Lys583 in the third extracellular loop of the lutropin/choriogonadotropin receptor is critical for signaling.

The lutropin/choriogonadotropin receptor (LH/CG-R) contains a relatively large extracellular domain, in addition to the seven transmembrane helices (TMH), three extracellular loops (ECL), and three intracellular loops typical of G protein-coupled receptors. While high affinity ligand binding has been attributed to the N-terminal extracellular domain, there is evidence that portions of the three ECLs may function in ligand binding and transmembrane signaling. We have investigated the role of several ionizable amino acid residues of rat LH/CG-R in human choriogonadotropin (hCG) binding and hCG-mediated cAMP production. COS-7 cells were transfected with the pSVL expression vector containing cDNAs of either wild-type or mutant rat LH/CG-R. Several point mutants of Lys583, located at the interface of ECL III and TMH VII, bound hCG like wild-type receptor but exhibited greatly diminished ligand-mediated signaling. Neither the point mutant, Lys401 → Asp (ECL I), nor the double mutant, Asp>397ed cells, only the double mutant bound hCG. The mutants Arg341 → Glu (interface of the extracellular domain and TMH I) and Lys488 → Glu (ECL II) proved to be similar to wild-type receptor in binding and signaling. Our results establish that Lys583 is important in signaling but not ligand binding. Its location on the opposite side of the membrane from Gs precludes a direct interaction, thus emphasizing the importance of a conformational change in the receptor and suggesting that ligand binding to receptor and ligand-mediated receptor activation are dissociable phenomena.

pal mediator of the actions of gonadotropins on most gonadal cells, there is evidence to support activation of the phospholipase C pathway as well, which results in the formation of inositol 1,4,5-trisphosphate and increased [Ca 2ϩ ] levels (3)(4)(5).
The LH/CG-R, FSH-R, and TSH-R are members of the glycoprotein hormone receptor family, characterized by a relatively large extracellular N-terminal region and a membraneembedded C-terminal region containing seven TMHs. This C-terminal region is homologous to the small ligand binding members of the G protein-coupled receptor superfamily. However, unlike the small ligand binding receptors, where binding occurs in a cleft formed by the TMHs, the glycoprotein hormone receptors utilize their extracellular domain as the high affinity binding site for the heterodimeric glycoprotein hormones with molecular masses of 30 -37 kDa (6 -10). This structural difference classifies these receptors as a distinct subfamily of the G protein-coupled receptor superfamily (11).
The N-and C-terminal domains of the rat LH/CG-R each contain over 300 amino acid residues, the latter distributed over three ECLs, seven TMHs, three intracellular loops, and a cytoplasmic tail (1). If the LH/CG-R spans the membrane similar to the small ligand binding G protein-coupled receptors, one would expect the seven putative TMHs to form a pocket like that of the bacteriorhodopsin and rhodopsin receptors (12,13); the six hydrophilic connecting loops are essential in maintaining this conformation. Additionally, although the high affinity binding site of the LH/CG-R is located in the ECD, there is evidence to support the presence of a lower affinity binding site in the C-terminal domain of the receptor (14,15). Therefore, the ECLs represent potential hormone contact sites.
In an attempt to define the role of the LH/CG-R ECLs in hormone binding and signaling, several Arg and Lys residues were replaced: Arg 341 (at the boundary between the ECD and TMH I), Lys 401 (ECL I), Lys 488 (ECL II) and Lys 583 (ECL III). Additionally, a reciprocal mutation of Asp 397 and Lys 583 was characterized. The relative positions of these amino acid residues are shown in Fig. 1. These particular residues are invariant at the homologous positions in the LH/CG-R and the FSH-R of all known species; in the TSH-R, the positions equivalent to residues 401 and 583 are His and Gly, respectively, and the other residues are invariant. Our results revealed that Lys 583 is not involved in hormone binding but is essential for full receptor activation. (Madison, WI). The plasmid Maxiprep DNA purification kit was obtained from Qiagen, Inc. (Chatsworth, CA). Lipofectamine, fetal bovine serum, trypsin-EDTA, Waymouth's medium, gentamicin, and penicillin-streptomycin were from Life Technologies, Inc. DEAE-dextran, isobutylmethylxanthine, chloroquine, triacetylchitotriose, BSA, protein molecular size markers, and Nonidet P-40 were purchased from Sigma. The Enhanced Chemiluminescence Western blotting analysis system was from Amersham Life Sciences, and the BCA protein assay system was purchased from Pierce. Agarose-bound wheat germ agglutinin was a product of Vector Laboratories (Burlingame, CA), and Centricon-30 columns were obtained from Amicon (Beverly, MA). Immobilon P transfer membrane was purchased from Millipore (Bedford, MA), and glass fiber filters were from Whatman (Maidstone, UK). Polyclonal anti-LH/ CG-R was kindly provided by Dr. Deborah Segaloff (University of Iowa, Iowa City, IA). Most other reagents were purchased from Sigma, Life Technologies, Inc., or Fisher.

Materials-
Mutant cDNAs of the Rat LH/CG-R-The cDNA for the rat LH/CG-R, inserted into the XbaI-BamHI site of the expression vector pSVL, was the generous gift of Dr. William Moyle (Robert Wood Johnson Medical School, Piscataway, NJ). The 21-base deoxyoligonucleotides coding for the appropriate codon changes were synthesized by Dr. Rudolf Werner (University of Miami, Miami, FL) and by the Molecular Genetics Instrumentation Facility at the University of Georgia. In vitro mutagenesis was performed (16) and mutant clones identified by dideoxy sequencing (17). Mutant cDNAs were amplified and DNA was obtained using the Qiagen plasmid Maxiprep kit.
Expression of the Rat LH/CG-R-COS-7 cells were kindly provided by Dr. Nevis Fregien (University of Miami, Miami, FL) and also purchased from the American Type Culture Collection (Rockville, MD). The cells, maintained at 37°C in humidified air containing 5% CO 2 in 90% DMEM and 10% fetal bovine serum, with 100 units/ml each penicillin and streptomycin, were transiently transfected with the eukaryotic expression plasmid pSVL-LH/CG-R, wild-type and mutants, using the DEAE-dextran method (18) or the Lipofectamine method as recommended by Life Technologies, Inc. For DEAE-dextran transfection, 20 g each of pSVL-LH/CG-R and pSV-␤-gal, to estimate transfection efficiencies, were added to a 10-ml solution of 90% DMEM, 10% NuSerum transfection medium containing a 0.4 mg/ml DEAE-dextran and 0.1 mM chloroquine, and the washed cells (ϳ4 ϫ 10 6 ) were incubated for 3.5-4 h at 37°C. After removal of the transfection medium, the cells were shocked for 2 min with 10% Me 2 SO; then COS-7 growth medium was added to the washed cells. For transfections with Lipofectamine, a 7.85-ml solution of serum-free DMEM, containing Lipofectamine (57 l/ml), 10 g each of the pSVL-LH/CG-R construct and the pSV-␤-gal plasmid and 10% (v/v) Opti-MEM media, was incubated with cells (ϳ2-3 ϫ 10 6 ) for 5 h at 37°C. The transfection medium was removed, the cells were washed, and COS-7 growth medium was added. With both types of transfection, the COS-7 cells were incubated overnight at 37°C in humidified air with 5% CO 2 . 5-Bromo-4-chloro-3indolyl-␤-D-galactoside staining of gluteraldehyde-fixed cells was used to estimate transfection efficiencies, which were roughly 10% for DEAE-dextran and 40% for Lipofectamine. 125 I-hCG Cell-surface Binding to Transfected Cells-The COS-7 cells were maintained for 16 h after transfection and then replated (5 ϫ 10 5 cells/well, six-well tissue culture plates). Some 48 -51 h post-transfection, the cells were about 70% confluent. Cells were then washed twice with serum-free Waymouth's medium containing 1 mg of BSA/ml, and 1 ml of this media was added to each well. Increasing concentrations of unlabeled hCG were then added to each well, followed by addition of 25 pM 125 I-hCG (approximately 10 5 cpm). Total and nonspecific binding were determined by addition of 125 I-hCG in the absence and presence of excess unlabeled hCG (54 nM). The plates were incubated at 25°C for 16 -18 h with gentle shaking. The cells were washed twice with cold phosphate-buffered saline, then trypsinized, collected, and counted in a ␥ counter. All determinations were performed in duplicate. Binding affinities and maximal binding capacities were calculated using the Ligand program (19).
125 I-hCG Binding to Detergent-soluble Extracts-The transfected cells were maintained for 16 h after transfection and then replated (2-2.5 ϫ 10 6 cells/dish, 10-cm tissue culture dishes). The protocol is adapted from that described by others (6,20,21). About 48 h posttransfection, the transfected cells were placed on ice for 15 min and then washed twice with 5 ml of ice-cold 0.15 M NaCl, 20 mM Hepes, pH 7.4 (buffer A). Cells were scraped into 2 ml of cold buffer A containing 1 mM phenylmethylsulfonyl fluoride, 2 mM EDTA, and 5 mM N-ethylmaleimide and pelleted by centrifugation at 2000 ϫ g for 20 min at 4°C. The pellet was resuspended in 0.25 ml of 1% Nonidet P-40, 20% glycerol in buffer A containing protease inhibitors and incubated on ice for 15 min. This mixture was centrifuged at 16,000 ϫ g for 15 min at 4°C. The supernatant was diluted with 2.25 ml of 20% glycerol in buffer A; 0.5 ml of the extract was incubated for 16 -18 h at 4°C with 50 pM 125 I-hCG (10 5 cpm). Total and nonspecific binding were determined in the pres- ence and absence of excess unlabeled hCG, respectively. Bound radioactivity was separated from unbound by filtration through Whatman GF/B filters that were previously soaked in 0.3% polyethylenimine in 10 mM Tris-HCl, pH 9.1 (22). The filters were washed five times with 0.1 M NaCl, 10 mM NaN 3 , 1 mg of BSA/ml in phosphate-buffered saline and counted in a ␥ counter. All determinations were performed in duplicate.
Intracellular cAMP Assay-Some 16 -18 h after transfection, the transfected cells were replated (1 ϫ 10 5 cells/well, 12-well tissue culture plates). At 48 -51 h post-transfection, the cells were washed twice with DMEM containing 1 mg of BSA/ml and incubated in 0.5 ml of this medium with 0.8 mM isobutylmethylxanthine for 15 min at 37°C. Increasing concentrations of hCG were then added and the incubation was continued for 30 min at 37°C. The cells were washed twice with fresh medium without isobutylmethylxanthine and then lysed in 100% ethanol at Ϫ20°C overnight. The extract was collected, dried under nitrogen gas, and resuspended in the buffer of the 125 I-cAMP assay kit, and cAMP concentrations were determined by radioimmunoassay. All measurements were performed in duplicate; means and standard errors were calculated using the Prism program.
Partial Purification of the LH/CG-R-Transfected cells were maintained for 16 -20 h after transfection and then replated (2-2.5 ϫ 10 6 cells/dish, 10-cm tissue culture dishes). Some 48 -51 h post-transfection, detergent-soluble extracts were prepared as described above. The procedures for cell lysis and partial purification of wild-type and mutant LH/CG-Rs were based on reports from other laboratories (21,23,24). Following cell lysis with 1% Nonidet P-40 and centrifugation as described above, the supernatant was diluted 2-fold with 10% glycerol in buffer A, then loaded onto a small agarose-bound wheat germ agglutinin column equilibrated with 0.1% Nonidet P-40 and 10% glycerol in buffer A, and rotated at 4°C overnight. Following extensive washing with this buffer, the LH/CG-R was eluted with one column volume of 3 mM triacetylchitotriose (24) containing 1 mM phenylmethylsulfonyl fluoride, 2 mM EDTA, and 5 mM N-ethylmaleimide. The eluted material was concentrated to approximately 0.1 ml in a Centricon-30 spin column, and the protein concentration was determined using the BCA assay.
SDS-PAGE and Western Blots-Equal amounts of purified cell lysates in sample buffer with no reducing agent were applied to a 7% SDS-polyacrylamide gel without boiling. After the electrophoresis, the proteins were electrophoretically transferred to polyvinylidene difluoride membranes, which were then washed twice in Tris-buffered saline (TBS: 20 mM Tris-HCl, pH 7.5, 0.5 M NaCl) and blocked for 2 h at room temperature in blocking solution (10% glycerol, 5% instant nonfat dry milk, 0.2% Tween 20 in phosphate-buffered saline). The filters were incubated overnight at room temperature with the same blocking solution containing 3 g/ml rabbit anti-LH/CG-R IgGs, obtained by protein A-Sepharose purification of anti-LH/CG-R antiserum (23). The membranes were washed five times for 5 min each with blocking solution and then incubated for 1 h at room temperature with a 1:5000 dilution of a horseradish peroxidase-labeled donkey anti-rabbit IgG whole antibody in blocking solution. Next, the membranes were washed as follows: twice with TBS, twice with 1% Nonidet P-40 in TBS, then once each with TBS, 1% Nonidet P-40 in TBS, and TBS alone. The membranes were exposed to an Amersham Enhanced Chemiluminescence developer solution for 1 min, wrapped in Saran Wrap, and exposed to Kodak XAR-5 film for 1 min. The films were densitometrically scanned (PDI System, Huntington, NY).

RESULTS
The mean K d from multiple independent transfections (n ϭ 9) for hCG binding to wild-type LH/CG-R on transfected intact COS-7 cells was 0.14 nM, with a range of 0.07-0.25 nM (Table I), there being no difference between transfections with DEAEdextran and Lipofectamine. On the other hand, receptor numbers/cell (uncorrected for transfection efficiencies) were greater with Lipofectamine transfection (mean of 2 ϫ 10 4 ) than with DEAE-dextran transfection (mean of 0.5 ϫ 10 4 ). The difference in receptor number/cell, however, had no significant impact on the maximal cAMP production elicited by hCG at 100 ng/ml in COS-7 cells transfected with the cDNA to wild-type LH/CG-R; the mean was 20.6 (range ϭ 12.3-31.7) pmol of cAMP/10 5 cells under the conditions used and was independent of receptor density over the range investigated (Table I). Likewise, the effective mean concentration of hCG necessary to increase the intracellular cAMP concentration from basal values to 50% of the maximal value in COS-7 cells expressing wild-type LH/ CG-R was 0.19 nM (range ϭ 0.05-0.4 nM) (Table I); this too was independent of receptor density. For ease in comparing the different experiments, the maximal hCG-mediated cAMP production over basal is normalized to 100% for each wild-type LH/CG-R control, and the value obtained with the various receptor mutants, also corrected for basal, is expressed as a percentage of that of wild-type. There was no evidence for a functional endogenous LH/CG-R in nontransfected COS-7 cells: the mean (range) of the basal cAMP production was 3.9 (1.6 -6.7) pmol cAMP/10 5 cells and that of control cells in the presence of 100 ng/ml hCG was 4.6(3.8 -5.2) pmol cAMP/10 5 cells.
The Lys 583 3 Glu mutation resulted in a LH/CG-R that specifically bound 125 I-hCG, and this binding was inhibited by unlabeled hCG in a concentration-dependent manner ( Fig. 2A). Although the estimated number of receptors varied in the transfected cells, e.g. 2 ϫ 10 3 for wild-type and 1 ϫ 10 3 for mutant LH/CG-R (uncorrected for transfection efficiency), the mutant LH/CG-R yielded a binding affinity equivalent to that of wild-type LH/CG-R (Table I). The Lys 583 3 Glu replacement resulted in markedly decreased production of cAMP in response to added hCG, e.g. 15% that of wild-type LH/CG-R at 100 ng/ml hCG (Fig. 3A, Table I). Since this result could be attributed to the decreased number of cell surface receptors, this mutant LH/CG-R was also transfected using Lipofectamine to obtain higher efficiency and increased receptor expression. The mutant receptor bound hCG with an affinity comparable to wild-type LH/CG-R (Table I). There was an increase in cell surface expression, e.g. 2 ϫ 10 4 receptors/cells (uncorrected for transfection efficiency), for both wild-type and mutant LH/CG-R, but the mutant receptor again stimulated cAMP production only 15% that of wild-type LH/CG-R at 100 ng/ml hCG (Table I).

TABLE I Values of K d , EC 50 , and R max in intact COS-7 cells expressing wildtype and mutant LH/CG-Rs
The results are given as mean Ϯ S.E. for the number of independent transfections listed (n). K d refers to the dissociation constant describing hCG-LH/CG-R binding, and EC 50 represents the effective concentration of hCG required to increase the intracellular cAMP concentration from the basal value (mean ϭ 3.9 and range ϭ 1.6 -6.7 pmol of cAMP/10 5 cells) to 50% of the maximal value (mean ϭ 20.6 and range ϭ 12.3-31.7 pmol of cAMP/10 5 cells). Both K d and EC 50 are given in units of nM. R max refers to the maximal cAMP produced by the transfected cells, corrected for basal production, in response to a saturating concentration of hCG added to cells expressing wild-type LH/CG-R, e.g. 100 ng/ml. Within each experimental group, the maximal values, corrected for basal production, were normalized to 100% for wild-type LH/CG-R, and the amount of cAMP produced by mutants at 100 ng/ml added hCG, also corrected for basal production, are given relative to the normalized value for wild-type in each particular set of transfections.
a Cell transfections were with DEAE-dextran. b Cell transfections were with Lipofectamine. c For these mutants the cAMP response to added hCG was too little to obtain an accurate EC 50 . At the highest concentration of hCG used (ϳ3 nM), the cAMP response was Ͻ30% that obtained with wild-type LH/CG-R.
To investigate the side-chain specificity of Lys 583 , replacements were also made with Arg, Gln, and Pro. In each case the mutations yielded LH/CG-Rs that bound 125 I-hCG with affinities comparable to that of wild-type LH/CG-R (Fig. 2B, Table I), but these mutant receptors also resulted in cAMP accumulation Ͻ30% that of wild-type LH/CG-R upon addition of hCG to transfected cells (Fig. 3B, Table I).
Competitive binding assays were unable to detect significant cell surface binding of 125 I-hCG to cells transfected with cDNAs to the Lys 401 3 Asp single mutation and the Asp 397 3 Lys/ Lys 583 3 Asp reciprocal mutation (Table II). 125 I-hCG binding to detergent-soluble extracts of transfected cells was found for the reciprocal mutant but not the point mutant (Table II). However, Western blot analysis indicated that the (Lys 401 3 Asp) LH/CG-R was expressed (Fig. 4). Wild-type and mutant LH/CG-Rs were about equally distributed among three bands of apparent molecular mass 101, 93, and 82 kDa; the total protein in the mutant LH/CG-R was much less than that of

FIG. 2. hCG binding to COS-7 cells transfected with wild-type and mutant cDNAs to LH/CG-R.
Competition binding of 125 I-hCG and hCG to wild-type LH/CG-R and six single mutants of LH/CG-R are given in panels A-D. The binding of 125 I-hCG with no added hCG was normalized to 100% in each case.

FIG. 3. cAMP levels in COS-7 cells transfected with wild-type and mutant cDNAs to LH/CG-R in the presence and absence of hCG.
Panels A-D give data on six single mutant forms of LH/CG-R; in each case, results are provided for cells expressing wild-type LH/CG-R and for control cells.
Arg 341 and Lys 488 were each replaced with Glu; both mutations yielded LH/CG-Rs that bound 125 I-hCG with affinities comparable to that of wild-type LH/CG-R (Fig. 2, C and D, Table I). These mutant LH/CG-Rs were capable of stimulating cAMP production in a dose-dependent manner, quite similar to wild-type LH/CG-R (Fig. 3, C and D, Table I).

DISCUSSION
Our results on the substitution of Lys 583 of the rat LH/CG-R with 4 amino acid residues, positively charged Arg, negatively charged Glu, polar but nonionizable Gln, and nonpolar Pro (an imino acid), show that the mutant LH/CG-Rs bind hCG as well as wild-type receptor, but coupling to adenylate cyclase is greatly diminished. These findings are particularly intriguing since Lys 583 is located extracellularly, at the boundary of ECL III and TMH VII, and is unable to interact directly with G s . Thus, one can conclude that ligand-mediated transmembrane signaling involves a conformational change of the LH/CG-R in which Lys 583 participates in some manner. Furthermore, this observation provides additional evidence that receptor activation/signaling can be dissociated from high affinity hormone binding in LH/CG-R. A similar conclusion was recently reached in reference to another portion of the rat LH/CG-R, namely the extracellular region just prior to TMH I (25). In that study we showed that individual replacements of Glu 332 and Asp 333 with Lys yielded mutant receptors that also bound hCG like wildtype LH/CG-R but exhibited greatly diminished signaling. Interestingly, recent results with FSH indicate that different sites are involved in receptor binding and signal transduction (26).
The finding that the [Arg 583 ]LH/CG-R mutant, in which one positively charged side chain is replaced with another, does not signal effectively indicates a stringent specificity for the lysine side chain. Interestingly, a charged residue at the interface of ECL III and TMH VII is not required for proper membrane localization since the replacements of Lys 583 with Gln and Pro yielded mutants that gave levels of cell surface expression comparable to that of wild-type receptor.
There is no information on the type of interaction between Lys 583 of LH/CG-R and either the ligand or another portion of the receptor. Since replacements of Lys 583 have no appreciable effect on hCG binding, one would conclude that any such interaction between the hormone and Lys 583 makes a negligible contribution to the overall free energy of binding. However, high affinity hormone binding may occur to the ECD, followed by a conformational change that "presents" the hormone to Lys 583 in ECL III where low affinity binding may occur. Using overlapping synthetic peptides, Roche et al. (15) found that a peptide based on the sequence of rat ECL III inhibited the binding of 125 I-hCG to rat luteal membranes with an IC 50 of 0.2 mM. The most obvious interaction of Lys 583 would be electrostatic, e.g. an ion-dipole or an ion-ion bond between the positively charged amino group and a negatively charged carboxyl group on hCG or on the receptor ECD or ECLs. Alternatively, a more hydrophobic environment could result in an abnormal pK of Lys 583 , which could lead to the formation of a strong hydrogen bond. Since our results suggest some form of ligandmediated conformational change of the receptor, it is reasonable to expect that Lys 583 of the LH/CG-R may flip from one type of interaction to another concomitant with hormone binding and that this change accompanies, or perhaps even triggers, a change in the conformation or relative interhelical position of TMH VII.
Ji et al. (27) reported that Asp 397 of the LH/CG-R, located at the boundary of TMH II and ECL I, formed an ion pair with Lys 91 of the ␣-subunit of hCG. They postulated that ligand binding may alter the conformation of the unoccupied receptor by reorienting TMH II (28) and concluded that this residue was important in signaling but was not essential for hormone binding (29), i.e. as we found for Lys 583 .
Whatever the exact role of Asp 397 in LH/CG-R function, we asked whether it and Lys 583 could participate in an ion-ion interaction. This was based in part on the close proximity of TMHs II and VII suggested by projection maps of rhodopsin (13) and by reciprocal ion-pair mutations in the gonadotropinreleasing hormone receptor (30). Moreover, interhelical interactions of TMH I, in proximity to II and VII, have been shown for the adrenergic receptors (31). We found that the reciprocal mutation failed to express in a functional form at the cell surface, although hormone binding could be detected in detergent-solubilized cells. Thus, Asp 397 and Lys 583 can be individually replaced with oppositely charged amino acid residues and cell surface expression is obtained, but replacing both residues results in intracellular trapping.
Lys 583 is invariant in the gonadotropin receptors, i.e. LH/ CG-R and FSH-R, but in TSH-R the equivalent position is occupied by Gly (2). In a revealing study, ECL III of the TSH-R was replaced with that from the ␤ 2 -adrenergic receptor; it was found that the mutant TSH-R bound TSH with high affinity, but the sensitivity of TSH-stimulated cAMP production and the maximal level of TSH-mediated cAMP production were significantly diminished in the mutant TSH-R (32). Coupled with our studies on Lys 583 in ECL III of the LH/CG-R, these results indicate an important role of ECL III in glycoprotein hormone signaling and a possible region of specificity delineating gonadotropin receptors from the TSH-R.
Two other positively charged residues of LH/CG-R, which are invariant in the glycoprotein hormone receptors, Arg 341 (locat-  583 3 Asp]LH/CG-R), and specific binding was determined using 50 pM 125 I-hCG with intact cells and following solubilization. For purposes of comparison, the specific binding to wild-type LH/CG-R was normalized to 100% and the specific binding of each of the two mutants is given relative to that value. The total binding of 125 I-hCG to wild-type LH/CG-R varied from about 4000 -9000 cpm, and nonspecific binding was generally Ͻ5% of the total binding to intact cells and about 20% in the soluble binding assay. The results are given as mean Ϯ S.E. from two transfections. ed in the ECD at the interface with TMH I) and Lys 488 (ECL II) were each replaced with Glu with no observable effect on hCG binding or receptor activation. Lys 401 (ECL I) was replaced with Asp, and the mutant receptor failed to exhibit significant hCG binding to intact or detergent-solubilized cells, suggesting that this amino acid residue may be involved in hormone binding. However, binding may occur but is difficult to detect if there is a low level of mutant receptor expression, if the receptor is rapidly degraded or if the K d is significantly increased. Lys is present at this position in LH/CG-R and FSH-R, while in TSH-R it is occupied by His; depending upon the pK of the His, the potential exists for retention of a positive charge at this location in ECL I of all glycoprotein hormone receptors.
Our results on Western analysis of wild-type LH/CG-R and (Lys 401 3 Asp) LH/CG-R revealed the presence of three bands of apparent molecular mass 101, 93, and 82 kDa. The original Western blot experiments on expressed LH/CG-R were performed using human 293 cells stably transfected with the wildtype LH/CG-R cDNA (24). A major 85-kDa form, considered the mature receptor, and a minor 68-kDa form, which appeared to be the incompletely glycosylated form, were found. The apparent 93-kDa band observed in transiently transfected COS-7 cells corresponds to the M r of purified rat ovarian LH/CG-R (2), although wide variations have been reported in the M r of LH/ CG-R from various sources. In another study from our laboratory using a slightly different set of protein standards, we found major and minor bands of apparent molecular mass 93 kDa (ϳ80%) and 78 kDa for expressed wild-type LH/CG-R (25). The greater amount of protein loaded in that study could easily prevent resolution of the bands of 93 and 101 kDa reported herein. Interestingly, the Lys 401 3 Asp LH/CG-R mutant also gave the same apparent M r forms as wild-type LH/CG-R, but at much lower levels.
In summary, these results enable us to conclude that Lys 583 of the rat LH/CG-R is critical in receptor activation following hormone binding. Since Lys 583 (ECL III) is located on the opposite side of the membrane from G proteins and does not appear to be involved in hormone binding, these findings offer considerable weight to the concept that receptor binding and activation are dissociable phenomena. Of interest are other observations we recently made that certain amino acid residues in TMH VII are also critical for signal transduction (33). Thus, this region of the receptor appears to be important in transmembrane signaling subsequent to hormone binding.