| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLE |
Department of Biochemistry and Molecular Biology, University of Georgia, Athens, Georgia 30602
Address all correspondence and requests for reprints to: Dr. David Puett, Department of Biochemistry and Molecular Biology, B129 Life Sciences Building, 120 Green Street, University of Georgia, Athens, Georgia 30602-7229. E-mail: puett{at}bmb.uga.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-receptor-C), to create the yoked (Y) complexes YCG/LHR, YCG/TSHR, and YCG/FSHR. The expression and bioactivity of these fusion proteins were examined in transiently transfected HEK 293 cells. Western blot analysis and antibody binding assays demonstrated that each of the proteins was expressed. In the case of YCG/LHR, minimal binding of exogenous hormone was observed due to the continued occupation of receptor by the fused ligand. The presence of hCG in the YCG/TSHR and YCG/FSHR, however, did not prevent binding of exogenous cognate ligand, presumably due to the lower affinity of hCG. The basal cAMP levels in cells expressing the YCG/LHR complex was approximately 20-fold higher than that in cells expressing LHR. Increases in basal cAMP production were also observed with YCG/TSHR and YCG/FSHR, e.g. 13- and 4-fold increases, respectively. Whereas the affinity and specificity of hCG for LHR are extraordinarily high, the hormone is capable of binding to and activating both TSHR and FSHR under these conditions that mimic high ligand concentrations. These findings were confirmed by adding high concentrations of hCG to cells expressing TSHR and FSHR. Although the functional interaction of hCG and TSHR has been recognized in gestational hyperthyroidism, there are no reports linking hCG to FSHR activation. This study, however, suggests that such a functional interaction is capable of occurring under conditions of high circulating levels of hCG, e.g. the first trimester of pregnancy and in patients with hCG-secreting tumors. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit and a receptor-specific ß-subunit, and confer activity by binding to their respective G protein-coupled receptors (GPCRs) (2). The glycoprotein hormone receptors are a special class of GPCRs, characterized by their relatively large ectodomain, which is responsible for high affinity hormone binding (3). Because of the high sequence similarity in their ß-subunits, both hCG and LH bind to the same receptor, LH receptor (LHR), whereas TSH and FSH bind to unique receptors, TSHR and FSHR, respectively (3). There is a high degree of sequence homology among the glycoprotein hormone ß-subunits (Fig. 1A
) and among the glycoprotein hormone receptor ectodomains (Fig. 1B).
|
We and others have used fusion protein techniques to further delve into structure-function studies within the families of glycoprotein hormones and their GPCRs (10, 11). Single chain, or yoked (Y), hormones have been designed in the manner N-hCGß--C (10, 11, 12) and N--hCGß-C (13, 14, 15). Our group (10) and that of Boime (11) first designed YhCG1 with hCGß at the N terminus; we then created YhCG3 in which hCGß is located at the C-terminus (13). Both forms of single chain hCG were effective in activating LHR; however, YhCG1 was able to bind LHR with a higher affinity than that found with YhCG3. This fusion protein technique was later expanded to include yoked hormone-receptor complexes in which YhCG1 and YhCG3 were fused to the N terminus of rat LHR: N-hCGß--CTP-LHR-C and N--CTP-hCGß-LHR-C, respectively (13, 16). This approach of using a fused ligand-receptor complex permits the study of receptor binding and activation by individual subunits, subunit mutants that fail to efficiently form holoproteins, and ligand mutants that bind with low affinity to their cognate receptor.
As the glycoprotein hormones are comprised of a common -subunit and homologous ß-subunits, and their receptors are homologous as well, we have prepared and characterized fusion proteins of single chain hCG with the three human receptors using the configuration N-ß--CTP-R to yield YCG/LHR, YCG/TSHR, and YCG/FSHR (Fig. 2) to investigate the anticipated weak binding of hCG to and the activation of its noncognate receptors, TSHR and FSHR. Although hCG stimulation of TSHR results in pathophysiological conditions such as gestational hyperthyroidism (17), there are no known clinical cases involving hCG activation of FSHR. However, the noncognate hormone-receptor interaction involving hCG and TSHR documents that cross-reactivity occurs within these families.
|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cell culture and transient transfections
HEK 293 cells were grown in monolayer culture in growth medium consisting of DMEM fortified with 10% (vol/vol) newborn calf serum, 10 mM HEPES buffer (pH 7.4), 50 U/ml penicillin, 50 µg/ml streptomycin, 50 µg/ml fungizone, and 0.125 µg/ml amphotericin B. Cells were grown and sustained at 37 C in humidified air containing 5% CO2. The cDNAs were transiently transfected using Lipofectamine 2000 as recommended by the manufacturer (Life Technologies, Inc., Grand Island, NY) into 75-cm2 tissue culture flasks containing HEK 293 cells (5 µg cDNA for each fusion protein).
Western blot analysis
Solubilized membrane fractions were electrophoresed on 10% SDS-PAGE and transferred to polyvinylidene difluoride membranes. The membrane was probed with a 1:1000 dilution of an anti-CTP polyclonal antibody (provided by Dr. Vernon Stevens, Ohio State University, Columbus, OH) and visualized by chemiluminescence via a secondary horseradish peroxidase-labeled antirabbit antibody.
Hormone binding
Cells were resuspended in 36 ml growth medium approximately 16–18 h after transfection, and 1 ml medium was added to each well of a 12-well tissue culture plate coated with 0.1% gelatin in PBS. The cells were assayed for 125I-labeled hormone binding approximately 24 h later. To perform competitive binding assays, increasing concentrations of unlabeled urinary hCG, bovine (bTSH; both hCG and bTSH were provided by Dr. Albert Parlow and the NIDDK), or hFSH (Sigma, St. Louis, MO) in Waymouth’s medium with 0.1% BSA (wt/vol) and 50 pM [125I]hCG, 50 pM [125I]bTSH, or 100 pM [125I]hFSH, respectively, were added to each well containing cells expressing the YCG/R protein or cognate receptor and incubated at 37 C for 6 h. Limited competitive binding studies were also performed with recombinant hCG (Organon, Oss, The Netherlands) and each of the 125I-labeled hormones and their cognate receptors. Nonspecific binding was determined by the addition of 1 µg/ml unlabeled hCG and bTSH and 3 µg/ml unlabeled hFSH. The incubation medium was aspirated, and the cells washed with PBS and then lysed by the addition of 0.5 ml 1 N NaOH. All binding assays were performed in duplicate. Mock-transfected cells treated in the same manner were used as negative controls.
Cell surface antibody binding assay
Approximately 16–18 h after transfection, the cells were processed as described above, assayed for binding of a receptor-specific primary antibody, raised against a synthetic peptide, and then detected by secondary [125I]antirabbit antibody. For hLHR, a 1:4000 dilution of an antirat LHR antibody raised against residues 15–38 (provided by Dr. Patrick Roche, Mayo Medical School, Rochester, MN) was used. Dr. Mariusz Szkudlinski (Trophogen, Inc., Rockville, MD) donated an anti-hTSHR antibody raised against residues 352–366, and Dr. Mario Ascoli (University of Iowa, Iowa City, IA) provided an antirat FSHR antibody raised against residues 19–29. Dilutions of 1:1000 and 1:500 were made of the TSHR and FSHR antibodies, respectively. The cells were incubated in the appropriate dilution of the primary antibody, as indicated, in Waymouth’s medium with 0.3% (wt/vol) BSA for the LHR antibody and 0.1% (wt/vol) BSA for the TSHR and FSHR antibodies for 4 h at room temperature while shaking. After washing twice with Waymouth’s medium containing 0.3% (wt/vol) BSA for the LHR antibody and 0.1% (wt/vol) BSA for the TSHR and FSHR antibodies, the cells were incubated with [125I]antirabbit secondary antibody (400,000 cpm/well) in Waymouth’s medium containing 0.3% or 0.1% BSA for 1 h at room temperature with shaking. The medium was aspirated, and the cells were washed with PBS after the second incubation period. Lysates were prepared as described above, and the samples were counted in a 
-counter. Each binding experiment was performed in duplicate.
cAMP assay
The cells were replated and washed as described above some 16–18 h after transfection, then incubated in Waymouth’s medium with 0.1% (wt/vol) BSA and 0.8 mM isobutylmethylxanthine for 15 min at 37 C. The cells were incubated with increasing concentrations of cognate hormone, urinary hCG, bTSH, or hFSH, for 30 min at 37 C in Waymouth’s medium containing 0.1% (wt/vol) BSA and 0.8 mM isobutylmethylxanthine immediately following the first incubation. In other studies increasing concentrations of recombinant hCG were added to cells expressing each of the three glycoprotein hormone receptors. The incubation medium was removed, and the cells were lysed in 100% ethanol at -20 C overnight. The extract was collected, dried under vacuum, and cAMP concentrations were determined by RIA as recommended (NEN Life Science Products, Boston, MA). Each data point was performed in duplicate, and mock-transfected cells were used as negative controls.
Data analysis
Competitive binding and cAMP results were analyzed by nonlinear regression using the PRISM software program (GraphPad Software, Inc., San Diego, CA). The results are given as the mean ± SEM, based on three to five independent transfections, and significance was determined by t test. The figures showing competitive binding and hormone-mediated cAMP production are representative experiments. The results in Table 1 refer to the average of each individual parameter from multiple transfections.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
|
, there were no significant differences in the expression of YCG/R complexes and their corresponding wild-type receptors, signifying that any differences in the functional assays were not due to differences in receptor densities.
Specific binding results for each of the YCG/R complexes and their corresponding wild-type receptors are depicted in Fig. 3C. There was no apparent binding of [125I]hCG to cells expressing YCG/LHR, probably due to the presence of the fused ligand. Cells expressing TSHR and YCG/TSHR were able to bind [125I]bTSH at approximately the same level, whereas those expressing YCG/FSHR displayed reduced binding of [125I]hFSH compared with cells expressing FSHR.
Yoked hCG-receptor complexes: characterization of single chain ligand receptor complexes
hLHR and YCG/LHR.
Cells expressing YCG/LHR showed essentially no binding of 50 pM [125I]hCG due to the occupancy of binding sites by the attached hormone (Fig. 3C), as expected (13, 16). These cells, however, exhibited a high constitutive level of cAMP production that was not elevated upon addition of exogenous hormone (Fig. 4). Comparisons of the binding and signaling parameters of cells expressing hLHR and YCG/LHR cells are shown in Table 1.
|
), and no differences were found in the IC50 values (Table 1). An approximately 13-fold increase in basal cAMP in cells expressing YCG/TSHR compared with that seen in cells expressing hTSHR was found. Interestingly, these cells were further stimulated upon addition of exogenous bTSH, and the same maximal cAMP production was observed as that obtained in cells expressing hTSHR (Fig. 5B). Moreover, there was no difference between the ED50 values for cells expressing hTSHR and YCG/TSHR (Table 1).
|
). Again, there was no difference in the IC50 values for hFSHR and YCG/FSHR (Table 1). Cells expressing YCG/FSHR exhibited an approximately 4-fold increase in basal cAMP production compared with that of hFSHR; moreover, as with cells expressing YCG/TSHR, those cells were further stimulated after the addition of exogenous hFSH, reaching a maximum cAMP level comparable to that of cells expressing hFSHR (Fig. 6B). No differences in the ED50 values were observed (Table 1).
|
shows a representative dose-response curve of cAMP production in hLHR-expressing cells at various concentrations of recombinant hCG. Interestingly, high doses of hCG can also lead to minimal activation of hTSHR and hFSHR (Fig. 7B). Although the responses were too limited to obtain ED50 and maximal cAMP values, these results document that the observed increase in basal cAMP in YCG/TSHR and YCG/FSHR is attributable to hCG and not to a nonspecific effect from fusing hCG to the N terminus of each noncognate receptor. Limited competitive binding studies were also conducted with recombinant hCG and [125I]bTSH plus hTSHR and [125I]hFSH plus hFSHR (data not shown). At about 300 µg/ml hCG, i.e. the highest dose used in Fig. 7B, competition with 50 pM [125I]bTSH and 100 pM [125I]hFSH and the cognate receptors reduced specific binding of the cognate hormone-receptor pairs by 30–40%.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit, and there is a high degree of sequence homology in the glycoprotein hormone ß-subunits and the glycoprotein hormone receptor ectodomains, some degree of cross-reactivity is evidently allowed within this hormone receptor family. An understanding of the differences and similarities between the abilities of these hormones to bind to and activate cognate and noncognate receptors will provide important information not only about the sequence determinants responsible for specificity, but also about common binding and receptor activation elements, e.g. and conserved ß contact sites to identical receptor domains. Such results will help guide the development of small molecule agonists and antagonists for these GPCRs.
Figure 1B is arranged to depict the three glycoprotein hormone receptor ectodomains in the context of the model proposed by Bhowmick et al. (5) for that of rat LHR. In their homology model based on the leucine-rich repeat (LRR) structure of pancreatic ribonuclease (18), residues 27–235 of rat LHR were proposed to form nine LRRs as shown in Fig. 1B for the human receptors. The N- and C-terminal regions of the ectodomain do not exhibit an LRR structure, but it was suggested that a portion of the hinge region may adopt a chemokine-like fold (5). Identical amino acid residues in the ectodomains of the three glycoprotein hormone receptors are shown by open boxes, and conserved residues, e.g. hydrophobic, ionizable/identical charge (K/R and D/E), and hydroxylated (S/T), are indicated by shaded boxes.
We know from several studies that high affinity hCG binding to LHR occurs primarily in the region encoded by exons 1–8 (3, 19, 20). There is, of course, no assurance that the binding of TSH and FSH to their cognate receptor ectodomains will be comparable to that of hCG and the LHR ectodomain. Indeed, in a systematic comparison of the roles of the LRRs in the ectodomains of LHR and FSHR, Song et al. (21), focusing on LRR1–8, concluded that the Leu/Ile-x-Leu/Ile motifs of the ß-strands are important, but not equally. Using Ala-scanning mutagenesis of the Leu/Ile residues in the above motif, where there is only one difference between LHR and FSHR in the aligned sequences, they found that hCG binding was eliminated or greatly reduced with Ala replacements of either of the ß-strand Leu/Ile residues in all eight LRRs of FSHR, whereas LRR5 and -6 of LHR contributed less than the others. Thus, even identical amino acid residues in the three receptor ectodomains may have distinct functions in binding cognate ligands. The same group also showed that LRR4 of LHR contained some amino acid residues critical for hCG binding and others important in signaling (22). Whether such distinctions will hold for FSHR and TSHR remains to be determined.
A careful examination of Fig. 1B shows that the regions encoded by exons 1 (most of the N-terminal Cys-rich region), 10, and a portion of 11 (most of the C-terminal Cys-rich region) are the most divergent in structure. The three ectodomains contain a highly conserved region encoded by exon 9, L(T/S)YPSHCCAFXN, and an identical region encoded by exon 11, FNPCEDIMGY. Both of these domains have been shown to influence hCG-mediated LHR activation, but not ligand binding (23, 24, 25, 26). Within the region of the LHR ectodomain believed to be responsible for high affinity ligand binding, the N-terminal Cys-rich region and that of LRR1 and -7 are the most variable and LRR2–6 and -8 exhibit about 55 ± 7% amino acid sequence identity/homology in the three receptors. It is tempting to assign some of the binding specificity to the hormone-specific regions of the ß-subunits making contacts with the N-terminal Cys-rich region, LRR1 and -7, and portions of LRR2–6 and -8, whereas the -subunit and the identical regions of the ß-subunits are more likely to have contact sites in the identical/conserved regions of LRR1–8. Those are only suggestions, but they provide a useful framework for data interpretation and to guide further experiments.
A number of amino acid residues on the - and ß-subunits have been suggested as possible contact sites for cognate receptor binding (2). The seatbelt regions of the ß-subunits, identified in the crystal structures of hCG (7, 8) and hFSH (9), have been studied by several groups using chimeras of the ß-subunits (27, 28, 29, 30). The results suggest that the N-terminal portion of the seatbelt, i.e. the determinant loop (2), appears more important in hCG binding to LHR, whereas the C-terminal region of the seatbelt has more of an influence on FSH and TSH binding to their cognate receptors. It will be of great interest to determine the contact sites of the seatbelts, if any (29), with the specific regions of the receptors discussed above.
It is interesting that when yoked to hTSHR and hFSHR, hCG is capable of activating the receptors with minimal, if any, interference with binding of the cognate ligand. This observation may indicate that yoked hCG is capable of partially activating the endodomain or may bind to and activate a region of the ectodomain independent of the site for cognate ligand binding. It is also possible that the low affinity of hCG for hTSHR and hFSHR is amplified in the yoked system, where hCG cannot diffuse far from the receptor ectodomain due to the constraint of the covalent linkage. As the reduced affinity of hCG to the noncognate receptors, relative to the cognate hormones, is probably manifested to some degree by the time of occupancy of the hormone on the receptor-binding site, one would expect a higher rate constant of dissociation of hCG from hTSHR and hFSHR than those of the cognate ligands. In the 30-min cAMP assay, any level of productive binding of yoked hCG to the noncognate receptors could lead to an increase in basal cAMP production. On the other hand, in a 6-h binding assay, the greater affinity of cognate ligand to the yoked hCG-receptor complex, e.g. expected Kd differences on the order of 104–105, would favor occupancy by the natural ligand rather than YhCG. If YhCG inhibits some of the binding of cognate ligand, it would probably reduce the apparent binding capacity and not the Kd, unless of course YhCG alters the conformation of the receptor-binding region on the ectodomain. The inherent difficulties of quantifying and comparing results from cognate ligand binding and receptor antibody binding prohibit sufficient accuracy to ascertain whether the apparent binding capacity values are altered.
It is known that in cases of gestational hyperthyroidism hCG is able to bind to and activate TSHR (17), consistent with our YCG/TSHR findings. Comparative studies of the abilities of hCG and hLH to activate hTSHR expressed in CHO cells showed that both were functional at high concentrations, with hLH being more effective than hCG (31). The difference was attributable to the hCGß-CTP. In addition, Rodien et al. (32) showed that a mutation in TSHR, in which Lys183 is replaced with Arg, increases the receptor’s affinity for hCG significantly above that of wild-type hTSHR. Again, this finding supports our studies and exemplifies the importance of the slight differences in primary structure within these families to direct hormone binding and allow for receptor activation. To the best of our knowledge, there are no reported clinical cases in which hypersecretion of hCG activates FSHR, but our results suggest that this can occur.
Yoked hormones and yoked hormone-receptor systems have been used to study interactions within the families of glycoprotein hormones and the glycoprotein hormone receptors. We and others have created YhCG complexes in which the fusion protein has been found to be active in either orientation, N-hCGß--C (10, 11, 12) and N--hCGß-C (13, 14, 15). In addition to single chain TSH (33) and FSH (34) hormones, both of which are as active as the heterodimers, Boime’s group has created multifunctional fusion proteins (35, 36). In one such protein, the common -subunit was fused to FSHß and hCGß, creating a hybrid that could react with FSHR as well as LHR (35). These studies were then expanded to include a fusion protein that included all three glycoprotein hormone ß-subunits and a single -subunit to create a trifunctional protein. This protein was found to bind with high affinity to each of the glycoprotein hormone receptors and result in cAMP production (36).
Although we cannot discount the possibility that the presence of hCG in the YCG/R complexes forces a constraint on receptor conformation that promotes an increase in basal cAMP levels, we do not believe that this happens, based on the documentation of hCG-TSHR interactions (17) as well as the absence of receptor activation when the -subunit, as a monomer or homodimer, was fused to hTSHR (37). In addition, recent studies from our group showed that the human -subunit, hCGß, CTP, or PRL, an unrelated protein, when fused to LHR, produced no increase in basal cAMP production (38). Given these data, we believe that the increases in basal cAMP levels that we have observed are caused by specific interactions of hCG with each of the glycoprotein hormone receptors. Of interest is the observation that YCG/LHR yields a high basal cAMP that is not increased by the addition of exogenous hormone. This mimics the functional response of many activating mutations of LHR and may reflect the possibility that the conformation of the partially activated receptor is somewhat different from that of the cognate (nonyoked) ligand-receptor complex (39). There is, unfortunately, no information available on the effects of yoking ligand to receptors (cognate and noncognate) on desensitization and internalization. In summary, the findings from this study have indicated an ability of hCG to bind to and activate its noncognate receptors, hTSHR and hFSHR.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: b, Bovine; FSHR, FSH receptor; GPCR, G protein-coupled receptor; h, human; hCG, human chorionic gonadotropin; LHR, LH receptor; LRR, leucine-rich repeat; TSHR, TSH receptor; Y, yoked.
Received August 8, 2002.
Accepted for publication October 1, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and chorionic gonadotropin ß subunits in tandem. Proc Natl Acad Sci USA 92:2041–2045ß is decreased upon deletion of five carboxyl terminal amino acids of the subunit without affecting its receptor binding. J Mol Endocrinol 24:157–164[Abstract]
and ß subunits in a single chain variant of human chorionic gondotropin affects the heterodimeric interaction of the subunits and receptor-binding epitopes. J Biol Chem 276:29871–29879 subunit and the follicle-stimulating hormone ß subunit. Role of a linker sequence. J Biol Chem 271:10445–10448-subunit gene to the follicle-stimulating hormone and chorionic gonadotropin-ß subunit genes: production of a bifunctional protein. Mol Endocrinol 13:1873–1881-subunit to the TSH receptor results in partial antagonism of TSH activation. Program of the 83rd Annual Meeting of The Endocrine Society, Denver, CO, 2001, p 265
This article has been cited by other articles:
 |
|
 |
|
  A. De Leener, L. Montanelli, J. Van Durme, H. Chae, G. Smits, G. Vassart, and S. Costagliola Presence and Absence of Follicle-Stimulating Hormone Receptor Mutations Provide Some Insights into Spontaneous Ovarian Hyperstimulation Syndrome Physiopathology J. Clin. Endocrinol. Metab., February 1, 2006; 91(2): 555 - 562. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Delbaere, G. Smits, O. Olatunbosun, R. Pierson, G. Vassart, and S. Costagliola New insights into the pathophysiology of ovarian hyperstimulation syndrome. What makes the difference between spontaneous and iatrogenic syndrome? Hum. Reprod., March 1, 2004; 19(3): 486 - 489. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. F. Vischer, J. C. M. Granneman, and J. Bogerd Opposite Contribution of Two Ligand-Selective Determinants in the N-Terminal Hormone-Binding Exodomain of Human Gonadotropin Receptors Mol. Endocrinol., October 1, 2003; 17(10): 1972 - 1981. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Smits, O. Olatunbosun, A. Delbaere, R. Pierson, G. Vassart, and S. Costagliola Ovarian Hyperstimulation Syndrome Due to a Mutation in the Follicle-Stimulating Hormone Receptor N. Engl. J. Med., August 21, 2003; 349(8): 760 - 766. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, The seat belt region of the ß-subunits. B, The alignment of the receptor ectodomains is separated to show the N-terminal region (
30 amino acid residues), the 9 (imperfect) LRR predicted by Bhowmick et al. (
, conservative replacements, e.g. hydrophobic, identically charged residues, and T/S. The frequently observed (N/C) at position 9 (
