Skip to main content
NCBI home page
Molecular Endocrinology logoLink to Molecular Endocrinology
. 2008 Dec 18;23(2):157–168. doi: 10.1210/me.2008-0384

Molecular Mechanism of Action of Pharmacoperone Rescue of Misrouted GPCR Mutants: The GnRH Receptor

Jo Ann Janovick 1, Akshay Patny 1, Ralph Mosley 1, Mark T Goulet 1, Michael D Altman 1, Thomas S Rush III 1, Anda Cornea 1, P Michael Conn 1
PMCID: PMC2646616  PMID: 19095769

Abstract

The human GnRH receptor (hGnRHR), a G protein-coupled receptor, is a useful model for studying pharmacological chaperones (pharmacoperones), drugs that rescue misfolded and misrouted protein mutants and restore them to function. This technique forms the basis of a therapeutic approach of rescuing mutants associated with human disease and restoring them to function. The present study relies on computational modeling, followed by site-directed mutagenesis, assessment of ligand binding, effector activation, and confocal microscopy. Our results show that two different chemical classes of pharmacoperones act to stabilize hGnRHR mutants by bridging residues D98 and K121. This ligand-mediated bridge serves as a surrogate for a naturally occurring and highly conserved salt bridge (E90–K121) that stabilizes the relation between transmembranes 2 and 3, which is required for passage of the receptor through the cellular quality control system and to the plasma membrane. Our model was used to reveal important pharmacophoric features, and then identify a novel chemical ligand, which was able to rescue a D98 mutant of the hGnRHR that could not be rescued as effectively by previously known pharmacoperones.

Computational modeling, site-directed mutagenesis, ligand binding, effector activation and confocal microscopy are used to show the molecular mechanism of pharmacoperone action on a GPCR, the hGnRHR.

G protein-coupled receptors (GPCRs) play central roles in almost all physiological functions; mutations in GPCRs are responsible for more than 30 disorders including cancers, heritable obesity, diabetes insipidus, blindness, and diseases involving the melanocortin type 4 and GnRH receptors [GnRHR (1,2)]. GPCRs constitute the most prevalent family of validated therapeutic targets in medicine, because 60–70% of approved drugs derive their benefits by selective targeting of this family (3,4). Many pathologies associated with misfolded mutant receptors occur because these are retained by the endoplasmic reticulum [ER (5)] and do not reach their normal site of function. Such mutants, when expressed in host cells, show greatly diminished expression at the plasma membrane, usually due to recognition/retention by the cellular quality control system [QCS (6,7,8)] of the ER. All of the first 17 reported point mutations of the human (h) GnRHR that lead to hypogonadotropic hypogonadism result in misfolded proteins (9). Such ER-retained mutants frequently show a change in residue charge compared with the wild-type (WT) receptor (12 instances), a gain or loss of either Cys (four instances) or a Pro (one instance). As misfolded proteins, these characteristically fail scrutiny by the cellular QCS, and misrouting occurs (8,9).

Pharmacoperones provide a folding template and enable many misfolded mutants to pass the QCS (8); they regain both the ability to bind agonist and the ability to transduce a signal. The ability to rescue and restore protein function presents a new therapeutic approach that is broadly applicable to the disease-causing mutants. In the case of the hGnRHR, pharmacoperones (9) rescue 15 of the first 17 mutants reported. These rescued mutants traffic correctly to the plasma membrane, where they bind ligand and successfully transduce signaling (9). The two mutants in this group that cannot be at least partially rescued by the pharmacoperone approach (10,11) are also misfolded and show significant structural distortion. In both of these cases [residues 168 and 217 in transmembrane (TM)4 and TM5, respectively], the mutation is identical, Ser→Arg, and both mutations occur in TM segments resulting in a change that rotates TM segments and precludes formation of the Cys14 –Cys200 bridge, essential for passage through the QCS (10).

Pharmacoperones for GnRHR mutants have now been identified from three different chemical classes [indoles, quinolones, and erythromycin macrolides (8)]. Despite the potential value of these agents for therapeutic approaches to mutant receptor rescue, little is known about their biochemical mechanism of action because they were selected in high throughput screens, rather than by design. We undertook elucidation of the biochemical mechanism by which pharmacoperones act with a view toward allowing rational design of such agents and expanding their use to other misfolded/misrouted proteins.

Results

Computational modeling

The homology model of the hGnRHR was predicted using the 2.8 Å resolution crystal structure of bovine rhodopsin; the β2-adrenergic receptor was not available at the time this work was done. The overall topology of the hGnRHR model structure is similar, showing seven TM helices and intracellular loop and extracellular loop (EL) regions (Fig. 1A). The distance between sulfur atoms of Cys14 and Cys200 is approximately 3.5Å, allowing formation of a disulfide bridge not found in other GPCRs; biochemical evidence suggests that this bridge is essential for routing of the human, but not rat or mouse, GnRHR to the plasma membrane (10). The developed hGnRHR model possesses a highly conserved disulfide bridge between EL1 Cys114 and EL2 Cys196, also essential for plasma membrane routing of the GnRHR in all species examined (10,11).

Figure 1.

Figure 1

Open in a new tab

A–D, Homology modeling of hGnRHR and docking of ligands. A, Three-dimensional topology of the predicted model of hGnRHR. 7-TM helices are shown in red, and intracellular loops, ELs, and N terminus are shown in green. The position of the predicted non-peptide-binding site is shown in the dotted circle. B, Predicted docking pose of Q89 in the hGnRHR binding cavity. Polar interactions (hydrogen bonding, ion pair, and ammonium-pi) between ligand-receptor are shown in green dotted lines. EL2 is shown as a cyan ribbon. Alternate side-chain rotamer of His306 is shown in pink (carbon) sticks. C, Predicted docking pose of In3 in hGnRHR binding cavity. D, Overlap of the docked pose of In3 onto Q89 docked in the hGnRHR binding cavity. Common pharmacophoric features of In3 and Q89 interacting with hGnRHR-binding site residues are highlighted by brown dotted circles. E and F, Chemical structures of indole and quinoline pharmacoperones used in this study showing IC50 values in parentheses. Common chemical cores among ligands are highlighted in blue.

Docking, followed by a flexible ligand-receptor minimization, was used to predict the binding sites of two classes of GnRHR antagonists, indoles (In3, In30, In31b) and quinolones (Q08, Q76, Q89; Fig. 1, E and F). The IC50 values previously obtained for hGnRHR binding are shown in parentheses. We identified how the common chemical features of these ligands (a basic amine, a heterocyclic core, and a 3.5-dimethyl aryl group) are oriented in the binding site, which helps rationalize the structural basis of pharmacoperone rescue ability for some of these compounds.

The binding pose was predicted for Q89 in the hGnRHR homology model (Fig. 1B). E90K is a naturally occurring mutant of hGnRHR that results in a misfolded receptor and human disease (12). Residues E90 (TM2) and K121 (TM3) form a salt bridge interaction in the apo-hGnRHR receptor model. Thus, the substitution of a basic residue at the E90 position creates an unfavorable charged interaction with the basic K121 residue that might be responsible for the misfolded E90K mutant receptor. Q89 appears docked in a pocket surrounded by TM2, TM3, TM6, and TM7 and lies just below EL2. The E90-K121 interaction appears to be disrupted in the Q89-docked model of the hGnRHR to accommodate the ligand. The E90 residue forms an ion-pair interaction with the basic piperazine nitrogen atom, and the K121 side chain forms a cation-π interaction with the substituted phenyl group of Q89. K121 mutations are known to decrease the ligand binding affinity (13). It can be hypothesized that the above-mentioned charged interactions between Q89 and the receptor make up for the disrupted E90–K121 salt bridge in the apo-receptor. It is also known that F313L mutation causes a 360-fold decrease in the binding affinity of Q89 (14). It can be seen that F313 (TM7) forms an orthogonal stacking hydrophobic interaction with the quinolone moiety of Q89. F309 (TM7) stacks parallel to the quinolone moiety from the other side. F309L/Q mutation is also known to decrease the nonpeptide ligand binding affinity (15). Y283 (TM6) and W280 (TM6) form the right wall of the binding pocket and are involved in hydrophobic interactions with the substituted phenyl group of Q89. It is known that the W280F mutation leads to decreased nonpeptide ligand binding affinity (14,15). The polar amide linker of Q89 points toward the water-solvated EL region. H306 (TM7) points away from the binding site; however, it was observed that an alternate rotamer position for H306 side chain (shown in pink carbon atoms; Fig. 1) can swing it toward the binding pocket to form a hydrogen bond with the amide carbonyl oxygen atom of Q89. Another key finding, based on the predicted binding site of Q89, is the identification of an additional acidic residue (D98) close to the binding pocket. D98 (TM2) is within ion-pair interaction distance from the basic piperazine nitrogen atom of Q89. Thus the basic nitrogen atom of the quinolone compounds forms a network of tight ion-pair interactions with E90 and D98. We hypothesize that in the absence of E90 interaction with K121, D98 can act as a surrogate anchor point for the critical ion-pair interaction between the receptor and ligands. The above possibility explains the pharmacoperone ability of Q89 to rescue the misfolded naturally occurring E90K mutant receptor (8).

Figure 1C shows the predicted binding site for In3 in the hGnRHR homology model. E90 forms an ion-pair interaction with the basic amine nitrogen atom, and K121 forms a cation-π interaction with the substituted phenyl group of In3. The indole moiety of In3 is sandwiched between F313 and F309 and is involved in hydrophobic interactions with these residues. Y283 and W280 form the right wall of the binding pocket and are involved in hydrophobic interactions with the substituted phenyl group of In3. The polar amide carbonyl group appears to be pointing toward the solvated EL region of the receptor. Similar to Q89, the basic amine nitrogen atom of In3 is also within appropriate distance from D98 to allow for an additional ion-pair interaction.

The docked poses of Q08 and Q76 and In30 and In31b also possess a very similar overall orientation within the GnRHR binding pocket to that of Q89 and In3, respectively. The docked poses of both the indole and quinolone series of compounds are consistent within each class as well as among both classes of ligands. In addition, several ligand-receptor interactions observed in the hGnRHR homology model appear to be consistent with the structure-activity relation and site-directed mutagenesis data (see above). Comparison of the docking poses of Q89 and In3 shows several key conserved interactions (Fig. 1D). The substituted phenyl moiety is involved in hydrophobic interactions with Y283 and W280. The quinolone/indole core is involved in hydrophobic interactions with F313 and F309. The polar amide linker in both Q89 and In3 is pointing toward the water-solvated EL region. It is very interesting to note that the protonated nitrogen atom of both Q89 and In3 occupies the same three-dimensional space and forms a network of ion-pair interactions with E90 and D98. We then set out to use site-directed mutagenesis to test this model.

Site-directed mutants: expression and rescue of single and double mutants at the plasma membrane

To test the postulated role for D98 in the biochemical mechanism of action of pharmacoperones, we constructed three single mutants of the hGnRHR (D98A, D98K, and D98N) and three double mutants (E90K/D98A, E90K/D98K, and E90K/D98N). The three single mutants (transfected at 95 ng) responded at basal levels only to the agonist, Buserelin (10−7 m; Hoechst-Roussel Pharmaceuticals, Somerville, NJ); Fig. 2A). The axes of all panels in Fig. 2 are identical to allow comparisons. Likewise, none of the three double mutants (also transfected at 95 ng) responded measurably to Buserelin (Fig. 2A). As controls, we also included WT hGnRHR, known to be only fractionally routed to the plasma membrane (16), as well as the mutant E90K [rescuable by pharmacoperones (8,12)] and two mutants, S168R and S217R, that cannot be rescued by pharmacoperones (10). These two mutations (E90 and D98) promote loss of the physical relation between the amino-terminal and EL2 that normally allows formation of the essential Cys14-Cys200 bridge. hGnRHR mutants that cannot form this bridge are recognized as misfolded by the cellular QCS (10) and are retained in the ER. None of these three control mutants produced a measurable response to Buserelin.

Figure 2.

Figure 2

Open in a new tab

Effect of mutation of residue D98 with A, K, or N on rescue with two classes of pharmacoperones. Total IP production was measured in response to a saturating concentration of Buserelin (10−7 m). A–C, Data are expressed as percent of WT without 1 μg/ml In3 or 2.5 μg/ml Q89 (at 95 ng, black bar, panel A; and 5 ng, black bar, panel D). Cells were transfected with 95 ng of WT or mutant cDNA (A–C) as described in Materials and Methods. Mutants S168R and S217R were used as controls, because these mutants cannot be rescued from ER retention with pharmacoperone. Mutant E90K was included as a positive control for rescue with pharmacoperone. Inset illustration shows the dose-response curves (log scale) for each class of pharmacoperone rescue. D–F, Dominant-negative effect of cotransfecting cells with 5 ng WT and 95 ng mutant cDNA (1:19) on pharmacoperone rescue as described in Materials and Methods. Averages and sems of at least three independent experiments performed in replicates of six are shown. DMSO, Dimethylsulfoxide.

Mutant D98N (95 ng) was rescuable by In3 (Fig. 2B; 33 ± 5% of the level of 95 ng hWT) and more modestly by Q89 (Fig. 2C; 13 ± 2% the level of 95 ng hWT). For comparison in Fig. 2, B and C, an inset is shown with the dose-response curves for rescue of WT hGnRHR and E90K hGnRHR with each pharmacoperone used. The (antagonist) pharmacoperones, to serve as rescue agents, are not present at the same time as the GnRHR agonist (see Materials and Methods). As expected E90K, S168R, and S217R (without pharmacoperone rescue) did not produce a response, although the E90K was rescuable by each of the two pharmacoperones, as previously shown (9,10,11,12).

Among the three double mutants (E90K/D98A, E90K/D98N, E90K/D98K), further encumbered by the inability to form the E90–K121 salt bridge, there was no response to Buserelin and no ability to rescue with any of the potential pharmacoperone molecules (Fig. 2, A–C).

In evaluating the preceding data, it is important to recognize that D98 and K121 (see mutants described below) are also believed to be points of contact for GnRH and for other GnRHR agonists (13,17). Accordingly, the inability to observe responsiveness might reflect inability to bind Buserelin or bind pharmacoperone, the retention of the mutant by the QCS, or a combination of these. To distinguish whether the loss of responsiveness resulted from the loss of GnRHR agonist binding or from the retention of the receptor by the ER QCS, we sought to use a methodology that did not rely on binding of ligand by the mutant.

Dominant-negative effect of single and double mutants on WT hGnRHR

We took advantage of the dominant-negative effect of GnRHR mutants (7,18). Because the movement of the newly synthesized receptor from the ER to the plasma membrane involves oligomerization, and the cellular QCS assesses the overall quality of the oligomer (potentially a combination of mutant and WT), the presence of the mutant also results in retention of WT GnRHR. Accordingly, we cotransfected WT hGnRHR (5 ng) in the presence of excess (95 ng) of each of the three single mutants, three double mutants, or control mutants described above and then assessed the ability to measure coupling due to WT receptor with or without each potential pharmacoperone (Fig. 2, D–F). The ratio of 1:19 (WT to mutant) has been shown to be optimally effective (7) because it increases the chances that individual cells that receive WT hGnRHR also receive the mutant. Moreover, this ratio minimizes the formation of WT:WT oligomers that would traffic correctly to the plasma membrane.

Cotransfection of WT with each of the D98 mutants (Fig. 2D) leads to the most retention of WT GnRHR by D98K, suggesting that this mutant is retained in the ER. These observations suggest that mutants D98A and D98N exert a more modest dominant-negative effect on WT hGnRHR (5 ng). In the case of D98N, there is measurable rescue by In3 (Fig. 2E) and by Q89 (Fig. 2F and Ref. 16).

When the dominant-negative effect on WT hGnRHR due to cotransfection with the double mutants was examined, it resulted in a very modest response, as occurred for E90K. E90K, however, could be rescued by In3 and Q89. Mutants S168R and S217R could not be rescued (Fig. 2, D–F), as previously reported and explained (10). We next used confocal microscopy to provide visual support for the modeling and biochemical findings.

Confocal studies and identification of a novel pharmacoperone

To identify compounds with improved pharmacoperone activity for rescue of mutants at position 98, a focused screening set was assembled based on the model described in Fig. 1. All of the compounds presented thus far (Fig. 1, E and F) contain either a basic secondary amine (indole series) or piperidine (quinolone series) that is predicted to form a bridge between residues D98 and E90. Upon mutation of D98 to A, N, or K, disruption of a salt bridge interaction would be expected, potentially leading to decreased binding affinity and a reduction in pharmacoperone activity for these compounds. To identify compounds that could potentially have improved activity against these mutations, we tested approximately 50 additional compounds in the indole and quinolone series that exhibited strong binding to GnRHR but which lacked this basic amine substituent. These compounds contained either a hydrophobic or polar group as a replacement for the basic amine, which could potentially interact more favorably with D98A or D98N while avoiding a charge-charge repulsion with D98K. One of these compounds (Q103; structure in Fig. 3A), containing a tetrahydrofuran moiety, showed improved rescue for both the D98A and D98N mutants as compared with In3 (Fig. 3B). We used In3 because that is our standard control in most studies.

Figure 3.

Figure 3

Open in a new tab

Mutational analysis of D98K-GFP and D98N-GFP chimeras. A, Structure of Q103. B, Curves showing pharmacoperone activity of Q103 or In3 are shown for WT hGnRHR, D98A, or D98N (25 ng). Total IP was measured in response to a saturating dose of Buserelin (10−7 m). Averages and sems were calculated from at least three independent experiments performed in replicates of four. Values are shown as percent of WT hGnRHR without pharmacoperone drugs. C, Total IP was measured for responsiveness of D98K-GFP and D98N-GFP (25 ng) with or without attempted rescued by In3 or with a highly effective quinolone pharmacoperone, Q103 (structure shown), to a saturating dose of 10−7 m Buserelin. Averages and sems were calculated from at least three independent experiments performed in replicates of four. D, Confocal images of D98K-GFP and D98N-GFP (25 ng, green) in cells colabeled with WGA (red) and ER tracker white (blue). Bar shows 10 μm for all images in the top panel. The lower left panel shows membrane regions of D98K and D98N in cells after pharmacoperone (2.5 μm) treatment, different from the cells used in the top panel. Images were enlarged to show lack of colocalization with WGA in the case of D98K and the presence of colocalization resulting in a yellow overlay in the case of D98N. The right lower image shows an example of ER formations suggesting extreme retention of proteins, frequently seen in the case of D98K expressing cells. The bar in the lower panel shows 5 μm. DMSO, Dimethylsulfoxide.

Although we have not explicitly modeled Q103 into the hGnRH homology model, its similarity to Q89 other than the replacement of the piperidine side chain of Q89 with a tetrahydrofuran (THF) ring suggests that Q103 might be able to assume a similar orientation of the quinolone scaffold. Clearly, the activity of both compounds in the WT receptor suggests that a different orientation of the side-chain and/or separate residues interacting differentially with the side chain must be possible because one would not expect a THF ring to favorably interact with the acid function of D98. Proximate to D98 and directed toward the putative binding cavity is the side chain of Q195 (Cβ–Cβ distance is ∼5.6 Å), a residue from EL2 preceding C196, which could provide a favorable hydrogen bond between the THF ring acceptor oxygen atom and the NεH atoms of Q195 (data not shown).

Previously, both Q195 and the nearby H306 have been modeled to interact with the terminal carbonyl oxygen atom of GnRH peptide G10 residue (19). This is further supported by an independent IP (inositol phosphate) readout assay for Q103 (hWT, 8,772 cpm; D98A, 9,404 cpm; D98N, 10,074 cpm). It is evident that by mutation to D98A and D98N, the IP readout increases. This suggests that Q103 interacts better in just the absence of a negatively charged D98 residue (D98A), avoiding unfavorable interactions with the polar THF ring. It gains a further potency boost with D98N, where it can most favorably hydrogen bond with the side chain of D98N, similar to the suggested interaction with Q195 in the WT phenotype (above).

Additionally, the improved ability of Q103 to rescue the above mutants could also be due to enhanced bioavailability within the cell compartment which, like binding affinity, is also a component of pharmacoperone rescue potency. The calculated logP values for In3, Q89, and Q103 are 6.9, 4.7, and 3.9, respectively. It is generally agreed that the closer is the clogP range of a small molecule to 0–3, the better is the balance between solubility and permeability, thus leading to a better presence in the cell (20). Therefore, considering a more modest clogP value of Q103 compared with In3 and Q89, and also noting that Q103 is an uncharged molecule unlike In3 and Q89, Q103 might have a better bioavailability within the cell, thus indirectly dictating its binding affinity and pharmacoperone ability.

Figure 3C shows the responsiveness of D98K-green fluorescent protein (GFP) and D98N-GFP (with or without In3 or Q103) to 10−7 m Buserelin. As shown above for D98K and D98N, the GFP chimera of those mutants were either nonresponsive (in the case of D98K-GFP) or both responsive and rescuable (in the case of D98N-GFP).

D98K-GFP and D98N-GFP (shown in green) were imaged in cells also labeled with ER stain to show the ER, and wheat germ agglutinin (WGA)-Alexa 633 (red) to stain plasma membrane (Fig. 3D). Single confocal images of GFP-expressing cells, approximately 0.5 μm thick, were acquired 1–2 μm above the coverslip. D98K-GFP was expressed rather uniformly throughout most cells, completely covering the area occupied by ER tracker (blue).

Treatment with In3 [2.5 μm (∼1.4 μg/ml)] or Q103 (2.5 μm) did not visibly affect this distribution (Fig. 3D, top row) of D98K-GFP, which was primarily in the ER. D98N-GFP was uniformly distributed in the intracellular space but showed increased concentration on the plasma membrane after treatment with In3 (overlap of WGA stain for the plasma membrane and GFP is shown in yellow). Treatment with Q103 produced even more pronounced concentration of receptor at the plasma membrane, and occasional aggregates were observed (Fig. 3D, smaller images). The distinction between receptors expressed throughout the cell, including those close to the plasma membrane (as in the case of D98K after Q103 treatment) or those concentrated at the plasma membrane (as in the case of D98N after Q103 treatment) is shown in cropped regions of cell membrane in the smaller images in Fig. 3D. An overlap (shown by the yellow display) is observed only for D98N.

Many of the D98K-GFP expressing cells showed large circular formations of organized smooth ER associated with extreme retention of proteins (21), shown in the lower right panel of Fig. 3D (ER tracker). The data suggest that D98K-GFP is retained in the ER and cannot be rescued by In3 or Q103. D98N-GFP, in contrast, can be modestly rescued.

These studies support the conclusions of localization of the mutants from mutational and dominant-negative studies, and we next used radioligand binding as marker for the presence of receptor and receptor mutants.

Binding studies

We performed binding studies (Fig. 4A) on WT hGnRHR and the D98A, N or K mutants alone (95 ng) and also assessed binding in a dominant-negative protocol (i.e. for Fig. 2, a dominant-negative protocol with a 1:19 ratio of WT:mutant [i.e. 5 ng WT + 95 ng mutant (as described above)]. In the former instance (Fig. 4A), very little binding could be detected in the cells, and in the latter instance (Fig. 4B), all three of the mutants appeared to cause diminution of measurable WT GnRHR, presumably due to retention of this moiety in the ER. The data from these studies are consonant with the confocal and IP studies and suggest that mutants D98A, N, and K are retained in the ER.

Figure 4.

Figure 4

Open in a new tab

Binding characteristics of mutating the D98 residue. A, Saturation binding studies were performed using 95 ng WT hGnRHR and the D98X mutant cDNAs as described in Materials and Methods. Very little binding of each mutant could be detected in the cells compared with WT. B, Saturation binding of a dominant-negative protocol with a 1:19 ratio of WT:mutant (5 ng and 95 ng, respectively). All three of the mutants appeared to cause diminution of measurable WT GnRHR, presumably due to retention of this moiety in the ER. Data are expressed as percent of WT for specific binding. Averages and sems were calculated from at least three independent experiments, each performed in replicates of four.

Effect of deleting K191, a residue that regulates trafficking of the hGnRHR

We also examined (Fig. 5) a means of rescuing the mutants that did not depend on pharmacoperones to address whether the D98 mutants could be rescued when drug interactions with this residue were not required. Residue K191 is associated with steric interference of the formation of the Cys14–Cys200 bridge in primates and, when deleted, increases expression at the plasma membrane. Mutants such as E90K that are rescuable by pharmacoperones are also readily rescuable by deletion of this amino acid alone (20). Accordingly, we prepared sequences in which K191 was deleted from the mutants of D98A, D98K, or D98N. Expressed alone D98AΔK191 and D98NΔK191 (25 ng), respond to 10−7 m Buserelin at nearly all WT levels (Fig. 5). This evidence indicates that the GnRH agonist, Buserelin, is able to bind adequately to the mutants that contain D98N and D98A, and these mutants can couple to effectors. This suggests that the modest-to-nil rescue by pharmacoperones of mutants D98A, D98K, or D98N is best explained by the failure of pharmacoperones to bind to such mutants. The observation that deletion of K191 from D98K does not rescue responsiveness suggests that it is, in addition, retained in the ER like other molecules in which substantial charge changes are made in amino acids in the TM helices (9).

Figure 5.

Figure 5

Open in a new tab

Effect of deleting K191 on D98 substitutions of A, K, and N. Cells were transfected with 25 ng WT hGnRHR or mutant cDNA as described in Materials and Methods. Total IP production was measured in response to a saturating concentration of Buserelin (10−7 m). D98A and D98N were rescuable after deletion of K191 whereas deletion (Δ) of K191 from the D98K mutant showed no rescue. Averages and sems of at least three independent experiments performed in replicates of six are shown.

Effect of mutation at residue 121

To assess the predicted role for residue K121, we constructed mutants in which that residue was replaced by A, D, E, G, N, Q, or R (Fig. 6A) and a similar series in which K191 was also deleted from the same mutants (Fig. 6B). Among those single mutants at residue 121 (Fig. 6A), only in the case of the conservative substitution, K121R, was there a response to Buserelin that was comparable to the WT hGnRHR. There was a slight rescue of K121A and K121Q and a more modest rescue of K121G and K121N. There is virtually no response when K121 is converted to the negatively charged D or E. Further deletion of K191 results in responses from K121A and K121Q, but not the negatively charged D or E. This latter observation suggests that the mutation K121Q itself does not preclude trafficking to the plasma membrane or binding its agonist (i.e. when K191 is deleted); however, this mutation does result in a protein that is unable to bind In3 or Q89 or be fully rescued by these pharmacoperones.

Figure 6.

Figure 6

Open in a new tab

Mutation analysis of K121 residues on the effect of the salt bridge and their relation to the deletion of K191 residue. A, Cells were transiently transfected with 25 ng of WT hGnRHR or mutant cDNA in which residue K121 was replaced by A, D, E, G, N, Q, or R. Cells were treated with or without pharmacoperone In3 or Q89 as described in Materials and Methods. B, Deletion of K191 shows pharmacoperone rescue in all mutants except K121D, and a nominal response in K121E, G, and N. Total IP was measured in response to a saturating dose of Buserelin (10−7 m). Empty vector (pcDNA3.1) was run as a control and was typically 175 ± 20 cpm. Averages and sems were calculated from at least three independent experiments performed in replicates of six. DMSO, Dimethylsulfoxide.

Dissociation constant (Kd), Bmax, and displacement of radioligand binding by pharmacoperones, In3 and Q89

Table 1 shows Kd (nm) and Bmax (fmol/105 cells) data for WT hGnRHR, K121R, and for the homologs of those two mutants lacking residue K191. These four moieties were the only ones for which Scatchard analysis could be performed. hD98A, hD98K, and hD98N do not express at measurable levels either assessed by IP response to the GnRH agonist (Fig. 2) or by binding (Fig. 4B). hD98NΔK191 and hK121QΔK191 have EC50 values in the IP assay greater than 5 × 10−8 m for Buserelin (data not shown) and hD98NΔK191, K121QΔK191, and K121R bind with low affinity in the Scatchard assay (Fig. 7 and data not shown). Other K121 mutants express at levels too low to measure (Fig. 6). The data show that replacement of K121 by R decreases both the affinity of tracer binding and the number of binding sites per cell. Removal of K191 increases the number of binding sites in the case of the WT receptor and hK121R, whereas decreasing the Kd of this mutant.

Table 1.

Bmax and Kd of select hGnRHR mutants and WT hGnRHR

hWT hWTΔK191 hK121R hK121RΔK191
Bmax (fmol/105 cells) 8.0 33.8 4.8 16.9
Kd(nm) 0.7 0.8 4.5 1.3
Open in a new tab

Cells were plated, transfected with 50 ng of the indicated vector, and binding parameters were determined as described in Materials and Methods. 

Figure 7.

Figure 7

Open in a new tab

Displacement of [125I]Buserelin binding by pharmacoperones In3 and Q89. Cells were plated, transfected with the indicated vector, and incubated in 1.25 × 105 cpm/ml of [125I]Buserelin in the presence of the indicated concentrations of pharmacoperones In3 and Q89. Binding was determined as described in Materials and Methods.

Figure 7 shows displacement of [125I]Buserelin by pharmacoperones In3 and Q89. Both of these displace binding of tracer, virtually to undetectable levels. In the case of moieties in which residue K191 is deleted, In3 is able to distinguish these mutants from those containing this reside. A higher concentration of In3 is required to displace an equal amount of [125I]Buserelin binding when K191 is absent. Q89 displacement of tracer binding is unable to distinguish WT GnRHR and the homolog lacking K191. This difference in dependence on the K191 suggests that these pharmacoperones occupy the molecular space relative to K191 differently. The inset shows the very low binding (note axis is counts per min) of mutants hD98NΔK191 and hK121QΔK191.

Discussion

Computational modeling of the human GnRHR predicted an E90–K121 salt bridge proximal to the binding site for pharmacoperone drugs. This bridge, broken in the naturally occurring hGnRHR mutant E90K, causes hypogonadotropic hypogonadism because the misfolded mutant receptor fails the cellular QCS and cannot traffic to the plasma membrane (8,22). This failure results in the inability to transduce signals from the natural ligand and therefore diminishes circulating gonadotropin levels. Pharmacoperone drugs correct the folding of E90K and other mutants and allow them to pass the cellular QCS (6,9,23). We used site-directed mutagenesis, radioligand binding, confocal microscopy, and assays for effector coupling, to confirm the model’s prediction that pharmacoperones from two different drug classes both rescue receptor mutants by creating a ligand-mediated bridge between residues D98 and K121. This serves as a surrogate for the E90–K121 salt bridge present in the WT hGnRHR. Stabilizing the relation between TM helices 2 and 3 increases the probability of mutant receptors being able to pass the cellular QCS. Other points of contact for these small molecule antagonists (distant from D98 and K121) have been previously reported (14,15), as described in Materials and Methods.

This study was complicated because residues D98 and K121 are also points of contact for binding by GnRH and its agonists (13,17) as well as for the pharmacoperones. In addition, because we lack antisera to this receptor, ELISA and Western techniques cannot be applied. We relied on three techniques that assessed the cellular localization of the mutants without direct binding of GnRH or its analog, Buserelin. These approaches were deletion of amino acid K191 [a primate-specific residue that inhibits routing of the hGnRHR to the plasma membrane (16)], confocal microscopy (24), and the dominant-negative effect that occurs because newly synthesized hGnRHR and mutants oligomerize for routing to the plasma membrane (7,18).

The charged residues in both the E90–K121 and the ligand-mediated D98–K121 bridges, although rare among the hydrophobic amino acids of the TM helices 2 and 3, are highly conserved in GnRHRs. D98 is absolutely conserved in all mammalian, reptilian, avian, and piscine GnRHRs sequenced to date. In the fruit fly, a conservative change is made to E98. Likewise, K121 is maintained in the same groups and in fruit flies, the residue is a conservative change, R121. E90 is conserved in mammals, but V90 is present in eels, reptiles, avians, flies, and perciform fish, and the residue is M90 in trout and catfish. Another apparent point of contact for quinolone pharmacoperones is the highly conserved F313 [L313 in canines and equines, and already reported to be the basis of the inability of these species to recognize these drugs, (14)], and it is certainly possible that this could result in further structural stability to the receptor.

It was initially curious to us that pharmacoperone drugs from different chemical classes all happened to interact identically by creating a surrogate bridge for E90–K121. Although it is conceivable that the correctly formed structure of the ligand-binding site is, in some way, related to the configuration of the receptor that is allowed to pass the cellular QCS, we considered other possibilities, including prejudice in the screening process used to select these drugs. In that regard, all the pharmacoperones used in the present study were selected from high-throughput screens for antagonism of the natural ligand. Accordingly, as competitors of the natural ligand, it is not surprising that they would interact at (or near) the ligand-binding site. This site resides in the lateral plane of the plasma membrane, a region characterized by a high percentage of hydrophobic residues. The linear sequence around E90 is, for example, LLE90TLIVMPLD98 and around 121 is VLSYLK121LFSM. This is a predominantly hydrophobic region with a modest number of ionic or polar groups. Accordingly, the observation of this common ionic site could reflect that the drugs were all selected with the same prejudice for this preferential ion pair and/or polar interaction with the charged residue sites. Accordingly, our data do not allow the conclusion that stabilization of the ligand-binding site is, itself, sufficient for a pharmacoperone to allow a molecule to pass the cellular QCS.

It is clear that pharmacoperones rescue most of the GnRHR mutants (7,8,9,10,11,12), even though mutations appear at many sites in the receptor, both in the TM component and in intra- and extracellular sites. It was initially curious that stabilization of the relation between TM2 and TM3 would successfully rescue such a diverse set of mutations. This may reflect the highly interactive nature of GPCRs and the critical requirement of this salt bridge for the chaperone system of the cell to recognize the protein as correctly folded.

The present study emphasizes the significance of the E90–K121 salt bridge for passage through the cellular QCS and the ability of pharmacoperone drugs to stabilize these mutants by creation of a substitute ligand-mediated bridge between D98 and K121, adding to our understanding of the physical attributes of the hGnRHR necessary for correct routing (25). The observation that pharmacoperones appear to create a surrogate D98–K121 bridge that replaces or augments the naturally occurring salt bridge (E90–K121) provides the basis of rational design of this class of drugs. The observation that many mutants and WT proteins, like the hGnRHR, itself are inefficiently expressed due to misfolding and subsequent retention in the ER (11,22,23,24,25,26,27,28,29,30) suggests that many proteins, including GnRHR mutants that form the basis of hypogonadotropic hypogonadism, may be candidates for rational drug design by this approach.

The use of pharmacoperones in vivo is recent, but there are some successes. In a mouse model, Pey et al. (31) successfully used compounds obtained from a chemical screen to treat rodents with phenylketonuria, an inherited metabolic disease caused by mutations in phenylalanine hydroxylase, the enzyme that converts Phe to Tyr. Another success involved patients with X-linked nephrogenic diabetes insipidus. Mutant vasopressin 2 receptors in nephrogenic diabetes insipidus results in misrouted proteins that are trapped in the ER, degraded, and do not reach the plasma membrane in the collecting ducts of the kidney where they would normally promote water reabsorption. In vitro studies indicated that a nonpeptide V1a receptor antagonist rescued cell surface expression and function of mutant V2 receptors. When applied in vivo, a short-term treatment with a V1a receptor antagonist showed that patients given this molecule decreased both 24-h urine volume and water intake. Maximum increase in urine osmolality was observed on d 3 and sodium, potassium, and creatinine excretions and plasma sodium were constant throughout the study (32). These studies suggest that rational design of these therapeutic agents, e.g. ones that do not compete with endogenous ligands, is likely to assist this therapeutic approach.

Materials and Methods

Computational modeling

The structural template used for homology modeling of the hGnRHR was the 2.8 Å resolution x-ray crystal structure of bovine rhodopsin (Protein Data Base identification no. 1F88). Two additional class A GPCR sequences were used for multiple sequence alignment, including the human melanocortin 4 receptor and human GPCR 119 (hGPR119). Multiple sequence alignment was performed using QUANTA (Accelrys, Inc., San Diego, CA). The highly conserved TM helix residues were aligned across all GPCR sequences. Wherever appropriate, gaps were inserted into the sequences to find an optimal alignment, ensuring that no gaps were present in the TM regions. The final multiple sequence alignment was submitted to MODELLER (version implemented in QUANTA) for generating homology models of the hGnRHR. During the model generation, the level of optimization was set to the high setting. Two models were selected from a set of 20 models based on protein health and structural and geometry checks. Further refinement of these two models was performed by energy minimization followed by a short molecular dynamics simulation within QUANTA. The model with overall good energy/structural parameters and appropriate distance for the second disulfide bridge (Cys14 and Cys200) was selected for further docking studies. A set of 300 and 600 low-energy conformations was generated for each quinolone and indole ligands, respectively, using implementation of distance geometry (33). These conformations were energy minimized using the Merck Molecular Force Field (MMFF) with a distance-dependent dielectric of 2r (this indirectly accounts for any artificial charge-charge interactions during minimization.34). For each ligand, 50 diverse conformations were selected for subsequent docking using FLOG (35). Because it was assumed that the ligands bind at the orthosteric binding site of the receptor as the majority of these ligands act as GnRHR antagonists, in addition to their possible pharmacoperone ability, a 20Å cube was centered on the approximate retinal binding pocket, and the entire residues, any portion of which is found inside the cube, were used to generate the docking grids. These are meant to represent seven physico-chemical atom types: cation, anion, donor, acceptor, polar, hydrophobic, and other which, in turn, are used to create match centers to nucleate matching complementary ligand-receptor interactions. During docking the protonation states of residues were kept charged, and visual inspection ensured correct assignment of tautomers where applicable. Based on existing structure activity relationships and mutation data for the GnRHR, two essential match centers were defined on the grid proximate to residues F313 and E90, which were used to guide docking. The docking-derived ligand poses were then submitted to a flexible ligand-receptor minimization using the MMFF with a distance-dependent dielectric constant of 2r. The backbone of protein was restrained with a force constant of 50 kcal/mol · Å2, and all residues lying within 8 Å of any given ligand-docked pose were kept flexible during minimization. This was surrounded by a second shell of residues beyond the 8 Å region around docked poses, which was kept rigid.

Materials

pcDNA3.1 (Invitrogen, Carlsbad, CA), the GnRH analog, d-tert-butyl-Ser6-des-Gly10-Pro9-ethylamide-GnRH (Buserelin), myo-[2-3H(N)]- inositol (PerkinElmer, Waltham, MA; NET-114A), DMEM, OPTI-MEM, lipofectamine, PBS (Life Technologies, Inc.; Invitrogen), competent cells (Promega Corp., Madison, WI), and Endofree maxi-prep kits (QIAGEN, Valencia, CA), were obtained as indicated. Small molecules, shown to serve as pharmacoperones (8), were obtained as described: In3, (2S)-2-[5-[2-(2-azabicyclo[2.2.2]oct-2-yl)-1,1-dimethyl-2-oxoethyl]-2-(3,5-dimethylphenyl)-1H-indol-3-yl]-N-(2-pyridin-4-ylethyl) propan-1-amine and its analogs (Merck & Co., Rahway, NJ, 8); Q89, (7-chloro-2-oxo-4-{2-[(2S)-piperidin-2-yl]ethoxy}-N-pyrimidin-4-yl-3-(3,4,5-trimethylphenyl)-1,2-dihydroquinoline-6-carboxamide) and its analogs [Merck & Co., Rahway, NJ (8)]. In the present study we used trypan blue exclusion to show cell viability after drug exposure. These studies were highly specific for the GnRHR and were screened for more than 100 other structures, including, α-adrenergics, β-adrenergics, adenosine receptor, bradykinin, CB1 and CB2, dopamine receptors, neurokinins, prostanoid receptors, serotonin receptors, somatostatin, calcium, sodium, potassium channels, monoamine oxidases, and several phosphatases. Other analogs used in the modeling studies are defined in a prior publication (8).

Mutant receptors

Human WT and mutant GnRHR cDNAs for transfection were prepared as reported elsewhere (12); the purity and identity of plasmid DNAs were verified by dye terminator cycle sequencing (Applied Biosystems, Foster City, CA). The green fluorescent chimeras of hGnRHR[D98K], D98K, and hGnRHR[D98N], D98N contain the spacer previously described (7).

Transient transfection

Cells were cultured in growth medium (DMEM, 10% fetal calf serum, 20 μg/ml gentamicin) at 37 C in a 5% CO2 humidified atmosphere. For transfection of WT or mutant receptors into COS-7 cells, 5 × 104 cells were plated in 0.25 ml growth medium in 48-well Costar cell culture plates. The cells were washed once, 24 h after plating, with 0.5 ml OPTI-MEM and then transfected with WT or mutant receptor DNA with pcDNA3.1 (empty vector) to keep the total amount of DNA constant (100 ng/well). Lipofectamine was used according to the manufacturer’s instructions. DMEM (0.125 ml) with 20% fetal calf serum and 20 μg/ml gentamicin was added 5 h after transfection. The medium was replaced 23 h after transfection with 0.25 ml fresh growth medium. Where indicated, pharmacoperones (indicated concentration) in 1% dimethylsulfoxide (vehicle) was added for 4 h in respective media to the cells, and then removed 18 h before agonist treatment (10).

Confocal microscopy

Cells (105 per well) were plated in 1 ml DMEM/10% fetal calf serum/20 μg/ml gentamicin in Lab-TekII Chamber no. 1.5 German Coverglass slides (Nalge Nunc, Naperville, IL) and transfected with 25 ng mutant with or without pharmacoperones as described above. The cells were pretreated, 23 h after transfection, with pharmacoperone for 4 h, washed, and incubated for an additional 18 h with DMEM/0.1% BSA/gentamicin. The cells were pretreated, 46 h after transfection, with 2.5μg/ml cycloheximide, for 5 h. ER-Tracker Blue-White DPX Dye was diluted in DMEM/0.1% BSA, supplemented with 10 mm HEPES, pH 7.4, to a final concentration of 500 nm for 30 min at 37 C. Cells were washed once and then incubated for 10 min with WGA-AlexaFluor 633 (both indicator molecules from Molecular Probes, Eugene, OR) and diluted to a final concentration of 12.5 μg/ml. After approximately 10 min with the tracking stains, cells were imaged with a Zeiss LSM 710 confocal microscope (Carl Zeiss Microimaging, Inc., Thornwood, NY) using a 40×/1.20w objective. ER tracker was excited at 405 nm, and emission was detected in a broad spectral domain from 420- to 580-nm interval. GFP was excited at 488 nm and detected in the 500- to 570-nm interval, and Alexa 633 was excited at 633 nm and detected at 650–720 nm. GFP and Alexa 633 were imaged simultaneously; ER tracker was imaged sequentially to eliminate the possibility of bleed through into the GFP channel. Images of single confocal planes were contrast enhanced in Adobe Photoshop CS3 (Adobe Systems Inc., San Jose, CA).

Inositol phosphate (IP) assays

Cells were washed twice, 27 h after transfection, with 0.50 ml DMEM/0.1% BSA/20 μg/ml gentamicin and then preloaded for 18 h with 0.25 ml of 4 μCi/ml myo-[2-3H(N)]-inositol in inositol free DMEM; cells were then washed twice with 0.30 ml DMEM (inositol free) containing 5 mm LiCl and treated for 2 h with 0.25 ml of a saturating concentration of Buserelin (10−7 m) in the same medium. Total IP was determined (36). This assay has been validated as a sensitive measure of plasma membrane expression for functional receptors when expressed at low amounts of DNA and stimulated by excess agonist (10).

Binding assays (Scatchard and displacement by pharmacoperones)

Cells were cultured and plated in growth medium as described previously (10), except 105 cells in 0.5 ml growth medium were added to 24-well Costar cell culture plates (cell transfection and medium volumes were doubled accordingly). The medium was removed, 23 h after transfection, and replaced with 0.5 ml fresh growth medium. Cells were washed twice, 27 h after transfection, with 0.5 ml DMEM containing 0.1% BSA and 20 μg/ml gentamicin, after which 0.5 ml of DMEM was added. After 18 h, cells were washed twice with 0.5 ml DMEM/0.1% BSA/10 mm HEPES; then a range of concentrations of [125I]Buserelin prepared in our laboratory [specific activity is 700–800 μCi/μg; the range is from 1.25 × 105 to 4 × 106 cpm/ml for Scatchard (4)] in 0.5 ml of the same medium was added to the cells and allowed to incubate at room temperature for 90 min, consonant with maximum binding (10). New receptor synthesis during this period is negligible at room temperature. For binding displacement assays, 1.25 × 105 cpm/ml of [125I]Buserelin was used with a range of increasing concentrations of antagonist (pharmacoperone) and added to the cells. After 90 min, the media were removed and radioactivity was measured (10). To determine nonspecific binding, the same concentrations of radioligand were added to similarly transfected cells in the presence of 5 μg/ml unlabeled GnRH. Saturation binding curve fits and calculations (Bmax and Kd) were computed with Sigma Plot 8.02 (Jandel Scientific Software, Chicago, IL), using a nonlinear one-site binding model.

Statistics

Data (n ≥ 3) were analyzed with one-way ANOVA (37) and then paired with Student’s t test (SigmaStat 3.1; Jandel Scientific Software); P < 0.05 was considered significant.

Acknowledgments

We thank Darren Kafka for technical assistance and Jo Ann Binkerd for formatting the manuscript. A.P. thanks the University of Mississippi, Department of Medicinal Chemistry, and Professor Mitchell Avery for allowing him to pursue a summer internship in the Merck Research Laboratories, during which time molecular modeling work was completed.

Footnotes

This work was supported by National Institutes of Health Grants HD-19899, RR-00163, and HD-18185.Present address for A.P.: Pfizer Global Research & Development, Structural & Computational Chemistry, BB2C, Chesterfield, Missouri 63017. Present address for R.M.: Pharmasset, Inc., Princeton, New Jersey 08540.

Disclosure Statement: J.J. and A.C. have nothing to declare; A.P. and R.M. were previously employed by Merck Research Laboratories; M.T.G., M.D.A., and T.S.R. are presently employed by Merck Research Laboratories; P.M.C. is an inventor on U.S. Patent (pending) 10/492,295.

First Published Online December 18, 2008

Abbreviations: EL, Extracellular loop; ER, endoplasmic reticulum; GFP, green fluorescent protein; GnRHR, GnRH receptor; GPCR, G protein-coupled receptor; In3, indole 3; IP, inositol phosphate; QCS, quality control system; Q89, quinolone 89;THF, tetrahydrofuran; TM, transmembrane; WGA, wheat germ agglutinin; WT, wild type.

References

  1. Thompson MD, Percy ME, McIntyre Burnham W, Cole DE 2008 G protein-coupled receptors disrupted in human genetic disease. Methods Mol Biol 448:109–137 [DOI] [PubMed] [Google Scholar]
  2. Brauner-Osborne H, Wellendorph P, Jensen AA 2007 Structure, pharmacology and therapeutic prospects of family C G-protein coupled receptors. Curr Drug Targets 8:169–184 [DOI] [PubMed] [Google Scholar]
  3. Landry Y, Gies J-P 2008 Drugs and their molecular targets: an updated overview. Fund Clin Pharmacol 22:1–18 [DOI] [PubMed] [Google Scholar]
  4. Taborqa M, Corcoran KE, Fernandes N, Ramkissoon SH, Rameshwar P 2007 G-coupled protein receptors and breast cancer progression: potential drug targets. Mini Rev Med Chem 7:245–251 [DOI] [PubMed] [Google Scholar]
  5. Castro-Fernandez C, Maya-Nunez G, Conn PM 2005 Beyond the signal sequence: protein routing in health and disease. Endocr Rev 26:479–503 [DOI] [PubMed] [Google Scholar]
  6. Morello J-P, Bouvier M, Petaja-Repo UE, Bichet DG 2000 Pharmacological chaperones: a new twist on receptor folding. Trends Pharmacol Sci 21:466–469 [DOI] [PubMed] [Google Scholar]
  7. Brothers SP, Cornea A, Janovick JA, Conn PM 2004 Human loss-of-function gonadotropin-releasing hormone receptor mutants retain wild-type receptors in the endoplasmic reticulum: molecular basis of the dominant-negative effect. Mol Endocrinol 18:1787–1797 [DOI] [PubMed] [Google Scholar]
  8. Janovick JA, Goulet M, Bush E, Greer J, Wettlauffer DG, Conn PM 2003 Structure-activity relations of successful pharmacologic chaperones for rescue of naturally occurring and manufactured mutants of the gonadotropin-releasing hormone receptor. J Pharmacol Exp Ther 305:608–614 [DOI] [PubMed] [Google Scholar]
  9. Leaños-Miranda A, Ulloa-Aguirre A, Janovick JA, Conn PM 2005 In vitro coexpression and pharmacological rescue of mutant gonadotropin-releasing hormone receptors causing hypogonadotropic hypogonadism in humans expressing compound heterozygous alleles. J Clin Endocrinol Metab 90:3001–3008 [DOI] [PubMed] [Google Scholar]
  10. Janovick JA, Knollman PE, Brothers SP, Ayala-Yanez R, Aziz AS, Conn PM 2006 Regulation of G protein-coupled receptor trafficking by inefficient plasma membrane expression: molecular basis of an evolved strategy. J Biol Chem 281:8417–8425 [DOI] [PubMed] [Google Scholar]
  11. Knollman PE, Janovick JA, Brothers SP, Conn PM 2005 Parallel regulation of membrane trafficking and dominant-negative effects by misrouted gonadotropin-releasing hormone receptor mutants. J Biol Chem 280:24506–24514 [DOI] [PubMed] [Google Scholar]
  12. Janovick JA, Maya-Nunez G, Conn PM 2002 Rescue of hypogonadotropic hypogonadism-causing and manufactured GnRH receptor mutants by a specific protein-folding template: misrouted proteins as a novel disease etiology and therapeutic target. J Clin Endocrinol Metab 87:3255–3262 [DOI] [PubMed] [Google Scholar]
  13. Zhou W, Rodie V, Kitanovic S, Flanagan CA, Chi L, Weinstein H, Maayani, S, Millar RP, Sealfon SC 1995 A locus of the gonadotropin-releasing hormone receptor that differentiates agonist and antagonist binding sites. J Biol Chem 270:18853–18857 [DOI] [PubMed] [Google Scholar]
  14. Cui J, Smith RG, Mount GR, Lo JL, Yu J, Walsh TF, Singh SB, DeVita RJ, Goulet MT, Schaeffer JM, Cheng K 2000 Identification of Phe313 of the gonadotropin-releasing hormone (GnRH) receptor as a site critical for the binding of nonpeptide GnRH antagonists. Mol Endocrinol 14:671–681 [DOI] [PubMed] [Google Scholar]
  15. Betz SF, Reinhart GJ, Lio FM, Chen C, Struthers RS 2006 Overlapping, nonidentical binding sites of different classes of nonpeptide antagonists for the human gonadotropin-releasing hormone receptor. J Med Chem 49:637–647 [DOI] [PubMed] [Google Scholar]
  16. Janovick JA, Brothers SP, Cornea A, Bush E, Goulet MT, Ashton WT, Sauer DR, Haviv F, Greer J, Conn PM 2007 Refolding of misfolded mutant GPCR: post-translational pharmacoperone action in vitro. Mol Cell Endocrinol 272:77–85 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Flanagan CA, Rodic V, Konvicka K, Yuen T, Chi L, Rivier JE, Millar RP, Weinstein H, Sealfon SC 2000 Multiple interactions of the Asp(2.61(98)) side-chain of the gonadotropin-releasing hormone receptor contribute differentially to ligand interaction. Biochemistry 39:8133–8141 [DOI] [PubMed] [Google Scholar]
  18. Leanõs-Miranda A, Ulloa-Aguirre A, Ji TH, Janovick JA, Conn PM 2003 Dominant-negative action of disease-causing gonadotropin-releasing hormone receptor (GnRHR) mutants: a trait that potentially coevolved with decreased plasma membrane expression of GnRHR in humans. J Clin Endocrinol Metab 88:3360–3367 [DOI] [PubMed] [Google Scholar]
  19. Soderhall JA, Polymeropoulos EE, Paulini K, Gunther E, Kuhne R 2005 Antagonist and agonist binding models of the human gonadotropin-releasing hormone receptor. Biochem Biophys Res Commun 333:568–582 [DOI] [PubMed] [Google Scholar]
  20. Kerns E, Di L 2008 Drug-like properties: concepts, structure design and methods: from ADME to toxicity optimization. New York: Academic Press [Google Scholar]
  21. Snapp EL, Hegde RS, Francolini M, Lombardo F, Colombo S, Pedrazzini E, Borgese N, Lippincott-Schwartz J 2003 The organization of engaged and quiescent translocons in the endoplasmic reticulum of mammalian cells. J Cell Biol 163:257–269 [Google Scholar]
  22. Maya-Nuñez G, Janovick JA, Ulloa-Aguirre A, Söderlund D, Conn PM, Mendez JP 2002 Molecular basis of hypogonadotropic hypogonadism: restoration of mutant (E(90)K) GnRH receptor function by a deletion at a distant site. J Clin Endocrinol Metab 87:2144–2149 [DOI] [PubMed] [Google Scholar]
  23. Petaja-Repo UE, Hogue M, Laperrier A, Walker P, Bouvier M 2000 Export from the endoplasmic reticulum represents the limiting step in the maturation and cell surface expression of the human δ opioid receptor. J Biol Chem 275:13727–13736 [DOI] [PubMed] [Google Scholar]
  24. Conn PM, Knollman PE, Brothers SP, Janovick JA 2006 Protein folding as posttranslational regulation: evolution of a mechanism for controlled plasma membrane expression of a G protein-coupled receptor. Mol Endocrinol 20:3035–3041 [DOI] [PubMed] [Google Scholar]
  25. Vandenberghe W, Nicoll RA, Bredt DS 2005 Interaction with the unfolded protein response reveals a role for stargazin in biosynthetic AMPA receptor transport. J Neurosci 25:1095–1102 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Petrovska R, Kapa I, Klovins J, Schioth HB, Uhlen S 2005 Addition of a signal peptide sequence to the α1D-adrenoceptor gene increases the density of receptors, as determined by [3H]-prazosin binding in the membranes. Br J Pharmacol 144:651–659 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Uberti MA, Hague C, Oller H, Minneman KP, Hall RA 2004 Heterodimerization with β2-adrenergic receptors promotes surface expression and functional activity of α1D-adrenergic receptors. J Pharmacol Exp Ther 313:16–23 [DOI] [PubMed] [Google Scholar]
  28. Saito H, Kubota M, Roberts RW, Chi Q, Matsunami H 2004 RTP family members induce functional expression of mammalian odorant receptors. Cell 119:679–691 [DOI] [PubMed] [Google Scholar]
  29. Pietila EM, Tuusa JT, Apaja PM, Aatsinki JT, Hakalahti AE, Rajaniemi HJ, Petaja-Repo UE 2005 Inefficient maturation of the rat luteinizing hormone receptor. A putative way to regulate receptor numbers at the cell surface. J Biol Chem 280:26622–26629 [DOI] [PubMed] [Google Scholar]
  30. Conn PM, Ulloa-Aguirre A, Ito J, Janovick JA 2007 G protein-coupled receptor trafficking in health and disease: lessons learned to prepare for therapeutic mutant rescue in vivo. Pharmacol Rev 59:225–250 [DOI] [PubMed] [Google Scholar]
  31. Pey AL, Ying M, Cremades N, Velazquez-Campoy A, Scherer T, Thöny B, Sancho J, Martinez A 2008 Identification of pharmacological chaperones as potential therapeutic agents to treat phenylketonuria. J Clin Invest 118:2858–2867 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Bichet DG 2006 Nephrogenic diabetes insipidus. Adv Chronic Kidney Dis 13:96–104 [DOI] [PubMed] [Google Scholar]
  33. Crippen GM, Havel TF 1988 Distance geometry and molecular conformation. New York: Research Study Press, Ltd. [Google Scholar]
  34. Halgren TA 1999 MMFF VII. Characterization of MMFF94, MMFF94s, and other widely available force fields for conformational energies and for intermolecular-interaction energies and geometries. J Comp Chem 20:730–748 [DOI] [PubMed] [Google Scholar]
  35. Miller MD, Kearsley SK, Underwood DJ, Sheridan RP 1994 FLOG: a system to select ‘quasi-flexible’ ligands complementary to a receptor of known three-dimensional structure. J Comput Aid Mol Des 8:153–174 [DOI] [PubMed] [Google Scholar]
  36. Huckle W, Conn PM 1987 Use of lithium ion in measurement of stimulated pituitary inositol phospholipid turnover. Methods Enzymol 141:149–155 [DOI] [PubMed] [Google Scholar]
  37. Klotz IM 1982 Numbers of receptor sites from Scatchard graphs: facts and fantasies. Science 217:1247–1249 [DOI] [PubMed] [Google Scholar]

Articles from Molecular Endocrinology are provided here courtesy of The Endocrine Society

RESOURCES