Roles of transmembrane prolines and proline-induced kinks of the lutropin/choriogonadotropin receptor.

The lutropin/choriogonadotropin receptor is a seven-helix transmembrane (TM) receptor. A unique feature of TM helices is the content of Pro, which generally is absent in α helices of globular proteins. Because Pro disrupts helices and introduces a ∼26° kink, it has been speculated that Pro plays a crucial role in the structure of TM helices, exoloops, and cytoloops of TM receptors. To examine the roles of the five TM Pros of the lutropin/choriogonadotropin receptor, these residues were individually substituted. Mutant receptors were examined for surface expression, hormone binding, and cAMP induction. Surface expression was monitored after introducing the flag epitope into the receptors. Flag epitopes slightly affected cAMP induction but not hormone binding or surface expression of receptors as monitored by immunofluorescence microscopy and 125I-anti-flag antibody. The results indicate that Pro479 in TM 4 and Pro598 in TM 7 play important yet contrasting roles. Pro479 is crucial for hormone binding at the cell surface but not after solubilization of the receptor. This is more likely due to the Pro side chain than the Pro-induced kink. Pro598 is important for surface expression. The kinks of Pro463 of TM 4, Pro562 of TM 6, or Pro591 of TM 7 are not important because the substitution of Phe for these residues did not significantly impact surface expression, hormone binding, and cAMP induction.

The LH/CG 1 receptor belongs to a subfamily of glycoprotein hormone receptors within the seven-transmembrane receptor family. Unlike most seven-TM receptors, it is comprised of a long extracellular N-terminal exodomain and a membraneassociated C-terminal endodomain containing 7 TM helices (1,2). The exodomain is capable of high affinity hormone binding (3)(4)(5) without hormone action (5,6). In contrast, the endodomain is capable of low affinity hormone binding (5,7) and receptor activation (6, 8 -10). Low affinity hormone binding and initial receptor activation are likely to involve the exoloops and TM helices of the endodomain. The structure of these domains probably is affected by the organization of TM helices.
A unique feature of TM helices is the presence of Pro (11), in contrast to the general lack of Pro in helices of globular proteins (12). Pro disrupts helices (13) and introduces a ϳ26°kink (14) in the helix backbone (15). The kink in TM helices is thought to orient so that the Pro-containing convex side is adjacent to neighboring TM helices, whereas the concave side faces the lipid phase (12). As a result, TM Pros are thought to influence the packing and organization of TM helices and thus, the mechanism of receptor activation (16). Despite their potential significance, the five TM Pros of the LH/CG receptor have not been examined together. Herein, we report the importance of two of the Pros in hormone binding and targeting of the receptor to the surface membrane.

Mutagenesis and Functional Expression of LH/CG Receptors-Mu-
tant LH/CG receptor cDNAs were prepared in the pSELECT vector using the Altered Sites Mutagenesis System (Promega) and subcloned into pcDNA3 (Invitrogen) as described previously (9). Mutated LH/CG receptor plasmids 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 G418. They were used for hormone binding, cAMP production, antibody binding, and fluorescence microscopy. 125 I-hCG Binding and Intracellular cAMP Assay-Stable cells were assayed for 125 I-hCG binding in the presence of 150,000 cpm of 125 I-hCG (17) and increasing concentrations of unlabeled hCG. The K d values were determined by Scatchard plots. hCG, batch CR 127, was supplied by the National Hormone and Pituitary Program. For intracellular cAMP assay, cells were washed twice with Dulbecco's modified Eagle's medium and incubated in medium containing isobutylmethylxanthine (0.1 mg/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°, the supernatant was collected, dried under vacuum, and resuspended in 10 l of the cAMP assay buffer, which was provided by the manufacturer (Amersham Corp.). cAMP concentrations were determined with an 125 I-cAMP assay kit (Amersham Corp.) following the manufacturer's instruction and validated for use in our laboratory. 125 I-hCG Binding to Solubilized LH/CG Receptor-Transfected cells were washed twice with ice-cold 150 mM NaCl, 20 mM HEPES, pH 7.4 (buffer A). Cells were scraped on ice, collected in buffer A containing protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 5 mM N-ethylmaleimide, and 10 mM EDTA), and pelleted by centrifugation at 1300 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 125 I-hCG and 6.5 l of solution containing increasing concentrations of unlabeled hCG. Varying concentrations of unlabeled hCG were dissolved in 0.9% NaCl and 10 mM Na 2 HPO 4 at pH 7.4. Binding assay mixtures were incubated at 4°C for 12 h. This solution was, then, 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 were removed. Pellets were resuspended in 1.5 ml of buffer A containing 20% polyethylene glycol 8000, centrifuged, and counted for radioactivity.
Immunofluorescence Microscopy-For fluorescence labeling of LH/CG receptors, the flag epitope (18), Asp-Tyr-Lys-Asp-Asp-Asp-Asp-* This work was supported by National Institutes of Health Grants HD-18702 and DK-51469. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ‡ The first two authors made equal contributions to this work. § To whom all correspondence should be addressed. Tel.: 307-766-6272; Fax: 307-766-5098; E-mail: Ji@plains.uwyo.edu. 1 The abbreviations used are: LH/CG receptor; LH, luteinizing hormone (lutropin); CG, chorionic gonadotropin (choriogonadotropin); TM, transmembrane; hCG, human chorionic gonadotropin; LH/CG-R, PBS, phosphate-buffered saline.
Lys (5Ј-GAC TAC AAG GAC GAT GAC GAT AAG-3Ј), was inserted between the C terminus (Ser 26 ) of the signal sequence and the N terminus (Arg 27 ) of mature receptors. The flag epitope (19) has successfully been used as a marker to identify, trace, and purify recombinant proteins carrying the tag without significantly impairing their biological activities (20,21). Intact cells were cultured on coverslips in 6-well plates for 2 days and fixed with 4% formaldehyde in PBS for 10 min at 25°. For labeling permeabilized cells, cells were fixed with 4% formaldehyde in PBS for 5 min at 4°and treated with 0.1% Triton X-100 in PBS for 5 min at 4°. Fixed intact or permeabilized cells were washed five times with PBS for 2.5 min each at 25°. They were sequentially treated with 0.4% type IV gelatin isolated from calf skin (Sigma) in modified Eagle's medium free of phenol red for 10 min at 25°and then with 5% goat serum and 1% fetal calf serum in the same medium for 20 min at 25°. The treated cells were incubated with 500 l/well of primary antibody solution (25 g of mouse anti-flag antibody in 1 ml of modified Eagle's medium containing 5% goat serum and 1% fetal calf serum) for 2 h at 37°. The cells were washed three times with PBS for 2.5 min each at 25°C and treated with a 400-fold dilution of Texas Red conjugated to goat anti-mouse IgG (Molecular Probes). Finally, the cells were washed with PBS six times for 2.5 min each at 25°C. The coverslip containing processed cells was mounted on glass slides using 50% glycerol in PBS and sealed using nail polish. Specimens were examined under a Leica TCS-4D laser scanning confocal microscope equipped with Scanware analysis software. The entire experiments were completed in a day to prevent increasing background fluorescence.
Radioimmunoassay for Flag-LH/CG Receptors-Mouse anti-flag monoclonal antibody M2 (Eastman Kodak Co.) was iodinated with 125 I according to the published procedure for radioiodination of hCG (17), and 125 I-anti-flag antibodies were purified on a Sephadex G-150 column. Binding of 125 I-anti-flag to 293 cells expressing flag-LH receptors was carried out following the 125 I-hCG binding assay described above.

RESULTS
There are five Pros in the seven TM helices of the LH/CG receptor, Pro 463 and Pro 479 in TM 4, Pro 562 in TM 6, and Pro 591 and Pro 598 in TM 7 (Fig. 1). These TM Pros were individually substituted with Phe to produce LH/CG-R P463F , LH/CG-R P479F , LH/CG-R P562F , LH/CG-R P591F , and LH/CG-R P598F . The mutant receptor plasmids were transfected into 293 cells. Stably transfected cells were assayed for 125 I-hCG binding and hCG induced cAMP synthesis. Counts of empty tubes (background) were ϳ50 CPM and nonspecific binding was ϳ100 CPM including background. Maximum specific binding counts/min are included in parentheses in the table section of the figure. Intact cells expressing LH/CG-R P463F , LH/CG-R P562F or LH/CG-R P591F bound hCG with K d values 44 -100% higher than the K d value of the wild type receptor ( Fig. 2A). They were also capable of inducing cAMP synthesis with EC 50 values 37-89% higher FIG. 2. Biological activities of Pro 3 Phe substitution mutants. Five TM Pros were individually substituted with Phe and the mutant receptors were stably expressed in human 293 cells. The cells were assayed for hormone binding and hCG dependent cAMP induction. Counts of empty tubes (background) were ϳ50 cpm, and nonspecific binding was ϳ100 cpm, including background. Maximum specific binding counts/min are included in parentheses in the table. Each experiment was performed in duplicate, and values were determined for K d , receptors/cell, EC 50 for cAMP synthesis, and maximal cAMP level. Experiments were repeated four to six times, and means and S.D. were presented in the table. The statistical significance of the data was analyzed by Student' t test. The resulting p values are presented as a for p Ͻ 0.001, b for p Ͻ 0.01, c for p Ͻ 0.02, and d for p Ͻ 0.05. NS, not significant; ND, not detectable. In addition, values for different mutants were compared with the corresponding values of the wild type receptor using analysis of variance with 95% confidence. A mutant with p Ͻ 0.05 was considered to be a significantly different from the wild type as discussed in the text. Nontransfected cells did not show specific binding of hCG. As a control to verify intended mutations and identify unintended changes, if any, in the cDNA sequence, mutant receptor cDNAs were reverted to wild type, and their sequences were confirmed. The revertant cDNAs were used to transfect cells and assayed for hormone binding and cAMP induction. All revertants behaved the same as the wild type (data not included).
FIG. 1. TM prolines and their conservation. The location of TM Pros is shown in the putative topology model of the LH/CG receptor. In some receptors, the flag epitope, Asp-Tyr-Lys-Asp-Asp-Asp-Asp-Lys (shown in squares), was inserted at the N terminus of the mature receptor lacking the signal sequence. The conservation of Pros is shown for glycoprotein hormone receptors and other seven-TM receptors. Conserved and diverse residues are marked as "ϩ" and "Ϫ," respectively. than that of the wild type receptor (Fig. 2C). These results suggest that the high EC 50 values are due to the corresponding high K d values. Also these hCG binding and hCG-dependent cAMP induction are specific for the receptor since nontransfected cells and mock transfected cells did not bind 125 I-hCG nor induced hCG-dependent cAMP production. In contrast to the functional mutants, LH/CG-R P479F and LH/CG-R P598F did not bind hCG or induce cAMP production at the normal hCG concentrations. However, at ϾM hCG concentrations, LH/CG-R P479F showed some hCG binding but not cAMP induction (data not shown). This result does not clearly distinguish whether the apparent low affinity hormone binding of LH/CG-R P479F is due to a reduction in binding affinity or to low surface expression.
Solubilized Receptors-Since these two mutant receptors might be synthesized and trapped within cells, 125 I-hCG binding studies were performed with cells solubilized in Nonidet P-40. hCG binding to solubilized LH/CG-R P479F was barely detectable. In contrast, solubilized LH/CG-R P598F bound hCG (Fig. 2B) with affinity similar that of solubilized wild type receptor, LH/CG-R P463F , LH/CG-R P562F and LH/CG-R P591F .
However, the K d values of solubilized receptors are ϳ2-fold higher than those of the corresponding receptors on intact cells. This small loss in apparent binding affinities of solubilized receptors is often observed for wild type and mutant receptors (22). It may reflect either the low sensitivity of the assay or slight denaturation of receptors during the solubilization and/or assay steps. In any event, our data indicate that LH/ CG-R P598F is synthesized and capable of hCG binding. Furthermore, the result suggests that the mutant receptor might be trapped inside of the cells. Alternatively, LH/CG-R P598F may be expressed on the cell surface but is incapable of binding hCG in situ, perhaps due to defective folding. In this case, solubilization in the nonionic detergent may facilitate proper folding and promote hormone binding. In contrast, LH/CG-R P479F is either not synthesized or synthesized but incapable of binding hormone.
Antibody Binding-To further analyze expression and localization of the Pro 479 3 Phe and Pro 598 3 Phe mutants, two independent immunological methods were utilized. Cells were transfected with plasmids encoding receptors carrying the flag epitope, flag wild type LH/CG receptor, flag-LH/CG-R P479F , and flag-LH/CG-R P598F . For immunofluorescent labeling studies, cells were either examined intact or after treatment with Triton X-100 to permeabilize the plasma membrane and to allow the antibody to enter the cytosol. Confocal laser fluorescence microscopy shows bright fluorescence of the flag wild type receptor on intact cells and in permeabilized cells (Fig. 3). Cells expressing the wild type receptor lacking the flag tag did not showed fluorescence, regardless of permeabilization. In addition, the cells expressing flag wild type receptor did not show fluorescence when treated for fluorescence labeling without anti-flag antibody. These controls demonstrate that the flag wild type receptor is expressed on the cell surface and within cells. Flag-LH/CG-R P479F was also observed on intact and permeabilized cells, indicating it is expressed on the cell surface and within cells. On the other hand, flag-LH/CG-R P598F was observed in permeabilized cells only, indicating that flag-LH/ CG-R P598F was not transported to the cell surface.
The other approach used to test expression of the receptors was radioimmune ligand binding using radioiodinated monoclonal anti-flag antibody. 125 I-anti-flag antibody bound to the cells expressing the flag wild type receptor or flag-LH/CG-R P479F but not to the cells expressing flag-LH/CG-R P598F (Fig.  4). The 138,000/cell, excluding the possibility that the binding observed resulted from nonspecific binding. Taken together, these and the fluorescence microscopy indicate that 125 I-antiflag was capable of specifically detecting the flag epitope in the receptors. More importantly, the results demonstrate that the Flag-wild type receptor and flag-LH/CG-R P479F were expressed on the cell surface, whereas flag-LH/CG-R P598F was not.
Activities of Flag Receptors-To test whether the flag epitope might interfere with processing and activities of the flag-LH/CG receptors, hCG binding and cAMP induction were examined. The flag wild type receptor and the wild type receptor on intact cells and in solution bound hCG with similar affinities (Fig. 5, A and B). In addition, the flag wild type receptor was capable of hCG-dependent cAMP induction, although the EC 50 value for cAMP induction was ϳ3-fold higher than the value of the wild type receptor (Fig. 5C). Furthermore, the K d values of solubilized LH/CG-R P598F and flag-LH/CG-R P598F were similar (Figs. 2B and 5B). These data show that the flag-LH/CG receptors are active, although their potency is somewhat different from LH/CG receptors lacking the flag epitope. With this in mind, we examined flag-LH/CG-R P479F . While hCG did not bind to flag-LH/CG-R P479F on intact cells, hCG bound to the receptor in detergent solution with low affinity (Fig. 5B). This is not entirely surprising since hCG binding to LH/CG-R P479F in solution was barely detectable.
The loss of binding on intact cells and the extremely low affinity binding in solution of flag-LH/CG-R P479F could be due to the introduction of the bulky phenyl side chain or the loss of the Pro induced kink at this position in TM helix 4. To distinguish these possibilities, Pro 479 was substituted with Gly and Ala to produce Flag-LH/CG-R P479G and Flag-LH/CG-R P479A , respectively. Both mutants failed to bind hCG on intact cells and induce hCG dependent cAMP synthesis (Fig. 6, A and B).
On the other hand, they were capable of binding hCG in detergent solution (Fig. 6, C and D). K d values were ϳ2-fold higher than that of the flag wild type receptor. In addition, receptor concentrations were greater than that of the flag wild type receptor. Again, it appeared possible that both mutants may not be expressed on the cell surface or may be expressed on the surface but defective in hCG binding. In the latter case, they may have undergone changes in the structure when solubilized in the detergent solution. To resolve these possibilities, cells transfected with plasmids carrying the mutant cDNAs were treated with anti-flag antibodies and examined under confocal microscopy. The results in Fig. 7 demonstrate that flag-LH/CG-R P479G and Flag-LH/CG-R P479A are present both on the surface of intact cells and within permeabilized cells.

DISCUSSION
In this study, the five TM Pros of the LH/CG receptor were individually substituted with Phe and the resulting mutant receptors were examined for hormone binding and cAMP in- duction. In addition, surface expression of nonbinding receptors were characterized with flag-LH/CG receptors using immunofluorescence microscopy and antibody binding.
The Pro 479 3 Phe and Pro 598 3 Phe substitutions caused dramatic, yet contrasting effects (Table I). The Pro 598 3 Phe substitution prevented surface expression without affecting hormone binding. In contrast, the Pro 479 3 Phe substitution allowed surface expression but impaired hCG binding to the receptor on the cell surface as did Pro 479 3 Gly and Pro 479 3 Ala substitutions. Surprisingly, solubilization in Nonidet P-40 rescued the ability of the mutants to bind the hormone. This novel observation demonstrates a striking difference in hormone binding on intact cells and in nonionic detergent solution. A simple explanation is that, during solubilization, the mutant receptors undergo structural changes leading to a structure that more closely resembles the native receptor. Alternatively, the structure of the mutant receptors may be subtly yet critically different depending on whether they are embedded in the lipid bilayer or associated with the nonionic detergent in solution. It is also possible that hormone binding of the mutants might be interfered with due to other membrane molecules which were present in the membrane but removed by detergent solubilization.
Although substitutions of Gly, Ala, and Phe for Pro 479 impacted hormone binding to solubilized receptors, the extents were significantly different. The K d value of flag-LH/CG-R P479F is significantly higher than those of flag-LH/CG-R P479G and flag-LH/CG-R P479A . One wonders whether the bulky phenyl side chain of Phe 479 interferes the interaction of TM 4 with a neighboring TM, hormone binding, and/or structural restoration during solubilization. A similar result was observed when Pro 556 of the thyrotropin receptor, which is equivalent to Pro 479 of the LH/CG receptor, was replaced with Leu in the hypothyroid hyt/hyt mouse (23,24). The thyrotropin receptor mutant was expressed on the cell surface but was incapable of binding hormone. Therefore, the substitution of bulky amino acids for the TM 4 Pro appears to be intolerable for hormone binding at the cell surface. Also, the Pro 479 -induced kink does not appear to play a significant role on hormone binding, since the Pro 479 3 Gly and Pro 479 3 Ala substitutions had the same effect, although the helix is expected to bend by Gly but not by Ala. Taken together, these results indicate that the side chain of Pro 479 is more important than the Pro 479 -induced kink for hormone binding, particularly at the cell surface. On the other hand, Pro 479 is not required for surface expression of the receptor.
The substitution of Phe for Pro 463 of TM 4, Pro 562 of TM 6, and Pro 591 of TM 7 slightly increased the K d values for hCG binding on intact cells and in detergent solution as well as the EC 50 values for cAMP induction. However, the substitutions did not significantly impact the maximal cAMP level. The results indicate that these three TM Pros play marginal roles on hormone binding and cAMP induction, and have no effect on surface expression.
The marginal effect of the Pro 463 3 Phe substitution is not surprising. Pro 463 in TM 4 is variant among the LH/CG receptors of different species and is generally absent in other glycoprotein hormone receptors and seven TM receptors. On the other hand, Pro 479 of TM 4 and Pro 598 of TM 7 are conserved throughout seven TM receptors. Likewise, Pro 562 of TM 6 and Pro 591 of TM 7 are conserved among the glycoprotein hormone receptors and are partially conserved among other seven TM receptors. Pro 463 may be nonessential because it is located at the third amino acid position from the N terminus of the TM 4 helix. Although internal Pros are helix destabilizers, a Pro within the first three positions does not destabilize an ␣ helix (12), because amide nitrogens of the first four residues in an ␣ helix do not form hydrogen bonds.
Pro 562 is located in the middle of TM 6 and Pro 591 in the N-terminal one-third of TM 7. Consequently, the loss of either Pro induced kink may affect exoloop 3. Since these Pro 3 Phe substitutions only slightly reduced the high affinity for hormone binding, exoloop 3 may not play a crucial role in high affinity hormone binding. These findings are consistent with the observation that exoloop 3 of the LH/CG receptor is essential for receptor activation but is only marginally important for high affinity hormone binding 2 (22). Also the results are in accord with studies showing that a synthetic peptide corresponding to the exoloop 3 sequence of the LH/CG receptor inhibits 125 I-hCG binding to the LH/CG receptor only slightly (25).
Finally, the data in this study also demonstrate the utility of flag epitopes for the localization, transport, and quantification of glycoprotein hormone receptors. Introduced epitopes do not significantly impact the biological activity of the receptor, which in turn provides a sound methodology for studying their surface expression and activation.