help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2005-1373
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Creemers, J. W. M.
Right arrow Articles by White, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Creemers, J. W. M.
Right arrow Articles by White, A.
Endocrinology Vol. 147, No. 4 1621-1631
Copyright © 2006 by The Endocrine Society

Agouti-Related Protein Is Posttranslationally Cleaved by Proprotein Convertase 1 to Generate Agouti-Related Protein (AGRP)83–132: Interaction between AGRP83–132 and Melanocortin Receptors Cannot Be Influenced by Syndecan-3

John W. M. Creemers1, Lynn E. Pritchard1, Amy Gyte, Philippe Le Rouzic, Sandra Meulemans, Sharon L. Wardlaw, Xiaorong Zhu, Donald F. Steiner, Nicola Davies, Duncan Armstrong, Catherine B. Lawrence, Simon M. Luckman, Catherine A. Schmitz, Rick A. Davies, John C. Brennand and Anne White

Department of Human Genetics (J.W.M.C., S.M.), University of Leuven and Flanders Interuniversity Institute for Biotechnology, B-3000 Leuven, Belgium; School of Medicine (L.E.P., A.G., A.W.) and Faculty of Life Sciences (L.E.P., A.G., P.L.R., N.D., C.B.L., S.M.L., A.W.), University of Manchester, Manchester M13 9PT, United Kingdom; Department of Medicine (S.L.W.), Columbia University College of Physicians and Surgeons, New York, New York 10032; Department of Biochemistry and Molecular Biology (X.Z., D.F.S.), University of Chicago, Chicago, Illinois 60637; and AstraZeneca (D.A., C.A.S., R.A.D., J.C.B.), Mereside, Cheshire SK10 4TG, United Kingdom

Address all correspondence and requests for reprints to: Professor Anne White, Stopford Building, University of Manchester, Oxford Road, Manchester M13 9PT, United Kingdom. E-mail: awhite{at}man.ac.uk.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Agouti-related protein (AGRP) plays a key role in energy homeostasis. The carboxyl-terminal domain of AGRP acts as an endogenous antagonist of the melanocortin-4 receptor (MC4-R). It has been suggested that the amino-terminal domain of AGRP binds to syndecan-3, thereby modulating the effects of carboxyl-terminal AGRP at the MC4-R. This model assumes that AGRP is secreted as a full-length peptide. In this study we found that AGRP is processed intracellularly after Arg79-Glu80-Pro81-Arg82. The processing site suggests cleavage by proprotein convertases (PCs). RNA interference and overexpression experiments showed that PC1/3 is primarily responsible for cleavage in vitro, although both PC2 and PC5/6A can also process AGRP. Dual in situ hybridization demonstrated that PC1/3 is expressed in AGRP neurons in the rat hypothalamus. Moreover, hypothalamic extracts from PC1-null mice contained 3.3-fold more unprocessed full-length AGRP, compared with wild-type mice, based on combined HPLC and RIA analysis, demonstrating that PC1/3 plays a role in AGRP cleavage in vivo. We also found that AGRP83–132 is more potent an antagonist than full-length AGRP, based on cAMP reporter assays, suggesting that posttranslational cleavage is required to potentiate the effect of AGRP at the MC4-R. Because AGRP is cleaved into distinct amino-terminal and carboxyl-terminal peptides, we tested whether amino-terminal peptides modulate food intake. However, intracerebroventricular injection of rat AGRP25–47 and AGRP50–80 had no effect on body weight, food intake, or core body temperature. Because AGRP is cleaved before secretion, syndecan-3 must influence food intake independently of the MC4-R.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
THERE IS LITTLE doubt that the melanocortin-4 receptor (MC4-R) plays an important role in coordinating appetite and metabolic rate with perceived metabolic requirement (reviewed in Ref.1). In this regard, two sets of neurons in the hypothalamic arcuate nucleus are particularly important: a) Proopiomelanocortin (POMC) neurons, which generate endogenous ligands for the MC4-R such as {alpha}MSH and ACTH (2), and agouti-related peptide (AGRP)/neuropeptide Y neurons, which generate AGRP, an endogenous MCR antagonist (3, 4). Both sets of neurons are sensitive to a wide range of peripheral signals that indicate metabolic status, such as leptin, insulin, glucocorticoids, and gut hormones (5, 6). AGRP expression is up-regulated in situations of negative energy balance (3, 4, 7, 8, 9). Also, genetic manipulation of AGRP expression levels (4, 10, 11) and physiological experiments (12, 13, 14, 15) demonstrate that AGRP has a potent and long-term anabolic effect on food intake and metabolic rate. Neuroanatomical data and pharmacological studies support the view that AGRP has this effect because it acts as a competitive antagonist at the MC4-R (16, 17, 18, 19), although alternative mechanisms have been proposed (20, 21, 22).

Despite the well-established role of AGRP in regulation of energy homeostasis, surprisingly little is known of its posttranslational regulation in the hypothalamus. This information is required to understand fully the physiological role of AGRP and the mechanism(s) by which it exerts its effects. To date, most physiological studies of AGRP function in vivo have used a chemically synthesized carboxyl-terminal AGRP fragment, AGRP83–132 (23). Pharmacological studies undertaken in vitro have indicated that this peptide and similar carboxyl-terminal derivatives, such as AGRP87–132, are sufficient to antagonize the MC4-R (16, 18, 24). Moreover, HPLC analysis of AGRP immunoreactivity in rat hypothalamic extracts indicate that AGRP undergoes posttranslational cleavage to generate a carboxyl-terminal fragment in vivo and very little full-length AGRP remains (25, 26). Likely candidate proteases that may be involved in AGRP processing include proprotein convertase (PC) 1/3, PC2, and PC5/6a, all of which have a neuroendocrine expression profile and are expressed in the hypothalamus (27).

However, observations that suggest AGRP is cleaved are not consistent with the current model of how AGRP and POMC derived peptides interact at the MC4-R in vivo. It has been proposed that syndecan-3, a central nervous system-specific proteoglycan that is implicated in food intake regulation, acts as a coreceptor for MC4-R by binding to amino-terminal AGRP via its heparin sulfate side chains and presenting carboxyl-terminal AGRP to the MC4-R (28, 29). This model is consistent with observations in genetically manipulated mouse models (28, 30) but implies that AGRP is secreted as a full-length molecule.

Based on these contradictory lines of evidence, it is important to determine whether AGRP undergoes posttranslational cleavage because if the syndecan-3 model is correct, then most physiological studies have been based on peptides that are not produced in vivo, and more appropriate studies are needed using full-length AGRP (30). In addition, if carboxyl-terminal AGRP fragments are produced by posttranslational cleavage, then one or more amino-terminal fragments must also be produced. These peptides may exert important physiological effects that are independent of the melanocortin system. Finally, if AGRP is posttranslationally processed, then the processing pathway may be tightly regulated as a means of controlling the amount of melanocortin antagonist synthesized and secreted at any given time.

In this present study we addressed five questions: 1) is AGRP posttranslationally processed; 2) which PCs (if any) are capable of cleaving AGRP and which AGRP peptides are secreted; 3) which proprotein convertases are expressed in AGRP neurons; 4) what are the relative potencies of secreted AGRP peptides at the MC4-R; and 5) what are the physiological effects of amino-terminal AGRP peptides in rats?


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Generation of human AGRP expression constructs
Full-length AGRP, including the signal peptide and a consensus Kozak initiation signal, was PCR amplified from a human hypothalamus cDNA library (Clontech Laboratories Inc., Mountain View, CA) using the following primers: sense, 5'-AGCCAGGCCATGCTGACCGCAGCGTTGC-3', antisense, 5'-CCCAAGCTTCTAGGTGCGGCTGCAGGGATT-3'. A separate reverse primer was designed to generate a Flag epitope-tagged version of full-length AGRP: 5'- CCCAAGCTTCTACTTGTCGTCGTCGTCCTTGTAGTCGGTGCGGCTGCAGGGATT-3'. PCR products were directly cloned into pCR-BluntII-TOPO (Invitrogen, Carlsbad, CA). Inserts were excised with EcoRI and subsequently cloned into pcDNA3+ (Invitrogen). Clones in the correct orientation were verified by sequence analysis. Subsequently, a series of mutated clones were generated in which Arg79-X-X-Arg82, Arg85-Arg86, and Arg86-X-X-Arg89 were converted to Ala79-X-X-Ala82, Ala85-Ala86, and Ala86-X-X-Ala89. Mutagenesis was undertaken using a Quickchange site-directed mutagenesis kit (Stratagene, La Jolla, CA) following the manufacturer’s protocol. A further two constructs were generated, in which internal FLAG epitope tags were incorporated immediately adjacent to the Arg79-X-X-Arg82 and Arg85-Arg86 sites. For the Arg79-X-X-Arg82-FLAG construct, two PCR products were generated using the following primer pairs: sense, 5'-AGCCAGGCCATGCTGACCGCAGCGTTGC-3', antisense, 5'-CTTGTCGTCGTCGTCCTTGTAGTCGCGGGGCTCGCGGTCCTG-3' and sense, 5'-GACTACAAGGACGACGACGACAAGTCCTCACGTCGCTGCGTA-3', antisense, 5'-CCCAAGCTTCTAGGTGCGGCTGCAGGGATT-3'. These two PCR products were mixed together, diluted, and amplified with the original full-length AGRP primers. The PCR product was cloned into pCR-BluntII-TOPO (Invitrogen) and subsequently pcDNA3+ (Invitrogen). For the Arg85-Arg86-FLAG construct, the following PCR primer pairs were used: sense, 5'-AGCCAGGCCATGCTGACCGCAGCGTTGC-3', antisense, 5'-CTTGTCGTCGTCGTCCTTGTAGTCGCGACGTGAGGAGCGGGG-3' and sense, 5'-GACTACAAGGACGACGACGACAAGTGCGTAAGGCTGCATGAGT-3', antisense, 5'-CCCAAGCTTCTAGGTGCGGCTGCAGGGATT-3'.

Transfection of mammalian cells and analysis of AGRP processing
AtT20 and ßTC3 cells were transfected using Lipofectamine (Invitrogen) and {alpha}TC1–6 cells using Lipofectamine 2000 (Invitrogen) as described previously (31). Regulated secretion experiments were performed essentially as described (32), except that secretion was induced for 3 h using 60 mM KCl. A truncated soluble form of furin (32) was used as a control for constitutive secretion. Albumin (25 µg/ml) was added to the medium samples before precipitation with 4 volumes of methanol at –20 C. Medium precipitates and cells were dissolved in sample buffer and size separated by SDS-PAGE. Western blotting was performed as described (33) using mouse anti-FLAG antibodies M1 or M2 (Sigma-Aldrich, St. Louis, MO) or a rabbit antibody directed against AGRP that recognizes both pro-AGRP and AGRP (kindly provided by Dr. G. Barsh, Stanford University School of Medicine, Stanford, CA).

Immunocytochemistry
Indirect immunofluorescence microscopy was performed as described (34) with some modifications. Briefly, AtT20 cells, fixed in 4% paraformaldehyde, were incubated with mouse anti-FLAG M1 antibody and a rabbit antibody directed against the amino terminus of POMC (kindly provided by Dr. P. Lowry, University of Reading, Reading, UK) diluted in PBS containing 0.5% blocking reagent (Roche, Indianapolis, IN) and 0.2% Triton X-100. Bound antibodies were detected with fluorescently labeled secondary antibodies (Alexa dyes; Molecular Probes Inc., Eugene, OR). Slides were mounted in Vectashield mounting medium (Vector Laboratories, Burlingame, CA) and analyzed on a Axiophot fluorescence microscope (Carl Zeiss, Inc., Oberkochen, Germany) equipped with UV optics. Images were recorded with a CE200A charge-coupled device camera system (Photometrics Inc., Huntington Beach, CA) using SmartCapture (Digital Scientific, Cambridge, UK) software.

Immunoelectron microscopy
Ultratructural analysis was performed based on the preembedding immunolabeling procedure described by Yi et al. (35). Cells were fixed in 3% paraformaldehyde and 0.15% glutaraldehyde. After quenching with 0.1% NaBH3, the cells were permeabilized with 0.035% Triton X-100 and incubated with primary antibody (1:1000 dilutions of rabbit anti-AGRP polyclonal or mouse anti-ACTH monoclonal). Ultrasmall gold-conjugated secondary antibodies (goat antirabbit or goat antimouse IgG; both Aurion, Wageningen, The Netherlands) were used at 1:100 dilutions. After postfixation in 2% glutaraldehyde, silver enhancement was performed using Aurion R-Gent SE-EM reagent (Aurion), according to the guidelines of the supplier. Finally, the cells were osmicated in 0.5% OsO4 and embedded in Agar 100 Resin (Agar Scientific, Essex, UK) Ultrathin sections were cut using the Leica Ultracut UCT ultramicrotome and stained with uranyl acetate and lead citrate. The sections were analyzed on a CM10 transmission electron microscope (Philips, Amsterdam, The Netherlands).

RNA interference
The 19-mer target sequences for fur and Pcsk6 (PACE4 gene) have been described previously (36). The 19-mer target regions of Pcsk1 (PC1/3), and Pcsk5 (PC5/6A) for RNA interference were selected using small interfering RNA (siRNA) Target Finder (Ambion: http://ambion.com/techlib/misc/siRNA_finder.html). The target sequences for Pcsk1 and Pcsk5 are 5'-GAAGCGCTCTTCATATCAC-3', and GACCATTCGACCAAACAGT-3', respectively. Upper and lower 60-mer oligonucleotides encoding the corresponding short hairpin (sh) RNAs were designed using pSilencer Converter (Ambion: http://ambion.com/techlib/misc/psilencer_converter.html). shRNAs contain the 19-mer target sequence, a short hairpin loop sequence (TTCAAGAGA) and the antisense target sequence, flanked by sequences necessary for RNA polymerase III termination (TTTTTT) and cloning. The double-stranded oligonucleotides were cloned in the mU6pro vector, kindly provided by Dr. D. Turner (37). PC2 was silenced using an engineered {delta}-ribozyme system kindly provided by Dr R. Day (38). The efficiency was confirmed by cotransfection of 0.8 µg of mU6 pro vector encoding shRNAs or the {delta}-ribozyme, with 0.2 µg expression vectors encoding the target mRNA and 1 µg empty vector.

Generation of recombinant full-length AGRP
Full-length AGRP (minus predicted signal peptide) was expressed in Escherichia coli, purified, and refolded essentially as previously described (17, 39). Briefly AGRP was PCR amplified from a human hypothalamus cDNA library (Clontech) using the primers: sense, 5'-CGGGATCCGGCTTGGCCCCCAT-3', antisense, 5'-CCCAAGCTTCTAGGTGCGGCTGCAGGGATT-3'. The PCR product was digested with BamHI and HindIII and cloned into pT7.36His, an in-house vector that incorporates a 6-His tag. E. coli BL21 (Invitrogen) were transformed with pT7.36His-AGRP. Expression of the recombinant protein was induced in the presence of 0.4 mmol/liter isopropyl-ß-D-thiogalactopyranoside and purified on a Ni-NTA agarose column using the QIAexpress kit (QIAGEN, Crawley, UK) according to manufacturer’s protocols. Recombinant AGRP was refolded following the protocol of Rosenfeld et al. (17). Fifty microliters of refolded material were subjected to analytical size-exclusion chromatography on a 2.4-ml Superdex 75 column (Amersham Biosciences, Chalfont St. Giles, UK) equilibrated in 50 mM Tris-HCl and 0.15 M NaCl (pH7.4). The column was eluted with the same buffer at 50 µl/min. Protein concentration was determined by both Dc protein assay (Bio-Rad Laboratories, Hercules, CA) and measurement of absorbance at 280 nm using a ND-1000 spectrophotometer (Nanodrop Technologies, Wilmington, DE), The extinction coefficient for AGRP was calculated using the equation of Gill and Von Hippel (40).

Dual in situ hybridization (ISH)
Male Sprague Dawley rats (Charles River Laboratories, Boston, MA) weighing 250–300 g were used. Coronal sections (15 µm) were cut through the entire rostrocaudal axis of the rat brain. Sections were thaw mounted onto slides, quickly dried, and stored at –80 C. Double-ISH studies were performed using 33P- plus digoxigenin-labeled ribonucleotide probes (riboprobes). To generate riboprobes, PCR-amplified cDNAs encoding rat PC1/3 (accession no. NM-017091, nucleotides 2072–2381), rat PC2 (accession no. NM-012746, nucleotides 1133–1463), and rat neuropeptide Y (accession no. NM-012614, nucleotides 76–426) were ligated into pGem-T (Promega, Madison, WI) using standard protocols. Linearized plasmids were transcribed with either T7 or SP6 according to manufacturer’s instructions. Reactions were terminated by digestion of the plasmid template and riboprobes were extracted (33P-labeled riboprobes only), precipitated, and resuspended in 50 µl 50% nuclease-free formamide+ 1 µl RNasin (Promega) and stored at 20 C. Riboprobes were heated to 65 C for 5 min and quenched on ice before addition to the hybridization buffer.

Before hybridization, slides were quickly brought to room temperature and sections were fixed for 15 min in cold 4% paraformaldehyde in 0.1 M PBS (pH 7.4). Slides were briefly rinsed in PBS (PB+ 0.9% NaCl), acetylated for 10 min in 0.25% acetic anhydride/0.1 M triethanolamine/0.9% NaCl, and then rinsed 3 x 2 min in PBS. Sections were taken through an increasing ethanol series, followed by 5 min in chloroform. Air-dried sections were incubated with antisense riboprobes (5 x 105 dpm of 33P-labeled riboprobe/slide plus 30 ng digoxigenin-labeled riboprobe/slide) in hybridization buffer [50% deionized formamide, 4x sodium saline citrate (SSC) (pH 7.0), 1 mM EDTA, 20 µg/ml yeast tRNA, 10% dextran sulfate, 1x Denhardt’s solution, and 0.25% sodium dodecyl sulfate], and incubated overnight at 65 C in a moist chamber. The following day, slides were washed at room temperature for 10 min in 2x SSC, followed by 2 x 30-min washes at 60 C, RNase treated [20 µg/ml in TEN buffer: 500 mM NaCl, 10 mM Tris (pH 7), and 1 mM EDTA] at 37 C for 30 min and then sequentially washed for 30 min at 60 C in 2x SSC/50% formamide, followed by 0.5x SSC.

For the detection of the digoxigenin-labeled riboprobe signal, slides (after high stringency wash) were washed in buffer 1 [100 mM Tris (pH 7.5), 150 mM NaCl] for 2 x 10 min and then blocked for 30 min in buffer 1 + 0.1% Triton X-100 + 2% heat-inactivated fetal bovine serum (Roche). Antidigoxigenin-alkaline phosphatase conjugated antibody (Invitrogen) was diluted 1:500 in buffer 1 + 0.1% Triton X-100 + 1% fetal bovine serum, and slides were incubated for 1 h in antibody solution at room temperature. Slides were then washed in buffer 1, and incubated for 10 min in buffer 2 [0.1 M Tris (pH 9.5), 0.1 M NaCl, 50 mM MgCl2] before color detection. Digoxigenin-labeled probes were visualized by incubating the slides in a chromogen solution containing nitroblue tetrazolium chloride and 5-bromo-4-chloro-3-indoyl-phosphate in buffer 2. The color reaction proceeded at room temperature, and once stopped, slides were extensively washed (>3 h) in 10 mM Tris (pH 8.0), 1 mM EDTA, and 150 mM NaCl. Sections were briefly dehydrated in 70% ethanol and air dried. Slides were then dipped in K5 nuclear emulsion (Ilford, Knutsford, UK) for autoradiography.

HPLC/RIA analysis of mouse hypothalamic extracts
Hypothalami from four PC1/3 null and four wild-type mice (41) were dissected using consistent landmarks and were individually homogenized in 0.5 ml of 0.1 N HCl, centrifuged at 16,000 x g, and the supernatant from each was analyzed by HPLC as previously described (26). One-milliliter fractions were collected, evaporated in a Speed Vac concentrator, and dissolved in buffer for AGRP RIA. The column was calibrated with 5 ng AGRP83–132 (Phoenix Peptides Inc., Belmont, CA) and with 5 ng of full-length AGRP (provided by Dr. G. Barsh, Stanford University School of Medicine, Stanford, CA). AGRP was measured by RIA as previously described (26) with an antiserum raised against human AGRP and directed at the C-terminal end of the molecule, provided by Dr. G. Barsh (25). AGRP83–132 (Phoenix Peptides) was used for the standard and tracer. Assay sensitivity is 2.5 pg with 50% displacement of tracer at 50 pg.

cAMP reporter assays
cAMP assays were undertaken as previously described (42). Briefly, CHOK1 cells were stably transfected with full-length human MC4-R and a cAMP reporter construct consisting of a cAMP response element and three vasoactive intestinal peptide enhancer elements upstream of a lac Z reporter gene (kindly provided by Drs. M. Needham and D. Scanlan, AstraZeneca, Cheshire, UK). Cells were grown to complete confluence in DMEM (Sigma), 10% fetal calf serum, 1% HT supplement (Invitrogen), 1% nonessential amino acids (Invitrogen), 200 µg/ml G418 (Invitrogen), and 500 µg/ml hygromycin B (Roche). Cells were washed in PBS and harvested. Ligand stocks (2.5 times) were prepared in indicator free DMEM and 40-µl aliquots were added, in quadruplicate, to poly-lysine coated 96-well plates. Stocks (10 times) of either full-length AGRP or AGRP83–132 (Phoenix Peptides) were added to appropriate wells in 10-µl aliquots. CHOK1 cells expressing the human MC4-R were added each well at a density of 50,000 cells/well, and the plate was incubated for 5 h at 37 C/5% CO2. cAMP was detected by addition of 1 mM chlorophenol red-ß-D-galactopyranoside (Roche) in buffer containing a final concentration of 40 mM Na2HPO4, 40 mM NaH2PO4, 7 mM KCl, and 0.7 mM MgSO4. ß-Galactosidase converts chlorophenol red-ß-D-galactopyranoside to give a red color. Results were quantified by reading absorbance at 590 nm on a Spectrafluor (Tecan, Männedorf, Switzerland) plate reader. Each experiment was performed a minimum of three times with quadruplicate wells. Dose-response data were fitted to a sigmoid curve using nonlinear squares regression (Origin 6.0, Microcal Software, Inc., Northampton, MA). Data from dose-response curves were transformed according to the method of Arunlakshana and Schild (43) to determine the affinity of the antagonist (pKb).

In vivo analysis of AGRP peptides
All experiments were performed using adult male Sprague Dawley rats (250–300 g, Charles River Laboratories, Sandwich, UK). Animals were kept in a 12-h light, 12-h dark cycle at 21 ± 1 C with 45 ± 10% humidity and free access to food (Beekay International, Hull, UK) and water. All experiments were performed in accordance with the United Kingdom’s Animals (Scientific Procedures) Act (1986). Animals underwent lateral cerebroventricular cannulation (0.8 mm posterior and 1.5 mm lateral to bregma and 3.0 mm down from dura) under halothane anesthesia. After 1 wk of recovery, animals were housed individually and left to acclimatize. The free-moving, conscious rats were given intracerebroventricular injections of 2 nmol AGRP83–132 (Phoenix Peptides) and rat equivalent sequences (XM_574228) for human AGRP25–51 (rat AGRP25–47 VAPLKGIRRSDQALFPEFSGLSL) and human AGRP54–82 (rat AGRP50–80 TAADRAEDVLLQKAEALAEVLDPQNRESRSP) (peptides custom synthesized by Bachem, Bubendorf, Switzerland) or vehicle (isotonic sterile saline) in a volume of 2 µl. Immediately after injections, a preweighed amount of food was presented to the animals. Food consumption was measured after 1, 2, 4, 8, 24, and 48 h. A temperature-sensitive, precalibrated radiotelemetery transmitter (TA10TA-F40, Data Sciences International, Minneapolis, MN) was implanted into the peritoneal cavity at the same time as the ventricular cannulation. The core body temperature of the animals was measured continuously throughout the 48 h experimental period. Correct cannulae placement was verified after the experiment by a positive dipsogenic response to a 2-µl icv injection of 100 ng human angiotensin (Sigma-Aldrich). Only animals that responded were included in the subsequent analysis.

Statistical analysis
Quantitative measurements of AGRP peptides in hypothalamic extracts were compared using a nonparametric Mann-Whitney U test. In the food-intake experiments, food intake and body weight gain were analyzed using a parametric one-way ANOVA. Core body temperature was expressed as change from mean basal values and was analyzed by calculating area under the curve (C/h) by the trapezoid method. In all tests, P < 0.05 was considered significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
AGRP is posttranslationally cleaved in the regulated secretory pathway
In the absence of a suitable hypothalamic cell line that endogenously expresses AGRP, we analyzed AGRP processing by transfecting three well-characterized murine neuroendocrine cell lines, {alpha}TC1–6, ßTC-3, and AtT20, with pcDNA3 (Invitrogen) encoding human full-length AGRP. These cells were chosen because they represent useful model systems for the regulated secretory pathway (44), and they endogenously express PCs that are likely to cleave AGRP in the hypothalamus. {alpha}TC1–6 and ßTC-3 were derived from mouse pancreatic islets and endogenously express PC2 and both PC1/3 and PC2, respectively. AtT20 cells were derived from mouse anterior pituitary corticotrophs and endogenously express PC1.

In ßTC3 cells transfected with human AGRP (Fig. 1AGo), Western blot analysis of the cell lysate demonstrates that full-length AGRP (12 kDa) is stored intracellularly and undergoes posttranslational cleavage to generate a carboxyl-terminal product of approximately 6 kDa. Stimulation of the cells with KCl greatly enhanced secretion of both full-length and carboxyl-terminal AGRP into the media, indicating that AGRP is stored in secretory granules. Figure 1BGo shows immunocytochemical evidence that AGRP colocalizes with POMC-derived peptides in transfected AtT20 cells. Analysis of cells using electron microscopy (Fig. 1BGo, lower right panel) demonstrates that AGRP is located in or near large dense-core vesicles. A similar labeling pattern was observed for ACTH (data not shown). These observations represent the first direct evidence that AGRP is sorted into the regulatory secretory pathway. Furthermore, it should be noted that the vast majority of AGRP, both intracellular and secreted, is in the processed 6-kDa form.


Figure 1
View larger version (45K):
[in this window]
[in a new window]
  		FIG. 1. A, pcDNA3. AGRP transfected into ßTC3 cells is predominantly stored intracellularly and is cleaved to generate a carboxyl-terminal 6-kDa fragment. A truncated soluble form of furin (32 ) is used as control for constitutive secretion and is more predominant in media than cell lysates. B, Coimmunofluorescence studies in AtT20 cells transfected with pcDNA3. AGRP indicate that AGRP- (red) and POMC-derived peptides (green) colocalize. Immunoelectron microscopy (lower right panel) shows silver-enhanced gold particles labeling AGRP predominantly in large dense-core vesicles.

 
AGRP is posttranslationally cleaved after Arg79-X-X-Arg82
Processing of AGRP is likely to involve PCs, a family of serine proteases that cleave prohormones at basic motifs, usually R/K-R or R-X-X-R (45). Family members include furin, PC1 (PC3), PC2, PC4, PC5 (PC6), PACE4, and LPC (PC7, PC8), and two more distantly related members, SKI-1 and NARC-1 (46, 47). Analysis of the protein sequence of human AGRP demonstrates that there are three potential prohormone cleavage sites that could potentially liberate a carboxyl-terminal peptide: Arg79-X-X-Arg82, Arg85-Arg86, and Arg86-X-X-Arg89. In an attempt to identify which cleavage sites are processed in vitro, we generated seven expression constructs in which potential cleavage sites have been disrupted by site-directed mutagenesis (Fig. 2AGo). These were transfected into ßTC3 cells, and precipitated media samples were analyzed by Western blot (Fig. 2BGo). In constructs in which the Arg79-X-X-Arg82 site was mutated, processing is blocked (constructs 1 and 3). Mutation of the Arg85-Arg86 and Arg86-X-X-Arg89 (constructs 2 and 4) did not affect cleavage. These results strongly suggest that AGRP undergoes a posttranslational cleavage event in vitro after Arg79-X-X-Arg82.


Figure 2
View larger version (36K):
[in this window]
[in a new window]
  		FIG. 2. A, Full-length AGRP expression constructs. In each case wild-type (W) putative cleavage sites were disrupted by site-directed mutagenesis, changing arginines to alanines (M). The position of inserted FLAG epitopes is indicated by black circles. B, Western blot analysis of media samples from ßTC3 cells transfected with each construct. C, Western blot analysis of media samples from ßTC3 cells transfected with FLAG tagged AGRP constructs. M2 antibody detects the FLAG epitope regardless of its position in the protein, whereas M1 detects only FLAG epitopes with a free amino-terminus, indicating that the predominant secreted peptide is AGRP83–132. The position of putative cleavage sites Arg79-X-X-Arg82 (REPR), Arg85-Arg86 (RR), and Arg86-X-X-Arg89 (RCVR) are indicated.

 
The main secreted peptide is AGRP83–132
In parallel to the above experiment, a series of FLAG-tagged constructs were generated to precisely define the secreted carboxyl-terminal AGRP peptide. Constructs were engineered so that a FLAG epitope was inserted between the P1 and P1' residues of the putative cleavage sites Arg79-X-X-Arg82 and Arg85-Arg86 (Fig. 2AGo). A cleavage event would therefore expose a FLAG epitope at the amino-terminus of the cleaved peptide. This could be detected by using an anti-FLAG M1 antibody, which detects only the FLAG-tag with a free amino terminus, whereas the M2 antibody detects the FLAG epitope independent of its position in the protein. Figure 2CGo shows Western blot analysis of media from ßTC3 cells transfected with the FLAG-tagged constructs. As expected, the M2 anti-FLAG antibody detects both a full-length band and a carboxyl-terminal cleaved fragment with all three constructs. The cleaved fragment is strongly detected by M1 in cells transfected with construct 6. This provides further evidence that AGRP is cleaved after Arg79-X-X-Arg82 and AGRP83–132 are secreted. A faint band was also detected in cells transfected with construct 5, suggesting some cleavage at Arg85-Arg86.

AGRP is predominantly cleaved by PC1
To assess which propeptide convertases cleave AGRP, we transfected {alpha}TC1–6, AtT20, and ßTC3 cells with pcDNA3 (Invitrogen) encoding full-length human AGRP. These cells endogenously express PC2, PC1/3, and PC1/3/PC2, respectively. Figure 3AGo demonstrates that cleavage occurs in all three of these cell lines, suggesting both PC1 and PC2 can cleave the Arg79-X-X-Arg82. Vectors encoding shRNA interference (shRNAi) fragments targeting proprotein convertases were then transfected into ßTC3 cells. Cotransfection experiments with vectors encoding targeted PCs were undertaken to assess efficacy of shRNAi silencing (Fig. 3BGo). In all cases complete or near complete suppression was achieved. We found that silencing of PC1/3 in ßTC3 cells resulted in a partial inhibition of AGRP processing, indicating that this enzyme is important in cleavage of AGRP (Fig. 3CGo). Silencing of other PCs in the presence of PC1/3 had no effect on AGRP processing, indicating that PC1/3 alone is sufficient. Figure 3DGo shows silencing of PC1/3 in AtT20 cells, which almost completely blocked processing, thereby defining a key role for PC1/3 in AGRP cleavage. To assess which other PCs can cleave AGRP, PC1/3 shRNAi-transfected AtT20 cells were cotransfected with furin, PACE4, PC5/6A, PC5/6B, and PC2/7B2. This rescue experiment demonstrated that both PC5/6A and PC2 have significant capacity to cleave AGRP in the absence of PC1/3. A potential role for PC2 is further indicated by the observation that AGRP is partially cleaved when transfected into {alpha}TC1–6 cells.


Figure 3
View larger version (39K):
[in this window]
[in a new window]
  		FIG. 3. A, pcDNA3. AGRP transfected into AtT20, {alpha}TC1–6, and ßTC3 cells, indicating that AGRP is cleaved in each cell line. B, To test shRNAi efficacy, ßTC3 cells were cotransfected with 0.2 µg expression vectors encoding target PC mRNA and either 0.8 mg mU6pro vector encoding shRNAi (+) or empty vector (–). In each case, complete or near complete suppression of recombinant target was achieved as assessed by Western blot. C, ßTC3 cells cotransfected with vectors encoding shRNAi particles targeted to PCs. AGRP is cleaved to near completion in cells transfected with shRNAi for furin, PC2, PACE4, and PC5/6A to generate AGRP83–132. PC1/3 shRNAi inhibits processing of AGRP by approximately 50%. D, AtT20 cells transfected with shRNAi particles targeted to PC1/3 blocks processing of full-length AGRP by approximately 80%. Cotransfection with PC2/7B2 and PC5/6A almost completely rescues cleavage of AGRP.

 
PC1/3 cleaves AGRP in vivo
To determine whether PC1/3 and PC2 colocalize with AGRP in the hypothalamus in vivo, we undertook dual-ISH experiments. Figure 4Go shows representative dark-field autoradiograms of coronal sections of the rat forebrain. PC1/3 generally exhibits a more restricted expression profile than PC2 but is particularly strongly expressed in the paraventricular nucleus and the supraoptic nucleus. PC2 is strongly expressed in the hippocampus and the thalamus. The consistency of these data with previous studies demonstrates the specificity of the riboprobes that we have designed (48, 49). High-power bright-field photomicrographs focusing on the arcuate nucleus demonstrate that both PC1/3 and PC2 (silver grains) are coexpressed in AGRP neurons (dark staining). We found that almost all AGRP neurons express both PC1/3 and PC2, thereby implicating a physiological role for these enzymes in AGRP posttranslational processing.


Figure 4
View larger version (108K):
[in this window]
[in a new window]
  		FIG. 4. Dark-field autoradiograms of coronal sections of rat forebrain incubated with riboprobes for PC1/3 (i, ii) and PC2 (iv, v). PVN, Paraventricular nucleus; TH, thalamus; VMH, ventromedial hypothalamus; HC, hippocampus; SON, supraoptic nucleus. Bright-field photomicrographs focusing on the arcuate nucleus demonstrate that PC1/3 (silver grains) (iii) and PC2 (silver grains) (iv) colocalize with AGRP (dark staining).

 
Based on experiments in vitro, PC1 appeared to be predominantly responsible for AGRP cleavage. To investigate the role of PC1 in vivo, we analyzed AGRP processing in hypothalamic lysates from PC1-null mice. Figure 5AGo shows a representative chromatograph of an individual wild-type and PC1-null mouse. In wild-type mice, the majority of AGRP immunoreactivity coeluted with AGRP83–132, indicating that processing occurs in vivo. However, in PC1-null mice, there was an increase in full-length AGRP immunoreactivity. Figure 5BGo shows quantitative analysis of full-length AGRP and AGRP83–132 in four wild-type and four PC1-null mice. There was no significant difference in AGRP83–132 levels between the two groups. However, there was significantly more full-length AGRP in the null mice vs. the wild-type mice (648 ± 237 vs. 198 ± 31 pg/hypothalamus; P = 0.04). The mean percentage full-length AGRP rose from 2.7% in wild-type mice to 9.5% in null mice (P = 0.02), indicating that PC1 processes AGRP in vivo, although other proteases, presumably PC2 and/or PC5/6a, compensate for PC1 in its absence.


Figure 5
View larger version (23K):
[in this window]
[in a new window]
  		FIG. 5. A, A representative chromatograph demonstrating the characterization of AGRP immunoreactivity in wild-type and PC1/3-null mice hypothalami by HPLC. Arrows indicate elution positions of synthetic AGRP83–132 and full-length AGRP. B, Quantification of AGRP peptides in four wild-type (hatched bars) and four PC1/3 null (black bars) hypothalami. *, P < 0.05.

 
AGRP83–132 is more potent than full-length AGRP at the MC4R
We undertook detailed pharmacological analysis to compare the properties of AGRP83–132, the predominant secreted AGRP peptide, with full-length AGRP. Recombinant full-length AGRP was generated and purified from E. coli AGRP was refolded as previously described (17). Figure 6AGo shows the recombinant protein as analyzed by size-exclusion chromatography. AGRP eluted as a single peak at a retention time consistent with a monomeric state, suggesting that the recombinant AGRP is correctly folded. To test the relative potencies of full-length AGRP and AGRP83–132 at the MC4-R, cAMP reporter assays were undertaken in CHOK1 cells stably expressed with human MC4-R and a ß-galactosidase reporter gene under the control of cAMP response elements. Data were subjected to Schild analysis to determine pKb values (Fig. 6Go, B and C). AGRP 83–132 [pKb = 8.75 ± 0.05 (1.8 nM)] was significantly more potent as an inhibitor than full-length AGRP [pKb = 7.95 ± 0.04 (11 nM)]. Interestingly, at low concentrations (e.g. 1 nM) full-length AGRP is ineffective, whereas AGRP83–132 significantly antagonizes {alpha}MSH. Given that concentrations of secreted AGRP in vivo are likely to be low (50), the observed modest differences in pKb values between full-length and truncated AGRP are likely to be physiologically significant.


Figure 6
View larger version (20K):
[in this window]
[in a new window]
  		FIG. 6. A, Recombinant full-length AGRP was generated in E. coli, purified, and refolded as described (17 39 ) and subjected to size-exclusion chromatography, which indicates that AGRP is correctly folded in a monomeric state. B, CHOK1 cells stably expressing hMC4-R and a ß-galactosidase reporter construct were treated with increasing concentrations of {alpha}MSH, and coincubated with 0 nM ({circ}), 1 nM (bullet), 5 nM ({square}), 10 nM ({blacksquare}), 50 nM ({triangleup}), and 100 nM ({blacktriangleup}) of either full-length AGRP or AGRP83–132. Data points represent means of quadruplicate measurements. Data shown are one representative of three independent experiments. For full-length AGRP the curves for 0 nM ({circ}) and 1 nM (bullet) are superimposed.

 
Rat AGRP25–47 and rat AGRP50–80 do not have a role in energy homeostasis
The observation that AGRP is cleaved raises questions concerning the role of the amino-terminal portion of the molecule. In particular, we postulated that interaction of amino-terminal peptides with syndecan-3 could affect food intake independently of the melanocortin system. We synthesized rat peptides corresponding to the commercially available human peptides AGRP25–51 and AGRP54–82 and determined cumulative food intake in groups of rats injected with a single bolus of 2 nmol AGRP83–132, AGRP25–47, and AGRP50–80 (Fig. 7AGo). As expected, AGRP83–132 had a potent effect on food intake that was apparent over 48 h, whereas the amino-terminal peptides did not stimulate food intake. AGRP83–132 also significantly increased body weight, compared with the vehicle-treated cohort (Fig. 7BGo), and decreased core body temperature (Fig. 7CGo). This was attributable to a decrease in expression of uncoupling protein-1 in brown adipose tissue (data not shown). Neither AGRP25–47 nor AGRP50–80 had a significant effect on body weight or body temperature.


Figure 7
View larger version (21K):
[in this window]
[in a new window]
  		FIG. 7. A, Mean food intake measurements following intracerebroventricular (icv) administration of a single 2-nmol bolus of vehicle (n = 6) ({blacksquare}), AGRP83–132 (n = 5) (bullet), AGRP25–47 (n = 5) ({diamondsuit}), and AGRP50–80 (n = 6) ({blacktriangleup}). B, Mean body weight changes in rats following icv injection of vehicle (open bars), AGRP83–132 (diagonal hatched bars), AGRP25–47 (horizontal hatched bars), and AGRP50–80 (black bars). C, Core body temperatures were measured over a 48-h period in rats injected with vehicle (black lines), AGRP83–132 (light gray lines), AGRP25–47 (dark gray lines), and AGRP50–80 (hatched lines). *, P < 0.05; **, P < 0.01; ***, P < 0.0001.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this study we addressed several important questions regarding the posttranslational processing and trafficking of AGRP. First, we provide direct evidence that AGRP is stored intracellularly in secretory granules and is secreted via the regulated pathway. This is consistent with it acting as an important regulatory neuropeptide. Indeed, analysis of AGRP content in hypothalamic extracts (7) and secretion experiments using perifused hypothalamic slices (25, 26) indicate that altered secretion of AGRP is more important than secretion of POMC-derived peptides in eliciting acute changes in melanocortinergic tone. Therefore, it is important to understand how AGRP release from secretory granules is regulated and which AGRP derived peptides are produced.

We clearly demonstrate that AGRP is posttranslationally cleaved to produce a carboxyl-terminal fragment. These data are consistent with previous HPLC analyses of AGRP immunoreactivity in rat serum and hypothalamic tissue (25, 26). However, the published studies did not precisely define the primary form of secreted AGRP. By undertaking a series of site-directed mutagenesis experiments, we have shown that cleavage occurs after the Arg79-Glu80-Pro81-Arg82 site to generate AGRP83–132. This observation is important because most physiological studies of AGRP function have used AGRP83–132 as it is the main commercially available form (Phoenix Peptides). This was synthesized based on analogy to the processing pattern of atrial-natriuretic factor, but there was no direct evidence that this peptide was produced in vivo (23). Our data support the concept that it is more important to consider the effects of AGRP83–132 in physiological studies rather than full-length AGRP (12, 13, 14).

By undertaking RNA interference and overexpression experiments in a series of neuroendocrine cell lines, we have shown that AGRP cleavage is predominantly catalyzed by PC1/3, although PC2 and PC5/6A have the capacity to cleave AGRP in the absence of PC1/3 in vitro. However, in contrast to PC1/3, RNA interference silencing of PC2 and PC5/6A did not inhibit AGRP processing in ßTC3 cells, indicating that they are not primarily important in the processing of AGRP. Combined HPLC and RIA analysis of PC1-null mice hypothalami indicate that PC1/3 cleaves AGRP in vivo because there is a significant accumulation of unprocessed full-length AGRP in null vs. wild-type hypothalami. However, it cannot be entirely excluded that processing by other PCs at the same or another cleavage site occurs in hypothalamic neurons in vivo. It is not surprising that genetic ablation of PC1/3 results in only a partial reduction of processing. Similar observations have been made for other neuropeptides in the PC1/3 null mice (51) and other PCs (52). It has become clear that for many, but not all substrates, a limited redundancy of PCs exists. Here it seems likely that PC2 and possibly other PCs compensate for the absence of PC1.

The observation of posttranslational cleavage of AGRP has important implications regarding the mechanism by which it elicits its physiological effects. It has been demonstrated previously that full-length AGRP, but not carboxyl-terminal AGRP, binds to syndecan-1. Based on this observation, it was proposed that syndecan-3, which unlike syndecan-1 is endogenously expressed in the hypothalamus, acts as a coreceptor for MC4-R (28). Experiments undertaken in syndecan-3-null mice (28, 29, 30, 53) and syndecan-1 transgenic mice (28) clearly indicate that syndecan-3 does indeed play an important role in energy homeostasis. However, our data show that AGRP is cleaved into distinct amino-terminal and carboxyl-terminal peptides before secretion. Therefore, it is difficult to envisage how syndecan-3 binding to an amino-terminal fragment could influence the effect of the carboxyl-terminal fragment. It is theoretically possible that despite intracellular cleavage, AGRP fragments remain associated and form a complex with syndecan-3, and MC4-R held together by disulfide bridges or noncovalent associations. However, we think this possibility is highly unlikely. First, covalent association of AGRP fragments through disulfide bridges can be ruled out because no cysteines are present in amino-terminal AGRP. Second, noncovalent association is unlikely because immunoprecipitation of processed AGRP under nondenaturing conditions did not result in coimmunoprecipitation of the propeptide (data not shown). These observations indicate that syndecan-3 cannot act as a coreceptor for the MC4-R. Consequently, syndecan-3 and the carboxyl terminus of AGRP must act independently in the regulation of food intake. This would explain the observation that syndecan-3-null mice are resistant to diet-induced obesity (29), whereas AGRP-null mice are not (54). Nevertheless, recent data do support the idea that syndecan-3 facilitates the actions of endogenous MC4-R antagonists because the obese phenotype observed in agouti lethal yellow mice is attenuated on a syndecan-3-null background (55). This phenomenon must, presumably, be a result of an indirect mechanism, possibly related to the role of syndecan-3 in central nervous system plasticity (53).

It is possible that amino-terminal AGRP fragments have a role in energy homeostasis independent of the MC4-R, and such a role may be mediated by syndecan-3. This possibility is supported by the observation that syndecan-3 is highly expressed in regions of the hypothalamus that receive dense innervation from AGRP neurons, such as the paraventricular nucleus (19, 28). A recent study implicated amino-terminal AGRP in energy homeostasis. Goto et al. (15) administered two commercially available human carboxyl-terminal AGRP peptides, AGRP25–51 and AGRP54–82, into rat brains via intracerebroventricular cannulae and found that both peptides increased body weight. However, these data are difficult to interpret because amino-terminal AGRP, unlike carboxyl-terminal AGRP, is not particularly well conserved between humans and rats and shows only 66% similarity. In our study we synthesized the equivalent rat peptides of human AGRP25–51 and AGRP54–82 and injected them into rat brains. We found that these peptides, in contrast to AGRP83–132, did not affect body weight, food intake, or core body temperature. Therefore, our results do not support a role for amino-terminal AGRP in the regulation of body weight. However, owing to a lack of amino-terminal AGRP antibodies, it has not been possible to ascertain which amino-terminal AGRP peptides are produced. The commercially available amino-terminal peptides have presumably been synthesized on the assumption that the Lys52-Lys53 site in human AGRP (Lys48-Lys49 in rat) is posttranslationally cleaved, although there is no direct evidence to support this. Further research is therefore required to determine which amino-terminal AGRP peptides are produced in vivo and what, if any, functional effect they have. Moreover, it has not been ascertained in this study whether amino-terminal fragments of AGRP can actually bind to syndecan-3. This will require further analysis to determine whether interaction between N-terminal AGRP and syndecan-3 has any physiological significance.

Both in vitro and in vivo studies have demonstrated that full-length AGRP displays some bioactivity (4, 17, 18, 39). In considering the implications of AGRP processing, we predicted that full-length pro-AGRP would be less potent as an antagonist than AGRP83–132. In this study we directly compared the pharmacological properties of recombinant full-length AGRP and AGRP83–132 in a cAMP reporter assay using CHO cells stably transfected with MC4-R. Based on Schild analysis, we demonstrated that full-length AGRP is 6.1-fold less potent than AGRP83–132. This finding is supported by another recent study that analyzed full-length human AGRP in a reporter gene assay (56). The differences between full-length AGRP and AGRP83–132 could translate into subtle differences in efficacy in vivo. Indeed, we previously demonstrated that subtle changes in POMC-derived peptide potency at the MC4-R can lead to profound obesity in vivo (42).

This study is the first to directly address posttranslational processing and trafficking of AGRP. We have found that AGRP is stored in secretory granules and is cleaved to generate AGRP83–132. Because amino-terminal and carboxyl-terminal AGRP are cleaved from one another before secretion, this study strongly suggests that syndecan-3 does not act as a coreceptor for the MC4-R. Further research is therefore required to understand the physiological role of syndecan-3. It would be interesting to study how AGRP processing is regulated in the hypothalamus. Our previous studies indicate that posttranslational processing of POMC is regulated in the hypothalamus with respect to energy balance (8). Other studies have shown that hypothalamic expression and activity of the PCs, PC1/3 and PC2, are altered in various rodent models of obesity (8, 57, 58, 59, 60). Extrapolating from these observations, it seems possible that AGRP processing may also be regulated as an additional mechanism of controlling melanocortin tone in the hypothalamus.


   Acknowledgments
 
We thank Mrs. Irene Conwell for technical assistance and Dr. David Smith and Dr. Andrew Turnbull of AstraZeneca for advice and helpful discussions.


   Footnotes
 
This work was supported by the Wellcome Trust, AstraZeneca, and National Institutes of Health Grant DK57561 (to S.L.W.). A.G. is funded by a Biotechnology and Biological Sciences Research Council studentship award.

J.W.M.C., L.E.P., A.G., P.L.R., S.M., S.L.W., X.Z., D.F.S., N.D., C.B.L., and S.M.L. have nothing to declare. D. A., C.A.S., R.A.D., and J.C.B. are employed by AstraZeneca. A.W. has received grant support (2003–2005) from AstraZeneca.

First Published Online December 29, 2005

1 J.W.M.C. and L.E.P. contributed equally to this work. Back

Abbreviations: AGRP, Agouti-related peptide; ISH, in situ hybridization; MCR, melanocortin-4 receptor; PC, proprotein convertase; pKb, affinity of the antagonist; POMC, proopiomelanocortin; sh, short hairpin; shRNAi, shRNA interference; SSC, sodium saline citrate.

Received October 28, 2005.

Accepted for publication December 20, 2005.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Cone RD 1999 The central melanocortin system and energy homeostasis. Trends Endocrinol Metab 10:211–216[CrossRef][Medline]
  2. Pritchard LE, Turnbull AV, White A 2002 Pro-opiomelanocortin processing in the hypothalamus: impact on melanocortin signalling and obesity. J Endocrinol 172:411–421[Abstract]
  3. Shutter JR, Graham M, Kinsey AC, Scully S, Luthy R, Stark KL 1997 Hypothalamic expression of ART, a novel gene related to agouti, is upregulated in obese and diabetic mutant mice. Genes Dev 11:593–602[Abstract/Free Full Text]
  4. Ollmann MM, Wilson BD, Yang Y-K, Kerns JA, Chen Y, Gantz I, Barsh GS 1997 Antagonism of central melanocortin receptors in vitro and in vivo by agouti-related protein. Science 278:135–138[Abstract/Free Full Text]
  5. Cone RD, Cowley MA, Butler AA, Fan W, Marks DL, Low MJ 2001 The arcuate nucleus as a conduit for diverse signals relevant to energy homeostasis. Int J Obesity 25:S63–S67
  6. Mizuno TM, Makimura H, Mobbs CV 2003 The physiological function of the agouti-related peptide gene: the control of weight and metabolic rate. Trends Mol Med 35:425–433
  7. Harrold JA, Williams G, Widdowson PS 1999 Changes in hypothalamic agouti-related protein (AGRP), but not {alpha}-MSH or pro-opiomelanocortin concentrations in dietary-obese and food-restricted rats. Biochem Biophys Res Commun 258:574–577[CrossRef][Medline]
  8. Pritchard LE, Oliver RL, McLoughlin JD, Birtles S, Lawrence CB, Turnbull AV, White A 2003 Pro-opiomelanocortin (POMC) derived peptides in rat cerebrospinal fluid and hypothalamic extracts: evidence that secretion is regulated with respect to energy balance. Endocrinology 144:760–766[Abstract/Free Full Text]
  9. Mizuno TM, Mobbs CV 1999 Hypothalamic agouti-related protein messenger ribonucleic acid is inhibited by leptin and stimulated by fasting. Endocrinology 140:814–817[Abstract/Free Full Text]
  10. Graham M, Shutter JR, Sarmiento U, Sarosi I, Stark KL 1997 Overexpression of Agrt leads to obesity in transgenic mice. Nat Genet 17:273–274[CrossRef][Medline]
  11. Makimura H, Mizuno TM, Mastaitis JW, Agami R, Mobbs CV 2002 Reducing hypothalamic AGRP by RNA interference increases metabolic rate and decreases body weight without influencing food intake. BMC Neurosci 3:18[CrossRef][Medline]
  12. Rossi M, Kim MS, Morgan DGA, Small CJ, Edwards CMB, Sunter D, Abusnana S, Goldstone AP, Russell SH, Stanley SA, Smith DM, Yagaloff K, Ghatei MA, Bloom SR 1998 A C-terminal fragment of agouti-related protein increases feeding and antagonizes the effect of {alpha}-melanocyte stimulating hormone in vivo. Endocrinology 139:4428–4431[Abstract/Free Full Text]
  13. Small CJ, Kim MS, Stanley SA, Mitchell JRD, Murphy K, Morgan DGA, Ghatei MA, Bloom SR 2001 Effects of chronic central nervous system administration of agouti-related protein in pair-fed animals. Diabetes 248–254
  14. Hagan MM, Rushing PA, Pritchard LM, Schwartz MW, Strack AM, Van der Ploeg LHT, Woods SC, Seeley RJ 2000 Long-term orexigenic effects of AgRP-(83–132) involve mechanisms other than melanocortin receptor blockade. Am J Physiol 279:R47–R52
  15. Goto K, Inui A, Takimoto Y, Yuzuriha H, Asakawa A, Kawamura Y, Tsuji H, Takahara Y, Takeyama C, Katsuura G, Kasuga M 2003 Acute intracerebroventricular administration of either carboxyl-terminal or amino-terminal fragments of agouti-related peptide produces a long-term decrease in energy expenditure in rats. Int J Mol Med 12:379–383[Medline]
  16. Pritchard LE, Armstrong D, Davies N, Oliver RL, Schmitz CA, Brennand JC, Wilkinson GF, White A 2004 Agouti related protein (83–132) is a competitive antagonist at the human melanocortin-4 receptor: no evidence for differential interactions with pro-opiomelanocortin derived ligands. J Endocrinol 180:183–191[Abstract]
  17. Rosenfeld RD, Zeni L, Welcher AA, Narhi LO, Hale C, Marasco J, Delaney J, Gleason T, Philo JS, Katta V, Hui J, Baumgartner J, Graham M, Stark KL, Karbon W 1998 Biochemical, biophysical and pharmacological characterization of bacterially expressed human agouti-related protein. Biochemistry 37:16041–16052[CrossRef][Medline]
  18. Yang Y-K, Thompson DA, Dickinson CJ, Wilken J, Barsh GS, Kent SBH, Gantz I 1999 Characterization of agouti-related protein binding to melanocortin receptors. Mol Endocrinol 13:148–155[Abstract/Free Full Text]
  19. Bagnol D, Lu X-Y, Kaelin CB, Day HEW, Ollmann M, Gantz I, Akil H, Barsh GS, Watson SJ 1999 Anatomy of an endogenous antagonist: relationship between agouti-related protein and proopiomelanocortin in brain. J Neurosci 19:RC26 (1–7)
  20. Nijenhuis WAJ, Oosterom J, Adan RAH 2001 AgRP (83–132) acts as an inverse agonist on the human melanocortin-4 receptor. Mol Endocrinol 15:164–171[Abstract/Free Full Text]
  21. Haskell-Luevano C, Monck EK 2001 Agouti-related protein functions as an inverse agonist at a constitutively active brain melanocortin-4 receptor. Regul Peptides 99:1–7[CrossRef][Medline]
  22. Chai B-X, Neubig RR, Millhauser GL, Thompson DA, Jackson PJ, Barsh GS, Dickinson CJ, Li J-Y, Lai Y-M, Gantz I 2003 Inverse agonist activity of agouti and agouti-related protein. Peptides 24:603–609[CrossRef][Medline]
  23. Quillan JM, Sadee W, Wei ET, Jimenez C, Li J, Chang JK 1998 A synthetic human agouti-related protein-(83–132)-NH2 fragment is a potent inhibitor of melanocortin receptor function. FEBS Lett 428:59–62[CrossRef][Medline]
  24. Jackson PJ, McNulty JC, Yang Y-K, Thompson DA, Chai B, Gantz I, Barsh GS, Millhauser GL 2002 Design, pharmacology, and NMR structure of a minimized cystine knot with agouti-related protein activity. Biochemistry 41:7565–7572[CrossRef][Medline]
  25. Li J-Y, Finniss S, Yang Y-K, Zeng Q, Qu S-Y, Barsh G, Dickinson C, Gantz I 2000 Agouti-related protein-like immunoreactivity: characterization of release from hypothalamic tissue and presence in serum. Endocrinology 141:1942–1950[Abstract/Free Full Text]
  26. Breen TL, Conwell IM, Wardlaw SL 2005 Effects of fasting, leptin, and insulin on AGRP and POMC peptide release in the hypothalamus. Brain Res 1032:141–148[CrossRef][Medline]
  27. Bergeron F, Leduc R, Day R 2000 Subtilase-like pro-protein convertases: from molecular specificity to therapeutic applications. J Mol Endocrinol 24:1–22[Abstract]
  28. Reizes O, Lincecum J, Wang Z, Goldberger O, Huang L, Kaksonen M, Ahima R, Hinkes MT, Barsh GS, Rauvala H, Bernfield M 2001 Transgenic expression of syndecan-1 uncovers a physiological control of feeding behaviour by syndecan-3. Cell 106:105–116[CrossRef][Medline]
  29. Strader AD, Reizes O, Woods SC, Benoit SC, Seeley RJ 2004 Mice lacking the syndecan-3 gene are resistant to diet-induced obesity. J Clin Invest 114:1354–1360[CrossRef][Medline]
  30. Reizes O, Benoit SC, Strader AD, Clegg DJ, Akunuru S, Seeley RJ 2003 Syndecan-3 modulates food intake by interacting with the melanocortin/AGRP pathway. Ann NY Acad Sci 994:66–73[Medline]
  31. Jackson RS, Creemers JWM, Farooqi IS, Raffin-Sanson M-L, Varro A, Dockray GJ, Holst JJ, Brubaker PL, Corvol P, Polonsky KS, Ostrega D, Becker KL, Bertagna X, Hutton JC, White A, Dattani MT, Hussain K, Middleton SJ, Nicole TM, Milla PJ, Lindley KL, O’Rahilly S 2003 Small-intestine dysfunction accompanies the complex endocrinopathy of human proprotein convertase 1 deficiency. J Clin Invest 112:1550–1560[CrossRef][Medline]
  32. Creemers JW, Usac EF, Bright NA, Van de Loo JW, Jansen E, Van de Ven WJ, Hutton JC 1996 Identification of a transferable sorting domain for the regulated pathway in the proprotein convertase PC2. J Biol Chem 271:252845–25291
  33. Creemers JW, van de Loo JW, Plets E, Hendershot LM, Van de Ven WJ 2000 Binding of BiP to the processing enzyme lymphoma proprotein convertase prevents aggregation, but slows down maturation. J Biol Chem 275:38842–38847[Abstract/Free Full Text]
  34. van de Loo JW, Teuchert M, Pauli I, Plets E, Van de Ven WJ, Creemers JW 2000 Dynamic palmitoylation of lymphoma proprotein convertase prolongs its half-life, but is not essential for trans-Golgi network localization. Biochem J 352:827–833
  35. Yi H, Leunissen J, Shi G, Gutekunst C, Hersch S 2001 A novel procedure for pre-embedding double immunogold-silver labeling at the ultrastructural level. J Histochem Cytochem 49:279–284[Abstract/Free Full Text]
  36. Jansen S, Stefan C, Creemers JW, Waelkens E, Van Eynde A, Stalmans W, Bollen M, 2005 Proteolytic maturation and activation of autotaxin (NPP2) a secreted metastasis-enhancing lysophospholipase. J Cell Sci 118:3081–3089[Abstract/Free Full Text]
  37. Yu JY, DeRuiter SL, Turner DL 2002 RNA interference by expression of short-interfering RNAs and hairpin RNAs in mammalian cells. Proc Natl Acad Sci USA 99:6047–6052[Abstract/Free Full Text]
  38. D’Anjou F, Bergeron LC, Larbi NB, Fournier I, Salzet M, Perreault J-P, Day R 2004 Silencing of SPC2 expression using an engineered {delta} ribozyme in the mouse ßTC-3 endocrine cell line. J Biol Chem 279:14232–14239[Abstract/Free Full Text]
  39. Ebihara K, Ogawa Y, Katsuura G, Numata Y, Masuzaki H, Satoh N, Tamaki M, Yoshioka T, Hayase M, Matsuoka N, Aizawa-Abe M, Yoshimasa Y, Nakao K 1999 Involvement of agouti-related protein, an endogenous antagonist of hypothalamic melanocortin receptor, in leptin action. Diabetes 48:2028–2033[Abstract]
  40. Gill SC, von Hippel PH 1989 Calculation of protein extinction coefficients from amino acid sequence data. Anal Biochem 182:319–326[CrossRef][Medline]
  41. Zhu X, Zhou A, Dey A, Norrbom C, Carroll R, Zhang C, Laurent V, Lindberg I, Ugleholdt R, Holst JJ, Steiner DF 2002 Disruption of PC1/3 expression in mice causes dwarfism and multiple neuroendocrine peptide processing defects. Proc Natl Acad Sci USA 99:10293–10298[Abstract/Free Full Text]
  42. Challis BG, Pritchard LE, Creemers JWM, Delplanque J, Keogh JM, Luan J, Wareham NJ, Yeo GSH, Bhattacharyya S, Froguel P, White A, Farooqi IS, O’Rahilly SO 2002 A missense mutation disrupting a dibasic prohormone processing site in pro-opiomelanocortin (POMC) increases susceptibility to early-onset obesity through a novel molecular mechanism. Hum Mol Genet 17:1997–2004
  43. Arunlakshana O, Schild HO 1959 Some quantitative uses of drug antagonists. Br J Pharmacol 14:48–58[Medline]
  44. Dey A, Xhu X, Carroll R, Turck CW, Stein J, Steiner DF 2003 Biological processing of the cocaine and amphetamine-regulated transcript precursors by prohormone convertases, PC2 and PC1/3. J Biol Chem 278:15007–15014[Abstract/Free Full Text]
  45. Taylor NA, Van de Ven W, Creemers JWM 2003 Curbing activation: Proprotein convertases in homeostasis and pathology. FASEB J 17:1215–1227[Abstract/Free Full Text]
  46. Seidah NG, Benjannet S, Wickman L, Marcinkiewicz J, Jasmin SB, Stifani S, Basak A, Prat A, Chretien M 2003 The secretory proprotein convertase neural apoptosis-regulated convertase 1 (NARC-1): liver regeneration and neuronal differentiation. Proc Natl Acad Sci USA 100:928–933[Abstract/Free Full Text]
  47. Seidah NG, Prat A 2002 Precursor convertases in the secretory pathway, cytosol and extracellular milieu. Essays Biochem 38:79–94[Medline]
  48. Schafer MKH, Day R 1995 In situ hybridisation techniques to map processing enzymes. Methods Neurosci 23:16–44
  49. Schafer MKH, Day R, Cullinan WE, Chretien M, Seidah NG, Watson SJ 1993 Gene expression of prohormone and proprotein convertases in the rat CNS: a comparative in situ hybridisation analysis. J Neurosci 13:1258–1279[Abstract]
  50. Shen CP, Wu KK, Shearman LP, Camacho R, Tota MR, Fong TM, Van der Ploeg LHT 2002 Plasma agouti-related protein level: a possible correlation with fasted and fed states in humans and rats. J Neuroendocrinol 14:607–610[CrossRef][Medline]
  51. Pan H, Nanno D, Che FY, Zhu X, Salton SR, Steiner DF, Fricker LD, Devi LA, 2005 Neuropeptide processing profile in mice lacking prohormone convertase 1. Biochemistry 44:4939–4948[CrossRef][Medline]
  52. Roebroek AJ, Taylor NA, Louagie E, Pauli I, Smeijers L, Snellinx A, Lauwers A, Van de Ven WJ, Hartmann D, Creemers JW 2004 Limited redundancy of the proprotein convertase furin in mouse liver. J Biol Chem 279:53442–53450[Abstract/Free Full Text]
  53. Kaksonen M, Pavlov I, Voikar V, Lauri SE, Hienola A, Riekki R, Lakso M, Taira T, Rauvala H 2002 Syndecan 3 deficient mice exhibit enhanced LTP and impaired hippocampus dependent memory. Mol Cell Neurosci 21:158–172[CrossRef][Medline]
  54. Qian S, Chen H, Weingarth D, Trumbauer ME, Novi DE, Guan X, Yu G, Shen X, Feng Y, Frazier E, Chen A, Camancho RE, Shearman LP, Gopal-Truter S, MacNeil DJ, Van der Ploeg LHT 2002 Neither agouti-related protein nor neuropeptide Y is critically required for the regulation of homeostasis in mice. Mol Cell Biol 22:5027–5035[Abstract/Free Full Text]
  55. Benoit SC, Clegg DJ, Strader AD, Reizes O, Deletion of the syndecan-3 gene attenuates the hyperphagia and obesity of the agouti lethal yellow (Ay) mouse. Program of the 87th Annual Meeting of The Endocrine Society, San Diego, CA, 2005, p 545 (Abstract P3-11)
  56. de Rijke CE, Jackson PJ, Garner KM, van Rozen RJ, Douglas NR, Kas MJH, Millhauser GL, Adan RAH 2005 Functional analysis of the Ala67Thr polymorphism in agouti related protein associated with anorexia nervosa and leanness. Biochem Pharmacol 70:308–316[CrossRef][Medline]
  57. Jing E, Nillni EA, Sanchez VC, Stuart RC, Good DJ 2004 Deletion of the Nhlh2 transcription factor decreases the levels of the anorexigenic peptides {alpha} melanocyte stimulating hormone and thyrotropin-releasing hormone and implicates prohormone convertases I and II in obesity. Endocrinology 145:1503–1513[Abstract/Free Full Text]
  58. Nilaweera KN, Ellis C, Barrett P, Mercer JG, Morgan PL 2003 Precursor protein convertase 1 gene expression in the mouse hypothalamus: differential regulation by ob gene mutation, energy deficit and administration of leptin, and coexpression with prepro-orexin. Neuroscience 119:713–720[CrossRef][Medline]
  59. Sanchez VC, Goldstein J, Stuart RC, Hovanesian V, Huo L, Munzberg H, Friedman TC, Bjorbaek C, Nillni EA 2004 Regulation of hypothalamic prohormone convertases 1 and 2 and effects on processing of prothyrotropin-releasing hormone. J Clin Invest 114:357–369[CrossRef][Medline]
  60. Berman Y, Mzhavia N, Polonskaia A, Devi LA 2001 Impaired prohormone convertases in Cpefat/Cpefat mice. J Biol Chem 276:1466–1473[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Am. J. Physiol. Regul. Integr. Comp. Physiol.Home page
E. Sanchez, V. C. Rubio, D. Thompson, J. Metz, G. Flik, G. L. Millhauser, and J. M. Cerda-Reverter
Phosphodiesterase inhibitor-dependent inverse agonism of agouti-related protein on melanocortin 4 receptor in sea bass (Dicentrarchus labrax)
Am J Physiol Regulatory Integrative Comp Physiol, May 1, 2009; 296(5): R1293 - R1306.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
J. W. M. Creemers, Y. S. Lee, R. L. Oliver, M. Bahceci, A. Tuzcu, D. Gokalp, J. Keogh, S. Herber, A. White, S. O'Rahilly, et al.
Mutations in the Amino-Terminal Region of Proopiomelanocortin (POMC) in Patients with Early-Onset Obesity Impair POMC Sorting to the Regulated Secretory Pathway
J. Clin. Endocrinol. Metab., November 1, 2008; 93(11): 4494 - 4499.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
E. Louagie, N. A. Taylor, D. Flamez, A. J. M. Roebroek, N. A. Bright, S. Meulemans, R. Quintens, P. L. Herrera, F. Schuit, W. J. M. Van de Ven, et al.
Role of furin in granular acidification in the endocrine pancreas: Identification of the V-ATPase subunit Ac45 as a candidate substrate
PNAS, August 26, 2008; 105(34): 12319 - 12324.
[Abstract] [Full Text] [PDF]

Home page
 DevelopmentHome page
D. Enshell-Seijffers, C. Lindon, and B. A. Morgan
The serine protease Corin is a novel modifier of the agouti pathway
Development, January 15, 2008; 135(2): 217 - 225.
[Abstract] [Full Text] [PDF]

Home page
 EndocrinologyHome page
L. E. Pritchard and A. White
Neuropeptide Processing and Its Impact on Melanocortin Pathways
Endocrinology, September 1, 2007; 148(9): 4201 - 4207.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
A. Breit, K. Wolff, H. Kalwa, H. Jarry, T. Buch, and T. Gudermann
The Natural Inverse Agonist Agouti-related Protein Induces Arrestin-mediated Endocytosis of Melanocortin-3 and -4 Receptors
J. Biol. Chem., December 8, 2006; 281(49): 37447 - 37456.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Creemers, J. W. M.
Right arrow Articles by White, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Creemers, J. W. M.
Right arrow Articles by White, A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals