Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Pancreatic polypeptide (PP) was the first peptide isolated (Kimmel et al., 1975) from a small family of structurally related regulatory peptides that includes peptide YY (PYY) and neuropeptide Y (NPY; Laburthe, 1990). As its original name implies, PP is primarily released postprandially from pancreatic endocrine type F cells (Schwartz, 1983). Although the major physiological role of PP remains to be determined, PP has been shown to exert prominent regulatory effects on gastrointestinal and accessory gland function. This includes inhibition of exocrine pancreatic secretion of enzyme, bicarbonate, and water (Schwartz, 1983); relaxation of gallbladder, thereby decreasing bile secretion (Ledeboer et al., 1997); stimulation of motility in stomach and intestine (McTigue et al., 1993); and inhibition of fluid secretion in small intestine (Holliday et al., 1997; Souli et al., 1997). PP also induces up-regulation of hepatic insulin receptors, resulting in modulation of glucose metabolism (Seymour et al., 1996). A few extradigestive effects of PP have also been described: stimulation of corticosterone secretion in rat adrenocortical cells (Andreis et al., 1993) and paracrine regulation of human and rat adrenal cortex (Nussdorfer et al., 1998).

PP receptors have been pharmacologically characterized by binding studies using labeled PP or by functional assays in intestine (Gilbert et al., 1988), liver (Nguyen et al., 1995), vas deferens (Jorgensen et al., 1990), spinal cord (Wager-Page et al., 1992), adrenal gland (Foucart et al., 1993), superior cervical ganglia (Schwartz et al., 1987), and brainstem nuclei (Whitcomb et al., 1990). A receptor displaying subnanomolar affinity for PP, the so-called Y4 or PP1 receptor, has been cloned recently in human (Bard et al., 1995; Lundell et al., 1995), rat (Lundell et al., 1996), and mouse (Gregor et al., 1996b). The human Y4 receptor cDNA encodes a predicted seven-transmembrane domain receptor of 375 amino acids. Tissue distribution studies in human and mouse suggest potential roles for Y4 receptor in the gastrointestinal tract, heart, and prostate, as well as in neural and endocrine signaling (Bard et al., 1995; Lundell et al., 1995; Gregor et al., 1996b; Yan et al., 1996). Human Y4 receptor mRNA is mainly expressed in small and large intestines and prostate, whereas various brain areas display low expression (Bard et al., 1995; Lundell et al., 1995).

The pharmacology of Y4 receptor with respect to PP, PYY, and NPY is still confusing in terms of inhibition of adenylyl cyclase activity and binding affinity inasmuch as functional characterization of recombinant receptor was unexpectedly carried out with ¹²⁵I-labeled PYY as tracer (Bard et al., 1995; Lundell et al., 1995). Moreover, few data are available regarding the molecular properties, coupling to G protein, and desensitization of human Y4 receptor. In this report, we described pharmacological and functional characterization of the recombinant human Y4 receptor stably expressed in a clonal Chinese hamster ovary (CHO) cell line. We show that the Y4 receptor is a 60-kDa monomeric glycoprotein that is negatively coupled to adenylyl cyclase via a Bordetella pertussis toxin-sensitive G_i protein. By using ¹²⁵I-hPP as a tracer in binding studies, we demonstrate that the order of binding affinity of natural peptides for the human Y4 receptor is in agreement with their relative potencies in inhibiting cAMP production: PP > PYY NPY. Finally, we present data suggesting that the human Y4 receptor is resistant to agonist-promoted desensitization.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Peptides and Reagents. PP [human and rat (hPP and rPP, respectively)], PYY [human (hPYY)], NPY [human (hNPY)], and vasoactive intestinal peptide [VIP; human (hVIP)] were purchased from Neosystem Laboratories (Strasbourg, France). ¹²⁵I-hPP (950 Ci/mmol) was prepared with ¹²⁵I-Na (IMS300; Amersham, Les Ulis, France) according to the chloramine T method and purified on a Sephadex G-50 column (Nguyen et al., 1995). ¹²⁵I-VIP (900 Ci/mmol) was prepared as previously described (Couvineau et al., 1996). The culture medium Ham's F-12 and other culture compounds were purchased from Life Technologies (Eragny, France). The eucaryotic vector pcDNA3 was purchased from InVitrogen (San Diego, CA). The CHO cell line was from European Collection of Animal Cell Cultures (cell line 85050302; Porton Down, UK). The monoclonal anti-Flag antibody was obtained from Kodak (New Haven, CT). The monoclonal anti-Tag antibody was obtained from the hybridoma MYC I-9E 10.2 (Couvineau et al., 1996), which was available through American Type Culture Collection (ATCC CRL-1729; Rockville, MD). The CHO clone 15 cell line expressing the recombinant human VPAC₁ receptor was established in our laboratory (Gaudin et al., 1996).

Transfection, Selection, and Culture of Cell Lines. The entire coding region of the human Y4 receptor from clone Hubert-pTEJ (Lundell et al., 1995) was subcloned in the pcDNA3 vector, which contains the selectable geneticin gene. In view of immunofluorescence experiments using confocal laser scanning microscopy (see later), the receptor was flagged at the C terminus with the following marker octapeptide GYKGGGGK (Flag). The Flag sequence was inserted through STOP-codon mutagenesis in EcoRI restriction site. The recombinant plasmid was transfected into the CHO cell line through calcium phosphate transfection. Briefly, 10⁶ cells were seeded onto a 100-mm Petri dish containing 10 ml of culture medium [Ham's F-12, 10% (v/v) FCS, 100 I.U./ml penicillin, 100 µg/ml streptomycin] 1 day before transfection. Then, 15 µg of plasmid DNA and 15 µg of sperm salmon DNA carrier were preincubated 30 min at room temperature in transfection buffer (125 mM CaCl₂, 140 mM NaCl, 0.75 mM Na₂HPO₄, 25 mM HEPES, pH 6.95). This mixture was added to the culture medium. Two days after transfection, cells were transferred into plastic 25-cm² culture flasks containing 10 ml of culture medium and were selected through the addition of geneticin (G418) at a final concentration of 1 mg/ml for 3 weeks. Resistant cells were cloned by limiting dilution. After cloning, clone 29 was selected on the basis of its high binding capacity for ¹²⁵I-hPP (see later) and then maintained in the above-described culture medium containing 10% FCS and 400 µg/ml G418 in a humidified atmosphere of 95% air/5% CO₂ at 37°C. Cells were passaged every 7 days in 25-cm² plastic culture flasks and used between the 4th and 12th passages. In some experiments, confluent cells grown in 12-well trays were treated overnight with pertussis toxin (0.4 µg/ml). This procedure was applied to confluent cells before the binding experiment and cellular cAMP assay (see later).

Preparation of Particulate Fraction of CHO Cells. Cells were grown to confluence for 3 to 4 days in 75-cm² plastic flasks, washed three times with 0.13 M PBS (pH 7), and then harvested with a rubber policeman and centrifuged at 2000g for 5 min at 4°C. The cell pellet was then exposed for 30 min to hypoosmotic 5 mM HEPES buffer (pH 7.4) as previously described (Voisin et al., 1996). The resulting broken cell suspension was centrifuged at 20,000g for 10 min, washed with 20 mM HEPES buffer (pH 7.4), repelleted, and stored at 80°C until used. This particulate fraction is referred to as the membrane preparation.

Binding of ¹²⁵I-hPP to Clone 29 Membrane- or Cell-Bound Receptors. Binding of ¹²⁵I-hPP to membrane preparations was conducted as previously described for PYY (Voisin et al., 1996). Briefly, membranes (100 µg protein/ml) were incubated for 45 min at 30°C in 250 µl of incubation buffer [20 mM HEPES buffer, pH 7.4, 2% (w/v) BSA, and 0.1% (w/v) bacitracin] containing 0.05 nM ¹²⁵I-hPP with or without unlabeled hPP or other competing peptides. At the end of the incubation, 150-µl aliquots of incubation medium were mixed with 150 µl of ice-cold incubation buffer. Bound and free peptides were separated through centrifugation at 20,000g for 10 min, and membrane pellets were washed twice with 10% (w/v) sucrose in 20 mM HEPES buffer, pH 7.4. The radioactivity was then counted with a gamma counter. The nonspecific binding, measured in the presence of 1 µM unlabeled hPP, represented about 2% of total binding. All binding data were analyzed using the LIGAND computer program developed by Munson and Rodbard (1980). The K_i values for the inhibition of ¹²⁵I-hPP binding by unlabeled peptides were calculated from the concentration of unlabeled peptides that induced half-maximal inhibition (IC₅₀) of the specific ¹²⁵I-hPP binding according to the following relation: K_i = IC₅₀ [K_d/(K_d + L), where K_d is the dissociation constant and L is the concentration of ¹²⁵I-hPP.

Kinetic Studies. Time course of association of ¹²⁵I-hPP was started by the addition of clone 29 cell membranes (100 µg protein/ml) as described earlier. The bound radioactivity was then measured at different times. When the binding of ¹²⁵I-hPP reached a steady state (45 min at 30°C), the dissociation of bound ¹²⁵I-hPP was followed after the addition of 1 µM unlabeled hPP by measurement of the radioactivity remaining bound at different times.

Cross-Linking of Bound ¹²⁵I-hPP to Y4 Receptors. Clone 29 cell membranes (500 µg protein/ml) were incubated for 45 min at 30°C with 0.5 nM ¹²⁵I-hPP in 20 mM HEPES buffer, pH 7.4, containing 2% (w/v) BSA and 0.1% (w/v) bacitracin. After incubation, membranes were washed with ice-cold 20 mM HEPES buffer, pH 7.4, and then incubated for 30 min at room temperature in 20 mM HEPES buffer, pH 7.4, containing 150 mM NaCl and 1 mM dithiobis(succinimidyl)propionate (DSP) as a cross-linker. The reaction was stopped by the addition of 50 µl of 1 M Tris-HCl buffer, pH 7.4. Cell membranes were then centrifuged at 20,000g for 10 min, and pellets were suspended in 60 mM Tris-HCl buffer, pH 6.8, containing 10% (v/v) glycerol, 0.001% (w/v) bromophenol blue, and 3% (w/v) SDS. The solubilization was carried out overnight at room temperature. In some experiments, cross-linked ¹²⁵I-hPP/receptor complexes were submitted to PNGase F treatment as follows: after cross-linking, membrane proteins were solubilized in 100 mM Tris-HCl buffer, pH 8.6, containing 1% (v/v) Triton X-100, 0.25% (w/v) SDS, 0.1% (v/v) -mercaptoethanol, and 2 mM phenylmethylsulfonyl fluoride. PNGase F (2 U/ml) was then added for 1 h at 37°C as previously described (Fabre et al., 1993). The medium was then centrifuged at 20,000g for 10 min, and the supernatant was applied to a 10% polyacrylamide gel with a 5% stacking gel. The gel was run and fixed as previously described (Voisin et al., 1993). It was dried and exposed for 2 days at 80°C to a Trimax type XM film (3M) with a Trimax intensifying screen.

Binding of ¹²⁵I-VIP to Recombinant VPAC₁ Receptors Expressed in CHO Clone 15 Membranes. Membranes from CHO clone 15 cells (80 µg protein/ml) were incubated for 60 min at 30°C with 0.05 nM ¹²⁵I-VIP as described in detail elsewhere (Gaudin et al., 1996), and bound and free peptides were separated through centrifugation (see earlier). The specific binding was calculated as the difference between the amount of ¹²⁵I-VIP bound in the absence (total binding) and presence (nonspecific binding) of 1 µM unlabeled hVIP. The nonspecific binding represented about 10% of total binding.

cAMP Measurement. Cellular cAMP content was assayed as previously described (Voisin et al., 1996). Clone 29 cells cultured in 12-well trays were incubated in the presence or absence of 10 mM forskolin in 1 ml of Ham's F-12 containing 2% (w/v) BSA, 0.1% (w/v) aprotinin, and 0.2 mM IBMX without or with varying concentrations of hPP for 45 min at 37°C. Clone 15 cells expressing the human VPAC₁ receptor were also cultured in 12-well trays. They were incubated in 1 ml of PBS, pH 7, containing 2% (w/v) BSA, 0.1% (w/v) aprotinin, and 0.2 mM IBMX without or with varying concentrations of hVIP for 45 min at 37°C as previously described (Gaudin et al., 1996). At the end of the incubation, the medium was rapidly removed, cells were washed in 1 ml of PBS (pH 7.4), and 1 ml of 1 M perchloric acid was added. After centrifugation for 10 min at 4000g, the cAMP present in the supernatant was succinylated, and its concentration was measured by radioimmunoassay as described (Voisin et al., 1996). Data are reported as picomoles of cAMP per 10⁶ cells.

Immunofluorescence Experiments with Confocal Microscopy. Clone 29 cells expressing the recombinant human Y4 receptor or clone 15 cells expressing the recombinant human VPAC₁ receptor (Gaudin et al., 1996) were grown on glass slides in 24-well trays as previously described (Gaudin et al., 1996). They were washed with PBS, fixed for 15 min with 2% (w/v) paraformaldehyde, and permeabilized with 0.075% (w/v) saponin in PBS (pH 7). Clone 15 cells expressing the VPAC₁ receptor tagged at the C terminus with the marker dodecapeptide (Tag) MEQKLISEEDLN (Gaudin et al., 1996) were incubated with anti-Tag antibodies (1:250^e) for 30 min. Clone 29 cells expressing the hY4 receptor flagged at the C terminus (see earlier) were incubated with anti-Flag antibodies (1:250^e) for 30 min. Clone 29 cells or clone 15 cells were then washed twice with PBS containing 0.075% (w/v) saponin and incubated for 30 min with anti-mouse IgG conjugated with fluorescein isothiocyanate. Slides were mounted in Glycergel, and selected fields were scanned with a True Confocal Scanner Leica TCS 4D consisting of a Leica Diaplan inverted microscope equipped with an argon-krypton ion laser (488 nm) with an output power of 2 to 50 mW and a VME bus MC 68020/6881 computer system coupled to an optical disk for image storage (Leica Laserchnik GmbH). The emitted light was collected through a long-pass filter on the target of the photo multiplier. Each sample was treated with a Lalman filter to increase the ratio of signal versus background. All image generating and processing operations were carried out with use of the Leica CLSM software package. Screen images were taken on Kodak TMAX with a 35-mm camera.

Desensitization Experiments. The possible down-regulation of recombinant human Y4 receptor expressed in clone 29 cells was investigated as previously described (Gaudin et al., 1996). The effect was studied of the pretreatment of cells with 10 nM hPP for 24 h on binding parameters, cAMP production, and receptor localization by immunofluorescence experiments. For binding experiments, clone 29 cells were grown to confluence for 3 to 4 days in 25-cm² plastic flasks. Cultured cell monolayers were pretreated (treated cells) or not (control cells) with 10 nM hPP in culture medium for 24 h at 37°C. Then, cultured cells were washed three times using 0.13 M PBS, pH 7, and membrane preparations from treated or control clone 29 cell were obtained as described earlier. Binding of ¹²⁵I-hPP to membranes from hPP-treated clone 29 cells or control clone 29 cells was conducted as described above. Cellular cAMP production was also assayed on hPP-treated clone 29 cells. Briefly, clone 29 cell monolayers, cultured in 12-well trays for 5 days, were pretreated (treated cells) or not (control cells) with 10 nM hPP in culture medium for 24 h at 37°C. The culture medium was then discarded, and monolayers were washed three times with 0.13 M PBS, pH 7. Treated cells or control cells were then incubated in the presence or absence of 10 µM forskolin in 1 ml of Ham's F-12 containing 2% (w/v) BSA, 0.1% (w/v) aprotinin, and 0.2 mM IBMX without or hPP (1 µM) for 45 min at 37°C. Cellular cAMP content was measured by radioimmunoassay as described earlier. In immunofluorescence experiments, clone 29 cells were grown on glass slides in 24-well trays. Cells were treated or not with 10 nM hPP for 24 h at 37°C. Cells were then washed with 0.13 M PBS and fixed for 15 min with 2% (w/v) paraformaldehyde. After permeabilization with 0.075% (w/v) saponin, treated and control cells expressing the hY4 receptor flagged at the C terminus were incubated with anti-Flag antibody, and immunofluorescence experiments using confocal microscopy were performed as described earlier.

Protein Determination. Protein concentration was measured according to the procedure of Bradford (1976) with BSA as standard.

Statistical Analysis. Results are expressed as means ± S.E. Statistical analysis of data was performed with ANOVA, followed by Dunnett's test when several groups were compared and by the Student's t test when only two groups were compared. Differences with values of P < .05 were considered statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

After selection of CHO cells transfected with the human Y4 receptor cDNA in G418 containing-culture medium, clone 29 was isolated and grown in culture medium containing 400 µg/ml G418 (see Materials and Methods). It was studied between passages 4 and 12 under conditions in which the level of recombinant human Y4 receptor appeared to be stable. Analysis (with the use of LIGAND) of the competitive inhibition of ¹²⁵I-hPP binding to membranes prepared from clone 29 cells by unlabeled hPP gave a linear Scatchard plot (Fig. 1), indicating the presence of one class of high-affinity receptors with a B_max value of 1.44 ± 0.12 pmol/mg protein and a dissociation constant of 0.26 ± 0.05 nM (five experiments). Specific binding of ¹²⁵I-hPP to clone 29 cell membranes was time-dependent and reversible. The kinetics of association of ¹²⁵I-hPP to Y4 receptors at 30°C showed that binding reached a plateau after 30 min (Fig. 2). The association rate constant k₁, calculated from the pseudo-first order rate constant k_obs, is 1.75 × 10⁸ M¹ × min¹ (Fig. 2, inset). When 1 µM unlabeled hPP was added at equilibrium, it could be observed that bound ¹²⁵I-hPP dissociated rapidly as a function of time (Fig. 2). The first order dissociation constant k₁ is 0.034 min¹. The dissociation constant calculated from these values (K_d = k₁/k₁) is 0.19 nM. The K_d values of Y4 receptors determined by Scatchard plots or by kinetic studies are therefore very similar (e.g., 0.26 versus 0.19 nM, respectively). Further experiments were carried out to determine the pharmacological profile of the recombinant hY4 receptor using several naturally occurring peptides of the PP-fold family. This receptor discriminates among hPP, rPP (78% identity between rPP and hPP), hPYY, and hNPY (Fig. 3). The K_i constants for the inhibition of ¹²⁵I-hPP binding by unlabeled peptides were (three experiments): hPP (K_i = 0.7 ± 0.1 nM) < rPP (K_i = 47 ± 10 nM) < hPYY (K_i = 94 ± 22 nM) < hNPY (K_i > 1 µM). The Y1-selective analog human [Leu³¹-Pro³⁴]NPY was slightly less potent than hPYY, with a K_i value of 124 ± 31 nM (three experiments) and substantially less potent than hPP (Fig. 3). The Y2-selective fragment porcine NPY(13-36) and the Y5-selective analog rat D-[Trp³²]NPY were also tested and did not significantly affect ¹²⁵I-hPP binding when tested up to 1 µM concentration (Fig. 3).

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Scatchard plot of hPP binding to clone 29 cells. Saturation analysis was conducted as described in Materials and Methods in the presence of a fixed concentration of 125I-hPP (0.05 nM) and varying concentrations of unlabeled hPP. Nonspecific binding was determined in the presence of 1 µM unlabeled hPP. Binding experiments were performed on membranes prepared from clone 29 cells () or clone 29 cells pretreated for 24 h with 10 nM hPP (). Data were analyzed using the LIGAND computer program (Munson and Rodbard, 1980). One representative experiment of four is shown.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Specific 125I-hPP binding to clone 29 cell membranes: Association and dissociation kinetics. The kinetics of association of 125I-hPP to clone 29 cell membranes was followed at 30°C (). Dissociation of bound 125I-hPP was measured after the addition () of 1 µM unlabeled hPP (). Insets, linearization of the association (left) or dissociation (right) data according to a pseudo-first order reaction. One representative experiment of two is shown.

View larger version (30K): [in this window] [in a new window]   Fig. 3.   Peptide specificity of Y4 receptors in clone 29 cells. Membranes were incubated with 0.05 nM 125I-hPP and varying concentrations of unlabeled hPP (), rPP (), hPYY (), hNPY (), human [Leu31-Pro34]NPY (), porcine NPY(13-36) (), and rat D-[Trp32]NPY (). Nonspecific binding was determined in the presence of 1 µM unlabeled hPP. Results are the means ± S.E. from three experiments.

In order to estimate the molecular mass of the recombinant human Y4 receptor expressed in clone 29 cells, cross-linking experiments were carried out using ¹²⁵I-hPP and DSP as a cross-linker (Fig. 4). SDS-polyacrylamide gel electrophoresis (PAGE) analysis of cross-linked material revealed a single M_r 64,000 band whose labeling was completely inhibited in the presence of 1 µM unlabeled hPP. Assuming one molecule of tracer (4000) was bound per molecule of receptor, an M_r 60,000 protein was identified as the recombinant human Y4 receptor in clone 29 cells. The calculated molecular weight corresponding to the hY4 receptor amino acid sequence deduced from the translated cloned cDNA nucleotide sequence is 42,194 (Lundell et al., 1995). Because four potential sites for N-linked glycosylation are present in the Y4 receptor protein sequence, we tried to determine the contribution of carbohydrate moiety in the apparent M_r of the recombinant receptor. For that purpose, the cross-linked ¹²⁵I-hPP-Y4 receptor complex was incubated with PNGase F. SDS-PAGE analysis of cross-linked material revealed a single M_r 47,000 band instead of the 64,000 band observed in the absence of PNGase F treatment (Fig. 4). These data indicate that the recombinant Y4 receptor expressed in CHO cells is occupied by a carbohydrate moiety, the apparent molecular weight of which can be roughly estimated at 17,000. As a control, cross-linking experiments with untransfected CHO cells gave no labeled band (Fig. 4). This supports that parent CHO cells do not express Y4 receptors, in keeping with the absence of specific ¹²⁵I-hPP binding to untransfected cells (not shown).

View larger version (34K): [in this window] [in a new window]   Fig. 4.   Covalent labeling of clone 29 cell membranes by 125I-hPP. Membranes prepared from clone 29 cells (lanes 1-3) or untransfected CHO cells (lane 4) were incubated with 125I-hPP in the absence (lanes 1, 2, and 4) or the presence (lane 3) of 1 µM unlabeled hPP. After washing, membranes were treated with 1 mM DSP. After cross-linking, membranes were treated (lane 2) or not (lanes 1, 3, and 4) with N-glycanase (PNGase F) and submitted to SDS-PAGE as described in Materials and Methods. Gels were calibrated with the following molecular weight marker proteins: -galactosidase (116,000), fructose-6-phosphate kinase (84,000), pyruvate kinase (58,000), ovalbumin (45,000), lactic dehydrogenase (36,500), and triosephosphate isomerase (26,600).

Several lines of evidence indicate that hY4 receptors in clone 29 cells are coupled to a pertussis toxin-sensitive G_i protein resulting in subsequent inhibition of production of cellular cAMP: 1) after overnight treatment of clone 29 cells to which we added pertussis toxin (0.4 µg/ml), an analysis of hPP binding to treated cells by Scatchard plots revealed a significant 2.5-fold decrease (P < .01) in receptor affinity compared with control values (K_d = 0.42 ± 0.08 nM versus 1.06 ± 0.13 nM, three experiments) with no change in receptor capacity (Table 1); 2) at maximally active dose (5 × 10⁴ M), GTP and its nonhydrolysable analog, guanosine-5'-(,-imido)triphosphate [Gpp(NH)p], inhibited the binding of ¹²⁵I-hPP to receptors in clone 29 cell membranes by 72 and 90%, respectively. Scatchard plot of hPP binding to clone 29 cell membranes, in the presence of 5 × 10⁴ M Gpp(NH)p, revealed a 4-fold decrease in receptor affinity compared with the control value with no change in receptor capacity (not shown). In contrast, ATP failed to alter hPP binding (not shown). 3) As shown in Fig. 5, hPP in the concentration range of 10¹¹ to 10⁶ M, inhibited forskolin-induced cAMP production in cultured clone 29 cells. A maximal 65% inhibition was observed at high hPP concentration (1 µM) and half-maximal inhibition (IC₅₀) was obtained for 0.56 ± 0.07 nM hPP. hPYY and hNPY were less potent than hPP, with IC₅₀ values of about 218 ± 20 nM and >1 µM, respectively. The order of potency of hPP, hPYY, and hNPY in inhibitory cAMP production is therefore very similar to the order of affinity of the peptides for the Y4 receptor, as determined in binding experiments (see Fig. 3). 4) The inhibitory effect of hPP on cAMP production was completely abolished when cultured clone 29 cells were preincubated with pertussis toxin (Table 1), indicating a pertussis toxin-sensitive G protein-mediated event.

View this table: [in this window] [in a new window]   TABLE 1 hPP binding parameters and inhibition of cAMP production by hPP in pertussis toxin-treated clone 29 cells Clone 29 cells were incubated overnight without () or with (+) 0.4 µg/ml Bordetella pertussis toxin (PTX). Y4 receptor binding parameters were determined by Scatchard analysis, and intracellular cAMP levels in cells were measured by radioimmunoassay. Results are means ± S.E. of three experiments.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Dose-response of hPP, hPYY, and hNPY in inhibiting forskolin-stimulated cAMP production in clone 29 cells. Clone 29 cells were incubated with 10 µM forskolin and varying concentrations of hPP (), hPYY (), and hNPY () for 45 min at 37°C. Compounds were added together with forskolin at time 0. The cellular cAMP content was then determined as described in Materials and Methods. Results are expressed as a percentage of forskolin-induced cAMP levels in absence of hPP, hPYY, or hNPY. The basal and forskolin-stimulated cAMP levels were 49 ± 10 and 325 ± 36 pmol/106 cells, respectively. Each value is the mean ± S.E. of three experiments.

Finally, we investigated the possible down-regulation of recombinant human Y4 receptors expressed in clone 29 cells on exposure to its natural agonist, hPP. It appeared that exposure of clone 29 cells to 10 nM hPP for 24 h did not result in any modification of the binding capacity (1.38 ± 0.08 and 1.44 ± 0.12 pmol/mg protein in treated cells and control cells, respectively) or binding affinity (K_d = 0.31 ± 0.09 and 0.26 ± 0.05 nM in treated cells and control cells, respectively; Fig. 1). We verified that preexposure of clone 29 cells to 10 nM hPP for shorter periods of time (1, 2, 4, and 8 h) did not result in any change in binding properties (not shown), nor was there any effect of treatment of clone 29 cells expressing human Y4 receptor with hPP on maximal agonist-induced inhibition of cAMP accumulation in clone 29 cells (Fig. 6A). A maximal 65% inhibition was observed with 1 µM hPP in control clone 29 cells and treated clone 29 cells (Fig. 6A). We verified that the absence of receptor desensitization was not associated with receptor density in clone 29 cells by investigating clone 21 cells, which have a 5-fold lower binding capacity than clone 29 cells. Y4 receptors in clone 21 cells also did not desensitize. Due to the presence of a Flag at the C terminus of the recombinant human Y4 receptor, it was possible to localize the receptor in clone 29 cells with confocal laser scanning microscopy using anti-Flag antibodies. As shown in Fig. 7, Y4 receptors were diffusely distributed over the plasma membrane of clone 29 cells. On the exposure of cells to 10 nM hPP for 24 h, the agonist treatment did not influence the subcellular distribution of Y4 receptors expressed in clone 29 cells, and no agonist-induced translocation of the Y4 receptor from plasma membrane to sequestered vesicles could be observed. It was verified that the presence of a Flag epitope at the C terminus of the recombinant human Y4 receptor has no effect on the binding affinity, pharmacological profile, and desensitization of the receptor compared with the wild-type Y4 receptor transfected in CHO cells (not shown). In this context, it was crucial to demonstrate that the absence of hPP-induced desensitization and sequestration of Y4 receptors was not due to intrinsic properties of CHO cells. Therefore, we also investigated CHO cells stably transfected with the recombinant hVIP (VPAC₁) receptor (Gaudin et al., 1996). Under conditions identical to those used for the investigation of Y4 receptors (i.e., after preincubation of CHO cells with 10 nM VIP for 24 h), we observed: 1) a sharp decrease in the binding capacity of CHO cells for VIP (0.43 ± 0.03 versus 1.62 ± 0.15 pmol/mg protein) with no change in the K_d value for VIP (i.e., 0.3 nM), 2) an important decrease in the efficacy of VIP in stimulating cAMP production (Fig. 6B), and 3) an internalization and a sequestration of VPAC₁ receptors in vesicular structures (Fig. 7). Because down-regulation experiments were carried out over a 24-h exposure period, the degree of peptide metabolism by CHO cells was investigated. Clone 29 cells and VPAC₁-expressing CHO cells were therefore incubated for 24 h at 37°C with hPP (10 nM) and VIP (10 nM), respectively. The amount of active peptide was thereafter measured by radioreceptorassay, showing that only 16 ± 6 and 18 ± 5% of hPP and VIP were inactivated, respectively (eight determinations). Therefore, it could be suggested that the absence of desensitization and internalization of Y4 receptors is an intrinsic property of this receptor, at least when transfected in CHO cells.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Desensitization of clone 29 cells or clone 15 cells: Effect on peptide-regulated cAMP level. A, clone 29 cells were preincubated in culture medium for 24 h at 37°C without (left) or with (right) 10 nM hPP. After washing with PBS, clone 29 cells were incubated for 45 min at 37°C with buffer alone (basal), 10 µM forskolin (Fk), or 10 µM forskolin and 1 µM hPP (Fk + hPP). Values are means ± S.E. from three experiments. ***P < .001 versus Fk. B, clone 15 cells were preincubated in culture medium for 24 h at 37°C without (left) or with (right) 10 nM hVIP. After washing with PBS, clone 15 cells were incubated for 45 min at 37°C with buffer alone (basal) or 1 µM hVIP (VIP). Values are means ± S.E. from three experiments. **P < .01, ***P < .001 versus basal. The cellular cAMP contents were then measured as described in Materials and Methods.

View larger version (61K): [in this window] [in a new window]   Fig. 7.   Desensitization of clone 29 cells or clone 15 cells: Confocal laser scanning detection of receptors. Clone 29 cells and clone 15 cells were preincubated with hPP and hVIP, respectively, as described in the legend to Fig. 5. The culture medium was then discarded, and cells were permeabilized with saponin. Clone 29 cells were incubated with a mouse monoclonal anti-Flag antibody, and clone 15 cells were incubated with a mouse monoclonal anti-Tag antibody. After washing, cells were incubated with anti-mouse IgG conjugated to fluorescein isothiocyanate and analyzed by confocal microscopy as described in Materials and Methods. A, untransfected CHO cells incubated with the anti-Flag antibody gave no staining. B, clone 29 cells preincubated in the absence of hPP. C, clone 29 cells preincubated in the presence of 10 nM hPP. D, untransfected CHO cells incubated with the anti-Tag antibody gave no staining. E, clone 15 cells preincubated in the absence of hVIP. F, clone 15 cells preincubated in the presence of 10 nM hVIP. VPAC1 receptors were diffusely distributed over the plasma membrane of clone 15 cells. On exposure of cells to 10 nM VIP for 24 h, receptors were internalized and sequestered in vesicular structures.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The present study represents the first extensive characterization of a stably transfected human Y4 receptor. It provides new insights regarding several aspects of the Y4 receptor, including molecular properties, peptide specificity, and negative coupling to adenylyl cyclase via a pertussis toxin-sensitive G protein. It also shows that the recombinant human Y4 receptor expressed in CHO cells is resistant to agonist-mediated desensitization and internalization.

The predicted molecular weight of the human Y4 receptor calculated from its amino acid sequence is 42,194 (Lundell et al., 1995). This receptor, as well as other Y receptors, is not endowed with signal peptide, as supported by sequence analysis of the N-terminal domain of receptor proteins (Bard et al., 1995; Lundell et al., 1995). Therefore, it could be anticipated that the molecular weight of the recombinant Y4 receptor expressed in CHO cells should be close to 42,000 in the absence of post-translational alteration. Cross-linking experiments using ¹²⁵I-hPP as tracer and DSP as a cross-linker revealed a single molecular receptor binding protein in transfected cells, the molecular weight of which is 60,000 (see Fig. 4). These data suggested that recombinant Y4 receptor undergoes N-glycosylation in CHO cells. This was demonstrated by treatment of the cross-linked binding protein with PNGase F, which resulted in an important decrease in its molecular weight, supporting that N-linked carbohydrates represent 17 kDa. This is consistent with the presence of three potential sites for N-linked glycosylation in the N terminus on asparagine residues 2, 19, and 29 (Bard et al., 1995). Because it was previously shown that all potential N-glycosylation sites in G protein-coupled receptors are not necessarily occupied by N-linked carbohydrates (Couvineau et al., 1996), further studies are certainly necessary to determine which sites are actually occupied in the human Y4 receptor. In a previous study of native PP receptors in rat liver membranes, the apparent molecular weight of the receptor protein was estimated at 42,000 (Nguyen et al., 1995), which is close to the molecular weight calculated from the rat Y4 receptor amino acid sequence (Lundell et al., 1996), suggesting a very low level of glycosylation of PP receptors in rat liver. Important differences in the N-glycosylation of G protein-coupled receptors according to tissues and/or species have been previously described (Fabre et al., 1993).

After the cloning of human Y4 receptor, its peptide specificity with respect to natural peptides of the PP-fold family was determined by binding studies that unexpectedly made use of ¹²⁵I-pPYY as a tracer instead of ¹²⁵I-hPP (Lundell et al., 1995). Lundell et al. (1995) reported that the K_i values of hPP, hPYY, and hNPY for inhibiting tracer binding to membranes from transiently transfected Cos1 cells were 0.014, 1.4, and 9.9 nM, respectively, supporting the idea that both hPP and hPYY are potent inhibitors of ¹²⁵I-pPYY binding to Cos1 cell membranes expressing human Y4 receptor. In sharp contrast, the same authors reported that half-maximal inhibition of cAMP production in transfected CHO cells was obtained with 7, 95, and >10,000 nM hPP, hPYY, and hNPY, respectively (Lundell et al., 1995). On the other hand, Bard et al. (1995) determined that the K_i values of hPP, hPYY, and hNPY for inhibiting ¹²⁵I-pPYY binding to membranes from transiently transfected Cos7 cells were 0.05, 0.87, and 2 nM, respectively. These data also suggested that the human Y4 receptor poorly discriminates between PP-fold family peptides. In the present study, we used ¹²⁵I-hPP as a tracer and found that the K_i values of hPP, hPYY, and hNPY for inhibiting tracer binding to membranes from stably transfected CHO cells were 0.7, 94, and >1000 nM, respectively. These data support the idea that the human Y4 receptor is highly specific for hPP. This is in close agreement with the inhibition of forskolin-stimulated cAMP production by hPP, hPYY, and hNPY in the same cells, with half-maximal inhibition obtained for 0.56, 218, and >1000 nM, respectively. The reason for the discrepancies between our study and those of Lundell et al. (1995) and Bard et al. (1995) may be tentatively ascribed to several factors, including transfection of different cell lines (CHO versus COS cells), use of different tracers (¹²⁵I-hPP versus ¹²⁵I-pPYY), and assay conditions. Previous studies of mouse Y6 receptors, also referred to as PP2 receptors, also pointed out sharp discrepancies regarding the pharmacological profile of this receptor transfected in CHO cells according to the tracer used [i.e., ¹²⁵I-PP (Gregor et al., 1996a) versus ¹²⁵I-PYY (Weinberg et al., 1996)]. Finally, the recent discovery of receptor activity-modifying proteins (RAMPs), the expression of which can regulate the ligand specificity of a G protein-coupled receptor (McLatchie et al., 1998), also supports the idea that the pharmacological profile of a given receptor can be largely modified by its molecular environment. Whether RAMPs or RAMP-like proteins do interact with Y4 receptors and, more widely, with Y receptors remains to be determined. Regardless of these possibilities, the present study clearly demonstrates by both binding assay and biological assay (inhibition of cAMP production) that the human Y4 receptor is actually a PP receptor with very low affinity for hPYY or hNPY. The use of selective ligands such as human [Leu³¹-Pro³⁴]NPY, porcine NPY(13-36), and rat D-[Trp³²]NPY (see Fig. 3) also clearly discriminated the human Y4 receptor against the human Y1, Y2, and Y5 receptors, respectively.

Previous studies have shown that PP mediates inhibition of cAMP production in Y4 receptor-transfected cells (Bard et al., 1995; Lundell et al., 1995, 1996). This was confirmed in the present work because PP at maximally active concentration inhibited forskolin-stimulated cAMP production in clone 29 cells by 65% (see Figs. 5 and 6A). This work further demonstrates that PP inhibits cAMP production by promoting the coupling of Y4 receptor to G_i protein because: 1) GTP or Gpp(NH)p inhibited the binding of ¹²⁵I-PP to Y4 receptors and 2) pretreatment of clone 29 cells with pertussis toxin completely abolished the inhibitory effect of PP on cAMP production in clone 29 cells and increased the K_d value of PP for binding to Y4 receptors. Although these data support the coupling of Y4 receptors with a pertussis toxin-sensitive G protein, it remains unclear which subtype of G protein is actually involved inasmuch as all G_i proteins, including G_i1, G_i2, and G_i3, and G_o protein are substrates for pertussis toxin and have been involved in inhibition of adenylyl cyclase (Taussig and Gilman, 1995). A role for G_o protein is unlikely because it is not involved in adenyl cyclase inhibition (Taussig and Gilman, 1995). A few previous studies of Y receptor coupling to G protein using G_i RNA antisense technology (Voisin et al., 1996) or antibodies to G_i (Freitag et al., 1995) have shown exclusive coupling to G_i2 in kidney epithelial cells (Voisin et al., 1996) or the involvement of both G_i2 and G_i3 in neuronal cells (Freitag et al., 1995). Finally, Bard et al. (1995) reported that human Y4 receptors expressed in LMTK cells (mouse fibroblast) mediated PP-induced increase of intracellular free Ca²⁺ concentration and PP-induced inhibition of cAMP production. This human Y4 receptor-dependent Ca²⁺ regulation was not evidenced in clone 29 cells stably transfected with human Y4 receptor because PP failed to increase intracellular Ca²⁺ as measured in Fura-2-loaded cells (T.V., M.G., A.-M.L., J.-J.M., and M.L., unpublished observations).

A surprising observation in the present work is that the Y4 receptor transfected in CHO cells is not prone to desensitization and internalization. Indeed, when clone 29 cells were preincubated with 10 nM hPP (i.e., a concentration about 10 times higher than the K_d value of the receptor or the EC₅₀ value of PP in inhibiting cAMP production) for 24 h, they were still fully responsive to a further challenge to PP with respect to inhibition of cAMP production (see Fig. 6A). Further studies also gave evidence of the absence of Y4 receptor internalization on challenge with PP. Indeed, preincubation of clone 29 cells with 10 nM PP did not result in any change in the binding capacity or K_d value of Y4 receptor at the cell surface. This was in good agreement with the fact that no internalization of flagged Y4 receptor could be evidenced by confocal microscopy of clone 29 cells on challenge with PP (see Fig. 7). None of these observations are related to the fact that CHO cells are not equipped with the molecular machinery necessary to desensitize and internalize G protein-coupled receptors. Indeed, desensitization and internalization of G protein-coupled receptors transfected in CHO cells were previously reported for various receptors coupled to adenylyl cyclase (Vilardaga et al., 1994; Gaudin et al., 1996; Fehmann et al., 1998), including G_i protein-coupled receptors mediating inhibition of adenylyl cyclase (Zhao et al., 1998). Moreover, it was verified in this study that VPAC₁ receptors stably transfected in CHO cells did desensitize and internalize under experimental conditions in which Y4 receptors did not (see Figs. 6B and 7). Although the possibility that CHO cells may lack some of the machinery involved in Y4 receptor regulation while such machinery is present for VPAC₁ receptor cannot be strictly ruled out, the present data suggest that the resistance to agonist-promoted desensitization and internalization is an intrinsic property of the human Y4 receptor. No data are available regarding the desensitization and sequestration of endogenously expressed Y4 receptors in cell lines or tissues. Such resistance has been previously observed for other G protein-coupled receptors, such as the human ₃-adrenergic receptor (Nantel et al., 1993). One possible hypothesis proposed to explain the resistance of the ₃-adrenergic receptor to agonist-promoted desensitization was the absence, in its primary structure, of most of the phosphorylation sites found in the ₂-adrenergic receptor that undergoes desensitization (Nantel et al., 1993). In particular, the absence of a consensus phosphorylation site for protein kinase A was emphasized (Nantel et al., 1993). Although nothing is known regarding the phosphorylation of Y4 receptor, it is interesting to note that the Y4 receptor, like the ₃-adrenergic receptor, lacks a phosphorylation site for protein kinase A in its primary sequence (Lundell et al., 1995). Regardless of the molecular basis for resistance to hPP-promoted desensitization of human Y4 receptors, this resistance may have important physiological consequences. In this context, it is interesting to note that contrary to the transient postprandial secretion of most digestive hormones (Brand and Schmidt, 1995), the secretion of PP into the blood may continue for hours after the ingestion of a large meal (Taylor, 1989). Whether long-term stimulation of Y4 receptors by a high and sustained postprandial plasma level of PP has a physiological meaning remains to be studied.

In conclusion, clone 29 CHO cells, which have been engineered to stably expressed the human Y4 receptor, provide clear evidence that the Y4 receptor is actually a PP-selective receptor. They also provide a powerful tool to study the pharmacological and molecular properties of this PP receptor and should be instrumental in developing new Y4 receptor agonists and antagonists.