Abstract
The antisecretory effects of several Y agonists, including pancreatic polypeptide (PP), indicate the presence of Y1, Y2, and Y4 receptors in mouse and human (h) colon mucosae. Here, we used preparations from human and from wild-type (WT), Y4, and Y1 receptor knockout (-/-) mice, alongside Y4 receptor-transfected cells to define the relative functional contribution of the Y4 receptor. First, rat (r) PP antisecretory responses were lost in murine Y4-/- preparations, but hPP and Pro34 peptide YY (PYY) costimulated Y4 and Y1 receptors in WT mucosa. The Y1 antagonist/Y4 agonist GR231118 [(Ile,Glu,Pro,Dpr,Tyr,Arg,Leu,Arg,Try-NH2)-2-cyclic(2,4′),(2′,4)-diamide] elicited small Y4-mediated antisecretory responses in human tissues pretreated with the Y1 antagonist, BIBO3304 [(R)-N-[[4-(aminocarbonylaminomethyl)-phenyl]methyl]-N2-(diphenylacetyl)-argininamide trifluoroacetate)], and attenuated Y4-mediated hPP responses in mouse and human mucosa. GR231118 and rPP were also antisecretory in hY4-transfected epithelial monolayers but were partial agonists compared with hPP at this receptor. In Y4-transfected human embryonic kidney (HEK) 293 cells, Y4 ligands displaced [125I]hPP binding with orders of affinity (pKi) at human (hPP = rPP > GR231118 > Pro34PYY = PYY) and mouse (rPP = hPP > GR231118 > Pro34PYY > PYY) Y4 receptors. GR231118- and rPP-stimulated guanosine 5′-3-O-(thio)triphosphate binding through hY4 receptors with significantly lower efficacy than hPP. GR231118 marginally increased basal but abolished further PP-induced hY4 internalization to recycling (transferrin-labeled) pathways in HEK293 cells. Taken together, these findings show that Y4 receptors play a definitive role in attenuating colonic anion transport and may be useful targets for novel antidiarrheal agents due to their limited peripheral expression.
Pancreatic polypeptide (PP) was the first reported member of a structurally related peptide family that comprises the neurotransmitter, neuropeptide Y (NPY), and hormones, peptide YY (PYY), and its product, PYY(3-36). The predominant source of PP is the pancreatic F-cell (Fiocca et al., 1983), although scattered PP-containing enteroendocrine cells are present in the human colon (El-Salhy et al., 1983). In contrast, PYY is abundant in mucosal L-type endocrine cells concentrated in the distal regions of murine (Arantes and Nogueira, 1997) and human (El-Salhy et al., 1983) intestine. NPY is expressed in enteric neurons of most mammalian species studied to date, including submucous plexus neurons that innervate the mucosa of human colon (Peaire et al., 1997).
PP and PYY are released postprandially and reach maximal plasma concentrations within 10 to 25 and 60 min, respectively. Their subsequent physiological actions include central appetite suppression (Murphy and Bloom, 2004) and peripheral effects on ion transport (Cox and Tough, 2002) and gastrointestinal motility. In ileostomy patients, PYY inhibits vasoactive intestinal polypeptide (VIP)-mediated fluid secretion (Playford et al., 1990), and in vitro NPY, PYY, PYY(3-36), and PP inhibit anion secretion in mouse and human colonic mucosa (Cox et al., 2001a; Cox and Tough, 2002). In addition, PYY is responsible for the “ileal brake” that slows gastrointestinal transit to facilitate digestion, thus highlighting the potential for Y-based agonists as natural antidiarrheal agents.
PP, PYY, and NPY act through four Gi-coupled receptors (Y1, Y2, Y4, or Y5), with a fifth functional type (Y6) present only in mice and rabbits (for review, see Michel et al., 1998). PP is the major endogenous agonist for the Y4 receptor, but it also binds and activates the Y5 receptor (Gerald et al., 1996). Y1, Y2, and Y5 receptors are stimulated by PYY or NPY, whereas C-terminal fragments, such as PYY(3-36), are Y2 and Y5 receptor-preferred agonists. Antisecretory responses to PP, PYY, and NPY in human and mouse colonic mucosa are mediated by a combination of Y1, Y2, or Y4 receptors, whereas Y5 agonists are ineffective (Cox et al., 2001a; Cox and Tough, 2002). Y1 and Y4 receptors are expressed by epithelia, whereas Y2 responses alone are sensitive to the neurotoxin tetrodotoxin, indicating that this receptor is expressed on submucous enteric neurons (Cox and Tough, 2002; Hyland et al., 2003). In addition, endogenous stimulation of epithelial Y1 receptors by PYY or neuronal Y2 receptors by NPY provides antisecretory tone in these two tissues, as revealed by competitive antagonists [i.e., BIBO3304 (Y1) or BIIE0246 (Y2)] (Cox and Tough, 2002; Hyland and Cox, 2005). Notably, human colon adenocarcinoma cell lines express Y4 receptors either alone (Cox et al., 2001b) or alongside Y1 receptors (Tough and Cox, 1996) but not Y2 receptors.
The restricted peripheral localization of the Y4 receptor (compared with widespread Y1 and Y2 distributions) provides an opportunity to develop selective antidiarrheal agonists to target the Y4 receptor, potentially with fewer side effects. To date, however, detailed characterization of Y4-mediated responses has been hampered by the lack of selective ligands. The Y1-preferring agonist Pro34PYY has efficacy at the Y4 receptor but only at high concentrations (Cox et al., 2001b). In addition GR231118, a homodimeric peptide based on the C-terminal sequence of NPY [(Ile, Glu,Pro,Dpr,Tyr,Arg,Leu,Arg,Try-NH2)-2-cyclic(2,4′),(2′,4)-diamide, also known as 1229U91] was originally identified as a competitive Y1 receptor antagonist with low Y2 affinity (Daniels et al., 1995; Hegde et al., 1995), but this dimeric nonapeptide also has Y4 affinity and efficacy (Matthews et al., 1997; Parker et al., 1998; Schober et al., 1998, 2000). The recent generation of Y4 knockout (-/-) mice provides the only current alternative to explore the loss of specific function associated with obesity (Sainsbury et al., 2002), cardiac function (Smith-White et al., 2002), and water intake (Wultsch et al., 2006). These studies found that Y4-/- mice are leaner, have decreased resting heart rate with lower arterial blood pressure, and increased dark phase water intake compared with wild-type (WT) mice. To define the role of the Y4 receptor in intestinal epithelia, we have used different Y4 ligands, including rat (r) PP, human (h) PP, Pro34PYY, and GR231118, to inhibit ion secretion across colonic mucosae from WT, Y4-/-, and Y1-/- mice, and supported these studies with investigations using human colon mucosa and cells expressing recombinant human or mouse (m) Y4 receptors. We demonstrate that human and mouse Y4 receptors are activated differentially by rPP and hPP and that at both orthologs, GR231118 acts as a partial Y4 agonist.
Materials and Methods
Materials. VIP, somatrophin release inhibitory factor, somatostatin-14 (SRIF-14), human (h) PP, hPYY(3-36), hPro34PYY, rat (r)PYY, and rPP (Bachem Laboratories Inc., St. Helens, UK) were dissolved in water and stored at -20°C, undergoing a single freezethaw cycle. [125I]hPP, [125I]PYY, and GTPγ[35S] were purchased from PerkinElmer Life and Analytical Sciences (Boston, MA). GR231118 was a gift from Dr. A. Daniels (GSK, Durham, NC), and the nonpeptide antagonists at the Y1 receptor (BIB03304) and the Y2 receptor (BIIE0246) were gifts from Boehringer Ingelheim Pharma KG (Biberach an der Riss, Germany). BIBO3304 and BIIE0246 solutions (1 mM stock concentration dissolved in 10% dimethyl sulfoxide) were stored at -20°C until required. Cell culture consumables were from the following suppliers: Dulbecco's modified Eagle's medium (DMEM) and fetal bovine serum (Invitrogen, Paisley, UK); trypsin (Worthington Biochemicals, Freehold, NJ); and kanamycin and amphotericin B (ICN Biomedicals, Oxford, UK). Y4 receptor cDNAs in pcDNA3.1 were kindly provided by Novo Nordisk (Copenhagen, Denmark; human sequence) and Dr. H. Herzog (Garvan Institute of Medical Research, Sydney, Australia; mouse sequence). Restriction enzymes and Pwo polymerase were purchased from New England Biolabs (Hitchin, UK). The rat monoclonal antibody 3f10 raised against the hemagglutinin (HA) epitope was supplied by Roche Molecular Biochemicals (Lewes, UK), and transferrin-Texas Red and goat anti-rat Alexa Fluor 488 were from Invitrogen. All other reagents were purchased from Sigma Chemical (Poole, Dorset, UK).
Animals. WT, Y4-/-, and Y1-/- mice, all maintained on a C57BL/6-129SvJ background, were provided by Dr. H. Herzog (Garvan Institute of Medical Research) and bred in-house. All mice were maintained in a 12-h light/dark cycle, with access to standard chow and water ad libitum. Mice were killed by 100% CO2 asphyxiation, after which the descending colon was removed and placed in oxygenated Krebs-Henseleit (KH) buffer containing 118 mM NaCl, 4.7 mM KCl, 25 mM NaHCO3, 1.2 mM KH2PO4, 1.2 mM MgSO4, 2.5 mM CaCl2, and 11.1 mM glucose, pH 7.4. Comparisons of mucosal responses from WT and Y4-/- or Y1-/- mice used tissues from age-matched males or females (≥10 weeks old).
Human Colon Preparations. Human distal colon specimens were obtained from consenting patients undergoing bowel resection surgery (five anterior resections, two sigmoid colectomies, one total colectomy, one right hemicolectomy, and one left hemicolectomy) for primary carcinoma, with the approval of the Guy's and St. Thomas' Hospitals Research Ethics Committee. Specimens were obtained from five males and five females (mean age, 64.6 ± 4.6 years). Colonic segments were taken no less than 15 cm from the tumor within 30 min of excision and were kept in KH until dissection.
Construction of HA-Tagged Y4 Receptor cDNA. The hY4 receptor coding sequence, lacking the initial Met codon, was amplified by polymerase chain reaction with a proofreading polymerase (Pwo). The following primers introduced flanking EcoRI/XbaI sites (underlined): forward, 5′-GAAGAATTCAACACCTCTCACCTCCTGGCCTTG-3′ and reverse, 5′-CCCTCTAGATTAAATGGGATTGGAC-3′. The purified product was digested with EcoRI and XbaI and ligated in-frame into the vector pCruzHA (Santa Cruz Biotechnology Inc., Santa Cruz, CA), followed by subcloning of the open reading frame (HindIII/XbaI) into pcDNA3.1 (Invitrogen). The HA-hY4 cDNA comprised the N-terminal HA epitope (MGSYPYDVPDYASLEF) followed by amino acids 2 to 375 of the Y4 receptor and was verified by full double-stranded sequencing.
Transfection Procedure and Cell Culture. HEK293T (from Prof. S. J. Hill, University of Nottingham, Nottingham, UK), and HT-29 human adenocarcinoma cells were grown to confluence in DMEM supplemented with 25 mM glucose, 10% fetal bovine serum, 100 μg/ml kanamycin, and 1.2 μg/ml amphotericin B at 37°C, as described by Holliday et al. (2005). They were passaged by trypsinization [0.25% (HEK293) or 0.5% (HT-29) (w/v) in versene]. Stable transfection with appropriate Y receptor cDNAs was performed by calcium phosphate coimmunoprecipitation (Holliday et al., 2005), selecting clones by G418 resistance (0.8 mg/ml) and screening for Y receptor expression. HEK293 clones expressing the native human or mouse Y4 receptors (hY4-13, abbreviated to hY4; and mY4-2, abbreviated to mY4) exhibited comparable levels of [125I]hPP-specific binding and were selected for displacement studies. GTPγ[35S] binding studies were undertaken using HEK293 cells transfected with either the HA-tagged human Y4 receptor (HA-hY4 [125I]hPP saturation, pKd, 10.05 ± 0.06; Bmax, 1.7 ± 0.2 pmol/mg; n = 3) or the murine Y4 receptor (mY4-3; also denoted below as mY4). Both these cell lines exhibited [125I]hPP (10 pM)-specific binding levels three times greater than the clones used in displacement studies. The hY4-transfected HT-29 clone was identified by the presence of functional PP responses.
Ussing Chamber Studies. Mucosal sheets from mouse and human descending colon were prepared as described previously (Cox et al., 2001a; Cox and Tough, 2002). Preparations with exposed areas of 0.6 (human) or 0.14 (mouse) cm2 were bathed in oxygenated KH solution at 37°C, and voltage-clamped at 0 mV in an Ussing chamber (DVC1000; WPI, Sarasota, FL). The resulting short-circuit current (Isc) was recorded continuously. A stable basal Isc was reached within 20 to 30 min, after which peptide additions were made to the basolateral reservoir. Mouse but not human tissues were pretreated with 30 nM VIP for 15 to 20 min to raise cAMP and consequently Isc, thus maximizing subsequent Y receptor-mediated antisecretory responses. Single additions of hPP, rPP, PYY, or GR231118 analogs were made at the indicated concentrations before the addition of the α2 adrenoceptor agonist UK14,304 (1 μM), which was used as an internal inhibitory control. The antagonists BIBO3304 (300 nM) or BIIE0246 (1 μM) were used to block Y1 or Y2 receptors, respectively, and tissues were pretreated with either vehicle (0.01% dimethyl sulfoxide) or the antagonists for 20 min. The effects of the antagonists were investigated on hPP and Pro34PYY (both 100 nM) responses in WT or Y4-/- mucosae and compared with GR231118 on rPP (30 nM), Pro34PYY (10 nM), and PYY(3-36) (30 nM) responses in WT or Y1-/- mucosae. In human mucosa, the effect of BIBO3304 was compared with 1 μM GR231118 on subsequent hPP or Pro34PYY (both 10 nM) responses, and the effects of the dipeptide were investigated on basal Isc in the presence and absence of BIBO3304.
HT-29 epithelial monolayers were grown on collagen-coated Millipore filters (area, 0.2 cm2) (Holliday et al., 2005) and were voltage-clamped at 0 mV. All monolayers were pretreated with 30 nM VIP before the addition of hPP, rPP, or GR231118 at the indicated concentration and a subsequent addition of 100 nM hPP 15 min later and SRIF-14 (100 nM) as an internal control.
Receptor Binding Assays. Fresh membranes (20 μl, 0.9 ± 0.1 μg/μl) from HEK293 clones, prepared as outlined by Holliday and Cox (1996), were incubated for 2 h at 22°C in binding buffer [10 mM HEPES, 5 mM KCl, 2.5 mM CaCl2, 1.2 mM K3PO4, 1.2 mM MgSO4, 25 mM NaHCO3, 0.1 mg/ml bacitracin, and 0.1% bovine serum albumin (BSA), pH 7.4] and 10 pM [125I]hPP, with or without increasing concentrations of unlabeled ligand (0.1 pM-1 μM). Membrane-bound radioactivity was separated by rapid filtration through Whatman GF/B filters (presoaked in 0.3% polyethylenimine), washed with ice-cold binding buffer, pH 7.4 at 4°C, and quantified in a gamma counter.
GTPγ[35S] Binding Assays. The GTPγ[35S] binding protocol has been described in detail previously (Holliday et al., 2005). In brief, fresh membranes prepared from HEK HA-hY4 or mY4 clones (30 μl) were pre-equilibrated for 90 min at 21°C in an optimized binding buffer, pH 7.4, containing 10 mM HEPES, 100 mM NaCl, 10 mM MgCl, 1 mM EDTA with 0.2% BSA, 0.1 mg/ml bacitracin, and 3 μM GDP, in the absence (basal binding) or presence of agonists hPP, rPP, PYY, and GR231118 (10 pM-1 μM). GTPγ[35S] (200 pM) was then added, and incubations were continued for an additional 20 min, over which the rates of basal and agonist-stimulated GTPγ[35S] binding were both linear. Reactions were terminated by rapid filtration over GF/B filters, and the bound membranes were washed with 10 ml of ice-cold incubation buffer without GDP. Filters were dried overnight before addition of scintillant (Ultima Gold MV; Packard Instruments, Berkshire, UK) and counting of β emissions.
Immunofluorescence Microscopy. HEK293 clones expressing HA-hY4 receptors were grown to 50 to 70% confluence on poly-l-lysine coverslips and preincubated for 1 h in serum-free DMEM including 1% BSA. Cells were labeled in the same medium with rat monoclonal anti HA antibody (8 μg/ml, 30 min at 37°C), before ligand addition, including transferrin-Texas Red (10 μg/ml) for the times indicated in the text. To terminate receptor trafficking, cells were washed twice (in PBS), fixed in 3% paraformaldehyde in PBS (15 min at 21°C), and subsequently permeabilized (0.075% Triton X-100 in PBS, 5 min at 21°C). Secondary detection was by goat anti-rat Alexa Fluor 488 conjugate (1:1000), and nuclei were visualized with 4′,6-diamidino-2-phenylindole, before postfixing and mounting in Mowiol 40-88 (Calbiochem, Nottingham, UK), as described before (Holliday et al., 2005).
Data Analysis. Pooled data from Ussing chamber studies are quoted as microamperes per centimeter squared, mean ± 1 S.E.M. GR231118 and subsequent hPP responses in HT-29 hY4 cells were also expressed as percentage reductions of VIP-prestimulated Isc levels. Y agonist concentration-response curves were constructed from single peptide additions to mucosa or epithelial monolayers, and pEC50 values ± 1 S.E.M. were calculated from sigmoidal curve fits to the combined data using GraphPad Prism 3.03 (GraphPad Software, San Diego, CA). Statistical analysis of two data sets was undertaken using Student's t test and multiple comparisons using one-way analysis of variance (ANOVA), with Dunnett's post-test where appropriate.
Radioligand binding studies were performed in duplicate using 100 nM hPP to determine nonspecific binding. Specific binding was ≥95% of total binding in HEK293 clone membranes. Displacement curves were fitted to pooled data groups from at least three experiments using GraphPad Prism. Conversion of calculated IC50 to pKi values was by the Cheng-Prusoff equation (Cheng and Prusoff, 1973). Estimates of [125I]hPP pKd and Bmax values were obtained from the direct fit of a one-site hyperbola using GraphPad Prism to individual saturation experiments.
In GTPγ[35S] binding assays, nonspecific binding, which was <5% total counts, was assessed in the presence of 10 μM unlabeled GTPγS. Each triplicate experiment included 1 μM hPP as an indicator of maximal Y4 receptor activation. Agonist-induced responses were expressed as a percentage over basal binding, which was 62.7 ± 7.8 fmol/mg (n = 15) for cells expressing HA-hY4 receptors and 115.6 ± 5.2 fmol/mg (n = 10) for mY4-transfected cells. Concentration-response curves were fitted to the combined data using Graph-Pad Prism, yielding the quoted pEC50 values.
For analysis and quantitation of immunofluorescence experiments (Sunyach et al., 2003), a vertical stack of images separated by 0.2-μm z-steps was acquired on a Zeiss Axiovert 100 microscope (63× oil objective; Omega optical excitation and emission filter sets; Omega Optical, Brattleboro, VT) using Openlab 4.03 (Improvision, Warwickshire, UK), and out of focus light was removed by deconvolution in Volocity 3.1 (automatic fast restoration; Improvision). Representative example images present the central 15 z-sections (3.0 μm, viewed from above), and this volume was used to measure the extent of receptor internalization and colocalization with transferrin (Volocity Measurements module). Data groups show the mean ± S.E.M. for nine to 10 analyzed cells (three to four from each of three independent experiments).
Results
Rat PP Stimulates Y4 Receptors, Whereas hPP and Pro34PYY Stimulate Both Y4 and Y1 Receptors in Mouse Mucosa. First, the sensitivity of WT mouse colonic mucosa to rPP, hPP, and Pro34PYY was compared with tissues from Y4-/- and Y1-/- mice. In WT colon mucosa, all three agonists reduced Isc, but only rPP responses reached a maximum, with a pEC50 value of 7.96 ± 0.01 (Fig. 1A). Responses to rPP were absent from Y4-/- tissue, whereas 30 nM rPP exerted near-maximal effects in WT and Y1-/- mucosae (Fig. 1A), indicating that this peptide selectively stimulates Y4 receptors. In contrast, hPP responses were present in all three genotype tissues (Fig. 1B), but only in Y1-/- mucosa did it reach a maximum with a pEC50 of 8.47 ± 0.13, whereas responses in WT and Y4-/- tissues were linear up to 300 nM. The Y1 receptor-preferring agonist, Pro34PYY (1-300 nM), was significantly more efficacious than rPP or hPP as observed previously (Cox et al., 2001a), and in WT and Y4-/- mucosae, the pEC50 values for Pro34PYY were 7.53 ± 0.06 and 7.59 ± 0.07 (Fig. 1C), respectively. Pro34PYY responses were significantly reduced in Y1-/- colon but not eliminated, suggesting that costimulation of Y1 and Y4 receptors occurs in WT tissues. The basal electrical parameters and VIP responses from mucosae of each genotype are shown in Table 1.
The Y1 or Y2 receptor antagonists BIBO3304 or BIIE0246, respectively, were used in conjunction with tissues from WT or Y4-/- mice to establish the extent to which Y1, Y2, and Y4 receptors contributed to hPP and Pro34PYY responses. Figure 2A shows that hPP responses were inhibited by 58% in BIBO3304-pretreated WT mucosa and were abolished in BIBO3304-pretreated Y4-/- mucosa, indicating costimulation of Y4 and Y1 receptors. Likewise, preincubation with BIBO3304 attenuated Pro34PYY responses by 84% in WT tissues and abolished them in Y4-/- preparations (Fig. 2B). BIIE0246 had no effect on either peptide response in WT or Y4-/- mucosa, demonstrating that Y2 receptors are not involved in these antisecretory responses (Fig. 2, A and B).
Effects of GR231118 in Mouse and Human Colon Mucosae. The effects of GR231118, a Y1 antagonist/Y4 agonist, were compared with BIBO3304 and BIIE0246 antagonism of Y4, Y1, or Y2-mediated responses in WT and Y1-/- mucosae. The Y4-mediated rPP responses were unaffected by the Y1 antagonist, BIBO3304, but they were significantly inhibited by 78 and 82% after GR231118 pretreatment in WT and Y1-/- tissues, respectively (Fig. 3A). Predictably, Pro34PYY responses were abolished by GR231118 or BIBO3304 in WT colon, but neither Y1 antagonist affected the residual responses to Pro34PYY in Y1-/- mucosa (Fig. 3B). PYY(3-36) responses were unaffected by GR231118 but were abolished by the Y2 antagonist, BIIE0246, in both genotypes (Fig. 3C). The α2 adrenoceptor agonist UK14,304 responses were -18.5 ± 3.1 μA/cm2 (n = 19) in WT mucosa and -51.9 ± 13.8 μA/cm2 (n = 16) in Y1-/- mucosa and were not attenuated by any of the antagonist pretreatments (data not shown).
GR231118 had no Y agonist-like actions in colonic mucosae; rather, it increased Isc in WT colon by 9.4 ± 1.3 μA/cm2 (n = 12). Surprisingly, GR231118 also raised Isc in BIBO3304-pretreated WT mucosa by 8.6 ± 1.8 μA/cm2 (n = 12) and in Y1-/- mucosa by 12.5 ± 2.9 μA/cm2 (n = 7). Furthermore, the GR231118 effects were not blocked by Y2 antagonism and were 15.1 ± 6.2 μA/cm2 (n = 5) in Y1-/- tissues pretreated with BIIE0246, suggesting that the effects of the dipeptide are not Y receptor-mediated in these tissues.
GR231118 also increased Isc in human colon mucosa, and this was comparable with the levels of Y1 tone revealed by BIBO3304 (Fig. 4A). In contrast to mouse mucosa, Y1 antagonism with BIBO3304 reversed GR231118 effects (Fig. 4A), converting them to significant reductions in Isc in human colon (Fig. 4B). In addition, GR231118 attenuated subsequent Y4-mediated hPP responses by 55% and abolished Y1 agonist Pro34PYY responses (Fig. 4B). UK14,304 responses were unaffected by any of the manipulations above (data not shown), and the basal electrical parameters for human colon mucosa are shown in Table 1.
Rat PP and GR231118 Are Partial Y4 Agonists in hY4-Transfected HT-29 Monolayers. We investigated the effects of hPP, rPP, and GR231118 in human epithelia expressing hY4 receptors alone. Here, hPP was antisecretory, with a pEC50 of 7.25 ± 0.06. However, rPP (100 nM) responses were significantly smaller than 100 nM hPP (Fig. 5A; P < 0.001), in marked contrast to their similar efficacy in mouse mucosae and their similar binding affinities at the hY4 receptor (below). Responses to higher hPP concentrations were transient in character, and the time-to-peak shortened from 4 (3 nM) to 2 (for 300 nM; Fig. 5B) min. Responses to 100 nM hPP reduced Isc by -9.9 ± 3.0 μA/cm2 (n = 3) and eliminated further responses to hPP added (without washout) 15 to 20 min later (-0.5 ± 0.3 μA/cm2; n = 3; P < 0.05). GR231118 stimulated agonist-like responses in hY4 monolayers (Fig. 5B), but these were slower (time-to-peak 10 min) and less transient than hPP responses. The Y4 agonist-like effects of GR231118 were abolished by 20-min pretreatment with 100 nM hPP (P < 0.05, Fig. 5C). Conversely, hPP responses were significantly attenuated by 56% by GR231118 prestimulation (Fig. 5D). SRIF-14 responses following hPP were -14.4 ± 2.0 μA/cm2, n = 4 and were unaffected by GR231118 (-11.0 ± 1.4 μA/cm2, n = 4).
Binding of Y4 Receptor Ligands to Human and Mouse Orthologs. Because rPP and hPP were equally effective in mouse mucosa, but not in hY4-expressing epithelia, and GR231118 was a low-efficacy agonist in the latter, the agonist binding affinities of each PP at hY4 and mY4 receptors were investigated. The rank order of [125I]hPP displacement from hY4 receptor-transfected HEK293 cells was: hPP ≥ rPP > GR231118 > PYY ≥ Pro34PYY (Fig. 6A; Table 2). In comparison, PYY bound mY4 receptors with a significantly lower affinity than hY4 receptors (P < 0.01), yielding an order of potency of rPP ≥ hPP > GR231118 > Pro34PYY >> PYY (Fig. 6B; Table 2). GR231118 bound to hY4 and mY4 receptors with similar affinities and with pKi values lower than hPP or rPP but greater than PYY. The Y1 receptor antagonist BIBO3304 (1 μM) had no effect on [125I]hPP binding to hY4 receptors but did partially inhibit binding to mY4 receptors by 23.2 ± 2.3% (n = 3; Fig. 6B).
Y4 Receptor Stimulation of GTPγ[35S] Binding. To directly measure the extent of Y4-coupled G protein activation by hPP, rPP, GR231118, or PYY, GTPγ[35S] binding assays were performed in HA-hY4 and mY4 HEK293 cells. In HA-hY4 (Fig. 6C) and mY4 (Fig. 6D) cells, hPP elicited the highest maximal response, with 66 and 41% increases in GTPγ[35S] accumulation over basal levels, respectively. In contrast, PYY was >1000-fold less potent at either hY4 or mY4 receptors, as predicted from the binding studies. Although rPP and hPP exhibited similar binding affinities for hY4 and mY4 receptors (Fig. 6, A and B; Table 2), rPP was less effective at stimulating G protein activation in the hY4 clone with a significantly reduced maximal response at 1 μM (P < 0.001) compared with 1 μM hPP (Fig. 6C). In addition, rPP was 100-fold less potent than hPP in mY4 membranes (Table 2). Furthermore, GR231118 clearly acted as a Y4 agonist, increasing basal GTPγ[35S] binding with comparable pEC50 values to hPP in both hY4 and mY4 membranes (Table 2). However, in each case, the maximal response to GR231118 (1 nM-1 μM) was only 47 to 57% of the 1 μM hPP response.
HA-hY4 Receptor Internalization. HA-hY4 receptors in HEK293 cells were surface-labeled with anti-HA antibody before treatment with peptides and Texas Red-conjugated transferrin to identify early and recycling endosomal compartments of the clathrin-mediated endocytic pathway. Basal HA-hY4 internalization was stimulated further by 15-min incubation with hPP (10 nM-1 μM) but not by rPP or PYY (1 μM; Fig. 7, A and B). Colocalization of punctate internalized Y4 receptors with transferrin (Fig. 7, A and C) increased after hPP. The maximal HA-hY4 internalization (after 1 μM hPP) was substantially lower than NPY-induced internalization of HA-rY1 receptors observed in HEK293 cells (see Supplemental Data; see also Holliday et al., 2005).
A marginal increase in HA-hY4 receptor endocytosis was apparent after 1 μM GR231118 treatment, but the presence of GR231118 did prevent further internalization to 100 nM hPP added 15 min subsequently (Fig. 7, A and B). GR231118 (1 μM) did not affect HA-rY1 receptor distribution, in contrast to a previous report demonstrating internalization of the radiolabeled ligand (Pheng et al., 2003). However, its preincubation efficiently prevented NPY-stimulated Y1 receptor endocytosis (see Supplemental Data).
Discussion
Because of the lack of selective tools, few studies have provided a detailed analysis of specific Y4 receptor functions to date. Our investigations have used colonic tissues expressing Y4 receptors in combination with Y1 and Y2 receptors (from human and mouse) with mucosae lacking either Y4 or Y1 receptors and cells transfected with hY4 or mY4 receptors to provide evidence for an inhibitory role of the Y4 receptor in epithelial ion transport.
When comparing loss of function in WT and Y4-/- colon, the striking feature of knockout tissue was the loss of rPP efficacy, indicating that this peptide inhibits ion secretion solely via Y4 receptors in WT mucosa. In contrast, hPP and Pro34PYY showed significantly greater efficacy than rPP in WT and Y4-/- mucosa, by virtue of their ability to costimulate Y1 and Y4 receptors. However, the sizes of hPP and rPP responses were similar in Y1-/- mucosa, whereas Pro34PYY responses were significantly inhibited. Taken together, these data indicate that hPP and Pro34PYY costimulate murine Y4 and Y1 receptors, hPP being less potent than the PYY analog as a Y1 agonist. Competitive antagonism of hPP and Pro34PYY Y1 responses by BIBO3304 (in WT and Y4-/- tissue) further confirmed activation of mY1 and mY4 receptors, whereas Y2 antagonism (BIIE0246) did nothing to either agonist response. Pro34PYY stimulation of Y4 receptors has been reported (Gerald et al., 1996; Cox et al., 2001b); however, hPP stimulation of mY1 receptors is noteworthy given that we have previously shown hPP to be inactive at human Y1 receptors (Cox and Tough, 2002) and rat colon mucosa (that expresses Y1 but not Y4 receptors) (Tough and Cox, 1996).
Although rPP responses were unaffected by BIBO3304, they were significantly inhibited by GR231118 in both WT and Y1-/- mucosa. Interestingly, rPP exhibited a high affinity for the hY4 as well as the mY4 receptor (see below), so this peptide could be considered a partial hY4 receptor agonist. GR231118 is a well documented Y1 antagonist that also has Y4 agonist properties (Schober et al., 1998; Berglund et al., 2003) but no Y2 binding affinity (Daniels et al., 1995). We found that GR231118 virtually abolished Pro34PYY responses and had no effect on Y2-mediated PYY(3-36) responses in WT mouse mucosa. In human colon, hPP responses were partially inhibited, and Pro34PYY responses were abolished by GR231118. However GR231118 per se produced no discernible inhibitory responses in untreated mouse or human colon but conversely increased Isc. Similar increases have been reported in mouse (Cox et al., 2001a; Hyland and Cox, 2005) and human (Cox and Tough, 2002) mucosa following Y1 antagonists (BIBO3304 and BIBP3226 but not the inactive enantiomer BIBP3435) and are attributed to blockade of endogenous Y1 inhibitory tone. Y1 receptor antagonism accordingly converted GR231118-induced increases in Isc to agonist-like decreases in Isc, thus revealing otherwise masked Y4 activation in human colon. However, in WT mouse mucosa, GR231118-induced increases in Isc were unchanged following Y1 (or Y2) antagonism and occurred to the same extent in Y1-/- mucosa. The NPY C terminus, upon which the GR231118 homodimer is based, evokes histamine release from rat peritoneal mast cells (Grundemar et al., 1994) by a non-Y receptor-mediated mechanism (Mousli et al., 1995). Thus, nonspecific actions of the dipeptide in mouse colon could include mast cell degranulation and subsequent release of secretagogues such as histamine and 5-HT (Keely et al., 1995; Borman and Burleigh, 1996). Under such circumstances, small Y4-mediated decreases in Isc induced by GR231118 would be easily masked by larger amine-mediated secretory responses.
In HT-29 epithelia expressing hY4 receptors through transfection, rPP was virtually inactive at 100 nM, whereas GR231118 was an agonist (as seen previously; Parker et al., 1998; Schober et al., 1998). The inhibitory Y4 responses to a maximal GR231118 concentration were slower, smaller, and less transient than hPP responses in these cells. The more transient character of high-concentration hPP responses (also seen in Y4 constitutively expressing epithelia; Cox et al., 2001b) indicate that hY4 receptors will desensitize following agonist exposure (as do epithelial Y1 receptors; Holliday et al., 2005). These findings contrast with the apparent lack of hY4 desensitization reported previously, perhaps because of the long-term (24-h) PP pretreatment used as a conditioning stimulus in that study (Voisin et al., 2000). More recent evidence suggests that hY4 receptors are able to recruit β-arrestin2, an adaptor protein with well known inhibitory actions on receptor-G protein coupling (Berglund et al., 2003). This interaction provides a plausible mechanism for the Y4 desensitization observed in mouse and human colon mucosa.
The measured binding affinities of Y4 receptor ligands are highly dependent on the choice of displaced radioligand ([125I]hPP, [125I]PYY, [125I]Pro34PYY, or [125I]GR231118; Gehlert et al., 1996, 1997; Matthews et al., 1997; Eriksson et al., 1998; Schober et al., 2000). We used [125I]hPP to radio-label Y4 receptors transfected in HEK293 cells and found that hPP had a similar binding affinity for hY4 and mY4 receptors, whereas rPP was equipotent, and GR231118 was displaced with similar high affinities at each receptor. The order of potency for hY4 cells (hPP ≥ rPP > GR231118 > PYY ≥ Pro34PYY) was the same as that described for hY4 receptors in CHO membranes (Gehlert et al., 1996; Eriksson et al., 1998). Relative peptide affinities at the mY4 receptor (rPP ≥ hPP > GR231118 ≥ Pro34PYY >> PYY) resemble those described for the rY4 receptor (Gehlert et al., 1997; Eriksson et al., 1998).
In contrast, our observed structure-activity relationship for hY4-stimulated GTPγ[35S] binding [hPP >> rPP (low efficacy) > PYY (low potency)] more accurately predicts the relative capacity of these peptides to produce functional responses mediated by endogenous hY4 receptors (Cox et al., 2001b). GR231118 has also been reported to activate Y4 receptors, inhibiting forskolin-induced cAMP accumulation as a full agonist, but with reduced potency compared with hPP (Parker et al., 1998; Schober et al., 1998). In studies that measure downstream signaling events where a Y4 receptor reserve exists, ligands with lower efficacy may still elicit a maximal response by occupying a greater proportion of the Y4 receptor population. GTPγ[35S] assays measure an early step in the cascade following Y4 activation (GDP/GTP exchange by the Giα subunits) and, importantly, before further amplification occurs, e.g. at adenylyl cyclase. Therefore, this assay provides a clear demonstration of the low efficacy that GR231118 has at both the hY4 and mY4 receptors compared with hPP (Fig. 6, C and D; Table 2), through a lower maximum response rather than a decrease in potency, which can be difficult to interpret. A similar discrepancy has been reported for the dopamine D4 receptor ligand, quinpirole, which is a full agonist in cAMP studies but a partial agonist as revealed by GTPγ[35S] assays using the same transfected cells (Gazi et al., 2000). Our GTPγ[35S] observations also indicate that rPP is a full agonist at the mY4 but not the hY4 receptor, despite high binding affinity for both species orthologs. This is consistent with the differential coevolution of endogenous PP and its cognate Y4 receptor in primate and rodent evolutionary lineages (Lundell et al., 1996; Eriksson et al., 1998).
The reduced internalization of agonist-stimulated HA-hY4 compared with HA-rY1 receptors is consistent with previous observations (Voisin et al., 2000; Parker et al., 2001; Gicquiaux et al., 2002; Pheng et al., 2003; Holliday et al., 2005) and with the relative abilities of each receptor to recruit β-arrestin2 as an adaptor for clathrin-coated pits (Berglund et al., 2003). After hPP endocytosis, Y4 receptors were substantially targeted to recycling compartments, suggesting their subsequent redelivery to the cell surface. Although the GR231118 effects on basal and hPP-stimulated Y4 internalization are entirely consistent with partial agonism, our demonstration that the dipeptide does not promote Y1 receptor endocytosis is in contrast to the significant sequestration of Y1 receptors by [125I]GR231118 (in transfected HEK293 cells; Pheng et al., 2003). A direct immunofluorescence approach offers benefits over the measurement of Y receptor internalization by radioligands, which requires the complete removal of surface-bound peptide in a problematic wash step. However, our results do not exclude an alternative mechanism of GR231118 endocytosis, involving sequestration in compartments very close to the cell surface (e.g., caveolae), particularly because inhibitors of clathrin-dependent and -independent pathways could apparently distinguish subtle differences observed between [125I]GR231118 and [125I][Leu31,Pro34]PYY internalization (Pheng et al., 2003). These discrepancies warrant future focused investigation.
In summary, this study demonstrates the species specificity of PP and selective stimulation of Y4 receptors in mouse and human colonic mucosae. The partial Y4 agonist qualities of rPP and GR231118 have been revealed in models expressing the hY4 receptor. Although the dual Y1/Y4 affinities of GR231118 (and nonspecific mucosal effects) would severely limit its therapeutic potential, its properties highlight the feasibility of developing lower efficacy Y4 agonists as novel treatments for gastrointestinal hypersecretory disorders. Such Y4 compounds could moderately inhibit diarrhea without eliminating endogenous physiological signals carried by circulating PP. More importantly, the restricted expression of Y4 receptors in intestinal tissues provides a realistic opportunity for developing PP-based Y4 agonists as novel antidiarrheals with limited side effects.
Acknowledgments
We thank Alex Daniels for providing GR231118 and Herbert Herzog for the WT, Y4-/-, and Y1-/- mice and for the mY4 cDNA used in this study. We also thank Niall Hyland for the hPP concentration-response data in Y1-/- mucosa.
Footnotes
-
This work was supported by a combination of Wellcome Trust and Biotechnology and Biological Sciences Research Council awards (to H.M.C.) and by a Wellcome Trust VIP Award (to N.D.H.).
-
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
-
doi:10.1124/jpet.106.106500.
-
ABBREVIATIONS: PP, pancreatic polypeptide; NPY, neuropeptide Y; PYY, peptide YY; VIP, vasoactive intestinal polypeptide; BIBO3304, (R)-N-[[4-(aminocarbonylaminomethyl)-phenyl]methyl]-N2-(diphenylacetyl)-argininamide trifluoroacetate); BIIE0246, (S)-N2-[[1-[2-[4-[(R,S)-5,11-dihydro-6(6h)-oxodibenz[b,e]azepin-11-yl]-1-piperazinyl]-2-oxoethyl]cyclopentyl]acetyl]-N-[2-[1,2-dihydro-3,5(4H)-dioxo-1,2-diphenyl-3H-1,2,4-triazol-4-yl]ethyl]-argininamide; GR231118 (GR), (Ile,Glu,Pro,Dpr,Tyr,Arg,Leu,Arg,Try-NH2)-2-cyclic(2,4′),(2′,4)-diamide; WT, wild type; m, mouse; r, rat; h, human; SRIF, somatrophin release inhibitory factor (somatostatin-14); GTPγ[35S], guanosine 5′-O-(3-[35S]thio)triphosphate; DMEM, Dulbecco's modified Eagle's medium; HA, hemagglutinin; KH, Krebs-Henseleit; HEK, human embryonic kidney; Isc, short circuit current; UK14,304, 5-bromo-N-(4,5-dihydro-1H-imidazol-2-yl)-6-quinoxalinamine; BSA, bovine serum albumin; ANOVA, analysis of variance; BIBP3226, ((R)-N2-diphenylacetyl)-N-[4-hydroxyphenyl)methyl]-argininamide; BIBP3435, ((S)-N2-diphenylacetyl)-N-[(4-hydroxyphenyl)methyl]-argininamide.
-
↵ The online version of this article (available at http://jpet.aspetjournals.org) contains supplemental material.
- Received April 19, 2006.
- Accepted June 27, 2006.
- The American Society for Pharmacology and Experimental Therapeutics
References
JPET articles become freely available 12 months after publication, and remain freely available for 5 years.Non-open access articles that fall outside this five year window are available only to institutional subscribers and current ASPET members, or through the article purchase feature at the bottom of the page.
|