help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
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 Gubitz, A. K.
Right arrow Articles by Reppert, S. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gubitz, A. K.
Right arrow Articles by Reppert, S. M.
Endocrinology Vol. 141, No. 3 1236-1244
Copyright © 2000 by The Endocrine Society

ARTICLES

Chimeric and Point-Mutated Receptors Reveal That a Single Glycine Residue in Transmembrane Domain 6 Is Critical for High Affinity Melatonin Binding1

Amelie K. Gubitz2,3 and Steven M. Reppert

Laboratory of Developmental Chronobiology, Pediatric Service, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts 02114

Address all correspondence and requests for reprints to: Dr. Steven M. Reppert, Jackson 1226, Massachusetts General Hospital, Boston, Massachusetts 02114. E-mail: reppert{at}helix.mgh.harvard.edu


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
To delineate domains of high affinity melatonin receptors that are essential for melatonin binding, we generated chimeras between the human Mel1a melatonin receptor and the melatonin-related orphan H9 receptor. The latter receptor displays no high affinity melatonin binding. The chimeric receptors were transiently expressed in COS-7 cells and analyzed by radioligand binding using 2-[125I]iodomelatonin ([125I]Mel). Replacement of individual transmembrane domains (TMs) of the Mel1a receptor by the corresponding H9 helixes revealed that TM6 plays a critical role in ligand binding. Substitution of H9-TM6 into the Mel1a receptor abolished any detectable [125I]Mel binding, whereas the remaining TMs could be readily exchanged without affecting ligand binding. Subsequent site-directed mutagenesis showed that glycine 20 in TM6 of the Mel1a receptor occupies an important position in the binding site. Thus, the mutation of glycine 20 to threonine, the corresponding H9 residue, severely reduced the receptor’s affinity for melatonin. Furthermore, the double mutation of alanine 14 to cysteine and of glycine 20 to threonine in TM6 completely eliminated high affinity [125I]Mel binding. This strongly suggests that molecular modifications in TM6 that involve glycine 20 lead to steric incompatibilities in the binding pocket that prohibit high affinity melatonin binding.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
THE PINEAL HORMONE melatonin controls the timing of reproduction in seasonally breeding animals and modulates circadian rhythms in mammals (for review, see Ref. 1). These photoperiodic and circadian effects are mediated through guanine nucleotide-binding protein (G protein)-coupled receptors that bind the physiologically active melatonin agonist 2-[125I]iodomelatonin ([125I]Mel) with an equilibrium dissociation constant (Kd) of less than 100 pM and inhibit cAMP accumulation as their common signaling mechanism (1, 2, 3). To date, three types of such high affinity receptors have been cloned, designated the Mel1a, Mel1b, and Mel1c melatonin receptors (1)4. These receptors share a 60% overall amino acid identity and a 73% identity if only the transmembrane domains are taken into account. Both the Mel1a and Mel1b receptors have been cloned from mammals, whereas the Mel1c receptor has to date only been cloned from Xenopus laevis, zebrafish, and chickens (4). For the two mammalian receptors, sites of high levels of receptor expression include the hypothalamic suprachiasmatic nuclei (mainly Mel1a), the hypophyseal pars tuberalis (mainly Mel1a), and the retina (mainly Mel1b) (5, 6).

In addition to the high affinity melatonin receptors, a further member of this receptor family has been cloned, designated the melatonin-related H9 receptor (7). This receptor shares a 45% amino acid identity with the Mel1a and Mel1b receptors. However, ligand binding assays using transiently expressed recombinant H9 receptor as well as in vitro autoradiography strongly suggest that this receptor is unable to bind melatonin (7, 8). To date, neither the endogenous ligand nor the physiological role of the orphan H9 receptor has been elucidated.

Despite our progress in understanding the biological actions of melatonin and the detailed knowledge of the sites of receptor expression, little is known about the interaction between melatonin and its high affinity receptors at the molecular level. As generally accepted for the binding of small molecule ligands to G protein-coupled receptors, melatonin is thought to dock to a binding site buried in the transmembrane region of its receptors (9). Using computer-based strategies, several models for the ligand-binding site of melatonin receptors have been proposed (9, 10, 11, 12). Although such models potentially provide an important insight into receptor/ligand interactions at the molecular level, computational modeling only provides circumstantial information on binding sites. As a consequence, such models always require verification by more direct, mutagenesis-based techniques.

In the case of the melatonin-binding site, sole reliance on the available computer models appeared particularly unsatisfactory. Firstly, the amino acid residues proposed as binding sites for melatonin by the various models overlap only partially (9, 10, 11, 12). Secondly, with the exception of a single residue (serine 280 in Laitinen’s model) (9), all residues inferred to function as direct melatonin-docking sites by computer modeling are conserved in the cloned orphan H9 receptors (9, 10, 11, 12). Given the complete lack of high affinity binding at H9 receptors, it thus appears that the available models are incomplete. Finally, although two previous mutagenesis studies (13, 14) indicate that a histidine residue in transmembrane domain 5 (TM5) is involved in ligand binding at the Mel1a receptor as suggested by several models (9, 11, 12), mutagenesis-based findings for the other hypothetical docking sites are much less conclusive (13, 14).

In the present study we took a biomolecular approach to identify domains of melatonin receptors that are critical for hormone binding. This involved the generation of chimeric receptor proteins between the human Mel1a and H9 receptors. A major advantage of this strategy was that no presumptions on the molecular structure of the ligand-binding site had to be made. Chimeric approaches have been widely and successfully used to study various aspects of G protein-coupled receptor function, such as subtype selectivity (15, 16, 17), species selectivity (18), and effector coupling (15). A certain degree of amino acid conservation (minimum of about 40%) between the receptors to be merged is usually seen as a prerequisite to render chimeric strategies insightful. Thus, the 45% overall and 55% regional (transmembrane domains) amino acid identity between the human Mel1a and H9 receptors was clearly suited for the generation of informative chimeras. Moreover, the H9 receptor was a uniquely suitable player in this chimeric study, as its inability to bind melatonin facilitated the identification of Mel1a domains essential for ligand binding. Our chimeric strategy was followed by the generation of four point-mutated constructs designed to examine the roles of individual amino acid residues in melatonin binding.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Materials
[125I]Mel was purchased from NEN Life Science Products (2200 Ci/mmol; Boston, MA). All cell culture reagents were obtained from Life Technologies, Inc. (Grand Island, NY). Unless otherwise stated, all other chemicals were purchased from Sigma (St. Louis, MO) or Fisher Scientific (Springfield, NJ).

Construction of chimeras
Chimeric receptors were synthesized according to the PCR method of overlap extension (19). Full-length complementary DNA (cDNA) clones of the human Mel1a receptor (GenBank accession no. U14108) and the human H9 receptor (GenBank accession no. U52219) were subcloned into the mammalian expression vector pcDNA 3.1 (Invitrogen, San Diego, CA) and used as PCR templates. To generate the Mel1a/H9 receptor chimeras, the desired segments of each receptor were initially amplified in separate high stringency PCR reactions (1.5 mM MgCl2) using 2.5 U Pfu DNA polymerase (Stratagene, La Jolla, CA). These reactions were conducted with Gene Amp Kit reagents (Perkin-Elmer Corp./Cetus, Norwalk, CT), performing 30 cycles at 94 C for 1 min, 66 C for 1 min, and 72 C for 2 min in the presence of 200 µM deoxy-NTPs and 200 nM oligonucleotide primers. For chimeras A–D, primers at the fusion site (internal primers) introduced complementary sequence overlaps of 21 bp between the segments to be fused. For chimeras E–L, internal primers incorporated sequence coding for the transmembrane domains to be fused as 39- to 57-bp long, partly complementary overhangs. In all constructs, flanking primers added sequence encoding the hemagglutinin (HA) epitope to their 5'-end and the c-Myc epitope to their 3'-end. In addition, the flanking primers introduced restriction endonuclease sites for HindIII (5'-end) and EcoRV (3'-end). Amplified receptor fragments were separated on 1–2.5% agarose gels (depending on the product size) and purified using the Qiaquick Gel Extraction Kit (QIAGEN, Chatsworth, CA). Subsequently, the receptor fragments were fused in a second round of high stringency PCR (same conditions as above) in the presence of flanking primers only. Fused PCR products were purified as described before, digested with HindIII and EcoRV (Promega Corp., Madison, WI), and then subcloned into the pcDNA 3.1 vector. The sequences of all chimeric constructs were verified by dideoxynucleotide chain termination sequencing (20) using the T7 Sequenase version 2 kit (U.S. Biochemical Corp., Cleveland, OH).

Synthesis of point-mutated receptors
Internal primers for the PCR-based overlap extension method (19) were designed to introduce specific point mutations into the cDNA encoding the human Mel1a receptor (point mutants M to P). Point-mutated receptors were generated according to the procedure described for chimeric receptors, and their sequences were verified by dideoxy sequencing (20).

Transient expression of mutant receptors in COS-7 cells
COS-7 cells (6 x 106) were seeded into 150-mm culture dishes containing 25 ml DMEM supplemented with 10% FBS, penicillin (50 U/ml), streptomycin (50 µg/ml), and fungizone (125 ng/ml). Eighteen to 20 h after seeding, cells were acutely transfected with plasmid DNA (10 µg/dish) according to the diethylaminoethyl-dextran method (21), and grown for another 72 h in 5% CO2 at 37 C.

Radioligand binding assays
Three days after transfection, medium was aspirated, and cells were mechanically harvested into PBS. Crude cell pellets were prepared by spinning the cells at 1,600 x g first and then at 12,000 x g for 10 min each at 4 C. The cell pellets (aliquoted in four pellets per 150-mm confluent dish) were stored at -80 C. Saturation binding assays were performed by resuspending the crude cell pellets in binding buffer (50 mM Tris-HCl, pH 7.4, and 5 mM MgCl2). The cell suspension was then incubated with nine concentrations of [125I]Mel (5 pM to 1.28 nM) in the presence (nonspecific binding) or absence (total binding) of 10 µM melatonin for 2 h at room temperature (total volume, 200 µl). Subsequently, the samples were rapidly filtered through presoaked glass fiber filters and washed three times with ice-cold binding buffer using a Brandel cell harvester (Gaithersburg, MD). The radioactivity bound to the crude cell pellets was assessed in a {gamma}-counter. Each data point was determined in triplicate (total binding) or duplicate (nonspecific binding). The numbers of repeats per binding experiment are given in Table 1Go. Protein measurements were conducted by the method of Bradford (22). Data analysis and curve fitting were performed using the KaleidaGraph program (version 3.08, Abelbeck/Synergy Software, Reading, PA).


View this table:
[in this window]
[in a new window]
  		Table 1. Binding affinities of chimeric and point-mutated receptor constructs for [125I]Mel

 
Western blot analysis
Crude cell pellets were resuspended in a buffer containing 10 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, and 1 x complete protease inhibitor cocktail (Roche Molecular Biochemicals, Indianapolis, IN). The cell suspension was spun at 12,000 x g for 10 min at 4 C, and 10–30 µg of the resulting supernatant were separated by denaturing SDS-12.5% PAGE (see Bio-Rad Laboratories, Inc., product guide for detailed information, Hercules, CA). After the transfer of the proteins onto Protran pure nitrocellulose (Schleicher & Schuell, Inc., Keene, NH), the blots were blocked with 10% blocking reagent (1 h at room temperature; Bio-Rad Laboratories, Inc.). Subsequently, they were incubated with a 1:10,000 dilution of mouse monoclonal anti-HA antibody 12CA5 (1 h at room temperature; Babco, Richmond, CA) or with a 1:1000 dilution of mouse monoclonal anti-c-Myc antibody 9E:10 (Babco). Blots were then washed and exposed to a 1:10,000 dilution of horseradish peroxidase-conjugated goat antimouse IgG (1 h at room temperature; The Jackson Laboratory, Bar Harbor, ME). Proteins were visualized by enhanced chemiluminescence (NEN Life Science Products) using ECL Hyperfilm (Amersham Pharmacia Biotech, Little Chalfont, UK).

Immunofluorescence
COS-7 cells (3.5 x 105) were seeded on coverslips in six-well plates and acutely transfected with plasmid DNA (1 µg DNA/well) using the diethylaminoethyl-dextran method (21) after they had reached 60–80% confluence. Forty-eight hours after transfection, cells were fixed in 4% paraformaldehyde (10 min at room temperature) and blocked with 10% normal goat serum in PBS containing 1% BSA (1 h at room temperature). Subsequently, cells were incubated with mouse monoclonal anti-HA antibody (1:1000; 1 h at room temperature; Babco), washed, and then submerged into a 1:1000 dilution of CY3-conjugated antimouse IgG (1 h at room temperature; The Jackson Laboratory). After rinsing the cells with PBS, nuclei were stained with bis-benzamide and mounted with 50% glycerol in PBS. Immunofluorescence was observed using a Leitz Dialux 22 microscope (Rockleight, NJ).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Carboxyl tail exchange between the Mel1a and H9 receptors does not significantly alter the binding properties of either receptor
In contrast to the over 50% amino acid (aa) conservation between the transmembrane domains of high affinity melatonin and H9 receptors, a high degree of sequence discrepancy was found between their carboxyl-termini. All cloned H9 receptors have exceptionally long carboxyl tails (human H9, 318 aa; mouse H9, 289 aa; sheep H9, 280 aa), whereas the high affinity melatonin receptors have tails in the range of 50 aa residues. To examine whether the long carboxyl-termini of the H9 receptors interfere with melatonin binding, we created a construct (chimera A) in which the human H9 carboxyl tail replaced the tail of the human Mel1a receptor (schematically depicted in Fig. 1aGo). Examination of the binding characteristics of this construct revealed that chimera A was still able to bind [125I]Mel with high affinity. The Kd value for chimera A was 222.9 ± 64.6 pM, which did not significantly differ from the affinity of the epitope-tagged wild-type receptor (by Student’s unpaired t test, P > 0.05; see Table 1Go and Fig. 1bGo).



View larger version (29K):
[in this window]
[in a new window]
  		Figure 1. a, Schematic representation of chimeras A–D. Black, Segments of the human Mel1a receptor; hatched, segments of the human H9 receptor. b, Representative saturation binding isotherms of COS-7 cells expressing the HA/Myc-tagged human Mel1a receptor (left panel) and chimera A (right panel). The mean Kd and binding capacity (Bmax) values for chimera A and the HA/Myc-tagged Mel1a receptor are given in Table 1Go. {circ}, Total binding; •, specific binding; {triangleup}, nonspecific binding (determined in the presence of 10 µM melatonin). c, Western blot analysis of chimeric receptor constructs. Detergent-solubilized protein fractions from cells acutely transfected with receptor constructs were subjected to SDS-PAGE and immunoblotted with a mouse monoclonal anti-HA antibody and horseradish peroxidase-conjugated antimouse IgG. Proteins of approximately 70.6 kDa (chimera A) and 39.4 kDa (HA/Myc-tagged Mel1a receptor and chimeras B–D) were visualized by enhanced chemiluminescene. For the HA/Myc-tagged Mel1a receptor and chimera A, 10 µg solubilized protein were loaded; for chimeras B–D, 30 µg solubilized protein each were loaded.

 
We then generated an inverted version of chimera A by replacing the carboxyl-terminus of the H9 receptor with the one of the Mel1a receptor (chimera B, Fig. 1aGo). The objective here was to test whether it would be possible to restore melatonin binding at the H9 receptor by removal of the long tail. However, when chimera B was assayed in saturation binding experiments, no specific binding of [125I]Mel was detected (at the concentration range of 5 pM to 1.28 nM). Importantly, this absence of binding was not due to the failure of COS-7 cells to express chimera B. Using the anti-HA antibody, Western blots of cells acutely transfected with chimera B yielded a strongly immunoreactive band of approximately 39.4 kDa, the predicted protein size for this construct (Fig. 1cGo). Accordingly, Western blot analysis of chimera A gave evidence for robust receptor expression (immunoreactive band of approximately 70.6 kDa; Fig. 1cGo). For both constructs, similar, but weaker, signals were obtained using the anti-c-Myc antibody (data not shown). Taken together the binding properties of chimeras A and B showed that the carboxyl-termini per se are not critically involved in permitting or inhibiting melatonin binding at the Mel1a and H9 receptors.

TM6 plays an important role in high affinity melatonin binding at the Mel1a receptor
To determine whether important sites for melatonin binding at the Mel1a receptor are clustered more toward the carboxyl-terminus of the receptor (i.e. in TM5–7), as suggested by some computer models (9), or nearer the amino-terminus (TM1–4), two chimeric constructs were synthesized in which these regions were targeted. In chimera C, TM5–7 (plus the connecting loops) of the Mel1a receptor were substituted by their H9 counterparts, whereas in chimera D TM1–4 (plus the connecting loops and the second extracellular loop) of the Mel1a receptor were replaced by the corresponding H9 domains (Fig. 1aGo). Examination of the binding properties of these chimeras showed that neither construct supported high affinity binding of [125I]Mel (Table 1Go). As before, expression of chimeras C and D in the COS cells was verified by Western blot analysis to ensure that the lack of detectable binding was not due to a loss of construct expression (Fig. 1cGo). The complete absence of melatonin binding at these two chimeras thus clearly demonstrated that large sequence exchanges with the H9 receptor are permitted in neither the amino- nor the carboxyl-portion of the Mel1a receptor.

As chimeras C and D had provided evidence that both the amino-half and the carboxyl-half of the Mel1a receptor contain important determinants for melatonin binding, we next performed single transmembrane domain substitutions. In chimeras E to K, each of the seven transmembrane helixes of the Mel1a receptor was individually replaced by the corresponding H9 domain (Fig. 2Go). Screening of these chimeras in saturation binding assays revealed that six of these seven constructs had binding affinities for [125I]Mel identical or very similar to the native (and epitope-tagged) Mel1a receptor. As listed in Table 1Go for chimeras E, F, G, H, I, and K, their Kd values ranged from about 40–110 pM. Representative saturation binding isotherms of COS cells transfected with chimeras H, I, and K are given as examples for high affinity binding at these chimeric receptors in Fig. 3Go. In contrast, chimera J, which contained TM6 of the H9 receptor, did not exhibit any specific binding for [125I]Mel (used at 5 pM to 1.28 nM). Typical counts per min values obtained in saturation binding assays of [125I]Mel at chimera J are depicted in Fig. 3Go (bottom left panel).



View larger version (25K):
[in this window]
[in a new window]
  		Figure 2. Schematic representation of chimeras E–L. Black, Segments of the human Mel1a receptor; hatched, segments of the human H9 receptor.

 


View larger version (30K):
[in this window]
[in a new window]
  		Figure 3. Representative saturation binding isotherms of COS-7 cells expressing chimeras H, I, J, and K. The mean Kd and binding capacity (Bmax) values for these chimeric constructs are given in Table 1Go. {circ}, Total binding; •, specific binding; {triangleup}, nonspecific binding (determined in the presence of 10 µM melatonin).

 
To confirm that chimera J was unable to bind melatonin while being properly expressed in the cell membrane, we performed Western blot analysis (Fig. 4aGo) and immunofluorescence experiments (Fig. 4bGo). The latter technique is particularly powerful in demonstrating that a recombinant receptor is correctly inserted into the plasma membrane of transfected cells. Thus, if the cell membrane is kept intact and an antibody used that targets the amino-terminus, fluorescent labeling of cells only occurs if the transfected construct is incorporated in the plasma membrane with the amino-terminus pointing extracellularly. Using the anti-HA antibody for detection of the hemagglutinin epitope at the amino-terminus, we observed fluorescent staining of individual, unpermeabilized COS-7 cells acutely transfected with chimera J (Fig. 4bGo, right panel). Similar fluorescent labeling was observed for cells acutely transfected with the HA/Myc-tagged Mel1a receptor construct (Fig. 4bGo, left panel). In contrast, cells transfected with an HA-tagged construct of the period gene mPer3, which is only expressed intracellularly (23), remained unlabeled (data not shown). These observations showed unequivocally that chimera J was trafficked to and incorporated into the plasma membrane of the transfected cells as generally expected for a G protein-coupled receptor.



View larger version (89K):
[in this window]
[in a new window]
  		Figure 4. a, Western blot analysis of chimeric receptor constructs. Detergent-solubilized protein fractions from cells acutely transfected with receptor constructs were subjected to SDS-PAGE and immunoblotted with a mouse monoclonal anti-HA antibody and horseradish peroxidase-conjugated antimouse IgG. Proteins of approximately 70.6 kDa (chimera L) and 39.4 kDa (all other receptor constructs) were visualized by enhanced chemiluminescene. COS-7 cells transfected with a FLAG-tagged (amino-terminus) human Mel1a receptor construct served as a negative control. For the HA/Myc-tagged Mel1a receptor, 10 µg solubilized protein were loaded; for the FLAG-tagged Mel1a receptor and chimeras E–L, 30 µg solubilized protein each were loaded. b, Immunofluorescence labeling of the HA/Myc-tagged Mel1a melatonin receptor (left panel) and chimera J (right panel) on the plasma membrane of acutely transfected, unpermeabilized COS-7 cells. Cells were prepared as described in Materials and Methods, incubated with mouse monoclonal anti-HA antibody and CY3-conjugated antimouse IgG, and observed under a fluorescent microscope.

 
The total loss of [125I]Mel binding at chimera J clearly implied that amino acid residues in TM6 of the H9 receptor inhibit the Mel1a receptor’s ability to properly interact with its ligand. To investigate whether replacement of TM6 at the H9 receptor by the corresponding Mel1a domain would have the reverse effect and facilitate melatonin binding, we synthesized a construct comprised of the H9 receptor plus TM6 of the Mel1a receptor (chimera L; Fig. 2Go). However, saturation binding assays using COS-7 cells transfected with chimera L failed to reveal the presence of specific high affinity binding for [125I]Mel (Table 1Go). Again, evidence for the expression of chimera L was provided by Western blotting (band of ~70.6 kDa; Fig. 4aGo). The failure to obtain melatonin binding at the H9 receptor upon exchange of TM6 with the Mel1a receptor agreed with our findings for chimera D and implied that multiple domains of the H9 receptor are involved in prohibiting melatonin binding.

A single glycine residue in TM6 is critical for melatonin binding
To define which amino acid residues in TM6 of the H9 receptor could so dramatically disrupt ligand binding at the Mel1a receptor, we subsequently conducted single amino acid substitutions. Initially, a sequence alignment of TM6 of all high affinity melatonin receptors and H9 receptors was carried out (Fig. 5Go). This enabled us to distinguish highly conserved amino acid residues from variable ones. Of the 25 residues thought to constitute TM6 in the high affinity melatonin and H9 receptors (7), 11 are fully conserved among all known members of this family (group 1). A second group of residues shows variance, but the amino acid changes are relatively conservative (residues 4, 5, 7, 8, 16, 19, and 21; Fig. 5Go). Finally, a third group displays considerable variability among the high affinity receptors (residues 11, 18, 22, 24, and 25; Fig. 5Go), indicating that the diversity of these residues is tolerated for melatonin binding. Residues 14 and 20, however, fit in neither of these groups and struck us as interesting candidates for mutagenesis. Residue 14 is a conserved alanine in high affinity melatonin receptors (with the exception of the Mel1a Xenopus X2.0 clone and the chicken Mel1c receptor, where it is a glycine). In all cloned H9 receptors this residue is replaced by cysteine. Secondly, residue 20 is a fully conserved glycine in the high affinity group and a threonine in all H9 receptors. Thus, residues 14 and 20 are conspicuously nonhomologous between the high affinity receptors and the H9 group.



View larger version (113K):
[in this window]
[in a new window]
  		Figure 5. Amino acid alignment for TM6 of members of the melatonin receptor family. Black background indicates fully conserved residues; dark gray background indicates residues conserved among the high affinity melatonin receptors; light gray background indicates residues conserved among the H9 receptors. Point-mutated amino acid residues are marked with an arrow.

 
To evaluate the roles of residues 14 and 20 in melatonin binding, we synthesized mutant Mel1a receptors in which these two amino acids were changed to their H9 counterparts (point mutants M and N; Fig. 6aGo) and tested their binding characteristics. In addition, a third mutant was created in which the fully conserved cysteine 12 was changed to alanine (point mutant P; Fig. 6aGo). The latter mutant served as a control to demonstrate that not any random mutation in TM6 would restrict the ability of the Mel1a receptor to bind [125I]Mel. Examination of point mutant M in binding assays revealed that the mutation of glycine 20 to threonine has a profound inhibitory effect on melatonin binding. As shown in Fig. 6bGo, this mutant exhibited only residual amounts of specific [125I]Mel binding (used from 5 pM to 1.28 nM). Furthermore, there was clearly no saturation apparent, thus indicating that the Kd value for point mutant M might be several orders of magnitude higher than that for the native, epitope-tagged Mel1a receptor. In contrast, point mutant N retained saturable, high affinity binding of [125I]Mel (Kd, 43.2 ± 22.8 pM; Table 1Go). The control mutant P also was able to bind melatonin with high affinity. In fact, the Kd value for point mutant P was slightly higher than that for the native receptor (Table 1Go).



View larger version (34K):
[in this window]
[in a new window]
  		Figure 6. a, Schematic representation of point mutants M–P. Black, Segments of the human Mel1a receptor; open circles, single amino acid substitutions as indicated in the title. b, Representative saturation binding isotherms of COS-7 cells expressing point mutant M and the HA/Myc-tagged human Mel1a receptor. Both receptors were assayed in parallel. Left panel, Double plotted isotherms of point mutant M and the HA/Myc-tagged human Mel1a receptor. Right panel, Isotherm of point mutant M alone with an increased scale. The mean Kd and binding capacity (Bmax) values for the HA/Myc-tagged Mel1a receptor are given in Table 1Go. Point mutant M: {circ}, total binding; •, specific binding; {triangleup}, nonspecific binding; HA/Myc-tagged Mel1a receptor: {diamond}, total binding; {diamondsuit}, specific binding; {triangledown}, nonspecific binding. c, Western blot analysis of mutant receptor constructs. Detergent-solubilized protein fractions from cells acutely transfected with receptor mutants were subjected to SDS-PAGE and immunoblotted with a mouse monoclonal anti-HA antibody and horseradish peroxidase-conjugated antimouse IgG. Proteins of approximately 39.4 kDa were visualized by enhanced chemiluminescene. Thirty micrograms of solubilized protein were loaded for each point mutant. d, Immunofluorescence labeling of point mutant O on the plasma membrane of acutely transfected, unpermeabilized COS-7 cells. Cells were prepared as described in Materials and Methods, incubated with mouse monoclonal anti-HA antibody and CY3-conjugated antimouse IgG, and observed under a fluorescent microscope.

 
Apart from introducing single point mutations, we also synthesized a double mutated construct in which alanine 14 and glycine 20 of the Mel1a receptor were simultaneously replaced by cysteine and threonine, respectively (point mutant O; Fig. 6aGo). Interestingly, this double mutation was found to have an even stronger obstructive effect on melatonin binding than the mutation of glycine 20 alone. For double mutant O, we were unable to detect any specific [125I]Mel binding between 5 pM and 1.28 nM. However, evidence for its expression in the COS-7 cells was readily obtained via Western blot analysis (Western blots for all point mutants are shown in Fig. 6cGo) and immunofluorescence (Fig. 6dGo). In sum, the binding properties of the point-mutated constructs demonstrated that within TM6 of the Mel1a receptor, glycine 20 plays a key role in enabling the receptor to bind its endogenous ligand, whereas alanine 14 is in a less critical position.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this study a chimeric approach was taken to delineate domains of the human Mel1a receptor that are critical for melatonin binding. Unlike site-directed mutagenesis, the use of chimeric proteins does not rely on targeting individual, preselected amino acids, but allows us to put larger segments of proteins under investigation. Given the relative paucity of information on ligand binding at high affinity melatonin receptors, this approach seemed advantageous. Furthermore, the existence of a closely related, yet pharmacologically different, family member strongly argued for the use of a chimeric strategy.

Based on the striking divergence between the carboxyl-termini of the human Mel1a and H9 receptors, we initially set out to investigate whether carboxyl tail exchange would alter the binding properties of either receptor (chimeras A and B). In general, the carboxyl-termini of G protein-coupled receptors are not believed to be directly involved in binding of small mol wt ligands such as melatonin (24). However, data from mutational studies have unequivocally shown that the carboxyl-termini control other important aspects of receptor function, such as coupling to effector G proteins, receptor desensitization via phosphorylation events, as well as receptor sequestration and down-regulation (25, 26).

Our observation that chimera A displayed binding properties comparable to the native Mel1a receptor clearly showed that this receptor tolerates large changes in the carboxyl tail without the loss of high affinity binding. Curiously, the calculated expression levels of chimera A were considerably lower than those for the epitope-tagged native receptor (see Table 1Go). However, the observed variability of Bmax values among all chimeric and point-mutated constructs prohibited any speculations on the existence of a lower (and undetected) affinity state for melatonin at chimera A. In our hands it has, in fact, to date been impossible to provide evidence for the existence of distinguishable low and high affinity states at melatonin receptors. Thus, the variable Bmax values of the receptor constructs were probably the result of slightly inconsistent transfection, expression, or trafficking efficiencies or divergent stabilities of the mutant proteins.

The finding that high affinity melatonin binding remained undetectable at the H9 receptor after removal of its long carboxyl-terminus showed convincingly that this receptor region alone is not responsible for inhibiting [125I]Mel binding at this receptor. Thus, taken together, the binding properties of chimeras A and B strongly argued that the carboxyl-termini of members of the melatonin receptor family do not play a key role in ligand binding. This fully agrees with our general understanding of the binding site for small ligands at G protein-coupled receptors (24, 25).

Subsequently, the focus of the study was shifted toward the transmembrane helixes of the Mel1a receptor, because generally these domains are thought to form a hydrophobic pocket in the cell membrane suited to bind small molecules (24). Two initial constructs (chimeras C and D) in which the amino and carboxyl halves of the Mel1a and H9 receptors had been interchanged provided evidence that both portions of the Mel1a receptor contain elements that critically contribute to melatonin binding. Important for the interpretation of these data is, however, the awareness that binding of a ligand can be inhibited in two ways. Firstly, ligand binding can be directly disrupted by removing amino acids that serve as docking sites. Secondly, binding can be indirectly obstructed by modifying the three-dimensional configuration of a receptor (misfolding). The absence of detectable [125I]Mel binding at chimeras C and D did not reveal whether this was due to a loss of docking sites or an alteration of the tertiary receptor structure. However, the fact that these constructs involved large sequence replacements suggested that receptor misfolding might have played an important role in prohibiting melatonin binding at chimeras C and D.

The exchange of individual transmembrane helixes between the Mel1a and H9 receptors revealed that fusion of H9-TM6 into the Mel1a receptor completely abolished [125I]Mel binding (chimera J). This striking loss of ligand binding unequivocally showed that TM6 of the H9 receptor contains residues that jeopardize melatonin binding, whereas the remaining TMs are structurally compatible with high affinity binding (if exchanged separately). Interestingly, these findings are not reflected by the degree of amino acid conservation between the Mel1a and H9 receptors in the individual TM helixes. Thus, transmembrane domains with a lower degree of sequence identity than TM6 (56%), such as TM1 (48%), TM4 (45%), and TM5 (52%), were exchangeable without impairing melatonin binding at the resulting chimeras.

Based on our finding that TM6 carries critical determinants for melatonin binding at the Mel1a receptor, individual amino acids in this helix were targeted by mutagenesis to examine their role in ligand binding. Two residues in TM6 of the Mel1a receptor were selected based on their biochemical and biophysical dissimilarities compared with the corresponding H9 residues. The fully conserved glycine 20 in TM6 of high affinity melatonin receptors is replaced by the bulkier and potentially hydrogen bond-forming threonine in all H9 receptors. Furthermore, alanine 14 in TM6 of most high affinity receptors is replaced by a cysteine in the H9 group, which, again, is biochemically different due to the sulfhydryl moiety. The substitution of glycine 20 by threonine led to a pronounced reduction in the affinity of the mutant for [125I]Mel. This inhibitory effect on melatonin binding provided compelling evidence that glycine 20 occupies a critical position in the ligand binding pocket. Due to the small size of this residue and its predicted position near the extracellular surface of the plasma membrane, it is conceivable that glycine 20 is important for maintaining the binding site accessible. Replacement of this residue by amino acids with bulky side-chains might therefore lead to an obstruction of the entrance to the binding pocket and/or disrupt the interface between two of the TM helixes.

The observation that the mutation of alanine 14 in TM6 to cysteine alone did not interfere with [125I]Mel binding, whereas its double mutation with glycine 20 completely abolished it, provided further proof that modifications involving glycine 20 severely compromise the binding pocket. However, this also demonstrated that alanine 14 itself does not serve as a direct melatonin-docking site, as its mutation was clearly tolerated. In the case of the double mutation, steric hindrance jointly originating from the introduced threonine and cysteine residues in TM6 might have been a key factor in prohibiting melatonin binding. As both residues introduce moieties that can form hydrogen and/or disulfide bonds, it is conceivable that such chemical interactions might prevent the ligand from accessing its binding site. Interestingly, our control point mutant P, where cysteine 12 in TM6 had been changed to an alanine residue, had a marginally higher affinity than the native receptor. Although this provided good evidence that not any random molecular modification in TM6 impairs ligand binding, it remains to be determined how this mutation can slightly enhance the binding affinity.

In conclusion, our chimeric strategy has revealed that TM6 of the Mel1a receptor contains critical determinants for melatonin binding. Furthermore, subsequent targeting of individual amino acids in this transmembrane helix has provided compelling evidence that a single glycine residue in TM6 plays a key role in melatonin binding. The high degree of amino acid conservation in the transmembrane regions among all high affinity melatonin receptors strongly suggests that our findings for the Mel1a receptor are relevant to the entire high affinity group. It is hoped that future mutagenesis studies along with biophysical strategies, such as x-ray crystallography, will further expand our knowledge on the molecular events underlying the interaction between melatonin and its high affinity receptors.


   Acknowledgments
 
We thank Drs. P. J. Richardson, J. D. Levine, and D. R. Weaver for valuable comments during the progress of this study; Dr. S. Sathyanarayanan for help with the immunofluorescence experiments; and Dr. M. J. Zylka for providing the HA-tagged mPer3 construct.


   Footnotes
 
1 This work was supported by NIH Grant DK-42125 and a Sponsored Research Agreement from Bristol-Myers Squibb Co. (to S.M.R.). Back

2 Supported by a Wellcome International Prize Travelling Research Fellowship (049087/Z/96/Z). Back

3 Current address: Department of Pharmacology, University of Cambridge, Tennis Court Road, Cambridge CB2 1QJ, United Kingdom. Back

4 According to the International Union of Pharmacology, the Mel1a melatonin receptor is referred to as the mt1 receptor, and the Mel1b melatonin receptor as the MT2 receptor. Back

Received October 6, 1999.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Reppert SM, Weaver DR 1995 Melatonin madness. Cell 83:1059–1062[CrossRef][Medline]
  2. Dubocovich ML, Takahashi, J 1987 Use of 2-[125I]iodomelatonin to characterize melatonin binding sites in chicken retina. Proc Natl Acad Sci USA 84:3916–3920[Abstract/Free Full Text]
  3. Reppert SM, Weaver DR, Rivkees SA, Stopa EG 1988 Putative melatonin receptors in a human biological clock. Science 242:78–81[Abstract/Free Full Text]
  4. Reppert SM, Weaver DR, Cassone VM, Godson C, Kolakowski Jr LF 1995 Melatonin receptors are for the birds: molecular analysis of two receptor subtypes differentially expressed in chick brain. Neuron 15:1003–1015[CrossRef][Medline]
  5. Reppert SM, Weaver DR, Ebisawa T 1994 Cloning and characterization of a mammalian melatonin receptor that mediates reproductive and circadian responses. Neuron 13:1177–1185[CrossRef][Medline]
  6. Reppert SM, Godson C, Mahle CD, Weaver DR, Slaugenhaupt SA, Gusella JF 1995 Molecular characterization of a second melatonin receptor expressed in human retina and brain: the Mel1b receptor. Proc Natl Acad Sci USA 92:8734–8738[Abstract/Free Full Text]
  7. Reppert SM, Weaver DR, Ebisawa T, Mahle CD, Kolakowski Jr LF 1996 Cloning of a melatonin-related receptor from human pituitary. FEBS Lett 386:219–224[CrossRef][Medline]
  8. Drew JE, Barrett P, Williams LM, Conway S, Morgan PJ 1998 The ovine melatonin-related receptor. J Neuroendocrinol. 10:651–661
  9. Navajas C, Kokkola T, Poso A, Honka N, Gynther J, Laitinen JT 1996 A rhodopsin-based model for melatonin recognition at its G protein-coupled receptor. Eur J Pharmacol 304:173–183[CrossRef][Medline]
  10. Sugden D, Chong NWS, Lewis DFV 1995 Structural requirements at the melatonin receptor. Br J Pharmacol 114:618–623[Medline]
  11. Garratt PJ, Travard S, Vonhoff S, Tsotinis A, Sugden D 1996 Mapping the melatonin receptor. IV. Comparison of the binding affinities of a series of substituted phenylalkyl amides. J Med Chem 39:1797–1805[CrossRef][Medline]
  12. Grol CJ, Jansen JM 1996 The high affinity melatonin binding site probed with conformationally restricted ligands. II. Homology modeling of the receptor. Bioorg Med Chem 4:1333–1339[CrossRef][Medline]
  13. Conway S, Canning SJ, Barrett P, Guardiola-Lemaitre B, Delagrange P, Morgan PJ 1997 The roles of valine 208 and histidine 211 in ligand binding and receptor function of the ovine Mel1aß melatonin receptor. Biochem Biophys Res Commun 239:418–423[CrossRef][Medline]
  14. Kokkola T, Watson MA, White J, Dowell S, Foord SM, Laitinen JT 1998 Mutagenesis of human Mel1a receptor expressed in yeast reveals domains important for receptor function. Biochem Biophys Res Commun 249:531–536[CrossRef][Medline]
  15. Kobilka BK, Kobilka TS, Daniel K, Regan JW, Caron MG, Lefkowitz RJ 1988 Chimeric {alpha}2-, ß2-adrenergic receptors: delineation of domains involved in effector coupling and ligand binding specificity. Science 240:1310–1316[Abstract/Free Full Text]
  16. Frielle T, Caron MG, Lefkowitz RJ 1989 Properties of the ß1- and ß2-adrenergic receptor subtypes revealed by molecular cloning. Clin Chem 35:721–725[Abstract/Free Full Text]
  17. Rivkees SA, Lasbury ME, Barbhaiya H 1995 Identification of domains of the human A1 adenosine receptor that are important for binding receptor subtype-selective ligands using chimeric A1/A2a adenosine receptors. J Biol Chem 270:20485–20490[Abstract/Free Full Text]
  18. Tucker AL, Robeva AS, Taylor HE, Holeton D, Bockner M, Lynch KR, Linden J 1994 A1 adenosine receptors. Two amino acids are responsible for species differences in ligand recognition. J Biol Chem 269:27900–27906[Abstract/Free Full Text]
  19. Ho SN, Hunt HD, Horton RM, Pullen JK, Pease LR 1989 Site-directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 77:51–59[CrossRef][Medline]
  20. Sanger F, Nicklen S, Coulson AR 1977 DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 74:5463–5467[Abstract/Free Full Text]
  21. Cullen BR 1987 Use of eukaryotic expression technology in the functional analysis of cloned genes. Methods Enzymol 152:684–704[Medline]
  22. Bradford MM 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254[CrossRef][Medline]
  23. Zylka MJ, Shearman LP, Weaver DR, Reppert SM 1998 Three period homologs in mammals: differential light responses in the suprachiasmatic circadian clock and oscillating transcripts outside of brain. Neuron 20:1103–1110[CrossRef][Medline]
  24. Strader CD, Fong TM, Graziano MP, Tota MR 1995 The family of G-protein-coupled receptors. FASEB J 9:745–754[Abstract]
  25. O’Dowd BF, Lefkowitz RJ, Caron MG 1989 Structure of the adrenergic and related receptors. Annu Rev Neurosci 12:67–83[CrossRef][Medline]
  26. Liggett SB, Freedman NJ, Schwinn DA, Lefkowitz RJ 1993 Structural basis for receptor subtype-specific regulation revealed by a ß2/ß3-adrenergic receptor. Proc Natl Acad Sci USA 90:3665–3669[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Mol. Endocrinol.Home page
C. S. Nelson, M. Ikeda, H. S. Gompf, M. L. Robinson, N. K. Fuchs, T. Yoshioka, K. A. Neve, and C. N. Allen
Regulation of Melatonin 1a Receptor Signaling and Trafficking by Asparagine-124
Mol. Endocrinol., August 1, 2001; 15(8): 1306 - 1317.
[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 Gubitz, A. K.
Right arrow Articles by Reppert, S. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gubitz, A. K.
Right arrow Articles by Reppert, S. M.

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