A three-dimensional model for the transmembrane domains of human endothelin-A receptor was built .... Endothelin- I was purchased from Novabiochem. AG.
Eur. J. Biochem. 231, 266-270 (1995) 0 FEBS 1995
Separable binding sites for the natural agonist endothelin-1 and the non-peptide antagonist bosentan on human endothelin-A receptors Volker BREU', Kento HASHIDO', Clemens BROGER', Chikara MIYAMOTO*, Yasuhiro FURUICHI*, Ashley HAYES', Barbara KALINA', Bernd-Michael LOFFLER', Henri RAMUZ' and Martine CLOZEL'
' Pharma Division, Reclinical Research, F. Hoffmann-La Roche Ltd., Basel, Switzerland ' Department of Molecular Genetics, Nippon Roche Research Center, Kamakura, Japan (Received 5 April 1995)
-
EJB 95 0549/3
A three-dimensional model for the transmembrane domains of human endothelin-A receptor was built using structural information from bacteriorhodopsin and sequence alignment to other guanine-nucleotidebinding regulatory((;) protein-coupled receptors. Based on this model, 18 amino acids located at the inside of the receptor were mutated and analyzed for binding of the natural ligand endothelin-I and bosentan, a recently described potent orally active endothelin antagonist [Clozel, M., Breu, V., Gray, G., Kalina, B., Loffler, B.-M., Bum, K., Cassal, J.-M., Hirth, G., Muller, M., Neidhart, W. & Ramuz, H. (1994) Pharmacological characterization of bosentan, a new potent orally active nonpeptide endothelin receptor antagonist, J. Pharmacol. Exp. Thel: 270, 228-2351. Mutation of Gly97, Lysl40, Lys159, Gln165 and Phe315, located in transmembrane region 1, 2, 3, 3, and 6, respectively, caused reduced specific binding of 1Z51-labelledendothelin-I , despite an expression level similar to wild-type endothelinA receptor. Mutation of Tyr263, Arg326 and Asp351 preserved endothelin-1 binding but caused reduced binding of bosentan. These amino acids, located on transmembrane regions 5, 6 and 7, respectively, are conserved among endothelin-A and endothelin-B receptors but not in other G-protein-coupled receptors. These observations demonstrate a dissociation of the binding site for the peptidic natural agonist endothelin-1 and the synthetic non-peptide antagonist bosentan. They provide the molecular basis for bosentan being a specific antagonist for both, endothelin-A as well as endothelin-B receptors and may in combination with studies on structure/activity relationship support the design of novel and more potent endothelin receptor antagonists. Keywords. Endothelin ; endothelin-A receptor; agonist ; antagonist ; binding-site.
Endothelin receptors belong to the family of guanine-nucleotide-binding-regulatory(G)-protein-coupled receptors characterized by seven hydrophobic transmembrane domains [l]. Their activation is suspected to play a role in a variety of diseases associated with focal vasoconstriction [2] as well as in chronic diseases, like hypertension [3] and atherosclerosis [4]. Thus, antagonism of endothelin receptors might offer a new therapeutic concept. Four distinct endothelin receptor subtypes, called endothelin-A [5], endothelin-B [6], endothelin-C [7] and endothelinAX [8] receptor have been cloned. However, only endothelin-A and endothelin-B receptors have been found in mammalian species. The endothelin-A receptor which selectively binds endothelin-1 and endothelin-2 mediates a major part of the vasoconstriction induced by endothelin-1 [9]. The endothelin-B subtype which is unselective for the different endothelin isoforms mediates endothelium-dependent vasodilatation [lo], but its stimulation by selective agonists can also induce vasoconstriction [ l l , 121. Recently, we have described Ro 46-2005, the first orally active endothelin antagonist [2, 131 and bosentan (Ro 47-0203) Correspondence to V. Breu, Pharma Division, Preclinical Research, F. Hoffmann-La Roche Ltd., Grenzacherstrasse 124, 70/436, CH-4002 Basel, Switzerland F a : +41 61 688 2438. Abbreviations. DMEM, Dulbecco's modified Eagle's medium; Me,SO, dimethysulfoxide; TM, transmembrane segment ; G protein, guanine-nucleotide-bindingregulatory protein.
[I41 a structurally optimized, more potent non-peptide mixed antagonist of endothelin-A and endothelin-B receptors. Both are important tools to study the pathophysiological role of endothelins. However, although a first study was performed on the binding mode of bosentan using chimeric receptors [15], the precise nature of the interaction with the human endothelin-A receptor remains to be established. Recent studies on chimeric endothelin receptors allowed identification of specific domains of the human endothelin-A receptor required for agonist binding and ligand selectivity [16191 as well as for signal transduction [20]. Using site-directed mutagenesis it was demonstrated that Tyr129, located in the second transmembrane domain plays an important role in the subtype-selective binding of ligands [21, 221. Based on the observation that there are at least two different ligand-interaction domains on the endothelin receptors [18] and comparable studies from other G-protein-coupled receptors [23], we speculated that the antagonist binding site of non-peptide antagonists like bosentan might be distinct from the interaction site of the natural agonist endothelin-1. Thus, we performed binding studies on recombinant human endothelin-A receptors with mutations of single amino acids in order to gain a more detailed picture of the antagonist-binding site. The individual mutations (Fig. 1) were selected using a three-dimensional model for the endothelin-A receptor which was built on the basis of recently published three-dimensional models for G-proteincoupled receptors [24-261.
Breu et al. (Eur: J. Biochem. 231)
267
Fig. 1. Schematic model of the human endothelin-A receptor showing putative transmembrane helices and performed amino acid exchanges. Amino acid numbering starts at the initiator methionine. The first 20 amino acids were removed by signal peptidase. N-linked glycosylation sites are indicated by arrows.
MATERIALS AND METHODS Reagents. Endothelin- I was purchased from Novabiochem AG. Bosentan (Ro 47-0203), [3H]bosentan (85 Cilmmol), BQ 123 (185 Cilmmol) and monoclonal mouse anti-(endothelin-A receptor) IgG (AH3A8) were provided by Prof. H. Ramuz, Dr Huguenin, Mr A. Trzeciak and Dr T. Watanabe, respectively, from F. Hoffmann-La Roche Ltd. 1251-Endothelin-l(specific activity, 2200 Cilmmol) and I3H1BQ 123 (185 Cilmmol) as well as '251-labelledsheep anti-mouse IgG were obtained from Anawa. Expression plasmids. The plasmid harboring the human endothelin-A receptor cDNA [20] was transfected to Escherichia coli CJ-236, and the recombinant phage was produced by the aid of a helper phage M13-K07. The mutation of endothelin-A receptor gene was performed by the Kunkel procedure using single-stranded phagemid DNA as a template [27]. Sequences of the mutated endothelin-A receptor genes were confirmed by DNA sequencing. Cells and transfection. COS-1 cells were seeded on culture dishes (60-mm diameter) at a density of approximately 90%. The tissue culture medium from the overnight cultures was removed and the cell monolayer was washed once with NaCIP, (0.2 g KCl, 0.2 g KH,PO,, 8 g NaCl and 1.5 g Na,HPO, dissolved in 1 1 H20, pH adjusted to 7.2). After addition of 1.25 ml Dulbecco's Modified Eagle's Medium (DMEM) without fetal calf serum, the dishes were incubated at 37 "C until required. 300 pl transfection mixture (530 p1 DMEM without fetal calf serum, 50 pl of DEAE-Dextran at 50 mg/ml and 20 p1 (8 pg) plasmid DNA) was added dropwise onto each dish. After 30 min at 37"C, 1.5 ml DMEM containing 200 pM chloroquine was added to each dish and incubated for an additional 90 min at 37°C. The transfection mixture was aspirated, 2.5 ml DMEM containing 10% dimethylsulfoxide (Me,SO) was added and the cells were incubated at 37°C for 3 min. Finally, the Me,SO medium was removed and replaced with 5 ml DMEM, 5% fetal calf serum supplemented with antibiotics. The cells were then incubated for 48 h before harvesting.
Membrane preparation and binding assays. Preparation of microsomal membranes and binding assays were performed as described previously [131. Competition binding experiments were performed in 250 p1 50 mM Tris buffer [pH 7.4, 25 mM MnCl,, 1 mM EDTA, 0.5% (mass/vol.) bovine serum albumin] in the presence of 10-25 pg protein and either 30000 cpm (32 pM) 1251-endothelin-l or 100000 cpm (2.1 nM) [3H]bosentan and varying amounts of unlabelled ligand. After incubation for 2 h at 22"C, bound and free ligand were separated by filtration. Non-specific binding was assessed in the presence of 100 nM unlabelled endothelin-I. Specific binding was defined as the difference between total binding and non-specific binding. Decrease of specific binding caused by the individual mutations was expressed as a percentage of the binding observed with the wild-type endothelin-A receptor. Untransfected COS-1 cells served as negative controls. KD and B,,, values were calculated from competition binding curves by direct fit analysis with the Ligand program [28]. Immunological detection. Transfected cells were harvested, resuspended in sample buffer, sonicated for 10 min and analyzed by polyacrylamide-gel electrophoresis. After electroblotting, the human endothelin-A receptors were detected by immunostaining using the monoclonal antibody AH3A8 and the immunoblot analysis kit ABC-AP (Vector Laboratories). Molecular modeling. A three-dimensional model of the transmembrane domain of the endothelin-A receptor was built by using the structure of bacteriorhodopsin [29] as a scaffold and the sequence alignment between bacteriorhodopsin and the G-protein-coupled receptors of Lewell [24]. Both sequence alignments published by Hibert's group [25, 261 differ from the used alignment in three helices. The amino acids of bacteriorhodopsin were exchanged into those of the endothelin-A receptor by using the Ca facilities of the modeling package Moloc [30]. Helix 5 was rotated by 60 degrees clockwise when viewed from the extracellular side. The model was energy optimized by using the force field of Moloc [31] in order to remove clashes between
268
Breu et al. (Eur: J. Biochem. 231)
Fig.2. Influence of point mutations on the specific binding of '*'Iendothelin-1 (black bars) and [3H]bosentan (grey bars) to membranes of COS-1 cells expressing human endothelin-A receptor. Decrease of specific binding caused by the individual mutations was expressed as a percentage of the specific binding observed with the wildtype endothe1in-A receptor, using exactly the same protein and tracer concentration. Untransfected COS-1 cells served as negative controls. Values are given as mean?SD of at least three different experiments performed in triplicates.
side chains keeping the Ca positions fixed. This model was then used to guide point-mutation experiments.
RESULTS AND DISCUSSION Competition binding experiments with 1251-endothelin-land unlabelled endothelin- 1 on recombinant wild-type endothelin-A receptor expressed transiently in COS-1 cells revealed a single class of binding sites with a KD of 75 pM for endothelin-I and a receptor density of 590 fmoVmg protein. This receptor density was sufficiently high to observe specific binding of 3H-labeled bosentan and BQ 123. In competition binding experiments with [3H]bosentan on recombinant wild-type endothelin-A receptor, bosentan exhibited a K,, of 73 nM and a maximal binding capacity similar to that of 1251-endothelin-1. The maximal specific binding of '251-endothelin-l and ['Hlbosentan was measured for each mutated endothelin-A receptor and compared to the corresponding results for the wildtype endothelin-A receptor (Fig. 2). Most mutations did not affect '251-endothelin-l binding. However, mutations G97A (transmembrane segment 1, TM l ) , K140I (TM 2), K159Q (TM 3), Q165D (TM 3) and F315L (TM 6) induced a reduction in specific binding of '251-endothelin-1to 0, 67, 54, 9 and 52%, respectively. Immunostaining proved that this lowered endothelin1 binding was not due to lower expression rates (Fig. 3). The receptors with mutations G97A, K159Q, Q165D and F315L also displayed very low ['Hlbosentan binding of 5, 0.5, 0 and 2%, respectively, indicating that the ligand binding site has been largely altered. The observation that mutation Y129F did affect neither endothelin-I nor bosentan binding is consistent with different studies where an Y129A mutation was introduced [21]. Most notably however, we observed three additional mutations which preserved binding of '251-endothelin-l but induced a partial or even total loss of the binding capacity for [3H]bosentan (Fig. 2): Y263F (TM 5 ) , 25%; R326Q (TM 6), 1 % and D351N (TM 7), 1.5%. This lower specific binding was associated with a lower potency of bosentan in competition binding experiments with '251-endothelin-l. The IC,, value was raised
Fig. 3. Immunoblot detection of the wild-type and mutated endothelin-A receptor. During separation by polyacrylamide-gel electrophoresis human endothelin-A receptor migrated at apparent molecular masses of 42, 45, 48 and 51 kDa. Removal of glycosylation sites by mutation of Asn29 and Am62 with Ala identified the bands at 45, 48 and 51 kDa as glycosylation products of the endothelin-A receptor. The expression levels of the different mutated endothelin-A receptors were similar to that of the wild-type endothelin-A receptors. Lane 1, COS-1 cells transfected with empty vector; lanes 2 and 8, wild-type endothelinA receptor; lane 3, [G97A]endothelin-A;lane 4, [K140I]endothelin-A; lane 5, [Q165D]endothelin-A; lane 6, [E220N]endothelin-A; lane 7,
[F315L]endothelin-A;lane 9, "29Alendothelin-A; lane 10, [N62A]endothelin-A; lane 11, [N29A/N62A]endothelin-A. from 1 5 O t 8 n M (wild type) to 580+105 nM (Y263F), 4502 130 nM (D351N) and greater than 10 pM (R326Q). Thus, the amino acids of endothelin-A receptor that are important for the binding of natural peptide agonists and synthetic low-molecular-mass non-peptide antagonists may differ. Tyr263, Asp35 1 and especially Arg326 are essential for the binding site of bosentan. As these amino acids are all conserved among endothelinA and endothelin-B receptors, this could explain why bosentan is a mixed antagonist for these receptors [14]. Similar interactions to those for bosentan seem to exist between the peptidic endothelin-A selective antagonist BQ-123 and the endothelin-A receptor, because for mutations R326Q and D351N we also observed a decreased binding of [3H]BQ-123 to 13% and 4%, respectively. However, in contrast to bosentdn, the binding of BQ-123 was not affected by the mutation Y263F, suggesting that Tyr263 does not play an important role for binding of BQ-123 to the endothelin receptor. The importance of amino acids from TM 7, like Asp351 for the binding of BQ 123 is consistent with other observations using chimeric receptors [18]. The selectivity of BQ 123 for the endothelin-A receptor may stem from additional contact sites with amino acids selectively present on endothelin-A receptors such as Tyrl29 [22]. It has been proposed that His323 (TM 6) could play a key role in endothelin binding [32]. This was concluded from chemical modification experiments of His residues with diethylpyrocarbonate. However, we did not observe a significant change in binding of endothelin-1 or bosentan when His323 was mutated to Phe. According to an alignment of all sequences of human Gprotein-coupled receptors present in release 27 of the SwissProt database [33], Tyr263 exists only in the neuromedin B receptor preferring bombesin, Arg326 is unique to endothelin receptors and Asp351 exists only in the anaphylatoxin chemotactic receptor and the calcitonin receptor. Thus, the specific interaction with 51-263, Arg326 and Asp351 may explain the specificity of bosentan for endothelin receptors 1141. In some other G-proteincoupled receptors, e.g., dopamine receptors and adrenergic receptors, the position of Tyr263 is occupied by Ser, another hydroxyl-group-containing amino acid, which is thought to interact with one of the hydroxyl groups of dopamine and adrenaline [26]. Considering the complete loss of binding affinity for bo-
Breu et al. (EUKJ. Biochem. 231)
sentan of receptors with mutations R326Q, and D351N it is tempting to speculate that Arg326 and Asp351 are involved in strong interactions with the sulfonamide group. These interactions would explain the high affinity of bosentan [I41 and Ro 46-2005 [2, 131 for endothelin receptors and characterize the sulfonamide part of the molecule as an important building block. This view is supported by the recent discovery of BMS-182874 [34] and L-749,329 [35], two other low-molecular-mass endothelin antagonists with a sulfonamide moiety. By introducing point mutations into the endothelin receptor it was not only possible to decrease, but also in the case of mutations K140I and F264L to increase the specific binding of [3H]bosentan to 168% and 196%, respectively (Fig. 2). Similar observations have been made for the p-adrenergic receptor [36]. When Phe412 was mutated to Asn, the receptor had a much higher affinity for the antagonists alprenolol, pindolol and propranolol. Single point mutations which differentially affect binding of agonist and antagonist have recently also been described for the cholecystokinin-B/gastrin receptor [37], the NK, receptor for substance P [38, 391 and the adrenergic p2 receptor [40, 411. For endothelin receptors it has been demonstrated, based on studies with chimeric receptor [16-181 that the C-terminal part of the natural agonist endothelin-1 interacts preferably with the region around transmembrane segments 1, 2, 3 and 7. The present study demonstrates a distinct binding site for non-peptide antagonists which differs from this agonist-binding site. It might overlap with a postulated second interaction site for natural agonists covering transmembrane segments 4, 5 and 6, which is believed to recognize the amino terminal loop domain of endothelin-1 [18]. This would provide an explanation why bosentan displays competitive antagonism for endothelin-1 binding. In general, the described data for mutated endothelin receptors allow us to obtain a more detailed picture of the agonistbinding and antagonist-binding site and the nature of the most important interactions. This is a pre-requisite for the improvement of the receptor model and for the design of more potent or selective antagonists. We wish to thank Mrs B. Butscha and Mr W. Lohrer for outstanding technical assistance and Dr W. Fischli for critical reading of the manuscript.
REFERENCES 1. Yanagisawa, M., Kurihara, H., Kimura, S., Tomobe, Y., Kobayashi, M., Mitsui, Y., Yazaki, Y., Goto, K. & Masaki, T. (1988) A novel potent vasoconstrictor peptide produced by vascular endothelial cells, Nature 332, 411 -415. 2. Clozel, M., Breu, V., Burri, K., Cassal, J.-M., Fischli, W., Gray, G. A,, Hirth, G., Loffler, B.-M., Miiller, M., Neidhart, W. & Ramuz, H. (1993) Pathophysiological role of endothelin revealed by the first orally active endothelin receptor antagonist, Nature 365, 759-761, 3. Dohi, Y. & Luscher, T. F. (1991) Endothelin in hypertensive resistance arteries, Hypertension (Dallas) 18, 543 -549. 4. Lerman, A,, Edwards, B. S., Hallet, J. W., Denise, M. D., Heublein, M., Sandberg, S. M. & Bumett, J. C. (1991) Circulating and tissue endothelin immunoreactivity in advanced atherosclerosis, N. Engl. J. Med. 325, 997-1001. 5. Arai, H., Hori, S., Aramori, I., Ohkubo, H. & Nakanishi, S. (1990) Cloning and expression of a cDNA encoding an endothelin receptor, Nature 348, 730-732. 6. Sakurai, T., Yanagisawa, M., Takuwa, Y., Miyazaki, H., Kimura, S., Goto, K. & Masaki, T. (1990) Cloning of a cDNA encoding a non-isopeptide-selective subtype of the endothelin receptor, Nature 348, 732-735. 7. Kame, S., Jayawickreme, C. K. & Lerner, M. R. (1993) Cloning and characterization of an endothelin-3 specific receptor (ET,)
269
receptor from Xenopus laevis dermal melanophores, J. Bid. Chem. 268, 19126-19133. 8. Kumar, C., Mwangi, V., Nuthulaganti, P., Wu, H.-L., Pullen, M., Brun, K., Aiyar, H., Morris, R. A,, Naughton, R. & Nambi, P. (1994) Cloning and characterization of a novel endothelin receptor from Xenopus heart, J. Biol. Chem. 269, 13414-13420. 9. Ihara, M., Noguchi, K., Saelu, T., Fukuroda, T., Tsuchida, S., Kimura, S., Fukami, T., Ishikawa, K., Nishikibe, M. & Yano, M. (1992) Biological profiles of highly potent novel endothelin antagonists selective for the ETAreceptor, Life Sci. 50, 247-255. 10. Takayanagi, R., Kitazumi, K., Takasaki, C., Ohnaka, K., Aimoto, S., Tasaka, K., Ohashi, M. & Nawata, H. (1991) Presence of nonselective type of endothelin receptor on vascular endothelium and its linkage to vasodilation, FEBS Lert. 282, 103-106. 11. Clozel, M., Gray, G. & Breu, V. (1992) The endothelin ET, receptor mediates both vasodilation and vasoconstriction in vivo, Biochem. Biophys. Res. Commun. 186, 867-873. 12. Moreland, S., McMullen, D. M., Delaney, C. L., Lee, V. G. & Hunt, J. T. (1992) Venous smooth muscle contains vasoconstrictor ET,like receptors, Biochem. Biophys. Res. Commun. 184, 100- 106. 13. Breu, V., Clozel, M. & Loffler, B.-M. (1993) In vitro characterization of Ro 46-2005, a novel synthetic non-peptide endothelin antagonist of ETA and ETB receptors, FEBS Lett. 334, 210-214. 14. Clozel, M., Breu, V., Gray, G., Kalina, B., Loffler, B.-M., Burri, K, Cassal, J.-M., Hirth, G., Muller, M., Neidhart, W. & Ramuz, H. (1994) Pharmacological characterization of bosentan, a new potent orally active nonpeptide endothelin receptor antagonist, J. Pharmacol. Exp. Thes 270, 228-235. 15. Adachi, M., Furuichi, Y. & Miyamoto, C. (1994) Identification of a region of the human endothelin ETAreceptor required for interaction with bosentan, Eus J. Pharmacol. 269, 225-234. 16. Adachi, M., Ymg, Y.-Y., Trzeciak, A., Furuichi, Y. & Miyamoto, C. (1992) Identification of a domain of ETA receptor required for ligand binding, FEBS Lett. 311, 179-183. 17. Adachi, M., Furuichi, Y. & Miyamoto, C. (1994) Identification of a ligand-binding site of the human endothelin-A receptor and specific regions required for ligand selectivity, EUKJ. Biochem. 220, 37-43. 18. Sakamoto, A., Yanagisawa, M., Sawamura, T., Enoki, T., Ohtani, T., Sakurai, T., Nakao, K., Toyo-Oka, T. & Masaki, T. (1993) Distinct subdomains of human endothelin receptors determine their selectivity to endothelin,-selective antagonist and endothelin,-selective agonists, J. Biol. Chem. 268, 8547-8553. 19. Adachi, M., Furuichi, Y. & Miyamoto, C. (1994) Identification of specific regions of the human endothelin-B receptor required for high affinity binding with endothelin-3, Biochim. Biophys. Actu 1223, 202-208. 20. Hashido, K., Adachi, M., Gamou, T., Watanabe, T., Furuichi, Y. & Miyamoto, C. (1993) Identification of specific intracellular domains of the human ETAreceptor required for ligand binding and signal transduction, Cell. Mol. Biol. Res. 39, 3-12. 21. Krystek, S. R., Pate, P. S., Rose, P. M., Fisher, S. M., Kienzle, B. K., Lach, D. A,, Liu, C.-K., Lynch, J. S., Novotny, J. & Webb, M. L. (1994) Mutation of peptide binding site in transmembrane region of a G protein-coupled receptor accounts for endothelin receptor subtype selectivity, J. Biol. Chem. 269, 12 383 - 12 386. 22. Lee, J. L., Elliot, J. D., Sutiphong, J. A,, Friesen, W. J., Ohlstein, E. H., Stadel, J. M., Gleason, J. G. & Peishoff, C. E. (1994) Tyr129 is important to the peptide ligand affinity and selectivity of human endothelin type A receptor, Proc. Nut1 Acad. Sci. 91, 7164-7168. 23. Schwartz, T. W. (1994) Locating ligand-binding sites in 7TM receptors by protein engineering, Curs Opin. Biotechnol. 5, 434-444. 24. Lewell, X. Q. (1992) A model of the adrenergic beta-2 receptor and binding sites for agonist and antagonist, Drug. Des. Discov. 9, 29-48. 25. Trumpp-Kallmeyer, S., Hoflack, J., Bruinvels, A. & Hibert, M. (1992) Modelling of G-protein-coupled receptors: application to dopamine, adrenaline, serotonin, acetylcholine and mammalian opsin receptors, J. Med. Chem. 35, 3448-3462. 26. Hibert, M. F., Trumpp-Kallmeyer, S., Bruinvels, A. & Hoflack, J. (1991) Three-dimensional models of neurotransmitter G-binding protein-coupled receptors, J. Mol. Phurmucol. 40, 8- 15.
270
Breu et al. (Eul: J. Biochem. 231)
27. Kunkel, T. A,, Roberts, J. D. & Zakour, R. A. (1987) Rapid and efficient site-specific mutagenesis without phenotypic selection, Methods Entymol. 154, 367-382. 28. Munson, P. J. & Rodbard, D. (1980) Ligand: a versatile computerized approach for characterization of ligand-binding systems, Anal. Biochem. 107, 220-239. 29. Henderson, R., Baldwin, J. M., Ceska, T. A., Zemlin, F., Beckmann, E. & Downing, K. H. (1990) Model for the structure of bacteriorhodopsin based on high-resolution electron cryo-microscopy, J. Mol. Biol. 213, 899-929. 30. Gerber, P. R. (1992) Peptide mechanics: A force field for peptides and proteins working with entire residues as smallest units, Biopolymers 32, 1003-1017. 31. Gerber, P. R. & Mueller, K. (1995) MAE, a generally applicable molecular force field for structure modelling in medicinal chemistry, J. Comput-Aided Mol. Design, in the press. 32. Huggins, J. P., Trumpp-Kallmeyer, S., Hibert, M. F., Hoflack, J. M., Fanger, B. 0. & Jones, C. R. (1993) Modelling and modification of the binding site of endothelin and other receptors, Eur: J. Pharmacot. 245, 203 -214. 33. Bairoch, A. & Boeckmann, B. (1992) The SWISS-PROT protein sequence data bank, Nucleic Acids Res. 20, 2019-2022. 34. Stein, P. D., Hunt, J. T., Floyd, D. M., Moreland, S . , Dickinson, K. E. J., Mitchell, C., Liu, E. C.-K., Webb, M. L., Murugesan, N., Dickey, J., McMullen, D., Zhang, R., Lee, V. G., Serafino, R., Delaney, C., Schaeffer, T. R. & Kozlowsky, M. (1994) The discovery of sulfonamide endothelin antagonists and the development of the orally active ETA antagonist 5-(dimethylamino)-N(3,4-dimethyl-5-isoxazolyl)-1-naphthalenesulfona~de, J. Med. Chem. 37, 329-331.
3.5. Walsh, T.F., Fitch, K. J., Chakravarty, P. K., Williams, D. L., Murphy, K. A., Nolan, N. A., O’Brien, J. A,, Lis, E. V., Pettibone, D. J., Kivlighn, S . D., Gabel, R. A., Zingaro, G. J., Krause, S . M., Siegl, P. K. S., Clineschmidt, B. V. & Greenlee, W. J. (1994) The discovery of L-749,329, a highly potent, orally active antagonist of endothelin receptors, Abstr: 13th Int. Symp. Med. Chem., Paris, September 19-23, 1994, p. 032. 36. Suryanarayana, S.,Daunt, D. A., Zastrow, M. V. & Kobilka, B. K. (1991) A point mutation in the seventh hydrophobic domain of the a2 adrenergic receptor increases its affinity for a family of b receptor antagonists, J. Biol. Chem. 266, 15488-15492. 37. Beinbom, M., Lee, Y.-M., McBride, E. W., Quinn, S. M. & Kopin, A. S.(1993) A single amino acid of the cholecystokinin-B/gastrin receptor determines specificity for non-peptide antagonists, Nature 362, 348-3.50. 38. Fong, T. M., Cascieri, M. A,, Yu, H., Bansal, A,, Swain, C. & Strader, C. D. (1993) Amino-aromatic interaction between histidine 197 of the neurokinin-I receptor and CP 96345, Nature 362, 350-353. 39. Gether, U.,Johansen, T. E., Snider, R. M., Lowe, J. A,, Nakanishi, S. & Schwartz, T. W. (1993) Different binding epitopes on the NKI receptor for substance P and a non-peptide antagonist, Nature 362, 345-347. 40. Chung, F. Z., Wang, C. D., Potter, P. C., Venter, J. C. & Fraser, C. M. (1988) Site-hrected mutagenesis and continuous expression of human b-adrenergic receptors, J. Biol. Chem. 263,4052-4055. 41. Strader, C. D., Sigal, I. S., Register, R. B., Candelore, M. R., Rands, E. & Dixon, R. A. F. (1987) Identification of residues required for ligand binding to the P-adrenergic receptor, Proc Natl Acad. Sci. 84, 4384-4388.
