Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Three members of the P2Y receptor subfamily, the P2Y₂, P2Y₄, and P2Y₆ receptors, are potently stimulated by uridine nucleotides. The P2Y₂ receptor, which has been cloned from human, rat, and mouse tissues, is activated equipotently by ATP and UTP (Lustig et al., 1993; Parr et al., 1994; Rice et al., 1995) but not by ADP or UDP (Nicholas et al., 1996). The P2Y₆ receptor has been cloned from human and rat tissues and in both cases is potently activated by UDP, and weakly by ADP (Chang et al., 1995; Communi et al., 1996; Nicholas et al., 1996). The rat and human P2Y₄ receptors share 83% identity at the amino acid level, but exhibit different nucleotide preferences. The rat P2Y₄ receptor is nearly equipotently activated by UTP and ATP (Bogdanov et al., 1998; Webb et al., 1998; Kennedy et al., 2000). In contrast, the human P2Y₄ receptor is activated by UTP but not by ATP (Communi et al., 1995; Nguyen et al., 1995). Indeed ATP is a relatively potent competitive antagonist of the human P2Y₄ receptor (Kennedy et al., 2000).

Extracellular uridine nucleotides regulate several components of airway epithelial mucociliary clearance. Mucosal administration of UTP (Mason et al., 1991), and to a lesser extent UDP (Lazarowski et al., 1997c), results in Ca²⁺-stimulated chloride secretion that is independent of the cystic fibrosis transmembrane conductance regulator (CFTR), the cyclic AMP-regulated epithelial chloride channel that is defective in cystic fibrosis. Mucosal UTP also increases ciliary beat frequency and mucin release (Davis et al., 1992; Lethem et al., 1993). These observations, together with data demonstrating extracellular UTP accumulation in primary cultures of nasal epithelial cells and in vivo in nasal secretions (Donaldson et al., 2000), have reinforced the concept that P2Y receptors are central coordinators of airway mucociliary clearance. Unambiguous identification of the relevant receptor(s) that control(s) airway epithelial cell responses has been difficult to established due to lack of selective P2Y receptor agonists and antagonists. However, the nucleotide selectivity of the receptor involved in regulation of ion transport responses in the airways suggests a major role for the P2Y₂ receptor with lesser involvement of the P2Y₆ receptor. This hypothesis was confirmed in recent studies demonstrating that the ion secretory actions of mucosal UTP and ATP in nasal and tracheal epithelia were essentially abolished in the P2Y₂(/) mouse, with only a small response to mucosal UDP preserved (Cressman et al., 1999).

Nucleotide regulation of ion transport processes also has been observed in nonairway epithelia. UTP stimulates transepithelial anion secretion (Clarke et al., 1999), ³⁵Cl efflux (Chinet et al., 1999), and bicarbonate secretion (Clarke et al., 2000) in gallbladder. UTP and/or ATP promote ductal ion transport in cholangiocytes (Roman and Fitz, 1999), regulate acid/base transport in biliary epithelial cells (Zsembery et al., 1998), electrogenic ion secretion in Sertoli cells (Ko et al., 1998), Cl secretion in endometrial epithelial cells (Chan et al., 1997), and K⁺ secretion in distal colonic mucosa (Kerstan et al., 1998). Most of these nucleotide actions in nonairway epithelia have been attributed to activation of the P2Y₂ receptor. However, studies with the P2Y₂(/) mouse indicate that additional or alternative receptors control ion transport responses to uridine nucleotides in the gastrointestinal system. For example, the potent actions of UTP and UDP on Cl secretion in jejunal and gallbladder epithelia, respectively, were unaffected after disrupting the P2Y₂ receptor gene (Cressman et al., 1999). Although the P2Y₄ and P2Y₆ receptors may be the logical candidates to account for the residual uridine nucleotide effects in P2Y₂(/) epithelia, the molecular and pharmacological properties of these murine orthologs are not known. To determine whether the nucleotide preferences of the (m)P2Y₄ and (m)P2Y₆ receptors could account for the effects of UTP and UDP in jejunal and gallbladder epithelium, we cloned these murine receptors and functionally expressed them in a null cell line to define their nucleotide selectivity for promotion of second messenger production. Moreover, we tested for the expression of P2Y₄ and P2Y₆ receptor transcripts in freshly isolated jejunum and in gallbladder epithelia cells, respectively, and ion transport studies were carried out on primary tissues to investigate the potential of these receptors for correcting defective Cl secretory transport in diseased CF gastrointestinal epithelia.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Cloning and Sequencing. Gallbladder epithelial cells from P2Y₂(/) mice (Cressman et al., 1999) were grown as a polarized primary culture on collagen-coated filters over a feeder layer of fibroblasts, as described (Kuver et al., 1997). Poly(A⁺) RNA (1 µg) was extracted from cultured P2Y₂(/) gallbladder epithelial cells following published protocols (Burch et al., 1995) and transcribed into cDNA using SuperScript RNase H⁺ reverse transcriptase (RT; Life Technologies, Gaithersburg, MD). The RT reaction (300 units RT in 53-µl final volume) proceeded for 1 h at 37°C. Degenerate oligonucleotide primers [forward, 5'-AGCATCCTCTTCCTCACCTGCATC/TAGC-3'; reverse, 5'-GGGTC(C/A)AG(G/T)(A/C)(A/C)(G/A)CTGTT(G/T/C)GC(A/G)CTGGC (G/T/C)A (A/G)GGCCG-3'] based on the coding sequence of the third and sixth putative transmembrane domains of previously cloned (h, r, m)P2Y₂, (h, r)P2Y₄, and (h, r)P2Y₆ receptors, were used to amplify by PCR (Taq polymerase; Life Technologies) a 495-bp fragment in mouse P2Y₂(/) gallbladder epithelial cell cDNA (named GBY6), which was found by sequence analysis (Fig. 5) to share high homology with a fragment of the rat P2Y₆ receptor gene. Genomic DNA from the P2Y₂(/) mouse was kindly provided by Dr Jean-Etienne Fabre (Department of Medicine, University of North Carolina, Chapel Hill, NC). A forward primer [5'-GTTCAAGTTCATCCTGTTGCC-3'] comprising the 84 to 104 bp starting at the 5' end of the coding sequence of the rat P2Y₄ receptor gene, and reverse primer [5'-GGGTCAAGGAAGCTGTTTGCA-3'] comprising the 754 to 734 region of the coding sequence of the rat P2Y₄ receptor, were used to amplify an ~670-bp-long sequence that was ~90% homologous to the corresponding sequence of the rat P2Y₄ receptor gene (hereafter named GENY4). The PCR products were ligated using a TA-cloning kit (Invitrogen, San Diego, CA) and the resulting plasmids were amplified in transformed One-Shot cells (Invitrogen) and subsequently purified using a Qiagen kit. Relevant species were identified by automatic sequencing. GENY4 and GBY6 clones were labeled by random priming and used to screen a mouse  phage 129SvEv genomic library (University of North Carolina Animal models Core Facility) with standard conditions at high stringency. Positive phages were purified and fragments subcloned into pBluescript (Stratagene, La Jolla, CA) for sequencing. Subclones positive for either P2Y₄ or P2Y₆ receptor DNA were sequenced completely through the coding regions by primer walking (University of North Carolina Automated Sequencing Facility). Sequencing confirmed the expected intronless nature of the two transcripts. Primers were designed containing either the ATG start or the stop codons of each gene and the entire genes were amplified with restriction sites to allow for cloning into the retroviral vectors.

In Situ Hybridization. Sections of gallbladder and jejunum were excised from 7- to 8-week-old mice of wild-type, mixed (BL6, 129, DBA) lineage; mounted in a rectangular, polypropylene embedding mold filled with Optimal Cutting Temperature medium (SAKURA Fine Tec. Co., Torrance, CA); and frozen on dry ice. Frozen blocks of tissues were sectioned (7 µm in thickness) and mounted on glass slides. Tissue sections were fixed in 4% paraformaldehyde, dehydrated, and stored at 20°C in an airtight box. Antisense and sense probes were obtained by PCR from GENY4 and GBY6 cDNA, cloned into PCR II vector (Invitrogen), and linearized with EcoRV or HindIII. In situ hybridization was carried out as described previously (Rochelle et al., 2000).

Expression of Recombinant Receptors in 1321N1 Cells. Retroviral vector-containing plasmids L(M)P2Y4USN and L(M)P2Y6USN were constructed by insertion of the cloned cDNAs into pLXSN. An amphotrophic packaging cell line, PA317, was used to produce the L(M)P2Y4USN and L(M)P2Y6USN vectors and a control vector containing only the neomycin-resistance (neo^r) gene (LN). Human astrocytoma cells (1321N1) were infected with L(M)P2Y4USN, L(M)P2Y6USN, or LN and selected with G418, as described previously (Parr et al., 1994). [³H]Inositol phosphate formation in [³H]myo-inositol-labeled cells and intracellular calcium ([Ca²⁺]_i) mobilization in Fura 2-loaded cells were measured as previously described (Lazarowski et al., 1997a,b).

Bioelectric Measurements. Adult mice [wild-type and CFTR(/) (Snouwaert et al., 1992)] of both sexes were used in this study. The techniques for mounting the intestinal and gallbladder tissues on Ussing chambers, and for performing the electrical measurements under short circuit (I_sc) conditions, have been described previously (Cressman et al., 1999).

Statistics. Data are presented as the mean value ± S.E.M. and considered significantly different (*) when p > 0.001 (unpaired t test).

Materials. ATP, UTP, CTP, and GTP were from Amersham Pharmacia Biotech (Piscataway, NJ). ITP was from Sigma (St. Louis, MO). ADP, UDP, and hexokinase were from Boehringer-Mannheim (Indianapolis, IN). Pertussis toxin was from Research Biochemicals International (Natick, MA). Fura-2 was obtained from Molecular Probes (Eugene, OR). [³H]myo-Inositol (specific activity 20 Ci/mmol) was purchased from American Radiolabeled Chemicals Inc. (St. Louis, MO). -³⁵S-UTP for the in situ hybridization studies was from New England Nuclear (Boston, MA). -³⁵S-dCTP for the phage library screen was from ICN Pharmaceuticals (Costa Mesa, CA).

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

P2Y₄ and P2Y6 Receptor Sequences. A mouse genomic clone that hybridized with the GENY4 probe encoded an open reading frame of 1083 bp with predicted 361 amino acids (Fig. 1A) that is 92 and 81% homologous to the rat (Bogdanov et al., 1998; Webb et al., 1998) and human (Communi et al., 1995; Nguyen et al., 1995) P2Y₄ receptor, respectively. We conclude based on sequence identity that this is the mouse P2Y₄ receptor. The sequence reported here is identical to a murine P2Y₄ receptor sequence recently submitted to the GenBank by Suarez-Huerta et al. (2001). A full-length clone obtained by screening the mouse genomic library with the GBY6 probe was found by sequence analysis to contain a 987-bp open reading frame. The nucleotide sequence of the murine P2Y₆ receptor reported in this study was deposited in the GenBank under the accession number AF298899. The deduced amino acid sequence of the GBY6-hybridizing clone (Fig. 1B) is 91 and 83% identical to the rat (Chang et al., 1995) and human (Communi et al., 1996) P2Y₆ receptors, respectively, indicating that this murine clone represents the mouse P2Y₆ receptor. By analogy to all G protein-coupled receptors, hydropathy analysis indicated that the predicted amino acid sequences of the (m)P2Y₄ and (m)P2Y₆ receptors encoded seven stretches of hydrophobic amino acids (putative transmembrane regions) that span between an extracellular amino terminus and the cytosolic carboxyl terminus (Kyte-Doolittle scale, Mac Vector software; Eastman-Kodak, Rochester, NY), with potential extracellular N-glycosylation sites and intracellular protein kinase C phosphorylation sites (Fig. 1).

View larger version (54K): [in this window] [in a new window]   Fig. 1.   Deduced amino acid sequences corresponding to the murine P2Y4 (A) and P2Y6 (B) receptors. The putative seven transmembrane regions are underlined. Potential N-glycosylation sites are indicated by *, and putative protein kinase C phosphorylation sites are indicated by .

Functional Expression of the (m)P2Y₄ and (m)P2Y₆ Receptors in 1321N1 Cells. Wild-type 1321N1 human astrocytoma cells do not express endogenous P2 receptors (Parr et al., 1994), and we observed no inositol phosphate formation or calcium mobilization in response to nucleotides in control experiments with LN (empty vector)-infected 1321N1 cells (data not shown). Since 1321N1 cells express an ectonucleotidase activity that hydrolyzes UTP and UDP with V_max = 0.6 to 1.1 nmol/min/million cells (Lazarowski et al., 1997a), 100 µM UTP and UDP (50 nmol/well; ~0.4 million cells/well) were used to examine the time courses for agonist-stimulated inositol phosphate formation in (m)P2Y₄ and (m)P2Y₆-1321N1 cells, respectively. P2Y₄ and P2Y6 receptor promoted a steady accumulation of inositol phosphates for up to 60 min (Figs. 2A and 3A). Although the rate of inositol phosphate accumulation promoted by UTP in P2Y₄ receptor-expressing cells appeared faster during the initial 2 min (Fig. 2A), the overall inositol phosphate accumulation was essentially linear during the 60-min incubation period (r² = 0.994). Both ATP and UTP activated the mouse P2Y₄ receptor with similar potencies. Although less potent, activation also was observed with ITP, GTP, and CTP (Fig. 2B). The EC₅₀ values for nucleotide-promoted inositol phosphate formation in (m)P2Y₄-1321N1 cells were as follows: UTP, 260 ± 84 nM; ATP, 435 ± 130 nM; ITP, 2 ± 0.6 µM; GTP, 7 ± 2 µM; and CTP, 25 ± 3 µM (mean ± S.E.M. values are from at least three experiments performed in duplicate). The (m)P2Y₄ receptor exhibited absolute triphosphonucleotide preferences, i.e., UDP and ADP [pretreated and assayed in the presence of hexokinase (Nicholas et al., 1996; Lazarowski et al., 1997c)] did not activate the (m)P2Y₄ receptor (Fig. 2B).

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Expression of the mP2Y4 receptor in 1321N1 cells. A, time course for inositol phosphate accumulation in the presence of 100 µM UTP (the data represent the mean ± S.E.M. from three experiments with duplicate samples). B, concentration-effect relationships for nucleotide-stimulated inositol phosphate formation were measured after 30 min and the results expressed as the percentage of the response observed with 3 µM UTP (control, 1316 ± 318 cpm; 3 µM UTP, 9793 ± 1368 cpm; mean values are from at least three experiments performed in triplicate, S.E.M. bars are omitted for clarity. C, [Ca2+]i changes in response to nucleotides in (m)P2Y4-1321N1 cells (the tracings are representative of at least three independent experiments).

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Expression of the mP2Y6 receptor in 1321N1 cells. A, time course for inositol phosphate accumulation in the presence of 100 µM UDP. B, concentration-effect relationships for nucleotide-stimulated inositol phosphate formation were measured after 20 min and the results expressed as the percentage of the response observed with 1 µM UDP (control, 1957 ± 179 cpm; 1 µM UDP, 9855 ± 527 cpm). In A and B, mean values and error bars are expressed as in Fig. 2. C, [Ca2+]i changes in response to nucleotides in (m)P2Y6-1321N1 cells (the tracings are representative of at least three independent experiments).

Expression of P2Y₄ Receptor mRNA in Jejunum. The nucleotide selectivity of the cloned (m)P2Y₄ receptor (UTP > ATP) most closely resembles the nucleotide selectivity of nucleotide-stimulated ion transport responses in murine P2Y₂(/) jejunum (Cressman et al., 1999), although a low potency but robust effect of ADP in jejunum suggested an additional purine nucleotide receptor. To test for expression of the P2Y₄ receptor transcripts in freshly excised mouse jejunum, in situ hybridization studies were carried out. As illustrated in Fig. 4A, P2Y₄ receptor mRNA is nonhomogeneously distributed through the entire length of the villi. No signal was detected in the muscle layer, suggesting that the P2Y₄ receptor is expressed in the jejunal epithelia. A detailed tissue distribution study by in situ hybridization of murine P2Y receptors will be reported elsewhere (L. Rochelle, personal communication).

View larger version (36K): [in this window] [in a new window]   Fig. 4.   Detection of the P2Y4 receptor mRNA by in situ hybridization in jejunum. Freshly excised jejunum was probed against antisense (center) or sense (right) GENY4-derived probe. The hematoxylin/eosin staining is shown on the left. Magnification, 100×.

Detection of the P2Y₆ Receptor mRNA in Gallbladder Epithelial Cells. The nucleotide selectivity for agonist-stimulated I_sc changes reported in murine P2Y₂(/) gallbladder epithelium (UDP > UTP) (Cressman et al., 1999) was similar to that observed with (m)P2Y₆-1321N1 cells (Fig. 3). Because a weak in situ hybridization signal for the P2Y₆ receptor mRNA was observed in freshly excised gallbladder (data not shown), expression of the P2Y₆ receptor was verified by RT-PCR analysis of cultured primary gallbladder epithelial cells. A 495-bp fragment was amplified, which correspond to the nucleotide fragment stretching from base 364 to 858 of the mP2Y₆ receptor-encoding sequence (Fig. 5).

View larger version (58K): [in this window] [in a new window]   Fig. 5.   Detection of P2Y6 receptor mRNA in murine gallbladder epithelial cells by RT-PCR. mRNA from gallbladder epithelial cells was isolated and reverse transcribed to cDNA as indicated under Experimental Procedures. Left, 557-bp PCR product (indicated by the arrow) was visualized after electrophoresis on 1% agarose gel and ethidium bromide staining. PCR [35× (1 min at 94°C + 2 min at 65°C + 2 min at 72°C)] was performed with 2 µl of gallbladder cDNA (b) or 2 µl of a negative control from a transcriptase reaction without RT (c). Nucleic acid size standards (indicated in kb) are shown in a. Right, nucleotide sequence of the PCR product indicating (underlined) the oligonucleotide primers flanking the 495-bp mP2Y6 receptor mRNA fragment.

Effect of Nucleotides on Ion Transport in CF Jejuna and Gallbladder. The defective cyclic AMP-regulated Cl channel CFTR, the hallmark for CF gastrointestinal dysfunction, can be bypassed by activation of Ca²⁺-dependent chloride channels. However, this alternative pathway is not present in the small and large intestine (for review, see Grubb and Gabriel, 1997). To test whether uridine nucleotide activation of jejunum and gallbladder epithelial ion transport (Cressman et al., 1999) reflected CFTR-dependent or CFTR-independent mechanisms, we examined the ability of UTP and UDP to promote I_sc changes in freshly excised jejunum and gallbladder tissues, respectively, from CFTR(/) mice (Fig. 6). In normal jejunal epithelia, luminal UTP (100 µM) induced a marked increase in I_sc, whereas virtually no effect of UTP was detected in tissues from CF mice (Fig. 6A). In contrast, luminal administration of UDP (100 µM) to gallbladder epithelia resulted in secretory responses that were preserved in the CF compared with normal mouse (Fig. 6B).

View larger version (25K): [in this window] [in a new window]   Fig. 6.   Bioelectrical measurements. Effect of mucosal addition of 100 µM UTP or 100 µM UDP on Isc in jejunum and gallbladder, respectively, from wild-type and CF mice. Representative tracings are illustrated in the top graphs. Changes in Isc (measured as peak values) from three independent experiments are indicated in the bottom graphs (mean ± S.D.).

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Two mouse gene products that are molecularly related (~90% identical) to the human and rat P2Y₄ and the P2Y₆ receptor genes were isolated and functionally expressed, and most likely they represent the murine P2Y₄ and P2Y₆ receptor orthologs. Pharmacological data indicate that the mouse P2Y₄ and P2Y₆ receptors are, in general, functionally similar to their rat and human orthologs. Both the (m)P2Y₄ and (m)P2Y₆ receptors couple to phospholipase C activation and calcium mobilization via pertussis toxin-insensitive, i.e., G_q/11, mechanisms. The (m)P2Y₄ receptor strictly recognizes nucleoside triphosphates, whereas the (m)P2Y₆ receptor is most potently activated by UDP and is only very weakly activated by ADP.

A major difference between the mouse and human P2Y₄ receptors is the effect of ATP on second messenger production. We previously reported that ATP had a weak and delayed effect on inositol phosphate formation in 1321N1 cells expressing the (h)P2Y₄ receptor (Lazarowski et al., 1997b), and that ATP did not trigger Ca²⁺ mobilization in (h)P2Y₄-1321N1 cells unless UDP was present (Lazarowski et al., 1997a,b). Because 1321N1 cells express an endogenous ectonucleoside diphosphokinase activity that transfers the -phosphate of exogenous ATP to endogenous UDP, resulting in formation of UTP (Lazarowski et al., 1997a, 2000), the effects of ATP on (h)P2Y₄-1321N1 cells likely reflected metabolic conversion of ATP to UTP rather than a true ATP/agonist effect. ATP interactions with the murine P2Y₄ receptor were similar to interactions with the rat P2Y₄ receptor (Bogdanov et al., 1998; Webb et al., 1998; Kennedy et al., 2000). That is, ATP is nearly as potent as UTP in promoting inositol phosphate formation and Ca²⁺ mobilization, and the [Ca²⁺]_i responses elicited by ATP are identical in time and magnitude to UTP responses. Moreover, the (m)P2Y₄ receptor had low selectivity within nucleoside triphosphates; ITP, GTP, and CTP (in this potency order) were full agonists on the (m)P2Y₄ receptor. Thus, the (m)P2Y₄ receptor is pharmacologically related to the (r)P2Y₄ and to the (h, r, m)P2Y₂ receptors, together comprising a functional P_2U-receptor group (UTP = ATP); on the other hand, the (h)P2Y₄ receptor (UTP ATP) along with (h, r, m) P2Y₆ receptors (UDP ADP) represent true uridine nucleotide-selective receptors.

The physiological role of the P2Y₄ receptor has remained elusive. The P2Y₄ receptor originally was cloned from a human placenta cDNA library, and P2Y₄ receptor transcripts could not be detected (by Northern blotting) in heart, liver, brain, testis, or kidney (Communi et al., 1995). Recently, RT-PCR screening for P2Y receptor in a variety of epithelial cell types revealed P2Y₄ receptor mRNA in 6CFSMEo, CFPAC-1, 16HBE14o, HASMSCl, and HAEo cells, although functional evidence was found only in the bronchial submucosal gland-derived 6CFSMEo cell line (Communi et al., 1999). Coincidentally, ion transport studies with freshly excised jejunal epithelial cells have indicated that UTP and ATP were nearly equipotent in promoting robust Cl secretory responses in the P2Y₂(/) mouse, whereas ADP was a weaker agonist and UDP had no effect (Cressman et al., 1999). These observations suggested a major role for the P2Y₄ receptor in this tissue and that an adenine nucleotide receptor (P2Y₁/P2Y₁₁/P2X?) may be additionally expressed. Our present study not only confirms that the P2Y₄ receptor is the most likely candidate for the UTP-promoted responses in jejunum but also provides evidence for expression of mRNA for P2Y₄ receptor in this tissue. To our knowledge, this is the first observation of a physiologically relevant function (e.g., intestinal ion transport) that can be associated with activation of the P2Y₄ receptor. Intestinal epithelium lacks the Ca²⁺-activated Cl channel that is typical of many other exocrine epithelia, e.g., airways and pancreas (Clarke et al., 1994; Grubb, 1997), whereas the gallbladder epithelium coexpresses the apical Ca²⁺-activated Cl channel with CFTR. CFTR is the prevalent mucosal Cl channel expressed in intestinal epithelial cells. Although CFTR is a cyclic AMP-activated Cl channel, it is also activated by protein kinase C (Jia et al., 1997) and by cyclic GMP-dependent kinase (Seidler et al., 1997). Ca²⁺-mobilizing receptors are known to activate protein kinase C, to promote prostanoid release secondary to phospholipase A₂ activation, and to raise intracellular cyclic GMP levels via interaction with nitric-oxide synthase. Moreover, Ca²⁺-mobilizing agents promoted activation of guanylate cyclase in specialized epithelia, e.g., enterocytes (Chaudhuri et al., 1998), and a functional CFTR protein was required for mouse intestinal calcium- and cyclic GMP-dependent anion secretion (Lohmann et al., 1997). We speculate that the effect of UTP on I_sc in normal jejunum indicates that the intestinal epithelial P2Y₄ receptor couples to a CFTR-dependent ion secretory pathway via either a protein kinase C- or protein kinase G-mediated mechanism, or via prostaglandin release and activation of adenylate cyclase-coupled prostaglandin E₂ receptors.

Based on the actions of ATP and UTP on ion secretion in both human and mouse gallbladder epithelium, it was proposed that the P2Y₂ receptor is the major nucleotide receptor expressed in this tissue (Chinet et al., 1999; Clarke et al., 1999, 2000). However, the potent and efficacious action of UDP on chloride secretion in gallbladder epithelium from the P2Y₂(/) mouse (Cressman et al., 1999) indicated that a UDP-selective receptor plays an important role in the regulation of ion transport in this tissue. Our data showing that UDP is the most potent agonist on (m)P2Y₆-1321N1 cells and that mRNA for the P2Y₆ receptor is expressed in gallbladder epithelial cells place the P2Y₆ receptor as the likely candidate for mediating the UDP effects in gallbladder. The idea that the P2Y₆ receptor is important for electrolyte movement across biliary secretory epithelium has important implications. The deficient cyclic AMP-regulated fluid and ion transport across biliary epithelium in CF largely impairs hepatobiliary function. The observation that gallbladder epithelial ion secretion can be restored by activation of a Ca²⁺-dependent chloride channel by the P2Y₆ receptor in gallbladder epithelium offers therapeutic possibilities for CF biliary disease. Our results demonstrate that the robust effect of mucosal UDP on gallbladder ion secretion is preserved in the CF mouse, and suggest that the P2Y₆ receptor efficiently couples to the CFTR-independent Ca²⁺-activated Cl secretory pathway in these cells.

In summary, our data demonstrate that 1) the (m)P2Y₄ receptor is a nonselective receptor for nucleoside triphosphates and likely is the major nucleotide receptor controlling CFTR-mediated Cl transport in jejunal epithelia, and 2) the (m)P2Y₆ receptor is highly selective for UDP and likely is a major nucleotide receptor in gallbladder epithelia. The P2Y₆ receptor may be a target candidate for correcting the defective Cl transport in CF-diseased gallbladder epithelia.