Abstract
We previously observed that (trans)-3,4-dichloro-N-methyl-N-[2-(1-pyrrolidinyl)-cyclohexyl]benzeneacetamide (U50,488H) promoted internalization and phosphorylation of the FLAG-tagged human κ opioid receptor (FLAG-hkor) stably expressed in Chinese hamster ovary (CHO) cells. In this study, we compared regulation of the FLAG-hkor expressed in CHO cells by U50,488H, dynorphin A, etorphine, and levorphanol, which were potent full agonists as determined by stimulation of guanosine 5′-O-(3-[35S]thio)triphosphate binding. Using fluorescence flow cytometry, we found that dynorphin A(1-17), like U50,488H, promoted internalization of the FLAG-hkor in a time- and dose-dependent manner. The antagonists naloxone and norbinaltorphimine, having no effect on FLAG-hkor internalization, effectively blocked dynorphin A(1-17)- and U50,488H-induced internalization. Interestingly, the full agonists etorphine and levorphanol did not cause internalization of the FLAG-hkor but significantly reduced dynorphin A(1-17)- and U50,488H-induced internalization in a dose-dependent manner. Immunofluorescence staining of FLAG-hkor yielded similar results. Dynorphin A(1-17) and U50,488H enhanced phosphorylation of FLAG-hkor to a greater extent than etorphine, but levorphanol did not increase FLAG-hkor phosphorylation. Etorphine or levorphanol decreased dynorphin- or U50,488H-induced phosphorylation. It is likely that conformations of the hkor required for phosphorylation and initiation of internalization are different from those for activation of G proteins. We also examined whether the four agonists had differential effects on superactivation of adenylate cyclase. Pretreatment with U50,488H, dynorphin A(1-17), or etorphine enhanced forskolin-stimulated adenylate cyclase activity to ∼200 to 250% of the control, whereas levorphanol pretreatment did not result in significant adenylate cyclase superactivation. Thus, the degree of superactivation caused by an agonist is unrelated to its ability to promote internalization of the hkor.
Opioid compounds and peptides act on opioid receptors to exert their pharmacological and physiological functions. Opioid receptors were classified into at least three types, μ, δ, and κ, based on pharmacological and anatomical analyses (for reviews, see Chang, 1984;Mansour et al., 1988). Subsequently, μ, δ, and κ opioid receptors were cloned, and these receptors belong to the rhodopsin subfamily of the G protein-coupled receptor (GPCR) family (for reviews, see Kieffer, 1995; Knapp et al., 1995). Opioid receptors are coupled through pertussis toxin-sensitive G proteins to affect a variety of effectors, including inhibition of adenylate cyclase, increase in potassium conductance, decrease in calcium conductance, and activation of the p42/p44 mitogen-activated protein kinase pathway (for a review, see Law et al., 2000). Activation of κ opioid receptors produces many effects including analgesia (von Voigtlander et al., 1983; Dykstra et al., 1987), dysphoria (Pfeiffer et al., 1986; Dykstra et al., 1987), and water diuresis (von Voigtlander et al., 1983; Dykstra et al., 1987).
Most GPCRs show attenuated responsiveness to agonists after prolonged or repeated activation. Three distinct processes have been characterized: desensitization (seconds to hours), internalization (minutes to hours) and down-regulation (hours to days) (Tsao et al., 2001; Pierce et al., 2002). Activation of the receptor, in addition to initiating signal transduction, enhances phosphorylation of the receptor in intracellular domains, mostly by G protein-coupled receptor kinases (GRKs). Phosphorylation of the receptor facilitates binding of arrestins, which uncouple the receptor from G proteins, causing desensitization. Arrestins also act as adapter proteins binding clathrin and adapter protein-2, which results in internalization of the receptor. More prolonged activation leads to degradation of the receptor in lysosomes, proteasomes, or membranes, resulting in a reduction of the receptor number, which is termed down-regulation.
We demonstrated previously that U50,488H enhanced phosphorylation of the human κ opioid receptor (hkor) expressed in Chinese hamster ovary (CHO) cells, which was mediated by GRKs (Li et al., 2002). Using a receptor binding technique, we found that the hkor underwent U50,488H-induced internalization via a GRK-, β-arrestin-, and dynamin-dependent process that likely involved clathrin-coated vesicles (Li et al., 1999; Zhang et al., 2002). In addition, GRK2 or GRK3 that was co-internalized with the hkor and G protein βγ subunits played a critical role for internalization of the hkor (Schulz et al., 2002). However, unlike U50,488H, etorphine did not promote internalization of the hkor (Li et al., 1999). Since receptor binding was employed to detect internalized receptors in these studies, we could not examine whether endogenous dynorphin peptides caused internalization because it is difficult to remove these peptides, due to their sticky nature.
In this study, we employed fluorescence flow cytometry and immunofluorescence staining to compare internalization of FLAG-tagged hkor (FLAG-hkor) induced by dynorphin A(1-17), U50,488H, etorphine, and levorphanol and their combinations. In addition, effects of the four agonists and their combinations on phosphorylation of FLAG-hkor were examined. Moreover, we addressed the question whether a compensatory increase in adenylate cyclase activities following agonist pretreatment was related to receptor internalization.
Materials and Methods
Materials.
[35S]GTPγS (∼1,250 Ci/mmol), [3H]diprenorphine (58 Ci/mmol), [32P]orthophosphate (8500–9100 Ci/mmol), and [3H]cAMP (30–40 Ci/mmol) were purchased from PerkinElmer Life Sciences (Boston, MA). (−)-U50,488H was provided by the Upjohn Co. (Kalamazoo, MI). Etorphine, levorphanol, and nor-BNI were provided by the National Institute on Drug Abuse (Bethesda, MD). Dynorphin A(1-17) and naloxone HCl were purchased from Peninsula Laboratories (Belmont, CA) and Sigma/RBI (Natick, MA), respectively. Rabbit polyclonal antibody against the FLAG epitope, anti-FLAG mouse M1 antibody, GDP, calyculin A, and FLAG peptide (DYKDDDDK) were obtained from Sigma-Aldrich (St. Louis, MO). Goat anti-mouse IgG (H+L) conjugated with Alexa Fluo 488 was obtained from Molecular Probes (Eugene, OR). Pansorbin was obtained from Calbiochem (San Diego, CA). Geneticin was purchased from Mediatech (Herndon, VA). Normal goat serum was purchased from Organon Teknika (Durham, NC); Opti-MEM I reduced serum was purchased from Invitrogen (Carlsbad, CA); Triton X-100 was obtained from Roche Diagnostics (Indianapolis, IN); Lab-Tek II Slide Chambers was purchased from Lab-Tek (Naperville, IL).
Stable Transfection of CHO and HEK 293 Cell Lines with the Human κ Opioid Receptor and Cell Culture.
FLAG-tagged human κ opioid receptor (FLAG-hkor) in the expression vector pcDNA3 was generated previously (Li et al., 2002). Clonal CHO cell lines stably expressing FLAG-hkor (CHO-FLAG-hkor) were established previously (Li et al., 2002). HEK 293 cells stably expressing FLAG-hkor (HEK-FLAG-hkor) were established according to our published methods (Zhu et al., 1997). Cells were cultured in 100-mm culture dishes in Dulbecco's modified Eagle's medium-Ham's F-12 medium (for CHO-FLAG-hkor) or minimum essential medium (for HEK-FLAG-hkor) supplemented with 10% fetal calf serum, 0.2 mg/ml geneticin, 100 units/ml penicillin, and 100 μg/ml streptomycin in a humidified atmosphere consisting of 5% CO2 and 95% air at 37°C.
κ Opioid Receptor Binding.
Receptor binding was conducted with [3H]diprenorphine in 50 mM Tris-HCl buffer containing 1 mM ethylene glycol-bis-(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid and 5 μM leupeptin (pH 7.4), as described previously (Li et al., 2002). Naloxone (10 μM) was used to define nonspecific binding. Saturation experiments were performed with various concentrations of [3H]diprenorphine (ranging from 0.02 nM to 2 nM). Competitive inhibition of [3H]diprenorphine binding was performed with [3H]diprenorphine at a concentration close to its Kd (∼0.2 nM) and various concentrations of (−)-U50,488H, dynorphin A(1-17), etorphine, or levorphanol. Binding was conducted at 25°C for 60 min in duplicate in a volume of 1 ml with 30 to 40 μg of protein. Bound and free ligands were separated by rapid filtration under reduced pressure over GF/B filters presoaked with 0.2% polyethyleneimine and 0.1% bovine serum albumin in 50 mM Tris-HCl (pH 7.4) for 1 h. Binding data were analyzed with EBDA and LIGAND programs.
[35S]GTPγS Binding Assay.
Membrane preparation and the [35S]GTPγS binding assay were performed as described previously (Huang et al., 2001). Cells were washed twice and harvested in Versene solution (0.54 mM EDTA, 140 mM NaCl, 2.7 mM KCl, 8.1 mM Na2HPO4, 1.46 mM KH2PO4, and 1 mM glucose) and centrifuged at 500g for 3 min. The cell pellet was suspended in buffer A (5 mM Tris, pH 7.4, 5 mM EDTA, 5 mM ethylene glycol-bis-(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid, and 0.1 mM phenylmethylsulfonyl fluoride), passed through a 26G3/8 needle five times, and then centrifuged at 46,000gfor 30 min. The pellet was resuspended in buffer A and centrifuged again. The membrane pellet was resuspended in buffer B (50 mM Tris-HCl, pH 7.0, 0.32 mM sucrose), aliquoted at about 600 μg/ml, frozen in dry ice/ethanol, and stored at −80°C. All procedures were performed at 4°C.
Prior to assay, membranes were thawed at 37°C, chilled on ice, passed through a 22-gauge needle, and diluted with buffer C (50 mM HEPES, 100 mM NaCl, 5 mM MgCl2, and 1 mM EDTA with 0.1% bovine serum albumin freshly added, pH 7.4). Membranes (10–20 μg of protein) were incubated in buffer C containing [35S]GTPγS (100,000–150,000 dpm, ∼80 pM), GDP (3 μM), and varying concentrations of the κ opioid agonist U50,488H (10−10 to 10−6M), dynorphin A(1-17) (10−11 to 10−7 M), etorphine (10−11to 10−7 M), or levorphanol (10−9 to 10−5 M) in a total volume of 0.5 ml for 60 min at 30°C. Nonspecific binding was defined by incubation in the presence of 10 μM GTPγS. Bound and free [35S]GTPγS were separated by filtration through GF/B filters under reduced pressure. Radioactivity on filters was determined by liquid scintillation counting. EC50 and maximal response values were calculated by use of the equation, y = [Emax/[1 + (x/EC50)s]] + background, in which y is the response at the dosex, Emax is the maximal response, and s is a slope factor.
Phosphorylation of the κ Opioid Receptors.
Phosphorylation was conducted according to a procedure modified from that of Li et al. (2002). CHO-FLAG-hkor cells were transferred from 100-mm dishes into six-well plates and cultured overnight to confluence. Cells were then grown in 1 ml/well phosphate-free medium at 37°C for 2 h. [32P]Orthophosphate (0.25 mCi/well) was added and incubated for another 2 h, and medium was aspirated. Cells were incubated with 1 μM (−)-U50,488H, 0.1 μM dynorphin A(1-17), 1 μM etorphine, or 10 μM levorphanol for 30 min at 37°C, cooled on ice, and washed three times with ice-cold phosphate-buffered saline. All subsequent steps were carried out at 4°C. Cells were solubilized for 1 h with solubilization buffer [1% Triton X-100, 50 mM Tris HCl, 150 mM NaCl, 1 mM EDTA, 20 nM calyculin A, and 10% complete protease inhibitor cocktail from Roche Diagnostics (Indianapolis, IN), pH7.5], and centrifuged at 100,000g for 1 h. Immunoprecipitation of FLAG-hkor was performed with rabbit anti-FLAG polyclonal antibody followed by Pansorbin (final dilution 1:20, 4°C, 1 h) according to our published procedure (Li et al., 2002). The mixture was centrifuged and the pellets were washed three times by centrifugation and resuspension. Immunoprecipitated materials were dissolved in 2× Lammeli sample buffer and subjected to 8% SDS-polyacrylamide gel electrophoresis (Chen et al., 1995), and32P was detected by use of a phosphoimager (Cyclone; PerkinElmer Life Sciences). Quantitation of receptor phosphorylation was performed with the OptiQuant software program.
Quantitation of Receptor Internalization by Fluorescence Flow Cytometry.
A fluorescence flow cytometry assay was performed according to a modification of a procedure described by Gage et al. (2001). Briefly, CHO-FLAG-hkor cells (5 × 105 cells) cultured in six-well plates were left untreated or were treated for 30 min at 37°C with U50,488H, dynorphin A(1-17), etorphine, or levorphanol at indicated concentrations. In some experiments, cells were pretreated with 10 μM naloxone, 0.1 μM nor-BNI, 0.1 μM etorphine or 10 μM levorphanol for 10 min before treatment with U50,488H or dynorphin A(1-17). Cells were washed three times with ice-cold buffer A (58 mM Na2HPO4, 17 mM NaH2PO4, and 68 mM NaCl) and lifted with buffer A containing 0.5 mM EDTA. Cells were incubated with M1 anti-FLAG antibody (1 μg/ml) in 500 μl of Opti-MEM I reduced serum medium containing 1 mM CaCl2 for 45 min at 4°C. After three washes with buffer A, cells were incubated with Alexa Fluo 488-conjugated goat anti-mouse IgG (1 μg/ml) in 500 μl of Opti-MEM I reduced serum medium containing 1 mM CaCl2 for 45 min at 4°C. Cells were washed three times with ice-cold buffer A and then resuspended with 300 μl of buffer A. Immunoreactivity of surface receptor was quantitated by fluorescence flow cytometry (FACScan; BD Biosciences, San Jose, CA). Fluorescence intensity of 10,000 cells was collected for each sample. Cellquest software (BD Biosciences) was used to calculate the mean fluorescence intensity of single cells in each population. The mean fluorescence of cells stained only with Alexa Fluo 488-conjugated goat anti-mouse IgG was also determined and subtracted from each sample. Internalized receptors were calculated according to the following equation: internalized receptors (% of surface receptors) = 100% − (the mean fluorescence of 10,000 live cells with drug treatment)/the mean fluorescence of 10,000 live cells without drug) × 100% (Keith et al., 1996). The dose-response relationship was fitted to the equation y = [Emax/[1 + (x/EC50)s]] + background, in which y is the response at the dosex, Emax is the maximal response, ands is a slope factor.
Immunofluorescence Staining.
HEK 293 cells stably transfected with the FLAG-hkor (HEK-FLAG-hkor) were cultured in 100-mm dishes, transferred into Lab-Tek II Slide Chambers, and cultured overnight. Cells were treated with or without (control) a drug or drugs at indicated concentration(s) for 30 min at 37°C, washed three times with ice-cold buffer B (8.1 mM Na2HPO4, 1.9 mM NaH2PO4, 154 mM NaCl, 1 mM CaCl2), fixed with 4% paraformaldehyde in buffer B for 10 min at room temperature, and washed three times with buffer B at room temperature to remove the fixative. Subsequently, cells were permeabilized with 0.05% Triton X-100 for 10 min at room temperature and incubated with 4% normal goat serum at room temperature for 10 min to block nonspecific binding. Cells were incubated with anti-FLAG mouse M1 antibody (4 μg/ml) in buffer B/4% normal goat serum/0.05% Triton X-100 at 37°C for 30 min, rinsed three times with buffer B/0.05% Triton X-100 at room temperature, and incubated with goat anti-mouse IgG (H + L) conjugated with Alexa Fluo 488 (2 μg/ml) in buffer B/4% normal goat serum/0.05% Triton X-100 at 37°C for 30 min. After three washes with buffer B/0.05% Triton X-100 at room temperature, cells were mounted with Slow-Fade mounting medium, and coverslips were sealed with nail polish. Two controls were used: anti-FLAG mouse M1 antibody (4 μg/ml), pretreated with an excessive amount of the FLAG peptide (100 μg/ml) before incubation, and omission of the anti-FLAG mouse M1 antibody from the procedures. Both controls showed no staining. Cells were examined under a fluorescence microscope (ELIPSE TE300; Nikon, Tokyo, Japan) equipped with a 60× NA 1.4 objective and fluorescein filter sets.
Determination of cAMP Level.
CHO-FLAG-hkor cells were cultured in 12-well culture plates overnight before experiments. For agonist pretreatment, cells were incubated at 37°C for 4 h with 1 μM (−)-U50,488H, 0.1 μM dynorphin A(1-17), 1 μM etorphine, or 10 μM levorphanol. After treatment, medium was removed and cells were washed three times with prewarmed (37°C) 0.1 M phosphate-buffered saline. Isobutylmethylxanthine (1 mM) in prewarmed (37°C) Opti-MEM I reduced serum medium was added at 0.5 ml/well and incubated for 10 min at 37°C followed by naloxone at 10 μM (final concentration) for another 10 min. Cells were then incubated with 10 μΜ forskolin for 10 min at 37°C, and the reaction was terminated by placing the plates in boiling water for 10 min. The contents of each well were collected and frozen at −80°C. On the day of cAMP determination, the samples were thawed on ice and sonicated. cAMP contents in each sample were determined with the cAMP binding protein method described by Huang et al. (2001). [3H]cAMP (∼250,000 dpm in 0.02 M citrate phosphate buffer, pH 5.0) was added on ice to all sample tubes and tubes containing known amounts of cAMP (from 1.25 to 40 pmol) for generation of a standard curve. cAMP binding protein partially purified from bovine adrenal glands was added to each tube, except the blanks, at an amount which gave 10,000 to 20,000 dpm of [3H]cAMP binding in the absence of cold cAMP. The mixture (final volume 170 μl) was incubated 2 h to overnight at 4°C. Bound and free [3H]cAMP were separated by adsorption of free [3H]cAMP by 100 μl of charcoal suspension (10% Norit A charcoal, 4% bovine serum albumin, 1% Antifoam A) and centrifugation (1,500g for 20 min). Radioactivity of bound [3H]cAMP in an aliquot of the supernatant was determined by liquid scintillation counting. The standard curve was analyzed with a logit-log equation and the KaleidaGraph 3.5 Program (Synergy Software, Inc., Reading, PA). The amounts of cAMP in samples were calculated based on the standard curve and converted to picomoles per well.
Statistical Analysis.
For comparison of multiple groups, data were analyzed with analysis of variance to determine whether there were significant differences among groups using Prism 3.0 (GraphPad Software, Inc., San Diego, CA). If so, Dunnett's post hoc test was performed to determine whether there was a significant difference between the control and each treatment group. For comparison of two groups, Student's t test was performed. P< 0.05 was the level of significance in all statistical analyses.
Results
Dynorphin A(1-17), Etorphine, U50,488H, and Levorphanol Were Potent Full Agonists for the hkor.
Binding affinity of dynorphin A(1-17), etorphine, U50,488H, and levorphanol for the hkor, and their potency and efficacy in stimulating [35S]GTPγS binding were determined. Dynorphin A(1-17), etorphine, U50,488H, and levorphanol inhibited [3H]diprenorphine binding to the hkor with high affinity, withKi values in the nanomolar or subnanomolar range (Table 1). We have shown previously that binding performed in Tris buffer and in [35S]GTPγS binding buffer yields similarKi values for agonists for the κ opioid receptor (Zhu et al., 1997). All four were potent full agonists in enhancing [35S]GTPγS binding, with EC50 values in the range of 0.14 to 17.9 nM (Table 1). Both the affinity and potency were in the order of dynorphin A(1-17) > etorphine > U50,488H > levorphanol. When a single concentration was used in some experiments, the concentration was 400- to 700-fold of its EC50 value unless specified otherwise.
Dynorphin A(1-17), like U50,488H, Promoted Internalization of FLAG-hkor, But Etorphine and Levorphanol Did Not.
When CHO-FLAG-hkor cells were incubated with M1anti-FLAG monoclonal antibody followed by Alexa Fluo 488-conjugated goat anti-mouse IgG, enhanced fluorescence level was detected using fluorescence flow cytometry, whereas the untransfected cells displayed little fluorescence. Dynorphin A(1-17) (0.1 μM) caused internalization of the receptor in a time-dependent manner, reaching a plateau at 30 min, similar to U50,488H (1 μM) (Fig.1). At the plateau, 30 to 40% of the receptors were internalized, and the extents of internalization achieved by dynorphin A(1-17) and U50,488H did not differ significantly. Pretreatment for 2 h with monensin (50 μM), a sodium ionophore which prevents acidification of intracellular vesicles and blocks the recycling of internalized receptors (Pippig et al., 1995), did not affect immunofluorescence of cell surface κ receptors, indicating that during the 30-min incubation period, there is no significant recycling of the receptor. A 30-min incubation was used in subsequent experiments. In contrast to dynorphin A(1-17) and U50,488H, etorphine (10−11, 10−10, 10−9, 10−8, 10−7, and 10−6 M) and levorphanol (10−6 or 10−5M) did not induce internalization of Flag-hkor (Fig.2, 10−6 M etorphine and 10−5 M levorphanol only). Pretreatment with monensin did not affect surface receptor immunofluorescence following etorphine or levorphanol treatment (data not shown). Thus, the lack of internalization by either drug is not the result of rapid recycling of internalized receptor. These results indicate that there are differences among agonists in promoting internalization of FLAG-hkor. Naloxone or nor-BNI, which itself did not affect internalization, blocked dynorphin A(1-17)- or U50,488H-induced internalization (Fig. 2).
Etorphine or Levorphanol Reduced Dynorphin A(1-17)- or U50,488H-Induced Internalization.
Whether etorphine or levorphanol had any effect on dynorphin A(1-17)- or U50,488H-induced internalization of FLAG-hkor was examined. Etorphine (0.1 and 1 μM) or levorphanol [1 and 10 μM for U50,488H, 10 μM for dynorphin A(1-17)] significantly reduced dynorphin A(1-17)- or U50,488H-induced internalization in a dose-dependent manner (Fig.3, A and B).
U50,488H and dynorphin A(1-17) dose dependently promoted internalization of the FLAG-hkor. The EC50 values of U50,488H and dynorphin A were 44.3 ± 8.0 and 2.6 ± 0.5 nM (n = 3 each, mean ± S.E.M.), respectively, and the maximal responses were reached at about 1 and 0.1 μM, respectively (Fig. 4, A and B). Etorphine (0.1 μM) and levorphanol (10 μM) increased the EC50 value of dynorphin A(1-17) by about 9- and 26-fold, respectively, without affecting the maximal response. No apparent maximal responses were reached up to 10−5 M U50,488H in the presence of etorphine (0.1 μM) and levorphanol (10 μM). The dose of U50,488H required to produce the same degree (13%) of internalization was increased by about 30- and 140-fold by etorphine (0.1 μM) and levorphanol (10 μM), respectively (Fig. 4, A and B).
Effects of Agonists and Antagonists and Combinations on Distribution of FLAG-hkor Immunofluorescence.
Immunofluorescence staining was carried out with M1 anti-FLAG antibody to detect surface and intracellular FLAG-hkor. Since CHO cells have large nuclei and small cytosol volumes, it is difficult to visualize internalized receptors. In contrast, HEK 293 cells have much smaller nuclei and a much larger cytosol volume; therefore, the cells are commonly used for visualizing internalized receptor. Using fluorescence flow cytometry, we found that HEK 293 cells and CHO cells were similar in the extent of U50,488H-promoted internalization of the FLAG-hkor; we thus used HEK 293 cells for immunofluorescence microscopy. Without drug treatment, immunofluorescence staining of FLAG-hkor was mostly on the cell surface. Dynorphin A(1-17) and U50,488H decreased cell-surface staining and caused punctate staining in the cytosol, indicating internalization of FLAG-hkor, but etorphine, levorphanol, nor-BNI, or naloxone did not (Fig.5). Treatment with etorphine, levorphanol, nor-BNI, or naloxone blocked or greatly reduced cytosolic punctate staining induced by dynorphin A(1-17) and U50,488H (Fig. 5). These results were consistent with those obtained with fluorescence flow cytometry analysis.
U50,488H or Dynorphin A(1-17) Did Not Facilitate Internalization of Etorphine- or Levorphanol-Occupied Receptors.
Whether U50,488H or dynorphin could facilitate etorphine or levorphanol to induce internalization of FLAG-hkor was investigated. Treatment with 10 nM U50,488H or 1 nM dynorphin A(1-17), which caused a low level of internalization, did not facilitate etorphine or levorphanol to promote internalization (data not shown). Rather, etorphine and levorphanol blocked the low level of internalization induced by 10 nM U50,488H or 1 nM dynorphin A(1-17) (data not shown).
Effects of the Four Agonists and Combinations on Phosphorylation of FLAG-hkor.
To investigate whether the differences among agonists in promoting receptor internalization are related to the abilities of the agonists in elevating receptor phosphorylation, we examined FLAG-hkor phosphorylation induced by the four agonists. The extent of phosphorylation of FLAG-hkor was in the order of dynorphin A(1-17) = U50,488H > etorphine, but levorphanol did not enhance phosphorylation of FLAG-hkor (Fig. 6). The molecular weight of phosphorylated FLAG-hkor was identical to what we reported previously (Li et al., 2002). U50,488H-induced phosphorylation of FLAG-hkor was shown to be blocked by naloxone (Li et al., 2002). We then examined whether etorphine or levorphanol had any effect on dynorphin A(1-17)- or U50,488H-induced phosphorylation. As shown in Fig. 6, etorphine or levorphanol reduced dynorphin A(1-17)- or U50,488H-induced phosphorylation of FLAG-hkor (Fig. 6).
Effects of Pretreatment with the Four Agonists on Forskolin-Stimulated Adenylate Cyclase Activity.
Finn and Whistler (2001) reported that lack of agonist-induced internalization of the μ opioid receptor or its mutant resulted in enhanced superactivation of adenylate cyclase. We thus examined whether pretreatment of the FLAG-hkor with the four agonists had differential effects on adenylate cyclase superactivation. As shown in Fig.7, pretreatment with U50,488H, dynorphin A(1-17), and etorphine for 4 h enhanced forskolin-stimulated adenylate cyclase to 200 to 250% compared with the untreated control. The extents of adenylate cyclase superactivation induced by dynorphin A(1-17), U50,488H, and etorphine were not significantly different. Etorphine at 0.1 μM and 1 μM produced similar effects. In contrast, levorphanol pretreatment did not cause significant superactivation. Thus, the four agonists have differential abilities to induce superactivation of adenylate cyclase, and the degree of superactivation is not related to whether the agonist causes internalization of the FLAG-hkor.
Discussion
Although etorphine and levorphanol were potent full agonists for the hkor in stimulating [35S]GTPγS binding, neither drug caused internalization of the FLAG-hkor, and etorphine slightly increased, but levorphanol did not enhance, phosphorylation of the FLAG-hkor. Rather, etorphine and levorphanol acted as antagonists in reducing internalization and phosphorylation induced by U50,488H and dynorphin A(1-17). To the best of our knowledge, this study represents the first report that a full agonist of a GPCR in receptor-mediated signaling acts as an antagonist in internalization and phosphorylation of the receptor. In addition, U50,488H, dynorphin A(1-17), and etorphine pretreatment enhanced forskolin-stimulated adenylate cyclase to 200 to 250% of that of the control, but levorphanol did not. Thus, the degree of adenylate cyclase superactivation is not related to receptor internalization.
Receptor Conformations Required for Internalization are Different from Those for G Protein Activation.
For stimulating [35S]GTPγS binding, the EC50 values of U50,488H and dynorphin A(1-17) were 2.07 ± 0.13 and 0.14 ± 0.01 nM, respectively. For promoting internalization, the EC50 values of U50,488H and dynorphin A(1-17) were 44.3 ± 8.0 and 2.6 ± 0.5 nM, respectively. The finding that EC50values of U50,488H and dynorphin A(1-17) in promoting internalization were 21 and 19 times, respectively, those for stimulating [35S]GTPγS binding supports the notion that different conformations are required for the two processes. Whereas dynorphin A(1-17) and U50,488H can induce conformational changes resulting in G protein coupling and receptor internalization, etorphine and levorphanol can only induce conformational alterations leading to G protein activation, but not receptor internalization. The study supports the notion that there are multiple activated states for GPCRs, the models for which have been proposed (for example, Scaramellini and Leff, 1998). Our results are consistent with those of Mhaouty-Kodja et al. (1999), that although two different constitutively active mutants of the α1B-adrenergic receptor had similar agonist-independent activities, the A6.34(293)E mutant had an enhanced basal phosphorylation level and underwent β-arrestin-mediated basal and agonist-induced internalization, but the D3.49(142)A mutant did not.
Differential Effects of Agonists in Promoting Phosphorylation and Internalization of the κ Receptors.
Our findings that levorphanol or etorphine did not promote internalization is in accord with the report of Blake et al. (1997b) that neither drug caused desensitization or down-regulation of the hkor. That dynorphin A and U50,488H induced phosphorylation and internalization is consistent with the observations that dynorphin A and U50,488H caused desensitization and down-regulation of the hkor (Blake et al., 1997b; Ling et al., 1998; Zhu et al., 1998; Li et al., 2000; Zhang et al., 2002). It is interesting that etorphine enhanced phosphorylation, albeit to a lesser extent than U50,488H or dynorphin A(1-17), but did not promote internalization under the same condition. Overexpression of GRK2 and arrestin-2 facilitated etorphine to induce down-regulation of the hkor (Li et al., 2000).
It should be noted that the rat κ opioid receptor stably expressed in CHO cells did not undergo U50,488H-induced regulation (Avidor-Reiss et al., 1995; Li et al., 1999, 2000; Jordan et al., 2000; Zhang et al., 2002), even with overexpression of GRK2 and arrestin-2 (Li et al., 2002). This species difference was attributed to the amino acid sequence difference in the C-terminal domain, particularly Asn358 versus Ser358 in the hkor (Li et al., 2002; Zhang et al., 2002). U50,488H caused phosphorylation and desensitization of the guinea pig κ opioid receptor, which has Ser358 as the hkor (Appleyard et al., 1997). In other cells, U69,593 or U50,488H did or did not cause desensitization of the mouse or rat κ receptor (Joseph and Bidlack, 1995; Tallent et al., 1998; Appleyard et al., 1999). Dynorphin peptides induced small degrees of internalization and down-regulation of the rat κ opioid receptor in CHO cells (Jordan et al., 2000). Etorphine did not induce internalization of the mouse κ receptor in HEK 293 cells (Chu et al., 1997).
Differences among agonists in promoting internalization have also been observed for μ and δ opioid receptors. Etorphine and various peptide agonists promoted internalization of both μ and δ opioid receptors, whereas morphine and levorphanol did not (Arden et al., 1995; Keith et al., 1996, 1998; Sternini et al., 1996; Bot et al., 1997). Zhang et al. (1998) demonstrated that morphine did not induce phosphorylation of the μ opioid receptor. Morphine partially inhibited etorphine-induced internalization of the μ opioid receptor (Sternini et al., 1996). Whether levorphanol or morphine has antagonistic effects on phosphorylation of the μ or δ receptor induced by other agonists has not been examined.
That four full agonists had differential abilities in promoting internalization of FLAG-hkor is in accord with the observations ofKeith et al. (1998) and Alvarez et al. (2002) that the relative ability of agonists in causing internalization of the μ opioid receptor did not correlate with their abilities to activate the receptor.
U50,488H- or dynorphin A(1-17)-activated receptors did not affect internalization of etorphine- or levorphanol-occupied receptors.
U50,488H or dynorphin A(1-17), at concentrations that caused a low level of internalization, did not facilitate etorphine or levorphanol to promote internalization. These findings are different from those ofHe et al. (2002), that a low concentration of DAMGO facilitated morphine to induce internalization of the μ opioid receptor. They attributed the observation to oligomerization of the μ opioid receptor. DAMGO-occupied receptors in the oligomer complexed with morphine-occupied receptors recruit internalization machinery to enable internalization of morphine-occupied receptors. Although the rat κ opioid receptors were shown to form homodimers or oligomers (Jordan and Devi, 1999), whether the human κ opioid receptors dimerize has not been examined.
Etorphine and Levorphanol Were More Potent in Inhibiting U50,488H- than Dynorphin A(1-17)-Promoted Internalization.
Etorphine (0.1 μM) or levorphanol (10 μM) shifted the dose-response curve of U50,488H-induced internalization to the right more than that of dynorphin A(1-17). The differences may be attributed to the finding that U50,488H and dynorphin A(1-17) bind to different domains of the κ opioid receptor (Xue et al., 1994), and etorphine and levorphanol may be able to compete more effectively with U50,488H for binding than with dynorphin A(1-17). In addition, it may be due to the difference in their relative potencies at the κ receptor, with dynorphin A (1-17) being ∼15 times more potent than U50,488H.
Relationship between Regulation of the FLAG-hkor to Superactivation of Adenylate Cyclase.
Pretreatment of CHO-hkor cells for 3 or 4 h with U50,488H or dynorphin A(1-17) caused desensitization of agonist-induced adenylate cyclase inhibition and down-regulation of the receptor, but levorphanol or etorphine pretreatment did not (Blake et al., 1997b; Zhu et al., 1998; Li et al., 2000). Our results indicate that lack of desensitization, internalization, and down-regulation did not enhance the degree of adenylate cyclase superactivation following agonist pretreatment, since U50,488H and dynorphin A(1-17), which induced regulation, and etorphine, which did not, caused similar degrees of superactivation. In addition, levorphanol, which did not cause internalization and down-regulation, did not induce superactivation of adenylate cyclase. These results are different from those of Finn and Whistler (2001). Using different agonists and mutants of the μ opioid receptor, these researchers found that lack of internalization enhanced the extent of superactivation of adenylate cyclase. They hypothesized that the persistent stimulation of cell-surface receptors caused higher degrees of cellular adaptation. The reasons for this difference are not clear. The events following activation of the κ opioid receptor leading to superactivation of adenylate cyclase may be different from those following μ receptor activation.
Regulation of Opioid Receptors by Etorphine and Levorphanol.
We found that etorphine and levorphanol did not promote internalization of the hkor. However, etorphine induced internalization of the μ and δ opioid receptors (Chu et al., 1997; Keith et al., 1998). Levorphanol pretreatment did not internalize the δ receptor (Bot et al., 1997), and whether it internalized the μ receptor has not been reported. Etorphine pretreatment of the hkor resulted in superactivation of adenylate cyclase, but levorphanol did not. Following treatment of the μ receptor with etorphine or levorphanol, no superactivaiton of adenylate cyclase was observed (Blake et al., 1997a).
Relationship between Receptor Regulation and Opioid Tolerance.
Repeated administration of κ opioid agonists leads to tolerance to the antinociceptive effect of κ agonists (Murray and Cowan, 1988;Bhargava et al., 1989), which may be partially accounted for at the receptor level (von Voigtlander et al., 1983; Bhargava et al., 1989;Morris and Herz, 1989; Joseph and Bidlack, 1995). Opioid tolerance is underlain by a multitude of biological mechanisms. There are two opposing views regarding the relationship between agonist-induced regulation of the μ opioid receptor and opioid tolerance. Bohn et al. (2000) showed that β-arrestin 2-deficient mice did not develop tolerance to morphine, indicating that β-arrestin 2-mediated biological events, which include agonist-induced desensitization, internalization, and down-regulation, contribute significantly to morphine tolerance. In contrast, He et al. (2002) reported that a small dose of DAMGO facilitated morphine to stimulate internalization of the μ opioid receptor, and rats treated chronically with both drugs showed reduced analgesic tolerance compared with rats treated with morphine alone, indicating that receptor internalization reduces morphine tolerance. An agonist that does not cause internalization may produce different degrees of tolerance from an agonist that promotes internalization. An agonist that blocks internalization induced by another agonist may modulate tolerance development of the second agonist. Needless to say, these hypotheses have to be tested in vivo.
Conclusion.
Dynorphin A(1-17) and U50,488H promoted internalization of the FLAG-hkor, but etorphine or levorphanol did not, although all were potent full agonists. Three agonists induced phosphorylation of FLAG-hkor in the order of dynorphin A = U50,488H > etorphine, but levorphanol did not. Etorphine or levorphanol reduced dynorphin- or U50,488H-induced phosphorylation and internalization. U50,488H, dynorphin A(1-17), and etorphine induced superactivation of adenylate cyclase, but levorphanol did not. Taken together, these results indicate that agonists have differential effects on regulation of the hkor, likely due to different receptor conformational changes induced by the drugs.
Footnotes
-
This work was supported by National Institutes of Health Grants DA 04745, DA11263, and DA13429.
-
DOI: 10.1124/jpet.102.045559
- Abbreviations:
- GPCR
- G protein-coupled receptor
- GRK
- G protein-coupled receptor kinase
- (−)-U50,488H
- (−)-(trans)-3,4-dichloro-N-methyl-N-[2-(1-pyrrolidinyl)-cyclohexyl]benzeneacetamide
- hkor
- human κ opioid receptor
- CHO
- Chinese hamster ovary
- CHO-hkor
- clonal CHO cell lines stably expressing the human κ opioid receptor
- CHO-FLAG-hkor
- clonal CHO cell lines stably expressing the FLAG-tagged human κ opioid receptor
- FLAG-hkor
- FLAG-tagged human κ opioid receptor
- GTPγS
- guanosine 5′-O-(3-thiotriphosphate)
- nor-BNI
- norbinaltorphimine
- U69,593
- (5α,7α,8β)-(−)-N-methyl-N-(7-(1-pyrrolidinyl)-1-oxaspiro(4,5)dec-8-yl) benzeneacetamide
- DAMGO
- [d-Ala2,N-Me-Phe4,Gly5-ol]-enkephalin
- ANOVA
- analysis of variance
- Received October 11, 2002.
- Accepted January 13, 2003.
- The American Society for Pharmacology and Experimental Therapeutics