Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Like many other cell surface receptors, the G protein-coupled receptors (GPCRs) are internalized by endocytosis. For most of these receptors, endocytosis is triggered by agonist-induced activation and requires the phosphorylation of the receptor, an event that promotes the association of the GPCR with a family of proteins called arrestins. The nonvisual arrestins are clathrin-binding proteins, which then target the phosphorylated GPCRs to clathrin-coated pits for subsequent endocytosis (Koenig and Edwardson, 1997; Krupnick and Benovic, 1998; Lefkowitz, 1998).

Studies from this and other laboratories have shown that the lutropin/choriogonadotropin receptor (LHR) generally follows the internalization pathway summarized above for the majority of GPCRs. The agonist-occupied LHR is internalized via coated pits (Ghinea et al., 1992) by a pathway that can be inhibited with a dominant-negative mutant of dynamin or the clathrin-binding domain of -arrestin (Lazari et al., 1998). The rate of internalization of the agonist-activated LHR is faster than that of the free receptor or that of the receptor activated by a weak partial agonist (Lloyd and Ascoli, 1983; Hoelscher et al., 1991; Min et al., 1998), and mutations of the receptor that impair or enhance agonist-induced activation have a parallel effect on agonist-induced internalization (Dhanwada et al., 1996; Min et al., 1998). Mutation of the sites phosphorylated in response to agonist-induced activation also impair agonist-induced internalization of the LHR (Wang et al., 1997; Lazari et al., 1998). Last, overexpression of arrestin-3 enhances agonist-induced internalization (Lazari et al., 1998). Unlike most other GPCRs, however, the internalized LHR does not recycle back to the plasma membrane. Once internalized, the agonist-LHR complex traverses the endosomal compartment and is delivered to the lysosomes without dissociation (Ascoli, 1984; Ghinea et al., 1992).

Additional studies indicate that there are other structural features of the LHR that participate in agonist-induced endocytosis; thus, the rate of internalization of the agonist-rat LHR (rLHR) complex is much slower (T_1/2 = 60-120 min, depending on the cell type) than that of most GPCRs (Lloyd and Ascoli, 1983; Hoelscher et al., 1991; Dhanwada et al., 1996; Wang et al., 1997; Lazari et al., 1998; Min et al., 1998). Studies using chimeras of the rLHR and the closely related rat follitropin receptor (Nakamura and Ascoli, 1999) indicate that the extracellular domain of these receptors, a region that is not phosphorylated or involved in arrestin binding, has a substantial effect on the rate of internalization. Truncations of increasing portions of the C-terminal tail can impair or enhance agonist-induced internalization independently of agonist-induced activation or agonist-induced phosphorylation (Rodríguez et al., 1992; Wang et al., 1996). Last, the mutation of two adjacent cysteines present in the C-terminal tail of the LHR has also been reported to impair the agonist-induced endocytosis of the LHR without affecting signaling (Kawate and Menon, 1994).

Clues about additional structural features that may participate in the endocytosis of GPCRs can be derived by scanning their intracellular regions for the existence of internalization signals previously described in other membrane proteins. Several internalization signals have been described in cell surface receptors that have a single membrane spanning domain (Kirchhausen et al., 1997). Two of these, a tyrosine-based motif (NPXXY in transmembrane helix 7) and a leucine-based motif (LL in the juxtamembrane region of the C-terminal cytoplasmic tail), are present and conserved among many GPCRs, including the LHR (see alignment in Schulein et al., 1998). Studies performed with the ₂-adrenergic receptor have shown that mutations of the NPXXY motif impair agonist-induced endocytosis, but they appear to do so indirectly, by preventing the agonist-induced activation and subsequent phosphorylation of this receptor, rather than by a direct participation of this motif in endocytosis (Ferguson et al., 1997). In contrast, a more direct participation of the dileucine motif in the agonist-induced internalization of the ₂-adrenergic receptor is indicated by the finding that mutation of this motif had little or no effect on signaling but impaired agonist-induced endocytosis (Gabilondo et al., 1997). More recently, however, it was reported that mutation of the dileucine motif of the thromboxane A₂ receptor has no effect on internalization (Parent et al., 1999).

The current study was designed to investigate the involvement of leucine-based motifs on the agonist-induced internalization and other aspects of the trafficking of the LHR.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Plasmids and Cells. A full-length cDNA encoding for the rLHR was obtained as described previously (McFarland et al., 1989) and was subcloned into pcDNAI/Neo for expression. A plasmid containing the entire coding region of the human LHR (hLHR; Minegishi et al., 1990) was initially provided to us by Ares Advanced Technology and was subcloned into pcDNA3.1 for expression. Mutants were constructed using polymerase chain reaction strategies, and their identity was verified by automated DNA sequencing (performed by the DNA Core of The Diabetes and Endocrinology Research Center of the University of Iowa).

Internalization Assays. The LHR-mediated endocytosis of ¹²⁵I-labeled human choriogonadotropin (¹²⁵I-hCG) was measured as follows. Transiently transfected cells (plated onto 35-mm wells) were preincubated in 1 ml of Waymouth's MB752/1 containing 1 mg/ml BSA and 20 mM HEPES, pH 7.4 (assay medium), for 30 to 60 min at 37°C. Each well then received 10 ng/ml ¹²⁵I-hCG, and the incubation was continued at 37°C. At the times indicated, cells were placed on ice and washed two or three times with 2-ml aliquots of cold Hanks' balanced salt solution containing 1 mg/ml BSA (wash medium). The surface-bound hormone was then released by incubating the cells in 1 ml of cold 50 mM glycine and 150 mM NaCl, pH 3, for 2 to 4 min (Ascoli, 1982). This buffer was removed, and the cells were washed once more. The acid buffer washes were combined and counted, and the cells were solubilized with 100 µl of 0.5 N NaOH, collected with a cotton swab, and counted. The radioactivity associated with the acid washes was considered to be surface-bound hormone, whereas that associated with the solubilized cells was considered to be internalized hormone.

Fate of Internalized Hormone. This was measured using modifications of previously published procedures (Lloyd and Ascoli, 1983; Ascoli and Segaloff, 1987; Hoelscher et al., 1991). Transfected cells were washed as described above and then incubated with ¹²⁵I-hCG (10 ng/ml) for 2 h at 37°C. The cells were then placed on ice and washed two or three times with 2-ml portions of cold wash medium. The surface-bound hormone was then released by incubation of the cells in 1 ml of cold 50 mM glycine and 150 mM NaCl, pH 3, for 2 to 4 min (Ascoli, 1982). This buffer was removed, and the cells were washed once more with the same acid buffer and then once with cold assay medium. The cells were placed back in 1 ml of warm assay medium containing 50 ng/ml hCG (to prevent the reassociation of any undegraded ¹²⁵I-hCG released from the cells back into the medium), and a second 2-h incubation at 37°C was conducted to allow the cells to process the hormone that had been internalized during the first incubation. At the end of the second incubation, the dishes were placed on ice, the medium was saved, and the cells were washed once with 2 ml of cold wash medium. The assay and wash media washes were combined and precipitated with 10% trichloroacetic acid to determine the amounts of degraded and undegraded hormone released (Ascoli, 1982). The cells were then solubilized with NaOH (see above) and used to determine the amount of hormone that remained cell associated.

Hormone Binding Experiments. Binding of ¹²⁵I-hCG to intact cells was performed during an overnight incubation with 100 ng/ml ¹²⁵I-hCG at 4°C as described elsewhere (Wang et al., 1993). Detergent extracts used to measure ¹²⁵I-hCG binding were obtained by solubilizing the cells in 0.5% Nonidet P-40, 20 mM HEPES, 100 mM NaCl, 20% glycerol, and 1 mM EDTA, pH 7.4, using a constant ratio of 100 µl of detergent solution/1 million cells as described elsewhere (Ascoli, 1983; Wang et al., 1993). The detergent concentration was diluted to 0.1%, and triplicate aliquots of the extracts were incubated with 100 ng/ml ¹²⁵I-hCG. The third aliquot also received 50 IU/ml crude hCG to correct for nonspecific binding. The free and bound hormones were separated as described previously (Ascoli, 1983; Wang et al., 1993).

cAMP Measurements. The cells were washed twice with 2 ml of warm assay medium and placed in 1 ml of the same medium containing 1 mM isobutylmethylxanthine. After a 15-min preincubation at 37°C, increasing concentrations of hCG were added, and the incubation was continued for 30 min at 37°C. The wells were then placed on ice, and the total cAMP (in the cells and medium) was extracted by adding 1 ml of 2 N perchloric acid containing 360 µg/ml theophylline. The samples were subjected to rapid cycle of freezing and thawing, and the debris was collected by centrifugation. The supernatants were neutralized and then used for cAMP measurement by radioimmunoassay.

Other Methods. Three wells were used for each data point in all experiments involving the measurement of ¹²⁵I-hCG binding (see sections about internalization, fate of internalized hormone, and hormone binding experiments). Two of these received ¹²⁵I-hCG only, and the third received ¹²⁵I-hCG and 50 IU/ml crude hCG to correct for nonspecific binding. Statistical analysis (t test with two-sided p values) was performed using InStat (GraphPAD Software, San Diego, CA).

Hormones and Supplies. Purified hCG (CR-127) was kindly provided by the National Hormone and Pituitary Agency of the National Institute of Diabetes and Digestive and Kidney Diseases. ¹²⁵I-hCG was prepared as described elsewhere (Ascoli and Puett, 1978). ¹²⁵I-cAMP and cell culture medium were obtained from the Iodination Core and the Media and Cell Production Core, respectively, of the Diabetes and Endocrinology Research Center of the University of Iowa. Other cell culture supplies and reagents were obtained from Corning (Palo Alto, CA)and Life Technologies (Grand Island, NY), respectively. All other chemicals were obtained from commonly used suppliers.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Mutation of Dileucine-Based Motifs Enhances Agonist-Induced Internalization of LHR without Affecting Signaling. Although a dileucine-based motif present in the juxtamembrane region of the C-terminal tail is conserved in most GPCRs (see alignment in Schulein et al., 1998), the LHR is unusual in that depending on the species of origin, it has three (mouse, human, and porcine) or four (rat) contiguous leucines in this position (Fig. 1). In the experiments described below, we chose to examine the involvement of this region on the internalization of the rLHR and hLHR.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Amino acid sequence alignment of the juxtamembrane region of the C-terminal tail of the LHR from several species, the human 2-adrenergic receptor, and the human arginine-vasopressin V2 receptor. Sequences were obtained from the SWISS-PROT databank and have the following accession numbers: rLHR (P16235), mouse LHR (P30730), porcine LHR (P29525), hLHR (P22888), human 2-adrenergic receptor (P07550), and human arginine V2 receptor (P30518). The partial sequences shown start at the conserved NPXXY motif (shaded in gray) present in the seventh transmembrane domain (enclosed in a box). Dileucine motifs and a conserved acidic residue located upstream and separated by at least three amino acids from the dileucine motif are in black. The additional dileucine motif present in the rLHR is marked with asterisks. Palmitoylated cysteines are also shaded in gray.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Rates of internalization of 125I-hCG in 293 cells transiently transfected with rLHR-wt or rLHR-L613,L614A. 293 cells transiently transfected with rLHR-wt (top) or rLHR-L613,L614A (bottom) were preincubated in warm medium for 1 h. At the end of this incubation (t = 0), the cells received 125I-hCG (10 ng/ml), and the incubation was continued at 37°C. A and D, surface-bound and internalized radioactivity was measured at the times indicated using the acid release procedure described in Materials and Methods. B and E, plots of the ratio of internalized to surface ligand versus time. The straight lines were obtained using a linear least-squares fit of the data points shown. C and F, the integral of the surface-bound radioactivity was calculated at each time point as described in Materials and Methods, and a plot of the internalized hormone versus the integral of the surface-bound hormone was generated as shown. The straight line shown through the points was obtained using a linear least-squares fit of the data points. The results of a representative experiment are shown. Note that the scales are different in each panel.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Internalization of 125I-hCG mediated by rLHR-wt, hLHR-wt, and mutants thereof. Cells were transiently transfected with the indicated constructs, and the T1/2 values for internalization were measured as described in Materials and Methods. Each bar represents the mean ± S.E.M. of three independent transfections. Note that the different numbering in rLHR and hLHR is simply because the codons in the hLHR have always been numbered from the N terminus of the hLHR precursor (i.e., including the signal peptide), whereas codons in the rLHR have always been numbered starting from the known N terminus of the mature protein. *p < .05 compared with corresponding LHR-wt.

View larger version (34K): [in this window] [in a new window]   Fig. 4.   Signaling properties of the rLHR-wt and mutants thereof. Cells were transiently transfected with the indicated constructs. 125I-hCG binding was measured during a 1-h incubation at room temperature using a single, saturating concentration of 125I-hCG (100 ng/ml). Total cAMP accumulation was measured during a 30-min incubation with the indicated concentrations of hCG as described in Materials and Methods. Each value represents the mean ± S.E.M. of three independent transfections.

Rat LHR-wt and LHR-L613,614A Are Internalized by an Arrestin- and Dynamin-Dependent Pathway. Because rLHR-wt is internalized by a pathway that requires the participation of a nonvisual arrestin and dynamin (Lazari et al., 1998), it was important to determine whether the enhancement of endocytosis induced by mutation of the dileucine motif also had an effect on the pathway used for endocytosis. This possibility was examined by testing the effects of overexpression of arrestin-3, a dominant-negative mutant of the nonvisual arrestins (Krupnick et al., 1997) and a dominant negative mutant of dynamin (Damke et al., 1994), on the agonist-induced internalization of rLHR-wt and rLHR-L613,614A. The results of these experiments are presented in Fig. 5 and Table 2 and show that overexpression of -arrestin(319-418) or dynamin-K44A lengthened the T_1/2 of internalization of rLHR-L613,614A and rLHR-wt. The T_1/2 of internalization of hCG in cells transfected with rLHR-wt alone is comparable with that detected in cells cotransfected with rLHR-L613,L614A and -arrestin(319-418); on the other hand, the T_1/2 of internalization of hCG in cells cotransfected with rLHR-wt and dynamin-K44A is comparable with that detected in cells cotransfected with rLHR-L613,L614A and dynamin-K44A. The data presented in Table 2 also show that overexpression of arrestin-3 shortens the T_1/2 of internalization of rLHR-wt and rLHR-L613,614A. It is interesting to note, however, that the T_1/2 of internalization of hCG in cells transfected with rLHR-L613,614A alone is comparable with that detected in cells cotransfected with rLHR-wt and arrestin-3.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Effect of -arrestin(319-418) and dynamin-K44A on the rates of internalization of 125I-hCG mediated by rLHR-wt or rLHR-L613,L614A. 293 cells were transiently cotransfected with the rLHR-wt (top) or rLHR-L613,L614A (bottom) and the other indicated constructs. The cells were preincubated in warm medium for 1 h. At the end of this incubation (t = 0), the cells received 125I-hCG (10 ng/ml), and the incubation was continued at 37°C. The surface-bound and internalized radioactivity was measured at 3- to 10-min intervals as described in Materials and Methods. A and C, plots of the ratio of internalized to surface ligand versus time are shown. The straight lines shown were obtained using a linear least-squares fit of the data points. B and D, the integral of the surface-bound radioactivity was calculated at each time point as described in Materials and Methods, and a plot of the internalized hormone versus the integral of the surface-bound hormone was generated as shown. The straight line shown through the points was obtained using a linear least-squares fit of the data points. The results of a representative experiment are shown. Note that the scales are different in each panel.

Effect of Leucine Mutations on Other Aspects of Intracellular Trafficking of rLHR. In addition to their participation in endocytosis, dileucine motifs have also been shown to function in the targeting of proteins to the basolateral surface of epithelial cells, in the sorting of proteins from the trans-Golgi network to endosomes and lysosomes (Kirchhausen et al., 1997), and in the sorting of the arginine vasopressin V₂ receptors from the endoplasmic reticulum to the plasma membrane (Schulein et al., 1998). In light of this information, we also examined the effect of leucine mutations on the degradation of the internalized hCG and on the targeting of the nascent rLHR to the plasma membrane.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Agonist-induced receptor activation is an important event in triggering the endocytosis of GPCRs in general (Lefkowitz, 1998) and the LHR in particular (Lloyd and Ascoli, 1983; Hoelscher et al., 1991; Min et al., 1998), so we have suggested that the same, or a very similar, active conformation of the rLHR is involved in the stimulation of G proteins and the endocytosis of the agonist-rLHR complex (Min et al., 1998). As such, mutations that affect agonist-induced rLHR activation are thought to have a corresponding effect on the endocytosis of the agonist-rLHR complex that may be considered indirect in nature (i.e., mediated by their effects on the active conformation of the rLHR). In contrast, mutations that affect the endocytosis of the agonist-rLHR complex without affecting signaling are likely to do so because of a more direct participation of the region or residue in question in the endocytic pathway. This contention is also supported by the finding that most of the mutations that affect signaling and endocytosis are located in the transmembrane domains of the rLHR (see above), a region of the rLHR that is unlikely to be directly exposed to the endocytic machinery, whereas all of the known mutations that affect endocytosis without affecting signaling are located in the C-terminal cytoplasmic tail of the rLHR, a region that is likely to be directly exposed to the endocytic machinery. Moreover, protein sorting motifs identified in other receptors that interact directly with the endocytic machinery are generally located in their cytoplasmic domains (Kirchhausen et al., 1997).

Dileucine motifs have been previously shown to be important in the targeting of proteins to the basolateral surface of epithelial cells, the sorting of proteins from the trans-Golgi network to the endosomal/lysosomal compartments (Kirchhausen et al., 1997), the exit of proteins from the endoplasmic reticulum to the plasma membrane (Schulein et al., 1998), and the endocytosis of cell surface receptors (Dietrich et al., 1997; Gabilondo et al., 1997; Hamer et al., 1997; Vincent et al., 1997; Geisler et al., 1998; Govers et al., 1998). The data presented here show that a tetraleucine motif (L613-L616) present in the C-terminal cytoplasmic tail of the rLHR (Fig. 1) affects the internalization and externalization of this receptor. The simultaneous mutation of L613 and L614 enhances agonist-induced internalization (Fig. 2 and Table 1) but has no effect on signaling (Fig. 4), degradation of the internalized hormone (Table 3), or the targeting of the rLHR to the cell surface (Table 4). The ability of this dileucine motif to inhibit internalization is further supported by the finding that grafting it into the C-terminal tail of the hLHR inhibits the internalization of the agonist-hLHR complex (Fig. 3). Mutation of the second two leucines (L615 and L616) of the tetraleucine motif of the rLHR has a lesser effect on agonist-induced internalization (Table 1) and no effect on the degradation of the internalized hormone (Table 3), but it impairs the targeting of the nascent rLHR to the plasma membrane (Table 4).

The finding that mutation of L615 and L616 of the rLHR or mutation of D611, an upstream acidic residue conserved in many GPCRs (Fig. 1 and Table 4), impairs the targeting of the nascent rLHR to the plasma membrane represents the second example of the involvement of a dileucine-based motif preceded by an acidic residue (D/ExxxLL; see Fig. 1) in the targeting of a nascent GPCR to the plasma membrane. Thus, mutation of the two adjacent leucines or the upstream glutamic acid of the D/ExxxLL motif present in the C-terminal tail of the vasopressin V₂ receptor (Fig. 1) has recently been shown to prevent the exit of the nascent vasopressin V₂ receptor from the endoplasmic reticulum to the plasma membrane (Schulein et al., 1998). Although we did not attempt to establish the intracellular location of the D/ExxxLL mutants of the rLHR, it is likely that they are trapped in the endoplasmic reticulum because all other intracellularly trapped mutants of the rLHR have been shown to be located in this compartment (Rozell et al., 1995; Fabritz et al., 1998). It should also be mentioned that the D/ExxxLL sequence is not the only motif involved in the proper membrane localization of the rLHR, however, because there are many other mutations of residues present in the extracellular, transmembrane, or intracellular domains of the rLHR that result in intracellular trapping (Segaloff and Ascoli, 1993; Wang et al., 1993; Rozell et al., 1995).

More importantly, the results presented here on the effect of mutation of L613 and L614 on the agonist-induced endocytosis of the LHR represent the first example of a dileucine-based motif that impairs endocytosis. All other reports on the involvement of dileucine-based motifs on the endocytosis of cell surface receptors have shown that such motifs facilitate endocytosis. Thus, mutation of dileucine motifs has been shown to inhibit the endocytosis of several single transmembrane receptors such as the T cell receptor (Dietrich et al., 1997; Geisler et al., 1998), the insulin receptor (Hamer et al., 1997), the growth hormone receptor (Govers et al., 1998), the prolactin receptor (Vincent et al., 1997), and one GPCR, the ₂-adrenergic receptor (Gabilondo et al., 1997). The intracellular location of L613 and L614, as well as the finding that their mutation enhances the agonist-induced endocytosis of the rLHR without affecting agonist-induced activation, suggests that these two residues are directly involved in the interaction of the rLHR with the endocytic machinery. This conclusion is in agreement with all the knowledge about dileucine motifs derived from the study of other receptors. Thus, all the dileucine motifs previously identified as being involved in the endocytosis of other receptors are located in their intracellular regions (Gabilondo et al., 1997; Hamer et al., 1997; Geisler et al., 1998; Govers et al., 1998). Moreover, like tyrosine-based motifs, dileucine-based motifs are thought to participate in protein sorting through their direct interaction with two clathrin adaptor protein (AP) complexes: AP-1, which is found associated with clathrin in the trans-Golgi network, and AP-2, which is found in association with clathrin at the plasma membrane (Dietrich et al., 1997; Kirchhausen et al., 1997; Rapoport et al., 1998). Although AP-2 is thought to function as an adaptor in the clathrin-dependent endocytosis of single transmembrane receptors with tyrosine- or dileucine-based motifs (Kirchhausen et al., 1997), the only clathrin adaptors that have been shown to participate in the endocytosis of GPCRs, including the LHR, are the nonvisual arrestins (Table 2 and Krupnick and Benovic, 1998; Lazari et al., 1998).

Because the nonvisual arrestins participate in the endocytosis of GPCRs by bridging them to clathrin (Krupnick and Benovic, 1998), one hypothesis that would explain the enhanced internalization of rLHR-L613,614A is that this mutation enhances the binding of the nonvisual arrestins to the rLHR. This hypothesis is supported by the findings summarized in Table 2. First, the T_1/2 of internalization of hCG in cells cotransfected with rLHR-L613,614A and an empty vector (~18 min) is comparable with the T_1/2 of internalization of hCG in cells cotransfected with the rLHR-wt and arrestin-3 (~20 min). Second, although cotransfection with -arrestin(319-418) shortens the T_1/2 of internalization of hCG mediated by rLHR-wt and by rLHR-L613,614A, the T_1/2 of internalization of hCG in cells expressing rLHR-L613,614A and -arrestin(319-418; ~118 min) is comparable with the T_1/2 of internalization of hCG in cells expressing rLHR-wt only (~100 min). Last, cotransfection with dynamin-K44A, a construct that blocks endocytosis at a step located downstream of the nonvisual arrestins (Damke et al., 1994), lengthens the T_1/2 of internalization of ¹²⁵I-hCG mediated by rLHR-wt or rLHR-L613,L614A to ~215 and ~231 min, respectively.

An alternative hypothesis that is consistent with the results presented here states that the rLHR-wt is internalized by a pathway that involves both a nonvisual arrestin and AP-2. In this scenario, the involvement of AP-2 would be considered to be more important than the involvement of the nonvisual arrestins in the internalization of rLHR-wt, and it would be responsible for the slow rate of internalization of this receptor. Because the leucine-to-alanine mutations reported here are expected to disrupt the putative interaction of the rLHR with AP-2 (Kirchhausen et al., 1997), the involvement of the nonvisual arrestins would become more important than the involvement of AP-2 in the internalization of rLHR-L613,L614A, thus shortening the T_1/2 of internalization of this mutant.