Structure Function Relationships of the Glycoprotein Hormone Receptors

The particularly wide spectrum of activating mutations identified in the TSHr correlates with the observation that wild-type TSHr displays readily detectable basal activity [10,11], whereas gonadotropin receptors are virtually silent in the absence of their ligands [12]. Another peculiarity of the TSHr is the spontaneous cleavage of a proportion of the molecules present at the cell surface into two subunits that remain linked by disulfide bridges [6].

The GPHrs show clear structural dichotomy between the ligand-recognizing amino-terminal ectodomain and the serpentine rhodopsin-like portion that transmits the signal to the G protein. How does binding of the hormone to the ectodomain result in the activation of the serpentine domain? We will summarize the three key steps in this process: recognition and binding of the hormones, activation of the serpentine portion of the receptors, and intramolecular transduction of the activation signal between the ectodomain and the serpentine portion.

Structure and Function of the Ectodomain of Glycoprotein Hormone Receptors

The ectodomain of all three receptors is made of nine leucine-rich repeats (LRRs), each « 25 amino acids, flanked by two-cysteine containing domains (Fig. 1A,B). LRR-containing proteins constitute a large family of both intra-and extracellular molecules specialized in protein-protein

extracellular domain il El 12 E2 13 E3

Figure 1 Schematic representation of the TSH receptor (A), ribbon representation of the model of the leucine rich repeats region (B), and illustration of the position of a series of activating mutations (C). (A) The seven transmembrane helices are drawn as helical nets, respecting the helice ends as observed in the crystal structure of rhodopsin [51]. Closed circles in the N-terminal extension represent the portion of the domain modeled by Kajava et al. [14], comprising residues 54 to 254. Some of the key residues discussed in the text are indicated as black circles with an indication of the amino acid and the position (numbering system of Ballesteros et al. [50] (B) The a-helices are drawn as solid tubes. The "horseshoe" curvature is clearly visible, and the position of the residue mutated in pregnancy hyperthyroidism (Lys 183) is indicated. (C) The TSH receptor is represented linearly, with the transmembrane helices indicated in Roman numerals. The positions of a series of activating mutations are indicated with the nature of the amino acid substitutions. The numbering is TSHr specific (e.g., D633 corresponds to D6.44). (Adapted from Smits, G. et al., Mol. Endocrinol., 16, 722-735, 2002; Parma, J. et al., J. Clin. Endocrinol. Metab., 82, 2695-2701, 1997. With permission.)

extracellular domain il El 12 E2 13 E3

Figure 1 Schematic representation of the TSH receptor (A), ribbon representation of the model of the leucine rich repeats region (B), and illustration of the position of a series of activating mutations (C). (A) The seven transmembrane helices are drawn as helical nets, respecting the helice ends as observed in the crystal structure of rhodopsin [51]. Closed circles in the N-terminal extension represent the portion of the domain modeled by Kajava et al. [14], comprising residues 54 to 254. Some of the key residues discussed in the text are indicated as black circles with an indication of the amino acid and the position (numbering system of Ballesteros et al. [50] (B) The a-helices are drawn as solid tubes. The "horseshoe" curvature is clearly visible, and the position of the residue mutated in pregnancy hyperthyroidism (Lys 183) is indicated. (C) The TSH receptor is represented linearly, with the transmembrane helices indicated in Roman numerals. The positions of a series of activating mutations are indicated with the nature of the amino acid substitutions. The numbering is TSHr specific (e.g., D633 corresponds to D6.44). (Adapted from Smits, G. et al., Mol. Endocrinol., 16, 722-735, 2002; Parma, J. et al., J. Clin. Endocrinol. Metab., 82, 2695-2701, 1997. With permission.)

interactions [13]. Structural models for the ectodomains of the TSHr and LH/CGr were elaborated on the basis of the three-dimensional structure of an LRR protein, ribonuclease inhibitor [14,15]. The models predict that the LRR portion of the ectodomains of the receptors would adopt a horseshoe (or segment of doughnut) shape, with alpha helices and beta sheets making the convex and concave surfaces of the structure, respectively (Fig. 1B). For the ribonuclease inhibitor, direct crystallographic evidence indicated that the concave surface was responsible for the majority of the binding interactions with the ligand (ribonuclease) [16]. The pertinence of this model has been tested for the LH/CGr, essentially by means of loss-of-function mutations [15,17,18]. In the case of the TSHr, a gain-of-function mutation was identified in a family presenting with pregnancy-dependent hyperthy-roidism due to a K183R amino acid substitution, located in the middle of the LRR portion of the ectodomain of the receptor (Fig. 1B), predicted to face inside the putative hormone binding domain [19]. Functional studies in transfected C cells show that the K183R mutant becomes abnormally sensitive to the pregnancy hormone hCG [19]. The gain of function, though modest, is enough to cause disease because of the extremely high concentration of hCG achieved during the first trimester of pregnancy. Extensive site-directed muta-genesis based on the putative structural model suggested that the gain of function was due to the unmasking of the negative charge of glutamic acid in position 157 from a salt bridge with lysine 183, not achieved with the arginine replacement

[20]. Any amino acid substitution in position 183 causes a gain of function similar to that of K183R. Definitive validation of the model of the TSHr based on the structure of the ribonuclease inhibitor has been obtained very recently. Conversion of eight carefully selected residues of the putative binding surface of the TSHr to their LH/CGr homologs yields a TSHr mutant displaying a sensitivity to hCG comparable to that of wild-type LH/CGr (G. Smits et al., in preparation).

A posttranslational modification with important functional significance has recently been identified in all three GPHrs. Close to the border between the ectodomain and the first transmembrane segment of the serpentine, the three receptors harbor a motif that undergoes tyrosine sulfation just before insertion of the molecule into the plasma membrane

[21]. The sulfated tyrosines are an important component of the binding surface, as mutant receptors unable to become sulfated lose sensitivity to their hormones by one order of magnitude [21]. This identifies the sulfated tyrosines as an important participant in the known ionic interactions between GPH and their receptors [22].

Activation of the Serpentine Portion

The GPHr and, in particular, the TSHr can be activated by a wide spectrum of amino acid substitutions or deletions affecting mainly but not exclusively the serpentine domain (Fig. 1C) [9]. Some of these are homologous to activating mutations identified initially in adrenergic receptors [23,24].

Others involve residues specific to the GPHr subfamily. Despite their high sequence similarity, the three receptors display great differences in the propensity to be activated by mutations, with the TSHr being more prone to activation than the LH/CGr and the FSHr being particularly refractory [25]. The structural bases for these differences are still unknown. Among the spontaneous gain of function mutations, those affecting residue D6.44, in the sixth transmembrane segment (numbering system of Ballesteros et al. [26]) deserve special attention. This residue is part of one of the sequence signatures specific to the GPHr in transmembrane VI [27]. D6.44 (D633 or D578 in the TSHr- or LH/CGr-specific numbering systems, respectively) is one of the residues most frequently mutated in precocious puberty of the male and toxic thyroid adenomas.

Experiments performed with the TSHr were driven by the observation that in LGR1 (a glycoprotein hormone receptor homolog of Drosophila) [28] Asp and Asn residues were naturally exchanged between 6.44 and 7.49, suggesting that these residues of transmembrane segments VI and VII interact with each other. Functional studies of single and double mutants transfected in COS cells led to the following model: In the inactive state of the receptor, D6.44 and N7.49 interact; release of the side chain of N7.49 from this interaction, caused by mutation of D6.44 (e.g., D6.44A), would make it available for interactions involved in stabilization of an active state of the receptor [29]. This conclusion is also drawn from the observation that the N7.49A mutant loses the ability to be stimulated by TSH. Addition of the N7.49A mutation to con-stitutively active TSHr mutants dramatically reduces their activity, to the level of the wt receptor or below [30]. These results are in agreement with others that point to N7.49, one of the most conserved residues in rhodopsin-like GPCRs, as a key residue involved in stabilizing both the inactive and the active conformations (see discussion in Meng and Bourne [31] and Lu et al. [32]). The partner(s) of N7.49 in the active conformation is (are) still subject to intense investigation; in several other GPCRs, experimental evidence points to D2.50 [33-36]. It is likely that a complex network of interactions implicating N7.49 and D2.50, but also other residues (e.g., N1.50), stabilizes the active conformation [31,32].

Intramolecular Signal Transduction Between the Ectodomain and the Serpentine Domain

The observation that ectodomains of the GPHr can bind their agonists with high affinity in the absence of the serpentine domain [1,37,38] is compatible with two models for the activation of the receptors. According to the first, high-affinity binding of the agonist would position the hormone for a low-affinity interaction with the extracellular loops (and/or crevice) of the serpentine, leading to activation. A candidate for this activating interaction is the alpha subunit common to the three hormones. Experimental support for this model has been provided by site-directed mutagen-esis experiments introducing reciprocal mutations in the LH/CGr and hCG and by affinity labeling [39,40].

The above model, however, does not account for the capacity of the three receptors to be fully activated by point mutations in their ectodomain. A serine in position 281 of the TSHr was found mutated to threonine, asparagine, or isoleucine in autonomous thyroid adenomas [41-43]. Subsequently, it was found that mutations introduced at homologous positions in the LH/CGr (S277) and FSHr (S273) were similarly active [44]. This led to the notion that the ectodomain normally exerts a silencing effect on the serpentine domain and that activation of the GPHr results from the release of this inhibitory interaction. Direct evidence for this silencing role of the ectodomain was obtained in two types of experiments. In the first, constructs containing only the serpentine domain of the TSHr were shown to increase basal cAMP levels when expressed in transfected cells [45,46]. In the second, chimeric molecules were made containing segments of LGR2 [47] (a Drosophila homolog of the GPHr with a high basal activity) and the LH/CGr (which is virtually devoid of basal activity). The results indicated the establishment of silencing interactions between a segment of the ectodomain (containing serine 277 of LH/CGr, see above) and the second extracellular loop of the transmembrane domain, provided they both originate from the LH/CGr [8]. From these experiments, one could propose that activation of GPHrs by their agonists results from the release of a silencing effect exerted by the unliganded ectodomain on an intrinsically active serpentine.

Whereas this would be in agreement qualitatively with the above experiments, it does not account for the observation that, when normalized to the level of receptor expression at the cell surface, the basal activity of serpentine-alone TSHr constructs is much lower than the maximal activity achieved after stimulation by saturating concentrations of the hormone, or in the most active serine 281 mutants [45]. In an attempt to integrate available information, we have proposed a model for the activation of the TSHr in which the ectodomain would act as a molecular switch (Fig. 2) [45]. In the "off" position, in the absence of hormone, the ecto domain acts as a tethered inverse agonist of the serpentine domain, minimizing basal activity. Binding of the hormone to the receptor stabilizes the "on" position, in which the ectodomain now behaves as a tethered full agonist. Mutations affecting serine 281 of the ectodomain similarly puts the switch in the "on" position. The relative potency of individual amino acid substitutions at S281 indicates a direct relation between the destructuring effects of the mutations and constitutive activity [44,48], suggesting that the gain of function results from a local loss of structure in the ectodomain.

Figure 2 Putative model of the intramolecular interactions involved in the activation of the TSH receptor. (A) The basal state of the receptor is characterized by an inhibitory interaction between the ectodomain and the serpentine domain; the ectodomain would function as a tethered inverse agonist. (B) Removal of the ectodomain releases the serpentine domain from the inhibitory interaction, resulting in partial activation. (C) Mutation of Ser281 into Leu switches the ectodomain from an inverse agonist into a full agonist of the serpentine domain. (D) Binding of TSH to the ectodomain is proposed to have a similar effect, converting it into a full agonist of the serpentine portion. It must be stressed that the scheme is purely illustrative. It emphasizes that, according to the model, activation does not require a direct interaction between the hormone and the serpentine domain. Such an interaction, however, is by no means excluded. (Adapted from Vlaeminck, V. et al., Activation of the cAMP pathway by the TSH receptor involves switching of the ectodomain from a tethered inverse agonist to an agonist. Mol. Endocrinol., 16, 736-746, 2002.)

Figure 2 Putative model of the intramolecular interactions involved in the activation of the TSH receptor. (A) The basal state of the receptor is characterized by an inhibitory interaction between the ectodomain and the serpentine domain; the ectodomain would function as a tethered inverse agonist. (B) Removal of the ectodomain releases the serpentine domain from the inhibitory interaction, resulting in partial activation. (C) Mutation of Ser281 into Leu switches the ectodomain from an inverse agonist into a full agonist of the serpentine domain. (D) Binding of TSH to the ectodomain is proposed to have a similar effect, converting it into a full agonist of the serpentine portion. It must be stressed that the scheme is purely illustrative. It emphasizes that, according to the model, activation does not require a direct interaction between the hormone and the serpentine domain. Such an interaction, however, is by no means excluded. (Adapted from Vlaeminck, V. et al., Activation of the cAMP pathway by the TSH receptor involves switching of the ectodomain from a tethered inverse agonist to an agonist. Mol. Endocrinol., 16, 736-746, 2002.)

A last set of experiments suggests that the molecular switch controlling activation of the serpentine domain must be a composite structure combining a portion of the ectodomain and the extracellular loops of the serpentine. A spectrum of well-defined activating mutations of the TSHr were engineered, either on a holoreceptor background or in serpentine-alone constructs. Whereas the mutations in the transmembrane segments or intracellular loops were equally effective on both backgrounds, mutations of the extracellular loops with a strong effect on the holoreceptor were totally ineffective on the serpentine-alone constructs [45]. This model does not rule out that activation of GPHr involves a direct interaction between the hormones and the serpentine portion of the receptors, but it indicates that such an interaction is not required to account for most observations. In the case of the TSHr, it also provides a rationale for the activation of the receptor by autoantibodies present in the plasma of patients with Graves'disease [6]. According to this model, stimulating autoantibodies would only need to have a "destructuring" effect on a segment of the ectosomain controlling the molecular switch.

Diabetes Sustenance

Diabetes Sustenance

Get All The Support And Guidance You Need To Be A Success At Dealing With Diabetes The Healthy Way. This Book Is One Of The Most Valuable Resources In The World When It Comes To Learning How Nutritional Supplements Can Control Sugar Levels.

Get My Free Ebook


Post a comment