Role of Cytoplasmic Tail of the Type 1A Angiotensin II Receptor in Agonist– and Phorbol Ester–Induced Desensitization
Abstract—To investigate mechanisms underlying the agonist-induced desensitization of the type 1A angiotensin II receptor (AT1A-R), we have stably expressed in Chinese hamster ovary (CHO) cells the wild-type receptor and truncated mutants lacking varying lengths of the cytoplasmic tail. Assay of inositol 1,4,5-trisphosphate (IP3) formation in response to agonist demonstrated that the truncated mutants T318, T328, and T348 lacking the last 42, 32, or 12 amino acid residues, respectively, couple with Gq protein with an efficiency similar to that of full-length receptors, whereas coupling of Gq protein was abolished in the T310 truncated mutant devoid of the carboxyl-terminal 50 amino acids. Exposure of CHO/AT1A-R cells expressing the wild-type AT1A-R to angiotensin II resulted in rapid and dose-dependent homologous desensitization of receptor-mediated IP3 formation, which was independent of the receptor internalization. Mastoparan, an activator of G protein–coupled receptor kinase (GRK), induced desensitization of the AT1A-R. The agonist-induced desensitization of the receptor was largely prevented by heparin, a potent inhibitor of GRK, whereas it was only partially attenuated by a protein kinase C (PKC)-specific inhibitor. The homologous or heterologous desensitization of the receptor was greatly impaired in the truncated mutants T318 and T328, lacking the Ser/Thr-rich (13 or 12 Ser/Thr residues) cytoplasmic tail of the AT1A-R. Deletion of the last two Ser residues, including one PKC consensus site in the receptor tail, prevented only phorbol 12-myristate 13-acetate–induced desensitization by 30%. Moreover, we found an agonist-induced translocation of a heparin-sensitive kinase activity. The angiotensin II–stimulated heparin-sensitive kinase could phosphorylate a thioredoxin fusion protein containing the entire AT1A-R cytoplasmic tail (N295 to E359), which lacks consensus phosphorylation sites for GRK1, GRK2, and GRK3. The heparin-sensitive kinase may not be GRK2, GRK3, or GRK6 expressed in CHO/AT1A-R cells, since angiotensin II did not induce translocation of these receptor kinases. Potential Ser/Thr phosphorylation sites located between S328 and S347 in the cytoplasmic tail of AT1A-R seem to play a critical role in the heterologous and homologous desensitization of the receptor. A heparin-sensitive kinase other than GRK2, GRK3, or GRK6 may be involved in the agonist-induced homologous desensitization of the AT1A-R.
The type 1 Ang II receptors, AT1A-R and AT1B-R, undergo rapid homologous downregulation on exposure to Ang II.1 2 3 Multiple mechanisms, such as desensitization and concurrent internalization, seem to be involved in this response to maintain cellular homeostasis. The mechanism of desensitization is complex; however, receptor phosphorylation seems to be an initial and major step. Extensive mutagenesis studies on the β2-adrenergic receptor revealed that phosphorylation of two Ser residues located on the third cytoplasmic loop and in the proximal portion of the cytoplasmic tail by protein kinase A is involved in the heterologous desensitization of the receptor, whereas agonist-induced homologous desensitization requires phosphorylation of many Ser/Thr residues in the distal portion of the receptor cytoplasmic tail by β-ARK. Phosphorylation of the cytoplasmic tail by β-ARK induces arrestin-like protein binding to the receptor, which has been proposed to block fully the physical coupling of receptor to Gs protein and adenylyl cyclase.4 This indicates multiple mechanisms for desensitization by phosphorylation. In the Gq/11-coupled AT1A-R, the desensitization mechanisms are not well understood. Although the receptor phosphorylation is associated with desensitization, whether they are causally related is not clear.3 Conflicting views exist as to the sites of phosphorylation. Thomas et al5 proposed that the second cytoplasmic loop may be the site of phosphorylation responsible for AT1A-R desensitization, whereas Balmforth et al6 reported observations indicating involvement of a long cytoplasmic tail in the receptor desensitization, without identifying specific domains.
The desensitization of the AT1 receptor was shown to occur independent of the receptor internalization in rat cardiomyocytes.1 We were able to show that the desensitization of AT1B-R can occur even if receptor internalization is inhibited and concluded that the two phenomena are distinct receptor functions even though they occur concurrently.2 Although AT1A-R and AT1B-R share a high degree of structural homology, they have some considerable differences in the cytoplasmic tail region, particularly in positions of Ser/Thr residues. Du et al7 have recently reported that AT1A-R and AT1B-R are regulated in opposite directions on low salt feeding. Since AT1A-R is the dominant subtype of the AT1 receptor in most of tissues, except for the adrenal and pituitary glands, determination of the mechanism of its desensitization has long been overdue.
Since we have been able to obtain cells expressing AT1A-R at a high level, which will enable us to quantify the downregulation of the receptor and its functions, we have performed studies directed to the delineation of the receptor domains involved in the desensitization using various mutants of AT1A-R and a thioredoxin fusion protein containing the entire carboxyl-terminal region of the receptor. To gain insight into the type of kinases involved in this process, we have also applied reagents that stimulate or inhibit PKC and GRKs. We have obtained evidence that the proximal portion (K310 to L317) of the receptor cytoplasmic tail is involved in the AT1A-R coupling to Gq and that the potential Ser/Thr phosphorylation sites located between S328 and S347 in the cytoplasmic tail play a critical role in the agonist- and phorbol ester (PMA)-induced desensitization. Finally, the type and role of protein kinases involved in agonist-induced desensitization of the AT1A-R were examined.
Materials and Methods
Cell culture media, fetal calf serum, geneticin (G418 sulfate), and protein tyrosine kinase inhibitors were obtained from Life Technologies, Inc; [125I]NaI, [γ-32P]ATP, and [3H]IP3 radioreceptor assay kits were purchased from Dupont NEN; CHO cells (CHO-K1) were obtained from American Type Culture Collection; Ang II and [Sar,1 Ile8 ]Ang II were obtained from Peninsula Laboratories; the nonpeptide receptor antagonist losartan and PD124319 were obtained from Research Biochemical Inc; PVDF membranes were purchased from Millipore; GF109203X, W-7, and H-89 were purchased from Calbiochem; the restriction enzymes and calcium phosphate transfection kits were obtained from Promega; expression vector pRc/CMV and the thiofusion expression system were purchased from Invitrogen Corp; PKC (α, β, and γ), Src kinase, Src kinase substrate peptide, and PKC assay kits were from Upstate Biotechnology Inc; and the polyclonal antibodies against GRKs and PLCs were from Santa Cruz. All other chemicals were purchased from Sigma Chemical Co.
Stable Expression of WT and Truncated AT1A-R
The entire coding region of rat kidney AT1A-R cDNA (an open reading frame encoding 359 amino acid residues) was subcloned into the BstXI site of the pRc/CMV expression vector. Truncated mutants lacking carboxyl-terminal segments of varying lengths were prepared by inserting a stop codon. Sites of truncation are shown in Fig 1⇓. The mutated DNA was confirmed by dideoxynucleotide sequencing using appropriate primers, as described by Sanger et al,8 and subcloned into the expression vector pRc/CMV. CHO-K1 cells were grown in Ham’s F-12 medium containing 10% fetal calf serum at 37°C in a humidified incubator with 5% CO2. At 50% confluence, CHO cells were transfected with WT or truncated AT1A-R plasmid DNA using a calcium phosphate precipitation/glycerol shock procedure previously described.2 After selection in G418 (500 μg/mL), resistant colonies were isolated, amplified, and screened for receptor expression by radioligand binding assay. Cells were propagated in flasks with Ham’s F-12 medium supplemented with 10% fetal calf serum under a selection pressure of 150 μg/mL G418 in an atmosphere of 95% air and 5% CO2 at 37°C.
Radioligand Binding Assay
Monoiodinated 125I-[Sar,1 Ile8 ]Ang II was prepared by the lactoperoxidase method9 and purified by reversed-phase HPLC as described previously10 except that a 0% to 80% acetonitrile gradient was applied for 60 minutes with a flow rate of 1.0 mL/min. Binding studies were performed with cells at or near confluence essentially as described previously.2 The assay buffer consisted of 50 mmol/L Tris-HCl, pH 7.4, 100 mmol/L NaCl, 5 mmol/L MgCl2, 0.25% BSA, and 0.5 mg/mL bacitracin. For determination of receptor affinity and density, competition binding studies were performed in the presence of 50 pmol/L 125I-[Sar,1 Ile8 ]Ang II and increasing concentrations of unlabeled ligands (10−12 to 10−6 mol/L). For measurement of the surface receptor binding capacity, the cells were pretreated as indicated and then incubated in the presence of 100 nmol/L Ang II for the indicated period at 37°C. Cells were then washed with ice-cold 50 mmol/L glycine and 150 mmol/L NaCl, pH 3.0, for 5 minutes at 4°C, followed by two saline washes. The binding studies were performed at near-saturating conditions with 2 nmol/L 125I-[Sar,1 Ile8 ]Ang II for 3 hours at 4°C.
Measurement of IP3 Production
Cells were subcultured into 12-well dishes and were exposed to Ang II at or near confluence; the treatment was stopped by the addition of 1/5 vol of 100% ice-cold trichloroacetic acid to the plates. The cells were then harvested by scraping and transferred to polypropylene tubes. The cell extract was vortexed thoroughly and then centrifuged for 10 minutes at 6000g at 4°C. The supernatant was removed and warmed to room temperature for 15 minutes. Levels of IP3 in each supernatant were determined by use of a radioimmunoassay kit from Dupont NEN following the manufacturer’s instructions.
Permeabilization of CHO Cells
CHO cells bearing AT1A-R at subconfluence in 12-well plates were washed twice with prewarmed calcium-free PBS and then washed twice with KG buffer (mmol/L: KCl 120, NaCl 30, MgCl2 2.5, K2HPO4 1, PIPES 10, EGTA 2, and CaCl2 0.5, pH 7.2). Cells were then incubated for 30 seconds with 0.0075% digitonin at room temperature in KG-A buffer (KG buffer supplemented with 5 mmol/L glucose and 2 mmol/L ATP), followed by three washes with KG buffer. Under this condition, >95% of the cells were permeabilized as assessed by staining with trypan blue. After permeabilization, the cells were loaded with mastoparan or low molecular weight heparin (average molecular weight, 3000) by incubation with 100 μmol/L mastoparan or 10 μmol/L heparin for 15 minutes in KG-A buffer at 37°C, respectively.
Desensitization of Receptors
Studies were performed essentially as described previously.2 In brief, the mastoparan- or heparin-loaded cells, or cells pretreated with chemicals as indicated, were incubated with or without 100 nmol/L Ang II in serum-free medium for indicated periods at 37°C. The reaction was stopped by washing the cells with 150 mmol/L NaCl/50 mmol/L glycine (pH 3.0) for 5 minutes at 4°C and with prewarmed PBS twice. For measurement of the IP3 level in response to Ang II, the cells were again stimulated with Ang II (100 nmol/L), and 15 seconds later the IP3 mass was assayed as described above.
Preparation of the Bacterial Thioredoxin-AT1A Fusion Protein
A region of the rat AT1A-R cDNA encoding the entire cytoplasmic tail (N295 to E359) was amplified by polymerase chain reaction and inserted into pBluescript KS(+) TA vector, and DNA was sequenced. Then, the BamHI-SalI fragment was cut and inserted into the same sites of the thiofusion expression plasmid pTrxFus vector in the frame. Induction of pTrxFus-AT1A–transformed Escherichia coli with tryptophan (100 μg/mL) resulted in the production of a fusion protein of ≈19 kD (11.6 kD of thioredoxin plus ≈7.5 kD for N295 to E359). The fusion protein was purified with activated ThioBond resin (Invitrogen Corp) according to manufacturer’s protocols.
Preparation of Cell Fractions and Western Blotting
Serum-starved cells were washed twice with PBS and stimulated with 100 nmol/L Ang II for the indicated periods. Then, the cells were washed with ice-cold PBS and scraped into 0.5 mL per dish of lysis buffer (10 mmol/L Tris-HCl, pH 7.4, 5 mmol/L EDTA, 0.1 mmol/L phenylmethylsulfonyl fluoride, 1 mmol/L benzamidine, and 10 μg/mL leupeptin and aprotinin). The cells were allowed to swell for 10 to 15 minutes and were then homogenized three times (30 seconds each) in a Polytron homogenizer (Brinkmann). Unbroken cells and cell nuclei were sedimented at 800g for 3 minutes and discarded. The plasma membranes were collected by centrifugation at 48 000g for 20 minutes and resuspended in lysis buffer. The supernatant, representing the cytosolic fraction, was centrifuged at 100 000g for 60 minutes. Preparations were kept at −80°C and used for Western blot or kinase assay. Samples containing equal amounts of protein (50 to 100 μg) were separated by 7.5% SDS-PAGE gels and transferred to PVDF membranes. Membranes were probed with 0.3 to 0.5 μg/mL GRK polyclonal antibodies, and immunoreactive bands were visualized with the enhanced chemiluminescence detection system (New England Biolabs). For Western blotting of PLCs, CHO/AT1A-R cells or VSMCs were lysed on ice in lysis buffer containing 1% Triton X-100. The detergent-lysed cell extracts were separated by 7.5% SDS-PAGE gels and subjected to Western blotting with PLC polyclonal antibodies as described above.
In Vitro Phosphorylation of Thioredoxin-AT1A Fusion Protein
For measuring the PKC-dependent or Src kinase–dependent phosphorylation of thioredoxin-AT1A fusion protein, purified thioredoxin-AT1A fusion protein (5 μg) was incubated with PKC (mixture of α, β, and γ, 50 ng per tube) or Src kinase (19 U per tube) in kinase buffer (mmol/L: Tris-HCl 20, MgCl2 10, EGTA 1, and dithiothreitol 2, pH 7.4) containing 1 μmol/L [γ-32P]ATP for 30 minutes or 60 minutes at 30°C (final volume, 100 μL). For phosphorylation of the fusion protein by membrane preparations from CHO/AT1A cells, an aliquot of cell membrane fractions (90 μg) that had been precleared by activated ThioBond resin, was incubated with purified thioredoxin-AT1A fusion protein (5 μg) in the above kinase buffer containing 50 μmol/L [γ-32P]ATP for 30 minutes at 30°C. The reaction was terminated by the addition of 1 mL ice-cold TE buffer (mmol/L: Tris-HCl 10 and EDTA 10, pH 7.4). ThioBond resin was added, collected by centrifugation (13 000g for 10 seconds), and washed four times with 1 mL of TE buffer. Bound fusion protein was dissociated by boiling in 1× SDS-PAGE sample buffer. Proteins were resolved by 15% SDS-PAGE. Gels were then dried and analyzed by autoradiography.
Desensitization of the AT1A-R–Mediated IP3 Formation
To investigate mechanisms underlying the agonist-induced desensitization of AT1A-R, we expressed cloned rat WT AT1A-R in CHO cells that lack endogenous AT1 receptors. The binding of 125I-[Sar,1 Ile8 ]Ang II, a peptidic Ang II receptor antagonist, to the cells bearing WT AT1A-R was dose-dependently inhibited by Ang II, [Sar,1 Ile8 ]Ang II, and an AT1A-R–selective antagonist, losartan. In contrast, PD124319, a selective AT2 receptor antagonist, had no effect on the binding of 125I-[Sar,1 Ile8 ]Ang II (data not shown). The AT1 receptor activates PLC through Gq/11.11 Previous studies of the AT1A-R have shown that Ang II–induced formation of IP3 was inhibited 75% by the tyrosine kinase inhibitor genistein and was associated with the activation of PLCγ1 in VSMCs lacking PLCβ1.12 Western blot analysis demonstrated that PLCβ1(150 kD), PLCβ3 (150 kD), and PLCγ1(145 kD) were clearly detected in CHO cells expressing high levels of WT AT1A-R (CHO/AT1A-R cells), whereas only PLCβ3 and PLCγ1 were detected in VSMCs (Fig 2⇓). To determine whether Ang II–induced formation of IP3 was sensitive to the tyrosine kinase inhibitors or calmodulin inhibitor W-7, we measured IP3 response to Ang II in CHO/AT1A-R cells with three different tyrosine inhibitors (genistein, herbimycin-A, and lavendustin A) or W-7. The results (not presented) showed that neither compound affected the IP3 response to Ang II in intact cells. This observation suggests that the IP3 response to Ang II in CHO/AT1A-R cells is probably mediated by PLCβ1 through Gq/11.11
To ascertain whether agonist-induced desensitization of AT1A-R expressed in CHO cells occurs, we determined the effect of a pretreatment with Ang II on the subsequent maximal Ang II–stimulated IP3 formation. The pretreatment with Ang II resulted in a subsequent attenuation of IP3 response to Ang II. By varying the time of preexposure to the agonist from 1 minute to 30 minutes, we found that Ang II rapidly induced receptor desensitization as early as 1 minute (P<.05), with complete desensitization observed by pretreatment for 15 minutes (Fig 3A⇓). Attenuation of IP3 production was maximal when cells were pretreated with 10 to 100 nmol/L Ang II, and half-maximal effects were observed after pretreatment with ≈3 nmol/L Ang II (Fig 3B⇓).
Involvement of GRKs in Agonist-Induced Homologous Desensitization of AT1A-R
We and others2 13 14 have shown that PKC is involved in the heterologous desensitization of the AT1 receptor. Treatment of CHO/AT1A-R cells with PMA (100 nmol/L) for 10 minutes completely abolished receptor-mediated production of IP3 with no effect on the basal level of IP3 (Fig 4⇓). This effect of PMA was completely antagonized by pretreating the cells with 10 μmol/L of a newly tested in vitro and in vivo nontoxic PKC-specific inhibitor, GF109203X15 (Fig 4⇓). Consistent with the effects of the PKC inhibitor, downregulation of PKC by overnight incubation with PMA (1 μmol/L)16 completely blocked the desensitization of the AT1A-R by PMA (data not shown). Like AT1B-R, desensitization of AT1A-R by a low Ang II concentration (1 nmol/L) was abolished by GF109203X, whereas the desensitizing effect at a high agonist concentration (100 nmol/L) was only partially prevented by GF109203X (Fig 4⇓). The partial attenuation was seen even at a high dose of this compound (20 μmol/L) or with PKC depletion by overnight treatment with 1 μmol/L PMA (data not shown). These results suggests that a PKC-independent pathway, probably involving GRKs, contributes to the homologous desensitization of the AT1A-R in transfected CHO cells.
We then studied the role of GRKs in agonist-induced homologous desensitization of AT1A-R. Mastoparan, a wasp venom peptide, has been known to mimic the agonist-bound (active form) GPCRs and activate GRKs.17 18 Incubation of permeabilized CHO/AT1A-R cells with 100 μmol/L mastoparan for 10 minutes completely abolished the receptor-mediated IP3 response to Ang II (Fig 5⇓). On the other hand, pretreatment of the permeabilized cells with 10 μmol/L heparin, a potent GRK inhibitor,19 20 for 20 minutes largely prevented the Ang II–induced desensitization of the AT1A-R–mediated IP3 response. The pretreatment with heparin decreased the basal level of IP3 only slightly (Fig 5⇓).
These results indicate that PKC plays a major role in the heterologous desensitization of AT1A-R and that a heparin-sensitive GRK(s) may be involved in the agonist-induced homologous desensitization of the receptor in transfected CHO cells.
Impaired Heterologous or Homologous Desensitization of Truncated Mutants of AT1A-R
To address the possible role of the cytoplasmic tail in the regulation of AT1A-R, we constructed truncated mutants of AT1A-R and stably expressed them in CHO cells. As shown in Fig 1⇑, T310 and T318 truncation removes all Ser/Thr residues, including three potential PKC phosphorylation sites in the cytoplasmic tail of the AT1A-R; T328 truncation removes all but the Ser326; and T348 truncation removes a potentially farnesylated cysteine and two Ser residues in the tail, including one PKC phosphorylation site, S348AK. Clonal, stably transfected cells expressing T310, T318, or T348 AT1A-R were prepared and matched with CHO/AT1A-R cells expressing WT AT1A-R at a high level. Cells expressing the T328 receptor were compared with cells bearing low levels of WT AT1A-R (Table⇓). As shown in the Table⇓, Scatchard analysis revealed that the dissociation constants (Kds) for 125I-[Sar,1 Ile8 ]Ang II among the WT AT1A-R and its mutants were similar, indicating that all the mutants possessed similar ligand binding affinity. Assay of IP3 formation in response to Ang II demonstrated that truncated mutants (T318, T328, and T348) of AT1A-R couple with effector PLC with an efficiency similar to that of full-length receptors. By contrast, the IP3 formation on Ang II stimulation was barely detected in cells expressing high levels of truncated mutant T310, indicating the region between K310 and L317 in the cytoplasmic tail of the AT1A-R that is involved in the receptor coupling to Gq protein (Table⇓).
As shown in Fig 6⇓, pretreatment of cells expressing WT AT1A-R or mutant T348 with 100 nmol/L Ang II abolished the IP3 response of the cells to a subsequent maximal stimulation with Ang II. By contrast, the Ang II–induced attenuation of the IP3 response was markedly impaired in both cell clones expressing T318 or T328 AT1A-R, indicating that potential Ser/Thr phosphorylation sites located between S328 and S347 in the cytoplasmic tail are important for the agonist-induced desensitization. In addition to the impaired agonist-induced desensitization, the PMA-dependent heterologous desensitization of the receptor was also dramatically impaired in the truncated mutants T318 and T328 (Fig 6⇓). Furthermore, deletion of the last two Ser residues (mutant T348), including one PKC phosphorylation site in the receptor tail, prevented PMA-induced desensitization of the AT1A-R by ≈30% (Fig 6⇓).
Phosphorylation of Thioredoxin-AT1A Fusion Protein by PKC
To determine whether the cytoplasmic tail of AT1A-R could be phosphorylated by PKC, we generated a thioredoxin fusion protein containing the entire carboxyl tail of the AT1A-R (Fig 7A⇓). The purified thioredoxin-AT1A fusion protein was then used in an in vitro PKC phosphorylation assay. It was not a surprise that the thioredoxin-AT1A fusion protein was highly phosphorylated by PKC (mixture of α, β, and γ) (Fig 7B⇓), since there are three potential PKC phosphorylation sites in the carboxyl tail of the AT1A-R. However, the thioredoxin-AT1A fusion protein was not phosphorylated by Src kinase, even though a Src kinase substrate peptide (KVEKIGEGTYGVVKK)21 was highly phosphorylated (205 728±17 235 cpm, 8-fold incorporation over background). There was no phosphorylation of the thioredoxin protein by PKC or Src kinase (Fig 7B⇓).
Ang II Did Not Induce Translocation of GRK2, GRK3, or GRK6
Agonist-induced translocation of GRK1, GRK2, or GRK3 is considered the first step involved in the homologous desensitization of rhodopsin or β-adrenergic receptors, respectively.22 To determine whether Ang II could induce translocation of GRK from cytosol to the plasma membranes, CHO/AT1A-R cells were pretreated with 1 μmol/L PMA overnight to deplete PKC and then exposed to 100 nmol/L Ang II for 15 minutes. Plasma membranes were prepared, and the kinase activity associated with the membranes was determined using thioredoxin-AT1A fusion protein as a substrate. As shown in Fig 7C⇑, the phosphorylation of the thioredoxin-AT1A fusion protein was increased dramatically by membranes derived from Ang II–treated CHO/AT1A-R cells depleted of PKC. This indicated that the cytoplasmic tail of AT1A-R could be phosphorylated by a PKC-independent mechanism. Moreover, the Ang II–induced kinase activity in the membrane fraction was completely inhibited by the GRK inhibitor heparin (100 nmol/L) (Fig 7C⇑) but not by the PKC-specific inhibitor GF109203X or the PKA inhibitor H-89 (data not shown). To determine which individual GRK is selectively activated (by translocation) on Ang II stimulation, CHO/AT1A-R cells or VSMCs were stimulated with Ang II for different time periods, and the cytosol and plasma membranes were prepared. Western blot experiments showed that GRK2 was almost equally distributed in cytosol or plasma membranes of the CHO/AT1A-R cells, whereas GRK3 was mainly present in cytosol (Fig 8A⇓). Ang II treatment did not induce translocation of GRK2 or GRK3 in CHO/AT1A-R cells, since the protein level of GRK2 or GRK3 in the cytosolic and membrane fractions was not changed by Ang II stimulation (Fig 8A⇓). Similar results with GRK2 and GRK3 were obtained before and after Ang II treatment in VSMCs (data not shown). In addition, acute PMA treatment had no effect on the translocation of GRK2 or GRK3 in either CHO/AT1A-R cells (Fig 8A⇓) or VSMCs (data not shown). However, translocation of GRK2 was observed by agonist UK14,304 treatment in MDCK cells stably expressing α2A-adrenergic receptors23 (Fig 8B⇓). GRK5 was undetectable in CHO/AT1A-R cells but was detected in VSMCs (Fig 8C⇓). GRK6 was detected only in the plasma membranes of both CHO/AT1A-R cells (Fig 8D⇓) and VSMCs (not shown). The localization of GRK6 was not changed on Ang II stimulation of CHO/AT1A-R cells (Fig 8D⇓) or VSMCs (not shown). On the other hand, PKC was rapidly activated by Ang II in CHO/AT1A-R cells as well as in VSMCs (not shown), as reported previously.24 25
In the present study, we have used four truncated mutants of AT1A-R (T310, T318, T328, and T348) lacking the potential carboxyl-terminal phosphorylation sites but retaining the normal binding of Ang II intact. We showed that the proximal portion (K310 to L317) of the receptor cytoplasmic tail is involved in the AT1A-R coupling to Gq, whereas the middle portion (S328 to S347) containing the potential Ser/Thr (eight Ser and two Thr residues) phosphorylation sites plays an important role in the agonist- and PMA-induced desensitization of the AT1A-R. Furthermore, the thioredoxin-AT1A fusion protein containing the entire cytoplasmic tail of the AT1A-R can be heavily phosphorylated by PKC and by a membrane-associated heparin-sensitive kinase from Ang II–treated CHO/AT1A-R cells depleted of PKC.
It has been noted that Ang II action is accompanied by a rapid and profound desensitization known as tachyphylaxis. However, the mechanisms by which Ang II receptors desensitize are still not clear. It seems that receptor sequestration or internalization does not play a major role in causing the homologous desensitization of the AT1A-R expressed in CHO cells, since the complete desensitization induced by 15 minutes of incubation with 100 nmol/L Ang II still occurred even when the receptor internalization was almost completely blocked by concanavalin A2 or phenylarsine oxide treatment (loss of membrane receptors: 12±2% with concanavalin A+Ang II or 8±0.6% with phenylarsine oxide+Ang II versus 50±4% with Ang II alone). Similar findings have been reported for the AT1 receptors in cardiomyocytes,1 AT1B-R expressed in CHO cells,2 AT1A-R expressed in 293 cells,3 and other PLC-linked GPCRs.26 27 Thus, rapid desensitization of the AT1A-R–mediated IP3 formation may well be a consequence of the agonist-induced biochemical modification of the receptor, such as receptor phosphorylation.
The cytoplasmic tail of the AT1A-R contains multiple Ser/Thr residues (13 of the last 59 amino acids), and there are three consensus PKC phosphorylation sites (S331TK, S338YR, and S348AK).28 29 Removal of all Ser/Thr residues in the tail (truncated mutant T318) or removal of all Ser/Thr residues but the S326 (mutant T328) dramatically impaired agonist- and PMA-induced desensitization of the receptor. Deletion of the last two Ser residues, including one PKC consensus site, prevented only the PMA-induced heterologous desensitization by ≈30%, suggesting that two other PKC consensus sites (S331TK and S338YR) may chiefly contribute to the PKC-dependent desensitization of the AT1A-R. Since the immunoprecipitable antibody against the AT1 receptor is not available, we cannot measure the receptor phosphorylation on agonist or PMA stimulation. We made and prepared a purified thioredoxin-AT1A fusion protein that contains the entire cytoplasmic tail of AT1A-R. The thioredoxin-AT1A fusion protein was highly phosphorylated by PKC and by a membrane-associated heparin-sensitive kinase from Ang II–treated CHO/AT1A-R cells depleted of PKC, but not by Src kinase in vitro despite the presence of four tyrosine residues in the receptor cytoplasmic tail. Taken together, these results indicate that the phosphorylation of the receptor cytoplasmic tail is involved in the homologous and heterologous desensitization of the AT1A-R and that the potential Ser/Thr phosphorylation sites important for the AT1A-R desensitization are located between S328 and S347 (containing eight Ser and two Thr residues) of the receptor cytoplasmic tail. Studies are now in progress to identify the major phosphorylation site(s) by the agonist or PMA treatment.
The PKC-specific inhibitor GF109203X completely suppressed the desensitization of AT1A-R by Ang II at a low concentration (1 nmol/L). This suggests that PKC plays a major role in heterologous desensitization at a near-physiological agonist concentration. The desensitizing effect at a higher agonist concentration (100 nmol/L) is only partially prevented by the inhibition of PKC. Previous studies have shown that the PKC inhibitor staurosporine suppressed agonist-induced Ang II receptor desensitization in Xenopus oocytes.14 Pfeilschifter and coworkers30 31 32 have obtained evidence implicating PKC in the Ang II–induced desensitization in glomerular mesangial cells. We have demonstrated that PKC is only partially involved in agonist-induced desensitization of AT1B-R in transfected CHO cells.2 Similar findings have also been reported for several other PLC-linked GPCRs.33 34 In contrast, PKC depletion or treatment with the selective PKC inhibitor RO31–7519 did not affect the rapid agonist-induced desensitization of the Ang II receptor in neonatal cardiac myocytes.1 Recently, Oppermann et al3 reported that PKC inhibition with staurosporine reduced agonist-induced phosphorylation of AT1A-R by 40% to 50% in transfected 293 cells but did not affect the agonist-induced desensitization of the receptor. In aggregate, these studies and ours suggest that the agonist-induced desensitization of AT1A-R may only be partially regulated by PKC in a cell type–specific manner.
A wasp venom peptide, mastoparan, has been shown to mimic the agonist-bound (active form) GPCRs and activate GRKs.17 18 Challenging the permeabilized cells expressing WT AT1A-R with mastoparan completely abolished the receptor-mediated IP3 response to Ang II. Pretreatment of the permeabilized cells with the GRK inhibitor heparin largely prevented the agonist-induced desensitization of AT1A-R. Furthermore, treatment of the cells depleted of PKC with Ang II induced the translocation of a heparin-sensitive kinase activity, which could phosphorylate the thioredoxin-AT1A fusion protein containing the entire cytoplasmic tail of AT1A-R. These data indicate that the heparin-sensitive kinase may be mainly involved in the agonist-induced homologous desensitization of AT1A-R. Six members of the GRK family have been cloned to date35 and, on the basis of structural similarities, have been divided into three subfamilies: (1) rhodopsin kinase (GRK1), (2) the β-adrenergic receptor kinase subfamily (GRK2 and GRK3), and (3) the GRK4 subfamily (GRK4, GRK5, and GRK6). GRKs also seem to have different patterns of expression in tissues. GRK1 and GRK4, for instance, are primarily expressed in retinal cells and testis, respectively, suggesting very specific functions. Whereas, other GRKs are expressed in many tissues, suggesting their broad roles in regulating the functions of GPCRs. Overexpression (>20-fold) of either GRK2, GRK3, or GRK5 was shown to enhance agonist-induced AT1A-R phosphorylation equally well,3 suggesting a role for GRKs (GRK2, GRK3, and GRK5) in the agonist-induced phosphorylation and desensitization of the AT1A-R. Since active GPCRs (agonist-bound form), in the presence of charged phospholipids, can directly associate with and activate GRKs,36 37 it is possible that overexpression (20-fold) of either GRK2, GRK3, or GRK5 could result in nonspecific direct interaction between Ang II–bound AT1A-R with the overexpressed GRK, leading to the enhanced agonist-induced phosphorylation of AT1A-R, since experiments did not show a significant difference between individual GRKs in their ability to enhance receptor phosphorylation in response to Ang II.3 As described above, we have demonstrated that the cytoplasmic tail of AT1A-R plays a critical role in the agonist-induced homologous desensitization of the receptor. The cytoplasmic tail of AT1A-R does not contain the consensus phosphorylation sites for GRK1 (the presence of acidic residue on the C-terminal side of a Ser/Thr), GRK2, and GRK3 (the presence of acidic residue localized to the N-terminal side of a Ser/Thr).37 38 Stimulation of CHO/AT1A-R cells or VSMCs with Ang II did not induce translocation of GRK2 and GRK3. GRK5 was not detected in CHO/AT1A-R cells but was detected only in the membrane fraction of quiescent VSMCs. GRK6 was detected only in the plasma membranes of both CHO/AT1A-R cells and VSMCs. The localization of GRK6 was not affected by Ang II stimulation of either CHO/AT1A-R cells or VSMCs. Moreover, we immunoprecipitated GRK6 and measured its catalytic activity using a soluble GRK substrate, casein, according to method of Pitcher et al.39 The activity of GRK6 was not affected by Ang II treatment in CHO/AT1A-R cells (not shown). Therefore, it seems that GRK2, GRK3, or GRK6 might not be involved in the agonist-induced homologous desensitization of the AT1A-R, at least in CHO/AT1A-R cells and VSMCs. Recently, a novel heparin-sensitive 40-kD receptor kinase, designated as MEK, was purified from porcine cerebellum. It was identified in the CHO cells expressing the m3-muscarinic receptor, and the activity of MEK in the membrane fraction was increased after the agonist stimulation.40
In summary, the data presented have demonstrated that potential Ser/Thr phosphorylation sites located between S328 and S347 in the cytoplasmic tail of AT1A-R play a critical role in the heterologous and homologous desensitization of the receptor. Although a recent study5 reported that the cytoplasmic tail of AT1A-R is not involved in desensitization, the preliminary observation needs to be reconsidered, since no acid-washing step was taken to remove the receptor-bound Ang II before a second challenge of the cells with agonist during the measurement of Ca2+ mobilization. However, the precise role played by the cytoplasmic tail of the AT1A-R in the complex series of biochemical mechanisms underlying receptor regulation needs to be explored further. The present results with the T310 truncated mutant indicate that the proximal portion (K310 to L317) of the AT1A-R cytoplasmic tail is directly involved in receptor coupling to Gq. Recently, T. Sano (unpublished data, 1997) detailed experiments showing that a specific sequence, Y312-F313-L314, in the proximal portion of the rat AT1A-R cytoplasmic tail is essential for coupling and activation of Gq. Clearly, more work is needed to identify the Ser/Thr residues between S328 and S347 involved in the homologous and heterologous desensitization of AT1A-R.
Selected Abbreviations and Acronyms
|Ang II||=||angiotensin II|
|β-ARK||=||β2-adrenergic receptor kinase|
|AT1A-R, AT1B-R||=||AT1A and AT1B receptor|
|CHO||=||Chinese hamster ovary|
|GPCR||=||G protein–coupled receptor|
|GRK||=||G protein–coupled receptor kinase|
|MDCK||=||Madin-Darby canine kidney|
|MEK||=||muscarinic receptor kinase|
|PKC||=||protein kinase C|
|PMA||=||phorbol 12-myristate 13-acetate|
|VSMC||=||vascular smooth muscle cell|
This study was supported in part by National Institutes of Health grants HL-58205, HL-35323, and DK-20593. The authors thank J.R. Keefer for the MDCK cells expressing α2A-adrenergic receptor and Dr E.J. Landon for reading this manuscript.
- Received May 28, 1997.
- Accepted January 5, 1998.
- © 1998 American Heart Association, Inc.
Abdellatif MM, Neubauer FC, Lederer WJ, Rogers TB. Angiotensin-induced desensitization of the phosphoinositide pathway in cardiac cells occurs at the level of the receptor. Circ Res. 1991;69:800–809.
Tang H, Shirai H, Inagami T. Inhibition of protein kinase C prevents rapid desensitization of type 1B angiotensin II receptor. Circ Res. 1995;77:239–248.
Oppermann M, Freedman NJ, Alexander RW, Lefkowitz RJ. Phosphorylation of the type 1A angiotensin II receptor by G protein-coupled receptor kinases and protein kinase C. J Biol Chem. 1996;271:13266–13272.
Thomas WG, Thekkumkara TJ, Motel TJ. Baker KM. Stable expression of a truncated AT1A receptor in CHO-K1 cells. J Biol Chem. 1995;270:207–213.
Du Y, Yao A, Guo D, Inagami T, Wang DH. Differential regulation of angiotensin II receptor subtypes in rat kidney by low dietary sodium. Hypertension. 1995;25(pt 2):872–877.
Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci U S A. 1977;74:5463–5467.
Benya RV, Kusui T, Shikado F, Battey JF, Jensen RT. Desensitization of neuromedin B receptors (NMB-R) on native and NMB-R-transfected cells involves down-regulation and internalization. J Biol Chem. 1994;269:11721–11728.
Marrero MB, Paxton WG, Duff JL, Berk BC, Bernstein KE. Angiotensin II stimulates tyrosine phosphorylation of phospholipase C-γ1 in vascular smooth muscle cells. J Biol Chem. 1994;269:10935–10939.
Brock TA, Rittenhouse SE, Powers CW, Ekstein LS, Gimbrone MA, Alexander RW. Phorbol ester and 1-oleoyl-2-acetylglycerol inhibit angiotensin activation of phospholipase C in cultured vascular smooth muscle cells. J Biol Chem. 1985;260:14158–14162.
Toullec D, Pianetti P, Coste H, Bellevergue P, Grand-Perret T, Ajakane M, Baudet V, Boissin P, Boursier E, Loriolle F, Duhamel L, Charon D, Kirilovsky J. The bisindolylmaleimide GF 109203X is a potent and selective inhibitor of protein kinase C. J Biol Chem. 1991;266:15771–15781.
Rogers TB, Gaa ST, Massey C, Dosemeci A. Protein kinase C inhibits Ca2+ accumulation in cardiac sarcoplasmic reticulum. J Biol Chem. 1990;265:4302–4308.
Higashijima T, Uzu S, Nakajima T, Ross EM. Mastoparan, a peptide toxin from wasp venom, mimics receptors by activating GTP-binding regulatory proteins (G proteins). J Biol Chem. 1988;263:6491–6494.
Haga K, Kameyama K, Haga T. Synergistic activation of a G protein-coupled receptor kinase by G protein βγ subunits and mastoparan or related peptides. J Biol Chem. 1994;269:12594–12599.
Benovic JL, Stone WC, Caron MG, Lefkowitz RJ. Inhibition of the β-adrenergic receptor kinase by polyanions. J Biol Chem. 1989;264:6707–6710.
Loudon RP, Benovic JL. Expression, purification, and characterization of the G protein-coupled receptor kinase GRK6. J Biol Chem. 1994;269:22691–22697.
Cheng HC, Litwin CM, Hwang DM, Wang JH. Structural basis of specific and efficient phosphorylation of peptides derived from p34cdc2 by a pp60src-related protein tyrosine kinase. J Biol Chem. 1991;266:17919–17925.
Keefer JR, Limbird LE. The alpha 2A-adrenergic receptor is targeted directly to the basolateral membrane domain of Madin-Darby canine kidney cells independent of coupling to pertussis toxin-sensitive GTP-binding proteins. J Biol Chem. 1993;268:11340–11347.
Haller H, Quass P, Lindschau C, Luft FC, Distler A. Platelet-derived growth factor and angiotensin II induce different spatial distribution of protein kinase C-α and -β in vascular smooth muscle cells. Hypertension. 1994;24:848–852.
Arai H, Escobedo JA. Angiotensin II type I receptor signals through raf-1 by a protein kinase C-dependent, ras-independent mechanism. Mol Pharmacol. 1996;50:522–528.
Lattion A-L, Diviani D, Cotecchia S. Truncation of the receptor carboxyl terminus impairs agonist-dependent phosphorylation and desensitization of the α1β-adrenergic receptor. J Biol Chem. 1994;269:22887–22893.
Tobin AB, Lambert DG, Nahorski SR. Rapid desensitization of muscarinic m3 receptor-stimulated polyphosphoinositide responses. Mol Pharmacol. 1992;42:1042–1048.
Pfeilschifter J, Ochsner M, Whitebread S, De Gasparo M. Down-regulation of protein kinase C potentiates angiotensin II-stimulated polyphosphoinositide hydrolysis in vascular smooth-muscle cells. Biochem J. 1989;262:285–291.
Brass LF. Homologous desensitization of HEL cell thrombin receptors. J Biol Chem. 1992;267:6044–6050.
Richardson RM, Haribabu B, Ali H, Snyderman R. Cross-desensitization among receptors for platelet activating factor and peptide chemoattractants. J Biol Chem. 1996;271:28717–28724.
Inglese J, Freedman NJ, Koch WJ, Lefkowitz RJ. Structure and mechanism of the G protein-coupled receptor kinases. J Biol Chem. 1993;268:23735–23738.
Palczewski K, Buczylko J, Lebioda L, Crabb JW, Polans AS. Identification of the N-terminal region in rhodopsin kinase involved in its interaction with rhodopsin. J Biol Chem. 1993;268:6004–6013.
Chen C-Y, Dion SB, Kim CM, Benovic JL. β-Adrenergic receptor kinase: agonist-dependent receptor binding promotes kinase activation. J Biol Chem. 1993;268:7825–7831.
Pitcher JA, Touhara K, Payne ES, Lefkowitz RJ. Pleckstrin homology domain-mediated membrane association and activation of the β-adrenergic receptor kinase requires coordinate interaction with Gβγ subunits and lipid. J Biol Chem. 1995;270:11707–11710.
Tobin AB, Keys B, Nahorski SR. Identification of a novel receptor kinase that phosphorylates a phospholipase C-linked muscarinic receptor. J Biol Chem. 1996;271:3907–3916.