Mammalian Type I Gonadotropin-Releasing Hormone Receptors ...

0013-7227/08/$15.00/0 Printed in U.S.A.

Endocrinology 149(3):1415–1422 Copyright © 2008 by The Endocrine Society doi: 10.1210/en.2007-1159

Mammalian Type I Gonadotropin-Releasing Hormone Receptors Undergo Slow, Constitutive, AgonistIndependent Internalization Adam J. Pawson, Elena Faccenda, Stuart Maudsley, Zhi-Liang Lu, Zvi Naor, and Robert P. Millar Medical Research Council Human Reproductive Sciences Unit, Edinburgh EH16 4TJ, United Kingdom Regulatory elements present in the cytoplasmic carboxyl-terminal tails of G protein-coupled receptors contribute to agonist-dependent receptor desensitization, internalization, and association with accessory proteins such as ␤-arrestin. The mammalian type I GnRH receptors are unique among the rhodopsin-like G protein-coupled receptors because they lack a cytoplasmic carboxyl-terminal tail. In addition, they do not recruit ␤-arrestin, nor do they undergo rapid desensitization. By measuring the internalization of labeled GnRH agonists, previous studies have reported that mammalian type I GnRH receptors undergo slow agonist-dependent internalization. In the present study, we have measured the internalization of epitope-tagged GnRH receptors, both in the absence and presence of GnRH stimulation. We demonstrate that mammalian type I GnRH receptors exhibit a low level of constitutive agonist-independent internalization. Stimulation with GnRH agonist did not significantly enhance the level of receptor internalization above the constitutive level. In contrast, the catfish GnRH and rat TRH receptors, which have cytoplasmic

T

HE SEMINAL STUDY on GnRH internalization was published by E. Hazum, P. M. Conn, and colleagues (1) over 25 yr ago. This important study demonstrated internalization of fluorescently labeled GnRH from the cell surface to intracellular sites when preincubated at 4 C with rat pituitary cells and then warmed to 37 C. Numerous studies have since followed and reported similar findings, by the use of fluorescently labeled, biochemically labeled, or radiolabeled GnRH analogs (2–21). The rate and extent to which labeled GnRH analogs have been reported to be internalized varies greatly depending on the cell type and GnRH receptor species and subtype. The internalization of GnRH into cells expressing nonmammalian GnRH receptors is generally rapid and quickly plateaus at a high level (50 – 80% of total bound labeled GnRH after 5–10 min) (21, 22). Similar findings have been reported for cells expressing the mammalian type II GnRH receptors (21– 23). In contrast, the internalization of labeled GnRH into cells expressing the cloned mammalian type I GnRH receptors is much slower and typically reaches a level of only around 25% First Published Online November 26, 2007 Abbreviations: FCS, Fetal calf serum; FITC, fluorescein isothiocyanate; GFP, green fluorescent protein; GPCR, G protein-coupled receptor; HA, hemagglutinin; ICC, immunocytochemistry; ICCB, ICC block solution; LPA, lysophosphatidic acid. Endocrinology is published monthly by The Endocrine Society (http:// www.endo-society.org), the foremost professional society serving the endocrine community.

carboxyl-terminal tails, displayed similar levels of constitutive agonist-independent internalization but underwent robust agonist-dependent internalization, as did chimeras of the mammalian type I GnRH receptor with the cytoplasmic carboxyl-terminal tails of the catfish GnRH receptor or the rat TRH receptor. When the carboxyl-terminal Tyr325 and Leu328 residues of the mammalian type I GnRH receptor were replaced with alanines, these two mutant receptors underwent significantly impaired internalization, suggesting a function for the Tyr-X-X-Leu sequence in mediating the constitutive agonist-independent internalization of mammalian type I GnRH receptors. These findings provide further support for the underlying notion that the absence of the cytoplasmic carboxyl-terminal tail of the mammalian type I GnRH receptors has been selected for during evolution to prevent rapid receptor desensitization and internalization to allow protracted GnRH signaling in mammals. (Endocrinology 149: 1415–1422, 2008)

of the total bound labeled GnRH after 30 min (21, 22). The mammalian type I GnRH receptors are unique among the rhodopsin-like G protein-coupled receptor (GPCR) superfamily, because they completely lack a cytoplasmic carboxylterminal tail (22, 24). In contrast to the mammalian type I GnRH receptors, but in common with other GPCRs, the cloned mammalian type II and nonmammalian GnRH receptors all have a cytoplasmic carboxyl-terminal tail (22). Because the cytoplasmic carboxyl-terminal tail of many GPCRs (25) and tailed GnRH receptors (11, 21, 22, 26) has been implicated in the regulation of their rapid agonistdependent internalization, it has been suggested that due to the lack of a cytoplasmic carboxyl-terminal tail, the mammalian type I GnRH receptors can be considered as natural internalization-defective mutants (11). The above internalization studies have monitored the internalization of labeled GnRH, typically involving a preincubation of cells expressing GnRH receptors with the labeled GnRH agonists at 4 C, followed by warming the cells to 37 C and then expressing the amount of internalized labeled GnRH agonist as a percentage of total bound labeled GnRH agonist. Therefore, these studies do not directly monitor internalization of the GnRH receptor but rather the internalization of labeled GnRH agonist as a surrogate marker for the GnRH receptor. Because many of these studies used [125I]GnRH agonist, it was not possible to assess constitutive agonist-independent internalization of the unoccupied receptor in the absence of GnRH. The present study reports the

1415

1416

Endocrinology, March 2008, 149(3):1415–1422

direct measurement of mammalian type I GnRH receptor internalization both in the presence and absence of unlabeled GnRH analogs. Our observations show that the rat type I GnRH receptor exhibits a low level of constitutive agonistindependent internalization. Stimulation with GnRH agonist did not change the level of receptor internalization above the constitutive level. In contrast, we demonstrated that the catfish GnRH receptor and rat TRH receptor, which have cytoplasmic carboxyl-terminal tails, displayed similar levels of constitutive agonist-independent internalization but underwent robust agonist-dependent internalization after exposure to their respective agonist ligands, as did chimeras of the rat type I GnRH receptor with the cytoplasmic carboxylterminal tails of the catfish GnRH receptor or the rat TRH receptor. Materials and Methods cDNA constructs, cell culture, and transient transfection The hemagglutinin (HA)-tagged human GnRH receptor and HAtagged Tyr325Ala and HA-tagged Leu328Ala human GnRH receptors were generated by a PCR method and subcloned into pcDNA3. The HA-tagged rat GnRH receptor, HA-tagged rat GnRH receptor/catfish GnRH receptor-C-tail, HA-tagged rat GnRH receptor/TRH receptor-Ctail, HA-tagged catfish GnRH receptor, and HA-tagged rat TRH receptor (8, 18, 27) were generated in our unit as previously described. Plasmid DNA for transient transfection was prepared using Maxi-Prep columns (QIAGEN, Valencia, CA) according to the manufacturer’s instructions. COS-7 cells were cultured as previously described (13, 28, 29), and transiently transfected using the SuperFect (QIAGEN) method according to manufacturer’s instructions. Wild-type HEK293 cells and HEK293 stable cell lines were generated and cultured as previously described (8, 18, 27, 30).

Receptor-binding assays Whole-cell receptor binding assays used the [125I]His5, d-Tyr6 GnRH analog (31). Transiently transfected COS-7 cells or HEK293 stable cell lines in 12-well culture plates were washed once with ice-cold HEPES/ DMEM/10% fetal calf serum (FCS) and incubated for 5 h on ice in HEPES/DMEM/10% FCS with 105 cpm/well [125I]His5, d-Tyr6 GnRH and varying concentrations of unlabeled His5, d-Tyr6 GnRH. Cell monolayers were rapidly washed twice in ice-cold PBS and solubilized in 0.1 m NaOH, which was counted using a ␥-counter to determine the amount bound radioligand. Nonspecific binding, consistently found to be less than 10% of total binding, was determined using vector-transfected cells (or wild-type HEK293 cells) and was subtracted from total binding to give specific binding.

Total inositol phosphate assays GnRH stimulation of total inositol phosphate production was as described (32). Briefly, transiently transfected COS-7 cells were incubated with inositol-free DMEM containing 1% dialyzed heat-inactivated FCS and 0.5 ␮Ci/well myo-[3H]inositol (Amersham Pharmacia Biotech, Piscataway, NJ) for 48 h. Medium was removed and the cells washed with 1 ml buffer (140 mm NaCl, 20 mm HEPES, 4 mm KCl, 8 mm glucose, 1 mm MgCl2, 1 mm CaCl2, 1 mg/ml BSA) containing 10 mm LiCl and incubated for 1 h at 37 C in 0.5 ml buffer containing 10 mm LiCl and GnRH agonist at the required concentration. Reactions were terminated by the removal of agonist and the addition of 1 ml ice-cold 10 mm formic acid, which was incubated for 30 min at 4 C. Total [3H]inositol phosphate was separated from the formic acid cell extracts on AG 1-X8 anion exchange resin (Bio-Rad, Hercules, CA) and eluted with a 1 m ammonium formate/0.1 m formic acid solution. The associated radioactivity was determined by liquid scintillation counting.

Pawson et al. • Mammalian Type I GnRH Receptor Internalization

[125I]GnRH analog internalization assays (acid/salt wash procedure) Receptor-mediated internalization of [125I]His5, d-Tyr6 GnRH and [125I]cetrorelix was determined by the acid/salt wash method as previously described (13, 29). Briefly, transiently transfected COS-7 cells and HEK293 cell lines in 12-well culture plates (poly-l-lysine coated for HEK293 cells) were washed once with ice-cold HEPES/DMEM/10% FCS and then incubated on ice in HEPES/DMEM/10% FCS with 105 cpm/well [125I]His5, d-Tyr6 GnRH for 5 h or [125I]cetrorelix for 3 h. Cells were rapidly warmed to 37 C for the indicated time points, and internalization was then stopped by placing the cells on ice and rapidly washing twice in 1 ml ice-cold PBS. Acid-sensitive bound radioligand, representing cell surface bound label, was removed by the addition of 1 ml ice-cold acid/salt solution (50 mm acetic acid, 150 mm NaCl, pH 2.8) for 12 min. After removal of the acid/salt wash, cells were solubilized with 1 ml 0.1 m NaOH to determine acid-resistant (internalized) radioligand content. The amount of radioligand remaining at the cell surface after warming to 37 C was expressed as the percentage of total cellassociated radioligand (acid-sensitive plus acid-resistant) at each time point.

Receptor internalization assay Cells were seeded into 12-well plates (4 ⫻ 105 cells per well) in growth medium and incubated overnight at 37 C in 5% CO2. After removal of growth medium, nonspecific cell surface antigens were blocked with 500 ␮l immunocytochemistry (ICC) block solution (ICCB: PBS, 10% FCS, 1% BSA) for 1 h at 4 C. This was replaced by 250 ␮l anti-HA antibody solution (mouse monoclonal clone 12CA5, 2 ␮g/ml in ICCB; Roche, Indianapolis, IN) at 4 C for 3–5 h. The cells were then washed three times in ICCB at 4 C, and 500 ␮l 1 ␮m ligand or ICCB was added and incubated at 4 C for 1 h. Plates were either kept at 4 C or moved to a 37 C CO2 incubator for the indicated time points. Plates were transferred to ice and fixed in 500 ␮l 4% paraformaldehyde fixing solution for 10 min at room temperature, followed by three ICCB washes. Antimouse [125I]Ig (1:200 dilution in ICCB of antimouse [125I]Ig from Amersham; 5–20 ␮Ci/mg protein, 100 ␮Ci/ml) was added and allowed to incubate for 3–5 h at room temperature, after which the cells were again washed three times in ICCB. Finally, the cells were solubilized in 500 ␮l 0.1 m NaOH and transferred to tubes, and associated radioactivity was counted for 1 min on a Perkin-Elmer (Norwalk, CT) Wizard 1470 automatic ␥-counter. Receptors remaining at the cell surface after warming to 37 C either in the absence or presence of ligand were expressed as percentage of receptors at the cell surface at the zero time point. COS-7 cells transiently transfected with pcDNA3 or wild-type HEK293 cells were used to determine nonspecific cell-associated radioactivity.

Confocal laser microscopy HEK293 stably cells expressing an HA-tagged rat GnRH receptor were seeded into eight-well chamber slides (5 ⫻ 104 cells per chamber) in growth medium and incubated overnight at 37 C in 5% CO2. To visualize the cell surface pool of HA-tagged receptors, growth medium was removed, and nonspecific cell surface antigens were blocked with 500 ␮l ICCB (PBS, 10% FCS, 1% BSA) for 1 h at 4 C. This was replaced by 250 ␮l anti-HA antibody solution (mouse monoclonal clone 12CA5, 2 ␮g/ml in ICCB; Roche) at 4 C for 2 h. The cells were then washed three times in ICCB at 4 C, and 500 ␮l 1 ␮m ligand or ICCB was added and incubated at 4 C for 1 h. Plates were either kept at 4 C or moved to a 37 C CO2 incubator for 1 h. Plates were transferred to ice. Cells were then fixed with 500 ␮l of 4% paraformaldehyde fixing solution for 10 min at room temperature, followed by three ICCB washes. Cells were then stained with secondary fluorescein isothiocyanate (FITC)-conjugated antimouse antibody for 2 h at room temperature or overnight and washed three times in ICCB, and slides were mounted for visualization on a Zeiss LSM510 laser scanning microscope. To visualize the total cellular pool of HA-tagged receptors, cells were first washed three times in ICCB at 4 C, and 500 ␮l 1 ␮m ligand or ICCB was added and incubated at 4 C for 1 h. Plates were either kept at 4 C or moved to a 37 C CO2 incubator for 1 h. Plates were transferred to ice. Cells were then fixed with 500 ␮l 4% paraformaldehyde fixing solution for 10 min at room temperature, followed by three ICCB washes. Cells were then perme-



1417

abilized in 500 ␮l ICCB/0.1% Nonidet P-40 where indicated and then stained with FITC-conjugated 12CA5 for 2 h at room temperature or overnight and washed three times in ICCB. Slides were mounted for visualization on a Zeiss LSM510 laser scanning microscope.

Data analysis Data are expressed as mean ⫾ se. Data were plotted and analyzed with Graph-Pad Prism software. Where appropriate, two-way ANOVA was performed with Bonferroni post-tests (with receptor as main factor and time as repeated measure).

Results Internalization of [125I]GnRH analogs into COS-7 and HEK293 cells expressing GnRH receptors

As in many previous studies, we measured the internalization of a radiolabeled GnRH agonist analog ([125I]His5, d-Tyr6 GnRH) into either transiently transfected COS-7 cells (Fig. 1A) or HEK293 stable cell lines (Fig. 1B) expressing various HA-tagged GnRH receptor species and chimeras, using the acid/salt wash procedure. The amount of [125I]His5, d-Tyr6 GnRH (prebound at 4 C) remaining at the cell surface after warming to 37 C was expressed as percentage of total cell-associated radioligand (acid-sensitive plus acid-resistant) at each time point. In both cell types, those expressing mammalian type I GnRH receptors (human and rat) apparently internalized [125I]His5, d-Tyr6 GnRH to a lesser extent (⬃25% loss of surface radiolabel) than cells expressing the catfish GnRH receptor or chicken GnRH receptor, as reported previously (13, 29), which both have a cytoplasmic carboxyl-terminal tail or chimeric tailed mammalian GnRH receptors (Fig. 1, A–C). These results suggest that tailed GnRH receptors are able to mediate the internalization of GnRH agonist more rapidly and to a greater extent than the tailless mammalian type I GnRH receptors. When we compared the internalization of [125I]His5, d-Tyr6 GnRH with that of the GnRH antagonist analog [125I]cetrorelix into transfected COS-7 cells transiently expressing either the human GnRH receptor or the chicken GnRH receptor, we observed that both the human and chicken GnRH receptors appeared to internalize the radiolabeled antagonist to the same lesser extent (Fig. 1C). The level of internalization of [125I]cetrorelix (⬃25% loss of surface radiolabel) was similar to internalization of [125I]His5, d-Tyr6 GnRH by cells expressing the human GnRH receptor (Fig. 1, A–C), the rat GnRH receptor (Fig. 1, A and B), and indeed a cytoplasmic carboxyl-terminal tail truncated chicken GnRH receptor mutant that we have previously reported (13, 29). These results suggest that the approximately 25% loss of surface radiolabel observed when measuring apparent agonist-dependent internalization of radiolabeled GnRH agonists into cells expressing the tailless mammalian type I GnRH receptors is instead constitutive agonist-independent internalization of these receptors from the cell surface (marked as X in Fig. 1C). Furthermore, only the tailed GnRH receptors exhibit a component of agonist-dependent internalization (marked as Y in Fig. 1C) above the constitutive component, which is similar to the constitutive agonist-independent internalization of the tailless mammalian type I GnRH receptors (marked as X in Fig. 1C). Clearly, a more direct approach to monitor receptor internalization in the absence of GnRH is required to substantiate these data.

FIG. 1. Internalization of [125I]labeled GnRH agonist and antagonist. Transiently transfected COS-7 cells (A and C) or stably transfected HEK293 cells (B) were preincubated with the radiolabeled agonist analog [125I]His5, D-Tyr6 GnRH or the radiolabeled antagonist analog [125I]cetrorelix and then warmed to 37 C for the indicated time points. Internalized radiolabeled agonist or antagonist was determined by the acid/salt wash procedure. Radioligand remaining at the cell surface after warming to 37 C was expressed as percentage of total cell-associated radioligand at each time point. The results shown are mean ⫾ SE of three experiments performed in triplicate. WT, Wild type.

Direct detection of epitope-tagged GnRH receptors at the cell surface

We first determined whether we could measure HAtagged receptors expressed at the surface of COS-7 (Fig. 2A) and HEK293 (Fig. 2B) cells using mouse anti-HA antisera (12CA5) as a primary detection, followed by fixing and secondary detection with antimouse [125I]Ig. The cell surface expression of all receptors was detectable and largely mirrored the expression levels when measured by a standard receptor binding assay with [125I]His5, d-Tyr6 GnRH (data not shown). To address the possibility that the mouse anti-HA antisera could interfere with the binding of GnRH to

1418


FIG. 2. Detection of cell surface HA-tagged receptors. A and B, Transiently transfected COS-7 cells (A) or stably transfected HEK293 cells (B) were preincubated with anti-HA antibody, washed, and fixed with 4% paraformaldehyde and then incubated with antimouse [125I]Ig, washed, and solubilized in 0.1 M NaOH, and radioactivity was measured. C and D, Transiently transfected COS-7 cells expressing the rat type I GnRH receptor were subjected to standard dose-response receptor binding (C) and inositol phosphate (D) assays in the absence and presence of anti-HA antibody. The results shown are mean ⫾ SE of three experiments performed in triplicate (A–C) and a representative experiment performed in triplicate (D). WT, Wild type.

the GnRH receptor, we coincubated the antisera with [125I]His5, d-Tyr6 GnRH in a standard receptor binding assay (Fig. 2C) and demonstrated that no interference occurred. Furthermore, the antisera did not significantly interfere with the receptor activation by GnRH and signaling to inositol phosphate production (Fig. 2D). The HA tag and presence of 12CA5 have previously been shown not to interfere with TRH receptor ligand binding, receptor activation, and signaling to inositol phosphate production (33). Constitutive agonist-independent internalization of GnRH receptors

We next sought to measure the internalization of receptors from the cell surface to intracellular sites in the absence of ligand over a time-course incubation at 37 C after prelabeling of the cell surface pool of receptors with the antisera at 4 C. Whether expressed in COS-7 (Fig. 3A) or HEK293 (Fig. 3B) cells, all receptors underwent constitutive internalization to intracellular sites to a similar extent (⬃25% loss of surface antisera-labeled HA-tagged receptor) and did not appear to be dependent of the presence/absence of a cytoplasmic carboxyl-terminal tail. Similar constitutive internalization dynamics were observed for the TRH receptor (Fig. 3). To discount the 2- to 3-fold variation in cell surface expression level


FIG. 3. Constitutive agonist-independent internalization. Transiently transfected COS-7 cells (A) or stably transfected HEK293 cells (B) were preincubated with anti-HA antibody and then washed and warmed to 37 C for the indicated time points. Cells were washed and fixed in 4% paraformaldehyde and then incubated with antimouse [125I]Ig, washed, and solubilized in 0.1 M NaOH, and associated radioactivity was determined. Receptors remaining at the cell surface after warming to 37 C were expressed as percentage of receptors at the cell surface at the zero time point. The results shown are mean ⫾ SE of three experiments performed in triplicate. WT, Wild type.

(Fig. 2, A and B) and therefore allow a direct comparison between the different receptors, the data were normalized such that the receptors remaining at the cell surface after warming to 37 C were expressed as the percentage of receptors at the cell surface at the zero time point. Internalization in the absence and presence of unlabeled GnRH

Having established that all the receptors tested displayed similar constitutive internalization dynamics, we next measured their agonist-dependent internalization over a timecourse incubation at 37 C after prelabeling with the antisera at 4 C. When expressed in COS-7 cells (Fig. 4), only the tailed receptors exhibited significant agonist-dependent internalization. The rate and degree of GnRH-induced internalization by tailed GnRH receptors was similar to that of TRHinduced internalization of the rat TRH receptor (Fig. 4F) and was largely in agreement with previously published data (33, 34). The tailless mammalian type I GnRH receptors (human and rat) did not undergo any significant enhancement in the level of internalization above the constitutive level in the presence of GnRH (Fig. 4, A and B). Similar results were



1419

FIG. 5. Agonist-dependent internalization in HEK293 cells. Stably transfected HEK293 cells were preincubated with anti-HA antibody and then washed and warmed to 37 C for the indicated time points either in the absence or presence of 1 ␮M ligand. Cells were washed and fixed in 4% paraformaldehyde and then incubated with antimouse [125I]Ig, washed, and solubilized in 0.1 M NaOH, and associated radioactivity was determined. Receptors remaining at the cell surface after warming to 37 C either in the absence or presence of ligand were expressed as percentage of receptors at the cell surface at the zero time point. The results shown are mean ⫾ SE of three experiments performed in triplicate. WT, Wild type.

FIG. 4. Agonist-dependent internalization in COS-7 cells. Transiently transfected COS-7 cells were preincubated with anti-HA antibody and then washed and warmed to 37 C for the indicated time points either in the absence or presence of 1 ␮M ligand. Cells were washed and fixed in 4% paraformaldehyde and then incubated with antimouse [125I]Ig, washed, and solubilized in 0.1 M NaOH, and associated radioactivity was determined. Receptors remaining at the cell surface after warming to 37 C either in the absence or presence of ligand were expressed as percentage of receptors at the cell surface at the zero time point. The results shown are mean ⫾ SE of three experiments performed in triplicate. WT, Wild type.

obtained in HEK293 cells, because only the tailed receptors exhibited significant agonist-dependent internalization levels (Fig. 5). To test the possibility that these results were not specific to the natural GnRH (GnRH I), we stimulated COS-7 cells expressing the rat GnRH receptor (Fig. 4G) and the catfish GnRH receptor (Fig. 4F) with GnRH II and His5, d-Tyr6 GnRH and observed similar results. The enhanced level of agonist-dependent internalization observed here for the chimeras of the mammalian type I GnRH receptors with the cytoplasmic carboxyl-terminal tails of the catfish GnRH receptor was more marked when compared with the levels observed when using the acid/salt wash procedure (Fig. 1).

This may be due to the levels of sensitivity of the two internalization assays. When we visualized the distribution of rat type I GnRH receptors stably expressed in HEK293 cells by confocal microscopy, a low level of receptors was observed at the plasma membrane (Fig. 6, A and B), and there was no obvious reduction of prelabeled cell surface receptors after warming to 37 C for 1 h both in the absence or presence of GnRH stimulation (Fig. 6, B and C). In cells that were fixed and then permeabilized and labeled, we observed that the majority of receptors were intracellular and appeared to be localized to endosomal compartments (Fig. 6), and there was no visually obvious redistribution of receptors in response to warming to 37 C and GnRH stimulation for 1 h (Fig. 6, F and G). We next screened a panel of GnRH analogs (both agonists and antagonists) for their ability to induce GnRH-dependent internalization of the rat GnRH receptor when stably expressed in HEK293 cells (supplemental Fig. 1, published as supplemental data on The Endocrine Society’s Journals Online web site at http://endo.endojournals.org). No significant loss of cell surface receptors above the constitutive (nonstimulated) level was detected in the presence of any of the GnRH analogs tested. Furthermore, we included a stimulation with lysophosphatidic acid (LPA) (LPA receptors are expressed in HEK293 cells) to determine whether induction of agonist-dependent internalization of a conventional tailed GPCR would nonspecifically mobilize rat GnRH receptors (supplemental Fig. 1). LPA treatment did not result in an

1420


FIG. 6. Visualization of rat type I GnRH receptors in HEK293 cells. A–C, HEK293 cells stably expressing the HA-tagged rat type I GnRH receptor were incubated with anti-HA antibody and then warmed to 37 C for the indicated time points either in the absence or presence of 1 ␮M GnRH. Cells were washed and fixed in 4% paraformaldehyde, stained with FITC-conjugated antimouse antibody, and visualized on a Zeiss LSM510 laser scanning microscope. D–F, Cells were first warmed to 37 C for the indicated time points either in the absence or presence of 1 ␮M GnRH and then fixed in 4% paraformaldehyde, permeabilized in and then stained with FITC-conjugated anti-HA antibody, and visualized on a Zeiss LSM510 laser scanning microscope. NS, Nonstimulated.

increase in rat GnRH receptor internalization above the constitutive level. Constitutive agonist-independent internalization is regulated by the putative carboxyl-terminal Tyr-X-X-Leu motif

Tyr-X-X-Leu tyrosine-based motifs have been implicated in mediating GPCR internalization (35). In all the cloned mammalian type I GnRH receptors, there is a totally conserved Tyr-Phe-Ser-Leu sequence exposed at the carboxy terminus. When we mutated the Tyr325 and Leu328 residues of the human GnRH receptor to alanines and measured the internalization of the mutant receptors compared with wildtype, both the Tyr325Ala and Leu328Ala receptors were internalized to a significantly lesser extent that the wild-type receptor, implicating the Tyr-Phe-Ser-Leu motif in mediating the constitutive agonist-independent internalization of mammalian type I GnRH receptors (Fig. 7). Discussion

Previous attempts to monitor the trafficking of unoccupied mammalian type I GnRH receptors using a carboxyl-terminally tagged rat GnRH receptor-green fluorescent protein (GFP) chimera suggested that there was an agonist-dependent component for internalization (5). However, the rat GnRH receptor-GFP chimera included the catfish GnRH receptor carboxyl-terminal tail as a spacer between the rat GnRH receptor and GFP tag, which was a requirement for successful cell surface expression of the chimeric receptor (5), suggesting that the observed agonist dependency may have arisen from the presence of the nonmammalian GnRH receptor tail.


FIG. 7. Tyr-X-X-Leu motif mediates constitutive agonist-independent internalization. Transiently transfected COS-7 cells were preincubated with anti-HA antibody and then washed and warmed to 37 C for the indicated time points either in the absence or presence of 1 ␮M ligand. Cells were washed and fixed in 4% paraformaldehyde and then incubated with antimouse [125I]Ig, washed, and solubilized in 0.1 M NaOH, and associated radioactivity was determined. Receptors remaining at the cell surface after warming to 37 C either in the absence (⫺) or presence (⫹) of ligand were expressed as percentage of receptors at the cell surface at the zero time point. The results shown are mean ⫾ SE of three experiments performed in triplicate. Two-way ANOVA was performed with Bonferroni post-tests (receptor as main factor and time as repeated measure). *, P ⬍ 0.05; ***, P ⬍ 0.001. WT, Wild type.

We have investigated whether the constitutive agonistindependent internalization of mammalian type I GnRH receptors is a process that is intrinsically regulated by elements within the receptor or merely a consequence of normal plasma membrane turnover dynamics. It does not appear to be the latter because treatment with LPA did not further increase the internalization of rat type I GnRH receptors in our stable HEK293 cell line, which express LPA receptors above the constitutive basal level. Rather, the constitutive agonist-independent internalization does appear to be intrinsically regulated, because when the carboxyl-terminal Tyr325 and Leu328 residues of the human GnRH receptor were replaced with alanines, these two mutant receptors underwent significantly impaired constitutive internalization, suggesting a function for the totally conserved TyrPhe-Ser-Leu sequence in mediating the constitutive internalization of mammalian type I GnRH receptors. Interestingly, the Tyr-Phe-Ser-Leu sequence has previously been suggested to play an important role in mammalian type I GnRH receptor function (20, 35), and the data from these studies reinforce our present findings. It was reported that truncation of the carboxyl-terminal Phe-Ser-Leu sequence of the rat type I GnRH receptor caused a complete loss of measurable ligand binding and effector coupling, whereas truncation of the both the Ser-Leu sequence or the carboxylterminal Leu residue alone caused a reduced ligand-binding affinity but retained effector coupling similar to that of the wild-type receptor. All these mutant receptors were expressed at the plasma membrane, and both the Ser-Leu and Leu truncated receptors apparently exhibited [125I]GnRH agonist internalization similar to the wild-type receptor, although a comparison of the internalization of these mutant


receptors both in the absence and presence of GnRH agonist was not determined (20). The paper by Paing et al. (35) provides an important comparison with our study. A highly conserved Tyr-X-X-Leu motif was identified within the cytoplasmic carboxyl-terminal tail of the protease-activated receptor-1 GPCR. When the Tyr and Leu residues in this internal motif were mutated to alanines, the mutated receptor underwent markedly impaired agonist-dependent internalization. When the cytoplasmic carboxyl-terminal of proteaseactivated receptor-1 was truncated so that the Tyr-X-X-Leu motif was now exposed at the carboxy terminus, the truncated receptor failed to undergo agonist-dependent internalization but exhibited constitutive agonist-independent internalization, much like the wild-type mammalian type I GnRH receptor in the present study, which also has the Tyr-X-X-Leu motif exposed at the carboxy terminus. Furthermore, based on the study by Paing et al. (35), we would predict that mutating this internal motif in the context of the nonmammalian and chimeric GnRH receptors would result in reduced agonist-dependent internalization, with no effect on the level of constitutive agonist-independent internalization. Taken together, these and several additional lines of evidence support the findings of the present study that the mammalian type I GnRH receptors undergo constitutive agonist-independent internalization. First, the mammalian type I GnRH receptors do not possess cytoplasmic carboxylterminal tails. This domain had been demonstrated play a critical role in mediating agonist-dependent internalization of GPCRs (25). Second, the mammalian type I GnRH receptors do not recruit ␤-arrestin, nor do they internalize via a ␤-arrestin-dependent pathway, whereas addition of a carboxyl-terminal tail to mammalian type I GnRH receptors confers ␤-arrestin dependency (8, 18). ␤-Arrestin has a wellestablished role in mediating agonist-dependent internalization of GPCRs (25, 36). Third, truncation of the cytoplasmic carboxyl-terminal tails of the mammalian type II and nonmammalian GnRH receptors (13, 21–23, 29, 37–39), and of other GPCRs, reduces their agonist-dependent internalization and ␤-arrestin dependency (25, 40). Fourth, the majority of mammalian type I GnRH receptors are intracellular in the nonstimulated state, suggesting that there exists a population of receptors in motion under basal conditions in the absence of GnRH stimulation (Fig. 6, E and F), and a recent study has provided strong data to reinforce this notion (41). The coupling to different signaling pathways can be selectively activated by different ligands, a concept called “agonist trafficking of receptor signals” by Kenakin (42). There is now evidence that this concept can be extended to the GnRH receptor (43, 44). Our recent studies on the binding and signaling characteristics of the human GnRH receptor in reproductive tissue tumor cell lines demonstrated a distinctly different pharmacology of ligand selectivity and signaling from that elucidated in pituitary gonadotropes. Specifically, the human GnRH receptor preferentially binds GnRH I and couples to Gq/11 protein in pituitary cells to mediate activation of phospholipase-C␤ and Ca2⫹ mobilization. However, GnRH I and II and certain antagonists of the receptor-mediated Gq/11 protein are potent inhibitors of proliferation in reproductive tumor cell lines through activation


1421

of Gi (44, 45). We have called the phenomenon GnRH ligandinduced selective signaling (22, 44). There is a precedent for differential ligand-induced signaling vs. internalization properties of GPCR agonists based on studies with the ␮-opioid receptor and several opioid ligands. It was demonstrated that opioid agonists that exhibit similar signaling properties by activating the same G protein subtypes, but they differed markedly in their ability to induce agonist-dependent internalization of the ␮-opioid receptor (46, 47). We therefore investigated whether different GnRH analogs could selectively stabilize the GnRH receptor in a conformation that would preferentially recruit internalization machinery that would promote its agonist-dependent internalization above the constitutive level. When we tested an array of analogs, we found that none of those tested increased the internalization of rat type I GnRH receptors in our stable HEK293 cell line above the constitutive basal level. Coupled with the fact that mammalian type I GnRH receptors do not undergo rapid agonist-dependent desensitization (27, 48), our finding that they do not undergo rapid agonist-dependent internalization provides further support for the underlying notion that the absence of the cytoplasmic carboxyl-terminal tail may have been selected for during evolution to prevent rapid desensitization and internalization, to allow a protracted LH surge over several hours, which is required for ovulation in mammals. This contrasts with the nonmammalian species, which have tailed GnRH receptors and undergo rapid agonist-dependent desensitization (8, 27) and internalization (8, 13, 21, 22, 29, 37, 49) and have a substantially shorter LH surge of typically no more than 120 min in the case of the domestic chicken (50). Acknowledgments Received August 21, 2007. Accepted November 12, 2007. Address all correspondence and requests for reprints to: Dr. Adam J. Pawson, Medical Research Council Human Reproductive Sciences Unit, The Queen’s Medical Research Institute, 47 Little France Crescent, Edinburgh EH16 4TJ, United Kingdom. E-mail: [email protected]. Disclosure Statement: A.J.P., E.F., S.M., Z.-L.L., and Z.N. have nothing to disclose. R.P.M. consults for Ardana plc.

References 1. Hazum E, Cuatrecasas P, Marian J, Conn PM 1980 Receptor-mediated internalization of fluorescent gonadotropin-releasing hormone by pituitary gonadotropes. Proc Natl Acad Sci USA 77:6692– 6695 2. Childs GV, Hazum E, Amsterdam A, Limor R, Naor Z 1986 Cytochemical evidence for different routes of gonadotropin-releasing hormone processing by large gonadotropes and granulosa cells. Endocrinology 119:1329 –1338 3. Conn PM, Hazum E 1981 Luteinizing hormone release and gonadotropinreleasing hormone (GnRH) receptor internalization: independent actions of GnRH. Endocrinology 109:2040 –2045 4. Conn PM, Smith RG, Rogers DC 1981 Stimulation of pituitary gonadotropin release does not require internalization of gonadotropin-releasing hormone. J Biol Chem 256:1098 –1100 5. Cornea A, Janovick JA, Lin X, Conn PM 1999 Simultaneous and independent visualization of the gonadotropin-releasing hormone receptor and its ligand: evidence for independent processing and recycling in living cells. Endocrinology 140:4272– 4280 6. Gorospe WC, Conn PM 1987 Agents that decrease gonadotropin-releasing hormone (GnRH) receptor internalization do not inhibit GnRH-mediated gonadotrope desensitization. Endocrinology 120:222–229 7. Hazum E, Meidan R, Liscovitch M, Keinan D, Lindner HR, Koch Y 1983 Receptor-mediated internalization of LHRH antagonists by pituitary cells. Mol Cell Endocrinol 30:291–301 8. Heding A, Vrecl M, Hanyaloglu AC, Sellar R, Taylor PL, Eidne KA 2000 The rat gonadotropin-releasing hormone receptor internalizes via a ␤-arrestin-

1422

9.

10.

11. 12.

13.

14. 15. 16. 17. 18.

19. 20.

21. 22. 23.

24. 25. 26. 27.

28.

29.


independent, but dynamin-dependent, pathway: addition of a carboxyl-terminal tail confers ␤-arrestin dependency. Endocrinology 141:299 –306 Huckle WR, McArdle CA, Conn PM 1988 Differential sensitivity of agonistand antagonist-occupied gonadotropin-releasing hormone receptors to protein kinase C activators. A marker for receptor activation. J Biol Chem 263: 3296 –3302 Loumaye E, Wynn PC, Coy D, Catt KJ 1984 Receptor-binding properties of gonadotropin-releasing hormone derivatives. Prolonged receptor occupancy and cell-surface localization of a potent antagonist analog. J Biol Chem 259: 12663–12671 McArdle CA, Davidson JS, Willars GB 1999 The tail of the gonadotrophinreleasing hormone receptor: desensitization at, and distal to, G protein-coupled receptors. Mol Cell Endocrinol 151:129 –136 Morel G, Dihl F, Aubert ML, Dubois PM 1987 Binding and internalization of native gonadoliberin (GnRH) by anterior pituitary gonadotrophs of the rat. A quantitative autoradiographic study after cryoultramicrotomy. Cell Tissue Res 248:541–550 Pawson AJ, Katz A, Sun YM, Lopes J, Illing N, Millar RP, Davidson JS 1998 Contrasting internalization kinetics of human and chicken gonadotropin-releasing hormone receptors mediated by C-terminal tail. J Endocrinol 156:R9 – R12 Pelletier G, Dube D, Guy J, Seguin C, Lefebvre FA 1982 Binding and internalization of a luteinizing hormone-releasing hormone agonist by rat gonadotrophic cells. A radioautographic study. Endocrinology 111:1068 –1076 Schvartz I, Hazum E 1986 Wheat germ agglutinin behaves as a GnRH antagonist but induces gonadotrope desensitization. Biochem Biophys Res Commun 140:750 –757 Schvartz I, Hazum E 1987 Internalization and recycling of receptor-bound gonadotropin-releasing hormone agonist in pituitary gonadotropes. J Biol Chem 262:17046 –17050 Suarez-Quian CA, Wynn PC, Catt KJ 1986 Receptor-mediated endocytosis of GnRH analogs: differential processing of gold-labeled agonist and antagonist derivatives. J Steroid Biochem 24:183–192 Vrecl M, Anderson L, Hanyaloglu A, McGregor AM, Groarke AD, Milligan G, Taylor PL, Eidne KA 1998 Agonist-induced endocytosis and recycling of the gonadotropin-releasing hormone receptor: effect of ␤-arrestin on internalization kinetics. Mol Endocrinol 12:1818 –1829 Wynn PC, Suarez-Quian CA, Childs GV, Catt KJ 1986 Pituitary binding and internalization of radioiodinated gonadotropin-releasing hormone agonist and antagonist ligands in vitro and in vivo. Endocrinology 119:1852–1863 Brothers SP, Janovick JA, Maya-Nunez G, Cornea A, Han XB, Conn PM 2002 Conserved mammalian gonadotropin-releasing hormone receptor carboxyl terminal amino acids regulate ligand binding, effector coupling and internalization. Mol Cell Endocrinol 190:19 –27 Pawson AJ, McNeilly AS 2005 The pituitary effects of GnRH. Anim Reprod Sci 88:75–94 Millar RP, Lu ZL, Pawson AJ, Flanagan CA, Morgan K, Maudsley SR 2004 Gonadotropin-releasing hormone receptors. Endocr Rev 25:235–275 Ronacher K, Matsiliza N, Nkwanyana N, Pawson AJ, Adam T, Flanagan CA, Millar RP, Katz AA 2004 Serine residues 338 and 339 in the carboxyl-terminal tail of the type II gonadotropin-releasing hormone receptor are critical for ␤-arrestin independent internalization. Endocrinology 145:4480 – 4488 Sealfon SC, Weinstein H, Millar RP 1997 Molecular mechanisms of ligand interaction with the gonadotropin-releasing hormone receptor. Endocr Rev 18:180 –205 Ferguson SS 2001 Evolving concepts in G protein-coupled receptor endocytosis: the role in receptor desensitization and signaling. Pharmacol Rev 53:1–24 McArdle CA, Franklin J, Green L, Hislop JN 2002 Signalling, cycling and desensitisation of gonadotrophin-releasing hormone receptors. J Endocrinol 173:1–11 Willars GB, Heding A, Vrecl M, Sellar R, Blomenrohr M, Nahorski SR, Eidne KA 1999 Lack of a C-terminal tail in the mammalian gonadotropin-releasing hormone receptor confers resistance to agonist-dependent phosphorylation and rapid desensitization. J Biol Chem 274:30146 –30153 Pawson AJ, Maudsley S, Morgan K, Davidson L, Naor Z, Millar RP 2005 Inhibition of human type I gonadotropin-releasing hormone receptor (GnRHR) function by expression of a human type II GnRHR gene fragment. Endocrinology 146:2639 –2649 Pawson AJ, Maudsley SR, Lopes J, Katz AA, Sun YM, Davidson JS, Millar RP 2003 Multiple determinants for rapid agonist-induced internalization of a nonmammalian gonadotropin-releasing hormone receptor: a putative palmitoylation site and threonine doublet within the carboxyl-terminal tail are critical. Endocrinology 144:3860 –3871


30. Anderson L, McGregor A, Cook JV, Chilvers E, Eidne KA 1995 Rapid desensitization of GnRH-stimulated intracellular signalling events in ␣T3-1 and HEK-293 cells expressing the GnRH receptor. Endocrinology 136:5228 –5231 31. Flanagan CA, Fromme BJ, Davidson JS, Millar RP 1998 A high affinity gonadotropin-releasing hormone (GnRH) tracer, radioiodinated at position 6, facilitates analysis of mutant GnRH receptors. Endocrinology 139:4115– 4119 32. Berg KA, Clarke WP, Sailstad C, Saltzman A, Maayani S 1994 Signal transduction differences between 5-hydroxytryptamine type 2A and type 2C receptor systems. Mol Pharmacol 46:477– 484 33. Petrou C, Chen L, Tashjian Jr AH 1997 A receptor-G protein coupling-independent step in the internalization of the thyrotropin-releasing hormone receptor. J Biol Chem 272:2326 –2333 34. Hanyaloglu AC, Vrecl M, Kroeger KM, Miles LE, Qian H, Thomas WG, Eidne KA 2001 Casein kinase II sites in the intracellular C-terminal domain of the thyrotropin-releasing hormone receptor and chimeric gonadotropin-releasing hormone receptors contribute to ␤-arrestin-dependent internalization. J Biol Chem 276:18066 –18074 35. Paing MM, Temple BR, Trejo J 2004 A tyrosine-based sorting signal regulates intracellular trafficking of protease-activated receptor-1: multiple regulatory mechanisms for agonist-induced G protein-coupled receptor internalization. J Biol Chem 279:21938 –21947 36. Ferguson SS, Downey 3rd WE, Colapietro AM, Barak LS, Menard L, Caron MG 1996 Role of ␤-arrestin in mediating agonist-promoted G protein-coupled receptor internalization. Science 271:363–366 37. Blomenrohr M, Heding A, Sellar R, Leurs R, Bogerd J, Eidne KA, Willars GB 1999 Pivotal role for the cytoplasmic carboxyl-terminal tail of a nonmammalian gonadotropin-releasing hormone receptor in cell surface expression, ligand binding, and receptor phosphorylation and internalization. Mol Pharmacol 56:1229 –1237 38. Hislop JN, Everest HM, Flynn A, Harding T, Uney JB, Troskie BE, Millar RP, McArdle CA 2001 Differential internalisation of mammalian and non-mammalian gonadotropin-releasing hormone receptors: uncoupling of dynamindependent internalisation from mitogen-activated protein kinase signalling. J Biol Chem 276:39685–39694 39. Hislop JN, Madziva MT, Everest HM, Harding T, Uney JB, Willars GB, Millar RP, Troskie BE, Davidson JS, McArdle CA 2000 Desensitization and internalization of human and Xenopus gonadotropin-releasing hormone receptors expressed in ␣T4 pituitary cells using recombinant adenovirus. Endocrinology 141:4564 – 4575 40. Gurevich VV, Gurevich EV 2006 The structural basis of arrestin-mediated regulation of G-protein-coupled receptors. Pharmacol Ther 110:465–502 41. Sedgley KR, Finch AR, Caunt CJ, McArdle CA 2006 Intracellular gonadotropin-releasing hormone receptors in breast cancer and gonadotrope lineage cells. J Endocrinol 191:625– 636 42. Kenakin T 1995 Agonist-receptor efficacy. II. Agonist trafficking of receptor signals. Trends Pharmacol Sci 16:232–338 43. Caunt CJ, Hislop JN, Kelly E, Matharu AL, Green LD, Sedgley KR, Finch AR, McArdle CA 2004 Regulation of gonadotropin-releasing hormone receptors by protein kinase C: inside out signalling and evidence for multiple active conformations. Endocrinology 145:3594 –3602 44. Millar RP, Pawson AJ 2004 Outside-in and inside-out signaling: the new concept that selectivity of ligand binding at the gonadotropin-releasing hormone receptor is modulated by the intracellular environment. Endocrinology 145:3590 –3593 45. Maudsley S, Davidson L, Pawson AJ, Chan R, de Maturana RL, Millar RP 2004 Gonadotropin-releasing hormone (GnRH) antagonists promote proapoptotic signaling in peripheral reproductive tumor cells by activating a G␣icoupling state of the type I GnRH receptor. Cancer Res 64:7533–7544 46. Keith DE, Anton B, Murray SR, Zaki PA, Chu PC, Lissin DV, MonteilletAgius G, Stewart PL, Evans CJ, von Zastrow M 1998 ␮-Opioid receptor internalization: opiate drugs have differential effects on a conserved endocytic mechanism in vitro and in the mammalian brain. Mol Pharmacol 53:377–384 47. Burford NT, Tolbert LM, Sadee W 1998 Specific G protein activation and ␮-opioid receptor internalization caused by morphine, DAMGO and endomorphin I. Eur J Pharmacol 342:123–126 48. Davidson JS, Wakefield IK, Millar RP 1994 Absence of rapid desensitization of the mouse gonadotropin-releasing hormone receptor. Biochem J 300:299 – 302 49. Acharjee S, Maiti K, Soh JM, Im WB, Seong JY, Kwon HB 2002 Differential desensitization and internalization of three different bullfrog gonadotropinreleasing hormone receptors. Mol Cells 14:101–107 50. Wilson SC, Sharp PJ 1975 Episodic release of luteinizing hormone in the domestic fowl. J Endocrinol 64:77– 86

Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.