The role of the hinge region of the luteinizing hormone receptor in hormone interaction and signal generation.

Luteinizing hormone receptor, a G protein-coupled receptor, consists of two halves, the N-terminal extracellular hormone binding domain (exodomain) and the C-terminal membrane-associated, signal-generating domain (endodomain). The exodomain has seven to nine Leu-rich repeats, which are generally thought to form a 1/3 donut-like structure and interact with human choriogonadotropin (hCG). The resulting hCG-exodomain complex adjusts the structure and its association with the endodomain, which results in signal generation in the endodomain. It is unclear whether the rigid 1/3 donut structure could provide the agility and versatility of this dynamic action. In addition, there is no clue as to where the endodomain contact point (the signal modulator) in the exodomain is. To address these issues, the exodomain was examined by Ala scan and multiple substitutions, while receptor peptides were used for photoaffinity labeling and affinity cross-linking. Our results show that the C-flanking sequence (hinge region), Thr(250)-Gln(268), of the Leu-rich repeats (LRRs) specifically interacts with hCG, preferentially hCGalpha. This interaction is inhibited by exoloop 2 of the endodomain but not by exoloops 1 and 3, suggesting an intimate relationship between Thr(250)-Gln(268), exoloop 2, and hCG. Taken together, our observations in this article suggest a new paradigm that the LRRs contact the front of hCG, while both flanking regions of the LRRs interact with the sides of hCG. This would trap hCG in the 1/3 donut structure of the LRRs and enhance the binding affinity. In addition, mutations of conserved Ser(255) in the sequence can constitutively activate the receptor. This provides a clue for the signal modulator in the exodomain. In contrast, a phenyl or phenolic group is necessary at conserved Tyr(253) for targeting the receptor to the surface.

The glycoprotein hormone receptors comprise two equal halves; an extracellular N-terminal half (exodomain) and a membrane-associated C-terminal half (endodomain). The exodomain is ϳ350 amino acids long, which alone is capable of high affinity hormone binding (1-3) without hormone action (3,4). The endodomain consists of seven transmembrane domains, three exoloops, three cytoloops, a C-terminal tail, and a short extracellular extension connected to transmembrane domain 1.
Receptor activation occurs in the endodomain (3,5,6), which is structurally equivalent to the entire molecule of many other G protein-coupled receptors (7). Existing data suggest that these two domains are associated together before and after hormone binding (8,9). A hormone initially binds to the exodomain, and the resulting hormone-exodomain complex adjusts its conformation (10) and then interacts with the endodomain (7). This change in the interaction in the ternary complex, hCG-exodomain-endodomain, is thought to generate a signal (6,7,11).
There is ample evidence that the exodomain of the luteinizing hormone/choriogonadotropin receptor (LH/CG-R) 1 alone is capable of binding hCG with high affinity. The truncated exodomain lacking the endodomain binds hCG with a high affinity similar to or slightly better than the wild type affinity (1)(2)(3). Exodomain fusion proteins are capable of binding hCG with a high affinity (12,13). A number of sites in the exodomain are important for hormone binding (14 -16) and hormone specificity (17)(18)(19). LHR peptides corresponding to three regions in the exodomain are capable of inhibiting hCG binding to the receptor (20).
A major feature in the structure of the LHR exodomain is the Leu-rich repeat (LRR) motif (21,22). LRRs are found in a large number of proteins (23). The existing crystal structures of a few LRR proteins show up to 15 LRRs, which form a rigid 2/3 donut structure (24,25). Based on these LRR crystal structures, the glycoprotein hormone receptors have been modeled (26 -28). They show a 1/3 donut structure consisting of 7-9 LRRs and suggest that the inner lining interacts (16,29,30) with the concave, front side of the hormones (31,32). Recently, it has been shown that the putative LRRs in the gonadotropin receptors are indeed active and that part of them interact with hCG (33)(34)(35).
However, it is unclear whether the rigid 1/3 donut structure of the LRRs alone interacts with hCG, adjusts its conformation, contacts with the endodomain, and activates the endodomain or if other regions of the exodomain are involved in the process. This agility and versatility may come from regions other than the LRRs. For example, we have shown that the N-terminal flanking region of the LRRs makes strong contact with both subunits of hCG (36,37). However, it is unknown whether the N-terminal flanking region of the LRRs is the only exodomain region outside of the LRRs to interact with hCG. In addition to the insufficient knowledge on the exodomain/hCG interaction, little is known about the contact sites between the exodomain and endodomain. There is evidence that exoloops 2 and 3 of the endodomain are involved in the interaction with unliganded exodomain and constrain the hormone binding at the exodomain (8,9). It has just been reported that LRR4 (34,35) and the C-terminal flanking (hinge) region of the LRRs (38) modulate signal generation in the endodomain. However, there is no clue to where the endodomain contact point is in the exodomain. In this article, the first evidence is presented that the hinge region interacts with hCG and that the hCG-exodomain complex makes contacts with exoloop 2 of the endodomain.

EXPERIMENTAL PROCEDURES
Materials-The N-hydroxysuccinimide of 4-azidobenzoic acid (NHS-AB) was synthesized as described previously (39). The N-hydroxysulfosuccinimide esters of ethylene glycolbis(sulfosuccinimidylsuccinate) (SES) were purchased from Pierce. The hCG CR 127 and hCG subunits were supplied by the National Hormone and Pituitary Program. Denatured hCG was prepared by boiling hCG in 8 M urea for 30 min. Receptor peptides were synthesized and N-acetylated and C-amidated by Biosynthesis (Lewisville, TX). They were purified on a Vydac C 18 high performance liquid chromatography column using a solvent gradient from 100% of 0.1% trifluoroacetic acid in water to 20% of 0.1% trifluoroacetic acid in water and 80% 1-propanol.
Mutagenesis and Functional Expression of LHR-Mutant LHR cDNAs were prepared in a pSELECT vector using the Altered Sites Mutagenesis System (Promega), sequenced, subcloned into pcDNA3 (Invitrogen) as described previously (40), and sequenced again to verify mutation sequences. This procedure does not involve polymerase chain reaction, and therefore, there are no polymerase chain reaction-associated unintended mutations. To produce truncated receptors, a codon was substituted with a stop codon. Mutant and truncated LHR constructs were transfected into human embryonic kidney 293 cells by the calcium phosphate method. Stable cell lines were established in minimum essential medium containing 10% horse serum and 500 g/ml of G-418. They were used for hormone binding and cAMP production. All assays were carried out in duplicate and repeated four to six times. Means and standard variations were calculated. 125 I-hCG Binding and Intracellular cAMP Assay-Stable cells were assayed for 125 I-hCG binding in the presence of a constant amount of 125 I-hCG (39) and increasing concentrations of unlabeled hCG. The K d values were determined by Scatchard plots. For intracellular cAMP assay, cells were washed twice with Dulbecco's modified Eagle's medium and incubated in the medium containing isobutylmethylxanthine (0.1 g/ml) for 15 min. Increasing concentrations of hCG were then added, and the incubation was continued for 45 min at 37°C. After removing the medium, the cells were rinsed once with fresh medium without isobutylmethylxanthine, lysed in 70% ethanol, freeze-thawed in liquid nitrogen, and scraped. After pelleting cell debris at 16,000 ϫ g for 10 min at 4°C, the supernatant was collected, dried under vacuum, and resuspended in 10 l of the cAMP assay buffer, which was provided by Amersham Pharmacia Biotech. cAMP concentrations were determined with an 125 I-cAMP assay kit (Amersham Pharmacia Biotech) following the manufacturer's instruction and validated for use in our laboratory. All assays for cAMP and binding to intact cells and solubilized receptors were repeated four to six times in duplicate to calculate means Ϯ S.D.

125
I-hCG Binding to Solubilized LHR-Transfected cells were washed twice with ice-cold 150 mM NaCl, 20 mM HEPES, pH 7.4 (buffer A). Cells were scraped on ice and collected in buffer A containing protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 5 mM N-ethylmaleimide, and 10 mM EDTA) and pelleted by centrifugation at 1,300 ϫ g for 10 min. Cells from a 10 cm plate were resuspended in 0.6 ml of buffer A containing 1% Nonidet P-40, 20% glycerol, and the above protease inhibitors (buffer B), incubated on ice for 15 min, and diluted with 5.4 ml of buffer A containing 20% glycerol plus the protease inhibitors (buffer C). The mixture was centrifuged at 100,000 ϫ g for 60 min. The supernatant (500 l) was mixed with 150,000 cpm of 125 I-hCG and 6.5 l of 0.9% NaCl and 10 mM Na 2 HPO 4 at pH 7.4 containing increasing concentrations of unlabeled hCG. after incubation at 4°C for 12 h, the solution was thoroughly mixed with 250 l of buffer A containing bovine ␥-globulin (5 g/ml) and 750 l of buffer A containing 20% polyethylene glycol 8000. After incubation at 4°C for 10 min, samples were pelleted at 1300 ϫ g for 30 min and supernatants removed. Pellets were resuspended in 1.5 ml of buffer A containing 20% polyethylene glycol 8000, centrifuged, and counted for radioactivity.
Derivatization and Radioiodination of Peptide-NHS-AB was freshly dissolved in dimethyl sulfoxide to a concentration of 50 mM in 0.1 M sodium phosphate, pH 7.5, to a concentration of 20 mM. This reagent solution was immediately used to derivatize receptor peptides. In the dark, 10 l of NHS-AB was added to 30 g of LHR 246 -269 in 40 l of 0.1 M sodium phosphate, pH 7.5. The mixture was incubated for 30 min for NHS-AB or 60 min for NHS-AB at 25°C. The following were added to the derivatization mixture: 1 mCi of Na 125 I in 10 l of 0.1 M NaOH and 7 l of chloramine T (1 mg/ml) in 10 mM Na 2 HPO 4 and 0.9% NaCl, pH 7.4 (PBS). After 20 s, 7 l of sodium metabisulfite (2.5 mg/ml) in PBS was introduced to terminate radioiodination. Derivatized and radioiodinated AB-125 I-LHR 246 -269 solution was mixed with 60 l of 16% sucrose solution in PBS and fractionated on Sephadex Superfine G-10 column (0.6 ϫ 15 cm) using PBS.
Affinity Cross-linking of 125 I-LHR 246 -269 to hCG-Disposable glass tubes were siliconized under dimethyldichlorosilane vapor overnight and autoclaved. In each siliconized tube, 20 l of PBS, hCG (70 ng in 10 l of PBS), and 125 I-LHR 246 -269 (100 ng in 10 l of PBS) were mixed and incubated in 37°C for 90 min. After incubation, 3 l of 0.1 mM SES in dimethyl sulfoxide was added to each tube and further incubated at 25°C for 20 min. The cross-linking reaction was terminated by adding 3 l of 5 mM Gly in PBS. The samples were boiled for 2 min in 2% sodium dodecyl sulfate, 100 mM dithiothreitol, and 8 M urea. The solubilized samples were electrophoresed on 8 -12% polyacrylamide gradient gels. Gels were dried on filter paper, then exposed to a molecular imaging screen (Bio-Rad) overnight. The imaging screen was scanned on a model GS-525 Molecular Imager System Scanner (Bio-Rad), and radioactive band intensity was analyzed using Image Analysis Systems Version 2.1 (Bio-Rad). Gels were exposed to X-Omat x-ray film at Ϫ75°C for ϳ4 days Photoaffinity Labeling of hCG-The following solutions were sequentially introduced to siliconized glass tubes: 20 l of PBS, 10 l of hCG (10 ng/l) in PBS, and 10 l of AB-125 I-LHR 246 -269 (10 ng/l) in PBS. The mixtures were incubated at 37°C for 90 min in the dark, irradiated with Mineralight R-52 UV lamp for 3 min as described previously (41), and solubilized in 2% SDS, 100 mM dithiothreitol, and 8 M urea. The samples were electrophoresed on 8 -12% polyacrylamide gradient gels. Gels were dried on filter paper and processed as described above.
Competitive Inhibition of Affinity Labeling of hCG and Inhibition of 125 I-hCG Binding to LHR-Competitive inhibition experiments were carried out as described for the affinity cross-linking and photoaffinity labeling experiments, except that 10 l instead of 20 l of PBS was introduced to each tube, and the mixture was incubated with 10 l of increasing concentrations of nonradioactive wild type or mutant LHR 246 -269 .
A human embryonic kidney 293 cell line stably expressing human LHR was incubated with 100,000 cpm of 125 I-hCG in the presence of increasing concentrations of nonradioactive wild type or mutant LHR 246 -269 as described previously (42). After washing the cells several times, the radioactivity associated with the cells was counted to determine the K d value.

RESULTS
Truncation of the C Terminus-LHR is encoded by 11 exons (22,43), and the exodomain is primarily comprised of exons 1-10, whereas the endodomain is mainly encoded in exon 11. When exons 1-10 were expressed in mammalian cells after truncating exon 11, the exodomain fragment was trapped within the cells and could not be detected on the intact cell surface or in the culture medium (2, 3). However, when the cells were solubilized in nonionic detergent, the exodomain fragment was capable of binding hCG with a high affinity similar to the hormone binding affinity of the wild type receptor (Table I) as reported previously (1)(2)(3). When the terminal region was further truncated by deleting both exons 10 and 11, the resulting exodomain fragment consisting of exons 1-9 was also trapped in cells and capable of hormone binding. The binding affinity was, however, significantly less as the K d value was ϳ6-fold greater than that of the wild type receptor in detergent solution (Table I). This result suggests the presence of a potentially important site around the exon 9 -10 boundary (36).
Sequence Comparison and Ala Scan of Conserved Residues-To screen important residues in the exon 9 -10 junction of LHR, the junction sequence was aligned with the corresponding sequence of the FSH receptor and TSH receptor (Fig.  1). A homology sequence was found from Tyr 253 to Phe 260 . In addition, the sequence corresponding to LHR Arg 261 -Gln 268 is highly conserved within each receptor type among species. We decided to Ala scan from Thr 250 to Gln 268 except Ala 259 that was substituted with Gly. The resulting mutant receptors were stably expressed in cells and intact cells were assayed for 125 I-hCG binding and cAMP induction (Fig. 2). In addition, cells that were unable to bind hormone were solubilized in Nonidet P-40 and assayed for 125 I-hCG binding in solution.
The results reveal two mutant classes: one that was capable of binding hCG on intact cells and the other incapable of hCG binding on intact cells. All of the latter were capable of binding hCG after solubilization in Nonidet P-40, indicating that Ala substitutions did not abolish the hCG binding activity and that the mutants were trapped in cells (36). The K d values were generally higher than the wild type value except Cys 257 3 Ala and Phe 260 3 Ala, indicating lower affinities. The lower affinities do not appear to be due to the variations in the receptor concentration, because they are in the range of 22,000 -57,000/ cell, except for Thr 252 3 Ala and Phe 260 3 Ala. At these concentrations, the receptors show normal activities (9,33,34,44). It is interesting to note that all mutants in the second half of the sequence from Ala 259 3 Gly to Gln 268 3 Ala were expressed on the surface, whereas only four mutants from Thr 250 to Cys 258 were expressed on the surface. This result suggests that the first half, but not the second half, of the sequence may be important for receptor targeting to the cell surface.
Substitutions for Tyr 253 Impair Surface Expression-To further investigate the impact on surface expression, Tyr 253 was substituted with an array of amino acids (Table II). The substitution of Phe did not impact the surface expression or the surface receptor concentration. In contrast, the His substitution significantly impaired the surface expression, and other substitutions blocked it. These results show that Tyr 253 3 Phe and a phenyl group at the Tyr 253 position are permissible for receptor targeting and hormone binding. Therefore, the hydroxyl moiety of the Tyr's phenolic group does not appear to influence the targeting and hormone binding. The partial blockage of surface expression by the substitution of His suggests that an imidazole at the Tyr 253 position is tolerable but not desirable. Substitutions of nonaromatic hydrophobic, hydrophilic, neutral, and acidic amino acids for Tyr 253 prevented surface expression of the receptor, although the concentration of the mutant receptors trapped in cells was significant and sometimes considerably higher than the total concentration of the wild type receptor. These results suggest that a phenolic or phenyl group is necessary at the Tyr 253 position for efficient receptor transport to the surface membrane. To test this possibility and the essential role of Tyr 253 , the residue was deleted. The resulting Tyr 253 deletion mutant was not expressed on the surface, supporting the hypothesis.
One may argue that cells expressing trapped receptors might have developed a defective mechanism for receptor transport, perhaps when stable cell lines were being established. To test this unlikely possibility, the cells stably expressing the Tyr 253 3 Ala mutant were transfected again with the wild type LH/ CG-R or the human FSH receptor plasmid (45). The transfected cells were capable of binding hCG or FSH, respectively (Table  III), indicating that the receptor transport mechanism was functional. To exclude the possibility that the second transfection rescued the receptor transport mechanism, additional experiments were performed. Cells stably transfected with the FSH receptor or wild type LH receptor were transfected again with the Tyr 253 3 Ala mutant plasmid. These doubly transfected intact cells were capable of binding FSH or hCG, respectively, indicating that the receptor-targeting mechanism was not affected by transfection of the Tyr 253 3 Ala mutant plasmid and expression of the Tyr 253 3 Ala mutant.
Constitutive Activation of cAMP Induction by Substitution for Ser 255 -The examination of cAMP induction discloses that the mutations impaired the EC 50 value, maximum level, or both for cAMP induction (Fig. 2, A and B). The only exception is Ser 255 3 Ala. It showed a similar EC 50 and higher maximum cAMP as compared with the wild type values, suggesting an improved efficacy of cAMP induction. To follow up this interesting observation, Ser 255 was replaced with several amino acids (Table IV). The resulting mutants were transiently surface-expressed in the ϳ20,000 -40,000/cell range. Some of mutations, Val and Ile, resulted in strikingly high basal cAMP levels, despite lower receptor concentrations on the surface. This is unusual because all activating mutations uncovered to date are found in the endodomain (11,46). A simple explanation is that the region may be involved in signal generation in the endodomain. If so, it may interact with hCG, the endodomain, or both. However, it is also possible that the mutations might have impacted protein processing including folding and thus, the global structure, which in turn resulted in the observed high basal cAMP level. One way to resolve this difficult issue is to determine its interaction, which can be done by affinity labeling.
Affinity Labeling of hCG with Receptor Peptide LHR-We synthesized a 24-mer peptide, Ac-Leu 246 -Asn 269 -NH 2 (LHR 246 -269), derivatized it with NHS-AB, and radioiodinated it to produce AB-125 I-LHR 246 -269 . NHS-AB couples to the Lys residues, and Lys 265 and Lys 266 are present in the peptide. To determine  1. Sequence alignment. The amino acid sequence around the exon 9 -10 junction is aligned for the LH, FSH, and TSH receptors. "*" represents a conserved residue.
whether AB-125 I-LHR 246 -269 would bind and label hCG, and if so, which subunit is labeled, the peptide derivative was incubated with hCG and irradiated with UV for increasing time periods (Fig. 3A). The samples were solubilized in SDS under reducing conditions and electrophoresed as described under "Experimental Procedures." The autoradiographic phosphoimage of the gel shows that AB-125 I-LHR 246 -269 labeled conspic-uously both the ␣ and ␤ subunits in hCG as well as faintly the ␣␤ dimer. The positions of hCG␣, hCG␤, and the hCG␣␤ dimer were determined by comparing the respective positions of 125 I-hCG␣, 125 I-hCG␤, and the cross-linked 125 I-hCG ␣␤ dimer on the autoradiograph (35, 37). The labeling was dependent on UV irradiation, reaching the maximum level at 60-s exposure. The labeling of the ␣␤ dimer implies that there were two AB groups

Contact Point of hCG and LH Receptor 3454
attached to the peptide and that one was apparently closer to hCG␣ and labeled it, whereas the other was closer to and labeled hCG␤. This is in contrast to an LHR peptide derivatized with an AB that is capable of labeling only one of the subunits but not both (35,37). It suggests the specificity of photoaffinity labeling.
Next, increasing concentrations of hCG were incubated with a constant amount of AB-125 I-LHR 246 -269 and irradiated with UV for 60 s (Fig. 3B). The results show that the hCG labeling was also dependent on the hCG concentration, reaching a maximum labeling at and above 200 ng. In the following experiment, 200 ng of hCG was incubated with increasing concentrations of AB-125 I-LHR 246 -269 and UV-photolyzed for 60 s (Fig.  4A). The result shows the dependence of the labeling on the AB-125 I-LHR 246 -269 concentration with the maximum labeling at and above 200 ng of AB-125 I-LHR 246 -269 . This experiment, taken together with the UV-and hCG-dependent experiments, indicate that the labeling of hCG␣, hCG␤, and hCG␣␤ is saturable and requires the peptide derivative, hCG, and UV photolysis. In addition, the results indicate that hCG was covalently labeled by AB-125 I-LHR 246 -269 and that this labeling was specific. To test whether the derivatization of AB impacted the activity of the peptide, AB-125 I-LHR 246 -269 was incubated with hCG in the presence of increasing concentrations of nonlabeled peptide and photolyzed (Fig. 4B). Nonlabeled peptide competitively inhibited the labeling. To further test the labeling specificity, AB-125 I-LHR 246 -269 was incubated with increasing concentrations of denatured hCG and photolyzed. No labeled bands appeared (data not included), indicating that AB-125 I-LHR 246 -269 labeled active, but not denatured, hCG.
Affinity Cross-linking of 125 I-LHR 246 -269 to hCG-Despite the specificity of photoaffinity labeling, one may raise an issue whether the AB attached to the peptide, particularly those at the contact point, might have interfered with the binding and labeling and distorted the result. To resolve this issue, the peptide without attached AB, 125 I-LHR 246 -269 , was crosslinked to hCG using an amino-specific homobifunctional reagent, SES. The results (Figs. 5 and 6) show that the peptide was cross-linked to both hCG subunits and the hCG␣␤ dimer. The affinity cross-linking requires SES, 125 I-LHR 246 -269 , and hCG and is dependent on their concentrations. The maximum levels of cross-linking reached around 15% of total peptide for   hCG␣ and 10% for hCG␤, indicating that the labeling is saturable and specific. In addition, hCG␣ was preferentially labeled compared with hCG, which is consistent with the preferential photoaffinity labeling of the hCG␣ subunit. This affinity crosslinking is more effective than photoaffinity labeling. A possible explanation is that SES can reach further than AB can, because of the maximum cross-linkable distance, ϳ13 Å for SES and ϳ7 Å for AB. The SES cross-linking results also show that either or both amino groups of Lys 265 and Lys 266 of 125 I-LHR 246 -269 , the only amino groups in the peptide, were crosslinked to one of seven hCG␣ amines, five hCG␤ amines, or those in both subunits. The SES-dependent cross-linking of 125 I-LHR 246 -269 to the hCG subunits was competitively inhibited by nonlabeled peptide, suggesting a specificity. Furthermore, SES failed to cross-link 125 I-LHR 246 -269 to denatured hCG (data not included). Differential Effects of Exoloop Peptides on Photoaffinity Labeling of hCG-Our mutational analysis suggests that the Thr 250 -Gln 268 region may be involved in signal generation in the endodomain, perhaps interacting with hCG, the endodomain, or both. Such interactions, if true, might impact the specific affinity labeling of hCG with LHR 246 -269 . To this end, hCG was incubated with AB-125 I-LHR 246 -269 in the presence of an excess amount of exoloop peptides and irradiated with UV (Fig. 7A). The labeling was inhibited most conspicuously by the exoloop 2 peptide, whereas the inhibition by the exoloop 3 peptide or exoloop 1 peptide was significantly less or barely noticeable, respectively (Fig. 7, B and C). DISCUSSION The results of our mutagenesis and affinity labeling studies indicate that the Thr 250 -Gln 268 sequence of the LHR exodomain interacts with both subunits of hCG, particularly with hCG␣. In previous reports, we have shown that the LRRs function in the LHR exodomain and interact with hCG (33)(34)(35) and that the N-flanking region of the LRRs also interacts with hCG (36,37). Receptor peptide mimics were useful in these studies, although synthetic peptide mimics have limitations. They do not necessarily mimic the function of the corresponding sequences in proteins, probably due to structural differences. This is underscored by the fact that an extension of either terminus or substitution of a single amino acid in peptides mimics deprives their activity, specificity, or both (20,35,37). Therefore, we have carefully chosen the sequence and size of the hinge peptide based on our mutational analysis and others' studies (20,36). Furthermore, short synthetic peptides can be structured for a stable tertiary structure (47) and assume a potent biological activity (48 -50).
Taken together, our observations show that the LRRs and their flanking regions interact with hCG. This suggests an intriguing possibility that the LRRs face hCG, probably at the front of hCG with the ␤ seat belt and ␣ C terminus (31,32), whereas the flanking regions of the LRRs face the sides of hCG and may reach the back of hCG. This would trap hCG in the 1/3 donut structure of the LRRs, slowing the dissociation rate of hCG and LHR (51) and enhancing the affinity. This is consistent with the fact that high affinities are generally ascribed to slow dissociation rates rather than fast association rates (52). For example, kinetic studies on interactions between hormones and receptors, antibodies and antigens, and enzymes and substrates show that association rate constants, k on , are in the FIG. 3. UV-and hCG-dependent photoaffinity labeling. AB-125 I-LHR 246 -269 (200 ng) was incubated with 200 ng of hCG and irradiated with UV for varying time periods (A) or incubated with increasing concentrations of hCG and exposed to UV for 60 s (B). After electrophoresis of the samples, gels were dried on filter paper and exposed to a molecular imaging screen (Bio-Rad) overnight. The imaging screen was scanned on a model GS-525 Molecular Imager System Scanner (Bio-Rad), and the radioactive band intensity was analyzed using Image Analysis Systems Version 2.1 (Bio-Rad). Gels were also exposed to X-Omat x-ray film at Ϫ75°C for ϳ4 days. The bar graphs show the percent radioactivity of the ␣ band and the ␤ band in a gel lane. narrow range of 10 5 -10 7 M Ϫ1 s Ϫ1 , whereas dissociation rate constants, k off , vary widely from 1 s -1 for low affinity complexes to a half-life of more than 3 months (10 Ϫ7 s Ϫ1 ) (52).
An hCG contact point in the Thr 250 -Gln 268 sequence is near or not too far from the two tandem Lys residues at 265 and 266 in the second half of the sequence. This is consistent with the fact that most Ala substitutions for the second half residues reduced the binding affinity by up to 2.3-fold but did not noticeably impact the surface expression and maximum cAMP induction level (Fig. 3, A and B). In contrast to this primary role of the second half in hCG binding, the first half of the sequence seems to be involved in the signal generation. For example, some substitutions for Ser 255 resulted in substantially high basal cAMP levels. In particular, the cAMP level was dramatically high for Val and Ile substitutions, indicating constitutive activation of adenylyl cyclase (38). Taken together with the results that the interaction of the Thr 250 -Gln 268 sequence with hCG is inhibited by the exoloop 2 peptide, our observations implicate the involvement of the Thr 250 -Gln 268 sequence in the interaction of hCG-exodomain complex with the endodomain and the subsequent signal generation. This is the first clue to the identification of the endodomain contact points in the exodomain. The exoloop 2 peptide may directly compete with LHR 246 -269 for the binding site or allosterically interfere with the interaction between LHR 246 -269 and hCG. On the other hand, the exoloop 3 peptide marginally inhibited the labeling of hCG by AB-125 I-LHR 246 -269 , and the inhibition by the exoloop 1 peptide was barely detectable. This suggests several possibilities that these peptides bind hCG, but the affinities are relatively low, do not share the binding site of LHR 246 -269 , or do not bind hCG at all. Since exoloop 3 impacts hCG binding to the exodomain, it is possible that exoloop 3 is capable of binding the exodomain or hCG-exodomain complex, but the affinity is low. In addition, exoloop 3 and the LHR 246 -269 sequence may not share a significant portion of their binding sites in hCG and the hCG-exodomain complex.
A phenyl or phenolic group at Tyr 253 is crucial for the recognition by the surface-targeting machinery, since both Tyr and Phe are permissible. It is striking to see that substitutions of other amino acids, including other hydrophobic amino acids, completely blocked the surface expression, suggesting a high specificity. The attenuation of surface expression by Ala substitution for Leu 251 -Thr 252 -Tyr 253 and His 256 -Cys 257 -Cys 258 is specific, and therefore, these residues may assume a particular structure. The two groups consist of 3 contiguous amino acids, and a secondary structure analysis suggests interesting possibilities. The Leu 251 -Thr 252 -Tyr 253 sequence is predicted to be part of a ␤ sheet (53), thus orienting Leu 251 and Tyr 253 on one side of the ␤ sheet and Thr 252 on the other. In such a structure, Leu 251 and Tyr 253 face one side, whereas Thr 252 would face the opposite side. This may explain why Leu 251 and Tyr 253 were more important for receptor transport than Thr 252 . Furthermore, it has been suggested that Leu and Tyr are part of a transport determinant of membrane proteins (54). Interestingly, the ␤ sheet is followed by a turn at Pro 254 and Ser 255 (53,55). Therefore, the two groups, Leu 251 -Thr 252 -Tyr 253 and His 256 -Cys 257 -Cys 258 , are likely to be in close proximity, which could allow them to form a contiguous determinant. This is also consistent with the observation that Pro 254 and Ser 255 play a role distinct from the flanking residues, in particular the unique role in modulating signal generation (38).
In conclusion, the results presented in this study show that the C-terminal flanking (hinge) region (38), Thr 250 -Gln 268 , of the LRRs in the LHR exodomain interacts with hCG, contacting primarily hCG␣. This interaction is specifically inhibited by exoloop 2 but not by exoloops 1 and 3 of the endodomain, suggesting an intimate relationship in the ternary complex of Thr 250 -Gln 268 , exoloop 2 and hCG. Taken together, our observations in this and previous articles (33)(34)(35) suggest a new paradigm that the LRRs face the front of hCG, while both of the flanking regions of the LRRs interact with the sides of hCG and possibly reach the back of hCG as well. This would trap hCG in the 1/3 donut structure of the LRRs, increasing the binding affinity. In addition, mutations of Ser 255 in the exodomain sequence can constitutively activate the receptor. This result, along with the ternary complex, provides a new clue for the identity of the enigmatic signal modulator in the exodomain.