Portal de Periódicos da CAPES

Model of Glycoprotein Hormone Receptor Ligand Binding and Signaling

2004; Elsevier BV; Volume: 279; Issue: 43 Linguagem: Inglês

10.1074/jbc.m406948200

1083-351X

William R. Moyle, Yongna Xing, Win Lin, Donghui Cao, Rebecca V. Myers, John E. Kerrigan, Michael P. Bernard,

Regulation of Appetite and Obesity

Studies described here were initiated to develop a model of glycoprotein hormone receptor structure and function. We found that the region that links the lutropin receptor leucine-rich repeat domain (LRD) to its transmembrane domain (TMD) has substantial roles in ligand binding and signaling, hence we term it the signaling specificity domain (SSD). Theoretical considerations indicated the short SSDs in marmoset lutropin and salmon follitropin receptors have KH domain folds. We assembled models of lutropin, follitropin, and thyrotropin receptors by aligning models of their LRD, TMD, and shortened SSD in a manner that explains how substitutions in follitropin and thyrotropin receptors distant from their apparent ligand binding sites enable them to recognize lutropins. In these models, the SSD is parallel to the concave surface of the LRD and makes extensive contacts with TMD outer loops 1 and 2. The LRD appears to contact TMD outer loop 3 and a few residues in helices 1, 5, 6, and 7. We propose that signaling results from contacts of the ligands with the SSD and LRD that alter the LRD, which then moves TMD helices 6 and 7. The positions of the LRD and SSD support the notion that the receptor can be activated by hormones that dock with these domains in either of two different orientations. This would account for the abilities of some ligands and ligand chimeras to bind multiple receptors and for some receptors to bind multiple ligands. This property of the receptor may have contributed significantly to ligand-receptor co-evolution. Studies described here were initiated to develop a model of glycoprotein hormone receptor structure and function. We found that the region that links the lutropin receptor leucine-rich repeat domain (LRD) to its transmembrane domain (TMD) has substantial roles in ligand binding and signaling, hence we term it the signaling specificity domain (SSD). Theoretical considerations indicated the short SSDs in marmoset lutropin and salmon follitropin receptors have KH domain folds. We assembled models of lutropin, follitropin, and thyrotropin receptors by aligning models of their LRD, TMD, and shortened SSD in a manner that explains how substitutions in follitropin and thyrotropin receptors distant from their apparent ligand binding sites enable them to recognize lutropins. In these models, the SSD is parallel to the concave surface of the LRD and makes extensive contacts with TMD outer loops 1 and 2. The LRD appears to contact TMD outer loop 3 and a few residues in helices 1, 5, 6, and 7. We propose that signaling results from contacts of the ligands with the SSD and LRD that alter the LRD, which then moves TMD helices 6 and 7. The positions of the LRD and SSD support the notion that the receptor can be activated by hormones that dock with these domains in either of two different orientations. This would account for the abilities of some ligands and ligand chimeras to bind multiple receptors and for some receptors to bind multiple ligands. This property of the receptor may have contributed significantly to ligand-receptor co-evolution. Several models have been devised to account for the interactions of hCG 1The abbreviations used are: hCG, human choriogonadotropin; hLH, human lutropin; bLH, bovine lutropin; hFSH, human follitropin; hTSH, human thyrotropin; LHR, LH receptor; FSHR, follicle-stimulating hormone receptor; TSHR, thyroid-stimulating hormone receptor; LRD, leucine-rich repeat domain; SSD, signaling specificity domain; TMD, transmembrane domain; IBMX, isobutylmethylxanthine; CHO, Chinese hamster ovary. and other glycoprotein hormones with their receptors, membrane proteins that contain a ligand binding NH2-terminal extracellular domain, a TMD consisting of seven-membrane spanning helices, and a cytosolic COOH-terminal domain (1McFarland K.C. Sprengel R. Phillips H.S. Kohler M. Rosemblit N. Nikolics K. Segaloff D.L. Seeburg P.H. Science. 1989; 245: 494-499Crossref PubMed Scopus (808) Google Scholar, 2Loosfelt H. Misrahi M. Atger M. Salesse R. M. T. Vu Hai Luu Thi Jolivet A. Guiochon Mantel A. Sar S. Jallal B. Garnier J. Milgrom E. Science. 1989; 245: 525-528Crossref PubMed Scopus (516) Google Scholar). The NH2-terminal three-fourths of the LHR extracellular domain, which we term the LRD, contains several leucine-rich repeats that are likely to give it a curved shape similar in structure to other leucine-rich repeat proteins such as the SCF ubiquitin ligases (3Schulman B.A. Carrano A.C. Jeffrey P.D. Bowen Z. Kinnucan E.R. Finnin M.S. Elledge S.J. Harper J.W. Pagano M. Pavletich N.P. Nature. 2000; 408: 381-386Crossref PubMed Scopus (493) Google Scholar). As shown here, the remaining quarter of the extracellular domain also makes important contributions to ligand binding and signaling and for this reason we refer to it as the SSD or signaling specificity domain (SSD). The amino acid sequence of the SSD is not similar to any known protein and its structure has not been modeled. The TMD appears to be similar in conformation to bovine rhodopsin (4Palczewski K. Kumasaka T. Hori T. Behnke C.A. Motoshima H. Fox B.A. LeTrong I. Teller D.C. Okada T. Stenkamp R.E. Yamamoto M. Miyano M. Science. 2000; 289: 739-745Crossref PubMed Scopus (5056) Google Scholar), but it was not known how it is coupled to the LRD and TMD. The manner in which ligands interact with these receptors has been controversial. One view suggests the α-subunit COOH terminus and the small seatbelt loop contact the concave surface of the LRD such that the ends of loops α1/α3 and β1/β3 are exposed (5Jiang X. Dreano M. Buckler D.R. Cheng S. Ythier A. Wu H. Hendrickson W.A. Tayar N.E. el Tayar N. Structure. 1995; 3: 1341-1353Abstract Full Text Full Text PDF PubMed Scopus (176) Google Scholar). This model does not provide an obvious means by which ligand binding results in signal transduction. Portions of the hormone (6Ji I. Ji T.H. J. Biol. Chem. 1991; 266: 14953-14957Abstract Full Text PDF PubMed Google Scholar), including α-subunit loops 1 and/or 3 (7Remy J.-J. Couture L. Pantel J. Haertle T. Rabesona H. Bozon V. Pajot-Augy E. Robert P. Troalen F. Salesse R. Bidart J-M. Mol. Cell. Endocrinol. 1996; 125: 79-91Crossref PubMed Scopus (33) Google Scholar), have also been suggested to contact the TMD. This view implies that the extracellular domain snares the ligand and delivers it to the transmembrane domain. We had proposed that interactions of the ligand with the LRD and SSD are needed for signaling (8Moyle W.R. Campbell R.K. Rao S.N.V. Ayad N.G. Bernard M.P. Han Y. Wang Y. J. Biol. Chem. 1995; 270: 20020-20031Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar, 9Bernard M.P. Myers R.V. Moyle W.R. Biochem. J. 1998; 335: 611-617Crossref PubMed Scopus (24) Google Scholar), even though the LRD by itself is sufficient for hCG binding to the LHR (10Braun T. Schofield P.R. Sprengel R. EMBO J. 1991; 10: 1885-1890Crossref PubMed Scopus (314) Google Scholar). A key postulate of our original model, namely that the groove between hormone loops α2 and β1/β3 contacts the rim of the LRD to form a high affinity binding site, is no longer tenable. We have found that portions of loop α2 facing this groove are unlikely to participate in high affinity rat LHR contacts, even though they appear to be near the hormone-receptor interface (75Xing Y. Lin W. Jiang M. Cao D. Myers R.V. Bernard M.P. Moyle W.R. J. Biol. Chem. 2004; 279: 44427-44437Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar). Observations described here suggest that β-subunit loops 1 and 3 of lutropins contact the SSD domain rather than the LRD. The SSD is the least understood region of the receptor extracellular domain. This portion of the human LHR is largely responsible for its ability to distinguish hCG and bovine LH (9Bernard M.P. Myers R.V. Moyle W.R. Biochem. J. 1998; 335: 611-617Crossref PubMed Scopus (24) Google Scholar), indicating that it contacts the ligand. The SSD may have a role in lutropin signaling as shown by the finding that some mutations increase the basal activity of the LHR (11Nakabayashi K. Kudo M. Kobilka B. Hsueh A.J. J. Biol. Chem. 2000; 275: 30264-30271Abstract Full Text Full Text PDF PubMed Scopus (94) Google Scholar). To increase the probability that we could deduce a structure for the SSD, we limited our studies to the smallest natural SSDs known, namely those of the marmoset LHR (12Zhang F.P. Rannikko A.S. Manna P.R. Fraser H.M. Huhtaniemi I. Endocrinology. 1997; 138: 2481-2490Crossref PubMed Scopus (60) Google Scholar) and the salmon FSHR (13Oba Y. Hirai T. Yoshirua Y. Yoshikuni M. Kawauchi H. Nagahama Y. Biochem. Biophys. Res. Commun. 1999; 265: 366-371Crossref PubMed Scopus (100) Google Scholar). The SSD of the marmoset receptor lacks residues derived from exon 10, which may be responsible for its unusual ability to distinguish hCG and LH (14Gromoll J. Wistuba J. Terwort N. Godmann M. Muller T. Simoni M. Biol. Reprod. 2003; 69: 75-80Crossref PubMed Scopus (63) Google Scholar). Human LHRs that lack exon 10 also respond better to hCG than to LH (15Muller T. Gromoll J. Simoni M. J. Clin. Endocrinol. Metab. 2003; 88: 2242-2249Crossref PubMed Scopus (94) Google Scholar). Because the SSD of the human LHR can limit its ability to bind hormones such as bLH (9Bernard M.P. Myers R.V. Moyle W.R. Biochem. J. 1998; 335: 611-617Crossref PubMed Scopus (24) Google Scholar), we concluded that the primate LHR might have a conformation that constrains ligand binding in a unique manner. Reasoning that a marmoset analog of the rat LHR, which interacts with lutropins from many species, would be a more useful tool to study the role of the SSD in hormone function, we characterized a rat LHR analog that lacks residues encoded by exon 10 and that has a histidine in place of Cys-314. It also contains a tyrosine in place of Trp-307, a change that appeared to increase receptor expression, and an NH2-terminal FLAG epitope tag. As shown here, these changes altered the ability of the rat LHR to respond to bLH, hCG partial agonists, and other hCG analogs in ways that provided new insights into receptor function. The small size of these SSD enabled us to develop models of the glycoprotein hormone receptors that are consistent with most data on ligand binding and signaling, including the finding that only two mutations of the FSHR are needed to enable it to interact with hFSH and hCG (16Vassart G. Pardo L. Costagliola S. Trends Biochem. Sci. 2004; 29: 119-126Abstract Full Text Full Text PDF PubMed Scopus (298) Google Scholar). Models described here will also explain many features of the TSHR, including its subunit nature (17Tanaka K. Chazenbalk G.D. McLachlan S.M. Rapoport B. J. Biol. Chem. 1999; 274: 33979-33984Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar), high basal activity (18Chen C.R. Chazenbalk G.D. McLachlan S.M. Rapoport B. Endocrinology. 2003; 144: 3821-3827Crossref PubMed Scopus (27) Google Scholar, 19Chen C.R. Chazenbalk G.D. McLachlan S.M. Rapoport B. Endocrinol. 2003; 144: 1324-1330Crossref PubMed Scopus (15) Google Scholar), and ability to be stimulated by anti-receptor antibodies (20Rapoport B. Chazenbalk G.D. Jaume J.C. McLachlan S.M. Endocr. Rev. 1998; 19: 673-716Crossref PubMed Scopus (517) Google Scholar). The sources of hCG and antibodies used in these studies have been described (8Moyle W.R. Campbell R.K. Rao S.N.V. Ayad N.G. Bernard M.P. Han Y. Wang Y. J. Biol. Chem. 1995; 270: 20020-20031Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar, 21Cosowsky L. Rao S.N.V. Macdonald G.J. Papkoff H. Campbell R.K. Moyle W.R. J. Biol. Chem. 1995; 270: 20011-20019Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar). hLH was obtained from Dr. Robert Campbell (Serono Reproductive Biology Institute, Rockland, MA). hCG-αδ2, an analog of hCG lacking the α-subunit loop 2 oligosaccharide was prepared by N-glycanase digestion of hCG after the subunits had been dissociated and permitted to recombine (22Xing Y. Moyle W.R. Biochem. Biophys. Res. Commun. 2003; 303: 201-205Crossref PubMed Scopus (10) Google Scholar). Other constructs encoding hormone analogs (Fig. 1) were produced by PCR. These included a disulfide cross-linked analog of hCG (α5-β8hCG) in which α-subunit residue Gln5 and β-subunit residue Arg8 were converted to cysteines, an analog of α5-β8hCG lacking the α-subunit glycosylation signal at loop 2 residue 52 (α5-β8hCG-αN52D), an analog of α5-β8hCG having a glycosylation signal at β-subunit loop 3 residue 77 made by converting βPro78 and βVal79 to serine and threonine (α5-β8hCG-βCHO77), and an analog of α5-β8hCG-αN52D having the β-subunit loop 3 glycosylation signal (α5-β8hCG-αN52DβCHO77). PCR mutagenesis was also used to construct rLHRδ10, an analog of the native rat LHR in which residue Glu267 was joined to Tyr295, residue Trp307 was converted to tyrosine, and residue Cys314 was converted to histidine. (Note, the numbering system used in these studies reflects the absence of the presumed signal peptide, which we assumed to be 26, 22, 17, and 21 residues for the rat LHR, human LHR, rat FSHR, and rat TSHR, respectively.) The hormone and receptor constructs were subcloned into the polylinker of pCI (Promega, Madison, WI), a mammalian expression vector that had been modified as described (23Myers R.V. Wang Y. Moyle W.R. Biochim. Biophys. Acta. 2000; 1475: 390-394Crossref PubMed Scopus (10) Google Scholar) and expressed transiently in COS-7 cells or stably in Chinese hamster ovary cells. Radioiodinated hCG and monoclonal antibodies were prepared using IODO-GEN (Pierce) as described (24Bernard M.P. Cao D. Myers R.V. Moyle W.R. Anal. Biochem. 2004; 327: 278-283Crossref PubMed Scopus (16) Google Scholar). Ligand binding was monitored by quantifying the abilities of hCG and bLH to compete with 125I-hCG for binding to the LHR and LHRδ10 on intact cells in physiological buffers (9Bernard M.P. Myers R.V. Moyle W.R. Biochem. J. 1998; 335: 611-617Crossref PubMed Scopus (24) Google Scholar). The total volume of the assay was 100 μl. Cyclic AMP accumulation was monitored by radioimmuno-assay using a rabbit cyclic AMP antibody (Strategic Biosolutions, Ramona, CA) and 2′-O-monosuccinyladenosine 3′5′-cyclic monophosphate tyrosine methylester (Sigma) that was radioiodinated as described (25Brooker J. Harper J.F. Terasaki W.L. Moylan R.D. Adv. Cyclic Nucleotide Res. 1979; 10: 1-33PubMed Google Scholar). The total volume of this assay was 60 μl. Statistical analyses were performed using Prism (GraphPad Software, San Diego CA). Protein threading was performed with the molecular modeling packages Look (26Lee C. Irizarry K. IBM Systems J. 2001; 40: 592-603Crossref Scopus (35) Google Scholar) and Sybyl (Tripos, St. Louis, MO). Depictions in Figs. 7 and 8 were prepared using Molscript (27Kraulis P.J. J. Appl. Crystallogr. 1991; 24: 946-950Crossref Google Scholar). Those in Fig. 12, center and right, were prepared using VMD (28Humphrey W. Dalke A. Schulten K. J. Mol. Graph. 1996; 14: 33-38Crossref PubMed Scopus (39882) Google Scholar) and Raster3D (29Mettitt E.A. Bacon D.J. Methods Enzymol. 1997; 277: 505-524Crossref PubMed Scopus (3878) Google Scholar) following molecular dynamics using the Amber force field (30Duan Y. Wu C. Chowdhury S. Lee M.C. Xiong G. Zhang W. Yang R. Cieplak P. Luo R. Lee T. Comput. Chem. 2003; 24: 1999-2012Crossref Scopus (3762) Google Scholar). Assumptions used during modeling are described under “Results.”Fig. 7Overview of receptor structure. Panel A, view of the LRD and SSD as would be seen looking toward the cell surface. The portions of the SSD that are missing in the SSD of models of the LHR and FSHR are indicated by arrows. The asterisks refer to residues shown in the table (panel C). Panels B and D, views of the LRD and SSD as seen from the transmembrane domain. That in panel D is rotated 90° relative to that in panel B. Panel C, table describing potential contacts in the LRD and SSD of all three glycoprotein hormone receptors. The single and double asterisks refer to residues in the LRD and SSD, respectively.View Large Image Figure ViewerDownload (PPT)Fig. 8hCG and hormone receptor complexes. Panel A, the structure of hCG illustrating the relative positions of antibodies that bind to the hormone-receptor complex (pointed arrows) and those that do not (blunted arrows). Although antibodies A113, B101, and B301 are potent inhibitors of hCG-LHR interaction, some of the antibodies that bind to the hormone-receptor complex such as B112 and to a lesser extent B105 and B110 also inhibit complex formation. The broken curved line is drawn to illustrate the top surface of the LRD. Panel B, this panel illustrates the docking of hCG to the extracellular domain of the receptor. Panels C and D, these panels illustrate the orientations that we anticipate are most favorable for docking of hCG and hFSH/hTSH, respectively. Note that the structure of the receptor enables either orientation of the ligand to contact the SSD.View Large Image Figure ViewerDownload (PPT)Fig. 12Comparison of sequences in regions of the LRD and TMD that we propose are essential for signaling. Panel A, several residues that we anticipate form important interactions between LRD repeats 3–6 (R3, R4, R5, R6), TMD helices 6 and 7 (H6, H7), and TMD outer loop 3 (OL3) are highly conserved in each receptor class. This is particularly evident for asparagine and aspartic acid in R4 and R5, for the lysine in H7, and for the sequence “PLITV” that forms all of outer loop 3. Backbone atoms of outer loop 3 form most of a binding site for the side chain of the lysine in H7 (panels B and C), which is shown boxed and that appears to be required for signaling. The presence of a glycine that corresponds to human TSHR residue Lys639 in one sequence of the rat TSHR (AAA53209.1) is the only receptor that we have been able to find that has a residue other than lysine at this position. A second sequence in the data base (AAG2421.1) suggests that there may be a lysine at this site. All mammalian TSHR have a lysine near the top of helix 6 that appears to face toward the TMD and helix 7. Molecular dynamics simulations suggest the side chains of this lysine and an arginine in the piscine TSHR are near a negatively charged residue in outer loop 2 of the TMD in the TSHR (panel C). This would be expected to attract the top of helix 6 to the TMD. Contacts between LRD repeats 3 (R3) and 6 (R6) and residues at the top of helices 6 (H6) and 7 (H7) appear to influence ligand binding specificity and differ between the hormone classes. The arrows indicate the proximity of residues in these repeats relative to residues in outer loop 3. Panel B, relative positions of key residues in a model of the rLHR lacking exon 10 following molecular dynamics simulation. Asp135, which is located in R5 of the LRD, appears to form a salt bridge with Lys573, which is located at the top or H6 in the TMD. Lys583 at the top of H7 appears to be stabilized by hydrogen bonds with backbone atoms of residues in outer loop 3 and with the backbone oxygen of a highly conserved glutamate in the SSD. The side chain of this glutamate also appears to have a role in stabilizing Lys488 in outer loop 2 of the TMD, an interaction that would be expected to contribute to the stability of the SSD-TMD complex. Panel C, relative positions of residues in a model of the human TSHR lacking residues 292–363. Unlike rLHR-Asp135, the corresponding hTSHR residue (i.e. Asp139) does not appear to participate in contacts with H6, most likely because an asparagine is substituted for the lysine corresponding to rLHR-Lys573. Several other contacts between residues in the LRD and outer loop 3 of the TSHR appear to stabilize interactions between the LRD and the TMD, however (not shown). Lys630 in H6 is located in a position in which it appears to form a salt bridge with Asp552 in outer loop 2, a phenomenon that would be expected to potentiate movements of H6 toward the TMD following TSHR binding, antibody binding, or trypsin digestion. As in the case of the LHR, the lysine in H7 of the TSHR appears to form hydrogen bonds with backbone atoms of residues in outer loop 3 and Glu388 of the SSD. Also, as in the case of the rLHR, the side chain of Glu388 appears to form a salt bridge with the side chain of Lys544, a lysine in outer loop 2. This interaction would be expected to help stabilize the positions of the SSD and TMD.View Large Image Figure ViewerDownload (PPT) Influence of the SSD on Ligand Binding—rLHRδ10 was derived from an analog of the rat LHR that contains a modified FLAG tag that does not require calcium for binding of the M1 anti-FLAG tag antibody. The presence of the epitope tag did not affect the ability of the rat LHR to bind hCG or bovine LH or to make cyclic AMP in response to hCG-δα2 (Figs. 2A and 3A). In most studies, the concentrations of hCG required to prevent binding of 125I-hCG to CHO cells that expressed rLHRδ10 were similar to those that inhibited binding of 125I-hCG to CHO cells that express the rat LHR (Fig. 2B). This suggested that both receptors had roughly equivalent affinities for hCG. Smaller amounts of hCG-δα2 were required to inhibit the binding of 125I-hCG to CHO to cells expressing rLHRδ10, however (Fig. 2B), indicating that it appeared to have a slightly greater ability to bind an hCG analog lacking the oligosaccharide on α-subunit loop 2. In contrast, bLH competed poorly with 125I-hCG for binding to rLHRδ10 (Fig. 2B). Thus, whereas 200–300 ng of bLH were required to inhibit the binding of 125I-hCG to the rat LHR by 50% in this assay, more than 10 μg of bLH was required to halve the binding of 125I-hCG to rLHRδ10 (Fig. 2). This showed that the ability of bLH to recognize hLHRδ10 was at least 30–50-fold lower than its ability to recognize the rat LHR. In this regard, rLHRδ10 behaved more like the human LHR than the rat LHR (9Bernard M.P. Myers R.V. Moyle W.R. Biochem. J. 1998; 335: 611-617Crossref PubMed Scopus (24) Google Scholar). These differences in the abilities of bLH to bind the rat LHR and rLHRδ10 showed that residues missing or replaced in the SSD of rLHRδ10 have roles in the binding of some lutropins. These finding are consistent with reports that regions of the rat LHR outside the LRD can influence the LHR binding of ovine LH, a lutropin similar to bLH (31Abell A. Liu X. Segaloff D.L. J. Biol. Chem. 1996; 271: 4518-4527Abstract Full Text Full Text PDF PubMed Scopus (27) Google Scholar).Fig. 3Influence of the SSD on signal transduction. Panel A, hCG and an analog prepared by deglycosylation of the same highly purified hCG preparation was used to stimulate the FLAG-tagged rat LHR in the presence of 2 mm IBMX, an inhibitor of phosphodiesterase. Values illustrate the total amounts of cyclic AMP produced (i.e. that in the medium plus that in the cells, means of triplicate incubations). Solid squares and circles represent the responses to hCG and hCG-δα2. Open squares and circles represent the responses to 10 and 30 ng of α5-β8hCG and α5-β8hCG-N52D, analogs containing an NH2-terminal cross-link. The efficacy of α5-β8hCG was not different from that of hCG. hCG-δα2 had ∼70% the efficacy of hCG in this assay (solid circles) as did the NH2-terminal cross-linked analog lacking the oligosaccharide on α-subunit loop 2 (open circles). Panel B, efficacy of hCG and hCG-δα2 in cells expressing the rat LHR and rLHRδ10. Cells expressing the rat LHR and rLHRδ10 were exposed to increasing amounts of hCG and hCG-δα2 in the presence of IBMX. hCG-δα2 had ∼55% the efficacy of hCG in cells expressing the rat LHR (solid lines). It had only 25% the efficacy of hCG in cells expressing the rLHRδ10 receptor (broken lines). Panel C, influence of an oligosaccharide added to β-subunit residue 77 on signal transduction through the rLHRδ10 receptor. The indicated hormone analogs were produced by transient expression of COS-7 cells. The media were concentrated by ultrafiltration and the concentrations of analogs were determined by sandwich immunoassays. As shown here, the presence of the intersubunit disulfide cross-link reduced the efficacy of hCG in assays that involve the rLHRδ10 receptor. Efficacy was diminished further by elimination of the α-subunit loop 2 oligosaccharide. Addition of an oligosaccharide near the tip of β-subunit loop 3 at residue 77 led to an increase in efficacy in both the cross-linked and cross-linked deglycosylated analogs. Panel D, signal transduction responses to hCG, hLH, and bLH in rat LHR and rLHRδ10 receptor expressing CHO cells. Cells expressing the rat LHR and the rLHRδ10 receptor were treated with hCG, hLH, and bLH. Note that the response to hLH was equivalent to that of hCG in both cell types. bLH had a significantly lower potency than either hCG or hLH in both cell types. These assays were performed in a volume of 60 μl.View Large Image Figure ViewerDownload (PPT) The seatbelt is responsible for much of the influence of the β-subunit on receptor binding specificity (32Campbell R.K. Dean Emig D.M. Moyle W.R. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 760-764Crossref PubMed Scopus (124) Google Scholar, 33Moyle W.R. Campbell R.K. Myers R.V. Bernard M.P. Han Y. Wang X. Nature. 1994; 368: 251-255Crossref PubMed Scopus (317) Google Scholar, 34Dias J.A. Zhang Y. Liu X. J. Biol. Chem. 1994; 269: 25289-25294Abstract Full Text PDF PubMed Google Scholar, 35Grossmann M. Szkudlinski M.W. Wong R. Dias J.A. Ji T.H. Weintraub B.D. J. Biol. Chem. 1997; 272: 15532-15540Abstract Full Text Full Text PDF PubMed Scopus (77) Google Scholar). To learn if the seatbelt can affect interactions of the glycoprotein hormones with the SSD, we tested the abilities of rLHRδ10 to recognize hCG/hFSH chimeras. One of these has its small seatbelt loop derived from the hFSH β-subunit and is known to bind LHR roughly 8–12% as well as hCG (32Campbell R.K. Dean Emig D.M. Moyle W.R. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 760-764Crossref PubMed Scopus (124) Google Scholar). This hormone chimera inhibited 125I-hCG binding to the rat LHR ∼10% as well as hCG, but had a much lower ability to inhibit the binding of 125I-hCG to the rLHRδ10 (Fig. 2D). This showed that the SSD domain may compensate for mutations in the small seat-belt loop that are known to reduce the ability of hCG to interact with the rat LHR. Substitution of hFSH residues for those in the hCG strap, i.e. the COOH-terminal half of the seatbelt, creates bifunctional chimeras that interact with the rat LHR like hCG and that bind the rat FSHR roughly one-third as well as hFSH (33Moyle W.R. Campbell R.K. Myers R.V. Bernard M.P. Han Y. Wang X. Nature. 1994; 368: 251-255Crossref PubMed Scopus (317) Google Scholar). Bifunctional chimera CF101–109 bound to the FLAG-tagged rat LHR receptor similar to hCG (Fig. 2A), but poorly to rLHRδ10 (Fig. 2C). Although this finding is consistent with the notion that hCG residues in the strap interact with parts of the SSD that are altered or missing in the rLHRδ10, this seems unlikely because many hCG analogs in which the seatbelt is latched to different parts of the α-subunit (36Xing Y. Lin W. Jiang M. Myers R.V. Cao D. Bernard M.P. Moyle W.R. J. Biol. Chem. 2001; 276: 46953-46960Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar) recognized the rat LHR much better than CF101–109 bound to rLHRδ10 (Fig. 2C). The hFSH residues in the straps of these chimeras alter the conformation of the heterodimer (37Wang Y.H. Bernard M.P. Moyle W.R. Mol. Cell. Endocrinol. 2000; 170: 67-77Crossref PubMed Scopus (13) Google Scholar), indicating that the reduced abilities of CF101–109 and CFC101–114 to bind rLHRδ10 might reflect their altered conformations. hFSH has less than 0.01% the activity of hCG in rat LHR binding assays; hCG/hFSH chimeras in which both the loop and the strap regions are derived from hCG have less than 1% of the activity of hCG in rat LHR assays (32Campbell R.K. Dean Emig D.M. Moyle W.R. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 760-764Crossref PubMed Scopus (124) Google Scholar). Both hFSH and these chimeras were inactive in assays employing the rLHRδ10 receptor (not shown). Influence of the SSD on Efficacy—To learn if the SSD contributes to hormone efficacy, we measured cyclic AMP accumulation in response to hCG and hCG-δα2, a partial agonist analog that lacks the oligosaccharide at residue Asn52 in loop α2. As noted earlier, the rLHRδ10 receptor has an NH2-terminal FLAG epitope tag. To learn if this would influence its ability to respond to hCG and hCG-δα2, we tested the abilities of these analogs to stimulate cyclic AMP accumulation in assays employing the FLAG-LHR. The presence of the epitope tag was not responsible for the reduced ability of the rLHRδ10 bearing cells to respond to hCG-δα2 (Fig. 3A). In assays employing cells that express the FLAG-LHR, the maximum amount of cyclic AMP accumulation observed in the presence of IBMX in response to hCG-δα2 was 70% that observed in response to hCG (Fig. 3A), a value that is equivalent to or greater than that observed in assays employing the rat LHR (Fig. 3B). Furthermore, the FLAG epitope did not affect the maximal responses to α5-β8hCG, an analog that has an NH2-terminal cross-link, or to α5-β8hCG,δα52, a cross-linked analog that is missing the oligosaccharide on α-subunit loop 2 (Fig. 3A). Cells expressing the rLHRδ10 receptor recognized hCG-δα2 better than they recognized hCG (Fig. 2B); their abilities to accumulate cyclic AMP in response to hCG-δα2 were less than half of those of cells that express the rat LHR in media lacking the in