Extracellular ATP induces significant functional changes in a wide variety of cell types by interacting with P2 purinoceptors. In the vascular system, ATP can be released as a cotransmitter from perivascular nerves. Moreover, ATP, which is copackaged with serotonin in platelet granules, can be released locally in significant amounts during platelet aggregation, suggesting that significant amounts of extracellular ATP may accumulate locally at vascular sites of thrombus formation or infection/inflammation. With use of selective agonists, P2 purinoceptors can be divided into the subclasses P2X, P2Y, P2U, P2Z, and P2T(1,2). The P2X purinoceptor was originally defined by a descending rank order of the agonist potency α,β-methylene ATP ≥ β,γ methylene ATP ≫ ATP ≥ 2-methylthio-ATP (2Me-SATP) in intact tissues, and of ATP = 2MeSATP > α,β-methylene ATP in isolated cells, in which the interference from ecto-ATPase on agonist potency is reduced (3). The most potent agonist for the P2Y purinoceptor is 2MeSATP. P2 Purinoceptors recognizing both ATP and UTP with equal affinity, termed P2U purinoceptors, have also been identified in various cells. UTP, as ATP, is stored in the granules of platelets and is as effective as ATP in the regulation of vascular systems, such as endothelium-dependent relaxation (4). However, as compared with ATP, UTP-mediated signal transduction in cultured endothelial cells (EC) has been little studied.
Before the identification of the P2U purinoceptors, the vasodilator action of ATP on various vascular beds was believed to be due to the release of prostacyclin (PGI2) and endothelium-derived relaxing factor (EDRF) by the P2Y purinoceptors on the endothelium, as has now been demonstrated in cultured EC (5). After the studies showing that P2U purinoceptors are stimulated by both ATP and UTP, O'Connor and co-workers proposed in 1991 that two types of ATP receptor coexist in EC (6). Support for this idea derived from the finding that cultured bovine aortic EC (BAEC) contain a mixed population of P2Y and P2U purinoceptors, both linked to phosphoinositide (PI)-specific phospholipase C (PLC) (7,8). Cultured adrenal and glomerular EC have also been shown to express P2U purinoceptors, which can induce Ca2+ mobilization (9,10) and PI turnover (11). However, different results have been obtained in other vascular beds; e.g., in microvascular EC from the human frontal lobe, [3H]-inositol phosphate ([3H]IP) accumulation did not change in response to 2MeSATP, whereas both ATP and UTP induced PLC responses (12). In addition, atypical P2Y purinoceptors, which are highly sensitive to 2MeSATP (EC50 value 27 nM) and induce Ca2+ mobilization without activation of PLC, have been identified in brain capillary EC (13). Recent comparisons of potencies of 2-thioether ATP derivatives also suggest that the P2Y purinoceptors in rabbit aorta and mesenteric artery may belong to different subtypes (14). Together these findings suggest that the presence of the various P2 purinoceptor subtypes in different types of EC may reflect species or tissue differences. Therefore, the distribution of P2 purinoceptor subtypes and the signal transduction pathways mediated by each subtype, especially P2U, require further study. In the present work, we determined the subtypes of P2 purinoceptors in the bovine pulmonary artery endothelium (CPAE) by examining their pharmacological activation of PLC.
Culture medium, fetal calf serum, and 0.25% trypsin/EDTA were obtained from GIBCO (Grand Island, NY, U.S.A.). [3H]Myoinositol (20 Ci/mmol) was purchased from New England Nuclear (Boston, MA, U.S.A.), 2MeSATP, 2-methylthio-ADP (2MeSADP), 2-chloroadenosine triphosphate (2ClATP), reactive blue, and suramin were products of Research Biochemicals (Natick, MA, U.S.A.). ATP, adenosine-5′-O-(2-thiodiphosphate) (ADPβS), adenosine-5′-O-(3-thiotriphosphate) (ATPγS), UTP, UDP, ADP, AMP, adenosine, adenosine-5′-tetraphospho-5′-adenosine (AP4A), adenosine-5′-pentaphospho-5′-adenosine (AP5A), and phorbol 12-myristate 13-acetate (PMA) were purchased from Sigma Chemical (St. Louis, MO, U.S.A.). FPL67156 was provided by Dr. P. Leff (Fisons, plc., U.K.). Dowex AG 1X8 (formate form) was from Bio-Rad Laboratories. Fura 2/AM and pluronic F-127 from Molecular Probes.
CPAE (ATCC, Rockville, MD, U.S.A.) were grown in 35-mm Petri dishes in minimum essential medium (MEM) supplemented with 10% fetal calf serum (FCS) at 37°C in 5% CO2 humidified air. Cells were passaged with 0.25% trypsin/l mM EDTA. All experiments were performed on confluent monolayers.
PI hydrolysis was measured by accumulation of IP in the presence of 10 mM LiCl. Confluent cells on 35-mm Petri dishes were labeled for 24 h with [3H]myo-inositol (2.5 μCi/ml) in growth medium for 24 h. The cells were then washed with physiological saline solution (PSS in mM: NaCl 118, KCl 4.7, CaCl2 1.8, MgCl2 1.2, KH2PO4 1.2, glucose 11, and HEPES 20, pH 7.4), containing 10 mM LiCl and incubated at 37°C for 20 min. The agonists were then added, the cultures were incubated for 30 min, and the reaction was terminated by aspiration of the reaction solution and addition of ice-cold methanol. The cells were then scraped off and [3H]IP was isolated with an AG 1 × 8 (formate form) column and eluted with 0.2 N ammonium formate/0.1 N formic acid.
Intracellular Ca2+ measurement
Cells grown on 24-mm coverslips were loaded with Fura II/AM (3 μM) at 37°C for 1 h, and washed twice with PSS before fluorescence measurements were made on a PTI spectrofluorometer with dual excitation wavelengths of 340 nm and 380 nm and an emission wavelength of 510 nm. [Ca2+]i was calculated from the fluorescence ratio according to the equation of Grynkiewiez and colleagues (15).
Data are mean ± SEM of values from several separate experiments performed in duplicate. Student's t test was used to evaluate the statistical differences and p < 0.05 was considered significant. The pA2 values for suramin and reactive blue as indexes of antagonistic potencies were calculated from the EC50 values (the concentrations of agonists that produce 50% of maximal effect) of right-ward-shifted dose-response curves.
Nucleotide analogue-induced PI turnover and additivity between agonists
The accumulation of [3H]IP induced by nucleotide analogues was used to evaluate the PI-specific PLC activation. As shown in Fig. 1, the rank order of potency, based on the threshold concentrations, was ATPγS > ADPβS = 2MeSATP = 2MeSADP ≥ 2ClATP > UTP = ATP = ADP and the order of response produced by 100 μM agonist was ATPγS > ADPβS > ATP > ADP = UTP = 2MeSATP = 2MeSADP = 2ClATP. At 100 μM, none of the other analogues (UDP, CTP, AP4A, AP5A, α,β-methylene ATP, β,γ-methylene ATP, AMP, and adenosine) had any significant effect on IP formation (data not shown). The concentration-response curves of ATPγS and ADPβS were shifted to the left as compared with those of ATP and ADP, and their potencies were increased by 30- and 10-fold, respectively. These results indicate that the degradative ability of ecto-ATPase in CPAE could attenuate the responses of some hydrolyzable nucleotide analogues.
To evaluate the specific action of 2MeSATP and 2MeSADP on P2Y purinoceptors and of UTP on P2U purinoceptors, the additive effects of the responses with 2MeSATP, 2MeSADP, UTP, ATP, and ADPβS were determined. The PI response induced by 2MeSATP or UTP was additive to that for 2MeSATP (1-100 μM) or UTP (10-300 μM) (Fig. 2). In contrast, no such effect was noted between the maximal ATP or ADPβS response (100 μM) and UTP (100 μM) or 2MeSATP (100 μM) (Fig. 3A). The response induced to 2MeSADP (100 μM) was additive to that for UTP (100 μM) but not to that for 2MeSATP (100 μM) (Fig. 3A).
To rule out the possible interference of ecto-ATPase for classification of P2 purinoceptor subtypes, the additive effect of 2MeSATP and UTP was further studied in the presence of FPL 67156, an inhibitor of ecto-ATPase. Figure 3B shows that FPL 67156 (30 μM) could indeed increase the response to UTP, but not to 2MeSATP itself. In addition, the additivity between these two agonists was still evident in the presence of FPL 67156. In the desensitization experiments, the homologous desensitization of 2MeSATP and UTP was not obvious; only ≈20% decrease in IP formation was observed after 30- to 60-min pretreatment with each agonist.
Effects of pertussis toxin, extracellular Ca2+, and PMA
As shown in Fig. 4A, the PI responses to 2MeSATP (100 μM) and UTP (100 μM) were slightly sensitive to 24-h pretreatment with pertussis toxin, with a concentration of 500 ng/ml producing only ≈15% decrease in agonist response. Removal of extracellular Ca2+ 10 min before stimulation significantly reduced the PI responses to 2MeSATP and UTP by 25 ± 6 and 55 ± 8, respectively (Fig. 4B).
The effects of PMA-induced protein kinase C (PKC) activation on IP production induced by 2MeSATP and UTP were investigated. The responses to 2MeSATP and UTP were reduced in a dose-dependent manner after 10-min pretreatment of the cells with 1-300 nM PMA (Fig. 5), with the 2MeSATP response being more sensitive to PMA treatment. The IC50 values for PMA were 1.8 and 7.6 nM for 2MeSATP and UTP, respectively. In addition, the degree of inhibition induced by PMA was greater in the case of 2MeSATP (80%) than of UTP (50%).
Effects of P2 antagonists
Over the range of 3-30 μM, suramin, a nonselective and competitive inhibitor of P2X and P2Y purinoceptors, produced dose-dependent inhibition of the PI responses to 2MeSATP (30 μM) and UTP (100 μM). At the concentrations tested, 2MeSATP and UTP induced similar levels of PI turnover (approximately sevenfold that of controls); under these conditions suramin was more effective against the 2MeSATP-induced response (Fig. 6A). The pA2 values for suramin were 5.5 and 4.4 for 2MeSATP and UTP, respectively. Reactive blue (0.1-3 μM) also produced a concentration-dependent inhibition of the responses to 2MeSATP and UTP, whereas no further inhibition was induced by 10 μM reactive blue (Fig. 6B). The pA2 values for reactive blue were 6.32 for 2MeSATP and 5.7 for UTP.
Fluorometric measurements of [Ca2+]i
The concentration-dependent increases in [Ca2+]i induced by 2MeSATP or UTP are shown in Fig. 7. Both induced a rapid increase in [Ca2+]i, followed by a decrease toward a later sustained [Ca2+]i increase and occasionally by oscillation (Fig. 7A). That the [Ca2+]i response to the lower concentration (0.1 μM) of 2MeSATP was greater than that for UTP was consistent with the greater potency of 2MeSATP on PI turnover, but the maximum response to the agonists was similar (Fig. 7B). In the absence of extracellular Ca2+, the peak [Ca2+]i response to 2MeSATP (1 μM) or UTP (1 μM) was reduced by 60 ± 6% (n = 3) and 53 ± 10% (n = 3), respectively. Moreover, the later increase and Ca2+ oscillation caused by both agonists were abolished by removal of extracellular Ca2+(Fig. 8).
P2 Purinoceptors may play a physiological role in EC, since ATP and UTP are released into the bloodstream after injury to death of the cells that constitute the cardiovascular system. In addition, aggregated platelets, the adrenal medulla, and neurons can release ATP and UTP from intracellular organelles (16,17).
In the present work, our extensive studies on the effect of selectivity and cross-interactions of agonists on PI turnover provided evidence that P2Y and P2U purinoceptors coexist in CPAE, based on the following results. In terms of agonist selectivity, 2MeSATP and 2ClATP, P2Y-selective agonists (1,14), are ≈10-fold more potent than UTP (P2U-selective) and ATP (nonselective for P2 purinoceptors). The EC50 values for 2MeSATP and UTP were 5.6 and 47 μM, respectively, Two P2X-selective agonists, α,β-methylene ATP and β,γ-methylene ATP, had no effect. 2MeSADP is equipotent to 2MeSATP in inducing IP accumulation by stimulation of P2Y purinoceptors (14) of turkey erythrocytes, but can also activate platelet P2T purinoceptors (18), in contrast to 2MeSATP, which acts as an antagonist of ADP responses mediated by P2T purinoceptors in platelets (19). 2MeSADP and 2MeSATP are also equipotent in CPAE. AP4A and AP5A are diadenosine polyphosphates, described as present in the secretory granules of neural and nonneural cells, such as platelets (20,21). Although both can activate P2X in sensory neurons (22) and P2Y in chromaffin cells (23), the existence of a different type of purinoceptor (P2D) for diadenosine polyphosphates has been suggested (1,24). In CPAE, AP4A and AP5A had no stimulatory effect on PI turnover.
ATP is a nonselective agonist for P2Y and P2U purinoceptors, as is indicated by its higher response than 2MeSATP, 2MeSADP, 2ClATP, and UTP at 100-300 μM. Results show that ADP is similar to ATP in inducing higher IP formation; it is only twofold less potent than ATP. Recent studies have emphasized the degradative role of ecto-ATPase in decreasing the potencies of some hydrolyzable adenine nucleotides, and thus confused the classification of P2 purinoceptor subtypes (3). In support of the function of ecto-ATPase in CPAE, we noted that the concentration-response curves of ATPγS and ADPβS (two agonists more resistant to degradation) were shifted to the left as compared with ATP and ADP.
In terms of cross-interactions between agonists, the additivity observed between 2MeSATP and UTP (Figs. 2 and 3B) and the nonadditivity observed between ATP and 2MeSATP or UTP (Fig. 3A) suggests that 2MeSATP and UTP, in the concentration range tested, act on different receptors, i.e., P2Y and P2U purinoceptors, whereas ATP acts on both. The additivity between 2MeSATP and UTP was further evidenced even when ecto-ATPase was inhibited by FPL67156 (25)(Fig. 3B), strongly supporting the existence of independent PI responses medicated by P2Y and P2U receptors. In addition, the mRNA of P2Y and P2U was detected in CPAE by reverse transcription-polymerase chain reaction analysis (unpublished observations). 2MeSADP acted additively with UTP, but not with 2MeSATP. ADPβS has been used to characterize turkey erythrocyte P2Y purinoceptors (2). In this additivity experiment, we noted that the responses between ADPβS and 2MeSATP or UTP, each at 100 μM, are not additive, which suggests that at least at 100 μM, ADPβS is nonselective for P2Y and P2U purinoceptors.
To evaluate susceptibility to putative P2 antagonists, we used suramin, a P2X and P2Y purinoceptor antagonist; it was more effective against 2MeSATP that UTP. The derived pA2 values for P2Y (5.5) or P2U (4.4) in CPAE are entirely consistent with the values for P2U purinoceptors in C2Cl2 myotubes (4.41) and C6 glioma (4.4) and P2Y purinoceptors in bovine aortic collateral artery rings (5.45) and C6 glioma cells (5.4) (26-29). Reactive blue, an anthraquinone-sulfonic acid derivative, has been proposed as a selective P2Y antagonist (30,31). In the present study, it was fourfold more potent against P2Y- than P2U-mediated PI turnover in CPAE. To confirm the selective effects of reactive blue at P2 purinoceptors, we showed that muscarinic M3-mediated PI turnover in human SH-SY5Y neuroblastoma cells was unaffected by the pretreatment with reactive blue (data not shown).
P2Y and P2U purinoceptor-mediated PLC activation in CPAE showed differential PMA sensitivity. The effect of 2MeSATP on IP formation was more sensitive to short-term PMA treatment than was the UTP effect. Activation of PKC by short-term PMA treatment inhibits PI turnover in response to various agonists by modulating postreceptor events (32). One possible regulatory site is the G protein that modulates coupling between the receptor and PLC (33), G proteins act as substrates for PKC, and their phosphorylation results in uncoupling of the receptor-effector system (34,35). At present, the mechanism by which PMA exerts differential inhibition of P2U- and P2Y-mediated PI responses is unknown. Whether the different PTX-insensitive G proteins that couple to each receptor type are involved requires further investigation. With respect to possible regulatory sites on the receptors themselves, the amino acid sequences of cloned P2U purinoceptors from humans, rats, and mice and of P2Y purinoceptors from turkeys, chicken, mice, and rats were analyzed, but showed no possible PKC phosphorylation sites (2). Differential susceptibility of P2 purinoceptor subtypes to PMA has also been reported in BAEC (36,37).
Although these results support the concept of the presence of P2Y and P2U purinoceptors in CPAE, some of the characteristics of PI turnover in the CPAE differ from those in BAEC (7,31); e.g., in CPAE, pertussis toxin is a rather weak inhibitor of the UTP and 2MeSATP responses, suggesting coupling of P2Y and P2U purinoceptors to PLC through PTX-insensitive Gq/G11 family, whereas in BAEC, the P2U purinoceptors are coupled to a PTX-sensitive Gi2 protein and the P2Y purinoceptors are coupled to Gq/G11. At present, we can not explain the discrepancy of P2U responses in these two EC types from different vascular beds of bovine. However, our results are consistent with those observed for human epithelial P2U purinoceptors. According to the report of Parr and associates (38), the ATP-induced intracellular Ca2+ mobilization through IP3 formation is inhibited only 20-30% by PTX in P2U-stably expressed cells. In conclusion, we showed the existence of P2Y and P2U purinoceptors in bovine pulmonary artery endothelial cells, which are coupled to PLC by a pertussis toxin-insensitive G protein and responsible for Ca2+ mobilization.
Acknowledgment: The study was supported by Grant No. DOH83-HR-301 from the National Institutes of Health, Center for Cardiovascular Research and Grant No. CRC 85-T04 from IBMS.
1. Fredholm BB, Abbracchio MP, Burnstock G, et al. Nomenclature and classification of purinoceptors. Pharmacol Rev
2. Harden TK, Boyer JL, Nicholas RA. P2
-Purinergic receptors: subtype-associated signaling responses and structure. Annu Rev Pharmacol Toxicol
3. Kennedy C, Leff P. How should P2X
purinoceptors be classified pharmacologically? Trends Pharmacol Sci
4. Martin W, Cusack NJ, Carleton JS, Gordon JL. Specificity of P2
-purinoceptor that mediates endothelium-dependent relaxation of the pig aorta. Eur J Pharmacol
5. Boeynaems JM, Pearson JD. P2
Purinoceptors on vascular endothelial cells: physiological significance and transduction mechanisms. Trends Pharmacol Sci
6. O'Connor SE, Dainty IA, Leff P. Further subclassification of ATP receptors based on agonist studies. Trends Pharmacol Sci
7. Motte S, Pirotton S, Boeynaems JM. Heterogeneity of ATP receptors in aortic endothelial cells. Involvement of P2Y
receptors in inositol phosphate response. Circ Res
8. Wilkinson GF, Purkiss JR, Boarder MR. The regulation of aortic endothelial cells by purines and pyrimidines involves co-existing P2Y
-purinoceptors and nucleotide receptors linked to phospholipase C. Br J Pharmacol
9. Castro E, Tome AR, Miras-Portugal MT, Rosario LM. Single-cell fura-2 microfluorometry reveals different purinoceptor subtypes coupled to Ca2+
influx and intracellular Ca2+
release in bovine adrenal chromaffin and endothelial cells. Pflugers Arch
10. Briner VA, Kern F. ATP stimulates Ca2+
mobilization by a nucleotide receptor in glomerular endothelial cells. Am J Physiol
11. Allsup DJ, Boarder MR. Comparison of P2
purinergic receptors of aortic endothelial cells with those of adrenal medulla: evidence for heterogeneity of receptor subtype and inositol phosphate response. Mol Pharmacol
12. Purkiss JR, West D, Wilkes LC, et al. Stimulation of phospholipase C in cultured microvascular endothelial cells from human frontal lobe by histamine, endothelin and purinoceptor agonists. Br J Pharmacol
13. Vigne P, Feolde E, Breittmayer JP, Frelin C. Characterization of the effects of 2-methylthio-ATP and 2-chloro-ATP on brain capillary endothelial cells: similarities to ADP and differences from ATP. Br J Pharmacol
14. Fisher B, Boyer JL, Hoyle CHV, et al. Identification of potent, selective P2Y
-purinoceptor agonists: structure-activity relationships for 2-thioether derivatives of adenosine 5′-triphosphate. J Med Chem
15. Grynkiewiez G, Poenie M, Tsien RM. A new generation of Ca2+
indicators with greatly improved fluorescence properties. J Biol Chem
16. Goetz K, Da Prada M, Pletscher A. Adenine-, guanine- and uridine-5′ -phosphonucleotides in blood platelets and storage organelles of various species. J Pharmacol Exp Ther
17. Keppler D, Rudiger J, Decker K. Enzymatic determination of uracil nucleotides in tissues. Anal Biochem
18. Cusack NJ, Hourani S. Design, syntheses and pharmacology of ATP analogues selective for subtypes of P2
-purinoceptors. Nucleosides Nucleotides
19. Cusack NJ, Hourani SMO. Specific but non competitive inhibition by 2-alkylthio analogues of adenosine 5′-monophosphate and adenosine 5′-triphosphate of human platelet aggregation induced by adenosine 5′-diphosphate. Br J Pharmacol
20. Flodgaard H, Klenow H. Abundant amounts of diadenosine 5′-5‴-P1
-tetraphosphate are present and releasable, but metabolically inactive in human platelets. Biochem J
21. Rodriguez del Castillo A, Torres M, Delicado EG, Miras-Portugal MT. Subcellular distribution studies of diadenosine polyphosphates-AP4
A and AP5
A-in bovine adrenal medulla: presence in chromaffin granules. J Neurochem
22. Kristhal OA, Marchenko SM, Obukhov AG, Volkova TM. Receptors for ATP in rat sensory neurones: the structure-function relationship for ligands. Br J Pharmacol
23. Pintor J, Torres M, Castro E, Miras-Portugal MT. Characterization of diadenosine tetraphosphate (AP4
A) binding sites in cultured chromaffin cells: evidence for a P2Y
site. Br J Pharmacol
24. Hoyle CHV. Pharmacological activity of adenine dinucleotides in the periphery: possible receptor classes and transmitter function. Gen Pharmacol
25. Crack BE, Pollard CE, Beukers MW, et al. Pharmacological and biochemical analysis of FPL 67156, a novel, selective inhibitor of ecto-ATPase. Br J Pharmacol
26. Henning RH, Duin M, Hertog A, Nelemans A. Activation of the phospholipase C pathway by ATP is mediated exclusively through nucleotide type P2
-purinoceptors in C2C12 myotubes. Br J Pharmacol
27. Lin WW, Chuang DM. Different signal transduction pathways are coupled to the nucleotide receptor and the P2Y
receptor in C6
glioma cells. J Pharmacol Exp Ther
28. Wilkinson GF, Mckechnie K, Dainty IA, Boarder MR. P2Y
purinoceptor and nucleotide receptor-induced relaxation of precontracted bovine aortic collateral artery rings: differential sensitivity to suramin and indomethacin. J Pharmacol Exp Ther
29. Boyer JL, Zohn IE, Jacobsen KA, Harden TK. Differential effects of P2
-purinoceptor antagonists on phospholipase C-and adenylyl cyclase-coupled P2Y
-purinoceptors. Br J Pharmacol
30. Burnstock G, Warland JJ. P2
-Purinoceptors of two subtypes in the rabbit mesenteric artery: reactive blue 2 selectively inhibits responses mediated via the P2Y
- but not the P2X
-purinoceptor. Br J Pharmacol
31. Connoly GP, Harrison PJ. Reactive blue 2 discriminates between responses medicated by UTP and those evoked by ATP or α,β-methylene-ATP on rat sympathetic ganglia. Eur J Pharmacol
32. Nishizuka T. Studies and perspectives of protein kinase C. Science
33. Orellana S, Solski PA, Brown JH. Guanosine 5′-O
-(thiotri-phosphate)-dependent inositol triphosphate formation in membranes is inhibited by phorbol ester and protein kinase C. J Biol Chem
34. Katada T, Gilman AG, Watanabe Y, Bauer S, Jakobs KH. Protein kinase C phosphorylates the inhibitory guanine-nucleotide-binding regulatory component and apparently suppresses its function in hormonal inhibition. Eur J Biochem
35. Zick Y, Sagi-Eisenberg K, Piner M, Gierschik P, Spiegel AM. Multisite phosphorylation of the α subunit of transductin by the insulin receptor kinase and protein kinase C. Proc Natl Acad Sci USA
36. Purkiss JR, Wilkinson GF, Boarder MR. Differential regulation of inositol 1,4,5-trisphosphate by co-existing P2Y
-purinoceptors and nucleotide receptor on bovine aortic endothelial cells. Br J Pharmacol
37. Communi D, Raspe E, Pirotton S, Boeynaems JM. Coexpression of P2Y
purinoceptors on aortic endothelial cells. Comparison of cell localization and signaling pathways. Circ Res
38. Parr CE, Sullivan DM, Paradiso AM, et al. Cloning and expression of human P2U
nucleotide receptor, a target for cystic fibrosis pharmacotherapy. Proc Natl Acad Sci USA