Simultaneous α_2B- and β₂-Adrenoceptor Activation Sensitizes the α_2B-Adrenoceptor for Agonist-Induced Down-Regulation

Over the last decade, considerable attention has been given to the short- and long-term regulation of α₂-adrenoceptor signaling. Primarily using transfected receptors in cell lines that do not endogenously express the α₂-adrenoceptor, Liggett and coworkers have conducted a series of experiments examining the short-term and long-term desensitization of the three α₂-adrenoceptor subtypes (α_2A/D,2B and _2C) (Liggett, 1998). Through these studies, structural components of the α₂-adrenoceptor have been identified as sites for GRK phosphorylation that are important for the short-term desensitization of these receptors. For example, four serines in the third intracellular loop are reported to be critical for short-term desensitization of the α_2A-adrenoceptor (Eason et al., 1995). However, there are structural differences within the α₂-adrenoceptor family that may lead to heterogeneity in the regulation of the three α₂-subtypes (Eason and Liggett, 1992). For example, significantly higher agonist concentrations are reportedly required for desensitization of α_2A compared with α_2B-adrenoceptors (Heck and Bylund, 1997). Moreover, studies have suggested that phosphorylation sites critical to short-term α₂-adrenoceptor desensitization are not important for long-term desensitization and down-regulation (Jewell-Motz and Liggett, 1995; Jewell-Motz et al., 1997). Data regarding long-term regulation of the α_2C-adrenoceptor also are controversial. A recent study by Bylund and coworkers suggests that GRK-mediated phosphorylation is critical for long-term agonist-induced down-regulation and desensitization of the α_2C-adrenoceptor (Deupree et al., 2002). They observed that the opossum kidney α_2C-adrenoceptor undergoes down-regulation, whereas the human receptor does not, and this difference seemed related to structural differences in the third intracellular loop of the receptor in these species. In contrast to this, down-regulation of the human α_2C-adrenoceptor in Hep2G cells was reported after long-term agonist exposure and was demonstrated to be the result of increased receptor degradation (Cayla et al., 1999). In summary, the role of GRK-mediated phosphorylation and receptor down-regulation in the long-term modulation of the α₂-adrenoceptor signaling is not well established.

Recently, we have demonstrated that 0.3 μM epinephrine (EPI) produces desensitization of α_2A-adrenoceptors after 24-h treatment of BE(2)-C human neuroblastoma cells in culture, whereas 1 μM norepinephrine (NE) has no effect (Bawa et al., 2003). These cells endogenously express both α_2A- and β₂-ARs. The desensitization by EPI was accompanied by α_2A-adrenoceptor down-regulation, and both desensitization and down-regulation were prevented by β-adrenoceptor blockade during the EPI treatment. Furthermore, the α_2A-adrenoceptor desensitization and down-regulation was accompanied by a selective up-regulation of GRK3 expression. β-Adrenoceptor blockade or GRK3 antisense treatment prevented the GRK3 up-regulation and also prevented the α_2A-adrenoceptor down-regulation by EPI treatment. The present study was initiated to investigate the potential cross talk regulation between the β₂-adrenoceptor and another α₂-adrenoceptor subtype, the α_2B. The model system chosen for this study was the NG108 cell, a neuroblastoma/glioma hybrid cell line that naturally expresses α_2B-adrenoceptors and has been used previously to study regulation of α₂-adrenoceptor signaling (Thomas and Hoffman, 1986; Convents et al., 1989). We also used the BN17 cell line, derived from the transfection of the human β₂-adrenoceptor gene into NG108 cells (Adie and Milligan, 1994). Our results suggest that simultaneous activation of the α_2B- and β₂-adrenoceptors enables the α_2B-adrenoceptor to be down-regulated by 67-fold lower EPI concentrations than when only the α₂-adrenoceptor is activated. This difference is due to the selective up-regulation of GRK3 produced by this treatment. In addition, we report here that in the absence of β₂-adrenoceptor activation, higher agonist concentrations produce α_2B-adrenoceptor down-regulation without up-regulation of GRK3 expression.

Materials and Methods

Materials. The following drugs were purchased from the indicated sources: (-)-EPI bitartrate, (+)-isoproterenol bitartrate, phenylmethylsulfonyl fluoride, alprenolol hydrochloride, phentolamine, cAMP, prostaglandin E1 (PGE₁), Dulbecco's modified Eagle's medium, alprenolol hydrochloride, adrenal cortex extract, hydroxyapatite, HAT supplement (0.1 mM hypoxanthine, 0.4 μM aminopterin, and 16 μM thymidine), sodium orthovanadate, sodium pyrophosphate, pepstatin, leupeptin, aprotinin, isobutylmethylxanthine, and poly-l-lysine hydrobromide (Sigma-Aldrich, St. Louis, MO); (-)-NE (Sigma/RBI, Natick, MA); [³H]cAMP and [³H]RX821002 (Amersham Biosciences UK, Ltd., Buckinghamshire, UK); G418 sulfate (Calbiochem, La Jolla, CA); fetal bovine serum and penicillin-streptomycin (Atlanta Biologicals, Norcross, GA); N,N,N′,N′-tetramethylethylenediamine and ammonium persulfate (Bio-Rad, Hercules, CA); GRK3 antibody (catalog no. sc-563), GRK2 antibody (catalog no. sc-562), anti-rabbit antibody (catalog no. sc-2301), anti-mouse antibody (catalog no. sc-2302), and enhanced chemiluminescence reagent (Santa Cruz Biotechnology, Inc., Santa Cruz, CA); glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibody (catalog no. RDITRK4G4C5) (Research Diagnostics, Flanders, NJ); and GRK3 antisense oligodeoxynucleotide (ODN) (5′-CCCGGTGTCTGCTTTCCT-3′) and sense (5′-AGGAAAGCAGACACCGGG-3′) (Sigma-Genosys, The Woodlands, TX).

Cells and Plasmids. NG108 and BN17 cells were obtained from Dr. Graeme Milligan (University of Glasgow, Glasgow, Scotland, UK). BN17 cells are NG108 cells transfected using plasmid pJM16, carrying a copy of the neomycin resistance gene and the cDNA encoding human β₂-AR, to express the human β₂-AR at about 300 fmol/mg protein (Adie and Milligan, 1994).

Cell Culture. The neuroblastoma/glioma hybrid NG108 cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% heat inactivated fetal calf serum, penicillin (100 U/ml), streptomycin (100 μg/ml), and HAT supplement. BN17 cells were grown in the same way as NG108 cells except that the media contained G418 (0.6 mg/ml) to maintain selection pressure. The NG108 and BN17 cells were grown in either 75-cm² flasks or 150-cm² plates coated with poly-l-lysine. Flasks or plates of cells that were more than 80% confluent were used throughout the study.

Pretreatment. NG108 cells and BN17 cells were pretreated with vehicle (serum-free medium containing 0.1 mM ascorbate and 1 μM sodium metabisulfite), or vehicle containing 0.3-20 μM NE or EPI for 24 h. In BN17 cells, NE or EPI pretreatment also was performed in the presence or absence of the β-AR antagonist alprenolol (1 μM). BN17 cells were also pretreated with vehicle, 1 μM ISO or 1 μM ISO in the presence of 0.3 μM NE for 24 h. For the antisense ODN pretreatment, BN17 cells were pretreated with 1 μM GRK3 antisense (5′-CCCGGTGTCTGCTTTCCT-3′) or sense (5′-AGGAAAGCAGACACCGGG-3′) ODN in serum-free media for 24 h followed by 24 h in the presence of serum, with or without the addition of 0.3 μM EPI (Dautzenberg et al., 2001; Thakker and Standifer, 2002).

α₂-AR and β₂-AR Agonist Concentration Response Curves. At the end of NE or EPI pretreatment, medium containing the drugs was aspirated, and the cells were harvested by pipetting fresh drug-free medium against the cells. Intact cells were sedimented by centrifugation at 1500g for 10 min, washed once with phosphate-buffered saline (PBS), and resuspended in Hanks' balanced salt solution (137 mM NaCl, 5 mM KCl, 0.6 mM Na₂HPO₄, 0.4 mM KH₂PO₄, 4 mM NaHCO₃, 6 mM d-glucose 6, 0.5 mM MgCl₂, 0.4 mM MgSO₄, and 1 mM CaCl₂) containing 0.1 mM ascorbate and the phosphodiesterase inhibitor isobutylmethylxanthine (0.5 mM). To determine the effect of α₂-AR agonist NE (10^-9-10^{-5 M}) on PGE₁-induced cAMP accumulation, intact cells were first incubated for 10 min at 37°C in Hanks' balanced salt solution buffer. Then, PGE₁ (10 nM), NE, and cells were added to assay tubes, and the tubes were incubated for an additional 10 min at 37°C. In all the experiments, the β-AR antagonist alprenolol (1 μM) also was included in this step. All assays were performed in duplicate in a total volume of 0.5 ml. The assay was terminated by removing the tubes to a boiling water bath for 5 min. After boiling, samples were centrifuged for 5 min at 14,000g, and cAMP levels from the supernatant fractions were determined in a [³H]cAMP (0.8 pmol) binding assay as described previously (Standifer et al., 1994). Preferential α₂-adrenoceptor agonists were not used in this study because NG108 cells express imidazole receptors, the activation of which inhibits cAMP accumulation (Greney et al., 2000). Because all preferential α₂-adrenoceptor agonists would activate both α₂ and imidazole receptors in NG108 cells, this would significantly complicate any data interpretation.

Membrane Preparation for Receptor Binding. To prepare membranes for receptor binding, the cells were first washed three times with PBS (pH 7.4) and then harvested from the plates by gentle scraping. The cells were sedimented by centrifugation at 3000g for 10 min. The cell pellet was then suspended in 10 volumes of Tris-HCl buffer (50 mM, pH 7.7) containing NaCl (100 mM), Na₂EDTA (10 mM), and phenylmethylsulfonyl fluoride (0.1 mM) and homogenized with a Polytron homogenizer (setting 5, 10 s). The membranes were incubated for 15 min at 25°C and sedimented by centrifugation (20,000 rpm) for 30 min at 4°C. The membranes were immediately used for the binding assay.

Radioligand Binding Assay to Determine Receptor Number. The α_2B-AR binding was performed with the α₂-AR antagonist, [³H]RX821002. For saturation studies, membranes (0.25-0.3 mg protein/ml) were incubated with [³H]RX821002 (0.1-30 nM) in potassium phosphate buffer (25 mM, pH 7.4) at 23°C for 30 min. Assays were performed in triplicate, and nonspecific binding was defined using 100 μM phentolamine. At the end of the incubation period, reaction was terminated by adding Tris-HCl buffer (50 mM, pH 8.0 at 4°C) and filtration over Whatman GF/B paper (Brandel Inc., Gaithersburg, MD). The filter paper was washed thrice with 3 to 4 ml of the filtration buffer (50 mM Tris-HCl, pH 8.0, 4°C). The amount of radioactivity in the sample was determined by scintillation spectroscopy in an LS6000 liquid scintillation counter (Beckman Coulter, Fullerton, CA).

Western Blot Analysis. Some of the cells collected either for the α_2B-AR response assay or receptor binding were used to prepare samples for Western blot analysis. Cell pellets were washed once with 1× PBS buffer (pH 7.4), lysed immediately in 100 to 150 μl of lysis buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 2 mM EDTA, 1% Triton X-100, 0.1% SDS, 0.02% sodium azide, 100 μg/ml phenylmethylsulfonyl fluoride, 1 μg/ml leupeptin, 1 μg/ml aprotinin, and 1 μg/ml pepstatin), vortexed, and digested for 30 min in an ice bath. The resultant cell lysate was diluted with 2× Laemmli gel loading buffer (50 mM Tris-HCl, pH 6.8, 10% glycerol, 2% SDS, and 0.1 mg/ml bromphenol blue). The proteins in the cell lysate were separated by SDS-polyacrylamide gel electrophoresis on a 10% polyacrylamide gel. The separated proteins were electrophoretically transferred to a polyvinylidene difluoride membrane (Amersham Biosciences UK, Ltd.) using a Mini Trans-Blot transfer gel unit (Bio-Rad). The polyvinylidene difluoride membrane was then incubated for 1 h in Tris-buffered saline, pH 7.6 (TBS-T; 20 mM Tris-HCl, 137 mM NaCl, and 0.1% Tween 20) containing 5% nonfat milk protein. After removal of blocking buffer, the membrane was washed with TBS-T and incubated overnight at 4°C with rabbit polyclonal antisera (primary antibody) directed against either G_i2α (1:1000; Upstate Biotechnology, Lake Placid, NY), GRK2, or GRK3 (1:1000; Santa Cruz Biotechnology, Inc.) in 2.5% nonfat milk. The blots were then washed with TBS-T to remove the primary antibody and incubated with an anti-rabbit horseradish peroxidase-conjugated secondary antibody (1:2500; Santa Cruz Biotechnology, Inc.) at room temperature for 45 min (G_i2α) or 1 h 30 min (GRK2/3) in 2.5% nonfat milk. Levels of GAPDH protein served as loading control. Therefore, the same blot was stripped with 62.5 mM Tris-HCl, pH 6.8, 2% SDS, 100 mM 2-mercaptoethanol in TBS-T, blocked, and reprobed by incubating either for 1 h at room temperature or overnight at 4°C with mouse anti-rabbit GAPDH antibody (1:8000; Research Diagnostics) in 2.5% nonfat milk. After removing the primary antibody, the blot was incubated with an anti-mouse horseradish peroxidase (1:10,000; Santa Cruz Biotechnology, Inc.) at room temperature for 45 min in 2.5% nonfat milk. Immunoreactive bands were visualized by enhanced chemiluminescence (Santa Cruz Biotechnology, Inc.). The intensity of each immunoreactive band was determined using either a FluorChem imaging system (Alpha Innotech Corp., San Leandro, CA) or Nucleovision Imaging Workstation (Nucleotech Corp., San Carlos, CA) and normalized to the GAPDH loading control.

Protein Estimation. Protein concentrations were determined by the Lowry method (Lowry et al., 1951).

Data Analysis.K_d, B_max, EC₅₀, and maximal response to the α₂-AR agonist was determined by nonlinear regression analysis using GraphPad Prism version 3.0 (GraphPad Software, Inc., San Diego, CA). For comparison between groups, the values were expressed as mean ± S.E.M. Comparisons between groups were made either by Student's t test or one way ANOVA followed by Tukey's post hoc test where appropriate (GraphPad Software Inc., San Diego, CA), and groups were considered significantly different if p < 0.05.

Results

In NG108 cells that express only α_2B-ARs, 24-h pretreatment with 20 μM NE was required to desensitize the α₂-AR-mediated inhibition of cAMP accumulation (Fig. 1A); 10 μM NE was insufficient to produce desensitization of the α_2B-ARs. Similarly, 20 μM but not 10 μM, EPI produced α₂-adrenoceptor desensitization (Fig. 1B). Neither NE nor EPI, at concentrations that resulted in α₂-adrenoceptor desensitization, produced any changes in the cellular expression of GRK2 or GRK3 (Fig. 2). However the desensitization was accompanied by down-regulation of specific α_2B-AR binding (Fig. 3).

In BN17 cells that express both α_2B- and β₂-ARs, 24-h pretreatment with only 0.3 μM EPI resulted in desensitization of the α_2B-ARs (Fig. 4). This concentration was 67-fold lower than the minimal concentration of EPI that resulted in α_2B-AR desensitization in NG108 cells. Pretreatment with 0.1 μM EPI had no significant effect on α_2B-AR responsiveness. Also, in BN17 cells, the concentration of NE (10 μM NE) required to desensitize the α_2B-AR (Fig. 5) was reduced by only 2-fold compared with the NG108 cells. More importantly, the effect of EPI or NE on α_2B-AR response was antagonized by the presence of the β-AR antagonist alprenolol (1 μM) during the EPI or NE pretreatment (Fig. 6).

In addition to desensitization of the α_2B-AR, pretreatment of BN17 cells with 0.3 μM EPI also resulted in down-regulation of the α_2B-ARs that was dependent on coactivation of the α_2B- and the β₂-AR (Fig. 7A). The pretreatment did not change the affinity of [³H]RX821002 for the α_2B-AR (Veh, 3.2 ± 0.8 nM; 0.3 μM NE, 3.6 ± 0.5 nM; 0.3 μM EPI, 3.2 ± 0.7 nM; vehicle + 1 μM alprenolol (Alp), 3.1 ± 0.4 nM; and 0.3 μM EPI + 1 μM Alp, 2.8 ± 0.3 nM). Similarly, pretreatment of BN17 cells with 10 μM NE resulted in a β₂-AR-dependent down-regulation of the α_2B-AR (Fig. 7B).

The effects of EPI and NE pretreatments on cellular levels of GRK3 also were different in BN17 cells, compared with NG108 cells. In BN17 cells, pretreatment with EPI or NE that resulted in α_2B-AR desensitization and down-regulation also produced a selective approximately 2-fold up-regulation of GRK3 (Fig. 8A). This increase in GRK3 expression also was prevented when the pretreatment was done in the presence of 1 μM alprenolol (Fig. 8A). No effect on GRK2 expression was observed with either treatment (Fig. 8B).

Because desensitization and down-regulation of the α_2B-ARs in BN17 cells were accompanied by a selective up-regulation of GRK3 expression, and because all of these effects were prevented by β-AR blockade with alprenolol, we determined whether α_2B-AR desensitization and down-regulation in BN17 cells was dependent upon GRK3 up-regulation. For this purpose, we performed the EPI pretreatment in cells treated with GRK3 antisense ODN to prevent the up-regulation of GRK3. Treatment with EPI in the presence of GRK3 antisense ODN prevented desensitization (Fig. 9A) and down-regulation (Fig. 9B) of the α_2B-AR, as well as prevented GRK3 up-regulation (Fig. 10A). Pretreatment with GRK3 antisense ODN or GRK3 sense ODN alone did not affect the acute α_2B-AR response in the BN17 cells (data not shown; see Fig. 9 legend). The GRK3 antisense ODN treatment also did not alter basal GRK3 (Fig. 10A) or GRK2 expression (Fig. 10B).

Finally, because desensitization and down-regulation of the α_2B-AR at low agonist concentrations was dependent upon β₂-AR activation in BN17 cells, we determined whether β₂-AR activation alone was sufficient to produce α_2B-AR desensitization by pretreating for 24 h with isoproterenol (1 μM). Isoproterenol pretreatment had no effect either on α_2B-AR responsiveness (Fig. 11A) or level (Fig. 11B) in BN17 cells, suggesting that simultaneous activation of the α_2B- and β₂-AR is necessary for α₂-AR desensitization at lower agonist concentrations in BN17 cells. Treating BN17 cells with 1 μM isoproterenol in the presence of 0.3 μM NE further tested this requirement of simultaneous activation of the α_2B- and β₂-AR. The NE and isoproterenol cotreatment desensitized (Fig. 11A) and down-regulated (Fig. 11B) the α_2B-AR. Also, pretreatment of the BN17 cells with 1 μM isoproterenol did not alter the cellular levels of either GRK2 or GRK3, but the cotreatment with 1 μM isoproterenol and 0.3 μM NE selectively increased the level of GRK3 (data not shown).

Discussion

In the present study, we demonstrate that simultaneous activation of the α_2B- and β₂-adrenoceptors results in down-regulation of the α_2B-adrenoceptor by EPI at concentrations ∼70-fold lower than those required when the β₂-adrenoceptor is not activated. The mechanism responsible for this effect of EPI at lower agonist concentrations is a selective up-regulation of GRK3 expression that requires simultaneous α_2B- and β₂-adrenoceptor activation. Prevention of GRK3 up-regulation by β-adrenoceptor blockade or GRK3 antisense ODN treatment during the EPI pretreatment eliminates the α_2B-adrenoceptor down-regulation at low EPI concentrations (0.3 μM). Without simultaneous activation of both α_2B- and β₂-adrenoceptors, EPI and NE still produced α_2B-adrenoceptor down-regulation, but significantly higher agonist concentrations (20 μM) are required.

The results of this study reveal an intriguing difference between the threshold concentrations of the physiological agonists NE and EPI required to induce α_2B-adrenoceptor down-regulation. This result is surprising in the context of the potency and efficacy of NE and EPI at the α₂-adrenoceptor. It has repeatedly been reported that these agonists are equipotent and equiefficacious at this receptor (Jasper et al., 1998). Therefore, one would not expect the observed difference in the minimal concentration of EPI versus NE required to induce down-regulation. However, the contribution of the β₂-adrenoceptor to this process immediately reveals the reason for this difference. In BN17 cells expressing both α_2B- and β₂-adrenoceptors, 0.3 μM EPI or 10 μM NE was the minimal concentration required to produce α_2B-adrenoceptor down-regulation, whereas in NG108 cells that express only the α_2B-adrenoceptor, the minimal concentration required to produce down-regulation was the same for NE or EPI, 20 μM. The difference between the minimal concentration of NE versus EPI required in BN17 cells is 33-fold, similar to the 50- to 70-fold greater potency for EPI versus NE at the β₂-adrenoceptor (Lands et al., 1967). Most likely, the greater degree of α_2B-adrenoceptor activation produced by NE at 10 μM compared with 0.3 μM, combined with the β_2B-adrenoceptor activation by NE at 10 μM, enabled NE to down-regulate the α_2B-adrenoceptor in BN17 cells at a slightly lower concentration than that required in NG108 cells. Therefore, the lower threshold concentration of EPI versus NE required to produce α_2B-adrenoceptor down-regulation, although surprising, is in good agreement with the pharmacology of EPI versus NE at the β₂-adrenoceptor.

The present study also clearly defines the role of the β₂-adrenoceptor in the down-regulation of the α_2B-adrenoceptor, i.e., to contribute to up-regulation of GRK3. However, it is important to recognize that neither α₂- nor β₂-adrenoceptor activation alone produces GKR3 up-regulation. First, α₂-adrenoceptor activation by NE at low concentrations in BN17 cells or by higher concentrations of NE or EPI in NG108 cells does not produce GRK3 up-regulation. Second, isoproterenol at concentrations that cause similar degrees of β₂-adrenoceptor activation as 0.3 μM EPI does not cause up-regulation of GKR3. Finally, up-regulation of GRK3 by EPI in BN17 cells is prevented by β-adrenoceptor blockade. Hence, simultaneous activation of both the α₂- and β₂-adrenoceptors was required for GRK3 up-regulation. The mechanisms responsible for GRK3 up-regulation were not studied further because this was not the focus of the present project. However, in a previous study of α_2A-adrenoceptor regulation, it was observed that activation of p42/44 MAP kinase by simultaneous α₂- and β₂-adrenoceptor activation was required for GRK3 up-regulation and that inhibition of protein synthesis or GRK3 antisense ODN prevented GRK3 up-regulation by EPI (K. M. Standifer, personal communication). Although these observations are suggestive of a regulation of GRK3 expression by a mitogen-activated protein kinase substrate, this is pure speculation because very limited information is available about the regulation of the expression of GRK3 or GRK2 (Penela et al., 2003).

Another somewhat surprising component of the results in the present study is the critical role of GRK3 in enabling EPI to produce α₂-adrenoceptor down-regulation at significantly lower agonist concentration. A role of GRKs in the short-term desensitization of the α_2B-adrenoceptor is clearly suggested by the work of Liggett and coworkers. These investigators have reported GRK-mediated phosphorylation of the α_2B-adrenoceptor and correlated this phosphorylation with desensitization of agonist-mediated inhibition of cAMP accumulation (Jewell-Motz and Liggett, 1995). Moreover, they have reported that mutations within the third intracellular loop of this receptor not only reduce GRK-mediated phosphorylation but also eliminate short-term agonist-induced desensitization of the α_2B-adrenoceptor. The physiological importance of this phosphorylation is suggested by recent reports of a polymorphism of the human α_2B-adrenoceptor that exhibits reduced agonist-induced phosphorylation and desensitization (Small et al., 2001) and is associated with an increased risk of myocardial infarction in human (Snapir et al., 2003). However, the role of GRK mediated phosphorylation in long-term regulation of the α_2B-adrenoceptor is more tenuous. For example, site-directed mutagenesis of the third intracellular loop of the α_2B-adrenoceptor, which reduced phosphorylation and eliminated short-term agonist-induced desensitization, was reported to have no effect on long-term agonist-induced receptor down-regulation (Jewell-Motz and Liggett, 1995). Nevertheless, the results of the present study support a critical role of GRK3 in α_2B-adrenoceptor down-regulation.

The important role of the GRKs in α_2B-adrenoceptor down-regulation suggested by the present study is in agreement with several recent reports regarding the role of GRKs in the down-regulation of other G protein-coupled receptors (GPCRs). For example, the δ-opioid receptor has been shown to undergo down-regulation in N18TG2 cells after exposure to etorpine in a GRK-dependent manner (Shapira et al., 2001). Similarly, the human κ-opioid receptor has been shown to undergo down-regulation in a GRK2-dependent manner in Chinese hamster ovary cells (Li et al., 2000). The expression of the metabotropic glutamate receptor 5 also was reported to be regulated by GRK2 (Sorensen and Conn, 2003). This effect of GRK2 requires the presence of a threonine residue (Thr 840) in the carboxyl terminal tail of metabotropic glutamate receptor 5, suggesting a role for GRK-mediated phosphorylation in down-regulating the receptor. Finally, a recent report suggests that GRK2 overexpression reduces the threshold agonist concentration required for down-regulation of the human muscarinic M2 receptor (Tsuga et al., 1998). Therefore, there is considerable evidence that down-regulation of several GPCRs is regulated by the GRKs.

The role of β₂-adrenoceptor mediated GRK3 up-regulation in enabling α_2B-adrenoceptor down-regulation by lower EPI concentrations highlights an emerging theme in receptor signal modulation, i.e., heterologous modulation of homologous receptor regulation (Chuang et al., 1996). Phosphorylation of GPCRs by GRKs requires an agonist-occupied receptor. For this reason, mechanisms involving GRKs were initially considered as strictly homologous. However, our results clearly demonstrate that, by contributing to the regulation of GRK3 expression, β₂-adrenoceptors heterologously modulate α_2B-adrenoceptor down-regulation. These observations are similar to our previous report where α_2A-adrenoceptor regulation by simultaneous α_2A- and β₂-adrenoceptor activation caused selective GRK3 up-regulation and α_2A-adrenoceptor down-regulation at low EPI concentrations (Bawa et al., 2003). As discussed in a recent review, accumulating evidence suggests that heterologous regulation of GRK expression may play an important role in the heterologous regulation of GPCR signaling (Penela et al., 2003). In the context of the present study, it is particularly important to note that this heterologous regulation allows long-term agonist-induced receptor down-regulation and desensitization to occur at significantly lower agonist concentrations than would be required for homologous regulation. Considering that the α₂-adrenoceptor generally requires micromolar agonist concentrations to produce desensitization compared with the β₂-adrenoceptor (Atkinson and Minneman, 1992), and considering the many physiologically important sites where α₂- and β₂-adrenoceptors are colocalized on cells such as vascular and uterine smooth muscle, fat cells, pancreatic β cells, and sympathetic nerve terminals to name a few, such heterologous regulation has the potential to significantly modify long-term regulation of many important functions.

Several investigators previously have reported heterologous regulation of α₂-adrenoceptor numbers. In HT29 cells, stimulation of cAMP accumulation led to an increase in the number of α₂-adrenoceptors (Sakaue and Hoffman, 1991). This increase was shown to be the result of an increase in gene transcription. By contrast, in cultured astroglia cells, increases in cellular cAMP levels or PKC activation was reported to cause a decrease in the steady-state levels of α_2A-adrenoceptor mRNA levels by decreasing gene transcription (Reutter et al., 1998). EPI also produced down-regulation of the α₂-adrenoceptor in these cells by a mechanism that involved a decrease in receptor transcription. However, the changes in α₂-adrenoceptor number observed in the present study seem to have no contribution from changes in cAMP. First, GRK3 antisense, which had no effect on adrenoceptor signaling, abolished EPI-induced α₂-adrenoceptor down-regulation. Second, isoproterenol at concentrations that produced β₂-adrenoceptor activation similar to that of EPI, and increased cAMP concentrations in BN17 cells, had no effect of α₂-adrenoceptor number. Therefore, the data in the present study are unique in that they emphasize a role for GRK3 expression in this heterologous regulation.

In summary, heterologous up-regulation of GRK3 by simultaneous activation of α_2B- and β₂-adrenoceptors enables the α_2B-adrenoceptor to be down-regulated by significantly lower concentrations of EPI compared with NE. Future studies will focus on the role of GRKs in α₂-adrenoceptor down-regulation and the influence of cellular levels of GRK expression on the threshold agonist concentrations required for α₂-adrenoceptor down-regulation.