Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Diminished receptor responsiveness after exposure to agonist is a feature common to many G-protein-coupled receptors (Dohlman et al., 1991). This loss of receptor responsiveness or desensitization may be agonist specific (homologous) or may attenuate responses to multiple agonists through their receptors (heterologous). These processes have been studied extensively in the cAMP-coupled beta-2 adrenergic receptor and the phosphodiesterase-coupled receptor rhodopsin. In these receptor systems, rapid desensitization is largely caused by direct phosphorylation of the receptor at serine and threonine residues, predominantly in the C-terminus (Dohlman et al., 1991). Receptor phosphorylation is mediated by receptor-specific kinases and general kinase systems such as PKA and PKC (Lefkowitz, 1993). These kinase systems provide a mechanism for regulating receptor activity through negative feedback loops (homologous desensitization) as well as cross-talk between different receptor systems (heterologous desensitization).

Less is known about the mechanisms of desensitization of receptors coupled to PLC. Although phosphorylation of receptors may play a role in regulation of PLC-linked receptors (Albus et al., 1995; Leeb-Lundberg et al., 1987; Takano et al., 1994; Tobin and Nahorski, 1993), accumulating evidence indicates that desensitization of these receptors may differ from the beta-2 adrenergic receptor and rhodopsin (Wojcikiewicz et al., 1993). In this regard, the effects of PLC-linked receptors may be modulated by mechanisms downstream of receptor-G-protein coupling (Wojcikiewicz et al., 1993). Thus, chronic stimulation of receptors coupled to PLC has been shown to: 1) down-regulate PKC isotypes (Kiley et al., 1991), 2) phosphorylate and inactivate PLC (Ryu et al., 1990) and 3) desensitize receptor-induced calcium signaling by inositol trisphosphate receptor down-regulation (Honda et al., 1995:, Wojcikiewicz and Nahorski, 1991). These regulatory processes may differentially regulate the dual signaling pathway stimulated by PLC activation through mechanisms not requiring direct receptor phosphorylation (Wojcikiewicz et al., 1993). Indeed, Thomas et al. (1995) found that calcium transients induced by a truncated mutant of the type 1A angiotensin II receptor are desensitized normally despite the absence of 13 potential phosphorylation sites in the C-terminus.

In the present study, we investigated the rapid regulation of receptors for TxA₂. This labile lipid mediator is a potent platelet-aggregating and vasoconstrictor eicosanoid which has been implicated in the pathogenesis of diseases affecting the heart, lungs, kidneys and peripheral vascular system (Fitzgerald et al., 1987; Oates et al., 1988; Stork et al., 1986). Its effects are mediated by activating specific cell surface receptors (Mais et al., 1985; Spurney et al., 1993a). In most cell systems, receptor activation stimulates PLC through pertussis toxin-insensitive G-proteins (Offermans et al., 1994; Shenker et al., 1991). Alternative splicing has also been shown to produce two isoforms of the human TxA₂ receptor (Raychowdhury et al., 1994) which, in addition to coupling to PLC, oppositely regulate adenylyl cyclase activity (Hirata et al., 1996). Because the half-life of TxA₂ is short (Oates et al., 1988), the actions of TxA₂ might be limited both by its biochemical instability and by receptor desensitization. Previous studies (Dorn and Davis, 1992; Spurney et al., 1994) have found that desensitization of TxA₂ receptors is mediated, at least in part, through activation of PKC, perhaps through direct phosphorylation of the receptor protein. In support of this hypothesis, Kinsella et al. (1994) demonstrated that PKC can phosphorylate C-terminal sequences of the TxA₂ receptor in vitro. This suggests that the TxA₂ receptor might be a substrate for PKC in vivo and that negative feedback loops involving protein kinases may regulate responsiveness of TxA₂ receptors.

To investigate the role of C-terminal domains in the rapid regulation of the TxA₂ receptor, we constructed a mutant TxA₂ receptor lacking 22 amino acids in the C-terminus, including four potential phosphorylation sites. Studies using this mutant receptor suggest that the C-terminus is involved in receptor-effector coupling and in mediating homologous desensitization, and that heterologous desensitization of the TxA₂ receptor by PKC is complex and depends, in part, on the C-terminal domains of the TxA₂ receptor.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Isolation and Mutagenesis of a Genomic Clone Encoding the Mouse TxA₂ Receptor

A genomic clone encoding the mouse TxA₂ receptor was isolated as previously described (Coffman et al., 1994) by use of the following strategy. PCR primers were prepared based on the published sequences of the mouse TxA₂ receptor cDNA (Namba et al., 1992). The primer pairs encompassed nucleotides 534-553 (CTCTTGGTGCTTCCTGACAC) and 972-953 (CTGGAGCTGTGAACTG AACC) of the TxA₂ receptor cDNA (Namba et al., 1992). A PCR product of the appropriate size was amplified from total RNA isolated from mouse lung, and its sequence was found to be homologous to previously described TxA₂ receptors (Namba et al., 1992). With this partial cDNA as a probe, we isolated three identical clones from a genomic DNA library made from 129/Ola mouse DNA. Portions of one of the clones were sequenced and found to contain the complete coding sequences for the TxA₂ receptor protein. We subcloned a XhoI/ApaI fragment of this genomic clone containing the complete amino acid encoding regions into the mammalian expression vector pcDNA 3 (Invitrogen, San Diego, CA). To create a C-terminal truncation mutant, we inserted a synthetic oligonucleotide containing an in-frame stop codon into an unique SacII restriction site located at nucleotide 955 (Namba et al., 1992) in the C-terminus of the TxA₂ receptor.

Culture and Transfection of a Mouse Mesangial Cell Line

Mouse mesangial cells derived from SV40 transgenic mice (Mackay et al., 1988) were obtained from American Type Culture Collection (Rockville, MD). Cells were grown in 75% DMEM and 25% F-12 nutrient medium (HAMS) supplemented with 5% heat-inactivated fetal calf serum, 14 mM HEPES, penicillin (100 U/ml) and streptomycin (100 µg/ml) (all from Gibco, Gaithersburg, MD) at 37°C in a humidified atmosphere of 95% air and 5% CO₂. Mesangial cells were subcultured every week after becoming confluent by use of 0.25% trypsin with 1 mM EDTA (Gibco, Gaithersburg, MD) and plated at a density of 2 to 5 × 10⁵ cells/ml. Cell viability was assessed by standard dye exclusion techniques (0.1% trypan blue) and was always greater than 95%.

To create cell lines expressing mutant TxA₂ receptors, our pcDNA 3 expression vector containing either the wild-type or mutant construct was transfected into SV40-transformed mouse mesangial cells by the calcium-phosphate method (Maniatis et al., 1989). To isolate permanent transfectants, G418-resistant cells were selected in complete medium containing 500 µg/l G418. After G418 selection, individual clones were screened for TxA₂ binding as described below.

Ligand Binding Assays

Preparation of mesangial cell suspensions. Confluent cultures of mesangial cells were incubated with 0.25% trypsin and 1 mM EDTA (Gibco, Gaithersburg, MD). After detachment, cells were immediately centrifuged at 500 × g for 5 min, and the pellet was suspended in ice-cold Tris-saline buffer (50 mM Tris-HCl, 150 mM NaCl, pH 7.4). The suspension was centrifuged at 500 × g for 5 min, and the pellet was resuspended in ice-cold Tris-saline buffer for the binding assays as described below. An aliquot of the cell suspensions was sonicated and frozen at 20°C until protein concentration was determined by the method of Bradford (1976).

[³H]SQ29548 binding studies. Immediately after preparation of mesangial cell suspensions, TxA₂ binding was assessed with the stable radiolabeled thromboxane receptor antagonist [³H]SQ29548 (Ogletree et al., 1985) (New England Nuclear, Boston, MA). Binding studies were performed at room temperature with Tris-saline buffer (50 mM Tris-HCl, 150 mM NaCl) adjusted to pH 7.4. In the standard binding assay, 20 µl of [³H]SQ29548 was added to 80 µl of the mesangial cell suspension (typically 40-80 µg of protein) to create a 100-µl solution containing 1 to 40 nM concentrations of [³H]SQ29548. Samples were incubated for 1 hr at room temperature to allow the reaction to reach equilibrium (Spurney et al., 1993b), and then the reaction was stopped by the addition of 4 ml of ice-cold Tris-saline buffer. Samples were rapidly filtered over Whatman GF/C filters, and each filter was washed twice with 4 ml Tris-saline buffer. After drying, filters were analyzed for ³H with a model 1191 TM Analytic gamma counter (Brandon, FL). Nonspecific binding was determined by measuring the amount of radioactivity bound in the presence of excess concentrations (10 µM final concentration) of unlabeled SQ29548. Specific binding of the radioligand typically averaged 5 to 30% of the total binding over the concentration range studied and varied directly with the protein concentration.

Equilibrium binding assays. Mesangial cell suspensions were incubated for 60 min with 0.1 to 40 nM concentrations of [³H]SQ29548. Equilibrium binding data were analyzed by the method of Scatchard (1949) to give estimates of the maximal number of specific binding sites (B_max) and apparent equilibrium K_d by fitting the data to a nonlinear model by use of the ENZFITTER computer program (Elsevier-Biosoft, Cambridge, UK). Data are expressed as femtomoles per milligram of protein.

Competitive displacement assays. Mesangial cell suspensions were incubated for 60 min with 10 nM [³H]SQ29548 and the following unlabeled compounds in concentrations varying from 10¹² to 10⁴ M: the thromboxane receptor antagonist SQ29548 (Ogletree et al., 1985) (Squibb Institute, Princeton, NJ), the thromboxane agonists U46619 (Coleman et al., 1981) (Cayman Chemicals, Ann Arbor MI) or [¹²⁷I]BOP (Morinelli et al., 1989) (Cayman Chemicals), the inactive thromboxane metabolite TxB₂ (Advanced Magnetics Inc., Cambridge, MA) or PGE₂ (Advanced Magnetics). Data were analyzed by the method of Cheng and Prusoff (1973) to calculate the dissociation constant for each inhibitor (K_i).

Signal Transduction

Measurement of inositol phosphate generation. Inositol phosphates were measured as described previously (Spurney et al., 1994). Mesangial cells were plated at a density 2 to 5 × 10⁵ cells/ml in six-well plastic culture dishes (9.5 cm²/well) (Costar, Cambridge, MA). After reaching confluence, cells were equilibrated for 24 hr in DMEM low inositol medium (Gibco, Gaithersburg, MD) with 0.1% dialyzed fetal calf serum, penicillin (100 U/ml) and streptomycin (100 µg/ml) (all from Gibco, Gaithersburg, MD) containing 5 µCi/ml myo-[³H]inositol (New England Nuclear, Boston, MA). The cultures were washed three times with 2 ml KRB and then incubated for the indicated times with the agents to be tested or their vehicle in 2 ml KRB at 37°C. For the desensitization experiments, 20 mM lithium chloride was included in the incubation medium to inhibit breakdown of inositol phosphates by the protocol described in detail below. The reaction was stopped by aspirating the medium, adding 240 µl of 3.3 N perchloric acid and then placing the samples on ice. After 15 min, cell monolayers were scraped off with a plastic spatula and transferred to a plastic microcentrifuge tube. The wells were washed with an additional 960 µl of 0.55 N perchloric acid, and this rinse was pooled with the initial 240-µl wash. Samples were centrifuged for 5 min at 10,000 × g, and exactly 1 ml of supernatant was transferred to another microcentrifuge tube. One milliliter of supernatant was neutralized by adding 55 µl of 10 N KOH and placing the samples on ice. After 15 min, neutralized samples were centrifuged at 10,000 × g for 5 min. For chromatography, 0.9 ml of supernatant was added to 9 ml of 5 mM sodium borate and applied to 1.0-ml columns packed with AG1-X8 formate anion exchange resin (Bio-Rad, Richmond, CA). Columns were washed twice with 10 ml of 5 mM sodium tetraborate containing 60 mM ammonium formate, and inositol phosphates were eluted sequentially with 3.5 ml of 0.1 M formic acid containing either 0.2, 0.4 or 1.0 M ammonium formate for IP1, IP2 and IP3, respectively. Columns were washed with 10-ml volumes of the appropriate elution buffer between collections of eluate fractions to remove any residual radioactivity. Eluate fractions were dissolved in 17 ml of Safety-Solve (Research Products International, Mount Prospect, IL) and were quantitated by liquid scintillation counting. With this method, radioactivity recovered in the IP3, IP2 and IP1 eluate fractions typically averaged 7%, 5% and 88% of the total radioactivity recovered in all eluate fractions, respectively. Basal IP3 generation was similar in wild-type and mutant receptors and averaged 2160 ± 397 dpm for cells expressing low numbers of wild-type receptors, 2014 ± 275 dpm for cells expressing high numbers of wild-type receptors, 1992 ± 317 for cells expressing low numbers of mutant receptors and 1815 ± 403 for cells expressing high numbers of mutant receptors.

Cytosolic calcium measurements. Intracellular calcium levels [Ca⁺⁺]_i were measured in confluent mesangial cells by fluorescence excitation of cells loaded with the fluorescent probe fura 2 as described previously (Spurney et al., 1994). Cells were grown on plastic Aclar cover slips (Pro Plastics, Linden, NJ) until confluence and then incubated for an additional 2 days in DMEM with 0.1% fetal calf serum, 14 mM HEPES and antibiotics. Cells were loaded for 1 hr in DMEM with fura 2-AM (Sigma, St. Louis, MO) added to the culture medium to a final concentration of 5 µM. Cells were then washed at 37°C in KRB containing 0.1% bovine albumin, pH 7.4, without fura 2-AM. After washing, cover slips were inserted diagonally into a cuvette and were continuously perfused with KRB. To ensure adequate loading and complete hydrolysis of the methylester, emission was measured at 510 nm over the excitation spectrum from 300 to 400 nm with a Perkin-Elmer LS50 fluorescence spectrometer (Norwalk, CT). Real-time measurements were monitored by alternating the excitation wavelength between 340 nm and 380 nm every 1.9 sec with a microcomputer, and data were derived from the ratio of emission at 510 nm. [Ca⁺⁺]_i was calculated as described by Grynkiewicz et al. (1985) by the following formula:

(1)

Statistical analysis. Data are presented as the mean ± S.E.M. Statistical significance was assessed by a paired or unpaired t test as indicated.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Membrane topology of the wild-type and mutant TxA₂ receptors. The predicted membrane topology of the TxA₂ receptor is shown in figure 1A. Black circles represent hydroxyl amino acids in the putative intracellular domains of the receptor which are potential phosphorylation sites for protein kinases. Our mutant TxA₂ receptor is truncated at amino acid 320, thus deleting 22 amino acids in the C-terminus including four potential phosphorylation sites. As shown in figure 1B, these four potential phosphorylation sites are highly conserved in the mouse, rat and human TxA₂ receptor.

View larger version (33K): [in this window] [in a new window]   Fig. 1.   Membrane topology of the TxA2 receptor. Panel A shows the proposed membrane topology of the mouse TxA2 receptor. Putative intracellular domains are rich in hydroxyl amino acids (shown by filled circles) which are potential phosphorylation sites for protein kinases. The mutant TxA2 receptor is truncated at amino acid 320 (indicated by the arrowhead), thus deleting 22 amino acids in the C-terminus including four potential phosphorylation sites. Panel B shows the amino acid residues deleted in the mouse TxA2 receptor mutant and the corresponding sequences in the rat and human TxA2 receptors (Abe et al., 1995; Hirata et al., 1991; Namba et al., 1992). Asterisks mark highly conserved hydroxyl amino acids.

The absence of C-terminal domains of the TxA₂ receptor does not affect TxA₂ binding. A mesangial cell line (Mackay et al., 1988) was transfected with plasmids containing either the wild-type or the mutant construct. After G418 selection, four wild-type and five mutant clones were isolated which exhibited stable levels of expression of the TxA₂ receptor protein. To control for the effect of receptor number on receptor coupling efficiency (Whaley et al., 1994), clones were matched for similar levels of "high" or "low" TxA₂ binding as shown in table 1. Competitive displacement assays found that the wild-type and mutant receptor displaced TxA₂ ligands with similar K_i values (table 2), which suggested that the absence of the C-terminus does not affect binding of TxA₂ to its receptor. TxA₂ binding was less than 20 fmol/mg of protein in nontransfected cells or cells transfected with vector alone (data not shown).

Receptor-effector coupling is impaired in TxA₂ receptors lacking C-terminal domains. The effects of the TxA₂ agonist U46619 on PLC activity is shown in figure 2 for clones expressing high receptor numbers (fig. 2A) and for clones expressing low receptor numbers (fig. 2B). In cells expressing the wild-type receptor, U46619 caused a prompt increase in IP3 generation, which decreased to 60% of the peak level by 10 min. In clones expressing the mutant TxA₂ receptor, a brisk increase in IP3 generation also occurred but the peak IP3 generation was only 50% of the wild type, and the rate of IP3 generation remained relatively stable during the 10-min period of the study. TxA₂-induced PI hydrolysis was negligible in nontransfected mesangial cells (fig. 2B).

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Time course of IP3 generation by wild-type and mutant TxA2 receptors. Clones were matched for similar levels of "high" (A) or "low" (B) TxA2 binding. After treatment with 10 µM U46619, IP3 generation increased promptly in clones expressing the wild-type receptor and then decreased to 60% of the peak level by 10 min. In clones expressing the mutant TxA2 receptor, the initial increase in IP3 generation was reduced 50% compared with the wild type, and the rate of IP3 generation remained relatively stable during the 10-min period of the study. TxA2-induced PI hydrolysis was not detectable in nontransfected mesangial cells. Experiments were performed in duplicate and data points are the mean of 4 to 12 experiments. *P < .05 versus the truncation mutant by paired t test.

Homologous desensitization of TxA₂ receptors is attenuated by deleting C-terminal domains. To investigate the role of C-terminal domains in TxA₂ receptor desensitization, mesangial cells expressing wild-type or mutant TxA₂ receptors were incubated with 10 µM U46619 before washing and rechallenge with either 10 nM, 100 nM, 1 µM or 10 µM concentrations of the TxA₂ agonist. As shown in figure 3A, prior incubation with U46619 reduced subsequent TxA₂-induced increases in IP3 generation in clones expressing wild-type receptors. Pretreatment with TxA₂ agonist also caused some degree of desensitization of mutant receptors (fig. 3B). To normalize for differences in the base-line response, data were expressed as the percent response in vehicle-treated cells, as shown in table 4. When the data are normalized in this fashion, there was significantly less desensitization in cells expressing the mutant TxA₂ receptor than in clones expressing the wild type. Similar results were seen for clones expressing lower numbers of TxA₂ receptors (table 5). These data suggest that desensitization of IP3 responses are attenuated by the absence of C-terminal domains of the TxA₂ receptor.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Concentration-response curves for wild-type and mutant TxA2 receptors. Clones with "high" levels of TxA2 binding were studied. For these experiments, clones expressing the wild-type receptor (A) or clones expressing the receptor mutant (B) were treated with vehicle or 10 µM U46619, washed to remove bound agonist and then rechallenged with either 10 nM, 100 nM, 1 µ M or 10 µM concentrations of the TxA2 agonist, as described under "Materials and Methods." IP3 generation was measured in cells rechallenged with U46619. Prior incubation with U46619 reduced subsequent U46619-induced IP3 generation in clones expressing the wild-type receptor. Although pretreatment with U46619 caused some desensitization of mutant receptors, the extent of TxA2 receptor desensitization was reduced compared with the wild type. Experiments were performed in duplicate and data points are the mean of 4 to 11 experiments. *P < .025 versus vehicle-treated cells by a paired t test.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Homologous desensitization of calcium responses. Clones with "high" levels of TxA2 binding were studied. For these experiments, cells expressing the wild-type receptor (A) or cells expressing the receptor mutant (B) were stimulated with 10 µM U46619 before washing and rechallenge with 10 µM U46619. [Ca++]i levels were measured by fluorescence excitation of fura 2-loaded cells as described under "Materials and Methods." Prior exposure to U46619 similarly reduced subsequent TxA2-induced increases in [Ca++]i levels in cells expressing either the wild-type receptor or the mutant receptor.

PKC-induced desensitization is altered by deleting C-terminal domains of the TxA₂ receptor. Previous studies suggest that PKC modulates TxA₂ receptor responsiveness (Dorn and Davis, 1992; Spurney et al., 1994). To investigate the role of C-terminal domains in PKC-induced desensitization of the TxA₂ receptor, we first tested the ability of the PKC inhibitor staurosporine (Tamaoki, 1991) to inhibit homologous desensitization of wild-type and mutant TxA₂ receptors. For these studies, clones expressing either the wild-type or mutant TxA₂ receptor were incubated with 10 µM U46619 in the presence or absence of 200 nM staurosporine. After 10 min, cells were washed and then rechallenged with 10 µM U46619. As shown in figure 5, staurosporine attenuated homologous desensitization in clones expressing the wild-type TxA₂ receptor. In contrast, staurosporine had little effect on homologous desensitization in clones expressing the mutant receptor. These data suggest that a staurosporine-sensitive kinase contributes to homologous desensitization of wild-type, but not mutant, TxA₂ receptors.

View larger version (36K): [in this window] [in a new window]   Fig. 5.   Effect of staurosporine on homologous desensitization. Clones with "high" levels of TxA2 binding were studied. For these experiments, clones expressing either the wild-type or mutant TxA2 receptor were incubated with 10 µM U46619 in the presence or absence of 200 nM staurosporine as described under "Materials and Methods." After 10 min, cells were washed and then rechallenged with 10 µM U46619. IP3 generation was measured in cells rechallenged with U46619. Staurosporine attenuated homologous desensitization in clones expressing the wild-type TxA2 receptor but had little effect on desensitization in clones expressing the mutant receptor. Experiments were performed in triplicate and data points are the mean of 20 experiments. *P < .05 versus vehicle-treated cells by an unpaired t test, P < .025 versus cells pretreated with U46619 by an unpaired t test.

View larger version (32K): [in this window] [in a new window]   Fig. 6.   Effect of phorbol esters on TxA2 receptor responsiveness. Clones with "high" levels of TxA2 binding were studied. For these experiments, clones expressing wild-type or mutant TxA2 receptors were incubated with either 10 nM, 100 nM or 1 µM concentrations of the PDBu before washing and stimulation with 10 µM U46619 as described under "Materials and Methods." IP3 generation was measured in cells stimulated with U46619. Pretreatment with 10 nM PDBu had little effect on TxA2 receptor responsiveness. In contrast, 100 nM PDBu significantly reduced U46619-induced PI hydrolysis by the wild-type receptor but had little effect on responsiveness of the receptor mutant. PDBu (1 µM) reduced responsiveness of both wild-type and mutant TxA2 receptors. Experiments were performed in triplicate and data points are the mean of 4 to 11 experiments. *P < .025 versus vehicle-treated cells by an unpaired t test.

GF109203X attenuates desensitization induced by phorbol esters. To determine whether PDBu-induced desensitization is mediated by PKC, we tested the ability of the specific PKC inhibitor GF109203X (Toullec et al., 1991) to block the loss of receptor responsiveness caused by 100 nM and 1 µM concentrations of PDBu in cells expressing high levels of wild-type TxA₂ receptors. As shown in table 6, 200 nM GF109203X blocked the effects of 100 nM PDBu, but was unable to completely inhibit desensitization induced by 1 µM PDBu. These data suggest that the effects of 100 nM PDBu are mediated through PKC activation. The inability of GF109203X to block the effects of 1 µM PDBu could result from either PKC-independent mechanisms of desensitization activated by the phorbol ester or from an inability of 200 nM GF109203X to completely block high levels of PKC activation induced by 1 µM PDBu.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

TxA₂ is a labile lipid mediator with potent platelet-aggregating and vasoconstrictor effects. These actions play a pivotal role in the pathogenesis of diseases affecting the heart, lungs, kidneys and peripheral vascular system (Fitzgerald et al., 1987; Oates et al., 1988; Stork et al., 1986). The effects of TxA₂ are tightly regulated. Persistent activation of the TxA₂ receptor leads to a loss of receptor responsiveness (Dorn et al., 1992; Spurney et al., 1994). This desensitization is mediated, at least in part, by protein kinases (Dorn et al., 1992; Spurney et al., 1994), perhaps by direct phosphorylation of the receptor protein (Kinsella et al., 1994). In the present study, we investigated the role of the C-terminus in the rapid regulation of TxA₂ receptor responsiveness. We found that TxA₂ receptors lacking C-terminal domains activated PLC less efficiently than wild type receptors, which suggested a role for the C-terminus in receptor-effector coupling. Moreover, agonist-specific desensitization of the mutant receptor was attenuated compared with wild-type receptors, which indicated that domains within the C-terminal 22 amino acids modulate the extent of homologous desensitization.

We and others have shown previously that homologous desensitization of TxA₂ receptors is mediated, at least in part, by PKC (Dorn et al., 1992; Spurney et al., 1994). To investigate the role of C-terminal domains in PKC-induced desensitization of the TxA₂ receptor, we first studied the effect of the PKC inhibitor staurosporine on homologous desensitization of TxA₂ receptors. We found that staurosporine attenuated TxA₂-induced desensitization of wild-type receptors, which suggested a role for a staurosporine-sensitive kinase such as PKC in modulating TxA₂ receptor responsiveness. We also found that staurosporine had little effect on TxA₂-induced desensitization of mutant receptors and that pretreatment with 100 nM PDBu decreased subsequent responsiveness of wild-type, but not mutant, TxA₂ receptors. Moreover, the specific PKC inhibitor GF109203X (Toullec et al., 1991) could block desensitization induced by 100 nM PDBu. Taken together, these findings suggest that domains within the C-terminus are involved in the regulation of the TxA₂ receptor by PKC. This conclusion is further supported by: 1) the presence of several consensus sequences for phosphorylation by PKC (Pearson and Kemp, 1991) in the C-terminal 22 amino acids deleted in the mutant receptor, and 2) the demonstration by Kinsella et al. (1994) that PKC can phosphorylate C-terminal sequences of the TxA₂ receptor in vitro.

At higher doses (1 µM), PDBu was equally effective in reducing responsiveness of both wild-type and mutant TxA₂ receptors. These findings indicate that regulation of TxA₂ receptor activity by PKC is complex and also involves mechanisms that do not require the C-terminus. These regulatory mechanisms might be activated by higher levels of PKC stimulation leading to either phosphorylation of additional domains of the TxA₂ receptor outside of the C-terminal 22 amino acids or phosphorylation of other substrates such as downstream components of the signaling pathway. Evidence for each of these possibilities has been demonstrated in other receptor systems. For example, phosphorylation of hydroxyl amino acids in the third intracellular loop has been implicated in heterologous desensitization of the beta adrenergic receptor (Dohlman et al., 1991). In addition, activation of PKC by phorbol esters has been shown to phosphorylate and inactivate PLC- (Ryu et al., 1990). Further studies will be needed to determine whether these regulatory mechanisms contribute to TxA₂ receptor regulation by PKC.

The reduced ability of our truncation mutant to activate PLC indicates that domains in the C-terminal 22 amino acids are important for coupling of the TxA₂ receptor to its G-protein. In other G-protein-coupled receptors, deletions or point mutations in C-terminal domains have also been shown to inhibit receptor effector coupling; however, these mutations have generally involved large deletions of the C-terminus including most of its amino-terminal domains (Buck et al., 1991; Ohyama et al., 1992), or substitution mutations of amino-terminal portions of the C-terminus (O'Dowd et al., 1988, 1989). With regard to the latter, a single point mutation of a cysteine residue located in the C-terminus of the beta-2 adrenergic receptor markedly inhibits receptor-effector coupling (O'Dowd et al., 1989). This cysteine residue is highly conserved in G-protein-coupled receptors (O'Dowd et al., 1989) but is not found in either the human, rat or mouse TxA₂ receptor (Abe et al., 1995; Hirata et al., 1991; Namba et al., 1992). These data indicate that TxA₂ receptor coupling may differ from other prototypical G-protein-coupled receptors. In this regard, Hirata et al. (1996) found that C-terminal domains of the human TxA₂ receptor determine the specificity of coupling to adenylyl cyclase. Moreover, portions of the first intracellular loop (Hirata et al., 1996) and third intracellular loop (D'Angelo et al., 1996) have both been implicated in coupling of the TxA₂ receptor to PLC. Taken together, these data suggest that multiple receptor determinants are involved in coupling of the TxA₂ receptor to its effector systems.

The absence of C-terminal domains attenuated, but did not prevent, homologous desensitization of the TxA₂ receptor. These data indicate that agonist-specific desensitization of the TxA₂ receptor is mediated by multiple mechanisms, some of which do not require the C-terminus. These mechanisms may include: 1) sequestration of receptors away from the cell surface, 2) inhibition of other components of the signaling cascade or 3) phosphorylation of domains other than the C-terminus. Regarding the third mechanism, receptor-specific kinases have been shown to play a key role in regulating receptor responsiveness in other receptor systems (Dohlman et al., 1991; Lefkowitz, 1993). Although the consensus motif for phosphorylation by receptor-specific kinases is not fully elucidated, serine or threonine residues flanked by acid amino acids (D/ES/T or D/EXS/T) appear to be preferred substrates for beta adrenergic receptor-specific kinases (Onorato et al., 1991). Motifs of this configuration are found in the third intracellular loop of both the mouse, rat and human TxA₂ receptor (Abe et al., 1995; Hirata et al., 1991; Namba et al., 1992) and might play a role in the regulation of TxA₂ responsiveness.

In contrast to PI hydrolysis, calcium responses desensitized similarly in clones expressing either the wild-type or mutant TxA₂ receptor. Moreover, in cells expressing high levels of TxA₂ receptors, calcium responses were similar in both wild-type and mutant clones despite lower rates of IP3 generation by the mutant receptor. These data suggest several possibilities. First, calcium responses appear to be nonlinear above a certain threshold; thus, further increases in PI hydrolysis do not necessarily result in further increases in [Ca⁺⁺]_i levels as has been reported in other receptor systems (Takano et al., 1994). Second, when PI hydrolysis falls below this threshold level, calcium responses rapidly attenuate in cells expressing either wild-type or mutant TxA₂ receptors, perhaps accounting for the similar desensitization of calcium responses in clones expressing either receptor type. Because of this nonlinear relationship between PI hydrolysis and calcium responses, measuring components of the signaling cascade distal to receptor-effector coupling may not detect subtle differences in receptor coupling or desensitization. This potential problem might account for the failure of investigators to find an effect of C-terminal truncation on coupling and desensitization in several other PLC-coupled receptors (Cyr et al., 1993; Thomas et al., 1995).

In summary, the present studies indicate that C-terminal domains of the TxA₂ receptor participate in coupling of the receptor to its effector systems and contribute to homologous desensitization of TxA₂ receptor. Heterologous regulation of the TxA₂ by PKC is complex and is mediated by mechanisms that are, in part, dependent on C-terminal domains of the TxA₂ receptor.