点击显示 收起
Hormone Research Center, School of Biological Sciences and Technology, Chonnam National University, Gwangju, Republic of Korea (J.H.L., A.F.W., K.M., C.W., M.A.S., S.Y.C., H.B.K., J.Y.S.)
Department of Physiology, Ulsan University College of Medicine, Seoul, Republic of Korea (H.C.)
Department of Aquaculture, Division of Aqua Life Science, Yeosu National University, Jeollanam-Do, Republic of Korea (W.-K.L.)
School of Biological Sciences, Seoul National University, Seoul, Republic of Korea (K.K.)
Abstract
Mammalian type I and II gonadotropin-releasing hormone (GnRH) receptors (GnRHRs) show differential ligand preference for GnRH-I and GnRH-II, respectively. Using a variety of chimeric receptors based on green monkey GnRHR-2 (gmGnRHR-2), a representative type II GnRHR, and rat GnRHR, a representative type I GnRHR, this study elucidated specific domains responsible for this ligand selectivity. A chimeric gmGnRHR-2 with the extracellular loop 3 (EL3) and EL3-proximal transmembrane helix 7 (TMH7) of rat GnRHR showed a great increase in ligand sensitivity to GnRH-I but not to GnRH-II. Point-mutation studies indicate that four amino acids, Leu/Phe7.38, Leu/Phe7.43, Ala/Pro7.46, and Pro/Cys7.47 in TMH7 are critical for ligand selectivity as well as receptor conformation. Furthermore, a combinatory mutation (Pro7.31-Pro7.32-Ser7.33 motif to Ser-Glu-Pro in EL3 and Leu7.38, Leu7.43, Ala7.46, and Pro7.47 to those of rat GnRHR) in gmGnRH-2 exhibited an approximately 500-fold increased sensitivity to GnRH-I, indicating that these residues are critical for discriminating GnRH-II from GnRH-I. [Trp7]GnRH-I and [Trp8]GnRH-I but not [His5]GnRH-I exhibit a higher potency in activating wild-type gmGnRHR-2 than native GnRH-I, indicating that amino acids at positions 7 and 8 of GnRHs are more important than position 5 for differential recognition by type I and type II GnRHRs. As a whole, these data suggest a molecular coevolution of ligands and their receptors and facilitate the understanding of the molecular interaction between GnRHs and their cognate receptors.
Gonadotropin-releasing hormone receptor (GnRHR), a rhodopsin-like G protein-coupled receptor (GPCR), is one of the most extensively studied receptors because of its dual significance both for understanding reproductive biology and for the development of medical therapies (Sealfon et al., 1997). It is now well-established that most vertebrates, including human, have at least two forms of GnRH (White et al., 1998; Fernald and White, 1999). One form, GnRH-I (also called mammalian GnRH), is primarily synthesized in the hypothalamus, whereas the other form, GnRH-II (also called chicken GnRH-II), is widely expressed in the brain and peripheral tissues. Although GnRH-I is known to regulate the secretion and synthesis of gonadotropins in the pituitary, the exact function of GnRH-II is largely unknown. The receptor for GnRH-I was first isolated from mammalian pituitary cells (Kaiser et al., 1992; Reinhart et al., 1992; Tsutsumi et al., 1992) and called mammalian type I GnRHR. Receptors that have a high affinity for GnRH-II have been identified in nonmammalian and mammalian species (Tensen et al., 1997; Illing et al., 1999; Millar et al., 2001; Neill et al., 2001; Wang et al., 2001; Bogerd et al., 2002; Seong et al., 2003). Mammalian type II GnRHR is closer in structure to nonmammalian GnRHRs than mammalian type I GnRHR. Mammalian type II GnRHR, like nonmammalian GnRHRs, contains the intracellular C-terminal tail, which is functionally important for desensitization and internalization (Heding et al., 1998; Willars et al., 1999), whereas mammalian type I GnRHR does not have a C-terminal tail. Mammalian type II and nonmammalian GnRHRs have Asp2.50 and Asp7.49 in the transmembrane helices (TMHs) 2 and 7, respectively, whereas mammalian type I GnRHRs contain Asp2.50 and Asn7.49, which are known to be important for receptor conformation and signal transduction (Blomenrhr et al., 1997; Mitchell et al., 1998).
Mammalian type II GnRHR has a higher affinity for GnRH-II than for GnRH-I, whereas the opposite is true for mammalian type I GnRHR. However, the factors that determine such differential ligand selectivity are poorly understood. Mutagenesis studies combined with computational modeling have identified a number of residues that are involved in ligand binding (Davidson et al., 1996; Flanagan et al., 2000; Hoffmann et al., 2000; Hvelmann et al., 2002). GnRH-II differs from GnRH-I by three amino acids at positions 5, 7, and 8; thus, searching for residues that may interact with them would help us to understand the mechanism underlying differential ligand selectivity. It has been proposed that Tyr5 and Leu7 of GnRH-I interact with Tyr6.58 and Trp2.64 of mammalian type I GnRHR (Hvelmann et al., 2002). However, because Tyr6.58 and Trp2.64 are also conserved in mammalian type II GnRHR, these residues alone cannot account for differential ligand selectivity. An acidic amino acid, Glu/Asp7.32 in EL3 of mammalian type I GnRHR is known to confer ligand specificity for GnRH-I by an electrostatic interaction with Arg8 of GnRH-I (Flanagan et al., 1994; Fromme et al., 2001). However, this is not fully explanatory, because some nonmammalian GnRHRs have an acidic amino acid (e.g., Glu7.32 for bfGnRHR-2 and Asp7.32 for catfish GnRHR) at this homologous position, yet these receptors respond better to GnRH-II than to GnRH-I (Wang et al., 2001). We have demonstrated that the positions of Ser and Pro flanking Glu/Asp7.32 are critical determinants for ligand selectivity (Wang et al., 2004). Replacement of the Ser-Glu-Pro (SEP) motif by Pro-Glu-Ser (PES) in mammalian type I GnRHR induced an increased sensitivity to GnRH-II but the opposite to GnRH-I. Moreover, mutation of a Ser-Gln-Ser (SQS) motif to SEP in bullfrog type I GnRHR (bfGnRHR-1) showed an increased sensitivity to GnRH-II but a decreased sensitivity to GnRH-I (Wang et al., 2004). However, this study found no reverse-ligand selectivity when the Pro-Glu-Tyr (PEY) motif in bfGnRHR-2 was replaced by SEP, suggesting the involvement of other residues in ligand selectivity.
Sequence alignments showed that the EL3-proximal TMH7 of mammalian type II GnRHR has a high degree of sequence identity with that of nonmammalian GnRHRs but not with that of mammalian type I GnRHR. In the present study, using rat GnRHR and gmGnRHR-2 as models for representative mammalian type I and type II GnRHRs, respectively, we addressed whether EL3 and/or EL3-proximal TMH7 determine differential ligand selectivity. Domain swapping and site-directed mutagenesis studies suggest that the Pro-Pro-Ser (PPS) motif in EL3 and Leu7.38, Leu7.43, Ala7.46, and Pro7.47 in TMH7 of gmGnRHR-2 are critical for discriminating GnRH-II from GnRH-I.
Materials and Methods
Materials. GnRH-I (pyro-Glu-His-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-GlyNH2), GnRH-II ([His5, Trp7, Tyr8]GnRH-I), [His5]GnRH-I, [Trp7]GnRH-I, [Trp8]GnRH-I, and [Trp7, Leu8]GnRH-I were synthesized by AnyGen (Gwangju, Korea). The c-fos-luc vector containing approximately -711 to +45 sequence of the human c-fos promoter constructed in the pFLASH vector was a kind gift from Dr. R. Prywes (Columbia University, New York, NY). Vent DNA polymerase was purchased from New England Biolabs (Beverly, MA). All oligonucleotides were synthesized from GenoTech (Daejon, Korea). GH3 cell lines stably expressing gmGnRHR-2 or rat GnRHR were established as described previously (Acharjee et al., 2002; Wang et al., 2003).
Amino Acid Residue Numbering Scheme. Amino acid residues are numbered according to their positions in gmGnRHR-2. To facilitate the comparison among different GnRHRs, the standard numbering system proposed by Ballesteros and Weinstein (1995) was also used.
Construction of Wild-Type and Mutant GnRHRs. The cDNA of gmGnRHR-2 subcloned into pcDNA3 (Invitrogen, Carlsbad, CA) at the KpnI and XbaI sites (Wang et al., 2003) was used as a template for creating domain-swapped or site-directed mutants. Domain-swapping and site-directed mutagenesis were performed by the polymerase chain reaction overlapping-extension method (Wang et al., 2003, 2004). To facilitate the construction of domain-swapped mutants, an exogenously introduced EcoRV site at the Asn7.34 residue and an intrinsic BstXI site or two intrinsic BamHI sites were used. EL3 or EL3-proximal TMH7 of rat GnRHR was amplified using a specific set of primers flanked by the overlapping sequence of gmGn-RHR-2 and the appropriate restriction endonuclease recognition site, producing rEL3S and rEL3L, respectively. The fragment from the N terminus to the EcoRV site at the Asn7.34 residue of rEL3S was replaced by the corresponding fragment of rat GnRHR, generating the r6TM chimera. Likewise, the fragment from the N terminus to the BamHI site at the Pro7.47 residue of gmGnRHR-2 was replaced by the corresponding part of rat GnRHR, producing the r6.5TM chimera. Mutated sequences were confirmed using the Sequenase Version 2.0 DNA Sequencing Kit (U.S. Biochemical Corporation, Cleveland, OH) according to the manufacturer's instructions.
Inositol Phosphate Production Assay. The inositol phosphate (IP) production assay was performed as described previously (Wang et al., 2003). GH3 cells (1 x 105/well) expressing gmGnRHR-2 or rat GnRHR were seeded in 12-well plates, and the following day, cells were incubated in inositol-free DMEM (Invitrogen) containing 2% dialyzed fetal bovine serum and labeled with 1 e藽i of [myo-3H]inositol/well (Amersham Biosciences UK, Ltd., Little Chalfont, Buckinghamshire, UK) for 18 h. Medium was then removed, and cells were washed with 0.5 ml of buffer A (140 mM NaCl, 20 mM HEPES, 4 mM KCl, 8 mM D-glucose, 1 mM MgCl2, 1 mM CaCl2, and 1 mg/ml fatty acid-free BSA). Cells were then preincubated with buffer A containing 10 mM LiCl for 15 min, followed by treatment with graded concentrations (0.01 nM to 10 e) of GnRHs at 37°C for 45 min. The reaction was terminated by removing the incubation medium and adding 0.5 ml of ice-cold 10 mM formic acid. After 30 min at 4°C, the formic acid extracts were transferred into columns containing Dowex anion exchange resin. Total IPs were then eluted with 1 ml of 1 M ammonium formate/0.1 M formic acid, and their radioactivity was determined.
Luciferase Assay. Wild-type and mutant GnRHRs were transiently transfected into CV-1 cells, which were maintained at 37°C in DMEM with 10% heat-inactivated fetal bovine serum, 1 mM glutamate, 100 U of penicillin, and 100 e/ml streptomycin. Cells were seeded in 24-well plates (1 x 105/well), and transfection was performed using the SuperFect transfection kit (QIAGEN, Valencia, CA) according to the manufacturer's instructions with a minor modification. For each transfection, 100 ng of each receptor cDNA, 200 ng of c-fos-luc vector, and 200 ng of internal control plasmid pCMV-Gal were used. One day after transfection, cells were serum-starved for 24 h and then challenged with GnRH for 6 h (Oh et al., 2003). Cells were harvested, and luciferase activity in the cell extract was determined according to standard methods in a Lumat LB9501 (Berthold Technologies, Bad Wildbad, Germany). The luciferase activities were normalized using -galactosidase values. Transfection experiments were performed in duplicate and were repeated three to five times.
Binding Assay. GnRH-II was radioiodinated using the chloramine-T method and purified by chromatography on a Sephadex G-25 (Sigma-Aldrich, St. Louis, MO) column in 0.01 M acetic acid and 0.1% BSA. HeLa cells were transfected with wild-type, individual mutant construct, or pcDNA3 (300 ng of DNA per well in 12-well plates) with Effectene (QIAGEN) according to the manufacturer's instructions. Thirty-six hours after transfection, intact cells were washed and incubated with binding buffer (DMEM supplemented with 0.1% BSA, pH 7.4) containing 250,000 cpm of 125I-GnRH-II (0.5 ml final volume) at 20°C for 1 h to achieve equilibrium. Specific binding was calculated by subtracting nonspecific binding (the presence of 10 e unlabeled GnRH-II) from total binding. For the displacement binding assay, 125I-GnRH-II was incubated in the presence of graded concentrations of unlabeled GnRH-I or GnRH-II.
Molecular Modeling. gmGnRHR-2 was built by MODELLER 6 version 2 (Sali and Blundell, 1993) on the basis of the crystal structure of bovine rhodopsin (Okada et al., 2002) as a template. GnRH-I or GnRH-II was docked onto putative binding sites of gmGnRHR-2 manually using Visual Molecular Dynamics (Humphrey et al., 1996): PyroGlu1 with Asn5.39, His2 with Asp2.61, Trp3 with Asn6.48, Tyr5 with Tyr6.58, and Gly10 with Asp2.61 and Asn2.65. The models for gmGnRHR-2/GnRH-II, mutant gmGnRHR-2/GnRH-I, and mutant gmGnRHR-2/GnRH-II were built by mutating corresponding residues in the gmGnRHR-2/GnRH-I model and underwent energy minimization and molecular dynamics annealing simulations in the MODELLER. The final models showing good geometry were confirmed by PROCHECK (Laskowski et al., 1993). The contacts between ligands and receptors were analyzed using Ligplot (Wallace et al., 1995). Figures of the models were drawn using Visual Molecular Dynamics (Humphrey et al., 1996).
Data Analysis. Analyses were performed using nonlinear regression, and the data were expressed as sigmoid dose-response curves. GnRH concentrations inducing half-maximal stimulation (EC50), halfmaximal inhibition (IC50), and maximal fold increases (Emax) were calculated using GraphPad Prism 3 software (GraphPad Software Inc., San Diego, CA). All data are presented as means ± S.E.M. The data were analyzed by one-way analysis of variance followed by the Bonferroni test. A p value <0.05 was considered statistically significant.
Results
Differential Ligand Selectivity of Mammalian and Nonmammalian GnRHRs. The ligand selectivities of rat GnRHR and gmGnRHR-2 were examined using two different methods: IP production, and c-fos promoter-driven luciferase (c-fos-luc) assays. For the IP assay, GH3 cells stably expressing rat GnRHR or gmGnRHR-2 were used (Wang et al., 2003; Maiti et al., 2003), and for c-fos-luc assay, CV-1 cells transiently expressing rat GnRHR or gmGnRHR-2 were used. As for rat GnRHR, GnRH-I showed a lower EC50 value than did GnRH-II, indicating that rat GnRHR has a higher sensitivity to GnRH-I than to GnRH-II. However, gmGnRHR-2 responded better to GnRH-II than did GnRH-I in both assay systems (Fig. 1 and Table 1). Regarding GnRHR, GnRH-I had a 7.4- and 5.9-fold higher potency than GnRH-II in IP and c-fos-luc assay systems, respectively. For gmGnRHR-2, GnRH-II was 204-fold (IP assay) and 239-fold (c-fos-luc assay) more potent than GnRH-I (Table 1). Because c-fos-luc was more sensitive than the IP assay system, we used the c-fos-luc system in ensuing experiments.
Values given form IP and c-fos-luc are log(EC50, M). Data represent the mean ± S.E.M. from three independent experiments. Numbers in parentheses represent fold difference in sensitivity between GnRH-I and GnRH-II.
EL3 and EL3-Proximal TMH7 Are Involved in Differential Ligand Selectivity. Sequence alignment showed that EL3-proximal TMH7 of mammalian type II GnRHR has a high degree of sequence identity with that of nonmammalian GnRHRs but not mammalian type I GnRHR. Furthermore, it was suggested that the proximal region of TMH7 of GnRHR affects the conformation of EL3 (Petry et al., 2002). We therefore presumed that both EL3 and the EL3-proximal TMH7 may be involved in differential ligand selectivity. To address this possibility, EL3 alone or together with EL3-proximal TMH7 of gmGnRHR-2 was swapped with that of rat GnRHR, designated rEL3S or rEL3L, respectively. Swapping EL3 alone did not induce a significant change in sensitivity for either GnRH-I or GnRH-II such that chimeric rEL3S, like wild-type gmGnRHR-2, showed a higher sensitivity to GnRH-II than to GnRH-I (Fig. 2A). It is interesting that rEL3L showed a great increase in sensitivity to GnRH-I but not to GnRH-II (Fig. 2B and Table 2), indicating that EL3-proximal TMH7 in gmGnRHR-2 is probably important for the discrimination between them. The functional importance of EL3-proximal TMH7 was further confirmed by additional chimeric receptors that have N termini to EL3 or to EL3-proxmial TMH7 of rat GnRHR, denoted r6TM or r6.5TM, respectively. The chimeric receptor r6TM, which has the EL3-proximal TMH7 sequence of gmGnRHR-2, has a high sensitivity to GnRH-II and a low sensitivity to GnRH-I, characteristics of a type II receptor (Fig. 2C). In contrast, r6.5TM containing the EL3-proximal TMH7 sequence of rat GnRHR has the ligand sensitivity, characteristic of a type I GnRHR (Fig. 2D), again confirming that EL3-proximal TMH7 in gmGnRHR-2 is critical for ligand selectivity.
Values represent the mean ± S.E.M. of three independent experiments performed in duplicate. Binding was expressed as a percentage of specific binding of gmGnRHR-2 (gm2).
Identification of rEL3L Amino Acids Involved in Ligand Selectivity. Because sequence alignment showed a six amino acid difference in EL3-proximal TMH7 between gmGnRHR-2 and rat GnRHR (Fig. 2E), we postulated that one of them may be responsible for ligand selectivity. Thus, six individual amino acids were reciprocally changed in the rEL3L chimeric receptor. Point mutation of Phe7.37 to isoleucine, Phe7.38 to leucine, or Ala7.42 to glycine did not induce significant changes in ligand selectivity compared with rEL3L (Fig. 3, A and B, and Table 2). Point mutation of Pro7.46 to alanine or Cys7.47 to proline completely suppressed receptor function in response to either GnRH-I or GnRH-II (Fig. 3, C and D). Finally, the mutation of Phe7.43 to leucine in rEL3L showed a significant decrease in sensitivity to GnRH-I (Fig. 3A).
It was shown previously that Glu7.32 of mouse GnRHR is a critical residue conferring ligand specificity for Arg8 of GnRH-I (Flanagan et al., 1994). Furthermore, we recently demonstrated that the positions of serine and proline flanking Glu7.32 are crucial for the ligand selectivity between mammalian and nonmammalian GnRHRs (Wang et al., 2004). Therefore, we sought to determine whether these amino acids are critical for the ligand selectivity in the chimeric receptor rEL3L. The Glu7.32 or SEP motif in rEL3L was changed to glycine or PPS, respectively. A mutation of Glu7.32 to glycine significantly decreased sensitivity to GnRH-I but not to GnRH-II (Fig. 3, C and D, and Table 2). Moreover, replacement of the SEP motif by PPS greatly increased sensitivity to GnRH-II, whereas it slightly decreased sensitivity to GnRH-I (Fig. 3, C and D, and Table 2). These results suggest that the SEP/PPS motif, together with amino acids in TMH7, is important in ligand sensitivity.
Identification of Amino Acids in Wild-Type gmGnRHR-2 That Confer Differential Ligand Selectivity. Because we observed that the amino acid residues Phe7.43, Pro7.46, and Cys7.47 in rEL3L critically affected receptor activation and ligand selectivity, we further examined the function of these residues in wild-type gmGnRHR-2. Two mutants, L7.43F and A7.46P, had no receptor activity (Fig. 4). The mutant P7.47C had essentially the same ligand selectivity as the wild-type gmGnRHR-2 (Fig. 4). Because we failed to observe reverse-ligand selectivity by a single mutation, we postulated that multiple amino acids are involved in ligand selectivity. To address this, double or triple mutants with different combinations of Leu7.43, Ala7.46, and Pro7.47 were constructed. It is interesting that a double mutant, A7.46P/P7.47C, exhibited an improvement in ligand selectivity for both GnRH-I and GnRH-II (Fig. 5, A and B). The double mutant L7.43F/P7.47C showed a decrease in receptor efficacy in both GnRH-I and GnRH-II but showed a slight increase in sensitivity for GnRH-I (Fig. 5, A and B). The double mutant L7.43F/A7.46P did not respond to GnRH stimulation (Fig. 5, A and B). A triple mutant, L7.43F/A7.46P/P7.47C, exhibited a large increase in ligand sensitivity for both GnRH-I and GnRH-II. Compared with wild-type gmGnRHR-2, the L7.43F/A7.46P/P7.47C mutant showed a 200-fold increased sensitivity to GnRH-I and a 20-fold increased sensitivity to GnRH-II (Fig. 5, A and B).
Because the PPS/SEP motif in EL3 and Leu7.43, Ala7.46, and Pro7.47 residues in TMH7 affect ligand selectivity, we examined a combinatory effect of these two motifs. The gmGnRHR-2 with SEP/L7.43F/A7.46P/P7.47C mutation revealed a slight decrease in sensitivity for both GnRH-I and GnRH-II (Fig. 5, C and D). This mutant also showed a decrease in Emax values for both GnRHs compared with the L7.43F/A7.46P/P7.47C mutant, which may be caused by low receptor expression (Table 2). It is interesting that additional mutations rEL3S/L7.43F/A7.46P/P7.47C (Fig. 7B), SEP/I7.37F/L.38F/L.43F/A7.46P/P7.47C (Fig. 5, C and D), or SEP/L7.38F/L7.43F/A7.46P/P7.47C (Fig. 5, C and D) increased sensitivity to GnRH-I and decreased sensitivity to GnRH-II (Table 2).
Ligand Binding Affinities. Ligand affinities of wild-type, rEL3L, and SEP/L7.38F/L7.43F/A7.46P/P7.47C mutants were determined using a competition binding assay. For the binding assay, HeLa cells were used because when transfected with the receptors, they have a much higher binding capacity than CV-1 cells. It should be noted that HeLa cells, in the c-fos-luc assay system, produce EC50 values similar to those of CV-1 cells when we applied the same receptor and ligand (data not shown). HeLa cells, however, have a high basal c-fos-luc activity; therefore, their fold increases are usually much lower than those in CV-1 cells (Oh et al., 2003). [125I]GnRH-II (250,000 cpm) was applied to HeLa cells expressing wild-type and mutant receptors in the presence of graded concentrations of unlabeled GnRH-I or GnRH-II. Log IC50 values for GnRH-I in cells expressing rEL3L (-7.87 ± 0.22) and SEP/L7.38F/L7.43F/A7.46P/P7.47C (-8.30 ± 0.11) were significantly lower compared with those in cells expressing the wild-type receptor (-6.69 ± 0.21) (Fig. 6A), indicating an increased affinity for GnRH-I in the mutant receptors. Log IC50 values for GnRH-II in cells expressing the wild-type, rEL3L, and SEP/L7.38F/L7.43F/A7.46P/P7.47C mutants were -9.58 ± 0.27, -10.19 ± 0.23, and -9.96 ± 0.27, respectively (Fig. 6B), showing that ligand affinities for GnRH-II in mutant receptors do not change as drastically as those for GnRH-I.
Relative ligand binding of mutant constructs was determined using [125I]GnRH-II in the absence or presence of unlabeled GnRH-II (10 e). For wild-type gmGnRHR-2, total and nonspecific binding were 2.3 ± 0.2% (5826 ± 50 cpm) and 0.60 ± 0.2% (1567 ± 30 cpm), respectively. Total binding for other mutant receptors ranged from 1.16 to 6.83%, whereas nonspecific bindings for other receptors were the same as that for the wild-type receptor. Mutants that did not respond to GnRHs (rEL3L/P313A, rEL3L/C314P, L7.43F, and A7.46P) were unable to bind radioiodinated GnRH-II. rEL3S, L7.43F/P7.47C, and SEP/L7.43F/A7.46P/P7.47C showed relatively low binding; rEL3L/PPS and A7.46P/P7.47C had higher binding than gmGnRHR-2. Other mutants exhibited 49.7 to 178.6% binding compared with wild-type gmGnRHR-2 (Table 2).
Ligand Sensitivity for Chimeric GnRHs. Natural and chimeric GnRHs, in which amino acids at positions 5, 7, and 8 were substituted, were used to examine ligand sensitivity of gmGnRHR-2, rEL3L, and SEP/L7.38F/L7.43F/A7.46P/P7.47C mutants. For wild-type gmGnRHR-2, all chimeric GnRHs ([His5]GnRH-I, [Trp7]GnRH-I, [Trp8]GnRH-I, and [Trp7, Leu8]GnRH-I) exhibited a higher potency than GnRH-I (Table 3). In particular, substitution of the amino acid residues at positions 7 and 8 of GnRH-I greatly increased potency to activate gmGnRHR-2. It should be noted that chimeric ligands [His5]GnRH-I and [Trp7]GnRH-I, which, like GnRH-I, retain Arg8, showed a 100- to 200-fold increased potency for either rEL3L or SEP/L7.38F/L7.43F/A7.46P/P7.47C that have enhanced sensitivity to GnRH-I (Fig. 7 and Table 3). [Trp8]GnRH-I revealed a 20- to 50-fold increased sensitivity for rEL3L and SEP/L7.38F/L7.43F/A7.46P/P7.47C compared with that for wild-type gmGnRHR-2. It is interesting that rEL3L and SEP/L7.38F/L7.43F/A7.46P/P7.47C had similar sensitivity to chimeric GnRHs. (Fig. 7 and Table 3). This result supports the idea that positions 7 and 8 in GnRH are important for conferring its specificity.
Data represent the mean ± S.E.M. of three independent experiments performed in duplicate.
Molecular Modeling. To support our biochemical data, we constructed models to simulate the interaction of GnRHs with wild-type gmGnRHR-2 and SEP/L7.38F/L7.43F/A7.46P/P7.47C mutant (Fig. 8). Overall, the models agree well with previous reports (Hvelmann et al., 2002; Wang et al., 2004): two cysteine residues (Cys113 and Cys188) of the receptors are close because they are involved in a disulfide bond; pGlu1 of the ligands formed hydrogen bonds with Asn5.39 of the receptors; Trp3 of the ligands was located in the aromatic cage formed by Trp6.48, Phe5.43, and Tyr6.52 of the receptors; Arg8 of GnRH-I formed an ionic interaction with Glu7.32 of the SEP/L7.38F/L7.43F/A7.46P/P7.47C mutant.
GnRH-II and GnRH-I differ by three residues: His/Tyr5, Trp/Leu7, and Tyr/Arg8. Trp7 of GnRH-II made a hydrophobic contact with Pro7.32 of gmGnRHR-2. Tyr8 had an interaction with Pro7.32 and His7.36 (Fig. 8A). In addition, Trp7 formed a hydrogen bond with the carbonyl oxygen of Val7.30, and Tyr8 did so with the backbone of Ser7.33. However, in the GnRH-I/gmGnRHR-2 complex, Leu7 formed a hydrophobic contact with Tyr6.58. Arg8 moved to EL2 and interacted with Val4.67 via a hydrophobic interaction (Fig. 8B). On the other hand, Trp7 of GnRH-II formed hydrophobic contacts with Glu7.32 and His7.36 of the SEP/L7.38F/L7.43F/A7.46P/P7.47C mutant, and Tyr8 made a hydrogen bond with the backbone nitrogen of Glu7.32 (Fig. 8C). In the complex of GnRH-I with the SEP/L7.38F/L7.43F/A7.46P/P7.47C mutant, Leu7 formed hydrophobic contact with Tyr6.58, and Arg8 had an ionic interaction with Glu7.32 (Fig. 8D). Mutation of the four amino acids at TMH7 altered intramolecular interactions. For instance, the hydrogen bond between Leu7.43 and Ala7.46 of gmGnRHR-2 was suppressed in the SEP/L7.38F/L7.43F/A7.46P/P7.47C mutant. In addition, novel hydrophobic contacts of Phe7.38 with Leu6.54, Phe7.43 with Asp2.61, and Pro7.46 with Leu7.44 were formed in the mutant receptor. It is also notable that the various intramolecular interactions were highly dependent on ligand type. Hydrophobic contacts of Phe7.43 with Leu1.42 and Val2.57, Cys7.47 with Gly1.49, and hydrogen bonds between Cys7.47 and Ser1.45 and Asn1.50 were present in the mutant receptor/GnRH-I complex, but these interactions were absent in the mutant receptor/GnRH-II complex. In contrast, the hydrophobic contacts of Phe7.38 with Leu6.53 and Pro7.46 with Val2.53 were present in the mutant receptor/GnRH-II complex.
Discussion
The present study demonstrates that replacement of EL3 and EL3-proximal TMH7 of gmGnRHR-2 with those of rat GnRHR greatly improves ligand sensitivity to GnRH-I but not to GnRH-II. Site-directed mutations on gmGnRHR-2 and back mutations on the domain-swapped receptor show that the PPS motif in EL3 and Leu7.38, Leu7.43, Ala7.46, and Pro7.47 in TMH7 of gmGnRHR-2 and the corresponding residues of rat GnRHR are responsible for differential ligand sensitivity to GnRH-I and GnRH-II.
It was suggested that not only Glu/Asp7.32 but also the positions of Ser and Pro flanking Glu/Asp7.32 in EL3 of mammalian type I GnRHR determine high selectivity for GnRH-I (Fromme et al., 2004; Wang et al., 2004). These findings indicate that a local conformation of EL3 is critical for differential ligand selectivity among nonmammalian and mammalian types I and II GnRHRs. However, replacement of EL3 from gmGnRHR-2 alone does not affect ligand selectivity to GnRH-I and GnRH-II. Likewise, substitution of SEP for the PEY motif of bfGnRHR-2 does not alter ligand sensitivity to GnRH-I and GnRH-II (Wang et al., 2004). These observations suggest that other amino acid residues/motifs are involved in the selectivity of GnRH. Our study strongly suggests that in mammalian type II GnRHR, EL3-proximal TMH7 in addition to EL3 participates in differential ligand selectivity. The importance of EL3-proximal TMH7 in ligand sensitivity is supported by the observation that rat GnRHR with the entire TMH7 of gmGnRHR-2 (r6TM) exhibits a significant decrease in sensitivity for GnRH-I. Mutations of Pro7.47 to cysteine combined with the mutation of Leu7.43 to phenylalanine and/or Ala7.46 to proline significantly increases sensitivity for GnRH-I, whereas mutations of a single amino acid residue at these positions does not affect ligand selectivity to GnRH-I, suggesting that the combination of each amino acid in TMH7 is critical for differential ligand selectivity.
On the basis of Millar's classification (Millar et al., 2004), we aligned the sequences of EL3 and proximal TMH7 of various GnRHR subtypes: human-1 and rat-1 for mammalian type I receptors, green monkey-2 and marmoset-2 for type II mammalian receptors, bullfrog-3 and Xenopus-2 for nonmammalian type II receptors, Japanese medaka-1 and bullfrog-2 for type III receptors, and finally Japanese medaka-2, bullfrog-2, and catfish-1 for nonmammalian type I receptors. Leu7.40, Leu7.44, and Asn7.45 residues are conserved between gmGnRHR-2 and the mammalian type I receptors, but these residues are also largely conserved in many other nonmammalian GnRHRs. Thus, these amino acid residues are not specific to mammalian GnRHR subtypes. After excluding the amino acids that are conserved throughout the GnRHR subtypes, we found that at least four residues, Iso7.37, Leu7.38, Gly7.42, and Leu7.43, in gmGn-RHR-2 are different from those in mammalian type I receptors but are highly conserved in nonmammalian GnRHRs (Fig. 9). Among these, Iso7.37 and Gly7.42 are not likely to contribute to GnRH-I selectivity, because back-mutations of these residues in the rEL3L receptor did not significantly affect sensitivity to GnRH-I. Two amino acids, Ala7.47 and Pro7.48, in gmGnRHR-2 are different from either those in type I mammalian GnRHR or in nonmammalian GnRHRs. Thus, they are unique to the type II mammalian receptors.
Double or triple mutations of Leu7.43, Ala7.46, and Pro7.47 in TMH7 significantly increase ligand sensitivity to GnRH-I. We cannot explain clearly how the combined mutation L7.38F/A7.46P/P7.47C increases GnRH-I sensitivity. It is unlikely that these residues have direct interactions with GnRH-I because they are deeply buried in the three-dimensional structure. Rather, the mutation on these residues may play a role in modulating conformation of the binding pocket in EL3. Our molecular modeling data consistently show no direct interactions of these residues with the ligand. It is of interest to note that inter- and intramolecular interactions of the mutant receptor could be modified by the ligand type applied, indicating that conformational changes in these residues may be closely related to the alteration in the ligand binding pocket of EL3.
Amino acids at positions 7.46 and 7.47 in TMH7 seem critical for receptor conformation and stability. The mutation of Ala7.46 to proline in wild-type gmGnRHR-2 and the mutation of Pro7.46 to alanine or Cys7.47 to proline in rEL3L impair receptor responsiveness. Extremely low binding of these mutant receptors to GnRH suggests that this impairment can be ascribed to the loss of binding activity or receptor stability. It is known that a proline residue leads to a local constraint on the polypeptide chain conformation because of its pyrrolidine ring structure. Thus, proline at a proper position in TMH7 seems to be important for receptor conformation/stability in wild-type and mutant gmGnRHR-2. The occurrence of two successive proline residues found in the rEL3L/C7.47P and A7.46P mutants might disrupt receptor conformation/stability because the loss of responsiveness of A7.46P mutant can be rescued by a double mutation (A7.46P/P7.47C). Furthermore, Pro7.46 in the SEP/L7.38F/L7.43F/A7.46P/P7.47C mutant has an intramolecular contact with Leu7.44, which is different from that of the wild-type receptor in which Ala7.46 at the same position has a hydrophobic interaction with Leu7.43. Pro7.47 alone in gmGnRHR-2 could not critically alter receptor conformation/stability because the mutation of Pro7.47 to cysteine did not affect ligand potency. The cysteine residue at position 7.47 is highly conserved in nonmammalian and mammalian type I GnRHRs, as well as in many other GPCRs, except for gmGnRHR-2 (Fig. 9). Thus, it may be possible that mutation of Pro7.47 to cysteine helps to form a more stable conformation. Mutation of Leu7.43 to phenylalanine in gmGnRHR-2 abolishes the receptor responsiveness to ligand, which can be rescued by a double mutation (L7.43F/P7.47C). In a three-dimensional structure, Leu7.43 and Pro7.47 are spatially very close. Therefore, it is postulated that a mutation of Leu7.43 to phenylalanine might cause a steric hindrance, which can be reversed by a further replacement of Pro7.47 to cysteine. Moreover, double mutations (L7.43F/P7.47C or A7.46P/P7.47C) not only rescue the activity of the L7.43F or A7.46P mutant, but also significantly increase the ligand sensitivity to both GnRH-I and GnRH-II. Furthermore, an approximately 100-fold increase in sensitivity toward GnRH-I was observed in the triple mutant L7.43F/A7.46P/P7.47C compared with that of wild-type gmGnRHR-2. Such an increase in sensitivity to ligands suggests that this motif is crucially involved in receptor activation. Thus, it is likely that Pro7.47 in the wild-type gmGnRHR-2 and Pro7.46 in mutant receptors are involved in TMH movements, contributing to GPCR activation/inactivation by forming molecular hinges or swivels (Sansom and Weinstein, 2000; Stitham et al., 2002).
It is noteworthy that replacement of EL3 and EL3-proximal TMH7 or mutations of amino acids in these regions did not decrease sensitivity to GnRH-II; instead, there was a slightly increased sensitivity to GnRH-II. It is well known that an acidic amino acid at position 7.32 in EL3 is required for high-affinity binding with Arg8 of GnRH-I. It seems that such an acidic residue also plays a certain role in interaction with Tyr8 of GnRH-II. Using a molecular model, Blomenrhr et al. (2002) suggested that Tyr8 of GnRH-II interacts with Glu7.32 in EL3 of the catfish GnRHR. Our molecular model also consistently showed that Tyr8 of GnRH-II has contact with Glu7.32 of the SEP/L7.38F/L7.43F/A7.46P/P7.47C mutant. It should be noted that Arg8 of GnRH-I did not interact with the PPS motif of wild-type gmGnRHR-2, whereas Tyr8 of GnRH-II had contact with Pro7.32 of wild-type gmGnRHR-2. Thus, at least in the gmGnRHR-2 structure, GnRH-II may not discriminate the receptor with the PPS motif from the receptor with the SEP motif.
Substitution of histidine for Tyr5 of GnRH-I did not alter its potency to activate wild-type gmGnRHR-2, rEL3L, or SEP/L7.38F/L7.43F/A7.46P/P7.47C, suggesting that position 5 of GnRH does not largely contribute to receptor-ligand interaction. This result is consistent with previous reports (Blomenrhr et al., 2002; Wang et al., 2003). Substitution of tryptophan for Leu7 or Arg8 in GnRH-I significantly increased the ability to activate gmGnRHR-2, indicating the importance of positions 7 and 8 in recognition of mammalian type II GnRHR. The SEP/L7.38F/L7.43F/A7.46P/P7.47C mutant showed ligand sensitivity similar to that of rEL3L, implying that multiple residues are required for distinguishing GnRH-II from GnRH-I. It is noteworthy that gmGnRHR-2 shows a high sequence identity in EL3-proximal TMH7 with nonmammalian GnRHRs but a relatively low sequence identity with that of mammalian type I GnRHR (Fig. 9). The evolutionary divergence of EL3 and TMH7 between mammalian types I and II GnRHR, therefore, may confer the differential selectivity toward GnRH-I and GnRH-II.
In summary, our studies demonstrate that EL3 and EL3-proximal TMH7 are responsible for differential ligand selectivity between mammalian types I and II GnRHRs. The elucidation of specific domains responsible for ligand selectivity may facilitate the understanding of ligand and receptor molecular coevolution, the mechanism of ligand-mediated GnRHR activation, and the development of novel drugs.
References
Acharjee S, Maiti K, Soh JM, Im WB, Seong JY, and Kwon HB (2002) Differential desensitization and internalization of three different bullfrog gonadotropin-releasing hormone receptors. Mol Cells 14: 101-107.
Ballesteros JA and Weinstein H (1995) Integrated methods for the construction of three-dimensional models and computational probing of structure-function relations in G protein-coupled receptors. Methods Neurosci 25: 366-428.
Blomenrhr M, Bogerd J, Leurs R, Schulz RW, Tensen CP, Zandbergen MA, and Goos HJ (1997) Differences in structure-function relations between nonmammalian and mammalian gonadotropin-releasing hormone receptors. Biochem Biophys Res Commun 238: 517-522.
Blomenrhr M, ter Laak T, Kuhne R, Beyermann M, Hund E, Bogerd J, and Leurs R (2002) Chimaeric gonadotropin-releasing hormone (GnRH) peptides with improved affinity for the catfish (Clarias gariepinus) GnRH receptor. Biochem J 361: 515-523.
Bogerd J, Diepenbroek WB, Hund E, van Oosterhout F, Teves AC, Leurs R, and Blomenrhr M (2002) Two gonadotropin-releasing hormone receptors in the African catfish: no differences in ligand selectivity, but differences in tissue distribution. Endocrinology 143: 4673-4682.
Davidson JS, McArdle CA, Davies P, Elario R, Flanagan CA, and Millar RP (1996) Asn102 of the gonadotropin-releasing hormone receptor is a critical determinant of potency for agonists containing C-terminal glycinamide. J Biol Chem 271: 15510-15514.
Fernald RD and White RB (1999) Gonadotropin-releasing hormone genes: phylogeny, structure and functions. Front Neuroendocrinol 20: 224-240.
Flanagan CA, Becker II, Davidson JS, Wakefield IK, Zhou W, Sealfon SC, and Millar RP (1994) Glutamate 301 of the mouse gonadotropin-releasing hormone receptor confers specificity for arginine 8 of mammalian gonadotropin-releasing hormone. J Biol Chem 269: 22636-22641.
Flanagan CA, Rodic V, Konvicka K, Yuen T, Chi L, Rivier JE, Millar RP, Weinstein H, and Sealfon SC (2000) Multiple interactions of the Asp(2.6198) side chain of the gonadotropin-releasing hormone receptor contribute differentially to ligand interaction. Biochemistry 39: 8133-8141.
Fromme BJ, Katz AA, Millar RP, and Flanagan CA (2004) Pro7.33303 of the human GnRH receptor regulates selective binding of mammalian GnRH. Mol Cell Endocrinol 219: 47-59.
Fromme BJ, Katz AA, Roeske RW, Millar RP, and Flanagan CA (2001) Role of aspartate7.32302 of the human gonadotropin-releasing hormone receptor in stabilizing a high-affinity ligand conformation. Mol Pharmacol 60: 1280-1287.
Heding A, Vrecl M, Bogerd J, McGregor A, Sellar R, Taylor PL, and Eidne KA (1998) Gonadotropin-releasing hormone receptors with intracellular carboxyl-terminal tails undergo acute desensitization of total inositol phosphate production and exhibit accelerated internalization kinetics. J Biol Chem 273: 11472-11477.
Hoffmann SH, ter Laak T, Kuhne R, Reilander H, and Beckers T (2000) Residues within transmembrane helices 2 and 5 of the human gonadotropin-releasing hormone receptor contribute to agonist and antagonist binding. Mol Endocrinol 14: 1099-1115.
Hvelmann S, Hoffmann SH, Kuhne R, ter Laak T, Reilander H, and Beckers T (2002) Impact of aromatic residues within transmembrane helix 6 of the human gonadotropin-releasing hormone receptor upon agonist and antagonist binding. Biochemistry 41: 1129-1136.
Humphrey W, Dalke A, and Schulten K (1996) VMD: visual molecular dynamics. J Mol Graph 14: 33-38.
Illing N, Troskie BE, Nahorniak CS, Hapgood JP, Peter RE, and Millar RP (1999) Two gonadotropin-releasing hormone receptor subtypes with distinct ligand selectivity and differential distribution in brain and pituitary in the goldfish (Carassius auratus). Proc Natl Acad Sci USA 96: 2526-2531.
Kaiser UB, Zhao D, Cardona GR, and Chin WW (1992) Isolation and characterization of cDNAs encoding the rat pituitary gonadotropin-releasing hormone receptor. Biochem Biophys Res Commun 189: 1645-1652.
Laskowski RA, Macarthur MW, Moss DS, and Thornton JM (1993) PROCHECK: a program to check the stereochemical quality of protein structures. J Appl Crystallogr 26: 283-291.
Maiti K, Li JH, Wang AF, Acharjee S, Kim WP, Im WB, Kwon HB, and Seong JY (2003) GnRH-II analogs for selective activation and inhibition of non-mammalian and type-II mammalian GnRH receptors. Mol Cells 16: 173-179.
Millar RP, Lowe S, Conklin D, Pawson A, Maudsley S, Troskie B, Ott T, Millar M, Lincoln G, Sellar R, et al. (2001) A novel mammalian receptor for the evolutionarily conserved type II GnRH. Proc Natl Acad Sci USA 98: 9636-9641.
Millar RP, Lu Z-L, Pawson AJ, Flanagan CA, Morgan K, and Maudsley SR (2004) Gonadotropin-releasing hormone receptors. Endocrine Rev 25: 235-275.
Mitchell R, McCulloch D, Lutz E, Johnson M, MacKenzie C, Fennell M, Fink G, Zhou W, and Sealfon SC (1998) Rhodopsin-family receptors associate with small G proteins to activate phospholipase D. Nature (Lond) 392: 411-414.
Neill JD, Duck LW, Sellers JC, and Musgrove LC (2001) A gonadotropin-releasing hormone (GnRH) receptor specific for GnRH II in primates. Biochem Biophys Res Commun 282: 1012-1018.
Oh DY, Wang L, Ahn RS, Park JY, Seong JY, and Kwon HB (2003) Differential G protein coupling preference of mammalian and nonmammalian gonadotropin-releasing hormone receptors. Mol Cell Endocrinol 205: 89-98.
Okada T, Fujiyoshi Y, Silow M, Navarro J, Landau EM, and Shichida Y (2002) Functional role of internal water molecules in rhodopsin revealed by X-ray crystallography. Proc Natl Acad Sci USA 99: 5982-5987.
Petry R, Craik D, Haaima G, Fromme B, Klump H, Kiefer W, Palm D, and Millar R (2002) Secondary structure of the third extracellular loop responsible for ligand selectivity of a mammalian gonadotropin-releasing hormone receptor. J Med Chem 45: 1026-1034.
Reinhart J, Mertz LM, and Catt KJ (1992) Molecular cloning and expression of cDNA encoding the murine gonadotropin-releasing hormone receptor. J Biol Chem 267: 21281-21284.
Sali A and Blundell TL (1993) Comparative protein modelling by satisfaction of spatial restraints. J Mol Biol 234: 779-815.
Sansom MS and Weinstein H (2000) Hinges, swivels and switches: the role of prolines in signalling via transmembrane alpha-helices. Trends Pharmacol Sci 21: 445-451.
Sealfon SC, Weinstein H, and Millar RP (1997) Molecular mechanisms of ligand interaction with the gonadotropin-releasing hormone receptor. Endocr Rev 18: 180-205.
Seong JY, Wang L, Oh DY, Yun O, Maiti K, Li JH, Soh JM, Choi HS, Kim K, Vaudry H, et al. (2003) Ala/Thr201 in extracellular loop 2 and Leu/Phe290 in transmembrane domain 6 of type 1 frog gonadotropin-releasing hormone receptor confer differential ligand sensitivity and signal transduction. Endocrinology 144: 454-466.
Stitham J, Martin KA, and Hwa J (2002) The critical role of transmembrane prolines in human prostacyclin receptor activation. Mol Pharmacol 61: 1202-1210.
Tensen C, Okuzawa K, Blomenrohr M, Rebers F, Leurs R, Bogerd J, Schulz R, and Goos H (1997) Distinct efficacies for two endogenous ligands on a single cognate gonadoliberin receptor. Eur J Biochem 243: 134-140.
Tsutsumi M, Zhou W, Millar RP, Mellon PL, Roberts JL, Flanagan CA, Dong K, Gillo B, and Sealfon SC (1992) Cloning and functional expression of a mouse gonadotropin-releasing hormone receptor. Mol Endocrinol 6: 1163-1169.
Wallace AC, Laskowski RA, and Thornton JM (1995) LIGPLOT: a program to generate schematic diagrams of protein-ligand interactions. Prot Eng 8: 127-134.
Wang AF, Li JH, Maiti K, Kim WP, Kang HM, Seong JY, and Kwon HB (2003) Preferential ligand selectivity of the monkey type II gonadotropin-releasing hormone (GnRH) receptor for GnRH-2 and its analogs. Mol Cell Endocrinol 209: 33-42.
Wang C, Yun O, Maiti K, Oh DY, Kim KK, Chae CH, Lee CJ, Seong JY, and Kwon HB (2004) Position of Pro and Ser near Glu7.32 in the extracellular loop 3 of mammalian and nonmammalian gonadotropin-releasing hormone (GnRH) receptors is a critical determinant for differential ligand selectivity for mammalian GnRH and chicken GnRH-II. Mol Endocrinol 18: 105-116.
Wang L, Bogerd J, Choi HS, Seong JY, Soh JM, Chun SY, Blomenrhr M, Troskie BE, Millar RP, Yu WH, et al. (2001) Three distinct types of GnRH receptor characterized in the bullfrog. Proc Natl Acad Sci USA 98: 361-366.
White RB, Eisen JA, Kasten TL, and Fernald RD (1998) Second gene for gonadotropin-releasing hormone in humans. Proc Natl Acad Sci USA 95: 305-309.
Willars GB, Heding A, Vrecl M, Sellar R, Blomenrhr M, Nahorski SR, and Eidne KA (1999) Lack of a C-terminal tail in the mammalian gonadotropin-releasing hormone receptor confers resistance to agonist-dependent phosphorylation and rapid desensitization. J Biol Chem 274: 30146-30153.