Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Thromboxane A₂ induces vasoconstriction, mitogenesis, and platelet aggregation (Hamburg et al., 1975; Hanasaki et al., 1990). It is biosynthesized from arachidonic acid through the consecutive actions of cyclooxygenase and thromboxane synthase (Shen and Tai, 1986). Its actions are initiated by its interaction with specific receptors called thromboxane receptors (TPs). Although studies on various structural ligands suggest the existence of different subtypes of TPs (Mais et al., 1985), a single gene encoding a member of the seven transmembrane G protein-coupled receptor (GPCR) family has been cloned (Hirata et al., 1991). However, two different isoforms of TP have been recognized as a result of alternate splicing. The first isoform, TP, was cloned initially from human placenta cells (Hirata et al., 1991) and subsequently from human cell lines (Angelo et al., 1994; Allan et al., 1996). The human TP cDNA encodes a protein of 343 amino acids with a calculated mol. wt. of 37.5 kDa. However, purified expressed receptor exhibited a protein band of 55 to 57 kDa, indicating that TP is a glycoprotein (Ushikubi et al., 1989; Borg et al., 1994). The second isoform, TP, was first cloned from human endothelial cells (Raychowdhury et al., 1994). The human TP cDNA encodes a protein of 406 amino acids, which is identical to that reported for TP for the first 328 amino acids. The two isoforms expressed in cultured cells show similar ligand-binding characteristics and phospholipase-C activation but regulate adenylyl cyclase activity in an opposite fashion (Hirata et al., 1996). TP activates adenylyl cyclase, whereas TP inhibits it.

One of the early regulatory mechanisms that is activated after GPCR stimulation is receptor-homologous desensitization (Chuang et al., 1996). This is an agonist-dependent adaptive process in biological systems that modulates responsiveness of the cell to repeated stimuli over time. TPs have been reported to undergo homologous desensitization after stimulation by agonists such as U-46619 and I-BOP (Liel et al., 1988; Murray and FitzGerald, 1989; Okwu et al., 1992). However, the molecular mechanisms that underlie TP-homologous desensitization remain to be elucidated. The general concept is that agonist-induced phosphorylation of the intracellular domains of the receptor results in uncoupling to the G proteins and the impairment of the receptor function (Freedman and Lefkowitz, 1996). Two types of kinases are known to mediate these modifications: second-messenger kinases and G protein-coupled receptor kinases (GRKs). In the former group, protein kinase C (PKC) has been suggested to be involved in agonist-induced phosphorylation of TP in some studies (Spurney and Coffman, 1997; Spurney, 1998) but is only minimally involved in other studies (Habib et al., 1997). The discrepancy seems to be due to the difference in the effect of PKC inhibitors on agonist-induced phosphorylation of TP. In the second group, GRK's involvement in agonist-induced phosphorylation of TP has been suggested but not directly demonstrated (Parent et al., 1999). Our studies as described in this article represent a direct demonstration of the effect of GRKs on TP phosphorylation and function.

GRKs are a family of novel protein kinases (Premont et al., 1995). These protein kinases have the unique ability to recognize and phosphorylate their GPCR substrates only in their active (i.e., agonist-occupied) conformations. There are six cloned members in this family. These include rhodopsin kinase (GRK1), -adrenergic receptor kinase 1 and 2 (GRK2 and GRK3, respectively), GRK4, GRK5, and GRK6. The significance of GRK-mediated phosphorylation for the regulation of receptor function under intact cell conditions has been demonstrated only in a few receptor systems (Freedman et al., 1995; Diviani et al., 1996; Neuschafer-Rube et al., 1999). The present study was designed to determine whether GRKs can phosphorylate and desensitize TP. To this end, we developed an inexpensive and efficient method of isolating phosphorylated receptors using metal ion chelation chromatography of His-tagged TP stably expressed in human embryonic kidney (HEK) 293 cells. GRK5 and GRK6 were chosen for this study because of their widespread tissue expression.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. The pcDNA3 expression vector was from Invitrogen (Carlsbad, CA). PGEX-2T plasmid vector was from Amersham Pharmacia Biotech (Piscataway, NJ). T4 DNA ligase, Vent DNA polymerase, BamHI, and XhoI were from New England Biolabs, Inc. (Beverly, MA). Pfu DNA polymerase was from Stratagene (La Jolla, CA). HEK293 cells were obtained from American Type Culture Collection (Manassas, VA). Taq DNA polymerase, heat-inactivated fetal bovine serum (FBS), antibiotic-antimycotic, and geneticin selective antibiotic (G418) were from Invitrogen. The QIAprep Spin Plasmid Miniprep Kit, QIAquick PCR Purification Kit, QIA Quick Gel Extraction Kit, and Effectene Transfection Reagent were obtained from QIAGEN (Valencia, CA). [³H]SQ29,548 (specific activity, 48 Ci/mmol) was from PerkinElmer Life Science (Boston, MA). I-BOP, U-46619, and SQ29,548 were from Cayman Chemical (Ann Arbor, MI). ³²P_i (500 mCi/ml) and [-³²P]ATP (10 Ci/mmol) were from ICN (Costa Mesa, CA). Fura-2/AM, staurosporine, Ro-31-8220, and recombinant human protein kinase C were purchased from Calbiochem-Novabiochem Corporation (San Diego, CA). GF 109203X and calphostin C were obtained from Alexis Corporation (Läufelfingen, Switzerland). Oligonucleotide primers were synthesized by Integrated DNA Technologies (Coralville, IA). Dulbecco's modified Eagle's medium (DMEM), Ni²⁺-NTA agarose, GSH agarose, phenylmethylsulfonyl fluoride (PMSF), phorbol 12-myristate 13-acetate (PMA), and other chemicals were from Sigma (St. Louis, MO). The cDNAs encoding human GRK5 and human GRK6 were kindly provided by Dr. Jeffrey Benovic of Jefferson University. The cDNA encoding GRK6-RDD (R215K, D484S, and D485T) was kindly supplied by Dr. Francois Boulay of the DBMS/Laboratoire de Biochimie et de Biophysique des Systemes Integres, France (Milcent et al., 1999).

Subcloning of C-Terminal His-Tagged Human TP cDNA. Human TP cDNA was amplified by PCR using Vent DNA polymerase as described (Chiang et al., 1996). The oligonucleotides used for PCR were designed on the basis of the amino terminal sequence (5'-CGGGATCCATGTGGCCCAACGGCAGTTC-3') and carboxyl terminal sequence (5'-ATAGAATTCTCAATGGTGATGGTGATGGTGCTGCAGCCCGGAGC-3') with extra BamHI and EcoRI sites on the ends, respectively, and six histidinyl residues on the C terminus. The PCR product was subcloned into the mammalian expression vector pcDNA3 at BamHI and EcoRI sites. The insertion of the TP-(His)₆ cDNA was confirmed by DNA sequencing.

Expression of the His-Tagged Human TP and GRKs in HEK293 Cells. HEK293 cells were grown in 90% DMEM supplemented with 10% heat-inactivated FBS, gentamicin, and antibiotic-antimycotic at 37°C in a humidified atmosphere of 95% air and 5% CO₂. Cells were subcultured every 3 days after confluence by use of 0.25% trypsin with 1 mM EDTA and plated at a density of 1 × 10⁵ cells/ml.

Preparation of HEK293 Cell Membranes. Confluent cultures of HEK293 cells were incubated with 0.25% trypsin and 1 mM EDTA. After detachment, cells were immediately centrifuged at 1000g for 5 min. The cell pellet from the 1 × 10⁷ cells was washed with 2 ml of phosphate-buffered saline (PBS) and then resuspended in 1 ml of homogenization buffer (10 mM Tris-HCl, pH 7.4, 50 µg/ml PMSF). The mixture was then sonicated four times for 10 s each with an ultrasonicator at a setting of 4. The homogenate was centrifuged at 100,000g for 30 min at 4°C. The pellet was resuspended in 1 ml of ice-cold binding buffer.

Preparation of GRK Cell Lysate. HEK293 cells (1 × 10⁷) transfected with pcDNA3-GRK5 or pcDNA3-GRK6 were homogenized by sonication in 0.5 ml of lysis buffer II (20 mM Hepes, pH 7.4, 1% Triton X-100, 0.15 M NaCl, 10 mM EDTA, 1 mM PMSF, and 10 µg/ml leupeptin) as described above. After centrifugation at 100,000g for 30 min, the cytosol (20 µg) was used for the kinase reaction.

Ligand Binding Assay. Ligand binding assay was conducted in 50 mM Tris-HCl (pH 7.4) buffer with 5 mM CaCl₂ as described (Chiang et al., 1996). For each assay, 100 µg of cell membrane fraction was incubated with 3 nM [³H]SQ29,548 in a 100-µl reaction volume at room temperature for 60 min. The reaction was terminated with the addition of 1 ml of ice-cold washing buffer (50 mM Tris-HCl and 150 mM NaCl, pH 7.4). The solution was filtered under vacuum through a GF/C glass filter (Whatman, Clifton, NJ), and the filter was then washed with 10 ml of washing buffer. The radioactivity retained on the filter was counted in 10 ml of scintillation cocktail. The nonspecific binding was determined by adding 10 µM unlabeled SQ29,548.

Measurement of Intracellular Calcium Release. Intracellular calcium release was measured by fluorescence excitation of cells loaded with the fluorescent probe Fura-2/AM as described (Woolkalis et al., 1995). Briefly, the HEK293 cells stably transfected with TP were collected and washed with Fura-2 assay buffer (4.3 mM Na₂HPO₄, 24.3 mM NaH₂PO₄, 4.3 mM K₂HPO₄, pH 6.8, 113 mM NaCl, 5 mM D-glucose, 1 mM CaCl₂, and 0.5% bovine serum albumin). The cells were resuspended in the same buffer at a concentration of 1 × 10⁶ cells/ml. The suspension was incubated with 10 µM Fura-2/AM at 37°C for 1 h, concentrated into pellet form by centrifugation at 1,000g for 10 min, and washed with the Fura-2 assay buffer, but without Ca²⁺, twice. The loaded cells were resuspended at approximately 5 × 10⁶ cells/ml in the Fura-2 assay buffer. The Ca²⁺ signal induced by TP agonist I-BOP was examined by Fura-2 fluorescence in an F-2000 fluorescence spectrophotometer (Hitachi Software Engineering, Yokohama, Japan) at 37°C with excitation and emission wavelengths at 340 and 510 nm, respectively. The reaction was initiated by adding I-BOP to a final concentration of 100 nM.

Western Blotting Analysis. The cell membranes (50 µg) prepared from HEK293 cells expressing the wild type, and the His-tagged receptors were subjected to SDS-PAGE on 10% polyacrylamide gel. The proteins were then transferred electrophoretically onto PVDF membranes. The membrane was blocked with 5% nonfat milk in 30 mM Tris-HCl, pH 7.4, containing 120 mM NaCl and 0.05% Tween-20 (TBST) at room temperature for 1 h. It was then incubated for 2 h at room temperature with a rabbit anti-serum against the N-terminal sequence of the TP in TBST with 5% nonfat milk (1:400 dilution), followed by incubation with horseradish peroxidase-linked protein A (1:5000 dilution in TBST with 5% nonfat milk) for 1 h at room temperature. The membrane was washed with TBST six times. The immunoreactive bands were detected with ECL⁺ Plus Western blotting detection system.

Expression of the Intracellular Loop (IL)-Specific Peptides of the TP as Glutathione S-Transferase (GST) Fusion Proteins. The amino acid sequences of each IL are shown in Table 1. The DNA sequences encoding the three ILs and C-terminal tail were generated by PCR using native Pfu DNA polymerase and pBacpak8 TP plasmid DNA as templates (Chiang et al., 1996). PCR amplification was carried out in 50 µl of volume, with 0.2 mM dNTP, 100 ng of each primer, 100 ng of pBacPak8 TP plasmid DNA, and 1.25 units Pfu DNA polymerase with cycle conditions of 94°C, 1 min; 60°C, 1 min; and 72°C, 1 min, for 35 cycles. Sequences of the oligonucleotides with extra BamHI or EcoRI site are: I IL-F, 5'-GTTAGGATCCATGCGGCAGGGGGGTT-3'; I IL-B, 5'-TGCAGAATTCCTAGAAGGTGAGGAAG-3'; II IL-F, 5'-GTCGGATCCATGTCAGAGC-GCTACCT-3'; II IL-B, 5'-ATAAGAATTCCTAGCGGCGCTGCGA-3'; III IL-F, 5'-GATCGGATCCATGGCCACCCTGTGCCA-3'; III IL-B, 5'-GCTCGAATTCTACTCCACCTCGGAG-3'; C-Tail-F, 5'-TATAGGATCCATGCGCCGCGCCGTGCT-3'; C-Tail-B, 5'-TATTGAATTCCTACTGCAGCCCGGAG-3'.

Preparation of Antiserum against TP N-Terminal Sequence. The DNA sequence encoding the N-terminal peptide (MWPNGSSLGPCFRPTNITLEERRLIASPW) of human TP was isolated by PCR using Taq DNA polymerase and subcloned into GST fusion protein expression vector pGEX-2T at BamHI and EcoRI sites. The GST fusion protein was expressed in E. coli BL21, a protease-deficient strain and purified by GSH agarose affinity chromatography as described (Smith and Johnson, 1988). The purified GST fusion protein was used to immunize the rabbit twice a month. The rabbit was bled through the ear veins 7 days after each boost. The titer of the antibody was determined by an enzyme-linked immunosorbent assay.

In Vitro Phosphorylation of the Intracellular Domains of the TP. In vitro phosphorylation experiments with the intracellular domain-specific peptides of the TP were performed as described (Diviani et al., 1996). Reactions were performed in 50 µl of 20 mM Tris-HCl buffer (pH 7.4) containing 10 mM MgCl₂ and 2 mM EDTA, 2 µg of GST fusion peptide, and 20 µg of crude GRK. The reaction was initiated by the addition of [-³²P]ATP (0.1 mM, 1000 cpm/pmol). After incubation for 30 min at 30°C, GSH agarose (20 µl of 50% suspension) was added and the incubation was allowed to continue for another 20 min. The GSH agarose was precipitated by centrifugation and washed six times with buffer (1 ml each). The GSH agarose was treated with SDS-PAGE loading buffer, boiled, and subjected to 10% SDS-PAGE. The ³²P-labeled peptide was detected by autoradiography.

In Vivo Phosphorylation of the TP. HEK293 cells (1 × 10⁶/ml) stably expressing His-tagged TP were washed once with phosphate-free DMEM and labeled with ³²P_i (100 µCi/ml) in the phosphate-free medium for 1.5 h at 37°C. Then cells were exposed to buffer alone, U46619 (1 µM), or I-BOP (100 nM) for 10 min. Reactions were terminated by the addition of ice-cold PBS buffer. After washing with the ice-cold PBS buffer three times, cells were scraped in lysis buffer (50 mM Tris-HCl, pH 7.4, 0.15 M NaCl, 5 mM EDTA, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 10 mM NaF, 10 mM sodium pyrophosphate, 10 µg/ml leupeptin, 10 µg of soybean trypsin inhibitor, 1 mM benzamidine, and 0.5 mM PMSF) and incubated for 30 min at 0°C. Cell debris was removed by centrifugation at 10,000g for 15 min. The His-tagged TP was isolated by adding 10 µl of 50% Ni²⁺-NTA agarose beads. After washing with lysis buffer for six times, the beads were suspended in SDS-loading buffer, boiled, and subjected to 10% SDS-PAGE. The ³²P-labeled receptor was detected by autoradiography.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

HEK293 cells were stably transfected with pcDNA3 encoding the wild-type receptor, the His-tagged receptor, or the vector alone. Successful expression of the wild-type and the His-tagged receptors in HEK293 cells were demonstrated by Western blotting as shown in Fig. 1. A distinct band of 55 kDa, an average size of the glycosylated human TP, was observed. The band appeared to be not as diffused as seen in other studies probably because of glycosylation to a similar degree. However, no detectable expression of the TP was found in HEK293 cells transfected with pcDNA3 vector alone. To further demonstrate that the recombinant receptors are functionally active, both ligand binding and calcium mobilization induced by an agonist were examined. Table 2 indicates that both recombinant wild-type and His-tagged TP receptors have comparable K_d and B_max values for antagonist ligand [³H]SQ29,548. Figure 2A shows that agonist I-BOP at 100 nM induced comparable Ca²⁺ signal in Fura-2/AM-loaded HEK293 cells expressing either wild-type or His-tagged receptor. Again, I-BOP did not induce any response in cells transfected with pcDNA3 alone. To ensure that the recombinant receptor still undergoes agonist-induced desensitization, cells expressing His-tagged receptor were pretreated with 100-nM I-BOP for 10 min. After washing twice, cells were challenged with 100-nM I-BOP again. Figure 2B indicates that cells pretreated with a vehicle responded normally with I-BOP challenge (a), whereas cells pretreated with I-BOP did not respond to a second addition of I-BOP (b), indicating that the I-BOP-pretreated cells had been desensitized.

View larger version (48K): [in this window] [in a new window]   Fig. 1.   Western blotting assay of the wild-type and His-tagged TP stably expressed in HEK293 cells. The cell membranes (100 µg) prepared from HEK293 cells stably transfected with pcDNA3 encoding the wild-type receptor or the His-tagged receptor or the pcDNA vector alone were respectively subjected to 10% SDS-PAGE and transferred onto PVDF membrane as described under Experimental Procedures. The immunoreactive band was detected by antiserum against N-terminal sequence of human TP at a dilution of 1:500. A representative Western blot analysis of two qualitatively similar results is shown.

View this table: [in this window] [in a new window]   TABLE 2 Kd and Bmax values of wild-type and C-terminal His-tagged TP expressed in HEK293 cells The HEK293 cell membranes (100 µg) stably transfected with the wild-type and C-terminal His-tagged TP were incubated with various concentrations of [3H]SQ29,548. The nonspecific binding was determined in the presence of 1 µM unlabeled SQ29,548. The Kd and Bmax values were determined by Scatchard Plot using GraphPad software. A representative table of two qualitatively similar results was shown.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   I-BOP-induced Ca2+ mobilization and desensitization in the HEK293 cells expressing TP. A, HEK293 cells were stably transfected with pcDNA3 vector, pcDNA3-TP, or pcDNA3-TP-(His)6. Cells (1 × 106) were respectively treated with 100 nM I-BOP, and the intracellular Ca2+ signaling was examined. B, HEK293 cells were stably transfected with pcDNA-TP-(His)6. Cells (1 × 106) were respectively treated with vehicle (a) and 100 nM I-BOP for 10 min (b). After washing twice with Fura-2 assay buffer, cells were treated with 100 nM I-BOP, and the intracellular Ca2+ signaling was examined as described under Experimental Procedures. A representative experiment of three qualitatively similar results is shown.

Using these well characterized HEK293 cells expressing recombinant His-tagged TP, we explored the phosphorylation of the receptor induced by various agents. Cells were prelabeled with ³²P_i and challenged with I-BOP. The ³²P-labeled receptor was isolated by Ni²⁺-NTA agarose followed by SDS-PAGE analysis. Figure 3A shows the time course of I-BOP-induced phosphorylation of the receptor. There was an agonist-induced time-dependent increase in phosphorylation since spontaneous phosphorylation was significantly less in the absence of an agonist. A distinct band of spontaneous phosphorylation was observed at 60 min probably because of endogenous kinase-mediated phosphorylation after prolonged incubation. Figure 3B shows a concentration-dependent phosphorylation of the receptor by I-BOP. Phosphorylation reached plateau at 100 nM I-BOP. Although the responsible kinase(s) involved in I-BOP-induced receptor phosphorylation is not clear, GRKs have been implicated as one of those kinase families involved. It was initiated to investigate whether any of those intracellular loops and C-terminal tail could be phosphorylated by GRKs in vitro. To facilitate the isolation of each intracellular domain after phosphorylation, each peptide was fused to GST and expressed as a GST fusion protein that could be readily purified by GSH-agarose affinity chromatography. The GST fusion proteins were separately used as substrates of GRK5 and GRK6 prepared from HEK293 cells overexpressing each respective GRK. Overexpression of GRKs was monitored by Western blot and was shown to be at a comparable level (data not shown). Figure 4 shows that endogenous kinase(s) (pcDNA3 alone) catalyzes the phosphorylation of GST-C-terminal tail fusion protein. A significant increase in the phosphorylation of the GST-C-terminal tail fusion protein was observed with both crude GRK5 and GRK6. Other intracellular domains were not found to be significantly phosphorylated by any GRK. Phosphorylation of TP by GRKs in intact cells was then carried out. HEK293 cells stably expressing the His-tagged TP were transiently transfected with pcDNA3 alone or with pcDNA3 encoding the cDNA of GRK5 or GRK6, respectively. The transfected cells were prelabeled with ³²P_i and then challenged with 100 nM I-BOP for 10 min. Figure 5A shows that I-BOP increased the phosphorylation of the receptor in the absence of the overexpression of any GRK, as demonstrated above. Transient overexpression of any GRK did not seem to increase significantly the phosphorylation of the receptor in the absence of I-BOP stimulation. However, a significant increase in the phosphorylation of the receptor was observed with overexpression of GRK5 or GRK6 in the presence of I-BOP stimulation. As a control, overexpression with an attenuated analog of GRK6, GRK6-RDD, resulted in significantly less receptor phosphorylation after stimulation with I-BOP. The time course of receptor phosphorylation was also examined in the presence and absence of GRK6 overexpression (Fig. 5B). The receptor phosphorylation followed a faster kinetics with GRK6 overexpression. The effect of GRK expression on I-BOP-induced Ca²⁺ release was further examined. HEK293 cells stably transfected with the wild-type TP were cotransfected with pcDNA3 harboring the cDNA of GRK5, GRK6, or no insert before stimulation with 100 nM I-BOP. The Ca²⁺ signal seemed to be attenuated by pcDNA3-GRKs transfected into cells expressing TP as shown in Fig. 6A. The concentration-dependent effect of pcDNA3-GRKs on the Ca²⁺ release induced by I-BOP is shown in Fig. 6B.

View larger version (57K): [in this window] [in a new window]   Fig. 3.   Time and dose-dependent phosphorylation of the HEK293 cells expressing TP-(His)6. HEK293 cells stably expressing His-tagged TP were metabolically labeled with 32Pi and stimulated with (A) 100 nM I-BOP for the indicated amount of time and (B) with the indicated concentrations of I-BOP for 10 min. His-tagged receptors were isolated by Ni2+-NTA agarose and separated on a 10% SDS-PAGE as described under Experimental Procedures. The autoradiography was then carried out. A representative autoradiogram of two qualitatively similar results was shown. The intensity of the autoradiogram was controlled by the length of film exposure. The film on time course studies was exposed for a shorter period of time to accommodate the clarity of the 60-min sample.

View larger version (43K): [in this window] [in a new window]   Fig. 4.   In vitro phosphorylation of cytoplasmic domains of TP by various GRKs. GST-IL fusion protein (2 µg) was used as a substrate to perform kinase reaction. The preparation of the crude GRKs and the kinase reaction conditions were as described under Experimental Procedures. 1, GST; 2, GST-TP first intracellular loop; 3, GST-TP second intracellular loop; 4, GST-TP third intracellular loop; 5, GST-TP-C-terminal tail. A representative autoradiogram of two qualitatively similar results is shown.

View larger version (40K): [in this window] [in a new window]   Fig. 5.   In vivo phosphorylation of TP by GRK5 and GRK6. A, HEK293 cells (1 × 106) stably expressing His-tagged TP were transiently transfected with pcDNA3, pcDNA3-GRK5, pcDNA3-GRK6, or a GRK6 mutant pcDNA3-GRK6-RDD, respectively. The transfected cells were metabolically labeled with 32Pi and stimulated in the absence and presence of 100 nM I-BOP. B, HEK293 cells (1 × 106) stably expressing His-tagged TP was transiently transfected with pcDNA3-GRK6. The transfected cells were metabolically labeled with 32Pi and stimulated with 100 nM I-BOP for the indicated amount of time. The phosphorylated receptors were isolated as described under Experimental Procedures. A representative autoradiogram of two qualitatively similar results is shown.

View larger version (12K): [in this window] [in a new window]   Fig. 6.   Inhibition of human TP-mediated intracellular Ca2+ release by coexpression of the GRKs. A, HEK293 cells (1 × 106) stably expressing TP were further transiently transfected with 0.5 µg of pcDNA3, pcDNA-GRK5, or pcDNA3-GRK6 for 2 days. Ca2+ release was monitored after the addition of 100 nM I-BOP in each case. B, HEK293 cells (1 × 106) stably expressing TP were transiently transfected with different amounts of pcDNA3-GRK5 or pcDNA3-GRK6 for 2 days. The total amount of plasmid DNA per transfection was kept constant (0.5 µg) by the addition of pcDNA3. Ca2+ release was induced by 100 nM I-BOP. Ca2+ release from cells transfected with pcDNA3 alone was taken as 100%. Ca2+ release from cells transfected with different amounts of pcDNA3-GRKs was calculated as a fraction of the release from the cells transfected with pcDNA3 alone. Ca2+ release was determined by the method described under Experimental Procedures. , GRK5; , GRK6

PKC was believed to be involved in agonist-induced phosphorylation of TP because of PKC inhibitor studies (Spurney and Coffman, 1997; Spurney, 1998). We repeated this experiment by examining the effect of a PKC inhibitor, GF 109203X, on I-BOP or PMA-induced phosphorylation of the receptor in HEK293 cells as shown in Fig. 7. Furthermore, we also used an antagonist, SQ29,548, to examine the authenticity of receptor action by I-BOP. As expected, SQ29,548 at 10 µM blocked completely I-BOP-induced receptor phosphorylation. GF 109203X at 5 µM also inhibited I-BOP-induced receptor phosphorylation to an extent comparable with the level of endogenous phosphorylation that occurred in the absence of agonist stimulation. Similar results were obtained by using other PKC inhibitors such as staurosporine, Ro-31-8220, and calphostin C at 1 µM (data not shown). Our results also suggest that PKC may be involved in I-BOP-induced receptor phosphorylation. Although PKC inhibitors have been used to explore the role of PKC in agonist-induced receptor phosphorylation, the specificity of these inhibitors toward other potential protein kinases involved in agonist-induced receptor phosphorylation has not been critically examined. Since GRKs augmented I-BOP-induced receptor phosphorylation, it was initiated to examine whether GF 109203X exhibited any effect on GRK-mediated I-BOP-induced receptor phosphorylation. Figure 8 shows that GF 109203X at 5 µM also inhibited GRK5- and GRK6-mediated I-BOP-induced receptor phosphorylation. These results indicate that GF 109203X is likely an inhibitor of GRK5 and GRK6. To examine whether GF 109203X is an inhibitor of GRK5 and GRK6, a GST-C-terminal tail fusion protein was used as a substrate. Figure 9 shows that crude preparations of GRK5 and GRK6 were inhibited in a dose-dependent fashion by GF 109203X. Almost total inhibition was observed at 10 µM. As a control, purified PKC was found to be inhibited by GF 109203X as expected. Nearly total inhibition was seen at 1 µM.

View larger version (78K): [in this window] [in a new window]   Fig. 7.   Inhibition of agonist- and PMA-induced phosphorylation of the TP by GF 109203X and SQ29,548. HEK293 cells (1 × 106) stably expressing His-tagged TP were stimulated with 100 nM I-BOP or 100 nM PMA for 10 min at 37°C in the absence or presence of 5 µM GF 109203X or 10 µM SQ29,548. The phosphorylated receptors ware isolated as described under Experimental Procedures. A representative autoradiogram of three qualitatively similar results is shown.

View larger version (67K): [in this window] [in a new window]   Fig. 8.   Inhibition of GRK5- and GRK6-induced phosphorylation of the TP by GF 109203X in intact cells. HEK293 cells (1 × 106) stably expressing His-tagged TP were transiently transfected with pcDNA3, pcDNA3-GRK5, or pcDNA3-GRK6, respectively. The transfected cells were metabolically labeled with 32Pi. The cells were incubated with 5 µM GF 109203X at 37°C for 10 min before stimulation with 100 nM I-BOP for another 10 min. The phosphorylated receptors were isolated as described under Experimental Procedures. A representative autoradiogram of three qualitatively similar results is shown.

View larger version (37K): [in this window] [in a new window]   Fig. 9.   Inhibition of PKC, GRK5, or GRK6 by GF 109203X in vitro. GST-TP-C-terminal tail (2 µg) was phosphorylated by purified PKC (10 ng) or crude preparation of GRK5 (20 µg) or GRK6 (20 µg) in the presence of increasing concentrations of GF 109203X as indicated. Phosphorylated GST-C-terminal tail was isolated as described under Experimental Procedures. A representative autoradiogram of two qualitatively similar results is shown.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

We have prepared stable C-terminal His-tagged TP-transfected HEK293 cells for studying receptor phosphorylation. The modified receptor exhibited binding and functional characteristics comparable to those of the wild-type receptor. His-tagged receptor is unique in that it can be specifically removed by inexpensive Ni²⁺-NTA agarose precipitation, which is a simple alternative to the immunoprecipitation commonly used in other receptor studies (Habib et al., 1997; Neuschafer-Rube et al., 1999).

The biological effects elicited by TP, like those of many other GPCRs, are regulated by homologous as well as heterologous desensitization (Liel et al., 1988; Murray and FitzGerald, 1989; Murray et al., 1990; Okwu et al., 1992; Sakai et al., 1996). Among several distinct mechanisms that have been implicated in the rapid desensitization of GPCRs, phosphorylation of the receptors seems to be particularly significant. A current model of homologous desensitization of GPCRs involves agonist-dependent translocation to the plasma membrane of a GRK that binds and phosphorylates the agonist-occupied receptor (Lefkowitz, 1993). The sites of phosphorylation are localized to the intracellular domains, particularly the C-terminal tail and the third intracellular loop. The subsequent binding of an arrestin-like protein is believed to dissociate the activated receptor from its associated G protein. Although the involvement of GRKs in the phosphorylation and desensitization of GPCRs has been demonstrated in a few cases (Freedman et al., 1995; Diviani et al., 1996; Neuschafer-Rube et al., 1999), their participation in the TP system has only been suggested. Parent et al. (1999) showed that TP but not TP underwent agonist-stimulated internalization. Internalization of TP was suggested to involve GRK as a dominant-negative mutant, GRK2-K220R, effectively reduced internalization. The ability of TP to internalize could be modestly enhanced by coexpression of GRKs, suggesting possible phosphorylation of the TP. In the present study, we prepared GST fusion proteins with each intracellular domain fused with GST to facilitate the isolation of labeled peptide following a protein kinase reaction in vitro. We have found that only the C-terminal tail was significantly phosphorylated by GRK5 and GRK6. This in vitro finding was further confirmed by intact cell-labeling of the receptor. Both GRK5 and GRK6 expressed in HEK293 cells augmented I-BOP-induced phosphorylation of the receptor. In fact, the kinetics of agonist-induced receptor phosphorylation was found to be more rapid with GRK than without GRK overexpression, indicating that there was a GRK-mediated receptor phosphorylation. GRKs alone did not induce any significant phosphorylation, supporting the concept that GRKs needed to be recruited to the receptor site by the agonist before phosphorylation can occur. The concept that phosphorylation is truly mediated by GRKs was further supported by a companion study in which an attenuated analog of GRK6, GRK6-RDD, was expressed in HEK293 cells expressing His-tagged TP. Overexpression of GRK6-RDD at an excess over endogenous GRKs may inhibit GRK-mediated receptor phosphorylation by acting as a competitive inhibitor of GRK activity. Phosphorylation induced by I-BOP was significantly curtailed compared with that in the presence of wild-type GRK6. The fact that GRK6-RDD induced greater phosphorylation than the control for GRK6 is probably due to the residual activity of mutant GRK6-RDD. Although the expression of GRKs was controlled at a comparable level by transfecting an equal amount of the plasmids as shown by the Western blot analysis, the expressed enzymes might have different specific activities. GRK5 and GRK6 are the two newest members of the receptor kinase family. In vitro studies have indicated that GRK5 can phosphorylate the ₂-adrenergic receptor, the M2-acetyl choline receptor, and rhodopsin in an agonist-dependent fashion (Kunapuli et al., 1994). GRK6 can also phosphorylate the same substrates but with stoichiometries significantly lower than those achieved by GRK5 (Loudon and Benovic, 1994). Subsequently, intact cell studies have shown that GRK5 can increase agonist-induced phosphorylation of the ₁-adrenergic receptor (Freedman et al., 1995) and the prostaglandin EP₄ receptor (Neuschafer-Rube et al., 1999). GRK6 can increase agonist-induced phosphorylation of the _1B-adrenergic receptor (Diviani et al., 1996). Our study has shown that either GRK5 or GRK6 exhibited little basal phosphorylation but increased dramatically I-BOP-induced phosphorylation of TP. These findings coupled with the in vitro studies clearly indicate that TP is an excellent substrate for both GRK5 and GRK6.

Not only was the phosphorylation of TP induced by I-BOP shown to be mediated by GRKs, but the Ca²⁺ release induced by I-BOP was also found to be inhibited by coexpression of GRKs in a dose-dependent manner. These findings imply an important role for GRKs in the agonist-induced desensitization and regulation of the TP. The agonist-stimulated desensitization of the TP has been shown in several reports (Habib et al., 1997; Spurney and Coffman, 1997; Spurney, 1998), although these studies did not demonstrate the participation of GRKs in the desensitization. However, a recent report by Parent et al. (1999) showed that the TP exhibited no agonist-induced internalization. Coexpression of GRKs increased receptor internalization only modestly. This seems to be in direct contrast with our finding of whether internalization is directly related to desensitization. In their study of the rapid desensitization of the -adrenergic receptors, Hausdorf et al. (1990) concluded that the molecular mechanisms underlying rapid desensitization do not seem to require internalization of the receptor, but rather an alteration in the functioning of the receptors themselves that uncouples the receptors from the associated G proteins. Our finding that GRKs mediate agonist-induced phosphorylation of the TP may provide a plausible mechanism by which an agonist induces desensitization through uncoupling the phosphorylated receptor from its affiliated G proteins.

In addition to the possible recruitment of endogenous GRKs by I-BOP and initiation of phosphorylation of TP, I-BOP may also activate phospholipase C and generate diacylglycerol, which turns on PKC (Hirata et al., 1996). Whether phosphorylation of TP can be also initiated by PKC was examined by the stimulation of His-tagged TP-transfected HEK293 cells with PMA. Indeed, PMA elicited significant phosphorylation of TP and this phosphorylation was totally inhibited by 5 µM of GF 109203X, a known inhibitor of PKC (Toullec et al., 1991), indicating the involvement of PKC in the phosphorylation of TP. However, the GRK-mediated phosphorylation induced by I-BOP was also found to be totally inhibited by the same inhibitor. Similar inhibition was also observed by other PKC inhibitors such as staurosporine, Ro-31-8220, and calphostin C, indicating that these compounds were not specific PKC inhibitors as previously claimed. GF 109203X, staurosporine, and Ro-31-8220 have been shown to act by interacting with the ATP-binding site of PKC. They may as well inhibit GRKs by interacting with a similar ATP-binding site. Calphostin C is a different type of PKC inhibitor and is known to interact with the regular domain of PKC by competing at the binding site of diacylglycerol and phorbol esters (Kobayashi et al., 1989). However, it also inhibited GRK-mediated receptor phosphorylation. There seems to be a possibility that GRKs are activated by PKC, and the inhibition of PKC by PKC inhibitors will lead to incomplete activation of GRKs, resulting in decreased receptor phosphorylation. This possibility is considered unlikely in light of the report by Pronin and Benovic (1997), who demonstrated that GRK5 was inactivated by PKC-mediated phosphorylation. It seems that these inhibitors, including calphostin C, may also inhibit GRKs. The direct effect of GF 109203X on the GRK5 and GRK6 enzyme preparations was then examined. It was found that GF 109203X was also an inhibitor of GRK5 and GRK6, although the IC₅₀ value was about 10 times higher than that for PKC. Although crude preparations of GRK5 and GRK6 were used, they should not alter the results that were shown in this study. It is not likely that phosphorylation of the C-terminal peptide was due to PKC in the crude preparation because the assay conditions were not optimal for PKC and the IC₅₀ values were about 10 times higher than that for purified PKC. Furthermore, GRK5 was found to be inactivated by PKC-mediated phosphorylation, as indicated above. Inhibition of PKC should result in a higher degree of phosphorylation by GRK if phosphorylation was due to a PKC-dependent mechanism. Therefore, the observed inhibition by GF 109203X was attributed to its effect on GRKs. Consequently, studies using this inhibitor to assess the role of PKC in agonist-induced phosphorylation and desensitization of the receptors need to be interpreted cautiously.

Spurney (1998) suggested that U-46619-induced phosphorylation of the mouse TP was mediated to a large extent by PKC because GF 109203X at 1 µM inhibited more than 70% of the phosphorylation. Our study also demonstrated nearly complete inhibition of phosphorylation induced by I-BOP at 5 µM GF 109203X. On the contrary, Habib et al. (1997) found that GF 109203X at 5 µM inhibited U-46619-dependent phosphorylation by only about 30%, suggesting that PKC was minimally involved in agonist-induced phosphorylation. Although a direct comparison of these studies is difficult because of the different experimental conditions used, it is very likely that both GRKs and PKC are involved in receptor phosphorylation following agonist stimulation. The relative contribution of each kinase in the phosphorylation of TP shall await the development of a truly specific inhibitor for each kinase. Considering the faster kinetics of GRK-mediated compared with PKC-mediated receptor phosphorylation found in ₂-adrenergic receptor system (Roth et al., 1991), it is likely that the majority of the TP phosphorylation induced by I-BOP is contributed by GRKs initially. PKC-mediated receptor phosphorylation becomes of increasing importance subsequently. Other protein kinases such as PKA can be activated also following agonist activation of the TP since I-BOP has been shown to increase intracellular levels of cAMP (Hirata et al., 1996; Habib et al., 1997). Habib et al. (1997) showed that the PKA inhibitor HA-89 at 50 µM affected marginally the U-46619-induced phosphorylation, whereas it blocked significantly forskolin-induced phosphorylation, indicating that PKA contributed little to rapid, agonist-induced phosphorylation of the TP.

In conclusion, the implications of our study can be severalfold. First, receptor tagged with oligohistidine is an inexpensive and effective means of studying the phosphorylation and desensitization of the receptor. Second, our results provide strong evidence that GRKs can play a role in agonist-dependent regulation of the TP. Third, they contribute to the elucidation of the receptor substrate specificity of different GRKs by identifying the C-terminal tail of the TP as phosphorylation substrate for GRK5 and GRK6. Phosphorylation of this domain may promote TP desensitization. Our study demonstrates directly a role for GRKs in the agonist-induced phosphorylation and desensitization of the TP.