| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
RESEARCH COMMUNICATION |
a Institut für Pharmakologie, Universitätsklinikum Benjamin Franklin, Freie Universität Berlin,D-14195 Berlin, Germany
b Klinik und Poliklinik für Kinderheilkunde, Virchow-Klinikum, Humboldt-Universität zu Berlin
c Institut für Molekulare Pharmakologie, Berlin
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) did not show restoration of constitutive activity and dual signaling, thus suggesting a functional role of a properly spaced hydroxyl group at position 601. Molecular modeling revealed that the formation of a hydrogen bond between the hydroxyl group of Y601 in TM5 and the carbonyl oxygen of A623 in the peptide backbone of TM6 is critical for the receptor to adopt active conformations that impart wild-type signaling properties. Our findings indicate that multiple active receptor states underlie coupling of a G-protein-coupled receptor to different G-proteins.—Biebermann, H., Schöneberg, T., Schulz, A., Krause, G., Grüters, A., Schultz, G., Gudermann, T. A conserved tyrosine residue (Y601) in transmembrane domain 5 of the human thyrotropin receptor serves as a molecular switch to determine G-protein coupling. FASEB J. 12, 1461–1471 (1998)
Key Words: G-protein-coupled receptor • molecular modeling • receptor conformation • thyroid • TSH receptor allele
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Several GPCRs have been shown to couple to more than one G-protein—Gs and Gq/11, for instance (7). In many cases it is unclear whether contact sites for different G-proteins on cytoplasmic receptor parts can be differentiated or are identical. In the case of the human TSHR, the middle portion of the second cytoplasmic loop has been suggested to be important for Gs coupling whereas no clearly defined region could be held responsible for receptor/Gq/11 interaction (8). When transiently expressed in COS-7 cells, two mutant TSHRs found in toxic thyroid adenomas (F631L and I486F) constitutively stimulated cAMP accumulation to a comparable extent, but only the I486F mutant constitutively activated phospholipase C (9). That the latter mutation is located in the first extracellular receptor loop (i.e., in a domain known not to be involved in G-protein interaction) suggests that stimulation of the inositol phosphate cascade may involve a distinct receptor conformation that is stabilized by the I486F mutation and differs from the active receptor conformation elicited by F631L.
Initial cloning of the dog TSHR (10) was soon followed by cDNA sequence deposition of the human receptor by three independent groups (11–13). The first human TSHR clone (11), usually serving as the cDNA and protein sequence of reference in most databases, differs from the other two published sequences in that a conserved Y residue at position 601 is exchanged for H. The same sequence alteration has recently been found on one allele in a patient with congenital primary hypothyroidism and was classified as a heterozygous silent polymorphism (14).
Because a Y residue at a corresponding position near the carboxyl-terminal border of transmembrane domain 5 (TM5) is conserved among many GPCRs, we set out to study the functional role of Y601 in the human TSHR.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Fig. 1)
were created by using standard polymerase chain reaction (PCR) mutagenesis techniques (15), using the human TSHR expression plasmid TSHR-pcD-PS (16) as a template. To compare signaling characteristics of TSHR-H601 and TSHR-Y601, a Bsu36I/BstEII fragment excised from TSHR-pcD-PS (TSHR-H601) (16) was replaced by a corresponding fragment amplified from genomic DNA to generate TSHR-Y601, which we classified as wild-type (wt) TSHR. Primers were designed to generate PCR products carrying various mutations. After digestion of amplified DNA with Bsu36 I and BstE II, mutant PCR fragments were used to replace the corresponding part in the TSHR-pcD-PS. The identity of the various constructs and the correctness of all PCR-derived sequences were confirmed by restriction analysis and dideoxy sequencing (Amersham, Buckinghamshire, U.K.).
|
Cell culture and gene transfer
COS-7 cells were grown as described previously (16). For transient transfections, COS-7 cells were seeded into 12-well dishes (1.5 x 105 cells/well), and 0.5–2 µg of plasmid DNA per well was transfected using lipofectamine (Life Technologies, Paisley, U.K.). For binding studies, 2 x 106 cells/10 cm dish were seeded and transfected with 20 µg DNA, using a calcium phosphate coprecipitation method (17).
Functional assays
Assessment of inositol phosphate and cyclic AMP accumulation was performed as described previously (18, 19). For radioligand binding studies, cells were harvested 72 h after transfection and binding assays were performed using membrane homogenates (16).
Enzyme-linked immunosorbent assay (ELISA)
To estimate cell surface expression of receptors, COS-7 cells were seeded into 48-well dishes (5 x 105 cells/well) and transfected with various TSHR constructs and a ß-galactosidase-coding plasmid as a negative control, as described above. Three days later, cells were washed with phosphate-buffered saline (PBS) and fixed with 4% formaldehyde in PBS. After washing and blocking with cell culture medium containing 10% FBS, cells were incubated with 150 µl/well of tissue culture supernatant from the hybridoma clone 2C11 directed against the extracellular domain of the human TSHR (20), and the immunological detection was performed as described previously (16).
Immunofluorescence studies
Immunofluorescence studies were carried out to examine the subcellular distribution of wt and mutant TSHRs and were performed as described elsewhere (16).
PCR screening for naturally occurring Y601H mutants
To study the frequency of the Y601H TSHR mutation within the normal population (healthy parents of children with endocrine diseases and laboratory staff), patients with thyroid diseases (children with congenital hypothyroidism and adults with hyperthyroidism) and children with other endocrine diseases (e.g., congenital adrenal hyperplasia, pubertas praecox, central diabetes insipidus, etc.), a simple PCR-based screening method was established. Genomic DNA was prepared from peripheral white blood cells using a DNA extraction kit (QIAamp Blot Kit). Subsequently, a 327 bp fragment of the TSHR was amplified by PCR using an antisense primer (5' GCC AAA CTT GCT GAG TAG GA 3') and a degenerated forward primer (5' CTC AAC ATA GTT GCC TTC GTC ATC GTC TGC TGC TGA CGT 3', degenerated bases are underlined); 0.5 µg of genomic DNA served as template. The degenerated primer was designed in a way to introduce an additional restriction site (AatII) into the PCR fragment if the Y601H mutation was present in the template DNA. PCR products were digested with AatII and separated on 2% agarose gels. DNA samples coding for the wt TSHR and mutant Y601H as well as a negative control were analyzed in all assays to guarantee correct assay performance.
Molecular modeling
The procedure used to construct a 3-dimensional (3D) structural model of the transmembrane regions and the connecting loops of the human TSHR was carried out in several steps, which included: sequence alignment of homologous glycoprotein hormone receptors and of 40 additional peptide hormone receptors; identification of TM boundaries by variability, hydrophobic profile, and secondary structure predictions; arrangement of the seven helix bundle based on a low resolution structure of rhodopsin (21); orientation (twist) of each TM helix on its own axis based on amphipathic profiles; 3D model screening of helix–helix interactions (22); and compilation of locations reported to affect receptor activation among glycoprotein hormone and other GPCRs and of mutations affecting ligand binding among homologous peptidergic GPCRs (23). With the exception of TM5, the resulting helix orientations were similar to those recently published by Baldwin et al. (24).
A first TM model with optimized helix packing subsequent to a side chain rotamere screening (Sybyl [Tripos] procedure) using the Amber force field (25) was extended by adding on receptor loops. A homology modeling approach was applied using the Biopolymer module of Sybyl (Tripos) and searches for suitable conformations of corresponding loop sequences within the Brookhaven 3D protein database.
The vast majority of the amino-terminal extracellular receptor domain is not included in the model. However, in order to have a geometrical counterpart for the neighboring extracellular loops 1 and 3, five amino acid residues preceding TM1 were added to the model. A portion of the intracellular carboxyl-terminal tail comprising the assumed intracellular loop 4 (i4) confined by the putative palmitoylation site at C699 was modeled by homology based on the nuclear magnetic resonance structure of cytosolic rhodopsin loop peptides (26). The stability of our receptor model was studied by means of molecular dynamics simulations. Amber force field conditions in vacuo were calibrated, keeping helix stability without restraints for such runs and using the experimental 3D structures of bacteriorhodopsin as a test case. The changes within the TM domains of our TSHR model occurring during a molecular dynamics simulation are directly related to the kinking of P-containing TM helices. Main differences from other receptors are manifested within TM5 because, in contrast to most other GPCRs, glycoprotein hormone receptors do not contain a P residue within TM5.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Fig. 1). TSHR variant Y601H was initially isolated from a cDNA library prepared from thyroid tissue of a patient with Graves' disease (11) and later classified as a polymorphic variant in one patient suffering from congenital hypothyroidism (14). To assess the prevalence of TSHR-Y601H in a European population, we analyzed genomic DNA of 263 subjects for the presence of the wt TSHR or TSHR-Y601H allele ( Table 1).
None of the patients with thyroid disease nor any other subject examined carried the mutant allele, and the exclusive presence of the wt receptor, TSHR-Y601, could be confirmed in each case. Therefore, we conclude that TSH-Y601H is a rarely occurring allelic variant, and only TSHR-Y601 should be regarded as a bona fide wt TSHR.
|
Functional assessment of conserved tyrosine residues at amino acid positions 601 and 605 of the human TSHR
To assess the functional importance of these conserved Y residues for TSH-induced signal transduction, we used a site-directed mutagenesis approach to introduce various amino acid substitutions and deletions at positions 601 and 605 as well as at the neighboring position 623 in TM6 ( Fig. 1). Mutant TSHRs were constructed by PCR-based in vitro mutagenesis using the wt TSHR as a template. All receptor constructs were transiently transfected into COS-7 cells, and TSH-induced cAMP and inositol phosphate (IP) formation was determined ( Fig. 2,
Table 2).
Stimulation of cells expressing the wt TSHR as well as mutants Y601H and Y605H with 100 mU/ml bovine TSH (bTSH) resulted in an increase in cAMP accumulation ( Fig. 2A). Basal and hormone-stimulated intracellular cAMP levels were slightly suppressed in cells expressing TSHR-Y601H, whereas the agonist-stimulated TSHR-Y605H behaved like the wt TSHR irrespective of reduced basal cAMP production (see Fig. 2A). A restriction of the transfected amount of plasmid DNA coding for the wt receptor in order to reduce receptor density resulted in wt TSHR expression that was similar to TSHR-Y601H in terms of cAMP formation (see Fig. 2A). In the case of the wt TSHR, determination of agonist-induced IP accumulation in transfected cells showed a pronounced response that was reduced when the amount of plasmid DNA coding for the receptor was restricted to one-third of the initial value ( Fig. 2B), demonstrating that a reduced expression of the wt TSHR as such did not abolish agonist-dependent IP accumulation. TSH-elicited IP formation was not observed in cells expressing TSHR mutants Y601H and Y605H. Because the determination of TSHR protein expression in membranes either by radioligand binding or cell surface ELISA assays ( Table 2) showed that TSHR-Y601H was expressed at densities comparable to the wt receptor, differences in membrane expression could not account for the different signaling potential of the wt TSHR vs. TSHR-Y601H.
|
|
The wt TSHR is characterized by considerable constitutive activity toward the Gs/adenylyl cyclase system whereas other glycoprotein hormone receptors, such as the LH receptor, are not (30). When expressed at comparable receptor densities, basal cAMP levels of cells expressing the wt TSHR are considerably higher than those in cells expressing TSHR-Y601H (see Fig. 2A, Table 2). Basal cAMP levels were systematically examined after transfecting cells with increasing amounts of plasmid DNA coding for the wt TSHR or the Y601H mutant ( Fig. 3).
Whereas the wt TSHR elicited a twofold increase in basal cAMP levels at higher DNA concentrations, no such effect was observed with the Y601H construct (see Fig. 3).
|
To better understand the functional role of conserved Y residues in the TM5/i3 region of the human TSHR, Y601 of the wt receptor was substituted by eight different amino acids or deleted, and the resulting TSHR constructs were characterized after expression in COS-7 cells. Receptor protein expression in cell membranes was assessed by displacement of 125I-bTSH by unlabeled bTSH (see Table 2) and, independently, by a cell surface ELISA procedure (see Table 2). Protein expression could not be demonstrated for mutants Y601P and Y601, although COS-7 cells transfected with these receptor constructs responded to TSH challenge with small yet reproducible increases in cAMP formation (see Table 2). The low responsiveness, however, did not allow for an accurate determination of ED50 values (see Table 2). The introduction of a helix-breaking P or the deletion of amino acid position 601 led to intracellular trapping of the mutant proteins, as indicated by the absence of membranous yet strong intracellular immunofluorescence of transiently transfected COS-7 cells (data not shown). Mutants Y601D and Y601K were characterized by a significantly reduced membrane insertion as determined by 125I-bTSH binding and cell surface ELISA. Both receptor mutants responded to TSH challenge with clearly measurable increases in cAMP formation (see Table 2), but the low level of expression precluded the determination of Ki values and of a meaningful ED50 value in the case of TSHR-Y601D. With regard to the more distally located Y605, mutant Y605A displayed receptor densities similar to the wt TSHR whereas membrane expression of TSHR-Y605H and TSHR-Y605 was significantly diminished (see Table 2). None of the eight remaining amino acid substitutions or deletions analyzed at positions 601 and 605 had a significant effect on the affinity toward the agonist TSH, as reflected by the respective Ki values (see Table 2).
All 13 TSHR constructs tested responded to TSH stimulation with a measurable increase in intracellular cAMP levels, and agonist concentrations yielding half-maximal cAMP formation (EC50 values) could be determined in 10 instances (see Table 2). The EC50 values obtained were comparable to the one characterizing the wt TSHR and, in general, did not differ by more than a factor of 2.3 if data obtained with the poorly expressed mutant Y601K were excluded (see Table 2). Exchange of the phenyl group at position 601 for an indolyl moiety (TSHR-Y601W), however, led to a marked rightward shift of the concentration–response curve toward higher agonist concentrations while hardly affecting cell surface expression (see Table 2).
The activated human TSHR not only couples to Gs, but also leads to pertussis toxin-insensitive (data not shown) phospholipase C stimulation (see Table 2). The EC50 value for the latter signal transduction pathway was shifted toward higher TSH concentrations by a factor of 3.4 when compared with the potency of TSH to elicit cAMP accumulation (see Table 2). Most significantly, none of the mutant TSHRs was able to functionally couple to the Gq/11/phospholipase C-ß system (see Table 2). Thus, our mutagenesis data highlight an essential role of Y residues at positions 601 and 605 to maintain the dual signaling capability of the human TSHR to adenylyl cyclase and phospholipase C.
Molecular modeling of the human thyrotropin receptor
In an attempt to decipher the molecular basis of the functional data obtained, 3D structural models of the transmembrane regions and the connecting loops of the human TSHR were constructed. Helix packing based on recent electron density maps of frog rhodopsin obtained for six different slices through the transmembrane region (21) led to an arrangement of helices in which TMs 3, 6, and 5 are very tightly packed along their intracellular halves ( Fig. 4A).
Our TSHR model stresses the following hydrophobic side chain interactions to contribute to the tight packing of TM5 and TM6: V597-I630, I604-A623 ( Fig. 4B, C). M626 in TM6, the TSHR equivalent to M571 of the LH receptor, was found not to be involved in the hydrophobic TM5/TM6 interaction, but rather appeared to be flexible and oriented in between TM3 and TM5.
|
The differential effect of Y601 vs. F601 on constitutive cAMP production and coupling to phospholipase C clearly indicates the participation of the Y601 hydroxyl group in the activation process. Rotamer analysis and torsion angle scanning of Y601 revealed that no other hydrophilic side chain interaction partners occur at a suitable distance. However, the hydroxyl group is principally able to reach a peptide backbone carboxyl group at position A623 in the TM6/i3 transition region to form a hydrogen bond (see Fig. 4B). In the inactive state, the side chain of A627 in TM6 disfavors an optimal hydrogen bond formation of Y601 toward the TM6 backbone. Therefore, Y601 gives rise to a van der Waals repulsion and simultaneously to an electrostatic attraction between TM5 and TM6, thus generating tension that presses the two helices apart. According to molecular dynamics trajectories, it appears most likely that an interaction of Y601 and Y605 via hydrogen bonding stabilizes the orientation of Y601 within TM6. Recent evidence suggests that agonist-dependent activation of GPCRs is accompanied by a movement of TM6 relative to other TMs (31–35). A clockwise (as viewed from the cell interior) rotational rigid body movement of TM6 has been noted during rhodopsin activation (31). Postulating a similar movement during TSHR activation, the formation of a hydrogen bond between Y601 and the backbone carbonyl oxygen atom of A623 in TM6 is immediately achieved ( Fig. 4D), thus stabilizing an active receptor conformation.
Experimental testing of the TSHR model
To further assess the functional relevance of our modeling approach, TSHR constructs were functionally characterized which contained additional amino acid substitutions at one or two positions. Substituting A at position 623 with a bulky Y residue resulted in a TSHR with considerable constitutive activity, reflected by a 2.5-fold elevation of basal intracellular cAMP levels ( Table 3).
Our TSHR model may offer an explanation for these data by predicting an unfavorable steric interference of the Y side chain of TSHR-A623Y (TM6) with I604 in TM5 (see Fig. 4B, C). Surprisingly, the double mutant Y601A/A623Y did not result in increased basal cAMP production and totally lost coupling to phospholipase C, although cell surface expression of this receptor was comparable to that of mutant A623Y (see Table 3). In summary, the experimental data are consistent with the relative orientation of helices in the model, in particular at the TM5/TM6 interface, and corroborate the earlier conclusion that a Y residue exactly at position 601 is indispensable for wt signal transduction properties of the TSHR.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Despite the conspicuous conservation of a Y residue within TM5, its functional importance has only been assessed for the AT1a angiotensin receptor (41). Y215 in the AT1a angiotensin II receptor has a critical role in receptor activation, because replacement of Y215 with F resulted in a receptor that was inserted into the plasma membrane and bound ligand with unaltered affinity, yet was devoid of signaling capacity (41). Therefore we set out to study the role of Y601 for signal transduction via the human TSHR.
A comparison of the signaling properties of TSHR-Y601H with the wt TSHR-Y601 (12, 13) revealed two important differences: 1) TSHR-Y601H was no longer able to activate the phospholipase C/IP signaling cascade whereas coupling to adenylyl cyclase was only modestly affected; and 2) TSHR-Y601H lost its endogenous constitutive activity toward Gs-dependent adenylyl cyclase activation, a characteristic feature of the wt TSHR-Y601 discriminating the latter from other glycoprotein hormone receptors (30). Phospholipase C activation by primarily Gs-coupled receptors requires a higher receptor density and higher agonist concentrations as opposed to stimulation of adenylyl cyclase by the same agonist (7). In accord with early data obtained with human thyroid slices and the recombinant wt TSHR expressed in CHO cells (42), we observed a three- to fourfold shift toward higher agonist concentrations of the concentration–response curve describing TSH-induced IP production. The loss of dual signaling ability of TSHR-Y601H cannot be accounted for by a reduced receptor density because cell surface expression of TSHR-Y601H amounted to more than 80% of wt receptor levels as determined by our ELISA approach. In summary, replacement of one highly conserved amino acid residue in TM5 selectively disables one signal transduction pathway of a receptor with originally dual signaling potential.
To address the issue of whether the functional consequences observed specifically depended on an H residue at position 601, seven additional amino acid replacements for Y601 and one deletion mutant were functionally analyzed. Mutant receptors displaying appreciable membrane expression showed 125I-TSH binding characteristics similar to the wt TSHR. This finding conforms to previous observations indicating that specific high-affinity ligand binding of glycoprotein hormone receptors is imparted by the large extracellular domain (2, 5) and that ligand binding and ligand-dependent receptor activation are clearly dissociable events in this receptor subfamily (43, 44). Introduction of a helix-breaking P or deletion of codon 601 interfered with proper receptor folding and precluded the expression of a significant number functionally competent receptors at the cell surface. Similar effects were observed with charged acidic (i. e., D) or basic (i. e., K) amino acids at codon 601. The amino acid A is generally considered to be the least disruptive mutation that can be made in the absence of specific knowledge about side chain interactions, and the high helical propensity makes it particularly favorable for substitution within 
-helices like TMs of GPCRs. Cell surface expression of TSHR-Y601A was therefore only marginally affected, yet this mutant functionally behaved like TSHR-Y601H and could not restore wt signaling. To further examine whether an aromatic side chain as such would account for wt receptor signaling properties, Y601 was replaced by W and F. Although the latter amino acid differs from Y only by the absence of a hydroxyl group, none of these mutants resembled the wt TSHR in functional terms. Considering that the hydroxyl group of Y601 may be involved in hydrogen bonding with vicinal amino acids, S (which is also capable of participating in a hydrogen bond) was inserted, but likewise failed to restore dual signaling capability and endogenous constitutive activity of the wt receptor. Therefore, our mutagenesis data emphasized the functional role of a properly spaced hydroxyl group as part of a hydrogen bond network involving TM5.
Structure–function studies of muscarinic acetylcholine receptors clarified the role of specific noncharged residues at the amino terminus of the i3 loop of the m3 muscarinic (28, 29) receptors. Extensive mutagenesis work revealed that a tyrosine residue (Y254 in the m3 receptor) conserved among the Gq/11-coupled m5 receptors is important for productive receptor/Gq/11 coupling. As Y605 in the human TSHR corresponds to Y254 in the m3 receptor, we addressed the importance of this amino acid residue for TSH-dependent IP production. Deletion of Y605 resulted in a TSHR that still responded to agonist challenge in terms of cAMP production, but was only poorly expressed at the cell surface. Replacement of Y605 with H or A residues generated TSHRs efficiently coupled to the Gs/adenylyl cyclase system, yet devoid of endogenous constitutive activity. None of the mutants restored TSH-dependent phosphoinositide breakdown, thus corroborating and extending structure–function insights initially worked out for the muscarinic receptor family to the TSHR and potentially to other glycoprotein hormone receptors. As Y605 is located four amino acids downstream of Y601, a distance that corresponds approximately to one turn in an -helical structure, it is conceivable that both Y601 and Y605 could function in concert in order to propagate agonist-induced conformational changes.
To gain further insight into the molecular mechanisms underlying our functional data, a 3D model of the human TSHR was constructed. The model is in good overall agreement with the recently published model of the human LH receptor (45). Both models predict a close packing at the intracellular aspects of TMs 3, 5, and 6. In particular, a P-induced kinking of TM6 leads to a tightly packed hydrophobic cluster with TM5 at the intracellular halves of these two helices. A hydrophobic `knob and hole' cluster between TM5 and TM6 in conjunction with an H bonding network between the central portions of TM6 and TM7 has been proposed to be essential to preserve the inactive form of the receptor (45). Recent evidence shows that activation of GPCRs is accompanied by a rigid body movement of TM6 relative to other TMs (31–35). The inability of disulfide cross-linked (31) and Zn2+ chelating (34) rhodopsin mutants to activate transducin underpins the functional importance of these helix movements. An essential role of TM6 in receptor activation and constitutive activity is indicated by activating mutations in this region shown for many GPCRs like the ß2-adrenergic receptor (46, 47). All these mutations may disrupt constraining interhelical interactions leading to agonist-independent movements of TM6. Of course, this assumption does not exclude the possibility that other domains also move in response to agonist binding. We propose that in the human TSHR, Y601 is one crucial element in the destabilization of helix packing, but certainly not the only one, as other glycoprotein hormone receptors that also carry a tyrosine residue at positions corresponding to Y601 in the TSHR do not display a similar level of spontaneous basal activity in terms of cAMP formation (30).
Our model further illustrates that the tight hydrophobic packing between A623 at the TM6/i3 transition and I604 within TM5 does not tolerate any other amino acid residues at position 623. Naturally occurring activating substitutions at this position (I623, V623, or S623) (48) are sterically unfavorable and would create tension between TM5 and TM6. This observation is consistent with in vitro mutagenesis results at the corresponding position of 1b-adrenergic receptor, in which A293 similarly contributes to a hydrophobic pair (32, 49). All mutants except the native A side chain were constitutively active.
Molecular modeling and the mutagenesis data presented here may explain our observation that a mutation of a conserved amino acid residue in TM5 selectively disables one signaling pathway of a receptor with dual signaling potential. In the allosteric ternary complex model describing activation of GPCRs (46, 50), an explicit isomerization step is introduced that regulates the transition from an inactive state R of the receptor to an active state R*. Only R* effectively interacts with G-proteins. Agonist binding shifts the equilibrium from R to R* and stabilizes the ternary complex consisting of agonist (A) -bound R* and G-protein (AR*G). The data presented here are in accord with the hypothesis of several distinct active states of a receptor as postulated by us before (7). Similar to light-activated rhodopsin, the agonist-bound TSHR may traffic between several active conformations that show distinct G-protein coupling predilections (7) due to alternating exposure of distinct intracellular domains relevant for productive G-protein interaction. According to our current concept ( Fig. 5),
the vast majority of wt TSHRs are assumed to reside in a preactivated state R', and agonist binding shifts the isomerization equilibrium either toward the preferred Gs coupling state R1* or to the Gq/11-interacting active conformation R2* (see Fig. 5), the latter requiring high receptor density and higher agonist concentrations as reflected in right-shifted concentration–response curves. In this scenario, Y601 can be regarded as a crucial molecular switch allowing the unliganded, inactive receptor R to preferentially adopt a preactivated conformation R' (see Fig. 5). The Y601H mutation disables the R/R' transition and fixes the receptor in the inactive ground state R (see Fig. 5). Whereas agonist binding is able to stabilize R1*, and thus allows Gs activation through the Y601H mutant, a direct R/R2* isomerization is not promoted by agonist resulting in selective abrogation of TSH-mediated phospholipase C activation.
|
This view is supported by the documentation of activating TSHR mutations in the transmembrane core and extracellular loops that differ in their constitutive activity toward the Gs/adenylyl cyclase system, as opposed to the Gq/11/phospholipase C-ß pathway (9), and by the functional role of K583 in exoloop 3 of the LH receptor. Any amino acid exchange at this position was found to be nonpermissible for agonist-induced cAMP production whereas IP signaling was unaffected (51).
If different receptor conformations are assumed that preferentially couple to distinct G-proteins, one may extend the model mentioned above and hypothesize that different agonists may stabilize distinct ternary complexes with different efficiencies (7). Recently, differentially glycosylated FSH isoforms have been shown to affect the coupling efficacy of the FSH receptor to Gs vs. Gi proteins (52). Thus, it may be possible to systematically develop pathway-selective glycoprotein hormone analogs as already indicated for recombinant human TSH (53). Apart from extending our knowledge on the specificity of receptor/G-protein interaction, such proteins may represent prototypes of new therapeutic agents.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Abbreviations: AR*G, agonist-bound R* and G-protein; 3D, 3-dimensional; ELISA, enzyme-linked immunosorbent assay; FSH, follicle-stimulating hormone; GPCRs, G-protein-coupled receptors (GPCRs); IP, inositol phosphate; LH/CG, lutropin/choriogonadotropin; PBS, phosphate-buffered saline; PCR, polymerase chain reaction; TSH, thyroid-stimulating hormone; TSHR, TSH or thyrotropin receptor; bTSH, bovine TSH; wt, wild-type; TM5, transmembrane domain 5; intracellular loop 3, i3.
Received for publication April 6, 1998. Accepted for publication May 27, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-helices. Nature (London) 389, 203–206[Medline]
-carbon template for the transmembrane helices in the rhodopsin family of G-protein-coupled receptors. J. Mol. Biol. 272, 144–164[Medline]
1B-adrenergic receptor: role of highly conserved polar amino acids in receptor activation. EMBO J. 15, 3566–3578[Medline]
1b-adrenergic receptor: potential role of protonation and hydrophobicity af a highly conserved aspartate. Proc. Natl. Acad. Sci. USA 94, 808–8131B-adrenergic receptor enhances mitogenesis and tumorigenicity. Proc. Natl. Acad. Sci. USA 88, 11354–11358 protein gene in 31 toxic thyroid nodules. J. Clin. Endocrinol. Metab. 82, 3885–38911B-adrenergic receptor by all amino acid substitution at a single site. Evidence for a region which constrains receptor activation. J. Biol. Chem. 267, 1430–1433This article has been cited by other articles:
|
|
 |
|
  T. R. H. Buch, H. Biebermann, H. Kalwa, O. Pinkenburg, D. Hager, H. Barth, K. Aktories, A. Breit, and T. Gudermann G13-dependent Activation of MAPK by Thyrotropin J. Biol. Chem., July 18, 2008; 283(29): 20330 - 20341. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-R. Chen, S. M. McLachlan, and B. Rapoport Suppression of Thyrotropin Receptor Constitutive Activity by a Monoclonal Antibody with Inverse Agonist Activity Endocrinology, May 1, 2007; 148(5): 2375 - 2382. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Nurwakagari, A. Breit, C. Hess, H. Salman-Livny, D. Ben-Menahem, and T. Gudermann A conformational contribution of the luteinizing hormone-receptor ectodomain to receptor activation J. Mol. Endocrinol., February 1, 2007; 38(2): 259 - 275. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Kleinau, M. Claus, H. Jaeschke, S. Mueller, S. Neumann, R. Paschke, and G. Krause Contacts between Extracellular Loop Two and Transmembrane Helix Six Determine Basal Activity of the Thyroid-stimulating Hormone Receptor J. Biol. Chem., January 5, 2007; 282(1): 518 - 525. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Karges, G. Krause, J. Homoki, K.-M. Debatin, N. de Roux, and W. Karges TSH receptor mutation V509A causes familial hyperthyroidism by release of interhelical constraints between transmembrane helices TMH3 and TMH5 J. Endocrinol., August 1, 2005; 186(2): 377 - 385. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Krohn, D. Fuhrer, Y. Bayer, M. Eszlinger, V. Brauer, S. Neumann, and R. Paschke Molecular Pathogenesis of Euthyroid and Toxic Multinodular Goiter Endocr. Rev., June 1, 2005; 26(4): 504 - 524. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Neumann, G. Krause, M. Claus, and R. Paschke Structural Determinants for G Protein Activation and Selectivity in the Second Intracellular Loop of the Thyrotropin Receptor Endocrinology, January 1, 2005; 146(1): 477 - 485. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Schulz and T. Schoneberg The Structural Evolution of a P2Y-like G-protein-coupled Receptor J. Biol. Chem., September 12, 2003; 278(37): 35531 - 35541. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-R. Chen, G. D. Chazenbalk, S. M. McLachlan, and B. Rapoport Evidence that the C Terminus of the A Subunit Suppresses Thyrotropin Receptor Constitutive Activity Endocrinology, September 1, 2003; 144(9): 3821 - 3827. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Fuhrer, M. D. Lewis, F. Alkhafaji, K. Starkey, R. Paschke, D. Wynford-Thomas, M. Eggo, and M. Ludgate Biological Activity of Activating Thyroid-Stimulating Hormone Receptor Mutants Depends on the Cellular Context Endocrinology, September 1, 2003; 144(9): 4018 - 4030. [Abstract] [Full Text] [PDF]
|
