differences between ATP- and UTP-induced increases in [Ca²⁺]i. On the other hand, the effect of 100 µM 2-methylthio-ATP (2-MeSATP) on [Ca²⁺]i was weak, and the same concentrations of ,-methylene-ATP (,-MeATP) and adenosine (ADO) were without effect.

View larger version (8K): [in this window] [in a new window]   Figure 2.   Representative recordings of Ca2+ responses induced by specific agonists for various prinoceptors in fura 2-loaded bovine tracheal epithelial cells. The addition of each agonist (100 µM) is shown by the arrows. Results are representatives of at least four independent experiments.

Analysis of ATP-Induced Increase in Isc

To determine whether Ca²⁺-dependent mechanism was involved in ATP-induced increase in Isc, the cells were pretreated with 2 mM EGTA for 10 min and subsequently with 10 µM thapsigargin (TG) for 5 min to chelate extracellular Ca²⁺ and to deplete intracellular Ca²⁺ store, respectively, and 100 µM ATP was added to the mucosal solution. Under this condition, an initial transient spike corresponding to the first peak was not observed (4.0 ± 1.9% versus ATP alone; n = 4; P < 0.001), although most of the subsequent responses were noted (Figure 3A). Next, we assessed the involvement of cAMP-PKA-dependent pathway in ATP-induced increase in Isc. Mucosal forskolin (1 µM) caused an increase in Isc, indicating that the cells have cAMP-dependent Cl channel (Figure 3B). Addition of 100 µM ADO caused an increase in Isc which was inhibited by pretreatment of cells with 10 µM H89, a specific PKA inhibitor (23), suggesting that PKA-dependent mechanism was operative (Figure 3C); and the H89-induced inhibition was also observed in the second peak (51.4 ± 8.2% versus ATP alone; n = 4; P < 0.05) and the following responses induced by ATP (Figure 3D).

View larger version (8K): [in this window] [in a new window]   Figure 3.   Ca2+- or cAMP-dependent effect on Isc of bovine cultured tracheal epithelium. (A) After treatment with EGTA (2 mM) for 10 min and successive TG (10 µM) for 5 min, ATP (100 µM) was added. Mucosal ATP did not show an initial transient spike corresponding to the first peak. (B) Mucosal forskolin (1 µM) caused an increase in Isc. (C) In the presence of H89 (10 µM), ADO (100 µM)-induced increase in Isc was inhibited. (D) H89 (10 µM) inhibited ATP (100 µM)-induced second peak, followed by a sustained response. Results are representatives of at least four independent experiments.

Extracellular ATP is known to be catabolized into adenosine diphosphate, AMP, or ADO by a ubiquitous ectoenzyme, ecto-5'-nucleotidase (24). Therefore, we addressed whether the ATP-induced increase in Isc was associated with ADO degraded from ATP. As shown in Figure 4A, simultaneous application of 100 µM UTP and 100 µM ADO mimicked the response to ATP. In the presence of 2 U/ml ADO deaminase (ADA), ADO-induced increase in Isc was markedly inhibited (Figure 4B), but the response to ATP was not affected (Figure 4C). Further, 300 µM ,-methyleneadenosine 5'-diphosphate (AMP-CP), an inhibitor of ectonucleotidase (25), did not alter ATP's action (Figure 4D). In addition, mucosal 100 µM ATPS, a metabolically stable analog of ATP, produced a similar response to ATP (Figure 4E). These findings indicate that the action of ATP is not due to the conversion of ATP to another form.

View larger version (8K): [in this window] [in a new window]   Figure 4.   Effects of various drugs on Isc of bovine cultured tracheal epithelium. (A) Mucosal UTP (100 µM) plus ADO (100 µM) showed an immediate increase in Isc containing the first and second peaks. (B) In the presence of ADA (2 U/ml), ADO (100 µM)-induced increase in Isc was markedly inhibited. (C) ADA (2 U/ml) did not affect the Isc response induced by ATP (100 µM). (D) AMP-CP (300 µM) did not affect the Isc response induced by ATP (100 µM). (E) Mucosal ATPS (100 µM) induced an initial transient increase in Isc, followed by a biphasic response. Results are representatives of at least four independent experiments.

Next, we examined which subtypes of prinoceptors were stimulated by ATP. After incubation of cells with 100 µM UTP, successive application of 100 µM UTP failed to cause an additional Isc elevation because of saturation of P2Y₂ receptor (formerly P_2U). However, subsequent addition of 100 µM ATP produced a further response with the plateau level within 100 s (Figure 5A), implying that this response was independent of P2Y₂ receptor stimulation. When 100 µM ATP was added after 100 µM ADO, the current was further increased, and then decreased within 30 s (Figure 5B). This result suggests that the spike response evoked by ATP was caused by the stimulation of prinoceptors other than P1 receptor. In the presence of 300 µM 8-SPT, an antagonist of P1 receptor, ADO- induced increase in Isc was inhibited (Figure 5C). 8-SPT showed a similar inhibitory effect on the second peak (57.8 ± 3.3% versus ATP alone; n = 4; P < 0.05) and the following response induced by ATP (Figure 5D), whereas it did not alter UTP-induced increase in Isc (data not shown).

View larger version (11K): [in this window] [in a new window]   Figure 5.   Effects of saturation of P2Y2 or P1 receptor and P1 receptor antagonist on ATP-induced Isc change in bovine tracheal epithelial cells. (A) After stimulation of the cells with UTP (100 µM), successive application of UTP (100 µM) failed to cause an additional Isc elevation. However, subsequent application of ATP (100 µM) showed a further Isc rise. (B) After stimulation of the cells with ADO (100 µM), successive application of ADO (100 µM) failed to cause an additional Isc elevation. However, subsequent application of ATP (100 µM) showed a rapid and transient increase in Isc. (C) In the presence of 8-SPT (300 µM), ADO (100 µM)-induced increase in Isc was inhibited. (D) In the presence of 8-SPT (300 µM), the second peak and the following response induced by ATP (100 µM) were inhibited. Results are representatives of at least four independent experiments.

PKA Assay

As shown in Figure 6, PKA activity was significantly elevated in the cells stimulated by forskolin, ATP, or ADO (n = 4; P < 0.01 for each), but not by UTP. In the presence of H89, ATP- or ADO-evoked PKA rise was significantly reduced (n = 4; P < 0.05 and 0.01, respectively), and was also inhibited by 8-SPT. Pretreatment of cells with EGTA and TG had little effect on the ATP- or ADO-induced increase in PKA activity.

View larger version (21K): [in this window] [in a new window]   Figure 6.   Effects of forskolin and prinoceptor agonists on PKA activity in bovine tracheal epithelial cells and their inhibition by H89, 8-SPT, or EGTA plus TG. After incubation with 1 µM forskolin (F) or 100 µM each of UTP, ATP, or ADO for 1 min, PKA activity was determined in the absence or presence of H89 (10 µM), 8-SPT (300 µM), or EGTA (2 mM) plus TG (10 µM). Data are means ± SE; n = 4 for each column. **P < 0.01, significantly different from the control values (C, no drug added). P < 0.05, P < 0.01, significantly different from the response to ATP or ADO alone, respectively.

Effects of Cl Channel Blockers on ATP- and UTP-Induced Increases in Isc

To assess the involvement of Cl channels in ATP- and UTP-induced increases in Isc, we examiend the effects of Cl channel blockers, DPC, DIDS, and glybenclamaide. In the presence of mucosal 100 µM DPC, the first and second peaks of Isc induced by ATP were reduced by approximately 80 and 85%, respectively (Figure 7). Similar inhibitory effects of DPC were observed with UTP-induced increase in Isc. The quantity of 100 µM DIDS, added to the mucosal solution, inhibited ATP- and UTP-induced first peaks of Isc by 50%. On the other hand, DIDS inhibited ATP-induced second peak of Isc only by approximately 30% compared with 45% inhibition of the UTP- induced second peak. In contrast, 100 µM glybenclamide inhibited the ATP-induced second peak by 40% but had little or no effect on ATP- and UTP-induced first peaks or UTP-induced second peak.

View larger version (23K): [in this window] [in a new window]   Figure 7.   Effects of Cl channel blockers, DPC (100 µM), DIDS (100 µM), and glybenclamide (100 µM) on the first (open columns) and second (hatched columns) peaks in Isc induced by ATP (100 µM) or UTP (100 µM). Values are expressed as the increase in Isc from the baseline levels. Data are means ± SE; n = 6 for each column. *P < 0.05, **P < 0.01, ***P < 0.001, significantly different form the control values (C, ATP, or UTP alone).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Our in vitro study showed that extracellular ATP increased Isc of bovine airway epithelium, which could be divided into at least two components reflecting first and second peaks. Although there have been a number of studies on transepithelial Cl transport by extracellular ATP and/or other nucleotides (5, 9, 14), few reports made reference to such an Isc having two distinct peaks (26, 27). We also found that the responses of Isc were different between ATP and UTP. The reason for this is not derived from the difference in [Ca²⁺]i response because these nucleotides increased [Ca²⁺]i in a similar way. In addition, the rank order of potency of agonists was UTP = ATP > 2-MeSATP > ,-MeATP = ADO, indicating P2Y₂ prinoceptor may be involved in the observed [Ca²⁺]i response (2, 10). Also, the Isc responses to 2-MeSATP and ,-MeATP were small or negligible (data not shown). These pharmacologic data are consistent with previous reports in airway epithelium (4, 5, 21), and the inhibition of Isc by EGTA plus TG suggests that Ca²⁺-sensitive Cl secretion was present in our cultured cells.

However, the Ca²⁺-dependent mechanism may be principally responsible for the first peak induced by ATP, and its substantial contribution to the second peak seems unlikely. Indeed, ATP was able to evoke transepithelial Cl transport through Ca²⁺-independent as well as Ca²⁺-dependent pathways. Stimulation of P2Y₂ receptor on airway epithelium activates PLC and produces inositol 1,4,5-trisphosphate (IP₃). Then the release of Ca²⁺ from internal stores occurs via IP₃ receptor, which results in the elevation of [Ca²⁺]i (5). Activation of PLC also leads to production of diacylglycerol, which in turn activates PKC (28). Accordingly, the involvement of PKC or Ca²⁺/calmodulin signaling mechanism in Cl channel activation is assumed (15, 29). These signal transductions downstream to PLC activation via P2Y₂ receptor are probably common pathways between ATP and UTP (10), but may be unable to explain the different response to Cl secretion. Therefore, we hypothesized that the second peak and the following response in Isc induced by ATP was evoked through the pathway other than PLC activation and its downstream. Because the PKA inhibitor H89, which inhibited ADO- induced increase in Isc, selectively reduced the second peak of Isc induced by ATP without affecting the first peak, cAMP-PKA-dependent pathway is likely involved in the response.

Nonetheless, it is uncertain whether ATP itself activates cAMP-PKA-dependent Cl secretion, because ADO, a degradation product of ATP, is known to potently stimulate transepithelial Cl secretion via P1 receptor (10, 30), and because a combined application of UTP and ADO to our cells was able to reconstitute the response induced by ATP. To clarify this possible concern, we tested the effects of ADA and AMP-CP on ATP-induced increase in Isc and found that neither compound influenced ATP's action. Further, ATPS elicited an increase in Isc having first and second peaks which were similar to that seen with ATP. These results confirmed that the second peak and the following response induced by ATP were not due to ADO or other ATP metabolites.

These findings indicate that the activation of ATP receptor may promote transepithelial Cl secretion through PKA-dependent pathway, but until now no convincing evidence of adenylate cyclase-coupled P2Y₂ receptor has been reported (10). Recent studies have demonstrated that ATP promotes cAMP formation via ATP receptor coupled to adenylate cyclase distinct from the P2Y₂ receptor in a newborn rat type II pneumocyte (31) and a neuronal cell line (32). This unique receptor is activated by both ADO and ATP in contrast to P1 and P2 receptors, and antagonized by methylxanthines such as 8-SPT that inhibit responses mediated by P1 but not P2 receptors. These characteristics are similar to those of the atypical prinoceptor, which was demonstrated in sympathetic nerves of rat caudal artery and provisionally termed P₃ prinoceptor by Shinozuka and colleagues (33). In the present study, stimulation with ATP in the presence of UTP or ADO at a nearly maximal effective concentration resulted in complete abolition of the first or second peak of Isc, respectively. It can thus be speculated that ATP and UTP share a common receptor (i.e., P2Y₂ receptor) whose stimulation produces the first peak, and that ATP and ADO act at the same receptor in the second peak. Further, the latter receptor was antagonized by 8-SPT, suggesting that this may be a P1-like or P₃ prinoceptor.

To confirm the existence of cAMP-PKA signaling pathway via ATP receptor, we examined whether ATP mediates PKA activation in bovine tracheal epithelium. As a result, both ATP and ADO, but not UTP, significantly increased PKA activities, an effect that was inhibited by H89 or 8-SPT, whereas removal of extra- and intracellular Ca²⁺ by EGTA and TG had little effect on the ATP-evoked PKA rise. Thus, ATP can stimulate PKA without requiring the increase in [Ca²⁺]i, and ATP may act on an adenylate cyclase-coupled receptor in addition to P2Y₂. These unusual pharmacologic characteristics of the ATP-mediated PKA activation were consistent with Isc studies, and the ATP receptor which activated PKA in bovine airway epithelium represents a novel purinoceptor subtype that cannot be explained by the P1/P2 receptor classification. To elucidate the precise identity of the adenylate cyclase- coupled ATP receptor, the molecular cloning is necessary.

On the basis of the findings described earlier, we speculated that different types of Cl channel may be activated by ATP and UTP. This concept was supported by the following findings: DPC eliminated ATP- and UTP-induced Isc responses to similar degree, and DIDS reduced both first peaks by approximately 50%, whereas DIDS more potently inhibited the second peak of UTP- than ATP-induced response. Further, glybenclamide significantly inhibited only the ATP-induced second peak. In our preliminary study, ADO-induced Isc was reduced by glybenclamide by approximately 70% but not by DIDS, implying that Cl currents mediated by cAMP-PKA-dependent pathway was DIDS-insensitive and glybenclamide-sensitive, and its characteristic is consistent with that of CFTR (9, 14, 15). That is, the component of ATP- but not UTP-induced second peak appeared to include CFTR-conducted Cl current. A series of results confirmed our hypothesis that the stimulation of P1-like receptor with ATP but not UTP mediates PKA-dependent signaling pathway, which activates glybenclamide-sensitive Cl channels such as CFTR and promotes Cl secretion across bovine cultured airway epithelium.

Recently, Schwiebert and coworkers (9) showed that CFTR activated by cAMP could conduct ATP out of the cell, and the released ATP in turn stimulated ORCC as an agonist. In addition to ATP regulation of CFTR-ORCC interaction, they observed that external ATP at high concentrations activated multiple types of whole-cell Cl currents in a time-dependent manner. Accordingly, we consider that multiplicity of ATP-activated Cl currents, which were probably mediated by several types of Cl channels, including Ca²⁺-activated Cl channel, ORCC, and CFTR, were involved in ATP-induced increase in Isc, thereby forming the dual peaks observed in this study. However, it remains unknown whether human airway epithelial cells also possess P1-like or P₃ receptor stimulated by both ATP and ADO.

In conclusion, external ATP elicited an increase in Isc with dual peaks associated with Cl secretion in bovine airway epithelium, and the current in the latter phase was larger than that evoked by stimulation of P2Y₂ receptor with UTP. The ATP-induced additional Cl current was glybenclamide-sensitive, and the involvement of PKA- dependent pathway via P1-like receptor is proposed as an underlying mechanism. Our present data may add new information that purinoceptors coupled to different signal transduction mechanisms would permit a variety of responses that are evoked by the same cell in response to extracellular ATP.