Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

NPY, a 36-amino-acid amidated peptide, is a member of the pancreatic polypeptide (PP) family, which also includes PP and the intestinal peptide, peptide YY (PYY). NPY is abundantly expressed in neurons and regulates a variety of physiological activities, including food intake (Heinrichs et al., 1998; Schwartz et al., 2000), energy expenditure (Hwa et al., 1999), anxiolysis (Heilig et al., 1989), vasoconstriction (Lundberg et al., 1982), ethanol consumption (Thiele et al., 1998), learning and memory (Wettstein et al., 1995; Redrobe et al., 1999), nociception (Broqua et al., 1996; Bannon et al., 2000) and anticonvulsant activity (Baraban, 1998; Vezzani et al., 1999). A family of six G-protein coupled receptors, termed Y₁, Y₂, Y₃, Y₄, Y₅, and y_6, has been identified that binds NPY, PYY and/or PP with high affinity. The NPY Y₃ receptor has not yet been cloned, and the NPY y6 receptor gene is a pseudogene in primates and is not present in rat (Burkhoff et al., 1998). Unlike NPY, which is expressed ubiquitously in the nervous system, the receptor subtypes show more restricted tissue-specific expression patterns.

NPY is one of a complex network of regulatory molecules produced both centrally and peripherally that act in the hypothalamus to regulate body weight (Williams et al., 2000). Injection of NPY into the cerebral ventricles (Clark et al., 1984) or the paraventricular nucleus (PVN) of the hypothalamus (Stanley and Leibowitz, 1984, 1985) elicits a robust feeding response and reduces body temperature in rodents (Bouali et al., 1995; Currie and Coscina, 1995; Pedrazzini et al., 1998). Chronic administration results in obesity (Vettor et al., 1994). Evidence that endogenous NPY may regulate feeding comes from studies demonstrating that food deprivation increases the content and release of NPY in the PVN and that refeeding reverses these effects (Sahu et al., 1988; Brady et al., 1990; Kalra et al., 1991). Furthermore, leptin, the adipocyte-derived satiety factor, is believed to act as a molecular signal informing NPY-ergic neurons in the PVN of the status of peripheral energy stores. High leptin levels resulting from increased adiposity suppress NPY production, whereas low leptin levels are permissive for hypothalamic NPY production (Campfield et al., 1996). The obesity syndrome of leptin-deficient ob/ob mice is partially attenuated in animals that are deficient in both leptin and NPY (Erickson et al., 1996).

Pharmacological and genetic approaches have been used to identify the NPY receptor subtypes that mediate feeding. The ability of various NPY analogs to stimulate food intake in rodents generally matches their in vitro affinities for the Y₁ and Y₅ receptor subtypes (Gerald et al., 1996; Hu et al., 1996). Additionally, Y₁- and Y₅-selective receptor antagonists have been shown to reduce spontaneous or NPY-stimulated food intake in rodents (Rudolf et al., 1994; Doods et al., 1995; Kanatani et al., 1996; Matthews et al., 1997; Criscione et al., 1998; Ishihara et al., 1998; Kask et al., 1998; Wieland et al., 1998). In vivo experiments in which expression of the Y₁ or Y₅ receptor subtypes has been blocked by antisense DNA (Lopez-Valpuesta et al., 1996; Tang-Christensen et al., 1998) or by targeted disruption of these genes (Marsh et al., 1998; Pedrazzini et al., 1998; Kanatani et al., 2000) have confirmed the involvement of the Y₁ and Y₅ receptor subtypes in feeding behavior. Within the hypothalamus, a region known to be involved in central regulation of feeding, the NPY Y₁ and/or Y₅ receptor subtypes are expressed in the paraventricular, medial preoptic, supraoptic, and arcuate nuclei and in the lateral hypothalamus (Gerald et al., 1996; Durkin et al., 2000). These areas are near the third ventricle; hence, the role of the Y₁ and Y₅ receptors in feeding can be probed by i.c.v. injection of peptide and nonpeptide agonists and antagonists.

Although selective Y₅ receptor agonists have been identified (Cabrele et al., 2000; Parker et al., 2000), the lack of Y₁-selective agonists has hampered the study of this receptor's role in feeding and obesity. In earlier studies several NPY analogs containing single-point D-amino acid substitutions were found to have different binding affinities for the NPY Y₁ and Y₂ receptor subtypes (Kirby et al., 1993a, 1995). Kirby et al. (1995) also identified a truncated NPY analog, Des-AA^10-17[Cys^7,21,Pro³⁴]-NPY, that is selective for the NPY Y₁ receptor over the Y₂ receptor. To determine whether any of these peptides might bind preferentially to the Y₁ receptor relative to the other NPY receptor subtypes, their affinities for all NPY receptors were measured. We report here the identification of several potent and selective NPY Y₁ receptor agonists and their effect on food intake in rats.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Peptide Synthesis and Characterization. Peptides were assembled by solid phase peptide synthesis techniques using the tertiary-butyloxycarbonyl strategy. Peptides were purified by preparative high-performance liquid chromatography and characterized using capillary zone electrophoresis, mass spectrometry, and analytical high-performance liquid chromatography as described previously (Kirby et al., 1993b).

Cloning and Expression of the NPY Receptors. Cloning and expression of the human and rat NPY Y₁, Y₂, Y₄, and Y₅ receptors and the construction of a chimeric rat/human NPY Y₅ receptor were described previously (Parker et al., 1998).

Cell Culture. Cell lines expressing one of the NPY receptor subtypes were grown in F12 medium (CHO-K1 cells) or Dulbecco's modified Eagle's medium (HEK 293 cells) supplemented with 10% (v/v) heat-inactivated fetal bovine serum, 2 mM glutamine, 100 units/ml penicillin, 100 µg/ml streptomycin, and 200 µg/ml G418 or Zeocin.

Competition Binding Assay. Cell membranes were prepared as described by Parker et al. (1998). Membrane protein (5-10 µg) was incubated with 0.2 nM ¹²⁵I-porcine PYY (NPY Y₁, Y₂, and Y₅ receptors) or ¹²⁵I-human PP (NPY Y₄ receptor) and the nonradiolabeled peptide of interest (10¹¹-10⁶ M) in 50 mM HEPES, pH 7.2, 2.5 mM CaCl₂, 1 mM MgCl₂, and 0.1% (w/v) bovine serum albumin in a total volume of 200 µl for 90 min at room temperature (NPY Y₂, Y₄, and Y₅ receptors) or at 30°C (NPY Y₁ receptor). Nonspecific binding was determined using 10⁶ M human/rat NPY (NPY Y₁, Y₂, and Y₅ receptors) or human PP (NPY Y₄ receptor). The reaction mixtures were filtered through Millipore MAFC glass fiber filter plates presoaked in 0.5% (v/v) polyethylenimine. The filters were washed twice with 150 µl of ice-cold Dulbecco's phosphate-buffered saline, and the filter-bound radioactivity was measured in a Packard TopCount scintillation counter (Packard, Meriden, CT).

cAMP Assay. HEK 293 cells expressing the human NPY Y₁ receptor, the rat NPY Y₄ receptor, or the rat NPY Y₅ receptor or CHO cells expressing the human NPY Y₂ receptor were used to measure cAMP production. Cells were plated at 1.5 × 10⁴ cells/well in 96-well dishes and reached confluence in 3 days. The cell monolayers were washed once in Hanks' balanced salt solution (HBSS), then incubated for 20 min at 37°C in HBSS containing 10 mM HEPES, pH 7.2, 4 mM MgCl₂, 0.2% (w/v) bovine serum albumin, and 1 mM 3-isobutyl-1-methylxanthine. This solution was then replaced with HBSS containing 10 mM HEPES, pH 7.2, 4 mM MgCl₂, 0.2% (w/v) bovine serum albumin, 1 mM 3-isobutyl-1-methylxanthine, forskolin (HEK 293 cells, 2.5 µM; CHO cells; 5 µM) and NPY or an NPY analog (10¹⁵-10⁴ M). The incubation was allowed to continue for an additional 10 min. Intracellular cAMP was then extracted with 75 µl of 100% ethanol and quantified by radioimmunoassay (Flash Plate; PerkinElmer, Boston, MA).

Animals. Adult male Long-Evans rats (250-300 g) were maintained in individual cages at 22°C with free access to food (Teklad rodent diet 8604; Harlan, Madison, WI) and water. The rats were exposed to a 12 h/12 h light/dark cycle with lights on at 4:00 AM. All studies were conducted in a facility accredited by the Association for Assessment and Accreditation of Laboratory Animal Care following protocols approved by the Animal Care and Use Committee of Schering-Plough Research Institute. The procedures were performed in accordance with the principles and guidelines established by the National Institutes of Health for the care and use of laboratory animals.

Surgery. Long-Evans rats were anesthetized by intramuscular injection of ketamine (100 mg/kg) and xylazine (10 mg/kg). A 22-gauge stainless steel cannula was implanted into the lateral ventricle using the following stereotaxic coordinates for cannula placement: 1.0 mm posterior to bregma, 1.5 mm lateral to midline, and 3.6 mm ventral to dura. Three weeks after surgery, the correct placement of the cannula was ascertained by measuring food intake after i.c.v. administration of NPY (0.3 nmol). Rats consuming at least 2.0 g of food within 60 min of NPY administration were judged to have correctly implanted cannulas.

Intracerebroventricular Infusion Protocols. All peptides administered to rats were dissolved in 0.9% sterile NaCl and administered at doses of 0.1, 0.3, or 1.0 nmol. Intracerebroventricular infusions were given in the middle of the light cycle (i.e., at 10:00 AM) to satiated rats. Peptides were infused into the lateral ventricle at 5 µl/min using a Hamilton infusion pump and syringe (Bioanalytical Systems, West Lafayette, IN). The guide cannula remained inserted for 1 min after infusion to prevent backflow of the infusion solution up the needle track. A known amount of chow was made available to the rats immediately after peptide infusion. Food consumption was measured by subtracting the amount of chow remaining 1, 2, or 4 h after peptide infusion from the amount presented at the beginning of the experiment. Baseline food consumption was determined 3 days before peptide administration by measuring food intake 1, 2, and 4 h after i.c.v. infusion of 0.9% NaCl.

Data Analysis. Binding and cAMP data were analyzed by nonlinear regression analysis using Prism (Graph Pad, San Diego, CA). In vivo results are given as means ± S.E. For each Y₁ agonist, changes in food intake were analyzed using analysis of variance followed by Dunnett's multiple comparisons test. Statistical significance was assessed by the p < 0.05 between the effect of the saline control and each dose of agonist or between the agonist alone and the agonist plus an antagonist.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Two Y₁-selective NPY analogs were identified from the complete series of D-amino acid substituted peptides, [D-Arg²⁵]-NPY (1) and [D-His²⁶]-NPY (2). 1 binds the Y₁ receptor with 12-, 82-, and 48-fold higher affinity than the Y₂, Y₄, and Y₅ receptor subtypes, respectively. Similarly, the affinity of 2 for the Y₁ receptor is 14-, 10-, and 17-fold greater than for the Y₂, Y₄, and Y₅ receptor subtypes, respectively. The affinity of 1 for the NPY Y₁ receptor is 3.2 times less than that of NPY, whereas the affinity of 2 is 7.1 times less than that of NPY (Table 1). Substitution of both residues{[D-Arg²⁵,D-His²⁶]-NPY (3)} yielded a molecule with a higher degree of selectivity for the NPY Y₁ receptor over the Y₂, Y₄, and Y₅ receptors but with a significantly higher Y₁ K_i value (9.7 nM) than either 1 or 2 (Table 1).

A second set of cyclized NPY analogs lacking the internal residues that make up the putative  turn were also evaluated for receptor subtype selectivity. Des-AA^10-17[Cys^7,21,Pro³⁴]-NPY (4) binds the NPY Y₁ receptor with a K_i value of 0.2 nM, whereas Des-AA^11-18[Cys^7,21,D-Lys⁹(Ac)]-NPY (5) has a K_i value of 0.7 nM. Both peptides are greater than 100-fold selective for the NPY Y₁ over the NPY Y₅ receptor (Table 1). However, 4 binds the NPY Y₄ receptor with relatively high affinity (5.0 nM), and 5 binds the NPY Y₂ receptor with a K_i value of 1.4 nM. Des-AA^11-18[Cys^7,21,D-Lys⁹(Ac),Pro³⁴]-NPY (6) has lower affinity for the Y₂ receptor (K_i = 124 nM) and higher affinity for the Y₁ receptor (K_i = 0.2 nM) but also higher affinity for the NPY Y₄ receptor (K_i = 7.3 nM). Substitution of D-His at position 26 {Des-AA^11-18[Cys^7,21,D-Lys⁹(Ac), D-His²⁶]-NPY (7)} results in marked loss of affinity for the NPY Y₁ (K_i = 2.2 nM), NPY Y₄ (K_i = 206 nM), and NPY Y₅ (K_i = 3023 nM) receptors but an increased affinity for the NPY Y₂ receptor (1.4 nM), and thus a loss of Y₁/Y₂ selectivity. However, substitution of both D-His²⁶ and Pro³⁴ {Des-AA^11-18[Cys^7,21,D-Lys⁹(Ac),D-His²⁶,Pro³⁴]-NPY (8)} yields a peptide that is 645-fold selective for the NPY Y₁ receptor over the NPY Y₂ receptor, 26-fold selective for the NPY Y₁ receptor over the NPY Y₄ receptor, and 1900-fold selective for the NPY Y₁ receptor over the NPY Y₅ receptor, although it retains a K_i value of 1.2 nM for the Y₁ receptor.

The functional activity of these NPY Y₁-selective peptides was determined by measuring their ability to block forskolin-stimulated cAMP production in cell lines expressing one of the NPY receptor subtypes. The results are shown in Table 2 and Fig. 1. The EC₅₀ value of NPY for the human NPY Y₁ receptor expressed in HEK 293 cells is 0.6 nM. All of the NPY Y₁-selective peptides tested are full agonists for the NPY Y₁ receptor, exhibiting varying degrees of potency ranging from an EC₅₀ value of 0.2 nM for 1 to 13.6 nM for 3. 1, 2, 7, and 8 are highly selectivity for the NPY Y₁ receptor subtype, whereas 4, 5, and 6 are also potent agonists for either the NPY Y₂ or NPY Y₄ receptors.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Inhibition of forskolin-stimulated cAMP production by NPY and selective NPY Y1 receptor agonists. HEK 293 cells expressing the human NPY Y1 receptor were exposed to forskolin (5 µM) and NPY () and 1 [D-Arg25]-NPY, 2 [D-His26]-NPY, 4 Des-AA10-17[Cys7,21,Pro34]-NPY, 5 Des- AA11-18[Cys7,21,D-Lys9(Ac)]-NPY, 6 Des-AA11-18[Cys7,21,D-Lys9(Ac), Pro34]-NPY, 7 Des-AA11-18[Cys7,21,D-Lys9(Ac), D-His26]-NPY or 8 Des-AA11-18 [Cys7,21,D-Lys9(Ac), D-His26,Pro34]-NPY () for 10 min. Intracellular cAMP levels were measured by radioimmunoassay. The values shown are the average ± S.E.M. of three experiments.

Four of the peptides identified as selective NPY Y₁ receptor agonists in vitro, 1, 2, 4, and 8, were compared with NPY for their ability to stimulate food intake in satiated Long-Evans rats. The peptides were administered at doses of 0.1, 0.3, or 1.0 nmol, and food consumption was monitored over a 4-h period (1, 2, and 8) or a 1-h period (4). Three of the agonists tested (1, 2, and 8) stimulated food intake in a dose- and time-dependent manner (Fig. 2, a-c). The increase in food consumption observed after administration of the three active NPY Y₁ agonists or NPY was apparent within 1 h and was sustained for at least 4 h. Relative to the saline control group, NPY and 2 significantly stimulated food intake 1, 2, and 4 h after dosing at doses of 0.3 and 1.0 nmol, whereas 1 significantly stimulated food intake at a dose of 1 nmol 1 and 2 h after administration and at the 0.3 nmol and 1 nmol doses 4 h after administration. The stimulation of food intake elicited by 8 was not statistically significant relative to the saline control but did show a significant linear (p < 0.05) overall dose-dependent trend at each time point. 4 did not stimulate food intake after 1 h at the 1.0 nmol dose relative to the saline control (data not shown).

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Stimulation of food intake by NPY and selective NPY Y1 receptor agonists. Cannulated Long Evans rats were injected i.c.v. with 0.9% NaCl (0 nmol) or peptide at doses of 0.1, 0.3 or 1 nmol. Food intake was measured 1 h (A), 2 h (B), and 4 h (C) after dosing. The values are the mean ± S.E.M. (n = 6). *p < 0.05; values are significantly different from the saline control group.

The dimeric nonapeptide 1229U91 (also referred to as GR231118 and GW1229) is a potent and selective antagonist of the NPY Y₁ receptor (Daniels et al., 1995) and has been reported to inhibit NPY-stimulated food intake in rats (Kanatani et al., 1996; Ishihara et al., 1998; Widdowson et al., 1999). We examined the ability of 1229U91 to block feeding stimulated by 1 (Fig. 3). Two-hour food intake was stimulated 2.7-fold by 1 (0.3 nmol). Coadministration of 1 (0.3 nmol) and 1229U91 (3 nmol) completely inhibited 1-stimulated food intake.

View larger version (29K): [in this window] [in a new window]   Fig. 3.   Inhibition of [D-Arg25]-NPY-stimulated food intake by 1229U91. Cannulated Long Evan rats were injected i.c.v. with 0.9% NaCl, 3 nmol of 1229U91, 0.3 nmol of 1, or 3 nmol of 1229U91 and 0.3 nmol of 1. Food intake was measured 1 h after dosing. The values are the mean ± S.E.M. (n = 11). *p < 0.05; value is significantly different from the saline control group. #p < 0.001; value is significantly different from 1 (0.3 nmol).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The study of the physiological responses mediated by each NPY receptor subtype has been facilitated by the identification of subtype-selective agonists. Agonists selective for the Y₂ receptor [C2-NPY (McLean et al., 1990)], the Y₄ receptor [rat PP (Bard et al., 1995)], and the Y₅ receptor {[DTrp³²]NPY (Balasubramaniam et al., 1994), [DTrp³⁴]NPY (Parker et al., 2000), [Ala³¹,Aib³²]NPY, [hPP^1-17,Ala³¹,Aib³²]hNPY, and [cPP^1-7,NPY^19-23,Ala³¹, Aib³²,Gln³⁴]hPP (Cabrele et al., 2000)} have been identified but, until now, no Y₁-selective agonist has been reported. Originally, [Leu³¹,Pro³⁴]-NPY had been described as Y₁-selective (Fuhlendorff et al., 1990). The Y₂ binding affinity of NPY was eliminated in [Leu³¹,Pro³⁴]-NPY by removing the C-terminal hexapeptide, which is known to be required for Y₂ binding, and substituting the C-terminal hexapeptide of PP. However, after identification of the Y₄ and Y₅ receptor subtypes, it was found that [Leu³¹,Pro³⁴]-NPY also has significant affinity and potency at these receptors (Gerald et al., 1996; Hu et al., 1996).

The identification of 1 and 2 as Y₁-selective agonists was based upon the results of an earlier study in which it was determined that substitution of the corresponding D-amino acid at either position 25 or 26 of NPY results in loss of Y₂ affinity but maintains Y₁ affinity (Boublik et al., 1989; Kirby et al., 1993a). 1 and 2 also inhibit forskolin-stimulated cAMP production in cells expressing the Y₁ receptor with EC₅₀ values similar to that of NPY but are significantly less potent activators of the Y₂ receptor. The present study demonstrates that 1 and 2 are also selective for the Y₁ receptor relative to the Y₄ and Y₅ receptors. The loss of Y₂, Y₄, and Y₅ receptor binding upon changing the chirality of a single amino acid in the C-terminal -helical segment of the molecule suggests that these receptors have more stringent requirements for the native conformation of this region of NPY than the Y₁ receptor. Work is in progress to further clarify these requirements.

The design of the internally truncated Y₁ agonists (4-8) was based on earlier observations that the putative  turn (amino acids 9-14) constituting the PP fold can be deleted without loss of Y₁ affinity, provided that the N- and C-terminal helices are held in close juxtaposition by a covalent linkage (Kirby et al., 1993b, 1995, 1997). Kirby et al. (1993b) found that a disulfide linkage between Cys⁷ and Cys²¹ was superior to other interchain bridges examined for maintaining Y₁ affinity. 4 and 5 had very high Y₁ affinity and potency but also had high affinity and potency at the Y₄ and Y₂ receptors, respectively. Substitution of Pro³⁴ in 5 to yield 6 reduced the affinity for Y₂ but, not unexpectedly, increased the affinity for Y₄. Because 2 is approximately 15-fold selective for Y₁ over Y₂ (K_i = 2.0 nM versus 29 nM), a D-His substitution in 5 was made to yield 7. This modification did not result in the intended loss of affinity for Y₂ but did result in a significant loss of affinity for Y₄. Thus, both Pro³⁴ and D-His²⁶ were included in 8, producing a peptide with excellent Y₁ selectivity and adequate Y₁ affinity and potency to be useful as a biological tool.

The stimulation of food intake by 1, 2, and 8 is the first demonstration that selective activation of the Y₁ receptor in animals expressing both the Y₁ and Y₅ receptors can induce feeding. The in vivo activity of both 1 and 2 is comparable with that obtained using equal doses of NPY. The stimulation of food intake observed after i.c.v. administration of 8 was more modest, and 4 failed to elicit food intake when administered at doses as high as 1 nmol. The failure of 4 to stimulate food intake was unexpected, because its K_i and EC₅₀ values at the Y₁ receptor are lower than those of NPY. Rapid elimination, degradation, or poor solubility may cause the poor efficacy of both cyclized analogs tested in vivo, limiting their access to the receptor sites. The complete inhibition of 1-stimulated food intake by the Y₁ selective antagonist 1229U91 confirms that the activity of 1 is not mediated by the Y₅ receptor. The 70% reduction in food intake measured after i.c.v. administration of 1229U91 alone suggests that Y₁ activation contributes to the spontaneous feeding and underscores the importance of the Y₁ receptor subtype in appetitive behavior.

The relative roles of the NPY Y₁ and Y₅ receptors in the regulation of food intake and energy expenditure are still unclear. The Y₁-selective agonists described here will be useful in characterizing more completely the role of the Y₁ receptor, not only in feeding behavior, but also in other physiological activities.