- Email: [email protected]
The Roles of Valine 208 and Histidine 211 in Ligand Binding and Receptor Function of the Ovine Mel1aβMelatonin Receptor
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
239, 418–423 (1997)
RC977482
The Roles of Valine 208 and Histidine 211 in Ligand Binding and Receptor Function of the Ovine Mel1ab Melatonin Receptor Shaun Conway,*,1 Sarah J. Canning,* Perry Barrett,* Beatrice Guardiola-Lemaitre,† Phillipe Delagrange,† and Peter J. Morgan* *Molecular Neuroendocrinology Unit, Rowett Research Institute, Greenburn Road, Bucksburn, Aberdeen, Scotland, United Kingdom AB21 9SB; and †Institut de recherches Internationales Servier, 6 place de Ple´iades, F-92415 Courbevoie Cedex, France
Received September 8, 1997
Site-directed mutagenesis was used to study two residues, valine 208 and histidine 211, in transmembrane domain 5 of the ovine Mel1ab melatonin receptor. A series of 4 mutants were constructed (V208A, V208L, H211F, H211L), and each engineered to contain a FLAG-epitope. Immunocytochemistry demonstrated that all the mutants were expressed in COS-7 cells at levels comparable to the FLAG-epitope tagged wildtype Mel1ab receptor (Ç120 fmol/mg protein). Ligand binding revealed however that all mutants had reduced affinities for 2-[125I]-iodomelatonin (Kd wildtype 139 pM, Kd mutants 320 to 989 pM). Competition studies, with a series of melatonin analogues, identified a probable interaction between histidine 211 and the 5-methoxy group of melatonin. The wild-type receptor and both valine 208 mutants displayed a dosedependent melatonin mediated inhibition of cyclic AMP levels in HEK293 cells, with IC50 values in the same rank-order as their melatonin binding affinities. Both H211F and H211L, however, did not display any melatonin mediated effects and may suggest that histidine 211 is critical for melatonin mediated receptor activation. q 1997 Academic Press
Melatonin synchronizes circadian rhythms in mammals (1), and regulates aspects of reproductive state 1 Corresponding author. Fax: /44 (0)1224 716653. E-mail: sco@ rri.sari.ac.uk. Abbreviations: 125I-Mel, 2-[125I]-iodomelatonin; GPCR(s), G protein-coupled receptor(s); TM, transmembrane segment; 5-HT, 5-hydroxytryptamine; N-AS, N-acetylserotonin; MSH, melanocyte stimulating hormone, DEAE-dextran, diethylaminoethyl-dextran; PBS, phosphate buffered saline (pH 7.4); EGTA, ethylene glycol-bis(b-aminoethyl ether) N,N,N*,N*-tetraacetic acid; DMEM, Dulbecco’s modified Eagle’s medium; EMEM, Eagle’s minimum essential medium; PEG 8000, polyethylene glycol (average molecular weight 8,000); IBMX, 3-isobutyl-1-methylxanthine; a-NDP-MSH, [Nle4,D-Phe7]-amelanocyte stimulating hormone; TCA, trichloroacetic acid.
0006-291X/97 $25.00
in seasonally breeding mammals (2). These effects are achieved via the binding of melatonin to high-affinity cell surface melatonin receptors (3,4,5). Despite our insight into the possible functions of melatonin receptors little is known about the molecular structure of the receptor or receptor-ligand interactions, information that could lead to the design of specific agonist and antagonist modulators of the melatonin signal. The clinical application of such drugs could prove beneficial as administered melatonin has already been shown to modulate the sleep-wake cycle in the blind (6,7) and provides an effective treatment for jet-lag (8). The recent cloning of a series of high-affinity melatonin receptors has identified them as members of the GPCR ‘superfamily’ (9,10,11). It has been proposed that melatonin receptors will have a heptahelical architecture containing an intrahelical ligand binding site, in common with other GPCRs (9,12). One feature of many GPCRs is that the general structure of the ligand binding site appears to remain essentially conserved. This is illustrated by the alignment of residues that have been shown to provide ligand binding interactions in diverse GPCRs. For example two TM5 serine residues in the b2-adrenergic have been proven to interact with the two hydroxyl groups on the aromatic ring of adrenaline (13). Residues which protein sequence align with these serine residues, placing them in the same presumed position within each receptor, have also been shown to interact with the substituted ring of 5-HT in certain 5-HT receptors (14,15). The two residues which are in analogous TM5 positions in melatonin receptors are a conserved valine and a conserved histidine (valine 208 and histidine 211, ovine Mel1a receptor). At present the structure of the melatonin binding site is unknown, but if it is essentially conserved with those of adrenergic and 5-HT receptors then these two residues may also be involved in ligand binding, potentially interacting
418
Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 7482
/
693b$$$221
09-22-97 21:16:03
bbrcgs
AP: BBRC
Vol. 239, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS studies the ovine Mel1ab receptor cDNA was ligated into the vector pBK-CMV (Stratagene) and the ovine MSH receptor cDNA into vector pcDNA I (Invitrogen). COS-7 cells and HEK293 cells were provided by ECACC. 125I-Mel (2200 Ci/mmol) was purchased from NEN DuPont. The drugs S20098 and N-NEA (N-[2-(1-naphthyl) ethyl] acetamide) were synthesised by Servier (see Fig. 2 for structures). Other general reagents were purchased from Sigma or Life Technologies Inc. Site-directed mutagenesis. Mutations were introduced into the ovine Mel1ab melatonin receptor in pBK-CMV using the Chameleon site-directed mutagenesis kit (Stratagene). Successful construction of each mutant was confirmed by DNA sequencing using the Prism dyedeoxy terminator kit (Applied Biosystems Inc.). In this study 4 mutants were generated, mutants V208A, V208L, H211F, and H211L. FLAG epitope tagging. A PCR based protocol was used to introduce the FLAG-epitope (DYKDDDDK), encoded by the nucleotide sequence 5*»GACTACAAGGACGACGACGATAAG…3 *, between codon 1 and codon 2 of the ovine Mel1ab receptor cDNA. FLAG-epitope tagged mutant receptors were generated by subcloning fragments containing the desired mutations into the FLAG-epitope tagged wild-type receptor construct. All constructs were confirmed by DNA sequencing. COS-7 cell tissue culture. COS-7 cells were grown as monolayers in DMEM supplemented with 10% fetal calf serum and 1% antibiotic/ antimycotic solution, in 5% CO2 at 377C. Confluent plates were reseeded at 15,000 cells/cm2 and used 24 hours later for transfection. Cells were transfected with receptor constructs using the DEAEDextran method of Cullen (18) and grown for 72 hours. Growth media was aspirated and cells washed twice with PBS before being harvested. Cells were pelleted in 1.5 ml microcentrifuge tubes (13,000 g, 1 min, 47C), and stored frozen at 0807C.
FIG. 1. 125I-Mel saturation binding. (A) Specific binding of 125I-Mel to COS-7 cells transfected with the ovine Mel1ab melatonin receptor, and (B) specific binding of 125I-Mel to COS-7 cells transfected with the FLAG-epitope tagged ovine Mel1ab melatonin receptor. Data shown are from a single experiment using mean values of triplicate determinations, and are representative of 3 similar experiments. Binding parameters calculated by Scatchard analysis (20) were, ovine Mel1ab melatonin receptor Kd 93.2 { 11.3 pM, Bmax 109.3 { 8.7 fmol/mg protein; FLAG-epitope tagged ovine Mel1ab melatonin receptor, Kd 138.9 { 16.7 pM, Bmax 121.8 { 16.8 fmol/mg protein (mean { S.E., nÅ3).
Immunocytochemistry. COS-7 cells were transfected as above and incubated for 24 hours. Cells were trypsinized and seeded onto poly-D-lysine coated glass Lab-Tek chamber-slides (Nunc), at 125,000 cells per chamber, and cultured for a further 24 hours. Slides were washed twice in PBS, fixed with 4% para-formaldehyde in PBS (47C, 20 min), then thoroughly washed with 3 changes of PBS followed by 3 serial 5 minute washes in antibody binding buffer (1% horse serum, 0.1% saponin in PBS). Fixed cells were incubated with M5 antiFLAG antibody (IBI, Kodak), at a concentration of 5mg/ml in antibody binding buffer (16 hours, 47C). Samples were washed 3 times in antibody binding buffer, and incubated for 2 hours at room temperature with a 1:200 dilution of fluorescein conjugated horse anti-mouse IgG antibody in PBS. Finally cells were washed in PBS and studied by immunofluorescence microscopy. 125
with the 5-methoxy substituted ring of melatonin. Evidence in support of such interactions was recently provided by a computationally derived model of the melatonin receptor, based on the structure of the visual GPCR rhodopsin (16). This theoretical model proposes a role for the histidine in interactions with the oxygen of the 5-methoxy group of melatonin, and a role to valine 208 in providing a aliphatic milieu for the methyl group. In this study we have used site-directed mutagenesis to investigate the requirement of the ovine Mel1ab melatonin receptor for valine 208 and histidine 211. This study represents the first such investigations into the structure/ function relationships of melatonin receptors. MATERIALS AND METHODS Materials. The cloning of the ovine Mel1ab receptor and the ovine MSH receptor have been previously reported (11,17). For expression
I-Mel equilibrium binding studies. 125I-Mel equilibrium binding experiments were performed in a final volume of 200ml of ‘ligand binding buffer’ (20 mM Tris-HCl pH 7.5, 1 mM EGTA), for 2 hours at 377C. Saturation analysis was performed with 125I-Mel between 2.5 and 1280 pM. Non-specific binding was determined in the presence of 1 mM melatonin. Separation of the specific binding component was performed by PEG 8000 precipitation and centrifugation at 47C (19). Determinations of Kd values were provided by Scatchard analysis (20), and protein determinations were performed in duplicate by the method of Bradford (21). Competitive displacement experiments were performed with 100 pM 125I-Mel tracer and competing drugs in the concentration range 10013 to 1005 M. All ligand binding experiments were performed using duplicate or triplicate samples, and repeated at least 3 times. Analysis of ligand binding was performed using the GRAFIT software package (Sigma). Cyclic AMP determinations. HEK293 cells were cultured as monolayers in EMEM supplemented with 10% fetal calf serum and 1% antibiotic/antimycotic solution, in 5% CO2 at 377C. Near confluent plates were co-transfected with melatonin receptor constructs together with a construct encoding the ovine MSH receptor, using the DEAEDextran method of Wong (22). Cells were cultured for 24 hours before
419
AID
BBRC 7482
/
693b$$$221
09-22-97 21:16:03
bbrcgs
AP: BBRC
Vol. 239, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1
Ligand Binding Affinities of the FLAG-Epitope Tagged Wild-Type and Site-Directed Mutant, Ovine Mel1ab Melatonin Receptors Kd [nM] 125
Receptor Wild-type-FLAG H211F H211L V208A V208L
Ki [nM]
I-Mel
0.14 0.83 0.99 0.32 0.71
{ { { { {
0.02 0.10 0.16 0.03 0.09
(05.9) (07.1) (02.3) (05.1)
Melatonin 0.63 3.54 5.01 1.08 4.01
{ { { { {
0.05 0.63 0.63 0.18 0.67
S20098 0.98 5.00 4.58 1.53 11.9
(05.6) (07.9) (01.7) (06.4)
{ { { { {
N-NEA
0.06 1.13 (05.1) 0.69 (04.7) 0.07 (01.6) 2.1 (012.1)
86.1 25.0 5.72 146 795
{ { { { {
3.6 4.0 (/3.4) 0.98 (/15.1) 13 (01.7) 140 (09.2)
N-acetylserotonin 754 { 32 832 { 134 (01.1) 501 { 49 (/1.5) 1560 { 190 (02.2) Ç10,000 (Ç013.8)
Note. Experiments were performed as stated under Materials and Methods, Kd values were determined by Scatchard analysis (20), and Ki values determined using the Cheng-Prusoff correction (25). Mutant receptors are referred to by the abbreviations stated in Materials and Methods. All values are the mean { S.E. from 3 or 4 identical experiments. The fold difference in affinity for the mutant receptors compared to the affinity for the FLAG-epitope tagged wild-type receptor is shown in parentheses for each drug.
being re-seeded into poly-D-lysine coated 24 well tissue culture plates at 21105 cells per well. Following incubation for a further 24 hours the cells were assayed. Medium was aspirated and the cells washed once with 0.5 ml of supplement free EMEM. The wash medium was removed and 0.5 ml EMEM (containing 1 mM IBMX) and appropriate concentrations of drugs, routinely 1 mM a-NDP-MSH and melatonin
in the range 10012 to 1007 M, added for 30 min at 377C. Reactions were stopped by removal of the assay medium and addition of 0.4 ml 5% TCA solution. Cyclic AMP determinations were performed on the TCA extract by radio-immuno assay as previously described (23). Data from individual experiments were averaged from duplicate determinations of 3 separate wells treated under identical conditions.
FIG. 2. Competitive displacement of 125I-Mel binding to the FLAG-epitope tagged ovine Mel1ab melatonin receptor, and site-directed mutant H211F, expressed in COS-7 cells. Competitive binding was performed as stated in Materials and Methods. Data for the FLAGepitope tagged ovine Mel1ab melatonin receptor are represented by open circles, and data for mutant H211F by filled circles. Data shown is for a single experiment using duplicate determinations (mean { S.E.), and is representative of 3 or 4 similar experiments. This figure illustrates some of the primary data used to calculate Ki values listed in Table 1.
420
AID
BBRC 7482
/
693b$$$221
09-22-97 21:16:03
bbrcgs
AP: BBRC
Vol. 239, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
RESULTS AND DISCUSSION The melatonin receptor used in this study was the ovine Mel1ab receptor (11), which had been modified by the inclusion of a FLAG-epitope within the aminoterminal domain prior to TM1 (data not shown). When tested against an unmodified ovine Mel1ab receptor the addition of the FLAG-epitope did not alter the expression level or 125I-Mel binding affinity of transfected COS-7 cells (Fig. 1). The FLAG-epitope tagged receptor was further characterized by competitive displacement of 125I-Mel by melatonin and several related drugs, producing affinity values consistent with those previously reported for other Mel1a and Mel1b melatonin receptors (Table 1 cf. refs. 10,24). Also the FLAG-tagged receptor was fully functional, displaying a melatonin mediated inhibition of stimulated cAMP levels in transfected HEK293 cells (Fig. 3). Therefore the FLAG-epitope tagged ovine Mel1ab receptor would appear a suitable model for investigations into melatonin receptor structure and function. A series of 4 site-directed FLAG-epitope tagged mutants were constructed, V208A, V208L, H211F, and H211L (data not shown), to investigate the potential roles of valine 208 and histidine 211 in the ligand binding site of the ovine Mel1ab melatonin receptor. Expression of mutants was monitored by immunocytochemical identification of the amino-terminal FLAG epitopes. This demonstrated that all the mutant receptors were expressed in COS-7 cells at similar levels to the FLAG-epitope tagged wild-type receptor (Ç120 fmol/mg protein), and that all receptors displayed similar membrane targeting (data not shown). Determination of Kd values for 125I-Mel were performed and identified that all the mutants had lower affinities than that observed for the wild-type receptor (Table 1). This suggested that both valine 208 and histidine 211 may be involved in the ligand binding site of the melatonin receptor. Interestingly 3 mutants, V208L, H211F, and H211L, all displayed quite large reductions in Kd (5 to 7 fold), while 1 mutant, V208A, had only a modest reduction in Kd (2.3 fold). Competitive displacement of 125I-Mel by 4 drugs (melatonin, N-AS, S20098, and N-NEA) was performed on all receptors and Ki values determined (25) (Table 1). The wild-type receptor and both the V208A and V208L mutants displayed the following rank order of affinities, 125I-Mel ú melatonin Å S20098 @ N-NEA ú N-AS, which is consistent with what is observed in ovine pars tuberalis tissue (H. E. Howell, personal communication). The V208A mutant had affinities consistently Ç2 fold lower than the wild-type receptor for all drugs, whereas V208L had affinities between 5 and 12 fold lower than the wild-type. One possible explanation of why mutant V208L has greater effects on ligand binding than V208A is the size of the substi-
FIG. 3. Melatonin mediated inhibition of a-NDP-MSH stimulated cyclic AMP levels in HEK293 cells transfected with the FLAGepitope tagged wild-type, and site-directed mutant, ovine Mel1ab melatonin receptors. Experimental procedures were performed as stated in Materials and Methods. Mutant receptors are referred to by the abbreviations stated in Materials and Methods. Data shown are the mean { S.E. from two experiments involving duplicate determinations of triplicate samples. Calculated IC50 values, FLAG-epitope tagged ovine Mel1ab melatonin receptor, 553 pM; mutant V208A, 900 pM; mutant V208L, 1.68 nM. Data is represented on two panels for clarity, with the data for the FLAG-epitope tagged wild-type receptor shown on both.
tuted group. All the three amino-acids studied at position 208 (valine, alanine, and leucine) have common aliphatic chemical properties, but differ in the size of the side-chain. Leucine is one carbon group bigger than the wild-type valine residue and therefore may impinge upon the binding pocket resulting in the consistent ú5 fold reduction in affinities for all the drugs tested. Alanine is however 2 carbon groups smaller than valine and the small reductions in affinity identified by mutant V208A may reflect a less significant effect on the ligand binding site structure. Mutants H211F and H211L displayed 5 to 8 fold reductions in affinity compared to the wild-type receptor for 125IMel, melatonin and S20098, but interestingly this was not apparent for the drugs lacking a methoxy group (N-AS and N-NEA). These drugs actually bound to H211F and H211L with affinities approximately equal to, or slightly greater than, the wildtype receptor (Fig. 2). One hypothesis that would explain these data is if histidine 211 makes a direct
421
AID
BBRC 7482
/
693b$$$221
09-22-97 21:16:03
bbrcgs
AP: BBRC
Vol. 239, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Model of the ovine Mel1ab melatonin receptor. The 7 TMs of the ovine Mel1ab melatonin receptor are shown as if viewed from the cytoplasmic side of the cell membrane. The model is based on the GPCR model of the adrenergic receptor proposed by Donnelly et al. (27), and highlights valine 208 and histidine 211 on the ligand binding face of TM 5.
interaction with the methoxy groups of 125I-Mel, melatonin and S20098, causing the wild-type receptor to bind these drugs with increased affinity compared to mutants H211F and H211L. However when structurally related drugs, which do not have a methoxy group, are tested the absence of this putative interaction results in the wild-type receptor displaying similar affinities for N-NEA and N-AS as those observed for mutants H211F and H211L. Such an interaction between histidine 211 and the 5-methoxy group of melatonin supports the recent theoretical melatonin receptor model of Navajas et al. (16). Functional analysis of mutant receptors was performed in transiently transfected HEK293 cells, by measuring the melatonin mediated inhibition of aNDP-MSH stimulated cAMP levels (Fig. 3). HEK293 cells were chosen for functional studies as COS-7 cells are not suitable for analysis of GPCRs which lower cAMP levels via a Gia G protein subunit (22). The wildtype receptor and mutants V208A and V208L all inhibited cAMP levels in transfected HEK293 cells, with IC50 values in the same rank order as their respective melatonin binding affinities (Fig. 3). This confirms that valine 208 appears to be in the melatonin receptor ligand binding pocket, but is not absolutely required for receptor function. Both mutants H211F and H211L did not show any melatonin mediated effects in HEK293 cells, even at concentrations of 1005 M (data not shown), suggesting that histidine 211 may be essential for melatonin mediated receptor function. Due to the low transient expression levels in HEK293, however, we were unable to determine expression levels of the wild-type
or any of the 4 mutant receptors by either ligand binding or immunocytochemistry techniques. Nevertheless, given that all receptors were expressed equally well in COS-7 cells it would appear unlikely that the lack of an observed response of H211F and H211L in HEK293 is due to a selective lack of expression of these mutants in this cell line. The lack of a functional response by these mutants is very interesting in the light of the proposal that a methoxy group is essential for agonist induced functional responses in melatonin receptors (26). Potentially the replacement of histidine 211 mimics the effect of a drug not having a methoxy group when bound to the wild-type receptor, i.e. the absence of the normal histidine 211-methoxy group interaction results in the receptor maintaining an inactive state. As valine 208 and histidine 211 are totally conserved in all identified melatonin receptors, it is most probable that the effects identified in this study will be common to the entire melatonin receptor family. This study represents the first investigations into melatonin receptor structure and function using sitedirected mutagenesis. These data identify that the melatonin receptor has a binding site involving residues in TM5, in common with other GPCRs such as the b2adrenergic receptor. A model of the ovine Mel1ab receptor, based on the adrenergic GPCR model of Donnelly et al. (27) is shown, and highlights valine 208 and histidine 211 on the predicted ligand binding face of TM 5 (Fig. 4). The use of this model will allow further experiments to be designed, targeting other residues on other helices which may be involved in melatonin receptor ligand binding interactions.
422
AID
BBRC 7482
/
693b$$$221
09-22-97 21:16:03
bbrcgs
AP: BBRC
Vol. 239, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
ACKNOWLEDGMENTS We thank April Newman for DNA sequencing, and Diane Slater, Lynda Williams, Janice Drew, Nigel Bunnett, and Patrick Gamp for technical information and assistance with the immunocytochemistry studies. This study was supported by SOAEFD and a grant from Servier, France.
REFERENCES 1. Cassone, V. M. (1990) Trends Neurosci. 13, 457–464. 2. Williams, L. M., Lincoln, G. A., and Morgan P. J. (1995) in Ruminant Physiology: Digestion, Metabolism, Growth and Reproduction: Proceedings of the Eigth International Symposium on Ruminant Physiology (Engelhardt, W. V., Leonhard-Marck, S., Breves, G., and Giesecke D., Eds.), pp. 541–561, Ferdinand Enke Verlag, Stuttgart, Germany. 3. Reppert, S. M., Weaver, D. R., Rivkees, S. A., and Stopa, E. G. (1988) Science 242, 78–81. 4. Morgan, P. J., Williams, L. M., Davidson, G., Lawson, W., and Howell, E. (1989) J. Neuroendocrinol. 1, 1–4. 5. Morgan, P. J. and Williams, L. M. (1996) Rev. Reprod. 1, 153– 161. 6. Arendt, J., Aldhous, M., and Wright, J. (1988) Lancet 8588, 772– 773. 7. Palm, L., Blennow, G., and Wetterberg, L. (1991) Annals Neurology 29, 193–198. 8. Arendt, J., Aldhous, M., and Marks, V. (1986) Br. Med. J. 292, 1170. 9. Ebisawa, T., Karne, S., Lerner, M. R., and Reppert, S. M. (1994) Proc. Natl. Acad. Sci. USA 91, 6133–6137.
10. Reppert, S. M., Weaver, D. R., and Ebisawa, T. (1994). Neuron 13, 1–20. 11. Barrett, P., Conway, S., Jockers, R., Strosberg, A. D., GuardiolaLemaitre, B., Delagrange, P., and Morgan, P. J. (1997) Biochim. Biophys. Acta. 1356, 299–307. 12. Sugden, D., Chong, N. W. S., and Lewis, D. F. V. (1995) Br. J. Pharmacol. 114, 618–623. 13. Strader, C. D., Candelore, M. R., Hill, W. S., Sigal, I. S., and Dixon, R. A. F. (1989) J. Biol. Chem. 264, 13572–13578. 14. Ho, B. Y., Karschin, A., Branchek, T., Davidson, N., and Lester, H. A. (1992) FEBS Lett. 312, 259–262. 15. Kao, H.-T., Adham, N., Olsen, M. A., Weinshank, R. L., Branchek, T. A., and Hartig, P. R. (1992) FEBS Lett. 307, 324–328. 16. Navajas, C., Kokkola, T., Poso, A., Honka, N., Gynther, J., and Laitinen, J. T. (1996) Eur. J. Pharmacol. 304, 173–183. 17. Barrett, P., MacDonald, A., Helliwell, R., Davidson, G., and Morgan, P. J. (1994) J. Mol. Endocrinol. 12, 203–213. 18. Cullen, B. R. (1987) Methods Enzymol. 152, 684–704. 19. Conway, S., Drew, J. E., Canning, S. J., Barrett, P., Jockers, R., Strosberg, A. D., Guardiola-Lemaitre, B., Delagrange, P., and Morgan. P. J. (1997) FEBS Lett. 407, 121–126. 20. Scatchard, G. (1949) Ann. NY Acad. Sci. 51, 660–672. 21. Bradford, M. M. (1976) Anal. Biochem. 72, 248–254. 22. Wong, Y. H. (1994) Methods Enzymol. 238, 81–94. 23. Morgan, P. J., Davidson, G., Lawson, W., and Barrett, P. (1990) J. Neuroendocrinol. 2, 773–776. 24. Reppert, S. M., Weaver, D. R., and Godson, C. (1996) Trends Pharmacol. Sci. 17, 100–102. 25. Cheng, Y.-C., and Prusoff, W. H. (1973) Biochem. Pharmacol. 22, 3099–3108. 26. Heward, C. B., and Hadley, M. E. (1975) Life Sci. 17, 1167–1178. 27. Donnelly, D., Findlay, J. B. C., and Blundell, T. L. (1994) Receptors and Channels 2, 61–78.
423
AID
BBRC 7482
/
693b$$$221
09-22-97 21:16:03
bbrcgs
AP: BBRC
239, 418–423 (1997)
RC977482
The Roles of Valine 208 and Histidine 211 in Ligand Binding and Receptor Function of the Ovine Mel1ab Melatonin Receptor Shaun Conway,*,1 Sarah J. Canning,* Perry Barrett,* Beatrice Guardiola-Lemaitre,† Phillipe Delagrange,† and Peter J. Morgan* *Molecular Neuroendocrinology Unit, Rowett Research Institute, Greenburn Road, Bucksburn, Aberdeen, Scotland, United Kingdom AB21 9SB; and †Institut de recherches Internationales Servier, 6 place de Ple´iades, F-92415 Courbevoie Cedex, France
Received September 8, 1997
Site-directed mutagenesis was used to study two residues, valine 208 and histidine 211, in transmembrane domain 5 of the ovine Mel1ab melatonin receptor. A series of 4 mutants were constructed (V208A, V208L, H211F, H211L), and each engineered to contain a FLAG-epitope. Immunocytochemistry demonstrated that all the mutants were expressed in COS-7 cells at levels comparable to the FLAG-epitope tagged wildtype Mel1ab receptor (Ç120 fmol/mg protein). Ligand binding revealed however that all mutants had reduced affinities for 2-[125I]-iodomelatonin (Kd wildtype 139 pM, Kd mutants 320 to 989 pM). Competition studies, with a series of melatonin analogues, identified a probable interaction between histidine 211 and the 5-methoxy group of melatonin. The wild-type receptor and both valine 208 mutants displayed a dosedependent melatonin mediated inhibition of cyclic AMP levels in HEK293 cells, with IC50 values in the same rank-order as their melatonin binding affinities. Both H211F and H211L, however, did not display any melatonin mediated effects and may suggest that histidine 211 is critical for melatonin mediated receptor activation. q 1997 Academic Press
Melatonin synchronizes circadian rhythms in mammals (1), and regulates aspects of reproductive state 1 Corresponding author. Fax: /44 (0)1224 716653. E-mail: sco@ rri.sari.ac.uk. Abbreviations: 125I-Mel, 2-[125I]-iodomelatonin; GPCR(s), G protein-coupled receptor(s); TM, transmembrane segment; 5-HT, 5-hydroxytryptamine; N-AS, N-acetylserotonin; MSH, melanocyte stimulating hormone, DEAE-dextran, diethylaminoethyl-dextran; PBS, phosphate buffered saline (pH 7.4); EGTA, ethylene glycol-bis(b-aminoethyl ether) N,N,N*,N*-tetraacetic acid; DMEM, Dulbecco’s modified Eagle’s medium; EMEM, Eagle’s minimum essential medium; PEG 8000, polyethylene glycol (average molecular weight 8,000); IBMX, 3-isobutyl-1-methylxanthine; a-NDP-MSH, [Nle4,D-Phe7]-amelanocyte stimulating hormone; TCA, trichloroacetic acid.
0006-291X/97 $25.00
in seasonally breeding mammals (2). These effects are achieved via the binding of melatonin to high-affinity cell surface melatonin receptors (3,4,5). Despite our insight into the possible functions of melatonin receptors little is known about the molecular structure of the receptor or receptor-ligand interactions, information that could lead to the design of specific agonist and antagonist modulators of the melatonin signal. The clinical application of such drugs could prove beneficial as administered melatonin has already been shown to modulate the sleep-wake cycle in the blind (6,7) and provides an effective treatment for jet-lag (8). The recent cloning of a series of high-affinity melatonin receptors has identified them as members of the GPCR ‘superfamily’ (9,10,11). It has been proposed that melatonin receptors will have a heptahelical architecture containing an intrahelical ligand binding site, in common with other GPCRs (9,12). One feature of many GPCRs is that the general structure of the ligand binding site appears to remain essentially conserved. This is illustrated by the alignment of residues that have been shown to provide ligand binding interactions in diverse GPCRs. For example two TM5 serine residues in the b2-adrenergic have been proven to interact with the two hydroxyl groups on the aromatic ring of adrenaline (13). Residues which protein sequence align with these serine residues, placing them in the same presumed position within each receptor, have also been shown to interact with the substituted ring of 5-HT in certain 5-HT receptors (14,15). The two residues which are in analogous TM5 positions in melatonin receptors are a conserved valine and a conserved histidine (valine 208 and histidine 211, ovine Mel1a receptor). At present the structure of the melatonin binding site is unknown, but if it is essentially conserved with those of adrenergic and 5-HT receptors then these two residues may also be involved in ligand binding, potentially interacting
418
Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 7482
/
693b$$$221
09-22-97 21:16:03
bbrcgs
AP: BBRC
Vol. 239, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS studies the ovine Mel1ab receptor cDNA was ligated into the vector pBK-CMV (Stratagene) and the ovine MSH receptor cDNA into vector pcDNA I (Invitrogen). COS-7 cells and HEK293 cells were provided by ECACC. 125I-Mel (2200 Ci/mmol) was purchased from NEN DuPont. The drugs S20098 and N-NEA (N-[2-(1-naphthyl) ethyl] acetamide) were synthesised by Servier (see Fig. 2 for structures). Other general reagents were purchased from Sigma or Life Technologies Inc. Site-directed mutagenesis. Mutations were introduced into the ovine Mel1ab melatonin receptor in pBK-CMV using the Chameleon site-directed mutagenesis kit (Stratagene). Successful construction of each mutant was confirmed by DNA sequencing using the Prism dyedeoxy terminator kit (Applied Biosystems Inc.). In this study 4 mutants were generated, mutants V208A, V208L, H211F, and H211L. FLAG epitope tagging. A PCR based protocol was used to introduce the FLAG-epitope (DYKDDDDK), encoded by the nucleotide sequence 5*»GACTACAAGGACGACGACGATAAG…3 *, between codon 1 and codon 2 of the ovine Mel1ab receptor cDNA. FLAG-epitope tagged mutant receptors were generated by subcloning fragments containing the desired mutations into the FLAG-epitope tagged wild-type receptor construct. All constructs were confirmed by DNA sequencing. COS-7 cell tissue culture. COS-7 cells were grown as monolayers in DMEM supplemented with 10% fetal calf serum and 1% antibiotic/ antimycotic solution, in 5% CO2 at 377C. Confluent plates were reseeded at 15,000 cells/cm2 and used 24 hours later for transfection. Cells were transfected with receptor constructs using the DEAEDextran method of Cullen (18) and grown for 72 hours. Growth media was aspirated and cells washed twice with PBS before being harvested. Cells were pelleted in 1.5 ml microcentrifuge tubes (13,000 g, 1 min, 47C), and stored frozen at 0807C.
FIG. 1. 125I-Mel saturation binding. (A) Specific binding of 125I-Mel to COS-7 cells transfected with the ovine Mel1ab melatonin receptor, and (B) specific binding of 125I-Mel to COS-7 cells transfected with the FLAG-epitope tagged ovine Mel1ab melatonin receptor. Data shown are from a single experiment using mean values of triplicate determinations, and are representative of 3 similar experiments. Binding parameters calculated by Scatchard analysis (20) were, ovine Mel1ab melatonin receptor Kd 93.2 { 11.3 pM, Bmax 109.3 { 8.7 fmol/mg protein; FLAG-epitope tagged ovine Mel1ab melatonin receptor, Kd 138.9 { 16.7 pM, Bmax 121.8 { 16.8 fmol/mg protein (mean { S.E., nÅ3).
Immunocytochemistry. COS-7 cells were transfected as above and incubated for 24 hours. Cells were trypsinized and seeded onto poly-D-lysine coated glass Lab-Tek chamber-slides (Nunc), at 125,000 cells per chamber, and cultured for a further 24 hours. Slides were washed twice in PBS, fixed with 4% para-formaldehyde in PBS (47C, 20 min), then thoroughly washed with 3 changes of PBS followed by 3 serial 5 minute washes in antibody binding buffer (1% horse serum, 0.1% saponin in PBS). Fixed cells were incubated with M5 antiFLAG antibody (IBI, Kodak), at a concentration of 5mg/ml in antibody binding buffer (16 hours, 47C). Samples were washed 3 times in antibody binding buffer, and incubated for 2 hours at room temperature with a 1:200 dilution of fluorescein conjugated horse anti-mouse IgG antibody in PBS. Finally cells were washed in PBS and studied by immunofluorescence microscopy. 125
with the 5-methoxy substituted ring of melatonin. Evidence in support of such interactions was recently provided by a computationally derived model of the melatonin receptor, based on the structure of the visual GPCR rhodopsin (16). This theoretical model proposes a role for the histidine in interactions with the oxygen of the 5-methoxy group of melatonin, and a role to valine 208 in providing a aliphatic milieu for the methyl group. In this study we have used site-directed mutagenesis to investigate the requirement of the ovine Mel1ab melatonin receptor for valine 208 and histidine 211. This study represents the first such investigations into the structure/ function relationships of melatonin receptors. MATERIALS AND METHODS Materials. The cloning of the ovine Mel1ab receptor and the ovine MSH receptor have been previously reported (11,17). For expression
I-Mel equilibrium binding studies. 125I-Mel equilibrium binding experiments were performed in a final volume of 200ml of ‘ligand binding buffer’ (20 mM Tris-HCl pH 7.5, 1 mM EGTA), for 2 hours at 377C. Saturation analysis was performed with 125I-Mel between 2.5 and 1280 pM. Non-specific binding was determined in the presence of 1 mM melatonin. Separation of the specific binding component was performed by PEG 8000 precipitation and centrifugation at 47C (19). Determinations of Kd values were provided by Scatchard analysis (20), and protein determinations were performed in duplicate by the method of Bradford (21). Competitive displacement experiments were performed with 100 pM 125I-Mel tracer and competing drugs in the concentration range 10013 to 1005 M. All ligand binding experiments were performed using duplicate or triplicate samples, and repeated at least 3 times. Analysis of ligand binding was performed using the GRAFIT software package (Sigma). Cyclic AMP determinations. HEK293 cells were cultured as monolayers in EMEM supplemented with 10% fetal calf serum and 1% antibiotic/antimycotic solution, in 5% CO2 at 377C. Near confluent plates were co-transfected with melatonin receptor constructs together with a construct encoding the ovine MSH receptor, using the DEAEDextran method of Wong (22). Cells were cultured for 24 hours before
419
AID
BBRC 7482
/
693b$$$221
09-22-97 21:16:03
bbrcgs
AP: BBRC
Vol. 239, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1
Ligand Binding Affinities of the FLAG-Epitope Tagged Wild-Type and Site-Directed Mutant, Ovine Mel1ab Melatonin Receptors Kd [nM] 125
Receptor Wild-type-FLAG H211F H211L V208A V208L
Ki [nM]
I-Mel
0.14 0.83 0.99 0.32 0.71
{ { { { {
0.02 0.10 0.16 0.03 0.09
(05.9) (07.1) (02.3) (05.1)
Melatonin 0.63 3.54 5.01 1.08 4.01
{ { { { {
0.05 0.63 0.63 0.18 0.67
S20098 0.98 5.00 4.58 1.53 11.9
(05.6) (07.9) (01.7) (06.4)
{ { { { {
N-NEA
0.06 1.13 (05.1) 0.69 (04.7) 0.07 (01.6) 2.1 (012.1)
86.1 25.0 5.72 146 795
{ { { { {
3.6 4.0 (/3.4) 0.98 (/15.1) 13 (01.7) 140 (09.2)
N-acetylserotonin 754 { 32 832 { 134 (01.1) 501 { 49 (/1.5) 1560 { 190 (02.2) Ç10,000 (Ç013.8)
Note. Experiments were performed as stated under Materials and Methods, Kd values were determined by Scatchard analysis (20), and Ki values determined using the Cheng-Prusoff correction (25). Mutant receptors are referred to by the abbreviations stated in Materials and Methods. All values are the mean { S.E. from 3 or 4 identical experiments. The fold difference in affinity for the mutant receptors compared to the affinity for the FLAG-epitope tagged wild-type receptor is shown in parentheses for each drug.
being re-seeded into poly-D-lysine coated 24 well tissue culture plates at 21105 cells per well. Following incubation for a further 24 hours the cells were assayed. Medium was aspirated and the cells washed once with 0.5 ml of supplement free EMEM. The wash medium was removed and 0.5 ml EMEM (containing 1 mM IBMX) and appropriate concentrations of drugs, routinely 1 mM a-NDP-MSH and melatonin
in the range 10012 to 1007 M, added for 30 min at 377C. Reactions were stopped by removal of the assay medium and addition of 0.4 ml 5% TCA solution. Cyclic AMP determinations were performed on the TCA extract by radio-immuno assay as previously described (23). Data from individual experiments were averaged from duplicate determinations of 3 separate wells treated under identical conditions.
FIG. 2. Competitive displacement of 125I-Mel binding to the FLAG-epitope tagged ovine Mel1ab melatonin receptor, and site-directed mutant H211F, expressed in COS-7 cells. Competitive binding was performed as stated in Materials and Methods. Data for the FLAGepitope tagged ovine Mel1ab melatonin receptor are represented by open circles, and data for mutant H211F by filled circles. Data shown is for a single experiment using duplicate determinations (mean { S.E.), and is representative of 3 or 4 similar experiments. This figure illustrates some of the primary data used to calculate Ki values listed in Table 1.
420
AID
BBRC 7482
/
693b$$$221
09-22-97 21:16:03
bbrcgs
AP: BBRC
Vol. 239, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
RESULTS AND DISCUSSION The melatonin receptor used in this study was the ovine Mel1ab receptor (11), which had been modified by the inclusion of a FLAG-epitope within the aminoterminal domain prior to TM1 (data not shown). When tested against an unmodified ovine Mel1ab receptor the addition of the FLAG-epitope did not alter the expression level or 125I-Mel binding affinity of transfected COS-7 cells (Fig. 1). The FLAG-epitope tagged receptor was further characterized by competitive displacement of 125I-Mel by melatonin and several related drugs, producing affinity values consistent with those previously reported for other Mel1a and Mel1b melatonin receptors (Table 1 cf. refs. 10,24). Also the FLAG-tagged receptor was fully functional, displaying a melatonin mediated inhibition of stimulated cAMP levels in transfected HEK293 cells (Fig. 3). Therefore the FLAG-epitope tagged ovine Mel1ab receptor would appear a suitable model for investigations into melatonin receptor structure and function. A series of 4 site-directed FLAG-epitope tagged mutants were constructed, V208A, V208L, H211F, and H211L (data not shown), to investigate the potential roles of valine 208 and histidine 211 in the ligand binding site of the ovine Mel1ab melatonin receptor. Expression of mutants was monitored by immunocytochemical identification of the amino-terminal FLAG epitopes. This demonstrated that all the mutant receptors were expressed in COS-7 cells at similar levels to the FLAG-epitope tagged wild-type receptor (Ç120 fmol/mg protein), and that all receptors displayed similar membrane targeting (data not shown). Determination of Kd values for 125I-Mel were performed and identified that all the mutants had lower affinities than that observed for the wild-type receptor (Table 1). This suggested that both valine 208 and histidine 211 may be involved in the ligand binding site of the melatonin receptor. Interestingly 3 mutants, V208L, H211F, and H211L, all displayed quite large reductions in Kd (5 to 7 fold), while 1 mutant, V208A, had only a modest reduction in Kd (2.3 fold). Competitive displacement of 125I-Mel by 4 drugs (melatonin, N-AS, S20098, and N-NEA) was performed on all receptors and Ki values determined (25) (Table 1). The wild-type receptor and both the V208A and V208L mutants displayed the following rank order of affinities, 125I-Mel ú melatonin Å S20098 @ N-NEA ú N-AS, which is consistent with what is observed in ovine pars tuberalis tissue (H. E. Howell, personal communication). The V208A mutant had affinities consistently Ç2 fold lower than the wild-type receptor for all drugs, whereas V208L had affinities between 5 and 12 fold lower than the wild-type. One possible explanation of why mutant V208L has greater effects on ligand binding than V208A is the size of the substi-
FIG. 3. Melatonin mediated inhibition of a-NDP-MSH stimulated cyclic AMP levels in HEK293 cells transfected with the FLAGepitope tagged wild-type, and site-directed mutant, ovine Mel1ab melatonin receptors. Experimental procedures were performed as stated in Materials and Methods. Mutant receptors are referred to by the abbreviations stated in Materials and Methods. Data shown are the mean { S.E. from two experiments involving duplicate determinations of triplicate samples. Calculated IC50 values, FLAG-epitope tagged ovine Mel1ab melatonin receptor, 553 pM; mutant V208A, 900 pM; mutant V208L, 1.68 nM. Data is represented on two panels for clarity, with the data for the FLAG-epitope tagged wild-type receptor shown on both.
tuted group. All the three amino-acids studied at position 208 (valine, alanine, and leucine) have common aliphatic chemical properties, but differ in the size of the side-chain. Leucine is one carbon group bigger than the wild-type valine residue and therefore may impinge upon the binding pocket resulting in the consistent ú5 fold reduction in affinities for all the drugs tested. Alanine is however 2 carbon groups smaller than valine and the small reductions in affinity identified by mutant V208A may reflect a less significant effect on the ligand binding site structure. Mutants H211F and H211L displayed 5 to 8 fold reductions in affinity compared to the wild-type receptor for 125IMel, melatonin and S20098, but interestingly this was not apparent for the drugs lacking a methoxy group (N-AS and N-NEA). These drugs actually bound to H211F and H211L with affinities approximately equal to, or slightly greater than, the wildtype receptor (Fig. 2). One hypothesis that would explain these data is if histidine 211 makes a direct
421
AID
BBRC 7482
/
693b$$$221
09-22-97 21:16:03
bbrcgs
AP: BBRC
Vol. 239, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Model of the ovine Mel1ab melatonin receptor. The 7 TMs of the ovine Mel1ab melatonin receptor are shown as if viewed from the cytoplasmic side of the cell membrane. The model is based on the GPCR model of the adrenergic receptor proposed by Donnelly et al. (27), and highlights valine 208 and histidine 211 on the ligand binding face of TM 5.
interaction with the methoxy groups of 125I-Mel, melatonin and S20098, causing the wild-type receptor to bind these drugs with increased affinity compared to mutants H211F and H211L. However when structurally related drugs, which do not have a methoxy group, are tested the absence of this putative interaction results in the wild-type receptor displaying similar affinities for N-NEA and N-AS as those observed for mutants H211F and H211L. Such an interaction between histidine 211 and the 5-methoxy group of melatonin supports the recent theoretical melatonin receptor model of Navajas et al. (16). Functional analysis of mutant receptors was performed in transiently transfected HEK293 cells, by measuring the melatonin mediated inhibition of aNDP-MSH stimulated cAMP levels (Fig. 3). HEK293 cells were chosen for functional studies as COS-7 cells are not suitable for analysis of GPCRs which lower cAMP levels via a Gia G protein subunit (22). The wildtype receptor and mutants V208A and V208L all inhibited cAMP levels in transfected HEK293 cells, with IC50 values in the same rank order as their respective melatonin binding affinities (Fig. 3). This confirms that valine 208 appears to be in the melatonin receptor ligand binding pocket, but is not absolutely required for receptor function. Both mutants H211F and H211L did not show any melatonin mediated effects in HEK293 cells, even at concentrations of 1005 M (data not shown), suggesting that histidine 211 may be essential for melatonin mediated receptor function. Due to the low transient expression levels in HEK293, however, we were unable to determine expression levels of the wild-type
or any of the 4 mutant receptors by either ligand binding or immunocytochemistry techniques. Nevertheless, given that all receptors were expressed equally well in COS-7 cells it would appear unlikely that the lack of an observed response of H211F and H211L in HEK293 is due to a selective lack of expression of these mutants in this cell line. The lack of a functional response by these mutants is very interesting in the light of the proposal that a methoxy group is essential for agonist induced functional responses in melatonin receptors (26). Potentially the replacement of histidine 211 mimics the effect of a drug not having a methoxy group when bound to the wild-type receptor, i.e. the absence of the normal histidine 211-methoxy group interaction results in the receptor maintaining an inactive state. As valine 208 and histidine 211 are totally conserved in all identified melatonin receptors, it is most probable that the effects identified in this study will be common to the entire melatonin receptor family. This study represents the first investigations into melatonin receptor structure and function using sitedirected mutagenesis. These data identify that the melatonin receptor has a binding site involving residues in TM5, in common with other GPCRs such as the b2adrenergic receptor. A model of the ovine Mel1ab receptor, based on the adrenergic GPCR model of Donnelly et al. (27) is shown, and highlights valine 208 and histidine 211 on the predicted ligand binding face of TM 5 (Fig. 4). The use of this model will allow further experiments to be designed, targeting other residues on other helices which may be involved in melatonin receptor ligand binding interactions.
422
AID
BBRC 7482
/
693b$$$221
09-22-97 21:16:03
bbrcgs
AP: BBRC
Vol. 239, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
ACKNOWLEDGMENTS We thank April Newman for DNA sequencing, and Diane Slater, Lynda Williams, Janice Drew, Nigel Bunnett, and Patrick Gamp for technical information and assistance with the immunocytochemistry studies. This study was supported by SOAEFD and a grant from Servier, France.
REFERENCES 1. Cassone, V. M. (1990) Trends Neurosci. 13, 457–464. 2. Williams, L. M., Lincoln, G. A., and Morgan P. J. (1995) in Ruminant Physiology: Digestion, Metabolism, Growth and Reproduction: Proceedings of the Eigth International Symposium on Ruminant Physiology (Engelhardt, W. V., Leonhard-Marck, S., Breves, G., and Giesecke D., Eds.), pp. 541–561, Ferdinand Enke Verlag, Stuttgart, Germany. 3. Reppert, S. M., Weaver, D. R., Rivkees, S. A., and Stopa, E. G. (1988) Science 242, 78–81. 4. Morgan, P. J., Williams, L. M., Davidson, G., Lawson, W., and Howell, E. (1989) J. Neuroendocrinol. 1, 1–4. 5. Morgan, P. J. and Williams, L. M. (1996) Rev. Reprod. 1, 153– 161. 6. Arendt, J., Aldhous, M., and Wright, J. (1988) Lancet 8588, 772– 773. 7. Palm, L., Blennow, G., and Wetterberg, L. (1991) Annals Neurology 29, 193–198. 8. Arendt, J., Aldhous, M., and Marks, V. (1986) Br. Med. J. 292, 1170. 9. Ebisawa, T., Karne, S., Lerner, M. R., and Reppert, S. M. (1994) Proc. Natl. Acad. Sci. USA 91, 6133–6137.
10. Reppert, S. M., Weaver, D. R., and Ebisawa, T. (1994). Neuron 13, 1–20. 11. Barrett, P., Conway, S., Jockers, R., Strosberg, A. D., GuardiolaLemaitre, B., Delagrange, P., and Morgan, P. J. (1997) Biochim. Biophys. Acta. 1356, 299–307. 12. Sugden, D., Chong, N. W. S., and Lewis, D. F. V. (1995) Br. J. Pharmacol. 114, 618–623. 13. Strader, C. D., Candelore, M. R., Hill, W. S., Sigal, I. S., and Dixon, R. A. F. (1989) J. Biol. Chem. 264, 13572–13578. 14. Ho, B. Y., Karschin, A., Branchek, T., Davidson, N., and Lester, H. A. (1992) FEBS Lett. 312, 259–262. 15. Kao, H.-T., Adham, N., Olsen, M. A., Weinshank, R. L., Branchek, T. A., and Hartig, P. R. (1992) FEBS Lett. 307, 324–328. 16. Navajas, C., Kokkola, T., Poso, A., Honka, N., Gynther, J., and Laitinen, J. T. (1996) Eur. J. Pharmacol. 304, 173–183. 17. Barrett, P., MacDonald, A., Helliwell, R., Davidson, G., and Morgan, P. J. (1994) J. Mol. Endocrinol. 12, 203–213. 18. Cullen, B. R. (1987) Methods Enzymol. 152, 684–704. 19. Conway, S., Drew, J. E., Canning, S. J., Barrett, P., Jockers, R., Strosberg, A. D., Guardiola-Lemaitre, B., Delagrange, P., and Morgan. P. J. (1997) FEBS Lett. 407, 121–126. 20. Scatchard, G. (1949) Ann. NY Acad. Sci. 51, 660–672. 21. Bradford, M. M. (1976) Anal. Biochem. 72, 248–254. 22. Wong, Y. H. (1994) Methods Enzymol. 238, 81–94. 23. Morgan, P. J., Davidson, G., Lawson, W., and Barrett, P. (1990) J. Neuroendocrinol. 2, 773–776. 24. Reppert, S. M., Weaver, D. R., and Godson, C. (1996) Trends Pharmacol. Sci. 17, 100–102. 25. Cheng, Y.-C., and Prusoff, W. H. (1973) Biochem. Pharmacol. 22, 3099–3108. 26. Heward, C. B., and Hadley, M. E. (1975) Life Sci. 17, 1167–1178. 27. Donnelly, D., Findlay, J. B. C., and Blundell, T. L. (1994) Receptors and Channels 2, 61–78.
423
AID
BBRC 7482
/
693b$$$221
09-22-97 21:16:03
bbrcgs
AP: BBRC