Proper targeting and activity of a nonfunctioning thyroid-stimulating hormone receptor (TSHr) combining an inactivating and activating TSHr mutation in one receptor Patrizia Agretti 1 , Giuseppina De Marco 1 , Paola Collecchi 2 , Luca Chiovato 3 , Paolo Vitti 1 , Aldo Pinchera 1 and Massimo Tonacchera 1 1 Dipartimento di Endocrinologia e Metabolismo, Ortopedia e Traumatologia, Medicina del Lavoro, Universita ` di Pisa, Pisa, Italy; 2 Dipartimento di Oncologia, Divisione di Anatomia Patologica, Universita ` di Pisa, Pisa, Italy; 3 Cattedra di Endocrinologia, Fondazione S Maugeri IRCCS, Pavia, Italy Activating mutations of the thyroid-stimulating hormone receptor (TSHr) have been identified as a cause of toxic adenomas. Germline-inactivating TSHr mutations have been described as a cause of congenital hypothyroidism. The effects of combining activating and inactivating mutations within a single receptor was studied. The double mutant T477I/P639S contained an activating TSHr mutation (P639S) together with an inactivating one (T477I). The other one (I486M/P639S) contained two activating mutations. Constructs were expressed in COS-7 cells and basal and TSH-stimulated cyclic AMP (cAMP) accumulation and inositol phosphate (IP) production were determined. The expression at the cell surface was studied both with binding and fluorescence-activated cell scanning analysis. Our results show that the effect of combining the two activating muta- tions is an increase in the constitutive activity only for the cAMP pathway and not for the IP pathway suggesting that different mutations result in receptor conformations with different relative abilities to couple to G s -alpha or G q -alpha. Surprisingly the double mutant containing the T477I behaves as an activating receptor with constitutive activity both for the cAMP and IP pathways. These data show that an inactive form of the TSHr which is trapped inside a cell after transfection is able to gain the membrane surface when combined with an activated form of the receptor. Keywords: TSH receptor; G-protein-coupled receptors; constitutive activity; site-directed mutagenesis; somatic mutations; germline mutations. G-protein-coupled, seven transmembrane segment recep- tors comprise the largest superfamily of proteins in the body [1]. Many G-protein coupled receptors have a certain basal activity (constitutive activity) and thus can activate G-proteins in the absence of the agonist [1,2]. Interestingly, it has been encountered that discrete mutations of these receptors are able to dramatically increase this constitutive agonist-independent receptor activity [3]. The thyroid- stimulating hormone receptor (TSHr), together with the follicle-stimulating hormone (FSH) and the luteinizing hormone (LH) receptors, is a member of a subfamily of seven transmembrane G-protein-coupled receptors, charac- terized by a large N-terminal extracellular domain involved in hormone binding [4,5]; the receptor is mainly coupled to adenylyl cyclase via G s -alpha and, in some species including man, it activates also the inositol phosphate cascade (IP) via aG q -alpha protein [6–8]. Current models of G-protein- coupled receptor activation consider that binding of the ligand, within the slit formed by the transmembrane helices (for biogenic amines), and/or to the extracellular loops (for peptide ligands), relieves a built-in negative constraint by stabilizing an active conformation of the receptor [9]. In this conformation, the new position of the transmembrane helices translates into an increased affinity of the intracel- lular loops for G-proteins. Somatic and germline activating mutations of the TSHr gene have been identified as a major cause of toxic thyroid adenoma [3,10,11] and hereditary or sporadic nonautoimmune toxic thyroid hyperplasia [12,13], respectively. On the contrary, inactivating mutations aboli- shing basal activity or affecting agonist induced response have been described in cases of congenital hypothyroidism with thyroid hypoplasia [14–16]. All activating TSHr mutations have been shown to activate adenylyl cyclase when expressed in eukaryotic cells [3,10]. Some of these mutations possess also constitutive activity for the IP pathway [3,11]. Inactivating mutations of the TSHr gene may be due to truncated forms of the receptor or to point mutations [14–16]. It has been demonstrated in vitro that point mutations can alter the routing of the receptor to the cell surface, resulting in loss of basal activity and loss of agonist induced cAMP production [14–16]. In order to study the mechanism of activation of the TSHr we explored the effects of combining previously Correspondence to M. Tonacchera, Dipartimento di Endocrinologia, Via Paradisa 2, 56124 Pisa, Italy. Fax: 050 578772, Tel.: 050 995048, E-mail: mtonacchera@hotmail.com Abbreviations: bTSH, bovine TSH; FACS, fluorescence-activated cell scanning; FSH, follicle stimulating hormone; G q -alpha, G-protein q alpha; G s -alpha, G-protein s alpha; IP, inositol phosphate; KRH, Krebs/Ringer/Hepes; LH, luteinizing hormone; TSH, thyroid-stimulating hormone. (Received 11 July 2003, accepted 1 August 2003) Eur. J. Biochem. 270, 3839–3847 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03778.x described mutations within a single receptor. We decided to combine two particularly potent activating TSHr mutations and to combine an activating mutation together with an inactivating one. Double mutant receptors (containing two activating TSHr mutations) harboring the amino acid substitution P639S (6th transmembrane segment) [17] and I486M (1st extracellular loop) [3], named I486M/P639S, and the double mutant receptor (containing an activating together with an inacti- vating TSHr mutation) P639S and T477I (1st extracellular loop) [16], named T477I/P639S, were constructed. Con- structs were subcloned in the expression vector pSVL and, after transient expression in COS-7 cells, basal and TSH- induced cAMP and IP production were determined. Cell- surface expression was evaluated with [ 125 I]bTSH (bovine TSH) binding, an enzyme immunosorbent assay (EIA) and a fluorescence-activated cell scanning (FACS) analysis using different monoclonal antibodies against the TSHr. Our results show that the effect of combining the two activating mutations is an increase in the constitutive activity only for the cAMP pathway and not for the IP pathway suggesting that different mutations result in receptor conformations with different relative abilities to couple to G s -alpha or G q -alpha. Surprisingly the double mutant containing the T477I behaves as an activating receptor with constitutive activity both for the cAMP and IP pathways. These data show that an inactive form of the TSHr which is trapped inside a cell after transfection is able to gain the membrane surface when combined with an activated form of the receptor. Materials and methods Construction of mutant TSH receptors The constructs harboring the single mutated TSH receptors T477I, I486M and P639S have been described [11,16,17]. In brief, a SpeI-CvnI segment (1660–1932) or a CvnI-BstEII segment (1932–2450) in the cDNA of the wild type (WT) TSHr in the expression vector pSVL, was replaced by a homologous segment harboring the mutation in position 477 and 486 or 639, respectively. These mutated sequences were cloned directly from DNA extracted from nodular tissues obtained from patients with toxic multinodular goiter (I486M, P639S) or from the blood of a patient with a germline inactivating TSHr mutation (T477I). The SpeI restriction site in the WT-TSHr was created by site-directed mutagenesis (the change in the coding region does not modify the encoded amino acid sequence). For the construction of the double mutants, a CvnI- BstEII fragment in the T477I and I486M TSHr was substituted by homologous segments harboring the P639S mutation, yielding the T477I/P639S or I486M/P639S dou- ble mutant, respectively (Fig. 1). The resulting constructs were sequenced directly by an ABI PRISM 310 Genetic Analyzer (PE Applied Biosystems, Foster City, CA, USA) to verify the presence of the mutations. Expression in eukaryotic cells of mutated genes For transient expression, COS-7 cells were seeded at the concentration of about 150 000 cells per 3-cm dish for binding, EIA and FACS analysis, cAMP and IP determination. COS-7 cells were grown in DMEM sup- plemented with 10% fetal bovine serum, penicillin 100 IUÆmL )1 , streptomycin 100 lgÆmL )1 , fungizone 2.5 lgÆmL )1 and 1 m M sodium pyruvate. One day after seeding, cells were transfected with the DEAE–dextran method followed by a 2-min 10% dimethylsulfoxide shock [18]. For functional assays, 48 h after transfection cells were used for cAMP or IP determinations and for EIA, FACS analysis and [ 125 I]bTSH binding studies. Triplicate dishes were used for each condition and each experiment was repeated at least three times. Results were expressed as mean ± SEM from one representative experiment. When not shown, SEM values were so small that they fall within the symbols. cAMP assay Cells were washed with Krebs/Ringer/Hepes buffer (KRH) and preincubated for 30 min at 37 °C. This was followed by a 1-h incubation at 37 °C in the presence of 0.5 m M isobutylmethyl xanthine as a cAMP phosphodiesterase inhibitor, in the absence of bTSH (basal values), or in the presence of various concentrations of bTSH (Sigma Chemical Co). At the end of the incubation the medium was removed and replaced by 0.1 M HCl. The cell extracts were dried in a vacuum concentrator and cAMP was determined as described [17] and expressed as picomoles per dish. Fig. 1. Schematic representation of the double mutants TSHr (I486M/ P639S; T477I/P639S). 3840 P. Agretti et al. (Eur. J. Biochem. 270) Ó FEBS 2003 Inositol phosphate assay For IP determinations, 24 h after transfection cells were incubated with 20 lCiÆmL )1 [ 3 H]inositol (Amersham Phar- macia Biotech Europe, Germany). The next day, dishes were washed three times with KRH, preincubated in KRH plus LiCl 10 m M for 30 min at 37 °C and incubated for 18 min at 37 °C in the presence of KRH plus LiCl 10 m M (basal values) or in the same medium plus different concentrations of bTSH. The incubation was stopped by addition of ice cold 3% HClO 4 ,and 3 H-labeled IP were isolated and assayed by stepwise chromatography on AG1 · 8 resin [19]. The cell debris in the bottom of the dishes was dissolved in 1 M NaOH and counted as phos- phatidyl inositols. Results were expressed as percentage radioactivity incorporated in inositol phosphates (IP 1 + IP 2 +IP 3 ) over the sum of radioactivity in inositol phosphates and phosphatidyl inositols [19]. Binding assay Forty-eight hours after transfection, cells were washed once with Hanks’ solution in which NaCl was replaced by sucrose 280 m M containing 0.2% bovine serum albumin (BSA) and 2.5% low fat milk. Binding studies were performed by incubating cells in that same medium at room temperature for 4 h in the presence of about 90 000 c.p.m. [ 125 I]bTSH (a gift of BRAHMS, Berlin, Germany) and the appropriate concentrations of cold bTSH. At the end of the incubation cells were rinsed twice with ice cold Hanks’ medium, solubilized with 1 M NaOH and bound radioactivity was determined in a gamma- counter. In the absence of a consensus about the bioactivity of pure bovine TSH [20], we have expressed all TSH or TSHr concentrations in mUÆmL )1 , assuming a 1/1 stechio- metry for TSH binding to its receptor. The competition binding curves have been fitted by nonlinear regression assuming a single receptor site [21]. Enzyme immunosorbent assay (EIA) EIA measurements were carried out with transfected cells, nonpermeabilized and in suspension. Forty-eight hours after transfection cells were detached from the plates with NaCl/P i containing 5 m M EDTA and EGTA. Cells were then pelleted and incubated with a mouse anti-(human TSHr Ig) (Novocastra Laboratories Ltd, UK) diluted at 1 lgÆmL )1 in NaCl/P i containing 0.5% BSA. After two washes with NaCl/P i , cells were incubated for 1 h at 4 °C with peroxidase-conjugated anti-(mouse IgG) as secondary antibody (Sigma Chemical Co.) diluted 1 : 70 000 in NaCl/ P i containing 0.5% BSA. The cells were washed three times with NaCl/P i and, finally, incubated with the o-phenylene- diamine dihydrochloride substrate (Sigma Chemical Co.) for 30 min at 37 °C. The reaction was stopped by adding 1 M H 2 SO 4 and color development was measured at 492 nm. FACS analysis Cells were detached from culture dishes with 5 mmolÆL )1 each of ethylenediamine tetraacetate and ethyleneglycol- bis-(beta-aminoethyl ether)-N,N,N¢,N¢-tetraacetic acid in NaCl/P i and transferred to Falcon tubes (2052, Falcon Labware, Cockeysville, MD, USA). Cells were washed with NaCl/P i plus 0.1% BSA, centrifuged at 500 g at 4 °Cfor 3 min, and treated appropriately for the nonpermeabilized or permeabilized cell assay as described previously [16]. Nonpermeabilized cells were incubated at room tempera- ture for 30 min with 200 lL of a monoclonal antibody directed against the TSHr (BA8 gently gifted from Dr Sabine Costagliola) diluted in NaCl/P i plus 0.1% BSA. A blank sample was prepared by incubating cells with 200 lL NaCl/P i plus 0.1% BSA. For the permeabilized cell assay, cells were fixed with 2% NaCl/P i /paraformaldehyde (UCB, Brussels, Belgium) and then treated for 30 min with NaCl/P i plus 0.1% BSA and 0.2% saponin (Sigma Chemical Co., St. Louis, MO, USA); all subsequent steps with antibodies were performed in 0.2% saponin. Cell-bound monoclonal antibodies were detected washing the cells with NaCl/P i plus 0.1% BSA and then incubating them for 30 min at 4 °Cin the dark with a goat anti-(mouse IgG) fluorescin-conjugated (Becton Dickinson and Co., San Jose, CA, USA) diluted 1:20inNaCl/P i with 0.1% BSA containing 10 lgÆmL )1 propidium iodide (Sigma). Propidium iodide (PI) was used to detect and exclude from the analysis damaged cells. Flow cytometric analysis was performed using a FACSort flow cytometer (Becton Dickinson and Co.) equipped with a laser for an excitation at 488 nm to detect monoclonal antibodies conjugated with fluorescin 5-isothiocyanate and PI. Fluorescence emission of fluorescin 5-isothiocyanate and PI from single cells were separated and measured using the standard optics of the FACSort. The CELLQUEST software program (Becton Dickinson and Co.) was used to acquire and analyze data. A minimum of at least 10 000 cells was analyzed. Computation of specific constitutive activity (SCA) and relative SCA (RSCA) Given that the transfection efficiency for each construct is constant for a given batch of cells, the SCA was calculated by: SCA ¼ðAr À AvÞ=ðFr À FvÞ where Ar and Av are the cAMP (or IP) of cells transfected with the mutant constructs and vector, respectively, and Fr and Fv are the corresponding mean fluorescence unit obtained with FACS analysis. RSCA, which is normalized to the WT-TSHr, was obtained by: RSCA ¼ SCAr=SCA-WT: Results cAMP cascade The effects of each TSH receptor construct harboring one or two mutations have been investigated after transient expression in COS-7 cells. For basal cAMP determinations, 250 ng per dish of the DNA of the various constructs, giving maximal cAMP stimulation [3], were used to transfect COS-7 cells. As expected, cells transfected with the WT-TSHr exhibited a Ó FEBS 2003 Double mutant TSHr mutations (Eur. J. Biochem. 270) 3841 threefold increase in basal cyclic AMP accumulation with respect to cells transfected with vector alone, showing constitutive activity (Table 1, Fig. 2A). Cells transfected with two of the constructs harboring a single mutation (I486M; P639S) displayed higher level of basal cAMP as compared to cells transfected with the WT-TSHr (Table 1, Fig. 2A), while cells transfected with the construct harbor- ing the T477I mutation showed levels of cAMP that were very similar to those exhibited from the cells transfected with the empty vector alone. The I486M/P639S double mutant showed a further increase of cAMP accumulation with respect to the single parental mutated receptors. Surprisingly the combination of the activating P639S with the inactivating T477I yielded a receptor T477I/P639S with a strong constitutive activity similar to that obtained with P639S. The biological response to bTSH of the cells transfected with the DNA of the different constructs was explored in terms of cAMP accumulation (Table 1, Fig. 2A). Cells transfected with the constructs harboring a single activating mutation (P639S; I486M) exhibited a maximal stimulation to bTSH that was similar to that observed with the WT-TSHr. Cells transfected with the inactivating mutation T477I showed a very low response to bTSH with values of cAMP production at 100 mUÆmL )1 of bTSH about seven- fold lower than the WT-TSHr. The I486M/P639S double mutant showed an increase of maximal cAMP accumula- tion after bTSH challenge with respect to the single mutated receptors. The T477I/P639S maintained a cAMP response to bTSH similar to P639S alone. Inositol phosphate cascade As expected from previous studies [3,11], no significant increase of basal levels of IP production was observed in the cells transfected with the WT-TSHr (Table 1, Fig. 2B); a slight increase over values obtained with an empty vector or the wild-type receptor construct, was observed in cells transfected with the P639S or I486M single mutant, showing constitutive activity for the IP production (Table 1, Table 1. Functional characteristics of TSHr mutants transfected in COS-7 cells as described in materials and methods. cAMP values, expressed as percentage basal cAMP levels of WT-TSHr, were measured in basal conditions and after stimulation with 10 and 100 mUÆmL )1 bTSH. IP values, expressed as percentage basal IP levels of WT-TSHr, were measured in basal conditions and after stimulation with 100 mUÆmL )1 bTSH. Results are mean ± SEM of three independent experiments. cAMP IP Basal 10 mUÆmL )1 100 mUÆmL )1 Basal 100 mUÆmL )1 WT-TSHr 100 1070 ± 46 1087 ± 37 100 338 ± 12 T477I 40 ± 8 95 ± 11 149 ± 16 96 ± 11 107 ± 10 I486M 430 ± 22 1053 ± 29 1214 ± 39 170 ± 10 592 ± 26 P639S 476 ± 17 1137 ± 35 1241 ± 27 238 ± 21 700 ± 31 T477I/P639S 403 ± 12 970 ± 18 1024 ± 69 123 ± 13 423 ± 18 I486M/P639S 586 ± 21 1186 ± 34 1671 ± 35 215 ± 15 600 ± 24 Fig. 2. Basal and bTSH-induced stimulation of cAMP and IP levels in COS-7 cells transfected with the single or double mutated receptors. Values are mean ± SEM from one representative experiment in which triplicate dishes were used. (A) Basal intracellular cAMP values (picomoles per dish) and levels of cAMP production after stimulation with 10 and 100 mUÆmL )1 bTSH in COS-7 cells transfected with saturating concentrations of DNA (250 ng per dish), of single mutated receptors (T477I, I486M, P639S) or double mutant receptors (T477I/P639S; I486M/P639S), WT-TSHr and pSVL alone, are shown. (B) Basal IP values, expressed as percentage of (IP + PI) and IP levels after bTSH stimulation (100 mUÆmL )1 )in COS-7 cells transfected with saturating concentrations of DNA (250 ng per dish), of single mutated receptors (T477I, I486M, P639S) or double mutant receptors (T477I/P639S; I486M/P639S), WT-TSHr and pSVL alone, are shown. 3842 P. Agretti et al. (Eur. J. Biochem. 270) Ó FEBS 2003 Fig. 2B). Cells transfected with the T477I mutation showed basal IP values similar to those obtained with pSVL or WT- TSHr alone. Cells transfected with the I486M/P639S double mutant showed basal IP values intermediate to those obtained in cells transfected with the single parental mutants, showing that the double mutant promoted the same higher basal IP values. Cells transfected with the T477I/P639S double mutant showed only a slight increase in basal IP production with respect to WT-TSHr or T477I alone, and a clear decrease with respect to P639S. Stimulation of IP accumulation by 100 mUÆmL )1 of bTSH exhibited a threefold stimulation of IP production in cells transfected with the WT-TSHr (Table 1, Fig. 2B). The I486M and P639S mutated receptors showed an increased production of IP with respect to the WT-TSHr after bTSH stimulation. The T477I completely lost the ability to respond to bTSH. The I486M/P639S maintained a similar production of IP after bTSH challenge with respect to the parental TSH receptors. The T477I/P639S was able to produce a significant IP stimulation after bTSH challenge, which was lower to that produced by P639S alone. Binding of [ 125 I]bTSH To measure the total number of receptors (or TSH binding capacity, B max ) expressed at the surface of the cells transfected with the different constructs, and their relative dissociation constants (K d ), binding studies were performed with a bovine [ 125 I]TSH tracer as described in Materials and methods. Cells transfected with the constructs harboring the single activating mutations I486M and P639S, exhibited a lower level of expression as compared to the WT-TSHr, indicating that the increased cAMP production (measured in the same experiment) was not due to overexpression of the mutated receptors (Fig. 3). When compared with the WT-TSHr, the level of receptor expression in COS-7 cells transfected with the T477I inactivating mutant was about four times lower (Fig. 3). The I486M/P639S double mutant showed a similar level of expression with respect to the single mutated receptors. Surprisingly, also the T477I/P639S showed a good ability to be expressed at the surface of the cells. As already noted earlier [3], constitutively active receptors showed a higher affinity for bTSH binding than the WT-TSHr (Fig. 3). EIA and FACS analysis The level of receptor expression on the cell surface for the different constructs was independently measured by EIA, using a monoclonal antibody directed against the extracel- lular domain of the TSH receptor (NCL-TSH-R2, Novo- castra Laboratories Ltd, Newcastle, UK) (data not shown), and by FACS analysis using BA8, a monoclonal antibody directed against the TSHr (Fig. 4). COS-7 cells transfected with the inactivating mutant T477I showed a very low level of expression of the mutated receptor at the cell surface (Fig. 4A). The T477I was however, clearly detectable within the cells after saponin permeabilization (Fig. 4B), suggest- ing that the T477I receptor was synthesized and recognized by the antibody, but might have a defect in folding of the mutant protein resulting in abnormal routing to the plasma membrane. The level of expression at the cell surface of the single mutated receptors (I486M, P639S) and the I486M/ P639S double mutant was similar and was the same according to the binding experiments (Table 2). The FACS analysis also confirmed that the T477I/P639S double mutant was able to reach the membrane surface. The comparison of WT-TSHr and mutated receptors expression at cell surface by using the three methods are shown in Table 2. The level of receptor expression on the cell surface was statistically different for all the single and double mutant receptors with respect to the WT-TSHr. The Student’s t-test with 95% confidence was used to establish significance (P £ 0.05). Fig. 3. Binding characteristics of the single and double mutants expressedinCOS-7cells.[ 125 ]IbTSH binding to COS-7 cells transfected with 250 ng per dish of the construct harboring the single mutated receptors (T477I, I486M, P639S) or double mutant receptors (T477I/ P639S; I486M/P639S) and WT-TSHr. The B max , total receptor amount (binding capacity; expressed as milliunits of TSH per milliliter) and K d (alsoexpressedinmUÆmL )1 of TSH) were computed as des- cribed in Materials and methods (the SEM values are so small that they fall within the symbols). Ó FEBS 2003 Double mutant TSHr mutations (Eur. J. Biochem. 270) 3843 Specific constitutive activity The data obtained from the FACS analysis allowed measurement of efficiency of transfection and computation of the increase in cAMP and IP within the effectively transfected cells. Assuming that the mutations did not affect recognition by the monoclonal antibodies, these data also allowed normalization of the increase in cAMP and IP accumulation to the amount of receptor expressed at the cell surface, yielding an estimation of specific constitutive activity. When transfecting COS-7 cells with increasing amounts of wild-type and mutant cDNAs we observed that the constitutive activity of the WT and mutated receptors was linear over the range of cell surface expression (data not shown). The I486M showed to be eightfold and the P639S 10-fold more active than the WT-TSHr. The combination of the two activating TSHr mutations produced a I486M/ P639S double mutant that showed an increase of SCA with respect to the parental constructs (11-fold with respect to the WT-TSHr) (Fig. 5A). The combination of the activating/ inactivating T477I/P639S double mutant produced a recep- tor that was 7.5 more active than the WT-TSHr but less active with respect to the P639S alone. These data were also used to compute IP accumulation with respect to the amount of receptor expressed at the cell surface. The data clearly showed that both the single activated TSHr receptors had constitutive activity with respect to the WT-TSHr (Fig. 5B) being 15-fold for the I486M and 33-fold for the P639S. The combination of the two activating TSHr mutations produced a receptor which Fig. 4. Expression analysis of the single and double mutants by FACS analysis using the BA8 monoclonal antibody. Fluorescence intensity is expressed in arbitrary units, as a function of cell number plotted on a logarithmic scale. (A) Nonpermeabilized cells assayed after transfection with pSVL, TSHr WT, the T477I, I486M and P639S single mutants, and the T477I/P639S and I486M/P639S double mutants. BA8 immunoreactivity (arbitrary units): vector, 3.20; TSHr WT, 19.67; T477I, 5.04; I486M, 15.00; P639S, 14.20; T477I/P639S, 15.52; I486M/P639S, 16.21. (B) Saponin- permeabilized cells identically transfected. BA8 immunoreactivity (arbitrary units): vector, 6.00; TSHr WT 11.80; T477I, 15.13; I486M, 17.70; P639S, 22.37; T477I/P639S, 20.16; I486M/P639S, 17.68. Table 2. The level of receptor expression on the cell surface for the different constructs after transfection was measured by [ 125 I]bTSH binding, EIA and FACS analysis. Data are represented as a percentage of the values corresponding to WT-TSHr minus the nonspecific absorbance readings for EIA, the non-specific binding and the non- specific fluorescence for FACS analysis. The nonspecific data were obtained transfecting COS-7 cells with the empty vector. Values are mean ± SEM of three independent experiments. [ 125 I]bTSH binding EIA FACS WT-TSHr 100 100 100 T477I 23 ± 3.8 44 ± 7.5 26 ± 3.9 I486M 81 ± 13.2 80 ± 13.8 77 ± 10.2 P639S 70 ± 8.4 65 ± 6.8 72 ± 9.9 T477I/P639S 77 ± 10.8 67 ± 7.1 78 ± 12.1 I486M/P639S 77 ± 9.5 72 ± 9.5 81 ± 8.8 3844 P. Agretti et al. (Eur. J. Biochem. 270) Ó FEBS 2003 was less potent that the P639S alone (24-fold over the WT-TSHr) while the combination of the activating/inacti- vating T477I/P639S produced a constitutive activity of sixfold with respect to the WT-TSHr. Discussion According to the extension of the ternary complex model [22] for activation of G-proteins by serpentine receptors, the receptor is assumed to exist in two interconvertible confor- mations: the ÔinactiveÕ one, enforced by a built-in negative constraint, corresponds to the major unliganded form of the receptor; the active conformation capable of promoting GDP/GTP exchange on the G-protein is achieved by a fraction of the unliganded receptors only, but is stabilized by binding of its ligand. The model predicts that unliganded receptors may display a basal activity (corresponding to the fraction of active receptors) which can be antagonized by inverse agonists, the effect of which is to stabilize the inactive conformation [22]. The TSH receptor offers a series of characteristics which make it an interesting model to study in this context: (a) it is capable of activating both G s -alpha and G q -alpha [6–8]; (b) it displays an easily measurable constitutive activity towards adenylyl cyclase after transfection in COS cells [1,3,10,11]; (c) it can be activated by a surprisingly diverse spectrum of point mutations, or (d) by stimulating autoantibodies; (e) loss- of-function mutations have also been described, abolishing basal activity or affecting agonist-induced response and (f) the mechanism by which binding of the hormone to the N-terminal domain of the receptor translates into activation of its serpentine domain is unknown. With the aim of improving our understanding of the mechanisms underlying some of these characteristics, we have explored the effects of combining activating and inactivating mutations of the TSHr within a single receptor on its constitutive activity towards G s -alpha and G q -alpha and its binding characteristics and responsiveness to bTSH. When two strong activating mutations of the TSHr gene (both possessing constitutive activity for both cAMP and IP pathways) have been combined a further increase of basal and bTSH-stimulated cAMP production was observed with respect to the single parental receptors. The expression of the I486M/P639S on the cell surface was similar to the single mutated receptors. These results suggest that the double mutant I486M/P639S may be one step further in a scale of constraint release and the two mutations cooperate to potentiate the activity of the receptor. The increase of basal activity by combining activating mutations is well known for glycoprotein hormone receptors [23,24]. No further increase of IP was evident in the double mutant I486M/ P639S with respect to the parental receptors. This observa- tion suggests the possible existence of multiple conformation states with different abilities to interact with G s -alpha and or G q -alpha. The mutational dissociation of functional con- formations have also been described for glycoprotein hormone receptors [25–27]. The T477I has been described as an inactivating mutation in a patient with congenital hypothyroidism and thyroid hypoplasia [16]. Similarly, an inactivating mutation of the FSH receptor gene that is located in the close vicinity of the amino acid 477 of the TSHr was found in a patient affected by primary ovarian failure [28]. When the T477I was transiently expressed in COS-7 cells we observed the complete loss of basal constitutive activity for the cAMP pathway, a strong reduction in bTSH stimulated cAMP production and absent bTSH stimulated IP production. The level of receptor expression in COS-7 cells transfected with the T477I measured both by binding with bTSH and by using a monoclonal antibody in FACS analysis directed against the TSHr was very low. The very low expression of the mutated receptor on the cell membrane is probably due to the poor routing of the receptor at the cell surface. Surprisingly the effect of combining this T477I with the activating P639S produced a double mutant with a similar level of expression at the cell surface with respect to the parental P639S. The T477I/P639S behaves like the P639S alone, in terms of basal and bTSH stimulated cAMP accumulation. Only a slight decrease in the basal and bTSH stimulated IP production was noticed transfecting this T477I/P639S with respect to P639S alone. These data clearly show that an inactivated G protein- coupled receptor was trapped intracellularly after transient expression in eukaryotic cells and when combined with an activated form of the receptor reconstituted a functional membrane receptor. Similarly, a D578H mutant LH receptor was shown to increase the cell surface expression of poorly expressed vasopressin-LH receptor’s chimera [29]. However, contrary to what has been observed in our model where the P639S has a lower expression at the cell surface Fig. 5. RSCA of single and double mutant receptors for cAMP (A) and IP (B) computed as described in Materials and methods. Results are the mean ± SEM from one representative experiment in which triplicate dishes were used. Ó FEBS 2003 Double mutant TSHr mutations (Eur. J. Biochem. 270) 3845 with respect to the WT-TSHr, the D578H mutant alone was shown to be expressed at the cell surface four times more than the WT-LH receptor [29]. Maya-Nunez et al.[30] showed that a mutant (E90K) gonadotropin-releasing hormone receptor (GnRHr) was rescued either by deleting K191 or by adding a C-terminal sequence. Either of these approaches alone supported high-membrane expression of the GnRHr. It has been proposed that the E90K mutation itself while resulting in protein misfolding does not irreversibly destroy the intrinsic ability of the mutant to bind ligand or to couple effectors. The rescued receptor, now stabilized in the plasma membrane, was able to activate the appropriate effector system. The proper folding and assembly of polypeptides occur in the endoplasmic reti- culum. If polypeptides cannot fold correctly, mechanisms of quality control ensure that the aberrant protein is not further processed along the secretory pathway [31]. The quality control mechanisms are mediated by a family of proteins called molecular chaperones. Recently, a class of compounds called chemical chaperones were shown to reverse the intracellular retention of several misfolded proteins [31]. Besides it has been described that small cell- permeable molecules can act as either chemical chaperones or pharmacological (ligand-mediated) chaperones to rescue mutant protein function [31]. Thus, the development of strategies aimed at promoting proper folding and matur- ation of mutant proteins could provide new therapies for a wide spectrum of diseases [31]. In conclusion, the combination of two activating muta- tions of the TSHr determined an increase in the activity only for the cAMP pathway and not for the IP pathway suggesting that different activating mutations result in receptor conformations with different relative abilities to stimulate the cAMP or IP regulatory cascades. Surprisingly the combination of a strong inactivating mutation with an activating one produced a receptor that was able to be expressed at the cell surface with high constitutive activity. Acknowledgements This work was supported by the following grants: Ministero dell’Uni- versita ` e della Ricerca Scientifica (MURST), Programma di Ricerca: Le malattie della tiroide: dalle basi molecolari alla clinica. Ministero dell’Universita ` e della Ricerca Scientifica (MURST), Programma di Ricerca: Strategie per la valutazione degli effetti disruptivi dei contaminanti ambientali sul sistema endocrino degli animali e dell’uomo. CNR Progetto Biotecnologie CTB 99.00.224.PF 31: Basi molecolari delle neoplasie benigne e maligne della tiroide. Istituto Superiore di Sanita ` : Basi Molecolari dell’ipotiroidismo congenito: predizione, prevenzione ed intervento I.S.S. Centro Eccellenza Ambi- SEN, Pisa. References 1. Gether, U. (2000) Uncovering molecular mechanisms involved in activation of G protein-coupled receptors. Endocr. Rev. 21,90– 113. 2. Morris, A.J. & Malbon, C.C. (1999) Physiological regulation of G protein-linked signaling. Physiol. Rev. 79, 1373–1430. 3. Van Sande, J., Parma, J., Tonacchera, M., Swillens, S., Dumont, J. & Vassart, G. (1995) Somatic and germline mutations of the TSH receptor gene in thyroid diseases. J. Clin. Endocrinol. Metab. 80, 2577–2585. 4. Vassart, G. & Dumont, J.E. (1992) The thyrotropin receptor and the regulation of thyreocyte function and growth. Endocr. Rev. 13, 596–611. 5. Nagayama, Y. & Rapoport, B. (1992) The thyrotropin receptor 25 years after is discovery: new insight after its molecular cloning. Mol. Endocrinol. 6, 145–156. 6. Laurent, E., Mockel, J., Van Sande, J., Graff, I. & Dumont, J.E. (1987) Dual activation by thyrotropin of the phospholipase C and cyclic AMP cascades in human thyroid. Mol. Cell Endocrinol. 52, 273–278. 7. Van Sande, J., Raspe, E., Perret, J., Lejeune, C., Maenhaut, C., Vassart, G. & Dumont, J.E. (1990) Thyrotropin activates both the cAMP and the PIP2 cascades in CHO cells expressing the human cDNA of TSH receptor. Mol. Cell Endocrinol. 74,R1–R6. 8. Allgeier, A., Offermans, S., Van Sande, J., Spicher, K., Schultz, G. & Dumont, J.E. (1994) The human thyrotropin receptor activates G-protein Gs and Gq11. J. Biol. Chem. 269, 13733–13735. 9. Lefkowitz, R.J., Cotecchia, S., Samama, P. & Costa, T. (1994) Constitutive activity of receptors coupled to guanine nucletide regulatory proteins. Trends Pharmacol. Sci. 14, 303–307. 10. Parma, J., Duprez, L., Van Sande, J., Cochaux, P., Gervy, C., Mockel,J.,Dumont,J.E.&Vassart,G.(1993)Somaticmutations in the thyrotropin receptor gene cause hyperfunctioning thyroid adenomas. Nature 365, 649–651. 11. Parma, J., Van Sande, J., Swillens, S., Tonacchera, M., Dumont, J.E. & Vassart, G. (1995) Somatic mutations causing constitutive activity of the TSH receptor are the major cause of hyper- functioning thyroid adenomas: identification of additional mutations activating both the cAMP and inositol phosphate- Ca ++ cascades. Mol. Endocrinol. 9, 725–733. 12. Dupre, L., Parma, J., Van Sande, J., Allegeier, A., Leclere, J., Schvartz, C., Delisle, M.J., Decoulx, M., Orgiazzi, J., Dumont, J.E. & Vassart, G. (1994) Germline mutations in the thyrotropin receptor gene cause non-autoimmune autosomal dominant hyperthyroidism. Nat. Genet. 7, 396–401. 13. Kopp,P.,VanSande,J.,Parma,J.,Duprez,L.,Gerber,H.,Joss, E.,Jameson,J.L.,Dumont,J.E.&Vassart,G.(1995)Congenital hyperthyroidism caused by a mutation in the thyrotropin receptor gene. N. Engl. J. Med. 332, 150–154. 14. Abramowicz, M.J., Duprez, L., Parma, J., Vassart, G. & Heinrichs, C. (1997) Familial congenital hypothyroidism due to inactivating mutation of the thyrotropin receptor causing profound hypoplasia of the thyroid gland. J. Clin. Invest. 99, 3018–3024. 15. Biebermann, H., Schoneberg, T., Krude, H., Schultz, G., Guder- mann, T. & Gruters, A. (1997) Mutations of the thyrotropin receptor gene causing thyroid hypoplasia and persistent congenital hypothyroidism. J. Clin. Endocrinol. Metab. 82, 3471–3480. 16. Tonacchera, M., Agretti, P., Pinchera, A., Rosellini, V., Perri, A., Collecchi, P., Vitti, P. & Chiovato, L. (2000) Congenital hyperthyroidism with impaired thyroid response to thyrotropin (TSH) and absent circulating thyroglobulin: evidence for a new inactivating mutation of TSH receptor gene. J. Clin. Endocrinol. Metab. 85, 1001–1008. 17. Tonacchera, M., Chiovato, L., Pinchera, A., Agretti, P., Fiore, E., Cetani, F., Rocchi, R., Vacava, P., Miccoli, P. & Vitti, P. (1998) Hyperfunctioning thyroid nodules in toxic multinodular goiter share activating thyrotropin receptor mutations with solitary toxic adenoma. J. Clin. Endocrinol. Metab. 83, 492–498. 18. Lopata, M.A., Cleveland, D.W. & Solmer-Wess, B. (1984) High level expression of a chloramphenicol-acetyltransferase gene by DEAE-dextrane mediated DNA trasfection coupled with a dimethyl-sulfoxide or glycerol shock treatment. Nucleic Acids Res. 12, 5707–5711. 19. Berridge, M.J. (1983) Rapid accumulation of inositol triphosphate reveals that agonists hydrolyse polyphosphoinosites instead of phosphathydilinositol. Biochem. J. 212, 849–858. 3846 P. Agretti et al. (Eur. J. Biochem. 270) Ó FEBS 2003 20. Rapoport, B. & Seto, P. (1985) Bovine thyrotropin as a specific bioactivity 5- to 10-fold that of previous estimates for highly purified hormone. Endocrinology 116, 1379–1382. 21. Swillens, S. (1995) Interpretation of binding curves obtained with high receptor concentrations: practical aid for computer analysis. Mol. Pharmacol. 47, 1197–1203. 22. Samama, P., Pei, G., Costa, T., Cotecchia, S. & Lefkowitz, R.J. (1994) Negative antagonists promote and inactive conformation of the B 2 -adrenergic receptor. Mol. Pharmacol. 45, 390–394. 23. Kosugi, S., Mori, T. & Shenker, A. (1998) An anionic residue at position 564 is important for maintaining the inactive conforma- tion of the human lutropin/chorigonadotropin receptor. Mol. Pharmacol. 53, 894–901. 24. Angelova, K., Fanelli, F. & Puett, D. (2002) A model for costi- tutive lutropin receptor activation based on molecular stimulation and engineered mutations in transmembrane elices 6 and 7. J. Biol. Chem. 277, 32202–32213. 25. Gilchrist, R.L., Ryu, K.S., Ji, I. & Ji, T.H. (1996) The luteinizing hormone/chorionic gonadotropin receptor has distinct trans- membrane conductors for cAMP and inositol phoshate signals. J. Biol. Chem. 271, 19283–19287. 26. Biebermann, H., Schoneberg, T., Schulz, A., Krause, G., Gruters, A., Schultz, G. & Gudermann, T. (1998) A conserved tyrosine residue (Y601) in transmembrane domain 5 of the human thyro- tropin receptor serves as a molecular switch to determine G-pro- tein coupling. FASEB J. 12, 1461–1471. 27. Schulz, A., Schoneberg, T., Paschke, R., Schultz, G. & Gudermann, T. (1999) Role of the third intracellular loop for the activation of gonadotropin receptors. Mol. Endocrinol. 13, 181–190. 28. Doherty, E., Pakarinen, P., Tiitinen, A., Kiilavuori, A., Huhtaniemi,I.,Forrest,S.&Aittomaki,C.(2002)Anovel mutation in the FSH receptor inhibiting signal transduction and causing primary ovarian failure. J. Clin. Endocrinol. Metab. 87, 1151–1155. 29. Sangkuhl, K., Schulz, A., Schultz, G. & Schoneberg, T. (2002) Structural requirements for mutational lutropin/choriogonadot- ropin receptor activation. J. Biol. Chem. 277, 47748–47755. 30. Maya-Nunez, G., Janovick, J., Ulloa-Aguirre, A., Soderlund, D., Conn, P.M. & Mende, J.P. (2002) Molecular basis of hypogona- dotropic hypogonadism: restoration of mutant (E 90 K) GnRH receptor function by a deletion at a distant site. J. Clin. Endocrinol. Metab. 87, 2144–2149. 31. Morello,J.P.,Petaja-Repo,U.E.,Bichet,D.G.&Bouvier,M. (2000) Pharmacological chaperones: a new twist on receptor folding. Trends Phamacol. Sci. 12, 466–469. Ó FEBS 2003 Double mutant TSHr mutations (Eur. J. Biochem. 270) 3847 . Proper targeting and activity of a nonfunctioning thyroid-stimulating hormone receptor (TSHr) combining an inactivating and activating TSHr mutation in. to combine two particularly potent activating TSHr mutations and to combine an activating mutation together with an inactivating one. Double mutant receptors