Regulation of Thromboxane Receptor Activation in Human Platelets (Desensitization/Uncoupling/Down-Regulation) ROSEMARY MURRAY and GARRET A

Home , Thromboxane receptor, U46619

Proc. Natl. Acad. Sci. USA Vol. 86, pp. 124-128, January 1989 Biochemistry Regulation of thromboxane receptor activation in human platelets (desensitization/uncoupling/down-regulation) ROSEMARY MURRAY AND GARRET A. FITZGERALD* Division of Clinical Pharmacology, Vanderbilt University, Nashville, TN 37232 Communicated by Stanley Cohen, October 7, 1988

ABSTRACT Thromboxane A2 (TxA2) is a potent platelet were gifts from P. Halushka (Medical University of South agonist that serves as an amplifying signal after exposure of Carolina); {1S-[1,2(5Z),3(1E,3S),4]}-7-{3-[3-hydroxy-1-ocen- platelets to other stimulants, such as thrombin, in vitro. tyl]-7-oxabicyclo-(2.2.2)hept-2-yl}-5-heptenoic acid (SQ Exposure of platelets to the TxA2 receptor agonists U46619 and 26,655) and {1S-[1,2 (5Z),3,4]}-7- (3-{[2-(phenylaminocarbon- SQ 26,655 (1.4 ,AM) resulted in a 60-90% decrease in subse- yl)hydrazino]methyl}-7-oxabicyclo(2.2. 1)hept-2-yl)-5-hep- quent TxA2 receptor-stimulated aggregation, calcium release, tenoic acid (SQ 29,548) were gifts from M. Ogletree (Squibb and protein kinase C activation. The desensitization was rapid, Institute, Princeton, NJ); (15S)-hydroxy-11,9-(epoxymetha- with a half-time of 2-3 min. The sequence of events involved in no)prosta-5Z,13E-dienoic acid (U46619) was a gift from R. TxA2 receptor desensitization involves initial uncoupling ofthe Gorman (Upjohn, Kalamazoo, MI). Reagents for polyacryl- receptor from a guanine nucleotide binding (G) protein fol- amide gel electrophoresis were from Bio-Rad. Other reagents lowed by eventual receptor down-regulation. Consistent with were from Sigma except where indicated otherwise. this hypothesis were (i) a 60-70% decrease in SQ 26,655- Desensitization and Preparation of Washed Platelets. Blood stimulated platelet GTPase activity, (ii) a shift to the right ofthe was drawn from healthy human volunteers (who had not dose-response curve for U46619-stimulated release of calcium taken any medication for at least 10 days) by venipuncture [ECso, 275 ± 51 nM (control)] vs. 475 ± 71 nM (desensitized); into EDTA (5 mM) and indomethacin (5 Iug/ml) (final P < 0.01], and (iii) a delayed loss ofreceptor sites. In summary, concentrations). The blood was centrifuged for 10 min at 160 exposure of platelets to TxA2 receptor agonists results in rapid x g, and platelet-rich plasma (PRP) was removed. For desensitization of the biochemical and functional responses to desensitization, PRP was incubated at room temperature for interaction with its receptor in human platelets. The kinetics of 30 min (or as indicated in individual experiments) with the these events are consistent with the hypothesis that this TxA2 receptor agonists U46619 (1.4 ,uM) (5), SQ 26,655 (1.4 icosanoid functions in the regulation as well as amplification of ,uM) (6), or vehicle (control). The platelets were washed as platelet activation in vivo. described (7). All buffers contained indomethacin at 5 ,ug/ml. Platelet Aggregation. Platelets (0.5 ml; -108 platelets per Thromboxane A2 (TxA2) is a potent agonist of platelet acti- ml) were preincubated at 37°C for 2 min, then stimulant was vation, causing shape change, secretion, and aggregation (1). added, and the extent of aggregation was monitored by light TxA2 is also an important potentiator ofplatelet activation by transmission in a Biodata Pap 4 aggregometer. other platelet stimulants, such as ADP and epinephrine. Calcium Measurement. Intracellular free calcium (Ca2+) These agents stimulate the release and subsequent metabo- measurements were made by monitoring the intensity of lism to TxA2 ofarachidonic acid (2, 3). TxA2, largely derived fura-2 fluorescence. PRP was incubated with 1 ,uM fura-2 AM from platelets, is synthesized in vivo even under physiolog- (Molecular Probes) at 37°C for 30 min. The fura-2-loaded ical conditions (4), implying continuous platelet activation, platelets were desensitized and washed as described above. which would seem likely to be subject to regulatory con- Changes in fluorescence of the fura-2-loaded platelets at straints. One potential mechanism whereby platelet function excitation wavelengths of 340 and 380 nm (less autofluores- might be controlled would be the decreased response of cence of the platelets in the absence of fura-2 loading) after platelets to TxA2 after exposure to agonist, or desensitiza- addition of various agonists were recorded. The ratio of tion. fluorescence at excitation wavelength 340 nm to fluorescence We have characterized the molecular events involved in at excitation wavelength 380 nm (relative fluorescence inten- TxA2 receptor desensitization by comparing the biochemical sity) is a measure of Ca71' concentration (8). responses to TxA2 receptor stimulation in untreated and Phosphorylation of 47-kDa Protein. Phosphorylation of desensitized platelets. Our results indicate that desensitiza- platelet 47-kDa protein, a protein kinase C substrate, was tion of platelet TxA2 receptors is mediated initially by assessed as described (9). After two washes in phosphate-free impaired coupling between the TxA2 receptor and its second buffer, control and desensitized platelets were resuspended messenger system-guanine nucleotide binding (G) protein- to a concentration of 2 x 109 platelets per ml in the same coupled phospholipase C activation. Uncoupling of receptor buffer [15 mM Tris-HCl, pH 7.4/150 mM NaCl/5 mM stimulation from phospholipase C activation is followed, dextrose/indomethacin (5 ,ug/ml)], also containing 500 ,uM upon extended exposure to agonist, by a decrease in platelet CaCl2 and 200 ,Ci of 32p orthophosphate per ml [carrier-free TxA2 receptor density. in HCl, neutralized with 1 M Tris (pH 7.4); 1 Ci/ml; 1 Ci = 37 GBq]. Platelets (5 x 107) were stimulated at room temperature for 60 sec, subjected to 10% SDS/polyacryl- MATERIALS AND METHODS amide gel electrophoresis (10), and autoradiographed. The Materials. 9,11-Dimethylmethano-11,12-methano-16-(4- hydroxyphenyl)-13,14-dihydro-13-aza-15-al8w-tetranor- Abbreviations: TxA2, thromboxane A2; G protein, guanine nucleo- TxA2 (PTA-OH) and 1271-labeled PTA-OH ('27I-PTA-OH) tide binding protein; Ca+,2 intracellular free calcium; PRP, platelet- rich plasma; PTA-OH, 9,11-dimethylmethano-11,12-methano- 16-(4-hydroxyphenyl)-13,14-dihydro-13-aza-15-acfw-tetranor-TxA2; The publication costs of this article were defrayed in part by page charge 127I-PTA-OH, 127I-labeled PTA-OH; 3H-SQ 29,548, 3H-labeled SQ payment. This article must therefore be hereby marked "advertisement" 29,548. in accordance with 18 U.S.C. §1734 solely to indicate this fact. *To whom reprint requests should be addressed.

124 Downloaded by guest on September 30, 2021 Biochemistry: Murray and FitzGerald Proc. Natl. Acad. Sci. USA 86 (1989) 125

molecular mass of the most heavily phosphorylated protein I 100 A * 100 B after stimulation was 47 kDa by comparison with concomi- 0 tantly run molecular mass standards. The extent ofphospho- 80 80 rylation in the 47-kDa protein was assessed by laser densitometry. o0 60 60 Binding Studies. Saturation binding. 125I-labeled PTA-OH [prepared by iodination of the TxA2 receptor antagonist -. 40- PTA-OH (11) or obtained from Amersham (2000 Ci/mmol)] 40 diluted 1:50 with 100 nM 1271-PTA-OH to give final concen- 0 trations of 5-60 nM; or 3H-labeled SQ 29,548 (3H-SQ 29,548) ~20- 20- CD, (from DuPont-New England Nuclear; 40 Ci/mmol) at final uJ concentrations of 2.5-30 nM, in 50 mM Tris-HCI, pH 7.4/150 1 2 3 20 40 60 mM NaCI/5 mM dextrose/1% bovine serum albumin (bind- CD ing buffer) was incubated with -5 x 107 washed platelets in STIMULATION TIME (min) a final volume of 200 ,ul. Nonspecific binding was measured FIG. 1. Desensitization of the platelet aggregation response. (A) by including 10 AM 127I-PTA-OH or unlabeled SQ 29,548. Platelets were desensitized with U46619 (open bars) or SQ 26,655 Incubations were done in duplicate at 370C for 30 min. The (solid bars) and subsequently stimulated with U46619 (1 MM), SQ reaction was terminated by dilution with 4 ml of ice-cold 10 26,655 (1 AM), or thrombin (0.1 unit/ml). (B) Platelets were desen- mM Tris HCl (pH 7.4) and the reaction mixture was imme- sitized with SQ 26,655 for 5, 15, 30, or 60 min prior to stimulation of diately filtered through Whatman GF/C glass fiber filters aggregation with 1 ,uM SQ 26,655. Results are expressed as percent- (Fisher). The filters were then washed with an additional 3 x age stimulation in control platelets and represent the mean + SEM 4 ml of 10 mM Tris HCl buffer. of 10 (A) and 5 (B) experiments. ***, P < 0.001; *, P < 0.05. Competition binding. 3H-SQ 29,548 (0.5 nM) was incubated with -5 x 107 platelets in the presence of various exposure to the TxA2 receptor agonists. The half-time of concentrations (1 nM to 10 AM) of U46619 or PTA-OH in desensitization was 2-3 min (Fig. 1B). binding buffer in a total volume of 200 ,ul. Incubations were The specificity of the induction of desensitization by done in duplicate at 37°C for 30 min and were terminated as interaction of the TxA2 receptor agonists with the TxA2 re- described for saturation binding assays. After addition of 10 ceptor was demonstrated by the inhibition ofthe induction of ml of scintillation fluid, the filters were assayed by liquid desensitization in the presence of the TxA2 receptor antag- scintillation counting. onist SQ 29,548 (Fig. 2). Binding analyses were performed by computerized non- Platelet aggregation has been associated with activation of linear curve fitting using the LIGAND program, as described phospholipase C, with resultant production of inositol by Munson and Robard (12). trisphosphate and 1,2-diacylglycerol, which stimulate release GTPase Assays. Stimulation of platelet high-affinity GTP- of Ca2' and the activation of protein kinase C, respectively. ase activity was assessed as an indication of receptor-G Resting Caj2+ concentrations were identical in control and protein coupling (13). Platelet membranes were prepared, desensitized platelets (0.68 ± 0.01 and 0.65 ± 0.03 fluores- following desensitization, as described (14). GTPase activity cence units, respectively). U46619 (1 MM) or SQ 26,655 (1 was measured essentially by the method of Cassel and AM) stimulation ofCa?2+ release in platelets desensitized with Selinger (15) with minor modifications. The reaction mixture, either TxA2 receptor agonist was attenuated to 60-70% of in a final volume of 50 ,ul, contained 5 mM MgCl2, 1 mM that obtained in control platelets (Fig. 3A). Thrombin stim- p[NH]ppA, 0.25 mM ATP, 1 mM phosphocreatine, creati- ulation ofCa?' release was significantly (P < 0.05) decreased nine phosphokinase (50 units/ml), 1 mM dithiothreitol, 1 mM by 27% ± 6% in desensitized compared with control EGTA, 0.75 ,uM ['y-32P]GTP (30 Ci/mmol; Amersham), 0.1 M plate- NaCl, and 10 mM Tris-HCl (pH 7.4). High-affinity GTPase lets. The decrease in agonist stimulation of Cai?' release was activity was defined as that inhibited by the inclusion of 1 mM significantly greater for the TxA2 receptor agonists than for measured in with thrombin (P < 0.001). [Control stimulation ofCa?+ release by GTP in the assay. Each point was triplicate 1 MM U46619 was 0.61 ± 0.1; by 1 mM SQ 26,655, it was 1.24 and without 1 mM GTP. After incubation for 10 min at 37°C, + ± 1 ml of ice-cold 5% charcoal (Norit A) in 20 mM phosphate 0.15; and by thrombin (0.1 unit/ml), it was 1.6 0.2 (pH 7.5) was added, and the mixture was vortexed and fluorescence units over basal.] centrifuged at 3000 rpm for 5 min in a Beckman Microfuge. 0 Aliquots (200,ul) were assayed by liquid scintillation counting. A blank value was obtained in the absence of added platelet membranes and always represented 2-5% of total z 20 [32P]GTP added. 0 Statistical Analysis. Pairwise comparisons were performed 40 by the Student's t test. LLI 60 RESULTS 0l 80 TxA2 Receptor Desensitization. Platelets preincubated with 0 either of the TxA2 receptor agonists (U46619 or SQ 26,655) were 60-70% less responsive to a subsequent stimulus with 100 either agonist, as measured by the aggregation response. Occasionally, full desensitization was observed (an example FIG. 2. Inhibition of induction of desensitization by the TxA2 is shown in Fig. 2). Thrombin (0.1 unit/ml)-stimulated ag- receptor antagonist SQ 29,548. Platelets were incubated with either vehicle (control), SQ 29,548 (1 AM), SQ 29,548 (1MM) + U46619 (1.4 gregation was slightly decreased (by 10%) by preincubation AM), or U46619 (1.4 ,uM) for 30 min, washed twice, stimulated with with either TxA2 mimetic (the decrease in thrombin stimu- U46619 (1 MAM), and aggregation was monitored. In this experiment, lation was significantly less than TxA2 receptor agonist U46619 stimulation was, as occurred occasionally, completely de- desensitization; P < 0.001) (Fig. 1A). The onset of desensi- sensitized. The effects of the antagonist were replicated in four tization was rapid-desensitization was maximal by 15 min of independent experiments. Downloaded by guest on September 30, 2021 126 Biochemistry: Murray and FitzGerald Proc. Natl. Acad. Sci. USA 86 (1989) activity was stimulated 41% ± 2% over basal by SQ 26,655 100 B (5 ,uM) and 27% ± 4% over basal by thrombin (0.5 unit/ml) in control was measured in Cl) ' platelets (basal activity the :z 80 presence of the antagonist SQ 29,548 and was 6.70 ± 0.3 LU pmol mg-l-min1). Desensitization resulted in no change in * basal or thrombin-stimulated activity, but SQ 26,655 stimu- C-) 0 40 * ] lation was decreased to 12% ± 4% over basal GTPase LU b-l activity, indicative of an impaired TxA2 receptor-G protein C/) 20 LU-lJ coupling in desensitized platelets (Fig. 4). TxA2 Receptor Binding Properties. We investigated the 0 3 1 2 3 contribution of changes in number and function of TxA2 U- receptors to desensitization. TxA2 receptor affinity for ago- STIMULATION nist was measured by competition of U46619 for 3H-SQ 29,548 binding sites (and analyzed by the programs LIGAND FIG. 3. Desensitization of Cake release antIdof 47-kDa protein and ALLFIT) and yielded a Kd value of 100 ± 10 nM for both phosphorylation. Control (open bars), U46619 dlesensitized (hatched control and desensitized platelets (n = 7). The EC50 for bars), and SQ 26,655 desensitized (solid bars) platelets were stimu- U46619 stimulation of Ca?+ release was increased in desen- lated with either U46619 (1 AM), SQ 26,655 (1 AlM), or thrombin (0.1 sitized compared with control platelets [EC50 = 475 + 71 nM unit/ml). (A) Cake release was measured as relative fluorescence and 275 ± 51 nM, respectively (Fig. 5)]. over at intensity basal excitation wavelengths Of 340 and 380 nm as The275 + 51 TxA 5)].for the described in the text. (B) The 47-kDa protein V)hosphorylation was The affinity of the TxA2 receptor for the antagonist quantitated by laser densitometry as described iin the text. The data PTA-OH was assessed, also by competition for 3H-SQ 29,548 are expressed as percentage of phosphorylation in control platelets. binding sites. There was an apparent increase in receptor Results represent the mean ± SEM of nine exKperiments. **, P < affinity for the antagonist in desensitized compared with 0.005; *, P < 0.05. control platelets. Analysis ofthe binding curves revealed that PTA-OH was binding to two affinity states ofthe receptor. In Phosphorylation of platelet 47-kDa protteminmcontrol and desensitized platelets, the proportion oflow affinity sites (Kd desensitized platelets was assessed as an index of protein = 6 ± 2.5 gM) was decreased from 50% ± 8% in control to kinase C activation (16). U46619 (1 ,uM), 'SQ 26,655 (1 ,uM), 21% ± 5% in desensitized platelets, resulting in an overall thrombin (0.1 unit/ml), and 12-0-tetradei canoylphorbol 13- apparent increase in affinity for PTA-OH. There was a acetate (10 nM) all stimulated 32p incorporaLtion into a 47-kDa concomitant increase in the proportion of high-affinity re- protein in control platelets. In platelets desensitized with ceptor sites (Kd = 130 + 19 nM; n = 7). either U46619 or SQ 26,655, stimulation o)f phosphorylation It has been demonstrated in other receptor systems that by the TxA2 receptor agonists was attenuaLted (to 27-37% of desensitization may be associated with a decrease in receptor control) (Fig. 3B). Thrombin stimulation of 47-kDa protein density (17-19). Saturation binding of 1251-PTA-OH to control phosphorylation was not significantly changed in desensi- platelets (Bmo = 295 ± 30 fmol per 108 platelets) and platelets tized compared to control platelets. desensitized with either U46619 or SQ 26,655 for 1 hr [Bmax TxA2 Receptor Coupling. We investigatted the possibility = 211 ± 25 and 232 ± 35 fmol per 108 platelets, respectively that TxA2 receptor desensitization resultled from impaired (n = 11)] revealed a 20-30% decrease in receptor sites after TxA2 receptor coupling. TxA2 receptor coupling to its asso- desensitization. However, using 3H-SQ 29,548 to measure ciated G protein in desensitized compatred with control receptor number, no significant decrease was observed in platelets was assessed by measurement of platelet high- desensitized (Bma = 212 ± 14 fmol per 108 platelets) affinity GTPase and by comparison of t]he dose-response compared to control platelets [Bma, = 245 ± 8 fmol per 108 curves for U46619 stimulation of calciuml release. GTPase platelets (n = 13)] at this time. The desensitization incubation was extended to 24 hr and 50 1 receptor density was assayed, using 3H-SQ 29,548, at various 100 -j 40 - cn t z 80

r T w ,> 30- to000 x 60 P: M 40 $ 20 - + 0 -j T 0 M.- 20 cl) 10

10-8 10-6

0- --26 [U46619] SO 26,655 Thrombin FIG. 5. Dose-response relationship of U46619 stimulation of FIG. 4. GTPase activity. Control platelets (open bars) and plate- Ca,2+ release. Control (i) or desensitized (I fura-2-loaded platelets lets desensitized with SQ 26,655 (stippled bars) were assayed for were stimulated with increasing concentrations of U46619 and Ca,+ high-affinity GTPase activity in the presence of 5 A±M SQ 26,655 or release assessed as described in the text. Results are expressed as thrombin (0.5 unit/ml). Results are expressed as percentage stimu- percentage of maximal Ca2+ stimulation for each treatment (maximal lation over basal (basal stimulation = 6.7 ± 0.3 pmol mg-l-mink) CaI+ stimulation in control and desensitized platelets = 1.1 + 0.1 and and are the mean ± SEM of five experiments performed in triplicate. 0.5 0.08, respectively) and are the mean SEM from four ***, P < 0.001. experiments. Downloaded by guest on September 30, 2021 Biochemistry: Murray and FitzGerald Proc. Natl. Acad. Sci. USA 86 (1989) 127 The sequence of events thought to be involved in homol- ogous desensitization of the B3-adrenergic receptor has been 100 - W. extensively defined (23). In the continued presence of ago- I nist, the fi-adrenergic receptor is rapidly phosphorylated- 80 - the receptor becomes uncoupled from Gs and, therefore, C mD from the adenylate cyclase enzyme. The receptor is then o 6 0 sequestered from the membrane surface and ultimately 000,.rcLS down-regulated, or converted to a form that can no longer 40 - bind radioligands and is not detectable by immunofluores- cence (24). We have made several observations consistent LU 20 - ws with a similar sequence of events during TxA2 receptor desensitization. TxA2 receptor stimulation has been shown to be mediated 5 10 15 20 25 by a G protein-in permeabilized platelets, U46619-induced DESENSITIZATION TIME (hrs) phosphatidic acid production and secretion were inhibited by the G protein inhibitor GDP[/3S] (25). Stimulation of a FIG. 6. Down-regulation of receptor number. Control (o) and high-affinity GTPase by TxA2 receptor agonists (U46619 and U46619 desensitized (A) platelets were washed after 1-24 hr of U44069) has also been demonstrated (26, 27). We assessed desensitization and were assayed, using 3H-SQ 29,548, for TxA2 receptor number. Results are expressed as percentage of receptor TxA2 receptor stimulation of GTPase activity as a measure of number in control platelets (263 + 27 fmol per 108 platelets) and receptor-G protein coupling in control and desensitized represent mean ± SEM of six experiments. platelets. After 30 min to 1 hr of desensitization, there was little change in TxA2 receptor properties-no change in time points (Fig. 6). After desensitization for 24 hr, the affinity for agonist and no loss of 3H-SQ 29,548 binding sites. receptor density had decreased to 50o of control (control There was, however, a 60% reduction in GTPase stimulation receptor number was 263 ± 27 fmol per 108 platelets). by SQ 26,655, consistent with a loss of TxA2 receptor-G Heterologous Desensitization of the TxA2 Receptor. While protein coupling. Thrombin stimulation of GTPase activity exposure of platelets to TxA2 receptor agonists only slightly was unchanged in desensitized platelets, suggesting that decreased the subsequent stimulation of aggregation by either the small (10-30%) decrease in aggregation and Ca2+ thrombin, pretreatment of EDTA anticoagulated PRP with responsiveness to thrombin stimulation in desensitized plate- thrombin (0.15 unit/ml) reduced the subsequent aggregation lets is not mediated by a change in thrombin receptor responsiveness of the platelets to stimulation to activation, or that the GTPase assay was not sensitive enough by U46619 to detect a small change in thrombin-stimulated GTPase 39% ± 8% of control aggregation (n = 4). activity. Since the TxA2 receptor and the thrombin receptor appear to couple to distinct G proteins (25), it is not clear DISCUSSION whether the locus of the modification responsible for impaired TxA2 receptor-G protein coupling is the TxA2 recep- This study was designed to explore the mechanisms involved tor or the G protein. in regulation of the platelet TxA2 receptor after exposure to The dose-response curves in control and desensitized TxA2 receptor agonists. Carmo et al. (20) reported U46619 platelets for U46619 stimulation of Ca2+ release also suggest and arachidonic acid-induced cross-desensitization of the that uncoupling of the TxA2 receptor from phospholipase C platelet aggregation response. This observation has been is the first stage in its desensitization-the EC50 for U46619 confirmed and extended to include desensitization induced was increased in desensitized platelets and may indicate an by a structurally distinct TxA2 receptor agonist, SQ 26,655. uncoupling of receptor compared to control. The induction of desensitization in EDTA-anticoagulated After 1 hr of desensitization, there was a 20-30% decrease PRP demonstrates that aggregation is not necessary for in TxA2 receptor number using the radioligand 15I-PTA-OH desensitization to occur and removes the necessity of pros- to measure binding. This receptor down-regulation was not tacyclin addition, preventing any possible contribution to or observed when the radioligand 3H-SQ 29,548 was used. inhibition of desensitization by elevated cyclic AMP (21). While it could be argued that the difference in 125I-PTA-OH Platelet biochemical responses (Ca?+ release and 47-kDa and 3H-SQ 29,548 binding in desensitized platelets may be protein phosphorylation) were correlated with changes in the due to residual agonist, we feel this is unlikely. If U46619 or aggregation response after desensitization. The stimulation SQ 26,655 had been retained after washing and were not of all three parameters by TxA2 receptor agonists was de- efficiently displaced by 125I-PTA-OH, we would expect to see creased by at least 60%o after desensitization. Although a decrease in affinity for the ligand. In fact, there was no thrombin stimulation of aggregation and Ca1?' stimulation by change in affinity for 1251-PTA-OH after desensitization (Kd thrombin were slightly decreased (by 10% + 4% and 27% ± = 23 + 2 nM in control and 16 + 2 nM in desensitized 6%, respectively) in desensitized platelets, the decreased platelets; P > 0.05). Heterogeneity in binding of TxA2 responsiveness to thrombin was significantly less (P < 0.001) receptor radioligands has been a feature of platelet TxA2 than to the TxA2 receptor agonists. It is important to note receptor studies-there have been discrepancies in the sites that, although in vivo thrombin stimulation is thought to be labeled by 3H-U46619 and 125I-PTA-OH (28). While the potentiated by release of TxA2, the decrease in thrombin 3H-U46619 and 125I-PTA-OH site difference has been attrib- stimulation of aggregation and Caj2+ release observed in our uted to different properties of agonist and antagonist binding, experiments was not due to a desensitization of the TxA2 there may be a more fundamental difference in the nature of contribution to stimulation-all of the experiments reported the sites identified or in the relative specificity of the ligands here were performed in the presence of indomethacin at used. Another indication of heterogeneity of platelet TxA2 concentrations that inhibit the production of endogenous receptor sites is the observation that the antagonist PTA-OH TxA2 (22). recognizes two affinity states of the receptor-states that are The rapid kinetics of desensitization are consistent with a not discriminated by the other agonists and antagonists used functional role for this process in regulating the effects of this in this study. One interpretation of the existence of two sites evanescent (t.2, =30 sec) platelet agonist in vivo. The platelets for PTA-OH and the relative decrease in the proportion of remained desensitized for at least 1 hr (data not shown). low-affinity sites following desensitization is that the low- and Downloaded by guest on September 30, 2021 128 Biochemistry: Murray and FitzGerald Proc. Natl. Acad. Sci. USA 86 (1989) high-affinity sites for PTA-OH represent coupled and uncou- 4. FitzGerald, G. A., Oates, J. A., Hawiger, J., Maas, R. L., pled forms ofthe receptor, respectively. This is supported by Roberts, L. J., II, Lawson, J. A. & Brash, A. R. (1983) J. Clin. the observation that the for U46619 stimulation of Invest. 71, 676-688. EC50 CaF+ 5. Coleman, R. A., Humphrey, P. P. A., Kennedy, I., Levy, release in control platelets is higher than the Kd for U46619 G. P. & Lumley, P. (1981) Br. J. Pharmacol. 73, 773-778. binding, suggesting that there are uncoupled or poorly coupled 6. Harris, D. N., Greenberg, R., Phillips, M. B., Michel, I. G., receptors in control platelets. After 30 min of desensitization, Goldenberg, J., Haslanger, M. F. & Steinbacher, T. E. (1984) there is an increase in the proportion of uncoupled receptors Eur. J. Pharmacol. 103, 9-18. (reflected in an increased EC50 for U46619 in desensitized 7. Mustard, J. F., Perry, D. W., Ardlie, N. G. & Packham, M. A. platelets). (1972) Br. J. Haematol. 22, 193-204. 8. Grynkiewicz, G., Poenie, M. & Tsein, R. Y. (1985) J. Biol. After the initial rapid uncoupling of TxA2 receptor and its Chem. 260, 3440-3450. associated G protein, there was, in the continued presence of 9. Lyons, R. M., Stanford, N. & Majerus, P. W. (1975) J. Clin. agonist, a slow down-regulation ofreceptor, demonstrated by Invest. 56, 924-936. a 50% ± 8% loss in receptor number following 24 hr of 10. Laemmli, U. K. (1970) Nature (London) 227, 680-685. desensitization, as measured using 3H-SQ 29,548. This long- 11. Saussy, D. L., Jr., Mais, D. E., Burch, R. M. & Halushka, term down-regulation of receptor number may have some P. V. (1986) J. Biol. Chem. 261, 3025-3029. clinical relevance; for example, in conditions associated with 12. Munson, P. J. & Robard, D. (1980) Anal. Biochem. 107, 220- sustained augmentation ofTxA2 biosynthesis, such as severe 239. 13. Vachon, L., Costa, T. & Herz, A. (1987) Biochem. Pharmacol. peripheral vascular disease (29). 36, 2889-2897. The TxA2 receptor was subject to heterologous desensiti- 14. Halushka, P. V., MacDermott, J., Knapp, D. R., Eller, T., zation following platelet exposure to thrombin, suggesting Saussy, D. L., Jr., Mais, D., Blair, I. A. & Dollery, C. T. that refractoriness to TxA2 stimulation may be of general (1985) Biochem. Pharmacol. 34, 1165-1170. importance in regulating platelet activation by stimulants 15. Cassel, D. & Selinger, Z. (1977) Biochem. Biophys. Res. other than TxA2 itself. Commun. 77, 868-873. Given the rapid amplification that occurs upon platelet 16. Nishizuka, Y. (1984) Nature (London) 308, 693-698. stimulation due to release of arachidonic acid and its metab- 17. Scarpace, P. J. & Abrass, I. B. (1982) J. Pharmacol. Exp. Ther. olism, an understanding of the regulation of the complex 223, 327-331. 18. Homburger, V., Lucas, M., Cantau, B., Barabe, J., Penit, J. & pathways of platelet activation is desirable. The rapid un- Bockaert, J. (1980) J. Biol. Chem. 255, 10436-10444. coupling of the TxA2 receptor upon exposure to TxA2 19. Shima, S., Komoriyama, K., Hirai, M. & Kouyama, H. (1983) receptor agonists and subsequent down-regulation of recep- J. Biol. Chem. 258, 2083-2086. tor number and the heterologous desensitization of the TxA2 20. Carmo, L. G., Hatmi, M., Rotilio, D. & Vargaftig, B. B. (1985) receptor by thrombin are consistent with the hypothesis that Br. J. Pharmacol. 85, 849-859. desensitization to TxA2 functions as an important element in 21. Hatmi, M., Rotilio, D., Haife, B., Antonicielli, F., Joseph, D. the physiological control of platelet activation in vivo. & Vargaftig, B. B. (1986) Eur. J. Pharmacol. 132, 219-228. 22. Rittenhouse-Simmons, S. (1981) J. Biol. Chem. 256, 4153- 4155. We gratefully acknowledge the technical assistance of Ms. Libby 23. Harden, T. K. (1983) Pharmacol. Rev. 35, 5-32. Shipp and the secretarial assistance of Ms. Esther Stuart. This work 24. Zemick, B. A. & Strader, C. D. (1988) Biochem. J. 251, 333- was supported in part by Grant HL30400 from the National Institutes 339. of Health. R.M. is the recipient of a fellowship from the American 25. Brass, L. F., Shaller, C. C. & Belmonte, E. J. (1987) J. Clin. Heart Association (Tennessee Affiliate). G.A.F. is an Established Invest. 79, 1269-1275. Investigator of the American Heart Association. 26. Avdonin, P. V., Svitina-Ulitina, I. V., Leytin, V. L. & Tka- chuk, V. A. (1985) Thromb. Res. 40, 101-112. 1. Hamberg, M., Svensson, J. & Samuelsson, B. (1975) Proc. 27. Houslay, M. D., Bojanic, D. & Wilson, A. (1986) Biochem. J. Natl. Acad. Sci. USA 72, 2994-2998. 234, 737-740. 2. Kambayashi, J., Morimoto, K., Hakata, N., Kobayasha, T. & 28. Liel, N., Mais, D. E. & Halushka, P. V. (1987) Prostaglandins Kosaki, G. (1981) Thromb. Res. 22, 553-556. 33, 789-797. 3. Watson, S. P., Ruggiero, M., Abrahams, S. & Lapetina, E. G. 29. Reilly, I. A. G., Doran, J., Smith, B. & FitzGerald, G. A. (1986) J. Biol. Chem. 261, 5368-5372. (1986) Circulation 73, 1300-1309. Downloaded by guest on September 30, 2021