| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Human Reproductive Sciences Unit, Medical Research Council (A.J.P., S.R.M., R.P.M.), Edinburgh, United Kingdom EH16 4SB; and Divisions of Chemical Pathology and Medical Biochemistry, University of Cape Town Medical School (A.J.P., J.L., A.A.K., Y.-M.S., J.S.D., R.P.M.), Cape Town 7925, South Africa
Address all correspondence and requests for reprints to: Prof. Robert P. Millar, Human Reproductive Sciences Unit, Medical Research Council, University of Edinburgh Chancellor’s Building, 49 Little France Crescent, Edinburgh, United Kingdom EH16 4SB. E-mail: r.millar{at}hrsu.mrc.ac.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1–81) mutants, effectively inhibited rapid agonist-induced internalization. Furthermore, cGnRH-Rs appeared to be mobilized to the ß-arrestin- and clathrin-coated, vesicle-mediated endocytic pathway upon ß-arrestin overexpression. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The mammalian GnRH-Rs are unique among the rhodopsin-like GPCR superfamily, because they completely lack a cytoplasmic carboxyl-terminal tail (2). In many GPCRs, this region has been demonstrated to play a crucial regulatory role in agonist-induced receptor phosphorylation, uncoupling, desensitization, internalization, and resensitization (3, 4, 5, 6, 7). In keeping with these observations, the mammalian GnRH-R does not show rapid desensitization (8). In contrast to the mammalian GnRH-Rs, but in common with other GPCRs, the cloned nonmammalian GnRH-Rs all have a carboxyl-terminal tail (2, 9, 10, 11, 12). We have previously demonstrated that the chicken GnRH-R (10) exhibits rapid internalization kinetics and requires the carboxyl-terminal tail for this process (13). Furthermore, it has been reported that the catfish (14), Xenopus laevis (15, 16), and bullfrog (17) GnRH-Rs also have enhanced internalization kinetics compared with the mammalian GnRH-Rs. The importance of the carboxyl-terminal tail of the catfish GnRH-R for cell surface expression, ligand binding, and receptor phosphorylation and internalization has also been shown (18, 19). Moreover, addition of the carboxyl-terminal tail of the catfish GnRH-R to the rat GnRH-R results in an increased level of cell surface expression (20), and introduction of the carboxyl-terminal tail of the rat TRH receptor at the carboxyl terminus of the rat GnRH-R led to increased internalization (14). Other studies have shown that the mammalian GnRH-Rs exhibit slow internalization kinetics (7, 14, 15, 16, 21). These studies thus suggest that the carboxyl-terminal tail of nonmammalian GnRH-Rs plays a pivotal role in their function and subcellular trafficking.
At least four pathways of agonist-induced internalization of GPCRs exist (22, 23), and they may be cell type specific. The classical GPCR internalization pathway involves GPCR kinases (GRKs), ß-arrestin, clathrin-coated pits, and the GTPase dynamin and is exemplified by the ß2-adrenergic receptor (3, 4, 5, 6, 7, 22, 23). The receptor is phosphorylated at specific serine and threonine residues by a GRK, followed by the binding of ß-arrestin, whereupon the receptor is targeted to clathrin-coated pits for internalization. Dynamin is thought to be necessary for the scission of clathrin-coated vesicles from the plasma membrane (24). GPCRs have also been reported to internalize independently of both ß-arrestin and dynamin or in pathways dependent on only one or the other, implying that GPCRs can undergo internalization via pathways distinct from clathrin-coated pits (3, 4, 5, 6, 22, 23). The localization of certain GPCRs to smooth noncoated membrane structures and vesicles (25, 26) suggests that GPCRs can use internalization pathways distinct from clathrin-coated vesicles. Caveolae are flask-shaped, nonclathrin-coated structures that have been implicated in the internalization of small molecules and certain GPCRs (27, 28, 29, 30, 31, 32, 33). In addition, dynamin has been implicated in caveolae function, although its exact functional role is not known. Dynamin may be responsible for the pinching off of caveolae from the plasma membrane (34, 35).
We have previously reported that the chicken GnRH receptor (cGnRH-R) undergoes rapid agonist-induced internalization and requires the carboxyl-terminal tail for this process (13). In this study we investigate the role of the carboxyl-terminal tail of the cGnRH-R in relation to its rapid internalization and identify the key residues and mechanisms involved. It is known that COS-7 cells endogenously express low levels of ß-arrestin, in contrast to the high levels present in HEK 293 cells (36). In this context we have sought to characterize the mechanisms involved in the rapid internalization of the cGnRH-R in COS-7 and HEK 293 cells by cotransfection of receptor constructs with wild-type and dominant-negative mutants of ß-arrestin, dynamin, and caveolin and by the inhibition of clathrin-mediated and caveolar endocytosis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cell culture and transient transfection
Plasmid DNA for transient transfection was prepared using Maxi-Prep columns (QIAGEN, Chatsworth, CA) according to the manufacturer’s instructions. COS-7 and HEK 293 cells were cultured as previously described for COS-1 cells (13) and transiently transfected using the SuperFect (QIAGEN) method according to the manufacturer’s instructions. Wild-type ß-arrestin-1 (ß-arr1) and (319–418)ß-arr1 cDNAs were provided by Dr. J. L. Benovic, wild-type dynamin-1 (dyn-1) and dominant-negative (K44A)dyn-1 cDNAs were provided by Dr. M. G. Caron, and the green fluorescence protein (GFP)-tagged, dominant-negative caveolin-1 cDNA, which includes the deletion of residues 1–81 [cav-1(1–81)], was provided by Dr. J. Eggermont. Pretreatments with monodansylcadavarine (MDC; Sigma-Aldrich Corp., St. Louis, MO), sucrose (Sigma-Aldrich Corp.), filipin (Sigma-Aldrich Corp.), and methyl-ß-cyclodextrin (mßCD; Sigma-Aldrich Corp.) were performed 37 C for 30 min before receptor internalization assays.
Receptor binding assays
Whole cell receptor binding assays used the 125I-[His5,D-Tyr6]GnRH analog (37). Transiently transfected COS-7 cells 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 of bound radioligand. Nonspecific binding, consistently found to be less than 10% of total binding, was determined using vector-transfected (pcDNA1/amp) COS-7 cells and was subtracted from total binding to give specific binding.
Total inositol phosphate assays
GnRH stimulation of total inositol phosphate production was previously described (38). Briefly, transiently transfected COS-7 cells were incubated with inositol-free DMEM containing 1% dialyzed heat-inactivated FCS and 0.5 µCi/well myo-[3H]inositol (Amersham Pharmacia Biotech, Piscataway, NJ) for 48 h. Medium was removed, and the cells were washed with 1 ml buffer (140 mM NaCl, 20 mM HEPES, 4 mM KCl, 8 mM glucose, 1 mM MgCl2, 1 mM CaCl2, and 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 indicated 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 phosphates were separated from the formic acid cell extracts on AG 1-X8 anion exchange resin (Bio-Rad Laboratories, 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.
Receptor internalization assays
Receptor-mediated internalization of 125I-[His5,D-Tyr6 ]GnRH was determined by the acid-wash method as previously described for COS-1 cells (13). Briefly, transiently transfected COS-7 and HEK 293 cells in 12-well culture plates (poly-L-lysine coated for HEK 293 cells) were washed once with ice-cold HEPES/DMEM/10% FCS, then incubated on ice in HEPES/DMEM/10% FCS with 105 cpm/well 125I-[His5,D-Tyr6]-GnRH for 5 h. Cells were rapidly warmed to 37 C for the indicated time periods, and internalization was 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 solution (50 mM acetic acid and 150 mM NaCl, pH 2.8) for 12 min. After removal of the acid wash, cells were solubilized with 1 ml 0.1 M NaOH to determine acid-resistant (internalized) radioligand content. Radioligand internalization was expressed as the percentage of total cell-associated label (acid-sensitive plus acid-resistant) at each time point.
Immunoprecipitation and Western blotting
GFP-tagged cav-1(1–81) protein was immunoprecipitated (39) from cell lysates by overnight incubation with GFP-agarose slurry (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) and washed. The immunoprecipitates were resolved by SDS-PAGE and electrotransferred to a polyvinylidene difluoride membrane (NEN Life Science Products, Boston, MA). Cav-1(1–81) protein was detected using antigoat GFP polyclonal antisera (Abcam, Cambridge, UK) and was visualized by enzyme-linked chemifluorescence (Amersham Pharmacia Biotech) and a PhosphorImager (Molecular Dynamics, Sunnyvale, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). All of the mutant receptors were expressed at the cell surface and generated an inositol phosphate response when exposed to GnRH agonist (Table 1). Progressive truncation of the carboxyl-terminal tail resulted in decreased levels of receptor expression at the cell surface (as measured by whole cell binding of 125I-[His5,D-Tyr6]GnRH), with maximal binding levels (Bmax) ranging from 77 ± 35% to 30 ± 10% of wild-type cGnRH-R levels. The T373A, T369T370AA, S366A, T362S363AA, and S337A mutant receptors were all expressed at the cell surface, with Bmax levels ranging from 86 ± 9% to 66 ± 23% of wild-type levels. The affinities (estimated using the 50% inhibitory concentration) of the mutant receptors were all within the nanomolar range and did not differ greatly from the wild-type receptor affinity (Table 1). With the exception of the S337stop (60 ± 12%), S337A (73 ± 8%), and T369T370AA (84 ± 19%) cGnRH-R mutants, agonist stimulation of all remaining mutant receptors resulted in maximal total inositol phosphate (IPmax) levels that were higher than those seen for the wild-type receptor (Table 1). The lower 50% inhibitory concentration of the S337A mutant may be due to partial uncoupling of the receptor from Gq/11, which is suggested by the low IPmax. In contrast, although the IPmax was equally low in the S337stop mutant, its affinity was the same as that of the wild-type receptor. The higher 50% effective concentration of the S337stop mutant is most likely due to the lower expression of this receptor (Bmax 
30% of wild-type).
|
|
). The relative levels of total inositol phosphate accumulation largely paralleled the maximal binding levels. The above data suggest that the length of the carboxyl-terminal tail of the cGnRH-R is important for efficient receptor expression at the cell surface, whereas specific serine and threonine residues located within the tail play a lesser role in expression. The decreased binding seen for the truncated mutants was due to decreased cell surface expression, rather than a perturbed ligand binding pocket caused by either truncation of the carboxyl-terminal tail or altered G protein coupling efficiencies, as the binding affinities (except for S337A) were unchanged. The decreased cell surface expression of the truncated mutants may be due to either general structural instability of the receptor, leading to a global reduction of both nascent and plasma membrane-associated receptor, or a reduction in specific cell surface expression.
Rapid agonist-induced internalization of cGnRH-R is mediated by a threonine-doublet at the carboxyl-terminal end of the cytoplasmic tail and a membrane-proximal cysteine residue
We have shown that the cGnRH-R undergoes rapid agonist-induced internalization and requires an intact carboxyl-terminal tail for this process (13). To more precisely localize the structural elements in the carboxyl terminus critical for the rapid internalization process, the cytoplasmic tail was progressively truncated (Fig. 1). The method used for measuring agonist-induced internalization discounts variations in receptor number due to decreased expression, because the data are presented as a percentage of the total agonist-bound receptors that are internalized at each time point measured. As the tail is progressively truncated, the mutant receptors undergo slower initial and decreased maximal levels of agonist-induced internalization (Fig. 2A). The most truncated S337stop mutant receptor exhibited a maximal steady state level of internalization only about 25% of that measured for the wild-type cGnRH-R after 90 min. This low level of internalization was similar to that of the wild-type human GnRH-R, which lacks a tail (13). There was little or no difference in both the initial and maximal levels of internalization of the S346stop, T351stop, and D356stop mutant receptors (internalization 70% of wild-type), suggesting that the region between Ile345 and Asp356 does not contain critical elements for the rapid internalization of the cGnRH-R. We therefore targeted putative serine and threonine phosphorylation sites flanking this region for mutation to alanine residues (Fig. 1) and measured the abilities of these single and double mutant receptors to undergo agonist-induced internalization. With the exception of the T369T370AA receptor (internalization 70% of wild-type), all of the remaining single and double mutants showed little or no difference in both the initial and maximal levels of internalization (Fig. 2B). These results suggest that the Thr369Thr370 doublet plays a critical role in the rapid agonist-induced internalization of the cGnRH-R. This conclusion is further strengthened by the finding that that the D356stop, T351stop, and S346stop mutant receptors all internalized at approximately the same maximum steady state level as the T369T370AA mutant, indicating that the threonine residues at positions 369 and 370 are critical for rapid internalization. Interestingly, the S366stop mutant receptor, which does not contain Thr373, Thr369Thr370, or Ser366 residues, exhibited a maximum steady state level of internalization only about 10% lower than that measured for the wild-type receptor. This result is unexpected and difficult to interpret. It is possible that truncating the tail to Gly365 (i.e. the S366stop mutant), which now places Thr362 and Ser363 close to the carboxyl terminus of the receptor, enables them to fulfill the important role of the Thr369Thr370 doublet near the carboxyl terminus of the wild-type receptor.
|
). Both mutant receptors were expressed at the cell surface and generated an inositol phosphate response when exposed to GnRH agonist (Table 1). The ability of the C328A mutant receptor to internalize was reduced to similar initial and maximal levels as those seen for the T369T370AA mutant (Figs. 2B and 3). This result suggests that Cys328 is additionally important to the rapid agonist-induced internalization process of the cGnRH-R. The approximately 20% component of internalization that remains for the S337stop truncation mutant (Fig. 2A) may be mediated by residues in the region upstream of Ser337, of which Cys328 is one. This approximately 20% remaining component is similar to the 20–25% inhibition of internalization observed for the C328A mutant receptor (Fig. 3). These results suggest that Cys328 is the most important residue in the tail region upstream of Ser337. Combination of the Cys328A and Thr369Thr370AA mutations led to a mutant receptor that internalized slightly lower than both the C328A or T369T370AA mutant receptors (Figs. 2A and 3). The fact that the creation of the double mutant did not produce an additive effect suggests distinct roles for Cys328 and the Thr369Thr370 doublet in the rapid agonist-induced internalization process. Interestingly, the C328AT369T370AA mutant receptor showed a 2.5-fold higher expression level at the cell surface compared with the wild-type receptor (Table 1). The higher IPmax value (IPmax 132% of wild type) of the C328AT369T370AA mutant is most likely due to the higher expression of this receptor (Bmax 250% of wild type).
|
, and 7A). To examine the effect of ß-arrestin on cGnRH-R internalization (Fig. 4), the cGnRH-R construct was cotransfected with vector (pcDNA1/amp) as a control, wild-type ß-arr1, dominant-negative (319–418)ß-arr1, or both ß-arr1 and (319–418)ß-arr1. Coexpression with ß-arr1 led to increased cGnRH-R internalization (Fig. 4, 
). This increase in internalization under conditions of ß-arr1 overexpression will henceforth be referred to as ß-arr1-promoted internalization.
|
|
|
|
, 
), indicating that the encoded protein was expressed and functional in the negative sense. Despite this, (319–418)ß-arr1 coexpression alone had virtually no effect on internalization (Fig. 4, 
), indicating that cGnRH-Rs are internalized via a ß-arr1-independent pathway in COS-7 cells unless ß-arr1 is overexpressed. This lack of effect is consistent with the reported low levels of endogenously expressed ß-arr1 in COS-7 cells (36).
Pretreatment with disrupters of clathrin-coated pit function, MDC (40) and hypertonic sucrose (41), failed to inhibit internalization (Fig. 5) even at later time points (data not shown), whereas MDC and sucrose pretreatments were effective at blocking ß-arr1-promoted internalization (Fig. 5). This is consistent with a role for clathrin-coated pit formation in the ß-arr1-dependent internalization pathway. MDC pretreatments showed similar effects when tested at concentrations from 50–400 µM for periods varying from 30–90 min (data not shown). The above results suggest that in COS-7 cells the wild-type cGnRH-R preferentially undergoes rapid agonist-induced internalization through a pathway that is independent of both ß-arr1 and clathrin-coated pits. However, under conditions of ß-arr1 overexpression, the cGnRH-R can be mobilized or recruited to internalize via a ß-arr1- and clathrin-mediated pathway.
The cGnRH-R expressed in COS-7 cells undergoes rapid agonist-induced internalization via a dynamin- and caveolae-mediated pathway
To examine the effect of dynamin on cGnRH-R internalization (Fig. 6), the cGnRH-R construct was cotransfected with vector (pcDNA1/amp) as a control, wild-type dynamin-1 (dyn-1), dominant-negative (K44A)dyn-1, or both wild-type ß-arr1 and (K44A)dyn-1. Coexpression with wild-type dyn-1 had no effect on internalization (data not shown), whereas coexpression with the dominant-negative (K44A)dyn-1 construct inhibited internalization (Fig. 6, ), suggesting the utilization of a dynamin-dependent pathway. Furthermore, the coexpression of receptors with both wild-type ß-arr1 and dominant-negative (K44A)dyn-1, revealed that ß-arr1-promoted internalization was blocked in the presence of coexpressed (K44A)dyn-1 (Fig. 6, ).
Pretreatment with filipin and mßCD, disrupters of caveolae formation (42), led to an inhibition of cGnRH-R internalization (Fig. 7A, , and Fig. 7B). In addition, filipin pretreatment was apparently able to effectively inhibit ß-arr1-promoted internalization (Fig. 7A, ). Filipin pretreatments were performed at concentrations from 0.5–5 µg/ml for time periods varying from 30–90 min with no alteration in the level of inhibition of internalization being evident (data not shown). The mßCD pretreatments were performed at 10 µM for 30 min. Caveolin-1 is the principal structural protein of caveolae membranes (43). Coexpression of a dominant-negative mutant [cav-1(1–81)] of caveolin-1, previously shown to disrupt the formation of caveolae lipid rafts (44), led to a significant inhibition of cGnRH-R internalization (Fig. 7D). We transiently transfected the GFP-tagged cav-1(1–81) construct into COS-7 cells and assessed its expression relative to mock-transfected cells by Western blot analysis with specific anti-GFP antisera (Fig. 7C). These results suggest that in COS-7 cells, the wild-type cGnRH-R preferentially undergoes rapid agonist-induced internalization in caveolae in a dynamin-dependent manner.
The cGnRH-R expressed in HEK 293 cells undergoes rapid agonist-induced internalization via a dynamin- and caveolae-mediated pathway
We were interested to compare the pathways of rapid agonist-induced wild-type cGnRH-Rs when expressed in COS-7 and HEK 293 cells, based on the significant differences in the background endogenous expression of ß-arr1 in these two cell lines (36). A 90-min course of internalization of transiently expressed wild-type receptor revealed that although the initial rates of internalization in the two cell types appeared to be very similar, the maximal steady state level was about 30% lower in HEK 293 cells than in COS-7 cells (Fig. 8A). This unexpected result was further complicated by our finding that pretreatment of wild-type cGnRH-R expressing HEK 293 cells with MDC or hypertonic sucrose had virtually no effect on internalization (Fig. 8B). However, coexpression with dominant-negative (319–418)ß-arr1 clearly retarded the initial rate of agonist-induced internalization, but had no effect on the maximal steady state level that wild-type cGnRH-Rs can reach in HEK 293 cells (Fig. 8B). In contrast, both coexpression with dominant-negative (K44A)dyn-1 and pretreatment with filipin effectively inhibited internalization of wild-type cGnRH-Rs expressed in HEK 293 cells, reducing both the initial rate and the maximal steady state level (Fig. 8C). These results suggest that despite the high background of endogenous ß-arr1 expression, the wild-type cGnRH-R preferentially undergoes rapid agonist-induced internalization in caveolae in a dynamin-dependent manner in HEK 293 cells.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To further characterize the internalization pathway of the cGnRH-R expressed in COS-7 cells, we employed the dominant-negative (K44A)dyn-1 mutant and disrupters of caveolar structure and function [filipin (42), mßCD (45), and a dominant-negative caveolin-1 mutant (44)]. Our data demonstrated that the cGnRH-R preferentially undergoes rapid agonist-induced internalization in a dynamin- and caveolae-dependent manner, based on the finding that internalization is inhibited in the presence of dominant-negative (K44A)dyn-1 and cav-1(1–81) overexpression, and pretreatments with filipin and mßCD.
We observed that under conditions of ß-arr1 overexpression, where cGnRH-Rs were additionally mobilized or recruited to the clathrin-coated pit-mediated internalization machinery, that filipin pretreatment was apparently effective at inhibiting ß-arr1-promoted internalization (Fig. 7A, ). It could be argued from the above result that the effect of filipin is nonspecific. However, this result more likely depicts the filipin-mediated inhibition of the caveolar component of internalization, and that the remaining fraction is, in fact, the filipin-resistant internalization of those receptors that have been recruited to the ß-arrestin-dependent pathway by ß-arr1 overexpression (i.e. the ß-arr1-promoted internalization component). Furthermore, coexpression with both wild-type ß-arr1 and dominant-negative (K44A)dyn-1 (Fig. 6, ) revealed that in addition to inhibition of the ß-arr1-promoted internalization component, (K44A)dyn-1 was effective at inhibiting the caveolar component as well. This result clearly implicates dynamin function in both clathrin- and caveolar-mediated internalization pathways.
Studies of the internalization pathways used by mammalian and nonmammalian GnRH-Rs show that the rat GnRH-R, which lacks a cytoplasmic tail, internalizes in a ß-arrestin-independent manner, but probably via a clathrin-dependent mechanism (21), and in a ß-arrestin-independent pathway that is dynamin dependent (46). Although these findings may appear to be inconsistent, it is cogent that a lack of a carboxyl-terminal domain in the mammalian GnRH-Rs probably confers their ß-arrestin independence, at least as far as internalization is concerned. Furthermore, it has been shown that the human GnRH-R undergoes internalization via a clathrin-mediated pathway, but in a dynamin-independent manner (16). A study characterizing agonist-induced internalization of the catfish GnRH-R (19) identified a serine residue in the carboxyl-terminal tail that is phosphorylated and may function as a ß-arrestin binding site. The conclusions of this study were based on the finding that substitution of this serine residue with alanine (or deleting the last 12 amino acids of the tail, which includes this serine) resulted in mutant receptors whose impaired internalization was not rescued by ß-arrestin overexpression in COS-7 cells (19). Furthermore, coexpression of the dominant-negative (319–418)ß-arrestin with the catfish GnRH-R in COS-7 cells produced a slight inhibition of the steady state level internalization (46). A recent study has characterized the internalization pathways of the three bullfrog GnRH-R subtypes (17). The bullfrog type II GnRH-R showed the most rapid rate and highest extent of internalization among the three receptors. Furthermore, internalization of the bullfrog type I GnRH-R was shown to be both ß-arrestin and dynamin-dependent, whereas bullfrog type II and III GnRH-Rs internalize via a pathway that is ß-arrestin independent, but dynamin dependent (17), similar to the pathway used by the chicken GnRH-R (present study). It is interesting to note that the last eight residues of the carboxyl-terminal tails of the chicken and bullfrog type II GnRH-Rs are surprisingly similar (GTTVNTVC for chicken, ATTVQSVF for bullfrog type II). The significance of this similarity is unclear, but may point to the importance of the threonine doublet identified here as critical for cGnRH-R internalization.
The apparent propensity of the cGnRH-R for undergoing rapid agonist-induced internalization via caveolae in a dynamin-dependent manner and independently of ß-arr1 and clathrin-coated vesicles is not restricted to COS-7 cells. Interestingly, the same internalization pathway is preferred in HEK 293 cells, which endogenously express about 70% more ß-arrestin than COS-7 cells (36). However, the maximal steady state level of internalization of the wild-type cGnRH-R was roughly 30% lower when expressed in HEK 293 cells than in COS-7 cells. It is interesting to note that the maximum steady state level of internalization of the wild-type catfish GnRH-R is much lower in COS-7 cells than in HEK 293 cells (19), underpinning the importance of ß-arrestin in the internalization pathway of the catfish GnRH-R. Clearly, both properties of a receptor and the cell type in which a receptor is expressed will determine the preferred pathway of agonist-induced internalization. It is therefore interesting to speculate about what, if any, the properties of the cGnRH-R are which make this receptor internalize preferentially in caveolae. We have shown that the presence of the carboxyl-terminal tail, the membrane-proximal Cys328, the Thr369Thr370 doublet are essential for rapid internalization of this receptor. Is it possible that these are the determinants for preferred internalization via caveolae?
A number of studies have reported caveolae-mediated internalization of GPCRs, including the ETA endothelin receptor (28), the cholecystokinin receptor (33), the B2 bradykinin receptor (29, 32, 47), and the m2 muscarinic receptor (30). The m2 muscarinic receptor was previously shown to preferentially internalize in a ß-arrestin-independent internalization pathway, but could be recruited to the ß-arrestin-dependent pathway by overexpression of ß-arrestin (48). Furthermore, the m1, m3, and m4 muscarinic receptors internalize via a ß-arrestin-independent, but dynamin-dependent, internalization pathway (49, 50), the same pathway used by the rat GnRH-R (21, 46). Internalization of the B2 bradykinin receptor has been shown to proceed independently of both ß-arrestin and dynamin, providing further support for its utilization of a caveolae-mediated pathway (47).
Most of the GPCRs to date shown to internalize via caveolae are coupled to Gq/11 and PLC activation. Apart from this common feature, there are no other apparent similarities between these GPCRs that could explain their preference for internalization via caveolae. To date, no receptor-specific elements, structural or sequence motifs, have been demonstrated to particularly target a GPCR to caveolae or promote its internalization via caveolae. It has been hypothesized that fatty acylation (palmitoylation and/or myristoylation) could function to target certain signaling molecules and cell surface receptors to caveolae compartments (43). Furthermore, many caveolar components share a common motif for dual acylation by palmitate and myristate at their amino termini (51). In this context, we note that the carboxyl-terminal tail of the cGnRH-R may be palmitoylated at the membrane-proximal Cys328 residue. Most GPCRs posses a putative palmitoylation site in the carboxyl-terminal tail proximal to the seventh transmembrane-spanning helix. Amino acid sequence comparison of this region of the cGnRH-R with that of the ß2-adrenergic receptor [known to be palmitoylated at Cys341 (52)] reveals a very similar sequence pattern, suggesting that should palmitoylation of Cys328 occur, this modification could serve to target the cGnRH-R to caveolae for rapid agonist-induced internalization.
In addition to the lack of identification of any sequence or structural motif being demonstrated to specifically target GPCRs to caveolae, the differences in dynamin dependency for caveolae-mediated internalization is a further complication (Refs. 31 , 34 , 35 , and 47 and this study). It appears that the involvement of ß-arrestin is not a prerequisite for internalization via a clathrin-mediated and dynamin-dependent pathway, and that the involvement of dynamin is not a prerequisite for internalization via a caveolae-mediated pathway. Alternatively, the pathways mediated by clathrin-coated pits and caveolae may be interlinked or positioned in tandem as suggested in a recent study (47). Certain receptors may therefore associate with clathrin-coated pits subsequent to their redistribution in caveolae. This scenario could explain the findings of a study demonstrating that rat GnRH-Rs colocalized with transferrin receptors (known to internalize via the clathrin-coated vesicle pathway), even though rat GnRH-R internalization is apparently ß-arrestin-independent (21). Recently, a study demonstrated that ß2-adrenergic receptors, which are known to internalize via a ß-arrestin-dependent, clathrin-mediated pathway, are initially localized within caveolae in neonatal myocytes (53). This localization was shown to be essential for signaling, which was disrupted by filipin treatment (53). As activation of ß2-adrenergic receptors leads to their rapid depalmitoylation and hyperphosphorylation (54), it is conceivable that this event could be the trigger to allow receptors to bind ß-arrestin and to be targeted from caveolae to the clathrin-mediated internalization pathway. Recently, palmitoylation-deficient mutant LH/CG receptors were demonstrated to undergo enhanced internalization and appeared to be more prone to ß-arrestin-dependent, clathrin-mediated internalization compared with wild-type receptors (55). In contrast, our C328A cGnRH-R mutant undergoes a markedly reduced level of internalization compared with the wild type. The above scenarios lend support to the idea that caveolae may be required by certain GPCRs to promote the initial formation of ligand receptor-G protein complexes and subsequent second messenger production by concentrating and increasing the proximity of the various components involved. Indeed, many species of G proteins, including Gq/11, have been shown to be enriched in caveolae (43, 56). There is evidence to suggest that the initial signaling events of other GPCRs coupled to Gq/11 and PLC activation may be predominantly restricted to caveolae (29). After this initial association in caveolae, certain receptors (e.g. ß2-adrenergic receptor) may then be targeted to clathrin-coated pits for internalization, whereas others (e.g. cGnRH-R, presented here) may remain in caveolae and undergo internalization via these compartments. Dynamin has certainly been implicated in caveolae-mediated endocytosis and may function in the scission of caveolae to form vesicles (34, 35). Furthermore, it has been shown that dynamin mediates caveolar internalization of muscarinic acetylcholine receptors (31).
In the current model for the regulation of archetypal agonist-activated GPCRs, the receptor must first be phosphorylated at specific serine and/or threonine residues, which promotes ß-arrestin binding. This results in rapid desensitization of the receptor by uncoupling it from its cognate G protein. The receptor is then targeted to clathrin-coated pits for internalization (3, 4, 5, 6, 7, 22, 23). A number of kinases have been implicated in GPCR regulation, including GRKs, protein kinase C (PKC), protein kinase A (PKA), and casein kinase II (CKII). The consensus sequence motifs for GRKs have not yet been clearly defined (although serine and/or threonine residues within an acidic environment are required), whereas those for PKC [(S/T)X(R/K)], PKA [(R/K)X(S/T)], and CKII [(S/T)XX(D/E)] have. The threonine doublet identified in this study does not lie in the consensus sequence motifs for PKC-, PKA-, or CKII-mediated phosphorylation, although the carboxyl-terminal tail of the cGnRH-R does contain putative PKC and CKII phosphorylation sites. These sites are, however, located in the region identified here as containing no elements essential for the rapid internalization of this receptor. This implies that PKC- and CKII-mediated phosphorylation of the cGnRH-R plays no apparent role in its rapid internalization, but may well be essential for other aspects of cGnRH-R regulation, such as desensitization and uncoupling, and resensitization. A recent study has suggested that CKII sites present in the carboxyl-terminal tail of the TRH receptor or introduced by mutation into the tail of the catfish GnRH-R play a role in ß-arrestin dependency for regulation of these receptors (57).
In conclusion, our data clearly demonstrate that there can be fundamental differences with respect to receptor trafficking function when deletion mutagenesis is compared with site-directed mutagenesis. This demonstrates to some extent a useful precedent, in that considering a truncated and therefore grossly distinct receptor, comparable to a point-mutated receptor, may be an oversimplification of the situation. We have identified a threonine doublet (Thr369Thr370) and a cysteine residue (Cys328) that are critical for rapid agonist-induced internalization of the cGnRH-R. Although the mechanism by which the threonine doublet promotes rapid internalization is unclear, it is possible that it serves a structural role rather than a site for protein-protein interaction. Palmitoylation of Cys328 may serve to target the cGnRH-R to caveolae microdomains for signaling and internalization. Whether expressed in COS-7 or HEK 293 cells, the cGnRH-R preferentially internalizes via caveolae in a dynamin-dependent manner, but may be recruited to the ß-arrestin- and clathrin-coated pit-dependent internalization pathway by overexpression of ß-arrestin in COS-7 cells. Finally, we provide evidence to support previous studies suggesting that clathrin-coated pits and caveolae are interlinked or positioned in tandem rather than existing as two mutually exclusive cell surface microdomains.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: ß-arr1, Wild-type ß-arrestin-1; Bmax, maximal binding level; cGnRH-R, chicken GnRH receptor; CKII, casein kinase II; dyn-1, wild-type dynamin-1; FCS, fetal calf serum; GFP, green fluorescence protein; GPCR, G protein-coupled receptor; GRK, GPCR kinase; IPmax, maximal total inositol phosphate; mßCD, methyl-ß-cyclodextrin; MDC, monodansylcadavarine; PKA, protein kinase A; PKC, protein kinase C; PLC, phospholipase C.
Received January 7, 2003.
Accepted for publication June 9, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T4 pituitary cells using recombinant adenovirus. Endocrinology 141:4564–4575 subunits Gq and Gi in caveolae in DDT1 MF-2 smooth muscle cells. J Biol Chem 272:17858–17866s complex leads to rapid depalmitoylation and inhibition of repalmitoylation of both the receptor and Gs. J Biol Chem 274:31014–31019-arrestin-dependent internalization. J Biol Chem 276:18066–18074This article has been cited by other articles:
 |
|
 |
|
  N. T Joseph, K. Morgan, R. Sellar, D. McBride, R. P Millar, and I. C Dunn The chicken type III GnRH receptor homologue is predominantly expressed in the pituitary, and exhibits similar ligand selectivity to the type I receptor J. Endocrinol., July 1, 2009; 202(1): 179 - 190. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. D. G. Pfleger, A. J. Pawson, and R. P. Millar Changes to Gonadotropin-Releasing Hormone (GnRH) Receptor Extracellular Loops Differentially Affect GnRH Analog Binding and Activation: Evidence for Distinct Ligand-Stabilized Receptor Conformations Endocrinology, June 1, 2008; 149(6): 3118 - 3129. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Pawson, E. Faccenda, S. Maudsley, Z.-L. Lu, Z. Naor, and R. P. Millar Mammalian Type I Gonadotropin-Releasing Hormone Receptors Undergo Slow, Constitutive, Agonist-Independent Internalization Endocrinology, March 1, 2008; 149(3): 1415 - 1422. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Naor, H. N. Jabbour, M. Naidich, A. J. Pawson, K. Morgan, S. Battersby, M. R. Millar, P. Brown, and R. P. Millar Reciprocal Cross Talk between Gonadotropin-Releasing Hormone (GnRH) and Prostaglandin Receptors Regulates GnRH Receptor Expression and Differential Gonadotropin Secretion Mol. Endocrinol., February 1, 2007; 21(2): 524 - 537. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M R Silver and S A Sower Functional characterization and kinetic studies of an ancestral lamprey GnRH-III selective type II GnRH receptor from the sea lamprey, Petromyzon marinus. J. Mol. Endocrinol., June 1, 2006; 36(3): 601 - 610. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Navratil, T. A. Farmerie, J. Bogerd, T. M. Nett, and C. M. Clay Differential Impact of Intracellular Carboxyl Terminal Domains on Lipid Raft Localization of the Murine Gonadotropin-Releasing Hormone Receptor Biol Reprod, May 1, 2006; 74(5): 788 - 797. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Orlichenko, B. Huang, E. Krueger, and M. A. McNiven Epithelial Growth Factor-induced Phosphorylation of Caveolin 1 at Tyrosine 14 Stimulates Caveolae Formation in Epithelial Cells J. Biol. Chem., February 24, 2006; 281(8): 4570 - 4579. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. E. Barran, R. W. Roeske, A. J. Pawson, R. Sellar, M. T. Bowers, K. Morgan, Z.-L. Lu, M. Tsuda, T. Kusakabe, and R. P. Millar Evolution of Constrained Gonadotropin-releasing Hormone Ligand Conformation and Receptor Selectivity J. Biol. Chem., November 18, 2005; 280(46): 38569 - 38575. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. N Hislop, C. J Caunt, K. R Sedgley, E. Kelly, S. Mundell, L. D Green, and C. A McArdle Internalization of gonadotropin-releasing hormone receptors (GnRHRs): does arrestin binding to the C-terminal tail target GnRHRs for dynamin-dependent internalization? J. Mol. Endocrinol., August 1, 2005; 35(1): 177 - 189. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Pawson, S. Maudsley, K. Morgan, L. Davidson, Z. Naor, and R. P. Millar Inhibition of Human Type I Gonadotropin-Releasing Hormone Receptor (GnRHR) Function by Expression of a Human Type II GnRHR Gene Fragment Endocrinology, June 1, 2005; 146(6): 2639 - 2649. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. K. Cheng and P. C. K. Leung Molecular Biology of Gonadotropin-Releasing Hormone (GnRH)-I, GnRH-II, and Their Receptors in Humans Endocr. Rev., April 1, 2005; 26(2): 283 - 306. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ronacher, N. Matsiliza, N. Nkwanyana, A. J. Pawson, T. Adam, C. A. Flanagan, R. P. Millar, and A. A. Katz Serine Residues 338 and 339 in the Carboxyl-Terminal Tail of the Type II Gonadotropin-Releasing Hormone Receptor Are Critical for {beta}-Arrestin-Independent Internalization Endocrinology, October 1, 2004; 145(10): 4480 - 4488. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. P. Millar and A. J. Pawson 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, August 1, 2004; 145(8): 3590 - 3593. [Full Text] [PDF]
|
|
|
|
|
|
A. S. Kauffman and E. F. Rissman A Critical Role for the Evolutionarily Conserved Gonadotropin-Releasing Hormone II: Mediation of Energy Status and Female Sexual Behavior Endocrinology, August 1, 2004; 145(8): 3639 - 3646. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Estall, B. Yusta, and D. J. Drucker Lipid Raft-dependent Glucagon-like Peptide-2 Receptor Trafficking Occurs Independently of Agonist-induced Desensitization Mol. Biol. Cell, August 1, 2004; 15(8): 3673 - 3687. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. P. Millar, Z.-L. Lu, A. J. Pawson, C. A. Flanagan, K. Morgan, and S. R. Maudsley Gonadotropin-Releasing Hormone Receptors Endocr. Rev., April 1, 2004; 25(2): 235 - 275. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. S. Kauffman and E. F. Rissman The Evolutionarily Conserved Gonadotropin-Releasing Hormone II Modifies Food Intake Endocrinology, February 1, 2004; 145(2): 686 - 691. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
), S366stop (
), D356stop (
), T351stop (
), S346stop (
), and S337stop (
) were expressed in COS-7 cells. Agonist-induced internalization was performed as described in Materials and Methods. The results shown are the mean ± SEM of six independent assays performed in duplicate.
), vector and sucrose (0.45 M) pretreatment (
). Agonist-induced internalization was performed as described in Materials and Methods. Pretreatments with MDC and sucrose were performed for 30 min at 37 C before the receptor internalization assay. The results shown are the mean ± SEM of three or four independent assays performed in triplicate.
). Agonist-induced internalization was performed as described in Materials and Methods. The results shown are the mean ± SEM of three independent assays performed in triplicate.