This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Shinyama, H. Articles by Flier, J. S. Search for Related Content PubMed PubMed Citation Articles by Shinyama, H. Articles by Flier, J. S.

Regulation of Melanocortin-4 Receptor Signaling: Agonist-Mediated Desensitization and Internalization

Division of Endocrinology, Diabetes, and Metabolism, Department of Medicine, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts 02215

Address all correspondence and requests for reprints to: Jeffrey S. Flier, M.D., Beth Israel Deaconess Medical Center, Finard 204, 330 Brookline Avenue, Boston, Massachusetts 02215. E-mail: jflier{at}caregroup.harvard.edu.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Disruption of the hypothalamic melanocortin-4 receptor (MC4R) pathway results in obesity both in humans and rodents, demonstrating a crucial role for hypothalamic MC4Rs in the regulation of energy homeostasis. Because even haploinsufficiency of the MC4R gene can cause obesity in humans and mice, subtle changes in receptor numbers or signaling are likely to impact upon the regulation of food intake and energy expenditure. Little is known about the intracellular regulation of MC4R signaling. Using GT1–7 cells, we show for the first time that the MC4R undergoes ligand-mediated desensitization. We then addressed the possible mechanisms underlying the desensitization using HEK293 and COS-1 cells transfected with hemagglutinin-tagged human MC4R. Preexposure of GT1–7 cells that express endogenous MC4R to the agonist for MC4R, -melanocyte-stimulating hormone, resulted in impaired cAMP formation to a second challenge of -melanocyte-stimulating hormone. The desensitization of MC4R was accompanied by time-dependent internalization of the receptor in HEK293 cells, which was partly inhibited by pretreatment with a specific protein kinase A (PKA) inhibitor, H89. In COS-1 cells, overexpression of dominant-negative G protein-coupled receptor kinase (GRK) 2-K220R partly inhibited the agonist-mediated internalization of MC4R, whereas it did not in HEK293 cells. Overexpression of dominant-negative mutants of ß-arrestin1-V53D and dynamin I-K44A prevented agonist-mediated internalization of MC4R. Mutagenesis studies revealed that Thr312 and Ser329/330 in the C-terminal tail are potential sites for PKA and GRK phosphorylation and may play an essential role in the recruitment of ß-arrestin to the activated receptor. Our data demonstrate that, through PKA-, GRK-, ß-arrestin-, and dynamin-dependent processes, MC4R undergoes internalization in response to agonist, thereby providing novel insights into the regulation of MC4R signaling.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
MELANOCORTIN RECEPTORS BELONG to the superfamily of G protein-coupled seven transmembrane receptors (GPCRs) (1). Each of the five subtypes identified so far couples in a stimulatory fashion to adenylate cyclase. The melanocortin 4 receptor (MC4R) is widely distributed in the brain, particularly in regions of the hypothalamus implicated in appetite and body weight regulation (2, 3). MC4R signaling is modulated by both an endogenous agonist, -melanocyte stimulation hormone (MSH), a peptide cleaved from proopiomelanocortin (POMC) and an endogenous antagonist, agouti-related protein (AGRP) (4). Leptin, an adipocyte-derived hormone, acts on POMC and AGRP neurons in the arcuate nucleus of the hypothalamus, resulting in increased MSH and decreased AGRP formation (5, 6). Several lines of evidence have indicated that activation of MC4R by MSH or synthetic peptide agonists reduce food intake, but suppression of MC4R signaling by AGRP or synthetic antagonists increase food intake and diminish the hypophagic response to leptin (7, 8). Targeted disruption of the MC4R gene in mice causes an obesity-diabetes syndrome characterized by hyperphagia, hyperinsulinemia, and hyperglycemia (9). Noteworthy is that heterozygotes for the null MC4R allele exhibit a phenotype intermediate between that of wild-type and homozygous littermates (9), unlike other monogenic animal models of obesity (10, 11, 12) and obese humans (13, 14, 15, 16). Although heterozygous loss-of-function mutations of MC4R in humans do not always display severe obesity (17, 18), these mutations represent the most common monogenic defect causing human obesity so far reported. Thus, the central melanocortin pathway is extremely important for normal energy homeostasis, and energy homeostasis through this pathway is highly susceptible to quantitative variation in MC4R expression.

GPCR signaling is strictly regulated by multiple mechanisms acting at different levels of signal propagation (19, 20). In many GPCRs, signaling is rapidly attenuated within minutes of agonist exposure, a process termed desensitization (20, 21). This process is associated with phosphorylation of serine/threonine residues in the third internal loop or carboxyl tail of the GPCR by G protein-coupled receptor kinases (GRKs) or second-messenger-dependent kinases (22). Phosphorylation by GRKs allows the GPCRs to interact with cytoplasmic proteins, arrestins, which uncouple the GPCRs from G proteins (23). For many GPCRs, arrestins target the receptor to clathrin-coated vesicles, and the receptor is sequestered into an intracellular vesicular compartment, probably endosomes, after pinching off the vesicles from plasma membrane by dynamin (24). After sequestration, the GPCRs may recycle to the surface membrane after resensitization (25, 26) or undergo lysosomal degradation (27). Certain GPCRs can be targeted selectively to lysosomes leading to down-regulation of the GPCR by the same membrane pathway that mediates internalization (28). Down-regulation of GPCR in various neural cell types is of particular interest because this may lead to certain neurological diseases, such as addiction (29) and opiate tolerance and dependence (30).

Limited information is available to date regarding the changes in MC4R signaling in obesity. It has been reported that diet-induced obesity in rats causes selective changes in the number of MC4R, but not of MC3R, in hypothalamic regions (31). Several mutations within the MC4R of obese patients are associated with decreased binding affinity of agonist to MC4R, resulting in the attenuation of MC4R signals (17, 18, 32, 33). However, it appears that mutations in genes involved in appetite control do not account for obesity in most humans. Taken together, changes in steady-state regulation of MC4R signaling could be involved in the pathogenesis of obesity, as seen with other GPCRs in certain neurological diseases (29) and opiate tolerance and dependence (30). However, little is known about the regulation of MC4R signaling, including termination of the signals, desensitization, and internalization of the receptor, all of which are important for the final status of MC4R signaling. In addition, efforts are underway to develop drugs for the treatment of obesity that act on the MC4R (34). These drugs may have potential problems, however, including rapid disappearance of action by receptor desensitization and internalization as well as tachyphylaxis and down-regulation as a consequence of repeated doses, which are a common and predictable phenomenon for GPCR agonists (35). In this context, we have used mouse hypothalamic GT1–7 cells that express endogenous MC4R to demonstrate for the first time that MC4Rs display agonist-mediated desensitization. Studies using human embryonic kidney (HEK) 293 and COS-1 cells, both of which were transfected with hemagglutinin (HA)-tagged MC4R, have also demonstrated that activation of MC4R by agonist is associated with protein kinase A (PKA) and GRK phosphorylation of serine/threonine residues in the C-terminal tail of MC4R, followed by ß-arrestin and dynamin-dependent internalization of the receptor.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Materials
DMEM, fetal bovine serum, Opti MEM I medium, and Lipofectamine 2000 reagent were obtained from Life Technology Inc. (Grand Island, NY). A pEGF1-N1 vector was from CLONTECH Laboratories, Inc. (Palo Alto, CA). The cDNA-encoding bovine-GRK2 and dominant-negative (DN) mutant GRK2-K220R in pcDNA3 were generously provided by Dr. J. L. Benovic (Thomas Jefferson University, Philadelphia, PA). Rat ß-arrestin-1 and rat dynamin I in pcDNA3 and their DN mutants, ß-arrestin-1-V53D and dynamin I-K44A, were kindly provided by Dr. R. J. Lefkowitz (Duke University, Durham, NC). MSH and 3-isobutyl-1-methylxanthine (IBMX) were from Sigma (St. Louis, MO). Forskolin and H-89 were from Calbiochem (San Diego, CA). AGRP (83–132) was purchased from Phoenix Pharmaceuticals, Inc. (Mountain View, CA). A cAMP enzyme immunoassay kit (RPN225) was obtained from Amersham (Piscataway, NJ). Antibodies for cAMP-responsive element-binding protein (CREB) and phosphorylated CREB were purchased from New England Biolabs, Inc. (Beverly, MA), and those for GRK2 (C-15), ß-arrestin1 (K-16), and dynamin I (C-16) were supplied from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Peroxidase-conjugated, anti-HA monoclonal antibody (3F10) was purchased from Roche Molecular Biochemicals (Québec, Canada). The peroxidase-conjugated second antibodies against goat IgG and rabbit IgG were from Santa Cruz Biotechnology, Inc. Five-week-old male C57BL/6J mice were purchased from The Jackson Laboratories (Bar Harbor, ME).

Construction of MC4R receptor-HA-tagged, green fluorescent protein (GFP) fusion protein and non-GFP-conjugated, HA-tagged MC4R
A human MC4R (hMC4R) cDNA encoded in pcDNA3 (Invitrogen, Carlsbad, CA) was obtained from Dr. A. Hollenberg (Beth Israel Deaconess Medical Center, Boston, MA). The HA epitope (YPYDVPDYA)-tagged-hMC4R was generated by PCR amplification using Pfu-Turbo polymerase (Stratagene, La Jolla, CA) with the sense and antisense primers using hMC4R cDNA as a template. Primers were designed to amplify the MC4R with a HindIII site (bold) at the N terminus, followed by Kozac sequence (italic), the initial methionine, and the HA-epitope (underlined) (5'-CTC AAG CTT CGA ATT CTG GCC ACC ATG TAT CCT TAT GAT GTG CCT GAT TAT GCC GTG AAC TCC ACC CAC CGT GGG-3') and a BamHI site at the C terminus (bold) (5'-CGG TGG ATCCCG GGC ATA TCT GCT AGA CAA GTC ACA-3') removing stop codon (TAA). The PCR product was purified, digested with HindIII and BamHI, purified, and then ligated into HindIII-BamHI-cut pEGFP-N1 to generate a plasmid that contains the HA-hMC4R sequence upstream of, and in frame with, the GFP coding sequence for expression of HA-hMC4R-GFP. Pure plasmid DNA for transfections into mammalian cells was isolated with the plasmid maxi kit (QIAGEN, Valencia, CA). Plasmid coding HA-tagged, non-GFP conjugated MC4R was made by inserting stop codon (TGA) immediately after the last Tyr332 by mutagenesis using QuikChange site-directed mutagenesis kit (Stratagene).

RT-PCR
RT-PCR analysis was carried out to analyze the expression pattern between GT1–7 cells and murine medial hypothalamus (C57BL/6J) of each subtype of the melanocortin receptors (MC1–5R), GRKs (GRK2–6), ß-arrestins (ß-arrestin 1, 2), and dynamin I mRNA. Medial hypothalamic tissue blocks were obtained from the mouse brain by dissection with a razor blade after killing the animal by CO₂ inhalation. Total RNA from GT1–7 cells and hypothalamic tissues were prepared by disruption of cells in RNA STAT-60 reagent (Tel-Test Inc., Friendswood, TX) according to the manufacturer’s instructions. RT-PCR was performed as described previously (36). Briefly, total RNA was treated with ribonuclease-free deoxyribonuclease (DNase) I for 30 min using a commercially available kit (DNA free, Ambion, Inc., Austin, TX) to eliminate contamination of genomic DNA. The RNA sample was reverse transcribed with 200 U Moloney murine leukemia virus reverse transcriptase (Ambion, Inc.) and 100 pmol oligo deoxythymidine primer for 1 h at 37 C and was then suspended in distilled water up to 100 µl.

The following oligonucleotides derived from the mouse MC1, MC2, MC3, MC4, and MC5R were synthesized and used as primers for PCR: MC1R (sense, 5'-GGT GTC CAG TCT CTG CTT CC-3', antisense, 5'-CAT GTG GGC ATA CAG AAT CG-3'); MC2R (sense, 5'-ATG CCC TGC AAT ACC ATA GC-3', antisense, 5'-GCA CCC TTC ATG TTG GTT CT-3'); MC3R (sense, 5'-CAT GTA CTT CTT CCT GTG CAG C-3', antisense, 5'-TGC TCT CGG AGT AGA TGA TGA A-3'); MC4R (sense, 5'-CTT TTA CGC GCT CCA GTA CC-3', antisense, 5'-CCA ATC AGG ATG GTC AAG GT-3'); and MC5R (sense, 5'-AAA TCC GAT GCC AAG AAG TG-3', antisense, 5'-GGG TGA GTG CAG GTT TTT GT-3'). Oligo PCR primers for GRK2–6 were synthesized according to the sequences reported by Li and Wang (37) (GRK2 and GRK3) and Horie and Insel (38) (GRK4, 5, and 6). Olig PCR primers for ß-arrestin1 and -2 and dynamin I were as follows: rat ß-arrestin1 (sense, 5'-CTC AGT ACA AGT GCC CAG TG-3', antisense, 5'-GAT GCA AGG TCT CCC AAC AG-3'); rat ß-arrestin2 (sense, 5'-TGA TGG GCA ACT CAA GCA CG-3', antisense, 5'-GAT GTC GTC GTC TGT GGC AT-3'); and mouse dynamin I (sense, 5'-ATC TGA AGC TGC GTG ATG TG-3', antisense, 5'-CAT CGA GTG CAT GAA GCT GT-3').

The PCR conditions for MC1–5R and GRK2 and -3 consisted of 34 thermal cycles at 60 C annealing temperature, GRK 4–6, and dynamin I consisted of 35 thermal cycles at 55 C annealing temperature. To visualize the PCR products, the samples were subjected to electrophoresis in 1% agarose gel containing ethidium bromide.

Cell culture and transfection
Mouse hypothalamic GT1–7 cells (39) were kindly provided by Dr. R. I. Weiner (University of California, San Francisco, CA). HEK 293 cells were purchased from ATCC (Manassas, VA). GT1–7, HEK 293, and COS-1 cells were grown in DMEM (Life Technologies, Inc.) supplemented with 10% (vol/vol) heat-inactivated fetal bovine serum, penicillin G (100 U/ml), and streptomycin (100 µg/ml) at 37 C in a humidified atmosphere containing 95% air 5% CO₂. When HEK 293 and COS-1 cells were transfected with cDNA, cells were grown to 80–90% confluence and transfected with Lipofectamine 2000 reagent (Life Technologies, Inc.) according to the manufacturer’s instruction. HEK 293 cells on 60-mm plates were transfected with 0.67 µg HA-tagged hMC4R and 4 µg GRK2, GRK2-K220R, ß-arrestin 1, ß-arrestin 1-V53D, dynamin I, dynamin I-K44A, or control vector pcDNA3 using 20 µl Lipofectamine 2000 reagent. Likewise, COS-1 cells were transfected with 0.03 µg HA-tagged hMC4R and 4 µg GRK2, GRK2-K220R, or pcDNA3 using 20 µl Lipofectamine 2000 reagent. Cells were transfected overnight and then split into 24-well plates coated with poly-D-lysine (Becton Dickinson and Co., Franklin Lakes, NJ). Twenty-four hours later, cells were washed once with Dulbecco’s PBS, and the culture medium was changed to serum-free Opti-MEMI (Life Technology Inc.) at least 2 h before the following experiments.

Assay of cAMP
To determine the effect of MSH and forskolin on cAMP formation, cells were incubated with Opti-MEM I containing 0.5 mM IBMX for 10 min and then stimulated with the indicated concentrations of MSH or forskolin for 15 min at 37 C. For the evaluation of MC4R desensitization, cells were stimulated with 100 nM MSH or 10 µM forskolin for the time indicated. After washing twice with serum-free Opti-MEM I, cells were incubated for 10 min with Opti-MEMI containing 0.5 mM IBMX and then stimulated with 100 nM MSH for 15 min at 37 C. To evaluate the effect of AGRP (83–132) on MSH-induced cAMP formation, cells were treated with AGRP (83–132) (0.01–10 nM) either for 30 min or 24 h and then washed three times with serum-free Opti-MEM I, followed by the stimulation with MSH for 15 min at 37 C. The cAMP formation was terminated by aspirating medium and adding 0.5 ml lysis solution 1B from the cAMP assay kit (Amersham). Intracellular cAMP content was measured by enzyme immunoassay using a commercially available assay kit (RPN225, Amersham) according to the manufacturer’s instruction.

Assay of receptor internalization
To rapidly quantify receptor internalization under a variety of conditions, we used an ELISA that measures the level of HA-epitope-tagged cell surface receptors. Thus, internalization is detected as a decrease in cell surface receptor levels, compared with nonagonist-treated controls. Cells were stimulated with indicated concentrations of MSH, forskolin, or AGRP (83–132) for the time indicated. Medium was aspirated, and cells were washed once with PBS and fixed for 10 min at room temperature with 3% paraformaldehyde in PBS (pH 7.4). Cells were washed once with PBS containing 100 mM glycine and then blocked for 30 min at 4 C with PBS containing 5% goat serum (Life Technologies, Inc.) and 1% BSA (fraction V, Sigma). Cells were then incubated for 1 h with a horseradish peroxidase-conjugated monoclonal antibody (3F10) directed against the HA epitope diluted 1:1000. Cells were washed five times in PBS, and antibody binding was visualized by adding 0.3 ml substrate solution for horseradish peroxidase (BM-blue POD solution, Roche Molecular Biochemicals). After incubation for 30 min at room temperature, development of the reaction was terminated by adding 0.3 ml of 2.0 M sulfuric acid. Three hundred microliters of samples were transferred into a 96-well plate, and the plate was read at 450 nm in a microplate reader (Molecular Devices, Sunnyvale, CA).

Fluorescence microscopy
To visualize the sequestration of MC4R in HEK 293, cells grown in 60-mm dishes were transfected with 0.67 µg HA-tagged, GFP-conjugated hMC4R cDNA and 4 µg pcDNA3 overnight, as described previously. Cells were then split into four-well chamber slides (Lab-Tek II, Nalgen Nunc International, Naperville, IL) coated with poly-D-lysine. Twenty-four hours later, medium was changed to Opti-MEM I and cells were treated with MSH for the time indicated at 37 C. Medium was then aspirated, and cells were washed once with PBS and fixed for 10 min at room temperature with 3% paraformaldehyde in PBS (pH 7.4). Cells were examined by microscopy on an Axioskop 2 fluorescence microscope (Carl Zeiss Inc., Thornwood, NY) using a Plan Fluor x100 objective. Images were collected using a QED camera (SPOT, Diagnostic Instruments, Inc., Sterling Heights, MI) software and processed with Adobe Photoshop.

Immunoblot analysis of total and phosphorylated CREB and wild-type and dominant-negative mutants of GRK2, ß-arrestin1, and dynamin I expression
GT1–7 cells grown on 100-mm dishes were serum deprived overnight with Opti-MEM I and treated with 100 nM MSH, in combination with or without IBMX for 5 min. Nuclear extracts were isolated following the procedures of Schreiber et al. (40), and equal amounts were loaded on 10% SDS-PAGE. After electrophoresis the proteins were transferred to nitrocellulose and probed overnight at 4 C with a 1:1000 dilution of anti-CREB or antiphosphorylated CREB antiserum (New England Biolabs, Inc.). Western analysis was performed by enhanced chemiluminescence (Amersham) after incubation with a murine secondary antibody (1:1000, Santa Cruz Biotechnology, Inc.).

HEK 293 or COS-1 cells were transiently transfected with either wild-type or DN mutants of GRK2, ß-arrestin1, or dynamin I. Forty-eight hours after transfection, cells were lysed by the addition of 1 ml RIPA buffer. Lysate was transferred to a microcentrifuge tube and passed through a 25-gauge needle five times, and samples were centrifuged 14,000 rpm for 20 min at 4 C. All samples were diluted in 2x Laemmli sodium dodecyl sulfate-loading buffer and heat denatured at 95 C for 5 min, and 20 µl of samples were size fractionated on 10% sodium dodecyl sulfate-polyacrylamide gels. After SDS-PAGE, protein was blotted onto nitrocellulose and probed overnight at 4 C with a 1:300 dilution of anti-GRK2, anti-ß-arrestin 1, or antidynamin I antiserum (Santa Cruz Biotechnology, Inc.). Western analysis was performed by enhanced chemiluminescence (Amersham) after incubation with a rabbit or goat secondary antibody (1:1000, Santa Cruz Biotechnology, Inc.).

Mutagenesis of HA-hMC4R
The HA-hMC4R-T232A, -S306A, -T312A, -S329A/S330A, and -T312A/S329A/S330A mutants were generated by replacing Thr232, Ser306, Thr312, Ser329/Ser330, and Thr312/Ser329/Ser330 to alanine residues of HA-tagged hMC4R using QuikChange site-directed mutagenesis kit (Stratagene) as described above. The authenticity of the mutant was confirmed by DNA sequencing of both strands (Core Sequencing Facility, Harvard Institute of Medicine).

Data analysis
Data are expressed as mean ± SEM of at least two independent experiments. Data were compared by t test and considered as significant when P < 0.05. Dose-response data were analyzed using Microcal Origin software (Origen Laboratory, Northampton, MA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Gene expression of five subtypes of melanocortin receptors in GT1–7 cells and mouse medial hypothalamus
One-half microgram DNase I-treated RNA obtained from GT1–7 cells and hypothalamic tissues was reverse transcribed to single-stranded DNA by reverse transcriptase, and then DNA was amplified by PCR for 34 cycles using specific oligo primers for MC1 to MC5R. As shown in Fig. 1A, PCR analysis revealed that only MC4R mRNA was detected by PCR amplification in GT1–7 cells, whereas both MC3R and MC4R mRNA were detected in the murine medial hypothalamic tissues, suggesting predominant expression of MC4R in GT1–7 cells.

View larger version (30K): [in this window] [in a new window]   Figure 1. RT-PCR analysis of gene expression of five subtypes of melanocortin receptors in GT1–7 cells and mouse medial hypothalamic tissues (A). MSH-mediated cAMP formation (B) and PKA phosphorylation of CREB (C) in GT1–7 cells. Time course (D) and concentration dependency (E) of MSH-mediated desensitization in GT1–7 cells. A, 0.5 µg DNase I-treated RNA obtained from GT1–7 cells and hypothalamic tissues was reverse transcribed to single-stranded DNA by reverse transcriptase, and then DNA was amplified by PCR using specific oligo primers for MC1-MC5R for 35 cycles, as described in Materials and Methods. B, GT1–7 cells were incubated with Opti-MEM I containing 0.5 mM IBMX. Ten minutes later, cells were stimulated with indicated concentrations of MSH for 15 min. Intracellular cAMP formation was measured by ELISA. C, GT1–7 cells grown on a 10-cm dish were stimulated with 100 nM MSH in the absence or presence of 0.5 mM IBMX for 15 min. Nuclear extract was obtained by the procedures as described in Materials and Methods. Twelve microliters of sample was subjected to SDS-PAGE followed by Western blotting and interaction with anitphosphorylated CREB or anti-CREB antibody. D, E, Cells were stimulated with 100 nM MSH for the time indicated (D) or varying concentrations of MSH for 60 min (E) and then washed with Opti-MEM I twice and incubated with Opti-MEM I containing 0.5 mM IBMX. Ten minutes later, cells were restimulated with the same concentration of MSH. Intracellular cAMP content was measured by ELISA. Data represent means ± SEM of two independent experiments performed in duplicate (B) or triplicate (D, E).

MSH-mediated cAMP formation and PKA phosphorylation of CREB

Desensitization of MC4R in response to MSH in GT1–7 cells
Pretreatment with 100 nM MSH for 10–180 min caused exposure-time-dependent attenuation of cAMP formation by a second treatment with 100 nM MSH (Fig. 1D), suggesting that MC4R undergoes desensitization in response to agonist. The desensitization developed rapidly by 30 min and gradually increased thereafter up to 180 min. After 30-min preexposure to 100 nM MSH, the capacity to produce cAMP was reduced to 30% of nonpretreated control (Fig. 1E). Pretreatment of the cell with varying concentrations of MSH (0.1–1000 nM) for 60 min caused concentration-dependent attenuation of cAMP response to a second challenge of 100 nM MSH (Fig 1E).

To examine whether MSH-mediated desensitization is responsible for an attenuated response in the downstream factors to MSH, we measured cAMP formation produced by forskolin, which stimulates adenylate cyclase directly following 24-h treatment with MSH. Twenty-four-hour pretreatment of GT1–7 cells with MSH at 1–100 nM for 24 h caused markedly attenuated cAMP formation in response to a second challenge of 100 nM MSH (Fig. 2A). In contrast, pretreatment with 10 nM MSH did not influence cAMP formation by forskolin at 50 µM. These results suggest that adenylate cyclase and its downstream signaling were not influenced by the pretreatment with MSH (Fig 2B).

View larger version (18K): [in this window] [in a new window]   Figure 2. Forskolin-mediated cAMP formation is not influenced by the desensitization of MC4R caused by MSH in GT1–7 cells. A, GT1–7 cells were pretreated with MSH (1–100 nM) for 24 h and then stimulated with 100 nM MSH for 15 min. B, GT1–7 cells were pretreated with MSH (10 nM) for 24 h and then stimulated with 50 µM forskolin for 15 min. Intracellular cAMP content was measured by ELISA. Data represent means ± SEM of two independent experiments performed in triplicate.

View larger version (17K): [in this window] [in a new window]   Figure 3. Dose-response curve of cellular cAMP formation by MSH (A) and time course of MSH-mediated desensitization (B) and internalization (C) of HA-hMC4R in HEK 293 cells. A, HA-tagged hMC4R was transiently transfected in HEK 293 cells. Cells were stimulated with indicated concentrations of MSH for 15 min. Intracellular cAMP formation was measured by ELISA. B, HEK 293 cells transfected with HA-hMC4R were exposed to 100 nM MSH for the indicated periods. Cells were stimulated with 100 nM MSH for the time indicated and then washed with Opti-MEM I twice and incubated with Opti-MEM I containing 0.5 mM IBMX. Ten minutes later, cells were restimulated with the same concentration of MSH. Intracellular cAMP content was measured by ELISA. C, HEK 293 cells transfected with HA-hMC4R were exposed to 100 nM MSH for the indicated periods. Cell surface HA-antigen was measured by ELISA as described in Materials and Methods. Data represent means ± SEM of three independent experiments performed in duplicate (A, B) or triplicate (C).

View larger version (126K): [in this window] [in a new window]   Figure 4. Fluorescence imaging of the agonist-stimulated redistribution of HA-hMC4R-GFP in HEK 293 cells. The pEGFP-N1 (A) or GFP-conjugated hMC4R was transiently transfected in HEK 293 cells. Cells were then split into four-well chamber slides. Twenty-four hours later, cells were treated with PBS (B) or 100 nM MSH for 10 min (C) and 60 min (D) at 37 C. Cells were fixed for 10 min at room temperature with 3% paraformaldehyde in PBS (pH 7.4) and examined by microscopy on a Axioskop 2 fluorescence microscope (Carl Zeiss) using a Plan Fluor x100 objective.

View larger version (25K): [in this window] [in a new window]   Figure 5. Effect of forskolin and H89, a specific PKA inhibitor on internalization of HA-hMC4R in HEK 293 cells. HA-tagged hMC4R was transiently transfected in HEK 293 cells. A, Cells were stimulated with indicated concentrations of forskolin for 15 min. Intracellular cAMP formation was measured by ELISA (inset). Cells were exposed to 10 µM forskolin for the indicated periods. Cell surface HA-antigen was measured by ELISA as described in Materials and Methods. Data represent means ± SEM of two independent experiments performed in triplicate. B, Cells were pretreated with 0.5% dimethylsulfoxide (DMSO) or H89 at 0.1–10 µM for 30 min and then stimulated with 100 nM MSH for the time indicated. Cell surface HA-antigen was measured by ELISA as described Materials and Methods. Data represent means ± SEM of three independent experiments performed in triplicate. *, P < 0.05 vs. DMSO.

View larger version (28K): [in this window] [in a new window]   Figure 6. RT-PCR analysis of gene expression of GRK 2–6, ß-arrestin 1 and 2, and dynamin I in GT1–7 cells and medial hypothalamic tissues of mice. Total RNA was obtained from GT1–7 cells and medial hypothalamic tissues of mice. One-half microgram total RNA was reverse transcribed and subjected to PCR using specific oligo primers for GRK2–6, ß-arrestin1 and 2, and dynamin I. To visualize the PCR products, the samples were subjected to electrophoresis in 1% agarose gel containing ethidium bromide.

Effect of overexpression of GRK2 or DN-GRK2 on MSH-induced MC4R internalization

As shown in Fig. 7A, Western blot analysis revealed that, in both HEK 293 and COS-1 cells, wild-type or mutant GRK2 is overexpressed by the transfection with these expression vectors, compared with the endogenous expression of GRK2 in cells transfected with pcDNA3. The relative expression level of GRK2 or GRK2-K220R to endogenous GRK2 (pcDNA3) was considerably higher in COS-1 cells than in HEK293 cells.

View larger version (24K): [in this window] [in a new window]   Figure 7. Effects of wild-type GRK2 and GRK2-K220R on MSH-mediated internalization of HA-hMC4R in HEK 293 cells (left) and COS-1 cells (right). GRK2, GRK2-K220R, or pcDNA3 was cotransfected with HA-tagged hMC4R in HEK 293 and COS-1 cells. A, Cells grown in a 10-cm dish were lysed with 1 ml RIPA buffer 48 h after transfection and centrifuged. Twenty microliters supernatant fraction were subjected to SDS-PAGE followed by Western blotting and interaction with anti-GRK2 antibody. HEK 293 (B) or COS-1 cells (C) expressing HA-hMC4R were stimulated with 100 nM MSH for the time indicated, and cell surface HA-antigen was measured by ELISA as described in Materials and Methods. Data represent means ± SEM of three independent experiments performed in triplicate and are expressed as percent loss of cell surface HA-antigen from nonstimulated.

Effect of overexpression of wild-type or dominant-negative ß-arrestin1 and dynamin I on MSH-induced MC4R internalization
We next addressed whether MC4R internalization was a ß-arrestin- or dynamin-dependent process. We cotransfected HEK 293 cells with HA-hMC4R and ß-arrestin1, ß-arrestin1-V53D, dynamin I, or dynamin-K44A. ß-Arrestin1-V53D binds better to clathrin than ß-arrestin1 but is impaired in its interaction with phosphorylated GPCRs (46). Dynamin-K44A is deficient in GTP binding and functions to inhibit dynamin-mediated scission of clathrin-coated vesicles from the plasma membrane (47).

As shown in Fig. 8A, Western blot analysis confirmed higher expression levels of ß-arrestin 1, ß-arrestin 1-V53D, dynamin I, and dynamin I-K44A in HEK 293 cells transfected with these expression vectors than those transfected with pcDNA3.

View larger version (23K): [in this window] [in a new window]   Figure 8. Effects of ß-arrestin 1, ß-arrestin 1-V53D, dynamin I, and dynamin I-K44A on MSH-mediated internalization of HA-hMC4R in HEK 293 cells. HEK 293 was transfected with ß-arrestin 1, ß-arrestin 1-V53D, dynamin I, and dynamin I-K44A, or pcDNA3 and HA-tagged hMC4R. A, Cells grown in a 10-cm dish were lysed with 1 ml RIPA buffer 48 h after transfection and centrifuged. Twenty microliters supernatant fraction were subjected to SDS-PAGE followed by Western blotting and interaction with anti-ß-arrestin 1 or anti-dynamin I antibody. C, Cells were stimulated with 100 nM MSH for the time indicated, and cell surface HA-antigen was measured by ELISA as described in Materials and Methods. Data represent means ± SEM of three independent experiments performed in triplicate and are expressed as percent loss of cell surface HA-antigen from nonstimulated. *, P < 0.05; **, P < 0.01 vs. pcDNA3.

Effect of mutation of serine/threonine residues in third internal loop and carboxy-terminal tail of hMC4R on internalization of the receptor
In general, GRK-mediated phosphorylation of serine or threonine residues located in third internal loop or C-terminal tail of activated GPCR allows the recruitment of ß-arrestin leading to internalization. We have found that PKA- or GRK-mediated phosphorylation of MC4R contributes to the agonist-mediated receptor internalization. To further address which seine/threonine residue would be phosphorylated by PKA and GRKs and has critical roles in the internalization of MC4R, we introduced multiple mutations into the third internal loop and C-terminal tail of MC4R, focusing on the serine and threonine residues. The third internal loop and C-terminal tail mutations are depicted schematically in Fig. 9A.

View larger version (20K): [in this window] [in a new window]   Figure 9. Effects of serine/threonine mutations in the third internal loop and C-terminal tail of HA-hMC4R on MSH-mediated internalization of the receptor in HEK 293 cells. A, Schematic of hMC4R third internal loop and C-terminal tail and mutations. The indicated serine and threonine residues were changed to alanine by site-directed mutagenesis. B, HEK 293 cells were transiently transfected with HA-tagged hMC4R or HA-tagged mutant hMC4R. Cells were stimulated with 100 nM MSH for the time indicated, and cell surface HA-antigen was measured by ELISA as described in Materials and Methods. Data represent means ± SEM of three independent experiments performed in triplicate and are expressed as percent loss of cell surface HA-antigen from nonstimulated. *, P < 0.05; **, P < 0.01 vs. wild-type hMC4R.

Effects of AGRP (83–132) on MSH-induced cAMP formation in GT1–7 cells and cell surface MC4R in HEK 293 cells
To explore whether AGRP influences cell surface MC4R expression and cAMP formation in response to agonist, GT1–7 cells or HEK 293 cells expressing HA-hMC4R was treated with various concentrations of AGRP (83–132) (Fig. 10). Pretreatment with 0.01–1 nM AGRP (83–132) for 30 min produced no obvious influence on MSH-induced cAMP formation in GT1–7 cells, whereas 10 nM AGRP (83–132) attenuated the MSH-mediated cAMP formation (Fig. 10A). The attenuation of cAMP formation by 10 nM AGRP (83–132) is probably due to the competitive antagonism of MSH binding to MC4R by AGRP (83–132), which still remained at MC4R following three consecutive washout. However, 24-h preexposure of AGRP (83–132) potentiated the cAMP formation in response to 35 nM MSH in a concentration-dependent manner. The attenuated cAMP formation by 10 nM AGRP (83–132) is presumably due to the competitive antagonism of AGRP as described above. These data clearly indicate that exposure of MC4R to AGRP results in the increased sensitivity to subsequent stimulation with MSH.

View larger version (20K): [in this window] [in a new window]   Figure 10. Effects of AGRP (83–132) on MSH-induced cAMP formation in GT1–7 cells (A) and cell surface MC4R in HEK 293 cells (B). A, GT1–7 cells were exposed to the indicated concentration of AGRP (83–132) either for 30 min or 24 h. Then cells were washed with Opti-MEM I three times and incubated with Opti-MEM I containing 0.5 mM IBMX. Ten minutes later, cells were restimulated with 35 nM MSH. Intracellular cAMP content was measured by ELISA. Data represent means ± SEM of representative data from two independent experiments with similar results performed in triplicate. *, P < 0.05; **, P < 0.01 vs. 30-min exposure. B, HEK 293 cells transfected with HA-hMC4R were exposed to the indicated concentrations of AGRP (83–132) for 6 h. Cell surface HA-antigen was measured by ELISA as described in Materials and Methods. Data represent means ± SEM of representative data from two independent experiments with similar results performed in triplicate. *, P < 0.05; **, P < 0.01 vs. PBS.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
The major findings of the present study are as follows: 1) MC4Rs display desensitization in response to agonist, as detected by impaired agonist-mediated cAMP formation after preexposure to the agonist in GT1–7 cells; 2) activation of MC4R by agonist is associated with time-dependent internalization of the receptor in HEK 293 cells; 3) fluorescence microscopy analysis demonstrates the sequestration of GFP-conjugated MC4R from the cell membrane, seen as a distinct punctate pattern, as early as 10 min after MSH exposure; 4) the internalization of MC4R by agonist is partly inhibited by treatment with H89, a specific PKA inhibitor, or overexpression of GRK2-K220R; 5) the internalization of MC4R by agonist is ß-arrestin and dynamin dependent; 6) Thr312 and Ser329/330 in the C-terminal tail of MC4R are potential sites for PKA and GRK phosphorylation and the subsequent recruitment of ß-arrestin to the activated receptor; and 7) AGRP, an endogenous antagonist of MC4R, increases cell surface MC4R.

RT-PCR analysis showed that hypothalamic GT1–7 cells predominantly express mRNA for MC4R, compared with MC3R, whereas medial hypothalamic tissues express both MC3R and MC4R mRNA, suggesting that GT1–7 cells are a suitable model for the evaluation of MC4R signaling. The desensitization of MC4R progressed rapidly during the first 30 min, and cAMP formation was reduced to less than 10% of control following 3-h preexposure of 100 nM MSH (Fig. 1D). Because forskolin-mediated cAMP formation was not influenced by the pretreatment with MSH (Fig. 2B), the desensitization by MSH in GT1–7 cells is probably due to the decrease in cell surface MC4R, uncoupling of G-protein from MC4R, or both. In our data, cAMP formation in response to 100 nM MSH was approximately 16 times less in GT1–7 cells than HEK 293 cells (Fig. 1A vs. Fig. 3A), whereas the magnitude of the desensitization caused by 100 nM MSH was greater in the former than the latter (Fig. 1D vs. Fig. 3B). The reason for the difference remains unclear at present, but it might be possible that, because MC4R in HEK 293 cells was overexpressed by transient transfection of MC4R cDNA, the expression level of MC4R was much greater in HEK 293 cells than GT1–7 cells; therefore, the endogenously expressed G protein that binds to the receptor was not sufficient to desensitize the receptor efficiently in HEK 293 cells. Of particular interest is that desensitization was observed by MSH at concentrations that produce minimal increase in cAMP formation (Figs. 1E and 2A). This may indicate that MC4R is readily desensitized by a small increase of cAMP associated with low receptor occupancy, as seen in ß₂-adrenoceptor, which is triggered at very low receptor occupancy, and only subtle increases in cAMP are needed to fully activate PKA (35).

It might be argued that the agonist-mediated decrease in cell-surface MC4R is caused not only by the internalization of the agonist-activated MC4R but also by a decrease of synthesis or translocation to the membrane of de novo synthesized receptors. It is not clear how many de novo receptors are newly synthesized and translocated to the cell membrane during a 60-min test period that was performed 48 h after the transfection of HA-hMC4R cDNA. It appears evident that MSH causes the rapid sequestration of GFP-conjugated MC4R from the cell membrane, as shown in Fig. 4. Furthermore, there was no obvious difference in the amounts of cell surface HA-epitope in HEK 293 cells between 48 h and 49 h after cDNA transfection (data not shown). Thus, although receptors might be synthesized de novo during a 60-min test period, it is unlikely that the 40–50% decrease in cell surface HA-MC4R would result from changes in this rate. To explore whether the translocation of MC4R to cell membrane is influenced by agonist, further studies would be required.

A large body of evidence has suggested that phosphorylation of serine/threonine residues in the third internal loop or C-terminal tail of GPCRs by serine/threonine kinases is involved in the desensitization and internalization of many GPCRs (21, 22, 48, 49). In our data, overexpression of GRK2-K220R lacking kinase activity and that of wild-type GRK2 had little or minimal effect on the agonist-promoted internalization of MC4R in HEK 293 cells. However, in COS-1 cells, overexpression of GRK2-K220R inhibited the agonist-mediated internalization of MC4R, whereas coexpression of wild-type GRK2 had little effect on the internalization. These findings suggest that GRKs play a crucial role in the agonist-mediated MC4R internalization, presumably through the phosphorylation of agonist-occupied receptor. The inability of GRK2-K220R to inhibit MC4R internalization in HEK 293 cells might result from a high affinity of endogenous GRKs for activated MC4R, resulting in insufficient competition of GRK2-K220R for receptor binding because HEK 293 cells are reported to express the highest level of endogenous GRKs and ß-arrestins, compared with other cell lines such as COS-7 and Chinese hamster ovary cells (44), as confirmed by Western blot analysis (Fig. 7A). Many in vitro studies have shown that most GPCRs can be phosphorylated by one or more different GRKs (22), although the extent of contribution or isoform specificity of GRK for the phosphorylation of the receptor remains unclear in most cases. The RT-PCR analysis indicates that mRNA for GRK2–6 is expressed both in hypothalamus and GT1–7 cells (Fig. 6). Future studies should shed light on which subtypes of GRKs are involved in the phosphorylation of MC4R expressed in the relevant hypothalamic neurons.

In HEK 293 cells expressing HA-hMC4R, although the exposure of the cells to forskolin at a concentration that increases cAMP 1.5-fold, compared with 100 nM MSH, this had little effect on the internalization of MC4R. In contrast, agonist-mediated internalization was suppressed by cotreatment with the specific PKA inhibitor H89, consistent with the fact that Thr312 (LRKTF) has the phosphorylation consensus motifs for PKA (XRRXS/T where X means any amino acid, means hydrophobic amino acid). The activation of MC4R by agonist causes PKA activation, as demonstrated by phosphorylation of CREB in GT1–7 cells. These results suggest that the agonist-mediated activation of G protein/adenylate cyclase and subsequent cAMP formation, a common stimulus that provokes activation of PKA, contribute to the desensitization and internalization of MC4R. Similar findings have been observed for the desensitization of MC2R: Adrenocortical ACTH-mediated desensitization of MC2R is inhibited by H89, whereas forskolin treatment does not cause desensitization of the receptor (43). The reason for the inability of forskolin to elicit the internalization of MC4R is unclear. Conformational change of the receptor following agonist binding might be necessary for the PKA-mediated phosphorylation of the receptor.

Several reports have established that GRK-mediated phosphorylation followed by ß-arrestin binding may represent a common mechanism required for the sequestration of many GPCRs (24, 50). ß-Arrestins can serve as adaptor proteins between the phosphorylated receptor and component of the endocytic machinery, such as adaptor protein 2 and clathrin (51, 52). In our data, overexpression of dominant-negative mutants of ß-arrestin1 and dynamin I inhibited the agonist-mediated internalization of MC4R by 54% and 90%, respectively, suggesting that ß-arrestin- and dynamin-dependent processes are involved in the internalization of MC4R. Because ß-arrestin1-V53D binds better to clathrin than ß-arrestin but is significantly impaired in its interaction with phosphorylated GPCRs (46), it competes with the wild-type ß-arrestin for clathrin binding. The incomplete blocking of DN-ß-arrestin1 on the agonist-mediated internalization might be due to the insufficient expression of DN-ß-arrestin1 to block the interaction of MC4R-bound ß-arrestin to clathrin. Alternatively, processes independent of ß-arrestin may also contribute to the internalization of the agonist-activated receptors. However, the latter is more likely because the MC4R-T312A/S329A/S330A mutant that lacks potential binding sites for ß-arrestin still undergoes internalization more than half of wild-type MC4R. RT-PCR analysis showed that GT1–7 cells expressed mRNAs of both ß-arrestins (Fig. 6), consistent with the finding that both ß-arrestin 1 and 2 are predominantly localized in neuronal tissues (53). ß-Arrestins share 78% homology and bind many receptors with comparable affinity (53). Dynamin I, a neuron-specific isoform, was equally expressed in GT1–7 cells and medial hypothalamic tissues (Fig. 6). Therefore, it is reasonable to assume that both ß-arrestin and dynamin are involved in the desensitization and internalization of MC4R in neurons, although it is still unclear which subtypes of arrestins contribute to the processes in the hypothalamic neuronal cells.

Replacement by alanine(s) of Thr312 and Ser329/330 in the C-terminal tail resulted in an impaired sequestration of mutated receptors to agonist, whereas mutations of Thr232 or Ser306 did not. This indicates that phosphorylation of these residues by kinases is critical for the internalization of MC4R. Although certain GPCRs are known to be internalized in a phosphorylation-independent manner (54) or in the absence of detectable agonist-induced phosphorylation (55), these results are consistent with the notion that, in many GPCRs, phosphorylation by GRK of serine/threonine residues in the C-terminal tail of GPCRs is a critical step for the desensitization and internalization (25, 56, 57). Because the Ser329/330 is consistent with the phosphorylation consensus motifs for GRK2 and GRK3 (the presence of acidic residue localized on the N-terminal side of target serines/threonines) (58), these serines may be phosphorylated by GRK, although the motif is not an essential determinant in directing phosphorylation. Gurevich et al. have demonstrated that only 2 mol of phosphate/mol are required for the high-affinity arrestin binding to rhodopsin (59), m₂ chorinergic (60), and ß₂-adrenergic receptors (61) using various truncated and chimeric arrestins. In contrast to other GPCRs that have several serine/threonine residues to be phosphorylated, the only two serine/threonine residues located in the C terminus are critical for the agonist-promoted internalization of MC4R. Taken together, it seems likely that ß-arrestin is recruited with high affinity to the C terminus of activated MC4R to phosphorylated Thr312 and Ser329/330. This hypothesis arises from the finding that the magnitude of internalization of MC4R-T312A/S329A/S330A mutant by agonist was almost identical with that of MC4R-T312A mutant, not synergically potentiated, compared with that of T312A or S329A/S330A mutant.

The MC4R-T312A/S329A/S330A mutant that lacks potential ß-arrestin binding sites still underwent agonist-mediated internalization by approximately 60% of wild-type MC4R, indicating that, in HEK 293 cells, less than half of agonist-mediated internalization is serine/threonine and arrestin dependent, and the rest is dynamin dependent but serine/threonine and arrestin independent. It might be therefore reasonable that the inhibition of the internalization by H89 (Fig. 5), GRK2 K220R (Fig. 7), and ß-arrestin1-V53D (Fig. 8) was incomplete (50% or less). The internalization of some GPCRs is less sensitive to the effects of ß-arrestin mutant. For example, the internalization of AT₁-angiotensin (62) and m₂ muscarinic (63) receptors is not blocked by DN-ß-arrestins. Caveolae have also been reported to mediate the internalization of several GPCRs (20). Caveolae are morphologically distinct from clathrin-coated vesicles, but their formation is reported to be dynamin dependent. Most GPCRs have a conserved NPXXY motif in the seventh transmembrane helices, and this motif is known as an internalization signal in ß₂-adrenergic receptors (64); initial asparagine is substituted to aspartic acid in all five subtypes of melanocortin receptors. Thus, alternative endocytic processes, in addition to the GRK/ß-arrestin-dependent process, may participate in the internalization of MC4R.

AGRP has been identified as competitively antagonizing the both MC4R and MC3R in the brain. Recently it has been suggested that AGRP acts as an inverse agonist for the MC4R in addition to a competitive antagonist (65, 66). In our data, preexposure of AGRP (83–132) to GT1–7 cells for 24 h potentiated the cAMP formation in response to MSH, whereas 30-min exposure did not. Furthermore, exposure of AGRP (83–132) alone for 6 h increased cell surface MC4R in HEK 293 cells expressing HA-hMC4R. These findings suggest that MC4R expressed in GT1–7 and HEK 293 cells internalizes spontaneously, even in the absence of agonist, and that AGRP (83–132) inhibits the internalization of the receptor from or recruits the receptor to the membrane, thereby leading to the increased cell surface MC4R number as well as cAMP formation in response to MSH. Thus, AGRP has a potential to up-regulate MC4R by stabilizing the inactive conformation of MC4R because inverse agonists have the ability to stabilize the inactive conformation of a receptor (67). The mechanism underlying this up-regulation is unclear at present and requires further investigation.

Although it has become clear from our data that MC4R signaling is regulated by desensitization and internalization of the receptor, there is no evidence that altered desensitization and internalization of MC4R are involved in the pathogenesis of obesity. Recently Pierroz et al. (68) have shown that treatment of mice with diet-induced obesity with MTII, an MSH analog, suppresses feeding during the first few days, with food intake returned to the level of nontreated mice thereafter. This may indicate that, even if leptin resistance is present, MC4R and downstream pathway is responsive to agonist given exogenously and that MC4R signaling may be readily desensitized by repeated doses. Several reports have shown that high-fat diet- or overfeeding-induced obesity in mice or rats is associated with increases in plasma leptin level and hypothalamic POMC expression (36, 69, 70), presumably leading to the enhanced MC4R signaling. Sustained rises in MC4R signaling may reduce the cell surface MC4R because of the enhanced desensitization and internalization of the receptor, thereby resulting in a sustained decrease in MC4R signaling, as reported by Harrold et al. (31), who showed that diet-induced obesity was associated with decreased hypothalamic MC4R expression in rats. In contrast, MTII is effective in reducing food intake and body weight in Zucker fatty rats, and the effects are more marked in obese rats than lean control rats (71). Zucker fatty rats are reported to show increased hypothalamic MC4R expression (31), probably because of decreased hypothalamic POMC expression associated with impaired leptin signaling.

Our data also suggest that AGRP has a potential to increase cell surface MC4R expression and cAMP formation in response to agonist. Taken together, the elevated melanocortin tone by overfeeding or a high-fat diet may facilitate the desensitization and internalization of MC4R, thereby leading to the down-regulation of MC4R. In addition, treatment with synthetic MC4R agonist may be effective in reducing food intake and body weight, but MC4R signaling may readily cause tachyphylaxis, especially in conditions of increased melanocortin tone. By contrast, it is tempting to speculate that the decreased melanocortin tone by food restriction or impaired leptin signaling may result in up-regulation of MC4R.

In summary, this study provides evidence that MC4R undergoes desensitization and internalization in response to agonist. MC4R desensitization and internalization involve PKA-, GRK-, ß-arrestin-, and dynamin-dependent pathways, indicating that these proteins play important roles in controlling the amplitude and duration of the MC4R signaling. The present study provides fresh insight into the regulation of MC4R from the viewpoint of both rapid (desensitization and internalization) and chronic (down-regulation) regulation of MC4R signaling. Impaired regulation of desensitization and internalization is a potential etiology of certain human diseases. Elevated GRK expression has been reported to accompany chronic heart failure (72) and hypertension (73). Recently Barak et al. (74) have found a naturally occurring loss-of-function mutation in the vasopressin receptor that is associated with familial nephrogenic diabetes insipidus by inducing constitutive arrestin-mediated desensitization. Thus, excessive GRK- and ß-arrestin-mediated desensitization and internalization of MC4R would be responsible for certain cases of human obesity. The regulation of central MC4R signaling is critical for the normal energy homeostasis. The present study provides a more complete understanding of the regulation of MC4R signaling and suggests new directions for research into MC4R signaling in the regulation of energy homeostasis.

	Acknowledgments

 
We are grateful to Dr. Richard I. Weiner, University of California San Francisco, for the gift of GT1–7 cells; Dr. Jeffrey L. Benovic, Thomas Jefferson University, for gifts of the GRK2 and GRK2-K220R cDNA; and Dr. Robert L. Lefkowitz, Duke University, for gifts of cDNAs encoding ß-arrestin1, ß-arrestin1-V53D, dynaimin I, and dynamin I-K44A. We thank Drs. Anthony T. Hollenberg, Christian Bjørbæk, Iphigenia Tzameli, Marisol Lopez, and Jonathan Hamm for their invaluable advice.

	Footnotes

 
This work was supported by grants from NIH (to J.S.F.).

Abbreviations: AGRP, Agouti-related protein; MSH, -melanocyte stimulating hormone; CREB, cAMP-responsive element binding protein; DN, dominant negative; DNase, deoxyribonuclease; GFP, green fluorescent protein; GPCR, G protein-coupled receptor; GRK, G protein-coupled receptor kinase; HA, hemagglutinin; HEK, human embryonic kidney; hMC4R, human melanocortin-4 receptor; IBMX, 3-isobutyl-1-methylxanthine; MC4R, melanocrtin-4 receptor; PKA, protein kinase A; POMC, proopiomelanocortin.

Received September 4, 2002.

Accepted for publication January 2, 2003.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Vergoni AV, Bertolini A 2000 Role of melanocortins in the central control of feeding. Eur J Pharmacol 405:25–32[CrossRef][Medline]
2. Cone R 1999 The central melanocortin system and energy homeostasis. Trends Endocrinol 10:211–216[CrossRef][Medline]
3. Mountjoy K, Mortrud M, Low M, Simerly R, Cone R 1994 Localization of the melanocortin-4 receptor (MC4-R) in neuroendocrine and autonomic control circuits in the brain. Mol Endocrinol 8:1298–1308[Abstract]
4. Schwartz M, Woods S, Porte DJ, Seeley R, Baskin D 2000 Central nervous system control of food intake. Nature 404:661–671[Medline]
5. Ahima R, Flier J 2000 Leptin. Annu Rev Physiol 62:413–437[CrossRef][Medline]
6. Elias C, Aschkenasi C, Lee C, Kelly J, Ahima R, Bjorbaek C, Flier J, Saper C, Elmquist J 1999 Leptin differentially regulates NPY and POMC neurons projecting to the lateral hypothalamic area. Neuron 23:775–786[CrossRef][Medline]
7. Fan W, Boston B, Kesterson R, Hruby V, Cone R 1997 Role of melanocortinergic neurons in feeding and the agouti obesity syndrome. Nature 385:165–168[CrossRef][Medline]
8. Seeley R, Yagaloff K, Fisher S, Burn P, Thiele T, van Dijk G, Baskin D, Schwartz M 1997 Melanocortin receptors in leptin effects. Nature 390:349[Medline]
9. Huszar D, Lynch C, Fairchild-Huntress V, Dunmore J, Fang Q, Berkemeier L, Gu W, Kesterson R, Boston B, Cone R, Smith F, Campfield L, Burn P, Lee F 1997 Targeted disruption of the melanocortin-4 receptor results in obesity in mice. Cell 88:131–141[CrossRef][Medline]
10. Chen H, Charlat O, Tartaglia L, Woolf E, Weng X, Ellis S, Lakey N, Culpepper J, Moore K, Breitbart R, Duyk G, Tepper R, Morgenstern J 1996 Evidence that the diabetes gene encodes the leptin receptor: identification of a mutation in the leptin receptor gene in db/db mice. Cell 84:491–495[CrossRef][Medline]
11. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman J 1994 Positional cloning of the mouse obese gene and its human homologue. Nature 372:425–432[CrossRef][Medline]
12. Yaswen L, Diehl N, Brennan M, Hochgeschwender U 1999 Obesity in the mouse model of pro-opiomelanocortin deficiency responds to peripheral melanocortin. Nat Med 5:1066–1070[CrossRef][Medline]
13. Krude H, Biebermann H, Luck W, Horn R, Brabant G, Gruters A 1998 Severe early-onset obesity, adrenal insufficiency and red hair pigmentation caused by POMC mutations in humans. Nat Genet 19:155–157[CrossRef][Medline]
14. Jackson R, Creemers J, Ohagi S, Raffin-Sanson M, Sanders L, Montague C, Hutton J, O’Rahilly S 1997 Obesity and impaired prohormone processing associated with mutations in the human prohormone convertase 1 gene. Nat Genet 16:303–306[CrossRef][Medline]
15. Montague C, Farooqi I, Whitehead J, Soos M, Rau H, Wareham N, Sewter C, Digby J, Mohammed S, Hurst J, Cheetham C, Earley A, Barnett A, Prins J, O’Rahilly S 1997 Congenital leptin deficiency is associated with severe early-onset obesity in humans. Nature 387:903–908[CrossRef][Medline]
16. Clement K, Vaisse C, Lahlou N, Cabrol S, Pelloux V, Cassuto D, Gourmelen M, Dina C, Chambaz J, Lacorte J, Basdevant A, Bougneres P, Lebouc Y, Froguel P, Guy-Grand B 1998 A mutation in the human leptin receptor gene causes obesity and pituitary dysfunction. Nature 392:398–401[CrossRef][Medline]
17. Vaisse C, Clement K, Durand E, Hercberg S, Guy-Grand B, Froguel P 2000 Melanocortin-4 receptor mutations are a frequent and heterogeneous cause of morbid obesity. J Clin Invest 106:253–262[Medline]
18. Farooqi I, Yeo G, Keogh J, Aminian S, Jebb S, Butler G, Cheetham T, O’Rahilly S 2000 Dominant and recessive inheritance of morbid obesity associated with melanocortin 4 receptor deficiency. J Clin Invest 106:271–279[Medline]
19. Böhm S, Grady E, Bunnet N 1997 Regulatory mechanisms that modulate signaling by G-protein-coupled receptors. Biochem J 322:1–18[Medline]
20. Ferguson S 2001 Evolving concepts in G protein-coupled receptor endocytosis: The role in receptor desensitization and signaling. Pharmacol Rev 53:1–24[Abstract/Free Full Text]
21. Chuang TT, Iacovelli L, Sallese M, De Blasi A 1996 G protein-coupled receptors: heterologous regulation of homologous desensitization and its implications. Trend Pharmacol Sci 17:416–421