A 5-HT4 Receptor Transmembrane Network Implicated in the Activity of Inverse Agonists but Not Agonists
2002; Elsevier BV; Volume: 277; Issue: 28 Linguagem: Inglês
10.1074/jbc.m202539200
ISSN1083-351X
AutoresLara Joubert, Sylvie Claeysen, Michèle Sebben, Anne‐Sophie Bessis, Robin D. Clark, Renée S. Martin, Joël Bockaert, Aline Dumuis,
Tópico(s)Neurobiology and Insect Physiology Research
ResumoActivation of G protein-coupled receptors is thought to involve disruption of intramolecular interactions that stabilize their inactive conformation. Such disruptions are induced by agonists or by constitutively active mutations. In the present study, novel potent inverse agonists are described to inhibit the constitutive activity of 5-HT4 receptors. Using these compounds and specific receptor mutations, we investigated the mechanisms by which inverse agonists may reverse the disruption of intramolecular interactions that causes constitutive activation. Two mutations (D1003.32A in transmembrane domain (TMD)-III and F2756.51A in TMD-VI) were found to completely block inverse agonist effects without impairing their binding properties nor the molecular activation switches induced by agonists. Based on the rhodopsin model, we propose that these mutated receptors are in equilibrium between two states R and R* but are unable to reach a third “silent” state stabilized by inverse agonists. We also found another mutation in TMD-VI (W2726.48A) that stabilized this silent state. This mutant remained fully activated by agonists. Molecular modeling indicated that Asp-100, Phe-275, and Trp-272 might constitute a network required for stabilization of the silent state by the described inverse agonists. However, this network is not necessary for agonist activity. Activation of G protein-coupled receptors is thought to involve disruption of intramolecular interactions that stabilize their inactive conformation. Such disruptions are induced by agonists or by constitutively active mutations. In the present study, novel potent inverse agonists are described to inhibit the constitutive activity of 5-HT4 receptors. Using these compounds and specific receptor mutations, we investigated the mechanisms by which inverse agonists may reverse the disruption of intramolecular interactions that causes constitutive activation. Two mutations (D1003.32A in transmembrane domain (TMD)-III and F2756.51A in TMD-VI) were found to completely block inverse agonist effects without impairing their binding properties nor the molecular activation switches induced by agonists. Based on the rhodopsin model, we propose that these mutated receptors are in equilibrium between two states R and R* but are unable to reach a third “silent” state stabilized by inverse agonists. We also found another mutation in TMD-VI (W2726.48A) that stabilized this silent state. This mutant remained fully activated by agonists. Molecular modeling indicated that Asp-100, Phe-275, and Trp-272 might constitute a network required for stabilization of the silent state by the described inverse agonists. However, this network is not necessary for agonist activity. G protein-coupled receptor 5-hydroxytryptamine 5-hydroxytryptamine receptor of class 4 G protein that stimulates adenylate cyclase transmembrane domain wild type inactive receptor conformation active receptor conformation ground state receptor conformation [1-[2(methylsulfonyl-amino)ethyl]4-piperidinyl]methyl-1-methylindole-3 carboxylate, maleate [1-[2-(methylsulfonylamino)ethyl]4-piperidinyl]methyl-5-fluoro-2-methoxy-1H-indole-3-carboxylate, hydrochloride [endo-N-8-methyl-8-azabicyclo(3.2.1)oct-3-yl]-2,3-dihydro-3-isopropyl-2-oxo-1H-benzimidazole-1-carboxamide hydrochloride 1-(8-amino-7-chloro-2,3-dihydrobenzo-(1,4)-dioxin-5-yl)-3-{1-[3-(3,4-dimethoxyphenyl)propyl]-piperidin-4-yl}propan-1-one hydrochloride salt 2,3-dihydrobenzo-(1,4)-dioxine-5-carboxylic acid 1-butylpiperidin-4-ylmethyl ester hydrochloride salt 2,3-dihydrobenzo-(1,4)-dioxine-5-carboxylic acid 1-butylpiperidin-4-ylmethylamide hydrochloride salt Dulbecco's modified Eagle's medium decomplemented fetal bovine serum G protein-coupled receptors (GPCRs)1 are the most numerous and versatile family of proteins that control cell-cell communications (1Bockaert J. Pin J.P. EMBO J. 1999; 18: 1723-1729Crossref PubMed Scopus (1215) Google Scholar, 2Gether U. Endocrinol. Rev. 2000; 21: 90-113Crossref PubMed Scopus (1095) Google Scholar, 3Wess J. FASEB J. 1997; 11: 346-354Crossref PubMed Scopus (506) Google Scholar). Their endogenous ligands include hormones, neurotransmitters, growth and survival factors, smells, tastes, and photons. Both wild type (WT) and mutant GPCRs have been discovered to be spontaneously active and to adopt an active (R*) conformation in absence of agonists (4Kjelsberg M.A. Cotecchia S. Ostrowski J. Caron M.C. Lefkowitz R.J. J. Biol. Chem. 1992; 267: 1430-1433Abstract Full Text PDF PubMed Google Scholar, 5Samama P. Cotecchia S. Costa T. Lefkowitz R.J. J. Biol. Chem. 1993; 268: 4625-4635Abstract Full Text PDF PubMed Google Scholar, 6Chidiac P. Hebert T.E. Valiquette M. Dennis M. Bouvier M. Mol. Pharmacol. 1994; 45: 490-499PubMed Google Scholar). In the simplest model, GPCRs exist in a dynamic equilibrium between active (R*) and inactive (R) states governed by an allosteric constant (J = [R]/[R*]). Different models have been proposed to simulate the transition of receptor between ligand or G protein-bound states. In the extended ternary complex, R* interacts with the G protein either spontaneously (R*G) or through agonist binding (AR*G). In the cubic ternary complex model, both the R and R* states interact with the G protein (for a review, see Ref. 7Kenakin T. FASEB J. 2001; 15: 598-611Crossref PubMed Scopus (352) Google Scholar). The efficacy of a ligand is thought to be a function of its relative affinity for R and R*. Therefore, neutral antagonists have equal affinity for both receptor states, agonists preferentially bind to R*, thereby shifting the position of equilibrium toward active conformation, and inverse agonists preferentially bind to R, shifting the position of equilibrium toward the inactive state. A growing body of evidence suggests the existence of multiple active and inactive conformations. A model of these multiple conformations has been proposed (2Gether U. Endocrinol. Rev. 2000; 21: 90-113Crossref PubMed Scopus (1095) Google Scholar, 7Kenakin T. FASEB J. 2001; 15: 598-611Crossref PubMed Scopus (352) Google Scholar), which accurately simulates the biochemical results of rhodopsin studies (8Surya A. Stadel J.M. Knox B.E. Trends Pharmacol. Sci. 1998; 19: 243-247Abstract Full Text PDF PubMed Scopus (23) Google Scholar, 9Bartl F.J. Ritter E. Hofmann K.P. J. Biol. Chem. 2001; 276: 30161-30166Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar). Understanding the function of GPCRs at a molecular level requires better understanding of the mechanisms by which the binding of agonists and inverse agonists preferentially stabilizes the R* and R conformations, respectively. An important contribution to this understanding was the identification of key residues, which, when altered by mutation, led to increased constitutive activity of GPCRs (i.e. accumulation of R* in the absence of agonists). These residues have been identified not only in the intracellular domains known to be involved in G protein activation, but also in transmembrane domains (TMDs) and even in some cases in extracellular domains. Most of these residues have been proposed to participate in networks of intramolecular interactions, which stabilize the receptors in inactive conformations R. Disruption of these networks results in a conformational shift from R to R* (2Gether U. Endocrinol. Rev. 2000; 21: 90-113Crossref PubMed Scopus (1095) Google Scholar, 10Robinson P.R. Cohen G.B. Zhukovsky E.A. Oprian D.D. Neuron. 1992; 9: 719-725Abstract Full Text PDF PubMed Scopus (438) Google Scholar, 11Porter J.E. Perez D.M. J. Biol. Chem. 1999; 274: 34535-34538Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar, 12Govaerts C. Lefort A. Costagliola S. Wodak S.J. Ballesteros J.A. Van Sande J. Pardo L. Vassart G. J. Biol. Chem. 2001; 276: 22991-22999Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar, 13Ballesteros J.A. Jensen A.D. Liapakis G. Rasmussen S.G. Shi L. Gether U. Javitch J.A. J. Biol. Chem. 2001; 276: 29171-29177Abstract Full Text Full Text PDF PubMed Scopus (525) Google Scholar, 14Ghanouni P. Gryczynski Z. Steenhuis J.J. Lee T.W. Farrens D.L. Lakowicz J.R. Kobilka B.K. J. Biol. Chem. 2001; 276: 24433-24436Abstract Full Text Full Text PDF PubMed Scopus (336) Google Scholar, 15Scheer A. Fanelli F. Costa T., De Benedetti P.G. Cotecchia S. EMBO J. 1996; 15: 3566-3578Crossref PubMed Scopus (356) Google Scholar). In the case of rhodopsin, a salt bridge constraint between Glu-1133.28 2For numbering of residues, we have followed either the convenient method proposed by Van Rhee and Jacobson, which includes both numbering in the sequence and position in a particular helix referred to its most conserved residue, or only the position in this helix.2For numbering of residues, we have followed either the convenient method proposed by Van Rhee and Jacobson, which includes both numbering in the sequence and position in a particular helix referred to its most conserved residue, or only the position in this helix. in TMD-III and Lys-2967.43 in TMD-VII stabilizes the receptor in the R state (10Robinson P.R. Cohen G.B. Zhukovsky E.A. Oprian D.D. Neuron. 1992; 9: 719-725Abstract Full Text PDF PubMed Scopus (438) Google Scholar). Similar intramolecular interactions have been demonstrated for α1A-adrenoceptors between Asp-1253.32 of TMD-III and Lys-3317.43 in TMD-VII. Activation of this receptor is believed to result from the disruption of the Asp-125-Lys-331 salt bridge by the neurotransmitter biogenic amine that acts as counter-ion of the Asp-125 (11Porter J.E. Perez D.M. J. Biol. Chem. 1999; 274: 34535-34538Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar). However, this mechanism of receptor activation is likely to be specific for α1B-adrenergic receptor, since the Lys in TMD-VII is not conserved within related GPCRs sequences. β2-Adrenergic receptors are kept under an inactive state by an ionic lock between the adjacent Arg and Asp of the DRY (D3.49R3.50Y3.51) sequence (cytoplasmic end of TMD-III) and a conserved Glu in the cytoplasmic end of TMD-VI (Glu-2686.30 in β2-adrenergic receptor) (13Ballesteros J.A. Jensen A.D. Liapakis G. Rasmussen S.G. Shi L. Gether U. Javitch J.A. J. Biol. Chem. 2001; 276: 29171-29177Abstract Full Text Full Text PDF PubMed Scopus (525) Google Scholar). For several GPCRs, charge-neutralizing mutations of these residues increase constitutive activity by disrupting this lock (15Scheer A. Fanelli F. Costa T., De Benedetti P.G. Cotecchia S. EMBO J. 1996; 15: 3566-3578Crossref PubMed Scopus (356) Google Scholar, 16Rasmussen S.G. Jensen A.D. Liapakis G. Ghanouni P. Javitch J.A. Gether U. Mol. Pharmacol. 1999; 56: 175-184Crossref PubMed Scopus (191) Google Scholar, 17Alewijnse A.E. Timmerman H. Jacobs E.H. Smit M.J. Roovers E. Cotecchia S. Leurs R. Mol. Pharmacol. 2000; 57: 890-898PubMed Google Scholar). In rhodopsin, similar disruption is observed following protonation of the Glu in the ERY sequence (equivalent to the DRY sequence) and another acidic residue that could be the conserved Glu of TMD-VI (18Arnis S. Fahmy K. Hofmann K.P. Sakmar T.P. J. Biol. Chem. 1994; 269: 23879-23881Abstract Full Text PDF PubMed Google Scholar). Taken together, these findings support the idea that the disruption of intramolecular interactions constitutes the receptor activation switch, although the precise residues involved in these interactions may vary between different receptors. Such a hypothesis is also supported by the observation that constitutively active receptors are characterized by structural instability (16Rasmussen S.G. Jensen A.D. Liapakis G. Ghanouni P. Javitch J.A. Gether U. Mol. Pharmacol. 1999; 56: 175-184Crossref PubMed Scopus (191) Google Scholar, 19Gether U. Ballesteros J.A. Seifert R. Sanders-Bush E. Weinstein H. Kobilka B.K. J. Biol. Chem. 1997; 272: 2587-2590Abstract Full Text Full Text PDF PubMed Scopus (259) Google Scholar, 20Claeysen S. Sebben M. Bécamel C. Parmentier M.-L. Dumuis A. Bockaert J. EMBO Rep. 2001; 2: 61-67Crossref PubMed Scopus (28) Google Scholar). In contrast to the rapid progress in understanding molecular mechanisms underlying the transition from R to R*, mechanisms by which inverse agonists favor the conversion from R* to R (inactive conformation) remain unclear. An important observation was that mutations of residues that constrain the receptor in the inactive state lead to constitutively activated receptors that can be switched back to an inactive state by inverse agonists. This suggests to us that inverse agonist action may implicate different constraining networks than those disrupted by agonists. The present study, combining mutagenesis, pharmacology, and molecular modeling of 5-HT4 receptor, sheds new light on this issue. We previously reported that 5-HT4 receptors display high constitutive activity under wild type and mutated forms (20Claeysen S. Sebben M. Bécamel C. Parmentier M.-L. Dumuis A. Bockaert J. EMBO Rep. 2001; 2: 61-67Crossref PubMed Scopus (28) Google Scholar, 21Claeysen S. Sebben M. Bécamel C. Eglen R. Clark R.D. Bockaert J. Dumuis A. Mol. Pharmacol. 2000; 58: 136-144Crossref PubMed Scopus (54) Google Scholar). In this study, three mutations are reported (D1003.32A in TMD-III and F2756.51A and W2726.48A in TMD-VI) that completely blocked the effects of inverse agonists while keeping their specific binding on the receptor. Besides, when combined with other potent constitutively activating mutations, these three mutations were discovered to be “dominant” as inverse agonist effects were still blocked. Surprisingly, they also left unchanged molecular activation switches induced by agonists. Molecular modeling revealed a network of interactions between these three residues. The mutants were generated by exchanging the endogenous residues to alanine in the m5-HT4(a) R cDNA sequence with the QuikChange site-directed mutagenesis kit from Stratagene. The sense primers used were: F2756.51A, 5′-CTG TTT CTG CTG GGC CCC CGC CTT TG-3′; D1003.32A, 5′-ACC TCT CTG GCT GTC CTA CTC ACC-3′; W2726.48A, 5′-CTG TTT CTG CGC GGC CCC CTT CTT-3′; S1975.43A, 5′-TAT GCT ATC ACC TGC GCT GTG GT-3′; R1183.50A, 5′-TTC CCT GGA CGC TTA TTA CGC CAT-3′; N3087.49D, 5′-GGT TGG ACC CTT TTC TCT ATG CCT-3′; T1344.38A, 5′-TGC GTA GAG GGG CCA TCT TGT TCC-3′. The cDNAs, subcloned into pRK5, were introduced into COS-7 cells by electroporation. Briefly, cells were trypsinized, centrifuged, and resuspended in EP buffer (50 mm K2HPO4, 20 mmCH3CO2K, 20 mm KOH, 26.7 mm MgSO4, pH 7.4) with 25–2000 ng of receptor cDNA. The total amount of cDNA was kept constant at 15 μg per transfection with wild type pRK5 vector. After 15 min at room temperature, 300 ml of cell suspension (107 cells) were transferred to a 0.4-cm electroporation cuvette (Bio-Rad, Ivry sur Seine, France) and pulsed with a gene pulser apparatus (setting 1000 millifarads, 270 V). Cells were diluted in Dulbecco's modified Eagle's medium (DMEM; 106 cells/ml) containing 10% dialyzed and decomplemented fetal bovine serum (dFBS) and plated on 10-cm Falcon Petri dishes or into 12-well clusters at the desired density. Six hours after transfection, the surrounding cell medium was exchanged for DMEM without dFBS with 2 mCi of [3H]adenine/ml to label the ATP pool and incubated overnight. cAMP accumulation was measured as described previously (22Dumuis A. Bouhelal R. Sebben M. Cory R. Bockaert J. Mol. Pharmacol. 1988; 34: 880-887PubMed Google Scholar). Membranes were prepared from transiently transfected cells plated on 15-cm dishes and grown in DMEM with 10% dFBS for 6 h, followed by incubation for 20 h in DMEM without dFBS. The cells were washed twice in phosphate-buffered saline, scraped with a rubber policeman, harvested in phosphate-buffered saline, and centrifuged at 4 °C (200 × g for 4 min). The pellet was resuspended in buffer containing 10 mm HEPES, pH 7.4, 5 mm EGTA, 1 mm EDTA, and 0.32 msucrose and homogenized 10 times with a glass-Teflon potter at 4 °C. The homogenate was centrifuged at 20,000 × g for 20 min, and the membrane pellet was resuspended in 50 mmHEPES, pH 7.4, (5 mg of protein in 1 ml of solution) and stored at −80 °C until use. 5-HT4 receptor densities were estimated with the specific radioligand [3H]GR113808 at saturating concentration (0.4–0.6 nm,K d = 0.12 nm) as described previously (23Ansanay H. Sebben M. Bockaert J. Dumuis A. Eur. J. Pharmacol. 1996; 298: 165-174Crossref PubMed Scopus (39) Google Scholar). 5-HT (50 μm) or RS100325 (10 μm) was used to determine nonspecific binding. Protein concentration in the samples was determined with the Bio-Rad protein assay (24Bradford M.M. Anal. Biochem. 1976; 72: 248-254Crossref PubMed Scopus (211983) Google Scholar). Using Kaleidagraph software, the dose-response curves were fitted according to the equation, Y=((ymax−ymin)/(1+x/EC50)nH))+yminEquation 1 where EC50 (or EC50inv) is the concentration of agonist (or inverse agonist) that evokes a half-maximal response; y max andy min correspond to the maximal and minimal responses, respectively; and n H is the Hill coefficient. Statistical differences were examined with the Stat-View Student program (Abacus Concepts, Berkeley, CA) with ttests. GR113808 ([1-[2(methylsulfonyl-amino)ethyl]4-piperidinyl]methyl-1-methyl-indole-3 carboxylate, maleate) and GR125487 ([1-[2-(methylsulfonyl-amino)ethyl]4-piperidinyl]methyl-5-fluoro-2-methoxy-1H-indole-3-carboxylate, hydrochloride) were generously donated by Glaxo (Ware, Herts, UK); [3H]GR 113808 was purchased from AmershamBiosciences; 5-HT (serotonin) was purchased from Sigma. BIMU8 ([endo-N-8-methyl-8-azabicyclo(3.2.1)oct-3-yl]-2,3-dihydro-3-isopropyl-2-oxo-1H-benzimidazole-1-carboxamide hydrochloride) was obtained from Boehringer Ingelheim (Milan, Italy). RS100235 (1-(8-amino-7-chloro-2,3-dihydrobenzo(1,4)dioxin-5-yl)-3-{1-[3-(3,4-dimethoxyphenyl)propyl]-piperidin-4-yl}-propan-1-one hydrochloride salt), RO116-0086 (2,3-dihydrobenzo-(1,4)-dioxine-5-carboxylic acid 1-butyl-piperidin-4-ylmethyl ester hydrochloride salt), RO116-1148 (2,3-dihydrobenzo-(1,4)-dioxine-5-carboxylic acid 1-butyl-piperidin-4-ylmethylamide hydrochloride salt) were generously donated by Roche Bioscience (Palo Alto, CA). A model of the transmembrane domain in the mouse 5-HT4receptor was constructed using the bovine rhodopsin structure (Protein Data Bank code: 1F88:A) as a template. Receptor sequence alignment was determined using GRAP Mutant data base (tGRAP.uit.no/fam1asel.html). All 5-HT4 receptor modeling was conducted following virtual truncation of certain amino acid residues: 1–10 of the N-terminal region, 163–183 of the e2 loop, 219–250 of the i3 loop, and 332–387 of the C-terminal region. Five different models were developed using the program Modeler in the Insight-II environment (Molecular Simulation Inc., San Diego, CA) on a Silicon Graphics R10000 O2 work station. Although these models generated very similar molecular structures, the findings reported here are based on the model that returned the best three-dimensional likeness of rhodopsin. This model was submitted to energy minimization using the program Discover 3 and the consistent force field. Cα-trace was tethered by applying a force constant value of the quadratic potential of 100. A steepest descent followed by a conjugate gradient method then was applied until convergence with 0.5 kcal mol−1 Å−1 root mean square energy gradient difference between successive minimization steps. GR125487 and SB207266 have previously been described to be inverse agonists at the 5-HT4 receptor (21Claeysen S. Sebben M. Bécamel C. Eglen R. Clark R.D. Bockaert J. Dumuis A. Mol. Pharmacol. 2000; 58: 136-144Crossref PubMed Scopus (54) Google Scholar). To find new 5-HT4 receptor inverse agonists we synthetized numerous compounds having common pharmacophoric features with the SB204070, a specific benzoate dioxane 5-HT4 receptor antagonist (25Gaster L.M. Sanger G.J. Drugs Future. 1994; 19: 1109-1121Crossref Scopus (20) Google Scholar). Among them, two highly potent inverse agonists were identified, RO116-0086, a benzoate dioxane that is the des-amino, des-chloro version of SB204070, and RO116-1148, which is the corresponding amide. By measurement of cAMP accumulation in cell lines that transiently express wild type 5-HT4(a) receptor, both RO116-0086 and RO116-1148 were found to have high potencies (EC50inv = 0.3 ± 0.1 nm) and high efficacies. Indeed, they inhibited 80% of basal constitutive activity, i.e. cAMP accumulation above mock cAMP accumulation, in the absence of agonist (Fig. 1 B). Both are much more potent and efficacious than previously described 5-HT4inverse agonists (21Claeysen S. Sebben M. Bécamel C. Eglen R. Clark R.D. Bockaert J. Dumuis A. Mol. Pharmacol. 2000; 58: 136-144Crossref PubMed Scopus (54) Google Scholar). Screening for mutations that would impair the efficacy of inverse agonists without modifying the agonist-induced activation, we analyzed the mutation of Asp-1003.32. This highly conserved residue within the biogenic amine receptor family (26Hibert M.F. Trumpp-Kallmeyer S. Bruinvels A. Hoflack J. Mol. Pharmacol. 1991; 40: 8-15PubMed Google Scholar) is believed to make an ionic interaction with neurotransmitters protonated amine. As expected, 5-HT was completely unable to stimulate the D100A mutated receptor (Fig.2 A), due to its loss of binding (Fig. 2 B). Interestingly, the binding of the labeled antagonist [3H]GR113808 was only slightly affected by the mutation (K D = 0.12 ± 0.05 nm and 0.26 ± 0.04 nm at WT and D100A receptors, respectively). In contrast to 5-HT, the benzimidazolone derivative BIMU8 remained fully and even showed a higher potency on the D100A mutant than on the WT 5-HT4 receptor. The EC50 values for cAMP stimulation were equal to 4 ± 1.5 and 1 ± 0.5 nm for WT and D100A, respectively (Fig. 2 A). Affinity of BIMU8 for 5-HT4 receptor was also slightly better on the D100A mutant than on the WT.K D values for BIMU8 measured by competition with the [3H]GR113808 radioligand were 30 ± 11 nm and 6.5 ± 3 nm for WT and D100A, respectively (Fig. 2 B). These data suggest that the D1003.32A mutant has not lost the ability to be activated as long as the agonist is able to bind. When expressed at densities ranging from 200 up to 10,000 fmol/mg of protein, the basal level of cAMP synthesis of the D100A mutant was 1.5–2-fold higher than that of WT. Receptor expression levels of WT or D100A were also not significantly different following transfection of 300 ng of cDNA (1350 ± 60 and 1580 ± 60 pmol/mg of protein for WT and D100A, respectively). Whereas, as already discussed, highly potent inverse agonists (RO116-0086, RO116-1148) blocked the WT constitutive activity, they were absolutely unable to inhibit basal constitutive activity of the D100A mutant (Fig. 3 A). This effect is not unique to these two newly highly potent inverse agonists; similar results were obtained with the previously described 5-HT4 receptor inverse agonists, GR125487 and SB207266 (21Claeysen S. Sebben M. Bécamel C. Eglen R. Clark R.D. Bockaert J. Dumuis A. Mol. Pharmacol. 2000; 58: 136-144Crossref PubMed Scopus (54) Google Scholar) (data not shown). This effect can therefore not be attributed to a loss of binding to the D100A mutant receptor, as shown in Fig.4, A and B. RO116-0086 competition binding experiments showed a very small decrease in affinity for D100A (K d = 3 ± 1.4 × 10−10m) relative to WT (K d = 8.4 ± 2.6 × 10−11m). RO116-1148 showed a 10-fold decrease in affinity for D100A (K d = 5.1 ± 1.8 × 10−9m) as compared with the WT (K d = 4.7 ± 1.5 × 10−10m).Figure 4The binding of the 5-HT4 receptor inverse agonists was conserved in the mutated D1003.32A receptor. Competition binding was performed for the inverse agonists (RO116-0086 (A) and RO116-1148 (B)), in the presence of 0.24 nm [3H]GR113808. The assays were carried out on membranes derived from COS-7 cells and expressing similar levels of WT or D100A receptor (2500 ± 180 and 2650 ± 210 fmol/mg of protein, respectively). Results are expressed as a percent of the specific binding in the absence of a competing ligand. Representative data are presented from one of six replicate experiments.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The binding of these inverse agonists to the D100A mutant was confirmed by the capacity of these compounds to behave as antagonist of the BIMU8-stimulated cAMP accumulation (Fig. 3, B andC). The influence of receptor expression on inverse agonist activity was then analyzed. While the receptor constitutive activity increased as expected as a function of the receptor concentration, RO116-1148 remained totally inactive as inverse agonist on the D100A mutant (Fig.5 B). In comparison, its effect on WT receptor was constant whatever the receptor concentration (Fig.5 A). Thus, a mutant was generated that was fully activated by but insensitive to inverse agonists despite the ability of these later to bind to the receptor. In a previous study we showed that either truncation of the C-terminal domain of 5-HT4 receptor (Δ3277.70) or the A2586.34L mutation in the third intracellular loop elevated the constitutive activity relative to the WT 5-HT4 receptor (21Claeysen S. Sebben M. Bécamel C. Eglen R. Clark R.D. Bockaert J. Dumuis A. Mol. Pharmacol. 2000; 58: 136-144Crossref PubMed Scopus (54) Google Scholar) (Fig. 6 A). At both of these mutations, 5-HT and BIMU8 increased cAMP synthesis over basal (Fig. 6 B), and RO116-1148 decreased cAMP synthesis relative to basal (Fig. 6, C and D). Following dual mutation of D100A+Δ327 or D100A+A258L, the inverse agonist property of RO116-1148 was lost (Fig. 6, C and D). Furthermore, as expected 5-HT was inactive on these double mutants, whereas as on D100A, BIMU8 remained fully efficient (Fig.6 B). Taken together, these results showed that the D100A+Δ327 and D100A+A258L dual mutants were qualitatively similar in their pharmacological behavior to the D100A mutant on its own. Thus, the D100A mutant is a dominant mutant. It blocked the effects of inverse agonists on constitutively activated 5-HT4receptors, whatever the nature or intensity of the constitutive activation. We investigated the effects of mutation of other residues known to be involved in the binding of 5-HT or the constitutive activation of the receptor (Fig.7 A). The T1344.38A mutation is located in the second intracellular loop (Thr-1344.38 is a conserved threonine in many 5-HT receptors (27Lembo P.M. Ghahremani M.H. Morris S.J. Albert P.R. Mol. Pharmacol. 1997; 52: 164-171Crossref PubMed Scopus (43) Google Scholar)). The R1183.50A mutation is located at the putative interface between TMD-III and the i2 loop (Arg-118 residue belongs to the DRY sequence). In addition to the D100A mutant, four mutants were generated in the membrane-spanning regions. The Ser-1975.43 of TMD-V, which likely interacts via a hydrogen bond with the OH of 5-HT, was mutated into Ala (S197A). Two highly conserved aromatic residues of TMD-VI (F2756.51A and W2726.48A) (28Strader C.D. Fong T.M. Graziano M.P. Tota M.R. FASEB J. 1995; 9: 745-754Crossref PubMed Scopus (327) Google Scholar) were mutated into Ala. The Asn residue in of TMD-VII (Asn-3087.49), which is conserved across the family 1 GPCRs, was mutated to Asp. The S197A, T134A, and F275A mutants were similar to the WT receptor in their levels of constitutive activity. The N308D and R118A mutants had levels of constitutive activity, similar to that of D100A (Fig.7 B). Of all the mutants investigated, we found only F275A to be similar to D100A in its insensitivity to the potent inverse agonist, RO116-1148 (Fig. 7, B and C). As with the D100A mutation, the absence of effect of RO116-1148 cannot be attributed to an absence of binding. Competition binding experiments with RO116-1148 demonstrated a very small decrease in the affinity for F275A (9 ± 2.3 × 10−10m) relative to WT (4.7 ± 1.5 × 10−10m) (Fig.8 A). In addition, the 5-HT-mediated cAMP accumulation at this F275A mutant was antagonized by RO116-1148 (Fig. 8 B). The W272A mutant expressed at a standard level of 450 fmol/mg, was totally silent, with a level of basal cAMP accumulation similar to that of mock-transfected controls (Fig. 7 B). However, as shown in Fig. 8 B, the 5-HT efficiency to stimulate cAMP accumulation was the same on the W272A mutant as on WT 5-HT4 receptor. Since it displayed no constitutive activity at all, this silent mutant was, by definition, insensitive to the inverse agonist activity of RO116-1148. To check if this mutation was also a dominant mutation that prevents the effect of inverse agonists, we performed dual mutation studies with the constitutively active mutant, A258L. The W272A+A258L mutant was found to have a modest level of constitutive activity (Fig. 7 B). Note that this modest constitutive activity could not be further increased, due to a low expression level of W272A mutant. The same phenomenon was already observed with a mutation of the 5-HT2A receptor when the analogous tryptophan was mutated into alanine at W3366.48A (29Roth B.L. Shoham M. Choudhary M.S. Khan N. Mol. Pharmacol. 1997; 52: 259-266Crossref PubMed Scopus (129) Google Scholar). However, if modest, this basal level of cAMP accumulation was sufficient to reveal that RO116-1148 did not induce detectable decreases in basal cAMP accumulation (Fig. 7 C). Competition binding studies with [3H]GR113808 demonstrated that RO116-1148 binds to the W272A mutant (1.14 ± 0.3 × 10−9m) with a very small decrease in affinity relative to WT (Fig. 8 A). In addition, the 5-HT-mediated cAMP accumulation at this W272A mutant was antagonized by RO116-1148 (Fig. 8 B). Taken together, these results revealed that two mutations (W2726.48A and F2756.51A), in addition to the D1003.32A mutatio
Referência(s)